Core Size Effects on the Reactivity of Organic Substrates as

Arnold J. Kell, Robert L. Donkers, and Mark S. Workentin*. Department of Chemistry, The University of Western Ontario, London, Ontario, Canada N6A 5B7...
0 downloads 0 Views 190KB Size
Download PDF
Langmuir 2005, 21, 735-742

735

Core Size Effects on the Reactivity of Organic Substrates as Monolayers on Gold Nanoparticles Arnold J. Kell, Robert L. Donkers,† and Mark S. Workentin* Department of Chemistry, The University of Western Ontario, London, Ontario, Canada N6A 5B7 Received April 19, 2004. In Final Form: August 5, 2004

Monolayer-protected nanoparticles (MPNs) with average core sizes of 1.7- (small), 2.2- (medium) and 4.5-nm (large) diameter have been prepared and functionalized with a variety of aryl ketone substrates, namely, 11-mercaptoundecaphenone (1), 1-(4-hexyl-phenyl)-11-mercaptoundecanone (2), 1-[4-(11-mercaptoundecyl)phenyl]hexanone (3), or 1-[4-(11-mercaptoundecyl)phenyl]undecanone (4). Upon irradiation in benzene solution, the aryl ketone-modified MPNs undergo the Norrish type II photoreaction and yield alkene- or acetophenone-modified MPNs exclusively, with no evidence for the generation of cyclobutanol. The extent of the photoreaction for the entire series of aryl ketones is dependent on the size of the MPN core. For 11-mercaptoundecaphenone, the reaction proceeds nearly to completion on the smallest MPN cores (99 ( 1%) but occurs to a much lesser extent on medium (85 ( 5%) and large cores (66 ( 6%). The differences in the extents of reaction are rationalized by the decreased reactivity of substrates on terrace regions, which become increasingly larger with the core size. In lending support to this hypothesis, the edge and vertex sites of medium-sized MPNs were selectively populated with an aryl ketone probe and shown to react quantitatively, whereas selective population of the terrace sites on the same-sized MPNs results in a much lower extent of reaction. Together, these results indicate differences in reactivity of monolayer substrates on terrace versus edge/vertex sites of MPNs. The differences in reactivity with site will play a role in the design of modified MPNs for applications.

Introduction Monolayer-protected nanoparticles (MPNs) are often used to model mobility and reactivity on 2-D selfassembled monolayers (SAMs) because of the ease with which they can be characterized via conventional analytical techniques, such as NMR spectroscopy. In many respects, MPNs are good mimics of SAMs;1,2 for example, alkyl chain dynamics and melting properties of solid-state MPNs are reportedly similar to those of the 2-D analogues,1 as are the solid-state IR stretching frequencies.2,3 However, there are inherent differences between 2-D SAMs and MPNs, such as the flat, uniform surface present on SAM surfaces as opposed to the highly faceted geometry of the MPN surface. Because MPNs are comprised of distinct edge, vertex, and terrace sites,4 changing the size of the MPN will alter the relative proportions of these sites, which may produce differences in monolayer reactivity or other properties as the MPN core is varied in size. Though many groups have explored how the properties of the metallic core change as its size is varied3,5,6 or as assemblies of MPNs are generated,7-9 surprisingly few studies have * Corresponding author. Phone: 519-661-2111 (86319). Fax: 519661-3022. E-mail: [email protected]. † Current address: Steacie Institute for Molecular Sciences, National Research Council of Canada. (1) Badia, A.; Lennox, R. B.; Reven, L. Acc. Chem. Res. 2000, 33, 475. (2) Hostetler, M. J.; Stokes, J. J.; Murray, R. W. Langmuir 1996, 12, 3604. (3) Hostetler, M. J.; Wingate, J. E.; Zhong, C.-J.; Harris, J. E.; Vachet, R. W.; Clark, M. R.; Londono, J. D.; Green, S. J.; Stokes, J. J.; Wignall, G. D.; Glish, G. L.; Porter, M. D.; Evans, N. D.; Murray, R. W. Langmiur 1998, 14, 17. (4) Templeton, A. C.; Wuelfing, W. P.; Murray, R. W. Acc. Chem. Res. 2000, 33, 27. (5) Alvarez, M. M.; Khoury, J. T.; Schaaff, T. G.; Shafigullin, M. N.; Vezmar, I.; Whetten, R. L. J. Phys. Chem. B 1997, 101, 3706. (6) Prasad, B. L. V.; Stoeva, S. I.; Sorensen, C. M.; Klabunde, K. J. Langmuir 2002, 18, 7515.

