Long-Lived Photoinduced Charge Separation Due to the Inverted

Oct 11, 2010 - Kesavapillai Sreenath , Tony George Thomas , and Karical R. Gopidas. Organic Letters 2011 13 (5), 1134-1137. Abstract | Full Text HTML ...
0 downloads 0 Views 2MB Size
Download PDF
J. Phys. Chem. C 2010, 114, 18725–18734

18725

Long-Lived Photoinduced Charge Separation Due to the Inverted Region Effect in 1,6-Bis(phenylethynyl)pyrene-Phenothiazine Dyad Chettiyam Veettil Suneesh and Karical R. Gopidas* Photosciences and Photonics, Chemical Sciences and Technology DiVision, National Institute for Interdisciplinary Science and Technology, Council of Scientific and Industrial Research, TriVandrums695 019, India ReceiVed: August 12, 2010; ReVised Manuscript ReceiVed: September 22, 2010

Photoinduced electron-transfer processes in a 1,6-bis(phenylethynyl)pyrene-phenothaiazine dyad, BPEP-PT, are examined using various techniques. The BPEP singlet excited state is quenched by electron transfer from PT leading to formation of BPEP radical anion and PT radical cation. Rate constants and quantum yields of the PET processes were determined from steady-state and time-resolved fluorescence experiments, and spectroscopic identification of the radical ion products was achieved using picosecond and nanosecond flash photolysis experiments. The charge-separated (CS) state was found to be long-lived but decayed to the BPEP triplet state under the influence of external heavy atom effect. The energy level diagram constructed on the basis of experimental data revealed the existence of the local triplet state below the CS state, yet the CS state did not exhibit any tendency to decay to this level. This showed that hyperfine interaction (HFI) or spin-orbit charge-transfer intersystem crossing (SOCT-ISC) mechanisms are not effective in inducing intersystem crossing in the CS state. It is suggested that absence of SOCT-ISC in the CS state may be a consequence of the total absence of ISC in the parent BPEP chromophore. The only option available to the CS state is a spin-allowed transition to the ground state, and this process is slow because of inverted region effects. 1. Introduction Photoinduced electron transfer (PET) is a thoroughly investigated process in natural and artificial systems, and it involves the transfer of an electron from a donor (D) to an acceptor (A) molecule following the excitation of D or A. PET results in the formation of a high-energy charge-separated (CS) state consisting of the radical cation of D and radical anion of A. In linked D-A systems the CS state in general is very short-lived and undergoes back electron transfer (BET) to regenerate the starting materials leading to wastage of the absorbed energy.1 D-A systems capable of generating long-lived CS states would be important in artificial photosynthesis and also in the design of organic photovoltaic cells.2-8 The term “conversion of solar energy into chemical energy” actually means the generation of a high-energy CS state by way of light absorption. In the natural photosynthetic system creation of a long-lived CS state is achieved through a number of relatively fast short-range electron-transfer steps leading to large overall charge separation distance which prevents the rapid BET process.9 Scientific efforts of the past few decades have resulted in the design and study of a large number of multichromophoric systems (triads, tetrads, pentads, etc.) capable of sustaining long-lived CS states.10-14 The multichromophoric systems, however, suffer from two major disadvantages. (1) Because of the difficulties involved in the synthesis of these complex systems they can seldom be used for any practical applications. (2) Every electrontransfer step in these polyads is exergonic, and hence, the energy stored in the final CS state is much smaller than the energy absorbed initially. Thus, the design and study of simple D-A systems capable of long-lived photoinduced charge separation is an important goal in PET research even today.15-18 * To whom correspondence should be addressed. E-mail: gopidaskr@ rediffmail.com.

Fukuzumi and co-workers have recently reported long-lived photoinduced charge separation, attributable to inverted region effects, in a few simple molecular dyads.19-23 The inverted region was a prediction of the Marcus theory,24 which states that the rate constant for nonadiabatic intramolecular electron transfer (ket) is given by

ket ) (2π/p)Hel2(4πλkBT)-1/2 exp[-(λ + ∆G0)2 /4λkBT] (1) where p is the Planck’s constant divided by 2π, Hel is the electronic coupling matrix element between the donor and acceptor, ∆G0 is the free energy change, λ is the reorganization energy of the electron transfer, kB is the Boltzmann constant, and T is the absolute temperature. The reorganization energy of electron transfer (λ) is the energy required to structurally reorganize the donor, acceptor, and their solvation spheres upon electron transfer. The Marcus equation defines a “normal” region (-∆G0 < λ) where ket increases with driving force (-∆G0) and an “inverted” region (-∆G0 > λ), where ket decreases with increase in driving force. According to Fukuzumi and coworkers D and A can be carefully selected in such a way that -∆G0PET ≈ λ and -∆G0BET . λ, and in such cases kPET/kBET would be very large and PET would result in long-lived CS states. Fukuzumi and co-workers claimed that the systems studied by them satisfy these conditions leading to the observation of long-lived CS states. Harriman, Verhoeven, and co-workers have challenged the claims of Fukuzumi’s group regarding the long-lived CS states in simple dyads.25-30 Harriman and Verhoeven pointed out that when -∆G0BET is very large, other deactivation channels such as jumps to low-lying local triplet levels may prevail over charge recombination to the ground state. Nuclear tunneling also may

