a I}...
.. 2. 3.6,:frl

 

spa (
:11 r...

f... y .51
oer...“ Ma“

.,y..
42313:...er
6. lo I:

0’ [at 44 !

 

v 1‘. ‘ , .
1... v7 2:3? .r. ,. :r..1.n...$...v.?finn.r¢r.?. 51.1119.

 

IIIIII3IIIIIII1III2IIIIIIIIIIIIIOIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIHI ’74 b l U :3 “1 / >4

W 00579 8065

 

LIBRARY
Michigan State
University

 

 

 

This is to certify that the

dissertation entitled

Spectroscopic Studies of Ionic Solvation
and Complexation in N-Methylformamide

presented by

Lee—Lin Soong

has been accepted towards fulfillment
of the requirements for

Ph.D. degreein Chemistry

flaw/WW
// 3%, W

 

MS U i: an Affirmative Action/Equal Opportunity Institution 0-12771

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative Action/Equal Opportunity Institution

 

_————_~_ _ . 7 e ,7,_ _

 

 

 

 

 

 

SPECTROSCOPIC STUDIES OF IONIC SOLVATION

AND COMPLEXATION IN N-METHYLFORMAMIDE

BY

Lee-Lin Soong

A DISSERTATION

Submitted to
Michigan State University
in partial fullfillment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Dpeartment of Chemistry

1988

 

 

 

ABSTRACT

. SPECTROSCOPIC STUDIES OF IONIC SOLVATION

AND COMPLEXATION'IN N-METHYLFORMAMIDE

BY

Lee-Lin Soong

Lithium-7 chemical shifts in NMF solution are virtually
concentration and counter-ion independent, presumably due to
a low probability of direct ionic encounter --- an
impenetrable solvation shell. Free cations and solvent-
separated ion pairs are the likely components for lithium
salts in NMF. For the sodium, potassium and cesium salts
studied, the chemical shifts are dependent on the anion and
change linearly with concentration. The order of increasing
shielding shift for cesium salts is:

2-.

I” < Br' < c1' < co < son"

3

' ._ < F' < N03- < c104' < TPB-.-

The observed behavior can be accounted for by the formation
of collisional ion pairs. Thallium-ZOS chemical shifts
change nonlinearly with concentrations of thallium nitrate
and perchlorate. This is believed to reflect ion-ion
interactions. Ion pair formation constants for these two

1

- salts are 2.6 i 0.4 M- for 111%., and 1.7 1: 0.5 14-1 for

 

I \—
_II
—II. I—ulll

Lee-Lin Soong

T1C10 4.
Solutions of LiSCN, NaSCN, CsSCN and TIAASCN
(tetraisoamylammonium thiocyanate) in NMF were studied in

the mid-ir region. The peak position of the (EN stretching

vibration for all of the salts studied was 2057 j; l cut-1:

ie. the peak frequency of the v1 mode of SCN- in NMF is

independent of the nature of the cation and the solute
concentration. This suggests the absence of contact ion

pairs in these solutions. Curve fitting analysis of these

broad bands reveals three bands, at 2045.0 cm-1 1

and 2065.5 cm-l. Since the positions of these three

, 2054.5 cm"

components are cation independent, the multiplet structure
must result from interactions between the thiocyanate anions
and solvent molecules. Definitive assignments cannot be
provided.

Studies of complexation reactions between cesium ions
and crown ethers and cryptands were also carried out. The
stabilities of Cs+ complexes with crown ethers and cryptands
decrease by 18C6 > DBZ7C9 z DBZ4C8 3 C221 2: C222 > DC18C6 >
DA18C6£ DBlSCS, DBZlC7 and c211 do not form complexes with
the cesium ion. The formation constants are usually small
because of high donicity of NMF. Solvent-ligand interaction
is a very important factor in determining the stability of
the complex. D8210? must have just the right environment
for NMF molecules, so there is almost no complex formed with

the cesium ion.

ACKNOWLEDGMENTS

The author would like to express his sincere gratitude
to Dr. Alexander I. Popov and Dr. George E. Leroi for their
guidance and encouragement during the course of this study.

He also wishes to thank the research group of Dr. Popov
and H.-H. Nam for their friendship, encouragement and
discussions.

Great thanks also go to the NMR group of the department
for their maintenance of the NMR spectrometers. .

Financial aids from his advisers, the National Science
Foundation and the Department of Chemistry, Michigan State
University are also acknowledged

The author also wishes to thank the strong support of
his family as well as his wife’s family.

Sophie’s and Steven’s loves and smiles, which work just
like the catalyst to the reaction, have cheered him a lot.
Finally, he ‘wishes to thank. his wife for her patience,

encouragement and good careness of children.

iv

Chapter Page
LIST OF TABLES O O O O 0 O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O x
LIST OF FIGURES O O O O O O O O ........ 0 ...... O O O I O O O O O O O I O O O 0 ”iii
LIST OF ABBREVIATIONS O O ..... O ..... O O O ...... C O O O I O O O O O O XXiii

CHAPTER I - INTRODUCTION................................. 1
A. Introduction.................................... 2
B. Historical Part................................. 3

l. Interactions in Electrolyte Solutions......... 3
1.1 Introduction.............................. 3
1.2 Electrical Conductance.................... 5
1.3' Nuclear Magnetic Resonance (NMR).......... 8
1.4 Infrared and Raman Spectroscopy........... 14
1.5 ESR and Electronic Spectroscopy........... 16

2. Complexation of Metal Ions by

- -~ Crown Ethers and Cryptands.................... 18
2.1 Introduction.............................. 18
2.2 Selectivity of Complexation............... 21
2.3 Thermodynamics of Complexation............ 27

3. N-methylformamide............................. 29
3.1 Properties and Structures................. 29
3.2 Solvation................................. 32

V

 

 

 

Chapter ‘ Page

3.3 Ion-Ion interactions...................... 38
CHAPTER II - EXPERIMENTAL PART........................... 44
A. Materials.... ...... ............................. 45
l. Salts......................................... 45
2. Solvents...................................... 47
3. Ligands.. ........... .......................... 47
B. Methods......................................... 48
1. Sample Preparation............................ 48
2. NMR Measurements ..... ......................... 48
3. IR Measurements ...... ... ..... ................. 49
C. Data Treatment .............. .................... 51
1. NMR Measurements ............ .................. 51
2. IR Measurements.... ..... ...................... 52
CHAPTER III - IONIC SOLVATION STUDIES IN
N-METHYLFORMAMIDE SOLUTIONS................ 54
A. Introduction.................................... 55
B. Ion-Ion Interactions in

N-Methylformamide Solutions..................... 56
1; Results..... ...... ............................ 56
2. Discussion.................................... 78
2.1 Infinite Dilution Chemical Shifts.......... 78

2.2 Concentration and Anion
Dependence of Chemical Shifts.............. 82
2.2-1 Lithium Salts........................... 82
2.2-2 Sodium, Potassium and Cesium Salts...... 87

vi

Chapter Page

2.2-3 Thallium Salts.......................... 94
2.2-4 Temperature Study of
the CsI/NMF System...................... 95
2.2-5 Chemical Shift as a Function of Mole
Fraction of Cesium Salts in NMF......... 95
2.3 Chemical Shift Theory...................... 99
2.3-1 Overlap Theory..........................102
2.3-2 Probability Theory......................105
2.3-3 H. R. S. Theory.........................109
2.3-4 Relaxation Theory.......................110
'3. Conclusion....................................112
C. Ion-Ion Interactions in N-Methylformamide
Mixed Electrolyte Solutions.....................115
1. Results.......................................115
2. Discussion....................................115
D. Ion-Solvent Interactions in
N-Methylformamide Solutions.....................131
1. Results and Discussion........................l3l
- --1.1 Infrared Measurements......................131
1.1-1 Thiocyanate Salts Studies...............131
1.1-2 Computer Resolution of
Thiocyanate Bands.......................136
1.1-3 1% (Volume) N-Methylformamide/
Acetonitrile Mixtures...................147
1.2 NMR Measurements.......................:...152

vii

Chapter Page

1.2-l H—l NMR Measurements....................152
1.2-2 N-14, N-lS and Cl-35 NMR Measurements...153
2. Conclusion....................................156
E. Suggestions for Further Studies.................156
CHAPTER IV - COMPLEXATION STUDIES IN N-METHYLFORMAMIDE...158
A. Introduction....................................159
B. Results and Discussion..........................160
1. Complexation of Cesium Ions by 1806
and Its Substituted Analogs ..................l60
1.1 18C6.......................................160
1.2 D318C6.....................................169
1.3 DC18C6.....................................l71
1.4 DAl8C6.....................................l7l
2. Complexation of Cesium Ions by
Larger Crown Ethers...........................l76
3. Complexation of Cesium Ions by Cryptands......183
C. Conclusion......................................190
D. Suggestions for Further Studies.................193
APPENDICES
APPENDIX I - DETERMINATION OF ION PAIR FORMATION
CONSTANTS BY THE NMR TECHNIQUE:
DESCRIPTION OF THE COMPUTER
PROGRAM KINFIT AND SUBROUTINE EQUATION...194
APPENDIX II- DETERMINATION OF COMPLEX FORMATION
CONSTANTS BY THE NMR TECHNIQUE:
viii

Page

DESCRIPTION OF THE COMPUTER PROGRAM
KINFIT AND SUBROUTINE EQUATIONS..........201
A. Determination of Formation Constants
for l:l Complexes...............................201
B. Determination of Formation Constants
for 1:1 and 2:1 Complexes.......................207

REFERENCESOOOOOOOO ..... O.........00.0.00....0000000000000215

ix

LIST OF TABLES

Table Page
1. Nuclear Properties of Some Elements.............. 9
2. Diameter in K of Some Univalent Cations

(in Crystals) and Macrocyclic Polyethers.. ....... 23
3. Physicochemical Properties of Formamide

(FA), N-Methylformamide (NMF) and

N,N-Dimethylformamide (DMF)...................... 30
4. Wavenumber Shifts (cm-1) of N-H and C80

Stretching Bands in Salt/NMF Solutions

(4 M) 34
5. Solubility of Salts in NMF at Different

Temperatures (g/100 g Solvent)................... 35
6. Lattice Energies, Heats of Solution and Beats of

Solvation of Some Alkali Metal Salts in NMF

-(kcal/mole) at 26.9°C............................ 36
7. Radii (r) for Solvated Ions and Solvation Numbers

(n) for Various Ions in NMF at 25°C.............. 39
8. Ion Pair Formation Constants of Electrolytes

in NMF at 25°C................................... 40
9. References for Multinuclear NMR Measurements and

Their Chemical Shifts (6) Relative to Those of

X

Table. - Page

Infinite Dilute Aqueous Solutions of Metal Ions.. 50
10. Lithium-7 Chemical Shifts of

LiCl and LiBr in NMF....... ...... ................ 57
11. Lithium-7 Chemical Shifts of

LiI and LiSCN in NMF .............. . ....... . . . . . . . 57
12. Lithium-7 Chemical Shifts of

LiTPB and LiNO3 in NMF.... ............. .......... 58
13. Lithium-7 Chemical Shifts of

LiClO4 and LiAsF6 in NMF......................... 58
14. Lithium-7 Chemical Shifts of LiPi in NMF......... 59
15. Sodium-23 Chemical Shifts of NaCl in NMF......... 59
16. Sodium-23 Chemical Shifts of NaBr in NMF......... 60
17. Sodium-23 Chemical Shifts of NaI in NMF.......... 60
18. Sodium-23 Chemical Shifts of NaTPB in NMF........ 61
19. Sodium-23 Chemical Shifts of NaAsF6 in NMF....... 61
20. Sodium-23 Chemical Shifts of NaPi in NMF......... 62
21. Sodium-23 Chemical Shifts of NaSCN in NMF........ 62
22. Sodium-23 Chemical Shifts of NaClO4 in NMF....... 63
23. -Sodium-23 Chemical Shifts of NaNo3 in NMF........ 63
24. Potassium-39 Chemical Shifts of RI in NMF........ 64
25. Potassium-39 Chemical Shifts of KSCN in NMF...... 64
26. Potassium-39 Chemical Shifts of KNO3 in NMF...... 65
27. Cesium-133 Chemical Shifts of CsF in NMF......... 65
28. Cesium-133 Chemical Shifts of CsCl in NMF........ 66
29. Cesium-133 Chemical Shifts of CsBr in NMF........ 66

xi

 

Table_
30.
31.
32.
33.
34.
35.
36.

37.

38.

39.

40.

41.

42.

43.

Cesium-133 Chemical
Cesium-133 Chemical
Cesium-133 Chemical
Cesium-133 Chemical
Cesium-133 Chemical

Cesium-133 Chemical

Shifts
Shifts
Shifts
Shifts
Shifts

Shifts

of
of
of
of
of
of

Page

CSI in NMF......... 67
CSSCN in NMF....... 67
CsClO4 in NMF...... 68
CSNO3 in NMF....... 68
CSTPB in NMF....... 69

Cs2C03 in NMF...... 69

Thallium(I)-205 Chemical Shifts

Of T1C1O4 in NMFOOOOOOOOO......OOOOOOOOOOOOOOOOOO 7o

Thallium(I)-205 Chemical Shifts

Of TlNo in NMFOOOOO..0.0000000000000000COOOOOOOO 7o

3

Parameters of Linear Regression for Pure Sodium

and Potassium Salts

in NMF

(Concentration vs.

Sodium-23 and Potassium-39 Chemical Shifts)... 71

Parameters of Linear Regression for Pure Cesium

Salts in NMF (Concentration vs. Cesium-133

Chemical ShiftS)0.0.0000000000000000000000000I... 72

Cesium-133 Chemical Shifts of CsI in NMF

at 273K, 296K and 341KOOOOOOOOOOOO......OOOOOOOOO 79

Parameters of Linear Regression for C31 in

NMF at Three Different Temperatures.............. 79

Infinite Dilution Chemical Shifts (ppm) Relative

to Water for the Cations

133 + 205 +

7

Li+, 23 + 39 +

Na , K ,

Cs and T1 in a Variety of Solvents....... 81

Parameters of Linear Regression for Cesium Iodide

in NMF with Different Values of the Ion Pair

xii

 

Table_

44.

45.

46.

47.

48.

49.

50.

51.

52.

Formation Constant (K) and Cs-133 Chemical Shift
Difference between Free Cesium Ions and Ion
Pairs (55).......................................
Parameters of Linear Regression for Pure Cesium
Salts in NMF (Mole Fraction vs. Cesium-133
Chemical Shifts).................................
Comparisons of Mole Fraction Shifts (50) with
Crystalline Chemical Shifts (61) for Cesium
Halides and Proportionality Constants (m) in
Equation 3.5.....................................
Concentration Shifts of Cesium Salts

in Water and NMF.................................
Concentration Shifts of CsTPB in DMF, DMSO,

EC and NMF.......................................

Concentration Shifts of C51 in MeCN, H 0,

2
EC and NMF.......................................
Linear Regression Parameters for Cesium Salts in
Water and NMF Using H. R. S. Theory's Empirical
.Formula..........................................
Summary of Ion-Association Studies in NMF, Via

Different Techniques.............................
Cesium-133 Chemical Shifts of NMF Solutions of

Mixed CsCl/O.10 M CsI and LiC1/0.10 M CsI........
Cesium-133 Chemical Shifts of NMF Solutions of

Mixed Neel/0.10 M CsI and NaTPB/0.10 M CsI.......

xiii

Page

92

98

100

107

108

108

111

113

116

116

Table. Page

53. Cesium-133 Chemical Shifts of NMF Solutions of

Mixed LiTPB/0.10 M CsI and CsTPB/0.1O M CsI..... 117
54. Cesium-133 Chemical Shifts of NMF Solutions of

Mixed LiC104/0.10 M C51 and NaClO4/0.10 M CsI.... 117
55. Cesium-133 Chemical Shifts of NMF Solutions of

Mixed esc1o4/o.1o M CsI and esno3/o.1o M CsI..... 113
56. Cesium-133 Chemical Shifts of NMF Solutions of

Mixed LiNO3/0.10 M CsI and NaN03/0.10 M CsI...... 118
57. Cesium-133 Chemical Shifts of NMF Solutions of

Mixed LiBr/0.10 M C31 and NaBr/0.10 M CsI........ 119
58. Cesium-133 Chemical Shifts of NMF Solutions of

Mixed CsBr/0.10 M CsI and LiSCN/0.10 M CsI....... 119
59. Cesium-133 Chemical Shifts of NMF Solutions of

Mixed NaSCN/0.10 M CsI and CsSCN/0.10 M CsI...... 120
60. Cesium-133 Chemical Shifts of NMF Solutions of

Mixed NaI/0.10 M CsI and KI/0.10 M CsI........... 120
61. Cesium-133 Chemical Shifts of NMF Solutions of

Mixed TPAI/0.10 M CsI............................ 121
62. .Cesium-133 Chemical Shifts of NMF Solutions of

Mixed CsSCN/0.2434 M CsI and CsSCN/0.4816 M

CsI.............................................. 121
63. Parameters of Linear Regression for Mixed Salt

Systems in NMF................................... 127
64. Molar Molecular Shift (ppm/M) for Mixed

Electrolyte Systems in NMF....................... 129

xiv

Table. Page

65. Peak Positions ‘V(cm-1) and Half-height Width

(HI/2) (cm-1) of C:N Stretching Bands of Four

Thiocyanate Salts in NMF......................... 132
66. Curve Fitting Results of Thiocyanate

Salts in NMF ....... .............................. 138
67. Curve Fitting Results of Thiocyanate Salts in

NMF at Different Temperatures.................... 139
68. Wavenumbers (cm-1) of N-H, C=0 and C-N Stretching

Bands in 1% (volume) NMF/MeCN Mixtures........... 148
69. Wavenumber Shifts (cm-1) of N-H, C=0 and C-N

Stretching Bands of NMF in 1% (volume) NMF/MeCN

Mixtures......................................... 150
70. Proton Chemical Shifts of NMF

and Its Solutions................................ 152
71. N-14 Chemical Shifts of Thiocyanate and Nitrate

Salts in NMF..................................... 154
72. N-15 Chemical Shifts of LiSCN-15 in NMF.......... 155
73. Cl-35 Chemical Shifts of

Perchlorate Salts in NMF......................... 155
74. Variation of Cesium-133 Chemical Shifts with the

Mole Ratio (18C6)/(CsI) in NMF................... 161
75. Variation of Cesium-133 Chemical Shifts with the

Mole Ratio (18C6)/(CsI) in NMF................... 162
76. Variation of Cesium-133 Chemical Shifts with the

Mole Ratio (18C6)/(CsTPB) in NMF................. 163

XV

Table_ ~ Page

77. Logarithms of Formation Constants of 1:1 and
2:1 (18C6/Cs+) Complexes of Cesium Salts with
18C6 in Various Solvents......................... 165
78. Limiting Chemical Shifts of Cs+, Cs+(18C6) and
Cs+(18C6)2 in Various Solvents................... 166
79. Variation of Cesium-133 Chemical Shifts with the
Mole Ratio (DB18C6)/(CsI) in NMF................. 170
80. Variation of Cesium-133 Chemical Shifts with the
(DC18C6)/(CSI) Mole Ratio in NMF................. 172
81. Variation of Cesium-133 Chemical Shifts with the
(DA18C6)/(CsI) Mole Ratio in NMF................. 173
82. Logarithms of the Formation Constants of Cesium
Complexes with Crown Ethers in Various Solvents.. 175
83. Variation of Cesium-133 Chemical Shifts with the
Mole Ratio (D821C7)/(CsI) in NMF................. 177
84. Variation of Cesium-133 Chemical Shifts with the
Mole Ratio (DBZ4C8)/(CsI) in NMF................. 178
85. Variation of Cesium-133 Chemical Shifts with the
‘Moie Ratio (DBZ?C9)/(CsI) in NMF................. 178
86. Logarithms of the Formation Constants of Cesium
Complexes with Crown Ethers in Various Solvents.. 181
87. Limiting Chemical Shifts of 1:1 Complexes of
Cesium ions with DBZlC7, DB24C8 and DBZ7C9 in
Various Solvents ........ ......................... 183
88. Variation of Cesium-133 Chemical Shifts with the

xvi

 

Table. Page

Mole Ratio (C222)/(CsI) in NMF................... 184
89. Variation of Cesium-133 Chemical Shifts with the
Mole Ratio (C222)/(CsTPB) in NMF................. 185
90. Variation of Cesium-133 Chemical Shifts with the
Mole Ratio (C221)/(CsI) in NMF................... 186
91. Variation of Cesium-133 Chemical Shifts with the
Mole Ratio (C211)/(CsI) in NMF................... 186
92. Logarithms of the Formation Constants of Cesium
Complexes with Cryptands in Various Solvents.. 189

3

93. C-4 and C-5 Separation in 1 C Spectra

of Cryptand C221 in Various Solvents............. 192

xvii

 

LIST OF E G RES

Figure Page
1. Infinite Dilution of Sodium-23 Chemical
Shifts vs. Gutmann Donor Numbers................ 11
2. Structures of Some Macrocyclic Polyethers........ 19
3. Various Stoichiometries of K+-Crown
Ether Complexes.................................. 20
4. Selectivity of 18-crown-6: log K (Formation
Constant) Values for the Reaction of 18-crown-6
with Metal Ions in Water vs. Ratio of Cation
Diameter to 18-crown-6 Cavity Diameter......... 24
5. Selectivity of Cryptands: log K (Formation
Constant) Values for the Reaction of Several
Cryptands with Alkali Metal Ions vs. Cation
Radius........................................... 25
6. (a) Cis- and Trans-Forms of NMF
(b) Linear Chain Structure of Liquid NMF......... 31
7. (a) A Proposed Structure of the Short-Lived
Solvent-Separated Ion Pair for NaBr in NMF
(b) Proposed Structures of Some Higher Ion
Associates for RbI in NMF.................... 42
8. Plot of Metal (Li-7 and K-39) Chemical Shifts
vs. Concentration of Metal Salts in NMF.......... 73
xviii

 

Figure Page

9. Plot of Sodium-23 Chemical Shifts vs.

Concentration of Sodium Salts in NMF............. 74
10. Plot of Cesium-133 Chemical Shifts vs.

Concentration of Cesium Salts in NMF............. 75
11. Plot of Cesium-133 Chemical Shifts vs.

Concentration of Cesium Salts in NMF............. 76
12. Plot of Thallium-205 Chemical Shifts vs.

Concentration of Thallium Salts in NMF........... 77
13. Plot of Cesium-133 Chemical Shifts vs.

Concentration of C51 in NMF at Three

Temperatures.... ........ ......................... 80
14. Sodium-23 Chemical Shifts vs. Gutmann Donor

Numbers. Reference - Aq. Na+ at Infinite

Dilution... .................. .................... 83
15. Potassium-39 Chemical Shifts vs. Gutmann Donor

Numbers. Reference - Aq. K+ at Infinite

Dilution........:................................ 84
16. Cesium-133 Chemical Shifts vs. Gutmann Donor

.Numbers. Reference - Ag. Cs+ at Infinite

Dilution......................................... 85
17. Thallium-205 Chemical Shifts vs. Gutmann Donor

Numbers. Reference - Aq. Tl+ at Infinite

Dilution............... ..... ..................... 86
18. Plot of Cesium-133 Chemical Shifts vs.

Concentration of C51 in NMF at Four Different K

xix

Figure

19.

20.

21.

22.

23.

24.

25.

26.

27.

Page

(Ion Pair Formation Constant) Values............. 90
Plot of Cesium-133 Chemical Shifts vs.

Concentration of C51 in NMF at Three

Different D (Chemical Shift Difference

between Free Cesium Ions and Ion Pairs)
Values........................................... 91
Plot of Cesium-133 Chemical Shifts vs. Mole

Fraction of Cesium Salts in NMF.................. 96
Plot of Cesium-133 Chemical Shifts vs. Mole

Fraction of Cesium Salts in NMF..................' 97
Plot of 6 / 51 vs. Mole Fraction

of Cesium Salts in NMF........................... 101
Plot of Cesium-133 Chemical Shifts vs.
Concentrations of Lithium Salts Added to

0.10 M CsI Solutions in NMF...................... 122
Plot of Cesium-133 Chemical Shifts vs.
Concentrations of Sodium Salts Added to

0.10 M CsI Solutions in NMF...................... 123
.Plot of Cesium-133 Chemical Shifts vs.
Concentrations of Cesium Salts Added to

0.10 M CsI Solutions in NMF...................... 124
Plot of Cesium-133 Chemical Shifts vs.
Concentrations of Iodide Salts Added to

0.10 M CsI Solutions in NMF...................... 125
Plot of Cesium-133 Chemical Shifts vs.

XX

Figure - Page

Concentrations of CsSCN Added to 0.10 M,

0.24 M and 0.48 M CsI Solutions in NMF.......... 126
28. Deconvolution of the GEN Stretching Band of

Thiocyanate Salts in NMF......................... 134
29. Deconvolution of GEN Stretching Bands of

Lithium Thiocyanate (0.24 M) in NMF at Three

Temperatures..................................... 135
30. Resolved Curves of the GEN Stretching Band

for Thiocyanate Salts in NMF..................... 140
31. Plot of Relative Intensities of Three GEN

Resolved Bands vs. Concentrations of LiSCN

in NMF........................................... 141
32. Plot of Relative Intensities of Three GEN

Resolved Bands vs. Concentrations of NaSCN

in NMF........................................... 142
33. Plot of Relative Intensities of Three GEN

Resolved Bands vs. Concentrations of CsSCN

in NMF........................................... 143
34. 'Plot of Relative Intensities of Three C:N

Resolved Bands vs. Temperatures in 0.24 M

LiSCN/NMF Solution............................... 144
35. Plot of Relative Intensities of Three GEN

Resolved Bands vs. Temperatures in 0.49 M

NaSCN/NMF Solution............................... 145
36. Plot of Relative Intensities of Three GEN

xxi

Figure Page

Resolved Bands vs. Temperatures in 0.40 M

CsSCN/NMF Solution...L........................... 146
37. Plot of Cesium-133 Chemical Shifts vs.

(18C6)/(Cs+) Mole Ratio in NMF................... 154
38. Plot of Cesium-133 Chemical Shifts vs.

(Ligand)/(Cs+) Mole Ratio in NMF................. 174
39. Plot of Cesium-133 Chemical Shifts vs.

(Ligand)/(Cs+) Mole Ratio in NMF................. 179
40. Plot of Cesium-133 Chemical Shifts vs.

(Ligand)/(Cs+) Mole Ratio in NMF................. 187

41. Carbon-13 NMR Spectrum of C221 in NMF............ 191

xxii

 

1- ELLIS

LiTPB
NaTPB
CsTPB
LiPi
NaPi
TPAI

TIAASCN

2- EQLEEEIE

3

-EC

DMF

FA

THF

DMSO

HMPA

LIS F BBR V ATION

Lithium tetraphenylborate
Sodium tetraphenylborate
Cesium tetraphenylborate
Lithium picrate

Sodium picrate
Tetrapentylammonium iodide

Tetraisoamylammonium thiocyanate

N-Methylformamide
Acetonitrile
Acetone
Nitromethane
Pyridine

Propylene carbonate
Ethylene carbonate
N,N-Dimethylformamide
Formamide
Tetrahydrofuran
Dimethylsulfoxide

Hexamethylphosphoramide

xxiii

 

3.

LIQANDE

12C4

15C5

18C6

DA18C6

DBlSCG

DC18C6

DBZlC7

DB24C8

DBZ7C9

DB30C10

C222

C221

C211

12-Crown-4
15-Crown-5
lB-Crown-G
Diaza-lB-Crown-6
Dibenzo-lB-Crown-G
Dicyclohexyl-lB-Crown-6
Dibenzo-Zl-Crown-7
Dibenzo-24-Crown—8
Dibenzo-27-Crown-9
Dibenzo-30-Crown-10
Cryptand 222
Cryptand 221

Cryptand 211

xxiv

 

 

A.I RO CTO

Fundamental studies of electrolyte solutions are
centered on ion-solvent and ion-ion interactions. A variety
of physicochemical techniques have been applied by numerous
investigators to study the behavior of electrolytes,
primarily in solvents with intermediate and low dielectric
constants. Studies of interionic interactions in solvents
of high dielectric constant have been sparse since it is
generally assumed that in these media electrolytes are
completely dissociated.

N-Methylformamide (NMF) is a valuable, but little
studied solvent which has a high dielectric constant (182.4
at 25°C, 308 at -40°C) and strong solvating ability
(Gutmann donor number : 25) . The combination of these two
properties would suggest an essentially complete absence of
interionic interactions. However, some studies have
indicated that there is ion association for certain salts
in NMF. These reports led us to a detailed study of the
extent and nature of ion-ion and ion-solvent interactions
in this solvent, through the use of several spectroscopic
techniques. The results of this investigation are presented
in Chapter III.

Complexation between crown ethers or cryptands and
metal ions has not been studied in NMF, so a systematic
study of the interaction between monovalent cations and
representative macrocyclic complexants in this solvent was

also carried out. It will be presented in Chapter IV.

B- W

1. WW5

1-1 M92129

There are many different kinds of interactions in
electrolyte solutions such as: ion-ion, ion-dipole, ion-
quadrupole, ion-induced dipole, dipole-dipole, dipole-
induced dipole and dispersion (London) interactions. The
sum total of the energies of these electrostatic
interactions can be represented as a power series in the
reciprocal of the interaction distance r. In this series,

the terms range from r-1 for the ion-ion energies

(strongest) to r-6

for the dispersion energies (weakest).
In addition, there may be specific chemical interactions in
electrolyte solutions such as hydrogen bonding, acid-base
interactions and charge transfer interactions.

