A detailed study was made of the rates of hydrolysis of trans-[Co(en)₂ (NO₂)₂] NO₃ as a function of acid concentration in perchloric, hydrochloric, hydrobromic, sulfuric, and nitric acids up to high concentrations of acid. All the data could be fitted 1 to the equations proposed by Bunnett¹. (1) [log k₁ - log (H +)] = w*log a_w + Ratio and (2) [log k₁ - log Hₒ] = wlog a_w + Constant. In these equations, k₁ is the first-order rate constant, (H+) is the hydrogen-ion concentration, a_w is the activity of water in the solution, and Hₒ is the Hammett acidity function. The w and w* are the slopes"Ratio" and"Constant" are the intercepts, of plots of the left side of eq. 1 or eq. 2 versus the log a_w. Bunnett proposed that for the hydrolysis of organic compounds, wand w* values are indicative of mechanism. From the constant values of w* obtained in hydrobromic, hydrochloric, sulfuric, and perchloric acids, a mechanism involving water acting as a nucleophile was indicated. The different value of w* obtained in nitric acid was consistent with the known difference in hydration in nitric acid. Values of w also were consistent with nucleophilic attack by water. Rates of hydrolysis of trans-[Co(en)₂ (NO₂)₂] NO₃ were also measured as a function of temperature. In perchloric acid, good linear Arrhenius plots were obtained from which the energy of activation was 23.6 + 0.8 kcal. and the activation entropy was 2.9 + 2.3 eu. at the 95% confidence level. Arrhenius plots in hydrochloric, hydrobromic, sulfuric, and nitric acids were curved, the amount of curvature increasing with acid concentration and temperature. At lower temperatures, the activation energy was 23-24 kcal. but decreased to as low as 12 kcal. at high acid concentrations and higher temperatures. This behavior is consistent with two consecutive reactions, the second reaction having the lower activation energy; the latter reaction was identified with addition of an anion to the intermediate nitroaquo complex. A series of dinitro complexes with"inert" ligand size varying from ammonia to 1, 10-phenanthroline was made; these complexes were studied in hydrochloric and perchloric acids. The first-order rate constant decreased as the size of the ligand increased, indicating direct participation of water in an SN₂ reaction. The values of w* in both acids were consistent with nucleophilic participation of water; for the bipyridine and orthophenanthroline complexes a lower value of w was observed. This observation could be related to the expectation that these large complexes would not be hydrated as much as the smaller compounds. A mechanism for the hydrolysis reaction consistent with the observed behavior was given. According to this mechanism, the complex is first protonated in an equilibrium step; the protonated species reacts with water to give the nitroaquo complex as the rate determaining step for all compounds· in all acids. Finally, the nitroaquo complex reacts with an anion Z⁻ (except when Z⁻ is ClO₄⁻) to give the final product of reaction. This investigation shows that Bunnett's equations are useful for correlating data, but some caution in accepting Bunnett's mechanistic interpretations from the values of w and w* is necessary. It is satisfying that application of these ideas to a totally different system (inorganic complexes) lead to a mechanism consistent with current ideas about the reactivity of coordination compounds of cobalt(III). / Doctor of Philosophy

An approximate spin-extended method is developed in which the Occupation Numbers of the Natural Orbitals of Charge are varied but the orbitals themselves are frozen. A number of techniques for obtaining appropriate Natural Orbitals of Charge are presented and these techniques are employed in semi-empirical and Ab-Initio studies. The molecules chosen for the Ab-Initio study, O₃, F₂ and C₂, are particularly interesting as Restricted Hartree-Fock (RHF) theory fails to provide an adequate description of these systems at their equilibrium geometries. The results of this work indicate that Approximate SpinExtended Hartree-Fock (ASEHF) theory provides a qualitatively acceptable description of bond dissociation, correctly predicts the symmetry of molecular states where RHF theory fails, and provides an efficient means for obtaining a significant amount of correlation energy in molecular calculations. Moreover, when the Generalized Valence Bond form of the ASEHF wavefunction is employed, a particularly useful description of the molecule is obtained. Specifically, the success or failure of the Woodward-Hoffman becomes transparent when the system is analyzed in terms of this wavefunction. / Ph. D.

