University of Southern California Dissertations and Theses

Nuclear Magnetic Resonance Studies Of The Rotational Isomerism In (Alpha,Beta)-Unsaturated Acyl Fluorides

(USC Thesis Other)

Nuclear Magnetic Resonance Studies Of The Rotational Isomerism In (Alpha,Beta)-Unsaturated Acyl Fluorides

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content NUCLEAR MAGNETIC RESONANCE STUDIES OF THE ROTATIONAL ISOMERISM IN a,B-UNSATURATED ACYL FLUORIDES by Frank Fang-Sheng Lin A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Chemistry) June 1971 72-3786 I LIN, Prank Fang-Sheng, 1941- ! NUCLEAR MAGNETIC RESONANCE STUDIES OF THE | ROTATIONAL ISOMERISM IN #, ytf-UNSATURATED } ACYL FLUORIDES. | I University of Southern California, Ph.D., 1971 I Chemistry, organic f | University Microfilms, A XEROX Com pany, Ann Arbor, Michigan THIS DISSERTATION HAS BEEN MICROFILMED EXACTLY AS RECEIVED U N IV E R S IT Y O F S O U T H E R N C A L IF O R N IA T H E G R A D U A T E S C H O O L U N IV E R S IT Y P A R K LO S A N G E L E S , C A L IF O R N IA 9 0 0 0 7 This dissertation, written by .....Frank. Fang-sheng;, .Lin......... under the direction of h..i . s „ „ Dissertation Com mittee, and approved by all its members, has been presented to and accepted by The Gradu ate School, in partial fulfillment of require ments of the degree of D O C T O R O F P H IL O S O P H Y Dean June 1971 Chairman PLEASE NOTE: Some Pages have indistinct p rin t. Filmed as received. UNIVERSITY MICROFILMS To my wife and my parents ACKNOWLEDGMENTS I wish to express my thanks to Professor ; Kenneth L. Servis, without his advices and help this | work would not have been possible. | To my wife, Jane, I extend my sincere j appreciation. Her contributions in moral support, in typing the reports and this dissertation can not be overemphasized. Teaching assistantship awards from Department of Chemistry, University of Southern California during my graduate years are gratefully acknowledged. Thanks also due to Dr. Eric Noe for helpful discussions. I would also like to thank Professor G. Binsch for making his DNMR computer program available to us. CONTENTS Page ACKNOWLEDGMENTS.................................... iii LIST OF TABLES..................................... vi LIST OF FIGURES.................................... viii ABSTRACT............................................ x ] i CHAPTER 1. INTRODUCTION................................ 1 1-1. Elements of Conformational Analysis... 1 1-2. The Nature of Potential Function Hindering the Internal Rotation Around the C-C Single Bond in Conjugated System..................... 14 1-3. Nuclear Magnetic Resonance Spectros copy and Conformational Analysis 17 2. RESULTS...................................... 24 2-1. Syntheses of Substituted a ,B-Unsatura ted Acyl Fluoride......... 24 2-2. Analysis of NMR Spectra of Acryloyl Fluoride......... 29 ! 2-3. Analysis of NMR Spectra of Crotonyl j Fluoride...................... 46 I i I 2-4. Analysis of NMR Spectra of Cinnamoyl Fluoride............... 56 2-5. Analysis of NMR Spectra of £-Methoxy- cinnamoyl Fluoride............. 60 3. DISCUSSION.............................. 69 4. EXPERIMENTAL................................ 101 | 4-1. Instrumentation and Techniques... 101 I Contents (cont'd) i Page 4-2. Preparation of Compounds Used in This Study................. 107 REFERENCES.......................................... 119 APPENDIX 1.......................................... 125 LIST OP TABLES Table Page I. NMR Parameters for Acryloyl Fluoride.......... 32 II. The Calculated 100 MHz NMR Spectrum of Acryloyl Fluoride in Carbon Tetrachloride Solution......... 33 III. The Calculated ^F Decoupled 100 MHz *H NMR Spectrum of Acryloyl Fluoride............. 37 IV. Chemical Shifts, Coupling Constants, Populations, and AG Value for Equilibrium of Acryloyl Fluoride at -l6l.5° in Vinyl Chloride Solution............................. 4l V. The Ratios of Populations of the s-cis to the s-trans Conformers of Acryloyl Fluoride As a Function of Temperature.................. 42 VI. Kinetic Data for Acryloyl Fluoride in Vinyl Chloride Solution....................... 45 VII. Chemical Shifts, Coupling Constants, Populations, and AG Value for Crotonyl Fluoride at -136°in Vinyl Chloride Solution............. 52 VIII. Kinetic Data for Crotonyl Fluoride in Vinyl Chloride Solution.......................... 56 IX. Chemical Shifts, Coupling Constants, Populations, and aG Value for Cinnamoyl Fluoride at -130.5° in 60% Vinyl Chloride and 40% Chlorodifluoromethane Solution. .... 59 X. Kinetic Data for Cinnamoyl Fluoride in 40% CHFC12 and 60% Vinyl Chloride............. 60 XI. Chemical Shifts, Coupling Constants, Populations, and aG Value for £-Methoxy- cinnamoyl Fluoride at -140° in 70% Vinyl Chloride and 30% Acetone Solution............. 65 XII. Kinetic Data for £-Methoxycinnamoyl Fluoride in 30% Acetone and 70% Vinyl Chloride Solution........ 66 vi Table Page | i XIII. Data of Some Three-Bond H-F Coupling Constants.......... ........................... XIV. The Values of Vlt V2# and Vo at Different Values of 9max............ 80 XV. The Ionization Potentials, Dipole Moments, and Carbonyl Stretching Frequencies of Some Aldehydes................. 8^ XVI. The Free Energies of Activation at -100° (aG*_iqqo), the Arrhenius Activation Energies (Ea), the Ground-State Free Energy Differences (AG), the Entropies of Activation (aS*), the Enthalpies of Activation (aH*), and the Frequency Factors of the a,0-Unsaturated Acyl Fluorides........ 88 vii LIST OP FIGURES i Figure Page | 1. The 100 MHz 1H nmr spectrum of acryloyl fluoride .............................. JO 2. The computer calculated 100 MHz *H nmr spectrum of acryloyl fluoride................. 3^ j ! 3» The decoupled 100 MHz ^H nmr spectrum of acryloyl fluoride ........ 35 4. The computer calculated decoupled 100 MHz !h nmr spectrum of acryloyl fluoride.............. 36 5. The ambient temperature 9^*1 MHz nmr spectrum of acryloyl fluoride in vinyl chloride solution............................ 38 6. Plot of log K against l/T for acryloyl fluoride*.............. ..................... 4 3 7. Experimental and calculated nmr spectra of acryloyl fluoride in vinyl chloride solution........... ^7 j 8. Arrhenius plot for acryloyl fluoride........ 4-8 9» The 100 MHz *H nmr spectrum of crotonyl fluoride............ 50 | 10. Experimental and calculated 1^F nmr i spectra of crotonyl fluoride in vinyl ! chloride solution........ 53 i ! ! 11. Arrhenius plot for crotonyl fluoride....... 5^ I 1 | 12. The 100 MHz H nmr spectrum of cinnamoyl fluoride in carbon tetrachloride solution.... 57 13* Experimental and calculated nmr spectra of cinnamoyl fluoride in 60% vinyl chloride | and ^0% chlorodifluoromethane solution. 6l 14. Arrhenius plot for cinnamoyl fluoride 62 viii Figure Page 15. The 100 MHz H nmr spectra of £-methoxy- cinnamoyl fluoride............. 63 16. Experimental and calculated 1^F nmr spectra of £-methoxycinnamoyl fluoride in ?0$ vinyl chloride and 30?S acetone solution........... 67 17. Arrhenius plot for £-methoxycinnamoyl fluoride..................................... 68 18. The conformations of 1,3-butadiene, acrolein, methyl vinyl ketone, and acryloyl fluoride... 92 19* The charge distributions on the carbon and oxygen atoms of the s-trans and the s-cis conformers of acrolein....................... 95 ABSTRACT Low temperature fluorine nuclear magnetic | resonance spectroscopy has been used to investigate the conformational equilibria of acryloyl fluoride, crotonyl fluoride, cinnamoyl fluoride, and £-methoxycinnamoyl fluoride. Interconversions of the s-trans and the s-cis conformers of these <*,6-unsaturated acyl fluorides were stopped on the nmr time scale at the low temperatures. Rate constants for conversion of the s-trans to the s- cis conformers were determined at each temperature by 19 analyses of the 'F nmr line-shapes. Activation para meters and the free energy differences for the s-trans and the s-cis conformers have been determined. The effect of structural changes on activation parameters i and free energy differences are also discussed. The i magnitude of the values of the various terms in the ; potential function for acryloyl fluoride are also re- i ported. x CHAPTER 1 INTRODUCTION 1-1. Elements of Conformational Analysis. The non-identical arrangements of the atoms in a molecule which are obtainable by rotation about one or more single bonds are called conformers. Thus, confor mational isomers are readily intercoverted at room tem perature by rotation, while configurational isomers are 1 2L not. Conformational analysis has been defined as "an analysis of the physical and chemical properties of a compound in terms of the conformation (or conformations) of the pertinent ground states, transition states, and X I) (in the case of spectra) excited states". Ethane is the simplest hydrocarbon exhibiting conformational isomerism, with an energy barrier of appro ximately 3 kcal/mole for rotation about the carbon-carbon 2 single bond, as first suggested by Kemp and Pitzer, The source of the potential barrier is not clear at the present time, although there has been much speculation on this point. The various proposed theories have been re- 'i 4 5 viewed by Wilson, and more recent by Millen, Dale, 6 7 Lowe, and Pethrick. Barrier heights restricting rotation are small compared with the energies involved in the formation of molecules, and, for this reason, are not easily computed or interpreted theoretically. However, it is generally accepted that the least stable arrangement for ethane occurs when the C-H bonds are eclipsed (la, dihedral angle 0 = 0°), and that the most stable confor- mation is a staggered one with 0 = 60° (lb). The confor mers (la and lb) are illustrated as projections along the G-C bond. H H H H H (la) Elipsed (lb) Staggered The potential energy of ethane molecule is a function of dihedral angle and may be expressed by the equation! V(0) * i V3 (1 + cos 3©) where is the height of the potential barrier. In n-butane, an examination of the potential energy as a function of the dihedral angle about the C2-Cj central bond reveals the existence of two different con- formational energy minima along the rotation coordinate. The most stable arrangement occurs when the two methyl groups are trans to each other. Steric repulsion of the two methyl groups causes the gauche form to be less stable. The two stable conformations were shown to differ in energy by 0.8 kcal/mole by infrared and Raman spectroscopy,^ The barrier for internal rotation was estimated to be 3.4 kcal/mole.10 Since a barrier of about 16-20 kcal/mole between interconvertible isomers is required to permit their separation at room temperature, the two conformations of n-butane interconvert rapidly. In such conformationally dynamic systems, properties of the molecules are directly related to the population weighted properties of each con- former. Factors other than steric repulsion need to be considered in more complex molecules. For example, the preferred conformation of 1,2-dichloroethane is the anti conformation. This is not so much because of the relative ly mild steric interaction of the chlorine atoms, but main ly because of the dipole-dipole repulsion in the gauche conformation.11 In the case of 1-chloropropane and 1- fluoropropane, the gauche forms were found to be more stable than the trans forms. It was rationalized that the CHyhalogen distance falls into the attractive part of the van der Waals curve, and that this attractive force be- tween CH^ and halogen groups is responsible for the sta bilization of the gauche form. ' * ' 1 The conformational characteristics of unsatu rated molecules are also of interest. Several studies of molecules containing either a carbonyl or an olefinic group have been reported. All available evidence indicates that both the C « 0 and the C * CH2 bonds prefer to be eclipsed by a bond from an adjacent carbon. For example, the more stable conformation of acetaldehyde as determined 12 by microwave spectroscopy*1 **’ is the one in which the car bonyl bond is eclipsed by a C-H bond. An analogous situ ation exists for propene (II).1- * In both propionaldehyde (III)1* * ' and methyl acetate,1^ an eclipsed methyl group leads to a lower energy, whereas for 1-butene1^ methyl eclipsing is as favorable as hydrogen (IV) eclipsing. For neopentylethylene, where a bulky t-butyl group has replaced the methyl of 1-butene, the double bond is largely eclipsed by a hydrogen.1^ There has been a great deal of interest in the 1,3-butadiene-type molecules, simply because the butadiene molecule has had an important role in the development of theories of chemical bonding. The length of the C2-C-j o 17 single bond in 1,3-butadiene is 1.46a, much less than the value (1.54a) found in saturated hydrocarbons or dia mond. The reason for this C2-C^ single bond shortening is H CH H CHq 0 H H H (II) Piropene (III) Propionaldehyde H H H ,GH CH H H (IVa) (IVb) 1-Butene still not completely understood. In spite of extensive theoretical effort, there is still a controversy as to whether ir-electron conjugation really produces any obser vable changes in the ground-state properties of the mole- oule.18*19 The existence of more than one stable rotational conformer in 1,3-butadiene- (V) or acrolein- type molecules 20 was recognized many years ago. A variety of experimen tal techniques have been applied in the study of this type H H H H \ / \ / C=C H C=C / \ / / \ H C=C H C— H / \ / H H H— C \ H (Va) (Vb) of isomerism. Electron-diffraction^ and infrared studied on 1,3-butadiene itself gave evidence that the planar s- trans conformer is the dominant (though not necessarily the exclusive) form at room temperature. Microwave studies of 1,1-difluorobutadiene by Beaudet,^ an(j 2-fluorobutadiene^' and isoprene^ by Lide failed to reveal the presence of the s-cis (or skew) conformer. On the other hand, 2,3-di-t- 7 butyl-1,3-butadiene is reported on the basis of uv studies 26 to exist in a non-transoid conformation. Steric inter action of the bulky t-butyl groups may prevent it from adopting either planar form. By infrared and Raman stu dies, chloroprene, 2,3-dichloro-l,3-butadiene and isoprene have been found to exist predominantly in the s-trans form, while hexachlorol,3-butadiene exists in a preferred non- 27 planar conformation, NMR studies of substituted 1,3-butadienes by Bothner-By and coworkers have also pointed to the possibi lity that the stable non-transoid form may adopt a skew 28,29 conformation. Based on the magnitude and sign of the four-bond H-H coupling constants, and some uv data, they deduced that 2-bromo-1,3-butadiene and 2-iodo-l,3-butadiene exist practically completely in the s-trans conformation, while 1,1,3-trichloro-1,3-butadiene (VI, X a Cl, Y - Cl) and 1,1,3-tribromo-1,3-butadiene (VI, X = Br, Y = Br) are (VI) entirely in a skew form. Furthermore, 1,l-dichloro-3- fluoro-1,3-butadiene (VI, X * Cl, F = F) and 1,l-dibromo-3- fluoro-1,3-butadiene (VI, X * Br, Y * F) were also reported to exist as a mixture of the s-trans and the skew con formers. From nmr, microwave, ir, and dipole-moment stu dies, Servis and Roberts concluded that 1,1,4,4-tetra- 30 fluoro-1,3-butadiene exists in the s-trans conformation. One of five-bond F-F coupling constant^J was reported 30 PF to be 35*7 Hz in that molecule. From the relatively small magnitude of the five-bond F-F coupling ( J , ^ Hn = 4,80, and = 11*3 Hz) obtained from an analysis of 19 the complex F nmr spectrum of hexafluoro-l,3-butadiene 31 (VII), Manatt and Bowers suggested that this molecule may not be in a planar form. (VII) 9 The source of the apparent instability of the s-cis form of 1,3-butadiene could be (i) unfavorable arran gement of the pi orbitalB as suggested from a quantum 32 mechanical calculation by Parr and Milliken, and/or (ii) severe steric interference of hydrogen atoms at the and positions. A different explanation for the con formational characteristics of 1,3-butadiene has been 2k offered by Lide. A two-term rotational potential func tion was proposed for 1,3-butadiene-type moleculei V (0) * i V2 (1 - cos 20) + i V3 (1 - cos 3©) The siim of these two terms yields two out-of plane minima (in addition to the s-trans minimum) which approach the s-cis form as V2/V^ is increased. The proposed function accounts for the "predominance of the s-trans form of bu tadiene without the necessity of introducing rather dubious 2k steric interaction". He also suggested that the exis tence of a stable non-planar form seemed at least as pro- 2k bable as the s-cis form. Attempts to determine the dihedral angle depen dence of the potential energy for 1,3-enones have also met with but limited success. In the case of acrolein, studies in the near ultraviolet-^'^ and in the far infrared^ have given values of the torsional frequencies of s-trans acrolein, but no spectrum of any other form has been de- 10 tected. A microwave study was carried out by Cherniak and 36 Costain^ in 1966, They reported the structural parameters of the planar s-trans acrolein, but again failed to find the spectrum of the s-cis conformer* The structure of acrolein was determined more recently by electron diffrac- 17 37 tion, and the results are in accord with the occur rence of the s-trans conformation. It was concluded from a microwave study of trans-crotonaldehyde that this mole- 38 cule also exists only in the s-trans form* Rotational barriers for molecules containing the 1,3-enone moiety have been obtained only in caseB where rotation by 180° produces identical (or nearly identical) 39 forms. From a low temperature nmr study, Anet and Ahmad-' estimated the free energies of activation for the symme trical C-C rotational barriers of js- substituted benzal- dehydes. The reported values were 7.9» 10.8, and 9.2 kcal/ mole for benzaldehyde (Villa), £-N,N-dimethylaminobenzal- dehyde (VUIb), and ])-methoxybenzaldehyde (VIIIc), respec tively. The effect of the substituent group on the rota tional barrier is as might be expected based on the known electron-donating effect of the dimethylamino and methoxy 39 40 substituents. These findings were later confirmed for (VUIb) and (Vllld) and extended to the substituted aceto- phenones (VUIe) and (VUIf). For the latter two com pounds, the free energies of activation are about 2 kcal/ 11 (VIII) Ri r2 r3 AF*t kcal/mole (Villa) H H H 7.9 (VUIb) (ch3)2n D H 10.8 (VIIlc) CH^O H H 9.2 (Vllld) ch3o D H 9.