Synthesis and kinetic investigations of macromolecular catalysts based upon the spacer atom concept

MISSING IMAGE

Material Information

Title:
Synthesis and kinetic investigations of macromolecular catalysts based upon the spacer atom concept
Physical Description:
xi, 108 leaves : ill. ; 28 cm.
Language:
English
Creator:
Langenmayr, Eric Jon, 1951-
Publication Date:

Subjects

Subjects / Keywords:
Macromolecules   ( lcsh )
Catalysts   ( lcsh )
Organic compounds -- Synthesis   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis--University of Florida.
Bibliography:
Includes bibliographical references (leaves 86-88).
Statement of Responsibility:
by Eric Jon Langenmayr.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000098608
notis - AAL4055
oclc - 06808273
System ID:
AA00003879:00001

Full Text











SYNTHESIS AND KINETIC INVESTIGATIONS
OF MACROMOLECULAR CATALYSTS BASED
UPON THE SPACER ATOM CONCEPT










BY


ERIC JON LANGENMAYR


A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL CF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA


1980
































Digitized by te Ir?.rnet Archive
in 2011 wTiPfu nnngafrom
University of Florida, George A. Smathers Libraries with support from LYRASIS and the Sloan Foundation


http://www.archive.org/details/synthesiskinetic001ang













ACKNOWLEDGEMENTS

The author wishes to express his appreciation to his

research director, Dr. James A. Deyrup, for suggesting this

problem and to acknowledge the large debt owed to Dr. Deyrup

for his assistance in the course of this research. Thanks

also to Donna McRitchie and Neil Weinstein for encouragement

and advice. Special thanks go to Curtis Lege for many en-

livened discussion about chemistry.

To the author's family-in-law goes his deep apprecia-

tion for their continued support and understanding and also

for copious supplies of lo-mein and sweet and sour pork.

To the author's family goes his heartfelt thanks for

their unrelenting support during the long period of his

education. He's finally out, Mr. Langenmayr!

To his wife and companion, Nina, who typed this dis-

sertation and put up with a cantankerous husband for the

past year, the author gives his love and appreciation. Now

its your turn.


iii

















TABLE OF CONTENTS


ACKNOWLEDGEMENTS . .

LIST OF TABLES . .

LIST OF FIGURES . .


ABSTRACTL *

CHAPTER I


CHAPTER II

















CHAPTER III


THE SPACER ATOM CONCEPT:
A NEW APPROACH TO THE SYNTHESIS
OF MACROMOLECULAR CATALYSTS. .

Introduction . .

Discussion . .

Design of Polymer Matrix .

Proposal of Research .


SYNTHESIS AND CLEAVAGE OF
MODEL COMPOUNDS AND POLYMERS .

Synthesis of Model Compounds .

Cleavage of Model Compounds. .

Synthesis of Spacer Atom Monomer

Synthesis of Polymers. .

Cleavage of Polymer Bound
Spacer Atom Monomer .


KINETIC EVALUATION OF POLYMERS .

Introduction . .

Preliminary Experiments. .


Page

iii

vii

viii

x










14

17




21

21


24

28

o 31


36


38

38

. 40









Burst Kinetics . .. .


Comparison of Catalytic Efficiencies
of Random-PS-DVB-Benz-OH 43 and
Ordered-PS-DVB-Benz-OH 45 for
Hydrolysis of DNPB . 57

Kinetics in Deuterium Oxide. .. .... 62

Discussion . 63


CHAPTER IV EXPERIMENTAL . 66

Introduction . 66

Syntheses . .. 67

N-Hydroxymethyl Urethane 17- 67
1-Carbethoxyaminomethyl-
benzimidazole 19
A. . 67
B. . 68
Phenyl Carbamate 20. ... 68
Benzyl Carbamate 22. ... 68
1-Carbenzyloxyaminomethyl-
benzimidazole 23 68
1,1'-Ureidomethylbenzimid-
azole 25 . 69
NaH/DMSO Cleavage of 1-Carbethoxy-
aminomethylbenzimidazole 19. 69
NaH/DMSO Cleavage of Urethane 16 70
NaH/DMSO Cleavage of 1,1'-
Ureidomethylbenzimid-
azole 25 . 71
Attempted Cleavage of 3-t-Butyl-
5-Phenyl-4-Oxazolin-2-One 29
with NaH/DMSO. ... 71
Di-(N-Benzimidazolyl)Methane 30. 71
NaH/DMSO Cleavage of Di-(N-
Benzimidazolyl)Methane 30. 72
5(6)-Aminobenzimidazole Hydro-
chloride Monohydrate 72
5(6)-Acrylamidobenzimidazole 34. 72
P-Acrylamidobenzyl Alcohol 36. 73
P-Acrylamidobenzyl Carbamate 37. 73
1-Carb-(4'-Acrylamido)Benzyloxy-
aminomethyl-5(6)-Acrylamido-
benzimidazole 38 74
P-Nitrobenzyl Carbamate 39 75
1-Carb-(4'-Nitro)Benzyloxyamino-
methyl-5(6)-Acrylamidobenz-
imidazole 40 . 76


. 49








General Synthetic Method for
Macroporous Polymer
Catalysts ... .. 77
NaH/DMSO Cleavage of Spacer
Atoms from PS-DVB-SAM 44. 78
Attempted Cleavage of Acet-
anilide with NaH/DMSO 79
Treatment of Random-PS-DVB-Benz-
OH 43 with NaH/DMSO 79
Development of Method for the
Selective Isocyanation of
Benzyl Alcohol in Random-
PS-DVB-Benz-OH 43 and Or-
dered-PS-DVB-Benz-OH 45 80
General Method of Reaction of
Polymers with Isocyanates 80
General Method of Selective
Cleavage of Isocyanate from
Random-PS-DVB-Benz-OH 43 and
Ordered-PS-DVB-Benz-OH 45 81
2, 4-Dinitrophenyl Benzoate
(DNPB) . 81
Polymer Surface Areas 82


Kinetics . 82

General Kinetic Method. .. .. 82
Error Analysis. .. 84
Preparation of Ester Solutions
Used in Kinetics. 84
Preparation of Buffer Solutions
Used in Kinetics. 84


REFERENCES . . 86

INFRARED SPECTRA . . 90

KINETIC RUNS . . 99

BIOGRAPHICAL SKETCH. . . 108














LIST OF TABLES

Table Page

I-1. Second order rate constants for hydrolysis
of 4-nitrocatechol sulfate. . 4

III-1. Effect of stirring on the pseudo first
order rate constant, kobs, for the
hydrolysis of PNPA catalyzed by PS-
DVB-Benz. . . .. 41

III-2 Effect of the amount of catalyst on the
pseudo first order rate constant, kobs'
for hydrolysis of DNPB catalyzed by
PS-DVB-Benz . . 41

III-3. Determination of degree of adsorption of
DNPB on PS-DVB. . ... 43

III-4. Effect of added DNP on the observed
pseudo first order rate constant, kobs 46

III-5. Effect of pH on the observed pseudo
first order rate constant, kobs, for
the hydrolysis of DNPB in the presence
of PS-DVB-Benz or PS-DVB. . ... 49

III-6. Pseudo first order rate constants for
hydrolysis of DNPB catalyzed by random-
PS-DVB-Benz-OH 43 and ordered-PS-DVB-
Benz-OH 45 .. . 57

III-7. Effect of isocyanation of random-PS-DVB-
Benz-OH 43 and ordered-PS-DVB-Benz-OH 45
on the pseudo first order rate constant,
Robs. .................... 60

III-8. Pseudo first order rate constants, kobs,
for the hydrolysis of DNPB catalyzed by
PS-DVB-Benz determined in D20 and H20 63

IV-1. Elemental analysis of macroporous
polymers. . . .. 78


v(ii














LIST OF FIGURES

Figure Page

I-1. The active site of a-lytic protease 2

1-2. Polyethylenimine (PEI). . .. 3

1-3. Binary copolymers containing N-methyl-
hydroxamate . . 4

1-4. Plot of ka,obs versus aHA for the hydrolysis
of p-nitrophenyl acetate. . 5

1-5. Mechanisms for the hydrolysis of 4-nitro-
phenyl acetate by MHA-VIM . 6

1-6. Push-pull mechanism . 10

1-7. General base mechanism. . ... 11

1-8. Gelular resin . . 15

1-9. Macroreticulate gelular resin . 16

I-10. Spacer atom model compounds . 18

I-11. Possible bifunctional mechanisms. 19

II-1. Electron micrograph of the surface of
macroporous polystyrene-divinylbenzene
polymer 1000x magnification . 33

11-2. Electron micrograph of the surface of
macroporous polystyrene-divinylbenzene
polymer 30,000x magnification 34

11-3. Macroporous polystyrene-divinylbenzene
polymers synthesized. . .. 35

II-4. Cleavage of polymer bound spacer atom
monomer (PS-DVB-SAM) . 37

III-1. Nucleophilic catalysis of hydrolysis of
p-nitrophenyl acetate (PNPA) 45

viii








III-2. Effect on absorbance versus time curve for
formation of DNP by addition of PS-DVB-Benz 48

III-3. The effect of substrate concentration on
the rate of an enzyme catalyzed reaction. 50

III-4. Burst kinetics of the a-chymotrypsin
catalyzed hydrolysis of p-nitrophenyl
ester . . 51

III-5. Hydrolysis of DNPB catalyzed by PS-DVB-Benz
under burst conditions. . ... 53

III-6. Nucleophilic catalysis by PS-DVB-Benz of
DNPB hydrolysis . 54

III-7. Surface of microsphere covered by a layer
of polymer of lower crosslink density 56

III-8. (a) Hydrogen bond formed between benzimid-
azole and benzyl alcohol. ... 62
(b) Blocked alcohol by reaction with
isocyanate prevents hydrogen bond
formation . 62

TV-1. Reaction tube . . 82














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

SYNTHESIS AND KINETIC INVESTIGATIONS
OF MACROMOLECULAR CATALYSTS BASED
UPON THE SPACER ATOM CONCEPT

BY

ERIC JON LANGENMAYR

March 1980

Chairman: James A. Deyrup
Major Department: Chemistry

A new approach to the synthesis of macromolecular cata-

lysts by construction of "active sites" containing functional

groups in close proximity to each other into a highly cross-

linked polymer matrix was investigated. This was accomplished

by copolymerization with technical divinylbenzene (55%) of

acrylamide derivatives of benzimidazole (A) and benzyl alco-

hol (B) temporarily united via spacer (s) atoms (A-s-s-s-B).

Macroporous polymers were obtained through use of DMSO as

inert diluent in the copolymerization. Removal of the spacer

atoms by NaH/DMSO treatment resulted in "active sites" con-

taining polymer bound benzimidazole and benzyl alcohol.

Enhanced catalytic efficiency of the non-random polymer

(ordered-PS-DVB-Benz-OH) for the hydrolysis of 2, 4-dinitro-

phenyl benzoate was not observed as evidenced by comparison

with polymer which contained randomly introduced benzimidazole

x








and benzyl alcohol (random-PS-DVB-Benz-OH). Selective

blocking of the alcohol moiety in both polymers resulted

in no change in catalytic efficiency of ester hydrolysis by

the random polymer, but increased the catalytic efficiency

of the nonrandom polymer by approximately 50% which demon-

strated the proximity of the functional groups in the non-

random polymer. The polymer catalysts were found to follow

burst kinetics and no solvent isotope effect was observed

when the kinetics were run in deuterium oxide buffer. This

was interpreted to mean that polymer bound benzimidazole

acted as a nucleophilic catalyst in the hydrolysis of 2, 4-

dinitrophenyl benzoate. Analysis of the burst kinetics

indicated that only about 5% of the benzimidazole incorporated

into the polymers was kinetically accessible. This result

implied that a hydrophobic polymer matrix was not the most

suitable for construction of a polymer catalyst for ester

hydrolysis.














CHAPTER I
THE SPACER ATOM CONCEPT:
A NEW APPROACH TO THE SYNTHESIS OF
MACROMOLECULAR CATALYSTS



Introduction

Enzymes catalyze the chemical reactions responsible

for the biological functions of life. The high catalytic

efficiency and specificity enzymes exhibit have stimulated

research for efficient enzyme models. Since enzymes are

really functionalized macromolecules, it is not surprising

that many of the model systems are based on synthetic poly-
1-4
mers having pendant functional groups. Two properties of

enzymes are felt to be of critical importance to catalytic

activity and must be incorporated into the design of poly-

meric catalysts. These are binding of the substrate mole-

cule at the active site of the enzyme and cooperative inter-

actions of functional groups present at the active site of

the enzyme.

It has been shown that most enzymatic catalysis proceeds

according to Michaelis-Menten Kinetics. In equation (1),

kl, kcat
E + S ES --- E + P (1)
k-1

E stands for enzyme, S for substrate and P for product. The

Michaelis complex, ES, refers to the binding between substrate

1








molecules and active site by secondary valence forces.

Hence, the rate of product formation depends directly upon

the concentration of ES. Therefore, binding of substrate to

catalyst is important in enzymatic catalysis and this prop-

erty should be included in the design of efficient polymer

catalysts.

The notion of cooperative effects in enzymatic cataly-

sis results from the observation that present at the active

site of many enzymes are functional groups necessary for

catalysis to occur.6 These groups have a particular spatial

relationship with respect to one another and it is thought

that they may interact in a cooperative manner to catalyze

the reaction. For example, the active site of a-lytic pro-

tease (Figure I-1), an enzyme which catalyzes the hydrolysis

of amide bonds, contains a carboxylic acid, an imidazole and

an alcohol. Recent studies7 utilizing nitrogen-15 nmr indi-

cate that the imidazole is hydrogen-bonded to the acid and

the alcohol in the following manner:




CHCH
CH 2 2 2H2
\4 5
C/ O
0 / O *H-- : 0 H2
O 31 1
2


Figure I-1. The active site of a-lytic protease.








