Micellar catalyzed reactions for flow injection analysis

MISSING IMAGE

Material Information

Title:
Micellar catalyzed reactions for flow injection analysis
Physical Description:
x, 77 leaves : ill. ; 28 cm.
Language:
English
Creator:
Hernández Torres, María A., 1958-
Publisher:
s.n.
Publication Date:

Subjects

Subjects / Keywords:
Instrumental analysis   ( lcsh )
Micelles   ( lcsh )
Chemistry thesis Ph. D
Dissertations, Academic -- Chemistry -- UF
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1986.
Bibliography:
Bibliography: leaves 73-76.
Statement of Responsibility:
by María A. Hernández Torres.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000989399
oclc - 17689643
notis - AEW6261
sobekcm - AA00004861_00001
System ID:
AA00004861:00001

Full Text













MICELLAR CATALYZED REACTIONS
FOR FLOW INJECTION ANALYSIS






BY






MARfA A. HERNANDEZ TORRES


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


1986



























To my parents,

Saul and Carmen,

with all my love.













ACKNOWLEDGMENTS


I would like to express my gratitude and appreciation to my

research director, Dr. John G. Dorsey, for his invaluable assistance,

guidance and encouragement throughout the years. I was very lucky to

have him as a teacher. His special qualities helped me to develop as

a professional, and at the same time he allowed me to grow as an

individual. His view of science and life created a good and healthy

environment for learning. His lessons will always be remembered.

Special thanks are given to Dr. Morteza G. Khaledi, for his

helpful suggestions and advice at the beginning of this project.

I would like to express my gratitude to my colleagues and friends

in Dr. Dorsey's research group, for their friendship and daily

encouragement. I treasure very much the time spent at the lab with

all of them even when life was tough. I will miss them all!

I would like to thank the faculty members at the University of

Florida with whom I was in contact during my graduate years. Special

thanks are due to the members of my committee, for their contributions

to my learning experience.

I would like to mention Cindy Zimmerman, for her promptness and

accuracy in typing the final draft of this thesis.

There are no words to express how thankful I am to my family and

friends. Their love, support, encouragement, patience, and faith in

me made possible this new achievement in my career. I want to thank









my grandparents, Avelina and Jose Antonio, for their care throughout

the years.














TABLE OF CONTENTS


Page
ACKNOWLEDGMENTS.............................. ............. ...... iii

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

LIST OF FIGURES ............. ................ .. .................vii

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

CHAPTERS

I INTRODUCTION................................. ............

Principles of Flow Injection Analysis....................1
Properties of Surfactants and Micelles in Aqueous
Solution ................... ........................ ..10

II THEORY AND BACKGROUND.................................... 17

Chemical Kinetics in a Flow Injection Analysis System....17
General Features of Micellar Catalysis...................19

III EXPERIMENTAL: FIA SYSTEM FOR PYRIDOXAL DETERMINATION....23

Apparatus ............................................... 23
Reagents............................... ........ ... ..... 25
Procedure...................... oo ................ ... .. 25

IV MICELLAR CATALYSIS IN THE DETERMINATION OF PYRIDOXAL
BY FLOW INJECTION ANALYSIS...............................26

Results and Discussion...................................26
Measurements of Dispersion...............................55

V CONCLUSIONS AND FUTURE WORK..............................70

REFERENCES ....... ................ ................................ 73

BIOGRAPHICAL SKETCH.... ................................ ....... ......77













LIST OF TABLES


Table Page

I Figures of merit for pyridoxal determination................40

II Variable parameters for the Modified Simplex Optimization
program .....................................................42

III Optimized and fixed variables for the determination of
pyridoxal ................................................... 45

IV Figures of merit for pyridoxal determination with
conditions obtained by Modified Simplex program.............47

V Variance and standard deviation values for pyridoxal
in aqueous media at 10% peak height.........................59

VI Variance and standard deviation values for pyridoxal
in aqueous media at 30% peak height.........................60

VII Variance and standard deviation values for pyridoxal
in aqueous media at 50% peak height..........................61

VIII Variance and standard deviation values for pyridoxal
in 0.05 M CTAB micellar media at 10% peak height............62

IX Variance and standard deviation values for pyridoxal
in 0.05 M CTAB micellar media at 30% peak height............63

X Variance and standard deviation values for pyridoxal
in 0.05 M CTAB micellar media at 50% peak height............64

XI Average values for variance and standard deviation
for pyridoxal in 0.05 M CTAB micellar and aqueous media
at 10, 30, and 50% peak height for aqueous and
0.05 M CTAB systems ........................................ 65

XII Dispersion values for aqueous and micellar system...........68

XIII Measurement of dispersion versus CTAB concentration.........69













LIST OF FIGURES


Figure Page

1 Diagram of an FIA system (a) and typical recording (b).......3

2 Dispersion in an FIA system .................................6

3 Velocity profiles and shapes of injected sample bolus........9

4 Dill-Flory's representation of a normal micelle.............12

5 A two dimensional schematic representation of the
regions of a spherical ionic micelle........................14

6 FIA manifold for the determination of pyridoxal..............24

7 Reaction and surfactant media used for the analysis
of pyridoxal ................................................27

8 Absorbance versus wavelength (nm) in aqueous system
for the determination of pyridoxal.........................29

9 Absorbance versus wavelength (nm) in 0.05 M CTAB
micellar system for the determination of pyridoxal..........30

10 Change in maximum absorbance (355 nm) versus time
(minutes) in aqueous media for the determination
of pyridoxal ................................................ 31

11 Change in maximum absorbance (355 nm) versus time
(minutes) in 0.05 M CTAB micellar media for the
determination of pyridoxal ................................32

12 Log (A,-At) versus time (minutes) for pyridoxal
in aqueous system...........................................34
13 Log (A.-At) versus time (minutes) for pyridoxal
in 0.05 M CTAB system......................................35
14 Absorbance calibration plots for pyridoxal in aqueous
( ] ) and 0.05 M CTAB (m ) systems.........................36









15 Fluorescence calibration plots for pyridoxal
determination in aqueous ( 0 ) and 0.05 M CTAB ( U )
systems..................................................... 38

16 Absorbance calibration plots for pyridoxal in aqueous
( 9 ) and 0.09 M CTAB micellar ( 0 ) media..................46

17 Absorbance recordings of a series of pyridoxal
standards in aqueous media..................................48

18 Absorbance recordings of a series of pyridoxal
standards in 0.05 M CTAB micellar media.....................49

19 Absorbance recordings of a series of pyridoxal
standards in 0.09 M CTAB micellar media.....................51

20 Fluorescence recordings of a series of pyridoxal
standards in aqueous media..................................52

21 Fluorescence recordings of a series of pyridoxal
standards in 0.05 M CTAB micellar media.....................54

22 Measurement of peak width, W, and asymmetry factor,
B/A, at 10, 30 and 50% peak height in an FIA peak...........57


viii













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

MICELLAR CATALYZED REACTIONS
FOR FLOW INJECTION ANALYSIS

BY

MARIA A. HERNANDEZ TORRES

August, 1986

Chairman: Dr. John G. Dorsey
Major Department: Chemistry


In this study, the applicability of micellar carrier streams for

the catalysis of reactions in FIA was investigated. Flow Injection

Analysis (FIA) is an automated kinetic method of analysis. An FIA

system with low dispersion and fast reaction kinetics will provide low

limits of detection and high sampling rates.

Micelles exhibit the ability to solubilize hydrophobic compounds

on or within their structures. The rate of many reactions in micellar

media is altered due to the proximity of reagents and analyte, changes

in the microenvironment and orientation of solutes.

The advantages of combining the solubilization property and

micellar catalysis for a given reaction that is taking place in an FIA

system are demonstrated. The reaction of pyridoxal (a B6 vitamer)

with cyanide was investigated in aqueous and micellar media. The

cationic surfactant, hexadecyltrimethylammonium bromide (CTAB), was

chosen for the micellar carrier solution. The oxidation product of








this reaction, 4-pyridoxolactone, was detected either fluorimetrically

or by ultraviolet absorbance. The reaction rates in the two media

were determined and compared. Calibration plots for pyridoxal were

made and the analytical figures of merit were compared for aqueous and

0.05 M CTAB micellar media.

A Simplex Optimization was carried out for the determination of

pyridoxal in micellar media. A new set of conditions for micellar

media was obtained from the Simplex program. With this new set of

conditions, a calibration plot for the micellar media was prepared and

subsequently compared to the calibration plot for the aqueous

system. The analytical figures of merit for the two carrier solutions

were calculated and compared.

The peak shape obtained in the FIA system was investigated. The

standard deviation and variance at 10, 30, and 50% peak height were

calculated by empirical equations. The agreement of these values

allows the peaks to be classified as exponentially modified gaussians.

Measurement of dispersion in both aqueous and micellar systems

was investigated. Dispersion values for both aqueous and micellar

systems were examined by two methods: by using the definition of

dispersion in FIA and by measuring the variance (second moment) at 104

peak height. In both methods, higher values for dispersion were

obtained for micellar media.













CHAPTER I
INTRODUCTION

Principles of Flow Injection Analysis

Flow injection analysis (FIA) is now established as a fast,

precise, accurate, efficient and extremely versatile analytical

tool. The FIA technique is used by many analytical chemists working

in a variety of different industries and institutions (1-9).

A historical review of the development of FIA reveals that since

its conception in the early 1970s, many of the concepts of flow

injection analysis have been adopted from other fields and many

workers have contributed to its development (10). However, the

technique of FIA became known by the simultaneous appearance of the

work of Stewart, Beecher and Hare (11) in the United States and

Ruzicka and Hansen (12) in Denmark in 1975. The Danish group

developed the method using primarily instrumentation normally

associated with segmented flow analyzers. In contrast, the American

group based their initial work on high performance liquid

chromatography components. For this reason, FIA is considered a

hybrid of the two techniques.

