Chemistry of poly(ether-ester) and siloxane based thermoplastic elastomers


Material Information

Chemistry of poly(ether-ester) and siloxane based thermoplastic elastomers
Physical Description:
ix, 120 leaves : ill. ; 29 cm.
Zuluaga, Hector Fabio, 1946-
Publication Date:


Subjects / Keywords:
Elastomers   ( lcsh )
Thermoplastics   ( lcsh )
Silicones   ( lcsh )
Graft copolymers   ( lcsh )
Chemistry thesis Ph. D
Dissertations, Academic -- Chemistry -- UF
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1993.
Includes bibliographical references (leaves 112-119).
Statement of Responsibility:
by Hector Fabio Zuluaga.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 030045691
oclc - 29897637
System ID:

This item is only available as the following downloads:

Full Text







To my wife, Maria Helena and the fruit of our love, our daughter


Several of the reasons why I came to the University of Florida
at this "late" age are: the encouragement of Dr. William R. Dolbier
Jr. and the support of my family.
The completion of this work would not have been possible
without the advice and moral and scientific support of the members
of the polymer floor: Professors G. B. Butler, R. Duran and J. R.
Reynolds, Drs. J. O'Gara, A. Wolf, K. Brzezinska, W. Rettig, M. Nauman,
H. Fadel, M. Elsabee, G. Adams, W. Sigmud, T. Viswanathan, P. Bernal
and S. Wanigatunga, and my former and present lab-mates, W. Cooke,
M. Murla, C. Matayabas, H. Zhou, N. Zhang, D. Patwardhan, D. Tao, J.
Portmess, J. Patton, D. Smith. J. Nel, C. Marmo, R. Advincula and his
wife Carolyn, A. Thibodeaux, R. Bodalia, J. Roberts, B. Sankaran, P.
Balanda and S. Kim.
Thanks go to Professor C. D. Batich, D. Amery and K. Mentak
from Materials Science for the surface analysis.
Special thanks go to Lori Engle and Jeff Linert for their
unfailing support, to Kathy Novak for widening my perspective of
this country and to Jim Konzelman for his understanding and his
technical support. I thank P. Hargraves, J. Poppell and Lorraine
Williams for their readiness to help. Lorraine and her husband David
were there specially in those tough moments.

Thanks go to the members of my committee: Professors W. R.
Dolbier Jr., W. M. Jones, C. D. Batich and J. R. Reynolds for their
academic assistance.
Sincere thanks go to all the friends who showed their support
specially Raul and Alejandra, Mauricio, Pim and little Erin, Wilma,
Elvia, Jorge Moreno and Jorge Morales, and of course the members of
the soccer group and the hosts in the "club", Francisco and Wilmot.
Special thanks go to my colleagues in Cali : Alvaro, Luz
Marina, Rodrigo, Omar, Nelly, Guillermo and Edison.
I am deeply grateful to my mother, Tulia, and my twelve
brothers and sisters, the Toro Puerta family, Ivan, Angela and their
families and William, for their continuous support.
This work was supported by a grant from the Army Research
office and the Universidad del Valle through a leave of absence.
Finally I want to thank my advisor Dr. K. B. Wagener for his
extraordinary support and his guidance through his example. He
certainly knows what a Ph.D. is about: independence.


AC KNO W LEDG EM ENTS......................................................... ................................. i i

A B S T R A C T .... .................................................................................... ........................... v ii


1 INTRO DUCTIO N ............................................................ ........................ 1

A Brief Introduction on Elastomers........................ .................. 1

The discovery of Elastic Copolymers.....................................................6

G raft C opolym ers ................................................... ........................ 1 1

Thermoplastic Elastomers ......................... ..... ................. 14

Microphase Separation................................... ........................ 16

Polysiloxane Based Thermoplastic Elastomers............................23

Objectives of this Dissertation..................... ....... ..............25

2 ALANINE MEDIATED POLYESTERIFICATION.........................................29

Introduction to Polyesterification................... ........................... 29

Alanine Mediated Polyesterification of a Hydroxyacid
Telechelom er ..................................................................................... 3 2

A Model Study of Alanine Mediated Esterification.......................38

The Alanine Catalyzed Esterification of Pivalic Acid............... 39

Transesterification of Methyl Benzoate..........................................46

C o nc lusio ns ........................................................................................... 4 9


Thermoplastic Elastomers Based on Soft Phase-Hard
Phase Copolymers ............................................................................. 50

Synthesis of the Soft Phase in Siloxane Based Multiphase
elastom ers ........................... .... ............... .......................... 5 6

Ring Opening Polymerization of Lactones....................... ...........61

4 SILOXANE-LACTAM GRAFT COPOLYMERS ............................................ 67

P o ly a m id e s ...................................... ...................................... ..................... 6 8

Polymerizability of Lactams ........................ .............................. 73

Polysiloxane-Lactam Graft Copolymers..........................................76

C o nclusio ns ................................. .................................................... 8 4

5 EXPERIMENTAL......................................................................................... 86

Instrum entation ........................ .................................................. 8 6

C he m ica ls ................................... ..................................................... 8 7

Esterification of Pivalic Acid. Model Studies...............................88

Synthesis of 2-(2-Methoxyethoxy)ethyl Pivalate(29).................88

Esterification Reactions...................... ..... ............................ 89

Chromatography of Reaction Mixtures..............................................90

Transesterification Reactions ........................ ......... ............ .... 91

Chromatography of Transesterification Reactions.....................92

Characterization of 2-(2-Methoxyethoxy)ethyl Benzoate.........94

Synthesis of Siloxane Random Copolymers ...................................95

Synthesis of Poly(dimethyl-co-

Synthesis of Poly(dimethyl-co-
methylcarboxypropylSiloxane) (11 a)............................................. 97

Synthesis of Siloxane Potassium Carboxylate (11b)...................97

Synthesis of End-capped Siloxanes from Cyclic
M onom ers ........................... .... ... ...... .......... ........... 98

Grafting of Lactones and Lactams to Poly(siloxane)s
(11a) a nd (11b) ........................................ ............ ....................... 9 9

Grafting of Pivalolactone (8).............................. ...................... 99

Grafting of D-Valerolactone (39) ...................................................1 00

Grafting of E-Caprolactam (43)....................... .......................1 00

Film Casting of Siloxane-Lactam Graft Copolymers................. 01

Molecular Weight Determinations............................ ............... 1 03

Vapor Pressure Osmometry ........................ ............ .............. 03

Gel Permeation Chromatography ........................................... .105

Universal Calibration for Molecular weights ..............................108

End G roup Analysis ............................................................. 111

M icroscopy........................ .................................................................... 1 2


BIOGRAPHICAL SKETCH ................. .....................................................1 21

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy


Hector Fabio Zuluaga

May, 1993

Chairman: Dr Kenneth B. Wagener
Major Department: Chemistry

This dissertation deals with thermoplastic elastomers.
Specifically it describes alanine mediated polyesterification as
well as the synthesis of siloxane graft copolymers with lactones
and lactams.
A model study of alanine (ALA) mediated polyesterification
was carried out using pivalic acid (PVA) and 2-(2-
methoxyethoxy)ethanol (MEEtOH) as model compounds. The effect of
alanine on the rate of esterification and on the catalytic activity of
titanium tetrabutoxide (TTBO) was examined. Simultaneous
reactions with alanine and with combinations of alanine and TTBO
were compared with a control reaction of PVA and MEEtOH. While
alanine showed a 3- to 6-fold enhancement of the rate of


esterification of PVA with MEEtOH, a small enhancement of the
catalytic activity of TTBO was observed. The work was extended to
transesterification reactions using methyl benzoate (MeBz) and
MEEtOH as model compounds and this time no "alanine" effect was
The synthesis of siloxane graft copolymers with lactones
(pivalolactone, y-valerolactone, 6-valerolactone) and e-caprolactam
led to graft copolymers in all cases except for y-valerolactone. This
work included the introduction of a carboxyl functional group in the
siloxane which was used to initiate grafting by ring opening
polymerization of the lactone or lactam. The reaction with
pivalolactone proceeded at room temperature in tetrahydrofuran and
18-crown-6, while the polymerization of 8-valerolactone required
higher temperatures (900C). Polymerization of e-caprolactam was
carried out in two steps, first heating the reagents in a sealed tube
to 260C and then heating to 180C under nitrogen atmosphere.
Siloxane graft copolymers with pivalolactone and E-
caprolactam showed a high degree of microphase separation and
formed good films from hexafluoroisopropanol solutions. Grafts
with different compositions of pivalolactone were prepared, which
could lead to the production of thermoplastic elastomers. Siloxane-
caprolactam graft copolymers, with siloxane contents up to 18%
were synthesized; however, only those copolymers containing 1-4
w% siloxane afforded films. These films showed a siloxane rich
surface, and thus the corresponding material is effectively a surface
modified nylon.


A Brief Introduction on Elastomers

This dissertation deals with polysiloxane and polyether-based
thermoplastic elastomers obtained by combining a soft polymeric
segment with a hard polymeric segment yielding either a block or a
graft copolymer. The work has been divided in three parts:
1- Study of the role of alanine in esterification and
transesterification reactions using model compounds. This study
stems from the promising results obtained in the alanine mediated
polyesterification of a poly(ester-ether) telechelomer.
2- Synthesis of siloxane-pivalolactone graft copolymers. A
polysiloxane has been substituted for the polyether as the soft
segment in the previously synthesized poly(ester-ether), and a change
from linear to graft copolymers was carried out.
3- Substitution of different lactones and lactams for
pivalolactone as the hard phase.
Most materials exhibit some degree of elasticity upon stress,
ranging from a very small value for metals to a very large value for
rubber. This difference is due to the fact that while metals are
made of very well packed atoms, rubber and rubberlike materials are
high molecular weight polymers which because of the length of their
chains, can adopt a large number of spatial arrangements.

Alteration of these arrangements explains the amazing response of
these materials to an imposed stress.
One of the first materials to exhibit a sizable elasticity was
discovered by the American Indians, who called it caoutchouc, from
caa wood and o-chu to flow or weep, because it was obtained as a
milky sap by making incisions on the trunk of the tree Hevea
brasiliensis.1 The Europeans discovered a number of uses given to
this material by the Indians: they spread it on clothes to make them
waterproof; they produced gumboots, flexible bottles and enema
syringes, etc.1 Rubber products, however, suffered from two major
drawbacks: first, the gum was sticky and remained permanently
sticky at room temperatures. In hot weather it turned softer and
stickier; second, in cold weather it became progressively hard and
stiff, and in extreme frost, it became almost completely inflexible.
A solution to this problem was found by an American inventor,
Charles Goodyear,2 who discovered the process now known as
vulcanization. Introduction of vulcanization on both sides of the
Atlantic was followed by the rapid successful application of rubber
to virtually all of its modern uses. The climax came with the
patenting of the pneumatic tire in 1888 by John B.Dunlop;1 his
invention came just at the right time to meet the needs of the infant
automobile industry.
The large demand for rubber over the next few years was an
incentive for the search of a synthetic substitute. It was already
known that rubber was made of isoprene (1), Figure 1.1, which was
converted to rubber by Bouchardat3 in France, Tilden in England and
Wallach4 in Germany. Eventually F. Hofmann5 in Germany succeeded

in synthesizing isoprene from mineral sources and, in 1909, was
awarded the world's first patent for a synthetic rubber.6 By this
time, Kondakov in Russia had discovered that dimethylbutadiene (2)
could be converted into an elastic polymer and was more easily

Isoprene (1) Dimethylbutadiene (2)

Figurel.1. Structures of isoprene and dimethyl-butadiene.

World War I put an end to further developments in England,
while in Germany, cut off from natural supplies, the effect was the
opposite and research was accelerated by the Nobel prize winning
discovery of Staudinger that rubber was made up of giant
Summarizing, the requirements needed for a rubberlike
material to be useful are as follows: 1- the material must consist of
polymeric chains; 2- the chains must have a high degree of
flexibility; 3- the chains must somehow be joined into a network
The first requirement arises from the fact that the molecules
in an elastomeric material must be able to alter their arrangements
and extensions in space dramatically in order to respond to an
imposed stress, and only long-chain molecules have the required

large number of spatial arrangements of very different extensions.
Figure 2 shows a two-dimensional projection of a random
arrangement of a 200 carbon polyethylene chain.9

0 to 40


Figure 1.2. Two-dimensional drawing of a 200 carbon
polyethylene chain.

