Conducting gel electrolytes with microporous structures for efficient quasi-solid-state dye-sensitized solar cells

Accepted Manuscript Conducting gel electrolytes with microporous structures for efficient quasi−solid−state dye−sensitized solar cells Shuangshuang Yuan, Qunwei Tang, Benlin He, Liangmin Yu PII:

S0378-7753(14)01628-0

DOI:

10.1016/j.jpowsour.2014.10.019

Reference:

POWER 19939

To appear in:

Journal of Power Sources

Received Date: 18 August 2014 Revised Date:

30 September 2014

Accepted Date: 1 October 2014

Please cite this article as: S. Yuan, Q. Tang, B. He, L. Yu, Conducting gel electrolytes with microporous structures for efficient quasi−solid−state dye−sensitized solar cells, Journal of Power Sources (2014), doi: 10.1016/j.jpowsour.2014.10.019. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Conducting gel electrolytes with microporous structures for efficient quasi−solid−state dye−sensitized solar cells

Shuangshuang Yuan1,2, Qunwei Tang1,2*, Benlin He2, Liangmin Yu1,3* Key Laboratory of Marine Chemistry Theory and Technology, Ministry of Education, Ocean

RI PT

1

University of China, Qingdao 266100, P.R. China;

Institute of Materials Science and Engineering, Ocean University of China, Qingdao 266100,

SC

2

Shandong Province, P.R. China;

Qingdao Colobrative Innovation Center of Marine Science and Technology, Ocean University of

M AN U

3

China, Qingdao 266100, P.R. China;

*E-mail address: [email protected]; [email protected]; Tel/Fax: 86−532−66781690;

Conducting

bromide/polyaniline

gel

electrolytes

from

poly(acrylic

acid)−cetyltrimethylammonium

and

poly(acrylic

acid)−cetyltrimethylammonium

TE D

Abstract:

(PAA−CTAB/PANi)

EP

bromide/polypyrrole (PAA−CTAB/PPy) are synthesized under driving forces of both osmotic pressure and capillary force within microporous PAA−CTAB matrix. The as−synthesized

AC C

PAA−CTAB/PANi or PAA−CTAB/PPy can extend the reduction reaction of triiodides from gel electrolyte/Pt counter electrode interface to both interface and three−dimensional framework of conducting gel electrolyte due to the electrical conduction of PANi or PPy toward reflux electrons (electrons from external circuit to Pt counter electrode). The enhanced kinetics for triiodides → iodide conversion is promising in elevating photovoltaic performances of quasi−solid−state dye−sensitized solar cells (DSSCs). Driving forces by both osmotic pressure across PAA−CTAB matrix and capillary force presenting in micropores can elevate the loading of PANi or PPy 1

ACCEPTED MANUSCRIPT incorporated liquid electrolyte in per unit volume, leading to further enhancement in charge transfer and electrocatalytic activity. The total power conversion efficiencies of 7.11% and 6.39% are recorded in the solar cells with PAA−CTAB/PANi and PAA−CTAB/PPy electrolytes under one sun irradiation, respectively, whereas it is 6.07% for the cell device with pure PAA−CTAB gel

RI PT

electrolyte. Electrical and electrochemical characterizations reveal that the electrical conduction and electrocatalytic performances have been significantly enhanced by incorporating electrical

SC

conducting PANi or PPy into microporous PAA−CTAB matrix. The concept opens a new approach of fabricating efficient polymer gel electrolytes for robust quasi−solid−state DSSC applications.

structure, Hydrogel, Conducting polymer

1. Introduction

M AN U

Keywords: Quasi−solid−state dye−sensitized solar cell, Polymer gel electrolyte, Microporous

TE D

Dye−sensitized solar cells (DSSCs) are promising solutions to energy crisis and environmental damage [1−5]. A typical DSSC device is composed of three components: dye−sensitized TiO2

EP

anode, Pt counter electrode, and electrolyte containing iodide/triiodide (I−/I3−) redox couples [6]. To date, an impressive light−to−electric power conversion efficiency of 12.3% has been measured from

AC C

liquid electrolyte based DSSC [7]. However, the leakage of liquid electrolyte as well as evaporation of organic solvent such as acetonitrile has been one of the main burdens in the commercial application of DSSCs [8,9]. By addressing this issue, full−solid−state electrolytes are preferred in enhancing the long−term stability [10,11]. Considering the low charge transfer kinetics and poor contact of electrolyte with electrodes, unsatisfactory conversion efficiencies are determined from their devices. Polymer gel electrolytes can combine the stability of full−solid−state electrolytes and rapid charge−transfer of liquid electrolyte [12−15]. The interconnected three−dimensional (3D) 2

ACCEPTED MANUSCRIPT frameworks of gel matrices provide space for redox couples storage and superhighways for charge transport. Traditional route of designing gel electrolytes is the direct imbibition of liquid electrolyte by amphiphilic gel matrix [16,17]. The imbibition kinetics is generally controlled by Flory theory [18],

RI PT

in which the driving force of liquid electrolyte is osmotic pressure across the gel matrix. The imbibition process is accompanied with the swollen of molecular chains in liquid electrolyte. An

SC

imbibition equilibrium can be obtained when the osmotic pressure is zero. Although the liquid electrolyte and organic solvent can be well sealed in the 3D framework of gel electrolyte, the poor

M AN U

contact between gel electrolyte/Pt counter electrode can increase loss of reflux electrons (electrons from external circuit to Pt counter electrode) in gel electrolyte/Pt interface. In our previous work [19−21], electron−conducting polymers such as polyaniline (PANi) or polypyrrole (PPy) have been incorporated into 3D framework of insulating gel matrices to fabricate conducting gel electrolytes

TE D

that can conduct both ions and electrons. The reduction reaction of I3− species in gel electrolyte/Pt counter electrode is extended to both interface and whole 3D framework of conducting gel

EP

electrolytes because of the promising electrocatalytic activity of PANi or PPy toward I3−. Significant enhancement in electron transfer kinetics and photovoltaic performances are achieved.

