Photophysical and Theoretical Insights on Fullerene

Article pubs.acs.org/JPCB

Photophysical and Theoretical Insights on Fullerene/ Zincphthalocyanine Supramolecular Interaction in Solution Anamika Ray,† Kotni Santhosh,‡ and Sumanta Bhattacharya*,† †

Department of Chemistry, The University of Burdwan, Golapbag, Burdwan - 713 104, India School of Chemistry, University of Hyderabad, Hyderabad, AP - 500 046, India

‡

S Supporting Information *

ABSTRACT: The present article reports photophysical studies on supramolecular interaction of a zinc phthalocyanine derivative, namely, zinc-2,9,16,23-tetra-tert-butyl phthalocyanine (1) with C60 and C70 in solvents having varying polarity, i.e., toluene and 1,2-dichlorobenzene (DCB). The interesting feature of the present work is the observation of charge transfer (CT) absorption bands of the fullerene/1 complexes in DCB. Utilizing the CT transition energy, many important physicochemical parameters like vertical ionization potential of 1, degrees of CT, oscillator strength, transition dipole moment, and resonance energy of interaction have been determined in the present case. The influences of 1 on the UV−vis spectral characteristics of C60 and C70 have been explained using a theoretical model that takes into account the interaction between electronic subsystems of 1 with fullerenes. Steady state fluorescence experiment elicits efficient quenching of the fluorescence intensity of 1 in the presence of both C60 and C70. The average binding constants of the C60 and C70 complexes of 1 (estimated by UV−vis and steady state fluorescence measurements) are determined to be 18 330 dm3·mol−1 (12 595 dm3·mol−1) and 19 160 dm3·mol−1 (15 292 dm3·mol−1) in toluene (DCB), respectively. Lifetime experiment yields a larger magnitude of charge separated rate constant for the C70/1 species. The faster charge recombination of the fullerene/1 systems observed in more polar solvent results from solvent reorganization energies. Quantum chemical calculations by the ab initio method explore the geometry and electronic structure of the supramolecules and testify the significant redistribution of charge between fullerenes and 1 during fullerene/1 interaction. A variable temperature 13C NMR study nicely demonstrates that the end-on orientation of C70 is very much responsible for the low selectivity in binding between C60/1 and C70/1 systems. Free energy of charge recombination and free energy of radical ion-pair formation signify that electron transfer from the excited 1 to C60 and C70 in the C60/1 and C70/1 complexes, respectively, is an unlikely process. Finally, transient absorption measurements in the visible region establish that energy transfer from TC60* (and TC70*) to 1 occurs predominantly in both toluene and DCB, which is subsequently confirmed by the consecutive appearance of the triplet state of 1.

1. INTRODUCTION Metallophthalocyanines (MPc's) have been successfully tested as photo sensitizers for photodynamic therapy.1−3 Photophysical properties of these dyes are strongly dependent on the central metal ion. Among different metals, zinc provides phthalocyanine (Pc) valuable fluorescence and singlet oxygen production properties.4,5 Efforts have been made to improve the selectivity and specificity of Pc toward different targets through variation of peripheral substitution on the macrocyclic ring.6,7 Pc exhibits particularly intense absorption characteristics in the red/near-infrared spectral region, where porphyrins, however, fail to absorb appreciably.8 The unique photophysical and electrochemical properties of Pc have led to their application as electron-donor or electron-acceptor moieties in multicomponent systems for energy and charge transfer processes.9 As a consequence of their outstanding lightharvesting properties combined with their redox features, phthalocyanines (Pc's) are most commonly encountered as the donor units in such photoactive molecules.10 Among MPc's, zinc phthalocyanine (ZnPc) has received special attention. The initial motivation is that ZnPc is ideally suited for the © 2012 American Chemical Society

characterization of the excited states of the Pc ring. Due to the d10 configuration of the central Zn2+ ion, the optical spectra of ZnPc complexes are not complicated indeed by the additional metal to ligand charge transfer and ligand to metal charge transfer bands that appear in the spectra of other transition metal phthalocyanines; however, the D4h symmetry of MPc's is retained. It has already been verified that the presence of the central metal ion along with peripheral and nonperipheral substitution strongly influences the photophysical behavior of Pc molecules.11 Diamaganetic metal such as Zn extends the triplet lifetime and thus can enhance the photoactivity of dye,12 whereas metal-free Pc's are rather inactive in photoconversion as observed by several researchers in recent past.13−15 Since the influence of different central metals (diamagnetic or paramagnetic) on both the radiative and nonradiative deactivation and on the intersystem quantum yields has been investigated rather intensely in the absence and Received: May 30, 2012 Revised: August 17, 2012 Published: August 21, 2012 11979

dx.doi.org/10.1021/jp3052483 | J. Phys. Chem. B 2012, 116, 11979−11998

The Journal of Physical Chemistry B

Article

presence of fullerenes,16−20 we have concentrated our focus on the noncovalent interaction between fullerenes and the Pc containing zinc as the central metal atom, i.e., 1, in our present investigations. After the initial discovery in 1984,21 the fortuitous contemporary growth of two apparently independent research lines, namely, synthetic fullerene chemistry and supramolecular fullerene photochemistry, has been reciprocally beneficial and contributed to boost activity in both fields. However, the formation of multicomponent “supermolecules” acting as artificial photosynthetic centers represents one of the most active research areas in fullerene science for the last 15 years.22 While this area of research is broad, the addition of novel building blocks such as Pc has led to numerous breakthroughs.23 The first report on fullerene/Pc arrays dates back to late 1995, when these two redox-active moieties were covalently linked following different synthetic strategies to form fullerene/ Pc hybrids.24 Fullerenes have been linked to Pc's as a consequence of their ability to delocalize charges over an extended spherical π-surface, broad absorption band,25a and low reorganization energy.25b In recent past, some covalently and noncovalently linked arrays comprised of fullerenes and Pc's have been prepared,26−28 some of which exhibit photoinduced electron transfer.29 An interesting aspect of the chemistry of fullerenes and Pc's is that they are spontaneously attracted to each other, as a result of ground state complexation.28,29 Our rationale, which is based on molecular interaction of fullerenes with a designed ZnPc, namely, 1 (Figure 1a), in solution

noncovalent interaction between fullerenes (C60 and C70) and 1 by means of various spectroscopic tools, like UV−vis, steady state, and time-resolved fluorescence, transient absorption studies, and variable temperature 13C NMR measurements supported by quantum chemical calculations at ab initio levels of theory. We anticipate that molecule 1 may impart some novel photophysical characteristics during fullerene/1 noncovalent interaction.

2. MATERIALS AND METHODS Both C60 and C70 were purchased from Aldrich, USA. Compound 1 is purchased from Aldrich, USA, in very pure form (dye content ∼96%). The compound is obtained as a single isomer. HPLC grade toluene and DCB (Spectrochem, India) have been used as solvents to favor noncovalent interaction between fullerene and 1 and, at the same time, to ensure good solubility and photostability of the samples. UV− vis spectral measurements are performed on a Shimadzu UV2450 model spectrophotometer using a quartz cell with 1 cm optical path length. Emission spectra have been recorded with a steady state Hitachi F-4500 model spectrofluorimeter. Fluorescence decay curves are measured with a HORIBA Jobin Yvon single photon counting setup employing a nanoled as the excitation source. C60 and C70 are selectively excited by 532 nm light from a Nd:YAG laser (6 ns fwhm) with 7 mJ power. For the transient absorption spectra in the visible region, a photomultiplier tube has been used as a detector for the pulsed Xe-monitoring light (150 W). Electrochemical experiments are performed with a BAS potentiostat. A glassy carbon electrode is used as the working electrode, a platinum wire is used as the counter electrode, and a Ag/AgCl electrode is used as the reference electrode. The Fc/Fc+ couple has been used to standardize the reference. Tetrabutyl ammonium hexafluorophosphate (0.5 mol·dm−3) is used as a supporting electrolyte. Theoretical calculations are performed with a Pentium IV computer using SPARTAN’06 V1.1.0 (USA) Windows version software. 13C NMR measurements are done in a 200 MHz Bruker NMR spectrometer. 3. RESULTS AND DISCUSSION 3.1. UV−vis Absorption Studies. The extensively conjugated aromatic chromophoric system of Pc generates intense bands in its absorption spectrum. It is already well established that MPc's are characterized by their electronic absorption with high extinction coefficients in the visible region, namely, Q-band, and weaker absorption in the lower wavelength region, i.e., Soret absorption band (or B band), resulting from S1 ← S0 and S2 ← S0 transitions, respectively.30 The stronger and the most well-resolved absorption band of 1 is detected in the visible region, ranging from 350 to 685 nm (Figure 2). Generally, in the case of Pc, the four fused benzene rings break the accidental degeneracy of the top-filled molecular orbitals as well. Because of the influence of the configuration interactions, only the 1a1u → 1eg* electronic transition is responsible for the generation of the Q-band.30a In our present case, peripheral substituents play an important role in the tuning of the absorption bands compared to unsubstituted ZnPc (1a, Figure 1b).31 The bonding of the alkyl chain to the Pc ring system influences the effectiveness of thermal conversion which may be shown by photo acoustic and absorption results.32 The rigidity of the molecular skeleton of Pc can be strongly affected by the kind of peripheral groups

Figure 1. Structures of (a) 1 and (b) unsubstituted ZnPc.

addresses this issue. Additionally, we would like to concentrate our attention on the measurements of various photophysical properties and the binding strength of fullerene/1 complexes in solvents having varying polarity. Tuning the solvent parameter would certainly enhance the possibility of finding new physicochemical insights regarding the nature of electronic perturbation in the donor as well as acceptor component of the fullerene/1 ensemble at ground state. The purpose of the present work is, therefore, to investigate some important physicochemical properties as well as to measure the extent of 11980



