a 2..
23w :0 .
‘ 9. )hk -

:41...
k

t
t 2."
a . ‘ H... .
firwuww-MWN‘ “WWW.“
A «9‘ v

» any
.
SIP
. 1141“!)
v Dal f

v1.7.3”...
.. iii...

"'5” ,
Qurdtgu.
.1: I3 .

7

J t... (a. i: .96. F}: o. E. Ava-(I. n. ,
. v .r 5.1.113: . Ll... I .

 

in . K a}. up: . A $13.. 55-, .r: .-

 

(New)

                                                                                                                           

93 01555 0449

This is to certify that the
dissertation entitled

TOWARD AN "INTERRUPTED 6-BOND":
ION BINDING STUDIES OF DIA- AND PARAMAGNETIC

TRIARYLPROPELLER IONOPHORES
presented by

Scott J. Stoudt
has been accepted towards fulfillment

of the requirements for

Ph . D. degree in Chemistry

Q.~~ KQVL.

Major pro ssor

 

Date I2 Awe]. qug‘

MS U is an Affirmatiw Action/Equal Opportunity Institution 0—12771

 

 

 

 

LIBRARY

Michigan State
University

 

 

 

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

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU le An Affirmative ActionlEquel Opportunlty lnetttuion
Warns-9.1

 

 

TOWARD AN “INTERRUPTED O-BOND”:
ION BINDING STUDIES OF DIA- AND PARAMAGNETIC
TRIARYLPROPELLER ION OPHORES

By

Scott J. Stoudt

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Chemistry

1995

 

ABSTRACT

TOWARD AN “INTERRUPTED O-BOND”:
ION BINDING STUDIES OF DIA- AND PARAMAGNETIC
TRIARYLPROPELLER IONOPHORES

By

Scott J. Stoudt

The ion binding properties of tris(2-methoxyphenyl)-Z (Z = C', N)
propellers and a related triarylborane (Z = B) species were probed to gain an
understanding of the relationship between structure and magnetic coupling
within pairs of these subunits. To promote pairing of the ligands about a
metal ion, two triarylamine moieties were tethered to make a covalently-
linked diamine complexant. Studies of the free diamagnetic ligands and their
metal complexes, using NMR, X—ray crystallography, ESR, and
computational methods, provided detailed information on stoichiometries,
energetics, and geometrical nature of metal ion binding by these systems.
Although the studies validated the ion binding strategy for self-assembly of
these triarylpropeller ionophores, ESR investigations failed to reveal
evidence of ion binding by tris(2-methoxyphenyl)methyl radical or a

biradical analogue.

To my parents

iii

TABLE OF CONTENTS

LIST OF TABLES ..................................................................................................... vi

LIST OF FIGURES ................................................................................................... viii
CHAPTER 1. INTRODUCTION ......................................................................... 1
1.1. Self-assembly ......................................................................................... 3

1.2. Electron Spin Coupling: The Pairwise Interaction ,,,,,,,,,,,,,,,,,,,,,,, 12

1.3. The “Tripod Ether” Approach to Electron Coupling ,,,,,,,,,,,,,,,,,, 23

CHAPTER 2. TRIARYL—Z PROPELLERS (Z = B, N) ................................. 27

2.1. Background. ........................................................................................... 28

2.2. Ion Binding Studies of Triarylpropeller Ionophores ,,,,,,,,,,,,,,,,,,, 31

2.3. Crystallographic Studies of Tripod Binding Sites ,,,,,,,,,,,,,,,,,,,,,,,, 47

CHAPTER 3. ION-BEARING PROPELLERS ................................................ 56

3.1. Background ........................................................................................... 57

3.2. Ion Binding Studies .............................................................................. 59

3.3. X-ray Studies of Metal Complexes ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 73

CHAPTER 4. TOWARD AN “INTERRUPTED o—BOND” ........................ 83

4.1. Syntheses of Paramagnetic Tripod Ether Ionophores ,,,,,,,,,,,,,,,,, 84

4.2. ESR Studies of Ion Binding .............................................................. 91

CHAPTER 5. EXPERIMENTAL SECTION ................................................... 99
APPENDIX ................................................................................................................. 130
REFERENCES ........................................................................................................... 147

iv

LIST OF TABLES

Table

2.1. Formation Constants and Chemical Shift Data for
Alkali Metal Complexes of 28 in CD3N02 Obtained
from Tandem 1I-I/7Li (23Na) NMR Titrations ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 42

2.2. Crystallographic Data for Compounds 39, 38, and 26-13 ,,,,,,,,,,,,,,,,,,, 49

2.3. Ring Twists for Hexamethoxytriphenyl-Z Propellers
and Related Species ..................................................................................... 52

3.1. NMR Data for 28, 45, and Related Li"', Na+, and K+
Complexes in CDC13 at 20 °C ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 61

3.7. Crystallographic data for 282'NaBPh4, 45-NaBPh4,
45-NaB(4-C1Ph)4, 45’KB(4'C1Ph)4'CH3N02, and 45 .......................... 75

3.8. Selected Distances (A), Angles (deg), and Torsion

Angles (deg) in 282-NaBPh4, 45°NaBPh4,
45-NaB(4-C1Ph)4, 4S-KB (4-C1Ph)4-CH3N02, and 45 .......................... 77

2.4A. Atomic Positional and Isotropic Thermal Parameters
for 39 ................................................................................................................ 130

2.5A. Atomic Positional and Isotropic Thermal Parameters
for 38 ................................................................................................................ 13 1

2.6A. Atomic Positional and Isotropic Thermal Parameters
for 26.13 .......................................................................................................... 132

3.2A. Atomic Positional and Isotropic Thermal Parameters
for 282'NaBPh4 ............................................................................................ 133

V

3.3A.

3.4A.

3.5A.

3.6A.

Atomic Positional and Isotropic Thermal Parameters
for 45-NaBPh4, ............................................................................................

Atomic Positional and Isotropic Thermal Parameters
for 45'NaB(4-C1Ph)4, .................................................................................

Atomic Positional and Isotropic Thermal Parameters
for 45'KB(4-C1Ph)4-CH3N02 .................................................................

Atomic Positional and Isotropic Thermal Parameters
for 45 ..............................................................................................................

vi

136

139

142

145

LIST OF FIGURES

Figure
l. 1. Self-assembly of tobacco mosaic virus. ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,

1.2. A schematic drawing of the three-dimensional
dimondoid network of hydrogen bonds formed by
a rigid tetrapyridone host. ...........................................................................

1.3. Illustration of various cation/ligand arrangements

in crystalline crown ether complexes:
(a) l8-crown—6-KSCN; (b) [benzo-lS-crown-SD'KI;
and (C) dibcnzo-24-crown-8° 2KSCN. ....................................................

1.4. Self-assembly of a double-stranded helicate. ........................................
1.5. Orthogonal p—orbitals illustrating Sab and Kab regions. ,,,,,,,,,,,,,,,,,,,,,

1.6. X-ray structure of 22. ...................................................................................

1.7. Possible orbital interactions between a nitroxide radical
and a transition metal ion: (a) 1t*—dxp__y2; (b) 1t*—dxz;

and (C) TB*-dzz- ...............................................................................................
1.8. Optimized structure of dimer 24 as calculated by Molecular

Mechanics. .....................................................................................................
2.1 X..ray structure of 23 ...................................................................................

2.2. Space-filling views of “tripod” binding sites: calculated
(MMP2) structure of 23 (left) and X-ray structure of
26-BF4 (right, BF4‘ counterion not shown). ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,

vii

10

13

21

22

29

29

2.3.

2.4.

2.5.

2.6.

2.7.

2.8.

3.1.

3.2

3.3

3.4.

3.5.

3.6.

3.7.

X-ray structure of 28. ................................................................................ 30

300 MHz 1H NMR spectra of 28 in CDC13: (a) with

no added salt; (b) with excess Mg(ClO4)2 at t = O;

(c) at t = 22 h; and (d) after filtration of the sample to

remove excess solid. .................................................................................... 33

300 MHz 1H N MR spectra of 28 in CDCl3: (a) with no
added salt; (b) with excess LiI; and (c) after treatment
with D20. ....................................................................................................... 37

300 MHz 1H NMR spectra of (a) 38 and (b) 38-Li1
in CDCI3. ........................................................................................................ 46

X-ray structures of the homologous series
tfis(2,6-dimethoxyphenyl)-Z (Z = B, C', N), showing

the near D3 symmetric structures of the borane 27 (left)

and the amine 39 (right), for comparison with the

unsymmetrical structure of radical 23 (center). ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 50

X-ray structures of (a) amine 39, (b) cation 26-13
(13’ counterion not shown) and (c) borane 38. ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 51

MNDO—calculated structures of Li+ complexes
presented with salient N MR data obtained on CDCl3

solutions: (a) 28-LiI; (b) 28-LiBPh4; and (c) 45'Li+. ,,,,,,,,,,,,,,,,,,,,, 62
300 MHz 1H NMR spectrum of 28-LiBPh4 in CDC13. ..................... 64
300 MHz 1H NMR spectrum of 45'LiBPh4 in CDC13. ..................... 65
300 MHz 1H NMR spectrum of 282-NaBPh4 in CDC13. .................. 66

300 MHz 1H NMR spectrum of (a) 45 and

300 MHz 1H NMR spectrum of (a) 45°NaBPh4 and
(b) 45'NaB(4-C1Ph)4 in CDC13. .............................................................. 69

300 MHz 1H NMR spectrum of 45°KB(4-C1Ph)4 in CDC13. .......... 70

viii

3.8.

3.9.

3.10.

3.11

4.1.

4.2.

4.3.

4.4.

4.5.

Contour plot of the 500 MHz 1H ROESY spectrum of
45-NaBPh4 (CDC13, 0.03 M, 25 °C) which shows no
interionic NOEs.,,,

Side and end-on views of the X—ray structure of

282-NaBPh4 (BPh4‘ counterion not shown). ...............................

(a) Side and end-on views of the X-ray structure of
45'NaB(4-C1Ph)4. (b) Side and end-on views of the

X-ray structure of 45°KB(4-C1Ph)4. For both structures,
the B(4-C1Ph)4" counterion is not shown. ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,

X-ray structure of 45. .........................................................................

300 MHz 1H N MR spectrum of pure chloride 49
obtained on treatment of carbinol 48 in benzene-d6

with excess SOClz in an NMR tube. ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,

300 MHz 1H NMR spectrum (CDC13 solvent) showing
the mixture of chloride 49 (major) and methane 36
(minor) obtained on treatment of carbinol 48 in CH2C12

with ~10 equiv of SOClz. ....................................................................

300 MHz 1H N MR spectrum of pure dichloride 51
obtained on treatment of diol 50 in benzene-d6

with excess SOClz in an NMR tube. ................................................

300 MHz 1H NMR spectrum (CDC13 solvent) of the
mixture of dichloride 51 (major) and hydroxylic
impurity (minor) obtained on treatment of carbinol 50

in benzene with ~20 equiv of SOC12. ..............................................

(a) Simulated ESR spectrum of monoradical 47
generated with the proton hyperfine couplings obtained
from ENDOR measurements. (b) ESR spectrum of 47
in 2-MeTHF at room temperature. (c) ESR spectrum

of biradical 46 in 2-MeTHF at room temperature. ,,,,,,,,,,,,,,,,,,,,

eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee

79

80

81

87

88

89

90

....... 93

4.6.

4.7.

ESR spectra obtained on frozen 2-MeTHF solutions
at 120 K: (a) monoradical 47 and (b) 47 with
excess NaBPh4. ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 95

ESR spectra of biradical 46 obtained on frozen

2-MeTHF solutions at 120 K: (a) with no added salt;

(b) with excess LiI; (c) with excess NaBPh4; and

(d) with excess KB(4-C1Ph)4 .................................................................... 98

CHAPTER 1

INTRODUCTION

The research described herein is motivated by the challenge of
designing organic magnetic materials. Our strategy is to “turn on” magnetic
interactions between simple organic paramagnets with the structurally well-
defined relationships enforced by metal ion complexation. Ultimately,
extended chains of organic radicals strung together via metal ions offer the

possibility of a designed molecular solid with unique magnetic properties.

In order to design extended systems of interacting subunits, it is
desirable to examine the behaviors of the individual subunits, and their
simple pairwise interactions, to gain an understanding of the relationship
between structure and magnetic interactions within pairs of simple
paramagnets. Given two weakly interacting electrons, one ultimately needs
to know which forces favor the low-spin singlet state and which favor the
much rarer high-spin coupled triplet state, which promises the most
interesting magnetic behaviors. By designing dia- and paramagnetic
molecular models of substructures within the potential extended system, it is
possible to individually characterize the structural and magnetic coupling

components of the system.

This thesis deals with spectroscopic and structural studies of ion
binding by dia- and paramagnetic triarylpropeller ionophores. Because of the
highly interdisciplinary nature of the project, I shall begin Chapter 1 by
introducing the concepts of self-assembly and pairwise electron coupling;
these two overviews, respectively, are primarily based on reviews by
Lindsey1 and Rajca.2 Detailed rationale and description of the project are
presented at the end of Chapter 1. The use of metal ion complexation to

assemble triarylpropeller ligands requires a detailed description of the ion

binding properties of the triarylpropeller framework. Chapters 2 and 3
describe studies of ion binding by diamagnetic systems; this work is
extended in Chapter 4 to paramagnetic triarylpropeller ligands. Lastly,

Chapter 5 provides detailed information on experimental procedures.

1.1. Self-Assembly

Self-assembly is the spontaneous formation of a well-defined, higher-
ordered structure from a given set of components under specific conditions.
The term “self-assembly” is often used loosely in the literature; for use in a
strict sense, the assembly process must be reversible and the product stable
at thermodynamic equilibrium. All information for assembly is contained
within the subunits or precursor molecules, and neither additional factors nor
energy input are required for assembly to occur.1 The self-assembly of a
given architecture involves three stages: recognition between the
components, correct orientation so as to allow growth, and termination of
the process leading to a discrete, finite species.3 In this dissertation, the term
self-assembly is viewed as applying only to those processes, as defined
above, in which three or more separate components are brought together

through multiple molecular recognition events.

Many components of biological systems contain the information for
their own assembly. From a structural vantage point, formation of specific
three-dimensional objects occurs because only those molecules having the
correct positioning of functional groups can fit together to form the
maximum number of bonding interactions.4 Biological assembly not only

affords a level of specificity and architectural control without parallel in

chemical synthesis, but does so with great efficiency under ambient
conditions. Well-known examples of self-assembling biological systems
include oligomers of DNA, microtubules,5 and tobacco mosaic virus.6 The
self-assembly of the latter system from RNA and protein subunits is

portrayed in Figure 1.1.

 

Figure 1.1. Self-assembly of tobacco mosaic virus. Reproduced with permission from
ref 1. Copyright 1991 Gauthier-Villars.

Although the concepts of self-assembly are rooted in biology, they are
also found in chemistry. For example, crystallization of molecular solids is
self-assembly practiced by chemists on a daily basis. Crystal formation
involves rapid and efficient generation of highly structured entities under a
wide variety of conditions, and gives rise to assemblies having long-range
three-dimensional order. Since the molecular packing patterns are
determined by noncovalent, intermolecular interactions, one can view the
process of crystal formation as a molecular recognition process at the surface
of the crystal; repeated recognition events involving subsequent molecules
ultimately gives an ordered solid. Crystallization thus provides a useful

precedent for contemplating the use of self-assembly in materials chemistry.

Constructing complex structures via self-assembly processes, rather
than by tedious bond-by-bond syntheses, is certainly an attractive idea. But
beyond “mere” self-assembly lies the prospect of designed self-assembly—
fitting molecules together in an arrangement predetermined by the chemist.
When applied to the design of molecular solids, this is referred to as crystal
engineering. The ability to predict crystal structure based on molecular
structure would be invaluable for designing crystals that exhibit nonlinear
optical properties,7 electrical conductivity,8 or ferromagnetic behavior.9
Each of these solid-state properties is influenced by solid-state structure: the
component molecules must have both the requisite molecular structure and

the correct orientation with respect to one another in the crystal lattice. 10

Rational control and predictability of molecular self-assembly
processes are critical for directed crystal engineering, but control of
molecular orientation in supramolecular structure is difficult and is
recognized as a major obstacle in materials design. The problem lies in the
weakness of intermolecular forces, which do not always align molecules in
their equilibrium positions before they are trapped in the growing solid
phase. Attempts at crystal engineering have explored such forces as charge-
transfer,8a electrostaticf‘d’e and halogen-«halogen11 interactions, but

hydrogen bonds have been the most utilized.

Considerable effort has been devoted to the study of molecular
association via hydrogen bonds in solution, especially for cases involving
the formation of dimeric species. 12 This important work has delineated some

of the fundamental features of molecular recognition, and has demonstrated

that a full appreciation of the recognition properties of the individual

components is needed to predict the shapes of larger aggregates.

Hydrogen bonds are moderately strong (1-5 kcal/mol) and
directional,13 and are thus more likely to enforce orientation than charge-
charge or van der Waals interactions. In solid-state studies, Etter,14
Leiserowitz,15 Taylor and Kennard,l6 and others17 have systematically
characterized hydrogen bonding interactions and hydrogen bonding patterns

occurring commonly between specific functional groups.

Work benefitting from the studies above has shown that
complementary components can be brought together to create solid-state
superstructures exhibiting desirable properties. Of particular interest is
Veciana’s use of hydrogen bonding in appropriately substituted nitronyl
nitroxide radicals to obtain molecular solids with high dimensional
organization.18 Moreover, this fundamental work shows that hydrogen
bonding between open-shell molecules can be used to prepare solids
exhibiting ferromagnetic behavior. Finally, a wonderful example of designed
self-assembly was reported by Wuest. Self-assembly of a rigid tetrapyridone
produces a dimondoid network with large chambers, as illustrated in Figure
1.2. This network selectively enclathrates guest molecules present during
crystallization. 19 The use of tetrahedrally disposed pyridones to control
molecular aggregation is particularly elegant since Wuest predicted the
three-dimensional structure based on knowledge of the hydrogen-bonding
motifs preferred by dipyridones.

 

Figure 1.2. A schematic drawing of the three-dimensional dimondoid network of
hydrogen bonds formed by a rigid tetrapyridone host. Adapted from ref 17c.

Coordinate bonds involving metal ions and organic ligands have been
used in the self-assembly of inorganic superstructures. As with other
molecular recognition processes, multiple binding sites are usually required
for structural recognition since binding energies at any one site are small
compared to those provided by covalent bonds. More data are available
concerning the roles of bonding and structure (ring size, number of rings,
sten'c factors, ligand basicity, etc.) in determining the stabilities of metal

chelates than for any other self-assembling system.20

Crown ethers211 and especially their three-dimensional cryptand22 or
spherand-type23 derivatives have been developed to an extraordinary degree
of refinement for the selective binding of metal ions. The hydrophilic,
electronegative cavities of these ligands are ideally suited for complexation
of alkali metal and alkaline earth metal cations according to their size.24 The
relationship between ligand cavity and cation radius is readily apparent in
the crystal structures of crown ether complexes. The structure of a classical

crown ether complex is typified by the symmetrical array of oxygen atoms.

These atoms contact the cation which lies exactly in the center of the ring, as
seen in the structure of the 18-crown-6-KSCN complex25 (Figure 1.3a).
Self-assembly of crown ether complexes may result from a drastic size
discrepancy between the cavity and the metal ion, as illustrated in Figures
1.3b and 1.3c. In cases where the metal ion is too big to fit in the cavity,
sandwich complexes of 2:1 stoichiometry are formed, as found for [12-
crown-4]2-NaCl-5HZO,26 [12-crown-4]2-NaOH'8HzO,27 [IS-crown-
5]2‘BaBr2°2HZO.28 and [benzo-lS-crown-5]2'KI.29 When the metal ion is
too small to fill the cavity, binuclear metal complexes of 1:2 stoichiometry
can be formed. Examples include dibenzo-24-crown-8-2Na(o-
dinitrophenolate),30 dibenzo-24-crown-8-2KSCN,31 dibenzo-30-
crown-10'2NaSCN,32 and numerous complexes of transition metal ions with

bis-chelating macromonocyclic ligands.33

It should be emphasized that although the cavity-metal ion size
relationship is significant, it is a gross oversimplification to attribute all ion
binding specificity to this factor alone. Many factors influence the self-
assembly of a complex and its resulting structure, including ligand
substituents and topology, choice of solvent, and the anion involved.
Infrared spectroscopic studies have revealed that benzo-lS-crown-S forms a
sandwich complex with Na"' (which could smoothly fit into the cavity), if the
anion is BPh4‘.34 Evidently this is due to the inability of the anion to
provide donor atoms for the N a+ ion which thus requires a second crown for

sufficient coordination.

Macropolycyclic ligands possessing binding sites for several metal

ions also may self-assemble to generate inorganic superstructures. Binuclear

transition metal complexes with macrobicyclic cryptands35 and cylindrical

macrotricycles36 have been described.37

(a) (b)

 

Figure 1.3. Illustration of various cation/ligand arrangements in crystalline crown ether
complexes: (a) 18—crown-60KSCN;25 (b) [benzo-lS-crown-5]2-KI;29 and (c) dibenzo-24—
crown-8°2KSCN.31 X-ray coordinates were obtained from the Cambridge Structural
Database.

The ordering of metal ions and organic ligands into defined arrays
presents intriguing prospects for the development of molecular materials and
devices.38 To this end, the controlled generation of novel types of arrays is
of utmost interest. In these self-assembly processes, the ligands must contain
the steric program that is “read” by the metal ions following the “algorithm”

represented by their coordination geometry.3 A beautiful demonstration of

10

this concept is provided by self-assembling inorganic helices. Addition of
Cu(I) ions to ligand strands consisting of three bipyridine units results in the
spontaneous assembly the double-stranded helicate, as illustrated in Figure
1.4. In the complex, two ligand strands are wrapped around each other with
three Cu(I) ions holding them together.39 The double-helix structure results
from the tetrahedral-like coordination imposed by each Cu(bpy)2+ site, and
from the design of the ligand which disfavors binding to only one strand.
These two features make up, respectively, the recognition process (the
“algorithm”) and the molecular steric “program” that leads to preferential
formation of the double-helical structures. Of particular interest is that
binding of one metal ion facilitates the binding of subsequent ions (positive
c00perativity),39 and that a given ligand forms a double-helix preferentially
with an identical strand if a mixture of ligands is used (self-recognition)”
Constable and coworkers have shown that double-helical structures are
formed when the metal ion is too small for the cavity in the planar
(bipyridine) ligand configuration, and that n-stacking interactions play a
critical role in the stability of the double-helical geometry.40

§+ O—pg
Key/

Double- Strmded
Helicotc

/.'-22 Bipyridinc
Octrm

Figure 1.4. Self-assembly of a double-stranded helicate. Reproduced with permission
from ref 1. Copyright 1991 Gauthier-Villars.

11

Using suitably designed ligands and selected metal ions, the self-
assembly of several types of architectures has been described for double38’41
and triple42 helical complexes, as well as circular,3 capped,43 cylindrical,3
grid-like,44 and rack-type45 multi-component species. An important aspect
of this work is that inorganic superstructures can indeed be generated on the

basis of the structural design envisioned by the chemist.

From a coordination chemistry perspective, helicates are polynuclear
complexes containing a string of metal ions. Although these systems exhibit
all of the essential features of a strict self—assembly process, a disadvantage
is that the size of the architecture formed depends on the size of the
oligobipyridine strand, the length of which is dictated by covalent bonds. A
more practical approach to materials involves the use of multidentate ligands
and appropriate metal ions to generate coordination polymers, where
coordinate bonds are responsible for both the assembly process and the size
of the superstructure. This kind of approach has been applied to the
preparation of conducting materials, albeit with varied success. Metal
derivatives of 1,5-diformyl-2,6-dihydroxynaphthalene46 are reported to have
conductivities on the order of 10‘2 (ohm cm)‘1, but these are exceptional
cases. Recently, Fox et al. have investigated the effect of perturbed ligand
structure on the stability of several possible redox levels accessible to nickel-
phosphine complexes.47 Their studies involving mono- and bimetallic model
complexes 1 and 2 ultimately led to the preparation of a semiconductive
coordination polymer 3 from 1,2,4,5-tetrakis(dimethylphosphino)benzene
and NiC12-6H20.48

12

 

 

M62 M62 M62 M62 M62 M62
is c. p i . a r: . a
:1 \Nli/ \Nli/ \Ni/
E,/ +\P Fi/ +\['D e/ +\fi>
M92 M92 M92 M92 M92 M92
1 2
M62 M92 x CI—
P P C'
’/ \nii/
‘ \
\I? I?/ + ~
_ M62 M62 _ X
3

1.2. Electron Spin Coupling: The Pairwise Interaction

A pair of electrons can couple to give a total spin (S) of either S = O
(“antiparallel spins”) or S = 1 (“parallel spins”). These spin values
correspond to singlet and triplet spin states, respectively, and reference can
be made to either antiferromagnetic (S = 0) or ferromagnetic (S: 1) spin
coupling. The energy difference between the two spin states (AEST)
measures the strength of the coupling (neglecting spin-orbit coupling). In a
sense, antiferromagnetic coupling may be considered as a weak chemical
bond, with AEST being a measure of the bond strength. The challenge is to
achieve and understand ferromagnetic coupling, which is antithetical to

bonding.

In considering the simple case of pairwise electron coupling, a

question that arises is: What factors dictate the preference for one spin state

13

over the other? To answer this question requires explicit consideration of the

effects of electron repulsion.

Because electrons are indistinguishable, the Pauli principle requires
that the wave function for an atom or a molecule must be antisymmetric to
electron interchange. The determinantal wave function that ensures this
antisymmetry gives rise to two types of two-electron integrals.49 The first is
the Coulomb integral (J ab); it is the electrostatic repulsion between two
electrons in orbitals a and b. The second is the exchange integral (Kab), an
interaction that has no classical analogue. It is defined as the energy of
repulsion of the overlap charge density with an identical overlap charge
density.50 The physical interpretation is complicated, but essentially Kab is a
measure of the degree to which electrons in different orbitals “feel” each
other’s presence. Since this interaction is electrostatic in nature, Kab is
intrinsically positive. Its size depends on the compactness versus diffuseness
of the overlap density, and on the value of the one-electron overlap integral
(Sab)-51 The AEST depends on Kab and Sab’ as well as the one-electron
energy difference between the two orbitals. The interplay of Kab and Sab in
determining the lowest spin state is described below for examples in which

the two electrons are in two degenerate (or nearly degenerate) orbitals.

0%

Sab=0 Kab>0

 

Figure 1.5. Orthogonal p—orbitals illustrating Sab and Kab regions.

14

The triplet ground state of atomic C results from ferromagnetic
coupling of the two electrons in the half-occupied, orthogonal p-orbitals. In
this case Sab = 0 because the regions of positive orbital overlap cancel the
regions of negative overlap. However, the two p-orbitals are partially
coextensive in space (as defined by the overlap density) and so cancellation
of Kab does not occur—its magnitude is quite substantial. Pictorial
representations of Sab and Kab are presented in Figure 1.5. The driving force
for ferromagnetic coupling of spins is the Pauli principle, which allows
electrons of like spin to avoid each other and therefore minimize electron
repulsion. This is the physical significance of the Kab term. Electrons of
opposite spin (singlet state) are free to occupy the “exchange region” and
therefore experience a greater destabilizing electron repulsion. In accord

with Hund’s rule, the triplet state is thus lower in energy than the singlet.

When Sab at 0, it usually becomes the dominant term in determining
spin orientations, leading to spin pairing. For H2 the singlet is the ground
state, even at very large internuclear distances where one has an essentially
degenerate pair of NBMOs. As the two orbitals get further apart, Sab is
greatly reduced. However, the Kab term falls off more rapidly,
approximately as (Sab)2-51 Since the orbitals share a decreasing region of
overlap in which the Pauli principle can operate, the electrostatic advantage
of parallel spin alignment is outweighed by the attractive interaction of the
electron of one nucleus with the other nucleus, favoring the singlet state.
This description of the bonding in H2, which is based on MO theory,
provides a reasonable picture of why the singlet is the ground state at small
internuclear separations. However, it fails to predict that the singlet is

preferred at large internuclear distances. The problem is that the MO wave

15

function is constructed by placing the pair of electrons in a o—bonding orbital
without considering the fact that the electrons would attempt to avoid each
other (electron correlation). Consequently, the MO wave function tends to
exaggerate the ionicity of H2. This effect is significant at large distances
where the atoms are essentially independent and should be described by
separate, localized wave functions; in this situation the triplet is incorrectly

predicted by M0 theory to be the lowest spin state.

The fact that the singlet is the ground state for H2 at all separations is
better addressed by a valence bond (VB) description. The VB wave function
is such that each electron in the electron-pair bond between the two
hydrogen atoms tends to “reside” on its “own” atom. In considering electron
interchange, the VB approach overemphasizes electron correlation, placing
less weight (than the MO approach) on the ionic terms in the wave function.
As a result, VB theory better describes H2 at large separations where the
isolated atoms provide an extreme case, whereas MO theory is better at

small distances where the combined atoms provide the extreme case.

The preceding paragraphs have described the importance of electron
repulsion in determining the state ordering in simple two-electron systems.
Based on these considerations, a prescription for obtaining a triplet ground
state is for the two electrons to occupy the same regions of space (i.e.
substantial overlap density) but have a net overlap (overlap integral Sab) near
0. An obvious strategy leading to ferromagnetic interaction between two
spin carriers A and B is to arrange them so that the two SOMOs a and b are
orthogonal (or nearly so).52’53 Unfortunately, the orthogonality condition is

difficult to impose on the SOMOs of neighboring molecules. Moreover, it is

16

not always sufficient to provide ferromagnetic interaction;54 significant
overlap density is required. Thus far, this approach has only been
intentionally achieved by controlling the geometry through the use of

bridges connecting the spin carriers, as described below.

Altemant hydrocarbons (AHs) are a class of molecules comprised of
conjugated rings with even numbers of carbon atoms and linear conjugated
chains. Compounds 4—8 are common examples of AHs. The connectivities
of 4—6 make drawing Kekulé structures for these molecules imposible. This
type of connectivity (called odd-alternant) for biradicals 4—6 gives rise to
two degenerate, singly-occupied N BMOs. Conversely, Kekulé molecules 7
and 8 (called even-altemant) are closed-shell species and therefore do not
possess NBMOs.

