Advance Publication by J-STAGE

ACCEPTED MANUSCRIPT
Electrochemistry Received: September 25, 2022
Accepted: September 29, 2022
Published online: See the URL below.

The Electrochemical Society of Japan https://doi.org/10.5796/electrochemistry.22-66111

Review Electrochemistry, (in press)

The 66th special feature

"Novel Aspects and Approaches to Experimental Methods for Electrochemistry"

Electrical Conductivity Measurement of Electrolyte Solution†,††

Minoru MIZUHATAa,b,*,§
a
Department of Chemical Science and Engineering, Graduate School of Engineering, Kobe University,
1-1 Rokkodai-cho, Nada, Kobe 657-8501, Japan
b
Faculty of Chemistry, Jagiellonian University, Gronostajowa 2, 30-387 Kraków, Poland
* Corresponding author: mizuhata@kobe-u.ac.jp

© The Author(s) 2022. Published by ECSJ. This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 License (CC BY,
http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse of the work in any medium provided the original work is properly cited. [DOI:
10.5796/electrochemistry.22-66111].

This is the final version of an authors’ manuscript, submitted by the author(s) and accepted for publication after peer review and technical editing by the Editorial Board of The
Electrochemical Society of Japan. This manuscript may contain minor errors or incomplete designs that do not affect the judgment for publication. It is the responsibility of the
authors to correct any errors in the Just Accepted manuscript during galley proof review.

³
This review is translated from an article in Denki Kagaku, 86, 42 (2018)1
with permission of the Electrochemical Society of Japan and is based on the
original text lectured by the author at the 35th Electrochemistry Workshop
organized by the Kansai Branch of the Electrochemical Society of Japan,
published in November 2005, with substantial revisions and additions to the
content to meet current IUPAC and Japanese Industrial Standards (JIS).
³³
This paper constitutes a collection of papers edited as the proceedings of
the 51st Electrochemistry Workshop organized by the Kansai Branch of the
Electrochemical Society of Japan.
§
ECSJ Active Member
M. Mizuhata orcid.org/0000-0002-4496-2215
Review

Electrical Conductivity Measurement of Electrolyte Solution†

IPT
Minoru MIZUHATAa,b,*,§

SCR
a
Department of Chemical Science and Engineering, Graduate School of Engineering, Kobe

University, 1-1 Rokkodai-cho, Nada, Kobe, 657-8501 Japan

b
Faculty of Chemistry, Jagiellonian University, Gronostajowa 2, 30-387 Kraków, Poland

*Corresponding Author: mizuhata@kobe-u.ac.jp

§

†
ECSJ Active Member
NU
This review is translated from an article in Denki Kagaku, 86,42 (2018)1 with permission of the

Electrochemical Society of Japan and is based on the original text lectured by the author at the 35th
MA
Electrochemistry Workshop organized by the Kansai Branch of the Electrochemical Society of Japan,

published in November 2005, with substantial revisions and additions to the content to meet current

IUPAC and Japanese Industrial Standards (JIS).

TED

ORCID

M. Mizuhata: 0000-0002-4496-2215
CEP
AC
Abstract

Methods for measuring and analyzing the electrical conductivity of electrolyte solutions are reviewed.

IPT
The accuracy of conductivity measurements, which have been around for more than 100 years,

depends on the physical properties related to ionic conductivity, especially the concentration of the

reference material, the electrical signal at the time of measurement, and the temperature, each of

SCR
which has been redefined each time according to changes in definitions and standards. This article

summarizes elemental factors in electrical conductivity measurement, including the concentration

and traceability of potassium chloride as a reference material, the history of the bridge method and

types of measuring cells, and important notes and recent topics regarding the concept of temperature

dependence.

Keywords:

principle of electrical conductivity measurement;

NU
MA
concentration and traceability of potassium chloride;

Kohlrausch bridge and automatic balancing bridge;

temperature dependence in Arrhenius and VTF equations

TED
CEP
AC
1. Introduction

In the field of physical science, the electrical conductivity, σ, is defined as the coefficient of the

IPT
electric field E on the current density i by Ohm’s law as follows:

i=σE (1)

SCR
Based on this definition, measuring the electrical conductivity σ means placing a charged particle

in an electric field and monitoring its movement. Most charged particles used in electrochemistry are

electrons and ions, and in liquid electrolytes, the charged particles are exclusively dissolved ions (this

paper does not deal with plasma in liquids). Therefore, the ionic conductivity is an important basic

NU
property of liquid electrolytes, and there are a wide range of considerations in its measurement, such

as the theory of ionic conduction, measurement techniques (AC or DC measurement), and the type

of conducting species (total or individual ionic conductivity) to be measured. For example, electrical
MA
conduction at the interface between the sample and the electrode (current collector) is caused by

charge transfer reactions in which ions transfer electrons. In this case, not only the steady-state current

caused by the transfer of charged particles, but also transient responses and polarization phenomena

such as the reaction current in electrochemical reactions and the charging current in capacitors
TED

(charging of charged particles near the electrode) due to the electric field response and adsorption

behavior of dipoles are involved. Therefore, it is necessary to select a measurement method that

corresponds to phenomena with different time constants. Thus, it can be said that electrical

conductivity measurement reflecting ionic conduction is a measurement and interpretation of the

CEP

current response to changes in electric potential, and an appropriate measurement method should be

selected in consideration of what kind of substance or charge transfer is occurring in the measurement

system. Since ionic conduction is a mass transferring phenomenon, it is necessary to always consider

the other conduction behaviors mentioned above, to distinguish between the factors that should be

eliminated, which are often the cause of measurement errors, and to make accurate and precise
AC

measurements.

Methods and theoretical development of ionic conductivity measurement have long been

conducted and established by Faraday, Hittorf, Kohlrausch, Debye, Hückel, Onseger, and others, and
their genealogy can be traced in books and textbooks.2 Most textbooks for beginners only describe

the measurement of electrical conductivity according to early methods, using KCl as the reference

material, using an AC Kohlrausch bridge at a given frequency, often 1 kHz, and measuring at a fixed

IPT
temperature. However, this description of measurement conditions must be revised, as more

meaningful measurement methods have been developed as concentration and temperature definitions

and measurement techniques have evolved.

SCR
In the early days of electrochemistry, the concept of ionic conduction summarized by Kohlrausch,

who systematically measured the electrical conductivity of electrolyte solutions, was the basis for

subsequent measurements of transport number and ionic conduction mechanisms, along with the

establishment of the conductivity measurement method using the Kohlrausch bridge, which he

modified from the Wheatstone bridge, and can be considered a sufficiently established method for

NU
measuring ordinary electrolyte solutions.3 The conductivity of concentrated aqueous electrolyte

solutions, which has been increasingly studied in recent years, has been considered as an extension

of the theory of dilute solutions, such as the decrease in conductivity due to ion-pair formation. On
MA
the other hand, since the Walden rule was proposed for the correlation between viscosity4 and

conductivity5−7 of viscous electrolyte solutions,8−10 many researchers including Angell,11,12

Watanabe,13,14 and others have studied ionic conductivity not only in concentrated electrolyte

solutions but also in molten salts, ionic liquids, and gel electrolytes from the viewpoint that ionic
TED

interactions affect physical properties.

Recently, due to restrictions on the amount of sample and atmosphere in which electrical

conductivity must be measured, the conventional method of measuring aqueous solutions at room

temperature and pressure (standard conditions) is becoming increasingly inapplicable. In addition,

CEP

electrolytes, which are liquid materials, are not only used as they are in solution, but are also used in

systems in which they coexist with porous solid materials (powders, fibers, fillers, etc.) to support

their shape. Also, the conductive mechanisms of those ions have also been extensively studied.15−28

However, since explaining everything would be confusing and would not be appropriate for this

paper, we will focus on how to accurately measure ionic conductivity, the reasons for doing so, and
AC

points to keep in mind. Reviews of ionic conductivity measurements have been published many times

in predecessor and this journals, and the author and his coauthors have had the opportunity to write

commentaries on electrical conductivity measurement methods29 and their interpretations, focusing

on impedance measurements in Japanese.1,30,31 This manuscript is a translation of a Japanese version

of a commentary previously written by the author,1 in which the description includes international

standards in addition to the content for Japan regarding industrial standards.

IPT
2. Specific Electrical Conductivity and Cell Constants

SCR
2.1 Resistance and electrical conductivity/specific electrical conductivity

The definition of electrical conductivity according to Ohm’s law has already been described as Eq.

(1), and additionally, the concept of concentration is introduced into the unit system convention and

conductors in the field of electrochemistry. Some properties are defined as follows, together with the

units used. When a current I (A) flows for an applied voltage E (V) in a measurement, its resistance

R (ohm, Ω, or S−1) is defined by Ohm’s law as

NU
R = E/I (2)

where R is the reciprocal of the electrical conductivity, σ. The resistance R depends on the shape of
MA
the sample and is proportional to the reciprocal of the sample length l (m) and the cross-sectional area

A (m2). In order to calculate the intrinsic electrical conductivity of a material, the cell constant Kcell

(m-1) is given by

Kcell = l/A (3)

TED

The specific electrical conductivity of each material is defined as κ (or σ) (S m−1) and is calculated

by the following equation. The specific resistance, ρ, is the reciprocal of κ.

