b... .9
a... .
.3... z

a! “4R
. 0:..5

              

 

 

   

   

7: .

               

  

4"
. .2333...
. .:u.....n.b.._...n .2,
aruflmrva z. y c «z.

 

                                 

Liwzixv out” .
~....c_....:H..~. Ht...
z. _
f

 

 

.1... 54......» .fi.
J...» ... 5.15%.
E .- .. .
--..r2...,u..
5.... 1....

 

    

.3.
fan: :1?
.3.

 

V “I...“ .
. .E
#31::‘5-330‘10 ,?(. .r, .- ,..
.11 5.33.313: 9.3L. &\Q
“h”... u‘au.1l.ilaw!um II}

ii...
\I.» 1“ 3713i )5 :3
5:. s. . .5 ...u.. .. .a
{121... :9... .
fist... sauna... 31.3.. afiuhrfi
.. .5 , .35.... .3 1.
"unfinisfiok fléiafin
fiwafi .n x fitriw
. u !. x i. A t4
. .uyu. h. ”uh; :rumuu
. svfifififiwx -
44...... .224 ..

52“.wa

         

   

                            

5.

2.3.1.3.... ..

. ... Wigwam?
0..

.rlv
: .r

    

             

as

3.. elviu is

. {at .1 .
t-.l‘fin4|o..w.vh§

flung...

.a

I"

 

   

           

    

I, utxul\
3.115....VI.
€3.15... SI. ..

 

oi.

        

 

      

    

, or?!

4.: n it: :I
Atatlfilulvtaulrrtmhmm. ’5???
fit... V13 i fix hi! .

                

 

3;;

   

 
  

..
a: 2 s , ..
a. . - ‘46....quifil. Radium
..:...,s,.., s. 3 :2. . .a
i .h..-_..1..x d? ##5... 5...... .. finhnfmri
. . i. .2. .. . ., a...
1,. ....-.......-..3....; .3...

5).}. 3r Eilul:

.. I: 7.4.13.2. "Wmllvo!
.. .31: a 3 ? ‘u

9. 91!. .1 $3.};

Viv;

 

 

 

 

.. 110...:

.93.?!I5.

.. nlvxn mi
0.0.5!

 

filoétl.
V1.3; ..)
:1... £056.

                             

 

    

 

 

 

 

 

 

.SV’I:Z‘.‘I:\ II.
.kvlllitll I .
y: I 3.)? Au 1!

 

 

   

         

DI‘V: .

 

~..I.a»(rl’\1’s|m{‘!.v. ,0?
: lial‘....:€:£ivi .
I: .

uv...)

 

 

 

 

 

 

 

9.3.1. I
559:..-

 

       

 

 

. EYE». at“... Qw.o...upu....mswm.r. ,
£3... . H

xiv f.....«...‘.:- .. .1. .
r.

«y E!) (I.

 

 
 

WIN:

C) .05;- ‘>«-

u 9-"

 

 

R LIBRARY I
" Michigan State te
L__ University

 

 

 

This is to certify that the

thesis entitled

SYNTHESIS AND CHARACTERIZATION OF
INCLUSION COMPLEXES OF OL-CYCLODEXTRIN
AND (AB)u BLOCK COPOLYMERS

presented by

Kirk L. Olson

has been accepted towards fulfillment
of the requirements for

M.S. degree in CHEMISTRY

 

,fl/fi%\

Major professor

Date ZI/crlqufi 20W

MS U is an Affirmative Action/Equal Opportunity Institution

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6101 cJCIRCIDmeDtnpes-nts

 

SYNTHESIS AND CHARACTERIZATION OF INCLUSION COMPLEXES OF
a-CYCLODEXTRIN AND (AB)n BLOCK COPOLYMERS

By

Kirk L. Olson

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF SCIENCE
Department of Chemistry

2001

ABSTRACT

SYNTHESIS AND CHARACTERIZATION OF INCLUSION COMPLEXES OF
a-CYCLODEXTRIN AND (AB)n BLOCK COPOLYMERS

By

Kirk L. Olson

Cyclodextrins thread onto polymer chains to form inclusion complexes, especially
when the polymer is hydrophobic relative to the solvent. Selective threading might occur
when the polymer architecture contains both hydrophobic and hydrophilic segments. on-
Cyclodextrin formed crystalline inclusion complexes with (AB)n microblock copolymers,
where the A block is a linear alkyl segment containing a single double bond and the B
block is an exact length segment of polyethylene oxide. The complexes were isolated
and characterized by solution and solid state NMR, XRD, DSC, and TGA. Each method
confirmed complex formation and showed that the physical properties of the complexes
are distinct from its individual components. The X-ray data are consistent with known
inclusion complexes that have a channel or column crystal structure. The stoichiometry
of complex formation, 2.3 a-cyclodextrin rings per polymer repeat unit, was determined
by NMR analysis of the complexes, and from an analysis of the yields of inclusion
complexes. The data suggest that the inclusion complex stoichiometry is defined by the
increasing insolubility of the polymer-cyclodextrin complex. Solid state NMR suggests

that there is a slight preference for threading onto hydrophobic segments of the (AB)n

polymer.

To my family

iii

ACKNOWLEDGEMENTS

First and foremost I would like to express my thanks and gratitude to my parents
and my sister for their undeniable love and support for me throughout my graduate school
experience. I write and dedicate this thesis for you.

I would also like to thank my advisor, boss, and friend, Gregory L. Baker, for his
knowledge, support, and guidance throughout the course of this research. Baker group
members have made this an enjoyable experience and I hope to stay in touch with all.
Gao (woo?) thanks for mentoring me and being a true fn'end. Micah, Chun, .IB, and
Tianqi you have been wonderful sources of knowledge and fun. Erin thanks for being a
lunch mate and a friend going through the same experiences. Fadi.... thanks for some
good} laughs and a source of bad coffee!

Finally I would. like to thank all my friends and family for their support! It has
been a fun ride! I have really enjoyed my graduate school experience and met some true

friends. Thanks, and God bless all.

iv

TABLE OF CONTENTS

LIST OF ABBREVIATIONS ............................................................... vii
LIST OF FIGURES ........................................................................... viii
LIST OF SCHEMES ......................................................................... x
LIST OF TABLES ............................................................................ xi
CHAPTER 1
INTRODUCTION
1.1 Cyclodextrins ........................................................................ l
1.2 Block copolymer properties in general ........................................... 5
1.2.1 Amphiphilic materials. . .‘ ................................................... 8
1.2.2 Synthesis and properties of (AB)n Copolymers and ABA triblock
oligomers ...................................................................... 10
1.3 Host-Guest chemistry ................................................................ 15
1.3.1 Host-Guest complexes with cyclodextrin ................................ 21
1.3.2 Polymer inclusion iomplexes .............................................. 26
1.4 References ............................................................................ 31
CHAPTER 2

SYNTHESIS AND CHARACTERIZATION OF INCLUSION COMPLEXES OF
a-CD AND (AB)n BLOCK COPOLYMERS

2.1 Acronyms for compounds used in the research .................................. 34
2.2 Results ................................................................................ 35
2.2.1 Complex formation between [erthEOy]n polymers and
a-CD .......................................................................... 35
2.2.2 Properties of (AB)n-a-CD complexes ..................................... 38
2.2.3 Complex formation between CXEOny model compounds
and a-CD ...................................................................... 51
2.2.4 Properties of ABA-a-CD complexes ..................................... 52

2.3 Discussion ........................................................................... 57

2.3.1 Solid state '3 C NMR of inclusion complexes ............................ 58
2.3.2 Stoichiometry of inclusion complexes .................................... 59
2.4 Conclusions ............................... - ........................................... 60
2.5 References ........................................................................... 62
CHAPTER 3
EXPERIMENTAL
3.1 General details ...................................................................... 64
3.2 Material synthesis .................................................................. 65
3.3 References ........................................................................... 67

vi

ADMET
BA

CD
Chaoycx
[ancxiaoy]n
CP

osc

DD

G

GPC

H/G

IC

LIST OF ABBREVIATIONS

Acyclic diene metathesis
Polybutyl acrylate

Cyclodextrin

See page 11

See page 11

Cross polarization

Differential scanning calorimetry
Dipolar decoupling

Guest molecule

Gel permeation chromatography
Host-guest complex

Host molecule

Inclusion complex

Magic angle spinning

Number average molecular weight
Weight average molecular weight
Nuclear magnetic resonance
Poly (e-caprolactone)

Poly (1,3-dioxolane)
Polyethylene

Polyethylene oxide
Polyisobutylene

Polypropylene oxide

Polystyrene

Ring opening metathesis polymerization
Thermogravimetric analysis
Crystallization temperature
Glass-transition temperature
Melting point

X-ray diffraction

Six glucose repeat units in CD
Seven glucose repeat units in CD
Eight glucose repeat units in CD
Chemical shift

Wavelength
Diffraction angle

vii

Figure 1.1

Figure 1.2

Figure 1.3
Figure 1.4
Figure 1.5

Figure 1.6

Figure 1.7
Figure 1.8
Figure 1.9
Figure 1.10
Figure 1.11

Figure 1.12

Figure 1.13

Figure 1.14

Figure 1.15

Figure 1.16

LIST OF FIGURES

Representative structures of cyclodextrin ................................ 1

Distribution of CD relevant papers published in 1996 by

C yclodextrin News .......................................................... 4
Types and examples of block copolymers ............................... 6
Alternating, random, and grafted block copolymers ................... 6
Morphologies of AB block copolymers .................................. 7
Structure of amphiphilic-block copolymers and the formation

of micelles in aqueous solutions .......................................... 9
Types and examples of surfactants ....................................... 10
Grubb and Schrock’s metal alkylidene metathesis catalyst ............ 12
Emil Fischer’s Lock and Key Model for enzyme activity ............. 15
Organic host molecules ..................................................... 16
Model of calixarene host-guest complex ................................. 20

A schematic representation of CD inclusion complex
formation with p-xylene as the guest molecule ......................... 22

X—ray crystal structure of a-CD hexahydrate viewed along
the a axis ..................................................................... 23

Crystal structures of (a) dimethyl-a-CD complexed with iodine
(herringbone-type cage) and (b) a-CD complexed with
p-nitrophenol (brick-type cage) .......................................... 24

Stereodrawing of the structure of a-CD complexed with an
iodine-iodide lithium salt (channel-type crystal structure) ............ 25

Diagram of the reversible formation of a pseudorotaxane (I) from
its acyclic and cyclic components, and the irreversible conversion
into a rotaxane (II) .......................................................... 27

viii

Figure 1.17
Figure 1.18

Figure 2.1

Figure 2.2

Figure 2.3

Figure 2.4

Figure 2.5

Figure 2.6

Figure 2.7

Figure 2.8

Figure -2.9

Figure 2.10

Figure 2.1 1

Diagrams of polypseudorotaxanes (III) and polyrotaxanes (IV). . . .. 27
Proposed superstructure for PEG-a-CD complexes .................... 28

Yield of [anCxEOy]n-a-CD inclusion complexes as a function
of polymer concentration ................................................... 37

Normalized DSC thermograms of a) [C37tC3EO6]n-a-CD IC,
b) the de-threaded complex and c) the pure [C31tC3EO6]n
homopolymer ................................................................ 39

TGA scans for a) [C47tC4E03]n, b) [C41rC4E03]n-or-CD IC, and
c) a-CD (the sample was dried for 12 h at 100 °C at the beginning
of the run) ..................................................................... 41

TGA scans for a) a-CD, b) [C3TCC3E06L1‘a'CD IC, and
c) [C31tC3EO6]n + a-CD mixture .......................................... 42

Powder X—ray diffraction patterns for a-CD complexes.
a) l-propanol-a—CD, b) [C31IC3EO6]n-a-CD, c) CMEOgCM-a-CD,
and d) octanol-a-CD ................. . ...................................... 44

‘3C CP/MAS NMR spectra of a) a-CD, b) CleOMCn-a-CD,
C) C12E014C12, d) [C31‘CC3EOg]n-a-CD, and C)[C31tC3EOg]n ........... 46

Continuous variation plot showing the yield of or-CD/(AB)n block
copolymer complexes obtained for various ratios of reactants ........ 49

The 300 MHz ‘H NMR spectrum of a) a-CD, b) the
[C47tC4EOg]n-a-CD IC, and c) [C47‘EC4E03]n in DMSO-d6 ............ 50

Yield of CXEOny-a-CD Complex as a function of the ABA
triblock concentration ........................................ g .............. 54

Continuous variation plot showing the yield of the

C12EOgC12-(X-CD IC (0 ) obtained for various ratios of the
reactants ..................................................................... 55

Normalized DSC thermograms of a) C12E08C12,
b) C12E03C12-a-CD IC, and C) a-CD .................................... 56

ix

Scheme 1.]

