Manual 21377040

Manual 21377040
Chapter 1:
CHAPTER 1
Introduction
INTRODUCTION
In this chapter, a review of some electrochemical principles necessary to be considered
when one is performing an electrochemical experiment (i.e. mass transport and its modes,
and double layer structure) is presented. A review of the theory on electrochemical studies
of cobalt inorganic and organometallic compounds, electrochemical flow cells, and a theory
of electrocatalysis is also presented.
1.1
PRINCIPLES OF ELECTROCHEMISTRY
The electrochemical reaction is a heterogeneous electron transfer. It takes place at the
interface between the electrode and the solution. Only redox active species present at the
interface can undergo this electron transfer. The electrochemical reaction at the interface
makes the composition in the nearby solution different from the bulk of the solution further
away. Because the solution tends to become homogeneous, redox active species diffuse
between the bulk and the perturbed zone, in which the reactant is depleted, widens.
Convection due to heating or stirring makes the solution rapidly more homogeneous, thus
reducing this zone from about 100 m to less than 1 m [1].
Unlike most spectroscopic methods, electrochemical measurements are actually made on
only a minute fraction of the sample confined to a highly inhomogeneous environment, the
electrode-solution interface. The coupling and interplay within this region of such
phenomena as interfacial charge transfer, diffusional mass transfer, adsorption,
chemisorptions, homogeneous phase chemical reaction, convection and dissolution can
cloud the interpretation of electrochemical data and discourage the practically minded
analyst. On the other hand, electrochemistry offers an invaluable tool for fundamental
investigation of these processes, each of importance in its own right. In either case, the
ultimate success of the experimenter will depend on a firm grasp of the underlying physical
principles [2].
The interface between an electrolyte solution and an electrode has come to be known as the
electrical double layer [2]. The double layer is much thinner than the diffusion layer because
it is only a few molecular layers thick. Under the applied potential, the liquid and solid
1
Chapter 1:
Introduction
layers are charged at their interface. Heterogeneity is a common thread binding all
electroanalytical methods. The act of placing an electrode in contact with a solution creates
a phase boundary that differentiates identical solute molecules into two types; those at a
distance from the electrode and those close enough to participate in the fascinating mutual
interactions known collectively as electrochemistry. This is not a trivial distinction, for often
it is the bulk-phase properties alone which are of analytical concern [2].
Imagine, for example, a positively polarized mercury surface immersed in a solution of
sodium chloride. In this case, the positive electrode surface attracts the negative chloride
ions because of electrostatic action, van der Waals’ forces, and specific chemical effects. As
a result, a layer of essentially non-hydrated chloride ions will accumulate very close to the
electrode surface, forming what is known as the inner Helmholtz layer. Because of the
presence of this negatively charged chloride ions, a double layer is said to exist. Just beyond
this layer is a second layer of tightly held hydrated chloride ions, a layer that marks the
boundary of the outer Helmholtz layer. Beyond this a diffuse layer extends with a net charge
whose ionic atmosphere contains ions of one sign in excess of their normal concentration
and those of the other sign in defect [3]. This assemblage of charged layers is commonly
referred to as simply the double layer (Fig. 1.1).
The simplest reactions involve only mass transfer of a reactant to the electrode,
heterogeneous electron transfer involving non adsorbed species and mass transfer of the
product to the bulk solution. Mass transfer, the movement of material from one location in
solution to another arises either from differences in electrical or chemical potential at the
two locations or from movement of a volume element of solution. The modes of mass
transfer are [4]:
1. Migration: movement of a charged body under the influence of an electric field (a
gradient of electrical potential).
2. Diffusion: movement of a species under the influence of a gradient of chemical
potential (i.e. a concentration gradient).
3. Convection: stirring or hydrodynamic transport. Generally fluid flow occurs because
of natural convection (convection caused by density gradients) and forced
convection, and may be characterized by stagnant regions, laminar flow, and
turbulent flow.
2
Chapter 1:
Introduction
Electrode
++++
Inner Helmholtz Layer
Cl– Cl– Cl– Cl–Cl– Cl– Cl– Cl–
–
–
–
–
H2O–Cl H2O–Cl H2O–Cl H2O–Cl
Na+ Cl–
Cl– Na+
Na+ Na+
Cl– Na+
Na+ Na+
Na+ Na+
Outer Helmholtz Layer
Na+ Na+ Na+
Na+ Cl– Na+
Cl– Na+ Na+ Na+
Na+ Na+ Cl– Na+
Cl– Na+ Na+ Na+
Solution
Diffuse Layer
Figure 1.1 Structure of an electrical double-layer.
Mass transfer to an electrode is governed by the Nernst-Planck equation, written for onedimensional mass transfer along the x-axis as [4]:
J i ( x) = − Di
δCi ( x) zi F
δφ ( x )
DiCi
−
+ Ci v( x)
δx
δx
RT
(1.1)
where Ji(x) is the flux of species i (mol s–1cm–2) at distance x from the surface, Di is the
diffusion coefficient (cm2 s–1), Ci(x)/ x is the concentration gradient at distance x, φ (x)/ x
is the potential gradient, zi and Ci are the charge (dimensionless) and the concentration
(mol.cm–3) of species i, respectively, and v(x) is the velocity (cm s–1) with which a volume
element in solution moves along the axis. The three terms on the right-hand side represent
the contributions of diffusion, migration, and convection, respectively, to the flux [4].
A rigorous solution is generally not very easy when all three forms of mass transfer are in
effect; hence electrochemical systems are frequently designed so that one or more of the
contributions to mass transfer are negligible. For example, the migrational component can
be reduced to negligible levels by addition of an inert electrolyte (a supporting electrolyte)
3
Chapter 1:
Introduction
at a concentration much larger than that of the electroactive species. Convection can be
avoided by preventing stirring and vibrations in the electrochemical cell [4].
In the bulk of solution (away from the electrode), concentration gradients are generally
small, and the total current is carried mainly by migration [4] where all charged species
contribute. For species j in the bulk region of a linear mass-transfer system having a crosssectional area A, ij = im,j
ij =
z 2 j F 2 AD j C j
RT
⋅
δφ
δx
(1.2)
where ij is the current component at any value of x arising from a flow of species j and im,j is
the migration current of species j.
In migration, the driving force responsible for moving the species to (or away from) the
electrode surface is the force exerted on a charged particle by a potential gradient existing in
the body of the solution. Thus, for example, a potential gradient in an electrolytic solution
will cause a current to flow within the body of the solution such that negative species will
move one way, positive the other, and neutral species will remain unaffected. The rate of
movement of a charged particle depends upon the magnitude of its charge, size, degree of
hydration, etc. In conductometric measurements, migration is the sole factor that limits the
current. In other cases, such as polarography, it is desirable to eliminate the contribution of
migration, and this is accomplished by addition of an excess (about 100-fold) of an inert
electrolyte that, in turn, decreases the potential gradient to a value sufficiently small so that
diffusion and/ or convection processes becomes current-limiting [3]. In general, it simplifies
the mathematical treatment of electrochemical systems by elimination of the φ / x term in
the mass transport equation [4].
Although migration carries the current in the bulk solution during electrolysis, diffusional
transport occurs in the vicinity of the electrodes, because concentration gradients of the
electroactive species arise there. Indeed, under some circumstances, the flux of electroactive
species to the electrode is due almost completely to diffusion [4].
4
Chapter 1:
Introduction
In many respects, the simplest and best understood process influencing electrochemistry is
diffusion. Diffusion is a factor virtually in every type of electroanalytical measurement, yet
it is most often introduced as a set of elemental laws, devoid of physical significance [2].
This transport mode has its origin in a gradient of chemical potential or, more simply, in a
concentration gradient. Thus, if the concentration of the species in the bulk of the solution is
greater than its concentration at the electrode surface (because of the electrode reaction), the
species will tend to diffuse from the bulk of the solution towards the electrode surface.
Whereas the direction of diffusion is from the regions of larger to smaller concentrations, its
rate is proportional to the magnitude of the concentration differences and to certain
characteristic properties of the diffusing species and medium. Three types of diffusion
transport exist, mainly linear, cylindrical and spherical diffusion. Diffusion that takes place
in a single direction (diffusion to a plane surface) is termed linear diffusion and is
mathematically and experimentally the simplest case [3].
Depending on the size of the electrode and the volume of electrolytic solution used, one can
distinguish between three limiting cases of diffusion. The simplest case is an electrode in a
thin-layer cell with a very low ratio of cell volume to electrode surface. Under these
conditions mass transport within the cell is negligible and diffusion is the only mode of
mass transport. By reducing the ratio between the electrode surface and the electrolyte
volume, one approximates the normal situation of a voltammetric experiment with semiinfinite planar diffusion. With the transition to extremely small electrode surfaces, the
conditions change yet again, and the diffusion process becomes dependent on the size and
geometry of the electrode [5].
As one might expect, the voltammetric current-voltage curves of these three cases differ
markedly. In the case of a thin electroactive layer, the cathodic and anodic waves appear as
perfect
mirror
images.
“Normal”
electrodes
produce
the
characteristic
cyclic
voltammograms, and extremely small electrodes (ultramicroelectrodes) yield steady-state
current-voltage curves, which resemble the classic polarograms as well as the currentvoltage curves of rotating electrodes (Figure 1.2) [5].
5
Chapter 1:
Introduction
5.50
Epa
5.E-06
4.E-06
4.50
E1/2
Ipa
3.E-06
E
3.50
I (A )
I (µ A )
2.E-06
2.50
1.E-06
0.E+00
1.50
Ipc
-1.E-06
0.50
-2.E-06
-3.E-06
-0.25
Epc
-0.15
-0.05
0.05
0.15
0.25
0.35
0.45
-0.50
-0.25
-0.15
-0.05
E (V)
0.05
E (V)
0.15
0.25
0.35
0.45
Figure 1.2 Typical voltammetric current-voltage curves. Left: for a semi-infinite diffusion obtained
using a “normal” electrode; right: for semi-infinite hemispheric diffusion obtained using extremely
small electrodes (ultramicroelectrodes). E = switching potential, E1/2 = half-wave potential, Epa and
Epc = anodic and cathodic peak potential, Ipa and Ipc = anodic and cathodic peak current. A typical
curve obtained in thin-layer solution can be obtained from Ref [5].
Convection is accomplished whenever the solution bearing the species is stirred into the
path of the electrode. This is often called hydrodynamic transport. This stirring action
increases the rate of transport of the species to the electrode and quantitative treatment of
this mode of mass transport is exceedingly difficult. Convection is frequently employed in
electroanalytical techniques (examples include the rotating electrode, mercury electrodes,
vibrating electrodes, and even the stirring action employed at the working electrode in
coulometric titrations, etc.), because of its increased sensitivity. Frequently one must resort
to the use of empirical expressions in the interpretation of these techniques [3].
1.2
ELECTROCHEMISTRY OF COBALT
ORGANOMETALLIC COMPOUNDS
INORGANIC
AND
Most electrochemical studies of cobalt were mainly for analytical purposes; simultaneous
determinations of cobalt, nickel, copper and so on were shown to be possible by using
complexing agents. The modern trend in the field of polarography of cobalt was to establish
the relationship between the structures or bonding types of cobalt complexes and their
electrode processes; this was because cobalt complexes, among transition metal complexes,
are large in number and exhibit substantial variety in their bonding nature from compound
6
Chapter 1:
Introduction
to compound. In these studies, non-aqueous polarography attracted special attention and
promised to provide clues for the elucidation of electron transfer mechanisms of the
electrode reactions and their correlation to the electronic configuration of a great variety of
mixed ligand complexes [6].
The most striking feature of non-aqueous polarography is that, by adopting aprotic nonaqueous solvents, the Co(III) complexes, which would otherwise lead to a loss of ligands in
the lower oxidation states of cobalt, are reduced in a stepwise fashion with a complete
retention of the original configuration. For example, the pathway Co(III)
Co(II)
Co(I)
Co(0) takes place without a loss of ligands in aprotic solvents, although the appropriate
valence orbital are, of course, delocalized over the entire complex molecule [6].
Voltammetric
techniques
such
as
current-controlled
oscillopolarography,
cyclic
voltammetry, controlled-current or –potential electrolytic studies, and investigations of
anodic oxidation have attracted much attention in the field of electrochemistry of cobalt as
powerful tools, along with classical or conventional polarographic and chronopotentiometric
studies, for probing the redox processes of the Co(III)/Co(II) and Co(II)/Co(I) couples in
solution. This is partly because a number of electrode processes of an inert-inert type with
-electron systems, which were recently discovered for
-bonded cobalt complexes in
organic solvents, made it possible to follow structurally the fate of the electrolyzed cobalt
complexes in solution, not only in the reduction process but also in the counter process of
oxidation with auxiliary instrumental methods of measurements such as ESR, NMR and
magnetic susceptibility [6].
The electrode processes of the completely “inert-inert” or “inert-labile” types are the ones in
which it is easiest and simplest to determine the assignments of the polarographic waves.
The terms “inert” and “labile” are used with regard to the “lability” of ligands bound to the
cobalt towards solvolysis, i.e., the ligand exchange reactions with solvent molecules, whilst
the terms as defined by Taube [7] are independent of the nature of the medium. For either
extreme case this consists of the complex remaining structurally intact or, alternatively,
undergoing a rapid dissociative equilibrium upon reduction. Moreover, the complete
retention of the structures in non-aqueous media makes it possible to relate the structures in
both the oxidized and reduced forms to the redox reaction of the electrode processes. For
this reason cyclic voltammetry or controlled-current oscillopolarography are powerful
7
Chapter 1:
Introduction
methods not only for the examination of the reversibility of the electrode reaction, but also
for the identification of the cobalt species responsible for the conventional polarographic
waves. Here the interpretation is facilitated because the cathodic reductions can be observed
and compared with the anodic oxidations in the same experiment [6].
The electrochemical studies of cobalt complexes with tetraphenylporphirins (TPP) are well
documented in the literature [6, 8–13]. The electrochemical behaviour of cobalt (II)
complexes with tetraphenylporphirins was studied by cyclic voltammetry. The [CoIITPP]
complex and its products, which were obtained by its controlled-potential electrochemical
oxidation, were studied by electrochemical spin resonance (ESR) spectra. The first
oxidation occurred at Ep = + 0.52 V, on the central Co(II) atom and all subsequent
oxidations occurred at the TPP ligand. The potentials of the central cobalt (II) oxidation
showed a linear dependence on the third ionization potential of the ion, whereas the ligand
oxidation potentials were approximately independent of the cobalt (II) ion. This
distinguishes between the cobalt oxidation change and ligand oxidation [6, 8]. Cyclic
voltammogram distinctly showed three one-electron reversible steps [8]:
+ 0.52 V
[Co II (TPP )] ←

→ [Co III (TPP )] +
(1.3)
+ 1.19 V
[Co III (TPP )]+ ←

→ [Co III (TPP )]2+
(1.4)
+ 1.42 V
[Co III (TPP)]2+ ←

→ [Co III (TPP)]3+
(1.5)
On the other hand, the free ligand H2TPP gave two one-electron irreversible oxidation steps
[8]:
+ 1.00 V
+ 1.20 V

→ [ H 2TPP ]+ ←

→ [ H 2TPP ] 2+
[ H 2TPP ] ←
The
voltammetric
behaviour
of
cobalt
(II)
(1.6)
with
chloro-N-methyl- , , , -
tetraphenylporphirins (ClNCH3TPP) was studied by cyclic voltammetry [9]. The half-wave
potentials for the reversible metal oxidation in acetonitrile occurred at 0.77 V for Co(II) –
Co(III) (all E1/2 values were reported vs. Ag AgCl). Following the oxidation of the metal
centre, the porphirin ligands were oxidized. The half-wave potentials for oxidation of the
tetraphenylporphirin ligand in ClNCH3TPP occurred at 1.30 and 1.60 V. The ligand
oxidations for the N-methyltetraphenylporphirin complexes in general were found at Ep
8
Chapter 1:
Introduction
values of 1.2 – 1.3 and 1.4 – 1.6 V. The cyclic voltammogram of the corresponding zinc (II)
complex was recorded to verify the assignment of metal centre and ligand oxidation. The
oxidation of the ClZnN-CH3TPP complex occurred at 1.05 and 1.5 V with no wave
appearing
in
the
region
0.0
–
0.9
V.
Cyclic
voltammetry
, ,, -
of
tetraphenylporphinatocobalt (II) in benzonitrile with 0.10 M tetraphenyl ammonium
perchlorate gave the values E1/2 = 0.50 V [Co(II) – Co(III)], 1.19 V (ligand oxidation), and
1.6 V (ligand oxidation) [9], agreeing reasonably with the literature values of 0.52 V, 1.19 V
and 1.42 V [8].
The electrochemistry of five- and six-coordinate cobalt (III)
-bonded porphyrins was
reported in pyridine (py), tetrahydrofuran (THF), and methylene chloride (CH2Cl2)
containing 0.1 M tetrabutyl ammonium perchlorate (TBAP) or 0.1 M tetrabutyl ammonium
hexafluorophosphate (TBAPF6) as supporting electrolyte [10]. Each complex undergoes up
to two reductions and two oxidations, all of which occur at the porphyrin
ring system.
Cyclic voltammogram of (TPP)Co(CH3) and (TPP)Co(CH2Cl2) each revealed two reversible
one-electron oxidations and a single one-electron or multi-electron reduction within the
potential range of the solvent. The singly reduced (TPP)Co(CH2Cl) and (TPP)Co(C2H5)
complexes were stable in THF, but this was not the case for (TPP)Co(CH3), which revealed
two
one-electron
waves
in
this
solvent.
Electrooxidized,
(TPP)Co(CH3)
and
(TPP)Co(CH2Cl) were relatively stable in CH2Cl2 on the conventional cyclic voltammetry
time scale, and both neutral derivatives were characterized by two well-defined one-electron
oxidations at potentials of 0.96 and 1.19 V. Cyclic voltammogram of (TPP)Co(C2H5)(py) in
pyridine showed one irreversible oxidation at Ep = 0.81 V and one irreversible reduction at
Ep = – 1.49 V leading to the formation of [(TPP)Co(py)2]+ and [C2H5N(C2H5)]+ [10].
Cyclic voltammetry and controlled-potential electrolysis were used to investigate the
electrochemical reduction of CoBr2 in dimethylformamide (DMF) and DMF + pyridine
mixtures [14]. Cyclic voltammogram of CoBr2 at a platinum electrode displayed an
irreversible reduction wave at – 1.17 V versus SCE. The reduction of Co(II) led to a Co(I)
species at – 1.4 V. On the time scale of slow cyclic voltammetry or of an electrolysis, Co(I)
led to solid Co(0) and Co(II) by disproportionation. The regeneration of Co(II) was effective
until its total consumption, which required two electrons per molecule. After the scan rate
was increased, a second reduction wave appeared at – 1.75 V, and it was assigned to the
product generated at – 1.4 V. As the scan rate was increased, the disproportionation reaction
9
Chapter 1:
Introduction
was progressively kinetically frozen and more Co(I) remained in the diffusion layer when its
reduction potential was reached (E = – 1.75 V), the reduction of Co(I) also led to solid
Co(0). The electrochemical reactions were as follows [14]:
Co II X 2 + e − ↔ Co I X + X −
(1.7)
C
2Co I X → Co( s ) + Co II X 2
(1.8)
E
Co II X 2 + e − ↔ Co I X + X −
(1.9)
Co I X + e − → Co( s ) + X −
(1.10)
E = – 1.4 V: E
E = – 1.75 V:
The electrochemical behaviour of CoBr2 was also studied in acetonitrile and pyridine
mixtures in the presence of allyl acetate (1 molar equivalent vs. CoBr2) and increasing
amounts of ethyl 4-iodobenzoate [15]. In the absence of aromatic halide the voltammogram
exhibited two successive reduction waves. The first wave occurred at -1.3 V ascribed to the
reduction of Co(II) to Co(I) followed by fast complexation with allyl acetate which led to
( 2-allylOAc)cobalt(I), the latter complex being more stable than the original Co(I) and
giving rise to a reoxidation wave Co(II). The resulting ( 2-allylOAc)cobalt(I) was reduced at
the second reduction wave Ep = -1.5 V to ( 2-allylOAc)cobalt(0). In the presence of ethyl 4iodobenzoate, the intensity of the wave at -1.3 V increased and reached a maximum (twice
the initial intensity), and the oxidation wave for Co(II) decreased [15].