10.1021/jp107606t  2010 American Chemical Society Published on Web 10/11/2010

18726

J. Phys. Chem. C, Vol. 114, No. 43, 2010

become very important, and these authors opined that the Marcus equation may not adequately describe electron transfer in the deep inverted region. They describe two possible cases. In the first case the 1CS state formed can undergo rapid interconversion to 3CS state which then undergoes a spinforbidden charge recombination to the ground state. In such cases the 3CS state can be long-lived if charge recombination to the ground state is the only decay pathway available to the 3 CS state. This, however, requires that the 3CS state is the lowest triplet state in the dyad as a whole because otherwise spinallowed decay of 3CS to local triplet states (3LE) is bound to prevail. Thus, wherever the 3CS state is the lowest state above the ground level, the CS state would be long-lived due to spin control and there is no need to invoke the inverted region effect to explain the long lifetime of the CS state. In the second scenario a 3LE level exists below the CS state, in which case CS state would decay to the 3LE and long-lived CS state would not be observed. We have observed inverted region kinetics in the past in PET reactions in the moderately large driving force regimes and did not come across any tunneling effects.31-34 We, however, agree with the view that in the deep inverted region existence of 3LE levels below the CS state is quite possible and in such cases decay of the CS state to 3LE levels must be taken into consideration. Formation of 3LE states in the photolysis of D-A systems is in fact well-documented in the literature.35,36 If the models put forward by Verhoeven and Harriman are the only possible models, then it would be impossible to have CS states exhibiting long lifetimes due to inverted region effects. Our studies show that other models which lead to long-lived charge separation due to inverted region effects can also exist. Recently, we have investigated the PET processes in two molecular dyads containing 9,10-bis(phenylethynyl)anthracene (BPEA) as acceptor and one or two phenothiazine (PT) units as donors.37 The dyads BPEA-PT and BPEA-PT2 have 3LE levels below the CS state, yet the CS states generated in these cases did not exhibit any tendency to decay to this level. A CS state generated in the singlet spin state can decay to a lowlying 3LE state through two possible mechanisms.38 The energy gap between 1CS and 3CS states is twice the exchange integral (2J). If |2J| is very small, the hyperfine interaction (HFI) between the electron and nuclear spins can lead to interconversion between 1CS and 3CS, and the 3CS thus formed can undergo a facile, spin-allowed transition to 3LE. The HFI energy is typically of the order of 50 G for organic radicals, and this corresponds to S T T interconversion frequency of 1.4 × 108 s-1. This implies that the radical ion pair formed in the singlet state will lose its pure singlet character within a few nanoseconds if |2J| < 50 G. If |2J| is substantially larger, the HFI mechanism can no longer drive S T T interconversion. It was estimated that if |2J| ) 1 cm-1, the S T T interconversion will be slowed down to microsecond time domain at ambient temperature in solution. In a series of D-bridge-A systems it has been shown that |2J| decreases exponentially with distance.38 When the distance separating D and A is small, |2J| is very large, and S T T interconversion will not take place by the HFI mechanism. If the orientation between the relevant orbitals of D and A is such that charge transfer between them is accompanied by a significant change in orbital angular momentum, recombination of 1(D•+-A•-) may be accompanied by a spin flip to directly populate low-lying local triplet levels without invoking 3 (D•+-A•-).39 Okada et al.40 observed that in a series of pyrene-bridge-dimethylaniline (PY-b-DMA) systems the rate of charge recombination to produce 3PY was strongly

Suneesh and Gopidas SCHEME 1: Scheme for Dyads That Can Exhibit Long-Lived CS Due to the Inverted Region Effect

dependent on the relative orientation of PY and DMA. In some of these cases generation of 3PY from the 1CS state occurred within 30 ps, and this could not be explained by the HFI mechanism, which usually requires a few nanoseconds. They concluded that when the electron-donating and electron-accepting molecular orbitals are approximately perpendicular to each other, the rate of 1CS f 3PY is increased. This mechanism for population of local triplet is formally similar to spin-orbit coupling (SO) ISC that occurs in an n-π* electronic transition within a chromophore. Dance et al. have coined the term spin-orbit charge-transfer intersystem crossing (SOCT-ISC) to describe this ISC mechanism.39 Since this mechanism is similar to SO-ISC, yields of 3LE states generated through this route can be enhanced by heavy atom effects. The BPEA-PT systems we reported earlier did not exhibit any tendency to decay to the 3LE state, and we may conclude that both HFI and SOCT-ISC mechanisms are not efficient in these systems. In intramolecular dyads the HFI mechanism becomes operational only if the distance between D and A is large. Hence, it is not surprising that this mechanism is not operating in BPEA-PT where the D and A moieties are separated by only three methylene units. It was, however, not clear as to why the SOCT-ISC mechanism is not able to operate here. Under these circumstances the 1CS state generated undergoes BET to the ground state as shown in Scheme 1. Such a possibility has not been considered by Verhoeven, Harriman, and co-workers. For systems obeying Scheme 1, long-lived charge separation can be attributed to inverted region effects and the long lifetime of the CS state observed in BPEA-PT systems was accordingly attributed to the inverted region effects. A new system which obeys Scheme 1 is reported here. In this paper we report the photoinduced electron-transfer processes in a dyad wherein 1,6-bis(methoxyphenylethynyl)pyrene (BPEP) serve as light absorber and acceptor and phenothiazine (PT) units serve as electron donor. The BPEP core is connected to the PT moieties through a short alkyl chain. Photoprocesses in the dyad BPEP-PT are compared with a model BPEP system. Structures of the model system and dyad are shown in Figure 1 along with a minimum energy conformation of BPEP-PT obtained by AM1 calculation using Gaussian 03 (revision D.01). We have probed the photoinduced charge separation in the BPEP-PT dyad using steady-state and timeresolved fluorescence measurements. Picosecond and nanosecond laser flash photolysis techniques confirmed the formation of a long-lived CS state in the dyad.

1,6(Bisphenylethynyl)pyrene-Phenothiazine Dyad

J. Phys. Chem. C, Vol. 114, No. 43, 2010 18727

Figure 1. Structures of compounds used in this study and the optimized structure of BPEP-PT.

2. Experimental Section

(2)

Quanta Ray Nd:YAG laser or by using an Applied Photophysics model LKS-60 laser kinetic spectrometer by using the third harmonic (355 nm) from an INDI-40-10-HG Quanta Ray Nd: YAG laser. The analyzing and laser beams were fixed at right angles to each other. Solutions for laser flash photolysis studies were deaerated by purging with argon for 20 min before experiments. Picosecond transient absorption experiments were carried out using mode-locked Nd:YAG laser system PY61C10 (355 nm, 5 mJ/pulse, fwhm 35 ps, 10 Hz repetition rate). The white light probe was generated by focusing fundamental laser output (1064 nm) onto a 10 mm quartz cuvette containing 10 mL water and 10 mL D2O mixture. The optical delay line provided a pump beam time window of 2.45 ns and step resolution of 3.33 ps. The transmitted probe light was focused on to a 200 µm core fiber connected to an Ocean Optics SD2000 UV-vis CCD spectrophotometer (400-800 nm). Typically, 100 excitation pulses were averaged to obtain the transient absorption at the set delay time. All the experiments were conducted at room temperature. Details of the synthetic procedures for compounds used in the study are given in the Supporting Information.

where the subscript R refers to the reference, OD is the optical density at the excitation wavelength (OD ≈ 0.1 at 390 nm for both reference and standard), n is the refractive index of the solvent, and A is the area under the fluorescence spectrum. Electrochemical experiments were performed by using a BAS 50W voltammetric analyzer. Solutions of the compounds (1 × 10-3 M) in dichloromethane containing 0.1 M tetra-n-butylammonium hexafluorophosphate were thoroughly deaerated and used for cyclic voltammetry (CV) experiments. Time-resolved fluorescence experiments were performed by using an IBH picosecond single-photon counting system employing a 401 nm nanoLED excitation source and a Hamamatsu C4878-02 microchannel plate (MCP) detector. Nanosecond laser flash photolysis experiments were either performed by using an Applied Photophysics model LKS-20 laser kinetic spectrometer by using the third harmonic (355 nm) from a GCR-12 series

