Transparent MoS2/PEDOT Composite Counter Electrodes for Bifacial Dye-Sensitized Solar Cells.

Dye-sensitized solar cells (DSSCs) are solar energy conversion devices with high efficiency and simple fabrication procedures. Developing transparent counter electrode (CE) materials for bifacial DSSCs can address the needs of window-type building-integrated photovoltaics (BIPVs). Herein, transparent organic-inorganic hybrid composite films of molybdenum disulfide and poly(3,4-ethylenedioxythiophene) (MoS2/PEDOT) are prepared to take full advantage of the conductivity and electrocatalytic ability of the two components. MoS2 is synthesized by hydrothermal method and spin-coated to form the MoS2 layer, and then PEDOT films are electrochemically polymerized on top of the MoS2 film to form the composite CEs. The DSSC with the optimized MoS2/PEDOT composite CE shows power conversion efficiency (PCE) of 7% under front illumination and 4.82% under back illumination. Compared with the DSSC made by the PEDOT CE and the Pt CE, the DSSC fabricated by the MoS2/PEDOT composite CE improves the PCE by 10.6% and 6.4% for front illumination, respectively. It proves that the transparent MoS2/PEDOT CE owes superior conductivity and catalytic properties, and it is an excellent candidate for bifacial DSSC in the application of BIPVs.

Introduction

Due to the energy crisis and environmental issues, solar cells have attracted great attention as important energy conversion devices for clean and renewable energy. Extensive investigations have been performed to explore cost-effective third-generation solar cell technologies such as polymer solar cells, perovskite solar cells, dye-sensitized solar cells (DSSCs), and other types of solar cells.[1−5] DSSCs have gained steady development since they achieved decent power conversion efficiency (PCE) of 7.9% in 1991.[6,7] A typical DSSC has a sandwiched cell structure with a dye-sensitized TiO2 photoanode, a counter electrode (CE), and redox electrolyte. CE plays an indispensable role in DSSCs operation, i.e., CE not only transports the electrons from an external circuit back to a redox system, but also is used as a catalyst to reduce redox couples of the electrolyte. Pt electrode is the most commonly used CE in DSSCs with excellent catalytic performance. However, due to the high cost of Pt and its poor corrosion resistance, the wide application of Pt CE is limited especially for large-scale production of DSSCs.[8]

Various platinum-substituted CE materials have been extensively studied, such as alloy CEs,[9−12] carbon CEs,[13−18] conductive polymer CEs,[19−22] metal oxide CEs,[23−25] composite CEs,[26−29] and so on. Among them, transparent conductive polymer CE has been widely considered as the superior candidate due to its good catalytic performance and transparency. Common conductive polymer materials mainly include poly(3,4-ethylenedioxythiophene) (PEDOT),[30,31] poly(3,4-propylenedioxythiophenes) (PProDOT),[32] polyaniline (PANI),[33] and polypyrrole (PPy).[34,35] Transparent or semitransparent electrodes can be used for building-integrated photovoltaics (BIPVs) to make use of the light from the interior of the building as well as the outside.[36−38] Conductive polymer PEDOT owns high conductivity, good transparency, and high stability, and it is also suitable for large-scale production with low cost.[39] However, the catalytic performance of PEDOT CE alone is not as good as that of the Pt CE.[40] Therefore, there is a strong need to find suitable materials in combination with the PEDOT films, improving the photovoltaic performance of DSSCs. Chen et al. developed transparent CEs of PEDOT/nitrogen-doped graphene (NGr) for bifacial DSSCs, achieving a PCE of 8.3% higher than that of the Pt CE.[40] Li et al. reported a 7.1% PCE achieved by a PEDOT/rGO composite CE in DSSCs, while the PCE of DSSCs based on the pristine PEDOT CE was only 6.4%.[41]

Herein, we combined MoS2 nanomaterials and PEDOT films in a very simple way to form transparent composite CEs for bifacial DSSCs. As a transition-metal compound, molybdenum sulfide (MoS2) has excellent conductivity and catalytic property.[42] Reports showed that MoS2 CE owes high catalytic activity to reduced I3–, and DSSCs using MoS2 CEs can reach similar PCE to that of the Pt CE.[43,44] We developed a new method to prepare transparent PEDOT/MoS2 composite CEs. Nanosized MoS2 was synthesized from a hydrothermal method, and MoS2 films were deposited onto a fluorine-doped tin oxide (FTO) substrate by spin-coating. Then, PEDOT films were electrochemically polymerized on top of the MoS2 films under potentiostatic conditions. The addition of MoS2 improves the continuity of the PEDOT film and reduces the pores of the PEDOT film, thereby improving the conductivity of the composite film. The as-fabricated bifacial DSSCs demonstrated excellent PCE of up to 7% from the front illumination. Particularly, our MoS2/PEDOT-based DSSCs showed a superior PCE value of 4.82%, illuminating from the CE side. This work has demonstrated that the DSSC with MoS2/PEDOT CE has better photovoltaic performance than that of the Pt CE. It provides a novel and efficient Pt-free CE material for bifacial DSSCs in the application of BIPVs.

