Excited Triplet State Interactions with Molecular Oxygen: Influence of Charge Transfer on the Bimolecular Quenching Rate Constants and the Yields of Singlet Oxygen [O*2(1.DELTA.g)] for Substituted Naphthalenes in Various Solvents

The Effect of Solvent Polarity on the Balance between Charge Transfer and Non-Charge Transfer Pathways in the Sensitization of Singlet Oxygen by ππ* Triplet States

A large set of literature kinetic data on triplet (T(1)) sensitization of singlet oxygen by two series of biphenyl and naphthalene sensitizers in solvents of strongly different polarity has been analyzed. The rate constants and the efficiencies of singlet oxygen formation are quantitatively reproduced by a model that assumes the competition of a non-charge transfer (nCT) and a CT deactivation channel. nCT deactivation occurs from a fully established spin-statistical equilibrium of (1)(T(1)(3)Sigma) and (3)(T(1)(3)Sigma) encounter complexes by internal conversion (IC) to lower excited complexes that dissociate to yield O(2)((1)Sigma(g)(+)), O(2)((1)Delta(g)), and O(2)((3)Sigma(g)(-)). IC of (1,3)(T(1)(3)Sigma) encounter complexes is controlled by an energy gap law that is generally valid for the transfer of electronic energy to and from O(2). (1,3)(T(1)(3)Sigma) nCT complexes form in competition to IC (1)(T(1)(3)Sigma) and (3)(T(1)(3)Sigma) exciplexes if CT interactions between T(1) and O(2) are important. The rate constants of exciplex formation depend via a Marcus type parabolic model on the corresponding free energy change DeltaG(CT), which varies with sensitizer triplet energy, oxidation potential, and solvent polarity. O(2)((1)Sigma(g)(+)), O(2)((1)Delta(g)), and O(2)((3)Sigma(g)(-)) are formed in the product ratio (1/6):(1/12):(3/4) in the CT deactivation channel. The balance between nCT and CT deactivation is described by the relative contribution p(CT) of CT induced deactivation calculated for a sensitizer of known triplet energy from its quenching rate constant. It is shown how the change of p(CT) influences the quenching rate constant and the efficiency of singlet oxygen formation in both series of sensitizers. p(CT) is sensitive to differences of solvent polarity and varies for the biphenyls and the naphthalenes as sigmoidal with DeltaG(CT). This quantitative model represents a realistic and general mechanism for the quenching of pipi triplet states by O(2), surpassing previous advanced models.

Quantitative Determination of 1Σg+ and 1Δg Singlet Oxygen in Solvents of Very Different Polarity. General Energy Gap Law for Rate Constants of Electronic Energy Transfer to and from O2 in the Absence of Charge Transfer Interactions

The quenching of excited triplet states of sufficient energy by O2 leads to O2(1sigma(g)+) and O2(1delta(g)) singlet oxygen and O2(3sigma(g)-) ground-state oxygen as well. The present work investigates the question whether in the absence of charge transfer (CT) interactions between triplet sensitizer and O2 the rate constants of formation of the three different O2 product states follow a generally valid energy gap law. For that purpose, lifetimes of the upper excited O2(1sigma(g)+) have been determined in a mixture of 7 vol % benzene in carbon tetrachloride, in chloroform, and in perdeuterated acetonitrile. They amount to 1.86, 1.40, and 0.58 ns, respectively. Furthermore, rate constants of O2(1sigma(g)+), O2(1delta(g)), and O2(3sigma(g)-) formation have been measured in these three solvents for five pi pi* triplet sensitizers with negligible CT interactions. The rate constants are independent of solvent polarity. After normalization for the multiplicity of the respective O2 product state, the rate constants follow a common dependence on the excess energies of the respective product channels. This empirical energy gap relation describes also quantitatively the rate constants of quenching of O2(1delta(g)) by 28 carotenoids. Therefore, it represents in the absence of CT interactions a generally valid energy gap law for the rate constants of electronic energy transfer to and from O2.

Quenching Mechanism of Rose Bengal Triplet State Involved in Photosensitization of Oxygen in Ethylene Glycol

The quenching rate constants of the excited triplet state of Rose Bengal (RB) by oxygen (k(obs)) were measured in ethylene glycol (EG) at different temperatures using nanosecond laser flash photolysis. Although a plot of the quenching rate constant k(obs) for RB triplet state vs oxygen concentration is linear at 20 degrees C, the oxygen dependence of k(obs) does not exhibit linearity but upward curvature at high temperatures from 130 to 140 degrees C. The upward curvature at high temperatures is not well-described by a kinetic scheme first postulated by Gijzeman et al., which is characterized by exciplex formation and a unimolecular dissociation of the exciplex to products, but instead by a more comprehensive mechanism involving a bimolecular dissociation in addition to a unimolecular one. The measurements of the oxygen dependence of k(obs) for RB triplet state at different temperatures yielded a reaction enthalpy for the exciplex formation of 150 kJ mol(-1). Due to the large exothermic reaction enthalpy, equilibrium was obtained for the exciplex at 20 degrees C even at low oxygen concentration and the bimolecular quenching by oxygen became the major dissociation process. The equilibrium attainment and bimolecular dissociation provide a linear oxygen dependence of k(obs) to all outward appearances. Therefore, linearity does not always mean that exciplex dissociation proceeds solely through a unimolecular mechanism.

Photosensitized Generation of Singlet Oxygen

This work gives an overview of what is currently known about the mechanisms of the photosensitized production of singlet oxygen. Quenching of pi pi* excited triplet states by O2 proceeds via internal conversion of excited encounter complexes and exciplexes of sensitizer and O2. Both deactivation channels lead with different efficiencies to singlet oxygen generation. The balance between the deactivation channels depends on the triplet-state energy and oxidation potential of the sensitizer, and on the solvent polarity. A model has been developed that reproduces rate constants and efficiencies of the competing processes quantitatively. Sensitization by excited singlet states is much more complex and hence only qualitative rules could be elaborated, despite serious efforts of many groups. However, the most important deactivation paths of fluorescence quenching by O2 are again directed by excess energies and charge-transfer interactions similar to triplet-state quenching by O2. Finally, two recent developments in photosensitization of singlet oxygen are reviewed: Two-photon sensitizers with particular application potential for photodynamic therapy and fluorescence imaging of biological samples and singlet oxygen sensitization by nanocrystalline porous silicon, a material with very different photophysics compared to molecular sensitizers.

Singlet Oxygen Quantum Yield Determination for a Fluorene-Based Two-Photon Photosensitizer

The quantum yield, Phi(Delta), of singlet oxygen generation under two-photon excitation has been determined for a fluorene derivative. A photochemical method was developed using 1,3-diphenylisobenzofuran (DPBF), a chemical quencher of 1O2, and 2-(9,9-didecyl-7-nitrofluoren-2-yl)benzothiazole (1) as a two-photon photosensitizer (PS). The photochemical kinetics of the quencher was measured by two different fluorescence methods. Fluorene 1 exhibited relatively high singlet oxygen quantum yield, Phi(Delta) approximately 0.4 +/- 0.1, and had a two-photon absorption cross-section of 28 +/- 5 GM. Thus, 1 may have potential for use as a two-photon PS in the near-IR spectral region for biomedical applications.

The balance between charge transfer and non-charge transfer pathways in the sensitization of singlet oxygen by ππ* triplet states

A charge transfer (CT) channel and a non-CT deactivation channel, both leading to formation of O(2)((1)Sigma (g)(+)), O(2)((1) Delta(g)) and O(2)((3)Sigma(g)(-)), compete in the quenching of triplet states by O(2). Recent studies by our group demonstrated that these channels are described by rather simple and general quantitative relations. In the present paper we use the detailed kinetic data on the quenching by O(2) of pi pi* triplet sensitizers of three homologous aromatic series in CCl(4) to derive a parameter, which describes the balance between CT and non-CT deactivation. This quantity, p(CT), is the relative contribution of CT mediated deactivation and is easily calculated for a sensitizer of known triplet energy from its quenching rate constant. The parameter p(CT) quantitatively describes the balance between both deactivation channels without requiring any knowledge of oxidation potentials. It is shown how the variation of p(CT) influences the efficiencies and the rate constants of O(2)((1)Sigma(g)(+)), O(2)((1)Delta(g)) and O(2)((3)Sigma(g)(-)) formation in the quenching process.

Direct detection of singlet oxygen via its phosphorescence from cellular and fungal cultures

In summary, we have developed methods to detect 1O2 spectrally from viable keratinocytes and fungal mycelial cultures. Both time-resolved and steady-state spectrophotometers were used, providing complementary information on optimal conditions for the unambiguous detection of 1O2 phosphorescence. By using our techniques, we were able to confirm that photosensitizers in contact with living cells indeed generate 1O2. The model systems and the procedures described can be adopted for other studies involving 1O2 and oxidative stress in living cells.

Sensitization of singlet oxygen via encounter complexes and via exciplexes of pipi* triplet excited sensitizers and oxygen

Both excited singlet states, 1sigma(g)+ and 1delta(g), and the triplet ground state, 3sigma(g)-, of molecular oxygen are competitively formed during the quenching by O2 of triplet (T1) excited sensitizers of sufficient energy. The corresponding overall rate constants kT(1sigma), kT(1delta), and kT(3sigma) as well as the T1 state energies E(T) and the oxidation potentials E(ox), have been determined for a series of six fluorene derivatives. Graduated and in part strong charge transfer (CT) effects on kT(1sigma), kT(1delta), and kT(3sigma) are observed. These and literature data strongly indicate that quenching occurs in two different channels each capable of producing O2(1sigma(g)-), O2(1delta(g)), and O2(3sigma(g)-). One proceeds via internal conversion (IC) of excited 1,3(T1 x 3sigma) complexes with no CT character (nCT), which cannot be distinguished from encounter complexes, the other via IC of 1,3(T1 x 3sigma) exciplexes with partial CT character (pCT). The contributions of nCT and pCT deactivation channels to the overall formation of O2(1sigma(g)+), O2(1delta(g)). and O2(3sigma(g)-) depend on E(T) and E(ox). The rate constants of the nCT channel are controlled by the excess energies of the respective IC processes by an energy gap law. The rate constants of the pCT channel depend on the change of free energy deltaG(CET) for complete electron transfer from T1 excited sensitizer to O2. Equations are presented which show the functional form of the dependence of the oxygen quenching rate constants on E(T) and E(ox). Particular emphasis is laid on the question of whether these relations could generally be valid for pipi* triplet sensitizers.

Photochemical studies of 4-tert-butyl-4'-methoxydibenzoylmethane (BM-DBM)

Kinetic UV-VIS absorption data following 355 and 266 nm nanosecond laser flash photolysis of 4-tert-butyl-4'-methoxydibenzoylmethane (BM-DBM) solutions is presented. The kinetics of the decay of the non-chelated enol (NCE) produced following 355 nm excitation of BM-DBM solutions are analysed in terms of mixed 1st- and 2nd-order kinetics. The temperature dependences of the component rate constants are unusual and both 1st- and 2nd-order components display negative activation energies, which are explained by invoking pre-equilibria. In addition, it is shown for the first time that 266 nm laser photolysis of BM-DBM solutions leads to formation of the triplet state of the keto (K) form with a lifetime of approximately 500 ns. Under these conditions the triplet state of the K form is quenched by oxygen.

Mechanistic and kinetic aspects of photosensitization in the presence of oxygen

Determining whether the first step of photooxygenation is Type I or Type II is a necessary prerequisite in order to establish the mechanism of photodynamic action. But this distinction is not sufficient, because other processes, both consecutive and competitive, commonly participate in the overall mechanism. Thus, in both Type I and Type II reactions, the initial products are often peroxides that can break down and induce free radical reactions. These aspects of photosensitization are discussed and illustrated by sensitizer/substrate systems involving (1) only radical reactions (decatungstate/alkane) and (2) reactions of singlet oxygen occurring in competitive and consecutive processes and possibly followed by radical reactions (methylene blue/2'-deoxyguanosine). Two other previously investigated systems involving, respectively, a Type II interaction followed by radical processes (methylene blue/alkene) and Type II reactions, some of which being competitive or consecutive (rose bengal/alkene), are briefly reconsidered.

Photophysics of the cationic 5,10,15,20-tetrakis (4-N-methylpyridyl) porphyrin bound to DNA, [poly (dA-dT)]2 and [poly (dG-dC)]2: interaction with molecular oxygen studied by porphyrin triplet-triplet absorption and singlet oxygen luminescence

Interaction between molecular oxygen and the cationic free-base 5,10,15,20-tetrakis (4-N-methylpyridyl) porphyrin (H2TMpyP4+) complexed with [poly (dA-dT)]2, [poly (dG-dC)]2 and calf thymus DNA, has been monitored in air-saturated heavy water solutions through porphyrin triplet-triplet absorption and singlet oxygen luminescence. Three different rate constants of porphyrin triplet state quenching have been found which correspond to different accessibilities of molecular oxygen to porphyrins embedded in the duplexes. The longest triplet state lifetime (30 microseconds), found for porphyrin bound to [poly (dG-dC)]2, corresponds to molecules well protected from oxygen. This supports the hypothesis of an intercalative binding mode of the porphyrin between GC base-pairs ('type A' sites). The fraction fT delta of the porphyrin triplet states quenched by molecular oxygen with singlet oxygen generation, is unity. In [poly (dA-dT)]2-porphyrin complexes, two sites ('type B' and 'C' sites of interaction) are involved, yielding very different triplet state lifetimes (5.5 microseconds and 20.5 microseconds) and efficiencies of singlet oxygen generation (fT delta = 0.50 and 0.82). The fT delta decreases can likely be explained in terms of competition between energy and electron transfer from the porphyrin excited triplet state to molecular oxygen. All three types (A, B and C) of interaction sites can be expected in porphyrin-DNA complexes.

Singlet oxygen and superoxide characteristics of a series of novel asymmetric photosensitizers

The singlet oxygen quantum yields and superoxide quantum yields for a series of novel compounds based on an asymmetrical protoporphyrin molecule have been examined. Electron spin resonance was used to measure superoxide yield and time resolved luminescence for singlet oxygen. A comparison between these results and previously published cell survival data was carried out. A broad association was found between singlet oxygen quantum yield and clonogenic cell kill.

Optimization of a pulse laser spectrometer for the measurement of the kinetics of singlet oxygen O2(delta 1g) decay in solution

A simple sensitive pulse laser spectrometer to measure the kinetics of singlet oxygen decay in solution is described. It utilizes a NdYag laser for pulse excitation of the 1O2 photosensitizers perinaphthenone and Rose Bengal or eosine at 355 nm and 532 nm, respectively. Singlet oxygen phosphorescence is detected by a germanium diode, to which collimated phosphorescence light is directed by concave and parabolic mirrors via a pair of Fresnel lenses and an infrared interference filter. The 1O2 transient decays are monitored using a Hewlett Packard digitizing oscilloscope and a PC computer. The excellent sensitivity of this spectrometer allows us to easily observe 1O2 decay (1 microsecond resolution) after a single pulse of laser excitation, which is important for quenching measurements of chemical substrates that react readily with 1O2. The small sample volume (0.1 ml) prepared in quartz tubes permits fast and inexpensive data acquisition for the determination of 1O2 quenching rate constants in most solvents.