Arrays of Double-Decker Porphyrins on Highly Oriented Pyrolytic

Alexander Falber, Benjamin P. Burton-Pye, Ivana Radivojevic, Louis Todaro, Raihan Saleh, Lynn C. Francesconi, Charles Michael Drain. . European Journa...
1 downloads 0 Views 488KB Size
Download PDF
5708

Langmuir 2006, 22, 5708-5715

Arrays of Double-Decker Porphyrins on Highly Oriented Pyrolytic Graphite Joe Otsuki,*,† Satoru Kawaguchi,† Toshihisa Yamakawa,† Masumi Asakawa,‡ and Koji Miyake§ College of Science and Technology, Nihon UniVersity, 1-8-14 Kanda Surugadai, Chiyoda-ku, Tokyo 101-8308, Japan, Nanoarchitectonics Research Center (NARC), National Institute of AdVanced Industrial Science and Technology (AIST), Tsukuba Central 5, 1-1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan, and AdVanced Manufacturing Research Institute (AMRI), AIST, 1-2-1 Namiki, Tsukuba, Ibaraki 305-8564, Japan ReceiVed March 30, 2006. In Final Form: April 17, 2006 Three double-decker complexes of cerium(IV) were synthesized, which commonly have a 5,10,15,20-tetrakis(4docosyloxyphenyl)porphyrin (C22OPP) moiety as one of the two tetrapyrrole rings. The three complexessCe(Pc)(C22OPP), Ce(C22OPP)2, and Ce(BPEPP)(C22OPP)sare distinguished by the other rings, which are Pc ()phthalocyanine), C22OPP, and BPEPP ()5,15-bis[4-(phenylethynyl)phenyl]porphyrin), respectively. The rate of inter-ring rotation of Ce(BPEPP)(C22OPP) was estimated to be ∼3 s-1 in solution at room temperature. These complexes assemble into ordered arrays at the interface of 1-phenyloctane and the highly oriented pyrolytic graphite surface, owing to the affinity of the long alkyl chains toward the surface, as revealed by means of scanning tunneling microscopy (STM) with molecular resolution. The shape of the upper ring is reflected in the STM image. Thus, Ce(Pc)(C22OPP), Ce(C22OPP)2, and Ce(BPEPP)(C22OPP) were observed as circular, square, and elliptic features, respectively. Possible molecular arrangements in the array of Ce(BPEPP)(C22OPP) are proposed by comparing STM images and molecular models. In the mixed arrays of Ce(BPEPP)(C22OPP) and H2(C22OPP), the double-decker complexes were distinguished by brighter features. Competitive adsorption experiments showed that the adsorption of Ce(BPEPP)(C22OPP) is less favorable than that of H2(C22OPP) by ∆Gapp ) 2.7 kJ mol-1. Ce(BPEPP)(C22OPP) molecules appeared elliptic when placed within their own row, while they appeared isotropic when flanked by H2(C22OPP) molecules. Implications of the differences in the observed shapes to the inter-ring rotation are discussed.

Introduction Ordered two-dimensional assemblies of molecules on atomically flat conductive surfaces, such as highly oriented pyrolytic graphite (HOPG), metals, and semiconductors, are attractive targets of research, since it is possible to probe the arrangement of molecules at the molecular resolution by means of scanning tunneling microscopy (STM).1-6 Information on precise supramolecular structures at an individual molecule level would provide the basis for the bottom-up construction of moleculebased nanomaterials, in which molecular arrangement is precisely controlled, by giving insights into molecule-molecule and molecule-substrate interactions. The approach using HOPG is especially attractive, since the assemblies can be prepared and observed under ambient conditions, without the need for ultrahigh vacuum or low-temperature conditions. Surface assemblies from porphyrins and phthalocyanines are of special interest for their ample electronic, photonic, and catalytic properties. In recent years, assemblies of various porphyrins and * To whom correspondence should be addressed. Fax: +81-3-32590817. E-mail: [email protected]. † Nihon University. ‡ NARC, AIST. § AMRI, AIST. (1) Cyr, D. M.; Venkataraman, B.; Flynn, G. W. Chem. Mater. 1996, 8, 16001615. (2) (a) De Feyter, S.; Hofkens, J.; Van der Auweraer, M.; Nolte, R. J. M.; Mu¨llen, K.; Schryver, F. C. Chem. Commun. 2001, 585-592. (b) De Feyter, S.; De Schryver, F. C. Chem. Soc. ReV. 2003, 32, 139-150. (c) De Feyter, S.; De Schryver, F. C. J. Phys. Chem. B 2005, 109, 4290-4302. (3) Samorı`, P.; Rabe, J. P. J. Phys. Condens. Matter. 2002, 14, 9955-9973. (4) Barth, J. V.; Weckesser, J.; Lin, N.; Dmitriev, A.; Kern, K. Appl. Phys. A 2003, 76, 645-652. (5) Hecht, S. Angew. Chem., Int. Ed. 2003, 42, 24-26. (6) Ruben, M. Angew. Chem. Int. Ed. 2005, 44, 1594-1596.