Article

In DCB, for the uncomplexed 1, the red shift in the wavelength maximum of the Q-band is observed compared to toluene (Figure 2b); 612 nm (toluene) to 616 nm (DCB), 650 nm (toluene) to 652 nm (DCB), and 678 nm (toluene) to 682 nm (DCB)). The peak maximum of the Soret absorption band remains unchanged in both toluene and DCB. Figure 2S in the Supporting Information and Figure 4a show the electronic absorption spectra of DCB solution of 1 with C60 and C70, respectively. Spectra of the above solutions have been recorded against the solvent as reference. The notable feature of such an experiment is that we have observed a very good isosbestic point (at 664 and 698 nm) in the corresponding UV−vis spectra of the mixture of C60 and 1 (Figure 4a). In the case of the C70/1 system, the isosbestic point does not appear well (Figure 2S, Supporting Information). In DCB, the systematic decrease in the absorbance value of the Q-absorption band, i.e., 682 nm, is monitored in connection with the estimation of K for the C60/1 and C70/1 systems. Excellent linear BH plots are obtained for both of the systems studied in the present investigations, and they are shown in Figure 4b and Figure 2S in the Supporting Information, respectively. Values of K are listed in Table 1. The most notable and interesting feature of the UV−vis experiment for the fullerene/1 complexation process in DCB is the observation of new absorption peaks for both the C60/1 and C70/1 systems, as depicted in Figure 4a and Figure 2S in the Supporting Information, respectively. The newly appeared bands do not exhibit any resemblance to the spectral characteristics of the parent donor and acceptor molecules. These peaks may be ascribed as charge transfer (CT) absorption peaks. Other than fullerenes, we have repeated the same experiment with different well-known electron acceptors, like o-chloranil, p-chloranil, 2,3-dichloro5,6-dicyano-p-benzoquinone, and 2,3,5,6-tetracyanoquinodimethane (TCNQ) in DCB. In all cases, very good and well resolved CT peaks are identified in the visible region of electronic spectra. CT peaks for the electron donor−acceptor (EDA) complexes of 1 with various electron acceptors are demonstrated in Figure 3S (Supporting Information). In Figures 2S and 3S (Supporting Information) and Figure 4a, the CT peaks are identified to be the peaks in the longest region of the visible spectra, because they normally have the longest wavelength among the peaks different from those obtained from the spectra of the components. It should be noted at this point that Imahori et al. have also noticed a broad and weak CT band for their particular designed supramolecular system, namely, ZnP−O34−C60 in benzene around 650−800 nm (λmax = 703 nm).34 In benzonitrile, the above CT band is slightly red-shifted (∼3 nm) relative to that of benzene.34 One important point to note is that, even though fullerenes and 1 undergo CT interaction with great ease, one may still argue that the way they approach each other is dictated by packing consideration rather than by the special affinity we have highlighted above. The choice of the polar solvent like DCB undoubtedly facilitates the possibility of CT phenomenon between hydrophobic fullerene and 1. The demonstration of fullerene/1 binding in DCB is, therefore, considered to be a spontaneous attraction. This indicates that the close contact between the ZnPc plane and the fullerene moiety achieved only in DCB is an essential criterion for observing the CT spectrum. The CT absorption spectra are analyzed by fitting to the Gaussian function y = y0 + [A/(w√(π/2)] exp[−2(x − xc)2/ w2], where x and y denote the wavenumber and molar extinction coefficient, respectively. One typical Gaussian

Figure 2. UV−vis absorption spectrum of 1 recorded in (a) toluene and (b) DCB against the solvent as reference; the concentration of 1 is fixed at 3.50 × 10−6 mol·dm−3 and 3.50 × 10−6 mol·dm−3 in toluene and DCB, respectively.

attached to the main core. Moreover, it is observed that the electron donating substituents (here tert-butyl group) at the peripheral position increase the electron density in the ring of 1 which causes a 6−8 nm red shift of the Q-band (i.e., 606 (1a) to 612 nm (1), 642 (1a) to 650 nm (1), and 671 (1a) to 678 nm (1)) and 14 nm red shift for the Soret absorption band (i.e., from 336 nm (1a) to 350 nm (1)).21−31 Evidence in favor of ground state interaction between fullerenes and 1 first comes from UV−vis spectroscopic measurement in toluene. Spectra of the above solutions have been recorded against the solvent as a reference. Addition of varying concentrations of C60 and C70 solutions (in toluene medium) to 1 (fixed concentration) produces a very good change in the absorbance value of the fullerene solutions. It is observed that the absorbance value of the mixture of fullerene/1 solution decreases systematically with the increasing amount of addition of fullerene (Figure 3a and Figure 1S in the Supporting Information for the C60/1 and C70/1 systems, respectively) in the Soret absorption band of 1. On the contrary, a systematic increase in the absorbance value is observed for the broad 400−700 nm absorption band of fullerene solution (resulting from a forbidden singlet−singlet transition in the case of C60− and C70),25a resulting from the molecular complex formation between fullerenes and 1. The solvent toluene does not absorb in the visible region. K values of the fullerene/1 systems have been determined in accordance with the Benesi−Hildebrand (BH) equation33 for cells with 1 cm optical path length. Excellent linear BH plots are obtained for both of the systems studied in the present investigations. Typical BH plots for the C60/1 and C70/1 systems in toluene are shown in Figure 3b and Figure 1S in the Supporting Information, respectively. Values of K are listed in Table 1. 11981



Article

Figure 3. (a) UV−vis absorption spectrum of (i) uncomplexed 1 (1.30 × 10−6 mol·dm−3), (ii) uncomplexed C60 (7.72 × 10−5 mol·dm−3), and (iii− vii) a mixture of C60 and 1 in which the C60 concentration varies in the range of 7.72 × 10−5 to 2.09 × 10−4 mol·dm−3 and the concentration of 1 is kept fixed at 1.30 × 10−6 mol·dm−3. (b) BH plot of the C60/1 system in toluene.

Table 1. Binding Constant (K) Determined by UV−vis and Steady State Fluorescence Titration Experiment in Toluene and DCB along with the Heat of Formation (ΔHf0) Values Determined by ab initio Calculations in vacuo for the Non-Covalent Complexes of 1 with C60 and C70 (Temperature 298 K) K (dm3·mol−1) toluene

ΔHf 0 (kcal·mol−1)

DCB

system

KUV−vis

Kfluores.

Kav

KUV−vis

Kfluores.

Kav

C60/1 C70/1

16 450 17 670

20 205 20 650

18 330 19 160

12 770 11 290

12 420 16 605

12 595 15 292

−0.624 −1.296 (side-on orientation of C70) −1.3045 (end-on orientation of C70)

a decent result near the maxima of the curve spread over a very small region. For this reason, although the errors in the center of the CT spectra for the complexes of 1 with various electron donors are very small, there are appreciable errors in the y0 value. The newly appeared bands, e.g., CT absorption bands of various fullerene/1 complexes, depend on the polarity of the

analysis plot is shown in Figure 5. The wavelengths at these new absorption maxima (λmax = xc) and the corresponding CT transition energies (hνCT) are summarized in Table 2. The Gaussian analysis fitting is done in accordance with the method developed by I. R. Gould et al.35 One important point to mention here is that Gaussian analysis of a curve generally gives 11982



Article

Figure 4. (a) UV−vis titration curve of the C60/1 system recorded in DCB: (i) uncomplexed 1 (8.90 × 10−6 mol·dm−3), (ii) uncomplexed C60 (6.25 × 10−5 mol·dm−3), and (iii−ix) mixture of C60 and 1 in which the concentration of C60 varies in the range of 1.0 × 10−4 to 2.25 × 10−4 mol·dm−3. (b) BH plot of the C60/1 system in DCB.

toluene, lacks the stabilization effect of such an EDA pair. Therefore, charge recombination, to yield the singlet groundstate or an excited-state product, is usually a fast phenomenon. Apart from that, the controlled formation of fullerene clusters in polar media is another versatile approach to improve the performance of fullerene-containing dyads or noncovalent conjugates under the aspect of charge separation.36 In particular, it is feasible that the electron delocalization within a fullerene cluster is distributed over all fullerene molecules. This, in turn, proved to be a beneficial factor behind stabilization of the charge-separated EDA state. On the other hand, εs plays a very important as well as useful role behind fruitful fullerene−ZnPc interaction. This may be well understood from the characteristic luminescence dependence on the value of ε. Specifically, increasing the solvent polarity from a nonpolar solvent, i.e., toluene (εs = 2.39) to DCB (εs = 9.93) results in a substantial decrease of the formation of the triplet state of 1. Additionally, it enhances the chance of the formation of the triplet state of fullerene which in turn facilitates the energy transfer phenomenon. In essence, the free-energy change (−ΔG0), associated with an intermolecular energy

Figure 5. Gaussian analysis plot of the CT band obtained for the C60/ 1 system recorded in DCB.

solvent and its dielectric constant (εs) value. This is because, in polar solvents, the charge-separated electron donor−acceptor (EDA) pair is regarded to diffuse semifreely, almost like two different entitites in solution. However, nonpolar solvent, e.g., 11983



Article

Table 2. CT Absorption Maximum (λCT), CT Transition Energy (hνCT), Vertical Electron Affinity of Various Electron Acceptors (EAv), Degrees of CT (α), Oscillator Strength (f), Transition Dipole Strength (μEN), and Resonance Energy of Interaction (RN) for the Complexes of 1 with Various Electron Acceptor Molecules Including C60 and C70 Recorded in DCB (Temperature 298 K) system

λCT (nm)

hνCT (eV)

EAV of electron acceptor (eV)

α

f

μEN (D)

RN (eV)

C60/1 C70/1 TCNQ/1 o-chloranil/1 p-chloranil/1 DDQ/1

712 713 724 729 715 730

1.74 1.74 1.71 1.70 1.735 1.70

2.31 2.59 2.80 2.87 1.37 3.27

0.020 0.014 0.0092 0.0075 0.036

0.285 0.276 0.096 0.0064 0.0178 0.0086

11.63 11.23 12.35 3.55 5.28 3.88

0.343 0.320 0.267 0.065 0.106 0.075

transfer, becomes more endothermic with increasing solvent polarity. This will be discussed in more detail in section 3.10. 3.2. Theoretical Model in Favor of Electric Dipole− Dipole Interaction between Fullerene and 1. Consider the interaction of C60 with 1. The interaction between the dipole− dipole transitions of C60 and 1 may be represented in the form Nmax

H=

∑ (dC60σxC60d1iσx ,i1)(1 − 3 cos2 θi)/ε∞ri 3 i=1

(1)

i

where d C60 and d 1 are the dipole moments of the corresponding transitions in C60 and the ith number of molecules in 1, σx and σx,i1 are the corresponding Pauly matrices, ri is the distance between the C60 and 1 molecules, and ε∞ in eq 1 marks the high-frequency dielectric constant. The reconstruction of the resulting spectrum, taking into account eq 1, is determined by mixing of the states of the C60 and the surrounding 1 molecules.

Figure 6. Variation in absorbance vs concentration of C60 for the C60/ 1 system done in toluene.

