Thesis (Ph. D.)--University of Chicago, 1943. / Lithoprinted. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references.

Thesis (M.S.)--University of Wisconsin--Madison, 1964. / eContent provider-neutral record in process. Description based on print version record. Bibliography: l. 67-71.

Thesis (Ph. D.)--Georgia State University, 2006. / Title from title screen. William H. Nelson, committee chair; Thomas L. Netzel , A.G. Unil Perera, Brian D. Thoms, Gary Hastings, committee members. Electronic text (243 p. : ill. (some col.)) : digital, PDF file. Description based on contents viewed Aug. 16, 2007. Includes bibliographical references.

Radicals, like other functional groups, may be classed as nucleophiles or electrophiles depending upon their character. Much has been done in the past to determine the· nucleophilicity or electrophilicity of a wide ra~ge of radicals, but it .. does not appear that the philicity ofthe aryl radical has yet been determined. A competitive cyclisation of an aryl radical to one of two alkyl chains containing olefins ofdiffering electron density was undertaken to determine aryl radical philicity. Prior to cyclisation, synthesis of the precursors was required, starting from simple commercially available aromatic compounds. A wide range of allyloxy aromatic compounds with different substituents on the two olefins was synthesised. In the course of the required multi-step synthesis many previously unreported compounds were isolated and fully characterised, including crystal X-ray diffraction studies where crystals were obtained. To investigate the effects of different electron density in the aryl ring a further range of competitive cyclisation precursors were synthesised with an ester group para to the site where the aryl radical is formed. Studies into the synthesis of a competitive cyclisation precursor with olefin bearing side chains with no heteroatom were also undertaken. The procedure for the cyclisation reactions and subsequent work-up was optimised under both conventional and microwave methodology. Analysis of the products was performed using both quantitative and qualitative analytical techniques. Quantitative analysis was performed using gas chromatography, IH NMR spectroscopy with an internal standard and isolation of products. Qualitative analysis was performed using both ID and 2D NMR spectroscopy. Analysis of the results of the competitive cyclisation reactions show beyond reasonable doubt that the aryl radical is nucleophilic in character.

Supervisor: Dr. R.E. Pincock In order to attempt to define more rigourously the geometric requirements of bridgehead free radicals, two t-butyl peresters of bridgehead acids are prepared — t-butyl adamantane-1-percarboxylate and t-butyl norbornane-1-percarboxylate. The decompositions of these peresters are examined, the rates determined, and the activation parameters for the decompositions calculated. Evidence is presented which suggests that the decomposition of t-butyl adamantane-1-percarboxylate proceeds via a concerted mechanism, splitting out CO₂ and producing the bridgehead free radical. An enthalpy of activation of 30.5 kcal./mole and an entropy of activation of 13 cal./deg. are determined. Similar evidence is presented which indicates that t-butyl norbornane-1-percarboxylate also decomposes via the concerted mechanism. The enthalpy of activation for this decomposition is found to be 32.2 kcal./mole and the entropy of activation found to be -3.6 cal./deg. The results obtained here are compared to those of other workers, and the inference that the entropy of activation plays a major role in determining rates of such decompositions is made. A synthetic pathway to another bridgehead perester, t-butyl bicyclo[2,2,2]octane-l-percarboxylate is explored, and a new and more practicable synthesis of bicyclo[2,2,2]octane-l-carboxylic acid presented. A comparison of bridgehead free radicals to bridgehead carbonium ions is made, and other evidence pertinent to the stereochemistry of bridgehead free radicals is discussed. / Science, Faculty of / Chemistry, Department of / Graduate

