In-situ Polymerization of Thiophene Derivatives Using a Gas-phase Oxidant to Form a Hole-transporting Layer

in Dye-sensitized Solar Cell

Yoshito Sasada, Fumiaki Kato, Kenichi Oyaizu, and Hiroyuki Nishide*

Department of Applied Chemistry, Waseda University, Shinjuku, Tokyo 169-8555, Japan

*nishide@waseda.jp

Poly(3,4-ethylenedioxythiophene) (PEDOT), for use as a hole-transporting layer in dye-sensitized solar cells (DSSCs), was synthesized via in-situ polymerization using a gas-phase oxidant. Highly volatile iodine was used as the gas-phase oxidant to fill up a porous TiO2 substrate. The conductivity of PEDOT increased to 0.9 S/cm upon the addition of a base, 4-tert-butylpyridine, to the monomer solution. A solid-state DSSC was fabricated using PEDOT as the hole-transporting layer, which displayed photo-electric conversion.

Keywords: conducting polymer, in-situ polymerization, polythiophene, dye-sensitized solar cell

1. Introduction Polythiophene derivatives (for example:

poly-(3,4-ethylenedioxythiophene) (PEDOT)), have been extensively studied as conducting polymers [1] and exploited as hole-transporting materials in electric and photoelectric devices such as organic light emitting diodes [2, 3], organic photovoltaic cells [4, 5], and dye-sensitized solar cells (DSSCs) [6], because of their high electric conductivity and tunable energy levels. To fabricate either the formation of a hole-transporting layer or polymerization of thiophene derivatives onto the porous substrates, in-situ electrochemical polymerization of thiophenes has been applied successfully; however, it was effective only on highly conducting substrates such as indium tin oxide (ITO) [7].

For DSSCs, a dye or sensitizer molecule is adsorbed onto the porous TiO2 substrate, which is photo-excited to inject electrons into the TiO2 semiconductor. The dye molecule is regenerated using a hole-transporting

mediator. A widely used mediator is the acetonitrile solution of an iodide/triiodide redox couple, which itself is regenerated at the counter electrode; iodide in the solution acts as the hole-transporting shuttle through diffusion [8]. Conducting polymers have been previously examined for suitability as the hole-transporting material in DSSCs in order to solidify the mediator solution and prevent the deterioration of cell performance caused by the volatilization of the solution [6]. Although the bulk solution phase was solidified by using polymers, it was difficult for the polymers to penetrate the porous TiO2 substrate. To address this problem, for example, in-situ photoelectrochemical polymerization of thiophene derivatives was carried out to form conducting polythiophene polymers on the porous TiO2 substrate. As a result, the resulting DSSC displayed photoelectric conversion [9, 10].

Jensen and coworkers reported the formation of conducting PEDOT via vapor-phase polymerization of

Journal of Photopolymer Science and Technology Volume 27, Number 3 (2014) 347–350 © 2014SPST

Received May 16, 2014 Accepted June 20, 2014 347

3,4-ethylenedioxythiophene (EDOT) by using iron(III)-tosylate as the oxidizing agent, which was pre-coated onto the substrate [11]. However, the residual iron derived from the oxidant often dented the hole-transporting performance of the PEDOT layer, and therefore, the applicable monomers were limited to highly volatile thiophene derivatives.

In this work, we polymerized the thiophene derivative monomers using gas-phase oxidants such as iodine and nitrosyl tetrafluoroborate. We found that the highly volatile and thiophene-soluble oxidants formed polymers of the thiophene derivatives coated onto or immersed into the porous substrate and that residues of the oxidants were easily removed from the formed polymer layers. The hole-transporting capability of the polymerized PEDOT on the porous TiO2 substrate was also examined by fabricating a solid-state DSSC.

2. Experimental

EDOT, 2,2':5',2''-terthiophene, 4-tert- butylpyridine (tBP), and iodine were purchased from Tokyo Chemical Industry Co. Nitrosyl tetrafluoroborate (NOBF4) was purchased from Sigma-Aldrich Co. Indoline dye molecule, 3-1,2,3,3a,4,8b-hexahydro- 4-[4-(2,2-diphenylvinyl)phenyl]cyclopenta[b]indole-7-yl-2-cyanoacrylic acid (D131), was purchased from Mitsubishi Paper Mills Ltd. All reagents were used without further purification. TiO2 paste (DSL 18 NR-T 20 nm) was purchased from Solaronix S.A. (Switzerland).

Porous TiO2 layer was prepared as follows: TiO2 paste was coated onto a glass or a fluorine-doped tin oxide (FTO) substrate using a bar-coating method. The coated substrate was sintered at 450 °C for 90 min and cooled down to room temperature.