The relationship between the hνCT of the lowest energy intermolecular CT band and the vertical ionization potential (IDv) of the donor for a series of complexes with a common acceptor species has been the source of much discussion. According to Mulliken’s theory,37 the ground state of the complex is a resonance hybrid of a “no-bond” state (D···A) and a dative state (D+A−) with the former predominating; the excited state is a resonance hybrid of the same two structures with the dative one predominating. CT transition energies of these complexes are related to the vertical ionization potentials (IDv) of the donors by the relation

E±i = (EC60 + E1)/2 ± {[(EC60 − E1)/2]2 + |Vi |2 }1/2 (2)

For one C60/1 pair, eq 1 gives eq 2, where EC60 and E1 are the energies of the dipole transitions of C60 and 1, respectively, and Vi = [dC60d1i − (1 − 3 cos2 θi)]ε∞−1ri−3 is the matrix element of the state mixing. The final expression has the form ε− = ε−(0) − (|Vi |N1/2)/2

(3)

where V is the amplitude of the nondiagonal flip-flop dipole− dipole matrix element for C60 and of the dipole transitions in neighboring 1 molecules and N is the number of neighboring 1 molecules. Such a dependence of the absorption band edge is valid only under the condition N < Nthr, where Nthr is the maximum number of 1 molecules that can take part in the dipole−dipole flip-flop interaction with C60. A further increase in the concentration of 1 does not increase the number of these molecules in the nearest environment of C60. Figure 6 shows that the dependence is saturated when the concentration of C60 exceeds 1.7 × 10−4 mol·dm−3 in the case of the C60/1 complexation process, in agreement with the theory. A similar sort of observation is noticed in the case of the C70/1 complexation process (see Figure 4S, Supporting Information). This mechanism, thus, allows us to explain the formation of the EDA type study. 3.3. Determination of Vertical Ionization Potential (IDv) of 1. For complexes with a neutral ground state, a CT band corresponds to a transfer of an electron from a donor to an acceptor molecule with the absorption of a quantum:

hvCT = (ID v − C1)] + [C2/(ID v − C1)]

(4)

Here

C1 = EA v + G0 + G1

(5)

EAv

where is the vertical electron affinity of the acceptor, G0 the sum of several energy terms (like dipole−dipole, van der Waals interaction, etc.) in the “no-bond” state, and G1 the sum of a number of energy terms in the “dative” state. In most cases, G0 is small and can be neglected, while G1 is largely the electrostatic energy of attraction between D+ and A−. The term C2 in eq 4 is related to the resonance energy of interaction between the “no-bond” and “dative” forms in the ground and excited states, and for a given acceptor, it may be supposed to be constant.37 A rearrangement of eq 4 yields 2C1 + hvCT = (1/ID v)C1(C1 + hvCT) + {(C2/ID v) + ID v} (6)

The vertical electron affinities of C60, C70, p-chloranil, ochloranil, TCNQ, and DDQ have been collected from the literature (see Table 2).38−42 Neglecting G0 and taking the

hvCT

Dδ + ··· Aδ − ⎯⎯⎯→ D(1 − δ) +A(1 − δ) − 11984



Article

typical D−A distance for all the CT complexes under study to be 3.5 Å, the major part of G1 is estimated to be e2/4πεor = 4.13 eV. Using these values, C1 is obtained from eq 5 for each of the acceptors. A plot of 2C1 + hνCT versus C1(C1 + hνCT) for a given donor and various electron acceptors yields a slope of 1/ IDv from which the value of IDv has been obtained for 1. The following linear regressions have been obtained with the present data: 2C1 + hvCT = (0.13742 ± 0.0029)C1(C1 + hvCT) + (7.3223 ± 0.1648);

correlation coefficient = 0.99 (7)

The above plot is demonstrated in Figure 7. The determined to be 7.28 eV in DCB.

IDv

of 1 is Figure 8. Variation of α vs EAV of various electron acceptors done in DCB.

spectra, we can enumerate the magnitude of the oscillator strength. The oscillator strength f is estimated using the formula f = 4.32 × 10−9

∫ εCT dv

(11)

where ∫ εCT dν is the area under the curve of the extinction coefficient of the absorption band in question vs frequency. To a first approximation,43 f = 4.32 × 10−9εmax Δv1/2

where εmax is the maximum extinction coefficient of the band and Δν1/2 is the half-width, i.e., the width of the band at half of the maximum extinction. The observed oscillator strengths of the CT bands are summarized in Table 2. It is worth mentioning that we need a proper calculation of oscillator strengths of fullerene/1 CT complexes. This is because oscillator strength is very sensitive to the molecular configuration and the electron charge distribution in the CT complex. In the C60/1 and C70/1 complexes, we could not use a simple model assuming a charge localized at a certain site of the fullerene sphere, because π-bonds of the fullerenes are directed radially with a node on the molecular cage. The equation for determining the oscillator strength, i.e., eq 11, is universal for only those complexes or conjugate (both covalent and noncovalent) which exhibits CT or EDA characteristics at the ground state. f is a dimensionless quantity, and values of f are normalized so that its maximum value is 1. The extinction coefficient is related to the transition dipole by

Figure 7. Plot of 2C1 + hνCT vs C1(C1 + hνCT) for the determination of IDV of 1.

3.4. Determination of Degrees of Charge Transfer (α). A rearrangement of eq 4 gives 2ID v − hvCT = (1/C1)ID v(ID v − hvCT) + [C1 + (C2/C1)] (8)

Utilizing the values of C1 and C2 obtained from eq 8, degrees of charge transfer (α) for the fullerene/1 CT complexes may be formulated as follows: α = (C2/2)/[(ID v − EA v + C1)2 + (C2/2)]

(12)

(9)

The values of α for all the EDA complexes of 1 are listed in Table 2. The values of α (calculated by using eq 9) are small and indicate that very little amount of charge transfer occurs in the ground state. In the present case, an excellent straight line relationship is obtained from the variation of α with EAv (Figure 8). The following correlation has been obtained with the present data:

μEN = 0.0952[εmax Δv1/2 /Δv]1/2

(13) −1

where Δν = full width of the band (in cm ), Δν1/2 = width of the band (in cm−1) at εmax/2, and μEN is defined as −e∫ Ψex∑iriΨg dτ. μEN is the transition dipole moment for the transition between the neutral ground state and the excited state of the CT complex, ri signifies the length of the position vector of the electron from the center of the coordinate axes, and dτ is the volume element in three-dimensional coordinate space. The quantity e is the electronic charge. The transition dipole strengths (μEN) for the complexes of 1 with various acceptors are given in Table 2. It is observed that there is not so much difference in the μEN values of the C60/1 and C70/1 systems. This trend in μEN is in conformity with the fact that the difference in electron susceptibility of C60 and C70 is very

α = (0.0623 ± 0.00188) − (0.0188 ± 7.67 × 10−4)EA v (10)

In the present case, with increasing electron affinity of the acceptors, the α value decreases. This is primarily because, apart from the electron affinity value of the electron acceptors, CT interaction is also controlled by complementary size of the electron acceptor or guest with the electron donor or host molecule. 3.5. Determination of Oscillator Strength ( f) and Transition Dipole Moments (μEN). From the CT absorption 11985



Article

Figure 9. (a) Fluorescence spectral variation of 1 (1.0 × 10−5 mol·dm−3) in the presence of C60 in toluene medium; the concentrations of C60 from top to bottom in the arrow mark (indicated in the figure) are as follows: 2.48 × 10−5, 3.73 × 10−5, 4.97 × 10−5, 6.22 × 10−5, 7.46 × 10−5, 8.70 × 10−5, 9.95 × 10−5, 11.20 × 10−5, 12.44 × 10−5, 13.70 × 10−5, 14.93 × 10−5, and 16.17 × 10−5 mol·dm−3 . A plot of relative fluorescence intensity vs [C60] for the C60/1 system in toluene medium is shown in the inset of Figure 9a. (b) BH fluorescence plot of the C60/1 system done in toluene.

small. However, the observed trend in μEN values of the fullerene/1 complexes corroborates fairly well with the determined K values of the respective complexes (see Table 1). It is expected that, if CT takes place at the ground state, the magnitude of the transition dipole moments (μEN) would be high. In this case, both C60 and C70 are expected to form strong complexes with 1 compared to other electron acceptors due to complimentary size and well preorganized stereoscopic fitting situation. For this reason, μEN values for the C60/1 and C70/1 systems are supposed to be higher compared to the CT complexes comprising 1 and other electron acceptors, viz., ochloranil, p-chloranil, and DDQ. 3.6. Determination of Resonance Energy (RN). Briegleb and Czekalla44 theoretically derived the relation εmax = 7.7 × 104 /[hvCT/|RN| − 3.5]

CT transition, and RN is the resonance energy of the complex in the ground state, which obviously is a contributing factor to the stability constant of the complex (a ground state property). RN values for the complexes of 1 with various electron acceptors are summarized in Table 2. A lower value of RN for the C70/1 complex than that of the C60/1 complex hints that there would be very less or no contribution of π−π interaction in the fullerene/1 complexation process. 3.7. Steady State and Time Resolved Fluorescence Studies. To study the photoinduced behavior of C60/1 and C70/1 supramolecular complexes and the recognition motif of C60 and C70 toward 1, steady-state emission measurements are carried out in both toluene and DCB. From the above discussions, we may infer that the direct π-stacking effect plays a very minor role, as the fullerene core may not get in close contact with the plane of 1. This can be demonstrated in terms of the electrostatic interactions prevailing between positively charged 1δ+ and the negatively charged fullereneδ− components

(14)

where εmax is the molar extinction coefficient of the complex at the maximum of the CT absorption, hνCT is the energy of the 11986



Article

Figure 10. (a) Fluorescence spectral variation of 1 (1.0 × 10−5 mol·dm−3) in the presence of C70 measured in toluene; the concentrations of C70 from top to bottom in the arrow mark (indicated in the figure) are as follows: 3.75 × 10−5, 6.23 × 10−5, 7.48 × 10−5, 8.73 × 10−5, 9.97 × 10−5, 11.22 × 10−5, 12.45× 10−5, and 13.70 × 10−5 mol·dm−3. A plot of relative fluorescence intensity vs [C70] for the C70/1 system (done in toluene) is shown in the inset. (b) BH fluorescence plot for C70/1 in toluene.

that the fluorescence of 1 is characterized by the maximum at 690 nm followed by one shoulder peak at 755 nm, upon excitation at the Soret band maximum. Evidence in favor of the energy-transfer deactivation is obtained from the titration experiment of a toluene or DCB solution of 1 with variable C60 (Figure 9a/Figure 5S (Supporting Information)) and C70 (Figure 10a/Figure 6S (Supporting Information)) concentration in the range from 2.20 × 10−5 to 35.40 × 10−5 mol·dm−3. It is observed that, upon excitation at Soret band maxima of 1, a C60 or C70 concentration dependent decrease in the intensity of the fluorescence maxima of 1 is seen in both toluene and DCB. Typical plots of the relative fluorescence intensity of 1 vs the concentration of C60 and C70 in toluene (DCB) are shown in the inset of Figures 9a (5S, Supporting Information) and 10a (6S, Supporting Information), respectively. At high fullerene concentration, a plateau feature is observed, at which the complexation of 1 is assumed to be complete (inset of Figures 9a (5S, Supporting Information) and

with a trend toward formation of well-directed and oriented assembly of 1:1 supramolecular fullerene/1 complexes. The simple mixing of the individual components, i.e., fullerene and 1, leads to a novel superstructure, for which we may expect that the highly fluorescent state of the singlet excited 1* is quenched by an intercomplex energy and/electron transfer to fullerene forming fullereneδ−. It has been reported earlier that charge separation can occur from the excited singlet state of 1 to fullerene in the fullerene/1 noncovalent hybrid system.45 Photophysical studies prove that, in the case of conformationally flexible dyads comprising fullerenes and macrocyclic receptor molecules, like porphyrin (Por), π-stacking interactions are facilitated due to through-space interactions between these two chromophores. This has been demonstrated by quenching of 1*Por fluorescence and formation of fullereneexcited states (by energy transfer) or generation of fullerene−/ Por+ ion-pair states (by electron transfer).46 In our present investigations, the steady state fluorescence experiments reveal 11987



Article

10a (6S, Supporting Information)). It should be mentioned at this point that a purely diffusion-driven process is ruled out, on the basis of the applied fullerene concentration. The decrease of the fluorescence intensity of 1 and the shift of the 1 fluorescence suggest a static quenching event inside the welldefined fullerene/1 supramolecular complexes. On the basis of the aforementioned results, we reach the conclusions that, in the fullerene/1 complexes, the fluorescence state of 1 is quenched by the addition of electron-accepting C60 and C70. K values of the fullerene/1 complexes are evaluated according to a modified BH equation and are listed in Table 1 along with the values obtained from UV−vis investigations.33 BH plots of fullerene/1 systems estimated in two different solvents are shown in Figures 9b (5S, Supporting Information) and 10b (6S, Supporting Information). As we use the Soret absorption band as our source of excitation wavelength in fluorescence experiment, the second excited singlet state of 1 has been deactivated by the following mechanism: internal conversion

0

energy tr.