The authors suggest that this results in favoring of the

imidazole tautomer with hydrogen on N-3 rather than on N-l,

which allows the imidazole to act as a general base catalyst

toward the alcohol. Furthermore, the imidazolium cation

formed as a result of proton removal from the alcohol is

stabilized by the carboxylate anion. The stabilized imidaz-

olium cation may act as a strong acid catalyst in the hydrol-

y -is of the anide.

These two properties, substrate binding and cooperative

effects, have been successfully incorporated into synthetic
8
polymer catalysts. For example, Klotz et al. used modified

polyethylenimine (PEI) as synthetic polymer catalysts for the

hydrolysis of activated esters such as p-nitrophenyl acetate.

Unmodified PEI (Figure 1-2) is a branched polymer containing

about 50% secondary, 25% primary and 25% tertiary amino groups.




H H H H
H


NH2

Figure 1-2. Polyethylenimine. (PEI)


In this modification, the authors lauroylated approximately 10%

of the amino groups and imidazolylmethylated approximately 15%

of the amino groups. This modified PEI, carrying both imidaz-

ole nucleophiles and hydrophobic carbon chains to insure

efficient binding between polymer and substrate, was compared

to monomeric imidazole and Type-IIA sulfatase, a hydrolytic








enzyme, for the hydrolysis of 4-nitrocatechol sulfate. They

obtained the second order rate constants shown in Table I-1.




Table I-1. Second order rate constants for hydrolysis of
4-nitrocatechol sulfate.

catalyst k2 (M -sec )

Imidazole (20C, pH 9.2) 5.0 x 10-12
Type-IIA sulfatase (370C, pH 6.9) 0.24
Modified PEI (20c, pH 9.2) 18.0




Therefore, modified PEI catalyzed the hydrolysis at a rate

1012 times faster than monomeric imidazole and 102 times

faster than the enzyme. These results demonstrate the impor-

tance of combination of efficient substrate binding with

simple catalytic groups.

Polymeric catalysts have also been synthesized which

show cooperative effects between functional groups on the pol-
9
ymer backbone. Kunitake and Okahata studied the hydrolysis of

p-nitrophenyl acetate catalyzed by a binary copolymer con-

taining N-methylhydroxamate and vinylimidazole CMHA-VIM)

compared to a binary copolymer containing N-methylhydroxamate

and acrylamide (MHA-AAM) shown in Figure 1-3.


-(CH -CH)- -CH-CH)- --(CH-CH)2-(CH2-CH)-7
C 0 N C = 0 =0
H 3-N-H N NH2
H3C' XOH H3C H 2

MHA-VIM MHA-AAM

Figure 1-3. Binary copolymers containing N-methylhydroxamate.








Figure 1-4 gives plots of the apparent rate constant, ka,obs'
a, obs'
of acylation of hydroxamic acid unit (MHA) against atA, the

degree of dissociation of hydroxamic acid. The graphs show








SMHA-VIM
k
a,obs




MHA-AAM

HA

Figure 1-4. Plot of k a,obs versus aHA for the hydrolysis of
p-nitrophenyl acetate.


that the neutral form of MAH-AAM is catalytically inactive,

i.e., k a,obs equals zero when aHA equals zero and the plot

passes through the origin. However, for MHA-VIM, ka, obs is

positive when a equals zero, i.e., the polymer is still

catalytically active even in the neutral form. For MHA-VIM,

when the rates were measured using D20 as a solvent, an iso-

tope effect (kH/kD = 1.6) was found for the intercept, but

not for the slope. The observation of an isotope effect for

the intercept implies that catalysis by undissociated hydrox-

amic acid includes proton transfer in the transition state.

Absence of an isotope effect for the slope means that catalysis

by hydroxamate anion does not include proton transfer in the








transition state. Kunitake concluded that for MHA-VIM,

acylation involved nucleophilic attack of undissociated

hydroxamic acid assisted by neighboring imidazole acting as

a general base catalyst, a cooperative mechanism (a), in

addition to direct attack of hydroxamate anion (b) and

imidazole (c), shown in Figure 1-5.


H


R-C-O-R'


C= 0

N

H3C O

R-C-O-R'
II
o


\-


R--C-O-R'
II
0


Figure 1-5. Mechanisms for the hydrolysis of p-nitrophenyl
acetate by MHA-VIM.








Although much success has been obtained in synthesizing

polymeric catalysts as models for hydrolytic enzymes, it must

be emphasized that there is a fundamental difference between

the model and the enzyme. Polymeric catalysts contain large

numbers of functional groups in the polymer backbone, while

enzymes usually have one active site where the functional

groups which participate in the catalysis are located. Coop-

erativity in polymeric catalysts depends on two or more func-

tional groups coming close enough to each other thus forming

an active site. This in turn depends on the random confor-

mational changes of the polymer chain. Therefore, in this

type of catalyst, cooperativity is random in nature. The ac-

tive site in an enzyme is nonrandom; hence, cooperativity is

nonrandom in enzymes. Imanishi has stated that this random

flexed polymer approach completely disregards the fact that

catalysis in enzymes is not a result of random occurrence, but

occurs as a result of the information implanted within the

structure of the enzyme itself.

The very property essential for polymeric systems of

this nature to exhibit cooperative effects, conformational

flexibility, makes it very difficult to separate and evalu-

ate the relative importance of cooperative effects, binding

effects and microenvironmental effects. Overberger et al.10

studied the hydrolysis of p-nitrophenyl acetate catalyzed by

poly [4(5)-vinylimidazole] or monomeric imidazole as a func-

tion of the fraction, a,, of neutral imidazole. A plot of

the second order rate constant, <- versus the fraction of

neutral imidazole shows a linear dependence of the rate








constant kIm for monomeric imidazole. A similar plot for

poly [4(5)-vinylimidazole] gives an upward curvature that

shows that at al values below 0.8, imidazole is the better

catalyst while at al values above 0.8, the polymer is the

better catalyst. The authors attributed this catalytic

enhancement at high pH to a cooperative interaction between

neutral and anionic imidazole on the polymer chain. Later,

however, Overberger and Morimoto showed that poly [4(5)-

vinylimidazole] underwent considerable conformational change

as a function of the fraction of neutral imidazole present.

In this case, it is difficult to attribute catalytic enhance-

ment simply to cooperative effects in view of the large con-

formational changes also occurring. It is noteworthy that

Kunitake and Okahata suggested that the catalytic enhance-

ment in this case could be the result of a microenvironmental
12
effect. Lege has presented evidence which indicates that

such catalytic enhancements may be explained as a change in

pKa, and, hence, in nucleophilicity of the nucleophiles with

changes in the microenvironment.



Discussion

As stated previously, most polymeric enzyme models have

been based upon a flexible polymer chain containing pendant
1-4
functional groups. Intramolecular cooperativity in a sys-

tem of this type depends upon the random flexing of the poly-

mer chain. In direct contrast would be a model based upon a

rigid polymeric matrix. A rigid matrix would allow the







construction of "active sites" within the polymer. These
"active sites," in analogy to the active sites present in
enzymes, would consist of clusters of two or more functional
groups in close proximity to each other. It should be pos-
sible to construct these "active sites" in the following
manner. Consider two functional groups, A and B, to be
placed into the "active site." A and B may be joined to each
other by a molecular bridge comprised of spacer atoms denoted
by "s." The resultant spacer atom molecule 1, if vinyl



A s s s B
1

groups were attached to both ends 2, could be incorporated
into a polymer with use of divinylbenzene or related divinyl
monomers to insure a rigid matrix 3. Active sites in




0
75A-s-s-s-B ) A-s-s-s-B (2)


0
2 3

the matrix would result if the spacer atoms could be cleaved
out leaving behind the functional groups A and B 4 as shown.
The spacer atoms must be stable to the conditions of








polymerization. Yet, it is important that the cleavage reac-

tion removing the spacer atoms be performed under the mildest

possible conditions. In this way, the integrity of the poly-

meric matrix would not be destroyed and the spatial relation-

ship between A and B, defined by the nature and number of

spacer atoms, would tend to remain unchanged. A and B, now





3 A B" (3)
cleavage

4

held in close proximity to each other by the crosslinked

matrix 4,should be able to show cooperative effects.

Such cooperative effects could operate in two distinct

ways. One way would be the direct interaction of A and B

simultaneously on the substrate molecule. For example, with

A as a nuclecphile (NuH) and B an electrophile (E), catalysis

of carbonyl substrates could occur by a push-pull mechanism.










Figure 1-6. Push-pull mechanism.

Alternatively, A could interact with B as B interacts with

the substrate molecule. An example of this type of cooperative

effect would be general base catalysis. With A as a general








base (GB) and B a nucleophile (NuH), catalysis of carbonyl

substrates could occur.




GB: H Nu: C = O
1
Figure 1-7. General base mechanism.


Consideration of these two ways in which A and B may

interact leads to another interesting possibility which is

a direct result of the spacer atom concept. With a suitably

rigid polymer matrix, the average distance between functional

groups, A and B, can be changed by varying the number of

spacer atoms used. The spacer atom molecule would be designed

with exactly the number of spacer atoms corresponding to the

distance between the functional groups in a conceivable coop-

erative interaction. For example, in a general base catalyzed

reaction, the transition state for proton removal would in-

clude one atom between the general base and the nucleophile.

Therefore, a spacer atom molecule 5 for this case would in-

clude one spacer atom.



/---GB- s Nu--/

5


Another possibility, which is a direct consequence of

the spacer atom concept, is the incorporation of a chiral

carbon atom as a spacer atom. A chiral cavity of this nature

containing functional groups in a stereospecific arrangement








is strictly analogous to the active site of an enzyme. This

should lead to chiral recognition of substrate molecules by

the polymer and, therefore, substrate specific reactions.

Wulff et al. have constructed chiral cavities in

the matrix of highly crosslinked macroporous polymers and

have used these polymers in the resolution of racemates. This

was accomplished by copolymerization of a-D-mannopyranoside-

2,3;4,6-di-O-(4-vinylphenylboronate)6 containing the chiral

template molecule 7, with bifunctional crosslinking monomers

and a monofunctional comonomer in the presence of an inert







O --B' H


R = CH or \ 2--NO
6 7

diluent. Cleavage of the template molecule 7 left a chiral

cavity containing two boronic acid groups 8. Such


OH


OH




13


polymers, when used in chromatographic separations of D- and
L-4-nitrophenyl pyranoside, gave optical enrichments of up
to 87%.
Shea and Thompson have reported evidence that func-
tional groups introduced into a highly crosslinked polymer
matrix using the template method of Wulff et al.1315 can
retain stereochemical information. Introduction of diester
9 into a highly crosslinked matrix followed by cleavage leaves
sites containing two benzyl alcohol groups 10. Polymer



CO2CH2 HOCH20
S(1) polymerize
H (4)
Z2 (2) cleavage
C02CH2 D HOCH D


9 10

memory is demonstrated in the following manner. Prochiral
fumaryl chloride reacts with the benzyl alcohol sites 10 to
form polymer bound fumarate ester 11. Methylene transfer

df2H CH2S-N-Me2 CO 2CH2
%0 22 2-H (5)


CH CO2CH2
CO 2-CH2
o y< ^

co -cHrj )








using (dimethylamino)phenyloxosulfonium methylide gives

polymer bound 1,2-cyclopropanedicarboxylic acid ester 12.

Cleavage of the ester allows recovery of 1,2-cyclopropane-

dicarboxylic acid. When racemic diester 9 is used, racemic

1,2-cyclopropanedicarboxylic acid 13 is recovered. When




12 H CO2H (5)


CO2H H

13

diester 9 is synthesized using (-)-trans-l,2-cyclobutane-

dicarboxylic acid, recovery of the diacid 13 shows an enan-

tiomeric enrichment of 0.05%. The authors attribute the

slight enantiomeric enrichment to the existence of a chiral

environment at the benzyl alcohol site.



Design of Polymer Matrix

The spacer atom concept as an approach to the synthesis

of efficient polymer catalysts requires that a rigid poly-

meric matrix be used. A rigid matrix is necessary to insure

that upon removal of the spacer atoms, the functional groups

are less likely to move away from each other in an irrevers-

ible manner. This type of rigidity can be achieved through

incorporation of large amounts of crosslinking agent into the

polymer matrix,_Polymers of this:type are called gelular resins.








Gelular resins containing sulfonic acid groups or tetralkyl-

ammonium hydroxide groups are called ion exchange resins.

In solvents which swell the polymeric matrix, ion exchange

resins can catalyze reactions normally catalyzed:by mineral

acids and bases. The polymeric matrix of an ion exchange

resin is usually composed of long chains of polystyrene

cross-linked by divinylbenzene. This produces a three

dimensional, insoluble polymeric phase. These resins

are produced in the form of rigid beads. The three dimen-

sional polymeric gelular matrix is a homogeneous structure

with no discontinuities as shown in Figure 1-8. In order


Figure 1-8.


Gelular resin.


to catalyze a reaction, gelular ion exchange resins must be

exposed to a solvent which swells the resin. In the absence

of swelling, only the catalytic groups on the surface of the








bead are accessible. This type of polymeric matrix is poor-

ly suited for the incorporation of active sites into the

matrix using spacer atom molecules. The distance between

catalytic groups at the active site would be dependant on the

degree of swelling. Furthermore, at divinylbenzene contents

above 12%, the polymeric matrix greatly resists swelling at

all.18

However, during the 1960's, a type of resin closely

related to gelular resins was developed which is well suited

for use with the spacer atom concept.19,20 The new resins

consist of agglomerates of very small gelular resin beads

called microspheres as shown is Figure 1-9. The microspheres,


Figure 1-9.