focused on correlating size-reactivity or -mobility relationships. Landman and co-workers have carried out computational studies concerning size-phase transition relationships for substrates anchored to MPNs of differing sizes,10,11 and more recently, size-photophysical relationships for porphyrin-12 and stilbene-modified MPNs have been studied.13 In a previous study, we carried out photochemical reactions on aryl ketone substrates anchored to monolayerprotected nanoparticles and have shown them to exhibit selective reactivity.14-16 For example, we have employed the Norrish-Yang type II photoreaction17,18 of a series of aryl ketones to probe the monolayer environment of gold MPNs comprised of 2.3 ( 0.6 nm cores. Photochemical excitation of aryl ketones capable of undergoing the Norrish-Yang type II reaction generally result in γ-hydrogen abstraction by the n,π* triplet excited state of the carbonyl group, leading to a biradical intermediate capable of fragmenting to a new ketone and an alkene (via the trans-biradical) or cyclizing to generate cyclobutanol (via the major pathway from the cis-biradical) (Scheme 1). (7) Storhoff, J. J.; Lazarides, A. A.; Mucic, R. C.; Mirkin, C. A.; Letsinger, R. L.; Schatz, G. C. J. Am. Chem. Soc. 2000, 122, 4640. (8) Storhoff, J. J.; Mirkin, C. A. Chem. Rev. 1999, 99, 1849. (9) Jin, R.; Wu, G.; Li, Z.; Mirkin, C. A.; Schatz, G. C. J. Am. Chem. Soc. 2003, 125, 1643. (10) Luedtke, W. D.; Landman, U. J. Phys. Chem. 1996, 100, 13323. (11) Luedtke, W. D.; Landman, U. J. Phys. Chem. B 1998, 102, 6566. (12) Imahori, H.; Kashiwagi, Y.; Hanada, T.; Endo, Y.; Nishimura, Y.; Yamazaki, I.; Fukuzumi, S. J. Mater. Chem. 2003, 13, 2890. (13) Zhang, J.; Whitesell, J. K.; Fox, M. A. J. Phys. Chem. B 2003, 107, 6051. (14) Kell, A. J.; Stringle, D. L. B.; Workentin, M. S. Org. Lett. 2000, 2, 3381. (15) Kell, A. J.; Workentin, M. S. Langmuir 2001, 17, 7355. (16) Kell, A. J.; Montcalm, C. C.; Workentin, M. S. Can. J. Chem. 2003, 81, 484. (17) Wagner, P. J. CRC Handbook of Photochemistry and Photobiology; CRC Press Inc.: Boca Raton, FL, 1995; Chapter 38. (18) Wagner, P. J. Acc. Chem. Res. 1971, 4, 168.

10.1021/la049016j CCC: $30.25 © 2005 American Chemical Society Published on Web 12/17/2004

736

Langmuir, Vol. 21, No. 2, 2005

Kell et al.

Scheme 1. Norrish-Yang Type II Photoreaction of an Aryl Ketone

However, in the MPN environment, irradiation of the aryl ketone probes results in the exclusive generation of an alkene and a new ketone product, suggesting that the biradical equilibrium favors the trans-biradical in the MPN environment. Another intriguing result of this earlier study of aryl ketone photochemistry on MPNs of this size is a general inability for the reactions to reach completion.14-16 The extents of the Norrish-Yang type II reaction for an entire series of aryl ketones were found to be 86 ( 6%.14,15 While numerous groups have employed different reactions or interactions of MPN-bound substrates for a variety of purposes, including the study of photophysical properties,12,13,19-30 the ability to carry out photoisomerization reactions/cycloaddition reactions,13,20,24,27 polymerizations,31-38 nanoparticle aggregation through DNA interaction,7-9 charge-transfer interactions39 and metal ion chelation,40-42 and molecular/ biomolecular recognition,43-46 few reports comment on the extent of reaction (19) Thomas, K. G.; Kamat, P. V. J. Am. Chem. Soc. 2000, 122, 2655. (20) Zhang, J.; Whitesell, J. K.; Fox, M. A. Chem. Mater. 2001, 13, 2323. (21) Wang, T.; Zhang, D.; Xu, W.; Yang, J.; Han, R.; Zhu, D. Langmuir 2002, 18, 1840. (22) Templeton, A. C.; Cliffel, D. E.; Murray, R. W. J. Am. Chem. Soc. 1999, 121, 7081. (23) Maxwell, D. J.; Taylor, J. R.; Nie, S. J. Am. Chem. Soc. 2002, 124, 9606. (24) Manna, A.; Chen, P.-L.; Akiyama, H.; Wei, T.-X.; Tamada, K.; Knoll, W. Chem. Mater. 2003, 15, 20. (25) Imahori, H.; Arimura, M.; Hanada, T.; Nishimura, Y.; Yamazaki, I.; Sakata, Y.; Fukuzumi, S. J. Am. Chem. Soc. 2001, 123, 335. (26) Huang, T.; Murray, R. W. Langmuir 2002, 18, 7077. (27) Hu, J.; Zhang, J.; Liu, F.; Kittredge, K.; Whitesell, J. K.; Fox, M. A. J. Am. Chem. Soc. 2001, 123, 1464. (28) Gu, T.; Whitesell, J. K.; Fox, M. A. Chem. Mater. 2003, 15, 1358. (29) Ipe, B. I.; Thomas, K. G.; Barazzouk, S.; Hotchandani, S.; Kamat, P. V. J. Phys. Chem. B 2002, 106, 18. (30) Kamat, P. V.; Barazzouk, S.; Hotchandani, S. Angew. Chem., Int. Ed. 2002, 41, 2764. (31) Smith, E. A.; Wanat, M. J.; Cheng, Y.; Barreira, S. V. P.; Frutos, A. G.; Corn, R. M. Langmuir 2001, 17, 2502. (32) Buining, P. A.; Humbel, B. M.; Philipse, P. A.; Verkleij, A. J. Langmuir 1997, 13, 3921. (33) Ohno, K.; Koh, K.; Tsujii, Y.; Fukuda, T. Macromolecules 2002, 35, 8989. (34) Wu, M.; O’Neill, S. A.; Brousseau, L. C.; McConnell, W. P.; Shultz, D. A.; Linderman, R. J.; Feldheim, D. L. Chem. Commun. 2000, 775. (35) Jordan, R.; West, N.; Ulman, A.; Chou, Y.-M.; Nukyen, O. Macromolecules 2001, 34, 1606. (36) Nuβ, S.; Bo¨ttcher, H.; Wurm, H.; Hallensleben, M. L. Angew. Chem., Int. Ed. 2001, 40, 4016. (37) Watson, K. J.; Zhu, J.; Nguyen, S.; Mirkin, C. A. J. Am. Chem. Soc. 1999, 121, 462. (38) Itoh, H.; Tahara, A.; Naka, K.; Chujo, Y. Langmuir 2004, 20, 1972. (39) Naka, K.; Itoh, H.; Chujo, Y. Langmuir 2003, 19, 5496. (40) Norston, T. B.; Frankcamp, B. L.; Rotello, V. M. Nano Lett. 2002, 2, 1345. (41) Kim, Y.; Johnson, R. C.; Hupp, J. T. Nano Lett. 2001, 1, 165. (42) Zamborini, F. P.; Leopold, M. C.; Hicks, J. F.; Kulesza, P. J.; Malik, M. A.; Murray, R. W. J. Am. Chem. Soc. 2002, 124, 8958.