1 + hv → 21* ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ 11* ⎯⎯⎯⎯⎯⎯⎯⎯→ 01/3fullerene*

A similar sort of experimental observations have already been reported by Yin et al.47 and Guldi et al.48 for their particular C60 supramolecule. It is interesting to note that the increase in magnitude of K values led to an increase of the fluorescence quenching efficiency. This can be viewed in terms of the fact that the rigidity of the Pc unit brings tight fixation of C60 and C70 unit in the fullerene/1 host−guest complex. To validate the presence of the static quenching mechanism in our present investigations, we have performed detailed picosecond time-resolved fluorescence measurements for fullerene/1 complexes in both toluene and DCB. The titration experiment is carried out at a fixed concentration of 1 and a variable concentration of C60 and C70. The time-resolved fluorescence of 1 reveals a single exponential decay with a singlet state lifetime (τs) value of 3.48 and 3.47 ns measured in toluene (Figure 11a(i)) and DCB (Figure 11b(i)), respectively. It is observed that, upon the gradual addition of fullerenes C60 and C70 in toluene (and DCB), the value of τs suffers little change, and mono-exponential decay is followed as depicted in Figure 11a(ii) and (iii) (11b(ii) and (iiii)), respectively. Since the value of τs does not change much in the presence of the quencher, there is no question of observing the lifetime value in another time range like the picosecond or femtosecond region. The values of τs of 1 in the absence and presence of fullerenes (measured in toluene and DCB) are tabulated in Table 3. From the lifetime experiment, the rate constant for charge separation for the singlet state (kCSs) is calculated using the equation k CSs = (1/τ s)complex − (1/τ s)ref s

Figure 11. (a) Fluorescence decay profiles of (i) 1 (2.18 × 10−5 mol·dm−3), (ii) C60 (9.35 × 10−5 mol·dm−3) + 1 (2.18 × 10−5 mol·dm−3), and (iii) C70 (8.92 × 10−5 mol·dm−3) + 1 (2.18 × 10−5 mol·dm−3) done in toluene. (b) Fluorescence decay profiles of (i) 1 (2.62 × 10−5 mol·dm−3), (ii) C60 (1.33 × 10−4 mol·dm−3) + 1 (2.62 × 10−5 mol·dm−3), and (iii) C70 (1.25 × 10−4 mol·dm−3) + 1 (2.62 × 10−5) done in DCB.

Table 3. Lifetime (τ), Rate Constant for Charge Separation (kCS), and Quantum Yield for Charge Separation (ΦCS) for 1 in the Absence and Presence of C60 and C70 Recorded in Toluene and DCB Medium (Temperature 298 K) solvent

τ (ns)

1

toluene DCB toluene DCB toluene DCB

3.48 3.47 3.25 3.09 3.23 3.05

C60/1 C70/1

kCS (s−1)

ΦCS

× × × ×

0.070 0.123 0.077 0.138

2.03 3.55 2.22 3.97

107 107 107 107

estimated to be higher in the case of the C70/1 complex. The values of both kCS and ΦCS are estimated to be higher for fullerene/1 complexes in DCB compared to toluene. From this very interesting observation, we may surmise that polar solvent facilitates electrostatic interaction51 between fullerenes and 1, which in turn is very much responsible for getting a fruitful CT phenomenon in this particular solvent. Therefore, it would be highly beneficial if quantum chemical calculations may invoke some light on the geometrical arrangement of both the fullerenes and 1 molecules in fullerene/1 systems. We are very much thankful to one of the learned reviewers for his/her comments, as it created the opportunity to find out the value of one important photophysical parameter, namely, the bimolecular quenching constant (kq) for the C60/1 and C70/1 systems. If we make a plot of the ratio of relative fluorescence intensity, i.e., F0/F (where F0 and F are the steady state fluorescence intensity of 1 in the absence and presence of fullerenes) vs concentration of the quencher (here fullerene), a nonlinear curve is obtained for all of the systems recorded under the present investigations. This is indicative of the phenomenon that both the static and dynamic quenching is observed in our present work. If the above plot is modified like eq 17 as stated

(15)

s

where (1/τ )ref and (1/τ )complex signify the lifetime of uncomplexed 1 and the fullerene/1 complexes, respectively. The quantum yield of the charge separated state for the singlet state (ΦCSs) is determined according to the equation ΦCSs = [(1/τ s)complex − (1/τ s)ref ]/(1/τ s)complex

system

(16)

The estimated values of kCS and ΦCS are summarized in Table 3. Table 3 indicates efficient charge separations for the C70/1 complex observed in both of the solvents studied under the present investigations. A similar sort of phenomenon is already observed for other fullerene/phthalocyanine supramolecular complexes.49,50 This is quite expected, as the value of K is 11988



Article

structures as well as association energies of the fullerene/1 ensembles is a tedious job due to the presence of a large number of basis functions resulting from the large size of the system which limits the use of methods. We will describe here an approach that appears to be capable of precisely mimicking such results with a far less expensive treatment of electron correlation. In our present investigations, the existence of the electrostatic interactions between fullerene and 1 has been evidenced by the results obtained on frontier molecular orbitals, viz., highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) employing ab initio or Hartree−Fock (HF) calculations with the help of the slater type orbital (STO)/3-21G as the basis set. The frontier orbitals like HOMO and LUMO have more importance because they can help explain the reactivity and electronic properties of the molecule. The HOMO−LUMO energy separation has been used as an index of kinetic stability for fullerenes, as for other types of molecules. HF calculations reveal that, in the ground state, the majority of the HOMO is located on 1, while the LUMO is positioned in the fullerene entity of the complex. The absence of HOMO on fullerene and LUMO on the ZnPc macrocycle suggests weak CT interactions between 1 and fullerenes, which is consistent with the fact that CT bands of low intensity are observed for the presently investigated supramolecules in the UV−vis experiments, as mentioned in section 3.3. It may be seen that LUMO energy (ELUMO) levels of fullerene/1 complexes compare well with the fullerene guest, while the HOMO energy (EHOMO) levels are similar to those of the uncomplexed receptor, 1. For example, in the case of the C60/1 complex (Figure 12a and b), ELUMO is computed to be −0.6592 eV, which is comparable to the ELUMO of C60, i.e., −0.7690 eV. Also, the EHOMO of the same complex was estimated to be −5.3013 eV, which corroborates excellently with that of 1, viz., −5.2903 eV. A similar sort of observation is also noticed for the C70/1 complexation process (Figure 12c− f). EHOMO and ELUMO of the C60/1, C70/1 (side-on orientation of C70), and C70/1 (end-on orientation of C70) complexes along with uncomplexed 1, C60, and C70 are given in Table 1S (Supporting Information). The present method of computations fails to produce corroboration between the theoretically obtained HOMO−LUMO energy gap and the hνCT values (as obtained from the analysis of CT absorption spectra) for the C60/1 and C70/1 systems. This is purely because of the fact that the theoretical calculations are done in vacuo. As a result of this, it fails to correlate exactly with the experimentally obtain CT transition energy. Lack of solvation energy calculation in the former case may be accounted for by this difference. In these particular systems, therefore, the electrostatic interaction primarily determined the absorption geometry of fullerene/1 complexes. It is also observed that, in the charged state, all the decisive molecular orbitals of the complex are leaning on one of the components, i.e., either on 1 or on the fullerene subunit. Neither the HOMO nor LUMO is found to be positioned on the tert-butyl portion of 1. It should be mentioned at this point that electrostatic interaction is only one of the important components (and not the most important one), which may contribute to the stabilization of molecular van der Waals complex between fullerenes and 1, since both components are not charged (very small differences in electronic density distribution over the neutral molecule cannot provide strong electrostatic interaction between fullerene and 1). This interaction is important for the complexes with charged

below, a straight line plot is obtained with the data obtained from steady state fluorescence measurements, for different fullerene/1 systems, recorded in toluene and DCB. {(F0/F − 1)/[Fullerene]} = (KSV + K s) + (KSVK s) [Fullerene]

(17)

In eq 17, KSV and Ks signify the value of diffusion controlled rate constant and static quenching constant, respectively; [Fullerene] indicates the concentration of fullerenes C60 and C70. The value of kq for the C60/1 and C70/1 systems in toluene (DCB) is reported to be 4.75 × 1012 s−1 (3.52 × 1012 s−1) and 3.65 × 1012 s−1 (1.85 × 1012 s−1), respectively. 3.8. Binding Constants. The binding constants (K) of fullerene/1 complexes determined by both UV−vis and steady state fluorescence measurements are summarized in Table 1. It shows that 1 undergoes an appreciable amount of complexation with both C60 and C70 in both toluene and DCB. Another interesting aspect of the present investigations is that 1 does not exhibit any sort of selectivity in binding, i.e., KC70/KC60. The values of KC70/KC60, measured in toluene and DCB, are determined to be 1.05 and 1.20, respectively. The present selectivity in binding for the fullerene/1 systems is, therefore, found to be lower than that of azacalix[m]arene[n]pyridine (∼1.940),52 and they are lower than those of calixarene diporphyrin (∼4.3), 53 cyclic dimers of Zn-porphyrins (∼25.5),54 and corresponding H2-diporphyrin (∼32)55 complexes. From the above results, we can say that, in spite of the similarity in structure of zinc porphyrin and zinc phthalocyanine, a remarkable decrease in selectivity of KC70/KC60 is observed in the latter case. The present selectivity ratio corroborates fairly well with the observed value for the fullerene/azulene complex (viz., KC70/KC60 = 1) reported in recent past by Rahman et al.56 However, C70 produces a somewhat higher value of K with 1 compared to C60 in both toluene and DCB. This phenomenon may be explained as follows: It is well-known that the stability of the supramolecular complexes depends upon attractive interactions between host and guest, and on solvation of the binding partners.57 In the present investigations, since the host and guest are separately solvated in solution, the desolvation of the host and guest is requisite in the association process. However, the process of desolvation is energetically an uphill task and, hence, an association takes place only when the gain in energy between the host and guest interaction exceeds this unfavorable energy. From the trends in the K values of the fullerene/1 complexes, we may infer that the extent of solvation and desolvation of the binding partners might play an important role in forming weak or strong supramolecular complexes. Haino et al.58 show that the stability of the complex increases as the solubility of fullerene in a solvent decreases, since less energy is required for the desolvation of fullerene which must necessarily precede its complexation with a host molecule. This would be one of the reasons for the higher K values of the C70 complexes in toluene, since the solubility of C60 is higher in toluene (4.1 mg/mL) than that of C70 (2.0 mg/mL).59 Incidentally, the solubility of fullerenes C60 and C70 is higher in DCB compared to toluene.60 For this reason, K for the fullerene/1 complexes is found to exhibit a lower value in DCB compared to toluene. 3.9. Theoretical Calculations. Making accurate quantum chemical calculations for the elucidations of electronic 11989