Macroreticulate gelular resin.


fused together, make up a macrosphere of about the same size

as a normal gelular resin bead. The agglomeration of








microspheres may be likened to many small marbles stuck to

one another resulting in a macrosphere possessing a perma-

nent pore structure. These macrospheres are characterized

by high internal surface areas. The high internal surface

area insures accessibility of many catalytic sites even with

crosslinking density of greater than 12%. These resins,

referred to as macroporous or macroreticulate, make avail-

able a rigid polymeric matrix necessary for construction of

active sites by use of the spacer atom concept, with a high

surface area so that many catalytic sites are available.

It was decided to use a polystyrene-divinylbenzene

macroporous polymer matrix for the following two reasons.

Technical grade divinylbenzene, containing about 55% divinyl-

benzene and 40% ethyl styrene, was readily available. Use

of technical grade divinylbenzene would yield a polystyrene-

divinylbenzene matrix which should be approximately 55% cross-

linked and should give a rigid matrix. It should be noted,

however, that the best possible matrix might result from

either using a higher or lower divinylbenzene content. Fur-

thermore, a hydrophobic matrix might not be the--most suitable

for catalysis of ester hydrolysis.



Proposal of Research

The first step in the design of a polymeric catalyst

is consideration of the type of reaction to be catalyzed.

Once the reaction type has been decided upon, the catalytic

groups to be used in the polymer can be chosen. Most of the








macromolecular enzyme models developed have been designed

to catalyze the hydrolysis of activated esters such as

p-nitrophenyl acetate since the mechanism of ester hydrol-

ysis is well worked out and the reaction may be followed

spectrally. Therefore, it was decided to synthesize a poly-

meric catalyst designed to catalyze the hydrolysis of acti-

vated esters. Imidazoles have been widely used as catalytic

groups in polymeric catalysts. Imidazole may catalyze ester

hydrolysis both as a nucleophilic catalyst and a general

base catalyst. Thus, spacer atom model compounds containing

imidazoles or benzimidazoles are attractive. Alcohols, phe-

nols, thiols, hydroxamic acids, pyridines and carboxylic

acids have also been used as catalytic groups in polymeric
1-4
catalysts. Some of the model compounds considered are

shown in Figure I-10. Shown in Figure I-11 are some typical





N I0
s s C
c=C
I I I
s /> O O

N N CHN 2






(a) (b) (c) (d)
Figure 1-10. Spacer atom model compounds.








transition states which are possible with polymeric catalysts

derived from spacer atom monomers shown in Figure 1-10.


R'O- C -R


R'O--C R

0


R

ScOR'=O

[ OR'


Figure 1-11.


H? R



R-N OR'

C -o0


(c) (d)

Possible bifunctional mechanisms.








Once the number of spacer atoms has been decided upon,

the type of atom to be used as spacer'atoms must be chosen,

as well as their sequence in the spacer atom linkage. These

two factors determine both the stability of the spacer atom

monomer towards conditions of polymerization as well as ease

of cleavage of the spacer atom linkage. With these consid-

erations in mind, it was decided to synthesize model com-

pounds of the type shown in Figure 1-10. A mild method for

cleavage of the spacer atoms would then be found. The

divinylated analog of the model compounds would be synthe-

sized and incorporated into a highly crosslinked macroporous

resin by copolymerization with divinylbenzene in the presence

of an inert diluent. Subsequent extrusion of the spacer atoms

as shown in the introduction results in two functional groups

in close proximity to one another in the polymer matrix. The

catalytic activity of polymers made in this way would then

be compared to polymers containing the same functional groups

introduced randomly into the polymer.














CHAPTER II
SYNTHESIS AND CLEAVAGE OF
MODEL COMPOUNDS AND POLYMERS



Synthesis of Model Compounds

Synthesis of a wide variety of model compounds of the

type shown in Figure I-10 was undertaken. It must be kept

in mind that the model compounds ultimately would be con-

verted to their divinyl analogs. Therefore, synthetic routes

must be compatible with the introduction of vinyl groups.

The first model compound to be synthesized in good yield by

a mild route consisted of a benzimidazole connected by a

three spacer atom carbamate bridge to an aliphatic alcohol.

The synthetic route was based on a modification of an
21
a-amidoalkylation reaction (6). A suitably substituted

amide 14 undergoes nucleophilic displacement under acidic

conditions by aromatic or aliphatic compounds containing


0 0
I H+ ||
R-C--N-CH2X + Nu: R-C-N-CH2Nu (6)

R' R'

14 15
R = Alkyl, Aryl 0
R'= H, Alkyl It
X = Halide, -OR, -NR, -NHCR, -OH


active methylene or methine groups. Although the reaction

21








is normally run under strongly acidic or basic conditions,

it was discovered that use of the relatively strong nucleo-

phile benzimidazole would allow use of mild conditions.

Urethane 16 was converted to N-hydroxymethyl urethane 17 by

treatment with paraformaldehyde (H2CO)x under alkaline con-
22
editions according to Soignet et al. N-hydroxymethyl


CH CH20-- C--NH2


(H2CO) x

Ba (OH) 2
600C


0
II
CH3CH20O-C --NHCH20H (7)


urethane 17 condensed with benzimidazole 18 in refluxing

acetonitrile with a crystal of p-toluenesulfonic acid (TsOH)

to form the model compound 1-carbethoxyaminomethylbenzimid-

azole 19 in good yield. It was found that the model compound


H

18


CH3CN

TsOH

A CH 2NHCOCH2CH3

19


19 could also be synthesized directly from condensation of

urethane 16 and benzimidazole 18 with parafornLaldehyde.


H
H


0 (H2CO)x

+ H2N OCH2CH3 CH 3CN
TsOH
A








Attempts to condense benzimidazole 18 and phenyl ure-

thane 20 with paraformaldehyde under the normal reaction

conditions resulted in the formation of a white precipitate

on the inside wall of the condenser. The white precipitate

was identified as paraformaldehyde. No product was observed.


(H2CO)x
CH CN

TsOH

18 + H NC-O-( (10)
(H2CO)x
CH3CN N
H 030
TsOH N
A -CH2 0H
sealed tube

20 21


Attempts to force the reaction under sealed tube conditions

gave only 1-hydroxymethylbenzimidazole 21 23 and .unreacted

phenyl urethane 20. The failure of phenyl urethane 20 to

undergo the condensation was probably the result of decreased

basicity of the nitrogen due to an increased resonance inter-

action of the lone electron pair on nitrogen with the carbon-

yl. This increased resonance interaction is explained by the

delocalization of the lone pair electrons on oxygen into the

phenyl ring. Consistent with this explanation is the fact

that benzyl carbamate 22 condensed with benzimidazole 18 to

form 1-carbenzyloxyaminomethylbenzimidazole 23. A five








(H2CO) x

18 + H2NCOCH2 H > (11)
TsOH N

CH2NHCOCH2

22 23


spacer atom molecule containing two benzimidazole rings 25

was formed readily from condensation of two moles of benz-

imidazole 18 and one mole of urea 24 with two moles of

paraformaldehyde.

(H 2CO) x

2 (18) + H 2NCNH2 (12)
TsOH N

CH2NHCNHCH2


24 25



Cleavage of Model Compounds

Nucleophilic attack at the carbonyl group of urethanes
25
is sluggish. In agreement with this, gentle warming of

1-carbethoxyaminomethylbenzimidazole 19.with 18% aqueous

sulfuric acid showed no hydrolysis. Reflux for two hours

gave some cleavage as evidenced by observation by nuclear

magnetic resonance spectroscopy of the formation of a small

amount of ethanol. Prolonged reflux led to further cleavage

along with formation of black insoluble material. Aqueous

sodium hydroxide, when added to a solution of 1-carbethoxy-

aminomethylbenzimidazole 19 in dimethylsulfoxide (DMSO)








failed to effect cleavage. Surprisingly, it was discovered

that cleavage of the spacer atoms could be effected at room

temperature by subjection of model compound 19 to excess

sodium hydride (NaH) in DMSO. The cleavage reaction was

run in DMSO-d6 and followed by nuclear magnetic resonance

spectroscopy. The nuclear magnetic resonance spectrum of

1-carbethoxyaminomethylbenzamidazole 19 in CDC13 shows a
-- 3
triplet at 61.22 (3H), a quartet at 64.13 (2H), a doublet

at 65.48 (2H), an aromatic multiple at 67.10-7.83 (4H) and

a singlet at 62.10 (1H). Treatment with NaH/DMSO led to

immediate formation of benzimidazole from 1-carbethoxyamino-

methylbenzimidazole 19 as evidenced by complete disappearance

of the doublet at 65.48 corresponding to the two aminal pro--

tons and concommitant formation of symmetrical peaks in the

aromatic region due to benzimidazole anion. Addition of

authentic benzimidazole 18 to the mixture resulted in the

enhancement of the aromatic region without appearance of new

peaks. A new peak appeared as a broad singlet at 64.50. The

quartet due to the ethoxy methylene protons decreased in

intensity and a new quartet about 0.5 ppm upfield appeared.

Addition of absolute ethanol enhanced the upfield quartet

without appearance of additional peaks. The mixture, when

allowed to stand 24 hours, underwent no additional change.

Addition of a drop of D20 caused the downfield quartet to

completely disappear and ethanol quartet to increase in in-

tensity. The broad singlet at 64.5 also disappeared. Ure-

thane 16 yielded ethanol quantitatively when treated with








excess NaH in DMSO. A mechanism consistent with these ob-

servations is shown in Scheme I. Under the strongly basic


SCHEME I


I
c= o0

OCH2-CH3

64.1361.22


CH

C2 -



OCH2CH3


27
+

I! 2
N

C==0

OCH2CH3

28


CH20

+


-NH

C -

(OCH2CH3


NCO"
+

HOCH2CH3


conditions, the carbamate nitrogen is deprotonated. The

anion 26 thus formed ejects the benzimidazole anion 27 to

form the imine 28. Hydroxide present captures the imine 28.

Cleavage is completed with loss of formaldehyde and subse-

quent breakdown of carbamate anion to isocyanate and ethanol.

Urethane 16 has been shown to cleave with treatment of potas-

sium amide in ammonia to form potassium isocyanate and eth-

anol.26


67.10D
to
7.83


rOHO

CH 2

N

C==0
I
OC CH2CH3


OH

CH2
N-

C== 0

OCH2CH3








Under the same reaction conditions, 1, l'-ureidomethyl-

benzimidazole 25 cleaves easily to form benzimidazole and

urea. The cyclic carbamate 29 27was inert to the NaH/DMSO

treatment. The fact that a carbamate without a proton on

the nitrogen is inert to these conditions is further evidence




O

H N




29


for the base catalyzed cleavage mechanism shown in Scheme I.

The requirement for using spacer atoms which are easily

cleaved was demonstrated by efforts to cleave di(N-benzyl-
28
imidazolyl)methane 30. Cleavage was not effected even by




N
N

CH2

,-N




30


refluxing in strong acids. Treatment with NaH/DMSO for pro-

longed periods of time effected cleavage; however, it was

felt that the conditions for cleavage were too drastic for








di-(N-benzylimidazolyl)methane 30 to be useful as a spacer

atom monomer.


Synthesis of Spacer Atom Monomer (SAM)

It was decided to synthesize a divinylated monomer

based on the model compound 1-carbenzyloxyaminomethylbenz-

imidazole 23. The synthetic route began with the synthesis

of 5(6)-acrylamidobenzimidazole 34 and p-acrylamidobenzyl-

carbamate 37. The two monomers were joined via condensation

with paraformaldehyde. The choice of acrylamide derivatives

over vinyl derivatives was based on two factors. Use of the

acrylamide derivatives allowed direct observation by infrared

spectroscopy of incorporation of monomer into the polymer.

Furthermore, the synthesis of 5(6)-acrylamidobenzimidazole 34


SCHEME II


0
0 2N N Sn/HC H 2N N Cl
\> \ HC1 H20
ON A N

H H


31 32


0 0


S.-HCI H20 2 H

Ir 20% Na2CO3 H
H








as published by Kunitake and Shinkai29 has fewer steps,

resulting in higher overall yield of product than the syn-

thesis of 5(6)-vinylbenzimidazole published by Overberger

and Posiadly. Synthesis of 5(6)-acrylamidobenzimidazole
29
34 according to Kunitake and Shinkai is shown in Scheme II.

The synthetic route to p-acrylamidobenzyl carbamate is shown

in Scheme III.


SCHEME III
0
CH OH CH2OH CH20-8-NH2




NH2 N-H N-H

;07-


35 36 37


5(6)-acrylamidobenzimidazole 34 was found to be insol-

uble in water and nonpolar organic solvents, slightly soluble

in methanol and refluxing acetonitrile and readily soluble

in dimethylformamide (DMF) and dimethylsulfoxide (DMSO). Due

to this slight solubility in acetonitrile, the condensation

reaction of 5(6)-acrylamidobenzimidazole 34 with p-acryl-

amidobenzyl carbamate 37 was carried out at low concentrations.

This extended the time for complete disappearance of start-

ing material, as observed in the nuclear magnetic resonance

spectrum from 3 to 10 days. Yields of 30% to 40% of the

diacrylamide 38 were obtained. In addition, large amounts










+ 37


(13)


of insoluble material formed. In an attempt to obtain the

diacrylamide 38 in larger yield, 5(6)-nitrobenzimidazole 31

and p-nitrobenzyl carbamate 39 were condensed with paraform-

aldehyde to form the dinitro compound 40. This unusual com-

pound precipitates from the refluxing acetonitrile as it is


0
CH20OCNH2




NO2






39


(H2CO)
CH3CN

TsOH


02NN
2 N N

N


(14)


CH 2
G2
NH

C O


02N H2
40


(H2CO) N
CH3CN N>

TsOH H

f2
NH
I
C =0

0
I
I 2



N
-8o


38








formed. Fortunately, elemental analysis of the precipitate

was satisfactory. Unfortunately, this route to the diacryl-

amide 38 proved fruitless as attempts to reduce the dintiro

compound 40 by catalytic hydrogenation or dissolving metal

reductions led to cleavage of the carbamate linkage.

