- Email: [email protected]
sulfide porous nanocubes as high-performance electrocatalysts for efficient dye-sensitized solar cells
Journal of Power Sources 369 (2017) 35e41
Contents lists available at ScienceDirect
Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour
Cobalt iron selenide/sulfide porous nanocubes as high-performance electrocatalysts for efficient dye-sensitized solar cells Yiqing Jiang, Xing Qian*, Yudi Niu, Li Shao, Changli Zhu, Linxi Hou** College of Chemical Engineering, Fuzhou University, Xueyuan Road No. 2, Fuzhou 350116, China
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
Novel Co-Fe-Se/S porous nanocubes (PNCs) were prepared by a template method. The PNCs were applied as efficient CE electrocatalysts for DSSCs. The PNCs had large surface area, high catalytic activity and good conductivity. The PCEs (9.58% and 9.06%) of the PNCs were higher than that of Pt (8.16%).
a r t i c l e i n f o
a b s t r a c t
Article history: Received 26 July 2017 Received in revised form 27 August 2017 Accepted 27 September 2017
A novel series of ternary compounds, namely cobalt iron selenide/sulfide nanocubes, are successfully synthesized as counter electrode (CE) materials for dye-sensitized solar cells (DSSCs), which deliver excellent performances. Homogeneous cobalt iron Prussian-blue-analog (PBA) nanocubes are prepared as the templates and are subsequently dealt with selenation/sulfidation processes via hydrothermal methods. Owing to their unique morphology, porous structure, high surface area, small charge transfer resistance and high diffusion coefficient, the Co-Fe-Se/S nanocubes possess high catalytic activity and excellent conductivity, which are tested and verified by electrochemical measurements. Meanwhile, cobalt iron selenide/sulfide nanocubes CEs achieve high efficiencies of 9.58% and 9.06%, respectively, which are both higher than that of Pt CE (8.16%). All these prominent merits make them outstanding and promising participants among Pt-free CE materials of DSSCs with lower production costs and higher power conversion efficiency. © 2017 Elsevier B.V. All rights reserved.
Keywords: Cobalt iron selenides Cobalt iron sulfides Porous nanocubes Photovoltaic performances Counter electrodes Dye-sensitized solar cells
1. Introduction Dye-sensitized solar cells (DSSCs) have been regarded as a kind of promising power conversion devices and have attracted great interest because of their characteristics, such as ease of fabrication, low expenses and high power conversion efficiency (PCE) [1,2]. A
* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (X. Qian), [email protected] (L. Hou). https://doi.org/10.1016/j.jpowsour.2017.09.080 0378-7753/© 2017 Elsevier B.V. All rights reserved.
typical DSSC device is composed of a counter electrode (CE), a dyeloaded TiO2 electrode and triiodide/iodide redox electrolyte. Each part has been studied to pursue better performance [3e11]. For example, Li et al. synthesized DTP-based dyes, which achieved a power conversion efficiency of 8.14% [6]. Among them, a counter electrode (CE) is viewed as one of the dominant parts of the classical DSSCs, transferring electron from outside circuit to redox electrolyte and reducing I 3 to I . Conventionally, platinum (Pt) was widely applied as common commercial CE material in DSSCs, on account of its favorable and famous conductivity and catalytic
36
Y. Jiang et al. / Journal of Power Sources 369 (2017) 35e41
activity [12]. However, the abundant commercial application and future development were limited and hindered by its obvious defects, such as low earth-abundance ratio, high price and ease of corrosion [13e15]. In this urgent situation, the further studies on alternatives of Pt are desperately desirable. To date, in order to seek ideal substitutes, various materials, including transition metal compounds (TMCs) [16e19], carbonaceous materials [20e22], conductive polymers [23,24] and their hybrids [25e27] have been researched. Among them, TMCs stand out because of their low costs, Pt-like catalytic activity and excellent conductivity [1]. Carbide [17], nitrides [16], oxides [19], phosphides [18], sulfides, selenides and their hybrids have been frequently investigated. It is the sulfides and selenides that win most of the attention for not only ease of synthesis via hydrothermal method but also favorable PCEs [14,28e30]. Various alloys, such as Co3S4 [31], NiS2 [26], NiS [32], FeS2 [33,34], CoS2 [35], CoS [36], MoS2 [37e40], NiSe2 [25], CoSe2 [37e39], FeSe2 [41] and so on, have been able to be approached and explored. For example, Cui et al. synthesized CoS2 embedded carbon nanocages with a high PCE of 8.20% (Pt, 7.88%) [42]. Except the chemical composition, the morphology mainly influences the performance of CE materials in DSSCs. Flower-like [43], nanofilm [25,44e46], nanotube [47], nanocages [42] and other morphologies have been designed and investigated. Thus, the roles that the chemical composition and morphology played could not be ignored in the design of catalyst. In this work, we chose cobalt iron Prussian-blue-analog (PBA) nanocubes as the precursors, which were hardly out of shape at high temperature but easy to convert, along with relatively high surface area [48,49]. Furthermore, the metal sulfides and selenides had great performance as CE materials in DSSCs with small charge resistance and high diffusion coefficient. For all the above reasons, we designed the synthetic routes that PBA nanocubes were prepared as the templates and subsequently treated with selenation/ sulfidation processes via hydrothermal methods. Benefiting from the unique morphology, structure and composition, the devices with Co-Fe selenide and sulfide porous nanocubes delivered PCEs of 9.58% and 9.06%, respectively, which are superior to that of Pt (8.16%). Hence, the results demonstrated the excellent electrocatalytic activity and remarkable conductivity of Co-Fe selenide and sulfide porous nanocubes.
2. Experimental section 2.1. Materials Potassium hexacyanoferrate (Ⅲ) (K3[Fe(CN)6], Sigma-Aldrich Corporation, 99.0%), cobalt nitrate hexahydrate (Co(NO3)2$6H2O, Aladdin Corporation, 99.0%), sodium citrate nonahydrate (C6H5Na3O7$9H2O, Aladdin Corporation, 99.0%), selenium powder (Se, Sigma-Aldrich Corporation, 99.5%), absolute ethanol (C2H5OH, Sigma-Aldrich Corporation, 99.7%), ethylenediamine (C2H8N2, Aladdin Corporation, 99.6%), sodium sulphide (Na2S, Aladdin Corporation, AR), 4-tert-butyl pyridine (C9H13N, Aladdin Corporation, 98.0%), titanium tetrachloride (TiCl4, Aladdin Corporation, AR), lithium perchlorate (LiClO4, Aladdin Corporation, 99.9%), iodine (I2, Aladdin Corporation, 99.8%), lithium iodide (LiI, Aladdin Corporation, 99.0%), acetonitrile (CH3CN, Sinopharm Chemical Reagent, 99.0%), the N719 dye (Solaronix Corporation, Switzerland, AR) and 1,2-dimethyl-3-propylimidazolium iodide (DMPII, Aladdin Corporation, 99.0%) were directly used without any further purification. The FTO glasses were obtained from Nippon Sheet Glass (Japan) with the sheet resistance of 15 U sq1.
2.2. Synthesis 2.2.1. Synthesis of Co-Fe prussian-blue-analog (PBA) cubes In a typical synthesis, 2 mmol of potassium hexacyanoferrate (Ⅲ) was dissolved in 100 mL of deionized (DI) water to form a uniform clean solution A. 3 mmol of cobalt nitrate and 4.5 mmol sodium citrate were dissolved in 100 mL of DI water to form an orange solution B. Then, the solution A was poured into solution B under vigorously stirring for 10 min. The obtained solution was aged for 2 days at ambient temperature. The black purple Co-Fe PBA nanocubes were precipitated by centrifugation and washed with water and ethanol for several times, and dried at 40 C for 12 h under vacuum. 2.2.2. Synthesis of Co-Fe-Se nanocubes 80 mg of Se powder was added into an ethanol solution (20 mL) containing 6 mL of ethylenediamine and the mixture was stirred for 0.5 h. Then, 40 mg of the as-prepared Co-Fe PBA cubes was dispersed in absolute ethanol (20 mL) under ultrasonication for 10 min to form a uniform suspension. The suspension was poured into the above solution. The mixture was transformed into a 50 mL teflon-lined stainless-steel autoclave and heated at 160 C for 6 h. The black sample was dried at 50 C for 12 h under vacuum after being centrifugated and washed with ethanol. Finally, in order to increase its crystallinity, the production was annealed at 350 C for 2 h at the heating rate of 1 C min1 under Ar atmosphere. The production was named as Co-Fe-Se. In comparison, the as-prepared Co-Fe PBA cubes were directly annealed as the same as the sample Co-Fe-Se and the sample named Co-Fe was obtained. 2.2.3. Synthesis of Co-Fe-S nanocubes 40 mg of the as-prepared Co-Fe PBA cubes and 10 mL of Na2S aqueous solution (4 mg mL1) were added into an ethanol solution (30 mL), which were dispersed under ultrasonication. The next experimental processes of this black production, named as Co-Fe-S, were the same as that of the sample Co-Fe-Se. 2.3. Fabrications of counter electrodes FTO glasses were washed with detergent, deionized water, acetone and ethanol in sequence. Typically, 100 mg of the obtained samples were dispersed in 10 mL of ethyl ethanol. After 30 min of sonication and stirring, a uniform ink suspension had been acquired before spinning coating on a piece of already pretreated FTO glass by spin-casting technique at a rotating speed of 500 rpm for 16 s and the covered FTO glass was heated and dried at 120 C for 5 min. This process was repeated for several times until the coating thickness was suitable. Approximately, the loading capacity of each FTO glass is 0.67 mg cm2. As a reference, the Pt CE was obtained by thermally depositing a platinum layer (0.02 M hexachloroplatinic acid in isopropanol) on the surface of FTO at 450 C for 30 min. 2.4. Fabrications of DSSCs The screen-printing method was utilized to cover the transparent nanocrystalline TiO2 layer (thickness, ~12 mm) and the scattering TiO2 layer (thickness, ~4 mm) on the conductive side of the pretreated FTO glass with 20 and 200 nm TiO2 sols, respectively. After being heated in air at 500 C for 1 h, the FTO glass was immersed in 0.04 M titanium tetrachloride aqueous solution at 70 C for 1 h and following sintered in air at 500 C for 1 h again. Then, the cooled photoanodes were soaked and sensitized in 0.3 mM N719 dye ethanol solution overnight in a dry and dark environment. Finally, to assemble a sandwich structure of DSSCs, the I/I 3 based electrolyte was injected into the vacuum space
Y. Jiang et al. / Journal of Power Sources 369 (2017) 35e41
between a photoanode and a CE. The electrolyte consisted of 0.3 M DMPII, 0.05 M I2, 0.1 M LiI and 0.5 M 4-tert-butylpyridine in acetonitrile solution. The active area of the cell was 0.16 cm2.