The first oxidative electrochemistry of cobalt (III) corroles was reported in THF, DMF,
benzonitrile (PhCN), and CH2Cl2 containing 0.10 M TBAP as supporting electrolyte [16].
The investigated compound was represented as (OMC)Co(PPh3) where OMC was the
trianion of 2,3,7,8,12,13,17,18-octamethyl corrole. The CV of the complex revealed up to
three oxidations and two reductions waves depending upon solvent. The (OMC)Co(PPh3)
complex revealed three one-electron reversible oxidations in PhCN at high scan rate, the
number of abstracted electrons was calculated by controlled-potential coulometry as well as
by analysis of the current-voltage curves obtained by cyclic voltammetry. The first
oxidation occurred at 0.19 V in PhCN while the latter two processes occurred at 0.76 and
1.54 V. The first two one-electron waves were reversible at all scan rates, but the third
oxidation became irreversible at lower scan rates (0.5 V/s). (OMC)Co(PPh3) complex
showed three one-electron oxidations in CH2Cl2, at 0.18 V, 0.80 V and 1.68 V.
(OMC)Co(PPh3) complex showed two one-electron oxidations in THF and DMF at 0.30 V,
10
Chapter 1:
Introduction
0.80 V (THF) and 0.25 V, 0.83 V (DMF) respectively. The electrochemical reactions were
as follows [16]:
(OMC )Co III ( PPh3 ) ↔ [(OMC )Co III ( PPh3 )]+ + e −
(1.11)
[(OMC )Co III ( PPh3 )] + ↔ [(OMC )Co III ( PPh3 )]2+ + e −
(1.12)
[(OMC )Co III ( PPh3 )] 2+ ↔ [(OMC )Co III ( PPh3 )]3+ + e −
(1.13)
Complexes
of
(5,10,15-tri-X-phenyl-2,3,7,8,12,13,17,18-octamethylcorrolato)cobalt(III)
triphenylphosphine, (OMTX-PC)Co(PPh3), where X = p-OCH3, p-CH3, p-Cl, m-Cl, m-F, oCl, o-F, or H, were synthesized and characterized in non-aqueous media using
electrochemical, spectroelectrochemical, and EPR techniques [17]. Each cobalt (III)
derivative showed two one-electron reductions, the first of which involved a Co(III)/Co(II)
conversion and concomitant loss of the bound PPh3 ligand. Four one-electron oxidations
were also observed for the investigated compounds, and this contrasts with the oxidative
properties of related cobalt (II) porphyrins which revealed a maximum of three one-electron
oxidation waves under similar conditions. The first one-electron oxidation of each
cobalt(III) corrole was metal-centred and resulted in formation of Co(IV) corrole as
ascertained by EPR spectroscopic characterization of the electrogenerated species [17].
The fact that Co(III) corroles can undergo four one-electron oxidations was not previously
reported, but presumably this reaction might also occur for (OMC)Co(PPh3) at very positive
potentials, i.e., at values of E1/2 greater than 1.9 V vs. SCE in 0.2 M TBAP in PhCN [17].
The fourth oxidation of (OMTXPC)Co(PPh3) was quasi-reversible for all eight compounds
investigated in that study, and this reaction was not examined in detail due to its proximity
to the anodic potential limit of the solvent. However, comparison of cyclic voltammograms
of (OMTp-ClPC)Co(PPh3) [E½ = 0.31, 0.85, 1.45, and 1.8 V] and (Tp-ClPP)Co(PPh3) [E½ =
0.30, 1.2, and 1.5 V] seemed to rule out the forth oxidation state as involving the bound
phosphine ligand, since the fourth oxidation state was not present on a cyclic
voltammogram of (Tp-ClPP)Co(PPh3). Further comparison of the (OMTp-ClPC)Co(PPh3)
and (Tp-ClPP)Co(PPh3) voltammograms showed that the first one-electron oxidation of
both macrocycles occurred at virtually the same potential, i.e., + 0.31 V for the corrole and
+ 0.30 V for the porphyrin. The first oxidation of the neutral porphyrin corresponded to the
11
Chapter 1:
Introduction
Co(II)/Co(III) electrode reaction, while that of the corrole was assigned to the
Co(III)/Co(IV) reaction [17].
Complexes of the type [Co(CO)L( -C5H5)] (where L = P(C6H11)3 or PPh3) were studied by
cyclic voltammetry and shown to undergo one-electron oxidation to the radical cations
[Co(CO)L( -C5H5)]+. The CV of the complex [Co(CO){P(C5H11)3}( -C5H5)] showed a
reversible one-electron oxidation in CH2Cl2 at a platinum electrode [18]. The other
compounds studied were similarly well behaved except for [Co(CO)(PPh3)( -C5H5)] which
displayed a tendency to deposit a film on the platinum electrode during oxidation in CH2Cl2.
For this compound the electrode had to be cleaned after every scan; the oxidation potential,
E°, was estimated at 0.13 V. In THF the voltammetry was cleaner (E° = 0.21 V) but the
oxidation was not as chemically reversible. In addition, there was a second irreversible wave
at 0.75 V, which was not present in CH2Cl2. Since the new wave was peculiar to THF it was
believed to be due to the oxidation of the solvated radical cation [Co(CO)(PPh3)(THF)( C5H5)]+ to the diamagnetic dication [18].
The compound [Co(CO)2( -C5H5)], was also studied in the presence of added [Fe( C5H5)2][PF6]. When [Fe( -C5H5)2][PF6] (E° = 0.43 V) was added to [Co(CO)2( -C5H5)] (E°
~ 0.97 V) in CH2Cl2, no reaction occurred on the basis of E° values [18]. In the presence of
PPh3, however, rapid oxidation occurred at room temperature to give [Co(PPh3)2( C5H5)][PF6]. It must be noted that neither the ferricinium ion nor the dicarbonyl undergone
detectable reactions with PPh3 at room temperature. This means that the formation of
[Co(PPh3)2( -C5H5)][PF6] must, therefore, occur via the mechanism shown in equation 1.14
to 1.16 below [18].
[Co(CO ) 2 (η − C5 H 5 )] + PPh3 ↔ [Co(CO )( PPh3 )(η − C 5 H 5 )] + CO
(1.14)
[Co(CO )( PPh3 )(η − C 5 H 5 )] ↔ [Co (CO )( PPh3 )(η − C 5 H 5 )] +
(1.15)
[Co(CO )( PPh3 )(η − C 5 H 5 )]+ + PPh3 ↔ [Co( PPh3 ) 2 (η − C 5 H 5 )]+ + CO
(1.16)
Electrochemical behaviour of [{Co( -NO)( -C5H5)}2][PF6] was obtained in CH2Cl2 by
cyclic voltammetry [19]. The compound was reduced to [{Co( -NO)( -C5H5)}2] at 0.34 V,
and oxidized to [{Co( -NO)( -C5H5)}2]2+ at 1.17 V. The cyclic voltammogram also showed
a second, irreversible, reduction wave at – 1.40 V. However, the peak current appeared
12
Chapter 1:
Introduction
larger than those of the other redox processes described above, but the wave was close to
that of the base electrolyte and may correspond to the initial formation of [{Co( -NO)( C5H5)}2]–. In the presence of PPh3, [{Co( -NO)( -C5H5)}2] was un-reactive but underwent
instant metal-metal bond cleavage to give [Co(PPh3)(NO)( -C5H5)]+. The voltammogram of
[{Co( -NO)( -C5H5)}2][PF6] revealed a one-electron reduction wave at 0.34 V
corresponding to the couple [{Co( -NO)( -C5H5)}2]+ – [{Co( -NO)( -C5H5)}2]. On adding
one equivalent of PPh3 a reaction occurred to give an orange-brown solution, a
voltammogram showed another wave at – 0.43 V, due to the irreversible one-electron
reduction to [Co(PPh3)(NO)( -C5H5)]+ [19].
The halogen-bridged tricobalt clusters, [Co3Cp3( 3-CPh)2( -Cl)]PF6
[Co3Cp3( 3-CPh)2( -Br)]SbF6 (3), and [Co3Cp3( 3-CPh)2( -I)]SbF6
.
.
MeCN (2),
CH2Cl2 (4), were
obtained from a reaction of a benzylidyne-capped tricobalt cluster, [Co3Cp3( 3-CPh)2] (1),
with halogens (X2 = Cl2, Br2 and I2) in CH2Cl2. The compounds were characterized by Xray diffraction, UV-Vis absorption spectra and cyclic voltammetry [20]. In a cyclic
voltammogram of 1 in 0.1 M BuNCl/MeCN an oxidation wave was observed at Epa = – 0.01
V, for the oxidation of 1 to 1+. On scan reversal an irreversible reduction wave appeared at
Epc = – 0.45 V. The complexes, 2+, 3+, and 4+ were studied by CV in CH2Cl2 with 0.1 M
Bu4NPF6. In the oxidation of 2+, a chemically reversible oxidation wave was observed at Epa
= 0.75 V versus Fc/Fc+. Very similar oxidation waves to those of 2+ were observed for 3+
and 4+. In the second scan, new redox waves were observed at Epa ~ 0.0 V, which indicated
that the oxidized species of 2 and 3 decompose slowly on a CV time-scale. In CH3CN, 4+
showed similar redox responses to those observed in CH2Cl2, but the oxidation waves of 2+
and 3+ were irreversible showing an Epa of 0.71 and 0.72 V, respectively [20]. Reduction
processes of the complexes in CH2Cl2 also resemble each other. An irreversible reduction
wave was observed in the potential region of – 0.57 to – 0.60 V. The irreversible reduction
of each complex exhibited a new chemically reversible redox couple at E1/2 = – 0.05 V,
which was the same as the oxidation potential of 1. These results suggested that the
reduction of the halogen-bridged complex reproduced the parent complex 1 [20].
The electrochemical properties of the redox mediator Co(III)/Co(II)(dbbip)2 (dbbip = 2,6bis(1’-butylbenzimidazol-2’-yl)pyridine) in a mixed acetonitrile/ethylene carbonate solvent
have been studied by a range of techniques in order to determine the rate constants for
electron transfer and the diffusion coefficients of the Co(II) and Co(III) species [21]. Cyclic
13
Chapter 1:
Introduction
voltammogram of Co(II)(dbbip)22+ in a mixture of 60 % ethylene carbonate/40 %
acetonitrile at a Pt disk electrode revealed a quasi-reversible electrode process at Ep = 0.39
V, assigned to the couple Co(III)/Co(II). A plot of the peak current density, jp,ox versus the
square root of the scan rate was used to estimate the diffusion coefficient of the Co(II)
complex. The plots gave a value of 1.9 × 10–6 cm2s–1 for the diffusion coefficient of the
bulky Co(II)(dbbip)22+ in the mixed solvent. The diffusion coefficient of Co(III) complex
ion was obtained using Fick’s first law and found to be equal to 1.1 × 10–6 cm2/s–1 [21].
The electrochemical behaviour of monomeric [Co2(CO)6(alkyne)] derivative is well known
[22]. At room temperature a one-electron diffusion controlled reduction process occurred at
Ep = – 1.0 V and was followed by a fast chemical complications in CH2Cl2 at a Pt electrode.
The chemical reversibility of the first process was enhanced by electron-withdrawing
substituents (e.g. CF3) or, to a lesser extent, by sterically demanding alkyne substituents
[22]. At ambient temperatures [{Co2(CO)6}2(PhC C–C CPh)] in CH2Cl2 the complex
undergone an apparent single two-electron reduction at Ep = – 0.94 V and another illdefined reduction wave was observed further at less negative potentials of ~ – 1.16 V. It was
found that a fast chemical decomposition following the reduction prevented proper
electrochemical analysis. It was also found that the two-electron peak at – 0.94 V gradually
split into two one-electron peaks as the temperature was lowered. At – 80 °C both waves
became reversible, as chemical complications were completely quenched, so that full
chemical reversibility was achieved, as shown by the directly associated re-oxidation peaks
[23].
1.3
THEORY OF ELECTROCHEMICAL FLOW CELLS
Detectors based on interactions between matter and an electrical current are another group
of major detection systems in FIA [24]. Electrochemical detection relies on the transfer of
electroactive species to the sensing surface. Because the sensor can only respond to species
in its vicinity, their concentration should be representative of the average concentration in
the bulk sample. This is strongly dependent on the characteristics of the hydrodynamic
system and flow–cell used. Thus, the flow should be fully uniform, pulse–free, and the
contact time of samples and standards with the sensor should be exactly the same. Kinetic
14
Chapter 1:
Introduction
discrimination of any side reactions taking place at the sensor surface has a very favourable
effect on selectivity [24].
The main assets of these detectors are their high sensitivity and selectivity which they
exhibit over wide concentration ranges [24]. Unlike optical detectors, they measure no
solution property, but rather respond to phenomenon occurring at the electrode surface; as a
result, they are better candidates for miniaturization. Electrochemical detectors are
compatible with a wide variety of cell shapes and volumes. Thus, some electrodes are
embedded in the cell walls, other planar electrodes are adhered to them, and still others are
of the open tubular type. Electrochemical flow cells are just as varied. They include the
early cascading models, commercially available electrodes and recent designs with built–in
electrodes. Also, electrodes can be placed in various positions (in the cavity, aligned with
the flow direction, in combination with others such as ion–selective sensors for multideterminations, etc.). In any case, the point measurement provided by electrodes should be
representative of the mean analyte concentration in the sample [24].
The flow through electrochemical cell designed by Burguera and co-workers [25] was made
from a polyethylene vial (45 cm long × 0.8 cm diameter filled with glass marbles) with an
effective volume of 0.5 ml. Two parallel glassy carbon rods with a total surface of 1.5 cm2
and of 2.0 cm2 were used as working and counter electrodes, respectively. Between them, a
1.3 cm long Pt wire was inserted as a pseudo-reference electrode. To further reduce the
inner volume of the cell to 0.5 ml, its body was filled with glass marbles of 2.0 mm of
diameter [25].
A multianalyte flow electrochemical cell for boianalysis was constructed by Maestre and
co-workers [26]. The upper and bottom parts of the cell were constructed from poly(methyl
methacrylate), providing high-level precision for the assembly of both surfaces. The
dimensions of the cell were 45 mm × 45 mm × 35 mm. A polyetheretherketone (PEEK)
plastic gasket (0.10 mm thickness) between both parts determined a cell dead volume of 70
µL. The six working electrodes and the reference electrodes were located in the upper part.
The inlet and the six outlets, corresponding to each working electrode, were placed in the
bottom part of the cell. The radial arrangement of the working electrodes, all of which were
equidistant from the inlet, guaranteed a laminar and identical hydrodynamic flow regime at
the six electrodes. Next to each working electrode, but at the bottom block, the six outlets
15
Chapter 1:
Introduction
were placed to prevent cross-talk from occurring and also to help contributing to identical
hydrodynamic pressure across the electrodes. The location of the reference electrode, as
well as the dimensions of the two platinum counter electrodes, prevented inadequate
conductance of the cell solution. The reference electrode was an Ag/AgClsat of the double
liquid junction type. The refillable outer electrolyte solution served as salt bridge to prevent
contamination of the reference element. This reference electrode was placed equidistantly
from each working electrode [26].
A wall-jet electrochemical detector was designed and characterized by Jaenicke and coworkers [27]. The cell was manufactured from Perspex with an internal volume of ~ 4 cm3.
The working electrode was a glassy-carbon disk 3.0 mm diameter and 2.0 mm thickness
(Tokai, Tokyo) which was press-fitted in a Teflon holder using an epoxy resin seal. The
electrode was prepared by first polishing with abrasive paper (# 1200) followed by a final
polish using 0.3 µm diamond paste until a mirror finish was obtained. A graphite disk
(Johnson Matthey) with the same dimension as that of the working electrode was used as
counter electrode. The reference electrode was Ag AgCl saturated KCl with a ‘Dycor’
polymer frit (Priceton Applied Research) as a liquid junction [27].
A modified Z-type flow-through cell for optical, electrochemical, and optoelectrochemical
flow injection analysis measurements was designed and constructed by Haghighi and coworkers [28]. The body of the flow-through cell was machined from a block of Plexiglas.
The flow cell had a 1 mm i.d. and was 20 or 10 mm long, so the volume of the cell was 16
or 8 µl, respectively. Two platinum sheets (0.1 mm thick) were placed on both ends of the
flow cell. A 1 mm i.d. hole was provided in each of these platinum sheets for transmission
of light. Both ends of the cell cavity were enclosed by two transparent glass windows. The
two platinum sheets were used as working and auxiliary electrodes with an
electrochemically active area of approximately 3 mm2. The reference electrode, a saturated
calomel electrode (SCE), was placed downstream in an overflow tube. The cell holder was
made of aluminium and was designed on the basis of the shape of the cell compartments of
the spectrophotometer. For spectrophotometric measurements in which organic solvent was
passed through the flow cell, the body of the flow cell was machined from stainless steel
[28].
16
Chapter 1:
Introduction
The home-made electrochemical flow-through cell was designed and modified for flow
injection analysis system by Masawat and co-workers [29]. The materials used were
polymethylmethacrylate or Perspex (A.C.S. Xenon Company Ltd, Thailand) as a working
block (19 mm thickness), stainless steel (Sahakol Machining, Chiang Mai, Thailand) as an
auxiliary block, home-made poly(ethyleneterephthalate) or PET (23 – 25 µm thickness) as a
spacer or gasket, 5H and 2B pencil lead (2 mm in diameter, Steadtler®) with only 5 mm
each in length of pencil lead was connected with silver wire (1 mm in diameter, 5 cm in
length) using conductive epoxy (Chemtronics®, USA) so that electrical contact can be made
to the pencil lead working electrode, and a 1 mm in diameter, 5 cm in length of silver wire
(99.9%, Prolabo, France) as a reference electrode. The Perspex body was drilled in the
centre and then was inserted with a piece of pencil lead which was drilled in the small
length at the other end. The silver wire was covered at one end with a small amount of
conductive epoxy and was inserted immediately into the drilled hole to connect with a piece
of pencil lead, then another silver wire was inserted in the drilled hole of Perspex body near
the one which was inserted first [29].
A simple and versatile cell suitable for spectroscopic and spectroelectrochemical studies in
the ultraviolet, visible, and infrared regions has been described [30]. Among the cell'
s
generally beneficial traits are its continuously adjustable path length, high degree of
chemical compatibility, ease of assembly and disassembly, and wide spectra range.
Regarding SEC applications, the cell permits adequate control of the OTE (optically
transparent working electrode) potential and exhibits a relatively short exhaustive
electrolysis time. The cell body was constructed from threaded glass connectors and a
Teflon stopcock assembly purchased from Ace Glass, Inc. The minimum liquid sample
volume of the cell is ~10 mL when the minimum path length configuration is employed. A
10 mm × 13 mm platinum screen (Aesar, 80-mesh) welded to a Pt lead at the end of a sealed
Pyrex tube was employed as the OTE in spectroelectrochemical studies. Platinum auxiliary
and quasi-reference electrodes were isolated in fritted Pyrex tubes (Ace Glass, C porosity)
[30].
Economou and co-workers have designed a thin layer electrochemical cell with a flow
channel thickness of 0.2 mm, for the detection of Co(II) by chemiluminescence [31]. The
flow cell consisted of a glassy carbon rod (3 mm in diameter) as a working electrode, a
home-made Ag/AgCl as a reference electrode positioned opposite a working electrode and a
17
Chapter 1:
Introduction
glassy carbon rod as a counter electrode was positioned downstream, near the outlet of the
cell [31].
