16 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim DOI: 10.1002/cphc.200400248 ChemPhysChem 2005,6,16–28
Determining the Gibbs Energy of Ion Transfer
Across Water–Organic Liquid Interfaces with
Three-Phase Electrodes
Fritz Scholz* and Rubin Gulaboski
[a]
1. Introduction
The lipophilicity of a compound is its affinity toward apolar sol-
vents, whereas its hydrophilicity describes its affinity toward
the polar solvent water. The higher the lipophilicity of a com-
pound is, the better its solubility in organic solvents, such as
benzene and paraffin oils. Lipo- and hydrophilicity are two
sides of one coin; that is, they are complementary properties.
Because in all living cells and higher biological structures there
are hydrophilic and lipophilic compartments in close proximity,
and because the chemical compounds in these systems have
to cross these different parts, for example, lipophilic mem-
branes, the lipophilicity of a compound decides—to a great
extent—its distribution, transport, metabolism, toxicity, and
thus its biologi cal activity. Although many other and much
more specific factors are also involved, a compound’s lipophi-
licity is recognized as one of the most important properties for
understanding their role in living systems. The common
method of assessing the lipophilicity of a compound is to de-
termine its partition coefficient in a two-solvent system using
water as one solvent and an organic, rather apolar solvent as
the other. n-Octanol was once proposed as the standard or-
ganic solvent because, although being reasonably apolar due
to its C
8
-tail, it still has a polar head group.
[1]
Thus, the n-octa-
nol molecule has similarities with lecithins, the major constitu-
ents of biological membranes. The n-octanol/water partition
coefficient, the so-called K
OW
, is routinely determined for all po-
tential pharmaceuticals
[1, 2a]
and is also used to assess the envi-
ronmental impact of a compound.
[2b]
These data are accessible
from simple shake-flask partition experiments. To avoid these
labor-intensive determinations, high performance liquid chro-
matography (HPLC) is routinely used, and empirical correla-
tions between retention data for compounds on specific col-
umns with the n-octanol/water data allow the calculation of
K
OW
for new compounds.
[2a]
For ionic species, the assessment
of their lipophilicities is more complex because the partition of
the counterions must always be taken into account, and the
distribution data of single ionic species have to rely on an
extra-thermodynamic assumption. Very early, Grunwald et al.
proposed the assumption that the Gibbs energy of transfer of
tetraphenylarsonium cations and tetraphenylborate anions are
equal.
[3]
Later, other systems have been proposed, but the
Grunwald assumption is still the most frequently accepted. In
principle, the partition coefficient of an ion, that is, its Gibbs
energy of transfer, is accessible in the same way as that of a
neutral compound. However, since ions posses a charge, elec-
trochemical techniques lend themselves to this purpose. A
conscious description of the evolution of ideas leading to the
present understanding of the potential difference at the
liquid–liquid interface of two immiscible electrolyte solutions
has been given by Girault and Schiffrin.
[4]
Based on the experi-
ments of Beutner,
[5a]
Leonor Michaelis, a pioneer of enzyme ki-
netics, gave, in 1922, a lucid description of the interfacial po-
tential (Phasengrenzpotentiale) at a liquid–liquid interface, and
an ordering of ions according to their lipophilicities.
[5b]
Follow-
ing the first experiments of Nernst and Riesenfeld with liquid–
liquid interfaces published in 1902,
[6]
it took almost 70 years
before Guastella and others undertook four-electrode measure-
ments at such interfaces.
[7, 8]
In the last three decades of the
20th century, voltammetric as well as other electrochemical
techniques have been applied to study the interface between
[a] Prof. Dr. F. Scholz, R. Gulaboski
Universitt Greifswald, Institut fr Chemie und Biochemie
Soldmannstr. 23, 17489 Greifswald (Germany)
Fax: (+ 49) 3834-864451
E-mail: fscholz@uni-greifswald.de
Supporting information for this article is available on the WWW under
http://www.chemphyschem.org or from the author.