2.5. Materials characterization The morphology and nanostructure of the products were characterized by a field-mission scanning electron microscopy (SEM, S4800, Hitachi) at 5 kV and 7 mA equipped with energy-dispersive Xray spectroscopy (EDS). The crystal structures of those nanocubes were identified by X-ray diffraction (XRD, X0 Pert PRO, Cu Ka, l ¼ 0.15406 nm) at 40 kV and 40 mA in the range of 10e70 within 10 min. The Raman spectrums were conducted on a microscopic confocal Raman spectrometer (Raman, in Via Reflex, UK). Special surface area and pore size distribution were measured by the Brunauer-Emmett-Teller (BET) analyzer (Micromeritics, ASAP 2020 M, USA). The chemical properties were researched by utilizing X-ray photoelectron spectrometer (XPS, ESCALAB 250, Mg Ka, USA). All the electrochemical measurements were conducted on the CHI660E electrochemical work station (CH Instruments). The cyclic voltammetry (CV) was employed in a three-electrode system. The as-prepared CE, platinum sheet and Ag/AgCl electrode were served as the working electrode, the counter electrode and the reference electrode in the system, respectively. All electrodes were inserted into an electrolyte containing 0.1 M LiClO4, 10 mM LiI and 1 mM I2 in acetonitrile solution. The scan rate was 50 mV s1 and the scan potential range was set from 0.5 to 0.9 V. For the electrochemical impedance spectroscopy (EIS) and Tafel polarization curves, a traditional symmetrical dummy cell configuration (CE/electrolyte/ CE) was fabricated. EIS was measured from 100 kHz to 100 mHz with an amplitude of 5 mV, while Tafel curves were operated at a scan rate of 10 mV s1 between 1.0 V and 1.0 V. To obtain the PCEs of DSSCs, the photocurrent density-voltage (J-V) curves were collected via putting the solar cells under AM 1.5 G illumination (100 mW cm2) irradiation provided by a standard solar simulator (XM-500W, Trusttech) which was calibrated with a standard silicon solar cell (91150V, Newport Corporation).
37
3. Results and discussion Fig. 1ac show the typical SEM images of the as-prepared uniform solid Co-Fe PBA nanocubes, which have smooth surfaces, sharp edges and corners with an average size of around 150 nm. The average size did not change distinctly during the follow processes. After hydrothermal treatment with Na2S solution and anneal, the solid precursors were etched and got round edges and corners as well as sunk surfaces with tiny particles (Fig. 1gei). As shown in Fig. 1def, after selenation, the smooth surface of precursors got much rougher and the particles on the surface of selenide were bigger than that of sulfide. Fig. S1 (Supporting information, SI) shows the SEM images of the Co-Fe sample. Commonly, a sample with good conductivity owned bright and clear SEM images. Therefore, the conductivity of precursors after sulfidation and selenation processes were enhanced, which could be proven intuitively by the SEM images. Fig. 2 and Fig. S2 (SI) exhibit the composition and phase structures of Co-Fe PBA, Co-Fe-Se, Co-Fe-S and Co-Fe nanocubes. It is obvious that diffraction peaks of the PBA are excellently assigned to the (ICDD PDF No. 01-075-0038), which indicates the pure quality of synthesized Co-Fe PBA nanocubes [50]. As for Co-Fe-Se nanocubes, the peaks at 23.8 , 29.1, 30.9 , 34.6 , 36.0 , 37.3 , 40.4 , 44.2 , 48.0 , 50.7, 53.9 , 55.7, 57.3 , 63.5 , 63.9 and 64.0 correspond to the (1 1 0), (0 1 1), (0 2 0), (1 0 1), (1 1 1), (1 2 0), (2 0 0), (2 1 0), (1 2 1), (1 1 1), (0 0 2), (0 3 1), (2 2 1), (1 3 1), (1 2 2) and (0 4 0) crystal planes of orthorhombic FeSe2 (ICDD PDF No. 00-012-0291) [51]. The peaks at 30.8 , 34.8 , 36.3 , 47.8 and 57.2 correspond to the (1 1 0), (1 1 1), (0 1 2), (1 2 1) and (2 1 1) crystal planes of orthorhombic CoSe2 (ICDD PDF No. 00-010-0408) [52]. Among those displayed peaks, the weak peaks at 29.9 and 54.5 show the existence of the O2, which is the reason for peak shifting. All the diffraction peaks of Co-Fe-S match well with cubic Co3S4 (ICDD PDF No. 01-073-1703) and hexagonal FeCoS2 (ICDD PDF No. 01-0750607). Several peaks appear at 30.7, 35.2 , 46.4 and 54.7 correspond to the (1 0 0), (1 0 1), (1 0 2) and (1 1 0) planes of hexagonal FeCoS2, which conforms with the earlier report [53]. Besides, the peaks at 16.3 , 26.7, 31.7, 38.3 , 47.4 , 50.6 and 55.4 correspond
Fig. 1. SEM images of (aec) Co-Fe PBA nanocubes, (def) Co-Fe-Se nanocubes, (gei) Co-Fe-S nanocubes.
38
Y. Jiang et al. / Journal of Power Sources 369 (2017) 35e41
Fig. 2. XRD patterns of Co-Fe-Se, Co-Fe-S and Co-Fe PBA.
to the (1 1 1), (2 2 0), (3 1 1), (4 0 0), (5 1 1) and (4 4 0) planes of cubic Co3S4 [54]. To further confirm the crystal structure, the Raman spectra are employed and the results are shown in Fig. S3 (SI). It can be seen in Fig. S3a (SI), the peaks at 316, 335, 385 and 410 cm1 determine the existence of hexagonal FeCoS2, while peaks at 472, 513, 609 and 679 cm1 determine the existence of cubic Co3S4 [55]. In addition, the Raman spectrum of Co-Fe-Se is composed of character peaks of orthorhombic CoSe2 (467, 516 and 663 cm1) and orthorhombic FeSe2 (180, 217 and 254 cm1) [51,56]. XPS was employed to determine the surface chemical compositions and the valence states of the samples Co-Fe-Se/S. The survey spectrum (Fig. 3a) presents the attendance of Co, Fe, Se, C and O elements in Co-Fe-Se. The results of Co (Figs. 3b and 4b), Fe (Figs. 3c and 4c), Se (Fig. 3d) and S (Fig. 4d) elements were analyzed via the Gaussian fitting method. Fig. 3b are constituted with two spin-orbit doublets and a pair of shakeup satellites (marked as “Sat.”). The peaks of Co 2p1/2 at 793.6 and 797.1 eV and the peaks of Co 2p3/2 at 778.6 and 780.8 eV are characteristic of Co3þ and Co2þ. The spectrum of Fe 2p in Fig. 3c can be fitted into a doublets and a satellite. The existence of Se2 is proved by the character speaks appearing at 55.6 and 54.7 eV in Fig. 3d and the broad satellite at 59.6 eV is for Se species with high oxide state [59]. There are Co2þ, Co3þ, Fe2þ and Se2 ions on the surface composition of Co-Fe-Se. Fig. 4a determines the presence of Co, Fe, S, C and O elements in Co-Fe-S. The corresponding peak positions in Fig. 4b are as the same as that of Fig. 3b, suggesting the attendance of Co3þ and Co2þ.
Fig. 3. XPS spectra of Co-Fe-Se, (a) survey spectrum, (b) Co 2p, (c) Fe 2p, (d) Se 3d.
Fig. 4. XPS spectra of Co-Fe-S, (a) survey spectrum, (b) Co 2p, (c) Fe 2p, (d) S 2p.
In Fig. 4c, besides the peaks at 708.5 and 722.2 eV reflect the Fe 2p3/ and Fe 2p1/2 of Fe2þ (typical Fe-S bond), the other peaks at 710.5 and 724.1 eV link with the Fe 2p3/2 and Fe 2p1/2 of Fe3þ. Furthermore, the S 2p1/2 and S 2p3/2 peaks locate at 163.5 and 162.3 eV in Fig. 4d, which is the evidence of the presence of S2 in Co-Fe-S. The peaks at 707.5 and 719.0 eV reflect the Fe 2p3/2 and Fe 2p1/2 of Fe2þ, which suggests the existence of typical Fe-Se bond in Co-Fe-Se. According to previous researches, Co2þ and Fe2þ can be easily oxidized into Co3þ and Fe3þ in Co/Fe-based alloy [57,58]. One thing can be noted that there are Co2þ, Co3þ, Fe2þ, Fe3þ and S2 ions on the surface composition of Co-Fe-S. The porosity and surface area of Co-Fe-S and Co-Fe-Se were studied by N2-sorption measurements. As shown in Fig. 5a, the
2
Fig. 5. Pore-structure analysis of Co-Fe-Se and Co-Fe-S, (a) N2-adsorption-desorptipn isotherm at 77 K, (b) pore-size distribution.