This projection was generated by using well known bond
lengths and bond angles, and by considering the preference of n-
alkanes for trans rotational states. Two features of this projection
are worth mentioning, these being (a) the high extension of part of

the chain, due to the preference for trans conformations and (b) the
tendency of some sections of the chain to fold in a random fashion,

forming more compact zones that decrease the end-to-end distance
of the chain. Even for this short chain, the extension could be
increased to 400% by rotation around the skeletal bonds without
altering bond angles nor breaking bonds.9
When an elastic polymer is stretched, the elastic work, W, can
be expressed as10 :
W= fdx
f = retractive force
dx = change in length
If the change in the free energy is aG,

(aG/aX)T,p = (aH/aX)T,p T(aS/aX)T,p

An ideal elastomer is defined by the condition

(aH/aX)T,p = 0 f = -T(aS/aX)Tp

The equation above shows that the retractive force in an ideal
elastomer is due to the decrease of entropy on extension, which
originates in the distortion of the polymer chains from their more
probable conformation.
The second requirement specifies that the different spatial
arrangements be accessible, not hindered by intrinsic rigidity of the
chains, extensive degree of crystallization, or high viscosity. This
needed flexibility implies that the polymer not only should be in its
amorphous state but, at use temperature, should be above its glass

transition temperature, Tg, in order to overcome the brittleness
characteristic of the glassy state.

Figure 1.3. Cross-linked polymer chains.

The third characteristic is required in order to obtain the
elastomeric recoverabilty. It is obtained by joining or cross-linking
segments of the chain, Figure 1.3, thereby preventing stretched
polymer chains from irreversively sliding by one another.

The Discovery of Elastic Copolymers

A new development in rubberlike materials was the discovery
that copolymers of dienes and olefins yielded better elastomers than
polydienes or polyolefins alone. One of the first to be prepared was
a copolymer of butadiene and styrene, similar to but not identical to
the styrene butadiene rubber of today.11 Now with the experience
gained over the years and a good understanding of the kinetics and
mechanism of copolymerization, polymer chemists are able to

synthesize products with predetermined properties. By carefully
choosing the components of a copolymer, the properties of the
respective homopolymers can be modified at will. The modification
of the polymers also can be achieved by mixing or blending them
with other materials which are effective in altering behaviour such
as antioxidants, plasticizers, flame retardants, fillers, etc. When
the additive is a high molecular weight polymer the resulting
mixture is called a polymer blend. A blend is a physical mixture of
two or more polymers as opposed to a copolymer, which posesses a
chemical bond between its components. Most blends are not stable
however, because of unfavorable free energy considerations.
Polymers prefer to intertwine among themselves rather than among
other polymers, and as a consequence, they eventually separate into
different phases.12
In a mixture of polymers a very small entropy of mixing (AS) is
observed, and even a slight positive change of enthalpy (AH), is
sufficient to produce a positive change in free energy; thus
incompatibility of the two polymers results.
Polymer blends can be useful when molecular weights are low
or when the expected useful life of the product is relatively short,
and therefore a simple mixture of polymers might be perfectly
suitable to achieve a specific property. On the other hand if
mixtures can be co-reacted, such as cross-linking of two
unsaturated rubbers, very useful long term behaviour results and the
product is really a copolymer.
Copolymers can be classified as (a) random, (b) alternating,
and (c) block, Figure 1.4.

-wAAABBAABAABBBABBAA random copolymer

ABABABABABABABABAB alternating copolymer


Figure 1.4. Classification of copolymers

Random copolymers are the product of a statistical placement
of the comonomers along the chain. A copolymer is considered
perfectly random when the two monomers show equal reactivity to
both propagating species. For example, methacrylonitrile (3) and
methyl methacrylate (4) have very similar reactivity ratios under
radical polymerization13 and consequently they react under these
conditions to produce a random copolymer, as shown in Figure 1.5.
Alternating copolymers result from the alternate placement of
the comonomer along the chain. This type of copolymer requires
pairs of monomers with very low reactivity ratios (no affinity for
their own propagating species). An example is the free radical
copolymerization of styrene (5) and maleic anhydride (6). The most
generally accepted mechanism for this reaction14-15 suggests the
formation of 1:1 complexes of an electron donor (styrene), and an
electron acceptor maleicc anhydride), which react to produce an
alternating copolymer, Figure 1.6.

C H3 CH H3

Methyl methacrylate (3)



bCH3 6CH3 6CH3

Figure 1.5. Production of a random copolymer from
methacrylonitrile and methylmethacrylate



Styrene (5) Maleic anhydride (6)


Figure 1.6. Formation of an alternating copolymer from
styrene and maleic anhydride.

Block copolymers are comprised of terminally connected
segments of comonomers. They are synthesized by "living" addition
polymerization or by step growth condensation techniques. A-B and
A-B-A architectures are primarily synthesized by anionic living
polymerization, while -(-A-B-)-n structures are more often prepared
by step growth methods. Anionic living polymerization discovered
by Szwarc and coworkers,16 allows control of molecular weight and
molecular weight distribution by fixing the monomer to initiator
ratio, while the order of addition of the monomers determines the
sequence of the blocks. The synthesis of an A-B-A, styrene-
butadiene- styrene block copolymer is shown in Figure 1.7.

Butadiene (7)


Styrene (5)


Figure 1.7. Synthesis of styrene-butadiene-styrene.

In the mid 1980's a combination of living anionic
polymerization and ring opening polymerization was employed to

synthesize a polyether-ester telechelomer with an OH group at one
end and a COOH group at the other. This prepolymer was polymerized
by step growth condensation, to obtain a multiblock or segmented
copolymer with (A-B)n architecture.17-18 The synthesis of this
segmented copolymer will be discussed in detail later.

Graft Copolymers

The copolymers displayed so far are all linear. Graft
copolymers, on the other hand, Figure 1.8, consist of a
functionalized, linear backbone to which a second monomer has been
attached, as a side chain, via chemical bond with the functional




Figure 1.8. Formation of graft copolymers.

Graft copolymers are generally prepared by free radical,
anionic, cationic or ring opening polymerization. Free radical graft
copolymerization is based, for the most part, on the reaction of

olefinic monomers in the presence of preformed polymers containing
labile hydrogens.19 The reaction is initiated by peroxides,
irradiation or thermal methods.20 Abstraction of the reactive
hydrogen produces radicals on the backbone which can polymerize
the second monomer. Graft copolymers prepared by this method are
characterized by a poor control on the composition.
Ionic methods allow better control of the structure, especially
when living anionic techniques are used. If the first polymer
backbone is functionalized with several anionic sites and the second
monomer is copolymerized by living techniques using the anionic
functional group as initiator, excellent control of the structure of
the graft can be achieved. This is the case when the cyanopropyl
group in poly(dimethyl-co-methylcyanopropylsiloxane) is converted
to the carboxylate and then reacted with pivalolactone21 (8) to
obtain polypivalolactone chains as pendants from the polysiloxane
backbone, Figure 1.9. The advantage of this synthesis is twofold.
First, carboxypropyl units are randomly placed on the polysiloxane,
due to similar reactivities in the two monomers,
dimethyldichlorosilane (9), and methylcyanopropyldichlorosilane
(10), and the carboxylic acid concentration on the backbone, can be
controlled by the ratio of the two monomers. Second, the living ring
opening polymerization of pivalolactone (8) proceeds easily at room
temperature and the amount of pivalolactone grafted can be
controlled as well. The combination of an amorphous polysiloxane
backbone with a highly crystalline segment may lead to very special
properties of the product as will be discussed in the next section.

Cl-Si-CI Cl-Si-CI
I (CH2)3
CH3 + CN
Dimethyldichloro Methylcyanopropyl
silane (9) dichlorosilane (10)

1- H20
2- H2S04

H3 ,H3 ,H3
.- Si-0-Si-O-n Si -O0-
CH3 CH3 (CH2)3
(1la) COOH

1. KOH

2. CH3 __ O
CHH3 I + 18-Crown-6
Pivalolactone (8)
Si-O-Si-O- Si -0-
CH3 CH3 (dH2) CH
C- C-CH2
6[ CH3 0 n

Poly(siloxane-g-pivalolactone) (12)

Figure 1.9. Synthesis of poly(siloxane-g-pivalolactone)

Thermoplastic Elastomers

Thermoplastic elastomers are materials with rubber-like
behaviour that soften and can be reshaped upon heating. In the late
1940s, Coleman at ICI attempted to improve the dyability of
polyethyleneterephthalate, PET, by copolymerizing it with
poly(oxyalkylene)glycol.22 A similar approach was used by Snyder at
Dupont23 to increase PET hydrophilicity; he carried out a controlled
ester interchange by melt mixing PET with a copolymer of
terephthalic acid, trimethyleneglycol and suberic acid. The
resulting copolyester had a higher strength, a higher stretch
modulus than any natural rubber threads and a very quick elastic
recovery (snap). The fibers were regarded as thermoplastic
elastomers since they could be extruded from their melt or spun
from solvents.
An important discovery was made by Schollenberg and
coworkers at Goodrich,24 who reported on a polyurethane elastomer,
Estane,T" which is a linear polyurethane described as being "virtually
cross-linked" by secondary rather than primary forces. At Shell, di
and triblock copolymers of styrene-butadiene and of isoprene-
styrene were successfully made in 1957 by Porter.25 During
subsequent work at this company, it was observed that
"unvulcanized" styrene-butadiene-styrene, S-B-S, triblock
copolymers, were unusually "snappy" and appeared to be "scorched";
stress-strain measurements showed high tensile strength and high
ultimate elongation, while solution data showed complete solubility
with no gel or microgel formation.22 This anomalous behaviour led

to the formulation of the domain theory.27-30 This theory considers
that in the bulk state, the polystyrene short end segments of these
block copolymers agglomerate; at temperatures significantly below
the glass transition temperature of polystyrene, these
agglomerations (domains) act as strong, multifunctional junction
points, and so the triblock copolymers behave as though they are
joined in a cross-linked network.22
Thermoplastic elastomers have been widely studied; the
poly(urethane-ether) thermoplastic elastomers discovered by
Schollenberg and Scott24 in 1958 which have been the subject of
intense research in the following years,31,32 are synthesized by step
polymerization of an aryl diisocyanate (13) and a diol (14), see
Figure 1.10. Other thermoplastic elastomers obtained by step
condensation polymerization are poly(ester-ethers),33 polyamides,34
and polysiloxane copolymers.


Polyethylene glycol (14) 4,4'-diphenylmethane
diisocyanate (13)


-lNH< \-CH2-Q H-(OCH2CHz)n-0-
S0 m

Figure 1.10. Poly(urethane-ether) thermoplastic elastomer.

Poly(siloxane) copolymers can be used as elastomers at very
low temperature due to the low glass transition of the siloxane
phase. Some examples of siloxane materials include copolymers
with polyurethanes,35 polycarbonates,36 polymethacrylates,37
polyamides,38 polymethylstyrene38 and polypeptides.39 Only a few
of them possess a semicrystalline phase, which in many respects is
preferred for thermally induced changes. Other advantages of the
siloxanes copolymers are biocompatibility, chemical inertness, gas
permeability, hydrophobicity and low surface energy. This last
property is very important for the synthesis of surface modified

Microphase Separation

The different constituents of block and graft copolymers are
bonded together covalently, thereby preventing inmiscibility of the
constituents from leading to physical separation as described
earlier. However, they show very special properties as a result of
morphological features unique to these systems. These features
have been explained by the domain theory in terms of "microphase"
separation of the incompatible components. The resulting
microphases, hard and soft segments, are of a size on the order of
the dimensions of the constituent polymer molecules themselves,
and these microphases often develop a highly organized domain
A pictorial representation of microphase separation is given in
Figure 1.11; the hard segment acts as a physical cross link while the
soft segment is responsible for the elastic behaviour. This idealized

structure was postulated22 from the mechanical and theological
properties of S-B-S triblock copolymer but there was no direct
observation to support it. Although microphase separation had
clearly been established by the Strasbourg School,40-42 the
phenomenon remained a curiosity until Holden and Milkovich27
discovered that microphase separation in block copolymers, of
appropriate composition and architecture, could yield materials with
unique and useful properties.

Soft Segment
Hard segment

Figure 1.11. Representation of the microphase separation in a
block copolymer.

The development of staining techniques using osmium
tetroxide43 allowed morphologies to be established by electron
microscopy.44 Using these techniques a more detailed picture of the
morphological changes with block copolymer composition was
proposed,45 as shown in Figure 1.12. Three basic types of
morphology for microphase separated copolymers are possible:
spheres, cylinders and alternating lamellae. As the styrene content
increases, the morphology of the polystyrene phase changes from
spheres to cylinders, both dispersed in a continuous diene phase.