^ (vine) (ch3)2 D CH3 8.5 (vnif) CH30 D CH3 7.3 12 mole lower than for the corresponding aldehydes. A careful dynamic nmr study of the internal rotation in 2-furaldehyde (IX) has been performed by ki Dahlqvist and Forsen. The free energies of activation H H (IXa) (IXb) for the C-C rotation were determined to be 10.93 and 10.35 kcal/mole for (IXa) and (IXb), respectively. In a number of recent papers, studies of rota tional isomerism in a,B-unsaturated acyl halide have been 1 ^ 2 -4 6 l q reported. Low temperature nmr study of perfluoro- 0 (Xa) (Xb) k-3 acrolein (X) was carried out by Brey and Ramey. J The molecule shows an averaged spectrum at room temperature but has two nmr-distinguishable conformers at -105° as a result of restricted rotation about the central C^-Cg bond of the conjugated system. Prom a study of the infrared spectra of txL acryloyl chloride. Katon and Feairheller concluded that two discrete rotational conformers coexist in the vapor and liquid states. The difference in energy of the two con formers could not be determined because of severe band overlap, but a gas phase value of 600 cal/mole was de- termined. For acryloyl fluoride (XI), Koster^2 examined the temperature dependence (+85° to -95°) of the nmr spectra. The averaged three-bond H-F coupling constants F 0 (XIa) (Xlb) were measured at high temperatures, and the rotational equilibrium of acryloyl fluoride was followed by the following equationt 14 3Tave „ D 3Ts-trans M 3Ts-cis HF HF + U - P) JHF where P is the population of the s-trans conformer. By a least squares treatment, the s-trans conformer was calculated to be 800+250 cal/mole more stable than the s- cis form of acryloyl fluoride. On the other hand, Carlson, j Fateley, and Whitkowski^ in an infrared and Raman study of acryloyl fluoride found absorption bands at 978 and 998 cm"1 and assigned them to the s-trans and the s-cis i conformers, respectively. From an investigation of the intensity of these bands over the temperature range +28° to -99°G, they determined the energy difference between the two forms of acryloyl fluoride to be 150+100 cal/mole in carbon disulfide solution and about the same for the neat liquid. A microwave study of acryloyl fluoride was 46 carried out by Keirns and Curl, The ground-state energy j difference between the planar s-trans and the planar s-cis | conformers of acryloyl fluoride was reported to be 90+100 | cal/mole, with the s-trans conformation being more stable. i | 1-2. The Nature of Potential Function Hindering the ! Internal Rotation Around the C-C Single Bond in t Con.iugated Systems. Generally, the potential energy function V(0) can be expressed as the Fourier cosine series,^7»48 V(0) = g1 (1 - cos 0) + g2 (1 - cos 20) + |3 (i - cos 30) (1) where 0 is the angle of internal rotation between the two conjugated double bondsi and that V^, V^t and V3 are one fold, two-fold, and three-fold rotational barriers respec tively* It is generally true that the higher terms (more than three-fold rotational barrier) are small and the series can be terminated after cosine 3®» The cos 2G term can be ascribed to the traditional resonance effect and has minima at 9 = 0° and 9 = 180°. V2 may be identified as the "resonance energy", since it indicates the addi tional stability gained from the maximum % orbitals over lapping of the two double bonds* The cos 39 term is a rotational potential similar to that in ethane or propy lene* The inclusion of term in conjugated -diene or -enone system is readily accounted for if the double bond is regarded as two bent single bonds as proposed by I lq Pauling* 7 The cos 9 term identifies the most stable con former. Obtaining enough data to disentangle V^, V2, and V-j becomes a major problem. The simplest case is benzal- dehyde molecule, where the symmetry requires that V (9) = V(K + 9) and hence VI = V3 = * ° Therefore, the potential function of benzaldehyde takes the simple form* In the case of 1,3-butadiene or acrolein, the rotation around the central C-C single bond gives two conformers with unequal energies* As a result of conjugation one expects Vg to be a major term in the diene or enone rotational potential function. Since it is well known that the s-trans conformers of 1,3-butadiene and acrolein are more stable than their corresponding s-cis (or non- planar) analogs, the inclusion of either or term (or both) becomes necessary* From the results of infrared and ultrasonic relaxation investigations of liquid acrolein, Fateley 47 et al. reported that the values of the three potential function terms were * 610, Vg 8 5 2095» and * = 110 cm*’\ It is difficult to measure the rotational bar riers of 1,3-butadiene and acrolein with good accuracy because that the corresponding s-cis (or non-planar) con formers can not be detected by spectroscopy methods. From 45 46 the result of ir and microwave studies, acryloyl fluo ride seemed to be a suitable tractable molecule for low temperature nmr work. Acryloyl fluoride also provides a resonable model of 1,3-diene or -enone since fluorine and 19 oxygen are of similar size. Low temperature F nmr studies were undertaken to determine the structural and dynamic properties of citp-unsaturated acyl fluorides. It was hoped that the analysis of the temperature dependent nmr spectra of ^.^-unsaturated acyl fluorides would give 17 the values of the rotational harriers and the ground- state free energy differences of the two possible con formers. From these data, we may be able to evaluate the relative magnitude of the various terms in the potential function for the conjugated acyl fluorides. The possible effects of structural changes on activation parameters and equilibrium constant will also be examined. 1-3. Nuclear Magnetic Resonance Spectroscopy and Confor mational Analysis. The application of dynamic nuclear magnetic resonance (DNMR) in the study of stereochemical problems, especially those associated with conformational analysis, 50 has recently been reviewed by Binsch. Rate processes which have been studied include those for (i) hindered rotation in substitued ethanes, amides, thioamides, carba mates, nitrosamines, nitrites, aldehydes and ketones, (ii) inversion of lone electron pairs in derivatives of ammonia, imines, and other elements possessing nonbonded electron pairs (P, S, ... etc.), (iii) ring inversions of homocyclic or heterocyclic compounds, (iv) valence iso merization and intramolecular rearrangements.^0 Using quantum-mechanical treatment, complex nmr spectra can be fully analyzed by computing the energies and probabilities of allowed transitions. In addition to chemical shifts, and coupling constants, nmr spectra are 18 also a function of certain time-dependent phenomena. Whenever a chemical exchange process exists* the chemical shifts and coupling constants are statistically weighted averages of the corresponding values in the exchanging species. The rate constants of the chemical exchange process can be calculated by nmr line-shape analysis.^ A large number of books and review articles have presented detailed accounts of nmr spectroscopy. J A brief dis cussion of the analysis of high resolution nmr spectra and chemical exchange line-shape theory which are relevant to the present study will be given below. In quantum theory, the energy of interaction be tween a magnetic moment ju. and an applied field H, is given by 31 = - H -rt) H*I = - h Iz (*0 where 3-c is the spin Hamiltonian, I is the nuclear spin operator and r is the magnetogyric ratio of the nucleus, V is the frequency of electromagnetic radiation. When there are many nuclei in a system, with nuclear spin-spin interactions, the Hamiltonian takes the form ■ Note that the Planck's constant h is dropped for con venience, so that the energy is expressed in frequency junits. I and I are raising and lowering operator |respectively and are defined by i1 = lX + ily and, the I+ and i” operations on the wave function V' I ,m yield d»») d ± - +1) * n , , I>B±1. while I_ | V^T > = m | > z 1 I ,m r I ,m Since the elements of the product basis (a and 3) are jeigenvectors of Izi. and Eq. (5) shows that the jdiagonal elements of X are generated by the terms con- | taining Iz^ and whereas 'terms containing i + X iI^—I.. generate off-diagonal elements. We have 20 <yniM I y »> ■ * JijD m ^ n (7) where U = 1 if Yfo differs from Yh by an interchange of spin i and j and zero otherwise. The transition proba bility is proportional to the matrix element, 2 P *= <1"n I !x IV.) Prom the values of the elements of the given spin Hamil tonian, the eigenvalues (energies) and the eigenvectors 52 53 (wavefunctions) can be calculated. * Hence, chemical shifts, coupling constants and the absorption intensities (but not the line-shape) can easily be obtained. Quantum Mechanical Line-shape Theory. The Bloch equations are useful for describing exchange effects in simple spectra. They apply to molecules with a single magnetic nucleus and a single line in the nmr spectrum. Two relaxation times, longitudinal relaxation time (T-^) and transverse relaxation time (T^) occur in the Bloch equations, and the line-shape depends on these parameters. However, when a system undergoing exchange is complicated by spin-spin coupling, the description of line-shapes and time-dependent effects requires a more sophisticated quan tum mechanical treatment.^ ^ All these treatments are 21 based on the density-matrix formalism, which is also suited for a refined discussion of Bloch's phenomenological equa tions. The following treatment is adopted directly from 50 Binsch. Consider a system of m identical coupled nuclei with spins 1/2 undergoing exchange between n different magnetic environments. Each environment k shall be cha racterized by the state function The total transverse magnetization G, G = Z P„G (8> It k k where p^ is the population of the spin system in the magnetic environment k, and G^ corresponds to the expec- + tation value of the operator -fir <I”>. Without loss of generality we may work with either I+ or I , and we will choose the minus sign (corresponding to spin flips from cC to & ). For convenience, Planck's constant h will be dropped, so that the energy is again expressed in fre quency units. Thus we have Gk =y< I_> I”^k> (9) Expandingy^ into a complete set of orthonormal stationary spin basis function < f > n - ? < 1 0 > J L 22 Eq. (9) becomes °k c* A i (u) k If one defines a density matrix y for the kth magnetic environment by fji ” Ckjcki (12> Eq. (11) can be written as G. = ¥ I l7.fk. =rtr(l~pk) (13) k i#j i3J0i where l7^ denotes the matrix elements <^| I | ) and tr the trace. In the absence of relaxation and exchange, i c each y obeys the equation of motion df* ,„k k k k k ^ - 3 t y ) W ^ = amrp1 ' t/i + ( ) I ( dpk ) dt 2lri* ° J + 1 dt 'r-olav 1 dt K In the presence of relaxation and exchange, then dpk \ + (4*— ) relax, ax exch. = 2ti[fk,xk] - 4 +1(& } (k^?1 - kkl?k)(1S) where k., is the first-order rate constant of the process by which the system switches from the magnetic environment 1 to the magnetic environment k. Under unsaturated steady- state conditions, the left-harld sides of Eq. (15) vanishes, and one obtains a system of linear equations for the _ k elements y . 23 <6 In a more recent paper, Binsch has developed the DNMR theory in the framework of the composite Liou- ville representation of quantum mechanics rather than in the usual Hilbert space representation. The equation of motion can be written as d?C c c c c c gf- = Mj> * ( -iL + R + E ) f (16) where c denotes the operator of interest is in the com posite Liouville representation. The Liouville operator c c c (L ), relaxation operator (R ), and exchange operator (E ) <6 are all defined in the original paper. Under unsaturated steady-state condition, -g|- * 0, Eq. (16) can be solved. Again, the total complex magnetization G in the xy plane can be obtained from Eq. (13)* Equations (13) and (16) are the master equations for line-shapes in dynamic nuclear magnetic resonance and they can be programmed for computer 56 calculation. in the present study, a computer program, 56 DNMR2, written by Binsch and Kleier"^ was adopted to cal culate theoretical spectra for various different values of exchange rate constant. Comparison of the experimental spectra with computer-calculated spectra permitted the relation of temperatures with exchange rate constants. 57 Thus Arrhenius parameters can easily be obtained. CHAPTER 2 RESULTS 2-1. Syntheses of Substituted q,B-Unsaturated Acyl Fluorides. A series of B-substituted a,B-unsaturated acyl fluorides was required for the determination of the rota tional barriers in equilibration of the s-cis and the s-trans conformers. The acyl fluorides can in general be obtained by treatment of the corresponding acyl chloride with a fluorinating reagent such as antimony trifluoride, eg hydrogen fluoride, and other metal fluorides. The SbF-a RC0C1 ---------► RCOF preparation of fluorinated compounds by treatment of organic halogen compounds with antimony fluoride is one of the oldest and still one of the most important laboratory methods for the production of organic fluorine derivatives. The great advantage of using antimony trifluoride is that the reaction can be carried out in a glass apparatus, and very often takes place at atmospheric pressure. Acryloyl fluoride was obtained by refluxing acryloyl chloride with antimony trifluoride under a nitro- 59 1 gen atmosphere for six hours. , The 100 MHz H nmr spec- 2b 25 trum of acryloyl fluoride is a complex ABCX system. From H H H H > = / sbg3 > NW ' / \ " * / \ H C0C1 H COF nmr and ir spectral analyses, no impurity could be de tected. When crotonyl chloride and antimony trifluoride were heated together under a nitrogen atmosphere, a mixture 1 19 of products was obtained. From H nmr and F nmr spectral data, it was clear that two isomers, crotonyl fluoride and isocrotonyl fluoride, were obtained in a ratio of 80*20. 19 The F chemical shift of crotonyl fluoride, was found to be 24.6 ppm downfield from the internal standard, fluoro- 3 trichloromethane, with = 7.84 Hz* for isocrotonyl 19 fluoride, the F chemical shift was 43.66 ppm downfield 3 from fluorotrichloromethane, with JHJ1 = 9.05 Hz, and ^jF,CH3 “ Hz. Apparently, antimony trifluoride, a weak acid, had catalyzed the double bond isomerization of the olefin during the course of the reaction and thereby led to the formation of two products. Because of the difficulty in separating crotonyl fluoride from isocrotonyl fluoride by conventional sepa- 26 ration methods (fractional distillation, vpc, etc.) other methods of preparing crotonyl fluoride were pursued. A neutral fluorinating reagent which would not catalyze the double bond isomerization was desirable for carrying out the reaction. Benzoyl fluoride^ is a known fluorinating reagent which can be used to prepare low molecular weight 6l acyl fluorides without isomerization. Treatment of ben zoyl chloride with hydrogen fluoride has afforded benzoyl fluoride.^0 The attempted reaction between crotonyl chlo ride and benzoyl fluoride gave none of the desired product. However, by treating crotonic acid with benzoyl fluoride in the presence of sodium fluoride, the desired crotonyl fluoride^* was obtained in 6596 yield. The 100 MHz nmr OH3 H OH H C=C + * COF — ^- aF > C=C + *C00H H ^ C00H H ^ COF spectrum of crotonyl fluoride exhibits the ABM3 part of an ABM3X system consisting of three multiplets centered at 1.96, 5*77* and 7*15 ppm downfield from TMS (relative areas 3*1*1)* In an attempt to prepare isocrotonyl fluoride, a different reaction sequence was investigated. The acid catalyzed bromination of 2-butanone was used to prepare 27 62 1 . 