're:—r-' LIBRARY

Michigan State
University A

 

This is to certify that the

thesis entitled

SPECTROSCOPIC STUDIES OF POTASSIUM SALT SOLVATION

AND COMPLEXATION IN VARIOUS SOLVENTS
presented by

Jeny-Shang Shih

has been accepted towards fulfillment
of the requirements for

Ph . D degree in Chemistry

 

 

ill/QM %/

Major professor

Date July 25 1978

 

0-7 639

© 1978

JENY-SHANG SHIH

ALL RI GHTS RESERVED

SPECTROSCOPIC STUDIES OF POTASSIUM SALT SOLVATION
AND COMPLEXATION IN VARIOUS SOLVENTS

BY

Jeny-Shang Shih

A DISSERATION

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

DOCTOR OF PHILOSOPHY

Department of Chemistry

1978

ABSTRACT

SPECTROSCOPIC STUDIES OF POTASSIUM SALT SOLVATION
AND COMPLEXATION IN VARIOUS SOLVENTS

BY
Jeny-Shang Shih

Potassium-39 and carbon-13 NMR measurements were
applied as a sensitive probes in the studies of solvation
and complexation of potassium salts in nonaqueous solvents.

In most cases, the chemical shifts of 39K were found
to be dependent on the concentration of K+ ion, which is
indicative of the formation of contact ion pairs. The
contact ion pairing formation was found to be dependent
on the dielectric constant and also on the donicity of the
solvents. Ion-solvent interaction was also studied by the
extrapolation of the chemical shift to infinitely dilute
concentration. In general, a correlation was observed
between the 39K infinite dilute chemical shift with the
Gutmann donor numbers of the solvents. A linear relation
between the infinite dilution chemical shifts and the
atomic numbers of alkali metals was found in some solvents
such as acetonitrile, dimethylsulfoxide and nitromethane.

Preferential salvation of K+ ion in binary mixed
solvents was studied qualitatively and quantitively

Jeny-Shang Shih

by 39K NMR. The geometric equilibrium constant, K1/n

and the free energy of preferential solvation were obtained
for each system by using Covington's treatment.

Complexation of potassium ion with macrocyclic
cryptands and with crown ethers was investigated in
several nonaqueous solvents by potassium-39 and carbon-
13 NMR. The 39K chemical shifts for the K* cryptate
0222 was found to be solvent independent which is
indicative of the formation of the inclusive complex.

The cryptand 0221 was found to form stable complexes with
K+ in some nonaqueous solvents, but the solvent dependent
39K chemical shift for K+0221 complexes seems to suggest
the formation of exclusive complexes. Carbon-13 NMR
studies seem to indicate that the cryptand 0221 forms
less stable complexes with K+ than the cryptand 0222.

The complexation of cryptand 0211 with K+ in
various solvents was studied by 39K NMR. The stabilities
of K+0211 complexes among these solvents are in the
order: acetone > ac etonitrile > pyridine > dimethylformamide
:>dimethylsulfoxide. The cation selectivities of cryptands
in nonaqueous solvents were monitored by 0-13 NMR.

The complexation reactions of K+ with some macro-
cyclic crown ethers, such as 18-crown-6, dibenzo-18-
crown-6. 15-crown-5, monobenzo-15-crown-5 and 12-crown-

A were investigated in various solvents by potassium-39

and carbon-13 NMR. The K+-18-crown-6 complexes were

Jeny-Shang Shih

found to be quite stable in nonaqueous solvents. The
stabilities of K+-18-crown-6 complex decrease in the

order: acetone > dimethylformamide > water >dimethylsulfoxide.
No ion-pairing formation was found between the complex
K+-18-crown-6 and the anion.

In the complexation study of 15-crown-5 with K+ both
1:1 and 2:1 sandwich ligand/K+ complexes seem to be formed
in all nonaqueous solvents used. The stability constants
of 1:1 complexes in these solvents were always large. The
solvent effect on the stability of 2:1 sandwich complexes
is in the order: nitromethane3>acetoneI>pr0pylene carbonate
> pyridine > acetonitrile >methanol > dimethylformamide >
dimethylsulfoxide, which with the exception of pyridine,
follows the inverse order of the donicities of these solvents.
In the 2:1 complexes, the anions were insulated from the
action of the solvent and the cation.

The complexation reactions of 12-crown-h with K+ in
various solvents were also studied by the same technique,
In most cases no evidence was found for the formation of
2:1 sandwich complexes and the 1:1 complexes seem to be
quite weak. Solvents influence on the stabilities of the
1 :1 complexes are in the order: acetonitrile}. acetone >
nitromethane > methanol >dimethylsulfoxide.

Finally, a recovery process for the cryptand 0222

and 0211 from cryptates was develoPed.

ACKNOWLEDGEMENTS

The author wishes to thank Professor Alexander I. POpov
for his guidance, counseling and encouragement throughout
this study.

The special contributions of Professor Michael J. Weaver
as second reader are appreciated. Many thanks also go to
all the members of Dr. A. I. Pepov's research group for
many months of friendship, and good times. And to Dianne,
Fred, Mojtaba and Dale, special thanks for their friendship,
discussion and stimulation.

Gratitude is also extended to the Department of
Chemistry, Michigan State University and the National
Science Foundation for financial aid.

Finally, I would like to acknowledge my parents and
my wife for their constant encouragement and cordial

consideration.

ii

TABLE OF CONTENTS
Chapter Page
I HISTORICAL REVIEW
INTRODUCTION................................... 1
(A) STUDIES OF IONIC SOLVATION AND ASSOCIATION
BY NMR..................................... 2
(B) POTASSIUM NUCLEAR MAGNETIC RESONANCE....... 8
(c) MACROCYCLIC CROWN ETHERS AND CRYPTATES..... 17
(a) MACROCYCLIC CROWN ETHERS...............17
(b) MACROBICYCLE CRYPTANDS................ 2#
III EXPERIMENTAL PROCEDURE
(A) INSTURMENTAL............................... 28
(B) CHEMICAL SHIFT MEASUREMENTS................ 31
(C) POTASSIUM SALTS............................ 34
(D) SOLVENTS................................... 3A
(E) PURIFICATION OF CRYPTANDS AND CROWN ETHERS. 35
(F) DATA HANDLING.............................. 36
III POTASSIUM-39 NUCLEAR MAGNETIC RESONANCE STUDIES
OF IONIC SOLVATION AND ASSOCIATION OF POTASSIUM
SALTS IN VARIOUS SOLVENT
INTRODUCTION................................... 37
(A) IONIC SOLVATION AND ASSOCIATION STUDIES OF
THE POTASSIUM IONS IN NEAT SOLVENT......... 37
(B) IONIC SOLVATION OF THE POTASSIUM ION IN
MIXED SOLVENTS............................. 72

iii

Chapter Page
IV POTASSIUM-39 AND CARBON-13 NMR STUDIES OF
POTASSIUM SALT COMPLEXATION IN VARIOUS SOLVENTS
INTRODUCTION.................................. 89
(A) COMPLEXATION OF THE K+ IONS BY CRYPTANDS.. 89
(B) COMPLEXATION OF THE K+ IONS BY CROWN

ETHF’RSOOOOOOOOOCOOOOOOOOOOOOOOOOOOOOOOOOOOO113

V RECOVERY OF CRYPTAND FROM CRYPTATE...............154

APPENDICES.............................................l64
APPENDIX I.......................................164
APPENDIX II......................................169
APPENDIX III.....................................l71
APPENDIX IV......................................174
APPENDIX V.......................................178

REFERENCESOOOOOOOOOOOOOOOO0..O...OOOOOOOOOOOOOOOOOOOOOO183

iv

Table

10

11

12

13

LIST OF TABLES

Page

Nuclear preperties of potassium isotopes........ 1O
Diameters of selected cations and macrocyclic
polyether cavities.............................. 20
Magnetic suceptivities corrections to the 39K
chemical shifts..................................33
39K chemical shifts Of potassium salt solutions. 39
Key solvent prOperties.......................... 46
The ion-pairing formation constants and limiting
chemical shifts of potassium in various solvents 61
Potassium-39 chemical shifts at infinite

dilution in various solvents................... 6#
39K chemical of potassium salt solutions in

mixed solvents................................. 7h
Summary of isosolvation point data for potassium
salts in the Binary solvent mixtures .......... 77
The equilibrium constants and free energy change
in the mixed solvents.......................... 85
The chemical shifts and line widths of potassium
~39 of K+0222 complexes at 0.5 mole ratio
(case/K+)....................................... 93
Carbon-13 chemical shift of potassium cryptates
0222............................................ 95
The change in 130 chemical shift of cryptand

CZZZ upon complexationOOOOOOOOOOOOOOOOOOOOOO...O 99

V

Table
1A

15

16

17

18

19

20

21

22

23

24

Page
Chemical shifts and line widths of K-39 of
K+-0221 complexes at 0.5 mole ratio (cam/16).... 103
Carbon-13 chemical shifts of potassium
cryptate 221.................................... 104

Mole ratio study of cryptand 0211 complexes

'ith KPF6 in various SOlVGDtS............... ....1O6

Formation constants and limiting chemical

shift for complexation of KPF6 by 0211...... .... 109
Carbon-13 chemical shift (Appm) of Li, Na

and Cs cryptates............................ .... 1‘0
Mole ratio studies of crown ethers complexes

with KPF6 in various solvents............... .... 11h
Formation constants and limiting Chemical

shift for the complexation of KPF6 by

18-crown-6 in various solvents................... 12‘
Formation constants of complexes of KPF6

with dibenzo-18-crown-6 in various solvents ..... ‘25
The elemental analysis for (1S-crown-§)2 KPF6
sandwich complex................................. 129
Formation constants and limiting chemical

shifts for 1:1 and 2:1 15-crown-5-K+ complexes

in various solvents.............................. ‘33
The limiting chemical shifts of 39K NMR for

the complexes of potassium salts with 15-

crown-S in various solvents...................,..,137

vi

Table

25

26

27

28

29

31

Elemental analysis of (monobenzo 15 0 5)2
KPF6........................................
Formation constants of complexes of KPF6
with monobenzo-15—crown-5 in various
solution....................................
The 0-13 NMR chemical shifts of M+-12-crown
-4 complexes in various solvents (M+ =

K+, 08+)....................................
Formation constants for the 12-crown-lt-K+
complexes in various solvents...............
Limiting chemical shifts and line widths

of K-39 for the complexation of KPF6 by
various macrocyclic ligands in acetone.....
Concentration of Na+ and 0222 in elution
solutions .................................

Carbon-13 chemical shifts of protonated

cryptandSOCOOOCOO...00......OOOOOOOCOOOO...

vii

Page

139

141

um

1A7

152

158

160

Figure

10

11

12

LIST OF FIGURES

Page
Structure of crown ethers .................... 18
Cryptands 0222, 0221 and 0211 (with internal
diameters) ................................. 25
The configurations of cryptand C222 .......... 26
The structures of cyclindrical macrotricyclic
cryptands..................................... 27
Potassium~39 NMR resonance of 0.005 M KPF6
in acetonitrile by Bruker 180 spectrometer.
(1000 scans, 15 minutes, line widtha~10 Hz)... 30
K-39 chemical shifts of potassium salts in
water and formamide....................... .... 45
K-39 chemical shifts of potassium salts in
dimethylformamide and dimethylsulfoxide .. ... #3
K-39 chemical shifts of pOtassium salts
in prepylene carbonate and formic acid......... 50
39K chemical shifts of potassium salts in
methanol and acetonitrile...................... 52
39K chemical shifts of potassium salts in
acetone and ethylenediamine.................... 53
39K chemical shifts of potassium iodide in
various solvents.......................... .... 55
39K chemical shifts of potassium thiocynate

in various BOlventS 0.000000000000000.....ooo 57

viii

—-'....~._.-

Figure Page

13

m

15

16

17

18

19

20

21

22

23

39K chemical shifts of potassium hexafluoro-
phosphate in various solvents............... ..... 58
The temperature dependent ionic association

Of potassium salts in water and acetone...... .... 62
The plot of the infinite dilution chemical

shift vs the Gutmann donor number............ .... 65
The range of infinite dilution chemical shifts
between nitromethane and pyridine for 23Na,

39K and 133Cs resonance...................... .... 68
The plot of infinite dilution Chemical shift

vs atomic number of alkali metal ions............. 71
39K chemical shifts of KPF6 in the Binary

mixtures of acetone with nitromethane, aceton-
itrile, water and pyridine........................ 76
39K chemical shifts of KPF6 in the mixtures

of acetonitrile with nitromethane, acetone

and water......................................... 78
39K chemical shifts of KPF6 in the acetonitr-
ile-prOpylene carbonate mixtures.................. 79
39K chemical shifts of KSCN in the mixtures

of DMSO with acetone, water and ethylene-
diamine........................................... 81
39K chemical shifts of KI in mixtures of

methanol with water and ethylenediamine........... 32

Convington plot for pyridine-acetone mixtures..... 86

ix

Figure
2A

25

26

27

28

29

30

31

32

33

3A

35

36

Page
Convington plot for the acetonitrile-
acetone mixtures............................... 88
39K NMR spectra of KPF6-cryptand 0222 solution..91
‘30 NMR spectra of K*-cryptand C222 in acetone. 97
Potassium-39 spectra of potassium-0221 cryptate
in various solvents; [0221]: 0.01 M[KPF6]= 0.02M 100
39K chemical shift vs mole ratio of 0211/K+
in various solvents........................... .107
39K chemical shift vs mole ratio of 1806/K+
in various solvents.......................... . 119
Chemical shifts of 18 0 6 K+ complexes as
function of concentration..................... .122
39K chemical shifts vs mole ratio of dibenzo
1806/K+ in various solvents................... .12#
59K chemical shift vs mole ratio of 1505/1:+
in nitromethane, acetone, methanol and
acetonitrile.................................. 12?
130 spectrum of 15-crown-5 at mole ratio 0.75
of K+/1505 in acetone........................ . 128
39K chemical shift vs mole ratio of 15C5/K+
in pyridine, propylene carbonate, dimethyl-
formamide and dimethylsulfoxide................ 131
Compurter fitting for the chemical shifts vs
mole ratio of 1505/K+ in methanol.............. 132
carbon-13 chemical shift vs mole ratio of

K+/1505 in various solvents.................... 136

3t

Figure

37

38

39

to

41

he

#3

AA

#5

#6

Page
39K chemical shift variation with the
concentration of 1:1 (1505) KPF6 complex
and potassium salt in acetonitrile............ 138
39K chemical shifts vs mole ratio of MB 1505
/K+ in nitromethane and acetonitrile.......... 140
39K chemical shift vs mole ratio of 12-crown
-l+/K+ in various solvents..................... 1h}
Carbon-13 chemical shift vs mole ratio of
K+/1204 in various solvents...................145
Carbon-13 chemical shift vs mole ratio of
Cs/ligand in methanol........................ 1A9
Mole ratio-59K chemical shift study for
various ligands in acetone....................151

Diagram for recovery of cryptand from

cryptateooooeoooooooooeoooocoo-00000000000 0000155

Elution curve for separation Of 0222H22+

from Na+oooooooooooooooooooooooeoooooooooe ....157
Carbon-13 sepctra of protonated and free

cryptands-coono...oooooooooooooooooooooooo c.0016]

1H spectra of protonated and free cryptands...162

xi

LIST OF ABBREVIATIONS

AC(M6200): Acetone
NM(MeN02): Nitromethane
P0: PrOpylene carbonate
ACN(Me20N): Acetonitrile
DMF: N-N—Dimethylformamide
DMSO: Dimethylsulfoxide
HFor: Formic acid
ForNHa: Formamide

PY: Pyridine

EN: Ethylenediamine
MeOH: Methanol

xii

CHAPTER I
HISTORICAL REVIEW

INTRODUCTION

Potassium and Other alkali metal ions play important
roles in chemistry as well as in biological system. However,
ionic interaction and binding process of these ions in
solution and especially in nonaqueous solvents still
remain largely unknown. The techniques usually applied to
study complexation reactions and ionic solvation are
potentiomtric methods using cation-selective electrodes,
calorimetry and conductometry.

Recently several new techniques which are very sen-
sitive to change of ionic enviroments in solution have
been deveIOped. One of the most important techniques is
Fourier-transform nuclear magnetic resonance. FITNMR
method is now used to yield the qualitative information
about the solvent-solute, solvent-solvent and solute-
solute interactions, as well as quantitative data concerning

ion-pair formation and complexation in alkali metal solutions.

HISTORICAL REVIEW

(A) STUDIES OF IONIC W

Since both the chemical shift and the relaxation
time of the nuclear resonances are very sensitive to
change of ionic enviroments in salution, the nuclear
resonance spectrosc0py method has become a very powerful
tool for the investigation of bath ion-ion and ion-
salvent interactions in alkali solutions.

The NMR studies of ionic salvation and association
of alkali salt solution can be performed using alkali
metal nuclei such as 7Li, 23Na, 39K, 87Rb and 133Cs,
anion nuclei such as 19F, 3501, 3‘Dr and ‘271 and the
solvents nuclei such as 130, 1H, th and 170. There are
many studies of ionic salvation and association in alkali
salt solutions performed by proton (1H) NMR (1-5) of the
solvent. However, relatively few studies an other nuclei
have been reported.

The pioneering NMR studies of ionic association in
alkali solutions were done by Shoolery and Alder (1).
They studied the ionic association in the HF aqueous
solution and reported that 19F chemical shifts of KF were
dependent on the concentration of potassium fluoride
aqueous solution and suggested some small amount of ion-
pairing formation. The same system was carefully studied
by Connick and Paulson (6). They reported that the 19F
chemical shift of KF aqueous solution moved down field

3

with increasing the potassium fluoride concentration at
the concentration <15.0‘M, and moved to high field at
higher concentration. It was interpreded in terms of the
formation of the solvent separated ion pairs at lower
concentration while the formation of the contact ion-
pairing formation at higher concentration. Recently,
DeWitte and Papov (7) extended these 19F resonance studies
of alkali hexafluoraphosphate solutions to a wide range

of nonaqueous media. They reported that in solvents of
medium polarity and donicity such as propylene carbonate
and acetonitrile, the ‘9F chemical shift for NaPF6 moved
upfield with increasing concentration of the salt. The
behavior is indicative of anion-cation interaction. However,
potassium hexafluoraphosphate solutions do not show any
concentration dependence of 19F chemical shift, which
indicates the absence of ionic association in these solu-
tions.

The 19F chemical shift for potassium fluoride in
water-organic solvent mixtures was monitored by Carrington
and coworkers (8). The linear variation of chemical shifts
of 19F with mole fraction of solvent were observed in
mixtures of water with methanol and formamide. It indicated
no preferential salvation of F- ions in either mixtures.
Similar results for the same system were also obtained
by several authors (9-12). However, water was reported
to appear strongly prefered in the primary salvation

shell of F- ion over acetone and acetonitrile (9). The

h
effect of paramagnetic ions on the transverse relaxation

times of 19F of PF6- in aqueous solutions was Observed
by Stengle and Langford (12). The decrease in relaxation
time of 19F was interpreted in terms of association
between PF6- and the paramagnetic ion.

The chemical shifts of the nuclear resonance of
3501, 81Br and 1271 in aqueous solutions of alkali
halides were studied by Deverell and Richards (13). They
reported that the magnitude of the shift of halide
nuclear resonance increases with increasing atomic number
of the halide ion and generally the shielding by solvent
showed dependence upon the partner cation: Na+>K+ >Li+>
Rb+2>Cs+ while the order of magnetic solvent shielding
effects produced by the anions was always faund to be
in the order I"< Br"< Cl"<E"< N031 Stengle, s; g; (11.)
monitored the 35Cl, 79Br and 1271 resonances of alkali
halide solutions in water and in mixed solvents of water
with methanol, acetonitrile, DMSO and DMF. The chemical
shift of halide nuclear resonance was found to be
strongly dependent on the solvent while the dependence
on the cation Li+, NaI, K+ were comparatively weak.

The NMR behavior of 01' and I" ions in mixtures of DMSO
with water indicated that there was no strong preference
for either solvent. However, the strong preferential
salvation of halide ions of water over acetonitrile

was observed.

Recently, 3501 was used to study quantitively the

salvation of LiCth in acetone-nitromethane mixtures by
Papov and Baum (15). In that work, the linewidth of the
35Cl resonance was monitored as a function of the
acetone/Li+ mole ratio. They reported that the solvation

number of Li+ by acetone is 4. The relaxation time of
35

Cl nuclei has been used to study the ionic association

of alkali chloride solution. Berman and Stengle (16) mani-
tored the spin-lattice relaxation time (T1) of 35CI

to study the ionic association behavior of the 010,+

ion with Li+ in DMF, acetonitrile and DMSO.

Another anion resonance such as 170 and 14N NMR can
be applied to the investigation of ionic salvation (17).
However, the use of ‘4N and 170 NMR are limited by the
need for special instrumentation due to their law magnetic
moments.

Alkali nuclei resonances such as 7Li, 23Na, 39K,
87Rb and 133Cs have also been applied as sensitive probes
for the study of ionic association and salvation of
alkali salt solutions. A review of the alkali nuclei
resonance in alkali salt solutions was published by
Papov (18).

Maciel and coworkers (19) monitored 7L1 chemical
shiftsof LiBr and L10104 in several nonaqueous solutions.
A linear correlation between chemical shifts of 7L1 and
Kosawer's 2 values (20) 0f the solvents was observed.

Recently, Papov and coworkers observed the chemical

shifts of the 7L1 resonance of Lithium salts in eleven

6

nonaqueous solvents (21)(22). They reported that in dimet-
hylformamide and lesser extent in prOpylene carbonate,
methanol and dimethylsulfoxide solution, the 7L1 chemical
shift were essentially independent of the counter ion
and the salt concentration. The results indicated either
that the salts were largely dissociated or that they -
exist in the form of solvents separated ion pairs.
Mishustin and Kessler (23) studied the lattice spin rela-
xation times (T1) of 7L1 of lithium salt solutions for
the investigation of the ion-solvent and ion-ion interaction.
They reported the linear correlation between the relaxation
time and Gutmann's donor number (24) of the solvents. The
donor number of the solvents was defined as the negative
enthalpy of the reaction between the solvent and antimony
pentachloride in 1, 2 dichloroethane solution. Similar
studies were made by Hertz and coworkers (25)(26), and
similar results were obtained. The ionic association of
organometallic lithium compounds such as LiAlMen was also
studied using 7Li resonance by Covington and coworkers (2?).
Recently, 7Li has been used to study the preferential
salvation of Li+ in mixed solvents. Covington, gt_gl (28)
studied the 7L1 resonance of LiNO3 in DMSO-H20 mixture.
They found that N03" ion are preferentially solvated by
DMSO and Li+ by water.

Sodium-23 NMR has been extensively investigated and
a comprehensive review of the literature available to

late 197A 18 presented in the Ph.D thesis of M.S Greenberg

(29). Kessler, 23’s; (30) studied the spin-lattice
relaxation time (T1) of 23Na and 133Cs of alkali salts
in eleven nonaqueous solvents. In the 23Na case they
also found that values of 1/T1 correlated with Gutmann's
donor numbers for these solvents (24) with the only
exception being H20. However, the correlation was less
markedly in the case of 08+. In that work, values of
donor number used for methanol (31) and water were 19
and 18 respectively. In a study of 23Na resonances of
sodium salts in different solvents, Bloor and Kidd (32)
found a correlation between thepKa of the solvent and
the chemical shift of the cation.

or the heavier alkali nuclei (K, Rb, Cs), the 133Cs
nucleus is the easiest to study by NMR technique. The
siganals were reported to be strong and sharp for the
cesium samples in the concentration range of 10'3‘M (33).

Halliday (34) reported that 133Cs chemical shift
changed nonlinearly with Cs salt concentration in water
and several nonaqueous solvents and the degree of nonli-
nearity depended upon the dielectric constants of the
solvents, the ‘33Cs shifts were explained in terms of
the cation-anion interaction. Recently Papov and DeWitte
(33) studies the 13303 chemical shift of cesium salts
in water and eleven nonaqueous solvents. In all cases,
the chemical shifts were found to be strongly concent-
ration dependent indicating some contact ion pair

formation. In that work, the chemical shift of the Cs+

at infinite dilution have been determined in these
solvents. Comparison of the limiting chemical shifts
with Gutmann's donor numbers for these solvents in
general shows an adequate but not straight line correlat-
ion as was observed with solvent-dependent chemical
shifts of the 23Na resonance (29)(35).

Since 87Rb nucleus has a large quadrupole moment,
the natural line width for 87Rb is large which limit the
use of 87Rb resonance for study on the ionic salvation
and association. The studies of the ionic association of
rubidium salt solution in water was reported by Deverell
and Richards (36). A very broad 87Rb signal (line width
«~400H2) for rubidium salts in dimethylsulfoxide have
been observed by Crawford and Gasser (37). Recently,
Neggla, 23's; (38) monitored the spin-lattice relaxation
time of 87Rb in methanol-water mixture. The Rb+ were
found to be preferentially solvated by methanol.

Carbon-13 NMR also can be applied to study the
salvation of cation. Stockton (39) monitored the 130
chemical shift of ethanol in A1013 ethanol solution as
a function of temperature, salvation shell signal appeared
below -20°, the 0H2 carbon showed three salvation signals.
At least two 0H3 carbon salvation shell signals were
discernible.

(B) POTASSIUM NUCLEAR MAGNETIC RESONANCE

There are three magnetic potassium nuclei is: 39K

40K and l+1K. The nuclear properties of these nuclei are
listed in Table 1. ,

As shown in the Table 1, all the potassium nuclei
(39K, “0K and l”K) have very low sensitivities and low
frequencies as compared to 1H. The sensitivity (S) is
function of natural abundance (No), magnetic moment (II)
and nuclear spin (I) by the relation: S = No u3(I + 1) /
IZ- Although 39K has a high natural abundance, it also
has a very low magnetic moment which leads to low sensit-
ivity. Both 40K and l”K nuclei have a low natural abund-
ance and a low magnetic moment. Due to the low sensitiv-
ities and low frequencies of potassium nuclei, the study
of potassium nuclei resonance is largely determined by
the sensitivity of the spectrometer. In the past rather
sparse studies on the potassium resonance have been reported.
In most cases, the measurements were performed at relat-
ively low fields (<(13 kilogauss) and continuous wave
techniques were used, and therefore they were limited to
solutions of fairly high concentration (:>0.2‘M) (36)(h0-A4).
Fortunately, due to the deveIOpment of Fourier transform
NMR technique and the availability of high field super-
conductive solenoids, potassium NMR studies are becoming
quite feasible.