Two on-line multidimensional HPLC systems are described; one coupling two liquid chromatographic columns (LC/LC), and one coupling liquid and gas chromatography (LC/GC). Theoretical equations relating the reproducibility, accuracy and transfer efficiency to system operating variables such as flow rate, retention time, column efficiency and transfer volume have been developed. These effects are explained and verified using experimental systems. Three major applications are shown for each of the systems. For the LC/LC systems, these include; the analysis of caffeine and theophylline in biological fluids; the analysis of hydrocarbon group types in fuels and oils and the analysis of catecholamines in urine with electrochemical detection. For the LC/GC system, the applications include; the analysis of pesticides in butter, the analysis of hydrocarbon group types in fuels and the analysis of polycyclic aromatic hydrocarbons in petroleum related samples. / Ph. D.

Two methods for the preparation of bisphenol-A polycarbonate-polydimethylsiloxane multi-block or segmented copolymers have been investigated. The first is a solution reaction which utilized the interaction of a preformed, α,ω-dimethylamino terminated polydimethylsiloxane with a preformed, α,ω-hydroxyl terminated polycarbonate. This synthesis produces perfectly alternating block copolymers wherein the block molecular weights are equal to the number average molecular weights of the respective well-characterized oligomeric precursors. This silylamine-hydroxyl reaction is clean, efficient and can be utilized to produce copolymers which consist of a variety of block sizes ranging from approximately dimeric to 25,000-30,000 g/mole. Intrinsic viscosities of 1.0 di/g. are easily attained and mechanical properties range from being elastomeric to rigid depending on the overall bulk composition. However, these copolymers do incorporate the relatively hydrolytically unstable [the molecular geometry of the bonds between Si—O—C] moiety as the link between the block structures. Thermal analysis of these perfectly alternating block copolymers reveals their two-phase nature even at block molecular weights as low as ~ 2000 g/mole. Surface analysis of the copolymer films performed by X-ray photoelectron spectroscopy (known as ESCA) indicates preferential surface migration of the siloxane component in all cases studied. In addition, preliminary ESCA studies of blends of these materials with a commercial polycarbonate homopolymer show siloxane on the surface even when bulk percentage of siloxane is as low as 0.05 wt.%. A method for estimating the percentage of the surface area sampled by ESCA which is comprised of siloxane was developed. Results derived from this procedure indicate the particular importance of the siloxane block size in determining surface composition. The second synthetic method investigated was a phase transfer catalyzed reaction which produced a more randomly coupled block copolymer. Preformed hydroxypropyl, hydroxybutyl, or carboxypropyl terminated siloxane oligomers were used whereas the polycarbonate blocks were produced in situ. These reactions, in general, were found to be less efficient than the "silylamine reaction." However, promising results were obtained using the hydroxyl-terminated siloxanes in two cases: 1, when siloxanes of <Mn> ≃ 5000 g/mole were used and, 2, when the siloxane oligomers were capped with the dichloroformate of bisphenol-A prior to the interfacial step. When these phase transfer catalyzed reactions were carried out utilizing the carboxyfunctional oligomeric siloxanes, reasonably high molecular weight copolymers could be obtained. The best results were achieved when an anhydrous "pre-phosgenation" step was utilized. In contrast to the perfectly alternating copolymer systems previously described, these randomly coupled block copolymers contain the ≡SiC≡ bond which may be preferable in some applications (e.g. biomaterials). One problem which must be addressed for the synthesis of block copolymers derived from preformed oligomers is the separate synthesis of those oligomers. The techniques for the preparation of hydroxybutyl and carboxypropyl terminated siloxane oligomers were developed in this research. Moreover, a novel, facile method for the production of polycarbonate oligomers of well-controlled number average molecular weights was also devised. This procedure involves monofunctional capping of a calculated fraction of the phenolic end-groups prior to oligomerization by the direct phosgenation route. The hydroxyl end-groups can then be regenerated by selective hydrolysis of the protecting groups. Trimethylsilyl chloride, trifluoroacetic acid and trifluoroacetic anhydride were shown to be suitable capping reagents. Additional information pertaining to the physical characteristics of these novel copolymers was derived from collaborative studies with other colleagues at this university and at the University of Akron. / Ph. D.