AC C

However, the driving force of liquid electrolyte by conducting gel matrices is still the osmotic pressure. Considering the potential enhancement in charge−transfer ability of I−/ I3− within gel matrix and electrocatalytic activity toward I3−, it is a prerequisite to elevate the content of liquid electrolyte and electron−conducting polymers in per unit volume of insulating gel matrix, respectively. In the current work, we propose a new approach of designing efficient conducting gel electrolytes by imbibing PANi (or PPy) and liquid electrolyte into microporous poly(acrylic 3

ACCEPTED MANUSCRIPT acid)−cetyltrimethylammonium bromide (PAA−CTAB) matrix. The imbibition kinetics is governed by both osmotic pressure and capillary force across the gel matrix and micropores, respectively. Results indicate that the electrical, electrochemical and therefore photovoltaic performances have

RI PT

been markedly enhanced in comparison with electron-insulating PAA−CTAB gel electrolyte.

2. Experimental

SC

2.1. Synthesis of PAA−CTAB matrix

PAA−CTAB matrix was synthesized by modifying the procedures: 1.0 g of CTAB and 10 g of

M AN U

acrylic acid were dispersed in 15 ml of deionized water. Subsequently, initiator ammonium peroxydisulfate (APS) (mass ratio of APS to acrylic acid was 0.0225) and crosslinker N,N’−methylene bisacrylamide (NMBA) (mass ratio of NMBA to acrylic acid was 0.001) were added to the mixed solution. Under a nitrogen atmosphere, the acrylic acid−CTAB monomers

TE D

would be initiated by the thermal decomposition of APS for forming PAA−CTAB prepolymer. With the proceeding of polymerization, the viscosity increased gradually. When the viscosity of the

EP

PAA−CTAB prepolymers reached around 180 mPa s−1 (The viscosity of the reagent was measured using a Haaker ReoStress RS75 rheometer at a shear rate of around 100 s−1), the reagent was poured

AC C

into a petri dish and cooled to room temperature until the formation of an elastic gel. After rinsing with excess deionized water, the samples were vacuum dried at 80 ◦C for more than 12 h. 2.2. Preparation of PANi and PPy 0.592 ml of aniline was dissolved in 20 ml of 1 M HCl aqueous solution to obtain a homogeneous mixture. 20 ml of 0.125 M APS aqueous solution was dipped in the above mixture within 30 min. The polymerization reaction was carried out at 0 oC. After 3 h, the resultant reactant was rinsed by 1M HCl aqueous solution, filtrated, and finally vacuum dried at 60 oC for 24 h. 4

ACCEPTED MANUSCRIPT 1.000 ml of pyrrole monomer was dipped in an aqueous solution of FeCl3 containing 7.788 g of FeCl3·6H2O and 58 ml of deionized water. Under vigorous agitation, the polymerization reaction was carried out at 5 oC for 24 h. After being rinsed by aqueous solution, filtrated, and finally vacuum dried at 60 oC for 24 h, the resultant PPy powders were obtained. The conductivities of

RI PT

resultant PANi and PPy were in the scale of ~0.5 and ~10 S cm-1, respectively. 2.3. Synthesis of conducting gel electrolytes

SC

The microporous PAA−CTAB matrices were prepared by immersing dried PAA−CTAB matrices in deionized water for 72 h to reach their swelling equilibrium, and subsequently

M AN U

freeze−dried under vacuum at −60 °C for 72 h. After that the matrices were immersed in liquid electrolyte consisting of redox electrolyte and PANi or PPy at room temperature for 20 days to reach absorption equilibrium. The contents of PANi and PPy in liquid electrolyte were 2.10 and 1.38 g L−1, respectively. A redox electrolyte consisted of 100 mM tetraethylammonium iodide, 100

TE D

mM tetramethylammonium iodide, 100 mM tetrabutylammonium iodide, 100 mM NaI, 100 mM KI, 100 mM LiI, 50 mM I2, and 500 mM 4−tert−butyl−pyridine in 50 ml acetonitrile.

EP

2.4. Assembly of quasi−solid−state DSSCs

TiO2 colloid was prepared according to a sol−hydrothermal method [22,23] and a layer of TiO2

AC C

nanocrystal anode film with a thickness of 10 µm and an area of 0.25 cm2 was prepared by coating TiO2 colloid onto conducting glass using a screen printing technique, followed by sintering in air at 450 ºC for 30 min. Subsequently, the TiO2 film was soaked in a 0.3 mM N719 [cis−di(thiocyanato)−N,N’−bis(2,2’−bipyridyl−4−carboxylic

acid−4−tetrabutylammonium

carboxylate, purchased from Solaronix, SA, Switzerland] ethanol solution for 24 h to uptake N719 dye for the fabrication of dye−sensitized TiO2 photoanode. Quasi−solid−state DSSC from pure PAA−CTAB, or PAA−CTAB/PANi, or PAA−CTAB/PPy gel electrolyte was fabricated by 5

ACCEPTED MANUSCRIPT sandwiching a slice of gel electrolyte with thickness of around 500 µm between dye−sensitized TiO2 anode and a standard Pt counter electrode (300~400 µm in thickness, purchased from Dalian HepatChroma SolarTech Co., Ltd). A black mask with an aperture area of around 0.25 cm2 was applied on the surface of cell device to avoid stray light.