The acetonitrile solution of EDOT or terthiophene was dropped onto a glass substrate or spin-coated onto the porous TiO2 substrate, and the substrate was exposed to I2 or NOBF4 atmosphere as the gas-phase oxidant in a sealed vessel to induce polymerization of the thiophene monomer. After 30－120 min, the substrate was rinsed with acetonitrile to remove the excess oxidant on the substrate.

The conductivity of PEDOT was estimated from the resistance and layer thickness. The resistance was measured using a four-probe method, and the layer thickness was measured using a contact stylus profiler (KLA Tencor P-6).

A solid-state DSSC was fabricated as follows: porous TiO2-coated FTO substrate was immersed into a 0.3 mM D131 dye solution (acetonitrile and tert-butyl alcohol, v/v = 1/1). The substrate was left to stand for overnight at room temperature, which was followed by rinsing with acetonitrile. On the substrate, EDOT was polymerized by the vapor-phase polymerization with I2. A drop of 1-ethyl-3-methylimidazolium bis(trifluoro- methylsulfonyl)imide was placed on the PEDOT-formed TiO2 substrate. The TiO2 substrate as the photo-anode was clipped with a Pt substrate as the counter electrode to fabricate the DSSC. The photovoltaic performance of the DSSC was measured using a computer-programmed digital sourcemeter (Keithley 2611A) under solar irradiation (100 mW/cm2).

A cross section of the TiO2 porous substrate was observed using scanning electron microscopy (SEM). The samples were scribed onto the glass substrate side and split using liquid nitrogen to observe the cross-sectional SEM-images (HITACHI High-Tech S-3000N). The IR spectrum of the PEDOT/TiO2 substrate was measured using reflection absorption method (NICOLET 6700 FT-IR). The UV-Vis spectrum of the PEDOT/TiO2 substrate was monitored using a UV-Vis spectrophotometer (JASCO UV-670).

3. Results and discussion

I2 and NOBF4 were used as the gas-phase oxidants. The thiophene monomer-coated glass substrate was set in a vessel filled with the oxidant atmosphere at room temperature. After 30 min, the substrate surface homogeneously turned black. The black layer formed under the I2 vapor was insoluble in any of the organic solvents. However, the layer containing NOBF4 was soluble in organic solvents, which suggested the formation of a polymer with I2 (I2 was used as a gas-phase oxidant in the subsequent experiment). The resulting layer formed with I2 was characterized as PEDOT using IR

348

J. Photopolym. Sci. Technol., Vol. 27, No. 3, 2014

spectroscopy (Figure 1, C-O-C = 1230 cm-1, S-C = 1080 cm-1,thiphene ring = 1400－1500 cm-1).

Figure 1. IR spectra of the PEDOT layer formed via in-situ polymerization with I2.

The electric resistance of the polymer layer

was reduced from >1 MΩ to 100－200 kΩ by adding either 4-tert-butylpyridine (tBP) as a base or tetrabutylammonium perchlorate as a salt into the monomer solution, which was coated onto the substrate. Both the highly volatile monomer EDOT and less volatile monomer terthiophene were polymerized using this method on any substrate surface.

The in-situ polymerization with the vapor phase oxidant, I2, was applied onto a TiO2 porous substrate (particle size = 20 nm). The acetonitrile solution of EDOT and tBP was spin-coated onto the TiO2 substrate, exposed to the I2 atmosphere, and washed with acetonitrile. The cross-sectional SEM images of the EDOT-polymerized TiO2 are shown in Figure 2. The porous space of TiO2 was thoroughly filled with a polymer after exposure to I2. This indicated that the in-situ polymerization of EDOT with vapor-phase I2 proceeded in the pores of TiO2 and the resulting polymer filled up the porous TiO2. The color of the EDOT-polymerized TiO2 substrate changed from dark green to dark red through the de-doping process by exposure to hydrazine vapor. Although the resulting layer showed absorption bands in the UV-Vis (broad) and infrared (800－1500 nm) region, the de-doping layer showed an absorption peak at 495 nm without any absorption in the infrared region. This result did not conflict with the formation of PEDOT.

Figure 2. Cross-sectional SEM images: (a) porous TiO2 substrate; (b) porous TiO2 substrate after in-situ polymerization of EDOT with I2 vapor (lower right: breakage through the freeze-fracture process) and (c) porous TiO2 substrate after in-situ polymerization of terthiophene with I2 vapor.

The electric conductivity of PEDOT-filled

TiO2 was measured using the four-probe method by taking into account the layer thickness of TiO2 (5 m). The conductivity of the PEDOT/TiO2 substrate was estimated to be 0.02 S/cm, which was reasonable for a conducting network of PEDOT formed in porous and semi-conducting TiO2. The conductivity was lower than those of electrochemical [7] and vapor-phase polymerized [11] PEDOTs, which could be ascribed to the lower molecular weight of PEDOT polymerized with I2.