 

In the case of odd-alternant systems, the half-filled NBMOs may be
confined to separate atom sets in the molecule so they do not span any
common atoms (disjoint MOS). In these systems the Kab term, which
corresponds to the simultaneous occupation of the same A0 and destabilizes
the singlet state, is insignificant. Consequently, the singlet and triplet states

for a disjoint system are nearly degenerate. Thus, predictions for the lowest

17

energy state are problematic, as in the case of tetramethyleneethane (5).55
On the other hand, the MOs may be orthogonal (so that Sab = 0) and
coincident at one or more atomic sites (non-disjoint MOS), resulting in
strong ferromagnetic coupling (since Kab is significant). The simplest and
best studied non-disjoint AH is trimethylenemethane (4). Spectroscopic
studies suggest a triplet ground state,56 and ab initio calculations predict
AEST 2: 15 kcal/mol.57 Compound 4 may be viewed as two methyl radicals
connected to the same end of ethylene (Ll-connection). Alternatively,
connecting two methyl radicals to opposite ends of ethylene gives butadiene,
a closed-shell (Kekulé) molecule (and therefore a ground state singlet) with
AEST = -74.3 kcal/mol.58 In a simplistic way, the ethylene moiety can be
thought of as a ferromagnetic coupling unit (bridge) when 1,1-connected,
and as an antiferromagnetic coupling unit when 1,2-connected. A large body
of spectroscopic work indicates that m-benzoquinodimethane (6), a non-
disjoint AH, has a triplet ground state59; ab initio calculations predict AEST
== 10 kcal/mol.60 In contrast, its 0-61 and p-isomers62 7 and 8, respectively,
are closed-shell (Kekulé) molecules having singlet ground states. Thus a
meta-connected benzene moiety is a ferromagnetic coupling unit, whereas
ortho- and para-linked benzenes are antiferromagnetic couplers. The
important observation here is that simple connectivities (bridges) that
produce non-disjoint NBMOs give ferromagnetically coupled systems. In
particular, meta-linkage of spin centers to a benzene ring has been realized
as a powerful paradigm for designing and synthesizing high-spin

molecules.63

One of the main drawbacks to biradicals such as 4 and 6 is their high

reactivity and poor stability; neither of these molecules is isolable. To

l8

prepare more robust biradicals, work has focused on linking stable
monoradicals via a spin coupling unit. Meta-connection of two
diphenylrnethyl moieties to a benzene ring corresponds to the Schlenk
hydrocarbon (9),64 which is almost completely oligomerized at room
temperature. Heating a solution of oligomerized 9, followed by rapid
cooling, gives an ESR signal at 77 K; Curie studies suggest that a minor
species possesses a triplet ground state.65 It is well-known that
triphenylrnethyl itself is highly associated in solution, dimerizing in a head-
to-tail fashion to a methylenecyclohexadienylidene structure.66 The use of
suitably bulky substituents, especially at the positions para to the benzylic
site, helps to improve the biradical’s stability and limits (or prevents)
association in solution.67 Examples include biradicals Ill—16,68 all of which

give triplet ESR spectra in frozen solutions.

Spin centers can also be connected via multiple coupling units.
Sequential linkage of meta-connected benzenes apparently leads to
ferromagnetically coupled systems despite the disjoint nature of this
connection. Weak ferromagnetic coupling is claimed in 1769; ESR Curie
studies on impure samples give AEST z 0.3 kcal/mol. Biradical derivatives
are also known that are based on sequential para-connection of benzene
units. Ground state singlets are found for Thiele’s hydrocarbon (18),70 a
close relative of 5; Chichibabin’s hydrocarbon (19);70’71 and the Miiller
hydrocarbon (20).72 It is important to point out that the lowest spin state is
often difficult to predict in these “stretched out” systems since coupling is
generally weak, and because structural or medium effects may be

significant},73

 

19

 

10 R=H,X=H
ll R=Me,X=H
12 R=i-Pr,X=H
l3 1R=Me,X=Me
14 R=CF3,X=H

 

20

—9— 9—9—
.Z

An example illustrating the use of single-atom bridges to link radical
centers is 1,3-cyclobutanediyl (21), a molecule that has been shown by
Dougherty74 to have a triplet ground state. The cause of the triplet ground
state in 21 is thought to be through-bond coupling,75 mediated by the
“bridging” CH2 groups. The through-space overlap of the radical p-orbitals
is substantial (S ab > 0), and would produce a large HOMO—LUMO gap
which favors a singlet ground state. However, the CH2 groups of 21 have
filled orbitals of 1!: symmetry that can mix with the p-orbitals, but only with
the symmetric combination (the HOMO). This through-bond interaction of
the radical p-orbitals raises the energy of the HOMO to a level that, by
coincidence, is nearly degenerate with the LUMO. Thus, the energetic
consequences of overlap between the radical p-orbitals are cancelled, but
Kab remains large, favoring a triplet ground state.76 This through-bond spin

coupling mechanism is analogous to the inorganic “superexchange”
model.77

21

The complex CuVO(fsa)2en-CH3OH (22)78 is a bimetallic system
that somewhat resembles biradical 21. The X-ray structure79 of 22 is shown
in Figure 1.6. In this complex the magnetic orbitals are orthogonal due to
molecular symmetry, and so Sab = 0. The unpaired electron of Cu (11)

(S = 1/2) occupies the dx2_y2 orbital, while the unpaired electron of V(IV)
(S = 1/2) occupies the dxy orbital. Here Kab is very large because the 2p
orbitals of the bridging oxygen atoms give both 0' overlap with a dxy orbital
on Cu and 1t overlap with a dxz—yZ orbital on V. This results in an
appreciable overlap density and therefore a large value of Kab- Compound

22 is thus a ground state triplet with a large AEST = 118 cm‘l.80

 

@Cu 0V

Figure 1.6. X-ray structure of 22.79 X-ray coordinates were obtained from the
Cambridge Structural Database.

A very interesting approach to “linking” two radical centers through
bridging units involves coordination of nitroxide ligands to a transition metal

ion.81 The coordination chemistry of these ligands has been reviewed.82

22

The nature of the coupling of a nitroxide directly bound to a metal ion
can be rationalized on the basis of orbital overlap considerations. The
unpaired electron of a free nitroxyl radical occupies a n* orbital shared by
the oxygen and nitrogen atoms. Metal orbitals of appropriate symmetry
include the de—yZ’ dxz, and dzz orbitals83 shown in Figure 1.7 for axially
bound nitroxides. Ferromagnetic coupling is expected when this orbital and
the de—yZ orbital of the metal are orthogonal (Sab = 0), as depicted in Figure
1.7a; this occurs when the M—O—N angle is 180°. An example of a complex
having this geometry is Cu(hfac)2(NITPh)2.84 Decreasing the M—O—N angle
in this situation results in greater overlap, which favors antiferromagnetic
coupling. Shortening of the M—O distance is expected to make any
interaction, either ferro- or antiferromagnetic, more intense; the nature of the

interaction just depends on the relative orientation of the interacting orbitals.

(a) (b) (C)

 

Figure 1.7. Possible orbital interactions between a nitroxide radical and a transition
metal ion: (a) 1t*—dx2_y2; (b) tt*-dxz; and (c) 1t*—dzz. Adapted from ref 81.

Antiferromagnetic coupling requires nonzero overlap (Sab > 0) of the

112* orbital of the radical and one or more magnetic orbitals of the metal ion.

23

Figures 1.7b and 1.70 are relevant to nitroxide complexes of Ni(II), Co(II),
and Mn(II), and they invariably lead to antiferromagnetic coupling. Thus,
the ground states of N i(hfac)2(proxyl)2,83 Co(hfac)2(proxyl)2,83 and
Mn(hfac)2(proxyl)285’86 have zero, one, and three unpaired electrons,
respectively. These spin states are due to antiferromagnetic coupling of the
nitroxyl spins (S = 1/2) with the metal-based unpaired spins (S = 1 (Ni),

S = 3/2 (Co), and S = 5/2 (Mn)).

In a recent overview,52 Kollmar and Kahn describe various strategies
for the design of ferromagnetically coupled systems, and summarize the
work as follows:

“To conclude, we would like to stress that the through-space
interactions on which we have focused in this Account are generally weak
except when p atomic orbitals belonging to adjacent molecules point to each
other. A way to increase the interaction is by linking the molecular units by
closed-shell bridges. Such an approach also allows one to control the relative
orientations of the units and therefore to impose the relative symmetries of

the interacting orbitals.”
1.3. The “Tripod Ether” Approach to Electron Coupling

We are working to achieve and understand electron coupling in radical
pairs or higher oligomers designed so that electron interactions are mediated
and enforced by metal cations. By “inverting” the concepts of Cram, Lehn,
and others—preorganization of an ion’s full complement of Lewis basic sites
and careful choice of cavity size—we seek to control the self-assembly of

organic free radical ionophores.

24

Tris(2,6-dimethoxyphenyl)methyl (23), originally synthesized by
Martin et al., is a remarkably stable free radical.87 It is monomeric and air—
stable, presumably by virtue of its D3 propeller conformation in which the
central carbon bearing the lone electron is protected from above and below
by tripods of methoxy groups. Studies on open-chain polyether ligands88
suggest that these ether oxygen “tripods” can serve as binding sites for metal
ions. Two molecules of 23 can self-assemble about a metal cation, fixing the
radicals’ relative orientation which, in turn, dictates electron coupling.
Scheme 1.1 portrays the proposed coupling in “interrupted G-bonds”—
radical pairs 24 or oligomers 25 in which metal cations mediate electron
interactions. For radicals like 23 which may bind metals on two faces,
alternating radical/metal cation stacks are anticipated; such chains would
center around linear arrays of one—electron carbon-centered p-orbitals
interacting through metal ions. Figure 1.8 shows an illustration of the

radical-metal ion—radical dimer 24 as calculated by molecular mechanics.

Scheme 1.1.

 

23 24 25

25

 

Figure 1.8. Optimized structure of dimer 24 as calculated by Molecular mechanics.

This work is concerned with the ion binding capabilities of dia- and
paramagnetic triaryl-Z propeller (Z = B, C', N) analogues of radical 23. The
triaryl-Z frameworks are new to the ion binding field, so their complexation
abilities have been studied using a variety of techniques, including NMR,
X-ray crystallography, ESR, and computational methods. Detailed
information has been obtained on the stoichiometry, energetics, and
geometrical nature of metal ion binding by these tripod ionophores. We have
found that the complexing abilities of the ether tripods depend strongly upon
solvent, metal cation, and counterion identities. In many cases, binding is
limited to one tripod per metal ion, as opposed to the hoped-for pairing of
tripods about the metal ion as in 24. To enhance their complexing abilities,
and to examine specifically the relationship induced between a pair of
tripods when they do coordinate about a single metal ion, we have tethered
two tripods to make a covalently-linked diamine ligand. Characterization of
simple 1:1 and 2:1 complexes between tripods and metal ions has provided a
basic picture of aggregation in extended structures, and has allowed us to
probe the requirements for the design of molecular solids with long-range

structure. Ion binding by the diamagnetic triaryl-Z tripods (Z = B, N) has

26

revealed the possibility of controlled assembly of paramagnetic tripod
ionophores (Z = C') whose pairwise electron interactions may then be turned
on by the binding event. Both a monoradical and a biradical analogue of 23
have been prepared, and ESR studies pursued to evaluate ion binding by

these systems.

CHAPTER 2

TRIARYL-Z PROPELLERS (Z = B, N)

27

28

2.1. Background

Establishing the ion binding potential of polydentate triarylmethyl
radicals is critical to the development of our approach. The structural basis
for ion complexation by these substrates is exemplified by the propeller-like
(helical) conformations adopted by per-ortho—substituted triaryl-Z
compounds (Z = B, C, N, or P). Surprisingly, radical 23 adopts an unusual
conformation in the solid state (Figure 2.1); one aryl ring is twisted out of
the central methyl carbon plane by only 12° while the other rwo rings are
twisted by 61°.89 This structure represents a point well along the way to the
transition state for the two—ring flip racemization pathway,90 and its large
deviation from the D3 ground state is unprecedented for triaryl—Z propellers.
The distorted “binding site” in solid 23 seems a poor representation of its
solution conformation, which is D3 in symmetry on the ESR timescale.87
The tripod binding site is more realistically shown by the X-ray structure of
carbocation 26. In contrast to radical 23, the more nearly D3 propeller
conformation of 26 positions the methoxy groups in such a way as to form a
pair of nucleophilic oxygen pockets in which the lone pairs project toward
the center of a small cavity. The crystal structure of 26°BF4 is shown in
Figure 2.2 alongside the calculated (MMP2) structure of radical 23.

\ \ \
O +
B
/03 /O3 /O3

23 26 27

 

Figure 2.2. Space-filling views of “tripod” binding sites: calculated (MMP2) structure of
23 (left) and X-ray structure89 of 26°BF4 (right, BF4— counterion not shown).

The structural similarity to podands88 suggests that these “tripod
ethers”9l can serve as hosts for metal cations. According to the principle of
preorganization,92 the smaller the conformational changes in host and guest
required for complexation, the stronger the binding. Although the
racemization barrier for 23 is predicted to be quite low,93 the symmetrical
substitution in all six ortho positions maintains the integrity of the ether
tripods. Thus 23 and its analogues are expected to require little

reorganization by the metal ion. With only three Lewis basic sites to offer an

30

ion, these substrates are not expected to be powerful ion complexants. Their

structures, however, appear to optimize the cavities they do exhibit.

\o
3 3 3
28 29 30

Ion binding studies involving radical 23 and the isostructural borane
27 had borne little fruit. The various difficulties and limited success with
these compounds prompted a study of the known tris(2-
methoxyphenyl)amine (28).94 The X-ray structure95 of 28 reported by
Miiller and Biirgi, shown in Figure 2.3, reveals a C 3 geometry with the three
aryl rings twisted from the C3 axis by ca. 45°. The nitrogen is slightly
pyramidalized to allow the methoxy substituents to move away from each
other, forming a nucleophilic pocket. This “binding conformation” is
apparently quite favorable despite the low rotational barriers observed for
tris-ortho-substituted triaryl-Z compounds,96 and the presence of only one

tripod binding site precludes formation of oligomeric chains.

 

Figure 2.3. X-ray structure of 28.95 X-ray coordinates were obtained from the
Cambridge Structural Database.

31

2.2. Ion Binding Studies of Triarylpropeller Ionophores

Compound 28 was prepared (70%) by a copper-catalyzed Ullmann-
type condensation of o-anisidine with 2—iodoanisole under phase-transfer97
conditions. The ability of amine 28 to bind metal ions was initially studied in
CDC13 solution by comparing the 1H NMR spectra in the absence and
presence of added salts. Unexpectedly, addition of excess solid Mg(ClO4)2
to 28 immediately produced an orange solution that exhibited broadening of
the methoxy proton resonance and coalescence of the aromatic multiplet, as
shown in Figure 2.4. These spectral changes could not be attributed to the
presence of excess solid since broadening persisted after filtration of the
sample (Figure 2.4d). Similarly, addition of excess LiBF4 to 28 also gave an
orange solution that showed analogous spectral changes; the unbroadened
1H NMR spectrum of 28 was recovered upon addition of D20. In a control
experiment, a solution of triphenylamine (29) in the presence of Mg(ClO4)2
instantly turned aqua-blue in color; the 1H NMR spectrum of the sample
revealed considerable line-broadening of the aromatic proton resonances.
The unbroadened 1H NMR spectrum of 29 was obtained on shaking the
sample with D20. In contrast, a solution of triphenylmethane treated with
excess Mg(ClO4)2 did not change color or exhibit the NMR line-broadening

phenomenon.

The observations recounted above suggested that both 28 and 29 were
being oxidized to their respective radical cations. A CDC13 solution of 28 (or
29) gave an unresolved ESR signal (no fine structure) when treated with
either Mg(ClO4)2 or LiBF4. A paramagnetic species was detected only when

the Mgz+ or Li+ salt was present with the amine—neither amines nor salts

32

Figure 2.4. 300 MHz 1H NMR spectra of 28 in CDCl3: (a) with no added salt; (b) with
excess Mg(ClO4)2 at t = O; (c) at t = 22 h; and (d) after filtration of sample (c) to remove
excess solid.

(a)

33

 

 

_t.._l

(b)

 

 

 

 

 

 

 

 

(C)
l
(J % r“ -/'/
((1)
4A L L a ”J L

 

 

rrrfirvv Tr—‘r—‘r—Ifirrri—VI'YI'
6.0 51'! 5.0 4.5 4.0

34

alone in CDCl; generated a radical species. It is interesting that tris(2-
methylphenyl)amine (30) and triethylamine do not undergo this oxidation
process. This may be a reflection of their higher oxidation potentials

(on = 1.0198 and 1.15 V99 vs. SCE, respectively) compared to those of 28
(on = 0.80 V vs. SCE)97 and 29 (on = 0.92 V vs. SCE). 100

It is known that the radical cations of both 28 and 29 (28“ and 29‘")
undergo facile self-reaction to form p-benzidines 3198 and 32,100,101
respectively, as shown in Scheme 2.1. From electrochemical studies
performed in acetonitrile, the bimolecular rate constant for benzidine
formation is 6.0 i 0.5 x 101 L mol‘1 sec‘1 in the case of 28"”, and
2.4 i 0.5 x 103 L mol'1 sec’1 for 29"".98 In our experiments, however,
solutions of the amines in the presence of Mg(ClO4)2 or LiBF4 remain
colored for a week or more, giving strong ESR signals. The 1H NMR spectra
of these samples never showed any evidence of a p-benzidine product, even
after “quenching” the radical species with D20. Earlier work has shown that,
when 29 is oxidized with 12, the ESR spectrum of the more stable 32"” is
observed. 101 Additionally, it has been observed that electrolysis of 29
initially gives the ESR spectrum of 29"”, which is soon replaced by the
spectrum of 32"" formed in the follow-up reaction. 100 The absence of fine
structure in our ESR spectra does not allow us to directly address this issue.
However, for the oxidation of both 28 and 29, our failure to observe
compounds 31 and 32 by NMR suggests that benzidine formation is not

appreciable, if it occurs at all.

35

Scheme 2.1.

— e +. dimerization
N -——> N p

 

3 3

Q“ Q Q Q
bNNR'NnNQ‘Qtl'NNN lit” 9.5

31 FI=OMe
32 R=H

In trying to determine the identity of the oxidizing species, 02 was
ruled out as the oxidant by the following experiment: Solutions of 28 were
prepared under argon in a glove-box using rigorously degassed (freeze-
pump-thaw cycles) CDC13; treatment of the solutions with either Mg(ClO4)2
or LiBF4 still resulted in oxidation of the amine, as confirmed by ESR.
Several other salts were also found to effect the oxidation process, including
LiClO4, LiPF6, and MgBrz. The likelihood of the anions playing a key role
in the oxidation process seemed small, especially since BF4‘ and PF6— are
used in electrochemistry as redox inactive counterions.102 However, the
presence of a metal ion appears important, since neither 28 nor 29 is

oxidized when treated with n-Bu4NBF4.

A salient result was the observation that a solution of 28, in the

presence of excess LiI, remained colorless and did not give an ESR signal.

36

Figure 2.5. 300 MHz 1H NMR spectra of 28 in CDCl3: (a) with no added salt; (b) with
excess Lil; and (c) after treatment of (b) with D20.

37

(a)

 

 

l_,._tl lL_

NNN W
.0
' ' 'N INI ' 7'0vava

 

 

 

 

 

.1121le ml
(C)
HOD
ti tr ti ' ts ' '5_‘FN ' 13‘ ll ' Tim

38

The 1H NMR spectrum of this sample is well-resolved and reveals
significant changes: The methoxy proton resonance is markedly shifted
downfield (A5 = 0.47 ppm) and the splitting pattern of the aromatic multiplet
has changed, as shown in Figure 2.5b. These spectral changes are not due to
a chemical transformation of the amine since the 1H NMR spectrum of
“unchanged” 28 is obtained upon addition of D20 (Figure 2.5c).
Additionally, treatment of 29 in CDCl3 with excess LiI did not produce an
ESR signal, nor significant shifting nor line-broadening in the 1H N MR
spectrum.

It appears that the oxidant is the CDCl3 solvent. This notion is
supported by the observation that experiments carried out in CD2C12 also
produce amine radical cations, but in poor electron-accepting media like
acetone, acetonitrile, benzene, or pyridine, oxidation is not observed.
Oxidation of 28 or 29 in the presence of counterions such as C104‘ and BF4‘
is interpreted as the inability of these anions to reduce the amine radical
cation once it is formed. For the purposes of NMR studies, the use of I‘ salts
effectively protects against the oxidation process: Evidently I’ is sacrificed
to make 12 instead of allowing the amine to be oxidized. Indeed, these
samples eventually show the yellow color characteristic of 12 in CDC13
solution. We have also found that oxidation is not a problem with BPh4" as
the counterion, although the reason for this is not understood. In any event,
complications arising from the oxidation of 28 can be surmounted by

judicious choice of anions.

Triarylamine radical cations are of interest since they have been

proposed as building blocks for the preparation of bulk magnetic materials.

39

Recently, Stickley and Blackstock103 have shown that two- and three-
electron oxidation of 33, a m-quinodimethane analogue, gives a triplet
diradical dication and a quartet triradical trication, respectively. In the long
run, the oxidation “difficulty” described above may give access to an
interesting new group of tripod paramagnets in the guise of the remarkably

long-lived amine radical cations we have observed.

p-Ap p-A|n

N N
p—An/ \p-An
p-An =«w.— OMe

N
p-An/ \p-An

33

The striking changes observed in the 1H NMR spectrum of 28 after
addition of excess Lil (vide supra) provides strong evidence for Li+
complexation. The different splitting pattern of the aromatic multiplet,
compared to that of the free host, is attributed to conformational
reorganization during complexation. Marked changes in the 13C NMR
spectrum of 28 are also observed upon addition of excess Lil, most notably
the methoxy 13C resonance (for which A5 = 2.56 ppm). Confirmation of Li"’
complexation is provided by 7Li NMR spectroscopy. Attempts to obtain a
7Li signal of LiI in CDC13 foundered due to the insolubility of LiI in this
solvent. However, when excess Lil was added to a solution of 28 in CDC13,

a 7Li resonance was observed at 2.11 ppm (relative to a 0.3 M LiCl/MeOH

40

external reference). A 1:1 complex stoichiometry (ZS-Lil) was determined
by comparing its 1H (OCH3) and 7Li NMR integrals to those of a
LiBPh4'3glyme reference sample; details are given in Chapter 5. Amine 28
cannot compete with H20 for Lil—hence the recovery of the “free” amine
1H NMR spectrum on shaking the sample with D20. In CDC13, compound
28 binds other salts as well; details of these binding studies are described in

Chapter 3.

The association of 28 with Lil in CDCl3 involves a rapid exchange
process. Fast exchange rates are favored by ligand flexibility and low
complex stability.104 The methoxy proton resonance of 28 (3.54 ppm) was
shifted to 3.62, 3.90, and 3.85 ppm on addition of ca. 0.5, 1.0, and 4.0
equivalents of Lil, respectively. Line-broadening of the methoxy proton
resonance and the aromatic multiplet was also observed (no ESR signals).
Equilibration of these samples over 4 days revealed further changes: The
methoxy signal had again shifted downfield (to 3.64, 4.01, and 4.03 ppm,
respectively), and all of the resonances had sharpened considerably. A
CDCl3 solution of 28 treated with Lil (0.5 equivalents) was examined by
variable temperature (VT) NMR. Decoalescence of the methoxy resonance
was observed between —20 and 0 °C, giving two signals at 3.54 and 4.04
ppm (ca. 1:1 by integration) which correspond to free and complexed host,
respectively. Broadening of the signals is due to the rapid exchange (on the
NMR timescale) of free and bound tripod ligands, which results in a
population-average signal. Unfortunately, we have been unable to determine
the molecularity of this exchange process since the kinetics are complicated
by the insolubility of LiI in CDC13 (in the absence of 28).

41

While ion binding in the relatively nonpolar chlorocarbon solvents is
interesting, we wished to measure binding constants and stoichiometries in a
solvent which could dissolve both host and guest. The solvents mentioned
previously—acetonitrile, acetone, and pyridine—are obvious candidates, but
studies with 28 have shown that evidence for ion binding is not readily
apparent in these media. These solvents, which have Gutmann donor
numbers105 of 14.1, 17, and 33.1, respectively, effectively compete with 28

for metal ions.

Nitromethane was selected for binding studies because of its low
donor number (2.7)105 and relatively high dielectric constant (35.9). As
observed in CDCl3, the exchange between free and complexed 28 in
nitromethane is fast on the NMR timescale. Again, only a single, population-
average signal is observed regardless of the amount of 28 or metal salt
present. Binding constants were determined from tandem 1H and 7Li (23Na)
NMR titrations. The salt (guest) concentration was held constant, and the
concentration of the host was incrementally changed to span a host to guest
ratio between 0 and 9. Complexation constants (Kf’s) were obtained from
plots of the 1H and 7Li (23Na) chemical shift variation as a function of the
host/ guest mole ratio; details of the experimental procedure and the data

treatment are given in Chapter 5.

Information obtained from the nonlinear least-squares curve-fits is
summarized in Table 2.1. The inflections in the binding isotherms are
indicative of 1:1 complex formation. Essentially the same Kf value is
obtained for 28-Li+ with the I’ and C104‘ salts. Both LiI and LiClO4 form

contact ion pairs in nitromethane,106 but this seems to have a negligible

42

effect on complex formation. The 28°Li+ complexes are ion-paired with the
anions, as evidenced by the anion dependence of the calculated 7Li chemical
shifts. Interestingly, the calculated 7Li shift for 28'LiI is nearly identical to
that seen in CDC13 solution for this complex. The calculated methoxy proton
chemical shift for 28-LiI (3.93 ppm) is also close to that observed for this
complex in CDC13 solution.

Table 2.1. Formation Constants and Chemical Shift Data for Alkali Metal Complexes of
28 in CD3N02 Obtained from Tandem 1H/7Li (23Na) NMR Titrationsa

 

 

Kr (M4) 5 (OCH3, ppm) 5 (Mt. ppm)
Lil 1.6 x 104 3.92 (3.57)b 2.51c (1.29)c.d
LiClO4 4.4 x 104 3.92 2.10C (0.84)“
NaBPh4 6.9 x 102 3.65 —8.94e (—12.9)e.d

 

3[M"] = 5 x 10'3 M. bChemical shift of 0.05 M 28 in CD3N02. cReferenced to 0.30 M
LiCl/MeOH. dChemical shift in the absence of 28. eReferenced to 0.30 M NaCl/HZO.

In contrast to LiI and LiClO4, NaBPh4 is completely dissociated in
nitromethane.107 The Kf value obtained for 28°Na+ is nearly two orders of
magnitude smaller than that of 28-Li+, indicating that Na+ is not complexed
as strongly as Li+. It is interesting to note that the data suggest a 1:1 complex
stoichiometry, but crystals of the 2:1 complex between 28 and N aBPh4 are
obtained on slow evaporation of a nitromethane solution (vide infra). The
calculated methoxy proton chemical shift for 28-NaBPh4 (3.65 ppm) is
significantly different from what is found in CDCl3 (3.14 ppm), for which a

2:1 complex is indicated (vide infra). In accord with this observation, the

43

calculated 23Na chemical shift for 28°NaBPh4 differs from that seen in
CDC13 for 282-NaBPh4 (—5.5 ppm), which is described in the next chapter.

To determine the importance of the ether tripod of 28 to ion binding,
the ligand properties of triarylarnines 34 and 35 were studied in CDCl3.
Compound 34 was synthesized (18%) from aniline and 2-iodoanisole using
the method described for 28, and 35 was similarly prepared (70%) from

diphenylamine and 2-iodoanisole.

\o \o
O NPh ‘,Nph2
2

34 35

When a solution of 34 was shaken with excess Lil, the complex 34'LiI
was formed. The 1H NMR spectrum of 34-LiI may be compared to that of
the free host: The methoxy resonance is shifted downfield (A5 = 0.40 ppm)
and the appearance of the aromatic region has changed. The 13C NMR
spectrum also exhibits significant changes, most notably the downfield shift
of the methoxy signal (A8 = 2.1 ppm). A 7Li resonance is observed at 8 2.74
ppm, which is considerably shifted (downfield) from that of 28-LiI. Addition
of D20 to a solution of 34°LiI recovers the 1H and 13C spectra of the free
amine. No salt uptake is observed (1H NMR) with NaI, NaBPh4, KB(4-
ClPh)4, KI, RbBPh4, RbI, CsI, or CsBPh4.

44

In contrast to 34, no changes are observed in the 1H or 13C NMR
spectra of 35 on addition of excess LiI, even after monitoring the sample
over several days. Furthermore, no salt uptake is observed (1H NMR) with
NaI, NaBPh4, KB(4-C1Ph)4, KI, RbBPh4, RbI, CsI, or CsBPh4 . Apparently
an ether tripod is not required for ion binding by these triarylpropeller
systems, but only one methoxy ether oxygen is not sufficient for
complexation. Not surprisingly, ion binding is not observed (1H, 13C, and
7Li or 23N a NMR) with any of the aforementioned salts when

methoxybenzene (anisole) or 1,3-dimethoxybenzene is the host in CDCl3.

It should be mentioned that both 34 and 35 are oxidized to amine
radical cations by Mg(ClO4)2 or LiBF4 in CDC13, giving light-blue and
emerald-green solutions, respectively. However, these oxidations appear to
be slower than those of 28 and 29, only producing ESR signals and line-
broadened 1H NMR spectra after ca. 1 day.

It is well-known from carbohydrate chemistry that, even in H20, as
few as three neutral oxygen atoms in a molecule suffice to form well-defined
and reasonably stable complexes with metal ions, provided that the oxygen
donors are suitably disposed. The 1:1 Li+ complex formation observed with
34 (and lack thereof with 35) suggested participation of the central nitrogen
as a third donor atom. The ion binding ability of tris(2-
methoxyphenyl)methane (36)108 was investigated to test this notion, but no
evidence of Li+ binding was found. A plausible interpretation is that the
methine hydrogen atom of 36 projects into the ether tripod, blocking
complex formation. The analogous tris(2-methoxyphenyl)borane, which

would be a more appropriate test system, turned out to be too sensitive for

45

isolation and use as an ion binding probe. Evidence suggests that this

compound is indeed formed, but decomposes during workup.

It was reasoned that since both borane 27 and trimesitylborane (37) 109
are fairly stable compounds, tris(2-methoxy-6-methylphenyl)borane (38)
should be as well. The compound was in fact prepared (65%) by addition of
2-methoxy-6—methylphenylmagnesium iodide to BF3°Et20 at 0 °C.
Stoichiometry determinations indicate that, like 28, this “one-faced” tripod
ether binds Lil in a 1:1 complex in CDC13. The 1H NMR spectra of 38 and
38°LiI are presented in Figure 2.6. No evidence (1H NMR) of complex
formation was obtained with LiClO4 or NaBPh4. While the results described
above are not direct proof of the nitrogen’s participation in complexes of 28
and 34, they do show that the ether tripod alone is sufficient to bind a metal

ion.

‘0 ‘o
‘0
CH 3 B N
. / o
3 3 / 3
36 37 38 39

Tris(2,6-dimethoxyphenyl)amine (39), the nitrogen analogue (Z = N)
of radical 23, was synthesized (vide infra) to examine the ion binding
properties of a “two-faced” triarylpropeller ionophore. Shaking a CDC13
solution of 39 with excess Lil gives a 1H NlVfll spectrum considerably

different from the free host: The methoxy signal is shifted downfield

46

Ea

- b — b

h h

m.m o.m m.m

b—hbbh—bbpp

.58 a 5% Be one an 3 .6 366% $22 E E: can .3 scam:—

c.v we 9m. Fm ob We cg

—bLlPN—»Ll~»—b~Lh——verb—bwhy—pp-p—h-ph—p-Lp

 

d

 

fl -

-3

m

til I 0 . . til ll

3V

 

 

 

 

 

47

(A5 = 0.32 ppm), as well as the meta- (A5 = 0.35 ppm) and para- (A5 = 0.35
ppm) ring proton resonances. A stoichiometry determination indicates that a
1:2 complex is formed between 39 and Lil, suggesting that one Lil is bound
per tripod unit of 39. The 39-2LiI complex is the only example to date in
which two metal ions are bound to a “simple” triarylpropeller ligand. 1 10 In

stark contrast, no ion complexation of any kind is seen with NaBPh4.