RI PT

2.5. Photovoltaic measurements

The photocurrent−voltage (J−V) curves of the DSSCs were recorded on an electrochemical

SC

workstation (CHI600E) under irradiation of a simulated solar light from a 100 W xenon arc lamp in ambient atmosphere. The incident light intensity was calibrated using a FZ−A type radiometer from

M AN U

Beijing Normal University Photoelectric Instrument Factory (calibrated by a standard Si solar cell) to control it at 100 mW cm−2 (AM 1.5). Each DSSC device was measured five times to eliminate experimental error and a compromise J−V curve was employed. 2.6. Characterizations

TE D

The morphology of the microporous PAA−CTAB matrix was captured with a Zeiss Ultra plus field emission scanning electron microscopy (FESEM). Fourier transform infrared spectrometry

EP

(FTIR) spectra were recorded on a Vertex 70 FTIR spectrometer (Bruker). The ionic conductivity of gel electrolyte was measured using a pocket conductivity meter (DSSJ−308A, LeiCi Instruments)

AC C

by filling gel electrolyte into interspace between two electrodes. The instrument was calibrated with 0.01 M KCl aqueous solution prior to experiments. Tafel polarization curves of the symmetrical dummy cells were measured by CHI660E electrochemical workstation. The symmetrical dummy cells fabricated with two identical Pt electrodes (Pt electrode/gel electrolyte/Pt electrode). The electrochemical impedance spectroscopy (EIS) plots of the ymmetrical dummy cells were carried out using a CHI660E electrochemical workstation at a constant temperature of 20 °C (controlled by an air conditioning) with an ac signal amplitude of 20 mV in the frequency range from 0.1 to 105 Hz 6

ACCEPTED MANUSCRIPT in the dark. The cyclic voltammetry was carried out on a CHI660E electrochemical workstation in a N2−purged liquid electrolyte to evaluate the catalytic activity of a conducting gel electrolyte toward I−/I3− couples: gel electrolyte was used as working electrode, Pt foil as the counter electrode, and an Ag/AgCl electrode as reference electrode, the supporting electrolyte was 0.1 M LiClO4 whereas the

3. Results and discussion Figure 1 here

SC

RI PT

redox couple was 10 mM LiI and 1 mM I2.

M AN U

The microporous structure of PAA−CTAB matrix provides 3D framework in reserving enormous electron−conducting polymers and liquid electrolyte. Polymerization of PAA−CTAB is believed as a free radical process [24]. As shown in Fig. 1, electrostatic force presenting in a negatively charged acrylic acid monomer and a positively CTAB molecule can form an acrylic

TE D

acid−CTAB complex monomer and release an HBr molecule. Thermal cleavage of initiator APS provides free radicals in initiating acrylic acid−CTAB complex monomers. Owing to the

EP

macrobiradical nature of NMBA, the resultant PAA−CTAB is a 3D framework with decoration of hydrophiphilic C=O groups and hydrophobic C16 alkyl chains [25]. Therefore, the as−synthesized

AC C

PAA−CTAB is an amphiphilic polymer. Figure 2 here

The pore size within PAA−CTAB matrix has been found to depend on a number of factors, such as initiator, crosslinker, reaction temperature, and concentration of monomer. Herein we focus on the dependence of microporous structure on conducting species and therefore liquid electrolyte loadings. Crosslinking of PAA−CTAB by NMBA occurs to form a 3D framework, in which PAA−CTAB chains are physically entangled and intermolecular or intramolecular−hydrogen 7

ACCEPTED MANUSCRIPT bonded in the dried state, giving a compact molecular configuration. Under the driving force of osmotic pressure, the 3D PAA−CTAB frameworks stretch and expand in deionized water, resulting in the formation of swollen PAA−CTAB hydrogels. The internal microscopic morphology of freeze−dried PAA−CTAB hydrogel is examined using SEM as shown in Fig. 3a, revealing a

RI PT

microporous structure. This peculiar structure along with imbibition of hydrophobic groups toward liquid electrolyte cause the diffusion of PANi or PPy incorporated liquid electrolyte into 3D

SC

framework of PAA−CTAB matrix. The extraordinary structure is expected to imbibe more PANi or PPy and liquid electrolyte for rapid charge transfer. As shown in Fig. 3b, partial micropores have

M AN U

been filled by PANi, still leaving pores for liquid electrolyte loading. However, it is relatively difficult for imbibing PANi or PPy chains into a dense PAA−CTAB matrix because of the entangled chains from matrix and conducting species. Driven by both osmotic pressure across gel matrix and capillary force presenting in micropores, PANi or PPy chains accompanied with liquid electrolyte

TE D

can diffuse into interconnected micropores of PAA−CTAB matrix and attach on the inner surface of these micropores by hydrogen bonds (Fig. 2). Homogeneous attachment of conjugated chains is

EP

expected to conduct reflux electrons from Pt counter electrode to conducting gel electrolyte and make contributions in reducing I3−. In comparison with pure PAA−CTAB gel electrolyte, the I3−

AC C

species have to transfer from anode/electrolyte interface to electrolyte/Pt interface for converting to I− ions, the long diffusion length for I−/I3− couples can apparently decrease the reaction kinetics and therefore photovoltaic performances. Figure 3 here Figure 4 here The FTIR spectra of PAA−CTAB/PANi, PAA−CTAB/PPy, PAA−CTAB, PANi, and PPy are shown in Fig. 4, where the absorption bands at 1682, 1507, and 1442 cm−1 in pure PAA−CTAB 8

ACCEPTED MANUSCRIPT matrix are originated from the vibrations of C=O stretching (PAA), C=O bending (PAA), and O−H distorting (PAA), respectively [26]. However, these bands shift to 1668, 1516, and 1432 cm−1 in PAA−CTAB/PANi and 1675, 1496, and 1448 cm−1 in PAA−CTAB/PPy, respectively. The change of absorption bands suggests the H−bonding interactions between PAA−CTAB framework and PANi

RI PT

or PPy backbone. From FTIR analysis, it is expected that the H−bonded PANi or PPy (−NH−) chains by 3D PAA−CTAB framework (C=O) can form interconnected channels for conducting

Figure 5 here

SC

electrons from Pt counter electrode to conducting gel electrolyte.

