CARBONUEE MAGNENC RESQNANCE OF
SOME ALWHAHC AME PHENYL CQMPOUNDS

AND $QW‘ENT swag-Es Q‘F fiakfi‘i ETHYL
CGMPQUMDS USING E’ROTQN
MAGNET“: RESQNAMCE

Thesis for Hm Degree 0? pk. D.
MICHIGAN STATE UNIVERSITY
Roy Lee Fciey

1964-

_.

“IBRARY ‘

‘ Michigan 5m
H University
L —

 

.3 ‘

£13....

MIC!‘IIQTT‘~I S'TATE L'L’IVERSITY
DEPARTMENT OF: CI-IEIVIISTRY
EAST LANSING, MICHIGAN

A BST RACT

CARBON- 13 MAGNETIC RESONANCE OF SOME ALIPHATIC
AND PHENYL COMPOUNDS AND SOLVENT STUDIES OF
SOME ETHYL COMPOUNDS USING PROTON
MAGNETIC RESONANCE

by Roy Lee Foley

The carbon-l3 magnetic resonance spectra of several aliphatic,
phenyl, diphenyl, and triphenyl compounds were obtained using a
Varian high-resolution nuclear magnetic resonance spectrometer
operating at a fixed radiofrequency of 15. 085 Mcps. The carbon-13
spectrawere obtained for all compounds in natural abundance using
15 mm. external diameter sample tubes. The low natural abundance
of the carbon-13 isotOpe and its longer relaxation times necessitated the
use of rapid passage dispersion mode conditions to obtain spectra.

The solvent studies on the ethyl compound were performed by using a
Varian A'-60 spectrometer to obtain the proton spectra.

The carbon-13 chemical shifts were found to vary over a wide
range in contrast to fluorine and proton chemical shifts. The carbon- 13
chemical shifts are interpreted in terms of inductive and anisotropic
effects and are shown in many cases to correlate quite well with atomic
or group electronegativities. In addition, the correlation of carbon- 13
chemical shifts with molar susceptibility data, supports the contention
of Pople that the paramagnetic contribution to the chemical shift is the
dominant contribution for carbon-13 shifts. The carbon-l3 proton and
carbon-13 fluorine spin- spin coupling constants have been recorded for

all compounds studied. In the o. -substituted toluenes very good

Roy Lee Foley

agreement between the spin-spin coupling constants calculated using
zeta values and measured coupling constants was observed.

A study of proton n.m. r. spectra of a series of ethyl compounds
M(CZH5)n in various solvents indicates that inter- and intramolecular

interactions for solute and solvent are absent.

CARBON-13 MAGNETIC RESONANCE OF SOME ALIPHATIC
AND PHENYL COMPOUNDS AND SOLVENT STUDIES OF
SOME ETHYL COMPOUNDS USING PROTON
MAGNETIC RESONANCE

BY

Roy Lee Foley

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

1964

ACKNOWLEDGMENTS

The writer wishes to express his sincere appreciation
to Professor Max T. Rogers for the supervision, encourage-
ment, and suggestions given during the formation of this
thesis. He also wishes to express his gratitude to the
National Science Foundation and the Atomic Energy Com-
mission for grants subsidizing this research.

>:< * >;< >1< >:< >:< >:< >{z >:< >:< >:< >',< >;< >:< >:< ):<

ii

TABLE OF CONTENTS

INTRODUCTION . . . . . . . . .

HISTORICAL REVIEW 0 O O 0 O O O O O O 0

THEORY O O O O O O O O O O O

Lamb's Theory of Chemical Shifts
Ramsey' 3 Theory of Shielding Constants .

An Approximate Theory of Chemical Shifts. .

A General Molecular Orbital Theory of Chemical

Approximation A
Approximation B
Approximation C
Approximation D
Approximation E
Approximation F

Atomic Contributions to Chemical Shifts '

Molecules.............
Carbon-Carbon Single Bonds .
Carbon-Carbon Double Bonds.
Carbon-Carbon Triple Bonds .
Allenes........... .

Ca rbonyl Groups .

Nuclear Spin-Spin Interaction . .

Nuclear Relaxation . . . . . . . . .
LineShapes.............
Line Broadening. . . . . . . . . . .

Line Asymmetry. . . .
Saturation . . . . . . .
Double Irradiation. . .

EXPERIMENTAL . . . . . .

Spectrometers. . . . .
Compounds Studied . .

0

iii

0 O O O

Shifts .

Page

16
16
16
17
18
18
21
22
23
26
28
29
31

33

33
34

TABLE OF CONTENTS - Continued

Determination of Spectral Parameters . . . . . . . . . . . 38
Proton Magnetic Resonance. . . . . . . . . . . . . . 38
Carbon-13 Magnetic Resonance. . . . . . . . . . . . 38

RESULTS....... ...... ....... 41

Effects of Solvent on the N.M. R. Spectra of Some Organo-
metallic Compounds . . . . -. . . . . . . . . . . . . 41
Substituted Acetic Acids . . . . . . . . . . . . . . . . . . . 44
Some Fluorinated Aliphatic Compounds .......... 55
a-Substituted Toluenes . . . . . . . . . . . . . . ..... 66
Diphenyl Compounds . . . . . . . . . . . . . . . . . . . . 86
Triphenyl Compounds. . . . . . . . . . . . . . . . . . . . 93

A1cohols..... ...... ................109
DISCUSSION OF RESULTS. . . . ................. 122
Solvent Studies of Ethyl Compounds ...... . . . . . 122

Correlation of Chemical Shifts with Molar Susceptibilities 123
Correlation of Chemical Shifts with Electronegativity . . . 123

SUMMARY. 0 O O O O O O O ...... O I O 0 O 0 0 0 O O O O O O 126

BIBLIOGRAPHY. . . . . . ....... . . . . . . . . ..... 127

iv

LIST OF TABLES

TABLE Page
I. Physical Constants of Alipahtic Compounds Studied. . . 35

II. Physical Constants of u-Substituted Toluenes Studied. . 36

III. Physical Constants of Diphenyl Compounds Studied . . . 36

IV. Physical Constants of Triphenyl Compounds Studied . . 37

V. Solvent Effects on the Carbon-13 Chemical Shifts of
CarbonDisulfide..................... 40

VI. Proton Chemical Shifts in Ethyl Compounds . . . . . . 41
VII. Spin-Spin Coupling Constants of Pb(CHZCH3)4 in CC14 . . 42
VIII. Spin-Spin Coupling Constants of Sn(CHZCH3)4 . . . . . . 42

IX. 15.1 Mc./sec. Carbon—13 Spectra of Some Substituted
AceticAcids........................ 54

X. 15.1 Mc./sec. Carbon-13 Spectra of Some Fluorinated
AliphaticCompounds................... 69

XI. Carbon-13 Chemical Shifts and Spin-Spin Coupling Con-
stants of Some Aliphatic and Fluorinated Aliphatic
Compounds................‘........ 71

XII. 15.1 Mc./sec. Carbon-13 Spectra of Some Substituted
TOIueneSO . O O O O O O O O O O O 0 O O O O O O O O 0 O O O 90

XIII. 15.1 Mc./sec. Carbon-13 Spectra of Some Diphenyl
Compounds........................ 97

XIV. Chemical Shifts (ppm.) and Spin-Spin Coupling Con-
stants (cps.) for Some Diphenyl Compounds. . . . . . . 98

LIST OF TABLES - Continued

TABLE

XV.

XVI.

XVII.

XVIII.

Page
15.1 Mc./sec. Carbon-13 Spectra of Some Triphenyl
Compounds. . .. . . . . . . .. . . . . . . . 107
Chemical Shifts (ppm.) and Spin-Spin Coupling Con-
stants (cps.) of Some Triphenyl Compounds ...... 108

Chemical Shifts for Some Aliphatic Alcohols . . . . . . 110

Calculated and Observed Spin-Spin Coupling Constants

in Substituted Toluenes. . . . . . . . . . . . . . . . . 111

V1

LIST OF FIGURES

FIGURE

. The shapes of n.m. r. signals: (a) Slow passage absorp-

tion, (b) Slow passage dispersion, (c) Rapid passage

dispersion......... .....
15-1Mc./sec. carbon-13 spectra of benzene . . . . .

Internal chemical shifts versus mole fraction of solute
forethylcompounds. . . . . . . . . . . . . .. . . . .

15.1 Mc./sec. carbon-13 spectra of trifluoroacetic

aCid. O O O O O O O O O O O I O O O O O O O O O O O O 0

15.1 Mc. /sec. carbon-13 spectra of difluorochloro-
acetic a'CidO 0 O O 0 O O 0 O 000000 O O I O O O O O

15.1 Mc./sec. carbon-13 spectra of fluorodichlor'o-
acetic acid OOOOOOOOOOOO O O O O O O O O O 0

15.1 Mc./sec. carbon-13 spectra of trichloroacetic

aCi-d. O O O O O I O O O O 0 O O O O O O O O O O O O O O O O

15.1 Mc. /sec. carbon-13 spectrum of dichloroacetic

aCidO O O O O O O O O O O O O O O O O G O O O O 0 O O O

15.1 Mc. /sec. carbon-13 spectrum of dichloroacetic

aCido O O O O O O O O O O O O O O O O 0 O O O O O O O O

15.1 Mc./sec. carbon-13 spectra of chloroacetic acid .

15. 1 MC. /sec. carbon-13 spectrum of phenylacetic
aCid. O I O O O O O 0 O O O O O O 0 O O O O O I 0 O O O

15. 1 Mc./sec. carbon~l3 spectrum of phenylacetic

aCidO O O O O O O 0 O O O O O O O 0 O O O O 0 O O O 0 0

vii

Page

27

3O

43

45

46

47

49

50

50

51

52

53

LIST OF FIGURES - Continued

FIGURE

11(a).

11(b).

12(a).
12(b).

13(a).

13(b).

14.

15(a).

15(b).

16(a).

16(1)).

17(a).

17(b).

18(a).

18(b).

15.1Mc./sec. carbon-13
anhydride. . . . . . . . .

15.1Mc./sec. carbon-l3
anhydride. . . . . . . . .

15.1Mc./sec. carbon-13
15.1Mc./sec. carbon-13

15.1Mc./sec. carbon-13
acetylacetone. . . . . . .

15.1Mc./sec. carbon-13
acetylacetone. . . . . .

15.1Mc./sec. carbon-13

l, 1, 2, 2-tetraf1uoroethane .

15.1Mc./sec. carbon—13
ethanol..........

15.1Mc./sec. carbon-13
ethan01.. . .. . .. . .

15.1Mc./,sec. carbon-13
2-propanol . . . . . . .

15.1Mc./sec. carbon-13
Z-pl‘OpanOI o o o o o o o o

15.1Mc./sec. carbon-13
2-methy1-2-propanol . .

15.1Mc./sec. carbon-13
2-methyl-2-propanol. . .

15.1Mc./sec. carbon-13

15.1Mc./sec. carbon-l3

spectrum of t rifluoroac etic

spectrum of trifluoroacetic

spectrum of acetylacetone. .
spectrum of acetylacetone.

spectrum of hexafluoro-

spectrum of hexafluoro-

spectra of 1, 2, -dibromo-

spectrum of 2, 2, 2-trifluoro-

spectrum of 2, 2, 2-trif1uoro-

spectrum of 1,1, l-trifluoro-

Spectrum of 1, 1, 1-trif1uoro-

O O O O 0 O O O O 0 O O O 0 O

spectrum of 1, 1, latrifluoro-

spectrum of 1, 1, 1-trif1uoro-

O O O O O O 0 ° 0 0 0 O O O 0 O

spectrum of toluene. . . . .

spectrum of toluene. . . . .

viii

Pace

56

56

57

58

59

6O

62

63

64

65

65

67

68

73

74

LIST OF FIGURES —' Continued

FTGURE

19(a).
19Ib).
20(a).
20 (b) .
21(a).
21(b).
22(a).
22(b).
23(a).
23(b).
'24(a).
24(b) .
25.
26.
27.
28.
29.
30.
31.

32.

15.

.1

1

15.1

15.1

15.1

15.1

15.1

.1

15.1

.1

h4c./sec.
L4c./sec.
LAc./sec.
hAc./sec.
bAc./sec.
hAc./sec.
L4c./sec.
h4c./sec.
LAc./sec.

/sec.

h4c./sec.

LAC.

hAc./sec.
h4c./sec.
hAc./sec.
L4c./sec.
L4c./sec.
bAc./sec.
bAc./sec.
/sec.

LAc./sec.

hdc.

carbon—13

carbon—13

carbon-13

carbon-13

carbonu13

carbon~13

carbon-13

carbon-13

carbon-13

carbon—13

carbon—13

carbon-13

carbon-13

carbon-13

carbon-13

carbon-13

carbon-13

carbon-13

carbon~13

carbon-13

spectruni

spectrunn

spectruni

spectrun1

spectrun1

spectrunn

spectruni

spectrunn

spectrun1

spectrunn

spectrun1

spectruni

Page

of benzylbromide. .
of benzylbromide. .
of benzylchloride. .
Cd benzylchloride. .
ofbenzyhnercapuni
ofbenzyhnercapuni
of diphenylmethane.
of diphenylmethane.
of benzylcyanide . .
of benzylcyanide . .
of benzylamine. . .

of benzylamine . . .

spectra of benzylalcohol . . .

spectra

spectra

spectra

spectra

spectra

spectra

spectra

ix

of diphenylether . . .
of diphenylsulfide . .
of diphenylselenide. .
ofchphenyhnercury

of triphenylamine . .

of triphenylarsine . .

75

76

77

78

80

81

82

83

84

85

87

88

89

92

94

95

96

100

of triphenylphosphine 101

102

LIST OF FIGURES - Continued

FIGURE

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

15.1 Mc./sec. carbon-13 spectra of triphenylstibine. .

15.1 Mc./sec. carbon- 13 spectra of triphenylbis-
mUthine. . O O O O O O I O O O O O I O O O O O O O O O O O

15. 1 Mc. /sec. carbon- 13 spectra of triphenylmethane .

Chemical shifts versus molar susceptibilities for some
aliphaticalcohols.....................

CX3 chemical shifts versus molar susceptibilities in
some substituted acetic acids . . . . . . . . . . . . . .

Chemical shifts of carbonyl carbon atoms versus molar
susceptibilities for some substituted acetic acids. . . .

The aliphatic carbon atom chemical shifts versus
molar susceptibilities insome phenylmethanes. . . . .

Chemical shifts of ortho carbon atoms versus molar
susceptibilities for some diphenyl and triphenyl com-
pounds. 0 O I O I O O O I O O. O O O O O O O O O I O O O O

CX3 chemical shift versus group electronegativity for
a series of substituted acetic acids. . . . . . . . . . .

Chemical shifts of carbonyl carbon atoms versus group
electronegativity for a series of substituted acetic

aCids. O O O O O O O O O O O O O O O I I O O O O O O O O O

CHZX chemical shift versus group electronegativity in
substitutedtoluenes. .. . .. . . ...... .... .

Chemical shifts of ortho carbon atoms versus Pauling's
electronegativity of central atoms for some diphenyl
and triphenyl compounds. . . . . . . . . . . . . . . . .

Page

104

105

106

113

114

115

116

117

118

119

120

121

INTRODUCTION

The applications of nuclear magnetic resonance spectroscopy
to the study of molecular structure are now well established.
However, it has only recently become possible to successfully apply
this technique to the carbon- 13 nuclei in natural abundance. The
long thermal relaxation times (T1) and low natural abundance of the
carbon- 13 isotope have tended to discourage the initiation of a compre-
hensive study of carbon-13 resonances in its compounds. The one
per cent natural abundance of the carbon-13 isotope is just sufficient
to allow its effects to be detectable in the proton and fluorine spectra
of simple molecules and, more important to permit the carbon-13
spectra themselves to be observed. The measurement of peak positions
and spin-spin coupling constants may provide a useful aid in the analysis
and understanding of molecular structure.

The research reported in this thesis consists of the following:
(1) an examination of the carbon-13 magnetic resonance spectra of
several aryl derivatives in order to ascertain which features of the
observed spectra may be correlated with known atomic and molecular
parameters, (2) an examination of the carbon-13 magnetic resonance
spectra of several fluorinated aliphatic compounds in order to determine
the effect of progressively replacing protons by fluorine atoms, and
(3) an examination of the proton magnetic resonance spectra of some
ethyl derivatives in carbon tetrachloride, cyclohexane, and chloroform
solution to determine possible effects of solvent on the internal chemi-

cal shifts and spin- spin coupling constants.