The diacrylamide 38 exists as two isomers since the

acrylamide moiety of the benzimidazole ring may be located at

the 5 or 6 position. However, both isomers should give essen-

tially the same active site in the polymer upon cleavage of

the spacer atoms since the benzimidazole ring can rotate

about the bonds attaching it to the polymer and has some free-

dom of movement.



Synthesis of Polymers

As previously mentioned, it was decided that the most

suitable polymeric matrix for our purposes was one having a

macroporous structure. Macroporous polymers have a continu-
18
ous permanent pore structure throughout the polymer. Such

polymers are made by suspension polymerization of the vinyl

and divinyl monomers in the presence of an inert diluent.3133

The pore structure and internal surface area depends on the

amount and type of diluent used. Diluents which are good

solvents for the monomers yield polymers which are charac-

terized by a small average pore diameter and a large inter-

nal surface area (50-500 m 2/g). Diluents which are poorer

solvents for the monomers yield polymers characterized by








larger pore diameters and smaller internal surface areas

(10-100 m2/g) 34

Due to the solubility properties of the diacrylamide 38,

it was not possible to do a suspension polymerization. How-

ever, the polymerization procedure used by Wulff et al.1315

in the synthesis of macroporous polymers containing chiral

cavities appeared to be a bulk polymerization with the mono-

mers diluted by acetonitrile. Since the diacrylamide 38 was

soluble in DMSO, it was decided to do a bulk polymerization

using DMSO as diluent.

Technical grade divinylbenzene (DVB) was used as the

source of monomers. Technical grade DVB consists of a mix-

ture of para- and meta- isomers of DVB (50-55%) and isomers

of ethylstyrene as a major contaminent (40%) along with some

minor contaminents.

Polymerizations were carried out under free radical

conditions using 1.0 ml DMSO as diluent per 1.0 g technical

DVB. The polymers obtained by this procedure were white

solids and had a chalky appearance. The chalky appearance

is characteristic of the porous nature of the polymer.31

Further evidence of the porous nature of the polymers re-

sulted when a small piece of polymer was placed in toluene.

This resulted in copious evolution of air bubbles as the
33
solvent displaced the air in the pores. The porous struc-

ture is shown clearly by the electron micrographs shown in

Figures II-1 and 11-2. From the electron micrographs, it






















































Figure II-1. Electron micrograph of the surface of
macroporous polystyrene-divinylbenzene
polymer lO00x magnification





















































Figure 11-2. Electron micrograph of the surface of
macroporous polystyrene-divinylbenzene
polymer 30,000x magnification









DMSO
AIBN


DVB + 34


6 hr at 80C
18-24 hr at 180C


DVB + 36


N N



H
PS-DVB-Benz 41


same conditions


PS-DVB-OH 42


DVB + 34 + 36


same conditions
7


random-PS-DVB-Benz-OH 43


DVB + 38


same conditions
_------------------------


I
C=0


-NH Q CH2

PS-DVB-SAM 44


Figure 11-3. Macroporous polystyrene-divinylbenzene polymers
synthesized.









can be seen that the polymer consists entirely of an agglom-

eration of spherical particles.

Analysis of totally insoluble polymers is problematic.

The two most direct approaches are elemental analysis and

infrared spectroscopy. Since the monomers used in this in-

vestigation contain an amide carbonyl, the carbonyl region

of the infrared spectrum is most informative. Examination

of the carbonyl region of the infrared spectrum of macro-

porous polystyrene-divinylbenzene (PS-DVB) prepared in the

absence of acrylamide monomer revealed a strong band due

to aromatic double bond stretch (ca. 1605 cm 1) and five

weaker bands due to aromatic overtones (ca. 1690 1950 cm- ).

When acrylamide monomer was present during polymerization,

incorporation could be directly observed by the presence

of a strong carbonyl band due to the amide (ca. 1695 cm- ).

The polymerizations shown in Figure 11-3 were carried out.



Cleavage of Polymer Bound
Spacer Atom Monomer

The infrared spectrum of uncleaved polymer bound spacer

atom monomer 44 shown in Figure 11-3 has a carbonyl band at
-1
1725 cm not completely resolved from the amide band at
-i
1695 cm Treatment with NaH/DMSO at room temperature for

12 hours resulted in cleavage of the carbamate group shown

in Figure 11-4 as evidenced by the disappearance in the in-

frared spectrum of the band at 1725 cm1. The nuclear

magnetic resonance spectrum of recovered DMSO showed no








peaks due to benzimidazole or benzyl alcohol. Therefore,

the amide bonds binding the catalytic groups to the polymer-

ic matrix were not cleaved by the NaH/DMSO treatment.


NaH/DMSO

RT


ordered-PS-DVB-Benz-OH

45


Figure 11-4.


Cleavage of polymer bound spacer atom monomer.


Acetanilide was used as a model compound for benzimidazole

bound to the polymer by an amide group. Treatment of acet-

anilide for 30 days with NaH/DMSO gave no cleavage.














CHAPTER III
KINETIC EVALUATION OF
POLYMER CATALYSTS



Introduction

Most of the polymeric catalysts designed as models of
1-4
hydrolytic enzymes have been homogeneous catalysts. Usu-

ally, the catalytic efficiency has been determined by the

ability of the polymer to catalyze the hydrolysis of acti-

vated esters such as p-nitrophenyl acetate. These homoge-

neous reactions may be followed easily by spectral observa-

tion of product.

However, as previously stated, the spacer atom concept

requires the use of a rigid polymeric matrix. Highly cross-

linked macroporous resins are the best suited for this pur-

pose, but they are insoluble. Hence, these polymeric cata-

lysts must be studied in a heterogeneous system. Two prob-

lems arise with the use of heterogeneous catalysts. The most

obvious problem is the potential difficulty of measuring

reaction rates of heterogeneous systems. The more subtle

problem is the effect of the polymeric matrix on a bound

functional group. That is, can it be assumed that a func-

tional group which catalyzes a homogeneous reaction will

still catalyze the reaction when bound to a polymer? Manas-

sen has shown that polybenzoquinones catalyze oxidative

38








dehydrogenation heterogeneously in the same manner as benzo-

quinones catalyze the reaction homogeneously. Manassen ob-

tained similar results for other organic catalysts and con-

cluded that, in the field of organic catalysis, the solid

state properties of the catalyst have little gross effect

on the bound functional group. Furthermore, the chemistry

of acidic or basic groups on ion exchange resins exhibits

the same chemical properties as free mineral acids and bases
18
in homogeneous catalysis. Therefore, it should be possi-

ble to chose functional groups to act as catalytic sites on

a polymeric matrix based upon the catalytic properties of

the functional group exhibited in a homogeneous system.

This is not to say that there are no advantages to binding

catalytic groups to an insoluble polymeric matrix. Such

systems may allow for enhanced catalysis through substrate

binding to the polymer and the ability to construct active

sites in the polymer matrix through use of the spacer atom

concept.

Many difficulties- exist in the measurement of reaction

rates in heterogenous systems employing a solid catalyst
37
which are absent in homogeneous systems. The kinetics of

heterogeneous systems are less reproducible than those of

a homogeneous system since it is difficult to reproducibly

construct a solid catalyst. Also, there is no simple way of

observing the concentration of reactant in the interfacial

region where the catalysis occurs. In order for reactant

molecules to come in contact with functional Qrcu's -on the







polymer matrix, they must diffuse through a boundary layer

separating the bulk solution from the solid phase. After

reaction has occurred, the product must then diffuse back

across the boundary layer to the bulk solution. To insure

that neither of the diffusion processes is rate determining,

heterogeneous systems are often stirred. An additional prob-

lem is that highly crosslinked macroporous polymers are very

brittle and mechanical stirring of a suspension of polymer

particles in a solvent leads to particle fragmentation which

precludes use of stirring to insure that the reaction is not

diffusion controlled.



Preliminary Experiments

The primary purpose of this investigation was to syn-

thesize polymeric catalysts which might show cooperative ef-

fects. In order to kinetically evaluate these polymers, it

was necessary to find a reliable technique for the measure-

ment of reaction rates. As previously mentioned, the poly-

meric catalysts synthesized in this investigation were de-

signed to catalyze the hydrolysis of activated esters. In

the course of choosing which specific ester to use, it was

discovered that by using a hydrophobic ester, such as

p-nitrophenyl caproate (PNPC), the observed pseudo first order

rate constant, kobs, was the same whether the system was

stirred or not. Results for the hydrolysis of PNPC catalyzed

by PS-DVB-Benz 41 are shown in Table III-1. These results








indicated that even in the absence of stirring, the rate

process is not a diffusion controlled one.




Table III-1. Effect of stirring on the pseudo first order
rate constant, kobs, for the hydrolysis of
PNPC catalyzed by PS-DVB-Benz.a/


catalyst

PS-DVB-Benz
PS-DVB-Benz


wt. catalyst

100 mg
100 mg


stir

yes
no


kobs* 104

3.8.3
4.2.3


a/pH 8.7, 0.02M TRIS Buffer; 25C; 9.0% (v/v) CH3CN




Evidence that the hydrolysis of 2,4-dinitrophenyl

benzoate (DNPB) catalyzed by PS-DVB-Benz in the absence of

stirring is not diffusion controlled is shown in Table III-2.




Table III-2. Effect of the amount of catalyst on the
pseudo first order rate constant, Kobs'
for hydrolysis of DNPB catalyzed by
PS-DVB-Benz.a/


catalyst

PS-DVB-Benz
PS-DVB-Benz
PS-DVB-Benz


wt. catalyst

25.0 mg
5.0 mg
5.5 mg


k o 103
obs

6. 8.4
2.7.2
2. 6.2


/pH 7.0; 0.02M Phosphate Buffer; 250c; 9.0% (v/v) CH3CN







If the catalyzed reaction was diffusion controlled, then

there would be no dependence of the pseudo first order rate

constant, kobs, on the amount of catalyst used. Table III-2

shows that k obs increases as the amount of catalyst increases.

It should be noted here that the reaction mixture was agi-

tated when a spectral measurement was taken. Therefore,

although continuous stirring was not used, the reaction mix-

ture did receive periodic agitation. However, even such in-

frequent agitation seemed unnecessary since measurements ob-

tained even an hour apart still gave good rate data.

The fact that diffusion was not rate-determining even

without continuous stirring seemed remarkable. It was de-

cided to try to determine how long it took for a hydrophobic

ester, introduced into a reaction mixture comprised of sol-

vent and PS-DVB, to be absorbed onto the polymer. To this

end, the following experiment was run. To three test tubes

was added 25 mg of PS-DVB. No polymer was added to the

fourth test tube. To all four test tubes was added the iden-

tical volume of the water-acetonitrile buffer used for the

kinetic runs. A 15 pl solution of DNPB was added to the

first test tube, and the test tube was vigorously swirled

for 15 seconds. The polymer was then filtered by suction

and the filtrate treated with aqueous potassium hydroxide.

The same procedure was done to test tubes two and three ex-

cept that swirling was done for 30 and 180 seconds, respec-

tively. The substrate was also added to test tube four con-

taining no polymer, and the solution was likewise filtered by








suction. The results, summarized in Table III-3, are star-

tling. When filtrate from test tube four was treated with

base, an intense yellow color due to dinitrophenylate

anion immediately appeared. The filtrate from the three

test tubes containing polymer was visually unchanged when

treated with base. Measurement of the absorbance and the

use of a Beers Law plot established that 99% of the ester

was adsorbed on the surface of the polymer within 15 sec-

onds.




Table III-3. Determination of degree of adsorption of
DNPB on PS-DVB.a/


test tube # polymer time(sec) Absorbance % Absorbed
on Polymer

1 yes 15 0.078 99
2 yes 30 0.056 > 99
3 yes 180 0.053 > 99
4 no 0.565 -

/pH 7.0, 0.02M Phosphate Buffer; 25C; 9.0% (v/v) CH3CN




Therefore, the hydrophobic ester, DNPB, when introduced into

the mixture composed of a polar solvent and the hydrophobic

polymer, rapidly and quantitatively adsorbs onto the surface

of the polymer. This behavior eliminates the need for stir-

ring to insure good contact of the reactants with the poly-

mer. It is more difficult to determine even qualitatively

how fast the product desorbs from the polymer. However, the








affinity of the product for the polymer can be qualitative-

ly determined. To determine the affinity of the product,

dinitrophenylate anion, for the polymer under the conditions

at which the kinetics were run, dinitrophenol (DNP) was

added to a mixture of PS-DVB and the water-acetonitrile buf-

fer. Periodically, the dinitrophenylate anion concentration

was determined spectrally. After a period of several days,

the dinitrophenylate anion concentration in solution reached

a limiting value of about 95% of the original concentration.

Therefore, the reaction product showed little affinity for

the polymer. However, this experiment was run under equi-

librium conditions and, while it is true that the product

had little affinity for the polymer surface, the experiment

did not reveal the rate Qf desorption of product from the

polymeric surface as it is formed by chemical reaction.

The use of PS-DVB in the two preceding experiments as

a model for the polymeric matrix containing benzimidazole

and/or benzyl alcohol may be criticized. However, these

functional groups comprise no more than about 2 1/2 mole

percent of the polymer as a whole. Therefore, it is not un-

reasonable to expect that the results obtained using PS-DVB

are qualitatively correct for the functionalized polymers.

Furthermore, the high affinity of substrate for polymer and

low affinity of product for polymer is consistent with the

kinetic behavior of the functionalized polymers. In view

of the low affinity of the product for the polymer, it was







decided to determine if added dinitrophenyl(DNP) would lead

to product inhibition.