The conductivity of PEDOT was improved by adding additives to the monomer solution. The addition of 25 mol% tBP as a base increased the conductivity of PEDOT/TiO2

100020003000

80

90

100

Re

flect

ion-

abso

rban

ce (

%)

Wavenumber (cm-1)

(a)

(b)

(c)

100 nm

100 nm

100 nm

349

J. Photopolym. Sci. Technol., Vol. 27, No. 3, 2014

substrate reached to 0.9 S/cm. It could be reasoned that tBP neutralized the protons generated during the EDOT polymerization process, which resulted in high-molecular weight PEDOT. The conductivity of PEDOT/TiO2 substrate might be applicable to the following DSSC fabrication.

Table 1. Conductivity of the PEDOT layer.

The valence band of PEDOT has been

reported to be 5.2 eV [12], which makes it suitable as a hole-transporting layer of the DSSC using the dye molecule D131 with a deep HOMO level (5.5 eV) [13] and a high molar absorption coefficient. The theoretical photo-voltage was calculated to be 1.2 V from the valence band of PEDOT and the conduction band of TiO2 (4.0 eV). A solid-state DSSC was fabricated using the PEDOT formed on the D131-adsorbed porous TiO2 in the presence of I2 vapor. The solid-state DSSC displayed photoelectric response as shown in Figure 3. However, the photo-current and photo-voltage remained 0.4 mA/cm2 and 0.3 V, respectively.

Figure 3. (a) Time course of open circuit potential and (b) chronoamperometry of the solid-state DSSC fabricated with PEDOT/TiO2 substrate as the photoanode.

PEDOT worked as a hole-transporting layer but the conductivity was not adequate to efficiently transport the hole. The conductivity of the PEDOT/TiO2 substrate could be improved by tuning the in-situ polymerization conditions.

Acknowledgement This work was partially supported by the

FIRST Program on “Development of Organic Photovoltaics toward a Low-Carbon Society” from JSPS, Japan and a Grant-in-Aid for Scientific Research (No. 24225003) from MEXT, Japan. References 1. A. Elschner, S. Kirchmeyer, W. Lovenich,

U. Merker, and K. Reuter, “PEDOT: Principles and Applications of an Intrinsically Conductive Polymer”, CRC Press, USA (2010) p.113.

2. Y.-J. Pu, T. Chiba, N. Aizawa, H. Sasabe, and J. Kido, J. Photopoly. Sci. Technol., 26 (2013) 403.

3. K. Fehse1, K. Walzer, K. Leo, W. Lövenich, and A. Elschner, Adv. Mater., 19 (2007) 441.

4. T. Kuwabara, S. Katori, K. Arima, Y. Omura, T. Yamaguchi, T. Taima, and K. Takahashi, Jpn. J. Appl. Phys., 53 (2014) 02BE07.

5. M. T. Lloyd, C. H. Peters, A. Garcia, I. V. Kauvar, J. J. Berry, M. O. Reese, M. D. McGehee, D. S. Ginley, and D. C. Olson, Sol. Energy Mater. Sol. Cells, 95 (2011) 1382.

6. S. Yanagida, Y. Yu, and K. Manseki, Acc. Chem. Res., 42 (2009) 1827.

7. L. Groenendaal, F. Jonas, D. Freitag, H. Pielartzik, and J.R. Reynolds, Adv. Mater., 12 (2000) 481.

8. A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo, and H. Pettersson, Chem. Rev., 110 (2010) 6595.

9. J. Xia, N. Masaki, M. Lira-Cantu, Y. Kim, K. Jiang, and S. Yanagida, J. Am. Chem. Soc., 130 (2008) 1258.

10. X. Liu, Y. Cheng, L. Wang, L. Cai, and B. Liu, Phys. Chem. Chem. Phys., 14 (2012) 7098.

11. B. Winther-Jensen, J. Chen, K. West, and G. Wallace, Macromolecules, 37 (2004) 4538.

12. S.S. Li, K.H. Tu, C.C. Lin, C.W. Chen, and M. Chhowalla, ACS Nano, 4 (2010) 3169.

13. T.L. Bahers, T. Pauporté, G. Scalmani, C. Adamo, and I. Ciofini, Phys. Chem. Chem. Phys., 11 (2009) 11276.

0

0.1

0.2

0.3

0.4

0.5

0 30 60 90

Ph

oto

volta

ge

(V

)

Time (s)

-0.6

-0.4

-0.2

0

0.2

0 50 100 150 200

Cur

ren

t de

nsi

ty(m

A /

cm2 )

Time (s)

(a) (b)

Conductivity

(S cm-1

)In-situ polymerization 0.02

In-situ polymerization in the presence of t BP 0.9

PEDOT/PSS 10-3‐500 [6]

Electrochemical polymerization 300 [7]Vapor-phase polymerization > 1000 [11]

PEDOT and types of polymerization

350

J. Photopolym. Sci. Technol., Vol. 27, No. 3, 2014