Michail Konstantinos Bogdos


Last Name

Bogdos

First Name

Michail Konstantinos

ORCID

0000-0003-1826-2706

Organisational unit

Filters

20241
Reset filters

Search Results

Publications 1 - 1 of 1
  • Mechanistically Guided Predictive Modeling of Stoichiometric Elimination Reactions of Palladium (II) Complexes
    Item type: Doctoral Thesis
    Bogdos, Michail Konstantinos (2024)
    The understanding of chemical mechanisms allows for the accurate prediction of the concentration of chemical species in a reaction over time and therefore the outcome of a reaction. Given the success of this approach, in the past decade focus has shifted to trying to predict the outcome of reactions given only the reaction conditions, with some degree of success. It is reasonable to assume that an approach to this problem that places emphasis on chemical mechanisms is a worthwhile endeavor. To achieve this, elementary steps need to be studied in isolation in order to be able to predict their relative or absolute rates given only the structures of reactants and the reaction conditions as inputs. This thesis describes two case studies of elementary steps which are examined through their chemical mechanism with the goal of using the chemical mechanistic information to build models which can predict their outcomes. The first part of this work investigates the competition of two 𝛽-elimination reactions, 𝛽-hydride (𝛽-H) and 𝛽-heteroatom (𝛽-X) elimination, that alkylpalladium complexes can undergo. These elementary steps were chosen as alkylpalladium complexes are intermediates in industrially relevant catalytic reactions, such as the Mizoroki-Heck reaction, alkyl C H activation and ethylene polymerization and these 𝛽-eliminations are the most common decomposition pathways for these intermediates. Despite this and the fact that these two eliminations are often in competition, the mechanism of the 𝛽-X elimination is poorly understood and the factors that control preference for either elimination when they compete are also not known. An investigation into the mechanism and these factors is reported in this thesis. Specifically, the influence of the identity of the X group and the ancillary ligands on the 𝛽-X/𝛽-H competition was probed, as well as the preference of a syn- or anti-𝛽-X pathway for 𝛽-X elimination. The information was captured in predictive models utilizing logistic regression. The second part of the thesis investigates the effect of sterics, electronics and coordination number on reductive elimination, using Pd(II) complexes which perform C N reductive elimination as model complexes. Reductive elimination is the bond-forming elementary organometallic step in many catalytic reactions. According to the canonical understanding found in the literature, the main factors affect the rate of reductive elimination are electronics at the metal center, ancillary ligand sterics and metal complex coordination number. The relative contribution of these factors to the rate of reductive elimination has not been quantified. In this chapter, the development of a new class of palladacycles which are suitable for studying these factors is described. Kinetics, electrochemical and computational characterization of these palladacycles yield the requisite data for the investigation. The use of statistical simulations and causal inference allows the reexamination of the canonical understanding, revealing that it cannot adequately account for the collected data. An updated model which incorporates the canonical understanding, contradictions in the literature and the newly collected data is described. In this new manifold, ancillary ligand sterics affect the rate in three distinct ways - directly, by atlering metal electronics and by causing a change in coordination number, which also affects metal electronics. Overall, it was found that the total effect of sterics is smaller than that of electronics and the coordination number does not directly influence the rate. Taken together, these two case studies serve to illustrate the way in which mathematical modeling of elementary steps can be facilitated by mechanistic considerations. They represent the first incremental steps towards developing a set of models which can work together to predict the outcomes of catalytic cycles.
Publications 1 - 1 of 1