In the past, it was generally assumed based on Skegg's concept

that air segmentation and attainment of a steady state signal were

essential for performing continuous flow analysis (6). The presence

of air bubbles in the analytical stream was thought to be necessary









to limit sample dispersion, to promote mixing of the sample with

reagents (by generating turbulent flow) and to scrub the walls of the

analytical conduits to prevent carryover of samples. However, it was

proven that analysis without air segmentation is not only possible but

also advantageous. In FIA, there is no air segmentation, the sample

is introduced as a plug via a valve or syringe, mixing is mainly by

diffusion-controlled processes, and the response curves do not reach

the steady-state plateau, but have the form of sharp peaks. The

absence of air segmentation leads to a higher sample throughput. The

presence of a sample carrier interface over which concentration

gradients develop during the course of analysis has opened new

analytical possibilities for continuous flow analysis. The

reproducibility is good, and there is no sample carryover. There is

no need to introduce and remove air bubbles, and an expensive high

quality pump is not necessary.

Flow injection analysis is based on the injection of a liquid

sample into a moving, nonsegmented reagent carrier stream. The

injected sample forms a zone that disperses and reacts with the

carrier on its way to a detector (5). The simplest FIA analyzer

(Figure la) consists of a pump (P) that propels the carrier stream

(R), an injector port (S), by which a well defined sample volume is

injected into a carrier stream, and a coil in which the sample zone

disperses and reacts with the components of the carrier stream,

forming a species to be sensed by a flow-through detector (D). A

typical recording has the form of a sharp peak (Figure lb) the height

of which is related to the concentration of analyte, and the time from




















S mt/min
(a)
S
P






2-30 s


(b)
IT --------







H






S z


Sit










Figure 1. Diagram of an FIA system (a) and typical recording (b)
(from 1).








injection to peak maxima is termed the residence time. Residence

times are typically from 3 to 30 seconds.

Between the points of injection and detection the sample plug

will have been physically dispersed to some degree and, in addition,

some chemical reactions will have taken place. The peak will reflect

both processes.

The injection of a sample has to be done in such a way that by

inserting a discrete slug of sample into a continuously moving stream

the movement of the stream is not disturbed. The residence time of

the sample in the analytical manifold should ideally be identical for

each sample, and the conditions to which the sample is exposed during

processing should also be the same. The reason for this is that not

only the physical dispersion but the chemistry involved requires a

reproducible travel pattern of the sample from the point of injection

to detection. In an FIA system, neither the mixing nor the chemical

reaction is complete; an equilibrium for either process is not

attained (1). By monitoring the reaction at a fixed precise time, the

concept of the steady state can be abandoned. This measurement at a

fixed time is just as analytically significant as the steady state

signal. Any fluctuations in residence time of the sample, i.e.,

variation in flow rate, will result in imprecise measurement of the

signal.

When a sample is first injected, it forms a well-defined sample

plug in the stream. As the sample is swept downstream through the

analytical conduits of narrow bore tubing, the plug disperses into

and, thus, mixes with the carrier stream under laminar flow conditions








to form a gradient. The magnitude of this dispersion is dependent on

the operating parameters applied to the system, including sample

volume, tubing bore size, tubing length, flow rate, sample volume and

coil diameter (1,6). Varying the values of these parameters confers a

significant degree of control over the dispersion characteristics and

facilitates optimization of a flow injection system for many diverse

applications, so that the optimum response is obtained at minimum time

and reagent expense.

The response curve has the shape of a peak reflecting a continuum

of concentrations. In contrast to all previous methods of automated

assay, there is no single element of fluid that has the same

concentration as the neighboring one (7) (Figure 2).

The dispersion coefficient, D, is the ratio of the concentration

of sample solution before (C) and after (C) the dispersion process

has taken place (Figure 2). The dispersion coefficient can be defined

at any point of the curve but, for convenience in the majority of FIA

methods, the dispersion coefficient is defined at maximum peak height.


C
D = (eq. 1.1)
max


In cases where a reaction is developed as a result of mixing

effected by dispersion, the peak height will increase as the reaction

proceeds toward completion. The maximum response will then be

attained when an optimum balance is reached between dispersion and

reaction time.








-- 3SNOdS3d


)c


CQ)


-'- NO I0VJ N30NO3








Flow injection analysis operates only in the laminar flow region

(3,13-16). In this region, FIA systems generate dispersion through

both diffusion and convection. Under such conditions of laminar flow,

the layer of liquid in contact with the tube surface is practically

stationary and the velocity of centrally placed molecules is twice the

mean velocity of the liquid. This gives rise to a parabolic velocity

profile (Figure 3a). In the absence of molecular diffusion, a sample

placed into a moving stream would have an infinitely long tail by the

time it reached the detector (Figure 3b). This results in

unacceptable carryover between samples. Diffusion of molecules

between the carrier and sample bolus serves to limit this convective

dispersion and effectively mixes sample and reagent. In coiled tubes,

it is the radial rather than the axial dispersion that contributes

most significantly to sample dispersion in FIA systems. This type of

dispersion, also called secondary flow, operates to move the fluid

both toward and away from the tubing wall and thus serves as an

efficient scrubbing mechanism (Figure 3d). A molecule located at the

center will tend to diffuse into a region of lower sample

concentration and by doing so it will move into a layer of liquid

moving at a slower longitudinal velocity. On the other hand, a

molecule located near the wall will diffuse toward the center of the

carrier solution and it will encounter a layer of faster moving

liquid, which will carry it away from the wall and toward the center

of the sample zone. This radial diffusion perpendicular to the

direction of the flow modifies the shape of the bolus head and the

result is low carryover and cross contamination. High sample

























Velocity profiles and shape of an injected sample bolus:
(a) Laminar flow, parabolic velocity profile; (b) Sample
dispersion caused by laminar flow without diffusion; (c)
Sample shape resulting from laminar flow with molecular
diffusion; (d) Secondary flow pattern in the cross-section
of a tightly coiled tube (from 3).


Figure 3.












(a) tube wall


(b)


(c)


(d)









throughput is obtained because of the limited band spreading. When

samples under these conditions reach the detector they have a shape as

in Figure 3c.

The versatility and simplicity of FIA systems permits this

technique to be widely used for performing chemical assays. The

recent increase in publications dealing with new FIA methods as well

as in separate symposia on this topic indicate the popularity of this

relatively new technique (2). On the other hand, more has to be done

on the theoretical side, where the dispersion of the sample zone must

be investigated in much greater detail. The theory of dispersion,

although very useful, is far from being exact and complete (17). Only

deeper theoretical studies will lead to design of even more advanced

continuous flow techniques that will allow chemical analyses to be

performed in new ways.



Properties of Surfactants and Micelles in Aqueous Solution

Scientists from around the world have shown special interest in

surfactants because of their unique characteristics (18-24).

Surfactants, or surface active agents, are amphiphilic molecules

having both pronounced hydrophobic and hydrophilic properties.

Surfactants are classified as cationic, anionic, nonionic or

zwitterionic according to the hydrophilic part (polar head group).

The hydrophobic tail consists of a hydrocarbon chain usually from 8 to

20 carbon atoms in length. Furthermore, this hydrophobic tail can

contain unsaturated portions or aromatic moieties, can be partly or








completely halogenated, and can be branched or consist of two or more

chains.

At low concentration, the surfactant is dispersed predominantly

as monomers, although dimers, trimers, etc. can exist. Over a narrow

concentration range, termed the critical micelle concentration (CMC),

surfactants have the important property of forming molecular

aggregates, called micelles. Above the CMC, there exists a dynamic

equilibrium between monomers and micelles. The amount of free monomer

remains approximately constant and equal to the CMC.

In aqueous solutions, surfactant molecules (typically from 60 to

100) associate to form a roughly spherical cluster (25) (Figure 4).

This micelle structure is such that the hydrophilic head groups are

directed toward and in contact with the aqueous solution, forming a

polar surface, while the hydrophobic tails are directed away from the

water forming a liquid-like hydrocarbon core. The microviscosity of

this core is considerably higher than in hydrocarbons. The micellar

surface is not uniform: some of the hydrocarbon chains are turned

towards the solvent or at least come into contact with it periodically

(26). On the whole, it may be supposed that the micellar surface is a

polar environment differing in properties from water itself.

Changes in temperature, concentration of surfactant, additives in

the liquid phase, and structural groups in the surfactant may cause

changes in the size, shape, and aggregation number of the micelle

(26-28).




















































Dill-Flory's representation of a normal micelle. The ionic
head groups are indicated by the circles, and the
hydrocarbon chains are pointing toward the center of the
micelle (from 25).


Figure 4.








In micelles of ionic surfactants, the charged head groups and the

counterions are located in a compact region, known as the Stern layer,

which extends from the core to within a few angstroms of the shear

surface of the micelle. Beyond the Stern layer, the remainder of the

counterions are located in the Gouy-Chapman electrical double layer

where they are completely dissociated from the charged aggregate and

are able to exchange with ions in the bulk of the solution (26)

(Figure 5).

The driving force for micelle formation in aqueous media is due

primarily to the hydrophobic effect, and the electrostatic repulsion

between the ionic head group limits the size that a micelle can

attain, thereby keeping the micelle size small.

One of the most important properties of micellar systems is their

ability to solubilize a variety of species (18-24,26-28).

Solubilization may be defined as the spontaneous dissolving of

substance (solid, liquid or gas) by reversible interaction with the

micelles of a surfactant in a solvent to form a thermodynamically

stable isotropic solution with reduced thermodynamic activity of the

solubilized material (24). The importance of the phenomenon of

solubilization from the practical point of view is that it makes

possible the dissolving of substances in solvents in which they are

normally insoluble or slightly soluble. Solubilization is a dynamic

equilibrium process and depends on temperature, nature of solute,

surfactant concentration and type of micellar system employed (26).

There are several possible solubilization sites and orientations

available in a micellar system and the site occupied by a solubilizate











Aqueous
bulk
phase


Range of
shear |I Core
surface I 0- 28
---- l-Stern layer,
up to a few A
Gouy-Chapman
double layer, up to
several hundred









Figure 5. A two dimensional schematic representation of the regions
of a spherical ionic micelle (from 24).








depends upon the nature of both the solute and the micelle. There is

a rapid equilibrium between various possible sites and also between

the solubilized state and the free state in the aqueous medium. The

site of solubilization is a topic of current debate. Some contend, if

the solute is nonpolar, it may pass completely into the hydrophobic

core or penetrate a particular depth into the surface layer. Others

contend it may be adsorbed at a hydrophobic region on the surface

(18).

