CHARACTERIZATION OF ION-SELECTIVEELECTRODES BY ELECTROCHEMICAL STUDIES OFION TRANSFER AT THE LIQUID/LIQUID INTERFACE

Item Type text; Thesis-Reproduction (electronic)

Authors Stevens, Anthony Clark, 1960-

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this materialis made possible by the University Libraries, University of Arizona.Further transmission, reproduction or presentation (such aspublic display or performance) of protected items is prohibitedexcept with permission of the author.

Download date 27/08/2021 21:55:53

Link to Item http://hdl.handle.net/10150/276364

INFORMATION TO USERS

This reproduction was made from a copy of a document sent to us for microfilming. While the most advanced technology has been used to photograph and reproduce this document, the quality of the reproduction is heavily dependent upon the quality of the material submitted.

The following explanation of techniques is provided to help clarify markings or notations which may appear on this reproduction.

1. The sign or "target" for pages apparently lacking from the document photographed is "Missing Page(s)". If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting through an image and duplicating adjacent pages to assure complete continuity.

2. When an image on the film is obliterated with a round black mark, it is an indication of either blurred copy because of movement during exposure, duplicate copy, or copyrighted materials that should not have been filmed. For blurred pages, a good image of the page can be found in the adjacent frame. If copyrighted materials were deleted, a target note will appear listing the pages in the adjacent frame.

3. When a map, drawing or chart, etc., is part of the material being photographed, a definite method of "sectioning" the material has been followed. It is customary to begin filming at the upper left hand corner of a large sheet and to continue from left to right in equal sections with small overlaps. If necessary, sectioning is continued again—beginning below the first row and continuing on until complete.

4. For illustrations that cannot be satisfactorily reproduced by xerographic means, photographic prints can be purchased at additional cost and inserted into your xerographic copy. These prints are available upon request from the Dissertations Customer Services Department.

5. Some pages in any document may have indistinct print. In all cases the best available copy has been filmed.

University Micrdnlrris

International 300 N. Zeeb Road Ann Arbor, Ml 48106

1329808

Stevens, Anthony Clark

CHARACTERIZATION OF ION-SELECTIVE ELECTRODES BY ELECTROCHEMICAL STUDIES OF ION TRANSFER AT THE LIQUID/LIQUID INTERFACE

The University of Arizona M.S. 1986

University Microfilms

International 300 N. Zeeb Road, Ann Arbor, Ml 48106

PLEASE NOTE:

In all cases this material has been filmed fn the best possible way from the available copy. Problems encountered with this document have been identified herewith a check mark V .

1. Glossy photographs or pages

2. Colored illustrations, paper or print

3. Photographs with dark background

4. Illustrations are poor copy

5. Pages with black marks, not original copy

6. Print shows through as there is text on both sides of page

7. Indistinct, broken or small print on several pages jX*

8. Print exceeds margin requirements

9. Tightly bound copy with print lost in spine

10. Computer printout pages with indistinct print

11. Page(s) lacking when material received, and not available from school or author.

12. Page(s) seem to be missing in numbering only as text follows.

13. Two pages numbered . Text follows.

14. Curling and wrinkled pages

15. Dissertation contains pages with print at a slant, filmed as received

16. Other

University Microfilms

International

CHARACTERIZATION OP ION-SELECTIVE ELECTRODES

BY ELECTROCHEMICAL STUDIES OP ION TRANSPER

AT THE LIQUID/LIQUID INTERFACE

by

Anthony Clark Stevens

A thesis submitted to the Faculty of the

DEPARTMENT OP CHEMISTRY

in Partial Fulfilment of the Requirements for the Degree of

MASTER OF SCIENCE

In the Graduate College

THE UNIVERSITY OF ARIZONA

1 9 8 6

STATEMENT BY AUTHOR

This thesis has been submitted in partial fulfilment of requirements for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this thesis are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his or her judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.

SIGNED:

APPROVAL BY THESIS DIRECTOR

This thesis has been approved on the date shown below

«. FREISER Date Professor of Chemistry

ACKNOWLEDGMENTS

I would like to express my gratitude to Professor

Henry Preiser for his encourgement and guidance, and the

faculty and staff in the Department of Chemistry at the

University of Arizona for their kind support during the

course of this work. I would also like to thank Dr.

Lawrence Cunningham for the development of the software and

instrumentation used.

I am especially grateful to my family for their love

and support.

iii

TABLE OF CONTENTS

Page

LIST OP ILLUSTRATIONS vi

LIST OF TABLES viii

ABSTRACT ix

1. INTRODUCTION 1

1.1 Polymer Membrane Ion-Selective Electrodes ... 2 1.1.1 Principles of PMISE Behavior 3 1.1.2 Neutral Carrier PMISEs 5 1.1.3 Coated-Wire Ion-Selective Electrodes . . 6

1.2 Electrochemistry at the Liqid/Liquid Interface ..... . 8

1.2.1 Principles of Electrochemistry at the Liquid/Liquid Interface 9

1.2.2 Electrocapillarity 11 1.3 Statement of Problem 14

2. EXPERIMENTAL 16

2.1 Materials 16 2.1.1 Reagents 16 2.1.2 Equipment 17

2.2 Procedures 23 2.2.1 Fabrication of CWISEs 23 2.2.2 Calibrations and Selectivity Studies . . 24 2.2.3 Current-Scan Polarography 25 2.2.4 Electrocapillary Measurements 28

3. RESULTS AND DISCUSSION 31

3.1 Characterization of a Cadmium PMISE using Current-Scan Polarography 31

3.1.1 Calibration Characteristics 31 3.1.2 Selectivity Characteristics 47 3.1.3 Cadmium Electrode Lifetime 58

3.2 Electrocapillary Studies of Neutral Surfactant Adsorption at the Liquid/Liquid Interface . . 70

3.2.1 Electrocapillary Measurements and Surface Excess Determinations .... 71

iv

V

TABLE OP CONTENTS—CONTINUED

3.2.2 Charge Density and Capacitance Data . . 77 3.2.3 The Effect of a Neutral Surfactant on

Facilitated Ion Transfer 82

4. CONCLUSION 85

APPENDIX A: DETERMINATION OF NONCOMPLEXED DNBP CONCENTRATION IN AN ISE POLYMER MEMBRANE 90

REFERENCES 91

LIST OP ILLUSTRATIONS

Page

Figure 1.1: Comparison of Moody-Thomas and Coated-wire ISE 7

Figure 2.1: Computer System for Calibration and Selectivity Studies for ISE 18

Figure 2.2: Computer System for Data Acquisition for AWE Experiments 21

Figure 2.3: Apparatus used for Current-Scan Polarography Experiments 22

Figure 2.4: Diagram of Electrodes used in Current-Scan Polarography ........... 26

Figure 2.5: List of Parameters Required for AWE Computer Program 29

Figure 3.1: Polarograms Showing Ion Transfer of Cd and Background Electrolyte 32

Figure 3.2: Schematic Representation of Ion Transfer Mechanism 34

Figure 3.3: Limiting Current Dependenc on DNBP Concentration 36

Figure 3.4: Plot Indicating Diffusion Controlled Kinetics for Cd-DNBP System 37

Figure 3.5: Plot Indicating Reversible Ion Transfer for Cd-DNBP System 39

Figure 3.6: Half-Wave Potential Dependence on Cd Concentration 41

Figure 3.7: Calibration Curves for Cd Electrodes with Initial Composition 43

Figure 3.8: Calibration Curves for Cd Electrodes with Altered Composition 44

vi

vii

LIST OF ILLUSTRATIONS—CONTINUED

Figure 3.9: Polarograms showing difference in Half-Wave Potential of Transfer for Potentiodynamic Selectivity Coefficient Determination 49

Figure 3.10: UV Spectrum of DNBP 61

Figure 3.11: tJV Spectrum of Cd-DNBP 62

Figure 3.12: UV Spectrum of Dissolved Cd Electrode Polymer Membrane 63

Figure 3.13: UV Spectral Analysis Calibration Curve for DNBP 64

Figure 3.14: UV Spectral Analysis Calibration Curve for Cd-DNBP 65

Figure 3.15: Plot Indicating Cd Loading in Polymer Membrane 67

Figure 3.16: Electrocapillary Curves for Triton X-100 72

Figure 3.17: Gibbs Plot for Surface Excess Determination of Triton X-100 74

Figure 3.18: Esin-Markov Type Plqt 76

Figure 3.19: Variation of Surface Excess with Charge Density 78

Figure 3.20: Charge Density Data for Triton X-100 80

Figure 3.21: Differential Capacitance Data for Triton X-100 81

Figure 3.22: Plot Showing Effect of Triton X-100 Concentration on Facilitated Ion Transfer 84

LIST OF TABLES Page

Table 3.1: Calibration Characteristics of Cadmium PMISE 46

Table 3.2: Selectivity Characteristics of Cadmium PMISE and Potentiodynamic Selectivity Coefficients 51

Table 3.3: The Results the Effect Plasticers have on Selectivity between Monovalent and Divalent Cations at the Liquid/Liquid Interface . . 56

Table 3.4: Results of UV Spectral Analysis on Cadmium PMISE Membranes 68

vii-i

ABSTRACT

A coated-wire PMISE based on the neutral carrier

dinonyl bipyridine (DNBP) was developed for the

determination of divalent cadmium. The PMISE yielded

nernstian responses, a dynamic range of three orders of

magnitude and a detection limit of 10"® M cadmium.

Selectivity coefficients were determined for Cu(II),

Zn(II), Co(II) and Ni(II). The short electrode lifetime was

shown by UV spectral analysis to be due to low free ligand

concentration in the membrane. Electrochemical studies of

Cd-DNBP ion transfer across the liquid/liquid interface were

used to examine the suitability of DNBP as a neutral carrier

for a PMISE. The ion transfer kinetics were determined to be

first order, diffusion controlled and reversible. The half-

wave potential showed a nernstian dependence with cadmium

concentration. Potentiodynamic selectivity coefficients

were determined and compared to the values. Ion

transfer kinetics were correlated to PMISE selectivty

characteristics for the first time.

The adsorption of Triton X-100 at the 1/1 interface

was investigated by electrocapillary measurements. Surface

excess, charge density and capacitance data was determined

for Triton X-100. The effect of Triton x-100 on facilitated

ion transfer was also investigated.

ix

CHAPTER 1

INTRODUCTION

In today's technological revolution, the demand for

chemical analysis has never been greater. To meet this

increasing demand, the development and understanding of

analytical disciplines is imperative. One of these

disciplines is the field of ion-selective electrode

potentiometry, which is important with respect to chemical

analysis for many reasons. It is a relatively simple

technique which measures the activity of a specific ion and

not the concentration. This becomes very beneficial in both

equilibrium and kinetic work. The technique is non­

destructive and does not affect the sample being measured in

any way. ion-selective electrodes are relatively

inexpensive and very portable. The methods of potentiometry

usually require minimal sample preparation and can be used

for direct measurement of a species (ie., use of a

calibration curve) or determination by titration. The

theory and methodology of ion-selective electrode

potentiometry has been reviewed extensively by several

authors (1-5).

Current-scan polarography at the liquid/liquid

interface is an electrochemical technique which can be

1

2

applied to the study of ion-selective electrodes. The

classical polarography experiment studies the transfer of

electrons between some species in solution and the dropping

mercury electrode; while at the liquid/liquid interface the

transfer of ions between an organic phase and an ascending

water electrode is observed.

l_il Polyroqg Membrane Ion-Selective Electrodes

Polymer membrane ion-selective electrodes (PMISE)

are a class of ISEs similar to the liquid-liquid membrane

type of electrode. The major difference is that the

electroactive material is dissolved in a plasticized polymer

membrane instead of an organic solvent. The result of this

difference is not only similar response characteristics

with respect to sensitivity and selectivity, but a more

durable electrode less prone to poisoning and in effect a

longer lifetime. Undoubtedly the most often used polymer

matrix is polyvinyl chloride (PVC) and a plasticizer. The

main role of the plasticizer is to reduce the glassy

transition temperture of the PVC below room temperture in

order to increase the ion moblility (6). Besides decreasing

the resistance through the membrane, the plasticizer can

also have an effect on the selectivity of the sensor. This

is particularly true in the case of discriminating between

mono and divalent cations with neutral carrier electrodes

(7). Other polymer matrices used in the construction of

3

PMISEs are Lucite 45, polymethyl acrylate and silicone

rubber.