Results and Discussion

PEDOT films can be obtained from a variety of routes. Solution processible aqueous solution of poly(3,4-ethylenedioxythiophene) and poly(styrene sulfonate) (PEDOT:PSS) is generally used to prepare conductive PEDOT films. Other methods including chemical polymerization and electrochemical polymerization are also commonly used to obtain PEDOT. Electrochemical polymerization to prepare PEDOT film can be classified as a potentiostatic mode, a galvanostatic method, and cyclic voltammetry (CV) method.[45−47] In the potentiostatic mode, the thickness of the electrochemically polymerized PEDOT films is controlled by polymerization voltage, polymerization time, and monomer concentration. Herein, a constant voltage was used to synthesize PEDOT film, and the schematic illustration of the bifacial DSSCs with MoS2/PEDOT CE is illustrated in Figure . The polymerization conditions were set for voltage steps of 1.1 V for 10 s and then 1.8 V for another 0.1 s. Suitable monomer concentration was optimized to ensure that the PEDOT film maintains both good light transmittance and excellent catalytic property. The thickness of the PEDOT film can be adjusted by changing the monomer concentration of EDOT.

Figure 1

Schematic illustrations of the bifacial DSSC with MoS2/PEDOT CE.

Figure a shows the transmittance of the PEDOT CEs prepared at EDOT monomer concentrations of 0.1, 0.05, 0.025, 0.01, and 0.005 M. The thickness of the PEDOT films varies with the monomer concentration. The transmittance of five PEDOT CEs all showed a rapid increase starting from 300 nm, reaching their maximum in the range of 450–600 nm, and then the transmittance gradually decreased. The transmittances of five PEDOT CEs are all more than 55% in the wavelength range of 400–800 nm. As the thickness of the CE decreases, the transmittance of the PEDOT CE increases. The highest transmittance of each PEDOT CE is 68, 74, 74, 74, and 75% for the films prepared under conditions of 0.1, 0.05, 0.025, 0.01, and 0.005, respectively, named as PEDOT (0.1 M), PEDOT (0.05 M), PEDOT (0.025 M), PEDOT (0.01 M), and PEDOT (0.005 M), respectively. Figure S1 shows the photograph of PEDOT/FTO films prepared under different monomer concentrations and the FTO substrate alone. It can be observed that blue films were formed on the FTO substrate. Taking the blank FTO glass as a reference, decreasing the EDOT monomer concentration from 0.1 to 0.005 M leads to a gradually diluted blue color, indicating the decreasing thickness of the PEDOT films.

Figure 2

(a) Transmittance of PEDOT CEs prepared with different EDOT monomer concentrations of 0.1, 0.05, 0.025, 0.01, and 0.005 M. (b) J–V curve of the DSSCs assembled by the different PEDOT CEs illuminated under 100 mW cm–2 from the photoanode (front) side.

Figure b exhibits the J–V curves of the DSSCs fabricated by the PEDOT CEs with different monomer concentrations. The photovoltaic parameters of these DSSCs are summarized in Table . It can be found that as the monomer concentration decreases, the PCE of the corresponding cells first increases and then decreases. A 6.2% PCE was achieved by the PEDOT (0.1 M) CE-based DSSCs. The DSSCs fabricated by the PEDOT (0.05 M) CE gained the highest PCE of 6.33% with a short-circuit current density (JSC) of 13.16%, open-circuit voltage (VOC) of 0.756 V, and fill factor (FF) of 63.6%. With a gradual reduction in the EDOT concentration, a lower PCE of 5.58% was obtained by the PEDOT (0.005 M) CE. The FF value of the DSSCs prepared by five different CEs showed a decreased trend with the decreasing EDOT concentration. It might be due to the decrease in the number of catalytic sites and conductivity in the thin PEDOT film. A too-thin PEDOT film would have poor uniformity, limited catalytic sites, and low conductivity. However, thick PEDOT films might have poor adhesion to the FTO substrate, and the interface transfer resistance will become large.[41] Considering the film transmittance and the overall photovoltaic performance, the optimized EDOT monomer concentration of 0.05 M was adopted for the subsequent synthesis of the PEDOT film and the MoS2/PEDOT composite film. The cross-sectional scanning electron microscopy (SEM) image of the PEDOT (0.05 M)/FTO film is shown in Figure S2, and about 250 nm thick PEDOT can be prepared via this method.