or interactions within the MPN environment.14-16,20,47-50 Because the MPN core is well-known to be inhomogeneous (there are edge, vertex, and terrace sites),4 we speculated at the time that aryl ketones positioned directly at edge and vertex sites (defect sites) and possibly those directly adjacent to the edge and vertex are able to undergo the conformationally demanding Norrish type II reaction, but those on the interior of the terrace could not.15 This sparked an interest in elucidating how the size of the MPN core (and the size of the terrace itself) affects the efficiency of the Norrish type II reaction, which is dependent on the mobility and order of the monolayer surrounding an MPN. Herein we use the Norrish type II photoreaction of a series of aryl ketones anchored to increasingly larger MPN cores (Scheme 2) to probe aspects of site reactivity, as it varies with core size. Ultimately we hope to extract information regarding how the overall mobility, orientational order, and reactivity of the monolayer changes as a function of core size. Selective functionalization of the MPN monolayer with the aryl ketone probe at defect sites and defect near-neighbor sites on the Au core allowed for the reactivity on these sites to be compared to that on terrace sites. We show that selective population of specific sites on the MPN core indicates that the photoreaction is siteselective, providing further insight into reactivity on MPNs of various sizes. The results should have wide-ranging implications on the use of MPNs as models for 2-D monolayers systems and allow better design of functionalized MPNs for applications. Results and Discussion In examining the relationship between reactivity/ mobility constraints in the monolayer and the core size of the MPNs, it is important to select and appropriately functionalize MPNs with distinct average core sizes. This task is not trivial because the general procedure employed in MPN synthesis does not generate MPNs with monodisperse core diameters.3 It is possible to prepare a series of MPNs where the core size distribution of each member of the series is narrowly dispersed and has little overlap (43) McIntosh, C. M.; Esposito, E. A., III; Boal, A. K.; Simard, J. M.; Martin, C. T.; Rotello, V. M. J. Am. Chem. Soc. 2001, 123, 7626. (44) Boal, A. K.; Rotello, V. M. J. Am. Chem. Soc. 2002, 124, 5019. (45) Fischer, N. O.; Verma, A.; Goodman, C. M.; Simard, J. M.; Rotello, V. M. J. Am. Chem. Soc. 2003, 125, 13387. (46) Hong, R.; Fischer, N. O.; Verma, A.; Goodman, C. M.; Emrick, T.; Rotello, V. M. J. Am. Chem. Soc. 2004, 126, 739. (47) Koenig, S.; Chechik, V. Langmuir 2003, 19, 9511. (48) Templeton, A. C.; Hostetler, M. J.; Kraft, C. T.; Murray, R. W. J. Am. Chem. Soc. 1998, 120, 1906. (49) Brust, M.; Fink, J.; Bethell, D.; Schiffrin, D. J.; Kiely, C. Chem. Commun. 1995, 1655. (50) Templeton, A. C.; Hostetler, M. J.; Warmoth, E. K.; Chen, S.; Hartshorn, C. M.; Krishnamurthy, V. M.; Forbes, M. D. E.; Murray, R. W. J. Am. Chem. Soc. 1998, 120, 4845.

Core Size and Reactivity of Organic Substrates

Langmuir, Vol. 21, No. 2, 2005 737

Scheme 2. Aryl Ketone-Modified MPNs Employed in the Study of Size-Dependent Reactivity on MPNsa

a In (a), an acetophenone is liberated from the MPN surface, while an alkene remains anchored to the surface. In (b), an alkene is liberated from the MPN surface, while a p-alkyl acetophenone remains anchored to the surface.