CEP

κ = ρ−1 = KcellR−1 (4)

The above resistivity values can be measured directly for solid materials where the sample geometry

can be fixed. However, it is difficult to directly measure the specific conductivity of liquids because

the sample geometry is not fixed, and in the case of ionic conductors, there is the Parker effect32,33
AC

where the electric lines of force between the sample cross-sectional area and electrodes do not always

coincide due to the state of the conductive path. Therefore, a measurement cell that is easy to
reproduce is used, and the cell constant, Kcell, is determined for each cell by measuring the resistance

Rstd of a reference material with a known specific conductivity κstd.

IPT
Kcell = κstdRstd (5)

The standard material is an aqueous KCl solution as described in Section 2.2. The electrical

SCR
conductivity of an ionic conductor is determined by the product of the number of ions as the

conducting species and their mobility. The number of ions is expressed as the amount of substance

of ions in a fixed volume of electrolyte. In usual electrochemical experiments, the concentration is

often defined by the volume molar concentration C, and the electrical conductivity per volume molar

concentration is called the equivalent conductivity Λ, as indicated as Eq. 6,

NU
Λ = κ/C (6)
MA
Sometimes, the SI unit system is “strongly” recommended in textbooks, and in general rules for

electrical conductivity measuring method in Japanese industrial standard: JIS K 0130−2008, which

will be described later, it is defined using only basic units, so it is shown as in Eq.6. However, in the

past, “centimeter (cm)” was conventionally used as the unit of length in electrochemistry, and liter,
TED

L (= dm−3), was often used for volume. Therefore, S cm−1 and mol L−1 were so widely used as units

of specific conductivity and concentration, respectively, that the following equation was used in the

definition of equivalent conductivity,

CEP

Λ = 1000 κ/C (6’)

in the definition formula of equivalent conductivity. The coefficient “1000” will remain in the

equation for the time being due to the customary notation of units.
AC

2.2 Reconsideration of principals for electrical conductivity measurement

In the conductivity measurement by Kohlrausch et al. in 1898,3 KCl was selected as the reference

material because it is a strong electrical solution containing K+ and Cl– as dissolved species, it is easy
to purify, weigh, and handle, and the mobility of the constituent ions, K+ and Cl–, is almost equal.

The method for determining the absolute conductivity of aqueous KCl solution, which is used as a

standard for calculating cell constants, requires an agreement on the substance and measurement

IPT
environment that guarantees traceability and reproducibility.

As the electrical conductivity of aqueous solutions obtained from KCl is not calculated

theoretically, it is determined by measuring the “absolute values” of electrical conductivity of

SCR
aqueous KCl solutions at a given concentration. This concept remains unchanged to this day. Since

the temperature standard of the SI unit system was redefined by ITS-90 in 1990,34 the electrical

conductivity of aqueous KCl solutions, whose measured values are greatly affected by temperature,

had to be measured again. The current electrical conductivity measurements of aqueous KCl solutions

based on the SI unit system standards are those reported based on the IUPAC verification in 2001.35

NU
The new values were established after a complete review of past measurements of the concentration

of KCl as the reference material, the measurement cell, and the measurement method, and after

identifying problems in the measurements. The Japanese Industrial Standards also revised the general
MA
rules for electrical conductivity measurement methods in accordance with these standards, which are

now in operation as JIS K 0130−2008.36 Therefore, by verifying the past measurement methods for

absolute values of electrical conductivity of aqueous KCl solutions, from which the IUPAC35 and

JIS36 standards were derived, it is possible to determine how general electrical conductivity
TED

measurements can be made and how the measurement of electrical conductivity can be improved.

The following is a description of the past methods for measuring the absolute value of electrical

conductivity of aqueous KCl solutions, from which the IUPAC and JIS standards were derived. The

purpose of electrical conductivity measurement of KCl aqueous solution is mentioned from the
CEP

viewpoint of traceability by the Research Institute of Standards for Measurement, National Institute

of Advanced Industrial Science and Technology (AIST).37,38, There is no ISO standard equivalent to

JIS K 0130-2008 mentioned above. ISO 7888:1985,39 which is equivalent to JIS K 0102,40 exists in

Japan for electrical conductivity measurements for water quality control that contribute to

environmental measurements. These differ in the concentration range of KCl applied, with JIS K
AC

0130 covering aqueous solutions with values from 5 S m‒1 to 200 S m‒1, whereas ISO 7888 (JIS K

0102) refers to measurements from 1 mS m‒1 to 500 S m‒1. The reason why the lower limit of
electrical conductivity for dilute solutions is set at 1 mS m‒1 is that the contribution of H+ and OH- to

electrical conductivity due to water dissociation cannot be ignored.40

IPT
2.2.1 Aqueous potassium chloride (KCl) solution as standard solution for conductivity

measurement

When measuring the electrical conductivity of liquids, aqueous potassium chloride (KCl) solutions

SCR
prepared by the experimenters are usually used in many fields including electrochemical research.

KCl A reagent of KCl specified in JIS K 8121-200741 deviated from ISO 6353-3:198742 is used.

Special grade reagents commercially available from reagent companies satisfy these regulations

except for the water content, and may be used in ordinary experiments if stored under careful control.

However, they must be dried in accordance with the regulations. The reagent should be powdered in

NU
an agate mortar, heated at 500°C for 4 hours, and cooled in a desiccator.41 By this process, the special

grade reagent satisfies the specification for “KCl for electrical conductivity measurement.” The

original IUPAC standard for storage describes drying in a desiccator with Mg(ClO4)2,35 but
MA
Mg(ClO4)2 is not suitable for drying solvents,43 and in Japan it is a class 1 hazardous material, so in

JIS, simply use airtight containers. Therefore, in JIS, an airtight container is simply used. Although

not specified in JIS, drying in a vacuum desiccator using P2O5 is a comparable drying method.

Water: Use pure water with an electrical conductivity of 0.2 mS m–1 (25 °C) or less.35,44
TED

Weigh the amount of KCl: According to the molecular weight of KCl (74.552 g mol‒1), weigh a

predetermined amount of KCl pretreated as described above and dissolve it in 1 kg of water. In

practice, it is difficult to precisely weigh the above amounts, so if the amount of KCl collected is

within 1% of the specified amount, the weight of water may be adjusted so that the KCl concentration
CEP

is identical.44

It should be noted that the concentration of the solution specified here is defined as a mass molar

concentration mol kg‒1, i.e., a solution of a certain amount of KCl dissolved in 1 kg of solvent (water).

JIS K 0130-2008, the general rule for electrical conductivity measurement methods specified in the

Japanese Industrial Standards (JIS), also uses the electrical conductivity value based on this IUPAC
AC

measurement.35 The three types of conductivity values, 1 mol kg‒1, 0.1 mol kg‒1, and 0.01 mol kg‒1,

are defined in the IUPAC standard above. 1 mol kg‒1, 0.1 mol kg‒1, and 0.01 mol kg‒1, and for 0.001

mol kg‒1, the literature value measured for aqueous KCl solutions specified in JIS K8121-2007
(potassium chloride (reagent)) is specified as the electrical conductivity of KCl.45 The reason for

using the molar concentration by weight is to ensure that changes in solution volume due to

temperature change do not affect the concentration.35 In a detailed experiment by Jones et al. in 1933,

IPT
the unit of concentration for electrolyte solutions was the demal (the amount of solute dissolved in 1

kg of solution, sometimes referred to as “D”),46 This unit was used for many years until the

determination of the electrical conductivity of KCl, which was defined by IUPAC, the SI system of

SCR
units, according to the IPTS-68 temperature standard.47 There was confusion due to the difference

between the academic field using the SI unit system as the standard and the industrial field using the

JIS standard as the standard. In JIS K0130-2008, for the first time in JIS standards, a measurement

method based on IUPAC units was made as the general rule, thereby defining a reference material

for electrical conductivity that is consistent both academically and industrially. Hence, the problem

NU
of determining the concentration of the reference solution has now been almost eliminated. It should

be noted that recent revisions can be found in currently published textbooks, handbooks, and

procedures, some of which are based on the previous plan such as JIS K0130-1995.
MA
Standard solutions of KCl prepared under the guarantee of traceability based on such concentration

specifications are supplied in many countries. In addition, disposable standard samples for which the

value of electrical conductivity is specified without specifying the type of electrolyte are sold as ISO

17025 Conductivity Calibration Standards,48,49 and reagent manufacturers in various countries are
TED

selling reagents complying with these standards.

