Cost-effective bifacial dye-sensitized solar cells with transparent iron selenide counter electrodes. An avenue of enhancing rear-side electricity generation capability

Accepted Manuscript Cost−effective bifacial dye−sensitized solar cells with transparent iron selenide counter electrodes. An avenue of enhancing rear−side electricity generation capability Juan Liu, Qunwei Tang, Benlin He, Liangmin Yu PII:

S0378-7753(14)01771-6

DOI:

10.1016/j.jpowsour.2014.10.152

Reference:

POWER 20072

To appear in:

Journal of Power Sources

Received Date: 16 September 2014 Revised Date:

8 October 2014

Accepted Date: 22 October 2014

Please cite this article as: J. Liu, Q. Tang, B. He, L. Yu, Cost−effective bifacial dye−sensitized solar cells with transparent iron selenide counter electrodes. An avenue of enhancing rear−side electricity generation capability, Journal of Power Sources (2014), doi: 10.1016/j.jpowsour.2014.10.152. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

SC

RI PT

ACCEPTED MANUSCRIPT

M AN U

Transparent and cost−effective iron selenide alloy counter electrodes were employed to fabricate bifacial dye−sensitized solar cells with front and rear efficiencies of

AC C

EP

TE D

7.64% and 4.95%, respectively

ACCEPTED MANUSCRIPT

Cost−effective bifacial dye−sensitized solar cells with transparent iron selenide counter electrodes. An avenue of enhancing rear−side

RI PT

electricity generation capability

Juan Liu,a,b Qunwei Tang,*a,b Benlin Heb, Liangmin Yu*a,c

Key Laboratory of Marine Chemistry Theory and Technology, Ministry of Education, Ocean

University of China, Qingdao 266100, P.R. China;

Institute of Materials Science and Engineering, Ocean University of China, Qingdao 266100, P.R.

China; c

M AN U

b

SC

a

Qingdao Colobrative Innovation Center of Marine Science and Technology, Ocean University of

China, Qingdao 266100, P.R. China;

TE D

E-mail: [email protected]; [email protected]; Tel/Fax: 86 532 66781690;

Abstract: Alloy materials have established themselves as alternative electrocatalysts for

EP

electrochemical devices because of their cost−effectiveness, high conductivity, good electrocatalytic activity, and reasonable stability. Aiming at reducing fabrication cost without sacrificing power

AC C

conversion efficiency of dye−sensitized solar cells (DSSCs), we report the feasibility of designing transparent and cost−effective Fe−Se alloy counter electrodes for bifacial DSSCs. Due to the rapid charge transfer ability and electrocatalytic activity, maximum front and rear efficiencies of 7.64% and 4.95% are measured for the DSSC with FeSe alloy electrode in comparison with 6.97% and 3.56% from Pt−based solar cell. The impressive results along with simple synthesis highlight the potential application of Fe−Se alloys in robust bifacial DSSCs. Keywords: Bifacial dye−sensitized solar cell; Iron selenide alloy; Transparent counter electrode; 1

ACCEPTED MANUSCRIPT

Mild solution synthesis; Rear−side electricity generation

1. Introduction

RI PT

Dye−sensitized solar cells (DSSCs), directly converting solar energy to electricity [1−6], are promising renewable photovoltaic devices with broad potential to solve future energy problems because of their easy fabrication, high power conversion efficiency in theory, and

SC

low−environmental impacts compared with silicon solar cells. So far, the best conversion efficiency of 12.3% has been measured by Grätzel with porphyrin−based dye together with

M AN U

Co(II/III)tris(bipyridyl) tetracyanoborate complex redox couples and platinum counter electrode (Pt CE) [7]. Upon optical excitation, the dye molecule absorbs a photon and injects an electron into conduction band (CB) of titanium dioxide (TiO2) nanocrystallite, and is subsequently recovered to its ground state by a redox couple. In this fashion, the irradiation of photosensitive dyes is crucial

TE D

for electron excitation and therefore light−to−electric conversion efficiency. However, the gradient descent in light intensity within a dye−sensitized TiO2 anode has been found a limit for dye

EP

illumination and electron injection [8]. That is not surprising to determine a low conversion efficiency from front−side (anode) irradiation. Viewed from this point, the enhancement in light

AC C

harvesting has been one of the promising avenues of elevating electricity generation. Many efforts on light harvesting have been made on setting blocking layer [9], reflecting layer [10], or adding light harvester [11], resulting in tedious procedures without reducing expenses. Recently, an emerging concept of either front or rear irradiation has been aroused to help bring down the cost of solar energy conversion [12−15]. Such bifacial DSSC collects photons from either side of device, accelerating its practical application [15]. To realize this technique of collecting sunlight from either side of a solar cell, transparent CEs are crucial in designing such bifacial 2

ACCEPTED MANUSCRIPT

DSSCs. Graphene and PANi, colored and semitransparent electrocatalyst, have been deposited on a conductive substrate for CE applications by Zhao et al in bifacial DSSC assembly [16], yielding front and rear efficiencies of 6.54% and 4.26%, respectively. However, the modest long−term

RI PT

stabilities and catalytic activities of carbonaceous materials and conductive polymers are limits for their commercial applications in DSSCs [17]. Although traditional Pt and its composites exhibit satisfactory stabilities and catalytic activities, the strong reflection to light from their metallic luster

SC

leads to a gigantic optical loss. Therefore, it is a prerequisite to explore transparent but cost−effective CE materials with excellent electrocatalysis and stability for robust bifacial DSSCs.

M AN U

By addressing these characteristics, we have successfully synthesized transparent and cost−effective Fe−Se binary alloy CEs by a mild solution strategy. The system of interest is the utilization of transparent alloy CEs, yielding maximum front and rear efficiencies of 7.64% and 4.95% under air mass 1.5 (AM1.5) solar light. To the best of our knowledge, this is so far no reports on bifacial

TE D

DSSCs with Fe−Se alloy CEs.

EP

2. Experimental section

2.1 Preparation of binary Fe−Se alloy CEs

AC C

The feasibility of this strategy was confirmed by following experimental procedures: A mixing aqueous solution consisting of 0.01 g of Se powders and 0.1 M FeCl3 (0.71 ml for Fe0.6Se, 0.94 ml for Fe0.8Se, 1.18 ml for FeSe, 1.41 ml for Fe1.2Se, and 1.64 ml for Fe1.4Se,) was made at 60 oC. The total volume of the solution was adjusted to 27.5 ml by deionized water. 2 ml of hydrazine hydrate (85 wt%) was dropped into the above solution, after vigorous agitating for 10 min, the reactant was transferred into a Teflon−lined autoclave and cleaned FTO glass substrate (sheet resistance 12 Ω sq−1, purchased from Hartford Glass Co., USA) with FTO layer downward was immersed in. After 3

ACCEPTED MANUSCRIPT

the reaction at 120 °C for 12 h, the FTO substrate was rinsed by deionized water and vacuumly dried at 50 oC. As references, pristine Fe or Se electrode was prepared according to above procedures by

RI PT

reducing 27.5 ml of FeCl3 (0.1 M) or 0.01 g of Se powders by 2 ml of hydrazine hydrate (85 wt%). 2.2 Fabrication of dye−sensitized TiO2 anodes

The TiO2 anodes were fabricated by coating TiO2 colloid synthesized by a sol−hydrothermal

SC

method onto cleaned FTO glass substrates [18]. The size of the resultant anodes was controlled at

furnace at 450 °C for 30 min.

M AN U

0.5 × 0.5 cm2 with an average thickness of 10 µm. The air−dried colloids were calcined in a muffle

The resultant TiO2 film was sensitized by immersing into a 0.50 mM ethanol solution of N719 dye (purchased from DYESOL LTD) for 24 h. A DSSC device was fabricated by sandwiching redox electrolyte between a dye−sensitized TiO2 anode and a CE. A redox electrolyte consisted of 100

TE D

mM of tetraethylammonium iodide, 100 mM of tetramethylammonium iodide, 100 mM of tetrabutylammonium iodide, 100 mM of NaI, 100 mM of KI, 100 mM of LiI, 50 mM of I2, and 500

EP

mM of 4−tert−butyl−pyridine in 50 ml acetonitrile. 2.3 Photovoltaic measurements

AC C

The photocurrent−voltage (J−V) curves of the assembled DSSCs with an active area of 0.25 cm2 were recorded on an Electrochemical Workstation (CHI600E) under irradiation of a simulated solar light (Xe Lamp Oriel Sol3A™ Class AAA Solar Simulators 94023A, USA) at a light intensity of 100 mW cm-2 (calibrated by a standard silicon solar cell) in ambient atmosphere. Each DSSC device was measured at least five times to eliminate experimental error and a compromise J−V curve was employed. A black mask with an aperture area of around 0.25 cm2 was applied on the surface of DSSCs to avoid stray light completely. 4

ACCEPTED MANUSCRIPT

2.4 Electrochemical characterizations The electrochemical performances were recorded on a conventional CHI660E setup comprising an Ag/AgCl reference electrode, a CE of platinum sheet, and a working electrode of FTO glass

RI PT

supported binary Fe−Se alloy. The cyclic voltammetry (CV) curves were recorded in a supporting electrolyte consisting of 50 mM M LiI, 10 mM I2, and 500 mM LiClO4 in acetonitrile. Electrochemical impedance spectroscopy (EIS) measurements were carried out in a frequency range

SC

of 0.01 Hz ~ 106 kHz and at an ac amplitude of 10 mV. Tafel polarization curves were recorded by

2.5 Other characterizations

M AN U

assembling symmetric cell consisting of Fe−Se alloy CE|redox electrolyte|Fe−Se alloy CE.

The compositions of the binary Fe−Se alloy CEs were detected by inductively coupled plasma-atomic emission spectra (ICP−AES). The morphology of the FeSe alloy CE was observed with a scanning electron microscope (SEM, S4800) and on a transmission electron microscopy

TE D

(TEM, JEM2010, JEOL). X−ray diffraction (XRD) profiles of the resultant alloys were recorded on an X−ray powder diffractometer (X’pert MPD Pro, Philips, Netherlands) with Cu Kα radiation (λ =

EP

1.5418 Å) in the 2θ range from 10 to 90° operating at 40 kV accelerating voltage and 40 mA current). XPS experiment was carried out on a RBD upgraded PHI-5000C ESCA system (Perkin

AC C

Elmer) with Mg Kα radiation (hν = 1253.6 eV). The optical transmission spectra of binary Fe−Se alloy CEs were recorded on a UV–vis spectrophotometer at room temperature. Incident photo−to−current conversion efficiency (IPCE) curves were obtained at the short−circuit condition on an IPCE measurement systems (MS260).

3. Results and discussion Figure 1 here 5

ACCEPTED MANUSCRIPT

The resultant Fe−Se alloys synthesized by the mild solution approach are determined by ICP−AES equipment, indicating atomic rations of 0.611: 1.000, 0.792: 1.000, 0.993: 1.000, 1.205: 1.000, and 1.387: 1.000 for Fe0.6Se, Fe0.8Se, FeSe, Fe1.2Se, and Fe1.4Se, respectively. There is a fact

RI PT

that the measured atomic ratios are close to the stoichiometries of Fe−Se alloys, therefore, the chemical formulae of the binary alloy CEs can be expressed according to their stoichiometric ratios. Top−view SEM images in Fig. 1a and Fig. 1b demonstrate a high surface coverage of porous FeSe

SC

nanostructures. The loose structure in deep examination provides large alloy/electrolyte interface and enormous channels for I3− reduction reaction. Lattice fringes are clearly observed in HRTEM

M AN U

image of FeSe alloy, as shown in Supplementary Fig. S1a, indicating the resultant alloy has a good crystallinity. Moreover, lots of lattice defects are determined, which also provide active sites for I3− adsorption and reduction. As shown in Fig. 1c, XRD results indicate that the alloying of Fe and Se can form Fe−Se alloys (PDF#74−0247) [19]. No diffraction peaks attributing to pure Se or pure Fe

TE D

are detected, demonstrating that Se and Fe have completely alloyed into Fe−Se alloys. The XPS spectrum of FeSe alloy in Supplementary Fig. S1b reveals that the Fe2p and Se3d are centered at

EP

707.1 and 55.3 eV, respectively. The binding energies are attributed to the features of FeSe alloy phase. Fe is a typical transition metal, having unfilled valence in d orbital. Therefore, the alloying of

AC C

Fe with Se can accept electrons to form coordinated intermediates, which is a prerequisite for robust CE materials.

Figure 2 here Table 1 here

Fig. 2a displays the optical transmittance spectra of various Fe−Se alloys along with Se, Fe, and Pt using FTO glass substrate as a baseline. Most of the Fe−Se alloys have high optical transparency (> 80%) in visible−light region, indicating that the incident light can penetrate Fe−Se alloy CEs for 6

ACCEPTED MANUSCRIPT

N719 dye excitation. Therefore, the CEs having high transparency in visible−light region are indispensable in designing bifacial DSSC devices. However, standard Pt electrode has a transparency of ~20% because Pt film has metallic luster and therefore the incident light can be

RI PT

reflected by luminous surfaces. The current density−voltage (J−V) characteristic of the cells under irradiation of AM1.5 solar irradiation from front side are shown in Fig. 2b and the photovoltaic parameters are summarized in Table 1. The device employing FeSe binary CE yields an optimal

SC

front η of 7.64% (Jsc = 17.15 mA cm−2, Voc = 0.738, FF = 60.4%). For references, the DSSCs from Se−only, Fe−only, and Pt−only CEs are also measured under the same conditions, giving a η of

M AN U

5.75%, 2.74%, and 6.97%, respectively. Fig. 2c displays the J−V curves of the DSSCs corresponding to rear irradiation, a maximum rear η is 4.95% due to the high optical transmittance for visible light. Statistical data on a larger batch of five bifacial DSSCs is shown in Supplementary Fig. S3. Although the compromising η of 7.64% ± 5% and 4.95% ± 13% have not been certified yet,

TE D

the small deviation and high reproducibility demonstrate that bifacial DSSCs with excellent photovoltaic performances can be realized using the method reported here. Notably, the Jsc and Voc

EP

values from rear irradiation are all lower than that extracted from front−side irradiation. As shown in Supplementary Fig. S2, gradiently decreased light intensity within TiO2 film leads to a gradient

AC C

increase in electron distribution (viewed from TiO2 anode to CE) [7]. In comparison with rear irradiation, the electron transfer suffers shorter diffusion lengths. Under full sunlight, an average injected electron may experience a million trapping events before either percolating to the collecting electrode or recombining with I3− species [20]. Figure 3 here To compare the catalytic activities, CV curves of various CEs toward liquid electrolyte containing I−/I3− redox couples are shown in Fig. 3a. The peak positions of the Fe−Se alloy CEs are 7

ACCEPTED MANUSCRIPT

very similar to that of Pt electrode [21], showing that Fe−Se alloys have a similar electrocatalytic effect to the Pt electrode. Considering that the function of a CE is to reduce I3− into I−, therefore peak reaction for Red1 can be employed to elevate the electrocatalytic activity of Fe−Se alloy CEs

RI PT

[22]. Judged by higher Red1 peak current density and nearly the same peak position, FeSe alloy has a superior catalytic activity. Moreover, the diffusion coefficient (Dn), a parameter for I3− diffusion kinetics, can be obtained by Randles-Sevcik theory and summarized in Supplementary Table S1

SC

[23]: 4.20 ×10−7 cm−2 s−1 for FeSe > 3.76 × 10−7 cm−2 s−1 for Fe0.8Se > 3.46 ×10−7 cm−2 s−1 for Fe1.2Se > 2.79 ×10−7 cm−2 s−1 for Fe1.4Se > 1.81 ×10−7 cm−2 s−1 for Fe0.6Se > 5.21 ×10−8 cm−2 s−1 for

M AN U

Se > 8.36 ×10−10 cm−2 s−1 for Fe. From the stacking CV curves of FeSe alloy CE at different scan rates, linear relationship between peak current densities and square root of scan rate as well as multiturn scan, as shown in Supplementary Fig. S4, one can make a conclusion that the redox reaction at FeSe surface is controlled by ionic diffusion in the electrolyte and is relatively stable

TE D

[24]. To further validate the catalytic activities of various CEs, EIS experiments are carried out using symmetric dummy cells by two identical CEs. Nyquist plots in Fig. 3b illustrate

EP

electrochemical impedence characteristics of Fe−Se alloy, Se, and Fe CEs, in which a semicircle is observed. A Randles−type circuit is obtained by fitting EIS spectra using a Z−view software, and

AC C

there is a good agreement between measured and fitted curves. The Rct values of all Fe−Se alloy CE are smaller than that of pure metal electrodes, indicating a promoting effect on charge−transfer ability by alloying Fe and Se. The charge−transfer ability of the CEs follows an order of FeSe > Fe0.8Se > Fe1.2Se > Fe1.4Se > Fe0.6Se > Se > Fe, which is in an agreement with the catalytic activity derived from CV analysis and electron lifetime on CE/electrolyte interface (Supplementary Fig. S5). Due to a fact of catalyzing I3− to I− (I3− + 2e = 3I−) by a CE, the average lifetime of electron in CE can be expressed as the reaction kinetics of Red1. The shortest electron lifetime of 24.1 µs is 8

ACCEPTED MANUSCRIPT

calculated from the symmetric dummy cells using FeSe electrode, indicating a fast catalytic kinetics. Tafel polarization curves in Fig. 3c were also confirmed with the symmetric dummy cells similar to those in EIS measurements to verify the electrocatalytic activities and charge transfer ability of

RI PT

CEs. The larger slope for the anodic or cathodic branch in Tafel polarization curves indicates a higher exchange current density (J0) on CE and better catalytic activity toward I3− reduction [25]. Apparently, the J0 follows an order of FeSe > Fe0.8Se > Fe1.2Se > Fe1.4Se > Fe0.6Se > Se > Fe, which

SC

matches the order of Rct because J0 is inversely proportional to Rct (J0 = RT/nFRct). Additionally, limiting diffusion current density (Jlim), obtained from the current density at low slope and high

M AN U

potential regions, depends on the diffusion coefficient (Dn) of I−/I3− couples in electrolyte (Jlim = 2nFCDn/l) [26]. The Dn of redox species is in linear with Jlim, indicating a good matching of Dn sequence to results from CV characterization (Jred1 = Kn1.5ACDn0.5v0.5). Fig. 3d presents the photon−to−current conversion efficiency (IPCE) of DSSC devices with various CEs. The broad

TE D

IPCE curves, covering the spectrum from 400 to 800 nm, exhibit a maximum IPCE value of ~80% for the cell with FeSe electrode and similar shape to the cell with Pt electrode. Moreover, the

EP

measured IPCE follows an order of FeSe ≈ Pt > Fe1.2Se > Se > Fe0.8Se > Fe > Fe1.4Se > Fe0.6Se, which is the same to front Jsc. The results from IPCE measurement reveal the number of electrons in

AC C

the external circuit produced is the largest using FeSe electrode [27,28], which is in consistent with the corresponding values of Jsc in Table 1. Figure 4 here

A fast start−up and long−term stability in engine, vehicle and power source applications have been two challenges for nanoenergy devices [29]. Therefore, the photocurrent dynamics of the bifacial DSSC was probed in order to determine its start−up, electron recombination, multiple start capability, and photocurrent stability. Fig. 4a shows the on−off effect in a time slot of 0 ~ 250 s. A 9

ACCEPTED MANUSCRIPT

sharp increase and no delay in photocurrent density suggest a fast start of cell operation at irradiation (front, rear, or both). As a reference, the on−off switch of cell device with standard Pt electrode is also provided, as shown in Fig. 4c, giving a delay in reaching maximum current density.

RI PT

The result demonstrates that FeSe alloy exhibits a higher catalytic activity than that of Pt electrode. For the quick start−up, fast kinetics at dye−sensitized TiO2 anode and Fe−Se alloy CE are crucial to promote the cell launch. Moreover, no obvious reduction in photocurrent density in each “on” state

SC

means a good multiple start capability and low electron recombination kinetics, which is in an agreement with Supplementary Fig. S7a). However, the decreased current density for Pt based

M AN U

DSSC in “on” state refers to a diffusion mechansim for I−/I3− species. The photocurrent density versus time plots over 1000 s display the photocurrent stability on prolonged exposure to light irradiation (100 mW cm−2). As shown in Fig. 4b, Fig. 4d, and Supplementary Fig. S8, the photocurrent densities decrease by 19.2% and 29.9% for the cells with FeSe and Pt electrodes,

TE D

respectively. Although 10 h−test is far from evaluating the long−term stability of a real cell deivce,

applications [30].

AC C

4. Conclusions

EP

the presented results demonstrate the superiority of FeSe alloy CE than Pt electrode for DSSC

In summary, we have demonstrated the feasibility of bifacial DSSC device comprising of a transparent Fe−Se binary alloy CE, a dye−sensitized TiO2 anode, and a liquid electrolyte containing I−/I3− redox couples. Due to the high optical transparency, charge−transfer ability, and catalytic activity of FeSe binary CE, promising power conversion efficiencies of 7.64% and 4.95% with a high reproducibility are measured in the bifacial DSSC device for front and rear irradiations, respectively. The merits on fast start−up, low electron recombination with I3− species, high multiple 10

ACCEPTED MANUSCRIPT

start capability, and good photocurrent stability highlight the potential applications in engine, vehicle and power sources. The research presented here is far from being optimized but these profound advantages in photovoltaic performances along with cost−effectiveness, mild solution

RI PT

synthesis, and scalable materials promise the transparent metal selenides competitive in well−established DSSCs.