Table 1. Stoichiometry and Extent of Reaction for Each of the MPNs Studied MPNa

diameter of MPN core (nm)b

stoichiometryc

final stoichiometryd,e

conversion to product

1-C12MPNsmall 1-C12MPNmedium 1-C12MPNlarge 1(edge)-C12MPN medium 1(terrace)-C12MP Nmedium 2-C12MPNsmall 2-C12MPNlarge 3-C12MPNlarge 4-C12MPNlarge

1.7 ( 0.4 2.2 ( 0.4 4.5 ( 0.7 2.2 ( 0.4 2.2 ( 0.4 1.7 ( 0.4 4.6 ( 0.9 4.2 ( 0.7 4.0 ( 0.8

(C12S)38(1)33Au201c (C12S)45(1)63Au314 c (C12S)37(1)334 Au2950c (C12S)88(1)20Au314 c (C12S)84(1)24Au314 c (C12S)42(2)29Au201 c (C12S)26(2)345 Au2950 c (C12S)82(3)289 Au2950 c (C12S)52(4)319 Au2950 c

(C12S)38(1)1(nonene)32Au201d (C12S)45(1)9(nonene)54Au314 d,e (C12S)37(1)127(nonene)207Au2 950e (C12S)88(nonene)20Au314 d (C12S)88(1)8(nonene)16Au314 d (C12S)42(2)2(nonene)27Au201 d (C12S)26(2)110(nonene)235Au2 950e (C12S)81(3)118(C11acetophenone)170Au2950 e (C12S)52(4)83(C11acetophenone)236Au2950 e

98 ( 2%f 85 ( 5% f 62 ( 6% f 99 ( 1% f 65 ( 5% f 93 ( 1% f 68 ( 4% f 59%g 74 ( 2% f

a For example, 1 place-exchanged onto an original dodecanethiolate MPN with a small core diameter would be represented as 1-C MPN 12 small. Determined via TEM. c Determined via 1H NMR spectroscopy after I2-induced decomposition to generate corresponding mixed disulfides 3,52 d and using the stoichiometries of MPNs prepared in this way presented by Hostetler et al. Determined directly from the integration of characteristic signals in the 1H NMR spectrum of the MPN, as described in the text. e Determined via integration of characteristic signals of the corresponding disulfides in the 1H NMR spectrum after I2-induced decomposition of the MPN. f Determined by comparison of the integration values for the protons for the original aryl ketone with those of the new acetophenone or nonene. These are average values based on the results from irradiation of at least three samples. g Result of a single irradiation.

b

with the core size distributions for the other members of the series. With this in mind, MPNs with three different average core size diameters were prepared by the standard two-phase synthesis developed by Brust51 and modified by Murray.3 We utilize dodecanethiol as the capping thiol, and the gold/dodecanethiol ratio is varied between 1:3, 1:1, and 6:1 to generate a series of MPNs with small, medium, and large cores, respectively. These base MPNs are represented as C12MPNX, where C12 represents dodecanethiolate and X represents the relative size of the MPN core, namely, small, medium, or large, respectively. Because the preparation of the MPNs requires the use of sodium borohydride (and hydride converts ketones to alcohols), the place-exchange reaction52,53 was employed to functionalize the C12MPNX with the probe aryl ketones, namely, 11-mercaptoundecaphenone (1), 1-(4-hexyl-phenyl)-11-mercaptoundecanone (2), 1-[4-(11-mercaptoundecyl)phenyl]hexanone (3), or 1-[4-(11-mercaptoundecyl)phenyl]undecanone (4) (Scheme 2). The synthesis of probes 1-4 was described in a previous report.15 The nomenclature adopted to describe the aryl ketone-modified MPNs is as follows: aryl ketone 1 place-exchanged onto a base (51) Brust, M.; Walker, M.; Bethell, D.; Schiffrin, D. J.; Whyman, R. J. Chem. Commun. 1994, 801. (52) Hostetler, M. J.; Templeton, A. C.; Murray, R. W. Langmuir 1999, 15, 3782. (53) Song, Y.; Murray, R. W. J. Am. Chem. Soc. 2002, 124, 7096.

C12MPN with a small core diameter is represented as 1-C12MPNsmall and likewise for other aryl ketones and core sizes. The stoichiometries of the MPNs were determined from the relative integration values for the characteristic protons of the resulting disulfides formed following I2-induced decomposition of the aryl ketonemodified MPN. In the cases of 2-4, there will be a contribution from both the terminal methyl group of the aryl ketone and the didodecyl disulfide at ∼0.9 ppm, so the contribution from the terminal methyl group on the aryl ketone must be subtracted from the total integration value in order to determine the contribution from the original dodecanethiolate only. It is assumed that the total number of substrates on the MPN surface remains constant prior to and after place-exchange has taken place.52 The sizes of the cores for the various MPNs were measured after the place-exchange reaction by transmission electron microscopy (TEM). Figure 1 shows the histograms for 1-C12MPNX, where 1-C12MPNsmall has an average diameter of 1.7 ( 0.4 nm (a), 1-C12MPNmedium has an average size of 2.2 ( 0.4 nm (b), and 1-C12MPNlarge has an average size of 4.5 ( 0.7 nm (c). There is some overlap in the core sizes for the small- and medium-sized MPNs, but the major populations for 1-C12MPNsmall are core sizes of 1.5 and 1.8 nm with only a minor population of 2.0 nm. For 1-C12MPNmedium, the major population includes core sizes of 2.0 and 2.3 nm, with only minor

738

Langmuir, Vol. 21, No. 2, 2005

Figure 1. The histograms for 1-C12MPNX, where X(core size) ) (a) small (1.7 nm), (b) medium (2.2 nm), and (c) large (4.5 nm).

populations of 1.5, 1.8, and 2.6 nm. These trends in core diameters are visibly evident in the TEM images and the plasmon absorption bands of the particles. The largest MPN in this series, 1-C12MPNlarge, has essentially no size overlap with the medium-size MPN, but does have a much broader distribution of core sizes. The MPNs modified with 2-4 display core size ranges similar to those described above (see Supporting Information), and the stoichiometry and core sizes for the entire series of MPNs studied are provided in Table 1. The sizes of the MPNs are consistent with published reports of similarly prepared MPNs, and they indicate that the place-exchange reaction does not significantly affect the expected size of the MPN cores.3 For example, Fox,28 Rotello,40 and Murray54 have shown that alkylthiolate MPNs undergo place-exchange reactions with long-chain aromatic substrates, resulting in MPNs (54) Aguila, A.; Murray, R. W. Langmuir 2000, 16, 5949.