Article

Figure 12. (a) HOMO of C60/1 system, (b) LUMO of C60/1 system, (c) HOMO of C70/1 system (when C70 is oriented in end-on), (d) LUMO of C70/1 system (when C70 is oriented in end-on), (e) HOMO of C70/1 system (when C70 is oriented in side-on) and (f) LUMO of C70/1 system (when C70 is oriented in side-on) done by HF//3-21G calculation in vacuo.

components.61 Together with electrostatic interaction, other types of fullerene/1 interaction like polarization interaction, π−π interaction, d−π interaction, etc., between neutral molecules also play a vital role in stabilizing the complex. Heat of formation (ΔHf0) values for the complexes of 1 with C60 and C70 also extend good support in interpreting the stability differences between corresponding C60/1 and C70/1 complexes (Table 1). The geometric parameters of the complexes have been obtained after complete energy minimization at the HF/3-21G level of theoretical calculations. Typical single projection geometric structures of the supramolecular complexes of 1 with C60, C70 (end-on), and C70 (side-on) are shown in Figure 13a, b, and c, respectively. For the energy classification of the optimized structures, we used the following (standard) ΔHf0 definitions. Heat of formations have been estimated within the supramolecular approach ΔHf0 = Ecomplex − [E1 + Efullerene] (where Ecomplex, E1, and Efullerene are the energy of the fullerene/1 complex, uncomplexed 1, and free fullerene, respectively) in order to determine the total strength of the various interaction patterns between the fullerenes and the receptor, e.g., 1. The value of ΔHf0 becomes negative in all the cases studied which indicates that the complexation processes of 1 with C60 and C70 are endothermic in nature.

Figure 13. Single projection geometric structures of (a) C60/1, (b) C70/1 (end-on orientation of C70), and (c) C70/1 (side-on orientation of C70) done by HF/3-21G calculations in vacuo.

Table 1 also suggests that the complexation of 1 with C70 is much more enthalpy favored than that of C60. Presumably, attractive π···π interaction is responsible for greater stability in the case of the C70/1 complex. One of the most significant features of the theoretical calculations is that 1 could not serve as a good discriminator receptor between C60 and C70 as revealed from K values with its complexes of C60 and C70 (see Table 1). This phenomenon stimulates us to look into detail regarding the orientation pattern of C70 toward the plane of 1. We anticipate that the surface area of the C60/1 complex (1191.33 Å2) would not differ much with respect to C70. Estimation of the surface area of the C70 complexes of 1 (side11990



Article

potential (MEP) maps have been generated for C60 (Figure 15a), C70 (Figure 15b), 1 (Figure 15c), C60/1 (Figure 15d), and C70/1 (Figure 15e) systems to visualize the electrostatic interactions. The MEP of 1 shows negative electrostatic potential (shown in red) on the Pc ring (mostly located on the nitrogen atoms) (Figure 15c). The MEPs for the fullerenes are blue-green, indicating positive electrostatic potential; bluegreen color of fullerenes corresponds to the center regions of the five- and six-membered rings (Figure 15a and b). However, along the 6:6 bonds, regions of negative potentials (shown in red) can be observed. Interestingly, in the supramolecular complexes, C60/1 (Figure 15d) and C70/1 (Figure 15e), the original blue-green color of the separated C60 and C70 are changed to green and the deep red color of Pc changed to reddish-yellow, indicating the possibility of energy or electron transfer between these two chromophores upon photoexcitation. It should be mentioned, at this point, that electrostatic interaction is one of the important components (and the most important one), which may contribute to the stabilization of the complexes through the van der Waals interaction, since both components, viz., fullerenes and 1, are not charged (the very small differences in electronic density distribution over the neutral molecule cannot provide strong electrostatic interaction between 1 and fullerenes). This interaction is important for the complexes with charged components, or it is important in the excited states of fullerene/1 assemblies. 3.10. Solvent Reorganization Energy (RS). The effect of solvent over electronic coupling phenomenon between fullerenes and 1 may be better understood by the estimation of solvent reorganization energy (Rs) for the fullerene/1 complexes. The total reorganization energy, in general, is a sum of the two terms, i.e., inner-sphere reorganization energy (solvent-independent) R066 and outer-sphere reorganization energy (solvent-dependent) Rs. In the case of fullerene, as far as the Rs contribution is concerned, this is believed to be small as well. The symmetrical shape and large size of the fullerene framework require little energy for the adjustment of an excited or reduced state to the new solvent environment. In the present investigations, the Rs values of C60/1 and C70/1 complexes have been estimated by applying the dielectric continuum model developed by Hauke et al. (see eq 4)67

on) and 1 (end-on) are estimated to be 1248.52 and 1248.60 Å2, respectively. Thus, unlike fullerene/porphyrin interaction,45,62 computations of the preferred conformations of fullerene/1 complexes confirm that van der Waals attractive forces play a negligible role for the investigated supramolecules which stands in completely reverse side of the concept that we observe in fullerene/porphyrin interaction.63 Table 1 summarizes the results obtained from ab initio calculations regarding the orientation behavior of C70 molecule toward 1 in terms of ΔHf0 value. It is observed that the alignment of C70 with the plane of 1 is favored in end-on orientation compared to familiar side-on orientation. The ΔHf0 values of the C70/1 complex at its end-on and side-on orientations are computed to be −1.305 and −1.296 kcal·mol−1, respectively. This promising feature of the C70/1 complexation process gets very reliable experimental evidence from the variable temperature 13C NMR experiment of the C70/1 complex recorded in tol-d8 medium (Figure 14). It

Figure 14. 13C NMR spectrum of the C70/1 system recorded in tol-d8 medium.

is already reported that chemical shifts of free C70 appear at 151.07, 147.52, 146.82, 144.77, and 131.0, due to the existence of five different types of carbon atoms, namely, polar, α, β, γ, and equatorial, respectively.64,65 In our present investigation, it is observed that, although the equatorial peak gets only 0.3 ppm (i.e., 131.3 ppm) perturbation in terms of difference in chemical shift value, the peak at the pole region suffers a formidable amount of chemical shift change, i.e., ∼1.8 ppm (peak value of 152.9 ppm). This observation strongly suggests that the pole region of C70 maintains its close distance with the plane of 1 as a result of its end-on orientation motif. Both theoretical and experimental observations, therefore, envisage that such a binding motif of C70 toward 1 occurs in order to maximize the electrostatic interactions. This feature proves that, unlike C70/ porphyrin interaction,64,65 the presence of dispersive forces associated with π−π interactions plays a negligible role in the present case. It should be mentioned at this point that the aromatic solvent induced shift (ASIS) for the noncovalent complexes generally ranges around the value of ∼0.05 Hz. For our presently investigated supramolecules, we have obtained a considerable extent of shift in 13C NMR investigations, i.e., 1.8 ppm or 360 Hz in a 200 MHz NMR instrument. Thus, the large amount of chemical shift value like 360 Hz cannot be identified as ASIS. The shift obtained in our present work is definitely due to the strong complexation between C70 and 1. Electrostatic interactions originating from the electron density at the surface of the fullerenes and 1 of the fullerene/ 1 supramolecules are supposed to play a vital role in the interaction between fullerenes and 1. Molecular electrostatic

R s = (e 2 /4πε0)[{(1/2R1) + (1/2R fullerene) − (1/RD − A )} (1/εs) − {(1/2R fullerene) + (1/2R1)}(1/εR )]

(18)

with the following parameters: radius of donor, R1 = 6.226 Å; radius of acceptor (Rfullerene): RC60 = 4.2 Å, RC70 = 4.4 Å; donor−acceptor separation (RD−A): RC60/1 = 3.514 Å, RC70/1 = 3.411 Å; solvent dielectric constant, εs (εtoluene = 2.39; εDCB = 9.93); solvent dielectric constant for electrochemical measurements, εR = 9.93. Values of Rs for the above supramolecular systems are given in Table 4. It is to be mentioned here that the solvent reorganization energies obtained in the present investigation do not corroborate well with that observed for the quinone/porphyrin system.68 The discrepancy in the value of Rs for quinone/porphyrin and fullerene/1 systems may be due to the subtle structural change in the host−guest complex which exerts a large influence upon the photoinduced electron and/energy transfer process. The Rs value for the fullerene/1 systems is much smaller than the bisporphyrin-linked system reported by Osuka et al. (0.5−1.5 eV).69 The small Rs value in the C60/1 and C70/1 systems may be ascribed to the large 11991



Article

Figure 15. MEPs of (a) 1, (b) C60, (c) C70, (d) C60/1, and (e) C70/1 systems (end-on orientation of C70) done by HF/3-21G calculations in vacuo.

RS(C60/1) determined in all the solvents studied in the present investigations. The free energy of electron or energy-transfer process may be calculated from the following equation:

Table 4. Electron Donor−Acceptor Distances (RD−A), Solvent Reorganization Energies (Rs), and Electronic Coupling Elements (V) of the C60/1 and C70/1 Systems Estimated in Toluene, Dichloromethane (CH2Cl2), and DCB (Temperature 298 K) system

RD−A (Å)

solvent

Rs (eV)

C60/1

3.514

toluene DCB CH2Cl2 toluene DCB CH2Cl2

−0.81 −0.41 −0.42 −0.88 −0.43 −0.44

C70/1

3.411

ΔG 0 ≈ 23.05(Eox − Ered) [in kcal/mol units]

V (cm−1)

(19)

Eox and Ered are the standard value of one electron oxidation and one electron reduction potential value of 1 and fullerenes (both C60 and C70), respectively, in solution. It should be mentioned at this point that we failed to get reliable data of both oxidation and reduction potential values of 1 in DCB. Therefore, Table 5 of the revised manuscript does not contain the data related to the physicochemical properties like free energy changes for the fullerene/1 noncovalent assemblies in DCB. Instead, we have recorded the oxidation and reduction potential values for both the fullerenes and 1 in dichloromethane. The main reason behind performing the experiment in dichloromethane is that there is a very negligible difference in the solvent dielectric constant value between DCB (εs = 9.93) and dichloromethane (εs = 8.93). For this reason, we may anticipate that dichloromethane may impart a similar sort of

5628.58

5721.72

electrostatic contribution originating from CT interaction at the ground state over a three-dimensional framework of the ZnPc and fullerenes, i.e., C60 and C70. Apart from that nonpolar solvent like toluene also reduces the value of Rs than those observed in DCB and dichloromethane (Table 4) which corroborates excellently well with the rationale given by Imahori et al.34 Table 4 also indicates that RS(C70/1) < 11992