Scheme 1.2

Scheme 1.3

Scheme 1.4

Scheme 1.5

Scheme 2.1

LIST OF SCHEMES

Generic structure of (AB)n polymer and ABA model compounds

used in this research ......................................................... 1 1
Synthesis of (AB)n multiblock copolymers .............................. 12
Synthesis of ABA triblock oligomers .................................... 13
Synthesis of exact length oligoethylene glycols ........................ 14
Synthesis of 18-Crown-6 ................................................... 17

Nomenclature of (AB)n block copolymers, ABA model
compounds, and inclusion complexes formed with a-CD ............ 34

Table 1.1.

Table 1.2

Table 1.3

Table 1.4

Table 2.1

Table 2.2

LIST OF TABLES

Characteristics of or, B, and y-cyclodextrins ............................. 2
Selected applications of cyclodextrins in industry ..................... 4
Comparison of various cation and crown ether cavity diameters. . . .. 17

Comparison of log KS values (in water solutions at 25 °C) of
various metal cations with organic host molecules 1 and 2 ............ l9

Characteristics of (AB)n block copolymers used in IC formation
with a-CD .................................................................... 36

Characteristics of ABA triblock oligomers used to form IC’s
with oc-CD .................................................................... 51

xi

CHAPTER 1
Introduction
1.1 Cyclodextrins
Cyclodextrins (CDs) are a series of cyclic oligosaccharides consisting of 6, 7, 8,
or larger (10, 14, or 26) glucose units joined by ct-1,4-glycosidic linkages. Cyclodextrins
are designated by the use of Greek letters (or, B, y. . .), which correspond to the number of
glucose units in the CD ring. The smallest cyclodextrin contains six (or) glucose units.
Cyclodextrins (Figure 1.1) resemble a hollow truncated cone with primary hydroxyl
groups located on the narrow side of the cone, and secondary hydroxyl groups on the
wide rim of the cone. The hydroxyl groups located on the exterior of the cone make CDs
hydrophilic and soluble in aqueous solutions, but the interior cavities of CD5 are
hydrophobic allowing CDs to incorporate small molecules or guests to form host-guest

inclusion complexes (ICs).I

 

¢ 0
0 £0
+ n
”0e ° + I
”J 0‘ P? Q‘% ,0 0 .:°
0

Figure 1.1. Representative structures of cyclodextrin as illustrated by Szejtli.2
Villiers discovered cyclodextrin in 1891 while he was digesting starch with a

' He collected two distinct crystalline products (on

microorganism, Bacillus amylobacter.
and B-CD) that he called cellulosines. In 1904, Schardinger, the first person to actually

characterize the crystalline extracts produced from starch, determined that the

cellulosines were actually cyclic oligosaccharides. Because of his discovery, CDs are
sometimes referred to as Schardinger dextrins.l The procedure used for the synthesis of
CDs today is very similar to that by Villiers, but involves the use of a different
microorganism, Bacillus macerans, which produces the cyclodextrin glucosyl transferase
enzyme (CGT-ase).2 a, [3, and y-CDs are formed upon the addition of CGT-ase to a
prehydrolyzed starch solution. The glycosyl transferase enzyme acts by fragmenting the
starch helix and then linking the ends of the fragment to form cyclic oligosaccharides.
From the starch digests, the components of the complicated mixture of CDs can be
separated by the use of selective complex forming agents.2 Toluene is used to complex
with B-CD, 1-decanol with a-CD, and cyclohexadecenol for the larger y-CD. After
complexation with these agents, the insoluble complexes are separated from the mixture
by filtration, washed and then the complex agent is extracted from the CD to give the
pure product. Today CDs are produced in quantities of greater than 1000 tons per year,
and in greater than 99% purity.2 Some of the characteristics and physical properties of
commercially available CDs are listed in Table 1.1.

Table 1.1. Characteristics of or, [3, and y-cyclodextrins.2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a B Y
Number of glucose units 6 7 8
Molecular weight 972 1135 1297
Solubility in water, g 100 mL'1 at 14.5 1.85 23.2
room temp
[a]D 25°C 150 i 0.5 162.5 i 0.5 177.4 :t .5
Cavity diameter, A 4.7-5.3 6.0-6.5 7.5-8.3
Height oftorus, A 7.9 3c 0.1 7.9 :1: 0.1 7.9 i 0.1
Diameter of outer periphery, A 14.6 :t 0.4 15.4 :t 0.4 17.5 :t 0.4
Approx. volume of cavity, A3 ' 174 262 427
Crystal water, wt % 10.2 13.2-14.5 8.13-17.7

 

The structure of CDs (Figure 1.1) and the cavity diameters listed in Table 1.1
suggests the possibility that CD molecules could form a complex (two or molecules held
together in a definable structural relationship) with other molecules in the interior cavity
of cyclodextrin. In the early 1950’s, Cramer discovered that CD and small organic
molecules could form complexes, and upon complexation with CD, the guest molecule
obtained certain benefits such as alteration of the solubility of the guest compound,
stabilization against the effects of heat, light, and oxidation, and also a reduction of
volatility.3 Because of these benefits, CDs are now employed in numerous industrial
applications including the pharmaceutical, food and cosmetic industries.4 Figure 1.2
describes the breakdown of CD-related papers in terms of their relevance to industrial
applications for the year 1996.2 The most common application of cyclodextrins is in the
pharmaceutical industry where drugs complexed with CD are found to have enhanced
solubility, stability, and bioavailability of the drug molecules. Table 1.2 summarizes
some of the uses of CD found in the pharmaceutical, food, and cosmetic industries.

Cyclodextrin is not only important because of its numerous applications found in
industry, but also because CD complexes have become model systems for studying the
non-covalent attractive forces that are found in complex biological systems, such as
enzymes, proteins, and DNA. Supramolecular structures, catenanes and rotaxanes, can
be made that mimic similar types of structures that are found in nature. Using the
complexing ability of CDs, one can easily synthesize these supramolecular structures,
which are not readily prepared by other methods. Recently, CD has been used as types of
ligands or sites for molecular recognition of various chemical compounds, which has

given renewed interest in the possible uses of cyclodextrin. In the work described in this

thesis, CD has been employed to study the properties of polymer inclusion complex
formation with an amphiphilic multiblock copolymer.

Foods and

Cosmetics , ,
Other 7% 1% Pest1c1des

16% .
D

Production of C

24% Pharmaceuticals

 
  
   

and Derivatives "’ . 22% Chemistry of CD
19%. complexes
Analytical
Chemistry

Figure 1.2. Distribution of CD relevant papers published in 1996 by Cyclodextrin News.2

Table 1.2. Selected applications of cyclodextrins in industryz‘4

of
Stabilize

Mustard
to a

Mask bitterness

Mask odor

Cholesterol
Pharmaceutical

Increased

Reduce irritation

Cosmetics and Personal Care Items

 

1.2 Block copolymer properties in general

Block copolymers are macromolecules comprised of two dissimilar repeat units
(A and B) that are connected (chemically bonded) to one another. Copolymers can be
synthesized in a variety of patterns that are distinguished by the number of blocks per
molecule (Figure 1.3). The blocks in these copolymers can be arranged so that repeating
units can be alternating, random, or grafted to one another (Figure 1.4).5

The combination of two homopolymers (A and B) to form a block copolymer
allows the polymer to not only have the properties of the individual segments, but also
give rise to new properties not found in a simple blend of polymers A and B. In a blend,
polymers A and B phase separate at the macroscopic level as a consequence of the
positive free energy change for mixing (AGmix).

AGmix = AHm -TAS,,,,x Equation 1.1

When two polymers are mixed together, the entropy of mixing (218m) is very small due
to the high molecular weight of the polymers. In most cases, the enthalpy of mixing
(AHM) is slightly positive, which leads to an overall slightly positive free energy change
(AGM) that results in incompatibility or phase separation. In block copolymers,
aggregation of individual polymer blocks with one another leads to phase separation, but
because the blocks are chemically linked to one another, macroscopic phase separation is

inhibited. The F lory-Huggins parameter ()0 is defined as,
= AHm,,./kTNA v3 Equation 1.2

where NA is the number of repeat units in block A, and vB is the volume fraction of block

B. x is a dimensionless parameter that describes the interactions or compatibility

between blocks A and B. If x is negative, polymers are miscible with one another

(homogenous), but if x is positive the copolymers will phase separate.

AxBy AXBYAX
Diblock Triblock

 

Polystyrene-b-polyethylene oxide Poly(styrene-b-butadiene-b-styrene)
(PS-PEO) (SBS)

(AxBy)n
Multiblock

WOMCK

[C47‘EC4EOy]n

Figure 1.3. Types and examples of block copolymers.

ABABABABAB -{—1AABABABBBAAB%— +AAAATAAAAA+
Alternating Random B Graft
B

—wwwww

Figure 1.4. Alternating, random, and grafted block copolymers.

Incompatibility of blocks leads to the formation of new polymer structures
(morphologies) that depend on the composition of the blocks in the copolymer (Figure
1.5). If the volume fractions, v, of the blocks are nearly equal, layered lattices or
lamellae (Figure 1.5. C) form, but if the volume fraction of one of the blocks is larger
than the other, then spheres (A), cylinders (B), or bicontinuous structures can be formed.
Phase diagrams can be made that describe the change in the micro-domain structure of

the polymer as a function of the volume fraction of the blocks.

   

Figure 1.5. Morphologies of AB block copolymers. White portions represent block A,
while dark portions represent block B of the AB polymer.6

Block copolymers are synthesized to alter the properties of a homopolymer to
achieve desired properties such as decreased crystallinity or improvements in the impact
resistance. The most common application of block copolymers is as thermoplastic
elastomers. Thermoplastic elastomers are polymers that possess thermally reversible
chemical or physical cross-links. An example of a thermoplastic elastomer is the ABA
triblock copolymer poly(styrene-b-butadiene-b-styrene) (SBS, Figure 1.3).5‘7 Styrene
and butadiene are incompatible with each another so they form separate phases joined at
the junctures where the various blocks are connected. This results in an elastomeric
material with soft (polybutadiene) and hard (polystyrene) segments. The glassy

polystyrene phases act as physical cross—links that are lost when the copolymer is heated

above 109 °C, the glass transition of polystyrene. Thermoplastic elastomers are used in
automobile tires, footwear, and various automobile parts. Another example of a
thermoplastic elastomer is Spandex or Lycra, which consists of a urea (hard block) and
diamine (soft block).5
1.2.1 Amphiphilic materials

A similar combination of incompatible blocks is found in amphiphilic block
copolymers. The term amphiphilic describes molecules that love both aqueous and non-
aqueous (oil) phases. The water-loving phase of an amphiphilic material is hydrophilic,

8 Amphiphilic

while the non-aqueous (water-hating or oil-loving) phase is hydrophobic.
materials have the unique ability to form self-assembled aggregates (micelles) when
dissolved in aqueous solutions (Figure 1.6). The driving force for the formation of
micelles is the minimization of contact of the hydrophobic segments with the aqueous
phase. As the concentration of the amphiphilic molecules increases, the hydrophobic
portions of the molecules aggregate to form solutions that contain both non-aggregated
species as well as clusters of various sized aggregates. At high concentrations, networks
composed of interpenetrating clusters are formed. The reverse process can happen when
the amphiphile is exposed to an oil or lipophilic solvent. The hydrophilic blocks
associate to form micelles with the outer-shell of the micelle composed of the
hydrophobic block. The ability of amphiphilic block copolymers to form micelles has led
to the use of these materials for the transportation of dyes in printing technology as well

as drug carrier systems.8

Hydrophobic block
0—

Air-water

T interQ:
Aqueous
Hydrophilic block W/ solution

 

 

 

o oo

8 00 *K
000 * o/
Micelle

Figure 1.6. Structure of amphiphilic block copolymers and the formation of

micelles in aqueous solutions.

