Journal of Power Sources 329 (2016) 57e71

Contents lists available at ScienceDirect

Journal of Power Sources

journal homepage: www.elsevier.com/locate/jpowsour

Review Article

Absorbency and conductivity of quasi-solid-state polymer electrolytes

for dye-sensitized solar cells: A characterization review
Ahmad Azmin Mohamad
School of Materials and Mineral Resources Engineering, Universiti Sains Malaysia, 14300, Nibong Tebal, Pulau Pinang, Malaysia

h i g h l i g h t s

Recent progress in quasi-solid state electrolytes for dye-sensitized solar cells.

Characterization of quasi-solid state electrolytes.

Characterizations of absorbency and conductivity.

Highlighted main outcomes and objectives of each characterizations.

a r t i c l e i n f o a b s t r a c t

Article history: The application of quasi-solid state electrolytes for dye-sensitized solar cells opens up an interesting
Received 7 April 2016 research eld to explore, which is evident from the increasing amount of publications on this topic. Since
Received in revised form 2010, signicant progress has been made with new and more complicated quasi-solid-states materials
24 July 2016
being produced. The optimization of new materials requires specic characterizations. This review
Accepted 11 August 2016
presents a comprehensive overview and recent progress of characterization methods for studying quasi-
solid-state electrolytes. Emphasis is then placed on the absorbency and conductivity characterizations.
Each characterization will be reviewed according to the objective, experimental set-up, summary of
Keywords:
Quasi-solid state electrolytes
important outcomes, and a few case studies worth discussing. Finally, strategies for future character-
Gel polymer electrolytes izations and developments are described.
Polymer electrolytes 2016 Elsevier B.V. All rights reserved.
Dye-sensitized solar cells
Characterizations

1. Introduction
Freely available solar energy is one of the major energy forms
that can be harnessed to fulll future energy needs. The harvesting
of solar energy has been introduced in many different technologies.
The solar cell is one such technology. Solar cells are typically named
after the semiconducting material they are made of, and the
number of layers of light-absorbing material, including bulk het-
erojunction solar cells, depleted heterojunction solar cells, crys-
talline silicon, thin lm, inorganic and organic.
Since the rst report in 1991 by Oregan and Gratzel [1], organic
dye-sensitized solar cells (DSSCs) are highlighted as promising
devices to convert solar energy to electrical energy, due to a pro-
spected low cost, environmentally friendly operation and relatively
high efciency. Inside the DSSC components, the electrodes (mainly
E-mail address: aam@usm.my.
the photoanode) are the limiting factors in terms of overall ca-
pacity, i.e. efciency. The electrolyte, on the other hand is not only
used to separate the electrodes or to complete the circuit, but acts
as a crucial medium to transfer electrons from the counter elec-
trode to the oxidized dye. If the properties of this redox electrolyte
are optimized by improving absorbency and ionic conductivity, the
efciency of DSSC will increase too.
Electrolytes used in DSSCs can generally can be classied into
three main categories [2]; liquid, quasi-solid-state (QSS) electro-
lytes and solid-state electrolytes. The usage of liquid electrolytes
can cause leaking of the electrolyte, volatilization of the electrolyte,
photodegradation of dye and corrosion of the counter electrode.
Meanwhile, solid-state electrolytes cause a contact problem at the
electrodes. To overcome these shortcomings, researchers have
proposed and attempted a technique where both liquid and solid
electrolytes are combined, to form a QSS electrolyte.
Although several recent review articles are available about
polymer electrolytes based on DSSCs components and cell

http://dx.doi.org/10.1016/j.jpowsour.2016.08.064
0378-7753/ 2016 Elsevier B.V. All rights reserved.
58 A.A. Mohamad / Journal of Power Sources 329 (2016) 57e71
performance [2e7], materials, preparation and properties of poly-
mer electrolytes [8], and solid polymer electrolytes [9], little
attention has been devoted to QSS as one of the main components
in DSSCs.
The rst application of gel electrolyte in a QSS-DSSC was re-
ported in 1995 [10]. Since then, QSS development has largely been
overshadowed by new electrode materials and the emphasis on the
power conversion efciency of DSSCs. To optimize the composition
of new materials or to produce commercially standard QSS, the
understanding of characterization for these purposes is very
important.
Since 2010, much work has been related to the characterization
of QSS electrolytes, chiey regarding materials. This makes them
very interesting and attractive, both from the point of view of
materials, techniques, basic science and as far as material charac-
terization is concerned. A bibliographic analysis using Web of Sci-
ence on topics Quasi Solid State, Gel, Electrolyte and Solar
Cells showed many publications (240 articles) (Fig. S1). The
number of articles is higher in Google Scholar (>8800 articles).
Based on these published papers, it is worth reviewing solely the
aspect of material characterization.
This review will focus upon the recent progress on the charac-
terization technique of choice for QSS and gel electrolyte evalua-
tion. Also, each characterization technique is followed by a review
of related examples. The properties of materials, effect of additive
and performance of modication are analyzed according to the
capability of the used technique. To cover all QSS and gel electro-
lytes characterizations would be beyond the scope of this review.
Therefore, this review provides selected characterizations in the
area of QSS and gel electrolytes for DSSCS over six years
(2010e2015). However, a few older papers are also cited to relate
the basic understanding and to act as a comparison.
This review is organized into two main characterizations for
QSS:
i. Absorbency, visual and liquid loading characterizations: the
unique and most important characterization for QSS. An
example includes imbibition kinetics.
ii. Electrochemical impedance spectroscopy (EIS) characteriza-
tions: characterization of ion transportation mechanism of
iodide/tri-iodide (I /I
3 ) redox couple in the QSS. Later will
discuss in detail as conductivity.
2. Components and mechanism of QSS-DSSCs
The standard conguration and operation principle for a QSS-
DSSC involves ve main components (Fig. 1a):
i. Anode: a transparent glass substrate covered with a trans-
parent conductive oxide layer, i.e. uorine-doped tin oxide
(FTO).
ii. Photocatalyst: a nanoporous oxide layer, i.e. titanium oxide
(TiO2) deposited on the anode to activate electronic
conduction.
iii. Dye: a monomolecule layer charge transfer dye, i.e. N3, N719
and Z709 are covalently bonded to the surface of the meso-
porous oxide layer to enhance light absorption. Components
of (i)e(iii) are also called photodanode electrode (PE).
iv. Electrolyte: a QSS polymer electrolyte lled with redox
mediator, i.e. I /I
3 couple for the recovery of dye and the
regeneration of the electrolyte itself during operation.
v. Cathode: a counter electrode (CE) made of a glass substrate
coated with a catalyst, i.e. platinum (Pt) to facilitate electron
collection.
The anode and cathode are connected through an external cir-
cuit, which allows photogenerated electrons to move from the PE to
the CE. The QSS can be considered as an electron acceptor (but it
does not allow electrons to pass through). All the reactions in QSS
involve ionic movement of the I /I
3 couple. The operating principle
of DSSC started as the light being transmitted through the CE (the
p). The redox electrolyte excites the dye (the n), thus allowing
electron transfer though the PE (the n) and out to the electrical
loop for application.
The subsequent process of oxidized dye to its original reduced
species, namely dye regeneration, is mediated via the redox QSS
electrolyte (Equations (1), (2) and (3)) in order to complete the
circuit loop [11].
I I2 4I
3 [1]
3I 2dye /I
3 2dye photoelectrode [2]
I  
3 2e 43I counter electrode [3]
3. What is a quasi-solid-state electrolyte?
A better understanding of the QSS polymer can be related to one
of the most common applications, which is used for urine ab-
sorption in baby diapers. The swollen gel holds the liquid in a solid,
rubbery state and prevents the liquid from leaking onto the skin
and clothing. Another common application is as a water-ingredient
storage in the agricultural industry. The granules of hydroponic gel
absorb the nutrient solution, again releasing this gradually over a
long period of time. This type of polymer is normally referred to as
super absorbent polymer (SAP) that can absorb and retain
extremely large amounts of a liquid relative to their own mass
(Fig. 1b). The properties of QSS range from swelling uid (gel) when
agitated to no-ow hard and tough matrix under steady-state
conditions (solid). By weight, QSS is mainly liquid, yet it behaves
as a solid, owing to a three-dimensional (3D) cross-linked matrix
within the liquid electrolyte. In this way, ions are dispersing within
a QSS.
Quasi solid-state is dened as semi solid electrolyte. It is
referring to a physical term as a system that lies along the state
between a solid and a liquid. The uniqueness of QSS is related to its
solid properties, such as the ability to support their own weight and
hold their shape. However, QSS also shares some properties of a
liquid. The high ionic conductivity and better interfacial contact are
among the important properties of liquid, but are also found in QSS
electrolytes. Some QSS are hard elastic and others are very soft,
turning to paste when force is applied. Nogueira et al. [12]
explained that the main root of a QSS polymer electrolyte is gel
polymer electrolyte (GPE). Thus, the denition of QSS is also rooted
in a GPE. Polymeric gel is a system that consists of a polymer
network swollen with a solvent. Owing to their unique hybrid
network structure, gels have both the cohesive properties of solids
and the diffusive transport properties of liquids. Gel electrolytes are
usually obtained by incorporating a large amount of a liquid plas-
ticizer and/or solvent (containing the desired ionic salts) into a
polymer matrix, giving rise to a stable gel with a polymer host
structure. When gelation occurs, a dilute or more viscous polymer
solution is converted into a system of innite viscosity: a gel.
The terms quasi-solid-state, quasi-solid-gel, quasi-solid gel-
polymer, polymer gel, jelly-like, apparently but not really solid-
state, gelation and quasi/gel are used interchangeably and
commonly found in the literature [13e21]. Generally, polymer
electrolytes consist of polymer, salt and solvent (solvent for
 A.A. Mohamad / Journal of Power Sources 329 (2016) 57e71 59

Fig. 1. (a) Schematic of a quasi-solid-state dye-sensitized solar cell and the electron transfer processes involved in energy conversion; (b) schematic of swelling of a dried polymer
network in water containing I/I 3 ions electrolyte to form QSS electrolyte (red box representing selected area for expansion); physical appearance of QSS (c) before added solution
and (d) expanded into loosely bound gel after immersed in solution.

polymer and salt). However, in terms of application in DSSC, QSS is
dened as polymeric or gel materials incorporating I /I
3 as a redox
couple to substitute the liquid electrolytes.
There are several mechanisms in the process of QSS swelling or
absorbency. For example, the process starts with a dried/shrink
cross-link network (Fig. 1c and d). When the polymer is immersed
in a water-based electrolyte, there will be a polymer-water inter-
action and polymer networks expand. The polymer backbone is
hydrophilic (i.e. water loving), then the hydration and the forma-
tion of hydrogen bonds take place. During the hydration process,
dissolved I /I
3 ions attract the polar water molecules and swell
together in QSS. The cross-links between the networks prevent the
solubility of polymer in water and also control the extent of
swelling or absorption by restricting innite swelling.
The literature on electrolytes can be roughly classied into three
main physical states: liquid, gel and solid. However, specic
60 A.A. Mohamad / Journal of Power Sources 329 (2016) 57e71
detailed classications of QSS are [2,8,12,22]:
i. Thermoplastic polymer electrolytes
- Mixture of polymer matrix with liquid electrolyte trans-
forming liquid state to gel state. Since the gelation process between
the polymer matrices and liquid electrolyte only involves a weak
interaction, its states are reversible depending on temperature.
These kinds of polymer electrolytes are generally known as GPE.
ii. Thermosetting polymer electrolytes
- Basically thermosetting polymer electrolytes are similar to
thermoplastic polymer electrolytes in terms of polymer electrolyte.
The difference is that the polymer of thermosetting forms a 3D
polymer network and hence wraps the liquid electrolytes inside.
Thermosetting polymer electrolytes are not reversible, due to the
cross-links. These kinds of polymer electrolytes are the most pop-
ular used by researchers for QSS.
iii. Composite polymer electrolytes
- The composite polymer electrolytes are thermoplastic or
thermosetting, incorporating inorganic nanoparticles (sometimes
called ceramic ller or metal oxide additive) to form QSS. The
presence of nanoparticles leads to a network structure, converting
the liquid mixture to a QSS with thixotropic properties. The
advantage of this kind of electrolyte lies in the fact that I /I
3 ions
are able to align on the nanoparticles network, thus leading to an
acceleration of the charge-transport dynamics.
iv. Ionic liquid electrolytes
- Design of QSS-based ionic liquid electrolytes exploit its prop-
erties as a salt in the liquid state. The ionic liquid is used as solvent
for QSS, but at the same time provides iodide for the system.
The driving force behind these extensive works on QSS is its
unique properties. QSS is able to produce higher conductivity
values rather than dry solid-state polymers. The conductivity of
the gel electrolytes can be expected to be very close to the corre-
sponding liquid electrolyte, which has led to their application in
many electrochemical solid-state ionic devices, not just DSSC but
high-energy density batteries, fuel cells, sensors and electro-
chemical display devices. The detailed uniqueness of QSS will not
be discussed here, but is simply listed as follows:
i. Shape form: matrix gel form (swelling) or in solid form (jelly-
like).
ii. Absorbency: ability to absorb, to retain and reducing the
leakage of liquid electrolyte.
iii. Wettability: electrodes/electrolyte interface similar to liquid.
iv. Conductivity: higher ambient ionic conductivity
>103 S cm1.
v. Stability: long-term thermostable, improving durability by
suppressing solvent evaporation and reducing electrode
degradation (compared to liquid electrolyte).
4. Absorbency
The absorbency of electrolyte (also known as liquid electrolyte
loading or swelling) is one of the main characterizations of QSS.
This characterization makes it unique compared to other electro-
lytes. The absorbency refers to the ability of polymeric material to
absorb electrolytes. This ability is a key factor affecting the ionic
conductivity of QSS. Consequently, the DSSC properties are also
depend strongly on this rst characterization.
For water-based electrolytes, the absorbing polymers are clas-
sied as hydrogels when cross-linked, absorbing aqueous solutions
through hydrogen bonding with water molecules. Meanwhile, the
higher degree of cross-linkers and the presence of ions in the so-
lution normally will impede the polymer ability to bond with the
water molecule. However, cross-linked and/or thermoset are used
to improve the form of irreversible gels.
In general, the absorbency of QSS characterization can be pre-
sented by:
i. Absorbency vs. solution or polymer concentration: to measure
the absorbency of the electrolyte at various concentrations and
different types of polymer.
ii. Absorbency vs. retention or time: to measure the time taken for
the polymer to be absorbed and to retain the solution.
Although plots of absorbency are presented by concentration
and retention, the effects of parameters will relate to time. Later,
time is used to calculate imbibition kinetics. Some SAP with 3D
networks and hydrophilic groups can absorb 20e100 times of the
liquid [18]. The uniqueness of SAP lies in its microporous network
structure. Owing to this structure and the hydrophilic group on it, a
large amount of organic solvent can be absorbed, which is barely
removed under some pressure or heat. An extensive review of
general methods to measure absorbency of SAP can be found
elsewhere [23] and possibly can be used to characterize QSS as well.
Absorbency is not complete without micro structure analysis,
i.e. by using scanning electron microscopy (SEM) characterization.
SEM is widely used since it has a large depth of eld, yielding a
characteristic three-dimensional (3D) appearance useful for un-
derstanding the surface morphology, porosity and network struc-
ture of a sample. These improved properties will help to absorb the
liquid electrolyte. Tang et al. [11] used the 3D appearance to
investigate the unique property of a poly (glycidyl acrylate) (PGA)
superabsorbent (Fig. S2). With the advantages of a honeycomb
microporous network structure and a hydrophilic group of PGA, a
large amount of organic solvent can be easily absorbed and stored.
To study porosity CE, Xiao et al. [24] used SEM as a selection tool
to nd the optimum columnar PEDOT electropolymerized onto the
nano-PEDOT thin lm and 3D PAA-PSS network. A pure sample of
nano-PEDOT was used as the control sample (Fig. S3). At a different
time of electropolymerization, a columnar PEDOT that continu-
ously grew was observed. They concluded that the best duration
time was selected at 1000 s, as the 3D PAA-PSS network was
properly lled up with the columnar PEDOT. This sample was used
for enhanced I 3 reduction and for storage of the liquid electrolyte in
the QSS-DSSC. The latter achieved the relatively high energy con-
version efciency of 6.35%.
The cross-section is benecial for QSS for comparison with the
surface layer (important for interface) and inside (important for
electrolyte absorption or storage). The cross-section technique
allow us to visualize the inside of the structure sample, such as a
spongy/porous nanostructure [25e30], a nanobowl structure [31],
and holes [32]. Some samples showed a spongy nanostructure on
the surface (Fig. S4), but not inside, as characterized by their cross-
sections [33]. Therefore, the cross-section method is a crucial
method for characterizing the porosity of QSS.
During the drying of porous QSS after synthesis, a porous
structure is formed inside the lm. However, some of the polymer
at the top tends to form a dense structure (solid skin) [32,34]. In
detail, Wang et al. [32] compared the top surface and cross-section
views. On the surface, the dried gel showed some typical
morphology of a skin (~1 mm), i.e. rough and continuously covering
 A.A. Mohamad / Journal of Power Sources 329 (2016) 57e71 61

the surface (Fig. S5). For the cross-sectional view, large holes Then, the liquid absorbency of the sample is measured accord-
(porous structures) were observed. In their study, holes of the ing to (Equation (4)) [35,36,40];
network were formed when the PVAc was dissolved in the liquid
electrolyte. Thus, the liquid electrolyte can transfer freely within   W2  W1
the polymer framework. Liquid absorbency g g1 [4]
W1

4.1. Absorbency with solution and polymers
In most QSS the polymer portion is usually less than 20% [35],
and the remainder is contributed signicantly by the solvent. As a
rst step in examining a QSS, it is usually worthwhile to have a look
at the physical expansion. For example, the expansion of the hybrid
poly (acrylic acid) (PAA)-poly (ethylene glycol) (PEG) polymer after
an uptake of liquid electrolyte can easily be noticed by the naked
eye (Fig. 2a). However, the most noteworthy change was the color
in a digital photograph. The color change has also been used to
indicate that the hybrid had fully absorbed the liquid electrolyte
[25,36e39].
Quantitative absorbency (also called liquid loading) is a
straightforward measurement [23,36,40]. Since the measurement
is based on weight, the sample can be powders, grains, granules,
blocks or thin lms. It should be noted that the shape of the
polymer is not a critical issue, but the concentration of electrolytes
is. A high concentration of electrolyte normally means less ab-
sorption into the polymer.
The swelling of a gel matrix in liquid electrolyte generally can be
divided into three processes [41]:
i. liquid electrolyte enters gel matrix.
ii. relaxation of the macro-molecular chains.
iii. stretching of the entire chains in the liquid electrolyte.
where W1 is the mass of dried sample (g), and W2 is the mass of
swollen gel-electrolyte (g).
Li et al. [36] reported the absorbency characterization (normally
combined with conductivity: Section 5.3) of the GPE in different
concentrations of LiI and KI (Fig. 2b). The liquid electrolyte used
consisted of I2 and LiI or KI in a mixed organic solvent of N-methyl-
2-pyrrolidone (NMP) and g-butyrolactone (GBL). Both iodide salts
had quite a similar inuence on the liquid electrolyte absorbency of
the gel produced. The liquid electrolyte absorbency increased,
achieving a maximum point and decreasing with the concentration
of LiI or KI.
Earlier Lan et al. [40] also used this technique to study the ab-
sorbency of thermosetting of PAA-PEG gel polymer electrolyte. The
gel was prepared by in-situ synthesized acetylacetone (acac)-pyri-
dine (Py)-I2 ionic conductors and the traditional method of soaking
PAA-PEG polymer matrix in liquid electrolyte to form the gel state.
The synthesized method shows higher liquid electrolyte absor-
bency (up to 17 g g1) than that of the traditional method (up to
11 g g1).
To characterize the effect of crosslinking of polymers on absor-
bency, Tang et al. [35] used PAA-gelatin and PAA-gelatin-
polyaniline (PANi) composite. The introduction of PANi to PAA/
gelatin polymer host results in a higher cross-linking density of the
composite, leading to a decreased liquid absorbency (16.45 g g1)
compared to PAA-gelatin (17.26 g g1). Higher cross-linking will

Fig. 2. Digital photograph of PAA-PEG (a) before and (b) after being soaked in liquid electrolyte (sample expanded and colored changed), adapted from Refs. [36]; (c) Dependence of
liquid electrolyte absorbency and ionic conductivity of polymer gel electrolyte on concentrations of alkali metal iodides (LiI and KI; lled square liquid electrolyte absorbency of
polymer gel electrolyte containing KI, lled circle liquid electrolyte absorbency of polymer gel electrolyte containing LiI), adapted from Refs. [36]; (d) Solvent retention ability of
PAA-PANi, EA-PAA-PANi and FEA-PAA-PANi gels. Reprinted with permission from Ref. [42] Copyright 2014 American Chemical Society.
62 A.A. Mohamad / Journal of Power Sources 329 (2016) 57e71
make the polymer matrix become rigid and unable to expand, even
though an enormous amount of electrolyte was used.
4.2. Absorbency with retention and time
Another important property associated with the electrolyte
absorbency of gel electrolytes is electrolyte retention ability or
gravimetric. The analysis is based on the percentage of solvent lost
over time. Good materials should retain electrolyte for a longer
time, even under pressure [36]. The electrolyte retention ability
depends on the density of cross-links formed in the gel, which act
as cages and prevents rapid evaporation of the electrolytes.
For this characterization, Kelkar et al. [42] used the swelled
hybrid gels that were taken out of the ionic liquid 1-ethyl-3-
methylimidazolium iodide (EMII), I2, KI and LiI in a mixed organic
solvent of NMP and acetonitrile electrolyte and stored in different
dry containers. Thereafter, their weights were measured over
specic time intervals as the solvent was evaporated. Later, the
percentage decrease in the weight of each gel was calculated and
plotted against time (Fig. 2c). Most interestingly, the result showed
that PAA-PANi hybrid gel had the poorest electrolyte retention
ability (but the highest solvent absorbency).
Another method that can be applied to study electrolyte
retention is absorbency under load testing, designed specically for
the effect of compression or expansion [18] (Fig. S6). Basically, the
characterization is used to determine the optimum volume that can
be absorbed by the QSS. To compare with electrolyte, water is also
used as control sample. The optimum volume will be used as
constant volume in characterizations such the equilibrium
swelling, expansion, and compressive stress-strain. The equilib-
rium swelling characterization is conducted to estimate the time
required for the gel to reach the equilibrium state.
Initially, the QSS grains are mostly sugar-like particles with
irregular shapes. When added with a solution, the gel started to
swell into loose jelly-like particles (soft and elastic) (Fig. S7). The
longer the gel stayed in the solution, the more it expanded. The
swelling of the electrolytes is lesser than that of the water. QSS with
electrolyte showed a gradual % of swelling. Unlike QSS with water,
the % of swelling signicantly increased during the rst 10 min and
then continuously increased with time (Fig. S8). However, regard-
less of the solution with which QSS, both samples reached the
equilibrium % of swelling at the same time within an average of
60 min, indicating that the osmotic pressure difference between
the gel and the solvent already became zero.
For the expansion and compression studies (Fig. S9), the QSS
immersed in electrolyte experienced lower strain compared with
those immersed in water. Considering that QSS can absorb higher
amounts of water, it expands and increases the strain value more
than the QSS with electrolyte. Meanwhile, the compressive stress-
strain curves, the strain amplitude increased with increasing
applied stress, and the strain amplitude for the QSS with electrolyte
is lower compared with that of QSS with water. The strain ampli-
tude is more likely to follow the trend of the percentage absorbency
of the QSS. Considering that the gel absorbed only a small amount
of electrolyte, the QSS only expanded slightly. Hence, it yielded a
smaller strain amplitude. On the other hand, the % absorbency of
the QSS with water is relatively high, and as a result, the QSS is able
to expand more, becoming a big bloated QSS. However, the gel is
relatively soft and thus experienced larger strain amplitude. Strain
amplitude can be used to interpret roughly the compressive
strength of the QSS. A larger strain amplitude means that the gel
experienced larger deformation and hence indicates that the
compressive strength of the gel is lower and vice-versa.
4.3. Imbibition kinetics
The absorbency measurement of liquid electrolytes can be
categorized as a basic characterization of QSS. To further under-
stand the mechanism of absorbency, imbibition kinetics is pre-
sented by absorbency vs. time [41] (or normally combined
absorbency and conductivity vs. time [28]) (Fig. 3). Imbibition ki-
netics is generally controlled by Flory theory, in which the driving
force of the liquid electrolyte is osmotic pressure across the gel
matrix. The imbibition process is accompanied by the swelling of
molecular chains in the liquid electrolyte. Imbibition equilibrium
can be obtained when the osmotic pressure is zero. The analysis of
imbibition kinetics will produce a transport number (n). n is ob-
tained by a linear best-t of the relation by Fickian theory (Equation
(5)):
Mt
kt n [5]
M
where Mt is the mass of the imbibed liquid electrolyte at time t, M
is the mass at equilibrium and k is a characteristic rate constant
relating to the properties of the polymer matrix.
By converting data from the plot of absorbency vs. time and
plotting log (Mt/M) vs. log (t), the n values are recorded; Fickian
behavior is interpreted as [30,37,41,43]:
i. n  0.5: a Fickian transport behavior in which the polymer
host relaxation is much faster than the diffusion.
ii. n 1: a non-Fickian mode of transport where liquid elec-
trolyte uptake is controlled by diffusion process (i.e.
swelling-controlled).
iii. 0.5 < n < 1: superposition of both phenomena (anomalous
transport) in which structural relaxation is comparable to
diffusion and loading of liquid electrolyte is controlled by
osmotic pressure.
Most reports give a value higher than 0.5, indicating that
anomalous mechanism mode and the loading of liquid electrolyte
are controlled by osmotic pressure. Examples of materials are the
microporous PAA-PEG [28], microporous PAA-
cetyltrimethylammonium bromide (CTAB)-graphene [29] and
PAA/glycerol (GR) and PAA/CTAB [26]. A few reports show n to be
less than 0.5 (Fickian transport mechanism), for example, PAA-PEG/
aniline and PAA-PEG/aniline-PANi [37].
Recently, an interesting report discussed the characterization of
osmatic pressure and capillary force mechanisms [30]. If the driving
factor for the liquid absorbency is only controlled by osmotic
pressure across the gel matrix, the absorbency will be limited. Then,
new materials were designed which can combine both mecha-
nisms to improve the efcacy of conducting gel electrolytes. By
imbibing PANi (or polypyrrole, PPy) and liquid electrolyte into a
microporous PAA-CTAB matrix, the driving forces by both osmotic
pressure and capillary force in the micropores improved the
loading of liquid electrolyte per unit volume, leading to further
enhancement in charge transfer and electrocatalytic activity.
However, the assumption of participation of capillary force is only
based on the existence of a capillary channel in the microporous
structure gel. No detail about quantitative osmotic pressure or
capillary force portion is available from this report.
5. Electrochemical impedance spectroscopy and conductivity
To complete the investigation of QSS, conductivity is critical. QSS
characterization, ionic conductivity in particular, is standard pro-
tocol prior to DSSC assembly. The highest conductive QSS will
 A.A. Mohamad / Journal of Power Sources 329 (2016) 57e71 63

Fig. 3. The time dependences of liquid electrolyte loading for pure PAAm and PAAm-7wt.% graphene (G) gel electrolytes (a) the rst plot is liquid electrolytes upload and (b) second
is logarithm plot for imbibition calculation, adapted from Refs. [41]; (c) Correlation between ionic conductivity, liquid electrolyte loading and imbibition kinetics of PAA-PEG in
PANi-containing liquid electrolyte, adapted from Ref. [28].

normally be used to fabricate DSSC. Among the available tech-
niques, EIS is considered the most popular and reliable character-
ization technique for the determination of the ionic properties of
QSS.
This method uses samples held between two symmetrical inert
blocking electrodes, i.e. stainless steel 316 (SS) [44,45], Au [46], Pt
[47,48], or synthesized CEs [49] (Fig. 4). This cell can also be used for
voltammetry or Tafel polarization characterization. The electrical
properties of QSS could be determined using an alternating current
(a.c.) method-based electrochemical study. The IS measurements
are carried out in a frequency range of 0.1e105 Hz [28,37],
0.01 Hze105 kHz [50] and 0.01 Hze106 kHz [49,51] with an AC
amplitude of 10 mV [49,50] and 20 V [28,37]. The specic designed
cell is also proposed under dark conditions [27,28,37].
5.1. Impedance plot
The simplest observation of EIS outcome is by using an
impedance plot or Nyquist plot. The latter provides information on
the electrochemical frequency behavior of the QSS system. The EIS
is a powerful technique in the characterization of impedance and
produces information such as:
i. Impedance spectra: to see spectra behavior by observing
semicircle and spike.
ii. Resistances: to measure ohmic series resistance (Rs) or bulk
resistance (Rb) taken from the onset of the rst semicircle and
the charge-transfer resistance (Rct). These parameters represent
the diffusion resistance of I /I
3 redox species.
Since QSS is a highly conductive medium for ionic conductivity,
the spike is usually recorded (Fig. 5a). Tan [52] studied the effect of
polymer concentration on the NaI/I2 electrolyte and all spectra are
represented by spikes. As the concentrations of polymer increase,
the spikes move to the right. The Rct for these types of spectra are in
the range of 40e200 U. Note that for correct plotting, the imaginary
64 A.A. Mohamad / Journal of Power Sources 329 (2016) 57e71

axis is plotted against a real axis quantity, and the scale on both
axes must be the same (i.e. 1 U: 1 scale quantity).
Normally when comparing the liquid to gel electrolyte, the spike
of the liquid is close to the origin as well as the gelled electrolyte.
However, Jung [51] shows a different nding: the resistance of the
PPy gel (0.20 U) was lower than the liquid (51.89 U) electrolytes
(Fig. 5b). This suggests that PPy gel electrolyte could effectively
improve the cell performance, even though the viscosity of PPy gel
electrolyte (1830 cP) is higher than that of liquid electrolyte
(10.94 cP).
Meanwhile, for QSS sandwiched by two asymmetric Pt glass
electrodes (Pt/gel electrolyte/Pt), the charge-transfer phenomena
at the electrode/gel electrolyte interface showed semicircular
Nyquist spectra [25,26,28,29,37,47,53e55] (Fig. 5c). As the Pt
counter electrode was used in assembling the symmetric cells, all
the Rs started at the same value, 18 U. The value of Rct was used to
show the decreases of composite in the order of PAA-CTAB/
graphene < PAA-CTAB/graphene oxide < PAA-CTAB/nano-graphite.
This order also showed an inverse order of electrocatalytic activity.
Although Nernst diffusion impedance Warburg element (W)
and constant phase element (CPE) were drawn together with the
Nyquist plot for QSS, no detailed calculation for CPE or electrical
double layer (Cdl) has been discussed in any studies. Most of the
analysis of QSS is only based on Rct. For more detailed information
on the tting EIS spectra using a Z-view software, see
Refs. [24,49,50,56,57]. These analyses of W, CPE and Cdl were
applied directly on DSSC devices, rather than on QSS. Since
impedance complex plane analysis of polymer electrolytes is not
great compared to metal oxide, unsurprisingly, most recent articles
do not report this raw data of the Nyquist plot, but jump directly to
the conductivity plot.
Fig. 4. (a) photograph and (b) schematic of the conductivity cell or symmetrical cell
using SS316 electrodes and Teon materials.

Fig. 5. (a) Nyquist plot of SAN (wt.%) in NaI/I2 based electrolyte, adapted from Refs. [52]; (b) Ionic resistance of the PPy gel and the liquid electrolyte from the Nyquist plot of EIS,
adapted from Refs. [51]; (c) Nyquist plots for symmetric cells fabricated by sandwiching a conducting gel electrolyte slice between two Pt counter electrodes, adapted from Ref. [29].
 A.A. Mohamad / Journal of Power Sources 329 (2016) 57e71
5.2. Conductivity at room temperature
After the Rct is determined, the conductivity (s) of the QSS can be
calculated (Equation (6)):
t
s [6]
Rb A
where t is the thickness of the electrolyte sample and A is the area
of contact between the sample and the electrode.
Conductivity is one of the major concerns in QSS studies. The
conductivity of GPE is mainly achieved by mobile I /I3 ions from
the conducting QSS electrolyte. The characterizations regarding
ionic conductivity are mostly focused on the effect of the type of
salts (separated or mixed) [36,58e60], salt concentrations
[48,61,62], llers [63], ionic conductors [40], and other external
factors.
The main observations from conductivity measurement data
are:
i. Conductivity vs. salts, plasticizer and polymer: to nd optimum
concentration of modication that produces the highest
conductivity.
ii. Physical properties: to relate these properties that will limit the
quest of the highest conductivity, i.e. dilute gel, dense gels or gel
ooded with liquid.
A typical plot of ionic conductivity with the effect of additional
single salt in agar gel was presented by Singh et al. [62] (Fig. 6a). It is
common to see increases of ionic conductivity during addition of KI
to Agar. After maximum conductivity, adding more KI salt, the ionic
conductivity value decreases. Such an enhancement in ionic con-
ductivity is attributed to the fact that adding KI provided more
mobile charge species (K ion and I anion) in the GPE matrix,
which enhanced overall ionic conductivity; while decreasing ionic
conductivity can be explained on the basis of the charge pair model.
For the effect of mixed salts, rst the optimum conductivity of
the rst salt (NaI) in PAA was identied [64,65]. The highest con-
ductivity of 9.8 0.2  102 S cm1 was attained in PAA with 1.0 M
NaI (Fig. 6b). Although 1.0 M NaI gave the best conductivity, the PAA
became a wet gel. The excess I ions caused the electrolyte to
become highly ionic with the charges tending to repel. The
decrease of solution uptake with the increase of I content forms a
wet polymer gel electrolyte (PGE). Thus, the optimal composition is
found to be 0.8 M NaI with a reasonably high conductivity of
7.6 0.3  102 S cm1.
The next step was to add a second salt, the ionic conductivity of
PAA 0.8 M NaI PGE was added with different concentrations of I2
(Fig. 6c). The highest conductivity of 8.0 0.1  102 S cm1 for the
mixed salts was achieved. The increase was attributed to the
increased amount of chaotropic ion/charge carriers (I 3 ) present at
higher concentrations of I2. Although additional second salts were
studied, they did not contribute to higher ionic conduction of the
gel, and only served as a redox solution. In some cases, the high salt
concentration caused the shrinkage of the polymer chains and
removed the contained solution. This effect reduced ion
transportation.
5.3. Conductivity and absorbency
For ionic transportation (later calculated as conductivity), liquid
electrolyte absorbency is vital. Since the absorbed electrolyte so-
lution acts as medium for ion conduction through the QSS, the QSS
must absorb a higher amount and retain the electrolyte solution
including mobile ions to achieve low internal resistance and high
65
ionic conductivity.
At the same time, several materials and liquid electrolytes have
been investigated in the recent past, in order to state the way in
which QSS parameters like concentration, soaking time, salts and
retention capability affect the QSS absorbency as well as the con-
ductivity. Although the conductivity and absorbency can be
measured separately, most researchers incorporate conductivity
and absorbency (vs. time) together [25,26,28,29,37]. This approach
offers a direct correlation between the quantity of absorbed liquid
electrolyte and conductivity.
Two main observations (two Y-axes) on the same gure are:
i. Liquid electrolyte loading vs. time: to measure the maximum
time taken for loading of liquid until reaching saturated
equilibrium.
ii. Conductivity vs. time: to analyze conductivity patterns and
compare with liquid electrolyte loading.
Li et al. [25] combined conductivity and absorbency to study a
poly (hydroxyethyl methacrylate) [poly (HEMA)]/GR-based mate-
rial (Fig. 6d). The samples were immersed in ionic liquid electrolyte
for a maximum of 22 days, and the ionic conductivity equilibria for
all samples were obtained at a swelling time of about 16 days. A
longer time for swelling indicated the diffusion of ionic liquid
electrolyte into a 3D framework of poly (HEMA/GR) matrix. In
terms of pattern, the curve of conductivity did not resemble the
curve of absorbency, especially at the end of the curve. The limited
transportation or saturated mobile ions in the matrix could be the
reason. However, a similar pattern of conductivity and absorbency
was observed for PAA-CTAB with graphene, graphene oxide, and
nanographite gel electrolytes [29], PAA-PEG/PANi [37], PAA-PEG
[28], PAA/GR and PAA/CTAB matrices [26].
In the case of the PGE at various concentrations of alkaline metal
LiI and KI, Li et al. [36] reported interesting results. At lower con-
centrations (<0.5 M of I), the PGE showed higher solubility than
conductivity. However, the trend of conductivity curve still
increased even at small amounts of salts. Only at the higher con-
centration (>0.5 M of I) did both curves follow the same trend (i.e.
downward trends). This behavior was similar to the KOH solution,
the QSS shrank at high concentration and volume characterizations
[18].
These behaviors were explained based on the adsorption
mechanism. Initially, mixed solvents penetrated into the PGE
hybrid by breaking parts of the hydrogen bonds in the hybrid. Later,
complexation interaction between cations (Li or K) and carbonyl
or ether groups occurred in the hybrid, destroying hydrogen bonds
and increasing the liquid electrolyte absorbency. Further increases
in the I concentration in the liquid electrolyte led to the formation
of many linkages in PGE, which caused the shrinkage and the
decrease of liquid electrolyte absorbency. At this concentration, the
viscosity of the liquid electrolyte also increased. This hindered
penetration of liquid electrolyte into the PGE hybrid. A careful
characterization of conductivity should be applied to avoid direct
measurement on excess liquid and not the PGE. Although there are
some changes in conductivity and absorbency, the curves are in a
similar pattern.
5.4. Conductivity at elevated temperature
The stability of QSS at elevated temperature (T) is very impor-
tant because the operation of DSSCs is related to a wide range of
temperature. Unfortunately, the electrolyte is the most sensitive
component when in contact with heat. At high temperature, QSS
would behave differently, and conductivity would differ from
ambient conductivity. The effect of temperature on conductivity
66 A.A. Mohamad / Journal of Power Sources 329 (2016) 57e71

Fig. 6. (a) Conductivity as a function of composition (wt.% of KI) for agarose doped with KI system, adapted from Refs. [62]; Ionic conductivity at 25  C (b) PAA GPE as a function of
NaI concentration, adapted from Refs. [64]; (c) PAA GPE in 0.8 M NaI and various concentrations of I2, adapted from Refs. [65]; (d) Poly (HEMA)/GR dependence of loading of ionic
liquid electrolyte and ionic conductivity on imbibition time, adapted from Ref. [25].
 A.A. Mohamad / Journal of Power Sources 329 (2016) 57e71 67

characterization, also known as temperature dependence, repre-
sents a detail in studying the ionic behavior of QSS. The experi-
mental setup is similar to normal conductivity characterization,
except that the cell is placed in the temperatures in the range of
25e100  C, i.e. in an oven or furnace.
Generally, the temperature-dependent characterization will be
plotted by:
i. Conductivity (s) vs. temperature (T): to characterize the in-
creases in conductivity under the inuence of temperature.
ii. Log conductivity (s) vs. 1/temperature (1/T): to characterize
conductivity behavior, to calculate activation energy (Ea), and to
relate to certain mechanisms.
5.4.1. Conductivity and temperature
The plot of s-T is rarely reported to explain conductivity
behavior, since it does not provide much information about the
mechanism. The main outcome is only to see the increasing con-
ductivity with added salts [58,59], or a comparison between gel and
liquid electrolytes [47] under the inuence of temperature (Fig. 7a).
Few reports exist: for example, Lan et al. [40] used this plot to
characterize polymer gel electrolytes based on synthesized
methods of polymer and ionic concentrations. The plot showed that
the synthesized method had little effect compared to concentration
of ionic concentrations. Sun et al. [66] studied the ratios of the ionic
conductivity of ionic liquid gel and the corresponding ionic liquid.
They found that the ratios of sgel/sIL increased with increasing
temperature.
5.4.2. Arrhenius
The more common investigation of temperature dependence is
further extending it to Ea calculation, which corresponds to the
gradients in linear plots. Since the QSS sample undergoes a ther-
mally activated process, it could be explained by the Arrhenius
equation (Equation (7)). The Arrhenius equation is a simple formula
originally given for the temperature dependence of the rate con-
stant, and therefore the rate of a chemical reaction. The ion trans-
port is also a thermally activated process, just like the rate of a
chemical reaction.
 
Ea
s so exp [7]
kT
where so is the pre-exponential factor, Ea is in J mol1 or eV, k is the
Boltzman constant, and T is temperature in K. Taking the logarithm

Fig. 7. (a) Ionic conductivity of the liquid electrolyte and the grafted cellulose gel electrolyte as a function of temperature, adapted from Ref. [47]; (b) Temperature dependence of
the ionic conductivity for gel-electrolytes with PAA/Gel and PAA/Gel/PANi as a polymer host, adapted from Ref. [35]; (c) Variation of activation energy and conductivity at room
temperature for agar paste with NH4I and I2, adapted from Ref. [44].
68 A.A. Mohamad / Journal of Power Sources 329 (2016) 57e71

(log or ln) of both sides of the equation gives (Equations (8) and
(9)):
 
EA
log s logso [8]
kT
  
EA 1000
log s log so [9]
1000T T
Hence, the Arrhenius plot of log s1/T gives a line whose slope
is Ea/1000k, from which the Ea can be calculated.
For example, a typical linear Arrhenius plot was presented by
Tang et al. [35]. This explained that ionic conduction by Arrhenius
type thermally assisted for PAA/Gel and PAA/Gel/PANi (Fig. 7b). The
Ea is the main parameter to justify the optimization of the sample
and, in this case PAA/Gel/PANI (8.69 kJ mol1) is lower compared to
pure PAA/Gel (10.13 kJ mol1). The data indicate that the faster
transportation rate of I /I3 in PAA/Gel/PANi hybrid as PAA/Gel/PANi
needs a lower energy to move. Using this method, the introduction
of PANi to the polymer system has a positive effect on the ionic
conductivity of gel-electrolyte.
Several groups have reported log s as a function of 1/T plot to
study cellulose, PAA, 1-butyl-3-methylimidazolium iodide ([Bmim]
I), azodiisobutyronitrile (AIBN), potassium iodide (KI), iodine, acetic
acid [47], PANi-Poly (HDT/PEG) [67], PANi-PAA-PEG [37], PPy-poly
(hydroxyethyl acrylate) (PHEA)/PEG [53], poly (HEMA/GR)/PANi
[54], poly (HEMA/GR) [25], PAA/GR and PAA/CTAB [26], poly
(HEMA/CTB)/PANi [27], PAA-g-gelatin and PPA-g-gelatin/PPy [68],
PPy-poly (HEMA/CTAB) [69], gelator 1-butyl-3-methyl-imidazo-
lium tetrauoroborate ([Bumim]BF4) [66], PGA-PPy [11], low mo-
lecular mass organogelators (LMOG) [70], PAA-PEG/PANi and PAA-
PEG/PPy [28], PAA-CTAB/graphene, PAA-CTAB/graphene oxide,
PAA-CTAB/nanographite [29], polyacrylonitrile (PAN) [71], and
polymethylmethacrylate (PMMA) [72].
Plotting the activation energy and conductivity together pro-
vides a graphical view of the effect of salt concentration (Fig. 7c).
Alias [44,73,74] studied agar GPE with double salts, NH4I and I2. The
conductivity of agar GPE decreased slightly after reaching a Fig. 8. Conductivities of the gel electrolyte at different temperatures and the VTF
maximum conductivity, which is attributed to the increase of redox tting: (a) ln s vs. 1/T plots and (b) ln (sT1/2) vs. 1/(TTo) plots. Reproduced from
couple species (I 3 ) at high I2 concentrations. When compared to Ea,
Ref. [76] with permission from The Royal Society of Chemistry.

it still follows the same pattern, but inversely. This shows that the
 
Ea energies needed for the ions to hop together on the coordinating Ea
s AT 1=2 exp [10]
site of the agar were lower compared to those of pure agar paste. RT  T0

5.4.3. Vogel-Tammann-Fulcher where s is the conductivity, A is a pre-exponential factor propor-

The s with a function of 1/T plots are not always linear, as ex-
pected from the Arrhenius law. Some of the plots have a distinct
curvature. For this kind of curve, the Vogel-Tammann-Fulcher (VTF)
equation can be used to interpret the behavior. The VTF behavior is
still thermally assisted, but there are changes, for example in vis-
cosity and free space of the polymer with temperature deviation
from the ideal Arrhenius behavior.
The available literature indicates that a material can have two
different ion-conduction mechanisms in the two different tem-
perature regimes [75]. When a temperature is lower than the
conductivity-transfer temperature (Tc), the material is in the gel
state and the solvent molecules are constrained by the polymer
host chains. At these temperatures, ion conduction depends on the
frequency of the intermolecular ion hopping (Arrhenius-type
conductivity-temperature behavior). When the temperature is
higher than Tc, the material shows VTF ion-conduction behavior.
This behavior is primarily related to the free volume of the solvent
as a gel in the solid-state, and I and I 3 are solvated at these
temperatures. The temperature dependence of ionic conductivity
and segmental mobility is expressed by VTF (Equation (10)) [76]:
tional to the number of carrier ions, Ea has the dimensions of en-
ergy, R is the molar gas constant, T is the absolute temperature and
To is the reference temperature that is usually 50 K below glass
transition temperature (Tg) (measured by DSC).
Yu et al. [76] used tting of the VTF equation to evaluate the
conductivity of (b-hydroxyethyl methacrylate) based organogel
electrolyte system. The plot of ln s1/T showed that the conduc-
tivity of the electrolyte increases with increasing temperature,
indicating that the sample was dominated by the relaxation of the
host polymer, the mobility ions and larger free volume of solvent
(Fig. 8a). However, when plotting using ln (sT1/2)1/(TTo) a
straight line was obtained (Fig. 8b). Then the Ea was calculated to be
1.376 kJ mol1. Meanwhile, PAN host polymers have also been
characterized by VTF, for example PAN Hex4NI MgI2
EC PC [58], PAN Hex4NI LiI I2 EC PC [59] and
PAN EC PC-n-Bu4NI I2 [77].
6. Conclusion and perspectives
The techniques for investigating QSS materials play an
 A.A. Mohamad / Journal of Power Sources 329 (2016) 57e71 69

Table 1
Summary of selected papers on absorbency and conductivity techniques to characterize QSS from 2010 to 2015.

Characterization Material Highlights Ref.

Visual observation: - Poly (acrylic acid) (PAA) poly (ethylene glycol) (PEG) in lithium - Sample expanded due to absorption [36]
digital photo iodide (LiI) or potassium iodide (KI) iodine (I2) N-methyl-2- - Transference color changes to dark orange
pyrrolidone (NMP) g-butyrolactone (GBL)
- PAA/Gel and PAA/Gel/polyaniline (PANi) - Dark red color changes to black [35]
- LiI, KI, and I2 - Proves the synthesis of modied PAA/Gel/PANI hybrid
superabsorbent
Absorbency- - Acac-Py-I2 in g-butyrolactone-N- methylpyrrolidone - Absorbency comparison for different QSS preparations [40]
concentration - Highest absorbency 17 g g1 (synthesis method)
- PAA PEG in LiI or KI I2 NMP GBL - Absorbency comparison between LiI and KI [36]
- Maximum value at 0.1 M LiI or KI, but the values attained
were different
Absorbency-EIS/ - PAA-PEG/aniline PAA-PEG/aniline PANi - Samples imbibe liquid electrolyte and swell at a higher rate in [37]
conductivity: - Tetraethylammonium iodide (TBAI) tetramethylammonium iodide the initial stage, later becoming constant and achieving their
Imbibition Kinetics (TMAI) tetrabutyl-ammonium iodide (TEAI) NaI KI LiI I2 in equilibrium
4-tertbutylpyridine (TBP) acetonitrile (AC) - Conductivity follows same behavior as absorbency
- n 0.08 and 0.17 with Fickian transport mechanism
- PAA-PEG - Swelling time of about 30 h, and no further diffusion occurs [28]
- PANi-polypyrrole (PPy) under longer immersion time
- TMAI NaI KI LiI I2 TBP in AC - Conductivity follows same behavior as absorbency
- n 0.61, 0.65, 0.60 with anomalous mechanism mode
- The loading of liquid electrolyte is controlled by osmotic
pressure
EIS/Nyquist plot - Styreneeacrylonitrile (SAN) NaI/I2 - Spike spectrum [45]
-conductivity - Rct in the range 40e200 U
- Conductivity decreases as viscosity increases
- PPy 1-buty-2,3-dimethylimidazolium iodide (BDI) poly (vinyl - Spike spectrum [51]
alcohol-co-ethylene) (PVA-EL) - Rct in the range 0.20e52 U
- Rct of gel lower than liquid solution
- The ionic conductivity plot of gel and liquid electrolytes
- Cellulose PAA - Triple semicircle spectra for gel and liquid electrolytes [47]
- 1-butyl-3-methylimidazolium iodide ([Bmim] - s vs. salts concentration
I) azodiisobutyronitrile (AIBN) KI I2 acetic acid - s vs. T
- ln s vs. 1/T
- s of the liquid > gel electrolytes
- Poly (ethylene glycol)dimethyl-ether (PEGDME) Na2S S KCl - Two semicircles attributed to the charge transfer at the [46]
counter/electrolyte interface and diffusion of ions in the
electrolyte
- The internal resistance showed a larger radius for the
semicircle
- The ionic conductivity was lowered as the concentration of
PEG oligomer increased
EIS/Conductivity- - PPy integrated poly (hydroxyethyl methacrylate [(poly [HEMA])/cety- - The increase in ionic conductivity with temperature indicates [69]
temperature trimethylammonium bromide) (CTAB) ion transportation enhancement
dependence: - KI I2 NMP AC - Ea was used to compare with the literature
Arrhenius - Polymethylmethacrylate (PMMA) - Ea from gradients in linear Arrhenius plots [71]
- Tetrapropylammoniumiodide (Pr4NI) KI - Single Ea value was used to predict an electrolytic solution
trapped in a PMMA matrix
- Poly (HEMA)/glycerol (GR) 1-butyl-3-methylimidazolium chloride - Lower Ea value for ionic transport along conducting channels [25]
([BMIM]Cl) 1-methyl-3-propylimidazolium iodide (MPII) - The interconnected pore structure provides a superhighway
- Poly (glycidyl acrylate) (PGA)-PPy for efcient ion transport
- Liquid electrolyte TBAI KI LiI I2 in AC and NMP - Ea and so calculation [11]
- Ea and so for PGA-PPy < PGA
- Lower Ea indicates a faster transportation rate of I /I
3 in PGA-
PPy hybrid
- The ion hopping increases with an increase of temperature
and enhances the conductivity of the system
- Low molecular mass organogelators (LMOGs) - Ea changed with Tgel [70]
- I2 LiI N-methylbenzimidazole (NMBI) 1,2-Methyl-3- - Ea is higher when temperature < Tgel
propylimidazoliumiodide (DMPII) in 3-methoxypropionitrile (MePN) - Ea is lower when temperature > Tgel (almost equal to the
liquid electrolytes)
- Different Ea is due to the bril sizes and types of the
microstructures
EIS/Conductivity- - Polyacrylonitrile (PAN) Hex4NI MgI2eEC PC - Tg measured for the To tting [58]
temperature - PAN Hex4NI LiI I2 EC PC - Pre-exponential factor (A) and Ea values obtained by tting [59]
dependence: VTF conductivity data to Equation (10)
- A is proportional to the number of ionic charge carriers in the
electrolyte
- Ea reveals the energy characteristics

important role in the fundamental understanding of the operation that enable spatially resolved measurements such as absorbency
of DSSC devices. It cannot be denied that much characterization has and conductivity characterizations (see Table 1 for a summary of
been performed in recent years, yet much still remains to be done. this review from selected articles).
Most existing research concerns the development of techniques The rapid development in polymer electrolyte characterizations
70 A.A. Mohamad / Journal of Power Sources 329 (2016) 57e71
for batteries, fuel cells, capacitors and others electrochemical de-
vices also gives some added value to new characterizations tech-
niques of QSS or DSSC. Then, more knowledge can be built upon
this foundation. However, most techniques have not changed
greatly since 2010 (or even the 2000s). To date, the main charac-
terizations of polymers and the I /I 3 system are almost mature,
which means that researchers are repeating the same character-
ization standards or must do list characterizations for QSS. Some
studies only change the materials, while the explanation is still the
same.
It is hoped that future characterization should be extended to
the mechanism of ions movements in order to enhance the charge
transfer of QSS/electrode, apart from obtaining the value of con-
ductivity. A good understanding of ionics conductivity is essential,
and a creative technique of proper characterization should be car-
ried out, however aiming at a deeper understanding on the QSS
side.
Another crucial part of the technique is the design of the char-
acterization setups that are tailored with the practical DSSCs.
Obviously, the combined use of multiple in-situ techniques is able
to probe absorbency and conductivity on the nanoscale in 3D in real
time. Examples of such combinations include: absorbency-
temperature, absorbency-conductivity-SEM and absorbency-
conductivity-X-ray diffraction. Given recent progress, it can be ex-
pected that such in-situ research will become increasingly impor-
tant in the characterization techniques of future QSS for DSSCs.
Currently, the power conversion efciencies related to QSS-
DSSC is typically about 7e10%. However, with continuous
improvement of QSS characterizations, new materials and under-
standing materials, the next generation of QSS-DSSCs can be real-
ized. Such cells may become alternatives to the organic liquid
electrolyte DSSC (efciency of 13%). Overall, the improvement of
QSS characterization is not only advantageous for DSSCs but also for
other electrochemical devices.
Acknowledgments
Financial support from USM-RUI 1001/PBahan/814262 is
gratefully appreciated.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.jpowsour.2016.08.064.
References
[1] B. O'regan, M. Gr atzel, Nature 353 (1991) 737e740.
[2] J. Wu, Z. Lan, J. Lin, M. Huang, Y. Huang, L. Fan, G. Luo, Chem. Rev. 115 (2015)
2136e2173.
[3] M.K. Nazeeruddin, E. Baranoff, M. Gra tzel, Sol. Energy 85 (2011) 1172e1178.
[4] M. Jrgensen, K. Norrman, S.A. Gevorgyan, T. Tromholt, B. Andreasen,
F.C. Krebs, Adv. Mater. 24 (2012) 580e612.
[5] J. Gong, J. Liang, K. Sumathy, Renew. Sustain. Energy Rev. 16 (2012)
5848e5860.
[6] H.K. Jun, M.A. Careem, A.K. Arof, Renew. Sustain. Energy Rev. 22 (2013)
148e167.
[7] M. Ye, X. Wen, M. Wang, J. Iocozzia, N. Zhang, C. Lin, Z. Lin, Mater. Today 18
(2015) 155e162.
[8] F. Bella, R. Bongiovanni, Journal of photochemistry and photobiology C, Pho-
tochem. Rev. 16 (2013) 1e21.
[9] P.K. Singh, R. Nagarale, S. Pandey, H. Rhee, B. Bhattacharya, Advances in nat-
ural sciences, Nanosci. Nanotechnol. 2 (2011) 023002.
[10] F. Cao, G. Oskam, P.C. Searson, J. Phys. Chem. 99 (1995) 17071e17073.
[11] Z. Tang, J. Wu, Q. Li, Z. Lan, L. Fan, J. Lin, M. Huang, Electrochim. Acta 55 (2010)
4883e4888.
[12] A.F. Nogueira, C. Longo, M.A. De Paoli, Coord. Chem. Rev. 248 (2004)
1455e1468.
[13] A.A. Mohamad, J. Power Sources 159 (2006) 752e757.
[14] S.C. Yap, A.A. Mohamad, Electrochem. Solid-state Lett. 10 (2007) A139eA141.
[15] A.A. Mohamad, Corros. Sci. 50 (2008) 3475e3479.
[16] N.H. Khalid, Y.M. Ismail, A.A. Mohamad, J. Power Sources 176 (2008) 393e395.
[17] M.N. Masri, M.F.M. Nazeri, A.A. Mohamad, Adv. Sci. Technol. 72 (2011)
305e308.
[18] Y.M.B. Ismail, H. Haliman, A.A. Mohamad, Int. J. Electrochem. Sci. 7 (2012)
3555e3566.
[19] S.C. Yap, Fabrikasi Dan Pencirian Bateri Proton Berasaskan Elektrolit Polimer
Gel Hidroponik, Final Year Project Report, School of Materials and Mineral
Resources Engineering, Universiti Sains Malaysia, Nibong Tebal, Pulau Pinang,
2006.
[20] A. Jamaludin, Z. Ahmad, Z.A. Ahmad, A.A. Mohamad, Int. J. Hydrogen Energy
35 (2010) 11229e11236.
[21] P.L. Ng, A. Jamaludin, Y. Alias, W.J. Basirun, Z.A. Ahmad, A.A. Mohamad, J. Appl.
Polym. Sci. 123 (2012) 2662e2666.
[22] F. Bella, C. Gerbaldi, C. Barolo, M. Gra tzel, Chem. Soc. Rev. 44 (2015)
3431e3473.
[23] M.J. Zohuriaan-Mehr, K. Kabiri, Iran. Polym. J. 17 (2008) 451e477.
[24] Y. Xiao, J. Wu, J.-Y. Lin, G. Yue, J. Lin, M. Huang, Z. Lan, L. Fan, J. Power Sources
241 (2013) 373e378.
[25] Q. Li, Q. Tang, N. Du, Y. Qin, J. Xiao, B. He, H. Chen, L. Chu, J. Power Sources 248
(2014) 816e821.
[26] Q. Li, Q. Tang, B. He, P. Yang, J. Power Sources 264 (2014) 83e91.
[27] Q. Li, Q. Tang, L. Lin, X. Chen, H. Chen, L. Chu, H. Xu, M. Li, Y. Qin, B. He, J. Power
Sources 245 (2014) 468e474.
[28] S. Yuan, Q. Tang, B. He, P. Yang, J. Power Sources 254 (2014) 98e105.
[29] S. Yuan, Q. Tang, B. He, Y. Zhao, J. Power Sources 260 (2014) 225e232.
[30] S. Yuan, Q. Tang, B. He, L. Yu, J. Power Sources 273 (2015) 1148e1155.
[31] S.Y. Heo, J.K. Koh, J.K. Kim, C.S. Lee, J.H. Kim, Electrochim. Acta 137 (2014)
34e40.
[32] L. Wang, H. Zhang, C. Wang, T. Ma, ACS Sustain. Chem. Eng. 1 (2013) 205e208.
[33] T. Makris, V. Dracopoulos, T. Stergiopoulos, P. Lianos, Electrochim. Acta 56
(2011) 2004e2008.
[34] B.C. Nath, D. Das, I.R. Kamrupi, K.J. Mohan, G.A. Ahmed, S.K. Dolui, RSC Ad-
vances 5 (2015) 95385e95393.
[35] Z. Tang, J. Wu, Q. Liu, M. Zheng, Q. Tang, Z. Lan, J. Lin, J. Power Sources 203
(2012) 282e287.
[36] X. Li, D. Zhang, X.J. Yin, S. Chen, J. Shi, Z. Sun, S. Huang, J. Solid State Elec-
trochem. 15 (2011) 1271e1277.
[37] P. Li, S. Yuan, Q. Tang, B. He, Electrochim. Acta 137 (2014) 57e64.
[38] J. Zhang, W. Zhang, J. Guo, C. Yuan, F. Yan, Electrochim. Acta 165 (2015)
98e104.
[39] J. Shi, J. Chen, Y. Li, Y. Zhu, G. Xu, J. Xu, J. Power Sources 282 (2015) 51e57.
[40] Z. Lan, J. Wu, J. Lin, M. Huang, Comptes Rendus Chim. 13 (2010) 1401e1405.
[41] J. Duan, Q. Tang, R. Li, B. He, L. Yu, P. Yang, J. Power Sources 284 (2015)
369e376.
[42] S. Kelkar, K. Pandey, S. Agarkar, N. Saikhedkar, M. Tathavadekar, I. Agrawal,
R.V. Gundloori, S. Ogale, ACS Sustain. Chem. Eng. 2 (2014) 2707e2714.
[43] Q. Tang, S. Yuan, H. Cai, J. Mater. Chem. A 1 (2013) 630e636.
[44] S.S. Alias, A.A. Mohamad, Ionics 19 (2013) 1185e1194.
[45] W.C. Tan, S.S. Alias, A.B. Ismail, A.A. Mohamad, J. Solid State Electrochem. 16
(2012) 2103e2112.
[46] H. Kim, I. Hwang, K. Yong, ACS Appl. Mater. Interfaces 6 (2014) 11245e11253.
[47] P. Li, Y. Zhang, W. Fa, Y. Zhang, B. Huang, Carbohydr. Polym. 86 (2011)
1216e1220.
[48] H.-Y. Chen, L. Lin, X.-Y. Yu, K.-Q. Qiu, X.-Y. L, D.-B. Kuang, C.-Y. Su, Electro-
chim. Acta 92 (2013) 117e123.
[49] B. He, Q. Tang, J. Luo, Q. Li, X. Chen, H. Cai, J. Power Sources 256 (2014)
170e177.
[50] B. He, Q. Tang, M. Wang, C. Ma, S. Yuan, J. Power Sources 256 (2014) 8e13.
[51] M.-H. Jung, J. Power Sources 268 (2014) 557e564.
[52] W.C. Tan, Synthesis of ZnO via Zinc-air and Application of Styrene-
acrylonitrile-based Electrolyte for Quasi-solid-state Dye-sensitized Solar
Cell, Master of Science Thesis, School of Materials and Mineral Resources
Engineering, Universiti Sains Malaysia, Nibong Tebal, Pulau Pinang, 2011.
[53] Q. Li, X. Chen, Q. Tang, H. Cai, Y. Qin, B. He, M. Li, S. Jin, Z. Liu, J. Power Sources
248 (2014) 923e930.
[54] Q. Li, Q. Tang, H. Chen, H. Xu, Y. Qin, B. He, Z. Liu, S. Jin, L. Chu, Mater. Chem.
Phys. 144 (2014) 287e292.
[55] Z. Seidalilir, R. Malekfar, H.-P. Wu, J.-W. Shiu, E.W.-G. Diau, ACS Appl. Mater.
interfaces 7 (2015) 12731e12739.
[56] H.-W. Chen, Y.-D. Chiang, C.-W. Kung, N. Sakai, M. Ikegami, Y. Yamauchi, K.C.-
W. Wu, T. Miyasaka, K.-C. Ho, J. Power Sources 245 (2014) 411e417.
[57] Z. Huo, C. Zhang, X. Fang, M. Cai, S. Dai, K. Wang, J. Power Sources 195 (2010)
4384e4390.
[58] T.M.W.J. Bandara, M.A.K.L. Dissanayake, W.J.M.J.S.R. Jayasundara, I. Albinsson,
B.-E. Mellander, Phys. Chem. Chem. Phys. 14 (2012) 8620e8627.
[59] T.M.W.J. Bandara, W.J.M.J.S.R. Jayasundara, M.A.K.L. Dissanayake,
H.D.N.S. Fernando, M. Furlani, I. Albinsson, B.-E. Mellander, Int. J. Hydrogen
Energy 39 (2014) 2997e3004.
[60] J. Fan, L. Li, H.-S. Rao, Q.-L. Yang, J. Zhang, H.-Y. Chen, L. Chen, D.-B. Kuang, C.-
Y. Su, J. Mater. Chem. A 2 (2014) 15406e15413.
[61] J. Chen, T. Peng, K. Fan, J. Xia, J. Mater. Chem. 21 (2011) 16448e16452.
[62] R. Singh, N.A. Jadhav, S. Majumder, B. Bhattacharya, P.K. Singh, Carbohydr.
Polym. 91 (2013) 682e685.
[63] K.-C. Huang, R. Vittal, K.-C. Ho, Sol. Energy Mater. Sol. Cells 94 (2010)
 A.A. Mohamad / Journal of Power Sources 329 (2016) 57e71 71

675e679. W.J.M.J.S.R. Jayasundera, B.-E. Mellander, J. Photochem. Photobiol. A Chem.

[64] A.L. Tan, Gel Polymer Electrolyte Based on Poly (Acrylic Acid), Sodium Iodide 246 (2012) 29e35.
and Iodine for Quasi-solid State Solar Cells, Final Year Project Report, School of [72] M.A.K.L. Dissanayake, R. Jayathissa, V.A. Seneviratne, C.A. Thotawatthage,
Materials and Mineral Resources Engineering, Universiti Sains Malaysia, G.K.R. Senadeera, B.-E. Mellander, Solid State Ionics 265 (2014) 85e91.
Nibong Tebal, Pulau Pinang, 2009. [73] S.S. Alias, Synthesis and Characterization of Agar Gel Polymer Electrolyte and
[65] A.L. Tan, L.J. Khoo, S.S. Alias, A.A. Mohamad, J. Sol-gel Sci. Technol. 64 (2012) Zinc Oxide for Photoelectrochemical Cell, Master of Science Thesis, School of
184e192. Materials and Mineral Resources Engineering, Universiti Sains Malaysia,
[66] S. Sun, J. Song, R. Feng, Z. Shan, Electrochim. Acta 69 (2012) 51e55. Nibong Tebal, Pulau Pinang, Malaysia, 2011.
[67] Q. Li, H. Chen, L. Lin, P. Li, Y. Qin, M. Li, B. He, L. Chu, Q. Tang, J. Mater. Chem. A [74] S.S. Alias, A.A. Mohamad, Synthesis of Zinc Oxide by Sol-gel Method for
1 (2013) 5326e5332. Photoelectrochemical Cells, Springer, 2014.
[68] Q. Li, J. Wu, Z. Tang, Y. Xiao, M. Huang, J. Lin, Electrochim. Acta 55 (2010) [75] J.H. Wu, S.C. Hao, Z. Lan, J.M. Lin, M.L. Huang, Y.F. Huang, L.Q. Fang, S. Yin,
2777e2781. T. Sato, Adv. Funct. Mater. 17 (2007) 2645e2652.
[69] Y. Qin, X. Chen, Q. Tang, Q. Li, B. He, K. Chen, S. Jin, W. Dai, M. Li, Y. Xie, Polym. [76] Z. Yu, D. Qin, Y. Zhang, H. Sun, Y. Luo, Q. Meng, D. Li, Energy & Environ. Sci. 4
Eng. Sci. 54 (2014) 2531e2535. (2011) 1298e1305.
[70] L. Tao, Z. Huo, Y. Ding, L. Wang, J. Zhu, C. Zhang, X. Pan, M.K. Nazeeruddin, [77] O.A. Ileperuma, G.R. Asoka Kumara, H.-S. Yang, K. Murakami, J. Photochem.
S. Dai, M. Gra tzel, J. Mater. Chem. A 2 (2014) 15921e15930. Photobiol. A Chem. 217 (2011) 308e312.
[71] M.A.K.L. Dissanayake, C.A. Thotawatthage, G.K.R. Senadeera, T.M.W.J. Bandara,