- Email: [email protected]
Photoelectrochromic devices: Influence of device architecture and electrolyte composition
Electrochimica Acta 219 (2016) 99–106
Contents lists available at ScienceDirect
Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta
Photoelectrochromic devices: Influence of device architecture and electrolyte composition Cláudia Costa, Isabel Mesquita, Luísa Andrade, Adélio Mendes* LEPABE, Departamento de Engenharia, Universidade do Porto – Faculdade de Engenharia, Rua Dr Roberto Frias, s/n 4200-465 Porto, Portugal
A R T I C L E I N F O
Article history: Received 23 June 2016 Received in revised form 9 September 2016 Accepted 27 September 2016 Available online 28 September 2016 Keywords: Solar cell Electrochromic device Smart windows Photoelectrochromic device Electrochromic solar cell
A B S T R A C T
Solar energy harvesting and chromogenic technologies can be integrated together to give self-powered and wireless photoelectrochromic devices (PECD). Due to the similarity in the architecture of both dyesensitized solar cells (DSC) and electrochromic devices (ECD), it is possible to merge these two devices into one combined solar-powered electrochromic device known as DSC-EC. The present work describes the preparation and characterization of electrochromic solar cells using PEDOT:PSS (poly(3,4ethylenedioxythiophene) polystyrene sulfonate) as electrochromic material in the counter-electrode. Different configurations and liquid and polymer-based electrolytes are studied. The resulting electrochromic solar cells were characterized focusing on the determination of solar to electricity conversion efficiency and the color contrast was assessed using color coordinates under simulated solar irradiation. The best DSC-EC configuration originates a color contrast of (DE = 30) at a potential difference of 0.4 V and energy conversion efficiency of 4.9% at VOC of 0.66 V and JSC of 11 mA cm2, when using the liquid electrolyte. On the other hand, the best performing DSC-EC using the polymer-based electrolyte showed a very good color contrast of (DE = 47) at short circuit and an energy conversion efficiency of 1% at VOC of 0.63 V and JSC of 4.5 mA cm2. ã 2016 Published by Elsevier Ltd.
1. Introduction A photoelectrochromic device (PECD) combines synergistically photovoltaic (PV) with electrochromic functions in a single device. When the photovoltaic function is given by a dye sensitized solar cell, the combined device is known as (DSC-EC) [1]; a DSC-EC is self-powered, changing color when illuminated by sunlight. In the last two decades, considerable research efforts have been directed towards materials and devices intended for a dynamic solar control such as smart windows in buildings [2], automotive or aeronautic industries, helmet [3] and smart displays. In particular, buildings integrating “smart” windows are able to improve two features: i) energy efficiency since they reduce the air conditioning needs; and ii) indoor comfort because they attenuate high solar intensity glare [4]. On the other hand, electrochromic windows with no external wiring, powered by the sunlight, are of great interest. DSCs are an important type of thin film photovoltaics, which have attracted much attention since the prominent work by Grätzel in 1991 [5]. These devices show very attractive advantages, like low-cost construct materials, efficient use of the diffuse light
* Corresponding author. E-mail address: [email protected] (A. Mendes). http://dx.doi.org/10.1016/j.electacta.2016.09.142 0013-4686/ã 2016 Published by Elsevier Ltd.
and transparency [5]. The working electrode of a DSC is a mesoporous oxide layer of nanometer size TiO2 particles. Attached to its surface is a monolayer of dye responsible for light absorption. Upon light absorption dye injects an electron into the conduction band of the semiconductor that percolate through the semiconductor network until the electrical contact. The oxidized sensitizer is regenerated by the transference of electrons from the redox couple present in the electrolyte, which in turn receives the electrons from the counter-electrode, normally platinum [6]. Electrochromism is the phenomenon exhibited by some materials that change color when an electric potential is applied; the electrochromic (EC) material, which can be an organic or an inorganic substance, is able to convert between two or more color states upon oxidation or reduction [3]. To take advantage of this phenomenon, namely for smart windows and low electrical consumption displays, an electrochromic device (ECD) should be built. An ECD is composed of two electrodes separated by an electrolyte, where at least one of the electrodes is transparent, such as a TCO coated glass substrate; in the most common configuration, the EC material is deposited on top of this TCO coated substrate [3]. When applied a potential difference between the two electrodes, the electrochromic layer changes the oxidation state, reacts with the electrolyte, and in consequence changes color [3].
100
C. Costa et al. / Electrochimica Acta 219 (2016) 99–106
For coupling a DSC with an ECD, several architectures can be considered. The most common and simple one considers a thin film of electrochromic material deposited on the photoelectrode or on the counter electrode of a DSC [7,8]. The DSC architecture used depends on the choice of the electrochromic material once the energy levels of the combined materials dictate the electron path through the device. For devices where the electrochromic material is applied on the counterelectrode (see Fig. 1), and upon light absorption, dye injects an electron into the conduction band of the semiconductor that percolates through the semiconductor network until the electrical contact. The oxidized sensitizer is regenerated by the transfer of electrons from the redox couple present in the electrolyte which in turn receives the electrons from the counter-electrode, the EC material film changes color during the process (by being reduced). For electrochromic materials with transparent/colored states, it is possible to have normally transparent devices or normally colored devices depending on the electrochromic material used and on the cell architecture. If the ground state of the electrochromic material is transparent the DSC-EC should be named normally transparent; likewise, if the ground state of the electrochromic material is colored this state should be named normally colored. The first DSC-EC device was reported in 1996 by Bechinger et al. [9] and concerns an integrated photoelectrochromic device using DSC technology, instead of coupling a solar energy harvesting device followed by a full electrochromic device in a sequential arrangement, as described before [10,11]. This work successfully describes a simple architecture where the counter-electrode of the DSC was replaced by a tungsten oxide layer in a normally transparent arrangement; a transmittance variation of 17% was reported. After Bechinger’s work, in 1999, Will et al., [12] reported a different DSC-EC architecture. These authors adsorbed the EC material (viologen) and the dye in the DSC photoelectrode; however, this architecture brought additional problems such as electron transfer between viologens adsorbed at the same or adjacent nanocrystals. In 2001, Hauch et al., [13] described a DSCEC device where the electrochromic WO3 layer was applied over the TCO glass substrate and under the TiO2 photoelectrode. The color contrast obtained was good (transmittance variation of 40%) but no solar to electricity conversion efficiency was reported. Later, Hsu et al. [14] and Wu et al. [7] used electrochromic material
PProDOT-Et2 as counter-electrode (instead of the platinum layer). This arrangement exhibited a solar to electrical energy conversion efficiency below 1% and transmittance variations of c.a. 40%. In 2012, Yang et al. [15] reported a device using a PProDOT-Me2 layer also as counter-electrode but now on top of the platinum layer; this arrangement exhibited a solar to electricity conversion efficiency of ca. 1% and a transmittance variation of 30%. More recently, in 2014, Cannavale et al. [16] described a micropatterned bifunctional counter-electrode made of platinum layer stripes intercalated with WO3 over ITO layer stripes; both types of stripes are not electrically connected. The platinum and WO3 strips were then electrically connected to the photoelectrode through different electric circuits. When the platinum circuit is connected, the cell displayed PV function, while when the WO3 circuit is connected, the cell change color displaying an electrochromic function; this complex cell displayed a conversion energy efficiency of ca. 3% and the transmittance variation was c.a. 25%. In 2015, Cannavale et al. [17] also presented for the first time a perovskite-based photovoltachromic device with self-adaptive transparency. The combination of semi-transparent perovskite photovoltaic and solid-state electrochromic cells enabled fully solid-state photovoltachromic devices with 26% (or 16%) average visible transmittance and 3.7% (or 5.5%) maximum light power conversion efficiency. The present work uses the well-known electrochromic material PEDOT:PSS (poly(3,4-ethylenedioxythiophene) polystyrene sulfonate) applied on the counter-electrode – Fig. 1. Despite being organic, PEDOT:PSS is quite stable and shows a high contrast color [18,25]. PEDOT:PSS layer was applied directly over the TCO glass substrate and over a platinum layer. Two electrolytes where considered, a commercial liquid electrolyte widely used in DSC devices (AN-50, Solaronix, Switzerland) and a semi-solid electrolyte (UV cured PEO (poly(ethylene oxide) based), used with various concentrations of lithium salts. UV-cured electrolytes are being extensively studied for applications such as in DSC devices [19], especially the UV-polymerization of PEO/PEG-based electrolytes [20–22]. This electrolyte was provided by company YD Ynvisible S. A. (http://www.ynvisible.com/) which works in the field of electrochromism; the exact composition of it is a trade secret. The electrolyte showed a quite good performance in electrochromic devices, [18,25,30]. The use of this electrolyte can be of
Fig. 1. EC-DSC device architecture with the PEDOT:PSS layer deposited on the counter-electrode: a) Applied directly on the TCO glass substrate; b) Applied on the platinum layer.