Table 1

Photovoltaic Parameters of DSSCs Fabricated by Different PEDOT CEs under Front Illumination

PEDOT CE (M)	η (%)	Voc (V)	Jsc (mA cm–2)	FF (%)
0.1	6.20	0.728	13.06	65.3
0.05	6.33	0.756	13.16	63.6
0.025	6.04	0.749	13.32	60.5
0.01	5.70	0.759	13.24	56.7
0.005	5.58	0.765	12.98	56.2

X-ray diffraction (XRD) and Raman spectroscopy were performed to confirm the composition of MoS2. The high-resolution transmission electron microscopy (HR-TEM) image of the MoS2 shows that the layer spacing of MoS2 is about 1.12 nm as shown in Figure S3. In Figure a, the 2θ diffraction peaks at 14.2, 32.9, 34.0, 38.0, and 58.1° can be assigned to the MoS2 planes (JCPDS No. 01-074-0932) of (003), (101), (012), (104), and (110), respectively. Compared to MoS2 with a layer spacing of 0.62 nm reported in the literature,[48] the diffraction peaks of MoS2 here had a broad layer spacing of 1.12 nm, showing a significant shift to the left side in the XRD pattern. The Raman spectrum in Figure b shows that the characteristic peaks of MoS2 located at 376.7 and 403.1 cm–1 correspond to the modes of E2g and A1g, respectively.[49] The XRD and Raman results indicated that the MoS2 nanomaterial was successfully synthesized. Figure c shows the Fourier transform infrared (FTIR) spectrum of PEDOT. The peaks around 1513 and 1326 cm–1 can be assigned to the saturated and unsaturated carbon stretching in the quinoid structure and the stretching in the thiophene ring. The vibration peaks at 1197, 1143, and 1084 cm–1 related to the C–O–C bond stretching in the ethylenedioxy group. The peaks around 972, 833, 779, and 685 cm–1 can be attributed to the C–S bond stretching in the thiophene ring.[50,51] Therefore, the FTIR spectrum proved that the PEDOT film was electrochemically polymerized on the FTO glass. The XRD spectrum of the MoS2/PEDOT composite film is presented in Figure d. The broad 2θ diffraction peaks at 25.5° corresponded to the (020) reflection of the polymer backbone.[52] The 2θ diffraction peaks at 32.6 and 57.6° can be assigned to the MoS2 planes. It indicates that the MoS2/PEDOT composite film was successfully prepared.

Figure 3

(a) XRD and (b) Raman spectra of the MoS2 sample. (c) FTIR spectrum of the PEDOT film. (d) XRD spectra of the MoS2/PEDOT composite film.

Morphologies of different CEs of FTO/Pt, FTO/MoS2, FTO/PEDOT, and FTO/MoS2/PEDOT were characterized by SEM, as shown in Figure . It can be observed that the Pt nanoparticles were filled in the concave area of the FTO substrate in Figure a,b. Figure b,f shows the morphology of two layers of the MoS2 film on the FTO substrate. The MoS2 film is composed of nanoparticles with nonuniform sizes. As can be seen from Figure c,g, the pristine PEDOT film has a porous structure. Chen et al. considered that the porous structure of PEDOT would have a high specific surface area to achieve high electrocatalytic capabilities.[40] However, this porous structure also indicates that the PEDOT film might lack a directional charge transfer pathway. In addition, it can be seen from Figure d,h that the morphology of the PEDOT/MoS2 film is different from that of the pure PEDOT film. The small pores and good uniformity of the MoS2/PEDOT film can be clearly observed, indicating that the MoS2/PEDOT film might have a high specific surface area and a large number of electrocatalytic active sites. The composition of the MoS2/PEDOT composite films was further surveyed by energy-dispersive spectroscopy (EDS) and elemental mapping analysis. Figure S4a–c shows the corresponding elemental mappings for sulfur, molybdenum, and carbon, respectively. It can be observed that these elements are uniformly distributed on the FTO substrate. The EDS spectrum in Figure S4d presents the coexistence of MoS2 and PEDOT in the composite films.

