Large Footprint Pyrene Chromophores Anchored to Planar and

May 1, 2009 - Absorption and fluorescence studies, and a comparison with a pyrene−rigid rod model compound, suggest that the new tripodal anchor gro...
2 downloads 10 Views 5MB Size
Download PDF
pubs.acs.org/Langmuir © 2009 American Chemical Society

Large Footprint Pyrene Chromophores Anchored to Planar and Colloidal Metal Oxide Thin Films Sujatha Thyagarajan and Elena Galoppini* Chemistry Department, Rutgers University, 73 Warren Street, Newark, New Jersey 07102

Petter Persson Department of Chemical Physics, Lund University, Box 124, SE-221 00 Lund, Sweden

Jovan M. Giaimuccio and Gerald J. Meyer Department of Chemistry and Department of Materials Science and Engineering, Johns Hopkins University, 3400 North Charles Street, Baltimore, Maryland 21218 Received March 3, 2009. Revised Manuscript Received April 11, 2009 Sensitization and binding of a large footprint pyrene chromophore to planar (sapphire) and colloidal metal oxide films (TiO2 and ZrO2) is investigated. The model compound combines a 1-pyrenyl-ethynylenephenylene unit with a new adamantane-tripodal linker that binds to the surface. The linker design, combining a large footprint (∼2 nm2) of the tripodal linker with the meta position of the COOH anchoring groups, was suggested from atomistic models, and it aims to provide improved spacing control. The pyrene chromophore unit provides a probe of sensitizer-sensitizer interactions through its propensity to form excimers, unless neighboring pyrene units are sufficiently spaced (g3.5 A˚). Absorption and fluorescence studies, and a comparison with a pyrene-rigid rod model compound, suggest that the new tripodal anchor group allows spacing control on planar surfaces. On colloidal films, the linker provides spacing control at low surface coverage but sensitizer-sensitizer interactions are still observed on colloidal films at high surface coverage. Implications for the functionalization of metal oxide films in hybrid molecule-metal oxide semiconductor material systems are discussed.

Introduction Molecular functionalization of metal oxide (MOn) films is of general interest for the development of hybrid molecule-metal oxide materials where examples of current interest include nanostructured photovoltaic devices such as dye-sensitized solar cells1,2 and photocatalysts,2,3 and hybrid bioinorganic materials such as nanostructured DNA-MOn heterosupramolecular assemblies for sensing applications.4 The molecules are often designed to have anchor groups for the covalent binding to the surface of the semiconductor. Carboxylic acid anchor groups (COOH) are frequently used to firmly graft organic dyes or metal complexes on nanostructured metal oxide thin films. There are, however, good reasons to develop new design of anchor groups that allow structural control of the dye/MOn interface.5 Redox and photoactive molecules interfaced with semiconductor or metal nanoparticles are used as components in electronic devices, and there is a great interest in controlling the nature of the attachment of such components, their spacing, distance, and orientation to the :: :: (1) (a) O’Regan, B.; Gratzel, M. Nature 1991, 335, 737. (b) Gratzel, M. Inorg. Chem. 2005, 44, 6841. (2) (a) Kamat, P. J. Phys. Chem. C 2007, 111, 2834. (b) Ardo, S.; Meyer, G. J. Chem. Soc. Rev. 2009, 38, 115. (3) (a) Fujishima, A.; Honda, K. Nature 1972, 238, 37. (b) Kato, H.; Asakura, K.; Kudo, A. J. Am. Chem. Soc. 2003, 125, 3082. (c) Hoertz, P. G.; Mallouk, T. E. Inorg. Chem. 2005, 44, 6828. (4) (a) Taratula, O.; Galoppini, E.; Mendelsohn, R.; Reyes, P. I.; Zhang, Z.; Duan, Z.; Zhong, J.; Lu, Y. Langmuir 2009, 25, 2107. (b) Dorfman, A.; Kumar, N.; Hahm, J.-I. Langmuir 2006, 22, 4890. (5) Galoppini, E. Coord. Chem. Rev. 2004, 248, 1283. (6) Adams, D. M..; et al. J. Phys. Chem. B. 2003, 107, 6668(see the Supporting Information for complete reference).

