A Theoretical Elucidation on the Solvent-dependent Photosensitive Behaviors of C60
ABSTRACT
In this paper, the solvent-dependent photosensitive behaviors of fullerene (C60) were investigated in polar and nonpolar solvents by time-dependent density functional theory (TD-DFT) calculation. Based on the calculated physicochemical parameters on triplet state, it is revealed that excited-state C60 only generates 1O2 via energy transfer in benzene, but can give birth to O2˙− and 1O2 in water via energy transfer and electron transfer, respectively. Considering the fact that electron transfer is more favorable compared with energy transfer in polar biological systems, especially with the presence of electron donors, the O2˙−-generating process will get predominant in physiological systems. These results account well for the experimental observations that O2˙− and ˙OH are primarily responsible for the photoinduced DNA cleavage by C60 under physiological conditions, whereas 1O2 plays a critical role in nonpolar solvents.
INTRODUCTION
Fullerene (C60) possesses many unusual properties (1–5), among which photosensitive activity was very attractive because of the following advantages (4–8): long wavelength of absorption, high yields of reactive oxygen species (ROS) and lack of acute toxicity in the absence of light. Accordingly, C60 is promising for the development of a new generation of photodynamic therapy (PDT) drugs.
Much effort has been devoted to investigating the photosensitive behaviors of C60 (5,8–11). It has been revealed that ROS generated by C60 at excited state are primarily responsible for its photo-dynamic effects, such as photocleavage of DNA (5,8–11). More interestingly, in different systems, the C60-induced photodamage is governed by different ROS (4,9–11). For instance, in nonpolar solvents (such as benzene). 1O2 is the initiator (4,9,10), whereas under simulated physiological conditions (water with the presence of physiological concentrations of electron donors, e.g. NADH), O2˙− and ˙OH play a key role in the photosensitization (11). The solvent-dependent photosensitive behaviors of C60 aroused our interest to provide a theoretical elucidation. Considering the successful use of time-dependent density functional theory (TD-DFT) in investigating the photo-physicochemical properties of photo-sensitizers (12–15), we attempt to achieve the goal by TD-DFT calculations, which will help to direct the further development of C60 as photodynamic drugs.
MATERIALS AND METHODS
The molecular structure of C60 was fully optimized by hybrid density functional theory (DFT) (16,17) and B3LYP functional (18–20) with 6–31G(d) Gaussian basis set (21), which has been proven sufficient to give accurate structural parameters of C60 (22). Then, the excited-state properties of C60 were calculated by TD-DFT formalism with the same basis set (23–25), by which the triplet-state (T1) energies (ET1) were obtained. The total electronic energies of anion and cation radicals were also calculated to estimate the ground-state vertical electron affinity (VEA) and vertical ion-ization potential (VIP) of C60. Then, the VEA and VIP in T1 state (VEAT1 and VIPT1) were estimated according to the following equations: VEAT1= VEASO - ET1; VIPT1= VIPSO - ET1. During the calculations, both in vacuo and in solvent models were considered. The solvent (benzene and water) effects were taken into account by employing the self-consistent reaction field (SCRF) method with polarizable continuum model (PCM) of Tomasi and coworkers (26–28). It has been demonstrated that the present method can precisely predict the absorption peak and triplet-state properties of photosensitizers (12–15). All of the calculations were performed with the Gaussian 03 package of programs (29).
RESULTS AND DISCUSSION
T1 excitation energy
As known to all, during the photosensitization, SO-state photosensitizers are initially excited to the singlet excited state (SI) and then intersystem cross to the T1 state, which is responsible for the photosensitive reactions. Thus, the lowest triplet excitation energy (ET1) of a photosensitizer is crucial to understanding the photosensitive mechanisms. Therefore, the ET1 of C60 were calculated first.
