Abstract:
In this thesis, we utilize DFT calculations to investigate the mechanism of
a fundamental chemical step that is widely implicated in homogenous catalysis,
namely insertion of unsaturated molecules into a metal-hydride bond of octahedral
saturated d6-transition metal complexes. Our first investigation pertains to
carbon dioxide insertion. For many complexes, the reaction has been experimentally
determined to follow an associative mechanism. There have been, however,
two views of the rate determining step (RDS) in such mechanism. A more prevailing
view assumes a linear transition state (TS) leading to a C-H bound formate
intermediate, the other assumes a cyclic TS. Herein, we conduct a detailed theoretical
investigation of the associative PES of CO2 insertion into the Re-H bond
of fac-(bpy)Re(CO)3H for which extensive experimental data are available. We
calculate both these transition states and study how their barrier and KIE change
as a function of the solvent and substitution at the 4,4’ bpy positions. Surprisingly,
results reveal that the initial stage of the reaction starts with a low energy
linear TS (ts1) for formation of a bond between the carbon and the metal hydride
bond leading to a bridged hydride species. A second stage of the reaction
involves cleavage of the bridged metal hydride bond and rearrangement of the
resulting ion-pair intermediate via a cyclic transition state (TScyc). To explore
the generality of this finding we also calculated CO2 insertion in the octahedral
(iP rPHNP)Ir(H)3. The amino functionality in on the ligand in this complex was
proposed to assist insertion by hydrogen bonding. Again, our calculations show
the PES to start with ts1 for bridge formation, followed by a RDS TScyc. We
show that consideration of TScyc as RDS affords activation free energies, solvent
effects, substituent effects and Kinetic Isotope Effects (KIE) that are all in excellent
agreement with experimental data available for the two systems. In light of
this finding, we initiated a second investigation aimed at understanding the mechanism of associative insertion of carbonyl compounds other than CO2. For this
purpose, we considered a series of eleven carbonyl compounds, including aldehydes,
ketones, esters, carboxamides, carbamates and urea derivatives. We worked
with Gusev’s osmium (PNN)Os(H)2(CO) catalyst. The insertion thermodynamics
spanned a range of around 30 kcal/mol. Surprisingly, the substrates that
disfavored insertion afforded equilibrium Isotope Effect (EIE) that were slightly
more inverse than the more favored insertions. This implies that the less favored
reactions result in formation of stronger C-H bonds. We rationalized this counterintuitive
finding using thermodynamic cycles starting with distortion of the
carbonyl moiety in the free substrates prior to formation of a C-H bond. Simply
put, the different substrates have very different distortion energies, and the
distorted substrates have comparable susceptibilities to reduction by formation
of a C-hydride bond. Kinetically, the calculations predict associative insertion of
all of the carbonyl compounds considered to follow initial bridge formation by a
linear ts1 followed by rate limiting TScyc. Associative insertion provides a major
shift from the more conventional insertion mechanism that requires initial coordination
of the substrate to a metal. The results obtained in the present work
provide new insights to understanding the associative mechanism for carbonyl
group insertion into the M-H bond of octahedral complexes.