CARBON-13 MAGNETIC RESONANCE SPECTROSCOPY Thesis by Frank Julian Weigert In Partial Fulfillment of the Requirements For the Degree of Doctor of Philosophy California Institute of Technology Pasadena, California 1968 (Submitted May 15, 1968) ii ACKNOWLEDGMENTS I should like to thank Professor John D. Roberts for both his direction and indirection during my graduate work. It is perhaps typical of his great foresight that he should have the DFS-60 ready and waiting for me at the time I began my graduate work. The latitude given me in the choice and execution of this thesis has been greatly appreciated. Special thanks are given to Joe Dence for his advice in the manipulation of the English language, Jim Magnust'n for training in the care and repair of nmr equipment, and Jacqueline Kroschwitz 13 for permission to discuss data on the c spectra of the continuous chain alcohols. Several helpful discussions with Stan Manatt (Jet Propulsion Laboratory) and Professor David Grant (University of Utah) are also acknowledged. I would like to thank Birgitta Isaacson for her artistic efforts and Jill Williamson for her fast and expert typing. Financial assistance in the form of a National Science Foundation predoctoral fellowship {1965-1968) and a part-time teaching assistantship (1965) are gratefully acknowledged. iii Abstract 13 High-resolution, natural-abundance C spectra have been 13 obtained from a wide variety of organic compounds; C chemical shifts and coupling constants have been correlated with other molecular properties. Geminal and vicinal, carbon-proton couplings in benzene and the five- and six-membered aromatic heterocycles have been related to the corresponding proton-proton couplings in substituted ethylenes. carbon-proton coupling constants in benzene are ~he JCCH = + 1. 0, JCCCH = + 7. 4 and JCCCCH = - 1.1 Hz. Extended Huckel wavefunctions are uniformly poor in explaining the long- range, carbon-proton couplings in aromatic systems. Couplings between carbon and elements other than hydrogen 13 have been observed in proton decoupled c spectra. All of the carbons in fluorobenzene and 1-fluoronaphthalene, but only six of the carbons in 2-fluoronaphthalene are coupled to the fluorine. One-bond, carbon-phosphorus coupling in trialkylphosphines is negative, while one-bond, carbon-phosphorus coupling in tetra alkylphosphonium ions is positive. Atoms which do not use hybrid orbitals to form bonds to carbon (F, P{III), Se, Te) may have negative, one-bond coupling constants because of the failure of the average energy approximation. One-bond between ~ouplings iv carbon and carbon, silicon, tin, lead and mercury appear to be explainable in terms of an effective nuclear charge and the s-bond order of the metal. Couplings between carbon and nitrogen and phosphorus(IV) have significant negative contributions to the Fermi contact coupling expression, though, within one series, correlations with s-bond order may be valid. Carbon-carbon coupling in cyclopropane derivatives (10-15 Hz) is consistent with a high degree of p character in the interior orbitals. Some two- and three-bond carbon-carbon coupling constants have also been observed. 13 Substituent effects of hydroxyl groups on the c chemical shifts of continuous-chain alkanes depend both on steric and elec· tronic factors. The hydroxyl substituent effects in the Long-chain, primary alcohols are a = -48. 3, {:3 = -10. 2, and y = +6. 0 ppm. The upfield y effect is attributed to steric crowding in the gauche conformations. Additivity of the hydroxyl and carbonyl and alkyl substituent effects in alkyl-substituted cyclohexanols and cyclo hexanones has been demonstrated. v TABLE OF CONTENTS PAGE CHAPTER I Introduction 1 CHAPTER II Carbon-Proton Coupling in Aromatic Compounds 21 CHAPTER ID Coupling of Carbon and Nuclei Other than Protons 63 CHAPTER IV Carbon-13 Chemical Shifts in Alcohols and Ket ones 160 Materials 216 Propositions 218 vi LIST OF TABLES Table Page 13 2. 1 c Chemical Shifts and Coupling Constants of the Five- Membered Nitrogen Heterocycles. 27 2. 2 Long-Range, Carbon-Proton Coupling Constants in the Monosubstituted Five-Membered Heterocycles. 28 2.3 Long-Range, Carbon-Proton Coupling Constants in Some Methyl-Substituted Heterocycles. 29 2. 4 Coupling Constants in Six-Membered Aromatic Compounds. 44 2. 5 Long-Range, Carbon-Proton Coupling in Methyl-Sub stituted Six-Membered Heterocycles. 