3. Results and Discussion 3.1. Synthesis and Characterization of BPEP and BPEPPT. The substrates used in this study, namely, BPEP and BPEP-PT, were synthesized as shown in Scheme 2. The various intermediates and final products were fully characterized by spectroscopic and analytical techniques. Data for the final products are presented here. 3.1.1. Data for BPEP. mp 261-262 °C. IR (KBr) νmax: 530, 631, 723, 812, 831, 852, 1028, 1105, 1165, 1178, 1250, 1290, 1447, 1460, 1510, 1597, 2199, 2549, 2839, 3005 cm-1. 1H NMR (CDCl3, 500 MHz) δ: 3.88 (s, 6H), 6.97 (d, 4H), 7.66 (d, 4H), 8.16 (m, 4H), 8.20 (d, 2H), 8.67 d, 2H). 13C NMR (CDCl3, 125 MHz) δ: 55.42, 92.42, 102.57, 114.22, 115.21, 118.44, 126.52, 127.52, 131.97, 133.17, 157.94. FAB-MS (M+) calcd for C34H22O2, 462.54; found, 462.38. 3.1.2. Data for BPEP-PT. mp 215 °C (dec). IR (KBr) νmax: 530, 741, 835, 851, 960, 1059, 1238, 1248, 1288, 1375, 1465,

Melting points were determined on a Mel-Temp II melting point apparatus and are uncorrected. Proton NMR data were obtained from either a 300 MHz Bruker Avance DPX spectrometer or a 500 MHz Bruker Avance DPX spectrometer. 13C NMR spectra were recorded using a 500 MHz Bruker Avance DPX spectrometer. FT-IR spectra were recorded on a Shimadzu IR Prestige 21 spectrometer. High-resolution mass spectra were obtained by using a JEOL JMS600 mass spectrometer. AbsorptionspectrawereobtainedusingaShimadzu3101PCUV-vis-NIR scanning spectrophotometer. Steady-state fluorescence experiments were performed with a SPEX Fluorolog F112X spectrofluorimeter by using optically dilute solutions. The fluorescence quantum yields in dichloromethane were determined with a relative method employing an optically matched solution of 9,10-diphenylanthracene in ethanol as reference (ΦR ) 0.95).41 The following equation was used:

ΦF ) ΦR

AODRn2 ARODnR2

18728

J. Phys. Chem. C, Vol. 114, No. 43, 2010

Suneesh and Gopidas

SCHEME 2: Scheme Used for the Synthesis of BPEP and BPEP-PTa

a

Conditions: (a) Pd(PPh3)2Cl2, CuI, NHiPr2, toluene, 24 h, 70 °C; (b) Pd(PPh3)2Cl2, CuI, NHiPr2, toluene, 24 h, 70 °C.

Figure 2. Absorption spectra of (a) BPEP and (b) BPEP-PT in DCM. The inset shows the absorption spectrum of N-methylphenothiazine in DCM.

Figure 3. Fluorescence spectra of (a) BPEP and (b) BPEP-PT in DCM.

1510, 1597, 2201, 2965 cm-1. 1H NMR (CDCl3, 500 MHz) δ: 2.31 (m, 4 H), 4.15 (m, 8 H), 6.89 (d, 2 H), 6.93 (m, 8 H), 7.17 (m, 8 H), 7.60 (d, 4 H), 8.15 (m, 6 H), 8.66 (d, 2 H). 13C NMR (CDCl3, 125 MHz) δ: 26.79, 43.88, 65.42, 92.41, 102.59, 114.27, 115.14, 118.43, 122.77, 125.32, 126.51, 127.19, 127.50, 129.88, 132.00, 133.22, 146.16, 157.98. FAB-MS (M+) calcd for C62H44N2O2S2, 912.28; found, 912.29. 3.2. Photophysical and Electrochemical Studies. The absorption spectrum of BPEP is considerably red-shifted when compared with that of the parent pyrene. The long-wavelength absorption due to BPEP in dichloromethane (DCM) occurs in the 350-450 nm region (Figure 2). The long-wavelength absorption in BPEP-PT is nearly identical to that in BPEP suggesting that the donor and acceptor moieties did not exhibit any electronic interaction in the ground state. The absorption due to the PT moieties in the dyad appeared at 250 and 310 nm. For the purpose of comparison the absorption spectrum of N-methylphenothiazine in DCM is shown in the inset of Figure 2. It can be safely assumed from Figure 2 that excitation above 350 nm populates only the BPEP excited states in the dyad. The fluorescence spectrum of the model compound BPEP is shown in Figure 3a. The emission maximum occurs at 438 nm. The Stokes shift observed is very small (810 cm-1) suggesting similar geometry for the S0 and S1 states of BPEP. Using the absorption and emission spectra the excitation energy (E00) was calculated, and the value obtained was 2.86 eV. The fluorescence quantum yield (ΦBPEP ) 0.93) was very high for BPEP. Substitution of PT leads to considerable quenching of the

fluorescence intensity as is evident from a comparison of the fluorescence spectra of BPEP and BPEP-PT shown in Figure 3. Spectra a and b in Figure 3 were recorded under identical conditions, and optical densities were matched at 390 nm. The fluorescence quantum yield for BPEP-PT (ΦBPEP-PT) was 0.1. Fluorescence lifetimes of BPEP and BPEP-PT were determined using the single-photon counting technique (Supporting Information). The samples were excited using a 401 nm nanoLED excitation source. BPEP exhibited monoexponential decay with a lifetime (τ0) of 1.3 ns. BPEP-PT also exhibited a monoexponential decay with a lifetime (τ1) of 0.14 ns. Decay profiles of BPEP and BPEP-PT along with the instrument function are presented in the Supporting Information (Figure S2). It is obvious from the above results that PT is an efficient quencher of BPEP fluorescence. Since the absorption due to PT occurs at higher energy compared to that of BPEP, quenching of BPEP fluorescence by energy transfer to PT is not possible, and we can safely assume an electron-transfer mechanism for the quenching. The reduced fluorescence quantum yield and lifetime of BPEP-PT are due to facile electron transfer from PT to the 1S* state of the BPEP moiety in the dyad. Redox potentials of the compounds (vs SCE) were measured in DCM using square wave voltammetry (see the Supporting Information). BPEP exhibited an oxidation peak at 1.10 V and reduction peak at -1.68 V (vs SCE). In BPEP-PT the redox potentials associated with BPEP remained the same, and an additional oxidation peak corresponding to PT/PT•+ oxidation

1,6(Bisphenylethynyl)pyrene-Phenothiazine Dyad

J. Phys. Chem. C, Vol. 114, No. 43, 2010 18729

appeared at 0.72 V (vs SCE). Thus, in BPEP-PT, the lowest reduction is associated with the BPEP chromophore and the lowest oxidation is associated with PT. Hence, in the BPEP-PT dyad, BPEP is the acceptor and PT is the donor. The free energy change associated with the electron transfer from PT to the 1S* state of BPEP can be calculated using the Weller equation:42

∆G0PET ) F(Eox - Ered) - E00 - e2 /εdcc

(3)

where F is the Faraday constant, Eox is the oxidation potential of PT, Ered is the BPEP-centered reduction potential, dcc is the center-to-center distance between the BPEP and PT moieties, and εs is the solvent dielectric constant. We used dcc ) 15.47 Å (obtained from AM1) for the calculation and obtained ∆GPET ) -0.56 eV. Thus, PET is exergonic in the BPEP-PT dyad, and upon excitation to the lowest energy band we expect the following processes to take place. hν > 350nm

BPEP - PT. 98 1BPEP* - PT

ET

BPEP* - PT f •-BPEP - PT•+

1

Figure 4. Picosecond transient absorption spectra obtained in flash photolysis of BPEP in DCM. The inset shows kinetic profiles of the transients at 730, 690, and 600 nm.

