Initial Radical Separation after Photolysis of 2,2′‐Azobis(isobutyronitrile) (AIBN) in Solution: Modeling the Primary Cage Effect for Polar Radicals

There is a contradiction as to the initial spatial separation r_i of the two transient 2‐cyanoprop‐2‐yl radicals (Me₂ ? CN) formed by flash photolysis of 2,2′‐azobis(isobutyronitrile) (AIBN) in solvents of various viscosities. The cage effect, expressed in terms of the in‐cage termination probability of the resulting radicals, is predicted correctly by classical Langevin models assuming a decrease of r_i with increasing viscosity. However, the electron‐spin polarization of the radicals escaping the primary cage clearly indicates that the initial separation distance r_i is independent of the solution viscosity. This obvious discrepancy can be reconciled by accounting for the strong electric dipole moments of these radicals and the resulting inter‐radical dipole? dipole interaction potential. We propose a primary‐caging model for polar radicals in solution based on an attractive inter‐radical mean‐force potential. The model is applied to the flash photolysis of AIBN and shown to describe properly the viscosity dependence of both the in‐cage termination probability (cage effect) and the electron‐spin polarization of the escaping 2‐cyanoprop‐2‐yl radicals.     

Microviscosity and wavelength effects on radical cage pair recombination

This study probed two aspects of the reactivity of geminate radical cage pairs formed by photolysis of (Cp′ = η⁵-C₅H₄CH₃). The first aspect studied examined whether the bulk viscosity has any predictive power in determining the magnitude of the cage recombination efficiency (F_cP). Although there is a clear relationship between the magnitude of F_cP and viscosity for systems where the bulk viscosity of the solution is altered by the addition of a non-macroscopic viscosity enhancer, the relationship is unclear for systems where the bulk viscosity is altered using polymeric viscosity enhancers. For this investigation, F_cP values were measured using femtosecond pump-probe transient absorption spectroscopy. The results clearly indicate that bulk viscosity can change drastically without affecting F_cP in systems containing small amounts of added polymers. The bulk viscosity is thus a poor parameter for predicting the cage effect in such systems. The second investigation looked at the effect of the photochemical excitation energy on F_cP for the [Cp′(CO)₃Mo, MoCp′(CO)₃] cage pair in hexane. The results showed that F_cP increased when the wavelength of irradiation was changed from 546 nm to 436 nm, and then remained constant as the wavelength of irradiation was changed from 436 nm to 404 nm to 366 nm. These results are somewhat surprising because the recombination efficiencies for diatomic and triatomic molecules have been shown to decrease monotonically with increasing excitation energy. Two explanations are offered for the reverse wavelength dependence observed in this study. The first explanation invokes the different dynamic behaviors of the two excited states involved at the selected wavelengths, and the second invokes the different speeds of radical separation following irradiation at the selected wavelengths.     

Modeling of recombination kinetics of radical pairs in magnetic field. Comparison with experimental dependences on the frequency of encounters in biradicals

The recombination kinetics of spin-correlated radical pairs (RPs) with three nonequivalent magnetic nuclei were calculated under conditions of enforced encounters between radicals at time-independent frequency n_dif. The simplest two-position model of a RP was used, which includes two states (contact state and distance-separated state) of the RP, differing in magnitude of isotropic spin-spin exchange interaction between radicals. The calculated kinetic curves were treated in terms of a three-exponential model. The dependences of corresponding rate constants (k _rec) on n_dif, external magnetic field strength (B ₀), and intensity, A _eff, of isotropic hyperfine coupling (HFC) were obtained. The k _rec-vs.-n_dif or k _rec-vs.-viscosity (n_dif varies simultaneously with the inverse lifetime of the contact state) plots pass through maxima whose positions are shifted from the n_dif region near the A _eff value at B ₀ = 0.5 G toward high n_dif values with an increase in B ₀. At n_dif ≫ A _eff, the k _rec-vs.-B ₀ plots pass through maxima in the region B ₀ = A _eff. The calculated dependences are compared with experimental data on recombination of biradicals. The results of calculations show that the experimentally observed maxima on the k _rec-vs.-B ₀ or k _rec-vs.-n_dif plots can be due to peculiar features of the spin dynamics induced by the hyperfine coupling rather than the exchange interaction effects, as is commonly accepted. __________ Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 5, pp. 1103–1110, May, 2005.     