Amphiphilic block copolymers can act as surfactants (a compound that reduces
the interfacial tension between two liquids or a liquid and a solid) and therefore have
applications as detergents, emulsifiers, and dispersants. There are three major classes of
surfactants (Figure 1.7), which are named according to the type of hydrophilic group
they contain: ionic (both cationic and anionic), zwitterionic, and nonionic. Ionic
surfactants have charged hydrophilic groups such as carboxylate ions (example: soaps) or
protonated amines. Zwitterionic surfactants contain both positive and negatively charged
hydrophilic groups, while the hydrophilic groups in nonionic surfactants contain no
charges but consist mainly of hydroxyl groups or ethers. The ABA triblock copolymers
made from hydrophilic poly(ethy1ene oxide) (PEO) and hydrophobic polyfiaropylene
oxide) (PPO) (Figure 1.7) are common examples of amphiphilic block copolymers that
are used as nonionic surfactants. These surfactants are commercially available, and are

known as Pluronics, Polaxamers, or Synperonics. Numerous researchers have studied

'— u -— ——--

 

these amphiphilic polymers to determine their phase behavior in selective solvents (both
“water” and “oil”) for one of the blocks.“10

0

/\/\/\/\/\/\/\/\/u\c-Na+

Anionic Surfactant-
sodium stearate

‘6

 

O a soap
NM 0 0
/ l 0' /( VH0 o/
Zwitterionic Surfactant- Nonionic Surfactant-
ammoniohexanoate PEOx-PPOy-PEOx

“ Pluronic”

Figure 1.7. Types and examples of surfactants.

1.2.2 Synthesis and properties of (AB)n copolymers and ABA triblock oligomers

The structures of the (AB)n block copolymers and ABA triblock oligomers used
in this research are shown in Scheme 1.1. The (AB)n block copolymers are abbreviated
as [anCxEOy]n, where C)( corresponds to x methylene units, 1: represents the two carbon
atoms of the double bond, and EOy refers to y ethylene oxide repeat units. The ABA
triblock oligomers have a similar shorthand notation and are abbreviated as CXEOny. As
seen in Scheme 1.1, the A block represents a linear alkyl segment, or a linear alkyl
segment containing a single double bond and the B block corresponds to an oligomeric
segment of ethylene oxide. These polymers have been previously synthesized and have

”‘12 The polymers are amphiphilic in

been found to have very interesting properties.
character with the ethylene oxide segment (B block) being hydrophilic, and the alkyl

segment (A block) being hydrophobic. The dual nature of these polymers allows for the

10

practical application of these polymers as nonionic surfactants. Additionally because the
polymers contain PEO blocks, these polymers can be used as solid polymer electrolytes

when doped with various Li+ salts. '3

CH —CH=CH— CH —O CH CH O
“{( 2)x ( 2)x i 2 2 l7}:
[erthEOy]n

CH3-(CH2)X_1-o(CHZCHzo)9—(CH2)X_1CH3
cxeoycx

Scheme 1.1. Generic structure of (AB)n polymer and ABA model compounds used in
this research

The polymers were synthesized by Acyclic Diene Metathesis (ADMET)
polymerization (Scheme 1.2) of oligoethylene oxide a,o)-dialkenyl ethers using
Schrock’s molybdenum alkylidene catalyst. ADMET polymerization is a step growth
polymerization of dienes. The polymerization cycle involves the formation of
metallocyclobutane intermediates that lead to dimer, trimer, and eventually high
molecular weight polymer formation.l4 Ethylene is a volatile by-product of the reaction
and is easily removed under vacuum to drive the equilibrium of the reaction to favor the
condensation process. The most widely used catalysts for ring opening metathesis
(ROMP) and ADMET polymerizations are those developed by GrubbsIS and Schrockl6
(Figure 1.8). Grubb’s ruthenium catalysts and the molybdenum and tungsten catalysts of
Schrock have been widely used for the synthesis of unsaturated polyethers, esters,

carbonates, and other (AB)n polymer structures.

11

W0 H + TSOVOil-s

NaH, THF

Scheme 1.2. Synthesis of (AB)n multiblock copolymers.

PCy3 Y@/

N
C11,“ l F3C _ "
'RU— H3C>l/OOM_-
cu’ CF3
i=i+c+cr3
PCY3 30

Figure 1.8. Grubbs and Schrock’s metal alkylidene metathesis catalysts.

12

 

A series of exact length triblock ABA oligomers (Scheme 1.3) with the generic structure
H(CH2)X(OCH2CH2)y(CH2)xH, were used to help understand the crystallization and
solubility behavior of the (AB)n multiblock copolymers. The ABA triblock oligomers
were synthesized (Scheme 1.3) by treating an appropriate exact length oligoethylene
glycol with sodium hydride, followed by addition of two equivalents of the desired alkyl

bromide. Depending on the chain lengths of x and y, the products of the reaction ranged

from viscous oils to crystalline solids.”18
NaH
H(OCHZCH2)yOH + 2 CH3(CH2)X~1BT V CXEOny

Scheme 1.3. Synthesis of ABA triblock oligomers.

Exact length oligoethylene glycols were used in the synthesis of both the (AB)n
block copolymers and the ABA triblock oligomers to help increase the regularity and
crystallinity of the B block. If ethylene oxide chain lengths with an average value
(random distribution) were used, the packing of the chains would be irregular, causing a
decrease in the crystallinity of the block copolymer and triblock oligomers. In addition,
exact length ethylene oxide chains enables systematic studies of the evolution of the
physical properties of polymers and oligomer (ex: solubility) with changes in the length
of the ethylene oxide segment. Oligoethylene glycols are commercially available up to
six repeat units but longer exact length EOy chains need to be synthesized. A modified
version of the approach of Keegstra19 was used to prepare glycols where y is 6-10 and 14.

The route to these exact length oligoethylene glycols is outlined in Scheme 1.4.

13

""7

 

I‘-,- .¢ , .

TrCl

 

 

H(OCH2CH2)XOH (excess) > H(OCH2CH2)XOTr
pryidine
NaH/THF
Ts(OCH2CH2)yOTs
Tr(OCH2CH2)x+yOTr 4 Na(OCHZCH2)xOTr

 

50 atm H2 Pd/C

V

H(OCH2CH2)X+YOH

 

Scheme 1.4. Synthesis of exact length oligoethylene glycols.

The polymers and triblock oligomers were used in this research to study the effect
that inclusion complex formation with cyclodextrin has on their physical properties. CD
is known to thread on various hydrophilic and hydrophobic polymers, so the amphiphilic
nature of the (AB)n block copolymers and ABA oligomers could lead to interesting

structures when cyclodextrin preferentially threads onto one of the blocks.

14

1.3 Host-Guest chemistry

The terms “host” and “guest” are used to name complexing partners in synthetic
organic chemistry.20 A host is defined as an organic molecule or ion whose binding sites
converge in the complex, while a guest is a molecule whose binding sites diverge in the
complex.20 Complexing partners are not unique or original to organic chemistry. Metal
ions (H) and ligands (G) are complexing partners found in inorganic chemistry, while
enzymes (H) and substrates (G) are complexing partners found in biochemistry. Host-
Guest (H-G) complex chemistry is based on the ideals of the organizational ability found
between membranes and enzymes. In 1894 Emil Fischer proposed the Lock and Key
model (Figure 1.9) to describe the interactions between the active site of an enzyme and

a site-specific substrate.2| The shape of the reacting substrates and the enzyme’s active

1:3

substrate enzyme substrate-enzyme breakdown enzyme
complex of substrate

site fit together like a key that opens a lock.

 

 

 

 

 

.....

Figure 1.9. Emil Fischer’s Lock and Key Model for enzyme activity.21
The shape compatibility between enzymes and substrates led to the belief that
organic molecules can recognize one another in the same manner. Molecular recognition
is defined as the ability of a ligand (molecular receptor) to select and to bind a specific

20-22

substrate out of a variety of particles. Size and shape differences, as well as

noncovalent attractive forces such as electrostatic interactions, H-bonding, and

15

 

solvophobic forces can affect the recognition of substrate (guest) molecules. For
complexation to occur, binding sites and steric barriers in potential partners must be
complementary to one another in electronic character and geometric arrangement?"22

Organic molecules (Figure 1.10) that act as host molecules in complex syntheses
are crown ethers (l), cryptands (2), calixarenes (3), and cyclodextrins (4). The crown
ethers 1 first synthesized by Pedersen in 1967 are the most studied host molecules for
complex synthesis. The shorthand notation for naming crown ethers is n-Crown-m,
where n is the ring size (total number of atoms), and m is the number of oxygen atoms in

the ring. There are two methods for the synthesis of crown ethers as shown in the

synthesis of 18-Crown-6 (1) (Scheme 1.5).20’23‘24

 

Figure 1.10. Organic host molecules.

16

.Can-py

 

In route A, 1 is synthesized by a nucleophile-electrophile approach, where a diol reacts
with a bis(electrophile). The reaction requires a strong base to deprotonate the alcohols,
and the electrophile is usually a di-chloro, bromo, or tosylated compound. In route B, a
cation is used to catalyze the reaction by coordination with donor 0 atoms to form the
cyclized polyethers. This is called the “template effect” and demonstrates the ability of

crown ethers to make H-G complexes with various cations.

TsO/\/ O\/\ O/\/ OTS (\O/fi
+ KOH [:
_____>

Route B KOH

-K+
QTs
OT
(\Q‘ o (— 3 °~~. 5 ,o
O ~~~~ 125'” j ’ OWEWO ——* E x3163; j
I, s\\ k I X J O i O
Q o o '

L.o- H °\—/ W
\.
0%
Scheme 1.5. Synthesis of 18-Crown-6.20

Table 1.3. Comparison of various cation and crown ether cavity diameters.20

 

 

 

 

 

cam“ diat::::::l[A] cm“ “h" diatgzttei:l[AL
Li+ 1.20 12-Crown-4 1.2-1.5
Na+ 1.90 15-Crown-5 1.7-2.2
K+ 2.66 18-Crown-6 2.6-2.8
Cs+ 3.38 21-Crown-7 3.4-4.3

 

 

 

 

 

 

17

o ‘- AA“

 

The hydrophilic cavity of various sized crown ethers has been found to match the
diameters of a wide variety of alkali, and alkaline earth metal ions (Table 1.3).”25 The
complexation selectivity of crown ethers for various metal ions (and other H-G
complexes) is represented by the stability constant (Ks) (Equation 1.4), which is equal to
the rate of complex formation (kr) divided by the rate of decomplexation (kd). K5 is a
measure of the thermodynamic stability of the complex and can be determined by
changes in the physical properties of the host or guest molecule upon formation of the
complex. Solubility, UV-vis spectroscopy, and shifts in 1H and '3 C NMR resonances are

some of the methods that have been used to determine the Ks of a H-G complex.20‘23’25

 

kt”
Host + Guest ——>‘k [Host-Guest] Equation 1.3
d
KS = LHost-Guestl = (Kr). Equation 1.4
[Host] [Guest] (kd)

Cryptands (2, Figure 1.10) are macropolycyclic molecules that contain an
intramolecular cavity of 3-dimensional shape. Cryptands are named according to the
number of oxygen donor atoms that are present in the bridges.”2 1’24 Cryptand 2 has three
bridges that contain two oxygen atoms in each bridge and is named as a [2.2.2] cryptand.
Cryptands have enhanced stabilities over crown ethers for alkali, and alkaline earth
metals as shown in Table 1.4.2023‘24’26 The reason for the pronounced selectivity of
cryptands for metal cations is the size complementarily between the cation and the 3-

dimensional intramolecular cavity (termed “spherical recognition”), and the enhanced

18

 

number of electron pairs that contact the cation (eight heteroatoms versus six heteroatoms
for 18-Crown-6).

Table 1.4. Comparison of log KS values (in water solutions at 25 °C) of various metal
cations with organic host molecules 1 and 2.