(4)

(5)

BET

•-

BPEP - PT•+ 98 BPEP - PT

(6)

Figure 5. Transient absorption spectrum of BPEP immediately after the laser pulse and decay traces at 690 nm in the absence (top) and presence (bottom) of oxygen.

The rate constant for the PET reaction can be calculated from the fluorescence quantum yields using eq 7.

kPET )

(

)

1 ΦBPEP -1 τ0 ΦDYAD

(7)

We obtained kPET values of 6.38 × 109 s-1. kPET values can also be calculated from the fluorescence lifetime data using eq 8.

kPET )

1 1 τl τ0

(8)

Using eq 8 we obtained kPET ) 6.37 × 109 s-1. kPET values calculated using the two equations agree very well. The quantum yield for PET was calculated using the kPET (Φet ) kPETτ1), and we obtained value of 0.89 for BPEP-PT, which indicated that PET is very efficient in this system. 3.3. Flash Photolysis Studies of Model Compound BPEP. 3.3.1. Picosecond Flash Photolysis. In order to gain a deeper understanding of the excited-state processes, model compound BPEP was subjected to picosecond and nanosecond flash photolysis in deaerated DCM solution. The transient absorption spectra obtained in the picosecond flash photolysis of BPEP in the 40 ps to 2.3 ns range are shown in Figure 4. The spectra taken at very short times after the laser flash showed absorption around 730 nm and bleaching below 500 nm. At longer times (1.5-2.3 ns), formation of a transient that absorbs in the entire visible region with a maximum at 690 nm is indicated. Curves A-C in the inset of Figure 4 show the ∆OD versus time profiles at 600, 690, and 730 nm. It can be seen that the 730 nm absorption exhibited a fast decay followed by a slow decay.

The profile for the 690 nm absorption was complex initially but showed a growth at later times. At 600 nm no decay or growth was observed in the 40-850 ps range, but a clear growth was observed thereafter. We assign the 730 nm absorption to the S1 f Sn absorption of BPEP. Decay of this absorption is complex because of the formation of new absorptions throughout the observation window. A probable candidate for the new absorption at 690 nm is the triplet state of BPEP. Such an assignment, however, is not supported by the experiment. The 690 nm species in fact is formed after an initial delay, which clearly indicates that this species did not form from the S1 state of BPEP directly. Our conclusion is that data obtained from the picosecond absorption studies is insufficient to identify the 690 nm absorbing species. 3.3.2. Nanosecond Flash Photolysis. Results of nanosecond flash photolysis experiments performed on BPEP using 355 nm (Nd:YAG, third harmonic) light is shown in Figure 5. The transient absorption exhibited a maximum around 690 nm and bleaching below 450 nm. This species appears to be the same as that observed at longer time scales (850 ps to 2.3 ns) in the picosecond flash photolysis of BPEP (Figure 4). The transient spectrum in Figure 5 was insensitive to oxygen. Nanosecond laser flash photolysis in the presence of oxygen also gave same spectrum. Kinetic traces at 690 nm obtained under argonsaturated and oxygen-saturated conditions are shown as insets in Figure 5, and these are very similar. It is well-known that the transient absorptions due to triplet excited states will be quenched by molecular oxygen. Since the transient absorption observed in the nanosecond time scale is unaffected by oxygen, we can conclude that this absorption is not due to 3BPEP*. We also conclude that this absorption did not arise from a triplet precursor.

18730

J. Phys. Chem. C, Vol. 114, No. 43, 2010

Figure 6. Transient absorption spectra obtained upon excitation of [Ru(bpy)3] · 2PF6 in the presence of BPEP (a) 3, (b) 8, and (c) 14 µs after the laser excitation. The inset shows the kinetic trace at 600 nm.

The kinetic traces shown in Figure 5 exhibited considerable amount of residual absorption indicating formation of stable products. We observed that continuous irradiation of BPEP with 355 nm laser light leads to considerable changes in its absorption spectrum. Initially the absorption in the 300-450 nm region decreases with formation of a weak absorption band in the 600-800 nm region (see Supporting Information Figures S4 and S5). Continued irradiation leads to decrease in the 600-800 nm band and formation of products absorbing at all wavelengths below 800 nm. The 690 nm absorption we see in the laser flash photolysis could be the same as the 600-800 nm absorption seen in the steady-state irradiation, and this can be attributed to intermediates of the photoreaction. We have not made detailed attempts to isolate or identify this intermediate or final products because our intention here is only to show that direct excitation of BPEP will not result in the formation of 3BPEP* and the 690 nm species seen in the flash photolysis of BPEP is not its triplet. The fluorescence quantum yield of BPEP is very high at 93%. The maximum ISC yield would be 7%, and hence, good transient absorption attributable to 3BPEP* was actually not expected upon direct excitation. The 3BPEP* can in principle be obtained through sensitization experiments. We have tried to generate 3BPEP* by energy transfer from the metal-to-ligand charge-transfer (MLCT) state of [Ru(bpy)3]2+. A solution of [Ru(bpy)3] · 2PF6 in DCM was flash photolyzed in the presence of excess BPEP using the 532 nm light from a Nd:YAG laser. Under the experimental conditions all the light is absorbed by [Ru(bpy)3]2+ (BPEP has no absorption at 532 nm) to form its MLCT triplet state which undergo energy transfer to BPEP to generate 3BPEP*. The transient absorption spectrum of Ru(bpy)32+ is well-characterized by a strong absorption around 370 nm, bleaching around 450 nm, and emission from the MLCT state in the 600-750 nm region. In the presence of BPEP the above transient absorptions and bleaching disappear, and the new spectrum shown in Figure 6 is obtained. The transient absorption exhibited maximum at 600 nm and decayed without leaving any residual absorption. Since the transient was formed from a triplet state, we assign it to 3BPEP*. We also attempted flash photolysis of BPEP in DCM containing 5% iodoethane to see if heavy atom induced ISC can generate 3BPEP. This experiment was not successful. It may be noted that the transient absorption spectrum obtained in the sensitization experiment (Figure 6) is different from the one obtained by direct excitation of BPEP (Figure 5). The lifetimes of the transients were also different. All these observations suggested that the transient species observed at the nanosecond time scale in the direct excitation of BPEP is

Suneesh and Gopidas

Figure 7. Picosecond transient absorption spectra obtained in the flash photolysis of BPEP-PT in DCM: (a) 40, (b) 200, and (c) 500 ps. The inset shows decay of the transient at 730 nm and growth at 600 nm.