C. Costa et al. / Electrochimica Acta 219 (2016) 99–106
significant value when merging electrochromic displays and DSC devices, being the electrolyte the bridging element. 2. Experimental Section 2.1. EC-DSC Device Preparation The working electrode substrate is FTO glass (TCO22-7, 2.2 mm thickness, 7 V/square, Solaronix, Switzerland) adequately cleaned [23]. FTO glass substrates were immersed into a 40 mM aqueous TiCl4 (titanium (IV) chloride, 99.99%, Acros Organic) solution at 70 C for 20 min to improve the adhesion of TiO2 paste to the substrate, washed with ethanol and dried with N2. A layer of TiO2 paste (Ti-Nanoxide T/SP, Solaronix, Switzerland) was coated on the FTO glass by screen-printing, kept at room temperature for 20 min and then dried for 5 min at 120 C. The screen-printing procedure was repeated two more times to reach a thickness of 12 mm of TiO2 paste (0.2 cm2 of circular active area). After drying the films at 120 C, they were gradually heated (10 C/min) up to 475 C and left at this temperature for 30 min. After cooling until room temperature, the TiO2 electrodes were immersed into a 1 mM ruthenizer 535-bisTBA dye (commonly called as N719, Solaronix, Switzerland) in ethanol and kept at room temperature for 12 h. To prepare the counter electrodes, two holes were drilled on the FTO glass using a drilling machine with diamond tip. The FTO substrates were washed and treated in UV-O3 system like the working electrodes. PEDOT:PSS PH1000 (Heraeus Clevios, Germany) diluted in water, mass fraction of 50%, was deposited by spray and then heated at 120 C for 10 min; PEDOT:PSS was deposited above the platinum layer and replacing it. The sensitized TiO2 electrodes and the EC material on the counter-electrode were assembled into a sandwich type cell and sealed with a hot-melt gasket of 25 mm thickness – Surlyn1 (Meltonix 1170-25, Solaronix, Switzerland) by hot-pressing. The electrolytes were injected into the cell through the holes made on the counter-electrode side. Two different electrolytes were used: a commercial available liquid electrolyte AN 50, often used for DSCs assembling (Solaronix, Switzerland), and an UV cured poly (ethylene oxide) (PEO) electrolyte. The UV cured PEO electrolyte, which presents a gel appearance, is composed by a polymer based on PEO containing organic groups that promote the polymerization of the polymeric chains, a mixture of aprotic solvents, a lithium salt and a photoinitiator, this formulation was provided by YD Ynvisible S.A. The electrolytes were tested with different dopants as lithium iodide and lithium perchlorate (Sigma-Aldrich). The cells filled with UV cured PEO electrolyte were cured under UV light for 2 min before closing the holes – the short time minimizes the degradation of the dye adsorbed onto TiO2. These same holes were then sealed by Surlyn1 and a cover glass using a soldering iron. In order to ensure good electrical contacts during the electrochemical measurements, the edges of the FTO outside of the cell were painted with conductive silver paste (186-3600, RS).
101
Fig. 1 and Table 1 show the architecture and electrolyte compositions considered; basically, the two selected electrolytes were used with the PEDOT:PSS layer (EC material). Devices without the platinum layer were also assessed. The compositions of the tested electrolytes are displayed in Table 1. To facilitate the description, hereafter the devices assembled are referred as D[composition number][PEDOT or Pt, corresponding to PEDOT:PSS or PEDOT:PSS/Pt as counter-electrode, respectively]. For example, D#4Pt refers to a device filled with electrolyte AN 50 doped with 8% of mass fraction of Lil and with a PEDOT:PSS layer applied on the Pt counter-electrode. For comparison, a reference DSC was analyzed and compared with the DSC-EC devices produced. The reference DSC is composed by a Pt counterelectrode and the electrolyte is AN 50 with mass fraction of 8% of LiI. 2.2. Photoelectrochemical Characterization A 150 W xenon light source (Oriel class A simulator, Newport USA) irradiating 96 mW cm2 at the surface of the EC-DSC device and equipped with an air mass filter of 1.5 (Newport, USA) was used. IV characteristics curves were obtained by applying an external potential bias and measuring the generated photocurrent with a ZENNIUM workstation (Ref. 2425-C, Zahner Elektrik, Germany). 2.3. Cyclic Voltammetry Cyclic voltammetry is a widely used technique to obtain the oxidation and reduction peaks of an electrochromic material. The measurements were performed using an AUTOLAB electrochemical station (PGSTAT 12/30/302) in three and two electrode configurations: a) Three electrode configuration – the working electrodes were 1) glass/FTO, 2) glass/FTO/Pt and 3) glass/FTO/PEDOT:PSS. These working electrodes were immersed in a synthetic electrolyte of propylene carbonate (PC) (Sigma-Aldrich) containing [NBu4] [BF4] (0.1 M - Sigma-Aldrich) as supporting electrolyte salt and the redox species LiI (lithium iodide - 3 mM) and I2 (iodine 2.9 mM) (both from Sigma-Aldrich). The counter-electrode used was a platinum wire while the reference electrode was Ag/AgCl electrode (Metrohm 6.0726.100). The solutions were bubbled with nitrogen and kept under a nitrogen atmosphere during the experiment. b) Two-electrode configuration – the working and the counterelectrodes were both made of a PEDOT:PSS layer applied on top of TCO glass substrates (symmetric devices), with the propylene carbonate electrolyte filling the gap between them. The dummy cell was sealed using a thin frame of Surlyn1 and the reference electrode was connected to the counter-electrode [24].
2.4. Color contrast Table 1 Electrolyte compositions considered. Composition number
Electrolyte
salt
Mass fraction of salt/%
1 2 3 4 5 6 7 8 9
AN 50
LiClO4 LiI
8 2 4 8 12 2 4 8 12
UV cured PEO
The use of color coordinates to calculate the color contrast of optical devices is widely used in electrochromism field because it is considered the most accurate way to quantify the human perception of color [25]. However, the use of this methodology is not common in the DSC-EC field. The electrochromic solar cell color contrast was obtained placing it beneath a light source side by side with a colorChecker1 board (Fig. 16). Digital photos were taken at defined intervals, starting with the electrochromic solar cell at its oxidized form and finishing with it at its reduced form, and the color on CIELAB coordinates was obtained using Matlab
102
C. Costa et al. / Electrochimica Acta 219 (2016) 99–106
software (version Matlab R2010b). The color contrast (DE) is calculated according to Eq. (1), where L* measures the lightness; +a* relates to the red direction; a* is the green direction; +b* is the yellow direction and b* is the blue direction [25]: qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð1Þ DE ¼ ðDL Þ2 þ ðDa Þ2 þ Db 2 whereDL ¼ jLreduction Loxidation j, Da ¼ jareduction aoxidation j and Db ¼ jbreduction boxidation j.