phthalocyanines have been investigated on surfaces of various substrates. Porphyrins and phthalocyanines which assemble themselves into ordered arrays on HOPG range from relatively simple compounds7-9 and alkylated derivatives10-13 to elaborate multiporphyrin compounds.14 Introducing long alkyl chains into (7) (a) Tao, N. J.; Cardenas, G.; Cunha, F.; Shi, Z. Langmuir 1995, 11, 44454448. (b) Tao, N. J. Phys. ReV. Lett. 1996, 76, 4066-4069. (8) (a) Xu, B.; Yin, S.; Wang, C.; Qiu, X.; Zeng, Q.; Bai, C. J. Phys. Chem. B 2000, 104, 10502-10505. (b) Lei, S. B.; Wang, C.; Yin, S. X.; Wang, H. N.; Xi, F.; Liu, H. W.; Xu, B.; Wan, L. J.; Bai, C. L. J. Phys. Chem. B 2001, 105, 10838-10841. (c) Lei, S. B.; Wang, C.; Yin, S. X.; Bai, C. L. J. Phys. Chem. B 2001, 105, 12272-12277. (d) Lei, S.; Xu, B.; Wang, C.; Xu, Q.; Wan, L.; Bai, C. Jpn. J. Appl. Phys. 2001, 4273-4276. (e) Yin. S.; Wang, C.; Xu, B.; Bai, C. J. Phys. Chem. B 2002, 106, 9044-9047. (f) Lei, S.-B.; Wang, C.; Wan, L.-J.; Bai, C.-L. Jpn. J. Appl. Phys. 2003, 42, 4729-4733. (g) Lei, S.; Wang, C.; Wan, L.; Bai, C. J. Phys. Chem. B 2004, 108, 1173-1175. (h) Lu, J.; Lei, S.; Zeng, Q.; Kang, S.; Wang, C.; Wan, L.; Bai, C. J. Phys. Chem. B 2004, 108, 51615165. (i) Yang, Z.-Y.; Lei, S.-B.; Gan, L.-H.; Wan, L.-J.; Wang, C.; Bai, C.-L. Chem. Phys. Chem. 2005, 6, 65-70. (9) (a) Walzer, K.; Hietschold, M. Surf. Sci. 2001, 471, 1-10. (b) Gopakumar, T. G.; Lackinger, M.; Hackert, M.; Mu¨ller, F.; Hietschold, M. J. Phys. Chem. B 2004, 108, 7839-7843. (10) (a) Qiu, X.; Wang, C.; Zeng, Q.; Xu, B.; Yin, S.; Wang, H.; Xu, S.; Bai, C. J. Am. Chem. Soc. 2000, 122, 5550-5556. (b) Qiu, X.; Wang, C.; Yin, S.; Zeng, Q.; Xu. B.; Bai, C. J. Phys. Chem. B 2000, 104, 3570-3574. (c) Wang, H.; Wang, C.; Zeng, Q.; Xu, S.; Yin, S.; Xu, B.; Bai, C. Surf. Interface Anal. 2001, 32, 266-270. (d) Liu, Y.; Lei, S.; Yin, S.; Xu, S.; Zheng, Q.; Zeng, Q.; Wang, C.; Wan, L.; Bai, C. J. Phys. Chem. B 2002, 106, 12569-12574. (e) Lei, S. B.; Wang, J.; Dong, Y. H.; Wang, C.; Wan, L. J.; Bai, C. L. Surf. Interface Anal. 2002, 34, 767-771. (f) Lei, S. B.; Yin, S. X.; Wang, C.; Wan, L. J.; Bai, C. L. Chem. Mater. 2002, 14, 2837-2838. (11) Ohshiro, T.; Ito, T.; Bu¨hlmann, P.; Umezawa, Y. Anal. Chem. 2001, 73, 878-883. (12) (a) Ikeda, T.; Asakawa, M.; Goto, M.; Miyake, K.; Ishida, T.; Shimizu, T. Langmuir 2004, 20, 5454-5459. (b) Ikeda, T.; Asakawa, M.; Miyake, K.; Shimizu, T. Chem. Lett. 2004, 33, 1418-1419. (c) Otsuki, J.; Nagamine, E.; Kondo, T.; Iwasaki, K.; Asakawa, M.; Miyake, K. J. Am. Chem. Soc. 2005, 127, 10400-10405. (13) Zhou, Y.; Wang, B.; Zhu, M.; Hou, J. G. Chem. Phys. Lett. 2005, 403, 140-145.

10.1021/la0608617 CCC: $33.50 © 2006 American Chemical Society Published on Web 05/18/2006

Arrays of Double-Decker Porphyrins

a molecule is a versatile approach to immobilize the alkylated molecule due to the tendency of alkyl moieties to form closely packed arrays on HOPG.1 Thus, 5,10,15,20-tetrakis(alkoxyphenyl)porphyrins form linear lamellar arrays, the alkyl chains from neighboring rows being interdigitated.10,12 Porphyrin and phthalocyanine macrocycles complex with rare earth metal ions to form sandwich-type double-decker compounds.15 There are only a few STM studies on homoleptic doubledecker phthalocyanines physisorbed on solid surfaces and are none on double-decker porphyrins or on heteroleptic complexes to the best of our knowledge. Early studies did not find an ordered array of double-decker phthalocyanines.16,17 Only recent studies have shown that introducing alkyl chains is helpful in immobilizing double-decker phthalocyanine complexes in ordered two-dimensional arrays on HOPG.18,19 These studies have revealed that alkylated double-decker phthalocyanines adsorb with the phthalocyanine units parallel to the graphite surface and the two-dimensional arrangement is practically identical to those by the corresponding monomeric alkylated phthalocyanines. Also, some double-decker phthalocyanines were observed on Au(111) in electrolyte solutions, although the adsorbed layers are less ordered.20 Herein, we present the first STM observations of doubledecker porphyrins, including heteroleptic complexes, the structures of which are shown in Chart 1. They contain 5,10,15,20tetrakis(4-docosyloxyphenyl)porphyrin (C22OPP) as one of the macrocycles as a common element for the adsorption on the surface of HOPG. The other ring, which will become the upper ring in the array, is phthalocyanine (Pc), C22OPP, or 5,15-bis[4-(phenylethynyl)phenyl]porphyrin (BPEPP). We have found that all these double-decker compounds form ordered monolayer arrays on the HOPG surface at the liquid-solid interface. We have investigated molecular packing in the array and relative adsorption affinity toward the surface as compared to free-base porphyrin. Besides, we have found that the shape of the upper ring is reflected in the STM image. Thus, Ce(Pc)(C22OPP), Ce(C22OPP)2, and Ce(BPEPP)(C22OPP) were observed circular, square, and elliptic features, respectively. In the mixed monolayer of Ce(BPEPP)(C22OPP) and H2(C22OPP), however, the doubledecker molecules flanked by the free-base porphyrin molecules are imaged as isotropic features. Implications of the differences in the observed shapes to the inter-ring rotation are discussed. (14) (a) Samorı´, P.; Engelkamp, H.; de Witte, P.; Rowan, A. E.; Nolte, R. J. M.; Rabe, J. P. Angew. Chem. Int. Ed. 2001, 40, 2348-2350. (b) Elemans, J. A. A. W.; Lensen, M. C.; Gerritsen, J. W.; van Kempen, H.; Speller, S.; Nolte, R. J. M.; Rowan, A. E. AdV. Mater. 2003, 15, 2070-2073. (c) Lensen, M. C.; van Dingenen, S. J. T.; Elemans, J. A. A. W.; Dijkstra, H. P.; van Klink, G. P. M.; van Koten, G.; Gerritsen, J. W.; Speller, S. Nolte, R. J. M.; Rowan, A. E. Chem. Commun. 2004, 762-763. (d) van Gerven, P. C. M.; Elemans, J. A. A. W.; Gerritsen, J. W.; Speller, S.; Nolte, R. J. M.; Rowan, A. E. Chem. Commun. 2005, 3535-3537. (15) For reviews, see: (a) Ng, D. K. P. Jiang, J. Chem. Soc. ReV. 1997, 26, 433-442. (b) Jiang, J.; Liu, W.; Arnold, D. P. J. Porphyrins Phthalocyanines 2003, 7, 459-473. (c) Jiang, J.; Kasuga, K.; Arnold, D. P. In Supramolecular PhotosensitiVe and ElectroactiVe Materials; Nalwa, H. S., Ed.; Academic Press: San Diego, 2001; pp 113-210. (d) Buchler, J. W.; Ng, D. K. P. In The Porphyrin Handbook; Kadish, K. M., Smith, K. M., Guilard, R., Eds.; Academic Press: San Diego, 2000; Vol. 3, pp 245-294. (e) Weiss, R.; Fischer, J. In The Porphyrin Handbook; Kadish, K. M., Smith, K. M., Guilard, R., Eds.; Elsevier Science: San Diego, 2003; Vol. 16, pp 171-246. (16) Jones, R.; Krier, A.; Davidson, K. Thin Solid Films 1997, 298, 228-236. (17) Capobianchi, A.; Paoletti, A. M.; Pennesi, G.; Rossi, G.; Scavia, G. Surf. Sci. 2003, 536, 88-96. (18) (a) Binnemans, K.; Sleven, J.; De Feyter, S.; De Schryver, F. C.; Donnio, B.; Guillon, D. Chem. Mater. 2003, 15, 3930-3938. (b) Klymchenko, A. S.; Sleven, J.; Binnemans, K.; De Feyter, S. Langmuir 2006, 22, 723-728. (19) Yang, Z.-Y.; Gan, L.-H.; Lei, S.-B.; Wan, L.-J.; Wang, C.; Jian, J.-Z. J. Phys. Chem. B 2005, 109, 19859-19865. (20) Ma, H.; Yang, L.-Y. O.; Pan, N.; Yau, S.-L.; Jiang, J.; Itaya, K. Langmuir 2006, 22, 2105-2111.