Y. Jiang et al. / Journal of Power Sources 369 (2017) 35e41
plots of Co-Fe-S and Co-Fe-Se display the properties of the typical type IV behavior (characteristic of mesoporous materials), indicating them contains mesoporous. Compared with the Co-Fe-S (33.7 m2 g1), the Co-Fe-Se (52.4 m2 g1) shows a distinct increase in Brunauer-Emmett-Teller (BET) surface area. Besides, the total pore volume of Co-Fe-S and Co-Fe-Se are 0.226 cm3 g1 and 0.213 cm3 g1. As the pore-size distribution shows (Fig. 5b), the samples Co-Fe-S and Co-Fe-Se armed with the main mesopores with the diameters of 3.54 and 3.94 nm, respectively. The high surface area and pore volume contribute to electrolyte penetration and provide more active sites, which is expected to improve the catalytic activity and delivers better performance in DSSCs [48,49,53,60]. The Co-Fe PBA nanocubes (7.90 m2 g1, 0.0607 cm3 g1) were tested to prove that the sulfation and selenation processes indeed increased the surface area and formed pores. Fig. 6a reveals the photocurrent density-voltage (J-V) curves of the DSSCs fabricated with the Co-Fe-Se, Co-Fe-S, Co-Fe and Ptbased CEs, and the summary of key photovoltaic parameters of these DSSCs are listed in Table 1. The Co-Fe PBA and Co-Fe worked
39
as comparisons to verify a result that sulfidation and selenation processes do enhance PCE. The fill factor (FF) and PCE of DSSCs fabricated with samples are calculated as follows. FF ¼ JmVm/JscVoc
(1)
PCE ¼ JmVm/Pin
(2)
Where Jm is the maximum current density, Vm is the maximum voltage, Jsc is the short circuit current density, Voc is the open circuit voltage and Pin is the power input (100 mW cm2) [61]. The PCEs of Co-Fe-Se, Co-Fe-S, Co-Fe, Pt and Co-Fe PBA based CEs are 9.58%, 9.06%, 2.89%, 8.16% and 1.77%, respectively. Obviously, the PCE of Co-Fe-Se based DSSC is highest and the PCEs of Co-Fe-Se and Co-FeS based DSSCs are both superior to that of Pt, Co-Fe and Co-Fe PBA. The results could be ascribed to the facts that the Co-Fe selenides owned higher catalytic activity than the Co-Fe sulfides, partly for the selenides have rougher surface and higher surface area, which means more electroactive sites and transfer channels [62]. Besides, Co-Fe PBA and Co-Fe nanocubes with low catalytic activity were endowed with high catalytic activity after the sulfidation and selenation processes. One could be noted that Co-Fe-Se based DSSC yielded the highest short-circuit photocurrent (Jsc: 18.4 mA cm2) and fill factor (FF: 0.689), implying that the Co-Fe-Se has the lowest diffusion impedance and highest diffusion coefficient for redox species [47]. To investigate electrochemical characters of the samples for I 3 reduction, Nyquist plots of EIS, consisting of two semicircles, were conducted on the typical sandwich-like structure (CE/electrolyte/ CE) dummy cells with different CEs. The plot of Co-Fe and the corresponding modeled equivalent circuit diagram are also displayed as insets in Fig. 6b. Four relevant vital impedance properties were presented in the equivalent circuit diagram. Charge transfer resistance (Rct) is the most significant parameter which reflects resistance between the interfaces of CEs and electrolyte, or the overpotential of the electrons transferring from CE to electrolyte, which corresponds to the diameter of the semicircle at the high frequency range. Series resistance (Rs) linking to the electrolytic resistance and adhesion of FTO substrate to electrocatalyst film, what is the intercept of the real axis of the highfrequency region; Nernst diffusion-limited impedance (ZN) of the I/I 3 redox couple attributes to another semicircle. The corresponding constant phase element (CPE) represents the standard deviation from ideal capacitance [48]. The related data are summarized in Table 2. It's obvious that the Rs values of all samples show little difference and the effect of Rs can be omitted. Meanwhile, the corresponding values of Rct are distinctly different and the lower value of this crucial parameter manifests the higher catalytic activity. The Co-Fe-Se based CE exhibits the lowest Rct value of 1.10 Ohm, while the Co-Fe-S based CE (2.73 Ohm) show a lower Rct value than that of the Pt CE (5.28 Ohm) and the Co-Fe based CE shows a much higher Rct (988 Ohm). Therefore, selenide possesses smaller value of Rct than that of sulfide and has the lowest resistance of the electron transferring from CE to electrolyte and the highest catalytic activity, which matches well with Table 1 Summary of photovoltaic parameters of DSSCs based on various CEs (the average data were calculated based on four DSSCs).
Fig. 6. (a) J-V curves of DSSCs fabricated with Co-Fe-Se, Co-Fe-S, Co-Fe, Pt CEs; (b) EIS Nyquist plots of dummy cells fabricated with Co-Fe-Se, Co-Fe-S, Co-Fe and Pt (the insets are the EIS Nyquist plots of cells with Co-Fe and the corresponding equivalent circuit model); (c) CV curves of Co-Fe-Se, Co-Fe-S, Co-Fe and Pt CEs for I/I 3 redox couples at a scan rate of 50 mV s1.
CE
Jsc (mA cme2)
Co-Fe-Se Co-Fe-S Co-Fe Pt Co-Fe PBA
18.4 17.5 13.8 15.9 13.8
± ± ± ± ±
0.3 0.2 0.4 0.4 0.3
Voc (mV) 762 785 762 780 728
± ± ± ± ±
8 6 12 7 13
FF 0.689 0.660 0.276 0.658 0.170
PCE (%) ± ± ± ± ±
0.013 0.010 0.008 0.016 0.003
9.58 9.06 2.89 8.16 1.77
± ± ± ± ±
0.11 0.10 0.05 0.12 0.04
40
Y. Jiang et al. / Journal of Power Sources 369 (2017) 35e41
Table 2 Summary of EIS, CV and Tafel Polarization parameters for various CEs (the mean values were calculated based on four samples). CE
Rs (Ohm)
Co-Fe-Se Co-Fe-S Co-Fe Pt
25.5 25.5 25.6 25.5
± ± ± ±
0.1 0.1 0.1 0.1
Rct (Ohm)
Epp (mA)
log (J0/mA cm2)
1.10 ± 0.06 2.73 ± 0.11 988 ± 30 5.28 ± 0.13
185 ± 4 226 ± 7 e 358 ± 14
1.97 1.84 0.81 1.75
the results of J-V measurements. To explore the kinetics for iodide/triiodide redox reaction, cyclic voltammetry (CV) measurements were carried on at a scan rate of 50 mV s1 with the dummy cells and the vital results were displayed in Fig. 6c and summed up in Table 2. As Fig. 6c shows, the CV curve is composed of two pairs of redox peaks, namely Red-1/Ox-1 and Red-2/Ox-2. The couple peaks of Red-1/Ox-1 correspond to Eq. (3) reaction and the other peaks correspond to Eq. (4) reaction. I 3 þ 2e / 3I
(3)
3I2 þ 2e / 3I
(4)
Since the oxidation and reduction of I/I 3 are spotted as the key processes, the higher intensity of the cathodic peak current density (JRed-1) reflects the higher reduction velocity of I 3 [25]. The lower peak-to-peak separation (Epp) is interrelated to the better reversibility of the redox reaction, which also means the higher catalytic activity [36]. Evidently, the JRed-1 values decrease in the order of CoFe-Se (4.32 mA cm2) > Co-Fe-S (3.72 mA cm2) > Pt (15.1 mA cm2). The order of values of Epp is Co-Fe-Se (185 mV) > Co-Fe-S (226 mV) > Pt (358 mV). As for Co-Fe, its curve doesn't have distinct peaks, which means that it hardly possesses catalytic activity. Therefore, the performance of Co-Fe-Se is superior to the others. Catalytic activities of Co-Fe-Se/S are better than that of Pt CE, which is in agreement with the results of EIS analysis. Fig. 7a and c exhibit the CV curves of Co-Fe-Se and Co-Fe-S CEs at different scan rates in the I/I 3 system. It is obvious that a fitting well linear relationship exists between the square root of the scan rates and redox current density in Fig. 7b and d. The absolute values
± ± ± ±
0.06 0.08 0.04 0.09
log (Jlim/mA cm2) 0.661 ± 0.018 0.562 ± 0.021 0.416 ± 0.015 0.399 ± 0.023
of current density of the anodic and cathodic peaks raise with the increasing value of the square root of the scan rates. Over the same increasing scan rates, the peak current density of the selenides grows more rapidly than that of the sulfides. According to the Langmuir isotherms principle, two crucial conclusions are drawn. Firstly, it is the way of iodide species diffusing in the electrolyte that controls I 3 reduction reaction on the surfaces of CEs and thus, the diffusion of iodide species in cell with the selenides CEs is faster than that of the sulfide CEs; Secondly, the adsorption of iodide species doesn't directly relate to the redox reaction on the surface of CEs and there is no specific interaction between the I/I 3 redox couple and CEs [46]. Tafel polarization curves measurements were carried on the same symmetric cells in the EIS researches and the corresponding plots were shown in Fig. 8. The key parameters were summarized in Table 2. The limiting diffusion current density (Jlim) is the intersection of anodic branch with y axis and the exchange current density (J0) is the slope for anodic or cathodic branch, and the higher value of J0 means the better catalytic activity [57]. As Table 2 shows, the values of J0 decrease in the order of Co-Fe-Se (0.661 log (mA cm2)) > Co-Fe-S (0.562 log (mA cm2)) > Pt (0.399 log (mA cm2)) > Co-Fe (0.416 log (mA cm2)). The relationship between the conclusions of the EIS and Tafel results shows in Eq. (5). J0 ¼ RT/nFRct
(5)
Where R is the gas constant, F is Faraday's constant, T is the temperature (298 K) and Rct is the charge transfer resistance acquired from the EIS [58]. The catalytic activities reflected by Tafel polarization curves share the same trend with that of EIS and CV results.
4. Conclusions In summary, by converting Prussian blue precursors through sulfidation and selenation processes, a series of novel cobalt iron sulfides and selenides were successfully synthesized. The Co-Fe-Se/ S with a size of around 150 nm maintained the similar cubic morphology with Co-Fe PBA, which have large surface area (52.4/
Fig. 7. (a,c) CV curves of Co-Fe-Se and Co-Fe-S CEs at different scan rates of 25, 50, 75, 100, 125 and 150 mV s1 and (b, d) the relationship between redox current density and square root of scan rates of CVs for Co-Fe-Se and Co-Fe-S CEs, respectively.
Fig. 8. Tafel polarization curves for the symmetric cells fabricated with Co-Fe-Se, CoFe-S, Co-Fe and Pt CEs.