The ET1 of C60in vacuo, benzene and water are similar to each other (1.59 vs 1.59 vs 1.60 electron volts [eV]; Table 1), indicating that ET1 is slightly influenced by the environment. In addition, the theoretical ET1 in benzene is close to the experimental value (1.63 ± 0.20 eV) (5), justifying the present method. As it is still a challenge to accurately determine ET1 of photosensitizers by experiments, the present results suggest that TD-DFT is an alternative approach for estimating the ET1 of other fullerenes.
| E TI | VEASO | VEATI* | VIPSO | VIPTI† | |
|---|---|---|---|---|---|
| In vacuo | 1.59 | −1.99 | −3.58 | 7.23 | 3.65 |
| In benzene | 1.59 | −2.69 | −4.28 | 6.44 | 4.85 |
| In water | 1.60 | −3.21 | −4.81 | 5.74 | 4.14 |
- *VEAT1= VEASO - ET1.
- † VIPT1= VIPSO - ET1.
Elucidation of photosensitive mechanisms of C60
As to pathway 2, the prerequisite for the reaction is that the summation of VIPT1 of C60 and the adiabatic electron affinity (AEA) of 3O2 is negative. As shown in Table 1, the VIPT1 of C60 is 4.85 eV in benzene, and the corresponding AEA of 3O2 is −2.33 eV calculated by the B3LYP/6–31+G(d,p) method (12,13). Therefore, O2˙− cannot be generated in benzene by electron transfer between T1-state C60 and 3O2. On the other hand, the VIPT1 of C60 is 4.14 eV in water, whereas the AEA of 3O2 is estimated to be −3.89 eV using the B3LYP/6–31+G(d,p) method in the same solvent. According to the above criterion, O2˙− could not be generated through this pathway in water either.
As to pathway 3 (Eqs. 3–5), the reaction precondition is that the energy summation of each reaction is negative. In benzene, the VEAT1 (VIPT1) of C60 is −4.28 eV (4.85 eV), whereas the VIPSO (VEASO) is 6.44 eV (−2.69 eV) (Table 1). Thus, the total energy of reaction 3 is positive, and thus the reaction is forbidden. Reaction 4 is also forbidden in benzene because of its positive total reaction energy (VEAT1 (−4.28 eV)+VIPT1 (4.85 eV) = 0.57 eV). Therefore, neither reaction 3 nor reaction 4 can proceed in benzene according to the present calculations. In fact, even if C60˙− is generated, it cannot pass its excess electron to 3O2 to form O2˙− (Eq. 5) due to its positive total reaction energy (AEAO2+ VIP (C60˙−) = 0.42 eV).
A similar analysis indicates that reaction 3 is forbidden in water. However, thanks to the negative total reaction energy (VEAT1 (-4.81 eV) + VIPT1 (4.14 eV) =−0.67 eV), reaction 4 is permitted. Furthermore, C60˙− can donate its excess electron to 3O2 to generate O2˙− (Eq. 5) in water, because the total energy of reaction 5 is negative (AEAO2+ VIP (C60˙2−) =−0.75 eV). This is consistent with the common notion that electron transfer reaction is favored in polar solvents. Once O2˙− is given, other ROS, such as H2O2 and ˙OH, can be produced through the Fenton reaction (30) or Haber-Weiss reaction (31).
Photosensitive mechanisms of C60.
CONCLUSION
TD-DFT-derived excited-state properties of C60, such as lowest triplet-state energy, vertical electron affinity and vertical ionization potential, can provide reasonable explanations for the experimentally observed photosensitive behaviors of the molecule. Although, in general, molecules in excited states react from their equilibrium geometries, it seems that the TD-DFT-derived vertical excited-state energies are applicable. The validity of the present calculation partially arises from the fact that the structure of C60 is highly symmetric, conjugative and rigid, which will result in small structural deviation of excited states from that of ground state and thus guarantees the accuracy of the calculations.
Acknowledgments
This work was supported by the National Basic Research Program of China (2003CB114400) and the National Natural Science Foundation of China (30100035 and 30570383).