In Japan, electrical conductivity standard solutions (aqueous KCl solutions) based on certification

issued by the National Metrology Institute of Japan, National Institute of Advanced Industrial Science

and Technology (AIST) began to be distributed in 2015.50 As of 2017, three types of KCl solutions
CEP

(1 mol kg‒1, 0.1 mol kg‒1, and 0.01 mol kg‒1. Currently, electrical conductivity reference materials

are sold by companies that comply with ISO 17034:2016 “General requirements for the competence

of reference material producers.”51 KCl is generally used and is a disposable product whose packaging

is defined by its electrical conductivity value, not by the concentration of the electrolyte solution.

However, these solutions are extremely expensive and are rarely used in practice, except when it is
AC

necessary to prove that the measurement equipment and data handling based on the measurement

results conform JIS. For ordinary measurements, standard solutions that conform to JIS standards

with sufficient accuracy are available from reagent manufacturers at low prices, and these can be used.
2.2.2 Measuring cell

Jones et al. precisely measured the electrical conductivity of KCl in the 1920s and 1930s firstly.

They made a major contribution to the establishment of the electrical conductivity measurement

IPT
method by improving the accuracy of measurements with the Kohlrausch bridge. They made a

number of improvements in elemental techniques related to the verification of actual measurements

of electrical conductivity by AC bridge,52 improvement of the power supply (AC oscillator) and

SCR
detector,53 cell shape,54 validity of Ohm’s law in electrolyte solutions,55 effect of polarization on the

application of AC,56 fabrication of platinum black electrodes,57 and use of oscilloscopes,58 and in

1937, he proposed a standard value for the electrical conductivity of aqueous KCl solutions

finally.46,59 In the IUPAC measurements reported in 2001, the Jones cell was used as one of the

measuring cells, and its accuracy was verified.35 Two measurement methods were used: 0.01 mol kg‒
1

NU
and 0.1 mol kg‒1 KCl solutions, the Jones cell shown in Fig. 1a was used to measure AC impedance

values at frequencies f = 1, 2, and 5 kHz, respectively using Jones bridges.46,52 The impedances at

each frequency were extrapolated to f → ∞ and the resulting values were used as resistance ones.
MA
However, this method is difficult to use for a 1 mol kg‒1 KCl solution with high electrical

conductivity because of its electrode polarization. Therefore, the cell shown in Fig. 1b was used and

the DC 4-terminal method was used. Here, nylon Swagelok joints are used to connect the vessels.

Resistance was determined by applying a given current between the outer electrode 1 and measuring
TED

the potential difference between the two electrodes. Since polarization occurs only at the current

probe, the conductivity of KCl can be determined directly from the cell shape since there is almost

no change in concentration between the potential probes. After confirming the reproducibility of the

conductivity obtained, the values shown in Table 1 were recorded and used as the IUPAC standard
CEP

values. However, there are some important points to be recognized when considering the method of

conductivity measurement of electrolyte solutions.

1. The appropriate measurement method must be used depending on the magnitude of the

electrical conductivity. That is, the best possible measurement method depends on the

conductivity of the material.

AC

2. AC measurements are made at 1, 2, and 5 kHz and extrapolated to f → ∞.

3. The position of the potential probe in a 4-terminal DC measurement is not defined by the

electrode insertion position, but by the liquid entanglement.

 These conditions suggest that the best method should be considered on a case-by-case basis,

depending on the nature of the liquid sample we are measuring, especially its electrical conductivity.

When measuring ionic conductivity in solids, the sample geometry directly defines the cell constants,

IPT
meaning that we are measuring absolute values of conductivity. In contrast, for liquid samples,

electrodes are placed at arbitrary positions in the measurement cell, so one should always be aware

that there are many things to be verified in terms of electrode size, contact conditions, and distance

SCR
between electrodes. General information on the measurement cell is described in section 4.

3. Circuit for Electrical Conductivity Measurement

3.1 Kohlrausch bridge

NU
A well-known circuit for electrical conductivity measurement is the Kohlrausch bridge which is a

derivative of the Wheatstone bridge equipped with an AC power supply. This circuit combines an

RC circuit with a known resistor element in parallel with a variable capacitor, a variable resistor, and
MA
a current meter as shown in Fig. 2, and uses the AC method for measurement. Here, the resistance

value of the sample, Rs, is given by

Rs = R1R2/R3 (7)
TED

Whereas this equation is valid for a circuit consisting of only a resistance component without a

capacity one, in the case of ionic conductors, there is a capacitance component due to the time delay

required for ion transfer and polarization phenomena at the electrodes. Jones proposed the use of a
CEP

variable capacitor to compensate for the phase difference in the AC current caused by the capacitance

of the sample in the conductivity measurement cell and made significant improvements to subsequent

measurements using bridges in which the description is written in the section of “Conditions of

Balance of the Bridge Analyzed” in Ref.52. As described above, the bridge method has been widely

used in conductivity measurements because it can provide very accurate values with a simple
AC

oscillation circuit. However, as shown in the next section, automatic balanced bridge circuits are now

widely used, and commercially available bridges for measuring the electrical conductivity of

electrolyte solutions are rarely available.

3.2 Automated balanced bridge

The auto-balancing bridge circuit is currently the most typical circuit used to measure electrical

IPT
conductivity by the AC method and is a typical circuit system for impedance meters and LCR meters.

The potential is generated by a power supply (AC oscillator) as shown in Fig. 3, and the potential

difference generated in the connected sample is measured. In this case, either the 4-terminal method

SCR
or the 2-terminal method is used, depending on the electrode configuration. The current value is

converted to a current value by measuring the potential difference generated in a fixed resistance

(shunt resistance) connected in series with the sample. Since a fixed resistor with a value similar to

that of the sample is suitable for measuring the potential, an appropriate resistance value is

automatically selected for the current measurement. The sample resistance is measured from the

NU
potential difference between the samples and the current value through an A/D converter and a

division operation circuit. Since an alternating current is applied, the resistance value obtained is

treated as an impedance as for the Kohlrausch bridge, and the phase of the current and potential is
MA
also measured. The applied AC frequency varies according to the manufacturer, but values in the

range of 1 – 100 kHz are used, and the higher the frequency, the smaller the resistance value (the

greater the electrical conductivity).

TED

3.3 DC four-terminal method

When measuring the resistance of a sample showing a low resistance value, such as a highly

concentrated electrolyte solution, AC resistance measurement is often difficult in terms of frequency

selection because of excessive current flow. Therefore, the DC four-terminal method (sometimes
CEP

called Kelvin method) is used. The cell configuration is shown in Fig. 1b. Since electrode reaction

and polarization are inevitable when using direct current, this method requires that the electrode

section and the voltage measurement section be separated in the cell to prevent the effect of electrode

polarization on the section measuring the potential difference. A DC power supply is connected to

the electrodes at both ends of the cell, and an electrometer with a high internal impedance of at least
AC

1013 Ω is used to measure the potential difference and calculate the resistance value. In this case, since

the amount of current can be varied over a wide range, the resistance should be calculated from the

slope of the I-V response after confirming linearity. A volt meter or tester with low internal impedance
is not suitable for the measurement because it itself becomes a power source. The method of

measuring conductivity with KCl standard solution, calculating the cell constant, measuring the

sample resistance, and calculating the electrical conductivity using Eq. 5 is the same as the AC method.

IPT
Although this method was partially applied in IUPAC measurement,31 the direct current method is

not recommended in JIS K 0130-2008.

SCR
4. Sample Cells

4.1 Shape of measuring cell

There are many designs of electrical conductivity cells as shown in Fig. 4, including cells of various

shapes invented by Kohlrausch and currently available on the market, and the most suitable shape

NU
depends on the amount of sample, resistance, and stability of the sample. In particular, electrical

conductivity is highly dependent on temperature and must be measured while maintaining the

electrolyte at a constant temperature. For this reason, there are cells in which the sample is injected
MA
into a solution bath for measurement for Figs. 4a-e, and cells in which the sample is flown through

the cell for measurement as shown in Fig. 4f.

Recently, commercially available cells for electrical conductometers are increasingly being used.

Generally, as shown in Fig. 5. a platinum resistance thermometer, thermistor, or thermocouple, and a

TED

glass tube containing conductors connected to the electrodes, with platinum electrodes installed above

and below on a ring, are widely used as an integrated type. In the case of using an integrated cell, a

protection cell is used. Regarding the thermometers provided here, the temperature standard ITS-90,

which was the reason for the revision of JIS K0130-2008, defines a platinum resistance thermometer
CEP

as a supplementary instrument at 13.8033 K-961.78°C. When an integrated cell is used, it is often

equipped with a protective tube. The following precautions should be taken as follows.

- The protective tube shall not be taken for the above purpose as shown in Fig. 5a. The protective

tube is used to avoid the influence of external diffusion currents and to keep the lines of electric

force only within the protective tube to avoid large changes in the cell constants due to large or
AC

small conductivity of the electrolyte. When the protective tube is removed, the value of the

electrical conductivity may increase. This value is due to the spread of the lines of electric force

inside the vessel and is not a correct value.

 - For the measurement under the condition for Fig.5b, no air bubbles shall be allowed to enter or

adhere to the protective tube around the electrode in order to avoid the internal volume of the

electrolyte solution changing and continuity being lost.