Ion-ion interaction is one of the vexing problems in
the study of electrolyte solutions. There is no clear
fundamental definition of the term "ion pair", and it can
be expressed by many adjectives such as "tight", "loose",
"contact", "solvent—shared" and "solvent—separated" ion
pair. -On a molecular level, distinctions of these
expressions might be made: a free ion is surrounded by
several shells of solvent molecules: a contact ion pair,
well separated from other ions, is composed of two ions

held together by electrostatic forces for at least one

vibrational period: one solvent molecule is interposed

' between the two oppositely charged ions of a solvent-shared

 

4

ion pair: each of the two interacting ions retains its
inner solvation shell in a solvent-separated ion pair.

Electrical conductance measurements, which depend only
on free ions and on clusters containing unequal numbers of
positive and negative charges, have been used to determine
ion pair formation constants since the early 1900s. The
experimental data have been treated by several different
models. More recently, spectrometric techniques, especially
infrared, Raman and NMR measurements, which probe solutions
on a molecular basis, have complemented the macroscopic
conductance experiments. NMR is sensitive to the local
environment of a magnetic nucleus, and the chemical shift
of a cation in a contact ion pair will be different from
its value in the free ion or in a solvent-mediated ion pair
(the latter two are indistinguishable) . In NMR
measurements, a single, population-averaged chemical shift
is usually observed, which results from fast exchange
between the free cation site and the two types of ion
pairs. Therefore, it is impossible to determine the
relative concentration of each component by NMR methods. It
is not possible to compare the ion pair formation constants
obtained by NMR techniques with those from conductance
measurements , because NMR technique monitors the
equilibrium between free ions (or solvent-separated ion
pairs) and contact ion pairs, whereas conductance monitors

the equilibrium between free ions and all kinds of ion

pairs. Vibrational spectroscopy is similarly sensitive to

 

5
the local environment of an ion: thus a band due to a
contact ion pair will be distinguishable from that arising
from a free ion or a solvent-mediated ion pair. A detailed
review of these techniques will be given in the following

sections.

1.2 W
Ostwald (1) was the first to derive an equation
relating the ion pair formation constant (K) for 1:1 weak
electrolytes to the measured equivalent conductance at a
given concentration C andAo the equivalent conductance at

infinite dilution:

K=A2c /I\o(l\o-l\)- (1.1)

However, this equation was found to be invalid for
solutions of strong electrolytes. Later, an empirical
equation for strong electrolytes was found (2):

A=Ao - SCI/2. (1.2)

This equation is explained well by the Debye-Hfikel-Onsager
theory, which shows that the equivalent conductance at low
concentration should be a linear function of the square
root of concentration.

Onsager (3,4) made an important step in the

electrostatic theory by attributing the changes in the

equivalent conductance with concentration to two long-range

 

 

6
effects of ion interactions, known as the relaxation and
electrophoretic effects. In the case of 1:1 electrolytes

the Onsager limiting law is given by

A =Ao - (o*l\° +0*)c1/2. (1.3)

8.2 x 105 / (135)3/2

*
where a

82 / (0T) “2 n

13
ll

(D:die1ectric constant, T:temperature(°l<) and n:viscosity)
The relaxation term (-0*C1/2) results from the perturbation
of the applied field by the asymmetry of the ionic

1/2) takes into

atmosphere. The electrophoretic term (-B*C
account the decrease in the velocity of an ion due to the
counter flow of the solvent in the ionic atmosphere. By an
extension of the Onsager limiting law, Fuoss and Kraus (5)
introduced the ion association constant into the

conductance equation:

>
II

a mo - Steal/21 (1.4-1)

7%
ll

(1-a) / (cozy; (1.4-2)

where S = 0*A +B*, V

o is the mean ionic activity

....

coefficient and d is the degree of dissociation of the ion

pair. Fuoss and Onsager (6) revised this equation by adding

I

higher order terms to the conductance equation:

A =I\o - S(CO)1/2 + Ecolog(co)

.. 2 - 3/2 _

+ cho KMcow: J2(co) (1.5 1)
__ 2 2
K — (1-0) / co v+ (1.5-2)
_ _ __ 2
where E - Ele E2 b - e / DkT
E1 = 2.302602a2b2 / 24c 5:2 = 2.302602abB* / 15c1/2
02 =xNe2c / 125DkT ln v_,_= -b¢/ 2(1+¢a)

The coefficients J1 and J2 are solvent dependent and can be

expressed as a function of the size parameter a. This
equation has been very successful in fitting conductance

data of 1:1 electrolytes with 10 < K < 1000 M-l. For K < 10

M"1 an almost horizontal line in the plot of equivalent
conductance versus Cl/2 is obtained, which makes the

calculation difficult. For K > 1000 14-1

the slope of the
conductance plot becomes (very large and is insensitive to
changes of the equivalent conductance at infinite dilution.

Many other models have been developed to obtain more

accurate fits to the conductance data since 1957, when the

Fuoss-Onsager equation was introduced. The conductance at

infinite dilution can be determined with better accuracy by

 

 

8

usingeither the Fuoss-Hsia or the Pitt equation (7,8,9).
Justice (10,11) modified the Fuoss-Onsager equation by
replacing the ion size parameter, a, with the Bjerrum
distance, q. Fuoss (12) proposed. a model in ‘which the
solvent-separated ion pair is an intermediate transition
state between ‘unpaired ions and. a contact ion. pair. A
distance parameter R was defined as the distance from a
reference ion beyond which continuum theory may be applied.
Ions in the region a 5 r g R were considered as ion pairs.

1.3 MW

The importance of NMR to the study of solution effects
was first noted as early as 1951 by Arnold and Packard
(13) . In the past 37 years there has been a spectacular
growth in the literature concerned with this broad topic.
In particular, electrolyte solutions have been widely
studied and the knowledge of the structure of these
solutions has been considerably advanced by the use of this
technique. Nuclear properties of some of the elements are
listed in Table 1. All alkali nuclei have the spin greater
than 1/2 and therefore the nuclei have a quadrupole moment
that should considerably broaden the resonance lines.
However, with the expection of rubidium the resonance
frequencies are measurable: and in the cases of Li-7 and Cs-
133 the natural linewidths are less than 1 Hz. Thus in most
cases chemical shifts can be measured quite precisely.

Halogen nuclei also show great sensitivities of both the

chemical shifts and, especially, the quadrupole relaxation

 

9

.Table 1. Nuclear Properties of Some Elements

Nucleus

Taken from Reference 127.

NMR Frequency
at 14.09 Kilo-

gauss (MHz)

(’3)

Abundance Spin Relative to

Natural Nuclear Sensitivity

1H at

Constant Field

Li-7
Na-23
K-39
Rb-87
Cs-133
C1-35
Br-79
I-127

T1-205

23.315

15.868

2.800

19.630

7.864

4.892

16.202

12.003

34.619

100
93.08
27.2

100
75.4
50.57

100

70.48

0.294

0.0927

0.000508

0.177

0.0474

0.0047

0.079

0.093

0.192

 

10

of halide ions to ion-solvent and ion-ion interactions.
Halogen NMR should be capable of providing detailed insight
into a number of structural and dynamic features of
electrolyte solutions. Thallium-205 is the third most
sensitive spin-1/2 nuclide (14) and its chemical shift is
extremely sensitive to solvent and concentration, which
make thallium-205 NMR a good probe of the interactions in
electrolyte solutions.

Chemical Shift Studies - Bloor and Kidd (15,16) and
Richards gt a]._,_ (17) showed that in aqueous solutions the
resonance frequencies of alkali cations are strongly
concentration and anion dependent. The variation in
cationic chemical shifts as a function of concentration and
of counter ions is undoubtedly due to cation-anion
interaction. Richards gt 1]”, (18,19) postulated three
types of ionic interactions which may contribute to the
chemical shifts: 1) short range repulsive overlap of the
ionic wave functions during collisions: 2) electrostatic
polarization of cationic wave functions by neighboring
ions; and 3) the van der Waals forces. These studies have
been extended to a variety of nonaqueous solvents by Popov
and co-workers (20). A linear relationship between
Gutmann's donor numbers* and Na-23 infinite dilution
chemical shifts were obtained (21) (Figure 1). However,
such a correlation doesn't exist in the case of Li-7, where

paramagnetic and diamagnetic contributions to the chemical

shift nearly cancel each other out. Concentration dependent

 

11

L! \
“[u\
\
n. \
\
a” \ t
\
tl- \ 40
”h \ 3i
, \
D
i u‘ “
9 .b \
a 7I- \
" \
t " ‘“
3- \
4- \\
3- \
\
a =- ~
In 7"
l \
0
' ” «x
In
2

2!
I

 

.3.
O
b-
.I-
an
‘-
8.
g.-
3..
“I.
Sr-
3.4

Figure 1. Infinite Dilution of Sodium-23 Chemical Shifts
vs. Gutmann Donor Numbers. 1. NM, 2. MeCN, 3.
ac, 4. Ethyl Acetate, 5. THF, 5. DMF, 7. DMSO,
8. Py, 9. HMPA, 10. Water. Taken from Reference

21.

 

 

12

Cs-133 chemical shifts were used to calculate contact ion
pair formation constants for some salts (22). Khazaeli gt
‘g1& (23) studied the concentration and temperature
dependence of the Cs-133 chemical shifts for different
cesium salts and fitted the data using a model involving
ion pairs and triple ions. The great discrepancy between
the results from ‘the different. methods was ascribed to
theoretical and experimental difficulties of the
conductance method. F-19 chemical shifts of the
hexafluorophosphate ion in a number of solvents have been
measured by DeWitte and Popov (24). Small upfield shifts as
a function of NaPF6 concentration are indicative of solvent-
separated ion pairs. Hinton and Metz (25) measured the T1-
205 chemical shifts as a function of the concentration for
TlNO3 and TlClO4 in liquid ammoniam and suggested the
presence of free, fully solvated thallium ions, as well as
ion pairs and of higher order aggregates as the
concentration increased. Analysis of the low concentration
*The Gutmann donor number is an empirical scale of the
donor ability of solvents. It is based on the enthalpy of
formation of the 1:1 complex between a dilute solution of a
given solvent S and antimony pentachloride in 1,2-
dichloroethane solution.

1,2 DCE
s + SbCls ------- -> S(SbC15)

-AH(Kcal/mol) = Gutmann donor number

 

13
data between 0 and 30 C’C allowed the determination of
TlNO3 ion. pair formation. constants and of thermodynamic
parameters (AH and.AS).

Quadrupole Relaxation Studies - For the majority of
nuclei with spin > 1/2, an important relaxation mechanism,
and usually the predominant one, is that of quadrupolar
relaxation. There are two major, but rather different
relaxation theories. One relaxation model treats the
electric field gradients at the relaxing ion as arising
from fluctuations in the electric environment of the ion
produced by solvent dipoles and counterions - i.e.,
electrostatic origin (26,27). The other treats the electric
field gradients as arising from short-range interactions
resulting when an ion collides with a solvent or other
solute particle and suffers a distortion in the symmetry of
the electron cloud. This mechanism is not electrostatic
because the electric field. gradient is propagated. by a
transient rearrangement of the observed ion-electron
distribution (28).

Since the formation of ion pairs in solution clearly
affects both mechanisms, an often-used strategy is to study
the concentration. dependence of 'the relaxation ‘time and
extrapolate to the limit of infinite dilution. A large
number of relaxation investigations have been carried out
for Li-7, Na-23 and Cs-133 nuclei (29,30,31). For alkali
halides (except fluorides) the concentration dependence of

the relaxation rates is quite small and the sequence of Cl-

l4

, Br__ and I- ions bears no simple relationship to the ion
size. A systematic investigation of ion quadrupole
relaxation in a broad range of solvents was also made by
Weingartner and Hertz (32). From the infinite dilution
values of the relaxation rate of the halide ions, they
found that halide ion solvation is weak in formamide,
acetone and dimethyl sulfoxide, whereas methanol, ethanol
and formic acid are characterized by much more effective
relaxation and thus stronger salvation.

1-4 W

In 1965, Evans and Lo (33) found a broad band in the

far infrared spectra (450-50 cm-1

1

) of tetraalkylammonium

salts in benzene, at z 120 cm- for the chlorides and at z

80 cm"1

for the bromides, which could be assigned to the
interionic vibration. Edgell and co-workers (34,35) studied
alkali solutions in tetrahydrofuran and dimethylsulfoxide.
They also observed broad bands whose frequencies varied
with the nature of the cation and the anion: therefore
these bands were assigned as interionic vibrations.

Popov and co-workers (36,37,38) studied numerous salts
and solvents and found that in the far infrared region the
vibration of a cation in a solvent cage could be observed.
The peak position depends on the mass of the cation and on
the solvent. In highly solvating solvents the peak position

is independent of the anion while in poorly solvating

solvents the band frequencies can vary with the anion,

- which showed that anions can penetrate into the inner

 

15

solvation shell of the cation and lead to the formation of
contact ion pairs. The above results indicate that
depending on the system, one can observe l) interionic
vibrations of a cation-anion pair, 2) vibration of the
cation in a solvent cage with one or more of the solvent
molecules replaced by the anion and 3) vibration of the
cation in the solvent cage without replacement by the
anion. It is interesting to note that when a cation
complexing agent (crown ether or cryptand) is added to the
above solutions the vibrational bands of the last two
vibrations disappear and a new band appears which is
completely independent of the solvent and of the anion and
which corresponds to the vibration of the cation in the
macrocyclic cage (39,40).

Edgell e1; ah (41,42) investigated the effects of
cation, anion, solvent and concentration on the far
infrared band of NaCo(CO)4 in tetrahydrofuran solution. The
additional Raman and IR data for the Co(CO) 4- anion were
explained by the existence of different sites for the anion
in tetrahydrofuran. It was concluded that in THF NaCo(CO) 4
exists -mainly as solvent-separated and contact ion pairs
(43,44,45).

Salts with polyatomic anions (9;th N03-, 0104-, SCN-,
etc.) , often show distinct bands for the free and the
contact ion paired anion. In the special case of a highly
hydrogen-bonding solvent with small polarizable molecules,

the vibrational modes of a polyatomic anion involved a

16
solvent-shared ion pair may be sufficiently perturbed to
permit their distinction from those of the free anion. This
has been observed in liquid ammoniam solutions of some
thiocyanate salts (46) . Deconvolution of the bands and
quantitative measurements of integrated band intensities
yield concentration quotients for the ionic association.
Several ion pair formation constants have been obtained in
this manner by IR and Raman spectroscopy (47) . Recently, a
detailed study of the variation in the Raman spectrum of
aqueous NaClO4 with changing Na+ concentration and
temperature was reported by Miller and Macklin (48) . Based
upon resolution and assignment of components in the
featureless contours assigned to u

(A and u 2 (E)

1 l)
vibrations of the T d perchlorate ion, they calculated the
equilibrium constant for contact ion association as 0.022

In"1 at 22 oC.

1.5 NW

The major chemical difference between the NMR and ESR
methods as applied to the study of ionic solvation is that
in ESR at least one component must be specifically
paramagnetic. Detailed study of ion pairing is possible
because of the appearance of hyperfine structures in the
ESR spectra of radical anions. The hyperfine splitting is
caused by the interaction between the unpaired electrons
and nuclei of the diamagnetic cations. Weissman and co-

workers (49,50) found by ESR technique that there are

associations of benzophenone, ketyl and naphthalene radical

 

17

anions with sodium cations. ESR also can distinguish
different types of ion pairs. For example, Hfifelmann and co-
workers (51) observed two sets of metal hyperfine couplings
for sodium naphthalenide. Sodium naphthalenide forms
contact ion pairs in tetrahydrofuran and solvent-separated
ion pairs are formed as tetraglyme is added. The intensity
of the line assigned to ‘the solvent-separated ion. pair
increases with the increase in tetraglyme concentration.
The equilibrium constant between the two kinds of ion pairs
was determined.

The spectrophotometric study of ion pair formation was
first reported by Symons and co-workers (52). They observed
marked shifts in the position of the first electronic
absorption band of a variety of iodides as a result of
changes in the solvent polarity, counterion and
temperature. Griffiths and Wijayanayake (53,54) extended
this work and made a systematic study of ion pair
formation. They found that interactions with
tetraalkylammonium ions cause a large shift to low energy
for the first electronic absorption band, and interactions
with alkali metal cations cause a small high-energy shift.
The former shift is a result of direct displacement of
solvent from the iodide and the latter is due to the
formation of solvent-shared ion pairs. Gilkerson and co-
workers (55,56,57) studied the ion association of alkali
picrate salts in 2-butanone and 2-propanol. A. new' band

assigned to the contact ion pair was found as an excess of

18
alkali cations was added to the alkali picrate solutions.
The ion pair formation constants were calculated and agreed
with those from the NMR. and electric conductance
measurements. They concluded. that for’ different
spectroscopic techniques, the extent of ion pair formation
should be the same provided that the same concentration
ranges are covered and the same expressions for the
activity coefficients are used.

2. ngpleggtign of Metal Long by grog; Ethers and
2mm

2-1 1.853.429.2123

Since Pedersen (58-60) and Lehn (61-64) discovered
macrocyclic crown ethers and cryptands which can form
stable complexes with metal ions, studies of these ligands
and their complexes have become a very popular field of
research (65). The structure of some crown ethers and
cryptands are given in Figure 2. Particularly their
complexes with alkali and alkaline earth metal cations can
be used as models for investigation of ion transport
through membranes in biological systems (66).

Complexes of different stoichiometries can be formed
between crown ethers and metal ions. For example, potassium
cation can form a 1:1 complex with 18-crown-6 (67), a 2:1
(crown/K+) complex with benzo-lS-crown-S (68), a 1:2
complex with dibenzo-24-crown-8 (69) and a 'wrap around'

1:1 complex with dibenzo-Bo-crown-lo (70) (Figure 3). Two

kinds of complexes between cesium ions and cryptand-222 can

 

19

o _ Fo/fi
(Si c”; c: ‘5
bk) K/
12C4 15C5 18C6
(:[gflm’fitfij [:j:€\v”E):D <_(—\_>
1,0»; k¢&\J *zflkdp’m

DC18C6 D318C6 DA18C6

new I». .° .
Vb 086%?) aiouoejg

. (5.101

DBZlC7 DBZ4C8 DBZ7C9
$513 w

W W
C222 C221 C211

Figure 2. Structures of Some Macrocyclic Polyethers.

 

20

K+(18C6) K+(DB15C5)2

W
(6°00

age 7.}
. @Eo:o:o§oj:® ‘06:,900 @

2+(Deuce) K+(DB30C10)

Figure 3. Various Stoichiometries of K+-Crown Ether

Complexes. Taken from References 67-70.

21

also be formed (71). The cation is inside the cavity of the
cryptand in a inclusive complex and there is no anion and
solvent dependence of the cesium-133 chemical shift. On the
other hand, the cation can be only partially inside the
cavity in an exclusive complex: then the cesium-133
chemical shifts of the complexes are dependent on the anion
and the solvent.

Different kinds of physicochemical techniques have been
used to study the thermodynamic and kinetic parameters of
the complexation reaction, such as potentiometry,
polarography, conductance, nuclear magnetic resonance
spectroscopy, ultraviolet-visible spectroscopy, infrared
spectroscopy, calorimetry and liquid-liquid partion. Of the
various methods used for such studies, metal NMR
(especially alkali nuclei and thallium-205) has been shown
to be a very powerful and sensitive technique since the
chemical shifts of metal nuclei are very sensitive to the
immediate environment of the metal ion in solution.
Recently, a competitive NMR technique has been developed to
measure formation constants for cations whose NMR
properties are not particularly well-suited to conventional
NMR study and formation constants which are too large to be
measured directly (72).

2.2 e ct v t Co exa 0

One of the most striking characteristics of the
macrocyclic polyethers is their ability to form complexes

selectively with various cations. The factors affecting the

—

22

formation and stability of these complexes include the
relative size of the ion and the macrocyclic cavity, the
electrical charge of the cation, the nature of the donor
atom in the ring, the number of binding sites in the ring,
steric hindrance in the ring, the solvent and extent of
solvation of the ion and the ligand, the binding sites and
the nature of the counterion.

The ionic diameters of some cations and the estimated
cavity sizes of the holes of selected cyclic polyethers are
listed in Table 2. Generally, crown compounds form the most
stable complexes with those metal cations whose ionic
radius best matches the radius of the cavity formed by the
ring upon complexation (58,59). Figure 4 shows how the
complex formation constants in the case of l8-crown-6 vary
with the ratio of cation to cavity diameter (73). The
maximum stability for complexes of 18-crown-6 with alkali
and alkaline earth cations occurs at a cation-to-cavity
diameter ratio of unity. Large crown ethers are not as
selective as 18-crown-6 because of the formation of
complexes of variable stoichiometry. In the case of
cryptands; there is a much better relationship between the
stability of the complexes and the relative sizes of the
cation and the cryptand cavity (Figure 5). Each alkali
cation is preferentially bound to the cryptand with the
proper size (74).

Generally, with a similar cation radius, a bivalent ion

has a higher complex formation constant than a monovalent

23
Table 2. Diameter in‘K of Some Univalent Cations
(in Crystals) and Macrocyclic Polyethers.

Taken from Reference 58.

Cation Cation Macrocyclic Macrocyclic

diameter polyether polyether diameter

(A) (R)

Li 1.20 12-crown-4 1.2-1.5
Na+ 1.90 15-crown-5 1.7-2.2
K+ 2.55 18-crown-6 2.5-3.2
Rb+ 2.95 21-crown-7 3.4-4.3
Cs+ 3.34 Larger than 21-crown-7 > 4
T1+ 2.80 Cryptand 222” 2.8

Cryptand 221 2.2

Cryptand 211 1.6

 

24

4.0 -I
8a"

3.5 -I

3.0-4

  
  

.0".

 

tog K 21H C j
o o
k/‘iJ
2-0 " Cavity Diameter were
1.5“
1 .0« Co’
0.5 1f r T fl 1 r 1

 

0.7 1.0 1.3
Die. MW! Oil. Cavity

Figure 4. Selectivity of 18-crown-6: log K (Formation
Constant) Values for the Reaction of 18-crown-6
with Metal Ions in Water vs. Ratio of Cation

Diameter to 18-crown-6 Cavity Diameter. Taken

from Reference 73.

 

 

 

 

25
10.0w Cwlb
9.0« 122.13”
8.0«
[2.1.1]b b
.{1zfl
7.o~ - -
‘0‘ [1321‘
beK =
s.o«
I.
‘ I
4.0- ,’ \.
I
I
I
I,
3.0- I
I .
I
I
10., I ‘.:_____:
o Tr W T W T
7': 100 125 | 1'50 175
i
no Mr K’ my c3:

Metal Ion Rodin (pm)

Figure 5. Selectivity of Cryptands: log K (Formation
Constant) Values for the Reaction of Several
Cryptands with Alkali Metal Ions vs. Cation

Radius. Taken form Reference 74.

26

ion. ,For example, it has been shown that complexes of
bivalent cations with dicyclohexyl-lS-crown-é are more
stable than those of univalent cations in aqueous solution
(75). However, for small cations, monovalent cations form
more stable complexes with dicyclohexyl-lB-crown-G than the
divalent cations. This is due to the fact that the divalent
cation is more solvated.

The type and number of donor atoms in the ring are
also important factors in determining the stability of the
complex. Substitution of sulfur or nitrogen for oxygen in
the crown ether ring reduces the affinity of the ligand for
alkali and alkaline earth cations because of the reduction
of the cavity size and the lower donor abilities of S and
N. However, an opposite stability trend is expected when
the cation-ligand bond has appreciable covalent character
as in the case of Ag+---N or Ag+---S interactions (76).
Cram et a1; (77) investigated the effect on complexation
by varying the number of donor atoms in the ring without
changing the size of the ring. They found that 18-crown-6
is a much better host for tetrabutylammonium cation than 18-
crown-5,— which has one less donor atom in the ring. The
formation of a complex in solution is not only a
competition for the metal ion between the ligand and the
solvent molecules, but also a competition for the ligand
between the metal ion and the solvent molecules. Therefore,
changes in the solvent can produce significant changes in

the formation constants of the complexes. Popov and co-

27

workers (78-83) have studied extensively the complexation
of alkali cations by macrocyclic ligands in a wide variety
of nonaqueous solvents. The results show that the stability
of complexes depends not only an the relative size of the
ion and macrocyclic cavity, but an the nature of the
solvent. Solvent-ion, solvent-ligand and ion-ion
interactions are important factors to the stability of the
complexes. For solvents with low dielectric constant, ion
pair will interfer with the complexation reaction. The
crown-solvent interaction is an important factor which
affects the complexation reaction. Formation constants of
acetonitrile and nitromethane complexes with 18-crawn-6
have been determined by Mosier-Bass and Popov (84) . The
results show clearly that these ligand-solvent interactions
can affect the complexation to a great extent.

2.3 111W

Kauffman et a1; (85) have discussed the contribution of
enthalpy and entropy to the complexation reaction. They
postulated that the enthalpy of formation of the complex is
influenced by: 1) the replacement of the first salvation
shell of the cation by the ligand, 2) the change in the
interaction with solvent molecules outside the complex as
compared to those outside the first salvation shell of the
cation, 3) the change in inter-binding site repulsions, 4)
the change in ligand salvation enthalpy upon complexation
and 5) the steric deformation of the ligand by the cation.

- They also attributed the changes of the entropy of the

 

28
cryptand complex to: 1) desalvation of the cation, 2)
release of solvent bound to the ligand, 3) changes in
ligand. internal entropy caused by orientation,
rigidification and conformational changes, 4) formation of
a single complex from two species, and 5) salvation of the
complex.

The equilibrium constant for the complexation of cesium
ions by cryptand-222 has been studied as a function of
temperature in several nonaqueous solvents (71). The values
of enthalpy and entropy showed that the complex is enthalpy
stablized but entropy destabilized. Popov and co-workers
(79,82) have studied the complexation of the cesium ion
with large crown ethers in nonaqueous solvents. In all
cases the complexation was enthalpy stabilized but entropy
destablized. They assumed that the decrease in entropy upon
complexation is related to a change in the conformational
entrapy of the ligand, although it is not the only factor
governing the change in entropy of complexation.

There have been many thermodynamic studies to analyze
the origins of the macrocyclic effect (63) and cryptate
effect.(86), and different conclusions were drawn. Hancock
and Martell (87) , in their recent review paper, concluded
that for both effects considerable stabilization is derived
from the greater basicity induced in donor atoms as
ethylene bridges are added. The important factors are: 1)
desalvation effects, where steric constraints to salvation

of the donor atom in the free ligand lead to increase

29
complex stability, 2) enforced dipole-dipole repulsion in
the ligand, which is relieved upon complex formation, and
3) structural preorganization of the ligand such that the
donor atoms in the free ligand are already correctly

oriented for complex formation.

3. - a d
3.1 Wm
N-Methylformamide (NMF) is a methyl-substituted

derivative of formamide. Many of its physicochemical
properties show intermediate values between those of
formamide (FA) and N,N-dimethylformamide (DMF). Comparisons
of some properties are made in Table 3 (88). The rotation
about the C-N bond of the NMF molecule is very severely
hindered (89) by an activation barrier of about 59 kJ/mol,
and the trans- and cis-conformations (Figure 6.a) both can
exist in the NMF system (The distinction between the cis-
and trans-forms is based on the ~C0-NH- bond in peptides).
Vibrational spectroscopic studies in the liquid phase
(90,91) and in carbon tetrachloride solutions (92) showed
that the trans-isomer was predominant in the liquid and in
the solution phase. These results agreed with conclusions
drawn from gas phase measurements obtained by infrared
(93) , microwave spectroscopy (94) and the electron
diffraction method (95). NMR spectra obtained in the liquid
phase (96) and dipole moment measurements of NMF in dioxane
solutions (97) indicated that 90-92% of the NMF molecules

were present in the trans-farm. Sugawara e3; 3],. (98) also

30

Table.3. Physicochemical Properties

of Formamide (FA),

N-methylformamide (NMF) and N,N-dimethylformamide

(DMF). Taken from Reference 88.

FA
Molecular weight 45.041
Boiling point(°C) 210.5
Melting point(°C) 2.55
Density(g/cm3,25°C) 1.12918

Refractive index(25°C) 1.44582
Viscosity(cp,25°C) 3.302

Dielectric constant 111.0

(20°C)
pKa(in water,20°C) -0.48
Dipole Moment(Debye) 3.37
Donor number ' 24*
Acceptor number 39.8*

59.058
180-185
-3.8
0.9988
1.4300
1.55
182.4
(25°C)
-0.04
3.85
24**

32.1*

73.095
153.0
-50.43
0.94397
1.42817
0.802
35.71
(25°C)

.000]-

* From "The Donor-Acceptor Approach to Molecular

Interactions" V. Gutmann, Plenum Press, New York (1978).

** From Yu.M. Kessler, A.I. Mishuatin and A.I. Podkovyrin,

£1 59181 989ml. 5. 111 (1977)-

31

(a)

H ,H H\ 40
\N"C /~-C\
H—c *0 H—o H
/
H/it '1 H
(trans) (cis)

(b)

._. ‘ 4: x

- - bondin
.. CH NH(CH3) hydrogen 9

Figure 6. (a) Cis- and Trans-forms of NMF.
(b) Linear Chain Structure of Liquid NMF.

Taken from Reference 99.

32

determined by theoretical calculation that the trans-form
is more stable than the cis-form. A linear and flexible
chain structure -(Figure 6.b) was proposed for the liquid
structure of NMF on the basis of scattered intensity data
by the X-ray diffraction method (99). The average degree of
association of NMF has been calculated from the dispersions
of the dielectric constants by Durov (100). The values
varied from 5.3 to 2.8 at temperatures 273 K - 393 K.

It is noted that the dielectric constant of NMF is
extremely large (182.4 at 25°C) (101), 40% larger than that
of formamide, although the dipole moment of NMF is only 15%
larger. Such a high dielectric constant can be explained as
the result of the fact that NMF molecules are chainwise
associated by hydrogen bonding with the single amino
hydrogen such that adjacent molecular dipole moments are
nearly parallel (102) . Moreover, cyclic hydrogen-bonded
dimers, which can lead to a decrease of the dielectric
constant, are less readily formed in NMF than in formamide
(103) . Thus NMF has a inuch larger dielectric constant than
that of formamide. However, the dielectric constant could
be decreased as salts are added. to NMF. For example, in
0.7 M NaI/NMF solution the dielectric constant is 100 at
25 °C.

