Ester hydrolysis catalysis by a zinc enzyme model complex and synthesis of a bioconjugatable model complex derivative

MISSING IMAGE

Material Information

Title:
Ester hydrolysis catalysis by a zinc enzyme model complex and synthesis of a bioconjugatable model complex derivative
Physical Description:
Book
Creator:
Boeringer-Hartnup, Melody Ann, 1964-
Publication Date:

Record Information

Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 20851180
oclc - 32838509
System ID:
AA00022795:00001

Table of Contents
    Title Page
        Page i
    Acknowledgement
        Page ii
    Table of Contents
        Page iii
        Page iv
    List of Tables
        Page v
        Page vi
    List of Figures
        Page vii
        Page viii
    Abstract
        Page ix
        Page x
    Chapter 1. Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
    Chapter 2. Synthesis of a bioconjugatable zinc complex
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
    Chapter 3. Kinetic methods and characterization of solution species
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
    Chapter 4. Temperature dependence of methyl acetate and p-Nitrophenyl acetate hydrolyses
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
    Chapter 5. Reaction profile for (Zn11-1,5,9-triazacyclododecane (OH))3(CIO4)3HCIO4
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
    Chapter 6. Summary and conclusions
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
    Appendix A. 500 MHZ NMR spectra
        Page 132
        Page 133
    Appendix B. Sample calculation
        Page 134
        Page 135
    References
        Page 136
        Page 137
        Page 138
    Biographical sketch
        Page 139
        Page 140
Full Text









ESTER HYDROLYSIS CATALYSIS BY A ZINC ENZYME MODEL COMPLEX
AND SYNTHESIS OF A BIOCONJUGATABLE MODEL COMPLEX DERIVATIVE









By

MELODY ANN BOERINGER-HARTNUP


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

UNIVERSITY OF FLORIDA


1994













ACKNOWLEDGEMENTS


Grateful appreciation is given to all the members of my doctoral
committee, whose input and advice were invaluable in the execution of this
project. Particular gratitude is given to Dr. David Richardson, my research

advisor, who has always been available for questions and advice. I wish to

thank my husband, Scot: without his support and love I would have been a far
unhappier soul. I wish to thank my parents, Wayne and Arline Hartnup, for their
unflagging confidence that I would, indeed, eventually finish my schooling.
Finally, I wish to acknowledge my baby daughter, Tabitha Selene, without
whom I would certainly have finished sooner, but with a great deal less joy.

Generous support for this study has been provided by the
Interdisciplinary Center of Biotechnology Research, the National Institutes of

Health, and the University of Florida.













TABLE OF CONTENTS



pAge
ACKNO W LEDG EM ENTS ..................................................................................... ii

LIST O F TABLES .................................................................................................. v

LIST O F FIG URES ................................................................................................ vi

ABSTRACT ............................................................................................................. ix

CHAPTERS

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

2 SYNTHESIS OF A BIOCONJUGATABLE ZINC COMPLEX ............ 10

Introduction ............................................................................................. 10
Results and Discussion ......................................................................... 13
Sum m ary and Conclusions ................................................................... 48
Experim ental ........................................................................................... 50

3 KINETIC METHODS AND CHARACTERIZATION OF
SO LUTIO N SPEC IES .................................................................. 57

Introduction ............................................................................................. 57
Experim ental ........................................................................................... 61
Results and Discussion ......................................................................... 67

Sum m ary and Conclusions ................................................................... 88

4 TEMPERATURE DEPENDENCE OF METHYL ACETATE AND
p-NITROPHENYL ACETATE HYDROLYSES ............................. 90

Introduction .............. ...... ..................... 90
Results and Discussion ......................................................................... 95
Experim ental ............................................................................................ 105
Sum m ary and Conclusions ................................................................... 105







5 REACTION PROFILE FOR (Zn"l-1,5,9-
TRIAZACYCLODODECANE (OH))3(CI04)3HC104 ..................... 106

In tro d u ctio n .............................................................................................. 1 0 6
Results and Discussion ...................................111
E x pe rim e nta l ........................................................................................... 123
Summary and Conclusions .................................................................. 125

6 SUMMARY AND CONCLUSIONS ...................................................... 126

APPENDICES

A 500 MHZ NMR SPECTRA .................................................................. 132

B SAMPLE CALCULATION .................................................................... 134

R E F E R E N C E S ..................................................................................................... .. 136

B IO G R A PH IC A L S K ETC H ...................................................................................... 139







LIST OF TABLES


page
Table 2.1
300 MHz 1 H NMR Resonances ...................................................... 22

Table 2.2
78 MHz 13C NMR Resonances ..................................................... 24


Table 2.3
500 M Hz 1H NM R Resonances ..................................................... 41

Table 2.4
Zinc Complex 1H NMR Resonances .............................................. 45

Table 2.5
Zinc Complex 13C NMR Resonances .......................................... 47

Table 3.1
Variation of HEPES Buffer pH With Temperature ...................... 73


Table 3.2
Temperature Dependence of the Extinction Coefficient of
p-Nitrophenolate and the pKa of p-Nitrophenol .............. 80

Table 3.3
Temperature Dependence of the Deprotonation Constant
(pKa) of the ZnlI-Bound H20 in Zn(OH2)- ...................... 86

Table 4.1
Temperature Dependence of the Cataysis of Hydrolysis of
Methyl Acetate by Zn(O H)-1 .................................................. 97

Table 4.2
Methyl Acetate Hydrolysis Activation Parameters ...................... 98


Table 4.3
Temperature Dependence of Catalysis of Nitrophenyl Acetate
Hydrolysis by Zn(O H)-1 .......................................................... 99


v








Table 4.4
Temperature Dependence of Catalysis of Nitrophenyl Acetate
Hydrolysis by Zn(OH)-1 in the Presence of KCI .................. 100

Table 4.5
p-Nitrophenyl Acetate Hydrolysis Activation Parameters .............. 101


Table 5-.1
Ester S eries H ydrolyses ...................................................................... 115


Table 5-1
C ata lytic C o nditio ns ............................................................................. 124







LIST OF FIGURES


page
Figure 1.1
C A a ctiv e site ............................................................................................. 2

Figure 1.2
Metalloenzyme Model Systems ............................................................ 2

Figure 1.3
Catalytic Mechanism for Hydrolysis of Methyl Acetate by Chin's
C o m p le x ...................................................................................... .. 3

Figure 1.4
Kimura Metalloenzyme Model Complex .............................................. 4

Figure 1.5
Adaption of Schultz Bioconjugation ...................................................... 6

Figure 1.6
Bioconjugate Molecular Modelling Diagram ...................................... 9

Figure 2.1
Scheme for Synthesis of a Bioconjugatable Derivative of 1,5,9-
triazacyclododecane ................................................................. 12

Figure 2.2
P o lym e rizatio n ..................................................................................... . 14

Figure 2.3
Richman-Atkins Synthesis of Polyaza Macrocycles ........................ 15

Figure 2.4
Aromatic Structural Assignments for 13C NMR ................................ 20

Figure 2.5
1,5,9-Triazacyclododecane Ring Structural Assignments .............. 21

Figure 2.6
S tereochem istry .................................................................................. .. 39

Figure 3.1
Flow Cell Kinetic Setup ......................................................................... 68

Figure 3.2
Modified UV-visible Cell Kinetic Setup .............................................. 71








Figure 3.3
Modified UV-visible Cell ....................................................................... 72



Figure 3.4
Base Addition Kinetic Setup ................................................................ 77

Figure 3.5
Comparison of Experimental Points and a Program-Generated
Titration Curve ............................................................................ 84

Figure 4.1
Zn(OH)-1 ................................................................................................. 91

Figure 4.2
Nucleophilic Catalysis .......................................................................... 92

Figure 4.3
General Base Catalysis ........................................................................ 92

Figure 5.1
DNPA Brdnsted Plot ............................................................................... 116

Figure 5.2
Methyl Acetate Brbnsted Plot ............................................................... 117

Figure 5.3
Correlation Diagram .............................................................................. 118

Figure 5.4
Solvent Mediated Mechanism ............................................................. 120












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


ESTER HYDROLYSIS CATALYSIS BY A ZINC ENZYME MODEL COMPLEX
AND SYNTHESIS OF A BIOCONJUGATABLE MODEL COMPLEX DERIVATIVE


By

Melody Ann Boeringer-Hartnup

August, 1994

Chairman: David Richardson
Major Department: Chemistry

Zinc is a cofactor in over 100 enzymes, including the functionally diverse

group containing carbonic anhydrases, carboxypeptidases, alcohol

dehydrogenases, aldolases, peptidases, proteases, phosphatases,
transphosphorylases, DNA and RNA polymerases, and a transcarbamylase.
(Zn1-[12]aneN3(OH))3(C104)3HC104, Zn(OH)-I, a stable complex investigated

by Kimura et al., has been chosen for detailed mechanistic study due to its
observed catalysis of ester hydrolysis as well as its close approximation of the

carbonic anhydrase active site environment. In this study the catalytic rate
constants for a variety of activated and unactivated carboxylic acid ester

substrates were determined. Primary kinetic solvent isotope effects for methyl
acetate and p-nitrophenyl acetate hydrolysis were also measured.
Measurement of the temperature-dependent rate constants, resulting in the
calculation of the activation parameters, was undertaken for methyl acetate and
p-nitrophenyl acetate catalyzed hydrolysis by Zn(OH)-1. These kinetic studies







led to formulation of possible catalytic mechanisms and comparison to

biological catalytic systems. The p-nitrophenyl acetate catalyzed hydrolysis
was assigned to a nucleophilic catalytic mechanism. The methyl acetate

catalyzed hydrolysis was assigned to a solvent-mediated bifunctional catalytic

mechanism.
The synthesis of a modified complex that can be conjugated to a

monoclonal antibody is also described. Products were characterized by NMR,
elemental analysis, and mass spectrometry. This bioconjugatable product is
now ready for conjugation and eventual catalytic assessment.













CHAPTER 1

INTRODUCTION


Investigation of complexes that mimic the reactivity and structure of

the active site of metalloenzymes has been a popular route to a better
understanding of the reaction mechanisms of enzymes and development of
nonenzymatic catalysts. Models for zinc metalloenzyme active sites have
been extensively investigated.1 Over 100 zinc-dependent enzymes have
been identified, including the functionally diverse group containing carbonic

anhydrases, alcohol dehydrogenases, aldolases, peptidases,
carboxypeptidases, proteases, phosphatases, transphophorylases, DNA

and RNA-polymerases, and a transcarbamylase.2 Several of these
enzymes, including carboxypeptidase3 and carbonic anhydrase,4 feature a
tetrahedrally coordinated zinc at the active site. While carbonic anhydrase

(CA) is best known for its biological function, hydration of carbon dioxide

(Equation 1-1),


H2003 H- H + HC03- (1-1)
CO2 + H 20 -. W_




it has also been shown to catalyze the hydrolysis of esters in vitro.5 The

active site of CA features a distorted tetrahedral coordination at zinc
consisting of three imidazole nitrogens (histidines) and one water molecule







(Figure 1.1).1,5 The pKa of the Zn-bound water in human CA-B has been

reported to be = 7.4



H20
Zn INO^,N" H

H-N -) I- N His 119

His 94 His 96

Figure 1.1
CA active site


Carboxypeptidase can also function as an esterase and has a pseudo
tetrahedral zinc at the active site coordinated by two hystidine imidazoles,

water, and a carboxylate from glutamate.3
Two recently studied metalloenzyme model systems (Figure 1.2)
reported by Chin and coworkers6,7 are [(2,2'-dipyridylamine)Cu(OH2)2]2+,

(A), and [(trpn)Co(OH)(OH2)]2+(trpn = tris(aminopropyl)amine), (_).

_120 12
S,OH2 NH2 OH2 72+

HN 'CUi
H "C u r / C N H2 %', "
) OH2 -N N H2u OH

A B

Figure 1.2
Metalloenzyme Model Systems

These complexes were considered successful reactivity models for zinc

esterases due to their significant catalytic hydrolysis of methyl acetate in
water at 250C. Compared to esters highly activated toward hydrolysis, such







as p-nitrophenyl acetate, methyl acetate is unreactive toward ester
hydrolysis (catalytic turnover of 30 minutes). For example, mechanisms for
catalytic hydrolysis of unactivated esters by A and B focus on the roles of the

metal bound hydroxides and the cis waters. It has been suggested that the
mechanism of hydrolysis for B (Figure 1.3) involves initial coordination of the

ester (via displacement of water) to the metal followed by intramolecular
attack of M-OH on the coordinated ester.6

2+
NH 2+ NH OCH3
0 0 _,, ,.
CN 1 1+ C01
NH2OH / CH3 HOH


0 ;-CH30H
C
C q2+
OH K N H 2
NH 2 %, ,,O -'
2 H20 co l.
--N H,2 "



Figure 1.3
Catalytic Mechanism for Hydrolysis of Methyl Acetate by Chin's Complex

Nucleophilic attack is followed by ester cleavage and subsequent

dissociation of the acylated catalyst through hydrolysis to regenerate the
catalyst and complete the hydrolytic process.
Enzyme model complexes A and B do not structurally mimic any
metalloenzyme active sites. These complexes also exhibit product inhibition
of ester hydrolysis due to tight association of the acetate to the complex (this
complex is so stable as to be isolable). They do not catalyze hydrolysis of







esters at a rate even remotely close to values for enzymes such as carbonic
anhydrase. Therefore, the contribution of these models to our
understanding of metalloenzyme mechanistic pathways is limited.
Kimura et al.1 recently reported the model complex (Zn1-1,5,9-
(triazacyclododecane)(OH))3(ClO4)3HClO4, Zn(OH)- 1, (Figure 1.4),


H H 2+ H 7+

-. -Zn-, -Zn-.

N- --> N PKa=7.3 N +,H,
NI..1 N,, + H+

N N



Zn(OH2)-1 Zn(OH)-I


Figure 1.4
Kimura Metalloenzyme Model Complex

which is a closer structural model for CA as the metal, number of

coordinated nitrogens, and number of bound waters were the same as for
CA. The pKa of the coordinated water, 7.3, is also very close to that of the
CA active site water, = 7,4 and that of carboxypeptidase, 7-8.3 Similar pKa
values are crucial as the mechanism of catalysis of ester hydrolysis by CA is

thought to involve the zinc-hydroxide ionization state. Product inhibition is
much less important in this system than in the non-Zn complexes discussed
previously as the LZn-acetate complex has a binding constant K=102.6 in
contrast with Chin's Co1 complex, which has an LCo-acetate value of

K=106.1.6,7







Kimura et al. reported a second order rate constant of 0.041 M1s1 for

p-nitrophenyl acetate hydrolysis at pH = 8.2, 250C, in 50 mM HEPES buffer.1

The in vitro catalytic rate constant value for p-nitrophenyl acetate hydrolysis
by CA is 400 M-1s-1, which is much faster than that of Zn(OH)-1 and any

other enzyme active site model that has been reported. The comparison of
CA catalysis of ester hydrolysis to that of Zn(OH)-1 illustrates the catalytic

enhancement possible in metalloenzyme systems and the difficulty of
producing functional models with the high reactivity of enzymes. While

enzymatic catalytic enhancement involves many features, such as transition
state stabilization and favorable enzyme-substrate orientation, the present
study focuses on simulating substrate binding and activation at the
molecular recognition site of the enzyme. If a substrate and catalytic center
are brought together in a favorable orientation and distance for ester

hydrolysis, catalysis should be improved over the free catalyst.8,9 Therefore,
a better metalloenzyme model would incorporate a molecular recognition

element for substrate binding as well as a catalytic metal cofactor.
Enzymes are not the only biological species that use molecular
recognition. Antibodies, for instance, have strong affinities for their target
antigens.8,9 They normally lack the ability to modify the substrate once

bound, however. An adaption of a method described by Schultz8,9 (Figure
1.5) would incorporate a catalyst bound near the recognition region of an

antibody. The monoclonal antibody used, MOPC-315, was selected due to
its specificity for molecules bearing the 2,4-dinitrophenyl group. This
conjugated complex incorporates two of the essential rate-accelerating
elements of metalloenzymes, i.e., molecular recognition and association of
the substrate near a catalytic cofactor.




