Kinetics of fast reactions of triplet states and radicals under photolysis of 4,4′-dimethylbenzophenone in the presence of 4-halophenols in micellar solutions of sodium dodecyl sulfate in magnetic field

Quenching kinetics of the 4,4′-dimethylbenzophenone triplet state with para-substituted phenol derivatives RC₆H₄OH (R = H, F, Cl, Br, I) was studied by nanosecond laser photolysis in aqueous micellar solutions of sodium dodecyl sulfate. The kinetic data were processed in the framework of a model with the Poisson distribution of phenols between micelles. The partition constants of RC₆H₄OH between the aqueous and micellar phases and the rate constants of their escape from a micelle and quenching of the 4,4′-dimethylbenzophenone triplet state with phenols in micelles were obtained. The quenching proceeds with high rate constants through hydrogen atom transfer to form the ketyl and phenoxyl radicals (no radicals are formed in the case of 4-iodophenol), which then recombine in a micelle or escape into the outer aqueous volume. The application of an external magnetic field retards radical pair recombination in a micelle and increases the fraction of radicals escaped into the aqueous phase. The quantum yield of radical pairs decreases 2.5-fold, and the rate of their recombination in micelles increases 2.5-fold on going from 4-chloro- to 4-bromophenol. This is caused by the acceleration of triplet radical pair recombination in the solvent cage. Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 6, pp. 1391–1396, June, 2005.     

Aging and degradation of polyolefins. I. Peroxide-initiated oxidations of atactic polypropylene

Efficiencies of polymer radical production by thermal decomposition of di-tert-butylperoxy oxalate (DBPO) have been measured in bulk atactic polypropylene (PP) at 25–55°C; they range from 1 to 26%, depending on [DBPO], temperature, and presence of oxygen. Most of the polymer radicals thus produced disproportionate in the absence of oxygen but form peroxy radicals in its presence. Most of the pairs of peroxy radicals interact by a first-order reaction in the polymer cage. The fraction that escapes gives hydroperoxide in a reaction that is half order in rate of initiation. In interactions of polymer peroxy radicals, in or out of the cage, about one-third give dialkyl peroxides and immediate chain termination, two-thirds give alkoxy radicals. About one-third of the later cleave at 45°C; the rest abstract hydrogen to give hydroxy groups and new polymer and polymer peroxy radicals. The primary peroxy radicals from cleavage account for the rest of the chain termination. Cleavage of alkoxy radicals and crosslinking of PP through dialkyl peroxides nearly compensate. Up to 70% of the oxygen absorbed has been found in hydroperoxides. The formation of these can be completely inhibited, but cage reactions are unaffected by inhibitors. Concentrations of free polymer peroxy radicals have been measured by electron spin resonance and found to be very high, about 10^?3M at 58–63°C. Comparison with results on 2,4-dimethylpentane indicate that rate constants for both chain propagation and termination in the polymer are much smaller than those for the model hydrocarbon but that the ratio, k_p/(2k_t)^½, is about the same.     

Recombination of Lophyl Radicals in Pyrrolidinium‐Based Ionic Liquids

The recombination of photolytically generated lophyl radicals has been investigated by UV/Vis spectroscopy in 1‐alkyl‐1‐methylpyrrolidinium bis(trifluoromethylsulfonyl)imides (NTf₂) in comparison with 1‐butyl‐3‐methylimidazolium NTf₂, dimethyl sulfoxide, and triacetin. The 1‐alkyl‐1‐methylpyrrolidinium‐based ionic liquids contain an alkyl substituent varying between butyl and decyl groups. Optically pure ionic liquids are used in these studies. Temperature‐dependent investigation of lophyl radical recombination shows an increase in the radical recombination rate with increasing temperature in each solvent, which is caused by decreasing viscosity with increasing temperature. Furthermore, the viscosity of the 1‐alkyl‐1‐methylpyrrolidinium NTf₂ increases nearly linearly within the row of these ionic liquids. In contrast, the recombination of the photolytically generated lophyl radicals is significantly faster in the ionic liquids than in the traditional organic solvents under investigation. Moreover, the recombination rate increases with the length of the alkyl chain bound at the cation of the ionic liquid at a given temperature. This may be caused by an increase in the extent of lophyl radical recombination within the solvent cage. Solvent cage effects dominate in the case of lophyl radical recombination in ionic liquids bearing a long alkyl chain or if the temperature is near the melting temperature of the ionic liquid. The positive value of the activation entropy supports this hypothesis. The results obtained are important for discussion of bimolecular radical reactions in ionic liquids.     