A study of the relation between mechanical properties and molecular weight distribution of polyvinyl alcohol was carried out. The polyvinyl alcohol was fractionated with respect to degree of polymerization (D.P.) by precipitation of the fractions from 2% solution in water with a n-propyl alcohol-water nonsolvent. The initial small scale procedure was modified so as to accumulate fractions in sufficient quantity for the testing of mechanical properties and for the preparation of blends. The fractions were refractionated until further refractionation gave no further change in D.P. The homogeneous fractions were cast into films using water as the solvent. These films were then used in determining mechanical properties. A series of three blends, all with normal distributions and the same D.P. at the maximum of the differential distribution curve but with variable heights at the maximum D.P., were prepared and the mechanical properties were determined. A second series of six blends, all with constant height and constant D.P. at the maximum, but all being skewed from the normal distribution, were prepared and their mechanical properties were determined. A correlation of tensile strength with the shape of the distribution curve was found to have the form: T = (8200-l0^{6 .87-.00476P_m}) + (2.786 x 10⁶H^4.767-1000) + (1450-10.4B), Where T is tensile strength; P_m is D.P. at maximum of the differential distribution curve; H is height at the maximum; B is area skew. The second term in brackets is dropped when H is greater than 0.190 and the entire last term is dropped when B = 0. A graphical correlation of mechanical properties with total percentage of low D.P. material gives fairly good results for polyvinyl alcohol and, using the data of other investigators, for cellulose nitrate, cellulose acetate, and polyvinyl acetate. / Ph. D.

The effects of temperature on the first and second acid dissociation constants of 8-hydroquinoline (oxine) have been measured and the following thermodynamic values assigned to solutions of 0.1M strength. For the process, H₂Ox⁺ ⇋ H⁺ + HOx at 25°C., pK₁ = 5.103, ΔH = 5.7 kcal/mole, and ΔS° = -4 e.u. For the second acid dissociation constant, HOx ⇋ H⁺ + Ox⁻ at 25°C., pK₂ = 9.76, ΔH = 6.3 kcal/mole, and ΔS° = -23 e.u. The values for the second acid dissociation constant agree with those previously published. The values for the first dissociation constant have not been previously determined but are consistent with the result expected for reactions of this charge and chemical type. The partition of 8-hydroxyquinoline was determined as a function of temperature for the following systems: water-benzene, water-nitrobenzene, water-toluene, and water-chloroform. The pertinent thermodynamic values for the partitions are as follows: a. Water-benzene: log P = 2.44; ΔH = -2.0 kcal/mole; ΔS° = 4.5 e.u. b. Water-nitrobenzene: log P = 2.67; ΔH = -1.7 kcal/mole; ΔS° = 6.6 e.u. c. Water-toluene: log P = 2.33; ΔH = -1.3 kcal/mole; ΔS° = 6.2 e.u. d. Water-chloroform log P = 2.703 ΔH = -2.5 kcal/mole; ΔS° = 3.9 e.u. The values obtained for the water-chloroform system agree well with tho literature and serve as a check on the rest of the data. The effect of temperature on the overall extraction constant of nickel oxinate and ferric oxinate was measured for the system water-chloroform. The thermodynamic values found for the process Me⁺ⁿ + nHOx ⇌ MeOx_n + nH⁺ are the following at 25°C. and an ionic strength of 0.1M: a. Nickel oxinate log Kₑ = -2.73; ΔH = -40.4 kcal/mole; ΔS° = -148 e.u. b. Ferric oxinate log Kₑ = -3.97; ΔH = -14.1 kcal/mole; ΔS° = -29 e.u. The effect of ionic strength on the partition of S-hydroxyquinoline between water and benzene was determined for four different uni-univalent salts: potassium chloride, sodium chloride, potassium nitrate and sodium perchlorate. The “salting out” of oxine observed is typical of non-electrolytes. / Doctor of Philosophy

The carbopalladation process has been extended to form carbocyclic systems. The intramolecular version of the carbopalladation process stereospecifically generates fused bicyclic palladium complexes upon treatment of cis and trans methyl-S-methyl-2-carbomethoxy-7-thio-hept-5-en-l-oate with lithium tetrachloropalladate and potassium tertbutoxide in moderate yield. The cyclization of the corresponding amines, cis and trans methyl-N,N-dimethyl-2-carhomethoxy-7-amino-hept-5-en-1-oate, proceeded more rapidly and stereospecifically generate stable fused bicyclic palladium complexes in excellent yield. NMR studies revealed the bicyclic complexes generated in this manner resulted from a trans addition of palladium and the carbon nucleophile. Direct reduction with sodium borohydride (for sulfides) or hydrogen (for amines) resulted in cyclopentyl sulfides and amines in good to excellent yields. The reaction was extended to ketoesters and homoallylic amines to give a variety of cyclopentanoids. Extensions to larger ring systems yielded unstable palladium complexes. These complexes underwent oxidation at the amine functionality upon warming. Hydrolysis of the intermediate complexes yielded cyclohexyl and cycloheptyl aldehydes whereas reduction with hydrogen in the presence of molecular sieves gave cyclohexyl and cycloheptyl amines in excellent yields. Attempts at utilizing olefin mid carbon monoxide insertions to form macrocycles proceeded poorly, if at all. / Ph. D.

The process of C. K. Bradsher for preparing mesosubstituted anthracene and 1,2-benzanthracene derivatives has been widely extended. The purpose of this research was to extend this process to the thiophene series, in addition to continuing some work on the 9-(dimethyl)- phenylanthracene series and some newly developed l,2-dimethyl-10-phenyl)-anthracene series. 2'-Thienyl-2-benzylbenzophenone was prepared and cyclized to 9-(2-thienyl)-anthracene by the conventional methods in good yields. 3'-Thienyl-2-benzylbenzophenone was prepared in a different way, indirect and tedious. Thiophene was iodinated to tetraiodothiophene and tetraiodothiophene was reduced with aluminum-amalgam to 3-iodothiophene. 3-Thienylmagnesiumiodide was prepared by the entrainment technique and condensed with 2-cyanodiphenylmethane to 3'-thienyl-2-benzylbenzophenone. This ketone was cyclized to 9-(3'-thienyl)-anthracene by the standard acid mixture. 2-(1-Naphthylmethyl)-phenyl-2-thienylketimine formed from the condensation between 2-(1'-naphthylmethyl)- benzonitrile and 2-thienylmagnesiumbromide resisted hydrolysis to the corresponding ketone but cyclized quantitatively to 10-(2-thienyl)-l,2-benzanthracene. Following the classical Bradsher process, the reaction between 2-thienylmagnesiumbromide and o-chlorobenzaldehyde was unsuccessful because the hydrol formed was so reactive that it polymerized rapidly and ended up with tars. A cross-condensation reaction between 2-ethynylmagnesium bromide and o-chlorobenzylchloride was utilized to give the expected 2-(2'-chlorobenzyl)- thiophene in one step. This new procedure has been widely extended to the other series. It is preferred to. the conventional Bradsher process and the Friedel-Crafts method in terms of time, chemicals, and unequivocal products formed. 2-(2'-Chlorobenzyl)-thiophene so formed by the new procedure was converted by von Braun reaction to the corresponding nitrile and the nitrile was allowed to react with phenylmagnesiumbromide to give 2-(2-thienylmethyl)-benzophenone. The above ketone was cyclized almost quantitatively to 4-phenyl-thiophanthrene. The thiophene-containing ketones prepared in this research are very viscous oils. In order to have crystalline derivatives for identification purposes these ketones were oxidized to the corresponding diketones. 2-Thienyl-2-benzylbenzophenone and 2-(2'-thienylmethyl)- benzophenone gave the identical diketone, 2-benzoyl-2'- thienyl-benzophenone. 3 1 -Thienyl-2-benzylbenzophenone was oxidized to anthraquinone by using 25% sulfuric .acid, acetic acid and sodium dichromate. The formation of anthraquinone indicates that the monoketone cyclized first and then split off the thiophene ring. Apparently, the 3-thienyl group is more susceptible to oxidation than the 2’-thienyl group. A study of the oxidation of anthracene derivatives for structure proof was made using 9-phenylanthracene as a model compound. It was oxidized to 10-phenyl-10-hydroxyanthrone which resisted further oxidation to anthraquinone. It is obvious that the tertiary alcohol system stabilized the phenyl ring. In continuing the work on the 9-(dimethyl)-phenylanthracene series, the yields of three ketimine salts and one ketone were improved. Satisfactory analytical data was obtained for 2’,6’-dimethyl-2-benzylbenzophenone. In an attempt to overcome the steric effects offered by 2,6-ortho groups, 2-(2',3'-dimethylbenzyl)-benzonitrile was prepared. The starting material for this series, l-bromo-2,3-dimethylbenzene has been prepared in 47% yield by diazotization in comparison with the previous yield of 30%. / Doctor of Philosophy

In the light of previous investigations of the cyclodehydration of o-benzylbenzophenones and analogous compounds to polycyclic aromatic hydrocarbons a study of 2'-substituted-2-benzylbenzophenones was undertaken. The previous work had been concerned with the rates of cyclization or 3'-substituted and 4'-substituted ketones in which no ortho effects were operating. The results of these investigations could be adequately interpreted according to the electronic theory of the English school. In the 2'-series ortho effects are superimposed upon the usual electronic effects. When the 2'-substituents were halogens a decrease in rate was observed with an increase in the size of the halogen atom. This indicates that one aspect of these ortho effects is steric. These reactions were carried out in a sealed tube in hydrobromic and acetic acids at 150°C rather than at 117.5°C which was the temperature used previously. This increase was necessary to get a reaction with all the ketones concerned. Even under these conditions the 2'-hydroxy-2- benzylbenzophenone cyclized very poorly. This was attributed to a second ortho effect due to intramolecular hydrogen-bonding between the hydroxy group and the keto group. Infrared spectroscopy demonstrated that this type of hydrogen-bonding is present. The 3'-hydroxy-2-benzylbenzophenone apparently cyclizes very readily since the 3'-methoxy compound is cyclized to 9-(3-hydroxyphenyl)anthracene quite readily. Thus, the two reactions in the 3'-series, ether cleavage and the cyclization reaction both proceed more rapidly than the cyclization of the 21-hydroxyketone. The 9-(2- methoxyphenyl)anthracene can be prepared in good yield by the use of phenyl acid phosphate or polyphosphoric acid as a cyclizing agent. This compound can be easily converted to 9-(2-hydroxyphenyl)anthracene by heating under reflux with a mixture of hydrobromic and acetic acids. The halogen substituted ketones are very viscous oils. In order to have solid derivatives for identification purposes these ketones and the methyl derivative were oxidized to the corresponding o-dibenzoylbenzenes which are all solids. Chromium trioxide in acetic acid was usually found to be the best reagent for this purpose. Two of the 9-phenylanthracenes were oxidized to the corresponding 9-phenyl-9-hydroxy-10-anthrones using sodium dichromate in acetic acid. This reaction is carried out very readily and in excellent yield. It may prove to be valuable in the identification of other aromatic polycyclic hydrocarbons. The reduction of 2'-fluoro-2-benzylbenzophenone to 2'-fluoro-2-benzylbenzhydrol was carried out using sodium borohydride in pyridine. The infrared spectrum showed no evidence of intramolecular hydrogen-bonding between the fluorine atom and the hydroxy group. A study of the use of solid surfaces in aromatic cyclodehydration was made using 4'-chloro-2-benzylbenzophenone as a model compound. Experiments with a large number of solids brought to light two factors which seem to contribute to the ability of a surface to catalyze the reaction. One is the acidity of the surface and the other is the ability of a surface to bind the ketone irreversibly. Surface acidity was measured in terms of the H₀ function by the use of Hammett indicators adsorbed on the surface. The color of the adsorbed indicators was used to obtain the H₀ values. The activity of each surface was measured in terms of the percent yield of the product. The solid of highest activity and highest acidity was a silica-alumina cracking catalyst containing 13% alumina. This catalyst gave a 94% yield with 4'-chloro-2-benzylbenzophenone compared to only about 25% on the commonly used Fisher activated alumina. 2'-Methyl-2-benzylbenzophenone was cyclized in 60% yield. This ketone gave a yield of only 20% when cyclized in the hydrobromic-acetic acid mixture at 180° for 9 hr. 2’,6’-Dimethyl-2-benzylbenzophenone has resisted all attempts to cyclize it including an attempt with the most active solid surface available, the silica-alumina cracking catalyst. / Ph. D.