Y. Jiang et al. / Journal of Power Sources 369 (2017) 35e41
33.7 m2 g1), loose 3D porous structures, small charge transfer resistance (1.10/2.74 U) and high diffusion coefficient, as well as high exchange current density. The catalytic activities of these samples for I 3 reduction were in the order of Co-Fe-Se > Co-FeS > Pt > Co-Fe. The DSSCs fabricated with the Co-Fe-Se and Co-Fe-S yielded PCEs of 9.58% and 9.06%, which were both much higher than that of Pt-based (8.16%) DSSC. The catalytic performance of bimetal sulfides and selenides achieved a leap forward and qualitative sublimation. Hence, Co-Fe-Se and Co-Fe-S porous nanocubes fully met with our expected requirements to substitute the rare Pt materials in DSSC devices. Acknowledgments We highly appreciate the National Natural Science Foundation of China (Nos: 21676057 and 21702031) for their financial support. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.jpowsour.2017.09.080. References €tzel, Acc. Chem. Res. 42 (2009) 1788e1798. [1] M. Gra [2] M. Urbani, M. Gr€ atzel, M.K. Nazeeruddin, T. Torres, Chem. Rev. 114 (2014) 12330e12396. [3] K. Zhang, M. Savage, X. Li, Y. Jiang, M. Ishida, K. Mitsuno, S. Karasawa, T. Kato, W. Zhu, S. Yang, H. Furuta, Y. Xie, Chem. Commun. 52 (2016) 5148e5151. [4] Y. Xie, W. Wu, H. Zhu, J. Liu, W. Zhang, H. Tian, W.-H. Zhu, Chem. Sci. 7 (2016) 544e549. [5] M. Liang, M. Lu, Q.-L. Wang, W.-Y. Chen, H.-Y. Han, Z. Sun, S. Xue, J. Power Sources 196 (2011) 1657e1664. [6] F. Li, Y. Chen, X. Zong, W. Qiao, H. Fan, M. Liang, S. Xue, J. Power Sources 332 (2016) 345e354. [7] K.R. Justin Thomas, A. Venkateswararao, C.-P. Lee, K.-C. Ho, Dyes Pigm 123 (2015) 154e165. [8] S. Fan, K. Lv, H. Sun, G. Zhou, Z.-S. Wang, J. Power Sources 279 (2015) 36e47. [9] Y. Ding, Y. Tang, W. Zhu, Y. Xie, Chem. Soc. Rev. 44 (2015) 1101e1112. [10] A. Baheti, K.R. Justin Thomas, C.-T. Li, C.-P. Lee, K.-C. Ho, ACS Appl. Mater. Interfaces 7 (2015) 2249e2262. [11] M. Liang, J. Chen, Chem. Soc. Rev. 42 (2013) 3453e3488. [12] Q. Tang, H. Zhang, Y. Meng, B. He, L. Yu, Angew. Chem. Int. Ed. 127 (2015) 11610e11614. [13] F. Hao, P. Dong, Q. Luo, J.B. Li, J. Lou, H. Lin, Energy Environ. Sci. 6 (2013) 2003e2019. [14] S.N. Yun, A. Hagfeldt, T.L. Ma, Adv. Mater 26 (2014) 6210e6237. [15] Z.W. Zheng, J. Chen, Y. Hu, W.J. Wu, J.L. Hua, H. Tian, J. Mater. Chem. C 2 (2014) 8497e8500. [16] G.R. Li, J. Song, G.L. Pan, X.P. Gao, Energy Environ. Sci. 4 (2011) 1680e1683. [17] M. Wu, X. Lin, A. Hagfeldt, T. Ma, Angew. Chem. Int. Ed. 50 (2011) 3520e3524. [18] M. Wu, J. Bai, Y. Wang, A. Wang, X. Lin, L. Wang, Y. Shen, Z. Wang, A. Hagfeldt, T. Ma, J. Mater. Chem. 22 (2012) 11121e11127. [19] E. Palomares, J.N. Clifford, S.A. Haque, T. Lutz, J.R. Durrant, J. Am. Chem. Soc. 125 (2003) 475e482. [20] X.J. Zheng, J. Deng, N. Wang, D.H. Deng, W.H. Zhang, X.H. Bao, C. Li, Angew. Chem. Int. Ed. 53 (2014) 7023e7027. [21] R. Leary, A. Westwood, Carbon 49 (2011) 741e772. [22] B. Anothumakkool, O. Game, S.N. Bhange, T. Kumari, S.B. Ogale, S. Kurungot, Nanoscale 6 (2014) 10332e10339. [23] M.S. Su'ait, M.Y.A. Rahman, A. Ahmad, Sol. Energy 115 (2015) 452e470.
41
[24] S. Yun, J.N. Freitas, A.F. Nogueira, Y. Wang, S. Ahmad, Z.-S. Wang, Prog. Polym. Sci. 59 (2016) 1e40. [25] X. Zhang, T.Z. Jing, S.Q. Guo, G.D. Gao, L. Liu, Rsc. Adv. 4 (2014) 50312e50317. [26] Z.Q. Li, F. Gong, G. Zhou, Z.S. Wang, J. Phys. Chem. C 117 (2013) 6561e6566. [27] M. Zheng, J.H. Huo, Y.G. Tu, J.H. Wu, L.H. Hu, S.Y. Dai, Electrochim. Acta 173 (2015) 252e259. [28] F. Gong, H. Wang, X. Xu, G. Zhou, Z.-S. Wang, J. Am. Chem. Soc. 134 (2012) 10953e10958. [29] Z. Jin, M. Zhang, M. Wang, C. Feng, Z.-S. Wang, Acc. Chem. Res. 50 (2017) 895e904. [30] Y. Duan, Q. Tang, J. Liu, B. He, L. Yu, Angew. Chem. Int. Ed. 53 (2014) 14569e14574. [31] Z. Hu, K. Xia, J. Zhang, Z. Hu, Y. Zhu, RSC Adv. 4 (2014) 42917e42923. [32] W. Zhao, X.L. Zhu, H. Bi, H.L. Cui, S.R. Sun, F.Q. Huang, J. Power Sources 242 (2013) 28e32. [33] Y. Hu, Z. Zheng, H. Jia, Y. Tang, L. Zhang, J. Phys. Chem. C 112 (2008) 13037e13042. [34] Y.C. Wang, D.Y. Wang, Y.T. Jiang, H.A. Chen, C.C. Chen, K.C. Ho, H.L. Chou, C.W. Chen, Angew. Chem. Int. Ed. 52 (2013) 6694e6698. [35] J.-C. Tsai, M.-H. Hon, I.-C. Leu, RSC Adv. 5 (2015) 4328e4333. [36] C.-W. Kung, H.-W. Chen, C.-Y. Lin, K.-C. Huang, R. Vittal, K.-C. Ho, ACS Nano 6 (2012) 7016e7025. [37] J. Zhang, S. Najmaei, H. Lin, J. Lou, Nanoscale 6 (2014) 5279e5283. [38] S. Jiang, X. Yin, J.T. Zhang, X.Y. Zhu, J.Y. Li, M. He, Nanoscale 7 (2015) 10459e10464. [39] D.D. Song, M.C. Li, Y.J. Jiang, Z. Chen, F. Bai, Y.F. Li, B. Jiang, J. Photochem. Photobiol. A 279 (2014) 47e51. [40] J. Ma, W. Shen, F. Yu, J. Power Sources 351 (2017) 58e66. [41] S. Huang, Q. He, W. Chen, J. Zai, Q. Qiao, X. Qian, Nano Energy 15 (2015) 205e215. [42] X. Cui, Z. Xie, Y. Wang, Nanoscale 8 (2016) 11984e11992. [43] Z.Y. Tang, Q.W. Tang, J.H. Wu, Y. Li, Q. Liu, M. Zheng, Y.M. Xiao, G.T. Yue, M.L. Huang, J.M. Lin, RSC Adv 2 (2012) 5034e5037. [44] X.B. Xu, D.K. Huang, K. Cao, M.K. Wang, S.M. Zakeeruddin, M. Gr€ atzel, Sci. Rep. 3 (2013) 4. [45] Z.Y. Zhang, S.P. Pang, H.X. Xu, Z.Z. Yang, X.Y. Zhang, Z.H. Liu, X.G. Wang, X.H. Zhou, S.M. Dong, X. Chen, L. Gu, G.L. Cui, RSC Adv. 3 (2013) 16528e16533. [46] X.Y. Zhang, S.P. Pang, X. Chen, K.J. Zhang, Z.H. Liu, X.H. Zhou, G.L. Cui, RSC Adv. 3 (2013) 9005e9010. [47] Z. Wei, J. Wang, S. Mao, D. Su, H. Jin, Y. Wang, F. Xu, H. Li, Y. Wang, ACS Catal. 5 (2015) 4783e4789. [48] X.Y. Yu, Y. Feng, Y. Jeon, B. Guan, X.W.D. Lou, U. Paik, Adv. Mater 28 (2016) 9006e9011. [49] H. Guo, T. Li, W. Chen, L. Liu, X. Yang, Y. Wang, Y. Guo, Nanoscale (2014) 15168e15174. [50] J. Lehto, S. Haukka, R. Harjula, M. Blomberg, J. Chem. Soc. Dalton Trans. (1990) 1007e1011. [51] B. Yuan, W. Luan, S.-t. Tu, Dalton Trans. 41 (2012) 772e776. [52] C.E.M. Campos, J.C. de Lima, T.A. Grandi, K.D. Machado, P.S. Pizani, Phys. B 324 (2002) 409e418. [53] A. Shetty, A. Saha, M. Makkar, R. Viswanatha, Phys. Chem. Chem. Phys. 18 (2016) 25887e25892. [54] Y. Li, J. Zhao, Y. Dan, D. Ma, Y. Zhao, S. Hou, H. Lin, Z. Wang, Chem. Eng. J. 166 (2011) 428e434. [55] X. Miao, K. Pan, G. Wang, Y. Liao, L. Wang, W. Zhou, B. Jiang, Q. Pan, G. Tian, Chem. Eur. J. 20 (2014) 474e482. [56] N. Berry, M. Cheng, C.L. Perkins, M. Limpinsel, J.C. Hemminger, M. Law, Adv. Energy Mater 2 (2012) 1124e1135. [57] A. Padhi, K. Nanjundaswamy, C. Masquelier, S. Okada, J. Goodenough, J. Electrochem. Soc. 144 (1997) 1609e1613. [58] R. Ma, K. Takada, K. Fukuda, N. Iyi, Y. Bando, T. Sasaki, Angew. Chem. Int. Ed. 47 (2008) 86e89. [59] T. Liu, A.M. Asiri, X. Sun, Nanoscale 8 (2016) 3911e3915. [60] J. Wang, N. Yang, H. Tang, Z. Dong, Q. Jin, M. Yang, D. Kisailus, H. Zhao, Z. Tang, D. Wang, Angew. Chem. Int. Ed. 52 (2013) 6417e6420. €tzel, J.H. Delcamp, ACS Appl. [61] N.P. Liyanage, A. Yella, M. Nazeeruddin, M. Gra Mater. Interfaces 8 (2016) 5376e5384. [62] B. Lei, G. Li, X. Gao, J. Mater. Chem. A 2 (2014) 3919e3925.