Unfortunately, the Lil complexes mentioned in the preceding
paragraphs appear to be stable only in solution. Attempts to crystallize these
complexes under various conditions have been unsuccessful. The following
section describes X-ray crystallographic studies of free ligands aimed at

structural characterization of tripod binding sites in the solid state.
2.3. Crystallographic Studies of Tripod Binding Sites

As mentioned earlier in this chapter, the free radical 23 was prepared
by J. C. Martin et al.; their ESR spectrum, obtained at —80 °C, indicated a D3
propeller structure with the three aryl rings twisted 45-50° out of the
coordination plane of the central carbon.87 CPK models of this species show
tripods of methoxy oxygens above and below the plane of the radical carbon;
these substructures call to mind the paired oxygen tripods which surround
alkali metal cations in Lehn’s [2.2.2] cryptates.111

As part of our binding site analysis, we reported X-ray structures of
23, 27, and 26-BF4.89 It was noted that the crystal geometry of radical 23 is
remarkably unsymmetrical, showing one ring nearly coplanar with the

carbon coordination plane, while the other two are steeply twisted (see Table

48

2.2). To gain insight into the geometrical preferences of per-ortho-
substituted Ar3Z propellers, we have synthesized and structurally
characterized three related tripod ethers, 39, 38, and 26°13. Crystallographic
data for these compounds are in Table 2.3, and the fractional coordinates can
be found in Tables 2.4A—2.6A in the Appendix. Amine 39 completes the
isosteric series Ar3Z, shown in Figure 2.7, where Z = B, C', N; borane 38 is
a structural intermediate between 27 and the known trimesitylborane (37);
and the salt 26°13112 may be compared to 26-BF4 previously described. The
new structures, 39, 38, and 26°13, are displayed in Figure 2.8.

Tris(2,6-dimethoxyphenyl)amine (39) was synthesized by the copper-
catalyzed Ullmann coupling of 2,6-dimethoxyaniline with 2,6-
dimethoxyiodobenzene under phase—transferg7 conditions. Intramolecular
congestion, which apparently blocks the formation of tris(2,6-
dimethylphenyl)amine and trimesitylamine by the usual condensation
methods,113 evidently does not prohibit triarylamine formation in this
system. However, prolonged reaction times did result in formation of the
cyclized amine 40; similar ring closures have been used to prepare 9-(2,6-
dimethoxyphenyl)-1,8-dimethoxyxanthydrol and sesquixanthydrol from
tris(2,6—dimethoxyphenyl)methanol.l 14

Table 2.3. Crystallographic Data for Compounds 39, 38, and 26°13

 

 

39 38 26-13
formula C24H27N06 C24H27BO3 C25H27I3O6
fw 425.49 374.29 804.20
F(000) 904 800 768
space group C2/c P21/n P21/c
crystal system monoclinic monoclinic monoclinic
z 4 4 2
a, A 10.813(2) 9.437(1) 740(2)

b, A 20.540(3) 15.364(2) 11.114(3)
c, A 9.982(2) 14.573(2) 16.682(2)
13, deg 9880(1) 9255(1) 9682(1)
v, A3 2190(1) 2110.8(8) 1362(4)
Dc, g cm-3 1.290 1.178 1.960
11(Mo Ka), cm—1 0.87 0.70 34.38
20m”, deg 55 50 45

final R3 0.057 0.079 0.057
final wa 0.076 0.093 0.067

 

are = 111701 — IFCIVZIFOI. bRw = {Zw(lFol — chl)2/EwlFo|2} 1/2; w = 1/62(|F0|).

50

.9258 ma .868 mo 2225a 3252:8me 2: 5:5 83388 8m .0235 an 658m 05 new 90: um 6:88 2: Co 8.5882
oEoEEmm mQ so: 2: @1505 AZ ..U .m n NV NA~>=on§x0€o§66dfi§ 8Com Some—080: 2: Co 85823 aarx 5N .9.sz

22 n. N am so u NV 3 a u NV S

 

51

The X-ray structure of 39 is similar to that of
perchlorotriphenylamine.113 The molecule occupies a crystallographic site
of C2 symmetry; the three C—N bond lengths (1.418, 1.416, and 1.416 A),
the three C—N—C bond angles (119.6, 120.2, and 120.2°), and the three aryl
ring twist angles (62.0, 61.0, and 610°) are all very similar. Thus, the crystal
geometry of 39 deviates only slightly from the ideal propeller shape (D3

symmetry), as found for the isostructural borane 27.89

(a) (b)

(C)

Figure 2.8. X-ray structures of (a) amine 39, (b) cation 26°13 (13" counterion not shown)
and (c) borane 38.

The X-ray structure of 38 reveals two methoxy groups on one face;
the third is on the other face. Consistent with this finding, the 1H NMR
resonances for the methoxy and the methyl groups of 38 split into 1:2 pairs

at low temperature. A variable temperature NMR study yielded ca. 9.7

kcal mol‘1 for the barrier to site interconversion. We attribute this barrier to

the coupled aryl ring rotations found by Mislow et al. for the related
dimesitylaryl boranes, which show barriers of 10—16 kcal mol‘l. 1 15

Table 2.2. Ring Twists in Hexamethoxytriphenyl-Z Propellers and Related Species

 

 

 

Compound (916 (926 93° GAMIC Comments
23a z = C' 12.2(2) 61.0(1) [61.0] 48.9f C2 Axis
27a z = B 62.8(2) 64.2(2) [64.2] 56.8 C2 Axis
39 z = N 62.0(1) 61.0(1) [61.0] 51.6 C2 Axis
26-BF4a z = C” 32.6(2) 46.1(2) 48.9(2) 43.48 General
215-13 z = C+ 35.8(2) 46.5(2) [46.5] 43.48 C2 Axis
38 z = B 57.3(3) 58.4(3) 61.6(3) 66.2, 64.5, 63.2h General
37b 2 = B 51.1 49.6 [49.6] 57.8 C2 Axis
41c z = B 25.8 52.9 56.3 27.0, 64.0, 64.0 General
44d 2 = C 4.3(6) 62.3(6) 65.5(6) — General
aRef 89.

bRef 112.

CRef 117.

dRei 121.

e’I‘wist angles (degrees) of mean aryl ring planes out of central atom’s three-carbon plane;
coplanar = 00°; square brackets enclose symmetry-defined twists.
fRadical AMI calculations used the half-electron methodlzo; UHF gives 46.4°.
SCation AMI calculations were run without counterion.
hX-ray and AMI values of G for the three unique aryl rings in 38 are listed in parallel
order. They may be labeled as 63 = unique (OCH3 “up”), 92 = pair (OCH3 “down”; ring
toward unique’s OCH3 side), 91 = pair (OCH3 “down”; ring toward unique’s CH3 side).

The series of boranes 27,89 38, 37 ,1 15,1 16 shows average twist angles

of 63.7°, 59.1°, and 501°, respectively. Significantly, the twist angles

become smaller as the methoxy groups are replaced by methyl groups.

53

Resonance between the n-donor methoxy groups and the boron center
should favor flatter structures. Most measures of steric bulk find methyl
groups to be larger than methoxy groups, again suggesting that the steeper
twists should be found in 37. Thus, the opposite trend might be expected on

grounds of both resonance and steric effects.

An interesting case in which, as in 23, one aryl ring prefers a
geometry coplanar with the central 2C3 plane is the Li+(12-crown-4)2 salt of
41, obtained by deprotonation of a para-methyl group in
trimesitylborane.117 Here, the twist angles of the neutral rings are 52.9 and
563°, while the quinoid ring remains 25 .8° out of the central coordination

plane despite the driving force for charge delocalization by n—overlap with

{)8}

41

the central boron.

Does the geometry of 23 reflect the radical’s intrinsic chemistry or is
it merely a statistical outlier in the normal range of planar triaryl-Z ring
twists? It is clear that the unsymmetrical geometry of 23 is not simply due to
its paramagnetic nature. Of the few known triarylmethyl radical
structures,89t118 the trimesitylborane radical anion, 1 16 and the tris(p-

biphenylyl)aminium radical cation,119 none show such extreme deviations

54

from threefold symmetry. However, the small number of Ar3Z structures
(where Z = B, C, N) prevents the use of purely structural arguments to
answer this question. In Martin’s original report and our own work, X-band
EPR (ca. 10 GHz frequency) of 23 shows three equivalent aromatic rings
down to —100 °C; if 23 is unsymmetrical, this finding places an upper bound
of 3-4 kcal mol‘1 on its ring equivalencing processes. Furthermore, two
structural comparisons hint that the propeller geometries are not grossly
affected by environmental influences: The twist angles in amine 39 closely
match those in borane 27; and in carbocations 26, the twist angles show little
variation between the BF4‘ and the 13‘ salts. Taken together, the above
observations make it seem unlikely that the strange geometry of 23 could

arise from packing forces alone.

If an electronic effect intrinsic to 23 produced this radical’s
unsymmetrical geometry, one might reasonably expect it to be reflected in
molecular orbital calculations. In fact, AMI calculations120 on compounds
23, 27 , 39, and 26 result in threefold symmetric structures in all cases, even
when the calculations begin at highly unsymmetrical structures or at the
crystal geometries; for comparison purposes, the computed twist angles are
included in Table 2.2. We have also computationally explored the energetic
cost of constraining the aryl ring twists in 23 to the twist angles (122°, 610°,
61 .0°) seen in the crystal structure. The AMI method finds that such a
distorted geometry is only 2.9 kcal mol‘1 higher in energy than the D3 form.
Thus, we believe that the solution structure of 23 is a threefold symmetric
propeller, but that a distortion of the type seen in the crystal structure is

easily accessed at room temperature.

55

 

43 44

A possible explanation for the apparently distorted crystal geometry of
23 was suggested by the discovery that oxygen slowly reacts with 23 to form
peroxide 42 and alcohol 43. As in 41, these two compounds should have
double bonds between the oxidized aryl ring and the erstwhile radical
carbon; thus the cyclohexadienylidene ring should be nearly coplanar with
the central carbon. The X-ray structure of the closely related peroxide 44
(Table 2.2) shows just such an unsymmetrical arrangement. 121 Both 42 and
43 are much less soluble than radical 23; it therefore seems reasonable that
these species could seed the crystallization of 23, inducing the radicals to

adopt contorted shapes.

CHAPTER 3

ION -BEARIN G PROPELLERS

56

57

3.1. Background

Metal ion complexation by tripod ethers is a new motif for self-
assembly of molecular systems. 122 Recent work from these labs has probed
the structural,89 dynamic,112 and electronic1 10b characteristics of tripod
ethers and their complexes. Our approach to organic molecular magnetic
materials rests on the abilities of paramagnetic tripod ethers (e. g. 23) to self
assemble by ion binding as in 24, putting the unpaired electrons into
magnetic communication in extended chains. This self-assembly strategy has
necessitated a detailed investigation of the ion binding properties achievable
within the tripod ether framework. Here we report detailed NMR and X-ray
structural studies of ion binding by tripod ether systems using stable,

diamagnetic analogues of 23.

As mentioned earlier, tris(2-methoxyphenyl)amine (28) is a useful
diamagnetic model for 23; it can offer only one tripod binding site,
precluding chain formation, and it crystallizes in a C3 geometry, placing the
three methoxy groups on one face of the propeller.95 Complex 282'Na+ then
mimics a radical-metal-radical subunit of 24. in the double tripod ligand 1,2-
bis[{2-bis(2-methoxyphenyl)aminophenoxy}]ethane (45), covalent linkage
of two molecules of 28 further promotes pairing of tripod sites about a metal
cation to form 45°M+. Vogtle and co-workers have examined three-armed
“open-chain cryptands” 123 related to 28 and 45, but to our knowledge, no
triarylamine propeller based complexes have been structurally characterized;
in particular, tris(2-benzyloxyphenyl)amine, a close analogue of 28, did not
yield crystalline complexes with alkali or alkaline-earth salts. 1 103 The NMR
and X-ray studies reported here for 28, 45, and related complexes validate

58

the ion binding strategy for self-assembly of tripod ethers. As detailed
below, the ion binding ability of ligand 45 augurs well for the use of
biradical 46 to probe pairwise metal cation-mediated radical-radical
coupling. Details concerning the synthesis of compound 45 are given in

Chapter 5.

 

45 (z = N)
46 (z = C')

59

3.2. Ion Binding Studies

Treatment of CDCl3 solutions of 28 with Lil, LiBPh4, or N aBPh4 in
excess leads to stoichiometric uptake of these otherwise insoluble salts, as
revealed by 1H, 13C, and alkali metal (7Li, 23Na) NMR spectra.
Stoichiometries and key NMR data are summarized in Table 3.1 and in
Chapter 5. With Li] and LiBPh4, 1:1 ligandzsalt ratios are found; NaBPh4,
however, yields a 2:1 ligand:salt stoichiometry, our first finding of twofold
complexation of a metal cation by a tripod ether. No salt uptake is seen in
CDC13 with NaI, KI, KBPh4, KB(4-C1Ph)4, RbBPh4, CsI, or CsBPh4.
Analogous studies of 45 show stoichiometric uptake of LiI, LiBPh4, N a1,
NaBPh4, and KB(4-C1Ph)4. With LiI, a 1:2 ligandzsalt ratio is found; it
appears that, as in 28, one tripod of 45 binds one equivalent of LiI (vide
infra). However, the other salts investigated—LiBPh4, NaI, NaBPh4, and
KB(4-C1Ph)4—show 1:1 stoichiometries upon complexation. No
complexation of K1, KBPh4, RbBPh4, CS], or CsBPh4 by ligand 45 was
observed in CDC13.

Beyond the above simple stoichiometries, solution structures of the
complexes may be inferred from detailed analysis of the 1H, 13C, and alkali
metal NMR spectra. For 28'Li1, time-averaged C3 symmetric coordination
of the Li+ ion in the ether tripod is indicated by a single methoxy lH
resonance and seven 13C signals. The narrow 7Li linewidth (AW/2 = 1.0 Hz)
suggests a symmetrical environment about Li".124 In a poorly solvating
medium like CDC13, the complex is most likely intimately ion-paired with
the 1‘ counterion,125 as seen in the recently reported structure of the Lil
complex of tris(2«-pyridylmethyl)amine;126 an MNDO-calculated structure

60

of 280LiI is presented in Figure 3.1a. Analogous complexation of an
equivalent of Li] in each tripod moiety of 45 is supported by 1H, 13C, and
7Li (8, AVl/z) data which are similar to those observed for 28'LiI. Thus, the
methoxy 1H and 13C shifts for 28 and 45 are nearly identical, both in the free
ligands (1H: 3.54 and 3.51 ppm; 13C: 55.7 and 55.8 ppm, respectively) and
in the resulting LiI complexes (1H: 4.02 and 4.01 ppm; 13C: 58.3 and 58.3
ppm), while the 7Li shifts are more variant (2.11 and 1.85 ppm).

For 28-LiBPh4, symmetrical coordination of Li+ by the three oxygen
atoms of the ligand is again suggested by 1H (Figure 3.2) and 13C NMR.
However, the methoxy 1H, 13C, and 7Li resonances of this complex are
shifted upfield from their counterparts in 28-LiI. Like the latter complex,
28'LiBPh4 can be envisioned as an intimate ion pair in which the Bth ion
completes the coordination sphere of the tripod-bound Li+ ion. In related
work, the ion pair formation constants in 1,2-dichloroethane solution are
reported to be 1660 M"1 and 7400 M‘1 for dibenzo-l8-crown-60LiBPh4 and
triphenylphosphine oxide-LiBPh4, respectively. 127

Figure 3.1b shows the structure of 28°LiBPh4 as calculated by
MNDO. The upfield chemical shifts above can be understood as resulting
from ring current effects due to the BPh4" phenyl groups. This View is
strongly supported by two additional findings: the 7Li resonance is much
broader in 28-LiBPh4 than in 28-Li1 (Av1/2 = 41 Hz vs. 1 Hz), and the
Bth ortho protons are shifted roughly 0.2 ppm downfield compared to
their chemical shifts in 45'LiBPh4, 45-NaBPh4, and 282-NaBPh4,
complexes in which the metal cation is shielded from direct contact with the

BPh4‘ counterion (vide infra). Analogous time-averaged interactions

61

8388338 m2 Z
888% 88 «2mm 38:88 £090 8 R773. .8 3:528 32 2pc .Um Z 88:? 2 QM 38080
8 32883? .omamsoxo +5 mo 88 05 mm 2:? 88 £22 08888 .3 3.888% A6... 8880988
8888308 ex 8888 28 8:88:80 05: 8:888 8 55 2 and 3888 8 30:80.35

 

 

S a 86 9.6 H o: 85 mam dam 28645.3
no a 8:. Ed 8 8; .31 8m .9} an a? 386-882.?
no it. a: N3 in end .me mam .m~.m £8 .43 £882.?
e e e :2 88 s? 82.9.

$3 8m wm.m £83.?

an? 8.4 2:. 33..

8m 3m me 8%: 8E

8.2. gm 428688288

8:- Rd 38888

85.? 2% 383.8

.8 $8 mmm 1 3 8.4 5.8

van 8 seem: new

fee 3 84. 9.8 a! 88 £28 88 .95on 88 .9508 868m

 

u. on 8 58 s 85358 +8 nee £2 .5 Baez as me .8 8.2 sea fizz .3” 2.3

62

(a) G?)

7Li: 5 =2.11
AV1/2 :1 Hz

     
   
 

7Li: 3 = —0.08

  
   
  

1H: 5 =4.02

ortho-1H shift
~O.2 ppm downfield“

(C)

Be,
7Li: 5 = 0.15

w. (AV1/2=4HZ

1H: 8 =3.3_8
U (not Spllt)

BAr4“ ortho-1 H
not shifted"

*Relative to any NaBAr4 complex

Figure 3.1. MNDO-calculated structures of Li+ complexes presented with salient NMR
data obtained on CDC13 solutions: (a) 28°LiI; (b) 28°LiBPh4; and (c) 45-Li+.

63

between Cs+i ions and Bth ortho protons have been uncovered by Bauer in
133Cs, 1H HOESY (heteronuclear Overhauser and exchange spectroscopy)
NMR studies of CsBPh4 contact ion pairs in pyridine solution. 128
Unfortunately, our 6Li, 1H HOESY129 NMR study of 28-6LiBPh4 in CDC13
did not reveal any interactions between the complexed Li+ ion and the

Bth counterion.

In each of the three Li+ complexes discussed above, the counterion
caps a Li+ ion bound in an ether tripod. The 1:1 ligandzsalt stoichiometry
found for complex 45-LiBPh4 implies a different binding motif, in which the
Li+ ion is sandwiched between the two triether tripods of 45. Two additional
comparisons lend support to this picture: First, the 7Li linewidth of
45°LiBPh4 is narrow (Avm = 4 Hz), indicating a more symmetrical 7Li
coordination sphere than that in 28°LiBPh4; second, as noted above, the
ortho protons of the BPh4“ phenyl rings do not exhibit the downfreld
shifting seen in 28-LiBPh4. Thus, the second tripod ether is preferred over
the BPh4‘ counterion as a capping ligand when the triarylamine subunits are
covalently linked. Consistent with this finding, the 7Li A8 between 45'2LiI
and 45-LiBPh4 (+1.70 ppm) is smaller than that between 28-LiI and
28'LiBPh4 (+2.19 ppm). The 1H NMR spectrum of 45-LiBPh4 is shown in
Figure 3.3, and the MNDO-calculated structure of the complex is presented

in Figure 3.10.

Although Li+ displays variable complex stoichiometries, Na+ is only
taken up by pairs of tripods. Complexes 282-NaBPh4 and 282-NaB(4-
C1Ph)4 are symmetrical on the NMR timescale, as shown by the simplicity
of their 1H and 13C spectra; the 1H NMR spectrum of 282°NaBPh4 is shown

.58 a 335.3 .8 5882; fizz E 5: 8m .3 253....

 

 

“oi: V5.0
vm.0m DMZ:
.LJ .|.[:LJ .tLJ
EQQ I.“ n A v m v n.w U m n m m. m nth m N.
M t
fl

 

 

+97“?
078'£—*

65

.maQU E vnmmfiomv Mo 8380mm M22 m: #3 8m .m.m charm

 

SSOZ'E—

H

SULB B

 

 

. c.mm . mi .

m. 1ft}? E
rm 0.0 mm 3.x WK
_ . . . . _ _ . .L . r t. . L . L|.l

5..
numb
mm

 

Wfi

66

./ ff/f i

 

 

 

Figure 3.4. 300 MHz 1H NMR spectrum of 282°NaBPh4 in CDCI3.

in Figure 3.4. As in the complexes of 45 with NaBPh4, NaB(4—ClPh)4, and
Na], their methoxy 1H resonances shift upfield relative to that of free ligand.
In the 45°Na+ complexes, the 23Na chemical shifts are independent of the
BAr4‘ counterion but are somewhat sensitive to the nature of the solvent.
The 23Na chemical shift for 45°NaBPh4 is —5 ppm (Av1/2 = 250 Hz) in
CDCl3/acetone—d6 (7:1, v/v) and —5.9 ppm (AVIIZ = 160 Hz) in
CDCl3/acetone-d6 (1: 1, v/v). 1n the presence of excess NaBPh4, two 23Na
signals are observed in CDCl3/acetone-d6 (7:1, v/v) at -5 ppm (complexed
Na+) and —10 ppm (solvated Na+). A similar experiment in CDC13/acetone-
d6 (1 : 1, v/v), however, shows only a single, population-averaged signal. The
23Na chemical shift for 0.03 M NaBPh4 is —10.4 ppm (AW/2 = 76 Hz) in
CDCl3/acetone-d6 (7:1, v/v) and -7.4 ppm (Avm = 14 Hz) in

67

CDCl3/acetone-d6 (1:1, v/v). In comparison, the 23Na chemical shift for
0.03 M 282°NaBPh4 is —8.2 ppm (AW/2 = 107 Hz) in CDC13/acetone-d6
(7:1, v/v) and —7.6 ppm (AV1/2 = 14 Hz) in CDCl3/acetone-d6 (1:1, v/v).
Addition of one equivalent of NaBPh4 to each of these solutions results in a
single, population-averaged signal at —8.4 ppm (Avm = 120 Hz) and at —7.6
ppm (AW/2 = 17 Hz), respectively. In CDC13 solution, the 23Na shifts for
282'NaBPh4 and 45-NaBPh4 are —11.9 ppm (Avllz = 240 Hz) and -5 .5 ppm
(Av1/2 = 260 Hz), respectively (see Table 3.1 and Chapter 5).

It is generally observed that 23Na chemical shifts move upfield both
with decreased solvent donicity130 and with lowered Na+ coordination
number by oxygen ligands. 131 Thus, the upfield changes in 23Na chemical
shifts from 45°NaBPh4 and 45-NaB(4-C1Ph)4 to 282-NaBPh4 and
282'NaB(4-C1Ph)4, and the greater dependence of the latter two on solvent
composition, imply that 28 is, as expected, the weaker ligand. As will be
seen below, X-ray structural data also support these ideas. Competition and
extraction experiments also indicate stronger binding of Na+ by 45 than by
28. Ligand 45 is competitive with 18-crown-6 for NaBPh4 in dry CDC13;
similarly, a CDC13 solution of 45 extracts N aBPh4 from D20 almost
quantitatively, as determined by 1H NMR. Unlike 45, ligand 28 is
completely unable to compete with D20 for NaBPh4. Finally, neither 28 nor
45 is a strong enough ligand to compete with D20 for Lil or LiBPh4.

In complexes 45-NaBPh4, 45'NaB(4-C1Ph)4, 45°NaI, and 45-KB(4-
ClPh)4, the methoxy resonances of 45 split into two equal intensity peaks;

the methylene resonances split into an apparent AB quartet with NaBPh4
(Figure 3.5) and NaB(4-C1Ph)4, but only broaden with NaI and

68

.58 a 3592.3 3 us me 3 co macaw $22 5 .5: 8m .3” 2:5

ran and m.m 7v mé o.m m.m ob

m6 TN mg
lb — n p — p n p _ b _ n n L Lt t- n n . _ P n .r .F p n P R bl _ n n n - PL m

b P . _ n h n .

__.__J

 

 

C
i I 1143i

3

 

 

 

 

69

.58 a xéuimazé 3 Rs 382.? 3 co seam fizz E E: 8m .3” 2:3...

 

 

 

 

3v

 

3

 

 

 

7O

.58 a zamzoimmé Lo 538% ~32 E £2 08 .5. 2:3..—

 

vofl 3K mm.om
vm§ hwd mmma
LquLJ} _LJ_|lL : L L
:50 o.m m.m c.v m.v ch. mb cm mb 0.x.
LIL: L L l— ? L Llr— L Lllrt_-l—.l. L L. L rrL L L p L LlL L — L L L L _ L L L L — L L P L — F TL L _

 

 

 

 

I1 I, l t 1‘ } ill» '11: it“; [ ill
% fl . _

 

 

L\\ \ \

 

71

.82
28:25 9. $9; 553 Go 2 .2 8d .598 3592.33 59.30% 88m E E: 8m 2: Lo 53 52:8 .3 2:3...

Lemmy an

o.m n.m o.v m.v o.n m.m o.w n6 o.h m.L.

  

 

o tn
b t~

llllllllllllolllllll
In 0 In 0 In

0
I
v

 

IIIIIIIIIIIIIIIIIII'IIII

I
r

I
I

E

if»:
N)

 

 

 

“l

 

 

W
o l E
h V
“52......
8 8
Z Z

72

KB(4-C1Ph)4. These results suggest that Na+ or K+ binding locks the
triarylamine subunits of 45 into propeller conformations, differentiating the
methoxy sites into two sets on the NMR timescale at room temperature.
Figure 3.6 shows the 1H NMR spectra of 45-NaBPh4 and 45-NaB(4-C1Ph)4,
while the spectrum of 45-KB(4-C1Ph)4 is displayed in Figure 3.7.

Lower limits on the binding energies of 45 with Na+ and K+ were
obtained by measuring the rates of methoxy group site exchange (kex) in
45-Na+ and 45°K+ using the saturation spin transfer (SST) technique. 132 In
CDC13 at 20 °C, the measured rate constants translate into AGi values of
71.8 and 66.8 k] mol‘l, respectively (see Table 3.1). The kc,x and AGit
values for complex 45-Na‘l' with BPh4‘ and B(4-ClPh)4‘ counterions are
equal within our uncertainties; evidently the site exchange barriers are not
differentially affected by these anions despite differences in the 1H N MR
spectra shown in Figure 3.6. Unfortunately, although the 1H NMR spectrum
of 45'NaI showed the methoxy group splitting characteristic of the other
Na+ complexes, its limited solubility precluded 13C, 23Na, and SST NMR

measurements.

In CDC13 solution, all of the systems discussed here are expected to
exist as tight ion pairs. 125,133 However, for the tripod ether complexes of
Na+ and K“, variations in the counterions have negligible effects on the 13C
and 23Na chemical shifts. The 13C spectra for 45-NaBPh4 and 45-NaB(4-
ClPh)4 show only minuscule differences, as do those for 282-NaBPh4 and
282-NaB(4-C1Ph)4. As noted above, 23Na spectra of 45-Na+ are insensitive
to the BAr4‘ counterion, as are the rate constants for methoxy group site

exchange. Excepting the case of 28°LiBPh4, evidence for ion interactions is

73

also lacking in the 1H and 13C spectra of the tetraarylborate anions, which
are essentially unaffected by the nature of the complex cations (i.e. 282-Na+,
45-Li+, and 45-Na"' with Bth; 282’Na“, 45-Na+, and 45-K+ with B(4-
ClPh)4)‘).

The probes that should be most sensitive to ion pairing are the nuclei
that make up the surfaces of groups around the “waist” of the ellipsoidal
complex; here an anion can most closely approach the center of positive
charge. Indeed, substantial variations are observed in the methoxy and
especially the methylene 1H chemical shifts for 45-NaX (where X = I",
Bth, B(4-C1Ph)4)"). However, both one—dimensional NOE and two-
dimensional 1H ROESY134 (rotating frame Overhauser effect spectroscopy)
NMR experiments on 45°NaBPh4 failed to show interionic N OEs between
the Bth counterion and the 45'Na+ core; the 1H ROESY NMR spectrum
is shown in Figure 3.8. Furthermore, the identical methoxy 1H chemical shift
values found in 282-NaBPh4 and 282-NaB (4-C1Ph)4 indicate that

interpretations of ion pairing should be made cautiously.
3.3. X-ray Studies of Metal Complexes

Crystals of 282-NaBPh4, 45°NaBPh4, 45-NaB(4-ClPh)4, 450KB (4-
ClPh)4'CH3N02, and free 45 have been analyzed by X-ray diffraction;
fractional coordinates for the compounds are given in Tables 3.2A—3.6A in
the Appendix. Crystallographic data for these compounds can be found in
Table 3.7, and selected geometrical information is compared in Table 3.8.

Drawings with partial atom-labeling schemes (counterions omitted) for
282'NaBPh4; 45'N3B(4-C1Ph)4 and 45'KB(4-C1Ph)4’CH3N02, and 45

74

Table 3.7. Crystallographic Data for 282°NaBPh4, 45-NaBPh4, 45'NaB(4-C1Ph)4,
45°KB(4-C1Ph)4°CH3N02, and 45

75

€58: n a ”S Sosawacoa t 1538 u 3% ._oL_w\__om_ t .05. u «a

 

 

z: d 38 Rod 3 .o Rod sax Ea

NS .c Rod God me .o 88 a: new

mm c. 5. mm 8 was $52

a; own 8a a; 3o 18 .3. 22V:

82 82 82 RE R2 Tao w am

8:3: 68: 83mg 8.5a €me 2 .>

88.8 @488 was .L

83.8 53.3 63$ Lean; $423 mac .n

25%: QSES wow 8

63.2 LESS: 68.8 @8me @432 < a

69%.: L382 €32 :: 5.: 68an a. d

€83 $95.2 388.2 85.2 68:: a s

N a 4 N 4 N

OLGSUOGOE OLE—00:05 08200608 08:09.55 OLGSOOGOE 839$ Rambo

wk am STAN wk .61an anew 8QO

SL 85 8mm 82 88 88:

8.80 8.3a 9:3: N3 § 3.22, E

eomzemso womzmsomemeo c923,18mammsu ooazmzaemsno oegg/88:80 2:88
m4 uozmmokiaimmev 2584828.. 382.3. Emmazxma

 

76

Table 3.8. Selected Distances (A), Angles (deg), and Torsion Angles (deg) in
282°NaBPh4, 45-NaBPh4, 45-NaB(4-C1Ph)4, 45°KB(4-C1Ph)4-CH3N02, and 45

.aomobm: 08$ 05 so 83:8 RE 35¢ 95 05 8a 2.0 can $28.: no £89.80 TC“ @832
.O 23 -Z 2: 2a 00 93 20 20:3 any—mg 002022.09 95 mo owfigm 2: ma mac 3.8 :03 é 3.5% .338 3:5 05 mo mg
28085 838 05 9 3:32 8239 68032 $25» 805 We .5"? .6825 we £8980 79w coxgd Ea - Z 2: 2w DU EH 29

 

77

 

as? 32%: €de 85% €34. «coonzonmzép

DE? 83.2: 398 €v.mm €68 “892012.29

£82. 83%. 3:8 A02? 653 “Buuzquép

p831 33%: €83. €23“ €53 «801292.29

DEM? 2: 3m. ENS €§N €an «892012.29

£591 832. Enmm €qmm €q$ “69292.2“.