0f the potassium nuclei, 39K has the highest sensit-
ivity and the signal of 39K resonance is easist to observe.
lFirst 39K resonance signal was observed at 1.59 M0 at a

10

Table 1. Nuclear PrOperties 0f Potassium IsotOpes

 

 

39K AOK 41K
Nuclear spin 3/2 4 3/2
QutrOpole moment 0.055 -0.07 0.067
(barn)
Magnetic moment<a) 0.390 -1.296 0.214
(llN)
Natural abundance 93.1% 0.012% 6.88%
Sensitivity(b) h.7 x 10‘# 6.2 x 10"7 5.8 x 10"6
Resonance at A.23 Teslgc) 8.4 MHz 10.4 MHz A.6 MHz

 

(a) The magnetic moment of potassium ion purely surrounded
by water molecules.

(b) Referred to the proton NMR signal of H20 at the same
magnetic field and with same probe volume.

(0) One Tesla = 10 kilogauss magnetic field. At the field
of A.23 Tesla, H resonanate at 180 MHz.

11

field of 8000 gauss by Collins (#0). He obtained the
ratio of the magnetic moment of 39K to proton:

u /'u = 0.13999. The value of I1 was found to be
39K 1H 1H

2.79268 IIN (£11). The value of u 39K was calculated to be
0.3909A i 0.000? 11N. Brun, gt al.(42) measured the
nuclear magnetic moments of 39K, 41K, 89Y and 109Ag at a
magnetic field of about 9000 gauss. Resonances of the two
potassium isotOpes 39 and #1 were obtained in a 15 molar
aqueous solution of potassium formate. For 39K the
magnetic was found to be 1“ 39 = +0.390873 : 0.000013 #N
in a agreement with the value of Collins's (A0). The signal
of 41K appeared with a signal to noise ratio of about
3:1. The ratio of the frequencies of the two isotOpes is
”#1 / 1139 = 0.50886 : 0.00008.

Hindman (A3) studied the potassium-39 resonance in
aqueous solution of potassium chloride. The 39K chemical
shift was found to be independent of the concentration of
potassium chloride. Deverell, 23.2; (36) monitored the
chemical shift of 39K in aqueous solution of potassium
halides and nitrates from concentration of 0.2 molar
up to saturation. The chemical shifts of the 39K nuclear
resonance in salt solutions were found to depend strongly
on the salt concentration in a non-linear manner. For
all the salts studies, however, the chemical shift does
vary linearly with the mean activity of the salt. This
suggests that the shifts are strongly influenced by

12

interactions between K+ ions with other ions. The importance
of ionic interactions is also shown by the strong dependence
of the chemical shifts on the nature of the anions present.
The authors also found that the effect of different anions
on the 39K chemical shifts varied in the order I" > Br' >
Cl'>>F'2>N03'. They suggested that it was caused by direct
collisional interactions between the cations and the

anions.

Bloor and Kidd (AA) monitored the 39K chemical shifts
of several aqueous potassium salt solutions. They reported
that in all cases the chemical shifts of 39K resonance
varies linearly with the salt concentration. The parama-
gnetic shielding Was decreased by the counter ions in the
order: I' > Er" > CN“ > P02- > OH‘ > 01' > cm" > 00%” > CNs'>
CH3000' > F" >N3" > N02' = H20 > S012; > Oral?" > N03" > Crzogn
The shift was explained in terms of the short range over-
lapping of the outer electron orbitals of the anion and
cation during random ionic collisions. The magnitudes of
the chemical shifts were shown to be directly prOportional
to the effectiveness of this overlap interaction.

Recently, the signal of the rare isotope #oK was
detected for the first time by Sahm and Schwenk (45). In
their work, the NMR lines of 39K and “K have also been
investigated in solution of some potassium salts in H20,
.DZO, methanol and ethylenediamine. The potassium salt

concentration range used in their work was from 0.2%! to

13

saturation. The magnetic moments of the 39K, l+OK and l”K
ions completely surrounded by water molecules were founded
to be 0.391, 1.296 and 0.21A IIN respectively. The natural
line width of 39K resonance was reported to be 12 Hz. At
infinite dilution the spin-spin relaxation time (T2) for

39K in H20 is 56 msec. The ratio of T2(39K) / T2(“‘K) is
1.36. The shielding constant ((7) of the 39K, 40K and 1”K
ions by the surrounding water are -0.105 x 10‘}, 0.13 x 10‘},
and -0.105 x 10"3 respectively.

Some studies of spin-lattice relaxation time of 39K
resonance have been reported. Shporer and Luz (46) monitored
the spin-lattice relaxation time (T1) of 39K resonance to
study the kinetics of complexation of potassium ions with
dibenzo-18-cr0wn-6 (DB18C6) in methanol. They reported
that in the solvated form, the NMR relaxation rate of 39K
was relatively long (..o.o1 sec), while in the (K+DB1806)
complex the summetry around the K+ ion was considerably
reduced and at same time the correlation time was increased
resulting in a fast nuclear relaxation. The activation
energy of the complexation of x* with DB18C6 in methanol
solution was 12.6 Kcal / mole.

A few 39K resonance studies on the biological system
have been reported. Damadian and coworkers (A?) found
that the spin-spin relaxation time (T2) of 39K of potassium
.ions in packed bacteria was much shorter than free K+ in

320. They also studied (1+8)U+9) the spin-lattice relaxation

1L1

time (T1) of 39K of potassium ions on several types of nor-
mal tissues and cancers of rate and mice. They reported
that the 39K spin-lattice relaxation time (T1) for normal
tissues was on the average 24% longer than that for the
cancerous tissues. Both T1 and T2 of 39K in fresh excised
rate muscle and brain were much shorter than for 39K in
aqueous solution. The results were interpreted in terms

of the association of K+ with fixed charges on macromole-
cules. They also studied the interaction of K+ with Dowex
50 exchange resin (50). They found both T1 and T2 of 39K
resonance were shortened by the association of K+ with the
exchange resin.

Although potassium-39 has low sensitivity and low
resonance frequency, fortunately, the quadrupole moment of
potassium is not large (0.055 barn) and it is smaller than
that for 23Na or 87Rb. Therefore, the natural line width
of the resonance is small (<12 Hz) (115). As with all
alkali nuclei, the chemical shift of potassium-39 is domi-
nated by the paramagnetic term (0p) in Ramasy equation
(51). According to Ramasy' equation the screening
constant,<7, which determines the position of the resonance
as the sum of various diamagnetic (shielding) and paramag-
netic (deshielding) contribution:

0 = 0p + 0d. (1.1)

‘7ci is the diamagnetic component and 0b is the paramag-

netic component.'The theory of paramagnetic interaction

15

was proposed by Honda and Yamashita (52), who suggested
that the chemical shift of cations and anions in alkali
halide crystals arieses from the overlap repulsive
forces between the closed shell of the ions. These
forces cause the excitation of p orbital electrons of
the alkali nuclei to higher states, the result being

a decrease in the shielding of the nucleus. For potassium
nuclei, the most important contribution to the chemical
shift must come from the overlap between the 3p orbitals
of the potassium ion and the outer s and p orbitals of
anions or solvent atoms. The overlap repulsive forces
will cause the excitation of electrons from the 3p

to the 4p levels. This pertubation produces a
paramagnetic (down field) contribution to the chemical
shift. This shift can be written in the following

equation:

 

 

a =-16c112 ‘ 1 . 32 (1.2)
P < r3 ZP ‘AEnp

where S is the overlap integral between p orbital of
the potassium ion and the orbitals of the neighboring

ions. —l—>is the radial function which is the average

r3
over the outer p orbitals of the potassium, 1AE is the
mean excitation energy from the 3p to the 4p orbital,
2
a is a constant = ——2-° where m and e are the mass
2mc
and charge of a electron respectively.

Since Kondo and Yamashita's theory made

16

satisfactory calculation of the chemical shifts not only
for the alkali halide crystals but also for the hexa-
hydrated 87Rb+ ion relative to the free ion (53).
Deverell and Richards (13) applied this theory to the
interpret ion of the chemical shift in solutions. They
suggested that the chemical shift at concentration c,

relative to the free ion can be written as :

0: -16d2 1 1 (“(0) (c) )

<r3>np AE ion-solvent + A ion-ion

 

(1.3)
wherejlis appropriate sum of the overlap integrals
of the orbitals of the central ion and surrounding

solvent molecules or neighboring ions. The values of
1

<r3>nP
following order, 23Na(5.9) <39K(7.98) <_87Rb(13.8) <
133Cs(18.7). Therefore, if the extents of both ion-

 

/.AE for alkali metal nuélei increase in the

solvent and ion-ion interactions are about the same

for an alkali nuclei with same anion and the same
solvent, the 6P will increase in this order 1330s:>
87Rb3>39x§>23Na. This assumption consistent with their
experimental results. They reported that the magnitudes
of the chemical shifts increased considerably with
increasing atomic number of the cation. Recently Sahm
and Schwenk (45) reported that the shielding of alkali
nuclei'ig 23Na, 39K, 87Rb and 133Cs in water was a

17

nearly linear function of the atomic number.

(C) MACROCYCLIC CROWN ETHERS AND QRYPTATES
(a) MACROCYCLIC CROWN ETHERS

Since Moore and Pressman (54) reported that the
biological effects of some antibiotic substances such as
valinomycin, nonactin and monensin depended on the
presence or absence of specific alkali metal cations
in the medium and suggested that these antibiotic
substances acted as ion-carriers (ionophorses) across
membranes with different specificities for different
ions.

After this discovery, the complexation of alkali
metals with these naturally antibiotic carrier have
been extensively investigated.

In 1967, the macrocyclic crown ethers which
resemble the antibiotic ligand were synthesized by
Pedersen (55). Since that time, Pedersen has reported
the synthesis of over sixty macrocyclic crown ethers
and discussed their abilities to complex alkali metal
ions inside the two dimensional cavities (56). Several
polyethers are shown in Figure 1. Generally, the alkali
Inetal ions are regarded as poor complexing cations,
and complexing of alkali cations by neutral molecules
:is an uncommon phenomenon. Because of these unusual
complexation prOperties, many alkali salts can be

dissolved in non polar organic solvents by forming

. 0 o (\o f
(O O) @0119
VV 00w"

18 CROWN 6 DIBENZO 18 CROWN 6

0””. (To
COL/o) C 39

L/
15 CROWN 5 MONOBENZO 15 CROWN 5

/—\

on. (a s.>
(Rf) H's—s OJNH

L/
'2 GROWN 4 CRYPTAND 22

Figure 1. Structure of Crown Ethers

19

complexes with these macrocyclic polyethers. After
this discovery, crown compounds and their complexes

have been extensively investigated.

Crown compounds have been studied as model systems
in cation transport through cellular membrance (57-60)
They have found use in organic chemistry to study
certain chemical reaction including the catalysis of
ionic organic reactions by solvolyzing cationic
species (61-6h).

The stability and thermodynamic studies for most
macrocyclic crown ethers in water or methanol have
been reported. Several excellent reviews of the crown
and hetero-crown compounds and their interaction with
metal cations have recently been published (65-71).

The stability of crown other complex was reported (71)
to be dependent on the several important parameters
discussed below.

(1) Relative sizes of cation and ligand gayity

In general, these ligands complex most strongly
those metal ions whose ionic crystal radius best
matches the radius of the cavity formed by the ring
upon complexation (55). The ionic diameters and the
sizes of cavities of crown ethers are listed in Table 2.

Frensdoff (72) reported that the Optimum polyether
ring size being such that the cation Just fits into

"hole" was benzo-15-crown-5 for Na+, benzo—18-crown-6

20

for K+ and 21-crown-7 for 05+. However, the relative
size of cation to ligand cavity have been found to be
not the only parameter which influence the stability
of crown ether. For example, Izatt,'§£.§; (73) reported
almost no cation selectivity was seen for 15-crown-5

in water.

Table 2. Diameters of Selected Cations and Macrocyclic
Polyether Cavities a

 

 

 

Cation Ionic Polyether Diameter
diameter of gavity
A A

Lithium 1.20 All 1h—crown-4 1.2-1.5
Sodium 1.90 All 15-crown-5 1.7-2.2
Potassium 2.66 All 18-Crown-6 2.6-3.2
Ammonium 2.8h All 21-crown-7 3.A—4.3
Rubidium 2.96
Cesium 5.3#
Silver 2.52
Barium 2.70

 

(a) Reference 72

(ii) Type and charge of cation

In solution, for alkali and alkali earth metals,
the selectivities of crown others for K+ and Ea2+ are
generally higher than those for smaller and larger

cations. Since smaller ions like Li+ are so strongly

21

solvated that considerably more energy must be expanded
in the desolvation step than for larger ions like 06+,
on the other hand the larger cations are unable to
attract and organize the ligand as well as smaller ones.
In general, large dipositive ions often have higher
stability constants than monOpositive ions of similar
size. For example, potassium ion has about the same
diameter as the barium ion, however, the stability

2+-dicyclohexyl 18-crown-6

constant (log K) of Ba
(Ba2*-DC1806) complex is about 3.6, but it is about 2.0 (67)
for the potassium complex. The result is an evidence that
the binding of alkali and alkaline earth metal ions to
macrocyclic ligands is electrostatic in nature.
(iii) Type of donor atom

The substitution of nitrogen or sulfur for oxygen
in crown ether reduces the latter's affinity for alkali
ions (72), the stability constants falling in the order
of decreasing electronegativity 0>NR=>NH1>S. However,
the effects of N or S substitution of Ag+ (72) and
ng+ (71) complexing were exactly the apposite. The
results were explained in terms of the covalent bonding,
not electrostatic force between donor atoms and metal

ions in the Ag+ (7A) and Hg2+ cases.

(iv) Number of donor atoms
The increasing in number of donor atoms without

changing the size of the ring can enhance the stability

22

of the complexes. Cram (75) reported 18-crown-5 is a
much poorer host for t-butyl ammonium ion than is
18-crown-6.
(v) Substitution on the macrocyclic ring

The addition of the benzo group to the 18-crown-6
caused decreasing in the stability of potassium-18-crown
‘-6 complex in methanol was reported by Frensdoff (72).
The result was explained in terms of decreasing in
both cavity of ligand and electron density of oxygen.
Meanwhile the addition of benzo group also alter the
selectivity of the ligand. In methanol, the formation

2+ complex of 18-crown-6 is larger

constant of the Ba
than that of the K... complex by a factor of ten. Dibenzo
18-crown-6, on the other hand, displays the opposite
preference, binding K+ better than Ea2*(76).

(vi) Solvent effect

 

Papov,‘gt'al (77) investigated the complexation of
18-crown-6 with Cs+ in several nonaqueous solvents.
They reported that both 1:1 and 2:1 ligand/06+ complexes
are formed, previously only 1:1 complex was found in
aqueous solution (73). The stability for 2:1 complexes
increasing among the nonaqueous solvents in the order
DMSO < DMF€ P0 € PY < A0. The authors concluded that, in
general, the increase in the donor number of the solvents
caused a decrease in the stability of complex. The

solvent effect on the stabilities of complexes of alkali

23

metal ions with dibenzo-18-crown-6 (DB18C6) were also
reported by several authors (78)(79). Matsura and Sasaki
(78) reported that the solvent influence on the stabilities
of the 1:1 DB1806/M+ complex were in the order DMSOA<DMF<I
P0. Evans and Cussler (79) found that alkali metal ions
formed stronger complex with DB1806 in acetonitrile than
that in methanol.

The order of preference for alkali metal ions affected
by the solvent was reported by Wong and coworkers (61).
They reported that sodium in THF solution formed a
stronger complex with dimethyldibenzo-18-crown-6 than
potassium. Arnett and Moriarity (80) reported that stabil-
ities of the complexes of large cations were less affected
by solvent than those of smaller ones.

Several crystal structures of complexes of alkali
metal ions with some crown ethers have been reported
by several authors (81-85). The X ray studies of alkali
metal 18-crown-6 complexes have shown that K... sits exac-
tly at the center of plane of the oxygen atoms while the
Cs... and Rb+ ions were displaced from the mean plane (81-83).
The X ray studies (84) of sodium complexes of benzo-15-
crown-5 have shown that the Na+ ion occupies the center
of the plane of the oxygen atoms. A 2:1 sandwich 12-crown-4
/NaI complex have been found by X ray analysis (85). It
was also found (84) that, for the potassium complex of
dibenzo-30-crown-10 and apparently for other large ring

cyclic polyethers, the complex consists of a wrap-around

24

structure where all the oxygen atoms are approximately
equidistant from the potassium ion but not in the same
plane.
( b) MACROBICYCLIC CRYPTANDS

The diaZOpolyoxa macrobicyclics cryptands was
synthesized by Lehn, et a1 (86). The cryptands are the
strongest complexing agent presently known for alkali
metal ions in aqueous solution (the stability constant
up to 105 1 mole‘l). Several typical cryptands are
shown in Figure 2. The Optimum cryptand cavity size for
cations was found to be cryptand 0222 for K+, cryptand
0221 for NaI and cryptand 0211 for L1... (86). The crystal
structure of several metal complexes of cryptand 0222
have been determined by using X-ray crystallography (87)(88).
In all cases, it was found that the metal ion was located
in the cavity of the macromolecule and that the two
nitrogen atoms participated in bonding to the meatl atom.
However, recently. Mei,.g§‘al (89) monitored the 133Cs
NMR of Cs cryptand 0222 complex in various solvents. They
suggested the presence of two types of 1:1 complexes in
solution, an exclusive complex in which the ion may
interact with the solvent, and an inclusive complex which
has a solvent-independent chemical shift.

These cryptands can exist in the three configuration
(90) exo-exo, endo-endo and endo-exo isomers which are

shown in Figure 3. The endo configuration has the

25

c222 (2.8A°) 02219.21?)

C211<1.6A°)

Figure 2. Cryptands 0222,0221 and 0211 (with

internal diameters)

26

nitrogen lone pair electrons directed toward the interior
of the cavity while the exo configuration has the nitrogen
lone pair electrons turning outside. The endo-endo configu-
rations for KI-Cryptate 222 and Rb-Cryptate 222 were found
by X ray crystallographly (87).

NH/ \
JV NC'M1T 1A
C¥"N
\\é3‘_: ,/ (w m-—<a 9) ‘::E;~‘4D

exo-exo endo-endo exo-endo

Figure 3. The Configurations of Cryptand 0222

The stability constant of alkali cryptates in water
and in methanol solution was reported by Lehn and
coworkers (92). The cation selectivities of cryptands
for alkali metal ions in water and methanol were found
‘to be cryptand 0222 for K*, cryptand 0221 for Na+ and
cryptand 0211 for Li+. Papov and coworkers (89)(92)(93)
extended this study to nonaqueous solvents by using
alkali metal NMR methods such as 7Li, 23Na and 1330s
NMR. They found that the formation constant of metal
ion complex in nonaqueous solvents was always higher

than that in water, and the stability constant of

 

27

complex was dependent on the donicity of the solvents.
Dye and coworkers (94) isolated the alkali metal anions
such as (Na+0222)Na' which was investigated by 23Na NMR.

Recently, cyclindrical macrotricyclic cryptands
have been synthesized (95) and shown in Figure 4.

N/’\frfib/\\
C V’s/K1”)

(:3,lr")a"_\2f’~‘\vi::] a; v.5"
boy/0d _

N
3

Y-cu

ll

[J

....

V:
s.

u
IR
-<
I

Figure 4. The Structures of Cyclindrical
Macrotricyclic Cryptands

The formation of macrotricyclic cation inclusion
complexes (96) have been reported. Cyclindrical
macrotricyclic ligands are topologically well suited
for the designed positioning of two metal cations

in a binuclear inclusion complex.

CHAPTER II

EXPERIMENTAL PROCEDURE

 

EXPERIMENTAL

(A) INSTRUMENTAL

(a) DA-60 Multinuclear NMR spectrometer

A modified DA-60 Fourier transform multinuclear NMR
spectrometer was used to obtain the 39K chemical shifts
in the studies of ionic association in neat nonaqueous
solutions and the preferential salvation in mixed solvents.
The spectrometer consists of modified NMR speclities MP-
1000 spectrometer, a Nicolet 1083 computer, interface and
the necessary accessories such as disk system, plotter etc.

The 39K chemical shift was measured at 2.8 MHz with
the magnetic field 14 kilogauss (1.4 Tesla). The frequency
source of thejRF unit is a 56.4 MHz crystal controlled
oscillator. The 39K nuclear magnetic resonance was Operated
with pulsed width «'30011sec. External lock system (H20)
is used. A 25 ml NMR tube with 5 ml sample solution were
used. The saturated KNO2 (31 M) in D20 solution was used as
the external reference. Using the DA-60 spectrometer, the
signal for 0.01‘M of KPF6 in acetonitrile can be observed.

(b) Bpuker WH-JQQ NMR spectrometer

A Bruker WH-180 sulerconducting multinuclear NMR
spectrometer was used to study the complexation of
potassium and macrocyclic ligand such as crown ethers
and cryptands. The spectrometer consists of a
superconducting solenoid, a Nicolet 1180 computer disk

system and temperature control units.

28

 

29

The quadrature detector system is used in WH-180.

In conventional FT NMR, a single phase-sensitive detector
(such as used on DA-60) can only determine the magnitude
of the frequency difference between the signal and rf
pulse, but not the sign of this difference, so the rf
pulse is usually set at one end of the spectral region to
avoid folding back of the resonance. This gives rise to
two problems: (1) The power bandwidth of the rf pulse
must be at least equal to twice of the total spectral
width (is: rH1 227T(SN)) and (2) Noise from the unused
side of the decreasing S/N by a factor of V2: Quadrature
detection can solve both of these problems directly.
Since the rf power requirement varies as the square of
the spectral width, quadrature techniques bring an
effective gain of a factor 4.

The data memory available in the Nicolet 1180 computer
in Bruker WH-180 is 16 K while 8 K in Nicolet 1083 in
DA-60 spectrometer. The 39K resonance was Operated at
resonance frequency 8.403 MHz at the magnetic field of
42.3 kilogauss. A 20 mm NMR tube and 10 ml of sample
solution were used in this work. A saturated (31'M)

KNOB in D20 solution was again used as the reference
sample. The signal of 39K resonance for 0.005‘M of KPF6
in acetonitrile can be obtained in the short time

(‘~45 minutes) and is shown in Figure 5.
(c) Varian CFT 20 spectrometer

 

3O

391: ma

0.005 n were/A011

 

Figure 5. Potassium-39 NMR Resonance of 0.005‘3
KPF6 in Acetonitrile by Bruker 180
Spectrometer. (2000 scans, 15 minutes,

line width «~10 Hz)

31

All the carbon-13 NMR chemical shift measurement
was performed on a varian CFT-20 high resolution NMR
spectrometer with a constant field of 18.7 kilogauss.
The 130 NMR was Operated at resonance frequency 20 MHz.
One ml of sample for measurement of the 130 chemical
shift was placed in 5 mm inner tube of coaxial tube.
While the mixture of lock and reference sample (vol 50%
of acetone in D20) in 8mm outer tube.

(13) CHEMICAL SHIFT MEASUREMENTS

All the 39K chemical shifts reported are referred
to the infinite dilution chemical shift of the potassium
ion in water. The paramagnetic shift (down field), was
designated as negative values.

The chemical shifts reported here are also corrected
for differences in bulk diamagnetic suceptibility between
sample and reference (H20) according to the relationship
of Live and Chan (97) for the spectrometer such as
DA-60 where the applied polarizing magnetic field is
transverse to the long axis of the cylindrial sample,
the bulk susceptibility correction to the observed
chemical shift is shown to be:

2 ref
0 sample)

corr = dobs +3(Xv " Xv (11.1)

where X vr ef and Xvsample are the magnetic susceptibities

32

for reference and sample respectively. For high field
spectrometers such as Bruker WH-180 where the polarzing
magnetic field is along the long axis of the sample.

The correction for the observed chemical shift is given by

ref sam 1e
doorr = bobs "%'(Xv ' ‘XV' p ) (II'Z)

Since Templeman and Van Cost (98) reported that
the contribution of the salt to the susceptibility of
the solution is very samll (for 9.65‘M NaI solution<<
0.1 ppm), no correction for the contribution of the
salt was applied. The respective susceptivity corrections
for the solvents, for both DA-60 and WH-180 spectrometers
are shown in Table 3.

For the mixed solvents, the volume diamagnetic
susceptivity of a given mixtures was calculated by the

Wiedemann's equation

 

 

‘ v (II.3)

VA + VB

where X vmix is the calculated volume susceptibility

of the mixture. Xk and tXB are the volume susceptibil-

ities of pure solvent A and B respectively, VA and VB

are the volumes of solvents of A and B respectively.
All the 0-13 NMR chemical shifts reported here

are referred to TMS resonance. The paramagnetic shift

(down field) was conventionly designated as positive

values.

 

33

mm monououom any

 

 

a..o+ no.0- mmm.o onasmasoocasmem
w:.o+ :m.or omw.o avfixomammahnaosdn
m:.o+ MN.OI O—m.o ocfldfiuhm

o o omuoo nouns
mm.o+ mn.ou mmm.o mafiapamOpoo<
Nw.o+ PM.OI oomco ouwamBHOMthuoadn
om.o+ mm.on .mmco ouHamBNOh
mw.o+ mace! omm.o Honmnaoz
.o.—+ mm.01 om:.o amassed
mm.o+ w..ou o:w.o ounmonnwo onoahdoam
owco+ 0:.01 mmm.o ufiod OHEhOh
on._+ mm.or .mm.o oqwnuosouaaz

Aowplmz homV Aoml<n 909v Ame. N XIV
Aenmvnuoo Aamdvaaoo Amvhufiafipapacomsm oaauossao> vqoeaom

 

mumanm Hausaono man on» on mmoauooauoo thHHnfipmoowsm afipoawmz .m canoe

34

(C) POTASSIUM SALTS
Potassium perchloride (Baker), thiocyanate, floride,
cloride, bromide, iodide (Fisher) and acetate (Baker)
were of reagent grade and were dried under vacuum at
..5000 for at least 48 hours before use. Potassium
hexafluoraphosphate (Pfaltz & Banner) was purified by
recrystallization from water and then dried under
vacuum at s11o°c for 72 hours. Potassium tetrapheno-
borate was prepared as the precipitate of the reaction
of potassium nitrite and sodium tetraphenoborate in
water. The precipitate was washed with conductance water
and recrystallize from acetone and dried under vacuum
at «~60°C for 72 hours. ‘
( D) SOLVENTS
Acetone (Mallinkrodt) was fractionally distilled

over calcium sulfate and then dried over Linde 4A
molecular sieves. Nitromethane, acetonitrile and
pyridine (Fisher) were fractionally distilled over
calcium hydride and dried over Linde 4A molecular
sieves. Propylene carbonate, dimethylformamide, dimeth-
ylsulfoxide and ethylenediamine were vacuum distilled
over calcium hydride and dried over 4A molecular
sieves. Formamide and formic acid were purified by
repeated fractional freezing and dried over Linde 4A
molecular sieves. Metahnol was fractionally distilled

over magnisum and iodine. The water content in the

35

solvent was analyzed with an automatic Karl Fischer

Titrator and was always below 100 ppm.
(E) PURIFICATION OF CRYPTANDS AND CROWN ETHERS

Cryptand 0222 was recrystallized from hexane and
dried under vacuum at about 40°C for 48 hours. Both
cryptand 0221 and 0211 were dried under vacuum before
use.