Table 2 Photovoltaic parameters of the DSC-EC with the two electrolytes tested and with PEDOT:PSS and/or Platinum as counter electrode. AN-50 electrolyte
VOC/V Jsc/mA cm2 FF h/% Color contrast DE
Reference DSC
D#4PEDOT
D#4Pt
0.690 10.95 0.711 5.37 –
0.660 11.04 0.668 4.87 30
0.670 10.11 0.712 4.75 9
3. Results and Discussion The electrolytes used in electrochromic devices are normally doped with lithium salt to promote the change of the redox state and, consequently, the color of the electrochromic material [26,27]. In the case of commercial AN 50 electrolyte, a lithium salt is already present, but the producer does disclose neither the exact substance nor its concentration; the concentration of the iodide/ triiodide pair is 50 mM [28]. When building DSC-EC devices using AN 50 electrolyte as received, the devices present photovoltaic activity but do not change color. This fact indicates that the lithium salt existing in the initial composition of the commercial electrolyte is not suitable to build an EC-DSC device. The most used salts for DSC-EC applications are LiI and LiClO4 in a mass fractions of 2%–12% [3,5,14]. To minimize leakage problems related to the use of liquid electrolytes it was decided to investigate also an UV cured electrolyte developed by company Ynvisible, which showed good results in electrochromic devices [29]. In this study, DSC-EC devices were optimized to produce the highest solar energy conversion efficiency and the color contrast was assessed for comparison. The architecture of the device was first addressed and then the electrolyte type and composition. 3.1. Study of the device architecture DSC-EC devices with PEDOT:PSS applied over the platinum counter-electrode layer and applied directly on the FTO layer were evaluated – Fig. 1; the electrolyte composition is given in Table 1, composition #4. Fig. 2 displays the characteristic curves of the devices and Table 2 show the performing parameters of the device: the open circuit potential, VOC, the short circuit current density, JSC, the fill factor, FF, and the energy conversion efficiency, h. Table 2 also indicates the color change observed between the shortcircuited and open circuit device. Device with PEDOT:PSS applied directly on the FTO layer (D#4PEDOT) presents a slightly better
photovoltaic performance than the other device (D#4Pt). Both characteristic curves show similar VOC, but D#4PEDOT shows higher JSC. For comparison, the photovoltaic performance of the reference DSC is also displayed. This reference DSC device exhibits slightly higher VOC than the other devices, but a JSC similar to that of the D#4PEDOT device. PEDOT:PSS layer alone in the counter-electrode displays an energy conversion efficiency similar to the devices using also a Pt layer, meaning that the PEDOT:PSS layer has a comparable electrochemical catalytic activity to the Pt layer alone. Fig. 3 shows the cyclic voltammetry in a three-electrode configuration of three counter-electrodes: a PEDOT:PSS layer alone, a Pt alone and just the bare FTO. The voltammogram shows that PEDOT:PSS layer catalyzes the redox reactions of I/I3 (redox peaks a, b, c and d) as the platinum layer [30]. For PEDOT:PSS working electrode, one more pair of peaks is found, which are related to the oxidation/reduction of the PEDOT:PSS itself (peaks e and f). The energy diagram of the two architectures is shown in Fig. 4. The PEDOT:PSS work function (5.2 eV) [31] is slightly higher than the platinum one (5.6 eV) [32]. In the PEDOT:PSS/Pt arrangement, the low energy electrons pathway is from the Pt layer to the PEDOT:PSS layer, originating the reduction of the electrochromic material (the PEDOT:PSS material) and the reduction of the electrolyte afterwards; but electrons can also short cut from the Pt layer to the electrolyte, since PEDOT:PSS layer is permeable and there is electrolyte contacting directly the Pt layer. In this event, the electrochromic layer will be less reduced exhibiting a lower color change. This conclusion is in agreement with the observation of a much stronger color change when only the PEDOT:PSS layer is present – see Table 2. Configuration with just a PEDOT:PSS layer at the counter-electrode was considered for the next experiments.
Fig. 2. IV curves of D#4PEDOT and D#4Pt devices. For comparison it is also shown the IV curve of the reference DSC.
C. Costa et al. / Electrochimica Acta 219 (2016) 99–106
103
Fig. 3. Cyclic voltammetry in a three electrode configuration, where the working electrode are A) single Pt layer and B) single PEDOT:PSS layer (electrolyte: 3 mM of LiI and 2.9 mM of I2 in PC). Bare FTO working electrode is used as a reference.
3.2. Electrolyte Optimization
Fig. 4. Energy diagram and working principle of a DSC-EC device with the electrochromic material in the counter-electrode.
Fig. 5 presents the IV curves of DSC-EC devices with a mass fraction of 8% (0.8 M) of salts LiClO4 and LiI (electrolyte compositions #1 and #4 respectively in Table 1). Salt LiI originates a significant photocurrent enhancement, though a small decrease on VOC is also observed; the energy conversion efficiency of both devices is 3.7% and 4.9%,for D#1PEDOT and D#4PEDOT respectively. The higher current density exhibited by device D#4PEDOT was assigned to the greater mobility of iodide (I) compared with perchlorate ion (ClO4)[33]. Fig. 6 shows two-electrode voltammograms of symmetric PEDOT:PSS electrochromic device loaded with PC electrolyte and salts LiI and LiClO4 at 8% of mass fraction. It is clear that device loaded with LiI salt exhibits much higher currents indicative of a much stronger electrochromic effect or color change. Salt LiI is related with better photovoltaic and electrochromic performances. Further optimization of the DSC-EC device was then directed to the LiI concentration. Devices D#2PEDOT to D#5PEDOT were built with mass fraction of 2%, 4%, 8% and 12% of LiI, respectively, and characterized – Fig. 7 and Table 3.
Fig. 5. IV curves of DSC-EC devices D#1PEDOT and D#4PEDOT.
104
C. Costa et al. / Electrochimica Acta 219 (2016) 99–106 Table 3 Photovoltaic parameters of the DSC-EC devices D#2PEDOT, D#3PEDOT, D#4PEDOT and D#5PEDOT. Mass fraction of LiI
VOC/V Jsc/mA cm2 FF Efficiency/%
Fig. 6. Cyclic voltammetry of symmetric PEDOT:PSS electrochromic devices with two electrolytes based on salts LiI and LiClO4.
The best energy conversion efficiency of the DSC-EC device was obtained adding 8% of mass fraction of LiI to electrolyte AN 50. The significant efficiency drop of device D#5PEDOT with 12% mass fraction of LiI was assigned to an osmotic effect that leads to the decrease of charge mobility at the PEDOT:PSS layer [34]. Along with the high solar to electricity conversion efficiency, device D#4PEDOT showed a good color contrast as shown in Fig. 9. Fig. 8 plots the characteristic curves for devices D#6PEDOT to D#9PEDOT, corresponding to UV cured PEO electrolyte with LiI added mass fraction from 2% to 12%. The best energy performing device was D#8PEDOT containing an added LiI mass fraction of 8%, with an energy efficiency of h = 1% (JSC = 4.49 mAcm2, VOC = 0.63 V, FF = 0.284). The IV curves with UV cured PEO electrolyte show very low values of fill factor, around 0.25. This was assigned to the use of a gel electrolyte with higher charge transport resistances. 3.3. Color Contrast Evaluation Fig. 9 shows D#4PEDOT changing color during the IV curve characterization at around 0.4 V. PEDOT:PSS displays a deep blue coloration in its reduced state and remains clear blue in its oxidized state. This device presents a color contrast of 30. In the case of device D#8PEDOT, the color change is observed at lower potential differences, close to 0 V. This system, however,
2 D#2PEDOT
4 D#3PEDOT
8 D#4PEDOT
12 D#5PEDOT
0.680 9.13 0.64 3.98
0.660 9.37 0.65 4.04
0.660 11.04 0.67 4.87
0.650 7.16 0.64 2.98
shows a slower color response compared with the previous one (D#4PEDOT). Fig. 10 shows device D#8PEDOT with a color contrast of 47; this value indicates a great color variation. As it can be seen in Fig. 10, the electrochromic area is higher than the area of dye, which is very relevant for potential industrial purposes. For both devices, it was observed that the bleaching process happens fully and the bleaching kinetic is faster than the coloring kinetic, as in most electrochromic systems [3]. 4. Conclusions DSC-EC device with AN-50 electrolyte modified with 8% of LiI mass fraction, and with a PEDOT:PSS single layer at the counterelectrode (device D#4PEDOT) displayed a color changed at ca. 0.4 V (DE = 30) and the highest energy conversion efficiency of 4.87% at 1 sun (VOC = 0.66 V and JSC = 11.04 mAcm2). DSC-EC devices with UV cured electrolyte modified with 8% of LiI and with a PEDOT:PSS single layer at the counter-electrode (device D#8PEDOT) exhibited a higher color contrast of 47 observed close to 0 V, but a lower energy conversion efficiency (1%). The best DSC-EC configuration was obtained when only a PEDOT:PSS layer is applied at the counter-electrode side; the use of the PEDOT:PSS layer over a Pt layer makes the color change less effective. On the other hand, adding LiI salt to the electrolyte proved to be more effective than using LiClO4 for obtaining larger energy conversion efficiency and a larger color contrast. For applications in smart windows the durability of devices can be compromised because the degradation of the PEDOT:PSS and of the UV-cured electrolyte, when exposed to UV light; in this event an UV filter should be added. A proof of concept DSC-EC with electrolyte AN-50 was aged for 300 hours under constant 0.6 sun illumination at 65 C and after an initial adaptation period of time, the cell displays a constant behavior showing no apparent
Fig. 7. IV curves of DSC-EC devices D#2PEDOT, D#3PEDOT, D#4PEDOT and D#5PEDOT.
C. Costa et al. / Electrochimica Acta 219 (2016) 99–106
105
Fig. 8. IV curves of DSC-EC devices D#6PEDOT to D#9PEDOT.
Biotechnology and Energy – LEPABE) funded by FEDER funds through COMPETE2020 Programa Operacional Competitividade e Internacionalização (POCI) – and by national funds through FCT Fundação para a Ciência e a Tecnologia. L. Andrade acknowledges European Commission through the Seventh Framework Programme, the Specific Programme “Ideas” of the European Research Council for research and technological development as part of an Advanced Grant under grant agreement No 321315. I. Mesquita acknowledges FCT for her PhD grant (PD/BD/105985/2014). References
Fig. 9. Color change of D#4PEDOT device, before and after 0.4 V. Observed during the IV curve characterization.