Article

equal and opposite to the reorganization energy according to the Marcus theory of electron transfer.73,74 It should be remembered in this context that the free energy for a spontaneous electron transfer is negative and the reorganization energy is positive. At some point, the sum will be zero. However, in our present investigations, the values of Rs for the fullerene/1 systems (see Table 4 of the revised manuscript) in three different solvents exhibit a negative value which does not corroborate well with the theory of electron transfer mechanism. In conclusion, we may say that the energy transfer mechanism is very much operative behind fruitful fluorescence quenching of the photoexcited 1 in the presence of C60 and C70 in solution. 3.11. Transient Absorption Study. From the steady state UV−visible spectra of 1 measured in both toluene and DCB (Figure 2), it is clearly evidenced that 1 does not have any appreciable absorption intensity at 532 nm. For this reason, we have predominantly excited C60 and C70 molecule by 532 nm laser light in our present investigations. Figure 16a demon-

Table 5. Static Energy (ΔGS), Free Energy of Charge Separation (ΔGCS), Free Energy of Charge Recombination (ΔGCR), Free Energy for Radical Ion Pair Formation (ΔGRIP), and Standard Free Energy Change Associated with the Formation of a Pair of Separated Ions from a Neutral Donor and Acceptor (ΔG0) for the C60/1 and C70/1 Systems Measured in Toluene and CH2Cl2 (Temperature 298 K) system 1 C60/1 C70/1

solvent

ΔGS (eV)

ΔGCS (eV)

ΔGCR (eV)

ΔGRIP (eV)

ΔG0 (eV)

toluenea CH2Cl2b toluenea CH2Cl2b

1.71 1.5991 1.76 1.5993

−1.736 −2.508 −1.798 −2.458

−0.094 −0.689 −0.032 −0.639

0.094 0.689 0.032 0.639

1.80 0.91 1.79 0.96

Eox (1•+/1) = 0.82 volt; Ered (C60/C60•−) = −0.98 volt; Ered (C70/ C70•−) = −0.97 volt. bEox (1•+/1) = 0.58 volt; Ered (C60/C60•−) = −0.33 volt; Ered (C70/C70•−) = −0.38 volt.

a

physicochemical responses toward the fullerene/1 noncovalent system during the complexation process when measured in such solvent. The magnitude and sign of ΔG0 as depicted in Table 5 clearly indicate that there is no possibility of energy wasting back electron transfer from the fullerene to 1 in both toluene and dichloroethane. It is customary to estimate the driving forces for the free energies of charge separation (ΔGCS) and free energies of charge recombination (ΔGCR) for the fullerene/1 complexation processes. ΔGCR is calculated using the Weller equation.70 −ΔGCR = Eox − Ered − ΔGS

(20)

In eq 20, the static energy (ΔGS) has been calculated according to the following equation:70 ΔGS = e 2 /4πε0εsR fullerene/ 1

(21)

The terms in eq 21, namely, e, ε0, and εs, refer to the elementary charge, vacuum permittivity, and static dielectric constant of the solvent used for rate measurements, respectively. Table 5 nicely demonstrates that the value of ΔGCR is more negative for the fullerene/1 systems estimated in dichloromethane compared to toluene. The faster charge recombination of the fullerene/1 systems observed in dichloromethane results from one important factor, Rs. On the basis of ΔGCR and excited state energy (E0,0) values of 1, the freeenergy changes of the charge separation process (ΔGCS) have been calculated using eq 22 and are listed in Table 5. −ΔGCS = E0,0 − ( −ΔGCR )

(22)

Table 5 reveals that the charge-separation process of 1 via the excited singlet state of C60 (1C60*) is sufficiently endothermic compared to C70 by 0.062 and 0.050 eV in toluene and dichloromethane, respectively. In our present study, we have also determined one important physicochemical parameter, namely, free energy of radical ion-pair formation (ΔGRIP).71,72 It is observed that the change in the value of radical ion-pair formation is less when the complexation process between fullerenes and 1 is monitored in dichloromethane compared to toluene. This phenomenon gives clear evidence in favor of energy transfer (rather than electron transfer) for the complexes studied under the present investigations, as we do not observe any sort of CT phenomenon in toluene. Moreover, it is already very much known to us that the maximum rate constant for electron transfer occurs when the free energy is

Figure 16. (a) Triplet−triplet absorption spectra of C60 recorded in DCB at different delay times of 1, 10, 20, and 40 μs. (b) Triplet decay profile of C60 measured in DCB (monitored at 750 nm).

strates nicely the triplet−triplet absorption spectra of C60 at different delay times measured in DCB. The absorption band appearing at 750 nm is attributed to the formation of T C60*.75−77 However, we fail to detect any absorption bands due to the formation of 1•+ and C60•−. From the above observations, we may infer that a photo induced energy transfer phenomenon via TC60* to 1 is confirmed in our present investigations. Figure 16b indicates the triplet decay profile of 11993



Article

C60 monitored at 750 nm, and the decay time is measured to be 30 μs. Since the decay of TC60* is accelerated on addition of 1, this is clearly indicative of the fact that a reaction other than electron transfer is taking place for the C60/1 complex. Figure 17a shows the C60 sensitized triplet−triplet absorption spectra

Figure 18. (a) Comparison of the decay profiles of the triplet−triplet energy transfer from C60 to 1 recorded in DCB observed at 750 and 530 nm. (b) Comparison of the decay profiles of C60 in the absence (filled circles in violet color) and presence (filled squares in black) of 1 in DCB.

observed at 480 nm of the triplet state of C70 in DCB produced after direct excitation at 532 nm. The decay and bleach recovery times are determined to be around 10 μs. Like C60, Figure 20a gives clear evidence in favor of the C70 sensitized triplet−triplet absorption spectra of 1 after the delay times of 0.6, 2.0, and 4.0 μs; in addition to that, rise time and decay time in the case of C70 are estimated to be 5 and 30 μs, respectively, as evidenced from the rise followed by decay profile of the triplet state of 1 measured in DCB (Figure 20b). It should be mentioned at this point that the triplet−triplet absorption spectrum of 1 in DCB does not provide any new physical insight (Figure 21a) and the triplet decay time of 1 in DCB is measured to be 22 μs from the triplet decay profile of such molecule (Figure 21b). Considering all of the above findings, the observed reactions may be illustrated in Appendix A and Appendix B for the C60/1 and C70/1 systems, respectively. 3.12. Electronic Coupling Element (V). The electronic coupling element (V) is related to the extent of overlap between the appropriate donor or host and acceptor or guest orbitals, and scales the dependency of free energy of electron transfer (i.e., driving force) on such a rate constant. The electronic interaction between fullerenes and 1 in fullerene/1 systems mixes electronic character and induces electron transfer (ET). It also creates an electronic basis for inducing dipoleallowed optical ET with the magnitude of the perturbation dictating the intensities of the intervalence transfer bands. This assumption implies that the value of electronic coupling element between 1 and fullerene is small or moderate assuming

Figure 17. (a) C60 sensitized triplet−triplet absorption spectra of 1 after delay times of 0.6, 1.0, and 2.5 μs in DCB. (b) Rise followed by the decay profile of the triplet state of 1 in DCB, the decay of which is monitored at 530 nm.

of 1 in DCB after the delay times of 0.6, 1.0, and 2.5 μs. Figure 17b demonstrates clearly the rise followed by decay profile of the triplet of 1 in DCB. Decay is monitored at 530 nm. The rise and decay times at 530 nm are determined to be 0.850 and 20 μs, respectively. The most fascinating feature of the transient absorption experiment comes from Figure 18a in which the comparison between the decay profiles at 750 and 530 nm of the triplet−triplet energy transfer from C60 to 1 is monitored. The decay time of C60 in the presence of 1 at 750 nm is calculated to be 1.2 μs, which is very close to the value obtained in rise time. The most concrete evidence in favor of triplet− triplet energy transfer from C60 may be ascribed in terms of the decay profile of such molecule in absence and presence of 1 done in DCB (Figure 18b); decays are monitored at 750 nm. It is observed that the decay time of C60 is decreased from 30 to 1.2 μs due to triplet−triplet energy transfer. The transient absorption experiment of C70 in the absence and presence of 1 is somewhat different than we observe in the case of C60. Figure 19a shows the triplet−triplet absorption spectrum of C70 (in DCB) after 1 μs delay time and excited at 532 nm. Figure 19b and c shows decay at 400 nm and bleach recovery curves 11994



Article

Figure 19. (a) Triplet−triplet absorption spectrum of C70 in DCB after 1 μs delay time, Ex. 532 nm; (b) decay and (c) bleach recovery curves of the triplet state of C70 in DCB produced after direct excitation at 532 nm.

Figure 20. (a) C70 sensitized triplet−triplet absorption spectra of 1 after the delay times of 0.6, 2.0, and 4.0 μs in DCB; the arrow indicates the triplet population of 1 in the early time scales. (b) Rise followed by the decay profile of the triplet of 1 in DCB, the decay of which is monitored at 530 nm.

Figure 21. (a) Triplet−triplet absorption spectrum of 1 recorded in DCB at 1 μs delay time. (b) Triplet decay profile of 1 in DCB monitored at 530 nm with a decay time of 22 μs.

11995



Article

an ET process in terms of the Marcus formalism.60 In the present investigation, V has been estimated from the absorption maxima of fullerene/1 complexes by applying the model proposed by Verhoeven et al.78 Values of V for the complexes of 1 with C60 and C70 are listed in Table 4. Estimated electronic coupling elements obtained for the above systems are very much comparable to the other donor−acceptor and host−guest systems found in the literature.79 V values determined in the present investigation are significantly larger than the typical value for the solvent-separated radical-ion pair (12 cm−1),79 indicating that both fullerenes and 1 are in close contact to form very good noncovalent assemblies. The higher V value in the case of the C70/1 complex arises due to the through space energy transfer pathway which originates from the close proximity between C70 and 1 as evidenced from distance calculation, i.e., 3.411 Å. There are some recent observations of large solvent dependence of energy transfer rate constants and electronic coupling elements in electron donor−acceptor systems where the donor and acceptor are in close proximity. Zimmt et al. have also reported the large solvent dependent electronic coupling matrix element for their C-clamp-shaped molecule.80 They propose that the solvent mediated super exchange coupling phenomenon accounts for higher V values. Therefore, the effect of solvent over V is likely to be important in the present systems. It should be mentioned at this point that the difference in electronic coupling might result from the energetic position of the CT state, which depends on the solvent polarity.81

(h)

(i) (j) (k)

■ ■

4. CONCLUSIONS From foregoing discussions, we reach the following conclusions: (a) Both C60 and C70 form ground state noncovalent complexes with a ZnPc derivative, 1, in toluene and DCB. (b) Polar medium facilitates the electrostatic interaction between fullerenes and 1 as evidenced from the well resolved CT absorption bands in the cases of both C60/1 and C70/1 systems when the UV−vis experiment is done in DCB. Utilizing CT transition energies for the fullerene complexes of 1, we have determined several important physicochemical properties like the vertical ionization potential of 1 in DCB, degrees of CT, oscillator strength, transition dipole strength, resonance energies of interaction, and electronic coupling elements. (c) The influences of 1 on the absorption spectra of C60 and C70 are explained using a theoretical model that takes into account the electronic subsystems of 1. (d) Efficient quenching of fluorescence intensity of 1 in the presence of fullerenes takes place in both of the solvents studied under the present investigations. (e) The magnitude of binding constant for fullerene/1 complexes as estimated from UV−vis and steady state fluorescence studies suggests that 1 may not be selectively employed as molecular tweezers for either C60 or C70. (f) Lifetime measurements of 1 in the absence and presence of fullerenes establish the presence of the static quenching mechanism in our present investigations. (g) Ab initio calculations well reproduce the geometry and binding pattern of both C60 and C70 toward 1 in forming effective but not selective supramolecular complexes.