02 NO2
1) / NSs,"-'H


H
I
s





H-N? N-H


H


L s 52H



5




2) NaCNBH3

3) DTT


MOPC-315
CFab


Figure 1.5
Adaption of Schultz Bioconjugation







The molecular modelling program Sybil was used in order to
determine if the linking compound and catalyst dimensions would result in

positioning of the catalyst that would be appropriate to react with a substrate
bound at the antibody binding site. In Figure 1.6 (courtesy of Dr. David
Richardson, Department of Chemistry, University of Florida, Gainesville,
Florida) a representation of a possible activated complex using the Fv region
of the mab, the tethered catalyst, and the bound substrate is displayed. It

appears that the catalyst has been tethered an appropriate distance from the
binding site in order to react with the substrate bound in the binding site.
Due to the overwhelming superiority of catalysis by enzymes relative

to standard catalysts under mild conditions, understanding the mechanistic
pathways for metalloenzymes would be extremely useful. Synthesis of
bioconjugates that would catalyze the synthesis of chiral compounds may be

possible. Synthesis of small amounts of pure chiral compounds is difficult
using traditional organic chemistry methods. Due to the ubiquity of zinc

containing enzyme active sites, production of a zinc metalloenzyme
biomimetic catalyst could be mechanistically enlightening and industrially

significant.
Synthesis of a bioconjugate requires the synthesis of the appropriate
bioconjugatable form of the model enzyme complex.8,9 The
bioconjugatable complex must have a functional group that can tether the
complex to the monoclonal antibody. This functional group is often attached
to the site of complex catalysis by a linking arm. The linking arm spaces the
catalyst a set distance from the binding site so that the catalyst is in proximity

to substrates when they are bound. Therefore, the linker must be of a length
to assure this positioning. Synthesis of a bioconjugatable complex is

described in Chapter 2.







In Chapter 3, the effect of experimental method, temperature, and
ionic strength upon the reaction rate for the catalysis of hydrolysis of

carboxylic acid esters by Zn(OH)- 1 is described. Evaluation of the
experimental method included initial kinetic studies that investigated
buffered vs. pH-stat controlled systems, reproducibility, and substrate

adsorption by components of the apparatus.
In Chapter 4, the hydrolytic mechanism was investigated using
temperature dependence for the catalytic hydrolysis of methyl acetate and p-
nitrophenyl acetate. The hydrolytic mechanism was further investigated in

an ester series study in Chapter 5. In this study, the catalytic rate constants
for carboxylic acid esters that are activated and unactivated towards
hydrolysis were measured. The catalytic rate constants were analyzed
using Br6nsted plots and correlation diagrams to elucidate the catalytic
mechanisms for hydrolysis of activated and unactivated carboxylic acid

esters. Primary kinetic solvent isotope effects for the catalytic hydrolysis of
methyl acetate and p-nitrophenyl acetate were also measured.















































Figure 1.6
Bioconjugate Molecular Modelling Diagram












CHAPTER 2

SYNTHESIS OF A BIOCONJUGATABLE ZINC COMPLEX


Introduction


This research is part of a study designed to produce a catalyst

bioconjugate that will have greatly enhanced catalytic properties over the
catalyst Zn(OH)-1 alone. Generally speaking, functional biomimetic model
catalysts are not as efficient as the native enzyme. For example, the
catalysis of hydrolysis of p-nitrophenyl acetate by CA has a rate constant of

400 M-is-1, 1 which is far higher than the value for the model system
(Zn(OH)-1) determined in this study, 0.058 M-is1 .5 This enhancement of

ca.104 is partially due to substrate binding by the enzyme. Association of
the substrate in the correct orientation to react with the metal cofactor results
in a unimolecular reaction rather than a bimolecular reaction with the free

catalyst. A higher reaction efficiency can be expected due to the
orientational effect and possible transition state stabilization by the enzyme

complex.8,9
As discussed in Chapter 1, Schultz's method8'9 encorporates many of

the aspects of enzyme catalysis into a bioconjugate. There are three basic
parts to the bioconjugate: the substrate binding site, the catalytic cofactor,
and the linking arm that connects them at the appropriate distance and
orientation. First, a monoclonal antibody, or mab, is chosen that is known to
bind a molecule that is similar in structure and functional groups to a class of







compounds whose catalysis is of interest. Monoclonal antibodies have
specific binding sites similar to the substrate binding sites of enzymes.8,9 In
this case, MOPC-315 was chosen as it has a binding site for dinitrophenyl
groups. When dinitrophenyl esters hydrolyze they produce

spectrophotometrically active species. This property is convenient for kinetic

studies.
Synthesis of a catalyst/antibody conjugate via the approach of
Schultz and coworkers8,9 requires the synthesis of an appropriate thiol

derivative of 1. (Figure 2.1). Finally, the Zn-thiol complex may be
incorporated into the MOPC-315. The completed bioconjugate can

potentially feature substrate specificity, orientational advantages, and
catalyst proximity to the reaction site.
In this chapter, syntheses of a thiol derivative of macrocycle 1 and its

zinc complex are described. Extensive characterizations of the products and

intermediates by NMR (300 and 500 MHz) are also given here.












0 0



H5C2 ~2H5


1.) NaH


EtOH, reflux
2 weeks
0 0 3x recrys.
\C / H /


H5C2 b2H5 H -NC


2




0 0

H-N N


Ts
4


H-N N-H


H


BH3-THF, ref lux 72 hrs.,
reflux 3 h in 6M HCI.


H -TN N-H

Ts


CHC13, triethylamine
p-toluenesulphonylchloride
24 h


H-N N-H


Ts
4


Na/Hg amalgam,
in MeOH, 48 h


H-N N-H


H
6


CISO3H,CHCI3
2.5 h


LiAIH4
Et20, 2 days


H-N N-H
NH
H


Figure 2.1
Scheme for Synthesis of a Bioconjugatable Derivative of 1,5,9-triazacyclododecane


O2C]


SH







QNJN-


Zn(CI04)2
EtOH


H
Zn-8


U\H









Results and Discussion


Synthesis
Synthesis of a p-aminomethyl benzyl substituted [12]-N3 macrocycle

has been reported by Helps and coworkers.10 Synthesis by other
investigators of other C-functionalized derivatives of polycycloaza

compounds has been accomplished by using the Richman-Atkins
approach11 for the initial cyclization. Due to the similarity of Helps and
coworkers' compound, 3-(p-cyanobenzyl)-1,5,9-triazacyclododecane, to our
target complex, their approach was chosen as the most promising.
Synthesis of Diethylbenzylmalonate, 2
Synthesis of diethylbenzylmalonate, 2 was modeled after the method

of Fonken and Johnson.12 This method features initial formation of
diethylsodiomalonate from diethylbenzylmalonate and NaH, followed by

addition of benzyl chloride. The yield was rather low, 20%, but was not

further investigated as a commercial source (Lancaster) became available
for the compound. The next three steps in the synthesis, cyclization,
tosylation, and reduction, were modeled after the synthesis by Helps and
coworkers.

Synthesis of 3-Benzyl-1,5,9-Triazacyclododecane-2,4-Dione, 3
Synthesis of 3-benzyl- 1,5,9-triazacyclododecane-2,4-dione, 3, was

straightforward in approach, but complex in execution.10 The reagents are
allowed to react in dry ethanol at reflux for 2 weeks. Initially, nucleophilic
attack of primary amines of the nonane on the carbonyl carbon releases
ethanol. At this point, cyclization may occur as the free primary amine can







attack the remaining ester carbonyl. However, the primary amine may also
attack a carbonyl of a different benzyl malonate, producing a polymer.

Low product yield, typical of such cyclization reactions, due to the
formation of noncyclized polymeric products (Figure 2.2) produced difficulty

in the workup of the reaction as the product is entrained in an intractable
orange polymeric substance. The product was obtained as white needle
crystals by precipitation; however, three recrystallizations were required to
produce a product free of contaminants.




000 0

N H
Et N Et Et
H

Figure 2.2
Polymerization

High dilution and addition of reactants in aliquots at time intervals
over the two weeks were utilized to improve reaction yield. High dilution

favors intramolecular attack by the remaining free amine to form the cyclized
product as the free amine is less likely to encounter another diethylester at
low concentrations of the ester. A reaction with lower reagent dilution

resulted in a lower product yield, as expected. Addition of the reagents in

aliquots did not appear to affect product yield significantly. Even higher
dilutions were not feasible as the reaction scale had already required the
use of a five liter reaction flask. As reaction times of bimolecular solution
reactions are tied to their dilutions, higher dilutions would also increase

reaction times. Theoretically, yield of the macrocycle relative to the polymer







would continue to improve as dilution increased. The formation of the initial
product (containing only one amide) would be slower, but the intramolecular

attack would occur at a much greater rate than would the intermolecular
reaction at higher dilution.

Synthesis of polyaza cycles has also been accomplished by the
Richman-Atkins method11 which features condensation of a substituted

diethylamine with good leaving groups (such as OTs, OMs, or halides) with
poly-sulfonamide sodium salts (Figure 2.3). This method typically has
several advantages over the simple cyclization, including faster reaction
times, higher yields, and lower reaction volumes. When X is a large group
such as tosyl or mesityl, much better yields are reported than for smaller
leaving groups such as halides. Observation of better yields for reactions
with tosyl or mesityl supports the theory that steric bulk reduces the formation

of the noncyclized polymer, thus favoring the formation of the cyclized
product. Reduced polymer formation avoids the necessity of high dilution
and improves the product yield, while the sulfonate ester leaving group
reduces the reaction time necessary.


Ts Ts
% Na+ + NNq
X Ts N NTs
Ts IN N' Na+ Ts IN N
T s



X = OTs, OMs, halides
Figure 2.3
Richman-Atkins Reaction Synthesis of Polyaza Macrocycles







As it was difficult to remove a single sulfonamide to form 6, cleaving the

sulfonamides of a polysulfonamide-derivatized amine would likely have

proved even more challenging.
Synthesis of 3-Benzyl-9-(p-Tolylsulfonyl)-1,5,9-Triazacyclododecane-2,4-
Dione, 4
In the synthesis of 4, formation of the sulfonanamide was necessary

as Helps and coworkers observed that unsubstituted ligands underwent
boron complexation during the subsequent BH3-THF amide reduction.10

The boron complex was also highly inert. Therefore, as the final product

desired is a zinc complex, this is highly undesirable. Sulfonamidation
appeared to protect the ligand from complexation, probably through the
disruption of the coordination site (by reducing the Lewis basicity of the

amine by addition of the highly electron-withdrawing tosyl group).
Following addition of triethylamine and toluene-p-sulfonyl chloride to

a solution of 3 in chloroform, the product, 4, was extremely difficult to purify
as it was very tacky and rather insoluble in most solvents, another negative
aspect of incorporating multiple sulfonamides. Several recrystallizations
were necessary in order to purify the compound. The yield (>80% in most
reactions) was high, which is common for reactions of this type.10
Synthesis of 3-Benzyl-9(p-Tolylsulfonyl)-1,5,9-Triazacyclododecane, 5
Synthesis of 3-benzyl-9-(p-tolylsulfonyl)-1,5,9-triazacyclododecane,

5, was based on an adaption of Helps and coworkers' method.10 The
reduction of the diamide to form 5 was accomplished by using BH3-THF.

Maintaining an inert atmosphere was necessary for this reaction. The
product was much more soluble than its predecessor, reducing the time
needed for purification as simple extraction with methylene chloride







produced relatively pure product. Further purification could then be attained

by chromatography, if needed.

Due to the onerous nature of the purification of 4, a method was

sought to transform 3 directly to 6. Various approaches were attempted
without success.10,11,13 Both reductive and acid hydrolytic methods were

attempted, but they produced decomposition products. A review of LiAIH4
chemistry revealed that it is an extremely powerful reducing agent that can

attack a wide variety of functional groups, often leading to decomposition
products in amide reductions.13 Birch reduction reduced the phenyl ring.
Synthesis of 3-Benzyl-1,5,9-Triazacyclododecane, 6

Cleavage of the sulfonamide group, forming 6, was then desired as
maintaining the tosyl group increased the difficulty of purification in following
reactions. Removing the tosyl group at this stage was ideal as there were

few functional groups present that could form side products. Synthesis of 3-
benzyl-1,5,9-triazacyclododecane, 6, was unsuccessful by using an
adaption of Helps and coworkers' method.10 They reported a 67% yield by
using HBr-acetic acid (45% w/v) and phenol mixture that was allowed to
reflux for 24 h. Reaction of compound 5 with this mixture produced a black

mixture after several hours and resulted in decomposition products. The
reaction was repeated a few times with the same result, and this approach

was abandoned. A similar hydrolytic cleavage method, 97% H2S04 at 100

C for 48 h, was described by Richman and Atkins, who had had success with
this method for tetraaza detosylations.11 The solution turned black in a few
hours and decomposition of the starting materials again occurred. Next,
reductive methods of sulfonamide cleavage were investigated.'3,14 LiAIH4
and Red-A again failed to produce product. The Birch reduction did have a
2% yield of product, but also reduced the aromatic group as a side product.







Attempts to maximize product by varying the Birch reduction conditions did
not improve yield.

Finally, a more gentle reducing agent (2% Na/Hg amalgam in dry
ethanol buffered by phosphate) was employed. This method had been
reported to detosylate polyazamacrocycles15 and worked without reducing

the aromatic group of the starting material. As the literature syntheses were
well after Richman and Atkins', it is likely that 1,5,9-triazacycles are

problematic when removing tosyl groups by using strongly acidic hydrolytic
methods. The yield was acceptable, 40%, and a combination of an

extraction series and chromatography were used to remove side products
and recover unreacted starting materials.
Synthesis of 3-(p-Benzenesulfonylchloride)-1, 5,9-Triazacyclododecane, 7
Having obtained 6, synthesis of the p-sulfonyl chloride, 3-(p-

benzylsulfonylchloride)-1,5,9-triazacyclododecane, Z, was based on the
literature synthesis of benzyl sulfonyl chloride.16 This reaction has been

well studied and various methods have been employed to make the reaction
require less chlorosulfonic acid, a very corrosive material.17 The reaction
requires two equivalents of acid to drive the product to the sulfonyl chloride

(Equations 2-1 and 2-2). However, as the second reaction is an equilibrium,

a large excess of acid is typically required.

RH + CIS3H -. RS03H + HQ (2-1)
RS03H + aCH RSO2CI + H2SO4 (2-2)


Addition of sodium chloride allows the reaction to be driven closer to
completion with less acid as the salt can react with the H2SO4, thus
removing it from solution and regenerating reactant (Equation 2-3). The







product was an oil that was separated from unreacted starting material by

using chromatography.




NaCI + H2SO4 NaHSO4 + HCI (2-3)


Synthesis of 3-(p-Thiobenzene)-1 ,5,9-Triazacyclododecane, 8 and Zn(OH)-
8

Synthesis of 3-(4-thiobenzyl)-1,5,9-triazacyclododecane, 8, employed
LiAIH4 in diethylether, a method well known for reduction of sulfonyl chloride

to thiols (Equations 2-4 and 2-5).18

RS02CI 0- RSO3H 0, RSOH )w RSSR RSH (2-4)

2RS02CI + 3LiAIH4 o- LiAICI2(SR)2 + 6 H2 + 2 LiAIO2 (2-5)



RSH
Formation of the Zn(OH)-8 complex (as well as Zn(OH)-6) was
accomplished similarly to that of model complex Zn(OH)-1.1 Simple addition
of the two components forms a soluble complex in methanol. The resulting

complex may be used in situ for the bioconjugation work.