1,3-dibromo-2-butanone. The H nmr spectrum 1,3-dibromo- -2-butanone (C^BrCOCHBrCH-j) has an interesting feature. The methylene protons at the 1-position were observed to be magnetically nonequivalent, with axspin-spin coupling cons- 2 gem tant of * 12.3 Hz. It is well recognized that the presence of an asymmetric center in a molecule can cause otherwise identical atoms to become magnetically non- 61 equivalent. Under basic reaction conditions, a,a1 -dihalogeno- ketones undergo Favorsky rearrangement to give the a,B - 64 unsaturated acids. Excess potassium bicarbonate was reacted with 1,3-dibromo-2-butanone in aqueous medium at room temperature for three hours.^ The crude product which was obtained by extracting the acidified solution with ether and evaporation of the ether was analyzed by nmr before further purification. The absence of absorption at 1.0 ppm charataristics of the methyl signal of crotonic acid indicated that the product consisted solely of iso- crotonic acid. This observation apparently contradicts the the presently accepted theory that the Favorsky rearrange ment of an a,a'-dihalogeno ketone gives both the cis and 64 trans geometric isomers of the a,$-unsaturated acid. The synthesis of isocrotonic acid provides an example of a stereospecific Favorsky rearrangement. Unfortunately, the reaction of isocrotonic acid and benzoyl fluoride in the presence of sodium fluoride again gave a mixture of crotonyl fluoride and isocrotonyl fluoride. NMR analysis of the product mixture revealed that the ratio of cis- to trans-acyl fluoride was 79 to 21. No attempt was made at this point to isolate iso crotonyl fluoride. Long range spin-spin coupling was observed in isocrotonyl fluoride. The peaks of the methyl protons were split by the fluorine atom with ^ J ** 1.50 GH^tF Hz through five bonds, but there was no observable coupling between the methyl protons and the fluorine atom in cro tonyl fluoride. This observations would be consistent with the occurrence of a "through space" coupling mechanism for the CHyF coupling. Cinnamoyl fluoride was prepared by treatment of cinnamoyl chloride with hydrogen fluoride in a polyethylene 61 bottle. Carbon tetrachloride was found to be a good solvent for extracting cinnamoyl fluoride from the reaction mixture. NMR spectrum indicated that no impurity was pre sent. The coupling constant of the two olefinic protons was found to be 15*9 Hz, suggesting that the two olefinic H HF C0C1 H COF protons are trans to each other. Treatment of ja-methoxycinnamic acid with thionyl chloride gave p-raethoxyeinnamoyl chloride in good yield. Employing a procedure similar to the one used in preparing cinnamoyl fluoride,^1 ^-methoxycinnamoyl fluoride was made by treating j>-methoxycinnamoyl chloride in carbon tetra chloride solution with hydrogen fluoride. Again nmr :c=c; C0C1 ch3°\ =0, OF spectrum indicated that no detectable amount of impurity was present. The coupling constant of the two olefinic protons was found to be 16.0 Hz, suggesting that these protons are trans to each other. 2-2. Analysis of NMR Spectra of Acryloyl Fluoride. Analysis of the 100 MHz nmr spectrum of acryloyl fluoride was performed by the use of computer program LA0C00N III.^ The 1H spectrum is best described 52 as resulting from the ABC part of an ABCX spin system (Fig. 1). The initial trial parameters were taken from i f 2 the values given by Koster. The program was then used Figure 1. The 100 MHz nmr spectrum of acryloyl fluoride. o to iterate to least-squares-error fit of the measured line positions from the observed 100 MHz 1H spectrum. The nmr parameters so obtained are shown in Table I. The calcu lated ABC part of the ABCX spectrum of acryloyl fluoride, as shown in Table II and Fig. 2, was in good agreement with the observed 100 MHz spectrum. Fluorine decoupling at 9^.1 MHz resulted in the simplification of the 1H spectrum, as shown in Fig. 3» Analysis of the ABC spectrum obtained by spin-decoupling using the LA0C00N III program yielded the same proton-proton coupling constants as in the ori ginal acryloyl fluoride system (see Fig. 4 and Table I). This provided confirmation of the choice of parameters (see Table I). The parameters so obtained were used to 19 calculate the 9^»1 MHz F nmr spectrum of acryloyl fluo ride, using the LA0C00N III program. The calculated spec trum was again in good agreement with the observed spec trum. The parameters we have obtained for the acryloyl fluoride system are only slightly different from what Koster has reported. The small discrepancy may be due to the difference in solvents employed (fluorotrichloro- methane was used in Koster's sample). 19 The 9^.1 MHz F nmr spectrum of a vinyl chloride solution of acryloyl fluoride at ambient temperature is a symmetrical multiplet, as illustrated in Fig. 5» centered at a chemical shift 23*9 ppm downfield from internal fluo- 32 Table I NMR Parameters for Acryloyl Fluoride COFi a Our values (Hz) Koster*s values1 3 (Hz) v2 “ v3 = 9,3 - 0tl 5.04 v - v = 38.4 + 0.1 5 4 23.16 J12 = ± O*1 8.51 J13 = 3»o + o.i 2.92 J^zj. = -0*3 + o.i -0.37 J23 = 10.5 it 0«l 10.56 Jgij, “ 17*3 ± o.i 17.3** J34 = 0.8 + 0.1 0.82 a. From 100 MHz *H spectrum, in GCl^ solution. b. From 60 MHz H spectrum, in 50fo CFCl^ solution. 33 Table II The Calculated 100 MHz *H NMR Spectrum of Acryloyl Fluoride in Carbon Tetrachloride Solution Line Experimental Calculated - Intensity Error frequency frequency 16 123.600 123.801 0.056 -0.201 30 126.600 126.792 0.118 -0.192 53 132.500 132.398 0.613 0.102 40 132.500 132.445 0.673 0.055 15 135.500 135.5^7 0.661 -0.047 31 137.100 137.210 0.566 -0.110 34 146.000 145.990 0.830 0.010 22 147.600 147.653 1.032 -0.053 4 151.000 150.756 1.367 0.244 8 151.000 150.802 1.431 0.198 35 156.400 156.408 0.442 —0.008 21 159.400 159.400 0.211 0.000 55 169.322 0.017 29 174.200 174.134 0.581 0.066 48 174.191 0.057 13 177.300 177.294 0.650 0.006 42 179.700 179.493 2.422 0.207 54 179.700 179.740 2.370 -0.040 6 184.304 1.304 28 184.600 184.454 2.182 0.146 52 186.000 185.938 2.270 0.062 3 187.600 187.557 1.266 0.043 33 193.200 193.332 1.032 -0.132 27 196.100 196.201 0.185 -0.101 19 201.100 201.146 1.022 -0.046 5 203.400 203.502 0.265 -0.102 2 211.300 211.409 0.368 -0.109 1 505.500 505.500 1.000 0.000 50 Hz Figure 2. The computer calculated 100 MHz nmr spectrum of acryloyl fluoride. VJ ■ p - Figure 3 acryloyl 50 Hz 19 • The F decoupled 100 MHz nmr spectrum of fluoride. K 50 Hz M 19 Figure 4. The computer calculated F decoupled 100 MHz nmr spectrum of acryloyl fluoride. On 37 Table III 19 The Calculated F Decoupled 100 MHz NMR Sepctrum of Acryloyl Fluoride Line Experimental Calculated Intensi frequency frequency 16 125.4 125.470 0.084 31 125.4 125.470 0.084 4-0 132.3 132.436 0.645 53 132.3 132.436 0.645 15 136.3 136.269 0.616 30 136.3 136.269 0.616 22 146.7 146.955 0.945 35 146.7 146.955 0.945 4 150.8 150.788 1.397 8 150.8 150.788 1.397 21 157.4 157.754 0.313 34 157.4 157.754 0.313 29 175.8 175.724 0.619 13 175.8 175.724 0,619 43 182.5 182.114 2.321 28 182.5 182.114 2.321 55 182.5 182.690 2.350 52 182.5 182.690 2.350 6 186.2 185.947 1.284 3 186.2 185.947 1.284 42 192.7 192.913 0.075 27 192.7 192.913 0.075 33 197.0 197.209 1.027 19 197.0 197.209 1.027 5 207.1 207.432 0.319 2 207.1 207.432 0.319 38 20 Hz Figure 5. The ambient temperature 9^*1 MHz nmr spectrum of acryloyl fluoride in vinyl chloride solution. 3 9 ! ! rotrichlororaethane. Profound changes occur in the *^F nmr | 1 spectrum of acryloyl fluoride as a result of cooling a | vinyl chloride solution of the sample* As the temperature ; is progressively lowered, the spectrum becomes increasingly j broadened, and a wide hump is observed at -135°» When the temperature is lowered to -l6l°, two distinct *^F nmr absorptions, with a population ratio of 0*76 to 0.2^ and well separated by 2075 Hz were observed. A coupling cons tant of 19*2 Hz associated with the more populated absorp tion (at higher field) was measured at this low tempera ture. These two nmr absorptions at very low temperature are attributed to the s-trans and the s-cis conformations of acryloyl fluoride. The most reasonable explanation for the observed temperature dependent nmr spectra is that i j interconversion of the two conformers is rapid at room temperature on the nmr time scale but slow at very low ; temperature. We have assigned the more intense signal (at ! higher field) to the s-trans conformer and the less in- ! tense signal (at lower field) to the s-ci3 conformer.* I The coupling constant of 19*2 Hz was assigned to be the I I three-bond H-F coupling of the s-trans conformer of aery- | | loyl fluoride. i ! It is difficult to obtain the three-bond H-f c o u - i " 5 s-cis | pling constant of the s-cis conformer, JHF » from the * See Chapter 3» ' low-temperature nmr spectrum, due to the extensive | coupling between the fluorine atom and the other three vinylic protons in this conformer. However, assuming AG is invariant with temperature,* and knowing the value of ^JHFtranS* ^HF°^S was calculated to be -7.50 Hz from the ! following two equationsi* P P * AG = -RT. In— ^- = -RT« In — (1) 1 pt ^ pt 3 ^;e = p 0- 3 j s ; o 1 s + P t13 ^;t r a n s m where Pc and P^ are the populations at the lower tempera ture T1# and Pc' and P^' are the populations at ambient temperature Tg for the s-cis and the s-trans conformers, respectively. The chemical shifts, coupling constants, popu- ; lations, and the energy difference for acryloyl fluoride | at -l6l° in vinyl chloride solution are given in Table IV. * For a chemical transformation accompanied by a change in free energy, AG, it can be shown from thermodynamics that I 3AG \ - AS (^ ~ )P ' A S * If AS = 0, then aG is independent of temperature as implied in Eq. (1). Table IV Chemical Shifts, Coupling Constants, Populations, and AG Value for Equilibrium of Acryloyl Fluoride at -161.5° in Vinyl Chloride Solution Population v a (Hz) s-trans O.76 + 0.03 -1548 + 5 (-16.4T ppm) 19.2 + 0.5 s-cis 0.24 + 0.03 -3623 ± 5 (-38.5 ppm) -7.5 ± 0.9b AG = 253 ± 67 cal/mole the s-trans and the s- fluoride. for the equilibrium between cis conformers of acryloyl a. All chemical shifts are downfield from internal fluorotrichloromethane. b. Calculated values, see text. From the area measurements of the *^F nmr signal below the coalescence temperature, the population ratios ; of the s-cis to the s-trans conformers of acryloyl fluoride | as a function of temperature are given in Table V and the ! log K vs. l/T plot is shown in Fig. 6. The enthalpy ! difference between the two forms of acryloyl fluoride was I calculated to be 229 ± 115 cal/mole from the slope of the 1 1 I straight line. However, due to the narrow temperature ! range, the entropy difference can not be accurately obtained from the extrapolation of the straight line. The uncertainty on the &H value obtained from the cut—and- kz \ I I Table V ! The Ratios of Populations of the s-cis to the s-trans Conformers of Acryloyl Fluoride As a Function of Temperature Temperature °C s-trans (%) s-cis (%) K -161 75.8 2*4-.2 0.319 -15^ 7^.7 25.3 0.339 -151 7^.3 25.7 0.3*4-6 -1^8 73.8 26,2 0.355 weigh area measurements method is admittedly large. | In order to calculate the C-C rotational barrier of acryloyl fluoride, rate constants of isomerization were determined at various temperatures between extremes of fastj i and slow exchange. These rates could be obtained from the line-shape a n a l y s i s ^ of the computer calculated and experimental nmr spectra at the various temperatures. i Theoretical spectra for this intramolecular exchange process were calculated from a computer program, DNMR2, written by Binsch and Kleier.^ The program is based on the density-matrix f&rmalism in quantum mechanical treat- -0.45 -0.46 -0.47 -0.48 -0.49 8.00 8.20 8.40 8.60 8.80 J_ X ,03 ----- T Figure 6. Plot of log K against l/T for acryloyl fluoride. ment and simulates complex exchange-broadened nmr spectra from chemical shifts, coupling constants, relaxation times, rate constants, populations and certain scaling parameters (see Experimental for details). A series of theoretical spectra was calculated by varying the rate constant for s-trans to B-cis conversion. Visual compa rison of the experimental spectra with computer-calculated spectra permitted the correlation of temperatures with exchange rate constants. Temperatures were determined by measuring the potentials from a copper-constantan thermo couple and were estimated to be accurate to + 2°C (upper limit). The kinetic data, determined from the line-shape analysis,^ are given in Table VI. Using Professor D. P. DeTar*s ACTENG computer program,^ the Arrhenius parameters and the values of £ £ + AS ,AH , and aG for C-C rotation in the s-trans conformer of acryloyl fluoride were calculated by least squares method to be as followsi A = (1.3 ± 0.28 ) x 1013 Ea - 6.06 + 0.07 kcal/mole AS4 1 = 0.93 ± 0.^1 e.u. AH = 5.7 8 + 0.07 kcal/mole AGC* = AH* - Tc AS* = 5.66 +0.07 kcal/mole Tq (coalescence temp.) - -1^1°C ^5 Table VI Kinetic Data for Acryloyl Fluoride in Vinyl Chloride Solution 0 Temp. C *ta b -3 Rate const, x 10 ^, sec“l -154.0 0.?26 0.110 -148.0 0.720 0*310 -143*0 0.727 0.780 -135*5 0.717 3*10 -130.5 0.717 6.00 -127*0 0.705 12.0 -122*5 0.700 20.0 -113*0 0.689 77.0 -108.0 0.684 116. a. The fractional population of the s-trans confor* mer. b. Rate constants for conversion of the s-trans conformer to the s-cis conformer. The reported values were estimated to be accurate to + The uncertainties are the actual standard de- ; viation from the least square evaluation based on tempera- | ture uncertainties of + 2° and rate constant uncertainties of ± 5 percent. Representative 1^F nmr spectra and an Arrhenius plot of In k vs. l/T of acryloyl fluoride in vinyl chloride solution are shown in Fig. 7 and Fig. 8, respectively. 2-3. Analysis of NMR Spectra of Crotonyl Fluoride. in carbon tetrachloride solution is shown in Fig. 9. The spectrum consists of three multiplets centered at I.96, 5«77» and 7*13 ppm downfield from TMS (relative area 3*1«1). The observed splitting pattern is consistent with that expected for an ABM3X spin system. Upon spin de coupling from the fluorine atom, the spectrum simplified to give a pattern characteristic of an ABM 3 system. The spectrum can be analyzed readily by first-order analysis. ^ The following parameters were obtained for crotonyl fluo ride in carbon tetrachloride solution. The 100 MHz 1H nmr spectrum of crotonyl fluoride C V V = 5*77 ppm (fi) J12 * 7.30 Hz = 7.13 ppm (6) J23 * 15*68 Hz CH3 “ 1,96 ppm (S 1 J2-0H3 “ 1,72 Hz J 3-CH3 * 6.90 Hz 47 -I08‘ If 2.5° -127 -143 -154 nmJ u mf c m u d^ K —2000 Hz— M k= 1 .1 6 x I05 sec“ i 7.70 x I04 3.10 x I03 7.80 x 1 0 1 .1 0 X I02 Figure 7* Experimental (left) and calculated (right) !9p nmr spectra of acryloyl fluoride in vinyl chloride solution. Figure 8. Arrhenius plot for acryloyl fluoride. 11.00 9.00 7.00 10VT 5.00 700 6.50 8.00 7.50 m i L j l iVvtyV 6 6,PPm 4 Figure 9• The 100 MHz nmr spectrum of crotonyl fluoride. 