3.2 8211511233

Salvation studies in the NMF-solution system have
received some previous attention . From viscos ity

measurements on some common salt solutions in NMF, Rastogi

33

(104,105) concluded that the salvation of the cations is
through electrostatic bonds and that of the anions occurs
through H-bonds: the tetraalkylammonium ions are not
solvated due to their large size and low surface charge
density. Bukowaka (106) did Raman and infrared studies on
LiClO4 and NaC1O4 in NMF solutions. Two possible salvation
models were suggested: 1) cation-solvent interaction occurs
through the oxygen and (or) the nitrogen atoms of the NMF;
2) very stable 1:2 chelate complexes between cations and
NMF molecules could be formed. Banner and Jordan (107)
studied the N-H stretching band of NMF for some lithium
salt/NMF solutions and found that with polyatomic anions
the N-H band shifts to higher frequencies (in the order:
PF6_ > C104- > NO3-) and to lower frequencies as the anions
are halides (in the order: F- > Cl-z Br- > I-) (Table 4).
It was concluded that the interaction of the N-H proton
with the polyatomic anions of low charge density is weaker
than with the NMF carbonyl group but it interacts strongly
with the halide ions in a manner of hydrogen bonding.

Paul e3; al. (108) determined the solubility of some
alkali and ammonium halides and thiocyanates in NMF (Table
5) . They found that thiocyanates and iodides are the most
soluble of the salts studied and attributed this to the
higher polarization of these ions and relatively higher
covalent character of these salts than that of the
corresponding chlorides and bromides. Heats of soultion,

heats of salvation (Table 6), ionic salvation enthalpies

 

34
Table 4. Wavenumber Shifts (cm-l) of N-H and C20
Stretching Bands in Salt/NMF Solutions

(4 M). Taken form Reference 107.

N-H C=O
Salt Av * Av *
NaPF6 -23 -26
LiClO4 -80 -38,-26
LiNO3 -l37 -30
Pure NMF -159 -36
Lil -183 -34
NaI -185 -30
LiBr -213 -32
LiCl -214 -34
CsF(dilute) -180,-260(sh) -33

* Calculated from reference 107, all wavenumbers
for both bands are subtracted from those of

dilute NMF in CCl4 solutions.

35

_Table 5. Solubility of Salts in NMF at Different

KBr
KI
KSCN
NH4C1
NH Br
NH I

NH SCN

Reference 108.

20.61
35.20

5.60
29.77
79.47

50.29

10.25
44.21
80.90

6.59
32.55
38.01

74.35

15.09
21.96

4.75
28.99
87.02

35.62

10.42
45.92
77.27

5.57
32.85
58.53

71.74

Temperatures (g/100g Solvent). Taken from

11.10
46.84

73.07

 

36

Table 6. Lattice Energies, Heats of Solution and Heats of
' Salvation of Some Alkali Metal Salts in NMF
(kcal/mole) at 26.9 0C. Taken from Reference 112.

Lattice Heat of Heat of
Salt Energy Solution Salvation
LiCl -201.20 -13.10 -214.30
LiBr -19l.20 -16.14 -207.34
LiSCN -182.70 -11.80 -194.50
LiClO4 -169.40 -13.80 -183.20
NaCl -185.80 -l.95 -187.75
NaBr -176.60 -4.39 -180.99
NaI -164.50 -7.50 -172.00
NaSCN -163.40 -4.78 -168.18
NaClO4 -153.80 -3.35 -157.15
KCl -168.90 +0.30 -168.60
KBr -l61.40 -0.80 -162.20
KI -151.10 -3.20 -154.30
KSCN -148.70 -1.05 -149.75
KClO4 -139.70 +1.50 -138.20
RbCl -162.80 +0.80 -162.00
RbBr -155.90 -----------
RbI -147.3O +1.50 -145.80
RbClO4 -134.90 +2.70 -132.20
CsCl -155.20 +0.55 -154.65
CsBr -149.30 -----------
CsI -140.30 +1.00 -139.30
CsClO4 -128.6O +3.10 ~125.50

37
and ionic entropies were measured by Weeda and Somsen
(109,110), Criss g al- (24) and 0111 sf g. (112) from
thermochemical studies on some ZNMF salt solutions. They
found that NMF has a tendency to approach nearer the anions
than the cations. It is known that anions are more solvated
than cations in protonic solvents.

Emsley gt al. (113) studied the NMF-fluoride systems,
using both ab inito LCAO-MO-SCF calculations and
spectroscopic techniques (IR, H-1 and F-19 NMR). From the
theoretical calculation they found the amide-fluoride
hydrogen bond is the second strongest hydrogen bond known.
The most stable conformation for the NMF-fluoride complex
is to have F- located trans to the carbonyl group with an N-
methyl C-H bond eclipsing the C-N band. In support of
theoretical calculations, some IR and NMR studies were

carried out. In the infrared studies, a "new" 1600 cm-1

band which shifts from 3465 cm-1

(for monomeric NMF) was
found as CsF was added to NMF solution. This band was
assigned to the N-H (or H-F) stretching band. H-1 and F-19
NMR studies also showed that there is an unusually strong
hydrogen bond between NMF and fluoride. However, Symons at
al. (114) disagreed with the conclusion that there is such
a strong hydrogen bond in the NMF-fluoride system, on the
basis of their IR and NMR studies of chlorides, fluorides
and xenons in various solvents. Moreover, they concluded

that F-19 chemical shifts are not a good measure of the

strengths of hydrogen bonds to fluoride.

38

Solvated radii and salvation numbers (Table 7) for
various ions in NMF were also calculated by Paul (115) and
Pontani gt g1. (116). The salvation numbers of anions have
been found to be higher than that of the cations of
comparable sizes. The anions may be arranged in the
following order on the basis of their salvation numbers in
NMF: C1’ > Br- > I‘ > scn‘ > c104'.

3.3 - o teractions

Although it is expected that in solvents of high
dielectric constant electrolytes should be completely
dissociated, interionic interactions have been studied in
NMF solutions by several groups of investigators. Paul gt
g1. (115) studied electrical conductance of 21 alkali and
alkylammonium salts. The data, were analyzed using
Shedlovsky’s modification of the Onsager equation (117),
and the ion pair formation constants so determined were
very small .(Table 8). It should be noted that when
association constants are that small, very different
results can be obtained. by using more recently proposed
conductance equations such as those of Pitts (8), Fuoss and
Hsia (9).or Justice (11). Moreover, it has been generally
observed that in alkali family salts salvation depends on
the charge density of the cations, and therefore varies in
the order Li+ > K+ > Na+ > Cs+== Rb+. Obviously, ion pair
formation constants should vary in the opposite direction.

In the above-cited paper the order for perchlorates is Cs+

sz+ > Na+ z K+ > Li+. Singh and Gopal (118) also

 

 

39
Table 7. Radii (r) for Solvated Ions and Salvation Numbers

(n) for Various Ions in NMF at 25°C.

Ion rd) a na nb
Li+ 5.33 5.45 5.4
Na+ 4.29 3.34 3.3
K+ 4.23 3.13 3.1
Rb+ 4.11 2.83 ---
Cs+ 3.99 2.51 2.4
NH4+ 3.33 1.50 ---
Cl“ 4.43 3.51 3.5
Br' 4.20 2.85 2.9
I" 4.05 2.41 2.4
SCN“ 4.01 2.33 ---
C104‘ 4.05 2.25 2.2

a: from reference 28.

b: from reference 29.

40
Table 8. Ion Pair Formation Constants of Electrolytes

in NMF at 25°C. Taken from Reference 115.

Ian Pair Formation

Electrolyte Constant(i0.02)
LiCl 1.52
NaCl 0.58
KCl 1.18
RbCl 1.09
CsCl 1.05
LiSCN 1.43
NaSCN 1.05
KSCN 1.43
NH4SCN 1.27
LiClO4 1.52
NaClO4 2.17
KClO4 2.04
RbClO4 3.03
CsClO4 2.86
NH4C104 2.17
MeNH3ClO4 3.70
MezNHZCIO4 3.03
Me3NHC104 4.17
Me4NC104 2.33
Et4NC104 2.50

BU4NClO4 2.27

41
concluded from their electric conductance results that
there is cation—anion interaction for some
tetraalkylammonium salt solutions in NMF. However, it
should be noted that the electric conductance studies of
French and Glover (119) show negligible association for
NaBr and KBr solutions in NMF and those of Johari and

Tewari (120) show that MgSO is completely dissociated in

4
NMF.

Thermochemical studies by Criss gt g1. initially seemed
to indicate some small degree of ion association in NMF
(121). However, their later work (122) withdrew the above
interpretation. A model of solvent-separated ion pairs was
proposed for NaI in NMF solution from vibrational
spectroscopy studies (123). It was suggested that a NMF
molecule could be attached to one iodide ion and one (or
two) sodium ions at the same time.

Thallium-205 NMR chemical shift measurements (124) on
r111103, TlClO4 and (CH3)2T1+N03' solutions in NMF showed
same concentration dependence of the salt, and the ion pair
formation constant for (CH3)2T1+N03- was reported as 11 i
10 M-1, indicating that the magnitude of the error makes it
impossible to confirm interionic association. Spin-lattice
relaxation measurements (125) on lithium ion and sodium ion
by lithium-7 and sodium-23 NMR measurements of LiCl and

NaBr in NMF show concentration dependence of T A short-

10
lived structure (Figure 7.a) for a solvent-separated ion

pair was proposed. The cation is tetrahedrally coordinated

42

 

“‘ 900 °

(13)

D“! 1

Figure 7. (a) A Proposed Structure of the Short-lived

Solvent-separated Ion Pair for NaBr in NMF.

(b) Proposed Structures of Some Higher Ion
Associates for RbI in NMF.

Taken From References 125 and 126.

 

43
by four solvent molecules. The anion approaches the cation,
penetrating into the dimple formed by three equatorial
solvent molecules. Furthermore, the same authors, on the
basis of studies of RbI in NMF (126) suggested that even

higher ion associates (Figure 7.b) could be present in this

high dielectric constant solvent.

W

XPERIMENTA ART

45

1- 3.91321

Lithium Chloride (Fisher), lithium bromide (Fisher),
lithium perchlorate (Alfa); lithium nitrate (Mallinckrodt)
and lithium hexafluoroarsenate (Alfa) were of reagent grade
and were used without further purification, except for
drying at 100°C for three days. Lithium iodide (Aldrich)
was dissolved in acetone and precipitated by placement in a
dry ice bath; the precipitate was dried under vacuum ' at
110°C for five days. Lithium thiocyanate (Alfa) was dried
under vacuum while the temperature was gradually increased
to 110°C over a five-day period. Lithium tetraphenylborate
was prepared by the method of Bhattacharyya, gt g1. (128)
and was dried at 80°C under vacuum for four days before
use. Lithium picrate was prepared by the reaction of 0.5
mole of lithium carbonate (Mallinckrodt) and 0.5 mole of
picric acid (MCB) in ethanol. It was recrystallized from a
‘bezene-acetone 'mixture and dried at 100°C for two days
before use.

Sodium chloride (Alfa), sodium bromide (MCB), sodium
iodide '(MCB), sodium perchlorate (G.F.Smith), sodium
nitrate (Baker) and‘ sodium hexafluoroarsenate (Aldrich)
were used as obtained, except for drying at 60°C for three
days. Sodium tetraphenylborate (Aldrich, Gold Label) and
sodium thiocyanate (Baker) were used as obtained, except
for drying for three days under vacuum at 45°C. Sodium

picrate was synthesized by mixing 0.5 mole of . sodium

46
hydroxide (Aldrich) and picric acid (MCB) in ethanol. It
was recrystallized three times from ethanol and dried under
vacuum at 70°C for five days before use.

Cesium fluoride, cesium chloride, cesium bromide,
cesium iodide, cesium perchlorate, cesium nitrate (Alfa)
and cesium carbonate (Aldrich) were used as obtained,
except for drying at 120°C for two days. Cesium thiocyanate
(Pfaltz and Bauer) was recrystallized twice from methanol
and then dried at 45°C under vacuum for two days. Cesium
tetraphenylborate was prepared by the metastatict reaction
as described by Mei (78) and then dried for two days under
vacuum at 70°C.

Thallium(I) nitrate (Alfa) and thallium perchlorate (K
a K) were purified by recrystallization from deionized
distilled water and then. dried at 120°C for two days.
Potassium iodide (Alfa) and potassium thiocyanate (Fisher)
were used as received except for drying at 45°C under
vacuum for two days. Tetraisoamylammonium thiocyanate (RSA)
and tetrapentylammonium iodide (Fisher) were dried under
high vacuum at 60°C for five days. Lithium thiocyanate(N-
15) was -prepared according to the method of Rykowsky and
Vanderplas (129), by which they synthesized potassium
thiocyanate(N-lS). The crude product was extracted with
several portions of acetone and the extract was evaporated
on a steam bath. The refined product was dried under vacuum

at 110°C for one day.

47
2- 22112222
N-Methylformamide (Aldrich) was refluxed over granulated
bariumroxide (Fisher) for one day, followed by fractional
distillation 'under reducedf pressure ‘with the ‘middle 50%
fraction retained. Acetonitrile (Baker) was refluxed with
CaH2 for one week, followed by fractionl distillation with
the middle 50% fraction retained. Both solvents were stored
in a dry box under dry nitrogen over freshly activated
Linde 3-3 molecular sieves. The water content of solvents
was determined by using a Varian Aerograph Model 920 gas
chromatograph and was found to be always below 70 ppm.
Deuterated water (KOR Isotopes) was used as received.
3- 1.125225
12-crown-4 (12C4, Aldrich) and 15-crown-5 (15C5,
Aldrich) were fractionally distilled under reduced pressure
and dried under vacuum for three days. 18-crown-6 (18C6,
Aldrich) was purified by converting it to the crystalline
acetonitrile complex (130), filtering the solution and
driving off acetonitrife under vacuum for two days at room
temperature. Dibenzo-18-crown-6 (D818C6, Parish) was
recrystallized twice from benzene and dried under vacuum
for three days. Dicyclohexyl-le-crown-6 (DC18C6, Parish)
was obtained as a mixture of isomers A and B and dried
under vacuum for three days. Diaza-lB-crown-G (DA18C6 or
[22], MCB), 21-crown-7 (21C7, Parish) and cryptands C222
(MCB), C221 (MCB) and C211 (MCB) were used as received

except for drying under vacuum for two days at room

48

temperature. Dibenzo-Zl-crown-7 (D821C7, Parish), dibenzo-
24-crown-8 (DBZ4C8, Parish) and dibenzo-27-crown-9 (DBZ7C9,
Parish) were recrystallized twice from n-heptane and then
dried under vacuum for two days at room temperature.
Pentaglyme (PG) was synthesized by N. Okoroafor following
the procedure of Izatt gt g1. (131). It was dried under
vacuum for three days before use.

B. METEQQQ

1. 22112122551321.2212:

For the NMR and IR concentration studies, the samples
were prepared by weighing out the desired amounts of salt
into 2 or 5 ml volumetric flasks and diluting to the mark
with solvents. For the complexation studies, the samples
were prepared by weighing out the various amount of the
complexing ligand into 2 or 5 ml volumetric flasks and
diluting to the mark with the metal ion solution which had
been. previously’ prepared. by dissolving the salt in the
desired amount of solvent.

2. EHB_H222!£222222

All nuclear magnetic resonance measurements were
carried ‘ out on a Bruker WH-180 multinuclear NMR
spectrometer with a field strength of 43.2 kilogauss. At
this field lithium-7, sodium-23, potassium-39, cesium-133,
thallium-205, nitrogen-14, nitrogen-15 and chlorine-35
resonate at 69.951, 47.61, 8.403, 23.62, 103.88, 13.01,
18.25 and 17.64 MHz, respectively.

All solutions were measured in 10 mm (or 20 mm for some

49

potassium salt solutions) od tubes (Wilmad) with a 4 mm od
insert (Wilmad) coaxially placed inside the larger tube.
The insert contained a chemical shift reference and the
lock compound (except for lithium-7 which was run without
lock). Table 9 lists the references for lithium-7, sodium-
23, potassium-39, cesium-133, thallium-205, nitrogen-14,
nitrogen-15 and chlorine-35 NMR measurements and their
chemical shifts relative to those of infinitely dilute
aqueous solutions of metal ions. All measurements (except
temperature studies) were taken at ambient probe
temperature, which is approximately 23°C.

Sensitivity enhancement. was achieved by signal
averaging and exponential line broadening of the free
induction decay (132). Since the signal/noise ratio is
proportional to (n)1/2, where n is the number of transients
scanned, the signal/noise ratio can be improved by
increasing the number of scans. Exponential line broadening
is the result of multiplying each point of the free
induction decay by a decaying exponent. This convolution
process improves the sensitivity and doesn’t introduce any
lineshape distortion (133). I

3- 13.322222292222

All infrared measurements were carried out on a Bomem-
DA3 FT-IR spectrometer with an RSX operating system and POP-
11 computer. The spectrometer has interferometric optics
and utilizes a globar source and a mercury cadmium tellwide

detector cooled by liquid nitrogen. The sample port was

50
Table 9. References for Multinuclear NMR Measurements and
Their Chemical Shifts (a) Relative to Those of

Infinite Dilute Aqueous Solutions of Metal Ions.

Nucleus Reference 5(ppm)
Li-7 0.1 M LiCl/DZO 0.00
Na-23 0.1 M NaCl/DZO -0.08

K-39 sat'd KNOz/DZO 3.00
Cs-133 0.5 M CsBr/020 8.94
Tl-205 0.1 M T1N03/D20 -l.50

N-l4 1.0 M NH4NO3/DZO ----

15
N 15 1.0 M NH4N 03/D20

C1-35 0.1 M NaCl/DZO ----

51
purged with nitrogen. For variable temperature studies the
sample was placed in a specially designed chamber through
which preheated air flowed. Temperature was varied by
changing the heating current and controlled. by a
thermistor. The spectra were obtained at a resolution of 2
cm-1. A standard demountable Barnes liquid cell with
calcium fluoride salt plates and a 0.05 mm path length was
used. In these studies the 4000-1100 cm.1 spectral region
was covered.
C- W
1. NMR Measurgmegts

All chemical shift measurements were corrected for the
difference in bulk diamagnetic susceptibility between
sample and reference by the equation of Martin, gt g1.
(127):

6

+ 4x/3 (X - X501) x 10 (2.1)

=
5 corr 5 obs ref

where xref and X sol are the volumetric susceptibility of
the reference and sample solvent respectively: 6 corr and

“obs are the corrected and observed chemical shifts. Since
the salt concentrations, were always low, the magnetic
susceptibility of the solution was taken to be equal to the
diamagnetic susceptibility of the solvent as indicated by
trampleman and Van Geet (134).

All chemical shifts are given versus infinitely dilute

aqueous solutions of lithium, sodium, potassium, cesium and

52

thallium ions for Li-7, Na-23, K-39, Cs-133 and Tl-205
measurements (Table 9). For N-14, N-15 and Cl-35
measurements, the chemical shifts were given versus the
references (Table 9). TheW chemical shift. was determined
(135) by 1) a computer fit (Lorentzian function) of the
three ‘most intense points or' 2) manual fit (Lorentzian
function) of the band profile. These methods are software
options in program NTCFTB which was installed in the WH-180
operating system. The choice between methods 1 and 2 was
made based on the width of the band. Method 1 was chosen
for all bands for which the line width was less than {2-
times the frequency' difference of adjacent. points.
Otherwise method {2— was used. The uncertainty of the
chemical shift was either 2 times the frequency difference
between adjacent points or one tenth of the line width.
Downfield chemical shifts were taken to be positive.

The formation constants of ion pairs and complexes were
calculated by fitting the experimental parameters (total
concentrations of metaI ions, chemical shift of free metal
ions and the observed chemical shifts) with the nonlinear
least-squares program KINFIT (136). Details on the use of
this program are given in the Appendices I and II (22).

2- IB.M22§E£2E2R£§

For all measurements the solvent spectra were recorded
and then subtracted from those of the solution samples to
eliminate interferences from the solvent bands. In the

studies of the CN-stretching vibration of various

53
thiocyanate salts, it was found that those bands are
composites. They were analyzed by the use of a PC program
which was written by H.-H Nam (137). The program includes
three parts: 1)deconvolution, 2)second derivative and
3)curve fitting. Details on the use of this program are

given in Chapter III.

C PT I

0 IO -80 O

- TH ORMAMIDE O UT ONS

55
A- IETBQDEQIIQN

Electrolyte solutions and ionic equilibria in solutions
are important not only in many branches of chemistry, but
also in biology and in -many industrial processes. In
academic research, the field of electrolyte solutions has
been largely neglected in recent years. Yet our knowledge
of the structure of electrolyte solutions still remains
unclear. For example, the over-all salvation number of an
ion is an inexact quantity and becomes largely a function
of the experimental technique. In order to unravel this
complex puzzle, the various Chemical species present in a
given solution should be identified, then their
interactions and equilibria could be studied.

A considerable amount of work is necessary to obtain a
better understanding of electrolyte solutions. Toward this
goal, one nonaqueous system (N-methylformamide) has been
investigated by a variety of physical techniques --- metal
ion NMR and infrared measurements, which have proven to be
very powerful tools for the electrolyte solution studies.

The purpose of this study is to elucidate the role of
the solvent in determining the type of species present in N-
methylformamide solutions of metal salts. By carefully
extending the earlier fundamental studies of Deverell and
Richards, Bloor and Kidd, and Popov and co-workers, it is
hoped to have more information about the electrolyte

solution structure in this solvent.

56
3- IQN:IQE_INIEBAQIIQN2_IE_N:EEIEXLZQBMAHIDB_§QLEIIQE§
1- 3222.12:

The concentration dependences of the Li-7, Na-23, K-39,
Cs—133 and Tl-205 chemical shifts of various metal salts in
N-methylformamide were examined. The results are given in
Tables 10-37 and Figures 8-12.

Lithium-7 chemical shifts show no change within
experimental errors as the concentrations of lithium salt
solutions are varied for nine lithium salts (chloride,
bromide, iodide, thiocyanate, nitrate, tetraphenylborate,
picrate, hexafluoroarsenate and perchlorate). For all the
sodium salts (chloride, bromide, iodide, thiocyanate,
tetraphenylborate, nitrate, picrate, perchlorate and
hexafluoroarsenate), potassium salts (iodide, thiocyanate
and nitrate) and cesium salts (fluoride, chloride, bromide,
iodide, thiocyanate, nitrate, tetraphenylborate,
perchlorate and carbonate), the chemical shifts change
linearly with the salt concentration. The slopes
(concentration shift), intercepts and correlation
coefficients (showing the quantitative relationship between
two variables, if this value equals -1 or 1, all points are
on a line) for each salt are listed in Tables 38 and 39.
Thallium-205 chemical shifts change nonlinearly with
concentrations of thallium nitrate and perchlorate. This is
believed to reflect ion-ion interactions. Ion. pair
formation constants for these two thallium salts were

determined by the procedure described in the Appendix

57
Table 10. Lithium-7 Chemical Shifts of LiCl

and LiBr in NMF.

Concentration(M) Chemical Shift
LiCl LiBr ($0.01ppm)
0.0127 0.0144 1.02 1.02
0.0425 0.0263 1.02 1.02
0.0802 0.0748 1.02 1.02
0.1286 0.1002 1.02 1.03
0.1557 0.1750 1.03 1.03
0.2371 0.2879 1.03 1.03
0.6346 0.4001 1.03 1.04
1.0061 0.8071 1.07 1.04

Table 11. Lithium-7 Chemical Shifts of LiI

and LiSCN in NMF.

Concentration(M) Chemical Shift
LiI LiSCN (10.01ppm)
0.1149 0.0139 1.02 1.02
0.1328 0.0198 1.02 1.02
0.1621 0.0387 1.03 1.02
0.2576 0.0956 1.03 1.03
0.3284 0.1434 1.04 1.03
0.7354 0.2944 1.05 1.03

58

Table 12. Lithium-7 Chemical Shifts of LiTPB

and LiNO3 in NMF.
Concentration(M) Chemical Shift
LiTPB Lino3 5 (:0 . 01ppm)
0.0045 0.0125 1.02 1.02
0.0112 0.0305 1.02 1.02
0.0173 0.0595 1.01 1.03
0.0307 0.0740 1.02 1.02
0.0464 0.1305 1.01 1.03
0.0565 0.2067 1.02 1.02
0.0627 0.2618 1.02 1.03

Table 13. Lithium-7 Chemical Shifts of LiClO4

and LiAsF6 in NMF.

Concentratian(M) Chemical Shift
LiClO4 LiAsF 6 a (10 . 01ppm)
0.0174 0.0153 1.02 1.02
0.0310 0.0345 1.02 1.02
0.0472 0.0569 1.02 1.02
0.0503 0.1136 1.02 1.02
0.0775 0.1432 1.01 1.02
0.1147 0.2019 1.02 1.02
0.1659 0.3919 1.02 1.02

59

Table 14. Lithium-7 Chemical Shifts of LiPi in NMF.

Concentration(M) 5(10.01ppm)
0.0138 1.02
0.0289 1.02
0.0694 1.02
0.1639 1.02
0.2278 1.02
0.2439 1.02
0.2662 1.02

Table 15. Sodium-23 Chemical Shifts of NaCl in NMF.

Concentration(M) 5(ppm) Linewidth(Hz)
0.0145 -4.59 42
0.0253 ' -4.70 43
0.0408 -4.59 43
0.0859 -4.68 44
0.0942 -4.55 45
0.1787 -4.57 48
0.2245 -4.55 49
0.2572 -4.55 51

0.2786 -4.65 51

60

Table 16. Sodium-23 Chemical Shifts of NaBr in NMF.

Concentration(M) 5(ppm) Linewidth(Hz)
0.0387 -4.66 44
0.1094 -4.65 49
0.2085 -4.62 50
0.2749 -4.61 51
0.4894 -4.56 55
0.6812 -4.51 58
0.9640 -4.39 63

Table 17. Sodium-23 Chemical Shifts of NaI in NMF.

Concentration (M) 5 (ppm) Linewidth (Hz)
0.0379 -4.65 44
0.1166 -4.63 46
0.1759 -4.58 48
0.3445 -4.52 51
0.5544 -4.47 55
0.8027 -4.33 61

61

Table 18. Sodium-23 Chemical Shifts of NaTPB in NMF.

Concentration(M) 6(ppm) Linewidth(Hz)
0.0199 -4.69 42
0.0427 -4.69 43
0.1008 -4.70 46
0.1774 -4.71 49
0.1823 -4.71 52
0.2059 -4.72 54

Table 19. Sodium-23 Chemical Shifts of NaAsF6 in NMF.
Concentration (M) 5 (ppm) Linewidth (Hz)
0.0229 -4.59 42
0.0337 -4.69 43
0.0538 -4.69 43
0.0661 -4.70 44
0.0799 -4.70 45

62

Table 20. Sodium-23 Chemical Shifts of NaPi in NMF.

Concentration(M) 6(ppm) Linewidth(Hz)
0.0265 -4.69 41
0.0611 -4.72 49
0.0681 -4.73 50
0.1846 -4.76 60
0.2752 -4.77 70
0.3318 -4.79 76

Table 21. Sodium-23 Chemical Shifts of NaSCN in NMF.

Concentration(M) 5(ppm) Linewidth(Hz)
0.0477 -4.69 40
0.1789 -4.67 46
0.1998 -4.66 46
0.2967 -4.65 48
0.3211 -4.53 49
0.4225 -4.62 50
0.5071 -4.60 50
0.8660 -4.58 52

63

Table 22. Sodium-23 Chemical Shifts of NaClo4 in NMF.

Concentration (M) 5 (ppm) Linewidth (Hz)
0.0143 -4.69 42
0.0321 -4.70 43
0.0422 -4.69 43
0.1208 -4.70 46
0.2055 -4.70 48
0.4276 -4.71 49
0.5266 -4.71 50

Table 23. Sodium-23 Chemical Shifts of NaNO3 in NMF.

- Concentration(M) 6(ppm) Linewidth(Hz)
0.0033 -4.69 43
0.0245 -4.69 43
0.0367 -4.68 44
0.0698 -4.69 45
0.1577 -4.67 46
0.4542 -4.67 49
0.6063 -4.67 51
0.7626 -4.67 53
0.8689 -4.68 56

64

Table 24. Potassium-39 Chemical Shifts of KI in NMF.

Concentration(M) 6(ppm) Linewidth(Hz)
0.0258 -2.72 51
0.0559 -2.65 52
0.1567 -2.39 52
0.2220 -2.31 54
0.3642 -1.89 56
0.4354 -l.52 60
0.6183 -1.27 61
0.8066 -0.91 63

Table 25. Potassium-39 Chemical Shifts of KSCN in NMF.

Concentration(M) 5(ppm) Linewidth(Hz)
0.0356 -2.74 52
0.0784 -2.73 53
0.1235 -2.69. 55
0.1642 -2.62 57
0.2498 -2.55 72
0.5262 -2.33 78
0.7500 -2.13 82
1.1012 -1.77 87

65

Table 26. Potassium-39 Chemical Shifts of KNo3 in NMF.

Concentration(M) 5(ppm) Linewidth(Hz)
0.0498 -2.78 52
0.1883 -2.80 55
0.2569 -2.85 67
0.4973 -2.91 74

Table 27. Cesium-133 Chemical Shifts of CsF in NMF.

Concentration(M) Mole Fraction 5(10.08ppm)
0.0296 0.00175 -1.76
0.0428 0.00253 -1.68
0.0629 0.00371 -1.64
0.1001 0.00589 -1.39
0.1294 0.00760 -1.27
0.1478 0.00867 -1.14

~0.1636 0.00959 -l.02
0.1906 0.01115 -0.98
0.2344 0.01368 -0.69
0.2525 0.01472 -0.57

66

Table 28. Cesium-133 Chemical Shifts of CsCl in NMF.