On-line electrochemistry/mass spectrometry was used to study the complex mechanism of
electrochemical oxidation of N,N-dimethyl-p-phenylenediamine in aqueous electrolytes in
the pH range 1.4 – 9.7 using a radial flow electrochemical cell [32]. The electrochemical
flow cell consisted of two cylindrical Delrin blocks separated by a 50 m Teflon spacer. The
electrochemical flow cell consisted of a Pt disk electrode (1.6 mm diameter) as a working
electrode pressed into the lower block. The electrolyte flowed in a radial, inward direction.
The auxiliary and reference electrodes were both Ag AgCl 1 M KCl aqueous electrodes
separated from the flow of the electrolyte by 2
m PEEK frits. The volume of the
compartment of the working electrode, which was the volume of electrolyte located between
the working electrode and the wall of the central tube, was only 0.1 L [32].
An electrochemical cell was developed to enable flow analysis with voltammetric detection
using a hanging mercury drop electrode (HMDE) [33]. The flow cell was made from a piece
of cylindrical extruded acrylic (Perpex) with a length of 2.5 cm and a diameter of 1.5 cm.
The cylindrical piece was flattened at the bottom to allow the positioning of a Teflon
support with a mirror for visual inspection of the mercury drop. A flow-channel was drilled
through the Perpex with a diameter of 0.7 mm and a length of 1cm (volume 4 L). The
capillary of the working electrode was inserted in the flow cell from the top, perpendicular
to the reference and counter electrodes. The reference electrode was a silver wire (1 cm
length; 0.5 mm diameter) and the chloride in the seawater was used as a counter ion. The
counter electrode was a platinum wire (1 cm length; 0.46 mm diameter). The counter
electrode and reference electrode were inserted in 3 mm diameter nylon 6-6 screws and
screwed into the Perpex cell with holes leading to the flow-cell [33].
The voltammetric flow cell was designed for operation in a manner compatible with FIA
[34]. The flow cell was machined from a 7-mm thick Perspex plate. It had a volume of < 4
L and accommodated three electrodes: the working electrode was a mercury hemisphere
which protruded from the bottom of the flow channel; the reference electrode consisting of a
Ag AgCl wire bathed in saturated potassium chloride was placed in a Pasteur pipette with
an asbestos fibre junction; the auxiliary electrode was a stainless steel syringe needle which
also served as solution outlet. A suction pipette with an asbestos fibre junction; the auxiliary
18
Chapter 1:
Introduction
electrode was held in place by a screw and silicone rubber washer. The distance between the
inlet and the surface of the working electrode (d, ca. 3 mm) was a critical parameter and
was, therefore, kept constant during all experiments [34].
Janata and co-workers have designed a flow-through cell and characterized it using the
combination of FIA and cyclic voltammetry [35]. The flow cell consisted of a mercury
microelectrode as a working electrode, a Ag AgCl (in saturated KCl) as a reference
electrode, a steel rod tube as an auxiliary electrode and outlet, a sample inlet, a mercury
reservoir, and drop-size adjustment. The microelectrode was placed just below the orifice of
the inlet tube. This arrangement ensured that any part of the sample plug comes into contact
with the electrode surface only once. The replacement of the mercury drop was done by
simply gently knocking off the old drop and then forming a new drop. The used up mercury
was allowed to accumulate at the bottom of the cell compartment from where it was
removed periodically [35].
1.4
THEORY OF ELECTRON TRANSFER CHAIN
CATALYSIS (ELECTROCATALYSIS) REACTIONS
(ETC)
Electrocatalysis or electron transfer chain (ETC) catalysis is the catalysis of reactions by
electrons without net current flow (as opposed to redox catalysis, which means catalysis of
reduction or oxidation by redox mediators, thus involving a net current flow) [36]. The
theory of electrocatalysis was first applied to an organometallic system by Feldberg, who
also set up the method of finite differences for the computer simulation of kinetic analysis of
the electrochemical data [37].
ETC catalysis has been shown to be a very efficient way to perform organic, inorganic and
organometallic reactions such as ligand exchange, isomerization, chelation, decomplexation,
insertion, extrusion, and oxidative addition. The coupling of electrocatalysis with
organometallic catalysis was also shown to be efficient for alkyne polymerization. The
simplest organometallic reaction, ligand exchange, has been the most studied one [36].
Electrocatalysis has been efficiently practiced using either an electrode in a preparative
electrolysis cell or a redox reagent as an initiator (electrocatalyst). It is thus extremely useful
19
Chapter 1:
Introduction
to have at hand a library of redox reagents and knowledge of their redox potentials [38]. The
chain induction can be affected by an oxidant (anode, 17e complex such as ferricinium,
organic or inorganic oxidant) or by a reducing agent (cathode, 19e complex such as Cp2Co
or CpFe(C6R6) (R = H or Me), or organic or inorganic reducing agent). Ligand exchange
reactions have been electrocatalyzed for mono- and polynuclear complexes. Note that 19e
intermediates or transition states are involved in both types of electrocatalysis induced by an
oxidizing or by a reducing agent [36].
The principle of ETC catalysis involves an initiation step, which is induced by an electron
or a hole of an electron [38]. This electron can be provided via an electrode or via a catalytic
amount of a judiciously chosen redox couple. It is followed by the propagation step which
may include many chemical steps. The cross electron transfer step closes the catalytic cycle
regenerating the radical obtained at the first stage of the initiation step. Side reactions due to
the high reactivity of 17- and 19-electron radicals may also disrupt the system at any stage
[38].
Let us say for example, one wishes to check the electrocatalytic reaction A
B, say a
ligand exchange reaction [36]. The cyclic voltammogram of a reactant A, gives a wave for
A which may be reversible or not (at least a certain degree of reversibility should be
observable on lowering the temperature and at high scan rate). In the presence of added
ligand L1, the CV of A will show the appearance of a new wave due to the reaction product,
B (A + L1
B – L2). If the wave of A is a cathodic one (A
A–), setting the potential at
this wave or scanning through this wave generates A– in the vicinity of the cathode. This
explains the observation of a CV wave for B. Meanwhile the wave for A is profoundly
affected: since A– reacts rapidly with L1 to give B, its concentration in the neighbourhood of
the cathode is strongly diminished and can reach zero (depending on the relative rates of the
scan and of the propagation reaction of A– with L1). A is also consumed by the cross redox
reaction (A + B–
A– + B); hence, its intensity is also diminished. In this case the wave of
B is located at a more negative potential than that of A. If a ligand exchange reaction is
electrocatalyzed by a small anodic current, the CV wave normally observed in the absence
of the ligand is modified in the presence of this ligand and a new wave due to the reaction
product appears at a more positive potential. This behaviour is characteristic of
electrocatalysis and gives an idea of the rate of both propagation steps [36].
20
Chapter 1:
Introduction
Ligand substitution of metal carbonyls plays a key role in the catalytic sequences of a
variety of important processes leading to carbon monoxide fixation. The conventional
associative and dissociative mechanisms for such exchanges are usually considered to
involve even-numbered, 16- and 18-electron intermediates [39]. Ligand exchange in metal
carbonyls and their derivatives has received extensive mechanistic scrutiny. Electrocatalysis
of ligand exchange is best illustrated by examining the effect of an extremely small anodic
current upon solutions of metal carbonyls containing added nucleophiles [40].
The role of 19-electron intermediates or transition states in oxidatively induced
electrocatalytic ligand exchange reactions was first demonstrated by Kochi [41], in his study
of the manganese complex [MeCpMn(CO)2]. The exchange of the ligands MeCN, pyridine,
and THF by the less electron-releasing ligands phosphines, phosphates, and isonitriles is
possible because cross ET propagation step is exergonic [41].
A cyclic voltammogram of MeCpMn(CO)2(MeCN) revealed a reversible one-electron wave
at Ep = 0.22 V, in acetonitrile at a Pt microelectrode [41]. In the presence of PPh3, a CV
revealed another reversible one-electron wave at Ep = 0.55 V and the wave at Ep = 0.22 V
for the reactant [R = MeCpMn(CO)2(MeCN)] became irreversible. The anodic peak current
for MeCpMn(CO)2(MeCN) continued to decrease in magnitude in proportion to the
concentration of PPh3, and the diffusion current fell to near zero in the presence of very high
concentrations of PPh3. It was concluded that addition of PPh3 led to the substitution
product [P = MeCpMn(CO)2(PPh3)], which occurred at Ep = 0.55 V. In other words the
anodic process leading to the depletion of MeCpMn(CO)2(MeCN) away from the electrode
surface, results in the concomitant formation of MeCpMn(CO)2(PPh3). The electrochemical
process involved was [41]:
[ MeCpMn(CO ) 2 ( MeCN )]+ + PPh3 → [ MeCpMn(CO ) 2 ( PPh3 )]+ + MeCN
(1.17)
A cyclic voltammogram of a monopyridine complex (py)W(CO)5 in acetonitrile consisted
of a single irreversible wave at Ep = 1.01 V. However, after addition of 20 equivalent moles
of tert-butyl isocyanide to a solution containing 1.0 × 10–3 M (py)W(CO)5, a CV wave at Ep
= 1.01 V disappeared and a new irreversible wave appeared at Ep = 1.18 V corresponding to
a substitution product, (t-BuNC)W(CO)5. When smaller amounts of tert-butyl isocyanide
21
Chapter 1:
Introduction
were employed, both waves, at Ep = 1.01 V and 1.18 V were observed on a cyclic
voltammogram. The electrochemical processes involved were [42]:
( py )W (CO )5 → [( py )W (CO )5 ]+ + e −
(1.18)
The electrocatalytic substitution of the monopyridine and monoacetonitrile complexes of
tungsten carbonyl, (py)W(CO)5 (Ep = 1.01 V) and (MeCN)W(CO)5 (Ep = 1.02 V), by
triphenylphosphine were found to be more difficult to interpret solely on the basis of the CV
experiments [42], since their anodic waves were not cleanly separated from the anodic wave
of triphenylphosphine which occurred at Ep = 1.3 V. In the presence of added PPh3, the
anodic wave at Ep = 1.02 V for (MeCN)W(CO)5 was absent. However, the cyclic
voltammogram of (py)W(CO)5 appeared to be unaffected by the presence of added
triphenylphosphine. In neither case the CV wave corresponding to the product
(PPh3)W(CO)5 could not be clearly discerned, owing to the presence of the phosphine wave.
Nonetheless, preparative scale electrolysis demonstrated that both complexes undergo
ligand substitution with PPh3 at the electrode potentials [42].
The thermodynamic parameters governing the electron transfer chain catalyzed substitution
of triphenylphosphine or iodide on CpFe(CO)2I have been studied [43]. The reaction is
driven by the much higher stability of the triphenylphosphine complex relative to the iodide
complex, and proceeds to completion even though the electron transfer which propagates
the catalytic chain is endergonic. A cyclic voltammogram of CpFe(CO)2I in CH2Cl2 at a
platinum disk electrode, revealed a chemically irreversible wave at Ep = – 1.64 V. Another
irreversible wave at Ep = – 2.24 V was also observed on a CV of CpFe(CO)2I, which was
assigned to the formation of a dimer, [CpFe(CO)2]2. The chemical irreversibility of the
reductions of CpFe(CO)2-halide complexes was established to be the result of rapid
dissociation of the halide following electron transfer. And the formation of I– was confirmed
by the observation of two oxidation peaks on the reverse CV scan at ca. 0.0 and + 0.1 V,
corresponding to the two-step oxidation of iodide to triiodide and then iodide. In the
presence of PPh3, a cyclic voltammogram revealed an irreversible wave at Ep = – 1.53 V for
the formation of product CpFe(CO)2(PPh3), and a wave at Ep = – 2.24 V was also observed.
On scan reversal a new anodic peak, which was not present on the CV wave of CpFe(CO)2I
was observed at Ep = + 0.95 V and was found to match that observed for PPh3 alone [43].
22
Chapter 1:
Introduction
Direct electroreduction of alkyl and aryl halides is mostly performed at potentials which are
more negative than – 2 V. But the picture changes radically when complexes of certain
transition metals, capable of reacting with aryl- or alkyl-halides (RX) and forming more
readily reducible compounds, were introduced into the electrolyte solution. The reductive
dehalogenation of RX can in this case be conducted at considerably more positive
potentials, most often at those of the regeneration of the metal complex active with respect
to RX [44].
The complexes which are catalytically active with respect to RX can be tentatively
subdivided into three groups: phosphine, tetraazamacrocyclic and polypyridyl complexes
[44]. Electrochemical behaviour of solutions of NiX2L2 (X = Cl, Br, I; L = PPh3) in Nmethylpyrrolidinone depends on the nature and concentration of the halide ions (X–); the
concentration of L and the presence of ethylene. Reduction of the dissociated nickel (II)
species leads to soluble zerovalent complexes only when Ni(II) is complexed to both
phosphine and halide ions. Additions of Cl– to solutions of NiCl2 lead to formation of NiCl3–
or NiCl42–, which are no longer electroactive. In the absence of ethylene the electroreduction
of Ni(II) yields successfully Ni(I) and Ni(0) complexes of the type Ni(0)L3. In the presence
of ethylene the total reduction of NiX2L2 can be achieved in the absence of added L leading
directly to the Ni(0) species NiL2C2H4. The zerovalent complexes may exist in the anionic
forms Ni(0)L3X– and Ni(0)L2C2H4X– depending on the nature and concentration of X [45].
Cyclic voltammetry of nickel (II) perchlorate was studied in 0.1 M TBAP in acetonitrile. In
the presence of added PPh3, it yielded an octahedral complex [NiII(PPh3)2(MeCN)4]2+,
which was reduced directly in a two-electron process to [Ni0(PPh3)4]. Another one-electron
anodic peak, reversible in character, was observed on the reverse scan and was due to the
oxidation of the obtained nickel(0) to [NiI(PPh3)4]+ which was further oxidized at the second
irreversible anodic peak, thus restoring the nickel(II) initially present. In the presence of
added C3H5Br, three new processes appeared, one was associated with a nickel(II) cathodic
peak which shifted to less negative potentials, and another two anodic peaks were due to
catalytic oxidation of free bromide ions in the presence of triphenylphosphine. However, the
anodic peaks relative to the stepwise oxidation of nickel(0) nearly disappeared. Further
additions of C3H5Br only caused a progressive increase in the new cathodic peak which also
took on a sigmoid shape. The overall reduction process was summarized as follows [46]:
23
Chapter 1:
Introduction
[ Ni II ( PPh3 ) 2 ( MeCN ) 4 ]2 + + 2e − + 2 PPh3 − 4 MeCN → [ Ni 0 ( PPh3 ) 4 ]
(1.19)
[ Ni 0 ( PPh3 ) 4 ] + C3 H 5 X − 3PPh3 → C3 H 5 Ni I ( PPh3 ) X + PPh3 → C3 H 5 Ni I ( PPh3 ) 2 X
(1.20)
Cyclic voltammetry of [Rh(PPh3)3]+ in the absence of excess PPh3, at low scan rate (0.02
Vs–1) showed two reversible one-electron reduction processes at Ep = – 1.40 V and – 1.73 V
and an irreversible anodic peak was also present on scan reversal at Ep = – 0.85 V.
Increasing the scan rate to 0.2 Vs–1, lead to a decrease in peak current of the couple at – 1.73
V, and another reversible couple appeared at more negative potentials of – 1.90 V, and also
an anodic peak at – 0.85 V decreased in magnitude [47]. At very high scan rate of 5.0 Vs–1,
two-well defined one-electron reversible processes appeared at – 1.40 V and – 1.90 V. Thus,
it appeared that [Rh(PPh3)3]+ was reduced in two reversible steps to [Rh(PPh3)3] and
[Rh(PPh3)3]–, and a species, responsible for the anodic process at – 0.85 V was tentatively
formulated as [Rh(PPh3)2]x. In the presence of added free PPh3, a rather complicated cyclic
voltammogram of [Rh(PPh3)3]+ became quite simple. [Rh(PPh3)3]+ was reduced in two oneelectron reversible steps at – 1.20 V and – 1.73 V. The second reduction potential was fixed
at – 1.73 V, whilst the first one was affected by an increase in concentration of PPh3, which
made it shift positively by 60 mV per decade of concentration. This implied that the
reduction process accompanied by coordination of a fourth phosphorus ligand. Since the
UV-Vis spectrum of [Rh(PPh3)3]+ was not modified by addition of PPh3 up to 0.1 M
concentration, coordination of the fourth ligand must take place after reduction of
[Rh(PPh3)3]+. The overall reduction process proposed was as follows [47]:
[ Rh( PPh3 )]+ + e − ↔ Rh( PPh3 )3 + PPh3 ↔ Rh( PPh3 ) 4
(1.21)
The electrochemistry of (OMC)Rh(PPh3) and (OMC)Co(PPh3) was investigated in the
presence of excess PPh3 to determine the fate of the bound PPh3 axial ligand after
electrooxidation or electroreduction of each complex [16]. Cyclic voltammetry of
(OMC)Co(PPh3) revealed three reversible one-electron oxidation peaks in benzonitrile
(PhCN) at Ep = 0.19 V, 0.76 V, and 1.54 V and two reduction waves at Ep = – 0.86 V and –
1.92 V. Both the first oxidation peak at 0.19 V and the two reduction peaks were unaffected
by PPh3 addition to solution, even after addition of 100 equivalent of PPh3. The second
oxidation peak at 0.76 V became irreversible after addition of 1.0 equivalent of PPh3 to
solution, and no further changes were observed up to addition of 500 equivalent of PPh3, the
24
Chapter 1:
Introduction
highest concentration investigated. The third oxidation at 1.54 V could not be investigated
in the PhCN/PPh3 mixtures due to an oxidation of PPh3, which occurred at a more positive
potential [16].
Cyclic voltammetry of (OMC)Rh(PPh3) revealed three oxidation peaks in benzonitrile
(PhCN) at Ep = 0.21 V, 0.66 V and 1.42 V and in addition two reduction peaks at Ep = –
1.27 V and – 1.34 V. Several changes were observed in the cyclic voltammograms of
(OMC)Rh(PPh3) in PhCN as PPh3 was added to solution. The first oxidation process at 0.21
V, shifted negatively in potential while the second oxidation process at 0.66 V, became
irreversible. The third oxidation at 1.42 V could not be investigated in the PhCN/PPh3
mixtures due to an oxidation of PPh3, which occurred at a more positive potential. Addition
of PPh3 also resulted in an increased anodic current for a reduction peak at – 1.34 V,
resulting in the formation of rhodium complex [(OMC)Rh(PPh3)2]+. The proposed
electrochemical reaction was as follows [16]:
[(OMC ) Rh III ( PPh3 )]+ + PPh3 ↔ [(OMC ) Rh III ( PPh3 ) 2 ]+
(1.22)
The anodic peak current for the process at – 1.34 V increased with an increase in PPh3
concentration. The ratio of anodic to cathodic peak current for process at – 1.34 V also
depends upon the scan rate, which was 0.23 at 0.02 Vs–1 and 0.90 at 20 Vs–1 in the presence
of 500 equivalent of PPh3 [16].
1.5
RESEARCH AIMS AND OBJECTIVES
The overall objectives of this dissertation were as follows:
•
Testing of the three flow-through cells designed and developed by Cukrowski [48]
that should allow us to gain on-line, in a closed system, fundamental information
about the cobalt organometallic compounds that are used as catalysts in industrial
streams in non-aqueous solutions.
•
Development of a method for correction of an uncompensated resistance using
ferrocene as a model compound in batch solutions.
•
Determine the experimental parameters to be used in each design of a flow cell (i.e.,
the influence of flow rate and scan rate) using ferrocene. This will enable us to
25
Chapter 1:
Introduction
establish which flow cell will be used for fundamental studies and which for
analytical purposes. Study the influence of on-line mixing on the CV curves and use
the established methodologies for quantitative analysis on-line.
•
Use of CoCl2(PPh3)2 as a cobalt standard aimed to provide the quantitative
information on the amount of cobalt organometallic compounds that are used as
catalysts in industrial streams, since it was fairly stable. Since it is known that most
cobalt catalysts are not stable under common room conditions under which the
monitoring system is envisaged to operate.