Imidazoles, including benzimidazoles, catalyze the

hydrolysis of activated esters, such as p-nitrophenylacetate

(PNPA), by nucleophilic catalysis. The mechanism involves

initial attack of the imidazole on the ester to form an

acyl-imidazole as shown in Figure III-1. The acyl-imidazole


N 0 0 H 0
\ + CH3 -0 C CH-3 2 CH3 C-OH
CN \=j

H 0+ +

NO2

2H -
0 H-\N

NO2

Figure III-1. Nucleophilic catalysis of hydrolysis of
p-nitrophenyl acetate(PNPA).

intermediate can react with water to form product or may

react with p-nitrophenolate anion to reform the reactant
38
molecule. The addition of p-nitrophenol (PNP) to the

system did result in a decrease in the observed reaction

rate and is considered proof for the existence of the

acyl-imidazole intermediate.

If attachment of benzimidazole to an insoluble polymer

matrix does not change the mode by which it catalyzes the

hydrolysis of activated esters, then it may be expected that








such catalysis occurs through a nucleophilic mechanism form-

ing an acylated benzimidazole on the surface of the polymer.

Experimental evidence that the catalysis by PS-DVB is nu-

cleophilic is given later. In the catalysis of the hydrol-

ysis of DNPB by PS-DVB-Benz, product inhibition could be

observed if the dinitrophenolate anion is present in suffi-

cient quantity for a long enough period of time at the poly-

mer surface. The effect of added dinitrophenol (DNP) on the

rate of hydrolysis of DNPB catalyzed by PS-DVB-Benz is

summarized in Table III-4.




Table III-4. Effect of added DNP on the observed pseudo
first order rate constant, k obs.a/


catalyst [DNP] M*108 k obs* 103

PS-DVB-Benz none 6.8.4
PS-DVB-Benz 5.0 6.9.3
PS-DVB-Benz 15.0 6.8.6

/pH 7.0; 0.02M Phosphate Buffer; 25C; 9.0% (v/v) CH3CN





The absence of product inhibition in the presence of

added DNP can be explained in two ways. The added DNP,

which at the pH of the buffer is present as the anion, does

not adsorb on the surface of the polymer at all. This would

result in no observable product inhibition since the added

DNP cannot approach acylated PS-DVB-Benz. Another explanation








is that a small amount of added DNP does adsorb to the poly-

mer surface, but the rate constant for desorption is much

larger than the rate constant for attack of dinitrophenylate

anion on the acylated benzimidazole. This results in no

observable product inhibition.

As previously shown in Table III-3, the hydrophobic

ester, DNPB, adsorbs quickly and quantitatively on PS-DVB.

However, the experiment did not show whether the adsorption

process was reversible or irreversible. An answer to this

question was obtained from the following experiment. Macro-

porous PS-DVB containing no functional groups was exposed

to the water-acetonitrile buffer in the usual manner. Next,

an excess of DNPB was added to the reaction mixture with

vigorous shaking. The reaction was followed spectrally

for a period of time by observation of formation of dinitro-

phenylate anion. To the reaction mixture was added an equal

weight of PS-DVB-Benz. Again, the reaction rate was moni-

tored. It can be seen from Figure III-2 that, with only

PS-DVB present, the reaction is slow as evidenced by the

first portion of the absorbance versus time plot. However,

at the time of addition of PS-DVB-Benz, a break in the line

occurs with the absorbance increasing much faster with time.

This means that the ester hydrolysis is occurring at a much

faster rate. Therefore, DNPB molecules originally adsorbed

on PS-DVB must become adsorbed on PS-DVB-Benz. This can

only occur if the DNPB molecules adsorbed on the hydrophobic








surface of PS-DVB are in equilibrium with bulk solvent. The

ester molecules may then be adsorbed onto the surface of

PS-DVB-Benz where the reaction is catalyzed.









absorbance







addition of PS-DVB-Benz


time

Figure III-2. Effect on absorbance versus time curve for
formation of DNP by addition of PS-DVB-Benz.


The effect of a change in pH on the observed pseudo

first order rate constant for the hydrolysis of DNPB in the

presence of PS-DVB-Benz and PS-DVB is shown in Table III-5.

As shown, in the presence of PS-DVB, an increase in the pH

from 6.9 to 8.5 led to an increase in the rate of ester hy-

drolysis by a factor of 30. In the absence of functional

groups on the polymer, this strong pH dependence must reflect

the fact that the major pathway is hydrolysis by hydroxide

ion. However, ester hydrolysis in the presence of PS-DVB-

Benz increases only by a factor of 1.7 when the pH is in-

creased from 6.9 to 8.5. This much smaller dependency of








rate constant on pH reflects the fact that ester hydrolysis

in this system is largely due to the catalytic action of the

benzimidazole functional groups bound to the polymer.




Table III-5. Effect of pH on the observed pseudo first
order rate constant, kobs' for the hydrolysis
of DNPB in the presence of PS-DVB-Benz or
PS-DVB.


catalyst pH kobs* 103

PS-DVB-Benz 7.0/ 6.80.4
PS-DVB-Benz 8. 6b/ 12.00.4
PS-DVB 7.0a/ 0.05E
PS-DVB 8.6- 1.50.2

a/0.02M Phosphate Buffer; 250C; 9.0% (v/v) CH3CN

b/0.02M TRIS Buffer; 25C; 9.0% (v/v) CH3CN

- The rate of this initially slow reaction decreased with
time. The value given is an upper limit based on the
initial rate.








Burst Kinetics

Essentially all enzymes are characterized by a satura-

tion phenomenon. That is, at low concentrations of the en-

zyme substrate, the velocity of the reaction is linearly

dependent on the concentration of the enzyme substrate while,

under conditions of a large excess of substrate, the reaction

velocity becomes independent of the substrate concentration.








This behavior is depicted in Figure III-3. The situation is

defined by equation (1) as shown in Chapter I and requires

that the enzyme bind the substrate before the reaction occurs.












velocity


[substrate]


Figure III-3.


The effect of substrate concentration on the
rate of an enzyme catalyzed reaction.


A limiting rate is reached at high enough substrate concen-

tration such that all the binding sites of the enzyme are

occupied by substrate.

The use of p-nitrophenyl acetate (PNPA) as substrate

for chymotrypsin led to further elucidation of the mecha-

nism.39,40 When the reaction was run under conditions of

a large excess of PNPA, the reaction was found to be biphas-

ic. There was an initial rapid release (burst) of p-nitro-

phenol (PNP) followed by a slower release of PNP as shown








in Figure III-4. Burst kinetics implies two rate steps in

the reaction, hence the existence of an intermediate. The












Intercept=R
absorbance /


Difference=Re-bt


time


Figure III-4.


Burst kinetics of the a-chymotrypsin catalyzed
hydrolysis of p-nitrophenyl ester.


situation is described by equation (15). The intermediate,

ES', was shown to be acetyl-chymotrypsin. The kinetics are



E + S E S ) ES' + P ) E + P2 (15)



explained by initial rapid acylation of the enzyme liberat-

ing a burst of PNP followed by a slower release of PNP which

is dependent on the rate of deacylation of the acetyl-enzyme








intermediate. Further information can be obtained from Fig-

ure III-4. Determination of the y-intercept, R, gives a_

direct measure of the concentration of the enzyme. From the

slope, Q, and the constant, b, the rates of acylation and

deacylation may be obtained.

Since observation of burst kinetics is evidence that

an intermediate exists along the reaction path, it was de-

cided to determine whether the polymeric catalysts synthe-

sized in this investigation followed burst kinetics. Obser-

vation of burst kinetics would rule out a mechanism whereby

PS-DVB-Benz acts as a general base catalyst on water to

directly catalyze ester hydrolysis without formation of an

intermediate. Figure III-5 shows a plot of absorbance versus

time for hydrolysis of DNPB by PS-DVB-Benz under burst con-

ditions. The plot shows the biphasic nature of the reaction.

Therefore, in the hydrolysis of DNPB catalyzed by PS-DVB-Benz,

an intermediate is formed along-the reaction path. Since

benzimidazole is the only functional group on the polymer,

it can be safely assumed that the intermediate is benzoylated

benzimidazole. Therefore, the hydrolysis is occurring

through nucleophilic catalysis by PS-DVB-Benz as shown in

Figure III-6. Random-PS-DVB-Benz-OH 43 and ordered-PS-DVB-

Benz-OH 45 give similar burst kinetics and, therefore, must

also catalyze the reaction by a nucleophilic mechanism.

As previously mentioned, determination of the inter-

cept, R, in Figure III-4 gives directly the concentration of














0




O
4
-4)


.0





m N

o


Do N

















n


o o
>1
C m








44






CD- 0 00 w

















Zo
\ oQ.
\ ( U














N N
H I I
C = O H




NO2 +
O-
NO2 0


C
NO2 gOH



Figure III-6. Nucleophilic catalysis by PS-DVB-Benz of
DNPB hydrolysis.


enzyme. Actually, it is the concentration of active sites

which is determined. For enzymes having one active site per

enzyme molecule, the concentration of active sites is equal

to the enzyme concentration. Similar analysis of Figure

III-5 should yield the concentration of kinetically active

polymer bound benzimidazole. The value of R in absorbance

units was converted to concentration units through use of

a Beers Law plot. Comparison of this concentration, which

is the concentration of kinetically active benzimidazole, to

the total amount of benzimidazole incorporated into the poly-

mer, as determined from elemental analysis, shows that only

about 5% of the benzimidazole in the polymer is kinetically








active. Similar results for ordered-PS-DVB-Benz-OH 45 (4%)

and random-PS-DVB-Benz-OH 43 (6%) were obtained.

These figures seemed surprisingly small and, initially,

seemed suspect. It had been thought that during the synthe-

sis of PS-DVB-Benz, the initially formed polymer chains

would be poorer in 5(6)-acrylamidobenzimidazole 34 while

the chains formed later would be richer in 5(6)-acrylamido-

benzimidazole 34. This would result in the presence of a

high concentration of benzimidazole at the polymer surface.

Treatment of PS-DVB-Benz with isocyanates led to polymers

whose infrared spectra showed appearance of carbonyl bands

nearly equal in intensity to the carbonyl band derived from

the amide linkage of benzimidazole to the polymer. This

indicated, at least qualitatively, that much more than 5%

of the benzimidazole was accessible. However, the burst

kinetics were carried out in a 37.5% (v/v) aqueous solution

of acetonitrile at 250C while the isocyanation reactions

were carried out in the non-polar solvent toluene at 50-600C.

Therefore, both results may be at least qualitatively cor-

rect, but the concentration of accessible benzimidazole on

the hydrophobic surface of the polymer is smaller when the

polymer is in contact with a polar medium than when the poly-

mer is in contact with a non-polar medium. This is an impor-

tant observation as it leads to the conclusion that, at least

for polymeric catalysts containing neutral catalytic groups,

the combination of a hydrophobic polymer in contact with an








aqueous medium is not the best system in which to evaluate

catalytic efficiency. These conclusions suggest that a

more efficient system would use the combination of a hydro-

philic polymer in contact with an aqueous medium, which

should result in a much higher concentration of kinetically

accessible active sites.
41
There is experimental evidence that the surface of

the highly crosslinked macroporous polymer is covered with

a layer of polymer of low crosslink density as shown in

Figure II1-7. This type of surface could result from the

copolymerization behavior of DVB with styrene. At low con-

versions, the forming polymer will be richer in DVB than the

monomer mixture. Toward the end of the polymerization, the

monomer mixture will be richer in styrene. The accessibil-

ity of functional groups present on the surface of the










S *



S

a




Figure III-7. Surface of microsphere covered by a layer of
polymer of lower crosslink density.








microspheres would then depend on whether or not a good

swelling solvent was used. A solvent such as toluene would

swell the less crosslinked layer and all functional groups

on the surface of the microsphere would be accessible. A

solvent such as the acetonitrile-water buffer used in the

kinetics would not swell the less crosslinked portion, thus

blocking the functional groups on the surface.



Comoarision of Catalytic Efficiencies of
Random-PS-DVB-Benz-OH 43
and
Ordered-PS-DVB-Benz-OH 45
for Hydrolysis of DNPB

Results of hydrolysis of DNPB catalyzed by random-PS-

DVB-Benz-OH 43 and ordered-PS-DVB-Benz-OH 45 are summarized

in Table III-6. It can be seen from the table that, in both

cases, the random-PS-DVB-Benz-OH 43 catalyzed the reaction

about 1.8 times faster than ordered-PS-DVB-Benz-OH 45 did.




Table III-6. Pseudo first order rate constants for hydrol-
ysis of DNPB catalyzed by random-PS-DVB-Benz-
OH 43 and ordered-PS-DVB-Benz-OH 45.


catalyst kobs. 103 pH

random-PS-DVB-Benz-OH 10.0.2 8.4
ordered-PS-DVB-Benz-OH 5.5.4 8.4
random-PS-DVB-Benz-OH 3.9.3 6.9
ordered-PS-DVB-Benz-OH 2.2.2 6.9
PS-DVB-Benz 2.6.1 6.9








In the absence of accurate knowledge of the surface concen-

tration of functional groups in the two polymers, it cannot

be determined whether this difference is a result of the

random versus ordered nature of the two polymers. Further-

more, since both polymers exhibit similar kinetics under

burst conditions, a cooperative mechanism for ordered-PS-

DVB-Benz-OH 45 is ruled out.

As a control, random-PS-DVB-Benz-OH was treated with

NaH/DMSO for 2 days at room temperature. The polymer was

recovered in the usual manner. The observed pseudo first

order rate constant obtained for the hydrolysis of DNPB was
3 .-1
9.4,10 min which is in good agreement with the pseudo

first order rate constants obtained with random-PS-DVB-Benz-

OH not previously treated with NaH/DMSO. Furthermore, the

infrared spectrum of NaH/DMSO treated random-PS-DVB-Benz-OH

was essentially the same as the infrared spectrum of un-

treated random-PS-DVB-Benz-OH.

In the absence of an observed cooperative effect, it

was decided to find evidence that, upon cleavage of the spac-

er atoms from PS-DVB-SAM 44, the benzimidazole and benzyl

alcohol remain close enough to interact with one another.