1.1.1 Principles of PMISE Behavior

When the membrane of a PMISE comes into contact with

a solution, a potential will be produced which can be

measured against a reference electrode. This potential is

the result of two different potential generating processes

(8). The more significant of these two potentials is the

boundary-potential. The boundary-potential stems from an

unequal distribution of oppositely charged ions at the

solution/membrane interface, resulting in a charge

separation. This unequal distribution arises from one of

the ions of a salt having a greater affinity for the

membrane phase. The increased affinity is due to the

electroactive species in the membrane having the ability to

complex or ion-pair with the ion of interest (assuming both

ions are equally lipophilic). The second type of process is

known as diffusion-migration. The diffusion-migration

potential occurs within the bulk of the membrane in a state

of non-equilibrium, thus a concentration gradient from the

internal reference and sample solutions exists. The

potential is generated through the movement of ion-pairs

with unequal mobilities. The movement of an individual ion-

pair creates a local electric field, many of these ion-pair

4

electric fields average out into the diffusion-migration

potential. The fundamentals of these potential generating

PMISE systems can be investigated by conducting

electrochemical studies (polarography, cyclic voltametry

etc.) at the liquid/liquid interface.

An ideally behaved PMISE, in the presence of only

the primary ion i, will give a response as a function of the

activity of i according to the Nernst equation (9).

E = Constant + RT/nF In a^ (1)

Where E is the observed potential, R is the gas constant

(8,31 JK~*mol~*) , T is temperture in degrees Kelvin, n is

the charge on the ion and F is the Faraday constant (96,487

coulombs). The "Constant" term is a result of the reference

electrode potential and various junction potentials which

are considered to be constant. Equation 1 predicts an

electrode for a monovalent and divalent ion to have

sensitivities of 59.2 and 29.6 mv per log concentration

unit, respectively. If the primary ion is accompanied by an

interfering ion j, our ideal PMISE will give a response

conforming to the Eisenman-Nikolsky equation (10).

11 /n E = Constant + RT/nF In (a^ + k^j a* j) (2)

Where all terms have the same meaning as before. The k^j

term is the selectivity constant. The selectivity constant

5

is a measure of how effectively the interfering ion competes

with the primary ion. The larger the selectivity constant

the greater ion j will interfere.

Sensitivity and selectivity are not the only PMISE

characteristics to determine whether an electrode is

successful. The linear dynamic range and the detection

limit of an electrode must also be considered. However,

for a PMISE to be economically feasible, it must have a

substantial lifetime. The lifetime of a PMISE is usually

dependent on the rate of loss of the electroactive species

and/or the plasticizer (11). PMISEs are most often rendered

useless due to the electroactive material or plasticizer

being too hydrophilic. Low hydrophobicity leads these

components to leach out of the membrane into solution,

creating either a membrane void of any ion-exchanging

ability or too high in resistance to function normally.

Loss of membrane components can also be accomplished through

decomposition or any other types of chemical reactions

leaving them nonfunctional.

1.1.2 Neutral Carrier PMISEs

Polymer membrane ISEs which are based on a neutral

ligand that forms a charged complex with metal ions are

known as neutral carrier PMISEs. Neutral carrier PMISEs,

for the most part, are based on very large macrocyclic

compounds that produce 1:1 metal-ligand complexes. Although

6

the ion-exchanging mechanism is different from the more

typical ion-pairing, neutral carrier PMISEs also yield a

theoretical nernstian response. In fact, the response of

these electrodes can be characterized by an overall

distribution constant (12), as shown in equation 3,

Ki,n = Bis,n ^is,n (cs/ks)n

The overall distribution constant, is a function of

the metal-ligand aqueous formation constant, 3iSrn, the

distribution constant of the complex, k£Sfn# and the ratio

of the free ligand concentration in the membrane, cs, to the

distribution constant of the ligand, ks. The selectivity of

such PMISEs can also be evaluated by the ratio of the

overall distribution constants of the primary and competing

ions.

1.1.3 Coated-Wire Ion-Selective Electrodes

Coated-Wire ISEs (CWISE) are polymer membrane ISES

with a simpler construction than the more conventional

Moody-Thomas PMISE (see Figure 1.1). CWISEs are simpler in

that the ion-sensitive polymer membrane is directly adhered

to a conducting wire, totally bypassing the use of an

internal reference solution. The fact that these electrodes

have similar response characteristics, with respect to

sensitivity and selectivity (13), as the conventional PMISEs

indicates that there must be some type of internal reference

COPPER WIRE INTERNAL

REFERENCE

ELECTRODE

AND

SOLUTION POLYMER INSULATION

MEMBRANE

f t / / / / m

MOODY-THOMAS CWISE Figure 1.1 Comparison of the Hoody-Thomas and Coated-Wire Ion Selective Electrode

8

system that keeps the potential constant at the

polymer/metal interface. When the metal wire is copper, the

most likely internal reference process is the corrosion of

the metal in the presence of dissolved oxygen, producing a

constant corrosion potential (14). Even though the internal

reference system is not well defined, there are many

advantages associated with the coated-wire design. CWISEs

can be fabricated quickly and easily resulting in a rugged

and durable electrode. The absence of the internal

reference solution allows the electrode to be small,

enabling low sample volumes to be used and not to mention

the possibility of further miniaturization for in vivo

applications (15). In addition to the above advantages, the

CWISE costs very little to produce (about 5 ) making these

devices suitable as disposable sensors.

lj.2. Electrochemistv at the Liquid/Liquid Interface

The first paper involved with electrochemistry

between two immiscible solvents was published by Nernst and

Kisenfeld in 1902 (16). After that first paper, most of the

work in this field focused on equilibrium studies until the

mid 1950s when Guastalla and Dupeyrat passed current through

the nitrobenzene/water interface (17). Since then the

amount vof research done in this area has been expanding

rapidly. This is witnessed by the number of papers and

review articles that have been published in recent years

9

(18-20). Vanysek has also written a short book covering the

basic theory, experimental set-ups and a broad range of

applications (21).

The importance of studying ion-transfer at the

interface of two immiscible electrolyte solutions (ITIES),

with regard to ion-selective electrodes, exists for several

reasons. Ion-selective electrodes are usually studied

through conventional potentiometrie methods, using the ISE

itself as the measuring device. Therefore, information

concerning the fundamental potential generating process is

not accesible. Not only will studying this potential

generating system allow us to obtain a better understanding

of ISE behavior, but will also give us a method in which

ion-exchangers and plasticizers can be evaluated with

respect to response and selectivity. Having the ability to

characterize ISEs using electrochemical techniques at the

ITIES will save the analyst time and materials in regard to

constructing the ISEs and characterizing them by

conventional potentiometry. Although very little research

has been done on modeling ISE behavior with electrochemical

techniques at the ITIES, there has been some accomplishment

in this area (22-24).

1.2.1 Principles of Electrochemistry at the Liquid/Liquid Interface

When two immiscible electrolytes are at equilibrium,

the two solvents are assumed to be mutually saturated and

10

the chemical potentials of each component are equal in both

phases. The chemical potential of each individual species

in a phase can be represented by the following general

equation

u = uiri + RTln a^ + zF<j> (4)

Where Uifi is the standard chemical potential of species i'

in phase 1, R and T are the gas constant and temperture,

is the activity of i, z is the charge on i, P is the faraday

constant and <}> is the potential difference between the two

phases. The difference in the standard chemical potential

of a species in each phase is equal to the standard Gibbs

energy of transfer, AGtr, and is also equal to the

difference in standard Gibbs energies of solvation.

The difference between the inner potentials of each

phase is known as the Galvani or Nernst potential. The

Nernst potential of a single ion between phases 1 and 2 can

be described by equation 5.

<|>2 -<|>1 = -1/zF (AGtr - RTln

For a single salt, present at low concentrations in both

phases, we can assume that the activities of the cation and

anion are equal. The potential difference between the

aqueous and organic phase will depend on the unequal ability

of the cation and anion to move from one phase to the other.

11

4*2 — 4*1 ~ (AGfcr_ - )/F (6)

This is known as the distribution potential and is related

to the boundary potential generating process in PMlSEs as

discussed in section 1.1.1.

In order to study ion transfer at the ITIES, the

background electrolytes must establish a potential "window"

in which essentially no current flows. This potential

"window" is a result of a polarized interface. For a

polarized interface to exist, a few conditions must be met:

each solvent must have dissolved salts which are completely

dissociated; the aqueous phase salt must be very hydrophilic

and the organic phase salt must be highly hydrophobic (25).

In the presence of an ideally polarized interface, the

observed current goes towards the charging of the double

layer, hence no charge is actually transferred across the

interface. Also in this potential "window" region, the

potential at the interface will not be determined by ion

activities or Gibbs transfer energies, the potential will be

a result of changes in the double layer.

1.2.2 Electrocapillarity

When either a positive or negative excess charge

accumulates at the interface between two immiscible liquids,

these units of charge repel each other. This repulsion at

the interface counteracts the normal tendency of the surface

12

to contract and results in the lowering of the interfaclal

tension. As the potential at the interface is varied the

amount of excess charge is also changing giving rise to a

variation in the interfaclal tension. A plot of potential

versus interfaclal tension is known as an electrocapillary

curve. The electrocapillary curve has a basic parabolic

shape but it is not necessarily a second order function. A

common feature of the electrocapillary curve is the

electrocapillary maximum (ECM). Since the slope at every

point along the electrocapillary curve is the surface charge

excess,

a = -OY/ 3E) (7)

the ECM is also identified as the potential of zero charge

(PZC). The electrocapillary curve is related to the charge

density on the electrode, a, which can be expressed as the

differential shown in equation 7, where Y is the interfaclal

tension and E is the potential. The differential

capacitance, Cd, is also related to the electrocapillary

curve as the derivative of equation 7.

Cd = (3Y/9E)2 (8)

The differential capacitance is probably the most useful

parameter in the study of interfacial structure.

Not only does an excess charge at the interface

create a change in interfacial tension, but an excess

13

concentration (relative to bulk concentration) of a chemical

species will produce a similar effect. This excess

concentration, known as a relative surface excess, is the

result of either specific or nonspecific adsorption of a

species at the interface. specific adsorption normally

refers to a species having actual chemical interactions with

the surface of the electrode, while nonspecific adsorption

usually creates a surface excess through coulombic i

interactions with the charged electrode. The relative

surface excess of a species i, can be determined through the

use of the Gibbs Adsorption Isotherm (26) shown in equation

9.

r = -i/RT aVBinc^ (9)

As seen from the above equation, surface excess can be

determined by measuring changes in the electrocapillary

curve (at constant potential) as a function of a particular

species concentration.

14

Cadmium is an extremely toxic element and therefore

a simple, convenient method for the determination in water

(in the divalent form) would be very desirable. Coated-wire

ion-selective electrodes represents a class of analytical

tools that are known for their convenience and simplicity in

chemical analysis. Neutral carrier polymer membrane ion-

selective electrodes (PMISEs) have for the most part been

based on large macrocyclic compounds that form 1:1 ligand-

metal complexes. The use of neutral chelating agents which

form 3:1 ligand-metal stoichiometrics have long been ignored

for PMISE applications. Dinonyl bipyridine (DNBP) is such a

neutral ligand, which forms a 3:1 complex with divalent

cadmium along with other transition metals. Thus a DNBP

based neutral carrier PHISE for cadmium will be developed in

order to examine the applicability of this class of ligands

towards ion-selective electrodes. The PMISE wll be examined

using the conventional potentiometric methods for

determining the calibration and selectivity characteristics.