51 19 The 9^.1 MHz F nmr spectrum of crotonyl fluo ride in vinyl chloride solution at ambient temperature consists of a doublet 23.93 ppm downfield from internal standard fluorotrichloromethane, with a coupling constant of 7.82 Hz for the three-bond H-F coupling. When the temperature was lowered to -136°, two distinct resonance absorptions separated by 1765 Hz appeared in the spectrum. As the temperature was gradually increased, these two absorptions quickly broadened into two wide humps, which o coalesced at about -107 • The absorption peak sharpened as the temperature increased, and finally became a sharp doublet at ambient temperature. The chemical shifts, coupling constants, populations, and free energy difference for the s-trans and the s-cis conformers of crotonyl fluo ride at -1360 in vinyl chloride solution are given in Table VII. Rates of conversion of the s-trans to the s-cis conformer of crotonyl fluoride were determined at tempera tures from -79° to -129°• The resultant rate constants are collected in Table VIII. The Arrhenius parameters and the values of As , AH , and AG for conversion of the s-trans to the s-cis conformers of crotonyl fluoride were calculated by the least squares method to be as follows» 52 Table VII Chemical Shifts, Coupling Constants, Populations, and AG Value for Crotonyl Fluoride at -1360 in Vinyl Chloride Solution Population v a (Hz) 3Jhf (Hz) s-trans 0.69 + 0.03 -1556 + 5 (-16,5 ppm) 21.5 ± 0.5 s-cis 0.31 + 0.03 -3321 t 5 (-35*3 ppm) -11.6 + 0.9b AG = 215 + 63 cal/mole for the equilibrium between the s-trans and the s-cis conformers. a. All chemical shifts are downfield from internal fluorotrichloromethane reference. b. Calculated value, see text. A = (0.85 ± 0.1) x 1013 E_ * * 6.99 + 0.05 kcal/mole & T r i AS* = 0.25 + 0.3 e.u. AH = 6.65 +0.05 kcal/mole AGC* = A H* - Tc AS* = 6.69 kcal/mole Tc (coalescence temp.) = -107°C 19 Representative F nmr spectra at the various temperatures and the Arrhenius plot of In k vs. l/T are shown in Figures 10 and 11, respectively. k =8.60 x I04 sec-1 -126 m i wwmmw r t l N P -129 k-2000 Hz— s i 2.70 x I04 1.02 x I03 J v_ 6.00 x I03 3.60 x I02 2.20 x I02 Figure 10, Experimental (left) and calculated (right) nmr spectra of crotonyl fluoride in vinyl chloride solution. Figure 11. Arrhenius plot for crotonyl fluoride. In k 12.00 10.00 8.00 6.00 10/ T 6.20 6.80 56 Table VIII Kinetic Data for Crotonyl Fluoride in Vinyl Chloride Solution Temp. °C a pt Rate const.1 3 x 10"3, sec”1 -129.0 0.680 .22 -126.0 0.676 .36 -119.5 0.669 1.02 -115.5 0.666 1.90 -110.5 0.660 3.10 -IO6.5 0.657 6.00 -102.0 0.653 10.2 -93.5 0.646 27.0 -89.0 0.643 44.0 -82.0 0.638 86.0 -79.0 0.636 120. a and b, see footnote on Table VI. 2-4. Analysis of NMR Spectra of Cinnamoyl Fluoride. i The 100 MHz H nmr spectrum of cinnamoyl fluoride in carbon tetrachloride solution is shown in Fig. 12. The spectrum consists of three multiplets centered at 6.25, 7.38, and 7.70 ppm downfield from TMS (relative area 115* 1)• The spectrum can be analyzed easily by first- order analysis. The following parameters were obtained 1 50 Hz K - Figure 12. The 100 MHz nmr spectrum of cinnamoyl fluoride in carbon tetrachloride solution. V J I -o COF1 \>2 = 6.25 ppm U) \>j * 7.70 ppm («) varomatic - ^ = 7.37 ppm (« ) J12 = J23 = 15,9 Hz 19 The 9^*1 MHz P nmr spectrum of cinnamoyl fluo ride in 60% vinyl chloride and *K)% chlorodifluoromethane solution at ambient temperature consists of a doublet 2^.63 ppm downfield from internal fluorotrichloromethane, with a coupling constant of 6.87 Hz for the three-bond H-F coupl ing. As the temperature is progressively lowered, the spectrum becomes increasingly broadened, and a wide hump !is observed at -92°. When the temperature is lowered to j -130°» the spectrum appears as two distinct absorptions with a population ratio of O.583 to 0.^17, well separated ;by 2010 Hz. The observed three-bond H-F coupling constants for the s-trans and the s-cis conformers are 19.^ and 1-5.7 Hz, respectively. The chemical shifts, coupling cons tants, populations, and free energy difference for the s- i trans and the s-cis conformers at -130° are given in Table | IX. It is noted that the calculated ^JhF°*S value (-7.8 Hz) |differs from the experimental value (-5.7 Hz) by 2.1 Hz. i Rates for conversion of the s-trans conformer of cinnamoyl fluoride to the s-cis analog were determined by Table IX 59 Chemical Shifts, Coupling Constants, Populations, and AG Value for Cinnamoyl Fluoride at -130.5° in 60% Vinyl Chloride and 40% Chlorodifluoromethane Solution Population va (Hz)' 3jhf ( H z > s-trans 0.58 + 0.03 -1400 + 5 (-14.9 ppm) 19.4 + 0.5 s-cis 0.42 + 0.03 -3410 + 5 (-36,2 ppm) -7.8 + 0.9b AG = 95 ± 58 cal/mole for the equilibrium between the s-trans and the s-cis conformers. a. All Chemical shifts are downfield from internal fluorotrichloromethane reference. b. Calculated value, see text. Iline-shape analysis^ at temperatures between -45° and 1-130.5°* The kinetic data so obtained are shown in Table X« The Arrhenius parameters and AS*,AH*, and aG* ;of activation^? for cinnamoyl fluoride were determined by |least squares analysis to be as followsi A = (0.68 + 0.11) x 10*3 Ea = 7*34 + 0.0? kcal/mole ! AS* = -0.88 + 0.34 e.u. AH* = 6,98 + 0.0? kcal/mole i ^ AGC* = AH* - Tc AS* * 7.14 + 0.07 kcal/mole | Tc (coalescence tempe.) = -92°C 60 Table X Kinetic Data for Cinnamoyl Fluoride in 40# CHFC12 and 60# Vinyl Chloride Temp. °C a pt Rate const.*5 x 10”3, sec"1 -121.5 0.610 0.180 -112.0 0.600 0.750 -108.0 0.598 1.40 -98.0 0.593 4.80 -86.0 0.586 16.0 -78.0 0.583 40.0 -72.0 0.581 7 8.0 -62.0 0.577 156.0 -57.0 0.575 246.0 -45.0 0.572 660.0 a and b, see footnote on Table VI. Representative ^F nmr spectra and the Arrhenius plot for cinnamoyl fluoride are shown in Figures 13 and 14. 2-5* Analysis of NMR Spectra of E-Methoxvcinnamoyl Fluoride. 1 The 100 MHz ■ LH nmr spectrum of £-methoxycinnamoyl fluoride in carbon tetrachloride is shown in Fig. 15, The spectrum consists of one singlet at 3.80 ppm (3H) and three 61 -57< fMN -72 -112 -121.5° wamAfrmiii < Him mi' I mi** -130.5° k = 2.46 x I05 sec-1 1.60 x I04 4.00 x 101 7.50 x I02 1.80 x I02 K — 2000 Hz—3 1 Figure 13. Experimental (left) and calculated (right) nmr spectra of cinnamoyl fluoride in 60# vinyl chloride and 40# chlorodifluoro- methane solution. 14.0 13.0 12.0 1 0 . 0 9.0 8.0 c 7.0 6.0 5.0 4.0 3.0 4.00 5.00 6.00 7.00 I03/ T Figure 1*J>. Arrhenius plot for cinnamoyl fluoride. ■20 0 Hz Figure 15* The 100 MHz nmr spectrum of £-methoxy- cinnamoyl fluoride. o\ u> 6b multiplets centered at 6.05» 7*15» and 7.68 ppm downfield from TMS (relative area li4il), with ^J12 * 6,30 Hz and •^22 “ 16.0 Hz. The 9^*1 MHz *^F nmr spectrum of £-metho- :c=c; COF xycinnamoyl fluoride in 30# acetone and 70% vinyl chloride solution is a doublet 2^.03 ppm downfield from fluorotri- chloromethane, with = 6.8 Hz. The spectrum broadens progressively as the temperature is lowered and finally splits into two absorptions of unequal intensity at -142°. These two absorptions were again ascribed to the s-cis and the s-trans single bond rotational conformers. The chemi cal shifts, coupling constants, populations, and free energy difference of the s-trans and the s-cis conformers of jj-methoxycinnamoyl fluoride at -142° in 70# vinyl chloride and 30# acetone solution are given in Table XI. Rates of conversion from the s-trans to the s-cis conformers of p-methoxycinnamoyl fluoride between -^3° and -12^,5° were determined by line-shape analysis*^ of the temperature-dependent ^F spectra. The resul tant kinetic data are presented in Table XII. Table XI Chemical Shifts, Coupling Constants, Populations, and AG Value for p-Methoxycinnamoyl Fluoride at -142° in 70% Vinyl Chloride and 30?S Acetone Solution Population va (Hz) 3Jhf (Hz) s-trans 0.61 + 0.03 -1376 ± 5 (-14.6 ppm) 20.4 + 0.5 s-cis 0.39 ± 0.03 -3407 + 5 (-36.2 ppm) -9.6 + 0.9b AG = 112 + the s-trans 54 cal/mole for the equilibri and the s-cis conformers. urn between a. All chemical shifts are downfield from internal fluorotrichloromethane f*eference. b. Calculated value, see text. i 4 " The Arrhenius parameters and AS , aH , aG of ! activation^ for j>-methoxycinnamoyl fluoride were deter mined by least squares analysis to be as follows 1 I A = (4.81 + 1.04) x 1013 Ea = 8,21 +0.08 kcal/mole AS* = 2.98 + 0.43 e.u. AH* = 7.84 + 0.08 kcal/mole | AGC+ = AH* - Tc AS* = 7.29 ± 0.08 kcal/mole Tc (coalescence temp.) - -87°C 66 Table XII Kinetic Data for j>-Methoxycinnamoyl Fluoride in 30% Acetone and ?0% Vinyl Chloride Solution Temp. °C *ta Rate const.1 3 x 10“-*, sec"1 -124.5 0.595 0.045 -118.5 0.615 0.135 -110.5 0.586 0.400 -83.5 0.573 16.0 -75.0 0.571 44.0 -59.5 0.566 175. -51.0 0.593 390. -43.0 0.590 850. a and b{ , see footnote on Table VI. Several of the fluorine nmr spectra are shown in Fig. 16 and the Arrhenius plot is reproduced in Fig. 17. 6? -59.5° -83.5° -110.5° k = 8.50 x I05 sec-1 1.75 x I05 4.40 x I04 1.60 x I04 4.00 x I02 K 2000 Hz — H Figure 16. Experimental (left) and calculated (right) 19f nmr spectra of p-methoxycinnamoyl fluoride in vinyl chloride and 30% acetone solution. 14.0 13.0 12.0 1 0 . 0 9.0 8.0 - 7.0 6.0 5.0 - 4.0 3.0 4.00 5.00 6.00 7.00 I03 /T Figure 17. Arrhenius plot for j>-methoxycinnamoyl fluoride. CN CD CHAPTER 3 DISCUSSION k6 From a microwave study, acryloyl fluoride was reported to exist in equilibrium between the s-trans and the s-cis conformations in the gas phase. The ground-state energy difference between the two forms was reported to be 90 + 100 cal/mole, with the s-trans thought to be the more stable conformer. The moments of inertia were reported to be virtually identical for the two forms (3*26 D and 3*21 D for the s-trans and the s-cis conformers respectively), consequently there is some uncertainty whether the spectra have been correctly assigned to s-trans and s-cis.^ The magnitude of the H-F coupling constants from the low tem perature nmr spectra of acryloyl fluoride provides the information as to which conformer is the more stable one. Two distinct ^ F nmr signals well separated by 2075 Hz were observed for acryloyl fluoride at -l6l°. A three-bond H-F coupling constant of + 19.2 Hz was associated with the more intense signal (at higher field), and a three-bond H-F coupling constant of ^ 7.5 Hz was calculated for the less intense signal (at lower field). The sign of the two coupling constants are opposite to each other. 69 It is well known that the vicinal H-H coupling constants show a dependence on dihedral angle described by 6 7 the modified Karplus aquation* ( J = A + B cos 0 + C cos 20 with A = 4.22, B = -0.5, C = 4.5 Hz (1) Equation (1) predicts that the coupling constants for the two vicinal protons trans to each other, J is slight- " nil ly greater than the corresponding cis coupling constants, ci.8 « The Karplus equation has been successfully applied to the ethane- and the allylic- type systems. In the con jugated systems, the three-bond H-H coupling across a 3 s*tr3ns single bond was reported to be = +10.41 Hz in 1,3-butadiene (Va), which is in the s-trans conformation. While in the near s-cis butadiene-type compound (XII) the three-bond H-H coupling, ^J^CiS is | 6.461 Hz.^ 68 \ / 3 C = < H / x h; h V ^ ° CH. (Va) (XII) 71 From the observations reported by Williamson,^0 it was deduced that vicinal H-F coupling constants also show an angular dependence similar to those of H-H coupling constants. If we further assume that the modified Karplus equation holds for the three-bond H-F coupling in conju gated system, we would expect that the three-bond H-F coupling across a single bond in the s-trans conformation should be greater than the corresponding s-cis analog. This is indeed observed experimentally. In 2-fluorobuta- diene (XIII), which is known to be in the s-trans confor mation,2^ the three-bond H-F coupling across C-C single 3 s-trans 71 bond, F is + 25.15. While in the near s-cis 69 butadiene-type compound (XIV), the three-bond H-F coupling, , is | 7 I Hs.* \ /3 C— G H / V / H (XIII) CH3 OCCH3 (XIV) ♦No vigorous attempt has been made in the determination of the sign of this coupling, therefore its sign was not certain. 72 From the above arguments, we assigned the signal 3 exhibiting JHF = + 19*2 Hz to the s-trans conformation, 3 and the signal with JHF “ ? 7*5 Hz to the s-cis confor mation. Since the coupling constant of + 19*2 Hz is asso ciated with the more intense signal in the low temperature nrar spectrum, the s-trans conformer of acryloyl fluo ride must be more stable than the s-cis analog. Our nmr results confirm the tentative assignment based on micro wave studies that the planar s-trans conformer is indeed more stable than the planar s-cis conformer of acryloyl fluoride.^ Our values of the three-bond H-F coupling cons tants of the a,0-unsaturated acryl fluorides, together with the values reported in the literature for the substituted 1,3-butadienes, are shown in Table XII. The values of the three-bond H-F coupling constants across the single bond for the s-trans conformation of the conjugated acyl fluo rides are smaller than the corresponding coupling constant in 2-fluoro-l,3-butadiene. The decrease in magnitude of the values of the three-bond H-F coupling constants in 1,1- dibromo-3-fluoro-l,3-butadiene and 1,l-dichloro-3-fluoro- 1»3-butadiene may indicate that these two compounds exist in a mixture of s-trans and skew conformations. 6q Bothner-By and Moser 7 have reported the angular dependence of the three-bond H-F coupling constants between 73 2 sp carbon atoms to be J = A + B cos 9 + C cos 29 (2) with A = 8.3, B = -0.9» and C = ?.? Equation (2) predicts that ^Jjjp = 7*0 Hz for the s-cis conformation and = 25 Hz for the s-trans conformation of the conjugated diene. The predicted coupling constants are not in good agreement with our values for the conju gated acyl fluorides (see Table XII). The main discre pancy is that they reported a positive three-bond H-F coupling constants, ^JHpC*S = +7»°6 Hz, in the near s-cis compound (XIV), while our results indicated that the three- ! bond H-F coupling for the s-cis conformers of acryloyl fluoride should be a negative value, with ^j|[pC*s = -7*5 Hz i for acryloyl fluoride. Since no vigorous attempt had been | made in the determination of the sign of this three-bond j H-F coupling in compound (XIV), the reported sign by I 69 ; Bothner-By and Moser should be reinvestigated. Recently, Jones and Ladd^2 reported that the j trans and the gauche three-bond H-F coupling constants were | found to have opposite sign. For example, in fluoroacetyl | fluoride, ^jjjptrans was found to be 31.6 Hz, while 3jj=|uche has the value of -6.5 Hz. From Table XIII, the averaged ^J®ptrans and " 5 ga oi s HF values for the conjugated acyl fluorides are 20 and Table XIII Data of Some Three-bond H-F Coupling Constants3 , f n , 3_s-trans 3,s-cis 3 ,ave _ _ Compounds HF HF HF Ref• H H ^}C=C 19.2 + 0.5b -7.5 + 0.9* 5 8.^ + 0.1 This work H COF OH, H >=/ / \OF \=< 21.5 ± 0.5b -11.6 + 0.9b 7*8 ± o.l This work :C 19,4 + 0.5 -8.7 + 0.9 6.9 + 0.1 This work ^COF si ■ P - Table XIII (cont'd) Compounds P-CH3O < t . H w / \ 3 s-trans JHF 20. I* + 0.5° 3 s-cisf HF 3Tave HF Ref. 9.6+0.9^ 6.8+0,1 This work 71 17.33® (J3J 4,) 29 -o V - n Table XIII (cont'd) Compound / 3 . c=c , h9 \ / 2 / c=\ H„ 3Ts-trans HF 3_s-cisf 3Tave HF HF 17.0^ (J^) All coupling constants are reported in Hz. In vinyl chloride solution. In 60% vinyl chloride and chlorodifluoromethane solution. In 70% vinyl chloride and J0% acetone solution. The reported conformations are a mixture of s-trans and skew forms. These coupling constants are the calculated values, see text. 77 -9 Hz respectively. From these two values, one may deduce that B = -14.5, and A + C = -5«5» in a equation of form (2) . The nature of the potential energy barriers restricting rotation around carbon-carbon single bonds between SP -type hybridized carbon atoms has been the 47.48 subject of great interest to chemists recently. '' As mentioned in Chapter 1, the general rotational potential energy function for a conjugated system (Eq. (3)) has three contributing terms, V(0) = g1 (1 - cos 6) + g2 (1 - cos 26) + ^3 (1 _ cos 39). (3) The rotational potential energy barrier measured from the lower minimum (0 * 0°) will be V« W - V <® °> = V < ® B ax> W where 0max is the angle of internal rotation at the tran sition state.* It can be shown** that if -Vj + 4v2 - 9V^ > 0, or 4v2 > Vj_ + 9V^» then the non-s-trans form will be in the s-cis conformation, and the simple expression holds * To find the value of ©max and 0min, see Appendix 1. ** See Appendix 1 for proof. 