HISTORICAL REVIEW

Most of the nuclear magnetic resonance (n. m. r.) data reported
in the literature to date have been concerned with nuclei whose
natural abundances are approximately one hundred per cent, mainly
hydrogen, fluorine, phosphorus, and boron. The commercial avail-
ability of sensitive high- resolution instruments has now made it
possible to observe carbon—13 satellites in hydrogen (1, 2) and fluorine
(3,4) spectra.

The first observation of carbon-13 spectra in natural abundance
was reported by Lauterbur (5). Employing a transmitter frequency
of 8. 5 Mc/sec. and a stationary magnetic field of 7940 gauss, Lauterbur
was able to obtain the carbon-13 chemical shifts and hydrogen-carbon-13
spin- spin coupling constants of several simple aliphatic compounds.
From this study Lauterbur concluded that the spin- spin coupling
between carbon- 13 and directly bonded hydrogen is comparable in
magnitude to that between hydrogen and other nuclei, and tends to in--
crease with decreasing magnetic shielding of the proton. Holm also
reported some of the earlier carbon-13 chemical shifts and spin-spin
coupling constants (6). The results of Holm essentially substantiated
the conclusions of Lauterbur.

Some comprehensive natural abundance studies of carbon- 13
shielding in aromatic systems have been made. In a study of aromatic
hydrocarbons (7) it has been found that the variations in pi-electron
densities are primarily responsible for the variations in the chemical
shifts in aromatic systems. Lauterbur (8) has concluded from a study
of phenols, anisole and dimethoxybenzenes that the methyl effects on

carbon-13 chemical shifts and coupling constants are relatively small.

2

The correlation of carbon-13 chemical shifts with electronegativity for
substituted aliphatic and aromatic compounds have been well illus-
trated by Lauterbur (9).

The analysis of carbon-13 spectra has been accomplished in
three ways. Lauterbur (7e12,) has applied the method of methyl substi-
tution to the analysis of spectra of benzene derivatives. Spiesecke
and Schneider (13, 14) have taken advantage of the fact that when deuterium
is substituted on a carbon in the benzene ring no peak will appear for
that carbon. Double irradiation methods have been applied by Friedel
and Retcofsky (15) in analyzing the carbon-13 spectra of olefins and
other hydrocarbons.

Karplus and POple (16) have recently published a theoretical treat-
ment of carbon-13 n.m. r. chemical shifts in conjugated molecules.

The empirical relation

ACT-A = QIPA'I)

suggested by Lauterbur (7), and by Spiesecke and Schneider (14), has
been derived using molecular orbital theory. AU'A is the chemical shift.
with reference to benzene, (1 is a positive constant that is approximately

160 ppm. , and P is the pi electron density. In addition, the shielding

A
is shown to be a function of the free valence of the atom under consider-
ation and of the polarity of the sigma bonds.

Savitsky and Namikawa (87) have shown that carbon- 13 chemical
shifts in CH3X, CZH5X, i- C3H7X, and t-C4H9X compounds are additive.
Their results indicate that the carbon-13 chemical shift toward lower
field per methyl substitution depends very little on the nature of X for
fi-carbons but vary considerably for a-carbons. Additivity relations of

the carbon-13 chemical shifts in para-disubstituted benzenes have also

been found by Savitsky (88). These relations are helpful in the

assignment of the carbon-13 n.m. r. spectra of polysubstituted ben-
zenes. Savitsky and Namikawa (89) have used carbon-13 n.m. r. to
investigate geometric isomerism about the ethylenic double bond.

It has been concluded that the carbon- 13 chemical shift differential
between cis and trans carbons could not be satisfactorily explained
solely in terms of long range anisotropy effects. An alternative
explanation involving possible deviation from coplanarity of bulky cis

substituents was pre sented.

THEORY

Lamb's Theory of Chemical Shifts

Chemical shifts have their origin in the magnetic screening of
the nucleus which arises from the orbital electronic currents induced
by an external magnetic field. These currents also give rise to
diamagnetic polarization (17). When the external magnetic field is

H the total magnetic moment of the induced current is XmHo where

o’
Xm is the molecular diamagnetic susceptibility. The secondary
magnetic field due to the induced currents at any given nucleus is
-()‘iHo where -0‘i is the magnetic screening constant. The total field
experienced by a given nucleus, which determines the n.m. r.
frequency, is given by

H. = HON-041). (1)

1

The chemical shift may then be defined as (18)

5=_i____1: (2)

where Hi is the resonant field of the signal being measured at a fixed
frequency and H1. is the corresponding field for a given reference.

7 From equation (1) it is seen that the differences in the shieldings
of various nuclei are reflected by the differences in the screening
constants. There have been several attempts to calculate the screen-
ing constants of atoms and molecules. Using a classical approach,

Lamb was able to derive an expression for the screening constants in

atoms (19). The resultant expression is generally known as Lamb's

formula

07- - 313—.— 2 (-.~) (3)

3m-1c Ti

where (-e) is the charge on the electron, mi is the electronic mass,

c is the velocity of light, and ri is distance of the ith electron from
the nucleus, the sum is over all the electrons. Equation (3) gives the
diamagnetic contribution of the electrons to the magnetic field and is
applicable only to atoms since it depends on the spherical symmetry of

the electric field of the nuclear electric potential.

Ramsey's Theory of Shielding Constants

Ramsey has made a quantum mechanical calculation of molecular
screening constants (20). Using second-order perturbation theory

Ramsey derived an equation for the molecular screening constants as

 

0'= .)(0|2;3.r|o>-3AE(012—41——km) (4)

3mc‘z k

where m0

j is the total angular momentum and AB is the average value
of all the electronic exitation energies. The first term of equation «4)
corresponds to Lamb's expression for the diamagnetic shielding of
single atoms. The second term arises from the lack of spherical
symmetry of the electric potential and is called paramagnetic term.
The most serious limitation of Ramsey's equation is that for large
molecules the diamagnetic and paramagnetic terms turn out to be

comparable in magnitude and opposite in sign thereby tending to cancel

each other.

An Approximate Theory of Chemical Shifts

The theory of chemical shifts has also been developed in terms of
breakdown of diamagnetic currents into a sum of atomic terms as was
first proposed by Saika and Slichter (21). The total screening constant

for any atom A may be broken down into the following four terms.

AA AA AB A '
G’A:0’d +g’p +2: 0'“ +0” ’rmg (5)
B(#A)
, AA . . .
The first term, O'd , gives the contr1bution to the secondary

magnetic field at nucleus A due to the diamagnetic currents on atom
A itself. For an isolated spherical atom, it is the only contribution and
is given explicitly by the Lamb's formula of equation (4).
The second term, prA is the contribution due to the para-
magnetic currents on atom A which gives the susceptibility term.
This term was first calculated for fluorine atoms by Saika and Slichter
(21) who showed that it was much more sensitive to chemical structure
than the first term. The total variation of O-pAA in going from F2 to
a highly ionic system is of the order of 10-3. For most other atoms
variations in this term will give the main contributions to the chemical
shifts. The hydrogen atom is the principle exception since the absence
of low lying p—orbitals makes this paramagnetic term negligibly small.
The third term, TAB, accounts for the contribution to the
screening of atom A by the atomic circulations on atom B. When the
magnetic effects of these neighboring currents are treated in a dipole
approximation, this term will include only the local anisotropy of the
local susceptibility on atom B (22, 23). If atom A is on or near the
axis of high diamagnetism of B, there will be an increase in the average
screening. This effect falls off as the inverse cube of the A-B inter-

nuclear distance and is not likely to exceed 10"5 in magnitude (l7).

A .
The fourth and final term of equation (5), 0‘ ’ r1ng, takes into

consideration the contribution to the screening due to ring currents
which cannot be localized on any atoms. The magnitude of this term
is usually less than 2 x 10'6 but it plays an important role in determin-

ing the spectra of aromatic compounds (24, 25).

A General Molecular Orbital Theory of
Chemical Shifts

An approximate theory of molecular diamagnetic susceptibilities
based on the use of gauge-invariant atomic orbitals (GIAO) has been
recently proposed by Pople (28). The theory is developed in terms of
an independent electron model and is applicable to the theory of chemi-
cal shifts.

The motion of each electron will be given by a one-electron

Hamiltonian
A )2 + V (6)

where A is the magnetic vector potential, P is the momentum, and V
an appropriately averaged effective electrostatic potential. When a
molecule is placed in an external magnetic field _I-_1_ with a single magnetic

nucleus that has a magnetic moment at the origin, A will be given by

é=iEXE+exz/r3- (7)

With a magnetic field present, the molecular orbital may be written

as a linear combination of atomic orbitals
Vi = ‘3 C’iu X 11 (8)

where X11 is a gauge-dependent atomic orbital (17) that may be repre-

sented as

Xu=¢uexp(-%Efi '3) (9)

¢u is the normal atomic orbital and A is the value of the vector potential
at the nuclear position of ¢u .
The use of the variation principle (26) will yield linear'equations
of the form
. = . 1
EHquiv £1 E SUV Civ ( 0)
where 6i is the eigenvalue of the secular equation and represents the
energy possessed by an electron in the molecular orbital.

The matrix elements Huv and Suv are the coulombic, resonance,

and overlap integrals of molecular orbital theory (27) and are given by

H

11V

I 713mg?
I exp { ($114.. zévl’f- I ‘1’: 5%(3‘15‘5'5) }z+g¢v¢r

c
(11)

II

and

:9:
SUN: f x“ Xydxr
-.-. f exp { (TIE-”Au .. fiVI' I} ¢u*¢vd,r (12)

It may be seen from equations (11) and (12) that the Hamiltonian and
overlap integrals depend on the magnetic field because of the gauge

factors.

The following approximations are those suggested by Pople in the

development of the theory of molecular diamagnetic susceptibilities (28).

Approximation A

 

The exponential part of equations (12) and (13) .is taken outside

of the integral sign and I. is replaced by )- (Ru-I» Ev): where Eu and Ev

10

are the nuclear positions of the atomic orbitals ¢u and ev, respectively.
Replacing r by%- (Ru-t RN) is required only when the atomic orbitals

are not on the same atom and does appear to be reasonable since the
midpoint between nuclei u and v will be approximately the center of

the "overlap distribution. " This approximation does give rise to the
possibility. that the matrix Huv may be non-Hermitian. Should this
possibility occur it is to be resolved by splitting the resulting matrix into
Hermitian and non-Hermitian parts with retention of only the Hermitian

pa rt.

. e
Equat1ng Luv = (mHAu - év). (Ru + RV) (13)
the matrix elements Huv and Suv become
_ , * 1 e 2
2H...- exp “Luv’ 14>. [‘75) I '3 + 216-" (1.21 +311 I’vd'r
, * l e z
+ exp (IL...) I .v [(75) {1: + flé'éd) + Xl¢ud7

(14)

and

SW = exp (111W) f q): ¢v d7. (15)

Approximation B

 

The assumption is made that the magnetic field is weak enough for
all magnetic energies to be small compared with the energy differences
between molecular orbitals Ei - Ej' This approximation allows the
magnetic effects to be treated as perturbations and all matrix elements

can be expanded in orders of magnitude in A
_ (0) (1) (Z) , , ,
H - H... + H... + H... + “6'

s = 5(0) + 5(1) + s”)

uv uv uv 11V

IIV

+ o . o (17)

11

more specifically

I . 0
Hss= + -——t 1.1
Herm

(Z) (0) ' *
H... =% Li... H..* ($32) Luvtf (hue-Aw: ¢vdrl (20>

He rm
2

e * z
+2mc2{f¢u (xi-6'11) ¢VdT1Herm

where H (0) is the Hamiltonian in the absence of a magnetic field.

For the overlap matrix elements

(0) __ *

Suv ‘ I 4’u ¢v d7“ (21)
Sm = 1L 5(0) + 5“” (22)
uv 11V uv uv

5“” = - 5- 1.2 5(0) + 1L 5‘01 +, 5(0) (23)
11V 1.1V 11V uv 11V uv

Approximation C

 

The atomic orbitals on the same atoms are assumed to be mutually
orthogonal and the overlap integral between atomic orbitals on different
atoms is negligible. The result of this approximation is to make the

overlap matrix take the form

S = 6 (24)

and make it independent of the magnetic field. The latter part of this
approximation, while not quantitatively accurate, leads to a major
simplification that is usually made in molecular orbital theories of non-

magnetic phenomena .

12

The following two approximations are made to simplify the matrix

elements Hiilv) and HLZV) (2).

Approximation D

 

The integral f (I): (é-éu)’ I: ¢vdT is to be neglected if dpu and rbv
are not on the same atom. This approximation may be justified as
follows: In a uniform magnetic field, the operator (é-év)' P is pro-
portional to the orbital angular momentum operator at the center of the
nucleus v. When ¢v is an atomic s orbital the integral vanishes.

When ¢v is a p or d orbital the operator converts it into another p or
d orbital. This results in an overlap-type integral between different

atoms which, in accordance with approximation C, is negligible. The

(Z) to

result of this approximation is to cause the second term in Huv

vanish as Luv vanishes, if the integral does not.

Approximation E

 

>:<
The integral f ¢u (A-Av)z¢vdr is negligible when ¢u and ¢v are
not identical.
In addition to the five preceding approximations for the theory of
molecular diamagnetic susceptibility, it is convenient to make an

additional approximation for the theory of chemical shift.

Apm'oximation F

 

The local vector potential (é'év) is replaced by the local vector

potential of a magnetic field that is given by

-1
é'S—v_TI-_—va(£'B-v) (25)

(‘1

(‘1

(‘1.

13

In view of the foregoing approximations, it is seen that the break-
down of the molecular screening constants into intra-atomic terms is
accomplished by neglecting certain integrals involving atomic orbitals
on more than one atom. The diamagnetic contribution to the suscepti-
bility, X3, arises from the gauge factor modification of the atomic
orbital, equation (9). The paramagnetic contribution to the susceptibility
comes from changes in the LCAO coefficients which corresponds to the
mixing of the ground and certain excited states of the molecule by the
magnetic field. The explicit relations for 7C: and X: in the z

direction are as follows (17):

 

A e‘2 A
(7‘ d)zz = - W i puu( x2 + 3’2qu (26)
(A: 1 — 4:122:2TAE QA (27)
QA = Z QAB (28)
CAB: PYAPxBaaAB-PYAPYB) + PYAPYB
(zaAB- PXAPXB) + ZPXAXBPYAYB (29)

where the various Puv elements are components of the charge-density

and bond-order matrix for the unperturbed molecule,

OCC
P = 2 z c. c (30)

uv Iu iv
and the sum is taken over occupied molecular orbitals. The para-
magnetic part, equation (27) has been. simplified by assuming that the
excitation energies required in the perturbation theory can be replaced

by an average value AE. The result can then be expressed in terms of

 

14

the Puv matrix for the molecular ground state; PxA PXB is the element

of the matrix for the 2px atomic orbitals on atoms A and B. When

A = B,“ PXA PXB

the bond order between the two atomic orbitals.

is the charge density in ZPxA orbital and when A 71 B,

The explicit relations for the components of the magnetic screen-

ing constant are:

(1) Local Diamagnetic Currents
-1
o-rd = ‘3——2‘ E Puu (r )uu (31)

This is similar to the Lamb formula derived from classical con-
sideration and is easily calculated if the LCAO molecular orbitals are

known.

(2) Local Paramagnetic Currents

AA
0.) : — ZN"l < r'3 > (32)
p 2p 7‘19

Here < r‘3 >2p is the mean value of r'3 for the 2p atomic orbital.
The direct proportionality to the local paramagnetic contribution 7(
should lead to a correlation of the chemical shift with diamagnetic
susceptibility data for atoms with 2p electrons. The < r'3 > proportion-
ality factor makes the local paramagnetic contribution large and is

likely to be dominant in the application to carbon, nitrogen, oxygen,

and fluorine.

(3) Currents on Neighboring Atoms

AB_1 _ B 5
0" ~19: N 1:126 1a,, (3RBQRBB - R2B6I1IB)/ RB (33)

15

Here a and 13 are the tensor suffixes and R o‘(a =x, y, z) are the

B
components of the vector from nucleus A to nucleus B. If the
B B
susceptibility on atom A is isotropic (X 05 = x 6 QB) this term is

zero so the contribution is frequently called the neighbor anisotropy
effect. When the local susceptibility tensor on atom B is axially

symmetric O’AB takes the form

J’AB=%-(A7LB)(1-3cosz{B)/R3B (34)
where A XB(= XE - XiB ) is the anisotropy of the susceptibility and
YB is the angle between the axis of anistropy and the AB internuclear line.