Micellar systems are very useful in the field of analytical

chemistry due to the unique properties of these organized

assemblies. Micelles can lead to the modification and improvement of

existing procedures and to the development of new methods of chemical

analysis. Micellar structures offer means to overcome solubility

problems, to speed reaction rates of analytical reactions and to

reduce side reactions, to shift acid base or redox equilibria, to

change spectral distribution or intensities, and to improve

selectivity and efficiency in extraction and chromatographic

methods. Several reviews on the use of surfactants in analytical

chemistry have appeared (29-31).

Our studies are directed mainly to observing the behavior of

combining micellar media with the technique of FIA. This represents

the first attempt to demonstrate the advantages of combining the

solubilization property and micellar catalysis for a given reaction

that is taking place in an FIA system. Higher sensitivity and/or

lower limits of detection are expected for a given FIA system when the

reaction is carried out with the appropriate micellar media. Kinetic





16


studies and measurement of the analytical figures of merit will be

compared for reactions taking place in aqueous and micellar media.













CHAPTER II
THEORY AND BACKGROUND


Chemical Kinetics in a Flow Injection Analysis System

It is well recognized that flow injection analysis yields a

response curve which is the result of two processes, both kinetic in

nature: the physical process of dispersion and the chemical process

of formation of reaction products (1-6,32).

The kinetics of physical dispersion, or incomplete mixing, has

been described in a number of papers (14-16,33-35) while attempts to

describe the effect of chemical kinetics has been practically

ignored. The shape of the transient peak profiles of FIA

determinations have been described only in terms of the dispersion

process. However, there has been an attempt recently to explore in

more detail the complexity of the overall kinetic process taking place

inside the sample plug and its boundaries in FIA systems and the

contribution of this kinetic chemistry to the peak profiles. Ruzicka

and Hansen recognized that "it is obvious that the comprehensive

theory of the flow method will eventually combine the theory of mixing

of liquids in continuous moving streams with the theory of chemical

kinetics" (36). From now on, more attention will be paid to the

chemistry process that is taking place (37-42).

In developing a flow injection method, one of the primary goals

is to maximize response together with sample throughput. These








parameters are intimately related to the sensitivity of the method and

the number of injections possible per hour.

Since there exists an interdependence between the reaction rate

and the rate of dispersion, both factors have to be weighed when

designing a new FIA system. The longer a sample stays within the

system, the greater will be the signal due to the chemistry that is

taking place. Longer residence times mean more time is allowed for

the reaction to take place, therefore more product is formed. This

increase is balanced against the point where dispersion will overcome

the formation of the product and cause the signal to decrease. The

longer the sample stays between the points of injection and detection,

the higher will be the dispersion and the signal measured (usually

peak height) will decrease. Another factor to be considered is time

needed for each determination. Longer residence times allow maximum

sensitivity of measurement but at the expense of decreasing sample

throughput. Therefore, short residence times are often preferred.

Fast chemical reactions are required for performing a simple FIA

analysis at a continuous carrier pumping rate with practical residence

times of about 30 seconds. Slow reactions must presently be performed

by the stopped flow mode or in a packed reactor (17). In stopped flow

mode, an electronic timer is used to cease the movement of the carrier

stream containing the sample zone in order to allow enough time to

produce an adequate amount of detectable product, and then the sample

is pumped through the flow cell while the peak is recorded in the

usual manner. Reactors packed with inert materials, such as glass,








enhance micromixing of sample and reagents in the carrier stream

without loss of peak height or loss of sample frequency.



General Features of Micellar Catalysis

It has been reported that with the proper choice of surfactant

the rate of a chemical reaction is substantially enhanced in a

micellar solution relative to that in the corresponding aqueous system

(18,24,26,27,31,43-45).

Research on the effect of surfactants on the kinetics of organic

reactions has demonstrated that it is the micelle structure, not the

individual molecules, that is responsible for the catalysis or

inhibition of these reactions. In accordance with this fact, the term

micellar catalysis has been applied to this phenomenon. Rate

acceleration or inhibition of organic reactions in micellar solution

arises from different rates of reaction of the substrate in the

micellar phase and in the bulk solution and the distribution of the

substrate between these two phases (26,32). Then, a prerequisite to

understanding reaction kinetics in micellar systems is to understand

the structure and solubilization properties of the micelles themselves

videe supra).

The kinetics of organic reactions occurring in micellar systems

are dominated by electrostatic and hydrophobic interactions between

the micelle structures, reactants, transition states and products.

The two physicochemical factors responsible for the efficiency of

micellar catalysis are








(1) the change in the reactivity of reagents on transfer from

water to the micellar phase and

(2) the concentration of reactants into the micellar phase.

The first factor, the differences in reactivity, can be explained by

the difference in the distribution of a substrate between these two

phases and by the difference in the degree and nature of substrate-

micelle binding. When a solute is solubilized in a micellar system,

the microenvironment about it is very different compared to that in

the bulk solvent. Micellar systems have the ability to change the

effective microenvironment and the microscopic properties of

solubilized solutes to that of aqueous media favoring the acceleration

of some organic and inorganic reactions.

For catalysis to occur, it is necessary that the substrate be

solubilized by the micelle and the site of solubilization be such that

the reactive site of the substrate is accessible to the attacking

reagent. It is here that hydrophobic interactions become important,

because they determine the extent and the site of solubilization in

the micelle. In general micellar effects on reactions follow several

rules, although there are exceptions. A hydrophobic reactant is

attached to a micelle by hydrophobic interactions, independently of

the charge on the micelle. If the second reactant is oppositely

charged to the micelle, it will be bound to the micelle and the

reaction is usually accelerated. When micelles and reactant ions bear

like charges, the reaction is inhibited due to the repulsion forces

between the ions and the micelle's surface. Nonionic or zwitterionic

micelles, generally, have no significant influence on the rates of









these reactions. The rate of certain organic reactions is unaffected

when one of the reactants is incorporated into the micellar phase and

the other is excluded from it. Exceptions to these rules can be

explained by the fact that sometimes hydrophobic effects overcome the

electrostatic repulsions and even when the micelle's surface charge

does not favor the reaction, catalysis does occur.

The orientation of the reagents in micellar media is different

from the bulk aqueous phase due to the different microenvironment that

the reagents experience on or within the micelles. If this micro-

environment is more attractive to reagents, the reagents are going to

spend more time on this phase; therefore, they will be concentrated in

this region. Also, the micelle structure provides a very specific and

reduced region where the reagents are being solubilized. If the

reagents are localized within this small region, they are being

concentrated and are closer to each other inducing the reaction to go

faster. These are called proximity and concentration effects.

Quite generally, increasing the hydrophobic character of the

surfactant, having longer alkyl chains, increases its efficiency as a

catalyst. At equal concentration of two surfactants, the more

hydrophobic may appear to be the better catalyst (or inhibitor) simply

because it has greater affinity for the substrate. Variation in

substrate structure has a profound influence, in many cases, on the

magnitude of micellar catalysis. The general rule seems to be that

the more hydrophobic the substrate, the more pronounced the micellar

catalysis.








The multiphase profile of surfactant concentration on the

reaction rate is as follows: below the CMC, the rate constants are

independent of surfactant concentration; above the CMC, the rate

constants rise rapidly with increasing surfactant concentration, level

off, and finally decrease with increasing concentration of

surfactant. This profile can be rationalized by the fact that the

rate constant increases as the concentration of micellar bound

reactants increases, but eventually an increase in surfactant

concentration dilutes the reactants in the micellar pseudophase, with

a decrease in the rate constant (18).

The influence of electrolytes on micellar catalysis is less

predictable. For most reactions micellar catalysis is inhibited by

counterions and the larger the ion, the greater the effect. This

behavior has been rationalized by assuming a competition between the

reactant and the electrolyte for a binding site on or in the

micelle. This salt inhibition may be explained principally by the

displacement of one reactant from the micellar surface by the

electrolyte. Enhancement of the micellar catalyzed reaction rate by

counterions has been suggested to be due to changes in micellar

structure by the salts and this new configuration of the surfactants

promotes the reactions.

Another advantage of using micellar media is that of favorable

substrate partitioning and binding in specific orientations and

configurations on the micelle structure. This makes the reaction very

selective to that substrate by decreasing the probability of other

interfering species competing for the place of the substrate.













CHAPTER III
EXPERIMENTAL: FIA SYSTEMS FOR
PYRIDOXAL DETERMINATION



Apparatus

The flow injection manifold used is shown in Figure 6. The

reagent streams were pumped by an Isco (Lincoln, NE) Tris model

peristaltic pump. Samples were introduced with a Rheodyne (Cotati,

CA) model 7125 sample injection valve with a 10 u1 loop. All tubing

was Teflon from the Anspec Company, Inc. (Ann Arbor, MI) with 0.5 mm

internal diameter. The reaction coil, 200 cm long, was thermostated

by immersing it in a water bath with a Techne TE-7 circulator

(Cambridge, England). For fluorescence measurement, a Varian (Palo

Alto, CA) model Fluorichrom detector with a 25 u1 total volume flow

cell was used. A combination of Varian filters was selected for

355 nm excitation wavelength and 435 nm emission wavelength. A Kratos

(Ramsey, NJ) model Spectroflow 757 absorbance detector with a 12 u1

flow cell was set at a wavelength of 355 nm to measure the absorbance

of the product. The output signals were recorded on a strip chart

OmniScribe recorder, Houston Instrument (Austin, TX). The pH of the

reagent solutions were measured with a Corning (Medfield, MA) model pH

meter 130.

For kinetic studies, the reaction was followed spectrophoto-

metrically by measuring the rate of change in the absorbance of






























10

II
E


x


ia-


w




O "
I
0-









4-pyridoxolactone, at 355 nm using a Hewlett Packard (San Diego, CA)

model 3450A Diode Array spectrophotometer connected to a Hewlett

Packard model 7470A plotter.