8% $3.: €38 €08 3:90.89

€22 €22 €32 €32 Zuzuzw

EMS €mw§ €wm§ €83 8.3:

SE €8§ €mm§ €83 8.2H

€22 €3§ €$§ €82 5.2H

€82 €33 €o$a €§n~ Bus:

3%..“ €w~§ €83 €23 8.2g

aims 8:3.“ €§n~ €$n~ 5.3:

Emma €33 €83 6:3 97%

2%? €P§ summed €22 Ens:

83s Emma €82 2:3 €me azizH
me Nozmmuaseuimmé Aguimuzé samdzé Emmazlnfi

 

78

alone are given in Figures 3.9, 3.10, and 3.11, respectively. Complex
45-NaBPh4 is not displayed since its 45°Na+ core is visually
indistinguishable from that in 45'NaB(4-C1Ph)4.

In 282-NaBPh4, the eight-coordinate Na+ ion is sandwiched between
two polyether tripods with Na—N distances of 2.677(7) and 2.710(6) A and
six Na—O distances averaging 2.565(6) A. One Na—O distance is elongated
(2.726(6) A) relative to the other five (2.503(6)-2.583(6) A; average =
2.533(6) A), distorting the coordination sphere from an ideal bicapped
octahedron. Excluding the long Na—O interaction, these distances are
comparable to the Na—N (2.682(8) and 2.726(8) A) and Na—O (2.453(8)-
2.524(8) A; average 2.491(8) A) lengths found in 45'NaBPh4, and the Na—N
(2.677(6) and 2.685(6) A) and Na—O (2.428(5)-2.524(5) A; average 2.475(5)
A) lengths in 45'NaB(4-C1Ph)4. In light of the threefold crystallographic
axis through the nitrogen atom in free ligand 28,95 the nearly linear N—Na—N
angles in 45-NaBPh4 (l78.6(3)°) and 45-NaB(4-C1Ph)4 (178.1(2)°), and the
NMR data discussed above, the low symmetry and bent (ZN—Na—N =
154.5(2)°) geometry of 282-NaBPh4 are unexpected. The structures of
282-NaBPh4, 45°NaBPh4, and 45°NaB(4-C1Ph)4 are similar in having
approximate C2 symmetry and hence homochiral pitches for the two
triarylamine propellers; for 45'NaBPh4 and 45'NaB(4-C1Ph)4, this geometry
is exactly as expected from the solution NMR data described above. The
disposition of the basic atoms in 282-NaBPh4, 450NaBPh4, and 450NaB (4-
ClPh)4 is reminiscent of that seen in [2.2.2] cryptand'Na+ (C222'Na+)
structures. For comparison to the above values, the average Na—N and Na—O
distances in C222-Nal are 2.75 and 2.57 A, respectively.135 The six benzene

rings in the tripod ether complexes enforce eclipsed N—C—C—O angles;

79

notably, the analogous torsions in C222-Na+ can be nearly eclipsed, 136
although they are typically skewed. 137

In contrast to 45-NaBPh4 and 45-NaB(4-C1Ph)4, the K+ complex of
45 crystallizes as a nitromethane solvate (45-KB(4—C1Ph)4-CH3N02) which
readily loses solvent on standing. The nitromethane molecule does not
appear to interact with the 45-K+ complex. The eight-coordinate K+ ion is
bound between two polyether tripods with K—N distances of 293(1) and
2.95(1) A, and six K—O lengths averaging 2.73(1) A. Relative to the 45-Na+
complexes, this structure shows an increased O—C—C—O torsion angle
(71.23(2)°) and a markedly bent N—K—N angle (164.5(5)°). Such differences
are as expected based on the larger size of K+ vs. Na+, and hence the longer
M+—heteroatom distances in 45-KB(4-C1Ph)4'CH3N02 compared to those in
45-NaBPh4 and 45-NaB(4-C1Ph)4. These distances are comparable to the
average K—N and K—O lengths of 2.87 and 2.79 A, respectively, found in
C222-KI.138

 

Figure 3.9. Side and end-on views of the X-ray structure of 282°NaBPh4 (BPh4-
counterion not shown).

80

   

(b)

 

Figure 3.10. (a) Side and end—on views of the X-ray structure of 4S-NaB(4-C1Ph)4. (b)
Side and end-on views of the X—ray structure of 45°KB(4-C1Ph)4. For both structures, the
B(4-ClPh)4’ counterion is not shown.

81

Free ligand 45 may be viewed as two linked molecules of 28 with
similar aryl ring twist angles and nitrogen pyramidalization; as in 28, each
propeller’s three alkoxy groups are on the same face.95 Thus, the tripods of
45 represent convergent functionalities poised to encapsulate a suitable
guest. Aryl methyl ethers ordinarily prefer a conformation with the methyl
group and aryl ring coplanar, as seen for the methoxy groups of 28,95 45,
and related o-methoxy substituted triaryl-Z propellers 139 described in
Chapter 3. In anisole itself, the preference for a coplanar geometry has been
estimated at 2-3 kcal/mo].140 It is noteworthy, therefore, that the aryl-O
bonds in the tether of 45 show large out-of-plane torsions (TCN—CO—O—CHZ
= 73.0° and 131.5°) which are relieved in 45°M+.

 

Figure 3.11. X-ray structure of 45.

Although 45 is formally a podand,88 it is a surprisingly strong ligand.
As an open chain, six-fold benzannelated C222 analogue, 45 might be

82

expected to be a feeble ionophore, since bridge cleavage88 and
benzannelation”1 are both generally found to weaken the binding ability of
C222.142 However, scission of two bridges of C222 and benzannelation
have opposite effects on the system’s conformational flexibility. Two
individually deleterious modifications thus offset each other, leaving 45 with

substantial complexing ability.

When the last bridge in 45 is snipped, the now disconnected tripods 28
become much less effective complexants, as demonstrated above. These
triarylamine propellers are free to rotate96 to conformations in which the
ether tripods are less ideally arranged than they appear in the crystal
structure of 28. Furthermore, the generally weak Lewis basicity of aryl
substituted ethers143 and amines144 make the weak complexing ability of 28
unsurprising. What seems out of place is the strength of binding exhibited by
45, since all of the above criticisms of 28 apply here as well. We can only
surmise that in addition to the binding entropy decrease conferred by the
tether, the complexation-induced strain relief suggested by the structures of
free 45 and 45-M+ may also enhance binding. In any case, our observations
indicate that biradical 46 should be an effective complexant, allowing further
work to probe the effects of complexed metal ions on radical-radical

interactions.

CHAPTER 4

TOWARD AN “INTERRUPTED O-BOND”

83

84

We have made substantial progress toward a detailed understanding
of the ion binding properties of our tripod ether systems. X-ray studies
have yielded valuable structural characterization of the binding pocket in
both free and complexed ligands, while NMR studies have examined the
stoichiometries and dynamics of ion binding in solution. In the context of
self-assembly, the modest success with amine 28 presents the possibilty of
using radical 47 for the assemblage of a radical—metal ion—radical dimer
structure like 24. As noted in Chapter 3, the ion binding ability of amine
45 suggests that biradical 46 should be an excellent candidate to probe

metal-ion—mediated pairwise electron coupling.

4.1. Syntheses of Paramagnetic Tripod Ether Ionophores

Triarylmethyl radicals are commonly generated by reduction of the
corresponding triarylmethyl chlorides. Thus, 49 and 51 were sought as
precursors to radical 47 and biradical 46, respectively, as shown in
Scheme 4.1.

Tris(2-methoxyphenyl)methanol (48)108 was prepared (80%) by
addition of methyl-Z-methoxybenzoate to 2—methoxyphenylmagnesium
bromide. Carbinol 48 in chloroform was treated with acetyl chloride, as
described by Lund;108 the results were as described, but the isolated
product was a ~1:1 mixture of 49 and 48 (by 1H NMR) which had the
same melting point reported by Lund. Attempts to purify the desired
chloride by fractional crystallization resulted in cocrystallization with
carbinol 48. Invariably, mixtures of 49 and 48 were obtained upon

treatment of 48 with various halogenating agents under conditions that

85

have been used successfully in the preparation of other triarylmethyl
halides.”5 Ishizu and coworkers mention the preparation of 49—by
treating 48 in methylene chloride with 2 equiv of SOClz—but no other
details are given.146 In our hands, this procedure affords not only chloride
49, but also tris(2-methoxyphenyl)methane (36);108 48 is recovered. In
contrast, reaction of 48 with SOC12 in benzene gave 49 as the major
product, but with carbinol 48 as the impurity. Treatment of a benzene-d6
solution of 48 with SOC12 and monitoring by 1H NMR has shown that 48
is completely converted to 49 (Figure 4.1); however, evaporating the
volatiles and then redissolving the residue in benzene-d6 gives a mixture
consisting of 49 and 48. For our purposes, chloride 49 is best prepared,
albeit as a mixture, by treating 48 in methylene chloride with ~10 equiv of
SOClz. Using this procedure, a typical product mixture gives a 49/36 ratio
of 31:1 by 1H NMR (Figure 4.2). Although methane 36 is present as an
impurity, it is diamagnetic, it shows no capacity for ion binding, and it

should tolerate the reducing conditions used to prepare radical 47.

Diol 50 was obtained (80%) by addition of 1,2-bis(2-
carbethoxyphenyl)ethane to 2-methoxyphenylmagnesium bromide. As for
the 48->49 transformation, difficulties were encountered in cleanly
converting diol 50 to dichloride 51. Saturation of a suspension of 50 in
ether with anhydrous HCl gas, and standing overnight, gave a product
mixture containing 51, 50, and the dimethane derivative 52; presumably
the latter product is formed by hydride abstraction from the solvent.147
Reaction of 50 in methylene chloride with SOC12 gave an extremely
complicated product mixture (by NMR), and no attempts were made to

isolate and characterize these products. NMR experiments have shown that

86

Scheme 4.1.

MgBr ‘0 \

o 2. W SOCI2
\ —'—_‘* COH -——> CCI
EtZO/benzene CHZCIZ

reflux 3 3
48 49
0 0E! E O
1 . 0V0
Ag
+ 2-MeTHF
2. H

Et20/benzene
reflux

 

?
0 Ho 0\ 47

E 2

62348
50 2

SOCI2
benzene

[0 Cl 0\ 2 A9 E O\ 2
——————>
0 CI 0’ 2-MeTHF o 0’

Cr"
2 2

51 46

87

.32 522 .3 5
£00m $088 5:» 0805589 5 we 8588 «o 8253: no 35820 3. 035:? 23 mo 82.58% MEZ E NEE 8m 4.9 charm

 

B on, E E E
E c m m m o v m; o m m n a m Kim 95 min
b b b _ n F k — h p I P b F I n — n n P I P h _ h h h b h P h h b _ h - P I b _ . LID

 

 

 

fl Eamd ad 95 NS 1.5 95 \ \ \ \
hp...—.p-bpbnpn—hppnunnup—hPth-.hbbP-pppnpup—hbnuthhhh

e

 

 

88

.59. co >83 2.. a? 90:6 5 a. 6:35 mo usages 8 853%
28:5 on 28508 new €0.55 3. ootoEo .«o 8:328 05 wEBofi C838 QUQUV 886on ”:22 5 N52 com .Né ensure

we 2 3. 2m
3 1J 1....
.m 3. . . Fm... . . lam. . . t 3.. . . 3 we 3

1..
.45 .
1‘14 1 4 -

 

 

 

 

it‘ll-lift.

mmmmmwmm

 

89

.093 $22 :0
E £00m 3088 2:3 080.2323 5 cm 86 mo E0830: so “005050 Hm 02.8206 23 mo 888.0% 52 Z T: NEE com .mé 0.53m

. . . 0.2 . . .
m m s m Nflv _ h 3 a me a Ma
3.! 9m m. oi m w o.m m. o m m6 oK m.“
n n p b P bl - n h h by P b h — b h L! h b h P h h .— hl b b h b — n n h b n — h b n n - Lh|b -

 

 

m
J? 11m J J

 

 

\ ........:\\

 

 

BW'E

9O

.NfiUOm .«0 >350 ON... 539 0:0NGOD 5 cm ~06 .«o E08305 co 605050
285 bean: 0:?er Ba 00an E 88206 .8 23E: 2: .8 0538 £080 8.58% mzz 9 £2 08 .3. 0.55

«.8 WM. «.8 Q”.
in o_.m m_.m 4% m6 “WA.“ m6. o_.m m6 AWN m_.n

 

- - at:

 

£557 ' E

 

 

r=

FB'E' ‘

91

diol 50 in benzene-d6 is completely converted to dichloride 51 with SOC12
(Figure 4.3), but evaporation of the sample produces a mixture of 51 and
50. We have found that dichloride 51 is best prepared, albeit as a mixture,
by treating a benzene solution of 50 with ~20 equiv of SOClz; the product
mixture consists of 51 (typically >95%), along with a small amount (<5%)
of a hydroxlyic impurity (see Figure 4.4).

Treatment of chloride 49 or dichloride 51 with silver powder in
dry, degassed 2-methyltetrahydrofuran (2-MeTHF) under argon produces a
yellowish-orange solution of radical 47 or biradical 46, respectively
(Scheme 4.1); similar results are obtained in benzene, toluene, and
chloroform. In contrast to radical 23, both 47 and 46 are extremely

oxygen sensitive and must be handled in the absence of air.

4.2. ESR Studies of Ion Binding

The room temperature ESR spectrum of 47 in 2-MeTHF is shown in
Figure 4.5b; this spectrum shows excellent agreement with the simulated
ESR spectrum (Figure 4.5a) that is generated using the proton hyperfine
couplings146 for 47 obtained from ENDOR measurements. For
comparison, the room temperature ESR spectrum of 46 in 2-MeTHF is

given in Figure 4.5c.

The AmS = 1 region of the ESR spectrum at 120 K for radical 47 in
2-MeTHF, shown in Figure 4.6a, is indicative of a doublet monoradical. In
contrast, the AmS = 1 region of 46, under the same conditions, consists of

four symmetrical signals which are characteristic of a randomly oriented

92

Figure 4.5. (a) Simulated ESR spectrum of monoradical 47 generated with the proton
hyperfine couplings”6 obtained from ENDOR measurements. (b) ESR spectrum of 47 in
2-MeTHF at room temperature. (c) ESR spectrum of biradical 46 in 2-MeTHF at room
temperature.

93

 

 

 

 

 

 

 

 

 

 

 

A Monoradical
2 G H n n n (simulation)
H F n

l l l J

 

 

 

Monoradical ‘

 

 

 

 

 

 

 

c:
2
C

 

 

 

 

n Biradical

W I“
l

3222 3246
GAUSS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

94

triplet state (S = 1) biradical with the approximate zero-field degeneracy,
lE/hcl z 0. The center peak in the spectrum corresponds to a doublet
impurity, and is considerably more intense than the triplet peaks (Figure
4.7a). A half-field AmS = 2 transition is also observed, confirming the
assignment of a triplet species. The other zero-field splitting parameter is
lD/hcl = 0.0051 cm‘l; within the point-dipole approximation, this D value
corresponds to an average distance of ~8 A between radical centers
according to Raw/A = 1.375 lD/cm"1l‘1/3.148 Kurreck and coworkers have
observed triplet ESR spectra for highly concentrated frozen solutions of
tris(4-biphenylyl)methyl radical, and have interpreted these results as
electron coupling between pairs of doublet monoradicals which form
intermolecular D3 rt-complexes.149 However, it seems unlikely that the
triplet state observed for 46 is due to intermolecular spin coupling, since
samples of monoradical 47 prepared at twice the concentration of 46 only
show doublet spectra at 120 K. Finally, we note that the ESR spectrum of
46 at 120 K, both in frozen chloroform and frozen chloroform/acetone
(7:1, v/v) solutions, shows a doublet signal in the AmS = 1 region; no half-

field transition is detected in either case.

Samples of 47 and 46 in 2-MeTHF were treated with ~10 equiv of
LiI, NaBPh4, and KB(4-C1Ph)4. However, the ESR spectra of these
homogeneous samples, either at room temperature or at 120 K, are
essentially identical to those of 47 or 46 in the absence of added salt. The
ESR spectrum of 47 in the presence of excess NaBPh4 is displayed in
Figure 4.6b, while the spectra of 46 in the presence of excess Lil, NaBPh4,
and KB(4-C1Ph)4 are presented in Figures 4.7b—d, respectively. Similarly,

chloroform solutions of either 47 or 46 treated with the aforementioned

95

(a)
3240 G

10 G monoradical / 2-MeTHF

(b)

+ NaBPh4 (xs)

Figure 4.6. ESR spectra obtained on frozen 2-MeTI-IF solutions at 120 K:
(a) monoradical 47 and (b) 47 with excess NaBPh4.

salts also showed the same doublet ESR spectrum of 47 or 46 that is
observed in the absence of salts. Additionally, these samples were filtered,
evaporated in air, and the resulting residues were treated with D20. For
both radical 47 and biradical 46, no evidence of the Li+ salt was found in
the D20 washing by 7Li NMR, and no resonances attributable to the Na+
or K+ salt were observed by 1H NMR. As found with 2-MeTHF as the

96

solvent, no changes were observed in the ESR spectra of 47 and 46 in
chloroform/acetone (7:1, v/v) upon salt addition; in all cases, doublet

spectra were obtained.

Lastly, we note that evidence for ion binding by 47 or 46 has not
been detected in 2-MeTHF, chloroform, and chloroform/acetone (7: 1, v/v)

using the ESEEM (electron spin echo envelope modulation) technique.

The observations recounted above suggest that both radical 47 and
biradical 46 are poorer ligands than we had expected. These results are
quite surprising in light of the ion binding abilities exhibited by the
diamagnetic model systems. The absence of any changes in the ESR spectra
of 47 and 46 in both 2-MeTHF and ch10roform/acetone (7:1, v/v) upon
salt addition indicates that if pairwise electron coupling is present, it is less
than kT. We have not been able to directly address whether ion binding is
occurring in these solvents with either ligand, since UV—vis studies are
hampered by the oxidation of both 47 and 46 upon salt addition to give
intensely colored triarylmethyl cations, as previously observed with radical
23. The situation is somewhat clearer for samples studied in chloroform.
Indirect evidence (vide supra) suggests that 47 and 46 do not solubilize
LiI, NaBPh4, or KB(4-C1Ph)4, as previously observed with amines 28 and
45.

97

Figure 4.7. ESR Spectra of biradical 46 obtained on frozen 2-MeTHF solutions at
120 K: (a) with no added salt; (b) with excess Lil; (c) with excess NaBPh4; and (d) with
excess KB(4-C1Ph)4.

 

 

 

 

Mfr mil ml. fill
J «J J J

CHAPTER 5

EXPERIMENTAL SECTION

99

100

General Methods

Melting points were determined on a Thomas-Hoover apparatus and
are uncorrected. Fourier-transform infrared (IR) spectra were obtained on a
N icolet IR/42 spectrophotometer; each sample was measured as a thin layer
prepared by evaporating a CHC13 solution on a NaCl plate. Electron impact
(EI) mass spectra were obtained on a Fisons VG Trio-1 mass spectrometer.
Fast atom bombardment (FAB) mass spectra were run on a J EOL JMS-
HX110 high resolution double-focusing mass spectrometer; m-nitrobenzyl
alcohol was used as the FAB matrix. Elemental analyses were performed by
Galbraith Laboratories, Inc., Knoxville, TN.

Routine 1H and ”um NMR spectra were obtained at 300 and 75.5
MHz, respectively, on Varian GEMINI 300 or VXR-300 spectrometers. The
1H NMR shifts are referenced to residual lH resonances in the deuterated
solvents: acetone-d6 (5 2.04); benzene-d6 (8 7.15); CDCl3 (8 7.24); and
CD3N02 (8 4.53). The 113C shifts are referenced to those of the deuterated
solvents: acetone-d6 (8 29.8); benzene-d6 (5 128.0); and CDC13 (8 77.0).
Peak multiplicities are abbreviated: s, singlet; d, doublet; t, triplet; q, quartet;
dd, doublet of doublets; m, multiplet. Coupling constants (J) are reported in
hertz. The 7Li (116.57 MHz) and 23N a (79.35 MHz) spectra were recorded
on a Varian VXR-300 spectrometer and referenced to 0.30 M LiCl/MeOH
and 3.0 M aqueous NaCl, respectively; no chemical shift corrections were
made for bulk diamagnetic susceptibility differences between the sample and
reference solvents. Two-dimensional 1H ROESY134 and 6Li, 1H HOESY129

NMR experiments were performed on a Varian VXR-500 spectrometer.

101

All air-sensitive reactions were performed in oven-dried glassware
using standard syringe/cannula techniques. 150 Gravity column
chromatography was performed on E. Merck silica gel 60 (230—400 mesh)
or Fisher neutral alumina (Brockman activity 1, 80—200 mesh). Thin-layer
chromatography was done on E. Merck plastic-backed plates (silica gel 60,
F254, 0.2 mm; aluminum oxide 60, type B, F254, 0.2 mm).

Nitromethane (Fluka) and 1,2—dichlorobenzene (Aldrich) were used as
received. Benzene, toluene, Et20, and THF were freshly distilled from
Na/benzophenone ketyl under argon; 2-MeTHF was distilled from Na under
argon. Absolute EtOH was dried according to Lund and Bjerrum.151
Acetone was stirred for 24 h over 3203 (5 wt%), and then filtered onto a
fresh charge of B203 and distilled under argon. 152 DMF was stirred for 24 h
over activated 3A molecular sieves (5 wt%), and then filtered onto a fresh
charge of 3A sieves and distilled under reduced pressure. 152 Methylene

chloride was refluxed over CaH2 and distilled under argon.

Acetone-d6 and benzene-d6 were dried as described above for the
nondeuterated solvents. Chloroform-d was passed through a short column of
basic alumina immediately before use. Nitromethane-d3 was refluxed over
CaH2 for 24 h, then decanted and allowed to stand over powdered 4A sieves
(5 wt%) under argon for 12 h, and finally decanted onto fresh 4A sieves and

distilled under reduced pressure.

The following salts were obtained from commercial sources and used
as received: n-Bu4NBF4, LiBF4, NaBPh4, RbBPh4, CsI (Aldrich);
LiBPh4-3glyme, LiPF6 (Alfa); NaCl (EM Science); Mg(ClO4)2 (Fisher);

102

MgBrz, LiCl, KCl, RbCl, NaI, KI (Strem). LiI (Aldrich) was twice
recrystallized from acetone and dried in vacuo at 100 °C for two days.
LiClO4 (GFS Chemicals) was dried in vacuo at 180 °C for 30 h. The salts
LiBPh4, 153 6LiBPh4, KBPh4,153 RbBPh4,153 and CsBPh4 154 were
prepared by reacting NaBPh4 with the appropriate alkali metal chloride;
sodium contamination (%) in the products was determined by metal analysis:
LiBPh4 (< 0.04%); KBPh4 (0.079%); RbBPh4 (<0.2%); CsBPh4 (0.10%).
KB(4-C1Ph)4 (Fluka) was used as received; metal analysis gave < 0.03%
sodium content. The 6LiCl (>96% 6Li) used to prepare 6LiBPh4 was a gift
from Mr. J. F. Remenar in Professor D. B. Collum’s group at Cornell

University.

BF3-Et20 (Aldrich) was treated with dry EtZO (2 wt%) to ensure an
excess, and then distilled from CaH2 under reduced pressure. 155 n—BuLi
(Aldrich) was titrated with 2,5-dimethoxybenzyl alcohol immediately prior
to use according to the procedure of Ronald. 156 SOC12 (Aldrich) was
distilled from triphenylphosphite.157

Unless specified, all other commercial chemicals were used as
supplied: o-anisidine, 2-bromoanisole, 18—crown-6, 1,3-dimethoxybenzene,
2,5-dimethoxybenzyl alcohol, diphenylamine, ethyl salicylate, 2-iodoanisole,
2-iodotoluene, MeOH (anhydrous), 2-methoxy-6-methy1aniline, methyl 2-
methoxybenzoate, 2-nitrophenol, o-toluidine, triphenylmethane (Aldrich);
Mg pieces (Alfa); 2-nitroresorcinol, triphenylamine (Eastman Kodak); and
Cu powder (Lancaster).

103

Syntheses

Tris(2-methoxyphenyl)amine (28). The synthesis of this compound
has been described by Frye et a1.94 and more recently by Soulié et al.,158 but
neither report gave 13C N MR data. A modified97 synthesis and complete
characterization details are as follows:

A mixture of o-anisidine (1.32 g, 10.7 mmol), 2-iodoanisole (6.27 g,
26.8 mmol), anhydrous K2CO3 (11.9 g, 86.1 mmol), 18-crown-6 (0.57 g,
2.16 mmol), Cu powder (2.73 g, 43.0 mmol), and 1,2-dichlorobenzene (20
mL) was refluxed under argon for 23 h. The hot mixture was filtered and the
filtrate distilled under reduced pressure. The resulting residue was twice
recrystallized from acetone to give 28 (1.58 g, 44%) as off-white crystals:
mp 145-146.5 °C (1it.94 mp 145—147 °C); 1H NMR (300 MHz, CDC13) 5
7.04—6.97 (m, 3 H), 6.85-6.75 (m, 9 H), 3.54 (s, 9 H); 13C{1H} NMR
(CDC13) 8 153.1, 137.8, 124.5, 123.7, 120.6, 112.6, 55.7.; EIMS m/z
(relative intensity) 336 (M+1, 24), 335 (M+, 100), 290 (16), 289 (72);
FABHRMS calcd for C21H21NO3 335.1522, found 335.1534.

Tris(2-methylphenyl)amine (30). A mixture of o-toluidine (2.30 g,
21.5 mmol), 2-iodotoluene (11.7 g, 53.7 mmol), 18-crown-6 (1.15 g, 4.35
mmol), anhydrous K2CO3 (23.8 g, 172 mmol), Cu powder (5.47 g, 86.1
mmol), and 1,2-diehlorobenzene (40 mL) was refluxed under argon for 16 h.
After cooling, the mixture was filtered through a thin layer (~2 cm) of silica
gel. The inorganic solids were washed with hot CHC13 and the combined
filtrates distilled under reduced pressure. The residue was twice
recrystallized from 95% EtOH to give 30 (1.48 g, 24%) as yellow crystals:
mp 104—105.5 °c (1it.94 mp 104—106 °C); 1H NMR (300 MHz, CDC13) 5

104

7.15—6.70 (m, 12 H), 1.88 (s, 9 H); 13C{1H) NMR (75.5 MHz, CDC13) 5
147.3, 133.2, 131.6,126.5, 125.2,123.6,18.6.

Phenyl-N,N-bis(2-methoxyphenyl)amine (34). A mixture of aniline
(1.99 g, 21.4 mmol), 2-iodoanisole (12.6 g, 53.8 mmol), anhydrous K2C03
(23.8 g, 172 mmol), 18-crown-6 (1.16 g, 4.40 mmol), Cu powder (5.46 g,
86.0 mmol), and 1,2-dichlorobenzene (40 mL) was refluxed under argon for
24 h. The hot mixture was filtered and the filtrate distilled under reduced
pressure. The resulting residue was twice recrystallized from 95% EtOH to
give 34 (1.19 g, 18%) as pale-yellow crystals: mp 91.5—92 °C (lit.159 mp 90
°C); 1H NMR (300 MHz, CDC13) 5 7.18—7.07 (m, 6 H), 6.93—6.85 (m, 4 H),
6.80—6.75 (m, 1 H), 6.66-6.63 (m, 2 H), 3.65 (s, 6 H); 13C{1H} NMR (75.5
MHz, CDCl3) 5 155.4, 148.3, 135.7, 128.7, 128.3, 125.9, 121.2, 119.1,
117.4, 112.9, 55.8; EIMS m/z (relative intensity) 306 (M+1, 19), 305 (M+,
100), 274 (7), 260 (10), 259 (49), 182 (10); FABHRMS calcd for
C20H19N02 305.1417, found 305.1421.

Diphenyl-N-(2-methoxyphenyl)amine (35). A mixture of
diphenylamine (4.04 g, 23.9 mmol), 2-iodoanisole (8.29 g, 35.4 mmol),
anhydrous K2C03 (13.1 g, 95.0 mmol), 18-crown—6 (0.63 g, 2.4 mmol), Cu
powder (3.06 g, 48.1 mmol), and 1,2—dichlorobenzene (40 mL) was refluxed
under argon for 14 h. The hot mixture was filtered and the filtrate distilled
under reduced pressure. The resulting residue was twice recrystallized from
95% EtOH to give 35 (4.63 g, 70%) as white crystals: mp 74.5—75 °c (1it.98
mp 74—75 °C); 1H NMR (300 MHz, CDC13) 5 7.23—7.13 (m, 6 H), 7.01—
6.87 (m, 8 H), 3.62 (s, 3 H); l3C{1H} NMR (75.5 MHz, CDC13) 8 156.1,
147.7, 135.5, 130.2, 128.8, 126.7, 121.6, 121.4, 113.3, 55.9 (remaining 13C

105

resonance not observed); EIMS m/z (relative intensity) 276 (M+1, 23), 275
(M"’, 100), 260 (15), 259 (10), 244 (12), 182 (15); FABHRMS calcd for
C19H17NO 275.1311, found 275.1304.

Tris(2-methoxyphenyl)methane (36). Carbinol 48 (2.00 g, 5.72
mmol) was dissolved in refluxing 95% EtOH (125 mL) and treated with
concentrated HCl (10 mL, 329 mmol). A deep purple color developed and
then began to fade after ca. 2 min. The mixture was refluxed for 12 h and the
colorless solution then cooled to room temperature. The product crystallized
at —20 it as a white solid (1.55 g, 81%): mp 136—137 °C (lit.108 mp 136—
137 °C); 1H NMR (300 MHz, CDC13) 5 7.20—7.14 (m, 3 H), 6.86—6.70 (m, 9
H), 6.40 (s, broad, 1 H), 3.67 (s, 9 H); 13C{1H} NMR (75.5 MHz, CDC13) 5
157.3, 132.5, 129.6, 127.0, 119.9, 110.7, 55.8, 36.9; EIMS m/z (relative
intensity) 335 (M+1, 25), 334 (M+, 100), 319 (15), 303(36), 227 (9), 226
(10), 195 (9), 181 (9), 121 (55), 107 (17), 91 (24).

2-Iodo-3-methylanisole. 2-Methoxy-6-methylaniline (8.23 g, 60.0
mmol) was dissolved in concentrated H2804 (50 mL), cooled to 0—5 °C, and
treated with a solution of NaN02 (4.28 g, 62.0 mmol) in H20 (20 mL).
When the addition was complete, the mixture was stirred for an additional
30 min at 0—5 °C, and then filtered through a plug of glass wool into a
solution of K1 (99.3 g, 0.60 mol) in H20 (150 mL). The resulting brown
mixture was heated on a steam bath until the evolution of N2 ceased, and
then EtZO (200 mL) and 1 M aqueous NaZSO3 (20 mL) were successively
added. The organic extracts were combined, washed with 2 M aqueous
NaOH (4 x 100 mL), dried over Na2804, and evaporated to give a red/black
solid. Chromatography over silica gel using CH2C12/hexanes (1:1, v/v)

106

afforded the iodide (4.08 g, 28%) as a white solid (Rf: 0.73): mp 43—43.5
°c (lit.160 mp 49 °C); 1H NMR (300 MHz, CDC13) 5 7.17 (t, 1 H, J = 8.0
Hz), 6.86 (d, 1 H, J = 8.2 Hz), 6.62 (d, 1 H, J = 8.3 Hz), 3.86 (s, 3 H), 2.46
(s, 3 H); 13C{1H} NMR (75.5 MHz, CDC13) 5 158.1, 143.4, 128.7, 122.4,
108.0, 93.1, 56.5, 28.8; EIMS m/z (relative intensity) 249 (M+l, 11), 248
(W, 100), 233 (18), 121 (3), 106 (14), 91 (15); FABHRMS calcd for
C3H910 247.9699, found 247.9706.

Tris(Z-methoxy-6-methylphenyl)borane (38). A solution of 2-iodo-
3-methylanisole (1.97 g, 7.94 mmol) in dry Et20 (10 mL) was added
dropwise with stirring to Mg pieces (0.19 g, 8.0 mmol) in EtzO (15 mL)
under argon. When the addition was complete, the mixture was stirred for an
additional 1 h, and cooled to 0—5 °C; freshly distilled BF30Et20 (0.32 mL,
2.64 mmol) in EtzO (10 mL) was then added dropwise. The mixture was
refluxed for 20 h, cooled, and poured onto crushed ice (60 mL) containing
5% aqueous HCl (5 mL). The aqueous layer was separated and extracted
with CH2C12 (4 x 50 mL). The organic extracts were combined, dried over
N a2SO4, and evaporated to give a brownish-orange solid. Recrystallization
of the crude product from 95% EtOH afforded 38 (0.60 g, 61%) as a white
solid: mp 169.5—170 °C; 1H NMR (300 MHz, CDC13) 5 7.13 (t, 3 H, J = 8.0
Hz), 6.69 (d, 3 H, J = 7.5 Hz), 6.63 (d, 3 H, J = 8.3 Hz), 3.41 (s, 9 H), 2.09
(s, 9 H); 13C{1H} NMR (75.5 MHz, CDC13) 5 161.2, 141.0, 129.4, 122.6,
108.9, 56.1, 21.2 (remaining 13C resonance not observed); EIMS m/z
(relative intensity) 375 ('OBM+2, 10), 374 (“’BM+1, 44), 373 (“’BM+, 10),
253 (13), 252 (53), 251 (17), 238 (35), 237 (100), 236 (25), 223 (17), 222
(37), 221 (20), 208 (12), 207 (54), 206 (15), 205 (10), 195 (11), 193 (25),
192 (13), 191 (23), 180 (11), 179 (46), 178 (51), 166 (15), 165 (42), 152

107

(15), 133 (30), 132 (12), 117 (10), 105 (64), 104 (11), 103 (15), 91 (28);
FABHRMS calcd for C24H27BO3 374.2054 (“13), found 374.2060.
Colorless needles of 38 suitable for X-ray analysis were obtained by

slowly evaporating a CDC13 solution in air at room temperature.

2,6-Dimethoxynitrobenzene. A mixture of 2-nitroresorcinol (15.0 g,
97.0 mmol), anhydrous K2C03 (26.8 g, 194 mmol), CH3I (13.5 mL, 217
mmol), and acetone (100 mL) was refluxed under argon for 10 h. The
solvent was evaporated, H20 (250 mL) added, and the mixture extracted
with EtZO (4 x 100 mL). The organic extracts were combined, washed with
2 M NaOH (3 x 100 mL), and dried over MgSO4. Recrystallization from
95% EtOH afforded the nitro compound (10.11 g, 57%) as pale yellow
needles: mp 129.5—130 °c (lit.161 mp 129—130 °C); 1H NMR (300 MHz,
acetone-d6) 8 7.44 (t, 1 H, J = 8.5 Hz), 6.85 (d, 2 H, J = 8.5 Hz), 3.90 (s,
6 H); 13C{ 1H} NMR (75.5 MHz, acetone-d6) 8 152.5, 132.2, 120.5, 105.7,
57.0; EIMS m/z (relative intensity) 184 (M+1, 10), 183 (M"', 100), 136 (44),
122 (12), 108 (11), 107 (48), 95 (20); FABHRMS calcd for C8H9NO4
183.0532, found 183.0535.

2,6-Dimethoxyaniline. A mixture of 2,6-dimethoxynitrobenzene
(2.78 g, 15.2 mmol), activated carbon (600 mg, 50-200 mesh), FeCl3-6H20
(254 mg), and MeOH (50 mL) was refluxed with stirring for 15 min.
Hydrazine hydrate (2.9 mL, 60.0 mmol) was then added in three portions
over 30 min. The mixture was refluxed an additional 14 h, cooled, and
evaporated. The resulting slurry was dissolved in CH2C12 and filtered.
Chromatography over silica gel using CH2C12 gave the aniline (1.52 g, 65%)
as a white solid: mp 76—77 °C (lit.162 mp 75.5—77 °C); 1H NMR (300 MHz,

108

CDC13) 5 6.61—6.51 (m, 3 H), 4.00 (s, broad, 2 H), 3.79 (s, 6 H); 13C{1H}
NMR (75.5 MHz, CDC13) 5 147.9, 127.0, 116.7, 105.0, 56.0; EIMs m/z
(relative intensity) 154 (M+1, 12), 153 (M+, 63), 139 (10), 138 (100), 95
(49); FABHRMS calcd for C8H11N02 153.0790, found 153.0794.

2,6-Dimethoxyiodobenzene. To a stirred solution of n-BuLi (48.0
mL, 0.12 mol, 2.5 M in hexanes) in dry THF (80 mL) at —5 °C under argon
was added 1,3-dimethoxybenzene (13.8 g, 0.10 mol) in one portion. The
mixture was allowed to warm to room temperature over 1 h, and was then
cooled to —70 °C. A solution of 12 (30.5 g, 0.12 mol) in dry THF (70 mL)
was added dropwise with vigorous stirring and, in the latter part of the
addition, the temperature was allowed to rise to —20 °C. When the addition
was complete, the mixture was treated with Na2S203 (10.0 g) in H20 (250
mL). The aqueous layer was separated and extracted with Et20 (3 x 100
mL). The organic extracts were combined, dried over Na2804, and
evaporated to give a yellow/white solid. Recrystallization from 95% EtOH
afforded the iodide (22.0 g, 84%) as white crystals: mp 103—104 °C (lit. 162
mp 102-103 °C); 1H NMR (300 MHz, CDC13) 5 7.24 (t, 1 H, J = 8.3 Hz),
6.49 (d, 2 H, J = 8.3 Hz), 3.87 (s, 6 H); 13C{1H} NMR (75.5 MHz, CDC13)
5 159.5, 129.8, 104.1, 77.6, 56.5; EIMS m/z (relative intensity) 264 (M+,
100), 249 (33), 234 (10), 221 (68), 206 (22), 132 (15), 122 (32); FABHRMS
calcd for C8H9102 263.9648, found 263.9649.

Tris(2,6-dimethoxyphenyl)amine (39). A mixture of 2,6-
dimethoxyaniline (0.51 g, 3.33 mmol), 2,6-dimethoxyiodobenzene (1.82 g,
6.89 mmol), anhydrous K2CO3 (3.65 g, 26.4 mmol), 18-crown-6 (0.18 g,
0.70 mmol), Cu powder (0.85 g, 13.3 mmol), and 1,2-dichlorobenzene

109

(6 mL) was refluxed under argon for 16 h. The cooled mixture was
chromatographed over silica gel using hexanes/EtOAc (5: 1, v/v), yielding an
orange/white solid (Rf: 0.15). Trituration of the colored product with
acetone afforded 39 (124 mg, 9%) as a white solid: mp 190-191 °C; IR (film
from CHC13) 3000, 2929, 2834, 1582, 1492, 1472, 1430, 1295, 1250, 1165,
1108 em-l; 1H NMR (300 MHz, CDC13) 5 6.87 (t, 3 H, J = 8.0 Hz), 6.50 (d,
6 H, J = 8.2 Hz), 3.43 (s, 18 H); 13C{1H} NMR (75.5 MHz, CDC13) 5
155.4, 129.8, 122.1, 107.3, 56.7; EIMS m/z (relative intensity) 426 (M+l,
29), 425 (M+, 100), 379 (48), 333 (7), 213 (13); FABHRMS calcd for
C24H27NO6 425.1839, found 425.1854.

Colorless needles of 39 suitable for X-ray analysis were obtained after

two recrystallizations from acetone at —20 °C.

Cyclized amine 40 was obtained by the procedure described for 39,
except that the mixture was refluxed for 72 h. Chromatography over silica
gel using hexanes/EtOAc (4: 1, v/v) gave 40 (5%) as a beige solid (Rf:
0.05): 1H NMR (300 MHz, CDC13) 5 7.11 (t, 1 H, J = 8.4 Hz), 6.76 (t, 2 H,
J = 8.2 Hz), 6.52 (d, 2 H, J = 8.3 Hz), 6.44 (dd, 2 H, J = 8.2, 1.3 Hz), 6.31
(dd, 2 H, J = 8.3, 1.3 Hz), 3.79 (s, 6 H), 3.51 (s, 6 H); 13C{1H} NMR (75.5
MHz, CDC13) 5 159.8, 152.0, 150.1, 127.3, 126.0, 124.8, 123.1, 107.9,
106.5, 104.4, 56.0, 55.6; EIMS m/z (relative intensity) 380 (M+1, 27), 379
(M+, 100), 333 (22), 287 (13), 226 (13), 189 (10); FABHRMS calcd for
C22H21N05 379.1420, found 379.1406.

l,2-Bis(2-nitrophenoxy)ethane. A solution of 2-nitrophenol (55.6 g,
0.40 mol) in dry DMF (200 mL) was added dropwise over 45 min to a
mechanically stirred suspension of NaH (9.6 g, 0.40 mol) in DMF (200 mL)

110

under argon. The orange-red mixture was stirred for 4 h, and then 1,2-
dichloroethane (15.8 mL, 0.20 mol) was added in one portion. The mixture
was refluxed for 10 h and the cooled mixture poured into ice-cold H20 (2
L). The precipitate was collected and washed successively with 2 M NaOH
(3 x 100 mL), H20 (5 x 200 mL), 95% EtOH (5 x 200 mL), and EtZO (3 x
200 mL). The dinitro compound (30.44 g, 50%) was obtained as an off-
white solid: mp 167—170 °c (lit.163 mp 165—168 °C); 1H NMR (CDC13) 5
7.81 (dd, 2 H, J = 8.1, 1.7 Hz), 7.58—7.52 (m, 2 H), 7.22 (dd, 2 H, J = 8.4,
1.2 Hz), 7.09—7.04 (m, 2 H), 4.52 (s, 4 H); ”mm NMR (CDC13) 5 157.0,
151.9, 134.3, 125.6, 121.4, 115.9, 68.7; EIMS m/z (relative intensity) 305
(M+1, 13), 304 (M+, 69), 167 (9), 166 (90), 123 (18), 122 (100).

l,2-Bis(2-aminophenoxy)ethane. Zinc dust (328 g, 5.0 mol) was
added to a mechanically stirred suspension of 1,2-bis(2-nitrophenoxy)ethane
(26.1 g, 85.6 mmol) in 78% EtOH (1.2 L), followed by a solution of CaC12
(12.0 g) in H20 (20 mL). The mixture was refluxed for 6 h and then filtered
through a coarse sintered glass funnel packed with Celite (3-cm) that was
topped with a layer of glass wool. The metal sludge was washed with boiling
78% EtOH (200 mL), and the cooled filtrate was poured into H20 (4 L). The
product was collected and washed with H20 (2 x 100 mL) to give the
desired ethane (18.8 g, 90%) as a pearly-white solid: mp 130—132 °C (lit.163
mp 127—130 °C); 1H NMR (CDC13) 5 6.87—6.68 (m, 8 H), 4.35 (s, 4 H),
3.82 (s, broad, 4 H); 13C{1H} NMR (CDC13) 5 146.2, 136.8, 121.9, 118.3,
115.3, 112.5, 67.4; EIMS m/z (relative intensity) 245 (M+1, 20), 244 (M"',
100), 136 (50), 135 (86), 120 (26), 109 (44), 108 (26).

111

1,2-Bis[{2-bis(2-methoxyphenyl)aminophenoxy}]ethane (45). A
mixture of 1,2-bis(2-aminophenoxy)ethane (2.01 g, 8.22 mol), 2-
iodoanisole (9.64 g, 41.2 mmol), l8-crown-6 (0.91 g, 3.4 mmol), anhydrous
K2C03 (18.3 g, 132 mmol), Cu powder (4.20 g, 66.1 mmol), and 1,2-
dichlorobenzene (80 mL) was refluxed under argon for 16 h. After cooling,
the mixture was filtered through a thin layer (~2 cm) of silica gel. The
inorganic solids were washed with hot CH2C12 and the combined filtrates
distilled under reduced pressure. After column chromatography (neutral
alumina, hexanes/CH2C12 (3:1)), the product was recrystallized from
acetone and dried at ~60 °C under high vacuum for 24 h, affording 45 (2.24
g, 41%) as an off-white solid: mp 137—137.5 °C; IR (film from CHCl3)
3061, 2934, 2834, 1588, 1497, 1455, 1319, 1267, 1248, 1181, 1119, 1048,
1028 em-l; 1H NMR (CDC13) 5 7.08—6.63 (m, 24 H), 3.51 (s, 12 H), 3.38 (s,
4 H); ”mm NMR(CDC13)8153.2, 151.6, 137.9, 137.6, 124.8, 124.0,
123.8, 123.1, 120.9, 120.8, 114.3, 112.9, 65.7, 55.8; EIMS m/z (relative
intensity) 669 (M+1, 16), 668 (M+, 46), 348 (15), 334 (31), 321 (10), 290
(17), 289 (100), 274 (10), 273 (12), 246 (18), 226 (11), 212 (11), 183 (21),
182 (35); FABHRMS calcd for C42H40N206 668.2888, found 668.2869.
Anal. Calcd for C42H40N206: C, 75.43; H, 6.03; N, 4.19. Found: C, 75.21;
H, 6.34; N, 3.96.

Crystals of 45 suitable for X-ray analysis were obtained by treating a
suspension of 45 in nitromethane with Lil (2 equiv). The ligand was
solubilized after shaking the mixture for ~10 min. Slow evaporation of the

solution at room temperature afforded free 45 as colorless crystals.

Tris(2-methoxyphenyl)methanol (48). A solution of 2-bromoanisole
(23.1 g, 0.124 mol) in dry EtZO (40 mL) was added dropwise to Mg pieces

112

(3.04 g, 0.125 mol) in dry EtZO (40 mL) under argon. When the addition
was complete (2 h), the mixture was stirred for an additional 30 min, and
then a solution of methyl 2-methoxybenzoate (9.31 g, 0.056 mol) in dry
benzene (120 mL) was added dropwise over 1 h. The mixture was refluxed
for 22 h, cooled, and quenched with saturated aqueous NH4C1 (400 mL).
Benzene (500 mL) was added to dissolve the precipitate that had formed,
and the aqueous layer was separated and extracted with benzene (3 x 150
mL). The organic layers were combined, dried over N a2SO4, filtered, and
evaporated to give a yellow/white solid. The crude product was
recrystallized from hexanes/benzene (1: 1) to give carbinol 48 (12.69 g, 65%)
as a white crystalline solid: mp 181—182 °C (lit.108 mp 1815—1825 °C); IR
3532, 3071, 2996, 2938, 2836, 1597, 1582, 1489, 1462, 1435, 1285, 1244,
1184, 1157, 1026 em-l; 1H NMR (300 MHz, benzene-d6) 5 7.64 (dd, 3 H,

J = 7.9, 1.8 Hz), 7.13-7.07 (m, 3 H), 6.91—6.85 (m, 3 H), 6.56 (d, 3 H, J =
7.3 Hz), 5.56 (s, 1 H), 3.02 (s, 9 H); 13C{1H} NMR (300 MHz, benzene-d6) '
5 158.0, 134.8, 130.5, 127.9, 120.3, 112.4, 80.7, 55.1; EIMS m/z (relative
intensity) 351 (M+1, 4), 350 (M+, 16), 244 (8), 243 (47), 215 (14), 136 (18),
135 (100), 121 (16); Anal. Calcd for C22H2204: C, 75.41; H, 6.33; O, 18.26.
Found: C, 75.18; H, 6.36; O, 17.51.

Tris(2-methoxyphenyl)methyl chloride (49). A solution of 48 (250
mg, 0.71 mmol) in dry CH2C12 (2 mL) in a Schlenk flask was treated with
SOC12 (0.5 mL, 6.85 mmol). The resulting deep purple solution was stirred
for 16 h while protected by a CaClz drying-tube. The mixture was
evaporated under reduced pressure, and the resulting dark purple residue

washed with small portions of dry EtZO. The product was obtained as a

113

mixture consisting of a 31:1 ratio of 49 and 36 (see Figure 4.2); the product
mixture was immediately used for the preparation of radical 47.

The NMR data for pure 49 were obtained by shaking a solution of 48
(25 mg, 0.071 mmol) in dry benzene—d6 (0.8 mL) with SOC12 (0.2 mL, 2.7
mmol) for 30 min in an NMR tube (see Figure 4.1):

49: 1H NMR (300 MHz, benzene-d6) 8 7.54 (d, broad, 3 H, J = 7.1
Hz), 7.11—7.05 (m, 3 H), 6.80—6.75 (m, 3 H), 6.56 (dd, 3 H, J = 8.2, 1.1 Hz),
3.12 (s, 9 H); 13C{ 1H} NMR (75.5 MHz, benzene-d6) 8 158.1, 132.2, 131.4,
129.2, 119.7, 113.0, 55.2 (oz-carbon resonance not observed).

However, evaporation of the volatiles resulted in hydrolysis to give

variable amounts of carbinol 48.

1,2-Bis[(2-carbethoxy)phenoxy]ethane. A solution of N aOEt was
prepared by adding Na metal pieces (13.9 g, 0.60 mol, ~1-cm cubes) to
anhydrous EtOH (350 mL) under argon. To the mechanically stirred solution
was then added ethyl salicylate (99.8 g, 0.60 mol) in anhydrous EtOH (150
mL) dropwise over 1 h. The mixture (now containing a white precipitate)
was stirred for an additional 30 min, and then 1,2-dichloroethane (29.64 g,
0.30 mol) was added in one portion. The mixture was refluxed for 18 h,
cooled, poured into H20 (1.2 L), and allowed to stand overnight. The
product was collected and recrystallized from 70% EtOH, affording the
diester (13.69 g, 13%) as colorless plates: mp 96.5—97 °C; IR 2981, 1723,
1599, 1590, 1495, 1446, 1368, 1291, 1251, 1167, 1143, 1083, 1058, 1019
cm-1; 1H NMR (300 MHz, CDC13) 5 7.76 (dd, 2 H, J = 7.7, 1.7 Hz), 7.46—
7.41 (m, 2 H), 7.07-6.96 (m, 4 H), 4.43 (s, 4 H), 4.29 (q, 4 H, J = 7.1 Hz),
1.29 (t, 6 H, J = 7.1 Hz); 13C{1H} NMR (75.5 MHz, CDC13) 5 166.3, 158.1,
133.3, 131.5, 121.4, 120.9, 114.3, 67.9, 60.8, 14.2; EIMS m/z (relative

114

intensity) 359 (M+1, <0.1), 358 (M"’, <1), 313 (6), 193 (25), 192 (9), 166
(37), 165 (12), 151 (5), 147 (19), 121 (100), 120 (54), 119 (8), 93 (9), 92
(13); FABHRMS calcd for C20H2206 358.1417, found 358.1422. Anal.
Calcd for C20H2206: C, 75.63; H, 6.06. Found: C, 75.34; H, 6.21.

1,2-Bis[{2-bis(2-methoxyphenyl)hydroxymethylphenoxy}]ethane
(50). A solution of 2-bromoanisole (11.5 g, 61.6 mmol) in dry EtzO (20 mL)
was added dropwise to Mg pieces (1.52 g, 62.4 mmol) in dry EtZO (20 mL)
under argon. When the addition was complete (2 h), the mixture was stirred
for an additional 30 min, and then a solution of 1,2—bis[(2-—
carbethoxy)phenoxy]ethane (5.00 g, 14.0 mmol) in dry benzene (60 mL)
was added dropwise over 45 min. The mixture was refluxed for 46 h, cooled,
and quenched with saturated aqueous NH4C1 (200 mL). The aqueous layer
was separated and extracted with benzene (2 x 100 mL). The organic layers
were combined, dried over Na2SO4, filtered, and evaporated to give a
viscous yellow/white residue. Slow evaporation of a CHC13 solution of the
crude product afforded 50 (6.38 g, 65%) as a white crystalline solid: mp
200—205 °C; IR 3503, 2936, 1584, 1487, 1462, 1287, 1238, 1181, 1113,
1024 em-l; 1H NMR (300 MHz, benzene-d6) 5 7.60 (dd, 6 H, J = 7.8, 1.7
Hz), 7.13—7.04 (m, 6 H), 6.89—6.81 (m, 6 H), 6.60—6.52 (m, 6 H), 6.15 (s,
CHC13), 5.52 (s, broad, 2 H), 3.45 (s, broad, 4 H), 3.00 (s, 12 H); 13C{1H}
NMR (75.5 MHz, benzene-d6) 8 158.0, 157.0, 135.3, 134.8, 130.6, 130.4,
128.1, 127.9, 120.5, 120.4, 113.6, 112.3, 80.3, 66.6, 55.0; EIMS m/z
(relative intensity) 681 (4), 680 (M—HZO, 8), 573 (5), 466 (3), 361 (5), 345
(9), 334 (10), 303 (20), 288 (17), 287 (78), 271 (21), 255 (11), 241 (17), 237
(13), 227 (17), 225 (16), 213 (21), 197 (16), 181 (14), 137 (14), 136 (19),

115

135 (100), 121 (93), 108 (11), 107 (52), 105 (17); Anal. Calcd for
C44H4208: C, 75.63; H, 6.06; O, 18.31. Found: C, 75.34; H, 6.21; O, 17.70.

l,2-Bis[{2-bis(2-methoxyphenyl)chloromethylphenoxy}]ethane
(51). A solution of di0150 (60.8 mg, 0.087 mmol) in dry benzene (3.2 mL)
in a Schlenk flask was treated with SOC12 (3211L, 0.44 mmol), and the
mixture stirred in the closed vessel for 24 h. The yellow supernatant was
removed via syringe from the white precipitate that had formed, and the
solid was washed successively with small portions of benzene/SOCIZ (100:1,
v/v) and dry benzene. The product was obtained as a mixture consisting of
51 (>95 %) and incompletely reacted 50 (<5 %) (see Figure 4.4); the product
mixture was immediately used for the generation of biradical 46.

The NMR data for pure 51 were obtained by shaking a solution of 50
(25 mg, 0.071 mmol) in dry benzene-d6 (0.8 mL) with SOC12 (0.2 mL, 2.7
mmol) for 30 min in an NMR tube (see Figure 4.3):

51: 1H N MR (300 MHz, benzene-d6) 8 7.49 (apparent d, broad, 6 H,
J = 7.7 Hz), 7.09—7.01 (m, 6 H), 6.78—6.69 (m, 6 H), 6.57—6.51 (m, 6 H),
6.32 (s, CHC13), 3.48 (s, broad, 4 H), 3.12 (s, broad, 12 H); 13C{1H} NMR
(75.5 MHz, benzene-d6) 8 158.1, 156.9, 132.3, 132.1, 131.6, 131.3, 129.4,
129.3, 127.9, 119.8, 114.0, 113.0, 66.5, 55.2 (oz-carbon resonance not
observed).

However, evaporation of the volatiles resulted in hydrolysis to give

variable amounts of carbinol 50.

l,2-Bis[{2-bis(2-methoxyphenyl)methanophenoxy}]ethane (52). A
suspension of di0150 (1.00 g, 1.43 mmol) in dry Et20 (10 mL) containing a
few granules of anhydrous CaC12 was saturated with anhydrous HCl gas.

116

The deep purple mixture was stoppered and allowed to stand overnight. The
supernatant (now brownish-orange in color) was removed via syinge under
argon, and the remaining solid residue was washed with dry EtZO. The
combined EtzO layers were stored under argon and, after ca. 8 h, an orange-
red solid precipitated. The product was filtered and washed with EtzO to
give 52 (7.2 mg, 0.8%) as a pale orange solid; 1H NMR (300 MHz, CDC13)
8 7.20—7.07 (m, 6 H), 6.90—6.65 (m, 18 H), 6.44 (m, 2 H), 3.71 (s, 4 H), 3.59
(s, 12 H); 13C{1H} NMR (75.5 MHz, CDC13) 5 157.4, 156.4, 133.7, 132.3,
129.9, 129.2, 127.1, 126.9, 120.3, 119.9, 112.4, 110.6, 66.6, 55.6, 37.1;
EIMS m/z (relative intensity) 668 (M+2, <1), 667 (M+1, <1), 666 (M+, 2),
560 (<1), 559 (3), 558 (7), 441 (<1), 440 (3), 439 (11), 288 (18), 287 (75),
271 (10), 227 (11), 213 (16), 181 (32), 135 (44), 121 (100), 107 (53);
FABHRMS calcd for C44H4206 666.2983, found 666.3026.

Generation of Tris(Z-methoxyphenyl)methyl (47) and l,2-Bis[{2-bis(2-
methoxyphenyl)methylphenoxy}]ethane (46)

A Schlenk flask containing either 49 (~0.26 mmol) or 51 (~0. 13
mmol) under vacuum was transferred to a nitrogen-filled dry-box. Silver
powder (470 mg, 4.36 mmol, Alfa, 99.95%, -100 mesh) was added, and the
flask transferred to a Schlenk line and evacuated. The flask was back-filled
with argon and evacuated again. This was repeated three times, and then a
volume (25 mL) of an appropriate solvent (2-MeTHF, benzene, toluene, or
CHC13) was added under argon. Stirring the mixture under argon for several
minutes produced a brilliant, yellow-orange solution of 47 or 46. Typically
the mixture was stirred for 2 h before samples were prepared for ESR

studies.

117

Semi-Empirical Calculations

Geometry optimizations were carried out using standard AMI120 and
MNDO164 procedures as implemented in the SPARTAN computer program
(SPARTAN version 3.1, Copyright 1994 Wavefunction, Inc.).

Stoichiometry Determinations for Lil Complexes

For the Lil complexes, stoichiometries were determined using 1H and
7Li NMR. As an example, the procedure is described below for the 28'LiI
complex:

A tube containing a reference sample of LiBPh4-3glyme in CDC13
was coaxially mounted in a 5-mm NMR tube containing 28 and excess LiI
(ca. 5 equiv) in CDC13. Lil is insoluble in CDC13 (by 7Li NMR) in the
absence of added ligand. The 28°xLiI:LiBPh4-3glyme ratio was determined
from the respective OCH3 1H integrals, and a LiI:LiBPh4 ratio was obtained
from the respective 7Li integrals. The stoichiometry (x) is the ratio of the 7Li
to 1H ratios. The 38-LiI and 45-xLiI stoichiometries were similarly

determined.
NMR Titration Experiments

Samples were prepared at 28/M+ mole ratios from 0 to 9 by mixing
appropriate volumes (11L syringe) of 28/CD3N02 and M+/CD3N02 stock
solutions in an NMR tube, and diluting with CD3N02 to the desired final

volume (0.7 mL); the metal ion concentration was held constant at

118

5.0 x 10‘3 M throughout. All samples were prepared in a dry-box under a N2
atmosphere, and the NMR spectra were obtained immediately after sample
preparation. Formation constants (Kf’s) were obtained from the variation of
chemical shifts with the concentration of 28 by means of a nonlinear least-
squares curve-fitting program (K1NF1T165). The 1H and 7Li/23Na shifts were
used in multiple-data-set fits to equations that assumed (1) formation of only
a 1:1 complex, and (2) formation of both 1:1 and 2:1 ligand/M+ complexes.

The equilibria are given below:

K 1 + K2
A-M

 

 

A2'M+

 

A+M+

where A is the amine ligand 28, M+ is the appropriate alkali metal ion (Li+
or Na+), A-M+ is the 1:1 complex between the amine and the metal ion,
Az-M+ is the 2:1 complex between the amine and the metal ion, and K1 and

K2 are the formation constants for the 1:1 and 2:1 complexes, respectively.
The equations describing case (1) are as follows:

80bS(M+) = {(K1[A]T + K1[M+]T + 1) — (K12[A]T2 + K12[M+]T2
— 2K12[A]T[M+]T + 2K1[A]T + 2K1[M+]T +1)1/2 / 2K1[M+]T}
{81.1(M+) — 5f(M+)} + 5gM+)

8obs(1H) = {(K1[A]T + K1[M+]T + 1) — (K12[A]T2 + K12[M+]T2
— 2K12[A]T[M+]T + 2K1[A]T + 2K1[M+]T +1)1/2/2K1[A]T}
{51.1(1H) — 8t(1H)} + 5f(1H)

119

where 8ObS(M+) is the observed alkali metal chemical shift, [A]T is the total
amine concentration, [M+]T is the total metal ion concentration, 81.1(M+) is
the alkali metal chemical shift for the 1:1 (A-M+) complex, 8f(M+) is the
alkali metal chemical shift for the free (solvated) metal ion, 81.1(1H) is the
OCH 3 proton chemical shift for the 1:1 (A0M+) complex, 8f(1H) is the
OCH 3 proton chemical shift for the free arrrine ligand, and K1 and K2 are as

defined above.
The equations describing case (2) are as follows:

50bS(M+) = {5,(M+) / [1 + K1[A]T + K1K2[A]T2]}
+{81.1(M+)/[1+(1/K1[A]T)+K2[A]T]}
+ {52.1(M+) / [1 + (1 IK2[A]T) + (1 /K1K2[A]T2)]}

801.3(1H) = {81;1(1H) — 5f(1H)} {[M+]T/ [A]T(1 +(1/K1[A]T)
+ KzlAlT)} + {52;1(1H) — 5f(lH)}
{2[M+]T / [A]T(1 + (1 IK2[A]T) + (1 /K1K2[A]T2))} +5f(1H)

where 82.1(M+) and 82.1(1H) are the alkali metal chemical shift and the
OCH 3 proton chemical shift, respectively, for the 2:1 (AZ-M1“) complex; the

remaining parameters are as defined above.

A visual comparison of the two curve-fits was used to determine the
complex stoichiometry and, in each case, only the curve-fits obtained using
the equations given for case (1) showed good agreement with the plots
obtained from the experimental data. The results are summarized in Table
2.1.

120

NMR Data for Metal Complexes

Typically, NMR samples were prepared by adding a CDC13 solution
of the ligand (0.03 M) to the appropriate alkali metal salt. All of the salts
used are insoluble in CDC13 (by NMR) in the absence of added ligand. After
shaking for ~10 nrin, the NMR spectra were recorded. Complexes of NaB (4-
ClPh)4 were prepared by treating CDC13 solutions of the respective NaBPh4
complexes with KB(4-C1Ph)4. After shaking for 10 min, the KBPh4
precipitate was removed by filtration. The 13C chemical shift assignments
for the BAr4' counterions are primarily based on known166 11B—13C

coupling constants.

28-Li1: 1H NMR 5 7.23-7.16 (m, 6 H), 70569 (no, 6 H), 4.02 (s,
9 H); 13C{1H} NMR 5 152.9, 135.7, 127.3, 126.9, 121.9, 112.1, 58.3; 7Li
NMR 8 2.11(AV1/2 =1.0 HZ).

28-LiBPh4: 1H NMR 5 7.62—7.51 (m, 8 H), 7.24—7.17 (m, 4 H),
7.11-6.85 (m, 20 H), 3.45 (s, 9 H); 13C{1H} NMR 5 164.0 (q, 1JBC = 49.3
Hz), 152.5, 135.9 (213C unresolved), 135.5, 127.7, 127.2, 126.1 (313C = 2.9
Hz) , 122.7, 122.2 (4JBC unresolved), 112.4, 56.7; 7Li NMR 5 —0.08
(Av1 ,2 = 41 Hz).

282-NaBPh4: 1H NMR 5 7.44—7.36 (m, 8 H), 7.19—7.11 (m, 6 H),
7.04—6.97 (m, 8 H), 6.94—6.85 (m, 16 H), 6.71 (dd, 6 H, J = 8.0, 1.0 Hz),
3.14 (s, 18 H); 13C{1H} NMR 5 163.9 (q, 1JBC = 49.3 Hz), 151.9, 137.1,
136.0 (ZJBC unresolved), 126.1, 125.6 (3JBC unresolved), 124.8, 122.3,
121.8 (413C unresolved), 113.4, 56.3; 23Na NMR 5 —-11.9 (At/1,2 = 240 Hz).

121

282-NaB(4-C1Ph)4: 1H NMR 5 7.23-7.08 (m, 14 H), 6.96 (d, 8 H,
J = 8.4 Hz), 6.91-6.83 (m, 12 H), 6.72 (dd, 6 H, J: 8.1, 1.8 Hz), 3.24 (s, 18
H); 13C{1H} NMR 5 160.9 (q, 1JBC = 49.7 Hz), 152.1, 137.3, 137.2 (2JBC
unresolved), 127.9, 125.6 (3JBC unresolved), 124.8, 122.1 (4JBC
unresolved), 113.3, 56.2 (remaining 13C resonance not observed); 23Na
NMR 8 —10.3 (AV1/2 = 200 Hz).

34-Lil: 1H NMR 5 7.31—7.17 (m, 6 H), 7.07—6.99 (m, 5 H), 6.73 (d,
2 H, J = 8.0 Hz), 4.01 (s, 6 H); 13C{1H} NMR 5 153.5, 147.2, 134.9, 129.5,
128.8, 128.3, 123.7, 123.1, 119.9, 112.5, 57.9; 7Li NMR 5 2.74(Av1,2 = 2.8
Hz).

34-LiBPh4: 1H NMR 5 7.54-7.44 (m, broad, 8 H), 7.3 1—7. 12 (m,
6 H), 7.08—6.93 (m, 12 H), 6.89-6.80 (m, 5 H), 6.50—6.45 (m, 2 H), 3.44 (s,
6 H); 13C{1H} NMR 5 163.8 (q, 1:1:1:1, JBC = 49.2 Hz), 153.5, 147.0,
138.5, 135.9 (JBC unresolved), 129.2, 128.4, 128.0, 126.0 (JBC unresolved),
123.4, 122.3, 122.1 (JBC unresolved), 118.1, 113.5, 57.1; 7Li NMR 8 —2.43
(Av1/2 = 8.2 Hz).

38-LiI: 1H NMR 5 7.26 (t, 3 H, J = 7.8 Hz), 6.95—6.86 (m, 6 H), 3.49
(s, 9 H), 2.13 (s, 9 H); 13C{1H} NMR 5 158.3, 140.1, 136.7, 131.2, 125.6,

113.3, 60.6, 21.1; 7Li NMR 5 0.97 (Av1 ,2 = 1.6 Hz).

39-2Li1: 1H NMR 5 7.22 (t, 3 H, J = 8.0 Hz), 6.85 (d, 6 H, J = 8.2
Hz), 3.75 (s, 18 H); 13c and 7Li NMR data were not obtained.

45-2Li1: 1H NMR 5 7.29—6.83 (m, 24 H), 4.04 (s, 4 H), 4.01 (s,

122

12 H); 13C{1H} NMR 5152.9, 150.7, 136.7, 135.3, 127.7, 127.2, 127.1,
125.3, 122.6, 122.1, 114.6, 112.4, 64.9, 58.3; 7Li NMR 5 1.85 (Avl,2 = 2.0
Hz).

45-LiBPh4: 1H NMR 5 7.40—7.32 (m, 8 H), 7.21—7.15 (m, 4 H),
7.10—7.04 (m, 2 H), 6.96—6.77 (m, 28 H), 6.57 (dd, 2 H, J = 7.1, 1.1 Hz),
3.38 (s, 12 H), 3.21 (s, 4 H); 13C{1H} NMR 5 164.2 (q, 1JBC = 49.3 Hz),
152.4, 150.8,137.6, 136.3, 136.2 (2JBC unresolved), 126.5, 126.0, 125.6,
125.4 (3JBC = 2.4 Hz), 124.4, 122.9, 122.4, 121.5 (4JBC unresolved), 114.5,
113.1, 67.8, 56.2; 7Li NMR 5 0.15 (At/1,2 = 4.0 Hz).

45-NaBPh4: 1H NMR 5 7.40—7.31 (m, 8 H), 7.30—6.63 (m, 34 H),
6.48 (dd, 2 H, J = 8.0, 1.2 Hz), 3.23 (m, 2 H), 3.14 (s, 6 H), 2.95 (m, 2 H),
2.88 (s, 6 H); 13C(1H} NMR 5164.1 (q, 1JBC = 49.4 Hz), 153.1, 151.6,
149.5, 138.3, 136.3, 136.1 (2JBC unresolved), 134.3, 128.9, 127.9, 126.7,
126.3, 125.7, 125.4 (3JBC = 2.8 Hz), 122.5, 122.2, 122.0, 121.9, 121.5 (4JBC
unresolved), 112.0, 111.8, 111.7, 65.1, 55.4, 55.2; 23Na NMR 5 —5.5
(Av1 ,2 = 260 Hz).

45-NaB(4-C1Ph)4: 1H NMR 5 7.24-7.07 (m, 14 H), 6.97—6.79 (m, 22
H), 6.70 (dd, 2 H, J = 7.8, 1.5 Hz), 6.61 (dd, 2 H, J = 8.0, 1.2 Hz), 3.47 (s,
br, 2 H), 3.39 (s, br, 2 H), 3.23 (s, 6 H), 2.95 (s, 6 H); 13C{1H} NMR 5
160.8 (q, 1JBC = 49.8 Hz), 153.0, 151.6, 149.4, 138.4, 137.2 (213C
unresolved), 136.3, 134.2, 128.8, 128.0, 127.9, 1269,1265, 125.8, 125.7
(31BC = 2.7 Hz), 122.7 (413C unresolved), 122.3, 122.2, 111.9, 111.8, 111.7,
65.0, 55.5, 55.3 (remaining l3C resonance not observed); 23Na NMR 8 —5.3
(Ax/1,2 = 240 Hz).

123

45-KB(4-ClPh)4: 1H NMR 5 7.23—7.05 (m, 14 H), 7.00—6.68 (m, 24
H), 6.58 (dd, 2 H, J = 8.0, 1.2 Hz), 3.40 (s, 6 H), 3.13 (s, 6 H), 3.23 (s, 4 H);
13C{1H} NMR 5 160.9 (q, 1JBC = 49.6 Hz), 151.7, 149.4, 137.7, 137.2
(2JBC unresolved), 136.7, 136.2, 127.9, 126.5, 126.2, 126.0, 125.7
(313C = 2.7 Hz), 125.5, 125.2, 123.6, 123.3, 123.0, 122.7 (4JBC unresolved),
114.7, 114.5, 112.5, 65.7, 56.6, 56.5 (remaining 13C resonance not

observed).
Saturation Spin Transfer (SST) Measurements

The SST technique132 was used to obtain rate constants for methoxy
group site-exchange (kex) in 45-NaBPh4, 45-NaB(4-C1Ph)4, and 45-KB(4-
ClPh) 4. Experiments were carried out on 0.03 M CDC13 solutions of the
complexes at 293 K by delivering a selective 180° pulse on the higher
frequency methoxy proton signal, followed by a nonselective 90° pulse after
increasingly longer delay times. Treatment of the change of intensities132 of
the methoxy signals as a function of time, using the KaleidagraphTM program
(version 2.1.4, Copyright 1992 Abelbeck Software), afforded exchange rate
constants; the reported standard deviations are those given by the curve-
fitting program. The AG1t values were calculated from the Eyring equation

assuming a transmission coefficient of unity.
Preparation of Crystalline Complexes

282°NaBPh4. A suspension of 28 (107.0 mg, 0.32 mmol) in
nitromethane (5 mL) was treated with N aBPh4 (55.5 mg, 0.16 mmol). A

clear solution resulted after shaking the mixture for ~2 min. Slow

124

evaporation of this solution at room temperature afforded 282°NaBPh4
(101.6 mg, 63%) as colorless crystals: mp 193—194.5 °C (dec); 1H NMR
(CDCl3) 8 7.44—7.36 (m, 8 H), 7 .19—7.11 (m, 6 H), 7.04—6.97 (m, 8 H),
6.94-6.85 (m, 16 H), 6.71 (dd, 6 H, J = 8.0, 1.5 Hz), 3.24 (s, 18 H); FABMS
m/z 693 (M+). Anal. Calcd for C66H62BN2NaO6: C, 78.25; H, 6.17; N,
2.76; Na, 2.27. Found: C, 78.47; H, 6.23; N, 2.66; Na, 2.12.

450NaBPh4. A solution of 45 (100 mg, 0.15 mmol) in CHCl3 (5 mL)
was treated with NaBPh4 (513 mg, 1.5 mmol). After shaking for 10 min, the
mixture was filtered to remove excess NaBPh4. Evaporation of the filtrate
afforded a viscous oil that solidified to white microcystals on trituration with
nitromethane. Recrystallization from nitromethane gave 45-NaBPh4 (64 mg,
42%) as colorless cubes that were suitable for X-ray analysis: mp 231—232
°C; 1H NMR (CDC13) 5 7.40—7.31 (m, 8 H), 7.30—6.63 (m, 34 H), 6.48 (dd,
2 H, J = 8.0, 1.2 Hz), 3.23 (m, 2 H), 3.14 (s, 6 H), 2.95 (m, 2 H), 2.88 (s, 6
H). Anal. Calcd for C66HGOBN2NaO6: C, 78.41; H, 5.98; N, 2.77; Na, 2.27.
Found: C, 78.07; H, 6.33; N, 2.46; Na, 2.31.

45-NaB(4-C1Ph)4. A suspension of 45 (32.1 mg, 0.05 mmol) in
nitromethane (1 mL) was treated with a solution of NaBPh4 (16.4 mg, 0.05
mmol) in nitromethane (1 mL). After shaking for 10 min, the mixture was
filtered through a plug of glass wool into a solution of KB(4-C1Ph)4 (23.7
mg, 0.05 mmol) in nitromethane (0.5 mL). The mixture was shaken for 10
rrrin, during which time a fine, white solid (KBPh4) precipitated. The
resulting suspension was centrifuged, and the supernatant was carefully
removed and slowly evaporated at room temperature; 45°NaB (4-C1Ph)4

(26.3 mg, 46%) was obtained as colorless cubes that were suitable for X-ray

125

analysis: mp 179—181 °C; 1H NMR (CDC13) 5 7.24—7.07 (m, 14 H), 6.97—
6.79 (m, 22 H), 6.70 (dd, 2 H, J = 7.8, 1.5 Hz), 6.61 (dd, 2 H, J = 8.0, 1.2
Hz), 3.47 (s, broad, 2 H), 3.39 (s, broad, 2 H), 3.23 (s, 6 H), 2.95 (s, 6 H).
Anal. Calcd for C66H56BC14N2NaO6: C, 69.01; H, 4.91; N, 2.44; Na, 2.00.
Found: C, 68.33; H, 4.83; N, 2.33; Na, 2.24.

45°KB(4-ClPh)40CH3N02. A solution of 45 (200 mg, 0.30 mmol) in
CHC13 (10 mL) was treated with KB(4-C1Ph)4 (345 mg, 0.70 mmol). After
shaking for 10 min, the mixture was filtered to remove excess KB(4—C1Ph)4.
Evaporation of the filtrate afforded a viscous oil that solidified overnight to a
waxy residue. Recrystallization of this residue by slowly evaporating a
nitromethane solution (ca. 1.5 mL) at room temperature yielded 45°KB(4-
ClPh)4'CH3N02 (111 mg, 30%) as colorless crystals that lose CH3N02 on
standing: mp 177—178 °C; 1H NMR (CDC13) 5 7.23—7.05 (m, 14 H),
7.00-6.68 (m, 24 H), 6.58 (dd, 2 H, J = 8.0, 1.2 Hz), 4.31 (s, CH3N02), 3.40
(s, 6 H), 3.13 (s, 6 H), 3.23 (s, 4 H). Anal. Calcd for
C66H56BC14N2KO6-CH3N02: C, 65.64; H, 4.85; N, 3.43; K, 3.19. Anal.
Calcd for C66H56BC14N2KO6'0.5CH3N02: C, 66.82; H, 4.85; N, 2.93; K,
3.27. Found: C, 67.26; H, 4.93; N, 2.30; K, 3.52.

X-ray Crystallography

Crystal data for 39, 38, and 26°13 are given in Table 2.3, while crystal
data for 282'NaBPh4, 45°NaBPh4, 45-NaB(4-C1Ph)4, 45°KB(4-
ClPh)4°CH3N02, and 45 are presented in Table 3.7. Intensity data were
collected at room temperature on an Enraf-Nonius CAD4 diffractometer

with graphite monochromated Mo K01 radiation. The measured intensities

126

were corrected for Lorentz and polarization effects; in the case of 38, an
empirical absorption correction167 was applied. The structures were solved
by direct methods (SHELXS-86168) and refined by full-matrix least-squares
procedures using the MolEN169 package of programs. Non-hydrogen atoms
were refined anisotropically. Hydrogen atoms were located in the succeeding
difference Fourier syntheses and added to the structure factor calculations,
but their positions were not refined.

All of the crystal data were obtained and analyzed at Purdue
University in Professor B. E. Kahr’s group.

39: A colorless needle of C24H27NO6 having approximate dimensions
0.36 x 0.21 x 0.20 mm was mounted on a glass fiber. A total of 2602
reflections (+h, +k, it) were collected in the range 4° < 26 < 55° with 1203
having ID > 36(10) being used in the refinement (142 variables). Final
R = 0.057 and RW = 0.076.

38: A colorless needle of C24H27BO3 having approximate dimensions
0.30 x 0.34 x 0.36 mm was mounted on a glass fiber. A total of 3872
reflections (ih, —k, +l) were collected in the range 4° < 29 < 50° with 1223
having I0 > 36(10) being used in the refinement (253 variables). Final
R = 0.079 and RW = 0.093.

26°13: A colorless needle of C25H27I3O6 having approximate
dimensions 0.32 x 0.13 x 0.05 mm was mounted on a glass fiber. A total of
1886 reflections (ih, +k, +1) were collected in the range 4° < 20 < 45° with
1082 having Io > 36(10) being used in the refinement (156 variables). Final
R = 0.057 and RW = 0.067.

127

282'NaBPh4: A colorless cube of C66H62BN2NaO6 having
approximate dimensions 0.26 x 0.22 x 0.22 mm was mounted on a glass
fiber. A total of 9814 reflections (ih, +k, +l) were collected in the range
4° < 29 < 50° with 2577 having 10 > 30(10) being used in the refinement
(560 variables). Final R = 0.056 and RW = 0.057. The maximum and
rrrinimum peaks in the final difference Fourier map corresponded to 0.22 and

0.10 e/A3, respectively.

45-NaBPh4: A clear needle of C66H6OBN2NaO6 having approximate
dimensions 0.30 x 0.25 x 0.20 mm was mounted on a glass fiber. A total of
12544 reflections (ih, ik, +1) were collected in the range 4° < 29 < 55° with
4100 having I0 > 36(10) being used in the refinement (560 variables). Final
R = 0.108 and Rw = 0.132. The high R-values are a consequence of
unresolved disorder in the BPh4" counterion. Coordinates of the ligand were
quickly convergent and the atomic displacement parameters were ordinary.
The maximum and minimum peaks in the final difference Fourier map

corresponded to 1.10 and —0.09 e/A3, respectively.

45'NaB(4-ClPh)4: A colorless cube of C66H56BC14N2NaO6 having
approximate dimensions 0.60 x 0.50 x 0.20 mm was mounted on a glass
fiber. A total of 9381 reflections (ih, —k, —l) were collected in the range
5° < 29 < 47° with 3210 having I0 > 30(10) being used in the refinement
(586 variables). Final R = 0.051 and RW = 0.056. The maximum and
minimum peaks in the final difference Fourier map corresponded to 0.37 and

0.23 e/A3, respectively.

128

45¢KB(4-C1Ph)4-CH3N02: A colorless cube of C67H59BC14KN308
having approximate dimensions 0.30 x 0.30 x 0.10 mm was mounted on a
glass fiber. A total of 5182 reflections (ih, —k, +l) were collected in the
range 5° < 29 < 47° with 2054 having Io > 36(10) being used in the
refinement (430 variables). Final R = 0.055 and RW = 0.061. The maximum
and minimum peaks in the final difference Fourier map corresponded to 0.22

and —0.07 e/A3, respectively.

45: A colorless crystal of C42H40N206 having approximate
dimensions 0.41 x 0.23 x 0.14 mm was mounted on a glass fiber. A total of
8103 reflections (ih, ik, +l) were collected in the range 4° < 29 < 55° with
2068 having I0 > 26(10) being used in the refinement (451 variables). Final
R = 0.102 and RW = 0.101. The maximum and minimum peaks in the final

difference Fourier map corresponded to 0.43 and —0.29 e/A3, respectively.
ESR Studies

ESR spectra were obtained with a Varian E4 X-band spectrometer
equipped with a variable-temperature unit. The temperature was controlled
by passing N2 gas through cooling coils which were immersed in liquid
nitrogen. Quartz ESR tubes (Wilmad, 4-mm O. D.) were connected to a
Schlenk line via a Cajon Ultra-Torr® reducing union (3/8”—>1/4” O. D.). The
ESR tubes were modified with quartz->Pyrex graded seals (1/4” 0. D.) so
they could be attached to the Cajon fitting. The opposite end of the Cajon
adapter was connected to a glass “Y” assembly (3/8” 0. D.); one end of this
assembly was connected to the Schlenk line, while the other end was sealed

with a rubber septum to allow for solution transfers.

129

A solution of radical 47 or biradical 46, generated as described above,
was filtered through a glass frit assembly under anaerobic conditons to
remove Ag metal and AgCl. Aliquots of the colored solution were
transferred via cannula under argon into quartz ESR tubes containing solid
metal salts; the samples were degassed (three freeze-pump—thaw cycles) and
sealed under vacuum. For samples prepared using a CHCl3/acetone (7: 1,
v/v) solvent mixture, the salts were dissolved in acetone, added to the ESR
tubes, and degassed (five freeze—pump—thaw cycles); the solution of 47 or
46 was then transferred, the samples degassed (three freeze—pump—thaw

cycles) and sealed under vacuum.

The simulated ESR spectrum of monoradical 47 (Figure 4.5a) was
generated using the ESRa program written by A. K. Rappé and C. J. Casewit,
Calleo Scientific Software, Colorado State University.

Table 2.4A. Atomic Positional and Isotropic Thermal Parameters for 39

130

 

 

Atom x y z B(AZ)
012 0.4453(3) 0.4239(2) 0.0341(3) 4.94(7)
016 0.3270(3) 0.3281(2) 0.4192(3) 5.18(7)
O22 0.4028(3) 0.2764(2) 0.0246(3) 549(8)
N1 1/2 0.3433(2) 1/4 298(9)
C11 0.3855(3) 0.3780(2) 0.2283(4) 3.03(8)
C 12 0.3578(4) 0.4197(2) 0.1178(4) 3.70(9)
C13 0.2463(4) 0.4549(2) 0.0985(5) 4.6(1)
C14 0.1632(4) 0.4479(2) 0.1859(5) 4.8( 1)
C 15 0.1863(4) 0.4063(2) 0.2944(5) 4.6(1)
C16 0.2977(4) 0.3717(2) 0.3167(4) 3.69(9)
C17 0.4043(5) 0.4413(3) —0. 1018(5) 6.9(1)
C18 0.2688(6) 0.3345(3) 0.5346(5) 7.7(2)
C21 1/2 0.2743(3) 1/4 3.5(1)
C22 0.4500(4) 0.2399(2) 0.1338(5) 4.44(9)
C23 0.4492(5) 0.1724(2) 0.1352(6) 6.4( 1)
C24 1/2 0.1412(3) 1/4 8.0(2)
C25 0.3815(6) 0.2454(3) —0.1053(5) 7.4(1)

 

Table 2.5A. Atomic Positional and Isotropic Thermal Parameters for 38

131

 

 

Atom x y z B(AZ)
012 0.1825(6) 0.8160(3) 0.5443(4) 6.1(1)
026 0.1636(6) 0.7154(4) 0.1892(4) 7.7(2)
036 0.3029(7) 0.6174(4) 0.3648(5) 9.2(2)
C11 0.0694(8) 0.7152(4) 0.4487(5) 4.3(2)
C12 0.0932(8) 0.7452(5) 0.5386(5) 4.3(2)
C13 0.032(1) 0.7100(5) 0.6129(5) 5.9(2)
C14 —0.054( 1) 0.6403(6) 0.5980(6) 7.1(3)
C15 -0.0819(2) 0.6065(5) 0.5124(6) 6.4(2)
C16 —0.0255(8) 0.6451(1) 0.4358(8) 5.0(2)
C17 —0.0568(2) 0.6072(5) 0.3450(6) 6.0(2)
C18 0.222(1) 0.8501(6) 0.6309(6) 7.4(3)
C21 0.0512(8) 0.8070(5) 0.2888(5) 4.7(2)
C22 -0.0493(8) 0.8703(5) 0.3042(6) 5.3(2)
C23 —0. 1282(2) 0.9063(5) 0.2310(6) 6.9(2)
C24 —0. 1077(9) 0.8767(6) 0.1440(6) 8.2(3)
C25 —0.013(1) 0.8149(6) 0.1252(6) 7.8(3)
C26 0.0661(9) 0.7803(5) 0.1980(6) 5.7(2)
C27 —0.0676(9) 0.9047(6) 0.3989(7) 6.8(3)
C28 0.191(1) 0.6855(8) 0.1018(7) 9.7(3)
C31 0.3104(8) 0.7646(5) 0.3647(5) 5.4(2)
C32 0.3852(8) 0.8451(6) 0.3653(6) 6.4(2)
C33 0.5322(9) 0.8374(6) 0.3649(6) 6.6(2)
C34 0.6058(9) 0.7572(6) 0.3663(6) 7.1(3)
C35 0.5325(8) 0.6816(5) 0.3655(6) 5.7(2)
C36 0.3888(8) 0.6890(5) 0.3672(5) 4.9(2)
C37 0.320(1) 0.9275(7) 0.3668(7) 9.7(3)
C38 0.3651(1) 0.5406(7) 0.3646(9) l 1.5(4)

B 0.1462(9) 0.7609(5) 0.3684(6) 3.8(2)

 

Table 2.6A. Atomic Positional and Isotropic Thermal Parameters for 26°13

132

 

 

Atom x y z B(Az)
012 0.103(1) 0.9865(9) 0.3881(6) 3.2(2)
022 —0.2 10(1) 0.6709(8) 0.1801(6) 2.4(2)
026 0.301(1) 0.9148(9) 0.1718(5) 2.2(2)
C11 0.000 0.855(2) 1/4 1.7(4)
C12 0.000 0.981(2) 1/4 1.9(4)
C13 0.051(2) 1.174(2) 0.321(1) 2.3(3)
C14 0.000 1.232(2) 1/4 4.4(6)
C21 0.052(2) 0.786(1) 0.1827(7) 1.8(3)
C22 —0.047(2) 0.682(1) 0.1569(8) 1.9(3)
C23 0.023(2) 0.603(1) 0.1022(9) 2.8(3)
C24 0.186(2) 0.627(1) 0.0746(9) 3.4(4)
C25 0.2482(2) 0.730(1) 0.0985(8) 2.6(3)
C26 0.214(2) 0.812(1) 0.1493(8) 2.0(3)

C121 0.178(2) 1.051(2) 0.4604(9) 4.8(4)

C221 —0.309(2) 0.563(2) 0.160(1) 4.4(5)

C261 0.480(2) 0.934(1) 0.1494(9) 2.9(4)

11 1/2 0.2790(2) 1/4 3.74(3)
12 0.6223(2) 0.2826(1) 0.422730) 4.83(3)

 

133

Table 3.2A. Atomic Positional and Isotropic Thermal Parameters for 282°NaBPh4

 

 

Atom x y z B(Az)
Na 0.6284(3) 0.1 1796(7) 0.3148(2) 3.66(7)
01 0.8117(4) 0.0837(1) 0.2611(3) 3.5(1)
02 0.5673(4) 0.0700(1) 0.4340(3) 4.7( 1)
03 0.4255(4) 0.1035(1) 0.2059(4) 4.2( 1)
04 0.5206(4) 0.1667(1) 0.3975(3) 4.1(1)
05 0.6478(4) 0.1476(1) 0.1369(3) 3.8(1)
06 0.8025(4) 0.1364(1) 0.4463(3) 4.1(1)
N1 0.5928(5) 0.0497(1) 0.2535(4) 2.7(1)
N2 0.7144(5) 0.1845(1) 0.3048(4) 3.0(1)
Cl 0.6514(6) 0.0506(2) 0.1657(5) 2.9(2)
C2 0.5984(7) 0.0366(2) 0.0782(5) 4.0(2)
C3 0.6634(7) 0.0378(2) —0.0026(5) 4.8(2)
C4 0.7833(7) 0.0518(2) 0.0077(5) 4.6(2)
C5 0.8375(6) 0.0663(2) 0.0956(5) 3.8(2)
C6 0.7701(6) 0.0674(2) 0.1743(4) 2.7(2)
C7 0.6628(6) 0.0307(2) 0.3351(5) 3.2(2)
C8 0.6440(6) 0.0412(2) 0.4291(5) 3.6(2)
C9 0.7058(8) 0.0238(2) 0.5108(5) 5.2(2)

C01 0.9414(7) 0.0921(2) 0.2829(6) 5.5(2)
C02 0.5166(8) 0.0759(3) 0.5219(6) 7.5(3)
C03 0.3409(8) 0.1316(2) 0.1723(6) 5.5(2)
C04 0.4989(8) 0.1755(2) 0.4954(6) 5.8(2)
C05 0.5958(7) 0.1367(2) 0.0401(6) 5.0(2)
C06 0.8583(7) 0.1113(2) 0.5179(6) 5.1(2)
C10 0.7891(8) —0.0034(2) 0.4983(6) 6.0(2)
C11 0.8102(8) —0.0034(2) 0.4063(6) 5 .3(2)
C12 0.7461(7) 0.0035(2) 0.3247(5) 4.2(2)
C13 0.4592(6) 0.0427(2) 0.2400(5) 3.1(2)
C 14 0.3734(6) 0.0704(2) 0.2153(5) 3.6(2)
C15 0.2453(6) 0.0639(2) 0.2004(5) 4.3(2)
C16 0.2013(7) 0.0293(2) 0.2097(6) 4.9(2)
C17 0.2852(7) 0.0027(2) 0.2366(5) 4.1(2)

Table 3.2A. (Continued)

134

 

 

Atom x y z B(AZ)
C18 0.4137(7) 0.0085(2) 0.2512(5) 4.0(2)
C19 0.5937(6) 0.2013(2) 0.2730(5) 3.3(2)
C20 0.4990(6) 0.1934(2) 0.3290(5) 3 .3(2)
C21 0.3850(7) 0.2114(2) 0.3106(6) 4.3(2)
C22 0.3641(7) 0.2360(2) 0.2354(6) 5.1(2)
C23 0.4566(7) 0.2417(2) 0.1777(6) 4.6(2)
C24 0.5707(7) 0.2250(2) 0.1960(5) 4.0(2)
C25 0.8011(6) 0.1810(2) 0.2341(5) 3.2(2)
C26 0.7670(6) 0.1616(2) 0.1473(5) 3.4(2)
C27 0.8511(7) 0.1572(2) 0.0805(5) 4.0(2)
C28 0.9718(7) 0.1712(2) 0.1018(5) 4.9(2)
C29 1.0083(7) 0.1898(2) 0.1879(6) 4.9(2)
C30 0.9235(6) 0.1949(2) 0.2539(5) 3.8(2)
C31 0.7672(6) 0.1958(2) 0.4015(5) 3.4(2)
C32 0.7682(7) 0.2319(2) 0.4253(5) 4.5(2)
C33 0.8146(7) 0.2433(2) 0.5196(6) 4.8(2)
C34 0.8585(7) 0.2187(2) 0.5891(5) 5.2(2)
C35 0.8555(7) 0.1829(2) 0.5666(5) 4.4(2)
C36 0.8102(6), 0.1716(2) 0.4735(5) 3.3(2)
C40 0.5734(6) —0. 1339(2) 0.2274(5) 3 .4(2)
C41 0.4998(7) —0.1031(2) 0.2178(6) 4.7(2)
C42 0.3677(8) -0. 1035(2) 0.2148(6) 5.6(2)
C43 0.3078(7) -0. 1349(2) 0.2236(6) 5.4(2)
C44 0.3744(7) —0. 1658(2) 0.2359(6) 5.1(2)
C45 0.5054(7) —0.1655(2) 0.2377(6) 4.7(2)
C50 0.7743(6) —0. 1399(2) 0.3556(5) 3 .0(2)
C51 0.7659(7) -0.11 15(2) 0.4204(5) 3.7(2)
C52 0.7968(7) -0.1142(2) 0.5219(5) 4.1(2)
C53 0.8360(7) —0. 1464(2) 0.5631(5) 4.0(2)
C54 0.8455(7) —0. 1748(2) 0.5032(5) 4.0(2)
C55 0.8144(6) —0. 1719(2) 0.4025(5) 3.6(2)
C60 0.7878(6) -0.0987(2) 0.1952(5) 3 .4(2)

Table 3.2A. (Continued)

135

 

 

Atom x y z B(Az)
C61 0.7351(7) —0.0828(2) 0.1078(6) 4.8(2)
C62 0.7898(8) —0.0534(2) 0.0660(6) 5.6(2)
C63 0.9003(8) —0.0409(2) 0.1 1 19(6) 5.5(2)
C64 0.9582(8) —0.0546(2) 0.1973(6) 5.8(2)
C65 0.9026(7) —0.0843(2) 0.2386(6) 4.9(2)
C70 0.7816(6) —0. 1675(2) 0.1762(5) 3.5(2)
C71 0.9101(8) —0. 1754(2) 0.1915(6) 5.6(2)
C72 0.9644(8) —0.2027(2) 0.1392(6) 5.8(2)
C73 0.8907(7) —0.2207(2) 0.0706(6) 5.5(2)
C74 0.7655(8) —0. 1869(2) 0.1027(6) 6.0(2)
C75 0.71 17(7) —0. 1348(2) 0.237796) 5.0(2)

B 0.7274(8) —0.1348(2) 0.2377(6) 3.3(2)

 

136

Table 3.3A. Atomic Positional and Isotropic Thermal Parameters for 45-NaBPh4

 

 

Atom x y 2 B09)
Na 0.1828(3) 0.2488(2) 0.1393(2) 3.65(8)
01 0.2355(5) 0.2209(4) —0.0176(4) 4.0(1)
02 0.1348(5) 0.0842(4) 0.1551(4) 4.7(2)
03 0.3527(5) 0.2510(4) 0.2315(4) 4.3(2)
04 0.1675(5) 0.3158(5) 0.2995(4) 4.9(2)
05 0.2356(5) 0.4136(4) 0.1425(4) 4.5(2)
06 —0.0048(5) 0.2144(4) 0.0690(4) 4.7(2)
N1 0.3260(5) 0.1218(4) 0.0816(4) 2.9(2)
N2 0.0334(6) 0.3742(5) 0.1986(5) 3.5(2)
Cl 0.2962(6) 0.0773(6) -0.0098(5) 3.3(2)
C2 0.2536(7) 0.1309(6) —0.0609(5) 3 .3(2)
C3 0.2338(7) 0.091 1(7) —0. 1503(6) 4.4(2)
C4 0.2488(8) —0.0028(7) —0. 1865(6) 5 . 1(3)
C5 0.2824(8) —0.0571(7) -0. 1368(6) 4.8(3)
C6 0.3072(7) —0.0164(6) —0.0485(6) 3.8(2)
C7 0.3228(7) 0.0589(6) 0.1367(5) 3.4(2)
C8 0.2212(8) 0.0378(6) 0.1708(6) 4.3(2)
C9 0.217(1) —0.0263(7) 0.2205(7) 6.8(3)

C01 0.1931(9) 0.2273(7) —0.0695(6) 5.2(3)
C02 0.0342(8) 0.0669(7) 0.1952(7) 5 .7(3)
C03 0.3587(8) 0.3137(7) 0.3169(6) 5.1(3)
C04 0.2496(9) 0.2927(8) 0.3531(6) 5.7(3)
C05 0.3398(8) 0.4349(7) 0.1087(7) 5.3(3)
C06 -0.0264(9) 0.1261(8) 0.0029(8) 6.3(3)
C10 0.313(1) —0.0671(7) 0.2380(8) 7.3(3)
C11 0.4095(9) —0.0432(8) 0.2065(7) 6.5(3)
C12 0.4143(8) 0.0192(6) 0.1549(6) 4.3(2)
C 13 0.4240(6) 0.1883(5) 0.0972(5) 2.9(2)
C14 0.4387(7) 0.2538(6) 0.1792(6) 3.5(2)
C15 0.5323(8) 0.3173(7) 0.2009(6) 4.4(2)
C16 0.6102(8) 0.3165(7) 0.1370(7) 4.9(2)
C17 0.5955(7) 0.2539(7) 0.0563(6) 4.5(2)

Table 3.3A. (Continued)

137

 

 

Atom x y z B(Az)
C18 0.5021(7) 0.1879(6) 0.0362(5) 3.5(2)
C 19 —0.0148(7) 0.3273(6) 0.2582(5) 3.5(2)
C20 0.0571(8) 0.2997(7) 0.3138(6) 4.2(2)
C21 0.0169(9) 0.2579(7) 0.3765(6) 5.4(3)
C22 -0.0932(9) 0.2403(8) 0.3826(6) 6.0(3)
C23 -0. 1668(9) 0.2651(8) 0.3250(7) 6.1(3)
C24 —0. 1258(8) 0.3097(7) 0.2638(6) 4.8(3)
C25 0.0973(7) 0.4633(6) 0.2397(6) 3.9(2)
C26 0.2004(8) 0.4828(6) 0.2094(6) 4.2(2)
C27 0.2614(9) 0.5683(7) 0.2462(7) 5.4(3)
C28 0.221(1) 0.6305(8) 0.3136(8) 7.1(4)
C29 0.120(1) 0.6127(8) 0.3464(8) 7.3(4)
C30 0.0594(9) 0.5282(7) 0.3082(7) 5.9(3)
C31 —0.0328(7) 0.3734(6) 0.1220(6) 3.9(2)
C32 —0.0535(7) 0.2900(6) 0.0546(6) 4.1(2)
C33 —0.1 153(8) 0.2872(8) —0.0205(7) 5.4(3)
C34 —0. 154(1) 0.3669(9) -0.0298(7) 7.4(3)
C35 -0.135(1) 0.4478(8) 0.0344(8) 8.1(3)
C36 —0.0748(9) 0.4529(7) 0.1092(8) 5.9(3)
C40 0.3854(7) 0.3199(6) —0.2709(6) 3.6(2)
C41 0.3692(8) 0.4048(6) —0.2107(6) 4.3(2)
C42 0.4144(8) 0.4313(7) —0. 1254(6) 4.7(2)
C43 0.4772(8) 0.3727(7) -0.0985(6) 4.8(2)
C44 0.4971(9) 0.2901(7) —0. 1532(7) 5.2(2)
C45 0.4509(8) 0.2641(7) —0.2385(6) 4.8(2)
C50 0.2970(8) 0.3784(6) —-0.4027(6) 4.2(2)
C51 0.1926(8) 0.3936(7) —0.4288(6) 4.7(2)
C52 0.1762(9) 0.4708(8) —0.4618(7) 6.1(3)
C53 0.263(1) 0.5304(8) -0.4665(8) 6.9(3)
C54 0.368(1) 0.5215(8) —-0.4404(7) 6.5(3)
C55 0.3823(9) 0.4457(8) —0.4089(7) 5.7(3)
C60 0.212(1) 0.2208(8) —0.3632(7) 6.4(3)

Table 3.3A. (Continued)

138

 

 

Atom x y z B(Az)
C61 0.132(1) 0.2477(8) —0.3080(8) 7.1(3)
C62 0.029(1) 0.195(1) —0.303(1) 9.8(4)
C63 0.020(2) 0.1 14(2) —0.356( 1) 15.7(7)
C64 0.103(3) 0.057(2) -—0.395(2) 25(1)
C65 0.205(2) 0.130(2) —0.407(2) 18.9(9)
C70 0.4051(9) 0.2274(8) —0.4390(7) 5.7(3)
C71 0.367(2) 0.177(1) —0.523(1) 1 13.4(6)
C72 0.547(2) 0.154(2) —0.575(1) 16.3(7) .
C73 0.446(2) 0.137(2) —0.598(2) 19.0(9)
C74 0.578(2) 0.216(1) —0.513(1) 14.8(7)
C75 0.499(2) 0.253(1) —0.454( 1) 13.1(6)

B 0.3250(9) 0.2850(8) —0.3708(7) 4.0(2)

 

Table 3.4A. Atomic Positional and Isotropic Thermal Parameters for 45°NaB(4-C1Ph)4

139

 

 

Atom x y z B(Az)
C11 0.1393(2) 1.0598(1) 0.1141(1) 825(6)
C12 0.7463(1) 0.7248(1) 0.12256(9) 7.07(6)
C13 0.4344(2) 0.81 1 1( 1) 0.49922(9) 7.86(6)
C14 0.0732(1) 0.5520(1) 0.14773(9) 675(5)
Na 0.6030(2) 0.2837(1) 0.1196(1) 3.72(6)
01 0.5289(3) 0.3724(2) 0.0522(2) 4.8(1)
02 0.5386(3) 0.3036(2) 0.2160(2) 4.3(1)
03 0.7589(3) 0.3380(2) 0.1625(2) 4.9(1)
04 0.7246(3) 0.1959(2) 0.1603(2) 4.5( 1)
05 0.6952(3) 0.2607(2) 0.0306(2) 5. 1( 1)
06 0.4407(3) 0.2253(2) 0.1038(2) 4.5(1)
N1 0.6024(3) 0.4191(2) 0.1565(2) 2.9(1)
N2 0.6017(3) 0.1493(2) 0.0789(2) 3.2(1)
C1 0.5112(4) 0.4489(3) 0.1332(3) 3.2(2)
C2 0.4755(4) 0.4254(3) 0.0779(3) 3.7(2)
C3 0.3899(4) 0.4534(4) 0.0536(3) 5.0(2)
C4 0.3378(4) 0.5025(4) 0.0840(3) 5.6(2)
C5 0.3712(4) 0.5248(4) 0.1378(3) 6.0(2)
C6 0.4582(4) 0.4984(3) 0.1627(3) 4.6(2)
C7 0.6111(4) 0.4171(3) 0.2202(3) 3.3(1)
C8 0.5739(4) 0.3590(3) 0.2491(3) 3.5(2)
C9 0.5774(4) 0.3594(4) 0.31 1 1(3) 4.8(2)
C10 0.6177(5) 0.4170(4) 0.3405(3) 5.9(2)
C11 0.6561(5) 0.4723(4) 0.3115(3) 6.4(2)
C12 0.6522(5) 0.4729(4) 0.2517(3) 5.1(2)
C13 0.6893(4) 0.4448(3) 0.1292(2) 3.2(1)
C14 0.7723(4) 0.4016(3) 0.1333(3) 4.1(2)
C15 0.8581(4) 0.4216(4) 0.1104(3) 4.9(2)
C16 0.8628(4) 0.4842(4) 0.0803(3) 5.9(2)
C17 0.7827(5) 0.5274(4) 0.0749(3) 5.4(2)
C18 0.6959(4) 0.5079(3) 0.1003(3) 4.4(2)
C19 0.5987(4) 0.1167(3) 0.1356(3) 3.5(2)

Table 3.4A. (Continued)

140

 

 

Atom x y z B(AZ)
C20 0.6671(4) 0.1401(3) 0.1788(3) 3.7(2)
C21 0.6730(4) 0.1 1 10(4) 0.2337(3) 4.6(2)
C22 0.6082(5) 0.0591(4) 0.2478(3) 5.4(2)
C23 0.5361(5) 0.0373(3) 0.2068(3) 5.2(2)
C24 0.5325(4) 0.0658(3) 0.1509(3) 4.2(2)
C25 0.6929(4) 0.1386(3) 0.0505(3) 3 .5(2)
C26 0.7391(4) 0.1956(3) 0.0247(3) 3.9(2)
C27 0.8260(4) 0.1862(4) —0.0027(3) 5.1(2)
C28 0.8669(5) 0.1 187(4) —0.0036(3) 6.3(2)
C29 0.8241(5) 0.0628(4) 0.0222(3) 6.0(2)
C30 0.7374(4) 0.0727(3) 0.0495(3) 4.5(2)
C31 0.5135(4) 0.1434(3) 0.0416(3) 3.6(2)
C32 0.4307(4) 0.1808(3) 0.0566(3) 3.7(2)
C33 0.3457(4) 0.1740(4) 0.0228(3) 5.1(2)
C34 0.3434(5) 0.1310(4) —0.0273(3) 6.1(2)
C35 0.4254(5) 0.0965(4) —0.0428(3) 5.6(2)
C36 0.5115(5) 0.1010(3) —0.0076(3) 4.3(2)
C50 0.3002(4) 0.8720(3) 0.2015(3) 3.6(1)
C51 0.2117(4) 0.8989(3) 0.2211(3) 4.4(2)
C52 0.1616(5) 0.9553(4) 0.1947(3) 5.0(2)
C53 0.2015(5) 0.9889(4) 0.1483(3) 5.0(2)
C54 0.2868(5) 0.9656(4) 0.1259(3) 5.0(2)
C55 0.3352(5) 0.9073(3) 0.1528(3) 4.7(2)
C60 0.4521(4) 0.7790(3) 0.2004(3) 3.4( 1)
C61 0.5305(4) 0.8272(3) 0.2010(3) 4.1(1)
C62 0.6199(4) 0.8115(3) 0.1767(3) 4.2(2)
C63 0.6337(4) 0.7467(3) 0.1518(3) 4.4(2)
C64 0.5616(5) 0.6974(4) 0.1510(3) 5.0(2)
C65 0.4723(4) 0.7142(3) 0.1743(3) 4.6(2)
C70 0.3506(4) 0.8638(3) 0.3351(3) 4.3(2)
C71 0.3712(4) 0.8050(3) 0.3006(3) 3 .5 ( l )
C72 0.4162(4) 0.7486(3) 0.3327(3) 3.7(1)

Table 3.4A. (Continued)

141

 

 

Atom x y z B(Az)
C73 0.4351(4) 0.7487(3) 0.3932(3) 4.0(1)
C74 0.4112(4) 0.8087(3) 0.4231(3) 4.1(1)
C75 0.3708(5) 0.8659(4) 0.3958(3) 4.9(2)
C80 0.2693(4) 0.7346(3) 0.2138(3) 3.4(1)
C81 0.2174(4) 0.7362(3) 0.1596(3) 4.4(2)
C82 0.1554(4) 0.6807(3) 0.1407(3) 4.6(2)
C83 0.1455(4) 0.6230(3) 0.1737(3) 4.0(1)
C84 0.1905(4) 0.6182(3) 0.2282(3) 4.5(2)
C85 0.2515(4) 0.6742(3) 0.2466(3) 3 .9( 1)
C01 0.4968(6) 0.3497(4) —0.0050(3) 6.5(2)
C02 0.4980(5) 0.2448(3) 0.2463(3) 4.9(2)
C03 0.8385(4) 0.2883(4) 0.1678(3) 5.6(2)
C04 0.7990(5) 0.2237(4) 0.1982(3) 5.6(2)
C05 0.7362(6) 0.3186(4) 0.0001(3) 6.9(2)
C06 0.3565(4) 0.2636(4) 0.1 195(3) 5.7(2)

B 0.3475(5) 0.7992(4) 0.2296(3) 3.4(2)

 

Table 3.5A. Atomic Positional and Isotropic Thermal Parameters for

45‘KB(4-C1Ph)4°CH3N02

142

 

 

Atom x y z B(Az)
K1 0.9349(2) 0.0178(2) 0.6953(2) 4790)
C11 0.08070 —0.5538(4) 0.94150 8.1(1)
C12 0.8384(3) —0.6852(3) 1.0548(2) 6.6(1)
C13 0.4063(3) —1. 1071(3) 0.6807(2) 7.7( 1)
C14 0.6423(4) —0.3077(3) 0.6285(2) 8.0(1)
01 0.7955(5) -0.0671(7) 0.7744(4) 4.9(2)
02 1.0090(7) —0.l840(7) 0.6774(4) 6.8(3)
03 1.0984(5) 0.0381(6) 0.7868(4) 4.5(2)
04 1.0958(5) 0.1260(7) 0.6513(4) 5.4(2)
05 0.8740(6) 0.2264(7) 0.7333(4) 5.2(2)
06 0.8332(6) 0.0116(7) 0.5680(4) 5.7(2)

C(31) 0.371(1) 0.241(1) 0.5496(7) 13.2(4)

C(32) 0.275(1) 0.342(2) 0.6031(9) 17.7(6)
N1 0.9806(6) —0. 1323(7) 0.8150(5) 3.4(2)
N2 0.9297(6) 0.2036(7) 0.5964(4) 3.2(2)

N(S) 0.351(1) 0.324(1) 0.5756(8) 11.2(4)
C1 0.8999(8) —0.2044(9) 0.8247(6) 3.6(2)
C2 0.8044(9) —0. 172(1) 0.8038(6) 4.2(3)
C3 0.7246(9) —0.243(1) 0.8130(7) 5.1(3)
C4 0.7417(9) —0.340(1) 0.8393(7) 5.7(3)
C5 0.8272(9) —0.376(1) 0.8610(7) 5.4(3)
C6 0.9114(9) —0.307(1) 0.8548(6) 5.2(3)
C7 1.0689(8) -0.1841(9) 0.7951(6) 3.5(2)
C8 1.0783(9) -0.214(1) 0.7241(6) 5.4(3)
C9 1.1619(9) —0.273(1) 0.7029(7) 5.8(3)

C(S) 0.418(1) 0.409(2) 0.583(1) 11.4(6)
C10 1.231(1) —0.297(1) 0.7509(8) 7.0(4)
C11 1.225(1) —0.271(1) 0.8208(8) 8.1(4)
C12 1.142(1) —0.210(1) 0.8418(7) 6.2(3)
C13 0.9942(8) —0.0450(9) 0.8667(6) 3.4(2)
C14 1.0550(8) 0.040(1) 0.8520(6) 3.9(3)

Table 3.5A. (Continued)

143

 

 

Atom x y z B(AZ)
C15 1.0691(9) 0.125(1) 0.9026(6) 4.8(3)
C16 1.0162(9) 0.125(1) 0.9629(7) 5.0(3)
C17 0.9537(9) 0.044(1) 0.9771(7) 4.8(3)
C18 0.9415(9) -0.044(1) 0.9288(6) 4.3(3)
C19 1.0002(8) 0.1600(9) 0.5502(6) 3.6(2)
C20 1.0880(8) 0.122(1) 0.5775(6) 4.1(3)
C21 1.1618(8) 0.084(1) 0.5355(6) 4.5(3)
C22 1.1452(9) 0.081(1) 0.4654(7) 6.1(3)
C23 1.061(1) 0.115(1) 0.4343(7) 6.5(4)
C24 0.9844(9) 0.157(1) 0.4772(7) 5 .5(3)
C25 0.9583(8) 0.2958(9) 0.6371(6) 3.5(2)
C26 0.9277(9) 0.310(1) 0.7056(6) 4.5(3)
C27 0.953(1) 0.404(1) 0.7455(7) 6.8(4)
C28 1.013(1) 0.481(1) 0.7144(8) 7.7(4)
C29 1.044(1) 0.466(1) 0.6493(9) 8.2(4)
C30 1.017(1) 0.375(1) 0.6076(7) 6.2(3)
C31 0.8292(8) 0.2020(9) 0.5732(6) 3.6(2)
C32 0.7796(8) 0.105(1) 0.5587(6) 4.2(3)
C33 0.6812(9) 0.106(1) 0.5405(7) 5.4(3)
C34 0.632(1) 0.203(1) 0.5352(7) 6.1(3)
C35 0.679(1) 0.299(1) 0.5461(8) 6.9(4)
C36 0.7783(9) 0.297(1) 0.5666(7) 4.9(3)
C50 0.3878(8) —0.6353(9) 0.8661(6) 3.6(3)
C51 0.3070(9) —0.625(1) 0.8209(6) 4.5(3)
C52 0.2137(9) —0.598(1) 0.8405(7) 5.3(3)
C53 0.2012(9) —0.584(1) 0.9140(7) 5.3(3)
C54 0.274(1) —0.593(1) 0.9610(7) 5.7(3)
C55 0.3717(9) —0.6 15(1) 0.9368(6) 4.8(3)
C60 0.5801(8) —0.6773(9) 0.8955(6) 3.5(3)
C61 0.6310(8) —0.7726(9) 0.9143(6) 3.9(3)
C62 0.7096(9) —0.774(1) 0.9640(6) 4.4(3)
C63 0.7384(8) —0.681(1) 0.9956(6) 4.2(3)

Table 3.5A. (Continued)

144

 

 

Atom x y z B(Az)
C64 0.6909(8) —0.586(1) 0.9829(6) 4.1(3)
C65 0.6128(8) —0.584(1) 0.9333(6) 4.6(3)
C70 0.4786(8) —0.7829(9) 0.7914(6) 3.6(3)
C71 0.4981(9) —0.808(1) 0.7231(6) 4.9(3)
C72 0.4815(9) —0.905( 1) 0.6857(6) 5.0(3)
C73 0.4378(9) —0.983(1) 0.7252(6) 4.3(3)
C74 0.4177(9) -0.973(1) 0.7937(7) 4.8(3)
C75 0.4369(9) 0.8659(4) 0.3958(3) 4.9(2)
C80 0.5304(8) —0.5731(9) 0.7788(6) 3.5(3)
C81 0.4825(9) —0.473(1) 0.7688(7) 4.9(3)
C82 0.5166(9) —0.390( 1) 0.7231(7) 5 .4(3)
C83 0.6016(9) —0.410(1) 0.6862(6) 4.4(3)
C84 0.6547(9) —0.506( 1) 0.6971(6) 5.0(3)
C85 0.6163(8) —0.583( 1) 0.7421(6) 4.4(3)
C01 0.6987(9) —0.030( 1) 0.7517(8) 7.0(4)
C02 1.019(2) —0.210(2) 0.611(1) 29.9(8)
C03 1.1625(8) 0.123(1) 0.7666(7) 5.6(4)
C04 1.1848(8) 0.101(1) 0.6896(7) 6.1(4)
C05 0.844(1) 0.232(1) 0.8063(6) 6.6(4)
C06 0.803(1) —0.084(1) 0.533(1) 13.4(7)

B 0.495(1) —0.665(1) 0.8316(7) 3.6(3)

 

Table 3.6A. Atomic Positional and Isotropic Thermal Parameters for 45

145

 

 

Atom x y z B(Az)
01 0.5913(9) —0.0880(6) 0.5966(5) 3.8(2)
02 0.774(1) 0.0751(5) 0.8657(5) 4.2(2)
03 0.3805(9) 0.1617(5) 0.7580(4) 3.4(2)
04 0.2454(9) 0.3734(5) 0.7979(4) 3.1(2)
05 -0.163(l) 0.2347(5) 0.6912(5) 3.8(2)
06 0.084(1) 0.3656(6) 0.5528(5) 4.2(2)
N1 0.485(1) -0.0108(6) 0.7676(5) 2.8(2)
N2 —0.041(1) 0.4192(6) 0.7176(5) 2.9(2)

C01 0.647(2) —0.122(1) 0.5101(8) 5.3(4)
Cl 0.369(1) —0.0565(7) 0.6932(6) 2.3(3)
C02 0.892(2) 0.124(1) 0.937(1) 9.3(6)
C2 0.422(1) —0.0913(7) 0.6053(7) 2.7(3)
C03 0.236(1) 0.1910(8) 0.7235(7) 3.7(3)
C3 0.307(1) —-0.1327(8) 0.5335(8) 3.8(3)
C04 0.266(2) 0.2858(8) 0.7142(7) 3.9(3)
C4 0.136(2) —0. 1319(9) 0.5459(8) 4.5(4)
C05 -0.214(2) 0.1342(9) 0.6784(9) 5.3(4)
C5 0.089(2) —0.0947(8) 0.6326(8) 4.2(4)
C06 0.150(2) 0.343(1) 0.4699(8) 5.6(4)
C6 0.197(1) -0.0580(8) 0.7055(7) 3.7(3)
C7 0.613(1) -0.0721(7) 0.7801(7) 3.0(3)
C8 0.759(1) —0.0291(8) 0.8302(7) 3.5(3)
C9 0.883(2) —0.0894(9) 0.8423(7) 4.5(3)
C 10 0.859(2) —0. 1930(9) 0.8024(8) 5.4(4)
C11 0.719(2) —0.2365(8) 0.7512(8) 4.7(3)
C12 0.593(2) —0. 1779(8) 0.7387(8) 4.2(3)
C 13 0.429(1) 0.0704(7) 0.8478(6) 2.1(2)
C14 0.372(1) 0.1578(8) 0.8405(7) 2.9(3)
C15 0.315(1) 0.2370(8) 0.9158(7) 3.4(3)
C16 0.315(1) 0.2317(9) 0.9964(7) 3.7(3)
C17 0.371(2) 0.1461(9) 1.0019(7) 3.9(3)
C18 0.425(1) 0.0675(8) 0.9286(6) 3.2(3)

Table 3.6A. (Continued)

146

 

 

Atom x y z B(AZ)
C19 0.104(1) 0.4860(8) 0.7444(7) 2.6(3)
C20 0.241(1) 0.4625(8) 0.7866(7) 3.0(3)
C21 0.375(1) 0.5292(9) 0.8213(7) 3.6(3)
C22 0.381(2) 0.6180(9) 0.8101(9) 4.9(4)
C23 0.248(2) 0.6388(8) 0.7664(9) 4.5(4)
C24 0.113(2) 0.5740(8) 0.7336(8) 4.0(3)
C25 —0.l 15(1) 0.4049(7) 0.7897(6) 2.6(3)
C26 -0. 179(1) 0.3122(8) 0.7761(7) 3.4(3)
C27 —0.258( 1) 0.2966(8) 0.8424(8) 4.1(3)
C28 —0.266(2) 0.3792(9) 0.9257(7) 4.6(4)
C29 —0. 199(2) 0.4727(9) 0.9417(8) 4.9(4)
C30 —0. 124(1) 0.4858(8) 0.8750(7) 3.6(3)
C31 —0.146(1) 0.4157(7) 0.6448(6) 2.3(3)
C32 —0.081(1) 0.3872(7) 0.5614(7) 3.2(3)
C33 —0. 185(2) 0.3827(8) 0.4905(8) 4.3(3)
C34 —0.355(2) 0.4033(9) 0.5027(7) 4.9(4)
C35 —0.414(2) 0.428(1) 0.5857(9) 5.8(4)
C36 —0.3 12(1) 0.4352(8) 0.6556(8) 3.7(3)

 

REFERENCES

10.

ll.

12.

147

Lindsey, J. S. New J. Chem. 1991, 5, 153-180.
Rajca, A. Chem. Rev. 1994, 94, 871-893.

Baxter, P.; Lehn, J .-M.; DeCian, A.; Fischer, J. Angew. Chem, Int. Ed. Engl.
1993, 32, 69—72.

Caspar, D. L. D.; Klug, A. Cold Spring Harbor Symp. Quant. Biol. 1962, 2 7, 1—
24.

Timasheff, S. N. In Protein-Protein Interactions; Frieden, C., Nichols, L. W.,
Eds.; Wiley: New York, 1981; pp 315-336.

Klug, A. Angew. Chem, Int. Ed. Engl. 1983, 22, 565-582.

(a) Etter, M. C.; Frankenbach, G. M. Chem. Mater. 1989, 1, 1—10. (h) Etter, M.
C.; Frankenbach, G. M.; Adsmond, D. A. Mol. Cryst. Liq. Cryst. 1990, 187, 25—
39. (c) Prasad, P. N.; Williams, D. J. Introduction to Nonlinear Optical Effects in
Molecules and Polymers; Wiley: New York, 1991. (d) Etter, M. C.; Huang, K. S.;
Frankenbach, G. M.; Adsmond, D. A. In Materials for Nonlinear Optics:
Chemical Perspectives; Marder, S. R., Sohn, J. E., Stucky, G. D., Eds.; American
Chemical Society: Washington, DC, 1991; Vol. 455, pp 446—456. (e) Russell, V.
A.; Etter, M. C.; Ward, M. D. Chem. Mater. 1994, 6, 1206—1217.

(a) Torrance, J. B. Acc. Chem. Res. 1979, 12, 79—86. (b) Wudl, F. Acc. Chem Res.
1984, 17, 227—232. (c) Williams, J. M.; Beno, M. A.; Wang, H. H.; Leung, P. C.
W.; Emge, T. J.; Geiser, U.; Carlson, K. D. Acc. Chem. Res. 1985, 18, 261—267.
((1) Fagan, P. J .; Ward, M. D.; Calabrese, J. C. J. Am. Chem. Soc. 1989, 111,
1689—1719. (e) Ward, M. D.; Pagan, P. J .; Calabrese, J. C.; Johnson, D. C. J. Am.
Chem. Soc. 1989, 111, 1719-1732.

(a) Torrance, J. B.; Oostra, S.; Nazzal, A. Synth. Met. 1987, 19, 709—714. (b)
Miller, J. S.; Epstein, A. J .; Reiff, W. M. Acc. Chem. Res. 1988, 21, 114-120. (c)
Miller, J. S.; Glazhofer, D. T.; Calabrese, J. C.; Epstein, A. J. J. Chem. Soc.,
Chem. Commun. 1988, 322—323. ((1) Miller, J. S.; Epstein, A. J. Angew. Chem,
Int. Ed. Engl. 1994, 33, 385—415.

(a) Desiraju, G. R. Crystal Engineering: The Design of Organic Solids; Materials
Science Monographs, Vol. 54; Elsevier: New York, 1989. (b) Wright, J. D.
Molecular Crystals, 2nd ed.; Cambridge University Press: Cambridge, NY, 1995.
Sarrna, J. A. R. P.; Desiraju, G. R. Acc. Chem. Res. 1986, 19, 222—228.

(a) Diederich, F. Angew. Chem, Int. Ed. Engl. 1988, 27, 362—386. (b) Hamilton,
A. D.; Pant, N.; Muehldorf, A. Pure Appl. Chem. 1988, 60, 533—538. (c)
Zimmerman, S. C.; Wu, W. J. Am. Chem. Soc. 1989, 111, 8054—8055. ((1) Rebek,
J ., Jr. Angew. Chem, Int. Ed. Engl. 1990, 29, 245—255. (e) Ebmeyer, F .; Rebek, J .,

l3.

14.

15.

16.

17.

18.

19.

20.

21.

22.

148

Jr. Angew. Chem, Int. Ed. Engl. 1990, 29, 1148-1150. (f) Wyler, R.; de Mendoza,
J .; Rebek, J ., Jr. Angew. Chem, Int. Ed. Engl. 1993, 32, 1699—1701.

(a) Vinogradov, S. N.; Linnel, R. H. Hydrogen Bonding; Van Nostrand Reinhold:
New York, 1971. (b) Joesten, M. D.; Schaad, L. J. Hydrogen Bonding; M.
Dekker: New York, 1974. (c) Jeffrey, G. A.; Saenger, W. Hydrogen Bonding in
Biological Structures; Springer-Verlag: Berlin, 1991.

(a) Etter, M. C. Acc. Chem. Res. 1990, 23, 120—126. (b) Etter, M. C. J. Phys.
Chem. 1991, 95, 4601-4610. (c) Reutzel, S. M.; Etter, M. C. J. Phys. Org. Chem.
1992, 5, 44—54. ((1) Etter, M. C.; MacDonald, J. C.; Wanke, R. A. J. Phys. Org.
Chem. 1992, 5, 191—200.

(a) Leiserowitz, L. Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem.
1976, B32, 775—802. (b) Leiserowitz, L. Acta Crystallogr., Sect. B: Struct.
Crystallogr. Cryst. Chem. 1977, B33, 2719—2733. (c) Leiserowitz, L.; Tuval, M.
Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem. 1978, B34, 1230—
1247. (d) Weinstein, S.; Leiserowitz, L. Acta Crystallogr., Sect. B: Struct.
Crystallogr. Cryst. Chem. 1980, B36, 1406—1418. (e) Weinstein, S.; Leiserowitz,
L.; Gil-Av, E. J. Am. Chem. Soc. 1980, 102, 2768-2772.

(a) Taylor, R.; Kennard, O. J. Am. Chem. Soc. 1982, 104, 5063—5070. (b) Taylor,
R.; Kennard, O. Acc. Chem. Res. 1984, 17, 320—326.

(a) Zhao, X.: Chang, Y.-L.; Fowler, F. W.; Lauher, J. W. J. Am. Chem. Soc. 1990,
112,‘ 6627—6634. (b) Fan, 15.; Vicent, C.; Geib, S. J .; Hamilton, A. D. Chem.
Mater. 1994, 6, 1113—1117. (c) MacDonald, J. C.; Whitesides, G. M. Chem. Rev.
1994, 94, 2383—2420. ((1) Whitesides, G. M.; Mathias, J. P.; Seto, C. T.; Chin, D.
N.; Mammen, M.; Gordon, D. M.; Acc. Chem. Res. 1995, 28, 37—44.