In addition to examining this new PMISE by the conventional

methods, the suitability of DNBP as a neutral carrier will

also be characterized by electrochemical experiments at the

liquid/liquid interface. The kinetics and thermodynamics of

DNBP facilitated cadmium ion (and other transition metals)

transfer will be used to relate both calibration and

selectivity characteristics of the PMISE.

The importance of a fundemental understanding of the

structure of the liquid/liquid interface and the adsorption

of molecules at that interface cannot be over emphasized.

Using a current scan polarography apparatus,

electrocapillary studies will be attempted to study the

adsorption of the neutral surfactant Triton X-100 at the

liquid/liquid interface. Both charge density and

capacitance data will be derived from the original data.

The interfacial adsorption studies will be used to

interpret the effect Triton X-100 has on facilitated ion

transfer.

CHAPTER 2

EXPERIMENTAL

2-m.I piaterf-als

2.1.i Reagents

Chromatographic grade polyvinyl chloride

(Polysciences; Warrington, PA), practical grade dioctyl

phthalate (J.T. Baker; Phillipsburg, NJ) and Gold Label

(99.9%) sodium tetraphenyl borate (Aldrich; Milwaukee, WI)

were used without further purification. Tetrahydrofuran

(Fishery Fair Lawn, NJ) was of reagent grade. 4,4'-Di(5-

nonyl) -2,2 '-bipyridine was obtained from Reilly Tar and

Chemical and was purified by recrystallization from ethanol.

Sixteen gauge-copper wire (Consolidated Wire; Chicago, IL)

was straightened, cut into ten centimeter sections, filed

flat on one end and subsequently cleaned in a Bransonic II

ultrasonic vibrator using water, acetone and chloroform.

Acetate salts of cadmium(II), zinc(II), nickel(II),

copper(II) and cobalt (II) were all of reagent grade and were

used as obtained from Mallinkrodt (Paris, KY). All water

used for ion-selective electrode work was house distilled

and deionized with a Crystalab Deeminizer ion-exchange

column.

16

17

1,2-Dichloroethane (Fisher) used in the current-scan

polarography experiments was always preeguilibrated with

water. Lithium chloride (Fisher) and magnesium sulfate

(Aldrich) were both reagent grade and used as recieved.

Tetraheptyl ammonium chloride (Fluka; Hauppauge, NY) was

recrystallized from acetone for further purification.

Tetraheptyl ammonium, tetraphenyl borate salt (THA+,TPB~)

was made by first dissolving the THA+,Cl~ in dichloroethane

(DCE), followed with ion-pair formation by extracting TPB"

out of the aqueous phase. The DCE phase containing

THA+,TPB"" was washed three times with deionized water.

Triton X-100 surfactant obtained from Sigma Chemical Co.

(St. Louis, MO) and plasticizers 2-nitrophenyl-octylether,

bis(2-ethylhexyl) adipate (both from Fluka) were used

without further purification. Reagent grade glacial acetic

acid (J.T. Baker) was diluted to the appropriate

concentrations and pH adjustment was done by dropwise

addition of 1M NaOH, All water used in the current-scan

polarography experiments was house distilled, deionized (as

above) and redistilled over permanganate.

2.1.2 Equipment

Electrode calibrations and selectivity studies could

be done on either of two computer controlled systems shown

in Figure 2.1. The major difference between the two systems

is the computer to which the experiment is interfaced. The

NOVA 3 PRINTER

IMSAI 8080

INTERFACE

pH METER

PLOTTER

TERMINAL

MULTIPLEXOR

ISE

CELL

BURET NO. I BURET NO. . 2

Figure 2.1 Computer fyster.i for Calibration and selectivity s tudies for ISEs.

19

original instrumentation was based on an Intel IMSAI 8080

microcomputer made available by Dr. C.R. Martin (27). The

experiment was later modified by Dr. L. Cunningham (28)

with a more versatile Nova 3 minicomputer (Data General

Corp.? Southboro, MA). The computer is interfaced to a 801A

pH/mV meter (Orion Research; Cambridge, MA), DV10 digital

burets (Mettler Instruments; Hightstown, NJ), a one-of-

five analog multiplexor and a relay to a stirrer motor. The

Nova 3 is also equipped with a DEC writer III LA-120

printer (Digital Equipment Corp.; Maynard, MA) and a Miplot

WX 4671 digital plotter (Watanabe Instruments; Tokyo, Japan)

for graphical representation of data. An ADM-5 graphics

terminal (Lear-Siegler; Anaheim, CA) allowed the

experimenter to enter the necessary experimental parameters

and to carry out data analysis. Dr. L. Cunningham wrote all

of the software which controls the above peripherals.

Programs are written in FORTRAN IV and assembly language for

Data General machines and in CONVERS for the IMSAI 8080.

Both the hardware and software have been described in

greater detail elsewhere (29),

The cell for CWISE calibrations consisted of a 100

ml beaker containing an initial 75 mis of water in which the

indicator (CWISE) and reference electrodes were immersed. A

schematic representation of the cell is shown below.

Cu/'Polymer Mem./Sample Sol*n//0.1M NH4N03/Sat. KCl/AgCl/Ag

20

The reference electrode is a double junction Ag/AgCl

electrode. The internal junction consists of a silver wire,

in which silver chloride was electrochemically deposited

(3) t in contact with a saturated KC1 solution. The external

junction is a 0.1M solution of NH4NO3. Ammonium nitrate

was used in the external junction because both NH^ and NO3

ions have similar mobilities and thus will keep the junction

potential to a minimum.

All cur rent-scan polar ography experiments were done

with the computer system and cell depicted in figures 2.2

and 2.3 respectively. The computer system is made up of a

Nova 800 minicomputer interfaced to the cell through various

hardware modules. These modules include a timer/drop

release detector, a galvanostat/mV meter with cell

resistance compensation, an analog-to-digital converter, a

digital-to-analog converter and a variable frequency clock.

Along with the above modules, the Nova 800 was also

interfaced to a Graphics Plus console (Zenith Data Sys.; St.

Joseph, Ml) and a Watanabe 1000 (Watanabe Instruments)

digital plotter for graphical display of acquired data. The

above computer system has been discussed in more detail

elsewhere (29). The cell used for current-scan polarography

has been described previously (30), except for some slight

modifications in the design and aqueous phase delivery

system. For the remainder of this work the current-scan

DATA SELECT LINES

TO PRINTER

-<) "TO TERMINAL

TO PLOTTER

mV METER GALVANOSTAT

D/A RELEASE

TIMER

DROP

DET.

A/D CLOCK

NOVA 800

CPU

CELL

Figure 2.2 Computer System for Data Acquisition for AWE Experiments.

22

ASCENDING WATER ELECTRODE

APPARATUS

ORGANIC COUNTER

AQUEOUS

REFERENCE

ORGANIC REFERENCE

AQUEOUS COUNTER

TO PERISTALTIO PUMP

Figure 2.3 Apparatus used for Current-Scan Polarography Experiments.

23

polarography cell will be referred to as the ascending water

electrode (AWE), In general, the AWE has a four electrode

configuration: two reference electrodes to measure the

potential difference between the organic and aqueous phases,

and two counter electrodes which apply a current through

these two phases. This cell configuration and the

galvanostatic nature of the AWE allows easy compensation for

ohmic potential drop through the high resistance of the

organic phase and capillary. Changes in the design of the

AWE have been minimal, making the placement of the organic

reference electrode more reproducible and taking the cell

apart for cleaning easier and quicker. In previous uses of

the AWE, a variable height reservoir was used in the control

of the aqueous solution flow rate. The present apparatus

has been improved by the addition of a Rabbit peristaltic

pump (Rainin Instruments; Woburn, MA) for constant and

reproducible flow rates (standard deviations between one and

two percent). This change was particularly applicable to

the acquisiton of electrocapillary data,

2umZ Procedures

2.2.1 Fabrication of OflSEs

Coated-wire ion-selective electrodes were made by a

method similar to that reported earlier (31). Precut and

clean copper wire (see section 2.1.1) was repeatedly dipped

in a 7.5% (w/w) solution of PVC in THP until a 1 mm film was

24

formed. The purpose of this film is to insulate the

conducting wire from the test solution hence preventing a

short circuit of the electrode. About 3 mm of insulation

material was then cut away to expose the copper wire, where

the ion sensitive polymer membrane solution was applied

repeatedly until the formation of a small uniform bead. It

was important to allow sufficient time (10 min.) for drying

of the membrane in order to prevent'air bubble formation.

The CWlSEs were conditioned by soaking in a 10"^M solution

of the appropriate ion a week prior to the first

calibration. Between calibrations the electrodes were also

stored in a millimolar solution of the primary ion.

2.2.2 Calibrations and Selectivity Studies

All electrode calibrations and selectivity studies

were performed using the computer system and cell described

in section 2.1.2. Five individual CWISEs can be

simultaneously calibrated in a single sample cell. The

electrodes are first allowed to equilibrate in a known

volume of water for about an hour to prevent any hysteresis

effects that may be caused by the storage solution. This

known volume, along with the concentration range,

concentration steps, the concentration of the standard, the

equilibrium time and the potential drift were entered into

the computer program. The above parameters allowed correct

25

dispensing of the concentration standard and monitored the

potential drift of the electrodes over a chosen time period.

Following a calibration, selectivity characteristics

of the CWISEs can be determined. Selectivity coefficients,

kirj* were calculated usng the two solution method (32).

Prior to the calibration, parameters such as the

concentration range of the interferent ion and the ion size

of both primary and competing ions were input to the

computer. A second 10 ml capacity buret was used for the

delivery of the competitive ion into the sample cell.

2.2.3 Cur rent-Scan Polarography

The basic construction of the electrodes used with

the AWE can be seen in figure 2.4. The electrodes are the

same as reported earlier (30), except for the organic

reference electrode. Instead of using the tetramethyl

ammonium ISE, the organic reference consists of a Ag/AgCl

reference electrode immersed in a IH LiCl solution, which is

in contact with 1 ml of the organic background electrolyte.

These internal solutions are separated from the organic

phase by the use of a porous (2 micron) teflon membrane.

This organic reference electrode will be referred to as the

Organic Background Electrolyte (OBE) reference electrode.

The TMA+ ISE reference electrode was less reproducible and

caused much more trouble with respect to false triggering

than the OBE reference. This is possibly due to the TMA+

26

SILICON RUBBER

STOPPER

Ag/AgCI

REFERENCE I M LiCI

I M LiCI

GLASS FRIT GLASS

FRITS

0.0 I M THA,TPB

IN DCE TEFLON

MEMBRANE TEFLON COLLAR

f 2

Figure 2.4 Diagram of Electrodes used in Current-Scan Polarographv Cell. (1) Aqueous, organic Counter Electrodes and Aqueous Reference Electrode; (2) Organic Reference Electrode.

diffusing from the electrode organic phase into the organic

electrolyte (which has no TMA+). Thus the diffussion of

TMA+ will cause a change in the activity of TMA+ in the

organic phase (in the ref. electrode), hence the variation

in potential of the reference electrode. Also, the

additional junction potential created at the organic

reference/organic electrolyte junction should be considered.