•
Study the electrochemical properties of CoCl2(PPh3)2 and identify species present in
solution during its oxidation, since the anodic reactions of CoCl2(PPh3)2 were never
studied using electrochemical techniques.
Synthesis of CoCl2(PPh3)2 and characterisation using elemental analysis and
Infrared (IR) spectroscopy.
Investigate the influence of a kind of working electrode material used on
recorded voltammograms.
Establish sensitivity of voltammetric measurements towards traces of
moisture present in a background solution.
Identify the oxidation reactions observed from the voltammograms.
Electrochemically monitored titration of CoCl(PPh3)3 with chloride to
investigate the binding ability of chloride.
Electrochemically monitored titration of CoCl2(PPh3)2 with PPh3 using
Cyclic Voltammetry (CV) to investigate the electrocatalytic properties of the
complex and establish if it is possible to monitor the free PPh3 in organic
solvents in the presence of organometallic compounds containing cobalt.
Investigate use of ferrocene as a possible internal standard to be used during
electrochemical measurements of CoCl2(PPh3)2.
Determine the number of electrons involved in each electrode process
observed from the voltammograms of CoCl2(PPh3)2.
•
Determine a diffusion coefficient of ferrocene in a mixture of acetonitrile and
pentanol (1:1).
26
Chapter 1:
1.6
Introduction
SUMMARY OF CHAPTERS
This dissertation contains six chapters, including the current chapter (Chapter 1). In Chapter
1 (Introduction) a detailed literature review and the aims and objectives of this research
project were presented.
In Chapter 2 the theory associated with the electrochemical techniques employed in the
dissertation were discussed in detail.
In Chapter 3 the experimental procedures employed for each electrochemical technique and
the types of instruments used including the instrumental parameters were presented. The
reagents and electrodes used were also listed.
Chapters 4 and 5 contain results and discussion regarding the preliminary studies using
ferrocene and the electrochemical properties of dichlorobis(triphenylphosphine)cobalt(II),
CoCl2(PPh3)2.
Chapter 6 contains the conclusions achieved from the results obtained.
Other results were also presented in an Appendix section A and B whilst references were
provided at the end of the dissertation.
27
Chapter 2:
CHAPTER 2
Theory of Experimental Techniques Employed
THEORY OF EXPERIMENTAL
TECHNIQUES EMPLOYED
In this chapter, a theory of the basic concepts of electrochemical techniques employed in
this work is reviewed. The interest in characterization of electrode mechanisms has
motivated the development of a multitude of electrochemical techniques. These techniques
enable the sequence of reactions to be determined and the rate constant(s) of the
homogeneous chemical reactions to be measured. Such information is deduced from the
effect of the coupled chemical reactions on the response signal to a particular excitation
signal that is impressed on the electrode. Spectroscopic techniques (UV-visible, IR, NMR
and ESR) have been effectively coupled with electrochemistry to enable monitoring of
homogeneous chemical reactions. Intermediates and products can sometimes be identified
from their spectra [2].
Electrochemistry has proven to be a valuable technique for generating reactive oxidation
states and studying the attendant solution chemistry of such electrogenerated species. The
ease of oxygen removal from electrochemical cells greatly facilitates the study of oxygensensitive species which are electrogenerated. Small quantities of valuable materials can be
studied with ease. Very rapid reactions can be monitored since some electrochemical
techniques are capable of measuring reactions up to the limit of diffusion-controlled rates.
Electrochemistry often has an additional advantage over traditional chemical approaches in
that the solution is not complicated by a reagent added to generate the redox state of interest
[2].
2.1
CYCLIC VOLTAMMETRY (CV)
Cyclic Voltammetry (CV) is perhaps the most effective and versatile electroanalytical
technique available for the mechanistic study of redox systems [2]. It enables the electrode
potential to be rapidly scanned in search of redox couples. Once located, a couple can then
be characterized from the potentials of peaks on the cyclic voltammogram and from changes
caused by variation of the scan rate. CV is often the first experiment performed in an
electrochemical study. The repetitive triangular potential excitation signal for CV causes the
28
Chapter 2:
Theory of Experimental Techniques Employed
potential of the working electrode to sweep back and forth between two designated values
(the switching potentials). Although the potential scan is frequently terminated at the end of
the first cycle, it can be continued for any number of cycles, hence the terminology cyclic
voltammetry. A scan in which the potential is becoming increasingly positive is termed a
positive scan and a scan in which the potential is becoming increasingly negative is a
negative (even though the potential may actually be positive). The scan profile used in a
particular experiment is generally determined by the location of the redox couple of interest
[2].
CV has the further attraction of providing information not only on the thermodynamics of
redox processes but also on the kinetics of heterogeneous electron-transfer reactions and
coupled chemical reactions. The characteristic shapes of the voltammetric waves and their
unequivocal position on the potential scale virtually fingerprint the individual
electrochemical properties of redox systems. For this reason the method has been labelled
“electrochemical spectroscopy” [49].
The cells for voltammetric experiments usually comprise a three-electrode arrangement,
with working and counter-electrodes sufficiently spaced, while the reference electrode is
brought close to the working electrode surface with a Haber-Luggin capillary to minimize
IR loss. In particular, in case of organic solvents and high scan rates, uncompensated
resistance with the resulting IR drop and double layer effects may affect the
voltammograms. The IR drop distorts the linear E/t curve, usually assumed in CV. Of
course, in turn, the current is affected. Thus, the IR compensation or correction is strongly
recommended [50].
The IR correction is then the potential drop between electrode and the capillary tip from the
salt bridge (note that there is no potential drop between the capillary tip and the reference
electrode itself, since only minute (< 10–12 A) currents flow in the reference electrode
circuit) [51]. It is obviously essential that the electrolyte conductivity should not vary with
capillary position, that the current flow should be uniform and that no substantial electrolyte
concentration profile should be set up in the diffusion layer, which can best be achieved
with highly conductive concentrated electrolyte. Experimental determination of the IR drop
is always possible. Earlier work involved the systematic variation of tip-electrode distance
and extrapolated the resultant measurement to zero separation, but nowadays the current29
Chapter 2:
Theory of Experimental Techniques Employed
interrupt method is most commonly employed. This exploits the fact that, on interrupting
the current, the IR contribution to the potential drops immediately to zero, whereas the
electrode potential falls only relatively slowly (in the order of ms) owing to the large
double-layer capacitance. The potential change can be measured with a storage oscilloscope
or by utilizing an appropriate electronic circuit within the potentiostat; IR compensation can
be built into the potential control by a variety of means, yielding automatically corrected
values of the electrode potential [51].
Another significant problem in cyclic voltammetry is the possibility of dissolution of the
counter electrode [51]. The potentiostat will force an equal and opposite current through the
counter electrode as the working electrode, and a small amount of metal at the counter
electrode may dissolve and re-deposit on the working electrode. For this reason, a counter
electrode of the same material as the working electrode is usually employed during
electrochemical investigations on gold, platinum etc. Cyclic voltammograms of a much
more complex form are found for more complex electrochemical processes such as the
oxidation of organic substances, especially in cases where the solution is unstirred, and the
CV shows a sensitive dependence on the type of electroactive substance in the electrolyte,
the electrolyte itself, and the electrode material. The existence of multiple peaks often
reflects the build-up and dissolution of chemisorbed inhibitor layers on the electrode
surface, for example, the inhibiting effect of a build-up of oxide layer on platinum on the
oxidation of small organic molecules [51].
Applicability of voltammetric techniques is in some cases limited due to several factors. In a
background electrolyte solution, we will observe oxidation or reduction of electrolyte
components, the electrode material itself, or impurities in the electrolyte at certain
potentials. These processes define the accessible potential window, and the observable
electrode reactions should yield voltammetric signals well inside this window to allow
analysis without interferences of background contributions [50].
If voltammetry is used for analytical purposes, an extremely large electrode area or volume
ratio is often chosen to increase sensitivity, e.g., in inverse voltammetry or when working
with thin-layer cells. Under these conditions, diffusion is no longer semi-infinite but finite,
i.e., the thickness of the diffusion layer is limited by the volume, and this changes the
characteristics of the diffusion gradient. In a voltammetric experiments the measurable
30
Chapter 2:
Theory of Experimental Techniques Employed
current at the working electrode has two components; one for the heterogeneous charge
transfer and one for the mass transport. However, there are two exceptions: the reversible
and the irreversible case [49].
Reversible System.
An example of an electrochemical reversible system is the simplest possible one-electron
oxidation or reduction of a chemical species in solution at the working electrode (WE). The
rate of heterogeneous charge transfer is so high that a dynamic equilibrium is established at
the phase boundary. The Butler-Volmer equation [eq. (2.1)] is reduced to the Nernst
equation [eq. (2.2)], i.e., the surface concentrations depend only on the actual electrode
potential and are no longer influenced by heterogeneous kinetic effects. The current, as a
measurable quantity for the charge flux at the electrode surface, is influenced solely by mass
transport, the slowest step (diffusion control). The characteristic shape of the cyclic
voltammogram is a result of the potential-dependent changes in the surface concentrations
of the redox system and the simultaneous diffusion processes [49].
j A (0, t ) =
i
nFA
= C A (0, t )k ο exp[− αnF RT ( E −E ο )] − C B (0, t )k ο exp[(1 − α ) nF RT ( E − E ο ) (2.1)
E = Eο −
RT
ln Q
nF
(2.2)
where, jA is a flux of species A (mol s–1cm–2); k° is a standard heterogeneous rate constant
(cm/s);
is a cathodic transfer coefficient, 1 –
is an anodic transfer coefficient, and Q is a
reaction quotient. The other variables were defined in the previous chapter and in the list of
symbols.
31
Chapter 2:
Theory of Experimental Techniques Employed
Epa
5.E-06
4.E-06
E1/2
Ipa
3.E-06
E
I (A )
2.E-06
1.E-06
0.E+00
Ipc
-1.E-06
-2.E-06
-3.E-06
-0.25
Epc
-0.15
-0.05
0.05
0.15
0.25
0.35
0.45
E (V)
Figure 2.1 Cyclic voltammetric curve of a reversible charge transfer obtained to show how the CV
parameters were evaluated. E = switching potential, E1/2 = half-wave potential, Epa and Epc = anodic
and cathodic peak potential, Ipa and Ipc = anodic and cathodic peak current.
The most important parameters are the two peak potentials Epa and Epc as well as the peak
currents Ipa and Ipc (Fig. 2.1). For reversible charge transfer ( = 0.5) without coupled
chemical reactions, Ipa/Ipc = 1 and Ep = 59 / n mV (at 25 °C) [49].
The peak current for a reversible electron transfer is given by the Randles-Sev ik equation
(at 25 °C):
i p = 2.686 × 10 5 n 3 / 2 CD 1 / 2 v1 / 2 A
(2.3)
where, n = number of moles (mol–1); C = concentration of the oxidized or reduced species in
a bulk solution (mol cm–3); D = diffusion coefficient (cm2 s–1); v = scan rate (V s–1) and A =
area of the WE electrode (cm2). The peak current is proportional to the square root of the
scan rate.
32
Chapter 2:
Theory of Experimental Techniques Employed
Irreversible System
Charge transfer at the electrode is extremely slow. Depending on the potential, only one of
the cathodic or anodic heterogeneous reactions has a measurable rate. Thus, the current is
largely controlled by the rate of the charge-transfer reaction (charge-transfer control). Since
the surface concentrations at the electrode are dependent on the heterogeneous reaction and
are far removed from thermodynamic equilibrium, one speaks of an irreversible process.
Under such conditions, the Nernst equation does not apply. Furthermore, this means that the
measured potential values cannot be compared with thermodynamic equilibrium potentials
[49].
Quasi-reversible System.
Both the charge transfer and the mass transport determine the current. The Nernst equation
is only approximately satisfied. The charge transfer is therefore termed quasi-reversible
[49].
Multielectron Transfer Processes.
Multielectron transfer usually takes place in separate steps. Depending on the separation
between the theoretical potentials of the redox reactions, several cases have to be
distinguished. If the separation of the potentials between redox transfers is large, the
resulting cyclic voltammogram consists of two typical, additively superimposed, oneelectron transfer waves [49].
Also, the second electron transfer might take place at
potentials negative to the first, giving rise to two overlapped one-electron peaks. Often there
is a chemical reaction upon reduction or oxidation, giving rise to an ECE process with an
apparent multielectron transfer [52].
Figure 2.2 presents a typical example of a two-step redox process of cobalt-tetra-{2-(2thienyl)ethoxy}phthalocyanine [CoTETPc]. The first wave corresponds to oxidation of
CoIITETPc to CoIIITETPc which was followed by complex oxidation, CoIIITETPc to
[CoIIITETPc]+. In this Figure a multielectron overall response arises. The product of the first
electron-transfer reaction (process I) undergoes a second electron-transfer step (process II)
at potentials more positive than that for the first step [4].
33
Chapter 2:
Theory of Experimental Techniques Employed
30 µA
II
I
0
100
200
300
400
500
600
700
800
900
1000
E / mV (vs. Ag|AgCl)
Figure 2.2 Cyclic voltammogram of two-electron transfer process obtained using 5 mM CoTETPc;
in DMF containing 0.1 M TBABF4, at a scan rate of 100 mV/s, obtained from Ref [53].
Electron Transfer with Coupled Chemical Reactions
One of the most intriguing aspects of electrochemistry involves the homogeneous chemical
reactions that often accompany heterogeneous electron-transfer processes occurring at the
electrode-solution interface. The addition or removal of an electron from a molecule
generates a new redox state, which can be chemically reactive. A variety of mechanisms,
some of which involve complicated sequences of electrode and chemical reactions, have
been characterized [2].
2.2
CHRONOAMPEROMETRY
In Chronoamperometry a current through a working electrode is recorded as a function of
time, while the constant potential is applied to this electrode. During the experiment the
electrode is stationary and usually the electrolyte is not agitated, but is at rest.
The
derivation of the response began at a potential step from a value at which no current flows,
to one at which the diffusion-limited current passes [50]. The response of the current to this
34
Chapter 2:
Theory of Experimental Techniques Employed
perturbation will be a sharp change from zero current followed by relaxation to a value close
to zero, the final steady state magnitude that is determined by the flow of species to the
electrode surface (Figure 2.3). Hence this varies according to electrode geometry and
solution convection [50].
For a uniformly accessible planar electrode, the diffusion process is known as semi-infinite
linear diffusion, since it can be assumed to occur only in one dimension perpendicular to the
electrode surface. The observed current depends directly on the observed concentration
gradient at the electrode surface [50].
2.E-05
I (A)
2.E-05
1.E-05
5.E-06
0.E+00
0
5
10
15
20
25
Time (s)
Figure 2.3 Potential step chronoamperometric response of a 7.7 × 10–4 mol/l ferrocene solution in
0.05 M TBAHFP in acetonitrile.
In fact, the current decays smoothly from an initial value at t = 0 and approaches zero with
increasing the time as described by the Cottrell equation for a planar electrode [2]:
it =
nFACD1 / 2
π 1 / 2t 1 / 2
where,
(2.4)
it = current of time (t), A
n = number of electrons, mol–1
35
Chapter 2:
Theory of Experimental Techniques Employed
F = Faraday’s constant, 96485 A.s
A = electrode area, cm2
C = concentration of electroactive species, mol/cm3
D = diffusion coefficient of electroactive species, cm2/s
Hence, the current is inversely proportional to the square root of time. The Cottrell equation
states that the product it1/2 should be a constant for a diffusion-controlled reaction at a planar
electrode. Deviation from this constancy can be caused by a number of situations, including
nonplanar diffusion, convection in the cell, slow charging of the electrode during the
potential step, and coupled chemical reactions. For each of these cases the variation of it1/2,
when plotted against t, is somewhat characteristic [2].
Chronoamperometry has proven useful for the measurement of diffusion coefficients of
electroactive species. An average value of it1/2 over a range of time is determined at an
electrode the area of which is accurately known and with a solution of known concentration.
The diffusion coefficient can then be calculated from it1/2 via the Cottrell equation. Although
the electrode area can be physically measured, a common practice is to measure it
electrochemically by performing the chronoamperometric experiment on a redox species
whose diffusion coefficient is known. The value of A is then calculated from it1/2. Such an
electrochemically measured surface area takes into account any unusual surface geometry
that may be difficult to measure geometrically [2].
If the heterogeneous electron transfer of the redox species with the electrode itself is slow,
the current after the potential step is necessarily less than in a system in which the electron
transfer is rapid. This aspect of electrochemistry has been used for the measurement of
heterogeneous rate constants. The behaviour of it1/2 as a function of time can be influenced
substantially by the presence of chemical reactions that are coupled to the electrode process.
Consequently, characteristic variations of it1/2 vs. t have been effectively utilized for the
quantitative study of such homogeneous chemical reactions [2].
36
Chapter 2:
2.3
Theory of Experimental Techniques Employed
FLOW INJECTION ANALYSIS (FIA)
Years ago scientists involved with making analytical measurements had very limited and
primitive equipment at their disposal. Light sources were flames and the sun. Most
wavelength dispersions or discrimination were performed using filters or prisms. The
detectors, in many cases, were the eyes or eventually photographic films. The readout
devise was a person capable of evaluating the observed signals. Data evaluation was again a
person. The sample was collected by a person and filtered, diluted, and in general handled
by a person using volumetric glassware, like pipettes and volumetric flasks. All
measurements were slow, tedious, and required a great deal of skill on the part of the
analysts in order to ensure precise and accurate results [54].
Any measurement in a chemical laboratory involving liquid materials comprises the
following operations: solution handling, analyte detection, data collection and computation
of results. Nowadays, there is no shortage of computers and sophisticated detectors to aid
chemists in performing data collection and computation of results, but solution handling
requires an arsenal of skills, which a practicing chemist has to master, since mixing,
decanting, pipetting, and other volumetric operations are still performed manually, even in
the most advanced laboratories, using tools that were designed more than 200 years ago
[55]. It might seem that robots would be suitable tools for automation of such manual tasks;
but, it is likely that their impact will remain limited to repetitive operations like weighing of
pulverized materials, mechanization of sample injection into chromatographic columns,
handling of radioactive materials, or sample preparation. Because manual handling using
robots requires extensive programming and active feedback control, the use of robots is
justified only if large series of repetitive operations is to be handled over prolonged periods
[55].
Truly, there seems to be no way of resolving the problem of automated solution handling
other than by manual operations, as long as we think in terms of batch operations, a concept
in which generations of chemists have been trained. Therefore, in freshmen courses, as well
as in advanced research laboratories, beakers, flasks, and volumetric glassware are still the
standard tools of the trade, coexisting with the electronics of advanced detectors and
computers [55].
37
Chapter 2:
Theory of Experimental Techniques Employed
In flow injection analysis (FIA) the sample is injected directly into a moving stream without
the addition of air (Fig. 2.4). The sample-reagent zone mixes and reacts as it moves
downstream towards the detector; the degree of dispersion is controlled by a variety of
factors, their impact being specific to the analytical system in use. It is the control of this
dispersion which is at the heart of the technique and coupled with short, highly reproducible
retention times and provides the potential for sampling rates up to 200 per hour [24]. Figure
2.5 below shows an amperometric response obtained from multiple injections of various
concentrations of ferrocene solution in the flow cell using FIA system, using a continuousflow method.
Sample
C
D
B
A
Waste
E
Figure 2.4 Schematic representation of a typical flow injection analysis system. A = background
electrolyte solution, B = Pump, C = mixing coil, D = detector and E = computer.
7
100 mg/l
100 mg/l
6
C u rre n t ( µ A )
5
50 mg/l
4
25 mg/l
3
20 mg/l
2
14.28 mg/l
1
0
-1
0
500
1000
1500
2000
2500
Time (s)
Figure 2.5 Typical detector output of an FIA system into which a ferrocene solution of various
concentrations is repeatedly injected.
