Article pubs.acs.org/ac
Matrix-Assisted Laser Desorption/Ionization Mass Spectrometric Analysis of Poly(3,4-ethylenedioxythiophene) in Solid-State DyeSensitized Solar Cells: Comparison of In Situ Photoelectrochemical Polymerization in Aqueous Micellar and Organic Media Jinbao Zhang,† Hanna Ellis,† Lei Yang,† Erik M. J. Johansson,† Gerrit Boschloo,† Nick Vlachopoulos,†,‡ Anders Hagfeldt,†,‡,§ Jonas Bergquist,∥ and Denys Shevchenko*,∥ †
Physical Chemistry, Centre of Molecular Devices, Department of Chemistry−Ångström Laboratory, Uppsala University, SE-75120 Uppsala, Sweden ‡ Laboratory of Photomolecular Science, Institute of Chemical Sciences and Engineering, École Polytechnique Fédérale de Lausanne, EPFL-FSB-ISIC-LSPM, Station 6, CH-1015 Lausanne, Switzerland § Center of Excellence for Advanced Materials Research, King Abdulaziz University, Jeddah 21589, Saudi Arabia ∥ Analytical Chemistry, Department of Chemistry−Biomedical Centre, Uppsala University, P.O. Box 599, SE-75124 Uppsala, Sweden S Supporting Information *
ABSTRACT: Solid-state dye-sensitized solar cells (sDSCs) are devoid of such issues as electrolyte evaporation or leakage and electrode corrosion, which are typical for traditional liquid electrolyte-based DSCs. Poly(3,4-ethylenedioxythiophene) (PEDOT) is one of the most popular and efficient p-type conducting polymers that are used in sDSCs as a solid-state hole-transporting material. The most convenient way to deposit this insoluble polymer into the dye-sensitized mesoporous working electrode is in situ photoelectrochemical polymerization. Apparently, the structure and the physicochemical properties of the generated conducting polymer, which determine the photovoltaic performance of the corresponding solar cell, can be significantly affected by the preparation conditions. Therefore, a simple and fast analytical method that can reveal information on polymer chain length, possible chemical modifications, and impurities is strongly required for the rapid development of efficient solar energy-converting devices. In this contribution, we applied matrix-assisted laser desorption/ionization mass spectrometry (MALDI MS) for the analysis of PEDOT directly on sDSCs. It was found that the PEDOT generated in aqueous micellar medium possesses relatively shorter polymeric chains than the PEDOT deposited from an organic medium. Furthermore, the micellar electrolyte promotes a transformation of one of the thiophene terminal units to thiophenone. The introduction of a carbonyl group into the PEDOT molecule impedes the growth of the polymer chain and reduces the conductivity of the final polymer film. Both the simplicity of sample preparation (only application of the organic matrix onto the solar cell is needed) and the rapidity of analysis hold the promise of making MALDI MS an essential tool for the physicochemical characterization of conducting polymer-based sDSCs.
D
cells (sDSCs), because of its high transparency in the visible range, high conductivity, and remarkable stability at room temperature.4 Since PEDOT is sparingly soluble in several common organic solvents, in situ photoelectrochemical polymerization (PEP) is the most favorable method to produce a PEDOT film that has a good interfacial contact with the porous dye/TiO2 electrode.5 Obviously, the properties of the generated conducting polymer can be significantly affected by the experimental conditions used in the process of PEP, such as
ye-sensitized solar cells (DSCs) have been considered as an attractive alternative to crystalline silicon photovoltaics, because of the low cost and the simple, as well as environmentally friendly, fabrication process.1,2 Traditional DSCs employ, as the charge-transport medium, a liquid electrolyte containing a regenerative redox mediator. However, electrolyte evaporation or leakage and electrode corrosion can limit the long-term stability of the devices. Replacement of the mediator-containing liquid electrolyte by a solid-state hole transporting material helps to overcome the above-mentioned issues.3 Among the various options, poly(3,4-ethylenedioxythiophene) (PEDOT) has attracted considerable attention as a hole-transporting phase for solid-state dye-sensitized solar © XXXX American Chemical Society
Received: December 28, 2014 Accepted: March 7, 2015
A
DOI: 10.1021/ac504851f Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry the nature of dye molecules,6,7 the composition of the electrolyte,5 and the light intensity.8 The chain length of the obtained polymer, as one of the most critical characteristics, can be directly related to its hole conductivity and redox behavior, finally affecting the photovoltaic performance of the corresponding sDSC. However, no work has been done on the detailed study of the chain length of PEDOT produced by in situ photoelectrochemical polymerization. Matrix-assisted laser desorption/ionization mass spectrometry (MALDI MS) has been widely applied for the analysis of synthetic polymers.9 It allows determination of the repeat-unit mass and the length of the polymer chains, the evaluation of the average molar mass and the molar mass distribution, the identification of the terminal groups, and the detection of impurities and additives. Examination of polymeric materials that are deposited on a conducting substrate (e.g., protective coatings of metal components, photoresist coatings of printed circuit boards, conductive polymers in organic light-emitting diodes) can additionally benefit from MALDI MS, since the mass spectrometric analysis can be conducted directly on the evaluated specimen without the need of pretreatment steps such as scratching off or extracting the analyzed polymer. For example, Crecelius and co-workers have successfully used MALDI MS imaging to inspect the quality of the audio operational amplifier wiring diagram imprinted onto the photoresist Novolac resin layer covering the conducting surface.10 The DSC working electrodes are also suitable samples for MALDI MS, because they are fabricated on conducting substrates and, in addition, contain UV-light absorbing materials such as transition-metal oxides and organic dyes. Recently, we have used the mass spectrometry with laserbased ionization to investigate molecular transformation on the surface of dye/TiO2 working electrodes in DSCs employing liquid electrolyte.11 In this work, we have extended the application of MALDI MS for analysis of sDSCs and performed the characterization of PEDOT generated in situ by photoelectrochemical polymerization in organic and aqueous micellar media directly on the sDSCs.