Figure 4

SEM images of FTO/Pt (a, e), FTO/MoS2 (2 layers) (b, f), FTO/PEDOT (c, g), and FTO/MoS2(2L)/PEDOT (d, h) at low and high magnifications.

Figure a shows the photocurrent density–voltage (J–V) characteristic curves of the DSSCs assembled by different thicknesses of the MoS2/PEDOT CEs. One to five layers of the MoS2 films were prepared via the spin-coating method. The as-prepared MoS2/PEDOT CEs were named as MoS2(1L)/PEDOT CE, MoS2(2L)/PEDOT CE, MoS2(3L)/PEDOT CE, MoS2(4L)/PEDOT CE, and MoS2(5L)/PEDOT CE, respectively. The photovoltaic parameters of these DSSCs are summarized in Table . The PCE of the as-fabricated DSSCs first increased and then decreased with the increases of the MoS2 thickness. The DSSCs fabricated by the MoS2(2L)/PEDOT composite CEs achieved the highest PCE of 7.0% compared with those fabricated under other conditions. The corresponding photovoltaic parameters of the optimized DSSCs are Voc of 0.743 V, Jsc of 13.73 mA cm–2, and FF of 0.686. Increase in the thickness of the MoS2 films leads to reduced Jsc and Voc. Thick MoS2 film would result in more active sites and more electron transfer pathways.[40] However, poor adhesion between MoS2 films and the FTO substrates causes large charge transfer resistance of the composite CEs. Therefore, as the number of MoS2 layer increases, the PCE of the DSSCs reaches the maximum and then decreases.

Figure 5

(a) J–V curves of the front illumination of the DSSCs with different types of MoS2/PEDOT CEs. (b) J–V curves of the DSSCs with Pt CE, MoS2 CE, PEDOT CE, and MoS2(2L)/PEDOT CE under front illumination. (c) J–V curves of the DSSCs with Pt CE, MoS2 CE, PEDOT CE, and MoS2(2L)/PEDOT CE under back illumination. (d) Transmittance of four different counter electrodes.

Table 2

Photovoltaic Parameters of the DSSCs with Different Types of MoS2/PEDOT CEs Tested under 100 mW cm–2 by Applying Front Illumination

CE	η (%)	Voc (V)	Jsc (mA cm–2)	FF (%)
MoS2(1L)/PEDOT	6.77	0.743	13.93	65.4
MoS2(2L)/PEDOT	7.00	0.743	13.73	68.6
MoS2(3L)/PEDOT	6.81	0.737	13.31	69.5
MoS2(4L)/PEDOT	6.73	0.736	13.30	68.7
MoS2(5L)/PEDOT	6.71	0.737	13.16	69.1

The photovoltaic performance of the DSSCs based on MoS2(2L)/PEDOT CEs is again compared with that of the CEs of Pt, pristine MoS2, and pristine PEDOT. Figure b shows the J–V curves based on the above four types of CEs under the front illumination. The related photovoltaic parameters are listed in Table . The DSSC with Pt CE showed a PCE of 6.58%, and the DSSCs using MoS2 CEs and PEDOT CEs showed PCEs of 3.81 and 6.33%, respectively. The PCE of DSSC with MoS2(2L)/PEDOT CE shows the best PCE of 7.0% among all the other types of CEs. Compared with the DSSCs made by PEDOT CE and MoS2 CE, the PCE of the DSSCs using MoS2/PEDOT CEs was increased by 10.6 and 83.7%, respectively. From Table , It can be found that the photovoltaic improvement of DSSCs made by MoS2(2L)/PEDOT composite CE is mainly due to the increase in the Jsc and FF values. It demonstrated that synergistic catalytic property can be gained by combining MoS2 and PEDOT.

Table 3

Photovoltaic Parameters of the DSSCs Made by Pt CE, MoS2 CE, PEDOT CE, and MoS2(2L)/PEDOT CE Tested under 100 mW cm–2 from Front and Back Illumination

	front illumination				back illumination
CE	η(%)	Voc (V)	Jsc (mA cm–2)	FF (%)	η (%)	Voc (V)	Jsc (mA cm–2)	FF (%)	retention (%)
Pt	6.58	0.767	14.28	60.0	4.78	0.749	8.59	74.4	72.6
MoS2(2L)	3.81	0.721	11.62	45.5	2.71	0.707	7.02	54.6	71.1
PEDOT	6.33	0.756	13.16	63.6	4.92	0.746	9.12	72.4	77.7
MoS2(2L)/PEDOT	7.00	0.743	13.73	68.6	4.82	0.730	8.75	75.5	68.9