Ions can be transferred between immiscible liquid phases across
a common interface, with the help of a three-electrode potentio-
stat, when one phase is an organic droplet attached to a solid
electrode and containing a redox probe. This novel approach has
been used in studies to determine the Gibbs energy of anion and
cation transfer, ranging from simple inorganic and organic ions
to the ionic forms of drugs and small peptides. This method of
studying ion transfer has the following advantages: 1) no base
electrolytes are necessary in the organic phase; 2) the aqueous
phase contains only the salt to be studied ; 3) a three-electrode
potentiostat is used; 4) organic solvents such as n-octanol and
chiral liquids such as d- and l-2-octanol can be used; 5) the
range of accessible Gibbs energies of transfer is wider than in the
classic 4-electrode experiments ; 6) the volume of the organic
phase can be very small, for example, 1 mL or less; 7) the experi-
ments can be performed routinely and fast. Herein, the basic
principle is outlined, as well as a summary of the results obtained
to date, and a discussion on the theoretical treatments concern-
ing the kinetic regime of the three-phase electrodes with immobi-
lized droplets.
ChemPhysChem 2005,6,16–28 DOI: 10.1002/cphc.200400248 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 17
two immiscible electrolyte solutions (ITIES) and determine the
partition coefficient of ions. These achievements are associated
with the names of Koryta,
[9]
Girault and Shiffrin,
[4]
Samec,
[10]
Va-
nysek,
[11]
Marecek,
[12]
Kakiuchi,
[13]
and Hundhammer,
[14]
and they
are reviewed in various places.
[15, 16]
All these studies are based
on polarizing the ITIES with the help of a four-electrode poten-
tiostat using cells as depicted in Figure 1. This technique al-
lowed, for the first time, polarization measurements at an ITIES
to be performed, and led to most significant advances in the
understanding of liquid–liquid interfaces.
[16, 17]
However, four-
electrode potentiostatic measurements at ITIES are not as
simple to perform as three-electrode potentiostatic measure-
ments at metal–electrolyte solution interfaces. The handling of
the cell needs skill and experience. To give an example, keep-
ing the interface in a stable position with respect to the two
Luggin capillaries approaching the interface from both sides,
needs special attention. Despite these experimental difficulties,
which kept the technique confined to use by specialists, there
are other more serious obstacles involved: 1) The necessity to
have foreign electrolytes in both solvents severely limits the
range of accessible potentials; 2) The number of organic sol-
vents for which suitable electrolytes could be found to polarize
the interface with an aqueous solution is rather small. Thus it
is a real misfortune that n-octanol could not be used in ITIES
experiments because, to date, no suitable base electrolyte has
been found that would allow to polarize the interface with an
aqueous solution; 3) For most cells, the necessary volume of
the organic solvent is rather large, preventing the use of ex-
pensive solvents.
In the last four years, a new, rather revolutionary approach
for performing an electrochemically driven phase-trans fer of
ions has been developed that uses a three-electrode potentio-
stat. This is possible when a three-phase electrode is used
where an electron transfer proceeds coupled to an ion transfer.
This Review is aimed at outlining the basic principle of such
measurements and summarizing the results obtained so far.
Eventually, we will discuss the future prospects of the metho d.
2. Three-Phase Electrodes
Electrodes with three phases participating actively in an elec-
trode reaction are very common. Each of the three phas es
must share an interface with both other phases. This situation
exists in most of the so-called surface-modified or film electro-
des, many battery and fuel cell electrodes, and electrodes of
the second kind. In fact, the majority of surface-modified elec-
trodes have arrays of particles that partially cover the electrode
surface. Instead of attaching particles to the surface of an elec-
trode, droplets of a water- immiscible liquid can also be attach-
ed. Figure 2 depicts the situation at a three-phase electr ode;
phase II is a droplet or a particle. Since the particle or droplet
contains neutral molecules and ions with equal amounts of
positive and negative charges, any electron transfer between
phase I and II must be accompanied by an ion transfer be-
tween phases II and III. The ion transfer is an indispensable re-
action for maintaining the electroneutrality of phase II, provid-
Fritz Scholz was born in 1955 and
studied chemistry at the Humboldt
University, Berlin, where he received a
Diploma in 1978, his Ph.D. in 1982,
and finished his habilitation in 1987.
At the Humboldt University he held
the position of a Dozent from 1989 to
1993 and of a Professor from 1993 to
1998. Since 1998 he is Professor at the
University of Greifswald. In 1987 and
1989 he worked with Alan M. Bond in
Australia. Beside various contributions
to electroanalysis, he developed a new kind of mercury flow-
through electrodes (bubble electrodes), established the voltamme-
try of immobilized particles as a technique to study the electro-
chemistry of solid compounds, and invented a method to measure
Gibbs energies of ion transfer with the help of three-phase electro-
des. His scientific achievements have been published in more than
200 papers and two books. He founded the Journal of Solid State
Electrochemistry and serves as its Editor-in-Chief since 1987. His cur-
rent scientific interests are focused on three-phase electrodes, elec-
trochemical transformations of solid particles, new pH-sensors, bio-
electrochemical fuel cells, and the interaction of suspended lipo-
somes and clay particles with electrodes.
Rubin Gulaboski, born in 1972 in
Prilep, Macedonia, studied chemistry
at the St. Kiril i Metodoij University,
Skopje, where he received the title of
Graduate Engineer in Chemistry, and
later a M.Sci. in Chemistry. He was the
first Macedonian student to be award-
ed a DAAD fellowship to perform
Ph.D. studies in Germany. He joined
the group of Fritz Scholz in 2001. In
2004 he received his Ph.D. from the
University of Greifswald. Currently, he
holds a postdoc position at the University of Porto, Portugal. He is
co-author of 30 papers and one book.
Figure 1. Representation of the arrangement used in four-electrode voltammet-
ric measurements at ITIES; Ref. = reference electrode, CE = counter electrode.
A
1
+
B
1
is a highly hydrophilic salt and A
2
+
B
2
is a highly lipophilic salt.
18 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemphyschem.org ChemPhysChem 2005,6,16–28
F. Scholz and R. Gulaboski
ed that phase II does not only pass the electrons on from
phase I to a redox species in solution. The latter may happen
in systems where the droplets (or particles) exhibit electrocata-
lytic activity. It is a most important feature of the three-phase
electrode with immobilized electroactive droplets or particles
that the electron and ion transfer must take place simultane-
ously at one electrode (compare with Figure 3). The electron
transfer between phase I and II can be disregarded only when
phase II is a metal to be oxidized because then both phases
are metals.
1
Three-phase electrodes are very well-known as re-
chargeable electrodes in batteries, where phase II is a redox-
active phase, capable of housing charge compensat ing ions
exchanged with an adjacent solution. The observed electro-
chemistry is called “insertion electrochemistry” since ions are
inserted and expelled parallel to the electron exchange. A very
similar electrochemistry can be arranged when phase II is a
droplet of an organic solvent containing a redox active com-
pound, or when the liquid organic phase II is itself electroac-
tive. Thermodynamically speaking, the three-phase electrode
comprising phase II possessing redox centers and ion conduc-
tivity, is a double electrode (the German expression is zweifache
Elektrode
[18]
). The formal thermodynamic analysis of a three-
phase electrode is rather simple: One can split an overall equi-
librium equation, Equation (1):
Ox
xþ
phase II
þ ne
phase I
þ nCat
þ
phase III
Ð Red
ðxnÞþ
phase II
þ nCat
þ
phase II
ð1Þ
with the Nernst equation, Equation (2):
E ¼ E
A
Ox=Red=Cat
þ
RT
nF
ln
a
Ox
xþ
phase II
a
n
Cat
þ
phase III
a
Red
ðxnÞþ
phase II
a
n
Cat
þ
phase II
ð2Þ
into two equilibria, one involving the transfer of electrons,
Equation (2 a):
Ox
xþ
phase II
þ ne
phase I
Ð Red
ðxnÞþ
phase II
ð2aÞ
and one involving the transfer of ions, Equation (2 b):
Cat
þ
phase III
Ð Cat
þ
phase II
ð2bÞ
with the following Nernst equations, Equations (3) and (4):
E
I=II
¼ E
A
Ox=Red
þ
RT
nF
ln
a
Ox
xþ
phase II
a
Red
ðxnÞþ
phase II
ð3Þ
E
II=III
¼ E
A
Cat
þ
RT
F
ln
a
Cat
þ
phase III
a
Cat
þ
phase II
ð4Þ
The standard potentials are interrelated by Equation (5):
E
A
Ox=Red=Cat
¼ E
A
Ox=Red
þ E
A
Cat
ð5Þ
Equations (2 a) and (2 b) both represent electrochemical equi-
libria, since a transfer of charged species between two phases
takes place. When phase II is a solid, it is not yet clear how the
activities of the species Ox, Red, and Cat
+
in the solid have to
be defined and how they could be determined. Furthermore,
experimentally, a separation of the free energies of the reac-
tions given by Equations (2 a) and (2b) is not possible in the
case of solids (because of the inaccessibility of single Galvani
potential differences). When phase II is a solution phase, the
activities of Ox, Red, and C
+
are in principle accessible, howev-
er, it remains that an extra-thermodynamic assumption is nec-
essary in order to quantify the free energy of ion transfer be-
tween the liquid phases II and III.
Figure 4 illustrates the situation in which an electroactive
compound dissolved in a droplet of an organic solvent is oxi-
Figure 2. Schematic drawing of a three-electrode cell with a working electrode
on a surface to which a particle or a droplet is attached (reproduced from
ref. [54]).
Figure 3. Scheme of the simultaneous electron and ion transfer at a three-
phase electrode (reproduced from ref. [54]).
1
When a metal particle, for example, silver, attached to a gold electrode is
anodically oxidized, the ion transfer is the transfer of Ag
+
ions from the
metal to the solution and the electron transfer occurs between gold and
silver. Such electron transfe r between electronically conducting phases will, of
course, always occur in electrochemistry, because any electr ode needs anoth-
er conductor at its terminal. It should be remembered that the interfacial po-
tentials that build up between the electrodes and its terminal conducting con-
nectors are responsible for the inaccessibility of single electrode potentials.
ChemPhysChem 2005,6,16–28 www.chemphyschem.org 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 19
Gibbs Energy of Ion Transfer
dized, accompanied by a simultaneous ion transfer between
the aqueous environment of the droplet and the organic
phase. In Figure 4, it is assumed that a neutral electroactive
compound is dissolved in the organic phase of the attached
droplet. This compound Red can be oxidized to Ox
+
. The oxi-
dation is accompanied by a charge compensating transfer of
anions An
from the aqueous solution to the organic droplet.
Reduction of Ox
+
will reverse the ion transfer. When the drop-
let does not contain any deliberately added electrolyte, its ini-
tial conductivity is very low and the entire electrode reaction
can start only at the three-phase junction, that is, the line sur-
rounding the droplet, since only near to that line will the inter-
face graphite j organic phase attain the applied potential to
drive the reaction given by Equation (1), at least as long as the
ionic conductivity of the droplet is very low. Figure 5 depicts
the potential drop at such a three-phase electrode. The poten-
tial drop Df
graph jaq
at the aqueous solution j graphite interface
is as adjusted by the potentiostat, however, the potential drop
Df
graph jorg
at the interface graphitejorganic solution is Df
graph jaq
diminished by the potential drop Df
aq jorg
at the organic solu-
tion jaqueous solution interface and by the ohmic drop DU
ohmic
inside the organic phase, Equation (6):
D
graphjorg
¼ D
graphjaq
DU
ohmic
D
aqjorg
ð6Þ
The three-phase junction line is a unique feature of
the three-phase electrode. It is a one-dimensional
entity at which the organic liquid, the aqueous elec-
trolyte, and the working electrode are in intimate
contact. At the electrode–aqueous solution interface,
as well as along the three-phase junction line, an
electric double layer is created, providing an abrupt
potential drop that is able to drive the electron-trans-
fer reaction. However, it is only along the three-
phase junction line that: 1) The organic phase pro-
vides the material to be transformed electrochemical-
ly; and 2) The aqueous phase provides the necessary
counter ions. Thus, it is understandable that the elec-
trochemical reaction commences along the three-
phase junction and proceeds later on in the course
of the voltammetric experiment, without any signifi-
cant constrains, into the interior of the droplet.
3. Three-Phase Electrodes with an
Immobilized Droplet of a Dissolved
Electroactive Compound
According to the scenario depicted in Figure 4, the overall re-
action proceeding at a droplet-modified electrode can be writ-
ten by Equation (7):
Red
ðoÞ
þ An
ðaqÞ
Ð Ox
þ
ðoÞ
þ An
ðoÞ
þ e
ð7Þ
If no kinetic constrains exist with respect to the electr on and
ion transfer, the thermodynamic treatment applied to the reac-
tion described by Equation (7) leads to Equation (8), a form of
the Nernst equation:
E ¼ E
A
Ox
þ
ðoÞ
=Red
ðoÞ
þ D
o A
aq,An
þ
RT
F
ln
a
Ox
þ
ðoÞ
a
An
ðoÞ
a
Red
ðoÞ
a
An
ðaqÞ
ð8Þ
Figure 4. A) Schematic representation of the situation in which an organic droplet contains
an electroactive species Red, which is oxidized to Ox
+
accompanied by the transfer of
anions An
from an aqueous electrolyte solution to the organic phase. B) Shows how the re-
action products Ox
+
and An
spread from the three-phase junction into the droplet, creat-
ing an ionic conductivity in the organic phase (reproduced from ref. [54]).
Figure 5. Potential drops at the different interfaces of a three-phase electrode with an immobilized droplet of an organic solvent immersed in an aqueous solution
(reproduced from ref. [55]).
20 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemphyschem.org ChemPhysChem 2005,6,16–28
F. Scholz and R. Gulaboski
![Figure 6. Normalized square-wave voltammograms recorded at droplet-modified electrodes with dmfc dissolved in nitrobenzene (0.1 mol L 1) and immersed in 1 molL 1 aqueous solutions of different sodium salts. The instrumental parameters were: frequency (f), 100 Hz; amplitude (Esw), 50 mV; scan increment (dE), 0.15 mV (reproduced from ref. [54]).](/figures/figure-6-normalized-square-wave-voltammograms-recorded-at-341py27z.webp)
![Figure 7. Forward (If), backward (Ib), and net (Inet) current components of the square-wave voltammetric response of a nitrobenzene droplet containing 0.1 mol L 1 dmfc attached to the surface of the working electrode and immersed in a 1 mol L 1 aqueous solution containing SCN anions. The experimental conditions were: frequency f=100 Hz; amplitude Esw=50 mV; and scan increment dE=0.15 mV (reproduced from ref. [27]).](/figures/figure-7-forward-if-backward-ib-and-net-inet-current-2f3ux9gj.webp)
![Figure 13. Differences in the standard Gibbs energies determined for the transfer of the anions of d- and l-phenylalanine from water to d- and l-2-octanol (reproduced from ref. [32]).](/figures/figure-13-differences-in-the-standard-gibbs-energies-1uo70kab.webp)
![Figure 11. SW voltammetric responses of an n-octanol droplet-modified electrode containing 0.05 mol L 1 dmfc attached to a paraffin-impregnated graphite electrode which is immersed in a 1 mol L 1 aqueous solutions of (C6H5)4B (TPB) = tetraphenylborate, BF4 (TFB) = tetrafluoroborate, SCN , ClO4 , and NO3 (reproduced from ref. [22]).](/figures/figure-11-sw-voltammetric-responses-of-an-n-octanol-droplet-1oj2z0fa.webp)
![Figure 12. Correlation between the standard partition coefficients of organic anions measured in the system water jNPOE, water jNB, and that measured in the system water jn-octanol for the compounds given in Figure 1 of ref. [30] (reproduced from ref. [30]).](/figures/figure-12-correlation-between-the-standard-partition-1yb1o0ln.webp)