Compound Characterizations
All comparisons and predictions of 1H/13C NMR peak positions for

the various products are based on information from Silverstein, Bassler, and
Morrill 19 for 1 H NMR peaks and from Breitmaier and Voelter 20 for 13C

NMR peaks unless otherwise noted. 1H NMR spin-spin coupling constant
values and assignments were also evaluated by using data in Silverstein,

Bassler and Morrill.19
13C NMR peak positions for the two aromatic systems appearing in

the various compounds were assigned by using the structural assignments
shown in Figure 2.4.

R3R4


b a b
a a
R1 R2 C'
CH3
x = H, SO2CI, SH

Figure 2.4
Aromatic Structural Assignments for 13C NMR

1H NMR peak positions for the triaza systems appearing in the

compounds were assigned by using the structural assignments shown in
Figure 2.5.




















Figure 2.5
1,5,9-triazacyclododecane Ring Structural Assignments

Attached proton tests, known as APT, are commonly used in 130 NMR
spectral assignment to distinguish between the various resonances due to
each type of carbon (10,20,30, and 40). The APT program available for the
QE-300 resolves resonances into those due to two types of carbons, ones

with an even number of protons attached to them (2and 40 carbons) and
those with an odd number of protons attached to them (1 0and 30 carbons).
This process21 is based upon simultaneous pulse sequences on the 1H and
13C channels of the NMR, which result in addition of the two types of FID for

"even" carbons and subtraction of the two FID for "odd" carbons.

In APT, the sequence produces spin-echos which are refocused at
time 2-. The coupling effect is dependent upon -. At time 2T, methine and

methyl groups have 130 magnetization components, which have phase
differences of 1800 with respect to the Larmor frequency. Therefore,
decoupling at time 2T gives a full negative signal. Methylene carbons,

however, have phase differences of 3600, 00, and, -3600 for their three
magnetization components, resulting in a full positive signal at 2-T. Finally,

quartenary carbons behave similarly to methylene carbons as they are

unaffected by the 1 H channel.









TABLE 2.1
300 mHz 1H NMR Resonances


Compound


diethylbenzyimalonate, 2
CDC13


3-benzyl- 1,5, 9-triazacyclo
dodecane-2,4-dione, 3
CD2CI2


3-benzyl-9-(p-tolylsulfonyl)-
1,5,9-triazacyclododecane-2,4-
dione, 4
CDC13


3-benzyl-9-(p-tolylsulfonyl)-
1,5,9-triazacyclododecane, 5
CD2Cl2


Peak(s),lnteg.a

7.25,m,5H
4.16,dqt,4H
3.67,t,1 H
3.24,d,2H
1.22,t,6H

7.61 ,s,2H
7.22,m,5H
3.57,tdd,2H
3.18,m,5H
2.86,ddd,2H
2.62,ddd,2H
1.72,m,2H
1.54,m,2H
1.15,s,1H


7.68,d,2H
7.35,d,2H
7.23,m,5H
6.50,s,2H
3.45-3.56
m,2H
3.27-3.45
m,3H
2.90-3.20
m + d,6H
2.40,s,3H
1.80,m,2H
1.60,m,2H

7.70,d,2H
7.30,d,2H
7.25,d,2H
7.23,m,3H
3.26-3.46
m,4H
2.60-2.72
m,4H
2.42-2.58
m + d,6H
2.40,s,3H
1.94,m,1 H
1.64,m,4H
1.40,s,2H


Assignment


Ar
CH2
CH
Ar-CH2
CH3

NHCOR
Ar
g
g, i, J
e
e
f
f
NH

Ts
Ts
Ar
NHCOR
g

g,i

e,j

Ts-CH3
f
f


Ts
Ts
Ar, ortho
Ar, meta + para
e?

g

h?
Ts-CH3
i
f
NH


j(Hz)


= 1.6, 7.1
= 15.8
7.7
= 14.9


=1.8,3.6,6.3

1.8,3.2,5.2
j= 1.8,3.2,5.2


j=8.6
j=8.6


j= 7.7


j= 8.0
j=8.2
j= 7.4


j= 7.7


appm from 300.15 MHz










Table 2.1-- continued


Compound


3-benzyl-1,5,9-triazacyclo dodecane,
6
MeOD


3-(p-benzylsulfonylchloride)
1,5,9-triazacyclododecane, 7
MeOD







3-(p-thiobenzyl)
1,5,9-triazacyclododecane, 8
MeOD


Peak(s), Integ.a

7.25,m,5H

3.00-2.70
m,10H
2.55-2.65
d + dd, 4H
2.13,m,1H
1.68,qnt,4H

7.76,d,2H
7.29,d,2H
3.05-2.72
m,1OH
2.54-2.70
m + d,4H
2.20,m,1H
1.76,qnt,4H


7.79,d,2H
7.30,d,2H
2.85-2.65
m,1OH
2.55-2.62
m + d,4H
2.10,m,1H
1.67,qnt,4H


Assignment


Ar

e, g, h


Ar
Ar
e,g, h


Ar
Ar
e, g, h


appm from 300.15 MHz


j values


j= 7.5
= 10.3, 12.3

= 5.5


=8.4
=8.4


7.5


=5.1


j=9.0
j= 9.0


j= 7.6


j= 5.2









TABLE 2.2
78 mHz 13C NMR Resonances


Compound


diethylbenzylmalonate, 2
CDC13




3-benzyl-1,5,9-triazacyclo
dodecane-2,4-dione, 3
CD2CI2




3-benzyl-9-(p-tolylsulfonyl)-
1,5, 9-triazacyclododecane-2,4-
dione, 4
DMSO





3-benzyl-9-(p-tolylsulfonyl)-
1,5,9-triazacyclododecane, 5
CD2Cl2






3-benzyl-1,5,9-triazacyclo
dodecane, 6
MeOD




appm from 78.481 MHz
* odd # protons attached


Aromatic
Peaksa

126.68*
128.45'
128.81"
137.87


125.50*
129.00*
129.65*
149.00



126.38*
127.15*
128.57*
129.07*
130.21"
138.49
140.09
143.80

125.77*
127.14*
128.21*
128.84*
129.24*
137.28
140.97
142.98

125.95*
128.10*
128.63*
139.28


Assignment


Aliphatic
Peaksa

13.97*
34.67*
53.82
61.38
168.79

27.10
33.10
40.05
49.00
56.48*
168.80

21.37*
27.20
32.77
35.98
42.51
57.00*
169.25


21.20*
26.11
37.84
40.15*
44.21
45.21
49.76


25.39
37.26
37.83*
47.94
48.24
52.73


Assignment


CH3
CH
Ar-CH2
CH2
COOR

f
I
e or g
e or g
i
h

Ts-CH3
f
I
e org
e org
i
h


Ts-CH3
f
J

e or g or h
e org or h
e or g or h


f


e org
e org
h










Table 2.2 -- continued


Compound


3-(p-sulfonylchloridebenzyl)
1,5,9-triazacyclododecane, 7
MeOD




3-(p-thiobenzyl)
1,5,9-triazacyclododecane, 8
MeOD




appm from 78.481 MHz
* odd # protons attached


Assignment


C
b
a ord
d ora


Aromatic
Peaksa

125.78*
128.62*
141.82
143.19



125.78*
128.48*
130.93
141.80


Aliphatic
Peaksa

24.82
36.80
37.33*
47.36
47.68
52.41

25.37
36.89
37.79*
47.97
48.23
52.65


Assignment


f
J

e org
e org
h

f


e org
e org
h









NMR of Diethylbenzylmalonate, 2
Diethylbenzylmalonate, 2, was characterized by 1H / 13C NMR

spectra and elemental analysis. The 1H NMR spectrum (Table 2.1) was

entirely consistent with the compound and relatively uncomplicated relative
to later products. A complex overlapping series of peaks at 7.25 ppm was

observed. Multiplets are characteristic of mono-substituted aromatics and
are due to the similar chemical environments experienced by the meta and
para position protons, which should have two doublet of doublets sets and a
triplet, respectively. The ortho position protons may exist as a resolved
doublet or doublets, but often they are lost in the multitude of peaks due to

the other aromatic protons. The doublet of quartets at 4.16 ppm, which
integrated at 4 protons, is due to the methylene protons on the methyl

groups. Typical values for methylene protons in similar environments (M-
OC(=O)R) are about 4.1 ppm. The triplet at 3.67 ppm, which integrated as
one proton, is due to the methine proton adjacent to the carbonyl carbon.

Typical values for methine protons in similar environments (adjacent to ester
functionalities) range from 3.6-3.7 ppm. The doublet at 3.24 ppm, which
integrated as two protons, can be assigned (by integration and splitting) to

the methylene group adjacent to the aromatic group. Finally, the triplet that
integrates as six protons at 1.22 ppm, is clearly (by position, integration, and
splitting) due to the methyl groups of the diethyl ester moieties. Typical
resonance values for methine protons in similar environments (once
removed from ester functionalities) are around 1.3 ppm.

The 13C NMR spectrum (Table 2.2) was consistent with the desired
product. Four peaks in the aromatic region were expected and were
observed for this mono-substituted aromatic. An APT indicated that three of







the peaks (128.81, 128.45, and 126.68 ppm) were due to carbons with an

odd number of protons attached, while one of the peaks (137.87 ppm) was
due to a carbon with an even number of protons attached. These results are

consistent with the three chemically equivalent types of carbons with one
proton and one carbon with no protons in this molecule. Further support for

these assignments comes from the spectrum of toluene, whose mono-
substituted aromatic structure is similar. Toluene's spectrum has a peak at
137.8 ppm for the substituted carbon and three peaks (128.6, 128.1, and

125.9 ppm) for the carbons at the ortho, meta, and para positions to
substitution. Comparison of the two spectra supports the assignments made

for diethylbenzylmalonate.

The 13C NMR peaks for the carbons associated with the ester
functional groups may be compared with those of esters such as
diethylpropanedioate. Diethylpropanedioate has peaks at 61.0 and 14.1

ppm for the methylene and methyl groups of its ethyl groups and a peak at
165.7 ppm for the carbonyl carbon of the ester functional group. These

resonances compare favorably with those observed for the product. These
assignments were also supported by an APT assignment of an odd number
of protons for the peaks at 14.1, 165.7, and 34.67 ppm and an even number

of protons for the peaks at 53.82 and 61.38 ppm. The final two peaks, 34.67
and 53.82 ppm, were thus easily assigned as they must be due to the i andj

carbons, respectively.
NMR of 3-Benzyl-1,5,9-Triazacyclododecane-2,4-Dione, 3
3-Benzyl-1,5,9-triazacyclododecane-2,4-dione, 3, was characterized
by 1H/130 NMR spectra and elemental analysis. The 1H NMR spectrum
(Table 2.1) of 3-benzyl-1,5,9-triazacyclododecane-2,4-dione was much
more challenging to interpret than that of diethylbenzyl malonate due to the







increasing complexity of the molecule. A broad singlet at 7.61 ppm, which

integrated as two protons, was assigned to the amide nitrogen protons.
Amides typically appear at a broad range, 8.5-5.0 ppm. A complex
overlapping series of peaks that integrated as five protons was observed at
7.22 ppm. Again, this is characteristic of mono-substituted aromatics. The

triplet of doublets of doublets at 3.57 ppm, which integrated as two protons,

is due to one of the two sets of two protons that are adjacent to the amide
functionality (g). There are two sets of peaks for several positions, which
correlate to protons that are on either face of the triaza cycle. The
assignment of these peaks as due to the protons on the carbon adjacent to

the amide (g) rather than the amine (e) was supported by the following
observations. Typical values for methylene protons in similar environments

(adjacent to amides) are about 3.4 ppm, which is close to the observed
value. Also, protons on carbons adjacent to amines typically range from 2.4-

2.7 ppm.

Other evidence supporting the assignments included resonances
observed for 1,5,9-triazanonane (one of the starting materials). It has
triplets at 2.65 and 2.55 ppm and a singlet at 1.1 ppm (in CDCI3). Since this

compound has protons that are adjacent to amines, it is unlikely that protons
adjacent to the amines in 3 would be remote from that region. Therefore, the

doublet of doublet of doublets observed at 2.86 and 2.62 ppm (both
integrating as two protons) should be assigned to the protons on the
carbons adjacent to the amines. Finally, all three splitting constants for the

peaks at 2.86 and 2.62 ppm are identical; therefore, they are from protons on
chemically equivalent sites and whose slightly different resonance positions
correlate to their location on opposite faces of the polyaza macrocycle. Also,







the splitting constants for the peaks at 3.57 ppm differ from the other two

sets.
As all three sets of splitting constants (due to protons in positions e

and g) had one value in common (1.8 Hz), it would be reasonable to assign
this splitting to the proton on the opposite face of the carbon in between

them (). Generally low splitting constant values are observed for protons
that are at large dihedral angles from the proton being split and because the

splitting due to protons on that middle carbon should be independent of the
proton signal being split. Therefore, the set of splitting constants that have
the values 3.2 and 3.6 hz was assigned to the splitting due to the other
proton of the carbon on f. This proton would be on the same face as the one

it is splitting, with a smaller dihedral angle; therefore, a larger splitting
constant value would be expected. The two values are not identical, but

they are very close (or at least much closer than the last set of splitting
constants), thus supporting this assignment. Finally, the large difference
between the magnitude and the value itself for the last set of splitting
constants, 5.2 and 6.3, indicates that this splitting is due to the geminal
proton interaction. This splitting would typically be of higher value than an

adjacent carbon's proton. The two splittings would also differ for a proton
attached to a carbon adjacent to an amine rather than an amide. As their

environments are not chemically equivalent, their geminal proton splitting
would be different for the two sets. Geminal proton splitting constants have a
0-30 Hz range, with a typical value being 12-15. This value is on the low
end of the range, but not as low as the other two sets would have been!
The multiplet at 3.18 ppm, which integrated as five protons, can be
assigned (by the process of elimination and relative position) to the second
set of two protons that are adjacent to the amide functionality (g), the







methylene group adjacent to the aromatic (), and the methine group

adjacent to the ester functionality (i).
The multiplets at 1.72 and 1.54 ppm, which integrate as two protons
each, can be assigned to the protons on the carbon P from the amine (f. The

two sets again reflect the two faces of the polyaza cycle. Their position is
also very similar to that of the protons on the carbon P from the amine for

1,5,9-triazanonane. Protons in similar environments have positions about

1.5-1.7 ppm. Finally, the broad singlet at 1.15 ppm can be assigned to the

amine nitrogen proton.
The 13C NMR spectrum (Table 2.2) was fairly simple to interpret. The
four expected peaks for the aromatic region were observed, and with the
exception of the peak for the carbon nearest modification, they are very

similar to the starting material. The carbonyl carbon remained at a very
similar field strength, which is consistent for carbonyl carbons of ester and

amide functionalities. Methanoic acid amide-N-methylamide-anti, for
instance, has a carbonyl carbon resonance of 168.7 ppm.
Predictably, in the aliphatic region, the resonances due to the ethyl
groups disappeared, to be replaced by peaks of the cyclic triaza compound.