Spheres Cylinders Lamellae Cylinders Spheres

Increasing A-Content
Decreasing B-Conternt

Figure 1.12. Effect of composition on block copolymer

When the volume fraction of styrene and diene are about the
same, the two form alternating lamellae. Further increase in
polystyrene content reverses the situation and now cylinders and
spheres of the diene appear in a polystyrene continuous phase.
Microphase separation in (A-B)n type multiblock copolymers, is
only observed if one of the segments is crystalline.46 Few studies
on microphase separation have been performed for graft copolymers.
Depending on their chain structure and homogeneity, graft
copolymers can have multiphase morphologies similar to A-B, A-B-
A, or (A-B)n. 47-49
Differential scanning calorimetry (DSC) has been very useful in
the study of microphase separation. The degree of phase mixing is
indicated by the deviation on a thermal transition such as glass
transition (Tg) and melting temperature (Tm). The presence of
separate values of Tg or Tm coinciding with those of the respective
homopolymers is an indication of good phase separation. Other

techniques employed to study phase separation in copolymers
include electron microscopy (EM), small angle X-ray scattering
(SAXS), and dynamic mechanical analysis (DMA).
Using DSC50,51 it was found that with a poly(pivalolactone)
segment length constant at 12 units, microphase separation begins
when the poly(oxyethylene) segment contains 14 units and is
complete upon extension to 24 units, (Figure 1.13). If the
poly(oxyethylene) unit is kept constant at 24 units and the PVL is
varied, (PVL)m-(POE)24-(PVL)m, a small degree of microphase
separation is observed for m= 5,7,9; complete microphase separation
is achieved when m reaches 16.


\ I 6"n \ I n

Figure 1.13. Structure of pivalolactone-polyoxyethylene
triblock copolymer (15).

Therefore, copolymers with segmental composition: (PVL)12-
16-(POE)14-4-24-(PVL)12-16, or higher, show excellent phase

Thermoplastic Elastomers Via Chain Propagation/Step Propagation

As described above polymer architecture can be varied to
achieve well phase separated, solvent resistant poly(ester-ether)
copolymers,17-18 Figure 1.14. Even so true poly(ester-ether)s are

deficient in terms of the physical properties that are required in
many applications, particularly when they are melt processed.5 2
Deficiencies in properties such as elastic recovery (less than 95%),
stress decay (often greater than 15%) and compression set (more
than 10%) are manifestations of the inefficiency of the physical
cross-link, which in turn is a result of poor phase separation. Phase
mixing can be attributed partly to irregularity of the polymer chain.
In the case of poly(ester-ether) copolymers prepared by step
condensation polymerization, the hard segment has a polydispersity
ratio approaching 2, and this broad molecular weight distribution
adversely affects phase separation.53 Therefore high regularity
within segments, i.e. monodispersity within each segment, should
lead to enhanced phase separation and better crystallization
phenomena.54 Furthermore, the model study mentioned before50,51
supplies information on the length of the POE and PVL blocks
necessary to ensure good phase separation.
A block copolymer from the soft segment ethylene oxide (16), and
the hard segment pivalolactone (8), was synthesized by anionic
living techniques, ensuring monodispersity within each segment.
Since the soft poly(oxyethylene) segment is incompatible with the
hard poly(pivalolactone) segment, these regular copolymers exhibit
enhanced phase separation. On the other hand, high molecular weight
is necessary to obtain a thermoplastic elastomer with good
mechanical properties. In order to further polymerize this block
copolymer by step growth propagation the carboxyl group provided by
polypivalactone on one end and the hydroxy group provided by

polyoxyethylene at the other end have been used, making the
difunctionalized block copolymer a telechelomer.
Figure 1.14 illustrates the approach used to synthesize the
poly(oxyethylene-b-pivalolactone) telechelomer (20).17,18 An acetal
capped anionic initiator21 (17) polymerizes ethylene oxide to give a
potassium alkoxide of a masked polyether and this new initiator is
used to polymerize pivalolactone. Since potassium alkoxides are
strong nucleophiles, they can randomly attack the carbonyl carbon
and the p-methylene carbon in lactones. Such a random attack would
result in a pivalolactone segment containing irregularities. A
weaker nucleophile such as carboxylate anion will attack the 3-
methylene carbon of pivalolactone.55-57 Thus it became necessary
to convert the alkoxide end to a carboxylate anion by reacting (19)
with succinic anhydride (18). The carboxylate anion was used to
polymerize pivalolactone (8) yielding the masked poly(oxyethylene-
b-pivalactone) copolymeric salt (19). Acid hydrolysis converted this
salt to the telechelomer, (POE)n-SA-(PVL)m, (20), where n=29, m=12.
Polymerization of this telechelomer by step growth
condensation was attempted using different -OH and -COOH
activating agents, and polyesterification catalysts.18 However,
none of the attempts were successful. High polymer was finally
achieved by alanine in a first "activation" step at 180"C, followed by
an increase in temperature to 250*C using Ti(OBu)4 a known
transesterification catalyst. It is the first time that alanine has

Ethylene oxide (16)

1) CH3-CH-O(CH2)30-K-
OCH2CH3 Acetal masked
initiator (17)
2) 0 O succinic
anhydride (18
3) CH3 0
3CH3 Pivalolactone


OCH2CH3 POE-b-PPVL salt (19) CH3


The telechelomer (20) CH3

1. 185C, alanine, 2hrs
2. 0.5 mm Hg, 0.5 hrs
3. Ti(OBu)4, 185-260"C,
Ar atmosphere, 1.5 hrs
4. 0.5 mm Hg, 0.5 hrs


Segmented copolymer (21) CH3

Figure 1.14. Chain propagation/step propagation
polymerization of ethylene oxide and pivalolactone.

/ 0 '*\


been used as an activator for polyesterification, and a more detailed
study of the "alanine effect" has been undertaken in this

Polysiloxane Based Thermoplastic Elastomers

Some of the advantages of polysiloxanes as the soft phase in
thermoplastic elastomers have already been mentioned in this
chapter. More detailed information may be found in the review on
siloxane containing copolymers by Yilgor and McGrath.58
A series of poly(siloxane-g-pivalolactone) copolymers were
synthesized59 in order to investigate composition/phase separation
relationships, and phase separation studies were carried out by DSC.

Table 1.1. DSC data of poly(siloxane-g-pivalolactone).


W% Tg ("C) Tm ("C)
10 -123 187.7
20 -123 191.1
30 -123 201.5
40 -123 202.2
50 -123 221.5

The results are presented in table 1.1. The Tg for the PDMS backbone
was the same for all the compositions of the copolymers and the Tm
for the poly(pivalolactone) segment was close to that of homopoly-
(pivalolactone) with similar molecular weights. From these results

silane (9)

+ CN
dichlorosilane (10)

1- H20
2- H2SO4

Si-O-Si-O- Si -0OF-
CH3 CH3 (CH2)3
(11a) COOH

1. KOH

2. CH3 0
CH31 + 18-Crown-6
Pivalolactone (8)
Si-O-Si-0- Si -0-
CH3 CH3 (CH2)3 CH
C -CH2-
0 CH3 n

Poly(siloxane-g-pivalolactone) (12)

Figure 1.15. Synthesis of poly(siloxane-g-pivalolactone).

it was concluded that there was excellent phase separation in the
copolymer with pivalolactone contents up to 60%. Water contact

angle measurements confirm this conclusion, as all copolymers,
except those containing 70% poly(pivalolactone), show a surface
richer in poly(dimethylsiloxane).
The synthesis of poly(siloxane-g-pivalolactone) is illustrated
in Figure 1.15. Cyanopropyl functionalized siloxane copolymers were
obtained by hydrolytic polymerization of the respective dichloro
silanes (9) and (10). The cyano group was converted to the
potassium carboxylate, and this functional group was used to graft
pivalolactone to the siloxane backbone.60, 61
Patwardhan60 synthesized a series of carboxypropyl
functionalized siloxanes with different spacing of the functional
group on the siloxane backbone in order to study the influence of the
carboxylic acid functional group concentration. 10, 20, 37, 60, and
120 Dimethylsiloxane repeat units per methylcarboxypropylsiloxane
unit were introduced in the backbone by controlling the mole ratio of
the two monomers, and grafting of pivalolactone on two of them was
carried out.

Obiectives of this Dissertation

This dissertation describes an extension of the work on
siloxane based graft copolymers. Other lactones and lactams have
been substituted for pivalolactone as the hard phase in order to
investigate the influence of the chain length of the lactone monomer
on the properties of the graft copoplymer.
The specific objectives of the research described in this
dissertation have been to conduct a model study of alanine mediated
esterification and transesterification for thermoplastic elastomers

and homopolymers and to synthesize siloxane based thermoplastic
elastomers grafted with new lactones and lactams as a hard phase.
For the first time, alanine has been reported to act as an
"activator" in polyesterification. This is a very interesting result,
as polyesterification is one of the most commercially exploited
reactions in the world, and as a consequence, large quantities of
different catalysts are consumed in the polymer industry. Many of
those catalysts are expensive and some of them are toxic; on the
other hand, alanine is a natural product, and nontoxic. It will not
only be more economical but will also be an environmental
alternative in the production of polyesters.
The synthesis of well phase separated, biocompatible, and
solvent and radiation resistant, siloxane based graft copolymers
that could be used as thermoplastic elastomers, is the other subject
of this dissertation. The research has been focused in two aspects,
these being control of the molecular weight of the siloxane soft
phase and substitution of other lactones and lactams for
pivalolactone as the hard phase. Control of the molecular weight of
the segments is highly desirable in order to assure good phase
separation, which is necessary for the synthesis of thermoplastic
elastomers. Figure 1.16, illustrates the synthesis of the siloxane
graft copolymers with lactones and lactams.
The soft segment is synthesized by acid catalyzed ring opening
of octamethylcyclotetrasiloxane (22) and its methylcyanopropyl
analog (23). This is an equilibrium reaction, where the linear
polysiloxane can regenerate cyclic siloxanes by backbiting, Figure
1.17. A decrease in the amount of cyclics can be obtained if the end

groups are blocked by reacting them with a monofunctional


CH3"-~0 HC3

CH3 I I 'CH 3

tetrasiloxane (22)

NC(CH2)3S 0 CH3

+ CH3, I (CH2)3CN
Si\ 0
NC(CH2)/ O-S(CH2)3CN
tetrasiloxane (23)

1) 50% H2S04, (PhaSiO)2
2) Aq. KOH

CH3 CH3 CH3 ICH2)3 CH3 CH3 CH3



CH3 CH3 CH3 CH2)3 CH3 CH3 CH3

O 0

Figure 1.16. Synthesis of polysiloxane graft copolymers with
lactones and lactams.


H3HH3 H3 H H H3 H3
CH3-Si-0- i-- 0i-- i-0-i-(0- i-)n-0- i-0-i--H

L3 H3 3 HA H3 H3 H3 H3 H3

\ / H3 H3 H3
CH3 Si Si H
1 I ( .3 HO-Si-O-Si-0 -w-Si-OH
O Si0 -L-

CH3 \H3 H3 H3 H

Figure 1.17. Backbiting to form cyclics.

Hexaphenyldisiloxane not only acts as an end capper which
blocks backbiting, but it also permits molecular weight
determination by 1H NMR. Correlation with other methods such as
vapor pressure osmometry (VPO), GPC, and viscometry allows a
better insight into molecular weight variations.
All these topics will be discussed in the chapters that follow.


Polyester elastomers are multiblock segmented copolymers
which can be represented by the general formula -(A-B)n-.
Commercialization of polyester elastomers has led to extensive
research in this area and, because these polymers offer unique
physical properties and processing characteristics, they have gained
worldwide acceptance.61 Any contribution that could lead to new
catalysts will be of academic and commercial interest. Before
discussing the study on the role of alanine in esterification, some
background information will be presented.

Introduction to Polyesterification

Polyesters have been studied since the last century,62 but it
was not until the 1930s that Carothers63 carried out a systematic
study on polyesterification; he prepared a series of linear aliphatic
polyesters that did not find commercial use due to their low melting
points. In 1946, Whinfield64 prepared poly(ethyleneterephthalate),
PET, and it was at this time that polyester became widely used for
the manufacture of fibers and films. In order to have commercial
application a polyester has to have high molecular weight, regular
chemical structure, and adequate physical properties to allow
processability. The synthesis of a particular polyester can often be

successfully performed by different methods: in bulk (usually at high
temperature), in solution, by an interfacial method or in the solid
state.65 When performed at low temperature, either in solution or
by an interfacial method, polyesterification may present problems
related to the removal of residual catalyst, side products, or
solvent. When polymerizations are carried out at high temperature,
chain scission may compete with the growing chain reactions, an
event which can become the most important limiting factor for the
molecular weight to increase.65
Industrial production of polyester is carried out by direct
esterification, figure 2.1, or by transesterification, figure 2.2.

0 O

1 mol 2.2 mol


0 0
Figure 2.1. Synthesis of "polyester" by direct esterification.

Figure 2.1. Synthesis of "polyester" by direct esterification.