A theory of carbon—13 n.m. r. chemical shifts in conjugated
molecules has been developed recently by Karplus and Pople (16). The
theory shows that the paramagnetic contribution to the carbon-13 chemi-
cal shifts is the dominant one. The derived expressions for the para-

magnetic contributions to the shielding in conjugated molecules are

AA 1132 4

p = — [fizz-rIAEII <r'3>2p [2 + I; (~1—- FA” (35)

AA__[eZBZ (AE3)] (1:3) [z+—4i)t(1-PAA)+-é(~f—-F)
{p _ ch 2P 9 H z z 9 A

(36)

where FA if the free valence of the atom and AH is the polarity para-
meter. Equation (35) defines the paramagnetic contribution for the
carbon atom which is in a planar conjugated system with sp‘2 hybridization
and bonded to three other carbon atoms. Equation (36) gives the para-
magnetic contribution for the carbon atom which is in a planar conjugated

system with sp2 hybridization but bonded to two carbon atoms and one

hydrogen atom.

16

Atomic Contributions to Chemical Shifts
in Organic Molecules

Pople's theory can be used to make approximate estimates of
atomic contributions to the diamagnetic susceptibility for some of the
more important organic groups. For atoms other than hydrogen, the
paramagnetic contribution X: is the most sensitive to electronic
structure and estimates of the relative shifts of carbon and the neighbor
anisotropy effect for hydrogen can be made by considering this term

only.

Carbon-Carbon Single Bonds

 

The theory predicts that all carbon atoms in saturated compounds
have the same X: (6.46 x 10'6) and are isotropic. Therefore all
carbons in unstrained paraffins should have the same chemical shift.
The experimental chemical shifts for paraffinic carbons are spread
over a range of approximately 30 p.p. m. to low field from methane.
Although these shifts have not been satisfactorily explained they may be
due to some delocalization of the bonding electrons.

Since the theory predicts that the carbon atoms are isotropic,
calculated chemical shifts for all paraffinic hydrogens are the same.
There is a range of approximately 1. 5 p.p.m. to low field from

methane in the experimental data,

o’(-CH) < M-CH.) <0’ (-CH.) < men.)

but this may be due to changing electron density.

Ca rbon- Ca rbon Double Bond 8

 

When the magnetic field is perpendicular to the carbon-carbon

double bond, the theory predicts larger 7C: terms on the doubly-bonded

(‘3

7"
H4

(7

In

17

carbon atoms for ethylene and other olefins. This contribution arises
from the mixing of the 04—? 1t and 17 +fexcited states with the
ground state by the magnetic field. The calculated and observed

chemical shifts for ethylene relative to ethane are given in Table I (17).

Table 1. Calculated and Observed Chemical Shifts For Ethylene

 

 

6 6
Atom (Calculated) (Observed)
(p-p-m-I (p-p.m.)
Carbon -46 - 120
Hydrogen -2.1 -4. 2

 

The calculated chemical shifts are in qualitative agreement with
the observed low diamagnetism of olefins and the low field shift of
carbon resonance from ethane. The calculated shifts are in the right

direction but are somewhat smaller than the observed values.

Carbon- Ca rbon T riple Bonds

 

The theory predicts that the acetylene molecule will, have a high
diamagnetic anisotropy along the molecular axis. This should cause a
high-field shift for the proton due to the neighbor anisotropy effect.

The fact that the acetylene proton resonance occurs to the low field of
ethane (-0. 6 p. p. m.) can be interpreted as a cancellation of neighbor
anisotropy effect by the low-field shift due to reduced charge density on
the hydrogen atom. The carbon chemical shift in acetylene should be
close to that of the paraffins. The observed carbon shifts lie at an

intermediate position between singly and doubly-bonded carbons.

18

Allenes

Calculations on the allene molecule are of particular interest
since they give an additional paramagnetic term A X: which is
twice as large on the central atom as on the two end atoms. This is
due to the fact the additional term in the plane of the double bond occurs
for both directions perpendicular to the C=C=C axis for the central atom,
but only for one direction for the end atoms. Hence it is to be expected
that the carbon resonance of the central atom should be shifted to low
field relative to the end atoms. The experimentally observed carbon
shifts for allene do exhibit the separations of the kind predicted.
However the observed difference in chemical shift between middle and
end carbon atoms in allenes is 140 ppm. whereas the calculated

separation is 40 ppm. The calculated shifts again are too small.

Carbonyl Groups

 

The theory provides an interpretation of the low diamagnetism of
aldehydes and ketones and also predicts anisotropies for the carbon

A
and oxygen atoms. Table II summarizes the paramagnetic terms X p (17).

Table II. Calculated Paramagnetic Contributions to the Atomic Molar
Susceptibilities for the Carbonyl Group (Units of 10)

 

 

(X) (X) (X) Y’l: ,
pXX p yy p ZZ X
c 6.46 10.76 6.46
0 10.83 10.00 4.31 /C=O

 

On the carbon atom the largest paramagnetic contribution is in the y-

direction and the anisotropy is comparable to that for carbon atoms in

19

X
P)zz for

oxygen can be interpreted similarly. The high paramagnetic term

the carbon-carbon double bond. The difference (Xp)yy - (

for oxygen in the x- direction arises from the low lying n -—> 11* excited
state, which is mixed with the ground state by a magnetic field in this
direction.

From Table II it can be seen that the anisotropies for both
atoms lead to negative (low field) contributions to the screening con-
stants of aldehydic protons. The total calculated neighbor anisotropy
effect is -2.6 p. p.m. of which -2. 2 p.p.m. comes from carbon.

The observed chemical shift of aldehydic protons relative to paraffinic
hydrogen is -8 p. p.m. It may be that the theory underestimates the
anisotropies in the carbonyl group.

The main advantages of the foregoing theoretical developments
lie in the fact that all the contributions to the screening are deduced
within a single theoretical framework and that the theory is particularly
suitable for the chemical shifts involving atoms of hydrogen, carbon,
nitrogen, oxygen, and fluorine. Equation (32) should be a suitable
starting point for discussing correlations between chemicals shifts
and atomic contributions to diamagnetic susceptibility.

Hameka, who has developed an exact theory of the diamagnetic
susceptibilities for diatomic molecules from SCF-LCAO-MO functions
by introducing GIAO (29, 30), has raised some serious objections to
Pople's approximate theory (31). Hameka contends that the deviations
caused by approximations A-D are not negligible but that they are small
when compared with the errors introduced by approximation E. In fact,
Hameka claims that the approximation E is incorrect and concludes
that it is generally not possible to express a molecular diamagnetic
susceptibility as a sum of atomic contributions since interatomic terms

are of major importance (31).

20

Pople has answered Hameka's objections by pointing out that
part of the error caused by neglect of the two-center integral has
been compensated for by associating larger coefficients with the one-
center terms (32). In addition, Pople has emphasized that the theory
based on atomic orbitals has a decided advantage in the application of
n. m. r. chemical shifts, since the magnetic field is measured at the
center of a particular atom.

This approximate theory has been successfully applied by Pople
to the hydrogen and carbon chemical shifts in carbon-carbon single,
double, and triple bonds, allenes, and carbonyl groups (17). In addition,
Karplus and Pople (33) have used a similar approach to develop a theory
of carbon-13 n.m. r. chemical shifts in conjugated molecules.

The theory does provide an interpretation of some of the empirical
features of carbon-13 chemical shifts (33) in spite of the severity of the
approximation. It should be emphasized that Pople's theory is to be
taken as a step to simplify another complex problem and is of consider-
able value in this respect.

The theory of chemical shifts has also been developed from the
valence bond approach. Some recent contributions to the theory of
shielding constants include those of O'Reilly (34) who used a group
theoretical method to calculate the chemical shifts of nuclei of ions in
crystalline fields. Kern and Lipscomb (35, 36) have calculated the
shielding constants in diatomic molecules by simplifying the paramagnetic
term of Ramsey's equation with a judicious choice of the gauge of the
vector potential and an approximation involving cancellation of average
excitation energies of the matrix elements of the linear and angular
momentum operators. In addition Hameka (37), and Ghosh and Sinka
(38), have surveyed the various valence bond approaches used in calcu-

lating shielding constants. Shielding constants in aromatic systems

21

have been investigated by Maddox and McWeeny (39), Hall and
Hardesson (40), and by Jonathan, Gordon and Dailey (41).

Nuclear Spin—Spin Interaction

In 1951 Gutowsky, McCall and Slichter observed that high resolu-
tion spectra frequently exhibited hyperfine structure which, in contrast
to the linear dependence of chemical shifts, was independent of the
applied magnetic field (42, 43, 44). Similar effects had been observed
by Hahn and Maxwell in the modulation of the spin-echo envelope in pulse
experiments (45, 46). It was concluded that the observed effects were
caused by an indirect coupling of the nuclear momentsii, which is
transmitted from nucleus to nucleus by the paired electrons comprising
the valence bonds. The spin- spin coupling constant between two nuclei

of spins i and j has the form (47):

Jij = X, If Ki, (37)

Where Ji’ is the spin-spin coupling constant, T is the magnetogyric
ratio,‘his Plancks constant divided by 2 17 , and K is a constant depend-
ing upon the molecular electronic structure. The magnitude of the
coupling constants for hydrogen are small (0-20 cps.) but may persist
over several chemical bonds (48, 49). The coupling constants between
other nuclei increase with atomic number and may be as large as several
kilocycles (50).

Recently the problem of the sign of the spin-spin coupling constants
has received considerable attention. Using double irradiation or spectrum
analysis, the relative but not the absolute, signs of spin-spin coupling
constants may be obtained. A knowledge of the absolute sign of spin- spin
coupling constants is of importance for comparison of theory with

experiment. Karplus (51) has suggested one method for the obtainment of

22

absolute signs which is based on the fact that the theory of spin- spin
coupling for directly bonded atoms is on much more solid ground than
that for more widely separated atoms and predicts a positive sign for
directly-bonded atoms. Karplus has further suggested that in organic
compounds the carbon-13-proton coupling provides a suitable reference
for determining absolute signs of proton-proton coupling constants.

Such absolute signs determinations have been carried out by Lauterbur
and Kurland (52) and Anet (53) who found only partial agreement with the
expectations from valence bond theory.

The current theoretical treatment of spin- spin coupling in organic
molecules suggests that there exists a correlation between the gem
proton coupling constants and the H-C-H angle and between vicinal
coupling constants and the H-C-C-H dihedral angle. A number of authors
have measured such spin- spin coupling constants in a number of relatively
simple molecules of known conformation, and compared the results with
theoretical predictions. These include Graham and Rogers (54),
Gutowsky and Juan (55), Williamson and Johnson (56), Anet (57), Barfield
and Grant (58), and Bernstein and Sheppard (59). Only partial agree-
ment between theory and experiment is generally found, and the actual
spin- spin coupling constants are no doubt modified by substituent effects

and other factors (60).

Nuclear Relaxation

When a collection of nuclei is placed in a magnetic field there will
be an equilibrium distribution of these nuclei into the various spin
states. For simplicity, nuclei with only two spin states will be con-
sidered. The populations of the two spin states will not be equal, but
will follow the Boltzmann distribution law with the state of lower energy

being more populated (61). With the absorption of energy from the

1_,

0*

.h-
.c

7'" '

A“!

Ti;

23

magnetic field, nuclei in the lower states will go into the upper state.
There will also be a tendency for nuclei in the upper state to return to
the lower state. The latter process is of extreme importance in the
theory of nuclear magnetic resonance and is usually referred to as the
relaxation process. The two general types of relaxation are the spin-
lattice and the spin— spin relaxation processes.

The spin~lattice or longitudinal relaxation, (T1), involves the
establishment of thermal equilibrium between a collection of nuclear
magnets with different quantum numbers and results in the establish-
ment of an equilibrium value of the nuclear magnetization along the
magnetic field axis (62). Among the interactions which contribute to
T1 are (a) dipole-dipole interactions, (b) nuclear dipole-other magnetic
moments interations such as electrons and molecular magnetic moments,
and (c) nuclear quadrupole-electric field gradients.

The spin-spin or transverse relaxation (T2) represents the loss
of phase coherence of the precessing collection of magnetic nuclei.

The interactions contributing to T2 are (a) magnetic field inhomogeneity,

(b) direct dipole-dipole coupling in solids, and (c) spin- spin interactions.

Line Shapes

The quantitative relationships between the shape of an n. m. r.
signal and the relaxation times T1 and T; can be obtained by solving the

Bloch equations (63)

' _ . MX

Mx - Y (MYHO + MLH1 Slnwt) — 7f;— (38)

M=f(MHcoswt-MH)-Mz- (39)
y Z I. X 0 T2

. _ M -

M2 = X ("MxHi sin wt - MyI-Ii cos wt) ___er_‘;1\_4_0__ (40)

24

where Ma( 0. = x, y, z) is the component of the magnetic moment per unit
volume, K is the magnetogyric ratio, 1:10 is the applied magnetic field,
Ii; is a field perpendicular to I_-I_o rotating with angular frequency 00, and
M0 is the equilibrium value of Mz.

The Bloch equations can be simplified by referring them to a set
of axes rotating with the applied field I_-1_1 rather than to the fixed axes
x, y, and z. This transformation will give a new set of axes rotating
with angular velocity 6) about the axes. The new components of M
which are parallel (in phase) and perpendicular (out of phase) to the

direction of H1, are called (.1 and v components. These components have

the following relations.

Mx =11 COS wt - V Sinwt (41)

My: -usin wt-vcoswt (42)

or alternately,

C.“
II

Mx cos cot - My sin wt (43)

v=-stinwt-Mycos wt (44)

Substitution of equations (41) and (42) into the equations (38), (39) and

(40) gives the following form for the Bloch equations:

11 +-*-+va=0 (45)
T2
. v
v+—-Awu+wl=0 (46)
T2
Mz+h—4-zf'fl9— -o.)v=0 (47)
1

where

25

Au) : COO-(.01, XHO '3 (.00, and r141: (.01.

The steady state solutions of equations (45), (46), and (47) can be
obtained by setting all time derivatives equal to zero. The steady state

solutions are

 

 

2
L01 T2 A0)
= M
u 0 1+T§ A00 + wlleTz (48)
V - MO (.01 T2 (49)

1+T§Aw + wfirlrz

Under experimental slow passage conditions, {1 , v, M; fr; 0, the rate of
absorption of energy is proportion to v and is given by equation (49).

If the magnitude of the oscillating field is small such that

a? T11“, << 1 (50)

then the rate of absorption of energy becomes proportional to

 

 

2T2
1+ TE Am: (51)
which gives the line-shape function
2T
2 (52)

3M = 1+417sz(}’0- )f)r

Equation (52) describes the shape of the signal for the absorption or

v mode, Figure 1a.

Under appropriate experimental conditions it is possible to observe
u, the in phase component of the magnetization. Under the conditions of

slow passage the line shape function is proportional to equation (49).

26

If saturation is negligible, that is, when equation (50) holds, the dis-

persion mode line-shape function is

If...)r

g(v) z ”4.213 (r.- r)?

 

(53)

which is shown in Figure lb.

Figure 1c illustrates the line shape for the rapid passage dis-
persion mode. Pure rapid passage conditions are not always achieved
experimentally and the dotted line in Figure 1c illustrates the line shape
under conditions intermediate between rapid passage and slow passage in

the dispersion mode .

Line Broadening

The absorption of a n.m. r. signal occurs over a range of fre-
quencies rather than a single frequency thus causing the signal to be
broadened. There are a variety of factors contributing to the broaden-
ing of a signal. Among these is the natural line width due to spontaneous
emission which is determined by the finite lifetime in the upper state.

The line widths are affected by the spin—lattice relaxation process.
The order of magnitude of the broadening can be approximated by the

uncertainty principle.
AEAt .1: H (54)

where AE and At is the uncertainty in energy and time, respectively.
From equation (52) it can be seen that the uncertainty in frequency of
absorption is (217At)"l and the line width measured in terms of frequency
is proportional to the reciprocal of T1.

In solids and highly viscous liquids, where the nuclei remain in

the same relative positions for a long period of time, the interactions

27

 

 

 

 

 

 

 

 

Figure 1. The shapes of n.m.r. signals: (a) slow passage absorption,
(b) slow passage dispersion, (c) rapid passage dispersion.