Fig. 10. Color contrast (DE) calculated for device D#8PEDOT. The image shows the devices in both redox states, under 1 sun irradiance (100 mW cm2 and 1.5 air mass filter).
degradation of the photovoltaic parameters. Finally, it was observed that the electrochemical phenomenon was observed within but also beyond the photovoltaic area, extending to the whole area of PEDOT:PSS application. Acknowledgments This work was financially supported by Project POCI-01-0145FEDER-006939 (Laboratory for Process Engineering, Environment,
[1] K.-S. Ahn, S.J. Yoo, M.-S. Kang, J.-W. Lee, Y.-E. Sung, Tandem dye-sensitized solar cell-powered electrochromic devices for the photovoltaic-powered smart window, Journal of Power Sources 168 (2007) 533–536. [2] R. Baetens, B.P. Jelle, A. Gustavsen, Properties, requirements and possibilities of smart windows for dynamic daylight and solar energy control in buildings: A state-of-the-art review, Solar Energy Materials and Solar Cells 94 (2) (2010) 87–105. [3] P. Monk, R. Mortimer, D. Rosseinsky, Electrochromism and Electrochromic Devices, Cambridge University Press, 2007. [4] C.M. Lampert, Large-area smart glass and integrated photovoltaics, Solar Energy Materials and Solar Cells 76 (4) (2003) 489–499. [5] B. O'Regan, M. Grätzel, A Low-Cost, High-Efficiency Solar-Cell Based on DyeSensitized Colloidal TiO2 Films, Nature 353 (6346) (1991) 737–740. [6] L. Andrade, H.A. Ribeiro, A. Mendes, Dye–Sensitized Solar Cells: An Overview, in Encyclopedia of Inorganic and Bioinorganic Chemistry, John Wiley & Sons, Ltd., 2011. [7] C. Wu, C. Hsu, K. Huang, P. Nien, A photoelectrochromic device based on gel electrolyte with a fast switching rate, Solar Energy Materials & Solar Cells 99 (2012) 148–153. [8] S.K. Deb, Opportunities and challenges in science and technology of WO3 for electrochromic and related applications, Solar Energy Materials & Solar Cells 92 (2008) 245–258. [9] C. Bechinger, S. Ferrere, A. Zaban, J. Sprague, B.A. Gregg, Photoelectrochromic windows and displays, Nature 383 (1996) 608–610. [10] D.K. Benson, H.M. Branz, Design goals and challenges for a photovoltaicpowered electrochromic window covering, Solar Energy Materials and Solar Cells 39 (1995) 203–211. [11] J.N. Bullock, C. Bechinger, D.K. Benson, H.M. Branz, Semi-transparent a-SiC:H solar cells for self-powered photovoltaic-electrochromic devices, Journal of Non-Crystalline Solids 198-200 (1996) 1163–1167. [12] G. Will, G. Boschloo, S.N. Rao, D. Fitzmaurice, Potentiostatic modulation of the lifetime of light-induced charge separation in a heterosupermolecule, J. Phys. Chem. B 103 (1999) 8067–8079. [13] A. Hauch, A. Georg, S. Baumgärtner, U.O. Krašovec, B. Orel, New photoelectrochromic device, Electrochimica Acta 46 (2001) 2131–2136. [14] C.-Y. Hsu, K.-M. Lee, J.-H. Huang, K.R. Justin Thomas, J.T. Lin, K.-C. Ho, A novel photoelectrochromic device with dual application based on poly(3,4alkylenedioxythiophene) thin film and an organic dye, Journal of Power Sources 185 (2008) 1505–1508.
106
C. Costa et al. / Electrochimica Acta 219 (2016) 99–106
[15] S. Yang, J. Zheng, M. Li, C. Xu, A novel photoelectrochromic device based on poly(3,4-(2,2- dimethylpropylenedioxy)thiophene) thin film and dyesensitized solar cell, Solar Energy Materials and Solar Cells 97 (2012) 186–190. [16] A. Cannavale, M. Manca, L. De Marco, R. Grisorio, S. Carallo, G.P. Suranna, G. Gigli, Photovoltachromic device with a micropatterned bifunctional counter electrode, ACS Applied Materials and Interfaces 6 (2014) 2415–2422. [17] A. Cannavale, G.E. Eperon, P. Cossari, A. Abate, H.J. Snaith, G. Gigli, Perovskite photovoltachromic cells for building integration, Energy Environ. Sci. 8 (2015) 1578–1584. [18] L. Gomes, A. Branco, T. Moreira, F. Feliciano, C. Pinheiro, C. Costa, Solar Energy Materials & Solar Cells 144 (2016) 631–640. [19] F. Bella, G. Leftheriotis, G. Griffini, G. Syrrokostas, S. Turri, M. Grätzel, C. Gerbaldi, A new design paradigm for smart windows: Photocurable polymers for quasi-solid photoelectrochromic devices with excellent long-term stability under real outdoor operating conditions, Advanced Functional Materials 26 (2016) 1127–1137. [20] J.R. Nair, L. Porcarelli, F. Bella, C. Gerbaldi, Newly Elaborated Multipurpose Polymer Electrolyte Encompassing RTILs for Smart Energy-Efficient Devices, ACS Applied Materials and Interfaces 7 (23) (2015) 12961–12971. [21] H. Zhao, X. Ding, P. Yang, L. Li, X. Li, Y. Zhang, A novel multi-armed and star-like poly(ethylene oxide) membrane for CO2 separation, Journal of Membrane Science 489 (2015) 258–263. [22] C. Porcarelli, F. Gerbaldi, Super Soft All-Ethylene Oxide Polymer Electrolyte for Safe All-Solid Lithium Batteries, Scientific Reports 6 (2016) art. no. 19892. [23] J. Maçaira, I. Mesquita, L. Andrade, A. Mendes, Temperature Role in the Recombination Reaction on Dye-Sensitized Solar Cells, Phys. Chem. Chem. Phys. 17 (2015) 22699–22710. [24] C. Costa, C. Pinheiro, I.D.S. Henriques, C.A.T. Laia, Inkjet Printing of Sol–Gel Synthesized Hydrated Tungsten Oxide Nanoparticles for Flexible Electrochromic Devices, ACS Appl. Mater. Interfaces 4 (2012) 1330–1340.
[25] P. Tehrani, L.-O. Hennerdal, A.L. Dyer, J.R. Reynolds, M. Berggren, Improving the contrast of all-printed electrochromic polymer on paper displays, Journal of Materials Chemistry 19 (2009) 1799–1802. [26] J. Jensen, M. Hösel, A.L. Dyer, F.C. Krebs, Development and manufacture of polymer-based electrochromic devices, Advanced Functional Materials 25 (2015) 2073–2090. [27] A. Kumar, J.R. Reynolds, Soluble alkyl-substituted poly (ethylenedioxythiophenes) as electrochromic materials, Macromolecules 29 (1996) 7629–7630. [28] http://shop.solaronix.com/iodolyte-an50. html [29] L. Gomes, A. Marques, A. Branco, J. Araújo, M. Simões, S. Cardoso, F. Silva, I. Henriques, C.A.T. Laia, C. Costa, IZO deposition by RF and DC sputtering on paper and application on flexible electrochromic devices, Displays 34 (2013) 326–333. [30] C.K. Hong, H.S. Ko, E.M. Han, K.H. Park, Electrochemical Properties of Electrodeposited PEDOT Counter Electrode for Dye-sensitized Solar Cells, Int. J. Electrochem. Sci. 10 (2015) 5521–5529. [31] J.A. Mikroyannidis, A.N. Kabanakis, D.V. Tsagkournos, P. Balraju, G.D. Sharma, Bulk heterojunction solar cells based on a low band gap soluble bisazopyrrole and the corresponding BF2-azopyrrole complex, J. Mater. Chem. 20 (2010) 6464–6471. [32] J. Wang, J. Zhao, F.E. Osterloh, Photochemical charge transfer observed in nanoscale hydrogen evolving photocatalysts using surface photovoltage spectroscopy, Energy Environ. Sci. 8 (2015) 2970–2976. [33] A.L. Smeigh, J.E. Katz, B.S. Brunschwig, N.S. Lewis, J.K. McCusker, Effect of the Presence of Iodide on the Electron Injection Dynamics of Dye-Sensitized TiO2Based Solar Cells, J. Phys. Chem. C 112 (2008) 18467–18473. [34] M. Eikerling, I. Yu. Kharkats, A.A. Kornyshev, Phenomenological Theory of Electro-osmotic Effect and Water Management in Polymer Electrolyte ProtonConducting Membranes, J. Electrochem. Soc. 145 (1998) 2684–2699.