Langmuir, Vol. 22, No. 13, 2006 5709 Chart 1. Array-Forming Double-Decker Complexes

Experimental Procedures Synthesis. Column chromatography was performed using Kanto Chemical silica gel (SiO2) 60N. Laser-desorption ionization (LDI) mass spectra were taken on an Applied Biosystems Voyager RPPRO with a time-of-flight (TOF) spectrometer. High-resolution mass spectra (HRMS) were taken on an Agilent G1969A system with an atmospheric pressure chemical ionization (CI) chamber and a TOF spectrometer. 1H NMR spectra were recorded on a JEOL GX400 spectrometer. Tetramethylsilane was used as an internal standard for chemical shifts (δ ) 0). Proton identifiers used in the assignment are shown in Chart 1 and Scheme 1. Recycle GPC was conducted using CHCl3 as eluent on a Japan Analytical Industry LC-9201 system equipped with Jaigel-2H and Jaigel-1H columns in tandem, with exclusion limits of 5000 and 1000, respectively. The double-decker complex, Ce(BPEPP)(C22OPP), was prepared as shown in Scheme 1, while Ce(Pc)(C22OPP) was prepared by reacting Ce(acac)3 (acac ) acetylacetonate) with H2(C22OPP) and the lithium salt of phthalocyanine (Li2Pc). C22OPP,21 4-(phenylethynyl)benzaldehyde,22 and dipyrromethane23 were prepared ac(21) Matile, S.; Berova, N.; Nakanishi, K.; Fleischhauer, J.; Woody, R. W. J. Am. Chem. Soc. 1996, 118, 5198-5206. (22) (a) De la Rosa, M. A.; Velarde, E.; Guzma´n, A. Synth. Commun. 1990, 20, 2059-2064. (b) Leadbeater, N. E.; Tominack, B. J. Tetrahedron Lett. 2003, 44, 8653-8656. (23) Wang, Q. M.; Bruce, D. W. Synlett. 1995, 1267-1268.

5710 Langmuir, Vol. 22, No. 13, 2006 Scheme 1. Synthesis of Ce(BEPP)(C22OPP)

cording to literature procedures. The purity of prepared materials was carefully confirmed with spectroscopic data, TLC, and GPC. Cerium Double-Decker Complex of Phthalocyanine and 5,10,15,20-Tetrakis(4-docosyloxyphenyl)porphyrin (Ce(Pc)(C22OPP)). A mixture of H2(C22OPP) (111 mg, 0.06 mmol) and Ce(acac)3‚ nH2O (60 mg, 1.14 mmol) in 1,2,4-trichlorobenzene (6 mL) was refluxed for 24 h under Ar. After the reaction was cooled to room temperature, Li2Pc (32 mg, 0.06 mmol) was added and the mixture was heated again to reflux for 4 h. The solvent was evaporated, and the residue was partitioned between CHCl3 and H2O. The insoluble material and the aqueous layer were extracted with CHCl3 several times. The combined CHCl3 layer was dried over Na2SO4. The removal of Na2SO4 by filtration and evaporation of the solvent gave a dark purple solid. The crude mixture was chromatographed (SiO2, CHCl3/hexane, 3:2) to give the following fractions in the eluted order, which were analyzed with LDI-MS: (1) a yellow solid, 10 mg, unidentified; (2) a purple solid, 32 mg, H2(C22OPP); (3) a green solid, 23 mg, Ce(Pc)(C22OPP); (4) a black solid, 23 mg, Ce(Pc)2. The desired third fraction was reprecipitated from CHCl3/MeOH to afford the pure Ce(Pc)(C22OPP), 8 mg, 0.003 mmol, 5%. 1H NMR (CDCl3): δ 0.88 (12H, t, J ) 6.8 Hz, CH3), 1.2-1.7 (152H, m, (CH2)19), 1.98 (8H, quintet, J ) 7.0 Hz, OCCH2), 4.19 (8H, t, J ) 7.0 Hz, OCH2), 6.33 (4H, d, J ) 8.2 Hz, oex), 6.71 (4H, d, J ) 8.2 Hz, oen), 7.21 (8H, br, m), 8.26 (8H, dd, J ) 5.6 Hz, 2.9 Hz, Pc2), 8.35 (8H, s, β), 9.25 (8H, dd, J ) 5.6 Hz, 2.9 Hz, Pc1). LDI-MS: m/z ) 2563.7 (MH+). CI-HRMS: m/z calcd for C164H221CeN12O4 (MH+), 2564.6570; found, 2564.6485. 5,15-Bis[4-(phenylethynyl)phenyl]porphyrin (H2(BPEPP)). Trifluoroacetic acid (0.07 mL) was added to a solution of dipyrromethane (210 mg, 1.4 mmol) and 4-(phenylethynyl)benzaldehyde (200 mg, 0.97 mmol) in CH2Cl2 (280 mL), which had been dearated by bubbling N2 for 10 min. The reaction was stirred for 3 h at room temperature in the dark. Then, 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (411 mg, 1.8 mmol) was added and the mixture was stirred for another 3 h. After Et3N (1.4 mL) was added, the reaction mixture was passed through a silica gel column to collect a fluorescent band. The solvent was evaporated and the residue was washed with MeOH and CH2Cl2 to afford a purple solid (51 mg, 0.077 mmol, 16%). This material was used without further purification due to its poor solubility. 1H