The peak at 56.48 ppm had an APT that indicated an odd number of protons
attached to the carbon. All other resonances had an even number of
protons attached to the carbon. Therefore, that resonance had to be due to
the methine carbon. The peaks at 49.00 and 40.05 ppm were assigned as
CH2NHR due to their position and because they were very similar to
resonances of the starting material, 1,5,9-triazanonane. Propylamine has an
c-carbon resonance of 42.3 ppm, a typical value for such amines. The

starting material 2 (whose carbon peaks should be very similar to the
product) with the possible exception of the carbon adjacent to the amide







functionality, had peaks at 33.79, 40.56, and 47.89 ppm in CDC13. The
product peak at 40.05 ppm should be due to the carbon adjacent to the

amine as it should not change significantly upon cyclization. The peak at
49.00 ppm was assigned to the carbon adjacent to the amide functionality

as it changes slightly from that of the starting material.
The last two peaks, 27.10 and 33.10 ppm, were difficult to assign as
they are within the range of values observed for similar. Initially, the 33.10
ppm resonance was assigned to the f carbon as the value for the starting
material had a resonance of 33.79 ppm for the corresponding carbon. Upon
synthesis of further products and evaluation of their 13C NMR spectra, it was

determined that this peak was due to the j carbon. Assignment was mainly

based on the observed 5 ppm downfield shift observed upon reduction of
the amide to the amine. The carbonyl carbon is p3 to the j carbon and y to

the f. Thus the j resonance should be affected more by the reduction than

should the f resonance.
NMR of 3-Benzyl-9-(p-Tolylsulfonyl)-1,5,9-Triazacyclododecane-2,4-Dione,

4
3- Benzyl-9-(p-tolylsulfonyl)- 1,5,9-triazacyclododecane-2,4-dione, 4,

was characterized by 1H/13C NMR spectra and elemental analysis. The 1 H
NMR spectrum (Table 2.1) was difficult to evaluate as there were many

overlapping peaks and the insolubility of the compound produced spectra of
relatively poor signal to noise ratio, even by using high acquisition numbers.
The aromatic resonances were simple to interpret as there was an addition
of two doublets, exactly what would be expected for addition of a p-
substituted aromatic to a system. The splitting constant value, 8.6 Hz for
both, is within the range of observed values, 6-10 Hz, and very close to the
typical value of 9 Hz. A broad singlet that integrated as two protons was







assigned to the amide protons. The multiplets were assigned by their

position relative to their position in the starting material as well as the

integrated number of protons. Therefore, the 3.45-3.56 ppm multiplet (two
protons) was assigned to one set (one face) of the g protons. The other set

was assigned to the 3.27-3.45 ppm multiplet (three protons) with the final
proton being assigned to the methine proton, i. The multiplet at 2.90-3.20

ppm (six protons) was assigned to both sets of e and the j protons. The

doublet due to the j protons was resolvable and had a splitting constant of
7.7 Hz. Typical splitting constants for CHA-CHB splitting with free rotation
are 6-8 Hz, with 7 Hz being typical. The singlet at 2.40 ppm integrated as
three protons and was located very close to the resonance of the methyl
group of the starting material, p-toluenesulfonyl chloride. Therefore, it was

assigned to the methyl group on the tosyl group. The last two multiplets,
1.80 and 1.60 ppm, integrating at two protons apiece, were assigned to the

two sets (faces) of the f protons.
The 13C NMR spectrum (Table 2.2) was uncomplicated in the

aliphatic region. Only one new resonance appears, which has an odd
number of protons by APT. Therefore, it was due to the methyl group on the

tosyl group. There were some resonance shifts for the carbons closest to the
site of substitution. The shifts were only a few ppm and easily accounted for

by the change in chemical environment due to the substitution by the tosyl
group.
The aromatic peaks were more difficult to assign given that another
aromatic sytem, the Ts group, had been added to the system. The Ts group
has two 30 and two 40 carbons, which should appear as two 'even' and two
'odd' peaks, repectively, by APT. These new peaks were observed.








The 40 carbons were assigned by comparison of observed
resonances with product 7, an aromatic sulfonyl chloride aliphatically

substituted in the para postition. The sulfonyl chloride has a very similar

aromatic system to that of the aromatic in 4. These assignments were also
consistent with the resonances for benzene sulfonyl chloride, 144.3, 129.9,
126.9, and 135.5 ppm for the carbons corresponding to a,b,c and d,

respectively. Finally, comparison with the 13C NMR spectrum for 6 (also in a
highly polar solvent) supports assignment of a (the only other 'even'

aromatic carbon) as there is only one aromatic system in 6 with a single
'even' carbon.

The 30 carbons were assigned due to appearance of new peak

positions relative to the starting material and to literature compounds, as

above. Further support for the assignments came from comparison of the
peak positions of 4 with13C NMR spectrum of compound 3, whose 30
carbons of its aromatic system are remote from the site of modification.
N MR of 3-Benzyl-9-(p-Tolylsulfonyl)-1 .5,9-Triazacyclododecane, 5

3-Benzyl-9-(p-tolylsulfonyl)-1,5,9-triazacyclododecane, 5, was

characterized by 1H/13C NMR spectra and elemental analysis. The 1H NMR
spectrum (Table 2.1) was somewhat easier to interpret than for the diamide
starting material due to improved peak resolution, which was partly due to
improved solubility. The resonances due to the tosyl group remain
essentially unchanged while a doublet was resolved from the multiplet due
to the original aromatic group. This doublet is due to the protons ortho to the
site of substitution. The splitting constant has a value of 7.4 Hz, which is
within the 6-10 Hz range for similar splitting.
The aliphatic region exhibited significant changes, especially for
groups in close proximity to the amide reduction. The methine proton, c to







the reaction site, is now at 1.94 ppm (integrated as one proton) rather than

around 3.3 ppm, which is consistent with replacement of an electron

withdrawing carbonyl with a methylene, resulting in an upfield shift. The j
resonance, P to the site of substitution, has now been assigned to 2.45 ppm

rather than 3.0 ppm for the same reasons. The broad peak at 1.40 ppm that
integrated at two protons was assigned to the new amine protons. This

position is similar to that of the amine proton earlier observed for 3-benzyl-
1,5,9-triazacyclododecane-2,4-dione. The multiplet at 1.64 ppm integrated
as four protons and was easily assigned to the f protons. As the two faces
are now unresolvable, this indicates improved symmetry for the compound.
Two rather rigid amide groups are being replaced by two methylene groups,
which have improved rotational and vibrational freedom over the amides,

promoting better overall symmetry for the molecule. The singlet at 2.42 ppm,
integrating as three protons, was relatively unchanged and was assigned to
the methyl group of the tosyl group. As it is remote to the site of modification,

this is expected. The remaining resonances were due to three multiplets,
each integrating as four protons, which are due to the three sets of
methylene protons (e, g, and h) that are adjacent to amines. Again, it

should be noted that rather than two resolved facial sets of two protons
each, the sets appear now as a multiplet worth four protons. This feature is

even more striking in the next product, where removal of the tosyl group
further improves chemical equivalency. Further analysis of the complex
multiplets by a 500 mHz 1H NMR (Table 2.3 and its relevant discussion) was
necessary in order to assign this region.
The 130 NMR spectrum (Table 2.2) was similar to the starting material
in many respects. The aromatic region was basically unchanged, which is
reasonable as the aromatic regions are remote from the site of modification.







The aliphatic region had some striking resonance shifts and an additional

resonance. Again, the APT results were very helpful in analysing the results

as there were several resonance shifts. The APT indicated that the
resonances at 21.20 and 40.15 ppm were due to carbons with an odd

number of protons attached. Therefore, the methine carbon has shifted
upfield 17 ppm, which is reasonable as explained in the proton spectrum

analysis. The resonance at 21.20 ppm was again assigned to the methyl of
the tosyl group. The j carbon ( 3 to the modification site) resonance was
shifted a few ppm downfield while the f carbon (y to the modication site)

resonance remained essentially unchanged. The three remaining peaks
are only 5 ppm apart, 44.21,45.21, and 49.76 ppm, and were assigned to the
three sets of carbons which are a to the amines. They should be very similar

as the three systems are chemically very similar.

Characterization of 3-Benzyl-1,5,9-Triazacyclododecane, 6
3-Benzyl-1,5,9-triazacyclododecane, 6, was characterized by 1H/13C
NMR spectra and mass spectrometry. The 1 H NMR spectrum (Table 2.1)
indicated increased similarity in the chemical environment, relative to the
starting material, for the three sets of protons which were a to the amines (e,

g, and h ). Overlapping peaks should be expected as detosylation of the
tosylated amine produces an amine that is more similar to the other two
amine sets. Therefore, a complex series of multiplets that integrated as ten
protons was assigned to both facial sets of protons for both sets of protons a

to the amines remote from modification (e and g) to the methine proton, i.
One facial set of the protons on the carbon adjacent to the methine proton (h
) was also included. Further analysis of the complex multiplets by a 500
mHz 1H NMR (Table 2.3 and its relevant discussion) was necessary in order
to assign this region. The other set was observed as a doublet of doublets







that integrated as two protons at 2.60 ppm, slightly further upfield The
observed splitting constants are within the range observed for splittings of
similar compounds. The doublet at 2.55 ppm which integrated as two
protons was assigned again as the j protons. The multiplet at 2.13 ppm that
integrated as one proton was assigned to the methine proton. Its' position
was observed slightly shifted downfield; however, the change in solvent from

methylene chloride to methanol is likely responsible. Finally, the increased
symmetry of the compound is best illustrated by the resolution of the two sets

of multiplets for the methylene groups at 1.68 ppm that integrated as four
protons into a single quintet. The splitting constant, 5.5 Hz, is reasonable for
compounds of this type (1,5,9-triazanonane has a splitting constant for the

corresponding protons of 6.8 hz). The amine protons are absent in this
spectrum due to exchange with the acidic methanolic proton.

The 13C NMR spectrum (Table 2.2) was very similar to that of 3-
benzyl-9-(p-tolylsulfonyl)-1,5,9-triazacyclododecane except for the absence

of the aromatic peaks due to the tosyl group and the peak in the aliphatic
region due to the methyl group of the tosyl. The spectrum may also be

compared with the spectrum of 3 as the aromatic group is relatively remote
to the reduction, affecting only the resonance due toa significantly. An APT
indicated tha there was only one carbon that had an odd number of protons

attached. This peak was assigned to the methine carbon as it was close to
the resonance of the methine carbon of the starting material. The rest of the

peaks had resonances similar to the starting materials and their
assignments were unchanged.
Chemical ionization mass spectrometry produces an (M + 1)
molecular ion peak. The calculated (M + 1) peak for 016H27N3 was







262.228 amu, as compared to the experimentally determined value of

262.220 amu.
NMR of 3-(p-Benzylsulfonylchloride)-1, 5,9-Triazacyclododecane, 7
3-(p-Benzylsulfonylchloride)-1,5,9-triazacyclododecane, 7, was

characterized by 1H/13C NMR spectra. The 1 H NMR spectrum (Table 2.1)
exibited two new doublets, each integrating as two protons, that is

characteristic of para-substituted aromatics. The splitting constants of 8.4
Hz are very similar to splitting constants for the tosyl group observed in

earlier products. As the substitution pattern is similar in both compounds,
similar splitting constants are expected. The rest of the spectrum is very
similar to the starting material, which is consistent with a site of modification

that is remote to nonaromatic protons.
The 13C NMR spectrum (Table 2.2) was basically unchanged from
the starting material except for the aromatic carbon resonances. An APT
indicated that only two aromatic peaks had carbons with an odd number of

protons attached, which is consistent with substitution at an additional site.

A profound downfield shift of about 15 ppm was also observed for the new
"even" peak, which is reasonable as the carbon was modified by a very

electronegative sulfonyl chloride group. Resonances observed for

benzenesulfonyl chloride (144.3,129.9,126.9, and 135.5 ppm) are similar to
the observed product, indicating a successful synthesis of the sulfonyl

chloride. The resonance for the para-substituted carbon was 144.3 ppm,
very close to the value of both resonances that are shifted downfield. It is
difficult to assign which of these peaks belongs to each carbon as the
resonance positions are similar and the ranges expected for each type of
carbon are within the other carbon's expected resonance range.








NMR of 3-(p-Thiobenzyl)-1,5,9-Triazacyclododecane, 8
3-(p-Thiobenzyl)-1,5,9-triazacyclododecane 8, was characterized by
1H/13C NMR spectra. The 1H spectrum (Table 2.1) of the product was

essentially unchanged from the starting material, making it difficult to

ascertain that reduction had occurred. The 13C NMR spectrum was much
more enlightening. Further analysis of the complex multiplets by a 500 mHz
1 H NMR (Table 2.3 and its relevant discussion) only slightly improved the

situation.
A slight upfield shift for one of the aromatic doublets, which would be

consistent with reduction of the sulfur, was observed. Generalized tables of
substituted aromatics typically do not differentiate between different sulfur-
containing groups in predicting resonance position. Therefore, presence of

the sulfur itself seems to be the dominant effect on resonance position, not

the specific type of sulfur-containing group. Thus, a small change in position
is not abnormal for their 1H NMR spectra.
The 13C NMR spectrum (Table 2.2) was very similar to that of the

sulfonyl chloride starting material with one very significant change, the
position of one of the aromatic carbon resonances. An 11 ppm upfield shift
in resonance was observed. Reduction of the sulfur would be expected to
affect the substituted carbon in this manner. Also, the spectrum of p-
thiophenol shows a similar trend relative to benzenesulfonyl chloride. The

sulfur-substituted carbon in the sulfonyl chloride had a resonance at 144.3
ppm and a resonance at 130.5 ppm for the thiol. Observation of a
resonance at 130.93 ppm supports the assertion of a successful reduction to

the thiol. Furthermore, a thiol 'stench' was observed upon completion of the
reaction.







500 MHz Proton NMR Spectra of Selected Compounds
Due to the region of overlapping peaks in the 300 MHz spectra of 5, 6,

8, and Zn(OH)-6, 500 MHz 1H NMR spectra were run to improve peak
resolution. The characterization by 500 MHz 1H NMR of the tosylated

triamine 5 resulted in detailed resonance assignments (Table 2.3).
Doublet/multiplet overlap at 2.42-2.58 ppm was resolved into a doublet and

a multiplet (see Appendix A for spectrum). The resonance position and
splitting constant of the doublet was consistent with earlier assignment to the
j protons. Furthermore, the protons adjacent to the chiral carbon (h) are
now assignable (Figure 2.6). As expected, both pairs of protons produce a

doublet of doublets where one splitting constant for both pairs is same (12.1
Hz) and the other splitting constant is quite different. One splitting constant

for both pairs should be the same as it is due to the geminal proton and the
other splitting constant should be quite different as the splitting is due to the
proton on the chiral carbon (which is cis to one pair of protons and trans to

the other pair of protons). This difference allows assignment of each pair of
protons, the trans set being the pair with a small (3.1 Hz) splitting constant

and the cis pair being the pair with the larger splitting constant (10.4 Hz).

N -.Ts

N~T


H HH Cb



Figure 2.6
Stereochemistry






40
Splitting constants were also calculated for the multiplet due to the protons
on the carbon adjacent to the Ts group (e). These two pairs of protons are

expected to exhibit a doublet of doublet of doublets. The splitting constants

should be largest for the geminal proton splitting, a good bit less for the cis
proton on the adjacent carbon, and smallest for the trans proton. The
splitting constants for one pair of protons (the other pair is obscured by one

pair of the f protons) are consistent with this prediction, 14.3 ,7.1, and 5.9

Hz.