Reagents

All reagents were used as received and prepared either in

deionized water or in surfactant solutions. The cationic surfactant

was hexadecyltrimethylammonium bromide (CTAB, purum grade) from Fluka

Chemical (Hauppauge, NY). Standard solutions of pyridoxal, Sigma

grade (Sigma Chemical Company, St. Louis, MO) and solutions of

potassium cyanide, certified ACS, from Fisher Scientific Company

(Fair Lawn, NJ) were used for this study. Phosphate buffer solutions

(0.6 M) certified ACS from Fisher Scientific Copmany were prepared and

the pH was adjusted with concentrated hydrochloric acid, ACS

certified, from Mallinckrodt (Paris, KY).



Procedure

The appropriate weight of surfactant was dissolved in distilled

water and the solution then filtered through a 0.45 An nylon-66

membrane filter (Rainin Instruments, Woburn, MA). Appropriate amounts

of pyridoxal and cyanide were dissolved either in distilled H20 or

micellar solution. All reported values are averages of at least four

determinations.













CHAPTER IV
MICELLAR CATALYSIS IN THE DETERMINATION OF
PYRIDOXAL BY FLOW INJECTION ANALYSIS



Results and Discussion

Pyridoxal is one of the three substances designated as vitamin

B6. Determination of pyridoxal and its derivatives is of great

interest, especially in clinical chemistry. A fundamental role for

pyridoxal has been postulated in the mechanism for active transport of

amino acids and metals ions across the cell membrane (46).

The analysis of pyridoxal in biological material has proven to be

difficult and unsatisfactory. Pyridoxal has been analyzed via high

pressure liquid chromatography (HPLC) with amperometric, enzymatic-

fluorometric or photometric detectors and via radiochemical means.

The most common detection procedure for pyridoxal is fluorimetry, in-

volving the use of Zn-glycine or formation of hydrazone derivatives

(47).

P. Linares and coworkers (47) reported a fluorimetric method for

determination of pyridoxal by flow injection analysis. This method

was based on the oxidation of pyridoxal in the presence of cyanide to

yield 4-pyridoxolactone (see Figure 7a).

There is no spectral evidence of reaction between cyanide and the

other vitamin B6 derivatives. These results indicate that the

carbonyl is the only group in the pyridoxal molecule capable of

reacting with cyanide (48-50).










N CH3

HOCH2 OH
CHO


+ CN-


pyridoxal cyanide


N HCH3

'OH


4-pyridoxol tone
4-pyri doxol ac tone


CTAB:

CH,(CH)N (CH ) Br
PA5(C3 3


Figure 7. Reaction and surfactant media used for the analysis of
pyridoxal. Reaction for pyridoxal with cyanide (a) and
CTAB's molecular formula (b).








Preliminary studies on the behavior of this reaction are shown in

Figures 8 and 9. At a wavelength interval of 280 to 400 nm, the

change in absorbance versus time was recorded for aqueous and micellar

media. The cationic surfactant, CTAB, was chosen to attempt to

promote the rate of this reaction (see Figure 7b). A CTAB

concentration of 0.05 M was used which is safely above the CMC of

0.0013 M at 25C (30). Maximum absorbance was found to be at 355 nm

for the pyridoxal-cyanide reaction product for both aqueous and

micellar media. In agreement with the expected results, the cationic

surfactant, CTAB, promotes the rate of the reaction. Greater changes

in absorbance were absorbed in micellar media during the same amount

of time. This increase in the rate of the reaction can be explained

by electrostatic attraction forces between the positive charge at the

micelle surface and the negative charge of the cyanide. Also, the

hydrophobic forces between the pyridoxal molecule and the nonpolar

portion of the micelle caused the reaction to proceed at faster

rate. Not only the solubilization and proximity effects contribute

(24,43) to micellar catalysis, but probably the stabilization of some

intermediate species with partial negative charge at the positive

micelle surface favored the formation of 4-pyridoxolactone.

Kinetics studies were performed to measure the rate constants for

this oxidation reaction in aqueous and 0.05 M CTAB media. Figures 10

and 11 show the curves of the change in absorbance versus time at a

maximum absorbance wavelength of 355 nm. For micellar media, the

reaction reached the plateau of the curve after 15 minutes, whereas in

aqueous media the plateau was reached after 30 minutes.






29









E
o- 0
x
0 0




0"0 E0
L u
X




Sr--
o a


oo













Soo
.- a,_



c






C C









O >












c fl3 Dl




7- 0 0 0
eS 0. 04
06 C




























33NV88OS9V ( o s
0















*O
C*
C*





U-

0*L




4V-
0 00










CO




rI-
Ij.c
















4- O
0
CE
o u
0
0 4_--

*r- S




0.0
C0
E--

x
SLO




o .,
ca



I I I / 0 Q.





0\ 0 T 00 0

o / i 0 0.1
E









,CD

.OC
( 0 =
U) *m

0J to

O X
C0 0. 5 0



C tj R Ln 0
*r /o















V "-0 40
SCc<
U-



1 1 0aS








S.0.


o en
*Ze
0cc
33..C S.. S..
4. a 4.
c)'-
C -

















































































0o 0 d 0







33NV80S8V


ooz


001







8


} Z


S-
o r
4- L







r-
0- 0
.C =















CL
S
4) 0
CO
0)



4-
-C















Eo
.9- 3









U 0 L
S-c T




LO
CL P
0
e) 0
CL
c
(o 4-)
0 CX m
S- *
- L


0r U
E U

.4-



0)oe
0 *4
XC -4
cr B
to *r-
J3 r
'r Pg





C <0












































































o o U) 0
it rf (n N N(
d o d 0 o


33NV8 0S8V


0





aL '-o -
09- 30

E*
i 0

LZ







OI *- (U
4-
n


0 B
m













41 0
I0








E *r- 0.

CA e



v 0






304--
0o
0
C > Q.










E 0-


cm4
2 o-

0 u

*1- 0 (
0 U



C C0 *
C









OS z









Assuming the reaction is pseudo first order, the rate constant

can be calculated by using the following equation:


-kt
log (A,-At) = 2.303 + log (A,-Ao) (eq. 4.1)


where A., Ao, At are the absorbance at infinite, initial and time t,

respectively; t is time in minutes; and k is the rate constant in

minutes-1 (51). From the slope of the curve (taking the negative and

multiplied by 2.303) the rate constants for the reaction of pyridoxal

taking place in water and 0.05 M CTAB media were 0.0490 and 0.0971

minutes-' (see Figures 12 and 13). The ratio of these rate constants,

kO.05M CTAB/kH20, is equal to 1.98. This value means that the
reaction is taking place at double the velocity in micellar CTAB than

in H20. Micellar catalysis does occur for this particular reaction.

To demonstrate the advantages of combining the technique of FIA

with micellar catalysis, measurements of the pyridoxal-cyanide

reaction product using absorbance and fluorescence detection were

carried out. Figure 14 shows the calibration plot for the pyridoxal

determination in aqueous and 0.05 M CTAB media. Comparing the slope

of both curves, 2.54X10-3 and 3.26X10-3 absorbance units/ppm of

pyridoxal, one can observe that by using a CTAB micellar media the

sensitivity of the system is increased 1.3 times. The differences

between the ratio of rate constants found for water and 0.05 M CTAB,

1.98, and the ratio of sensitivities of these two systems by FIA, 1.3,

may be explained by the fact that the kinetics throughout the entire

sample plug is not constant. The work of Paiton and Mottola (37)






































-I


I I I I I I I I I I I I L I I


( V -9SLZ') 60o


In f~Oc)


0 0.



4 -
04

CA
a) 0



a -
U0)
C 4-





0
4,,
-c




Q.




4-
c f



V) 0o














OS
E E


ac
41 c
S- C
E


4 .
(D I-0
cu







**
OS
4-)





u,
Le











0(

0 2

S -



-4
a)m
s- x ro-













*-

Lj.






35






-'o


0



Cud


o* 0




Emn
c S







E)
Q
LE

00
/ o
/ --" 0





4- "
>, 0
c0 o







Qc



y N ox E
E
eCL


C n M:


S0 -
S04-
















D M CL .
V ) r. W



















S 5-










LL
/ s i- <1
/ Ld S "
y 2 ^g
H -^- I-
/ *- c
/
/ rO n O rl

















O



*a



*r
03

O


00
4-)
E-














E
O
O M.



e*u
o 0

r' O



EaE







0 *o
Ln 0






o-
\A3 4-









\ \ <0
s-d
0 C



0 0 00 0
















(oaom) stiufl ouoqiosqV
\. USC
\ \ s- "
\ \ -Q -0 S



^P( C. +-
\\ \\aoM sCu Efuqog *c









demonstrated that the assumption of having a constant rate coefficient

throughout the entire body of the sample plug is invalid. The

kinetics involved within the sample plug are more complex. Paiton and

Mottola suggested that the rate coefficient changes with time. This

has been rationalized by assuming that each fluctuation in rate

coefficient with time corresponds with one of three regions within a

sample plug, namely, the leading region, the central region, and the

trailing region. In both the leading and trailing regions, the

carrier/sample interfaces induce molecular diffusion, while the

velocity profile induces convection. In the central region, where no

sample/carrier boundary exists, convection becomes the primary

dispersion force. Because the physical dispersion in these three

regions of the sample plug differ from one another, the rate

coefficients along the length of the plug are expected to vary with a

wave pattern. The fact that the reaction rate varies throughout the

sample plug may mean also that the kinetic order is not constant

within the sample plug.

Figure 15 shows the calibration plot with a fluorescence detector

for aqueous and micellar systems. From the slope of the curves, the

calculated sensitivity of the pyridoxal system is three times greater

when using CTAB micellar media compared with the same system in

aqueous media. The sensitivities for micellar and aqueous solutions

were 0.294 and 0.102 cm (peak height) per ppm of pyridoxal,

respectively. A greater change in the ratio of sensitivities was

observed for fluorescence determination due to the fact that not only

micellar catalysis was taking place but fluorescence enhancement was









40






30






..20
w E

O.i5&
z ,-o

OG
S

,, Io


0
0





Figure 15.