Both ab initio calculations and variable temperature 13C NMR measurements prove that the direction of binding pattern of C70 is end-on toward the plane of 1, which facilitates the electrostatic motif of binding in our present case. Free energy of charge recombination and free energy of charge separation signify that electron transfer from the excited 1 to C60 and C70 in the C60/1 and C70/1 complexes is an unlikely process. Photoinduced energy transfer via TC60* and TC70* from 1 is confirmed by observing the transient absorption spectra in DCB medium. Electronic coupling element values scale the dependency of CT interaction on binding affinity between fullerenes and 1. Finally, we may infer that the foregoing spectroscopic and theoretical studies on fullerene/1 model systems may be of immense interest for interpreting various photophysical and physicochemical parameters of fullerene/1 hybrid systems.

APPENDIX A hv

ISC

C60 ⎯→ ⎯ sC60* ⎯⎯→ TC60*;

T

Ent

T

Ent

C60* + 1 ⎯→ ⎯ C60 + T1*

532

APPENDIX B hv

ISC

C70 ⎯→ ⎯ sC70* ⎯⎯→ TC70*; 532

C70* + 1 ⎯→ ⎯ C70 + T1*

where ISC stands for intersystem crossing and Ent energy transfer.

■

ASSOCIATED CONTENT

S Supporting Information *

HOMO and LUMO energy done by HF calculations for uncomplexed 1, uncomplexed C60, uncomplexed C70, C60/1 and C70/1 systems done in vacuo, UV−vis titration experiment for the C70/1 system in toluene medium, BH plot for the C70/1 system in toluene medium, UV−vis titration experiment for the C70/1 system in DCB medium, BH plot for the C70/1 system in DCB medium, UV−vis absorption spectra of only 1 in DCB along with mixture DDQ, o-chloranil, p-chloranil, and TCNQ in DCB medium, variation of absorbance for the C70/1 system against the molar concentration of C70 in DCB medium, fluorescence spectral variation of 1 in the presence of C60 in DCB medium, BH fluorescence plot of the C60/1 system in DCB, fluorescence spectral variation of 1 in the presence of C70 in DCB medium, BH fluorescence plot of the C70/1 system in DCB given as Table 1S and Figures 1S−6S, respectively. This material is available free of charge via the Internet at http:// pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: sum_9974@rediffmail.com. Fax: +91-342-2530452. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS A.R. thanks Council of Scientific & Industrial Research, New Delhi, for providing a Senior Research Fellowship to her. 11996



Article

(12) Leznoff, C. C.; Lever, A. B. P. Phthalocyanines: Properties and Applications; VCH: New York, 1996. (13) Wróbel, D.; Goc, J.; Ion, R. M. J. Mol. Struct. 1998, 450, 239− 246. (14) Wróbel, D.; Boguta, A.; Ion, R. M. J. Mol. Struct. 2001, 595, 127−138. (15) Kossanyi, J.; Chahraoui, D. Int. J. Photoenergy 2000, 2, 9. (16) Bishop, S. M.; Beeby, A.; Parker, A. W.; Foley, M. S. C.; Philips, D. J. Photochem. Photobiol., A 1995, 90, 39−44. (17) Fernandez, D. A.; Awruch, J.; Dicelio, L. Photochem. Photobiol. 1996, 63, 784−792. (18) Wróbel, D.; Boguta, A.; Ion, R. M. J. Photochem. Photobiol., A 2001, 138, 7−22. (19) Wróbel, D.; Hanyz, I.; Bartkowiak, R.; Ion, R. M. J. Fluoresc. 1998, 8, 191−198. (20) Frackowiak, D.; Planner, A.; Waszokowiak, A.; Boguta, A.; Ion, R. M.; Wiktorowicz, K. J. Photochem. Photobiol., A 2001, 141, 101−108. (21) (a) Kroto, H. W.; Heath, J. R.; O’Brien, S. C.; Curl, R. F.; Smalley, R. E. Nature 1985, 318, 162−163. (b) Rohlfing, E. A.; Cox, D. M.; Kaldor., A. J. Chem. Phys. 1984, 81, 3322−3330. (c) Lee, K.; Song, H.; Park, J. T. Acc. Chem. Res. 2003, 36, 78−86. (22) Martín, N. Chem. Commun. 2006, 2093−2104. (23) (a) Sastre, A.; Gouloumis, A.; Vázquez, P.; Torres, T.; Doan, V.; Schwartz, b. J.; Wudl, Echegoyen, L.; Rivera, J. Org. Lett. 1999, 1, 1807−1810. (b) Guldi, D. M.; Ramey, J.; Martinez-Diaz, M. V.; de la Escosura, A.; Torres, T.; Da Ros, T.; Prato, M. Chem. Commun. 2002, 2774−2775. (c) Claessens, C. G.; González-Rodríguez, D.; Torres, T. Chem. Rev. 2002, 102, 835−854. (d) de la Torre, G.; Claessens, C. G.; Torres, T. Chem. Commun. 2007, 2000−2015. (e) Niemi, M.; Tkachenko, N. V.; Efimov, A.; Lehtivuori, H.; Ohkubo, K.; Fukuzumi, S.; Lemmetyinen, H. J. Phys. Chem. A 2008, 112, 6884− 6892. (f) Lehtivuori, H.; Kumpulainen, T.; Hietala, M.; Efimov, A.; Lemmetyinen, H.; Kira, A.; Imahori, H.; Tkachenko, N. V. J. Phys. Chem. C 2009, 113, 1984−1992. (24) Linssen, T. G.; Dürr, K.; Hanack, M.; Hirsch, A. J. Chem. Soc., Chem. Commum. 1995, 103−104. (25) (a) Taylor, R.; Hare, J. P.; Abdul-Sada, A. K.; Kroto, H. W. J. Chem. Soc., Chem. Commun. 1990, 1423−1425. (b) Guldi, D. M. Chem. Commun. 2000, 321−327. (26) Torres, T.; Gouloumis, A.; Sanchez-Garcia, D.; Jayawickramarajah, J.; Seitz, W.; Guldi, D. M.; Sessler, J. L. Chem. Commun. 2007, 292−294. (27) Bottari, G.; Olea, D.; Navarro, C. G.; Zamora, F.; Herrero, J. G.; Torres, T. Angew. Chem., Int. Ed. 2008, 47, 2026−2031. (28) Ray, A.; Santhosh, K.; Chattopadhyay, S.; Samanta, A.; Bhattacharya, S. J. Phys. Chem. A 2010, 114, 5544−5550. (29) Nojiri, T.; Alam, M. M.; Konami, H.; Watanabe, A.; Ito, O. J. Phys. Chem. A 1997, 101, 7943−7947. (30) Author: (a) Mack, J.; Stillman, M. J. Electronic Structures of Metal Pc and Porphyrin Complexes from Analysis of the UV-Visible Absorption and Magnetic Circular Dichromism Spectra and Molecular Orbital Calculations. In The Porphyrin Handbook; Kadish, K. M., Smith, K. M., Guilard, R., Eds.; Academic Press: San Diego, CA, 2003; Vol. 16, p 43. (b) Gouterman, M. In The Porphyrins; Dolphin, D., Ed.; Academic Press: New York, 1978; Part A. Physical Chemistry. (31) Ray, A.; Goswami, D.; Chattopadhyay, S.; Bhattacharya, S. J. Phys. Chem. A 2008, 112, 11627−11640. (32) Wróbel, D.; Boguta, A. J. Photochem. Photobiol., A 2002, 150, 67−76. (33) Benesi, H. A.; Hildebrand, J. H. J. Am. Chem. Soc. 1949, 71, 2703−2707. (34) Imahori, H.; Tkachenko, N. V.; Vehmancen, V.; Tamaki, K.; Lemmetyinen, H.; Sakata, Y.; Fukuzumi, S. J. Phys. Chem. A 2001, 105, 1750−1756. (35) Gould, I. R.; Noukakis, D.; Gomez-Jahn, L.; Young, R. H.; Goodman, J. L.; Farid, S. Chem. Phys. 1993, 176, 439−456. (36) Thomas, K. G.; Biju, V.; Guldi, D. M.; Kamat, P. V.; George, M. V. J. Phys. Chem. A 1999, 103, 10755. (37) Mulliken, R. S. J. Am. Chem. Soc. 1952, 74, 811−824.

Financial assistance provided by the Department of Science & Technology, New Delhi, through the SERB-DST Project of Ref. No. SR/S1/PC-39/2011 and Centre for Advance Studies, UGC, New Delhi, is also gratefully acknowledged. We also wish to record our gratitude to Prof. Anunay Samanta, School of Chemistry, University of Hyderabad, Hyderabad, India, for his helpful co-operations in this work. We are also very much thankful to Prof. Goodson and the learned reviewers for making valuable comments.