Bimolecular self-reactions of 2-arylindandione- 1,3-yl radicals studied by flash photolysis

Kinetic and thermodynamic data for reaction (1) of certain C-centered aromatic radicals (referred to in this paper by the numbers I to X) in chlorobenzene: have been obtained. The k₁ values of radicals varied between (1.1 ± 0.2) × 10⁶M^?1·sec^?1 (radical VIII) and (3.6 ± 0.7) × 10⁹M^?1 sec^?1 (radical VI) at 20°C. An investigation of the relationship between the recombination rates of radicals I–VIII and X and the solvent viscosity (mixture of toluene and dibutylphthalate, 0.6 < η < 18.4 cP) has shown that the recombination reactions involving radicals I–IV are limited by diffusion in solvents having a viscosity η> 10 cP and are activation reactions in solvents having a viscosity η < 10 cP. The recombination of radicals VIII and IX is an activation reaction, while that of radicals V–VII is diffusion-controlled in the entire viscosity range. The recombination of radical X is limited, in the viscosity range of 18.4 to 2 cP, by intrusion into the first coordination sphere of the partner, the effect of viscosity on the radical X recombination rate in the specified range being the same as its effect on diffusion-controlled reactions. The possible reasons of the discrepancies between the experimental fast recombination rate constants and the theoretical values calculated by the Debye–Smoluchowski theory are discussed. The equilibrium constant depends strongly on the nature of the substituent in the phenyl fragment: the substituents which increase unpaired electron delocalization in the radical intensify the dissociation of the respective dimer. Long-wave absorption bands have been recorded for radicals I–X and their extinction coefficients obtained. Dimers I–V are thermo- and photochromic compounds.     

Effect of radical scavenger on initiation efficiency

In order to analyze the effect of the scavenger on initiator efficiency f or ratio of combination between two radicals (1 ? f), an equation was derived by solving the modified Smoluchowski equation under the boundary condition based on the law of conservation of mass. At time t approaching infinity in a stationary state, this equation becomes: f = {D + a(Dk[S])^1/2}/{ak₀ + D + a(Dk[S])^1/2}, where D is the sum of the diffusion constants of two radicals, a is the diameter of the radical particle, k₀ is a specific rate of recombination in the cage between two radicals, k is the rate constant of reaction between the radical and the scavenger as a pair, and [S] is the scavenger concentration. When ak₀ ? D and k[S] is small, this equation approximates the previously derived Wijsman's equation. An equation of the same type as Noyes' equation is also derived from this equation, when [S] = 0. By using the equations derived in this paper, the effect of α,α′-diphenyl-β-picrylhydrazyl on the ratio of recombination of the (CH₃)-(CN)C^* radical and the effect of monomer on the initiation efficiency are satisfactorily explained.     

Kinetics of radical formation and decay in photooxidation of 4-halophenols sensitized by 4-carboxybenzophenone in aqueous solutions