Kell et al.

with core sizes and deviations (as measured by TEM) similar to those reported here. It should be noted that the stoichiometries in Table 1 represent average compositions for the MPNs. The compositions reported in Table 1 are based on the average sizes of our MPNs determined by TEM and using the results of a study by Murray and coworkers where the synthetic conditions used to prepare MPNs were varied, producing a series of MPNs with many different core sizes.3 The MPNs were well characterized, and the average diameters (by TEM) of the MPNs we prepared were compared to the average sizes determined in their study (by TEM) in order to estimate the average number of gold atoms in the core for 1-4-C12MPNX and the average number of substrates surrounding the core. The shapes of MPN nanoparticles are assumed to be truncated octahedrons, as described by Whetten55 and Murray.3 While this information is provided in Table 1, more important for this study than the actual composition of the MPN is the proportion of 1-4 to C12 in each of the MPNs, since this allows us to better follow the extent of reaction with irradiation. In the present study we are using the Norrish type II photoreaction of a series of aryl ketones (1-4) anchored to MPNs with increasingly larger cores in order to probe how the extent of the Norrish type II reaction is affected by core size. In doing so, 1H NMR spectra of 15-20-mg samples of 1-4-C12MPNX, where X ) small, medium, and large, respectively, dissolved in nitrogen-saturated benzene-d6 were recorded prior to irradiation and at intermittent times during the irradiation. The spectral changes in each case are described below. In general, where the probe is 1 or 2, the broad resonances at 7.89 (protons ortho to the carbonyl on the phenyl ring), 7.10 (meta, and in some cases para, protons on the phenyl ring), and 2.7 ppm (protons R to the carbonyl on the alkyl chain) decrease in intensity as new broad resonances at 5.90 (vinyl proton), 5.10 (vinyl protons), and 2.11 ppm (allyl protons) increase in intensity, indicating that MPN-bound alkene is generated, as depicted in Scheme 2a (Figure 2). Accompanying these changes is the increase in intensity for sharp resonances attributed to acetophenone at 7.88 (protons ortho to the carbonyl on the phenyl ring), 7.10 (meta, and in some cases the para, protons on the phenyl ring), and 2.1 ppm (methyl ketone singlet). Any liberated acetophenone was washed free of the modified MPN partway through the irradiation to prevent competitive absorption, which could slow or prevent efficient photoreaction on the MPN surface. In addition, the liberated acetophenone was also washed away from the MPN when the photoreaction was complete, before the final stoichiometry of the MPN was determined. There was no evidence for the generation of any disulfide during the irradiation, which would be evident from the growing in of a sharp triplet at 2.4 ppm in benzene-d6. Evidence of disulfide would indicate photoinduced liberation of the probe thiolate from the Au particle. In addition, there is no evidence for precipitation or changes in the size of the nanoparticles with irradiation. Probe aryl ketones 3 and 4 were also anchored to C12MPNlarge. These probes reacted similarly to 1 and 2, except that an alkene is liberated to solution and the product para-alkyl acetophenone remains anchored to the MPN surface (Scheme 2b). The spectral changes accompanying irradiation of 3- and 4-C12MPNlarge are provided as Supporting Information. The initial and final 1H NMR spectra for all 1-4-C12MPNX are shown in the Supporting (55) Whetten, R. L.; Khoury, J. T.; Alvarez, M. M.; Murthy, S.; Vezmar, I.; Wang, Z. L.; Stephens, P. W.; Cleveland, C. L.; Luedtke, W. D.; Landman, U. Adv. Mater. 1996, 8, 428.

Core Size and Reactivity of Organic Substrates

Figure 2. The 1H NMR spectra of (a) 1-C12MPNsmall, (b) 1-C12MPNmedium, and (c) 1-C12MPNlarge measured in benzene-d6 (solvent peak indicated with *) (i) prior to irradiation, and (ii) when the Norrish type II reaction was complete. The filled arrows indicate the decreased intensity of the resonances associated with 1, and the hollow arrows indicate the increase in intensity for the resonances associated with nonene-modified MPN. The liberated acetophenone has been washed away from the MPN prior to recording the final 1H NMR spectra in (a), (b), and (c).