Subtraction spectra utilizing Fourier transform infrared

spectroscopy to observe differences in hydrogen bonding be-

tween random-PS-DVB-Benz-OH and ordered-PS-DVB-Benz-OH were

not informative. Attempts to reconnect the benzimidazole

and benzyl alcohol on ordered-PS-DVB-Benz-OH with several

bifunctional electrophiles were unsuccessful. A kinetic





SCHEME IV


43 or 45


selective
cleavage


0
0


43 or 45


c




60



difference was observed by selectively blocking the benzyl

alcohol by reaction with a-napthyl isocyanate as shown in

Scheme IV. This was done by isocyanating both benzimidazole

and benzyl alcohol functional groups and then selectively

removing the isocyanate group from benzimidazole. The ki-

netic results for the modified polymers are shown in Table

III-7. These results show that there is no effect of iso-

cyanating the benzyl alcohol in random-PS-DVB-Benz-OH. For

ordered-PS-DVB-Benz-OH, there was a rate increase of over

50%. This is surprising as the original intent of isocya-

nating the benzyl alcohol was to sterically hinder the




Table III-7. Effect of isocyanation of random-PS-DVB-Benz-
OH 43 and ordered-PS-DVB-Benz-OH 45 on the
pseudo first order rate constant, kobs.


random-PS-DVB-Benz-OH

isocyanate k *b10 # runs

none 4.1.3 3
a-napthyl- 3.9+ .,3 2
t-butyl- 3.7a/.l 1

ordered-PS-DVB-Benz-OH

isocyanate kobs *10 # runs

none 2.1i.2 2
a-napthyl- 3.2.2 2
t-butyl- 3.0.l 1
control 2.2.l 2

-/The t-butyl carbamyl group was cleaved more slowly than
the a-napthyl carbamyl group. Therefore, the slightly
smaller kobs could be the result of incomplete cleavage.








benzimidazole of the ordered polymer, thus causing a rate

decrease. To determine whether or not the method used to

isocyanate the polymers and then to cleave the group was

somehow activating ordered-PS-DVB-Benz-OH, a control was

run by subjecting the ordered polymer to conditions of iso-

cyanation without the isocyanate present. The polymer was

then exposed to the exact conditions of cleavage. The re-

sultant polymer showed no rate enhancement. To determine

whether the effect was the result of the naphthyl ring,

t-butyl isocyanate was used to selectively isocyanate the

benzyl alcohol. Once again, isocyanation has no effect on

the catalytic efficiency of random-PS-DVB-Benz-OH, but

caused a rate increase in the hydrolysis catalyzed by ordered-

PS-DVB-Benz-OH of nearly 50%.

It is difficult to explain this peculiar rate effect.

One possible explanation is that in ordered-PS-DVB-Benz-OH,

the benzyl alcohol is strongly hydrogen bonded to the lone

electron pair of the pyridine type nitrogen of the benzimid-

azole as shown in Figure III-8. It might be expected that

the polar hydroxyl group would be tightly associated with

the polar benzimidazole in the non-polar:environment of-the

polymer. The strong hydrogen bond would tend to decrease

the nucleophilicity of the benzimidazole, therefore reducing

the catalytic efficiency of the polymer. Isocyanation of

the benzyl alcohol blocks the hydroxyl group and eliminates

the hydrogen bond. The non-hydrogen bonded benzimidazole








would then be a stronger nucleophile and catalytic efficien-

cy would be enhanced.










)N N

HNH




N











(a) (b)

Figure III-8. (a) Hydrogen bond formed between benzimidazole
and benzyl alcohol.
(b) Blocked alcohol by reaction with isocyanate
prevents hydrogen bond formation.






Kinetics in Deuterium Oxide

To rule out a general base mechanism for the hydrol-

ysis of DNPB catalyzed by PS-DVB-Benz, the rates were









determined in D 20. Deuterium oxide is slightly denser than

H20 and the polymer aggregated near the surface. This made

it difficult to wash .the polymer off the walls of the reac-

tion-tube. Much of the polymer was frequently exposed to air.

The pseudo first order rate constants in a D20 buffer as

reported in Table III-8 are not as reproducible as rate

constants determined in H20. However, it can be seen that

there seems to be no significant difference between the two

solvents; hence, general base catalysis does not occur.


Table III-8.





polymer

PS-DVB-Benz
PS-DVB-Benz
PS-DVB-Benz
PS-DVB-Benz
PS-DVB-Benz
PS-DVB-Benz


Pseudo first order rate constants, kobs, for
the hydrolysis of DNPB catalyzed by PS-DVB-
Benz determined in D20 and H20 "a


DO
2

yes
yes
yes
yes
no
no
no


k -103
obs

4.1.3
3.6.2
4.0.2
3.0+.1
4.1.2
3. 8.2


/PS-DVB-Benz used in this study contained about 1 mole per-
cent benzimidazole.


Discussion

The kinetic results of selectively isocyanated random-

PS-DVB-Benz-OH 43 and ordered-PS-DVB-Benz-OH 45 show that








a rigid polymer matrix containing catalytic sites comprised

of functional groups in close proximity to one another can

be constructed by use of the spacer atom concept. For the

specific case of the relative catalytic efficiency of ran-

dom-PS-DVB-Benz-OH 43 and ordered-PS-DVB-Benz-OH 45, the

results were disappointing. Polymer 43 actually catalyzed

the hydrolysis of DNPB about 1.8 times faster than polymer

45. In the absence of accurate knowledge of the concentra-

tion of kinetically accessible functional groups, it is not

possible to attribute this difference to an effect resulting

from the random versus non-random nature of the polymers.

However, accessibility of functional groups under various

conditions seems similar for the two polymers as evidenced

by burst kinetics and chemical modification of the polymers

with isocyanates. Therefore, it may be concluded that no

significant cooperative effects are present.

This is not an unexpected result since imidazole

and benzimidazole catalyze the hydrolysis of esters with

good leaving groups by a nucleophilic mechanism, and show

no cooperative effects for these esters.38 However, small

model compounds combining imidazole with hydroxamic acids

or thiols do show cooperative effects in the catalysis of

activated ester hydrolysis.19'42 Therefore, construction of

active sites containing imidazole or benzimidazole with

hydroxamic acids or thiols should lead to cooperative effects

with activated esters.








Preliminary kinetic experiments showed that a kinetic

system comprised of a polar liquid phase in contact with a

hydrophobic polymer catalyst for the hydrolysis of a hydro-

phobic ester had both advantages and disadvantages. The

combination of a high affinity of ester molecule for the

polymeric catalyst with a low affinity of product for the

polymer resulted in no complications due to diffusion. In

addition, the reaction was easy to follow since the product

desorbed quickly into the liquid phase where it was measured

spectrally. However, the results obtained by burst kinetics

indicated that under these conditions only a small propor-

tion of benzimidazole was kinetically accessible. It is

likely that this is the result of the hydrophobic nature of

the polymer. A better catalyst with more kinetically ac-

cessible benzimidazole might be obtained by use of a hydro-

philic polymer. Favorable binding could be maintained in

such a system by the incorporation of hydrophobic groups at

the active site.















CHAPTER IV
EXPERIMENTAL



Introduction

Melting points are recorded in degrees Centigrade and

are uncorrected. Melting points were determined with a

Thomas-Hoover Unimelt capillary melting point apparatus.

Infrared spectra were recorded on a Beckman IR10 or a Perkin-

Elmer 283B infrared spectrophotometer. Nuclear magnetic

resonance spectra were recorded on a Varian Model A60-A

spectrometer for 60 MHz proton spectra and on a JEOL FT 100

for 100 MHz proton spectra. All chemical shifts (6) are

reported in parts per million (ppm) downfield from tetra-

methylsilane as internal reference. Low resolution mass

spectra, exact mass and molecular weight data were measured

on an AEI-MS-30 double beam spectrometer. Microanalyses

were carried out by Atlantic Microlab, Inc., Atlanta, Geor-

gia. Surface area measurements were carried out by Micro-

meritics Instrument Corp., Norcross, Georgia. Solvent evap-

oration was performed at reduced pressure on a Bichi

Rotavapor-R rotary evaporator equipped with either a water

aspirator or mechanical vacuum pump. UV-visible spectro-

photometric measurements were made on either a Beckman DB-G,

a Cary 15 or a Cary 17D spectrophotometer. The pH

66







measurements were made with a Beckman Research pH meter

equipped with a Radiometer GK 2321C electrode.



Syntheses


N-Hydroxymethyl Urethane (17)

N-hydroxymethyl urethane 17 was prepared from ethyl

carbamate and paraformaldehyde in the presence of barium
22
hydroxide. Pure product was obtained after one recrystal-

lization from benzene: mp 49.5-51, lit mp 530; nmr (CDC1 3)

6 1.25 (triplet, 3H), 4.17 (quartet, 2H), 4.73 (broad dou-

blet, 3H), 6.37 (broad hump, 1H).


1-Carbethoxyaminomethylbenzimidazole (19)

A. Benzimidazole (0.100 g, 0.85 mmol), N-hydroxymethyl

urethane (0.101 g, 0.85 mmol) and 2 ml reagent grade aceto-

nitrile were placed into a 25 ml round-bottomed flask fitted

with magnetic stir bar, condenser and drying tube. One crys-

tal p-toluenesulfonic acid was added and the solution was

brought to reflux. After overnight reflux, the acetonitrile

was removed under reduced pressure, leaving a white solid.

Two recrystallizations from benzene gave white crystals:

mp 139-1400; ir (nujol) 1727 cm (C=O); nmr (CDC13) 6 1.22

(triplet, 3H), 4.13 (quartet, 2H). 5.48 doublett, 2H), 6.53

(broad triplet, 1H), 7.10-7.83 multiplee, 4H), 7.92 (sing-

let, 1H);

Anal. Calcd for C11H13 N 302: C, 60.26; H, 5.97; N, 19.17.

Found: C, 60.18; H, 5.97; N, 19.15.








B. Benzimidazole (1.00 g, 8.5 mmol), urethane (0.75

g, 8.5 mmol), paraformaldehyde (0.25 g, 8.5 mmol), 10 ml

of reagent grade acetonitrile and a few crystals of p-tolu-

enesulfonic acid were placed into a 25 ml round-bottomed

flask fitted with magnetic stir bar, condenser and drying

tube. The solution was refluxed three days, cooled and the

acetonitrile removed under reduced pressure. The white,

solid residue (1.87 g) was taken up in chloroform, washed

once with 20% aqueous sodium carbonate and dried over anhy-

drous magnesium sulfate. The drying agent was filtered and

the chloroform removed under reduced pressure yielding 1.64 g

solid residue. Recrystallization from benzene gave white

crystals (1.12 g, 60%): mp 139-140.


Phenyl Carbamate (20)

Pheny] carbamate was prepared by the slow addition of

phenyl chloroformate with stirring to 5 volumes of cold
24
concentrated ammonium hydroxide: mp 146-148, lit mp 149-

1520.


Benzyl Carbamate (22)

Benzyl carbamate was prepared by the slow addition of

benzyl chloroformate with stirring to 5 volumes of cold con-
24
centrated ammonium hydroxide: mp 84-86*, lit mp 87.


1-Carbenzyloxyaminomethylbenzimidazole (23)

Benzimidazole (0.10 g, 0.85 mmol), benzyl carbamate

(0.15 g, 0.85 mmol), paraformaldehyde (0.025 g, 0.85 mmol),






2 ml of reagent grade acetonitrile and 1 large crystal of

p-toluenesulfonic acid were stirred and heated at 550 for

3 days in a 25 ml round-bottomed flask fitted with condenser.

Upon cooling, the solution yielded white crystals which were

collected by suction filtration and allowed to air dry: mp

158-1590; ir (nujol) 1730 cm-1 (C=O); nmr (CDC13) 6 5.15

(singlet, 2H, benzylic), 5.55 doublett, 2H, aminal), 6.25

(broad humpr, 1H, NH), 7.22-7.92 multiplee, 9H, aromatic),

8.00 (singlet, 1H, amidine); mass spectrum m/e 281 (P+).

Anal. Calcd for C 6H15 N 302: C, 68.31; H, 5.36; N, 14.94.

Found: C, 68.30; H, 5.37; N, 14.93.


1,1'-Ureidomethylbenzimidazole (25)

Benzimidazole (0.200 g, 1.7 mmol), urea (0.051 g, 0.85

mmol), paraformaldehyde (0.051 g, 1.7 mmol) and a crystal of

p-toluenesulfonic acid as catalyst were placed into a 25 ml

round-bottomed flask fitted with magnetic stir bar, condenser

and drying tube. Two milliliters of reagent grade acetoni-

trile was added and the mixture heated to 50 after which the

reactants dissolved. The solution was heated overnight at

50. A white precipitate formed from the hot solution and

was filtered by suction (0.270 g, 98%): mp 195 (sublimes);

ir (nujol) 1680 cm-1 (C=0); nmr (DMSO-d6) 6 5.56 doublett,

4H), 7.10-7.83 multiplee, 10H), 8.20 (singlet, 2H).


NaH/DMSO Cleavage of 1-Carbethoxyaminomethylbenzimidazole (19)

In a small test tube, NaH (0.080 g, 57% oil dispersion)

was added to 1.0 ml reagent grade DMSO. The test tube was








swirled to effect thorough mixing and sealed with parafilm.

After 8 to 10 hours, the NaH/DMSO was added to an nmr tube

containing 1-carbethoxyaminomethylbenzimidazole 19 (0.050 g)

and the nmr spectrum obtained immediately. The nmr spectra

showed immediate quantitative formation of benzimidazole.

Formation of ethanol was complete after about 6 hours. When

the reaction was run in DMSO-d6, the NaH was added to the

nmr tube containing 1-carbethoxyaminomethylbenzimidazole dis-

solved in DMSO-d6. The nmr tube was capped and was shaken

vigorously. The cap was removed-to release H2 gas. Addition

of ethanol and authentic benzimidazole after cleavage was

complete resulted in the appearance of no new peaks in the

nmr spectrum and enhanced the peaks attributed to benzimid-

azole anion, 6 6.67-7.5 (aromatic, 4H), 7.68 (singlet, 1H),

and ethanol, 3.47 (quartet, 2H), 1.07 (triplet, 3H).