38
Chapter 2:
Theory of Experimental Techniques Employed
According to the type of flow used, flow methods can be classified as: (i) segmented-flow,
(ii) continuous-flow, and (iii) stopped-flow methods [56]. The term ‘stopped-flow methods’
is applied to kinetic methods involving mixing of the sample and the reagents in the detector
cell in order to perform periodic measurements for monitoring reaction development. This
type of method is rarely considered to be of the automatic type. We should note that the
‘continuous-flow’ concept does not exclude occasionally stopping the flow (for example, to
allow the reaction to proceed without increasing sample dispersion in the carrier).
Continuous-flow methods are also kinetic methods: measurements are performed during the
course of the reaction without the need to wait for equilibrium to be reached. Therefore,
some continuous-flow methods frequently include halting of the flow. In segmented-flow
methods the flowing stream is segmented by air bubbles that are primarily intended to avoid
carry-over between successively processed samples [56].
The three basics of FIA, as defined by R ži ka and Hansen, are reproducible timing (which
leads to the reproducible physical conditions), sample injection and controlled dispersion.
Dispersion is described as the amount that the chemical signal is reduced by injecting a
sample plug into an FIA system [54]. This is represented mathematically by
D* = C C max
(2.5)
where D* is the dispersion coefficient at the peak maximum produced by the ratio between
C°, the concentration of a pure dye, and Cmax, the concentration of that same injected dye as
it passes through the detector [54].
The experimental conditions usually involved in FIA result in incomplete mixing of the
injected sample plug with the carrier stream, with two important consequences, (i) mixing is
time-dependent, and therefore occurs to different extents at different points along the flowline, (ii) the extent of mixing is highly reproducible from sample to sample. Thus, the
technique gives rise to the creation of a time-dependent concentration gradient of sample
within the carrier streams. The physical foundations of FIA are related to dispersion, which
is defined as the dilution undergone by a sample volume injected into the flowing stream.
The dispersion is characterized by the concentration profile adopted by a zone or plug
inserted at a given point in the system without stopping the flow [57].
39
Chapter 2:
Theory of Experimental Techniques Employed
There are two mechanisms contributing to the dispersion of the injected sample,
(i)
Convective transport, occurring under laminar flow conditions (Fig. 2.6). This
yields a parabolic velocity profile with sample molecules at the tube walls
having zero linear velocity and those at the centre of the tube having twice the
average velocity [57].
(ii)
Diffusional transport, due to the presence of concentration gradients in the
convective transport regime, gives rise to axial and radial diffusion (Fig. 2.6).
The former, due to horizontal concentration gradients at the leading and trailing
edges of the injected sample zone contributes insignificantly to the overall
dispersion, whereas the latter, resulting from concentration differences
perpendicular to the direction of the flow makes an important contribution to the
overall dispersion. If the flow is considered to be made up of a large number of
superimposed fluid cylinders travelling convectively at different speeds, radial
diffusion tends to balance concentrations in such a manner that the molecules
located at the tube walls tend to move to the centre, whereas those at the centre
travel outwards. This process is of transcendental importance in accounting for
the fact that every sample injected maintains its integrity. Indeed, this motion
slows down convective transport, thus hindering progressive dilution of the zone
in the carrier stream [57].
Figure 2.6 Types of transport mechanisms present in a closed system.
Flow operations are much easier to automate, since they replace the mechanical handling of
oddly shaped (and often fragile) containers by sequential movements of liquids in tubes.
40
Chapter 2:
Theory of Experimental Techniques Employed
Flow operations are much easier to miniaturize by using small bore tubing, and the micro
volumes are conveniently manipulated and metered by pumping devices, which (unlike
pipettes) are not affected by surface tension (or by shaking hands) [55]. Flow operations are
much easier to control in space and time, since using closed tubing avoids evaporation of
liquids, provides exactly repeatable path(s) through which measured solutions move, and
provides an environment for a highly reproducible mixing of components and formation of
reaction products. Flow operations are very versatile, since flows can be mixed, stopped,
restarted, reversed, split, recombined, and sampled, while contact times with selected
sections of reactive or sensing surfaces can be precisely controlled. Finally, flow operations
allow most detectors and sensors to be used in a more reproducible manner than when used
in batch operations and by hand-as is obvious to anyone who has used both conventional
and flow-through cuvettes [55].
While many of the advantages of flow operations have been exploited in chromatography,
why has the batch approach not yet been replaced by flow systems in all areas of laboratory
practice? The reason must be tradition, and the fact that most chemists are used to thinking
in terms to batch operations, where homogeneous mixing is thought to be the only
reproducible way to bring reactants together and where the homogeneously mixed solution
is regarded as the only suitable form in which a reproducible measurement can be taken
[55].
FIA technique can be applied to agriculture, water and soil analysis, environmental
laboratories, biochemistry, biotechnology, pharmaceuticals, clinical laboratories and in food
and feed, and in process and quality control [54]. Predominantly FIA has been applied to
colorimetric methods using well established chemistries also found on both air segmented
and discrete analysis systems. Such analysis can be simple involving no more than the
addition of the sample to the moving reagent/carrier stream or involve dialysis; solvent
extraction; multiple reagent addition as intermediate steps. Improved sensitivity can be
achieved from a stopped flow technique, on line pretreatment etc, but at the expense of
sampling rate. FIA can also be applied to other detection systems e.g. ion selective
electrodes or as a means of sample introduction to Atomic Absorption Spectrophotometry
[24].
41
Chapter 2:
Theory of Experimental Techniques Employed
Modern electrochemistry offers a wide range of methods that can be used for continuous or
discrete measurements in flowing liquids. The combination of flow injection analysis and
cyclic voltammetry is attractive because of the flexibility of the former and the diagnostic
power of the latter. Recently developed flow injection techniques with controlled dispersion
offer a unique possibility to prepare solutions under controlled conditions in flowing
systems [58]. Two classes of electrochemical measurement are employed in flow detection:
one class is based on charge transfer between a liquid or gaseous phase containing the
analytes and a solid or immiscible liquid phase that is electrically conductive or semi
conductive, and includes the most common potentiometric, voltammetric and coulometric
detection techniques, and the other class involves the measurement of the electrical
properties of liquids, i.e. the electrical conductivity and relative permissivity [58].
There are many designs of flow-through detectors described in the literature, mainly for
continuous, on-line monitoring and for chromatographic applications [58]. The best design
for both solid-state and polarographic detectors involves a geometry in which the fluid
stream points at a right angle to the electrode surface or a disk upon which the fluid system
stream impinges perpendicularly. For the polarographic detectors, the best geometry is that
the fluid and mercury streams are at right angle to each other. The two kinds of detectors
then perform equally well [59]. Figure 2.7, shows an example of a cyclic voltammetric
curve of ferrocene solution in a flowing solution, using a kind of detector cell where the
fluid stream points at a right angle to the electrode surface.
Flowers and Callender, designed and characterized a transmittance cell (volume 10 ml) for
ultraviolet, visible, and infrared spectroscopy and spectroelectrochemistry [30]. Cyclic
voltammograms were measured at various scan rates for ~ 1.5 mM ferricyanide in 0.1 M
KNO3. A rather severe edge effect manifested the voltammograms due to exposure of the
thin layer to the bulk solution about its entire perimeter. Separation of the cathodic and
anodic peaks was observed to increase from ~ 50 to 100 mV as the scan rate was increased
from 1 to 4 mV/s. This was likely a result of ohmic drop across the optically transparent
working electrode due to the resistance of the thin layer solution. Despite these
shortcomings, the quality of the voltammograms indicated that the cell design permitted
reasonably accurate control of the optically transparent working electrode [30].
42
Chapter 2:
Theory of Experimental Techniques Employed
6.E-06
5.E-06
5
4
3
4.E-06
2
3.E-06
1
I (A)
2.E-06
1.E-06
0.E+00
-1.E-06
-2.E-06
1 - 50 mV/s
2 - 100 mV/s
-3.E-06
3 - 200 mV/s
4 - 300 mV/s
-4.E-06
-0.25
5 - 400 mV/s
-0.15
-0.05
0.05
0.15
0.25
0.35
0.45
E (V)
Figure 2.7 Cyclic Voltammograms of a 50 mg/l Ferrocene in 0.01 M TBAHFP in acetonitrile on a
Pt disk electrode at various scan rates, using a flow-through cell, at a flow rate of 1 ml/min.
Chen and Long designed a flow micro-cell (volume 5 L) and characterized it using cyclic
voltammetry of potassium hexacyanoferrate(III) at various scan rates [60]. A separation of
the cathodic and anodic peaks from 26 to 89 mV was observed, when the scan rate was
increased from 2 to 18 mV/s. They found that the edge effect of the flow micro-cell was less
than that previously reported by Flowers and Callender [30], the symmetry of the cyclic
voltammogram for this thin layer cell was improved and the IR drop was decreased. The
quality of the data demonstrated that the flow microcell can be successfully used for
electrochemical studies [60].
43
Chapter 3:
CHAPTER 3
Experimental
EXPERIMENTAL
This chapter describes in general the experimental procedures employed for each
electrochemical technique and the types of instruments used including the instrumental
parameters. The description of the three flow cells is included, the reagents and types of
electrodes used. The synthetic procedure of the two cobalt organometallic complexes is also
presented.
3.1
REAGENTS AND ELECTRODES
3.1.1 Reagents
•
Tetrabutyl ammonium hexaflourophosphate (TBAPF6), and ferrocene were purchased
from Fluka and used without further purification.
•
Acetonitrile C.P (99 %) was purchased from Sigma-Aldrich and purified by distillation
over P2O5 (98 %, from Riedel-deHa n).
•
Pentanol (99 %) was purchased from Fluka and purified by distillation over dried
potassium carbonate (99.85 %, from Ssangyong) (dried at 125 °C).
•
Acetone A.R (99.8 %), from Promark Chemicals was used as received.
•
Methanol A.R (99.5 %) and Hexane C.P (96 %) were purchased from Radchem
Laboratory Supplies and used as received.
•
Tetrahydrofuran A.R (99.8 %), from Lab-Scan Analytical Sciences was use as received.
•
Ethanol (99.5 %), from Merck was used as received.
•
Ethyl acetate, ACS reagent (99.5 %); petroleum ether; Hydranal-Coulomat AD (reagent
for coulometric Karl Fischer titration for cells without diaphragm), were purchased from
Riedel-deHa n and used as received.
•
Cobalt (II) chloride hexahydrate (CoCl2.6H2O) was purchased from Riedel-deHa n and
dried in an oven to a blue color before use.
•
Nitric acid C.P (HNO3, 55 %), from Bio-zone chemicals was used as received.
•
Sodium tetrahydroborate (NaBH4, 98 %), and AgNO3 were purchased from Saarchem
and used as received.
•
Triphenylphosphine was purchased from Sigma-Aldrich and used as received.
44
Chapter 3:
•
Experimental
Bis-(triphenylphosphine)dichlorocobalt(II)
[CoCl2(PPh3)2]
and
chlorotris-
(triphenylphosphine)cobalt(I) [CoCl(PPh3)3] were synthesized using the literature
method [61 – 62].
3.1.2 Electrodes
•
Platinum disk, glassy carbon, silver disk, and gold disk electrodes, each 2 mm in
diameter, were purchased from Metrohm, Switzerland.
•
Steel rod tube outlet (1/16 x 0.040 x 5 cm), from Upchurch Scientific, Anatech.
•
Platinum sheet electrode was from Metrohm, Switzerland.
•
Autolab dummy Cell, from Echo Chemie, Netherlands.
3.2
APPARATUS
3.2.1 Flow Injection Analysis
The flow injection apparatus consisted of two Dosimats (Metrohm), with a 10 and a 5 ml
exchange units, a PEEK mixing tee (25 x 23 x 20 mm in size), a six-port Hamilton injection
valve, and a flow cell. The tube length between the injectors and the mixing tee was 40 cm
Teflon tubing (2 mm ID, Metrohm). The tube length between the mixing tee and the
manifold and the manifold and the detector were both 15 cm Teflon tubing (2 mm ID,
Metrohm). The solutions were degassed by bubbling with argon before pumping through the
flow injection analysis system. The descriptions of the three flow cells designed by
Cukrowski [48] are as follows:
Flow Cell 1
The flow–by electrochemical cell body was made of a rectangular piece of PEEK (49 x 32 x
25 mm in size). The flow cell consisted of a platinum disk electrode (2 mm diameter) as a
working electrode, a gold disk electrode (2 mm diameter) as a counter electrode and a
Ag AgNO3 (0.01 M) in acetonitrile as a reference electrode, connected to a frit as a liquid
junction, and two bores (inlet and outlet) and a spacer. In this flow cell, the working
electrode was positioned in such a way that the fluid stream points at a right angle to the
electrode surface. The counter and the reference electrode were facing each other at opposite
45
Chapter 3:
Experimental
positions. The working electrode and the spacer were also facing each other, and the spacer
was used to vary the flow cell volume.
RE
Cell
RE
Spacer
Outlet
PEEK
Inlet
WE
Teflon
AE
AE
Figure 3.1 Diagram of the representative electrochemical flow–by cell 1, where WE =
working electrode, RE = reference electrode and AE = auxiliary electrode.
Flow Cell 2
The electrochemical flow cell was a wall-jet type. The cell body was made of a rectangular
piece of PEEK (42 x 32 x 25 mm in size). The flow cell consisted of the same electrodes as
those described in flow cell 1. The only difference between the two flow cells was the
position of the inlet tube with respect to the working electrode. In this flow cell the working
electrode was facing the inlet tube. This arrangement ensures that any part of the sample
plug comes into contact with the working electrode surface only once. And also an inlet
tube was incorporated inside a spacer.
46
Chapter 3:
Experimental
RE
Cell
RE
Spacer and Inlet
Outlet
PEEK
WE
Teflon
AE
AE
Figure 3.2 Diagram of the electrochemical wall–jet cell 2, with a gold disk as a counter
electrode.
Flow Cell 3
The electrochemical flow cell was a wall-jet type. The cell body was made of a rectangular
piece of PEEK (42 x 27 x 25 mm in size). The position and type of the counter electrode in
this flow cell differs from that of flow cell 2. The counter electrode was a steel rod tube and
it was positioned perpendicular to the reference electrode, and it served as a solution.
AE
RE
Cell
RE
Spacer and inlet
Outlet and AE
PEEK
WE
Teflon
Figure 3.3 Diagram of the electrochemical wall–jet cell 3, with the steel rod tube as a
counter electrode.
47
Chapter 3:
Experimental
3.2.2 Cyclic Voltammetry (CV) and Chronoamperometry
An Autolab Type II system (Eco Chemie, Utrech, Netherlands) connected to a model VA
Stand 663 (Metrohm, Herisau, Switzerland) was used for all experiments. The instrument
was controlled by PC via GEPS version 4.8 software. All electrochemical experiments were
carried out at room temperature using a three-electrode cell. The working electrodes were
either a Pt disk; gold disk or a glassy carbon disk (all from Metrohm, each with diameter of
2 mm). The counter electrode was a Pt sheet. The reference electrode consisted of a silver
wire immersed in a solution of silver nitrate (0.01 M) in acetonitrile. The reference electrode
was immersed in a Luggin capillary salt bridge containing 0.1 M TBAHFP dissolved in a
desired background solvent. The distance between the Pt disk working electrode and the tip
of the salt bridge was ~ 1 mm to minimize ohmic drop. The stability of the reference
electrode was checked periodically against a 0.77 mM ferrocene solution in the solvent of
interest by measuring the change in peak potential,
Ep, of Fe2+Cp2/Fe3+Cp2 after
measurement of the current-potential curve of the investigated compound. All potentials
have been referenced to the Fe2+Cp2/Fe3+Cp2 couple. The standard IR drop correction
facility of the instrument was used to minimize the effects of resistance in the electrolyte,
using the resistance value obtained with ferrocene.
3.2.3 Digital Fitting and Simulation
Theoretical cyclic voltammograms were fitted using the Fit and Simulation option located in
the Analysis menu of the Data Presentation window obtained from a GPES version 4.8
software. A reversible model was selected and a model name was set to Fit. The number of
exchanged electrons was set to 1, temperature to 298 K and the initial guesses of the
parameters were obtained. The Full Fit control parameter was selected followed by a Fast
Fit and the fitted curve was saved.
3.2.4 Karl Fischer Titration
A Hydranal-Coulomat AD reagent for coulometric titration for cells without diaphragm was
used. The reagent contains methanol, imidazole, sulfur dioxide and diethanolamine. The
instrument used for moisture determination was 831 KF Coulometer (Metrohm), and a 728
magnetic stirrer (Metrohm). This instrument consisted of a generator electrode, a platinum
wire electrode (with two wires), a 300 ml titration vessel, a stirrer bar and a keyboard.
48
Chapter 3:
Experimental
3.2.5 UV-Visible Spectroscopy
A Perkin Elmer UV-Vis (Lambda E2 201) spectrometer was used. The experimental
parameters were as follows: scan speed = 400 nm/min; path length = 10 mm; response =
medium and data range = 190 – 800 or 500 – 800 nm.
3.2.6 Infrared Spectrum
The IR spectra were obtained from KBr disks. The disks were run on the Perkin Elmer (RX
1 FT-IR system) spectrum using 32 scans at 2.0 cm–1 resolution and in the region 400 to
4000 cm–1.
3.2.7 Nuclear Magnetic Resonance (NMR) Spectroscopy
Room-temperature 31P NMR measurements of the free ligand, and complex were performed
on a Bruker ARX 300. H3PO4 ( = 0) was used as an internal standard for 31P spectrum.
3.2.8 Other Equipment
•
Hyprez Five-Star 1µm synthetic diamond paste for use on IMP cloth was purchased
from IMP Sampling Analysis.
•
Diamond extender blue, used as a lubricant with synthetic diamond paste was purchased
from IMP Sampling Analysis.
•
An IMP polishing cloth was purchased from IMP Sampling Analysis.
•
A DPA-2 polishing wheel was from Struers.
•
Whatman filter paper (44 Ashless Circles 110 mm diameter), from Sigma-Aldrich.
•
Magnetic stirrer hot plate was purchased from Metrohm.
•
A PTFE magnetic stirrer bar.
•
A PTFE magnetic retriever rod.
•
Drying Oven Series 2000, with a maximum temperature of 250 °C was purchased from
Apollo Scientific.
•
Precision balance (Mettler Toledo, 4 decimal places in grams), was purchased from
Microsep, Switzerland.
•
763 Dosimats and exchange units were purchased from Metrohm, Switzerland.
•
A six port distribution valve (model-R77810) and valve head (model-R36719) were
purchased from Hamilton, Switzerland.
49
Chapter 3:
Experimental
•
PEEK material was purchased from Metrohm, Switzerland.
•
Teflon tubing’s were purchased from Metrohm, Switzerland.
•
Millipore (Milli-Q Synergy), water supply system capable of producing ultra pure water
with a resistivity of 18.2 M
•
cm was purchased from Microsep.
Argon UHP gas (minimum purity 99.999 %), from Afrox Scientific.
3.3
SYNTHESIS
3.3.1 Synthesis of Bis-(triphenylphosphine)-dichlorocobalt(II) [CoCl2(PPh3)2]
CoCl2(PPh3)2 solid was prepared by dissolving 2.4007 g of dried CoCl2 in 60 ml of ethanol
and heated at 60–70 °C. 8.0003 g of triphenylphosphine was also dissolved in 60 ml of
ethanol and heated at 60–70 °C. The hot ethanol solutions of CoCl2 and triphenylphosphine
were mixed slowly. The complexes began to precipitate before the mixing was complete.