IPT
- Although the cell shown in Fig. 5c appears to have “three poles,” the upper and lower poles

have the same polarity. Therefore, the leakage of lines of electric force does not occur as in Fig.

5b, and the cell constants are extremely accurate.

SCR
- The quadrupole cell shown in Fig. 5d has current and voltage probes arranged independently.

For conductivity measurements of highly resistive solutions, the distance between electrodes

should be narrower and the electrode area should be larger. A conductivity cell with extreme

geometries, such as Keysight 16452A liquid test fixture is commercially available for measuring the

NU
capacitance component of electrolyte solutions and solvents with very high resistance values. It has

a diameter of 38 mm and a very small cell constant with spacers that allow the distance between

electrodes to be adjusted from 1.3 mm to 3.0 mm for capacitive measurements. The cross-sectional
MA
area and distance between the electrodes are defined, so the cell is capable of measuring absolute

resistance.

4.2 Platinum-black electrode

TED

Electrodes are often made of platinum or platinum-black because they must be resistant to

corrosion by the electrolyte, but since JIS K 0130 specifies that electrodes must be “impervious to

specimens,” titanium and stainless steel (SUS316 etc.) are sometimes permitted in commercial

products. When using a platinum electrode, it is recommended to use a platinum-black electrode in

CEP

which a platinum electrode is inserted into a platinum chloride acid solution to secure a sufficient

effective area at the electrode, and a black velvety platinum dendrite is formed by electrolysis as

shown in Fig. 6. The method of producing platinum black is as follows. (See Ref. 33, p.5)

(a) Removal of platinum black

Platinum black can be easily removed by electrolyzing a platinum (black) electrode as an anode in
AC

about 1 mol L‒1 hydrochloric acid.

(b) Plating method

 Weighed 30 g of H2PtCl6･6H2O and 0.25 g of (CH3COO)2Pb･3H2O are dissolved in water, and 1 L

of electrolyte is prepared. A platinum electrode is placed in the electrolyte, and the electrolyte is

energized (35‒140 kC m‒2) for approximately 10 min, switching polarity several times at a constant

IPT
current density (100-400 A m‒2) while stirring the electrolyte using a power supply of approximately

6 V DC.

Next, the platinum black is energized in 0.033 mol L‒1 sulfuric acid for about 30 min, changing the

SCR
current direction occasionally, to remove adhered or absorbed hexachloroplatin(IV) acid and chlorine

gas. Since polarization is inhibited once Pt black is applied, care should be taken with the applied

voltage to prevent dissolution of the Pt black on the anode side.

Once a Pt/black electrode is dried, surface wetting is impaired and diffusion of the electrolyte into

the Pt/black is inhibited. Therefore, it is necessary to take care of the maintenance of both commercial

NU
and home-built cells. This indicates that the cell is not suitable for non-aqueous electrolytes,

electrolyte solutions in which dissolved species are adsorbed or react with the Pt black, or colloidal

solutions. In the case of the measurement of the electrical conductivity of non-aqueous electrolytes,
MA
a Pt plate can be used. However, since polarization tends to occur, it is desirable to measure at a

frequency at which polarization hardly occurs, based on an understanding of the frequency dispersion

state by AC impedance. Recent electrical conductivity measurement equipment often measure

conductivity in a relatively high-frequency range and have a circuit system that minimizes the
TED

polarization effect at the electrode, so it may be better to use a platinum plate electrode when the

electrical conductivity is somewhat small. In any case, even when platinum plate electrodes are used,

care should be taken to ensure that the shape and arrangement of the electrodes, the orientation of the

electrodes, and the parallelism between the electrodes in the case of opposing plate electrodes do not
CEP

change.

4.3 Two-pole and four-pole cells

JIS K 0130-2008 recommends the AC 2-electrode method, the AC 4-electrode method, and the

electromagnetic induction method as electrical conductivity meters as mentioned in 4.2. The former
AC

two methods are usually applied in electrochemical measurements, and often comprise a 2-pole cell

and a 4-pole cell, respectively. The characteristics of these methods are shown in Table 2. The two-

pole method serves as both a current collector and a voltage measuring device, and is affected by the
polarization of the electrodes. The quadrupole method, on the other hand, separates the current-

carrying electrode from the voltage-measuring electrode, and since the voltage-measuring electrode

is connected to a high-impedance voltage measuring device, only a very small amount of current

IPT
flows through it and is not affected by electrode polarization. Therefore, the polarization resistance

indicated by R2 and R3 is almost negligible. The quadrupole method minimizes the electric field effect

of the sample container in the sample cell because the area where the electric field is applied is limited

SCR
to the vicinity of the electrodes. Therefore, the sample volume and container volume have little

influence on the measured values.

4.4 Connection between cell and device

The connection between the cell and the device is made with a conductor of good electrical

NU
conductivity. Normally, commercially available conductors are used as they are, and little attention

is paid to the following points, including the AC method.

- Avoid using unnecessarily long connectors to avoid errors due to low terminal voltage. When
MA
connecting electrodes with clips, etc., avoid increasing contact resistance due to corrosion by

cleaning, polishing, etc.

- In the case of the AC method, long conductors cause a decrease in terminal voltage and a

decrease in line-to-line capacitance. In particular, when using an impedance meter to measure

TED

conductivity, it is necessary to use a coaxial cable and ground the conductor itself as an

inductance component in the high-frequency range and to ground the conductor in parallel

between the cell and the device because of the contribution of the stray component of

superimposed interline capacitance. It should be noted that coaxial cables must be used and
CEP

grounded to equipment.

5. Electrical Conductivity Measurement by AC Method and Frequency Dependence

However, the appropriate measurement frequency range differs depending on the sample
AC

resistance. Since the frequency at which polarization occurs is often different for each sample, the

value measured with the KCl standard solution is not necessarily applicable at all frequencies.

When measuring liquid electrolyte solutions by frequency dispersion, it is necessary to apply an

 AC voltage within the potential window where no electrode reaction occurs, and the semicircle

corresponding to the charge transfer resistance component in the electrochemical impedance is not

seen, and only the polarization impedance on the electrode at the low frequency side is obtained.

IPT
Only the polarization impedance can be obtained, so only a straight line is often seen in the complex

impedance plot, as shown in Fig. 7. If conductivity is calculated by impedance measurement,

conductivity can be easily calculated by taking the solution resistance as the point where the

SCR
straight line and the real axis intersect. Here, we would like to make a note on the frequency

dependence of the AC measurement at an arbitrary AC frequency without impedance measurement.

Figure 8 shows the frequency dependence of various electrolytes as a Bode diagram. Due to the

effect of electrode polarization, the absolute value of impedance |Z| increases at low frequencies.

However, the frequency at which the increase is observed depends on the characteristics of each

NU
sample and shows different values, indicating that it is not desirable to calculate all conductivity

values using only the value measured at 1 kHz, for example. The resistance of each electrolyte

should be measured at the higher frequencies where the polarization resistance is no longer
MA
observed, thus indicating that impedance measurement is suitable for liquid samples as well, since

the frequency dispersion can be easily determined. In practice, it is known that commercially

available cells select an appropriate AC frequency according to the resistance of the sample to be

measured. For example, Radiometer Analytical uses 94 Hz for the 4‒40 µS range and 46.9 kHz for
TED

the higher range. The relaxation times corresponding to these phenomena correspond to the

reciprocal of the frequency, and thus capture phenomena on the order of millisecond to

microsecond. The adsorption and desorption of ions near the electrode and the orientation of the

solvent occur on this timescale and are responsible for the polarization phenomena. However, the
CEP

Debye-Falkenhagen effect,60-63 a phenomenon in which the ionic atmosphere follows the electric

field based on the electric field gradient around the ion in the bulk, is a relaxation phenomenon in

the picosecond to nanosecond range, and thus the resistance value varies in the GHz range. If the

conductivity of an aqueous electrolyte solution measured in the kHz range shows no frequency

dependence, it is reasonable to assume that the conductivity is the reciprocal of the resistance at
AC

that time.

6. Temperature Dependence of Electrical Conductivity

6.1 Accuracy of measurement temperature

It is widely known that electrical conductivity is a transport property of ions and that it depends

IPT
sensitively on temperature. In other words, the temperature control of the measurement system must

be precise. In general, a 1:1 electrolyte such as KCl at a concentration of about 1 mol kg‒1 shows a

conductivity change of about 2% per degree. Therefore, if we require three significant digits of

SCR
conductivity (expanded uncertainty 0.1%), stability at temperatures below 0.01 °C must be ensured.

To guarantee this accuracy, in Japan, the temperature measurement is equivalent to using a

calibrated platinum resistor guaranteed by a calibration laboratory registered under the Japan

Calibration Service System (JCSS),37 so it is necessary to fully consider how much accuracy can be

achieved in the interpretation of the measured value of electrical conductivity. In general,

NU
thermocouples and thermistors are often used because relative comparisons of electrical conductivity

are often acceptable from an academic standpoint, rather than absolute measurements. On the other

hand, when summarizing results for which a description of accuracy is required, specifically the
MA
temperature measurement method and its instruments should also be described in the paper.

When measuring temperature, it is desirable to insert the sensor part of the thermometer into the

cell and measure the temperature of the electrolyte directly. If this is not possible due to the shape of

the cell or the nature of the electrolyte, the temperature stability of the thermostatic bath should be
TED

maintained to the same degree, and the stability of the temperature in the cell should be confirmed by

a constant value of electrical conductivity to be used as the measurement value.

6.2 Differences between industrial standards and scientific analysis

CEP

Interpretations of the temperature dependence of electrical conductivity vary widely depending on

the purpose of the measurement. Commercially available electrical conductivity meters must conform

to some industrial standard, and the same applies to the handling of temperature. In conclusion, the

temperature compensation function of electrical conductivity meters should not be used when

measuring the variation of electrical conductivity with temperature (temperature dependence) from
AC

an academic standpoint or when measuring an unknown sample at an arbitrary temperature. The

reason for this is explained in a note using JIS K0130-2008 or ISO 7888:1985, which is used in

domestic commercial products as an example. (See Appendix A) The temperature dependence of the
electrical conductivity of KCl is expressed by a polynomial approximation in Jones’s study,46

IUPAC,35 and JIS,36 and has no physicochemical meaning. In the case of temperature compensation

by nonlinear approximation, the function used is also a polynomial function centered on the electrical

IPT
conductivity of an aqueous solution of KCl at 25 °C, and it is clear that it cannot be applied to other

samples. In other words, if the temperature compensation function of the conductometer is turned on

when measuring the electrical conductivity of a sample, the electrical conductivity will only be

SCR
corrected with respect to the 25 °C KCl solution according to the temperature compensation

coefficient of KCl.

6.3 Temperature dependence and activation energy of conductivity

The temperature dependence of the electrical conductivity of electrolytes in which ions move as

NU
charged particles is generally explained by Eyring’s Transition State Theory. In other words, charged

particles (ions) respond to an electric field, and position exchange (rearrangement of materials) occurs

through the transition state, which is considered as an electrical conductivity phenomenon. Based on
MA
the transition state theory, it is assumed that the energy required to reach the transition state in the

movement of ions (meaning the rearrangement of matter) is the activation energy ΔEa, and that even

if charged particles exist in the system, only charged particles that have obtained this energy can

move. As a result, using the same approach as for general chemical reactions, the electrical
TED

conductivity σ is

σ = σ0 exp(‒ΔEa/RT) (8)

or

log σ = ‒ΔEa / 2.303RT + log σ0 (9)

CEP

where σ0 and R are a constant and a gas constant (8.314 kJ mol‒1 K‒1). Therefore, if a linear correlation

is obtained on a log σ vs. 1/T plot, commonly called Arrhenius plots. Here, the slope d log σ/d(1/T) is

d log σ/d(1/T) = ‒ΔEa/2.303R (10)

AC

and the activation energy ΔEa is calculated as follows:

 ΔEa = ‒2.303R[d log σ/d(1/T)] (11)

The activation energy is often used as a measure of the ease of ionic migration, and is applicable to

IPT
any shape of ionic conductors such as liquids, solids, and gels.

6.4 Nonlinear Arrhenius plots and VTF equation

SCR
When the temperature dependence of electrical conductivity is measured, a convex upper

temperature dependence is often observed. As a typical example, Fig. 9 shows the electrical

conductivity of an aqueous solution of Ca(NO3)2 measured by Angell et al. over a wide concentration

range.64 When the temperature dependence has a nonlinear response in a log σ vs. 1/T plot as in this

example, an empirical formula called the VTF formula

NU
W = A exp[‒B/(T ‒ T0)] (12)
MA
is often applied, where W is a number indicating the possibility of rearrangement of the material (e.g.,

electrical conductivity, fluidity, diffusivity, etc., indicating flow), A is the pre-exponential term

constant, and B is a constant. This equation was derived by Vogel65 and Fulcher66 from measurements

of the temperature dependence of the viscosity constant of glass, and was applied to a wide range of
TED

glassy materials by Tammann67 and thus became known as the VTF equation. Since then, Angell has

systematized not only viscous flow but also various transport properties of electrolyte, such as ionic

conduction in electric field and ionic diffusion in concentration gradient.

For the VTF equation for electrical conductivity proposed by Adam and Gibbs, Angell derived
CEP

from free volume theory.68 The VTF equation is indicated that

W = AT‒1/2 exp[‒B/(T ‒ T0)] (13)

of which pre-exponential term is proportional to T‒1/2.69‒73

AC

The derivation of the pre-exponential term AT ‒1/2 by free volume theory was performed in a system

with molten salts by Angell et al.73,74. After the derivation of this equation, the comparison with the
experimental data was done using the equation derived by Adams et al. as described in Appendix B.

The term “1/2” attached to the pre-exponential term has a less significant effect on the equivalent

conductivity.

IPT
The application of transition state theory to the interpretation of ionic conduction assumes that

given the energy required to activate a charged particle in an electric field, there is always an exchange

of ionic positions that contributes to ionic conduction. Thus, in the situation described by the

SCR
Arrhenius equation, is the following conditions must be satisfied.

- Ion migration is not disturbed by other dissolved species.

- The law of independent ion transfer at infinite dilution (Kohlrausch) holds.

- Debye-Hückel theory (point charge transfer in a continuous medium) is assumed.

These conditions are unlikely to be satisfied because charged particles in reality have finite sizes

the VTF equation to hold in electrolyte solute. NU

and the interactions exist between charged particles in ion transfer. Therefore, it is more common for

When the VTF equation is applied, the apparent activation energy can have a variety of meanings.
MA
If one were to adapt transition state theory and take only the transition energies, it would be strictly

arithmetic to go back and find the value of z*Δμ in Eq. 24 in Appendix B.68 However, if the

phenomenon of ionic conduction is not only caused by the movement of individual ions, but also by

the coordination of other dissolved species (ions, solvents, molecular chains in case of polymers,
TED

etc...), the calculation of transition energy based on the VTF equation, of which gradient varies with

temperature, should take into account the molecular motion and entropy effect of the system. From

this point of view, assuming that the linearity is maintained in the microscopic temperature range, it

is not meaningless to treat the differential coefficient of the VTF equation at each temperature (the
CEP

slope in the VTF plot) as the apparent activation energy. Calculating the differential coefficients of

Eqs. 12 and 13, they are expressed as follows:

from Eq. 12,

d ln W(T) ΔEw(T) T
= = ‒B
AC

(14)
d (1/T) R T ‒ T0

and from Eq. 13,

 d ln W(T) ΔEw(T) 1 T
= =‒ T‒ B (15)
d (1/T) R 2 T ‒ T0

IPT
On the other hand, the discussion of various parameters in the VTF equation is tied to the

correlation with the structure of the electrolyte. In particular, the discussion of T0 has been applied to

various ionic conductors with random structures, such as concentrated solutions,74 ionic liquids,75,76

SCR
molten salts77 and polymer gels,78 in addition to the application to glass materials, which were the

origin of the introduction of VTF, and the correlation with structural properties is often made based

on this interpretation. The interpretation of the results is often correlated with the structural properties.

In addition to cases where the temperature dependence follows either the Arrhenius or VTF

NU
equation, which is a continuous function within the measurement temperature range, there are also

cases where the temperature change is abrupt and discontinuous based on phase change (melting point,

transition point) or structural change (change in coordination number to ions, etc.) in the electrolyte
MA
solution. If the activation energy is calculated based on the gradient of the plot across these abrupt

changes, an extremely large activation energy value is obtained. Since these abrupt changes are due

to factors other than migration phenomena in the solution, the gradient should not be considered as

the activation energy. At the very least, it is necessary to confirm the cause of the abrupt change in
TED

electrical conductivity, and to calculate the activation energy only in the temperature region where

the stability of the system with temperature is maintained.18,24,25

6.5 Choice between log(σT) and log σ on the vertical axis of an “Arrhenius plot” of ionic

conductivity
CEP

The temperature dependence of electrical conductivity is often plotted over a wide temperature

range, and as noted in 6.3 and 6.4, the temperature dependence is shown in semi-logarithmic plots.

The vertical axis is denoted here as log σ, and on the basis of the Arrhenius equation derived from the

transition state theory in ionic conduction, the plots show that a linear correlation can be obtained
AC

between log σ and the reciprocal of temperature, 1/T.

On the other hand, based on the Nernst-Einstein equation, the relationship between the

electrophoretic force (ionic mobility, u) acting on ions in an electric field and the driving force
(diffusion coefficient, D) due to the concentration gradient in ion migration in an ion-conducting

medium is expressed by the following formula,79

IPT
uRT ukT
D= = (16)
|z|F |z|e

where k = R/NA, e = F/NA, and substituting λ≡ |z|Fu into this equation yields the following relationship,

SCR
uRT λRT
D= = (17)
|z|F z2F2

By the way, based on the transition state theory and free volume theory described in 6.3, as shown in

Appendix B, all of the physical properties related to mass transfer, such as electrical conductivity,

NU
fluidity, and diffusion constant, show a temperature dependence based on the Arrhenius equation.

However, if it is assumed that the electrical conductivity is based on diffusion phenomena in a

material placed in an electric field based on the Nernst-Einstein equation, the temperature dependence
MA
of electrical conductivity is to be derived based on Eq.16 as follows,

–ΔEa
D = D0 exp (18)
RT

Then,
TED

ΔEa
logD = log D0 – (19)
2.303RT

substituting Eq.17 into Eq. 19,

ΔEa
logD = log(RλT/ z2F2) = log D0 –
CEP

(20)
2.303RT

ΔEa
log λT = log(D0 z2F2/R) – (21)
2.303RT

If it is assumed that only specific (single) ion contributes to conductivity at certain concentration, i.e.,
AC

λ= σ/C, the Arrhenius-type temperature dependence is shown to be satisfied by setting log σT on the

vertical axis. This difference is reflected as a slight difference when the same data is illustrated.

Temperature dependence of the calculated electrical conductivities of a material following the

equation (a) log σ =log A – ΔEa/2.303RT and (b) log(σT) = log A – ΔEa/2.303RT for Arrhenius plots

in Fig. 10. In the case of the electrical conductivity σ = 1 S m–1 at the measured temperature T =

500 K and the activation energy is ΔEa = 38.29 kJ mol–1, adjusted electrical conductivity is 10–6 S m–

IPT
1
at 200 K for equation (a). Using equation (b), the value of σ should be compensated by the

temperature change and a little more than the value for equation (a).

In considering the temperature dependence, the plots based on the transition state theory or the free

SCR
volume theory, or based on the diffusion behavior are chosen. This is not a question of which to

choose, but rather a necessary choice in the interpretation of the data obtained. For this reason, “log

σT” is used when specific ions are moving or in solids such as crystals and semiconductors where the

path of movement is fixed, and “log σ” is used when multiple ions are moving along random paths.

7. Summary
NU
In a scientific purpose, the study of electrical conductivity and temperature dependence is very
MA
useful in discussing the physical properties of other dissolved species in solution, such as viscosity

and diffusion coefficient, and since these physical properties are very sensitive to changes in these

properties and vary greatly, more accurate and precise measurements are necessary.

The contents described in this paper can be understood by constructing a measurement system
TED

using a self-made cell, impedance meter, LCR meter, etc., and understanding the characteristics of

each device and verifying its operation. On the other hand, commercially available electrical

conductivity meters, whose primary purpose is to be used for environmental analysis, often have an

interface system that prioritizes convenience, such as a temperature compensation function and other
CEP

measurement modes compliant with industrial standards. Therefore, it is necessary to carefully check

the operational status of the device to ensure that there is no discrepancy between theory and practice,

as described in this paper.

No matter how much the accuracy and simplicity of the measurement device improves, the last

remaining issue is to understand the liquid entanglement and contact around the cell, especially at the
AC

interface between the electrode and the sample, and the polarization phenomena that occur there.

Even though the AC method is now commonly used, care must be taken in handling the sample.

Conductivity measurement should be performed in a condition that does not affect the composition
as much as possible, contrary to CV measurement that traces electrochemical reactions, and it is

necessary to consider how to remove the influence of polarization while obtaining accurate

measurement values.

IPT
Appendix A. Temperature Compensation of Electrical Conductivity

SCR
In ISO7888:1985 and JIS K 0130-2008, the temperature correction of the electrical conductivity of

an electrolyte solution is treated as follows, assuming that the value of the electrical conductivity

depends linearly on temperature.

The reference temperature for the temperature correction is 25 °C. If the electrical conductivity at

that temperature is κ25 and the electrical conductivity at the measurement temperature at T is κT, the

αθ,25 =
NU
temperature compensation coefficients αθ,25 are calculated by the following equation.

κθ – κ25
(22)
MA
κ25(θ – 25)

Using the coefficients obtained here, the electrical conductivity κ of the sample is calculated from the

electrical conductivity κθ at any temperature θ using the following equation.

TED

κ = κθ/αθ,25 (23)

As can be seen from this relationship, the temperature dependence of the electrical conductivity varies

linearly and is based on the relationship calculated from the temperature coefficient. Since the
CEP

temperature coefficient of a sample is not linear from a point of view in physicochemical properties,

and the differences are very large depending on the substance and its concentration, if the electrical

conductivity needs to be measured in scientific purpose, the temperature compensation function of

the electrical conductivity meter or all the input temperature coefficient values must be made
AC

ineffective.

The need to introduce a temperature coefficient is due to the high need for temperature

compensation when developing conductivity meters with excellent outdoor portability, such as for
environmental measurements. In other words, in conductivity measurements of electrolytes

containing trace amounts of ions, such as the measurement of ionic quantities in both seawater and

freshwater, it is not necessary to discuss the temperature coefficient so strictly, and in many cases,

IPT
emphasis is placed on concentration detection. Therefore, the priority is to calculate the temperature

coefficient of typical electrolytes in advance, and to use a conductivity cell with a temperature

compensation function to easily measure and compare electrical conductivity outdoors.

SCR
Appendix B. Derivation Process of the Equation Proposed by Adam and Gibbs in Ref. 68

Adam and Gibbs68 assumed that in highly concentrated electrolyte solutions with low fluidity, ion

transfer is due to the cooperative movement of all particles in a locally independent microscopic space

NU
in the liquid from a statistical point of view. The Arrhenius-type temperature dependence of the

transition state theory induced by the statistical interpretation is expressed by the following Eq. 24 in

terms of the probability function W for the rearrangement to occur.

MA
W = A exp(–z*Δμ/kT) (24)

where

z*: Size of the smallest region where cooperative migration occurs (number of particles in the
TED

region)

z*Δμ: Potential energy required for rearrangement of particles in the region

k: Boltzmann constant
CEP

By the way, the rearrangement entropy per mole Sc in the macroscopic range is the sum of the product

of the entropy sc* in the smallest region where cooperative rearrangement occurs and the number of

smallest regions in the system from a statistical mechanics viewpoint. Therefore, since all phenomena

in the minimum region can be assumed to be identical, the following equation is obtained using the

Avogadro constant, NA,

AC

sc*/z* = Sc/NA (25)

Substituting Eq. 25 into Eq. 24, respectively, we obtain

 W = A exp(–sc*NAΔμ/kT Sc) (26)

= A exp(–C/T Sc) (27)

IPT
where C = sc*NAΔμ/k (constant), where sc* and Δμ are statistical constants independent of

temperature.

On the other hand, using the entropy of the macroscopic system, the heat required for

SCR
rearrangement in the system is expressed as the integral of the constant pressure molar specific heat,

ΔCp, integrated over a given temperature range, and thus,

∫
T
Sc = (ΔCp/T)dT (28)
T0

where T0 is the theoretical glass transition temperature at which the entropy with respect to the particle

NU
array of the entire system is zero. According to the experimental results by Angell et al., the value of

ΔCp inversely proportional to temperature in such systems.64 Therefore, substituting the integral from

Eq. 28 into equation Eq. 27, we obtain,

MA
W = A exp[–C′T/(T – T0)] (29)

= A exp[–B/(T – T0)] (30)

TED

Where, C′ and B are constants. The assumption used to derive this equation is that ion transfer occurs

cooperatively in a medium in a locally independent microscopic space. The temperature dependence

of the electrical conductivity of a material with this assumption does not show a linear Arrhenius plot.

A derivation of the equation for the same reason has been calculated by Angell et al. from Bond-
CEP

Lattice theory.80 The temperature dependence of the various physical properties obtained from the

derivation of this probability function are expressed as follows.

Ionic conductivity λ = λ0 exp[–B/(T – T0)] (31)

Fluidity φ = φ0 exp[–B/(T – T0)] (32)

AC

Ionic diffusivity Di = Di0 exp[–B/(T – T0)] (33)

References

1. M. Mizuhata, Denki Kagaku, 86, 42 (2018). [in Japanese]

2. H. M. Goodwin, Harper’s Scientific Memoirs (Ed. J. S. Ames), VII. The Fundamental Laws of

IPT
Electrolytic Conduction, Harper & Brothers Publishers (1899).

3. F. Kohlrausch, L. Holborn, and H. Diesselhorst, Wied. Ann., 300, 417 (1898).

4. P. Walden, H. Ulich, and E. J. Birr, Z. Phys. Chem., 131, 21 (1927).

SCR
5. P. Walden, Bull. Acad. Sci., (St. Perterburg), 6, 405 (1914).

6. P. Walden, H. Ulich, and E. J. Birr, Z. Phys. Chem., 130, 495 (1927).

7. P. Walden, H. Ulich, and E. J. Birr, Z. Phys. Chem., 131, 1 (1927).

8. P. Walden and E. J. Birr, Z. Phys. Chem., 160A, 45 (1932).

9. P. Walden and E. J. Birr, Z. Phys. Chem., 160A, 57 (1932).

10. P. Walden, Z. Phys. Chem., 157A, 389 (1931).

NU
11. C. A. Angell, J. Phys. Chem., 70, 3988 (1966).

12. W. Xu, E. I. Cooper, and C. A. Angell, J. Phys. Chem. B, 107, 6170 (2003).
MA
13. H. Tokuda, K. Hayamizu, K. Ishii, S. Tsuzuki, M. A. B. H. Susan, and M. Watanabe, J. Phys.

Chem. B, 109, 6103 (2005).

14. H. Tokuda, S. Tsuzuki, M. A. B. H. Susan, K. Hayamizu, and M. Watanabe, J. Phys. Chem.

B, 110, 19593 (2006).

TED

15. G. E. Archie, Trans AIME, 146, 54 (1942).

16. S. Deki, M. Mizuhata, S. Nakamura, K. Nakamura, A. Kajinami, and Y. Kanaji, J. Electrochem.

Soc., 139, 996 (1992).

17. S. Deki, M. Mizuhata, K. Nakamura, A. Kajinami, and Y. Kanaji, J. Electrochem. Soc., 139,
CEP

1544 (1992).

18. S. Deki, S. Nakamura, K. Nakamura, A. Kajinami, and M. Mizuhata, J. Chem. Soc., Faraday

Trans., 89, 3811 (1993).

19. M. Mizuhata, Y. Sumihiro, and S. Deki, Phys. Chem. Chem. Phys., 6, 1944 (2004).

20. M. Mizuhata, F. Ito, and S. Deki, J. Power Sources, 146, 365 (2005).
AC

21. M. Mizuhata, Y. Ito, K. Yaso, and S. Deki, Electrochemistry, 73, 603 (2005).

22. M. Mizuhata, K. Yaso, Y. Ito, and S. Deki, Electrochemistry, 73, 606 (2005).

23. Y. Suzuki, H. Maki, M. Matsui, and M. Mizuhata, Electrochemistry, 87, 294 (2019).
24. M. Mizuhata, A. B. Béléké, H. Watanabe, Y. Harada, and S. Deki, Electrochim. Acta, 53, 71

(2007).

25. M. Mizuhata, H. Kubo, Y. Ichikawa, H. Maki, and M. Matsui, J. Electrochem. Soc., 168,

IPT
046516 (2021).

26. A. J. Bhattacharyya and J. Maier, Adv. Mater., 16, 811 (2004).

27. C. Pfaffenhuber, M. Gobel, J. Popovic, and J. Maier, Phys. Chem. Chem. Phys., 15, 18318

SCR
(2013).

28. T. Fukutsuka, K. Koyamada, S. Maruyama, K. Miyazaki, and T. Abe, Electrochim. Acta, 199,

380 (2016).

29. M. Mizuhata and K. Amezawa, Denki Kagaku (presently Electrochemistry), 74, 417 (2006). [in

Japanese]

NU
30. M. Mizuhata, J. Adhes. Soc. Japan, 44, 401 (2008).

31. M. Mizuhata, Denki Kagaku (presently Electrochemistry), 76, 831 (2008).

32. C. A. Kraus and H. C. Parker, J. Am. Chem. Soc., 44, 2422 (1922).
MA
33. H. C. Parker and E. W. Parker, J. Am. Chem. Soc., 46, 312 (1924).

34. H. Prestion-Thomas, Metrologia, 27, 3 (1990).

35. K. W. Pratt, W. F. Koch, Y. C. Wu, and P. A. Berzansky, Pure Appl. Chem., 73, 1783 (2001).

36. JIS K 0130–2008, p. 1. [in Japanese]

TED

37. T. Asakai, AIST Bull. Metrology, 9, 135 (2014). [in Japanese]

38. T. Asakai, I. Maksimov, S. Onuma, T. Suzuki, T. Miura, and A. Hioki, Accreditation Qual.

Assur., 22, 73 (2017).

39. ISO 7888:1985, https://www.iso.org/standard/14838.html

CEP

40. JIS K 0102–2021, p.28-29 [in Japanese]

41. JIS K 8121–2007, p. 5. [in Japanese]

42. ISO 6353-3:1987, https://www.iso.org/standard/12659.html

43. W. L. F. Armarego, Purification of Laboratory Chemicals, 8th Ed., Butterworth-Heinemann, p.

31 (2017).
AC

44. JIS K 0130–2008, p. 5. [in Japanese]

45. T. Shedlovsky, J. Am. Chem. Soc., 54, 1411 (1932).

46. G. Jones and B. C. Bradshaw, J. Am. Chem. Soc., 55, 1780 (1933).
47. E. Juhász and K. N. Marsh, Pure Appl. Chem., 53, 1841 (1981).

48. ISO/IEC 17025 Testing and calibration laboratories, https://www.iso.org/ISO-IEC-17025-

testing-and-calibration-laboratories.html

IPT
49. ISO 17025 Conductivity Calibration Standards, https://www.traceable.com/traceable-

products/conductivity-ph/iso-17025-conductivity-calibration-standards.html

50. NMIJ certified reference material (CRM), https://unit.aist.go.jp/qualmanmet/refmate/ [in

SCR
Japanese]

51. ISO 17034:2016, https://www.iso.org/standard/29357.html

52. G. Jones and R. C. Josephs, J. Am. Chem. Soc., 50, 1049 (1928).

53. G. Jones and G. M. Bollinger, J. Am. Chem. Soc., 51, 2407 (1929).

54. G. Jones and G. M. Bollinger, J. Am. Chem. Soc., 53, 411 (1931).

NU
55. G. Jones and G. M. Bollinger, J. Am. Chem. Soc., 53, 1207 (1931).

56. G. Jones and S. M. Christian, J. Am. Chem. Soc., 57, 272 (1935).

57. G. Jones and D. M. Bollinger, J. Am. Chem. Soc., 57, 280 (1935).
MA
58. G. Jones, K. J. Mysels, and W. Juda, J. Am. Chem. Soc., 62, 2919 (1940).

59. G. Jones and M. J. Prendergast, J. Am. Chem. Soc., 59, 731 (1937).

60. P. Debye and H. Halkenhagen, Phys. Z., 29, 121 (1928).

61. P. Debye and H. Halkenhagen, Phys. Z., 29, 401 (1928).

TED

62. K. Ibuki and M. Nakahara, J. Chem. Phys., 92, 7323 (1990).

63. A. Chandra and B. Bagchi, J. Chem. Phys., 112, 1876 (2000).

64. C. A. Angell and R. D. Bressel, J. Phys. Chem., 76, 3244 (1972).

65. H. Vogel, Phys. Z., 22, 645 (1921).

CEP

66. G. S. Fulcher, J. Am. Ceram. Soc., 8, 339 (1925).

67. G. Tammann and W. Hasse, Z. Anorg. Allg. Chem., 156, 245 (1926).

68. G. Adam and J. H. Gibbs, J. Chem. Phys., 43, 139 (1965).

69. C. A. Angell, J. Phys. Chem., 68, 218 (1964).

70. C. A. Angell, J. Phys. Chem., 68, 1917 (1964).

AC

71. C. A. Angell, J. Phys. Chem., 70, 2793 (1966).

72. C. A. Angell, J. Phys. Chem., 70, 3988 (1966).

73. C. A. Angell, J. Chem. Phys., 46, 4673 (1967).

74. E. Rhodes, Water and Aqueous Solutions —Structure, Thermodynamics, and Transport

Process— (Ed. R. A. Horne), Fused Salts as Liquids, p. 175 (1972).

75. M. A. B. H. Susan, A. Noda, and M. Watanabe, Electrochemical Aspects of Ionic Liquids (Ed.

IPT
H. Ohno), Diffusion in Ionic Liquids and Correlation with Ionic Transport Behavior, p. 55

(2005).

76. H. Ohno, M. Yoshizawa, and T. Mizumo, Electrochemical Aspects of Ionic Liquids (Ed. H.

SCR
Ohno), Ionic Conductivity, p. 75 (2005).

77. M. Spiro and F. King, Ionic Liquids (Eds. D. Inman and D. G. Lovering), Transport Properties

in Concentrated Aqueous Electrolyte Solutions, p. 57 (1981).

78. D. Bamford, G. Dlubek, A. Reiche, M. A. Alam, W. Meyer, P. Galvosas, and F. Rittig, J. Chem.

Phys., 115, 7260 (2001).

NU
79. A. Nitzan, Chemical Dynamics in Condensed Phases: Relaxation, Transfer and Reactions in

Condensed Molecular Systems, Oxford Graduate Texts; Oxford University Press: Oxford, New

York, (2006).
MA
80. C. A. Angell and K. J. Rao, J. Chem. Phys., 57, 12 (1972).
TED
CEP
AC
Figure Captions
Figure 1. Jones’ Cell for electrical conductivity measurement. (a) Diagram of the AC conductivity
cell used for the absolute measurements: a; end sections (half cells), b; center extension tube, c;
flanges, d; Pt electrodes, e; O-rings. A plastic band (not shown) surrounds the outer circumference of

IPT
the mating flanges to hold them coaxially and a compression clamp (also not shown) presses on the
O-rings to hold the mating flanges together, and (b)Diagram of the DC conductivity cell used for the
absolute measurements: 1; current electrodes, 2; potential difference sensing electrodes, 3; Nylon tee
joints, 4; O-rings. Reprinted from Ref. 35, Copyright 2001 IUPAC.

SCR
Figure 2. Circuit diagram of a Kohlrausch bridge, R1: fixed resistor, R2, R3: variable resistors, G:
galvanometer.

Figure 3. Automatic balancing of linear AC bridge circuits.

NU
Figure 4. Typical structures of electrodes for an electrical conductivity measurement. (a) Simplified
type, (b) and (c) Insertion type, (d) For dilute solution, (e) For precise measurement, and (f) Flow
cells.
MA
Figure 5. Schematics of commercially available insertion-type conductivity cell with electric lines of
force generated during measurement. (a) Shield removed two-pole cell. An unshielded electrode
causes the lines of electric force to spread so much that they depend on the shape of the measurement
bath. (b) The two-pole cell with the shield. Since lines of electric force leak through the air holes in
the upper part of the cell a little, they are considered to have little effect. (c) A two-pole cell with
TED

identical poles on the top and bottom to prevent leakage of lines of electric force to the outside. (d) A
four-pole cell. The current and voltage probes are independent.

Figure 6. (a) Surface SEM image and (b) cross-sectional TEM image of a platinum electrode with
platinum black deposited by Pt plating from H2PtCl6 solution as described in Section 4.2.
CEP

Figure 7. (a) complex impedance plot and (b) corresponding Bode plot of a typical solution
resistance measurement. When polarization at the electrode interface does not occur at high
frequencies, the phase difference between the applied voltage and current disappears and the plot is
on the real axis. On the Bode diagram, the resistance is independent of frequency and shows a
constant value as shown in (b).
AC

Figure 8. Frequency dependence of absolute AC impedance |Z| for various electrolytes (Bode’s plots).
16452A in TMPATFSA (Tetramethylpropylammonium tetrafluoro-bisulfonamide) indicates a
Keysight (formerly Agilent Technology) cell.
Figure 9. Temperature dependence of electrical conductivity of aqueous Ca(NO3)2 solutions.
Concentration unit is % by weight. Reprinted from Ref. 64, Copyright 1972 American Chemical
Society.

IPT
Figure 10. Temperature dependence of the calculated electrical conductivities of a material following
the equation (a) log σ =log A ‒ ΔEa/2.303RT and (b) log(σT) = log A ‒ ΔEa/2.303RT for Arrhenius
plots, where the electrical conductivity σ = 1 S m‒1 at the measured temperature T = 500 K and the
activation energy is ΔEa = 38.29 kJ mol‒1 for which value, the electrical conductivity is adjusted
σ=10‒6 at 200 K for equation (a).

SCR
NU
MA
TED
CEP
AC
 IPT
SCR
NU
MA
TED
CEP

Figure 1. Jones’ Cell for electrical conductivity measurement. (a) Diagram of the AC conductivity
cell used for the absolute measurements: a; end sections (half cells), b; center extension tube, c;
flanges, d; Pt electrodes, e; O-rings. A plastic band (not shown) surrounds the outer circumference of
the mating flanges to hold them coaxially and a compression clamp (also not shown) presses on the
AC

O-rings to hold the mating flanges together, and (b)Diagram of the DC conductivity cell used for the
absolute measurements: 1; current electrodes, 2; potential difference sensing electrodes, 3; Nylon tee
joints, 4; O-rings. Reprinted from Ref. 35, Copyright 2001 IUPAC.
 Variable
C condenser
Sample
Rs

IPT
G

R2 R3

SCR
Power source (Oscillator)

NU
Figure 2. Circuit diagram of a Kohlrausch bridge, R1: fixed resistor, R2, R3: variable resistors, G:
galvanometer.
MA
TED
CEP
AC
 Current
measurements

IPT
Shunt I
resistors

I
/V Conductance
2-pole 4-pole

SCR
electrode electrode
±200 mV

V
Potential
measurements

NU
Current
source
Reference potential
Source Er
MA
Figure 3. Automatic balancing of linear AC bridge circuits.
TED
CEP
AC
 (a) (b) (c)

IPT
SCR
(d) (e) (f)

NU
MA
Figure 4. Typical structures of electrodes for an electrical conductivity measurement. (a) Simplified
type, (b) and (c) Insertion type, (d) For dilute solution, (e) For precise measurement, and (f) Flow
cells.
TED
CEP
AC
 (a) (b) (c) (d)

IPT
SCR
-

+
i V

NU -
MA
Figure 5. Schematics of commercially available insertion-type conductivity cell with electric lines of
force generated during measurement. (a) Shield removed two-pole cell. An unshielded electrode
causes the lines of electric force to spread so much that they depend on the shape of the measurement
bath. (b) The two-pole cell with the shield. Since lines of electric force leak through the air holes in
the upper part of the cell a little, they are considered to have little effect. (c) A two-pole cell with
TED

identical poles on the top and bottom to prevent leakage of lines of electric force to the outside. (d) A
four-pole cell. The current and voltage probes are independent.
CEP
AC
 (a)

IPT
SCR
500 nm

(b)
NU
MA
TED

100 nm
CEP

Figure 6. (a) Surface SEM image and (b) cross-sectional TEM image of a platinum electrode with
platinum black deposited by Pt plating from H2PtCl6 solution as described in Section 4.2.
AC
 ‒Z” Impedance caused by diffusion

IPT
Resistance
of solution

SCR
R Z’

NU
MA
TED

log f

Figure 7. (a) complex impedance plot and (b) corresponding Bode plot of a typical solution
resistance measurement. When polarization at the electrode interface does not occur at high
frequencies, the phase difference between the applied voltage and current disappears and the plot is
CEP

on the real axis. On the Bode diagram, the resistance is independent of frequency and shows a
constant value as shown in (b).
AC
 IPT
SCR
NU
MA
Figure 8. Frequency dependence of absolute AC impedance |Z| for various electrolytes (Bode’s plots).
16452A in TMPATFSA (Tetramethylpropylammonium tetrafluoro-bisulfonamide) indicates a
Keysight (formerly Agilent Technology) cell.
TED
CEP
AC
 IPT
SCR
NU
MA
TED
CEP

Figure 9. Temperature dependence of electrical conductivity of aqueous Ca(NO3)2 solutions.

Concentration unit is % by weight. Reprinted from Ref. 64, Copyright 1972 American Chemical
Society.
AC
 IPT
SCR
(b)

(a)

NU
MA
Figure 10. Temperature dependence of the calculated electrical conductivities of a material following
TED

the equation (a) log σ =log A ‒ ΔEa/2.303RT and (b) log(σT) = log A ‒ ΔEa/2.303RT for Arrhenius
plots, where the electrical conductivity σ = 1 S m‒1 at the measured temperature T = 500 K and the
activation energy is ΔEa = 38.29 kJ mol‒1 for which value, the electrical conductivity is adjusted
σ=10‒6 at 200 K for equation (a).
CEP
AC
Table 1. The value of specific electrical conductivity κ of KCl standard solution newly established
by IUPAC based on the temperature indication of ITS-90.35

IPT
t/°C 0.01 mol/kg KCl 0.1 mol/kg KCl 1 mol/kg KCl
(104 κ/S m−1)
0 772.92 ± 0.23 7116.85 ± 2.85 63488 ± 25
18 1219.93 ± 0.37 11140.6 ± 4.5 —

SCR
20 1273.03 ± 0.38 11615.9 ± 4.6 99170 ± 40
25 1408.23 ± 0.42 12824.6 ± 5.1 108620 ± 43
30 1546.63 ± 0.46 14059.2 ± 5.6 118240 ± 47

NU
MA
TED
CEP
AC
 Table 2. Comparison of 2-pole and 4-pole methods.

2-pole method 4-pole method

IPT
Schematic
of equivalent
circuits

- Easy maintenance - The current-potential response is

SCR
- Inexpensive because of their linear in any conductivity region.
simple structure. - The conductivity regions in the
- Suitable for use with highly calibration and conductivity
Features and viscous samples and colloidal measurements can be different.
advantages solutions. - Ideal for samples with high
conductivity.
- Applicable for samples with low
conductivity if the effect of stray

-
NU
The electrodes must be placed in
the center of the container to avoid
electric field effects.
The effect of polarization cannot
-
capacitance is compensated.
Not suitable for small samples
because all four electrodes must be
in contact (immersed) with the
sample. (3-4 cm distance between
MA
be ignored in the case of samples electrodes is required.)
Disadvantage with high conductivity. - Not suitable for flow method.
- It is desirable to use two types of
samples with conductivities that
differ by about two orders of
magnitude when calibrating
conductivity.
TED

With Pt black: applicable in wide

Measurement With Pt black: >3 mS cm−1
range of conductivity
range Without Pt black: >1 mS cm−1
Without Pt black: >100 mS cm−1
CEP
AC