To prove the idea of bifacial illumination, the DSSCs assembled by CEs of Pt, MoS2, PEDOT, and MoS2(2L)/PEDOT were tested under back illumination as well. Figure c shows the J–V curves of the cells illuminated from the backside (CE side). The photovoltaic parameters of the corresponding DSSCs are listed in Table . When the photovoltaic performance is measured from the backside, a certain amount of light is absorbed by CE and electrolyte, and then absorbed by dye molecules in the photoanode. Thus, the Jsc of the DSSCs drops dramatically compared to that of the same cells irradiated from the front side. Taking the DSSC made by the CEs of MoS2(2L)/PEDOT as an example, the maximum Jsc achieved by the back irradiation is only 8.75 mA/cm2 in contrast with the Jsc of 13.73 mA/cm2 under the front illumination. Besides, the DSSCs with the PEDOT CE and MoS2(2L)/PEDOT CE showed better PCE performance than that of the Pt CE. Correspondingly, the DSSC with PEDOT CE has the highest Jsc of 9.12 mA cm–2 and PCE of 4.92%, while the DSSCs with Pt CE and MoS2(2L)/PEDOT CE have close Jsc and PCE. When irradiated from the backside, the photovoltaic performance of DSSCs is not only related to the catalytic performance of the CE itself but the transmittance of CE also has an important impact on the DSSCs performance.

Figure d shows the transmittance spectra of four different CEs of Pt, MoS2, PEDOT, and MoS2(2L)/PEDOT, and they all have more than 60% transmittance in the wavelength range of 400–800 nm. Among them, the MoS2 CE has the highest light transmittance of up to 80%, showing its excellent transparency for bifacial DSSCs. However, due to the limited electrocatalytic performance of pristine MoS2, the DSSC made by MoS2 CE presented low PCEs from both front and back illuminations. The PEDOT film has the best light transmittance among the other three CEs. The MoS2(2L)/PEDOT CE showed similar light transmittance to the Pt film. The transmittance results confirmed that the MoS2(2L)/PEDOT CE is suitable CEs for bifacial DSSCs.

A parameter of retention was calculated and is listed in Table . Retention is the ratio of the cell PCE under back illumination to the PCE of the same cell under front illumination. As can be seen from Table , the retention value of the DSSCs with PEDOT CE is up to 77.7%. After incorporating MoS2 into the PEDOT films, the retention value of the DSSCs can still remain 68.9%. This value is much higher than the reported value of 28.3% for the bifacial DSSCs using Si3N4/MoS2-PEDOT:PSS CE.[53] The as-prepared MoS2/PEDOT CE shows good property to meet the requirements of building-integrated photovoltaics (BIPVs).

A comparison of the reported DSSCs fabricated by different CEs for bifacial illumination under standard conditions is summarized in Table S1. By comparison, it can be seen that the as-synthesized MoS2(2L)/PEDOT CE and PEDOT CE in this work have both high PCEs and light transmittance. For the first time, we combined the MoS2 with PEDOT by electrochemical synthesis to form a transparent counter electrode, which are excellent candidates for BIPVs.

Cyclic voltammetry is generally used to evaluate the electrocatalytic activity of the counter electrode in DSSCs.[54] Herein, the electrocatalytic capabilities of various CEs of Pt, MoS2, PEDOT, and MoS2(2L)/PEDOT for triiodide ions reduction were assessed by cyclic voltammetry, as shown in Figure a. Two pairs of typical oxidation–reduction current density peaks can be observed. The oxidation–reduction peaks labeled as Ox1 and Red1 on the low potential side (left side) is attributed to the chemical reaction shown in eq . The oxidation–reduction peaks on the high potential side marked as Ox2 and Red2 (right side) is attributed to the chemical reaction shown in eq .[55]Generally, the oxidation–reduction peak on the low potential side has a crucial influence on the photovoltaic performance of the DSSCs. Therefore, the peak spacing (ΔEp) and cathode peak current density (JOx1, JRed1) on the low potential side were carefully examined to estimate the catalytic activity of the CEs. The main parameters are summarized in Table . The kinetic redox ability of the CEs to the I–/I3– redox couple is associated with the magnitude of the ΔEp value. A small ΔEp value reveals the superior electrocatalytic ability of the corresponding CEs.[40] The composite CE of MoS2(2L)/PEDOT showed a lower ΔEp value (0.318 mV) than that of the MoS2 (0.686 mV) and the PEDOT (0.380 mV), indicating an improved catalytic capability by combining the MoS2 and PEDOT. In addition, the cathode peak current density (JOx1, JRed1) also represents the reduction ability of CEs to I3–. The excellent electrocatalytic capability of the corresponding CE is attributed to the higher absolute values of JOx1 and JRed1. In Table , the absolute values of JOx1 and JRed1 of MoS2(2L)/PEDOT CE (JOx1 = 1.851 mA cm–2, JRed1 = −1.726 mA cm–2) are significantly improved compared to those of the pristine PEDOT CE (JOx1 = 1.586 mA cm–2, JRed1 = −1.517 mA cm–2) and the pristine MoS2 CE (JOx1 = 0.989 mA cm–2, JRed1 = −0.935 mA cm–2). The result suggests that the MoS2(2L)/PEDOT CE has a stronger reduction ability to I3– than the MoS2 CE and the PEDOT CE, which is consistent with the result of the ΔEp value.