M AN U

The dependences of liquid electrolyte loading and room−temperature ionic conductivity of PAA−CTAB/PANi, PAA−CTAB/PPy, and PAA−CTAB gel electrolytes on imbibition time are shown in Fig. 5. It is expected that the swelling of gel matrix in liquid electrolyte is governed by Flory theory from osmotic pressure across the PAA−CTAB [18]. However, the as−synthesized

TE D

PAA−CTAB matrix has been swollen by deionized water and freeze−dried to synthesize microporous structure, therefore, capillary force also participates in the driving of liquid electrolyte

EP

[27]. A rapid increase in initial stage (0−8 days) occurs because of gradual diffusion of liquid electrolyte into PAA−CTAB matrix. An imbibition equilibrium can be obtained at swelling time of

AC C

around 15 days, and no further diffusion occurs under longer immersion time. In order to determine the nature of liquid electrolyte loading by microporous PAA−CTAB, the accumulative liquid electrolyte loading over time have been fitted using the Fickian theory [28]. Mt = kt n M∞

(1)

where Mt and M∞ are the mass of the imbibed liquid electrolyte at time t and at equilibrium, respectively. K is a characteristic rate constant relating to the properties of PAA−CTAB matrix, and

n is a transport number characterizing the transport mechanism. n ≤ 0.5 suggests a Fickian or Case I 9

ACCEPTED MANUSCRIPT transport behavior in which the PAA−CTAB relaxation is much faster than the diffusion; n = 1 gives a non−Fickian or Case II mode of transport where liquid electrolyte uptake is controlled by diffusion process. 0.5 < n < 1 refers to an anomalous or a Case III mode in which structural relaxation is comparable to diffusion. By plotting log(Mt/M∞) vs log(t), the n values are recorded

RI PT

and shown in inserts. The n values from the microporous PAA−CTAB/PANi, PAA−CTAB/PPy, and pure PAA−CTAB matrices by swelling in various liquid electrolytes are 0.64, 0.76, and 0.62,

SC

respectively, indicating an anomalous mechanism mode in which structural relaxation is comparable to diffusion. The result indicates that the loading of liquid electrolyte is mainly controlled by

M AN U

osmotic pressure.

The liquid electrolyte loading can be obtained by calculating: Liquid electrolyte loading ( g g −1 ) =

M gel − M polymer M polymer

(2)

TE D

where Mgel (g) and Mpolymer are mass of anhydrous PAA−CTAB matrix and polymer gel electrolyte, respectively. It is apparent that the PAA−CTAB/PANi has a liquid electrolyte loading of 13.70 g g−1, which is much higher than 8.65 and 6.72 g g−1 for PAA−CTAB/PPy and PAA−CTAB. Higher liquid

EP

electrolyte loading is expected to give a better ionic conductivity owing to a higher content of liquid

AC C

electrolyte in per volume unit. Room-temperature ionic conductivities of PAA−CTAB/PANi, PAA−CTAB/PPy, and pure PAA−CTAB gel electrolytes are 8.55, 7.93, and 6.38 mS cm−1, which is in a good agreement with liquid electrolyte loading. The enhancement in ionic conductivity maybe assigned to the facts of increased liquid electrolyte content in per unit volume and polycation nature of PANi or PPy.

Figure 6 here The conductivity−temperature plots of various gel electrolytes follow an Arrhenius relationship reasonably well, which is a typical feature of ion−conduction nature [29], as shown in Fig. 6. 10

ACCEPTED MANUSCRIPT Apparently, the ionic conductivities of the gel electrolytes are in an order of PAA−CTAB/PANi > PAA−CTAB/PPy > pure PAA−CTAB. According to Fig. 6, the ionic conductivities at 75 °C are 13.21, 9.77, and 8.38 mS cm−1 for PAA−CTAB/PANi, PAA−CTAB/PPy, and pure PAA−CTAB, respectively. Therefore, the interconnected microporous structure of PAA−CTAB/PANi provides a

RI PT

superhighway for facile ion transport. The increased ionic conductivity is expected to significantly enhance the reaction kinetics of DSSCs. There is a fact that the conductivity of PANi or PPy has

SC

influence on the total conductivity of gel electrolyte. The electrical conductivities of PANi and PPy can be optimized by adjusting synthesis conditions such as dosage of oxidant, dopant, concentration,

M AN U

reaction temperature. In the current work, we focus on the investigation of dependence of PANi or PPy on electrical and electrochemical performances of gel electrolytes. Figure 7 here

Fig. 7 shows the CV curves recorded using PAA−CTAB/PANi or PAA−CTAB/PPy gel

TE D

electrolyte as a working electrode and liquid electrolyte as supporting electrolyte. As a reference, pure PAA−CTAB gel electrolyte is also utilized as a working electrode in recording its CV curve.

EP

No redox peaks are observed using PAA−CTAB gel electrolyte as working electrode, indicating no electrocatalytic activity of PAA−CTAB to I−/I3− redox couples. However, a pair of redox peaks are

AC C

determined from PAA−CTAB/PANi or PAA−CTAB/PPy conducting gel electrolyte. As a reference, the CV curve of standard Pt electrode in liquid electrolyte was also measured under the same conditions. The peak shapes and positions of the CV curves from PAA−CTAB/PANi and PAA−CTAB/PPy are similar to that from Pt electrode, suggesting that the conducting gel electrolytes have a similar electrocatalytic activity to standard Pt electrode [30,31]. It is noteworthy to mention that the disappearance of Ox2 peaks for the conducting gel electrolytes may reveal a relatively low catalytic activity toward I−/I3− redox couples. Generally, the whole catalytic kinetics 11

ACCEPTED MANUSCRIPT is governed by Red1/Ox1, therefore, the detection of Red1/Ox1 peaks in the conducting gel electrolytes demonstrates the catalytic activity toward I−/I3− redox couples. It have been widely reported that PANi and PPy are good candidates for Pt−free counter electrode materials toward I3− reduction [32−35]. Similarly, the imbibed PANi or PPy is also believed to form interconnected

RI PT

channels, which can conduct reflux electrons from Pt counter electrode to the whole 3D conducting gel electrolyte. Therefore, the electrocatalytic reaction of iodides can be extended from gel

SC

electrolyte/Pt interface to both interface and 3D conducting gel electrolyte. The combination of electron−conducting PANi or PPy with insulating PAA−CTAB matrix has a promotion effect on

M AN U

enhancing electrocatalytic activity because of the incremental quantity in active sites for electroreduction reaction. Moreover, the ratio of Jox1/|Jred1| is a parameter to elevate the reversibility of the redox reaction toward I−/I3− [36]. The obtained value from PAA−CTAB/PANi gel electrolyte is 1.11 which is closer to 1.0 than 1.15 from PAA−CTAB/PPy, indicating that PAA−CTAB/PANi

TE D

gel electrolyte has a more reversible redox reaction for I3− ↔ I−. The rapid recovery of I− to their ground state facilitates the participation in subsequent circles and improves the long−term stability.