100 200


300


Pyridoxol (ppm)


Fluorescence calibration plots for pyridoxal determination
in aqueous ( E ) and 0.05 M CTAB ( ) systems. Flow rate
1.4 ml/min; temperature 45C; phosphate buffer (0.6 M); pH
7.3; KCN 1.5X10' M; 355 and 435 nm excitation and
emission wavelengths, respectively.


40






"30






E
I.-"

Wa
Oe-
S gC
-10 U


1








also observed. In micellar-enhanced fluorescence, the emission

intensity of the analyte is usually many times greater than in the

corresponding homogeneous media (52,53). This increase in sensitivity

of solutes in solutions containing micellar aggregates has been

explained by the diminution of deactivation processes for the excited

states. These phenomena occur due to a decrease in polarity, and an

increase in viscosity and shielding against quenching in micellar

media (29).

Table I summarizes the analytical figures of merit for both

aqueous and micellar systems. Very good coefficients of correlation

were obtained for the four curves, all being 0.999. These calibration

curves recorded under working conditions are linear over a wide range

of concentrations. The linear range for the aqueous system with a

fluorescence detector was found to be from 0.42 ng to 2.0X103 ng and

from 94 ng to 2.0X103 ng of pyridoxal for absorbance measurement. For

0.05 M CTAB media the linear dynamic range was from 0.17 ng to 1.1X103

ng of pyridoxal for fluorescence and from 77 ng to 2.0X103 ng of

pyridoxal for absorbance. Due to the large concentration range for

the recorded curves, it was necessary to work at several values of the

instrument sensitivity. At higher concentrations of pyridoxal, the

fluorescence intensity is beyond the spectrofluorimeter range. The

reproducibility of the system was measured by manual injection of 11

replicates of pyridoxal solution at 25.53 ppm. Relative standard

deviation of peak height in percent was calculated and was found to

vary between 0.97 and 3.25 for the studied systems. Limits of

detection were found to be lower for the reaction taking place in










Table I. Figures of merit for pyridoxal determination. Fluorescence
detector: excitation 355 nm, emission 435 nm, range 500, or
variable UV-visible absorbance detector at 355 nm, range .1,
1.5X10" M cyanide, pH 7.3, phosphate buffer 0.6 M, flow
rate 1.4 ml/min, temperature 30C, chart speed 1 cm/min,
10 pl sample loop, pyridoxal in deionized water, tube length
200 cm, i.d. 0.5 mm.


Fluorescence Absorbance

Aqueous 0.05 M CTAB Aqueous 0.05 M CTAB

Sensitivity 0.102 0.294 2.54X10-3 3.26X10-3

Coefficient of
correlation 0.9996 0.9992 0.9999 0.9999

Limits of
detection (ng) 0.42 0.17 94 77

Relative standard
deviation (%) 2.06 3.25 0.97 1.76

a Slope of the calibration plot for fluorescence, cm/ppm of pyridoxal
and absorbance units/ppm of pyridoxal for absorbance measurements.
b Eleven determinations at 25.53 ppm of pyridoxal.









micellar solutions. A more significant change in limits of detection

was not obtained due to an increase in the background signal for

micellar solutions. This needs further study in the future the

examination of other systems in micellar media by FIA, since a

significant change in limits of detection usually accompanies micellar

catalysis (52).

Optimum conditions may be different for aqueous and micellar

systems. The reason for this difference is that micelles can change

the microenvironment of the solubilized molecules (22). Then, an

optimization of conditions for FIA system is required.

The output signal, in this case peak height, is influenced by the

dispersion of the sample in the reagent stream and the degree of

completeness of the reaction taking place. These two are affected by

experimental parameters such as flow rate, reagent concentrations,

length of the reaction coil, etc. Since these experimental parameters

interact with each other, optimization of FIA methods using univariate

design (optimization of every parameter by separate studies) is time

consuming and may be inadequate to determine the best set of experi-

mental conditions (54). Sequential simplex optimization procedures

have been found to be valuable in development of new FIA methods

(54-56).

A modified variable size simplex method (57,58) was used for

optimization of pyridoxal determination. The optimization of this

system was performed by changing four variables: pH, temperature,

flow rate and surfactant concentration. Table II shows the initial,

final and increment values for each changing parameter. The reason to











Table II. Variable parameters for the Modified Simplex Optimization
program.


Initial Range Increment
Parameter Value Value Value


1. pH 7.3 6.5-8.0 0.5

2. Flow rate (ml/min) 1.4 1.0-1.7 0.1

3. Temperature (C) 45 30-50 5

4. CTAB concentration (M) 0.05 0.05-0.15 0.05








include pH as one of the parameters to be optimized is that the

reaction of pyridoxal with cyanide is pH dependent. It is reported in

the literature that micelles affect the pH of solutions, by changing

acid dissociation constants (59). The range of pH from 6.5 to 8.0 was

chosen because previous studies showed this to be the optimal pH range

for the oxidation reaction. An increase in temperature increased the

reaction rate in such a manner that higher values for the response

function resulted, up to a certain point where the signal started to

decrease as the temperature increased due to deactivation of

fluorescence. At temperatures below 30C, the response of the system

(peak height) is greatly diminished; at temperatures above 500C,

bubble formation in the FIA system can prove detrimental to

reproducibility (47). Flow rate is closely related with the output

signal videe supra). Faster flow rates will usually lead to a

decrease in signal because less time is available for the reaction to

take place and vice versa. The surfactant concentration was included

as a variable since increasing the number of micellar structures

increases the number of sites available for solute solubilization and,

therefore, promotes the formation of 4-pyridoxolactone. If the

surfactant concentration is increased too much, the reaction will

occur at a lower rate due to the dilution factor. In other words, as

the number of micelle structures increases, solute molecules will be

solubilized on different micelle structures apart from each other.

There will be a physical impediment for a pyridoxal molecule to

encounter a cyanide ion preventing the oxidation reaction to

proceed. The cyanide concentration and the length of reaction coil








were kept constant. At a cyanide concentration of fivefold the

pyridoxal concentration, the intensity of the output signal is not

influenced by a change in the cyanide concentration (47). The simplex

was finished after 18 experiments (one reflection, two expansions and

four contractions). Optimum values found, together with those

variables kept constant throughout the development of the simplex, are

summarized in Table III.

With these new optimum values for absorbance measurement in CTAB

media, a set of standard pyridoxal solutions were run and a

calibration plot was recorded and compared with the calibration plot

for the water system run with previous conditions (see Figure 16). A

ratio of 1.8 was found, by comparing the slopes of the two curves, for

water and CTAB, 2.09X10-3 and 3.84X10-3 absorbance units per parts per

million of pyridoxal. As was expected, a higher sensitivity was

observed. This can be explained by the fact that now the conditions

are more favorable for the reaction to take place. Table IV shows the

analytical figures of merit for the reaction in optimized CTAB

conditions and previous conditions for aqueous system. Relative

standard deviation, measured by 11 determinations of 24.14 ppm of

pyridoxal, were 1.06 and 2.12% for CTAB and water systems. Very good

coefficients of correlation were observed for both curves, 0.999,

assuring the linearity of the curve. Lower limits of detection were

obtained by using 0.09 M CTAB and 49C, 64 ng, compared with the

aqueous system, 86 ng.

Figures 17 to 21 show a set of pyridoxal standard solutions run

by FIA under different studied conditions.










Table III. Optimized and fixed variables for the determination of
pyridoxal.

Variables Fixed
in the Optimization Optimized Variables

1. 10 ul sample: Pyridoxal: 1. pH: 6.74
9.18X10-4 M in 0.05 M CTAB


2. Cyanide concentration: 2. Flow rate: 1.3 ml/min
1.5X10-2 M


3. Phosphate buffer solution: 0.6 M 3. Temperature: 49C


4. Length of reaction coil: 200 cm 4. CTAB concentration: 0.09 M









.50-




.40




.30-




.20




.10-




0


0 20


40 60 80 100 120


PYR IDOXAL


(ppm)


Figure 16.


Absorbance calibration plots for pyridoxal in aqueous
( 0 ) and 0.09 M CTAB micellar ( a ) media. Absorbance
measured at 355 nm, flow rate 1.3 ml/min, temperature
49C, pH 6.74, phosphate buffer (0.6 M), KCN 1.5X10- M.


I a I I I I


.50




.40


.30




.20


.10












Table IV.


Figures of merit for pyridoxal determination with
conditions obtained by Modified Simplex program. Variable
UV-visible ab orbance detector at 355 nm, range .1, 0.09 M
CTAB, 1.5X10- M cyanide, pH 6.74, 0.6 M phosphate buffer,
flow rate 1.3 ml/min, temperature 49C, chart speed
1 cm/min, 10 ul sample loop, pyridoxal in deionized water,
tube length 200 cm, i.d. 0.5 mm.


Aqueous


CTAB


Sensitivity 2.09X10-3 3.84X10-3
absorbance units
ppm of pyridoxal

Coefficient of correlation 0.9994 0.9998

Limits of detection (ng) 86 64

Relative standard deviation (%)a 2.12 1.06

a Eleven determinations at 24.14 ppm of pyridoxal.


_ ~ ~





















4

d




z
a 0.05 c

0, b
0
C, b








TIME (min)

Figure 17. Absorbance recordings of a series of pyridoxal standards
in aqueous media. Pyridoxal: (a) 8.10 ppm, (b) 24.31
ppm, (c) 40.52 ppm, (d) 81.04 ppm, (e) 105.4 ppm.
Absorbance measured at 355 nm, range 0.05, sample volume
10 u1, all tubes 0.5 mm I.D., flow rate 1.4 ml/min,
tempera ure 45C, phosphate buffer 0.6 M, pH 7.3, KCN
1.5X10" M, chart speed 0.25 cm/min.





















4








S0.05


b
a


TIME (min)


Figure 18.