The next question is why the OBE reference produces such a

steady potential. Since there is essentially no current

flowing through the organic reference electrode, the

potential will be determined by the charging of the double

layer between the 1 M LiCl aqueous phase and the organic

background electrolyte phase. As long as there is no

perturbation in the double layer structure (ie. no current

flowing and a constant composition in both phases), the

potential at that interface should remain constant. The OBE

reference electrode produces a potential 106 +/- 5 mV more

positive than the previously used TMA+ ISE reference

(resulting in a current voltage curve shifted 106 mV in the

negative direction when compared to the same curve measured

against the TMA+ ISE) (33). Throughout this work, all

potentials reported will have been measured against the OBE

reference electrode.

In the course of obtaining a single data point of

the current-scan polarogram, the hardware modules mentioned

earlier (section 2.1.2) have the following roles. First of

all, as the aqueous phase ascends through the more dense

organic phase, the digital-to-analog converter sends a

predetermined voltage to the galvanostat which in turn

applies a current through the cell. The drop-release

detector is disabled for several seconds in order to allow

the resistance through the capillary to decrease to the

point (when the ascending water drop has sufficient surface

area) where the detector won't falsely trigger. Once the

drop-release detector is enabled the analog-to-digital

converter starts taking potential readings from the mV meter

until drop release occurs. In order for all of these

modules to work in harmony, the the correct computer program

parameters have to be entered. Shown in figure 2.5 are the

program parameters and typical values that work correctly.

2.2.4 Electrocapillary Measurements

Along with the current-voltage curves acquired with

the above apparatus, electrocapillary data is taken

concurrently. The variable frequency clock determines the

drop time at each current step. Knowing the drop time, the

aqueous flow rate and the phase densities, the interfacial

tension can be calculated by Young and Harkins method (34).

When taking electrocapillary data, generally the flow rate

is kept very slow (0.002 ml/sec) to give more precise drop

time data and to minimize the effects of eddy currents

29

PARAMETERS DESCRIPTION TYPICAL VALDES

INITIAL I

PINAL I

IRANGE

ISTEP

REPEAT CNT

FLAG DELAY

CLOCK FREQ

HSECS/PT

POINTS/SCAN

PTS AVGED

PTS BEFORE TRIG

TIME WINDOW

CAP RADIUS

Where to start current scan (uA)

Where to stop current scan (uA)

Microamp range set on interface

Size of current steps in uA

How many times to repeat at a point if apparatus mistriggers

How long to keep drop release detector disabled (milliseconds)

Frequency of clock

Sampling frequency of potential (milliseconds)

Number of potential readings (pts) taken during one drop lifetime

Number of potential readings averaged for the stored value at one i step

Number of pts to come off of potential spike to average potential readings

Always keep at 0.7

Radius of capillary in cm

-5

15

100

0.5

7000

100

20

1000

25

35

0.7

0.048

Figure 2.5 List of Parameters Required for AWE Computer Program with Description and Typical Values Entered.

30

produced by the water drop ascending through the organic

phase after it has been released. Due to the low agueous

solubility of the tetraheptyl ammonium/acetate salt, when

THA+ transfers from the organic to the aqueous phase it will

tend to precipitate out on the capillary. Thus it is

imperative to clean the teflon capillary with acetone after

each scan since this precipitate formation, eddy currents

and other perturbations (table vibrations, air bubbles near

the capillary etc.) can cause the drop to release

prematurely to give poor electrocapillary results. The

flow rate of the peristlatic pump is calibrated prior to an

electrocapillary experiment by weighing the amount of water

collected over a period of time.

CHAPTER 3

RESULTS AND DISCUSSION

2-lL Characterization q£ a Cadmium EfllSE uging Current Scan Polaroaraphv

The polymer membrane ISE to be characterized is

based on the neutral bidentate ligand dinonyl bipyridine

(DNBP). The electrode was conditioned to be selective

towards divalent cadmium ion. Cadmium was chosen for its

favorable formation constant with DNBP. The selection of

cadmium was also in hope that the rapid kinetics of

cadmium(II) would give additional selectivity over other

slower transition metals such as nickel(II), cobalt(II) and

zinc(II), which also exhibit favorable formation constants

with dinonyl bipyridine.

The suitability of Dinonyl bipyridine as an

electroactive component for a neutral carrier PMISE was

studied by various experiments. The PMISE chemical system

was modeled by placing the DNBP in the organic phase and

cadmium or other possible interfering ions in the aqueous

phase of the current scan polarography apparatus.

3.1.1 Calibration Characteristics

Figure 3.1 shows an anodic wave superimposed on a

current-voltage curve of the background elecrolytes. The

31

32

17.0

- 12.0 to Q_ z: a: o oc

z tu a: a:

c_>

-.01 0C0 . 10 .20 .30 .40

POTENTIRL (VJ

Figure 3.1 Polarogram of Cd-DNBP lonlransfer (1); polarogram of Background Electrolyte (2).

33

cathodic final is due to the transfer of tetraheptyl

ammonium ions from the organic to the aqueous phase; the

anodic final is a result of the transfer of tetraphenyl

borate ions from the organic to the aqueous phase. The

anodic wave arises from the transfer of cadmium(II) from the

aqueous phase (0.2 M sodium acetate, pH 7) to the organic

(0.01 M THA,TPB in DCE), facilitated by the ligand DNBP.

The appearance of a proton-ligand wave will depend on the pH

of the aqueous phase and the water exchange rate of the

competing metal ion (35).

The question of the mechanism of facilitated ion

transfer at the liquid/liquid interface has not been fully

resolved. In general, there are two different schools of

thought. The mechanism in which the following data will be

interpreted is described below and depicted in Figure 3.2.

The cation Mf with charge n, and ligand L diffuse from their

bulk phases to the aqueous and organic reaction layer,

respectively. Some of the ligand, depending on the

distribution constant, will cross the interface into the

aqueous reaction layer and complex with Mn+. The complex,

(Mn+La), will subsequently cross the interface from the

aqueous to the organic reaction layer producing the observed

ion transfer or current. In the organic reaction layer,

(Mn+La) will complex with additional b ligands and ion pair

with m tetraphenyl borate anions. The complex

(Mn+La+k,mTPB"") then diffuses into the bulk organic phase.

34

AQUEOUS

BULK PHASE

AQUEOUS

REACTION

LAYER

Mn M +&L ^

CM"!) B

ORGANIC

REACTION

LAYER •

S \ \ N % V

ORGANIC

BULK PHASE

CMla) °;l^vp^L

bL*

1 <M,0

+ mTPB-

CM1!. braTPB—

I

Figure 3.2 Schematic Representation of Mechanise of IonTransfer Across the L i qui d/L i quid Interface.

35

The other school of thought believes that KDR for L is too

small and that complex formation for the most part takes

place at the interface or in the organic reaction layer.

When DNBP is present in the organic electrolyte at

varying concentrations and the cadmium concentration (lmM)

is held constant in the aqueous phase, it can be seen that

the limiting current of the Cd-DNBP polarogram varies

linearly as a function of DNBP concentration (Figure 3.3).

In the above conditions the cadmium ion is in excess, hence

there is no observable change in the half-wave potential of

the facilitated cadmium transfer. The above experiment

simply illustrates the first order dependence of the ion

transfer kinetics with the concentration of our ligand.

The next question is whether the kinetics are

controlled by the diffusion of our ligand from the bulk to

the organic reaction layer or by the water exchange rate of

the cadmium ion. The question of diffusion or kinetic

control of ion transfer is answered by the dependence of the

limiting current on the aqueous flow rate. According to

equation 10, a diffusion

Log i^. = 0.5 Log(Plow rate) + Constant (10)

controlled process will yield a limiting current which has a

square root dependence on the flow rate. Figure 3.4 shows a

plot of Log limiting current versus Log flow rate producing

36

CE d

2: tu oc on ZD o

CD z

5.50 ..

4.50 ..

3.50 ..

2.50 ..

ac 1.50 ..

.50 ..

DNBP CONCENTRATION (xl0M

Figure 3.3 Limiting Current Dependence on DNBP Concentration {mol/l).

37

. 8 0

.70 .. UJ Q£ 0£

O

O . 6 0 . .

CD O .50 ..

-2.9 -2.8 -2.7 -2.6 -2.5

LOG (FLOW RRTEJ

Figure 3.4 Plot Indicating Diffusion Controlled Kinetics for Cd-DNBP system.

38

a slope of 0.56, Indicating the kinetics of formation of the

cadmium-DNBP complex is rapid enough for ion transfer to be

diffusion controlled. Since the transfer of cadmium is

diffusion controlled, we are able to calculate the diffusion

coefficient of DNBP in the organic phase by applying the

Ilkovic equation shown below.

ix « 627nPCD1/2t1/6m2/3 (11)

Where i^ is the limiting current (uA), n and F have there

usual meaning, C is the concentration of ligand (mol/liter),

D is the diffusion coefficient (cm2/sec), t is the drop

time (sec) and m is the aqueous flow rate (g/sec). The

diffusion coeffecient of DNBP in DCE was determined to be

l.OxlO""7 cm2/sec. The modified Ilkovic equation was not

used since the radius of the teflon capillary was determined

to be insignificant.

Another important parameter with respect to the

behavior of an ISE is the reversibility of ion transfer.

This is particularly evident in the case of neutral carrier

PMISEs as depicted by equation 3, where the overall

distribution constant of the ion is dependent on the

concentration of free ligand in the membrane. The

reversibility of ion transfer is exhibited by the shape of

the current-voltage wave. The wave shape can be evaluated

by the conformity to equation 11. Figure 3.5 is a plot of

10.5 1 | | ) 1 1 1 | 1 h—H -

1 0 . 0 . .

9 . 5 . . . . .

9.0 ..

8.5 ..

8 . 0 . .

7.5 ..

1 1 1 1 1 1 1 1 1 1 1. -.50 -.30 -.01 .10 .30 .50

LOG

Figure 3.5 Plot Indicating Reversible Ion Transfer for Cd-DNBP System.

i 1 1 1 1 1 1 1 1 \ - • -i

1 1 1 1 1 1 1 1 1

40

E = Ej/j + RT/nF lnftij - i)/i} (11)

the potential at some point on the wave against the log term

shown in equation 11 (where i is the current at that point

on the wave and i-^ is the limiting current). At a DNBP

concentration of 0.75 millimolar, a slope of 28 was obtained

indicating the ion transfer is reversible.

The above results have mainly dealt with the kinetics

of our PMISE chemical system. Thermodynamically, we would

expect the system to comply with the Nernst equation

(equation 1) if the PMISE were to behave in a normal sense.

When the DNBP concentration was held constant at 0.5 mM and

the cadmium concentration was increased from 5xl0~4 to 0.01

molar, a negative shift in the half-wave potential of 34 +/-

3 millivolts (mV) per log unit of concentration was

observed. This value is close to the theoretical nernstian

slope (2.3RT/nF) of 30 mV per log concentration.

The results of the above experiments suggest that a

PMISE for cadmium based on the neutral carrier DNBP should

yield a good working calibration curve. The demonstration

of rapid reversible kinetics of ion transfer should be

manifested by an electrode with a fast response time (36)

and one that behaves in accordance with the thermodynamics

of the system. Thermodynamicallly speaking, the half-wave

potential dependence on cadmium concentration is indicative

of a PMISE that will respond in a Nernstian fashion.

41

> E

Z LU I— O Q_

UJ > tr

1 u.

-3.0 -2.5 - 2 . 0 -1.5

LOG Cd CONCENTRRTION

Figure 3.6 Half-Wave Potential Dependence on Cd Concentration (in mol/1) for Cd-DNBP tystem.