Macrocyclic 18-crown-6 was purified by forming
a complex with acetonitrile. When about 5 grams of
18-crown-6 was dissolved in 25 ml of acetonitrile,
fine white crystals of the complex were formed. The
flask was cooled in an ice-acetone hath (not dry ice)
to precipitate as much of the complex as possible and
the solid was then collected by rapid filtration. The
weakly bound MOON was removed from the complex by
pumping under vacuum. The melting point of recrystallized
18-Crown-6 was 39-4000, identical to the literature
value (100). Dibenzo-18-crown-6 was purified by recry-
stallization from benzene and dried under'vacuum for
at least 48 hours.

Both 15-crown-5 and 12-crown-4 were purified by
vacuum distillation (pressure ~10 torr) at 80°C~120°C.
Carbon-13 and 1H NMR were used to detect impurities and
water contents.

Monobenzo-15-crown-5 was synthesized by M. Shamsipur

in our laboratory. It was recrystallized from heptane

36

before use.
(NW

The time averaging of NMR spectra and Fourier
transformation of 39K data were done on the Nicolet
computers (Nicolet 1180 on Bruker WH-180 and Nicolet
1083 on varian DA-60) using program FTNMRD and QFN
for the WH-180 and DA-60 spectrometers respectively.
Chemical shift readout was directly obtained from
the spectra during the experiment. All the chemical shift
data reported in this thesis are relative to the chemical
shift of 39K* in an infinitely dilute solution. The
measurements were also corrected for the differences in
bulk diamagnetic susceptibility between nonaqueous
solutions and the reference (H20).

Chemical shift data obtained from salvation and
complexation studies were all fitted with apprOpriate
equations on a CDC-6500 computer using the least squares
program KINFIT (101) to obtain the respective formation
constants of ion-pairs and complexes. The applications
of the related subroutine equations and KINFIT program
are described in the Appendices.

CHAPTER III

POTASSIUM-39 NUCLEAR MAGNETIC RESONANCE STUDIES
OF IONIC SOLVATION AND ASSOCIATION OF POTASSIUM
SALTS IN VARIOUS SOLVENTS

CHAPTER III (A)

IONIC SOLVATION AND ASSOCIATION STUDIES OF
THE POTASSIUM IONS IN NEAT SOLVENTS

NT ODUCTION

Nuclear magnetic resonance of metallic nuclei is a
very sensitive probe of the ionic enviroment. Therefore,
nuclear magnetic resonance has become a powerful method
for the investigation of electrolyte solutions. The
measurement of chemical shift of metal nuclear resonance
yield valuable quantitative and qualitative information
about ion-solvent and ion-ion interactions.

In the past, the salvation and association of Lithium,
sodium cesium salts have extensively studied by 7Li,23Na
and 133Cs NMR respectively. However, since 39K nucleus
resonates at very low frequency and, in addition, has low
sensitivity, the studies on the 39K resonance have been
much more sparse and, in most cases, confined to aqueous
solution of fairly high concentration (:>0.2‘M). The
recent development of Fourier transform NMR technique
and of superconducting soleaoids with high magnetic fields
renders the study of 39'K resonance a much easier task.

In this study, the potassium-39 chemical shifts of potassium
salts were studied in eleven solvents in the 0.01-1.0‘M
concentration range.

RESULTS AND DISCUSSION

The 39K chemical shifts of potassium salts such as
potassium hexafluoraphosphate, perchlorate, tetraphenol-
borate, thiocynate, chloride, fluoride, bromide and

iodide were measured in various solvents. The data are

37

38

presented in Table 4. The 39K chemical shifts as fun-
ction of K+ ion concentration are illustrated in Figure 6-
10. As can be seen in most cases, the 39K chemical shift
of potassium salts show concentration dependence. Since
39K chemical shift of K+ ion is only sensitive to the
short range interaction, it seems reasonable to assume
that the variation of the chemical shifts with concen-
tration is an indication of cation-anion interaction,
presumably the formation of contact ion pairs.

It is also seen that in general, increasing conce-
ntration of the potassium halides leads to a paramagnetic
(downfield) chemical shift while potassium salts with
polyatomic anions such as PF6', 0104‘, and BPhh' give
the diamagnetic (upfield) shifts with increasing
concentration. Since the potassium-39 chemical shift
is dominated by the paramagnetic term in the chemical
shift equation (Ramsey's equation) (51), according to
Kondo-Yamashita theory (52), the increase in electron
density around the K+ ion results in rise of a strong
short-range repulsive force which induce the excitation of
3p electron of the K+ ion to higher energy states, and
decrease the shielding of the potassium nucleus. In
other words, the increase in electron density around
the K+ ion will result in a downfield shift. Replace-
ment of solvent molecules in the potassium ions inner

salvation sphere by anions may either increase or

 

39

 

 

 

 

we: gal
o.m No.0 0... no.0
0.0. no.0 m.m 00.0 0.0. 050.0
5.0. 000.0 m.m ..0 0.0. ..o Housman:
5.0. 0..o 0.0 00.0 ..51 ..o
¢.m. a~.o s.: 0.0 o.s1 0.0
m.:. 0.0 0.m 00.0 020 0.01 m.0
0.0. 5.0 sao< cannon 5.: 0..0 0.01 0.0
:.0 000.0 0.m mm.o p.01 m.o
0.0 no.0 m.0 0.0 0.01 no.0
... ..0 m.m 0.0 n.01 0.. 00:0
5.. 0.0 5.0. 0.. ouwmmsuom 0.0. .o.o
m.m 0.0 .... ..0 :.m. ..o
0.0 :.0 ..m. 0.0 m.m. 0.0
m.m m.o seem; 0.~. 0.0 0.0. :.o
0.0 .o.o 0.0. 0.0 0.0. 0.0
0.n 00.0 m.m. 0.0 0.0. 0.0
0.: ..o 0.0. 0.. ..0. 0..
m.a ~.o 0.m. mm.. .ooa.10 n.0. mm.. .ooonv
m.m 0.0 oaanpssouoo< 0.m. 0.. onoaoo< 0.0. 0.. 0:00004
some. .zv.omo0 uso>H00 ammo sz.ono0 pqobaom some. sz.omoo pmobaom
who.
msoausaom pawm anamomuom no mamasm Headsomo man .: capes

40

 

 

 

0.0+ 00.0
:..1 0..0
0.001 00.0 0.01 00.0
0.001 o..0 ..01 0.0
0.001 00.0 ..01 0.0
0.001 0.0 ..01 0.. 020 0.00 .0.0
0.001 0.0 0.0 00.0 0.00 000.0
..001 0.. 20 0.0 0.0 0..0 00.0
s.m. no.0 m.s 00.0 o..m ..o 002.:
0... 000.0 0.0 0.0 0002 0..1 00.0
0.0. ..0 0.01 .0.0 0.01 ..0
..0. 0.0 om 0.01 00.0 0.0 0.0
0.0+ .o.o 0.01 00.0 0.0 00.0 asaoanhm
0.01 00.0 0.01 ..0 0.0. 00.0
0.01 ..0 0.01 0.0 0.0. ..0
...1 0.0 0.01 0.0 0.0. 0.0
0..1 0.. 0.01 0.0 0.0. 0.0
0.:1 0.. have; 0.0.1 0.0 0020 0.0. 000.0 00
ammo. .zv.omoo uno>aom ammo. .zv.omoo unebaom ammo. nzv.omoo unobaom
Ha HM mama

ooaaaenoo .a canoe

L11

 

 

 

 

mam01 1000001
0.0. 00.0
0.0. 00.0
s.o. .0.0 0.0. 00.0
0.0. 00.0 n.o. mso.o 002.2
..0. 000.0 0.0+ 00.0
0.0 00.0 0.01 o..0
0.0 000.0 0.01 00.0
0.0 ..o 0.01 0.0.
0.0 0.0 0.01 00.0 0... 00.0
0.0 0.0 0.01 00.0 0... o..0
0.0 0.0 . 0.01 00.0 mawauamO»oo< 0... 00.0
0.0 0.0 0.0. 000.0 0.0 0.0
0.0 00.0 0000000 0.0. 000.0 ..0 0.0
0.01 ..0 0.0. ..0 0.0 0.. 0000 cannon
0.01 0.0 0.0. 0.0 ..0 00.0
0.01 00.0 0.0. 0.0 0.0 ..0
0.01 0.0 0.0. 0.0 0.0 00.0
0.01 00.0 0.0. 0.0 ..0 0.0
0.01 0.0 0.0. 00.0 0.. 0.0
0.01 0.. 00:0 0.0. 0.. 00 0.0 0.. moaswsuom
EH 4 CC . 0000 000500 sam 4 .5 . 0000 000500 39. 4 .5 . 0000 000500
z000 2000 Hz .
000000000 .0 cases

 

 

42

..0: no.0 0.mmu no.0

..0- 000.0 0.00- o..0

0.0- ..0 0.00: 00.0

0.0- m.0 0.mmu 0.0

..0... m.0 000... 3.0

..0- 0.0 0000 0.m~u no.0 0000000
0.0 00.0 0... mo.o_ 0.n~u 0.. -00000000
0.0 000.0 0.~_ 00.0 0.0 00.0

0.0 00.0 0... 0.0 0.0 ..0

0.0 000.0 0... 00.0 0.0 00.0

0.0 ..o 0... 00.0 0000 000000 0.0 0.0

0.~ 0.0 0.0 000.0 ..0 00.0

0.0 0.0 0.0 00.0 0.0 0.. 00000000
0.0 0.0 020 0.0 ..0 m.o+ 00.0

0.0. 00.0 0.0 0.0 0.0: 0.0

0... 000.0 0.0 0.0 0.0: mn.0

0.0. 00.0 0.0 00.0 0.0- 0.0

0... 000.0 0.0 00.0 0.0- 00.0

0... ..0 0000000 0.0 0.. 000 0.0- 0.. 0000;

z 0000 000>Hom
00000 2000 0000
000000000 .0 0200.

 

 

 

0.0- n~o.o

5.: nno.o 0.0- no.0

0.n op.o 0.0: noo.o

o.m n~.0 0.0- ..0

m.on .0.0 ..m n.0 m.0u m.0

m.ou no.0 n.0 no.0 Hog moaamanoo m.0a n.0

...- n.0 o... .0.0 o.na n.0
n..- 0.0 0.0. nmo.o m.0u no.0 :oaou 0020

o..- 0.. n.0. no.0 0.nma no.0

o.mu 0.. umx «.0. nuo.o n.0mu noo.o
m.o+ .0.0 0.0 ..0 pox 0.0.- o..0 oouaoaon
~.ou ..0 0.0. no.0 n.0.1 n..0 :oaou onoamopm

w 0.0.. n.0 0.0. ..0 o.m 8.0

_._- 0.F 0.0 nn.o o.m nmo.o

n.mu 0.. Hog 0.0 n.0 mu o.~ no.0

m.o+ .o.o m.o no.0 o.~ nno.o

m.o- ..0 m.o n_o.o o.m ..0

0.0- n.0 n.0 m.o o.~ m.o

...: 0.. m.o nn.o o.m n.0
o..- 0.. mg 0000: o.n n.0 o<u Hoamnooz o.~ 0.0 :oaou 0:0
ammqfizv.oqou aawm aqo>aom .ammqo sz.ocoo yawn aao>aom sand sz.onoo paam paopaom

 

coznfiunoo .: oanma

1+4

decrease the electron density at the cation. The symm-
etric polyanions apparently decrease the electron density
resulting in the upfield shift. 0n the other hand,
replacement of a solvent molecule by the halides increase
the electron density around the K... ion resulting in a
downfield shift. This indicates that both the halides
and thiocynate anion are better electron donors than
the solvent molecule they replace. On the other hand,
these symmetric polyanions are poorer electron donors
to the K+ ion than the solvent molecule.

The data obtained for solutions of potassium salts
in formamide and water are illustrated in Figure 6.
Both water and formamide have high dielectric constants
(109.5 and 78.5 respectively) and high donor numbers of
24.7 and 33 respectively. The dielectric constants and
Gutmann donor numbers of the solvents are shown in
Table 5. As Figure 6 shows, in the formamide and aqueous
solutions, potassium salts except KPF6 in H20 give a
linear chemical shift dependence on salt concentration.
Similar linearity of chemical shift lg salt concentr-
ation plots in the case of halide NMR study for alkali
halides have been reported by Deverell and Richards (13).
This phenomanon was interpreted in terms of the formation
of "collisional" ion pairs or very weak contact ion pairs.
In the KPF6 aqueous solution, the slight curvature seems

to indicate a slight amount of ionic association. The

45

 

 

 

 

 

 

 

 

 

 

E 1.4
a I
Q 12- FORMAMIDE
KPFb
8 . M/A
M,
4 ~‘ “ ~~ ~—...
“““‘0 ~~~~~~~~ .
\8‘“*\---~O.‘E [O
o. KCI
...4.
-8- L .
o 0.2 04 0.6 0.8

CONCENTRATION (M)

Figure 6. K-39 Chemical Shifts of Potassium Salts in
Water and Formamide

46

Table 5, Key Solvent PrOperties

Solvent Donor Dielectric

Number(a) Constant
Nitromethane 2.7 35.9
Acetonitrile 1h.1 38.8
PrOpylene Carbonate 15.1 65.0
Acetone . 17.0 20.7
Formic Acid 17.0 56.1
Tetrahydrofuran (THF) 20.0 7.6
Formamide 2h.7 109.5
Methanol 25.7(b) 32.7
Dimethylformamide (DMF) 26.6 136.7
Dimethylsulfoxide (DMSO) 28.9 46.7
Ethanol 31.5 24.6
Pyridine 33.1 12.}
Water 33 78.5
Ethylenediamine 55 12.9

(a) Reference 24

(b) Reference 29
(a)

Gutmann donor number scale. The scale is based on the
enthalpy of the reaction 3 + SbC15--~S.SbCl5 in dilute
1,2-dichloroethane solution. The donor number of the

Solvent S is defined as DN(S) = - 485,5bC15°

47

slight ionic association (for KPF6 in H20) has also been
reported from a conductance study by Robinson and
Stokes (102). It can be seen that for the halides, the
extent of the paramagnetic shift increase in the order
I'>Br">F", which is always observed for alkali metal
NMR. The result was explained in terms of the increased
collision probability with increasing the anion size
(7)613)(36). The smaller shifts for the thiocynate

than the halides is probably attributed to a greater
basicity of the halides (32). The greater availability
of the electrons on the more basic anions results in

a proportionately greater pertubation of the spherical
symmetry of the outer electron cloud of the potassium
ion.

Potassium-39 NMR measurements were also made on
potassium salt solutions in dimethylformamide and
dimethylsulfoxide. Both dimethylformamide and dimethy-
lsulfoxide have high donor ability (with donor number of
26.6 and 29.8 respectively) and average dielectric
constants (36.1 and 45.0 respectively}. The results
are shown in Figure 7. In dimethylsulfoxide solutions,
linear chemical shifts dependence on concentration
were also observed in the K0104 and KPF6 cases. However,
the 39K chemical shifts is independent of salt concen-
tration in the KSCN and KBPh4 cases which can be
explained either by absence of contact ion pairing

48

 

 

 

DIME THYL FORMAMIDE "

 

 

 

 

 

 

   

 

 

—6'
0 0.2 0.4 0.6 0.8
-4. DIMETHYL SULFOXIDE
2 KOO4
2: *6 ”We KPF.
<1 "”"f. KBPh. KSCN
A k A
-8.
*‘0' K1
...]2
0 0.2 0,4 0.6 0.8 1.0 1.2 1.4

CONCENTRATION (M)

Figure 7. K-39 Chemical Shifts of Potassium Salts in

Dimethylformamide and Dimethylsulfoxide

 

 

#9

formation or by a coincidental equality of electron
density at the K... ion upon replacement of solvent
molecules by the anions. In the KI case, a small con-
centration dependence of the chemical shift is indica-
tive of the some amount of the ionic association. As
can be seen, the results in dimethylformamide are
similar to those obtained for the dimethylsulfoxide
solutions.

The data Obtained for solutions of potassium
salts in formic acid and prOpylene carbonate are
illustrated in Figure 8. Both formic acid and propylene
carbonate also have high dielectric constants (55.0
and 69.0 respectively). The results in the formic
acid case are similar to those obtained for the forman-
ide solutions, in which the potassium salts used gave
a linear chemical shift dependence on salt concentration.
Once again the results are indicative of the formation
of very weak contact ion pairs or the collisional ion
pair. However, prOpylene carbonate (PC) shows a little
different behavior. The small extent of ionic association
of potassium salts in PC would be expected due to the
high dielectric constant of the solvent. However for
KPF6 solution in PC, the ionic association seems to be
significant, as seen by the some degree of curvature in
the plot of chemical shift‘zg concentration. Although

prOpylene carbonate has a high dielectric constant of

A'PPM

50

 

 

 

 

 

 

 

 

PROPYLENE CARBONATE
22' '
I8 '3pr
14. p/o/jH/afi _I.<SCN
)Or ‘3 KI
8..
T-___-_--_._._______ _ _ _ ..-—-4:
2‘2’
FORMIC ACID
18' ‘
"’_fl,,.—1>I(PF6
L4' Aflrfl__.—¢r*""fl”fl’a °
T‘8=a==::IZ3L_________‘;L—~:::::-——‘—:c¥;bl
10' .Kl
6.
2D
"I ‘ l D
o 0.2 0.4 C.6 0.8 1.0

CONCENTRATlON (M)

Figure 8. K-39 Chemical Shifts of Potassium Salts
in PrOpylene Carbonate and Formic Acid

51

65, it has a low donor number of 15.0. The data suggest
that the dielectric constants of the solvent is not
the only parameter which influence the ionic association.
It is evident that the donicity of the solvent is also
an important parameter in the formation of ion-pairs.
Potassium salt solutions in solvents of low donicity
and medium dielectric constant, such as methanol and
acetonitrile (see Table 5), were also investigated by
using the 39K NMR. The results are presented in Figure 9.
As expected, all the potassium salts in acetonitrile
and methanol undergo a significant extent of ionic
association, as indicated by the significant degree of
curvature of the plot of the chemical shift gs, salt
concentration. As can be seen from Figure 10, concentr-
ation dependent chemical shifts were also Observed for
KPF6 and KSCN in acetone which has a low donicity of
17.0 and a low dielectric constant of 20. However, the
398 chemical shift of potassium tetraphenylborate shows
concentration independence. It is not surprising, since
BPhh' is a large and relatively nonpolarizable ion which
would only weakly interaction with a cation. Unfortunately,
potassium tetraphenylborate is only slightly soluble
in most solvents, which limits the study of the ionic
associtation of this salt.
The concentration dependence chemical shift of the

solutions in ethylenediamine were also measured and the

 

52

 

 

 

 

 

 

 

 

 

l8’
METHANOL
W5L
e—s KP
KSCN
_9 s
2 KI
Q.
G.
<1 fir-——-——— ———.—_....._. ...... .... .. ..--
ACETONITRI LE
3' -
a7 KPEb
4” A
o.
, KSCN
‘4, ‘!———"‘—-‘.
-3-
0 0.2 ' 0T4 0.6 0.8
CONCENTRATION M
Figure 9. 39K Chemical Shifts of Potassium Salts in

Methanol and Acetonitrile

55

 

 

 

 

 

 

 

 

 

20. ACETONE
16- a KPF
“of O O b
12. '
aim KBPI’IA
8. N
flKSCN
4.
+----..-.._-...._....-_.. ...—.... .. _. _. _. _ ..--
2" ETHYLENEDIAMINE 7“"
E -18. A KC|O4
342/ KSCN
...—L o e 4 A #7
, -26\\)\
'30’ O\o
-34. Q
0 0.2 0.4 016 0.8 1.0

CONCENTRATION of POTASSIUM IM)

Figure 10. 39K Chemical Shifts of Potassium Salts in
Acetone and Ethylenediamine

58

results are presented in Figure 10. Ethylenediamine has a
very high donor number of 55, but a very low dielectric
constant of 12. Unfortunately, most potassium salts are
only sparingly soluble in this solvent. As Figure 10
shows, the high donicity of ethylenediamine does not
prevent the ionic associations in KI and KC104 solutions,
as indicated in the plot of chemical shift is concentra-
tion. However, a chemical shift that was independent of
the salt concentration was observed in the KSCN case.

Nitromethane has very low donicity of 2.7 but a
relatively large dielectric constant of 36. However, the
solubilities of most potassium salts in nitromethane are
low (<0.1 21). Only the solutions of KSCN and KPF6 at
low concentration (<10.11§) were studied, and the data
are presented in Table A. The slight change in chemical
shift in low concentration range (from 0.02 to 0.1 :4) does
not provide much information about the ionic association,
but it is qualitatively indicative of the ionic associa-
tion in such low concentration range.

It was of interest to us to investigate the effect
of the solvent on the ionic association of potassium
salts with a same anion. As can be seen from Figure 11,
in the case of the potassium iodide, in all solvents used,
the chemical shifts of K1 show concentration dependence,
indicative of some extent of ionic association. Even

in the solvents of high donicity, such as

55

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ah. K1
8 A in MeOH
M J
4 ’ . FORM MIDE
0 film. _________ H 0
. """ “---u ........... 2 ‘+
+ ‘_.——-—-——-J'DMF" -'
E -—4
o.
a...
‘Q -—6
’B‘k
D D D
’10‘ fl DMSO
441;
' L EN
'32 ‘ W 4)
.7 0.2 0.4 0.6 6.8
CONCENTRAHON (M1
Figure 11. 39K Chemical Shifts of Potassium Iodide in

Various Solvents

 

 

56

ethylenediamine and dimethylsulfoxide,there is significant
ionic association, which is indicated by curvatures of
the plot. The results seem to suggest the donicity of the
solvents seems not to be a predominant factor for the
ionic association of KI. However, in the solvents of

high dielectric constant such as formamide, water and
formic acid, the linear concentration dependence of 39K
chemical shift of KI were Observed, which are indicative
of very small extent of ionic association or only the
"collisional" ion pairs. These results suggest that for
potassium iodide, the dielectric constant of the solvent
seems to be a more important factor than the donicity

of the solvent for the contact ion-pairing formation.
However, in the KSCN case, as shown in Figure 12, the
chemical shifts of KSCN in the solvents with either high
donicity, such as dimethylsulfoxide, ethylenediamine,
dimethylformamide or in the solvents with high dielectric
constant, such as prOpylene carbonate, formic acid (see
Table 5) show concentration independence. As can be

seen, the ionic association is only exhibited in the
solvent of medium dielectric constant and low donicity,
such as acetonitrile, acetone and methanol.

In the case of KPF6 (Figure 13), in all cases, the
chemical shift show concentration dependence, indicative
of the ionic association. Although prOpylene carbonate
(PC) has a much higher dielectric constant than dimethyl-
formamide (DMF), the larger degree of curvature of the

Figure

 

57

 

 

KSC N
:o-oo-o-u----o-o---o----o-----------o— ....... E. Q)
0 e e z 2 F0 R H

Riz‘ $001.]

T

EN

an-.-— ---.—---- .---—-o.—-- --.-.— -‘

L

 

 

0.2 0.4 015 08 I0
CONCENTRATION (M)

12. 39K Chemical Shifts of Potassium
Thiocynate in Various Solvents

 

58

 

 

0.2 0.4 016
CONCENTRATIO N M

 

 

FOINH2

.0
K 1

0.8

DMF

 

LO

Figure 13. 39K Chemical Shifts of Potassium

Hexafluor0phosphate in Various Solvents

 

59

plot was found in PC and DMF, the result seems to be the

evidence for the effect of donicity of solvent on the
ionic association, since PC has lower donicity than DMF.
However, compared to pyridine, DMF has a lower donicity
but a higher dielectric constant. The smaller extent of
ionic association of KPF6 was observed in DMF than that
in pyrinine, as indicated by the degree of curvature of
the plot. The results suggest that the dielectric cons-
tant is also an important factor for the ionic associ-
ation of KPF6.

The ion pairing formation also can be determined
by the measurements of the chemical shifts as function
of the salt concentration. The ion-pairing formation

constants can be determined by the following equations (33)

M 2
-1 + (1 + AKIPCT ' 7+ )1/2

 

8 =

,obs " (6F ’ 81p) * 6IP

 

M , 2

1 2 (IIIA.1)
(1.823 x 106) I /

(DT)3/2

 

 

and -log 7+ =‘

9
50029 X 10 0
5‘. 11/2
1/2

1 +

 

(DT)
(IIIA.2)

 

60

where 8 is the observed chemical shift, 6F and <51?

Obs
are the chemical shifts of potassium ions in the free
solvated state an in the ion pair respectively. K is
the ion-pairing formation constant, C? is the total
concentration of the K... ions in solution, 7+ is the
activity coeffinient, I is the ionic strength, D is the
dielectric constant of the solvent, T is temperature in
OK, 2 is the size parameter of potassium salt. The ion
pairing formation constant was calculated by fitting the
concentration dependent chemical shift data using the
above equation by the KINFIT program (101). The application
of KINFIT and subroutine equations to calculate the ion-
pairing formation constant are described in Appendix I.

The ion pairing formation constants for potassium
salts in various solvents are shown in Table 6. The
equilibrium constant KIP is thermodynamic constant
where activity corrections are applied, and Kc is the
concentration equilibrium constant. As can be seen in
most cases, the values of KIP are small. In the case of
KPF6, the very weak interactions between the potassium
ions and PF6' anions in some solvents were also observed
using ‘9F NMR by DeWitte and Popov (7). This result is
not unexpected, since both K+ and PF6' are large and
relatively nonpolarizable ions.