NaH/DMSO Cleavage of Urethane (16)

In a dry box under a N2 atmosphere, 16 _(0.040 g)

was dissolved in DMSO-d6 in an nmr tube. A small portion of

NaH (57% oil dispersion) was added. The nmr spectra were ob-

tained and revealed the presence of two quartets of about

equal intensity. Further addition of NaH resulted in an

increase in intensity of the upfield quartet and concomitant

decrease in intensity of the downfield quartet. Further

addition of NaH led to the complete disappearance of the

downfield quartet. Addition of absolute ethanol resulted

in the appearance of no new peaks and increased the intensity








of the upfield quartet. Complete cleavage of urethane was

also effected by NaH in reagent grade DMSO.


NaH/DMSO Cleavage of l,l'-Ureidomethylbenzimidazole (25)

The bis-benzimidazole (0.030 g) was dissolved in DMSO-

d6 and the nmr spectrum obtained. A small amount of NaH

(57% oil dispersion) was added to the nmr tube; the tube was

capped and was shaken vigorously. The cap was removed to

release H2 gas. The nmr spectrum obtained within 5 minutes

of addition of NaH showed only benzimidazole with no start-

ing material left.


Attempted Cleavage of 3-t-Butyl-5-Phenyl-4-Oxazolin-2-One
(29) with NaH/DMSO

Compound 29, generously supplied by H. Gingrich, was

dissolved in DMSO-d6 and the nmr spectrum recorded. A small

portion of NaH (57% oil dispersion) was added and the nmr

tube capped and shaken. The nmr spectrum, recorded after 1

hour, showed no change. Additional NaH was added. After

10 days, the nmr spectrum was obtained. The spectrum showed

unchanged starting material.


Di-(N-Benzimidazolyl)Methane (30)

Benzimidazole (2.3 g, 19 mmol) was dissolved in a solu-

tion of 1.4 g 80% potassium hydroxide in 14 ml of ethanol.

Dibromomethane (1.7 g, 9.8 mmol) was added dropwise with

stirring. The solution was then refluxed for 4 hours. The

precipitate (KBr) was filtered by suction and the filtrate

was evaporated to dryness under reduced pressure. The








residue was washed with water. One recrystallization from

ethanol/water gave colorless crystals: mp 248-2490, lit mp28

252.


NaH/DMSO Cleavage of Di-(N-Benzimidazolyl)Methane (30)

Di-(N-benzimidazolyl)methane was dissolved in DMSO-d6

and the nmr spectrum was obtained. A small portion of NaH

(57% oil dispersion) was added. The nmr spectrum showed no

change. The reaction was followed a few days by monitoring

with nmr spectroscopy and showed no change. After a period

of 6 weeks, an. nmr spectrum indicated approximately 50%

cleavage occurred to form benzimidazole.


5(6)-Aminobenzimidazole Hydrochloride Monohydrate (32)

5(6)-aminobenzimidazole-HC1-H20 was synthesized by the

reduction of 5(6)-nitrobenzimidazole with tin and hydrochlo-
29
ric acid: mp 103-107, lit mp 108-109. The free base

was obtained by treatment with aqueous sodium carbonate:

nmr (DMSO-d6) 6 6.6 doublett of doublets, 1H), 6.8 doublett,

1H), 7.3 doublett, 1H), 7.9 (singlet, 1H).


5 (6)-Acrylamidobenzimidazole (34)

The benzimidazole 34 was prepared from acryoyl chloride

and 5(6)-aminobenzimidazole hydrochloride monohydrate follow-
29
ing the procedure described by Kunitake and Shinkai, with

the following modifications: sufficient sodium dithionite

was added to the reaction mixture to discharge the red color;

the reaction product was isolated as the free base by _-







dissolution of the hydrochloride salt in water followed by

neutralization of the aqueous solution with 20% sodium

carbonate. This resulted in the precipitation of the free

base as a white solid (65%): mp 217-217.5; ir (nujol)
-i
1670 cm-1 (C=0); nmr (DMSO-d6) 6 5.71 doublett of doublets,

1H), 6.38-6.58 multiplee, 2H), 7.30-7.68 (aromatic, 2H),

8.15 (singlet, 1H), 8.27 (broad singlet, 1H), 11.85 (broad

hump, 1H).

Anal. Calcd for C10H9H30: C, 64.16; H, 4.85; N, 22.45.

Found: C, 64.00; H, 4.90; N, 22.35.


p-Acrylamidobenzyl Alcohol (36)

Acryolyl chloride (0.73 g, 8.0 mmol) was added drop-

wise with stirring to p-aminobenzyl alcohol (2.00 g, 16.0

mmol) dissolved in 40 ml tetrahydrofuran at 0. After the

addition was complete, the reaction mixture was allowed to

warm to room temperature. The precipitate was filtered by

suction. Treatment with activated charcoal in boiling water

followed by hot filtration over a celite pad gave a colorless

solution which formed short white crystals (1.5 g, 51%) upon
-i
cooling: mp 106.5-108; ir (nujol) 1670 cm- (C=0); nmr

(DMSO-d6) 6 4.55 doublett, 2H), 5.67 doublett of doublets, IH),

6.40 multiplee, 2H) 7.10-7.80 (quartet, 4H).


p-Acrylamidobenzyl Carbamate (37)

p-Acrylamidobenzyl alcohol (0.80 g, 4.5 mmol) in 5 ml

of reagent grade tetrahydrofuran was added dropwise to a








stirred solution of carbonyl diimidazole (0.80 g, 4.9 mmol)

in 15 ml tetrahydrofuran. After addition, the reaction was

stirred for 30 minutes. The solution was cooled in an ice

bath for 15 minutes after which 3 ml of concentrated ammo-

nium hydroxide was added dropwise with stirring. After the

addition was complete, the ice bath was removed and the re-

action was stirred for 60 minutes. The tetrahydrofuran was

removed in vacuo to leave a light yellow residue which was

washed with water and dried (0.92 g, 93%). Recrystallization

from boiling water gave white crystals: mp 153-1560; ir

(nujol) 1710 cm-1 (C=O), 1665 cm-1 (C=O), nmr (DMSO-d6) 5

4.93 (singlet, 2H), 5.68 doublett of doublets, 1H), 6.38

multiplee, 2H), 7.18-7.72 (quartet, 4H), 10.0 (broad sing-

let, 1H).

Anal. Calcd for C GHI2H203: C, 60.00; H, 5.49; N, 12.72.

Found: C, 59.90; H, 5.53; N, 12.68.


1-Carb-(4'-Acrylamido)Benzyloxyaminomethyl-5(6)-Acrylamido-
benzimidazole (38)

P-Acrylamidobenzyl carbamate (0.50 g, 23.0 mmol), 5(6)-

acrylamidobenzimidazole(0.43 g, 23.0 mmol) and a crystal of

p-toluenesulfonic acid were dissolved in 150 ml of reagent

grade acetonitrile at 600 and then the temperature was raised

tQ 80. Excess paraformaldehyde was added periodically.

The reaction was monitored for product formation as evidenced

by the appearance of a broad doublet (ca. 6 5.4) in the nmr

spectrum. During the reaction, a precipitate formed which

was insoluble in most solvents. The reaction was complete








after 10 days. The hot solution was decanted from the

precipitate and the solvent was evaporated under reduced

pressure to leave 0.48 g of an off white solid. The solid

was dissolved in methanol. The methanol solution was added

in small portions to 0.5 g silica gel and the methanol re-

moved under reduced pressure. The silica gel was added dry

to a column of silica gel (8.0 g). The column was eluted

first with 200 ml methylene chloride. The product was

eluted with methanol. Removal of methanol under reduced

pressure left 0.36 g (38%) of the desired product as a white
-i
solid: mp >200 (decomposes); ir (nujol) 1670 cm- (C=0),

1720 cm-1 (C=O); nmr (DMSO-d6) 6 5.03 (singlet, 2H), 5.45

(broad doublet, collapses to singlet with D20, 2H), 5.85

doublett of doublets, 2H), 6.30-6.57 multiplee, 4H), 7.20-

7.83 (quartet, 7H), 8.17-8.27 (broad doublet, 2H), 10.2

(broad hump, 2H).

Anal. Calcd for C22H21N504 H20: C, 60.40; H, 5.30; N, 16.01.

Found: C, 60.49; H, 5.33; N, 16.00.


p-Nitrobenzyl Carbamate (39)

p-Nitrobenzyl alcohol (0.500 g, 3.27 mmol) in 10 ml

tetrahydrofuran was added dropwise to 4 stirred solution of

carbonyl diimidazole (0.583 g, 3.60 mmol) in 20 ml tetra-

hydrofuran. After the addition was complete, the solution

was cooled in an ice water bath. To the cooled, stirred

solution was added dropwise 3 ml concentrated ammonium

hydroxide. After addition of the ammonium hydroxide, the








ice bath was removed and the solution was stirred for 45

minutes. The solvent was evaporated in vacuo and the resi-

due taken up in methylene chloride, washed three times with

water and dried over anhydrous magnesium sulfate. The dry-

ing agent was filtered and the methylene chloride evaporated

in vacuo leaving a yellow solid (0.61 g, 94%). Recrystal-

lization from boiling water followed by drying in vacuo

gave an analytical sample: mp 154-155 ; ir (nujol) 1690

cm-1 (C=O); nmr (DMSO-d6) 6 5.18 (singlet, 2H), 6.56 (broad

hump, 2H), 7.53-8.30 multiplee, 4H).

Anal. Calcd for C8H8N204: C, 48.98; H, 4.11; N, 14.28.

Found: C, 48.92; H, 4.11; N, 14.28.


1-Carb-(4'-Nitro)Benzyloxyaminomethyl-5(6)-Acrylamido-
benzimidazole (40)

p-Nitrobenzyl carbamate (0.300 g, 1.53 mmol) and 5(6)-

nitrobenzimidazole (0.250 g, 1.53 mmol) were dissolved in

20 ml of hot acetonitrile and the resultant solution was fil-

tered hot through a sintered glass funnel into a 50 ml pear

shaped flask. The funnel was then rinsed with 2 additional

portions of hot acetonitrile. Paraformaldehyde (0.046 g,

1.53 mmol) and a large crystal of p-toluenesulfonic acid were

added to the reaction flask. The reaction was stirred and

heated at 65 for 2 hours, whereupon a white precipitate was

observed. The temperature was raised to 850 and the reaction

was refluxed overnight. After cooling to room temperature,

the reaction mixture was filtered by suction and the white








precipitate washed several times with cold acetonitrile

and, finally, cold ethanol. Drying overnight in vacuo

yielded 0.44 g (77%): mp >300 ; ir (nujol) 1730 cm-1 (C=O);

nmr (DMSO-d6, 100 ) 6 5.20 (singlet, 2H), 5.64 and 5.70

(2 doublets, 2H), 7.50-8.74 multiplee, 8H).

Anal. Calcd for CI6 13 506: C, 51.75; H, 3.53; N, 18.86.

Found: C, 51.69; H, 3.57; N, 18.81


General Synthetic Method for Macroporous Polymer Catalysts

The following technique was used to synthesize all

polymers used in this study. Technical divinylbenzene was

washed 3 times with equal volumes of 5% sodium hydroxide

solution, followed by washing with copious amounts of dis-

tilled water until the washings were neutral. The divinyl-

benzene was dried over anhydrous calcium chloride. The

appropriate monomers and AIBN were dissolved in 1.0 ml of

DMSO in a small test tube and 1.0 g of dried divinylbenzene

was added. The test tubes were sealed with parafilm and

placed in an oil bath at 800 for 6 hours, then at 1100 for

18-24 hours. The chalky white polymers were removed from

the test tubes, broken up with a mortar and pestle and

washed exhaustively with acetonitrile. The acetonitrile

washings were removed under reduced pressure leaving the

DMSO behind. The DMSO was checked for unreacted monomer by

nuclear magnetic resonance spectroscopy. The polymerization

procedure given above resulted in little or no unreacted

monomer. The polymers were dried at 80 in vacuo in a








drying pistol. Analysis for incorporation of functional

groups into the polymers was by infrared spectroscopy and

elemental analysis.




Table IV-1. Elemental analysis of macroporous polymers.


polymer 34 36 C H -N
(mg) (mg) calcd found calcd found calcd found


PS-DVB 91.19 91.31 8.46 8.52 0.33 0.26
PS-DVB-OH 78 89.74 89.74 8.34 8.40 0.79 0.71
PS-DVB-OH 19 90.77 90.94 8.43 8.52 0.50 0.41
PS-DVB-OH 9.5 90.92 90.82 8.44 8.49 0.45 0.40
PS-DVB-Benz 40 90.37 90.21 8.36 8.35 0.99 0.95
random-PS-DVB-
Benz-OH 20 19 90.39 90.28 8.68 8.70 0.80 0.75
random-PS-DVB-
Benz-OH 40 38 89.78 86.55 8.31 8.38 1.17 1.15
PS-DVB-SAM 90.33 88.16 8.37 8.29 0.84 1.19







NaH/DMSO Cleavage of Spacer Atoms from PS-DVB-SAM (44)

Sodium hydride (NaH, 0.50 g, 57% oil dispersion) was

added to 3 ml of reagent grade DMSO and left overnight. The

resultant NaH/DMSO mixture was centrifuged and the super-

natant was pipetted into a test tube containing 0.33 g of

PS-DVB-SAM 44 and a magnetic stir bar. The mixture was

stirred for 1 minute. After 2 days, the polymer was washed

extensively with aqueous DMSO until the washings were neu-

tral and then with reagent grade acetonitrile. The polymer

was dried under vaccum at 80. An infrared spectra (nujol)







-1
showed loss of carbonyl band at 1720 cm- The carbonyl
-i
band at 1695 cm remained.