to the back (glass) side of the FTO/TiO2 layer. A current density of 8 μA/cm2 was applied for 2000 s. The electrode potential was monitored against the reference electrode, so that it did not overstep a stipulated positive potential limit of 0.6 V vs. SHE in order to prevent the overoxidation of the generated conducting polymer. After completion of PEP and rinsing off the electrolyte, the films were washed with ethanol and then dried by a N2 flow. As a result, four types of solar cell, termed D35-O, D35-M, LEG4-O, and LEG4-M (the D35 or LEG4 term specifies which dye was employed; the O or M suffix indicates whether an organic or aqueous micellar electrolyte, respectively, was used for PEDOT PEP), were fabricated. In addition, a MALDI MS analysis of the PEDOT film electrochemically deposited from an aqueous micellar electrolyte directly on the FTO substrate was carried out. Such a PEDOT-coated substrate is usually used as a counter electrode in DSCs with liquid electrolyte. The preparation of this type of sample was previously reported.14 MALDI MS Measurements. The MALDI MS experiments were performed in the positive reflector mode with delayed extraction (150 ns) on a Bruker Daltonics Ultraflex II MALDI TOF/TOF spectrometer equipped with a pulsed N2 laser (337 nm). The instrument settings were as follows: ion source I, 25.0 kV; ion source II, 21.7 kV; lens voltage, 10.1 kV; reflector voltage I, 26.3 kV; and reflector voltage II, 13.8 kV. The laser attenuation was adjusted to 10% above the threshold value. Calibration was performed externally using a mixture of the CHCA matrix, the ruthenium tris-bipyridine and tris-phenanthroline complexes [Ru(bpy)3](ClO4)2 and [Ru(phen)3]Cl2, and the Bruker Daltonics peptide calibration standard. Specification of the calibrants’ peaks are given in Table S1 in the Supporting Information (SI). DHB and 2-[(2E)-3-(4-tert-butylphenyl)-2-methylprop-2enylidene]-malononitrile (DCTB) were used as matrices. Solution of DHB was prepared at a concentration of 20 mg/ mL in a 70:30:0.1 (v/v/v) mixture of H2O, MeCN, and TFA, while the DCTB matrix was dissolved in MeOH at a concentration of 15 mg/mL. Matrix solution (0.5 μL) was spotted onto the PEDOT film covering a round area 5 mm in diameter and was allowed to dry. The spotting procedure was repeated two times, until a thin microcrystalline film was formed on the surface of the solar cell. The examined solar cells were mounted onto a homemade aluminum holder11 and contacted by copper tape. The holder was inserted into a standard Bruker Daltonics MTP Slide Adapter II (No. 235380) and loaded into the instrument (see Figure S1 in the SI). Each spectrum was the sum of 500 single laser shots. Spectra were acquired for the m/z range of 200− 9000. However, no signals were observed at m/z >1300. For each sample, 30 spectra were collected. The mean, median, and standard deviation of the relative intensities for each peak considered in this work are given in Table S2 in the SI. The tandem mass spectrometric (MS/MS) experiment was performed in the positive LIFT reflectron mode. The precursor ion selector (PCIS) range was 0.6% of parent ion mass. The voltage parameters were set at the following values: IS1, 8 kV; IS2, 7.2 kV; lens, 3.6 kV; reflector 1, 29.5 kV; reflector 2, 13.85 kV; LIFT1, 19 kV; and LIFT2, 3.6 kV. In total, 3500 laser shots were accumulated, including 1500 for a parent ion and 2500 for fragments.
■
EXPERIMENTAL SECTION Chemicals Used. Bis(3,4-ethylenedioxythiophene) (bisEDOT) from Kairon Chemicals (France) was used as a precursor for photoelectrochemical polymerization. Dyes D3512 and LEG413 were supplied by Dyenamo AB (Sweden). Peptide calibration standard, 2,5-dihydroxybenzoic acid (DHB), and α-cyano-4-hydroxycinnamic acid (CHCA) were obtained from Bruker Daltonics. All other chemicals (analytical grade) were purchased from Sigma−Aldrich. Sample Fabrication. The porous TiO2 layer was applied on the fluorine-doped tin oxide (FTO) glass substrates (15 mm × 25 mm) and then sensitized with the D35 or LEG4 dyes, according to previously described procedures.5 For in situ photoelectrochemical polymerization of bis-EDOT, a threeelectrode cell was employed. Dye/TiO2/FTO was the working electrode. The counter and reference electrodes were stainless steel and aqueous Ag/AgCl (3 M NaCl), respectively. Two alternative electrolytes for the polymerization of bis-EDOT were used: (a) organic electrolyte, with 5 mM bis-EDOT and 0.1 M LiN(CF3SO2)2 in acetonitrile; (b) aqueous micellar electrolyte, a saturated solution of EDOT (