Langmuir 2009, 25(16), 9219–9226

surface.6 Improved control of the molecular layer is desirable from a practical standpoint, both for anticipated improvements in devices performance and, more fundamentally, for the ability to perform control experiments to investigate interfacial properties. For example, the development of dyes that are bound via rigid linkers is important for studies of charge transfer processes (sensitization) at the interface of wide band gap metal oxides,5 particularly TiO2 and ZnO, which find applications in dye sensitized solar cells and other devices. Organic aromatic hydrocarbons such as perylene,7 azulene,8 anthracene,9 and pyrene,10 attached through rigid linkers, are useful models to study the dependence of the mechanism of injection into TiO2 surfaces from the semiconductor-chromophore distance. The use of efficient anchor groups also offers significant prospects for simultaneous coadsorption of different molecular species, something that (7) (a) Gundlach, L.; Ernstorfer, R.; Willig, F. J. Phys. Chem. C 2007, 111, 13586. (b) Persson, P.; Lundqvist, M. J.; Ernstorfer, R.; Goddard, W. A.; Willig, F. J. Chem. Theory Comput. 2006, 2, 441. (c) Ernstorfer, R.; Gundlach, L.; Felber, S.; Storck, W.; Eichberger, R.; Willig, F. J. Phys. Chem. B 2006, 110, 25383. (d) Szarko, J. M.; Neubauer, A.; Bartelt, A.; Socaciu-Siebert, L.; Birkner, F.; Schwarzburg, K.; Hannappel, T.; Eichberger, R. J. Phys. Chem. C 2008, 112, 10542. (8) Lamberto, M.; Pagba, C.; Piotrowiak, P.; Galoppini, E. Tetrahedron Lett. 2005, 46, 4895. (9) (a) Kamat, P. V. Langmuir 1990, 6, 512. (b) Martini, I.; H. Hodak, J.; Hartland, G. V. J. Phys. Chem. B 1999, 103, 9104. (c) Giaimo, J. M.; Rowley, J. G.; Meyer, G. J.; Wang, D.; Galoppini, E. Chem. Phys. 2007, 339, 146. (10) (a) Matsunaga, Y.; Takechi, K.; Akasaka, T.; Ramesh, A. R.; James, P. V.; Thomas, K. G.; Kamat, P. V. J. Phys. Chem. B 2008, 112, 14539. (b) Taratula, O.; Rochford, J.; Piotrowiak, P.; Galoppini, E.; Carlisle, R. A.; Meyer, G. J. J. Phys. Chem. B. 2006, 110, 15734. (c) Hoertz, P. G.; Carlisle, R.; Meyer, G. J.; Wang, D.; Piotrowiak, P.; Galoppini, E. Nano Lett. 2003, 3, 325. (d) Thomas, K. G.; Kamat, P. V. Acc. Chem. Res. 2003, 36, 888.