Attempted Cleavage of Acetanilide with NaH/DMSO

Acetanilide (0.060 g) was dissolved in DMSO-d6 and

the solution transferred to an nmr tube. The nmr spectrum

was recorded. A small amount of NaH was added. The nmr

spectrum showed no change. Additional NaH was added. After

17 days, the nmr spectrum was essentially unchanged. The

DMSO-d6 was poured over ice and made slightly acidic. A

white precipitate formed. The precipitate was collected by

suction filtration and dried under reduced pressure (0.045 g,

75%). An nmr spectrum showed the white precipitate to be

acetanilide.


Treatment of Random-PS-DVB-Benz-OH (43) with NaH/DMSO

Sodium hydride (NaH, 0.050 g, 57% oil dispersion) was

added to 0.5 ml DMSO in a small test tube. The test tube

was sealed with parafilm and swirled vigorously. The NaH/

DMSO was left overnight, then centrifuged. The supernatant

was added to random-PS-DVB-Benz-OH 43 (0.030 g). After 2

days, the polymer was recovered by suction filtration and

washed with aqueous DMSO until the washings were neutral.

The polymer was then washed 5 times with reagent grade aceto-

nitrile. The polymer was dried under vacuum at 800. An

infrared spectrum (nujol) showed no loss in intensity of the

carbonyl band at 1695 cm-
carbonyl band at 1695 cm.








Development of Method for the Selective Isocyanation of
Benzyl Alcohol in Random-PS-DVB-Benz-OH (43) and
Ordered-PS-DVB-Benz-OH (45)

PS-DVB-Benz 41 and PS-DVB-OH 42 were treated with

excess a-napthylisocyanate in toluene at 50 for 10 hours

with triethylamine as catalyst. The polymers were collected

by suction filtration and washed 5 times with toluene, 5

times with DMSO and 5 times with acetonitrile. The polymers

were dried in vacuo at 800. Infrared spectra (nujol) showed

a new carbonyl band at 1730 cm- due to the carbamate formed.

Treatment of the polymers with benzyl amine in the presence

of 4-dimethylaminopyridine in DMSO as solvent at 800 for

8 hours resulted in cleavage of the carbamate bond in PS-

DVB-Benz as evidenced by the disappearance of the carbonyl
-1
absorbanace at 1730 cm- An infrared spectra of the car-

bonyl absorption at 1730 cm-1 was still present, although

there seemed to be a slight decrease in the intensity of

the absorption.


General Method of Reaction of Polymers with Isocyanates

The polymers (0.020 g) were placed in a small test

tube and sufficient toluene was added to completely cover

the polymers. One drop of a solution of 1 volume of tri-

ethylamine in 3 volumes of toluene was added. The isocyan-

ate (0.020 g) was then added and the test tubes were tight-

ly stoppered and placed in an oil bath at 600 overnight.

The polymers were collected by suction filtration and

washed successively with toluene, DMSO and acetonitrile.








The washed polymers were placed in a drying pistol and dried

in vacuo at 800. An infrared spectrum (nujol) showed the

appearance of a new absorbance in the carbonyl region (ca.

1730-1740 cm-1).


General Method of Selective Cleavage of Isocyanate from
Random-PS-DVB-Benz-OH (43) and Ordered-PS-DVB-Benz-
OH (45)

The isocyanated polymers were treated with benzyl

amine in the presence of 4-dimethylaminopyridine in DMSO as

solvent at 800 for 8 hours. The polymers were collected

by suction filtration and washed successively with aqueous

alkaline DMSO, aqueous DMSO until the washings were neutral,

DMSO and acetonitrile. The polymers were dried at 800 in

vacuo in a drying pistol. The infrared spectrum (nujol)

showed the presence of the carbamate carbonyl band at 1730-
-i
1740 cm but with reduced intensity.


2, 4-Dinitrophenyl Benzoate (DNPB)

Dinitrophenol (4.0 g, 22 mmol) was measured into a

100 ml round-bottomed flask equipped with a magnetic stir

bar and drying tube. Pyridine (25 ml) was added to the

flask and benzoyl chloride (3.7 g, 26 mmol) was then added

dropwise with stirring. A yellow precipitate formed. The

contents of the flask were poured over ice water and the

resultant mixture stirred. The mixture was filtered by

suction and the solid triturated 4 times with absolute

ethanol. One recrystallization from t-butyl alcohol gave








slightly yellow-tinted crystals (2.0 g, 32%); mp 129-131

lit mp 128-129.


Polymer Surface Areas

Surface areas for two of the polymers used in this

study were determined by a modified, single point B.E.T.

method using nitrogen as the adsorbate. PS-DVB had a spec-

ific surface area of 333.3 m 2/g and random-PS-DVB-Benz-OH

had a specific surface area of 374.3 m2/g.



Kinetics


General Kinetic Method

Kinetics were performed using polymer fractions which

passed through a No. 40 U.S. Standard Sieve, but were re-

tained by a No. 80 U.S. Standard Sieve. The amount of poly-

mer catalyst (normally 25 mg except in the case of isocya-

nated polymer when 5 mg quantities were used) was weighed

out on an analytical balance and then placed in the bottom of

a reaction tube, as shown in Figure IV-1, fitted with a side

arm and ASTM medium glass frit. Dried acetonitrile (0.25 ml)

was pipetted through the side arm into the bottom of the reac-

tion tube. The reaction tube was then placed in a thermostat-

ted bath at 2500.30 for 30 minutes. The ester solution was

glass frit ---side arm


TOP BOTTOM


Figure IV-1. Reaction tube.








added to the reaction tube through the side arm using a

Drummond Microtrol syringe equipped with a teflon tip. Im-

mediately upon addition of the ester solution, the side arm

of the reaction tube was tightly stoppered with a rubber

stopper and shaken vigorously for 30 seconds. The time of

addition of ester solution was recorded and the reaction

tube replaced in the thermostatted bath. To take absorbance

measurements, the reaction tubes were removed from the ther-

mostatted bath and shaken vigorously. The reaction tube

was then inverted with the funnel end over a 4 ml cuvette.

A positive pressure of N2 was applied to the side arm which

forced the water-acetonitrile buffer-into the cuvette. After

an absorbance measurement, the water-acetonitrile buffer was

returned to the reaction tube. Care was taken to wash any

polymer which adhered to the walls of the reaction tube.

The formation of p-nitrophenolate anion was observed at

400 nm. Infinity absorbances were obtained by periodically

measuring the absorbances until two consecutive measurements

were the same. The observed rate constants were obtained

both by a least squares program and by the slope of the best

straight line through a first order plot of the data. Al-

most always, the two methods gave identical results. Occa-

sionally, the least squares result was significantly larger

than the result obtained graphically. In these cases, the

graphical method was used. All reactions were followed to

at least 50% completion. Almost all kinetic runs were made

twice, the exception occurring when only a small amount of








polymer catalyst was available. Burst kinetics were done in

the same manner except that the buffer used consisted of 1.5

ml of acetonitrile and 2.5 ml aqueous buffer. The larger pro-

portion of acetonitrile used was necessary due to the low sol-

ubility of DNPB in the aqueous buffer. A 40 fold excess of

substrate over catalyst was used in the burst kinetics.


Error Analysis

First order plots for the appearance of product with

time were constructed for each kinetic run. The residuals

were determined using the observed data points and the calcu-

lated points of the least squares line. From the residuals,

the standard error was calculated and the error reported as

95% confidence limits using the appropriate value from the

students T table.


Preparation of Ester Solutions Used in Kinetics

A solution of p-nitrophenyl caproate (8.2.10-3M) in

dried acetonitrile was generously supplied by C. Lege. To

a 10 ml volumetric flask was added 2, 4-dinitrophenyl ben-

zoate (0.020 g, 6.94-10-5 mole). The flask was filled to

the mark with dried acetonitrile. The acetonitrile was dried

by refluxing 3 hours over P205 followed by distillation.


Preparation of Buffer Solutions Used in Kinetics

The phosphate buffers were prepared from two stock buf-

fer solutions (0.02 m) made with glass distilled water using

primary standard sodium hydrogen phosphate and dipotassium

phosphate. These solutions were combined to form solutions




85



of the correct pH. Deuterium oxide buffers were prepared

using deuterium oxide of 99.7 atom %D minimum isotopic puri-

ty supplied by Merck and Co., Inc. TRIS buffer was prepared

with primary standard TRIS and was generously supplied by

C. Lege. Certified ACS potassium chloride was used to make

the buffers 0.1 M ionic strength.













REFERENCES


1. Y. Imanishi, J. Polym. Sci., Macromol. Rev., 14, 1
(1979).

2. C. G. Overberger and J. C. Salamone, Acc. Chem. Res., 2,
217 (1969).

3. T. Kunitake and Y. Okahata in "Advances in Polymer Sci-
ence," Vol. 20, H.-J. Cantow, Ed. (Springer-Verlag,
Berlin, 1976), pp. 159-221.

4. T. Shimidzu in "Advances in Polymer Science,"
Vol. 23, H.-J. Cantow, Ed. (Springer-Verlag, Berlin,
1977), pp 55-102.

5. L. Michaelis and M. L. Menten, Biochem. Z., 49, 333
(1913).

6. K. J. Laidler and P. S. Bunting in "The Chemical
Kinetics of Enzyme Action," 2nd ed. (Clarendon Press,
Oxford, 1973), Chap.:! ....

7. J. D. Roberts and W. W. Bachovchin, J. Amer. Chem. Soc.,
100, 2041 (1978).

8. I. M. Klotz, H. C. Kiefer, W. I. Congdon and I. S.
Scarpa, Proc. Natl. Acad. Sci. USA., 69, 2155 (1972).

9. T. Kunitake and Y. Okahata, J. Amer. Chem. Soc., 98, 7793
(1976).

10. C. G. Overberger, T. St. Pierre, N. Vorchheimer, J. Lee
and Y. Yaroslavsky, J. Amer. Chem. Soc., 87, 296 (1965).

11. C. G. Overberger and M. J. Morimoto, J. Amer. Chem. Soc.,
93, 3222 (1971).

12. C. Lege, Ph. D. Dissertation (University of Florida,
Gainesville, Florida, 1980).

13. G. Wulff, W. Vesper, R. Grobe-Einsler and A. Sarhan,
Makromol. Chem., 178, 2799 (1977).

14. G. Wulff, W. Vesper, R. Grobe-Einsler and A. Sarhan,
Makromol. Chem., 178, 2817 (1977).








15. G. Wulff, A. Sarhan and K. Zabrocki, Tetrahedron Lett.,
44, 4329 (1973).

16. K. J. Shea and E. A. Thompson, J. Org. Chem. 43, 4255
(1978).

17. R. Kunin, "Ion Exchange Resins" (Wiley, New York, 1950).

18. A. R. Pitochelli, "Ion Exchange Catalysis and Matrix
Effects"" (Rohm and Haas, Philadelphia, 1975).

19. J. R. Millar, D. G. Smith and T. R. E. Kressman,
J. Chem. Soc., 304 (1965).

20. J. R. Millar, D. G. Smith, W. E. Marr and T. R. E.
Kressman, J. Chem. Soc., 2740 (1964).

21. H. E. Zaugg and W. B. Martin, Org. Reactions, 14, 52
(1965).

22. D. M. Soignet, R. J. Berni, J. I. Wadsworth and
R. R. Benerito, Anal. Letters, 5, 395 (1972).

23. G. B. Bachman and L. W. Heisey, J. Amer. Chem. Soc.,
68, 2496 (1946).

24. H. E. Carter, R. L. Frank and H. W. Johnston in
"Org. Syn.," Coll. Vol. III, E. C. Horning, Ed.
(Wiley, New York, 1955), p.167.

25. P. A. S. Smith, "Open Chain Nitrogen Compounds,"
1st Edition (W. A. Benjamin, New York, 1965) Chapter
Six.

26. J. S. Blair, J. Amer. Chem. Soc., 48, 96 (1926).

27. H. L. Gingrich, Ph. D. Dissertation (University of
Florida, Gainesville, Florida, 1977).

28. A. F. Pozharskii, A. M. Simonev and E. A. Zvezdina,
Khim. Geterotsiki. Soedin., 5, 889 (1967).

29. T. Kunitake and S. Shinkai, J. Amer. Chem. Soc., 93,
4247 (1971).

30. C. G. Overberger and C. J. Podsiadly, Bioorganic Chem.,
3, 16 (1974).

31. R. J. Kunin and K. A. Kun, J. Poly. Sci., A-I 6, 2689
(1969).

32. R. Kunin, E. Meitzner and N. Bortnick, J. Amer. Chem.
Soc., 84, 305 (1962).








33. J. R. Millar, D. G. Smith, W. E. Marr and T. R. E.
Kressman, J. Chem. Soc., 218 (1963).

34. W. L. Sederel and G. J. DeJong, J. Applied Poly. Sci.,
17, 2835 (1973).

35. W. G. Lloyd and T. J. Alfrey, J. Poly. Sci., 62, 301
(1962).

36. J. Manassen, Chim Ind. (Milan), 51, 1058 (1969).

37. S. J. Thomas and G. Webb, "Heterogeneous Catalysis,"
1st Edition (Wiley, New York, 1968)

38. W. P. Jencks and J. F. Kirsch, J. Amer. Chem. Soc., 86,
833 (1964).

39. B. S. Hartley and B. A. Kilby, Biochem. J., 56, 288
(1954).

40. H. Gutfreund and J. M. Sturtevant, Biochem. J., 63,
656 (1956).

41. W. Heitz and W. Kern, Angew. Makromolekulure Chem., 1,
150 (1967).

42. F. Schneider, Hoppe-Seyler's Z. Physiol. Chem., 348,
1034 (1967).





































INFRARED SPECTRA