Most polyesterification reactions only occur at sufficiently
high rates in the presence of suitable catalysts, and due to the
academic and industrial relevance of catalysis, an enormous number
of publications have appeared.65

0 O

1- 180C
2- 260C


Figure 2.2. Synthesis of "polyester" by

Most polyesters can be obtained by direct esterification at
high temperature either from dicarboxylic acids and diols or from
hydroxy acids. Esterification is an equilibrium reaction that can be
driven to completion by removal of the water produced in the
process. The reaction is normally carried out under reflux, where
water is removed by either application of vacuum to the system or
by the introduction of a stream of an inert gas.
Direct esterification can proceed at high temperature (1800C-
280C) even in the absence of added catalysts; in this case the
carboxyl group of the acid provides protons that catalyze the

reaction. Small amounts (0.1-0.5 wt %) of an external catalyst are
nevertheless added in order to increase the reaction rate An
extensive list of catalysts employed in direct esterification has
been reported by Fradet and Marechal.66 This list includes strong
protonic acid such as H2S04, H3P04, CF3SOsH, p-toluensulphonic
acid, etc., oxides or salts of heavy metals, and organometallic
compounds of titanium, tin, zirconium and lead. It appears that for a
metal to be effective it has to be able to exchange ligands and to
coordinate with reactants forming a complex that it is not too
strong. The catalytic activities of many metal compounds have been
compared, and it has been found that alkoxide derivatives of
titanium and zirconium are the most efficient.67
Most of the polyesters made by direct esterification can also
be prepared by transesterification. This reaction is of great
commercial importance, since monomers as ester derivatives have
lower melting points, higher solubilities in diols, and can usually be
obtained at a higher purity grade than the corresponding acids, and
therefore often allow production of better quality products and
easier process control. However higher costs, resulting from more
expensive raw materials and more expensive plants, sometimes
makes direct esterification more convenient, as in the case of

Alanine Mediated Polvesterification of a Hydroxvacid Telechelomer

A recent study of polyesterification catalysis can be found in
the synthesis of well phase separated thermoplastic elastomers
combining chain and step propagation polymerization.17-18 This

synthetic scheme was presented in figure 1.14, where the chain
propagation process involved the use of an acetal masked initiator
to polymerize ethylene oxide, followed by reaction with succinic
anhydride to yield a carboxylate terminated oligomer. Addition of
pivalolactone to this oligomer, followed by hydrolysis, produced a
monodisperse poly(oxyethylene-b-pivalolactone) telechelomer (20)
having a Mn of 2700 and containing -OH and -COOH end groups.
Subsequent step propagation of this hydroxy acid terminated
telechelomer was attempted without any success, even when the
telechelomer was treated with -OH and -COOH activators. COOH
activating agents such as N,N'-bis(2-oxo3-oxo-azolidinyl)-
phosphoramidic chloride/triethylamine,69 1-methyl-2-bromo-
pyridinium chloride/tri n-butylamine,70 and OH activating agents
such as triphenylphosphite,71 hexachlorocyclotriphosphatriazene,72
and dehydrating agents such as trifluoroacetic acid/methylene
chloride and methylsulphonic acid /phophorus pentoxide,73 had no
effect on the polymerization. Well known polyesterification
catalysts such as antimony oxide and titanium tetrabutoxide also did
not induce step polymerization.
The use of alanine followed by titanium tetrabutoxide, figure
2.3, yielded a high molecular weight -[(POE)n-(PVL)m]x multiblock
segmented copolymer (20).18 Surprisingly, alanine was almost
totally recovered via sublimation when the telechelomer/alanine
mixture was subjected to vacuum (0,5 mmHg). Furthermore,
elemental analysis of the copolymer (21), revealed the presence of
0.1% nitrogen, indicating a catalytic amount of alanine left in the
product. The theoretical amount of nitrogen calculated on the basis

of a 1:1 reaction of the telechelomer and alanine is 0.5%, and thus it
is evident that alanine did not form a 1:1 adduct with the
telechelomer.18 This is the first time that an amino acid of natural
ocurrence has been reported as an activator for polyesterification
and, if confirmed, it could be of great commercial value as a
substitute for antimony and titanium based catalysts which present
a certain degree of toxicity.

Telechelomer (20), n=29, m=12

1-85C, alanine, 2hrs.
2-0.5 mm Hg for 1/2 hr.
3-Ti(OBu)4, 185C to 2600C,
under argon, 1.5 hrs.
4-Apply vacuum (0.5mm Hg)
for 0.5 hr.
Segmented copolymer (21) H3

Figure 2.3. Alanine mediated polyesterification of the

The selection of alanine (24) as a possible "activator", was
aimed at the alleviation of the steric hindrance posed by the two
methyl groups a to the carboxyl end group of the telechelomer,
assuming that an amide linkage is to be formed,18 see figure 2.4. If
the copolymer is on average a tetramer of the telechelomer, then
0.1% N in the copolymer is approximately equivalent to a 1:1

alanine:copolymer adduct, a fact overlooked previously. This
suggests that alanine (24) reacts with the growing chain to form
either an amide (25), or a mixed anhydride (26). Propagation of this
reaction would require a second molecule of the telechelomer to
perform a nucleophilic attack on either the amide or the anhydride
derived from the first telechelomer molecule.

HOCH2CH "",,w- OCH2-C -C-OH
Telechelomer (20) CH3

185C CH3- 0-I-C-OH
Alanine (24)

HOCH2CH2^""""^w OCH2-C --C-N--- CH -C-OH
OH3 Amide (25)


CH3 0 0 NH2
1 II II 1
HOCH2CHg~^vvv OCH2-C -C-O-C- CH --CH
CH3 Mixed anhydride (26)
Mixed anhydride (26)

Figure 2.4. Reaction of alanine with the telechelomer

These derivatives are even more sterically crowded on the site
of propagation than the telechelomer, and therefore the steric
reasons invoked as the argument for using alanine could be
The fact still remains that the only way to convert the
telechelomer to the segmented copolymer involved the use of alanine
in a preliminary step and three hypotheses can be proposed as a
possible explanation for the action of alanine in this reaction. a)
Alanine could simply act as a dehydrating agent; b) It could act as a
template for the two reacting ends of the telechelomer, c) It could
increase the effectiveness of Ti(OBu)4 catalyst itself.18
The first hypothesis seems reasonable, as alanine is known to
form a dimer upon heating with the loss of two molecules of water.
It may well be that when alanine sublimes as a result of the vacuum
applied, figure 2.3, the dimer regenerates alanine monomer, carrying
water out of the system.
The second hypothesis requires that alanine bring two
telechelomer molecules together in such a way that the carboxyl end
group of one molecule finds the hydroxyl end group of the other. This
implies a high degree of order in the chains, which at this
temperature, 1850C, and given the length of the telechelomer is
The third hypothesis is not very reasonable, as the catalytic
action of titanium alkoxides implies coordination of the carboxylic
acid to the titanium and exchange of alkoxide groups for the hydroxyl
group.67 A telechelomer alanine adduct would in fact place the
carboxyl group further away from the coordination center.

It seems more reasonable to consider that in forming the
mixed anhydride with the telechelomer, alanine becomes a better
leaving group (as the carboxylate) rather than the hydroxyl group.
Furthermore, hydrogen bonding of the amino group of the mixed
anhydride with the carbonyl of the telechelomer could help labilize
the carbonyl for the nucleophilic attack of the OH end group of
another chain (figure 2.5). Similar hydrogen bonding could be true in
the amide intermediate; however, the amide group is not as good
leaving group.




CH3 ) -CH3
S------ C C=O
I II /
CH3 ... ,O

Figure 2.5. Hydrogen bonded assisted nucleophilic attack of
the hydroxyl end on the carboxyl end of the intermediate.

A model study was undertaken in order to investigate the
specific role of alanine in this process.

A Model Study of Alanine Mediated Esterification

Model studies help to determine if a reaction that has been
successful under specific conditions can be generalized to a whole
class of compounds holding the same functional groups. In the
present case, the telechelomer (20) can be considered a hydroxy
acid, and its polymerization yields a polyester.
Pivalic acid, PVA, and 2-(2-methoxyethoxy)-ethanol, MEEtOH,
were chosen as model compounds in an attempt to mimic the
chemistry involved in the synthesis of the telechelomer, see figure

Pivalic acid (27) 2-(2-methoxyethoxy)ethanol (28)


Alanine (24)

/O O O\ + H20

2-(2-methoxyethoxy)ethy aate (29)
2-(2-methoxyethoxy)ethyl pivalate (29)

Figure 2.6. Alanine mediated esterification of model

Pivalic acid (27) is similar to the pivalolactone hard segment
of the telechelomer, and 2-(2-methoxyethoxy)ethanol (28) is
equivalent to three ethylene oxide (soft segment) repeat units. If
alanine is a catalyst in the polymerization of the telechelomer (20)
to a multiblock segmented copolymer (21) then a rate enhancement
in the model reaction should be observed. Could alanine be a
catalyst for polyesterification in general? A second model study to
mimic formation of polyethyleneterephthalate is presented as well.

Esterification of Pivalic Acid

The reactions were carried either under vacuum or under argon
atmosphere. In each case the reagents were dried under vacuum at
room temperature, and then heated until they refluxed; samples were
syringed out and analyzed by high performance liquid
chromatography (HPLC).

Reactions under vacuum
1- The effect of alanine alone.
Simultaneous reactions with and without alanine under
moderate vacuum (0.1 mm Hg) and reflux (93-1240C) were carried
out. Comparison of the retention factor (Rf) of the reaction mixture
by thin layer chromatography (TLC) with the Rfs of the starting
materials showed the increase of ester (29) with time as confirmed
by infrared spectroscopy (IR) and 13C nuclear magnetic resonance,
(13C NMR). The change in the carbonyl stretching frequency in the IR,
from 1696 to 1725 cm-1, as well as the change in the 13 C NMR
chemical shift from 183 to 178 ppm indicated the conversion of
pivalic acid (27) to the corresponding ester (29), (figures 2.7 and

2.8). Furthermore, quantitative analysis of the reaction mixture was
carried out by high performance liquid chromatography, HPLC, which
allowed calculation of the ester acid ratio. Table 2.1 shows a 4-6
fold catalytic effect by alanine on the esterification reaction.

Table 2.1. Ester/acid ratio vs time for alanine mediated

Time Reaction number
(h:min) 1 2 3
1:10 0.06 0.25 0.56
2:10 0.23 0.43 1.10
3:30 0.59 0.67 2.40
5:40 0.61 0.99 4.30
8:05 0.93 1.00 6.70

Alanine mol% : reaction 1: 0.0, reaction 2: 5.2, reaction 3: 0.52.

When alanine mole % is 5.2 the rate of the reaction is
increased in the first three hrs, but reaches a constant value after
that. However when alanine mole% is reduced an order of magnitude
to 0.52, all the samples analyzed show a significant increase in the
reaction rate.
2-The combined effect of alanine and titanium tetrabutoxide.
According to the work discussed before,18 the combination of
alanine and titanium tetrabutoxide (TTBO) was the key to a
successful polymerization of the hydroxy acid telechelomer (20).

Figure 2.7. IR of pivalic acid, top, and 2-(methoxyethoxy)ethyl
pivalate, bottom.



183 ppm
Il -r i S W 111 ,01

183 ppm

178 ppm


178 ppm

Figure 2.8. 13C NMR of pivalic acid, top, reaction mixture,
center, methyl pivalate, bottom.

One of the hypotheses was that alanine could enhance the
catalytic activity of titanium tetrabutoxide, and thus experiments
were designed to compare the catalytic action of alanine and TTBO
separately with the catalytic activity of a mixture of both. In
separate experiments, equimolar amounts of pivalic acid and the
alcohol (MEEtOH) were treated with 3% molar TTBO and varying
amounts of alanine; 6.5 and 1 mole percent alanine was added to the
4th and 5th reactions, respectively, (see table 2.2).

Table 2.2. Ester/acid ratio vs time
Time Reaction number

(h:min) 3 4 5
1:25 1.05 0.47 0.77
2:03 7.46 2.85 22.5
4:05 19.2 17.0 100

Alanine, TTBO mole%: reaction 3: 0.0, 3.1; reaction 4: 6.5, 2.9;
reaction 5: 1.0, 3.0.

While the presence of 1% alanine showed a 1-3 fold increase in
the production of the ester with respect to the control, the presence
of 6.5% alanine had just the opposite effect. This result indicates
that alanine in amounts close to 1.0% enhances the catalytic activity
of TTBO. Confirmation of this "alanine effect" was obtained when
three separate reactions, each one with 1.5% TTBO and varying
alanine mole % of 0.0, 1.0 and 1.0, were run simultaneously. The
reactions containing 1.0% alanine, showed a 1-3 fold increase in the
production of the ester (table 2.3). These encouraging results led us

to carry out reactions with catalyst concentrations similar to those
used in industrial polyesterification in order to explore potential
commercial applications.

Table 2.3. Ester/acid ratio vs time.
Time Reaction number
(h:min) 7 8 9
1:08 0.10 0.15 0.30
4:00 0.34 0.47 0.43
8:05 0.60 1.90 1.50
12:30 0.70 2.60 13.1

Alanine, TTBO mole%: reaction 7: 0.0, 1.5; reaction 8: 1.0, 1.5;
reaction 9: 1.0, 1.5.