After complete mixing the hot solution was allowed to cool for about 2 minutes, filtered
using canular technique while still warm, and the crystalline products were washed several
times with ethanol and finally with ethyl acetate. The products were then dried in vacuum
[61 – 62]. The solid product was blue and it was stored in a desiccator over CoCl2. Yield =
4.4 g (66 %). Anal. Calcd for CoCl2(PPh3)2 (CoC12H10P2Cl2): C, 66.06; H, 4.62; Found: C,
65.80; H, 4.60. IR (KBr) v/cm–1: 3364m, 3047s [v(C–H)], 1658mw, 1480s [m v(C–C)],
1434vs [n v(C–C)], 1303m, 1181vs, 1164s [c (C–H) ip], 1096s [q X-sens], 1070s [d (C–
H) ip], 1028m [b (C–H) ip], 997s [p ring], 920 m [l (C–H) oop], 843m [g (C–H) oop],
744s [f (C–H) oop], 708vs [v (C–C) oop], 520s [y X-sens], 498s [y X-sens], 456m, 439m
[t X-sens], 420w [[t X-sens]. UV-Vis [acenonitrile/pentanol (1:1)]
max/nm:
297, 591, 678
nm.
3.3.2 Synthesis of chlorotris(triphenylphosphine)cobalt(I) [CoCl(PPh3)3]
Cobalt (II) chloride hexahydrate (2.4006 g) and triphenylphosphine (8.0004 g) were added
to ethanol (60 ml), and the mixture was vigorously stirred at 60–70 °C for 30 min to form
blue CoCl2(PPh3)2. The mixture was cooled to 30 °C, and sodium tetrahydridoborate was
added (total of 0.3206 g, added in 10 small portions) over ca. 10 min. The reaction mixture
turned dark-green, and finally minute brown crystals were formed. The crystals of
50
Chapter 3:
Experimental
chlorotris(triphenylphosphine)cobalt(I) were vacuum-filtered under argon, washed with
ethanol, water, ethanol, and petroleum ether, and dried under vacuum [62]. Yield = 5.1 g (57
%). Anal. Calcd for CoCl(PPh3)3 (CoC54H45P3Cl): C, 73.59; H, 5.15; Found: C, 68.17; H,
4.75. IR (KBr) v/cm–1: 3542m, 3418m, 3047s [v(C–H)], 1974mw, 1811mw, 1583m [k v(C–
C)], 1479s [m v(C–C)], 1434vs [n v[C–C], 1307m, 1282mw, 1189vs, 1154s [c (C–H)],
1119vs, 1090s [q X-sens], 1070s [d (C–H) ip], 1025m [b (C–H) ip], 997s [p ring], 918m
[l (C–H) oop], 852m [g (C–H) oop], 742s [f (C–H) oop], 722vs, 693vs [v (C–C) oop],
618w [s (C–C) ip], 541vs, 507m [y X-sens], 484s [y X-sens], 406m [w (C–C) oop].
3.4
SOLUTION PREPARATION
3.4.1 Background Electrolyte Solution
•
0.3874 g (1 mmol) TBAHFP salt was weighed and transferred into a 20 ml
volumetric flask. 10 ml of acetonitrile was added to the flask to dissolve the salt.
Pentanol was later added to the mark to make up a 20 ml background electrolyte
solution.
3.4.2 Salt Bridge Solution
•
A 0.3874 g (1 mmol) of TBAHFP salt was weighed and transferred into a 10 ml
volumetric flask. 5 ml of acetonitrile was added to the flask to dissolve the salt.
Pentanol was later added to the flask to the mark to make up a 10 ml solution.
3.4.3 Reference Electrode Solution
•
A 0.0425 g (0.25 mmol) of AgNO3 salt was weighed and transferred into a 25 ml
volumetric flask. Acetonitrile was added to dissolve the salt. Once dissolution was
complete, the flask was filled to the mark with acetonitrile, to make up a 0.01 M
solution. The reference electrode solution was prepared weekly. The reference
electrode was stored in a 0.01 M AgNO3 solution in acetonitrile (solution was
changed weekly).
51
Chapter 3:
Experimental
3.4.4 Analyte (i.e. Ferrocene, CoCl2, etc.) Solution
•
For example, a 0.05 × 10–3 mol/l analyte stock solution was prepared by weighing a
required mass of solid and transferring it into a 20 ml volumetric flask and diluting it
with the background solvent (i.e. mixture of acetonitrile and pentanol (1:1))
containing the supporting electrolyte.
3.5
PROCEDURE
3.5.1 Cleaning of Electrodes and Cells
•
A titration vessel and a Luggin-capillary salt bridge were rinsed first with deionised
water, followed by nitric acid (0.5 M), then rinsed with deionized water and finally
dried by rinsing with acetone (three times), wiped with a filter paper. During repeats
of experiments the cell was rinsed with a background solvent and dried with acetone.
•
The flow cells were rinsed first with deionized water, followed by rinsing quickly
with nitric acid (0.5 M) (to avoid swelling of the cells, since they were made of
PEEK material), then rinsed thoroughly with deionized water and finally dried by
rinsing with acetone (three times). During repeats of experiments, the cell was
cleaned by passing the background electrolyte solution into the cell, and scanning
the CV at the same potential range, until no redox features indicating the presence of
the studied electroactive species can be detected.
•
A Pt sheet auxiliary electrode was cleaned in the similar manner as the salt bridge.
•
The reference electrode was rinsed with acetone and wiped with a filter paper.
•
The Pt disk working electrode was cleaned by polishing with a 1 µm diamond paste
mixed with diamond extender blue, using a polishing cloth. The disk electrode was
mantled on a polishing wheel and rolled for several minutes over a wet polishing
cloth, until it was shiny. It was then rinsed with deionized water, followed by a 0.5
M HNO3 acid, deionized water, and finally rinsed with acetone (three times) and
wiped with a filter paper.
•
During repeats of experiments, the electrodes were removed from the solution,
carefully rinsed with a background solvent, dried with acetone and scanned over the
52
Chapter 3:
Experimental
same potential range in the same background electrolyte solution, until no redox
features indicating the presence of the studied electroactive species were detected.
3.5.2 Karl Fischer Titration
•
A 5 ml syringe was used to suck the distilled solvent from the flask.
•
The syringe was weighed.
•
The start button form the KF instrument was pressed, and 2 ml of the solvent in a
syringe was added into the KF cell, the syringe was weighed again and the obtained
mass (i.e. the mass of a 2 ml solution) was entered into the KF parameters using a
keyboard.
•
The enter button was pressed to resume the titration and the results were recorded on
the KF screen.
3.5.3 CV in Stationary Solution
•
An electrochemical cell was filled with 20 ml of a background electrolyte solution.
•
The cell was equipped with all the electrodes. All necessary electrical connections
between the electrodes and the potentiostat were made.
•
The solution was purged, by bubbling with argon for approximately 10 minutes. The
argon was turned off during measurement, but a blanket of argon was maintained
over a solution.
•
The CV parameters were set as follows.
(a) Ferrocene:
Estart = – 0.25 V
Esweep = + 0.45 V
Efinal = – 0.25 V
Estep and Scan Rate = 0.004 V (50 mV/s).
(b) CoCl2(PPh3)2:
Estart = 0 V
Esweep = + 1.8 V
Efinal = 0 V
Estep and Scan Rate = 0.004 V (50 mV/s).
53
Chapter 3:
•
Experimental
Before adding the analyte to the background solution, the background curves were
recorded. These curves were subtracted later from the curves in the presence of the
analyte.
•
The background electrolyte solution was stirred between measurements in order to
restore initial conditions, but it was not stirred during the experiment.
•
A desired amount of the analyte was injected into the cell and the CV for the redox
couple was recorded.
•
The solution was purged between measurements.
3.5.4 Chronoamperometry in Stationary Solution
•
Chronoamperometric curves were recorded immediately after recording a cyclic
voltammogram (i.e. from the same solution).
•
Before measurement of any chronoamperometric curves, each compound was
first evaluated using cyclic voltammetry; the potentials determined by CV were
used to set the potential limits for the chronoamperometry experiments.
•
Example of the chronoamperometric experiment:
o The background current of the electrode was recorded in 0.05 M
TBAHFP in acetonitrile (20 ml), applying the constant potential.
o The potential was stepped from –0.2 to 0.2 V.
o A desired amount of ferrocene was injected into the cell, and current was
recorded as a function of time at an appropriate potential.
o The background curve was subtracted from the ferrocene curve.
o The Cottrell plot of current versus the square root of time was
constructed, and the slope was determined using linear regression
analysis.
o The electrochemically active area of the electrode was determined from
the slope and the diffusion coefficient for ferrocene obtained from the
literature (D = 2.37 x 10–5 cm2/s, at 24 °C [63]).
o After the determination of the electrode area, the cell and electrodes were
cleaned as described in section 3.3, and the same procedure as above was
repeated to determine the diffusion coefficients of the other compounds.
o Chronoamperometric parameters in mixture of acetonitrile and pentanol
(1:1) were set as follows:
54
Chapter 3:
Experimental
(a) Ferrocene:
Standby Potential = – 0.2 V
Number of Potential Steps = 1
Applied Potential = 0.25 V
Duration Time = 25 s
Sample Time = 0.01 s
(b) CoCl2:
Standby Potential = 0.2 V
Number of Potential Steps = 1
Applied Potential = 1.0 V
Duration Time = 25 s
Sample Time = 0.01 s
(c) CoCl2(PPh3)2:
Standby Potential = 0.2 V
Number of Potential Steps = 1
Applied Potential = 1.1 V
Duration Time = 25 s
Sample Time = 0.01 s
3.5.5 Combination of FIA and CV.
•
200 ml of a background electrolyte solution was added into a 250 ml bottle and
connected to a dosimat (with a 10 ml exchange unit).
•
A 100 ml solution of an analyte was added into another 250ml bottle and connected
to another dosimat (with a 5 ml exchange unit).
•
The two solutions were degassed by bubbling with argon before pumping through
the flow injection analysis system.
•
Then the background solution was pumped through the cell at a flow rate of 1 ml
min–1 and the CV obtained.
•
Both the background solution and the sample solution were pumped at a flow rate of
4 ml min–1 each to afford a total flow rate of 8 ml min–1, to the mixing tee where
they were mixed.
•
After leaving the mixing tee, the mixed sample and background solution passed
through a length of tubing and went to waste.
55
Chapter 3:
•
Experimental
They were then pumped at a flow rate of 0.5 ml min–1 each to afford a total flow rate
of 1 ml min–1, to the detector.
•
The analyte was oxidized at the platinum disk working electrode in the flow cell.
After travelling through the flow cell, the stream exited the cell, flowed past a
reference and counter electrode, and then was delivered to waste.
•
The CV was then obtained on a flowing solution.
•
The flow was then stopped for a few seconds and the CV was obtained in a stopped
flowing solution.
3.6
ELECTROCHEMICAL DATA ANALYSIS
•
Peak current ratios for peaks of nearly reversible redox couples, where obtained
from:
ipa / ipc
•
(3.1)
For the evaluation of the separation of peak potentials the equation below was used,
Ep = Epa – Epc
•
(3.2)
Ep was also used for the measurement of the reversibility of the charge transfer. For
a fully reversible system, it was given by,
Ep = 2.3RT/nF = 59/n mV (25 °C)
•
(3.3)
The half-wave potential was determined from the following equation,
E1/2 = Epa - Ep/2 = (Epa + Epc)/2
•
(3.4)
Analysis of the chronoamperometry data was done by fitting the current-time data to
the Cottrell equation (eq. 2.4) using i versus t-1/2 graph.
•
The slope of this plot was obtained.
•
The active electrode area was calculated by substituting the value of it1/2 into the
Cottrell equation.
•
The same working electrode was immediately used to determine the diffusion
coefficient of ferrocene (in another solvent) and that of cobalt compounds studied.
•
The values of it1/2 and the active electrode area were then substituted into the Cottrell
equation and the diffusion coefficient was calculated.
•
The number of electrons involved was determined from the slope of a Cottrell plot
of i
1/2
/FACD1/2 vs. t–1/2.
56
Chapter 4:
CHAPTER 4
Preliminary Studies Involving Ferrocene
PRELIMINARY STUDIES INVOLVING
FERROCENE
This chapter will focus on the preliminary studies involving ferrocene as a model compound
using electrochemical techniques. Ferrocene/ferrocinium couple will be used as a reference
electrode potential during the studies involving cobalt organometallic compounds. In this
chapter the value of resistance present in solvent mixtures that will be used in the study of
cobalt organometallic compounds will be measured using ferrocene as an electroactive
species. This value will then be used to correct the IR voltage drop present in solutions
when studying the electrochemical properties of cobalt organometallic compounds. The
effect of electrolyte concentration on the peak potential separation value will also be
evaluated in this chapter. The effect that electrode positioning has on the cyclic
voltammogram will also be evaluated in a flowing solution using three different types of
electrochemical flow cells designed by Cukrowski [48] during the course of this project.
This was done by investigating the influence of solution flow-rate and scan rate on the
appearance of the voltammogram and also investigating the possibility of sample/reagent
mixing on-line.
Typical experimental conditions of CV were as follows, unless otherwise stated:
•
Platinum (Pt) disk was used as a working electrode (WE).
•
Acetonitrile was used as a background solvent.
•
Tetrabutyl ammonium hexafluorophospate (TBAPF6) was used as a supporting
electrolyte.
•
An analyte concentration of 7.7 × 10–4 mol/l was employed.
•
A scan rate was 50 mV/s.
•
Only the third (or last) CV scan was used for analysis.
•
All measurements were made at room temperature.
57
Chapter 4:
4.1
Preliminary Studies Involving Ferrocene
A MEASUREMENT OF UNCOMPENSATED RESISTANCE
PRESENT IN VOLTAMMOGRAMS OF FERROCENE
The oxidation of ferrocene is widely used in cyclic voltammetric studies in highly resistive
organic solvents as a means of reference electrode potential calibration, because of its
presumed ideal reversible behaviour [64]. It has become common procedure to add
ferrocene, typically in the range 0.5-1.0 mM concentration, to the electrolyte solution when
cyclic voltammetry is practised in non-aqueous solvents. The main purpose of this
preliminary work is to make use of the resistive distortion present in the ferrocene
voltammogram to determine a reliable value of the uncompensated resistance, which in turn
will be used to correct for IR voltage drop present in the voltammogram of the compound of
interest. This was done by introducing increasing amounts of resistance into the cyclic
voltammetric program until a good match was found between the theoretical and the
experimental voltammogram. This method can also be used to confirm that the ferrocene
process is indeed behaving in the ideal manner required for both reference potential and
resistance calibration procedures to be valid. The uncompensated resistance in a cell is that
resistance which, when multiplied by the total cell current, gives the difference in potential
between the solution at the surface of the working electrode (but outside the electrical
double layer) and the solution at the tip of the reference electrode probe [65].
In this work ferrocene is investigated over the appropriate range of scan rate and at various
concentrations. IR drop correction was applied until the properties of the cyclic
voltammograms adhere to the criteria for reversibility: difference between half-peak and
peak potentials (Epa – Ep/2) equalling 56.5 mV, difference between the anodic and cathodic
peak potentials (Epa – Epc =
Ep) of about 59 mV (with some dependence on switching
potential), peak potentials that are independent of scan rate (numerical criteria are for a oneelectron reaction at 298 K) [65], the peak current that is directly proportional to the square
root of scan rate (Ip/v1/2 = k) and the ratio between anodic and cathodic peak current (Ipa/Ipc)
must be equal to one. The solvent used in this experiment is a mixture of acetonitrile and
pentanol (1:1 volume ratio), with 0.05 M tetrabutyl ammonium hexaflourophosphate
(TBAPF6) as supporting electrolyte.
A typical cyclic voltammogram with resistance compensation for the oxidation of ferrocene
is given in Figure 4.1 below. The diamonds curve is the theoretical current-potential curve
58
Chapter 4:
Preliminary Studies Involving Ferrocene
for a reversible one-electron process whilst the bold line is the experimental currentpotential curve obtained using Ferrocene after IR voltage drop correction. The theoretical
CV curve was obtained using a digital fitting and simulation method (the procedure is given
in the experimental section chapter 3). The resistance compensation was adjusted in this
case by using the method described in the above paragraph. The agreement between
experiment and theory indicates that the instrumental compensation has introduced no
significant distortion in the data.
8.E-06
6.E-06
4.E-06
I (A )
2.E-06
0.E+00
-2.E-06
-4.E-06
-6.E-06
-0.25
-0.15
-0.05
0.05
0.15
0.25
0.35
0.45
E (V)
Figure 4.1 Comparison of a theoretical (diamonds) and experimental (solid line) corrected for IR
with, R = 2200 ) CV curves of ferrocene.
Under the same conditions with no compensation, the curve resulted in large Ep of about
86 mV at low scan rates of 100 mV/s (Figure 4.2). Data from 8 experiments for scan rates
between 0.05 and 0.5 V/s were analysed and several important parameters describing the
shape of the voltammograms were obtained. The data is summarised in Table 4.1 and 4.2.
59
Chapter 4:
Preliminary Studies Involving Ferrocene
8.E-06
6.E-06
I (A )
4.E-06
2.E-06
0.E+00
-2.E-06
-4.E-06
-0.25
-0.15
-0.05
0.05
0.15
0.25
0.35
0.45
E (V)
Figure 4.2 Comparison of an experimental uncorrected (solid line) and corrected for IR with R =
2200 (diamonds) CV curves of ferrocene.
Ep remains constant up to scan rates of 500 mV/s, a fact attesting to the accuracy of the
compensation (Table 4.2). Uncompensated voltammograms obtained under the same
conditions showed peak potential shifts of 8 mV at 500 mV/s (Table 4.1). Epa – Ep/2, Ipa/Ipc,
and Ep values are all in accord with theory. Anodic peak current over square root of scan
rate (Ipa/v1/2) remained constant regardless of scan rate, which is ideal for a reversible oneelectron transfer process.
60
Chapter 4:
Preliminary Studies Involving Ferrocene
Table 4.1 Analysis of data obtained from CV curves of ferrocene after background subtraction.
Scan rate,
Epa (V)
Epc (V)
v (V/s)
0.05
0.075
0.100
0.150
Ep
Ipa (A)
Ipc (A)
Ipa/Ipc
(mV )
0.182
0.186
0.186
0.186
0.100
0.100
0.100
0.100
82
86
86
86
Ipa/v1/2
Epa–Ep/2
1/2
(mV)
–5
57
–5
60
–5
60
–5
60
–5
[A/(V/s) ]
–6
3.60 × 10
–6
4.50 × 10
–6
5.22 × 10
–6
6.38 × 10
–6
–6
3.58 × 10
–6
4.48 × 10
–6
5.19 × 10
–6
6.36 × 10
–6
1.00
1.00
1.00
1.00
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
0.200
0.190
0.100
90
7.37 × 10
7.34 × 10
1.00
1.6 × 10
62
0.250
0.190
0.100
90
8.25 × 10–6
8.25 × 10–6
1.00
1.6 × 10–5
63
90
–5
–5
0.95
–5
63
–5
63
0.400
0.500
0.190
0.190
0.100
0.100
90
1.04 × 10
–5
1.16 × 10
1.09 × 10
–5
1.20 × 10
0.97
1.6 × 10
1.6 × 10
Table 4.2 Analysis of data obtained from CV curves of ferrocene after background subtraction and
after correction of IR voltage drop using an R value of 2200 .
v (V/s)
Epa (V)
Epc (V)
Ep (mV )
Ipa (A)
Ipc (A)
Ipa/Ipc
Ipa/v1/2
Epa–Ep/2
1/2
(mV)
–5
[A/(V/s) ]
–6
–6
0.05
0.173
0.113
60
3.60 × 10
3.58 × 10
1.00
1.6 × 10
54
0.075
0.173
0.113
60
4.50 × 10–6
4.48 × 10–6
1.00
1.6 × 10–5
54
60
–6
–6
1.00
–5
54
–5
54
–5
54
–5
54
–5
54
–5
54
0.100
0.150
0.200
0.250
0.400
0.500
4.2
0.173
0.173
0.173
0.173
0.173
0.173
0.113
0.113
0.113
0.113
0.113
0.113
60
60
60
60
60
5.22 × 10
–6
6.38 × 10
–6
7.37 × 10
–6
8.25 × 10
–5
1.04 × 10
–5
1.16 × 10
5.19 × 10
–6
6.36 × 10
–6
7.34 × 10
–6
8.25 × 10
–5
1.09 × 10
–5
1.20 × 10
1.00
1.00
1.00
0.95
0.97
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
CV OF FERROCENE IN A FLOWING SOLUTION
We needed to analyse the compounds of interest in a closed system by combining flow
injection analysis and cyclic voltammetry technique. Combination of flow injection analysis
and cyclic voltammetry is attractive because of the flexibility of the former and the
diagnostic power of the latter [35]. In order to do this we first determined the suitable flow
rate, scan rate, and a type of flow cell to use in order to obtain important parameters from a
CV curve. Ferrocene was used as a model compound for this study since it is a well-studied
compound that behaves reversibly. The flow rate and the scan rate are the two important
parameters. Their influence on the appearance of the voltammogram is examined in detail
using the three flow cells designed in our group by Cukrowski [48]. Full descriptions of the
61
Chapter 4:
Preliminary Studies Involving Ferrocene
cell designs are presented in the experimental section (Chapter 3). CV’s of ferrocene were
obtained first in a stationary solution before obtaining them in a flowing solution.
