Photoinduced Electron Transfer in Langmuir− Blodgett Monolayers of

May 3, 2005 - Tommi Vuorinen,* Kimmo Kaunisto, Nikolai V. Tkachenko,. Alexander ... General Physics Institute, Russian Academy of Science, Vavilov st...
0 downloads 0 Views 279KB Size
Download PDF
Langmuir 2005, 21, 5383-5390

5383

Photoinduced Electron Transfer in Langmuir-Blodgett Monolayers of Porphyrin-Fullerene Dyads Tommi Vuorinen,* Kimmo Kaunisto, Nikolai V. Tkachenko, Alexander Efimov, and Helge Lemmetyinen Institute of Materials Chemistry, Tampere University of Technology, P.O. Box 541, 33101 Tampere, Finland

Alexander S. Alekseev General Physics Institute, Russian Academy of Science, Vavilov st. 38, 117942 Moscow, Russia

Kohei Hosomizu and Hiroshi Imahori Department of Molecular Engineering, Graduate School of Engineering, Kyoto University, PRESTO, Japan Science and Technology Agency (JST), Katsura, Nishikyo-ku, Kyoto 615-8510, Japan Received February 7, 2005. In Final Form: March 29, 2005 A series of electron donor-acceptor (DA) dyads, composed of a porphyrin donor and a fullerene acceptor covalently linked with two molecular chains, were used to fabricate solid molecular films with the LangmuirBlodgett (LB) technique. By means of the LB technique, the DA molecules can be oriented perpendicular to the plane of the substrate. In DHD6ee and its zinc derivative hydrophilic groups are attached to the phenyl moieties in the porphyrin end of the molecule; while in the other three dyads, TBD6a, TBD6hp, and TBD4hp, the hydrophilic groups are in the fullerene end of the molecule. This makes it possible to alternate the orientation of the molecules in two opposite directions with respect to the air-water interface and to fabricate molecular assemblies in which the direction of the primary photoinduced vectorial electron transfer can be controlled both by the deposition direction of the LB monolayer and by the selection of the used DA molecule. This was proved by the time-resolved Maxwell displacement charge measurements. The spectroscopic properties of the DA films were studied with the steady-state absorption and fluorescence methods. In addition, the time correlated single photon counting technique was used to determine the fluorescence properties of the dyad films.

Introduction Already for a few decades the design of molecular systems capable of performing light-induced charge separation has been a subject of intensive research.1,2 Specifically, the interest in organic photovoltaics has risen. An essential question is how to create efficient charge separation in an organic photovoltaic system. One promising way is to use electron donor-acceptor molecules to create the primary charge separation. Fullerenes, in particular C60, have excellent electron-accepting properties.3 Because fullerenes have a low absorbance of the visible light they have to be used together with some chromophoric compounds to create the photoinduced electron-transfer reaction. As a result of their conjugated π-electron system, the compounds, for example, porphyrins and phythochlorins, possess good chromophoric activity at the visible region of the spectrum and good electrondonating properties.4 Considering the photoinduced intramolecular electron transfer the porphyrin and fullerene molecules are good building blocks for the donor-acceptor dyad molecules. Covalently linked porphyrin-fullerene * To whom correspondence should be addressed. E-mail: [email protected]. (1) Wasielewski, M. R. Chem Rev. 1992, 92, 435-461. (2) Gust, D.; Moore, T. A. Topics in Current Chemistry; SpringerVerlag: Berlin, 1991; Vol. 159. (3) Echegoyen, L.; Echegoyen, L. E. Acc. Chem. Res. 1998, 31, 593601. (4) Guldi, D. M.; Zilbermann, I.; Anderson, G. A.; Kordatos, K.; Prato, M.; Tafuro, R.; Valli, L. J. Mater. Chem. 2004, 14, 303-309.

or phythochlorin-fullerene dyads have shown in solutions excellent properties for the photoinduced charge separation.5,6 For photovoltaic applications the donor-acceptor molecules have to be immobilized. There are many useful methods to fabricate molecular films. It is essential to have a solid film of DA molecules with an anisotropic structure, that is, all the donor-acceptor (DA) pairs have the same orientation. In this case one may obtain a photoactive DA interface capable of pumping charges in the desired direction under a light illumination. The Langmuir-Blodgett (LB) technique is a well-known method for the preparation of organized monomolecular thin films.7 The technique provides a unique tool for working with amphiphilic DA dyads. By using the LB technique one can fabricate molecular assemblies where (5) (a) Vehmanen, V.; Tkachenko, N. V.; Efimov, A.; Damlin, P.; Ivaska, A.; Lemmetyinen, H. J. Phys. Chem. A 2002, 106, 8029-8038. (b) Tkachenko, N. V.; Lemmetyinen, H.; Sonoda, J.; Ohkubo, K.; Sato, T.; Imahori, H.; Fukuzumi, S. J. Phys. Chem. A 2003, 107, 8834-8844. (c) Armaroli, N.; Marconi, G.; Echegoyen, L.; Bourgeois, J.-P.; Diederich, F. Chem.sEur. J. 2000, 6 (9), 1629-1645. (d) Imahori, H.; Ozawa, S.; Ushida, K.; Takahashi, M.; Azuma, T.; Ajavakom, A.; Akiyama, T.; Hasegawa, M.; Taniguchi, S.; Okada, T.; Sakata, Y. Bull. Chem. Soc. Jpn. 1999, 72, 485-502. (e) Tkachenko, N. V.; Guenther, C.; Imahori, H.; Tamaki, K.; Sakata, Y.; Fukuzumi, S.; Lemmetyinen, H. Chem. Phys. Lett. 2000, 326, 344-350. (6) Chukharev, V.; Tkachenko, N. V.; Efimov, A.; Guldi, D. M.; Hirsch, A.; Scheloske, M.; Lemmetyinen, H. J. Phys. Chem. B 2004, 108, 1637716385. (7) Roberts, G. Langmuir-Blodgett Films; Plenum Press: New York, 1990.

10.1021/la050347l CCC: $30.25 © 2005 American Chemical Society Published on Web 05/03/2005

5384

Langmuir, Vol. 21, No. 12, 2005

Vuorinen et al.

Figure 1. Structures for the porphyrin-fullerene dyads studied.

the DA molecules are oriented all to the same direction, to obtain vectorial photoinduced electron transfer, and the DA layer can be deposited, for example, between the electron and the hole transfer layers and, thus, form part of a photovoltaic device. There are several studies about organized fullerene dyad films made with self-assembling8,9 and LB methods.11,12 In previous studies the fullerene acceptor has been connected to the donor moiety with one linker,11,12 resulting in in-line orientation of porphyrin and fullerene. In the present work, the doubly bridged porphyrin-fullerene dyad has the face-to-face orientation between the donor and the acceptor moieties.6,13 This paper describes in detail the LB film formation properties of various amphiphilic porphyrin-fullerene dyads. The spectroscopic properties of the dyad films are studied and discussed. The steady-state absorption and fluorescence properties as well as the time-resolved fluorescence are in particular discussed. The time-resolved (8) Yamada, H.; Imahori, H.; Fukuzumi, S. J. Mater. Chem. 2002, 12 (7), 2034-2040. (9) Yamada, H.; Imahori, H.; Nishimura, Y.; Yamazaki, I.; Ahn, T. K.; Kim, S. K.; Kim, D.; Fukuzumi, S. J. Am. Chem. Soc. 2003, 125, 9129-9139. (10) (a) Ikonen, M.; Sharonov, A.; Tkachenko, N.; Lemmetyinen, H. Adv. Mater. Opt. Electron. 1993, 2, 6371-6379. (b) Tkachenko, N. V.; Hynninen, P. H.; Lemmetyinen, H. Chem. Phys. Lett. 1996, 261, 234240. (11) (a) Tkachenko, N. V.; Vuorimaa, E.; Kesti, T.; Alekseev, A. S.; Tauber, A. Y.; Hynninen, P. H.; Lemmetyinen, H. J. Phys. Chem. B 2000, 104, 6371-6379. (b) Alekseev, A. S.; Tkachenko, N. V.; Tauber, A. Y.; Hynninen, P. H.; O ¨ sterbacka, R.; Stubb, H.; Lemmetyinen, H. Chem. Phys. 2002, 275, 243-251. (12) Tkachenko, N. V.; Vehmanen, V.; Efimov, A.; Alekseev, A. S.; Lemmetyinen, H. J. Porphyrins Phthalocyanines 2003, 7 (7), 255-263. (13) Efimov, A.; Vainiotalo, P.; Tkachenko, N. V.; Lemmetyinen, H. J. Porphyrins Phthalocyanines 2003, 7 (9), 610-616.

Maxwell displacement charge (TRMDC) method is applied to verify the vectorial photoinduced electron transfer in the DA dyad monolayers. The dependence of the photoelectric signal on the excitation light intensity is used for the estimation of the saturation properties. Materials and Methods Materials. Chloroform of analytical grade (Merck) was used for solution preparation and spreading solvent. Octadecylamine (ODA) was of 99% grade (Sigma). The synthesis for the studied dyad molecules is described elsewhere.13 The structures for the dyads are presented in Figure 1. The dyads having polar groups at the fullerene end, TBD6a, TBD6hp, and TBD4hp, and the four tert-butyl (TB) groups in the porphyrin end are referred to as TB-dyads. The solutions were prepared with the concentration of approximately 1 mg/mL of dyad or ODA in chloroform. The spreading solutions were diluted from these stock solutions to the total concentration equal or less than 1.0 mM. Film Preparation. The LB 5000 and LB Minitrough systems (KSV Instruments, Helsinki, Finland) were used for isotherm determinations and film depositions. The subphase was a phosphate buffer containing 0.5 mM Na2HPO4 and 0.1 mM NaH2PO4 (pH ∼ 7) in ion-exchanged Milli-Q water. The subphase temperature was adjusted with a thermostat to 18 ( 0.5 °C. Samples for spectroscopic studies were deposited onto quartz substrates which were cleaned by the standard procedure.7 In addition, just prior to use, the quartz slides were plasma etched for 15 min in a low-pressure nitrogen atmosphere with plasma cleaner PDC-23G (Harrick). For the photoelectrical measurements glass slides covered by a semitransparent indium tin oxide (ITO) electrode with a sheet resistance of approximately 10 Ω per square were used. The glasses with ITO electrodes were cleaned with an ultrasonic bath first in acetone and then in chloroform, 30 min in each solvent, and plasma etched in nitrogen for 10 min prior to use. Eleven layers of ODA were deposited onto ITO slides to prevent interactions between active dyads

ET in Porphyrin-Fullerene LB Films

Langmuir, Vol. 21, No. 12, 2005 5385

Figure 3. Isotherms for pure dyad monolayers: (1) DHD6ee, (2) ZnDHD6ee, (3) TBD6a, (4) TBD6hp, and (5) TBD4hp. Table 1. Limiting Areas for 100% Dyad Films (S100) and Calculated Areas for Dyad (SDyad) and Matrix (SODA) in Mixed Monolayers

Figure 2. Schematic illustration of the TRMDC measurement cell and dyad monolayer samples.

dyad

S100, Å2

Sdyad, Å2

SODA, Å2

DHD6ee ZnDHD6ee

239 185

TBD6a TBD6hp TBD4hp

250 227 212

231 107a 182b 246 222 211

14.5 23.4a -10.1b 23.3 9.4 21.2

a ZnDHD6ee concentration < 30 mol %. b ZnDHD6ee concentration > 30 mol %.

and the ITO electrode. In addition, samples were finished with 12 layers of ODA to prevent interactions between the active molecules and the top electrode. The deposition conditions and sample structures are described in more detail later on. Spectroscopic Studies. The steady-state absorption spectra for dyad mono- and multilayers were recorded by a Shimadzu UV-2501PC spectrophotometer. Steady-state fluorescence spectra were measured with a Fluorolog 3 fluorimeter (SPEX, Inc.) equipped with a cooled IR sensitive photomultiplier (Hamamatsu R2658). The emission spectra were corrected by using the correction function supplied by the manufacturer. The timeresolved fluorescence measurements were done with a timeresolved single photon counting method, and the decay associated spectra (DAS) were obtained as described elsewhere.14 In short, samples were excited at 590 nm and emission decays were recorded in the wavelength range 620-800 nm. The instrumental response (full width at half-maximum, fwhm) for the used system was about 100 ps. For the DAS, the decays were measured at different wavelengths with a constant collection time (typically 3 min) and fitted globally to a multiexponential model I(t) ) a0 + ∑ai exp(-t/τi). The obtained pre-exponential factors, ai(λ), were corrected by using the sensitivity spectrum of the microchannel plate photomulplier (Hamamatsu R3809U-50) provided by the manufacturer. Electrical Measurements. The vectorial photoinduced electron transfer in the films was studied with the TRMDC method.10,11,12 Samples were excited by a 10 ns laser pulse from the second harmonic of a titanium-sapphire laser (adjustable in range 410-450 nm) pumped by the second harmonic of a Q-switched Nd:YAG laser (532 nm). The time resolution was determined by the excitation pulse width. Schematic illustrations for typical sample structure and detailed structure for a complementary DHD6ee sample pair are given in Figure 2. In general, the sample structure for the photoelectric measurements was ITO|11-12 ODA layers|dyad monolayer|12-13 ODA layers|InGa liquid-metal drop electrode. The ODA layers prevent interactions between the active dyad monolayer and the electrodes. The measured TRMDC signals are, thus, caused only by the photoinduced electron movements inside the dyad layer perpendicular to the plane of the film.11,12 The active dyad layers were deposited either upward (from water to air) or downward (from air to water) directions. For all the dyads sample pairs with complementary DA orientations were prepared. The samples had extremely low conductivity (Rs > 1012 Ω), and, therefore, they could be treated as capacitors with capacitances of typically

Film Preparation and Their Properties. For the monolayer preparations the compressing rate used for the Langmuir film was 15 cm2 min-1. The Langmuir film surface pressure-area (Π-A) isotherms were measured for the 100% dyad films and for the mixed films in an ODA matrix. The isotherms for pure dyad monolayers had a smooth start for the pressure rising and a gentle collapse. All pure isotherms suggested an expanded dyad monolayer on the water surface rather than a condensed one (Figure 3). The limiting areas per molecule were extrapolated from the steepest part of the isotherm curve to the zero surface pressure, and the values are listed in Table 1. The obtained limiting areas for the 100% dyad films varied from 183 to 250 Å2, ZnDHD6ee having the smallest and TBD6a the biggest areas. For a two-molecule immiscible or ideally miscible system, the limiting area can be calculated with an equation Stot ) x1 × (S1 - S2) + S2, where x1 is the molar fraction and S1,2 is the limiting area for the molecule in question.15 This equation was used to estimate the areas for the pure molecules when the dyads were mixed into the ODA matrix with different molar ratios. The obtained limiting areas as a function of the dyad fraction for DHD6ee and ZnDHD6ee are shown in Figure 4. For the other dyads, except ZnDHD6ee, this was fitted with a single slope over the whole range of the molar composition. For ZnDHD6ee the fitting was performed with two separate linear fits (Figure 4). The limiting areas calculated for the dyads and the matrix molecules in the mixed

(14) Vehmanen, V.; Tkachenko, N. V.; Imahori, H.; Fukuzumi, S.; Lemmetyinen, H. Spectrochim. Acta, Part A 2001, 57, 2229-2244.

(15) Gaines, G. L., Jr., Insoluble Monolayers at Liquid-gas Interfaces; John Wiley & Sons: New York, 1966.

100-200 pF. The preamplifier input resistance Rin was 100 MΩ or higher. Thus, with the smallest input resistance the instrumental time constant, τRC, was RinCs ∼ 10 ms. The TRMDC measurements in time domains much shorter than τRC were done in the photovoltage mode.10b,12 In the photovoltage mode the measured signal amplitudes are proportional to the charge displacement, and the signal decay describes the recombination of the charge-separated state in the studied dyad layer.

Results and Discussion

5386

Langmuir, Vol. 21, No. 12, 2005

Vuorinen et al.

Figure 4. DHD6ee and ZnDHD6ee limiting areas as a function of dyad fraction in the ODA matrix. The solid line represents linear fit for DHD6ee limiting areas, and the dashed lines show two fits for ZnDHD6ee areas. Table 2. Surface Density (n), Absorbance (A), Fraction of Absorbed Light (r), and Absorption Cross Section (σ) Values for the Dyad Monolayersa dyad

n, nm-2

A

R

σ, ×10-16 cm2

DHD6ee ZnDHD6ee TBD6a TBD6hp TBD4hp

0.30 0.47 0.25 0.29 0.29

0.01 0.01 0.015 0.011 0.01

0.023 0.023 0.034 0.025 0.023

7.7 4.8 14 8.5 7.7

a ZnDHD6ee has a fraction of 18 mol %, and the rest of the dyads have a fraction of 10 mol % in ODA.

monolayers are listed in Table 1. As can be seen in Figure 4, the change between the slopes for the ZnDHD6ee/ODA monolayer is at approximately 30 mol %. At low concentrations (less than 30 mol %) the limiting area obtained from the fit for ZnDHD6ee is roughly 110 Å2, which is less than the measured area for the 100% ZnDHD6ee isotherm, 183 Å2. At high ZnDHD6ee concentrations (over 30 mol %) the fitted dyad limiting area is 182 Å2 and that for ODA is -10 Å2. In other words, a negative deviation from the linear dependence is obtained. The positive deviation from the linearity is an indication of the repulsive forces between two components in the monolayer.16 Thus, an explanation for the observed negative deviation is that the rigid matrix molecules impede the motion of flexible molecules mixed into it.15 The obtained ODA areas for the DHD6ee and TBD6hp films are also smaller than those expected, but still the area-molar fraction behavior is linear in the whole range. There is, thus, some flexibility in the dyad structure that is seen when mixed in the rigid matrix. In the case of active molecules it is essential to have an optimal coverage of the molecules over the surface area. The relative covers of free-base dyads for the 10 mol % concentration at a 15 mN m-1 surface pressure are approximately 55%, but for ZnDHD6ee the cover is about 36%. When the ZnDHD6ee concentration is increased to 20 mol % the relative cover of the dyad is approximately 56%. Another way to view the coverage is the surface density, n, of the active molecules. Basically, the surface density tells how many molecules there are per unit area of the film. To estimate the surface density one needs to know the mean molecular area, S, of the mixed monolayer and the molar fraction, x, of the active molecule in the matrix. The surface density is n ) x/S. The surface densities of the dyads are shown in Table 2. The concentrations for the samples were selected as a compromise between the highest possible surface coverage of the dyad and the LB film transfer ratio. The LB (16) Wang, S.; Li, Y.; Shan, L.; Ramirez, J.; Wang, P. G.; Leblanc, R. M. Langmuir 1997, 13, 1677-1681.