Published on Web 05/01/2009

DOI: 10.1021/la9007679

9219

Article

normally would be problematic due to competitive binding replacement for species with different surface binding affinities. We have previously reported the study of pyrene chromophores bound to the semiconductor surface through a series of rigid-rod linkers made of an oligophenylenethynylene (OPE)11 bridge varying in the number of units and terminating with an isophthalic acid (Ipa) group to anchor to the semiconductor surface.10b,10c The OPE spacers induced a red shift in the absorbance of the pyrene unit and an increase in the extinction coefficient, and, as a result, pyrene modified with OPE-Ipa linkers10 (termed “pyrene-Ipa-rods”, such as 3 in Figure 1) were used to make efficient working TiO2 solar cells.10c In that study, we observed pyrene excimer emission on films prepared from pyrene-Ipa-rods such as 3 bound to ZrO2 nanoparticles which was dependent on the surface coverage.10c Zirconium oxide was selected to study the fluorescence spectra because, while the morphology of ZrO2 films is similar to that of TiO2, the much wider band gap (Ebg(ZrO2) ∼ 5.0 eV, Ecb ∼ -1.78 V vs NHE)12 ensures that the pyrene excited state, with a redox potential ∼ -1.5 V vs NHE, is not quenched. For this reason ZrO2 films are useful substrates to study the excited state of dyes bound to semiconductor nanoparticles. The observation of excimer suggested that, at typical surface coverages used for the solar cells preparation (∼2  10-8 mol/cm2), the dye units were closely packed on the surface and that the excimer, rather than pyrene monomer, could be the sensitizing species. Hence, the need to take advantage of the propensity of pyrene model compounds to form excimers in order to study the effect of sensitizer coverage and film morphology on the binding of the dye. In an attempt to also separate the dye units on the semiconductor surface and study the pyrene monomeric species, we synthesized model dye 1, consisting of a 1-pyrenyl-ethynylenephenylene moiety attached to a 1,3,5,7-adamantyl core with three anchoring groups (Figure 1).13 The design was based on atomistic models pertaining to interfacial structure matching considerations. First, 1 has the COOH functional groups in the meta position of the phenyl ring (Figure 1) rather than in para as was the case with the early tripodal model compounds that we had previously prepared.14 Second, compared to tripodal linkers prepared in the past, the footprint spanned by the elongated OPE “feet” is also much larger (∼200 A˚2). The footprint was designed to be large enough to prevent the formation of pyrene aggregates on the surface and also to study the influence of footprint size on injection rates, a study now in progress. The large footprint tripodal sensitizer 1 is thus designed to allow twodimensional surface control in which lateral (sensitizer-sensitizer) distance control is achieved by controlling the footprint size, and perpendicular (sensitizer-surface) distance control is achieved by controlling the length of the chromophore-tripodal anchor rigid rod length. The synthesis of 1 and the IR and absorption spectra in solution and bound to TiO2 surfaces were recently reported.13 In this paper, we report a study of the excited state properties of 1 on ZrO2 films and on planar sapphire surfaces in order to provide insight into the two-dimensional structural and electronic control provided by the new large footprint sensitizer. Tetracarboxylic acid 2, having the same anchoring group and similar structure but without the chromophoric unit, was used as a cobinder, and (11) Sudeep, P. K.; James, P. V.; Thomas, K. G.; Kamat, P. V. J. Phys. Chem. A 2006, 110, 5642. :: (12) Kalyanasundaraman, K; Gratzel, M Coord. Chem. Rev. 1998, 177, 347. (13) Thyagarajan, S.; Liu, A.; Famoyin, O. A.; Lamberto, M.; Galoppini, E. Tetrahedron Lett. 2007, 63, 7550. (14) Wei, Q.; Galoppini, E. Tetrahedron 2004, 60, 8497.

9220 DOI: 10.1021/la9007679

Thyagarajan et al.

Figure 1. Structure of 1, 2, and 3. The footprint of 1 spans an area of about 2 nm2.

pyrene-Ipa-rod (3),10b,10c which closely packs on the surface of MOn nanoparticle films,10c was used as a reference compound.

Experimental Section Materials. Acetonitrile (Burdick and Jackson, biosynthetic grade), ethanol (Warner-Graham, 200 proof), methanol (Fisher, HPLC grade), chloroform (Fisher, ACS grade and HPLC grade), zirconium(IV) n-propoxide (70 wt % in 1-propanol, Alfa-Aesar), nitric acid (Fisher, ACS grade), and Carbowax (15 000-20 000 Da from Sigma) were all used as received. Anatase TiO2 films (∼1.0  2.5 cm2) were prepared as previously described.15 The synthesis of 1 has been previously reported.13 The synthesis of 3 along with binding studies to TiO2 and ZrO have been reported.10c The tetracarboxylic acid 2 was prepared by basic hydrolysis of the corresponding methyl ester, which had been isolated as a byproduct in the synthesis of 1.13 1,3,5,7-(3-Carboxyphenyl-4-ethynylphenyl)adamantane (2): Mp 280-287 °C. 1H NMR (THF) δ 8.82 (4H, s), 8.43 (8H, d, J = 7.5 Hz), 7.86 (8H, d, J = 7.5 Hz), 7.65 (4H, d, J = 8.5 Hz), 7.60 (4H, d, J = 8.5 Hz), 7.45 (4H, d, J = 8.5 Hz), 2.23 (12H, s). 13C NMR (THF) δ 167.3, 149.8, 135.7, 132.8, 131.2, 130.4, 129.1, 128.6, 125.1, 123.6, 120.8, 90.3, 88.6, 46.9, 39.5. IR (cm-1): 3035, 2927, 2210, 1695, 1500, 1249. HRMS (FAB) calcd for C70H48O8 (MH+), 1017.3427; found, 1017.3435. UV-Vis Absorbance and Fluorescence. UV-vis absorption spectra were collected at room temperature on a Varian Cary 50 spectrophotometer or on a VARIAN Cary-500 UV-vis spectrophotometer in a 1 cm square quartz cuvette. The spectra of the thin films were collected with the sensitized film oriented toward the excitation light. Fluorescence spectra were collected on a Spex Fluorolog instrument or on a VARIAN Cary-Eclipse instrument. Each film was placed in the cuvette at a 45° angle to both the excitation light and the detector. FT-IR-ATR. Fourier-transform infrared attenuated total reflectance (FT-IR-ATR) measurements (see the Supporting Information) were collected on a Thermo Electron Corporation Nicolet 6700 FT-IR spectrometer using a ZnSe crystal. ZrO2 Film Preparation. The ZrO2 nanoparticle films were prepared following a published method15 or the improved16 procedure described here. We observed that the results were independent from the preparation method. Under ambient conditions, the solution of zirconium n-propoxide (12.5 mL) was added dropwise over a period of 20-30 min to a vigorously stirred (15) Rochford, J.; Chu, D.; Hagfeldt, A.; Galoppini, E. J. Am. Chem. Soc. 2007, 129, 4655. (16) We found that ZrO2 films tend to peel or crack more easily than the TiO2 films, and the procedure reported here ensured films of excellent quality and reproducibility.