Information. No attempt was made to determine the quantum yields of the photoreactions. The conversions for each of the MPNs studied are outlined in Table 1 along with the stoichiometry of the MPNs prior to and after irradiation. Once again, the stoichiometries outlined in Table 1 are based on the average size of the MPN cores as estimated by previously reported work.3 The importance is not the actual stoichiometry, but the percentage of 1-4 capable of undergoing the reaction as the core size is varied. Irradiations were repeated a minimum of three times. There was no evidence for the generation of cyclobutanol via cyclization of the 1,4-biradical intermediate (Scheme 1) by IR or 1H NMR spectroscopy for any of the modified MPNs, regardless of core size. As mentioned above, following purification (that is, washing the MPNs with ethanol to remove liberated acetophenone or alkene), the extents of reaction or conversions for the MPNs were determined by comparison of the integration value for the appropriate resonances from the 1H NMR spectra of either the modified MPN itself (1- and 2-C12MPNsmall

Langmuir, Vol. 21, No. 2, 2005 739

and 1-C12MPNmedium), an I2-promoted decomposition reaction where the substrate is liberated from the surface as a mixed disulfide (1-4-C12MPNlarge), or both. In the case of 1- and 2-C12MPNX, the integration value of the vinyl protons of the product alkene was compared to that of the aromatic resonance of the ortho protons on the phenyl ring of the aryl ketone, whereas for 3- and 4-C12MPNlarge the integration value of the aromatic resonance of the protons ortho to the carbonyl on the phenyl ring for the product acetophenone was compared to that of original aryl ketone. The decomposition reaction was required for 1-4-C12MPNlarge, because the broadness of the 1H NMR spectra makes integration more prone to error for the determination of the final stoichiometry directly from the MPN. As a control, we carried out both the I2-promoted decomposition method and the direct integration of the MPN, and they provide the same estimate. While ligands are known to migrate from the terrace sites to the edge/vertex sites (and visa versa), these migration rates are slow on the time scale of our irradiations carried out at room temperature.52,53,56,57 The results in Table 1 are for samples irradiated until no further change in conversion to products is observed. However, these same samples, when left for several weeks, can be reacted further, suggesting migration of the ligands on this longer time scale (vide infra). Thus, the results we present as extent of reaction represent the reactivity of the MPNs with ligand composition and site on the MPN as they are first prepared. With this caveat, the extent of reaction for 1-C12MPNsmall was nearly quantitative (99 ( 1%), whereas 1-C12MPNmedium had an extent of reaction of 85 ( 5% and 1-C12MPNlarge had a conversion of only 62 ( 6%, indicating that as the core size increases, the reactivity of 1 significantly decreases. This trend is general for all 1-4-C12MPNX studied, as presented in Table 1. The smaller MPN cores, 1-C12MPNsmall and 2-C12MPNsmall, facilitate a more efficient Norrish type II photoreaction (99 ( 1% and 93 ( 1%, respectively) than both 1-C12MPNmedium (85 ( 5%) and 1-4-C12MPNlarge (collectively 66 ( 6%) (Table 1). Interestingly, 2-C12MPNsmall shows an extent of reaction of 93 ( 1%, which is intermediate to that of 1-C12MPNsmall (99 ( 1%) and 1-C12MPNmedium (85 ( 5%) (Table 1). Analysis of the TEM micrographs (the histograms are presented in the Supporting Information) shows that 2-C12MPNsmall has a larger population of 2.0-nm cores (major core size populations at 1.5, 1.8, and 2.0 nm) with respect to 1-C12MPNsmall (major core size populations at 1.5 and 1.8 nm). The larger population of 2.0 nm cores for 2-C12MPNsmall translates to a decrease in reactivity as compared to 1-C12MPNsmall, so it is not surprising that the extent of reaction falls between that of 1-C12MPNsmall and 1-C12MPNmedium. Not only do these data show that the extent of reaction for the Norrish type II reaction is quite sensitive to small changes in the average core size (extent of reaction increases ∼85% to ∼99% as the average core size decreases from 2.2 to 1.7 nm), but overall we see a 30% decrease in the extent of reaction as the average size of the MPN core increases from 1.7 to 4.5 nm. The data also imply that only the MPNs with core sizes in excess of 2.0 nm provide large enough terraces to observe any differences in reactivity, at least for this reaction type. It is worth noting, however, that even in the case of 1- and 2-C12MPNsmall, where the reaction is nearly quantitative, there was no evidence for the generation of cyclobutanol (Scheme 1), indicating that (56) Donkers, R. L.; Song, Y.; Murray, R. W. Langmuir 2004, 20, 4703-4707. (57) Montalti, M.; Prodi, L.; Zaccheroni, N.; Baxter, R.; Teobaldi, G.; Zerbetto, F. Langmuir 2003, 19, 5172-5174.