AC C

and presented [37]:

EP

To elucidate the diffusion of I3− in a conducting gel electrolyte, Randles−Sevcik theory is employed

J red = Kn1.5 ACDn 0.5v 0.5

(3)

where Jred is the peak current density of Red1 (mA·cm−2), K is 2.69 × 105, n is the number of electrons of reduction reaction, A is the electrode area (cm2), C represents the bulk concentration of I3− (mol L−1), Dn is the diffusion coefficient (cm2 s−1), v is scan rate in recording CV curves. The Dn values of PAA−CTAB/PANi and PAA−CTAB/PPy gel electrolytes are 5.47 × 10−6 and 1.37 × 10−6 cm2 s−1, respectively, which are comparable to 6.24 × 10−6 cm2 s−1 in pure PANi [38] and 4.02 ×

12

ACCEPTED MANUSCRIPT −6

2

−1

10 cm s in pure PPy counter electrodes [39]. The calculated values are also in the same scale with the reports from Grätzel et al [40−42]. Results indicate that the conducting gel electrolytes have a comparably electrocatalytic performance with pure PANi or PPy counter electrode. Figure 8 here

RI PT

EIS has been widely employed in testing the electrocatalytic activity for the regeneration of a redox couple by assembling a symmetric dummy cell composed of a gel electrolyte sandwiched by

SC

two identical FTO glass supported Pt electrodes, as shown in Fig. 8a. According to the Randles−type circuit (inset of Fig. 8a) [30], the intercept on the real axis represents the series

M AN U

resistance (Rs). The arc arises from the charge−transfer resistance (Rct) at counter electrode/gel electrolyte interface, which changes inversely with the electrocatalytic activity of gel electrolytes on the reduction of I3−, whereas W represents the Nernst diffusion impedance corresponding to the diffusion resistance of I−/I3− redox species. CPE is a constant phase element and is frequently used

TE D

as a substitute for a capacitor in an equivalent circuit to fit the impedance behavior of the electrical double layer. These electrochemical parameters were obtained by fitting EIS spectra using a Z−view

EP

software and there is a good agreement between measured plots and fitted curves. The Rct values are 61.9, 178.2, and 219.0 Ω cm2 for PAA−CTAB/PANi, PAA−CTAB/PPy, and pure PAA−CTAB gel

AC C

electrolytes, respectively. The attachment of PANi or PPy chains on inside surface of 3D PAA−CTAB matrix can form electron−conducting channels for charge migration. Lower Rct reveals that the charge−transfer ability in conducting gel electrolyte/Pt interface is higher than that in PAA−CTAB/Pt interface. Prior to the synthesis of conducting gel electrolytes, PANi or PPy has been completely dissolved into liquid electrolyte containing I−/I3− redox couples. Under driving by both osmotic pressure and capillary force, liquid electrolyte along with PANi (or PPy) diffuse into 3D framework of microporous PAA−CTAB matrix. The imbibed PANi (or PPy) chains are attached 13

ACCEPTED MANUSCRIPT onto inner surface of PAA−CTAB matrix by H−bonding, whereas the migration modes of I−/I3− redox couples migrate within the gel electrolyte and liquid electrolyte are considered as the same because of a supporting function of PAA−CTAB matrix. The exposed PANi (or PPy) chains that closely contact Pt CE can conduct electrons from Pt surface to the 3D system of conducting gel

RI PT

electrolyte. The extension of electrocatalytic reaction from gel/Pt interface to both interface and 3D framework of conducting gel electrolyte can dramatically accelerate the reduction reaction of I3−

SC

→I−.

Tafel-polarization plots were also recorded to determine the electrocatalytic activity of the gel

M AN U

electrolytes, which was also performed with a symmetric similar to those used in EIS measurements, are shown in Fig. 8b. A larger slope in the anodic or cathodic branch indicates a higher exchange current density (J0). Considering that the pure Tafel region is not observable, therefore the low field region is used to assess J0 variation. The extracted J0 has an order of PAA−CTAB/PANi >

TE D

PAA−CTAB/PPy > PAA−CTAB. J0 is inversely proportional to Rct [43]: J0 = RT/nFRct, where R is the gas constant, T is absolute temperature, F is Faraday’s constant. Apparently, the calculated Rct

EP

(PAA−CTAB/PANi > PAA−CTAB/PPy > PAA−CTAB) match the order in EIS and CV analysis. Additionally, the intersection of the cathodic branch with Y−axis is determined as limiting diffusion

AC C

current density (Jlim), a parameter depends on the diffusion coefficient (Dn) of I−/I3− redox couples at CE/electrolyte interface. Jlim is in proportion to Dn [44]: Jlim = 2nFCDn/l, where l is the distance between electrodes in a dummy cell, n is the number of electrons involved in the reduction of I3−, and C is the I3− concentration. Apparently, both Jlim and Dn have a sequence of PAA−CTAB/PANi > PAA−CTAB/PPy > PAA−CTAB. Alternatively, the Dn can also be described by Randles−Sevcik theory. As shown in Fig. 7, the calculated Dn in CV measurement is consistent with that from Tafel analysis. 14

ACCEPTED MANUSCRIPT Figure 9 here Table 1 here Fig. 9 shows the photovoltaic characteristics of DSSCs from PAA−CTAB/PANi, PAA−CTAB/PPy, and pure PAA−CTAB gel electrolytes and the parameters reflecting DSSC

RI PT

properties are summarized in Table 1. The DSSC employing PAA−CTAB/PANi gel electrolyte achieves the highest power conversion efficiency (the efficiency of converting solar energy into

SC

electricity by a DSSC device) of 7.11% in comparison with 6.39% from PAA−CTAB/PPy and 6.07% from pure PAA−CTAB based DSSCs. This might be attributed to the high liquid content of

M AN U

liquid electrolyte in per volume unit, enhanced ionic conductivity, and elevated electrocatalytic behavior toward I3−. After performing a comprehensive analysis of the DSSCs, it can be concluded that the photovoltaic performance is in agreement with the EIS and Tafel polarization results. Moreover, the FF for PAA−CTAB/PPy based solar cell is relatively low in comparison with that

TE D

from PAA−CTAB/PANi electrolyte, which maybe a result of entangled conjugated structure for

electrode.