deposition was unsuccessful for the 100% dyad monolayers. In the ODA matrix the highest dyad concentration for a good-quality LB film (i.e., transfer ratio close to unity) was approximately 20 mol %. For all the dyads other than ZnDHD6ee the selected concentration was 10 mol %. For ZnDHD6ee, the used dyad fraction was 18 mol %. If not otherwise mentioned these are the used dyad concentrations. The depositions were done at the surface pressures of 15, 30, and 45 mN m-1 for the DHD6ee and TBD6a monolayers. The deposition of TBD6a succeeded at all three surface pressures in both directions, but for the DHD6ee monolayer the deposition at the highest surface pressure was possible only in the air-to-water direction. The deposition surface pressure was optimized on the basis of the LB transfer parameters and the electrical measurements (see “TRMDC measurements”). The best surface pressure for the DHD6ee and TBD6a monolayer depositions was 15 mN m-1. The deposition rate for DHD6ee and ZnDHD6ee was 5 mm min-1 and for the TB-dyads was 10 mm min-1. The drying time after every water-to-air deposition was 15 min. Atomic force microscopy (AFM). The AFM measurements, applying the tapping mode, were performed with a Nanoscope IIIa (Digital Instruments, Santa Barbara, CA). The micrographs were recorded for all the dyad films, except for TBD4hp. For the microscopic measurements, the dyad monolayers with a concentration of 20 mol % were deposited in the water-to-air direction onto silicon substrates. The obtained pictures of the film topography show that the TB-dyads have formed a different microstructure than the DHD6ee and ZnDHD6ee dyads. All the monolayers were extremely flat having the difference between the highest and the deepest parts of the film roughly 0.5 nm. On the basis of the number of bonds, the dyad and ODA heights can be approximated to be 1.9 and 2.5 nm, respectively. The difference in the heights corresponds to the difference in the molecular height between the dyad and ODA. As seen in Figure 5a, the ZnDHD6ee film has aggregated areas (dark parts) with respectively irregular shapes. Figure 5b shows that TBD6a forms round aggregates with a diameter of hundreds of nanometers. The DHD6ee and TBD6hp monolayers show structures similar to those of the ZnDHD6ee and TBD6a monolayers, respectively. When areas for different molecular domains are estimated from Figure 5 the obtained ratio between the dyad and the matrix areas does not correlate very well with the calculations based on the isotherm measurements. The TBD6a monolayer micrograph taken on a larger scale showed that the dyad-rich domains have a rather inhomogeneous distribution over the matrix-rich area. In other words, some parts of the monolayer have more than half of the area covered by the dyad while some areas have less than half covered by dyad. Absorption Spectra. For the optical measurements, the dyad samples were deposited either onto a pure quartz substrate or onto a substrate containing nine ODA layers. In the first case, the deposition of the first dyad layer took place in the water-to-air direction and in the latter case in the opposite direction. The absorption spectra were recorded after every water-to-air deposition to monitor the growth of the multilayer film. For all dyad films the increase in the absorbance depended linearly on the number of the deposited films at least for the first seven layers. When the deposition was continued further the linearity disappeared, especially for the TBD4hp and TBD6hp films (Figure 6). When a clean substrate was passed through a Langmuir film, used for several depositions, the transfer ratio and the monolayer absorbance

ET in Porphyrin-Fullerene LB Films

Langmuir, Vol. 21, No. 12, 2005 5387

Figure 5. Atomic force micrographs for 20 mol % (a) ZnDHD6ee and (b) TBD6a monolayers on silicon. Table 3. Positions and Widths of Dyad Soret Bands in Toluene Solution and in 10 mol % LB Films Soret λmax, nm

Figure 6. Absorbance as a function of the number of layers for 10 mol % DHD6ee and TBD6hp. The dashed line represents a linear fit for DHD6ee in the whole range, and the solid line is for the first seven layers of TBD6hp.

Soret fwhm, nm

dyad

toluene

LB

toluene

LB

DHD6ee ZnDHD6ee

428 432

16 16

TBD6a TBD6hp TBD4hp

429 429 429

432 438a 441b 434 430 429

23 20a 21b 19 24 24

15 15 14

a With a ZnDHD6ee concentration of 10 mol %. b With a ZnDHD6ee concentration of 20 mol %.

Figure 8. Normalized steady-state fluorescence emission for different 10 mol % dyad multilayers. Excitation wavelength was 430 nm. DHD6ee and TBD6a films were 10 layer films, but TBD4hp and TBD4hp were 13 layer films. Figure 7. Normalized absorption spectra for DHD6ee in toluene (solid line) and in the Y-type multilayer LB film (dashed line). The Soret band is broadened and shifted to longer wavelengths when DHD6ee is incorporated in the solid film. In the inset is the Soret band absorption.

were the same as for the film just deposited on a multilayer structure. Thus, the decrease in the film transfer ratio was due to the chancing of the dyad/ODA monolayer at the air-water interface during the repeated depositions. In the absorption spectra of the LB films the Soret band is broadened for all dyads and also red-shifted about 5 nm except in the case of TBD6hp and TBD4hp. The absorption spectra of DHD6ee in toluene and in the LB film are shown in Figure 7. The Soret bandwidths and the band positions of the dyads in toluene and in the LB film are listed in Table 3. The spectral red shifts and the band broadenings indicate a side-by-side π aggregation or J aggregation.17,18