 

 

 

 

 

 

Log Ks values of various H-G complexes in an aqueous solution
Host Li+ Na+ K+ Cs+
18-C-6 0 0.8 2.03 2.72
[2.2.2]
Cryptand 2.0 3.9 5.4 8.0

 

 

 

 

 

Crown ethers and cryptands are organic host molecules that form complexes with
small guests (Group 1 and other metals). These host molecules have established that
structural complementarity between the guest molecule and the receptor is the key factor
for accurate molecular recognition of the guest.23 Crown ethers and cryptands have been
used in various industrial applications as metal extraction agents or “scavengers”.24 The
ability to recognize large macromolecules plays an essential role in the building up of
supramolecular structures such as enzyme-substrate or antigen-antibody complexes, and
therefore host molecules that can recognize large organic guest molecules have received
much attention recently.2 1’22

Organic host molecules that can be involved in binding and recognition of small
to large organic guest molecules are calixarenes and cyclodextrins (3 and 4, Figure 1.10).
The name calixarene comes from the Greek work calix, meaning chalice (cup), and arene

indicating the incorporation of aromatic rings in the macroeyclic array. Calixarenes are

named by specifying the size of the macrocycle by a bracketed number inserted between

19

 

calix and arene and specifying the nature and position of substitution on the aromatic
rings.20 Calixarene 3 is named 5,11,17,23,30-pentamethy1-31,32,33,34,35-
pentahydroxycalix[5]arene according to this nomenclature. Calixarenes are formed from
the condensation of phenol with formaldehyde in the presence of base and usually
contain 4 to 8 aromatic rings.”24 Calixarenes, as their name states, are cup or tub-shaped
molecules, with a hydrophobic cavity that is able to incorporate organic guest molecules
(Figure 1.11). Complexes have been made with a variety of organic guest molecules
including haloforms, aromatics, nitromethane, and even C60 carbon fullerenes (a 1:1
complex with p—l‘ert-butylcalix[8]arene). The application of calixarenes are similar to
crown ethers and cryptands (as scavengers), but because calixarenes can recognize
organic molecules, the applications are more numerous. Beer and Chen27 have
synthesized substituted calixarenes and incorporated them into an electrode structure for
use as a sensor capable of detecting toxic vapours (PCBs) in air, while Nolan and
Diamond have developed sensors that are selective for lead and mercury, as well as

plutonium from nuclear wastes.28

 

Figure 1.11. Model of a calixarene host-guest complex.

20

 

Cyclodextrin (4) is an example of a host molecule that is able to include large
organic guest molecules. The most remarkable property of the cyclodextrins is their
ability to form H-G complexes with a variety of molecules. The one condition for
formation of a complex is that the guest molecule must fit entirely or at least partially into
the CD cavity.“29 The properties of H-G complexes with CD are discussed in the next
two sections and are directly related to the research in this manuscript.

1.3.1 Host-Guest complexes with cyclodextrin

Cyclodextrins (Figure 1.1) are unique host molecules because their shape (a
hollow truncated cone) allows guest molecules to be included inside the cavity of the CD
molecule (called an “inclusion complex” or IC).3 CD inclusion complexes are
synthesized in aqueous solutions and complex formation is driven by the attractive
hydrophobic interactions involved between the interior cavity of CD and the guest
molecule (Figure 1.12).2 In aqueous solutions, water molecules are trapped inside the
cyclodextrin cavity, which is energetically unfavorable due to polar-apolar interactions.
Guest molecules that are less polar than water associate with the hydrophobic cavity of
CD and release the water molecules into the bulk solution (a gain in entropy), and the
complex is stabilized by Coulombic, van der Waals, or H-bonding interactions between
the host and guest.2

ICs usually form in a 1:1 stoichiometric ratio of CDzGuest molecule, but other
ratios (2:1, 1:2) have been discovered.30 The stoichiometric ratio depends on the
structure of the guest molecule, the. concentration of the guest molecule in solution, as

well as the solvent used for complex formation. ICs are isolated as a stable crystalline

21

 

substance, but in solution an equilibrium is established between the complexed species

and the decomplexed species (represented by Equations 1.3 and 1.4).

 

000000000000 00 00 0000000000
00000 Oooooooooooooooogooo
000 CH. 000 0 000000 ooCH 000

1'

 

0 0°

00

2 "To 157'" 88
O

. o
oooio o o 0 0000000 00000
00 00000 0 0 000000 0000000000 0

Figure 1.12. A schematic representation of CD inclusion complex formation with p-
xylene as the guest molecule.2 In this illustration by Szejtli, the small circles represent
water molecules.

 

 

 

 

 

O
O I ‘0000

    

 

 

In the formation of ICs, guest molecules are arranged in the crystal lattice in
either a channel or caged structure.3 1'32 The solid-state crystal structure of ct-CD hydrate
is shown in Figure 1.13, and is an example of a cage-type crystal structure.32 There are
two types of cage structures that exist for CD inclusion complexes: herringbone and

brick-type (Figure 1-14). Research by Saengerzg'“:33

and Harata32 have shown that
depending on the size and structure of the guest molecules, different crystal structures are
formed. The crystal structure of oc-CD hydrate is an example of the herringbone cage-
type crystal structure. In this crystal structure CDs are packed crosswise so that the
interior cavity of CD is blocked on both sides by other CD molecules. Herringbone
crystal structures occur in CD complexes when the guest is a small molecule such as
methanol or propanol. Because small guest molecules are included inside the cavity of
CD and don’t extend beyond the dimensions of the CD cavity, the inclusion complex

crystal structure resembles that of uncomplexed CD (Figure 1.13). Brick-type cage

structures (b, Figure 1.14”) are produced from complexes with para-disubstituted

22

 

benzenes. The crystal packing is reminiscent of the offset layering of bricks in a wall,
where layers of CD molecules above and below block individual CD cavities. Harata
discovered that the more polar groups of a number of 1,4-substituted aromatic
compounds are located near the wide rim of the CD torus (2° OH groups), while nonpolar
groups are directly inserted through the CD cavity (near narrow rim, 1° OH groups).32
Besides p-disubstituted aromatic compounds, a-CD complexes with N,N-
dimethylforrnamide, and 2-pyrrolidone have been discovered to have a layered type of
crystal structure.32 The space group for both the herringbone and brick-caged crystal
structure is D2}, and orthorhombic. The unit cell is primitive (P) and is defined as P212121

by the International Tables of X-ray Crystallography.

 

Figure 1.13. X-ray crystal structure of a-CD hexahydrate viewed along the a axis.32

Dark circles represent water molecules in the crystal structure.

23

Lt,”

 

 

 

”V,

 

1., o
a was“...

 

at.
its.

..\

 

 

 

-CD complexed with iodine

(1

Crystal structures of (a) dimethyl
(herringbone-type cage) and (b) a-CD complexed with p-nitrophenol (brick-type cage).32

Both crystal structures are viewed along the a axis.

Figure 1.14.

24

As was shown in the cage-type crystal structure, the crystal structure (herringbone
or brick) is dependent on the structure and size of the guest, which is also true in
distinguishing between channel-type and cage-type crystal structures. Small molecular
guests (< 5 carbon atoms in a straight chain) crystallize in cage structures whereas
organic guest molecules larger than five carbon atoms long, form channel-type crystal
structures.”32 In channel-type complexes (Figure 1.15), CDs are stacked head-to-tail or
head-to-head to form columnar crystals in which the guest molecules are embedded

inside the cavity of CD.

 

Figure 1.15. Stereodrawing of the structure of a-CD complexed with an iodine-
iodide lithium salt (channel-type crystal structure).32

25

 

X-ray crystal structures of inclusion complexes provide important information on
the three-dimensional structure of H-G complexes as well as information on the
noncovalent interactions involved in complex formation. The importance of this
knowledge has led to the interest in synthesizing other inclusion complexes with CD.
The relative ease of complex formation is also responsible for the explosion in the
number of CD inclusion complexes. The belief that CD can recognize differences in
organic molecules has drawn attention to the use of CD for the self-assembly of
nanoscale structures. Molecular machines, sensors, and artificial receptors are some of
the proposed uses for these nanoscale structures. Cyclodextrin is also able to form H-G
complexes with macromolecular compounds (polymers) and the properties of these
complexes are discussed in the next section.

1.3.2 Polymeric inclusion complexes

Since the development of host/guest chemistry scientists began to look at Nature
for their next inspirations in the development of supramolecular complexes (association
of two or more chemical species held together by intermolecular forces)?"22 Dendrimers
and C60 (fullerene) are some examples of chemical species that mimic those structures
found in Nature or architectural design.34 lnterlocked and intertwined structures are
examples of supramolecular complexes that have gained much interest not only for their
beauty but also because of their possible applications as nanoscale machines.

When an acyclic molecule is inserted inside the cavity of a macrocyclic molecule
the supramolecular complex is termed a pseudorotaxane (I, Figure 1.16).”37

Pseudorotaxanes are in equilibrium with their uncomplexed state. The attachment of

bulky groups to the ends of the acyclic molecule prevents dissociation into its free

26

 

components. Such molecules are termed rotaxanes (11, Figure 1.16), from the Latin rota

meaning wheel, and axis meaning axle.
—

+- - 21x Ild'l’
0 .—_~» o- =- vo-
I [I

Figure 1.16. Diagram of the reversible formation of a pseudorotaxane (I) from its
acyclic and cyclic components, and the irreversible conversion into a rotaxane (II).

Pseudorotaxanes form by threading the macrocycle onto the acyclic polymer
chain in aqueous solutions. The free and bound states of these mechanically interlocked
complexes have different physical properties, which has led to studying these compounds
as models for self-assembly in nanoscale structures. The driving force for the formation
of these supramolecules are noncovalent attractive forces (which are found to be
responsible for the formation of highly organized biological systems such as membranes
and ribososmes), so these molecules can be used as models for the investigation of
complex biological entities.3

Since the early 1990’s it has been known that cyclodextrins thread onto polymer
chains to form supramolecular inclusion complexes such as polypseudorotaxanes (III)
and polyrotaxanes (IV).3-‘35'37 In these structures many CDs are located on the polymer

backbone structure to form a type of molecular abacus.

69an- 091919-

11
III IV

 

Figure 1.17. Diagrams of polypseudorotaxanes (III) and polyrotaxanes (IV).

Harada et al. was the first to report the complex formation between CD and a
polymer, an IC formed from a-CD and poly(ethy1ene glycol) (PEG).38’39 The complex
formed upon the addition of PEG to a saturated aqueous solution of a-CD. After stirring
the solution overnight, Harada et a1. collected a white solid precipitate (inclusion
complex) by filtration. Numerous experiments were performed on the synthesized
inclusion complexes to confirm their composition and properties. lH NMR showed a 2:1
ratio of the proton resonances from a-CD and PEG, corresponding to a stoichiometry of
two ethylene glycol units per Ot-CD. X—ray powder diffraction of the complexes shows
that they are crystalline and the patterns correspond to a column or channel crystal
structure. '3 C CP/MAS NMR also confirmed that a polymer chain was threaded through
the CD cavity. Figure 1.18 is a proposed model by Harada36 showing the complex
formed between a-CD and PEG. Harada has also observed that complex formation was
selective with only a-CD forming a complex with PEG. B-CD did not form a complex
with PEG because its interior cavity is too large to establish strong interactions with the
acyclic polymer. Harada has since made inclusion complexes with various polymers
including poly(isobutylene)4O (PIB), poly(s-caprolactone)“ (PCL), and most recently

poly(dimethylsiloxane).42

mmmm

_/ ‘ ’
’1} .- "”\ ' n’ I, I, \/P‘ / T‘
«x; /“‘? I' \/ \. L/ \ ’ .I’""'“\..r ii." 1.5/‘\ ”(x j../\ P\

.{ i. _4 i: . ‘J l

QMHU

Figure l. 18 Proposed superstructure for PEG- --01 -CD complexes as illustrated by
Harada.36

\

‘,/ {f I:

 

28

The combination of CD’S affinity to bind almost an infinite number of guest
molecules with its ability to form self-assembled supramolecular structures has provided
numerous research possibilities and industrial applications. Recently Hayakawa et a].43
synthesized nanotubes consisting of CD molecules. Nanotubes are fine capillaries with
an inside diameter of nm order which have attracted great interest for their use as sites for
molecular recognition. The nanotube was prepared by threading CD onto a polymer
chain (PEG-BA) with bulky ends to form a polyrotaxane. A molecular tube was then
formed by cross-linking adjacent CD units in the polyrotaxane structure, and when the
bulky end groups were removed, the nanotube unthreaded from the polymer chain. The
synthesized nanotube was studied as a host for the reversible binding of small molecules
such as iodine.