AC C

4. Conclusions

EP

PPy chains and increased interfacial resistance at anode/gel electrolyte, and gel electrolyte/Pt

In summary, microporous PAA−CTAB matrix has been successfully synthesized and employed as placeholder for PANi or PPy and liquid electrolyte loading. The imbibed PANi or PPy conjugated chains are attached onto the inside surface of 3D PAA−CTAB framework by H−bonding. Compared with pure PAA−CTAB gel electrolyte, the liquid electrolyte loading and therefore ionic conductivity have been significantly enhanced because of increase of liquid electrolyte content in per volume unit. The imbibed PANi or PPy in conducting gel electrolyte can 15

ACCEPTED MANUSCRIPT conduct reflux electrons from Pt counter electrode into its 3D framework, resulting in an elevated electrocatalytic reaction kinetics of I3−. Quasi−solid−state DSSC from PAA−CTAB/PANi gave a power conversion efficiency of 7.11% in comparison with 6.07% from pure PAA−CTAB based DSSC device. This research opened a gateway to improve the photovoltaic performances of

RI PT

quasi−solid−state DSSCs and highlighted competitive capacity of the quasi−solid−state DSSC

SC

among photovoltaic devices.

Acknowledgements

M AN U

The authors gratefully acknowledge Fundamental Research Funds for the Central Universities (201313001, 201312005), Shandong Province Outstanding Youth Scientist Foundation Plan (BS2013CL015), Shandong Provincial Natural Science Foundation (ZR2011BQ017), Research Project for the Application Foundation in Qingdao (13−1−4−198−jch), National Natural Science

TE D

Foundation of China (51102219, 51342008), National High Technology Research and Development Program of China (2010AA09Z203, 2010AA065104), and National Key Technology Support

AC C

References

EP

Program (2012BAB15B02).

[1]

B. O’Regan, M. Grätzel, Nature 353 (1991) 737−740.

[2]

U. Bach, D. Lupo, P. Comte, J.E. Moser, F.Weissortel, J. Salbeck, et al, Nature 395 (1998) 583−585.

[3]

M. Grätzel, Nature 414 (2001) 338−344.

[4]

X. X. Chen, Q. W. Tang, B. L. He, L. Lin, L. M. Yu, Angew. Chem. Int. Ed. 53 (2014) 10799−10803. 16

ACCEPTED MANUSCRIPT [5]

Y. Y. Duan, Q. W. Tang, B. L. He, R. Li, L. M. Yu, Nanoscale (2014) doi: 10.1039/C4NR03900A.

[6]

S. F. Zhang, X. D. Yang, Y. H. Numata, L. Y. Han, Energy Environ. Sci. 6 (2013) 1443−1464. A. Yella, H. W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran, M. K. Nazeeruddin, et al,

RI PT

[7]

Science 334 (2011) 629−634.

J. N. Freitas, A. F. Nogueira, M. A. D. Paoli, J. Mater. Chem. 19 (2009) 5279−5294.

[9]

C. Y. Neo, J. Ouyang, J. Mater. Chem. A 1 (2013) 14392−14401.

[10]

P. Wang, Q. Dai, S. M. Zakeeruddin, M. Forsyth, D. R. MacFarlane, M. Grätzel, J. Am.

M AN U

SC

[8]

Chem. Soc. 126 (2004) 13590−13591. [11]

J. H. Wu, S. Hao, Z. Lan, J. Lin, M. Huang, Y. Huang, et al, J. Am. Chem. Soc. 130 (2008) 11568−11569.

R. X. Dong, S. Y. Shen, H. W. Chen, C. C. Wang, P. T. Shih, C. T. Liu, et al, J. Mater.

TE D

[12]

Chem. A 2013, 1, 8471−8478.

Q. H. Li, Q. W. Tang, L. Lin, X. X. Chen, H. Y. Chen, L. Chu, et al, J. Power Sources 245

EP

[13]

(2014) 468−474.

L. Wang, H. Zhang, C. Wang, T. L. Ma, ACS Sustainable Chem. Eng. 1 (2013) 205−208.

[15]

J. Shi, S. Peng, J. Pei, Y. Liang, F. Cheng, J. Chen, ACS Appl. Mater. Interfaces 1 (2009)

AC C

[14]

944−950. [16]

Z. Tang, J. Wu, Q. Li, Z. Lan, L. Fan, J. Lin, et al, Electrochim. Acta 55 (2010) 4883−4888.

[17]

Q. H. Li, J. H. Wu, Z. Tang, Y. Xiao, M. Huang, J. Lin, Electrochim. Acta 55 (2010) 2777−2781. 17

ACCEPTED MANUSCRIPT [18]

P. J. Flory, 1953, Principles of Polymer Chemistry (New York: Cornel University Press).

[19]

Q. H. Li, H. Chen, L. Lin, P. J. Li, Y. C. Qin, M. Li, et al, J. Mater. Chem. A 1 (2013) 5326−5332.

[20]

Q. H. Li, X. X. Chen, Q. W. Tang, H. T. Xu, B. L. He, Y. C. Qin, J. Mater. Chem. A 1 (2013)

[21]

RI PT

8055−8060.

Q. H. Li, X. X. Chen, Q. W. Tang, H. Y. Cai, Y. C. Qin, B. L. He, et al, J. Power Sources

[22]

SC

248 (2014) 923−930.

P. P. Xu, Q. W. Tang, B. L. He, Q. H. Li, H. Y. Chen, Electrochim. Acta 134 (2014)

[23]

M AN U

281−286.