(a) Hernandez, E.; Mas, M.; Molins, E.; Rovira, C.; Veciana, J. Angew. Chem,
Int. Ed. Engl. 1993, 32, 882—884. (b) Cirujeda, J .; Ochando, L. E.; Amigo, J. M.;
Rovira, C.; Rius, J .; Veciana, J. Angew. Chem, Int. Ed. Engl. 1995, 32, 55—57. (c)
Cirujeda, J .; Mas, M.; Molins, E.; de Panthou, F. L.; Laugier, J.; Park, J. G.;
Paulsen, C.; Rey, P.; Rovira, C.; Veciana, J. J. Chem. Soc., Chem. Commun. 1995,
709—710.

Simard, M.; Su, D.; Wuest, J. D. J. Am. Chem. Soc. 1991, 113, 4696—4698.

Martel], A. E.; Calvin, M. Chemistry of the Metal Chelate Compounds; Prentice-
Hall: New York, 1952.

Pedersen, C. J. Angew. Chem, Int. Ed. Engl. 1988, 27, 1021—1027.

Lehn, J.-M. Acc. Chem. Res. 1978, II, 49—57.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

149

Cram, D. J. Angew. Chem, Int. Ed. Engl. 1986, 25, 1039-1057.
Lehn, J.-M. In Struct. Bonding, 1973, 16, 1—69.

Seiler, P.; Dobler, M.; Dunitz, J. D. Acta Crystallogr., Sect. B: Struct. Crystallogr.
Cryst. Chem, 1974, B30, 2744-2745.

van Remoortere, F. P.; Boer, F. P. Inorg. Chem. 1974, 13, 2071-2078.

Boer, F. P.; Neuman, M. A.; van Remoortere, F. P.; Steiner, E. C. Inorg. Chem.
1974, 13, 2826—2834.

Feneau-Dupont, P. J .; Arte, E.; Declercq, J. P.; Germain, G.; Van Meerssche, M.
Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem, 1979, B35, 1217—
1220.

Mallinson, P. R.; Truter, M. R. J. Chem. Soc., Perkin Trans. 11 1972, 1818—1823.
Hughes, D. L. J. Chem. Soc., Dalton Trans. 1975, 2374-2378.

(a) Fenton, D. E.; Mercer, M.; Poonia, N. S.; Truter, M. R. J. Chem. Soc., Chem.
Commun. 1972, 66—67. (b) Mercer, M.; Truter, M. R. J. Chem. Soc., Dalton
Trans. 1973, 2469—2473.

Owen, J. D.; Truter, M. R. J. Chem. Soc., Dalton Trans. 1979, 1831—1835.

(a) Groh, S. E. Isr. J. Chem 1976/77, 15, 277—307. (b) Drew, M. G. B.; Rodgers,
A.; McCann, M.; Nelson, S. M. J. Chem. Soc., Chem. Commun. 1978, 415—416.
(c) Coughlin, P. K.; Dewan, J. C.; Lippard, S. J.; Watanabe, E. 1.; Lehn, J .-M. J.
Am. Chem. Soc. 1979, 101, 265—266. ((1) Agnus, Y.; Louis, R.; Weiss, R. J. Am.
Chem. Soc. 1979, 101, 3381—3384.

Parsons, D. G.; Truter, M. R.; Wingfield, J. N. Inorg. Chim. Acta 1975, 14, 45—
48.

Lehn, J .-M.; Pine, S. H.; Watanabe, E. 1.; Willard, A. K. J. Am. Chem. Soc. 1977,
99, 6766—6768.

(a) Melinger, M.; Fischer, J .; Weiss, R. Angew. Chem, Int. Ed. Engl. 1973, 12,
771. (b) Wiest, R.; Weiss, R. J. Chem. Soc., Chem. Commun. 1973, 678—679. (c)
Fischer, J .; Mellinger, M.; Weiss, R. Inorg. Chim. Acta 1977, 21, 259—263.
Lehn, J.-M. Pure Appl. Chem. 1980, 52, 2441—2459.

Lehn, J.-M. Angew. Chem, Int. Ed. Engl. 1990, 29, 1304—1319.

39.

40.

41.

42.

43.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

150

Lehn, J .-M.; Rigault, A.; Siege], J .; Harrowfield, J .; Chevrier, B.; Moras, D. Proc.
Natl. Acad. Sci. USA, 1987, 2565—2569.

(a) Constable, E. C.; Ward, M. D. J. Am. Chem. Soc. 1990, 112, 1256—1258. (b)
Constable, E. C. Tetrahedron 1992, 48, 10013—10059.

(a) Lehn, J.-M.; Rigault, A. Angew. Chem, Int. Ed. Engl. 1988, 27, 1095—1097.
(b) Garrett, T. M.; Koert, U.; Lehn, J .-M.; Rigault, A.; Meyer, D.; Fischer, J. J.
Chem. Soc., Chem. Commun. 1990, 557—558.

Kramer, R.; Lehn, J.-M.; DeCian, A.; Fischer, J. Angew. Chem, Int. Ed. Engl.
1993, 32, 703—706.

Leize, A.; Van Dorsselaer, A.; Kramer, R.; Lehn, J .-M. J. Chem. Soc., Chem.
Commun. 1993, 990—993.

Baxter, P. N. W.; Lehn, J .-M.; Fischer, J .; Youinou, M.-T. Angew. Chem, Int. Ed.
Engl. 1994, 33, 2284—2287.

Hanan, G. S.; Arana, C. R.; Lehn, J .-M.; Fenske, D. Angew. Chem, Int. Ed. Engl.
1995, 34, 1122—1124.

(a) Dewar, M. J. S.; Talati, A. M. J. Am. Chem. Soc. 1963, 85, 1874. (b) Dewar,
M. J. S.; Talati, A. M. J. Am. Chem. Soc. 1964, 86, 1592—1595.

(a) Fox, M. A.; Chandler, D. A.; Wang, P. Macromolecules 1991, 24, 4626—4636.
(b) Fox, M. A.; Chandler, D. A.; Kyba, E. P. J. Coord. Chem. 1992, 25, 1-19.
Fox, M. A.; Chandler, D. A. Adv. Mater. 1991, 3, 381—384.

Flurry, R. L., Jr. Quantum Chemistry: An Introduction; Prentice—Hall: Englewood
Cliffs, New Jersey, (1983.

Mich], J; Bonacic-Koutecky, V. Electronic Aspects of Organic Photochemistry;
Wiley: New York, 1990.

Salem, L. Electrons in Chemical Reactions: First Principles; Wiley: New York,
1982. ’

Veciana, J.; Vidal, J .; Jullian, N. Mol. Cryst. Liq. Cryst. 1989, 176, 443—450.
Kollmar, C.; Kahn, 0. ACC. Chem. Res. 1993, 26, 259—265.
Kollmar, H.; Staemmler, V. Theor. Chim. Acta 1978, 48, 223—239.

(a) Dowd, P.; Chang, W.; Paik, Y. H. J. Am. Chem. Soc. 1986, 108, 7416—7417.
(b) Roth, W. R.; Kowalczik, U.; Maier, G.; Reisenauer, H. P.; Sustmann, R.;

56.

57.

58.

59.

60.

61.

62.

63.

65.

151

Mfiller, W. Angew. Chem, Int. Ed. Engl. 1987, 26, 1285—1287. (c) Du, P.;
Borden, W. T. J. Am. Chem. Soc. 1987, 109, 930—931.

(a) Dowd, P. J. Am. Chem. Soc. 1966, 88, 2587—2589. (b) Baseman, R. J .; Pratt,
D. W.; Chow, M.; Dowd, P. J. Am Chem. Soc. 1976, 98, 5726—5727.

Dixon, D. A.; Dunning, T. H., Jr.; Eades, R. A.; Kleier, D. A. J. Am. Chem. Soc.
1981, 103, 2878—2880.

Hudson, B. S.; Kohler, B. E.; Schulten, K. In Excited States; Lim, E. C., Ed.;
Academic Press: New York, 1982, Vol. 6, pp 54—55.

(a) Migirdicyan, E.; Baudet, J. J. Am. Chem. Soc. 1975, 97, 7400—7404. (b) Platz,
M. S. In Diradicals; Borden, W. T., Ed.; Wiley: New York, 1982, pp 195—258. (c)
Rule, M.; Matlin, A. R.; Seeger, D. E.; Hilinski, E. F.; Dougherty, D. A.; Berson,
J. A. Tetrahedron 1982, 38, 787—798. ((1) Wright, B. B.; Platz, M. S. J. Am. Chem.
Soc. 1983, 105, 628—630. (e) Goodman, J.; Berson, J. A. J. Am. Chem. Soc. 1985,
107, 5409—5424. (1) Berson, J. A. In The Chemistry of Quinoid Compounds; Patai,
S., Rappaport, 2., Eds.; Wiley: New York, 1988; Vol. 2, Chapter 10.

(a) Kato, S.; Morokuma, K.; Feller, D.; Davidson, B. R.; Borden, W. T. J. Am.
Chem. Soc. 1983, 105, 1791—1795. (b) Fort, R. C., Jr.; Getty, S. J .; Hrovat, D. A.;
Lahti, P. M.; Borden, W. T. J. Am. Chem. Soc. 1992, 114, 7549—7552.

(a) Flynn, C. R.; Michl, J. J. Am. Chem. Soc. 1974, 96, 3280—3288. (b)
Trahanovsky, W. S.; Chou, C.-H.; Fischer, D. R.; Gerstein, B. C. J. Am. Chem.
Soc. 1988, 110, 6579—6581.

Williams, D. J.; Pearson, J. M.; Levy, M. J. Am. Chem. Soc. 1970, 92, 1436—1438.

Polycarbenes with spins up to S = 9 have been prepared. See: (a) Fujita, I.; Teki,
Y.; Takui, T.; Kinoshita, T.; Itoh, K.; Miko, F.; Sawaki, Y.; Iwamura, H. Izuoka,
A.; Sugawara, T. J. Am. Chem. Soc. 1990, 112, 4074—4075. (b) Nakamura, N.;
Inoue, K.; Iwamura, H.; Fujioka, T.; Sawaki, Y. J. Am. Chem. Soc. 1992, 114,
1484-1485. (c) Iwamura, H.; Koga, N. Acc. Chem. Res. 1993, 26, 346—351. (d)
Nakamura, N.; Inoue, K.; Iwamura, H. Angew. Chem, Int. Ed. Engl. 1993, 32,
872—874.

Schlenk, W.; Brauns, M. Chem. Ber. 1915, 48, 661—669; 716—728.
(a) Kothe, G.; Denkel, K.-H.; Sfimmermann, W. Angew. Chem, Int. Ed. Engl.

1970, 9, 906—907. (b) Lockhurst, G. R.; Pedulli, G. F.; Tiecco, M. J. Chem. Soc. B
1971, 329—334.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

152

(a) Lankamp, H.; Nauta, W. Th.; MacLean, C. Tetrahedron Lett. 1968, 249—254.
(b) For a review on the “hexaphenylethane riddle”, see: McBride, J. M.
Tetrahedron 1974, 30, 2009—2022.

(a) Neumann, W. P.; Uzick, W.; Zarkadis, A. K. J. Am. Chem. Soc. 1986, 108,
3762—3770. (b) Dfinnebacke, D.; Neumann, W. P.; Penenory, A.; Stewen, U.
Chem. Ber. 1989, 122, 533—535. (c) Neumann, W. P.; Penenory, A.; Stewen, U.;
Lehnig, M. J. Am. Chem. Soc. 1989, 111, 5845—5851.

(a) Veciana, J .; Rovira, C.; Crespo, M. 1.; Armet, 0.; Domingo, V. M.; Palacio, F.
J. Am. Chem. Soc. 1991, 113, 2552-2561. (b) Rajca, A.; Utamapanya, S.; Xu, J. J.
Am. Chem. Soc. 1991, 113, 9235—9241. (c) Rajca, A.; Utamapanya, S. J. Org.
Chem. 1992, 57, 1760—1767.

Rajca, A.; Utamapanya, S.; Smithhisler, D. J. Org. Chem. 1993, 58, 5650—5652.
(a) Jarrett, H. S.; Sloan, G. J .; Vaughn, W. R. J. Chem. Phys. 1956, 25, 697-701.
(b) Sloan, G. J .; Vaughn, W. R. J. Org. Chem. 1957, 22, 750—761. (0)
Montgomery, L. K.; Huffman, J. C.; Jurczak, E. A.; Grendze, M. P. J. Am. Chem.
Soc. 1986, 108, 6004—6011.

For an excellent discussion of the “diradical paradox” associated with 19, see ref
59b.

Schmidt, R.; Brauer, H.-D. Angew. Chem, Int. Ed. Engl. 1971, 10, 506—507.
Silverman, S. K; Dougherty, D. A. J. Phys. Chem. 1993, 97, 13273—13283.

(a) Pranata, J .; Dougherty, D. A. J. Phys. Org. Chem. 1989, 2, 161—176. (b)
Goldberg, A. H.; Dougherty, D. A. J. Am. Chem. Soc. 1983, 105, 284—290.

Hoffmann, R. Acc. Chem. Res. 1971, 4, 1—9.
Dougherty, D. A. Acc. Chem. Res. 1991, 24, 88—94.

(a) Sinn, E. Coord. Chem. Rev. 1970, 5, 313—347. (b) Hay, P. J .; Thibeault, J. C.;
Hoffmann, R. J. Am. Chem. Soc. 1975, 97, 4884—4899.

Kahn, 0.; Tola, P.; Galy, J .; Coudanne, H. J. Am. Chem. Soc. 1978, 100, 3931—
3933.

Kahn, 0.; Galy, J.; Journaux, Y.; Jaud, J .; Morgenstem-Badarau, I. J. Am. Chem.
Soc. 1982, 104, 2165—2176.

(a) Kahn, 0. Comments Inorg. Chem. 1984, 3, 105—132. (b) Kahn, O. Angew.
Chem, Int. Ed. Engl. 1985, 24, 834—850.

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

96.

153

Caneschi, A.; Gatteschi, D.; Sessoli, R. Acc. Chem. Res. 1989, 22, 392—398.

(a) Eaton, S. S.; Eaton, G. R. Coord. Chem. Rev. 1978, 26, 207—262. (b) Eaton, S.
S.; Eaton, G. R. Coord. Chem. Rev. 1988, 83, 29—72.

Porter, L. C.; Dickman, M. H.; Doedens, R. J. Inorg. Chem. 1988, 27, 1548—1552.
Gatteschi, D.; Laugier, J .; Rey, P.; Zanchini, C. Inorg. Chem. 1987, 26, 938—943.
Dickman, M. H.; Porter, L. C.; Doedens, R. J. Inorg. Chem. 1986, 25, 2595—2599.

Benelli, C.; Gatteschi, D.; Zanchini, C.; Doedens, R. J .; Dickman, M. H.; Porter,
L. C. Inorg. Chem. 1986, 25, 3453-3457.

Sabacky, M. J.; Johnson, C. S., Jr.; Smith, R. G.; Gutowsky, H. S.; Martin, J. C. J.
Am. Chem. Soc. 1967, 89, 2054-2058.

(a) V6gtle, F.; Weber, E. Angew. Chem, Int. Ed. Engl. 1979, 18, 753-776. (b)
Weber, E.; Vogtle, F. Top. Current Chem. 1981, 98, 1-41. (c) Hilgenfeld, R.;
Saenger, W. In Host Guest Complex Chemistry—Macrocycles—Synthesis,
Structures, Applications; Vogtle, F.; Weber, E., Eds.; Springer-Verlag: Berlin,
1985, pp 43—124.

Kahr, B.; Jackson, J. E.; Ward, D. L.; Jang, S.-H.; Blount, J. F. Acta Crystallogr.,
Sect. B: Struct. Sci. 1992, B48, 324—329.

Mislow, K. Acc. Chem. Res. 1976, 9, 26—33.

By analogy to the crown ethers, we propose the name “tripod ether” for the
propeller-like tris(o-alkoxyary1)-Z compounds, where Z may be any sp2 or sp3
hybridized element.

Artz, S. P.; Cram, D. J. J. Am. Chem. Soc. 1984, 106, 2160—2171.

Force-field calculations suggest a racemization barrier of ca. 10 kcal/mol
(Jackson, J. E., unpublished results).

Frye, C. L.; Vincent, G. A.; Hauschildt, G. L. J. Am. Chem. Soc. 1966, 88, 2727—
2730.

Muller, E.; Burgi, H.-B. Acta Crystallogr., Sect. C: Struct. Commun. 1989, C45,
1400—1403.

Barriers of ca. 12—13 kcal/mol have been determined for compounds closely
related to 28. See: Hellwinkel, D.; Melan, M.; Degel, C. R. Tetrahedron 1973, 29,
1895-1907.

97.

98.

99.

100.

101.

102.

103.

104.

105.

106.

107.

108.

109.

110.

111.

112.

113.

114.

115.

154

Gauthier, S.; Fréchet, J. M. J. Synthesis 1987, 383—385.
Nelson, R. R; Adams, R, N. J. Am. Chem. Soc. 1968, 90, 3925—3930.
Loveland, J. W.; Dimeler, G. R. Anal. Chem. 1961, 33, 1196—1201.

Seo, E. T.; Nelson, R. F.; Fritsch, J. M.; Marcoux, L. S.; Leedy, D. W.; Adams, R.
N. J. Am. Chem. Soc. 1966, 88, 3498—3503.

Dollish, F. R.; Hall, W. K. J. Phys. Chem. 1965, 59, 2127—2129.
Fleischmann, M.; Pletcher, D. Tetrahedron Lett. 1968, 6255—6258.
Stickley, K. R.; Blackstock, S. C. J. Am. Chem. Soc. 1995, 116, 11576—11577.

Vogtle, F.; Weber, E. In The Chemistry of the Ether Linkage; Patai, S., Ed.;
Supplement E; Wiley: London, 1981, part 1, p 59.

Gutmann, V. Chimia 1977, 31, 1—7.

Cahen, Y. M.; Handy, P. R.; Roach, E. T.; Popov, A. I. J. Phys. Chem. 1975, 79,
80—85.

Greenberg, M. S.; Bodner, R. L.; Popov, A. I. J. Phys. Chem. 1973, 77, 2449—
2454.

Lund, H.; J. Am. Chem. Soc. 1927, 49, 1346—1360.

Brown, H. C.; Dodson, V. H. J. Am. Chem. Soc. 1957, 79, 2302—2306.

(a) Heimann, U.; Herzhoff, M.; Vogtle, F. Chem. Ber. 1979, 112, 1392—1399. (b)
Jang, S.-H.; Lee, H.-I.; McCracken, J.; Jackson, J. E. J. Am. Chem. Soc. 1993,
115, 12623-12624.

Lehn, J .-M. Angew. Chem, Int. Ed. Engl. 1988, 27, 89—112.

Dostal, S.; Stoudt, S. J .; Fanwick, P.; Sereatan, W. F.; Kahr, B.; Jackson, J. E.
Organometallics 1993, 12, 2284—2291.

Hayes, K. S.; Nagumo, M.; Blount, J. F.; Mislow, K. J. Am. Chem. Soc. 1980,
102, 2273-2276.

Martin, J. C.; Smith, R. G. J. Am. Chem. Soc. 1964, 86, 2252-2256.

Blount, J. F.; Finocciaro, P.; Gust, D.; Mislow, K. J. Am. Chem. Soc. 1973, 95,
7019-7029.

116.

117.

118.

119.

120.

121.

122.

123.

124.

125.

126.

127.

128.

129.

130.

155

Olmstead, M. M.; Power, P. P. J. Am. Chem. Soc. 1986, 108, 4235-4236.
Bartlett, R. A.; Power, P. P. Organometallics 1985, 5, 1916-1917.
Kahr, B.; Van Engen, D.; Gopalan, P.Chem. Mat. 1993, 5, 729-732

Brown, G. M.; Freeman, G. R.; Walter, R. I. J. Am. Chem. Soc. 1977, 99, 6910-
6915.

AMl method: Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J. J. P. J.
Am. Chem. Soc. 1985, 107, 3902-3909.

Jang, S.-H.; Gopalan, R; Jackson, J. E.; Kahr, B. Angew. Chem, Int. Ed. Engl.,
1994, 33, 775—777.

Jang, S.-H.; Bertsch, R. A.; Jackson, J. E.; Kahr, B. Mol. Cryst. Liq. Cryst. 1992,
211, 289—303.

(a) V6gtle, F.; Miiller, W. M.; Wehner, W.; Buhleier, E. Angew. Chem, Int. Ed.
Engl. 1977, 16, 548—549; (b) Vogtle, F.; Miiller, W. M.; Buhleier, E.; Wehner, W.
Chem. Ber. 1979, 112, 899—907; (c) Tiimmler, B.; Maass, G.; Vogtle, F.; Sieger,
H.; Heimann, U.; Weber, E. J. Am. Chem. Soc. 1979, 101, 2588—2598. For
reviews, see: ((1) Vogtle, F.; Weber, E. Angew. Chem, Int. Ed. Engl. 1979, 18,
753—776; (e) Vogtle, F. Chimia 1979, 33, 239—251.

Green, D. P.; Reich, H. J. J. Am. Chem. Soc. 1989, 11 1 , 8729-8731.

In acetone, which is a better solvating medium than CDC13, the ion pair formation
constant for Lil is reported to be 145. See: Savedoff, L. G. J. Am. Chem. Soc.
1966, 88, 664-667.

Brownstein, S. K.; Plouffe, P.-Y.; Bensimon, C.; Tse, J. Inorg. Chem. 1994, 33,
354—358.

Gilkerson, W. R.; Jackson, M. D. J. Am. Chem. Soc. 1982, 104, 1218—1223.
Bauer, W. Magn. Reson. Chem. 1991, 29, 494—499.

(a) Bauer, W.; Schleyer, P. v. R. Magn. Reson. Chem. 1988, 26, 827—833. (b)
Hoffmann, D.; Bauer, W.; Schleyer, P. v. R. J. Chem. Soc., Chem. Commun.
1990, 208—211.4(c) Bauer, W.; Lochmann, L. J. Am. Chem. Soc. 1992, 114, 7482—
7489. (d) Hilmersson, G.; Davisson, G. J. Organomet. Chem. 1995, 489, 175-
179. (e) Hilmersson, G.; Davisson, O. Organometallics 1995, 14, 912—918.

Erlich, R. H.; Popov, A. I. J. Am. Chem. Soc. 1971, 93, 5620—5623.

131.

132.

133.

134.

135.

136.

137.

138.

139.

140.

141.

142.

143.

156

Kintzinger, J. P.; Lehn, J.-M. J. Am. Chem. Soc. 1974, 96, 3313—3314.

Dahlquist, F. W.; Longmuir, K. J .; Du Vernet, R. B. J. Magn. Reson. 1975, 17,
406—410.

In THF, which is a better solvating medium than CDC13, the ion pair formation
constants for LiBPh4 and NaBPh4 at 25 °C are reported to be 1.26 x 104 and 1.23
x 104, respectively. See: Comyn, J .; Dainton, F. S.; Ivin, K. J. Electrochim. Acta
1968, 13, 1851—1860.

(a) Bothner-By, A. A.; Stephens, R. L.; Lee, J .-M.; Warren, C. D.; Jeanholz, R.
W. J. Am. Chem. Soc. 1984, 106, 811—813. (b) Bax, A.; Davis, D. G. J. Magn.
Reson. 1985, 63, 207-213.

Moras, D.; Weiss, R.- Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem.
1973, 329, 396—399.

Tehan, F. J.; Barnett, B. L.; Dye, J. L. J. Am. Chem. Soc. 1974, 96, 7203—7208.

(a) Adolphson, D. G.; Corbett, J. D.; Merryman, D. J. J. Am. Chem. Soc. 1976, 98,
7234—7239; (b) Teller, R. G.; Finke, R. G.; Collman, J. P.; Chin, H. B.; Bau, R. J.
Am. Chem. Soc. 1977, 99, 1104-1111.

Moras, D.; Metz, B.; Weiss, R. Acta Crystallogr., Sect. B: Struct. Crystallogr.
Cryst. Chem. 1973, 329, 383-388.

Stoudt, S. J .; Gopalan, P.; Kahr, B.; Jackson, J. E. Struct. Chem. 1994, 5, 335—
340.

The rotational barrier in anisole is a matter of recent debate. Estimates from both
theory and experiment are converging on a value of 2—3 kcal mol-1 for the
perpendicular transition state. See: (a) Spellmeyer, D. C.; Grootenhuis, P. D. J .;
Miller, M. D.; Kuyper, L. F.; Kollman, P. A. J. Phys. Chem. 1990, 94, 4483—
4491. (b) Vincent, M. A.; Hillier, I. A. Chem. Phys. 1990, 140, 35—40.

(a) Cox, B. G.; Knop, D.; Schneider, 11. J. Phys. Chem. 1980, 84, 320—323; (b)
Cox, B. G.; Van Truong, Ng.; Garcia-Rosas, J .; Schneider, H. J. Phys. Chem.
1984, 88, 996—1001; (c) Buschmann, H.-J. Chem. Ber. 1985, 118, 3408—3412.

For a more complete compendium of binding data, see: Izatt, R. M.; Pawlak, K.;
Bradshaw, J. S.; Bruening, R. L. Chem. Rev. 1991, 91, 1721—2085.

The anisyl group is an intrinsically poor ligand, as determined by its weakness in
hydrogen bonding. See: (a) Amett, E. M.; Joris, L.; Mitchell, E.; Murty, T. S. S.
R.; Schleyer, P. v. R. J. Am. Chem. Soc. 1970, 92, 2365—2377. (b) Mitsky, J .;
Joris, L.; Taft, R. W. J. Am. Chem. Soc. 1972, 94, 3442-3445.

144.

145.

146.

147.

148.

149.

150.

151.

152.

153.

154.

155.

156.

157

Triphenylamine itself forms exceedingly unstable adducts with Lewis acids. See:
Kemmitt, R. D. W.; Nuttall, R. H.; Sharp, D. W. A. J. Chem. Soc. 1960, 46—50.

For example, see: (a) Gomberg, M.; Nishida, D. J. Am. Chem. Soc. 1923, 45, 190—
207. (b) Gomberg, M.; Buchler, C. C. J. Am. Chem. Soc. 1923, 45, 207—222. (c)
Gomberg, M.; Forrester, G. C. J. Am. Chem. Soc. 1925, 47, 2373—2391. ((1) Bent,
H. E.; Dorfman, M.; Bruce, W. F. J. Am. Chem. Soc. 1932, 54, 3250—3258. (e)
Bachmann, W. E.; Kloetzel, M. C. J. Org. Chem. 1935, 2, 356—375. (1) Marvel, C.
S.; Whitson, J.; Johnston, H. W. J. Am. Chem. Soc. 1944, 66, 415—417. (g)
Bowden, S. T. J. Chem. Soc. 1957, 4235—4239. (h) Bowden, S. T.; Zalichi, D. T.
J. Chem. Soc. 1957, 4240—4243. (i) Zarkadis, A. K.; Neumann, W. P.; Marx, R.;
Uzick, W. Chem. Ber. 1985, 118, 450—456.

Ishizu, K.; Mukai, K.; Shibayama, A.; Kondo, K. Bull. Chem. Soc. Jpn. 1977, 50,
2269—2271.

Compound 49 has been used as a hydride abstraction reagent. See: Huszthy, P.;
Lempert, K.; Simig, G.; Tamas, J. H.-V.; Almasy, A. J. ChemResearch (S) 1985,
228—229.

Tukada, H. J. Chem. Soc., Chem. Commun. 1994, 2293—2294.

(a) Broser, W.; Kurreck, H.; Niemeier, W. Tetrahedron 1976, 32, 1183—1187.
Also see: (b) Hinrichs, K.; Kurreck, H.; Niemeier, W. Tetrahedron 1974, 30, 315—
320. (c) Kurreck, H.; Niemeier, W. Tetrahedron Lett. 1974, 3523—3526. (d)
Broser, W.; Kurreck, H.; Plato, M. Tetrahedron 1975, 31, 1769—1779.

Shriver, D. F.; Drezdzon, M. A. The Manipulation of Air-Sensitive Compounds,
2nd ed.; Wiley: New York, 1986.

Lund, H.; Bjerrum, J. Ber. 1931, 64, 210—213.

(a) Burfield, D. R.; Gan, G.-H.; Smithers, R. H. J. Appl. Chem. Biotechnol. 1978,
28, 23-30. (b) Burfield, D. R.; Smithers, R. H. J. Org. Chem. 1978, 43, 3966-
3968.

Bhattacharyya, D. N.; Lee, C. L.; Szwarc, M. J. Phys. Chem. 1965, 69, 608—61 1.

DeWitte, W. J .; Liu, L.; Mei, E.; Dye, J. L.; Popov, A. I. J. Solution Chem. 1977,
6, 337—348.

Furniss, B. S.; Hannaford, A. J .; Smith, P. W. G.; Tatchell, A. R. Vogel ’s
Textbook of Practical Organic Chemistry, 5th ed.; Wiley: New York, 1989, p 421.

Winkle, M. R.; Lansinger, J. M.; Ronald, R. C. J. Chem. Soc., Chem. Commun.
1980, 87—88.

158

157. Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory Chemicals, 3rd ed.;
Pergamon: New York, 1988.

158. Soulié, C.; Bassoul, R; Simon, J. New J. Chem. 1993, I 7, 787—791.

159. Flad, G.; Demerseman, P.; Royer, R. Bull. Soc. Chim. Fr. 1976, 11—12, 1823—
1824.

160. Sugii, Y.; Shindo, H.; J. Pharm. Soc. Jpn. 1934, 54, 829—844.

161. VanArendonk, A. M.; Cupery, M. E.; Adams, R. J. Am. Chem. Soc. 1933, 55,
4225—4230.

162. Lofgren, N. M.; Takman, B. Acta Chem. Scand. 1952, 6, 1016—1019.

163. Angelici, R. J.; Quick, M. H.; Kraus, G. A.; Plummer, D. T. Inorg. Chem. 1982,
21, 2178—2184.

164. MNDO method: (a) Dewar, M. J. S.; Thiel, W. J. Am. Chem. Soc. 1977, 99,
4899—4907. (b) Dewar, M. J. S.; Thiel, W. J. Am. Chem. Soc. 1977, 99, 4907-
4917.

165. Dye, J. L.; Nicely, V. A. J. Chem. Ed. 1971, 48, 443—448.

166. (a) Weigert, F. J .; Roberts, J. D. J. Am. Chem. Soc. 1969, 91, 4940—4941. (b)
Odom, J .; Hall, L. W.; Ellis, P. D. Org. Magn. Reson. 1974, 6, 360—361.

167. Walker, N.; Stuart, D. Acta Crystallogr., Sect. A: Cryst. Phys. Dijfr. Theor. Gen.
Crystallogr. 1983, A39, 158-166.

168. Sheldrick, G. M., Institut fiir Anorganische Chemie der Universitat Gottingen,
FRG, 1986.

169. MolEN, An Interactive Structure Solution Procedure, Enraf-Nonius, Delft, The
Netherlands (1990).

"llllllllllllllll