Concentration(M) Mole Fraction. 6(ifl.08ppm)
0.0006 0.00004 -1.85
0.0131 0.00077 -1.54
0.0306 0.00181 -1.43
0.0463 0.00273 -0.81
0.1155 0.00679 0.43
0.2616 0.01524 3.22
0.2933 0.01706 3.98
0.3431 0.01990 4.96
0.4214 0.02433 6.52

Table 29. Cesium-133 Chemical Shifts of CsBr in NMF.

Concentration(M) Mole Fraction 5 (i0.08ppm)
0.0038 0.00022 -l.69
I '0.0118 0.00070 -1.43
0.0219 0.00129 -1.12
0.0465 0.00274 -0.50
0.0980 0.00577 1.15
0.1450 0.00851 2.50
0.2213 0.01293 4.94

67

Table 30. Cesium-133 Chemical Shifts of C51 in NMF.

Concentration(M) Mole Fraction 6(10.08ppm)
0.0062 0.00037 -1.64
0.0133 0.00079 -1.43
0.0204 0.00121 -1.23
0.0296 0.00175 -0.61
0.0778 0.00458 0.84
0.1316 0.00773 2.91
0.1555 0.00912 3.74
0.1751 0.01025 4.36
0.2354 0.01374 6.80
0.2434 0.01421 6.84

Table 31. Cesium-133 Chemical Shifts of CsSCN in NMF.

Concentration(M) b Mole Fraction 6(10.08ppm)

. -0.0149 0.00088 -1.54
0.0199 0.00117 -1.54
0.0455 0.00275 ‘1.33
0.0571 0.00337 ‘1.23
0.0903 0.00531 '0.81
0.1605 0.00941 '0.09
0.2443 0.01419 0.74

Table 32.

Concentration(M)

0.0039

0.0157

0.0411

0.1035

0.1506

68

Cesium-133 Chemical Shifts

Mole Fraction

0.00023

0.00093

0.00243

0.00609

0.00883

of CsClO4 in NMF.

6(:0.08ppm)

-1.85

-1.86

-1.90

-1095

-2.05

Table 33. Cesium-133 Chemical Shifts of CsNO3 in NMF.

Concentration(M) Mole Fraction 6(10.08ppm)
0.0092 0.00054 -1.85
0.0180 0.00106 -1.80
0.0398 0.00235 -1.74
0.0611 0.00360 -1.69
0.1119 0.00658 -1.54

°0.1809 0.01059 -1.38

69

Table 34. Cesium-133 Chemical Shifts of CsTPB in NMF.

Concentration(M) Mole Fraction 6 (10.08ppm)
0.0043 0.00025 -1.95
0.0056 0.00033 -2.05
0.0113 0.00067 -2.47
0.0361 0.00213 -3.81
0.0749 0.00441 -6.19
0.0940 0.00553 -7.32

Table 35. Cesium-133 Chemical Shifts of Cs CO in NMF.

2 3
Concentration(M) Mole Fraction 5(10.08ppm)
0.0054 0.00032 -1.85
0.0140 0.00083 -1.74
0.0241 0.00142 -1.64
0.0344 0.00203 -1.33
‘0.0490 0.00289 -1.23
0.0664 0.00391 -1.02
0.0832 0.00490 -0.81
0.0953 0.00561 -0.69
0.1361 0.00799 -0.19
0.1799 0.01053 0.32
0.2105 0.01230 0.84

70

Table 36. Thallium(I)-205 Chemical Shifts of TlClO4 in NMF.

Concentration(M) 6(:0.03ppm)
0.0041 159.58
0.0072 159.03
0.0184 158.13
0.0402 156.45
0.0868 153.10
0.1432 149.82
0.2252 145.46
0.3045 141.82
0.3697 139.90
0.4728 136.41

Table 37. Thallium(I)-205 Chemical Shifts of TlNO3 in NMF.

Concentration(M) 5(i0.03ppm)
0.0044 160.48
0.0155 160.00
0.0284 159.42
0.0607 158.13
0.1056 156.61
0.1591 155.42
0.2093 154.77

71

Table 38. Parameters of Linear Regression for Pure Sodium

Sodium-23

Concentration

Shift (ppm/M)

Intercept

(PPm)

and Potassium Salts in NMF (Concentration vs.

and Potassium-39 Chemical Shifts).

Correlation

Coefficient

NaTPB
NaPi
NaSCN
KI

KSCN

-4.69

-4068

-4.67

-4069

-4.70

-4.69

-2.77

-2.79

-2.76

0.8835

0.9887

0.9907

-0.9580

0.9547

0.9915

0.9986

-0.9768

72
Table 39. Parameters of Linear Regression for Pure Cesium
Salts in NMF (Concentration vs. Cesium-133

Chemical Shifts).

Concentration Intercept Correlation
Salt Shift (ppm/M) (ppm) Coefficient
CsF 5.31 -1.933 0.9970
CsCl 19.83 -1.864 0.9996
CsBr 30.29 -l.824 0.9997
CsI 36.15 -1.877 0.9995
C82C03 12.61 -1.873 0.9969
CsSCN 10.38 -1.782 0.9994
CsNO3 2.67 -1.853 0.9971
CsClO4 -1.34 -1.836 0.9840
CsTPB -59.50 -1.722 0.9998

73

E
Q.
3 0—
if
U)
E
.9
E-l—
m
.c
Q
:E .
g

..2_

     

UCI
Kl

KSCN
KN03

I I I
(10 C13 C16 (19 1.2

ODDO

 

 

Concentration(M)

Figure 8. Plot of Metal (Li-7 and K-39) Chemical Shifts

vs. Concentration of Metal Salts in NMF.

74

-4.3

Sodium—23 Chemical Shift(ppm)

 

     

' 0.0 0.3 0.5 0.9 1.2

Concentration(M)

Figure 9. Plot of Sodium—23 Chemical Shifts vs.

Concentration of Sodium Salts in NMF.

75

 

 

8—-
E 51
o.
3
i3
00
6 4-
.9
E
o
.c
L)
ro 2—
pf)
|
E
.2
8 09
C) o 0 cs
0 A CsBr
_ D CsCl
o O CSF
"2 I I I I 1
Oi) 0.1 0J2 On3 C14 C15

Concenhofion(M)

Figure 10. Plot of Cesium-133 Chemical Shifts vs.

Concentration of Cesium Salts in NMF.

76

 

 

O-

E

Q.

:3 .

E -2-

(fl

'5

.9

E

m

.c

L) —4fi

I"?

PI)

|

E

3

'6 -6-

m x CsZC03

L) o CsSCN
A CsN03
Cl CsC|O4
o CsTPB

‘—8 I I I
OI) 0.1 C12 013

Concentration(M)

Figure 11. Plot of Cesium-133 Chemical Shifts vs.

Concentration of Cesium Salts in NMF.

77

 

 

170-1
E
Q.
3 1609.
:E
(J)
‘5
.9 °
E
CD 150‘ o
.c
C)
03
O 0
OJ
l
E
.2 140-
'5
.c
F— o
A—A THWD3
o TKNO4
130 , 121

I I
01) 0.1 C12 2 C13 CI4 C15

Concentration(M)

Figure 12. Plot of Thallium-205 Chemical Shifts vs.

Concentration of Thallium Salts in NMF.

78

I, and the results are 2.6 i 0.4 M-1

1

for TlNo3 and 1.7 i.
0.5 M' for TlClO4.

The concentration dependences of Cs-133 chemical shifts
of cesium iodide at three temperatures (273K, 296K and
341K) were examined. The results are given in Table 40. For
all three temperatures, the cesium-133 chemical shifts also
change linearly (Figure 13). The slopes, intercepts and
correlation coefficients are shown in Table 41.

2. Qiscussgan
2.1 Infinite Dilution Chemica; Shigtg

Infinite dilution chemical shifts for alkali metal and
thallium ions in NMF are given in Table 42 along with those
obtained in a variety of other solvents. This result
provides the starting point for discussion of the factors
which may determine salvation-induced chemical shifts.
According to expectation, the range of infinite dilution
chemical shifts of alkali metal ions increases with the ion
size, being (for common solvents) ca. 5-6 ppm for Li+, ca.
15-20 ppm for Na+, ca. 30-40 ppm for K+, and ca. 130 ppm
for Cs+. Bloor and Kidd (15) provided good arguments that
Na+-23i ~shielding differences between solvents are
predominantly due to the paramagnetic term, and this
certainly applies for larger cations as well. According to
the overlap model of Chemical shifts, the overlap of the
ion's outer p orbitals with outer orbitals of neighboring
solvent molecules should determine the infinite dilution

chemical shift. Attempts have been made to correlate the

79
Table 40. Cesium-133 Chemical Shifts of CsI in NMF at 273K,

296K and 341K.

Cesium-133 Chemical Shifts (10.08 ppm)

Concentration (M) 273K 296K 341K
0.0062 -5.35 -1.64 2.95
0.0133 -4.92 -1.43 3.28
0.0204 -4.76 -l.23 3.50
0.0296 -4.30 -0.61 3.87
0.0778 -2.70 0.84 5.76
0.1316 -0.79 2.91 7.80
0.1555 0.04 3.74 8.80
0.1751 0.45 4.36 9.57
0.2354 2.73 6.80 12.06

Table 41. Parameters of Linear Regression for CsI in NMF at

5

Three Different Temperatures.

Concentration Intercept Correlation

Temperature Shift (ppm/M) (ppm) Coefficient
273K 34.59 -5.43 0.9994
296K 36.38 -1.89 0.9994

341K 39.42 2.70 0.9999

80

 

 

14—
E 10—
Q.
3
:5 -
U7
6 6-
.9
E
m -
.c
C)
r0 2%
P0
I '1
E
.2
8’3 —2-
C)
_ A 341K
0 296K
6 0 273K
‘ l I I l ”i
01) 0.1 C12 C13 C14 C15

Concentration(M)

Figure 13. Plot of Cesium-133 Chemical Shifts vs.
Concentration of Cal in NMF at Three

Temperatures.

81

Table 42. Infinite Dilution Chemical Shifts (ppm) Relative

to Water for the Cations 7Li+, 23Na+, 39K+,

133Cs+ and 205T1+ in a Variety of Solvents.

Solvent D.N.* Infinite Dilution Chemical Shift(ppm)

Li+ Na+ K+ Cs+ T1+

MeNO2 2.7 -0.36 -15.6 -21.1 -59.8 ----

MeCN 14.1 -2.80 -7 -0.41 32.0 ----
DMSO 29.8 -1.01 -0.11 7.77 68.0 359.7
PC 15.1 -0.61 -9.4 -11.48 -35.2 ----.

MeOH 19 -0.54 -3.8 -10.05‘ -45.2 ----
DMF 26.6 0.45 -5.0 -2.77 -0.5 146.8

AC 17.0 1.34 -8.4 -10.48 -26.8 ----
Py 33.1 2.54 1.35 0.82 -31 783.2

Water 18 0 0 0 0 0
NMF 25 1.02 -4.69 -2.77 -1.84 160.8

* Gutmann's donor number
Except for NMF data, all data are taken from references

140 and 150.

82
infinite dilution chemical shifts with quantities
expressing the electron-donating ability of the solvent. In
our group, we have considered the Gutmann solvent donor
numbers, which were mentioned in the introduction. There is
a rather good linear relation between Na-23 infinite
dilution chemical shift and solvent donicity (Figure 14)
(21). The correlations are also relatively good for K-39
(138), Cs-133 (139) and Tl-205 (140) (Figures 15,16 and
17). No correlation was found in the case of Li-7 (38,141),
because for lithium-7 nucleus the diamagnetic and
paramagnetic terms are of the same order of magnitude. The
donor number of NMF, which was not determined by Gutmann,
can be obtained from Figures 14-17 and they are 24.3, 24,
26.2 and 24 respectively. These numbers are in very good
agreement with each other and show that NMF is a solvent
with medium solvating ability.
2.2 o oentrat d Anion Do a nos 0 c 8

2-2'1 LEW

Li-7 chemical shifts in NMF solution are virtually
concentration and counter-ion independent, presumably due
to a low probability of direct. ionic encounter --- an
impenetrable solvation shell. It is reasonable to say that
there is no contact ion pairs formed for lithium salts,
because the chemical shifts of a cation in a contact ion
Pair (nearest-neighbor interaction) should be different
from its value in the free ion or in a solvent-separated

1011 pair. However, the NMR method cannot distinguish

83

 

 

'3 \
FN‘
Ur- \
n»- \
\
Ul- \
\
"I. \ ‘.
m- ‘ 3'
‘ a
a " "‘
5 " \
a ’b \
.- \
t " \‘i
a 3!- \.
..-.--..--..--..‘
4|- .\
D \
3'- . \
2h- \
3 ' ' 70“
l C ' \
0- ‘0‘ ' I“
-l- I \\
I
-2. ' a‘
-3 L P L l_ 1 L g 1 I 1
O 4 I 12 H N 24 a 33 1‘ O
comm

Figure 14. Sodium-23 Chemical Shifts vs. Gutmann Donor
Numbers. Reference - Aq. Na+ at Infinite
Dilution. Taken From Reference 21.
1. NM, 2. MeCN, 3. AC, 4. Ethyl Acetate, 5. THF,

6. DMF, 7. DMSO, 8. Py, 9. HMPA, 10. H20.

84

 

‘hnl
o 3 3

O

0
b

. 15“.. OM... CM“! Sam
3 0 h 6 a

 

 

 

 

Figure 15. Potassium-39 Chemical Shifts vs. Gutmann Donor

Numbers. Reference - Ag. K+ at Infinite

Dilution. Taken from Reference 138.

85

 

 

 

 

 

so ‘ufinoz
#- “
~‘ .MoOH
4O " x
6“”
- ~‘HFOf
M0
20 " ‘\2
E :- “~“ FG'NHZ
a Q
3 O .......-....-.--;‘.DMF .H20
«0 I
I I \‘
-20 t I \
n \‘
’ omen : OP9~
-40 I- .
p I
I
‘5° ’ 2 DMSO
1 1 L 1 1 1 1 #1 1
2 9 l4 20 26 32 36

DONOR NUMBER

Figure 16. Cesium-133 Chemical Shifts vs. Gutmann Donor
Numbers. Reference - Aq. Cs+ at Infinite

Dilution. Taken from Reference 139.

86

 

       

 

I I I r I r r ‘ I 1 I
2000’
b .0 '1
y- d
l500" -
1- .1
1000- ‘
a P ‘
l " d
a .. .1
\ 500— 4
.- I- d
h _
_ I- 11
: E---------z-§ 94 :
I
.4 0 ’ ' 7
‘ I- I "
2 p I «I
a - : -
III I- q
z __ ' b
o -800 I
I- ' "
I- . 4
r l ‘
I- I d
I
“000 1 11 1 1 11 1 1, 1 1 l 1 7

 

 

20 30 40 IO
DONOR nausea

o
3

Figure 17. Thallium-205 Chemical Shifts vs. Gutmann Donor
Numbers. Reference - Ag. Tl+ at Infinite
Dilution. Taken from Reference 140.

1. H O, 2. FA, 3. NMF, 4. DMF, 5. DMSO, 6. DNA,

2
7. HMPA, 8. Py, 9. n-Butylamine.

87

between free ions and solvent—separated ion pairs.
Weingartner's nuclear magnetic relaxation results (125)
showed that LiCl- can form solvent-separated ion pairs in
NMF and Perelygin’s infrared results (123) also showed the
possibility of the formation of the solvent-separated ion
pairs. Thus, free cations and solvent-separated ion pairs
are the likely components for lithium salts in NMF. The
conclusion does not conflict with the conductance results
(115), which showed that ion-association occurs for lithium
salts in NMF, because conductance measurements cannot
distinguish between contact and solvent-separated ion
pairs, whereas NMR is sensitive to the formation of contact
ion pairs.

2.2-2 godium. gotassium and Cesium gals:

For the sodium, potassium and cesium salts studied, the
chemical shifts change linearly with concentration and also
depend on anions. The order of increasing shielding shift

for cesium salts is:

2-
3

< 3' < N03- < c104" < TPB'.

This sequence is the same as those in the KI/HZO (l6) and

I- < Br- < c1" < do < SCN-

Rb+/K20 (17) systems. Bloor and Kidd related this sequence
to the anion basicity and with the nephelauxetic series
(163). In the cases of tetraphenylborate, picrate and
perchlorate salts, increase in the salt concentration
results in an upfield shift of the metal resonance. Whereas

in the cases of halide, thiocyanate, carbonate and nitrate

sa
co
te
so
de

th

ac

E10

an
do
wh
si.

we

po
RU
i0:
mi:
pm
of

are
in

SUC

88

salts, a downfield shift of the metal resonance is found as
concentration increases. For the bulky, Symmetric
tetraphenylborate and perchlorate anions, replacement of a
solvent molecule by the anion decreases the electron
density around the cation, resulting in upfield shifts of
the metal resonance. On the other hand, other anions
(except the nitrate ion in the case of potassium nitrate)
act as better electron donors than the displaced solvent
molecule, and the metal resonance shifts downfield. Akitt
(142) suggested that the upfield shift caused by oxyanions
and tetrafluoroborate in water arises because these anions
do not overlap the cation’s orbitals, but‘displace water,
which does have an overlap contribution. This is unlikely,
since in NMF downfield shifts for carbonate and nitrate
were observed.

Then questions arise: what kind of role do the anions
play? Perhaps the most obvious possible mechanism is the
polarization of the electronic environment of the magnetic
nucleus by the strong electrostatic field produced by other
ions in its vicinity. If this were an important factor we
might expect the smaller, more strongly polarizing ions to
produce the largest downfield shifts, so that the shielding
of the metal nuclei would show the following anion
dependence Cl- < Br- < 1'. However, the observed effects
are quite the reverse, and it seems unlikely that changes
in the strength of solvation of the ions could account for

such a reversal . Electrostatic effects are also

89

unsatisfactory explanations since they are not able to
explain the observed upfield shifts of metal resonance in
solutions of some nitrate, picrate and tetraphenylborate
salts. The interaction must be of another kind.

What causes this linear character? Do the salts form
contact ion pairs? It is interesting to investigate the
possibility in detail by calculating the ion pair formation

constant from the equations in Appendix I:

a = [-1 + (1 + 4K¥§C>V21<af - aipmzx vie)

+ ‘ip' (3.1)
-1og V2: [(4.198x106) z+z_ 11/2/(DT)3/2] /
[1 + (5.029x109aIl/2)/(DT)1/2]. (3.2)

Different values of the ion pair formation constant and the
chemical shift difference between the free ion and contact
ion pair were substituted into equation (3.1) for
concentrations up to 1.0 M. Figure 18 shows that, at a

fixed value of (afree the lower the ion pair

- “ion pair”
formation constant is, the straighter is the line in the

chemical shift vs. concentration plot. Figure 19 and Table

43 show that the value of (I ) determines

free - 6ion pair
the magnitude of the concentration shift (slope) if the

plot is a straight line. Experimentally, the concentration

 

9O

 

 

10-—
E 8-
Q.
E;
if
U)
6 6~
.9
E
m
.c
L)
m 4—
N7
7‘ K=1
E
.2
8 2—
C)
K=OD1
O I I l I I I I I l j
013 0.2 0:4 0.6 0.8 1.0
Concentration(M)
Figure 18. Plot of Cesium-133 Chemical Shifts vs.

Concentration of CsI in NMF at Four
Different K (Ion Pair Formation Constant)

Values.

91

 

 

 

10—
0:1000
E 3.—
Cl
ii
E
(f)
"5 6-
.9
E
G)
.C
L)
N) 4—
pr)
l
E
.2
8 2—
L)
0:100
0 0:10
1 I I I I I I I I ’7
0.0 0.2 0.4 0.6 0.8 1.0

Concenhofion(M)

Figure 19. Plot'of Cesium-133 Chemical Shifts vs.
Concentration of C31 in NMF at Three
Different D (Chemical Shift Difference
between Free Cesium Ions and Ion Pairs)

Values.

92
Table 43. Parameters of Linear Regression for Cesium Iodide
in NMF with Different Values of the Ion Pair
Formation Constant (K) and Cs-133 Chemical Shift
Difference between Free Cesium Ions and Ion

Pairs (A6).

Concentration Correlation

K(M-1) 55(ppm) Shift (ppm/M) Coefficient
1 10 3.66 0.9798
0.1 10 0.85 0.9992
0.01 10 0.10 0.9999
0.01 100 0.98 1.0000
0.01 1000 9.82 1.0000

93
shift is 36.15 ppm/M and the correlation coefficient is
0.9995 for CsI solutions. Comparison of the experimental
value with the calculated values in Table 43 suggests that
the ion pair formation constant for the cesium iodide in

1 and 0.01 M’l. Then the

the NMF system is between 0.1 M-
chemical shift difference between the free cesium ion and
cesium iodide contact ion pair can be estimated,” by the
ratio between the concentration shift and AI . According to
the calculation, the chemical shift difference could be
as high as 425-3681 ppm. It is hard to imagine that there
exists such a big chemical shift difference between
free cesium ions and contact ion pairs. So formations of
contact ion pairs could not be the explanations to this
linear character. Are solvent-separated ion pairs formed?
If so, the cation is directly surrounded only by NMF
molecules. The shift then reflects a change in the
interaction between the cation and NMF as the salt
concentration increases. The shift would be long range,
which is analogous to proton chemical shifts in organic
compounds. Since the shift would have to be transmitted
through at least three bonds, it should be fairly small.
Thus, it is concluded that thermodynamic ion pairs are not

*This calculation is analogous to the equation proposed by

Dye (167):

“IKCVI A6 +6free. (3.3)

94
formed in these systems.

Deverell and Richards (18,19,28,143) developed a theory
for the chemical shift in aqueous solution from that for
the chemical shift in alkali halide crystals (144-146), and
postulated that the shift depends on overlap contributions
from solvent molecules, the anions, and the cations and
that the measured quantity is the average of all
possibilities in the molecular ensemble. Overlap requires
direct ion-ion contact during which anions continually
displace solvent molecules to form 'short lived contact ion
pairs’ (collisional ion pairs). This theory seems to
explain well the NMF results also. Detailed discussions
will be given in section 2.3.

2.2-3 Thallium Salts

Thallium nitrate and perchlorate form contact ion pairs
in NMF, while alkali metal salts do not. The non-linear
plot associated with the thallium ion is symptomatic of an
equilibrium. This sort of change would give a curved plot.
The anion must displace the solvent and remains associated
with the cation solvation sphere in some way and changes
the attitude of the solvating molecules. However, thallium
nitrate and perchlorate form collisional ion pairs in water
(the thallium-205 chemical shift vs. concentration plot is
linear (164)). In NMF the ion-ion energy is much larger
than the ion-dipole energy, and thermodynamic ion pairs
could be formed. In water the thallium ion is more highly

solvated, and the ion-dipole energy could be as

95

large as ion-ion energy. Therefore water could compete with
anions to coordinate with thallium cations, thus
discriminating against thermodynamic ion pairs.

2.2-4 W

The linear relationship between the observed cesium-133
chemical shifts and the salt concentrations holds,
regardless of the solution temperature in the range from
273K to 341K. The mechanism responsible for the changes of
the chemical shifts does not change in this temperature
range. Table 41 shows that the higher the temperature, the
higher is the infinite dilution chemical shift and the
concentration shift. This result indicates that as
temperature increases, the NMF molecules become less
structured and the dielectric constant decreases, which
leads to more ion-solvent (or ion-ion) interactions, larger
overlap between cations and solvent molecules (or anions),
and results in higher infinite dilution chemical shift (or
higher concentration shift).

2.2-5 Chgmicai shift as 5 Function of 321; Igggtiog of

WW

From Figures 20 and 21 and Table 44, it is apparent

that the cesium—133 chemical shift I is proportional to the

mole fraction X of the cesium salts:

5 = X50! (3.4)

where 50 is the mole fraction shift of the salt and its

96

8—-

E 5..

Q.

8

E o

(f)

'5 4‘

.2

E 1

m

.c

L)

m 2—

N7

|

E

.2

8 0—

C) .- 0 CsI

. . D CsBr
;J. A CsCl

._2 . O CsF

 

 

I I I I ‘—7
.0000 0.005 0.010 0.015 0.020 0.025

Mole Fraction of Cesium Salts

Figure 20. Plot of Cesium-133 Chemical Shifts vs. Mole

Fraction of Cesium Salts in NMF.

97

 

 

2-

E 0..

a.

3 .

$2 . /*”’

.C 0

U1 // ,

E “2“ “U : km

.2

E

m

_C

L)

m —4—

N3

1

E

3

'6

Q)__5_ o C52C03

0 .1 CsSCN
A CSNO3
D CsC|04

-—8 I a CsTPB

 

I I I I
0.000 0.003 0.006 0.009 0.012 0.015

Male Fraction of Cesium Salts

Figure 21. Plot of Cesium-133 Chemical Shifts vs. Mole

Fraction of Cesium Salts in NMF.

98
Table 44. Parameters of Linear Regression for Pure Cesium
Salts in NMF (Mole Fraction vs. Cesium-133

Chemical Shifts).

Mole Fraction Intercept Correlation
Salt Shift (ppm) (ppm) Coefficient
CsF 91.31 -1.937 0.9970
CBCl 342.89 -1.881 0.9995
CsBr 518.05 -1.832 0.9997
CsI 619.51 -1.890 0.9995
C82C03 216.71 -1.887 0.9979
CsSCN 178.75 -1.789 0.9994
CSNO3 45.60 -1.853 . 0.9973
CsClO4 -22.79 -1.836 0.9836
CsTPB -1011.29 -1.720 0.9998

99
value ’should' be equal to the chemical shift of the
crystalline salt. However, for the cesium halide salts
(except CsF) a o is far away from the chemical shift of the
crystalline salt (Table 45). Thus a proportionality

constant m is added:
I =mX61, (3.5)

where 51 is the chemical shift of the crystalline salt. In
a plot of 5 / 31 as a function of X (Figure 22) the
experimental points for all four halides fall around four
straight lines. Table 45 shows that the m values for cesium
salts increase with the size of anions and are in between
1.01-2.18. The m value is equal to 1.01 for cesium
fluoride. This might imply that cesium fluoride has
different mechanism leading to the changes of the chemical
shifts from those of other cesium halide salts. Fluoride
ion might be able to penetrate into the solvation shell
very easily to interact. with cations without any
interference from the NMF molecules: to mimic the
crystalline cesium fluoride salt. The m value may be
regarded as a measure of the effectiveness of NMF molecules
in excluding anions from the cations.

2-3 §h1n1£§1_§hili_1h£2£2

To date, no theoretical treatment of chemical shifts
has been generally accepted. However, there are some

theories explaining the origin of the chemical shift in the

100
Table 45. Comparisons of Mole Fraction Shifts (do) with
Crystalline Chemical Shifts (51) for Cesium
Halides and Proportionality Constants (m) in

Equation 3.5.

Salt 6 0(ppm) 6 1 (ppm) m

CsF 91.31 9oa 1.01
CsCl 342.89 232b 1.48
CsBr 518.05 258c 2.01
CsI 619.51 284b 2.18
CsSCN 178.75 109b 1.64

a: from reference 144.
b: from reference 165.

c: from reference 166.

101

{1050—
a Csl
a CsBr
0.0409 A C89
0 CsF
0.0302

0.020—

o.\°'

0.010—

0.000—

—.010

 

-._020 1 I
0.000 0.006

 

I I I

 

I I
0.018 0.024

f.
0.012

lvlale Fraction of Cesium Salts

Figure 22. Plot of a / I 1 vs. Mole Fraction of Cesium
Salts in NMF.

102

water system in physical and chemical terms. A brief
introduction to each theory will be made, followed by
discussions and- comparisons between the water and NMF
systems.

2.3-1 Qvegigg ngon

Deverell and Richards (17,143) studied the effect of
concentration and counter ion on both the magnetically
active alkali metals and the halide ions. For the cation
NMR, they noted that the downfield shift produced by the
counter ions was in the same order as that produced in the
crystalline state. The Kondo-Yamashita (144) approach
yields the following equation for the shielding constant

for an ion in a cubic crystal:
_ 3 2
a - K<1/R >(1/A)(:s ), (3.6)

where K is a constant, <1/R3> is the expectation value
taken over all the outer p electrons of the central ion, A

2 is the sum of the

is the average excitation energy and SS
squares of appropriate overlap integrals of the p orbitals
of the central atom with s orbitals and with p orbitals in
both a and a: bonding configurations of the surrounding
ions. Deverell and Richards used the Kondo-Yamashita

formalism and obtained the chemical shift of an ion in

solution as the following:

6 ' K<1/R3>(1/A) [Qion-ionm) + ¢ion-water(c)

103

‘4’ (0)]: (3.7)

ion-water

where °ion- (C) and m ion-water(c) are the appropriate

ion
sums of overlap integrals with all surrounding ions and
water molecules at a solute concentration C, and “’ion-
waterm) is the sum at infinite dilution. They concluded
from the observed linearity of the chemical shift with
solute activity that the first term, 41 ion-ion‘c)’ being
dependent on ion-ion collisions, is predominant. Since °ion-
ion‘c) and °ion-water‘c) are proportional to the activity
(or concentration), the slope (activity shift or
concentration shift) of the straight line in the chemical
shift vs. activity plot is equal to K<1/R3>(l/A) and the
intercept. his -K<1/R3>(l/A)¢ion_water(0). ¢1°n_water(0)
cannot be obtained, so experimentally the intercept
(chemical shift of the metal ion in infinitely dilute
aqueous solution) was assumed to be equal to zero.

In the NMF system, the same trend of the results was
obtained as that in water, and some discussion follows:

l)Because of the high value of the dielectric constant,
the mean-ionic activity coefficients of various metal salts
in NMF solution are all almost close to unity. Thus, in the
NMF system the concentration shift is the same as the
activity shift. There is a linear relationship between the
concentration (or activity) and the chemical shift. This is

different from the water system, in which the linear

relationship was found only when activity was used. There

104

are two reasons for this. One is that the dielectric
constant of NMF is much higher than that of water. The
other is that a different concentration range was used in
these two systems. For the water system the concentration
went up as high as 10 M; in NMF the highest concentration
used was around 1 M.

2)The cesium cation has the largest concentration shift
in the alkali metal ion series, and Li+ has no
concentration shift. This could be due to the strong
solvation of the lithium ion, which reduces considerably
the probability of ion-ion contact; on the other hand
cesium ion is the least solvated, which leads to the
largest overlap. For halide salts, iodide has the largest
concentration shift because of its largest overlap with the
metal ions.

3)For the NMF system, equation (3.7) can be modified

into:
- 3 b
6 "K<1/R >NMF(1AA)NMF[°ion-ion(c) +‘9ion-NMF(C)

3 .
-‘pion;NMF(o)] - K<1/R >water(IAA)water¢ion-water(°)(3'8)

In the same way, the intercept (infinite dilution chemical

shift) is equal to

K<1/R3>NMF ( VA) NMFq’ion-NMF (0)

105

- K<1/R3> (l/A)

water water°ion-water(o)'

and the slope (concentration shift) is equal to

3
K<1/R >NMF(1/A)NMF'
Table 42 shows that sodium-23, potassium-39 and cesium-133
ions have negative values for the infinite dilution
chemical shift. In these cases, it might be assumed that
. 3
°ion-water(°) 15 larger than °ion-NMF(°) if <1/R > and A

are approximately the same for water and NMF systems.
2.3-2 zreeabilisz_1999r1
Covington and Newman (147) wrote a more realistic

equation than equation (3.8):

5(C) = K<l/R3>(1/‘A)Wion-ion - °ion-solvent] x P(C)' (3‘9)

where 0 10 is the change of overlap when a solvent

n-solvent
molecule is removed from the close vicinity of an ion, ¢ion-
ion is the change in overlap when an ion is placed in the
hole left by the solvent molecule, and p is the probability
that such an encounter takes place. It is possible that
high probability will affect the curvature of the chemical
shift vs. concentration plot, whereas large changes in
overlap integral will affect the shift magnitude.

From equation (3.9), the concentration shift should be

equal t° K<1/R3>(1/A)k if ”ion-ion“) -°ion-solvent(cn

106

x P(C) = k. x (concentration). Then three variables will be
solvent dependent -A , <R3> and k. It might be reasonable
3

to assume that A and <R > are at approximately same order
of magnitude for the same metal salts. Then for cesium
salts, k(NMF) must be larger than k(water) because the
concentration shift in the NMF system is always larger than
that in the water system (Table 46). Of course, k is the
combination of many factors, such as solvation energy, the
closest distance to which the ions can approach, dielectric
constant, reorientation of solvent molecules, strength of
hydrogen-bond, and solvation number. Not one of them is
predominant. For example, dielectric constant is not the
major factor leading to the linear change in the chemical
shift. If dielectric constant was the major factor, then
the concentration shift should be lower for those solvent:
with higher dielectric constant values because of the lower
probability for ion-ion encounter. However, Tables 47 and
48 show that, generally the higher the dielectric constant
is, the higher the concentration shift is for cesium iodide
and cesium tetraphenylborate in various solvents. For
solvents, with high dielectric constants, the presumption
that a dielectric continuum can represent the solvent is no
longer tenable. The molecular nature of the solvent and
specific ion-solvent interactions must be taken into
account. The region in space where dielectric saturation
takes place is an appreciable fraction of the space within

which ion-ion interactions are considered to happen.

107

Table 46. Concentration Shifts of Cesium Salts in Water and

NMF.

Concentration Shift (ppm/M)
Salt Water* NMF
CSCl 5.03 19.83
CsBr 11.15 30.29
CSI 19.64 36.15
CSN03 “2.74 2.67

* Calculated from the data in reference 18.

108
Table 47. Concentration Shifts of CsTPB in DMF, DMSO, EC

and NMF.

Dielectric Gutmann Concentration Correlation

Solvent Constant D.N. Shift (ppm/M) Coefficient
*

DMF 36.7 26.6 -19.54 -0.9977

DMSO* 45.0 29.8 -11.32 -0.9925

8c** 89.1 16.4 -3s.4o -0.9980

NMF 182.4 25 -59.50 -0.9998

Table 48. Concentration Shifts of C51 in MeCN, H 0, EC

2
and NMF.

Dielectric Gutmann Concentration Correlation

Solvent Constant D.N. Shift(ppm/M) Coefficient

MeCN* 36.0 14.1 1.07 0.9958
***

H20 78.6 18 19.64 0.9947

sc** 89.1 16.4 38.78 0.9979

NMF . 182.4 25 36.67 0.9997

* Lu Lu L. Hsu, Master Thesis, Michigan State University
(1976).

** Lee-Lin Soong, Research Report, Summer (1985) and
Winter (1986).

*** Reference 18.

109

However, the presence of another ion strongly affects the
electric field in the region between them. It is then
extremely difficult to estimate the effective dielectric
constant in the non-spherically symmetric region in between
the ions. Non-electrostatic effects: donor-acceptor
interactions, hydrogen bonding, etc., are superimposed on
the Coulombic interactions. Under such conditions the
Bjerrum (148) approach must be abandoned.

2.3-3 5, g. 8. Theory

In another paper, Richards and co-workers (18) provided

the formula:

5 = P(CHJI '1' 62 + 53]. (3-10)

where 51, 5 2 and a 3 are the shift contributions due to
ionic encounter, solvent displacement and solvent
reorientation respectively. P(C) is the probability of
ionic encounter. As before, they neglected the latter two
contributions. Then they used Debye-H’ticker potential
theory, modified by adding an oscillating potential to take
into account the particular nature of the solvent and the
tendency for solvent molecules to form well-defined shells
around an ion. Empirically, they found that the shift obeys

the following relationship:

log ‘obs = g x log (concentration) + D, (3.11)

110
where g is a parameter which is anion dependent and is
related to the closest distance to which ions can approach.
D is a constant. It would be interesting to test this
equation with the experimental data from NMF solutions.
There is a problem for those systems for which chemical
shifts are all (or partly) negative values, because it is
not possible to take the logarithm of negative numbers. For
the CsCl, CsBr and C51 / NMF systems, only points with
positive values of the chemical shift were picked. In the
C82C03, 3
negative chemical shift values were converted into positive

CsF, CsSCN, CsClO4, CsNO and CsTPB / NMF systems,
values by multiplying each by -1. The results are listed in
Table 49. It is found that the experimental data for CsCl,
CsBr and C31 / NMF systems have better correlation with
equation (3.11) than those of the other cesium salt/NMF
systems.

2-3-4 W

Mishuatin and Kessler (149) hypothesized that the
significant solvation symmetry effect should make it
possible to perform reliable discriminations between
contact ‘ and solvent-separated ion pairs. However,
Weingartner (125) opposed the significant electronic
contribution to the relaxation process which was suggested
by Mishustin and Kessler, and proposed a cation-anion-
solvent model for the LiCl and NaBr / NMF systems by
analyzing the concentration dependence of 1/T1 in terms of

two effects: 1) a direct contribution to q2 (mean. square

111
Table 49. Linear Regression Parameters for Cesium Salts in

Water and NMF Using H. R. S. Theory's Empirical

Formula.
Correlation
System g value Coefficient
CsCl/NMF 2.173 0.9920
CsBr/NMF 1.785 0.9983
CsI/NMF 1.815 0.9887
CsCl/H20* 0.785 0.9999
CsBr/H20* 0.814 0.9981
CsI/H20* 0.813 0.9985
CsF/NMF -0.469 -0.8957
CsZCO3/NMF -0.537 -0.7874
CsSCN/NMF -0.991 -0.7990
CsClO4/NMF 0.026 0.8780
CsNO3/NMF -0.091 -0.9278
CsTPB/NMF 0.423 0.9850

* take from reference 18.

112

electric field gradient) due to the Coulomb charges of
neighboring ions, and 2) an indirect effect caused by
distortion of the solvent structure around the reference
ions by short duration collisions with counter ions. No
contact ion pairs or triple ion aggregates were found, and
a short-lived structure for the solvent-separated ion pair
was proposed. This result agrees with that from the Li-7
chemical shift measurements, which show that there is no
concentration dependence of the Li-7 chemical shift.
However, in another recent paper (126) Weingértner proposed
that there are various kinds of higher ion aggregates in
the RbI/NMF system and suggested that this could be true
also in the CsI/NMF system. It is hard to rationlize this
predication with the results of the cesium-133 chemical
shift measurements.

3- 9292111191193

It seems that there is no single theory which can
satisfactorily explain the origin of the chemical shifts
for all systems. Overlap theory seems the best, but it
requires more complicated and accurate quantum mechanical
calculations, which may be powerless if too many
assumptions are assumed. This is true especially in the
case of NMF, for which the molecular structure is more
complicated than that of water.

Table 50 summarizes some results on ion-association
studies obtained by different techniques. It should be

noted that it is questionable whether these experimental

113
Table 50. Summary of Ion-Association Studies in NMF, Via

Different Techniques.

IR NMR Measurement Conductance

Salt Measurement Chemical Shift Relaxation Measurement
Lithium Free or SSIP Short-lived SSIP Ion Pairs
(9 Salts) (LiCl) (3 Salts)

Sodium SSIP Collisional IP Short-lived SSIP Ion Pairs

(NaI) (9 Salts) (NaBr) (3 Salts)
Potassium Collisional IP Ion Pairs
(3 Salts) (3 Salts)
Rubidium Higher Ion Aggregate Ion Pairs
(RbI) (2 Salts)
Cesium Collisional IP Ion Pairs
(9 Salts) (2 Salts)
Thallium Contact IP
(3 Salts)
Tetraalkyl- Ion Pairs
Ammonium (7 Salts)

SSIP: Solvent-separated Ion Pair. Free: Free Ion.

114

methods yield direct information on first coordination
sphere effects. However, different kinds of ion-ion
interactions have been proposed for many salt/NMF systems.
This shows that even in this high donor, high dielectric
constant solvent (NMF), and even at low salt concentration,
there may be some kind of ion-association. These results
also lead us to think about the phenomena and nomenclature
of ion pairs because there are several ambiguous terms ---
contact ion pair, solvent-separated ion pair, collisional
ion pair, transient contact ion pair, weak contact ion
pair, short-lived contact ion pair and short-lived solventé
separated ion pair.

A thermodynamic ion pair arises in dilute solutions
when. oppositely charged ions approach one another more
closely than average, approach sufficiently close to affect
the total level of electrostatic interactions. As mentioned
in section 2.3-2, this type of ion pair cannot exist in a
high dielectric constant solvent at low concentration.
Instead, ions collide with other ions within a very short
range and collisional ion pairs could be formed. They can
be detected by the NMR method only, because NMR parameters
are sensitive to very short-range interactions. Thus, the
difference between thermodynamic and collisional ion pairs
is in the time scale of association (150). A collisional
ion pair is formed if the collision between ions lasts for
a 'minimum’ time which is long compared with the general

rate of motion within the radial distribution around a

115
given ion, whereas a thermodynamic ion pair is formed if

there is a very long association time.

C. - O 8 ONE I - B
W
1- 39.81113};

Cesium-133 chemical shifts were measured for eighteen
mixed electrolyte systems including lithium, sodium and
cesium salts (chloride, bromide, thiocyanate, perchlorate,
tetraphenylborate and nitrate) and three iodide salts
(tetrapentylammonium, sodium and potassium iodide) in 0.10
M CsI/NMF solutions. The results are given in Tables 51-61
and Figures 23-26 (Cesium-133 chemical shifts are given
relative to that of 0.10 M CsI). Cesium-133 chemical shifts
also change linearly with the concentration of the salts
added to 0.10 M CsI/NMF solutions for all systems studied.
CsSCN was also mixed with 0.24 M and 0.48 M C31, in
addition to 0.10 M CsI. The results are given in Table 62
and Figure 27 (Cesium-133 chemical shifts are given
relative to those of 0.24 M CsI and 0.48 M CsI) . Again,
cesium-133 chemical shifts change linearly with the
concentrations. Concentration shifts for all systems
(including those for pure cesium salts) are summarized in
Table 63.

2. 218.999.91.911

Covalency, electrostatic polarization and short range

repulsive forces have been previously considered as

possible sources of chemical shifts in the single salt

116
Table 51. Cesium-133 Chemical Shifts of NMF Solutions of

Mixed CsCl/0.10 M CsI and LiC1/0.10 M CsI.

Concentration Concentration
of CsCl(M) 6(10.08ppm) of LiCl(M) 6(i0.08ppm)
0.0273 0.58 0.0401 0.81
0.0603 1.41 0.1050 2.18
0.0879 1.91 0.1239 2.55
0.1319 2.73 0.2029 4.36
0.1550 3.56 0.2300 4.86
0.2593 5.71 0.3055 6.41
0.3940 8.41
0.4647 10.03

Table 52. Cesium-133 Chemical Shifts of NMF Solutions of

Mixed NaCl/O.10 M CsI and NaTPB/0.10 M CsI.

Concentration Concentration

of NaCl(M) 6(30.08ppm) of NaTPB(M) 5(10.08ppm)
0.0402 0.81 0.0237 -1.43
0.1095 2.30 0.0602 -3.67

0.2173 4.73

117
Table 53. Cesium-133 Chemical Shifts of NMF Solutions of

Mixed LiTPB/0.10 M CsI and CsTPB/0.10 M CSI.

Concentration Concentration

of LiTPB(M) 6(10.08ppm) of CsTPB(M) 5(10.08ppm)

0.0175 -0.87 0.0045 -0.14
0.0242 -1.25 0.0157 -0.89
0.0382 -1.93 0.0284 -1.55
0.0478 -2.49 0.0336 -1.88
0.0563 -2.93 0.0457 -2.29
0.0583 -3.00

Table 54. Cesium-133 Chemical Shifts of NMF Solutions of
Mixed LiClO4/0.10 M CsI and NaClO4/O.10 M CsI.
Concentration Concentration

of LiClO4(M) a(:0.08ppm) of NaClO4(M) 5(10.08ppm)

0.0202 -0.09 0.0580 -0.16
0.0494- -0.21 '0.1070 -0.37
0.0729 -0.29 0.1548 -0.52
0.1081 -0.46 0.1768 -0.57
0.1264 -0.53 0.2622 -0.88
0.1504 -0.63 0.3022 -0.99
0.1701 -0.71 0.3430 -1.09
0.1988 -0.83 .

118

Table 55. Cesium-133 Chemical Shifts of NMF Solutions of
Mixed CsClO4/0.10 M CsI and CsNO3/0.1O M CsI.
Concentration Concentration
of CsClO4(M) 6(10.08ppm) of CsN03(M) 5(i0.08ppm)

0.0323 -0.106 0.0205 0.04
0.0575 -0.137 0.0416 0.07
0.0708 -0.189 0.0705 0.12
0.0990 -0.272
0.1270 -0.354

Table 56. Cesium-133 Chemical Shifts of NMF Solutions of

Mixed LiN03/0.10 M CsI and NaNO3/0.10 M CsI.

Concentration. Concentration

of LiN03(M) 6(10.08ppm) of NaNO3(M) 6(10.08ppm)
0.0384 0.00 0.0235 0.02
0.0566. 0.00 0.0465 0.03
0.0899 0.00 0.0700 0.04
0.1175 -0.04 0.1141 0.08
0.1842 -0.04 0.1724 0.13
0.2633 -0.04 0.2188 0.17
0.3119 -0.04 0.3018 0.21

0.3322 -0.04 0.3618 0.25

119

Table 57. Cesium-133 Chemical Shifts of NMF Solutions of
Mixed LiBr/0.10 M CsI and NaBr/0.10 M CsI.

Concentration Concentration

of LiBr(M) 6(:0.08ppm) of NaBr(M) 5(10.08ppm)
0.0115 0.31 0.0165 0.44
0.0691 1.57 0.0544 1.50
0.0835 2.18 0.0729 1.99
0.1186 3.11 0.0972 2.74
0.1307 3.49 0.1176 3.30
0.1480 3.92 0.1399 3.86
0.1624 4.30 0.1613 4.55
0.1773 4.80 0.1987 5.60

Table 58. Cesium-133 Chemical Shifts of NMF Solutions of
Mixed CsBr/0.10 M CsI and LiSCN/0.10 M CsI.

Concentration Concentration

of CsBr(M) 5(10.08ppm) of LiSCN(M) a(¢0.08ppm)
0.0249 0.78 0.0308 0.24
0.0566 1.61 0.0484 0.37
0.0905 2.60 0.0723 0.57
0.1022 3.01 _0.0923 0.74
0.1262- 3.67 0.1584 1.27
0.1433 4.09 0.1707 1.36
0.1835 5.33 0.1992 1.56
0.2129 6.24 0.2407 1.89

120

Table 59. Cesium-133 Chemical Shifts of NMF Solutions of
Mixed NaSCN/0.10 M CsI and CSSCN/0.10 M CsI.

Concentration Concentration
of NaSCN(M) 5(.~I_-0.08ppm) of CsSCN(M) 6 (10.08ppm)

0.0204 0.20 0.0236 0.24
0.0537 0.51 0.0416 0.41
0.1326 1.23 0.0730 0.74
0.3071 2.88 0.1068 1.07
0.4040 3.76 0.1343 1.40
0.5212 4.95 0.1545 1.57
0.6149 5.73 0.1759 1.70

0.2262 2.31

Table 60. Cesium-133 Chemical Shifts of NMF Solutions of
Mixed NaI/0.10 M C31 and KI/0.10 M CsI.

Concentration Concentration

of NaI(M) I5(¢p.08ppm) of KI(M) 6(10.08ppm)
0.0240 0.88 0.0199 0.67
0.0717 2.45 0.0506 1.83
0.0927 3.28 0.0672 2.32
0.1348 4.60 0.0979 3.07
0.1541‘ 5.35 0.1657 5.55
0.1721 6.01 0.1958 6.70
0.1968 6.92

121
Table 61. Cesium-133 Chemical Shifts of NMF Solutions of
Mixed TPAI/0.10 M CsI.

Concentration of TPAI(M) 5 ($0.08ppm)

0.0146 0.49
0.0403 1.42
0.0728 2.39
0.0988 3.39

Table 62. Cesium-133 Chemical Shifts of NMF Solutions of

Mixed CsSCN/0.2434 M CsI* and CsSCN/0.4816 M
**

CsI

Concentration Concentration

of CsSCN(M)* 5(10.08ppm) of CsSCN(M)** I (10.08ppm)
0.0139 0.16 0.0319 0.37
0.0387 0.33 0.0568 0.62
0.0780 0.83 0.1063 1.11
0.0982. 1.07 0.1503 1.53
0.1126 1.16 0.2309 2.44
0.1340 1.40 0.3136 3.26
0.1919 2.06 0.4744 5.00
0.3814 3.97 0.7422 7.81

122

O
l

O‘)
l

l\)
l

.//

L

Cesium—133 Chemical Shift(ppm)

 

'4 A A L A A A A
J Gr u ‘- fl w f ‘3

 

x LiBr
_2 d 0 LiCl
x LiSCN
. . A LJNO3
a LKNO4
o lJTPB
"6 T7 I I I I
C10 0.1 0.2 0.3 004 0.5

Concentration of Lithium sdlts Added
to 0.1 M Csl Solution in NMF

Figure 23. Plot of Cesium-133 Chemical Shifts vs.

Concentrations of Lithium Salts Added to

0.10 M CsI Solutions in NMF.

123

 

 

NaBr
NaCl
NaSCN
NaNO3
NaClO4
NaTPB

I I I ”*1
0.00 0.15 0.30 0.45 0.60

Cesium—133 Chemical Shift(ppm)

OUDXOBK

 

‘—6

 

Concentration of Sodium Salts Added
to 0.1 M Csl Solution in NMF

Figure 24. Plot of Cesium-133 Chemical Shifts vs.
Concentrations of Sodium Salts Added to

0.10 M CsI Solutions in NMF.

124

 

 

8—

A 6‘

E

Q.

O.

E“ 4-

33 0

(fl

.8 2-

E

Q)

.C

O 0‘0 t—r—fi

1“) WM

N7

l

E —2—

.3 X CsBr

(3 0 CsCl
_4_ X CsSCN

A CsNO3
Cl CsC|O4
-6 O CsTPB
I I l I I j

0.0 0.1 0.2 0.3

Concentration of Cesium Salts Added
to 0.1 M Csl Solution in NMF

Figure 25. Plot of Cesium-133 Chemical Shifts vs.
Concentrations of Cesium Salts Added to

0.10 M CsI Solutions in NMF.

125

 

 

 

8—~

E

o.

e- 6—

{E

m ..

B

.9

t 4—

_c

U

1")

Pf) -4

|

E 2—

'5

m

U
1 A NaI/0.1MCsl

a Kl/O.1MCsl
o TPAI/0.1MCsl

0 I I I I
0.00 0.05 0.10 0.15 0.20

Concentration of Iodide Salts Added
to 0.1 M Csl Solution in NMF

Figure 26. Plot of Cesium-133 Chemical Shifts vs.
Concentrations of Iodide Salts Added to

0.10 M CsI Solutions in NMF.

126

 

 

 

8-—
E
o.
e- 6-
{E
(f) 1.11
E
.2
GE. 4_
.c
U
“3
m -1
l
E 2—
'5
m
U
" A CsSCN/0.48MCSI
a CsSCN/0.24MCSI
0 o CsSCN/0.1OMCsl
I I I *7
0.0 0.2 0.4 0.6 0.8

‘ "Concentration of Thiocyanate Salts Added
to Csl Solution in NMF

Figure 27. Plot of Cesium-133 Chemical Shifts vs.
Concentrations of CsSCN Added to 0.10 M,

0.24 M and 0.48 M CsI Solutions in NMF.

127
Table 63. Parameters of Linear Regression for Mixed Salt

Systems in NMF.

Concentration Correlation

System Shift(ppm/M) Coefficient
CsCl 19.83 0.9996
CSC1+0.10MCSI 22.05 0.9982
NaC1+0.10MCSI 21.95 0.9961
LiCl+0.10MCSI 21.62 0.9998
CsBr 30.29 0.9997
CsBr+0.10MCsI 29.05 0.9997
NaBr+0.10MCsI 28.37 0.9998
LiBr+0.10MCsI 27.36 0.9979
CsTPB -59.50 ~0.9998
CSTPB+0.10MCSI -52.79 -0.9930
NaTPB+0.lOMCsI ” -62.01 -0.9999
LiTPB+0.10MCSI -52.47 -0.9997
CSClO4 -1.34 -0.9840
CSC104+0.10MCSI -2.73 -0.9904
NaClO4+0.10MCsI -3.26 -0.9979
LiC104+0.10MCsI -4.18 -0.9997

128

Table 63 (Cont’d).

System Concentration Correlation

Shift(ppm/M) Coefficient
CSNO3 2.67 0.9971
CSNO3+0.10MCSI 1.67 0.9984
NaNO3+0.10MCSI 0.70 0.9969
LiN03+0.10MCSI 0.00 ------
CsSCN 10.38 0.9994
CSSCN+0.10MCSI 10.09 0.9987
NaSCN+0.10MCSI 9.37 0.9999
LiSCN+0.10MCsI 7.91 0.9998
CSSCN+0.24MCSI 10.47 0.9994
CSSCN+0.48MCSI 10.51 0.9999
CsI 36.15 0.9995
NaI+0.10MCsI 34.83 0.9996
KI+0.10MCSI 33.72 0.9984
TPAI+0.10MCSI 33.87 0.9991

129

solutions, and only the last enables a satisfactory

interpretation. of the experimental data. In *the present
case it is easy to see that covalency and electrostatic
effects are not plausible mechanisms. The short range
repulsive forces present during direct collisions between
the ions are again suggested as the most likely mechanism.
The molar molecular shift for the salts added to 0.10
M CsI can be defined as the concentration shift difference
between the mixed electrolyte system and 0.10 M CsI/NMF
(the concentration shift of 0.10 M CsI/NMF is assumed to be
zero). Since cesium iodide has the largest concentration
shift (or the strongest overlap between cation and anion),
molar molecular shifts for all the systems studied are all
negative (Table 64). The values clearly show that the order
of increasing shielding shift molar

(or' decreasing

molecular shift) is in agreement with that in the case of

Table 64. Molar Molecular Shift (ppm/M) for Mixed
Electrolyte Systems in NMF.
1' Cl' Br" TPB- c104' N03- SCN-
Cs+ 0 -14.10 -7.10 -88.94 -38.88 -34.48 -26.06
Na+ -1.32 -14.20 -7.78 -98.16 -39.41 -35.45 ~26.78
Li+ ---- -14.50 -8.79 -88.62 -40.33 -36.15 -28.24

130
single cesium salt solutions:
I" < 8:" < 01‘ < SCN- < N03“ < 0104' < T98".
The molar molecular shifts are also slightly cation
dependent. Except for the case of TPB-, all systems show
shielding shift increases in the order:

Cs+ < Na+

< Li+.
The dependence of the cation chemical shifts on the other
types of cation in NMF solution indicates that the effects
are quite specific. The existence of interactions between
ions of like charge has previously been suggested by
thermodynamic, calorimetric (151—155) and NMR measurements
(156). If the collision frequency remains essentially
constant, ‘which. is likely for interactions between
oppositely charged ions, then replacement of some cesium
cations by sodium or lithium ions will lead to a smaller
degree of mutual overlap and a upfield shift of the cesium-
133 resonance. This type of process will involve
interactions between ions of like charge. For the CsSCN/Csl
system, the Cs-133 chemical shift was measured as CsSCN was
added. to different concentrations of cesium. iodide. The
results .show that the concentration shift is essentially
constant. The processes determining the chemical shifts are
more or less independent of the total concentration.
Tetraphenylborate salts are always interesting salts.
In single electrolyte systems, the cesium-133 resonance of
cesium tetraphenylborate always shows a very large upfield

shift no matter what the solvent is. The magnitude of the

131
shifts (absolute value) are always as large as those of
cesium iodide (downfield shift) . However, sodium
tetraphenylborate shows no concentration dependence of the
sodium-23 chemical shift in a variety of solvents
eventhough sodium iodide still has the largest shift. This
implies that for sodium tetraphenylborate the anion doesn't
penetrate the cation solvation sphere and no significant
chemical shift occurs. On the other hand, in the NaTPB/CsI
mixed electrolyte system there is an unusually large molar
molecular shift (absolute value) compared to those in the
CsTPB/CsI and LiTPB/CsI systems. Again, this might be
explained on the basis of the phenomena of cation-cation
interactions. Since sodium ions don't interact with
tetraphenylborate ions, then cesium ions have more chances
to interact with sodium ions relatively. This will lead to
a more negative value of the molar molecular shift relative
to that of cesium iodide.
D. - ON N 8 ION
1. W

1-1 1W

1.1-1 WW

Solutions of LiSCN (0.06 to 0.60 M concentration
range), NaSCN (0.05-0.94 M), CSSCN(0.05-0.92 M) and TIAASCN
(0.06-0.50 M) in NMF were studied in the mid-ir region. The
Peak position of the C54 stretching vibration for all of

1

the salts studied was 2057 i 1 cm- (Table 65), ie., the

Peak frequency of the v mode of SCN- in NMF is independent

1

132

Table 65. Peak Positions v (cm-1) and Half-height Width

(Kl/2) (cm-1) of CEN Stretching Bands of Four
Thiocyanate Salts in NMF.
Salt Concentration 11(cm-1) H1/2(cm-l)
LiSCN 0.0560 2057.5 30
0.0823 2057.5 33
0.1946 2057.5 33
0.2376 2057.5 34
0.3162 2058.1 33
0.4263 2058.1 33
0.5775 2057.5 32
0.6040 2057.5 33
0.7951 2058.1 33
NaSCN 0.1034 2056.8 31
0.2122 2056.8 32
0.2815 2057.5 32
0.3989 2057.5 31
0.4441 2057.5 32
0.8024 2058.1 33
0.9412 2057.5 32
CsSCN 0.0527 2056.8 30
0.0947 2057.5 32
0.2181 2056.8 31
0.2820 2056.8 33
0.4011 2056.8 32
0.4916 2056.8 31
0.9239 2056.8 31
TIAASCN 0.0545 2056.8 31
0.1639 2056.8 32
0.2587 2057.5 32
0.4632 2057.5 32

133
of the nature of the cation and the solute concentration.
This suggests the absence of contact ion pairs in these
solutions: the thiocyanate ions are highly solvated. The
GEN band half-height width for all of the solutions
investigated is 32 i 2 cm-l. From this very broad band it
can be concluded that there is probably a nonuniform
environment surrounding the thiocyanate anions. Therefore
mathematical deconvolution was used. As shown in Figure 28,
three bands appear for all systems studied following the

deconvolution. Their positions are 2045 i l cm-l, 2055 i 1

cm'1 and 2066 i 1 cm?1

. Since the positions of these three
components are cation independent, the multiplet structure
must result from interactions between the thiocyanate
anions and solvent molecules.

Excess crown ethers 12C4, 18C6 and cryptand C222 were
added to LiSCN (0.06 M and 0.43 M), NaSCN (0.05 M and 0.44
M) and CsSCN (0.05 M and 0.40 M) solutions respectively.
These macrocyclic molecules are known to efficiently
complex the particular cations. Deconvolution shows that
addition of the complexing reagents produces no effect on
the systems studied: the positions of the three bands are
unchanged. This shows again that on the infrared timescale
there is essentially no cation-anion interaction in alkali
thiocyanate NMF solutions. FTIR spectra of 0.24 M LiSCN,
0.27 M NaSCN and 0.40 M CsSCN solutions were also obtained
at different temperatures from 264K to 328K. The

deconvolution results are shown in Figure 29. It Can be

134

 

 

 

‘ h 1 F T I
1950 2000 2050 2100 2150

Wavenumber(cm—l)

Figure 28. Deconvolution of the C N Stretching Band of

Thiocyanate Salts in NMF.

135

 

 

 

328K

298K

 

266K

17

 

I I I . I ‘7
.1950 2000 2050 2100 2150

Wavenumber(cm—l)

Figure 29. Deconvolution of GEN Stretching Bands of
Lithium Thiocyanate (0.24 M) in NMF at Three

Temperatures.

136

clearly seen that the 2045 cm-1.

peak changes its shape with
changing temperatures whereas there is no change in shape
or position of the other two components. The results
confirm that several species exist in the systems studied.

1.1-2 Compute; Resolution of Thiocyanate Bang;

The infrared contours given by the GEN stretching
vibration of the thiocyanate ion in NMF do not have
features that clearly indicate the individual components of
these composite bands. Resolution of these contours was
based on fitting their variations in spectra of solutions
having different compositions. Each contour was fitted by a
systematic series of approximations beginning with manual
adjustment of band parameters until contours in spectra of
several different solutions were matched by changing only
the height of the components and ending with computer
refinement (48). The quality of a match was visually judged
by a display of the difference between the experimental and
synthetic contours.

The band shapes were obtained from spectra of TIAASCN
solutions that contain primarily the unassociated SCN-
ions. The EN bands were fit by synthetic bands that had
358 Lorentzian and 65% Gaussian character. After an
iterative fitting' procedure for spectra. of several
solutions to determine the positions and widths, the
heights of the three bands could be varied independently to

fit the contours given by several other concentrations. The

final envelope fitting was again done by the computer with

137

the positions of the component bands fixed (2045.0 cm-l,

1 1).

Curve fitting results are given in Tables 66-67 and

2054.5 cm” and 2065.5 cm-

Figure 30. The results show that the three-peak fit is
better than a two-peak fit, and that the half-height widths

for the three peaks are relatively constant (30 i; 2 cm-l,

1 and 12 i 1 cm-1 respectively). Plots of

15 i, 1 cm-
relative integrated intensities of these three peaks versus
thiocyanate concentrations and solution temperature are
given in Figures 31-36. It seems that for lithium, sodium
and cesium thiocyanate there is no temperature dependence
of the relative intensity (-9 - 55°C). This shows the
uniqueness of solvation structures of thiocyanate anions in
NMF. Kabisch and Klose (157), from their studies of the
temperature dependence of the N-H band in NMF, also found
that NMF liquid structures are almost temperature
independent over the temperature range 295-390K. However,
all three thiocyanate salts do show some concentration
dependence of the relatiVe intensities of the three peaks.
With increasing salt concentration, the intensity of the

1 1

2054.5 cm- peak increases whereas that of the 2065.5 cm-

1

peak decreases. The intensity of the 2045.0 cm- peak is

essentially concentration independent.
It is difficult to provide assignments for these three

peaks. However, the 2054.5 cm-1

peak is the dominant
species in the system. Thus, they must represent free

thiocyanate anions. Thiocyanate ions may be wrapped around

138

Table 66. Curve Fitting Results of Thiocyanate
Salts in NMF.

Salt Concentration Relative Intensity(%) Chi-
(M) lst’ 2nd* 3rd* Square

LiSCN 0.0560 11.41 59.03 29.56 0.000298
0.0823 10.67 61.61 27.72 0.001984
0.1946 5.93 74.26 19.81 0.000745
0.2376 5.23 72.49 22.28 0.000480
0.3162 1.93 73.47 24.60 0.004981
0.4263 6.76 64.59 28.65 0.000636
0.5775 8.63 53.90 37.47 0.001094
0.6040 11.35 51.90 36.75 0.000773
0.7951 11.39 46.95 41.66 0.003362“

NaSCN 0.1034 13.10 70.37 16.53 0.000028
0.2122 12.22 69.97 17.81 0.000148
0.2815 10.05 69.66 ' 20.29 0.000345
0.3989 10.97 67.55 21.48 0.001410
0.4441 10.94 67.18 21.88 0.002871
0.8024 10.39 65.53 24.08 0.002050
0.9412 10.04 62.59 27.37 0.007214

CsSCN 0.0527 7.62 77.86 14.52 0.000788
0.0947 9.94 73.71 16.35 0.000293
0.2181 12.34 68.87 18.79 0.003519
0.2820 16.51 65.04 18.45 0.000509
0.4011 16.05 63.60 20.35 0.002272
0.4916 14.47 66.94 18.59 0.000978
0.9239 11.84 69.33 18.83 0.002371

* lst --- 2045.0 cm"1
2nd --- 2054.5 cm-:

3rd --- 2065.5 cm-

139

Table 67. Curve Fitting Results of Thiocyanate salts in
MMF at Different Temperatures.

Solution Temperature Relative Intensity (%) Chi-
(OC) lst* 2nd* 3rd* Square

0.2376 M -9.0 6.58 70.65 22.77 0.001977
LiSCN -1.0 7.14 70.18 22.68 0.001804
5.0 7.33 70.09 22.58 0.001439

13.0 11.66 66.68 21.66 0.004171

25.0 5.23 72.49 22.28 0.000480

34.0 9.99 66.59 23.42 0.003517

41.0 9.07 66.37 24.56 0.001080

48.0 9.08 65.84 25.08 0.002122
55.0 8.72 67.55 23.73 0.002167=

0.4885 M -1.0 11.61 66.81 21.58 0.001381
NaSCN 10.0 9.51 67.94 22.55 0.006390
15.0 12.68 66.77 20.55 0.001581

25.0 10.04 69.31 20.65 0.001292

30.0 12.03 67.76 20.21 0.001247

39.0 8.13 70.54 21.33 0.007727

44.0 11.08 69.22 19.70 0.003326

0.4011 M -7.0 14.30 69.69 16.01 0.002437
CsSCN 0.0 12.15 70.80 17.05 0.002144
5.0 10.75 72.22 17.03 0.002052

15.0 12.10 71.03 16.87 0.001871

25.0 16.05 63.60 20.35 0.002272

38.0 16.30 66.41 17.29 0.002331

45.0 10.09 65.78 24.13 0.001855

'* lst --- 2045.0 cm'1
2nd --- 2054.5 cm'1

3rd --- 2065.5 cm'1

140

 

 

 

 

(18—
C16—
8 2054.5
C
o
{3 (14—
o
3 2065.5
<
0.2- \
2045.0
0.0 ' l b '2 1 r l ' l
1950 2000 2050 2100 2150

Wavenumber(cm - 1)

Figure 30. Resolved Curves of the CEN Stretching Band

for Thiocyanate Salts in NMF.

141

100-1
0 2054.5cm—1
a 2065.5cm—1
A 2045.0cm—1
802
g: 1
,3 604 .°
m
C o
B 0
.E o
G) a
s 40— .
2 O
m
a 7 00 o
20— 11°
0 A I

 

 

l l l
0.0 072 0.4 0.6 0.8 1.0
Concentration of LiSCN (M)

Figure 31. Plot of Relative Intensities of Three (24

Resolved Bands vs. Concentrations of LiSCN

in NMF.

142

 

 

100-—
O 2054.5cm—1
D 2065.5cm—1
A 2045.0cm—1
80—~
g d o 0 ° 00
:1? 60- .
m
c
.93
.E
m
.5 40—
.2
w
0:
Cl
20‘ 0 0°
0

 

 

l l
0.0 0.2 0.4 06 0:8 T0
Concentration of NaSCN (M)

Figure 32. Plot of Relative Intensities of Three C'=‘N

Resolved Bands vs. Concentrations of NaSCN

in NMF.

143

 

100-1
o 2054.5cm—1
. CI 2065.5cm-1
A 2045.0cm—1
80- o
4:" 60—
m
c
8
£2 .
m
L; 40-
.2
m
m a
20‘ 0 o a a c:
O I

 

l l l l
0.0 0.2 0.4 0.6 0.8 1.0
Concentration of CsSCN (M)

Figure 33. Plot of Relative Intensities of Three GEN
Resolved Bands vs. Concentrations of CsSCN

in NMF.

144

 

 

100..
O 2054.5cm—1
D 2065.5cm—1
A 2045.0cm-1
80—<
E o o o o o o o
.43? 60—
(f)
C
.3.)
.E
m
:3 40-
.2
m
0:
20— ° ° °
0 l I I i I i 1
—10 10 30 50

Temperature (°C) [0.2376 M uscn]

Figure 34. Plot of Relative Intensities of Three can
Resolved Bands vs. Temperatures in 0.24 M

LiSCN/NMF Solution.

145

 

 

100-—
O 2054.5cm—1
Cl 2065.5cm—1
A 2045.0cm—1
80~
g 4 ° 0 o o o 0 °
,3? 60—
m
C
2
£5 .
m
.3 40—
2
0
Q: .
20‘ o o o a 0 D o
O I I f I I I fi
-10 10 30 50

Temperature (°C) [04885 M NaSCN]

Figure 35. Plot of Relative Intensities of Three CEN
Resolved Bands vs. Temperatures in 0.49 M

NaSCN/NMF Solution.

146

 

 

100—
O 2054.5cm—1
D 2065.5cm—1
A 2045.0cm—1
80-<
g 0 O O
.4? 60~
U)
c
B
m
.3 4o—
.9
m
0: .
20“ o
2 o 0 o A 2
O I I ' I ' I fi
-10 1O 30 50

Temperature (°C) [0.4011 M CsSCN]

Figure 36. Plot of Relative Intensities of Three GEN

Resolved Bands vs. Temperatures in 0.40 M

CsSCN/NMF Solution.

147
by the flexible NMF chain structure or thiocyanate ions may
be solvated through hydrogen bonds between either end of
the thiocyanate ion (158) and the terminal N—H group of the
NMF molecules, with cations attached to the oxygen (and
nitrogen) atoms of the NMF molecules. Because of their low

1

intensities, the 2045 cm- and 2065.5 cm'1 peaks could

represent the less solvated thiocyanate ions. For the 2045
cm.1 peak its relatively low wavenumber position may
reflect less triple bond character, and the possibility
that the nitrogen atom of the thiocyanate anion act as a
bidentate or tridentate ligand.

1-1-3 WWW:

mm

Some studies of the solvent vibrations were also
carried out. The absorption bands due to the stretching
vibrations of the N-H, c=o and C-NI groups of NMF in 1%
NMF/acetonitrile under the influence of dissolved LiSCN
(the concentration of the saturated solution is less than
0.5 M), NaSCN, CsSCN(sparingly soluble) and Iron(III)
perchlorate were examined. Acetonitrile was used as a
medium in order to break the hydrogen bonds among NMF
molecules since this hinders observation of the effects
being studied.

The results are shown in Table 68. From the observed
shifts it is clear that the hydrogen bonds between NMF

molecules are broken in the mixed solvent solution. For

example, the N-H band shifts to higher wavenumber (+101

148
Table 68. Wavenumbers (cm-1) of N-H, c=o and C-N Stretching

Bands in 1% (volume) NMF/MeCN Mixtures.

NMF Absorption Bands

N-H c=o C-N
Solution y Au v Av v Av
NMF 3301 --- 1667 --- 1243 ---
1% NMF/MeCN 3402 --- 1690 --- 1223 ---
0.15M LiSCN 3336 -16 1637 -3 1245 +17
0.38M LiSCN 3331 -21 1630 -10 1247 +19
sat'd LiSCN 3379 -23 1679 -11 1249 +21
0.10M NaSCN 3399 -3 1690 0 1229 +1
0.22M NaSCN 3397 -5 1639 -1 1231 +3
0.37M NaSCN 3391 -11 1636 -4 1237 +9
0.52M NaSCN 3337 -15 1635 -5 1240 +12
0.98M NaSCN 3379 -23 1633 -7 1242 +14
sat'd CsSCN 3403 +1 1690 0 1229 +1
0.09H Fe(ClO4)3 ----* --- 1643 -42 1274 +46

* can’t be determined because of low signal/noise ratio.

149

cm"1 ) in 1% NMF/MeCN compared to that of pure NMF liquid.
In this kind of ternary system (NMF, MeCN and salt), all
three components can interact with one another. Perelygin
(159) and Pominov e_t a]; (160) investigated the effect of
salts on the amide absorption bands in the ternary system
by comparing the absorption spectra of the binary system
(amide and acetonitrile) . They found that the interaction
of acetonitrile with the salts has no effect on the
positions of the absorption bands of amide by interaction
with salts.

The positions of the three bands of NMF change with
salt concentration. The more salt added, the greater is the
shift. Comparisons of the three stretching bands among
Perelygin’s (123), Pominov's (161) and this work for
different lithium and sodium salts are given in Table 69.

The wavenumber shifts for the C-N and c=o stretching
bands are the results of the coordination with the metal
ions, mainly through the oxygen atoms and possibly the
nitrogen atoms (123,159) . The wavenumber shifts of both
bands are dependent on the charge densities of metal ions.
The largest shifts of both bands are found in the case of
li‘e(ClO4)3 which has the highest charge density of metal
ions studied. Similarly, lithium ions have higher charge
densities than sodium ions, so that the larger wavenumber
shifts are found in the cases of lithium salts.

The N-H stretching bands are very broad (linewidth z 50

cm-l). With increase in the salt concentration, this band

Table 69. Wavenumber Shifts

Stretching Bands of NMF in

Mixtures.

Solution

a
NaClO4
NaSCN

NaI

a
Mg(ClO4)2

Fe(Clo4)3

150

(cm-1

) 0f N-H, C=0 and C-N

1% (volume) NMF/MeCN

+21
+21

+25

+10

+14

+11

+35

+46

a: Perelygin and co-workers (reference 123).

b: Pominov and Pavlova (reference 159).

151
moves toward lower frequencies, which results from the
interaction between metal ions and the nitrogen atoms of
NMF, only as indicated by Perelygin. This kind of
interaction would be stronger for lithium ions than for
sodium ions so larger shifts are observed for the lithium
salts. In Table 68, lithium thiocyanate has the higher
wavenumber shifts than sodium thiocyanate if their
concentrations are equal. However, the wavenumber shifts
are also dependent on the anions (Table 68). This indicates
that different anions interact with the N-H group of NMF
molecules to a different extent. It is hard to compare the
magnitude of the wavenumber shift in different anion
systems because the N-H band is very sensitive to the salt
concentration, whereas the concentrations are different for
the salts listed in Table 69. Bonner and Jordan (107)
studied the N—H stretch of NMF for several salts with the
same concentration (4 M) and found from the wavenumber
shifts that the anions interact in decreasing order (Table
4):
F'.’ > Cl'z Br- > 1" > N03- > c104" > PF6-.

Halide salts can interact with the N-H proton via hydrogen
bonding. Accordingly, thiocyanate ions could also form
hydrogen bonds with the N-H proton of NMF. However, the

hydrogen bonds must be weaker in thiocyanate systems than

that of the NaI system.

152

1-2 EMBL§25§EISQQBS§
1-2‘1 §:1_E!B_!23§B££!2n£§

Proton chemical shifts were measured for 2.11 M
LiCl/NMF, 1.21 M: NaSCN/NMF and. pure NMF solutions. The
results, given in Table 70, show that the chemical shifts
of the methyl and aldehyde protons of NMF in the IJCl and
NaSCN solutions shift to higher field compared to those of
pure INMF. These shifts reflect ‘the interactions between
metal ions and the oxygen and nitrogen atoms of the NMF
molecules. The interesting signal is that of the N-H
group. The N-H proton shifts downfield compared to that of
pure NMF. It is known that protons involved in hydrogen
bonding undergo resonance at a lower applied magnetic field
than non-associated protons (162). Then it can be concluded
that the order of the hydrogen bond strength in NMF
decreases by: chloride ion > thiocyanate ion > NMF

molecule. This is consistent with the results from 1% NMF

b

Table 70. Proton Chemical Shifts of NMF and Its Solutions.

-c§3 -cgo -Ng
Solution 6 A6 6 A6
NMF ‘2.75 ---- 3.09 ---- 7.91 ----
1.21M NaSCN 2.67 -0.03 3.01 -0.03 7.93 +0.02

2.11M LiCl 2.61 -0.14 7.97 -0.12 8.19 +0.28

153

/MeCN. Additions of LiCl and NaSCN partially destroy the
hydrogen bonds already existing in solution. In turn, the
loss of structure will be compensated by a greater degree
of ionic solvation. Further, in the LiCl and NaSCN
solutions the linewidth of the N-H proton signal
decreases and becomes more observable. This may result from
the decrease in proton exchange or reduction in quadrupole
effects.

1-2‘2 H:111_E:1é_QBQ.QL:1§.EBB.§2§§Q££E§E&£

Values of the N-l4, N-15 and Cl-35 chemical shifts of
thiocyanate, nitrate and perchlorate salts at various salt
concentrations, are given in ‘Tables 71-73. There is no
concentration dependence of chemical shift for these three
nuclei. The N-14 and Cl-35 chemical shift measurements are
not sensitive probes of the ionic solvation because of
their broad linewidths. Moreover, for nitrate and
perchlorate salts, the nitrogen and chlorine atoms are
surrounded by three and four oxygen atoms respectively.
This makes the measurements even less sensitive because the
nitrogen and chlorine atoms would not be easily disturbed,
which leads to very small chemical shift changes. The N-15
chemical shift measurement should be a better probe because
of its very narrow linewidth. However, the N-15 chemical
shifts of LiSCN-15 are still concentration independent
because lithium ions are well solvated and the interaction
between thiocyanate ions and the N-H proton of NMF is very

. weak in this concentration range (0.04-0.34 M).

154
Table 71. N-14 Chemical Shifts of Thiocyanate and Nitrate

Salts in NMF.

N-14 Chemical Shift (il.8ppm)

Salt Concentration(M) -SCfl C-H(NMF) -303
LiSCN 0.1102 -166.6 -263.7 -----
0.4231 -165.1 -262.3 -----
0.6437 -166.6 -262.7 -----
NaSCN 0.1135 -166.3 -267.0 -----
0.2356 -167.2 -266.9 -----
0.5373 -166.3 -266.6 -----
0.3051 -166.3 -266.5 -----
CsSCN 0.1032 -165.9 -266.7 -----
0.2639 -164.5 -265.4 -----
0.4095 -164.1 -265.3 -----
0.7652 -163.6 -265.3 -----
TIAASCN 0.0325 -164.5 -266.6 -311.5*
0.1127 -163.6 -266.5 -311.5*
0.2423 —165.1 -266.5 -311.5*
NaNO3 0.0302 ------ -265.5 3.0
0.2233 ------ -265.7 3.0
0.4672 ------ -265.0 3.0
0.7073 ------ -265.0 2.9
CsN03 0.0433 ------ -266.5 3.0
0.0929 ------ -266.5 3.0

* -n(isoamyl)4+

155

Table 72. N-15 Chemical Shifts of LiSCN-15 in NMF.

Concentration(M) N-15 Chemical Shift(i0.04ppm)
0.0423 -162.60
0.1664 -162.58
002146 -162058
0.3387 -162.56

Table 73. Cl-35 Chemical Shifts of Perchlorate Salts

in NMF.

Salt Concentration(M) 6(io.14ppm)

LiClO4 0.1870 -125.84
0.1104 -125.81
0.0630 -125.81

NaClO4 0.2899 -125.81
0.2287 ~125.81
0.0939 -125.81

CSCIO4 0.1835 -125.67
0.0919 -125.67
0.0695 -125.67

TlClO4 0.5023 -126.08
0.3594 -125.95
0.2067 -125.81
0.1302 -125.81
0.0648 -125.67

156
2- 9959.13.91.99:
The results presented above clearly show that solvation
of cations in NMF is through ion-dipole interaction with
=0 and probably C-N groups, while that of anions is
through interactions with the N-H groups. Halide and
thiocyanate ions form hydrogen bonds with N-H groups,
whereas perchlorate and nitrate ions interact with protons
in a different way. Infrared measurements of thiocyanate
systems do not reveal ion-ion interactions, unlike the NMR
measurements. In fact, the infrared time scale is much
faster than that of NMR and ion-ion interactions should
also be observable using the infrared technique. The
possible answer for this is that the ion-ion interactions
in NMF are very weak and the solvation structures of the
thiocyanate ions are not very much perturbed: this might
lead to a very small change in the infrared spectra,
unobservable under the experimental conditions.
E. 0N8 F 38
The studies already stimulate the following suggestions
for further studies:
1) A systematic study on ion-solvent interactions by H-
1 and C-13 NMR measurements.
2) Halogen NMR measurements on halide salt solutions in
NMF. This should provide more information about
interactions between anions and cations or solvent

molecules.

3) N-14 linewidth measurements have been used to study

157
the ion pair formation in the TlNo3/liquid ammonia system.
It must be interesting to use the same technique to study
the TlNOB/NMF system in which contact ion pairs are formed.

4) Infrared measurements of perchlorate and nitrate NMF
solutions.

5) Since solvent-separated ion pair has been found in
formamide system by infrared studies on the N-D band, it
must be also interesting to carry them out in NMF system.

6) Raman studies of highly concentrated thiocyanate
solutions in NMF. Hopefully, some bands resulting from

cation-anion interactions could be found.

CO

159
A- IEIBQDHQIIQ!

Previous studies in our laboratories (71,72,78-84) and
elsewhere (65,66,68-70,73,75,76) have shown that nuclear
magnetic resonance of alkali nuclei offers a very sensitive
technique for the studies of changes in the immediate
chemical environment of the alkali ions in solution. The
chemical shifts and linewidths of the resonances can give
information about ion-ion, ion-solvent, and ion-ligand
interactions. During the past decade alkali metal NMR has
been used extensively to study the thermodynamics and
kinetics of the complexation reaction between alkali metal
ions and crown ethers and cryptands.

Complexation between crown ethers or cryptands and
metal ions has not been previously studied in N-
methylformamide solutions. The work presented in this
chapter describes the complexation reaction of cesium ions
with a series of crown ethers and cryptands, such as 18C6,
DBl8C6, DA18C6, DC18C6, D321C7, DBZ4C8, DBZ7C9, C222, 0221
and C211, in NMF by using cesium-133 and C-13 NMR
measurements. These studies not only provide the stability
constants of different complexes but also give useful
information about the effects of the solvent, and the

different donor atoms of the ligands.

160
B. REEQLQQ MD QISCUSBLOH
1. m e at o o e um Ions b 806 n
8 b tut na o

1.1 139;

Complexation between 18C6 and two cesium salts (0.05 M
cesium ‘tetraphenylborate and 0.05 M (and 0.01. M: cesium
iodide ) were studied. The cesium-133 chemical shift was
determined as a function of the ligand to cesium. mole
ratio. The results are listed in Tables 74-76 and Figure
37. A downfield shift followed by a break and then an
upfield shift which gradually approaches a limiting value
can be explained by the formation of a strong 1:1 complex
followed by the addition of a second ligand to form a 2:1
sandwich complex (168) . It is known that the paramagnetic
shift is determined by the overlap of donor's lone-pair
electrons with p or d orbital of the cation (143). So the
upfield. shift of the 2:1 complex ‘probably results from
large decrease in the individual overlap integrals in the
2:1 complex as the short range repulsions are relaxed. The
complexation formation constants were determined by the
procedure described in the experimental part and Appendix
II, and the results are given in Tables 77 and 78. Three
effects are discussed as follows:

Concentration effect: Different curves were found for
0.01 M CsI/18C6 and 0.05 M CsI/18C6 systems. In the case of
0.05 M CsI a stronger, sharp break ‘was observed. The

difference in the formation constants (1:1 and 2:1) are not

161
Table 74. Variation of Cesium-133 Chemical Shifts with the
Mole Ratio (18C6)/(CsI) in NMF.

[CsI] = 0.0502 + 0.0002 M.

Mole Ratio Chemical Shift
(18C6)/(Cs+) (+0.04ppm)

0 1.16 0.01 2.22
0.12 1.56 0.43 1.41
0.25 2.25 0.63 -0.27
0.38 2.46 1.04 -0.95
0.57 3.42 1.54 -3.38
0.77 4.28 2.04 -5.56
0.82 7.04 2.16 -11.25
0.88 7.48 2.22 -12.29
0.95 9.01 2.35 -14.87
1.01 9.89 2.35 -16.63
1.04 11.50 2.29 -19.00

162
Table 75. Variation of Cesium-133 Chemical Shifts with the
Mole Ratio (18C6)/(CsI) in NMF.

[CsI] = 0.0101 + 0.0002 M.

Mole Ratio Chemical Shift

(18C6)/(Cs+) (+0.04ppm)

0 1.22 -1.43 1.26
0.20 1.42 -0.31 1.46
0.41 1.73 -0.19 1.46
0.61 2.03 0.32 1.36
0.31 2.54 0.74 1.26
0.91 3.05 0.95 0.95
0.96 4.06 1.05 0.63
1.02 7.61 1.05 -1.33
1.07 3.49 1.15 -1.95

1.12 9.22 1.15 -2.05

163
Table 76. Variation of Cesium-133 Chemical Shifts with the
Mole Ratio (18C6)/(CsTPB) in NMF.

[CsTPB] = 0.0506 + 0.0002 M.

Mole Ratio Chemical Shift

(1806)/(Cs+) (10.04ppm)

0 1.23 -4.31 0.35
0.10 1.29 -4.07 0.73
0.22 1.96 -3.33 -0.39
0.41 2.39 -2.14 -1.95
0.57 3.75 -1.03 -5.56
0.73 5.50 -0.21 -9.92
0.35 6.67 0.29 -12.60
0.90 7.39 0.43 -15.23
1.00 9.16 0.79 -17.35
1.03 10.49 0.79 -19.42
1.09 12.10 0.35 —21.69

1.12 0.92

164

10—

O
n l
“a.

Cesium—133 Chemical Shift(ppm)

 

 

_.1Q_.
_20_
a 0131MCsl
A 0.05MCsl
o 0.05MC TPB
—30 I I I I I I ' ' I 1 I [S I fi
0 3 6 9 12

Male Ratia(18C6)/(Cesium Ian)

Figure 37. Plot of Cesium-133 Chemical Shifts vs.

(18C6)/(Cs+) Mole Ratio in NMF.

Table 77. Logarithms

165

of Formation Constants of 1:1 and 2:1

(18C6/Cs+) Complexes of Cesium Salts with 18C6 in

Various Solvents.

Dielectric

Solvent Constant

Gutmann

Donor No. logK(1:l) logK(2:l)
MeCN 37.5 14.1 >5 0.57
PC 65.0 15.1 4.18 1.04
AC 20.7 17.0 >5.30 1.53
DMF 36.71 26.6 3.95 0.39
DMSO 46.68 29.8 3.04 0.0
Py 12.4 33.1 >4 1.04
NMF 182.4 25 0.01MCsI 2.74610.001 0.210.?

0.05MCSTPB 2.7:0.1 0.0510.03

0.05HCSI 2.9:0.1 -0.1110.03

Except for NMF, {CsTPB} = 0.01 M and all data are taken

from reference 78.

166
Table 78. Limiting Chemical Shifts of Cs+, Cs+(13c5) and

Cs+(18C6)2 in Various Solvents.

Solvent Cs+ [Cs+(l8C6)] [Cs+(18C6)2]
(ppm) (ppm) (ppm)

pc -36.5 -3.1¢0.2 ' -44.5¢0.3

AC -35.3 -6.4 -47+9

Py -32.4 10.2io.2 -43.0:0.2

ouso 63.0 23.610.4 -49¢24

our -0.3 3.37:0.05 ~48-210-7 .

Mean 24.1 14.310.2 -53¢7

NMF -1.34 0.01xcsx 2.6:0.3 -43i62
0.05MCsI 3.1+0.1 -75¢4
0.05MCsTPB 2.7+0.2 -65¢3

Except for NMF, [CsTPB] = 0.01 M and all data are taken

from reference 78.

167
significant within experimental error. This result is
reasonable because the equilibrium constant should be
concentration independent. However, the limiting chemical
shifts of the 1:1 complexes are slightly different (about 1
ppm). From Chapter III, we note that for pure cesium salts
in NMF the magnitude of the shifts is proportional to the
salt concentration: 0.05 M CsI has larger paramagnetic
shift than 0.01 M CsI in NMF ‘because of more ion-ion
interactions. In the same way, one would expect that the
1:1 Cs+-18C6 complex in 0.05 M CsI could show a larger
paramagnetic shift than that in 0.01 M CsI because cesium
ions are not completely inside the 18C6 cavity, which
leaves them accessible to iodide ions. The cesium ions in
the 1:1 complex in 0.05 M CsI system have more interactions
with iodide ions than those in the 1:1 complex in 0.01 M
C31. The chemical shifts for 2:1 complexes in both cases
are almost equal. This is understandable because in
Cs+(18C6)2 complexes the cesium ions are sandwiched between
two 18C6 molecules and‘have no interaction with the iodide
ions. Thus the chemical shift of the 2:1 complex is
concentration independent. It is? noted that the error is
much larger in the 0.01 M CsI case because the total
chemical shift change is very small (2 ppm), which makes
the fitting difficult. Since there is no concentration
dependence of formation constants, 0.05 M cesium ion
concentration were used for all systems studied because the

results would be more precise.

168

Anion effect: The 0.05 M CsTPB/18C6 system has the same
shape of the curve as that of CsI/18C6 at the same Cs+
concentration. This result indicates that 1:1 and 2:1
complexes are also formed 'in this system. The formation
constant is the same as that in the 0.05 M CsI system. The
1:1 complex in 0.05 M CsTPB has less paramagnetic shift
than the 1:1 complex in 0.05 M CsI. This is expected
because the tetraphenylborate anion is an electron-
withdrawer, which makes the interaction between Cs+ in 1:1
complexes and anions weaker. There are differences in the
formation constants and chemical shifts for 2:1 complexes
between the C81 and CsTPB systems. It seems possible that
the 2:1 formation constant for CsTPB(18C6)2 is larger than
that in the CsI system because tetraphenylborate ions are
less capable than iodide ions to compete with 18C6
molecules for’ 1:1 Cs+(18C6) complexes. However, as
mentioned earlier chemical shifts for 2:1 complexes should
be concentration and anion independent: yet there is some
variation. Tw0 problems should be considered. One is that
the chemical shifts of free cesium ions are not constant,
whereas they are assumed constant in the program used to
determine the formation constants. The other is that there
are differences in the chemical shift change rate before
reaching the top of the curves between the two cases, which
results from the differences in concentration shifts
between pure CsI and CsTPB solutions in NMF. These two

problems lead to the different observed chemical shifts for

169
the 2:1 complexes.

Solvent effect: In Tables 77 and 78, data from previous
studies (78) of the complexation between the cesium ion and
18C6 in six other solvents are reported. There is no simple
relationship between the formation constants and either the
dielectric constant or the donor number of the solvent. The
complexation reaction is essentially an‘ ion-dipole
interaction, and donor number should play a more important
role than the dielectric constant. Generally, solvents with
lower donor number should have higher formation constants.
However, NMF has approximately the same donor ability as
DMF, and the formation constant in NMF is about ten times
smaller than that in DMF. This might imply that there is
some kind of interaction between 18C6 and NMF molecules,
just as formamide molecules can form 2:1 complexes with
18C6 (169) via hydrogen bonding. Thus the formation
constant in NMF is smaller than that in DMF.

1.2 M

The attachment of two benzo groups on 1806 to form
D81806 was expected to result in weaker complexes. This
aromatic group decreases the 0---0 distance, which causes a
decrease in cavity size (170). A mole ratio study of D818C6
complexes with CsI in NMF was made. The results (Tables 79
and 82, and Figure 38) show that the Cs+-D318C6 complex
must be very weak because there is not even an inflection
point in the plot of chemical shift versus mole ratio. The

upfield shift is expected because benzo groups make the

170
Table 79. Variation of Cesium-133 Chemical Shifts with the
Mole Ratio (D818C6)/(CsI) in NMF.

[CsI] = 0.0493 + 0.0002 M.

Mole Ratio Chemical Shift
(D818C6)/(Cs+) (+0.04 ppm)

0 0.01
0.21 -0.63

0.44 —1.30
0.54 -2.30
0.72 -3.23
0.31 -3.79
0.87 -4.04

0.97 -4.73

1.03 -5.04

1.10 -5.41
1.30 -6.53

-7.84

171

electron density of the cavity smaller and the interaction
between D818C6 and Cs+ is weaker than that between NMF
molecules and cesium ions. Again, the formation constant in
DMF is about ten times larger than that in NMF.
Interactions between DBlBCG and NMF molecules are possible.

1.3 m

It would be interesting to test the effect of a
complexing agent which does not have aromaticity: therefore
dicyclohexyl-18C6 (DC18C6) was used. This ligand has five
isomers, of which only two have been isolated (termed
isomers A and B) . A mixture of these two isomers was used
in this study. The results (Tables 80 and 82, and Figure
38) show that the extent of the downfield shift is much
greater than that in the case of 18C6. This phenomenon may
be caused by the higher rigidity of the DC18C6 framework
which results in a larger overlap of the lone pair
electrons of oxygen with the cesium ion. However this
rigidity also leads to a drop in the stability of Cs+-
DC18C6 complex as compared with the Cs+-18C6 complex (164).
Again (Table 82), the formation constant in DMF is much
higher than that in NMF.

1.4 951306

The replacement of two oxygen atoms by two sets of N-H
atoms on 18C6 to form diaza-18C6 (DA18C6) is expected to
result in weaker complexes too (76,171-173), because the
electronegativity decreases as oxygen atoms are replaced by

N-H groups. The results (Tables 81 and 82, and Figure 38)

172
Table 80. Variation of Cesium-133 Chemical Shifts with the
(DC1BC6)/(CsI) Mole Ratio in NMF.

[C81] 5 0.0493 + 0.0002 M.

Mole Ratio Chemical Shift
(DC18C6)/(Cs+) (+0.04 ppm)

0 0.01
0.13 1.13
0.29 3.31
0.46 6.54
0.53 3.97
0.97 15.76
1.00 16.63

1.09 13.93

1.13 19.00
1.22 20.99
1.37 22.43
1.99 23.40
2.72 29.27
3.13 29.39
4.04 29.33
5.00 29.33

173
Table 81. Variation of Cesium-133 Chemical Shifts with the
(DA18C6)/(CsI) Mole Ratio in NMF.

[CsI] = 0.0493 + 0.0002 M.

Mole Ratio Chemical Shift
(DA18C6)/(Cs+) (+0.04 ppm)
0 0.01
0.14 0.51
0.25 0.94
0.43 1.69
0.54 2.12
0.34 3.49
0.94 3.37
0.96 3.99
1.00 3.99
(1.12 4.61
1.27 5.13
1.51 6.30
1.95 8.10
2.25 9.03
3.02 11.34
3.64 13.77

5.40 18.56

 

174

 

30-—

E

a.

3 20—

{E

U)

__ 1

a

.2

g 10—

.C a

U 0

N?

pr)

I

E o

3 0-4

’6

m

L)
A—A DB18C6
a DA18C6
G—e DC18C

-10 I , I , j
0 2 4 6

Male Ratio (Ligand)/(Cesium Ian)

Figure 38. Plot of Cesium-133 Chemical Shifts vs.

(Ligand)/(Cs+) Mole Ratio in NMF.

175
Table 82. Logarithms of the Formation Constants of Cesium

Complexes with Crown Ethers in Various Solvents.

Solvent D818C6 DC18C6 DA18C6
PC 3 4 ......
MeCN 1.54 > 4 2.48
AC > 3 > 4 ......
DMF 1.48 3.45 ------
DMSO 1.34 2.20 ------
PY 3.35 (1:1) > 5 ......

2.36 (2:1)
Me0H 3.55 (1:1) 4.25 ------
2.92 (2:1)
“20 0083 ---- ......
NMF (R z 0) 1.6+0.1 0.05+0.04

Except for NMF, all data are taken from reference 65.

176
show that the formation constant is much smaller than that
of the Cs+-18C6 complex.

2- W

The cesium cation has a diameter of 3.383 , which is
just the right size to fit conveniently into the cavity of
ligands D321C7, DBZ4C8 and D827C9, which range from 3.4 to
4.3K . Consequently, these Cs+-ligand complexes should be
very stable. The results (Tables 83-85 and Figure 39) show
that the cesium-133 chemical shift is essentially
unaffected by changes in the mole ratio of ligand to cesium
ions in the case of D321C7, and shifts upfield as
expected in the cases of D824C8 and D827C9.

There. are two possible explanations for the surprising
result in the case of D321C7. One is that the chemical
shifts of the complexes are very close to those of free
cesium ions. However, this is very unlikely because the
environment of complexed Cs+ should differ markedly from
that of the free cation. Indeed, the environment of the
cesium ions should be similar for all three ligand
complexes. There is no reason for Cs+-D821C7 to have a very
different chemical shift from those of other two complexes.
The second explanation is more likely. There is a
possibility of reaction between D821C7 and NMF molecules.
As mentioned before, NMF might interact with 18C6 and its
substituted analogs. However, in general the interaction is
not strong enough to prevent formation of Cs+-crown ether

complexes. Perhaps D821C7 has just the right size for NMF

177
Table 83. Variation of Cesium-133 Chemical Shifts with the
Mole Ratio (DBZlC7)/(CsI) in NMF.

[CsI] = 0.0493 + 0.0002 M.

Mole Ratio Chemical Shift

(DBZlC7)/(Cs+) (+0.04ppm)

0 0.99 0.00 0.21
0.17 1.04 0.04 0.24
0.44 1.16 0.09 0.26
0.53 1.23 0.12 0.23
0.66 1.31 0.15 0.30
0.33 2.37 0.17 0.33
0.94 0.19

178
Table 84. Variation of Cesium-133 Chemical Shifts with the
Mole Ratio (DBZ4C8)/(CSI) in NMF.

[CsI] = 0.0493 + 0.0002 M.

Mole Ratio Chemical Shift

(DBZ4C8)/(Cs+) (+0.04ppm)

0 0.94 0.00 -15.37
0.22 1.00 -4.35 -16.12
0.39 1.06 -7.65 -16.99
0.64 1.12 -11.32 —17.37
0.75 1.23 -13.33 -13.11

0.91 -15.37

Table 85. Variation of Cesium-133 Chemical Shifts with the
Mole Ratio (D827C9)/(CsI) in NMF.

[C311 8 0.0498 1 0.0002 M.

Mole Ratio Chemical Shift

(D327C9)/(Cs+) (+0.04ppm)

0 0.93 0.00 -16.43
0.09 0.99 -2.11 -17.24
0.27 1.03 -5.35 -17.30
0.51 1.15 -10.39 -13.36
0.70 1.13 -13.63 -13.61
0.36 -15.62 '

179

(3
4»
p

1
1’

 

-121

CeSUnT—133 Chenflcalfflflfi(ppafl
$6

I
-—O
03

l

A 082107
a 0824C8
_20 o 0827C9 T r 1

0.0 0.5 ' 11.0 1.5
Nkfle RaUa (Ugands)/(Ceskwnians)

 

Figure 39. Plot of Cesium-133 Chemical Shifts vs.

(Ligand) /(Cs+) Mole Ratio in NMF.

f

180
molecules to form very strong complexes. Then cesium ions
cannot compete effectively with NMF molecules for D821C7,
and almost no Cs+-0821C7 complex will be formed.

Cesium ions do form complexes with DBZ4C8 and DBZ7C9.
Their formation constants are about the same as those in
DMF (Table 86), which is different from what is found in
the cases of 1806 and its analogs. This could be explained
if D824C8 and D827C9 are too big for NMF molecules to form
complexes.

For large crown ethers, which are capable of forming
stable three-dimensional 'wrap around' complexes with
smaller cations (70,79,174), it can be expected that the
size of the cation will influence strongly the extent of
the complexation reaction. If the cation is too large, the
three-dimensional structures cannot be formed and only some
of the oxygen atoms can bond to the cation; consequently a
weaker complex results. On the other hand, if the ring size
is much larger than the cation, the ligand can still form
the ’wrap around' structure, but in this case the oxygen
atoms of the ligand are in close proximity and the
resulting repulsive force will weaken the complex.

The formation constant and the limiting chemical shifts
obtained in the study for the complexation of cesium ions
by D824C8 support the above conclusions. The cesium ion is
still too large to form a ’wrap around’ complex. Thus the
cation remains exposed to the solvent molecules, which

results in the solvent dependence of the cesium-133

181
Table 86. Logarithms of the Formation Constants of Cesium

Complexes with Crown Ethers in Various Solvents.

Solvent 032107 0324c3 032709
PC 3.30 3.25 3.64
MeCN 3.95 3.94 3.39
AC 3.93 3.71 4.24
DMF 2.34 2.10 2.20
DMSO 1.72 1.61 1.33
Py 4.27 4.00 4.15
Me0H 3.96 3.65 3.52
NMF (K:= 0) 2.24+0.09 2.4+0.3

Except for NMF, all data are taken from reference 65.

182
resonance for the complexed cesium ion (Table 87). an the
other hand, in the case of D827C9 the cesium ion is more
effectively insulated from the medium as a result of the
formation of a three-dimensional ’wrap around’ complex; the
limiting chemical shifts of the complexed cesium ion are
very nearly independent of the solvent (82).

3. WWW

Complexation equilibria between 0222, C221 and 0211 and
two cesium salts (0.05 M cesium iodide and cesium
tetraphenylborate) were studied. The results are given in
Tables 88-92 and Figure 40. In all cases, complexation of
the cation results in a paramagnetic shift of the cesium-
133 resonance. For the same ligand/Cs+ mole ratio the
magnitude of the chemical shift is in the order C222 > C221
> C211.

Cryptands containing three-dimensional cavities of
suitable size can form both inclusive and exclusive stable
complexes with the metal ions (71). In the studies of
complexation between C222 and two cesium salts (iodide and
tetraphenylborate), the results show that there is no anion
dependence of the formation constants, and of the limiting
chemical shifts of Cs+-C222. The limiting chemical shifts
of Cs+-C222 for both salts (227 + 22 ppm and 215 i 17 ppm)
are very close to that of the inclusive Cs+-C222 complex
(245 :3; 5 ppm). This implies that most of the complexes in
NMF are inclusive cryptate complexes, so no anion

dependence of the limiting chemical shifts of cryptate

183
Table 87. Limiting Chemical Shifts of 1:1 Complexes of
Cesium Ions with D821C7, D824C8 and DBZ7C9 in

Various Solvents.

Limiting Chemical shift (ppm)

Solvent 0821C? DBZ4C8 DBZ7C9
MeCN 8.17 -14.86 '22.34
AC -7.98 -29.24 ----
Me0H -19.07 -36.47 ----
DMF -0.92 -24.93 -24.19
DMSO 0.96 8.32 -20.14
Py 6.21 -20.75 -21.98
NMF ---- -22.98 -23.09

Except for

NMF, all data are taken from reference 82.

184
Table 88. Variation of Cesium-133 Chemical Shifts with the
Mole Ratio (C222)/(CsI) in NMF.

[CsI] = 0.0503 1 0.0002 M.

Mole Ratio Chemical Shift
(€222)/(Cs+) (ppm)

0 1.00 -0.01i0.04 137.59:0.12
0.11 1.08 ll.80:0.21 143.6910.12
0.28 1.18 40.74:0.21 149.9910.12
0.50 1.22 72.99:0.21 152.89i0.12¢
0.66 1.68 99.03i0.21 167.67i0.11
0.69 1.95 101.72i0.16 172.11i0.11
0.82 2.35 116.71i0.l6 174.9010.11
0.87 4.08 120.74i0.16 179.66+O.11
0.95 6.22 132.42i0.16 181.3110.11

185
Table 89. Variation of Cesium-133 Chemical Shifts with the
Mole Ratio (C222)/(CsTPB) in NMF.

[CsTPB] = 0.0506 + 0.0002 M.

Mole Ratio Chemical Shift
(c222)/<Cs+) (ppm)

0 0.99 -4.31+0.04 139.24+0.12
0.16 1.06 12.37+0.21 144.00+0.12
0.22 1.10 25.26+0.21 147.32+0.12
0.35 1.13 46.63+0.21 153.09+0.12,
0.62 1.61 90.35+0.21 169.94+0.11
0.72 1.97 104.62+0.16 175.11+0.11
0.73 2.40 112.57+0.16 173.52+0.11
0.37 3.47 124.56+0.16 131.52+0.11
0.97 6.03 136.45+0.16 133.59+0.11

186

Table 90. Variation of Cesium-133 Chemical Shifts with the

Mole Ratio (C221)/(CsI) in NMF.

[CsI] = 0.0494 + 0.0002 M.

Mole Ratio
(C221)/(CS+)

0 1.09 0.04+0.04
0.56 1.14 27.32+0.22
0.76 1.19 35.39+0.22
0.34 1.47 33.06+0.22
0.96 1.61 41.60+0.14
1.04 2.70 43.53+0.14

(PPm)

Chemical Shift

44.52+0.14
45.64+0.14
46.39+0.14
49.92+0.11:
51.06+0.ll

54.7210.11

Table 91. Variation of Cesium-133 Chemical Shifts with the

Mole Ratio (C211)/(CsI) in NMF.

[CsI] 8 0.0494 1 0.0002 M.

Mole Ratio Chemical Shift

(C211)/(Cs+) (+0.04ppm)

0 0.98 0.04 0.66
0.13 1.02 0.04 0.70
0.32 1.10 0.16 0.74
0.66 1.18 0.45 0.82
0.81 1.25 0.58 0.91
0.94 1.45 0.66 1.07

187

 

 

180‘ Q/\/——=r‘”"_’_”s
/—'\ /M
E
Q.
Q.
j; 140—
LE
(f7
3 3
-- 1
E 00
m
.c
C.)
I")
:3 60~
I
E
3
'3
Q 20“
I A C222
.» 0 C221
0 C211
"20 I l I 1
O 2 4 6 8

Male Ratio (Ligana)/(Cesium Ian)

Figure 40. Plot of Cesium-133 Chemical Shifts vs.

(Ligand)/(Cs+) Mole Ratio in NMF.

188

complexes is anticipated. In general, it is expected that
in solvents of high donicity the competition for the cation
between the ligand and the solvent molecules should
decrease the stability of the complex. As seen in Table 92,
with the exception of pyridine there does seem to be a
tough inverse correlation between the stability of the
complex and the donor ability of the solvent as expressed
by the Gutmann donor numbers. Pyridine, has the highest
donicity, and yet the most stable complex is formed in this
medium. It has been pointed out previously that pyridine,
being a nitrogen donor, or a 'soft' base, does not solvate
strongly a 'hard’ acid such as an alkali ion (175,176):
thus this apparent anomaly is removed.

In the case of cryptand C211, it is evident the
difference between the Cs+ radius and the size of the
ligand cavity (3.38 X vesrus 1.63) is so large that there
is no possibility for the formation of a complex in
solvents of high donicity such as NMF, DMF and DMSO.

The results obtained with cryptand C221 show that the
variation of the chemical shift with the ligand/cation mole
ratio'is indicative of the formation of a rather stable
complex in NMF. The formation constant of Cs+-C221 complex
is about the same as that of the Cs+-C222 complex in NMF
despite the fact that the cavity diameter of C222 (2.8K)
is larger than that of C221 (2.2K). One of the three
chains of C221 is shorter than the other two: perhaps this

may force the ligand into a configuration in which the

Sc

PC

He

AC

189
Table 92. Logarithms of the Formation Constants of Cesium

Complexes with Cryptands in Various Solvents.

Limiting Chemical Shift

Solvent 0222 0221 0211 of Cs+-C222(ppm)
p0 3.97 4.9 0.7 193.3

MGCN 4057 ----- 1014 21004

30 4.03 ----- 0.45 203.0

DMF 2.16 3.33 1.2 155.3

DMSO 1.45 2.94 (K.z 0) 144.3

Py > 5 ----------- 224.3

NMF 2.27+0.01 2.3+0.2 (K:= 0) 227+22 (CsTPB)

215+17 (CsI)

Except for NMF, all data are taken from reference 80.

190

macrocycle has a more widely open cavity and may be better
able to accept the cesium ion than is the case with the
C222 ligand (80). C221 forms a more stable complex with
cesium ion in DMF than that in NMF even though they have
almost the same donor number. As was mentioned earlier,
there must be some kind of interaction between the ligands
and NMF solvent molecules. The C-13 NMR spectrum of C221 in
NMF was measured (Figure 41). The difference in C-13
chemical shifts between C-4 and C-5 in C221 at room
temperature is 0.82 ppm (Table 93) . It has been reported
(177) if that a difference in C-13 chemical shifts between
C-4 and C-5 of C221 (Table 93 Solvent A) is indicative of
specific solvent-ligand interactions. A portion of the MMF
molecule (N-H or CH3) might penetrate into the cavity of
the ligand and form hydrogen bonds with the heteroatoms of
0221.

C. W

The stabilities of complexes between cesium ions and
crown ethers and cryptands decrease in the order: 18C6 >
D827C9 z D824C8 z C221 z C222 > DC18C6 > DA18C6. DBl8C6,
DBZlC7'.and C211 do not form complexes with Cs+. The
formation constants are usually small because of the high
donicity of NMF. Solvent-ligand interaction seems to be a
very important factor in determining the stability of the
complex. D821C7 might have just the right geometrical
environment for NMF molecules, which would account for the

absence of Cs+/D821C7 complexation. In the cases of 18C6,

191

”2 .
4’, ‘ t. Q ~ 0 /“\
/ :\ /\
:22. "f V0 V/N
x / o o .
\/ \\/

 

 

 

Figure 41. Carbon-13 NMR Spectrum of C221 in NMF.

192
Table 93. C-4 and C-5 Separation in 13C Spectra

of Cryptand C221 in Various Solvents.

Solvent A t(°0) a 4-55(ppm) Solvent 3 t(°0) 5 4-,5(ppm)

NMF 24 0.81 DMF -53 0.0
FA 34 2.32 THF -88 0.0
H20 30 2.00 1,3-Dioxolane -93 0.0
Nitromethane -20 1.78 Py -39 0.0
Nitroethane —70 1.49 AC -90 0.0
Nitropropane -80 1.26 Toluene —80 0.0
MeCN -42 1.64 Anisole -37 0.0
MeCN 24 0.47 Methyl Acetate 34 0.0
MeOH -80 1.70
MeOH -43 0.92
Ethanol -84 1.15

Except for NMF, all data are taken from reference 177.

b.

193
its analogs and 0221, the ligands are partially solvated by
NMF ‘molecules. This makes the Cs+/ligand complexation
formation constants in NMF are about ten times smaller than
those in DMF. Dibenzo-24-crown-8, dibenzo-27-crown-9 and
cryptand 0222 are too big to form complexes with NMF, so
complexation reactions between cesium ions and these
ligands are dominant. For these ligands there is little
difference between the formation constants in NMF and in
DMF.
D. EEQQEflIIQ!§_£QB_IQBI§EB_§IHDIB§

1) Create a new program which can solve the formation
of 1:1, 2:1 complexes and collisional ion pairs. This would
help us understand the influence of collisional ion pairs
on the complexation reaction.

2) Perform complexation studies on lithium, sodium and
thallium ions with crown ethers and cryptands.

3) Investigate the interactions. between. ligands and
NMF, FA and DMF molecules via H-1 and C-13 NMR

spectroscopy.

W

194

W

DETERNINATION OP ION PAIR PORMATION CONSTANTS BY
TN! NMR TECHNIQUE: DESCRIPTION OF THE COMPUTER

PROGRAN KINPIT AND SUBROUTINE EQUATION

The equilibrium for ion pair formation can be

expressed as
M +X =MX (1.1)
and

Kip = [M+X-] / [14+][X-1v: = Kc / v: (L2)

in which Kip’ K and v+are the thermodynamic ion pair

0 _

formation constant, the concentration equilibrium constant
and the mean ionic activity coefficient respectively. By
using the well known Debye-H'Lickel equation, v+ can thus

calculated as follows:

-logY:= [(4.198x106) z+z_ 11/2/(0'r)3/2] /

[1 + (5.029x109a11/2)/(DT)1/2]. (1.3)

In this equation 2+, z_ are the charges of the ions, I is

the molar ionic strength which is 1/2 (iciziz) (C =

195
concentration summed over all species in the solution). D
is the dielectric constant of the solvent and T and a are
the temperature (K) and the closet distance of approach of
the ions inK.
The observed chemical shift is a population average of
those of the free ion and the ion pair: i+e+

)X (1.4)

aobs =6fo + 5ipxip = (‘f - lip f + ‘ip'

Where X a [M+] / Ct; and C is the total concentration of

f t
metal ions in the system. Material balance gives

0t = [M+] + [M+x'] = [M+] + KC[M+]2. (1.5)

Therefore
+_._ 1/2
[M ] - [ l + (1 + 4KcCt) ] / (ZKC) (I.6)

e + a- 1/2
xf [M 1 / 0t [ 1 + (1 + 4cht) 1 / (2Kc)0t(1.7)

, 2
and Kc ..-Kip V4,. so that, finally

sob, = [-1 + (1 + 4K1PV:C)1/21<6f - sip/(wig?)

+5 (1.8)

ip’

Three constants and two unknowns are used in the FORTRAN

196

code:

COSNST (1) = CONST (2) = (01)“2 CONST(3) = a

5 f
The two input variables are the concentration of the salt
Ct and the observed chemical shift a obs which are
designated as XX(1) and XX(2) respectively in FORTRAN code.
Starting with an estimated values of 6 i

and K i the

p p'
program fits the calculated chemical shift observed values
by an iteration method. The SUBROUTINE EQUATION is listed

on the following pages.

0000000

0000000

00000000

197

SUBROUTINE EQN

IMPLICIT REAL*8(A-H,O-Z)
IMPLICIT INTEGER*4(I-N)
INCLUDE 'KINFIT:KINFITCOM.FOR/LIST'
INCLUDE 'KINFIT:KINEQNCOM.FOR/LIST'
Entry/Control Point
GOTO (2,3,U,S,1,7,8,9,10,11,12,13) ITYPE
ITYPE = 5 Initial call. No input has taken place
1 CONTINUE
RETURN
ITYPE = 6 Control card #1 and CONST have been input
7 CONTINUE
NOUNK = 2 ! Set the dimensions of the
problem
NOVAR = 2
RETURN
ITYPE = 3: 1 Experimental data has been
read
8 CONTINUE
RETURN

ITYPE = 1: Evaluate algebraic equation and residual

000 00

35

00000000

0000000

CONTINUE

03MA=3XP(-4198000.0*SQRT(XX(1))/(C0NST(2)**3*

(1.0+50.29*00NST(3)*SQRT(xx(1))/CONST(2))))

CONK=U(2)*(GEMA"2.0)

3:4.0*00NK*XX(1)

S:((-1.0+SQRT(1.+F))/(2.*CONK*XX(1)))*(CONST(1)-
U(1))+U(1)

IF (IMETH .NE. -1) 0030 35

Simulations only
RETURN

Fits
CONTINUE

BRESID:S-XX(2)
RETURN

ITYPE = 2: Set the initial conditions for
differential eqn's

CONTINUE

Y(1) = 1.0E-20

NOEQN = 1 ! Set the nuumber of
equations

RETURN

°CONTINUE

IF (0(1) .LE. 0.0) U(1) : ABS(U(1))

CONC1 = CONST(1)

CONC2 = CONST(2)

DY(1) = 0(1) ' (CONC1 - Y(1))*(CONCZ-Y(1))
-(U(1)/CONST(3))*Y(1)

RETURN

ITYPE = 4° Calculate the residual for differential

eqn's ,

0000000

0000 0000000

00000000

00000000

199

CONTINUE
IF (IMETH .NE. -1) GOTO 20

CONTINUE

CONST(6) is the molar absorptivity -- substitute exp
value if known.

CONST(6) = 5000.0
RESID : Y(1)+DLOG10(XX(2)/CONST(H))/(CONST(6)*0.2)
RETURN

ITYPE = 8: Calculate X(KVAR,I) the IPLT = 2 plotting
mode

RETURN

-ITYPE = 9: FOP(I) : X(KVAR+1,I) for the IPLT : 3

-- mode

CONTINUE
RETURN

ITYPE = 10: FOP(I) : X(KVAR+2,I) for the IPLT = 4
mode

200

11 CONTINUE
' RETURN

ITYPE = 11: FU(I) <<<< x-axis; FO(I) <<<< yaxis
(IPLT = 5)

00000000

12 CONTINUE
RETURN

0000000
H
.—3
l-<
"U
I!)
II
N
0
‘D
p—a
H
(D
D.
ID
"9
n
(D
"I
(D
p.
B
C
p—o
W
('1‘
H0
O
D

13 CONTINUE
RETURN
END

201

DETERMINATION OP COMPLEX FORMATION CONSTANTS BY
THE NMR TECHNIQUE: DESCRIPTION OP THE COMPUTER

PROGRAM KINPIT AND SUBROUTINE EQUATION

A. CO PO 1'
In the complexation reaction. between ligand L and
metal ion M+ in solution, the equilibrium for a 1:1

complexation reaction can be expressed as:
M + L = ML . (II.1)

In the same way as described in Appendix I, the observed
chemical shift can be expressed by:

*' 2 2 2 2 2
d Obs [(KCM KCL 1) + (K CL + K CM - 2K CLCM + ZKCL

fl, + 2x0" + 1)1/2](aM -6ML) / 2K0M +5“, (11.2)

where K is the formation constant, CM is the total
concentration of metal ions, CL is the total concentration
of ligands, 5 ML is the chemical shift of complexes. In
order to fit this equation, two constants and two unknowns

are used in the FORTRAN code:

202

0(1) 0(2) = K

="ML

cons'r (1) =0M CONST (2) =5"
The two input variables are the concentration of the ligand
CL and the observed chemlcla shift 5 obs which are
designated ae XX(1) and XX(2) respectively in the FORTRAN
code. Starting with an estimated values of K and ‘ML' the
program fits the calculated chemical shift to the observed

values by an iterative method. The SUBROUTINE EQUATION is

listed on the following pages.

0000000

0000000

0000000

00000000

000000

203

SUBROUTINE EQN

IMPLICIT REAL*8(A-H,0-Z)

IMPLICIT INTEGER*N(I-N)

INCLUDE 'KINFIT:KINFITCOM.FOR/LIST'

INCLUDE 'KINFIT:KINEQNCOM.FOR/LIST'

Entry/Control Point

0010 (2,3,u,5,1,7,8,9,10,11,12,13) ITYPE

ITYPE = 5 Initial call. No input has taken place
1 CONTINUE

WRITE (LUNOUT,8500) 1 Log the ID of this routine
8500 FORMAT (‘ *** EON: 1:1 Complex (16-JUN-86) ***')

RETURN

ITYPE : 6 Control card #1 and CONST have been input
7 CONTINUE

NOUNK = 2 ! Set the dimensions of the

problem

NOVAR = 2

RETURN

ITYPE = 3: ! Experimental data has been

read

8 CONTINUE

RETURN

204

2 ' CONTINUE
IF(U(2).GT.O.) GO TO 1000
U(2)=1000
1000 CONTINUE
A: U(2)*CONST(1)
B: U(2)'XX(1) ‘
C= (CONST(2)- U(1))/(2. 'A)
D: (B- A+1. )"2
CALC=((A-B-1.)+SQRT(D+H.*A))*C+U(1)
IF (IMETH .NE. -1) GOTO 35

C
C Simulations only
C
RETURN
C
c Fits
C
35 CONTINUE
RESID=CALC-XX(2)
RETURN
C
c ..............................
C
C ITYPE = 2: Set the initial conditions for
C differential eqn's
C
c ..............................
C
3 CONTINUE
RETURN
C
C ..............................
C
C ITYPE : 3 Evaluate the differential eqn 3
C
c ..............................
C
4 CONTINUE
RETURN
C ’ -
C ..............................
C
C ITYPE : 4 Calculate the residual for differential
C eqn's
C
C ..............................
C
5 CONTINUE
IF (IMETH .NE. -1) GOTO 20
Simulations only

000

RETURN

205

X(KVAR+2,I) for the IPLT : 4

F0(I) <<<< yaxis

C
C Fits
C
20 CONTINUE
C
C CONST(6) is the molar absorptivity -- substitute exp
C value if known.
C
CONST(6) = 5000.0
RESID : Y(1)+DLOG10(XX(2)/CONST(N))/(CONST(6)*0.2)
RETURN
C
c ..............................
C
C ITYPE = 8: Calculate X(KVAR,I) the IPLT = 2 plotting
C mode
C
c ..............................
C
9 CONTINUE
RETURN
C
c ..............................
C
C ITYPE = 9 FOP(I) : X(KVAR+1,I) for the IPLT = 3
C mode
C
c ..............................
C
10 CONTINUE
RETURN
C
c ..............................
C
C ITYPE : 10: FOP(I) :
C mode
C
c ..............................
C
11 ,CONTINUE
RETURN
C
C ..............................
C
C ITYPE = 11 FU(I) <<<< x-axis;
c (IPLT = 5)
C
C ..............................
C
12 CONTINUE
RETURN
C
c ..............................

C

206

C ITYPE : 12: Called after simulation
C .
c ..............................
C
13 CONTINUE
RETURN

END

207
B. , ON 0 S 0 ~ :
QQNBLEZEE

The equilibria for this reaction can be expressed as
+ +
M + L = ML K = [ML 1 / [M 1[L1 (11.3)
ML+ + L = ML K = [ML2+] / [ML+][L]. (11.4)

Let [M+].T and [L]T denote the total concentrations of the

metal ions and ligands respectively. Then

+ + + +
[M 1T = [M ] + [ML 1 + [MLZ ]
= [M+](l + K [L] + K K [L]2) (11 5)
1 1 2 °
+ +
[L1T = [L] 4» [ML 1 + 2[ML2 1

[L1(1 + K1[M+] + 2K1K2[M+][L]). (11.6)
(II.5) and (II.6) can be arranged as
+ _ + 2
[M 1 - [M 1T / (1 + K1[L] + K1K2[L] ) (11.7)
2K K [M+][L]2 + [L](l + K [M+]) - [L1 = 0 (11 3)
1 2 1 T ' '

Therefore

208

+ +2 + 12
[L] = {-(1 + 311M 1) + [<1 + K1 [M 1 + 8K1K21M ][L]TJ / 1
+
/ (4K1K2[M 1) (11.9)
The observed chemical shift is:

5 obs = XM‘5M + XML‘ML + XML21’ML2 (“'10)
In order to fit the calculated result with the experimental
data, an expression for the relative mole fraction of all
three species in terms of [M],r and [L]T is required. Two
constants and four unknowns are used in the FORTRAN code:

5 U(4)

U(l) = K U(2) = K 0(3)

1 2 ML 3 a ML2

+
CONST (1) = [M 1T CONST (2) = ‘M

The two input ‘variables are the total concentration. of
ligand [L]T and the observed chemical shift ‘obs which are
designated as XX(1) and XX(2) respectively. The SUBROUTINE

EQUATION is listed on the folowwing pages.

0000000000 0 0 0 0 00 0 00 000000

0000000 00 0

00000

209

SUBROUTINE EQN

This EON deals with the case of two competing
equilibria.
M+ + L = ML+ K1
ML+ + L = ML; K2
Unknowns:
U(1) Formation constant, K1, for ML+.
U(2) Formation constant, K2, for MLE.
0(3) Limiting chemical shift for ML+
U(4) Limiting chemical shift for MLE.

Variables:
XX(1) Concentration of ligand, (L).
XX(2) Observed chemical shift of M nuclei.

Constants:
CONST(1) Total concentration of metal, M.
CONST(Z) Chemical Shift for free metal.

.----------------------------------------------“----‘---.
.---------------------------------------------------------

IMPLICIT REAL*8(A-H,O-Z)
IMPLICIT INTEGER'4(I-N)

INCLUDE 'KINFIT:KINFITCOH.FOR/LIST'
INCLUDE 'KINFIT:KINEQNCOM.FOR/LIST'

INTEGER.“ NUMITR ! Maximum iterations
of metal series

”DATA NUMITR/1000/

11133 = 5: Initial call. No input has taken place

0000000

0000000

0000

0000000000

CONTINUE

WRITE (LUNOUT,8500) ! Log the ID of this routine
FORMAT (' *** EON: SANDCP { Sandwich Complexation',
'26-SEP-86) } ***')

RETURN

CONTINUE

NOUNK = 4 ! Set the dimensions of the
problem

NOVAR : 2

RETURN

CONTINUE
RETURN

-This subroutine will solve a general polynominal

The polynominal must be set so that the constant is
positive.

U(1) = ABS(U(1))

The Single letter variables that follow are the
coefficients for the polynominal that is to be solved

A=U(1)*U(2)
A1=-A
B = 2.0”A1'CONST(1) + A'XX(1) - U(1)

000 00

2005

200”

2003

2000

3000
3001

2001

2002

2006

999
2008

211

C = U(1)*XX(1) - U(1)'CONST(1) - 1
IF(XX(1).EQ.0.0) GOTO 2008

FREELi is the current value of the ligand being
tested to see if it is the true free ligand
concentration.

VALUEi is the value of the polynominal

FREELO = 0.0

FREEL1 = 0.00
VALUE1=XX(1)
FREEL2=XX(1)/1.0E+12
CONTINUE

VALUE2=((A1*FREEL2+B)‘FREEL2+C)*FREEL2+XX(1)
IF(VALUE2.EQ.0.0) 0010 2001
IF(VALUE2.GT.0.0) 0010 2000

CONTINUE
FREEL3=(FREEL2-FREEL1)/(VALUE1-VALUE2)*VALUE1+FREEL1
IF(ABS((FREEL3-FREELO)/FREEL3).LT.0.0000001)
0010 2002
VALUE3=((A1'FREEL3+B)*FREEL3+C)“FREEL3+XX(1)
IF(ABS(VALUE3).LT.0.00000001) 0010 2002
IF(VALUE3.GT.0.0) 0010 2003

VALUE2=VALUE3

FREEL2=FREEL3

FREELO=FREEL3

0010 2004

CONTINUE

VALUE1:VALUEB

FREEL1:FREEL3

FREELO=FREEL3

0010 2004

CONTINUE

VALUE1=VALUE2

FREEL12FREEL2

IF(FREEL2.LT.XX(JJ/100.) 0010 3000
FREEL2=FREEL2+XX(1)/100.

0010 3001

CONTINUE

FREEL2:FREEL2*10.

'CONTINUE

IF(FREEL1.GT.XX(1)) 0010 2006
0010 2005

CONTINUE

FREEL=FREEL2

0010 2007

CONTINUE

EREEL=EREEL3

0010 2007

CONTINUE

IF(LAP.NE.3) 0010 2008

NRITE (LUNOU1,999)
EORMAT(/,5x,29au NO NEGATIVE VALUE FOUND **)
CONTINUE

3002

000

000

35

00000000

0000000

212

FREEL=0.0
CONTINUE

EREEM = CONST(1)/ (1.0 + U(1)*FREEL +

U(1)'U(2)*EREEL*FREEL)

FREEML - U(1)*FREEM*FREEL

FREMLL - U(2)”?REEML'FREEL

CALC = ( FREEH'CONST(2) + FREEHL*U(3) + FREMLL*U(H))/
CONST(1)

CONTINUE
IF (IMETH .NE. -1) GOTO 35

Simulations only
RETURN

Fits

CONTINUE

RESID:CALC-XX(2)
RETURN

ITYPE = 2: Set the initial conditions for
differential eqn's

CONTINUE
RETURN

CONTINUE
RETURN

ITYPE = fl: Calculate the residual for differential
eqn's

213

C .-
5 ' CONTINUE
IF (IMETH .NE. -1) GOTO 20
C
C Simulations only
C
RETURN
C
c Fits
C
20 CONTINUE
RETURN
C
c ..............................
C
C ITYPE : 8: Calculate X(KVAR,I) the IPLT = 2 plotting
C mode
C
c ..............................
C
9 CONTINUE
RETURN
C
c ..............................
C
C ITYPE : 9: FOP(I) : X(KVAR+1,I) for the IPLT = 3
C mode
C
c ..............................
C

10 CONTINUE
RETURN

ITYPE : 10: FOP(I)
mode

X(KVAR+2,I) for the IPLT : 3

00000000

11 'CONTINUE
RETURN

ITYPE = 11: FU(I) (<<< x-axis; F0(I) <<<< yaxis
(IPLT = 5)

00000000

12 CONTINUE
RETURN

c ..............................

00000

21H

ITYPE : 12: Called after simulation

13 CONTINUE
RETURN
END

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

215

w. Ostwald, Z1 £0!§ik;.§h§mi. 2, 270 (1888).

F. Kohlrausch and L. Holborn, "Das Leitverm gen der
Elektrolyte", Teuber, Leipzig (1916).

L. Onsager, Z; Phx§,, 26, 277 (1927).

L. Onsager and R. M. Fuoss, l1 Bugs, Qhem6, 66, 2689
(1932).

R.M. Fuoss and C. A. Kraus, 66 Am, gngm6 6666, 66, 476
(1933). '

L. Onsager and R.M. Fuoss, 66 23266 gh6m6, 61, 668
(1957).

R.F. Prini, 116366 Eazaday §QC., 66, 3311 (1969).
E. Pitt. 2:991 8221 S921 Ar. 211. 43 (1953)-

R.M. Fuoss and K. L. Hsia. 2:921 £2112 82292 $911. 52.
1550 (1967).

J.C. Justice, 16 636m; Ehxg6, 66, 353 (1968).
J.C. Justice, ELQQSIQQDLEE Acta., 16, 701 (1971).
R.M. Fuoss, 16 Eth6 thmi. 12, 525 (1975).

T._Arnold and M. E. Packard, g6 chem, Enxg6, 12,
507,(1951).

J.F. Hinton and K. R. Metz, " NMR of Newly Accessible
Nuclei", Volume 2, Academic Press, New York, Chapter
14 (1933).

E. G. Bloor and R. G. Kidd, 6636 16 gngm6, 56,
3425 (1968).

E. G. Bloor and R. G. Kidd, 6636 Q6 Qngm6, 66,
3926 (1972).

C. Deverell and R. E. Richards, M211 Ehng, 16,
551 (1966).

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

216

J.D. Halliday, R.E. Richards and R.R. Sharp, 2199; Boy,
9991 Landau. 3135. 45 (1969)-

C. Hall, R.E. Richards and R. R. Sharp, 2:221 8921 §OQ.
Landgn. 111A. 297 (1974)-

A-I- POPOV. Burs 82211 Qnsm1. i1. 101 (1979)-

R. H. Erlich and A. I. Popov, is An; Qngm6 $666, 22,
5620 (1971).

W.J. DeWitte, L. Liu, E. Mei, J.L. Dye and A.I. Popov,
11 591n1 Chem1. 5. 337 (1977)-

S. Khazaeli, A.I. Popov and J. L. Dye, 11 Eh231 gngm6,
95. 4238 (1982).

W.J. Dewitte and A. I. Popov, 21.52101 gngm6, 6, 231
(1976).

J.F. Hinton and K. R. Metz, 26 fian; Qth6, 2, 197
(1930).

H.G. Hertz, V. Versmold and C. Yoon, 26:6 Bungen; £h2§1
thm1. El. 577 (1933)-

H. G. Hertz and R. Mazitov, 53:1 Bunssn1 Rhys; thm1,
2:, 1103 (1981).

C. Deverell, M211 £h¥§1p 16, 491 (1969).

H. G. Hertz, R. Tutsh and H. Versmold, 36:6 finnggn;
RDXEL 9h9m1. 15. 1177 (1971)-

H.G. Hertz, H. Holz, G. Keler, H. Versmold and C. Yoon,
59:1 2909901 20221 thm1. 19. 493 (1974)-

H. G. Hertz, G. Stalidis and H. Versmold, is thm;
Bh¥§1: §§. 177 (1959)-

H2.Weing§rtner and H. G. Hertz, 26:6 231116611L BBXQL
Qhfils. 51. 1204 (1977)-

J.C. Evans and G. Y.-S. Lo, 11 Eh¥§1 Qn6m6, 62, 3223
(1965).

W.F. Edgell, A. T. Watts, J. Lyford and W. H. Risen, 16
Am; thm1 £991. 22. 1855 (1966).

w.r. Edgell, J. Lyford, R. Wright, W. M. Risen and A.T.
Watts. 11 Ami thm1 5991. 22. 2240 (1970).

B.w. Haxey and A. I. Popov, 11 Am; Chem; £921. 21, 20
(1969).

41

49

50

51.

520

53.

54.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

217

J. L. Wuepper and A. I. Popov, 26 Am; thm; 2666, 21,
4352 (1969).

A- 1- Popov, Burs 82211 98281. 11. 275 (1975).

A.T. Tsatsas, R.H. Stearns and W.M. Risen, 26 Am; ghgm6
5991. 25, 5247 (1972).

Y.M. Cahen and A. I. Popov, 21 26166163 6h6m6, 6, 599
(1975).

W.F. Edgell, A. T. Watts, J. Lyford and W. M. Risen, 16
Am1 Chem1 £921. 88. 1855 (1966)-

W.F. Edgell, J. Lyford, R. Wright, w. M. Risen and A.T.
WattS. 21 881 98281 5991. 22. 2240 (1970)-

W.F. Edgell, J. Lyford, A. Barbetta and C. I. Lose, 26
An1 §h§m1 £921. 21. 6403 (1971)-

W.F. Edgell and J. Lyford, l1 Am6 Cbsm1.§991. 22, 6407
(1971).

W.P. Edgell and A. Barbetta, 11 Am; Qhfimi £221. 96, 415
(1974).

J. B. Gill. BRIE ABEL; §h§m1p.ii. 1355 (1931)-

D.N. Waters, " Vibrational Spectra of Electrolyte
Solutions "in pecialist Periodical Reports," Molecular
Spectroscopy" Vol. 6, The Chemical Society, London
(1979).

A.G. Miller and J. W. Macklin, l1 Rhys, gngm6, 62, 1193
(1985).

R.H. Atherton and S. I. Weisman, J; Am; thm; $221. 61,
1330 (1961).

P.J. Zandstra and S. I. Weisman, 16 Am; §h6m6 $666, 61,
4408 (1962).

K. Hételmann, J. Jagur-Grodzinski and M. Szwarc, J6 AmL
Chan; £291. 21. 4545 (1959)-

H. Smith and H. C. R. Symons, IIQn§1 {graggy £921. 61,
338 (1957).

T.R. Griffiths and R. H. Wijayanayake, 116n§6 {gigggy
£991. 55. 1563 (1970).

T.R. Griffiths and R.H. Wijayanayake, £1£1§1 EQIQQQ! 1,
62, 1899 (1973).

6.

69
70
71.
72.
73.

.74...

55.

56.

57.

58.
59.
60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

218

W. R. Gilkerson and A. M. Roberts, 11 221 63661 §oc,,
192. 5181 (1980).

W.R. Gilkerson and K. L. Kendrick, 11 23261 66631, 26,
5352 (1984).

K.L. Kendrick and W. R. Gilkerson, 21 661n1 gngm1, 16,
253 (1986). '

C.J. Pedersen, 11 Am1 CREE; 6661, 62, 7017 (1967).
C.J. Pedersen, 11 Am1 gngm1 6661, 22, 391 (1971).
C.J. Pedersen, 11 6:61 gngm1, 66, 254 (1971).

B. Dietrich, J. H. Lehn and J. P. Sauvage, Iggrgnggzgn
19111. 2885 (1969).

E. Graf and J. M. Lehn, 21 561 CREE; 6661, 21, 5022
(1975).

J-M- Lehn. 8921 98981 3921. 11. 49 (1978)-

J.H. Lehn, E. Souveaux and A. K. Willard, 61 Am1 thm;
£221. 129: 4915 (1973)-

R.H. Izatt, J. S. Bradshaw, s.A. Nielsen, J.D. Lamb, D.
Sen and J.J. Christensen, £82m; 3621, 66, 271 (1985).

R.H. Izatt and D.P. Nelson, 6616366, 165, 443 (1969).

P. Seiler, H. Dobler and J. D. Dunitz, 56;;
911518119911. 125, 2744 (1974).

R.H. Izatt, R. E. Terry, D. P. Nelson, Y._Chan, D. J.
Eatough, J.S. Bradshaw, L. D. Hansen and J. J.

Christensen, 21 Am1 QhfiE; 6661, 26, 7626 (1976).

D.E. Fenton, H. Mercer, N. S. Poonia and H.R. Truters,
£89m; 9988281. 66 (1972)-

H.-A. Bush and H. R. Truter, 11 gngm1 6661, Egrkin 11,
345 (1972).

E. Hei, A.I. Popov and J.L. Dye, 11 261 thm1 6691, 22,
6532 (1977).

R. D. Boss and A. I. Popov, 136161 ghgm1, 26, 1747
(1986).

J.J. Christensen, D. J. Eatough and R. H. Izatt, Chem;
3321, 15, 351 (1974).

J.H. Lehn and J.P. Sauvage, 11 561 thm1 6661, 21, 6700

81

87

88

89.

90.

910

92.

75.

76.

77.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

89.

90.

91.

92.

219
(1975).

R.H. Izatt, D. J. Eatough and J.J. Christensen, 6616661
2989189. 15. 161 (1973)- ,

H.F. Frensdorff, 11 Am1 §h6m1 $991..21. 600 (1971).
D.J. Cram, R. C. Helgeson, L. R. Sousa, J. M. Timko, M.

Newcomb, P. Moreau, F. Dejong, G. W. Gokel, D. H.
Hoffnam, L. A. Domeier, S. C. Peacock, K. M. Madan and

L- Kaplan. Burg 8 89211 98981. 11. 327 (1975)-

E. Mei, A.I. Popov and J. L. Dye, 11 thg; thm;. 61.
1877 (1977).

M. Shamsipur and A. I. Popov, 11 Am1 thm; £991. 121.
4051 (1979).

E. Mei, L. Liu, J. L. Dye and A. I. Popov, 61 661n1
QDEEL. 9. 771 (1977)-

A.J. Smetana and A. I. Popov, 11 661n1 gn6m1, 2, 183
(1980).

M. Shamsipur, G. Rounaghi and A. I. Popov, 11 661n1
98981. 2. 701 (1980)-

J.D. Lin and A. I. Popov, 11 261 63661 6691, 122, 3773
(1981).

P. A. Mosier-Boss and A. I. Popov, 11 Am1 thm1 629;,
121. 6168 (1985).

E. Kauffman, J. M. Lehn and J. P. Sauvage, 36121 ERIE;
AQSQ1. 52. 1099 (1975)-

D. K. Cabbiness and D. W. Margerum, 11 Am1 SEEM; 6921,
21, 6540 (1969).

R. D. Hancock and A. E. Martell, ngmgntg 6n Ingrg;
m: Q: 237 (1988)“

J.A. Riddick and W.R. Hunger, "Techniques of Chemistry,
II, Organic Solvents", Wiley-Interscience, New York
(1970).

R-L- JoneS. 11.M91L'§292312§£1. 2. 581 (1958)-

T. Miyazawa, T. Shimanouchi and S. Mizushima, Qngm1
ERXEE. 21. 408 (1955)-

I- Suzuki. 59112.§hng £921 £281. 35. 540 (1962)-
R.A. Russel and H.W. Thompson, 622622663161 56661, 2,

93.
94.

95.

96.

97.

98.

99.

100.

101.

102.

103.

104.

105.

106.

107.

108.

109.

110.

111.

220
138 (1956).
T. Miyazawa, 11 M211 6266616661, 1, 155 (1960).
R.A. Elzaro, 61661 522:1; Int; B. 11.
52111 ghgmL £291 Inn21‘fill

1052 (1973).

M. Kitano and K. Kuchitsu.

631 (1974).

L.A. LaPlanche and M.T.
337 (1964).

RogerS. 11 881 989m; 5991. 85.
M.J. Aroney, R.J.W.

£991, 1, 3179 (1965).

Y. Sugawara, A.I. Hirakawa, M. Tsuboi, S. Kato and K.
Morokuma, CREE; 2n261, 62, 339 (1981).

H. Ohtaki, S. Itoh and B. M. Rode, 66111 thm; 6921
innl. 52. 271 (1985)-

V.A. DurOV. 39581 11 £8251 98982. 25.

Le Fevre and A.N. Singh, 11 thm1

1616 (1981).

G.R. Leader and J. F. Gormley, 11.5m1.§h3n1.5991.
11, 5731 (1951)
S.J. Bass, W. I. Nathan, R.H. Meighan and R.H. Cole,

11 20281 dhsm1. 58. 509 (1964)-

P. Winsor and R. H. Cole, 61 26261 6n6n1, 66, 2486
(1982).

P-P- Rastogi. £3112 98981 $991 inn2. $3. 2442 (1970)-

P-P- Rastogi. 2:291 Indian N2111 £211 Adddl. Eart_8.

51, 345 (1981).

J. Bukowska, 11 8611,6636661, 26, 1 (1983).

0. D. Bonner and C. F. Jordan, 26261611 gngm1 8 2626166,
6, 293 (1976).

R.C. Paul, J. P. singla, M. S. Lamba, D.S. Gill and.P.
Narnia. Indian 11 98981. 11. 1024 (1973)-

L. Weeda and G. Somsen, 3661 1:621 gn1m1, 66, 159
(1966).
L. Weeda and G. Somsen, 3661 {11621 66161, 66,

263 (1967).

C.H. Criss, R.P. Held and E. Luksha,
12, 2970 (1968).

l1.2h¥§2.§h2n;r

112.

113.

114.

115.

116.

117.

118.

119.

120.

121.

122.

123.

124.

125.

126.

127.

128.

129.

221

D.S. Gill, J. P. Singla, R.C. Paul and S.P. Narula,

.11 Chani 2991. 522 (1972)-

J. Emsley, D.J. Jones, J.M. Miller, R.E. Overill and
R.A. Waddilove, 61 An1 Qngm1 6661, 162, 24 (1981).

C. Carmona, G. Eaton and M.C.R. Symons, 11 gngm1 £291.
snani damnunI. 873 (1987)-

R.C. Paul, D.S. Gill, J.P. Singla and S.P. Narula,
Indian 11 Chami. 2. 63 (1971)-

T. Pontani and B. M. Rode, 36n662h1 thm1, 166, 1153
(1977).

T- Shedlovsky. II Eranklin Insti. 222. 739 (1938)-

R-D- Singh and R- Gopal. 22111 Chani Sada Inna. 42.

2088 (1972).

C.M. French and K. H. Glover, 116n§1 Kazaflax1 $221. 51.
1418 (1955).

G.P. Johari and P. H. Tewari, 11 BREE; Qn6m1, 62, 3167
(1965).

R.P. Held and C. M. Criss, 61 23261 gngm1, 62, 2611

(1965).
E. Lnksha and C.M. Criss, 21 2ny§1 gngn1, 16, 1496
(1966).
I. S. Perelygin, S. V. Izosimova and Yu. M. Kessler,

znnri Strnki KhinI. 2. 390 (1968).

R.W. Briggs, K.R. Metz and J.P.
gnani. 2. 479 (1979)-

H. Weingfirtner. II 20251 Qnani.
(1982).

I'Io'WBi-ng'a'l’tner. it. 91191.. 5.9.9.2.: m m .1! 18.!
3053 (1982).

Hinton. II Saintian

(Wiesbaden), 166, 23

M. Q. Martin, G. J. Martin and J.
"Practical NMR Spectroscopy",
London. (1980)

-J. Delpeuch,
Heyden and Sons, Ltd.,

D.N. Bhattacharyya, C. L. Lee, J. Smid and M. Szwarc,

11 £8221 98281. 22. 608 (1965)-

A. Rykowski and H. L. Vanderplas,
£221. 21. 204 (1975).

£991 11 8921 Natni

130.

131.

132.

133.

134.

135.

136.

137.

138.

139.

140.

141.

142.

143.
144.

145.

146.

147.

.F.L. Cook. 11 9:91 ChanI. 32.

222

G.W. Gokel, D. J. Cram, C. L. Liotta, H. P. Harris and
2445 (1974).

B.L. Haymore, J. D. Lamb, R. M. Izatt and J. J.
Christensen, 1ngzg1 gngm1, 21, 1598 (1982).
C. Brevard and P. Granger, " Handbook of High

Resolution Multinuclear NMR", John Wiley and Sons

(1981).
J. C. Lindon and A. G. Ferridge, 21291 in NMR
2229119221. 11. 27 (1980)-

G.C. Templeman and A. L. Van Geet, 61 Am1 Qh6m1 6661,
21, 5578 (1970).
R. D. Boss, Ph.D. Thesis, Michigan State Univrsity,

East Lansing, Michigan (1985).

J. L. Dye and
443 (1971).

v. A.

N1celY1 1.7. gm; 351119.... 5.8.:

H.-H. Nam, Private Communication.

J- 8- Shih and A- I- POPOV. Inardi 82911 Chani Latti.
11, 105 (1977).

W. J. Dewitte, R. C. Schoening and A. I. Popov, 136:61

HEEL; £83m; Lfitta. 12. 251 (1975)-

R. W. Briggs and J. F. Hinton, 21 Eana gngm1, 6, 827
(1977).

Y. M. Cahen, P.R. Handy, E.T. Roach and A.I. Popov, 11
BREE; Shana. 12. 30 (1975)

J- W- Akitt. II QbanI.2991 Iaradax Iranai 1. 11. 1557
(1975).

C. Deverell,

Bragri NEE 2222222221. 1. 278 (1969)-

J..Kondo and J. Yamishita, 21 Rhys. SEER; §Qli§§a.19.

245 (1959).

D. Ikenberry and T.P. Das, 11.6ngm1 Enyg1, 16, 2199
(1965).
D. Ikenberry and T.P. Das, 11 6n6m1 2h261, 66, 1361

(1966).

A.K. Covington and K. E. Newman, In: Modern Aspects of
Electrochemistry, Vol. 12, Eds. J.D. Bockris and B.E.
Conway, Plenum Press, New York (1977).

148.

149.

150.

151.

152.

153.

154.

155.

156.

157.

158.

159.

160.

161.

162.

163.

164.

165.

223

N- Bjerrum. Kali Dani Yidanski Salaki Matizzxai Maddi.

~1, No. 9 (1926).

A. I. Mishustin and Y. M. Kessler, 61 691n1 66661, 1,
779 (1975).

J. W. Akitt, " Multinuclear NMR ",
Press (1987). '

T.P. Young, Y.C. Wu and A.A. Krawetz, 01229221 Egzgdax
£921. 21. 37 (1957)-

Y. c. Wu, N. B. Smith and T. P. Young, 11 BREE; gngm1,

Chapter 7, Plenum

62, 1868 (1965).

J.H. Stern and G.W. Anderson, 11 £hx§1 gn6m1, 66, 2528
(1964).

R.H. Wood and R. W. Smith, 61,2h261 gngm1, 62, 2974
(1965).

M.S. Stakhanova, V. Vasilev and Yu. A. Epikhin, 36661

11 beaI 98:81. 21. 182 (1953)-

C. Deverell and R. E. Richards, u611 EDYRa. 16, 421
(1969).
6- Kabisch and M. Klose. 21 28221.9bania. Leipzig.

266.

V.I. Baranovskil, T.T. Ermolaeva, V. I. Chizhik, 0. V.

Sizova and N.V. Ivanova, 6621 11 gngm1 £h261, 6, 861
(1987).

I.S. Perelygin, 6661 11 6663661, 13, 360 (1962).

I. S. Pominov and A. z. Gadzhiev, 1121,261611 61113:,
No.6, 19 (1965).

I. S. Pominov and T. E. Pavlova, zngrngl 2213166661
SRQKEIQEBQRIII. 11. 277 (1959)-

J.A; Pople, W. G. Schneider. and H. J. Bernstein, High
-resolution Nuclear Magnetic Resonance, Chap. 15. (New
York, McGraw-Hill Book Co.) (1959).

687 (1985).

C. E. Schafter and C. K. Jorgensen, 61 1n6161 H3211
Shani. 2. 143 (1958)-
Y.-C. Lee, Ph.D. Thesis, Michigan State University,

East Lansing, Michigan (1983).

A. Ellaboudy and J. L. Dye,

11.881 Shani 292;. 125.
6490 (1983).

166.

167.

168.

169

170.

171.

172.

173.

174.

175.

176.

177.

224

S. Mooibroek, R. E. Wasylishen, R. Dickson, G. Facey

_and B.A. Pettitt, 11 M661 366661, 66, S42 (1986).

M. S. Greenberg and A.I. Popov, 11 661n1 66661, 6, 653

(1976).

C.J. Pedersen, 11 231 ghgm1 6661, 22, 386 (1970).

W. H. Watson, J. Galloy, D. A. Grossie, E. vagtle and
W.M. Miller, 61 6:61 6h6m1, 162, 347 (1984).

J-M- Lehn. Strnainra Banding. 1a. 1 (1973)-

D.L. Johnston and W.D. Horrocks, 182191 thm;. 12.
(1971).

687

R.W. Kluiber and G. Sasso, 226

(1970).

Inargi China Astai. A.

M. Herceg and R.
435 (1970).

WeiSS. Indrsi 82911 dbani Latai. a.

D. Live and S. I. Chen, 61 Am1 66661 6661, 26, 3769

(1976).

299

A. Hourdakis and A. I. Popov, 11(661n1 £hgm1. 6.

(1977).

N. Ahmad and M. C.
(1977). '

Day, 21 Ana §h§m1 fiflfia. 22. 941

E. Schmidt, J.-M. Tremillon, J.-P. Kintzinger and A.I.
POPOV. 11.8n1 98981 2991. 125. 7563 (1983)-