740

Langmuir, Vol. 21, No. 2, 2005

even when the monolayer is sufficiently mobile to allow the Norrish type II reaction, the trans-biradical is favored, so the MPN environment exerts some control over reactivity in all of 1-4-C12MPNX. It is well understood that the MPN core is inhomogeneous, comprised of edge, vertex, and terrace sites. The relative number of edge and vertex sites as compared with the number of terrace sites is going to depend on the size of the MPN core. Specifically, as the core size increases, so too will the relative number of terrace sites with respect to edge and vertex sites. Because each of the 1-4-C12MPNX is able to react via the Norrish type II reaction but the conversions are different, we hypothesize that the aryl ketones anchored to the edge and vertex sites on all of the MPNs employed in this study will be more mobile and more reactive, whereas those on the terrace will be much less conformationally mobile and unreactive with respect to the Norrish type II reaction. To provide further support for this hypothesis, we manipulated the place-exchange reaction to selectively populate the terrace or edge/vertex sites independently, allowing us to test the reactivity at the sites individually. A model for the place-exchange reaction described by Murray suggests that there is a fast exchange between the incoming thiol and the thiol anchored to the edge and vertex sites.52,53,56,57 This means that the edge and vertex sites can be populated very quickly, with the extent of exchange depending roughly on the statistical amount of each thiol in solution (that is, if there is a large excess of incoming thiol, essentially all of the edge and vertex sites will be populated with incoming thiol). Further monolayer exchange requires migration of ligands to defect sites from the neighboring terrace. Migration from terrace sites is slower than ligand exchange and becomes increasingly more difficult as the proximity from the edge/vertex site is increased. This model suggests that we can employ rather dilute solutions of incoming ligand (1 for example) and short (∼1 h) reaction times in order to populate the edge and vertex (defect) sites predominantly, whereas much longer reaction times and high concentrations of incoming ligand can be used to populate all possible exchangeable sites on the MPN surface. Indeed, the results presented below provide support for this mechanism. We prepared MPNs functionalized with 1 predominantly at the edge and vertex sites by stirring 1 with C12MPNmedium where the ratio of 1 to dodecanethiolate is 1:1.3 (assuming that the MPN is made up of ∼25 wt % dodecanethiolate)3 in toluene for 45 min. The mixture was then concentrated and washed free of 1 and dodecanethiol with warm 95% ethanol. The resulting MPN is referred to as 1(edge)-C12MPNmedium, because the edge and vertex sites are populated predominantly with 1. In a separate reaction, a 5 times excess of 1 was added to C12MPNmedium and the resulting solution was stirred for 5 days, at which time the MPN was concentrated and purified by washing with warm 95% ethanol. Half of this 1-C12MPNmedium (which contains 1 distributed over edge, vertex, and neighboring terrace sites) was then subjected to a short place-exchange reaction with an 8 times excess of dodecanethiol, resulting in the displacement of 1 at the edge and vertex sites, trapping any remaining 1 predominantly at the near-defect terrace sites. This MPN is referred to as 1(terrace)-C12MPNmedium. Each of these MPNs (∼15 mg) was irradiated in nitrogen-saturated benzene-d6 solution, and the same spectral changes described earlier were observed for this series of MPNs (Figure 3). As described above, the extents of reaction were determined after prolonged irradiation generated no further changes in the 1H NMR spectrum. Interestingly, when 1 is anchored

Kell et al.

Figure 3. The 1H NMR spectra of (a) 1(full)-C12MPNmedium, (b) 1(edge)-C12MPNmedium, and (c) 1(terrace)-C12MPNmedium measured in benzene-d6 (solvent peak indicated with *) (i) prior to irradiation, and (ii) when the Norrish type II reaction was complete. The filled arrows indicate the decreased intensity of the resonances associated with 1, and the hollow arrows indicate the increase in intensity for the resonances associated with nonene-modified MPN. In (a), (b), and (c), the liberated acetophenone was washed away from the MPN before the final 1H NMR spectra were recorded.

predominantly at the edge and vertex sites (1(edge)-C12MPNmedium), the photoreaction proceeds quantitatively (99 ( 1%), similar to the 1-C12MPNsmall, whereas when these edge/vertex sites are exchanged with the unreactive ligand, (1(terrace)-C12MPNmedium), irradiation results in a much lower extent of reaction (65 ( 5%), as was observed with 1-C12MPNlarge (Table 1). The data for the 1-C12MPNmedium shows clearly that the extent of reaction for the Norrish type II reaction on MPNs is dependent on the binding site on the core. When 1 is distributed over all exchangeable MPN sites (1-C12MPNmedium), the photoreaction proceeds to 85 ( 5% conversion. Population of the edge and vertex sites predominantly (1(edge)-C12MPNmedium), results in extents of reaction much higher (99 ( 1%). When the edge/vertex sites are populated with an unreactive ligand, (1(terrace)C12MPNmedium), the result is much less efficient photoreactions (65 ( 5%). In fact, similar numbers of unreacted

Core Size and Reactivity of Organic Substrates

1 remain on the MPN when all exchangeable sites on the MPN are populated, compared to when the near-defect terrace and terrace sites are populated predominantly. These results are consistent with the idea that the photoreaction proceeds much more efficiently at the edge and vertex sites than at terrace sites and may react in an intermediate manner in regions between these two types of sites. This may be the result of mobility constraints imposed on terrace-bound substrates, which affect conformational reaction dynamics or excited state/biradical quenching. As pointed out in the description of the place-exchange reaction, the substrates surrounding MPN cores are capable of migrating on the surface. This was a concern to us, particularly if 1-4 could migrate from unreactive sites (the terrace) to more reactive sites (edge and vertex sites) during the course of the irradiation. In general, the extent of reaction for 1(terrace)-C12MPNmedium was dependent on the length of time the MPN was allowed to stand before irradiation. When the MPN was irradiated until no further changes were detected and then left to stand as a solid film for 1 week, re-irradiation increased the extent of reaction by ∼5%. In addition, when 1(terrace)-C12MPNmedium was left in solution for 2 weeks prior to irradiation, it behaves much like 1-C12MPNmedium where its extent of reaction was 82 ( 4%. This suggests that migration does occur on the MPN surface, but the migration rate is slow enough that, provided the MPN is used immediately after preparation, we see the effect of 1 residing predominantly at the terrace for the duration of the irradiation. The extent of reaction for (1(terrace)C12MPNmedium) shows that even near-defect terrace sites still exhibit reactivity. A gradual decrease in the photoreactivity with the probe increasing distance away from an edge or vertex may account for this. It is also possible that migration at the closest defect distances occurs significantly enough to account for the changes in the extent of reaction. Increasing the core diameter decreases the surface curvature at the edge between two terraces, resulting in reduced free space to allow the photoprobe to reach its reaction orientation. This is also consistent with the MPN becoming more 2-D as the diameter increases, and in the present case, that results in defect sites becoming more terracelike. Any one or a combination of them is responsible for the observed photoreactivity. When the irradiations of 1-4-C12MPNlarge were carried out at elevated temperatures (40 ( 5 °C), the extent of reaction was quantitative. We propose that the lowered extents of reaction at 23 °C are either related to the increased ordering of the substrates on the terrace or to a slower rate of migration between the terrace and the edge and vertex sites. For solid-state samples of alkylthiolate MPNs, there are phase transitions where the alkyl chains go from a highly ordered crystalline state to a more disordered liquidlike state. In general, this transition occurs between 3 °C (for C12MPN) and 51 °C (for C18), depending on the length of the alkyl chain.1 When the photoreaction is carried out at these elevated temperatures, there is presumably a similar phase transition whereby the substrates would become more disordered and conformationally mobile, allowing them to undergo the Norrish type II photoreaction more efficiently. There is also the possibility that the elevated temperature increases the rate of migration from the unreactive terrace sites to the more reactive edge and vertex sites; however, no studies have been carried out correlating temperature with the rate of migration. There is likely some contribution from both of these factors, and we are currently

Langmuir, Vol. 21, No. 2, 2005 741

exploring the effect of temperature on reactivity and the rate of migration on MPN surfaces. Conclusions We have utilized triplet-state Norrish type II reactions of 1-4-C12MPNX to probe the effects that core size has on reactivity. The results establish that as the core size of MPNs get larger, the extent of the probe reaction decreases significantly. This trend in reactivity is attributed to the defect regions of MPNs becoming more like their 2-D SAM counterparts as the MPN core diameter increases. For smaller cores, with a larger surface curvature, the increased free space allows the photoreactive probe to achieve a more favorable reactive orientation compared than on the terrace (nondefect) site where it is constrained by other monolayer ligands. We postulate that the ligand properties undergo a transition from defect to nondefect on the MPN surface where the most efficient photoreaction occurs for ligands of closest proximity to the defect sites. This trend was also evidenced by the selective place-exchange study where there was a lower extent of reaction for MPNs with 1 positioned on all exchangeable sites on the MPN in comparison to 1 occupying edge and vertex sites only. Most other studies concerning conformational mobility or dynamics of substrates surrounding MPNs have been carried out in the solid state. Our studies here show that, even in solution, there exist core size (site selective) effects on reactivity. It is also of interest that the onset of mobility constraints preventing the Norrish type II reaction from reaching completion can be related to significant populations of MPN cores greater than 2.0 nm in diameter (this is the minimum size where there is a large enough terrace to start to exhibit effects on the reaction). The probe reaction is subject to different constraints as the attaching site on the MPN core undergoes a defect-to-nondefect transition. The decreased reactivity in the present case can be due to the probe becoming increasingly less mobile as the ligand moves toward the interior of the monolayer, which may affect the ability to achieve the required conformation for reaction, the dynamics of quenching of excited states, or the dynamics of the resulting intermediates. We do not wish to imply that reactivity will always be depressed for substrates on terrace sites, because it is possible that there are cases where reactivity is increased as a result of the limited mobility. For example, Fox and co-workers reported that the extent of reaction for the [2 + 2] photocycloaddition of stilbene-modified MPNs with 2.5-nm cores is ∼15%,20 close to the percentage of aryl ketone that does not react on the MPN we employed with a similar diameter. Extending that study, the authors more recently reported that the emission intensity of a stilbene moiety increases on increasingly larger MPN cores, which they attribute to fewer stilbene substrates being at edge and vertex sites where there is more efficient quenching from the bending motions of the substrates that allow them to more closely approach the surface, resulting in more efficient fluorescence quenching.13 We are currently probing other reaction types that may exemplify this effect. Generally speaking, our results suggest that, when in solution, an MPN will only become a decent mimic of 2-D SAMs, at least with respect to mobility and reactivity, when the terrace becomes extremely large, that is, when the proportion of terrace sites greatly outnumbers the edge and vertex sites. Acknowledgment. We thank the Natural Sciences and Engineering Research Council of Canada (NSERC), the Canadian Foundation of Innovation (CFI), the Ontario Research and Development Challenge Fund (ORDCF),

742

Langmuir, Vol. 21, No. 2, 2005

and The University of Western Ontario (ADF) for supporting this research. A.J.K. thanks NSERC Canada for a PGSB and the Province of Ontario for an OGS. M.S.W. thanks the Province of Ontario for a PREA. Supporting Information Available: The experimental conditions for the synthesis for 1-4 and all of the MPNs used

Kell et al. in this study, as well as 1H NMR spectra for the irradiated MPNs and the decomposed MPNs and TEM images and histograms for each 1-4-C12MPNX, where X ) small, medium, and large, are available free of charge via the Internet at http://pubs.acs.org.

LA049016J