28

between magnetic dipoles can lead to a greater broadening than that
caused by T1. The broadening is caused by the spin— spin relaxation
time, T2, and is of the order of 105 cycles/sec. which is substantially
larger than normal values of l/Tl. In liquids and gases the molecules
are rotating rapidly and the dipole-dipole interaction is effectively
averaged out. Under these circumstances T1 and T3 are approximately
equal in magnitude.

When nuclei have spin greater than one-half their electric
quadropole moments will interact with the electric field gradients.

This will give rise to additional mechanisms by which relaxation will
occur. The net effect will be to decrease the value of T1 and T2 causing
additional broadening.

The essential instrumental limitation which contributes to the
broadening of n.m. r. signals is the inhomogeneity of the static magnetic
field. Ho, over the dimensions of the sample. This effect is of major
importance in carbon- 13 natural abundance spectra where it is customary
to use non- spinning sample tubes of large external diameter.

Of particular importance in carbon-l3 spectra studies is the
broadening caused by unresolved spin- spin couplings with distant hydro-
gen atoms. These can be resolved in a few carbon- 13 spectra and have
been measured in proton (65, 66) and fluorine spectra (67, 68); variations
of these distant couplings will result in changes in the widths and heights
of peaks and will affect intensity measurements which are usually based

solely on peak heights.
Line Asymmetry

In addition to those factors contributing to the shapes and broaden-
ing of n.m. r. signals discussed in the previous sections, there are
factors which are of particular significance to natural abundance
carbon-13 studies. These factors affect the symmetry of the carbon-L13

resonance signal and are discussed below.

29

There is an inherent asymmetry in the rapid passage dispersion
mode carbon-13 spectra. In Figure 2 are shown the 15.1 Mc./sec.
spectra of benzene under such conditions. The inversion of the spectrum
with reversal of sweep directions is to be noted. It has often been
0 bserved that the appearance of the carbon-l3 spectra exhibits a strong
dependence on the sweep direction. This makes it desirable that all
spectra be presented for both sweep directions.

Another source of asymmetry of carbon-13 spectra is the magneti-
zation-transfer effect. This causes the second peak in the respective
sweep directions to be decreased in intensity by an amount which depends
upon the sweep rate and the relaxation times of both carbon-l3 and
hydrogen. Similar effects are observed for higher multiplets. The
magnetization-transfer effect is caused by a rapid transfer of non-
equilibrium nuclear spin magnetization from one magnetic environment
to another (69). Magnetization transfer in chemical reactions has been
studied in detail by McConnell and Thompson (70). However, transfer
by relaxation of spin-coupled nuclei although quite similar has not been

treated quantiatively to date (71).

Saturation

The absorption of energy from the radio frequency field will not
only cause a reduction in the population of nuclei in the lower state but
also a reduction in the probability for further absorption. The magnitude
of the latter will increase as the amplitude of the oscillating field in-
creases. This effect which is known as saturation can be defined in terms

of a saturations factor, Z, given by

1
z = (1 + VHETJZ)’ (55)

3O

 

 

 

 

 

 

 

 

Ho Increase \ HO decrease

T 7

 

 

 

 

 

Figure 2. 15.1 Mc./sec. carbon-13 spectra of benzene.

 

31

It is to be emphasized that in equation (55) the greater the radio frequency
power applied to the sample, the greater the degree of saturation. In
addition, the combination of a large radio frequency field, long spin-
lattice relaxation times, and narrow absorption lines may prevent the
observation of a given line or spectrum. This is of particular significance
in natural abundance carbon-13 studies where the long relaxation times

of the carbon-13 nuclei and the large radio frequency power customarily

used necessitate special precautions.

Double Ir radiation

The hyperfine splitting due to nuclear spin-spin interactions between
nuclei possessing magnetic moments can be removed by double resonance.
This can be accomplished by simultaneously irradiating one set of coupled
nuclei with a strong radio frequency field close to its resonance frequency
while observing the second set of nuclei with a weak radio frequency field.
The essential objective is to cause the nuclei irradiated by the strong radio
frequency field to undergo rapid transitions which reduces the lifetimes
in states with a given m value. When the lifetime becomes much less
than J‘l, the group observed with the weak field is effectively decoupled.
The double resonance technique is more easily applied to nuclei with
different magnetogyric ratios than when nuclei have the same magneto-
gyric ratio. Hence, carbon-13-proton decoupling is more readily obtained
than proton-proton decoupling.

The theory of double irradiation has been developed on a rigorous
statistical quantum mechanical basis by Bloch (72). An approximate
theoretical treatment, limited to the case where J << 6 has been given
by Arnold (73). More recently the theory of double irradiation has been
treated by Freeman and Whiffen (74), Anderson and Baldeschwieler (75),

and Anderson and Freeman (76, 77). The latter papers are far more

32

comprehensive and contain graphs and tables from which frequencies
and intensities in double irradiations spectra for Aan(m, n f 3) can

be obtained.

EXPERIMENTAL

Spectrometer s

The Varian A-60 Spectrometer was used to obtain the proton
magnetic spectra. The internal chemical shift measurements for
the ethyl compounds were obtained using 250 or 100 cps. sweep widths.
The probe temperature was approximately 350C.

The carbon-13 magnetic resonance spectra were obtained using
a Varian Associates high-resolution nuclear magnetic resonance
spectrometer, model V-4300-2. A Varian Associates model V-4331A
RF probe was used to obtain spectra at a fixed frequency of 15. 085
Mc. /sec.

The constant magnetic field was obtained with a Varian Associates
model V-4012A twelve-inch high-resolution electro-magnet equipped
with a model VK-3513 field trimmer and a model V-4365 field homo-
geneity control unit; the power supply for the electromagnets was a model
V—21OOA power supply equipped with a field reversing mechanism.

The stability of the spectrometer system was enhanced by use of
a Varian Associates model VK-3506 magnetic flux stabilizer. Further
stability was achieved by using a Sorenson 10005 a. c. voltage regulator
to regulate the line voltage to the spectrometer console. The magnetic
cooling system, consisting of distilled water circulated through the
cooling coils and recycled through a refrigerated copper tank, was
maintained at a temperature of 19. 7 i 0. 30C. The room temperature
was regulated by a commercial air—conditioning unit.

Calibration of the spectra was obtained by using audio frequencies
from a Hewlett-Packard Model 200 CD wide range oscillator, to modulate

the field. Modulation of the field by audio frequencies produces side-bands

33

34

on either side of the main signal. The separation of the side—bands
from the main peak is equal to the modulation frequency. A Hewlett-
Packard Model 521~A electronic counter was used to measure the
audio modulation frequencies. The spectra were recorded on a Varian
Associates Model G-lO Graphic recorder.

The probe insert had an internal diameter of 16 mm. which
permitted the use of 15mm. external diameter sample tubes. All spectra
were obtained under rapid passage dispersion mode conditions. The
sweep rate employed was approximately 300 cps. per minute. Since
the line widths increase with increasing radio frequency power H1,

a compromise had to be reached between the signal strength and resolution
in a given spectrum. It has been empirically established that better
spectra are usually obtained when the radio frequency field is at a setting
of coarse-2 and fine-5. To obtain the dispersion mode spectra, the

probe was tuned at maximum radio frequency power. The R—F field

was then set at coarse-2 and fine-5. After minimizing the direct current
signal level meter with the fine red paddle (u-mode) the fine blue paddle
(v-mode) was turned counter clockwise to introduce a leakage of 5 to 10
microamperes. The resultant direct current signal level meter reading

was usually 50 to 60 microamperes.

Compound 5 Studied

All compounds studied in this research are listed in Tables I, II,
III, and IV. The compounds were purchased from Eastman Organic
Chemicals, Peninsular ChemResearch, Inc. , Columbia Organic Chemi-
cals Co. , Inc. , and The Matheson Company. The solvents carbon
disulfide and chloroform were J. T. Baker AR grade and were used with-
out further purification. The carbon- 13 spectra of both solvents

exhibited no impurity peaks. Table I lists the aliphatic compounds studied.

35

The compounds were dried for at least twenty-four hours over Drierite
and distilled at atmosphere pressure. Chloroacetic acid and tri-
chloroacetic acid which are solid at room temperature were distilled
without drying. Table 11 lists some physical constants of toluene and
its derivatives. Tables III and IV list some physical constants of the

diphenyl and triphenyl compounds, respectively.

Table I. Physical Constants of Aliphatic Compounds Studied

 

 

Compound Boiling Point (OC.) Molecular Weight
oncogn 71. 2 114
CFZC1COZH 119.7 130
CFClzCOzH 160. 3 147
CC1,cozH 195.4 163
CHClzCOZH 193. 8 129
CHZC1COZH 187. 9 95 '
(CF3CO)ZO ‘ 38. 2 210
(CF3CO)ZCHZ 63. 7 208
(CH3CO)2CHZ 174. 6 100
.CFzBrCFzBr 45. 3 260
CF3CHZOH 72. 5 100
CF3CH(OH)CH3 76. 7 114

CF3C(CH3)ZOH 80.

p—I

128

 

(I-

If}.

36

Table II. Physical Constants of a—Substituted Toluenes Studied

 

 

Compound Boiling Point (0C) Molecular Weight
¢CH3 110. 6 92
¢CH2C1 176. 3 127
¢CHZBr 196. 8 171
¢CH2NH2 182. 4 107
¢CHZCN 230.6 117
¢CHZSH 193.5 124
¢CHZOH 203. 2 108
¢CHZ¢ 263. 1 168

 

Table III. Physical Constants of Diphenyl Compounds Studied

 

 

Compound Boiling Point (OC.) Molecular Weight
(1)20 255. 7 170
(1’25 293. 9 186
¢zSe 297. 4 233
¢zHg 12381 355

 

aMelting point.

37

Table IV. Physical Constants of Triphenyl Compounds Studied

 

 

 

Compound Melting Point (°c.) Molecular Weight
¢3N 125. 5 245
¢3P 79. 2 262
63As 60. 7 306
4135b 51.9 353
e381 76.5 440
¢,CH 92. 8 244

 

Sample Preparation

The samples for which proton spectra were to be obtained were
degassed and sealed in 5mm. external diameter A- 60 tubes. For the
oxygen sensitive ethyl compounds the solvent was added to the sample
tube and the solute was then vacuum distilled into the sample tube.

For the carbon- 13 studies the liquid samples were mixed with
CS; to give thirty to fifty volume per cent solutions which were placed
in a 15mm. external diameter sample tube filled to a height of 20 to
30 mm. , and the tubes sealed under vacuum. Alternately, the neat
liquids were placed in a 15mm. external diameter sample tube attached
to a 14/35 standard-taper female joint. A 14/35 standard-taper male
joint sealed at one end was used as a stopper. The internal diameter
of the male joint was reduced to 5 mm. This permitted a 5mm. sample
tube containing 57. 3% carbon— 13 enriched methanol to be placed within
the sample tube to be used as a secondary standard. Where solubility
was sufficient to give a good carbon-13 signal, solids were dissolved

in carbon disulfide to yield a saturated solution. The saturated solution

38

was then sealed under vacuum in the 15 mm. sample tube. Compounds
having low solubility in carbon disulfide were dissolved in chloroform

and the spectra obtained using the above standard-taper tubes.

Determination of Spectral Parameters

Proton Magnetic Re sonanc e

 

To obtain the internal chemical shifts and spin- spin coupling con-
stants for A383 and A3B2X systems the experimental spectra must be
compared with the theoretical spectra. This is necessary because the
ethyl compounds considered in the solvent studies do not have first order
spectra at 60. 000 mcs. The theoretical spectra were obtained from the
table of line frequencies and relative intensities for the A3Bz system
listed by Corio (48), with one modification. Corio gives the data for
the cases where the internal chemical shifts are positive--that is,
when the resonance of the methyl protons, A group, occur at a higher
field than the resonance of the methylene protons, B group. For the
systems considered the resonance peak for the methylene protons occurs
at a higher frequency than that of the methyl protons, resulting in a
negative internal chemical shift. This requires a change of sign for
the frequency of each line position. The internal chemical shifts were
obtained by measuring the frequency separation between the A6 line
and the mean of the B4 and B5 lines. The values for the spin- spin
coupling constants were obtained by simultaneous solution of two corres-

ponding equations for the line frequencies.

Carbon-l3 Magnetic Resonance

 

The inversion of rapid passage dispersion mode spectra with re-
versal of sweep direction and the strong dependence of the appearance

of a spectrum upon the sweep direction have made it necessary to record

39

the spectra in both sweep directions. The direction of increasing field
will have the peaks pointing up and the higher applied fields to the right.
The direction of decreasmg field has the peaks pointing down and the
higher applied fields to the left. For both spectra, the direction of
sweep is from left to right, increasing in the first case and decreasing
in the second. The measurements of peak positions were accomplished
by measuring equivalent points in spectra taken with increasing and
decreasing sweep field and averaging the results. The spectra were
calibrated using a 1000 cps. modulation frequency. A secondary cali-
bration used was the spacing between the two outer members of the
quartet from the enriched sample of methanol.

All chemical shift measurements are given using carbon disulfide
as the internal and external standard. This follows the convention
established by Lauterbur who has shown that solvent effects on the
chemical shifts of carbon disulfide are negligible. Table V gives the
solvent effects on the chemical shifts of carbon disulfide in various
solvents (78). Due to the negligible solvent effects on the chemical shift
of carbon disulfide no corrections for the diamagnetic susceptibility
have been made.

The assignments of the peaks in the aliphatic compounds can be
made on the basis of the multiplicity caused by the spin- spin interactions.
This was possible because the overlapping of peaks occurred infrequently.
One exception was (CF3CO)ZCHZ where the spin-spin coupling caused over-
lapping peaks. The assignment of peaks was greatly aided by the
unambiguous assignments that had been made for (CH3CO)ZCHZ.

For the a- substituted toluenes the aromatic portion of the spectrum
consists of two or three peaks. The five ring carbon atoms can be
assigned by consideration of the magnetization transfer effects and

relative peak heights.

Table V. Solvent Effects on the Carbon-13 Chemical Shifts of Carbon

 

 

 

Disulfide'"
6
Solvent (ppm.) Corrected (ppm.)
Nitrobenzene O. 0. 5
Acetone 0. 0-0
Aniline O. 0. 2
Cyclohexane 0. 0.1
Carbon Disulfide (0 (0)
Bromoform 0. O. 4

 

3):
Reference (78).

Note: All solutions contained one-third carbon disulfide by volume.
The maximum deviation of an individual measurement from the

average was 0. 2.

The assignments of peaks occurring in the spectra of the diphenyl

and triphenyl derivatives are somewhat more difficult to make. The

height of each peak was measured in both sweep directions and averaged.

The assignments for diphenyl ether were based on the similarity of the

carbon- 13 spectra of phenol which was analyzed by Lauterbur with the

aid of methyl substitution (8).

For the remaining phenyl compounds the

assignments were made on the basis of the factors discussed previously.

In addition, quadrupole broadening and larger spin values for the

arsenic, antimony, and bismuth (I: g- ; %,%~; 521-) compounds will cause

the substituted carbon peaks not to be observed. The para carbon atoms

can usually be distinguished from the ortho and meta carbon atoms by

relative peaks heights .

could not be made with complete certainty.

The choice between ortho and meta carbons

However, Spiesecke and

Schneider have shown on the basis of deuterium substitution that the range

of chemical shifts for meta carbons is usually very small, 2.4 ppm.

RESULTS

Effects of Solvent on the N.M. R. Spectra
of Some Organometallic Compounds

The results from the solvent studies of the n.m. r. spectra of
a series of ethyl compounds are listed in Tables I and II, and
Figure 1. Table VI gives the internal chemical shifts of the pure
compounds and also their internal chemical shifts at infinite dilution.
Figure 1 is a graph of chemical shift against mole fraction of solute
from which the internal chemical shifts at infinite dilution were ob-
tained. The variation of spin-spin coupling constants with concen-
tration for all of these compounds is essentially negligible. Table VII
lists the spin- spin coupling constants as a function of mole fraction in
CC14 for Pb(CI-IZCH3)4 which has an internal chemical shift of approxi-

mately zero.