78 V(ir) - V(0°) = Vx + Vy (5) The energy difference between the two forms then gives an indication of the relative importance of the Vx and terms in Eq. (3)* From our low temperature investigation, we found that AG = 0.25 ± 0.0? kcal/mole and a H* = 5*78 + 0.07 kcal/mole for acryloyl fluoride in vinyl chloride solution. We have also established that the two equili brium conformations are s-trans and s-cis. with the former being more stable. If we neglect zero-point energy contributions to the entropies of the s-trans and the s-cis conformers, and assume AS = 0 (vide infra) for these two conformers, then the measured free energy difference between the two forms can be set equal to the sum of the and terms. We have Vx + - 250 cal/mole (6) The rotational potential energy barrier can be taken as the enthalpy of activation for s-trans to s-cis interconver- sion, V« W ’ = i (Vj ♦ v2 + v3) - i (Vj cos enax + v2 cos 26max + V3 cos 38max). (7) 79 In order to find 6__v value, one sets the derivative of IudX V with respect to 0 equal to zero and then determines the roots of the following equation, 12V3 cos2emax + I*v2 008 em ax + Vx - 3V3 = o (8) We have three equations, (6), (7), and (8), yet there are four unknowns (Vlt V2, V-j, and ®max) to be solved. Addi tional information is needed in order to solve for all the unknowns, At this point, we want to examine the values of Vt, V2, and at the different values of 0max. Using Equations (6), (7)» (8), and the assumed values of 0_QV, UlcLX values for V^, V2, and are calculated, as shown in Table XIV. The term is a rotational potential similar to 12 13 that in acetaldehyde or propylene J and is expected to be 73 small for the conjugated systems. Miyazawa and Pitzer reported that * 2.1, V2 - 9»9» and V3 = -0.1 kcal/mole for formic acid* For acrolein in the gas phase, was reported to be small and negative (and also somewhat uncertain), in evaluating three-term rotational potential function for substituted benzaldehydes, the V3 term was 74 even ignored by some authors. If the torsional frequency of acryloyl fluoride in the liquid phase can be measured by the infrared spec- 80 Table XIV The Values of Vlt V2, and V^ at Different Values of 9 „ max a max vab v 2 b v a b 85° -325 5735 575 o 00 00 -15 5773 265 89° 85 5778 165 906 187 5780 63 91° 289 5778 -39 92° 389 5773 -139 95° 698 5735 -*448 a. 9max is the angle of torsion at the transition state. b. In cal/mole. troscopy, then the needed information can be obtained from the following equation V* = vx + 4V2 + 9Vj = y (9) where vis the 0 +1 torsional transition frequency, and F is the reduced moment of inertia constant. Unfortunately, only the value of v in the vapor phase had been reported,and the v value in liquid phase could not be found. Usually, the values differ appreciably (20 to 81 25%) in the different phases I 4 . K * Carlson et al. reported that v_ - 115 cm ■ * ■ , and s Vg* = Vi + 4V2 + 9V-j * 18.65 kcal/mole for acryloyl fluoric de in the gas phase. It is unfortunate that the torsional frequency (v ) of acryloyl fluoride in the liquid phase was not measured. The torsional frequencies of acrolein were ^ -j reported to be v = 157, and v = 182 cm” for the gas S 1 phase and the liquid phase respectively. The torsional frequency of acryloyl fluoride in the liquid phase (v1) has 1 ffP been estimated by multiplying v by a factor of ~ — Hence, S 1 b f one obtains the approximate values of = 133 cm'1, and # = 21,6 kcal/mole for acryloyl fluoride in the liquid phase. Using the estimated value of 21.6 kcal/mole coupled with Eqs. (6), (7)» and (8), one obtains » ^00, Vg = 56^0, V-j » -150 cal/mole. As expected, the term is small. If V-j term is completely ignored, then one has = 250, V2 = 5655 cal/mole, and ©max = 90.6°. The (1 — cos ©) term in the potential function has a periodicity of 2" in G and has a maximum value at G = it. The main factor contributing to should be the dif ference in interatomic interaction of the cis Cj hydrogen with the fluorine or oxygen atoms in the planar forms of acryloyl fluoride. The analysis given above indicates that must have a positive value (0.1 to 0.4- kcal/mole). This implies that the cis-3H...0 interaction (if repulsive) 82 S jmust be greater than the corresponding cis-3H.*. F inter- i action. If both Vj and are small, as indicated, the |major contribution to the rotational energy barrier must i jcome from the V% term. Our analysis indicated that V2 has the value 5*78 kcal/mole and the 0max is-92°. The two fold barrier arises mainly from the overlap between the ir-orbitals of the conjugated double bonds and was expected to be the m&jor term in the potential function. The possibility that the non-s-trans forms of the subsituted 1,3-butadienes may adopt a skew conformation has been pointed out by Bothner-By and coworkers,28*29 our analysis of the potential function indicates that the meta stable non-s-trans form is in fact the s-cis form with 0 = 180°. One further way of distinguishing between the ! 1 i j js-cis and the non-planar forms would be the determination ! lof the entropy difference between the s-trans and the non- i s-trans forms. A consideration of the theoretical value of AS will be helpful and is given below. The difference be- j tween the entropy of two conformers can be expressed as a j jsum of contribution from translational (Astr), vibrational i O £ ;(ASvib)» and rotational motion (ASro^), i AS = AStr + ASvife + ASrot. i Since the masses are equal for both conformers of acryloyl fluoride, AStr = 0. The barrier to internal rotation (Ea) | is high compared with AG in acryloyl fluoride, the low- : lying vibrational frequencies of the two conformers are likely to be nearly equal and therefore ASV^^ = 0. The ro tational contribtuion to the entropy difference is given by ASrot = £ R In (At Bt Ct / Ac Bc Cc), (11) i where Aj_, B^, and are the principal moments of inertia for conformer i. Because the values of moments of inertia are very close for the two conformers,^ the ASrot term is ; likely to be very small, ASro- t = 0. Thus, the entropy difference between the s-trans and the s-cis conformers of acryloyl fluoride is likely close to zero. o g—cis At the slow exchange limit, the value of cinnamoyl fluoride was found to be -5*7 Hz from the low temperature nmr spectra. The experimental value differs from the calculated value (-6.8 Hz) by 2.1 Hz. This deviation probably comes from the assumption that | | AS = 0 for the s-trans and the s-cis conformers. From the | experimental values of 3jS"trans (19.^ Hz), (-5*7 Hz), and the averaged (6.9 Hz) at 35° for cinnamoyl fluoride, the fractional populations for the two conformers i and aG^ value at 35° can be calculated. The value of AG^ so obtained was 5 ± 98 cal/mole. The aS value was calcu lated to be 0.5 ± 0.5 e.u. from the two values of AG at I -130° and 35°• From the above considerations, we can con clude that the AS values for the s-trans and the s-cis | 84 conformers of these a ,e-unsaturated acyl fluorides must be I lo small. Koster's reported value of AS = 1.3 e.u., for acryloyl fluoride seems to be questionable and undoubtedly contributed to his incorrect analysis. Before discussing the substituent effect on the rotational energy barriers of the four a,$-unsaturated acyl fluorides we have studied, it is of interest to examine the j physical properties of some a,e- unsaturated aldehydes. When a double bond is conjugated to an aldehyde group, some | physical properties undergo significant changes. This can be seen by examination of the ionization potentials, dipole moments,^ and C=0 stretching frequencies (in ■ vapor phaseJ??*?® of the aldehydes shown in Table XV. Table XV The Ionization Potentials, Dipole Moments, and Carbonyl Stretching Frequencies of Some Aldehydes Substance Ionization potential3, (e.u.) Dipole moment*5 (D) c vc=o (cm”*) HCHO 10.88 2.34 1744 ch2=chcho 10.25 3.11 1724 CH3-CH=CHCHO 9.81 3.67 1715 a. See Ref. 76, b. See Ref. 46 and 76. c. See Ref. 77 and 78. The interaction between a double bond and a carbonyl group is shown in the differences in the ioniza tion potentials, dipole moments, and the C-0 stretching frequency in formaldehyde and acrolein. Acrolein was found to have lower ionization potential, higher flipole moments, and lower carbonyl stretching frequency than formaldehyde, All these property changes can be explained by the contri bution of dipolar resonance form (XVa) as result of conju- gati on. the ionization potential decreases by 0.44 e.u., whereas the dipole moment increases by 0.56 Debye on the intro duction of one methyl group into the 0-carbon atom of acro lein. This is in accord with the simple view that the inductive effect and/or the hyperconjugative effect of the methyl group stabilizes the polar resonance form (XVa), and the electron transfer from the double bond toward the more electronegative oxygen atom is facilitated. In other words, crotonaldehyde is expected to be more strongly conjugated than acrolein. From the preceding discussion R ,H H H (XV) (XVa) By comparison of acrolein and crotonaldehyde, 86 | of the ionization potentials, dipole moments, and infrared | carbonyl stretching frequencies of the a-unsaturated aldehydes, it appears that the polar resonance form (XVa) makes a significant contribution to the resonance hybrid In acyl fluorides, a similar trend in increase j : of dipole moments is oberved as in the aldehydes. Formyl fluoride (HCOF) has dipole moment of 2.2 Debyes, while the dipole moments of acryloyl fluoride were reported to be 3*26 and 3»21 Debyes for the s-trans and the s-cis, res pectively. A similar dipolar resonance structures again can be drawn for a,3-unsaturated acyl fluorides. If the (XV) (XVI) (XVIa) R \ / ,C=CV H R H H C— F H 0 (XVII) (XVIIa) 87 dipolar resonance structures (XVIa) and (XVIIa) contribute significantly to the resonance hybrids, (XVI) and (XVII), one would expect that the two stable conformers of acryloyl fluoride take the planar s-trans and the s-cis forms, and that the rotational barriers are mainly two-fold. This is in accord with our experimental results. The Arrhenius activation energies (E_), the free c l energies of activation at -100 (aG*_iqqO ), the ground- state free energy differences (AG), the entropies of acti vation (AS*), and the frequency factors (A) of the four acyl fluorides studied are shown in Table XVI. The free energy of activation at -100 (aG*_10Qo ) separating the s-trans and the s-cis conformers of acryloyl fluoride was found to be 5*62 kcal/mole. Upon methyl sub stitution at the 0-position, the barrier of crotonyl fluo ride is raised to 6.69 kcal/mole, an increase of 1.0? kcal/ mole, The inductive effect and/or the hyperconjugation effect of the methyl group stabilizes the polar resonance structure (XVIa) of crotonyl fluoride, and result in the lowering of the ground-state energy of crotonyl fluoride. At the transition state, the conjugation between the C=C double bond and C=0 double bond is virtually destroyed. The methyl group would be expected to stabilize the iso lated double bond but to a lesser extent than in the con jugated planar form. Consequently, the methyl substituent Table XVI The Free Energies of Activation at -100° (AG*_10Qo), bhe Arr^en^ - Us Activation Energies (E ), the Ground-State Free Energy Differences (AG), the Entropies of £ L Activation (AS*), the Enthalpies of Activation aH*, and the Frequency Factors of the 0,6 -Unsaturated Acyl Fluorides Fluorides Acryloyld Crotonyl^ Cinnamoyle p-CH^O-Cinnamoyl^ p a AG -100° 5.62 + 0.0? 6.69 ± 0.05 7.13 ± 0.07 7.32 ± 0.08 a Ea 6.60 + 0.07 6.99 + 0.05 7.34 + 0.07 8.21 + 0.08 AGb 253 ± 67 215 ± 63 95 ± 58 112 + 54 4=° AS 0.93 + 0.4 -0.25 ± 0.30 -0.88 ± 0.3^ 2.98 + 0.43 AH 5.78 + 0.07 6.65 ± 0.05 6.98 + 0.07 7.84 + 0.08 A (1.30 + 0.28) x 1013 (0.85 ± 0.10) x 1013 (0.68 + 0.11) x IO13 (4.81 + 0.04) x 1013 CD Table XVI (cont’d) 89 a. In kcal/mole. b. In cal/mole. c. In e.u.. d. In vinyl chloride solution. e. In 60$ vinyl chloride and 1+0$ chlorodifluoromethane solution. f. In 70% vinyl chloride and 30$ acetone solution. 90 at the 8-position will have a lower stabilization effect in the transition state than in the ground-state. This results in the raising of the effective height of the rotational barrier of crotonyl fluoride in comparison to acryloyl fluoride. It is noted that an electron-releasing methyl group at the exposition in resonance form (XVIa) will tend to stabilize the dipolar form and increase the partial double bond character of the C^-Cg single bond. This increased double bond order should result in an increased rotational energy barrier. The introduction of a benzene ring at the g- position of the a,g-unsaturated acyl fluoride, will delo calize the positive charge of resonance form (XVIa) further into the benzene ring, and hence stabilize the ground-state of cinnamoyl fluoride substantially. The increased con jugation raises the rotational energy barrier for cinnamoyl fluoride in comparison to acryloyl fluoride and crotonyl fluoride. The stabilization of the ground-state due to the resonance effect of the benzene ring is greater than the inductive effect and/or the hyperconjugative effect of a methyl substituent at the 0-position. The £-methoxy group on a benzene ring generally has two effects: (i) electron releasing through the re sonance effect (-R), and (ii) electron withdrawing through the inductive effect (+1). The observation that £-metho- l I xycinnamoyl fluoride has a higher rotational energy barrier [than cinnamoyl fluorides reveals that the resonance effect !of the methoxy group is more important than its inductive i i jeffect in this system. c—C :c=o F (XVII) (XVIIa) j Our values of the ground-state energy differences i I (AG) between the s-cis and the s-trans conformers of the j Substituted a,0-unsaturated acyl fluorides are much lower i ^ A 1 7 4 8 jthan that of acrolein,1,3-butadiene, ’ and methyl 79 [vinyl ketone, all of which exist predominantly in the s- t itrans forms (Fig. 18). Assuming the same bond angles and bond distances as found for the s-trans forms*? of 1,3-buta diene, the calculated cis-cis H-H distance in the s-cis ! 0 [form of 1,3-butadiene is 1.93A. In the case of acrolein, i ! Ithe distance between cis C-^-H and the oxygen atom in the j o s-cis conformation is 2.53A. The van der Waals radii for Figure 18. The conformations of 1t3-butadiene, acrolein, methyl vinyl ketone, and acryloyl fluoride. 93 o , o o 80 H, 0, and F are 1.21A,* 1.40A, and 1.35A, respectively. Steric repulsion may account for part of the destabili zation of the non-s-trans form, but is not the determining factor. Because the s-trans form of methyl vinyl ketone is sterically less favorable than the s-cis analogi also, less severe steric interaction is expected in the s-cis form of acrolein. It is questionable whether non-bonded H-H re pulsion in the s-cis form of 1,3-butadiene could produce the necessary 2.3 kcal/mole destabilization.®1 The in stability of the non-s-trans metastable conformers of 1,3- butadiene, acrolein, and methyl vinyl ketone must be due to the increased contributions of and terms to the potential functions. The metastable forms will be s-cis if 4v2 - - 9V3 > Oi or the raetastable forms will be non- planar (skew) if 4v2 - - 9V^ <0. A recent self-con sistent molecular orbital calculation on 1,3-butadiene was 48 reported by Radom and Pople. They adopted a flexible rotor model (i.e. allow the widening of the CCCC angle in the s-cis form) in the calculation, and the results reveal that the metastable conformer is in the planar s-cis form. The activation energy barrier for the s-trans conformer to the s-cis is 6.73 kcal/mole, while the energy difference *The hydrogen van der Waals radius value of 1.21.8 has been disputed. It is felt that 1.5-1.68 is probably a more realistic value.“2,83 is 2,05 kcal/mole. Qualitatively, from an electrostatic point of view, the interactions between n-electron clouds of two C=C bonds or one C=C bond and one O O bond linked by a C-C single bond are repulsive in nature. The repulsion should be maximum when the two double bonds are eclipsed (0 = 180°), and minimum when they are trans (0 = 0°) to each other. The electrostatic repulsions between these double bonds will be much greater than the interaction between a C=C bond and a C-H bond. In fact, all the available evidence indicates that both the C=0 and C=CH2 bonds prefer to be eclipsed by an adjacent single bond.1 Consequently, it is possible that the interaction between two eclipsed double bonds renders all the non-s-trans forms of acrolein, 1,3-butadiene, and methyl vinyl ketone un stable. In the case of acryloyl fluoride, both the s- trans and the s-cis forms were found to exist in comparable amounts (about 60 to 40 at fcoom temperature, respectively) from our low-temperature nmr study. Apparently, the interaction between a C=C moiety and C-F moiety becomes comparable to the interaction between a C=C moiety and a C=0 moiety, with the latter slightly greater in magnitude. It is noted that fluorine atom has six non-bonded electrons and is just slightly smaller than the oxygen atom in size. 95 From Table XVI, the ground-state energy diffe rences between the s-cis and the s-trans conformers of the acyl fluorides decrease in the following orderi acryloyl> crotonyl> cinnamoyl^ p-methoxycinnamoyl fluoride. The substituent effects on the changes in the ground-state energy differences need to be explained. From a SCF-MO- ' 8^ ; LCAO-CI calculation of the acrolein moleoule, the charge distributions on the carbon and oxygen atoms were reported as followsi The residual charges centered at the atoms are shown in parenthesis. The distance between the C3 carbon atom and the oxygen atom is smaller in the s-cis conformer than in the s-trans conformer. The electrostatic interaction between the C-j carbon atom and the oxygen atom should Figure 19* The charge distributions on the carbon and oxygen atoms of the s-trans and the s-cis conformers of acrolein. favor the s-cis conformer over the s-trans conformer. Thus, the increased charge separation due to the increased conjugation should diminish the energy difference between the s-cis and the s-trans forms. As discussed earlier, the order of increasing conjugation isl&s followsi acryloyl <crotonyl <cinnamoyl < jo-methoxycinnamoyl fluoride Therefore, the ground-state energy differences between the s-cis and the s-trans conformers of the acyl fluorides should decrease in the following orderi acryloyl >crotonyl> cinnamoyl> £-methoxycinnamoyl fluoride Experimentally, the difference in AG between cinnamoyl fluoride and £-methoxycinnamoyl fluoride (17 cal/mole) is negligible compared to the uncertainty {+ 55 cal/mole) in AG values. The jj-methoxy substitution on benzene ring does not produce an appreciable effect on the AG value. The predicted decreasing trend in the A G values is therefore consistent with our experimental results. The ground-state free energy difference between the two conformers of acryloyl fluoride was found to be AG = 253 + 67 cal/mole, from our results. This value is in fair agreement with the value, 90 + 100 cal/mole, reported from a microwave study,^ and 150 + 100 cal/mole, i f , 5 from an infrared study. ^ The small differences presumably are due to the fact that the measurements were made in 97 different media. The microwave experiment was carried out in the gas-phase j pur nmr experiment was performed in vinyl chloride solutionj and the infrared measurement was carried out in carbon disulfide solution. However, the reported value that AH = 800 + 250 cal/mole by Koster for the two conformers of acryloyl fluoride in fluorotrichlo- romethane solution seems to be too high in oomparison with our result. Koster's nmr data treatment was closely similar to the procedure developed by Gutowsky, Belford, and OC McMahon (hereafter referred to as GBM method). Since the GBM method had been widely adopted by many nmr workers, it seems to be worthwhile to give a brief account on this method.