The synthesis of supramolecular structures that contain molecularly imprinted
polymers is another example where the complexing ability of CDs is used. Komiyama et
al.44 designed polymeric receptors for cholesterol by crosslinking B-CD with
diisocyanates in the presence of cholesterol as a template. Initial studies showed that
70% of the cholesterol in a water-THF mixture was absorbed by the templated polymer
receptor. Nicholls et a].45 developed a method to synthesize synthetic receptors for
water-soluble substances. In Nicholls’ approach, B-CD was used as the receptor for
effective recognition of D and L-phenylalanine in aqueous solutions.

The above examples show the versatility of polymer inclusion complexes with
CD. The understanding of the interactions involved in the self-assembly of
supramolecular structures is important for the efficient design and recognition of guest

molecules. The research presented in this thesis gives information on the ability of CD to

29

form complexes with a series of amphiphilic (AB)n block copolymers. Possible molecular
recognition of CD for one of the polymer blocks, as well as changes in the physical

properties of both reactants are discussed.

30

1.4

(1)

(2)
(3)

(4)
(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

(13)

(14)

(15)

(16)

References

Bender, M., Komiyama, M. C yclodextrin Chemistry; Springer-Verlag: New York,
1978.

Szejtli, .1. Chemical Reviews 1998, 98, 1743-1753.

Wenz, G. Angewandte Chemie-International Edition in English 1994, 33, 803-
822.

Hedges, A. R. Chemical Reviews 1998, 98, 2035-2044.

Odian, G. Principles of Polymerization. 2nd Ed.; John Wiley & Sons: New York,
1981.

Stroble, G. R. The Physics of Polymers: Concepts for Understanding Their
Structures and Behavior; 2nd ed.; Springer-Verlag: New York, 1997.

Seymour, R. B.; Carraher, C. E., Jr. Polymer Chemistry. An Introduction; Marcel
Dekker: New York, 1981.

Porter, M. R. Handbook of Surfactants, lSt Ed.; Chapman and Hall: New York,
1991.

Nace, V. M.; Editor Nonionic Surfactants: Polyoxyalkylene Block Copolymers;
Marcel Dekker, Inc.: New York, 1996.

Alexandridis, P.; Lindman, B.; Editors Amphiphilic Block Copolymers: Self-
Assembly and Applications.; Elsevier Science: Amsterdam, 2000.

Qiao, 1.; Chen, Y. Y.; Baker, G. L. Chemistry ofMaterials 1999, 11, 2542-2547.
Chen, Y. Y.; Ph.D. Dissertation. Michigan State University, 1999.

Swan, K. A.; Stowe, M. K.; Chen, Y.; Qiao, J.; Baker, G. L. Polymer Preprints
2000, 41(2), 1365.

Wagener, K. B.; Boncella, J. M.; Nel, J. G. Macromolecules 1991, 24, 2649-57.

Schwab, P.; Grubbs, R. H.; Ziller, J. W. Journal of the American Chemical
Society 1996, 118, 100-10.

Schrock, R. R.; Murdzek, J. S.; Bazan, G. C.; Robbins, J.; DiMare, M.; O'Regan,
M. Journal of the American Chemical Society 1990, 112, 3875-86.

31

(17)
(18)

(19)

(20)

(21)

(22)

(23)

(24)

(25)

(26)

(27)

(28)

(29)

(30)

(31)

(32)

(33)

Chen, Y. Y.; Baker, G. L. Journal of Organic Chemistry 1999, 64, 6870-6873.

Chen, Y. Y.; Baker, G. L.; Ding, Y. Q.; Rabolt, J. F. Journal of the American
Chemical Society 1999, 121, 6962-6963.

Keegstra, E. D.; Zwikker J. W.; Roest, M. R.; Jenneskens, L. W. Journal of
Organic Chemistry 1992, 57, 6678-6680.

Host Guest Complex Chemistry: Macrocycles: Synthesis, Structures,
Applications; Vogtle, F. W., E, Ed.; Springer-Verlag: New York, 1985.

Supramolecular Chemistry: Concepts and Perspectives; Lehn, J.-M., Editor.;
VCH Publishers: New York, 1995.

Tsoucaris, G.; Editor Current Challenges on Large Supramolecular Assemblies;
Kluwer Academic Publishers: Boston, 1999.

Crown Compounds: Toward Future Applications; Cooper, S. R., Editor.; VCH
Publishers: New York, 1992.

Constable, E. C. Coordination Chemistry of Macrocyclic Compounds; Oxford
University Press: New York, 1999.

Crown Ethers and Analogous Compounds; Hiraoka, M., Ed.; Elsevier Science:
New York, 1992; Vol. 45.

Crown Ethers and Analogs; Patai, S. R., Zvi, Editors; John Wiley & Sons: New
York, 1989.

Beer, P.; Shade, M.; Chen, Z.; PCT International Applications: Patent No.
WO9511449, 1995.

Diamond, D.; Nolan, K. Analytical Chemistry 2001, 73, 22A-29A.

Saenger, W. Angewandte Chemie-International Edition in English 1980, 19, 344-
362.

Saenger, W.; Steiner, T. Acta Crystallographica Section A 1998, 54, 798-805.

Saenger, W. R.; Jacob, J.; Gessler, K.; Steiner, T.; Hoffmann, D.; Sanbe, H.;
Koizumi, K; Smith, S. M.; Takaha, T. Chemical Reviews 1998, 98, 1787-1802.

Harata, K. Chemical Reviews 1998, 98, 1803-1827.

Saenger, W.; McMullan, R. K.; Fayos, J.; Mootz, D. Acta Crystallography, Sect.
B 1974, B30, 2019-28.

32

(34)

(35)

(36)
(37)
(38)
(39)

(40)

(41)

(42)

(43)

(44)

(45)

Amabilino, D. B.; Stoddart, J. F. Chemical Reviews 1995, 95, 2725-829.

Philp, D.; Stoddart, J. F. Angewandte Chemie-International Edition in English
1996, 35, 1155-1196.

Nepogodiev, S. A.; Stoddart, J. F. Chemical Reviews 1998, 98, 1959-1976.
Raymo, F. M.; Stoddart, J. F. Chemical Reviews 1999, 99, 1643-1663.
Harada, A.; Kamachi, M. Macromolecules 1990, 23, 2821-2823.

Harada, A.; Li, J .; Kamachi, M. Macromolecules 1993, 26, 5698-5703.

Harada, A.; Suzuki, 8.; Okada, M.; Kamachi, M. Macromolecules 1996, 29,
5611-5614.

Kawaguchi, Y.; Nishiyama, T.; Okada, M.; Kamachi, M.; Harada, A.
Macromolecules 2000, 33, 4472-4477.

Okumura, H.; Okada, M.; Kawaguchi, Y.; Harada, A. Macromolecules 2000, 33,
4297-4298.

Ikeda, E.; Okumura, Y.; Shimomura, T.; Ito, K.; Hayakawa, R. Journal of
Chemical Physics 2000, 112, 4321-4325.

Asanuma, H.; Kakazu, M.; Shibata, M.; Hishiya, T.; Komiyama, M. Chemical
Communications 1997, 1971 - 1 972.

Piletsky, S. A.; Andersson, H. S.; Nicholls, I. A. Macromolecules 1999, 32, 633-
636.

33

CHAPTER 2

Synthesis and Characterization of Inclusion Complexes of a-CD and (AB).. Block
Copolymers

2.1 Acronyms for compounds used in this research

The polymers and triblock oligomers used in this research contain two segments,
an ethylene oxide chain and either an n-alkyl or an n-alkyl chain that contains a single
double bond. For simplicity, the nomenclature system used to refer to these compounds is

similar to that of PEO/PE surfactants. (AB)n block copolymers are abbreviated as
[CxtthEOy]n, Where Cx corresponds to x methylene units, it represents the two carbon
atoms of the double bond, and EOy refers to y ethylene oxide repeat units. ABA triblock

oligomers are abbreviated as CXEOny and follow the same scheme used to represent the
block copolymers. a-Cyclodextrin is abbreviated as a-CD. Inclusion complexes (ICs)
with (AB)n block copolymers and ABA triblock oligomers are abbreviated as (AB)n-a-
CD or ABA-a-CD ICs respectively. Specific polymer or triblock oligomer inclusion
complexes use the surfactant nomenclature along with a-CD. Examples of the naming

protocol are shown in Scheme 2.1.

—{(CH2)x—CH=CH—(CH2);o(CHZCHzo);+—n
[CxtthEQy]n

[C3flC3EOgln-a-CD

C12E03C12-a-CD
CH3—(CH2)X_1-o(CHZCHzo)y—(CH2)X.1CH3

cxeoycx

 

Scheme 2.1. Nomenclature of (AB)n block copolymers, ABA model compounds, and
inclusion complexes formed with a-CD. A proposed structure of a (AB)n-a-CD complex
is also shown.

34

2.2 Results
2.2.1 Complex formation between [CxtthEOy]n polymers and oc-CD.

Inclusion complexes of a-CD and [CxtthEOy]n polymers were prepared at room
temperature by adding 2 mL of a 20 wt% solution of a-CD in water to a solution of the
(AB)n block copolymer in 0.5 mL methanol. Inclusion complexes were synthesized at
various polymer concentrations to determine the stoichiometry of the complexes (see
below). For these experiments, polymer concentrations (2-40 mg/mL) were calculated by
dividing the weight of the polymer sample (in milligrams) by the total volume of solution
(water + methanol = 2.5 mLs). For low polymer concentrations (< 8 mg/mL), the oc-CD
and polymer solutions became turbid on mixing, and the complex precipitated as white
crystals which were collected by filtration and washed with water to remove
uncomplexed a-CD. Uncomplexed polymer formed a milky suspension that passed
through the filter paper, and a control experiment run in the absence of oc-CD showed that
98% of the polymer could be recovered from the filtrate. Thus, the polymer/a-CD
complexes reported here are not contaminated by free polymer. All complexes were
dried overnight in a vacuum oven (150°C) to remove residual water from the complex.
Higher polymer concentrations (> 12 mg/mL) led to formation of awhite gel, which was
washed to remove uncomplexed polymer and a-CD and isolated as a crystalline IC by
freeze-drying. The gel formed more rapidly at higher polymer concentrations. Our
observations are in agreement with those of Baglioni et al. , who reported the formation of
white solid gels during the synthesis of ICs from (x-CD and PEG in aqueous solutions."2
He also described the threading process for the formation of these complexes, and the

effects that temperature and composition have on complex formation. Table 2.1 lists the

35

(AB), block copolymers that were used in the synthesis of ICs with a-CD. The
characteristics of the syntheses of ICs from a-CD and [CxtthEOy]n polymers were all

similar in their formation.

Table 2.1 Characteristics of (AB), block copolymers used in IC formation with a-CD.

 

 

 

 

 

 

(AB), block copolymer Mn'(g/mol) Concentrations of polymer (mg/mL) in
solution
[C31tC3EO6],, 25,000 2, 4, 8, 12, and 16
[C31tC3E07],, 26,000 2, 4, 8, 12, 16, and 20
[C31tC3EOg],, 47,000 2, 4, 8, 10, 12, 16, 20, 24, and 40
[CanC4EOg],, 19,000 2, 4, 8, 10, 12, 16, 20, 24, and 40

 

 

 

1 Molecular weight data for (AB), polymers was obtained by Chen.3

The yield of the [CxtthEOy],,-01-CD ICs, defined as (mass of the isolated
complex)/(mass of on-CD + mass of polymer) are shown in Figure 2.1. The yields
increase linearly with the concentration of polymer used in the synthesis, and saturation
of the yield was observed in all cases. The yields of complexes at different
concentrations were repeatable within 2%. The saturation point corresponds to the
polymer concentrations where gelation of the polymer-CD solution was observed,
(generally > 12 mg/mL), and depends on the chemical structure of the (AB), polymer.
Similar saturation effects were seen during the synthesis of inclusion complexes between
a-CD and PEO4 and PCL.5 The explanation given for saturation of the yield is that the
complexes have a characteristic stoichiometry in terms of the number of CDs threaded

per polymer repeat unit. In ICs with PEO, one a-CD molecule is threaded for every two

ethylene oxide units, while PCL-a-CD complexes have a 1:1 stoichiometry. The

36

 

stoichiometry of (AB),-0t-CD le was determined by two different methods and is

discussed in the next section.