The effect of concentration of the supporting electrolyte was evaluated in order to make an
educated decision on which concentration to use in a flowing solution. This was done with
the aim of minimizing the cost of an electrolyte since high concentrations of background
electrolyte will demand huge amounts of supporting electrolytes (bearing in mind that large
amounts of electrolytes will be consumed daily during monitoring). Ferrocene was also
investigated in this solvent over a wide range of scan rates and at various concentrations and
IR drop correction was applied until the properties of the cyclic voltammograms adhere to
the criteria for reversibility.
4.2.1 Studies in a Stationary Solution.
First we looked at the effect of supporting electrolyte concentration to determine a suitable
concentration to be used for the preliminary studies of ferrocene in a flowing solution; the
electrolyte concentrations investigated were 0.01 M and 0.05 M. Table 4.3 and 4.4 present
tabulated results of the peak current separations before and after IR drop correction with R =
6200
over a wide range of scan rates and on samples containing 50 and 100 mg/l
concentration of ferrocene using 0.01 M TBAPF6 as supporting electrolyte concentration.
Data from 9 experiments for scan rates between 0.05 and 0.5 V/s were analysed and several
important parameters describing the shape of the voltammograms were obtained (Table 4.3
and 4.4). From Table 4.3 before IR voltage drop correction, Ep increases with increase in
scan rate and concentration of ferrocene at high scan rates (300 mV/s). At low scan rates of
up to 200 mV/s and at a concentration of 50 mg/l, Ep remains constant but is above 59 mV.
Epa – Ep/2 is 55 mV, Ipa/Ipc ratio is close to one and Ipa/v1/2 remained constant. Increasing the
scan rate to above 200 mV/s resulted in an increase in Ep, Epa – Ep/2 and decrease in Ipa/v1/2
values. Increasing the concentration to 100 mg/l resulted in a huge increase in Ep, Epa –
Ep/2 and decrease in Ipa/v1/2 values at all studied scan rates. The ratio between anodic and
cathodic peak current is far from unity.
62
Chapter 4:
Preliminary Studies Involving Ferrocene
Table 4.3 Analysis of data obtained from CV curves of ferrocene in acetonitrile containing 0.01 M
TBAPF6 after background subtraction.
Conc.
(mg/l)
50
v
(V/s)
0.05
0.075
0.100
0.125
(V)
0.135
0.135
0.135
0.135
Epc
(V)
0.053
0.053
0.053
0.053
Ep
Ipa (A)
Ipc (A)
Ipa/Ipc
82
82
82
Epa–Ep/2
1/2
(mV)
–6
55
–6
55
–6
55
–6
55
–6
[A/(V/s) ]
(mV)
82
Ipa/v1/2
–6
1.80 × 10
–6
2.20 × 10
–6
2.53 × 10
–6
2.84 × 10
–6
–6
1.75 × 10
–6
2.18 × 10
–6
2.50 × 10
–6
2.79 × 10
–6
1.03
1.01
1.01
1.02
8.0 × 10
8.0 × 10
8.0 × 10
8.0 × 10
0.150
0.135
0.053
82
3.10 × 10
3.04 × 10
1.02
8.0 × 10
55
0.200
0.135
0.053
82
3.60 × 10–6
3.55 × 10–6
1.01
8.0 × 10–6
55
86
–6
–6
1.00
–6
57
–6
62
–6
62
–5
63
–5
66
–5
68
–5
68
–5
70
–5
72
–5
74
–5
76
–5
80
0.300
0.400
0.500
100
Epa
0.050
0.075
0.100
0.125
0.150
0.200
0.300
0.400
0.500
0.139
0.143
0.143
0.147
0.156
0.160
0.160
0.164
0.169
0.173
0.177
0.182
0.053
0.045
0.045
0.049
0.045
0.045
0.045
0.041
0.041
0.036
0.032
0.032
98
98
98
111
115
115
123
128
137
145
150
4.20 × 10
–6
4.77 × 10
–6
5.26 × 10
–6
3.49 × 10
–6
4.10 × 10
–6
4.63 × 10
–6
5.00 × 10
–6
5.35 × 10
–6
5.95 × 10
–6
6.97 × 10
–6
7.73 × 10
–6
8.27 × 10
4.18 × 10
–6
4.73 × 10
–6
5.24 × 10
–6
3.08 × 10
–6
3.38 × 10
–6
3.84 × 10
–6
4.15 × 10
–6
4.44 × 10
–6
4.98 × 10
–6
5.92 × 10
–6
6.73 × 10
–6
7.18 × 10
1.00
1.00
1.13
1.21
1.21
1.21
1.21
1.20
1.18
1.15
1.15
7.7 × 10
7.5 × 10
7.4 × 10
1.6 × 10
1.5 × 10
1.5 × 10
1.4 × 10
1.4 × 10
1.3 × 10
1.3 × 10
1.2 × 10
1.2 × 10
Introduction of IR drop correction only improved the Ep value, which remained constant
with increase in scan rate with a value of 60 mV (Table 4.4). The Epa – Ep/2 deviated from a
theoretical value by about 5 mV at all studied scan rates at a concentration of 50 mg/l and
deviated by 4 mV at a concentration of 100 mg/l. The deviation in a (Epa – Ep/2) value from
56 mV predicted for the process controlled only by diffusion, may be due to
overcompensation resulting in a steep curve. The peak current ratio remained unchanged
after compensation, so was the value of Ipa/v1/2.
63
Chapter 4:
Preliminary Studies Involving Ferrocene
Table 4.4 Analysis of data in Table 4.3, obtained from CV curves of ferrocene after correction of IR
voltage drop using an R value of 6200 .
Conc.
(mg/l)
50
v
(V/s)
0.050
0.075
0.100
0.125
(V)
0.120
0.120
0.120
0.120
Epc
(V)
0.060
0.060
0.060
0.060
Ep
Ipa (A)
Ipc (A)
Ipa/Ipc
(mV)
60
60
60
60
Ipa/v1/2
Epa–Ep/2
1/2
(mV)
–6
51
–6
51
–6
51
–6
51
–6
[A/(V/s) ]
–6
1.80 × 10
–6
2.20 × 10
–6
2.53 × 10
–6
2.84 × 10
–6
–6
1.75 × 10
–6
2.18 × 10
–6
2.50 × 10
–6
2.79 × 10
–6
1.03
1.01
1.01
1.02
8.0 × 10
8.0 × 10
8.0 × 10
8.0 × 10
0.150
0.120
0.060
60
3.10 × 10
3.04 × 10
1.02
8.0 × 10
51
0.200
0.120
0.060
60
3.60 × 10–6
3.55 × 10–6
1.01
8.0 × 10–6
51
60
–6
–6
1.00
–6
51
–6
51
–6
51
–5
52
–5
0.300
0.400
0.500
100
Epa
0.050
0.120
0.120
0.120
0.122
0.060
0.060
0.060
0.062
60
60
60
4.20 × 10
–6
4.77 × 10
–6
5.26 × 10
–6
3.49 × 10
–6
4.18 × 10
–6
4.73 × 10
–6
5.24 × 10
–6
3.08 × 10
–6
1.00
1.00
1.13
7.7 × 10
7.5 × 10
7.4 × 10
1.6 × 10
0.075
0.122
0.062
60
4.10 × 10
3.38 ×10
1.21
1.5 × 10
52
0.100
0.122
0.062
60
4.63 × 10–6
3.84 × 10–6
1.21
1.5 × 10–5
52
60
–6
–6
1.21
–5
52
–5
52
–5
52
–5
52
–5
52
–5
52
0.125
0.150
0.200
0.300
0.400
0.500
0.122
0.122
0.122
0.122
0.122
0.122
0.062
0.062
0.062
0.062
0.062
0.062
60
60
60
60
60
5.00 × 10
–6
5.35 × 10
–6
5.95 × 10
–6
6.97 × 10
–6
7.73 × 10
–6
8.27 × 10
4.15 × 10
–6
4.44 × 10
–6
4.98 × 10
–6
5.92 × 10
–6
6.73 × 10
–6
7.18 × 10
1.21
1.20
1.18
1.15
1.15
1.4 × 10
1.4 × 10
1.3 × 10
1.3 × 10
1.2 × 10
1.2 × 10
Increasing the electrolyte solution concentration to 0.05 M, results in improved results.
Table 4.5 presents data before IR voltage drop correction for scan rates between 0.05 and
0.5 V/s; Ep remains constant but is above 59 mV. The Epa – Ep/2 value is 55 mV and Ipa/Ipc
is close to one; this is all in accord with theory at all studied scan rates and concentrations.
64
Chapter 4:
Preliminary Studies Involving Ferrocene
Table 4.5 Analysis of data obtained from CV curves of ferrocene in acetonitrile containing 0.05 M
TBAPF6 after background subtraction.
Conc.
(mg/l)
50
v
(V/s)
0.050
0.075
0.100
0.125
0.130
0.130
0.130
0.130
(V)
0.062
0.062
0.062
0.062
Ep
Ipa (A)
Ipc (A)
Ipa/Ipc
68
68
68
Epa–Ep/2
1/2
(mV)
–6
55
–6
55
–6
55
–6
55
–6
[A/(V/s) ]
(mV)
68
Ipa/v1/2
–6
1.80 × 10
–6
2.20 × 10
–6
2.55 × 10
–6
2.85 × 10
–6
–6
1.82 × 10
–6
2.20 × 10
–6
2.62 × 10
–6
2.83 × 10
–6
0.99
1.00
0.97
1.00
8.0 × 10
8.0 × 10
8.0 × 10
8.0 × 10
0.130
0.062
68
3.10 × 10
3.12 × 10
0.99
8.0 × 10
55
0.175
0.130
0.062
68
3.36 × 10–6
3.31 × 10–6
1.01
8.0 × 10–6
55
68
–6
–6
1.01
–6
55
–6
55
–6
55
–6
55
–5
55
–5
55
–5
55
–5
55
–5
55
–5
55
–5
55
–5
55
–5
55
–5
55
–5
55
–5
55
–5
55
–5
55
–5
55
–5
55
–5
55
–5
55
0.300
0.400
0.500
0.050
0.075
0.100
0.125
0.150
0.200
0.300
0.400
0.500
200
(V)
Epc
0.150
0.200
100
Epa
0.050
0.075
0.100
0.125
0.150
0.200
0.300
0.400
0.500
0.130
0.130
0.130
0.130
0.130
0.130
0.130
0.130
0.130
0.130
0.130
0.130
0.130
0.130
0.130
0.130
0.130
0.130
0.130
0.130
0.130
0.130
0.062
0.062
0.062
0.062
0.062
0.062
0.062
0.062
0.062
0.062
0.062
0.062
0.062
0.062
0.062
0.062
0.062
0.062
0.062
0.062
0.062
0.062
68
68
68
68
68
68
68
68
68
68
68
68
68
68
68
68
68
68
68
68
68
3.60 × 10
–6
4.41 × 10
–6
5.09 × 10
–6
5.66 × 10
–6
3.61 × 10
–6
4.41 × 10
–6
5.08 × 10
–6
5.65 × 10
–6
6.10 × 10
–6
7.00 × 10
–6
8.52 × 10
–6
9.96 × 10
–5
1.15 × 10
–6
6.88 × 10
–6
8.56 × 10
–6
9.93 × 10
–5
1.11 × 10
–5
1.21 × 10
–5
1.40 × 10
–5
1.69 × 10
–5
1.95 × 10
–5
2.19 × 10
3.55 × 10
–6
4.37 × 10
–6
5.05 × 10
–6
5.54 × 10
–6
3.58 × 10
–6
4.42 × 10
–6
5.05 × 10
–6
5.66 × 10
–6
6.05 × 10
–6
6.96 × 10
–6
8.50 × 10
–6
9.94 × 10
–5
1.14 × 10
–6
7.04 × 10
–6
8.66 × 10
–5
1.00 × 10
–5
1.12 × 10
–5
1.22 × 10
–5
1.40 × 10
–5
1.69 × 10
–5
1.94 × 10
–5
2.16 × 10
1.01
1.01
1.02
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
0.98
0.99
0.99
0.99
0.99
1.00
1.00
1.00
1.00
8.0 × 10
8.0 × 10
8.0 × 10
8.0 × 10
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
3.1 × 10
3.1 × 10
3.1 × 10
3.1 × 10
3.1 × 10
3.1 × 10
3.1 × 10
3.1 × 10
3.1 × 10
Correction of IR drop only improved the Ep value, which remained constant with increase
in scan rate and concentration and had a value of 60 mV (Table 4.6 and 4.7). The Epa – Ep/2
deviated from a theoretical value of 56 mV by about 3 mV at all studied concentrations.
65
Chapter 4:
Preliminary Studies Involving Ferrocene
Table 4.6 Analysis of data in Table 4.5 obtained from CV curves of ferrocene after correction of IR
voltage drop using an R value of 3000 .
Conc.
(mg/l)
50
v
(V/s)
0.050
0.075
0.100
0.125
(V)
0.126
0.126
0.126
0.126
Epc
(V)
0.066
0.066
0.066
0.066
Ep
Ipa (A)
Ipc (A)
Ipa/Ipc
60
60
60
Epa–Ep/2
1/2
[A/ (V/s) ]
(mV)
60
Ipa/v1/2
–6
1.80 × 10
–6
2.20 × 10
–6
2.55 × 10
–6
2.85 × 10
–6
–6
1.82 × 10
–6
2.20 × 10
–6
2.62 × 10
–6
2.83 × 10
–6
0.99
1.00
0.97
1.00
(mV)
–6
53
–6
53
–6
53
–6
53
–6
8.0 × 10
8.0 × 10
8.0 × 10
8.0 × 10
0.150
0.126
0.066
60
3.10 × 10
3.12 × 10
0.99
8.0 × 10
53
0.175
0.126
0.066
60
3.36 × 10–6
3.31 × 10–6
1.01
8.0 × 10–6
53
60
–6
–6
1.01
–6
53
–6
53
–6
53
–6
53
–5
53
–5
53
–5
53
–5
53
–5
53
–5
53
–5
53
–5
53
–5
53
0.200
0.300
0.400
0.500
100
Epa
0.050
0.075
0.100
0.125
0.150
0.200
0.300
0.400
0.500
0.126
0.126
0.126
0.126
0.126
0.126
0.126
0.126
0.126
0.126
0.126
0.126
0.126
0.066
0.066
0.066
0.066
0.066
0.066
0.066
0.066
0.066
0.066
0.066
0.066
0.066
60
60
60
60
60
60
60
60
60
60
60
60
3.60 × 10
–6
4.41 × 10
–6
5.09 × 10
–6
5.66 × 10
–6
3.61 × 10
–6
4.41 × 10
–6
5.08 × 10
–6
5.65 × 10
–6
6.10 × 10
–6
7.00 × 10
–6
8.52 × 10
–6
9.96 × 10
–5
1.15 × 10
3.55 × 10
–6
4.37 × 10
–6
5.05 × 10
–6
5.54 × 10
–6
3.58 × 10
–6
4.42 × 10
–6
5.05 × 10
–6
5.66 × 10
–6
6.05 × 10
–6
6.96 × 10
–6
8.50 × 10
–6
9.94 × 10
–5
1.14 × 10
1.01
1.01
1.02
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
8.0 × 10
8.0 × 10
8.0 × 10
8.0 × 10
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
1.6 × 10
Table 4.7 Analysis of data obtained from CV curves of ferrocene after background subtraction and
after correction of IR voltage drop using an R value of 1000 .
Conc.
(mg/l)
200
v
(V/s)
0.050
0.075
0.100
0.125
0.150
0.200
0.300
Epa
(V)
0.122
0.122
0.122
0.122
0.122
0.122
0.122
Epc
(V)
0.062
0.062
0.062
0.062
0.062
0.062
0.062
Ep
Ipa (A)
Ipc (A)
Ipa/Ipc
60
60
60
60
60
60
Epa–Ep/2
1/2
[A/ (V/s) ]
(mV)
60
Ipa/v1/2
–6
6.88 × 10
–6
8.56 × 10
–6
9.93 × 10
–5
1.11 × 10
–5
1.21 × 10
–5
1.40 × 10
–5
1.69 × 10
–5
–6
7.04 × 10
–6
8.66 × 10
–5
1.00 × 10
–5
1.12 × 10
–5
1.22 × 10
–5
1.40 × 10
–5
1.69 × 10
–5
0.98
0.99
0.99
0.99
0.99
1.00
1.00
(mV)
–5
53
–5
53
–5
53
–5
53
–5
53
–5
53
–5
53
–5
3.1 × 10
3.1 × 10
3.1 × 10
3.1 × 10
3.1 × 10
3.1 × 10
3.1 × 10
0.400
0.122
0.062
60
1.95 × 10
1.94 × 10
1.00
3.1 × 10
53
0.500
0.122
0.062
60
2.19 × 10–5
2.16 × 10–5
1.00
3.1 × 10–5
53
The amount of resistance required for compensation of curves at high concentrations of
ferrocene (200 mg/l) was smaller than the amount required for low concentrations. The
66
Chapter 4:
Preliminary Studies Involving Ferrocene
resistance amount required for compensation was only 1000 Ohm. This shows that analyte
made significant contribution to the overall conductivity of solution that most likely resulted
in migration current. Therefore, for any rigorous studies involving e.g. mechanisms of
electrochemical reaction, either the concentration of the supporting electrolyte would have
to be increased or concentration of the analyte (i.e. ferrocene) would have to be decreased.
4.2.2 Studies in a Flowing Solution.
Here we investigated the effect of scan rate and flow rate on the appearance of a cyclic
voltammogram using different flow cells. The main difference in these flow cells was
positioning of electrodes. The performances of the flow through cells were evaluated by
cyclic voltammetry using ferrocene as an analyte. First we analysed the CV curves obtained
using the so-called flow-by cell, whereby the background electrolyte solution is flowing
along the working electrode. The reference and counter electrodes were facing each other
and so were the outlet and inlet tubes.
3.E-06
2.E-06
Flow rate = 0.46 ml/min
Scan rate = 50 mV/s
Ipa/Ipc = 2.19E-06/2.22E-06 =0.99
Ep = (0.196 - 0.0171)V = 179 mV
2.E-06
I (A )
1.E-06
5.E-07
0.E+00
Background
-5.E-07
-1.E-06
-2.E-06
-0.25
-0.15
-0.05
0.05
0.15
0.25
0.35
0.45
E (V)
Figure 4.3 CV curve of a 50 mg/l ferrocene obtained using a flow-by electrochemical cell, flow rate
= 0.46 ml min–1.
67
Chapter 4:
Preliminary Studies Involving Ferrocene
The employed concentration of the supporting electrolyte (TBAPF6) was 0.01 M. A low
concentration of electrolyte solutions was used to minimize cost and because the amount of
resistance introduced could be corrected for. Figure 4.3 shows a cyclic voltammogram of
ferrocene at a scan rate of 50 mV/s and a flow rate of 0.46 ml min-1. Well-defined anodic
and cathodic peaks were observed from the CV curve. The anodic wave represents the
oxidation of Fe2+Cp2 to Fe3+Cp2, while the cathodic wave represents the reduction of
Fe3+Cp2 to Fe2+Cp2 (the reverse reaction).