42

The PMISEs were constructed with two different

membrane compositions. Initially the polymer membrane was

made up of 50% PVC, 45% dioctyl phthalate (plasticizer, DOP)

and 5% DNBP. These electrodes produced calibration curves

shown in Figure 3.7. The membrane composition was modified

so that it now had 47.5% PVC, 45% DOC, 5% DNBP and 2.5%

sodium tetraphenyl borate. The calibration curves produced

by these electrodes are shown in Figure 3.8. The major

difference in the response characteristics of these two

PMISEs is the linear dynamic range. Both types of

electrodes are linear down to about 10"5,4 M cadmium and

have detection limits near 10"® M. However, at the high

concentration end of the calibration curve, the first set of

electrodes stop responding at 10"^*® M while the composition

modified electrodes were linear all the way up to 10"^ M

cadmium. The limited dynamic range at high concentrations

of the electrode is due to an anion interference. The anion

effect is a well known phenomenon with neutral carrier

PMISEs. At high concentrations of cation there will be a

large proportion of charged complexes in the interfacial

region of the membrane. These positively charged complexes

will behave as anion exchange sites. In this particular

case, acetate was the interfering anion present as the

cadmium counter ion and as a pH buffer {0.1 M, pH 4.8). The

incorporation of a lipophilic anion in the polymer membrane

provides anionic sites which decreases the affinity of an

43

- 1 0 .0. .

-30,0..

-50.0..

-70.0..

-90.0..

-5.5 -4.5 -3.5 -2.5

LOG Ca CONCENTRATION

Figure 3.7 Calibration Curves for Cd Electrodes ofComposition: 5056 PVC, 4556 DOP, 556 DNBP. Concentration in mol/1. Different Symbols R epresent Different Electrodes.

-40.

-60.

-80 .

-100

-120

-5.5 -4.5 -3.5 -2.5

LOG Cd CONCENTRATION

Figure 3.8 Calibration Curves for Cd Electrodes of composition: 47.53> PVC, 45% DOP, 5% DNBP, 2.5% TPB. Concentration in mol/1. Different symbols Represent a different Electrode.

-5.5 -4.5 -3.5

45

anion for the membrane. By the addition of sodium

tetraphenyl borate in the polymer membrane and the

elimination of the acetate buffer, the linear dynamic range

of the electrode was extended by nearly an order of

magnitude.

The cadmium PMISEs were calibrated once a day over

the lifetime of the electode of approximately a week to ten

days. Nernstian responses were obtained for all five

electrodes in the set. The calibration results are

summarized in Table 3.1. The reproducibility from one

electrode to the next was fairly good; the slopes and

intercepts between the five electrodes had standard

deviations of a half of one millivolt and seven millivolts

respectively. However, the reproducibility of one electrode

over the lifetime of the electrode should be better. In

some instances the slope of an electrode would remain

constant and the intercept would change only about a

millivolt over a two day period. But for most part, the

variations were larger and the intercepts tended to vary

towards more positive potentials. Perhaps the

reproducibility could have been improved by thermostating

the sample cell and by using a pH buffer.

The response times of these electrodes were

determined to be the amount of time it took to reach a

steady-state potential in which there was less than a 0.5

46

Table 3.1 Calibration Characteristics of Cadmium PMISE

ELECTRODE SLOPE SLOPE INTERCEPT INTERCEPT # STD STD

1 30.16 1.59 53.93 6.10

2 29.67 1.55 51.51 6.16

3 30.85 1.24 63.96 5.10

4 29.65 1.48 54.93 6.85

5 30.77 1.51 66.94 6.67

47

mv/min drift. Although the computer allows five minutes for

the electrodes to reach a steady-state potential, the

cadmium PMISEs had much faster response times. The response

time of the cadmium electrode was usually less than one

minute, but varied depending on the concentration range.

3.1.2 Selectivity Characteristics

The most often studied and probably the PMISE

characteristic of greatest importance is the selectivity

of an electrode with respect to the primary ion.

Selectivity characteristics of PMISEs are most often studied

through potentiometry alone and reported as selectivity

coefficients, However, other electrochemical

techniques including alternating-current impedance methods

and electrodialysis have been applied to correlating

physical parameters such as transport numbers to the

selectivity of a membrane (37). This section will focus on

correlating the selectivity of the cadmium electrode with

the differences in potential at which the primary and

interfering ions transfer across the liquid/liquid

interface.

In a purely thermodynamic sense, the selectivity of

an ISE could be characterized by measurelng how easily ions

will transfer across the liquid/liquid interface in the

presence of the ligand of interest. The accuracy of this

characterization will largely depend on how well the liquid

48

organic phase mimics the electrode membrane, and the

similarities in the rate of ion transfer between the primary

and interferring ions. The measure of ion transfer facility

will be the half-wave potential at the metal ion

concentration equal to that used in the actual ISE

selectivity determination. The difference in half-wave

potentials for the primary and interfering ions will be used

in the calculation of the Potentiodynamic selectivity

coefficient (k^y"j). Potentiodynamic selectivity

coefficients are calculated using the same equation

(equation 12) used for the calculation of except half-

wave potential differences are used instead of the measured

ISE potential differences.

ki,j = ^ exP (AE np/RT) ai ~ ai)/a2 t12*

Samec et. al. determined potentiodynamic selectivity

coefficients using the formal potentials of ion transfer for

a number of alkali and alkaline earth metal ions (24).

Although a similar trend between and the

potentiodynamic selectivity coefficient was observed, there

was a substantial difference between the actual values of

k^°j and k^nj for any particular ion. It would be highly

advantagous to also have the ability to acquire values

that reflected the actual k£°j values.

The cadmium electrode membrane was again

approximated by incorporating 0.5 mM DNBP in DCE along with

49

1 1 . 0

CO Q.

<r o cc. o

z: LU Q£ 0£ => O

- 1 . 0

-3.0

-.20-.i00.0 .10 .20 .30 .40 .50

POTENTIAL (V)

Figure 3.9 Polarograms Showing Difference in Half-Wave potential for the Transfer of (1) Cd-DNBP and (2) Cu-DNBP.

50

the 0.01 M THA+,TPB~ electrolyte. Polarograms were obtained

at cadmium and interferring ion concentrations of 10~3 M in

the aqueous 0.2 M acetic acid/sodium acetate electrolyte (pH

6.85). The resulting differences in half-wave potentials

are tabulated in Table 3.2. Both copper(II) and zinc(II)

yielded cathodic waves at less positive potentials that

cadmium(II), indicating the transfer of these two

interfering ions is more favorable than the transfer of our

primary ion. The interfering ions cobalt(II) and nickel(II)

did not produce any type of waves at all. This is

undoubtedly due to the very slow kinetics (water exchange

rates) exhibited by these cations. In fact, a trend of

decreasing wave height with decreasing water exchange rate

exists (as long as the rates are slow enough so they're not

diffusion controlled). The order of wave height and water

exchange rate are as follows: Cd,Cu > Zn > Co,Ni. This

same trend was also observed in Yoshida's work with 1,10-

phenanthroline (35).

The selectivity of neutral carrier based PMISEs can

be approximated by the ratios of the overall distribution

constant of the two ions in question. However, it must be

understood that this evaluation takes only the thermodynamic

properties of the distribution of an ion into question. The

distinction between thermodynamic and kinetic contributions

to the selectivity of an ISE becomes more critical when

Table 3.2 Selectivity Characteristics of Cadmium PMISE and kYnj values determined by Cur rent- Scan' Pol arogr aphy

INTERFERENT Log Iffo AE1/2 Log

Copper(II) 5.77 172 5.72

Zinc(II) 0.85 50 1.68

Cobalt(II) -0.97

Nickel(II) -2.40

dealing with transition metals as opposed to the alkali and

alkaline earth metals. This is simply due to the fact that

the alkali and alkaline earth metals have fast water

exchange rates and therefore the kinetics will be diffusion

controlled. While the transition metal ions have a much

broader range of water exchange rates, where some will be

controlled by diffusion and the others controlled by the

kinetics of complex formation. The rate of ion transfer

will be diffusion controlled for both copper and cadmium

ions, and thus should be accommodated by the thermodynamic

model. But the cations such as zinc(n), cobalt (II) and

nickel(II) have significantly slower water exchange rates

than cadmium and copper (10, 1000 and 10,000 times slower,

respectively), and should deviate from selectivity values

predicted by thermodynamics exclusively. Assuming all of the

above transition metal-DNBP complexes are equally

extractable into the polymer membrane, equation 3 leads one

to believe that increasing selectivity should follow a trend

of increasing formation constant. Therefore, the order of

selectivity should be: Ni > Cu > Co > Zn > Cd (formation

constant values are for the parent compound of DNBP, 2,2'-

bipyridyl (38)). However, the actual order of selectivity

is: Cu > Zn > Cd > Co > Ni. The above discrepancies are

surely due to the differences in the rate of ion transfer

across the solution/membrane boundary.

53

As can be seen from Table 3.2, the current-scan

polarography studies yielded potentlodynamic selectivity

coefficients in the same order as the selectivity

coefficents obtained from the PMlSEs. In the case of

copper(II) being the interfering ion, a potentlodynamic

selectivity coefficient was produced within less than ten

percent of the determined k£°j value for the PMISE. It

should also be remembered that the rate of ion transfer for

both Cu(II) and Cdfll) cations is diffusion controlled. The

kjyri value for Zn(II) deviated to a much greater extent from AR J

the k^otj value of the PMISE. The PMISE is less selective 11 J

towards zinc over cadmium than the differences in the

potential of ion transfer would infer. The enhanced

selectivity of cadmium is again due to the slow rate of ion

transfer for zinc. It is interesting to note that by simply

multiplying the differences in half-wave potential by the

ratio of wave heights (zincscadmium) resulted in a A Ejy2

value that produces a within fifteen percent of the

kJj^j value. Perhaps this could be used to correct for any

large deviations due to a difference in the kinetics of ion

transfer. It was impossible to evaluate both cobalt and

nickel since their polarograms lacked any distinction from

the pure background electrolyte. However, it should be

noted that the k^^ values are much smaller than would be

54

expected for cations with formation constants 104 and 109

times larger than the formation constant of the primary ion.

The role a plasticizer plays in a neutral carrier

PMISE goes beyond the normal application as a means to

increase the mobility of ions within the polymer membrane.

It has been shown that the dielectric constant of the

membrane, which can be altered by choosing the appropriate

plasticizer, has a significant effect on the selectivity

between monvalent and divalent cations for a neutral carrier

PMISE. As the dielectric constant is varied, especially

when e <10, a low dielectric constant will increase the

selectivity towards singly charged cations while high

dielectric constants favor the extraction of cations with a

charge of two (7). Therefore the selection of a plasticizer

for a neutral carrier PMISE becomes much less trivial.

Using current-scan polarography at the liquid/liquid

interface, the role the plasticizer plays in the selectivity

between monovalent and divalent cations was investigated.

Three different plasticlzers were studied: dioctyl phthalate

(DOP), bis (2-ethyl hexyl) adipate (EHA) and 2-nitrophenyl

octylether {NPOE}. The plasticizer was incorporated into the

organic electrolyte at a concentration of 50% (by weight)

along with 0.5 mM DNBP. The proton, at a pH of 4.8, was

used to represent the monovalent cation. The divalent

cation used was cadmium(II) at a millimolar concentration.

Selectivity was evaluated by the differences in the

55

potential at which the proton and cadmium transfered. For

the PMISE to be selective towards Cd2 + versus H+, a

plasticizer that yielded the greatest difference in the

potentials of transfer (with Cd2+ transferring at a less

positive potential than H+) would give the most desirable

selectivity. If the only effect of the plasticizer was to

change the dielectric of the organic phase, then the

selectivity should increase with an increase in the

dielectric constant of the plasticizer. According to the

dielectric constants of the plasticizers used in this study,

we would expect NPOE to have the most beneficial effect on

the selectivity with DOP being second and EHA third.