S. S. Yuan, Q. W. Tang, B. L. He, L. Men, H. Y. Chen, Electrochim. Acta 125 (2014) 646−651.

I. Rintoul, C. Wandrey, Polymer 46 (2005) 4525−4532.

[25]

S. S. Yuan, Q. W. Tang, B. L. He, H. Y. Chen, Q. H. Li, C. Q. Ma, et al, J. Power Sources

TE D

[24]

249 (2014) 277−284.

Q. W. Tang, X. M. Sun, Q. H. Li, J. M. Lin, J. H. Wu, J. Mater. Sci. 44 (2009) 726−733.

[27]

Q. W. Tang, S. S. Yuan, H. Y. Cai, J. Mater. Chem. A 1 (2013) 630−636.

[28]

N. W. Franson, N. A. Peppas, J. Appl. Polym. Sci. 28 (1983) 1299−1310.

[29]

A. M. Fleshman, M. Petrowsky, J. D. Jernigen, R. S. P. Bokalawela, M. B. Johnson, R.

AC C

EP

[26]

Frech, Electrochim. Acta 57 (2011) 147−152. [30]

S. S. Yuan, Q. W. Tang, B. L. He, Y. Zhao, J. Power Sources 260 (2014) 225−232.

[31]

S. S. Yuan, Q. W. Tang, B. L. He, P. Z. Yang, J. Power Sources 254 (2014) 98−105.

[32]

S. Ahmad, E. Guillen, L. Kavan, M. Grätzel, M. K. Nazeeruddin, Energy Environ. Sci. 6 (2013) 3439−3466. 18

ACCEPTED MANUSCRIPT [33]

Q. W. Tang, H. Y. Cai, S. S. Yuan, X. Wang, J. Mater. Chem. A 1 (2013) 317−323.

[34]

J. Xia, L. Chen, S. Yanagida, J. Mater. Chem. 21 (2011) 4644−4649.

[35]

J. H. Wu, Q. H. Li, L. Q. Fan, Z. Lan, P. J. Li, J. M. Lin, et al, J. Power Sources 181 (2008) 172−176. H. Tributsch, Coordin. Chem. Rev. 248 (2004) 1511−1530.

[37]

T. Daeneke, A. J. Mozer, T. H. Kwon, N. W. Duffy, A. B. Holmes, U. Bach, et al, Energy

[38]

SC

Environ. Sci. 5 (2012) 7090−7099.

Y. M. Xiao, J. Y. Lin, W. Y. Wang, S. Y. Tai, G. T. Yue, J. H. Wu, Electrochim. Acta 90

M AN U

(2013) 468−474. [39]

B. L. He, Q. W. Tang, J. H. Luo, Q. H. Li, X. X. Chen, H. Y. Cai, J. Power Sources 256 (2014) 170−177.

[40]

RI PT

[36]

P. Wang, Q. Dai, S. M. Zakeeruddin, M. Forsyth, D. R. MacFarlane, M. Grätzel, J. Am.

[41]

TE D

Chem. Soc. 126 (2004) 13590−13591.

P. Wang, S. M. Zakeeruddin, J. E. Moser, M. Grätzel, J. Phys. Chem. B 107 (2003)

EP

13280−13285.

L. Kavan, J. H. Yum, M. K. Nazeeruddin, M. Grätzel, ACS Nano 5 (2011) 9171−9178.

[43]

C. J. Raj, K. Prabakar, A. D. Savariraj, H. J. Kim, Electrochim. Acta 103 (2013) 231−236.

[44]

B. L. He, X. Meng, Q. W. Tang, ACS Appl. Mater. Interfaces 6 (2014) 4812−4818.

AC C

[42]

19

ACCEPTED MANUSCRIPT

Figure and table captions Fig. 1. Synthesis routes of 3D PAA−CTAB framework. Fig. 2. Molecular structures of the resultant PAA−CTAB/PANi and PAA−CTAB/PPy gel matrices. Fig. 3. Cross−sectional SEM photographs of (a) PAA−CTAB matrix and (b) PAA−CTAB/PANi.

RI PT

Fig. 4. (a) Full and (b) magnified FTIR spectra of PAA−CTAB/PANi, PAA−CTAB/PPy, pure PAA−CTAB, PANi, and PPy.

SC

Fig. 5. Time dependences of liquid electrolyte loading and ionic conductivity for (a) PAA−CTAB, (b) PAA−CTAB/PPy, and (c) PAA−CTAB/PANi gel electrolytes.

M AN U

Fig. 6. Arrhenius plots of various gel electrolytes from PAA−CTAB/PANi, PAA−CTAB/PPy, and pure PAA−CTAB.

Fig. 7. CV curves of various gel electrolytes toward the liquid electrolyte containing I−/I3− redox couples. The active area of gel electrolytes was 2 cm2 and all the curves were scanned at a scan rate

reference.

TE D

of 50 mV s−1. The CV curve of standard Pt electrode in liquid electrolyte was provided as a

EP

Fig. 8. (a) Nyquist plots and (b) Tafel polarization curves of PAA−CTAB/PANi, PAA−CTAB/PPy, and PAA−CTAB gel electrolytes. The inserts in figure (a) and figure (b) display a Randle’s

AC C

equivalent circuit and magnified Tafel polarization curves in an abscissa range of −0.2 ~ 0.2 V, respectively.

Fig. 9. Characteristics J−V curves of quasi-solid-state DSSCs from PAA−CTAB/PANi, PAA−CTAB/PPy, and PAA−CTAB gel electrolytes. Table 1. The EIS parameters in Fig. 8a and photovoltaic performances of the quasi−solid−state DSSCs based on different gel electrolytes.