Figure 6

(a) Cyclic voltammograms of Pt CE, MoS2 CE, PEDOT CE, and MoS2(2L)/PEDOT CE tested at a scan rate of 50 mV s–1. (b) Tafel polarization curves of the Pt CE, MoS2 CE, PEDOT CE, and MoS2(2L)/PEDOT CE. (c) Nyquist plots with the symmetrical cell made by Pt CE, MoS2 CE, PEDOT CE, and MoS2(2L)/PEDOT CE for full view. (d) Enlarged Nyquist plots inserted with an equivalent circuit.

Table 4

Electrochemical Parameters of CEs of Pt, MoS2, PEDOT, and MoS2(2L)/PEDOTa

CE	ΔEp (mV)	JOx1 (mA cm–2)	JRed1 (mA cm–2)	RS (Ω cm2)	Rct-EIS (Ω cm2)	J0 (mA cm–2)	Rct-Tafel (Ω cm2)
Pt	0.331	1.601	–1.381	7.27	3.94	3.39	3.79
MoS2	0.686	0.989	–0.935	7.66	103.5	0.11	116.82
PEDOT	0.380	1.586	–1.517	7.20	6.03	2.31	5.56
MoS2(2L)/PEDOT	0.318	1.851	–1.726	5.26	5.55	2.54	5.06

The working areas of the measured sample are 1 cm2 for CV, electrochemical impedance spectroscopy (EIS), and Tafel measurements.

Figure b shows Tafel polarization plots with Pt CE, MoS2 CE, PEDOT CE, and MoS2(2L)/PEDOT CE. Tafel polarization is also an important method for evaluating the catalytic performance of the electrode. Generally, the Tafel polarization curve can be divided into three regions according to the potential difference: the polarization region (overpotential |V| < 120 mV); the Tafel region (120 mV < |V| < 400 mV); and the diffusion region (|V| > 400 mv).[40] To estimate the catalytic ability of the DSSC CE, the reduction ability of the CEs to I3– was mainly investigated. The ability of the CEs to reduce iodide ions can be evaluated by the exchange current density (J0), which is closely connected to the catalytic activity of the CE. Generally, the Tafel linear extrapolation method can indirectly obtain the J0 value. A larger J0 value indicates that CE has a better catalytic activity for the reduction of I3–. It can be seen from Table that the J0 value of the MoS2(2L)/PEDOT CE (J0 = 2.54 mA cm–2) is larger than that of the PEDOT CE (J0 = 2.31 mA cm–2). It proves that the MoS2(2L)/PEDOT CE has a superior capability for the reduction of I3– than the PEDOT CE, which is consistent with the results of the JOx1 and JRed1 values studied above.

Figure c shows the Nyquist plots of symmetric cells with the Pt CE, MoS2 CE, PEDOT CE, and MoS2(2L)/PEDOT CE. Enlarged Nyquist plots of four CEs in the high-frequency range and the inserted equivalent circuit are shown in Figure d. Rs represents series resistance originating from the substrates and CE films. Its value can be derived from the starting point of the first semicircle in the horizontal axis. A lower Rs indicates a higher conductivity of the catalytic CEs, and the as-fabricated DSSC would have a better FF value.[40] The Rs value can be adjusted by changing the CE thickness. It can be seen from Table that MoS2(2L)/PEDOT CE has the smallest Rs value of 5.25 Ω cm2, proving its superior conductivity. Meanwhile, Rct shows the charge transfer resistance between the electrode and the electrolyte interface. A smaller Rct indicates that the CE has a better electrocatalytic activity for the reduction of I3–. The MoS2(2L)/PEDOT CE has a lower Rct value of 5.55 Ω cm2 compared with that of MoS2 CE of 103.5 Ω cm2 and PEDOT CE of 6.03 Ω cm2. It revealed that the MoS2/PEDOT CEs have better catalytic performances. The results from the EIS are consistent with the results obtained from the CV tests.