500 mHz


TABLE 2.3
1 H NMR Resonances


Compound
3-benzyl-9-(p-tolylsulfonyl)-
1,5,9-triazacyclododecane, 5
MeoD


3-benzyl-1,5,9-triazacyclododecane,
6

MeoD










3-(p-thiobenzyl)
1,5,9-triazacyclododecane, 8
MeOD


Peak(s), Integ
7.70,d,2H
7.30,d,2H
7.25,d,2H
7.23,m,3H
3.08-3.15
ddd,2H
2.85-2.94
dd +m, 6H
2.80-2.85
dd, 2H
2.50-2.60
dd + m, 4H
2.45, d, 2H
2.40,s,3H
1.94,m,1H
1.64,m,4H


7.23,m,3H
2.72-2.78
m, 4H
2.67-2.70
dd, 2H
2.59-2.65
ddd, 2H
2.51-2.57
ddd, 2H
2.41-2.48
dd +d, 4H
1.94,m,1 H
1.64,qnt,4H

7.75,d,2H
7.25,d,2H
2.85-2.65
m,1OH
2.55-2.62
m + d,4H
2.10,m,1H
1.67,qnt,4H


Assignment
Ts
Ts
Ar
Ar
e

e+g


htrans


I
Ts-CH3
i
f


hrans


g
g


hcis
i
f

Ar
Ar
e, g, and htrans


j(Hz)
= 8.0
j= 8.2
= 7.4

= 5.9,7.1,
and 14.3


j = 3.1, 12.1

j = 10.4, 12.1
(for the dd)
j=7.4


j = 3.0, 11.9

j = 4.6, 7.3,
and 12.3
j = 4.6, 7.3,
and 12.3
j= 7.8
j = 10.1, 11.9

j=5.5

j= 8.4
j= 8.4


j, hcis


j=5.2


Note: ppm from 499.943 MHz


hcis








Characterization of 6 by 500 mHz 1H NMR was very enlightening
(Table 2.3). First, the doublet which overlapped a doublet of doublets (2.55-

2.65 ppm) is much better resolved than in the 300 mHz spectrum (see
Appendix A for spectrum). Therefore, the peak positions and splitting

constants were much easier to determine. Next, the mutiplet (2.70-3.00
ppm) that integrated as ten protons in the 300 mHz spectrum resulted in four
sets of resonances in the 500 mHz spectrum. Using the same reasoning as
in assignment of resonances in 5, the resolution of another doublet of
doublets (2.67-2.70 ppm) and calculation of its splitting constants allowed

the assignment of both this doublet of doublets as well as the doublet of
doublets at 2.41-2.48 ppm. There are slight upfield shifts in position for
these resonances relative to 5, as expected due to subtle electronic

reorganizations (these protons are remote from the site of modification)
caused by the removal of the highly electronegative tosyl group.
In the 300 mHz 1H NMR spectrum, all three sets of methylene protons

adjacent to amines have very similar resonances. Only by the use of a high
field instrument were these resonances split into four constituents.
Therefore, assignment of the remaining two sets of protons required several

steps. First, the integration of the resonance sets had the following pattern:
four protons (2.72-2.78 ppm), two protons (2.67-2.70 ppm), 2 protons (2.54-

2.65 ppm), and two protons (2.51-2.57 ppm). Next, the two higher field
resonances were identical in appearance (complex doublet of doublet of

doublets) and have identical splitting constants. This pattern had been
observed several times for the two facial sets of protons for the triaza ring in
earlier compounds, which suggests that they may be assigned to the two
pairs of g protons. Finally, the resonances furthest downfield, integrating as
four protons, were examined. Due to the remote position that the a








protons have relative to the modification site, it would not be unusual for both
sets to have resonances very close to one another. The resolution of the f

protons into a clean quintet supports this argument. In this case they

overlap, causing a messy, unresolvable multiplet. Furthermore, the
resonance of the protons on the carbon a to the amines in the unmodified

ligand 1, are at 2.73-2.80 ppm. Therefore, the 2.72-2.78 ppm multiplet has
the closest resonance to the e protons of this closely related reference
compound.

Characterization of 8 by 500 MHz 1H NMR did not greatly improve
resonance assignments relative to the 300 MHz spectrum. The only
significant improvement was assignment of the h cis and trans, protons,
2.55-2.62 ppm and 2.85-2.65 ppm multiplets, respectively. These two
doublet of doublets were more clearly resolved than for the 300 MHz

spectrum.
NMR of Zinc Complexes
Zinc complexes were formed for three compounds, the final
bioconjugatable product 8, the unmodified ligand 1 (which forms the kinetic

studies model complex), and an intermediate product 6 (Table 2.4). These
complexes are designated Zn(OH)-8, Zn(OH)-I, and Zn(OH)-6, respectively.
In 1H NMR, all of these complexes exhibited a similar spectrum, reflecting

the loss of symmetry due to the zinc complexation. The inequivalent protons
result from the formation of facial sets for previously equivalent protons and

substantial splitting constant changes for the protons which were already
inequivalent or split into facial sets. These facial sets represent protons on

each side of the plane formed by the three amine nitrogens. The simpler the
molecule, the easier the interpretation of the splitting. Zn(OH)-1 was easily

compared to its parent ligand as the spectra were both relatively






44
uncomplicated and each set of inequivalent protons was resolved. Zn(OH)-

6, however, had a 1H NMR spectrum which was extremely complex as the

parent ligand had a complex spectrum featuring a large overlapping

multiplet region. As these protons were nearly equivalent, the splitting in
facial sets (or change in existing facial set splitting) resulted in complex
overlapping resonances. The g proton sets are no longer isolated and the
htrans protons cannot be recognized within the multiplet. Specific
assignments for each set of e, g, and h protons, possible for 6, cannot be

made for this complex.









TABLE 2.4
Zinc Complex 1H NMR Resonances


Compound

12-AneN3, a
MeOD



Zn(OH)-1Ia
D20, pD = 10


Peak(s),Integ

2.73-2.80
t, 12H
1.65-1.70,
qnt, 6H

3.16-3.25,
ddd, 6H
2.82- 2.93,
ddd, 6H
1.95- 2.10,
ttd, 3H
1.60-1.73,
ttd, 3H


Assignment


j(Hz)


j=5.7

j= 5.7


= 2.2, 7.9, 13.2

=2.2, 9.0, 13.2

=2.1, 7.9, 16.5

=2.1, 9.0, 16.5


Zn(OH)-6b
MeOD


Zn(OH)-8a
MeOD








appm from 300.15 MHz
bppm from 499.843 MHz


7.16-7.30,
m, 5H
2.70-2.94
m,1OH
2.50-2.63
d + m, 4H
1.90-2.10,
m,1H
1.56-1.70,
m,4H

7.75,d,2H
7.25,d,2H
2.65-2.90
m,1OH
2.45-2.65
m + d,4H
2.25-2.41
,m,1H
1.60-1.70,
m,4H


Ar

e, g, and brans

j, hcis

i

f


Ar
Ar
e, g, and htrans

j, hcis

i
f


j= 7.0


= 8.4
j= 8.4


j= 7.6









Resonances in the aromatic region are easier to interpret. Upon
complexation the resonance of the methylene f resonances is a multiplet,

compared to the clean quintet in the parent ligand. Both of these areas

appear to show the splitting, especially the f protons, into inequivalent sets
of protons. This splitting was more clearly observed for Zn(OH)-1. Finally,
the 1H NMR of Zn(OH)-8 had a very similar appearance to that of Zn(OH)-6
for the same underlying reasons. Splitting (very slight) of the doublet of
doublets to doublet of doublets of doublets in the aromatic region of Zn(OH)-

8 is likely due to zinc complexation rather than formation of a disulfide.
Disulfide formation was tested by exposure of an uncomplexed sample, 8, to
air and subsequently taking its' 1H NMR spectrum. It did not exhibit further

aromatic splitting.
130 NMR resonances for Zn(OH)-6 on the 125 MHz NMR mirrored the

loss of symmetry relative to the ligands that was observed in the 1H NMR

spectrum (Table 2.5). While there are resonances for each carbon, there is
also another complete set of resonances having much smaller intensities
next to them. This smaller set of resonances are particularly evident in the

aromatic region and support the theory that formation of another zinc
complex, perhaps a dimer, has occurred. Broadening of peaks noted in the
1 H NMR spectrum may also be due to the presence of two zinc complexes of
the ligand. Two zinc complexes, therefore, could result in two different sets
of spectra. Decreased symmetry of the complex relative to the ligand alone
was forwarded earlier as an explanation for the 1 H NMR spectrum.












Zinc Comple;


TABLE 2.5
(13C NMR


Aromatic

140.714
130.326
130.035
129.669
129.523
127.379


Resonances


Aliphatic

56.843
54.228
52.742
51.393
49.765
49.484
39.418
38.675
25.946
25.766


appm from 125.697 MHz


Compound

Zn(OH)-6a
MeOD







Summary and Conclusions


There are several possible approaches to synthesis of C-substituted
polyaza macrocycles.10,11 The approaches by Helps and coworkers 10 and
Richman-Atkins11 are two popular methods in the literature. The similarity of

the target compound to Helps and coworkers' was a major factor in our
choice of synthetic method. The Richman-Atkins method is a general

method for synthesis of polyaza cycles rather than a specific reaction for
synthesis of substituted [12]-N3 ring systems. As was observed in the
synthesis of the target compound, changes in polyaza macrocycle structure
and derivatization caused profound differences in their reactivity. Thus,
successful adaptions of literature methods were mainly observed in those

taken from triazamacrocycle syntheses.
In the synthesis of 3, the formation of noncyclized polymeric products

complicated the reaction workup as the product was entrained in an orange

polymeric tar. Low product yield was also a problem as the polymer was the
major product. High dilution and addition of reactants in aliquots at time
intervals over the two weeks were utilized to improve reaction yield.
Addition of the reagents in aliquots did not appear to affect product yield

significantly.A reaction with lower dilution of the reagents produced a lower
product yield, as expected.
The sulfonamide, 4, was extremely difficult to purify as it was very

tacky and rather insoluble in most solvents. The reduction of the diamide to
form 5 was accomplished by using BH3-THF.10 Initially, cleavage of the tosyl

group, forming 6, was attempted by using the HBr/Acetic acid method
proposed by Helps and coworkers.10 These conditions caused reagent

decomposition. Further attempts at acid hydrolysis by using H2S04 also








produced decomposition.11 Reductive cleavage by using the Birch

reduction produced a small percentage of product. The major product,
however, resulted from the reduction of the aromatic. Finally, a 2%Na/Hg

amalgam was tried in the hopes of improving the yield. This method worked
without reducing the aromatic group of the starting material.
Chlorosulfonation of 6 to produce 7 was accomplished by adaption of

literature preparations for the production of aromatic sufonyl chlorides,

specifically p-toluenesulfonyl chloride.17 Observation of two doublets in the
aromatic region and loss of the multiplet observed for the aromatic in the
starting material was a clear indication of product formation. Reduction of

this product to the thiol by LiAIH4, a traditional reagent for the reduction of
sulfur species, produced 8.17 Due to the small quantity of 8 synthesized, the
Zn(OH)-_complex was formed in situ for bioconjugative use. Precipitation

and recrystallization would have been very difficult at this scale.
Characterization of all the compounds and complexes was primarily

accomplished by using NMR. Difficulties included low solubilities for the
earlier products, extremely hygroscopic properties of the polyamines, and
overlapping resonances for the later products. Low solubilities made NMR
spectra difficult to obtain, especially 13C NMR. This problem was
compensated for by using high acquisition numbers. Hygroscopic

compounds were primarily soluble in deuterated methanol. Thus, the amine
protons exchanged with the acidic proton in the solvent and they were
unobservable. This exchange also caused the methanol peak to be large,
occasionally obscuring spectral characteristics. Therefore, extensive
dehydrating efforts were necessary in order to get an acceptable spectrum.
This same property made it difficult to obtain acceptable elemental analyses

for the polyamines. Overlapping peaks in the 1 H NMR spectra were








analyzed by using a higher field instrument in order to resolve separate

resonances.


Experimental


General Methods
1H/13C NMR spectra were obtained on a GE QE-300 spectrometer at

300.15 and 78.48 MHZ respectively. FT-IR spectra were obtained on a
Perkin-Elmer 1600. Mass Spectra were obtained by Dr. George Alameddin

on an lonSpec Omega/386 ICR Mass Spectrometer with a 3 tesla magnet.
Elemental analyses were determined at the University of Florida

microanalysis center. All reagents and solvents used were of analytical
grade. Specialty chemicals used included: NaH (Aldrich), 1M BH3-THF

(Aldrich), diethylmalonate (Aldrich), diethylbenzylmalonate (Lancaster),

1,5,9-triaazanonane (sold as 3,5'-iminobispropylamine from Aldrich), 2%
Na/Hg amalgam (Lancaster), chlorosulfonic acid (Fisher), LiAIH4 (Aldrich),

and Red-Al (Aldrich). Unless otherwise noted, dry solvents were HPLC
grade solvents which were freshly-distilled over appropriate drying agents.
Care was taken in this study to minimize concentrations of extraneous metal
ions (which might complex the free ligand) in the bioconjugateable complex

synthesis. All water used (including rinse water for glassware) was purified
in a Barnstead nanopure water system and had a resistance of :=17

megaohms.
Synthesis of Diethyl benzylmalonate, 2
The synthesis of 2 utilized a modification of the method reported by
Fonken and Johnson12. To a 3-necked round-bottomed flask outfitted with a
magnetic stirring bar, a reflux condenser, and sealed with a septum, 2.54 g








(0.11 mol) of NaH was added. The flask was placed in a 65C water bath.

Slowly, 75 ml of dry ethyl alcohol was added by syringe. Steady bubbling
occurred during the addition and the subsequent stirring (5 minutes). Then
26.34 g (0.16 mol) of di-ethyl malonate was added in the same manner over

5 minutes, at which time vigorous bubbling commenced. After stirring for

several minutes the mixture transformed from a gray sludge to an amber
liquid. Then 20.89 g (0.16 mol) of benzyl chloride was added slowly over 5
minutes. The solution became opaque and creme colored. It was allowed

to stir at 650C for 1.5 hrs. The resultant mixture was then allowed to cool

under nitrogen.
Once the solution cooled, 100 ml of water were added. The solution
was transferred to a separatory funnel, and the organic layer was removed.
The remaining aqueous layer was extracted three times with diethyl ether

and the organic layers were combined and added to the original organic
layer. At this point, K2CO3 was added to the organic layer and the mixture

was allowed to sit overnight. The diethyl ether was removed by rotary

evaporation. Distillation under reduced pressure (10 mm), yielded 5.25g,
at 120-130C (19% yield). This fraction was characterized by 1H NMR
(300.15 mHz, CDCL3) d 7.25 (m, 5H), 4.16 (d of q, 4H), 3.67 (t, 1H), 3.24 (d,

2H), 1.22(t, 6H) and 13C NMR (78.48 mHz, CDCI3) d 168.79 (ester),
137.87, 128.81, 128.45, 126.68 (benzyl), 61.38, 53.82, 34.67, 13.97.
Analysis: calculated for C14H1804: C, 67.17; H, 7.25; found: C, 67.30; H,

7.37.
Synthesis of 3-Benzyl-1,5,9-Triazacyclododecane-2,4-Dione, 3.
Synthesis of 3 was accomplished by using an adaption of a method
proposed by Helps and coworkers.10 A solution of 41.25g (300 mmol, 100-
fold excess) 1,5,9-triazanonane in 5000 ml ethanol was prepared and left







magnetically stirring under nitrogen. Then 75.00 g (300 mmol) of 2 was
slowly added by syringe. The solution was heated to reflux. The reaction

mixture turned light yellow-orange after a day. Under these conditions the
reaction solution was allowed to reflux for two weeks. Aliquots taken at

various intervals were analyzed by using 1H NMR, which indicated a
complex mixture. The solution was then allowed to cool to room
temperature. The ethanol was almost entirely removed by rotary
evaporation and the resulting orange syrup was treated with 40 mL acetone
and placed in the cold room (50C). After several hours, copious amounts of
white ppt (11.29 g, 13% yield) came out of solution. This precipitate was
removed by filtration and rinsed with cold ethanol. It was then recrystallized
in ethanol. Fine white crystals resulted, 8.24g (10% yield), which were
characterized by elemental analysis and NMR. 1H NMR (300.15 mHz,

0D2012) d 7.61 (s, 2H), 7.22 (m, 5H), 3.57 (tdd, 2H), 3.18 (m, 5H), 2.86
(ddd, 2H), 2.62 (ddd, 2H), 1.72 (m, 2H) 1.54 (m, 2H). 1.15 (s, 1H). 130 NMR
(78.48 mHz, CDC13) d 168.80 (amide), 149.00, 129.65, 129.00, 125.50

(benzyl), 56.48,49.00,40.05, 33.10, 27.10. Analysis: calculated for

C16H23N302: C, 66.39; H, 8.02; N, 14.53; found: C, 66.38; H, 8.10; N, 14.57.