20

(NH4)2S2O8

2SO4

2NH4

RI PT

ACCEPTED MANUSCRIPT

O

O HO OH

OH

HO

O

O

HO

2n

n

OH

OH2

O

O

O

SC

HO

HSO4

HO

H2O

SO4

OH

OH

OH

O

O

Br

O

M AN U

N

OH

N

O

O

HO

O

HO

N

O

N

O HO O

O

HO

N H

TE D

O

N H

N

O

O

EP AC C

O O

O

OH

OH

OH

OH

O N

OH O HN

NH HN HO

HO

HO

O

O

O O O

O

n O

O

O

O

NH

O

N

OH

n

O

N

n O

HO

Fig. 1

21

O

N

OH

ACCEPTED MANUSCRIPT

O

O O

RI PT

N

O

OH

n

O

N

O

O

O

OH

OH

OH O

NH HN H N

H N

NH

H N

H N

N H

NH

N

HO

HO

HO

O

O

n

O O

SC

O

N

O

O

n HO

O O

O

O

N

O

O

O

OH

OH

OH

O

NH

H N

H N

O

M AN U

N

O

OH

n

O

HN H N

H N

H N

NH

HN

O

N

HO

HO

HO

O

O

O

O

O

O

n

O

HO

AC C

EP

TE D

Fig. 2

22

H N

O

N

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

Fig. 3

23

ACCEPTED MANUSCRIPT

a PAA-CTAB/PANi PAA-CTAB/PPy PAA-CTAB PANi PPy 3500

3000

2500

1507 1442

1682

2000

1500

1000

500

Wavenumber ( cm ) -1

2000

PAA-CTAB/PANi PAA-CTAB/PPy PAA-CTAB

SC

4000

RI PT

Transmittance (a.u.)

Transmittance (a.u.)

b

1800

1600

1400

M AN U

Wavenumber ( cm )

AC C

EP

TE D

Fig. 4

24

-1

1200

PANi PPy

1000

ACCEPTED MANUSCRIPT

b

4

-0.2

3

-0.3

3

-0.4

n=0.62

-0.5

1

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

0

2

4

6

8

10

12

14

0 16

-0.1

-0.3

2

-0.6

1

18

0

2

6

0.0

8

8

9 8 7 6

Log (Mt/M8 )

-0.1

6

-0.2

n=0.64

-0.3

4

-0.4

0.4

2

4

6

8

10

Time (day)

EP

Fig. 5

25

0.8

1.0

12

14

4 3

1.2

Log t (day)

0

0

0.6

5

-1

-0.5

2

TE D

-1

4

2 16

18

Ionic conductivity (mS cm )

10

2

-0.5

M AN U

12

Liquid electrolyte load (g g )

n=0.76

-0.4

3

0.6

0.8

1.0

1.2

Log t (day)

10

Time (day)

c

4

-0.2

4

Time (day) 14

6

0.0

5

0.4

Log t (day)

1

6

-1

-0.6

-1

2

2

RI PT

4

-0.1

7

SC

Log (Mt/M8 )

0.0

Ionic conductivity (mS cm )

5

5

8

8

Log (Mt/M8 )

6

-1

6

9

Liquid electrolyte load (g g )

7

AC C

-1

7

Ionic conductivity (mS cm )

Liquid electrolyte load (g g )

a

12

14

0 16

18

ACCEPTED MANUSCRIPT

Ε = a 5 .4

2.2

2.0

9k

Ε = 4.69 a k

mo

Jm

ol - 1

J mo

l -1

1.8 3.3

3.4

3.5

3.6

3.7

M AN U

3.2

l -1

SC

Ε = a 6. 4 6 kJ

2.4

RI PT

PAA-CTAB PAA-CTAB/PPy PAA-CTAB/PANi

-1

ln ( ionic conductivity, mS cm )

2.6

-1

-1

1000*T ( K )

AC C

EP

TE D

Fig. 6

26

3.8

ACCEPTED MANUSCRIPT

PAA-CTAB/PANi PAA-CTAB/PPy PAA-CTAB Pt

8

-

Ox2, 2I3 -2e=3I2 -

Ox1, 3I -2e=I3

-

4

0

RI PT

-2

Current density ( mA cm )

12

Red2, 3I2+2e=2I3

-4 -

-

-1.2

-0.6

0.0

SC

Red1, I3 +2e=3I

-8

0.6

M AN U

Potential (V vs Ag/AgCl)

AC C

EP

TE D

Fig. 7

27

-

1.2

ACCEPTED MANUSCRIPT

CPE 250

1

-2

Log(j, mA cm )

Rct

W

150 PAA-CTAB/PANI PAA-CTAB/PPY PAA-CTAB

100

-1

Polarization zone

Limiting diffusion zone

J0 Tafel zone

-2

PAA-CTAB/PANi PAA-CTAB/PPy PAA-CTAB

-3

50

-4

0 0

50

100

150 200 2 Z' (ohm cm )

250

300

-1.0

-0.5

EP

TE D

M AN U

Fig. 8

AC C

2

-Z'' (ohm cm )

0

Rs

200

Jlim

RI PT

b

300

28

-0.2

-0.1

0.0

0.1

0.2

Potential (V)

0.0

SC

a

Potential (V)

0.5

1.0

ACCEPTED MANUSCRIPT

18

-2

Curretn density (mA cm )

15

RI PT

12 9 6

0 0.0

0.2

0.4

0.6

M AN U

Voltage (V)

SC

PAA-CTAB/PANi PAA-CTAB/PPy PAA-CTAB

3

AC C

EP

TE D

Fig. 9

29

0.8

ACCEPTED MANUSCRIPT

Table 1 Jsc (mA cm−2)

Voc (V)

FF (%)

Rs (Ω cm2)

Rct (Ω cm2)

PAA−CTAB/PANi

7.11

14.49

0.704

69.7

22.3

61.9

PAA−CTAB/PPy

6.39

15.73

0.689

59.0

26.0

178.2

PAA−CTAB

6.07

12.15

0.696

71.8

27.1

219.0

AC C

EP

TE D

M AN U

SC

RI PT

Efficiency (%)

Gel electrolytes

30

ACCEPTED MANUSCRIPT ● PANi or PPy is imbibed into 3D framework of microporous PAA−CTAB matrix ● The electrocatalytic reaction of I3− is conducted into conducting gel electrolyte ● Liquid electrolyte is driven by both osmotic pressure and capillary diffusion

AC C

EP

TE D

M AN U

SC

RI PT

● A power conversion efficiency of 7.11% is measured under one sun illumination.