 

 

 

 

 

70
A
9 A
60
’o‘ I I -
25 50
x
2
o.
E 40
o
0
«5 30
2
o
5'. 20
10
O l l l l l l
0 4 8 12 16 20 24

Concentration of Polymer in mglmL

Figure 2.1. Yield of [anCxEOy],,-a-CD inclusion complexes as a function of polymer
concentration. ( I [C31tC3EOg],,-01-CD, A [CartCaEOg].,-a-CD, C [C31tC3EO6],,-a-CD)

37

2.2.2 Properties of (AB),-01-CD complexes

1) Thermal behavior

Thermal analysis was used to confirm formation of (AB),-cx-CDs complexes.
Representative Differential Scanning Calorimetry (DSC) and Thermal Gravimetric
Analysis (TGA) data for a-CD, [CxtthEOy],,, and the [anCxEOy],,—01-CD ICs are shown
in Figures 2.2 and 2.3-2.4 respectively. The data for complexes synthesized from the
other polymers are similar. These results are similar to those observed by Harada and
Tonelli who have synthesized ICs with a-CD and PEG."6'8 Figure 2.2 shows DSC scans
for the polymer [C37tC3EOb],, (bottom trace), and the [C37IC3E06]n-0.-CD inclusion
complex (top trace). All samples were heated to 150°C to erase their thermal history and
flash-cooled to —100°C. The scans shown are the second heating scans, run at 10°C/min.
under He, and normalized with respect to sample mass. The DSC scan of the pure
polymer has a well-defined Tg as well as a complicated sequence of exo and endotherms
associated with crystalline transitions in the polymer. These transitions are barely visible
in the scan of the IC, indicating that the amount of free polymer in the complex is small.
Integration of the melting transition leads to an estimate of about 10% free polymer in the
sample. Integration of the initial heating scan (from 25 to 150°C) indicates only 1-2%
free polymer in the complex sample, and thus most of the free polymer seen in the second
scan must result from partial dethreading during the scan used to erase the thermal
history. As a control, we de-threaded an IC by stirring the complex in THF. The solution
was filtered to remove the insoluble a-CD, concentrated, and dried under vacuum. The
DSC scan of the dethreaded product shows the characteristic melting and crystallization

peaks of the free polymer.

38

 

 

 

 

 

 

 

 

 

 

 

a
0 lb -
'U
C
E
3
.9
LL
a
CB
0)
I C J

F

-100 -50 0 50

Temperature (°C)

Figure 2.2. Normalized DSC thermograms of a) [C31tC3EO6],,-a-CD IC, b) the de-
threaded complex and c) the pure [C31tC3E06],, homopolymer. All scans were run at
10°C/min under helium.

39

Shown in Figure 2.3 are TGA data for ct-CD, [C47TC4EOg]n, and the
[C41tC4EOg],,-a-CD 1C. a-CD shows an onset for weight loss at 310°C, while the
polymer (trace a) has an onset for weight loss at 360°C. Both Ot-CD and the polymer
show small residues. The TGA data for the [CanC4EOg]n-a-CD IC is more complex.
The onset for decomposition is 340°C, between that of a-CD and the pure polymer, and
the decomposition profile shows a rapid weight loss to ~25% of the original weight,
followed by completion of the decomposition process at a higher temperature. Because
the DSC results imply that the ICs contain little free polymer, we surmise that the second
step in the TGA curve for the polymer-CD IC results from dethreading of the [C during
the TGA run.

For controls, physical mixtures of a-CD and (AB), polymers were also
characterized by TGA and DSC. The TGA data for a physical mixture of [C3nC3EOé],,
and oc-CD (trace c in Figure 2.4) show a different decomposition profile than that of the
synthesized [C37tC3EOé]n-01-CD IC. The TGA scan of the mixture clearly shows two
distinct decomposition profiles associated with CD and the (AB), block copolymer. The
inclusion complex’s profile shows an increase in the onset for decomposition over a-CD
and the physical mixture. The DSC data of a physical mixture of a-CD and [C37tC3EO7],,
(data not shown) has barely observable thermal transitions associated with the free
polymer, and look like previous scans of synthesized inclusion complexes (Figure 2.2,
trace a). Although the DSC scans of synthesized (AB),-0t-CD ICs look similar to a

physical mixture of the two components, the dethreading experiment proves complex

formation occurs and the thermal properties of ICs are quite different than their

individual components.

40

 

 

 

 

 

 

 

100 ' ‘Q— -- V. a
C
80 -
°\° .-
E 60 -
.2’
w n
3
4o -
20 -
0 n l I l 1 l l I
0 100 200 300‘ 400

Temperature (°C)

Figure 2.3. TGA scans for a) [C41tC4EOg],,, b) [C41tC4EOg],,-a-CD IC, and c) or-
CD (the sample was dried for 12 h at 100°C at the beginning of the run.) Samples
were heated at 10°C/min. under a flow of nitrogen.

41

Weight %

 

 

 

 

 

 

 

100 - M 1-11.1.0
b
a
80 r
60 P
40 -
20 -
0 j I 1 I 1 I n I
O 100 200 300 400

Temperature (°C)

Figure 2.4. TGA scans for a) [C31tC3E06],, + a-CD mixture, b) [C31tC3E06],,-a-
CD IC, and c) a-CD. Samples were heated at 10°C/min. under a flow of nitrogen.

42

2) Solid state properties of ICs

Powder x-ray diffraction and solid-state 13 C CP/MAS NMR was used to
determine the structure of the inclusion complexes as well as confirm complex formation.
Diffraction data from inclusion complexes with known crystal structures were compared
with those from (AB),-0t-CD ICs. The propanol-a-CD and octanol-a-CD inclusion

‘2 are known to have caged (herringbone) and column (channel) crystal

complexesg’
structures respectively, and the XRD patterns of these complexes were used to identify
the structures of the (AB),-a-CD inclusion complexes. Figure 2.5 shows the x-ray
powder patterns for a) the propanol-a-CD complex, b) [C31tC3EO6],,-0L-CD, c)
C14EOgC14-a-CD, and d) the octanol-a-CD IC. The packing of the propanol-a-CD caged
type crystal structure is orthorhombic with space group P212121 (similar to crystal
structure of a-CD hexahydrate, Figure 1.13).l3 The cell dimensions are: a = 14.292 A, b
= 37.515 A, c = 9.393 A. The octanol-a-CD IC shows the x-ray diffraction pattern
associated with the channel type crystal structure of a-CD guest molecules. The channels
are packed in a hexagonal unit cell and show characteristic Bragg diffraction peaks at l,
\/3, and \17 corresponding to 20 = 7.45°, 12.97°, 19.94°, and 22.65°. These peaks have
been assigned to be 100, 110, and 210 reflections by Topchieva in PEG-a-CD le.6’7""'l7
The prominent peak at 200° (210 reflection) has been reported to be diagnostic for the
formation of an inclusion compound inside the channel of the CD. A comparison of the
diffraction patterns of the known inclusion complexes (octanol and propanol-a-CD)
shows that the (AB), ICs are crystalline and are consistent with a columnar structure.
These results are similar to the crystal structure of other polymeric inclusion complexes

(PCL, PIB, and PDXL)5 ‘16’18'20 with cyclodextrin. Recently, Ripmeester and coworkers

43

obtained a single crystal X-ray structure of PEG-B-CD 1C.” The crystal of this complex
was monoclinic, with a C2 space group and unit cell parameters a = 18.726 A, b = 24.475 ,

A, c = 15.398 A.

 

 

 

 

 

 

 

d W

T I l I l I

2 6 10 14 18 22 26 3O
29

Figure 2.5. Powder X-ray diffraction patterns for a-CD complexes. a) 1-
propanol-oc-CD, b) [C31tC3E06],,-a-CD, c) C14E03C14-a-CD, and d) octanol-a-
CD. Data were collected at a rate of 20 = 2°/min over the range 20 = 0-40°.

 

 

 

44

Solid-state NMR data also confirmed that polymer chains were threaded through
the cyclodextrin cavity. Solid-state NMR probes the structure of the complex at the
molecular level and is more sensitive to didordered molecular architectures than powder
X-ray diffraction. Figure 2.6 shows the 13C CP/MAS NMR spectra of the a-CD
complexes of [C3nC3EOg],, and CleOiaClz, a model compound for the (AB), system.
Spectra of oc-CD, [C37tC3EOg],,, and CleOMClz are shown for comparison. In the
uncomplexed state, a-CD exists in an unusual high-energy linkage conformation because
one of the glucose residues is rotated toward the axis of the CD torus. This rotation
causes the CD molecule to exist in a partially “collapsed” or distorted conformationzz'24
In this high-energy conformation a-CD is non-symmetrical and thus, the 13C CP/MAS
NMR spectrum of a-CD (spectrum a in Figure 2.6) shows multiple resonances for each
type of carbon in a-CD. Upon complexation with guest molecules, the a-CD macrocycle
expands to adopt a nearly symmetrical hexagonal structure, thereby relieving the
distorted conformation, which allows each glucose unit of a-CD to exist in a similar
environment. '3 C CP/MAS NMR spectrum were taken of the individual (AB), block
copolymer (spectrum e) and the ABA triblock oligomer (spectrum c) to probe for
chemical shift changes associated with complexation with ct-CD. In the spectrum of
[anC3EOg]n, resonances at ~70 ppm correspond to the carbon atoms of the ethylene
oxide block of the polymer, while resonances at ~30 ppm are the methylene carbons of
the alkylene block. The spectrum of the triblock oligomer CrzEOMClz show similar
resonances plus a peak at ~10 ppm from the methyl group of the alkyl blocks. The spectra
of the [C31tC3EOg],,-01-CD (spectrum (I) and C12E014C12-a-CD (spectrum b) ICs show

only a broad resonance for each type of carbon atom in the glucose structure, a change

45

that is consistent with threading of the polymer chain through the CD cavity. These
results are consistent with the results of X-ray studies of single crystals of a-CD

performed by McMullan and Takeo.9’25

 

 

 

 

‘fl—ET111l'Yll’lllTi.11T111WWWWWT

120 110 100 90 80 7E» 60 50 4o 30 20 10
ppm

WTT—TT—TT—FT—VTTITfi—ffli .1 'I‘TTI—Y‘Tf—T— 1
1

Figure 2.6. ‘3 C CP/MAS NMR spectra of a) a-CD, b) C12E014C12-01-CD, c)C12EOr4C12,
d) [C3nC3EOg]n-a-CD, and e)[C31tC3EOg],,. Arrows indicate ’3 C resonances for the alkyl
chain of the ABA oligomer that shift when complexed with a-CD. Spectra were taken at
a spinning rate of 4.0 KHz.

46

3) Stoichiometry of complexes

To determine the stoichiometry (number of CD’s threaded per polymer repeat
unit) of [anCxEOy],,-0L-CD ICs, continuous variation plots were made and the results
were compared to stoichiometries determined from 1H NMR spectra of the complexes
measured in DMSO‘df). Harada et al. has used this method to determine the
stoichiometry of CD complexes of PEO,4 PCL,5 and poly(dimethylsiloxane”). The
continuous variation method (also known as Job’s method) is a simple and effective
method to determine stoichiometry. Given the chemical reaction:

aA + b8 2 cC Equation 2.1
the reaction can be rewritten as:

A + kB 2 mC Equation 2.2
where k = b/a and m = c/a. If a series of solutions is prepared, each containing the same
total number of moles of reactants A (or-CD) and B ([anCxEOyh), but with a different
mole ratio, R, of B to A, the maximum yield of product C ([anCxEOyIn-a-CD) is
obtained when R = k (the stoichiometric ratio of B to A). A plot of the product yield
versus R has a maximum value at the initially unknown value of k.

The method of continuous variation can be justified mathematically through the
following procedure. If “x” represent the moles of A in a particular solution, and the total
moles of A and B are fixed at 1.0 throughout a series of solutions, then for each solution x
= moles of A, and l-x = moles of B. The maximum yield of product is obtained when R
= moles B/moles A = (1-x)/x, which is equal to k. From Equation 2.2, if x is less than
the stoichiometric amount of A, then A is the limiting reagent and the moles of the

product = mx. A plot of the moles of product versus x over a series of solutions should be

47

linear, with slope m. If x exceeds the stoichiometric amount of A, then B is the limiting
reagent and moles of the product = m(1-x)/k. A plot of moles product versus x over a
series of solutions should be linear, with slope = -m/k. The first plot proceeds up to the
right as x increases, while the second plot proceeds down to the right as x increases. At
the intersection of these two lines: mx = m(l-x)/k, which when solved for x gives: x =
l/(l+k). Substitution for x in the expression of R = (1-x)/x = { 1 — [1/(l+k)]} / [1/(1+k)]
= k, which demonstrates that the maximum product is obtained when R = k.

Figure 2.7 shows continuous variation plots for the same series of [CxtthEOy],,
polymers found in Figure 2.1, with the yield of complex plotted as a function of the mole
fraction of a-CD. The sum of the reactants was fixed at 1.8 x 10" mole (based on the
molecular weight of the polymer repeat unit). All of the plots show maximum yields
when the molar ratio of a-CD is ~0.70 suggesting that the stoichiometries of the
complexes are ~2.3 :1 (oz-CD : polymer repeat unit).

The stoichiometries obtained from the continuous variation plots agree with those
calculated from 1H NMR spectra of the complexes in DMSO-d6. All (AB),-01-CD
complexes are soluble in DMSO and de-thread in solution. Thus, as shown in the 300
MHz ‘H NMR spectra of OL-CD, [C4712C4E03]n, and the de-threaded [CanC4EOg],,-a-CD
inclusion complex (Figure 2.8), the spectra only show resonances for a-CD and the
polymer. The stoichiometry of the complex was calculated by comparing the integration
values for H-l of a-CD (see structure in Figure 2.8) and the —CH2 (signal 4, or to the
C=C bond) of the polymer repeat unit. H-l of a-CD integrates for six protons while the

methylene signal integrates for four protons in the uncomplexed state. In the 1H NMR of

48

the complex, the same methylene signal integrates for only 1.7 protons, which

corresponds to (4.0/1.17) = 2.3 a-CD ‘s threaded per polymer AB repeat unit.

 

 

 

 

 

 

180
160 -
A 140 -
‘e”
V 120 -
X
2
9- 100 -
E
O
0 8O -
H—
O
E 60 -
.2
>- 40 -
20 '-
0 r J I I I l J
0 0.2 0.4 0.6

Mol fraction a-CD

Figure 2.7. Continuous variation plot showing the yield of a-CD/(AB), block
copolymer complexes obtained for various ratios of reactants. The sum of the reactants
was fixed at 1.8 x 104 mole (based on the molecular weight of the polymer repeat unit).

(I[C31tC3EOg],,, A [CanCaEOg]n, o [C31tC3E06].,).

49

 

 

 

 

 

 

 

 

H-1
3 L
_
r..- 1.":
U
b m‘ JL f ,
2.3
,2 4 o 4 13.31
s . ° «8 l—Jg. .______.
1

 

 

5.0 4.0 ppm 3.0 2.0

Figure 2.8. The 300 MHz 1H NMR spectrum of a) a-CD, b) the [C4fiC4EOgln-a-CD IC,
and c) [C41tC4EOg],, in DMSO-d6. The IC stoichiometry was calculated by comparing the

integration ratios of H-l of a-CD (6 H) to the -CH2 units (signals 2-4) in the polymer
repeat unit.

50

2.2.3 Complex formation between CxEOyCx model compounds and a-CD
Complexes synthesized from a series of ABA triblock oligomers (Table 2.2)
were used as model compounds for understanding the behavior of polymer ICs. In
particular, CXEOny-a-CD ICs were synthesized to explore how differences in the
hydrophobicity of the model compounds (increased alkyl length or decreased ethylene

oxide chain length) affect yield and stoichiometry of the complexes.

Table 2.2. Characteristics of ABA triblock oligomers used to form ICs with a-CD.

 

 

 

 

 

 

 

 

 

 

 

ABA triblock MW (g/mol) Concentrations of CXEOyC, (mglmL)
C12E03C12 486.8 3.3, 6.6, 10, 15, 20, and 33
02130602 618.9 3.3, 6.6, 10, 15, 20, and 33
oppose12 707.1 3.3, 6.6, 10, 15, 20, 33, and 72.7
C,zlso,.,c,2 971.4 3.33, 6.6, 10, 15,20, and 33
C,.,1~:o3c,4 542.9 8.1, 16.3, and 24.4
C,4Eo,c,, 675.1 10.1, 203,304, and 40.5
onshore,1 763.2 3.3, 6.6, 10, 15, 20, and 33
C,,1=.o3c,6 599.0 9, 18, 27, and 36
(3,650.0, 731.2 11, 22, 33, and 44

 

 

 

51

 

When an aqueous solution of a-CD (100 mg/mL) was added to the ABA model
compounds above their melting temperature, a white precipitate formed immediately.
The solutions were stirred overnight, filtered, washed with water, and dried under
vacuum. In contrast to the results seen with the polymers, no gel formed at any
concentration of the ABA model compound. Complexes also were synthesized by the
dissolution of the model compound in methanol before the addition of the 01-CD solution.
The yields were lower for this synthetic method, and again no gels formed. As in the
polymer case, the yields of the ABA-a-CD ICs were studied as a function of the
concentration of ABA model compound. Data for four complexes are shown in Figure
2.9. For each series of complexes, the yield saturated at slightly higher concentrations
than was seen for the polymers, and at similar concentrations, the isolated yields and
saturation levels were higher for the ABA model compounds than for the polymers. The

slopes of the plots were similar to that seen for the synthesis of the polymer ICs.

2.2.4 Properties of ABA-a-CD complexes

The CXEOny-oc-CD le were analyzed by 1H and '3 C solid-state NMR, powder
X-ray diffraction, DSC, and the stoichiometry of the complexes were determined by the
continuous variation method. Figure 2.10 shows a representative continuous variation

plot used to determine the number of CDs threaded per C12E03C12 triblock oligomer. As
seen in the figure, the maxima of the plots (0.7 mol fraction a-CD) corresponds to a
stoichiometry of 2.3 ct-CD’S per triblock molecule, similar to that observed for the (AB),-

a-CD ICs. 1H NMR data confirmed the stoichiometries of the C12E03C12 and

52

CrzEOMC ,2 complexes; the other ABA model compound-CD complexes were insoluble
in deuterated DMSO and all other deuterated solvents tested.

DSC scans were taken of each ABA-a-CD IC and compared to its individual
components. A representative example, the C1 zEOgClz-a-CD IC, (Figure 2.11) shows a
melting transition at 33°C for the triblock oligomer, which is lost upon complexation
with a-CD. The scan of the complex is featureless like that of a-CD, confirming that
there is no free ABA compound in the samples. To further prove that a-CD was threaded
onto the model compounds, the ABA-a-CD ICs were analyzed by powder X-ray
diffraction and 13C solid-state NMR. Figure 2.5 (trace 0) shows a representative
example, the C14EOgC14-a-CD IC, which has a diffraction pattern consistent with the
formation of columnar (channel) crystal structure. 13C solid-state CP/MAS NMR
(Figure 2.6) was taken of the triblock oligomer, CMEOIZCM (spectrum C), and its
inclusion complex with a-CD (spectrum B). Like the (AB),-01-CD ICs, a-CD becomes
symmetrical in the inclusion complex formation with the ABA model compounds, a

change that is consistent with threading of triblock oligomer through the CD cavity.

53

 

Yield of Complex (%)
N (to A 01 O) \1 oo (0
O O O O O O O O
O
O

_x
O

 

 

 

O '7 .‘-' 1 l 4 I 1 I l I 1

0 1O 20 30 40 50
Concentration of CXEOny (mglmL)

Figure 2.9. Yield of CXEOny-a-CD complex as a function of the ABA triblock
concentration. Complexes were prepared by adding an aqueous solution of a-CD (150
mg/1.0 mL H20) to the molten ABA oligomer (5 — 60 mg).

(13 (314130604,A CloEO3Cl6,O CloEoocm,<> (314130304)-

54

 

 

 

 

180
160 -
140-
‘6’, 120
15.
X
2 100
a.
E
O
U 30 -
H-
O
2
.92 601
>.
40 4
20«
o . 2 L
0 0.2 0.4 0.6 . 0.8 1

Mol Fraction a-CD

Figure 2.10. Continuous variation plot showing the yield of the C rzEOgCiz-a-CD IC
(0) obtained for various ratios of the reactants. The sum of the reactants was fixed at 1.5
x 104 mole.

55

 

 

 

 

 

‘1'
1

 

Heat Flow (Endo ——>)

A o

 

 

 

1 I 1 I a I 1

-100 -50 O 50 100
Temperature (°C)

 

Figure 2.11. Normalized DSC thermograms of a) C12E03C12, b) C12E03C12-a-CD IC,
and c) a-CD. The scans shown are the second heating scans taken after heating the
samples to 100°C to erase the thermal history, and flash quenching to -100°C. All scans
were run at 10°C/min under helium.

56

2.3 Discussion
Since the discovery by Harada that a-CD threads onto a polyethylene glycol

' numerous inclusion complexes have been synthesized from a wide variety of

chain,
polymers. The design of nanometer scale supramolecular structures organized by
noncovalent interactions in the hydrophobic cavity of CD has gained much interest by
many researchers. Besides studying the noncovalent binding forces (found in biological
systems), the selectivity and molecular recognition of CD towards guest molecules has
been of major interest. Molecular recognition is based on the selectivity-structure
correlation between the host cavity of CD and the guest-solvent mixture. Previous
researchers have shown a-CD can recognize guest molecules based on the cross-sectional
areas of the polymer. Harada has shown that a-CD will form ICs with PEG, but will not
form complexes with PPG because of size constraints of the polymer backbone.26 A goal
of this research was to see if a-CD could distinguish (recognize) between two types of
polymers based on their difference of properties, not in size.

To date there have been few reports of ICs formed from block copolymers. Such
complexes could have interesting structures and properties since the CDs could
preferentially thread onto one of the blocks. Tonelli et al. prepared 01 and y-CD
complexes with a (PCL-PEO-PCL) ABA triblock copolymer,l6 to study the behavior of
isolated and segregated polymer chains. Since only a single chain can be incorporated
inside the a-CD cavity and two parallel, side-by-side chains can reside in the y-CD
channel, one can compare properties of single and pairs of parallel, side-by-side chains

with those of the bulk polymer. Yui et al. synthesized stimuli-responsive polyrotaxanes

from B-CD and a PEO-PPO-PEO triblock copolymer that can fimction as a molecular

57

27:28 and Topchieva et al. studied complex formation with a series of low molecular

piston,
weight nonionic surfactants: PEO-FPO, PEO-PPO-PEO, and PPO-PEO-PPO”.

In this research, we reported the first synthesis of a-cyclodextrin ICs with (AB),
microblock copolymers. The structures of these high molecular weight polymers are
related to nonionic surfactants, and differ in that the blocks have relatively short lengths
and the AB pattern is repeated many times. We also reported results for a series of ABA
triblock model compounds30 that have structures similar to the (AB), block copolymers.

The ICs formed from these small molecule analogs helped in interpreting the results

obtained for the a-CD complexes of (AB), block copolymers.

2.3.1 Solid state 13C NMR of inclusion complexes

The experimental data show that oc-CD threads onto the (AB), and ABA model
compounds to form columnar inclusion complexes, and the NMR results support a facile
threading and de-threading process in solution. There is however, inconclusive evidence
for selectivity in the threading process. One would expect that a-CD would prefer to
reside over the hydrophobic segments of the polymer chain. One sign of selective
threading would be a shift in the '3 C resonances for the alkyl chains, without a
corresponding change in the signal from the carbon atoms of the ethylene oxide chains.
The solid state NMR spectrum of [C3TIC3E03]n (Figure 2.6, spectrum e) shows a sharp
resonance at about 29 ppm from the methylene units of the alkylene blocks and a peak at
69 ppm from the ethylene oxide units of the polymer chain. In the spectrum of the
[C31tC3EOg],,-01-CD complex (spectrum d), the methylene resonance broadens and shifts

to 26 ppm. The results for the ABA model system are similar; the '3 C resonances for the

58

alkyl chain shift from 30 ppm in the ABA oligomer to 28 ppm in the a-CD complex.
However, the ethylene oxide signals in the (AB), block copolymers and ABA model
compounds overlap with the signals observed for carbons C-2, C-3, and C-5 of or-CD,

which makes any changes in the chemical shift for these resonances unobservable.