Although the observed spectral red shifts are moderate they indicate together with the band broadening that porphyrins tend to be in side-by-side contact to each other. Thus, the absorption spectra support film structures where dyads form aggregates rather than are isolated by the ODA matrix molecules. In addition, the observed spectral shifts suggest that porphyrin moieties form more or less uniform and flat layers, which is due to the high orientation of dyads in the monolayer. Steady-State Fluorescence Spectra. The steadystate emission spectra for DHD6ee, TBD6a, TBD4hp, and TBD6hp multilayer films, normalized at a wavelength of 660 nm, are shown in Figure 8. The fluorescence spectra of DHD6ee, TBD4hp, and TBD6hp are almost similar to that of the porphyrin emission indicating that the length of bridges (4 or 6) has no effect on the fluorescence. The

(17) Imahori, H.; Yamada, H.; Nishimura, Y.; Yamazaka, I.; Sakata, Y. J. Phys. Chem. B 2000, 104, 2099-2108.

(18) Osuka, A.; Maruyama, K. J. Am. Chem. Soc. 1988, 110, 44544456.

5388

Langmuir, Vol. 21, No. 12, 2005

Vuorinen et al.

Figure 9. Decay associated fluorescence emission spectra for (a) DHD6ee and (b) TBD6a.

dyads have, however, a roughly 400 times lower emission intensity compared to that of 5,10,15,20-tetrakis(3,5-tertbutylphenyl)porphyrin (TBP) film.19 In addition, the emissions are shifted to the longer wavelengths and the bands are broadened compared to the TBP films. Practically no emission was detected from the ZnDHD6ee LB films. For TBD6a an intense broad exciplex emission 5a,6 with a maximum at approximately 740 nm was observed. The well-pronounced exciplex emission was similar to that in the phytochlorin-fullerene films.11,12 Time-Resolved Fluorescence. Time-resolved fluorescence measurements were performed for the DHD6ee and TBD6a multilayers. For both dyad films the emission decays are clearly multiexponential. The obtained lifetimes for three exponential fits and the component spectra of the samples are alike (Figure 9). The shortest-lived component, with a lifetime of approximately 70 ps, has two bands around 660 and 720 nm. This component can be attributed to the porphyrin monomer emission. The fitted lifetime is close to the time resolution of the measuring instrument and, thus, the value should be considered as the upper limit for the real lifetime. The second component has the emission band at 720 nm and the lifetimes of approximately 300 and 600 ps for the DHD6ee and TBD6a films, respectively. These bands correspond the fullerene fluorescence emission.11,20,21 The observed lifetimes for fullerene emission are less than half of the fluorescence lifetime in solution for the pristine C60, τ ≈ 1.2 ns.22 The longest living component has lifetimes of 1.3 and 2 ns for DHD6ee and TBD6a, respectively. This component has a broad band in the region of 700-800 nm and is attributed to the exciplex emission.5a,11,12 The time constants of the DHD6ee exciplex emission in nonpolar and polar solvents are 2.9 and 0.38 ns, respectively.6 Thus, the observed time-constant in the LB film falls between the values measured in different types of solvents. TRMDC Measurements. The polarity of the TRMDC signal depends on the orientation of the donor and the acceptor relative to the ITO electrode. The TRMDC measurements provide, thus, an elegant way to verify the orientation of the DA dyads in an LB monolayer. For example, when the deposition direction of the DA dyad monolayer is changed from the air-to-water to the waterto-air direction, also the polarity of the observed electrical signal is changed to opposite. Moreover, the relative orientation of the donor and the acceptor moieties can be preserved by changing the type of the dyad but maintain(19) Anikin, M.; Tkachenko, N. V.; Lemmetyinen, H. Langmuir 1997, 13, 3002-3008. (20) Sun, Y.-P.; Wang, P.; Hamilton, N. B. J. Am. Chem. Soc. 1993, 115, 6378. (21) Catala´n, J.; Elguero, J. J. Am. Chem. Soc. 1993, 115, 9249. (22) Kim, D.; Lee, M.; Suh, Y. D.; Kim, S. K. J. Am. Chem. Soc. 1992, 114, 4429.