Otsuki et al. NMR (CDCl3): δ -3.10 (2H, s, NH), 7.45-7.47 (6H, m, H4, H5), 7.72 (4H, d, J ) 6.4 Hz, H3), 8.00 (4H, d, J ) 8.0 Hz, H2), 8.28 (4H, d, J ) 8.0 Hz, H1), 9.11 (4H, d, J ) 4.4 Hz, β1), 9.44 (4H, d, J ) 4.4 Hz, β2), 10.36 (2H, s, meso). LDI-MS: m/z ) 662.4 (M+). CI-HRMS: m/z calcd for C48H31N4 (MH+), 663.2543; found, 663.2545. Cerium Double-Decker Complexes of 5,15-Bis[4-(phenylethynyl)phenyl]porphyrin and 5,10,15,20-Tetrakis(4-docosyloxyphenyl)porphyrin (Ce(BPEPP)(C22OPP), Ce(C22OPP)2, and Ce(BPEPP)2). A mixture of Ce(acac)3‚n(H2O) (22 mg, 0.05 mmol), H2(BPEPP) (17 mg, 0.025 mmol), H2(C22OPP) (48 mg, 0.025 mmol), 1-octanol (2 mL), 1,8-diazabicyclo[5.4.0]undec-7-ene (0.025 mL), and toluene (1 mL) was refluxed for 16 h under Ar. The solvent was evaporated, and the residue was extracted with CHCl3. The CHCl3soluble material was chromatographed (SiO2, CHCl3/hexane, 1:1) to give several fractions, each of which was analyzed by LDI-MS. A fraction containing the heteroleptic double-decker complex Ce(BPEPP)(C22OPP) was collected, which was further purified through recycle GPC to afford a dark orange solid (8 mg, 0.003 mmol, 12%). Other products included various double- and triple-decker complexes, from which the homoleptic double-decker complexes, Ce(BPEEP)2 and Ce(C22OPP), were also isolated. Data for Ce(BPEPP)(C22OPP). 1H NMR (CDCl3): δ 0.88 (12H, t, J ) 7.2 Hz, CH3), 1.25-1.52 (144H, m, (CH2)18), 1.67 (8H, quintet, J ) 7.2 Hz, OCCCH2), 2.02 (8H, quintet, J ) 7.2 Hz, OCCH2), 4.30 (8H, t, J ) 7.2 Hz, OCH2), 6.35 (4H, br, d, J ) 7.2 Hz, oex), 6.55 (2H, br, d, J ) 6.0 Hz, H1ex), 6.81 (4H, br, d, J ) 7.2 Hz, mex), 7.46-7.53 (8H, m, H4, H5, H2ex), 7.76-7.78 (8H, m, H3, men), 8.09 (4H, s, β3 or β4), 8.46-8.48 (10H, m, β1, H2en, β4 or β3), 8.72 (4H, d, J ) 4.4 Hz, β2), 9.05 (2H, s, meso), 9.57 (4H, br, d, J ) 7.2 Hz, oen), 9.73 (2H, br, d, J ) 6.0 Hz, H1en). LDI-MS: m/z ) 2712.9 (MH+). CI-HRMS: m/z calcd for C180H233CeN8O4 (MH+), 2712.7388; found, 2712.7423. Data for Ce(C22OPP)2. 1H NMR (CDCl3): δ 0.88 (24H, t, J ) 6.8 Hz, CH3), 1.26-1.65 (304H, m, (CH2)19), 2.00 (16H, quintet, J ) 6.8 Hz, OCCH2), 4.27 (16H, t, J ) 6.8 Hz, OCH2), 6.30 (8H, br, oex), 6.77 (8H, br, mex), 7.64 (8H, br, men), 8.28 (16H, s, β), 9.48 (8H, br, oen). LDI-MS: m/z ) 3962.4 (M+). CI-HRMS: m/z calcd for C264H409CeN8O8 (MH+) 3963.0998; found, 3963.0787. Data for Ce(BPEPP)2. 1H NMR (CDCl3): δ 6.46 (4H, br, d, J ) 7.6 Hz, H1ex), 7.44-7.52 (16H, m, H2ex, H4, H5), 7.76 (8H, d, J ) 7.6 Hz, H3), 8.40-8.50 (12H, m, H2en, β1), 8.80 (8H, br, β2), 9.16 (4H, s, meso), 9.77 (4H, br, d, J ) 7.6 Hz, H1en). LDI-MS: m/z ) 1460.6 (M+). CI-HRMS: m/z calcd for C96H57CeN8 (MH+), 1462.3787; found, 1462.3772. STM Measurements. 1-Phenyloctane was purchased from Kanto Chemical and used as received. STM measurements were carried out with an SII SPI3800N-SPA400 microscope or a Veeco Instruments Nanoscope IIIa multimode SPM under ambient conditions. These microscopes are equipped with low-current modules. STM tips were made mechanically with Pt/Ir (9:1) wire. STM-1 grade HOPG was purchased from GE Advanced Ceramics. The uppermost layers of HOPG were peeled off with adhesive tape immediately before use. After the freshly cleaved HOPG was observed to confirm the atomic resolution of the tip, a droplet of a sample solution in 1-phenyloctane was applied on the surface just below the tip using a syringe. Bias voltages in figure captions refer to the substrate voltage with respect to the tip. No particular bias dependence were noted between -1.1 and -1.5 V.

Results and Discussion Structural Characterization of Double-Decker Complexes and Inter-ring Rotation in Solution. The LDI-MS is a convenient indicator to show the presence of porphyrin and phthalocyanine complexes, as these compounds give strong molecular ions (M+ or MH+) without the need for any matrix. A high degree of the purity of samples of specific double-decker complexes was indicated by the complete lack of MS peaks for other double- and triple-decker complexes, which had been contained in crude reaction products.

Arrays of Double-Decker Porphyrins

The 1H NMR spectra for Ce(C22OPP)2 and Ce(Pc)(C22OPP) are as expected from reported data for similar double-decker complexes.24,25 Large differences in chemical shifts are noted for phenylene protons depending on the exo and endo orientations. The exo protons appear at higher fields, and the endo protons are in the lower fields due to the ring current effect exerted by the opposite porphyrin or phthalocyanine macrocycle. In the 1H NMR spectrum for Ce(BPEPP)2, the protons on the terminal phenyl groups (H3, H4, and H5) and ones at the meso positions show sharp signals. The resonances for the phenylene protons are distinguished depending on the exo and endo orientations, as in the case for Ce(C22OPP)2. There appear two resonances for the pyrrolic β protons, which are broadened at room temperature as opposed to the case for Ce(C22OPP)2. In the minimum-energy conformation, the angle of rotation of one porphyrin ring away from the eclipsed conformation of the two rings is most likely close to 45° due to an antiprismatic coordination geometry. This means that there are four chemically inequivalent β protons in the molecule. Therefore, each of the two broad resonances is actually a coalesced resonance for two kinds of β protons. This evidences that a 90° oscillation about the main rotational axis of the porphyrin ligands is occurring in a NMR time scale at room temperature (∼25 °C). The 1H NMR spectrum for the heteroleptic Ce(BPEPP)(C22OPP) is nearly a superposition of the spectra for the halves of homoleptic Ce(C22OPP)2 and Ce(BPEPP)2. Two separate singlet peaks appear for β protons (β3 and β4)26 on the C22OPP ring due to the C2 symmetry of the opposing BPEPP ring. This suggests that the interporphyrin rotation, if any, is slower than the NMR time scale at room temperature. Two phenylene groups would have to pass across the opposing phenylene groups simultaneously to effect a 90° rotation in this molecule, providing an effective barrier. To gain quantitative information on the rate of the inter-ring rotation of Ce(BPEPP)(C22OPP), variable-temperature 1H NMR measurements were carried out in CDCl2CDCl2, as shown in Figure 1. The protons on phenylene groups exhibit broad signals even at 20 °C, indicating that the phenylene groups on the both macrocycles are rotating about the C(meso)-C(ipso) bonds in an NMR time scale. These signals become broadened further as temperature is increased to the extent that the signals almost disappear around 80 °C. Another broadening is appreciated for the signals of β3/β4 protons when the temperature is raised above 60 °C. This broadening is due to an increased rate of interporphyrin ring rotation. The spectral shapes were simulated using the formulation for exchanging protons,27 in combination with a least squares curve fitting procedure, as shown with the red lines in the figure. From the simulation, the rates of exchange, i.e., a 90° rotation, are determined as k ) 10 (60 °C), 24 (80 °C), and 91 s-1 (100 °C). The ln k values are linear against reciprocal temperature, indicating that the rotation is an Arrhenius-type activated process, from which the activation energy is determined as Ea ) 51 kJ mol-1. The rate of exchange at 25 °C is estimated to be 3.2 s-1 by extrapolation. Aida28 and Shinkai29 have independently shown that doubledecker porphyrins do rotate and the rate of interporphyrin rotation depends on substituents and temperature. For example, it takes tens of minutes at room temperature for the cerium double(24) Davoras, E. M.; Spyroulias, G. A.; Mikros, E.; Coutsolelos, A. G. Inorg. Chem. 1994, 33, 3430-3434. (25) Lachkar, M.; De Cian, A.; Fischer, J.; Weiss, R. New J. Chem. 1988, 12, 729-731. (26) Identification of each signal to the β3 (facing to the nonsubstituted side of the porphyrin core of the BPEPP moiety) or β4 (facing to the phenylethynylphenyl group) proton has not been attempted. (27) Gutowsky, H. S.; Holm, C. H. J. Chem. Phys. 1956, 25, 1228-1234.

Langmuir, Vol. 22, No. 13, 2006 5711

Figure 1. Variable-temperature 1H NMR spectra for Ce(BEPP)(C22OPP) in CDCl2CDCl2. Assignments for temperature-dependent signals are given on the 20 °C spectrum. Simulated spectra for β3 and β4 protons are shown in red, which are offset for visibility.

decker complex of 5,10,15-tris(4-methoxyphenyl)-20-(4-pyridyl)porphyrin to make a 90° turn.29a This 90° turn demands for the four meso-aryl groups from one of the macrocycles to pass across their counterparts on the opposite macrocycle simultaneously, which gives rise to the most demanding transition barrier. When there is no aryl substituent, a 90° turn is more facile. In the case of the cerium double-decker complex with 5,20-bistolylporphyrin, which is chiral due to its D2 symmetry, the enantiopure complexes racemize quickly at a rate of 80 s-1 at 13 °C28a due to a 90° turn in a way that no phenyl group crossover is involved. The fast oscillation in Ce(BPEPP)2 in the present study is consistent with these literature data. As for Ce(BPEPP)(C22OPP), which has four aryl groups in one macrocycle and two in the other, a 90° turn involves simultaneous crossing of two pairs of phenyl groups. Hence, the steric barrier may lie somewhere between those in double-decker complexes of homoleptic meso-tetraarylporphyrins and those of 5,15-diarylporphyrins. The estimated rotation rate of ∼3 s-1 at 25 °C indeed falls between the values described above and closer in magnitude to that for a previously reported cerium 5,15diarylporphyrin double-decker complex (k ) 0.37 s-1 at 10 °C), which is modified such that a 90° turn is possible only in a way that the two phenylene groups on one porphyrin ring pass across the counterparts in the other porphyrin ring.28b STM Observations of Ce(Pc)(C22OPP) and Ce(C22OPP)2. A droplet of a solution of a double-decker complex in 1-phenyloctane was dropped on an HOPG surface. Then, the complexes spontaneously form ordered monolayer arrays at the liquidsolid interface, which can be observed with STM with molecular resolution. An STM image for Ce(Pc)(C22OPP) is shown in Figure (28) (a) Tashiro, K.; Konishi, K.; Aida, T. Angew. Chem., Int. Ed. Engl. 1997, 36, 856-858. (b) Tashiro, K.; Fujwara, T.; Konishi, K.; Aida, T. Chem. Commun. 1998, 1121-1122. (c) Tashiro, K.; Konishi, K.; Aida, T. J. Am. Chem. Soc. 2000, 122, 7921-7926. (29) (a) Takeuchi, M.; Imada, T.; Ikeda, M.; Shinkai, S. Tetrahedron Lett. 1998, 39, 7897-7900. (b) Sugasaki, A.; Ikeda, M.; Takeuchi, M.; Robertson, A.; Shinkai, S. J. Chem. Soc., Perkin Trans. 1 1999, 3259-3264. (c) Ikeda, M.; Takeuchi, M.; Shinkai, S.; Tani, F.; Naruta, Y. Bull. Chem. Soc. Jpn. 2001, 74, 739-746. (d) Takeuchi, M.; Ikeda, M.; Sugasaki, A.; Shinkai, S. Acc. Chem. Res. 2001, 34, 865-873. (e) Ikeda, M.; Takeuchi, M.; Shinkai, S.; Tani, F.; Naruta, Y.; Sakamoto, S.; Yamaguchi, K. Chem. Eur. J. 2002, 8, 5542-5550.

5712 Langmuir, Vol. 22, No. 13, 2006

Figure 2. STM image for Ce(Pc)(C22OPP) on HOPG; ∼10 µM in 1-phenyloctane; 100 × 100 nm2; tunneling conditions: I ) 2 pA, V ) -1.2 V. Inset shows a magnified image (10 × 10 nm2).

Figure 3. Array of Ce(C22OPP)2 (20 µM) at the 1-phenyloctaneHOPG interface. (a) 50 × 50 nm2; tunneling conditions: I ) 2 pA, V ) -1.3 V. (b) Magnified view (10 × 10 nm2), highlighting the square shaped molecular images. (c) Molecular models consistent with the STM image. Alkyl chains on the upper porphyrin rings are omitted for clarity.

2. Bright round spots make up rows separated by a darker background. The bright and dark regions correspond to the aromatic and aliphatic parts, respectively, of the molecule. This striped arrangement is typical for the assemblies of 5,10,15,20-tetrakis(4-alkoxyphenyl)porphyrins.10,12 Furthermore, the lattice parameters (a ) 4.0 nm, b ) 1.9 nm, θ ) 84°) are close to those for the array of monomeric H2(C22OPP) (a ) 4.2 nm, b ) 2.1 nm, θ ) 86°). This indicates that the alkoxyporphyrin side of the double-decker compound adsorbs on the HOPG surface. Thus, the alkoxyporphyrin ring constitutes the lower ring, while the phthalocyanine ring constitutes the upper ring. An STM image of the array of Ce(C22OPP)2 on HOPG at the liquid-solid interface is shown in Figure 3a. The lattice parameters of the striped arrangement are a ) 4.4 nm, b ) 2.0 nm, and θ ) 84°, again close to those for the monomeric H2(C22OPP)2, which indicates that these molecules make an array in such a way that the porphyrin plane is flat on the surface with alkyl chains from neighboring rows being interdigitated. It also shows that the alkyl chains on the upper rings do not exert a detrimental effect on the array formation and STM measurements. There is no information available as to the arrangement of the upper alkyl chains. It is unlikely that the upper alkyl chains also interdigitate as do the lower alkyl chains since the orientation of the upper porphyrin ring is not optimum for the interdigitation due to a twist from the lower porphyrin rings, as described in the following. The magnified view (Figure 3b) shows that the double-decker molecules are imaged as squares. It is most likely that the corners correspond to the phenylene groups. The orientation of the squares

Otsuki et al.

Figure 4. STM image for Ce(BPEPP)(C22OPP) on HPOG; 20 µM in 1-phenyloctane; 50 × 50 nm2; tunneling conditions: I ) 26 pA, V ) -1.3 V.

is rotated by 45° from that seen in 5,10,15,20-tetrakis(4alkoxyphenyl)porphyrin arrays. A likely arrangement of the molecules is shown in Figure 3c. Note that the four oxygen atoms in the upper ring correspond to the corners of the squares in the STM image (upper alkyl chains are omitted for clarity), consistent with the upper rings being away from the lower ring by 45°. This is consistent with antiprismatic coordination geometry of double-decker complexes. STM Observations of Ce(BPEPP)(C22OPP). The BPEPP moiety has an oblong shape. The distance between the phenyl hydrogens on the both ends of the molecule is 3.16 nm, while the distance between the meso hydrogens at the 10- and 20positions across the porphyrin core is 0.90 nm, according to an MM3 modeling. A typical STM image is shown in Figure 4. The lattice parameters for the two-dimensional array are a ) 4.44 (0.35) nm, b ) 1.98 (0.08) nm, and θ ) 83 (4)°, where values in parentheses are the standard deviations for the data collected from 20 different STM images. These values are again close to those for H2(C22OPP). Each molecule appears as an ellipse, reflecting the shape of the upper BPEPP moiety.30 This is a significant observation since we are able to know the orientation of the molecule. In these orientations of BPEPP, there would be no direct interaction among neighboring BPEPP moieties in a row, as the distance separating them (1.98 nm) is large compared to the width of the BPEPP moiety (0.90 nm). However, a 90° turn of the BPEPP ring would be impossible due to steric hindrance at the tip of the phenylethynyl groups. Inspection of the 20 images revealed that there are two types of orientations for the major axis of the ellipse. In one orientation (11 images), the major axis of an ellipse is along the a axis of the unit cell, as shown in Figure 5a, while in the other orientation (9 images), the axis is in the direction of one of the diagonals of the unit cell (Figure 5b).31 We refer the former type as the rectangular pattern, the latter as the parallelogram pattern. These two different orientations of the BPEPP moiety most likely reflect the difference in the arrangement of underlying porphyrins, since all BPEPP moieties in a single domain adopt the same orientation even though direct interactions between BPEPP moieties are not (30) A slight asymmetry in the ellipses is probably due to an asymmetry in the STM tip. We have confirmed that the elliptic shape is reproducibly obtained in repeated measurements using different tips. Also, the evidence that the elliptic shape is not an artifact of the tip irregularity but intrinsic to the molecule is provided by the image in Figure 7b, wherein the ellipses in different domains orient differently. If the elliptic shape were due to a tip irregularity, all ellipses in a single image would orient in the same direction. (31) Only occasionally, an orientation between these two types was observed. In this orientation, the major axis of each ellipse directs toward the middle of the molecules in the next row, resulting in a staggered arrangement. It seems that this pattern is more often encountered in less-ordered domains with many defect sites nearby, suggesting that the staggered pattern is a transient arrangement.