In addition, in the polarization region of the Tafel polarization curve, the slope of the linear portion of the V–J curve measured near the equilibrium potential is the polarization resistance. In the case of DSSCs, it is equivalent to the Rct value of the electrode/electrolyte interface, which can be obtained according to the formula (40)Here R is the general gas molar constant; T is the thermodynamic temperature; F is the Faraday constant; and n is the number of electrons transferred per chemical reaction between the electrolyte and the counter electrode interface, n = 2.

As can be seen from Table , the Rct value calculated by J0 from the Tafel test is close to the Rct value obtained from the EIS measurement. The values of ΔEp, JOx1, and JRed1 obtained by CV measurements, the values of Rs and Rct-EIS estimated by the EIS, and the values of J0 and Rct-Tafel gained by the Tafel polarization curve suggest that all obtained electrochemical parameters of the CEs have the consistent trend as that of the photovoltaic parameters of the respective DSSCs.

Conclusions

Composite materials of MoS2/PEDOT were successfully prepared and investigated as the counter electrode (CE) for bifacial DSSCs. The PEDOT films were electrochemically polymerized under potentiostatic conditions. Nanomaterials MoS2 were synthesized by a hydrothermal method. Composite CEs of MoS2/PEDOT were adjusted via optimizing the synthesis condition of PEDOT and the thickness of MoS2 films to ensure high PCE of the corresponding DSSCs. The DSSCs based on composite CEs of MoS2 (two layers) combined with the PEDOT (0.05 M) showed the best photovoltaic performance, achieving the maximum PCE of 7% from the front illumination and 4.82% from the back illumination. The measurements of cyclic voltammetry, electrochemical impedance spectroscopy, and Tafel polarization test show that the MoS2(2L)/PEDOT composite CEs exhibit excellent conductivity and superior catalytic properties. Moreover, the DSSCs made by MoS2(2L)/PEDOT composite CEs gained high retention value over 68%. This work demonstrated a new strategy to prepare transparent composite CEs of MoS2/PEDOT for efficient bifacial DSSCs in BIPVs.

Experimental Section

Materials

All reagents were purchased from commercially available sources and used without further purification. Sodium molybdate (Na2MoO4, 99%), thiourea (CH4N2S, 99%), and glucose (C6H12O6, 99%) were purchased from Sinopharm Chemical Reagent Co., Ltd. 3,4-Ethylenedioxythiophene (EDOT, 99%), lithium perchlorate trihydrate (LiClO4·3H2O, 99%), acetonitrile (AN, 99.5%), and valeronitrile (VN, 99%) were purchased from Aladdin. Titanium tetrachloride (TiCl4, 99.9%), chloroplatinic acid (H2PtCl6·6H20, 99.95%), and 1-butyl-3-methylimidazolium iodide (BMII, 99%) were purchased from Macklin. Lithium iodide (LiI, 99.9%), iodine (I2, 99.5%), guanidinium thiocyanate (GuSCN, 97%), and 4-tert-butylpyridine (TBP, 96%) were purchased from Sigma-Aldrich. Titanium dioxide paste (TiO2) and cis-diisothiocyanato-bis(2,20-bipyridyl-4,40-dicarboxylato) ruthenium(II) bis(tetrabutylam-monium) (N-719 dye) were purchased from Dyesol. Fluorine-doped tin oxide (FTO) coated conducting glass plates were purchased from Opvtech.

Synthesis of MoS2

Ten milliliters of Na2MoO4 (0.03 M) in deionized water, 20 mL of thiourea (0.12 M) in deionized water, and 0.1 g of glucose were added into an autoclave. The autoclave was heated at 200 °C for 14 h, and then cooled to room temperature. After the reaction, black solid was precipitated and collected by centrifuging at 9000 rpm. Then, the black precipitate was washed by deionized water. The cleaning–centrifugation process was repeated three times to obtain the final MoS2 with a layer spacing of 1.12 nm. The as-obtained MoS2 powder was dispersed in deionized water with a concentration of 15 mg/mL for further use.