Figure 10. Photovoltage response signals for DHD6ee (dashed lines) and TBD6hp (solid lines) monolayers (10 mol % in ODA). “Up” means deposition has taken place at a surface pressure of 15 mN m-1 in the water-to-air direction, and “down” means in the air-to-water direction. The excitation wavelength was 430 nm, and the excitation energy densities were 0.5 mJ cm-2.

ing the deposition direction. The behavior described above is observed for the photovoltage response signals for the DHD6ee and TBD6hp monolayers (Figure 10). The DA molecules are oriented at the air-water interface as expected according to the hydrophilic groups in the different ends of the molecules and will be transferred onto the solid substrate without changes in the orientation. For the DHD6ee and ZnDHD6ee monolayers the change in the deposition direction results in photoelectric signals which are practically mirror images. Instead, the change in the deposition direction of the TB-dyad monolayers results in photovoltage response signals which are not mirror images with respect to the horizontal axis at the zero voltage. All the TB-dyads have a characteristic shape for the TRMDC signal (Figure 10), in which a fast formation to the expected direction is followed by a fast relaxation leading to inversion of the signal polarity after about 100 ns after the excitation. One possible explanation for the observed charge recombination in the TB-dyad monolayers is the environment inside the monolayer. The charge separation in a TB-dyad monolayer takes place in a different direction with respect to the polar hydrophilic groups of the film molecules than in the DHD6ee and ZnDHD6ee monolayers. Previously we have observed similar dependence of the photoelectric signal polarity on the deposition direction of the phytochlorin-fullerene LB films.11,12 Effect of the Deposition Pressure on the TRMDC Signal. The effect of the deposition surface pressure on the photovoltage response signal was studied for the DHD6ee and TBD6a films (Figure 11). The used deposition surface pressures were 15, 30, and 45 mN m-1. The relative photovoltage response amplitudes of the DHD6ee monolayers, deposited in the air-to-water direction at different surface pressures, are 1, 0.53, and 0.26 at the pressures

ET in Porphyrin-Fullerene LB Films

Langmuir, Vol. 21, No. 12, 2005 5389

Figure 11. Dependence of the photovoltage signals on the deposition surface pressure for the monolayers of (a) DHD6ee and (b) TBD6a.

Figure 12. (a) Dependence of the PV response amplitude on the excitation energy density and the saturation fits for the dyad monolayers deposited in the water-to-air direction. (b) The saturation fit with eq 1 (dashed line) and the initial slope from the measured data (solid line) for the DHD6ee monolayer.

15, 30, and 45 mN m-1, respectively. The shape of the photovoltage signal for DHD6ee was practically independent of the deposition pressure and direction. The characteristic behavior of the TB-dyads is seen also at different deposition pressures of the TBD6a monolayers. For the monolayer deposited in the air-to-water directions the maximum signal amplitudes are insensitive to the deposition surface pressure, but the decay profiles differ from each other (Figure 11b). For the film deposited upward at the pressure 45 mN m-1, the fast positive component vanished. The deposition surface pressure has influence on the photovoltage signals for both DHD6ee and TBD6a monolayers. Increase in the deposition surface pressure resulted in a decrease of the DHD6ee photovoltage response amplitude. Thus, 15 mN m-1 was selected as the deposition surface pressure for the dyad monolayers. No significant changes in the spectroscopic properties of the monolayer were observed when the deposition surface pressure was changed. Thus, influence of the deposition surface pressure on the electrical signals is most probably due to small changes in the mutual orientation of the porphyrin and fullerene moieties caused by the compression of the monolayer to higher surface pressures. In a similar way it was observed in solution that the spectroscopic properties of the porphyrin-fullerene dyads depend on the length of the two molecule bridges connecting the moieties.6 Effect of Excitation Energy Density on the TRMDC Signal. The saturation of the photovoltage signal amplitude was determined for the DHD6ee, TBD6a, and ZnDHD6ee monolayers by measuring the photoelectric response amplitudes at different excitation energy densities. The results are shown in Figure 12a. The excitation wavelengths were 430 and 440 nm for free-base and zinc porphyrins, respectively. The used excitation energy was controlled by the set of neutral density filters, and the

Table 4. Saturation Amplitudes (U0), Saturation Energy Densities (I0), and Absorption Cross Sections (σ) Calculated from the I0 Values for the Dyad Monolayers Deposited in the Water-to-Air Direction dyad

U0, V

I0, mJ cm-2

σ, ×10-16 cm2

DHD6ee ZnDHD6ee TBD6a

0.56 0.83 0.39

0.71 0.87 1.04

6.5 5.2 4.4

maximum energy was roughly 5 mJ cm-2, corresponding approximately 1 × 1016 photons cm-2. The obtained photovoltage amplitude (Uout) versus excitation energy density (Iexc) is fitted by using a simple saturation model:

(

[ ])

Uout ) U0 1 - exp -

Iexc I0

(1)

where U0 and I0 are the saturation amplitude and the saturation excitation density, respectively. Basically, I0 is the light intensity at which the photon density is equal to the inverse of the chromophore absorption cross section, I0 ) hν/σ. The fits for all monolayers are shown in Figure 12a. The saturation amplitude, U0, is the maximum amplitude at infinite excitation energy. The obtained saturation values from the fit with eq 1 for the dyad monolayers deposited from water to air are shown in Table 4. The used saturation model shows a good agreement with the measured data, although it is based on the probability of photon absorption and does not take into account the intermolecular interaction between the excited species. Thus, it is not necessary to introduce a more complex saturation model, as it was the case of the phytochlorin-fullerene dyad LB films.12 The initial slope of the Uout versus Iexc curve reveals the sensitivity of the charge separation to the photon flux. The slope can be obtained from the ratio of the saturation