The oxidation of arsenic(III) by chromium(VI) requires the postulation of uncommon oxidation states as reactive intermediates. The possible intermediates of chromium are the divalent, tetravalent, and pentavalent states. Recently arsenic(IV) has been postulated as a reactive intermediate in certain oxidations of As(III). A detailed study of the reaction rate law in various solutions has been carried out in order to determine the nature of the intermediates found in the Cr(VI)-As(III) reaction. The effects of hydrogen ion and buffer concentrations have been determined. The mechanism postulated depends upon the solution in which the reaction occurs. The kinetics of the chromium(VI)-arsenic(III) reaction have been measured in perchloric acid solutions, acetic acid solutions, acetic acid-acetate buffers, ammonium nitrate solutions, and dihydrogen phosphatemonohydrogen phosphate buffers. The rate laws are: Perchloric Acid - -d[Cr(VI)]/ dt = [k₀ + k_H[H⁺] + k_H² [H⁺]²] (K’ [As (III)][Cr(VI)])/(l+K’[As(III)]) Acetic Acid-Acetate Buffers - -d[Cr(VI)]/ dt = [k₀ + k_HOAc[HOAc]} (K’ [As (III)][Cr(VI)])/(l+K’[As(III)]) Acetic Acid - -d[Cr(VI)]/ dt = [k₀ + k’_HOAc[HOAc]] (K’ [As (III)][Cr(VI)])/(l+K’[As(III)]) H₂PO₄⁻- HPO₄⁻² Buffers - -d[Cr(VI)]/ dt = (kK”[As(III)][Cr(VI)][H₂PO₄⁻])/(1+k”[H₂PO₄⁻]) Ammonium Nitrate Solutions - -d[Cr(VI)]/ dt = k₂ [As(III)][Cr(VI)] The reaction studied was the oxidation of As(III) by Cr(VI} according to the stoichiometry 3 As(III) + 2 Cr(VI) = 3 As(V) + 2 Cr (III) Rate data were obtained spectrophotometrically and by iodometric titration depending upon the region being investigated. At high As(III)-low Cr(VI), the Cr(VI) concentration was followed as a function of time at 350 mu using a Beckman DU spectrophotometer. The validity of Beer's law with respect to Cr(VI) was checked. At low As(III)-high Cr(VI), the unreacted As(III) concentration was determined by iodometric titration. In phosphate buffers, the HCrPO₇⁻² complex is indicated as an oxidizing agent. In the other systems, the following mechanism is consistent with the experimental facts. As(III) + HCrO₄⁻ [stacked right and left arrows with K above] As(III) · HCrO₄ As (III), As(III) · HCrO₄⁻ [right arrow with kₒ above] products, rate determining As(III) · HCrO₄⁻ + HOAc [right arrow with K_HOAc above] products, rate determining As(III) · HCrO₄⁻ + H⁺ [right arrow with k_H] products, rate determining In perchloric acid, H₂CrO₄is indicated as participating in the oxidation by the following steps: HCrO₄⁻ + H⁺ [stacked right and left arrows with Kₐ above] H₂CrO₄, ₂CrO₄ + As(III) [stacked right and left arrows with K above] As(III) · H₂CrO₄ The decomposition of As(III}•H₂CrO₄ is also acid catalyzed. In the mechanism described above, the products of the rate determining steps are arsenic(V) and chromium(IV). The existence of chromium(IV) is supported by the induced oxidation of ferrocyanide by the As(III)-Cr(VI) reaction. The chromium(IV) reacts immediately with chromium(VI) to form chromium(V) which can oxidize As(III) to As(V) directly. This is in agreement with the experimentally observed stoichiometry and accounts for the use of K' in the rate laws where K’ is actually 2K. The value of K is obtained from plots of 1/k_m versus 1/As(III). The value of K which describes all data reported is 17.2 1.mole⁻¹. In all three systems, perchloric acid, acetic acid, and acetic acid-acetate, the same kₒ is obtained. The specific rate constant for the acetic acid catalysis is 1.99 x 10⁻³ 1.⁻² mole² sec.⁻¹. Comparison of k"_HOAc terms for acetic acid and acetic acid-acetate solutions shows that k”_HOAc for the acid solution is larger than that for the buffers. This difference is explained on the basis of a medium effect. / Doctor of Philosophy