Absorbance recordings of a series of pyridoxal standards
in 0.05 M CTAB micellar media. Pyridoxal: (a) 8.10 ppm,
(b) 24.31 ppm, (c) 40.52 ppm, (d) 81.04 ppm, (e) 105.4
ppm. Absorbance measured at 355 nm, range 0.05, sample
volume 10 u1, all tubes 0.5 mm 1.0., flow rate 1.4 ml/min,
temperature 45C, phosphate buffer 0.6 M, pH 7.3, KCN
1.5X10- M, chart speed 0.25 cm/min.

























Figure 19.


Absorbance recordings of a series of pyridoxal standards
in 0.09 M CTAB micellar media. Pyridoxal: (a) 8.10 ppm,
(b) 24.31 ppm, (c) 40.52 ppm, (d) 81.04 ppm, (e) 105.4
ppm. Absorbance measured at 355 nm, range 0.05, sample
volume 10 ul, all tubes 0.5 mm I.D., flow rate 1.3 ml/min,
temperature 49C, phosphate buffer 0.6 M, pH 6.73, KCN
1.5X10- M, chart speed 0.25 cm/min.





TIME (min)


4
l*-*


0.05


j















E
o




-





0
z
U




U>



0
LrJ







U-
0

IJ
CO
UJ




.L_


TIME (min)


Figure 20.


Fluorescence recordings of a series of pyridoxal standards
in aqueous media. Pyridoxal: (a) 8.10 ppm, (b) 24.31
ppm, (c) 40.52 ppm, (d) 81.04 ppm, (e) 105.4 ppm. Sample
volume 10 ul; all tubes 0.5 mm I.D.; flow rate 1.4 ml/min;
temperature 45C; phosphate buffer 0.6 M; pH 7.4; KCN
1.5X10- M; range 1000; chart speed 0.25 cm/min; 355 nm
and 435 nm excitation and emission wavelengths,
respectively.


e


4

d








C
























Figure 21.


Fluorescence recordings of a series of pyridoxal standards
in 0.05 M CTAB micellar media. Pyridoxal: (a) 8.10 ppm,
(b) 24.31 ppm, (c) 40.52 ppm, (d) 81.04 ppm. Sample
volume 10 1l; all tubes 0.5 mm I.D.; flow rate 1.4 ml/min;
temperature 45C; phosphate buffer 0.6 M; pH 7.4; KCN
1.5X10" M; 355 nm and 435 nm excitation and emission
wavelengths, respectively; range 1000; chart speed 0.25
cm/min.
















E


o
0


S4 C





~ (b
w



0

0
I-J








TIME (min)









Measurements of Dispersion

In flow injection analysis, both sample throughput and sample

dilution are directly related to dispersion. The dispersion process

which takes place during the transport of the sample from the

injection device toward the detector is one of the less understood

aspects of FIA videe supra). In analogy with chromatographic systems,

Poppe (60) observed that the total peak broadening in FIA is the sum

of the contributions from the injection process, the flow through

reactors and connectors, the holdup volume of the flow through

detector, and the time constants of associated electronics. These

processes can be described by the individual peak variances:


a2 = a2 + a2 C2
overall injection flow detector


Provided that the detector and electronics are well designed, the

variance of detection may be neglected (if it is at least five times

smaller than the standard deviations due to injection and

transport). One of the models frequently used to describe the

dispersion process is the tank in series model. According to this

model, the flow reactor can be considered as a series of N ideal

mixers. If the number of mixing stages, N, is sufficiently high, the

resulting curve has a Gaussian shape. However, in FIA, peaks mostly

show a tailing character (1,42). In 1981, Reijn and coworkers (33)

described the distribution curve of an FIA to be a modified Gaussian

function. Note that only the physical aspects of dispersion are taken

into consideration. It was assumed that there is no contribution to








dispersion due to chemical reaction between the sample and reagent

stream. This assumption, however, is known to be invalid (38,42).

For our system, the FIA peaks were examined for their resemblance

to Gaussian, exponentially modified Gaussian and other peak shapes by

measurement of the variance, M2 and the standard deviation of the

parent Gaussian function, aG. Foley and Dorsey (61) demonstrated that

the type of peak shape can be assigned by the agreement of the values

M2 and aG determined from the asymmetry factor, B/A, and peak width,

W, at 10, 30, and 50% peak height (see Figure 22). Equations 4.2 to

4.7 were used to calculate M2 and oG at different peak heights.


M2 =
2 1.764(B/A)2
0.1


W2
0.1

- 11.15(B/A) .1 + 28


2
0.3
3 2
-3.85(/A)0.3 + 23(B/A)0.3 47.9(3/A)0.3 + 38.7


W2
0.5
2 -8.28(B/A) + 41.8(8/A)
0.5 0.5


- 72.3(B/A).5 + 44.5


W0.1
G =3.27(B/A)0.1 + 1.2



W0.3
OG =2.8(B/A)0.3 + 0.48


M =


(eq. 4.2)


(eq. 4.3)


(eq. 4.4)


(eq. 4.5)


(eq. 4.6)

















































Figure 22.


A Wo.s Ao3+ Bo.s




A B Wo.s = A 1 + Bo.3








TIME




Measurement of peak width, W, and asymmetry factor, B/A,
at 10, 30 and 50% peak height in an FIA peak.








0a.5
and G 2.5(/A)5 (eq. 4.7)




Tables V to X show the calculated values for the variance and

standard deviation for pyridoxal in aqueous and micellar media at

different peak heights. For this study, the variable wavelength

absorbance detector was set at 292.6 nm, the maximum absorbance

wavelength for pyridoxal. Triplicate injections of 10 ul of

pyridoxal, 2.81X10-4 M in aqueous solution, were made into the aqueous

stream and peaks were recorded. Manual measurement of peak height, h,

peak width and asymmetry factor at 10, 30 and 50% peak height were

done for each peak and substitution of these values into equations 4.2

to 4.7 gave M2 and aG. Averages of three values were calculated

for M2 and aG at each peak height. The same procedure was followed

for 2.83X10-4 M pyridoxal in 0.05 M CTAB injected into 0.05 M CTAB

stream. Table XI shows the average values for A2 and oG at 10, 30 and

50% peak height. The relative standard deviation of the values from

the three peak heights was calculated for pyridoxal in the aqueous

system to be 16.06% for M2 and 4.88% for aG. For 0.05 M CTAB the

relative standard deviation was 5.68% for M2 and 5.20% for oG.

According to the agreement of the values calculated at the three

heights, these peaks can be classified as exponentially modified

Gaussians. Foley and Dorsey (61) reported a relative standard

deviation of *7.7% for M2 or *3.2% for oG values calculated at the

three heights was necessary to ensure the validity of the











Table V. Variance and standard deviation values for pyridoxal in
aqueous media at 10% peak height. Variable UV-visible
absorbance detector at 292.6 nm, range .1, temperature 45"C,
flow rate 1.4 ml/min, tube length 200 cm, i.d. 0.5 mm, 10 il
sample loop, pyridoxal 2.81X10' M, chart speed 10 cm/min.


h (Peak height in cm)

W0.1 (Width at 10% peak height in cm)

A

B

B/A (Asymmetry ratio)

TM2 (Variance in cm2)a


14.15

2.70

1.09

1.61

1.48

0.47


14.35

2.70

1.10

1.60

1.45

0.47


14.15

2.70

1.08

1.62

1.50

0.48


S= 0.473 cm2

RSDb = 1.48%


aG (Standard deviation in cm2)


0.447


0.454


0.442


X = 0.448 cm2

RSD = 1.35%


a 1 cm2 = 36 s2


b Relative standard deviation.











Table VI.


Variance and standard deviation values for pyridoxal in
aqueous media at 30% peak height. Variable UV-visible
absorbance detector at 292.6 nm, range .1, temperature
45"C, flow rate 1.4 ml/min, tube length 200 cm, i.d.
0.5 mm, 10 ul sample loop, pyridoxal 2.81X10- M, chart
speed 10 cm/min.


h (Peak height in cm)

WO.3 (Width at 30% peak height in cm)

A

B

B/A (Asymmetry ratio)

M2 (Variance in cm2)a


7 = 0.563 cm2

RSDb = 2.05%



aG (Standard deviation in cm2)

j = 0.448 cm2

RSD = 4.12%


14.15

1.90

0.80

1.10

1.38

0.57


0.437


14.35

2.00

0.85

1.15

1.35

0.55


0.469


a 1 cm2 = 36 s2.


b Relative standard deviation.


14.15

1.90

0.80

1.10

1.38

0.57


0.437











Table VII.


Variance and standard deviation values for pyridoxal in
aqueous media at 50% peak height. Variable UV-visible
absorbance detector at 292.6 nm, range .1, temperature
45*C, flow rate 1.4 ml/min, tube length 200 ct, i.d.
0.5 mm, 10 u1 sample loop, pyridoxal 2.81X10- M, chart
speed 10 cm/min.


h (Peak height in cm)

WO.5 (Width at 50% peak height in cm)
A

B

B/A (Asymmetry ratio)

M2 (Variance in cm2)a


T = 0.47 cm2

RSDb = 0%



oG (Standard deviation in cm2)

X = 0.487 cm2

RSD = 0%


a 1 cm2 = 36 s2


b Relative standard deviation.


14.15

1.40

0.65

0.75

1.15

0.47


0.487


14.35

1.40

0.65

0.75

1.15

0.47


0.487


14.15

1.40

0.65

0.75

1.15

0.47


0.487











Table VIII.


Variance and standard deviation values for pyridoxal in
0.05 M CTAB micellar media at 10% peak height. Variable
UV-visible absorbance detector at 292.6 nm, range .1,
temperature 45C, flow rate 1.4 ml/min, tube length
200 cm, j.d. 0.5 mm, 10 ul sample loop, pyridoxal
2.83X10" M in 0.05 M CTAB, chart speed 10 cm/min.


h (Peak height in cm)

WO.1 (Width at 10% peak height in cm)

A

B

B/A (Asymmetry ratio)

M2 (Variance in cm2)a


7 = 0.55 cm2

RSDb = 0%



CG (Standard deviation in cm2)

X = 0.462 cm2

RSD = 3.12%


a 1 cm2 = 36 s2.


b Relative standard deviation.