Kinetics of formation and recombination of radicals formed by quenching of the triplet state of 4-carboxybenzophenone (CB) with para-substituted phenol derivatives RC₆H₄OH (R = OMe, H, Cl, Br, I) in aqueous solutions was studied by nanosecond laser photolysis. At pH ≥ 5.4, quenching proceeds with high rate constants ((1–3)⋅10⁹ L mol⁻¹ s⁻¹) through electron transfer to form the radical anion CB^⋅− and radical cation RC₆H₄OH^⋅+. The latter is transformed into the phenoxyl radical within ≤10 ns. At pH ≤ 8, the CB^⋅− radical anion is protonated in a phosphate buffer with the rate constant increasing from 4⋅10⁶ to 15⋅10⁶ s⁻¹ with a decrease in the pH from 8 to 5.4. The yield of radicals decreases from 100 to 13% as the atomic weight of halogen in the RC₆H₄OH molecule increases due to an increase in the probability of recombination of the primary triplet radical pair in the solvent cage and partial intersystem crossing in an encounter complex (³CB, RC₆H₄OH). The effect of heavy atom is also observed in the kinetics of volume recombination of the radicals, the magnitude of effect corresponds to the acceleration of the primary recombination of the triplet radical pair. Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 6, pp. 1397–1402, June, 2005.     

Effect of excess halide on the palladium‐catalyzed insertion‐triggered radical copolymerization of methyl acrylate and 1‐hexene

[Pd₂(μ‐Cl)₂(C₆F₅)₂(tht)₂] ( 1 ) is a very efficient initiator of the radical polymerization of methyl acrylate, but it is not active in the polymerization of methyl methacrylate or in the copolymerization with 1‐hexene. The addition of an excess of NBu₄Cl to solutions of [Pd₂(μ‐Cl)₂(C₆F₅)₂(tht)₂] ( 1 ) provides an initiator system that copolymerizes methyl acrylate and 1‐hexene by an insertion‐triggered radical mechanism. Random copolymers are obtained with 11% incorporation of 1‐hexene in moderate yields (about 35%). Studies of the decomposition products obtained after the first insertion of methyl acrylate in the Pd? C₆F₅ bond of 1 show that the addition of excess halide in the presence of monomer favors the homolytic cleavage of the Pd? C bond, and the generation of the radicals that are active species in the polymerization, versus alternative evolution pathways. © 2006 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 44: 5682–5691, 2006     

Sc2O@Td(19151)‐C76: Hindered Cluster Motion inside a Tetrahedral Carbon Cage Probed by Crystallographic and Computational Studies

A new cluster fullerene, Sc₂O@T_d(19151)‐C₇₆, has been isolated and characterized by mass spectrometry, UV/Vis/NIR absorption, ⁴⁵Sc NMR spectroscopy, cyclic voltammetry, and single‐crystal X‐ray diffraction. The crystallographic analysis unambiguously assigned the cage structure as T_d(19151)‐C₇₆, which is the first tetrahedral fullerene cage characterized by single‐crystal X‐ray diffraction. This study also demonstrated that the Sc₂O cluster has a much smaller Sc?O?Sc angle than that of Sc₂O@C_s(6)‐C₈₂ and the Sc₂O unit is fully ordered inside the T_d(19151)‐C₇₆ cage. Computational studies further revealed that the cluster motion of the Sc₂O is more restrained in the T_d(19151)‐C₇₆ cage than that in the C_s(6)‐C₈₂ cage. These results suggest that cage size affects not only the shapes but also the cluster motion inside fullerene cages.     

Effect of solvent on the termination rate constant in initial stages of free radical polymerization

The effect of solvent on the termination rate constant K_t, in the initial stages of free radical polymerizations has been estimated by considering its effect on the viscosity of the medium and on the overall dimensions of the macroradicals. The expression derived predicts that K_t is inversely proportional to the viscosity of the reaction medium, η₀, and that K_t, increases as the overall dimensions of the radicals decrease in poorer solvents. The effect of solvent on the η₀K_t, product depends on the average size and concentration of polymer in the system. For relatively high concentrations of high-molecular-weight polymers η₀K_t can be greater in a better solvent than in a poorer solvent. This trend would be reversed for low concentrations and/or low polymer molecular weights. Good agreement has been found between experimental and estimated values of η₀K_t.     