®^ The experimentally observed average nmr para meters (coupling constants of chemical shifts) can be related to the nmr parameters of the individual conformers by <J>T = E XjJ^ (12) Here x^ is the mole fraction of the conformer i. For the system involves only ’ two conformers, the equilibrium constant can be written as X1 ni K 1 * — *- exp (AS/R) exp (-aH/RT), (13) x2 n2 where n^ is the number of degeneracies of conformation i, 98 A S - - S2 is the entropy difference, and aH is the corresponding enthalpjjt difference. Since + xg = 1, substitution of (13) into (12) gives <J>T = (J2 + KJX) / (1 + K) (14) or K = (J2 - <J>T) / (<J>T - Jx). (15) It is possible to determine K at different temperatures only if Ji and are known and assumed to be temperature independent so that all the temperature dependence of arises from population changes. In the GBM approach, AS is usually assumed to be zero, and the equilibrium constants and the mole fractions depend only on the value of AH. A function $ is defined according to 2 * = E (<J>T - x1J1 - x2J2} (16) and is minimized by variation of the unknown parameters AH, J^, and J2, In practice this process is performed by starting with certain assumed values of AH to calculate and x2 at each temperature. The latter are substituted intfc a set of equation (12) which are solved by the method of least squares to give the best values of ^ and J2. These are then fed back into Eq. (16) to obtain the func tion < f > for the particular value of aH. The process is 99 repeated with a systematic variation of AH to obtain the most probable values of both the nmr and thermodynamic parameters. The GBM method has been applied to a variety of substituted ethanes®^ and claimed a high accuracy in the measurement of AH in some cases. However, there are cases that the values derived from GBM method were proven to be inaccurate. For example, in 1,l,2-trichloro-l,2,2-tri- fluoroethane (CF2CI-CFCI2)1 the AH value of 2000 to 2760 cal/mole is an order of magnitude different from that measured from vibrational spectra (350 + 150).®^ Newmark and Sederholm®^ "froze out" the conformers in CFgBr-CFBrCl and found that the coupling constants of the conformers did not agree with those found by GBM. Govil and Bern- 86 stein had investigated the rotational isomerism of 1,1,2,2-tetrabromofluoroethane (CHBrg-CFB^) by using low- temperature *H and ^^F nmr spectra. They found that the values of ah, and coupling constants obtained from the GBM method and the "froze out" method are completely different. The two sets of results are compared as follows 1 AH J (Hz) J, (Hz) AS (cal/mole) ® GBM method -800+25 +5.64+0.02 +17+0.16 0 (assumed) "Froze out" -198.9 1.15 22.2 -0.137 method ioo j The most probable reason for these discrepancies is that | the function $ defined in Eq. (16) in GBM method varies ; j j very slowly with AH. This means that a wide range of j I AH values are acceptable, and that the ♦ . value so j m m | obtained may not be the true solution and can lead to the incorrect results in the values of AH, and the coupling constants of the individual conformers, Thus, Govil and Of. Bernstein concluded that the measurements of the average coupling constants at high temperatures, <J>T, may not give accurate values for the thermodynamic and nmr para- iip meters. Since Koster had employed the similar procedure as GBM did (except that he had AS = 1.3 e.u.), it is thus not surprising that Koster's results are quite different from ours. The two sets of parameters are compared as ! shown belowi A Ga AHa i c,b 3Ts-transc 3Ts-cisc A b JHF JHp i Koster's I 800+250 1.3 1^.5+0.5 -3*5±0.5 Ours 253+67 * * * 1 - 19.2+0.5 -7.5±0.9 a. In cal/mole. b. In e.u.. c. In Hz. CHAPTER 4 EXPERIMENTAL ^-1. Instrumentation and Techniques. Melting points were taken on a Biichi melting point apparatus and are uncorrected. All infrared spectra were recorded on Perkin-Elmer spectrometers Model 337 or ^57* All nmr spectra were recorded on a Varian Associates HA-100, unless otherwise specified. The instrument was equipped with an integrator/decoupler (#V3521A) and a variable-temperature controller {#V^3i *'3) • Fluorine chemi cal shifts are reported in Hertz downfield from fluorotri- chloromethane as internal standard. The 9^*1 MHz low- temperature nmr spectra were recorded in HR mode and were calibrated with audio side bands of the reference signal, using a Hewlett-Packard Model 200AB audio oscilla tor and Model 5512A frequency counter. The ambient-tempe- rature spectra were normally recorded in the frequency- sweep mode using an internal lock. At the low tempera tures, the sample was thermostated with pre-cooled nitrogen gas passing through a vacuum-jacketed Dewar into the probe. In order to ensure that the true Lorentzian line-shapes are actually observed in our line-shape analysis, pre cautions are made to avoid saturation of the nmr signals, 101 102 and not to use excessive filtering of noise. Whenever the temperature was changed by an increment, the field homogeneity controls were readjusted by observing fluoro- trichloromethane signal which is not affected by the exchange process. The temperatures were determined using a precalibrated copper-constantan thermocouple inserted in an nmr tube containing suitable solvent. For each temperature measurement, the solution was allowed to stand until the equlibrium temperature was reached. The samples were prepared by dissolving about 15# (v/v) acyl fluoride in vinyl chloride (and other mixed solvents) using fluorotrichloromethane as internal standard. The sample tube was then sealed under vacuum. The concentrations of the various solutions are shown below in approximate percentage by volume. a) Acryloyl fluoride 15% acryloyl fluoride, 10# fluorotrichloromethane, and 75% vinyl chloride. b) Crotonyl fluoride 15# crotonyl fluoride, 10# fluorotrichloromethane, and 75# vinyl chloride. c) Cinnamoyl fluoride 10# cinnamoyl fluoride, 10# fluorotrichloromethane, 30# chlorodifluoromethane, and 50# vinyl chloride. d) ]3-Methoxycinnamoyl fluoride 10# ]D-methoxycinnamoyl fluoride, 10# fluorotrichloro methane, 25# acetone, and 55# vinyl chloride. 103 QQ Construction of the thermocouple. Copper and constantan ! wires were used for constructing the thermocouple. The junctions were made by tightly twisting about 1 cm of the | ends of the two wires (well cleaned with sand paper), warming the ends with a flame, and then dipping them into "Brazo Flux" (from the Linde Air Product Co.). The junc tion was then heated in a hot Bunsen flame until the wire were fused. The wiring of the thermocouple is shown in the following diagrami To potentiometer To nmr probe > To ice-water bath ------- Copper Constantan | Calibration of the thermocouple. The known melting j I points of water, carbon tetrachloride, chloroform, and carbon disulfide, and the boiling point of fresh liquid nitrogen were used as standards to calibrate the thermo- j couple. The liquid-solid equilibrium mixtures of carbon tetrachloride, chloroform, and carbon disulfide were obtained by pouring liquid nitrogen into a well-cleaned Dewar containing the above mentioned spectral grade pure 104 ] J ! solvents. The liquid-solid slush gave stable EMF readings j I I (from a Leeds & Northrup Co. Potentiometer, serial no. ; ! 718429), indicating that the solid-liquid equilibrium had been reached. The results are shown as followsi Substance EMF (mv) Temp, (obs.) Temp* (lit.) ice-water 0.00 0° 0° CCl^ 0.854 0 00 • <V 1 CM 1 -22.98° CHCI3 2.281 -64.3° -63.5° 0S2 3.710 -113,3° -110.8° liq. N2 5.541 -200.0 -195.8° Computer calculation of theoretical curves. All computer calculations were performed at the University of Southern California on an IBM 360/65 computer equipped with a University Computing Company (UCC) Digital Incremental Plotter. The computer program, DNMR2, used for calcula tion of the theoretical curves for the intramolecular exchange process, was written by Binsch and Kleier.^ The program simulates complex exchange-broadened nmr spectra from chemical shifts, coupling constants, rela xation times, rate constants, populations and certain scaling parameters. Detailed information about the program can be obtained from QCPE (Quantum Chemistry Program Exchange), Indiana University. For the computer j calculations in this project, there are several input ! parameters which need to be explained and are given below. i I 1) Populations — At the very low temperatures, when the I slow-exchange limit has been reached, the populations of 1 the two isomers can be determined by integration of the two 19 distinct signals from the low temperature ^F nmr spectra. Populations at the higher temperatures were estimated by the following equation,* AG = -RTX In Px / P2 = -RT2 In P' / P'2 or r S1 log P1 / P2 = T2 log p*x / P'2. Sometimes the populations had to be adjusted slightly in order to have a good fit with the experimental spectrum. 2) Chemical shifts — These were obtained from the j spectrum at the lowest temperature taken, presumably at i the slow-exchange limit, and were kept constant throughout i j the calculation. 3) Coupling constants -- The three-bond H-F coupling 3 S _______ . Jjjp , were taken from the low-temperature spectra. The corresponding coupling constants of the s-cis conformation analogs, s - < HF o s-cis » were calculated from the following equation, V/ - * 3 C r a n s m i - p> 3^ ; cis * See footnote on p. ^0. I 106 i | 3 9.V 6 where is the averaged coupling constant at ambient temperature, and P is the population of the s-trans ! | conformer of the acyl fluoride at ambient temperature. | j 4-) T2 — The effective transverse relaxation time (in sec.) was obtained from the line width at half-height (^l/2) eternal fluorotrichloromethane reference signal by the relationship T2 = 1 / ( it 5) Scale — The horizontal plotting scale (in mm/Hz) j was determined from the separation of the reference peak : (CGl^F) and the generated sidebands (1 KHz or 2 KHz). In most cases, only two nuclei (the fluorine and | the C2 proton) were included in the calculation, since no difference was observed for the 2- and ^-nuclei calcula- : tions for the range of rate constants reported. From the input data, several spectra were generated by the UCC i ! plotter. By placing the theoretical spectrum on top of the j | experimental spectrum, the line-shapes of the two spectra | could be compared. The exchange rate constant at the particular sample temperature was obtained from the calcu lated spectrum considered to be in best agreement with the experimental spectrum. One way to roughly estimate the rate constant is to plot the half-height width (W1//2) of the computer generated spectra vs. rate constants. From the observed experimental value of Wjy2» ‘ fche rate constant k could be estimated. Activation parameters 107 j were calculated by a program ACTENG, written by DeTar,57 The calculations are based on a weighted least squares treatment which allows the use of data of different pre- | cision. i | 4-2. Preparation of Compounds Used in This Study. Acryloyl fluoride.^ A mixture of 10 g (0.11 mole) of acryloyl chloride and 19.6 g (0.11 mole) of antimony trifluoride was refluxed for six hours under a nitrogen atmosphere. The mixture was then fractionally distilled; the fraction boiling at 30 - 33° was collected. The yield of the desired product was 5*0 g (6l$). The product was kept in a Dewar cylinder filled with Dry Ice-acetone. No impurity could be detected in the 60 MHz nmr spectrum | which was identical to that reported by Koster.42 The 100 MHz proton nmr spectrum of a carbon tetrachloride i solution showed three sets of multiplets centered at 6.1, i 6.5, and 6.65 ppm downfield from TMS. The analysis of the i nmr spectrum was carried out by using the computer program i tc | LA0C00N III. The nmr parameters are shown in Table I in i 19 | Chapter 2. The F nmr spectrum of a vinyl chloride solution at ambient temperature was a multiplet 23*94 ppm downfield from internal fluorotrichloromethane. Attempted preparation of crotonyl fluoride. A mixture of 6.0 g (0.58 mole) of crotonyl chloride and 10.2 g (0.58 mole) of antimony trifluoride was refluxed for six 108 hours under a nitrogen atmosphere. The mixture was then fractionally distilled. The yield was 2.30 g (45%)» bp 84-87°. The product was stored at the temperature of Dry Ice-acetone. The nmr spectrum (60 MHz) indicated that a mixture of acyl fluorides in a ratio of 20«80 had 19 been obtained. The F nmr spectra also indicated that there were two components, assumed to be crotonyl and 19 isocrotonyl fluoride. The F nmr spectral data (in car bon tetrachloride solution) were as followst CH3CH=CHC0F Crotonyl fluorfide vt = -2316 Hz (-24.61 ppm) from CCI3F a doublet with ^J^f = 7*84 Hz Isocrotonyl fluoride Vc = (”^3*66 ppm) from CCI3F o 3 two quartets with «Jftp = 9*05 Hz = 1.13 Hz. Benzoyl fluoride.^0 Hydrogen fluoride (10 g, 0.5 mole) was distilled from the cylinder through a polyethylene tube into a 250 ml polyethylene transfer bottle which had been previously weighed. No protection against atmos pheric moisture was necessaryj the bottle was cooled in Dry Ice-acetone bath, and about 10 ml of liquid hydrggen fluoride was collected. The reaction was carried out in a 250 ml polyethylene bottle fitted with an inlet tube 109 leading to the bottom# A long polyethylene tube was con nected to the top of the bottle to allow the hydrogen chloride and excess hydrogen fluoride to escape. Benzoyl chloride (56.2 g, 0.4 mole) was placed in the reaction vessel, and the hydrogen fluoride gas was then introduced by distillation from the transfer bottle through the inlet tube. The hydrogen fluoride was added over a period of approximately two hours. After the addition was complete, the reaction mixture was warmed to 3Q-b0° and kept at this temperature for one hour. The mixture was then washed in an ordinary glass separatory funnel with 100 ml of ice-water in which 2.5 g (0.04 mole) of boric acid had been dissolved. The organic layer was quickly separated, and 2 g of anhydrous sodium fluoride and 2 g of anhydrous sodium sulfate were added. The mixture was allowed to stand for 30 minutes and was then filtered and distilled through a short co lumn. Yield of benzoyl fluoride was 21,3 g (43$), bp 159- 160° (lit., 159-160°).^ The 94.1 MHz nmr spectrum in vinyl chloride solution at ambient temperature showed a singlet 1?.04 ppm downfield from internal fluorotrichlo romethane. The ir spectrum (thin film) showed absorption at 3055. 1805, 1596, 1448, 1235, '1178, 1032, 1010, 797, 768, 700, 646 cm”^. Attempted synthesis of crotonyl fluoride. A mixture of 110 2.1 g (0.02 mole) of crotonyl chloride, 3*1 g (0.025 mole) of benzoyl fluoride, and 0.2 g of sodium fluoride was re fluxed for two hours under a nitrogen atmosphere. The mixture was then distilled using a microdistillation apparatus. The distillate, bp 120-130°, was identified as starting material (acid chloride) by nmr spectroscopy. 6i Crotonyl fluoride. A mixture of 2.0 g (0.023 mole) of crotonic acid, 3»1 g (0.025 mole) of benzoyl fluoride, and 0.2 g of sodium fluoride was slowly distilled from a microdistillation apparatus. The distillation was stopped before the distillate temperature reached 100°. The yield was 1.31 g (657S). Spectral data for crotonyl fluoride 1 I.R. VCG14 2950 (broad), 1805 (C=0), 1605, 164-5 max (C=C), 14-40, 1375. 1308, 1292, 1210, 1090, 1020, 962, 944, 820 cm"1. 100 MHz 1H nmr spectrum (in CCl^ solution) CH \ / : c=c H' COFj v2 = 5«77 ppm («) J12 = 7.29 Hz v-j = 7.13 ppm (6) = 0 vCH3 ” 1*^ ppm ^ J23 = 15*68 Hz J2 ,CH-j * 1,72 HZ 1.3-Dibromo-2-butanone. 