■

REFERENCES

(1) Kaestner, L.; Cesson, M.; Kassab, K.; Christensen, T.; Edminson, P. D.; Cook, M. J.; Chambrier, I.; Jori, G. Photochem. Photobiol. Sci. 2003, 2, 660−667. (2) Fabris, C.; Soncin, M.; Miotto, G.; Fantetti, L.; Chiti, G.; Dei, D.; Roncucci, G.; Jori, G. J. Photochem. Photobiol., B 2006, 83, 48−54. (3) Cauchon, N.; Nader, M.; Bkaily, G.; van Lier, J. E.; Hunting, D. Photochem. Photobiol. 2006, 82, 1712−1720. (4) Fernández, D. A.; Awruch, J.; Lelia E. Dicelio, L. E. Photochem. Photobiol. 1996, 63, 784−792. (5) De Rosa, M. C.; Crutchley, R. J. Coord. Chem. Rev. 2002, 233− 234, 351−371. (6) Sharman, W. M.; van Lier, J. E. In The Porphyrin Handbook; Kadish, K. M., Smith, K. M., Guillard, R., Eds.; Academic Press: San Diego, CA, 2003; Vol. 15, p 1. ́ (7) (a) Kobayashi, N.; Ogata, H.; Nonaka, N.; Lukyanets, E. A. Chem.Eur. J. 2003, 9, 5123−5134. (b) Li, R.; Zhang, X; Zhu, P.; Dennis, K. P. N; Kobayashi, N.; Jiang, J. Inorg. Chem. 2006, 45, 2327− 2334. (c) Wróbel, D.; Boguta, A. J. Photochem. Photobiol., A 2002, 150, 67−76. (8) (a) Kadish, K. M., Smith, K. M., Guillard, R., Eds. The Porphyrin Handbook; Academic Press: San Diego, CA, 2003; Vol. 16. (b) Rio, Y.; Rodríguez-Morgade, S.; Torres, T. Org. Biomol. Chem. 2008, 6, 1877− 1894. (9) Claessens, C. G.; Hahn, U.; Torres, T. Chem. Rec. 2008, 8, 75− 97. (10) (a) Malig, J.; Jux, N.; Cid, J.-J.; Vázquez, P.; Torres, T.; Guldi, D. M. Angew. Chem. 2011, 123, 3623−3627. (b) Malig, J.; Priv.-Doz., N. J.; Kiessling, D.; Cid, J.-J.; Vázquez, P.; Torres, T.; Guldi, D. M. Angew. Chem., Int. Ed. 2011, 50, 3561−3565. (c) Martínez-Díaz, M. V.; de la Torre, G.; Torres, T. Chem. Commun. 2010, 46, 7090−7108. (d) Bessho, T.; Zakeeruddin, S. M.; Yeh, C.-Y.; Diau, E. W.-G.; Grätzel, M. Angew. Chem. 2010, 122, 6796−6799. (e) Bessho, T.; Zakeeruddin, S. M.; Yeh, C.-Y.; Diau, E. W.-G.; Grätzel, M. Angew. Chem., Int. Ed. 2010, 49, 6646−6649. (f) Mori, S.; Nagata, M.; Nakahata, Y.; Yasuta, K.; Goto, R.; Kimura, M.; Taya, M. J. Am. Chem. Soc. 2010, 132, 4054−4055. (g) Martínez-Díaz, M. V.; Torres, T. In Handbook of Porphyrin Science; Kadish, K., Smith, K. M., Guilard, R., Eds.; World Scientific Press: Singapore, 2010; Vol. 10, pp 141−181. (h) Walter, M. G.; Rudine, A. B.; Wamser, C. C. J. Porphyrins Phthalocyanines 2010, 14, 759−792. (i) Cid, J.-J.; García-Iglesias, M.; Yum, J.-H.; Forneli, A.; Albero, J.; Martínez-Ferrero, E.; Vázquez, P.; Grätzel, M.; Nazeeruddin, M. K.; Torres, T; Palomares, E. Chem. Eur. J. 2009, 15, 5130−5137. (j) Hardin, B. E.; Hoke, E. T.; Armstrong, P. B.; Yum, J.-H.; Comte, P.; Torres, T.; Fréchet, J. M. J.; Nazeeruddin, M. K.; Grätzel, M.; McGehee, M. D. Nat. Photonics 2009, 3, 406−411. (k) O’Regan, B. C.; López-Duarte, I.; MartínezDíaz, M. V.; Forneli, A.; Albero, J.; Morandeira, A.; Palomares, E.; Torres, T.; Durrant, J. R. J. Am. Chem. Soc. 2008, 130, 2906−2907. (l) Cid, J.-J.; Yum, J.-H.; Jang, S.-R.; Nazeeruddin, M. K.; MartínezFerrero, E.; Palomares, E.; Ko, J.; Grätzel, M.; Torres, T. Angew. Chem. 2007, 119, 8510−8514. (m) Cid, J.-J.; Jun-Ho Yum, J.-H.; Song-Rim Jang, S.-R.; Nazeeruddin, M. K.; Martínez-Ferrero, E.; Emilio Palomares, E.; Ko, J.; Grätzel, M.; Torres, T. Angew. Chem., Int. Ed. 2007, 46, 8358−8362. (11) Yanık, H.; Aydın, D.; Durmus, M.; Ahsen, V. J. Photochem. Photobiol., A 2009, 206, 18−26. 11997



Article

(38) Bhattacharya, S.; Nayak, S. K.; Chattopadhyay, S.; Banerjee, M.; Mukherjee, A. K. Spectrochim. Acta, Part A 2002, 58, 289−298. (39) Bhattacharya, S.; Banerjee, M.; Mukherjee, A. K. Spectrochim. Acta, Part A 2001, 57, 1463−1470. (40) (a) Foster, R. Organic Charge Transfer Complexes; Academic Press: New York, 1969; Chapters 3 and 13. (41) Peover, M. E. J. Chem. Soc. 1962, 4540−4549. (42) Bhattacharya, S.; Banerjee, M.; Mukherjee, A. K. Spectrochim. Acta, Part A 2001, 57, 2409−2416. (43) Mulliken, R. S.; Person, W. B. Molecular Complexes; Wiley: New York, 1969. (44) Briegleb, G.; Czekalla, J. Z. Phys. Chem. 1960, 24, 37−78. (45) Sun, D. Y.; Tham, F. S.; Reed, C. A.; Chaker, L.; Boyd, P. D. W. J. Am. Chem. Soc. 2002, 124, 6604−6612. (46) Bhattacharya, S.; Chattopadhyay, S.; Nayak, S. K.; Banerjee Bhattacharya, S.; Banerjee, M. Spectrochim. Acta, Part A 2007, 68, 427−431. (47) Yin, G.; Xu, D.; Xu, Z. Chem. Phys. Lett. 2002, 365, 232−236. (48) Guldi, D. M.; Da Ros, T.; Braiuca, P.; Pra to, M. Photochem. Photobiol. Sci. 2003, 2, 1067−1073. (49) Ray, A.; Santhosh, K.; Bhattacharya, A. J. Phys. Chem. A 2011, 115, 9929−9940. (50) Ray, A.; Chattopadhyay, S.; Bhattacharya, S. Spectrochim. Acta, Part A 2011, 79, 1435−1442. (51) Pal, D.; Goswami, D.; Chattopadhyay, S.; Bhattacharya, S. J. Phys. Chem. A 2010, 114, 6776−6786. (52) Claessens, C. G. ; Martínez-Díaz, M. V.; Torres, T. Supramolecular Chemistry: From Molecules To Nanomaterials; John Wiley & Sons Ltd.: U.K., 2012. (53) Wang, M.-X.; Zhang, X.-H.; Zheng, Q.-Y. Angew. Chem., Int. Ed. 2004, 43, 838−842. (54) Dudič, M.; Lhoták, P.; Stibor, I.; Petříčková, H.; Lang, K. New. J. Chem. 2004, 28, 85−90. (55) Shoji, Y.; Tashiro, K.; Aida, T. J. Am. Chem. Soc. 2004, 126, 6570−6571. (56) Mustafizur, R. A. F. M.; Bhattacharya, S.; Peng, X.; Kimura, T.; Komatsu, N. Chem. Commun. 2008, 1196−1198. (57) Mizyed, S.; Tremaine, P. R.; Georghiou, P. E. J. Chem. Soc., Perkin Trans. 2 2001, 3−6. (58) Haino, T.; Yanase, M.; Fukuzawa, Y. Angew. Chem., Int. Ed. 1997, 36, 259−260. (59) Xihuang, Z.; Zhennan, G.; Yongqing, G.; Yiliang, S.; Zhaoxia, J.; Yan, X.; Biyun, S.; Yi, W.; Hua, F.; Jingzun, W. Carbon 1994, 32, 935− 937. (60) Marcus, Y.; Smith, A. L.; Korobov, M. V.; Mirakyan, A. L.; Avramenko, N. V.; Stukalin, E. B. J. Phys. Chem. B 2001, 105, 2499− 2506. (61) Tkachenko, N. V.; Efimov, A.; Lemmetyinen, H. J. Porphyrins Phthalocyanines 2011, 15, 780−790. (62) Takai, A.; Chkounda, M.; Eggenspiller, A.; Gros, C. P.; Lachkar, M.; Barbe, J.-M.; Fukuzumi, S. J. Am. Chem. Soc. 2010, 132, 4477− 4489. (63) Schuster, D. I.; Jarowski, P. D.; Kirschner, A. N.; Wilson, S. R. J. Mater. Chem. 2002, 12, 2041−2047. (64) Hanna, J. V.; Wilson, M. A. J. Phys. Chem. 1992, 96, 6518. (65) Taylor, R. Lecture Notes on Fullerene Chemistry: A Handbook for Chemist; Imperial Press: London, 1999. (66) Guldi, D. M. Pure Appl. Chem. 2003, 75, 1069−1075. (67) Hauke, F.; Hirsch, A.; Liu, S. G.; Echegoyen, L.; Swartz, A.; Luo, C.; Guldi, D. M. ChemPhysChem 2002, 3, 195−205. (68) Kumar, K.; Lin, Z.; Waldeck, D. H.; Zimmt, M. B. J. Am. Chem. Soc. 1996, 118, 243−285. (69) Osuka, A.; Noya, G.; Taniguchi, S.; Okada, T.; Nishimura, Y.; Yamazaki, I.; Mataga, N. Chem.Eur. J. 2000, 6, 33−46. (70) Rehm, D.; Weller, A. Isr. J. Chem. 1970, 7, 259−271. (71) El-Khouly, M. E.; Araki, Y.; Ito, O.; Gadde, S.; McCarty, A. L.; Karr, P. A.; Zandler, M. E.; D’Souza, F. Phys. Chem. Chem. Phys. 2005, 7, 3163−3171.

(72) Sandanayaka, A. S. D.; Araki, Y.; Ito, O.; Deviprasad, G. R.; Smith, P. M.; Rogers, L. M.; Zandler, M. E.; D’Souza, F. Chem. Phys. Lett. 2006, 325, 452−460. (73) Marcus, R. A. J. Chem. Phys. 1999, 110, 5307. Marcus, R. A.; Sutin, N. Biochim. Biophys. Acta 1985, 811, 265. (74) Barbara, P. F.; Meyer, T. J.; Ratner, M. A. J. Phys. Chem. 1996, 100, 13148 (and references therein). (75) Kajii, Y.; Nakagawa, T.; Suzuki, S.; Achiba, Y.; Obi, K.; Shibuya, K. Chem. Phys. Lett. 1991, 181, 100−104. (76) Sension, R. J.; Phillips, C. M.; Szarka, A. Z.; Romanow, W. J.; McGhie, A. R.; McCauely, J. P.; Smith, A. B.; Hochstrasser, R. M. J. Phys. Chem. 1991, 95, 6075−6078. (77) Ebbesen, T. W.; Tanigaki, K.; Kuroshima, S. Chem. Phys. Lett. 1991, 181, 501−504. (78) Bixon, M.; Jortner, J.; Verhoeven, J. W. J. Am. Chem. Soc. 1994, 116, 7349−7355. (79) Gould, I. R.; Farid, S. Acc. Chem. Res. 1996, 29, 522−528. (80) Kumar, K.; Lin, Z.; Waldeck, D. H.; Zimmt, M. B. J. Am. Chem. Soc. 1996, 118, 243. (81) Guldi, D. M.; Zibermann, I.; Gouloumis, A.; Vázquez; Torres, T. J. Phys. Chem. B 2004, 108, 18485−18494.

11998


Photophysical and Theoretical Insights on Fullerene

Recommend Documents