Graduate Thesis Or Dissertation
 

Insights and Patterns from Quantitative and Statistical Analyses of Chemical Systems

Public Deposited

Downloadable Content

Download PDF
https://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/6t053p55s

Descriptions

Attribute NameValues
Creator
Abstract
  • The work presented here is the culmination of efforts in describing complex chemical systems with statistical methods, thereby providing relatively simple quantitative models and easily comprehensible qualitative insights. The following investigations have various goals, such as development of new methods, fundamental understanding, molecular control of chemical species, and advancement of synthetic methods. Unifying these projects is the purpose of contributing essential understanding to the chemical community for the advancement of their respective fields. A General Group Additivity Method is developed which provides a platform methodology for the quantitative description of any chemical system of interest. There are three improvements defined in contrast to previous work: a) use of Ordinary Least Squares to fit a model instead of finding an analytical solution to a saturated model b) calculation of standard error of each group in the model to provide a description of variance in each group c) a process to guide the aggregation or disaggregation of groups directed by evaluation of error compared to energetic contribution of the group as well as the geometry of the group as it appears in multiple clusters in the training set. The aluminum oxo-hydroxy cluster system serves as a case study for the method, and the construction of the model provides insights into the formation of Al clusters by putting observed quantities in context with observations from experiment. The models produced are quite accurate when compared with quantum mechanics computations, but require a fraction of the time investment when predicting the Gibbs energy of cluster condensation. Finally, the impact of the model is improved through a code and workflow used to acquire thermodynamic predictions from the model using common softwares. The alkyltin cluster system is addressed with two separate projects. Known clusters are computed with density functional theory (DFT) and the condensation energies are provided in the context of monoalkyltin trialkoxide hydrolysis. It is found that the dodecamer species, with general form [RSn12O14(OH)6][X]2, is thermodynamically most favorable to form and requires the lowest ratio of H2O:Sn, implying that it may be difficult to prevent monoalkyltin trialkoxide solutions from condensing toward this species. The general group additivity method is applied to this cluster system, and the results are contrasted with the Al oxo-hydroxy cluster system. Notably, ƞH2O ligands have no statistical significance in the monoalkyltin oxo-hydroxy cluster system, while their impact on condensation of Al clusters is significant. Three unique µ3O groups are identified, defined by the number of bridging units by which they are surrounded. These are described as double bridges, and it is shown that the µ3O group with 2 double bridges (µ3O-2DB) is energetically more favorable than the group surrounded by 3 double bridges (µ3O-3DB). The group with only 1 double bridge (µ3O-1DB) was found to have a similar energetic contribution as µ3O-2DB. The model is produced with only one-third of the cluster species investigated, and is validated against the remaining cluster species. The group additivity model is shown to have excellent agreement with quantum mechanics (QM) computed energies, and may be implemented as a predictive tool for investigation of theorized cluster species. Expanding on the alkyltin cluster system, the chloride-containing dimer (C4H9Sn)2(OH)2Cl4(H2O)2 is shown to produce excellent quality thin-films for use in nanolithographic patterning when spin-coated from solution. This result is contrasted with the performance of a solution containing a larger dodecamer species. As the cluster is synthesized from the slow hydrolysis of n-butyl tin trichloride, the hydrolysis and coordination of aqua ligands to the monoalkyltin trichloride monomer is computationally investigated. We find that all hydrolyses and coordinations are energetically unfavorable. DFT computations show that the formation of the dimer is thermodynamically favorable, and that hydrolysis of chloride ligands is significantly unfavorable, indicating it is a stable species in air. Characterization of the film species suggests that the dimer may condense into a dodecamer during the spin-coat process, or possibly during dissolution of the thin film. This change is also investigated with DFT computations, which show that condensation from the dimer into the dodecamer is unfavorable unless Cl ligands are replaced by ƞOH ligands. It is theorized that these Cl ligands are lost during the rapid desolvation involved in the spin-coat process. As the solvent is quickly lost, there is a high concentration gradient between the solution and the air, and as HCl quickly evolves from the solution, the drop in HCl concentration drives hydrolysis of Cl ligands from dimer species. A uniquely selective acylation of a biaryl diol is investigated in which the sequential desymmetrization and kinetic resolution steps are disaggregated through implementation of a kinetic modeling software. The origins of selectivity are elucidated through DFT computations of the transition states, identifying the distortion from planarity of the 1,5-O•••S interaction within the key acyl ammonium intermediate as a destabilizing feature that determines atropselective acylation and thus product enantioselectivity. As a result of the computational discoveries, the observed high enantioselectivity that results from two transition states occurring concurrently in the novel synthetic approach is quantitatively described.
License
Resource Type
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Advisor
Committee Member
Academic Affiliation
Rights Statement
Publisher
Peer Reviewed
Language
Embargo reason
  • Pending Publication
Embargo date range
  • 2020-08-26 to 2022-09-26

Relationships

Parents:

This work has no parents.

In Collection:

Items