Electron impact fragmentation of some mixed cyclotetraphosphazenes

The electron impact fragmentations of some cyclotetraphosphazenes are reported and discussed. The major fragmentation path involves loss of two amine radicals and one chlorine radical in the series P₄N₄Cl_8-n(NMe₂)_n when n=2, and subsequent stages involve a ring contraction process with elimination of a P = N fragment, when n = 5 loss of amine radicals predominates on statistical grounds with little evidence of ring contraction. In the series P₄N₄F_8-n(NMe₂)_n fragmentation is dominated by loss of amino radicals when n = 4 and loss of fluorine radicals predominates on statistical grounds when n = 2. In the series P₄N₄F_8-nX_n (n = 2 or 4, X = Cl or Br), when n = 2 and X = Br the major fragmentation path is the loss of two bromine radicals, whereas when X = Cl the more favoured path is the loss of two chlorine radicals. In both, subsequent stages involve ring contraction reactions with elimination of a PN fragment. When n = 4 and X = Br or Cl on bond energy grounds the more favoured fragmentation pattern is the loss of bromine or chlorine radicals, respectively.     

Peresters XIII. Solvent effects in the decomposition of t-butylperoxy α-phenylisobutyrate with special reference to the cage effect

t-Butylperoxy α-phenylisobutyrate ( I ) decomposes thermally by concerted formation of carbon dioxide, t-butoxy, and cumyl radicals. Radical pair return in the solvent cage therefore does not affect the observed rate of decomposition, but is readily determined by means of galvinoxyl and other scavengers. In a series of 15 solvents the rate constant varies over a 2.8 fold range, being fastest in aromatic solvents. In the same solvent series the relative rates of diffusion and combination of radicals, measured by the cage effect, change by tenfold and are largely determined by the viscosity of the solvent. In all solvents of η > 8 mP, the reciprocal of the cage effect is a linear function of (T^1/2/η), as recently observed for trifluoromethyl and methyl radicals [16]. This property of the cage effect provides a test by which it can be distinguished from other processes that reduce the efficiency of free-radical production from an initiator.     

Transient free radicals in viscous solvents

The majority of free radicals are highly reactive species which participate in bimolecular reactions with each other. Validation of the theory of molecular diffusion and reactivity in the liquid state requires knowledge of rate constants of radical–radical reactions (recombination, disproportionation) and their viscosity dependencies. An accurate comparison of theory and experiment has become available due to experimentally measured diffusion coefficients of reactive radicals by transient grating technique. Initial distribution of radicals in solution can be not random but pair-wise as in photo- or thermoinitiation of free radical polymerization reactions. Probability of a radical escape of a partner (cage escape) characterizes the initiator efficiency. Despite decades of measurement of cage effect values, cage effect dynamics with free radicals have only been investigated quite recently. The present tutorial review considers the effect of viscosity of Newtonian liquid on two types of recombination—in the solvent bulk and in a cage. Further, since radicals are paramagnetic species, external magnetic field affects probability of their reactions in pairs. These effects are also observed in viscous liquids, and reasons for such observations are explained. The recently discovered low magnetic field effect is also observed on radical pairs in viscous liquids.     

Intercage Electron Transfer Driven by Electric Field in Robin–Day‐Type Molecules

A new class of isomers, namely, intercage electron‐transfer isomers, is reported for fluorinated double‐cage molecular anion e^?@C₂₀F₁₈(NH)₂C₂₀F₁₈ with C₂₀F₁₈ cages: 1 with the excess electron inside the left cage, 2 with the excess electron inside both cages, and 3 with the excess electron inside the right cage. Interestingly, the C₂₀F₁₈ cages may be considered as two redox sites existing in a rare nonmetal mixed‐valent (0 and ?1) molecular anion. The three isomers with two redox sites may be the founding members of a new class of mixed‐valent compounds, namely, nonmetal Robin–Day Class II with localized redox centers for 1 and 3 , and Class III with delocalized redox centers for 2 . Two intercage electron‐transfers pathways involving transfer of one or half an excess electron from one cage to the other are found: 1) Manipulating the external electric field (?0.001 a.u. for 1 → 3 and ?0.0005 a.u. for 1 → 2 ) and 2) Exciting the transition from ground to first excited state and subsequent radiationless transition from the excited state to another ground state for 1 and 3 . For the exhibited microscopic electron‐transfer process 1 → 3 , 2 may be the transition state, and the electron‐transfer barrier of 6.021 kcal mol^?1 is close to the electric field work of 8.04 kcal mol^?1.     