However, as can be seen in Table 3.3, the actual order of

preference is EH A, NPOE and then DOP. These results

indicate that the plasticizer bis-(2-ethylhexyl) adipate

would be the most effective. By changing the plasticizer

from dioctyl phthalate to EHA, a selectivity enhancement of

three times would be realized between the proton and

cadmium(II). Whether or not these results could be

generalized to say that EHA would produce the best

selectivity of divalent cations over monovalent is hard to

say. It is difficult to evaluate the extent of specific

intermolecular interaction between the organic functional

group of the plasticizer and the two cations studied. The

proton was probably a poor choice for the monovalent cation.

56

Table 3.3 Results of the Effect Plastlcizers have on the Selectivity between Monovalent and Divalent Cations at the Liquid/Liquid Interface.

PLASTICIZER CATION DIELECTRIC E1/2 Ei/, CONSTANT (mVf (mV)

NPOE Cd2+ 23 .6 120 48

NPOE H+ 23 .6 168

DOP Cd2 + 5.15 164 26

DOP H+ 5.15 190

EH A Cd2 + 4.13 128 58

EH A H+ 4.13 186

57

The small size of H+ (with respect to cadmium), thus a high

charge density, would make it more apt to participate in

intermolecular interactions than cadmium(II). A more

appropriate choice would have been copper(I).

The plasticizer with the largest dielectric

constant, NPOE, should increase the facility of ion transfer

for the divalent cation to the greatest extent; in the same

sense, that plasticizer should also hinder the transfer of

the monovalent cation to the greatest extent. Nitrophenyl

octylether does indeed yield the least positive potential of

transfer for divalent cadmium. But, NPOE also produces the

most facile proton transfer. This behavior is probably due

to the interaction between the nitro group of the

plasticizer and the proton. With the aqueous phase having a

positive charge, the nitro group (the two oxygens sharing a

negative charge) would be orientated towards the interface,

thus allowing favorable interactions with H+. These

favorable interactions will result in an effective increase

in the proton concentration at the interface giving rise to

an ion-transfer potential less positive than expected. Both

DOP and EHA have d i-{2-e t hy 1 he xy 1) ester group

functionalities. 3 The difference being that the ester is

bonded to a phenyl group with DOP and a hydrocarbon for EHA.

Therefore, there should be a minimal difference in cation-

plasticizer interaction. The proton transfers at a slightly

less negative potential for EHA, as would be expected since

58

EHA has a slightly smaller dielectric constant. However,

for a reason not apparent to the author, EHA causes

cadmium(II) to transfer 36 mV less positive than DOP. The

above results illustrate the large role the organic

functional group of the plasticizer can play in the transfer

of cations across the liquid/liquid interface. The question

of extrapolating these results to actual PMISE behavior can

only be answered by runninig selectivity studies on

electrodes composed of the above plasticizers.

3.1.3 Cadmium PMISE Lifetime

As was mentioned in section 3.1.1, these cadmium

electrodes have short lifetimes of approximately a week to

ten days. The concern of the short lifetime is not only to

learn how to extend the working time of these electrodes

(for they are so inexpensive to fabricate they could easily

be used as disposable sensors), but the mechanism by which

these electrodes become nonresponsive may give further

insight to the behavior of these PMISEs.

When the electrodes are calibrated once a day for

about a week, the response of the electrode begins to decay.

This decay in sensitivity is exhibited by a steady decrease

in the slope of the calibration curve. The slope continues

to decrease until ultimately the slope of the electrode

becomes negative (about -10 mV/log conc.) as if it were

responding to anions. As was discussed in section 1.1.3,

59

the lifetime of a PMISE is usually dependent on the rate of

loss of the ion exchanging ability of the electroactive

component. In most cases this is due to the electroactive

material leaching into the aqueous solution until the

concentration is too low for an effective response. In the

case of the cadmium PMISE, dinonyl bipyridine is a very

lipophilic compound; it containing two alkyl chains of nine

carbons, only two nitrogen atoms and no oxygen atoms. Thus

the high lipophilicity of DNBP makes it a poor candidate for

the above mechanism. The anionic response of the electrode

after it had stopped responding to cadmium pointed towards

the possibility of the membrane being loaded with cadmium to

the extent where there was an insufficient number free

ligand sites in the membrane. This mechanism of electrode

inactivity was first tested by observing the effect of

extracting the cadmium out of the membrane using a water

soluble chelate. The dead electrodes were conditioned in a

pH 11, 0.1 M EDTA solution overnight. These electrodes were

reconditioned in a millimolar cadmium acetate solution for a

day and then calibrated. The original Nernstian response of

the electrode returned with all the other PMISE

characteristics exhibited by the electrode before it died.

In order to confirm that the PMISEs were becoming

nonresponsive due to cadmium loading and therefore a lack of

exchange sites in the membrane, a quantitative OV spectral

60

analysis was done. The objective was to monitor the

concentration of the complexed DNBP in the membrane as a

function of time. In preparation for this study, spectra

for all membrane components were taken against a tetahydro

furan background. Dioctyl phthalate, tetraphenyl borate,

dinonyl bipyridine and the cadmium complex all absorb in the

region from 233 to 280 nm. However, DNBP and the cadmium-

DNBP complex had an additional absorption peak(s) in the 280

to 320 nm region. It is this second region that will allow

the monitoring of the ligand concentration. For the

spectral analysis to be successful, there must be either a

shift in an absorption peak or a significant difference in

the molar absorptivities between the complexed and

noncomplexed ligand.

As can be seen by Figure 3.12, which shows the

spectra of the dissolved membrane (dissolved in THF), all of

the individual peaks in the 280 to 320 nm region have been

averaged into one peak, therefore no peak shift analysis can

be done. Fortunately, there is a large difference in the

molar absorptivities between the complexed and noncomplexed

ligand. Better linearity was obtained when peak areas over

the 280-320 nm region was used as opposed to peak height at

one wavelength (292 nm). Therefore, a proportionality

constant which is proportional to all of the single molar

absorptivities in the region of interest is used instead of

a single molar absorptivity.

1 . 6 0 +

1.40 ..

1 . 2 0 . .

1 . 0 0 . .

. 8 0 . .

. 6 0 . .

.40 ..

. 2 0 . .

240 260 2B0 300 320

WAVE LENGTH Cnm)

Figure 3.10 UV Spectra of DNBP.

ID (_>

<X CD or O in cc cr

1.80 j.

1 . 6 0 . .

1.40 „

1 .20 . .

1 . 0 0 . .

. 8 0 . .

. 6 0 . .

.40 ..

. 2 0 . .

240 260 260 300 320

VRVE LENGTH (nn)

Figure 3.11 UV Spectra of Cd-DNBP.

3.40

63

3.00

2 . 6 0

2 .20

UJ CJ £ 1 . 8 0 CD OC Q cn CQ a: 1.40

1 . 0 0

. 6 0

. 2 0

240 260 280 300 320

WRVE LENGTH (nm)

Figure 3.12 UVSpetrum of Dissolved Cd Electrode P olymer ffembrane (dissolved in THF).

64

<£ UJ 0£ CC

5d cr UJ Q_

1 2 . 0

DNBP CONCENTRATION UK)

Figure 3.13 UV spectral Analysis Calibration C.urve for DNBP D atermination at (1) 234-280 nm and (2) 280-320 nm.

65

2 6 . 0

2 2 . 0

18.0 <r UJ tc. = 14.0 xz CC LU

10 .0

6 . 0

2 . 0

Cd-DNBP CONC. (mM]

Figure 3.14 UV Spectral Analysis calibration Curve for Cd-DNBP Determination at (1) 234-280 and (2) 280-320 nm.

+ •f .01 .04 .06 .09 . 11

66

The percentage of complexed sites and concentration

of free ligand within the membrane was calculated according

to the procedure in Appendix 1. The summary of these

calculations are shown in Table 3.4 and in Figure 3.15. The

data from the UV analysis indicates there indeed is an

increase in the number of complexed sites as a function of

time or the number of calibrations. After the electrodes

had been conditioned for 3 days in millimolar cadmium

acetate, 50% of the DNBP is in the complexed form. Ten

calibrations later, 67% of the DNBP is complexed and the

electrodes are no longer functional. The next question is

whether this decrease of 17% in the available free exchange

sites could make the difference from a functioning to a

nonfunctioning electrode.

In section 1.1.2 we saw that the overall

distribution constant of an ion into a membrane containing a

neutral carrier is described by equation 3. The last term

in equation 3 is of particular interest: (cs/ks)n shows the

dependence of the overall distribution constant on the

concentration of noncomplexed ligand sites. Simon and Oesch

(39) showed that the limiting ligand concentration in the

membrane (the concentration at which the electrode no longer

responds properly) was 10~4 mol/1 for ligands with a 1:1

stoichiometry with the primary ion. Hence, according to

equation 3, a neutral carrier that forms a 3:1 ligand to

67

6 0 . 0

70.0 ..

a. m z a

i

S 6 0 . 0 . .

a cc c_J

50.0 ..

NUMBER OF CALIBRATIONS

Figure 3.15 Plot Indicating Cd Loading of P-olymer Membrane as a Function of the Number of Calibrations orT.ime.

68

Table 3.4 Results of DV Spectral Analysis on Cacbnium PMISE Membranes.

NUMBER WEIGHT OF PEAK PEAK AREA FREE OF MEMBRANE AREA PER LIGAND

CALIBRATIONS (mg) GRAM CONC.(M)

0 13 29.7 2285 0.085

1 13 31.0 2384 0.078

2 12 28.5 2375 0.078

3 14 35.0 2460 0.068

10 15 40.5 2700 0.056

69

metal complex, such as DNBP, should have a limiting

concentration of about 0.05 M. Fifty percent DNBP

corresponds to a concentration of 0.085 M, thus we would

expect the electrode to behave in a normal sense. However,

33% DNBP translates into a concentration of 0.056 M, which

is very close to the limiting concentration. Therefore it

should not be surprising that under the above conditions the

electrode response begins to deteriorate.

The excessive uptake of cadmium into the membrane is

most likely due to the extremely large formation constant

exhibited by DNBP. Dinonyl bipyridine, like most other

ligands that form 3:1 complexes, have overall log formation

constants two to three times larger than the 1:1 complex

forming macrocycles. Section 3.1.2 illustrates that the

DNBP based electrode was more selective towards copper than

any of the other transition metals studied. However, when a

DNBP based PMISE was conditioned to respond to copper it

didn't function at all. Again, this is undoubtedly due to

the lack of free ligand sites, for the CufDNBPjj complex has

a formation constant six log units greater than that of the

Cd{DNBP)g complex. The continuous loading of the polymer

membrane with cadmium ions will give rise to an inconsistent

concentration of free ligand in the membrane. According to

equation 3, a change in the free ligand concentration

produces a change in the distribution of the ion into the

membrane. This variation in the free ligand site

70

concentration explains the poor reproducibility demonstrated

by the cadmium PHISE. Also, with the DNBP concentration

being so near the limiting concentration, small changes in

pH, temperature or ionic strength could perhaps be

manifested as larger than expected deviations in the slope

and intercept.

3.2 Electrocapillary Studies of Neutral Surfactant Adsorption at £he Liquid/Liquid Interface

The electrocapillary curve describes the variation

in the surface or interfacial tension as a function of

electrical potential. When electrocapillary measurements

are made while varying the activity (which will be

approximated by concentration) of a species, important

information dealing with interfacial adsorption of that

species can be acquired. Information on the extent a

particular species adsorbs to the interface between two

liquids can be applied to improving and obtaining a better

understanding of such processes as solvent extraction. Not

only does electrocapillary studies give insight into

processes occuring at the liquid/liquid interface, but also

allows one to get a better understanding of the interfacial

structure itself. Interfacial structure studies are usually

.carried out by one of two types of experiments:

electrocapillary measurements as discussed above or

capacitance measurements via AC impedance techniques. For

71

the most part, capacitance measurements are more popular

since integration is required to obtain charge density and

electrocapillary curves (although the point of zero charge

must be determined independently) which tends to average out

experimental uncertainties. Electrocapillary curves have to

be differentiated to get charge density and capacitance

data. Electrocapillary experiments were chosen purely out

of convenience since ion transfer studies were being carried

out using current-scan polarography. The polarographic

apparatus was shown to produce accurate interfacial tension

measurements from drop time data (40) and therefore yield

good electrocapillary curves.