Arrays of Double-Decker Porphyrins

Figure 5. STM images (10 × 20 nm2) for 20 µM Ce(BPEPP)(C22OPP) in 1-phenyloctane on HOPG. The arrows indicate the direction of the major axes of the ellipses. (a) The rectangular pattern; tunneling conditions: I ) 26 pA, V ) -1.3 V. (b) The parallelogram pattern; tunneling conditions: I ) 10 pA, V ) -1.3 V.

expected. The orientation of lower porphyrins directly adsorbed on the HOPG surface can be determined if the antiprismatic coordination geometry around the cerium ion is assumed. Then, the arrangement of alkyl chains can be estimated from the known dimensions of alkyl chains, assuming that alkyl chains take the all-trans extended conformation. Two possible arrangements of alkyl chains consistent with the unit cell parameters and the orientation of porphyrins discussed above have been found for each of the rectangular and parallelogram patterns. These possible arrangements are shown in Figure 6. Molecular models drawn to scale are superimposed in the upper left diagram to show the correspondence between the molecular structures and the diagrams. For the rectangular pattern, the angle between the porphyrin row (b axis) and the alkyl chains can be 67° or 79°. The alkyl-alkyl separations in these cases are 4.6 and 4.9 Å, respectively. For the parallelogram pattern, the possible angles made with the b axis and the alkyl chains are found to be 57° and 87°, the alkyl-alkyl separations being 4.0 and 4.8 Å, respectively. Two energy-minimized configurations have been found for alkane molecules adsorbed on HOPG by molecular mechanics calculations.32 In one configuration, the plane including the zigzag carbons is parallel to the basal plane of graphite, which is denoted as flat orientation. In the other configuration, the zigzag carbons are on a plane perpendicular to the surface,

Langmuir, Vol. 22, No. 13, 2006 5713

which is called vertical orientation. The distances between the adjacent alkane fragments are 4.5 Å for the flat orientation and 4.1 Å for the vertical orientation.32 These values are consistent with the arrangements shown in Figure 6. STM Observations of Mixtures of Ce(BEPP)(C22OPP) and H2(C22OPP). STM images were also taken for mixed solutions of the double-decker complex, Ce(BPEPP)(C22OPP), and the free-base porphyrin, H2(C22OPP). Representative images are shown in Figure 7, prepared from solutions of different component ratios. In these images, some of the free-base porphyrin molecules are replaced by higher features, which can be attributed to the double-decker complexes, since the number of the higher spots increases on increasing the feed ratio of the double-decker complex. The free-base porphyrins appear as spots with a height of ∼0.2 nm, whereas the double-decker complexes appear as protrusions of a height of 0.7-0.8 nm as shown by the height profile. Thus, the Ce(BEPP)(C22OPP) and H2(C22OPP) form mixed domains due to the matching of the lattice parameters. In Figure 7a, in which most double-decker molecules are isolated in the array of the free base porphyrin molecules, the shape of the individual double-decker complexes is isotropic, which is in contrast to the elongated shape observed in the array consisting purely of Ce(BEPP)(C22OPP). In Figure 7b, some double-decker complexes are placed in a row comprising its own species, while some double-decker complexes are flanked by free-base porphyrin molecules. The former species appear as an ellipse (examples are indicated by a brace), while the latter species (examples are indicated by arrows) appear as a more isotropic round shape. The observed numbers of the double-decker molecules are consistently smaller than expected from the concentration ratio in the feed solution. This observation suggests a weaker adsorption of the double-decker complex onto the surface as compared to the free-base porphyrin. To gain quantitative information on the adsorption behavior in the mixed system, the fraction of H2(C22OPP) molecules in the total number of molecules in STM images were counted as a function of the fraction in the feed solution, [H2(C22OPP)]/([H2(C22OPP)] + [Ce(BPEPP)(C22OPP)]), as shown in Figure 8. As a first approximation, the adsorption

Figure 6. Possible molecular arrangements of Ce(BPEPP)(C22OPP) consistent with the STM images for rectangular (upper two) and parallelogram (lower two) patterns. Angles between the b axis and alkyl chains and widths per alkyl chain are indicated.

5714 Langmuir, Vol. 22, No. 13, 2006

Otsuki et al.

Figure 7. STM images for mixtures of Ce(BPEPP)(C22OPP) and H2(C22OPP) at the 1-phenyloctane-HOPG interface; 50 × 50 nm2; tunneling conditions: I ) 6 pA, V ) -1.3 V. (a) Feed ratio Ce(BPEPP)(C22OPP)/H2(C22OPP) ) 1:2; 20 µM in all. The lower panel shows the height profile along the line in the upper image. (b) Feed ratio 4:1; 20 µM in all. The brace indicates a portion of a row of Ce(BPEPP)(C22OPP), and the arrows indicate double-decker molecules flanked by free-base porphyrins.

Figure 8. Relationship between the fractional solution concentration of H2(C22OPP) in the mixture with Ce(BPEPP)(C22OPP) and the fractional surface coverage at the 1-phenyloctane-HOPG interface. The total concentration is kept constant at 20 µM. The solid curve is based on the Boltzmann distribution with ∆Gapp ) 2.7 kJ mol-1.

behavior may be related to the Boltzmann distribution between different apparent adsorption free energies (Gapp).12a The relationship between the surface coverage (θ) and the concentration in the solution (C) can be expressed as eq 1.

( )

θm C m ∆Gapp ) exp θd Cd RT

(1)

where R and T are the gas constant and temperature (297 K), respectively, and the subscripts m and d stand for monomeric H2(C22OPP) and double-decker Ce(BPEPP)(C22OPP), respectively. The value of ∆Gapp ) 2.7 kJ mol-1 gives the best fit to the data, which is shown as a solid curve in Figure 8. The difference in the adsorption free energy may be partly explained

