Extension of Structure–Reactivity Correlations for the Hydrogen Abstraction Reaction by Bromine Atom and Comparison to Chlorine Atom and Hydroxyl Radical

Abstract

Recently we presented structure-reactivity correlations for the gas-phase ambient-temperature rate constants for hydrogen abstraction from sp(3)-hybridized carbon by chlorine atom and hydroxyl radical (Cl•/HO• + HCR3 → HCl/HOH + •CR3); the reaction enthalpy effect was represented by the independent variable ΔrH and the "polar effect" by the independent variables F and R, the Hammett constants for field/inductive and resonance effects. Both these reactions are predominantly exothermic and have early transition states. Here, we present a parallel treatment for Br• whose reaction is significantly endothermic with a correspondingly late transition state. Despite lower expectations because the available database is less extensive and much more scattered and because long temperature extrapolations are often required, the resulting least-squares fit (log k298,Br = -0.147 ΔrH - 4.32 ΣF - 4.28 ΣR - 12.38 with r(2) = 0.92) was modestly successful and useful for initial predictions. The coefficient of ΔrH was ∼4-fold greater, indicative of the change from an early to a late transition state; meanwhile the sizable coefficients of ΣF and ΣR indicate the persistence of the "polar effect". Although the mean unsigned deviation of 0.79 log k298 units is rather large, it must be considered in the context of a total span of over 15 log units in the data set. The major outliers are briefly discussed.

Extension of Structure–Reactivity Correlations for the Hydrogen Abstraction Reaction to the Methyl Radical and Comparison to the Chlorine Atom, Bromine Atom, and Hydroxyl Radical

“…Summary results from this multivariable approach are shown in eqs  for Δ r H and Σσ para as the independent variables, where k 298 is in units of cm 3 molecule –1 s –1 per H, Δ r H is in kJ/mol, and MUD is the mean unsigned deviation from the best least-squares fit correlation line. (The coefficients shown here for Br • are slightly different from those reported in ref because of the present use in eq of averages for k 298 rather than individual values for substrates for which multiple rate constants have been reported.) To codify the expected trend that k will increase with decreasing Δ r H (increasing exothermicity), the “ a ” coefficient in the Evans–Polanyi term must be negative, because (Δ r H ) is of course negative for an exothermic reaction and positive for an endothermic reaction.…”

“…We have recently explored − the use of the venerable empirical Hammett constants as the independent variable to represent the polar effect. This was combined with the Evans–Polanyi formulation to represent the enthalpic effect, except that the dependent variable was log k rather than the more usual E .…”

“…The choice of Hammett constants to indicate the electron donor/acceptor properties of substituent groups was based on their much wider availability rather than, for example, ionization energies and electron affinities of the substrates. The reaction range was limited to gas-phase abstractions from sp 3 -hybridized carbons and the attacking radicals were Cl • , HO • , and Br • . Summary results from this multivariable approach are shown in eqs  for Δ r H and Σσ para as the independent variables, where k 298 is in units of cm 3 molecule –1 s –1 per H, Δ r H is in kJ/mol, and MUD is the mean unsigned deviation from the best least-squares fit correlation line.…”

Antioxidant and Cytoprotective Effects of Kukoamines A and B: Comparison and Positional Isomeric Effect

“…Such positional isomeric effect is assumed to be from the interaction between functional groups, such as a macrocycle by a hydrogen bridge between the hydrogen of the chelate and middle N-atom, or a field inductive effect between the dihydrocaffeoyl moiety and middle N-atom. Field inductive effects, however, have been reported to affect hydrogen abstraction (i.e., hydrogen-atom-transfer) or proton dissociation (i.e., proton-transfer) of phytophenols [ 35 , 36 , 37 , 38 ]. Of course, the detailed interactions may be complicated and require further investigation in the future.…”

Comment on “Radicalicity: A scale to compare reactivities of radicals” (Chem. Phys. Lett. 618 (2015) 99–101)

“…While it is indeed true that the activation energy of a radical reaction typically decreases as its exothermicity increases, as expressed historically in the Evans-Polanyi relationship [4], it has also been recognized for over half a century that the relationship is usually not strictly linear because the rate constants are also often dependent on polar effects and steric effects [5][6][7] as well as thermochemical effects. Our recent studies [8][9][10] Therefore we suggest that the term "radicalicity", presented [2] as a measure of radical reactivity, not be adopted because it is in reality simply another expression for thermochemistry for which there is already adequate terminology (BDE).…”