3.2.1 Electrocapillary Measurements and Surface Excess Determinations

The neutral surfactant studied in this work is

Triton X-100. Trition X-100 was present in the aqueous

electrolyte at concentrations varying from 5x10"^ M to

1x10"^ M. The 1 M magnesium sulfate aqueous electrolyte

(with Triton X-100) and 0.02 M THA+,TPB~ (in DCE) organic

electrolyte were preequilibrated two hours prior to

electrocapillary measurements. The electrocapillary curves

collected for the four different concentrations of Triton

X-100 are illustrated in Figure 3.16. Each individual

electrocapillary curve represents the average of five curves

with relative standard deviations of less than five percent.

As the concentration of the surfactant is increased, there

72

27.0

2 6 . 0

25.0

24.0 a aJ%O® o m * A A a* 154 _ O Offl A

• CE ® ©A 23.0

2 2 . 0

2 1 . 0

* .. 2 0 . 0

19.0

-.20-.100.0 .10 .20 .30 .40 .50

POTENTIAL (V)

Figure 3.16 Electrocapillary Curves for Varying Concentrations of Triton X-100. Interfacial Tension in D.yne/cm.

73

is a significant decrease in the interfacial tension as

depicted in Figure 3.16. This variation in the interfacial

tension as a function of concentration can be utilized by

the Gibbs Adsorption Isotherm (equation 9) to determine the

relative surface excess of Triton X-100. Figure 3.17 shows

a plot of the interfacial tension, at the point of zero

charge, versus the log concentration of Trition X-100. The

data is linear through the concentration range used. More

concentrated solutions of surfactant were avoided because of

micelle formation. The surface excess at the PZC was

determined to be 2.44 +/- 0.118 xlO""** mol/cm2. This

surface excess value is of the same magnitude as the surface

excess of Triton X-100 at the pure chloroform/water

interface (obtained by the drop volume method) (41).

Frumkin (42) assumed that adsorption at the double

layer could be represented by a model composed of two

capacitors in parallel corresponding to the plate areas

which are uncovered and covered with adsorbate,

respectively. Such a model predicts a linear plot of

potential against log concentration of neutral adsorbate at

constant charge. Therefore, the question of whether Triton

X-100 is specifically adsorbed or is just residing in the

diffuse layer can be answered by this type of Esin-Markov

plot. Frumkin explained PZCs shifting in the positive

direction to be due to dipole orientation of neutral

2 6 . 0

25.5 ..

74

o KH tn

UJ 25.0 ..

C_) cc li­ft: UJ

24.5 ..

24.0 ..

9.5 10.5 11.5 12.8

- In CONCENTRATION

Figure 3.17 Gibbs Plot for the Determination of Triton X-100 Surface Excess. Interfacial Tension in dyne/cm and concen­tration in mol/1.

75

aliphatic adsorbates. Aromatic adsorbates were observed to

cause a negative shift in the PZC. This negative shift was

thought to be caused by pi electron interactions with the

surface of the electrode. Figure 3.18 shows such a plot

for Triton X-100 at the PZC. The point of zero charge is

observed to shift in the negative direction as the

concentration of Triton X-100 is increased. However,

although Triton X-100 does possess an aromatic

functionality, it seems unlikely that pi electron

interactions with mercury could be extrapolated to the case

of the liquid/liquid interface. In this particular

instance, the orientation of the neutral adsorbate, Triton

X-100, would be expected to have the hydroxyl groups in

contact with the aqueous phase and the aliphatic/aromatic

portion in the organic phase.

How the surface excess of a neutral surfactant

varies with the charge on an electrode is also an

interesting aspect of interfacial structure. When a neutral

molecule adsorbs to the mercury electrode, the molecules

maximum coverage occurs at low charges and goes to zero or

nearly to zero at high charges. This behavior is mainly due

to the neutral molecule having to compete with water

adsorption at the electrode. At high charges it is much

more difficult for the molecule to displace the strong

dipole interactions of water and the electrode. However,

the ascending water electrode at the aqueous/organic

200

g * 6 0 -q: < x o

O '20-q: LU N

O 80 -(— z o CL

10 I I 13 12

-In CONCENTRATION

Figure 3.18 Esin-Markov T,ype Plot for Triton X-100. Point of Zero Charge in mV and C.oncentration in mol/1. .

77

interface shows a much different type of behavior, as shown

in Figure 3.19. At high negative charges the Triton X-100

has a high surface excess. As the charge decreases the

surface excess remains fairly constant or increases slightly

until a charge of about minus one. The surface excess then

decreases through the PZC and reaches a minimum at about

plus three microcoulombs/cm^. This peculiar, unsymetrical

behavior is perhaps due to Triton X-100 interacting with the

aqueous electrolyte, MgSOjj. It has been demonstrated that

the sulfate anion forms a distinct hydrogen bond with an

alcohol functional group (44). At high negative charges,

the aqueous side of the interface has a high excess of

sulfate anions. The high excess of sulfate anions

interacting with the alcohol groups on Triton X-100 might

create a larger than expected excess of our neutral

surfactant. The decrease in surface excess coincides with

decreasing concentration of sulfate in the double layer as

the charge goes from negative to positive. The sudden

increase of surface excess at high positive charges may be

the result of electrostatic Interactions between the highly

positive aqueous double layer and the hydroxyl groups of

Triton X-100.

3.2.2 Charge Density and Capacitance Data.

As can be seen from equations 7 and 8, charge

density and differential capacitance data can be derived

in tn UJ u x UJ

LU (J cr li-IC =3 V)

4. SB..

S.50 ..

2.5B ..

I .S t ..

.50 ..

CHRRGE DENSITY

Figure 3.19 ?Plot,I.indicating Variation of Surface Excess {in mol/cm xlO ) with Charge Density (uC/cm ).

79

from the initial electrocapillary curves. Both charge

density and differential capacitance data for the Triton X-

100 experiments were acquired by graphical differentiation

of the electrocapillary curves. Figure 3.20 shows the

results of the first derivative, or the charge density

versus potential data. These four curves (each representing

a different concentration of Triton X-100) are fairly

similar to curves produced at the aqueous/mercury interface.

Both aqueous/metal and aqueous/organic (in the presence of

an adsorbed neutral species) charge density curves exhibit a

flat region at low charges, except this region is more

defined for the aqueous/organic case. This flat region

becomes more pronounced as the concentration of neutral

surfactant is increased.

Differential capacitance data was acquired by taking

the derivative of the charge density-potential curve. The

Gouy-Chapman Theory predicts differential capacitance to

vary with potential in a way-to produce a U-shaped curve.

The differential capacitance-potential curves from the

Triton X-100 experiments all have this basic TJ-shape (see

Figure 3,21). The four capacitance curves show a general

decrease in the differential capacitance as the

concentration of Triton X-100 is increased. This is due to

both a decrease in the dielectric constant in the double

layer (on the aqueous side) and an increase in the plate

80

-10.0

—8.0 ..

« 3 -2.0 a bl (9 QC S s CJ

2.0 ..

6.0 . .

H 1 1 1 t 1 H

«

•

H 1 1 1 1 1 1 -.20 0.0 .20 .40

-10.0

-6.0 ..

-2.0 .. ui cs

u 2. 0 . .

6.0 ..

H 1 1 1 1 1 1 V

H 1 1 1 1 1 1 -.20 0.0 .20 .40

POTENTIAL (V)

(a)

-10.0 •+—H 1 1 I 1 1—4-

-6.0 ..

CO

S -2. 0 a hi C9 tc ec CJ

2.0 . .

6.0 . .

H 1 1 1 1 1 1 1 -.20 0.0 .20 .40

-10.0

tn

tu <9

-8.0 ..

-2.0 . .

2.0. . u

6.0..

POTENTIAL tV)

(b)

H 1 1 1 1 i h

r

H 1 1 i 1 1 1 -.21 0.0 .20 .40

POTENTIAL IVJ

(c)

POTENTIAL tV)

(U>

Figure 3.20 Charge .Density .irata (uC/cr .i*") for F our D.ifferer.tC.Ot.cc::-trations of Triton X-IOC: (a) 5x10"^, (b) 1x13"^, (c) 5x1u"- d,iC-

(cl) lxlD"4 i;io 1 /1.

DIPFERENTIRL CRPRCITflNCE

ui pj oo u o c C\ Zl -s o n> -— o c. n co

c+ * -s ro —» gj —1 x c+ oo u • -* _J.

.raw -h -«S 2 O Q

o -h —• n ' -J r+*

• —j. _~u

CT Qj

O —' Z3 O X &I 1 *1^ - ' CJ o o o -»•

ci-£i

CJ Q %,v

o tn cu x r> — J CJ

! -—-r.ic » -n

TO - r.j

r> —j t'J X —' —h «:> o i -j LT

o CJ ui

>c -*i -J -J,

'? C-

I i c h

tn N ui M UI

n

•* o H m sc

r_

•B a H m z -t H 33 r~

<

DIFFERENTIRL CRPflCITRNCE

DIFFERENTIAL CRPRCITflNCE

UI M M U • ui ui ui • • • I • m m

DIFFERENTIRL CRPRCITflNCE

UI M M U • UI UI UI

82

distance of the "capacitor". A peculiarity in the capacity-

potential curve is the hump located between -100 and +100

millivolts. These humps are usually indicative of

adsorption of some charged species, thus it could possibly

be explained by the adsorption of tetraheptyl ammonium from

the organic phase electrolyte. However, Senda showed for

the tetrabutyl ammonium, tetraphenyl borate organic

electrolyte that no specific adsorption occurs (45). In the

presence of a neutral surfactant, these humps are caused by

large changes in adsorbate coverage (46). The location of

the hump coincides with the charge density on the electrode

where a large decrease in surface excess is observed (Figure

3.19). Along this same line, the large decrease in surface

coverage could lead to a significant decrease in the

"capacitor" plate distance, and therefore the small increase

in capacitance is observed.

3.2.3 The Effect of a Neutral Surfactant on Facilitated Ion Transfer

The effect of Triton X-100 on the nonfacilitated

transfer of potassium at the liquid/liquid interface has

been observed by Yoshida (47). Half-wave potential shifts

to less positive potentials was observed as the

concentration of Triton X-100 is increased. This behavior

indicates that Triton X-100 is facilitating the transfer of

potassium, such as a ligand like valinomycin would

83

facilitate the transfer of potassium across the

liquid/liquid interface.

A similar experiment with cadmium(II) was carried

out. The effect Triton X-100 has on the facilitated

transfer of Cd(II) in the presence of the ligand DNBP was

investigated. Cadmium acetate was present in the aqueous

electrolyte (0.2 M sodium acetate) at a concentration of one

millimolar. The organic phase consisted of 0.01 M THA+,TPB""

and 0.25 millimolar DNBP. The concentration of Triton X-100

was varied from 1x10"® M to 1x10"^ M. As the concentration

of the neutral surfactant was increased there was no

apparent shift in the half-wave potential of transfer.

However, as the concentration of Triton X-100 was increased

there was a significant kinetic enhancement of ion transfer.

The limiting current increases with increasing Triton X-100

concentration (see Figure 3.22) in a nonlinear fashion.

According to the Ilkovic equation, this increase in wave

height can only be accounted for by an increase in the

diffusion of DNBP in the organic phase. However, the

Ilkovic equation shows a square root dependence on the

diffusion constant and a log-log plot of the data produced a

straight line with a slope of 0.2, not the expected 0.5.