Langmuir 2009, 25(16), 9219–9226

Thyagarajan et al. solution of 0.52 mL of 70% nitric acid and 75 mL of deionized water. The solvent was evaporated as the stirred solution was heated to approximately 95 °C. The volume was reduced to ∼17 mL over 4-6 h, and water was added when the volume decreased below 17 mL. The solution was cooled to room temperature and transferred into a Teflon cup, which was placed in a Parr 4749 autoclave17 and heated at 200 °C for 12 h. After the autoclave returned to room temperature, Carbowax (avg mol wt 2,000) (1 g) was added to the ZrO2 aqueous colloid and the mixture was stirred vigorously until a homogeneous white and viscous paste was obtained. The ZrO2 paste was applied via the “doctor blade” method (casting the paste on the glass with a blade or a glass rod) using Scotch tape as a spacer to glass cover slides (VWR) to form films with an area of ∼2 cm2. The films were left to dry for ∼1 h in air at ambient temperature and then heated to 420 °C under an oxygen atmosphere for 30 min and slowly allowed to cool to room temperature. Prior to binding, the films were “acid treated” by immersion in a pH 1 aqueous solution of HNO3 or “base treated” by immersion in pH 11 KOH aqueous solution. In both cases, the films were immersed for 5 min in acid (or base) and then rinsed with acetonitrile and allowed to dry. Films that were used without acid or base treatment are identified as untreated in the text. The SEM image of a typical ZrO2 film obtained by this method is shown in Figure 2. Binding. The binding of 1 on TiO2 from THF solutions has been reported before.13 It was observed, however, that, depending on the quality of THF,18 1 degraded within hours to a few days. In this paper, we describe binding experiments from MeOH/CHCl3 solutions, which exhibited no sign of degradation over days (UV-vis, see the Supporting Information Figure S-2). In a typical procedure, the films were immersed into a 0.5 mM solution of 1 for at least 18 h and then rinsed with a small amount of neat solvent. As a precaution, the binding experiments were done in the dark.

Binding Method A (Immersion for 5 days in Diluted Solutions). The ZrO2 films were placed in 3 mL solutions of 1

with concentrations varying from 43 to 0.22 μM in 10% MeOH/ CHCl3 and allowed to bind for up to 5 days. The films were then rinsed by immersion in a 25 mL solution of CHCl3.

Binding Method B (Variable Time from a Concentrated Solution). Base treated ZrO2 films were immersed in a 50 μM solution of 1 in MeOH/CHCl3 (10/90) for times that ranged from 10 s to 20 min and then rinsed by immersing the slide in a neat solution of CHCl3. Binding Method C (Cobinding of 1 and 2). Base treated ZrO2 films were placed in dye solutions containing both 1 and the cobinder tetraacid 2 in MeOH/CHCl3 (10/90). After being immersed for 18 h, the films were removed from the dye solutions, rinsed with neat CHCl3, and allowed to dry.