In order to obtain chemical evidence concerning the configuration of free alkyl radicals, 1-(carbo-t-butylperoxy)-cis- and trans-1,4-dimethylcyclohexane and exo- and endo-2-(carbo-t-butylperoxy)-2-methylnorbornane have been synthesized, and their decomposition rates and products determined. The cis- and trans-cyclohexyl peresters undergo thermal decomposition at 60° in cumene at essentially the same rates (cis isomer ∆H⁺ = 27.5 kcal/ mole, ∆S⁺ = 5.07 eu; trans isomer ∆H⁺ = 27.8 kcal/mole, ∆S⁺ = 5.75 eu) and with direct formation of 1,4-dimethylcyclohexyl radicals, carbon dioxide, and t-butoxy radicals. The products, cis- and trans-1,4-dimethylcyclohexane, 1,4-dimethylcyclohexene, and 4-methylmethylenecyclohexane, formed from the 1,4-dimethylcyclohexyl radicals in degassed cumene, are the same from both precursors. In the presence of oxygen in 1,2-dimethoxyethane (glyme) or 25 ml Freon TF:1 ml glyme, both the cis and trans peresters gave identical yields (after reduction) of trans-1,4-dimethylcyclohexanol (58%) and cis-1,4-dimethylcyclohexanol (42%) at oxygen pressures from 1 to approximately 600 atm. A common intermediate is thus indicated. In cumene at 75°, exo-2-(carbo-t-butylperoxy)-2-methylnorbornane decomposes 6.6 - 6.7 times faster than its endo isomer. The activation parameters are ∆H⁺ = 27.9 kcal/mole, ∆S⁺ = 6.31 eu, and ∆H⁺ = 30.3 kcal/mole, ∆S⁺ = 9.33 eu, respectively. The products, exo- and endo-2-methylnorbornane, 2-methyl-2-norbornene, and 2-methylenenorbornane, resulting from 2-methyl-2-norbornyl radicals in degassed cumene, are essentially the same from either isomeric source. The same ratio of exo- to endo-2-methyl-2-norbornyl hydroperoxide (15%:85%) is formed from either perester in the presence of oxygen at pressures up to -600 atm. Unlike the apparently unique case of distinctly different cis- and trans-9-decalyl radicals, the alkyl radicals from the decomposition of isomeric norbornyl and cyclohexyl peresters have a single configuration or they invert much faster than they react with oxygen. The results suggest that if radicals with a stereochemical "memory" of their origin exist in these systems, their lifetimes are of the order of 10⁻¹⁰ seconds or less. For the 1,4-dimethylcyclohexyl radical, hydrogen atom transfer and reaction with oxygen were found to occur preferentially (82% and 58%, respectively) from an axial direction. On the other hand, cumene and oxygen react with the 2-methyl-2-norbornyl radical predominantly from an exo direction (90% and 85%, respectively). The carbonylation of tetrahydrolinalool, and several synthetic routes to 1-alkyl-4-t-butylcyclohexanecarboxylic acids also have been explored. / Science, Faculty of / Chemistry, Department of / Graduate

The technique of electron paramagnetic resonance (EPR) has been used to investigate the electronic structure and geometry of several small triatomic or tetraatomic radicals which have been trapped at 4.2 K in the inert matrices of neon, argon or krypton. The EPR spectra were analyzed by using an extensive computer simulation program based on a detailed general spin Hamiltonian which includes the Zeeman, nuclear Zeeman, hyperfine and quadrupole interaction terms. A general method for analyzing the EPR spectrum of a poly-crystalline sample has been discussed with particular emphasis being placed on systems for which the principal axes of the electronic g-tensor and hyperfine tensors are not coincident. The chloroperoxyl radical (C100) was previously studied in several polar environments and in an argon matrix, but the EPR parameters were somewhat in disagreement. A study of this radical, formed by the UV irradiation of chlorine dioxide (ClO₂) trapped in an inert matrix was undertaken with a view to improving the accuracy of the EPR parameters. The electronic g-tensor and hyperfine tensor were found to be non-coincident. The radical species FSO has been produced by the far UV photolysis of thionyl fluoride (F₂SO) in an argon matrix. The spin Hamiltonian parameters were obtained for this radical and found to be remarkably similar to the well known species F00. The hyperfine components were interpreted in a manner similar to F00. The chlorodisulfonyl radical (C1SS) was produced by the near UV photolysis of the dichlorodisufane molecule (S₂Cl₂) in an argon matrix. The electronic g-tensor, hyperfine coupling and quadrupole coupling constants were determined from comparison of the observed spectrum with the computer simulated spectrum. To obtain good agreement, the principal axes of the hyperfine and quadrupole tensors had to be rotated by ~10° from those of the electronic g-tensor. The radical species HSO₂ and FSO₂ were produced by the UV photolysis of HI/S0₂ or trifluoromethylhypofluorite (CF₃OF)/S0₂ mixtures in the rare gas matrices. The F₂SO₂ radical was also produced by the far UV photolysis of sulfuryl fluoride (F₂SO₂). The EPR parameters were determined assuming non-coincident axes for the hyperfine and electronic g-tensors. Another species was produced in the reaction of methyl radicals with SC>2 but the nature of the adduct is uncertain. Chlorine dioxide (ClO₂) was studied in several inert matrices and the spin Hamiltonian parameters were determined for each matrix. Partial orientation has been observed in all the matrices. The preferred orientation is with the plane of the radical parallel to the deposition surface. The spectrum has been simulated using an approximate distribution function for the molecular orientations. A second trapping site has been observed on annealing the argon matrix. This site exhibits slightly different hyperfine and Zeeman interactions and is thought to mean that the site is a substitutional one. The matrix shifts of the electronic g-tensor and hyperfine tensor are discussed in terras of the van der Waals and Pauli interaction forces. / Science, Faculty of / Chemistry, Department of / Graduate