Table VI. Proton Chemical Shifts in Ethyl Compounds

 

 

 

Internal Chemical Internal Chemical Shifts

Shifts (cps.) (cps.) At Infinite Dilution
Compound Pure Compound CHC13, CCl4 C6le
Zn(CHzCH3)2 -51.5 :1: 0.4 - -48.8 -
Hg(CHzCH3)Z -16.2 i0.1 -18.3 '17.0 '1595
Ge(CH2CH3)4 -18.1:t 0. 2 - -18.8 -
Sn(CHZCH3)4 -22. 5 :1: 0. 2 -24.0 -23.0 -22.4
Pb(CHzCH3)4 ’\‘ O - 40 -

 

41

42

Table VII. Spin-Spin Coupling Constants of Pb(CHZCH3)4 in CC14

 

—'

 

Mole Fract1on JAB(cps.) JAX(cps.) JBX(cps.)
0.099 7.83*0.15 125.7i0.2 41.6i0.2
0.332 7.82i0.16 125.3i0.l 41.6:h0.3
1.000 7.9Zi0.22 125.3i0.2 41.0:h0.2

 

The value of JBX for Sn(CHzCH3)4 reported by Narasimhan and
Rogers (79) is in disagreement with the value 51.4 cps. reported by
Klose (90). In order to determine this coupling precisely the n.m. r.
spectrum of Sn(CHzCH3)4 has been obtained at a spectrometer frequency
of 100 Mc/sec. . Using an internal chemical shift of 37. 6 cps. at 100
Mc./sec. values for JAX and JBX can be obtained. These are listed in

Table VIII along with the values previously reported (79).

Table VIII. Spin-Spin Coupling Constants of Sn(CHZCH3)4

 

 

 

Compound JAB(cps.) JAX(cps.) JBX(cps.)
sn“7(CHZCH3). 8. 1 73. 2 55. 7
Sn1'9(CHZCH,). 8. 1 76. 7 58. 4
Sn“7(CHzCI-l3)4a 8. 2 68. 1 30. 8
sn'"9(c112c113).f‘1 8. 2 71. 2 32. 2

 

3”Reference 79.

 

 

 

 

 

 

 

 

 

 

 

 

19. O .
e V
18. 0 g B
17.0 e a A.
.6 .
hp") 16. 0- . G 1
15. 0 380493211
24- 0W
23. 0.. O V” )4
m C if ‘r
"6/ V
22. O»
21. summon).
-5 20.0
(0130)
19.
18.
17.
52.
51.
50.
-8 .
(0930) 49.
48. 0-
Zamora).
47. A i I I l ' 1 1

 

 

 

0.0 0.10.2 0.30.4 0.5 0.6 0.7 0.8 0.91.0
Mole Fraction of Solute

Figure 3. Internal Chemical Shifts Versus Mole
Fraction of Solute for Ethyl Compounds.

A-CC
B-CHC
C'GBH”

44

The n.m. r. spectra which are reproduced on the following pages
were obtained using a Varian high- resolution spectrometer operating
at 15. 085 Mc./sec. The spectra for each compound will be presented
in pairs. For the increasing sweep direction the peaks will be point-
ing up and for the decreasing sweep direction the peaks will be point—

ing down.
2 Substituted Acetic Acids

The chemical shifts, spin- spin coupling constants, and peak
assignments for the series of substituted acetic acids are listed in
Table IX. The carbon-13 n.m. r. spectra are shown in Figures 4

through 10 .

T rifluoroac etic Acid

 

The carbon-13 n. m. r. spectra of trifluoroacetic acid are shown
in Figure 4. The most intense peak to the low magnetic field is due to
the carbonyl carbon. The quartet at higher magnetic field is due to

the trifluorom ethyl carbon.

Difluorochloroac etic Acid

 

The carbon-13 n.m. r. spectra of difluorochloroacetic acid are
shown in Figure 5. The spectra consist of four peaks. The carbonyl
peak occurs at lowest magnetic field. The triplet at higher magnetic

field is due to the difluorochloromethyl carbon.

Fluorodichloroac etic Acid

 

The carbon-13 n.m. r. spectra of fluorodichloroacetic acid are
shown in Figure 6. The spectra contain three peaks. The most in-
tense peak to the low magnetic field is due to the carbonyl carbon. The

doublet at higher field is due to the fluorodichloromethyl carbon.

45

 

 

meoo

 

.Eom ofloomouosfiffi mo museum mfiuconnmo 503.32 HA: .w “:5th
.mmo com
I
A O \ om
/ ommouoop I smashes“

 

ENS

 

 

3.1.3

5.7.0

 

ammo

Enho

 

 

meoo

 

 

46

 

 

 

.Eom ofloomoHofiaoosoajfic mo @30on Mascoflhmo .oom\dz HA: .m osapmflh
I. QmNoHUo
e EGQU
3.00 38.80
Amvsommo
Amzoemo
meoo
3:0...er
:zofio
A O
)/ ommonocfi mm
.mflU OON

 

 

 

 

47

.Eom ofioomouozowpouodd mo mspoomm mfiuconsmo .mmoz H .ma

 

 

T 03
QMNUHUMU

 

3.00 5.0.3
3.3er

.mmu CON

 

K
)

 

Eromo

:Eomo

smashes“

$.00

 

.o oudmfim

 

48

Trichloroac etic Acid

 

The carbon-13 n.m. r. spectra of trichloroacetic acid in chloro-
form are shown in Figure 7. The spectra contain two peaks. The peak
for the carbonyl carbon occurs at low field and that for the trichloro-

methyl carbon at higher field.

Dichloroacetic Acid

 

The carbon—l3 n.m. r. spectra of dichloroacetic acid are shown
in Figures 8(a) and 8(b). The most intense peak at low magnetic field
is due to the carbonyl carbon. The high field doublet is due to the

dichloromethyl carbon.

Chloroacetic Acid

 

The carbon-13 n.m. r. spectra of Chloroacetic acid are shown in
Figure 9. The most intense peak at low magnetic field is due to the
carbonyl carbon. The high field triplet is due to the chloromethyl

carbon.

PhenLlacetic Acid

 

The carbon-13 n.m. r. spectra of phenylacetic acid are shown
in Figure 10(a) and 10(b). The spectrum consists of three groups of
peaks. The carbonyl carbon occurs at lowest magnetic field. The
doublet is due to the aromatic ring carbons. The triplet occurring at
higher magnetic field is due to the carbon that has the phenyl group and

two protons attached to it.

49

 

 

 

 

 

:300 0300moso30€u mo 0:033 2:02:00 003.02 Ema .N. enamfih
/ ommon0op m 4 0.33:9: m
meoo
:00
.36 com
. . meoo

 

n60

 

 

50

 

 

 

COZH
)———-1
200 cps. CHC12(1)
CHC13(2)
H increase sf
0

 

 

 

 

Figure 8(a). 15. 1 Mo/sec. carbon- 13 spectrum of dichloroacetic acid.

 

CHc1.(1)
cncnm)

decrease
H %
o

 

 

 

COZH

 

 

 

Figure 8(b). 15.1 Mc/sec. carbon-13 spectrum of dichloroacetic acid.

51

.300 030000H0H£0 mo 0:00am MHIGOQSU

.00m\.0§ Ema .0 Headwfih

 

 

meoo

 

 

.mmo CON

0

 

Fl
/ osmouump 3

50:0 56.48

Exomo

23030

A a a
Gem

 

3:060

for
30.30

Amzoemo 23030

0
E

A

ommouocfl

$.00

 

 

.300 03000H>G0£m mo §h00omm mfiuconnso .00m\.02 Ema .Amvod ousmfih

 

52

 

o
1 033.35 E

0&0 cow

 

N
5 no 33030

5.30 :38 :Eomo

Emu

 

 

:50

 

 

53

.200 030001393 mo 85:00am MHIGOQHMG .00m\.0§ H .ma

 

mo
:50

meoo

 

.mmo OON

 

Emu

  
    

 

o
.A mm

ommohomb

 

5:25

3:20 5.30 £58

21:0

 
  

nHomo

 

.32 $st

 

Table IX. Carbon- 13 Spectra of Some Substituted Acetic Acids

54

 

 

 

 

Compound Group Peak (cps.) (ppm.) JC13_F
(cps.)
CF3COZH COZH 1 511 33. 94:0. 7
CF3(1) 2 818
cr3(2) 3 1106 82.8409 287:1:2
cr,(3) 4 1393
CF2(4) 5 1680
crzcmozn cozH 1 489 32. 240. 5
CFZC1(1) 2 906 297:1:4
crzcuz) 3 1202 79. 640. 4
crzcm) 4 1499
CFClzCOZH cozn 1 461 30. 540. 6
CFC12(1) 2 1179 98. 240. 9 30344
CFC12(Z) 3 1482
CC13COZH COZH 1 447 29. 63:0. 3
CC)3 2 1603 106. 240. 8
1
JC 3-H
CleCOzH COZH 1 368 24. 440. 3
CC12H(1) 2 1853 128.8:I:0.4 181:1:3
CC12H(Z) 3 2034
CClHZCOZH c0211 1 338 23. 340. 7
CC1H2(1) 2 2190
CC1H2(2) 3 2345 155. 4:1:1. 0 155:1:4
OCH-12(3) 4 2500
(bonzcozn cpZH 1 236 14. 340. 3
C ,CH(1) 2 912 60640.5
CH(2) 3 1066 65. 6:1:0. 5 154i3
CHZ(1) 4 2177
CHZ(Z) 5 2313 153.440. 5 13744
0142(3) 6 2451

 

assig
given
Figu)
and s

fiuor

shoe?
Begir
Thef

Ineth

Acen

Figu:
The)
the c
expec
pounc
Centr
tons)

quart

:1:

(D
>4:
Q)
M

/

ShOWn

55

Some Fluorinated Aliphatic Compounds

The chemical shifts, spin- spin coupling constants, and peak
assignments for some fluorinated aliphatic compounds studied are
given in Table X. The carbon-l3 n.m. r. spectra are shown in
Figures 11 through 17. Table XI lists the carbon-13 chemical shifts
and spin-spin coupling constants of some aliphatic compounds and

fluorinated aliphatic c ompounds .

T rifluoroac etic Anhydride

 

The carbon-13 n.m. r. spectra of trifluoroacetic anhydride are
shown in Figures 11(a) and 11(b). The spectra consist of five peaks.
Beginning at low field, the first peak is assigned to the carbonyl carbon.
The four remaining peaks comprise the quartet due to the trifluoro-

methyl carbon.

Ac etylac etone

 

The carbon-13 n.m. r. spectra of acetylacetone are shown in
Figures 12(a) and 12(b). The spectra consist of three groups of peaks.
The two peaks occurring at the lower magnetic field are assigned to
the carbonyl carbons. Two types of carbonyl carbon atoms are to be
expected on the basis of the keto-enol equilibria existing in this com-
pound. The two lines appearing at intermediate field are due to the
central carbon. The doublet is to be expected since one of the two pro-
tons on the central carbon takes part in the keto-enol equilibria. The

quartet occurring at high field is due to the methyl carbons.

Hexafluoroac etylac etone

 

The carbon-13 n.m. r. spectra of hexafluoroacetylacetone are

shown in Figures 13(a) and 13(b). The spectra consist of six peaks.

56

 

CF3(2)
CO
CF3I3)
CF3(1)

CF3I4)

increase
H >
o

 

 

 

200 cps.

 

Figure 11(a). 15. 1 Mc/sec. carbon-13 spectrum of trifluoroacetic anhydride.

 

CFsil)

 

 

 

CF3I4) CO
CF30)
H decrease; CF3(2)
o

 

 

Figure 11(b). 15.1 Mc./sec. carbon-13 spectrum of trifluoroacetic anhydride.

.0Gou000H>u000 mo 93.30090 mfiuconnmd 003.02 Ema ASNH 0ndmfm

 

57

 

\ o
i 0000HUGM m

 

.mmo com

650
O0

:50
3.50 3.50

5 50

 

 

EOU

 

 

58

0:800030000 mo 593025 Mauconumo .00m\02 1m: .AQVNH 0stwm

 

 

00

$00

 

 

 

O

K
A 0mm0n00p m

 

.mmo com

550

Samoa 550

350
3.50.
:50

 

 

59

.0cou003300mouosamx0a mo Eds000mm m: #5930 .00m\ .02 H .2

40:; 0H5wfirm

 

 

E50

:50

5.50

5.00

 

0&0 CON .

O

 

\
I

0mm0hocfi

300

E

 

 

60

.0Cou000300omou03G0x0: mo 83.3025 mausonumu .00m\.02 A .mH .3va 0Hdmfih

 

 

ZOO

Amvmmne

AmvaHV

v
ai

A::5o

.eee oom

Amvmnu

 

 

«)1-

.1/8

508

101

por

61

Beginning at the low field part of the spectrum, the first peak is that
of the carbonyl carbon atom. The next three peaks are part of the
quartet from the trifluoromethyl carbons. The fourth member of this
quartet cannot be seen due to overlap with the doublet from the central
carbon. The two peaks occurring on the high field portion of the

spectrum are due to the central carbons.

1, 2 Dibromo- 1, 1, 2, 2, - tetrafluoroethane

 

The carbon-13 n.m. r. spectra of 1, 2-dibromo—l, l, 2, 2, -tetra—
fluoroethane is shown in Figure 14. The spectrum consists of three
peaks that comprise the triplet to be expected due to the spin- spin inter-

action between carbon- 13 and the bonded fluorine nuclei.

2 , 2 , 2 - T rifluoroethanol

 

The carbon-13 n.m. r. spectra of 2, 2, 2-trifluoroethanol are
shown in Figures 15(a) and 15(b). The spectra consist of two groups of
peaks. The quartet occurring on the low field portion of the spectrum
is due to the trifluoromethyl carbon. The triplet occurring on the high

field portion of the spectrum is due to the alcoholic carbon.

1, 1, l-Trifluoro-Z- propanol

 

The carbon-13 n.m. r. spectra of 1, 1, l-trifluoro-Z-propanol are
shown in Figures 16(a) and 16 (b). The spectra consist of two quartets
and a doublet. The quartet occurring at the low field portion of the
spectrum is assigned to the trifluoromethyl group. The doublet is due
to the alcoholic carbon. The second quartet occurring at the high field

portion of the spectrum is due to the methyl carbon.

62

 

 

 

 

 

 

 

CF 1
CF2BI’(2) 2Br( )
1
CF r‘ ) CFZBr(
CFzBr(3)
CF2B1‘(2)
increase decrease
‘r H >
o O
4____J
200 cps.
Figure 14. 15.1 Mc./sec. carbon-13 spectra of 1, 2-dibromo-1,1, 2, 2—

tetrafluoroethane.

 

63

.Hocmaamouosfiwuu- .N .N .N mo €350on 2-4909th .ovm\.02 1mg .3va vndmwh

 

 

 

 

 

[.mmo CON
A omMopufi
sEONmo
n
Emommu 25098 3 mo

 

5P3

Esme

O

I

 

 

64

.Hocmsuvouosflcuum .N .N mo 8.950on Mauconnumu .oom\.o§ imd .3va 0&9me

 

 

game

820

 

 

ammo

 

EMS

I,

.mmo com

o
Acmmvhuov E

 

Emofo EmONmu

330:5

 

 

65

.HOCMQOHQuNuouOSGCHJ J J Ho 9330QO mauconymo .uvm\.02 H .m~ 43A: 059me

 

 

5:15 @130
amen Asmomo ammo
ammo Aavnmo
:mmo .Cmomu :1 o
Tll. A ommmnoov om

 

 

 

.Hocmmoumumuouogflhud J J we 5.930QO mauaonnmo .uom\.02 12 4.3.: vudwwh

 

   

Ede 3an

Asmomu
ammo
Emmo
O

O a \
E / UvaHuGfi m

ammo Asmomo

 

 

 

 

Fig
the

the

66

1,1, 1 -Trif1uoro- Z-methyl- Z— prOpanol

 

The carbon-13 n.m. r. spectra of 1, l, l-trifluoro-Z-methyl-Z-
propanol are shown in Figures 17(a) and 17(b). The spectra consist
of nine peaks. The quartet occurring at the low field portion of the
spectrum is due to the trifluoromethyl carbon. The single peak at the
intermediate field portion is due to the alcoholic carbon. The quartet
occurring at the high field portion of the spectrum is due to the

methyl carbons.

a -Substituted Toluenes

The chemical shifts, spin-spin coupling constants, and peak
assignments for a series of a- substituted toluenes are listed in Table

XII. The carbon-l3 n.m. r. spectra are shown in Figures 18 through 25.

Toluene

The carbon-l3 n.m. r. spectra of toluene are shown in Figures
18(a) and 18(b). The spectrum consists of two groups of peaks. The
lower field peaks are those of the aromatic carbons. The quartet at

high field is due to the aliphatic carbon.