The above results indicate that Triton X-100 is effecting

the diffusion of DNBP not in a straight forward manner.

10

<

UJ cc 0C 3 O

O

8 -

C 6-

4-

10 30 50

TRI X- I 00 CONC.

Figure 3.22 (-Plot howing E-ffect of Triton X-100 Concentration (mol/1 xlO ) on |_ inviting Current of Cd-DNBPwave.

CHAPTER 4

CONCLUSIONS

The use of current scan polarography at the

liquid/liquid interface in the characterization of neutral

carrier PMISEs was demonstrated. However, it should be

emphasized that the suitability of a ligand as a neutral

carrier can only be evaluated with respect to a few of the

characteristics of a PMISE. Basic information on the

kinetics of ion transfer (kinetic or diffusion controlled,

reversiblity etc.) to indicate whether the electrode will

respond rapidly and reversibly, and whether the behavior of

the PMISE will conform to the thermodynamics of the system.

The response of the PMISE as a function of the primary ion

concentration; a nernstian or iionnernstian responding ion

selective electrode. Evaluation of selectivity

characteristics when kinetics are rapid enough to produce a

polarographic wave. Due to the rapid kinetics of the alkali

and alkaline earth cations, the prediction of selectivity

coefficients by this method would be most promising for

those metals. PMISE Characteristics which are inaccessible

by this technique are detection limits, dynamic range and

the PMISE lifetime. As far as using the organic phase to

model the PMISE membrane, a possible improvement would be to

85

86

use a polyvinyl chloride/DCE mixture at a stationary

electrode implementing an electrochemical technique such as

cyclic voltametry.

The use of the dinonyl bipyridine based PMISE in the

determination of divalent cadmium was evaluated. Good

detection limits were obtained (10~® M) and a reasonable

dynamic range of. three orders of magnitude. Prom a

practical stand point, this PMISE could be used in the

presence of cobalt{II), nickel(II), the alkali and alkaline

earth metals. However, the large interference of copper(II)

and zinc(II) will render the electrode useless in the

presence of these two metals. The PMISE lifetime is short,

but can be extended by EDTA conditioning.

The kinetics of ion transfer was for the first time

correlated to the selectivity of an ion-selective electrode.

It was shown that the selectivity coefficient between the

primary ion and an ion which has rapid kinetics (diffusion

controlled) could be predicted by differences in the

potential of ion transfer across the liquid/liquid

interface. In the case of ions with slow kinetics (slow

water exchange rates) where ion transfer is kinetically

controlled, the predicted selectivity coefficient was

demonstrated to be larger than the actual PMISE selectivity

coefficient. The data showed that the slower the water

exchange rate the larger the difference between the actual

87

selectivity coefficient and the selectivity coefficient

expected from thermodynamics exclusively. This large impact

the kinetics of ion transfer has on selectivity is most

apparent when there is a broad range of water exchange

rates, such as with the transition metal ions. However, the

applicability of this concept is not limited only to ion-

selective electrodes, but to any kinetically controlled

system. It would be particularly interesting to apply the

above argument to biological systems, such as membrane

transport, where thermodynamic equilibrium does not exist.

The results indicate that bidentate ligands, such

as dinonyl bypyridine, are suitable for applications as

neutral carriers for PMISEs with a couple of inherent

drawbacks. One disadvantage is the large formation

constants exhibited by such ligands, and the other is the

fact that the critical concentration of the free ligand in

the membrane is considerably higher than the more typical

1:1 stoichiometry complex forming macrocycles. The above

two problems combine to produce an ion-selective electrode

with a very short lifetime. Excessive ion loading caused by

the large formation constants produce free ligand

concentrations in the membrane below the critical

concentration. There are two different approaches to

eliminating the above problem. It was shown by simply

conditioning the electrodes in a solution of EDTA increased

the free ligand concentration above the critical

88

concentration to yield normal responding PMISEs. This

method of revitalizing the electrodes should be examined to

a greater extent; determine the effect of EDTA conditioning

on electrode characteristics over a long period of time.

The other approach which should be investigated is to

increase the initial concentration of the neutral carrier in

the membrane. Not only would the effect of increasing

ligand concentration on electrode lifetime be studied, but

how other electrode characteristics (reproducibility,

response, detection limit, selectivity etc.) are effected

would also have to be considered. In general, as far as the

role of a ligand as a neutral carrier, the large macrocylic

molecules are more superior to the type of ligands in

question here. However, this does not mean that these type

of ligands should be excluded, especially in the case of

transition metal ions where selective macrocycles are few.

The application of current scan polarography to the

study of interfacial adsorption at the liquid/liquid

interface was exhibited using Triton X-100 as an example.

It was shown by taking the appropriate experimental

precautions, good reproducible electrocapillary curves are

attainable. Application of the Gibbs adsorption isotherm to

the electrocapillary data allows the determination of

reliable relative surface excess values. It was possible to

yield good derivatives of the electrocapillarycurves to

89

produce charge density and differential capacitance data to

be used in the further study of interfacial structure. For

future studies of this sort, a computer program to take the

first and second derivatives (numerically) of the

electrocapillary curves should be written. Since the

electrocapillary curve cannot be generalized as any one type

of polynomial (second or third order), a moving polynomial

routine would have to be used. Future experiments should

investigate the adsorption of charged surfactants and their

effect on ion transfer. The surface activity of metal

chelates could also be studied using the above technique.

APPENDIX A

DETERMINATION OF NONCOMPLEXED DNBP CONCENTRATION

IN AN ISE POLYMER MEMBRANE

Example calculation for membrane 1 (0 calibrations)r

Initial amount of DNBP in membrane = (0.013g)(0.05)

= 6,5xl0"4g

Intitial number of mols of DNBP « 6.5xl0_4g/408g mol"1

= 1.6xl0~®mols

Concentration of DNBP in cuvette = 1.6xl0""6mols/0.011

= 1.6X10~4K

Predicted absorbance if all DNBP in noncomplexed form:

A = (Kdnbp)(cell path length)(Concentration)

where Kougp is proportionality constant for DNBP

A = (75,410) (1) (1.6xl0-4) =12.1

Difference in absorbance must be due to complexed DNBP,

therefore AA/AK = AC ; AA = 29.7 - 12.1 = 17.6 AK = 217,490

and

AC = 17.6/217,490 = 8.1xlO"5M

Percentage of complexed DNBP = (8. 1x10" Vl. 6xl0~4) 100

= 50%

Number of mols of noncomplexed DNBP = (0.5)(1.6x10"®)

= 8xl0"7mols

Concentration of noncomplexed DNBP = 8xl0~^mols/9,3xl0~®l

= 0.085M

90

REFERENCES

1. Freiser, H. (Ed.), Ion Selective Electrodes in Analytical Chemistry, Vol. 1, Plenum Press, New York, 1978

2. Freiser, H. (Ed.), Ion Selective Electrodes in Analytical chemistry, Vol. 2, Plenum Press, New York, 1980

3. Eisenman, G. (Ed.), Glass Electrodes for Hydrogen and Other Cations, Marcel Dekker, New York, 1967

4. Morf, W.E., The Principles of Ion Selective Electrodes and Membrane Transport, Elsevier Sci. Pub. Co., Budapest, Hungary, 1981

5. Covington, A.K. (Ed.), Ion Selective Electrode Methodology, Vol. 1 and Vol. 2, CRC Press, Boca Raton, FL, 1979

6. Moody, G.J., Thomas, J.D.R., in Reference 1, page 291

7. Simon, W., Morf, W.E., Meier, P., Structure and Bonding, Vol. 16 (1973), 113

8. Buck, R.P., in Reference 1, Chapter 1

9. Reference 4, page 5

10. Reference 4, page 8

11. Ammann, D., Morf, W.E., Anker, P.C., Meier, P., Simon, W., Ion Selective Electrode Rev., Vol. 5 (1983), 3

12. Morf, W.E., Simon, W., in Reference 1, Chapter 3, page 227

13. James, H., Carmack, G., Freiser, R., Anal. Chem., 44 (1972) , 856

14. Cattrall, R.W., Hamiton, I.e., Ion Selective Electrode Rev., Vol. 6 (1984) , 125

15. Freiser, H., in Reference 2, Chapter 2

91

92

16. Nernst, W., Riesenfeld, E.H., Ann. Phys., Vol. 8 (1902), 600

17. Dupeyrat, H., Guastalla, J., J. Chim. Phys., Vol. 53 (1956), 469

18. Koryta, J., Vanysek, P., in H. Gerischer and C.W. Tobias (Eds.), Advances in Electrochemistry and Electrochemical Engineering, Vol. 12, John Wiley & Sons, New York, 1981

19. Vanysek, P., Buck, P..P., J. Electroanal. Chem., 163 (1984), 1

20. Koryta, J., Ion Selective Electrode Rev., 5 (1983), 131

21. Vanysek, P., Electrochemistry on Liquid/Liquid Interfaces, Springer-Verlag, New York, 1985

22. Fujinaga, T., Phil. Trans. R. Soc. Lond., A305 (1982), 631

23. Kakiuchi, K., Obi, I., Senda, M., The Bull. Chem. Soc. Japan, 58 (1985), 1636

24. Samec, Z., Horoolka, D., Haracek, v., J. Electroanal. Chem., 135 (1982), 265

25. Koryta, J., Vanysek, P., Brezina, M., J. Electroanal. Chem., 75 (1977), 211

26. Bard, A., Faulkner, L., Electrochemical Methods: Fundamentals and Applications, page 491, John Wiley & Sons, New York, 1980

27. Martin, C.R., Investigations of Polymer Membrane Ion Selective Electrodes, Ph.D. Dissertation, University of Arizona, Tucson, AZ, 1980 .

28. Cunningham, L., Ion Selective Electrodes for Organic Cations and Pharmaceuticals, Ph.D. Dissertation, University of Arizona, Tucson, AZ, 1984

29. Reference 28, Appendices A, B and C

30. Yoshida, Z., Freiser, H., J. Electroanal. Chem., 162 (1984) , 307

93

31. Martin, C.R., Freiser, H., J. Chem. Educ., 57 (1980), 512

32. Rechnitz, G., Srinivasan, K., Anal. Chem., 41 (1969), 1203

33. Lin, S., Personal Communication

34. Harkins, W.D., Brown, in J. Am. Chem. S., 41: 499; 19

35. Yoshida, Z., Preiser, H., Inorg. Chem., 23 (1984), 3931

36. Tuth, K., Pungor, E., Anal. Chim. Acta., 64 (1973), 417

37. Ogyenfuss, P., Morf, W.E., Funck, R.J., Pham, H.V., Zund, R.E., Pretsch, E., Simon, W., in E. Pungor (Ed.) , 3rd Symposium on Ion Selective Electrodes, Elsevier, New York, page 73, 1981

3 8. Smith, R.M., Martell, A.E., Critical Stability Constants, Vol. 2, Plenum Press, New York, 1975

39. Oesch, U., Simon, W., Anal. Chem., 52 (1980), 692

40. Cunningham, I>., Freiser, H., Langmuir, 1 (1985), 537

41. Aprahamian, E., Role of the Interface in the Kinetics and Mechanism of Solvent Extraction Systems, Ph.D. Dissertation, University of Arizona, Tucson, AZ, 1985

42. Frumkin, A.N., Z. Physik. Chem., 35 (1926), 792

43. Frumkin, A.N., Donde, A., Kulvarskii, P., Z. Physik. Chem., 123A (1926), 321

44. Moroi, Y., Motomura, K., Matuura, R., Bull. Chem. Soc. Japan, 45(9), 2697

45. Kakiuchi, T., Senda, M., The Bull. Chem. Soc. Japan., 56 (1983), 1753

46. Delahay, P., in Double Layer and Electrode Kinetics, John Wiley. & Sons, New York, 1966

47. Yoshida, Z., Personal Communication