Preparation of Various Counter Electrodes (CEs)

The FTO glass was cleaned by sonicating in a bath of detergent, deionized water, acetone, and ethanol for 15 min in each step. Constant voltage was applied to synthesize PEDOT films in electrochemical station. In detail, a three-electrode setup consisted of a Pt sheet as the counter electrode, FTO glass as the working electrode, and Ag/AgCl as the reference electrode. The electrolyte was prepared by dissolving 0.05 M EDOT monomer and 0.1 M LiClO4·3H2O in acetonitrile. A constant voltage of 1.1 V was set to trigger the EDOT monomer for 10 s and then further polymerized at a higher voltage of 1.8 V for 0.1 s. Then, a uniform and transparent PEDOT film was evenly coated on the FTO substrate. For the preparation of MoS2 CE, a suspension of MoS2 (5 mg/mL) in water/isopropanol mixture (volume ratio: 1:2) was prepared, and then layers of MoS2 were spun-coated onto the clean FTO glass at the spinning speed of 2000 rpm for 20 s. The as-formed MoS2 film was dried up at 100 °C for 10 min. To prepare the composite MoS2/PEDOT CE, the FTO glass coated with MoS2 film was used as the working electrode, and the remained polymerization method to form PEDOT was kept similarly as described above. The as-obtained both PEDOT CE and MoS2/PEDOT CE were washed with acetonitrile and dried at 40 °C. As for the Pt CE, 20 μL of H2PtCl6 in isopropanol solution (10 mM) was drop-casted on the cleaned FTO glass and then the Pt CE was sintered at 400 °C for 15 min.

Fabrication of DSSC

A compact TiO2 layer was prepared by treating the cleaned FTO substrate with TiCl4 (40 mM) aqueous solution at 70 °C for 30 min and then sintering at 500 °C for 30 min. The TiO2 paste was subsequently screen-printed onto the above TiO2 films and sintered at 500 °C for 30 min. The N719 dye (0.5 mM) was dissolved in a mixture of absolute ethanol and dimethyl sulfoxide (DMSO) (volume ratio of 9:1). After the TiO2 photoanode was cooled to 80 °C, it was immediately immersed in the N719 dye solution for dye soaking for 24 h, and then the dye-sensitized TiO2 photoanode was rinsed by ethanol and air-dried. After attaching the dye-sensitized photoanode and the CEs, the I–/I3– redox electrolyte was injected into the sandwiched cell to complete the cell assembly. The electrolyte is composed of 0.1 M LiI, 0.03 M I2, 0.5 M 4-tert-butylpyridine (TBP), 0.1 M guanidinium thiocyanate (GuSCN), and 0.6 M 1-butyl-3-methylimidazolium iodide (BMII) dissolved in the mixed solvents of acetonitrile and valeronitrile (volume ratio of 0.85:0.15).

Characterization

The transmittance of the different CEs was measured by the ultraviolet–visible (UV–vis) absorption spectroscopy (Shimadzu UV-3600 spectrometer) in the wavelength range of 200–800 nm. The crystallographic structures of the samples were confirmed by a powder X-ray diffractometer (Rigaku D/Max 2500), using Cu Kα as the radiation source (λ = 1.5418 Å) in the range of 5–70°. The morphology was characterized by field-emission scanning electron microscopy (FE-SEM, NanoSEM 450, FEI) and transmission electron microscopy (TEM, Talos F200X, FEI). The Raman spectra were collected from a Raman spectrometer (Renishaw Invia RM200). The electrocatalytic activity of the CEs was measured by cyclic voltammetry (CV) using an electrochemical workstation (Princeton 4000). In the three-electrode system, an FTO/Pt, FTO/PEDOT, or FTO/MoS2/PEDOT CE was used as the working electrode (WE), a Pt sheet as the counter electrode (CE), and an Ag/AgCl electrode as the reference electrode (RE). The scan rate of the CV measurement was set to 50 mV s–1. The electrolyte contained 10.0 mM LiI, 1.0 mM I2, 0.1 M LiClO4·3H2O, and acetonitrile. Electrochemical impedance spectroscopy (EIS) measurement was performed by testing the symmetrical counter electrodes using an impedance analyzer from the electrochemical workstation (Princeton 4000) at room temperature. The EIS test frequency ranges is 1 MHz to 0.1 Hz with an amplitude of 10 mV. Tafel polarization curves were obtained from the symmetrical counter electrodes with a scan range from −0.6 to +0.6 V at a step of 10 mV. The photovoltaic performance of the as-prepared DSSCs was determined by measuring the photocurrent density–voltage (J–V) curves using a Keithley 2401 source meter under a solar simulator (Nowdata SXDN-150E) at 100 mW cm–2 intensity. The light intensity was calibrated by a standard silicon solar cell.