45 2.6 A Comparison of Carbon-Proton Coupling in Benzene with Proton-Proton Coupling in Ethylenes. 50 2. 7 A Comparison of Proton-Proton Coupling in Substituted Ethylenes with Carbon-Proton Coupling in Six Membered Aromatic Heterocycles. 58 2. 8 Carbon-Proton Coupling Constants in Aromatic Compounds Calculated from Extended Hlickel Wave- functions. 59 3. 1 Carbon-Fluorine Coupling in Ortho Substituted Fluoro- benzenes. 66 3. 2 Carbon-Fluorine Coupling- in Meta Substituted Fluoro- benzenes. 67 3. 3 Carbon-Fluorine Coupling in Para Substituted Fluoro- benzenes. 68 vii Table Page 3.4 Chemical Shifts and Coupling Constants Obtained from 13 the Analysis of the Proton Decoupled c Spectra of the Difluorobenzenes. 76 3.5 Carbon -Fluorine C oup iing in Polyfluorobenzenes. 79 3.6 Carbon-Fluorine Coupling in Pentafluoroiodobenzene. 80 3.7 Carbon-Fluorine Coupling in 1- and 2-Fluoro- naphthalene. 83 3.8 Carbon-Fluorine Coupling in 1, 5-Difluoronaphthalene. 91 13 3.9 Carbon-Fluorine Coupling and c Chemical Shifts of Some Continuous -Chain, Primary Fluorides. 94 13 3.10 c Chemical Shifts of Some AliI?hatic Geminal Di- fluorides. 96 3.11 NMR Parameters of Phosphorus-Containing Compounds. 98 3.12 Carbon-Carbon Coupling in Cyclopropane Derivatives. 107 3.13 Carbon-Carbon Coupling in Aliphatic Compounds. 110 3. 14 Carbon-Carbon Coupling in Alicyclic Compounds. 111 3.15 Carbon-Carbon Coupling in Aromatic and Olefinic Compounds. 114 3. 16 Long-Range, Carbon-Carbon Coupling. 117 3.17 Carbon-Nitrogen Coupling. 119 3. 18 Carbon-Mercury Coupling. 121 3.19 Coupling of Carbon and Miscellaneous Elements. 123 3.20 Coupling Constants Calculated from Extended Hiickel Wavefunctions. 127 viii Table Page 3.21 Comparison of the Calculation of the One-Bond Carbon- Metal Coupling by the Method of Karabatsos and Smith with the Observed Values. 131 3.22 A Comparison of Coupling Constants Involving Carbon with Those Involving Protons. 133 3.23 Coupling to the Carbonyl Carbon in Some Esters. 141 3.24 Long-Range Coupling to Carbon in Aromatic Derivatives. 148 3.25 Geminal Coupling to Carbon in Aliphatic Derivatives. 149 3.26 A Comparison of Vicinal Coupling to Carbon and Protons in Aliphatic Derivatives. l!JO 13 4. 1 C Chemical Shifts in Continuous-Chain Alcohols. 161 13c 4.2 Substituent Effects of the Hydroxyl Group on the Chemical Shifts of the Continuous-Chain Alcohols. 163 4.3 Gauche and 'I'rans Hydrogen-Hydrogen Interactions in Various Conformations of the Continuous-Chain Alkanes and Alcohols. 174. 13c 4.4 Chemical Shifts and Substituent Effects in Some Branched Alcohols. 180 13c 4.5 Chemical Shifts in Cyclohexanols. 183 13c 4.6 Chemical Shifts in Alkyl Cyclohexanes. 184 4.7 Hydroxyl Substituent Effects in Substituted Cyclo- hexanols. 186 4.8 13c Chemical Shifts in Cyclic Alcohols. 194 13c 4.9 Chemical Shifts in Alkyl Cyclohexanones. 199 ix Table Page 13 c 4. 10 Substituent Effects on the Chemical Shifts of Alkyl Cyclohexanones Relative to Cyclohexane. 203 13 4.11 Carbonyl Substituent Effects on the C Chemical Shifts of Alkyl Cyclohexanones. 204 13 4. 12 c Chemical Shifts and Substituent Effects in Cyclic Ketones. 207 1 I. Introduction 13c The practical use of nuclear magnetic resonance has been hindered by low natural abundance of this nucleus (1. 1%), low inherent sensitivity to nmr detection (1. 6% relative to protons at constant field) and long relaxation times (1, 2). Various methods have been used to overcome these difficulties. 13 Isotopic enrichment increases the number of C nuclei in the active sample volume. Furthermore, specific introduction 13 of c allows unambiguous assignments of the carbon resonances to be made. The main disadvantage of this technique is the expense 13 of preparing the labelled mat~rials. The possibility of using c as a non-degradative tracer in labelling studies has received only limited attention because of the experimental difficulties in observing 13 c resonances (3). 13 The INDOR technique developed by Baker (4) allows the c resonances to be detected with the sensitivity of proton nmr. For 13 observation of c resonances by INDOR the field of the spectrometer 13 is adjusted so that a c satellite resonance in the proton spectrum is 13c on resonance. A weak rf field is then swept through the spectrum. 13c When a transition in the spectrum is irradiated which has an energy level in common with the proton line, a perturbation of the proton resonance is seen. This technique demands excellent