Benz ylbromide

 

The carbon-13 n.m. r. spectra of benzylbromide are shown in
Figures 19(a) and 19(b). The doublet at low magnetic field is due to
the aromatic carbons. The triplet occurring at higher field is due to

the aliphatic carbon.

Benzylchloride

 

The carbon-13 n.m. r. spectra of benzylchloride are shown in

Figures 20(a) and 20(b). The aromatic carbon resonances are in the low

 

/

‘

Suva: nu

P-<Il‘

67

.Hocmmonmumuafioaum-onodflvfiuH J .H mo 5330on Mansonsmno .oom\o§ H .3

.3: Sumsm

 

 

Enmo

 

3 no
335

 

ammo

$00

0
\
“ vmmohofi m

 

.mau com
Tllllll.

Enmo

Enmo

Ede
Enmo

 

Jocmmoumumurfifiogum-OMOSGCTJ J J mo 89.30QO mfiucoofimo .oom\.o§ Add 40;: whamfim

 

68

 

 

.mmo CON
Tl;
K o
/ vmmopoop 3
5:5
300
Enmo Enmo
_

 

ammo Ears

Ears
SP3 :vnmo

 

 

 

 

 

Table X. Carbon-13 Spectra of Some Fluorinated Aliphatic Compounds

 

 

 

vC 6C Joli}? JcilH
Compound Group Peak (cps. ) (ppm. ) (cps. ) (cps. )
(CF3CO)2 (3:0 1 699 48. 941. 3
CF3(1) 2 823
CF3(2) 3 1109 28641
CF3(3) 4 1394 82.9i0.5
CF3(4) 5 1680
(CH3CO)ZCHZ c=0 1 -37.0 -2.4540.5
C—OH 2. 120 80.0i0.4
CH(I) 3 1391 97.640.5 16444
CH(2) 4 1554
CH3(1) 5 2409
CH3(2) 6 2536 172.440.8 12848
CH3(3) 7 2668
CH3(4) 8 2796
(oncohcn,Z C-OH 1 263 17.44o.9
CF3(1) 2 752
CF3(Z) 3 1045 77.941.2 28344
(3173(3) 4 1317
(314(1) 5 1441 101.241.1 17244
CH(2) 6 1614
CFZBrCFZBr CF2(1) 1 956
0172(2) 2 1249 83.440.6 30348
CF2(3) 3 1662
CF3CHZOH CF3(1) 1 670
CF3(2) 2 953 72.640.5 28341
CF3(3) 3 1245
CF44) 4 1518
CHZOHU) 5 1914
CHZOH(2) 6 2057 136.340.5 14342
CHZOH(3) 7 2201
CF3CH(OH)CH3 CF3(1) 1 654
CF3(2) 2 934 71.240. 5 28043
CF3(3) 3 1214
CF3(4) 4 1494
CHOHU) 5 1891 129.5i0.6 127i2
CHOH(2) 6 2017
0143(1) 7 2442
CH3(2) 8 2562 173.840.? 12046
(3413(3) 9 2681
(3113(4) 10 2801

 

continued

11

Co

CF

Table X - Continued

70

 

 

 

V6 5 c Jc1217 Jc1114

Compound Group Peak (cps.) (ppm.) (cps.) (cps.)
CF3C(CH})ZOH 0173(1) 1 643

(353(2) 2 922

(3173(3) 3 1200 70.340. 5 27842

(2173(4) 4 1478

COH 5 1888 125. 2:1:0. 9

(3143(1) 6 2429

CH3(2) 7 2554 173.840.4 12641

(3143(3) 8 2680

CH3(4) 9 2806

 

\\

 

 

Table XI. Carbon—13 Chemical Shifts and Spin-Spin Coupling Constants
of Some Aliphatic and Fluorinated Aliphatic Compounds
CH3COZH CF3COZH

15H: 130 (78)

(CH,C0),o

CH, 172(5)
JCH 130(5)
C02 26(5)

(CH,C0),CH2

CH, 172
CHz 98
co -2.5
JCH3 128
J 164
CH
corf8.o

CH,CH,OH

CH, 176(80)
JCH3 125(80)

CHZOI-l 137(80)
JCHZOH 140(80)

(CH,),CH0H

CH3 169(80)
CHOH 131(80)
JCH3 124(80)
JCHOH 147(80)

CF,(78)(5); 82.8
JCF3 283(91); 275(5); 287
coz 28(5); 34

(CF,CO),o

CF, 829
CF, 284(78) 286
C0, 49

(CF,C0),CH,

CF, 78
JCF3 283
CH, 101

JCHZ 172

COH 17

CF,CH,0H

CF3 73
JCF 283
CHzéH 136

JCHZOH 143
CF,CH(0H)CH,

CH, 173
CHOH 129

JCH3 119

JCHOH 126
CF, 71

JCF3 280

 

continued

Table XI - Continued

72

 

(CH,),C- OH

CH, 162(80)
JCH3 125(80)
COH 125(80)

CH,C1CH,C1

CHZCI 148(78)
JCHz 154(78)

CF‘3C(CH3)ZOH

CH, 174
JCH3 126
COH 125
CF,70

JCF3 278

CFzBrCFzBr

CFzBr 83
JCFZ 303

 

73

.ocmdaou mo 53.30QO mfiuconhmo .oom\.02 ima .vawH oudmfih

 

 

$446

5430 3:40

5.30

\

4
ommouocw

\

O

I

.646 com

Emu

 

mo
:50

 

 

74

6:04:06 Ho 9330on mansoofiamo .omm\.oz H6.” .3va oudmfim

 

 

Emu

 

1 0
m0 } ommonoov I

:50

 

.mmo oom

ammo

Esme
£58

3430

 

 

.oEEoungnsmo. mo 53.30on mausonumo .oom\ .62 H .5 .Amme ousmfm

 

75

 

.mmo cow.

5th440
35430

5.5430

 

Emu

O

k
/ mmmouocfl TH

 

 

Lm

m0
Emu

 

 

 

76

.CBEOHQENCCQ mo 55.30on

Mansonnmo

66656212 483 6.34:.”

 

 

:Eu

 

Emu

 

£6

.666 cow

0

\
/ mmmohomp E

 

armfio
55.30
35:40

 

 

77

 

CH(l)

 

 

200 cps.

 

CH(2)

 

 

H

0

 

1nC reaseIL

CH,C1(2)

CHZCI(1)

 

 

Figure 20(a). 15.1Mc./sec.

carbon- 13 spectrum of benzylchloride.

78

 

 
  

    

CH2C1(1)

 

 

 

CHZCL(3
CH2C1(2)
3
Ho decrease %
CH(l)
r———o
200 cps.
CH(2)

 

 

 

Figure 20(b). 15. 1 Mc. /sec. carbon-13 spectrum of benzylchloride.

79

field portion of the spectrum. The peak for the substituted carbon is
barely observable. The other two more intense peaks are due to the
remaining five carbon atoms of the ring. The triplet occurring in the

higher field portion of the spectrum is due to the aliphatic carbon.

Benzylmercaptan

 

The carbon- 13 n.m. r. spectra of benzylmercaptan are shown in
Figures 21(a) and 21(b). The peaks for the aromatic carbons occur in
the lower field portion of the spectrum. The peak occurring at lowest
field is that of the substituted carbon. The other two more intense
peaks are due to the five remaining ring carbons. Theii triplet occurr-
ing in the higher field portions of the spectrum is due to the aliphatic

carbon.

Diphenylmethane

 

The carbon-13 n.m. r. spectra of diphenylmethane are shown in
Figures 22(a) and 22(b). The three peaks assigned to the aromatic
carbons are composed of a single peak for the substituted carbon and a
doublet for the five remaining ring carbons. The triplet occurring at

highest field is due to the aliphatic carbon.

Benzylcyanide

 

The carbon-13 n. m. r. spectra of benzylcyanide are shown in
Figures 23(a) and 23(b). The doublet occurring in the lower field portion
of the spectrum is due to the aromatic carbons. The triplet occurring

at high field is due to the aliphatic carbon.

80

.GmpmmoHoEH>Nson mo 85.30696 mH uCOnHHmo .omm\ .02 H .mH

.HmvHN ouderm

 

 

EmmNmo :Emfio
Hmvmmmmo

o
4 ommoHoCH HH

 

.mnHo OON

Emu

 

Emu

 

 

.GmummonEHCHNCon Ho 5530QO mH Iconumo .oom\ .02 H .mH

.SIN 634E

 

81

 

:50

Emu

)

 

  
       

 

.mmu CON

0
\
4 ommouoop TH

N
Emm monEwNCU

 

 

82

.ocmfiuoEHCHCoHHQHC mo 8.9.3.0on mHuconHmo .oom\.oH>H H .mH

.Hmvmm madmfm

 

 

Nmu

 

\
I

ommoHoCH

O

I

Emu

.646 com
1||+

 

SHED

Mme

 

 

83

.oCCHHoEH>CoHQHC mo gnuoomm mHucooflHwo .oom\1.oH>H H .mH .HQVNN oudeh

 

 

N.mo

 

.23 com
1., o
/ ommohoop H.H
Emu
Emu
U
6 £0

 

 

84

.oCH:m>oH>Nsvn Ho 5530on mHusonHmo .oom\.oH>H H.mH .7“.va oudmfim

 

 

Euro
:35
£38

 

K
ll

vmmoHoGH

0

mm

.wmo com

.llln.

Emu

 

 

Emu
mo

 

 

85

.opHCm>oH>NCon mo 85.30on mHuconHwo .oom\.uH>H H .mH

Ham... 634E.

 

 

Emu

 

 

Emu

 

 

O

 

ommmnuop H.H

Efo
3.248

Efo

 

 

86

Benzylamine

 

The carbon-13 n. m. r. spectra of benzylamine are shown in
Figures 24(a) and 24(b). The peaks for the aromatic carbons occur in
the lower field portion of the spectrum. The peak occurring at lowest
field is that of the substituted carbon. The other two more intense
peaks are due to the five remaining ring carbons. The triplet occurr-

ing in the higher field portion of spectrum is due to the aliphatic carbon.

Benzylalcohol

 

The carbon-13 n.m. r. spectra of benzylalcohol are shown. in
Figure 25. The aromatic portion of the spectrum, occurring at low
field consists of three peaks. The peak at lowest field position is due to
the substituted carbon. The other two more intense peaks are due to the
remaining five ring carbons. The triplet occurring at high field is due

to the carbon atom of the CHZOH group.
Diphenyl Compounds

The peak positions, observed and calculated peak heights, and
peak assignments for a series of diphenyl compounds are listed in
Table XIII. The chemical shifts for the various ring carbons are listed
in Table XIV. The 15. 1 Mc. /sec. carbon-13 spectra for these com-

pounds are shown in Figures 26 through 29.

Diphenylether

 

The carbon-13 n.m. r. spectra of diphenylether are shown in
Figure 26. The spectra consist of six distinguishable peaks. The assign-
ments given are based on the spectra of phenol which were analyzed in

detail by Lauterbur (8) with the aid of methyl substitutions.

87

 

 

 

 

CH(l)
CH(2)
H increase >
o
C8
CHZNHZ(2)
CH2 (1)
HZNHZ(3)

{-————4
200 cps.

 

 

 

Figure 24(a). 15. 1 Mc./sec. carbon-13 spectrum of benzylamine.

88

 

CH,NH,(1)
‘CH,NH,(3)

CH,NH,(2)

 

 

 

 

H decreasekr
o
CH(l)
1—1
200 cps. CH(2)

 

 

 

Figure 24(b). 15. 1 Mc./sec. carbon-13 spectrum of benzylamine.

89

.HoaHoonH>Nc®nH mo whuoomm mHucoonHmu .uom\.oH>H H .mH .mN oudmfim

 

 

Em

 

Emu

U

k

0

A
vmwmhuvv

Emofo

:EONmo

E

 

 

 

K

i
omonoGH

€30.35

0

TH

 

.mmo com

«I'll

Emu m

 

 

 

:Eo

 

 

90

 

 

 

Table XII. 15.1 Mc./sec. Carbon-13 Spectra of Some a-Substituted
Toluenes
V”c 6 c JCU-H
Compound Peak Group (cps. ) (ppm. ) (cps .)
Toluene 1 CS 706 60.140. 3
2 CH(l) 958 68.840.4 16141
3 CH(2) 1118
4 CH,(1) 2474
5 CH,(Z) 2596 176.140. 5 12241
6 CH,(3) 2718
7 CH,(4) 2839
Benzylbromide 1 CS 890 59. 040. 5
Z CH(I) 943 67. 940. 3 16241
3 CH(2) 1105
4 CH,(1) 2302
5 CI-lz(2) 2456 162. 840. 5 15443
6 CHZ(3) 2610
Benzylchloride 1 CS 894 59. 343
2 C140) 938 67. 643 16242
3 CH(2) 1100
4 CH,(1) 2125
5 CH,(2) 2275 150. 640. 5 15145
6 CHZ(3) 2426
Benzylmercaptan 1 CS 835 55.440. 2
2 CH(l) 958 68. 740. 2 15943
3 CH(2) 1116
4 CH2“) 2383
5 CH2(2) 2523 167. 240. 3 14042
6 CHZ(3) 2663
Diphenylmethane 1 CS 796 52. 842
2 CH(l) 907 65. 543 16843
3 CH(2) 1069
4 CH,(1) 2204
5 CHZ(2) 2332 154. 640. 0 13042
6 CH2(3) 2454

 

continued

91

Table XII - Continued

 

 

 

 

Vc 6c JCl3-H
Compound Peak Group (cps. ) (ppm. ) ( cps.)
Benzylcyanide 1 CS 947 62. 843
2 CH(1) 947 67.940.11 15742
3 CH(2) 1104
4 CH,(1) 2488
5 CH,(Z) 2620 170.740.6 13243
6 CH,(3) 2753
Benzylamine 1 Cs 802 53. 240. 5
2 CH0) 967 69.540.5 16145
3 CH(2) 1129
4 CH,(1) 2137
5 CH,(z) 2273 150.740.8 13644
6 CH2(3) 2408
Benzylalcohol 1 Cs 841 55.740.4
2 CH(l) 963 69.1-40.5 15741
3 CH(2) 1121
4 CH,(1) 1868
5 CH2(2) 2010 133.241.0 14142
6 CH3(3) 2151

 

92

.HoaHuoHlaaofiHmHHu «0 0300mm mHunonHmu .0om\.0H2 H .mH 6N musmHh

 

 

$

Iommouoov

nu

3m

 

.m

336
36 .N

0
HH

 

 

 

H:

56:“.

 

 

_

56 .N H34

 

3.: .m
:3.~

\
I

an .m

.mmo com

o
unmouoaH

mm

 

 

 

93

Diphenyl sulfide

 

The carbon-13 n.m. r. spectra of diphenylsulfide are shown in
Figure 27. The spectra consist of two peaks indicating that the five
ring carbons are equivalent. The substituted carbon overlaps the

lower peak of the doublet from the ring carbons.

Diphenylselenide

 

The carbon-13 n.m. r. spectra of diphenylselenide are shown in
Figure 28. The spectrum consist of four partially resolved peaks.
In the assignment of the peaks advantage is taken of the fact that
resonance frequencies of meta carbon atoms have a range of 2.4 ppm. (14).
The 15. 1 Mc./sec. carbon-13 chemical shifts of meta carbons occur
near 65 ppm. on the carbon disulfide scale. Therefore the lines giving
a chemical shift near 65 ppm. relative to carbon disulfide will be

assigned to the meta carbon.

Diphenylmercury

 

The carbon-13 n.m. r. spectra of diphenylmercury are shown in
Figure 29. The substituted carbon resonance occurs at a lower

magnetic field than the other ring carbons.

T riphenyl Compounds

The peak positions, observed and calculated peak heights, and
peak assignments for some triphenyl compounds studied in this research
are listed in Table XV. The chemical shifts for the various ring
carbons are listed in Table XVI. The 15.1 Mc./sec. carbon-13 spectra

for these compounds are shown in Figures 30 through 35.

94

.mvaHSmH>C0nHQHv «0 8.3.0090 2:654:00 .00m\ .02 H .mH

.3. 6.33.4

 

 

Emu

SEC

0
A I
0mmvnoop

 

mmo

 

3:40

 

 

 

Emu

\

11 0mm?“ 00H

0

HIH

 

 

95

.0H0H00H00H>COHHQHH0 Ho 0300mm mHuconHHmo .0om\.0H2

12 .3 654E

 

 

4m0

0

 

0mmou0op

v

 

8

.AmmmouosH

0

mm

.68 cow.
.11..