13.60

2.85

1.10

1.75

1.59

0.55


0.445


13.50

2.90

1.15

1.75

1.59

0.55


0.470


13.70

2.90

1.15

1.75

1.59

0.55


0.470











Variance and standard deviation values for pyridoxal in
0.05 M CTAB micellar media at 30% peak height. Variable
UV-visible absorbance detector at 292.6 nm, range .1,
temperature 450C, flow rate 1.4 ml/min, tube length 200 cm,
i.d. 0.5 mm, 10 l1 sample loop, pyridoxal 2.83X10- M in


0.05 M CTAB,


chart speed 10 cm/min.


h (Peak height in cm)

WO.3 (Width at 30% peak height in cm)
A

B

B/A (Asymmetry ratio)

'M2 (Variance in cm2)a


j = 0.58 cm2

RSDb = 0%



aG (Standard deviation in cm2)

i = 0.543 cm2

RSD = 0%


a 1 cm2 = 36 s2


b Relative standard deviation.


Table IX.


13.60

2.10

0.95

1.15

1.21

0.58


0.543


13.50

2.10

0.95

1.15

1.21

0.58


0.543


13.70

2.10

0.95

1.15

1.21

0.58


0.543










Table X. Variance and standard deviation values for pyridoxal in
0.05 M CTAB micellar media at 50% peak height. Variable UV-
visible absorbance detector at 292.6 nm, range .1,
temperature 45C, flow rate 1.4 ml/min, tube length 200 cm,
i.d. 0.5 mm, 10 l1 sample loop, pyridoxal 2.83X10" M in


0.05 M CTAB, chart


speed 10 cm/min.


h (Peak height in cm)

WO.5 (Width at 50% peak height in cm)
A

B

B/A (Asymmetry ratio)

M2 (Variance in cm2)a

> = 0.616 cm2

RSDb = 7.2%



aG (Standard deviation in cm2)

ST= 0.559 cm2

RSD = 2.36%


a 1 cm2 = 36 s2.


b Relative standard deviation.


13.60

1.60

0.75

0.80

1.13

0.59


0.566


13.50

1.70

0.80

0.90

1.125

0.667


0.544


13.70

1.60

0.75

0.85

1.13

0.59


0.566










Table XI.


Average values for variance and standard deviation for
pyridoxal in 0.05 M CTAB micellar and aqueous media at 10,
30, and 50% peak height for aqueous and 0.05 M CTAB
systems.


X RSD
10% 30% 50% (cm2) (%)

0.05 M CTAB

M2 (cm2) 0.55 0.58 0.616 0.582 5.68

oG (cm2) 0.462 0.559 0.543 0.521 5.20


H20


M2 (cm2) 0.473 0.559 0.473 0.503 6.06

aG (cm2) 0.448 0.448 0.487 0.461 4.88









exponentially modified Gaussian peak shape. Proven that the FIA peaks

in our system were exponentially modified Gaussian, studies of the

dispersion in aqueous and micellar system were performed by using

equation 4.2. Variance measurements at 10% peak height are more

precise than those calculated at 30% and 50% peak height (61).

Given the conditions in Tables V and VIII, the dispersion between

aqueous and micellar media was compared in terms of the variance,

M2. The values for M2 were found to be 0.473 and 0.55 cm2 for

aqueous and 0.05 M CTAB media, respectively.

The dispersion of both aqueous and micellar systems was

calculated also by using equation 1.1, which is the definition of FIA

dispersion. The concentration of pyridoxal before the dispersion

process was measured by passing through the detector a solution of

pyridoxal in water (2.81X10-4 M) or in 0.05 M CTAB (2.83X10-4 M) and

measuring the response of the detector. The response of the detector

was obtained by the average of six measurements. The concentration of

pyridoxal after the dispersion process has taken place was obtained by

measuring the maximum peak height of 11 injections of 10 pl of

pyridoxal solutions (above) into pure aqueous or 0.05 M CTAB. Note

again that dispersion occurring under this circumstance is only due to

physical contributions; no chemical reaction is taking place. The

ratio of the detector response before and after the dispersion process

occurred was 17.62 for the aqueous system and 19.14 for the micellar

system. Contrary to what was expected higher dispersion was found for

micellar media. Due to the higher viscosity of 0.05 M CTAB solution

compared with water, the dispersion was expected to be less. To prove








the validity of these results, the same type of experiment was

performed on two different days and the same results were obtained

(see Table XII).

To investigate the effect of surfactant concentration on the

dispersion process, a set of dispersion measurements (M2) versus CTAB

concentration were performed. The concentration of CTAB was varied

from 5X10-6 M to 5X10-2 M. This range of concentration includes

solutions above and below the critical micelle concentration. Ten

microliters of pyridoxal (2.81X10-4 M) aqueous solution was injected

into the stream and from the recorded peaks the variance at 10% peak

height was calculated. From Table XIII, one can say that for a

solution of 5X10-6 M the dispersion obtained is very similar to that

calculated for water. Above this concentration an increase in

surfactant concentration had no effect on the dispersion up to a

concentration of 5X10-2 M CTAB, at which point an increase in

dispersion was observed. Micellar concentration changes the

dispersion of the FIA system. This may be explained by the fact that

the pyridoxal molecules are localized on or within the micelle

structure. Micellar media is more viscous than aqueous media (43);

therefore, the mass transfer in a radial direction decreases. An

increase in peak dispersion results because a decrease in mixing

across the stream tends to increase dilution of the solute by

longitudinal dispersion (42). The increase of dispersion with

micellar media needs further study.












Table XII. Dispersion values for aqueous and micellar systems.


0.05 M CTAB Aqueous

M2 M2
C/Cmax (cm2) C/Cmax (cm2)

1. 19.54 0.55 17.62 0.47

2. 19.14 0.56 17.05 0.45

3. 19.26 0.54 17.25 0.45

C/Cmax = ratio of concentration of pyridoxal before and after the
dispersion process has taken place.

T2 = variance or second moment in cm2,











Table XIII. Measurement of dispersion versus CTAB concentration.
Variable UV-visible absorbance detector at 292.6 nm,
range .1, temperature 45"C, flow rate 1.4 ml/min, tube
length 2Q0 cm, i.d. 0.5 mm, 10 ul sample loop, pyridoxal
2.81X10 M in deionized water, chart speed 10 cm/min.



2 M2 M2 M2 Average
(cm2) (cm2) (cm2) (cm2)

CTAB (M)

5.0X10-6 0.44 0.43 0.46 0.44
2.5X10-5 0.52 0.46 0.52 0.50
5.0X10-5 0.47 0.52 0.49 0.49
2.5X10-4 0.48 0.50 0.46 0.48
5.0X10-4 0.49 0.46 0.52 0.49
2.5X10-3 0.52 0.50 0.49 0.50
5.0X10-3 0.50 0.52 0.52 0.51
2.5X10-2 0.47 0.52 0.49 0.49
5.0X10-2 0.52 0.57 0.53 0.54

H20 0.43 0.44 0.48 0.45


M2 = variance or second moment in cm2













CHAPTER V
CONCLUSIONS AND FUTURE WORK


The applicability of combining the technique of Flow Injection

Analysis with micellar catalysis has been shown. The results obtained

from the determination of pyridoxal in micellar media compared to

those in aqueous media are promising. Higher response, as measured by

peak height, was recorded at all times for micellar carrier

solutions. Higher sensitivities and lower limits of detection were

obtained for the micellar system when the oxidation product of

pyridoxal and cyanide was detected either fluorimetrically or by UV

absorbance. Higher sensitivity ratios, for micellar to aqueous

systems, were obtained when using fluorescence detection. In this

case, not only is the reaction taking place at a faster rate but the

solubilization of reagents within the micelle structure contributes to

increase the signal due to the shielding effect.

With this example, it has been proven that due to the kinetic

nature of the FIA technique the use of micellar media can be very

advantageous. To support this investigation, the utility of using

micellar carrier solution in an FIA system, it will be necessary to

perform similar experiments by running different reactions taking

place in aqueous and micellar media. Keeping in mind the kind of

surfactant that will catalyze the specific reaction (anionic,








cationic, nonionic or zwitterionic), the rate of the reaction should

be measured and if it is favorable applied to FIA.

One of the reactions that will be very interesting to look at is

the determination of metals by their complexation with dyes in the

presence of micelle media (62-68). Some characteristics observed for

these metal-dye complexes in micellar media are an increase in molar

absorptivity and red shifts in the wavelength of maximum absorbance

(63). These substantial changes in the UV visible spectrometry of

these complexes together with the technique of FIA can be developed as

a new spectrophotometric method for determining micro amounts of metal

ions. The resulting method should be fast, easy and inexpensive.

Lower background signals were found for the aqueous systems

compared to CTAB micellar systems. Therefore, the limits of detection

for micellar media were not as low as expected. This increase in

background noise for micellar media needs further study to elucidate

if the increase in noise is due to the presence of micelles or if it

is just observed with this specific reaction.

Due to the specific interactions between micelles and solutes the

selectivity of a particular reaction can be increased, reducing the

amount and kind of interference. An investigation on this topic will

be very valuable especially for the determination of very small

amounts of analytes.

It has been mentioned that catalysis of organic and inorganic

reactions also occur in apolar media in the presence of reversed

micelles (19,69). Reversed micelles offer similar and at the same

time different characteristics from normal micelles. Future research









on the application of reversed micelles for reactions occurring in

apolar media in combination with the technique of FIA will be very

interesting.

According to the agreement of the values for the second moment

and the standard deviation measured at 10, 30 and 50% peak height, the

peaks recorded from an FIA system can be designated to be

exponentially modified gaussian.

Higher values for dispersion were found for micellar media.

These results were not as expected and need further study to

understand better the reason for this increase in dispersion. More

detailed experiments should be performed under different conditions,

e.g., varying the type and concentration of surfactant, temperature,

etc.