2-Butanone (36 g, 0.05 mole) | was mixed with 50 ml of 48$ hydrobromic acid and chilled with ice-water. Bromine (160 gf 1.0 mole) was added drop- wise to the ketone solution. After the addition of bro mine was completed, 100 ml of water was added to the solu tion, and the heavier organic layer separated. The organic layer was then fractionally distilled under reduced pres sure. The first fraction,bp 36-39°/8 mm* was identified by nmr spectroscopy as 3-bromo-2-butanone. The second fraction bp 72-7^°/6 mm, was identified by nmr spectroscopy as the desired product, 1,3-dibromo-2-butanone. The yield was 61 g (53$)• Spectral data for CB^BrCCHBrCH^ 1 I.R. vneat 2960, 2940, 1720 (00), 1450, 1390, max 1340, 1260, 1218, 1185* 1152, 1107, I 1060, 1024, 976, 870 cm"1. 100 MHz *H nmr (in carbon tetrachloride solution) j 1.74 ppm (6), 3H, doublet, CH3- 4.10 ppm (6), 2H, AB quartet, CHgBr- ; 4.80 ppm (6), 1H, quartet, -CHBr- | with 2J§5m = 12.3 Hz, Av * 28 Hz HH AB 112 3 J. * 6.8 Hz H-CH3 Isocrotonic acid. 1,3-Dibromo-2-butanone (46 g, 0.20 | mole) was added to a solution of 100 g (1.0 mole) of The mixture was thoroughly stirred for 3 hours. The solution was then extracted with ether (2 x 150 ml)t and the ether solution was discarded. The aqueous portion was acidified with dilute hydrochloric acid and again extrac ted with ether (6 x 150 ml). After drying over magnesium sulfate, the ether phase was evaporated at the vacuum of the water aspirator to prevent superheating and isomeri- | zation. The last traces of ether were removed under high | vacuum (0.4 mm) during 15 minutes. The crude product j ; (12.4 g» 72$) was dissolved in 25 ml of petroleum ether ; and left at -20° for several days. Crystals separated and : were collected by filtration while cold. ; Spectral data for isocrotonic acidi | potassium bicarbonate in 1 1 of water during 5 minutes \ C00H :c=c \ I.R CC14 max 3000 (broad), 1690 (C=0), 1636 (C=C) 1442, 1291, 1235. 1227, 925. 818, 672 (cis CH-CH out of plane) cm”*. 1 100 MHz H nmr (in carbon tetrachloride solution) 2.1 ppm (<5), 3H, doublet of doublets, CH^ 5.74- ppm (6), 1H, two separate quartets, 6*35 ppm (fi). 1H, two overlapping quartets, H a 3jHaHb = 11,5° Hz 3jOH3-Ha ■ 7-24 HZ ^Jpu u = 1.80 Hz Isocrotonyl fluoride. Isocrotonic acid (2.0 g, 0.02 mole), benzoyl fluoride (3.1 g, 0.025 mole) and sodium fluoride (0.2 g), were mixed in a 10 ml round-bottomed ; flask. The mixture was slowly distilled over a period : of 3 hours from a small-scale distillation apparatus I equipped with a drying tube. A fraction boiling at 35- j 4-0° at ?60 mm was taken. The yield was 150 mg (8* 59^) • ! Analysis of the nmr spectra of the product so obtained I revealed that the product was a mixture of isocrotonyl fluoride and crotonyl fluoride, with a ratio of 79 to 21, isocrotonyl fluoride being the major product. NMR para meters for isocrotonyl fluoride in carbon tetrachloride solution were as followst 114 VCH3 “ *'lk ppm (S) JCH3-Ha = 7.33 Hz vHb * 5>76 PP"1 (5) JCH3-Hb = U 7 9 H8 VHa = 6,62 ppm JHa-Hb = 11<2^ Hz JOH3-P - 1-50 HZ JHa-F = 8-81* Ha ^Hb-F = ^z* Cinnamoyl fluoride.^1 Cinnamoyl chloride (5.0 g, 0.03 mole) was cooled to 0° in a 250 ml polyethylene bottle. Excess hydrogen fluoride gas (about 2 g) was introduced through an inlet tube over a period of 1 hour while keep ing the reaction mixture magnetically stirred. The mix ture was allowed to warm to room temperature for 20 minu tes and then warmed to 40° for another 20 minutes. The residue was extracted with carbon tetrachloride (2 x 30 ml), and carbon tetrachloride was removed under vacuum (0.4 mm). The extraction was repeated once more. The product weighed 3*3 g (73*4 %), Analysis of the nmr spectra indicated that no detectable amount of impurity was present. Spectral data for cinnamoyl fluoride I.R. vccli+ 3050, 1800 (C=0), 1688, 1630 (C=C), max 1450, 1327* 1300, 1187, 1108 (s, C-F stretch) cm"1. 1 100 MHz H nmr (in carbon tetrachloride solution) 7*70 ppm (6), 1H, Ha 6.25 ppm (6), 1H, Hb 7*38 ppm (6), 5H, aromatic protons JHaHb = 15'9 H* JHbF " 6** H- ! £-Methoxycinnamoyl chloride. In a 1 1 round-bottomed | flask was placed 17.8 g (0.10 mole) of £-methoxycinnamic I acid and 23,8 g (0.20 mole) of thionyl chloride. The flask was fitted with a reflux condenser and a calcium I chloride drying tube on top of the condenser, and the : mixture was heated to 80° for three hours. The excess thionyl chloride was removed by distillation with a bath temperature of 100°. The residue was distilled at re duced pressure, bp 125-130°/l mm. The yield was 17•5 g (89%), mp 51-51.5°. Spectral data for jj-methoxy cinnamoyl chloride* I,R* ''max4 3°00f 295°* 2925» 2830 (ch3°“)» 1750 (C=0), 1593* 1566, 1505, 1455, 1420, 1325, 1308, 1280, 1255, 1170, 1115, 1100, 1030, 972, 823, 708, 643, 621 cm”*. 100 MHz *H nmr (in CCl^ solution) singlet, CH^O- AA'BB* pattern one half of an AB quartet one half of an AB quartet In a 250 ml polyethylene | bottle was placed 1.96 g (0.010 mole) of j>-methoxycinnamoyl I chloride and 50 ml of carbon tetrachloride. A long poly- ; ethylene tube was connected to the top of the bottle for | ! removal of the hydrogen chloride and excess hydrogen | | fluoride. The mixture was cooled to 0° in a ice-water | bath. Hydrogen fluoride (about 2 g) was introduced through | the inlet polyethylene tube ovor a period of 1.5 hours | I while keeping the reaction mixture magnetically stirred. i | The reaction mixture was then allowed to warm to about j 40° for 20 minutes. Additional carbon tetrachloride (20 . n i l ) 3.90 ppm (6), 3H, 7.25 ppm (6), 4H, 6.45 ppm (6), 1H, 7.80 ppm (6), 1H, Ttrans „ - „ J = 15*5 Hz. AB I E-Methoxycinnamoyl fluoride. 117 was added to the solution. The reaction mixture was then filtered, and the carbon tetrachloride was removed under vacuum (0.4 mm). The yield was 1.2 g (66$), Analysis of nmr spectrum indicated that no detectable amount of impurity was present. Spectral data for j>-methoxycinnamoyl fluoride* I,R* vmax* 29^° (broad)» 2830 (CH3O-), 1792 (C=0), 1624, 1600 (C=C), 1505. 1420, 1252, 1189, 1170, 1100, 1033# 980, 828 cm"1. 100 MHz 1H nmr (in CCljj, solution) COF 3.80 ppm (6), 3H, singlet 6.05 ppm (6), 1H, doublet of one half of an AB quartet 7.68 ppm (6), 1H, one half of an AB quartet 7.15 ppm U)» 4h, aromatic AA'BB* pattern J12 = 6,30 Hz J23 ” l6.0 Hz 118 19 The F chemical shift of E-methoxycinnamoyl fluoride is a doublet 24.03 ppm downfield from fluorotrichloromethane in 30% acetone and 70% vinyl chloride solution, with 3jhf = 6,80 Hz* REFERENCES 1. (a) E. L. Eliel, "Stereochemistry of Carbon Com pounds", McGraw-Hill Book Co., Inc., New York, 1962, p. 133. (b) E. L. Eliel, N. L. Allinger, S. J. Angyal, and G. A. Morrison, "Conformational Analysis", Interscience Publishers, Inc., New York, N. Y., 1965i p. 1. 2. J. D. Kemp and K. S. Pitzer, J. Chem. Phvs., 4. 7^9 (1936). 3. E. B. Wilson, Jr., Advan. Chem. Phvs.. 2, 367 (1959). 4. D. J. Millen, Prog, in Stereochem., 3» 138 (1962), P. B. D. de la Mare and w. Klyne, eds., Butterworths Inc., Washington, D. C. 5. J. Dale, Tetrahedron. 22, 3373 (1966). 6. J. P. Lowe, Prog. Phvs. Org. Chem., 6, 1 (1968). 7» R. A. Pethrick and E. Wyn-Jones, Quart. Rev., 23. 301 (1969). 8. E. L. Eliel, N. L. Allinger, S. J. Angyal, and G. A. Morrison, "Conformational Analysis", Interscience Publishers, Inc., New York, N. Y., 1965» P« 1* 9* (a) K. S. Pitzer, J. Chem. Phvs., 8, 711 (19^0)1 (b) G. J. Szasz, N. Sheppard, and D. H. Rank, ibid., 16. 70^ (19^8); (c) R. A. Bonham and L. S. Bartell, J. Am. Chem. Soc.. 81, 3^91 (1959). 10. E. L. Eliel, N. L. Allinger, S. J. Angyal, and G. A. Morrison, "Conformational Analysis", Interscience Publishers, Inc., New York, N. Y., 1965* P» 9. 11. E. L. Eliel, N. L. Allinger, S. J. Angyal, and G. A. Morrison, "Conformational Analysis", Interscience Publishers, Inc., New York, N. Y., 1965, pp. 13-19. 12. R. W. Kilb, C. C. Lin, and E. B. Wilson, J. Chem. Phys.. 26, 1695 (1957). 13. D. R. Herschbach and L. C. Krischer, J. Chem. Phys., 28, 728 (1958). 14-. R. J. Abraham and J. A. Pople, Mol. Phys., 2* ^09 (I960). J19 ____________ 120 15* J. K. Wilmhurst, J. Mol. Spectr.. 1^ 201 (1957) i T. Miyazawa, Bull. Chem. Soc. Japan. 34, 691 (1961). 16. A. A. Bothner-By, C. Naar-Colin, and H. Gunther, J. Am. Chem. Soc.. 84. 2748 (1962). 17* K. Kuchitsu, T. Fukuyama, and Y. Morino, J. Mol, Structure, i., 463 (1967)* 18. M. J. S. Dewar and H. N. Schmeising, Tetrahedron, 5, 166 (1959). 19. R. S. Mulliken, ibid. 6, 68 (1959). 20. H. Mackle and L. E. Sutton, Trans. Faraday Soc.. 47. 691 (1951), and references therein. 21. A. Almenningen, 0. Bastiansen, and M. Taraetteberg, Acta Chem. Scand.. 12, 1221 (1958). 22. D. J. Marais, N. Sheppard, and B. P. Stoicheff, Tetrahedron. 17. 163 (1962). 23. R. A. Beaudet, J. Chem. Phys., 42, 3758 (1965)• 24. D. R. Lide, Jr., J. Chem. Phys.. 2£, 2074 (1962). 25. D. R. Lide. Jr. and M. Jen, J. Chem. Phys.. 40. 252 (1964). “ 26. H. Wynberg, A. De Groot, and D. W. Davies, Tetrahe dron Letters. 1083 (1963). 27. G. J. Szasz and H. Sheppard, Trans. Farad. Soc.. 49, 358 (1953). 28. A. A. Bothner-By and D. Jung, J. Am. Chem. Soc.. 90. 2342 (1968). 29. A. A. Bothner-By and D. F. Koster, J. Am. Chem. Soc.. 20, 2351 (1968). 30. K. L. Servis and J. D. Roberts, J. Am. Chem. Soc.. 87. 1339 (1965). 31. S. L. Manatt and M. T. Bowers, J. Am. Chem. Soc.. 91. 4381 (1969). 32. R. G. Parr and R. S. Mulliken, J. Chem. Phvs.. 18. 1338 (1950). 121 : 33* J. C. D. Brand and D. G. Williamson, Disc. Faraday Soc.. 3£» 184- (1963). 34. J. M. Hollas, Spectrochim. Acta., 19. 1425 (1963). 35* R. K. Harris, Spectrochim. Acta.. 20, 1129 (196*0. 36. E. A. Cherniak and C. C. Costain, J. Chem. Phys., 4j>, 104 (1966). j 37« M. Traettebert, Acta. Chem. Scand.. 24, 373 (1970). ; 38* M. Suzuki and K. Kozima, Bull. Chem. Soc. Jap., 42, 2183 (1969). 39. F. A. L. Anet and M. Ahmad, J. Am. Chem. Soc., 86, 119 (1964). 40. R. E. Klinck, D. H. Marr, and J. B. Stothers, Chem. Commun.. 409 (1967). 41. K. -I. Dahlavist and S. Forsen, J. Phys. Chem., 69. 4062 (1965). 42. D. F. Koster, J. Am. Chem. Soc., 88. 5067 (1966). 43. w. S. Brey, Jr. and K. C. Ramey, J. Chem. Phys., 39. 844 (1963). 44. J. E. Katon and W. R. Feairheller, Jr., J. Chem. Phys., 4Z, 1248 (1967). 45. G. L. Carlson, W. G. Fateley, and R. E. Witkowski, J. Am. Chem. Soc.. 89. 6437 (1967). 1 46. J. J. Keirns and R. F. Curl, Jr., J. Chem. Phys., 48, ! 3773 (1968). 47. W. G. Fateley, R. K. Harris, F. A. Miller, and R. E* Witkowski, Spectrochim. Acta.. 21, 231 (1965). 48. L. Radom and J. A. Pople, J. Am. Chem. Soc., 92, 4?86 (1970). 49. L. Pauling, "The Nature of the Chemical Bond", 3rd ed., Cornell University Press, Ithaca, New York, I960, pp. 130-142, 292. i 50. G. Binsch, "Topics in Stereochemistry", Vol. 3* 97 (1968), E. L. Eliel and N. L. Allinger, eds., Inter science, New York. 122 51* J« D. Memory, "Quantum Theory of Magnetic Resonance Parameters", McGraw-Hill, inc., New York, 1968. 52. J. W. Emsley, J. Feeney, and L. H. Sutcliffe, "High Resolution Nuclear Magnetic Resonance Spectroscopy", Vols. 1 and 2, Pergamon Press, Oxford, 1965* 53. P. L. Corio, "Structure of High Resolution NMR Spec tra", Academic Press, New York, 1966. 54. E. W. Garbisch, J. Chem. Educ.. *4£, 311 (1968)• 55* M. Lynden-Bell, in "Progress in NMR Spectroscopy", J. W. Emsley, J. Feeney and L. H. Sutcliffe, eds., Vol. 2, p. 163, 1967, Pergamon Press, Oxford, 56. a) G, Binsch, J. Am. Chem. Soc.. 91. 130*4- (1969)* b) G. Binsch and D. A. Kleier, "Quantum Chemistry Program Exchange", No. 1*4-0, DNMR Computer Program, Chemistry Department, Indiana University, Blooming ton, Indiana. 57. a) D. F. DeTar, J. Chem. Educ.. 44, 759 (1967)* b) D. F. DeTar, "Quantum Chemistry Program Exchange", No. 79, ACTENG Program, Chemistry Deaprtment, Indiana University, Bloomington, Indiana. 58. M. Hudlicky, "Chemistry of Organic Fluorine Com pounds", MacMillan Co., New York, 1962. 59. W. Brugel, Z. Elektrochem.. 64, 11*49 (I960), 60. G. A. Olah and S./J. Kuhn, "Organic Syntheses", Vol. * 45, p. 3, 1965» John Wiley & Sons, Inc., New York. 61. G. A. Olah and M. B. Comisarow, J. Am. Chem. Soc., 69. 269*4 (1967). 62. C. Rappe, Arkiv. Kemi.. 21. 503 (1963). 63. a) E. I. Snyder, J. Am. Chem. Soc.. 85, 262*4 (1963)1 b) W. L. Meyer et al., J. Am. Chem. Soc.. 87. 1573 (1965). 6*4. A. Kende, "Organic Reactions", Vol. 11, p. 278, i960, R. Adams et al. eds., John Wiley & Sons, Inc., New York. 65. C. Rappe, Acta Chem. Scan.. 17. 2766 (1963). 123 66. a) A. A. Bothner-By and S. M. Castellano, J. Chem. Phys., 41., 3863 (1964)* b) A. A. Bothner-By and S. M. Castellano,"Quantum Chemistry Program Exchange", No. Ill, LAOCOON III, Chemistry Department, Indiana University, Bloomington, Indiana. ! 67. M. Karplus, J. Am. Chem. Soc., 85, 2870 (1963), i ; 68. R. T. Hobgood and J. H* Goldstein, J. Mol. Spectrosc.. 12, 76 (1964). 69. A. A. Bothner-By and E. Moser, J. Am. Ghem. Soc., 90, 2347 (1968). 70. K. L. Williamson, Y. F. Li, F. H. Hall and S. Swager, J. Am. Chem. Soc.. 88, 5678 (1966), 71. A. A. Bothner-By and R. K. Harris.oJ. Am. Chem. Soc., 82, 3445 (1965). 72. V. I. P. Jones and J. A. Ladd, J. Chem. Soc., B, 1179 (1970). 73* T« Miyazawa and K. S. Pitzer, J. Chem. Phys., 30. 1076 (1959). 74. F. A. Miller, W. G. Fateley, and R. E. Witkowski, Spectrochim. Acta, 23A. 891 (1967). 75. H. S. Taylor and S. Glasstone, "A Treatise on Physi cal Chemistry", Vol. 1, 1942, D. Van Nostrand Co., New York. 76. I* Omura, K. Higashi, and H. Baba, Bull. Chem. Soc. Ja£.. 2£, 504 (1956). 77* G. Berthier and J. Serre, "The Chemistry of the Carbonyl Group", Interscience Pub., New York, 1966, p. 16. i | 78. D. Cook, J. Am. Chem. Soc.. 80, 49 (1958). I 79. P. D. Foster, V. M. Rao, and R. F. Curl, Jr., J. Chem. | Phys.. 42, 1064 (1965). I 80, L. Pauling, "The Nature of the Chemical Bond", 3rd ed., Cornell University Press, Ithaca, New York, ! i960, p. 260, 124 81. J. G. Aston, G. Szasz, H, W. Woolley, and F. G. Brickwedde, J. Chem. Phys,, 14, 67 (1946), j t ! 82. N. L. Allinger, M. A. Miller, F* A. Van Catledge, and! J. A. Hirsch, J. Am. Chem. Soc., 89. 4345 (1967). j 83. J. B. Hendrickson, J. Am. Chem. Soc.. 89. 7036 (1967). 84. F. L. Pilar, J. Chem. Phys.. 4£, 5375 (1967)1 for results of simple Huckel MO calculations, see C. A. Coulson and A. Streitweiser, Jr., "Dictionary of n-Electron Calculations", W. H. Freeman and Co., San Francisco, California, 1965. 85. H. S. Gutowsky, G. G. Belford, and P. E. McMahon, J. Chem. Phvs.. J6, 3353 (1962). 86. G. Govil and H. J. Bernstein, J. Chem. Phys.. 47. 2818 (1967). 87. R. A. Newmark and C. H. Sederholm, J. Chem. Phvs.. 22, 3131 (1963). 88. K. B. Wiberg, "Laboratory Techniques in Organic Chemistry", McGraw-Hill Book Co., New York, i960, p. 86. APPENDIX 1 The Relationship Between the Conformations of the Metastable Form of the Conjugated System and the Magnitude of Vlt Vg, and V^ The potential function V(©) describing the internal rotation of a conjugated system may be expanded as a Fourier cosine series. where 9 is the angle of torsion between the two conjugated double bonds. It is generally true that the series can be terminated after three terms, since the higher terms (more than four-fold rotational barrier) are usually very small. | We have I j ! V i v2 v * 3 ; V(0) = -1 (1 - cos 9) + £ (1 - cos 2 9) + (1 - cos 3©). C L £ + ( 1) (2) In order to find the 9__„ (at the transition state) and | max j ©min (a-t the metastable form), one sets the derivative of V with respect to 9 to zero. i or i ( sin 9 + 2 V2 sin 29 + 3 sin 3©) = 0 125 _________ (3) 126 i ! Equation (3) can be transformed to | 2 ; sin © (12V3 c o s © + * * - V2 cos © + V1 - 3V^) = 0. (4) ' One of the solutions is i i J sin © = 0 or © = Oj * . Note that 0 = 0 corresponds to the s-trahs conformation, and © = * corresponds to the s-cis conformation* The sign of second derivative of V with respect d2V to 0, ( — 5- ), will determine whether the potential at the d© specific angle of torsion is at maximum or minimum. The d2 v : conditions are VQ at minimum when ( ) has positive * d©2 9 d2y value, and V at maximum when ( — 7 ? has negative value. ; 9 d© d2V From Eq. (2), ( — 5- ) can be derived as d© i d2V ! (---r~) = £ (V, cos © + cos 2© + 9V~ cos 30). (5) d©2 1 2 J d2V It is clear that when © = 0 (s-trans), {--~ is -------d©2 0 positive, the s-trans form is at minimum as defined. However, when © - w (s-cis), 12? ) = i (-Vi + 4v, - 9V,) (6) de2 i c j only if -v1 + ^V2 - 9V3 > 0, or 4v2 > Vx + 9Vy the s-cis conformation is at potential minimum (the metastable form). The angle of torsion at the transition state (© ), can be calculated from the second part of Eq. (4). max One obtains _ -v2 + J v 2z - 3V3 (Vi - 3V,) (7) cos max ---------------------- 6V3 If + 4-Vg - 9V^< 0, or 4V2< + 9V3, then the s-cis form is at potential maximum, and the metastable form has to be in skew conformation. The angle of torsion at the potential minimum (®min) other than the s-trans conformation (© = 0) can be calculated by the following equation: -V2 - cos ®min V22 ~ 3V3 (Vi - 3V3) 6V3 while ©„QV (at transition state) can again be calculated lucLX from Eq. (7).