Nmo

Em.m :34 .4

area .m
Em; 5:. .m
36:“. “6.0 .H

 

 

3:50.58 100:me Ho 08.00am mHnsonHmo .00m\ .02 H .mH .om whamHh

 

96

 

£0
0
1030.000 H.H

 

 

56.2.34
Em.m

 

 

 

SKYN

2:. 3:.

 

34
3.3:“.
:35

\

1’

 

0m00H0CH

.mm0 com
.11

 

 

97

 

 

 

Table X111. 15. 1 Mc./sec. Carbon-13 Spectra of Some Diphenyl Compounds
Height

Compound Peak (cps .) Obsd. Calcd. Assignments
Diphenylether 1 555 1. 70 2.00 C—1

2 894 2.15 2.00 CI-I-3, 5(1)

3 1008 0.90 1.00 . CPI-4, (1)

4 1054 4.20 4.00 CH-3,5(2);CH-2,6(l)

5 1159 1.05 1.00 CH-4(2)

6 1222 2.00 2.00 CI-l-Z, 6(2)
Diphenylsulfide 1 888 6.61 7.00 C—1;CH-2,3,4,5,6(1)

2 1053 5.39 5.00 CH-2,3,4,5,6(2)
Diphenylselenide 1 857 4.16 4.00 C-l;CI-l-2, 6(1)

2 905 2.71 3.00 CH-4(l);CH-3,5(1)

3 1004 1.19 2.00 CH-Z, 6(2)

4 1067 3.93 3.00 CI-I-4(2);CH-3, 5(2)
Diphenylmercury 1 340 1. 73 2. 00 C-1

2 779 2. 28 2.00 CH-Z, 6(1)

3 922 4.29 5.00 CH-4(1);CH-2, 6(2);

CPI-3, 5(1)
4 1072 3.70 3.00 CH-4(2);CI~I-3, 5(2)

 

98

Table XIV. Chemical Shifts (ppm.) and Spin-Spin Coupling Constants

(cps.) For Some Diphenyl Compounds

 

 

 

Compound Cs Ortho Meta Para Cn-H
Diphenylether 36. 75.6 64.7 71. 160
Diphenylsulfide 58. 64. 2 64. 2 64. 165
Diphenylselenide 56. 61. 7 65. 3 65. 157
Diphenylmercury 22. 56. 4 66.1 66. 148
Diphenylmethane 52. 65. 5 65. 5 65. 162

 

99

T riphenylamine

 

The carbon-13 n.m. r. spectra of triphenylamine are shown in
Figure 30. The spectra consist of five peaks. The first peak has
been assigned to the substituted carbon. The second peak is the first
of the doublet due to the meta carbon. The third peak includes the
first members of the doublets for the ortho and para carbons. The
fourth peak is the second member of the meta carbon doublet. The
fifth peak is the second member of the doublet for the ortho and para

carbons.

T riphenylphosphine

 

The carbon-13 n.m. r. spectra of triphenylphosphine are shown
in Figure 31. The spectra consist of four peaks. The first peak is
due to the first members of the doublets from the ortho and para
carbons. The second peak is due to the first member of the doublet
from the meta carbons. The third peak is due to the second members
of the doublets from the ortho and para carbons. The fourth peak is

the second member of the doublet for the meta carbons.

Triphenylarsine

 

The carbon—13 n.m. r. spectra of triphenylarsine are shown in
Figure 32. The spectra consist of four peaks with assignments as
follows: The first peak is due to the first members of the doublets
from the ortho and para carbons. The second peak is due to the first
member of the meta carbon doublet. The third peak is due to the second
members of the doublets from the ortho and para carbons. The fourth

peak is due to the second member of the doublet from the meta carbons.

6558130433» mo 0:00am mHuconumo .00m\.0H>H H .mH .om 4.;th

 

 

100

 

 

.36 SN

_ . 5:36 N .

£38 .
:FEEJ .
. Ema .
m To .

HNMV‘LO

 

1A m
0m00h000 m

0

81
J 0mmohucH TH

 

 

 

 

101

.msfinamogma>G0£QHHuwo mnuoomm mHuconHmo .00m\.0v& H.mH

.3 654E

 

 

£0

18
I

ommuu0op

0

mm

 

 

 

.9? COM
.1.

RI

I

33....
36.234
Ems
36.82:.

ammonocH

O

 

mm

 

 

102

 

.0sHmHmH>00Hmwuumo mhpoomm mH1Co3Hmo .00m\.0<4.H.mH

.Nm 634;

 

 

k
/

0m800000

0

4H

 

 

.mmu oom

 

 

 

 

 

\
1A

Ems
Staged
Ema
:ILSEJ

0m00H0CH

0

mm

.4
.m
.N
.H

 

 

 

 

103

Triphenylstibine

 

The carbon-13 n.m. r. spectra of triphenylstibine are shown in
Figure 33. The spectra consist of five peaks with assignments as
follows: The first peak is due to the first member of the doublet from
the ortho carbons. The second peak is due to the first member of the
doublet from the para carbon. The third peak is due to the first member
of the doublet from the meta carbons. The fourth peak is due to the
second member of the doublets from the ortho and para carbons. The
fifth peak is due to the second member of the doublet from the meta

carbons.

T riphenylbi smuthine

 

The carbon-13 n.m. r. spectra of triphenylbismuthine are shown

in Figure 34. The spectra consist of five peaks with assignments as
follows: The first peak is due to the first member of the doublet of the
ortho carbons. The second peak is due to the first member of the doublet
from the para carbon. The third peak is due to the first member of the
doublet from the meta carbons and the second member of the doublet
from the ortho carbons. The fourth peak is due to the second member of
the doublet from the para carbon. The fifth peak is due to the second

member of the doublet from the meta carbons.

T riph enylm ethane

 

The carbon-l3 n.m. r. spectra of triphenylmethane, aromatic
portion only, are shown in Figure 35. The spectrum consists of three
peaks with assignments as follows: The peak occurring at lowest magnetic
field is due to the substituted carbon. The other two more intense peaks
are due to the remaining five carbon atoms of the ring. The aliphatic

C13 doublet was at higher field and is not shown in the figures.

104

.ocHnHumH>c0~HmHnu mo muuoomm mHLuofiumu .0om\.02 H .mH .mm 0.33%

 

 

£0

\
A

ommouoov

O

 

 

m 36 con 541m .m
T1 Etfiid .4
:35 .m
_ 2:. .N
36:“. .H

 

 

 

 

   

o
1.
A. coach 2: H.H

 

 

 

 

 

105

055953313043?» mo ouuoomm mHuconumo .000\ .02 H .mH

.3 35E

 

 

Nm0

K
I4

ammouuov

4H

 

.36 com
Tlllli

H84

Ema
34236:“.
341m
:SJ

N

‘

ommokunH

.
o-CNMV‘ID

N.610

o
mm

 

 

 

.vqmfiofiacmxmwuu mo muuommm mdunonumo .umm\.o§ H .2 .mm 9:.th

 

106

 

.mmo CON

0

N
/ ommouoov E

 

Nmo

Emu Emu

Emu

 

o
N
lmmdmu a”: E

 

 

Emu

 

 

 

107

 

 

 

Table XV. 15.1 Mc./sec. Carbon—13 Spectra of Some Triphenyl Compounds
V: Height

Compound Peak (cps. Obsd. Calcd. Assignments
Triphenylamine 1 701 3. 30 3.00 C-1

2 904 3.66 3.00 CH-3, 5(1)

3 984 4.10 4.50 CH—4(1);CH-2,6(l)

4 1062 2.41 3.00 CH-3, 5(2)

5 1147 4.53 4.50 CH-4(2);CH-2,6(2)
Triphenylphosphine 1 842 5.00 4. 50 CH-4(1);CH-2, 6(1)

2 907 3.85 3.00 CH-3, 5(1)

3 994 1.65 4.50 CH-4(2);CH-2,6(2)

4 1074 4.40 3.00 CH-3, 5(2)
Triphenylarsine 1 824 4.45 4.50 CH-4(1);CH-2, 6(1)

2 904 4. 35 3.00 CH-3, 5(1)

3 990 1.89 4.50 (SH-4(2), CH-Z, 6(2)

4 1071 4. 30 3.00 CH-3, 5(2)
Triphenylstibine 1 793 3.43 3.00 CH—Z, 6(1)

2 826 1.78 1.50 CH-4(1)

3 907 2.89 3.00 CPI-3, 5(1)

4 956 3.03 4.50 CH-Z, 6(2);CH-4(2)

5 1065 3.89 3.00 CH-3, 5(2)
Triphenylbismuthine 1 768 2. 76 3.00 CH-Z, 6(1)

2 882 3.06 1.50 CH-4(l)

3 931 3.72 6.00 CH-Z, 6(2);

CPI-3, 5(1)

4 1040 2. 39 1.50 CPL-4(2)

5 1074 3.06 3.00 CH-3, 5(2)
Triphenylm ethane 1 754 C -l

2 906 CH-2,3,4,5,6(1)

3 1068 CH-2,3,4,5,6(2)

4 2024 CH(l)

5 2169 CH(2)

 

108

Table XVI. Chemical Shifts (ppm.) and Spin-Spin Coupling Constants
(cps.) of Some Triphenyl Compounds

 

 

 

Compounds CS Ortho Meta Para JCU-H
Triphenylamine 46.5 70.7 65.2 70.6 150
Triphenylphosphine ---- 60.9 65. 7 60.9 156
Triphenylarsine ---- 60.1 65. 5 60.1 166
Triphenylstibine ---- 58. 0 65.4 59. 0 150
Triphenylbismuthine —--- 56. 3 66. 4 63.7 155

Triphenylmethane 50.0 65.4 65.4 65.4 162

 

 

109

Alcohols

The chemical shifts for some aliphatic alcohols are listed in
Table XVII.

It has been empirically found by Malinowski (93) that carbon—13
proton spin- spin coupling constants, for compounds of the type CHXYZ,

can be separated into three components according to the equation

qu-H = 7x133?L ‘72 (56)

In equation (56) J is the carbon-13 proton spin- spin coupling

13
constants; compofent:I 7X, 3Y' 32 are contributions which are
associated with substituents X, Y, and Z, respectively. The additivity
relation expressed in equation (56) has been derived using valence bond
theory by Juan and Gutowsky (94). The valence bond approach used
with this model gives a linear relation between the 5 character of
the carbon hybrid orbital involved in the C-H bond (aHz ) and the
observed CU-H coupling constant.
= 500 a

J (57)

C13'H HZ

Where J is the observed Cl3—H coupling constant and 0H2 is the

c‘3-H
fractional 5 character of the C-H bond.
The experimental values for the spin- spin coupling constants

J for the aliphatic carbon in the a-substituted toluenes are listed

C‘3-H
in Table XVIII. The calculated spin-spin coupling constant and the
fractional s-bond character based on the assumption that the contact term

is dominant are also listed in Table XVIII.

110

Table XVII. Chemical Shifts for Some Aliphatic Alcohols);

 

 

 

 

C-O C2 C3

Alcohol (ppm.) (ppm.) (ppm.)

Methyl 145 --- ---

Ethyl 137 176 --- (F
Iso-propyl 131 169 --—-

n-Propyl 129 167 183

n-Butyl 132 158 --- :-
iso-Butyl 124 162 174

sec-Butyl 125 167 175

tert-Butyl 125 167 175

 

:1:
These data were supplied by J. Stothers (80).

111

Table XVIII. Calculated and Observed Spin-Spin Coupling Constants in
a-Substituted Toluenes

 

 

 

JCl3-H(Cps') JC13-H(Cps°) Fractional
Compound Calculated Observed s-Character
¢CH3 126 122 0. 244
(bongo 127 130 o. 260
cbcnzcozn 131 137 0.274
CbCHZNI-Iz 134 136 o. 272
(bCHZCN 137 132 o. 264
cbcnzsn 139 140 o. 280
(bCHZOH 144 141 o. 282
6CHZC1 153 151 o. 302

¢)CHZBr 153 154 o. 308

 

112

An attempt has been made to correlate the carbon-13 chemical
shifts with the molar susceptibility of the molecules,__.YSLithin a series
the change in molar susceptibility should reflect the change in bond
magnetic anisotropy. The results are shown in Figures 36 through 40.
Figure 36 shows the correlation of the carbon-13 chemical shifts with
molar susceptibilities (81) for some aliphatic alcohols- Figures 37
and 38 show the correlation of the carbon-l3 chemical shifts with
molar susceptibilities for the substituted carbon and the carbonyl
carbon atoms in some substituted acetic acids, respectively. Molar
susceptibility data are not available for all of the compounds studied
in this research but have been estimated from Pascal constants in some
cases. Figure 39 shows a linear relationship between carbon-13 chemi-
cal shifts and molar susceptibilities for some phenylmethanes. Chemical
shift data for tetraphenylmethane are not available due to its low solu-
bility. Figure 40 shows the correlation of carbon-13 chemical shifts
with molar susceptibilities for some diphenyl and triphenyl compounds.

It has been shown that carbon-13 chemical shifts depend in part,
on the electronegativities of the group substituted on the carbon (13, 14).
The carbon-13 chemical shifts for the substituent carbons and for the
carbonyl carbons in substituted acetic acids have been plotted versus
the group electronegativities (82) in Figures 41 and 42, respectively.
Figure 43 shows the correlation between carbon-13 chemical shifts of
the aliphatic carbon atom and the group electronegativities (83) of
the CHZX group in the OCHZX series. Figure 44 shows the correlation
between carbon-13 chemical shifts of the ortho carbon atoms and
Pauling's electronegativities (84) for the central atom of a series of

diphenyl and triphenyl compounds.

 

113

.303002 03.3334 080m HON mowfidnwumvumdw HMHOE msmum> muflfim HMdeumSU .om 0.8:th
EH N ax...
00 cm ON 00 cm ow om ON 0..” o

. . . . . . 1 . . . . 2:
Endgamé436624683629A J 4 :3

3585-9339694 .2 .H 8: .

Hofifimouofimueé .N .N E

aofimoumé £5620 E .o:

Hondmonm gnawed/TN .3

3535A .3 .

3533-: 3..

39865-». a;

 

#1 4 u a q a

             

.02
aondmoumua A3
. 353
3566...... E 5 .o
35:32 A:
.62
n
.3;

1 9:

 

 

 

114

200 r

 

190 ~
180 ~
170 '.
160 '

.150 —

140 -

   

130 _ cmrco,11

(ppm.)
120-

. ‘7
n

 

110 "
100*

90 ~
cr,oo,n

so. 1

70'

60-
\

so;

40.

 

 

3O1 1 1 1 1 I I .- 0 .C W11- .
0 10 20 ‘30 40 50 60 7O 80 90 100 .110 120
~x. 2 10°

Figure 3?. OX. Chemical Shifts Versus Molarh'Susceptibilitie's
in Some Substituted Acetic Acids.

115

 

 

 

 

80 - .
70-
6o.
50'
0
(ppm.)
40—
01?,ch
6 0301.00.11
¢
30» 01',0100,H ’5
00
o . a 1.03.8
. 01m,oo,n
zo~ 0150100,}!
9
03300.11
10» -
1 l 1 a I
o 20 4o - 6o 80 . 100 120
ix. x 10'

Figure 38. Chemical Shifts of Carbonyl Carbon Atoms Versus
Molar Susceptibilities for Some Substituted Acetic Acids.

116

200 .

 

U

190

150 '

140

 

.130-

 

 

120 ' 1 . . .
60 80 100 ,120 '140 160 180
-11. 2’10“ ~

I

 

Figure 39. The Aliphatic Carbon Atom. Chemical Shifts
Versus Molar Susceptibilities in Some Phenylmethanes.

110

117

 

100 1

90'

801'

70*

5o
O

60*

50.

40)

 

30 -
100

003,89 0 .

' 1

120 140 160 , 180
ax. : 1C,

'0

«p.31 ‘

. 200

 

220

Figure 40. Chemical Shifts of Ortho Carbons Atoms Versus

Molar Susceptibilities for'Some Diphenyl and Triphenyl

Compounds.