Contents lists available at ScienceDirect
Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta
Photoelectrochromic devices: Influence of device architecture and electrolyte composition Cláudia Costa, Isabel Mesquita, Luísa Andrade, Adélio Mendes* LEPABE, Departamento de Engenharia, Universidade do Porto – Faculdade de Engenharia, Rua Dr Roberto Frias, s/n 4200-465 Porto, Portugal
A R T I C L E I N F O
Article history: Received 23 June 2016 Received in revised form 9 September 2016 Accepted 27 September 2016 Available online 28 September 2016 Keywords: Solar cell Electrochromic device Smart windows Photoelectrochromic device Electrochromic solar cell
A B S T R A C T
Solar energy harvesting and chromogenic technologies can be integrated together to give self-powered and wireless photoelectrochromic devices (PECD). Due to the similarity in the architecture of both dyesensitized solar cells (DSC) and electrochromic devices (ECD), it is possible to merge these two devices into one combined solar-powered electrochromic device known as DSC-EC. The present work describes the preparation and characterization of electrochromic solar cells using PEDOT:PSS (poly(3,4ethylenedioxythiophene) polystyrene sulfonate) as electrochromic material in the counter-electrode. Different configurations and liquid and polymer-based electrolytes are studied. The resulting electrochromic solar cells were characterized focusing on the determination of solar to electricity conversion efficiency and the color contrast was assessed using color coordinates under simulated solar irradiation. The best DSC-EC configuration originates a color contrast of (DE = 30) at a potential difference of 0.4 V and energy conversion efficiency of 4.9% at VOC of 0.66 V and JSC of 11 mA cm2, when using the liquid electrolyte. On the other hand, the best performing DSC-EC using the polymer-based electrolyte showed a very good color contrast of (DE = 47) at short circuit and an energy conversion efficiency of 1% at VOC of 0.63 V and JSC of 4.5 mA cm2. ã 2016 Published by Elsevier Ltd.
1. Introduction A photoelectrochromic device (PECD) combines synergistically photovoltaic (PV) with electrochromic functions in a single device. When the photovoltaic function is given by a dye sensitized solar cell, the combined device is known as (DSC-EC) [1]; a DSC-EC is self-powered, changing color when illuminated by sunlight. In the last two decades, considerable research efforts have been directed towards materials and devices intended for a dynamic solar control such as smart windows in buildings [2], automotive or aeronautic industries, helmet [3] and smart displays. In particular, buildings integrating “smart” windows are able to improve two features: i) energy efficiency since they reduce the air conditioning needs; and ii) indoor comfort because they attenuate high solar intensity glare [4]. On the other hand, electrochromic windows with no external wiring, powered by the sunlight, are of great interest. DSCs are an important type of thin film photovoltaics, which have attracted much attention since the prominent work by Grätzel in 1991 [5]. These devices show very attractive advantages, like low-cost construct materials, efficient use of the diffuse light
* Corresponding author. E-mail address: [email protected] (A. Mendes). http://dx.doi.org/10.1016/j.electacta.2016.09.142 0013-4686/ã 2016 Published by Elsevier Ltd.
and transparency [5]. The working electrode of a DSC is a mesoporous oxide layer of nanometer size TiO2 particles. Attached to its surface is a monolayer of dye responsible for light absorption. Upon light absorption dye injects an electron into the conduction band of the semiconductor that percolate through the semiconductor network until the electrical contact. The oxidized sensitizer is regenerated by the transference of electrons from the redox couple present in the electrolyte, which in turn receives the electrons from the counter-electrode, normally platinum [6]. Electrochromism is the phenomenon exhibited by some materials that change color when an electric potential is applied; the electrochromic (EC) material, which can be an organic or an inorganic substance, is able to convert between two or more color states upon oxidation or reduction [3]. To take advantage of this phenomenon, namely for smart windows and low electrical consumption displays, an electrochromic device (ECD) should be built. An ECD is composed of two electrodes separated by an electrolyte, where at least one of the electrodes is transparent, such as a TCO coated glass substrate; in the most common configuration, the EC material is deposited on top of this TCO coated substrate [3]. When applied a potential difference between the two electrodes, the electrochromic layer changes the oxidation state, reacts with the electrolyte, and in consequence changes color [3].
100
C. Costa et al. / Electrochimica Acta 219 (2016) 99–106
For coupling a DSC with an ECD, several architectures can be considered. The most common and simple one considers a thin film of electrochromic material deposited on the photoelectrode or on the counter electrode of a DSC [7,8]. The DSC architecture used depends on the choice of the electrochromic material once the energy levels of the combined materials dictate the electron path through the device. For devices where the electrochromic material is applied on the counterelectrode (see Fig. 1), and upon light absorption, dye injects an electron into the conduction band of the semiconductor that percolates through the semiconductor network until the electrical contact. The oxidized sensitizer is regenerated by the transfer of electrons from the redox couple present in the electrolyte which in turn receives the electrons from the counter-electrode, the EC material film changes color during the process (by being reduced). For electrochromic materials with transparent/colored states, it is possible to have normally transparent devices or normally colored devices depending on the electrochromic material used and on the cell architecture. If the ground state of the electrochromic material is transparent the DSC-EC should be named normally transparent; likewise, if the ground state of the electrochromic material is colored this state should be named normally colored. The first DSC-EC device was reported in 1996 by Bechinger et al. [9] and concerns an integrated photoelectrochromic device using DSC technology, instead of coupling a solar energy harvesting device followed by a full electrochromic device in a sequential arrangement, as described before [10,11]. This work successfully describes a simple architecture where the counter-electrode of the DSC was replaced by a tungsten oxide layer in a normally transparent arrangement; a transmittance variation of 17% was reported. After Bechinger’s work, in 1999, Will et al., [12] reported a different DSC-EC architecture. These authors adsorbed the EC material (viologen) and the dye in the DSC photoelectrode; however, this architecture brought additional problems such as electron transfer between viologens adsorbed at the same or adjacent nanocrystals. In 2001, Hauch et al., [13] described a DSCEC device where the electrochromic WO3 layer was applied over the TCO glass substrate and under the TiO2 photoelectrode. The color contrast obtained was good (transmittance variation of 40%) but no solar to electricity conversion efficiency was reported. Later, Hsu et al. [14] and Wu et al. [7] used electrochromic material
PProDOT-Et2 as counter-electrode (instead of the platinum layer). This arrangement exhibited a solar to electrical energy conversion efficiency below 1% and transmittance variations of c.a. 40%. In 2012, Yang et al. [15] reported a device using a PProDOT-Me2 layer also as counter-electrode but now on top of the platinum layer; this arrangement exhibited a solar to electricity conversion efficiency of ca. 1% and a transmittance variation of 30%. More recently, in 2014, Cannavale et al. [16] described a micropatterned bifunctional counter-electrode made of platinum layer stripes intercalated with WO3 over ITO layer stripes; both types of stripes are not electrically connected. The platinum and WO3 strips were then electrically connected to the photoelectrode through different electric circuits. When the platinum circuit is connected, the cell displayed PV function, while when the WO3 circuit is connected, the cell change color displaying an electrochromic function; this complex cell displayed a conversion energy efficiency of ca. 3% and the transmittance variation was c.a. 25%. In 2015, Cannavale et al. [17] also presented for the first time a perovskite-based photovoltachromic device with self-adaptive transparency. The combination of semi-transparent perovskite photovoltaic and solid-state electrochromic cells enabled fully solid-state photovoltachromic devices with 26% (or 16%) average visible transmittance and 3.7% (or 5.5%) maximum light power conversion efficiency. The present work uses the well-known electrochromic material PEDOT:PSS (poly(3,4-ethylenedioxythiophene) polystyrene sulfonate) applied on the counter-electrode – Fig. 1. Despite being organic, PEDOT:PSS is quite stable and shows a high contrast color [18,25]. PEDOT:PSS layer was applied directly over the TCO glass substrate and over a platinum layer. Two electrolytes where considered, a commercial liquid electrolyte widely used in DSC devices (AN-50, Solaronix, Switzerland) and a semi-solid electrolyte (UV cured PEO (poly(ethylene oxide) based), used with various concentrations of lithium salts. UV-cured electrolytes are being extensively studied for applications such as in DSC devices [19], especially the UV-polymerization of PEO/PEG-based electrolytes [20–22]. This electrolyte was provided by company YD Ynvisible S. A. (http://www.ynvisible.com/) which works in the field of electrochromism; the exact composition of it is a trade secret. The electrolyte showed a quite good performance in electrochromic devices, [18,25,30]. The use of this electrolyte can be of
Fig. 1. EC-DSC device architecture with the PEDOT:PSS layer deposited on the counter-electrode: a) Applied directly on the TCO glass substrate; b) Applied on the platinum layer.