As seen from Figure 1A, the degree of ionic associ-

ation shows temperature dependence. In H20 and acetone

cases, the degree of curvature of the chemical shift-

 

61

unnumsoo ESHApHHHsoo oHamshooanonp on» ma mHM any
uncommon asaanaaasoo soapmapsoosoo one ma ox any

 

 

00:1 0.. o). zomx noon;
n.m sn.n H sm.s nn.o H an.m sons assesses
ms.m1 mo._ H o~.n sm.o H ns._ zone cannons
nm.n1 .... H nn.o. n:.o H nn.o zone maansascsccs
nn.n_1 n..o H on.o 01. Ha coasmssom
o.s_1 00.n H mo.s .... H on._ Ha coaxoaasnasssosao
s.nn- 01. os.o H no.. Ha csaamascncasssm
ooms. 01. 01. He soon:
n.0 no.0 H nn., ...m H ns.n Ha assesses
0:1. 01. 01. mass coauooasmassscsao
ooms 01. 01. name sacs cannon
ooms. 01. 01. mass cossmanom
n.mm 01. u..o H n~.0 mass oneness
o.sm no.0 H nn.o 01. name monenssooasnsosao
o.s on.. H os.n ms.. H mo.s mama soon:
s.o. s_.n H ns.0 oo.m H .o.s mama summonses mscasoons
0.0 on.o_ H om.n. 0.0 H so.o_ name cassoasoscos
253 3H... n 3» fix as v 3m aflmm nook—”om

 

mpqo>aom msOHam> ad asammmuom Ho

mscasm Hsossoso msaoasss can mnemonsoo sosscsnos moanamoasoH one .0 cases

62

 

K .
16- PFbjgéf’Jyfi,,.4rvv““° 3dfi3
Ar””n
14‘ O
, A ——-—éT———-1r——--m;l5(3

1
I

 

 

 

 

CONCENTRATION (M)

Figure 1A. The Temperature Dependent Ionic Association
of Potassium Salts in Water and Acetone

63

concentration plot increases with decreasing temperature,
which is indicative of a stronger ion-ion interaction.

The concentration dependent chemical shift study
also can provide the information about the ion-solvent
interaction. At infinite dilution, it is reasonable to
assume that only ion-solvent interaction exist. Therefore,
the study of the infinite dilution chemical shift can
give some information about the ion-solvent interaction.
The "infinite dilution" chemical shift was obtained by
extrapolation of the curve to zero concentration by
using the computer program KINFIT, the application of
KINFIT program and subroutine equation for this study
are described in Appendix II.

The mean infinite dilution chemical shift of the
K+ ion in various solvents are presented in Table 7.
The plot of infinite dilution chemical shift gs the
Gutmann donor number (24) is illustrated in Figure 15.
It is readily seen that in general, there is a correl-
ation between the magnitude of the downfield chemical
shift and the donor number. Nine of the twelve solvents
seem to fall on a respectable straight line but the
correlation is not good for methanol, acetonitrile and
dimethylsulfoxide solution. It is interesting to note
that DeWitte and Popov (33) observed identical behavior
in the case of cesium-133 resonance, where the same

solvents show deviation from linearity in the chemical

6h

Table 7. IPotassium-39 Chemical Shifts at Infinite
Dilution in Various Solvent

Donor (b)

Solvent Gogppm) Number
Nitromethane 21.1 : O.h 2.7
Formic Acid '11.6 : 0.h 17.0
Propylene Carbonate 11.5 I 0.4 15.0
Acetone 10.5 i 0.9 17.0
Methanol 10.1 t 0.7 25.7
Formamide 4.6 i 0.6 24.7
Dimethylformamide 2.8 i 0.8 26.6.
Acetonitrile 0.t 1 0.8 14.1
water 0(a) 33.0
Pyridine -0.8 i 0.4 33.1
Dimethylsulfoxide -7.3 i 0.6 29.8
Ethylenediamine -23.6 i 0.7 55.0(c)

(a) Reference

(b) Gutmann donor number scale. The scale is based on the
enthalpy of the reaction S + SbCls-—+S.SbCl5 in dilute
1,2-dichloroethane solution. The donor number of the
Solvent S is defined as DN(S) = "AHS.SbC15’

(c) Estimated from 23Na chemical shifts. M. Herlem and
A. I. Papov, J. Amer. Chem. Soc.,‘25 , 1431 (1972)

65

.uonssz Homoa commune
one. mm 32m Hmoasoso 8338 3225 on» 00 sea one .n. 2&2

3533 do «$202 «028

               

 

 

     
    

IPWINPIBI @113 we 3. on on on ow. @1318 .2 0 s

m effigfivcorfw 'VNI
@ON1

’07

W91

m1
V
.VIW

zu.zm .0

n20
9.2.0.. .1
.m
«220%. :Ooim

im—

.0—

ocoquo:.Z .ON

 

 

 

 

66

shift-donor number plot. The sign and the magnitude of
the deviations were in the same direction and, approx-
imately, of the same magnitude. However, a good linear
correlation has been observed for Na infinite dilution
chemical shift with solvent donicities without
exception (29).

The correlation between the infinite dilution chemical
shift ( 80) of 39K resonance and donor number (DN) of the

solvents can be expressed by the following equation:

a = 21+.“ -0082 DN

0

The donor number of methanol from this 39K chemical
shift-donor plot can be predicted to be about 18.0, not
25.7. The value of 19.0 for methanol donicity was reported
by Olofsson (31). The deviation of acetonitrile from
linearity in the plot is probably due to some covalent
interaction between the nitrogen on acetonitrile and the
K+ ion (32). As can be seen from Table 7 with the exception
of DMSO and H20, the infinite dilution chemical shifts of
the K... ion in the nitrogen donor solvents go more down
field than that in oxygen donor solvents. Similar behav-
iors were reported in the Na and Cs cases (29)(32)(33).
The phenomenon was interpreted (32) in terms of the small
amount of covalent interaction between the cation and
the solvent. Overlap between 8 and p orbitals of the
solvent and p orbital of cation would permit the partial

electron transfer from solvent to cation.

67

It is interesting to compare the magnitude and the

"range" of the potassium-39 chemical shift to that of
sodium-23 (29) and that of cesium-133 (33). As can be
seen from Figure 16, the difference between chemical
shifts of Na+, K+ and Cs+ in nitromethane and pyridine
are 15, 23, and 90 ppm respectively. These results are
not unexpected, since according to Ramsay's theory,

one electron donated by the solvent to the cation induce

the change in paramagnetic chemical shift by the following

equation:
ea A A
cop = (- WK Tirol): (Lprv/rp3) |w0>
2 - -
-.- .. LSEEaZ =(constant) & (IIIA.3)
MECZ A A

where A is the average exciation energy of one electron
from np to (n+1)p orbital. e and M are charge and mass
of electron respectively. C is the velocity of light.

L is angular momentum of the p orbital electron and

P
r is the radial distance of the p orbital electron

fiom the origin at the nucleus. According to this equation,
the change in chemical shift of salvation is function of
the quantity ‘<:ag . The values 0:353: for Na+, K+,

Rb and Cs reported (13) tofibe 5. 9, _7. 9, 13. 8 and 18. 7

(a.u./Rydbers) respectively. Therefore change in chemical

68

Figure 16. The Range of Infinite Dilution Chemical

Shifts between Nitromethane and Pyridine
for 23Na, 3'9K and 133Cs Resonance

 

 

69

 

 

 

 

 

 

 

 

 

.3. 15 Nitromethane 23
1 NO
15ppm (ppm)
L 0
3 Donor Number 33
3. 39
«1,— 22-Jr Nitromethane
23ppm (Rom)
L -1... PY
. 5
3 Donor Number 33
8° N1 tho 133
_r— 60+- Irome "9 CS
90ppm (ppm)
OP
-°—-30.- y
33

3 Donor Number

Q
80, Infinite dilute chemical shift

70

shift ( 6p) due to the solvation of alkali metal ions
should increase in the order; Na+< K+< Rb+<:Cs+.

As can be seen from Figure 17, in the cases of
nitromethane, dimethylsulfoxide and acetonitrile, there
are the nearly linear correlations between the infinite
dilution chemical shifts of the 39K resonances and the

atomic numbers of alkali metals.

71

 

60. ,.. NITROMETHANE
40- ,/

20" A0”.

APPM

\\ \‘WACETONITRILE

-80. DIMETHYLSULFOXIDE

 

 

I l l l l I 1

10 20 30 40 50 60
ATOMIC NUMBER

 

Figure 17. The Plot of Infinite Dilution Chemical Shift
1g Atomic Number of Alkali Metal Ions

CHAPTER III (B)

IONIC SOLVATION OF THE POTASSIUM ION IN MIXED
SOLVENTS

 

 

(B) 01 ..:.\O .'O:. U.\\U.'.. I-

It is well known that when a solute is dissolved
in binary solvent mixtures, the primary solvation shell of
the solute need not maintain the composition of the bulk
solvent. It is likely that it may prefer one solvent over
the other. Thermodynamic methods has usually been applied
to study the preferential solvation. However, they often
can not differentiate between short and long range effects.
Since the chemical shift of nuclear magnetic resonance is
only sensitive to contact solvation. NMR technique has
become a powerful tool for the study of preferential
solvation.

The study of preferential solvation of ions by metal
nuclei resonance have been reported by several authors (11)
(29)(103). The variation of the NMR chemical shift with so-
lvent composition was explained in terms of preferential
solvation of the metal ion by one of the solvents in the
mixture as indicated by isosolvation point (equisolvation
point). This isosolvation point is the composition at
which the chemical shift lie halfway between the two pure
solvent values. It has been postulated that it corresponds
to the composition at which both solvents participate equal
in the contact solvation shell. In addition, it is assumed
that constant solvation number in solution at various sol-
vent composition, and no solvent-solvent interaction in

the mixtures.

72

75

RESULTS AND DISCUSSION

The potassium-39 chemical shifts of potassium salts
were measured as function of the solvent composition in
twelve binary mixtures and the data are presented in
Table 8. The results for the mixtures of acetone with
nitromethane, acetonitrile, water and pyridine are illust-
rated in Figure 18 and Table 9. It can be seen, that in
all cases there are the smooth transition as a function of
solvent composition from the chemical shift characteristic
of acetone to other solvents. The isosolvation points in
the mixtures of acetone with nitromethane, acetonitrile,
pyridine and water are at 0.20, 0.34. 0.38 and 0.75 mole
fraction of acetone respectively. The data seem to suggest
that the relative order of solvating ability is nitromethane
< acetonitrile < pyridine < acetone < water.

The results for the mixtures of acetonitrile with
nitromethane, prOpylene carbonate, acetone and water are
presented in Figure 19 and 20 and Table 9 and the isosol-
vation points are at 0.h1, 0.47. 0.66 and 0.92 mole
fraction of acetonitrile. Again the solvating abilities
for K+ ions among these solvents increase in the order:
nitromethane<:propylene carbonate<:acetonitrile<cacetone
< water.

The results from the studies of the acetone mixtures
and acetonitrile mixtures suggest that for KPF6 the solv-
ating abilities of these solvents increase in the following

7L1

 

o.w1
o.w1
o.w1
o.®1
N.BI
n.w1
n.n1
m.—1
..J
ammAV
o.m1
m.o—1
m.~.1
m.m_1
m.w.1
N.ON1
m.m~1

some. omzo .3 a:

o
o..0
ON.o
mm.o
m:.o

n.0
n.0
m.o
0..

o< no

0..
ne.o
no.0
os.o
mm.o

Azomx z n.00

m:

o<lom2fi

ousuxfiz

s..o Azomm z n.00

o

zmnomzn
ouauxflz

o.n+ o
a.m+ :n.o
..o+ mn.o
0.~+ m:.o
s.m+ .n.0
..0+ nm.o
$.01 ...o
s._1 o
and 04 no
o.n. 0..
n.n. :n.o
n.0. nn.o
o.o m¢.0
m.m mm.o
n.n ...o
s.n 0

some on so a:

h:

:62 2 ~00
zosuos

onsvxfiz

Emma 2 ~00
zosuos

ousuufiz

n.m— 0..
N.N— mm.o
n.0. mm.o
m.m mm.o
n.0 «n.0
n.m $9.0
n.0 o
9.34 3 Ho
u.n— 0.,
m.¢. _w.o
m.n— mm.o
N.w— N:.o
:.w. s~.o
m.o~ N_.o
m.nm 0

Band 04 HO h:

“name = ~.oo
Manon
runawaz

Annex.= ..00
22104

oasuxax

38:8 scan: on occasion 58 5:838 06 H0230 won .m 3an

75

 

~.o 0..
0.0 0.0
n.0 05.0
0.. 00.0
s.: on.0 AHs : n.00
0.0 0 ommumoez
sass 0mm 00 m: 23.22
0.. 0..
:.m 00.0
~.: mn.0
..m 0m.0
m.u .n.o Amman z ~.00
0.s_ 0 ommuos

sass omm so .0.

0.3ng

m; o;
0.. mw.0
m._ mn.0
n.. wm.0
~._ n.0 Amman : ~.oo
u.: o ommuzos
soda 0mm Ho m: 0.333”:
m.m ..0
0.0 N.0.
N. : n.0
n.n. n.0 Annex 2 «.00
5.5— O... radium
sand on we a: 0.383”:

0.0+ 0.,
T... 3.0
..51 3.0
Naomi wm.0
n.0m- n_.o “Haws n.00
0.0m... 0 387%
sam a mom: «0 ms 05.503”:
mim 0..
n.n_ no.0
n.~_ 00.0
s.m no.0
..0 mm.0
m.s n_.o Anmnu.x ..oo
0.n 0 zzuzos
5004‘ :2 «0 h: oasvxfiz
0.0... o.—
061 00.0
0.m1 nn.0
0.... ~.0
m.01 o_.o 520mm,: n.00
m.0- o ommuomzo
334 0020 «o a: 0.3003:

 

00323300 .0 0.388

76

 

24’

 

NI TROMETHANE

A

ACETONE

 

 

* \
ISOSOLVATION POINT ‘2.
O. PYRIDINE
0

072 04.4 06 0:8 10
MOLE FRACTION OF SOLVENT '

Figure 18. 39K Chemical Shifts of KPF6 in the Binary

Mixtures of Acetone with Nitromethane,
Acetonitrile, Water and Pyridine.

77

.095H0m mm now: was Hg : m.0 .ommo monspxfis Hoqmnpos nH on

opsaom we own: one: 20mm : n.0 .moHNouHsmHnmposdo Ho mouauxaa one Ham 0H “90
.oom: mums whim mo 2 ..0 one; osmsposoapam mo mouszws on» sH pmooxo .oazaom

.moazuxfia oHHApfisopoom can oaopoom on» sH Amy

mm 00m: one; 000% no 2 m.0

 

_—
moez no moauomnm 0H0:
=
2
omzm no GOHpomum 0H0:
3
=
.—
oHHAuficopmom Ho soHaowum 0H0:
:
Z
=

'—
osouoom mo :oHpomnm 0H0:

mm.0
Pm.0
0:.0
nn.0
m—.0
N0.0
00.0
50.0
P:.0
n:.0
mu.0
wm.0
:n.0
0m.0

HoumnquIosHsmdoosoannpm
Hosanna: IAoVAopma
nsasmascseasssmaooaxooHomasneosao
Aopmaloodxouasmannpoadn
omOpoodIAnvoodxouasmannvoafin
nopmgloafluofinopoo<
onopoodIoHHapHsopoo¢

commonamo osoahmonm1oHHavaOpoo<
osmnposoupHZIAmvoHHupHsopoo<
Azommv oHHupHsopoodiosoooo<
nopmslon0000<

omfiofiuhmIoaovoo<

Anmdao cassossonnoaaesoomcs
oqmnuosou0HZIAmvosouood

 

usaom soHpm>HomomH

ouspxnz

 

unhapxaz pso>aom nnmmam can

on madam ssfimmmpom pom mama venom soHpm>HomomH no nnmsasm .0 dance

 

78

 

 

 

 

 

‘ NITROMETHANE
241 ACETONITRILE /.
20- .
16-
,, .. -..-..O
E O /’O"
o. // ACETONE
Q 12 I/
<1 /
}/
31 x’
[0”
4th ’/
WATER
:5 A A
O .
0:2 0:4 0:6 08 1.0

MOLE FRACTION OF SOLVENT

Figure 19. 39K Chemical Shifts of KPF6 in the Mixtures
of Acetonitrile with Nitromethane, Acetone

and Water.

 

 

79

 

PC-ACN mixture.
'3 "‘ 0.2 M KPF6

MP

11 -

   
 

839K (ppm)

0.53

 

 

l l

l l I
0.2 0.4 ofs 0.8 1.0

 

MOLE FRACTION OF PC

Figure 20. 39K Chemical Shifts of KPF6 in the
Acetonitrile-Propylene Carbonate Mixtures.

80

orders, nitromethane<:pr0pylene carbonatefLacetonitrile
< pyridine < acetone <water.

The 59K chemical shifts of KSCN as function of the
composition of the mixtures of dimethylsulfoxide with
acetone, water and ethylenediamine are presented in
Figure 21 and Table 9 and the isosolvation points are
exhibited at 0.15, 0.35 and O.h8 mole fraction of DMSO.
It seems to indicate that the relative order of solvating
ability is DMSOEZethylenediamine >water >acetone. Despite
the very high donicity of ethylenediamine, it does not
appear to be a better solvating agent than dimethylsul-
foxide. The enhancement of donicity of DMSO by introdu-
ction of even small amount of another solvent into neat
DMSO was reported (29) previously. It was explained in
terms of the break up of the polymeric structure of DMSO
by addition of the other solvent. Therefore, it seems
reasonable to assume that the high solvating ability
of DMSO in the DMSO-ethylenediamine mixtures results
from the same causes.

Finally, the preferential solvation of K1 in the
mixture of methanol with water and ethylenediamine were
investigated. The results are shown in Figure 22 and
Table 9. The isosolvation points for the mixtures of
methanol with water and ethylenediamine are exhibited
at 0.51 and 0.53 respectively, which indicates that the
relative solvating ability in ethylenediamine >water:>

81

 

DMSO ACETONE

 

ETHYLENEDIAMI
-24- NE

 

J

0.2 0:4 0:6 0:8 {,0
MOLE FRACTION OF SOLVENT

Figure 21. 39K Chemical Shifts of KSCN in the Mixtures
of DMSO with Acetone, Water and Ethylene-
diamine.

82

 

     
 

MeOH-EN(K|)

 

 

MOLE FRACTION or MeOH
0.2 03: ole 0.3 1.0

4‘
MeOH-H20(KV
2 .
E
3' 0»
Q
...-2‘ C O

0.2 6.4 ole 6.8 {.0
MOLE FRACTION or H20

Figure 22. 39K Chemical Shifts of KI in Mixtures of
Methanol with Water and Ethylenediamine.

 

 

 

 

 

 

83

methanol.

Recently, Covington,'g§'al (28) developed a quantit-
ative model for preferential solvation of ions in binary
solvent mixtures. They presented a equation that allows
the calculation of equilibrium constants and the Changes
in free energy of preferential solvation. The equation
is

—‘—-=_‘ (1+ ;) (11113.1)

6 8p K1/n9—(B-
XA

where: a = observed chemical shift relative to

the resonance of M+ in pure A(solvent)
8p: total range of the chemical shift
(is: 5A0 - cBo ) '
K1/n = the geometric equilibrium constant
n = the solvation number
XA, XB = the mole fraction of A and B(solvents)

respectively

The KI/n and 1/'op can be calculated from the slope
and intercept of the plot of 1/6 .zg XB/XA respectively,
and finally the free energy of preferential solvation,

AG/n can be obtained as following:

AGO/n = -RT ln XV“ (11113.2)

The two typical plots of 1/3 z§ XB/XA are shown

84

in Figure 23 and 2n, and the values of 1:1/‘1 and AG for
each system were obtained by linear least squares
procedure by KINFIT program and summaried in Table 10.
The computer subroutine equations used to calculate KVn
is described in Appendix II. In spite of a number of
idealized assumption, in all cases shown in Table 10,
the plot of 1/6 ‘zg XB/XA yield straight lines. The
Covington's quantitative approach seems to be successful

in the 39K NMR preferential solvation studies.

85

.msaxcsasmasssessa house .Hosmsscs "mom: .msssmsse
-scaassm.azm .esssssaa «as .csmssmscapaz azz .easssssosmes have .mscsees was Ass
x
Am ano>aomv+M elmll.a¢ pdobaomv

M
.o4 +
soapmswm on» ma pneumsoo Hafiunwafisvo can use mmquo menu on» one x was u<.on

.onopmoo mo N.o soapomum oaoa pm psaom soHum>HomomH any unwom soaps>aomomH Adv

 

 

mom.ou mw.o m: mm.o mom: Hm zmumocz
sum.o us._ ms m.o zos mama zzuzos
so..o no.0 m: mm.o omza zoms ommuomza
mms.o mm.P m: as.o cuss zone zmnomzn
om:.. .m.. m: mm.o omzn macs wauomza
om_._- no.0 m: mu.o as mass ommnos
mm:.o .m.. m: mn.o as mass amuos
we:.o a... m: s:.o us zoms zosuos
smm._ m_.m m: :m.o as mass zosuos
wmm.m wo.s amps: m.o as mass “sizzumm
mfs
Asiaeacs\fixv s\maV Ace s\as Reyna mH sash memapxwz

 

muco>aom cox“: on» ma owsmno hmuosm mosh use musmpmnoo asaunaaasam mas .o— mHnwe

86

Figure 23. Convington Plot for Pyridine-Acetone Mixtures

87

0:

o N
o m
o

 

..o
and
ad

.v.o

dev
I

6.0
.56

.md
3%: 3T:

 

 

 

 

88

monspxfiz esopmo<loawupfinopoo< map you poam nouwsa>soo .JN ousmah

a2 on
83x

ox. own o.~. 0..

 

‘

 

nzomvco<lzo<

 

 

 

 

CHAPTER IV

POTASSIUM-39 AND CARBON-13 NMR STUDIES OF POTASSIUM
SALT COMPLEXATION IN VARIOUS SOLVENTS

CHAPTER IV (A)

COMPLEXA‘I‘ION OF THE K+ IONS BY CRYP‘I'ANDS

INTRODUCTION

 

Since the advent of macrocyclic crown ethers and
cryptands (55)(66), the complexation of crown ethers with
alkali metals in water and methanol have been studied
extensively. However, the complexation of crown ether
with metal ion in nonaqueous solvents (except methanol)
have received much less attention.

Potassium-39 and carbon-13 NMR can be applied as
sensitive and useful probes for the quantitative and
qualitative studies of the complexation reactions of
potassium salt with macrocyclic crown ethers and cryptands

in various solvents.

(A) COMPLEXATION OF THE K+ IONS BY CRYPTANDS

 

The complexation studies of potassium ions with
cryptand 0222, 0221 and 0211 have been carried out in
various solvents. In the 39K NMR mole ratio study, the
concentration of KPF6 was held constant (0.02‘3) and
the ligand concentration varied. On the other hand,
in the carbon-13 NMR study, the ligand concentration
was kept constant (0.05‘fl) and KPF6 concentration varied.
The cavity size of cryptand 0222 was reported to be
2.63 (66) which matches the diameter of the potassium
(«v2.7A). Therefore, the 0222-K+ complexes should be
stable. In acetone, methanol and nitromethance solution

cases, at mole ratio of 0.5 of 0222/K+, two 39K signals

89

90

are observed corresponding to the free K+ and complexed
K+ ions respectively (Figure 25). Therefore, the exchange
between free K+ ions and complexed K+ is slow on the

39K NMR time scale. In general, if 11 and V

free complex
are the resonance frequencies of free and complexed

K+ respectively and

) (IVA.1)

1
exchan e r t -———7r V - V .
g a 6 ((V2 ) ( free complex ’

than the two NMR signals corresponding to the two K+

sites will be observed. 0n the other hand when

1

exchange rate_>_(\f—2:7T) ( V V ) (IVA.2)

free complex

than, only one pOpuiation averaged NMR signal will be
observed. The 39K chemical shifts for free K... and com-
plexed K+ are listed in Table 11. As can be seen, the
chemical shifts for complexed K+ ion in various solvents
are about the same, which is indicative of the formation
of inclusive K+-0222 complexes..In an inclusive complex,
the potassium ion is inside the ligand cavity and is
completely enclosed by the macrocyclic ligand, it is
essentially isolated from the solvent. In order to
obtain further information about the strength and stru-
cture of the K... 0222 complexes in various solvents,

carbon-13 NMR of ligand was also used to study the system.

91

Figure 25. 39K NMR Spectra of KPF6-Cryptand 0222 Solution

92

KPF '-
i I 39

C
ACETONE

 

 

METHANOL

 

”F

NITROMETHANE

 

 

COM
PLEX FR
EE K+

93

Table 11. The Chemical Shifts and Line Widths of
Potassium-39 of K*C222 Complexes at 0.5 Mole
Ratio (0222/K*)

 

 

Solvent Chemical Shift (ppm) Line Width (Hz)
Nitromethane 22.27 (F)(a) hu.o
-2.#8 (c) 92.8
Methanol 11.51 (F) 24.5
-2.35 (c) 73.3
Acetone 12.01 (F) 19.3
-2.78 (c) 87.9
Dimethylformamide(b) 6.28 (F) 25.2
-2.04 (C) ' 90.2
Acetonitrile(b) 1.96 (F) 17.1
-2.u1 (c) 87.9

 

(a) F = Free K+, C = Complexed K+ ion
(b) The signals for free K+ and complexed K+ are overlaped
in dimethylformamide and acetonitrile.

9A

The resulting C-13 chemical shifts are presented in
Table 12 and Figure 26. The three observed C-13 NMR
signals of free cryptands correspond to three kind of
carbons in C222. The assignments of peaks for the three
carbons are shown in Figure 26. Carbon(1) and carbon(2)
are the QQHZ carbons and carbon(3) is N_C_H2 carbon.