Contents lists available at ScienceDirect
Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour
Cobalt iron selenide/sulfide porous nanocubes as high-performance electrocatalysts for efficient dye-sensitized solar cells Yiqing Jiang, Xing Qian*, Yudi Niu, Li Shao, Changli Zhu, Linxi Hou** College of Chemical Engineering, Fuzhou University, Xueyuan Road No. 2, Fuzhou 350116, China
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
Novel Co-Fe-Se/S porous nanocubes (PNCs) were prepared by a template method. The PNCs were applied as efficient CE electrocatalysts for DSSCs. The PNCs had large surface area, high catalytic activity and good conductivity. The PCEs (9.58% and 9.06%) of the PNCs were higher than that of Pt (8.16%).
a r t i c l e i n f o
a b s t r a c t
Article history: Received 26 July 2017 Received in revised form 27 August 2017 Accepted 27 September 2017
A novel series of ternary compounds, namely cobalt iron selenide/sulfide nanocubes, are successfully synthesized as counter electrode (CE) materials for dye-sensitized solar cells (DSSCs), which deliver excellent performances. Homogeneous cobalt iron Prussian-blue-analog (PBA) nanocubes are prepared as the templates and are subsequently dealt with selenation/sulfidation processes via hydrothermal methods. Owing to their unique morphology, porous structure, high surface area, small charge transfer resistance and high diffusion coefficient, the Co-Fe-Se/S nanocubes possess high catalytic activity and excellent conductivity, which are tested and verified by electrochemical measurements. Meanwhile, cobalt iron selenide/sulfide nanocubes CEs achieve high efficiencies of 9.58% and 9.06%, respectively, which are both higher than that of Pt CE (8.16%). All these prominent merits make them outstanding and promising participants among Pt-free CE materials of DSSCs with lower production costs and higher power conversion efficiency. © 2017 Elsevier B.V. All rights reserved.
Keywords: Cobalt iron selenides Cobalt iron sulfides Porous nanocubes Photovoltaic performances Counter electrodes Dye-sensitized solar cells
1. Introduction Dye-sensitized solar cells (DSSCs) have been regarded as a kind of promising power conversion devices and have attracted great interest because of their characteristics, such as ease of fabrication, low expenses and high power conversion efficiency (PCE) [1,2]. A
* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (X. Qian), [email protected] (L. Hou). https://doi.org/10.1016/j.jpowsour.2017.09.080 0378-7753/© 2017 Elsevier B.V. All rights reserved.
typical DSSC device is composed of a counter electrode (CE), a dyeloaded TiO2 electrode and triiodide/iodide redox electrolyte. Each part has been studied to pursue better performance [3e11]. For example, Li et al. synthesized DTP-based dyes, which achieved a power conversion efficiency of 8.14% [6]. Among them, a counter electrode (CE) is viewed as one of the dominant parts of the classical DSSCs, transferring electron from outside circuit to redox electrolyte and reducing I 3 to I . Conventionally, platinum (Pt) was widely applied as common commercial CE material in DSSCs, on account of its favorable and famous conductivity and catalytic
36
Y. Jiang et al. / Journal of Power Sources 369 (2017) 35e41
activity [12]. However, the abundant commercial application and future development were limited and hindered by its obvious defects, such as low earth-abundance ratio, high price and ease of corrosion [13e15]. In this urgent situation, the further studies on alternatives of Pt are desperately desirable. To date, in order to seek ideal substitutes, various materials, including transition metal compounds (TMCs) [16e19], carbonaceous materials [20e22], conductive polymers [23,24] and their hybrids [25e27] have been researched. Among them, TMCs stand out because of their low costs, Pt-like catalytic activity and excellent conductivity [1]. Carbide [17], nitrides [16], oxides [19], phosphides [18], sulfides, selenides and their hybrids have been frequently investigated. It is the sulfides and selenides that win most of the attention for not only ease of synthesis via hydrothermal method but also favorable PCEs [14,28e30]. Various alloys, such as Co3S4 [31], NiS2 [26], NiS [32], FeS2 [33,34], CoS2 [35], CoS [36], MoS2 [37e40], NiSe2 [25], CoSe2 [37e39], FeSe2 [41] and so on, have been able to be approached and explored. For example, Cui et al. synthesized CoS2 embedded carbon nanocages with a high PCE of 8.20% (Pt, 7.88%) [42]. Except the chemical composition, the morphology mainly influences the performance of CE materials in DSSCs. Flower-like [43], nanofilm [25,44e46], nanotube [47], nanocages [42] and other morphologies have been designed and investigated. Thus, the roles that the chemical composition and morphology played could not be ignored in the design of catalyst. In this work, we chose cobalt iron Prussian-blue-analog (PBA) nanocubes as the precursors, which were hardly out of shape at high temperature but easy to convert, along with relatively high surface area [48,49]. Furthermore, the metal sulfides and selenides had great performance as CE materials in DSSCs with small charge resistance and high diffusion coefficient. For all the above reasons, we designed the synthetic routes that PBA nanocubes were prepared as the templates and subsequently treated with selenation/ sulfidation processes via hydrothermal methods. Benefiting from the unique morphology, structure and composition, the devices with Co-Fe selenide and sulfide porous nanocubes delivered PCEs of 9.58% and 9.06%, respectively, which are superior to that of Pt (8.16%). Hence, the results demonstrated the excellent electrocatalytic activity and remarkable conductivity of Co-Fe selenide and sulfide porous nanocubes.
2. Experimental section 2.1. Materials Potassium hexacyanoferrate (Ⅲ) (K3[Fe(CN)6], Sigma-Aldrich Corporation, 99.0%), cobalt nitrate hexahydrate (Co(NO3)2$6H2O, Aladdin Corporation, 99.0%), sodium citrate nonahydrate (C6H5Na3O7$9H2O, Aladdin Corporation, 99.0%), selenium powder (Se, Sigma-Aldrich Corporation, 99.5%), absolute ethanol (C2H5OH, Sigma-Aldrich Corporation, 99.7%), ethylenediamine (C2H8N2, Aladdin Corporation, 99.6%), sodium sulphide (Na2S, Aladdin Corporation, AR), 4-tert-butyl pyridine (C9H13N, Aladdin Corporation, 98.0%), titanium tetrachloride (TiCl4, Aladdin Corporation, AR), lithium perchlorate (LiClO4, Aladdin Corporation, 99.9%), iodine (I2, Aladdin Corporation, 99.8%), lithium iodide (LiI, Aladdin Corporation, 99.0%), acetonitrile (CH3CN, Sinopharm Chemical Reagent, 99.0%), the N719 dye (Solaronix Corporation, Switzerland, AR) and 1,2-dimethyl-3-propylimidazolium iodide (DMPII, Aladdin Corporation, 99.0%) were directly used without any further purification. The FTO glasses were obtained from Nippon Sheet Glass (Japan) with the sheet resistance of 15 U sq1.
2.2. Synthesis 2.2.1. Synthesis of Co-Fe prussian-blue-analog (PBA) cubes In a typical synthesis, 2 mmol of potassium hexacyanoferrate (Ⅲ) was dissolved in 100 mL of deionized (DI) water to form a uniform clean solution A. 3 mmol of cobalt nitrate and 4.5 mmol sodium citrate were dissolved in 100 mL of DI water to form an orange solution B. Then, the solution A was poured into solution B under vigorously stirring for 10 min. The obtained solution was aged for 2 days at ambient temperature. The black purple Co-Fe PBA nanocubes were precipitated by centrifugation and washed with water and ethanol for several times, and dried at 40 C for 12 h under vacuum. 2.2.2. Synthesis of Co-Fe-Se nanocubes 80 mg of Se powder was added into an ethanol solution (20 mL) containing 6 mL of ethylenediamine and the mixture was stirred for 0.5 h. Then, 40 mg of the as-prepared Co-Fe PBA cubes was dispersed in absolute ethanol (20 mL) under ultrasonication for 10 min to form a uniform suspension. The suspension was poured into the above solution. The mixture was transformed into a 50 mL teflon-lined stainless-steel autoclave and heated at 160 C for 6 h. The black sample was dried at 50 C for 12 h under vacuum after being centrifugated and washed with ethanol. Finally, in order to increase its crystallinity, the production was annealed at 350 C for 2 h at the heating rate of 1 C min1 under Ar atmosphere. The production was named as Co-Fe-Se. In comparison, the as-prepared Co-Fe PBA cubes were directly annealed as the same as the sample Co-Fe-Se and the sample named Co-Fe was obtained. 2.2.3. Synthesis of Co-Fe-S nanocubes 40 mg of the as-prepared Co-Fe PBA cubes and 10 mL of Na2S aqueous solution (4 mg mL1) were added into an ethanol solution (30 mL), which were dispersed under ultrasonication. The next experimental processes of this black production, named as Co-Fe-S, were the same as that of the sample Co-Fe-Se. 2.3. Fabrications of counter electrodes FTO glasses were washed with detergent, deionized water, acetone and ethanol in sequence. Typically, 100 mg of the obtained samples were dispersed in 10 mL of ethyl ethanol. After 30 min of sonication and stirring, a uniform ink suspension had been acquired before spinning coating on a piece of already pretreated FTO glass by spin-casting technique at a rotating speed of 500 rpm for 16 s and the covered FTO glass was heated and dried at 120 C for 5 min. This process was repeated for several times until the coating thickness was suitable. Approximately, the loading capacity of each FTO glass is 0.67 mg cm2. As a reference, the Pt CE was obtained by thermally depositing a platinum layer (0.02 M hexachloroplatinic acid in isopropanol) on the surface of FTO at 450 C for 30 min. 2.4. Fabrications of DSSCs The screen-printing method was utilized to cover the transparent nanocrystalline TiO2 layer (thickness, ~12 mm) and the scattering TiO2 layer (thickness, ~4 mm) on the conductive side of the pretreated FTO glass with 20 and 200 nm TiO2 sols, respectively. After being heated in air at 500 C for 1 h, the FTO glass was immersed in 0.04 M titanium tetrachloride aqueous solution at 70 C for 1 h and following sintered in air at 500 C for 1 h again. Then, the cooled photoanodes were soaked and sensitized in 0.3 mM N719 dye ethanol solution overnight in a dry and dark environment. Finally, to assemble a sandwich structure of DSSCs, the I/I 3 based electrolyte was injected into the vacuum space
Y. Jiang et al. / Journal of Power Sources 369 (2017) 35e41
between a photoanode and a CE. The electrolyte consisted of 0.3 M DMPII, 0.05 M I2, 0.1 M LiI and 0.5 M 4-tert-butylpyridine in acetonitrile solution. The active area of the cell was 0.16 cm2.