C. Costa et al. / Electrochimica Acta 219 (2016) 99–106
significant value when merging electrochromic displays and DSC devices, being the electrolyte the bridging element. 2. Experimental Section 2.1. EC-DSC Device Preparation The working electrode substrate is FTO glass (TCO22-7, 2.2 mm thickness, 7 V/square, Solaronix, Switzerland) adequately cleaned [23]. FTO glass substrates were immersed into a 40 mM aqueous TiCl4 (titanium (IV) chloride, 99.99%, Acros Organic) solution at 70 C for 20 min to improve the adhesion of TiO2 paste to the substrate, washed with ethanol and dried with N2. A layer of TiO2 paste (Ti-Nanoxide T/SP, Solaronix, Switzerland) was coated on the FTO glass by screen-printing, kept at room temperature for 20 min and then dried for 5 min at 120 C. The screen-printing procedure was repeated two more times to reach a thickness of 12 mm of TiO2 paste (0.2 cm2 of circular active area). After drying the films at 120 C, they were gradually heated (10 C/min) up to 475 C and left at this temperature for 30 min. After cooling until room temperature, the TiO2 electrodes were immersed into a 1 mM ruthenizer 535-bisTBA dye (commonly called as N719, Solaronix, Switzerland) in ethanol and kept at room temperature for 12 h. To prepare the counter electrodes, two holes were drilled on the FTO glass using a drilling machine with diamond tip. The FTO substrates were washed and treated in UV-O3 system like the working electrodes. PEDOT:PSS PH1000 (Heraeus Clevios, Germany) diluted in water, mass fraction of 50%, was deposited by spray and then heated at 120 C for 10 min; PEDOT:PSS was deposited above the platinum layer and replacing it. The sensitized TiO2 electrodes and the EC material on the counter-electrode were assembled into a sandwich type cell and sealed with a hot-melt gasket of 25 mm thickness – Surlyn1 (Meltonix 1170-25, Solaronix, Switzerland) by hot-pressing. The electrolytes were injected into the cell through the holes made on the counter-electrode side. Two different electrolytes were used: a commercial available liquid electrolyte AN 50, often used for DSCs assembling (Solaronix, Switzerland), and an UV cured poly (ethylene oxide) (PEO) electrolyte. The UV cured PEO electrolyte, which presents a gel appearance, is composed by a polymer based on PEO containing organic groups that promote the polymerization of the polymeric chains, a mixture of aprotic solvents, a lithium salt and a photoinitiator, this formulation was provided by YD Ynvisible S.A. The electrolytes were tested with different dopants as lithium iodide and lithium perchlorate (Sigma-Aldrich). The cells filled with UV cured PEO electrolyte were cured under UV light for 2 min before closing the holes – the short time minimizes the degradation of the dye adsorbed onto TiO2. These same holes were then sealed by Surlyn1 and a cover glass using a soldering iron. In order to ensure good electrical contacts during the electrochemical measurements, the edges of the FTO outside of the cell were painted with conductive silver paste (186-3600, RS).
101
Fig. 1 and Table 1 show the architecture and electrolyte compositions considered; basically, the two selected electrolytes were used with the PEDOT:PSS layer (EC material). Devices without the platinum layer were also assessed. The compositions of the tested electrolytes are displayed in Table 1. To facilitate the description, hereafter the devices assembled are referred as D[composition number][PEDOT or Pt, corresponding to PEDOT:PSS or PEDOT:PSS/Pt as counter-electrode, respectively]. For example, D#4Pt refers to a device filled with electrolyte AN 50 doped with 8% of mass fraction of Lil and with a PEDOT:PSS layer applied on the Pt counter-electrode. For comparison, a reference DSC was analyzed and compared with the DSC-EC devices produced. The reference DSC is composed by a Pt counterelectrode and the electrolyte is AN 50 with mass fraction of 8% of LiI. 2.2. Photoelectrochemical Characterization A 150 W xenon light source (Oriel class A simulator, Newport USA) irradiating 96 mW cm2 at the surface of the EC-DSC device and equipped with an air mass filter of 1.5 (Newport, USA) was used. IV characteristics curves were obtained by applying an external potential bias and measuring the generated photocurrent with a ZENNIUM workstation (Ref. 2425-C, Zahner Elektrik, Germany). 2.3. Cyclic Voltammetry Cyclic voltammetry is a widely used technique to obtain the oxidation and reduction peaks of an electrochromic material. The measurements were performed using an AUTOLAB electrochemical station (PGSTAT 12/30/302) in three and two electrode configurations: a) Three electrode configuration – the working electrodes were 1) glass/FTO, 2) glass/FTO/Pt and 3) glass/FTO/PEDOT:PSS. These working electrodes were immersed in a synthetic electrolyte of propylene carbonate (PC) (Sigma-Aldrich) containing [NBu4] [BF4] (0.1 M - Sigma-Aldrich) as supporting electrolyte salt and the redox species LiI (lithium iodide - 3 mM) and I2 (iodine 2.9 mM) (both from Sigma-Aldrich). The counter-electrode used was a platinum wire while the reference electrode was Ag/AgCl electrode (Metrohm 6.0726.100). The solutions were bubbled with nitrogen and kept under a nitrogen atmosphere during the experiment. b) Two-electrode configuration – the working and the counterelectrodes were both made of a PEDOT:PSS layer applied on top of TCO glass substrates (symmetric devices), with the propylene carbonate electrolyte filling the gap between them. The dummy cell was sealed using a thin frame of Surlyn1 and the reference electrode was connected to the counter-electrode [24].
2.4. Color contrast Table 1 Electrolyte compositions considered. Composition number
Electrolyte
salt
Mass fraction of salt/%
1 2 3 4 5 6 7 8 9
AN 50
LiClO4 LiI
8 2 4 8 12 2 4 8 12
UV cured PEO
The use of color coordinates to calculate the color contrast of optical devices is widely used in electrochromism field because it is considered the most accurate way to quantify the human perception of color [25]. However, the use of this methodology is not common in the DSC-EC field. The electrochromic solar cell color contrast was obtained placing it beneath a light source side by side with a colorChecker1 board (Fig. 16). Digital photos were taken at defined intervals, starting with the electrochromic solar cell at its oxidized form and finishing with it at its reduced form, and the color on CIELAB coordinates was obtained using Matlab
102
C. Costa et al. / Electrochimica Acta 219 (2016) 99–106
software (version Matlab R2010b). The color contrast (DE) is calculated according to Eq. (1), where L* measures the lightness; +a* relates to the red direction; a* is the green direction; +b* is the yellow direction and b* is the blue direction [25]: qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð1Þ DE ¼ ðDL Þ2 þ ðDa Þ2 þ Db 2 whereDL ¼ jLreduction Loxidation j, Da ¼ jareduction aoxidation j and Db ¼ jbreduction boxidation j.