Synthesis of 3-Benzyl-9-(p-Tolylsulfonyl)-1,5,9-Triazacyclododecane-2,4-
Dione.4
First, 3.6 g (12.4 mol) 3-benzyl-1,5,9-triazacyclododecane-2,4-dione
was added to 300 mL CHC13. stirring magnetically under an N2 atmosphere.
2.34g of triethylamine was then added over 5 min. Then 3.24 g of toluene
sulfonyl chloride was added and the solution brought to reflux. The solution
was allowed to remain at reflux for 24 h. After 2 h it was observed that a

heavy white microcrystalline product was precipitating out of solution. After
24 h, copious ppt was observed in the bottom of the flask. The crude ppt








(9.90 g) was filtered by using a fine glass frit and rinsed thoroughly with cold
ethanol. The crude precipitate was recrystalized in hot methanol to produce
5.2 g (22.3 mmol) product (94% yield). 1H NMR (300.15 mHz, CDCL3) d
7.68 (d, 2H), 7.35 (d, 2H), 7.23(m,5H), 6.50 (s, 2H), 3.45-3.56(m,2H),3.27-

3.45(t + m, 3H),2.90-3.20 (d + m, 6H), 2.40 (s, 3H), 1.80(m,2H), 1.60 (m, 2H),
13C NMR (78.48 mHz, CDCI3) d 169.25 (amide), 143.80, 140.09,138.49,

130.21, 129.07, 128.57, 127.15, 126.38 (benzyl), 57.00, 42.51, 35.98, 32.77,
27.20, 21.37. Analysis: calculated for C23H29N304S: C, 62.28; H, 6.59; N,

9.48; found: C, 62.18; H, 6.47; N, 9.31.
Synthesis of 3-Benzyl-9-(p-Tolylsulfonyl)-l,5,9-Triazacyclododecane, 5
A slurry was produced by addition of 3.78g of 3-benzyl-9-(p-
tolylsulfonyl)-1,5,9-triazacyclododecane-2,4-dione (8.56 mmol) to 50 mL dry
THF. This slurry was added to a round-bottomed flask fitted with a graham

condenser under a brisk N2 flow. Then 120 mL of 1M BH3-THF was added
by syringe over 10 minutes. The reaction was heated to reflux and allowed
to react for 3 days. The solution was allowed to cool to room temperature

and was then quenched with cold methanol. The solution was then taken to

a white, foamy residue by rotoevaporation. This residue was then dissolved
in KOH solution (pH>13) and extracted with methylene chloride three times.

The combined organic extracts were reduced by rotary evaporation to a

bright white foamy residue which subsided to a light yellow oil (3.2g, 90 %

yield). 1H NMR (499.943 mHz, MeOD) d 7.70 (d, 2H), 7.30 (d, 2H), 7.25
(d,2H) 7.23 (m, 3H), 3.08-3.15 (ddd, 2H), 2.85-2.94 (dd + m, 6H), 2.80-2.85
(dd,2H), 2.50-2.60 (dd + m,4H), 2.45 (d, 2H), 2.40 (s, 3H), 1.94 (m, 1 H), 1.64
(qnt,4H), 13C NMR (78.48 mHz, CDCI3) 142.98, 140.97, 137.28, 129.24,
128.84, 128.21, 127.14, 125.77 (benzyl), 49.76, 45.21, 44.21, 40.15, 37.84,








26.11, 21.20, Analysis: calculated for C23H33N302S: C, 66.47; H, 8.01; N,
10.12; found: C, 66.38; H, 8.15; N, 9.72.

Synthesis of 3-Benzyl-1,5,9-Triazacyclododecane, 6.
A solution 0.30g 3-benzyl-9-(p-tolysulfonyl)-1,5,9-
triazocyclododecane (0.72 mmol) in 10 mL freshly distilled MeOH was
prepared. Then 0.4g Na2HPO4 and 6 g finely divided 2% Na/Hg amalgam

was added under a brisk N2 flow.. The solution turned a murky brownish-

pink for 1 h. The solution became colorless with suspended white
precipitate thereafter. After 48 hrs, 20 mL H20 was added and the solution
decanted off from the metallic Hg. The solution was then extracted with
CHCI3. The CHCI3 fraction was then evaporated to a light yellow oil. The

1H NMR of the oil indicated the presence of product. The oil was then
dissolved in CHCI3 and run over a 60 A silica column. The eluant order was
CHCL3, 5% MeOH/95% CHCI3, 20% MeOH/80% CHCI3, 35MeOH/64%
CHCI3/1% NH3, and 48% MeOH/48% CHCI3/4% NH3. Two bands came off

close together with the final eluant. The first band was starting material and
the second band, 0.1 g (0.29 mmol) colorless oil, had a 1 H NMR indicating
product (40 % yield). 1H NMR (499.943 mHz, MeOD) d 7.23 (m, 5H), 2.72-
2.78 (m, 4H), 2.67-2.70 (dd,2H), 2.59-2.65 (ddd,2H), 2.51-2.57 (ddd,2H),
2.41-2.48 (dd +d, 4H), 1.94 (m, 1H), 1.64 (qnt,4H), 13C NMR (78.48 mHz,
MeOD) d 139.28, 128.63, 128.10, 125.95 (benzyl), 52.73, 48.24, 47.94,

37.83, 37.26, 25.39. Mass Spectrum calculated for C16H27N3 (C I) m/z (M +
1) 262.228, m/z (M + 1) 262.220 found.
Synthesis of 3-(p-Benzylsulfonyl Chloride)-1,5,9-Triazacyclododecane, 7.
A solution was prepared from 50 mg (0.14 mmol) 3-benzyl-1,5,9-
triazacyclododecane in dry (sieves only) CHCI3. Then 0.43 g chlorosulfonic
acid was added to a 50 mL round-bottomed flask under N2 atmosphere and







sitting in an ice bath. The CHC13 solution was added slowly with brisk
magnetic stirring over 15 minutes. A vigorous reaction leading to the

formation of an orange lower acid layer and an colorless upper CHCI3 layer.
The reaction was left in the ice bath for 2.5 hrs and the temperature ranged

from 0-120C. The reaction was then quenched by addition to a small beaker
full of ice. Again, the reaction was vigorous. The entire contents were then
reduced to an light yellow-orange oil by rotoevaporation which had a 1H
NMR indicating relatively clean product. Chromatography was then
performed on 60 A silica gel. The eluant order was CHCL3, 5% MeOH/95%
CHCI3, 20% MeOH/80% CHCI3, 35%MeOH/64% CHCI3/1% NH3,

48MeOH/48% CHCI3/4% NH3, and 45% MeOH/45% CHC13/10% NH3.

Subsequent concentration of the fraction, 15 mg, from the final eluant
produced an 1H NMR indicating the pure product (0.068 mmol, 49% yield).
1H NMR (300.15 mHz, MeOD) d 7.75 (d, 2H), 7.30 (d, 2H)3.20-2.72 (m,
1OH), 2.54-2.70 (d + m, 4H), 2.27(m, 1H), 1.90(qt,4H,), 13C NMR (78.48
mHz, MeOD) d 143.19, 141.82, 128.62, 125.78 (benzyl), 52.41, 47.68,

47.36, 37.33, 36.80, 24.82.
Synthesis of 3-(p-Thiobenzyl)-l,5,9-Triazacyclododecane, 8.
First, 30 mL freshly distilled Et20 was added to a 50 mL round-
bottomed flask equipped with a Graham condenser and containing 30 mg
(0.068 mmol) of 3-(4-benzylsulfonylchloride)-1,5,9-triazacyclododecane

under N2. Then the solution was heated to reflux and a solution containing
3 mg LiAIH4 in 7 mL in freshly distilled Et20 was added over 30 minutes
through the septum top on the graham condenser. The reaction was stirred
at reflux overnight and quenched with wet, degassed MeOH the next day.
The solution was taken to a residue, 15 mg, and a 1H NMR indicated product

(0.040 mmol, 59% yield). 1H NMR (300.15 mHz, MeOD) d 7.75 (d, 2H),







7.25(d, 2H), 2.85-2.65(m, 1OH), 2.55-2.62 ( d + m, 4H), 2.10 (m, 1H), 1.67(qt,
4H), 13C NMR (78.48 mHz, MeOD) d 141.80, 130.93, 128.48, 125.78,

52.65, 48.23, 47.97, 37.79, 36.89, 25.37.
Synthesis of (Zn-1-1,5,9-Triazacyclododecane(OH%(01.43 H0IO4 (Zn(OH)-
1)

Adapting a method of Kimura et a[.,1 an equal volume of 0.88 mM
ethanolic solution of Zn(CIO4)26H20 was added to a 0.88 mM ethanolic

solution of 1,5,9-triazacyclododecane (1). Immediate precipitation of an off-
white powder was observed. The mixture was allowed to stir about 5
minutes and collected by filtration with a glass frit. The resultant powder was

twice recrystallized from nanopure water, resulting in colorless crystals. An
FT-IR spectrum (KBr pellet) agreed with the literature spectrum. Anal. Calcd
for C27H66N9O3Zn3(CIO4)3HC104 : C, 27.96; H, 5.82; N, 10.87. Found: C,

28.51; H, 5.43; N, 10.63.

Synthesis of Zn(OH)-6 and Zn(OH)-8.
Adapting a method of Kimura et al.,1 an equal volume of degassed
0.12 mM deuterated methanol solution of Zn(CI04)26H20 was added to a

degassed 0.12 mM deuterated methanol solution of 6 or 8. The resultant
solution was characterized by 1H and 13C NMR spectra (Tables 2.4 and
2.5).












CHAPTER 3


KINETIC METHODS AND
CHARACTERIZATION OF SOLUTION SPECIES


Introduction


Determination of catalytic rate constants requires extensive

knowledge of the experimental kinetic system. For each reaction an
appropriate monitoring method must be determined and its utility assessed.
Reproducibility of catalytic rate constants between kinetic runs and

acceptable precision must be achieved. Variations in reaction conditions,
such as pH, temperature or ionic strength, that are assumed to remain
constant must be accounted for and corrected. Finally, the concentration of
any solution species which affects the catalytic rate constant calculations,
such as p-nitrophenol and Zn(OH2)-1 in this study, must be measured and

corrections for their presence must be included in the calculations.
Therefore, the acid dissociation constants for Zn(OH2)-1 and p-nitrophenol
must be determined at all temperatures used in the study. The AH and ASO

for acid dissociation of p-nitrophenol and Zn(OH2)-1 were calculated from a
van't Hoff plot of their temperature dependence. For the p-nitrophenyl
acetate hydrolyses in HEPES buffer, the pH was measured as the
temperature was varied from 20-400C for 50 mM HEPES buffer. Synthesis
and purification of the catalyst and reactants were necessary. Water
purification and reaction setup cleansing are also described.








Hydrolyses of p-nitrophenyl acetate (p-NPA), and other para-
substituted phenyl acetates which release the corresponding phenol, are

often followed spectrophotometrically. Depending upon the phenyl acetate,
either the phenol or the phenolate may be the product species that is
monitored. At pH 8.2, p-NPA hydrolysis was monitored by using the 404 nm

p-nitrophenolate (p-NP-) absorbance (Equation 3-1).

02N 02N
+ 0
Zn(OH)-1 + H 1



c=U
H (3-1)
CH3 Ka

p-NITROPHENYLACETATE
(p-NPA)

02N

+ H+

b-




Many investigators have used spectrophotometric detection of products for
monitoring ester hydrolyses, including Jencks, Bruice, and Benkovic.22-24
Due to the vast library of data for esters which these investigators produced,
the reaction conditions for these studies were patterned after reaction
conditions in the earlier studies.
Few carboxylic esters with poor leaving groups, such as the alkyl
esters, produce products with UV or visible absorption bands in a








Few carboxylic esters with poor leaving groups, such as the alkyl

esters, produce products with UV or visible absorption bands in a

convenient wavelength range. Therefore, a common approach to study
reaction rates of these esters involves neutralization by addition of a

standardized base to determine the amount of carboxylic acid produced at
specific time intervals (Equation 3-2).

0 Zn(OH)- 0
II %H (3-2)
H3C-C-O-CH3 H3CCOH + CH3OH

METHYL ACETATE Ka




0
II + H+
H3C-C- +




Historically, aliquots of the reaction solution were removed at many intervals,
the reaction quenched, and the acid titrated with a standardized base.25,26
Problems with this approach include potential contamination of the reaction
during the procedure. Quenching the reaction aliquot quickly and
thoroughly was also necessary. Potential sources of error due to titration

delays may include dissolved carbon dioxide or potential product
decomposition or further reaction in the quenching solution. Therefore,
recent development of inexpensive pH stat controllers allowed sensitive,








Catalyses of carboxylic acid ester hydrolyses have been shown to be
consistent with second order reaction kinetics -- first order in both catalyst

and ester concentration.25,27 Examination of the mechanisms in equations
3-1 and 3-2, therefore, results in the following expression for the observed
reaction rate (Equation 3-3).


Vobs = k[ester][catalyst] (3-3)



If data is only taken during times before significant production of product and
if the esters' concentration is large relative to that of the catalyst, then the

esters' concentration remains constant during this period. As one of the
reactant concentrations is now a constant, the reaction is pseudo-first order.

The observed reaction rate for catalyzed hydrolyses of carboxylic

acid esters has at least four components, catalysis due to hydroxide ion,
protons, water, and Zn(OH )-1 (Equation 3-4).


Vobs = koH[OH'-][S]+ kH[H+][S]+ kH20[S]+ kzn(OH)-![Zn(OH)-j[S] (3-4)


Of these, only hydroxide and Zn(OH)-1 catalyze a significant percentage of
the observed reaction rate. In order to determine the contribution of the

hydroxide ion catalysis to tfle overall observed rate, the background reaction
rate at each temperature is measured and the koH- is calculated. As the pH,

the ester concentration, and the koH value are known, a background
reaction rate is generated for catalyzed reactions. The background value is

subtracted from the observed reaction rate, resulting in the reaction rate due
solely to Zn(OH)-1. The kZn(OH)-I is then calculated.










