Paul M. Zimmerman

Multi-exciton generation-the creation of multiple charge carrier pairs from a single photon-has been reported for several materials and may dramatically increase solar cell efficiency. Singlet fission, its molecular analogue, may govern multi-exciton generation in a variety of materials, but a fundamental mechanism for singlet fission has yet to be described. Here, we use sophisticated ab initio calculations to show that singlet fission in pentacene proceeds through rapid internal conversion of the photoexcited state into a dark state of multi-exciton character that efficiently splits into two triplets. We show that singlet fission to produce a pair of triplet excitons must involve an intermediate state that (i) has a multi-exciton character, (ii) is energetically accessible from the optically allowed excited state, and (iii) efficiently dissociates into multiple electron-hole pairs. The rational design of photovoltaic materials that make use of singlet fission will require similar ab initio analysis of multi-exciton states such as the dark state studied here.

The ground and lowest three adiabatic excited states of methylene are computed using the variational Monte Carlo and diffusion Monte Carlo (DMC) methods using progressively larger Jastrow–Slater multideterminant complete active space (CAS) wave functions. The highest of these states has the same symmetry, A11, as the first excited state. The DMC excitation energies obtained using any of the CAS wave functions are in excellent agreement with experiment, but single-determinant wave functions do not yield accurate DMC energies of the states of A11 symmetry, indicating that it is important to include in the wave function Slater determinants that describe static (strong) correlation. Excitation energies obtained using recently proposed pseudopotentials [Burkatzki et al., J. Chem. Phys. 126, 234105 (2007)] differ from the all-electron excitation energies by at most 0.04 eV.

The reactivity of NH(2)BH(2) in the presence of ammonia-borane (AB) is investigated using ab initio CCSD(T) simulations to answer the following three questions: How do AB and NH(2)BH(2) react? How do aminoborane species oligomerize apart from catalytic centers? Can the formation of experimentally observed products, especially cyclic N(2)B(2)H(7)-NH(2)BH(3), be explained through the kinetics of NH(2)BH(2) oligomerization in the presence of AB? AB is shown to react with NH(2)BH(2) by the addition of NH(3)-BH(3) across the N=B double bond, generating linear NH(3)BH(2)NH(2)BH(3). This species decomposes by surmounting a reasonable barrier to produce two NH(2)BH(2) and H(2). The generation of additional NH(2)BH(2) from NH(2)BH(2) and AB provides a pathway for autocatalytic NH(2)BH(2) production. The important intermediates along the oligomerization pathway include cyclic (NH(2)BH(2))(2) and linear NH(3)BH(2)NH(2)BH(3), both of which have been observed experimentally. Calculations show cyclic N(2)B(2)H(7)-NH(2)BH(3), an aminoborane analogue of ethylcyclobutane, to be the kinetically preferred stable intermediate resulting from oligomerization of free NH(2)BH(2) over its isomers, cyclic B(2)N(2)H(7)-BH(2)NH(3) and cyclotriborazane, cyclic (NH(2)BH(2))(3). Simulations show cyclotriborazane formation to be kinetically slower than cyclic B(2)N(2)H(7)-NH(2)BH(3) formation and imply that formation of the cyclic species cyclotriborazane and cyclopentaborazane may be catalyzed by binding of NH(2)BH(2) to a catalytic metal center. Routes that may lead to larger noncyclic oligomers are suggested to be kinetically competitive. The highly reactive N=B double bonds of NH(2)BH(2) are shown to be of central importance in understanding aminoborane oligomerization.