2.5. Materials characterization The morphology and nanostructure of the products were characterized by a field-mission scanning electron microscopy (SEM, S4800, Hitachi) at 5 kV and 7 mA equipped with energy-dispersive Xray spectroscopy (EDS). The crystal structures of those nanocubes were identified by X-ray diffraction (XRD, X0 Pert PRO, Cu Ka, l ¼ 0.15406 nm) at 40 kV and 40 mA in the range of 10e70 within 10 min. The Raman spectrums were conducted on a microscopic confocal Raman spectrometer (Raman, in Via Reflex, UK). Special surface area and pore size distribution were measured by the Brunauer-Emmett-Teller (BET) analyzer (Micromeritics, ASAP 2020 M, USA). The chemical properties were researched by utilizing X-ray photoelectron spectrometer (XPS, ESCALAB 250, Mg Ka, USA). All the electrochemical measurements were conducted on the CHI660E electrochemical work station (CH Instruments). The cyclic voltammetry (CV) was employed in a three-electrode system. The as-prepared CE, platinum sheet and Ag/AgCl electrode were served as the working electrode, the counter electrode and the reference electrode in the system, respectively. All electrodes were inserted into an electrolyte containing 0.1 M LiClO4, 10 mM LiI and 1 mM I2 in acetonitrile solution. The scan rate was 50 mV s1 and the scan potential range was set from 0.5 to 0.9 V. For the electrochemical impedance spectroscopy (EIS) and Tafel polarization curves, a traditional symmetrical dummy cell configuration (CE/electrolyte/ CE) was fabricated. EIS was measured from 100 kHz to 100 mHz with an amplitude of 5 mV, while Tafel curves were operated at a scan rate of 10 mV s1 between 1.0 V and 1.0 V. To obtain the PCEs of DSSCs, the photocurrent density-voltage (J-V) curves were collected via putting the solar cells under AM 1.5 G illumination (100 mW cm2) irradiation provided by a standard solar simulator (XM-500W, Trusttech) which was calibrated with a standard silicon solar cell (91150V, Newport Corporation).
37
3. Results and discussion Fig. 1ac show the typical SEM images of the as-prepared uniform solid Co-Fe PBA nanocubes, which have smooth surfaces, sharp edges and corners with an average size of around 150 nm. The average size did not change distinctly during the follow processes. After hydrothermal treatment with Na2S solution and anneal, the solid precursors were etched and got round edges and corners as well as sunk surfaces with tiny particles (Fig. 1gei). As shown in Fig. 1def, after selenation, the smooth surface of precursors got much rougher and the particles on the surface of selenide were bigger than that of sulfide. Fig. S1 (Supporting information, SI) shows the SEM images of the Co-Fe sample. Commonly, a sample with good conductivity owned bright and clear SEM images. Therefore, the conductivity of precursors after sulfidation and selenation processes were enhanced, which could be proven intuitively by the SEM images. Fig. 2 and Fig. S2 (SI) exhibit the composition and phase structures of Co-Fe PBA, Co-Fe-Se, Co-Fe-S and Co-Fe nanocubes. It is obvious that diffraction peaks of the PBA are excellently assigned to the (ICDD PDF No. 01-075-0038), which indicates the pure quality of synthesized Co-Fe PBA nanocubes [50]. As for Co-Fe-Se nanocubes, the peaks at 23.8 , 29.1, 30.9 , 34.6 , 36.0 , 37.3 , 40.4 , 44.2 , 48.0 , 50.7, 53.9 , 55.7, 57.3 , 63.5 , 63.9 and 64.0 correspond to the (1 1 0), (0 1 1), (0 2 0), (1 0 1), (1 1 1), (1 2 0), (2 0 0), (2 1 0), (1 2 1), (1 1 1), (0 0 2), (0 3 1), (2 2 1), (1 3 1), (1 2 2) and (0 4 0) crystal planes of orthorhombic FeSe2 (ICDD PDF No. 00-012-0291) [51]. The peaks at 30.8 , 34.8 , 36.3 , 47.8 and 57.2 correspond to the (1 1 0), (1 1 1), (0 1 2), (1 2 1) and (2 1 1) crystal planes of orthorhombic CoSe2 (ICDD PDF No. 00-010-0408) [52]. Among those displayed peaks, the weak peaks at 29.9 and 54.5 show the existence of the O2, which is the reason for peak shifting. All the diffraction peaks of Co-Fe-S match well with cubic Co3S4 (ICDD PDF No. 01-073-1703) and hexagonal FeCoS2 (ICDD PDF No. 01-0750607). Several peaks appear at 30.7, 35.2 , 46.4 and 54.7 correspond to the (1 0 0), (1 0 1), (1 0 2) and (1 1 0) planes of hexagonal FeCoS2, which conforms with the earlier report [53]. Besides, the peaks at 16.3 , 26.7, 31.7, 38.3 , 47.4 , 50.6 and 55.4 correspond
Fig. 1. SEM images of (aec) Co-Fe PBA nanocubes, (def) Co-Fe-Se nanocubes, (gei) Co-Fe-S nanocubes.
38
Y. Jiang et al. / Journal of Power Sources 369 (2017) 35e41
Fig. 2. XRD patterns of Co-Fe-Se, Co-Fe-S and Co-Fe PBA.
to the (1 1 1), (2 2 0), (3 1 1), (4 0 0), (5 1 1) and (4 4 0) planes of cubic Co3S4 [54]. To further confirm the crystal structure, the Raman spectra are employed and the results are shown in Fig. S3 (SI). It can be seen in Fig. S3a (SI), the peaks at 316, 335, 385 and 410 cm1 determine the existence of hexagonal FeCoS2, while peaks at 472, 513, 609 and 679 cm1 determine the existence of cubic Co3S4 [55]. In addition, the Raman spectrum of Co-Fe-Se is composed of character peaks of orthorhombic CoSe2 (467, 516 and 663 cm1) and orthorhombic FeSe2 (180, 217 and 254 cm1) [51,56]. XPS was employed to determine the surface chemical compositions and the valence states of the samples Co-Fe-Se/S. The survey spectrum (Fig. 3a) presents the attendance of Co, Fe, Se, C and O elements in Co-Fe-Se. The results of Co (Figs. 3b and 4b), Fe (Figs. 3c and 4c), Se (Fig. 3d) and S (Fig. 4d) elements were analyzed via the Gaussian fitting method. Fig. 3b are constituted with two spin-orbit doublets and a pair of shakeup satellites (marked as “Sat.”). The peaks of Co 2p1/2 at 793.6 and 797.1 eV and the peaks of Co 2p3/2 at 778.6 and 780.8 eV are characteristic of Co3þ and Co2þ. The spectrum of Fe 2p in Fig. 3c can be fitted into a doublets and a satellite. The existence of Se2 is proved by the character speaks appearing at 55.6 and 54.7 eV in Fig. 3d and the broad satellite at 59.6 eV is for Se species with high oxide state [59]. There are Co2þ, Co3þ, Fe2þ and Se2 ions on the surface composition of Co-Fe-Se. Fig. 4a determines the presence of Co, Fe, S, C and O elements in Co-Fe-S. The corresponding peak positions in Fig. 4b are as the same as that of Fig. 3b, suggesting the attendance of Co3þ and Co2þ.
Fig. 3. XPS spectra of Co-Fe-Se, (a) survey spectrum, (b) Co 2p, (c) Fe 2p, (d) Se 3d.
Fig. 4. XPS spectra of Co-Fe-S, (a) survey spectrum, (b) Co 2p, (c) Fe 2p, (d) S 2p.
In Fig. 4c, besides the peaks at 708.5 and 722.2 eV reflect the Fe 2p3/ and Fe 2p1/2 of Fe2þ (typical Fe-S bond), the other peaks at 710.5 and 724.1 eV link with the Fe 2p3/2 and Fe 2p1/2 of Fe3þ. Furthermore, the S 2p1/2 and S 2p3/2 peaks locate at 163.5 and 162.3 eV in Fig. 4d, which is the evidence of the presence of S2 in Co-Fe-S. The peaks at 707.5 and 719.0 eV reflect the Fe 2p3/2 and Fe 2p1/2 of Fe2þ, which suggests the existence of typical Fe-Se bond in Co-Fe-Se. According to previous researches, Co2þ and Fe2þ can be easily oxidized into Co3þ and Fe3þ in Co/Fe-based alloy [57,58]. One thing can be noted that there are Co2þ, Co3þ, Fe2þ, Fe3þ and S2 ions on the surface composition of Co-Fe-S. The porosity and surface area of Co-Fe-S and Co-Fe-Se were studied by N2-sorption measurements. As shown in Fig. 5a, the
2
Fig. 5. Pore-structure analysis of Co-Fe-Se and Co-Fe-S, (a) N2-adsorption-desorptipn isotherm at 77 K, (b) pore-size distribution.