Figure 8. Transient absorption spectrum of BPEP-PT immediately after the laser pulse. The insets show decay profiles at 515 and 600 nm.

not due to 3BPEP*. Therefore, we conclude that ISC to the triplet state is not a significant process in the photophysics of BPEP. We can also conclude that the 3BPEP* generated through sensitization does not initiate any photoreaction, and hence, the photoproduct formation seen in Supporting Information Figures S4 and S5 does not involve triplet-derived intermediates. Since 3BPEP* can be obtained readily by energy transfer from 3[Ru(bpy)3]*, the energy of the 3BPEP* state must be lower than that of the 3[Ru(bpy)3]* state. This puts an upper limit for the 3BPEP* state at 2.1 eV. 3.4. Flash Photolysis of Dyad BPEP-PT. 3.4.1. Picosecond Flash Photolysis. Figure 7 shows the transient absorption spectra obtained in the picosecond flash photolysis of BPEP-PT. At short time scales the absorption due to the 730 nm species (which we assign to the S1 f Sn absorption of BPEP in Figure 4) is dominant. This absorption decays within the observation window with concomitant formation of new absorptions at 600 and 515 nm. The inset of Figure 7 shows the decay at 730 nm and matching growth at 600 nm, indicating that the 600 nm species is formed directly from the 1S* of BPEP. Since the number of data points was limited, the decay and growth rates could not be calculated accurately from the inset in Figure 7. Since the enhanced fluorescence decay in BPEP-PT is attributed to electron transfer from PT to 1BPEP* (eq 5), the transient absorptions arising from the decay of the 1BPEP* absorption can be attributed to products of this electron-transfer reaction. 3.4.2. Nanosecond Flash Photolysis. BPEP-PT was also subjected to nanosecond flash photolysis in deaerated DCM, and the transient spectrum obtained is shown in Figure 8. The transient absorption spectrum exhibited maxima at 515 and 600 nm. Under deaerated conditions decays of both the absorptions were similar and exhibited a lifetime of 30 µs (see insets in Figure 8). According to eqs 4 and 5, excitation of the BPEP chromophore in BPEP-PT can initiate electron transfer from

1,6(Bisphenylethynyl)pyrene-Phenothiazine Dyad

Figure 9. Transient absorption spectrum of BPEP-PT under oxygensaturated condition immediately after the laser pulse. The insets show decay profiles at 515 and 600 nm.

J. Phys. Chem. C, Vol. 114, No. 43, 2010 18731

Figure 10. Transient absorption spectra of BPEP-PT in the presence of TMPA (2 × 10-4 M) in deaerated DCM solution (a) 1.5, (b) 4, and (c) 8 µs after excitation.

SCHEME 3

PT to 1BPEP* leading to formation of •-BPEP-PT•+. A large number of studies are available with PT as the electron donor, and the PT•+ is known to have an absorption maximum around 515 nm.43-46 BPEP•- is not known in the literature. To the best of our knowledge, there are no reports on the photochemical or electrochemical reduction of bis(phenylethynyl)pyrene in the literature. Our attempts to generate BPEP•- through spectroelectrochemistry were also not successful. Since electron transfer from PT to 1BPEP* is implicated and the 515 nm absorption can be assigned to PT•+, the 600 nm absorption is assigned to BPEP•-. In fact, picosecond transient absorption studies also indicated formation of transient absorptions at 515 and 600 nm directly from the singlet excited state of the BPEP moiety (Figure 7). The results indicated that the same transients are seen in the picosecond-microsecond time scale. These transients are very long-lived (τ ) 30 µs) and decay by the BET process, for which we obtained a rate constant (kBET ) 1/τ) of 3.34 × 104 s-1. In order to further confirm the assignment of the transient absorptions, we have carried out the nanosecond flash photolysis of BPEP-PT in the presence of oxygen. It is well-accepted that radical anions are quenched efficiently by oxygen, whereas radical cations are relatively unaffected, and hence, flash photolysis in the presence of oxygen can be used to identify the radical ion species. The transient absorption spectrum obtained in the presence of oxygen along with the kinetic traces obtained at 515 and 600 nm are shown in Figure 9. It can be seen that the decay of the 515 nm transient has become somewhat slower and the 600 nm transient disappeared completely in the presence of oxygen. This result suggested that the absorptions at 515 and 600 nm are due to different species. Disappearance of the 600 nm transient in the presence of oxygen confirms our assignment of this species to a radical anion. Previously we reported that N-methylphenothiazine radical cation can be prepared in a stable form by treating Nmethylphenothiazine with Cu(ClO4)2 in acetonitrile,46 and the spectrum of the radical cation obtained (see the Supporting Information) was nearly identical to that shown in Figure 9. It is to be mentioned here that laser irradiation in the absence of oxygen did not lead to any decomposition in BPEP-PT. Absorption spectra of BPEP-PT (3 mL, 10-4 M) were taken before and after laser irradiation (∼100 laser shots), and identical spectra were obtained. This is in contrast to the model BPEP, which exhibited significant decomposition under these conditions. In order to further confirm the assignment of transient absorptions in Figure 8 to the CS state of the BPEP-PT dyad, we have carried out secondary electron-transfer experiments. When the CS state is long-lived, it can be involved in exergonic electron-transfer reactions with better donors or acceptors. We

have carried out this experiment using tris(4-methoxyphenyl)amine (TMPA), which is a better electron donor than PT. A solution of BPEP-PT was flash photolyzed in the presence of TMPA in DCM solution, and the transient spectra obtained at different times following the laser flash are shown in Figure 10. The spectrum at short time scales is similar to that in Figure 8 and is assigned to •-BPEP-PT•+. At longer times the absorption due to PT•+ at 515 nm decays along with formation of absorption at 715 nm, which is assigned to TMPA•+.47 The inset shows the decay of the transient at 515 nm and formation of absorption at 720 nm. Since TMPA is a better electron donor (Eox ) 0.57 V vs SCE in DCM) compared to PT (Eox ) 0.75 V vs SCE), secondary electron transfer as shown in eq 9 is expected to take place. From the oxidation potentials of the PT and TMPA moieties, ∆G0 for reaction 9 was obtained as -0.18 eV. •-

ET

BPEP - PT•+ + TMPA f •-BPEP - PT + TMPA•+

(9) This experiment further confirmed the assignment of the transients in Figure 8 to the long-lived CS state of BPEP-PT. Since PET occurs in the singlet state, •-BPEP-PT•+ is formed with overall singlet multiplicity. On the basis of electrochemical studies energy of this state is placed at 2.4 eV. From the sensitization experiments we arrived at an upper limit for the triplet state of BPEP at 2.1 eV. Since a 3LE state is placed below the 1CS state, it is energetically possible for the 1 CS state to decay to the 3LE state through the HFI or SOCTISC pathways as described earlier and shown in Scheme 3. If any of these mechanisms were operating we would have ended up with 3BPEP*-PT, which exhibited the absorption spectrum shown in Figure 6. The HFI mechanism may not be operating in this system because of the high value of |2J|. Previous reports suggested that SOCT-ISC can be enhanced by the use of heavy atom effects.30 In order to see if the heavy atom effect can be used to induce SOCT-ISC in BPEP-PT we have carried out the flash photolysis in the presence of iodoethane. We observed that the transients (absorbing at 515