The contribution of a chemical reaction to the dispersion of a

peak in FIA is not well understood. By comparing the dispersion of

different reactions taking place in different types of surfactants to

the dispersion obtained in aqueous media could help to explain the

contribution of the chemistry to the total dispersion process. The

differences in rate of reaction between different micellar and aqueous

systems can help to explain this phenomena. By taking a reaction

whose rate of reaction varies in the presence of anionic, cationic,

nonionic and aqueous media, the contribution of chemical kinetics may

be clarified.














REFERENCES


1. Ruzicka, J.; Hansen, E.H. "Flow Injection Analysis"; Wiley-
Interscience: New York, 1981.

2. Ruzicka, J.; Hansen, E.H. Anal. Chim. Acta, 1986, 179, 1.

3. Rocks, B.F.; Riley, C. Clin. Chem., 1982, 23, 409.

4. Stewart, K.K. Anal. Chem., 1983, 55, 931A.

5. Ruzicka, J.; Hansen, E.H. Anal. Chim. Acta, 1979, 106, 207.

6. Ranger, C.B. Anal. Chem., 1981, 53, 20A.

7. Ruzicka, J. Anal. Chem., 1983, 55, 1040A.

8. Valcircel Cases, M.; Luque de Castro, M.O. "Analisis por
Inyecci6n en Flujo"; Departamento de Qufmica Analftica,
Universidad de C6rdova, C6rdoba, 1984.

9. Mottola, H.A. Anal. Chem., 1981, 53, 1312A.

10. Stewart, K.K. Talanta, 1981, 28, 789.

11. Stewart, K.K.; Beecher, G.R.; Hare, P.E. Anal. Biochem., 1976,
70, 167.

12. Ruzicka, J.; Hansen, E.H. Anal. Chim. Acta, 1975, 78, 145.

13. Vanderslice, J.T.; Stewart, K.K.; Rosenfeld, A.G.; Higgs, D.J.
Talanta, 1981, 28, 11.

14. Reinj, J.M.; Van Der Linden, W.E.; Poppe, H. Anal. Chim. Acta,
1981, 123, 229.

15. Tijssen, R. Anal. Chim. Acta, 1980, 114, 71.

16. Van Den Berg, J.H.M.; Deeler, R.S.; Egberink, H.G.M. Anal. Chim.
Acta, 1980, 114, 91.

17. Ruzicka, J.; Hansen, E.H. Anal. Chim. Acta, 1980, 114, 19.

18. Mittal, K.L., Ed. "Micellization, Solubilization, and
Microemulsions"; Plenum Press: New York, 1977; vols. 1 and 2.










19. Lindman, B.; Wennerstrom, H.; Eicke, H.F. "Micelles"; Topics in
Current Chemistry No. 87; Springer-Verlag: New York, 1980.

20. Tanford, C. "The Hydrophobic Effect"; 2nd ed.; Wiley-
Interscience: New York, 1980.

21. Rossen, M.J. "Surfactants and Interfacial Phenomena"; Wiley-
Interscience: New York, 1978.

22. Mittal, K.L.; Fendler, E.J. "Solution Behavior of Surfactants";
Plenum Press: New York, 1982; vols. 1 and 2.

23. Fendler, J.H. "Membrane Mimetic Chemistry"; John Wiley: New
York, 1982.

24. Fendler, J.H.; Fendler, E.J. "Catalysis in Micellar and
Micromolecular Systems"; Academic Press: New York, 1975.

25. Dill, K.A.; Flory, P.J. Proc. Natl. Acad. Sci. USA, 1981, 78,
676.

26. Berezin, I.V.; Martinek, K.; Yatsimirskii, A.K. Russ. Chem.
Rev., 1973, 42, 787.

27. Menger, F.M. Acc. Chem. Res., 1979, 12, 111.

28. Fisher, L.R.; Oakenfull, D.G. Chem. Soc. Rev., 1977, 6, 25.

29. Pelizzetti, E.; Pramauro, E. Anal. Chim. Acta, 1985, 169, 1.

30. Cline-Love, L.J.; Habarta, J.G.; Dorsey, J.G. Anal. Chem., 1984,
56, 1133A.

31. Hinze, W.L. In "Solution Chemistry of Surfactants"; Mittal,
K.L., Ed.; Plenum Press: New York, 1979; vol. 1, p. 79.

32. Betteridge, D. Anal. Chem., 1978, 50, 832A.

33. Reijn, J.M.; Van Der Linden, W.E.; Poppe, H. Anal. Chim. Acta,
1981, 126, 1.

34. Reijn, J.M.; Van Der Linden, W.E.; Poppe, H. Anal. Chim. Acta,
1980, 114, 105.

35. Reijn, J.M.; Van Der Linden, W.E.; Poppe, H. Anal. Chim. Acta,
1983, 145, 59.

36. Ruzicka, J.; Hansen, E.H. Anal. Chim. Acta, 1978, 99, 37.

37. Paiton, C.C.; Mottola, H.A. Anal. Chim. Acta, 1984, 158, 67.









38. Reijn, J.M.; Poppe, H.; Van Der Linden, W.E. Anal. Chem., 1984,
56, 943.

39. Hungerford, J.M.; Christan, G.D.; Ruzicka, J.; Middings, J.C.
Anal. Chem., 1985, 57, 1794.

40. Pardue, H.L.; Fields, B. Anal. Chim. Acta, 1981, 124, 39.

41. Pardue, H.L.; Fields, B. Anal. Chim. Acta, 1981, 124, 65.

42. Painton, C.C.; Mottola, H.A. Anal. Chem., 1981, 53, 1713.

43. Cordes, E.H.; Dunlap, R.B. Acc. Chem. Res., 1969, 2, 329.

44. Mittal, K.L.; Lindman, B., Ed. "Surfactant in Solutions"; Plenum
Press: New York, 1982; vols. 1 and 2.

45. Cordes, E., Ed. "Reaction Kinetics in Micelles"; Plenum Press:
New York, 1973.

46. Abraham, W.; Handler, P.; Smith, E.L.; Stetten, D.W. "Principles
of Biochemistry"; McGraw Hill Book Company, Inc.: New York,
1954; pp. 1008-1010.

47. Linares, P.; Luque de Castro, M.D.; Varcarcel, M. Anal. Lett.,
1985, 18, 67.

48. Bonavita, V. Arch. Biochem. Biophys., 1960, 88, 366.

49. Oshishi, N.; Fukui, S. Arch. Biochem. Biophys., 1968, 128, 606.

50. Linares, P.; Luque de Castro, M.D.; Varcarcel, M. Anal. Chim.
Acta, 1984, 18, 67.

51. Moore, J.W.; Pearsen, R.G. "Kinetics and Mechanism"; Wiley-
Interscience: New York, 1981.

52. Hinze, W.L.; Singh, H.N.; Baba, Y.; Harvey, N.G. Trends Anal.
Chem., 1984, 3, 193.

53. Turro, N.J.; Gratzel, M.; Braun, A.M. Angew. Chem. Int. Ed.
Engl., 1980, 19, 675.

54. Anagnostopoulen, P.I.; Koupparis, M.A. Anal. Chem., 1986, 58,
322.

55. Betteridge, D.; Sly, T.J.; Wade, A.P.; Tillman, J.E. Anal.
Chem., 1983, 55, 1292.

56. Burke, G.C.; Stedman, G.; Wade, A.P. Anal. Chim. Acta, 1983,
153, 277.








57. Legett, D.J. J. Chem. Ed., 1983, 60, 707.

58. Shavers, C.L.; Parsons, M.L.; Deming, S.N. J. Chem. Ed., 1979,
56, 307.

59. Buton, C.A.; Wolfe, B. J. Am. Chem. Soc., 1973, 95, 3742.

60. Poppe, H. Anal. Chim. Acta, 1980, 114, 59.

61. Foley, J.P.; Dorsey, J.G. Anal. Chem., 1983, 55, 730.

62. Diaz Garcia, M.E.; Sanz-Medel, A. Talanta, 1986, 33, 255.

63. Callahan, J.H.; Cook, K.D. Anal. Chem., 1984, 56, 1632.

64. Jarosz, M.; Marczenko, Z. Analyst, 1984, 109, 35.

65. Marczenko, Z.; Jarosz, M. Analyst, 1982, 107, 1431.

66. Marczenko, Z.; Kalowska, H. Anal. Chim. Acta, 1981, 123, 279.

67. Xi-Wen, H.; Poe, D.P. Anal. Chim. Acta, 1981, 131, 195.

68. Medina, J.; Hernandez, F.; Marin, R.; Lopez, F.J. Analyst, 1986,
111, 235.

69. Fendler, J.H. Acc. Chem. Res., 1976, 9, 153.













BIOGRAPHICAL SKETCH


Maria A. Hernindez Torres was born in Mayaguez, Puerto Rico, on

May 29, 1958. She received her elementary and high school education

at the Colegio de La Milagrosa, Mayaguez, Puerto Rico. In 1980, she

completed her Bachelor of Science degree in chemistry at the

University of Puerto Rico Mayaguez Campus Magnaa cum laude). In 1983,

she pursued a Master of Science degree in analytical chemistry at the

University of Florida. She is now receiving a Doctor of Philosophy

degree in analytical chemistry at the University of Florida.











I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.




JohrfdG. horsey, Chairman
Assciat Professor of Chemistr


I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.




Times D. Winefordne
graduate Research professor of Chemistry


I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.




Anna F. Brajter oth
Assistant Professor of Chemistry


I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.



Martin Vala
Professor of Chemistry










I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.




Chri eptr .--re y y
Assistant Profesor of Pharmacy




This dissertation was submitted to the Graduate Faculty of the
Department of Chemistry in the College of Liberal Arts and Sciences
and to the Graduate School and was accepted as partial fulfillment of
the requirements for the degree of Doctor of Philosophy.



August, 1986
Dean, Graduate School









































UNIVERSITY OF FLORIDA
IIIIIIII1il lll 1 I lll 112 I I0ll 1 111111111 t
3 1262 08554 1471