Binding Method D (Sequential Binding; Binding of 2 Followed by 1). Base treated ZrO2 films were placed in cobinder tetraacid 2 in MeOH/CHCl3 (10/90) solution for 18 h, rinsed with CHCl3, then placed in sensitizer 1 in MeOH/CHCl3 (10/90) solution for 18 h, rinsed with CHCl3, and allowed to dry. Base treated ZrO2 films were placed in spacer 2 in MeOH/CHCl3 (10/90) solution for 18 h, rinsed with CHCl3, then placed in sensitizer 1 in MeOH/CHCl3 (10/90) solution for 18 h, rinsed with CHCl3, and allowed to dry in air. Binding to Planar Sapphire. Sapphire optical substrates were placed in 5 mL of 50 μM solutions of 1 and 3 in EtOH/ CHCl3 (2/98) for 18 h and then rinsed by immersion in 25 mL neat CHCl3. The substrates were dried by touching an edge with a Kimwipe. UV-vis absorption and fluorescence experiments were performed under nitrogen atmosphere.

(17) The preparation of TiO2 in a custom-made Ti autoclave (Parr) led to improved colloids, as corrosion of steel autoclaves is often observed. (18) Slower degradation was observed when the solutions were prepared from anhydrous THF (Aldrich) which was freshly distilled from sodium-benzophenone.

Langmuir 2009, 25(16), 9219–9226

Article

Figure 2. SEM image of a typical colloidal ZrO2 thin film utilized in this study.

Results Atomistic modeling considerations are useful for investigating the anchoring of dye molecules to semiconductor substrates, for example, in dye-sensitized solar cells.19 Modeling is particularly valuable for assessing the surface binding capabilities of molecules with more than one anchor group where there is typically a tradeoff between strong surface binding at several points of attachment on one hand and conformational strain on the other hand. The usefulness of simple modeling based on adsorbatesubstrate structural matching considerations has, for example, been demonstrated for Ru-dyes with several anchor groups on anatase TiO2 by Shklover et al.20 Advanced quantum-chemical calculations and molecular dynamics simulations are also gradually emerging as a valuable tool for atomistic studies of dyesemiconductor interfaces for accurate atomistic modeling,19,21 but such approaches remain computationally demanding for such large-footprint adsorbates as tripodal sensitizers. We recently investigated optoelectronic properties of ruthenium tris-bipyridine dyes containing oligophenyleneethynylene rigid rod linkers using first principles calculations.22 As a first step in the predictive modeling of tripodal sensitizers binding to metal oxide semiconductor surfaces, we here visualize simple structural matching models also of tripodal sensitizers on an unrelaxed TiO2 anatase (101) surface.23 The aim of this simple modeling is restricted to guide the experimental efforts to develop better tripods by identifying design strategies that favor binding to the substrate via all three anchor groups. Detailed computations of adsorption energies of specific binding modes for large-footprint tripods, for example, using first principles quantum chemical calculations, remains an ambitious goal beyond the scope of the present investigation. :: :: (19) (a) Persson, P.; Bergstrom, R.; Ojamae, L.; Lunell, S. Adv. Quantum Chem. 2002, 41, 203. (b) Persson, P.; Lundqvist, M. J. J. Phys. Chem. B 2005, 109, 11918. (c) Persson, P.; Lundqvist, M. J.; Nilsing, M.; van Duin, A. C. T.; Goddard, W. A., III. In Proceedings of SPIE (The International Society for Optical Engineering): Physical Chemistry of Interfaces and Nanomaterials V; Spitler, M., Willig, F., Eds.; 2006; Vol. 6325, 63250P. (20) Shklover, V.; Ovchinnikov, Yu. E.; Braginsky, L. S.; Zakeeruddin, S. M.; :: Gratzel, M. Chem. Mater. 1998, 10, 2533. (21) Duncan, W. R.; Prezhdo, O. V. Annu. Rev. Phys. Chem. 2007, 58, 143. (22) Lundqvist, M. J.; Galoppini, E.; Meyer, G. J.; Persson, P. J. Phys. Chem. A 2007, 111, 1487. (23) Visualization using GaussView, version 3.0. Dennington, R., II; Keith, T.; Millam, J.; Eppinnett, K.; Hovell, W. L.; Gilliland, R. GaussView, version 3.0; Semichem, Inc.: Shawnee Mission, KS, 2003.

DOI: 10.1021/la9007679

9221

Article

Thyagarajan et al.