18732

J. Phys. Chem. C, Vol. 114, No. 43, 2010

Suneesh and Gopidas

Figure 11. Kinetic traces obtained on laser flash photolysis of BPEP-PT in the presence of (a) 0%, (b) 2%, and (c) 5% iodoethane in DCM and (d) nanosecond transient absorption spectrum of BPEP in the presence of 5% iodoethane in DCM obtained immediately after laser irradiation.

and 600 nm in Figure 8) are quenched by the addition of small quantities of iodoethane (Figure 11). Upon addition of small quantities of iodoethane the lifetime of the transient at 600 nm decreased with an increase in ∆OD, until the solution contained 5% iodoethane (Figure 11a-c). The spectrum of the transient observed in the presence of 5% iodoethane immediately after the laser pulse is shown in Figure 11d. A comparison of Figure 11d with Figure 6 confirms that the species obtained is 3 BPEP*-PT. Formation of 3BPEP*-PT from •-BPEP-PT•+ in the above experiment confirms a few important conclusions of this study. (1) 3BPEP*-PT lies at a lower energy level compared to the 1 CS state, (2) neither HFI nor SOCT-ISC is an important pathway for the decay of 1CS, and what we are observing is actually 1CS and not 3CS, (3) SOCT-ISC can be induced in 1 CS using the heavy atom effect to generate the 3LE state, and most importantly (4) it is possible to have a 3LE level below a long-lived 1CS state, without the latter getting converted into the former. This experiment also confirms that the model presented in Scheme 1 is practical. Long-lived photoinduced charge separation in dyads is a controversial subject, and after examining several systems Harriman and Verhoeven suggested that a CS state can be classified as “unusually long-lived” if the ratio of the charge separation and charge recombination rate constants exceed 5000.28 These authors also suggested certain criteria that must be fulfilled before this classification can be made. We would like to confirm here that most of the conditions laid down by Harriman and Verhoeven are fulfilled in the case of BPEP-PT. For example, we have established 3LE is not formed in the flash photolysis of BPEP-PT. The triplet state of BPEP was generated and identified through sensitization experiments or under the influence of the heavy atom effect, and this species is not observed in the picosecond-microsecond time window. The PT triplet state is known in the literature, and this species is also not detected in the flash photolysis experiments.48,49 The transient formed in the picosecond time window and that observed in the nanosecond time window are similar. We further showed that the transient spectrum observed in the nanosecond

SCHEME 4: Schematic Representation of Photophysical Processes in BPEP-PT

time window actually is due to two species, one of which is unambiguously identified as PT•+, and by virtue of eq 5, the other species can be identified as BPEP•-. Thus, from the flash photolysis experiments we conclude that irradiation of BPEP-PT results in the formation of a long-lived CS state identified as •-BPEP-PT•+. Heavy atoms induce formation of the 3LE state, which further confirms that the long-lived CS state has singlet character. For this dyad kPET and kBET were 6.38 × 109 and 3.34 × 104 s-1, respectively, resulting in kPET/ kBET ) 1.9 × 105. Since this value is much larger than 5000, BPEP-PT can be identified as dyad with an unusually longlived CS state. 3.5. Energy Level Diagram for BPEP-PT. The energy level diagram of BPEP-PT, constructed on the basis of available information, is given in Scheme 4. The S1 state in BPEP-PT is placed at 2.86 eV, based on the positions of the absorption and emission bands. Electrochemical studies have placed the energy of the CS state in BPEP-PT at 2.4 eV. The reported value of triplet energy in nonpolar solvents for PT is 2.62 eV.50 For BPEP the triplet energy level is not known in the literature. Our attempts to obtain a phosphorescence

1,6(Bisphenylethynyl)pyrene-Phenothiazine Dyad spectrum for BPEP also failed. We generated the BPEP triplet by sensitization experiments using [Ru(bpy)3]2+, which indicated that the triplet energy of BPEP is lower than that of [Ru(bpy)3]2+. Energy of the MLCT state of [Ru(bpy)3]2+ is 2.1 eV.50 In Scheme 4 we placed 3BPEP* at this value, which is a higher limit. Since the electron transfer occurs in the singlet manifold, the CS state formed is also in the singlet state. As mentioned earlier interconversion between 1CS and 3CS is not occurring in BPEP-PT most probably because of a large |2J| value. If we assume that energy difference between 1CS and 3CS is ≈ 0.1 eV () 806 cm-1), we can place the 3(-•BPEP-PT•+) state at 2.3 eV, which is 0.2 eV above the local triplet level 3 BPEP*-PT. Although excitation of BPEP-PT can lead to the formation of several high-energy states, evidence exists only for the formation of 1BPEP*-PT and -•BPEP-PT•+. Excitation of the BPEP chromophore leads to 1BPEP*-PT, where the excitation is localized on BPEP. The excited chromophore can undergo fluorescence decay to the ground state or accept an electron from PT to generate 1(-•BPEP-PT•+). Since the HFI and SOCT-ISC mechanisms for the interconversion between 1 CS and 3CS are inefficient the CS state remains in the singlet manifold. Energy transfer from 1BPEP*-PT to BPEP-1PT* will be endergonic and hence ruled out. Although the phenothiazine-based local triplet BPEP-3PT* lies 0.24 eV lower than 1 BPEP*-PT, singlet f triplet energy transfer is a forbidden process, and hence, formation of BPEP-3PT* is also ruled out. The BPEP chromophore did not exhibit any propensity to undergo ISC, and hence, we do not expect 1BPEP*-PT also to undergo ISC to 3BPEP*-PT. The only decay pathway available to the 1CS state of BPEP-PT is a spin-allowed BET to the ground state. The rate constant for this process was found to be very low, and according to eq 1 this can only be attributed to two reasons: (1) an extremely weak coupling Hel between the donor radical cation and acceptor radical anion and/or (2) the inverted region effects. We rule out the former possibility based on the following arguments. (1) The donor and acceptor moieties are separated only by three methylene groups, and at this distance the electronic coupling is expected to be good. For the PET process Hel can be calculated using observed value of ket in eq 1, and we obtained Hel ) 40 cm-1. There is no reason for Hel for the BET process to be several orders of magnitude smaller than this value. (2) Hel can depend on the relative orientation of D and A moieties. In this case, however, orientational effects are not expected to play any role because of the flexible nature of the linker unit. (3) Electronic coupling between donor radical cation and acceptor radical anion will also affect the SOCTISC process, and it may be argued that absence of SOCT-ISC in the 1(•+BPEP-PT•-) system is also due to poor electronic coupling. We observed that SOCT-ISC can be induced in 1 •+ ( BPEP-PT•-) by the external heavy atom effect, and it is accepted that the heavy atom effect is employed to circumvent the spin-forbidden nature of transitions and its dominant influence is to enhance spin-orbit coupling. We attribute the long lifetime of the CS state to inverted region effects. The inverted region effects operate when -∆G0 . λ. For the BET reaction between BPEP•- and PT•+ the ∆G0 value is highly negative at -2.4 eV. The reorganization energy λ is the sum of λo and λi. The outer-sphere reorganization energy λo depends on the radii of the donor (rD) and acceptor (rA), the center-to-center distance dcc, and the optical (εop) and static (εs) dielectric constants of the solvent as given in eq 10.