Table 2 Photovoltaic parameters of the DSC-EC with the two electrolytes tested and with PEDOT:PSS and/or Platinum as counter electrode. AN-50 electrolyte
VOC/V Jsc/mA cm2 FF h/% Color contrast DE
Reference DSC
D#4PEDOT
D#4Pt
0.690 10.95 0.711 5.37 –
0.660 11.04 0.668 4.87 30
0.670 10.11 0.712 4.75 9
3. Results and Discussion The electrolytes used in electrochromic devices are normally doped with lithium salt to promote the change of the redox state and, consequently, the color of the electrochromic material [26,27]. In the case of commercial AN 50 electrolyte, a lithium salt is already present, but the producer does disclose neither the exact substance nor its concentration; the concentration of the iodide/ triiodide pair is 50 mM [28]. When building DSC-EC devices using AN 50 electrolyte as received, the devices present photovoltaic activity but do not change color. This fact indicates that the lithium salt existing in the initial composition of the commercial electrolyte is not suitable to build an EC-DSC device. The most used salts for DSC-EC applications are LiI and LiClO4 in a mass fractions of 2%–12% [3,5,14]. To minimize leakage problems related to the use of liquid electrolytes it was decided to investigate also an UV cured electrolyte developed by company Ynvisible, which showed good results in electrochromic devices [29]. In this study, DSC-EC devices were optimized to produce the highest solar energy conversion efficiency and the color contrast was assessed for comparison. The architecture of the device was first addressed and then the electrolyte type and composition. 3.1. Study of the device architecture DSC-EC devices with PEDOT:PSS applied over the platinum counter-electrode layer and applied directly on the FTO layer were evaluated – Fig. 1; the electrolyte composition is given in Table 1, composition #4. Fig. 2 displays the characteristic curves of the devices and Table 2 show the performing parameters of the device: the open circuit potential, VOC, the short circuit current density, JSC, the fill factor, FF, and the energy conversion efficiency, h. Table 2 also indicates the color change observed between the shortcircuited and open circuit device. Device with PEDOT:PSS applied directly on the FTO layer (D#4PEDOT) presents a slightly better
photovoltaic performance than the other device (D#4Pt). Both characteristic curves show similar VOC, but D#4PEDOT shows higher JSC. For comparison, the photovoltaic performance of the reference DSC is also displayed. This reference DSC device exhibits slightly higher VOC than the other devices, but a JSC similar to that of the D#4PEDOT device. PEDOT:PSS layer alone in the counter-electrode displays an energy conversion efficiency similar to the devices using also a Pt layer, meaning that the PEDOT:PSS layer has a comparable electrochemical catalytic activity to the Pt layer alone. Fig. 3 shows the cyclic voltammetry in a three-electrode configuration of three counter-electrodes: a PEDOT:PSS layer alone, a Pt alone and just the bare FTO. The voltammogram shows that PEDOT:PSS layer catalyzes the redox reactions of I/I3 (redox peaks a, b, c and d) as the platinum layer [30]. For PEDOT:PSS working electrode, one more pair of peaks is found, which are related to the oxidation/reduction of the PEDOT:PSS itself (peaks e and f). The energy diagram of the two architectures is shown in Fig. 4. The PEDOT:PSS work function (5.2 eV) [31] is slightly higher than the platinum one (5.6 eV) [32]. In the PEDOT:PSS/Pt arrangement, the low energy electrons pathway is from the Pt layer to the PEDOT:PSS layer, originating the reduction of the electrochromic material (the PEDOT:PSS material) and the reduction of the electrolyte afterwards; but electrons can also short cut from the Pt layer to the electrolyte, since PEDOT:PSS layer is permeable and there is electrolyte contacting directly the Pt layer. In this event, the electrochromic layer will be less reduced exhibiting a lower color change. This conclusion is in agreement with the observation of a much stronger color change when only the PEDOT:PSS layer is present – see Table 2. Configuration with just a PEDOT:PSS layer at the counter-electrode was considered for the next experiments.
Fig. 2. IV curves of D#4PEDOT and D#4Pt devices. For comparison it is also shown the IV curve of the reference DSC.
C. Costa et al. / Electrochimica Acta 219 (2016) 99–106
103
Fig. 3. Cyclic voltammetry in a three electrode configuration, where the working electrode are A) single Pt layer and B) single PEDOT:PSS layer (electrolyte: 3 mM of LiI and 2.9 mM of I2 in PC). Bare FTO working electrode is used as a reference.
3.2. Electrolyte Optimization
Fig. 4. Energy diagram and working principle of a DSC-EC device with the electrochromic material in the counter-electrode.
Fig. 5 presents the IV curves of DSC-EC devices with a mass fraction of 8% (0.8 M) of salts LiClO4 and LiI (electrolyte compositions #1 and #4 respectively in Table 1). Salt LiI originates a significant photocurrent enhancement, though a small decrease on VOC is also observed; the energy conversion efficiency of both devices is 3.7% and 4.9%,for D#1PEDOT and D#4PEDOT respectively. The higher current density exhibited by device D#4PEDOT was assigned to the greater mobility of iodide (I) compared with perchlorate ion (ClO4)[33]. Fig. 6 shows two-electrode voltammograms of symmetric PEDOT:PSS electrochromic device loaded with PC electrolyte and salts LiI and LiClO4 at 8% of mass fraction. It is clear that device loaded with LiI salt exhibits much higher currents indicative of a much stronger electrochromic effect or color change. Salt LiI is related with better photovoltaic and electrochromic performances. Further optimization of the DSC-EC device was then directed to the LiI concentration. Devices D#2PEDOT to D#5PEDOT were built with mass fraction of 2%, 4%, 8% and 12% of LiI, respectively, and characterized – Fig. 7 and Table 3.
Fig. 5. IV curves of DSC-EC devices D#1PEDOT and D#4PEDOT.
104
C. Costa et al. / Electrochimica Acta 219 (2016) 99–106 Table 3 Photovoltaic parameters of the DSC-EC devices D#2PEDOT, D#3PEDOT, D#4PEDOT and D#5PEDOT. Mass fraction of LiI
VOC/V Jsc/mA cm2 FF Efficiency/%
Fig. 6. Cyclic voltammetry of symmetric PEDOT:PSS electrochromic devices with two electrolytes based on salts LiI and LiClO4.
The best energy conversion efficiency of the DSC-EC device was obtained adding 8% of mass fraction of LiI to electrolyte AN 50. The significant efficiency drop of device D#5PEDOT with 12% mass fraction of LiI was assigned to an osmotic effect that leads to the decrease of charge mobility at the PEDOT:PSS layer [34]. Along with the high solar to electricity conversion efficiency, device D#4PEDOT showed a good color contrast as shown in Fig. 9. Fig. 8 plots the characteristic curves for devices D#6PEDOT to D#9PEDOT, corresponding to UV cured PEO electrolyte with LiI added mass fraction from 2% to 12%. The best energy performing device was D#8PEDOT containing an added LiI mass fraction of 8%, with an energy efficiency of h = 1% (JSC = 4.49 mAcm2, VOC = 0.63 V, FF = 0.284). The IV curves with UV cured PEO electrolyte show very low values of fill factor, around 0.25. This was assigned to the use of a gel electrolyte with higher charge transport resistances. 3.3. Color Contrast Evaluation Fig. 9 shows D#4PEDOT changing color during the IV curve characterization at around 0.4 V. PEDOT:PSS displays a deep blue coloration in its reduced state and remains clear blue in its oxidized state. This device presents a color contrast of 30. In the case of device D#8PEDOT, the color change is observed at lower potential differences, close to 0 V. This system, however,
2 D#2PEDOT
4 D#3PEDOT
8 D#4PEDOT
12 D#5PEDOT
0.680 9.13 0.64 3.98
0.660 9.37 0.65 4.04
0.660 11.04 0.67 4.87
0.650 7.16 0.64 2.98
shows a slower color response compared with the previous one (D#4PEDOT). Fig. 10 shows device D#8PEDOT with a color contrast of 47; this value indicates a great color variation. As it can be seen in Fig. 10, the electrochromic area is higher than the area of dye, which is very relevant for potential industrial purposes. For both devices, it was observed that the bleaching process happens fully and the bleaching kinetic is faster than the coloring kinetic, as in most electrochromic systems [3]. 4. Conclusions DSC-EC device with AN-50 electrolyte modified with 8% of LiI mass fraction, and with a PEDOT:PSS single layer at the counterelectrode (device D#4PEDOT) displayed a color changed at ca. 0.4 V (DE = 30) and the highest energy conversion efficiency of 4.87% at 1 sun (VOC = 0.66 V and JSC = 11.04 mAcm2). DSC-EC devices with UV cured electrolyte modified with 8% of LiI and with a PEDOT:PSS single layer at the counter-electrode (device D#8PEDOT) exhibited a higher color contrast of 47 observed close to 0 V, but a lower energy conversion efficiency (1%). The best DSC-EC configuration was obtained when only a PEDOT:PSS layer is applied at the counter-electrode side; the use of the PEDOT:PSS layer over a Pt layer makes the color change less effective. On the other hand, adding LiI salt to the electrolyte proved to be more effective than using LiClO4 for obtaining larger energy conversion efficiency and a larger color contrast. For applications in smart windows the durability of devices can be compromised because the degradation of the PEDOT:PSS and of the UV-cured electrolyte, when exposed to UV light; in this event an UV filter should be added. A proof of concept DSC-EC with electrolyte AN-50 was aged for 300 hours under constant 0.6 sun illumination at 65 C and after an initial adaptation period of time, the cell displays a constant behavior showing no apparent
Fig. 7. IV curves of DSC-EC devices D#2PEDOT, D#3PEDOT, D#4PEDOT and D#5PEDOT.
C. Costa et al. / Electrochimica Acta 219 (2016) 99–106
105
Fig. 8. IV curves of DSC-EC devices D#6PEDOT to D#9PEDOT.
Biotechnology and Energy – LEPABE) funded by FEDER funds through COMPETE2020 Programa Operacional Competitividade e Internacionalização (POCI) – and by national funds through FCT Fundação para a Ciência e a Tecnologia. L. Andrade acknowledges European Commission through the Seventh Framework Programme, the Specific Programme “Ideas” of the European Research Council for research and technological development as part of an Advanced Grant under grant agreement No 321315. I. Mesquita acknowledges FCT for her PhD grant (PD/BD/105985/2014). References
Fig. 9. Color change of D#4PEDOT device, before and after 0.4 V. Observed during the IV curve characterization.
