2011ApJ...729...15C

The excitation temperature T₀₁ derived from the relative intensities of the J = 0 (para) and J = 1 (ortho) rotational levels of H₂ has been assumed to be an accurate measure of the kinetic temperature in interstellar environments. In diffuse molecular clouds, the average value of T₀₁is ∼70 K. However, the excitation temperature T(H⁺₃) derived from the (J, K) = (1, 1) (para) and (1, 0) (ortho) rotational levels of H⁺₃ has been observed to be ∼30 K in the same types of environments. In this work, we present observations of H⁺₃ in three additional diffuse cloud sight lines for which H₂ measurements are available, showing that in four of five cases T₀₁ and T(H⁺₃) are discrepant. We then examine the thermalization mechanisms for the ortho:para ratios of H⁺₃ and H₂, concluding that indeed T₀₁ is an accurate measure of the cloud kinetic temperature, while the ortho:para ratio of H⁺₃ need not be thermal. By constructing a steady-state chemical model taking into account the nuclear spin dependence of reactions involving H⁺₃, we show that the ortho:para ratio of H⁺₃ in diffuse molecular clouds is likely governed by a competition between dissociative recombination with electrons and thermalization via reactive collisions with H₂.