J. Phys. Chem. C, Vol. 114, No. 43, 2010 18733

λ0 ) ∆e2

(

)(

1 1 1 1 1 + 2rA 2rD dcc εop εs

)

(10)

Since the donor and acceptor moieties are not spherical, an approximate radius ) 1/2(length × breadth)1/2 was assumed. Using these values we obtained λo ) 0.93 eV in DCM. In the case of PT, oxidation to PT•+ leads to substantial changes that involve bending of the ring system, and hence, a high value of 0.27 eV was reported for λi.51 BPEP is a very rigid organic molecule for which we can assume a low value of 0.1 eV for λi. Thus, the total internal reorganization energy ) 0.37 eV and the sum of λo and λi for the BET process in •-BPEP-PT•+ would be 1.3 eV, which is approximately 1.0 eV lower than -∆G0, pushing the system into the deep inverted region (the only assumption here is about the λi value of BPEP. Even if this value is larger by 0.1 eV, -∆G0 would still be much larger than λ). As far as interconversion between 1CS and 3CS in BPEP-PT is concerned, we can invoke a large value of |2J| to explain the absence of the HFI mechanism. The total absence of the SOCTISC mechanism in 1CS was surprising, and an explanation for the same is not available at this time. We made a proposal in our earlier study that ISC is forbidden in the parent chromophore and this forbiddingness is carried over to the 1CS state also. This explanation seems to be valid in the present case also, but whether this conclusion can be generalized is yet to be established. The factors that influence ISC in the parent BPEP are very different from those in the 1CS state of the dyad. The energetics of the ISC processes are also vastly different in the two systems. The time scales of the competing processes in the respective states are also different. For example, in BPEP kF () ΦBPEP/τ0) ) 7.15 × 108 s-1. Since we do not observe ISC in BPEP we assume that kISC is at least 2 orders of magnitude slower than kF, suggesting that kISC < 7.15 × 106 s-1. In the 1CS state kBET ) 3.34 × 104 s-1, and since the SOCTISC is not occurring we assume that kSOCT-ISC < 3.34 × 102 s-1. Thus, the difference in the time scales of kISC in the parent chromophore and kSOCT-ISC in the 1CS state is very large, and hence, the question of whether one can translate the forbiddingness of the ISC process in the parent chromophore to the forbiddingness of the SOCT-ISC process in the 1CS state remains to be answered. However, our success with BPEAand BPEP-based systems encourages us to believe that our proposal can be a good starting point for designing dyads capable of exhibiting long-lived CS states. We are currently involved in the design of more dyad systems based on this proposal in order to establish its generality. 4. Conclusions In this paper we have examined the photoinduced electrontransfer processes in a bis(phenylethynyl)pyrene-phenothaiazine dyad, BPEP-PT. Absorption spectra of BPEP-PT indicated the absence of any ground-state interactions between BPEP and PT moieties. Fluorescence of the BPEP moiety is highly quenched in the dyad due to facile electron transfer from the PT unit. The rate constants and quantum yields of the PET process were determined from the steady-state and time-resolved fluorescence studies. Picosecond and nanosecond transient absorption studies suggested formation of a long-lived CS state in BPEP-PT. Oxygen quenching studies and secondary electron-transfer experiments confirmed the assignment of the long-lived transient to the CS state. The CS state underwent heavy atom induced SOCT-ISC to the local triplet state, which

18734

J. Phys. Chem. C, Vol. 114, No. 43, 2010

confirmed the singlet multiplicity of the CS state. The only decay pathway available to the 1CS state is a spin-allowed transition to the ground state, and the slow kinetics of this process is attributed to inverted region effects. On the basis of available data a scheme summarizing the dynamics and energetics of the probable photophysical processes in the dyad was constructed. This study confirms that long-lived charge separation attributable to inverted region effects is possible in D-A dyads. Acknowledgment. The authors thank the Council of Scientific and Industrial Research (CSIR-NWP 23) and the Department of Science and Technology (DST Grant No. SR/S5/OC15/2003), Government of India, for financial support. C.V.S. thanks CSIR for a research fellowship. The authors also thank Professor P. Ramamurthy, Director, National Centre for Ultra Fast Processes, Madras University, Chennai for helping with the picosecond transient absorption measurements. This is contribution NIIST-PPG 302. Supporting Information Available: Detailed synthetic procedures of the compounds, electrochemical data and fluorescence decay profiles of the compounds, radical cation spectra of N-methyl phenothiazine obtained by chemical method, and changes in the absorption spectrum of BPEP upon laser irradiation. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Photoinduced Electron Transfer, Parts A-D; Fox, M. A., Channon, M., Eds.; Elsevier: Amsterdam, The Netherlands, 1988. (2) Wasielewski, M. R. Acc. Chem. Res. 2009, 42, 1910–1921. (3) Balzani, V.; Credi, A.; Venturi, M. ChemSusChem 2008, 1, 26– 58. (4) Kamat, P. V. J. Phys. Chem. C 2007, 111, 2834–2860. (5) Benniston, A. C.; Harriman, A. Mater. Today 2008, 11, 26–34. (6) Fukuzumi, S. Phys. Chem. Chem. Phys. 2008, 10, 2283–2297. (7) Guldi, D. M. Chem. Soc. ReV. 2002, 31, 22–36. (8) Holten, D.; Bocian, D. F.; Lindsey, J. S. Acc. Chem. Res. 2002, 35, 57–69. (9) The Photosynthetic Reaction Center; Deisenhofer, J., Norris, J. R., Eds.; Academic Press: San Diego, CA, 1993. (10) Gust, D.; Moore, T. A.; Moore, A. L. Acc. Chem. Res. 2001, 34, 40–48. (11) Gust, D.; Moore, T. A.; Moore, A. L.; Lee, S. J.; Bittersmann, E.; Luttrull, D. K.; Rehms, A. A.; De Graziano, J. M.; Ma, X. C.; Gao, F.; Belford, R. E.; Trier, T. T. Science 1990, 248, 199–201. (12) Flamigni, L.; Baranoff, E.; Collin, J. P.; Sauvage, J. P. Chem.sEur. J. 2006, 12, 6592–6606. (13) Imahori, H.; Guldi, D. M.; Tamaki, K.; Yoshida, Y.; Luo, C.; Sakata, Y.; Fukuzumi, S. J. Am. Chem. Soc. 2001, 123, 6617–6628. (14) Wasielewski, M. R.; Gaines, G. L., III; Wiederrecht, G. P.; Svec, W. A.; Niemczyk, M. P. J. Am. Chem. Soc. 1993, 115, 10442–10443. (15) Fukuzumi, S. Bull. Chem. Soc. Jpn. 2006, 79, 177–195. (16) Geiss, B.; Lambert, C. Chem. Commun. 2009, 1670–1672. (17) Hauke, F.; Atalick, S.; Guldi, D. M.; Hirsch, A. Tetrahedron 2006, 62, 1923–1927. (18) Wan, J.; Ferreira, A.; Xia, W.; Chow, C. H.; Takechi, K.; Kamat, P. V.; Jones, G., II; Vullev, V. I. J. Photochem. Photobiol., A 2008, 197, 364–374. (19) Ohkubo, K.; Imahori, H.; Shao, J.; Ou, Z.; Kadish, K. M.; Chen, Y.; Zheng, G.; Pandey, R. K.; Fujitsuka, M.; Ito, O.; Fukuzumi, S. J. Phys. Chem. A 2002, 106, 10991–10998.