The amount of titanium tetrabutoxide used in the industrial
production of polyethyleneterephthalate PET is of the order of 0.1-
0.05 mole %, however, reactions using this amount of catalyst were
very slow. As expected, an increase in the amount of catalyst to
1.4% reduced reaction times to those observed in previous
experiments. However, the "alanine" effect was not observed and
further experiments did not always reproduce the earlier findings.
This research then leads to three conclusions: 1- Molar
concentrations of alanine close to 1.0% appear to enhance the rate
for esterification of pivalic acid with methoxyethoxyethanol 2- In
order to observe catalytic activity of titanium tetrabutoxide its
molar concentrations should be close to 1.0%. 3- At low pressure
(0.1-10 mm Hg) and relatively low temperatures (70-110C), alanine

concentrations close to 1.0% molar appear to produce a small
enhancement of titanium tetrabutoxide catalytic activity.

Model reactions under argon
Further esterification experiments were carried out, this time
under argon atmosphere, just above atmospheric pressure, allowing
study of the reactions at higher temperatures.
The reagents were dried under vacuum at room temperature, then
mixed and heated to reflux temperatures (160-170C). Samples
were syringed out from the reactions and analyzed by HPLC. Table
2.4 shows the results obtained when three separate reactions were
run; reaction 10 contained no alanine nor TTBO, reaction 11
contained 0.75% TTBO, and reaction 12 contained 0.75% TTBO and
1.0% alanine.

Table 2.4. Ester/acid ratio vs time.

Time Reaction number

(h:min) 10 11 12
1:00 0.24 2.2 1.9
1:55 0.54 3.7 2.6
2:20 0.41 3.9 2.9
2:50 1.0 5.0 3.5
3:20 1.4 5.8 4.2
3:58 1.8 7.0 4.5
4:50 3.2 7.4 4.8
9:00 3.6 18 7.0

Alanine, TTBO mole% : reaction 10: 0.0,0.0;
0.75; reaction 12: 1.0, 0.75.

reaction 11: 0.0,

Two interesting observations can be extracted from these
data. Esterification reactions are faster when run under argon at
higher temperatures. 2- No enhancement by alanine of the catalytic
activity of TTBO, was observed under these conditions.
In fact, comparison of tables 2.3 and 2.4 shows that higher
ester/acid ratios are observed for reactions under argon at higher
temperatures. This, of course, is expected as higher temperatures
increase the number of molecules with enough energy to overcome
the activation energy of the reaction. The second observation, no
"alanine" effect, was contradicted by a similar experiment, where a
slight increase in the activity of TTBO was observed when a mixture
of 1.2/0.75 mole% alanine/TTBO was added to the reagents.
Finally, a transesterification model study was carried out to
complete this work and this research is discussed in the next

Transesterification of Methyl Benzoate

One of the problems in the direct esterification of pivalic acid
is the precipitation of a white solid attributed to hydrolysis of
titanium alkoxides.67 Transesterification of methyl pivalate was
not feasible due to its low boiling point (101-102C), and therefore
methyl benzoate (b.p. 198-1990C) was chosen as a suitable model
compound. Besides its higher boiling point, methyl benzoate has a
very close resemblance to dimethyl terephthalate, the most used
monomer in the polyester industry, and the side product of the
reaction, methanol, is easier to remove (b.p. 64-650C).

Since previous esterification of pivalic acid proceeded at a
higher rate when carried out at atmospheric pressure and refluxing
temperatures, the same conditions were applied to the following
Equimolar amounts of methyl benzoate (MeBz) and 2-(2-
methoxyethoxy)ethanol (MEEtOH) were reacted under argon
atmosphere; refluxing temperatures started at 150C and increased
to 1900C as the transesterification proceeded to completion. Four
reactions were carried out (table 2.5); when 1.2 mole % alanine was
added, no positive effect on the reaction rate was observed.
Addition of 0.71 mole % of TTBO instead of alanine, produced a
positive effect at the beginning but after 3 hours no enhancement of
the reaction rate was observed.
Twelve more reactions were completed after modification of
the apparatus in order to establish identical conditions for all the
experiments. Two of these experiments showed the "alanine" effect,
while the other 10 did not not exhibit this phenomenon.

Table 2.5. MEETBz/Me Bz ratio vs time.
Time Reaction number

(h:min) 13 14 15 16
0.40 0.05 1.9 2.6 0.02
1:50 0.35 2.6 3.3 0.94
3:20 2.2 3.0 3.5 1.7
4:20 8.3 5.1 3.6 1.9
6:40 18 5.7 3.8 2.4

Mole % alanine, TTBO: reaction 13: 0.0, 0.0; reaction 14: 1.2, 0.71;
reaction 15: 0.0, 0.71 ; reaction 16:.1.2, 0.0.

These results led us to conclude that alanine does not enhance
the catalytic activity of TTBO in the transesterification of methyl
benzoate and MEEtOH.
It should be noted that the conditions in the step propagation
of the telechelomer could not be reproduced exactly with model
compounds. The telechelomer was a solid with number average
molecular weight (Mn) of 2700, which could be heated to high
temperatures under vacuum. The exact conditions of the reactions
were as follows: equimolar amounts of d,l-alanine and telechelomer
were heated, under argon, to 185C for two hours, then, maintaining
the temperature, the reaction was switched from 760 mm Hg to 0.1
mm Hg for 1/2 hour; at this point alanine sublimed and was
withdrawn from the reaction mixture. The product obtained did not
form a film when cast from a solution in methylene chloride.
Addition of a catalytic amount of TTBO to this product and heating
to 185C, under argon, produced a highly viscous polymer which
could not be magnetically stirred. Finally the temperature was
gradually increased to 2600C for 1.5 hours and then vacuum applied
for 1/2 hour.
As can be inferred from the previous paragraph, alanine was
required in a first step, but then removed and TTBO added as
catalyst. Furthermore, low pressure (0.5 mm Hg) and high
temperatures can be applied in the second step of the reaction. The
application of such conditions was not possible in our model studies
due to the volatility of pivalic acid and methyl benzoate. For
example, alanine was present all the time in these alanine-mediated

model reactions. Sublimation of alanine was not possible without
evaporating pivalic acid or methyl benzoate.


Model studies for transesterification and esterification have
shown that in most cases alanine enhances the rate of esterification
of pivalic acid (PVA) and 2(2-methoxyethoxy)ethanol (MEEtOH).
Furthermore, alanine appears to enhance the catalytic activity of
titanium tetrabutoxide (TTBO) in the same reaction. Alanine may
play a role in a preliminary activation step of the polyesterification
of poly(pivalolactone)-poly(oxyethylene) telechelomers, as shown by
the formation of dimers, trimers and occasionally pentamers of the
telechelomers. However, reproduction of the exact conditions of the
original reaction was not possible with low molecular weight model


Thermoplastic Elastomers Based on Soft Phase-Hard Phase

A part of this research has dealt with the catalyzed synthesis
of polyester elastomers. This section describes work on siloxane
graft copolymers as potential thermoplastic elastomers where one
goal has been to improve the thermal stability of a siloxane soft
phase and assure its analytical evaluation. A second goal has
consisted of examining the effect of ring size of a lactone monomer
on the hard phase behaviour.
In view of the promising previous results,18,59 the synthesis
of poly(siloxane-g-lactones) was undertaken in a search for
thermoplastic elastomers with good microphase separation, solvent
and radiation resistance, and biocompatibility. One of the
requirements for the production of thermoplastic elastomers.27-30
is the display of microphase separation between the hard and soft
The soft phase, responsible for the elasticity, should be a
flexible polymer with a low glass transition temperature (Tg) while
the hard phase should be either a highly crystalline polymer, or, an
amorphous polymer with a high Tg, such that at use temperature, the
hard phase is in the glassy state. The hard phase acts as a

physical cross-link that prevents the copolymer chains from
slipping past each other.
Polydimethylsiloxanes, figure 3.1, are an excellent choice for
the soft phase, due to the unusual molecular structure of the
polymer chains themselves.

HOw- Si-O-Si-O w -H
Polydimethylsiloxane, PDMS, (30)

Figure 3.1. Structure of poly(dimethylsiloxane).

The Si-O bond (1.64A) is longer than the C-O bond (1.41A) but
substantially shorter than the Si-O predicted by calculation from
the atomic radii (1.83A). This shortening of the bond length can be
explained by its large ionic character (40-50%) and by its partially
double bond character due to the pp dp interaction between the p
lone pair in the oxygen and the empty d orbital in the silicon.74 The
Si-O bond is highly polarized as a result of silicon's low
electronegativity (1.74); hence, large intermolecular forces would
be expected.
Surprisingly, siloxanes show large molar volume (75,5
cc/mole),75 low cohesion energy,76 low surface tension and surface
energy, low solubility parameter and low dielectric constant.74,77
These unusual properties have been explained by Rochow on the basis
of a high mobility of the methyl groups and confirmed by

spectroscopic76 and thermodynamic studies.74 The polymer's low
intermolecular forces and flexibility are also responsible for the
low Tg of polydimethylsiloxanes which, together with a high thermal
stability, make them unique for their use as elastomers in a very
wide temperature range. Additionally, polydimethylsiloxanes are
transparent to UV light, very resistant to ozone and atomic
oxygen,74,77 highly permeable to various gases, hydrophobic, and
both chemically and physiologically inert.77,78
Despite their outstanding properties, polydimethylsiloxanes
(PDMS) require high molecular weights to develop useful mechanical
properties. Even at molecular weights of the order of 106 they
exhibit cold flow and very weak rubbery properties. Chemical cross-
linking has been used to improve mechanical properties; however,
they still show low tensile and tear strength. A description of
silicon rubbers and their many uses can be found in the book by
Lynch79 and references given there.
Noshay and McGrath80 show that the two preferred general
methods for the synthesis of block copolymers are forms of living
addition polymerization, or step growth condensation
polymerization. Figure 3.2 illustrates an example of living
polymerization using butyl lithium as initiator producing a
butadiene-dimethylsiloxane diblock copolymer. An example of the
step growth condensation polymerization technique is shown in
figure 3.3.


CH2=CH-C=CH2 Bu -CH2-CH=C-CH-- Li
Isoprene (1)

Hexamethyl (Si-O)
D3, (31) CH3


Bu H2-CH=C-CH2 Si-O Li
CH3 ~H3
Poly(butadiene-b-dimethylsiloxane) (32)

Figure 3.2. Synthesis of poly(butadiene-b-dimethylsiloxane)



H2N-(OH2)3 f{Si-O Si-(CH,)3-NH2 + H2
a,m amino terminated
siloxane (33) NCO
Diisocyanate MDI, (34)

r CH3 CH3


C!H3 CH3
Siloxane-urethane segmented copolymer (35)

Figure 3.3. Synthesis of siloxane-urethane phase separated

Three desirable features ensure the success of these
polymerization techniques : 1- The location and concentration of the
active sites are known. 2- Contamination by homopolymer is
minimal. This results from stoichiometry and purity control in step
condensation and from the absence in terminating reactions in living
systems. 3- Phase length and sequence are controlled by the order
of monomer addition in living polymerization and by a good selection
of telechelomer molecular weight in step polymerization.
Living polymerization techniques can be used to produce
different types of architectures of the copolymer such as ABA, AB
and (A-B)n. This technique represents one of the best methods to
produce regular blocks exhibititng a very low molecular weight
distribution. On the other hand, step condensation methods produce
broader molecular weight distributions but they are less sensitive
to reactive impurities.
Ring opening polymerization, has been frequently applied to the
synthesis of siloxane copolymers.58 Reactive organofunctional
terminated siloxane oligomers initiate the polymerization of these
cyclic monomers where the functionalities include amino, carboxyl,
epoxy, piperazine, chloro and hydroxyl groups. Block copolymers of
the ABA type have been synthesized using a,w-aminopropyl
terminated polydimethylsiloxane as an initiator for the ring opening
polymerization of N-carboxyanhydrides of several aminoacids, figure
3.4. A PDMS having aminopropyl side chains,39.81 was employed in
the synthesis of graft copolymers.
Teyssie and co-workers82 used tetraalkylammonium salts of
carboxy terminated siloxane oligomers to initiate the ring opening

polymerization of pivalolactone (3,3-dimethyl-2-oxetanone).
Pivalolactone is an excellent choice, as the hard phase, because of
its high degree of crystallinity. Furthermore its ring strain
facilitates ring opening polymerization.

CH3 CH3 0

H2N-(CH2)3 i-O -i-(CH2)3-NH2 + PhCH,- =O
CH3 n CH3 |
amino terminated Phenylalanine N-carboxy
polysiloxane (36) anhydride (37)

SHN-(CH2) i-O i-(CH2)3-NH --CO-CH-NH-

CH3 n CHI3 IH2
Ph x
Siloxane-phenylalanine segmented copolymer (38)

Figure 3.4. Synthesis of PDMS-Polypeptide block copolymer.