. . .5: 636... 2.5664
wouaflmadm mo .mownom d HON hugfldmoaonuovflm 95.20 mdmue> ”GEM 328030 «X0 .3. onfiwmh
rgwudwoaohugm 95.3”

cm .m co .m cm .N 00 .N o

 

cm

2.
2:
SH Aév

.0.
0mm

1.18

 

mpg

 

com

1 mNN

1

omN

 

 

 

. - th

119

. .5: 636... 6:34 363036 mo 328 a.
new hugflmwoaouuoeflm macaw mdmuo> m53< £03.30 7:39.30 mo “HEW #038030 .Nv mafimfm

bgflmwoconuooam 990.20
cm .m as .m om .N - co .N o

4

. 2
ado-mu
_ o

 

eased".

 

a...
3.3

 

 

 

120

4me monogoh
wouafiumnam . U 5 figfldwmnouuomfim macho m5mh0> 53w 33.9250 X n30 .mw onamwh

, . humzudwmconuuofim @5050
cm .v mN .v co xv mu .m cm .m mm .m co .m ms .N om .N‘ mm .N
1 + . 1 A q . OCH

.7

4 1

o2 .
23.
3:5

9:

cm“

 

.. .OON.

 

 

 

121

m.m

.mvndomEoU 1:333." .H. van 15:35 080m new 83< Hahn—~30
mo >3>3dmoaouuoo~m n.mafidmm mamno> m83< nonumo 9*qu mo mufinm 3.35050 4% uudmwm

figfldwoao 503m m .mfigdnm

o.m

m.N

o.N

m4

 

 

1

 

 

 

3»

co

DISC USSION OF RESU LT S

Solvent Studies of Ethyl Compounds

The data in Table VI and Figure 3 show that the effects of the
solvents chloroform, carbon tetrachloride, and cyclohexane on the
proton n.m. r. spectra of the ethyl compounds of the metals are small.
The absence of large changes in the internal chemical shifts with
dilution indicates that there is essentially no intermolecular association
between the solute molecules. Furthermore, there does not appear to
be any appreciable intermolecular interactions between the solvent and
solute molecules. The proton magnetic resonance spectra of the ethyl
compounds have been analyzed and published previously (79, 85, 86).

In the analysis of the proton spectra for these compounds the assump-
tions are made that for a given compound the ethyl groups are equivalent
and that all the methyl and methylene hydrogens are equivalent,
respectively. The small differences between the internal chemicals
shifts in the pure solute and at infinite dilution indicates that these
assumptions are valid.

Corrections of the internal chemical shifts for differences in the
bulk diamagnetic susceptibility were not made due to the lack of
susceptibility data for the ethyl compounds. In addition, the chemical
shifts of the methyl and methylene protons were measured in the same

environment with respect to solvent.

The JAB’ JAX' and JBX

compounds do not show any significant changes with concentration.

spin-spin coupling constants for the ethyl

Table VII lists the changes in these spin- spin coupling constants as a

function of mole fraction of Pb(CHzCH3)4. Similar results are found for

122

123

the other ethyl compounds. Table VIII lists the values of the three
coupling constants for Sn(CZH5)4 determined in this work and also the
values determined previously by Narasimhan and Rogers (79).

A significant disagreement occurs in the values of the JB coupling

X

constants. It now appears that the previously reported value for JBX

is wrong and the correct value is approximately 55 cps.

Correlation of Chemical Shifts with
Molar Susceptibilities

I

 

Figures 36 through 40 show that the carbon-l3 chemical shifts
correlate quite well with molar susceptibilities in many cases. In all i
the figures with the exception of Figure 38 chemical shifts increases
as the molar susceptibilities increase. The correlation of carbon-13
chemical shifts with diamagnetic susceptibility data has been predicted
theoretically by Pople (17) for atoms with 2p electrons. As predicted
theoretically it is observed that the magnetic moment per molecule
increase as the shielding increases.

Thus, Figures 36, 37, 39, and 40 all have positive slopes.

However, in Figure 38 it can be seen that as the magnetic moment per
mole increases the shielding decreases. This reversal of the slope

for carbonyl carbons may be due to the extra paramagnetic contributions
to the atomic susceptibility due to the C = 0 double bond. Whenever the
applied field is oriented perpendicular to a double bond the theory
predicts larger X P . terms on the atoms participating in the double

bond.

Correlation of Chemical Shifts with
Electronegativity

The correlations of carbon-13 chemical shifts with electronega-

tivity for the compounds studied in this research are shown in

124

Figures 41 through 44. In general it can be seen that the carbon-l3
chemical shift of a carbon atom in a group decreases with increasing
group electronegativity. Thus the shielding of a given atom decreases
as the group electronegativity increases as would be expected if
inductive effects were dominant.

The case of the carbonyl carbons is different. InFigure 42,
the carbon-13 chemical shifts of the carbonyl carbon are seen to in-
crease with increasing group electronegativity of the substituted methyl
group in a series of acids. It is probable that inductive effects'are not
as important as magnetic anisotr0py effects here. Spiesecke and
Schneider (13), in a study of the carbon-l3 resonances in compounds of
the CH3CH2X type, concluded that the anisotropy contribution causes a
high-field shift of the a-carbon resonance and a low-field shift of the [3-
carbon resonance a similar effect may operate here.

In Figure 43 the correlation of carbon-l3 chemical shift with
group electronegativity is shown for some substituted toluenes, OCHZX.
It seems reasonable to expect that the correlation. with electronegativity
would be similar to that observed in CH3X compounds. It should be
noted that ¢CHzBr, CbCHzCl, and (LCHZCN deviate from the simple
electronegativity correlation line. This may be due to an appreciable
neighbor anisotropy effect in these cases. Similar effects for CH3C1,
CH3Br, and CH3I have been noted in correlation plots of carbon- 13
chemical shifts with electronegativity of the substituent atoms (13).

Figure 44 shows a plot of ortho carbon-l3 chemical shifts in
diphenyl and triphenyl compounds as a function of the electronegatiyity
of the substituent atoms. As might be expected the inductive effect
would be small at the ortho carbons. Hence it appears that anisotropy

contribution is the dominent factor.

125

The approximate agreement of the J coupling constants

c‘3—H

calculated using zeta values with the observed J constants given

c‘3-H
in Table XVII indicates that the contact term is the dominant factor

determining this coupling in the a- substituted toluenes.

SUMMARY

The natural abundance carbon-13 magnetic resonance spectra of
several aliphatic, phenyl, diphenyl, and triphenyl compounds have
been obtained at 15. 085 Mc. /sec. The spectra have been analyzed and
peak assignments made. The chemical shift data show a good corre-
lation with group and atomic electronegativities and molar susceptibilities.
The spin- spin coupling constants have been recorded in all cases.
In the d-substituted toluenes it was shown that the experimentally ob-
served carbon-l3 proton spin- spin coupling constants agree quite
well with the spin- spin coupling constants calculated using zeta values.
Solvent studies of some metal-organic ethyl compounds were per-
formed by observing the proton magnetic resonance spectra at 60. 000
Mc./sec. The solvent studies results indicated that in chloroform,
carbon tetrachloride, and cyclohexane the amount of intermolecular
association is negligible. The spin— spin coupling constants were ob-

served not to vary with concentration.

126

10. P.

11. P.

12. P.

13. H.

14. H.

15. R.

16. M.

17. J.

18. J.

BIBLIOGRAPHY

. D. Cohen, N. Sheppard, and J. J. Turner, Proc. Chem. Soc.,

1958, 118.

. Muller and D. E. Pritchard, J. Chem. Phys., 1, 768 (1959).

V. D. Tiers, J. Chem. Phys., 22, 2263 (1961).

. K. Harris, J. Mol. Spectroscopy, 12, 309 (1963).

C. Lauterbur, J. Chem. Phys., 26, 217 (1957).

. H. Holm, J. Chem. Phys., 2_6_, 707 (1957).

C. Lauterbur, J. Am. Chem. Soc., 83,, 1838 (1961).

C. Lauterbur, J. Am. Chem. Soc., 83, 1846 (1961).

C. Lauterbur, Ann. N. Y. Acad. Sci., 7_0, 841 (1958).

C. Lauterbur, J. Chem. Phys., _3_8_, 1406 (1963).

C. Lauterbur, J. Chem. Phys., 3_8_, 1415 (1963).

C. Lauterbur, J. Chem. Phys., _3_8, 1432 (1963).

Spiesecke and W. G. Schneider, J. Chem. Phys., _3_5, 722 (1961).
Spiesecke and W. G. Schneider, J. Chem. Phys.,_3_5, 731 (1961).

A. Friedel and H. L. Retcofsky, J. Am. Chem. Soc., _§_5_, 1300
(1963).

Karplus and J. A. Pople, J. Chem. Phys., 3_8, 2803 (1963).

A. Pople, Disc. Faraday Society, 34, 7 (1962).

A. Pople, W. G. Schneider, and H. J. Bernstein, "High Resolu—
tion Nuclear Magnetic Resonance, " Chapter 5, McGraw-Hill,

New York, 1959.

127

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

128

W. E. Lamb, Phys. Rev.,_§_0_. 817 (1941).

N. F. Ramsey, Phys. Rev.,_7_8_, 699 (1950).

A. Saika and C. P. Slichter, J. Chem. Phys., 22, 26 (1954).

J. A. Pople, Proc. Roy. Soc. (London), 5232, 550 (1957).

H. M. McConnell, J. Chem. Phys., 21, 226 (1957).

F. London, J. Phys. Radium, 8, 397 (1937).

J. A. Pople, J. Chem. Phys., .2_4, 1111 (1956). E
E

H. Eyring, J. Walter, and G. E. Kimball, "Quantum Chemistry, "
p. 99, John Wiley and Sons, New York, 1944.

A. Streitweiser, Jr. , "Molecular Orbital Theory for Organic
Chemists, " p. 39. John Wiley and Sons, New York, 1961.

J. A. Pople, J. Chem. Phys., 17, 53 (1962).

H. F. Hameka, M01. Phys., 1, 203 (1948).

H. F. Hameka, J. Chem. Phys., .33, 1996 (1961).

H. F. Hameka, J. Chem. Phys., _3_,'_7__, 3008 (1962).

J. A. POple, J. Chem. Phys., 31, 3009 (1962).

M. Karplus, and J. A. Pople, J. Chem. Phys., _3_8_, 2803 (1963).
D. F. O'Reilly, J. Chem. Phys., _36, 274 (1962).

C. W. Kern and W. N. Lipscomb, J. Chem. Phys., _3_Z_, 260 (1962).
C. W. Kern and W. N. Lipscomb, J. Chem. Phys., _3_1, 275 (1962).
H. F. Hameka, Rev. Mod. Phys., 3_4, 87 (1962).

S. K. Ghosh and S..K. Sinha, J. Chem. Phys., _3_6_, 737 (1962).

I. J. Maddox and R. McWeeny, J. Chem. Phys., 3_6_, 2353 (1962).

C. C. Hall and A. Hardisson, Proc. Roy.Soc. (London), 268,
328 (1962).

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

129
N. Jonathan, S. Gordon, and B. P. Dailey, J. Chem. Phys., 26.
2443 (1962).

H. S. Gutowsky, D. W. McCall, and C. P. Slichter, Phys. Rev.,
_8__2_, 748 (1951).

H. S. Gutowsky, D. W. McCall, and C. P. Slichter, Phys. Rev.,
§_4_, 589 (1951).

H. S. Gutowsky, D. W. McCall, and C. P. Slichter, J. Chem.
Phys., 21, 279 (1953).

E. L. Hahn and D. E. Maxwell, Phys. Rev., .8__4,, 1246 (1951).
E. L. Hahn and D. E. Maxwell, Phys. Rev., _8_8_, 1070 (1952).
P. L. Corio, Chem. Rev., 62, 363 (1960).

P. L. Corio, and I. Wienberg, J. Chem. Phys., _3_1, 569 (1959).
N. Muller, and D. E. Pritchard, J. Chem. Phys., _3_l_, 768 (1959).
Reference 18, Chap. 8.

M. Karplus, J. Am. Chem. Soc., 84, 2458 (1962).

P. C. Lauterbur and R. J. Kurland, J. Am. Chem. Soc., _8_4_,
3405 (1962).

F. A. L. Anet, J. Am. Chem. Soc., _8_4_, 3767 (1962).

J. D. Graham and M. T. Rogers, J. Am. Chem. Soc., _8_4, 2249
(1962).

H. S. Gutowsky and C. Juan, J. Chem. Phys., _3_7, 120 (1962).

K. L. Williamson and W. S. Johnson, J. Am. Chem. Soc., _8_3_,
4623 (1961).

F. A. L. Anet, J. Am. Chem. Soc., _8_‘_1_, 747 (1962).

M. Barfield and D. M. Grant, J. Am. Chem. Soc., 83, 4726
(1961).

 

59.
60.

61.

62.

63.

64.

65.

66.

67.
68.

69.

70.

71.
72.
73.

74.

75.

76.

H.

J.

W.

130

J. Bernstein and N. Sheppard, J. Chem. Phys., _3_7_, 3012 (1962).

. Meiboom, Ann. Rev. Phys. Chem., E, 335 (1963).

M. Jackman, "Applications of Nuclear Magnetic Resonance
Spectroscopy in Organic Chemistry, " Chapter 2, Pergamon Press,
New York, 1959.

. D. Roberts, "Nuclear Magnetic Resonance, " p. 12, McGraw-Hill,

1959.

Bloch, Phys. Rev., :72, 460 (1946).

. Bloembergen, E. M. Purcell, and R. V. Pound, Phys. Rev.,

_7_3_, 679 (1948).

J. Karabatsos,J. Am. Chem. Soc., fl, 1230 (1961).

. Muller, and D. E. Pritchard, J. Chem. Phys., _3_1_, 1471

(1959).
K. Harris, J. M01. Spectroscopy, _12, 309 (1963).
Muller, and D. T. Carr, J. Phys. Chem., 91, 112 (1963).

M. McConnell, and D. D. Thompson, J. Chem. Phys., _2_6_,
958 (1957).

M. McConnell, and D. D. Thompson, J. Chem. Phys., _3_1.
85 (1959).

. C. Lauterbur, J. Am. Chem. Soc., _83, 1838 (1961).

Bloch, Phys. Rev., 102, 104 (1956).
T. Arnold, Phys. Rev., 102, 136 (1956).

Freeman and D. H. Whiffen, Proc. Phys. Soc. (London), 22,
794 (1962).

M. Anderson and J. D. Baldeschwieler, J. Chem. Phys., _3_Z
39 (1962).

A. Anderson and R. Freeman, J. Chem. Phys., _3_1, 85 (1962).

77.

78.

.79.

80.

81.

82.

83.

84.

85.

86.

87.
88.
89.
90.
91.

92.

93.

94.

131

R. Freeman and W. A. Anderson, J. Chem. Phys., _3_7_, 2053 (1962).
P. C. Lauterbur, in "Determination of Organic Structures by
Physical Methods, " F. C. Nachod and W. D. Phillips, Ed. ,
p. 485, Academic Press, New York, N. Y., 1962.

P. T. Narasimhan and M. T. Rogers, J. Chem. Phys., _3_4, 1049
(1961).

J. B. Stothers, Private Communication.

G. W. Smith, General Motors Research Laboratories, Report.
GMR-317, December 20, 1960.

J. Hinze, M. A. Whitehead, and H. H. Jaffe, J. Am. Chem. Soc.,
£5, 148 (1963).

J. K. Wilmshurst, J. Chem. Phys., _2_Z, 1129 (1957).

L. Pauling, "The Nature of the Chemical Bond, 3d Ed. , p. 93,
Cornell University Press, Ithaca, New York, 1960.

P. T. Narasimhan and M. T. Rogers, J. Am. Chem. Soc.,_§_2_,
34 (1960).

P. T. Narasimhan and M. T. Rogers, J. Am. Chem. Soc., 8_2_,
5983 (1960).

. B. Savitsky and K. Namikawa, J. Phys. Chem., fl, 2430 (1963).
. B. Savitsky, J. Phys. Chem., 61, 2723 (1963).
. B. Savitsky and K. Namikawa, J. Phys. Chem., _6_7_, 2754 (1963).

G
G
G
G. Klose, Ann. Physik, _8_, 220 (1961).
G. V. D. Tiers, J. Phys. Soc. Japan, _1_5_, 354 (1960).
N

. Sheppard and J. J. Turner, Proc. Roy. Soc. (London), A252,
506 (1959).

E. R. Malenowski, J. Am. Chem. Soc., 83, 4479 (1961).

C. Juan and H. S. Gutowsky, J. Chem. Phys., 31, 2198 (1962).

mmmTPY UBP‘M‘N