Experimental


General Methods
UV-visible spectra were recorded on a Nicolet 9420

spectrophotometer. 1H and 13C NMR spectra were obtained on a GE QE-

300 spectrometer at 300.15 and 78.18 MHZ respectively. Quantitative
elemental analyses were determined at the University of Florida
microanalysis center. A Haake-F3 thermostatted circulator bath was used in

cases where recirculated temperature control was necessary. All reagents
and solvents used were of analytical grade except acetonitrile and NaOH,
which were spectrochemical grade, and the HCl standardized solution,
which was an Acculite solution.
Care was taken in this study to minimize concentrations of extraneous

metal ions (which might catalyze ester hydrolysis) in the catalyst synthesis
and kinetic studies. All water used (including rinse water for glassware) was
purified in a Barnstead nanopure water system and had a resistance of >17

megaohms. In kinetic studies, Barnstead water was further filtered through
Sigma's C-7901 chelating resin to remove all adventitious metal ions. All
glassware used for stock solutions, as well as the pH-stat tubing of the flow
cell pH-stat setup, were washed with Na2EDTA and then rinsed thoroughly

with this water. Stock solutions were transferred to new, plastic bottles and

stored at 50C. Synthesis and purification of (Znll-1,5,9-
triazacyclododecane(OH))3(C0O4)3 HCIO4, Zn(OH)-I was described in

Chapter 2.
Purification of p-nitrophenol








The p-nitrophenol was obtained at 99+% purity from Aldrich. It was

then sublimed under 4 mm Hg for 2h at 1000C, producing light yellow
crystals. The freshly sublimed crystals were crushed, dried in a vacuum

oven at low heat for 2h, and stored under N2 in a desiccator.
pH-stat Controlled System, Flow Cell.
The rate of p-NPA hydrolysis was measured by using a thermostatted
pH stat controlled system. The increase in absorbance at 404 nm, due to the
release of the nitrophenolate product, was monitored. Unless otherwise

stated, a final concentration of 1.00 mM in ester and 0.600 mM in Zn(OH)-I
was used.
A schematic of the experimental apparatus is shown in Figure 3.1.
The temperature was maintained by using a thermostatted UV-visible cell
and reaction vessel. The reactions occurred in aqueous solution maintained

at the desired pH (t 0.1 ) with an Omega PHCN-36 pH controller coupled
with a Ross model 8103 combination pH electrode. Base delivery was

accomplished by using a Buchler multistatic (peristaltic) pump equipped with
silicone tubing. It was used as a slave to the controller, switching on when
the low pH set point was reached and switching off as the high set point was

exceeded. The temperature was maintained ( 0.010C) by a Haake-F3
thermostatted circulator bath. Any tubing which entered the solution used
drawn quartz tips. Reaction volume was 12.0 ml and circulation volume was

3.2 ml. A 24.40 mM NaOH solution was used as the titrant, which was
standardized with a 0.100 N HCl Acculite solution. The reactions took place
in 50 mM HEPES buffer and the ionic strength was 0.10 (maintained with
NaCIO4). The catalyst was dissolved in 10% CH3CN/buffer and then added

the reaction cell. In a separate vial, the ester was dissolved in 0.020 ml
acetonitrile. The ester was then added to the circulating solution, including








the catalyst, which was already at the correct temperature and pH. An
average mixing/transport delay was about 45 seconds. The reaction was

followed for about 10 minutes, or until UV-visible absorbance at 404.0 nm
exceeded 1. The second-order rate constant due to catalyst alone was

determined from the net initial slope resulting from subtraction of the
hydroxide-catalyzed ester hydrolysis contribution (determined in separate

experiments) to the observed rate. Corrections were applied to account for

the presence of significant percentages of nitrophenol and Zn(OH2)-I in
solution (see Apendix B for sample calculations).
Irreproducible catalysis rate constant values and observation of
nitrophenolate production during system rinses (rapid development of a

yellow color in the rinse water) between kinetic runs indicated ester
adsorption. The reaction vessel, pH electrode, and flow cell were examined
and determined to be nonadsorbing. Subsequent circulation of a reaction

solution through the system resulted in 10% adsorption of the ester on the
tubing walls. Further studies resulted in significant adsorption of ester on

silicone, tygon, and polyethylene tubing.

pH-stat Controlled System, Modified UV-visible Cell.
The rate of p-NPA hydrolysis was measured by using a modified
thermostatted pH stat controlled system. Again, the increase in absorbance
at 404 nm, due to the release of the nitrophenolate product, was monitored.

Unless otherwise stated, a final concentration of 1.00 mM in ester and 0.600
mM in Zn(OH)-I was used. The same pH stat controller, base delivery
system, and water bath was used as for the flow cell system. However, a
specialized UV-visible cell was fabricated to eliminate the need for an
external reaction vessel. The temperature was controlled to 0.01 C as the
UV-visible cell was double-walled and the cell holder was thermostatted.








Space limitations of the system required a small, portable solution stirring
device. The portable stirrer was based upon a cheap, commercially

available, battery operated vitamin/diet drink mixer. A 13 mm diameter
propeller-shaped blade was machined from a teflon block. The propeller

was positioned as to maximize solution circulation, which was tested by
using a drop of red food coloring dye.
The reactions were studied in aqueous solution buffered with 50 mM

HEPES buffer at 25.0 0.10C and an ionic strength of 0.10 (maintained with
NaC1O4). Initially the catalyst was dissolved in 10% CH3CN/buffer. In a

separate vial, the ester was dissolved in 0.020 ml acetonitrile. The ester

solution was added quickly to the catalyst solution, mixed thoroughly, and
transferred to the modified UV-visible cell. An average mixing/transport
delay was about 45 seconds. The reaction was followed for about 10

minutes, or until UV-visible absorbance at 404.0 nm exceeded 1. The rate
constants were calculated in the same manner as detailed in the flow cell

system description.
pH-stat Controlled System, Base Addition Method.
The experimental setup included the exterior vessel, peristaltic pump,

and water bath from the flow cell system. Due to the critical nature of exact
pH control for these reactions relative to the spectrophotometric
experiments, the pH stat controller was modified to produce a 0.03 pH

range rather than 0.1. Modification allowed the controllor to sense more
subtle changes in pH. Therefore, it started pumping very close to the low set
mark and ceased at the high set point. Overshooting the high set point was

a common problem in the other systems.
The reactions occurred in aqueous solution maintained at the desired
pH (+ 0.03) with an Omega PHCN-36 pH controller coupled with a Ross








model 8103 combination pH electrode. Ionic strength was maintained at 1.0

by KCI. Reactant concentrations varied as the esters varied. As with the
phenyl esters, the base concentration used varied with the reaction rate,
higher concentrations being used for the faster reactions. Constant pH was

accomplished by base addition. A measuring pipet filled with base was
attached to a Buchler multistatic (peristaltic) pump equipped with silicone

tubing. The base was added as needed to maintain constant pH. The

peristaltic pump was a slave to the pH stat controller, switching on when the
low pH set point was reached and switching off as the high set point was
exceeded. Any tubing which entered the solution used drawn quartz tips.
The fine quartz tips improved base delivery sensitivity and were far easier to

clean than the tubing.
Extinction Coefficient for p-NP-

A nitrophenolate solution was prepared with [NPO-] = 5.075 x 10-3
and [t= 1.0 M (KCI as the ionic strength source) and its absorbance at 404

nm for each temperature determined. The extinction coefficient was then
determined by using Beer's Law.
p-NPOH Titration
The modified UV-visible cell, pH electrode and spectrophotometer
from the p-NPA kinetic setup was used. The temperature probe from the pH
meter was calibrated. The pH electrode was calibrated with pH 7.00 and

10.00 buffer. The constant temperature water bath was then started and its
temperature set. Then 1.00 mL p-NP stock solution (0.025 mmol/mL in dry,

spectral grade acetone) was added to the modified UV-visible cell. Then
0.500 mL 2.00 M KCI solution was then added (for an final It = 0.10). Finally,

8.500 mL chelex-treated nanopure water was added. The final solution was
then degassed by bubbling wet Ar through the solution for 20-30 minutes. At








this point the pH was checked and adjusted to around 5 with 0.028 mL

0.100 M HCI and the temperature adjusted to 200C. The 0.0995 M base
was then added in a 0.020 mL increment by using an E2-1000 motorized

Rainin auto-pipettor and the pH taken. The temperature was then increased,
50C at a time, and the pH taken at each step. Another increment of base
was then added, the temperature increased, and the pH again taken at each
temperature. This procedure was repeated throughout the titration. The
temperature probe and pH electrode were checked several times to insure

that their calibration was not drifting.
Zn(OH2)-1 Titration
The temperature probe from the pH meter was calibrated. The pH
electrode was calibrated with pH 7.00 and 10.00 buffer. The constant
temperature water bath was then started and its temperature set. First,

0.0039 g Zn(OH)-1 (for an initial concentration of .0010 M) was placed in the

modified UV-visible cell. Then 0.500 mL 2.00 M KCI solution was then
added (for an final u = 0.10). Finally, 8.500 mL chelex-treated nanopure

water was added. The final solution was then degassed by bubbling wet Ar

through the solution for 20-30 minutes. At this point the pH was checked
and adjusted to 3.90 with 0.260 mL 0.10 M HCI, the pH brought up to 5.16
with 0.028 mL of the 0.0969 M KOH, and the temperature adjusted to 200C.
Another standardized base, 0.00916 M, was then added in a 50 uL
increment by using an E2-1000 motorized Rainin auto-pipettor and the pH

taken. The temperature was then increased, 50C at a time, and the pH taken
at each step. Another increment of base was then added, the temperature
increased, and the pH again taken at each temperature. This procedure
was repeated throughout the titration. When the pH reached 7, the 0.0969 M
KOH was then used in 0.010 mL increments. The temperature probe and pH








electrode were checked several times to insure that their calibration was not

drifting.




Results and Discussion


Flow Cell pH Stat Controlled System
Initially, a pH-stat controlled system (Figure 3.1) that featured an

external reaction vessel, peristaltic pumping of reaction solution, and a UV-
visible flow cell was investigated. Flow cell methods allow the use of larger

solution volumes than can typically be used with a UV-visible
spectrophotometer as the majority of the solution volume is in the exterior

vessel rather than the UV-visible flow cell. As shown in Figure 3.1, crowding
would be a problem if a regular UV-visible cell was used. Solution mixing
was accomplished by using a Teflon magnetic stirring bar and a magnetic

stirrer. A double-walled quartz reaction vessel and a thermostatted UV-
visible cell holder insured constant temperature throughout the setup. The
temperature was tested at several points along the setup with a

thermometer. Water from a recirculating constant temperature water bath
was used for the thermostatted areas. The tubing carrying the reaction

solution was wrapped with foam to prevent heat loss. Maintenance of
constant reaction solution pH required a pH stat controller. The controller

activated a finely adjustable peristaltic pump attached to a pipet filled with
base. The tubing for the base delivery was tipped in finely drawn quartz tips
to increase the sensitivity of base delivery. Using fine tips reduced
overshooting of the high set mark on the pH state controller.









KOH


water block
UV-VISIBLE SPECTROPHOTOMETER


Figure 3.1
Flow Cell Kinetic Setup








Drawbacks of this system include sluggish pH adjustments due to

reaction volume circulation and sporadic temperature fluctuations due to the

use of an external reaction vessel. During replicate runs at one temperature,
it became obvious that the reaction rates were also steadily decreasing.

Speculation that ester was adsorbing in the system was confirmed when
adsorption tests on the tubing used yielded 10% adsorption of 10 mM

ester on the circulating tubing. Tests on silicone, tygon, and polyethylene

tubing all resulted in ester adsorption. This factor, as well as unsuitability of
other tubing types for the solvent system and peristaltic pumps, led to
modification of the reaction apparatus as described in the next section.

Modified pH Stat Controlled System
A modified pH stat controlled system (Figure 3.2) was then

constructed by using a fabricated UV-visible cell with a double-walled
solution reservoir (Figure 3.3) rather than a flow cell/exterior vessel
recirculation method. Due to the confined area around the reaction cell, this
method necessitated a thorough, metal-free, portable stirring source. After

several failed attempts, a portable stirrer was developed. It was based upon
a cheap, commercially available, battery operated vitamin/diet drink mixer.

A new propeller-shaped blade was machined from a teflon block. The

teflon blade proved an efficient stirring device and was free of metallic or
organic contamination. Stirring efficacy of the blade was tested by addition
of red dye to the solution in the stirred cell. The rest of the setup was the
same as in the flow cell system. The temperature dependence of pH of
HEPES buffer was measured at the same temperature, ionic strength, and
buffer strength used in the rate constant temperature dependence study
(Table 3.1). The reconfigured system allowed measurement of catalytic rate

constants when pH variation and substrate adhesion, which limited the flow






70
cell system, were not factors. Therefore, it was considered a good design for
measurement of catalytic rate constants for p-NPA and other carboxylic

acids with spectrophotometrically active leaving groups. The temperature

dependence study for p-NPA hydrolysis also used this system.



















































Figure 3.2
Modified UV-visible Cell Kinetic Setup




















Stirrer
Titrant
pH electrode-
Temperature probe


-H20


51,J-


Figure 3.3
Modified UV-visible Cell










Table 3.1

Variation of HEPES Buffer pH With Temperature


[OH]


14.1669 1.03 x 10-6


13.9965 1.36 x 10-6


13.8330 1.65 x 10-6


13.6801 1.99 x 10-6


13.5348 2.37 x 10-6


a 50 mM HEPES buffer in water treated with


metal chelating resin; p = 0.1


(sodium perchlorate); 0.010C. b CRC Handbook of Chemistry and Physics
ed. Robert C. Weast, CRC Press: Boca Raton, 1984; p D-168.


T (C)a


8.18


8.13


8.05


7.98


7.91


pOH


5.99


5.87


5.78


5.70


5.63









The Zn(OH)-I catalytic rate constants obtained by using this system
were reproducible (-5-10%). The original p-NPA temperature dependent

hydrolyses were done in a HEPES buffer solution. Prelimary kinetic runs at
25C produced values for koH- of 9.45 0.10 M-1s-1 at pH 8.2, comparing
very well with the Jencks et al. value of 9.5 M1s"1 .28 The kzn(OH)_l value

was 0.058 0.001 M-1s-1 by using the same conditions, which is slightly
higher than the 0.041 M1s-1 value reported by Kimura.1 The other phenyl

acetates used the same apparatus and conditions except that 1.0 M KCI was
used to maintain ionic strength and HEPES buffer was eliminated

altogether. These changes were implemented in order to more closely
resemble conditions used by investigators of organic catalysts, such as
Jencks, Bruice, and Benkovic22-24 A later p-NPA temperature dependence
conducted by Jeni Shepardson (unpublished work, Chemistry Department,

University of Florida, Gainesville, FL, Kinetic Studies.), another investigator
in the lab, also used the modified conditions as well as the modified pH stat
controller used in the methyl acetate hydrolyses. The koH- value for p-NPA

hydrolysis in this later study was 7.49 0.34 M-s1 at 250C and pH 8.2. The
calculated value for kzn(OH).l at pH 8.2 was 0.0116 0.0016 M-1s-1. These

values were obtained by using [ester] = 1.00 mM with 0.0100 M KOH as

titrant.
Modification of the system eliminated the external reaction vessel
improved the pH variance, temperature fluctuation, and ester adsorption
problems but did not eliminate the potential buffer interaction. A later

temperature dependent p-NPA hydrolysis study by Jeni Shepardson
(unpublished work, Chemistry Department, University of Florida, Gainesville,
FL, Kinetic Studies.) which eliminated HEPES buffer and replaced it with








KCI produced similar background reaction rate constants and lower Zn(OH)-
1 reaction rate constants. Evaluation of these values is further complicated
by the anion association constant values for Cl- published by Kimura.29 C-

exchange for the complexed hydroxide would dominate the solution species

if this is accurate. Certainly, the reduction of the catalyzed reaction rate

constant is consistent with his observations. He supports his findings by

using anion association constant values derived from both potentiometric
and inhibition study sources. All of his studies were done in HEPES buffer
solution. Therefore, the potential interaction of the buffer has not been
addressed. The possibility of an interaction is further supported by an
increase in reaction rate noted for the methyl acetate catalyzed hydrolyses.
These were also done by using KCI as the ionic strength source, yet they

display an increase in reaction rate constants rather than the decrease
expected for constants uncorrected for C- exchange. HEPES buffer was

absent in the methyl acetate studies.
Base Addition pH Stat Controlled System

Carboxylic ester hydrolyses which do not produce
spectrophotometrically active species can be folllowed by the consumption
of the base required by the pH stat controller to maintain constant pH As

one acid functionality is produced per hydrolysis of one carboxylic acid
ester, there is a one to one stoichiometry between the KOH consumed to
maintain constant pH and the acid produced. Therefore, the rate of
formation of product may be calculated by the amount of base consumed.