As mentioned at the beginning of this chapter, two aspects
will be considered in the synthesis of siloxane based thermoplastic
elastomers: improvement of the thermal stability of the siloxane
soft phase and the effect of ring size of the lactone hard phase, in
the properties of the graft copolymers. The synthesis of these
copolymers is described in the next section.

Synthesis of the Soft Phase in Siloxane Based Multiphase elastomers

Two synthetic schemes were carried out in an attempt to
study the influence of the length of the siloxane backbone on the
microphase separation of the graft copolymer :
1- Step polymerization from dichlorosilanes.
2- Ring opening polymerization from cyclic monomers plus an

silane (9)

S (CH2)3
+ I
dichlorosilane (10)

1- H20
2- H2SO4

"~ Si-O-Si-O-Si ,0-*
CH3 CH3 (H2)3
Carboxy functionalized H
siloxane (lla)

Figure 3.5. Synthesis of polysiloxane with

Synthesis of Polvsiloxanes from Dichlorosilanes

pendant carboxyl

This method, described by Katayana,83 consists in the addition
of an ether solution of dimethyldichlorosilane (9) and

methylcyanopropyldichlorosilane (10) to an excess amount of water,

followed by refluxing of the resulting product with aqueous
sulphuric acid in order to hydrolyze the cyano group to the
carboxylic acid, figure 3.5.
Progress of the hydrolysis was monitored by IR spectroscopy,
figure 3.6, following the slow disappearance of the cyano stretching
signal (2240 cm-1) and the increasing intensity of the signal for
carbonyl stretching (1700 cm-1).


2240 cm"1

1700 cmTL1

4000 Wavenumber cm1 600

Figure 3.6. IR of polysiloxane

The mixture of the linear polysiloxane (11a) and cyclic oligomers
was separated by distillation under reduced pressure and analysis of
the volatile fraction by 1H NMR and IR spectroscopy showed that the
carboxypropyl functional group was incorporated in the linear
polymer. Furthermore, the 1H NMR spectrum of the polymer, figure

3.8, allowed calculation of the number of methylcarboxypropyl units

per dimethylsiloxane unit by comparing the relative intensities of

the signal at 2.4 ppm (CH2-COOH) and the signal at 0.0 ppm (_Ha-Si),

used as reference. This number varied from 27 to 33, smaller than

the expected value of 37, as a consequence of the formation of the
cyclic oligomers in the reaction.


H3 CH CH3 H3 H3 a
Ph3-Si-O- i-O-; i-O- i-Oi-OPh3
CH3 CH3 H2 b CH3 CH3


7:0 6.60 50 .0 3.0 20 1.0 0.0 PPM

Figure 3.7. 1H NMR of polysiloxane.

Synthesis of Polvsiloxanes from Cyclic Monomers

This second method of synthesis involved straightforward

hydrolysis with aqueous sulfuric acid of a mixture of octamethyl-
cyclo-tetrasiloxane (D4) and the corresponding analog with

cyanopropyl groups substituting for one of the methyls, figure 3.8.

Hexaphenyldisiloxane was added to cap the end groups and minimize
scission of the polysiloxane chains by backbiting, as shown in figure

1.17, and to allow molecular weight calculation by 1H NMR end group

analysis. In fact the 3 phenyl groups on each side of the chain
account for a 30 proton signal as opposed to a 2 proton signal for the
hydroxy terminated chains in the non capped polysiloxane. Addition
of hexaphenyldisiloxane as a chain terminating agent was somewhat
troublesome as most of this compound precipitated out of the
reaction mixture (a viscous colorless oil), and its removal required
several recrystallizations from ether.

3 Si C Si\ /3
CH3. I 'CH3 + CH3... (CH2)3CN
Si O Si O
CH3/ O--Si-CH NC(CH2)3/ O-S(CHz)CN
Octamethylcyclo Tetramethyltetracyanopropyl
tetrasiloxane (22) tetrasiloxane (23)

50% H2SO4, (Ph3SiO)2

Ph3 Si-O-Si-0-Si--O Si--O"-Si-O-Si-O-Si-Ph3
CH3 CH3 (CH2)3 CH3 CH3


Figure 3.8. Synthesis of end-capped polysiloxane from cyclic

Molecular weight determination of samples of the polymers
obtained by these two different synthetic schemes was undertaken
by four different methods:

1- End group analysis.
2- Gel permeation chromatography, GPC.
3- Vapor pressure osmometry, VPO.
4- Viscometry.
The first three methods measure the number average molecular
weight, while the fourth measures the viscosity average molecular
weight. Considering that these polysiloxanes have been fractionated
by distillation under reduce pressure, it can be assumed that the
polymer samples consist of a distribution of linear molecules where
cyclic oligomers and very low molecular weight linear chains have
been eliminated. Molecular weight determined by GPC is relative to
polystyrene standards and should be taken just as a reference. It
should be noted that it is safe to compare molecular weights of
polysiloxanes prepared from dichloro silanes (uncapped) or cyclic
monomers (endcapped) only if themolecular weights are determined
by the same method. Molecular weights determined by GPC are in the
range of 25,000 to 33,000 for both products, suggesting that the
endcapping reaction did not occur efficiently, an observation
anticipated by the precipitation of the endcapper from the reaction
mixture. A more realistic figure of Mn = 5000 was obtained by VPO,
since the method was independently checked by determining the
molecular weight of a polystyrene standard Mn =12,000, obtaining a
measurement of 10,800.
The synthetic scheme is continued by treating the polymer
containing carboxyl pendant groups (11a) with a concentrated
solution of KOH for a short time (only five minutes to avoid
degradation), and the result is the corresponding salt (11b), figure

3.9. Formation of the salt was confirmed by IR, and by increase in
water solubility which led to emulsions when the product was
extracted with diethylether. This carboxylate salt is used as the
initiator for the ring opening polymerization of lactones.


,w Si-O -W -0
OH3 (CH2)3
Carboxypropyl functionalized IO
siloxane (11a)

SAq. KOH, 5 min.


S Si-O i-O 0

CH3 (CH2)3
Siloxane potassium COCOK+
carboxylate (11b)

Figure 3.9. Formation of the polysiloxane potassium

Ring Opening Polymerization of Lactones

Polymerization of Pivalolactone

Pivalolactone reacts with nucleophiles at two sites, figure
3.10. Strong nucleophiles attack the carbonyl group84,85 while weak

nucleophiles attack at the methylene carbon.86 Lenz,55 Hall,56 and
Beaman57 studied the polymerization of pivalolactone and found that



Figure 3.10. Two sites of attack at pivalolactone.

the less nucleophilic carboxylate anion polymerized pivalolactone
smoothly, opening the ring with alkyl oxygen cleavage to form a
carboxylate propagating species. Polymerization of pivalolactone
with carboxylates affords living polymers, and no termination or
chain transfer occurs. This process requires the absence of water;
therefore, drying of the carboxylate is essential.
Synthesis of poly(siloxane-g-pivalolactone), figure 1.15, was
carried out in tetrahydrofuran, THF, using the potassium salt of the
carboxyl functionalized siloxane (11b) as the initiator.59 The
reaction became heterogeneous in five minutes, yet the
polymerization continued, and a 24 hr period was sufficient to
produce monomer conversions greater than 90%. Heterogeneous
graft copolymerizations of this nature are rare, and in this case the
reaction could be done only in the presence of 18-crown-6. This
crown ether is known to complex with the potassium counterion
rendering the carboxylate anion an efficient initiator for lactone
polymerization. Control experiments showed that the crown ether

by itself did not polymerize pivalolactone, nor did hydroxy
terminated polysiloxanes. This indicates that hydroxy groups are
much slower initiators than carboxylate anions. Copolymers with
compositions ranging from 10 to 90 weight % pivalolactone were
obtained; DSC analysis showed excellent microphase separation of
the hard phase (pivalolactone) and soft phase (siloxane), while
contact angle studies indicated a surface rich in siloxane.59
A variation in this synthetic procedure, substitution of KH in
THF for aqueous KOH in figure 3.9, was attempted in order to obtain
the siloxane carboxylate, (11b). This approach simplifies the overall
procedure, as the THF solution of the carboxylate is carried directly
on to the next step, addition of pivalolactone monomer, thus avoiding
the time consuming extraction, purification and drying of the
carboxylate. Synthesis of copolymers with 30 and 50 weight %
pivalolactone was attempted; however, homopivalolactone was
obtained in most cases, and mixtures of the homopolymer and the
copolymers in the rest. The cause of the homopolymerization of
pivalolactone may well be the presence of an excess of KH besides
extraneous moisture in the solvent, which would generate potassium
hydroxide, also would produce homopolymer.
Following the method reported before,59 poly(siloxane-g-
pivalolactone)s containing 30 and 70 weight % pivalolactone were
synthesized, and these polymers afforded free standing films from
10 % solutions in hexafluoroisopropanol, HFIP. Film formation is
indicative of high molecular weight which is responsible for the
unique properties of the polymer.

High molecular weight, together with a narrow molecular
weight distribution, assured by the living nature of the synthetic
technique and the high percent crystallinity of polypivalolactone,
contribute to the achievement of excellent phase separation in the

Polymerization of Other Lactones

The polymerization of the five and six membered lactones, y
and 8 valerolactone, figure 3.11, was attempted in order to study the
effect of the ring size on the properties of the graft copolymer.

0 0



y-valerolactone (39) 8-valerolactone (40)

Figure 3.11. Structures of y and 8 valerolactones.

Carothers found that while six membered ring lactones
polymerized, five membered ring lactones did not.63 This was
confirmed by Hall and Schneider,87.88 who systematized the
information on the polymerization of cyclic esters, urethanes, ureas,
imides and amides, and added their own contribution to supply a
wealth of information on this field.

In our hands the ring opening polymerization of 8-valerolactone
by the carboxylate functionalized siloxane proceeded in the presence
of 18-crown-6 in THF at 90C, figure 3.12. A soluble graft
copolymer was obtained, as confirmed by GPC, where the siloxane
(11b) molecular weight was calculated as 5000 while the copolymer
showed a single signal, molecular weight 15000. A temperature
much higher than that required for the polymerization of
pivalolactone (room temperature) was necessary for the
polymerization of 8-valerolactone since the latter contains less ring
strain. On the other hand, y-valerolactone did not polymerize,
regardless of the experimental conditions employed.


O + SiO \.-Si-O-0 .
CH3 (CH2)3COO-K+
(40) (11b)
18-crown-6, 90C

-\\ Si-0-lSi-O0J-Y

CH3 (CH2)3C O-(CH2)4-C
O O nf

poly(siloxane-g-8-valerolactone) (41)

Figure 3.12. Synthesis of poly(siloxane-g-8-valerolactone).


In conclusion the ring strain of the lactone is directly related
to its polymerizability and to the conditions under which
polymerization occurs. The polymerization required the presence of
18-crown-6 to increase the nucleophilicity of the carboxylate anion
by trapping the potassium counterion. While the siloxane-
pivalolactone graft copolymer (12) was insoluble in most organic
solvents, the siloxane-8-valerolactone counterpart (41) was easily
soluble in organic solvents due to its relatively low molecular
weight, 15,000 (GPC).
Improving the solvent resistance of siloxane based multiphase
polymers has been one of the goals of this dissertation. With this in
mind, other cyclic compounds such as lactams were considered as
hard phases. This chemistry is discussed in the next chapter.


The synthesis of siloxane-lactone graft copolymers was
discussed in the previous chapter, where the combination of soft and
hard phases produced film forming thermoplastic elastomers. A
similar approach will be described in this chapter, substituting
lactams for lactones.
While siloxane-polyamide block and segmented copolymers
have been studied before,89,90 little is known about siloxane-
polyamide graft copolymers. Polyamides are mechanically very
strong and have a wide range of applications in industry. The
combination of siloxanes and polyamides could lead to other
desirable properties such as enhanced impact resistance, excellent
solvent resistance, elastomeric behavior and altered surface
Amide linkages have cohesive energies (8.5 kcal/mole) that are
intermediate between ureas (14.3 kcal/mole) and imides (4-6
kcal/mole). Although polyurethanes produce stronger intermolecular
hydrogen bonding than polyamides, polyurethanes are thermally
unstable. Therefore when hydrogen bonding and thermally stable
polyamides are combined with elastomeric low Tg siloxanes, novel
elastomers with improved solvent resistance and good thermal
stability might be obtained.


Polyamides are produced by the reaction of diamines with
dicarboxylic acids in which the linkage forming the main chain of
the macromolecule occurs through the formation of amide bonds.

Nylon 6 and Nylon 66

The generic term "nylon" has been adopted by the U.S. Federal
Trade Commission for designating fibers of the polyamide type.91
The common nomenclature uses the word nylon, followed by numbers
designating the number of carbons constituting the repeat unit of
the polymer; for example, the reaction between hexamethylene
diamine (42) and adipic acid (43) produces nylon 66 (44), figure 4.1,

H2-N-CH2-(CH2)4-CH2-NH2 + HOOC-(CH2)4-COOH
hexamethylenediamine (42) Adipic acid (43)


NH-CH2-(CH2)4CH2) 2-NH-CO-(CH2)4-C
Nylon 66 (44)

Figure 4.1. Synthesis of nylon 66.

while the polyamide from caprolactam (45), which contains 6 carbon
atoms, is known as nylon 6, (46) figure 4.2. Polyamides, together
with polyester and polyacrylonitrile, constitute the most important
group of polymers in the production of textile fibers.91

-H -- 4 NH-(CH2)5-CO-

Caprolactam (45) Nylon 6 (46)

Figure 4.2. Synthesis of nylon 6.

The first polyamide was synthesized by Gabriel and Maas in
1899.92 They observed that upon heating E-caprolactam a very tough
material was produced; however, they did not realize the
importance of their observation. The synthesis and development of
nylon as the first synthetic fiber was the product of the genius of
Wallace Hume Carothers and the observation of his chief associate
Julian Hill. Carothers, then a professor at Harvard University, was
attracted to E. I. du Pont de Nemours & Co. in 1928, where he was
allowed to conduct fundamental research, without regard for any
immediate commercial objective.93 Du Pont had become involved in
the production of cellulose acetate fibers in 1929, and it is a credit
to the vision of its management that Carothers was allowed
complete freedom in his work and given all the resources he
requested, which in the end led not only to nylon but to neoprene and
to macrocyclic compounds such as the synthetic musks.93

Carothers' initial interest was only general polymer research.
For some time he had been mentally exploring the possibility of
imitating nature in the creation of "giant" molecules such as cotton,
silk and wool.94 He reasoned that just as ethyl alcohol and acetic
acid will react with each other to form ethyl acetate so, in a similar
way, would a dialcohol react with a diacid, leading to a product with
reactive ends that could further react in the same manner to produce
a compound with a longer carbon backbone, as shown in figure 4.3.

Dialcohol Diacid

[ 0 0
--O-R-O-C-R'-C 1--


Figure 4.3. Synthesis of Carothers superpolymer.

Nylon Fibers

The need for difunctional starting materials is one of the
fundamental principles of step polymerization. By applying this
method and using a molecular still, Carothers assistant Julian Hill
observed that the compounds he had been working with had displayed
profound changes. The new material was tough and elastic and more
importantly, fibers were formed when a glass rod was dipped into

the material and lifted. Carothers called these materials
superpolymers. It was suggested that the discovery might be useful
as a textile fiber; however, the product had low melting point, was
soluble in hot water and in common cleansing fluids.
Carothers turned his attention to polyamides, prepared a wide
range of diamines, from C2 to C18, evaluated them as monomers and
reacted them with a variety of aliphatic diacids to produce the
corresponding polyamides.63 More than 100 superpolymers were
prepared which produced fibers that resisted heat, washing and dry
cleaning. This new group of synthetic materials was given the
generic name nylon. Nylon was the first purely synthetic fiber, and
newspapers announced its discovery as one of the most important in
the century.94
Following publication in the early 1930s of Carothers research
on polyamides and polyesters, Schlack95 in 1937, synthesized nylon
6 from e-caprolactam using e-aminocaproic acid as catalyst, at I.G.
Farbenindustrie. The I.G. Farben group succeeded in commercially
producing nylon 6 monofilaments under the trade name of PerluranT
in 1939. Du Pont produced nylon 66 fine denier fibers for
experimental evaluation in April 1937,96 its first commercial plant
for the production of nylon 66 began operations in December 1939.
In May 1940, fine denier nylon was available nationwide.97 It is a
tribute to the vision of the earlier developers such as Carothers and
Schlack that still today nylon 6 and nylon 66 represent the largest
volume of total polyamides converted to fibers.
In 1932 Carothers and Hill98 stated the requirements for fiber
forming polymers as follows: 1- The polymer must consist of very

large molecules having a number average molecular weight of not
less than 12,000 and a molecular chain length of not less than 1000
A. 2- The polymer must be capable of crystallizing. 3- Three
dimensional polymers such as cross-linked polyamides and
polyesters are generally unsuitable.
These requirements have remained relatively unchanged
despite the large amount of research that has followed the original
work. Since small molecules and polymer molecules with molecular
weights below a critical value (called the molecular weight for
entanglement) are weak and readily attacked by solvents and other
reactants, it is apparent that the unique properties of polymers are
related to molecular weight. Thus toughness, impact strength, melt
viscosity, modulus, tensile strength, thermal stability and
resistance to corrosives are dependent on molecular weight and
molecular weight distribution. However, while a molecular weight
above the critical value is essential for most practical applications,
extremely high molecular weight polymers are very expensive to
The critical value for molecular weight depends on Tg, the
cohesive energy density for amorphous polymers and the degree of
crystallinity for crystalline polymers. The degree of crystallinity
depends on the molecular constitution and stereoregularity of the
chains. With the exception of polymers with high stereoregularity,
such as isotactic polypropylene, strong intermolecular forces such
as hydrogen bonding are required for fibers. Thus, while a molecular
weight of 12,000-15,000 is sufficient for a commercially useful

polyamide fiber, molecular weights of the order of 106 are
necessary for polyethylene body implants.
It is important to note that in 1945 it was reported that
Farbenindustrie had evaluated more than 3,000 polyamide
constituents without finding any commercially feasible polyamides
with important improvements over nylon 6 and nylon 6,6. At the
same time research on this field was being carried out at Du Pont
and several other companies.93

Polymerizabilitv of Lactams

While at du Pont Hall put together the existing information on
the polymerizability of lactams88 and contributed his own share to
this field. With the information gathered he made some
generalizations: (a) "Water and sodium hydride were equally
effective catalysts at temperatures above 200 C. At room
temperature water was ineffective but sodium hydride was still
potent". (b) "Ring opening polymerization of lactams was a
restricted technique, with the majority of lactams failing to
polymerize". (c) "Of the five 8- and 9-membered ring lactams
examined all polymerized. Approximately one half of the various 7-
membered ring compounds studied to that date (1958) polymerized.
Most six membered ring lactams did not polymerize." (d) "Aryl and
alkyl substituents on a ring decrease polymerizability, especially if
they are on the nitrogen atom of the lactam".
These statements have been revised as a consequence of the
vast amount of work performed on this field in the following years.
For example, water and sodium hydride are not catalysts but

initiators of the reaction and their mode of action is very different.
Polymerization of lactams can be initiated not only by water and
bases but also by acids.99-102 Water initiated polyamidation is the
method most frequently used in industry followed by anionic
polymerization, while cationic initiation is not useful because the
molecular weights of the polymers that result are not sufficiently
high. Nylon 6 is the material of major importance in commercial
production, while nylon 12 and nylon 8 are of lesser importance.
Hydrolytic polymerization of caprolactam is carried out
commercially in batch and continuous processes by heating the
monomer in the presence of 5-10% water to temperatures of 250-
2700C. Three equilibria have been proposed for the water initiated
polymerization of lactams :
1-Hydrolysis of the lactam to the amino acid;


H + H20 HO2C(CH2)5NH2 eq 4.1

2-Step polymerization of the amino acid with itself;

2 HOgC(CH2)5NH2 --- w CO-NH + H20 eq 4.2

3-Ring opening polymerization of the lactam by the amine
group of the amino acid;

HO2C(CH2)5 NH -[OC(CH2)5NH]-H eq 4.3

followed by propagation in the same way as reaction 4.3.
Ring opening polymerization is the most likely route, while
step polymerization of the amino acid with itself accounts only for
a very low percentage of the reaction. Polymerization is acid
catalyzed as indicated by the fact that amines or NH2(CH2)5COONa
are poor initiators in the absence of water. Therefore initiation and
propagation are believed to occur by the following sequence

S + ""NH2 -,NHCO(CH2)5NH3+

0 0
H+ .NHCO(CH2)5NH3+ H2+ -NHCO(CH2)5NH2

On the other hand, anionic polymerization of lactams occurs
via the lactam anion which reacts with another molecule of
monomer by ring opening transamidation.99-102
A series of new lactams have been synthesized and, in most
cases, successfully converted into the corresponding polymers. The
traditional hypothesis that five and six membered lactams are very
stable and therefore unpolymerizable has been revaluated. Even
substituted rings which until recently were considered
unpolymerizable successfully incorporated as linear units into

polymer or copolymers.109 Most of these lactams have been
polymerized under anionic conditions, which require the use of dry
solvents, and solubility of the forming polymer becomes the factor
controlling its molecular weight. As mentioned earlier, most
industrial processes for the production of nylon 6 use water as
Bearing in mind that the main goal of this dissertation is the
synthesis of thermoplastic elastomers using the soft-phase, hard-
phase approach, E-caprolactam was chosen as the monomer to graft
onto the functionalized siloxane, because of polycaprolactam (nylon
6) high crystallinity, high melting temperature and commercial
availability. According to M.I. Kohan,104 nylon was a new concept in
plastics for several reasons. It was crystalline which meant a sharp
transition from solid to melt and it meant higher service
temperatures. A combination of toughness, rigidity and "lubrication
free" performance, peculiarly suited it to mechanical bearing and
gear applications. Nylon acquired the reputation of a quality
material by showing that a thermoplastic could be tough as well as
stiff and do jobs better than metals in some cases. This
performance gave nylon the label "an engineering thermoplastic".

Polysiloxane-Lactam Graft Copolvmers

Polymerization of caprolactam was carried out in two steps:
1-the monomer was mixed in a thick reaction tube with catalytic
amounts of water, frozen, evacuated and sealed. The reaction
mixture was then heated to 250-2600C for several hours; 2- the
tube was cooled opened and heated again at 1800C to distill the

water as well as to increase the molecular weight of the

0 JH3 3
N-H + O-S- i-O-

SH3 H2)3
e-caprolactam (45) Carboxypropyl
siloxane (11a)
1-H20, 260C
sealed tube
2-N2, 180C

cpH3 3

(IH3 ( H2)3
--1-(c2)5- c

Poly(siloxane-g-e-caprolactam) (47)

Figure 4.4. Synthesis of poly(siloxane-g-E-caprolactam)

The chemistry was conducted combining a one to one mixture
(W/W) of caprolactam and functionalized siloxane. The final product
was extracted with dichloromethane, ether and acetone to eliminate
any unreacted mixture, then dissolved in m-cresol and recrystallized
by adding diethylether. The purified polymer was dissolved in
hexafluoro isopropanol (2 % by weight) and free standing films were
easily produced by pouring the solution in a Petri dish and letting
the solvent evaporate. Formation of free standing films is a

condition "sine qua non" for the potential use of a polymer as
thermoplastic elastomer or engineering material. Only high
molecular weight polymers fulfill this condition as the
intermolecular interactions, through hydrogen bonding of the amide
groups, must overcome intramolecular forces and entropy driven
random coils and thus induce crystallization.

Characterization of Polv(siloxane-g-E-caprolactam)

The siloxane-lactam copolymer was analyzed by 1H NMR, 13C
NMR, IR, XPS and water contact angle techniques. The 1H NMR
spectrum, figure 4.5, allowed analysis of siloxane-lactam ratios.

0.0 ppm 3.2 ppm
-L- 1.2-1.6 ppm
I I 2.2ppm
-Si-O-Si-O- ppm A
(C H) 0.0 1.0
CH3 2.2 5.3

3.8 PP I 1.5 -68.

Figure 4.5. 1H NMR of poly(siloxane-g-E-caprolactam) 4/96

By comparing the integration of the signals at 8 = 0.00 ppm
CH3-Si ) and 8 = 2.2 ppm ( CH2-CO ), a siloxane content, in weight %,
of 1.7-4.0 % was calculated. Although the weight % of both
monomers was initially the same, in all reactions, most of the
siloxane was recovered by extraction with ether, dichloromethane or

acetone. Carbon NMR, figure 4.6, which is less sensitive than proton

NMR, did not show signals for the CH3-Si groups due to the low
siloxane content in the copolymer, while the signal for the C=O of

the lactam segment (178.4 ppm) was noticeably intense. It also

showed individual signals for the other five lactam carbons.
IR of copolymer films, figure 4.7, showed the characteristic

signals of polyamides (cm-1) : 3250, 3050, N-H stretching of
polymeric amides,106 1625 and 1530, amide bands 1 and 2,

corresponding to C=O stretching and 1070, Si-O stretching.


I18 PPH 148 128 188 88 60 48 28 0

Figure 4.6. 13C NMR of poly(siloxane-g-e-lactam) 4/96 w%.




cm 3000 2500 2000 1500 1000 600

Figure 4.7. IR of poly(siloxane-g-E-caprolactam).

Surface analysis of the films by XPS, figure 4.8, shows a

61.1/18.3, C/Si ratio which indicates a surface rich in siloxane

when compared to 50/25 for PDMS and 74.6/0.6 for a copolymer

where both the siloxane and lactam segment shared the surface.

This siloxane rich surface should be hydrophobic, as confirmed by

the water contact angle = 90.50 measured on the copolymer films.