2.3.2 Stoichiometry of inclusion complexes

The stoichiometry of the polymer-a-CD complexes studied was 2.3 a-CD
units/polymer repeat unit. The same stoichiometry was obtained for complexes prepared
from the ABA model compounds. Comparisons of the length of the polymer repeat unit
with the depth of the oc-CD cavity (8 A) suggest a maximum threading of 6-7 a-CD
units/repeat unit. The low observed stoichiometry suggests that the CD units are not
threaded unifome onto the polymer, due to kinetic control of the stoichiometry of the
complex. Considering the solid state NMR results and the stoichiometry of the
complexes, we conclude that our data are consistent with preferential threading onto the
alkylene segments.

Non-uniform threading can stem from the precipitation of the complex from
solution, and the data of Figure 2.1 implicate solubility as a major factor in determining
the yield and stoichiometry of the complexes. The yield plots for the polymers show
slightly different slopes that depend on the mole fraction of the hydrophobic alkylene
chain in the polymer. In Figure 2.1, the least hydrophobic polymer ([C37tC3EOg],,, 8 of
33 chain atoms in the alkylene segment) has the smallest slope and the lowest yield at
saturation. The other two polymers, [C37tC3EO6],, and [CanC4E03],,, are essentially

identical in terms of the fraction of chain atoms in the hydrophobic segment, and have

59

slopes and limiting yields that are nearly identical. For the IC stoichiometry to be nearly
independent of the polymeric or ABA model compound structure, the solubility limit for
the ICs must be exceeded when 2-3 a-CD units thread onto the chain. Since the
amphiphilic nature of the polymers renders the (AB), polymers sparingly soluble in the
reaction solvent, it should not be surprising that threading a few CD units onto a chain is
sufficient to make the IC insoluble.

An alternative explanation for the shift of the slope in Figure 2.1, is a molecular
weight dependence on the threading rate. Of the three polymers used in the study,
[C 37tC3EOg],, had the highest molecular weight (Mn: 44 kg/mol) whereas those of
[C4TEC4E03]n and [C3TEC3EO6]n were lower (M.,= 19 and 25 kg/mol, respectively). Thus
consistent with the results, [C37tC3EOg],, should be the slowest polymer to completely
thread and might be expected to give the lowest yield. A molecular weight effect is not
observed in the ABA triblock model compound series (Figure 2.9) nor when ICs were
made at the same concentration with various (AB), polymers that have different Mn’s
(data not shown). A molecular weight effect cannot be ruled out though because samples

covering a broad range of molecular weights were not available for this study.

2.4 Conclusions

A variety of spectroscopic and analytical techniques confirm that (x-CD threads
onto (AB), microblock copolymers and ABA triblock model compounds to give
columnar inclusion complexes. Relatively small changes are seen in the yield and
stoichiometry of the complexes as the lengths of A and B units are changed. The

stoichiometry of the complexes is dominated by the decrease in the solubility of the

60

polymers upon complexation. Solid state NMR data on complexes formed from ABA
triblock model compounds suggest preferential threading over hydrophobic segments,
although the 2.311 stoichiometry observed for the complexes does not lead to a simple

regular structure for the complexes.

61

2.5
(1)
(2)
(3)
(4)
(5)

(6)
(7)
(8)

(9)

(10)

(11)

(12)

(13)

(14)

(15)
(16)
(17)
(18)

References

Harada, A.; Kamachi, M. Macromolecules 1990, 23, 2821-2823.

Ceccato, M.; LoNostro, P.; Baglioni, P. Langmuir 1997, 13, 2436-2439.

Qiao, J .; Chen, Y. Y.; Baker, G. L. Chemistry of Materials 1999, 11, 2542-2547.
Harada, A.; Li, J .; Kamachi, M. Macromolecules 1993, 26, 5698-5703.

Kawaguchi, Y.; Nishiyama, T.; Okada, M.; Kamachi, M.; Harada, A.
Macromolecules 2000, 33 , 4472-4477.

Rusa, C. C.; Luca, C.; Tonelli, A. E. Macromolecules 2001, 34, 1318-1322.
Rusa, C. C.; Tonelli, A. E. Macromolecules 2000, 33, 1813-1818.

Harada, A.; Li, J .; Kamachi, M.; Kitagawa, Y.; Katsube, Y. Abstracts Papers of
the American Chemical Society 1997, 213, 28-POLY.

McMullan R.K.; Saenger. W.; Fayos J .; Mootz D. Carbohydrate Research 1973,
31, 37.

Saenger, W. Angewandte Chemie-lnternational Edition in English 1980, 19, 344-
362.

Saenger, W. R.; Jacob, J .; Gessler, K.; Steiner, T.; Hoffmann, D.; Sanbe, H.;
Koizumi, K.; Smith, S. M.; Takaha, T. Chemical Reviews 1998, 98, 1787-1802.

Harata, K. Chemical Reviews 1998, 98, 1803-1827.

Saenger, W.; McMullan, R. K.; Fayos, J .; Mootz, D. Acta Crystallography, Sect.
B 1974, B30, 2019—28.

Panova, I. G.; Gerasimov, V. 1.; Topchieva, I. N. Vysokomolekulyarnye
Soedineniya Seriya A & Seriya B 1998, 40, 1681-1686.

Harada, A. Coordination Chemistry Reviews 1996, 148, 115-133.
Lu, J.; Shin, I. D.; Nojima, S.; Tonelli, A. E. Polymer 2000, 41, 5871-5883.
Huh, K. M.; Ooya, T.; Sasaki, 8.; Yui, N. Macromolecules 2001, 34, 2402-2404.

Harada, A.; Suzuki, 8.; Okada, M.; Kamachi, M. Macromolecules 1996, 29,
5611-5614.

(19)

(20)

(21)

(22)

(23)

(24)

(25)

(26)

(27)

(28)

(29)
(30)

(31)

Li, J .; Yan, D. Macromolecules 2001, 34, 1542-1544.
Lu, J.; Mirau, P. A.; Tonelli, A. E. Macromolecules 2001, 34, 3276-3284.

Udachin, K. A.; Wilson, L. D.; Ripmeester, J. A. Journal of the American
Chemical Society 2000, 122, 12375-12376.

Manor, P. C.; Saenger W. Journal of the American Chemical Society 1974, 1974,
3630-3639.

Gidley, M. J .; Bociek, S. M. Journal of the American Chemical Society 1988,
110, 3820-3829.

Schneider, H. J .; Hacket, F.; Rudiger, V.; Ikeda, H. Chemical Reviews 1998, 98,
1755-1785.
Takeo, K.; Kuge, T Agricultural and Biological Chemistry 1970, 34, 1787.

Harada, A.; Okada, M.; Kawaguchi, Y.; Kamachi, M. Polymers For Advanced
Technologies 1999, 10, 3-12.

F ujita, H.; Ooya, T.; Yui, N. Macromolecules 1999, 32, 2534-2541.

F ujita, H.; Ooya, T.; Yui, N. Macromolecular Chemistry and Physics 1999, 200,
706-713.

Topchieva, I. N.; Karezin, K. I. Colloid Journal 1999, 61, 514-519.
Chen, Y. Y.; Baker, G. L. Journal of Organic Chemistry 1999, 64, 6870-6873.

Okumura, H.; Okada, M.; Kawaguchi, Y.; Harada A. Macromolecules 2000, 33,
4297-4298.

63

CHAPTER 3

Experimental

3.1 General details

Unless otherwise specified, ACS reagent grade starting materials were used as
received from commercial suppliers. THF was dried by refluxing over CaHz overnight,
and was then distilled from Na/benzophenone. House nitrogen was used in air and
moisture sensitive reactions.

Proton and carbon nuclear magnetic resonance ('H and '3 C) analyses were carried
out at room temperature in deuterated solvents (CDCI3, DMSO-d6, and D20) on a Varian
Gemini-300 spectrometer. The chemical shifts were calibrated using the signals from
residual non-deuterated solvent and are reported relative to tetramethylsilane. Solid-state
'3 C CP/MAS NMR spectra were measured at 100.4 MHz on a Varian VXR-400
spectrometer with sample spinning rates of 4 0 kHz at room temperature.

Molecular weights of polymer samples were determined by gel permeation
chromatography (GPC) using a PLgel 20;: Mixed A column at 30 °C with THF as eluting
solvent at a flow rate of lmL/min. A Waters R410 Differential Refractometer was used
as the detector. The concentration of the polymer samples was 1mg/mL, and the results
were calibrated with monodisperse polystyrene standards.

Differential scanning calorimetry (DSC) analyses of compounds were run on a
Perkin Elmer DSC 7 instrument calibrated with indium and hexyl bromide standards.
The samples were sealed in aluminum pans and were heated at 10°C/min under a helium
atmosphere. Liquid nitrogen was used as a coolant. Samples were melted and held at

150°C for five minutes to erase their thermal history. They then were quenched to —100

64

°C at a rate of 200°C/min, and then held at —100°C for five minutes before the sample run
was started. Samples were heated to 150°C, cooled to ~100°C, and heated t0150°C, all at
a rate of 10°C/min. The DSC melting point was taken as the onset of the peak of the
melting endotherm. Heats of fusion were calculated from the endothermic peak using the
accompanied functions of the DSC 7 software. Thermogravimetric analyses (TGA) of
the compounds were obtained from a Perkin Elmer TGA 7 instrument at a heating rate of
10°C/min, usually over the temperature range of 30 to 600°C.

X-ray diffraction patterns were collected from powdered samples on a Rigaku
rotaflex 200B diffractometer equipped with a rotating anode, Cu K01 x-ray radiation
(1:1.541838 A) and a curved crystal graphite monochrometer. Diffraction patterns were
collected at a rate of 20 = 2°/min over the range 20 = 0-40°. Powdered samples were
prepared by grinding the samples and then spreading the solid onto the window of a glass
sample holder with a spatula.

3.2 Material synthesis

The (AB), polymers were prepared by the metathesis polymerization (ADMET)
of oligoethylene oxide 01,o)-dialkenyl ethers using Schrock’s molybdenum alkylidene
catalyst as described earlier."3 The number average molecular weights (Mn) of the
polymers were [CanCaEOg],, = 19x103, {cameos}, = 44x103, and [emu-:06], =
25x103 g/mol. The polydispersity of each polymer is ~2. The polymers are soluble in
methanol and THF, but are insoluble in hexanes and water. ABA triblock oligomers
were prepared by treating the appropriate exact length oligoethylene glycol with NaH and

2 equivalents of the desired alkyl bromide.

65

1 Complex Formation between [CxflCxEOyIn Polymers and a-CD.

Inclusion complexes of a-CD and [CxtthEOy]n polymers were prepared at room
temperature by adding a 2 mL solution of oc-CD in water to a solution of the (AB), block
c0polymer in 0.5 mL methanol. For low polymer concentrations (< 8 mg/mL), the
solution became turbid and the complex precipitated as white crystals, which were
collected by filtration and washed with water to remove any uncomplexed a-CD. All
complexes were dried overnight in a vacuum oven to remove any water from the
complex. Higher polymer concentrations led to formation of a white gel, which was
isolated as a crystalline inclusion complex by freeze-drying and washing the resulting
solid with water.

2 Complex Formation between C,,EOyCx model compounds and a-CD.

Complexes were synthesized from three series of model compounds, CleOnd
where y= 3, 6, 8, and 14, CMEOyCM where y = 3, 6, and 8, and C16EOyC16 where y = 3
and 6, to explore how the differences in the hydrophobicity of the model compounds
(increased alkyl length or decreased ethylene oxide chain length) have on the complex
yield and stoichiometry of the complexes. When an aqueous solution of (x-CD (100
mg/mL) was added to the ABA model compounds above their melting temperature, a
white precipitate formed immediately. The solutions were stirred overnight, filtered,

washed with water, and dried under vacuum.

66

3.3

(1)
(2)
(3)

References

Qiao, J.; Chen, Y. Y.; Baker, G. L. Chemistry ofMaterials 1999, 11, 2542-2547.
Chen, Y. Y.; Baker, G. L. Journal of Organic Chemistry 1999, 64, 6870-6873.

Chen, Y. Y.; Baker, G. L.; Ding, Y. Q.; Rabolt, J. F. Journal of the American
Chemical Society 1999, 121, 6962-6963.

67