by an entropy factor resulting from the difference in symmetry; that is, the free-base porphyrin has two equivalent faces to adsorb to the surface while the double-decker complex has only one.12a,33 This contribution to the adsorption free energy amounts to RT ln 2 ()1.7 kJ mol-1 at 297 K). The doming or saucerlike deformation of the C22OPP ring in the double-decker complex15d may account for the remaining 1 kJ mol-1 destabilization. On the Observed Shapes of Individual Molecules. The STM images presented herein clearly distinguish individual molecules. Furthermore, the overall shapes of individual molecules can be discerned from the images. As for Ce(Pc)(C22OPP), circular images were recorded. The high symmetry of phthalocyanine makes it difficult to observe its orientation in the array. In contrast, Ce(C22OPP)2 gave a square image. In the case of 5,10,15,20tetrakis(alkoxyphenyl)porphyrins (e.g., H2(C22OPP)), the tetraphenylporphyrin core is often observed as a square or X-shaped feature, the corners or the ends corresponding to the four peripheral phenyl rings. The orientation of the squares for Ce(C22OPP)2 is rotated from those observed for monomeric free-base porphyrins by 45° and is consistent with the upper ring orientation which is related to the lower ring with an antiprismatic geometry. Thus, the STM image of Ce(C22OPP)2 is more influenced by the upper ring than by the lower. The orientation is even more clearly manifested in the STM image in the case of Ce(BPEPP)(C22OPP), which has an oblongshaped upper ring. Elliptic images are obtained for individual molecules reflecting the shape of the upper ring. In the pure array, all the upper rings are oriented with the long axis more or less perpendicular to the row of molecules. This is again consistent with an antiprismatic geometry. There should be another stable orientation for the upper ring, that is in parallel with the molecular row. However, it is impossible to adopt this orientation since the molecular length is longer than the separation between the molecules in a row. In contrast to the array comprising solely of Ce(BPEPP)(C22OPP), the isolated higher spots in the STM images for the mixed arrays with free-base porphyrin appear as isotropic features without showing the elliptic shape as shown in Figure 7. In the mixed array, the lower porphyrin ring of a double-decker complex should be immobilized in the array by the neighboring free-base porphyrins via interdigitating alkyl chains. Hence, as a reason for the isotropic feature, an interesting possibility can be invoked that the double-decker complex is rotating around the cerium ion as far as the space allows. The estimated rate of a 90° rotation of this double-decker complex in solution is ∼3 s-1 at room temperature. It could be expected from this value that molecules have been rotated many times before one isotropic image of a single molecule is complete.34 However, it may not be possible to predict the rotation rate of surface adsorbed double-decker complexes from the data in solution, since adsorption to surface may affect the conformation of the porphyrin rings,35 which in turn will influence the inter-ring interactions and, consequently, the rotation rate. Also, the strong electric field and possible redox reactions under the influence of the STM measurement might affect the rotation of the double-decker complexes. It should also be pointed out that inelastically tunneling electrons could (32) Yin, S.; Wang, C.; Qiu, X.; Xu, B.; Bai, C. Surf. Interface Anal. 2001, 32, 248-252. (33) Bailey, W. F.; Monahan, A. S. J. Chem. Educ. 1978, 55, 489-493. (34) Under our typical imaging conditions for a scan size of 50 × 50 nm, it takes about 15 ms for the tip to pass over one 3-nm-sized molecule in one scan, about 4 s to complete the image for a single molecule, and 1 min to complete a whole image. (35) Yokoyama, T.; Yokoyama, S.; Kamikado, T.; Mashiko, S. J. Chem. Phys. 2001, 115, 3814-3818.

Arrays of Double-Decker Porphyrins

provide enough energy to excite relevant vibrational modes36 and induce rotation.

Conclusions and Prospects In conclusion, we have presented here the ordered arrays of double-decker porphyrin/phthalocyanine complexes at the 1-phenyloctane-HOPG interface. The ordered arrays are made possible by attaching long alkyl chains on one of the rings in the doubledecker complexes. The alkylated ring adsorbs onto the HOPG surface with the aid of alkyl-surface and alkyl-alkyl interactions. Then, the other macrocycle becomes the upper ring. As for Ce(Pc)(C22OPP), circular images are recorded in STM. In contrast, the molecules of Ce(C22OPP)2 are imaged as squares. The BPEPP ring in Ce(BPEPP)(C22OPP) is of an oblong shape, which is reflected in the STM images. Thus, the orientation of the individual molecules can be discerned from the STM image. Two patterns differing in the orientation of the BPEPP moiety have been recognized. Competitive adsorption experiments have shown that the affinity toward the HOPG surface of Ce(BPEPP)(C22OPP) is 2.7 kJ mol-1 less than that of H2(C22OPP). Only isotropic images are observed when the molecule of Ce(BPEPP)(C22OPP) is flanked by the free base molecules, H2(C22OPP), as opposed to the double-decker molecule in its own row. This result raises an interesting possibility that isotropic images may be due to the rotation of the upper ring. More clear evidence may be needed to unequivocally prove the rotation of these molecules immobilized on surface. (36) Henzl, J.; Mehlhorn, M.; Gawronski, H.; Rieder, K.-H.; Morgenstern, K. Angew. Chem., Int. Ed. 2006, 45, 603-606.

Langmuir, Vol. 22, No. 13, 2006 5715

Let us conclude the discussion with a few comments on why the array of rotatable molecules is of interest.37 If the rotation of one ring is transmitted through intermolecular interactions to the next ring within the array, or in other words, a concerted rotation is realized, we could exploit the phenomenon for the development of molecular gears or cogwheels. Further, if the array is so constructed that the orientation of a ring is determined by those of surrounding rings, e.g., via electrostatic interactions, this is exactly what is required for the operation of the molecular cellular automatum,38 which has a major advantage as a way to realize a molecular computer, since it can dispense with the need of wiring individual molecules, which is a daunting issue without a foreseeable solution at present. Acknowledgment. This work was supported by the Ministry of Education, Culture, Sports, Science and Technology, Grantin-Aid for Scientific Research (C), 16510090, 2004 and the HighTech Research Center Project for Private Universities. LA0608617 (37) For surface-mounted molecular rotors, see: (a) Gimzewski, J. K.; Joachim, C.; Schlittler, R. R.; Langlais, V.; Tang, H.; Johannsen, I. Science 1998, 281, 531-533. (b) Zheng, X.; Mulcahy, M. E.; Horinek, D.; Galeotti, F.; Magnera, T. F.; Michl, J. J. Am. Chem. Soc. 2004, 126, 4540-4542. (c) Hersam, M. C.; Guisinger, N. P.; Lyding, J. W. Nanotechnol. 2000, 11, 70-76. (d) Clarke, L. I.; Horinek, D.; Kottas, G. S.; Varaksa, N.; Magnera, T. F.; Hinderer, T. P.; Horansky, R. D.; Michl, J.; Price, J. C. Nanotechnol. 2002, 13, 533-540. (e) van Delden, R. A.; ter Wiel, M. K. J.; Pollard, M. M.; Vicario, J.; Koumura, N.; Feringa, B. L. Nature 2005, 437, 1337-1340. (38) (a) Lent, C. S. Science, 2000, 288, 1597-1599. (b) Imre, A.; Csaba, G.; Ji, L.; Orlov, A.; Bernstein, G. H.; Porod, W. Science 2006, 311, 205-208.