This thesis reports studies of the electronic spectra of some gaseous oxide molecules. The (0,0) band of the C⁴ Σ⁻ - X⁴ Σ⁻ electronic transition of V0 has been recorded by intermodulated laser-induced fluorescence at a resolution of about 100 MHz over the range 17300 - 17427 cm⁻¹. The hyper- fine structure caused by the ⁵¹V nucleus (I = 7/2) is almost completely resolved. Internal hyperfine perturbations between the F₂ and F₃ electron spin components (where N = J – ½ and J + ½ , respectively) occur in both electronic states; these are caused by hyperfine matrix elements of the type ΔJ = ±1. The C⁴ Σ⁻ state has many local electronic-rotational perturbations, and also suffers from large spin-orbit perturbations by distant electronic states, for which it has been necessary to introduce a second spin-rotation parameter, Y[sub=s], and the corresponding isotropic hyperfine parameter, b[sub=s]. The background theory for this new hyperfine parameter and the calculation of its matrix elements are described. The A⁴π – X⁴ Σ⁻ electronic transition of V0 in the near infra-red has been recorded at Doppler-limited resolution by Fourier transform spectroscopy, and rotational analyses performed for the (0,0) band at 1.5μ and the (0,1) band in 1.18 μ. The hyperfine structure is prominent in the A⁴π ⁴n[sub=5/2] – x⁴ Σ⁻ subband, and in many of the spin satellite branches. As shown by the value of the Fermi contact hyperfine parameter in the A⁴π its electron configuration is (4sσ)¹ (3dδ)¹ (4pπ)¹ in the single configuration approximation. Laser-induced fluorescence spectra of gaseous FeO have proved that the bands whose P and R branches have been analysed rotationally by Harris and Barrow (and which are known to involve the ground state) are Ω'= 4 – Ω''= 4 transitions. The electron configuration (4sσ)¹ (3dδ)³ (3dπ)² ⁵ Δ[sub=i], is the only reasonable assignment for the ground state of FeO. The rotational structure of the 000-000 band of the 2490 Å system of ¹⁵N0₂ (2²B₂ - X²A₁) has been analysed from high dispersion grating spectrograph plates. The band is found to be slightly predissociated, exactly as in the ¹⁴N0₂ isotope, which suggests that it might be usable for laser separation of the isotopes of nitrogen. / Science, Faculty of / Chemistry, Department of / Graduate

No description available.

The methods used for free radical detection are chemical titration, spectrophotometry, Gouy magnetic measurements and NMR shift technique. The free radicals chosen for this study are Fremy' s salt (ON(SO_3 )K_2),2,2-diphenyl-1-picryl hydrazyl (DPPH) 3-carbamoyl-2,2,5,5-tetramethylpyrrolidin-1-yloxy (CTM-pyrrolidin), 3-carbamoyl-2,2,5,5-tetramethyl-3-pyrrolin-1-yloxy (CTM-pyrrolin), 4-acetamido-2,2,6,6-tetramethylpiperidino-1- oxyl (ATMP) and galvinoxyl. The detectable range for each method is given. The NMR method is a fast and convenient method for the magnetic susceptibility determination of a sample in solution, with. accuracy up to ± 1%.