Y. Jiang et al. / Journal of Power Sources 369 (2017) 35e41
plots of Co-Fe-S and Co-Fe-Se display the properties of the typical type IV behavior (characteristic of mesoporous materials), indicating them contains mesoporous. Compared with the Co-Fe-S (33.7 m2 g1), the Co-Fe-Se (52.4 m2 g1) shows a distinct increase in Brunauer-Emmett-Teller (BET) surface area. Besides, the total pore volume of Co-Fe-S and Co-Fe-Se are 0.226 cm3 g1 and 0.213 cm3 g1. As the pore-size distribution shows (Fig. 5b), the samples Co-Fe-S and Co-Fe-Se armed with the main mesopores with the diameters of 3.54 and 3.94 nm, respectively. The high surface area and pore volume contribute to electrolyte penetration and provide more active sites, which is expected to improve the catalytic activity and delivers better performance in DSSCs [48,49,53,60]. The Co-Fe PBA nanocubes (7.90 m2 g1, 0.0607 cm3 g1) were tested to prove that the sulfation and selenation processes indeed increased the surface area and formed pores. Fig. 6a reveals the photocurrent density-voltage (J-V) curves of the DSSCs fabricated with the Co-Fe-Se, Co-Fe-S, Co-Fe and Ptbased CEs, and the summary of key photovoltaic parameters of these DSSCs are listed in Table 1. The Co-Fe PBA and Co-Fe worked
39
as comparisons to verify a result that sulfidation and selenation processes do enhance PCE. The fill factor (FF) and PCE of DSSCs fabricated with samples are calculated as follows. FF ¼ JmVm/JscVoc
(1)
PCE ¼ JmVm/Pin
(2)
Where Jm is the maximum current density, Vm is the maximum voltage, Jsc is the short circuit current density, Voc is the open circuit voltage and Pin is the power input (100 mW cm2) [61]. The PCEs of Co-Fe-Se, Co-Fe-S, Co-Fe, Pt and Co-Fe PBA based CEs are 9.58%, 9.06%, 2.89%, 8.16% and 1.77%, respectively. Obviously, the PCE of Co-Fe-Se based DSSC is highest and the PCEs of Co-Fe-Se and Co-FeS based DSSCs are both superior to that of Pt, Co-Fe and Co-Fe PBA. The results could be ascribed to the facts that the Co-Fe selenides owned higher catalytic activity than the Co-Fe sulfides, partly for the selenides have rougher surface and higher surface area, which means more electroactive sites and transfer channels [62]. Besides, Co-Fe PBA and Co-Fe nanocubes with low catalytic activity were endowed with high catalytic activity after the sulfidation and selenation processes. One could be noted that Co-Fe-Se based DSSC yielded the highest short-circuit photocurrent (Jsc: 18.4 mA cm2) and fill factor (FF: 0.689), implying that the Co-Fe-Se has the lowest diffusion impedance and highest diffusion coefficient for redox species [47]. To investigate electrochemical characters of the samples for I 3 reduction, Nyquist plots of EIS, consisting of two semicircles, were conducted on the typical sandwich-like structure (CE/electrolyte/ CE) dummy cells with different CEs. The plot of Co-Fe and the corresponding modeled equivalent circuit diagram are also displayed as insets in Fig. 6b. Four relevant vital impedance properties were presented in the equivalent circuit diagram. Charge transfer resistance (Rct) is the most significant parameter which reflects resistance between the interfaces of CEs and electrolyte, or the overpotential of the electrons transferring from CE to electrolyte, which corresponds to the diameter of the semicircle at the high frequency range. Series resistance (Rs) linking to the electrolytic resistance and adhesion of FTO substrate to electrocatalyst film, what is the intercept of the real axis of the highfrequency region; Nernst diffusion-limited impedance (ZN) of the I/I 3 redox couple attributes to another semicircle. The corresponding constant phase element (CPE) represents the standard deviation from ideal capacitance [48]. The related data are summarized in Table 2. It's obvious that the Rs values of all samples show little difference and the effect of Rs can be omitted. Meanwhile, the corresponding values of Rct are distinctly different and the lower value of this crucial parameter manifests the higher catalytic activity. The Co-Fe-Se based CE exhibits the lowest Rct value of 1.10 Ohm, while the Co-Fe-S based CE (2.73 Ohm) show a lower Rct value than that of the Pt CE (5.28 Ohm) and the Co-Fe based CE shows a much higher Rct (988 Ohm). Therefore, selenide possesses smaller value of Rct than that of sulfide and has the lowest resistance of the electron transferring from CE to electrolyte and the highest catalytic activity, which matches well with Table 1 Summary of photovoltaic parameters of DSSCs based on various CEs (the average data were calculated based on four DSSCs).
Fig. 6. (a) J-V curves of DSSCs fabricated with Co-Fe-Se, Co-Fe-S, Co-Fe, Pt CEs; (b) EIS Nyquist plots of dummy cells fabricated with Co-Fe-Se, Co-Fe-S, Co-Fe and Pt (the insets are the EIS Nyquist plots of cells with Co-Fe and the corresponding equivalent circuit model); (c) CV curves of Co-Fe-Se, Co-Fe-S, Co-Fe and Pt CEs for I/I 3 redox couples at a scan rate of 50 mV s1.
CE
Jsc (mA cme2)
Co-Fe-Se Co-Fe-S Co-Fe Pt Co-Fe PBA
18.4 17.5 13.8 15.9 13.8
± ± ± ± ±
0.3 0.2 0.4 0.4 0.3
Voc (mV) 762 785 762 780 728
± ± ± ± ±
8 6 12 7 13
FF 0.689 0.660 0.276 0.658 0.170
PCE (%) ± ± ± ± ±
0.013 0.010 0.008 0.016 0.003
9.58 9.06 2.89 8.16 1.77
± ± ± ± ±
0.11 0.10 0.05 0.12 0.04
40
Y. Jiang et al. / Journal of Power Sources 369 (2017) 35e41
Table 2 Summary of EIS, CV and Tafel Polarization parameters for various CEs (the mean values were calculated based on four samples). CE
Rs (Ohm)
Co-Fe-Se Co-Fe-S Co-Fe Pt
25.5 25.5 25.6 25.5
± ± ± ±
0.1 0.1 0.1 0.1
Rct (Ohm)
Epp (mA)
log (J0/mA cm2)
1.10 ± 0.06 2.73 ± 0.11 988 ± 30 5.28 ± 0.13
185 ± 4 226 ± 7 e 358 ± 14
1.97 1.84 0.81 1.75
the results of J-V measurements. To explore the kinetics for iodide/triiodide redox reaction, cyclic voltammetry (CV) measurements were carried on at a scan rate of 50 mV s1 with the dummy cells and the vital results were displayed in Fig. 6c and summed up in Table 2. As Fig. 6c shows, the CV curve is composed of two pairs of redox peaks, namely Red-1/Ox-1 and Red-2/Ox-2. The couple peaks of Red-1/Ox-1 correspond to Eq. (3) reaction and the other peaks correspond to Eq. (4) reaction. I 3 þ 2e / 3I
(3)
3I2 þ 2e / 3I
(4)
Since the oxidation and reduction of I/I 3 are spotted as the key processes, the higher intensity of the cathodic peak current density (JRed-1) reflects the higher reduction velocity of I 3 [25]. The lower peak-to-peak separation (Epp) is interrelated to the better reversibility of the redox reaction, which also means the higher catalytic activity [36]. Evidently, the JRed-1 values decrease in the order of CoFe-Se (4.32 mA cm2) > Co-Fe-S (3.72 mA cm2) > Pt (15.1 mA cm2). The order of values of Epp is Co-Fe-Se (185 mV) > Co-Fe-S (226 mV) > Pt (358 mV). As for Co-Fe, its curve doesn't have distinct peaks, which means that it hardly possesses catalytic activity. Therefore, the performance of Co-Fe-Se is superior to the others. Catalytic activities of Co-Fe-Se/S are better than that of Pt CE, which is in agreement with the results of EIS analysis. Fig. 7a and c exhibit the CV curves of Co-Fe-Se and Co-Fe-S CEs at different scan rates in the I/I 3 system. It is obvious that a fitting well linear relationship exists between the square root of the scan rates and redox current density in Fig. 7b and d. The absolute values
± ± ± ±
0.06 0.08 0.04 0.09
log (Jlim/mA cm2) 0.661 ± 0.018 0.562 ± 0.021 0.416 ± 0.015 0.399 ± 0.023
of current density of the anodic and cathodic peaks raise with the increasing value of the square root of the scan rates. Over the same increasing scan rates, the peak current density of the selenides grows more rapidly than that of the sulfides. According to the Langmuir isotherms principle, two crucial conclusions are drawn. Firstly, it is the way of iodide species diffusing in the electrolyte that controls I 3 reduction reaction on the surfaces of CEs and thus, the diffusion of iodide species in cell with the selenides CEs is faster than that of the sulfide CEs; Secondly, the adsorption of iodide species doesn't directly relate to the redox reaction on the surface of CEs and there is no specific interaction between the I/I 3 redox couple and CEs [46]. Tafel polarization curves measurements were carried on the same symmetric cells in the EIS researches and the corresponding plots were shown in Fig. 8. The key parameters were summarized in Table 2. The limiting diffusion current density (Jlim) is the intersection of anodic branch with y axis and the exchange current density (J0) is the slope for anodic or cathodic branch, and the higher value of J0 means the better catalytic activity [57]. As Table 2 shows, the values of J0 decrease in the order of Co-Fe-Se (0.661 log (mA cm2)) > Co-Fe-S (0.562 log (mA cm2)) > Pt (0.399 log (mA cm2)) > Co-Fe (0.416 log (mA cm2)). The relationship between the conclusions of the EIS and Tafel results shows in Eq. (5). J0 ¼ RT/nFRct
(5)
Where R is the gas constant, F is Faraday's constant, T is the temperature (298 K) and Rct is the charge transfer resistance acquired from the EIS [58]. The catalytic activities reflected by Tafel polarization curves share the same trend with that of EIS and CV results.
4. Conclusions In summary, by converting Prussian blue precursors through sulfidation and selenation processes, a series of novel cobalt iron sulfides and selenides were successfully synthesized. The Co-Fe-Se/ S with a size of around 150 nm maintained the similar cubic morphology with Co-Fe PBA, which have large surface area (52.4/
Fig. 7. (a,c) CV curves of Co-Fe-Se and Co-Fe-S CEs at different scan rates of 25, 50, 75, 100, 125 and 150 mV s1 and (b, d) the relationship between redox current density and square root of scan rates of CVs for Co-Fe-Se and Co-Fe-S CEs, respectively.
Fig. 8. Tafel polarization curves for the symmetric cells fabricated with Co-Fe-Se, CoFe-S, Co-Fe and Pt CEs.