A technique which allows the monitoring of base consumption is
simple to design yet difficult to perfect (Figure 3.4). The exterior vessel used
in the flow cell system was utilized, as well as the base delivery system. The






76
temperature bath was also the same. Solution homogeneity, however, was
accomplished by using a Teflon magnetic stirring bar and a magnetic stirrer.


















pH ELECTRODE pH-STAT,,..

TEMPERATURE PROBE


WATER
JACKET


I. H20


Figure 3.4
Base Addition Kinetic Setup


TITRANT











The accuracy of the rate constant depends entirely upon base

consumption in these reactions. As base consumption is linked to pH stat

controller limitations, the pH state controller was modified for the unactivated
esters to improve the accuracy of the rate constants. Modifying the controller
resulted in pH ranges of 0.03 rather than 0.1. The improvement was

determined by monitoring the reaction pH with a pH meter ( 0.01).
The base consumption must also be large enough to be easily

measured at short intervals but small enough not to significantly change the

solution volume. As most of these reactions are relatively slow, high

concentrations of the ester were typically used to produce a measurable
reaction rate. Short intervals are desirable as many data points may be

acquired early in the reaction. Early in the reaction the ester concentration
remains essentially constant, allowing the assumption of pseudo-first order
reaction kinetics. Changing the solution volume by excessive base addition

would also change the concentration of all solution species, adding a
variable to the system studied. Therefore, the base used to titrate the acid

needed to be concentrated enough to match acid production, yet not too
concentrated so that the end point was overshot. It was necessary to
standardize several KOH solutions in order to strike this balance for each

temperature and ester studied.
The reactions were studied in aqueous solution at 25.00 0.01C

and an ionic strength of 1.0 (maintained with KCI). The methyl acetate koH
value at pH 8.20 was 0.28 0.03 Mls-1, which falls within the range of
values observed over the years by many investigators.26,30,31 The methyl

acetate kZn(OH)-I value at pH 8.20 was (3.3 0.2) x 10-4 M-1s-1, which is in







good agreement with the Zn(OH)-1 catalyzed rate constant reported by
Kimura, 3.1 x 10-4 M-1s-1.1 The temperature dependence study for methyl
acetate hydrolysis (and subsequent determination of AH1 and ASt) also

used this system. Jeni Shepardson (unpublished work, Chemistry
Department, University of Florida, Gainesville, FL, Kinetic Studies.) also

repeated the methyl acetate hydrolyses at 250C and 400C for both the

background and catalyzed reactions. At 250C a second-order rate constant
of 0.300 0.013 M-1s-1 was obtained for the background reaction and
(5.55 0.73) x 10-4 M-1s-1 for the catalyzed reaction. At 400C rate constants

of 0.712 0.088 M-1s1 and (1.27 0.12) x 10-3 M-1s-1 were obtained for
the background and catalyzed reactions, respectively. These values were in

agreement with the values in the earlier temperature dependent study. Ms.
Shepardson was unaware of values obtained in the earlier study or

available in the literature at the time of her study. The 400C background
value was a little lower than the one found in the original study but the

overall agreement was good. Therefore, it is considered a good design for
measurement of catalytic rate constants for methyl acetate and other

carboxylic acids with poor leaving groups.







Dissociation Constant Titrations
To calculate accurate reaction rate constants for p-NPA hydrolysis,

determination of both the temperature dependence of the extinction
coefficient of p-NP- (the spectrophotometrically monitored species) and the
pKa of p-NP were necessary (Table 3.2). Although p-NP- is the dominant

solution species at pH 8.2, its conjugate acid exists in solution as well.
Since the monitored species was the basic form, calculation of the reaction
rate from the p-NP- concentration alone would result in a reaction rate which
was too low. Since correction for the percentage of the product in the acid

form requires the percentage of the acid present at each temperature, the
temperature dependence of the pKa of p-NP was necessary. The extinction
coefficient, as well as the Xmax, for p-NP- was also temperature dependent.
Therefore, determination of the extinction coefficient and the Xmax at each

temperature measured was also necessary.










Table 3.2
Temperature Dependence of the Extinction Coefficient of p-
Nitrophenolate and the pKa of p-Nitrophenol

Temperature p-NP- p-NP
$404a p bPa
(0C) (M-1cm-1) experimental literature

20 18,200 7.19 0.03 7.21


25 18,300 7.12 0.02 7.15


30 18,400 7.04 0.04 7.10


35 18,500 6.99 0.02 7.04


40 18,600 6.95 0.02 7.00
ap= 1.00 (KCI), pH = 10.0, 0.01C. b/= 0.10 (KCI), + 0.01C;
errors are 1 s.d. c Allen, G. F.; Robinson, R. A.; Bower, V. E. J. Phys.
Chem. (1962), 66, 171.









Initially, the dependence of the extinction coefficient upon
temperature was measured by using conditions from a literature source.32
These conditions were [p-NP-] = 1 mM at pH 10. A capped UV-visible cell,

UV-visible spectrophotometer, and constant temperature water bath were

also used in this study. The p-NP- was added by using a stock solution in
dry, spectral grade acetone. All water used was Barnstead nanopure water
which was treated with a metal chelating resin. The temperature ( 0.010C)

was measured as it varied from 20 to 400C. The values agreed well with
literature sources. It was, however, necessary to measure them again by

using the exact solution which would be used for the reactions. The study
was repeated by using conditions similar to those used for the reactions, [p-
NP-] = 5.075 x 10-4 and ,L = 0.10 ( KCI as the source of ionic strength). The

values (Table 3.2) agreed well with literature sources.33,34 The van't Hoff
plot for the temperature dependence was linear. A consistent shift to slightly
lower values may be due to condition variations or to the experimental
method itself. It is precisely for these reasons that the value is determined by

using the experimental setup to be used for the later studies. The AHO and
ASO values, 5.1 0.4 kcal mol-1 and -16 2 cal deg-' mo1-1 respectively,
also agree well with literature values. Allen's potentiometric titration method
produced values of 4.5 0.2 kcal mol-1 and -17.9 1.0 cal deg1mo-1 for
AH and ASO, respectively.33 A calorimetric study by Fernandez and
Heppler produced a AH value of 4.7 kcal mol-1.34 The highly negative

entropy value results from a neutral reactant producing two charged
products, a highly unfavorable entropic situation in highly polar aqueous

conditions. Charged species produce extended solvation spheres, a highly

ordered situation.








The pKa's of p-NP at each temperature were determined by using a
method developed by Martel and Motekaitis.35 They developed exacting

methods to determine pKa's (often of polyprotic acids, which have difficult
titration curves to evaluate) and complex stability constants, as well as

developing computer programs to complement their methods. The
agreement between the experimental points and the computer-generated
titration curve was excellent (Figure 3.5). Their method for determining pKa
values was necessary in order to determine the temperature dependence of

the dissociation constant of p-NP. The pKa does not vary much as the
temperature is increased by 50C; therefore, titration curves would be difficult
to analyze for such subtle changes. Their protocol includes the use of water
treated vigorously to remove inpurities (especially metal ions), carefully

standardized, C02-free KOH solutions, 1 M KCl solution, and reactant
solution concentrations of 0.01 M. Careful calibration of the meter-

electrode system was also undertaken. Other equipment used included the
modified UV-visible cell from the p-NPA study, a constant temperature water

bath, and an automated pipettor for base addition.








2


1

1


1



T
CL


0.05 O'


o.15 0.2
V KOH (mL)


0.25 0.3 0.35


Figure 3.5

Comparison of Experimental Points and a Program-Generated Titration

Curve


0 ..............................................................--...-....-...... . . . ................


............................. -..-......----------------------------- ............~..



8 ............................. ................ .. ..~....~ .............................................

--- ------.......... ..............................................................- ---------


61 . .- .- - --. . . .- - -. . .- -.
I






85


Determination of the temperature dependence of the Zn(OH2)-I
dissociation constant for its complexed water was also necessary (Table

3.3). As its pKa is also close to the pH used in the catalysis reactions,
Zn(OH2)-_ (the catalytically inactive species) exists in solution as well.1 The

percentage present can be calculated once the pKa at each temperature is
known. Therefore, the corrected [Zn(OH)-i] can be calculated and the rate

constant determined by using the rate law. The titration itself was
accomplished in the same manner as for p-NP. Again, the values agreed
well with what information is available for this dissociation (Table 3.2).
Kimura et al reported the values 7.30 and 7.89 for the dissociation at 250C

and 0C.1 Zompa reported a value of 7.51, by using a different ionic
strength source than either Kimura or this study, at 250C.36 The value
observed in this study for 250C, 7.43, was right in between the two literature
values. The value is consistent with the data obtained in this work. The
van't Hoff plot for the temperature dependence was linear. Evaluation of the
van't Hoff plot produced AH and ASO values of 10.7 0.9 kcal mo1-1 and

2 2 cal deg-l mol-1, respectively. The small entropy change represents a

system where solvent ordering is not changed significantly, such as when a
dication produces two monocations. The overall charge in solution is
constant, a favorable entropic factor compared to acid dissociation of a

neutral species.
It was difficult to correlate the values observed with the anion
association study values by Kimura.29 Kimura reported that Cl- had an
association constant of 101.5. As the ionic strength is maintained by using
KCl in this study, the Cl- should displace the OH- in Zn(OH)-_ in over 80% of
the catalyst in solution. If this is accurate, the catalyzed rate constants








calculated should be 5 times lower for solutions where the ionic strength is
maintained by KCI rather than NaC104. In the MA study, the value was

actually higher than by Kimura and coworkers. These results were
supported by kinetic studies done by another investigator, Jeni Shepardson

(unpublished work, Chemistry Department, University of Florida, Gainesville,
FL, Kinetic Studies.). The p-NPA catalyzed rate constant was, indeed, lower

for KCI solutions that for NaC104. It is not, however, as low as expected.
These results were also supported by kinetic runs done by Jeni

Shepardson. The acid dissociation constant values were also in excellent
agreement with literature values. The dissociation constants would not be
close to their expected values by using a potentiometric method where only

the amount of base added, moles of catalyst present, and position of the
steepest slope of the curve are used to determine the pKa. According to
Kimura's values, the moles of acid present would have actually been much
lower; therefore, the value would have been corrected for that. The titration
curve in this study was that of a well-shaped, clearly monoprotic acid.









Table 3.3
Temperature Dependence of the Deprotonation Constant (pKa) of
the Zn"-Bound H20 in Zn(OH2)-I


Tem pe ratu re
(C)


pKa experimentala


pKa literature


7.89 0.03b


7.62 0.13


7.43 0.08


7.32 0.07


7.21 0.07


7.09 + 0.04


7.51c


a = 1.00 (KCI), 0.01C; errors are 1 s.d. b Kimura, Eiichi J. Am.
Chem. Soc. (1990), 112, 5805. c Zompa, L. J. Inorg. Chem. (1978),
17,2531.











Summary and Conclusions


Using Kimura's method, Zn(OH)-l was synthesized and twice
recrystallized from water to produce colorless crystals. Elemental analysis,

NMR, and FT-IR spectra agreed well with literature values.1
Drawbacks of the flow cell spectrophotometric system included pH
variance, sporadic temperature fluctuations, adsorption of ester upon interior

tubing walls, and unknown reactivity of buffer species with the ester in the
presence of Zn(OH)-I. A modified pH stat controlled system (Figure 3.2)
was then constructed by using a fabricated UV-visible cell with a double-
walled solution reservoir (Figure 3.3) rather than a flow cell/exterior vessel

recirculation method. The reconfigured system allowed measurement of

catalytic rate constants when pH variation and substrate adhesion, which
limited the flow cell system, were not factors. The Zn(OH)-I catalytic rate
constants by using this system were reproducible ( 5-10%). Therefore, it

was considered a good design.
The catalyzed p-NPA hydrolysis value at 250C was higher in this

study, 0.058 M-1s1 than the value obtained by Kimura, 0.041 M's"1.1 It is
unclear, however, whether Kimura took the percentage of the monitored
product in the p-NP form and Zn(OH2)-1 present into account in his

calculations. With these corrections the values would be in good
agreement. The difference in Zn(OH )-1 catalyzed values between the
HEPES/NaCIO4 solution and the KCI will be further discussed in light of the

association constant work done by Kimura in the dissociation constant
conclusions. The catalyzed methyl acetate hydrolysis value at 250C in this








study, 3.3 x 10-4 M-s-1, was in good agreement with the value obtained by

Kimura, 3.1 x 10-4 M-ls1.1
In order to calculate the catalytic rate constants, it was necessary to

determine the temperature dependence of the pKa of Zn(OH2)-I. For the p-

NPA temperature dependence study calculations, it was also necessary to
redetermine the pKa for p-NP by using our experimental setup. These
studies also allowed calculation of the AH and ASO for both acid

dissociations. The specific value for the p-NP pKa was determined for each
temperature used in the p-NPA temperature study. The pKa values obtained

agreed well with literature values (Table 3.2).33,34 A consistent shift to
slightly lower values may be due to condition variations or to the
experimental method itself. The AH and ASO values were 5.1 0.4 kcal
mol-1 and -16 t 2 cal deg-lmol-1, respectively. In contrast, the
thermodynamics of the acid dissociation of Zn(OH2)-I is quite different

energetically. The dissociation of Zn(OH2)-I was also studied at each
temperature to be used later in the MA and p-NPA temperature dependent
catalysis studies. The values agreed well with literature values for this
dissociation (Table 3.3).1 The AH and AS0 values were 10.7 0.9 kcal

mol-1 and 2 2 cal deg-1 mol-1, respectively.
The anion association study by Kimura was difficult to correlate with
the values observed.29 Kimura reported that C- had an association constant

of 101-5. As the ionic strength is maintained by using KCI in this study, the
Cl- should displace the OH- in Zn(OH)-I in over 80% of the catalyst in
solution. The observed curve in this study was that of a well-shaped, clearly

monoprotic acid.













CHAPTER 4


TEMPERATURE DEPENDENCE OF METHYL ACETATE AND p-
NITROPHENYL ACETATE HYDROLYSES


Introduction


The temperature dependence of carboxylic acid ester hydrolyses by
organic catalysts has been studied by several investigators.26,23,30 The
AHt and ASl values obtained from such studies can provide clues about the

catalytic mechanism, and understanding the catalytic mechanism may allow
optimization of reaction conditions and synthesis of improved catalysts. For
comparison to the literature data on organic catalysts, a study of the
temperature dependence of ester hydrolysis by Zn(OH)-1 was undertaken.
Relatively little data are available on the thermodynamic parameters for zinc
catalyzed reactions of this type.
The activation parameters for p-nitrophenyl acetate (p-NPA) and
methyl acetate catalyzed hydrolyses were determined from Eyring plots of

their temperature-dependent rate constants in solvent alone and in the
presence of Zn(OH)-1. Comparison of the values for p-NPA, an ester which

typically undergoes nucleophilic hydrolysis, and methyl acetate, an ester
which typically undergoes general base hydrolysis, can often provide

detailed information on the mechanism of action for the catalyst under
investigation.39