Suneesh and Gopidas (20) Kashiwagi, Y.; Ohkubo, K.; McDonald, J. A.; Blake, I. M.; Crossley, M. J.; Araki, Y.; Ito, O.; Imahori, H.; Fukuzumi, S. Org. Lett. 2003, 5, 2719–2721. (21) Ohkubo, K.; Kotani, H.; Shao, J.; Ou, Z.; Kadish, K. M.; Li, G.; Pandey, R. K.; Fujitsuka, M.; Ito, O.; Imahori, H.; Fukuzumi, S. Angew. Chem., Int. Ed. 2004, 43, 853–856. (22) Fukuzumi, S.; Kotani, H.; Ohkubo, K.; Ogo, S.; Tkachenko, N. V.; Lemmetyinen, H. J. Am. Chem. Soc. 2004, 126, 1600–1601. (23) Okamoto, K.; Hasobe, T.; Tkachenko, N. V.; Lemmetyinen, H.; Kamat, P. V.; Fukuzumi, S. J. Phys. Chem. A 2005, 109, 4662–4670. (24) Marcus, R. A.; Sutin, N. Biochem. Biophys. Acta 1985, 811, 265– 322. (25) Benniston, A. C.; Harriman, A.; Li, P.; Rostron, J. P.; Verhoeven, J. W. Chem. Commun. 2005, 2701–2703. (26) Benniston, A. C.; Harriman, A.; Li, P.; Rostron, J. P.; Ramesdonk, H. J.; Groeneveld, M. M.; Zhang, H.; Verhoeven, J. W. J. Am. Chem. Soc. 2005, 127, 16054–16064. (27) Verhoeven, J.; van Ramesdonk, H. J.; Zhang, H.; Groeneveld, M. M.; Benniston, A. C.; Harriman, A. Int. J. Photoenergy 2005, 7, 103– 108. (28) van Ramesdonk, H. J.; Bakker, B. H.; Groeneveld, M. M.; Verhoeven, J. W.; Allen, B. D.; Rostron, J. P.; Harriman, A. J. Phys. Chem. A 2006, 110, 13145–13150. (29) Benniston, A. C.; Harriman, A.; Verhoeven, J. W. Phys. Chem. Chem. Phys. 2008, 10, 5156–5158. (30) Verhoeven, J. W.; van Ramesdonk, H. J.; Groeneveld, M. M.; Benniston, A. C.; Harriman, A. ChemPhysChem 2005, 6, 2251–2260. (31) Prasad, E.; Gopidas, K. R. J. Am. Chem. Soc. 2000, 122, 3191– 3196. (32) Smitha, M. A.; Prasad, E.; Gopidas, K. R. J. Am. Chem. Soc. 2001, 123, 1159–1165. (33) Balan, B.; Gopidas, K. R. Chem.sEur. J. 2006, 12, 6701–6710. (34) Balan, B.; Gopidas, K. R. Chem.sEur. J. 2007, 13, 5173–5185. (35) Hasharoni, K.; Levanon, H.; Greenfield, S. R.; Gosztola, D. J.; Svec, W. A.; Wasielewski, M. R. J. Am. Chem. Soc. 1995, 117, 8055–8056. (36) Brun, A. M.; Harriman, A.; Tsuboi, Y.; Okada, T.; Mataga, N. J. Chem. Soc., Faraday Trans. 1995, 91, 4047–4057. (37) Suneesh, C. V.; Gopidas, K. R. J. Phys. Chem. C 2009, 113, 1606– 1614. (38) Verhoeven, J. W. J. Photochem. Photobiol., C 2006, 7, 40–60. (39) Dance, Z. E. X.; Mickley, S. M.; Wilson, T. M.; Ricks, A. B.; Scott, A. M.; Ratner, M. A.; Wasielewski, M. R. J. Phys. Chem. A 2008, 112, 4194–4201. (40) Okada, T.; Karaki, I.; Matsuzawa, E.; Mataga, N.; Sakata, Y.; Misumi, S. J. Phys. Chem. 1981, 85, 3957–3960. (41) Morris, J. V.; Mahaney, M. A.; Huber, J. R. J. Phys. Chem. 1976, 80, 969–974. (42) Rehm, D.; Weller, A. Ber. Bunsen-Ges. Phys. Chem. 1969, 73, 834– 839. (43) Moroi, M.; Braun, A. M.; Gra¨tzel, M. J. Am. Chem. Soc. 1979, 101, 567–572. (44) Klumpp, T.; Linsenmann, M.; Larson, S. L.; Limoges, B. R.; Bu¨rssner, D.; Krissinel, E. B.; Elliott, C. M.; Steiner, U. E. J. Am. Chem. Soc. 1999, 121, 1076–1087. (45) Ajayakumar, G.; Gopidas, K. R. Photochem. Photobiol. Sci. 2008, 7, 826–833. (46) Sumalekshmy, S.; Gopidas, K. R. Chem. Phys. Lett. 2005, 413, 294–299. (47) Gould, I. R.; Ege, D.; Moser, J. E.; Farid, S. J. Am. Chem. Soc. 1990, 112, 4290–4301. (48) Sarata, G.; Noda, Y.; Sakai, M.; Takahashi, H. J. Mol. Struct. 1997, 413-414, 49–59. (49) Sakaguchi, Y.; Hayashi, H. J. Phys. Chem. A 1997, 101, 549. (50) Murov, S. L.; Carmichael, I.; Hug, G. L. In Handbook of Photochemistry, 2nd ed.; Marcel Dekker: New York, 1993. (51) Borowicz, P.; Herbich, J.; Kapturkiewicz, A.; Opallo, M.; Nowacki, J. Chem. Phys. 1999, 249, 49–62.

JP107606T