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

I. Nuclear Magnetic Resonance (Nmr) Studies Of Methanol - Boron-Trifluoride Complexes. Ii. Competing Nucleophilic Substitution Processes In The Solvolysis Of Cyclopropylcarbinyl Sulfonates

PDF

The Stereochemical Studies On The Alpha-Phenyl Beta-Methylvinyl And The Alpha-Chlorocyclopropyl Radicals

PDF

Conformational Analysis Of Cis, Cis-1,6-Cyclodecadienes By Nuclear Magnetic Resonance Techniques

PDF

Nuclear Magnetic Resonance Studies Of Long-Range Fluorine Spin Interactions

PDF

Nucleotides And Enzyme Regulation: Cooperativity In Phosphorylase B And Cyclic Nucleotide Phosphodiesterase

PDF

The Microwave Spectra Of Methylcyclopropane And 2,2-Difluoro-1-Methylcyclopropane

PDF

Studies On The Effects Of The Plant Growth Substances On Algae

PDF

Synthesis And Biological Evaluation Of Some Phenylureido Thiazolylmethoxyamines With Potential Antiinflammatory Activity

PDF

Stress Relaxation Behavior In The Primary Transition Region

PDF

Electric Field Induced Spectra Of Propynal And Other Molecules

PDF

The Microwave Spectra Of 1,1-Difluoro-3-Methyl Butadiene And 1-Methyl Cyclobutene

PDF

Absorption-Spectrum Of Oxygen In The Even Singlet 0A-Delta Metastable Stable In The Region From 1090 To 1700 Angstroms

PDF

Studies on acyl transfer reactions on n-acylimidazoles

PDF

Intermolecular Forces, Exciton Splittings And Lattice Vibrations Of Crystalline Pyrazine

PDF

Kinetic investigation of the hydrolysis of aryl β-D-glucopyranosiduronic acids by β-glucuronidase

PDF

Sedimentary History Of The Early Pliocene In The Los Angeles Basin, California

PDF

Enzymatic dithiol criteria

PDF

Structural Studies Of Selected Boron-Carbon Compounds

PDF

Basic studies in the analytical chemistry of beryllium

PDF

The Chemistry Of Some Pentaborane(9) Adducts

Asset Metadata

Creator Lin, Frank Fang-Sheng (author)

Core Title Nuclear Magnetic Resonance Studies Of The Rotational Isomerism In (Alpha,Beta)-Unsaturated Acyl Fluorides

Contributor Digitized by ProQuest (provenance)

Degree Doctor of Philosophy

Degree Program Chemistry

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag chemistry, organic,OAI-PMH Harvest

Language English

Advisor Servis, Kenneth L. (committee chair), Beaudet, Robert A. (committee member), Fife, Thomas H. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-546313

Unique identifier UC11362341

Identifier 7203786.pdf (filename),usctheses-c18-546313 (legacy record id)

Legacy Identifier 7203786

Dmrecord 546313

Document Type Dissertation

Rights Lin, Frank Fang-Sheng

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA