Photodissociation of N-Arylmethylanilines: A Laser Flash Photolysis, Fluorescence, and Product Analysis Study(1)

Design and Photo-Induced Dynamics of Radical-Chromophore Adducts with One- or Two-Atom Separation: Toward Potential Probes for High Field Optical DNP Experiments

Breaking the maximum enhancement barrier of 660 at room temperature in a conventional dynamic nuclear polarization (DNP) experiment has the immense potential of practical applications. Optical DNP experiments with radical-chromophore (RC) adducts, which harnesses hyperpolarized radicals, instead of thermalized radicals, offers a powerful way to achieve this. Typical DNP and NMR experiments, however, are carried out at high magnetic fields of about 5-10 T, whereas the large electron spin hyperpolarization (ESP) demonstrated in the RC adducts so far are at a much low field of 0.3 T. Thus, in order to realize a successful optical DNP experiment, it is imperative to ask whether the RC adducts, which are currently available, can achieve a large ESP even at high fields. The present work poses this question and shows that the current RC adducts would not generate a large ESP at high fields unless the separation between the chromophore and nitroxyl moiety is reduced to less than four bonds. Two serious bottlenecks in this direction are the near impossibility of synthesizing such RC adducts using the common nitroxyl radicals and the absence of any photophysical studies on RC adducts with such short spacer groups. In this regard, the present work exploits the spin trapping methodology to synthesize one- and two-atom separated naphthalene-nitroxyl RC adducts. Good yields and excellent stability of the adducts have been demonstrated. Furthermore, the present work presents their detailed photophysical and photochemical studies by transient optical and time-resolved EPR studies. On the basis of the present results, a potential RC adduct is proposed for the high field optical DNP experiments. Finally, the prospect of exploiting the large EPR signal enhancement due to ESP in the field of spin trapping studies has been discussed.

Ground- and triplet excited-state properties correlation: a computational CASSCF/CASPT2 approach based on the photodissociation of allylsilanes

Excited-state properties, although extremely useful, are hardly accessible. One indirect way would be to derive them from relationships to ground-state properties which are usually more readily available. Herewith, we present quantitative correlations between triplet excited-state (T₁) properties (bond dissociation energy, D₀(T₁), homolytic activation energy, E(a)(T₁), and rate constant, k(r)) and the ground-state bond dissociation energy (D₀), taking as an example the photodissociation of the C-Si bond of simple substituted allylsilanes CH₂=CHC(R¹R²)-SiH₃ (R¹ and R² = H, Me, and Et). By applying the complete-active-space self-consistent field CASSCF(6,6) and CASPT2(6,6) quantum chemical methodologies, we have found that the consecutive introduction of Me/Et groups has little effect on the geometry and energy of the T₁ state; however, it reduces the magnitudes of D₀, D₀(T₁) and E(a)(T₁). Moreover, these energetic parameters have been plotted giving good linear correlations: D₀(T₁) = α₁ + β₁ · D₀, E(a)(T₁) = α₂ + β₂ · D₀(T₁), and E(a)(T₁) = α₃ + β₃ · D₀ (α and β being constants), while k(r) correlates very well to E(a)(T₁). The key factor behind these useful correlations is the validity of the Evans-Polanyi-Semenov relation (second equation) and its extended form (third equation) applied for excited systems. Additionally, the unexpectedly high values obtained for E(a)(T₁) demonstrate a new application of the principle of nonperfect synchronization (PNS) in excited-state chemistry issues.

Novel pathway for N1-acetyl-5-methoxykynuramine: UVB-induced liberation of carbon monoxide from precursor N1-acetyl-N2-formyl-5-methoxykynuramine

Irradiation of the melatonin metabolite N(1)-acetyl-N(2)-formyl-5-methoxykynuramine (AFMK) with UV light of 254 nm causes the release of carbon monoxide (CO) and, thus, deformylation to N(1)-acetyl-5-methoxykynuramine (AMK). Liberation of CO was demonstrated by reduction of PdCl(2) to metallic palladium, under avoidance of actions by other reductants. Photochemical AMK formation was not due to UV-induced hydroxyl radicals, because the reaction also took place with high efficiency in ethanol and 2-propanol. Moreover, AMK was generated from AFMK by UVB on a dry thin layer chromatographic plate. Although AMK seems to be the major primary product generated by UVB radiation, prolonged exposure of AFMK led to various other products, especially formed by destruction of AMK, as shown by irradiation of this latter compound. With regard to the demonstration of melatonin in skin and substantial amounts of AFMK in keratinocytes, these findings may be of dermatologic relevance.

Competitive Consecutive Electron Transfer in Determination of Ionization Potentials: Ketene Derivatives

Kinetics of competitive consecutive electron transfer was used to determine ionization potentials of transient species. Kinetics of two-stage electron transfer reactions in aprotic solvent was studied using 355 nm laser flash photolysis. The concentrations of transients produced by the laser flash photolysis were monitored by their light absorption. Triplet-excited tetrachloro-p-benzo-quinone (p-chloranil) generated by a 355 nm laser flash oxidized diethyl ketene, diphenyl ketene, or phenyl ethyl ketene to form radical cations. The ketene radical cations, in turn, oxidized tertiary amine, forming ground state ketene and ammonium radical cation. The kinetics of the disappearance of ketene radical cations (and/or appearance of ammonium radical cations) due to consecutive, competitive electron transfer to ketene and p-chloranil radical cations was monitored. By monitoring kinetics in the presence of tertiary amines with different oxidation potentials, it was established that in acetonitrile the oxidation potential of diethyl ketene was 5.4 eV; for phenyl ethyl ketene, it was approximately 4.8 eV; and for diphenyl ketene, it was 4.6 eV. The results were in agreement with the oxidation potentials of ketenes computed using published data.

Triplet- vs. singlet-state imposed photochemistry. The role of substituent effects on the photo-Fries and photodissociation reaction of triphenylmethyl silanes

The photochemistry of three structurally very similar triphenylmethylsilanes 1, 2, 3 [p-X-C(6)H(4)-CPh(2)-SiMe(3): X = PhCO, 1; H, ; Ph(OCH(2)CH(2)O)C, 3] is described by means of 248 and 308 nm nanosecond laser flash photolysis (ns-LFP), femtosecond LFP, EPR spectroscopy, emission spectroscopy (fluorescence, phosphorescence), ns-pulse radiolysis (ns-PR), photoproduct analysis studies in MeCN, and X-ray crystallographic analysis of the two key-compounds 1 and 2. The photochemical behavior of 1, 2 and 3 is discussed and compared with that of a fourth one, 4, bearing on the p-position an amino group (X = Me(2)N) and whose detailed photochemistry we reported earlier (J. Org. Chem., 2000, 65, 4274-4280). Silane 1 undergoes on irradiation with 248 and 308 nm laser light a fast photodissociation of the C-Si bond giving the p-(benzoyl)triphenylmethyl radical (1*) with a rate constant of k(diss)= 3 x 10(7) s(-1). The formation of 1* is a one-quantum process and takes place via the carbonyl triplet excited state with high quantum yield (Phi(rad)= 0.9); the intervention of the triplet state is clearly demonstrated through the phosphorescence spectrum and quenching experiments with ferrocene (k(q)= 9.3 x 10(9) M(-1) s(-1)), Et(3)N (1.1 x 10(9) M(-1) s(-1)), and styrene (3.1 x 10(9) M(-1) s(-1)) giving quenching rate constants very similar to those of benzophenone. For comparative reasons radical 1* was generated independently from p-(benzoyl)triphenylmethyl bromide via pulse radiolysis in THF and its absorption coefficient at lambda(max)= 340 nm was determined ([epsilon]= 27770 M(-1) cm(-1)). We found thus that the p-PhCO-derivative 1 behaves similar to the p-Me(2)N one (the latter giving the p-(dimethylamino)triphenylmethyl radical with Phi(rad)= 0.9), irrespective of their completely different ground state electronic properties. In contrast, compounds 2, 3 that bear only the aromatic chromophore give by laser or lamp irradiation both, (i) radical products [Ph(3)C* and p-Ph(OCH(2)CH(2)O)C-C(6)H(4)-C(*)Ph(2), respectively] after dissociation of the central C-Si bond (Phi(rad)= 0.16), and (ii) persistent photo-Fries rearrangement products (of the type of 5-methylidene-6-trimethylsilyl-1,3-cyclohexadiene) absorbing at 300-450 nm and arising from a 1,3-shift of the SiMe(3) group from the benzylic to the ortho-position of the aromatic ring (Phi approximately 0.85 for 2). Using fs-LFP on 2 we showed that the S(1) state recorded at 100 fs after the pulse decays on a time scale of 500 fs giving Ph(3)C* through C-Si bond dissociation. In a second step and within the next 10 ps trityl radicals either escape from the solvent cage (the quantum yield of Ph(3)C* formation Phi(rad)= 0.16 was measured with ns-LFP), or undergo in-cage recombination to photo-Fries products. Thus, singlet excited states (S(1)) of the aromatic organosilanes (2, 3) prefer photo-Fries rearrangement products, while triplet excited states (1, 4) favor free radicals. Both reactions proceed via a common primary photodissociation step (C-Si bond homolysis) and differentiate obviously in the multiplicity of the resulting geminate radical pairs; singlet radical pairs give preferably photo-Fries products following an in-cage recombination, while triplet radical pairs escape the solvent cage (MeCN). The results demonstrate the crucial role which is played by the chromophore which prescribes in a sense, (i) the multiplicity of the intervening excited state and consequently that of the resulting geminate radical pair, and (ii) the dominant reaction path to be followed: the benzophenone- and anilino-chromophore present in silanes 1 and 4, respectively, impose effective intersystem crossing transitions (k(isc)= 10(11) s(-1) and 6 x 10(8) s(-1), respectively) leading to triplet states and finally to free radical products, while the phenyl chromophore in 2 and 3, possessing ineffective isc (k(isc)= 6 x 10(6) s(-1)) leads to photo-Fries product formation via the energetic high lying S(1) state [approximately 443 kJ mol(-1)(106 kcal mol(-1))].

Photochemistry and photophysics of triarylmethane dye leuconitriles

The photochemical reactions of crystal violet leuconitrile (CVCN) were investigated by the means of product analysis and trapping experiments, laser flash and steady-state photolysis, and steady-state fluorescence. The influence of oxygen on the reaction was examined in detail. The photochemistry of malachite green leuconitrile (MGCN), basic fuchsin leuconitrile (BFCN), and crystal violet leucomethyl (CVMe) and leucobenzyl (CVBn), as well as triphenylacetonitrile, was studied. The results suggest ionization occurs from S1, while the di-pi-methane reaction is the photochemical route from T1.

1,3-dimethoxy-5-methylene-1,3-cyclohexadiene compounds with leaving groups at C6: generation, solvolytic reactivity, and their importance in the photochemistry of 3,5-dimethoxybenzyl derivatives

The photochemistry of 3,5-dimethoxybenzyl compounds with the leaving groups acetate (1a), chloride (1b), bromide (1c), iodide (1d), diethyl phosphate (1e), and trimethylamine (1f), as the chloride, was examined by both product studies and flash photolysis. The isomeric triene, 5-methylene-1,3-cyclohexadiene derivative was observed for the acetate (2a), diethyl phosphate (2e) and trimethylammonium chloride (2f). The solvolysis of these derivatives, 2, was examined in alcohol solvents and the rate correlation with YOTS values gave m = 0.47 (2a) and 0.63 (2e), suggesting SN1 reactivity but with an early transition state. Quantum yields for formation of 2a and 2e indicated that these trienes play only a minor role (approximately 16%) in the overall photochemistry of the corresponding arylmethyl substrates.

Mechanism of the Photodissociation of 4-Diphenyl(trimethylsilyl)methyl-N,N-dimethylaniline

On irradiation in hexane (248- and 308-nm laser light) 4-diphenyl(trimethylsilyl)methyl-N,N-dimethylaniline, 2, undergoes photodissociation of the C-Si bond giving 4-N,N-dimethylamino-triphenylmethyl radical, 3(*) (lambda(max) at 343 and 403 nm), in very high quantum yield (Phi = 0.92). The intervention of the triplet state of 2 (lambda(max) at 515 nm) is clearly demonstrated through quenching experiments with 2,3-dimethylbuta-1,3-diene, styrene, and methyl methacrylate using nanosecond laser flash photolysis (LFP). The formation of 3(*) is further demonstrated using EPR spectroscopy. The detection of the S(1) state of 2 was achieved using 266-nm picosecond LFP, and its lifetime was found to be 1400 ps, in agreement with the fluorescence lifetime (tau(f) = 1500 ps, Phi(f) = 0.085). The S(1) state is converted almost exclusively to the T(1) state (Phi(T) = 0.92). In polar solvents such as MeCN, 2 undergoes (1) photoionization to its radical cation 2(*)(+), and (2) photodissociation of the C-Si bond, giving radical 3(*) as before in hexane. The formation of 2(*)(+) occurs through a two-photon process. Radical cation 2(*)(+) does not fragment further, as would be expected, to 3(*) via a nucleophile(MeCN)-assisted C-Si bond cleavage but regenerates the parent compound 2. Obviously, the bulkiness of the triphenylmethyl group prevents interaction of 2(*)(+) with the solvent (MeCN) and transfer to it of the electrofugal group Me(3)Si(+). The above results of the laser flash photolysis are supported by pulse radiolysis, fluorescence measurements, and product analysis.