The Origin of Magnetic Field Dependent Recombination in Alkylcobalamin Radical Pairs

Magnetic field effect studies of alkylcobalamin photolysis provide evidence for the formation of a reactive radical pair that is born in the singlet spin state. The radical pair recombination process that is responsible for the magnetic field dependence of the continuous-wave (CW) quantum yield is limited to the diffusive radical pair. Although the geminate radical pair of adenosylcob(III)alamin also undergoes magnetic field dependent recombination (A. M. Chagovetz and C. B. Grissom, J. Am. Chem. Soc. 115, 12152–12157, 1993), this process does not account for the magnetic field dependence of the CW quantum yield that is only observed in viscous solvents. Glycerol and ethylene glycol increase the microviscosity of the solution and thereby increase the lifetime of the spin-correlated diffusive radical pair. This enables magnetic field dependent recombination among spin-correlated diffusive radical pairs in the solvent cage. Magnetic field dependent recombination is not observed in the presence of nonviscosigenic alcohols such as isopropanol, thereby indicating the importance of the increased microviscosity of the medium. Paramagnetic radical scavengers that trap alkyl radicals that escape the solvent cage do not diminish the magnetic field effect on the CW quantum yield, thereby ruling out radical pair recombination among randomly diffusing radical pairs, as well as excluding the involvement of solvent-derived radicals. Magnetic field dependent recombination among alkylcobalamin radical pairs has been simulated by a semiclassical model of radical pair dynamics and recombination. These calculations support the existence of a singlet radical pair precursor.     

Aging and degradation of polyolefins. II. γ-initiated oxidations of atactic polypropylene

Oxidations of bulk atactic polypropylene (PP) have been carried out at 22 and 45°C, and the dependence of rate of formation of each product on rate of initiation has been determined. The principal product is PP hydroperoxide, formed in a half-order reaction. One termination product is polymeric dialkyl peroxide, formed in a first-order reaction. Other termination and propagation products, alcohols and carbonyl compounds, are formed in reactions that are mostly first-order in initiation. At 22°C, G is 9–63. G is about three times as great at 45°C as at 22°C. Experiments with 2,6-di-tert-butyl-p-cresol shows that it can inhibit all non-cage propagation and all formation of PP hydroperoxide, but that it does not affect cage reactions of initiating radicals and their successors. Only about 16% of the initiating PPO₂· radicals escape the cage at 45°C. Oxidations of PP, n-hexane, and their mixture with both peroxide and γ-ray initiation show that nearly all the initiating radicals escape the cage in solution but that the concentration of PPO₂· radicals is much less than in bulk because of a much faster chain termination. Both the propagation and termination constants for PP oxidation are much faster in solution, but the changes compensate so that k_p/(2k_t)^½ is about the same in solution as in bulk.     

Kinetics of polymerization rate based on the dependence of termination rate on chain length

When the structure of a primary radical resembles that of the chain end of the polymer radical, the rate of the primary radical termination is approximately the same as the termination rate between the oligomer radical and the polymer radical. The rate constant of termination between polymer radicals of chain length n and s, which involve the primary radicals, is k_t,ns = const.(ns)^?a. In the polymerization of methacrylonitrile initiated by 2,2′-azobisisobutyronitrile in dimethylformamide at 60.0°C, the value of a is found to be 0.091. From data obtained previously in the bulk polymerization of styrene initiated by 1-azobis-2-phenylethane at 60.0°C, the value of a is found to be 0.167. Because such a values are so large that they are not estimated by the excluded volume, the termination rates are discussed by adding the dependence of the diffusion of the segments to that for chain length.     

Photo CIDNP d'acridines dans le diphenylmethane

It is shown by CIDNP that the photoreaction of acridine with diphenylmethane involves the radical pair acridinyle-diphenylmethyl^T formed from T_nπ^* acridine. Unsuccessful attempts to trap radicals by CCl₄ prove that the recombination of the radicals produced during the photoreaction proceeds essentially in the solvent cage in which they are formed. An explanation of the variation of the acridine reactive state multiplicity with the reacting solvent is proposed.