5390

Langmuir, Vol. 21, No. 12, 2005

Vuorinen et al. Table 5. Obtained β Values from the U(t) ∼ t-β Fit

Figure 13. Photovoltage decay for the 18 mol % ZnDHD6ee monolayer deposited in the air-to-water direction. In the inset, the fast part of the ZnDHD6ee decay is shown.

values, U0/I0. The initial slope can also be taken directly from the measured data (Figure 12b) at low excitation densities, Iexc , I0. The slopes (Uout/Iexc) from the measured data are 0.95, 1.19, and 0.64 V mJ-1 cm2 for the DHD6ee, ZnDHD6ee, and TBD6a monolayers, respectively. The corresponding ratios from the saturation values (U0/I0) are about 20% smaller for DHD6ee and ZnDHD6ee and roughly 50% smaller for TBD6a than the values taken directly from the data. The saturation model describes the Uout versus Iexc dependence at low excitation densities better for DHD6ee and its zinc derivative than for TBD6a. The saturation energy densities and the absorption cross sections calculated from those values (σ ) hv/I0) are presented in Table 4. For DHD6ee and ZnDHD6ee the values for the absorption cross sections obtained from the steady-state absorption measurements (Table 2) and the photovoltage saturation measurements correlate very well. But for TBD6a the σ value obtained from the absorption measurements is two times higher than that from the photovoltage measurements. Decay of the Charge-Separated State. The charge recombination decay in DHD6ee and ZnDHD6ee follows the power law U(t) ∼ t-β, as shown in Figure 13. When the input resistance of the TRMDC instrument is set to 10 GΩ, the instrumental time constant is increased to 500 ms. The slower time constant enables one to measure the photovoltage signals in the longer time domains without perturbation of the signal by the instrumental decay. The β values obtained from power-law fits are on the order of 0.25 for the DHD6ee and ZnDHD6ee monolayers; see Table 5. The parameter β remained constant for five decades in time (Figure 13), in other words, in a double logarithmic plot the linear part of the signal had a length of 5 orders of magnitude of time. At the time of roughly 0.1 s, the decay order changes because of the RC time constant of

monolayer

β

monolayer

β

DHD6ee (up) DHD6ee (down)

0.13 0.23

ZnDHD6ee (up) ZnDHD6ee (down)

0.23 0.28

the measurement circuit. For example, in conductive polymer-fullerene blends the density of photoinduced charge-separated states is obeying the power law with β values of approximately 0.4.23,24 In the case of polymerC60 mixtures, the sides of charge traps have an isotropic distribution over the bulk material while in the present system the electron donors and acceptors are located in two parallel layers into which the charges are separated. In other words, in the present DA system the charge recombination occurs only in one definite direction, while in the polymer-C60 samples all the directions for the charge recombination are possible. Thus, different charge recombination kinetics for these two systems are expected to be observed. Conclusions The LB technique was utilized to fabricate molecular thin films of novel porphyrin-fullerene dyads in which the electron donor and the acceptor moieties are covalently linked with two bridges. The multilayer film preparation was possible when dyads were mixed into an ODA matrix. The flexibility of the dyad molecules is seen, for example, when the limiting areas for the mixed monolayers are examined. In the cases of DHD6ee, ZnDHD6ee, and TBD6hp, negative deviations from the expected areas are observed. The TRMDC measurements verify that the orientation of the DA dyads in the monolayer is controlled by the position of the hydrophilic groups in the molecule. The recombination of the charge-separated states in the monolayers of DHD6ee and its zinc derivative follows the power law (nCS ∼ t-β) with β being approximately 0.25. Acknowledgment. This work was supported by the Academy of Finland, Project “New Artificial DonorAcceptor Materials”, the National Technology Agency of Finland. This work was supported partially by Grantin-Aid (No. 16310073 to H.I.) from MEXT, Japan. H.I. also thanks Grant-in-Aid from MEXT, Japan (21st Century COE on Kyoto University Alliance for Chemistry), for financial support. LA050347L (23) Nogueira, A. F.; Montanari, I.; Nelso, J.; Durrant, R.; Winder, C.; Sariciftci, N. S.; Brabec, C. J. Phys. Chem. B 2003, 107, 1567. (24) Montanari, I.; Noqueira, A. F.; Nelson, J.; Durrant, J. R.; Winder, C.; Loi, M. A.; Sariciftci, N. S.; Brabec, C. Appl. Phys. Lett. 2002, 81, 3001.