Y. Jiang et al. / Journal of Power Sources 369 (2017) 35e41
33.7 m2 g1), loose 3D porous structures, small charge transfer resistance (1.10/2.74 U) and high diffusion coefficient, as well as high exchange current density. The catalytic activities of these samples for I 3 reduction were in the order of Co-Fe-Se > Co-FeS > Pt > Co-Fe. The DSSCs fabricated with the Co-Fe-Se and Co-Fe-S yielded PCEs of 9.58% and 9.06%, which were both much higher than that of Pt-based (8.16%) DSSC. The catalytic performance of bimetal sulfides and selenides achieved a leap forward and qualitative sublimation. Hence, Co-Fe-Se and Co-Fe-S porous nanocubes fully met with our expected requirements to substitute the rare Pt materials in DSSC devices. Acknowledgments We highly appreciate the National Natural Science Foundation of China (Nos: 21676057 and 21702031) for their financial support. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.jpowsour.2017.09.080. References €tzel, Acc. Chem. Res. 42 (2009) 1788e1798. [1] M. Gra [2] M. Urbani, M. Gr€ atzel, M.K. Nazeeruddin, T. Torres, Chem. Rev. 114 (2014) 12330e12396. [3] K. Zhang, M. Savage, X. Li, Y. Jiang, M. Ishida, K. Mitsuno, S. Karasawa, T. Kato, W. Zhu, S. Yang, H. Furuta, Y. Xie, Chem. Commun. 52 (2016) 5148e5151. [4] Y. Xie, W. Wu, H. Zhu, J. Liu, W. Zhang, H. Tian, W.-H. Zhu, Chem. Sci. 7 (2016) 544e549. [5] M. Liang, M. Lu, Q.-L. Wang, W.-Y. Chen, H.-Y. Han, Z. Sun, S. Xue, J. Power Sources 196 (2011) 1657e1664. [6] F. Li, Y. Chen, X. Zong, W. Qiao, H. Fan, M. Liang, S. Xue, J. Power Sources 332 (2016) 345e354. [7] K.R. Justin Thomas, A. Venkateswararao, C.-P. Lee, K.-C. Ho, Dyes Pigm 123 (2015) 154e165. [8] S. Fan, K. Lv, H. Sun, G. Zhou, Z.-S. Wang, J. Power Sources 279 (2015) 36e47. [9] Y. Ding, Y. Tang, W. Zhu, Y. Xie, Chem. Soc. Rev. 44 (2015) 1101e1112. [10] A. Baheti, K.R. Justin Thomas, C.-T. Li, C.-P. Lee, K.-C. Ho, ACS Appl. Mater. Interfaces 7 (2015) 2249e2262. [11] M. Liang, J. Chen, Chem. Soc. Rev. 42 (2013) 3453e3488. [12] Q. Tang, H. Zhang, Y. Meng, B. He, L. Yu, Angew. Chem. Int. Ed. 127 (2015) 11610e11614. [13] F. Hao, P. Dong, Q. Luo, J.B. Li, J. Lou, H. Lin, Energy Environ. Sci. 6 (2013) 2003e2019. [14] S.N. Yun, A. Hagfeldt, T.L. Ma, Adv. Mater 26 (2014) 6210e6237. [15] Z.W. Zheng, J. Chen, Y. Hu, W.J. Wu, J.L. Hua, H. Tian, J. Mater. Chem. C 2 (2014) 8497e8500. [16] G.R. Li, J. Song, G.L. Pan, X.P. Gao, Energy Environ. Sci. 4 (2011) 1680e1683. [17] M. Wu, X. Lin, A. Hagfeldt, T. Ma, Angew. Chem. Int. Ed. 50 (2011) 3520e3524. [18] M. Wu, J. Bai, Y. Wang, A. Wang, X. Lin, L. Wang, Y. Shen, Z. Wang, A. Hagfeldt, T. Ma, J. Mater. Chem. 22 (2012) 11121e11127. [19] E. Palomares, J.N. Clifford, S.A. Haque, T. Lutz, J.R. Durrant, J. Am. Chem. Soc. 125 (2003) 475e482. [20] X.J. Zheng, J. Deng, N. Wang, D.H. Deng, W.H. Zhang, X.H. Bao, C. Li, Angew. Chem. Int. Ed. 53 (2014) 7023e7027. [21] R. Leary, A. Westwood, Carbon 49 (2011) 741e772. [22] B. Anothumakkool, O. Game, S.N. Bhange, T. Kumari, S.B. Ogale, S. Kurungot, Nanoscale 6 (2014) 10332e10339. [23] M.S. Su'ait, M.Y.A. Rahman, A. Ahmad, Sol. Energy 115 (2015) 452e470.
41
[24] S. Yun, J.N. Freitas, A.F. Nogueira, Y. Wang, S. Ahmad, Z.-S. Wang, Prog. Polym. Sci. 59 (2016) 1e40. [25] X. Zhang, T.Z. Jing, S.Q. Guo, G.D. Gao, L. Liu, Rsc. Adv. 4 (2014) 50312e50317. [26] Z.Q. Li, F. Gong, G. Zhou, Z.S. Wang, J. Phys. Chem. C 117 (2013) 6561e6566. [27] M. Zheng, J.H. Huo, Y.G. Tu, J.H. Wu, L.H. Hu, S.Y. Dai, Electrochim. Acta 173 (2015) 252e259. [28] F. Gong, H. Wang, X. Xu, G. Zhou, Z.-S. Wang, J. Am. Chem. Soc. 134 (2012) 10953e10958. [29] Z. Jin, M. Zhang, M. Wang, C. Feng, Z.-S. Wang, Acc. Chem. Res. 50 (2017) 895e904. [30] Y. Duan, Q. Tang, J. Liu, B. He, L. Yu, Angew. Chem. Int. Ed. 53 (2014) 14569e14574. [31] Z. Hu, K. Xia, J. Zhang, Z. Hu, Y. Zhu, RSC Adv. 4 (2014) 42917e42923. [32] W. Zhao, X.L. Zhu, H. Bi, H.L. Cui, S.R. Sun, F.Q. Huang, J. Power Sources 242 (2013) 28e32. [33] Y. Hu, Z. Zheng, H. Jia, Y. Tang, L. Zhang, J. Phys. Chem. C 112 (2008) 13037e13042. [34] Y.C. Wang, D.Y. Wang, Y.T. Jiang, H.A. Chen, C.C. Chen, K.C. Ho, H.L. Chou, C.W. Chen, Angew. Chem. Int. Ed. 52 (2013) 6694e6698. [35] J.-C. Tsai, M.-H. Hon, I.-C. Leu, RSC Adv. 5 (2015) 4328e4333. [36] C.-W. Kung, H.-W. Chen, C.-Y. Lin, K.-C. Huang, R. Vittal, K.-C. Ho, ACS Nano 6 (2012) 7016e7025. [37] J. Zhang, S. Najmaei, H. Lin, J. Lou, Nanoscale 6 (2014) 5279e5283. [38] S. Jiang, X. Yin, J.T. Zhang, X.Y. Zhu, J.Y. Li, M. He, Nanoscale 7 (2015) 10459e10464. [39] D.D. Song, M.C. Li, Y.J. Jiang, Z. Chen, F. Bai, Y.F. Li, B. Jiang, J. Photochem. Photobiol. A 279 (2014) 47e51. [40] J. Ma, W. Shen, F. Yu, J. Power Sources 351 (2017) 58e66. [41] S. Huang, Q. He, W. Chen, J. Zai, Q. Qiao, X. Qian, Nano Energy 15 (2015) 205e215. [42] X. Cui, Z. Xie, Y. Wang, Nanoscale 8 (2016) 11984e11992. [43] Z.Y. Tang, Q.W. Tang, J.H. Wu, Y. Li, Q. Liu, M. Zheng, Y.M. Xiao, G.T. Yue, M.L. Huang, J.M. Lin, RSC Adv 2 (2012) 5034e5037. [44] X.B. Xu, D.K. Huang, K. Cao, M.K. Wang, S.M. Zakeeruddin, M. Gr€ atzel, Sci. Rep. 3 (2013) 4. [45] Z.Y. Zhang, S.P. Pang, H.X. Xu, Z.Z. Yang, X.Y. Zhang, Z.H. Liu, X.G. Wang, X.H. Zhou, S.M. Dong, X. Chen, L. Gu, G.L. Cui, RSC Adv. 3 (2013) 16528e16533. [46] X.Y. Zhang, S.P. Pang, X. Chen, K.J. Zhang, Z.H. Liu, X.H. Zhou, G.L. Cui, RSC Adv. 3 (2013) 9005e9010. [47] Z. Wei, J. Wang, S. Mao, D. Su, H. Jin, Y. Wang, F. Xu, H. Li, Y. Wang, ACS Catal. 5 (2015) 4783e4789. [48] X.Y. Yu, Y. Feng, Y. Jeon, B. Guan, X.W.D. Lou, U. Paik, Adv. Mater 28 (2016) 9006e9011. [49] H. Guo, T. Li, W. Chen, L. Liu, X. Yang, Y. Wang, Y. Guo, Nanoscale (2014) 15168e15174. [50] J. Lehto, S. Haukka, R. Harjula, M. Blomberg, J. Chem. Soc. Dalton Trans. (1990) 1007e1011. [51] B. Yuan, W. Luan, S.-t. Tu, Dalton Trans. 41 (2012) 772e776. [52] C.E.M. Campos, J.C. de Lima, T.A. Grandi, K.D. Machado, P.S. Pizani, Phys. B 324 (2002) 409e418. [53] A. Shetty, A. Saha, M. Makkar, R. Viswanatha, Phys. Chem. Chem. Phys. 18 (2016) 25887e25892. [54] Y. Li, J. Zhao, Y. Dan, D. Ma, Y. Zhao, S. Hou, H. Lin, Z. Wang, Chem. Eng. J. 166 (2011) 428e434. [55] X. Miao, K. Pan, G. Wang, Y. Liao, L. Wang, W. Zhou, B. Jiang, Q. Pan, G. Tian, Chem. Eur. J. 20 (2014) 474e482. [56] N. Berry, M. Cheng, C.L. Perkins, M. Limpinsel, J.C. Hemminger, M. Law, Adv. Energy Mater 2 (2012) 1124e1135. [57] A. Padhi, K. Nanjundaswamy, C. Masquelier, S. Okada, J. Goodenough, J. Electrochem. Soc. 144 (1997) 1609e1613. [58] R. Ma, K. Takada, K. Fukuda, N. Iyi, Y. Bando, T. Sasaki, Angew. Chem. Int. Ed. 47 (2008) 86e89. [59] T. Liu, A.M. Asiri, X. Sun, Nanoscale 8 (2016) 3911e3915. [60] J. Wang, N. Yang, H. Tang, Z. Dong, Q. Jin, M. Yang, D. Kisailus, H. Zhao, Z. Tang, D. Wang, Angew. Chem. Int. Ed. 52 (2013) 6417e6420. €tzel, J.H. Delcamp, ACS Appl. [61] N.P. Liyanage, A. Yella, M. Nazeeruddin, M. Gra Mater. Interfaces 8 (2016) 5376e5384. [62] B. Lei, G. Li, X. Gao, J. Mater. Chem. A 2 (2014) 3919e3925.