Figure 3. Models of pyrene tripodal compounds anchored to a TiO2 anatase (101) surface through COOH groups that are in para (a) and meta (b) positions on the benzene rings of the footprint, illustrating the effect on the molecule and anchor group orientation relative to the surface. (c) Binding model of tripod 1 on a TiO2 anatase (101) surface.

A structural model of a pyrene attached to the type of tripodal linker previously described14 shown in Figure 3a indicates that at least one of the three para-COOH anchor groups is likely to have an unfavorable angle of attachment relative to the surface. The model moreover suggests that poor bite angles of the anchor groups relative to the surface are likely to be a more significant problem for the successful design of tripods compared to the lattice matching of the anchor groups, and this problem is likely to persist also for other para-COOH tripods with longer rigid legs. Modeling of alternative tripods instead suggests that metaCOOH tripods, exemplified in Figure 3b, are likely to be more suitable for surface attachment with all three anchor groups favorably oriented relative to the substrate. The possibilities for favorable anchor group orientations with meta-COOH tripods is seen to persist also for the large footprint tripod 1 as shown in Figure 3c. In addition, model 1 maintains a perpendicular orientation with respect to the surface. The FT-IR-ATR spectra of 1 bound to acid and base treated TiO2 films are in accordance with the published data,13 and suggest that 1 binds to the TiO2 films via carboxylate binding mode (the type of binding modeled in Figure 3c), regardless of the surface treatment. Spectra obtained on ZrO2 (Supporting Information Figure S-3) were very similar to those obtained on TiO2 and showed the disappearance of the CdO band and the appearance of broad bands in the carboxylate region. Absorption and Fluorescence Spectra in Solution. The solution absorption spectrum of 1 in MeOH/CHCl3 (10/90), shown in Figure 4, was independent of the concentration and did not significantly differ from those reported previously in THF,13 with the typical pyrene vibronic structure and the π-π* transitions below the 300 nm region being assigned to the phenyleneethynylene units. Tripod 1 exhibits almost quantitative fluorescence quantum yield (Φ ∼ 0.9 in THF).13 The solution fluorescence spectrum of 1 in EtOH/CHCl3 (10/90) was moderately concentration dependent. The fluorescence spectrum of a 2.5 μM concentration of 1 had a maximum at 394 nm with a less intense band at 418 nm. At increasing concentrations, the band at 394 nm shifted to 398 nm. Additionally, the intensity of the 418 nm band increased relative to the 398 nm band. A small Stokes shift was observed for 1. This small Stokes shift can account for the concentration dependence of the fluorescence spectra as the higher energy emission is 9222 DOI: 10.1021/la9007679

Figure 4. Solution absorption (a) and fluorescence (b) spectra of 1 in EtOH/CHCl3 (10/90) at varying concentrations; λexc = 355 nm.

preferentially reabsorbed, giving rise to the blue shift and decreased intensity observed at the higher solution concentrations (Figure 4b).24 The sensitizer solutions produced no observable emission beyond 475 nm, indicating the absence of excimer formation, as expected in relatively diluted solutions of a pyrene derivative in polar solvents. (24) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 2nd ed.; Kluwer Academic/Plenum: New York, 1999; p 698.

Langmuir 2009, 25(16), 9219–9226

Thyagarajan et al.

Figure 5. Absorption (a) and fluorescence (b) spectra of 1 bound

to ZrO2; λexc = 355 nm at different surface coverage, method A. Surface coverages are given as a percentage of saturation coverage (0.54=54% of saturation).

Absorption and Fluorescence Spectra Bound. The binding of 1 was studied in different conditions: binding from diluted solutions of varying concentration for a long time (method A, Experimental Section) and from a concentrated solution at varying times (method B, Experimental Section). The absorption and fluorescence spectra of untreated ZrO2 films that were immersed in 0.5 M sensitizer solutions of 1 according to method A are shown in Figure 5. The surface coverage of the films in Figure 5 were estimated by comparing the absorption of the films with the absorption of the highest optical density film of 1 obtained on a base pretreated ZrO2 film having an absorbance of 1.88 at 381 nm. The absorbance spectrum was concentration independent. The fluorescence spectrum of the film of 1 with a surface coverage about 54% of saturation had a broad peak centered at 505 nm. The band at 505 nm is broad, featureless, and significantly red-shifted compared to the spectral features of 1 in solution. This change in the fluorescence spectrum is identical to the characteristic pyrene excimer emission. The bands at 394 and 417 nm (monomer) were more intense for films that had lower surface coverage of 1 and are similar to diluted solutions of 1. The intensity of the monomer spectrum was highest at a surface coverage of ∼16% to saturation and then decreased as the surface coverage of 1 was decreased to ∼2% to saturation. The surface coverage of 1 on ZrO2 was quantified by immersing the films in a concentrated sensitizer solution of 1 and measuring the absorption spectrum of the film over time (binding method B, Experimental Section) (Figure 6). Langmuir 2009, 25(16), 9219–9226