Table 4.8 Data of peak potential separations and peak current ratio obtained from a 50 mg/l
ferrocene stock solution at different scan rates and flow rates using a flow-by cell 1.
Before IR drop correction
Flow rate
–1
v (V s )
Epa (V)
Epc (V)
–1
(ml min )
0.00
0.46
0.98
1.52
Ep
After IR drop correction
Ipa/Ipc
Epa (V)
Epc (V)
(mV)
Ep (mV)
[R = 34000 ]
0.050
0.183
0.00427
178
0.99
0.120
0.060
60
0.100
0.206
0.0159
190
0.99
0.120
0.060
60
0.200
0.233
0.0412
192
0.99
0.120
0.060
60
0.300
0.262
0.062
200
0.98
0.120
0.060
60
0.400
0.271
0.0677
203
0.98
0.120
0.060
60
0.050
0.196
0.0171
179
0.99
0.120
0.060
60
0.100
0.214
0.0275
186
0.96
0.120
0.060
60
0.200
0.247
0.0412
206
0.91
0.120
0.060
60
0.300
0.262
0.0476
214
0.86
0.120
0.060
60
0.400
0.271
0.0510
220
0.85
0.120
0.060
60
0.050
0.201
0.0256
175
0.99
0.120
0.060
60
0.100
0.222
0.0159
206
0.99
0.120
0.060
60
0.200
0.247
0.0275
219
0.98
0.120
0.060
60
0.300
0.286
0.0476
238
0.94
0.120
0.060
60
0.400
0.305
0.0577
247
0.89
0.120
0.060
60
0.050
0.214
0.0299
184
0.98
0.120
0.060
60
0.100
0.230
0.0159
214
0.97
0.120
0.060
60
0.200
0.247
0.0137
233
0.95
0.120
0.060
60
0.300
0.284
0.0473
237
0.93
0.120
0.060
60
0.400
0.304
0.0550
249
0.93
0.120
0.060
60
The CV curves obtained with increase in scan rate are presented in Appendix A (Figure A1
– A4). Increasing the flow rate and scan rate result in large potential separation between the
anodic and cathodic peaks which can be interpreted in terms of IR drop due to the working
68
Chapter 4:
Preliminary Studies Involving Ferrocene
electrode and the reference electrode being far apart (Table 4.8). It is interesting to see that
after IR drop correction (R = 34000
) the Ep value approached 60 mV for all studied scan
rates as predicted for one-electron reversible charge transfer.
Figures 4.4 to 4.6 present comparisons of the CV curves of ferrocene obtained in a flowing
solution, non-flowing solution in a flow cell and bulk solution in an open cell before and
after IR voltage drop correction. This was done to see how the peak potential separation
value vary in different solution types and how much resistance would be required in each
CV to bring the peaks closer together (to ~ 60 mV). The amount of resistance used to bring
the peak separation value to 60 mV was 34000
solution and 6200
in both the flowing and non-flowing
in a bulk solution. The difference in IR is due to the geometry of the
cells, namely the distance between the reference and the working electrode.
2.5E-06
2+
Fe Cp2
3+
-
Fe Cp2 + e
2.0E-06
1.5E-06
1.0E-06
I (A )
5.0E-07
0.0E+00
-5.0E-07
-1.0E-06
-1.5E-06
-2.0E-06
-0.25
3+
-
Fe Cp2 + e
-0.15
-0.05
2+
Fe Cp2
0.05
0.15
0.25
0.35
0.45
E (V)
Figure 4.4 Comparison of the CV curves of a 50 mg/l ferrocene in a flowing (— 0.46 ml min–1) and
non-flowing (—) solution using a flow-by electrochemical cell and in a bulk solution (—) using an
open cell.
69
Chapter 4:
Preliminary Studies Involving Ferrocene
2.5E-06
2+
-
3+
Fe Cp2
Fe Cp2 + e
2.0E-06
1.5E-06
I (A )
1.0E-06
5.0E-07
0.0E+00
-5.0E-07
-1.0E-06
-1.5E-06
-2.0E-06
-0.25
-
3+
2+
Fe Cp2 + e
-0.15
-0.05
Fe Cp2
0.05
0.15
0.25
0.35
0.45
E (V)
Figure 4.5 Comparison of the CV curves of ferrocene shown in Figure 4.4 after IR voltage drop
correction.
2.5E-06
2+
3+
Fe Cp2
-
Fe Cp2 + e
2.0E-06
1.5E-06
I (A)
1.0E-06
5.0E-07
0.0E+00
-5.0E-07
3+
Fe Cp2 + e
-1.0E-06
-0.25
-0.15
-0.05
0.05
-
2+
Fe Cp2
0.15
0.25
0.35
0.45
E (V)
Figure 4.6 Comparison of CV curves of a 50 mg/l ferrocene in a flowing solution before (—) and
after (—) IR voltage drop correction (R = 34000 ), flow rate = 0.46 ml min–1.
70
Chapter 4:
Preliminary Studies Involving Ferrocene
Next, we repeated the above experiments of ferrocene in a flowing solution using a wall-jet
type of a flow cell whereby the solution is flowing perpendicularly towards the working
electrode. Two types of wall-jet flow cells were used in this study, with the main difference
being a type of the auxiliary electrode used and its position in a flow cell as described in the
experimental section chapter 3. Figures 4.7 and 4.8 show CV’s of ferrocene in a flowing
solution using two wall-jet flow cells at a flow rate 0.46 ml min-1. CV exhibits a
characteristic polarogram-like shape at slow scan rate (50 mV/s) and the reverse trace is
almost superimposable on the forward trace, hence no peak-like responses are observed. In
this respect the reverse trace yield no useful information as oxidized products are swept
away from the electrode before they could eventually be reduced. This behaviour arises
because in wall-jet cells a thin layer of stagnant solution is present adjacent to the electrode
surface and flow is away from the electrode ensuring that a fresh solution is effectively
brought to the electrode surface. Under such hydrodynamic conditions, the electrochemical
process is controlled by both convection and diffusion.
4.0E-06
3.5E-06
3.0E-06
2.5E-06
I(A )
2.0E-06
1.5E-06
1.0E-06
5.0E-07
0.0E+00
-5.0E-07
-0.25
Background
-0.15
-0.05
0.05
0.15
0.25
0.35
0.45
E(V)
Figure 4.7 CV curves of a 50 mg/l ferrocene in a flowing solution using a wall-jet electrochemical
cell with a gold-disk auxiliary electrode, flow rate = 0.46 ml min–1.
71
Chapter 4:
Preliminary Studies Involving Ferrocene
4.0E-06
3.5E-06
3.0E-06
2.5E-06
I(A )
2.0E-06
1.5E-06
1.0E-06
5.0E-07
0.0E+00
-5.0E-07
-0.25
Background
-0.15
-0.05
0.05
0.15
0.25
0.35
0.45
E(V)
Figure 4.8 CV curves of a 50 mg/l ferrocene in a flowing solution using a wall-jet electrochemical
cell with a steel rod tube auxiliary electrode and outlet, flow rate = 0.46 ml min–1.
4.0E-06
3.5E-06
3.0E-06
2.5E-06
I(A)
2.0E-06
1.5E-06
1.0E-06
5.0E-07
0.0E+00
-5.0E-07
-1.0E-06
-0.25
-0.15
-0.05
0.05
0.15
0.25
0.35
0.45
E(V)
Figure 4.9 CV curves of a 50 mg/l ferrocene in a flowing solution using a wall-jet electrochemical
cell with a gold-disk auxiliary electrode, flow rate = 0.46 ml min–1 and v = 100 mV/s.
72
Chapter 4:
Preliminary Studies Involving Ferrocene
As scan rate is increased to 100 mV/s, the voltammograms deviates a little from the typical
polarogram shape, as a small peak component is present at the onset of the limiting current
plateau (Figure 4.9).
Figure 4.10 shows the effect of increased scan rate on the cyclic voltammogram of
ferrocene. As the scan rate is increased to above 100 mV.s-1, the voltammograms deviate
from a convection/diffusion-controlled polarogram shaped curve to nearly diffusion
controlled peak-shaped curves. The peak intensity increased with increasing scan rate, while
the anodic peak shifts towards higher potentials as expected due to IR drop. The oxidized
compound will still be present in the stagnant layer attached to the electrode surface during
the reverse scan, and the reversibly oxidized ferrocene clearly shows a reduction wave since
increasing the scan rate results in a decrease of the diffusion layer thickness. There seems to
be no significant difference between curves obtained in the two wall-jet flow cells, which
confirms that the type and position of the auxiliary electrode does not play a significant role.
The CV data obtained for the peak potential separations are summarised in Tables 4.9 and
4.10.
6.E-06
400 mV.s
5.E-06
-1
Fe2+Cp2
Fe3+Cp2 + e-
300 mV.s-1
200 mV.s
4.E-06
100 mV.s
3.E-06
-1
-1
-1
50 mV.s
I(A)
2.E-06
1.E-06
0.E+00
-1.E-06
-2.E-06
Fe3+Cp2 + e-
-3.E-06
-0.25
-0.15
-0.05
Fe2+Cp2
0.05
0.15
0.25
0.35
0.45
E(V)
Figure 4.10 CV curves of a 50 mg/l ferrocene after background subtraction at different scan rates
and a flow rate of 0.46 ml/min using a wall-jet electrochemical cell with a gold-disk auxiliary
electrode.
73
Chapter 4:
Preliminary Studies Involving Ferrocene
Table 4.9 Data of peak potential separations and peak current ratio obtained from a ferrocene
solution using a wall-jet flow cell with a gold disk auxiliary electrode.
Flow rate (ml min–1)
v (V s–1)
0.46
0.050
–
0.100
0.230
0.0476
182
0.97
0.200
0.252
0.0296
222
0.97
0.300
0.260
0.0237
236
0.97
0.400
0.270
0.0000
270
0.99
0.050
–
–
–
–
0.100
–
–
–
–
0.200
–
–
–
–
0.300
0.286
0.0238
262
1.03
0.400
0.305
0.0000
305
0.99
0.98
Epa (V)
Epc (V)
–
Ep (mV)
Ipa/Ipc
–
–
– = no peak was observed
Table 4.10 Data of peak potential separations and peak current ratio obtained from ferrocene
solution using a wall-jet flow cell with a steel rod tube auxiliary electrode.
Flow rate (ml min–1)
v (V s–1)
0.46
0.050
0.98
Epa (V)
–
Epc (V)
–
Ep (mV)
Ipa/Ipc
–
–
0.100
0.230
0.0159
214
1.0
0.200
0.237
0.0148
222
1.0
0.300
0.262
0.0238
238
0.99
0.400
0.271
0.0310
240
0.99
0.050
–
–
–
–
0.100
–
–
–
–
0.200
–
–
–
–
0.300
0.286
0.000
286
1.0
0.400
0.305
0.000
305
1.0
– = no peak was observed
We arrived at the following conclusions from the above data:
•
With all the flow cells it is possible to choose the experimental conditions that
favour convection/diffusion-controlled (polarogram-shaped) voltammograms or
diffusion-controlled (peak-shaped) voltammograms.
•
The main distinguishing factor between the three home-made designs of the flow
cells is that with the flow–by electrochemical cell 1, there is no convection, so
74
Chapter 4:
Preliminary Studies Involving Ferrocene
processes are controlled only by diffusion and results are very much similar to those
obtained in the bulk solution in an open cell. It is obvious that diffusion-controlled
electrochemical processes recorded here might change to mixed mode (convection
and diffusion) at slow scan rates and increased flow rates. Flow-by cell is by far
more versatile as it can be recommended for fundamental, mechanisms, speciation,
quantitative and qualitative studies performed on-line.
•
The wall-jet cells should be recommended for quantitative analysis as it generates
significantly higher analytical signal when compared with the flow-by cell.
4.2.3 Investigations of the Possibility of Sample/Reagent Mixing On-Line Using
Ferrocene in Acetonitrile Containing 0.01 M TBAPF6.
The above flow injection analysis experiments were performed in solutions that were
prepared in batch. We attempted to mix the electrolyte and sample solution on-line by
incorporating a mixing T-piece onto our FIA system. During on-line mixing the mixing Tpiece originated a reaction (mixing) zone that was subsequently carried through a six port
manifold into the detector. The efficiency of the mixing approach is of importance for
convenient reaction development, to improve the analytical signal and reproducibility.
Before analysis, we performed an experiment over a range of resultant flow rates to see its
effect on the anodic peak height. The resultant mixing flow rate is the total flow rate of the
background electrolyte and the sample solution.
For example, if one desires to measure a CV curve of a 50 mg/l ferrocene solution from a
100 mg/l ferrocene stock solution on-line one will pump both the background and sample
solution at a flow rate of 4 ml/min resulting in a mixed solution with a total flow of 8
ml/min to the cell.
Q
q
=
C
c
Where Q = is the total or resultant flow rate to the cell
q = is the sample solution flow rate to the mixing T-piece
C = is the concentration of the stock solution
c = is the concentration of the sample solution to the cell
Therefore: q = Q c
C
75
Chapter 4:
Preliminary Studies Involving Ferrocene
q = (8 ml / min) × (50 mg / l )
100 mg / l
q = 4 ml / min
∴ Background flow rate to the mixing T − piece = Q − q
= 8 ml / min − 4 ml / min
= 4 ml / min
1.19
1.18
1.17
1.16
I pa (µ A)
1.15
1.14
1.13
1.12
1.11
1.1
1.09
1.08
0
5
10
15
20
25
Resultant flow rate (ml/min)
Figure 4.11 Plot of resultant flow rate vs. anodic peak current of ferrocene, v = 100 mV/s.
Figure 4.11 shows a plot of resultant mixing flow rate versus anodic peak current obtained
in a ferrocene solution. The anodic peak current increased with an increase in mixing flow
rate from 2 ml min–1 to 7 ml min–1 and remained constant from 7 ml min–1 to 20 ml min–1
(the highest studied mixing flow rate). It appears that from the resultant flow rate of 7
ml/min the stagnant layer remains constant as it has reached the smallest thickness, hence
the signal does not increase further. From this data we decided to use 10 ml min–1 as the
mixing flow rate in all experiments performed with on-line mixing.
The influence of on-line mixing on the CV curves of ferrocene was demonstrated by
comparing the curve obtained with manual mixing to the one obtained with on-line mixing
76
Chapter 4:
Preliminary Studies Involving Ferrocene
(Figure 4.12). The peaks were enhanced when CV curves were recorded during on-line
mixing and the analytical signal was improved. This behaviour arises because the solutions
were prepared under controlled conditions with exclusion of moisture and any other
particulate matters that can lead to contamination of solution. Comparison with a
voltammogram obtained in a bulk solution (Figure 4.12, dotted curve) showed that the
situation in a bulk solution approaches that in a flowing solution, when the analysed
concentration of an analyte is prepared by hand using a syringe, in terms of signal intensity.
It was not clear at this stage why the curve measured using an on-line mixing method was
much higher or led to greater sensitivity as compared to those prepared in batch.
2.5E-06
2.0E-06
1.5E-06
I (A)
1.0E-06
5.0E-07
0.0E+00
-5.0E-07
-1.0E-06
-0.25
-0.15
-0.05
0.05
0.15
0.25
0.35
0.45
E (V)
Figure 4.12 Comparison of the CV curves of a 20 mg/l ferrocene obtained in a flowing solution
when the stock solution was prepared by hand (---) and using on-line (—) mixing technique, flow
rate = 0.98 ml min–1 and v = 100 mV/s. The CV curve of 20 mg/l ferrocene obtained in a bulk
solution (….) using an open cell at the same scan rate is also overlaid for comparison.
Quantitative analysis test of ferrocene at different concentrations was performed using online mixing approach. It must be noted that, one can prepare solutions of different
concentrations by varying the flow rates of both the background and the sample solution.
Figures 4.13 and 4.14 shows CV curves obtained in different concentrations of ferrocene in
a flowing solution with on-line mixing using a flow-by electrochemical cell. The peak
intensity increases and the anodic peak shifts to higher potentials with an increase in
77
Chapter 4:
Preliminary Studies Involving Ferrocene
concentration. As a result the peak potential separation increased with an increase in
concentration. This behaviour arises due to large IR voltage drop present in solution, which
becomes significant in flowing solutions. Linearity was observed from a calibration curve
constructed from CV curves obtained after background subtraction (Figure 4.15).
2.5E-06
2.0E-06
25 mg/l Ferrocene
1.5E-06
20 mg/l Ferrocene
15 mg/l Ferrocene
I(A )
1.0E-06
10 mg/l Ferrocene
5.0E-07
5 mg/l Ferrocene
0.0E+00
Background
-5.0E-07
-1.0E-06
-0.25
-0.15
-0.05
0.05
E(V)
0.15
0.25
0.35
0.45
Figure 4.13 CV curves obtained from various concentrations of ferrocene in a flowing solution,
flow rate = 0.98 ml min–1, mixing rate = 10 ml min–1 and v = 100 mV/s.
78
Chapter 4:
2.5E-06
2.0E-06
Preliminary Studies Involving Ferrocene
Flow rate = 1 ml/min
Mixing flow rate = 10 ml/min
Scan rate = 100 mV/s
25 mg/l Ferrocene
1.5E-06
20 mg/l Ferrocene
15 mg/l Ferrocene
1.0E-06
I(A)
10 mg/l Ferrocene
5.0E-07
5 mg/l Ferrocene
0.0E+00
-5.0E-07
-1.0E-06
-1.5E-06
-0.25
-0.15
-0.05
0.05
0.15
0.25
0.35
0.45
E(V)
Figure 4.14 CV curves obtained in Figure 4.13 after background subtraction.
3
2.5
y = 0.0983x
2
R = 0.9965
I pa (µ A )
2
1.5
1
0.5
0
0
5
10
15
Concentration (mg/l)
Figure 4.15 Calibration plot for data presented in Figure 4.14.
79
20
25
30
Chapter 4:
Preliminary Studies Involving Ferrocene
2.5E-06
25 mg/l Ferrocene
2.0E-06
20 mg/l Ferrocene
1.5E-06
15 mg/l Ferrocene
I (A)
1.0E-06
10 mg/l Ferrocene
5.0E-07
5 mg/l Ferrocene
0.0E+00
-5.0E-07
-1.0E-06
-0.25
-0.15
-0.05
0.05
0.15
0.25
0.35
0.45
E (V)
Figure 4.16 CV curves obtained in Figure 4.14 after IR voltage drop correction.
3.0E-06
2.5E-06
y = 1E-07x
R2 = 0.9997
Ipa (A)
2.0E-06
1.5E-06
1.0E-06
5.0E-07
0.0E+00
0
5
10
15
Concentration (mg/l)
Figure 4.17 Calibration plot for data presented in Figure 4.16.
80
20
25
30
Chapter 4:
Preliminary Studies Involving Ferrocene
The CV curves obtained at different concentrations of ferrocene in Figure 4.14 where
corrected for IR voltage drop with R = 34000
and the new calibration plot was
constructed. Figure 4.16 shows the CV curves obtained at different concentrations of
ferrocene after IR voltage drop correction. The peak potential separation remained constant
with increase in concentration. The calibration plot gave a linear relationship with
concentration (Figure 4.17). IR voltage drop correction improved a calibration plot of
ferrocene – almost a perfect straight line was obtained.
Most importantly, the anodic peak potential (Epa) remained independent on the analyte
concentration. This would allow applying constant potential, e.g. between 0.15 and 0.20 V
for quantitative monitoring of the analyte. Without IR correction, it would be required to
apply a large positive potential, which will vary with concentration of the analyte and might
result in interferences from other oxidizable materials in cases where more than one
compound is analysed or electroactive. It follows that IR drop plays a major role so one
should redesign the cells to bring the reference electrode much closer to the working
electrode to minimize IR drop and this can be achieved when using flow-by cell.
81
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement