REVIEW

Recent Trends in High Efficiency Photo-Electrochemical SolarCell Using Dye-Sensitised Photo-Electrodes and Ionic LiquidBased Redox Electrolytes

Suresh Chandra

Received: 1 January 2011 / Accepted: 8 April 2011 / Published online: 1 March 2012

� The National Academy of Sciences, India 2012

Abstract Low cost and long life photovoltaic solar cell is

one of the most viable renewable energy technologies

needed for the future. The development of commercial

solid–solid junction type solar cells (e.g., p–n junction

based on semiconductor like Si, GaAs, CdS, CdTe etc.) is

still limited by cost. A new technology of semiconductor-

electrolyte junction viz. Photo-electrochemical solar cells

(PESC) has recently evolved. Dye-sensitization of PESC

photo-electrodes has revolutionized the concept of PESC

and many low cost ‘‘dye-sensitized solar cells (DSSC)’’

with fairly high efficiency have been reported which has

made them as a safe bet for the future energy needs. This

paper gives an overview of the recent progress made in

DSSC after elucidating the principles and limitations of

earlier PESC’s. The future prospects are also discussed.

Keywords Solar cells � Photoelectrochemical cells �Dye sensitised solar cells � Ionic liquids

Introduction

Direct conversion of solar energy to electrical energy

through the use of photo-voltaic solar cells has been the

strategy followed for more than four decades to develop an

alternate renewable source of energy in order to replace, at

least partly, the fossil fuel based energy systems. Genera-

tion of photovolatage in photovoltaic solar cells involves

the following steps:

(i) Each photon having energy greater than the bandgap

of semiconductor (i.e., hm[ Eg) results in the transfer

or excitation of one valence band electron to the

conduction band, leaving a ‘hole’ in the valence band.

In other words, such photoexcitation leads to the

generation of electron–hole pair:

semi conductor�!hme� þ hþ

(ii) Subsequently photo-voltage is generated when these

photogenerated electron–hole pairs are separated by

some electric field ‘‘built-in’’ within the solar cell.

In general, the ‘‘built-in’’ field is produced by some

charge discontinuity or inhomogeneity. The earliest

approach was to have a p–n semiconductor junction. It is

well known in semiconductor physics [1] that electrons

diffusing from n-type semiconductor to p-type semicon-

ductor and hole diffusing from p-side to n-side annihilate

near the junction resulting in the formation of ‘‘depletion

layer’’ consisting of sheaths of positively and negatively

charged atoms on n-side and p-side respectively. This

results in a junction field. Most earlier developed conven-

tional solar cells were p–n junction solar cells. Subse-

quently, many more variants of junction solar cells were

developed like Metal–Semiconductor (M–S), Metal–

Oxide–Semiconductor (M–O–S), Metal–Insulator–Semi-

conductor (M–I–S) etc. All these are ‘Solid–Solid junction’

solar cells. The technology of fabricating such solar cells is

expensive both in terms of actual money as well as power

needed to produce them. Nonetheless, most of the currently

marketed solar cells belong to this category. Finding

alternatives of these solar cells was and still is a challenge.

It is clear from the above discussion that the creation of

depletion layer and junction field in the conventional solid–

solid junction solar cell was as a result of charge transfer

S. Chandra (&)

Physics Department, Banaras Hindu University, Varanasi

221005, India

e-mail: sureshchandra_bhu@yahoo.co.in

123

Proc. Natl. Acad. Sci. Sect. A Phys. Sci. (January–March 2012) 82(1):5–19

DOI 10.1007/s40010-012-0001-4

across p and n sides (in p–n junction) or across metal and

semiconductor (in M–S junction). A different type of

charge transfer can possibly occur from the semiconductor

to liquid electrolyte (consisting of ?ve and -ve ions like

holes/electrons in semiconductors). Such an electrochemi-

cal charge transfer under photo-excitation was visualized

by Fujushima and Honda [2] to split water (electrolysis)

and by Gerisher [3] for producing electricity directly from

solar energy. The formation of junction simply implied

dipping a semiconductor into a liquid electrolyte. This was

the beginning of replacing solid state junction of conven-

tional solar cells by solid–liquid electrolyte junction solar

cells. Considering the nature of charge transfer, such solar

cells were termed as ‘‘photo-electrochemical solar cells

(PESC)’’ [4]. The story of development of PESC is full of

ups and downs. It started with hype in 1970s; sulked in late

1980s when satisfactory solutions to problems like stabil-

ity, photo-corrosion, life and efficiency could not be found

out; got an upward kick in 1990s when Gratzel’s group [5]

in Switzerland suggested the new concept of dye sensiti-

zation of photo electrodes of PESC resulting in high effi-

ciency dye sensitized solar cells (DSSC) which was further

boosted in 2000s when ionic liquids offered a novel route

for obtaining high performance electrolytes for DSSC (see

[6] and references cited there in). The present paper briefly

introduces the basic principles, problems and perspectives

of early Photo-electrochemical solar cells followed by an

overview of the recent trends of dye sensitization and use

of ionic liquids in developing long- life, sustainable, effi-

cient and inexpensive photo-electrochemical solar cells.

Photo-electrochemical Solar Cells: Basics

As discussed above, the photo generated charge carriers in

semiconductor junction solar cells are separated by the

field across the depletion layer. Let us consider the most

common p–n junction shown in Fig. 1a. The p-side has

more free holes while n-side has more electrons. So,

electrons from n-side would diffuse towards p-side and

holes will move the other way. At or near the junction, they

may collide and annihilate leaving behind immobile ion-

ized donor and acceptor atoms on the two sides of the

interface. These ionized donors and acceptor atoms result

in a sheath of positive and negative charges separated by a

distance, called as depletion layer. Hence, an electric field

is created across the depletion layer. The diffusion of

electrons and holes will continue till the junction field is

large enough to stop such cross-over. At equilibrium, the

Fermi levels on the p-side and n-side align themselves. The

energy level diagram for such a p–n junction is shown in

Fig. 1b. On photoexcitation, the photogenerated electron–

hole pairs modify the number of electrons and holes near

the junction and hence modify the junction voltage and

current. It may be noted that the energy levels on both

p- and n-sides are bent depending upon the respective

charge carrier concentrations.

Suppose one side of junction in Fig. 1a is replaced by an

electrolyte. This can be attained by dipping the semicon-

ductor into an electrolyte as shown in Fig. 2.

This forms the basis of photo-electrochemical solar cell.

The junction can now be visualized as a semiconductor-

electrolyte junction in which the electrons–holes from

semiconductor and ions from electrolyte may try to cross

the junction (as in p–n junction) resulting in reduction–

oxidation (REDOX) reaction and the formation of deple-

tion layer. To understand this, we will have to study the

potential and charge distributions on the semiconductor

and electrolyte side separately in order to be able to con-

struct the energy level diagram near the junction.

Charge and Potential Distribution on the Electrolyte

and Semiconductor Sides of a PESC

For proper understanding of semiconductor-electrolyte

interface, the charge and potential distributions on both

electrolyte and semiconductor sides have to be understood.

The charge distribution on the electrolyte side of the

interface in a photoelectrochemical solar cell is more

Fig. 1 a Formation of p–n junction, b energy level diagram near the

junction

6 Proc. Natl. Acad. Sci. Sect. A Phys. Sci. (January–March 2012) 82(1):5–19

123

complicated than that in p–n junction. Generally, a dense

layer of ions get stuck to the semiconductor photo electrode

surface (termed as Helmholtz layer) while the ions form a

diffuse layer beyond this (termed as Gouy layer) as shown

in Fig. 3.

Most of the cations and anions in aqueous electrolyte are

‘‘solvated’’ by water molecules. The attractive image as

well as dipole induced forces due to the water dipole,

dispersive forces and chemical bonding are collectively

responsible for sticking of the water dipoles to the semi-

conductor electrode surface. Apart from this, partially

desolvated ions also can get stuck to the semiconductor

surface. Some possible manners of sticking of ions to the

semiconductor electrode surface to form the inner dense

layer (Helmholtz layer) are shown in Fig. 4.

The free as well as the solvated cations/anions in the

bulk of the electrolyte beyond the dense layer form a

diffused layer (also termed as Gouy layer) as shown

earlier in Fig. 3. Potential distribution, /G(x), in the

Gouy layer can be obtained by solving the Poisson’s

equation:

d2uGðxÞdx2

¼ �qelðxÞ=eoeGð1Þ

where qel is the ionic charge in the electrolyte at a distance

x from the junction, eo and eG are free space and Gouy

layers permittivities, respectively. The charge distribution

for cations and anions can be approximated by Boltzmann

distribution:

qelðxÞ¼X

iZieCo

i exp½�ZieuðxÞ=kT � ð2Þ

where Zi is the valence of the ith ion, Coi is the ion con-

centration at x = 0.

Assuming concentration of both the cations and anions

to be same and taking Zi = 1, Eq. 1 can easily be solved

for uGðxÞwhich is given by

uGðxÞ ¼ uo exp[ � LGx� ð3Þ

where /ois potential at x = 0 and LG is Gouy layer

thickness given by the following expression

LG ¼ eoeGkT.

2Coe2Z2� �1=2 ð4Þ

Similarly we can get the expression for the potential dis-

tribution on the ‘‘semiconductor-side’’ of the semiconductor-

electrolyte interface /SCð Þ by solving relevant Poisson’s

equation:

d2uscðxÞdx2

¼ � 1

eoesc

:qscðxÞ ð5Þ

where esc and qsc, respectively, are dielectric constant and

charge density on the semiconductor side given by

qscðxÞ ¼ e½ND � nðxÞ þ pðxÞ � NA� ð6Þ

where ND and NA are number of ionized donors and

acceptors, n(x) and p(x) are electrons and hole

concentration at different distances (x) from the interfaces.

Assuming exponential variation of n and p with x, the

potential distribution on the semiconductor side can be

obtained as (see [4]):

uscðxÞ ¼ uo �LDxð Þ ð7Þ

where LD is termed as Debye length given by

LD ¼eoesckT

2ne2

� �1=2

ð8Þ

The total interface potential, uGa (galvanic potential) will

be the sum of potential drop across space charge layer in

Fig. 2 Basic photo electrochemical solar cell

Fig. 3 Hypothesized ionic charge distribution on the electrolyte side

of the interface

Proc. Natl. Acad. Sci. Sect. A Phys. Sci. (January–March 2012) 82(1):5–19 7

123

semiconductor (usc), dense or Helmholtz layer on the

electrode side (uH) and Gouy layer (uG):

uGa ¼ usc þ uH þ uG

An order of magnitude calculation for a typical PESC

shows that [4]

usc : uH : uG ¼ 750 : 1 : 1:5

So, usc is only important. This is approximately similar to

metal–semiconductor junction where most of the drop is

known to occur on the semiconductor side. Thus, a PESC

behaves like a M–S junction solar cell.

With this background, we are in a position to discuss

the basics of PESC. The four energy levels involved are

conduction and valance bands (Ec and Ev) of the semi-

conductor and oxidation and reduction energy levels of

redox specie (Wox and Wred) in the electrolyte. On

illumination of semiconductor, electron–hole pair gener-

ation occurs which are separated in the space charge layer

on the semiconductor side with subsequent lowering of

depletion layer barrier. This results in increased number

of ‘‘holes’’ crossing over to the electrolyte side to be

captured by the reduced ionic species. On short circuiting

(or connecting through a load) the photoelectrode and

counter electrode of the PESC (see Fig. 2), the electrons

are driven via the external circuit towards the metal

counter electrode where it participates in the reduction

reaction of the oxidized species. If the redox couple is

such that the total cathodic ? anodic reactions do not

lead to a net chemical change (DG = 0) and the two

electrodes only serve as a ‘‘shuttle’’ for the charge

transfer, then the cell is an ‘‘electrochemical photovoltaic

device or solar cell’’. If net chemical reaction is taking

place as a result of charge transfer with DG \ 0 or

DG [ 0, then such cells are respectively called as pho-

toelectrolysis cell and photocatalytic cells. This paper is

concerned with electrochemical photovoltaic solar cell

(i.e., DG = 0; no net chemical change). It is obvious that

the overall efficiency of such solar cells will depend upon

Ec and Ev of semiconductor, charge transfer kinetics

between redox electrolyte and semiconductor as well as

counterelectrode.

Fig. 4 The formation of ionic

dense layer on the electrolyte

side of PESC near the

semiconductor electrode a Two

possible orientation of water

dipole. b Solvated ions sticking

in the electrode. c Partially

desolvated ions sticking to the

electrode

8 Proc. Natl. Acad. Sci. Sect. A Phys. Sci. (January–March 2012) 82(1):5–19

123

Materials Choice and Problems of Conventional PESC

The materials choice depends mainly upon the following

criteria:

(i) high efficiency (g)

(ii) long life (negligible corrosion)

(iii) suitable band gap of the photoelectrode so as to be

able to absorb major portion of solar spectrum and

produce electron–hole pair. Further factors like

surface states, recombination, etc. should not reduce

efficiency much.

(iv) Compatibility of the electrolyte with the semicon-

ductor to yield fast charge transfer kinetics.

The efficiency of solar cell g is defined as:

g ¼ output power

input power (solar energy)� 100 %

¼Eg

R1Eg

aðEÞNðEÞdER1

0NðEÞdE

ð9Þ

where Eg is the band-gap of semiconductor, N(E) is number

of photons coming from sun having energy E and a(E) is the

fraction of photons absorbed. The photons with energy

greater than Eg (i.e., hm[ Eg) are only important for solar

energy conversion since only these photons would generate

electron–hole pairs. This is why the lower limit of integra-

tion in the numerator of Eq. 9 is Eg. It is obvious that for

higher g, Eg and a(E) both must be large which often are self-

contradictory. Considering the energy distribution in ter-

restrial solar spectrum, it is found from Eq. 9 that highest

efficiency will be found for Eg & 1.5 eV. The popular

semiconductor materials in the above range of Eg are GaAs,

GaP, CdTe, CdSe etc. (see [4] for detailed discussion).

Because of the high promise and potential of PESC, hun-

dreds of studies were reported between 1975 and 1985 using

single crystals, thin films, electro-deposited thin films etc. of

GaAs, CdS, CdTe, MoSe2 etc. Reasonably high efficiency

(*7 to 19%) using small area n-GaAs single crystals could

be attained using S2�=S2�2 redox electrolyte [7, 8]. Single

crystals of n-CdS Eg * 2.4 eV with a variety of redox

couple like S2�=S2�2 ; I�=I2; Fe(CN)3�

6 =Fe(CN)4�6 also

were found to give efficiencies in the range 1.5–6.5% but

were generally unstable [9–11]. The stable cells had much

lower efficiency. Single crystalline CdTe and CdSe

(Eg & 1.4 and 1.7 eV), whose band-gaps are near to the

optimum values gave higher efficiencies as expected [12–

17]. Large semiconductor crystals are difficult to prepare

and therefore, thin film large area PESC’s were developed

but their efficiencies were considerably less [18–20]. Most

physical techniques of thin film deposition are expensive.

Chandra’s group [21–25] in India tried a cost effective

electrodeposition technique for preparing large area thin

films of semiconductors and used them to fabricate PESC’s.

The efficiencies as well as stability of these PESC’s were

also less than those of single crystalline PESC’s.

The limitation imposed by the band gaps on the effi-

ciency of the solar cell can be partly taken care of by

designing multiple band-gap semiconductor electrolyte

solar cells (for a review see [26]). This concept (Fig. 4) is

equally applicable to the solid–solid junction solar cells

(Fig. 5).

It is well known that higher band gap materials will give

higher photo-potential but low photo-current because the

lower energy photons (which are more abundant in ter-

restrial solar radiation) are not absorbed by the wide band-

gap semiconductors (ht\ Eg) and hence do not contribute

to the photo-current. On the contrary, the low band-gap

semiconductors will give large photocurrent but lower

photovolatage. Therefore, multiple band-gap systems will

provide better matching/utilization of incident solar radia-

tion. Consider a solar cell of four semiconductor photo-

electrodes with different band gaps Eg1, Eg2, Eg3, Eg4 such

that Eg1 [ Eg2 [ Eg3 [ Eg4. The higher energy part of

solar insolation ht1 \ Eg1 will be absorbed in semicon-

ductor 1 and the lower energy photons will pass to the

semiconductor 2 in the stack where photons whose energy

is more than Eg2will be absorbed. This process will con-

tinue from one semiconductor to the next in the stack. This

results in better utilization of the whole part of solar

spectrum. Many variants of PESC’s based on multi-band

gap structures are designed by Licht [26]. The achievable

theoretical efficiencies are predicted to be higher than 30–

50% which still remains elusive.

Attempts were made not only to try different band gap

semiconductors in PESC, but some interesting configurations

Fig. 5 Conceptual visualization of utilization of different energy

photons in a stacked multijunction solar cell

Proc. Natl. Acad. Sci. Sect. A Phys. Sci. (January–March 2012) 82(1):5–19 9

123

were also tried in which the liquid electrolyte was replaced

by polymer electrolyte. The first attempt in this direction was

by Skotheim’s group [27, 28] using PEO-based polymer

electrolytes. The efficiency as well as the stability was very

low. The lower efficiency was due to lower conductivity as

well as the interfacial contact problem because of web-like

contact (not smooth contact) between polymer electrolyte

film and the semiconductor due to the shrinkage of the former

[29].

Apart from the low efficiency, the more serious problem

in all types of PESC’s was the stability and short life of the

cell due to the decomposition of semiconductor surface

(electrochemical corrosion) in contact with the electrolyte.

It may be remarked here that the mobile charges of both

signs from semiconductor (electrons and holes) and elec-

trolyte (cations and anions) participate in the charge

transfer reaction at the semiconductor electrolyte interface.

Consider a compound semiconductor AB. The decom-

position takes place in the following two ways:

(a) Cathodic decomposition: electrons (n) participate in

the reaction like

AB + Z:e� ! AþBZ�solv: ð10Þ

The more electronegative element B goes as the sol-

vated ion in the electrolyte. The free energy for the

above reduction reaction should exceed the thermo-

dynamic decomposition potential nVD.

(b) Anodic decomposition: holes (p) participate in the

decomposition reaction

AB + Z:hþ ! Aþsolv:þB ð11Þ

The more electropositive element A goes as the sol-

vated ion in the electrolyte. The corresponding ther-

modynamic decomposition potential is pVd.

The electrode decomposition is a thermodynamically as

well as chemically controlled process. The semiconductor

is ‘‘stable’’ if

pVD [ Vredox [ nVD ð12Þ

It is unstable, if

Vredox [ pVD or Vredox\nVD ð13Þ

Sometimes, even when energetics of Eq. 12 predicts

stability, the decomposition occurs due to ‘‘slow charge

transfer reaction kinetics’’. Many efforts were made to stop

corrosion in late 1970s and 1980s (see [4]) like surface

decoration by a protective layer, proper choice of redox

electrolyte, use of a fast redox couple, etc. However, the

success was limited. The journey of research efforts for

replacing conventional solid–solid junction (like p–n) solar

cells by PESC were found to be heading towards

insurmountable road-block.

It was felt that wide band semiconductors are likely to

be less corrosive (but not good for photo generation of

charge carriers). The photo generation with ht & 1.5 eV

will be ideal, but such semiconductor would corrode if its

charge-transfer reaction at the interface involves electro-

lyte. These are self contradictory and the solution of this

problem eluded till the novel concept of dye-sensitization

of a wide band gap semiconductor (TiO2) was propounded

by Gerischer and his group [3] in 1991 giving the ‘‘dye-

sensitized solar cells (DSSC)’’ described below.

Dye sensitized solar cells (DSSC): working principle

In the first ‘‘Dye-Sensitized Solar Cell’’, O’Regan and

Gratzel [5] used TiO2 sensitized by a ruthenium complex

dye as photoelectrode. The photo excitation took place in

the dye and excited dye transferred the electron to the wide

bandgap nano-TiO2. Since the first report of O’Regan and

Gratzel [5], many groups became active worldwide and

numerous cell configurations have been studied. Cost

effective and commercially viable DSSC’s with efficiency

*10% have already been reported. (see for example

[6, 30–33], and references cited therein). The basic design

of DSSC is schematically shown in Fig. 6.

The photoelectrode is usually obtained by forming a thin

layer of nano-TiO2 paste on to a conducting glass plate

(ITO or FTO). Two most common methods for spreading a

uniform layer of TiO2 are:

1. Doctor Blade method. In this method the TiO2 paste

goes between a blade edge and the ITO/FTO substrate.

Fig. 6 Schematic representation of a dye sensitized solar cell

10 Proc. Natl. Acad. Sci. Sect. A Phys. Sci. (January–March 2012) 82(1):5–19

123

The film spreads as blade moves along the substrate.

The thickness is controlled by adjusting the blade edge

height.

2. Conventional ‘‘Screen Printing’’ technique. The ITO/

TiO2 assembly as obtained above is subsequently

dipped in the dye solution. The dye molecules cover

most of the TiO2 particles. This composite assembly

becomes the ‘‘photoelectrode’ of DSSC. Some impor-

tant dyes used in DSSC are described later. The

counter electrode could be either sputter coated

platinum on to a glass or metal substrate. A suitable

electrolyte is sandwiched between the counter elec-

trode and the photoelectrode which acts as ‘‘charge

mediator’’ between them. The photo generated elec-

trons enter the DSSC through the counter electrode

after flowing through the external load. The whole

assembly is sealed to make the DSSC leak proof and

mechanically stable. Most commonly used redox

couple is I�=I�3 , though other types have also been

used which are briefly described in a later section.

The total charge transfer mechanism in the DSSC is a

multi step phenomenon as shown in Fig. 7 and described

below:

Step 1 Dye (D) absorbs spectral light and goes to the

excited state D*:

TiO2=D + hm! TiO2=D* ð14Þ

Step 2 The energy levels of chosen dye are such that the

excited dye D* is able to inject electrons from the

excited state rapidly to the conduction band and,

in turn, dye gets ‘‘oxidized’’ to D?:

TiO2=D*! TiO2=Dþ + e�=TiO2 ð15Þ

Step 3 These injected electrons (e-), after flowing

through the external circuit, return back into the

DSSC—system through the counter electrode and

reduce the ‘‘oxidized component’’ of the redox

couple ½RC]þox :

½RC]þox + e�=CE ! ½RC] ð16Þ

Step 4 The oxidized dye D? gets regenerated to D by

receiving an electron from the reduced

component of the redox couple, i.e., [RC]

TiO2=Dþ + [RC] ! TiO2=½RC]þox ð17Þ

After step 4, the complete regeneration of DSSC

(inclusive of TiO2, dye and redox couple) has taken place

and the cell is ready for the next cycle of operation. From

the above discussion, it is clear that the over all efficiency

of DSSC’s heavily rests on the following:

1. Suitable dye sensitizer

2. Proper redox electrolyte

Therefore, much of the recent research is focussed on

these aspects. Some important developments are described

below.

Some important Sensitizer Dyes for DSSC

Dye sensitization is a crucial step in the realization of high

efficiency DSSC’s. Some of the necessary requirements for

a good dye sensitizer are: (i) broad absorption spectrum,

particularly in the visible range. (ii) suitable ground and

excited energy states compatible with the semi-conductor

electrode energy levels (as well as that of redox mediator)

for efficient energy transfer to and from it as shown in

Fig. 7. (iii) ability to form well contacted interface with the

semi-conductor surface. (iv) should regenerate itself via

electron donation from the redox electrolyte (or from the

hole conductor being used as mediator in a special type

‘‘solid state DSSC’’ described later). (v) non-Toxic (vi)

long stability. It should be able to sustain about 108 turn-

over cycles corresponding to about 20 years of exposure to

natural light [32].

Most popular dye sensitizers for DSSC are mostly based

on Ruthenium Complexes [6, 31, 33–38] shown in Fig. 8,

in particular Dye coded as N3–dye [cis-di (isothiocyanoto)

bis (2,20 bipyridyl-4,40 dicaxboxylate) ruthenium (II)]. The

N3 dye suffers from the drawback that it lacks absorption in

Fig. 7 Schematic representation of energy levels of TiO2 and dye

along with charge transfer amongst photoexcited dye, TiO2, redox

couple mediator and the counter electrode

Proc. Natl. Acad. Sci. Sect. A Phys. Sci. (January–March 2012) 82(1):5–19 11

123

the red spectral region. Nonetheless, this still continues to

be one of the best dye sensitizer and many studies are

reported on DSSC’s with efficiencies in the range 5–10%

using N3 dye.

A new ruthenium dye giving DSSC’s with efficiency

greater than 10% is ‘‘black dye’’, coded as N–749,[tri(cy-

anoto)–2,20,200,ter-pyridyl-4,40,400,tricarboxylate) ruthenium

(II)]. The highest efficiency of 11.1% has already been

achieved. Apart from ruthenium charge- transfer com-

plexes, some metal porphyrins like chlorophyll, Zinc

metalloporphyrins, etc. have also been tried. However, the

efficiencies of DSSC’s obtained for these dyes are much

lower than those of cells which used ruthenium dyes (see

Goncalves et al. [34] and references cited therein).

Many ‘‘metal free organic dyes’’ like indoline, couma-

rin, polyene, cynanine, perylene, pthalocyanine, etc. have

also been successfully used in DSSC, with efficiencies

ranging from 1.5 to 9%. Li et al. [39], recently reported

DSSC using two newly synthesized metal free organic

dyes. These dyes contained thienothiophene or thiophene

segments as p–conjugation system (dyes were called as

D-ST and D–SS, respectively). The structure of these two

dyes along with some other important organic dyes [34, 39]

are shown in Fig. 9. The advantage of metal free organic

dyes over the Ru-complexes is that they, apart from having

high molar extinction coefficient, can be synthesized in

pure form at lower cost.

Electrolytes for DSSC

Electrolyte is a very critical part of DSSC’s since it acts as

a ‘‘mediator’’ to regenerate the dye at the photoelectrode

through the electronic charge received from the counter

electrode as explained earlier. Many redox couples have

been tried and the following couples have given promising

results:

(i) I�3 =I�: This has so far given the highest efficiency and

is most popular. Nearly 90% of studies reported in

literature on DSSC’s use this redox system. The

mechanism of electron charge transfer between redox

Fig. 8 Some common

ruthenium complex based dye-

sensitizers for DSSC (i) N-13

(red), (ii) Z-49 (black dye), (iii)Z-907, (iv) N-719, (v) K-19 and

(vi) K-60

12 Proc. Natl. Acad. Sci. Sect. A Phys. Sci. (January–March 2012) 82(1):5–19

123

couple I�=I�3 and the counter electrode as well as

between redox couple and dye is schematically

shown in Fig. 10. The diffusion of I�3 is generally

Grotthus-type charge transport mechanism [40].

(ii) Co(II)/Co(III): This is probably the most promising

new redox system because of some advantages over

Iodide-couple like its non-volatility and non-toxicity

[41, 42], (iii) (SeCN)2 (SeCN)- and (SCN)2/(SCN)-

[43, 44]. (iv) Br�3 =Br� [45]

For efficient charge transfer from or to the redox system,

the redox system should be capable of forming good

interface with the dye as well as with the counter electrode.

The different manners in which it has been achieved are

briefly described below.

Traditional Volatile Solvent Based Electrolytes

Most dyes are water sensitive. Therefore, instead of

aqueous solvent, non-aqueous solvents are used. Some of

the common non aqueous solvents used are acctonitrile,

valernitrile, 3-methoxy propionitrile, ethylene carbonate,

propylene carbonate, c-butyrolactone etc. In fact, many

other solvents can also be used so long they show good

redox couple solubility, low volatility, low toxicity and low

viscosity allowing fast diffusion of ions in the electrolyte.

The DSSC’s using volatile–organic solvents, inspite of

having high efficiency suffer from the problems of long

term stability (solvent evaporates), leakage and sealing–

packaging problems. To overcome some of these difficul-

ties, ionic liquids are increasingly being used as solvent.

Ionic Liquid Based Electrolytes

Ionic liquids (IL) are a new class of material [46], which

are self-dissociating and composed of large organic cations

and inorganic–organic anions of varying sizes. This leads

to high ionic conductivity in IL. Apart from high conduc-

tivity, the IL’s have low vapour pressure, high thermal and

chemical stability, good solubility for many organic–inor-

ganic compounds and large electrochemical window. This

has resulted in their application in many electrochemical

solar cells, including DSSC. Papageorgiou et al. [40] were

the first to develop a DSSC with ionic liquid 1-methyl-3-

hexyl imidazolium iodide (MHIm). They reported long

Fig. 9 Some important metal

free organic dyes used in DSSC

(i) indolin, (ii) coumarin, (iii)polyene, (iv) dyes with

thienothiophene and thiophene

segment, D-ST and D-SS

Fig. 10 Mechanism of electron exchange involving redox couple

mediator I2/I32

Proc. Natl. Acad. Sci. Sect. A Phys. Sci. (January–March 2012) 82(1):5–19 13

123

term stability of this cell. Since then, many 1,3-dialky-

limidazolium cation based ionic liquid have been used.

(See for example [6, 34]). The general molecular structure

of dialkylimidazolium based ionic liquid is given in

Fig. 11.

By changing R1 and R2, one can obtain a variety of ionic

liquids. It may be noted that if the cation of the ionic liquid

is I�; SCN�; SeCN�; (BCN4Þ�, etc., the ionic liquid as

electrolyte serves dual role of acting as redox couple source

(I�; SCN�, etc.) as well as solvent. The high viscosity of

IL’s, coupled with their negligible volatility, reduces the

sealing problem considerably. However, the high viscosity

tends to reduce ion diffusivity and hence, slow charge

transfer for dye regeneration. This decreases the over all

efficiency of the DSSC’s.

Many approaches have been used for increasing the Voc,

Jsc and the efficiency of ionic liquid based DSSC’s. Some

of the important approaches adopted are:

(i) Use of low viscosity ionic liquids has been proposed

because it is expected that this will increase the ion

diffusivity which, in turn, will enhance the dye –

regeneration rate. Most of the Imidazolium Iodides

have high viscosity. As mentioned earlier, iodides

offer preferred redox system because of its good

charge transfer behaviour. To overcome the diffi-

culty posed by the high viscosity of imidazolium

iodides, it has been mixed with low viscosity

imidazolium based IL’s with anions other than

iodide (like SCN�; SeCN�; N(CN)�2 ; and BF�4 ,

etc.) [6, 47] giving binary and ternary mixtures. A

DSSC using a ternary mixture of ionic liquid has

recently been reported with efficiency as high as

8.2% [45].

(ii) Some inorganic, e.g., CuI or organic additives, e.g.,

tetra-butylpyridine,N-alkylbenzimidazoles, guanidi-

nium thiocyoanate, etc. have also been tried. (see [6]

and references cited therein). It is believed that the

additive cations are adsorbed to the porous photo-

electrode of TiO2 shifting the conduction band edge

of the semiconductor.

(iii) Non-imidazolium based ionic liquids have also been

used like sufonium [49], ammonium [50], guanidinium

[51], phosphoniun [52], or tetrahydrothiopheniun [53].

The former four showed low efficiencies but for the

latter the efficiency seems promising.

(iv) An interesting modification to the DSSC’s has been

made in which the solar cell is irradiated both from

the front side (see Fig. 6) as well as from the back

side viz. counterelectrode side. Such a cell has been

termed as ‘‘Bifacial dye-sensitized solar cell’’ (see

for example [54] and references cited therein). In

‘‘front-side alone’’ type DSSCs, most of the radiation

is absorbed in the first few layers of dye-sensitized

TiO2 while in bifacial cells, the solar radiation from

rear side also contribute to the photo-effect in the

remaining back portion of TiO2 photo-electrode

resulting in enhanced efficiency. This has been

attained by a minor design change by pressing

porous silica on the counterelectrode side between it

and the electrolyte.

Inspite of the success achieved in developing high effi-

ciency DSSC’S using electrolyte containing IL’s, the

problem of leakage in cells stored or used for long time

persisted. Two approaches viz. using ‘‘Ionic gel’’ and ‘‘Ion-

conducting polymer electrolyte’’ are being extensively

investigated which yield quasi-solid state and all solid state

DSSC’s. These are briefly described below.

Quasi-solid State DSSC’s

Using Ion Gels as Electrolytes

The IL-electrolytes used in DSSC’s have been converted

into quasi-solid state ionic gels using two main approaches:

(i) Use of inorganic materials in nano and micron sizes

to produce thickened gel-like composite material is

very common. Many dispersed materials have been

tried like SiO2 [55–57], carbon nano-tubes, carbon

black [58], a-ZrP [59], etc. DSSC efficiencies *5 to

6% could be obtained, particularly with SiO2 nano-

composites.

(ii) The other approach is to use organic gelators. Some

gelators are: (PVP ? different dicarboxylic acids)

[60], PVdF-HFP [61], PEG [62], low molecular

weight gelators [63], natural polysaccharide agarose

[64], cross-linked gelataion agents like NMBI (1-N-

methylbenzimidazole) [65], etc.

In general, the efficiency of DSSC’s with gel-electrolyte

was less than those having IL’s without gelators. This is

understandable because the I�=I�3 diffusion in gels will be

slower than in pure IL’s resulting in slower charge transfer

kinetics. A comparison of short circuit current (Jsc), open

circuit voltage (Voc) and efficiency (g) for DSSC’s using

gelled (obtained by adding by PVDF-HFP) and non-gelled

electrolyte viz. ionic liquid, 1-ethyl-3-methylimidazolium

N+NR2

X-

R1

Fig. 11 Structure of di-alkyl imidazolium cation (and X- anion)

based ionic liquid

14 Proc. Natl. Acad. Sci. Sect. A Phys. Sci. (January–March 2012) 82(1):5–19

123

bis(trifluoromethane sulphonyl)imide, (EMIm-TFSI) ?

[LiI ? tetrabutyl pyridinium] is given in Table 1 along

with DSSC’s using organic volatile solvent (methoxy-

acetonitrile) instead of IL.

Large area (100 mm 9 100 mm) DSSC’s using ion gels

have already been demonstrated which promises their

commercial viability [58, 66].

DSSC’s Using Plasticized Polymer Electrolyte

Membrane

As discussed earlier, the polymer electrolytes have been

tried earlier as a replacement of liquid electrolyte in many

studies on conventional PESC’s [25, 27, 29]. Similar

modified membranes have been used recently in DSSC’s.

Polyethylene oxide (PEO) doped with a variety of

salts have been studied for quite a long time as

Liþ; Naþ; Kþ; Hþ , etc. conductors [67–72] for applica-

tion in fuel cells and polymer batteries. Most of these were

conducting through cations as well as anions. The mobil-

ities of iodide anion in PEO ? NH4I and SCN- in Poly-

ethylene Succinate ? NH4SCN systems were so high that,

given proper aggregating sites, they grew into large size

fractals [73, 74]. There are a large number of studies on

polymer electrolytes (see [67] for earlier work). The rele-

vant point with reference to DSSC’s is that the anions (like

iodide ions) are also mobile which can be made to act as

charge transfer ‘‘mediator’’ between the dye and the

counter-electrode.

De Paoli et al. [75–79] has reported many studies on

DSSC’s using membranes of PEO-based elastomer (viz.

polyethylene oxide-co-epichlorohydrin) doped with

NaI ? I2. They have integrated many such cells to fabri-

cate a ‘‘Solar Module’’ of *8 V. Kang et al. [80] devel-

oped a DSSC with polymer electrolytes obtained by

blending PEO, polysiloxanes and a series of quaternary

ammonium iodides (PS-NQAS). Many plasticizers (e.g.

propylene carbonate, PC; c-butyrolactone, c-BL etc.) were

also added to enhance the conductivities. They reported

efficiency of *1.39%. Recently effort has shifted to much

simpler systems like PEO ? (KI ? I2), PEO ? (NaI ? I2)

etc. containing a variety of ionic liquids to enhance

conductivity as well as to obtain plasticization effect [81–

84]. The efficiencies attained with all types of polymer

electrolyte based DSSC are still much lower than those

which use ‘‘gel electrolytes’’ or ‘‘organic solvent electro-

lytes’’. However, the flexible nature of polymer electrolytes

and associated avoidance of the problem of leakage,

packaging etc. still make them a promising alternative for

long term applications.

All Solid State DSSC’s by Replacing Electrolyte with p-

Semiconductor

This is a radical and fundamental departure from the con-

ventional DSSC’s. As explained earlier, an electron from

valance band of dye, after photoexcitation, goes to the

conduction band which gets transported through the TiO2

electrode and subsequently through the external circuit

goes to the counter electrode. To regenerate the dye,

electron transfer from the electrolyte to dye occurs which,

in turn, is regenerated by receiving the electron from the

counter electrode. In the new concept of all solid state

DSSC’s, the electrolyte is replaced by p-semiconductor (as

shown in Fig. 12).

The photoexcited dye is regenerated by hole transfer

(from Dye valance band) to the p-semiconductor through

which it hops and ultimately meets the electron from the

counter electrode side. The above mechanism is schemati-

cally given in Fig. 13). However, the p-semiconductors to be

used in such DSSC’s have to satisfy some basic criteria viz.

(a) The p-semiconductor should be such that it can be

deposited easily into the porous composite (TiO2 ?

dye) photo electrode by low temperature processes

without removing the dye layer over the TiO2.

Table 1 Comparison of dye sensitized solar cell parameters with

iodine couple in normal organic volatile solvent, IL, Ionic gel and

Ionic gel sheet (Matsui et al. [66])

Iodine redox couple solvent Jsc (mA/cm2) Voc (mV) g (%)

Volatile 14.6 750 6.7

Ionic liquid 12.1 640 4.5

Gel 9.6 640 3.8

Ion gel sheet 8.0 480 2.4

Fig. 12 Schematic representation of the basic assembly of all solid

state DSSC in which p-semiconductor is responsible for charge

mediation and dye-regeneration

Proc. Natl. Acad. Sci. Sect. A Phys. Sci. (January–March 2012) 82(1):5–19 15

123

(b) Since hole transfer from the valance band of dye has

to take place to the p-semiconductor, the valance

band of the semiconductor should lie above the dye

valance band as shown in Fig. 13.

Both inorganic and organic p-semiconductors have been

used. DSSC’s using CuI, CuSCN and NiO have been

studied. Tennakone et al. [85] were first to report the solid

state DSSC using p-CuI. The same group later reported

[86] enhanced efficiency for such cells with efficiency as

high as 2.6%. The stabilities were low because CuI tends to

be oxidized under illumination. It has been reported that by

incorporating small quantities (1 mM) of thiocyanates like

1-methyl-3-ethylimidazolium thiocyanate [87] or triethy-

lammoniumhydrothiocyanate [88] improves the stability.

Meng et al. [89] using the former additive reported a cell

with efficiency *3.8% which was stable for nearly

2 weeks.

Cells using CuSCN [90–93] and NiO [94] have also

been reported but the efficiency of these cells is still lower

than those using CuI but the stability is comparative or

sometimes even better.

Gratzel et al. [95] was first to report a DSSC with p-type

‘‘organic semiconductor’’. The material used by them was

OMeTAD [2,20,7,70-tetrakis (N, N-di-p-methoxyphenyl-

amine) 9.90-spiro-bifluorene]. They found that efficiency

was very low unless it is doped with some ionic salts (say

lithium salts) but it needed co-solvents. Many co-solvents

have been tried [96, 97] to yield cells with efficiency of

*3%. Some novel supramolecules have also been reported

[98, 99] which dissolve Li?-ions without using co-solvents.

Many other hole transporting materials like PEDOT(Poly-

ethylenedioxythiophene, polyaniline, polypyrrole etc.)

have also been tried but the cells so obtained have com-

paratively low efficiency (see [30] and references cited

therein).

In conclusion, the science and technology of DSSC have

come to a stage where it can be stated that DSSC’s are a

safe bet as a future replacement of conventional semicon-

ductor junction solar cells because of its low cost, easy

processibility and possibility of scaling up.

Acknowledgments Thanks are due to National Academy of Sci-

ences, India for the award of NASI-Platinum Jubilee Senior Scientist

Fellowship. Thanks are also due to S. K. Chaurasia, M. P. Singh, Y.

L. Verma and Abhishek Kumar Gupta for help in preparing the

manuscript.

References

1. Sze SM (1974) Physics of semiconductor devices. Wiley, New

York

2. Fujishima A, Honda K (1972) Electrochemical photolysis of

water at a semiconductor electrode. Nature 238:37–38

3. Gerischer H (1975) Electrochemical photo and solar cells: prin-

ciples and some experiments. Electroanal Chem Interface Elect-

rochem. 58:263–274

4. Chandra S (1985) Photoelectrochemical solar cells. Gordon &

Breach, New York

5. O’Regan B, Gratzel (1991) A low cost higher efficiency solar cell

based on dye-sensitized colloidal TiO2 film. Nature 353:737–740

6. Grolov M, Kloo L (2008) Ionic Liquid electrolyte for dye-sen-

sitized solar cell. Dalton Trans 2655–2666

7. Heller A, Miller BI, Chu SS, Lee YT (1979) 7.3% Efficient thin

film, polycrystalline n-gallium arsenide semiconductor liquid

junction solar cell. J Am Chem Soc 101:7633–7634

8. Noufi R, Tench D (1980) High efficiency GaAs photoanodes.

J Electroche Soc 127:188–190

9. Anderson WW, Chai YG (1976) Becquerel effect solar cell.

Energy Conversion 15:85–94

10. Ellis AB, Kaiser SW, Wrighton MS (1976) Optical to electrical

energy conversion. Characterization of cadmium sulfide and

cadmium selenide based photoelectrochemical cells. J Am Chem

Soc 98:6855–6866

11. Nakatani K, Matsudaira M, Tsubomura H (1978) Photoanodic

behaviour of n-type cadmium sulfide in acetonitrile solutions

containing Iodine ion. J Electrochem Soc 125:406–409

12. Heller A, Schwartz GP, Vadimasky RG, Menezes S, Miller B

(1978) Output stability of n-CdSe/Na2S-S-NaOH/C solar cells.

J Electrochem Soc 125:1156–1160

13. Wrighton MS, Morse DL, Ellis AB, Ginley DS, Abrahamson HB

(1976) Photoassisted electrolysis of water by ultraviolet irradia-

tion of an antimony doped stannic oxide electrode. J Am Chem

Soc 98:44–48

14. Ellis AB, Kaiser SW, Wrighton MS (1976) Visible light to

electrical energy conversion. Stable cadmium sulfide and cad-

mium selenide photoelectrodes in aqueous electrolytes. J Am

Chem Soc 98:1635–1637

15. Ellis AB, Kaiser SW, Wrighton MS (1976) Semiconducting

potassium tantalate electrodes. Photoassistance agents for the

efficient electrolysis of water. J Phys Chem 80:1325–1328

16. Heller A, Miller B (1980) Some recent progress in semiconduc-

tor-liquid junction solar cells. Electrochimca Acta. 25:29–41

17. Muller N, Tenne R, Cahen D (1981) Photoelectrochemical per-

formance of the n-CdSe/aqueous polysulfide system at room- and

sub-zero ambient temperatures. J Electroanal Chem 130:373–379

18. Owen JR (1977) A proposed cadmium–selenium film-electrolyte

Schottky barrier solar cell. Nature 267:504–505

Fig. 13 Relative energy band positions and charge transfer mecha-

nism in an all solid state DSSC

16 Proc. Natl. Acad. Sci. Sect. A Phys. Sci. (January–March 2012) 82(1):5–19

123

19. Gerrard WA, Owen JR (1977) A stable photo-electrochemical

solar cell employing a CdSe photoanode. Mat Res Bull 12:

677–1980

20. Hodes G, Cahen D, Manassen J, David M (1980) Painted,

polycrystalline thin film photoelectrodes for photoelectrochemi-

cal solar cells. J Electrochem Soc 127:2252

21. Pandey RK, Sahu SN, Chandra S (1996) Handbook of Semi-

conductor Electrodeposition. Marcel Dekker, New York

22. Chandra S, Singh DP, Srivastava PC, Sahu SN (1984) Electro-

deposited semiconducting molydenium selenide films, optical,

electrical, electrochemical, and photoelectrochemical solar cell

studies. J Phys D: App Phys 17:2125–2138

23. Chandra S, Pandey RK (1980) A photoelectrochemical cell for

solar energy conversion using electrocodeposited CdSe films.

Phys Stat Solidi (a) 59:787–794

24. Chandra S, Khare N (1987) Electro-deposited gallium arsenide

film: II. Electrochemical and photoelectrochemical solar cell

studies. Semicond Sci Techno 2:220–225

25. Pandey RK, Kumar SR, Rooz AJN, Chandra S (1991) Photo-

electrochemical and solar cell studies on n-CdSe films grown by

repeated electrodeposition cycles. Thin Solid Films 200:1–9

26. Licht S (2001) Multiple band gap semiconductor/electrolyte solar

energy conversion. J Phys Chem B 105:6281–6294

27. Skotheim TA (1981) A tandem photovoltaic cell using a thin film

polymer electrolyte. Appl Phys Lett 38:712–714

28. Skotheim T, Lundstrom (1982) Solid polymer electrolyte pho-

tovoltaic Cell. J Electrochem Soc 129:894–896

29. Bhattacharya B, Upadhyaya HM, Chandra S (1996) Photoelect-

rochemical studies of an ion conducting polymer (PEO)/semi-

conductor (Si) junction. Solid State Comm 98:633–638

30. Li B, Wang L, Kang B, Wang P, Yong Q (2006) Review of recent

projects in solid state dye-sensitized solar cells. Solar Energy

Mater Solar Cells 90:549–573

31. Lenzmann FO, Kroon JM (2007) Recent advances in dye-sensi-

tized solar cells. Adv Optoelect 2007, Art. ID 65073, p 10. doi:

10.1155/2007/65073

32. Gratzel M (2003) Dye-sensitised solar cells. J Photo Chem,

Phtobobiol C Photochem Rev 4:145–153

33. Dong FT, Dai S & Wang KJ. Review of recent progress in dye-

sensitized solar cells. Adv Optoelect 2007, Art. ID. 75384, p 13.

doi:10.1155/2007/75384)

34. Goncalves LM, Bermudez VZ, Ribeiro HA, Mendes AM (2008)

Dye-sensitized solar cells: a safe bet for the future. Energy

Environ Sci 1:655–667

35. Nazeeruddin MK, Kay A, Rodico I, Humphry-Baker R, Muller E,

Liska P, Vlachoponloss N, Gratzel M (1993) Conversison of light

to electricity by cis-X2 (2,20-bipyridyl-4,40 dicarboxylate) ruthe-

nium (II) Charge transfer sensitizers (X = Cl-, Br-, I-, CN-,

SCN-) on nono-crystalline TiO2 electrode. J Am Chem Soc

115:6382–6390

36. Nazeeruddin MK, Pechy P, Renouord T, Zakeeruddin SM,

Humphry-Baker R, Comte P, Liska P, Cevey L, Costa E, Shk-

lover V, Spiecia L, Deacon GB, Bignozzi CA, Gratzel M (2001)

Engineering of efficient panchromatic sensitizers for nanocrys-

talline TiO2-based Solar Cells. J Am Chem Soc 123:1613–1624

37. Chiba Y, Islam A, Watanabe Y, Komiya R, Koide N, Han L

(2006) Dye-sensitized solar cells with conversion efficiency of

11.1%. Jpn J Appl Phys 45:L638–L640

38. Nazeeruddin MK, Angelis FD, Fantacci S et al (2005) Combined

experimental and DFT–TDDFT computational study of photo-

electrochemical cell ruthenium senstizers. J Am Chem Soc 127:

16835–16847

39. Li S, Jiang K, Shao K, Yang L (2006) Novel organic dyes for

efficient dye-sensitized solar cells. Chem Commun 2792–2794

40. Papageorgiou N, Athanassov Y, Armand M, Bonhote P, Pet-

tersson H, Azam A, Gratzel M (1996) The performance and

stability of ambient temperature molten salts for solar cell

applications. J Electrochem Soc 143:3099–3108

41. Sapp SA, Elliot CM, Contado C, Caramori S, Bignozzi CA

(2002) Substituted polypyridene complex of cobalt (II/III)as

efficient electron transfer mediators in dye sensitized solar cell.

J Am Chem Soc 124:11215–11222

42. Nusbaumer H, Zakeeruddin SM, Moser JE, Gratzel M (2003) An

alternative efficient redox couple for the dye sensitized solar

system. Chem Eur Journ 9:3756–3763

43. Oskam G, Bergeron B, Meyer GJ, Searson PC (2001) Pseudo-

halogen for dye-sensitized TiO2 photoelecrochemical cells.

J Phys Chem B 105:6867–6873

44. Bergeron BV, Meyer A, Oskam G, Meyer GJ (2005) Dye-sen-

sitized SnO2 electrodes with iodide and pseudohalide redox

mediators. J Phys Chem B 109:937–943

45. Wang ZS, Sayama K, Sugihara H (2005) Efficient eusin Y dye-

sensitized solar cell containing Br-/Br3- electrolyte. J Phys Chem

B 109:22449–22455

46. Ohno H (ed) (2005) Electrochemical aspect of ionic liquids.

Wiley, New Jersey

47. Cao Y, Zhang J, Bai Y, Li R, Zakeeruddin SM, Gratzel M, Wang

P (2008) Dye sensitized solar cells with solvent free ionic liquid

electrolytes. J Phy Chem C 112:13775–13781

48. Bai Y, Cao Y, Zhang J, Wang M, Li R, Wang P, Zakeeruddin

SM, Gratzel M (2008) High performance dye-sensitized solar

cells based on solvent-free electrolytes produced from eutectic

melts. Nature Mater 7:626–630

49. Paulsson H, Hagfeldt A, Kloo L (2003) Molten and solid trial-

kylsulfonium iodides and their polyiodides as electrolytes in dye-

sensitized nanocrystalline solar cells. J Phys Chem B 107:

13665–13670

50. Santa - Nokki H, Busi S, Kallioinen J, Lahtinen M, Korppi

Tommola J (2007) Quaternary ammonium polyiodides as ionic

liquid/soft solid electrolytes in dye-sensitized solar cells. J Pho-

tochem Photobiol A: Chem 186:29–33

51. Wang P, Zakeeruddin SM, Gratzel M, Kantleher W, Mezger J,

Stoyanov EV, Scherr O (2004) Novel room temperature ionic

liquids of hexaalkyl substituted guanidinium salts for dye-

sensitized solar cells. Appl Phys A, Mater Sc Process 79:

73–77

52. Kunugi Y, Hayakawa H, Tsunashima K, Sugiya M (2007) Dye-

sensitized solar cells based on quaternary phosphonium ionic

liquids as electrolytes. Bull Chem Soc Jpn 80:2473–2475

53. Xi C, Cao Y, Cheng Y, Wang M, Jing X, Zakeeruddin SM,

Gratzel M, Wang P (2008) Tetratydrothiophenium based ionic

liquid for high efficiency dye-sensitized solar cell. J Phy Chem C.

112:11063–11067

54. Ito S, Zakeeruddin SM, Comte P, Liska P, Kuang D, Gratzel M

(2008) Bifacial dye-sensitised solar cells based on an ionic liquid.

Nature Photonics 2:693–698

55. Wang P, Zakeeruddin SM, Comte P, Exnar I, Gratzel M (2003)

Gelation of ionic liquid based electrolytes with silica nano par-

ticles for quasi solid state dye sensitized solar cells. J Am Chem

Soc 125:1166–1167

56. Zhao Y, Zhai J, Tan S, Wang L, Jiang L, Zhu D (2006) TiO2

micro/nano-composite structured electrodes for quasi-solid-state

dye-sensitized solar cells. Nanotechnology 17:2090–2097

57. Kim JH, Kang MS, Kim YJ, Won J, Park NG, Kang YS (2004)

Dye-sensitized nanocrystalline solar cells based on composite

polymer electrolytes containing fumed silica nanoparticles. Chem

Commun 1662–1663

58. Usui H, Matsui H, Tanabe N, Yanagida S (2004) Imporved dye

sensitized solar cells using ionic nono composite gel electrolysis.

J Photochem Photobiol A 164:97–101

59. Wang N, Lin H, Li J, Li X (2006) Improved quasi-solid dye-

sensitized solar cells by composite ionic liquid electrolyte

Proc. Natl. Acad. Sci. Sect. A Phys. Sci. (January–March 2012) 82(1):5–19 17

123

including layered a-zirconium phosphate. Appl Phys Letts

89:194104–194104-3

60. Kato T, Okazaki A, Hayase S (2006) Latent gel electrolyte pre-

cursors for quasi-solid dye sensitized solar cells. The comparison

of nano-particle cross-linkers with polymer cross-linkers. J Pho-

tochem Photobiol A 179:42–48

61. Wang P, Zakeeruddin S.M., Exnar I, Gratzel M (2002) High

efficiency dye-sensitized nanocrystalline solar cells based on

ionic liquid polymer gel electrolyte. Chem Commun 2972–

2973

62. Lee H, Kim W, Park S, Shin W, Jin S, Lee J, Han S, Jung K, Kim

M (2006) Effects of nanocrystalline porous TiO2 films on inter-

face adsorption of phthalocyanines and polymer electrolytes in

dye-sensitized solar cells. Macromol Symp 235:230–236

63. Kubo W, Kitamura T, Hanabusa K, Wada Y, Yanagida, S (2002)

Quasi-solid-state dye-sensitized solar cells using room tempera-

ture molten salts and a low molecular weight gelator. Chem

Commun 374–375

64. Suzuki K, Yamaguchi M, Kumagai M, Tanabe N, Yanagida S

(2006) Dye sensitized solar cells with ionic gel electrolyte

prepared from imidazolium salts and agarose. CR Chimie 9:

611–616

65. Wang L, Fang S-B, Lin Y (2006) Novel polymer electrolytes

containing chemically crosslinked gelators for dye-sensitized

solar cells. Polym Adv Technol 17:512–517

66. Matsui H, Okada K, Kawashima T, Ezure T, Tanabe N (2004)

Dye sensitized solar cells using ionic liquid based electrolytes.

Fujikura Technic Rev 48–52

67. Grey FM (1991) Solid polymer electrolytes: fundamental and

technological applications. VCH, New York

68. Hashmi SA, Kumar A, Maurya KK, Chandra S (1990) Proton

conducting polymer electrolyte I polyethelyne oxide ? NH4-

ClO4. J Phys D Appl Phys 23:1307–1314

69. Chandra A, Srivastava PC, Chandra S (1995) Ion transport

studies in PEO ? NH4I polymer electrolysis with dispersed

Al2O3. J Mater Sci 30:3633–3638

70. Hashmi SA, Chandra S (1996) Experimental investigation on a

sodium ion conducting polymer electrolyte base on polyethylene

oxide complexed with NaPF6. Mat Sci Eng B 34:18–26

71. Chandra A, Chandra S (1994) Mixed anion effect in polyethylene

oxide base sodium ion conducting polymer electrolytes. J Phys D:

Appl Phys 27:2171–2179

72. Singh PK, Chandra A (2002) Ion conducting polymer electrolyte

composites dispersed with ferroelectric Ba0.70Sr0.30TiO3 ceramic

powder. Natl Acad Sci Lett India 25:286–293

73. Chandra A, Chandra S (1994) Experimental observation of large

size fractal in ion conducting polymer electrolytes. Phys Rev B

49:633–636

74. Srivastava N, Chandra A, Chandra S (1995) Studies on dense

branched growth of (SCN)x and ion transpose in PEO: NH4SCN

polymer electrolyte. Phys Rev B 52:225–229

75. Nogueira AF, De Paoli MA (2000) A dye sensitized TiO2

photovolatic cell constructed with an elastomeric electrolyte.

Solar Energy Mater Soler Cells 61:135–141

76. Nogueira AF, Durrant JR, De Paoli MA (2001) Dye Sensitised

nanocrystalline solar cells employing a polymer electrolyte. Adv

Mate 13:826–830

77. De Freitas JN, Nogueira VC, Ito BL, Soto-Ovideo MA, Longo C,

De Paoli MA, Nogueira AF (2006) Dye-sensitized solar cells and

solar module using polymer electrolyte: stability and perfor-

mance investigations. Int J Photoenergy 2006, Art. ID. 75483,

pp 1–6. doi: 10.1155/IJP/2006/75483

78. Floves IC, De Freitas JN, Longo C, De Paoli MA, Winnischfer H,

Nogueira AF (2007) Dye-senstisized solar cells based on TiO2

nanotubes and a solid-state electrolyte. J Photochem Photobiol A:

Chem 189:153–160

79. Nogueira VC, Longo C, Nogueira AF, Soto-Oviedo MA, De

Paoli MA (2006) Solid state dye sensitized cell: Imporved per-

formance and stability using plasticized polymer electrolyte.

J Phochem Photobilol A: Chem 181:226–232

80. Kang J, Li W, Wang X, Lin Y, Xiao X, Fang S (2003)

Polymer electrolytes from PEO and novel quaternary ammo-

nium iodides for dye sensitised solar cells. Electrochim Acta

48:2487–2491

81. Singh PK, Kim KI, Lee JW, Rhee HW (2006) Polymer electro-

lyte with ionic liquid for DSSC application. Phys Status Solid

(a) 203:88–90

82. Singh PK, Kim KW, Rhee HW (2009) Quantum dot doped solid

polymer electrolyte for device application. Electrochem Comm

11:1247–1250

83. Singh PK, Kim KW, Rhee HW (2009) Ionic liquid (1-methyl

3-propyl imidazolium iodide) with polymer electrolyte for DSSC

application. Polymer Eng Sci 49:862–865

84. Singh PK, Kim KW, Nagarale RK, Rhee HW (2009) Preparation,

characterization and application of ionic liquid doped solid

polymer electrolyte membranes. J Phys D Appl Phys 42:

125101–125104

85. Tennakone K, Kumarasinghe AR, Wijayantha KGU, Sirimanne

PM (1995) A dye-sensitized nano-porous solid-state photovoltaic

cell. Semicond Sci Technol 10:1689–1693

86. Tennakone K, Perera VPS, Kottegoda IRM, Kumara GRRA

(1999) Dye-sensitized solid state photovoltaic cell based on

composite zinc oxide/tin (IV) oxide films. J Phys D Appl Phys

32:374–379

87. Kumara GRA, Konno A, Shiratsuchi K, Tsukahara J, Tennakone

K (2002) Dye-sensitized solid-state solar cells: use of crystal

growth inhibitors for deposition of the hole collector. Chem

Mater 14:954–955

88. Kumara GRA, Kaneko S, Okuya M, Tennakone K (2002) Fab-

rication of dye-sensitized solar cells using triethylamine hyd-

rothiocyanate as a CuI crystal growth inhibitor. Langumuir

18:10493–10495

89. Meng QB, Takahashi K, Zhang XT, Sutanto I, Rao TN, Sato O,

Fujishima A, Watanabe H, Nakamori T, Uragami M (2003)

Fabrication of an efficient solid-state dye-sensitized solar cell.

Langmuir 19:3572–3574

90. O’Regan B, Schwartz DT (1995) Efficient photo-hole injection

from adsorbed cyanine dyes into electrodeposited copper(I) thio-

cyanate thin films. Chem Mater 7:1349–1354

91. O’Regan B, Schwartz DT, Zakeeruddin SM, Gratzel M (2000)

Electrodeposited nanocomposite n–p heterojunctions for solid-

state dye-sensitized photovoltaics. Adv Mat 12:1263–1267

92. O’Regan B, Lenzmann F, Muis R, Wienke J (2002) A solid-state

dye-sensitized solar cell fabricated with pressure-treated P25–

TiO2 and CuSCN: analysis of pore filling and I–V characteristics.

J Chem Mater 14:5023–5029

93. Kumara GRRA, Konno A, Senadeera GKR, Jayaweera PVV,

De Silva DBRA, Tennakone K (2001) Dye-sensitized solar

cell with the hole collector p-CuSCN deposited from a solu-

tion in n-propyl sulphide. Solar Energy Mater Solar Cells 69:

195–199

94. Bandara J, Weerasinghe H (2005) Solid-state dye-sensitized solar

cell with p-type NiO as a hole collector. Solar Energy Mater Solar

Cells 85:385–390

95. Bach U, Lupo D, Comte P, Moser JE, Weissortel F, Salbeck J,

Spreitzer H, Gratzel M (1998) Solid state dye-sensitized meso-

porous TiO2 solar cells with high photon-to-electron conversion

efficiencies. Nature 395:583–585

18 Proc. Natl. Acad. Sci. Sect. A Phys. Sci. (January–March 2012) 82(1):5–19

123

96. Kruger J, Plass R, Cevey L, Piccirelli M, Gratzel M, Bach U

(2001) High efficiency solid-state photovoltaic device due to

inhibition of interface charge recombination. Appl Phys Letts

79:2085–2087

97. Kruger J, Plass R, Gratzel M, Matthieu HJ (2002) Improvement

of the photovoltaic performance of solid-state dye-sensitized

device by silver complexation of the sensitizercis-bis(4,40dicar-

box2,20bipyridine)-bis(isothiocyanato) ruthenium(II). Appl Phys

Letts 81:367–369

98. Park T, Haque SA, Potter RJ, Holmes AB, Durrant JR (2003) A

supramolecular approach to lithium ion solvation at nanostruc-

tured dye sensitised inorganic/organic heterojunctions. J Chem

Commun 11:2878–2879

99. Haque SA, Park T, Xu C, Koops S, Schulte N, Potter RJ, Holmes

AB, Durrant JR (2004) Interface engineering for solid-state dye-

sensitized nanocrystalline solar cells: the use of ion-solvating

hole-transporting polymers. Adv Funct Mater 14:435–440

Proc. Natl. Acad. Sci. Sect. A Phys. Sci. (January–March 2012) 82(1):5–19 19

123