Article

Figure 6. Absorbance (a) and normalized fluorescence (b) spectra of 1 bound to ZrO2. Spectra were recorded after immersing the film in a 50 μM dye solution (acetonitrile) for the indicated periods of time; λexc = 355 nm.

As expected, the absorption of the films increased with the binding time; however, in all cases, the normalized fluorescence spectra were the same within experimental error (Figure 6b). Interestingly, even the fluorescence spectrum of the film with a surface coverage of less than 10% to saturation was almost identical to the fluorescence spectrum observed at saturation coverage. The absorption and fluorescence spectra of ZrO2 films prepared by binding 1 and tetraacid 2 are provided in Figure 7. Two methods were used: either binding from a solution of both 1 and 2 (method C, Experimental Section) or by first binding 2 followed by 1 (method D, Experimental Section). The absorption and fluorescence properties of the films prepared by cobinding from a solution containing both the dye 1 and the tetraacid 2 were similar to those of the films that were prepared by sequential binding (method D). The surface coverage of 1 was proportional to the amount of 1 in the dye solution and inversely proportional to the amount of 2. The fluorescence spectra of the films obtained by cobinding and sequential binding are similar to the fluorescence spectra shown in Figure 7b. The intensities of the monomer and excimer emission were similar for all films. Tripod 1 did not bind to ZrO2 films that were saturated with spacer 2. However, tetraacid 2 displaced 1 on ZrO2 films in competitive binding experiments. In conclusion, the results suggest that the tetraacid 2 displaces 1 and not vice versa. The absorption and fluorescence spectra of 1 and the previously reported pyrene-Ipa-rod 310 on a sapphire substrate are shown in Figure 8. Sapphire (Al2O3) was selected as an insulating, but planar, surface and as a comparison with the ZrO2 films made of nanoparticles (Figure 2). Weak absorption bands at 370 and DOI: 10.1021/la9007679

9223

Article

Thyagarajan et al.

Figure 8. Absorbance (solid line) and fluorescence (dashed line) spectra of tripod 1 (a) and pyrene-Ipa-rod 3 (b) anchored to a planar sapphire substrate; λexc = 355 nm. Figure 7. Absorption (a) and fluorescence (b) spectra of ZrO2 films treated with a solution of 1 and 2 in MeOH/CHCl3 (10/90) varying in concentrations; λexc = 355 nm. See text for details.

390 nm were observed for 1 on sapphire. The fluorescence spectrum of 1 on sapphire, with λmax = 394 nm and a less intense band at 418 nm, was similar to that of 1 in solution, with no indication of excimer emission. The data were compared with those obtained for pyrene-Ipa-rod 3, which served as a control experiment, as it is known that the rodlike dyes closely aggregate on the surface.10b The fluorescence spectrum of 3 had a maximum at 422 nm with the broad peak centered at ∼463 nm typical of excimer emission.

Discussion 1. Binding to Nanostructured MOn. Room temperature reactions of the tripodal compounds with the metal oxide thin films result in surface binding. The UV-vis absorption data in Figure 5a indicate that 1 binds to ZrO2 films with saturation surface coverage of ∼3  10-8 mol/cm2 which is a typical value obtained for dyes bound to metal oxide nanoparticle films. For the point of discussion, this surface coverage is considered 100% for the colloidal films and lower coverages are reported as a percentage of it. Infrared studies of the surface bound tripods are consistent with complete conversion of the COOH groups of the compounds to carboxylates.13 Carboxylate binding is commonly observed for binding sensitizers to metal oxides12 and thus was not unexpected. In summary, the visible and infrared spectroscopic data of 1 are consistent with the idealized three point attachment to the surface with the pyrene oriented perpendicular to and away from the surface (Figure 3c). Low fractional surface coverage (