As Figure 26 shows, at mole ratio 0.5 of K*/C222, the
two signals for each carbon were observed for carbons

of the free ligand and complexed ligand respectively.
Again we have a slow exchange between the free ligand
and the complexed ligand. As seen in Table 12 a slow
exchange occurs even in solvents of high

donor number and high dielectric constant, such as
dimethylsulfoxide and dimethylformamide. At a mole

ratio of 1.0, only one peak for each carbon was observed
which corresponds to the complex. It is interesting to
see that chemical shifts of both free ligand and
complexed ligand are about the same in various solvents
except H20. It appears, therefore, that the structures
of either free cryptands or complexed cryptands in these
solvents are about the same. However, in the water case,
the chemical shifts for free cryptand show large differ-
ence from that in another solvents. For example, for

the NQHZ carbon, the ‘30 chemical shift is 5h.76 ppm

in the H20 case, but it is about 58.10 ppm in another

solvents. The different behavior in aqueous solutions

95

onmwaa moxoamaoo you Hammam on a

 

 

1 . ass sm.mm loo ma.sm
oo.mm Ammozv mo.mm “my
1 1 on .m.mm on .m.mm
mm..s Ammoov mm..s “NV
1 1 on so.ms on mo.ms
mm.ms Ammoov sm.ms A_v mans omzs
1 1 ADV :m.mm on Pw.mm
mo.mm Ammozv mo.mm Amy
1 1 ADV Pm.mw ADV mm.mm
sm._s Ammoov .m._s “NV
1 1 Aussa._s .aov se._s
mm.ms Ammoov mm.ms “_V mans man
1 on pw.mm on Pm.mm on Fo.mm
sm.mm Assess s~.sm any
1 ADV —N.mm ADV pm.mw ADV mmomm .
ms..s Ammoov ms.is Amy
1 on so.mu on s..ms Auv so.~s
mm.ms Ammoov sm.ms A_V cams ososoos
sam Q
o.m o.m o.— m.o o n
Ammmo\+xv oases mac: amen pamm ssosflom

 

mmmo mosmsssoo sasmmmsom oo osssm Houseman n.1soosmo

.Np magma

96

anomaa uoxoamaoo mom Hogwam ADV t

 

 

 

Ram 1 1 2.3 $55 63m 5
em.es 1 1 mm.ao assess _s.os Amy
mm._s 1 1 om._u Ammoov oo..s A.V 20mm 0mm
1 1 1 on m:.mm
om.sm Ammozv mo.om Amy
1 1 1 on _~.mm
mm..s Ammoov mm..s Amy
1 1 1 on so.ms
mm.~s Ammoov sn.~s APV zone omza
o.n o.m 0.. m.o o n
Ammmo\+mv oases sac: some uses sooeaon
possesses .m. oases

97

1 2 FREE C222/ACETONE
3

F ‘ (KW/(c222) 0.5
C

 

 

 

 

411411181 . MWNW

3 2 ’}__\\3/’\\.
ML

W

13
Figure 26. c NMR Spectra of K*-Cryptand 0222 in Acetone

98

may indicate some interaction between the free 0222 and
water, such as hydrogen bonding between the oxygen of the
cryptand and the water molecule.

As can be seen from Table 13, in all cases there is
a large Change in chemical shifts for NQHa carbon and one of
09H2 carbons (carbon 2 in Table 13) upon complexation,
but a small change in the other 0_C_H2 carbon (carbon 1
in Table 13). If the potassium ion is located in the
center of cavity and the structure of cryptand does not
change upon complexation, the changes in chemical shifts
for two 09H2 carbons should be about the same and larger
than that for NQHZ carbon since usually ion—dipole
interaction for O-M+ is stronger than for N-M+. However,
actually, as can be seen from Table 13, the change in
chemical shift for NQHZ carbon is larger than that of
either QQHZ carbons. The most probable explanation is
that there is the strong interaction between K+ and
nitrogen atom which makes the change in structure of
cryptand from the exo-exo (90) to the endo-endo confor-
mation upon complexation, which makes a extremely change
in the enviorment of NgH2 carbon.

The complexation of cryptand 0221 and potassium
ion was monitored by 39K NMR. As Figure 27 shows, at
mole ratio 0.5 of 0221/K+, in acetone, methanol, diemth-
ylformamide and acetonitrile, the two 39K signals which

correspond to the free K+ ion and complexed K+ ion were

99

Table 15. The Change in 13C Chemical Shift of

Cryptand C222 upon Complexation

 

 

Solvent Acetone Dimethyl- Dimethyl-
sulfoxide formamide
A6,(0CH2) 0.21 ppm 0.51 ppm 0.39 PPm

 

100

Figure 27. Potassium-39 Spectra of Potassium-C221 Cryptate
in Various Solvents; [C221] = 0.01 .131 ,
[KPF6] = 0.02 M.

101

Acetonitrile

A
”...—-7

Cryptand 221 - KPF

|C221| -
KPF6 - 0.5

6

 

01' methyl formami de
1 Jim/f
v. r 1
'V "WNW

-_.. fi—1.~

Methanol

MW ‘01” M411

H O

 

102

observed. It indicates the slow exchange between the free
K+ and complexed K+ ion on the 39K NMR time scale. However,
in the water case, the only one broad peak observed at
mole ratio 0.5 indicates the fast exchange between free K+
and complexed K+. The chemical shifts and line widths

for the free K+ and complexed K+ in various solvents are
shown in Table 14. The chemical shifts of K*-0221 cryptate
seem to be dependent on the solvent, which is indicative
of some contact interaction between K+ ion and the solvent.
It is reasonable to assume the formation of exclusive
K+-0221 complexes in these solvents.

The carbon-13 NMR was also applied for the investig-
ation of complexation of K+ and cryptand C221. The ‘50
chemical shift as function of the K+/C221 mole ratio
in acetone, dimethylsulfoxide and water are presented
in Table 15. In the acetone case, 13C chemical shift for
each carbon on cryptand seems to reach a limiting value
after the mole ratio of 1.0 (the experimental error is
about : 0.05 ppm) which is indicative of the formation
of a stable complex. However, at mole ratio 0.5 of
K+/C221, only one averaging peak for each carbon was
observed, which is indicative of the fast exchange
reaction between free and complexed cryptand C221 while
the slow exchange reaction in the K+-C222 case was observed
as metioned previously. Therefroe, the K+-C222 cryptate

in nonaqueous solvents is a relatively inert complex

103

Table‘Hp. Chemical Shifts and Line Widths of K-39
of K+-C221 Complexes at 0.5 Mole Ratio

 

 

 

(0221/K*)
Solvent 611m(ppm) V1/2(Hz)
Acetone (1) +12.31 (F) 31.7
(2) -10.94 (C) 53.8
Methanol (1) +10.86 (F) 2t.#
(2) -1h.13 (c) 78.1
Dimethylformamide (1) +5.05 (F) 39.0
(2) -1h.13 (C) 68.0
Pyridine (1) -0.18 (F) 30.3
(2) -17.60 (C) 82.h
Acetonitrile (1) +1.56 (F)* 19.5
(a) -13.54 (c1' 68.3
320 (1) -#.83 102.5
(F)' 2 Chemical shift for free potassium ion.

(of

complex KPF6= 0.02 M.

Chemical shift for complexed K+ in K+-C221

1011

Ao<w.mv _NNU

 

 

sm.mm 1 1 seamsozv no.6m “no
_o.mm 1 1 seam:ozv om.um Asv
om.on 1 1 Amv
mn._s 1 1 mwamsoov mo._s Amy
mm.ms 1 1 _oAm:oov ms._u “_v Hm om:
+Z.mm m RV
ss.em mw.mm 1 swAmsozv ms.mm Asv
z mm.mm mm.mm 1 mowmzoov mw.is Amy
//<\\r.eo._s em._s 1 meamsoov sm.ms Amy
mm..s ms._s 1 _oxmzoov mw.ms A_v mans omza
mm.mm si.mm mo lav
ss.am ss.am as.am seamsozv ao.am ass
om.mm mm.mm ww.os mofimzoov _m._s Anv
mm._s os._s sm..s moamzoov mm.ms Amy
ss..s mw._s m_.ms _oxmsoovsmo ss.ms AFV mans ososoos
o.N OJ n.0 0 xx
A_mmo\+sv oases use: seem namm ssosHom

 

FNN mpopmmho ESfimepom mo Sufism. awofiuono Mplconumo.m—0Hnma

105

while K+-C221 cryptate is a relative labile complex.

The 13C Chemical shifts for each carbon on free
0221 in aqueous solution can be seen to show some differe-
nce from that in another solvents. For example, the
chemical shift for carbon 5 (NQHZ carbon) is «'56 ppm
in the water case, but it isa~f§9 PPm in acetone and
DMSO cases. Again, this result may be due to the inter-
action between the cyptand C221 and the water.

The data of potassium-39 NMR Chemical shifts for
K+-0211 cryptates are tabulated in Table 16. The mole
ratio studies of the complexation in various solvents
are illustrated in Figure 28. Since the exchange between
free K... and complexed K... was fast on 39K NMR time scale,
only one 39K signal was observed at various mole ratio of
0211/K+. In the cases of acetone, pyridine and aceton-
itrile solutions, the chemical shifts reach the limiting
value at a low mole ratio of ligand/K+ (..2,5), which
is indicative of the formation of relatively strong comple-
xes. The formation constants for these complexes are listed
in Table 16. In the dimethylformamide case, the chemical
shift tends to change continually even after mole ratio
h.0, which indicates the formation of weak complexes.
However, in the DMSO case the chemical shift of 39K seems
not change with increasing ligand concentration. The
formation constants of K+-C211 complex was obtained by

fitting the curve using KINFIT program, (the application

106

mass 2 ~0.0 spasm .

 

0.01 0.n 0.01 0.:
n.01 0.m 1 0.n ,
s.s1 m._ 0.01 0.m
0.01 0.. 0.01 0..
0.01 n.0 0.01 n.0
0._1 0.0 :.~1 0 an 0.01 0 00:0
0.01 0.m. m.._1 0.m s.m1 0.n
..01 m.m m._.1 m.m s.m1 m.~
0.0 0.m ~.__1 0.m 0.:1 0.m
0.. m.. s.s1 0.. «.01 m..
m.m 0.. 0.01 0.. 0., 0..
m.: n.0 ..n1 n.0 0.m n.0
0.m 0 man 0..+ 0 20s 0.~_ 0 0m

.3.
£994 +x\:mo unobaom 5&4 +M\:No ..Eobaom ammo +g\:~o asobamm

30652... snows; 5 0mm: fin. moxoaaaoo :3 compass 0o soaps 30mm 3o: .0. cases

107

 

  

 

 

12
‘0 ' Cryptand 211
0.02 M KPFb
3 ..
61
4 — ° .
g 2
J3-
ao 0 '
° CMAF
-2 o
-4
Acetone
-cL _____________ 1-
CMASC)
-8 ~13— Pyridine
40 _- Acetonitrile
'12 1 4 L 1

 

1.0 2.0 3.0 4.0
(C 211]/ RH

Figure 28. 39K Chemical Shift 1g Mole Ratio of
0211/K+ in Various Solvents

108

of KINFIT and subroutine equation are described in
Appendix IV). As Table 17 shows, the formation constants
of K+-C211 complexes among these solvents decrease in

the order; acetone:>acetonitrile >pyridine >dimethyl-
formamide. It is not surprising that the most stable
complex in acetone solution, since acetone has a low
donicity and a low dielectric constant. Acetonitrile

and dimethylformamide have about same dielectric constant
(“'38), but the difference in stabilities of complexes
in both solvents was observed, which may be due to the
lower donicity of acetonitrile than that of dimethyl-
formamide. Pyridine has higher donicity than dimethyl-
formamide and it would be expected that the complex in
pyridine should be weaker than that in dimethylformamide.
However the reverse experimental result was obtained.

It should be noted that.according to the Pearson's
Hard-soft-acid-base (HSAB) theory (104), pryidine is a
relatively soft solvent (base) and K... is a hard ion
(acid), the K+ ion should be expected undergo weak inter-
action with pyridine, and then the K+-cryptand intera-
ction would be expected to be strong.

The carbon-13 NMR chemical shift measurements were
also made at mole ratio (M+/cryptand) of 0.5 in alkali
metal salt (NaI, K+ and Cs+) nonaqueous solutions. The
data are presented in Table 18. The results seem to

indicate that inert complexes are formed in the cases of

109

Table 1?. Formation Constants and Limiting Chemical
Complexation of KPF6 by C211

Shift for

 

 

 

Solvent Log Kf 611m(ppm)
Acetone >h. -5.Ah : 0.05
Acetonitrile 2.80 i 0.21 -11.78 i 0.37
Pyridine 2.17 i 0.06 -9.15 i 0.08
Dimethylformamide 0.99 i 0.06 -12.98 i 0.20
Dimethylsulfoxide(a) - -

 

(a) The ligand concentration show no effect on

shift of KPF6 in dimethylsulfoxide.»

the chemical

O

ccmmfia voxoamsoo pom mamsmam on cswmfia 00am pom wasnMHm any mamsmwm cmoum Amy

 

 

moaovmm.0m
. 1 . Salesma.00
so am as am 00.00 ae.sm A00
onmm.w0
_ .,s 1 .0s
0 00 so..s 00.05 A00
. 1 . “0000.0s
am ms as is 0~.~s s_._s Amv
ss.ms 1 0_.ms A0000.os
PNNU mohh Ahvmhomb unopn Apv moo mn0900< FNNO
m . . Aevms.mm
. . A00.m.00
Am000 _s 00 _s Amv0m..s any 00._s Amv
o o AOvaov300N5
havsm mm mm mm Anyasvmm.ms Amv om.ms A_V 0.0 0020 mmmo
£009
unsymmuo oapmm 0H0:
mosh +00 +M +02 +Hg AA\+MV psobaom unopmhuo

 

ncsmsnsso no use s .82 .HS 0o Asdmasv omens Hmossoso n.1sosan0 .0_ cases

111

 

 

8002mm

 

0m.sm 0..s0 _m.sm 00.n0
0~.s0 A00
so.am sa.am 00.00 0s.am ae00o mm
00.00 A00
onsm.s0
0s._s 0m._s mn..s s_._s
0n..s Amv
0 ms.
09mm. 0000 00.3. 3.: A0 00
asvsm.ms A—V 0.0 0020 ..m0
:No mosh 030m
sassosso oases oao:
cons +00 +0 +82 «0 A0\+sv .ssosaom esososso

 

connfianoo cm. 0.3.98

112

Li+ for cryptand C211, Na+ for cryptand C221 and K+ for
cryptand 0222, as indicated by two 13C signals observed
for each carbon on cryptands which are corresponding

to free ligand and complexed ligand respectively. It
indicates that in nonaqueous solvents, cryptands still
have very sharp cation selectivities; cryptand 0222 for
K+, cryptand 0221 for Na*, and cryptand 0211 for Li+.

CHAPTER Iv (B)

COMPLEXATION OF THE K+ IONS BY CROWN ETHERS

(B) COMPLEXATION OF THE K+ IONS BY CROWN ETHERS

In this work, the complexation reactions of potassium
hexafluorOphosphate with crown ethers, such as 18-Crown-6,
dibenzo-18—crown-6, 15-crown—5, monobenzo-15-crown-5
and 12-Crown-4 were investigated in various solvents
by 39K NMR and 130 NMR.

BFSULTS AND DISCUSSION

The chemical shifts of 39K as function of mole
ratio of ligand/K+ were measured and the data in various
cases are presented in Table 19. In the 39K NMR compl-
exation study, the concentration of KPF6 was held const-
ant (0.04‘M) and the ligand concentration varied.

In the 18-crown-6 case, the plots of the 39K chemical
shift vs mole ratio of 1806/K+ in acetone, dimethyl-
formamide, water and dimethylsulfoxide are illustrated
in Figure 29. In the case of acetone, the chemical
shift reachs the limiting value after mole ratio 1.0,
which is indicative of the formation of very strong
K+-18C6 complex. It is not surprising, since the diameter
of potassium ion is 2.73, which is very close to the
cavity size of 18-crown-6 (2.62) (72). The potassium
ion would be expected to form a stable complex with
18-crown-6. However, in the solvents of high donicity
and dielectric constant, such as dimethylsulfoxide,
water and dimethylformamide, as can be seen from Figure

29, the curves level off at higher mole ratio (:>3.0),

113

111+

com: mm; H0

.3.

 

 

 

0.0+ 0.0
0.0+ n.0
~61 0.0.
0. .1 o.—
..n1 0.0
0.0 0.0 0._ 0.0 0.01 0 00:0
0.0 0.0 0.. 0.0 0.: 0.0
0.0 0.0 m._+ 0.0 0.: 0.0
0.: m.— 0.0+ 0.. 0.0 n.—
u.: 0.. 0.0+ 0.. 0.: 0..
0.0 0.0 0.0+ 0.0 0.5 0.0
0.0 0 .08 0.01 0 + 0N0 0.0. 0 00.
an 4 +0\0o0_ 000500 sam 4 +0330— »nmpaom and +0000. anon—How
muqo>aom msoaum> as 00mm spat moxoanaoo muosam stone «0 0000590 oaamm 0H0: .m— manna

115

 

0.0 0.0
0.0 0.: 0.0 0.0
0.0 0.0 0.0 0.0
0.0 0.0 0.0. 0.0
0.0 0.0 0... 0..
..0 00.. 0.0. 00..
0.0 0.. 0.0. 0..
0.0 0.. 0.0. 00.0
0.0 00.0 ..0. 0.0
0.0 0.0 0.00 00.0
0.. 0 0.00 0
000 4 +0.00 .0: 0000 +0000. 0:
200 :2 sscsaom 000. 02
0.0.1 0.0
000 0.. 0.0.1 0.0 0.01 0.0
0.01 00.. 0.0.1 0.0 0.01 0..
0.01 0.. 0.01 0.. ..01 0..
0.01 00.0 0.01. 0.. 0.01 00.0
0.: 0.0 0.01 0.0 0.01 0.0
0.0 00.0 0.01 00.0 ...1 00.0
0.0. 0 0.0 0 00.0 0
00.0 d +x\omo 0.3 4 +x\omo ammo. +v<oma
00 an zo< uncsaom 000. 000

 

000000000

.m— 3300...

116

 

 

0.0+ 00.0

0.0+ 0.:

0.0+ 0.0

0.0+ 0.0

0.0. 0.: 10.0+ n.0

o... o.n 0.0+ 0.0

0... 0.: 0... 0.0 0... 0.: 0.01 00..
0... o.m 0.0. 0.0 0... o.m ...1 0..
0... 0.0 0.0 0.. 0.0. 0.0 0..: mm.—
m.o. 0.. 0.: mm.— n.0. m.— o.mu o.—
m.o. 0.. m.m 0.. 0.0. o.— m.:1 00.0
0.0. 0.0 0.0 0.0 0.4. 00.0 0.01 0.0
0... 00.0 0.0 0.0 0.m. 0.0 n.0: 00.0
0.0. 0 :.o 0 0.00 o 0.01 0

sum 4 +x\mom . awn 4 +x\mom . 5.3 4 +x\mom . Emma +M\mom .
o< 0m :2 00:0 900>Hom

 

0osaaaaoo .0. manna

117

000: 003 Hm

.I.

 

 

0.0. 0.:

0.0. 00.0 .... 0.0
:... 00.0 .... 0.:
0... 00.0 0... 0.: 0.0. 0.0
0.0 00.0 0... 0.0 .... 0.: 0.0 0.0
0.0 00.. 0... 0.0 .... 0.0 ..0 0.0
0.0 0.. 0... 0.0 0.0. 0.0 0.0 00..
0.0 00.. 0.0 0.0 ..0. 0.0 0.0 0..
0.: 0.. 0.0 0.. 0.0 0.. 0.0 00..
0.0 0.0 :.0 00.. 0.0 0.. 0.0 0..
0.0 00.0 0.0 0.. 0.0 00.0 0.0 00.0
0.0 0.0 0.0 00.0 0.0. 0.0 0.0 0.0
:.0 00.0 ..0 0.0 0... 00.0 :.0 00.0
0.. 0 0.0 0 0.0. 0 ..0 0
2004 +0000 . 304 ..0.000. as: 10.000. 0004 +0000.

20: 110mooz 00 0:0 pqo>aom

 

000000000 .0. 00000

118

 

 

0.0 0.:
0.0 0.0
0.01 0.0 0.0 0.0
0..1 0.: 0.0. 0.0
0.01 0.0 ..0. 00..
0.01 0.0 0.0. 0..
:.01 0.. 0.0. 0..
..0: 0.. m.c. 00.0 0.0 0.0
:.01 0.0 n.00 0.0 0.0 0.0
0.01 0 Oman 0.00 00.0 ..0 0..
:.0 0.0 0.00 0 zz 0.: 0..
:.0 0.: 0.01 0.: 0.: 00.0
0.0 0.0 0.01 0.0 0.0 0.0
0.m o.n 0.01 0.0 :.0 00.0
0.0 0.0 0.01 0.0 0.. 0 20¢
0.0 0.0 0.01 0.0 0.0 0.0
0.0 0.. 0.01 0.. 0.0 0.0
0.0. 0.. :.01 0.. 0.0 0.0
0.0. 00.0 :.01 00.0 n.0. 0..
m... 0.0 :.01 0.0 0.0. 0..
n... 00.0 :.01 00.0 0.0. 0.0
0... 0 mom: :.01 0 0000 0.0. 0 0<
8.3 a +5.80. unobaom sand +0330— »:gaom 8004 +0330. unopaow

 

coaaapnou .0. manna

8 (ppm)

119

 

 

 

18-crown-6

0.04 M KPf-g>

+ Acetone
‘~-A.___A. -_A ...1-A"- DM F

" H20 (K I)
M01100

 

 

1 J I l

 

1.0 3.0 4.0

2.0
[18C6]/[K+]

39

Figure 29. K Chemical Shift y_§ Mole Ratio of

1806/K+ in Various Solvents

 

120

which is indicative of the formation of weaker complexes
than that in acetone.

The formation constants of complexes in these solvents
were calculated by use of the KINFIT program (101). The
computer subroutine equations are decribed in Appbndix IV.
The formation constants of K+-18C6 complexes in various
solvents are presented in Table 20. The formation const-
ants among these solvents are in the order; acetone:>
dimethylformamidei>dimethylsulfoxide)»water, which follows
the inverse order of donicity of these solvents. These
results are not'surprising, since, as mentioned in chapter
III, the K+-solvent interaction is a function of the
donicity of the solvent. A strong-K+-sOlvent interaction
would be expected in a solvent of high donicity and con-
sequently a strong K+-solvent interaction will prevent
the interaction between K+ ions and 18-crown-6. The
logarithm formation constant of K+-18-crown-6 in H20
obtained in this work is in good agreement with the value
2.06 reported from a calorimetric titration study (73).

As can be seen from Figure 30, the chemical shift
of the 18-crown-6-K+ complex is concentration independent,
which is indicative of an absence of ion pairing between
18-crown-6-K+ and the PF6‘ anion. The salt solution
however, shows some K+ PF6' ion pair formation.

The complexation reactions of the potassium ion

with dibenzo-18-crown-6 (DB1806) were also studied by

121

Table 20.Formation Constants and Limiting Chemical

Shift for the Complexation of KPF6 by

18-crown-6 in Various Solvents

 

 

Solvent Log Kf 611m (ppm)
Acetone >u. 4.46
Dimethylformamide 2.70 i 0.04 3.85
deethylsulfoxide 2.19 i 0.23 1.u3
* +
Water 2.17 _ 0.13 1.56

 

* KI was used instead of KPF6

122

. KPF6
13C)

12.5-

 

(K18C6fPFg COMPLEX

__.__().-———----1-1C‘r _ v-
(15-

 

nj

0 01 0.2 0.3 0.4

CONCENTRA110N1M)

Figure 30. Chemical Shifts of 18 C 6 K+ Complexes as
Function of Concentration

123

59K NMR. Unfortunately, this study is limited by the
low solubilities of dibenzo-18-crown-6 in most nonaqueous
solvents. In this study, the K+ concentration was held
constant at 0.02 3, instead of 0.04‘3, which was used
in the previous study.

The results are presented in Figure 31. As can
be seen in the case of acetonitrile and pyridine,
the curves level off at a mole ratio of ligand/K*—~1.5
and 2.5 respectively, which indicates that the complex
of dibenzo-18-crown-6-K+ in acetonitrile is stronger than
that in pyridine. This may be due to the much higher
donicity of pyridine which weakens the ion-ligand inter-
action. Unfortunately, in the case of acetone, the low
solubility of dibenzo-18-crown—6 limited the study after
mole ratio 1.25, however, obviously, the curve for acetone
solutions does not level off at mole ratio of 1.0 which
indicates that the K+-DB18C6 complex is weaker than
K+-1806 complex which the curve levels off exactly mole
ratio of 1.0 (see Figure 29 at page 119 ). The formation
constants of K+DB18C6 complexes in these solvents are
presented in Table 21.

The value of the formation constant of K+-dibenzo
18 C 6 complex in acetone must be suspect since not
enough data were obtained after mole ratio of 1.0.

The complexation of the potassium ion with 15-crown-

S has been investigated the same technique. Since the

121+

 

 
  
  
  
 

DIBENZO 18 CROWN 6

 

‘\

"a-.- ACETONE

 

 

 

‘8' “\A—ACETONITRILE
.10.
PYRIDINE
1.0 21.0 3.0
[CROW N]/ [KJ'J
Figure 31- 39K Chemical Shifts _Yg Mole Ratio Of

Dibenzo 18c6/K+ in Various Solvents

125

Table 21. Formation Constants of Complexes of KPF6

with Dibenzo-18-crown-6 in Various Solvents

 

 

Solvent Log K1 61(ppm)
Acetonitrile >4 -7.32
pyridine 3.42 i 0.11 -13.65

Acetone >3 -8.26

 

126

diameter of potassium ion and the cavity size of 15-
crown-5 are 2.62 and 1.72 (72) respectively, the potassium
ion is too large to fit into the cavity of the ligand.
The variation of the chemical shifts as a function of the
mole ratio of 15-crown-5/K+ is presented in Figure 32.
In acetone and methanol solutions, the chemical shift
goes down field until mole ratio of 1.0. Further addition
of the ligand reverses the direction of the shift. It is
reasonable to assume that the results indicate that
formation of a stable 1:1 (15C5/K+) complex followed by
the addition of a second 15-crown-5 molecule to form a
2:1 (15C5/K+) sandwich complex (81).

The formation of 2:1 (15c5/K+) sandwich complex
at the 15C5/K+ mole ratio 1.0 in acetone was also sugge-
sted by a carbon-13 NMR study. Since 15-crown-5 is a
symmetric ligand, it shows only one 130 NMR signal.
However, as Figure 33 shows, at mole ratio of 0.75 of
K+/ligand, two carbon-13 NMR signals were observed, which
correspond to the 1:1 and 2:1 complexes respectively.
The solid (15-crown-5)2KPF6 complex was precipitated
from methanol solution of 0.1‘g'KPF6 and an excess of
15-crown-5 (:>O.2'fl). The elemental analysis data are
shown in Table 22. The solid decomposed at 255°C.

As Figure 32 shows, there seems to be a weak break

in the curve at mole ratio 1.0 in nitromethane and

acetonitrile solutions, which is also indicative of a

127

 

APPM
i
J

 

Figure 320

NITROMETHANE

\ . ACETONE
" METHANOL

   

ACETONITRILE

1 1 1

1.0 2.0 3.0 41.0 5.0
[15 CROWN 5]/[1<+]

39
K Chemical Shift 1s Mole Ratio of 1505/K+

in Nitromethane, Acetone, Methanol and
Acetonitrile.

 

128

 

WW

13 .
Figure 33. C Spectrum of 15—Crown—5 at Mole Ratio

0.75 of K*/15c5 in Acetone

129

Table 22. The elemental analysis for (15-crown-5)2 KPF6

sandwich complex

 

c% 11% P%

 

Observation: 38.#8 6.56 4.97
Calculation: 38.k6 6.h1 #.96

 

 

130

probable formation of 2:1 complexes. Also, it is interest-
ing to note that in all of these solvents, the limiting
chemical shifts of the curves seem to approach the same
value, which is another evidence for the 2:1 sandwich
complex in these solvents. It is reasonable to assume
that in all the nonaqueous solvents used including
nitromethane and acetonitrile, the 2:1 sandwich (15-crown
-5)2K+ complex can be formed with excess ligand, where
as in water only the 1:1 complex has been reported (73).
The complexation reaction of K+ with 15-crown-5 in
pyridine, propylene carbonate, dimethylformamide and
dimethylsulfoxide were also investigated, and the results
are illustrated in Figure 34. In prOpylene carbonate
another V shaped curve is obtained. In pyridine, dimethyl-
formamide and dimethylsulfoxide solutions. There are
also very weak breaks at mole ratio of 1.0. These results
seem to indicate the formation of 1:1 and 2:1 complexes
in these solvents. The formation constants for 1:1 and
2:1 complexes can be obtained by fitting these chemical
shift data by using KINFIT computer program. One typical
fitting is shown in Figure 35. The computer subroutine
equations are described in Appendix V. The formation
constants for 1:1 and 2:1 complexes in various solvents
are shown in Table 23. As can be seen, the formation
constants of 2:1 complexes seem to decrease in the order;

nitromethane:>acetone:>pr0pylene carbonatez>pyridine >

131

 

-l
«E

 

 

 

1.6 2.0 31.0 4‘0 51.0
[15 CROWN 51/[K+J

Figure 31.. 391: Chemical Shift .Yé Mole Ratio of 1505/16“
in Pyridine, PrOpylene Carbonate, Dimethyl-
formamide and Dimethylsulfoxide.

 

1:322

w

 

u—c-n-no—u-fif --_~v-.-_b-" u—o—u-c—‘f

83% So: m». 3.3%

'
m"'-m"' ‘Alv-'

.Hoqmnpoz ad +x\mom_ no

vamUn. VO_._.<~_ EOE

IOUSmUW—anxfi 0000000000000.00000.
_m-u--m----m----m----m----m----m----m----v----m----m----m----m-u--w----v----r----w----m.-.-v----"----.
I I

I u

u u w

w

n O

O n

a
u

a ..

m

o m.
‘ m
u

.u

OK
- ....U .-...0.

OK

0

d

I'm----"--I'V'-'nlf---lml'--MUU‘I'V'---V"'-mII--.L'-'-m---'¢----¢--'aim--I'w"|I." I‘--o 0-.-.

Hmoaamso on» now mnappam nonhumaoo .mm ouzmam

dev

133

 

maw>fiuoommou moxwamaoo paw uaw P". you qunm Haofiaoso mnfipfiaaa on» and N» van .6

hao>auoommon m¢xmanaoo _«m was

 

pup non mpampmaoo mafipwauou on» ohm NM can —x Adv

i

 

:5... .o.mu no.0 H mm._ o..0 n _m.~ ovfixouaamamnpoafia
mo.N— mo.w m..o H um.~ mo.o H F—.: ovaaaahomahnaoadn
3.2 moé 9.0 H mo.m : A 35:32
mo.m— n_.¢ ...o H :o.m mAn maanufinoucod
mm.m. mm.. $0.0 H m..m m.“ onavfiumm
m:.m— mm.m m_.o H mm.m .gA muaqonuau onoammoum
mm.—. mm.m o..0 H N_.m mA“ cacaoo<
2.... m9? 86 H is :A ofifioaoufiz
Aammvma Aammv—e Na won Adv—M won pa¢>aom

 

munobaom msoanw> :H moxmamaoo +x|muabonuump

."m cam .u. no“ mpgwnm Hmofiaono mnapaaaq can mpqaumqoo aoaamsgom .mm.manaa

134

acetonitrilei>methanolI>dimethylformamide:>dimethylsu-
lfoxide. The order observed is the inverse order of
donicity for these solvents except for pyridine, but
does not correlate with the dielectric constants of these
solvents. For example, although prOpylene carbonate (PC)
has the highest dielectric constant of all the solvents
used, the formation constant of the complex in PC is
larger than that in most solvents. Therefore the donicity
of the solvent seems to be a very important parameter
in the complexation reaction.

Pyridine has the very low dielectric constant of
12.0 which may have some effect on the complexation,
however, it is also a "soft' base and despite its high
donicity, it may not solvate strongly the alkali ions
which are "hard" acid.

It is seen in Table 23 that the chemical shifts
for the 1:1 complexes ( 61) seem to be solvent depend-
ent. This is reasonable, since, in the 1:1 complex, the
solvent molecules have a ready access to the cation.
However, for the sandwich complexes: the cation should
be insulated from the solvent, the chemical shifts are
largely solvent independent.

It is surprising that the formation constants for
1:1 complexes in various solvents are always very large.
Even in solvents of high donicity such as dimethylfor-
mamide and dimethylsulfoxide, the values of log K1 are

 

135

4.11 and 2.91 respectively. The results were checked by
carbon-13 NMR. The results are illustrated in Figure 36.
As can be seen in the acetone case, the C-13 chemical
shifts reach a limiting value after mole ratio 1.0 of
K+/1505, which indicates the formation of a very strong
complex. Even in dimethylsulfoxide, in which the chemical
shifts reach a limiting value after mole ratio 2.0, the
formation constant (log K) for the 1:1 complex obtained
by fitting the curve is 2.78. The results are in good
agreement with the K-39 NMR result.

The influence of the anion on the formation of the
sandwich complexes was investigated in these solvents.
The results shown in Table 24 clearly show that as expected,
the nature of the anion has no influenCe on the 39K limiting
chemical shift and therefore, there is no evidence for
the complexed cation-anion interaction. Even for the 1:1
complex the data in Figure 37 shows only a very small
degree of ion pairing between (1505K)+ and PF6', there is
only a very small change in 39K chemical shift with com-
plex concentration.

The complexation study of the potassium ion with
monobenzo-lS-crown-5 in various solvents was also perfo-
rmed by potassium-39 NMR. Since the attachment of the
benzo group on crown ethers has been reported to diminish
the bascity of the oxygens and to reduce the cavity size
of the ligand (55), weaker complexes of potassium with

136

 

.mpao>aom

 

 

 

 

$3.3, ca menial ac caucm 30: mm 32m 3338 2.5898 .mm enema
G 2305 2:35:
as n... a... 3 3 no o 2 ca 3 o.F ....o
q . . a . a q _ u _ a
. ..Nduu
.1 1 . o< .99
on. .

 

 

 

 

de V

137

 

o.m cause caea +xxmom. he caeawan man ac caeeae chad one Aev

 

 

m.mn «.0 u m.~_ Ha
m.:m «.0 H m.m_ zone accesses
¢.mo m.o H o.~. Hm
m.nm ~.o u o.~. zone
u.¢m m.o u o.~_ mama ceaaaancuashaeaaa
..mm «.0 u m.m. 20mm
m.:m m.o H m.m— mama mafinaaaouood
Aaaav heasm Hacaacho madam paeeaom

Anmvxcvhhea: enag

 

muaobaom macaum> as munsououm— and: madam

aaamncaca ac nexeaaaco an» ace mzz man no cheahm aceaaeau waaeaaaa one .em canes

138

 

l: 1 COMPLEX
5 (K15 CS)+PF3

’\ o

1!

IAI>PAA

6 «PF(,

 

 

 

ql 02 03 CA Q5

CONCENTRATION (M)

Figure 37. 39K Chemical Shift Variation with the

Concentration of 1:1 (1505) KPF6 Complex
and Potassium Salt in Acetonitrile.

139

monobenzo -15-crown-5 than with 15-crown-5 would be
expected.

As can be seen from Figure 58, in nitromethane and
acetonitrile, the limiting chemical shifts seem to reach
the same value, which is probably due to the formation of
2:1 sandwich complexes. The formation of both 1:1 and
2:1 complexes of monobenzo-15-crown-5 with K+ in the
methanol-water mixture has been reported by Izatt (105).
A solid MB 15 C 5 KPF6 complex was obtained by precipit-
ating it in methanol solution. The melting point was 2A8
“925000 and the elemental analysis is given below (Table
25).

Table 25. Elemental Analysis of (monobenzo 15 C 5)2 KPF6

 

 

0% H% P%
Observation: h6.62 5.5h h.41
Calculation: 46.67 5.55 4.31

 

The formation constants for both 1:1 and 2:1 complexes
obtained are presented in Table 26. In the acetonitrile
case, the formation constant of 1:1 K+-monobenzo 15 c 5
complex is smaller than that of 1:1 K*-15 c 5 complex.

The 2:1 complex, however, shows a formation constant

(log K2) for the MB 1505-K+ complex which is larger than

“+0

MONOBENZO 15 CROWN 5

C>

I

 

A PPM
11":

 

 

 

 

1.6 5.0 31.0 4.0
[MB 15c 5mm

Figure 33. 39K Chemical Shifts l§ Mole Ratio of
MB 1505/K+ in Nitromethane and Acetonitrile.

“+1

 

 

 

as... mm.m_ mo.o H s..¢ .sa m o m.

ss.m mm.m_ sA ea . m o m. m: oamnaoaoaoaz

mm.~_ n..s ...o H so.~ ma m o m.

am.m ~_.m s_.o H ms.m sm.o H ms.m m o m. m: oaaaaasooooa
anmm a 3mm. ¢ mw moq ; m3 «Emma anofinom

 

noggom 36.2.; a.“

mlgouoIm—Iouaopoqoz gawk mag no moxoamaoo no mandamsoo dogma—nah .mm 933..

1#2

that observed in the 15-crown-5 case. In the nitromethane
case, formation constants for 2:1 complexes in both 15-
crown-5 and benzo-15-crown-5 cases are large. It is not
unexpected since the nitromethane has low donicity of
2.7, it may solvate the K+ ions weakly.

The complexation of 12-crown-h and potassium ion in
various solvents was studied by K-39 and C-13 NMR, and
the results of 39K NMR mole ratio studies are illustrated
in Finger 39. As can be seen, in all cases there are no
obvious inflection points or breaks in the curves. It is
very difficult to say whether or not a 2:1 complex is
formed. The cavity size of 12-crown-h is —~1.22 (72)
which is very small for the K+ ion (~v2.7X). Thus it is
possible to form both 1:1 and 2:1 (1204/K+) complexes in
these solutions. However, the 2:1 (1204/h+) complex could
not be isolated from KPF6 methanol solution in the same
manner in which we obtained the 2:1 (15C5/K+) complex.
Therefore, no evidence of the formation of the 2:1 complex
was obtained.

In order to get more information about the behavior
of the complexation in this system, the ‘30 NMR mole
ratio study for this system was performed. The data are
presented in Table 27. The plot of C-13 chemical shift of
12-crown-u'zg mole ratio of K+/12-crown-4 is shown in
Figure 40. Since the 12-crown-4 is a symmetric ligand and
because the exchange between free ligand and complexed

ligand is fast, only one 130 NMR signal was observed at

 

143

 

12 CDRCDVVPQ A.

16' \ ' CH3NO?
[JACETONE

   

. MeOH

 

 

 

 

 

 

1.10 210 310 1.0 EU 6.0
[CROWN]/[K+]

Figure 59. 59K Chemical Shift‘zg Mole Ratio of
12—crown-4/K+ in Various Solvents

144

 

 

 

 

m_.mm o.m
w:.mm m.—
oo .30 o. P
_w.wm m.m mm.mm mn.o
No.00 o.m mh.om n.0
_s.mm m.. mo._s o onoooo< mama
mm.mm 0.. no.5m o.m
Om.ou n.0 wmfib m.—
sm._s o Hoaosaoz zommo sm.om o.—
:m.mw m.m Po.mm no.0
mw.om O.N mm.mw no.0
:m.Pm 0.. mu.mm om.o
MN.NN. n.0 ww.o~. mm.o
ms.ms o omza mama sw..s o oaaaoanoaoos mama
S\+5 . 3\+5
as“ Q 0.3mm mac: pso>aom 3mm 8mm Q 0.3mm oaoz psopaom 3mm
n+mo .+x n +zv muso>aom
msoflum> aw mmxoamsoo slssouo1mpl+z mo mamasm Hmoaamno mzz m—Io 0:9 .mm manna

145

 

74

  

72

APPM

70‘

68 "

13C

  

K"- 12 GROWN 4

 

0
\‘F‘ ACN

 

 

0.5 1.0 1.5 2.0 '2.5 3.0 3.5

+.
(Kl/(CROWN)

Figure 40. Carbon-13 Chemical Shift‘vg Mole Ratio of
K+/1204 in various Solvents.

146

various mole ratios. Even in the solvents of low donicity,
such as acetone and acetonitrile, the 13C resonance does
not tend to reach the limiting value before the mole
ratio 2.0, indicative of the formation of a weak 1:1
complex. In the dimethylsulfoxide solutions, the C-13
chemical shift continue to change even after mole ratio
3.0, which suggests that the 1:1 complex is quite weak

in this solvent. It is seen that in all cases only

evidence for the formation of very weak 1:1 complexes

 

was observed. There is no indication of a 2:1 complex.

If only the 1:1 complex is assumed to be present in
various solvents, the formation constants of 1:1 complexes
can be obtained by fitting the curves for both carbon-

13 and potassium-39 NMR and are presented in Table 28.

As can be seen, the formation constants of complexes

from both C-13 and K-39 NMR studies are in good agreement
with each other in acetonitrile, acetone and dimethyl-
sulfoxide cases. The results do not definitely rule

out the formation of a very weak 2:1 complex. As Table

28 shows the formation constants of 1:1 complex in methanol
and dimethylfoxide are obviously smaller than in aceton-
itrile, acetone and nitromethane. In the methanol case,

it is possible to envisage hydrogen bonding between the
methanol molecule and the oxygen of ligand which would
probably prevent or reduce the extent of complex format-

ion. The very weak 1:1 complex in DMSO is not unexpected,

1#7

 

 

 

N... H 0s.00 s..0 H s0.0 mans

0m.0 H 00.0w no.0 H _n.0 anus coaxoaaanaasoosaa
0m.0 H ,0.0 s0.0 H m0._ onus Hosanna:
55.0 H mm.m Pp.o H um.. onlx onmanSOAuaz
mm.0 H ms.s0 50.0 H s0.a Mano

0s.0 H m0.s o..0 H 0a., onus onerous
00.0 H .0.s0 s0.0 H 0m.m mzz n.10
sm.0 H s0.m o..0 H 0..m mzz anus oaaanaaooooa

Assay chasm
Hmofiaono AM moqv
wqauasfiq pampmsou soapmsuom 00:90: usm>aom
musm>aom
wsoaum> ca moxoamaoo +Musussouolmp on» you mpqmumnoo mafiamsnom .wN canoe

1#8

since DMSO has high donicity and can solvate K+ easily.

It is of interest to study the Cs+ ion complexation
with 12-crown-h, since both K+ and Cs+ ions are too large
for 12-crown-4 which has about same size as the Li+ ion.
The 13C NMR mole ratio study for 120u in the Cs+ case
is shown in Figure 41. In the 12-crown-4 case, the 13C
chemical shift of ligand does not tend to reach the limit-
ing value even at mole ratio of 2.5, which is indicative
of the formation of a very weak 1:1 Cs*-12-crown-h complex.
The formation constant (log K) of the 1:1 Cs*-12-crown-

4 complex obtained by fitting the curve is only 1.09
(the data is presented in Table 27, page‘H#+ ). Since
the Li+ ion has about same size as the cavity of 12-
crown-h, the Li+-12-crown-h complex would be eXpected
to be strong. Surprising, the Li+ ion was also reported
to form only a weak complex with 12-crown-4 in acetone
(106). It may be that due to the very small size of
12—crown-h (only four oxygen atoms on the crown ether
ring), the attraction between the metal ion and the ligand
is so weak. As can be seen from Figure #1, in the 15-
crown-5 case the curve levels off at mole ratio 1.0,
which indicates the formation of strong 1:1 complex
(formation constant (log K):>4). This is not surprising,
since 15-crown-5 is a larger ligand than 12-crown-h and
the attraction between metal ion and ligand is stronger.

Even for Li+, which is too small for 15-crown-5, the

 

1119

 

13
C Cs /MeOH

74 "

72
2
CL
0. 15 CROWN 5
<1 A '

70"

12 CROWN 4
681

 

 

 

 

4

0,5 1.0 1.5 2.0 2.5

Cosh/(CROWN)

Figure 1+1. Carbon-15 Chemical Shift 1s Mole Ratio of
Cs/Ligand in Methanol.

150

formation of a strong Li+-15-crown—5 complex in acetone
was reported (formation constant :>103) (106).

Haynes, gt'gl (107) found that a good correlation
between the 23Na NMR chemical shift of the complex of
the Na+ ion with antibiotic ligands such as monensin,
enniatin B and valinomycin and the stability constant of

the complex formation in the same solvent as

5 complex = 60 + m log Ks (IVB.1)

 

where 6 o and m are constants and Ks is the 'stability
constant for the complex. According to the above equation,
the strongest complex would be expected to have the lar-
gest shift. However, as shown in Figure #2, in acetone
solution, DB18C6 has the largest 39K'NMR shift but it
does not form the strongest complex as mentioned previou-
sly (page 123 ). For cryptands, as shown in Table 29,
cryptand 0221, not cryptand C222, gives the largest shift
(A5) even though it does not form the more' stable
complex with the K+ ion than cryptand 0222. These
results indicate that the 39K NMR chemical shifts of
complexes do not correlate with the stabilities of
complexes in these macrocyclic polyether systems. This
probably suggests that not only K+-ligand attraction
force but also another factor such as repulsion force
contributes to the paramagnetic shift of 39K* upon
complexation. The K+ ion was reported to nicely fit into

151

 

 

 

 

ACETONE
14.
15C5
12x. ”.A-——--'O'-----..
- ‘. \ . ,‘O",
10* .\' . I2C4
81 . ‘
6»
E 4 A n o 18C6
O.
O. 2»
‘4
O.
-2.
-41 A
-6~ ‘\\‘__ oa1aco
o 1.0 2.0 310 4:0
(CRoww)/(K+)

Figure he. Mole Ratio-39K Chemical Shift Study for

Various Ligands in Acetone.

 

 

152

m
A.Hnecsv \ P n ma mpficscwcscnsd cacflm caposwms was spud: mafia

a.
Hanspmn anon scam :cfiusnfiupsco mswcsacsa cad» scdpmxmaon dammlsfimm on» ma :9 Anv

.
cscpccm dd +m 009% you guano Hmcaacso u comma menu» 1 paaflas “lac Amy

 

 

 

 

mn.m1 m0.m s.m0 mm.s+ sussoacnm.
s0.m1 s..0 ..0n no.0. mussoacnm_
00.0w- ms.m . s__ 0m.01 01seca010_1csscsa0
s..01 ms.m . 0._n_ 0s.s+ muasoacuma
:o.w.1 wm.m 0.5m ¢:.m1 _pmo uqmumauo
mm.mm1 m0.m 0.mm no.0.1 .mmo camsssao
as.s.1 ~0.m 0.50 0s.m1 mmmo oscbasao

Access usasaasv mxsmv 0.. m: as Asaav scamsco .a. com

3320: 3cm some 3.20 £3; 83 $20 Hocascno 3333 0503

apps .1.
mscpco< :H mocmmHA caachocaomz mscHum> hp mama no
scfipmxcansco can you mmIm up mnpuas mafia 0am mumanm Hmofiacso msfipfiaflq .mm canoe

153

the cavity of 18 C 6 (81). As mentioned previously,

the attachment of the benzo group on the crown ether
ring results in decreasing the cavity size of the crown
ether (55). the DB1806 would be expected to have smaller
cavity size than 1806. Therefore, when the K+ ion is
inside of the cavity (81) the repulsive interaction F
between the K+ ion and the oxygen atom on the crown
ether would be larger in the DBC18C6 case than that in
the 18C6 case. The large repulsive interaction probably

 

 

result in a large paramagnetic shift. The similar behavior
is observed in the cryptand C221 case. The cavity of

0221 is too large for the K+ ion, when the K+ ion tries

to fit into the cavity (The K+ ion was found on the
margin of the cavity of 0221 from X-raystudy of K+-

C221 complex crystal (108), the repulsive interaction

also will be induced.

CHAPTER V

RECOVERY OF CRYPTAND FROM CRYPTATE

 

 

INTRODUCTION

 

Cryptands C211, 0221 and C222 are commercially
available but at a rather high price. Therefore, it
seemed useful to deveIOp a technique by which the ligands
can be recovered from used solutions of their complexes.

It has been shown by Lehn,lgt.§l(86) that when the
two nitrogen atoms of the bimacrocycle are protonated,
the ligand is in the exo-exo form and has very little
complexing ability. Lok (109) also reported that the
sodium C222 cryptate dissociates into the protonated
cryptand and free sodium ion at pH 6.7. The recovery
procedure is based on these observations.
‘RESULTS AND DISCUSSIONS

The recovery procedure, shown diagramatically in
Figure 43 involves four steps. 1. Recovery of solid
cryptate complex from solutions. 2. Preparation of
aqueous solutions of the cryptate and release of the
captured cation at low pH. 3. Separation of the protonated
cryptand from the metal ion(s) on a cation exchange column.
4. Conversion of the protonated cryptand to the basic
form and purification.

Solutions of metal cryptates in various solvents
were dried in m at ~1O'2 tcrr at room temperature (M
must not be done if the solutions contain Cloh' anion).
The solids (O.1~?0.3 g) were dissolved inv~c20 m1 of aqueous
6 H HCl with gentle heating and the solution again

154

 

155

 

C222 M+X— I

solution

 

 

5011

[11+ (2)

0222K; , [W I 1
i CATION EXCHANGE (3)

 

622-5M+X‘ J 1

 

 

 

 

022m;

J ANION EXCHANGE (4)

0222 FREE 1

 

 

 

 

 

 

'Figure 43. . Diagram for Recovery of Cryptand from

Cryptate

156

evaporated to dryness at «v10'2 torr. The residue,
containing alkali salts and diprotonated cryptand, was
dissolved in 20 ml of aqueous 0.1 fl HCl. (If the original
solutions contain a variety of anions, it is useful at

this point to convert the salts to the chloride form by

 

an anion exchange column in C1" form). The solution II
was then passed through a cation exchange column in H+

form (Dowex 50 x 8, 100 -200 mesh, 1.2 x 22 cm column).

Metal ions were eluted with 150 ml of 1.0{M 301 while

the diprotonated cryptand remained on the column. Second 5

elution was then carried out with ..120 ml of 6{fi HCl.

In an experiment involving sodium-222 cryptate the elution
of the alkali cation was followed by atomic absorption
while that of the cryptand, by proton NMR. The elution
curves are illustrated in Figure 44. The experimental

data for atomic absorption and 1H NMR are presented in
Table 30.

Then, crystals of the diprotonated ligand (C222
o2HCl) were obtained by evaporating the solvent from
the cryptand solution under vacuum. The crystals are
redissolved in'~'2 ml of water, placed on an anion exch-
ange column in the 0H" form (Dowex 1 x 2, 100t~v200 mesh,
1.2 x #5 cm) and the free base eluted with conductance
water. Sometimes it is necessary to repeat several times
this step until most protonated cryptand was converted

to the free cryptand C222. Finally the free cryptand

157

 

 

mmmmmo m0 acdpmawmcm mom c>asu ncausam .J: casmfim

 

 

+mz Scam +m .
QEV w_>_DI_O> FZMDJM
OWN Om_ ON_
_ _ . 0.111.. n pl
_UI Z® iA-1 _UI Z—

 

 

 

 

 

 

 

_ 0.0
No.0
mod
v0.0
mod
$0.0

(W)

$142223 sum 40 No1lvsiN33Noo

158

 

soaosaon scaoaac H002 0 as n0 can on csaccaom .sv

 

 

 

n10. a s..m 1 no.0 1 0N.

N1o. a 0a.. 1 nn.0 1 00

N10. a 0m.s 1 0.. no.0 n0

N1o. a n0.. 1 no.0 ... 00

n10. a 0m.s 1 ..0 n.n n:
n10. a :..m 1 no.0 0.. on Homz o

1 1 1 1 on.

1 N10. a 00.0 1 0.n om.

1 N10. a 0..0 1 0.: no.

1 N10. a n0.o 1 n.0. 00

1 N10. a 00.n 1 0.00. ns

1 N10. a os.n 1 0.00 00

1 ~10. a nm.o 1 0.n n:

1 1 1 0.. on
1 1 1 0.. n. H002 .

ad .5 mmmo .80 .0103

.mmmoo .+szo Asvaoamscan mzz meannessH .m.s .Hso osaac> soassam
macasaacm soassam as mmmo one +cz so scassaosocaco .0n canny

159

was obtained by dring in vacuum and recrystallized from
hexane. The purity of cryptand 0222 was confirmed by
melting point, carbon-13 proton-NMR and flame emission
technique which was applied to detect the content of Na+
impurity. The yield is about 601~c80%. The recovery of
the cryptand C211 from cryptate was performed in the
similar procedures.

Since the recovery is based on the protonation
of cryptand. It was of interest to investigate the pro-
tonation of cryptand in more detail. The carbon-13 and
1H NMR were applied to study the protonation. These
results studies are presented in Table 31 and Figure 45
and #6. As can be seen, both 1H and 13C NMR spectra of
protonated cryptand show so much difference from that
of free cryptand. As can be seen from Table 31 in the
H20 case the change in C-13 chemical shift by protonation
for carbon 1 (QQHZ), carbon 2 (QQHZ) and carbon 3 (NQHZ)
are -0.56, 5.3A and -0.66 ppm respectively ("+" means
upfield shift). The assignments of peaks for carbons
was made by Lehn (86). As can be seen, the one of QQHZ
carbon, not NQHZ carbon give the largest shift. The
identical phenomenon was also observed by Lehn, 33,3; (110).
Since it is well known that the protonation is easier
occurred on nitrogen atom, not on oxygen atom, it is
reasonable to expect the largest shift for NQHZ carbon.

These results are surprising. There are some probable

 

 

 

 

 

 

Au\ / .
.oooo.o . fi\\// \\\//4
2:3 .3 3 an Sontnn zx/b/b/xz
o . .1111. .
.000..0+.mo a no +0 .oo.m 00 n n
Axmamsccv.:.01 0. mm..m ANV .
AooaMV m.O1A.V 0 m0 onao.m0
0 mm.mu A:
6 Owe
1 no vampmhao Ammmo\+xv n.0 n m: an mmmo
00.. ccosacocam ossoaaao + mmm cons
Am sonamcv mm.m+ MNHMM .N.mm Anv
0m.o1 os.mw .1Hmfl1 on..0 .m.
00.01 mo.nn 0..m0 ..0 000: mmmo
Am conumcv :m.m+ mm.m mm.:m Amv
on.01 on.mw +2 .0.00 Amv
2004 00.; CV 0mm mmmo
nag“; 8994 £509
ocpmscucamo 1 comm 1 campmhuo
1 ad 0 .
0 unscucam uncumaao 009m uqc>acm ucmumhuo

 

mfld
o cameo sonsccooam 0c mssasm Hocascso n.1sohaso

..m canoe

161

‘30 FREE 0222/H20
1 1 3 70
71.8 72.4 54.7 3 n

 

 

 

 

 

WVWWJWTNMJAKV 3l/2\oflo/\ 11

72.4 65.4 55.6

 

 

1

1’1‘11111‘

1 r1

 

 

NI" “$11441 \LWAVJ 111““. “1‘1 (“111

Figure 45. Carbon-13 Spectra of Protonated and Free
Cryptands

‘HNMR
NH

(\O/TO/\ FREE 0222
N/\\\J5h\\\~fif\\\JN (:11

 

 

C222 H3“

. . 1 l 1 1 A i_n_J
4 3 1313'“

Figure #6. 1H Spectra of Protonated and Free Cryptands

163

eXponation as following:

(1) It is caused by 3 effect. The attachment of H+ on
nitrogen causes the big chemical shift for S carbon
(carbon (2)) (110).

(2) The assignments of 13C NMR peaks for carbons is
probably not correct, however, this is less possible
since the QQH2 carbon, (not NQHZ carbon) usually resonates
around 70 ppm, for example, in the cases of 18-crown-6,
15-crown-5 and 12-crown-#, the QQHZ carbons all resonated
around 70 ppm as mentioned in Chapter IV(B).

(3) The big change in 130 chemical shift for carbon (2)
(QQHB) probably suggest an extremely structure change
upon protonation. As also can be seen from the 1H NMR
spectra shown in Figure #6, the peaks for NCHZ and OCH2

protons in protonated cryptand show so close, which

indicates that the all protons have the similar enviroments

while the so different enviroments were found for NCH2
and OCHZ protons on free cryptand. This is probably
another evidence for an extremely cryptand structure

change upon protonation.

.. fi".1
\.

 

APPENDICES

 

APPENDIX I
DESCRIPTION OF COMPUTER PROGRAM KINFIT AND SUBROUTINE
EQN FOR THE CALCULATION OF ION PAIR FORMATION
CONSTANTS BY THE NMR TECHNIQUE

The equilibrium for an ion pair reaction can be

expressed as

 

 

 

 

KIP _
M” + x“ .——— '- 11“ - x (11.1)
and + _ C E
K _ (11 - X) _ (11x) . 1
IP ' _ - Y 2
11+ - x CM+ OX- 1:
1

where Kc is the concentration equilibrium constant,

1’: is the mean activity coefficient which can be calculated

by using Debye-Huckel equation:

6
1.82 10
(:53/2 I“! I

' 108 7+ = (A03)
50.29 -o

1 + —————_ ' a

(UNI/2

 

where Z+ and Z- are the charges of the ions, I is the

molar ionic strength, D is the dielectric constant of

16#

165

the solvent, T is temperature (OK), 8 is the closest

0
distance of approach of the ions in A. The values of

8 for KPF6, KI and KSCN were used to be #.0, 3.0 and

5.0 X (102)(111).

The observed chemical shift is a pOpulation average

 

 

 

 

of these of the free ion and ion pair I"
dobs = GFXF + bipxip ‘
= (6 F " 61p)XF '1' in (A04)
CM
where F
CT
Mass balance leads to
_ M M
[M+A]= CT " CF (11.6)
and charge balance to
- M
[W] = [A] 5" CF (11.7)
substitution of (5) (6) (7) into (#) yields
1 2
M -1 i (1 + 4chg) /
CF 2 (A.8)
2Kc
substitution of (8) to (2) we obtain
M 2 1/2
-1 + (1 + #KIPCT 7t )

obs ZKIPCT +

(A.9)

166

In order to fit this equation, four constants and two

parameters are used; namely

 

Const (1) = chemical shift of free K+
Const (2) = dielectric constant of solvent
Const (3) = temperature of solution (OK)
Const (#) = ion ion closest distance (2) F
U (1) = chemical shift of ion pair
11 (2) = Kn,
In the cases of solvents of low dielectric constant 1
1

such as acetone, pyridine and ethylenediamine, the
Debye-Hukel (DH) equation seems to be invalid at high
concentration of the potassium salt ()>0.2,M). For
example, the value of 7; obtained for 0.2,! of KPF6 in
acetone is only 0.17 (aCtivity = 0.03h). The value of

KIP for KPF6 in acetone obtained by fitting the data

to equation (A-9) and the DH equation is only 0.03

while the concentration equilibrium constant for ion-pair

formation (Kc) is 8.25 I 0.17.
The value of Kc can be obtained using equation (A-9)

with?+ = .0 and only one constant (ie: chemical shift

of free K”). The subroutine EQN for calculations of

KIP and Kc are listed the next two pages.

(“WOFHNOCWD

167

K

1P
5”“ROHYINE EON
C0“N0V KOUN1. 11APE1JTADE.141.Lnn.x[uCc.uno1.nov :.‘1wv-.(.
1H1‘01ffi101 AVOD C110 IthE‘varYr-XI04"Y“0 l110‘Opo‘O-' .7.
210VAL-v51.1.nt.L q.111.v.111.v£C1.ucst.to':SY.-1Lm 3.11.1 01,,

JYYYoCnv51S

Covucw/FREDTIIVEYH

C0”“c~7001~1/Kcu1.Jnnr.xt\

OIPCh§1o~ x1e.zoa1.n1?o1.w1x1o.100111x~1t1.r101 lo1orv1‘-o~.
1.p(?g.>11.vsc11201211.zL11001.11120%1.I|r\:v(PC .\<'1"v1.‘l
2011101.CONS 5(Soclb1.~cfi1150101W”111“11-G'11°1 11.11111-c1.
3.00p11sc1.Lop11501.vvv15011c0V5111b!~111||31

GU 10 1?.31“.51|07030°010011017‘ 11‘”

‘ CUIethO'E

1150 :A0

J!Ap[:h1

welt: 1JTA°E161

s ro~HA11/////////.° KINF11 CU°VL Ex1°\’0Lh 09’)
nonuxza

NOVAG:7

pElHQN

7 CONTINUE
at1uaw
A CU'-11‘1HE
R€1ucw
? (DNYIN'E
1“! £1 H.~E.-11 GO 10 3%
a: 1uc-1
35 CUH1INHE

CU“51111= S OF F9 8: K

cons1171=015L . C HSY. of SOLVLNI

CUH§11118 15"”. Or SOL.

CUhS11b)* 10H [ON 01<1. 1A1 Cu

“Untouc. (1" K

XI171= 08%. CNE“. 9H1?

U111: < or 10H PAID

H121:K[P

AVY2!XCI1

cOhcxsxx111

no 0? Pa” JPHtlolo
S Uw=Cn~CK
1‘1910u182163.83
a; «911(117Ap:.ns1 xx111.JrH.u1?‘ Gfi"“'
as rujgnt16x.£1a.e.15.r16.n.tla.~
UHIINHE

SlnN=1= $0911510u1

'0 £031./2.

V? AD:I./2.

a on:-1,e?3E 06/1cow511?1-C0a511311-0-F10

H nn:§,079[oo 09/1C0w51171eCnh<11311o-vEI-

6:HL0421H10A0510~011/11.00H11RecowsttaI'SIOucII

nn~«n:10.oeonA~LOG

C4ID:II(?1'GAHHAOOP

cuhc::1-},csnu1(AUS(‘.0¢,OCK|°'IK1|111|'12.°t(l°|

olfr ans cnurK-ANY1

rv‘C2 AHY°0.05

11101'f.L1.FpAC1 Go to 290

go luv: (“MCK
o xf=C0erlxx111
onanssxxecor<11|101l.‘xK)OU(11

H1510=0EL0ns-XX1?1

7 111014

(Ur-1 11111E

UHN

COhYINHE

nzluc:

5 C0"1HE
xrljnfivu.N£.-|I 00 10 ’0
Htlupa

23 COhYIMHC

RtIUDV

Q CU'-11N'1E
atlnqw

Id CONIINHE
RL1UGN

I 1Jh1 N11
‘ SVIHAM E

 

PG

h.

40.1

as

on
..

b

15

.I‘IC‘U

_- d .— V
\l w k 0
mnnnnm¢nnn9m~ntncnpn

168

71/!1 0P1=1 11H a.u'~11 17113/17

( rpnuvlv J' ['11:

’“-» ~H111.I1AU1.11‘I[.141 Lhucllwcn. 11-1.vHJLn.“n11'.a.1.114A1. Long
1vl'o1'k1obran-«$111.IL-J‘HWHI1'“.-1.--11....-H.111y.r1,.1..,ph..11).: ('3...
’1' "LLO'rIeIeHICL‘."I'IOVOI'I‘JLk'":L"0'2"".)Ie'l)‘IeJ10h10V"‘”Ie1 1". L)..-
3vvv (u1c1

( ""’I"‘|1/|'11H

1 ‘1‘./|'11:1/'LUI.Inn1.17a

111W! [(6.1031. I11,’ 1310:1111".1u01bell1u1d‘1101300 101-01.,00107111‘00) En
le“"~0’I"-'L"/.O/l‘e.w1(JuU'OAU‘3010111VA|(70).!SICIJU).Y11L).
*“"‘ 1-1'11'5150'1”|-Ur‘"5v1'|5”lw|>0|cV111015H1-H111(Su1.1Ul1501
Jo'49‘1"10LH01‘HJ’0YVV1HJDOLHJ 11101.1I-1|,1

I. If. I'leCohesolo7eaogeIUe|10|"I Ill"

C.-111HE

l:~u€:0:

1' J{:a1

u't1i 1113u1.n1
‘J‘J5'1/////////e
N -'I‘J'=9

H MIN-3:9

$4 .' be"

H
C
‘1
C 111””:

1'3"" IHOHEO'I, (‘0 I", 39

P”I1~H

Crfillw E

S=‘1- .V°\O"7“HS‘I.J°(&.0*U(£’°A81l)11))/12.0°U(2)oxn('1g'.¢c3~91
1.1-11111.u1|1

zglrzs-ax121

,.|1;'.'

I‘c,“J'|E

IUfia

M-Il' ”E

vtlurtHcNE.-l1 60 10 20
INC"

1INHE

III:'\

I 0".’.'E

HEY

O r

 

 

APPENDIX II

DESCRIPTION OF COMPUTER PROGRAM KINFIT AND SUBROUTINE
EQN FOR THE EXTRAPOLATION OF NMR CHEMICAL SHIFTS TO
INFINITELY DILUTE CONCENTRATION

It was discovered that the curves described by
the chemical shift vs concentration plots could be
adequately described using a simple power series in

concentration

8 a do + AC + BCZ + 1303 +204 + F05 + so6
+ 807

dob

where bobs is the observed chemical shift, C is the
° salt concentration in molarity, A. B. D. E. F. G.

i and H are unknowns and 50 is the chemical shift at
infinite dilution, which is also unknown. The above

equations expressed in Fortran notation are as follows:

3 = u(1) + U(2)*XX(n) + U(3>’XX(1)"2
+ umfxxm"; + U(5)'xx(1)"4
+ 0(6)‘xx<1>'_"5 + u<7>fxxm“6
+ u(a>*xx<1)**7
Subroutine EQN of this calculation is listed on the

next page

169

 

 

 

2O

25

36

35

46

65

so

r.o:,:_sr~..v9u-r [hp
-. . . . .

170

SnunchxN£ [on
r""“’3'a ‘r*"'l1c"’.n' o '15"! eid‘ol fil'.xl°](ll.nnl‘].uf)vnu.ufl'l'ly .x.v’.‘1q’.’. r0,“-

l?‘l.ILRI.l'Av'u'QI”"A"°'“H-I'Y“olloU7!vn.UAllcrnoorn-rH.P.7to10oE {Duo
30

..‘::2.',:T;Ii‘.n1.t'u. ' l.’IYo"VcV|.r‘c’1rH‘ OCH'J'J .‘th‘ c.1081 04001 .| h“! 0 (”Nu

( (.HHHHN wt UY/IHCYH

C?7MON{UOINYI'QDYo0001.!!! ,

.r) ”var. UN I'l‘o 00!out/9b.911Hoo1fl'nqutld.'(vn¢100\.rn(105$.rou-q (’5‘
)s' (fut/llo‘.v':I‘/CO’/llo’l ‘JOOIQ'Hl/0’.rl\.\'(,: ‘l/O'O‘s""'JU'I")H)QO to
""'°’SF“WU ‘ibovltbow"'the)olkulutuo\.uxtvocsov.nxllt90).101(%o>
3."ODYthe!olnnttsor.vvvc50).cous1«us..xxx¢|q.

'fi 7f ',§1O“qu‘O,IfiUQOI0‘|ltl), [IYD

t’211n1f

)1Ap(:ho

JYflD£=Hl

vfiltf (JTLDE.6)

rnnuflT(/////’///O. (In!!! CURvC ExTDAngnvooo)

oil.oO-.p:q

'Ovn=:7

DEYHDN

7 COM INUC
CF‘HSN
fl CO‘JY INUE

35

\l
nnnnnnno—nnnnnun O man—n

OETUDN

n )oxx(|voucaroxxtl’00?.ncaroxxtliooaou¢sioxxcgiooa
yous.ut71oxx(1yooeou¢etoxxtltoo7
t

Z)

ZZZIOO'ZZP‘

C .1010:

M I)! "I
”not
I—Fv

.,

IUE
u

omomomom nanomm : u on “o
Z-a

I 'a: ‘3'
-I
.. .

 

.u‘m

 

APPENDIX III

DESCRIPTION OF COMPUTER PROGRAM KINFIT AND SUBROUTINE
EQN FOR THE EQUILIBRIUM CONSTANT OF PREFERENTIAL
SOLVATION IN BINARY SOLVENT MIXTURES

The equilibra in mixed solvents can be expressed

as K1 K2
M+(SA)n + SE ______—. M+(SA)n_1(SB) -—————----'

K
M+(SA)n-2(SB)2 ——__n'—' M+<SB)II + (SB)n

(8.1)
where SA and SB are the solvent A and B respectively
n is the solvation number K1 ..........Kn are the
equilibrium constants for each step. If o:p is the total'
shift in the resonance of M+ from pure A to pure B
solvents. Hence, the intrinsic shifts of the various
solvated species was assumed to be prOportional to the

amount of B which they contain and then

1
oMAn = o; awn-18 = {or Oman = 6p (B.2)
1/n .. coco-cc 1/n
Let K' = K - (K1K2K3 Kn) (B.3)
K] = nK', Kn = %K' (Bali) .

The final equation in this treatment allows calculation
of K”11 as follows:

171

 

172

1
T '—(1 + K1/n

 

35.13
XA
a = observed chemical shift relative to the resonance

of M+ in pure A

o .
- o -
5p " ‘M(solv A) ‘M+(801V B) q,

K1/n = the geometric equilibrium constant

n = solvation number l

 

XA , XB = the mole fractions of A and B respectively b:

The subroutine equation is listed on the next page.

 

173

S“ W01” 1hr. tun

LVHWUN <UOMT.IIAHLoJTAquiwIoLachlNCRqNOPIoNOVARoNOUN
li'fialcstvloHVQHLbIUQlARcthilIYP9XX9RXTvaDXlioFUPQFO9
(1”J5kafitvTOUTfLOHgJJJoYoUYfVECE9N§blQCUNSIqNJAToJUATv

JYY10C\’JSI\
COAHON/FHLUT/IMLIH
LLCJ |m0N/i-’U1I'H/l\0r’ I v JOP
Ul”uhblHN X(eo3uolvu

10V\Zuc)l’9VLCI ((992

qutI.).CUNSIb(Su91n
39MUDI(fiu)9LUPI‘Du)9
ulMiwalow'ulfF(duu)
GI 10 (21J9495019Iv

L Ly-"lhilt. ' . .
1'! "9;,sz
Jt—ptznl
NQ.Q‘I"K=/

IVJVA“="
KL'UKJ
7 CQSIINHL
R. ’r.KI‘
.-‘ LL-‘cllNHc,
R.-.l.lh30
/ CuWIINHL ‘
1rt1MLIH.Nt.—l) b0 10 35

C
)
9
Y
C
9

‘-

R1.‘UR”€
3: CC3|1NHE
xa\1)=x(A)/X(n) ‘ _ _ -
x~\a)=sH=1/s . $=LHtM. SHIFI Or A 39 RtLArIVt TO 5 OF
U(‘) = 1/(S(A)-(UI)' "
U(c) = Ktlln) 9 LUUIH. CONSI.
N=3ULVATIUH N0. ' '
A:...oxx(A)/u(2)

5N=AGU(1)
n-JlU=Sfi-XX(4)
d:.'UKI\

i Cur-I lNUC
R - .tht‘l

MCI;

”(is

eprn.NL.-n) o0 T0 29

V
r201

\‘("

RA'UNN
)' C':'.'!1|H.L
-3UKN
l; LC-‘l 1m";

H.'URN
L3IIHUE
'. _‘UKH

o
O J

I:

A)

W-ZIC.

 

 

APPENDIX IV

DESCRIPTION OF COMPUTER SUBROUTINE EQN FOR THE
CALCULATION OF FORMATION STANTS FOR ONLY ONE STEP
REACTION

When the exchange between the free and complexed
K+ ion is fast on the NMR time scale, only a pOpulation

averaged chemical shift is observed

! - ' (D.2)
F XF ‘ 6FXF * bIPXIP

In which 61? , élpf and 5c are the chemical shifts

where b

for free, ion pair and complexed K+ ions respectively
and XF, XIP and Xc are the relative mole fractions

for each species. Since

(D.3)

C)
05
H
C)

M M * CML

and T

D.
CL = CML +CL ( A)

Substitution of (D.3), (D.h) and (D.2) to (D.1)
yields

' T
bobs = [R. 6M... + (KI‘.R.CL ' 6.MX)J/[(1+ d PIXOR)

+ (KIP’R'CMT. JIP )/ (1+ KIP.R)] / GMT

(v.5)

where T T
R = cM / (Kr-RIP) 'CL (1+Kf) + (1/KIP) (D.6)

171+

 

 

 

175

In which GMT and 0LT are the total concentration

of metal and ligand respectively. Kf and KIP are the

formation constant of complex and ion-pair respectively.
In order to fit these equations, four constants

and two parameters are used; namely

Const (1) = Ion pair formation constant
Const (2) = Total cone of ion pair
Const (3) a Chemical shift of ion pair
Const (4) = Chemical shift of metal
U (1) = Chemical constant of complexed
I U (2) = Formation constant of complex

If only very small amount of the ion pair formation
is in the solution, the (C.6) equation can be simplified

to 2
T T 2 T
= - C -
bobs (KfCM K L 1) + (Kf CL

2 T
' 2xf cL

2 T3
+ Kf CM

T T T T

+ ZKfCMT + 1)‘/2 6f - 5C 4- SC

M
2KCT

Subroutine EQN of the calculation of Kf with and
without ion pairing consideration are listed on the

next two pages.

 

 

 

n (HWHFWWNWG

W

C C

C 1
1;

C C

C 1
3|

l
c currx run anxnx
;

‘a

176

QHfiRruvluf {0"
Cpuuno «nunr.[YiDL.|tAur. |wl.Lz.U.(|q(Q,unpy nfivn:.qq” ...,u c.‘ J? .
lU'XcY‘CYol.nv.°[§ln.1n0.tus.ltYP.xx.L(Ivo.nl||.rno.rn.ru.-.,..1...
PIOVAL.v<I.Y .nT.L.N.III.v.n\.v; gl.u.\l.;n-;1.unat.|;.nt.u.-t..nvv.
1V4'W

..0. “.-.' 0‘..',

"an ,l‘nl.'/V¢5[.l’\‘3f.(ul
nl'orxcvnuvtl..1.;:|.nt))l.nlvPHIu(I.xt(..\,rm‘(Ino1.rql\.w.tn-L_U
I.“!?:-9|I.vtFY123oPlioln‘Juct-IO‘/)'-ll vtt"v!-""‘ d‘t“'”"
)thlnl.CnIQYQIGOcIHI.HCKIlHOIOIS“lu( U'c”\1\”‘\0‘o”ll"Hhiolc'|“)‘
3,UH0YICQIoLfiUYIRJIoYVY(HO|oCUVSI‘IhiollIlle
nl“E\<lnN nxrft?coloC‘)Ou!
GP 10 t?01c“oqolo7oHoQoIOOIIOI7‘ IIYP
cwaixnnr

l"°E=~C
J‘roitfil

N a‘a’q|(=3
~.)Vh.’):‘
phIHSN
C’loYl‘nlf
RL'II~"
C;)'IIIHF
”CH
)4 NW
C (A..ch1y 3 {ON DAIH FnRMAIIUN CONSIAN!
cnscvcav : tntnL CONC. 0‘ "EIAI-
Cnuartl) = ' CHE". Rnlll 0‘ IOV Phi)
cqucttb) : CHCU.<H1FI ”7 "’1“;
”(\\ = CNEM. culrt of Cuuntrxtl utYnL
...-n : rmmnlnu cousin” H' CD'N’LEI
11¢.) : Intnt (ONC. nr LIHAHO
lit?! a OHS‘QVIH (”Epo $“lrI
CNN»: .12,

Eviluc 1. to MT '3C1tursg n? YhL VEHMS
DU=!!‘I\0¢l./H( otl MRI I
nO=¢I./¢Hc7)-CONST¢1H ?x& K” l'CUNSI(7"/C”""“‘
n°=-ro~s1t73/(u12b~cnu H! H
an=(1.-no- -ooooaill.

Hq:(;.n(p0001.-q .Ip ponn.77,0HQI/?7.

arr~ 0([u1 rnu HHICH roan 0F IHL SOLUTION 1% t3 u! “SEC
1Fi1=can.0)\/«.o(AA003I/?7.
1r (15.<1)11.1!.17

HF eroolac rnuw IS uscn HERE

V‘:'-""/?. -§O"l¢ICRII‘

‘An: (nuqctaggoocuu-

1u:(-uu/? -5notllCSYil

DfnztonstfflllOOCHNR
HECFlkfi TH‘ <1n~ or In: cuatuuol ILDVS

l‘ (anl7. l7 18

..M: .‘5A

1‘ (1Hi\°. 10.?5

q-lflg-uuq

Rv=ALA.uHD

p:-U-¢cD/3.I

(.v 1n 1'. .
NE 'alnnunucroxc roaru or IME SOLUtxnv It ”can «so:
r'. .‘r‘:

95.:2. ~Acn<(-l.
aavLE->An/1
C‘c:l-'IR/7,!/ fianihufifl-Ah0'3‘/?7, |‘)
tv 0' (”slut 1£¥1
‘.09.1C‘L.Gl.l.‘| no to «0
L‘

s

(Obi/1.)IIOCOSCFE/3.0AHELEOFL53|

lr-tfrr LY,-

r.:DAn-a(ޤ(
R~’- 7. ‘ ocuorc
“9"”:‘0‘DI‘ ‘

I
C
A“

C CNFC* FHU ViCl"l'Y OF YHE CALCULettD MEIAL cnuc

an
??

[iIU\|I.|t.1

HUIUEI ITQO'027, .

r' I'HAII. CHQINE N’ 70, H'IUSIHE n} 0] A'JH 0"!
gt“:

C an N -c~ nun rxun Anoyuro gnu!

l1
I5

r. in:r' “(3" .

l"1.-rl.AG!|Ho:to|“

UJIT‘CIYADEQ;I

7“"H5'l . QLI INULE 9001‘ FLUVKED.‘

Q! .
C snuk'llnvfi CALCULAILD "t'AL CONC. IN ORIGINAL Fluatlnq

qulNE ran

1?

1s

"”‘ ”“'=' rtn a.h°«11

U'L1A=auornuk1luiofn!?l000xx (ll .H(|\!/ ( ou(;|¢uioqrfiq-‘(|g¢ocj ~g

IH’H‘C'WQTllH/H.0('n'H-'.IHI' NH Igamslc )
u-~ln:\;Ltn- 11!?! ,

I‘llv5'n4.ur.- -II (:0 m 1:.

n-lncn
Ctotlfi;uf
"HSu

‘glvite.ur.—ti no In ’0

41300.“?yfuu?n
I4 -.
% R

.19
2v
25
30

3%

hfi

suunouIINE EO~

V

39

177

73/73 OPI=I FIN “.6‘“33 01/05/78

SUHROHIIHE EnN

CUHHnV KOUNIottfipEQJ'JOEQIHIOL‘pOXIVCRoNOpYQNQVARQNOHNKoIQUQII‘AXO EON“
‘HIAOIEQTQ[0&VORESIDOIARQEPSQ[IIPOXXORXIYPQDXI[croporn'FUQDQZlOIUOE EON“
g1¢¥AE512¥£I.OI.LoflvJJvacovchLIoNCSIoCONSt.NDAIanAToNuptcLo’Io EON“
Y o l

CJHHnN/FREUI/INEIH

CU‘HGVIDOIHT/KOOIoJODIoXXK

nldtksInn XIhoJJOIoU()0IoHIlIBoJOOIoXXIbIoFOPIJOOIoFOIJUOIoFUI300I E0
[,0I9J.DII9VFCII?JQIII92L(JOJIOIOIZOIQEIGVALI7OIcXSIIIUuIoYIIOI‘
ZDYIIGI.COIKISISqolfiIQNCQIISvIvIS“IN(§oI.C(IYD(SOIuDXllIhUIle&(§0I
3.“V9IIécloLnPIISCIoYYYISUIoCOVbIIIhIoX(XII§I

0|“iNSInW DIF'IPLGIoC(?UOI

G IO I7OJQBQQQIO708990IJOIIOIK’ [1Y9

C‘LIIENHE

[IAD :5;

JI~95=sI

d.|!fi IJIADE.E)

F”F‘AII///////o0, 'UQMAIION CONSIAHI OI

OJ. Ion-(z)

NIVAD=$

9 luau

CIHIINHE

aquou

C . -'!."'r

R quq

C '-Il*nI£

I'EI"€YH.“Eo-II 00 70 35

R 0H9]

CCHIIVHE

A‘IHIIIO‘PI'IXXIII002I

H=IHI3I°°PI0IC0uSIIII002I

C’-’.I0IHI?I00;I0IxxI|IIOCONSIIII

? f.;’ln()II.I(lIIII

".J0IH(?II~IC1HSIIIII

AdzthI’IP'ICWISIIIII

h”‘¢04I)II’XlIII

F ’Irfi '{IPI-wtIII/I9.0¢Cnn$ttIIIOIJI?III
fifcIIaaoflu-I.I‘\HdIIANuILOH°L°U'L°I.III°CCIOHIII

H-ol‘=:-(lI/I

fl I”).

C III' If

n {Ion

L -'!.g Ir.

D-IHDM

9 II'I?

I'i3"' ~..'.-II a) II ?3

I. '13.:

~' ..Ilon"

I _I!

I II'I'

H In-

I. I .' "l co

n I '

0
Q
o

 

 

APPENDIX V

DESCRIPTION OF COMPUTER PROGRAM KINFIT AND SUBROUTINE

EQN FOR THE CALCULATION OF FORMATION CONSTANTS
FOR TWO-STEP REACTION

The equilibria for this two step reaction can be

 

m
expresses as ‘-
K ..
K“ + pF6' ——.IP K+PF6
+ + .
15 c 5 15 c 5 J
. I;
K1 1] , '1 K1
+ - KIP - +
K «1505 + PF6 = PF6 . K o15cs
+ +
15 c 5 15 c 5
I
K2 “ K v 11 K2
+ IP + _
K (15cm = (K 15c5)?_pp6

As shown in Figure 379 the 1:1 complex of 15 C 5 with
K+ gives no evidence for contact ion pair formation.
That is, KIP'-O. Also, as can be seen from Table 24,,
anions show no interaction with 2:1 complexes, it _
suggested KIP"—~O too. Therefore the above scheme can be
simplified to KIP

K” + PF6' =. K+PFg

+ K1 +
K + 1505 = (K 1565)

K
(K+15C5) + 1505 'é K+(ISCS)2

178

179

The observed chemical shift can be expressed as

+ 6 X (E'I)

6 - +
' 6FXF 6IPXIP ‘5 c1xc, ca c2

-obs
Where F, IP, 01, 02 denote uncomplexed K+, ion-paired
+
K+, (K 1505) and K+(15C5)2 respectively.
The equilibria for two step reaction can be

rewritted in general form:

 

 

M+ + L=ML, K1 = CML/ CMCL (3.2)
+ ——3
ML + L,_.ML2, K2 = cMLZ / cMLcL (3.3)
C
w + X-g-t MX, KIP = MX (3.1+)
CM+CX-

The concentration balance leads to

T _ E.

T - 5.6
cL - cL + cML + acMLZ ( )

T _ T _

substition of (3.2) (E.7) to (3.5) and (E.6) to (3.3)
yields

C T

2
M CM(1+KICL+K1KZCL IKIPCX) (E.8)

and T 2 (E )
CL = CL + KICMCL + ZKIKZCMCL .9

substition of (D.8) to (D.9) we obtain

V’

 

4K1K2°M
(3.10)

and also since

180

x01 a KICMCL (E.11)
x62 = KIKZCMCLZ/ GMT (E.12)
XF g CM / GMT (E.13)
xIP = KIPCMCX‘ / GMT 03.14)

In most solvents, KPF6 show very weak ionic

association, in addition, the results from 130 NMR

study (shown in Figure 36) indicate that the formation

constant for 1:1 complex always large, therefore

ion pair formation can not compete with the complexation

formation. Therefore in most solvents, the ion pairing

formation can be neglect (XIP21 O). In this case, in

order to fit these above equations, four parameters

and two constants

Const (1)

Const (2)

U (1)

U (2)

U (3)

U (4)

subroutine EQN of

are used

total concentration of metal
chemical shift of free metal
formation of 1:] complex

Chemical shift of 1:1 complex
Formation constant of 2:1 complex

chemical shift of 2:1 complex

the calculation for with and without

ion pairing are listed on the next two pages.

 

 

1 £31

:OOOOC£::O:2:.:O::2.::§:.:::€102:000000......OOOOOOOOOOOOOOO 0.0000000...

 

 

0
IA}! ' an
uAL1L°8111Iluiltoo
5'
0
11.0 "In C
on o '0 O‘ 3 . ' 00" V" UM} I H "IOIl O“
g°lo§'$=?§?i ?:€:€5:l ::$i:;ol ';:U::SI;"0°:?Y0'° ‘1'. i9‘3. .b.' 0101‘ I 0
I1V‘L.IS'0'OK'OLo'0’JJ0'0070V C1.~C\11C0Is11~0‘lw '0 O
)v'v.Cn~::$ ' It! fg::
‘833831-n$3153n117.»73..11...‘ o 1'111e1 r )00 f 13001 1)001 i9::
0 o o o o
g-"mm-‘ézilmh1:21.139:159m:1625132122532‘51112:fizz... .2:
93n3911 9?‘.§132353:§u;s::.2.451.l.....;.... on.
1 1
gongus:9?3.e.!.1:1.0.‘.t1.11.121 l'!’
‘30:; 1411‘
I
I 00 02'
gVUQH
1 C 11111.11!
hfimlofi
N VIC.)
Ynfiu
0 gguilnul
fiCCCCCi?(($Cf$SC SSCCCCC'CccscgSCCPCCCC'E;2PCC$CCSCS$CSSECCCCCCCCCCCCCk
r 121 .
1:°~:;§§%$.u:'§ g t" 1n~. 21Cfl~c. or a~10~ E
“CCCCC (CECC C$C8CCO CECCCCCCz'CCCCECCCCCCCSCCCCCCCCCCCCCCCCCCCCCCCCCCCCU
7:81: l:tn:3.0 ?' (Onl'c '1-1 .E3::::::;Q.OCOM$IIQDOCNN‘VIIIOIII
.171'l1t1.(HoCO~S'1|" CO~‘01-|.°:37:ll.’0.°01|'°C0~"1|’DII
r1111 1. n 1.0~c.u111 .1. 1 co '0 810
111-11.1 3 :Eow 11:11 C0~'-|.C- ‘ "
1"111-
. O
S:2§": 1 a 11
3 0
391.3,..2111 ’Olns 101.-
101 “on-11:11::Ou .1 1 o
I
S:£é‘111?.1::0;:;:;:!0n131:llll11
I o .0 O
8:7.0CIN1;H::1BUI|13U;1OC1~100)1Cnus11o100
I 1 I
£923“:9?I1111.OCOQ371010Coqn1
1'111111.(0.1.01 o 10 )5
3" 3~1.to.1.!1 C1~|'|.C'?0
1 oClN-I:I 13" .
31$c1:}7:730*'1,1 126:1 1111 .
loo. 00-111111.- 111-1'1 1s 01v£0htu 1
"‘C|"II.ok'oo.....' ‘0 '0 , .
$°' 9‘12232.
20 $07.0utl10CIIDOUII..U"|.CCNO..,
9~.1(o~s1111lo .
t” {it 3:;$?:?1é3:...,co~.11 1
(on! oul 10Cnhi’C1N1ICfl~§’CO
w'1"."’:'¢°1~:2.1~2:'..:22':2'.2: ' mm...
C 1 10 0H1 1 v o ’0
3973:§?:‘~c.-11 on 10 11 Inn.
111 11<n s
II111-C0~d
.0101...
. 28:17:.
, .1‘100‘0“..II" CNN.
\ :A'VTqu 'P-
01:v9~ Duo
83:“. l~U(
Iv uq
‘ ir:; -? s.~¢.-11 on 1o ,1
I,
’. S ?'|2" I1~o
I“ .
0 ¢§~1lzu¢ 00k.
O 1qu
I0 C ~II~U¢
I 1 Cu
" $92,121:
1! ~ ~1
10~e
Egon.“ 10~o
7
0

C h
1::ifl2: "‘Aflo
~. I‘L .§:"."‘ 0 0 0 0000c0.0000000.0000000000000...coo...
éczz:03;:300000000000000000 00 00 0
'0
O

 

000001011

1'51

16?
:0;

1t?
"01

‘9

1

‘- " 'Y'.o Y'J’J
C‘ “‘2" " .6701'~“"{0."-""0141?.b‘nil' -3... ’70 I'. 9.1‘ 'VI‘O"0:Tv-.‘O
l-T'oZL‘T-2.-‘-'.'.~‘-‘\;.'.|"Jo1 “".|_"‘1|l~‘.lYV ¢"',1" ' .7“— "n'lL‘.10t
(‘:".L .1“. .'c.)71.,o‘111JvYoUY V: L71‘.CSI9C ‘11"-TO L~I'“QDI‘LC)I'
‘rY‘1tz'll'xTQ
C," > './‘..v' ,/i"-:'ID‘
CI °'./‘\)I .l/'( 'o.’-r"’TcI’) -
LK“IH\IO' :1~.‘J31.nvPC1..II<~o‘J;:.Iv'ci.rfd'1t7)-FC‘I“D)~F““00
It"‘?:'?l‘°V1' 119v. )1).," 1 KU',)Q‘..>",':).1,L.. .l 1,)Jl.k'\1“ 'k'o{‘1030_
2u"lli.ffl .R.‘"c-ZHI.NLKTKECloiH“I. “ !.-IT' (Hoiovkl1&SC)olCK¢50
3.111 9111.01”.;J'15:).‘(V1'1HJ).C1,;‘.)I(U131)£A1lS)
X'TLF'~ 11-6l
RLLL P71")
1.”): "‘1'“!
(1L If 17131‘oog11171q101IUO‘IOA’) XIYP
Ci-TIVIE
r.-..‘T‘.:)
L2mu-=a
III-13:“?
1,..1—(‘
:t‘lllz'o
(A'I'c::.'.'l€
~111..2.,
Cl-c'wwf
ML '10-'4
c-"":l'1'£
Ci.1:"'1‘
C'Y=Cnnsrllt
U‘L.=Fo591121
ri=H1|1
01-1:V191
-?=H111
D'.L?=U""
C17: 1n1A; CCuCEnTSA'XCN C7 LlfiiNO (I? (JO-V 5'
CHI: TGTLL C-".C4_7.7’-.‘.’lf:.‘1 Jr “:7“. (SIDE-"'1
F9 = ‘03”ATIHN CUNRTflhT Of THL P TO I L?:~ - 1A CDV9LLX
D‘LG = INITIAL C~f”;rnL SHIFY (716E HE'&.1
nzL> = szxtzuc Cr=v2caL s~1rt or Ins 2 t: 1 C ~=.:v
Urt| : L1u171u6 CwCHICAL SHIP! 0F TnE l I: I CC“DLLK
Uzi = oWRE:VfF C“E“IC&L fiwlfl
gzl((‘OfLT-q0,'U',)O1(}Oy;)oCMT°,(190?)-CL70(.“OO? )/f..er)ov?- '~"’Z’
c:llae~n10K/°CM1°1CU7-CLT)OKX'CLT°(<2'CLI°I))/‘ "1'K2' Kl"2’
e=-r“7/1a°/1022-K1°°?1
£=i1°L-§‘S1/3
L:.?cre01— Cog-Co“9J1/P7
UC7=1N0'?1/~01¢‘°J1/97
1“»111 1:32.70ul.?'01
cw5r121-4/p1/1s1911-1;1131/2711 '
r1=nro<1C09111
01c:ACnC1-].1
TfiL7L=F]/3
C“:?°§T¥Yl-A/BD'COSITHETAD-S/3
fi' 70 1;)!
1fi5T12-M/70gfioTanTl
l‘ "fc7l’ 10112311201
111712-7L911
h‘Lle‘ °
bi‘l't"l"’ll./3
I‘ 1‘<‘7l.NL.II 60 70 30]
L5=-LA
N~LTK=
CIGT!%HC
Y:bl’=-"/?- fiQHTlD’71
1"T?§T?1 IO?02079?0?
Tl‘T?:-]E§
1.8L17=1
HH=11E§T?‘O'1I./3.l
1F 1n9r1?.~£.11 60 T0 30?
HH=-l"H
N1L1?=‘
(0".IIDI.E
c"=anoan-S/3.
x =CUIruT
C =«(LY-7°Cu102-CM1/1I-xlocn1
C“L='|'C“'FL
x1=CUL Y
0£L=x¢~1UCLo DEL?)ol|'(DELl-DLL?)°DELZ
l ‘1“VY“.VE.-|! GO TO 15
1'1’1-HFL .
14.111143
CO' I INHE
#111123.
CJLIIHHE
HLDHF-.‘
C)HTI‘.HE
I): Ufi-N
(d-ayxh11r
1"!”il”. NC.-|) 60 10 ?0
arlu~1.
(UNTIHHE
p-IHSu
CU‘TI'Q'IE
HP‘Hhu
CJHIl'HE
-Iuog
Cu 01]‘.I!E
#111,2',
(,1.[,11[

n1 11.3'1

f u

 

REFERENCES

1O

11

12

13

J. N. Shoolery and B. J. Alder, J. Chem. Phys.,
£11 805 (1955)

. P. Fabrikand, S. S. Goldberg, R. Leifer and
. G. Agen., Mol. Phys., 2, 425 (1963)
Fratiello, R. E. Lee, D. P. Miller and

. M. Nishida, Mol. Phys., l9. 551 (1966)

ou<13>mou

. W. Maxey and A. I. P0pov, J. Amer. Chem. Soc.,
29. 4470 (1968)

E. Schasched and M. C. Day, J. Amer. Chem. Soc.,
12. 503 (1968)

R. E. Connick and R. E. Poulson, J. Phys. Chem.,
6;, 1002 (1958)

w. DeWitte and A. I. Popov, J. Solution Chem.,

5, 231. (1976)

A. Carrington, F. Dravnick and M. C. R. Symons,
Mol. Phys.,lfi, 174 (1960)

J. P. K. Tong and C. H. Langford, Can. J. Chem.,
fig. 1721 (197#)

L. S. Frankel, C. H. Langford and T. R. Stengle,
J. Phys. Chem.,,zg,‘1376 (1976)

A. K. Covington, T. H. Lieley and G. A. Porthouse,
Trans. Faraday Soc. 1., 62, 963 (1973)

T. R. Stengle and C. H. Langford, J. Phys. Chem.,
62. 3299 (1965)

C. Deverell and R. E. Richards, Mol. Phys., 16,
421 (1969)

183

14.

15

16

17

18

19

20

21

22

23

24

25

26

18h

J. S. Stengle, Y. E. Pan and C. H. Langford, J.

Amer. Chem. Soc.,lgg, 9037 (1972)

A. I. POpov and R. G. Baum, J. Solution Chem., A.

#41 (1975)

H. A. Berman and T. R. Stengle, J. Phys. Chem.,

22, 2551 (1975)

J. Sadley and A. J. Sadeley, J. Mag. Res.,'lg

(97), 103 (1974)

A. I. POpov, in "Solute-Solvent Interactions"

M. Dekker, New York, N.Y. (1976), Vol. II. pp 271-330
R. E. Maciel, J. K. Hancock, L. E. Lafferty,

P. A. Muller and W. K. Musker, Inorg. Chem., 5,

554 (1966)

E. M. Kosower, J. Amer. Chem. Soc.,‘QQ, 3253 (1958)
R. H. Erlich, E. Roach and A. I. POpov, J. Amer.
Chem. Soc., 2;, 4989 (1970)

Y. M. Cahen, P. R. Handy, E. T. Roach and A. I. POpov,
J. Phys. Chem.,‘ZQ. 80 (1975)

A. I. Mishystin and Y. M. Kessler, J. Solution Chem.,
5. 799 (1975)

V. Gutmann "Coordination Chemistry in Nonaqueous
Solvents" Springer Vienna, (1968)

H. G. Hertz and H. Weingartner, J} Solution Chem.,

Eh 790 (1975)
C. A. Melendres and H. G. Hertz, J. Chem. Phys.,

519 #156 (197#)

27

28

29

3O

31
32

33

311

35

36

37

38

39

185

J. P. Oliver, P. A. Scherr and A. J. Weibel,

J. Organometal Chem., 85, 265 (1975)

K. Covington, R. Lamtzkea and J. M. Thain, J. Chem.
Soc. Faraday, 2Q, 1869 (1974)

M. S. Greenberg, Ph.D thesis,Michigan State University
(1974)

Y. M. Kesseler, A. I. Mishnstim and A. I. Podkovyrin,
J. Solution Chem., 6, 111 (1977)

G. Olofsson, Acta. Chem. Scan.,.2§, 1352 (1968)

E. G. Bloor and G. R. Kidd, Can. J. Chem.,fié,

5A25 (1968)

W. J. DeWitte, L. Liu , E. Mei, J. C. Dye and

A. I. P0pov, J. Solution Chem., 6, 337 (1977)

J. D. Halleday, R. E. Richards and R. R. Sharp,
Proc. Roy. Soc. London, All}, 1+5 (1969)

M. S. Greenberg, R. L. Bondner and A. I. Popov,

J. Phys. Chem.,zz, 2hh9 (1973)

C. Deverell and R. E. Richards, M01. PhyS..lQ,

551 (1966)

J. W. Crawford and R. P. Gasser, Trans. Faraday
Soc.,fi}, 2758 (1967)

P. Neggla, M. Holtz and H. G. Hertz, J. Chim. Phys.,
2.1! 56 (19711)

G. W. Stockton and J. S. Martin, Can. J. Chem.,

5;. 7## (1974)

no
41

A2

43

1+5

116

£17
118

1+9

50

51
52

53

51+

186

T. L. Collins, Phys. Rev.,,lQfi, 80 (1950)

H. Sommer, H. A. Thomas and J. A. Hipple, Phys.
Rev.,.§§, 697 (1951)

E. Brun, J. 0eser, H. H. Stans and C. G. Telschow,
Phys. Rev., 23, 172 (1953)

J. C. Hindman, J. Chem. Phys.,'3§, 1000 (1962)

E. G. Bloor and R. G. Kidd, Can. J. Chem., 59,
3926 (1972)

W. Sahm and A. Schweenk, Z. Naturforsch,‘;2A,
1754 (197A)

M. Shporer and Z. Luz, J. Amer. Chem. Soc., 97.
665 (1975)

F. W. COpe and R. Damadian, Nature, .2_2_§, 76 (1970)
M. Goldsmith and R. Damadian, Physiol. Chem. &
Physics,,2, 263 (1975)
R. Damadian and F. W. COpe, Physiol Chem. &
Physics,Ԥ. 511 (1973)

F. W. Cape and R. Damadian, Physiol Chem. &
Physics,Ԥ, 17 (197#)
N. F. Ramsey, Phys. Rev.,'Z§, 699 (1950)

J. Kondo and J. Yamashita, J. Phys. Chem. Solids,
.12: 245 (1959)

D. Ikenberry and T. P. Das, Chem. Phys.,‘gfi, 2199
(1965)

C. Moore and B. C. Pressman, Biochem. BiOphys.

Res. Commun.,,l§, 562 (196A)

55
56
57

58

59
60

61

62
63

64

65

66
67

68
69
70

187

C. J. Pedersen, J. Amer. Chem. Soc.,.fig, 7071 (1967)
C. J. Pedersen, J. Amer. Chem. Soc.,‘gg, 391 (1970)
S. Ciani, G. Eisenman and G. Szabo, J. Membrane
Biol. ,1_ 294 (1969)

D. C. Tostenson, Fred. Proc. Rad. Amer. Soc. Exp.
Bio., 21, 1269 (1968)
H. Lardy, Ibid.,|§2, 1278 (1968)

F. L. Cook, T. C. Caruso and C. L. Liotta,
Textrahedron. Lett., 4029 (1974)
K. H. Wong, G. Knoizer and J. Smid, J. Amer. Chem.
Soc.,,flg, 666 (1970)
M. J. Maskornick, Tetrahedron Lett., 1797 (1972)
S. G. Smith and M. P. Hanson, J. Org. Chem.,
.16. 1931 (1971) *
D. J. Samand and H. E. Simmons, J. Amer. Chem.
Soc.,‘2a, 4024 (1972)
R. M. Izatt, D. T. Eatough and J. J. Christensen,
Structure and Bonding,‘l§, 161 (1973)
J. M. Lehn, Ibid, _16, 1 (1975)

J. J. Christensen, D. J. Eatough and R. M. Izatt,
Chem. Rev.,'24, 351 (1974)

J. S. Bradshaw and J. Y. Hui, Review, 649 (1974)

G. W. Gokel and H. D. Durst, Synthesis, 168 (1976)
A. I. Papov and J. M. Lehn, Chapter in Chemistry

of Macrocyclic Compounds, Gordon A. Melson, Editor

Plenum Publishing, N. Y, in press.

71

72

73

74

75

76

77

78

79

8O

81

188

J. D. Lamb, R. M. Izatt, J. J. Christensen and

D. J. Eatough, "Thermodynamics and Kinetics of
Cation-Macrocycle Interaction" Chapter in Chemistry
of Macrocyclic compounds, Gordon A. Melson, Editor,
Plenum Publishing, N. Y. in press.
H. F. Frensdorff, J. Amer. Chem. Soc., 23, 600
(1971)
R. M. Izatt, R. E. Terry, B. L. Haymore, L. D. Hansen,
N. K. Dalley, A. G. Avondet and J. J. Christensen,
J. Amer. Chem. Soc., 28, 7620 (1976)
J. R. Lotz, B. P. Block and W. C. Fermelius, J.
Phys. Chem.,‘éfi. 541 (1959)
D. T. Cram, R. C. Helgeson, L. R. Sousa and K. Kaplan,
Pure Appl. Chem., 43. 327 (1975)
S. C. Shah, K. K0polow and J. Smid, J. Polym.
Sci., 33. 2023 (1976)

E. Mei, A. I. Papov and J. L. Dye, J. Phys. Chem.,
Q1, 1677 (1977)
N. Matsura and A. Sasaki, Bull. Chem. Soc., Jap.
9.9 1246 (1976)

D. F. Evans and E. L. Cusseler, J. Solution Chem.,
.1. 499 (1972)

E. M. Arnett and T. C. Moriavity, J. Amer. Chem.
Soc., 23, 4908 (1971)
M. R. Truter, Structure and Bonding,'l§, 71 (1973)

82

83

84

85

86

87

88

89

9O

91

92

93

94

189

F. P. Boer, M. A. Neuman, F. P. Van-Remortere
and E. C. Steiner, Inorg. Chem.,'13, 2826 (1974)
F. P. Vanremoortere and F. P Boer, Inorg. Chem.,
.129 2071 (1971+)

C. Riche and C. P.Billy, J. C. S. Chem. Comm.,
185 (1977)

M. A. Busch and M. R. Truter, Chem. Commun.,
1349 (1970)

B. Dietrich, J. M. Lehn and J. P. Sanvage,
Tetrahedron Lett., 2885 (1969)

D. Moras, B. Metz and R. Weiss, Acta Cryst.,
‘Eéfi. 388 (1973)

B. Metz, D. Moras and R. Weiss, Chem. Commun.,
440 (1971)

E. Mei, A. I. P0pov and J. L. Dye, J. Amer. Chem.
Soc.,‘22 6532 (1977)

P. B. Dietrich, J. M. Lehn and J. P. Saurage,
Tetrahedron Lett.,.fig, 2889 (1969)

P. B. Dietrich, J. M. Lehn and J. P. Saurage,
Chem. Commun., 1055 (1970)

Y. M. Cahen, J. L. Dye and A. I. Papov, J. Phys.
Chem.,,ZS 1289 (1975)

Y. M. Cahen, J. L. Dye and A. I. Popov, J. Phys.
Chem., 22, 1292 (1975)

J. L. Dye, C. W. Andrews and S. E. Mathews, J.
Phys.,‘22. 3065 (1975)

95
96

97

98

99

100

101

102

103

104
105

106
107

108

190

J. M. Lehn, Pure and appl. Chem., 42, 857 (1977)
J. M. Lehn and J. Simon, Hel. Chim. Acta, 69,

141 (1977)

D. H. Live and S. I. Chem., Anal. Chem.,‘gg, 791
(1970)

G. J. Templeman and A. L. VanGeet, J. Amer.

Chem. Soc.,.24. 5578 (1972)

G. Foex, C. J. Gorter and L. J. Smiths, "Constants
Selectionnees, Diamagnetisme et. Paramagnetism,
Relaxation Paramagnetique" Masson & Cie Editeurs
Paris (1957)

C. W. Gokel, D. J. Cram, C. L. Liotta, H. P. Harris
and F. L. Cook, J. Org. Chem., 32, 2445 (1974)

V. A. Nicely and J. L. Dye, J. Chem. Educ.,.4§,
443 (1971)

R. A. Robinson, J. M. Stokes and R. H. Stokes

J. Phys. Chem., 65, 542 (1961)

L. S. Frankel, T. R. Stengle and C. H. Langford
Chem. Commun.,.323 (1965)

R. G. pearson, J. Amer. Chem. Soc., 85, 3533 (1963)
R. M. Izatt, R. E. Terry and J. J. Christensen

J. Amer. Chem. Soc.‘2§ 7626 (1976)

A. Smetana, Private communication

D. H. Haynes, B. C. Presman and A. K. Walsky,
Biochem., 19_ 852 (1971)

R. Weiss private communication

 

191

109 w. Lok, Ph.D Thesis Michigan State University (1975)
110 J. M. Lehn, private communication
111 J. Kielland, J. Amer. Chem. Soc. 52, 1675 (1937)