Fig. 10. Color contrast (DE) calculated for device D#8PEDOT. The image shows the devices in both redox states, under 1 sun irradiance (100 mW cm2 and 1.5 air mass filter).
degradation of the photovoltaic parameters. Finally, it was observed that the electrochemical phenomenon was observed within but also beyond the photovoltaic area, extending to the whole area of PEDOT:PSS application. Acknowledgments This work was financially supported by Project POCI-01-0145FEDER-006939 (Laboratory for Process Engineering, Environment,
[1] K.-S. Ahn, S.J. Yoo, M.-S. Kang, J.-W. Lee, Y.-E. Sung, Tandem dye-sensitized solar cell-powered electrochromic devices for the photovoltaic-powered smart window, Journal of Power Sources 168 (2007) 533–536. [2] R. Baetens, B.P. Jelle, A. Gustavsen, Properties, requirements and possibilities of smart windows for dynamic daylight and solar energy control in buildings: A state-of-the-art review, Solar Energy Materials and Solar Cells 94 (2) (2010) 87–105. [3] P. Monk, R. Mortimer, D. Rosseinsky, Electrochromism and Electrochromic Devices, Cambridge University Press, 2007. [4] C.M. Lampert, Large-area smart glass and integrated photovoltaics, Solar Energy Materials and Solar Cells 76 (4) (2003) 489–499. [5] B. O'Regan, M. Grätzel, A Low-Cost, High-Efficiency Solar-Cell Based on DyeSensitized Colloidal TiO2 Films, Nature 353 (6346) (1991) 737–740. [6] L. Andrade, H.A. Ribeiro, A. Mendes, Dye–Sensitized Solar Cells: An Overview, in Encyclopedia of Inorganic and Bioinorganic Chemistry, John Wiley & Sons, Ltd., 2011. [7] C. Wu, C. Hsu, K. Huang, P. Nien, A photoelectrochromic device based on gel electrolyte with a fast switching rate, Solar Energy Materials & Solar Cells 99 (2012) 148–153. [8] S.K. Deb, Opportunities and challenges in science and technology of WO3 for electrochromic and related applications, Solar Energy Materials & Solar Cells 92 (2008) 245–258. [9] C. Bechinger, S. Ferrere, A. Zaban, J. Sprague, B.A. Gregg, Photoelectrochromic windows and displays, Nature 383 (1996) 608–610. [10] D.K. Benson, H.M. Branz, Design goals and challenges for a photovoltaicpowered electrochromic window covering, Solar Energy Materials and Solar Cells 39 (1995) 203–211. [11] J.N. Bullock, C. Bechinger, D.K. Benson, H.M. Branz, Semi-transparent a-SiC:H solar cells for self-powered photovoltaic-electrochromic devices, Journal of Non-Crystalline Solids 198-200 (1996) 1163–1167. [12] G. Will, G. Boschloo, S.N. Rao, D. Fitzmaurice, Potentiostatic modulation of the lifetime of light-induced charge separation in a heterosupermolecule, J. Phys. Chem. B 103 (1999) 8067–8079. [13] A. Hauch, A. Georg, S. Baumgärtner, U.O. Krašovec, B. Orel, New photoelectrochromic device, Electrochimica Acta 46 (2001) 2131–2136. [14] C.-Y. Hsu, K.-M. Lee, J.-H. Huang, K.R. Justin Thomas, J.T. Lin, K.-C. Ho, A novel photoelectrochromic device with dual application based on poly(3,4alkylenedioxythiophene) thin film and an organic dye, Journal of Power Sources 185 (2008) 1505–1508.
106
C. Costa et al. / Electrochimica Acta 219 (2016) 99–106
[15] S. Yang, J. Zheng, M. Li, C. Xu, A novel photoelectrochromic device based on poly(3,4-(2,2- dimethylpropylenedioxy)thiophene) thin film and dyesensitized solar cell, Solar Energy Materials and Solar Cells 97 (2012) 186–190. [16] A. Cannavale, M. Manca, L. De Marco, R. Grisorio, S. Carallo, G.P. Suranna, G. Gigli, Photovoltachromic device with a micropatterned bifunctional counter electrode, ACS Applied Materials and Interfaces 6 (2014) 2415–2422. [17] A. Cannavale, G.E. Eperon, P. Cossari, A. Abate, H.J. Snaith, G. Gigli, Perovskite photovoltachromic cells for building integration, Energy Environ. Sci. 8 (2015) 1578–1584. [18] L. Gomes, A. Branco, T. Moreira, F. Feliciano, C. Pinheiro, C. Costa, Solar Energy Materials & Solar Cells 144 (2016) 631–640. [19] F. Bella, G. Leftheriotis, G. Griffini, G. Syrrokostas, S. Turri, M. Grätzel, C. Gerbaldi, A new design paradigm for smart windows: Photocurable polymers for quasi-solid photoelectrochromic devices with excellent long-term stability under real outdoor operating conditions, Advanced Functional Materials 26 (2016) 1127–1137. [20] J.R. Nair, L. Porcarelli, F. Bella, C. Gerbaldi, Newly Elaborated Multipurpose Polymer Electrolyte Encompassing RTILs for Smart Energy-Efficient Devices, ACS Applied Materials and Interfaces 7 (23) (2015) 12961–12971. [21] H. Zhao, X. Ding, P. Yang, L. Li, X. Li, Y. Zhang, A novel multi-armed and star-like poly(ethylene oxide) membrane for CO2 separation, Journal of Membrane Science 489 (2015) 258–263. [22] C. Porcarelli, F. Gerbaldi, Super Soft All-Ethylene Oxide Polymer Electrolyte for Safe All-Solid Lithium Batteries, Scientific Reports 6 (2016) art. no. 19892. [23] J. Maçaira, I. Mesquita, L. Andrade, A. Mendes, Temperature Role in the Recombination Reaction on Dye-Sensitized Solar Cells, Phys. Chem. Chem. Phys. 17 (2015) 22699–22710. [24] C. Costa, C. Pinheiro, I.D.S. Henriques, C.A.T. Laia, Inkjet Printing of Sol–Gel Synthesized Hydrated Tungsten Oxide Nanoparticles for Flexible Electrochromic Devices, ACS Appl. Mater. Interfaces 4 (2012) 1330–1340.
[25] P. Tehrani, L.-O. Hennerdal, A.L. Dyer, J.R. Reynolds, M. Berggren, Improving the contrast of all-printed electrochromic polymer on paper displays, Journal of Materials Chemistry 19 (2009) 1799–1802. [26] J. Jensen, M. Hösel, A.L. Dyer, F.C. Krebs, Development and manufacture of polymer-based electrochromic devices, Advanced Functional Materials 25 (2015) 2073–2090. [27] A. Kumar, J.R. Reynolds, Soluble alkyl-substituted poly (ethylenedioxythiophenes) as electrochromic materials, Macromolecules 29 (1996) 7629–7630. [28] http://shop.solaronix.com/iodolyte-an50. html [29] L. Gomes, A. Marques, A. Branco, J. Araújo, M. Simões, S. Cardoso, F. Silva, I. Henriques, C.A.T. Laia, C. Costa, IZO deposition by RF and DC sputtering on paper and application on flexible electrochromic devices, Displays 34 (2013) 326–333. [30] C.K. Hong, H.S. Ko, E.M. Han, K.H. Park, Electrochemical Properties of Electrodeposited PEDOT Counter Electrode for Dye-sensitized Solar Cells, Int. J. Electrochem. Sci. 10 (2015) 5521–5529. [31] J.A. Mikroyannidis, A.N. Kabanakis, D.V. Tsagkournos, P. Balraju, G.D. Sharma, Bulk heterojunction solar cells based on a low band gap soluble bisazopyrrole and the corresponding BF2-azopyrrole complex, J. Mater. Chem. 20 (2010) 6464–6471. [32] J. Wang, J. Zhao, F.E. Osterloh, Photochemical charge transfer observed in nanoscale hydrogen evolving photocatalysts using surface photovoltage spectroscopy, Energy Environ. Sci. 8 (2015) 2970–2976. [33] A.L. Smeigh, J.E. Katz, B.S. Brunschwig, N.S. Lewis, J.K. McCusker, Effect of the Presence of Iodide on the Electron Injection Dynamics of Dye-Sensitized TiO2Based Solar Cells, J. Phys. Chem. C 112 (2008) 18467–18473. [34] M. Eikerling, I. Yu. Kharkats, A.A. Kornyshev, Phenomenological Theory of Electro-osmotic Effect and Water Management in Polymer Electrolyte ProtonConducting Membranes, J. Electrochem. Soc. 145 (1998) 2684–2699.