532.20, 01S
-285.05, ClS

399.95 N1S r/

SI-- 102.65, Si2p

600 400 0oo
Binding energy (eV)

Figure 4.8. XPS of poly(siloxane-g-E-caprolactam).

Phase segregation of the siloxane and the lactam components

is also observed by optical microscopy using 90 polarized light; a
picture of a film of poly(siloxane-g-e-caprolactam) is shown in

figure 4.9. The crystalline phase, show birefringence and appears as
a white solid while the amorphous phase constitute the black

background in the picture.

Copolymers like the one mentioned above are interesting as a
surface modified nylon, and could be of commercial importance.
Nylon is a hygroscopic material and this is one of the main causes
for its deterioration upon long exposures, specially in very humid
environments. Given the commercial importance of nylon, extensive
research has been dedicated to its surface modification. Siloxanes
and highly fluorinated polymers are most frequently used as surface
modifiers. Due to their large molar volumes, low cohesive energy
densitities and high flexibility (low Tg), polydimethylsiloxanes have
low surface tension and surface energies.107 The surface tension
for high molecular weight PDMS 21-22 dynes/cm is at least 10
dynes/cm lower than that of many other polymers such as
polystyrene (30-34 dynes/cm), poly(ethylene terephthalate) (40-43
dynes/cm), and nylon (38-42 dynes/cm).107 Most polymers
interfacial surface tensions are below 10 dynes/cm, and as a result
the air-polymer surfaces of siloxane containing copolymers as well
as their blends are enriched in the lower surface energy siloxane.108
Properties which can be improved by surface modification with
siloxane are biocompatibility, hydrophobicity, surface finish and
gloss, release properties, reduction in friction and atomic oxygen
Siloxane segmented, block and graft copolymers have been
shown to be very useful in polymer surface modification through
blending.109 The organic component of the siloxane can be chosen to
be similar with the base polymer in order to achieve compatibility.
Siloxane homopolymers are usually incompatible with the base
polymer and thus may eventually separate from the material surface.

As opposed to blends, the siloxane phase in siloxane copolymers is
chemically bonded to the organic counterpart therefore macrophase
separation is not a problem.
Until recently surface modification of nylon through siloxane
graft copolymers has been elusive. However the successful
synthesis of surface modified nylon as described herein has been
accomplished through the grafting of caprolactam to polysiloxane
with carboxyl pendant groups, as confirmed by XPS and water
contact angle measurements.
Since our original goal was to synthesize thermoplastic
elastomers, it is important to be able to produce a copolymer with
higher percentage of siloxane to introduce the necessary elasticity.
The initial method used to prepare the copolymer yielded products
with up to 4 % siloxane.
In an attempt to increase the incorporation of siloxane in the
copolymer (47) the initial reaction mixture was dissolved in toluene
and refluxed for several hours at 1200C, then the temperature was
increased to distill the solvent to a maximum of 250C. The product
was purified as before and characterized by IR, 1H NMR and 13C NMR.
Although the copolymer (47) contained 18 % by weight of the
siloxane segment, calculated by 1H NMR, figure 4.10, it did not,
produce free standing films. Calculation of the chain length of the
lactam segment for this copolymer revealed 111 repeat units while
the one containing 4 % siloxane, corresponds to 695 repeat units.
This is equivalent to molecular weights of 12,500 and 78,000

8 ppm 3.5 3.2 2.2 1.7-1.1 0.0
Area 1 2 2 6 2

3.5 p Is '

Figure 4.10. 1H NMR of poly(siloxane-g-E-caprolactam) 18/82

In contrast to the copolymer containing 4 % siloxane, the 13C
NMR of the copolymer containing 18 % siloxane, figure 4.11, showed

a signal for the CH3-Si while the signal for the C=0 was very small.

It may well be that in the initial step in the reaction using
toluene the lactam/linear poyamide equilibrium leans towards the
lactam in a higher percent than in melt reactions, and consequently
the molecular weight of the graft is decreased. The final product in

the commercial production of nylon 6, contains 8-9% of caprolactam

as well as 3% of cyclic oligomers.101


180 100 0.0 (ppm)

Figure 4.11. 13C NMR of poly(siloxane-g-E-caprolactam) 18/82


Reaction of E-caprolactam with carboxy functionalized
siloxanes, in the melt, afforded film forming polysiloxane-nylon
graft copolymers with 1.7-4.0% siloxane. These copolymers show
excellent phase segregation as indicated by surface analysis (XPS
and water contact angle measurements).
Similar reactions carried out under dilute conditions provided
graft copolymers with a higher siloxane content (18%) however, the
molecular weight of the nylon segment was not high enough to
induce formation of free standing films. Most likely these dilute
conditions favored the formation of cyclic products over high
molecular weight linear poycaprolactam chains. The film forming
poly(siloxane-g-e-caprolactam) could be of commercial use as a self
lubricating nylon.



All NMR spectra were obtained on a Varian XL-200 spectrometer
using deuterated chloroform as solvent and TMS as reference;
chemical shifts reported in ppm. For polymers containing siloxanes,
no TMS was used, the methyl signal of the methyl siloxane was used
as reference instead. IR spectra were recorded on a Perkin Elmer FT
spectrophotometer or a Perkin Elmer 281. GC was performed on a
Hewlett Packard 5880A series gas chromatograph, equipped with a
FID detector and a fused silica 0.31mmx50m capillary column
packed with a 0.17m film of SE-54 (methyl-phenylsilicone), HPLC on
a liquid chromatograph with the following conformation: U6K
injector, a Waters 590 pump, a variable wavelength UV-VIS
Spectroflow 757 detector and a Waters Novapak C18 column
150x3.9mm. GPC was carried out on a liquid chromatograph equipped
with a Waters M6000A pump, a concentration sensitive differential
refractometer and two attached Phenogel columns (300x 7.8 mm,
cross-linked polystyrene-divinylbenzene, 500 and 50000A), all data
was collected and analyzed on a Zenith personal computer model 48
equipped with a Metrabyte multi-IO card and an Epson dot matrix
printer. VPO was performed on a Wescan 233 molecular weight

apparatus, XPS on a Kratos XSSAM 800instrument with Mg anode at
12 KV, 19 mAmp, 3x10-8 Torr, software DS800.


All solvents were reagent grade, THF was refluxed overnight
over a sodium and potassium alloy (2:1) and distilled under argon
before use. The argon was purchased from Air Products and dried by
passing it through sulfuric acid, concentrated sodium hydroxide and
calcium sulfate. Pivalic acid (PVA), 2-(2-methoxyethoxy)ethanol
(MEEtOH), DMAP, DCC, were purchased from Aldrich and vacuum dried
before use and titanium tetrabutoxide from Fisher Scientific.
D,L-alanine (Eastman Kodak) was recrystallized from water,
Vacuum sublimed and stored in a desiccator containing calcium
Chlorosilanes, cyclic siloxanes hexaphenyl disiloxane, were
purchased from Petrarch (Huls America) and stored in a refrigerator.
One gallon of dimethyl,dichlorosilane was donated by Dr Paul Kremer
of PCR Inc at Gainesville. Pivalolactone was donated by Dr Gus Ibay
of Southern Research Institute; 8-valerolactone and y-valerolactone
were purchased from Aldrich. All lactones were processed before
use by stirring them overnight with cacium hydride and distilling
under reduced pressure.
E-caprolactam, (gold label, Aldrich) was recrystallized from
cyclohexane, dried in a vacuum oven at 400C for 10 hrs and stored in
a desiccator containing calcium sulfate.

Esterification of Pivalic Acid. Model Studies

Synthesis of 2-(2-Methoxvethoxv)ethyl Pivalate(29M

A solution of pivalic acid (27) (3.9 nmol), MEEtOH (28)(5.0
mmol), DMAP (0.53 mmol), DCC (3.8 mmol), in diethylether (25 cc)
was left stirring under argon at RT for 15 hrs. The N,N-dicyclohexyl
urea formed was filtered and the filtrate washed with water (3x25
cc), 5% acetic acid (3x25 cc) and dried with magnesium sulfate. The
crude product was recovered after evaporation of the ether under
reduced pressure, purified by CC (Al203, hexane-dichloromethane
gradient) and characterized as follows:
White solid, m.p.: 178-90C.
IR (film, cm-1): 2890(s), C-H str.; 1725(m), C=O str.; 1120(s),
C-O str.
1H NMR (6,ppm): 1.95,s,3H; 1.30,s,3H; 1.32,s,3H, corresponding
to pivalic acid methyl groups; 3.9,s,3H, OCtH, 3.5-3.75,m,6H,
corresponding to the internal methylene groups of the ester; 4.25 ,t,
(j=7Hz), 2H, COO.-2.
13C NMR (8,ppm): 27.02 (CHJ)3C; 38.59 (CH3)3.; 58.81, OaH3;
61.44,COOCH2; 178.3 C=0.
M. Sp. (m/e, %1): 204, 21,M+; 173, 0.29, loss OCH3; 129, 100,
loss OCH2CH20CH3, 103, 19.7 pivalate cation; 85, 14, pivalate minus
CH3; 57, 42, (CH3)3C+.

Esterification Reactions.

These reactions were carried out in 50 cc 3 necked flasks with
a thermometer in one neck, a rubber septum in the other and a
condenser in the center, connected to a Schlenk line (argon, vacuum).
Equimolar amounts of pivalic acid (27) and MEEtOH (29) were dried
separately under vacuum, overnight, then mixed together under argon
and magnetic stirring. Two different ways of removing the water
formed in the reaction were employed: 1- Vacuum, at room
temperature. 2- A stream of argon and heating to the refluxing
temperature of the mixture, 160C at the beginning and 190C at the
end. Samples (0.05-0.10 cc) were syringed out through the septum
with argon filled syringes and analyzed by HPLC. When the reactions
were under vacuum, the vacuum was quenched with argon before
Typically four simultaneous reactions were run; the first
reaction was a control containing pivalic acid and MEEtOH only, the
second one contained the acid and the alcohol plus a catalytic
amount of alanine, the third one contained a catalytic amount of
Ti(OBu)4 instead of alanine, and the fourth contained both alanine
and Ti(OBu)4, Table 5.1

Table 5.1. Pivalic acid esterification.
Reaction Reagents in mmol
number PVA MEEtOH ALA Ti(OBu)4

1 20 20 0.00 0.00
2 20 20 0.49 0.00
3 20 20 0.00 0.29
4 20 20 0.47 0.29

Chromatoaraphv of Reaction Mixtures

Samples of the reaction mixture were collected (0.05-0.10 cc),
diluted to 4.0-5.0 cc with HPLC methanol and filtered through 0.45i
Millipore filters. Samples containing Ti(OBu)4 were passed through
a 1 cm layer of alumina (80-20 mesh) on a 2.3 cm o.d. coarse pore
sinter glass funnel, and filtered again through a 0.45 l filter. The
analysis was performed in a liquid chromatograph, described
previously, using a 6/4 mixture of acetonitrile/water as solvent at
a flow of 0.5 ml/min. Acid ester ratios were calculated by the
respective areas measured at 215 nm. Figure 5.1 shows HPLC
chromatograms for samples taken at different times from from the
esterification reaction of pivalic acid.

Pivalic ester


1 2 3 4

Figure 5.1. HPLC of pivalic acid esterification.

Transesterification of Methyl Benzoate. Model Studies

Transesterification Reactions

Transesterification reactions were carried out in a similar
way as the esterification of pivalic acid. In order to assure
uniformity in the four reactions run simultaneously, a new glass
line was designed, see Figure, consisting of a horizontal tube
connected to a cold trap in one end and a three way stopcock on the
other and four outlets in between. Each reaction flask ( 50 cc, 3N
round bottom) was attached to the line through 25.0 cm vertical
tubes. The horizontal tube was wrapped with a heating tape, set at
1200C, to assure evaporation of the methanol liberated in the trans
esterication. Reactions were carried out at RT under vacuum or

heated to the refluxing temperature, under argon. Samples were
syringed out through a septum, purified in the same way as the
pivalates and analyzed by HPLC at 230 nm or by GC. A typical set of
reactions, is presented in Table 5.2.

Table 5.2. Methyl benzoate transesterification
Reaction Reagents in mmol
number Me Bz MEEtOH ALA Ti(OBu)4

1 20 20 0.00 0.00
2 20 20 0.00 0.15
3 20 20 0.34 0.00

Chromatoaraohv of Transesterification Reactions

High performance liquid chromatograohv

Samples of the reaction were purified and analyzed by HPLC
under the same conditions as the pivalate samples, except for the
maximum absorption wavelength, 230 nm for the benzoate as
opposed to 215 for the pivalate, see Figure 5.2.
A sample of the transesterification product was collected
from the HPLC effluent and used for mass spectrometry.

Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EVE1ZS75K_UB2O6P INGEST_TIME 2013-01-23T16:19:36Z PACKAGE AA00012985_00001