SC

Acknowledgments

The authors gratefully acknowledge Fundamental Research Funds for the Central Universities

M AN U

(201313001, 201312005), Shandong Province Outstanding Youth Scientist Foundation Plan (BS2013CL015), Shandong Provincial Natural Science Foundation (ZR2011BQ017), Research Project for the Application Foundation in Qingdao (13−1−4−198−jch), National Natural Science Foundation of China (51102219, 51342008), National High Technology Research and

TE D

Development Program of China (2010AA09Z203, 2010AA065104), and National Key Technology

EP

Support Program (2012BAB15B02).

Supporting information

AC C

Electronic Supplementary Information (ESI) available: EIS data, HRTEM image, XPS spectrum, scheme figure, repeated J−V curves, CV curves, Bode EIS plots, dark J−V curves, EIS plots of cell devices, and Bode EIS plots of DSSCs.

References [1]

B. O’Regan, M. Grätzel, Nature 353 (1991) 737−740.

[2]

M. Grätzel, Nature 414 (2001) 338−344. 11

ACCEPTED MANUSCRIPT

[3]

Y. Peng, J. Zhong, K. Wang, B. Xue, Y. B. Cheng, Nano Energy 2 (2013) 235−240.

[4]

H. J. Zhang, L. T. Yang, Z. Liu, M. Ge, Z. Zhou, W. Chen, Q. Z. Li and L. Liu, J. Mater. Chem. 22 (2012) 18572−18577. Z. Y. Tang, J. H. Wu, M. Zheng, J. Huo, Z. Lan, Nano Energy 2 (2013) 622−627.

[6]

X. X. Chen, Q. W. Tang, B. L. He, L. Lin, L. M. Yu, Angew. Chem. Int. Ed. 53 (2014)

RI PT

[5]

10799−10803.

A. Yella, H. W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran, M. K. Nazeeruddin, E. W. Diau, C.

SC

[7]

Y. Yeh, S. M. Zakeeruddin, M. Grätzel, Science 334 (2011) 629−634.

S. S. Yuan, Q. W. Tang, B. L. He, L. Men, H. Y. Chen, Electrochim. Acta 125 (2014)

M AN U

[8]

646−651. [9]

M. H. Kim, Y. U. Kwon, J. Phys. Chem. C 115 (2011) 23120−23125.

[10]

Q. F. Zhang, D. Myers, J. Lan, S. A. Jenekhe, G. Z. Cao, Phys. Chem. Chem. Phys. 14

TE D

(2012) 14982−14998.

B. Kumar, S. W. Kim, Nano Energy 1 (2012) 342−355.

[12]

S. Ito, S. M. Zakeeruddin, P. Comte, P. Liska, D. B. Kuang, M. Grätzel, Nat. Photonics 2

EP

[11]

(2008) 693−698.

Q. D. Tai, B. Chen, F. Guo, S. Xu, H. Hu, B. Sebo, X. Z. Zhao, ACS Nano 5 (2011)

AC C

[13]

3795−3799.

[14] [15]

Y. Y. Duan, Q. W. Tang, B. L. He, R. Li, L. M. Yu, Nanoscale, 2014, doi: 10.103900A. G. E. Jabbour, D. Doderer, Nat. Photonics 4 (2010) 604−605.

[16]

Q. D. Tai, X. Z. Zhao, J. Mater. Chem. A 2 (2014) 13207−13218.

[17]

F. Gong, H. Wang, X. Xu, G. Zhou, Z. S. Wang, J. Am. Chem. Soc. 134 (2012) 10953−10958. 12

ACCEPTED MANUSCRIPT

[18]

B. B. Hu, Q. W. Tang, B. L. He, L. Lin, H. Y. Chen, J. Power Sources 267 (2014) 445−451.

[19]

W. J. Wang, X. Pan, W. Q. Liu, B. Zhang, H. W. Chen, X. Q. Fang, J. X. Yao, S. Y. Dai,

[20]

RI PT

Chem. Commun. 50 (2014) 2618−2620.

M. Law, L. E. Greene, J. C. Johnson, R. Saykally, P. D. Yang, Nat. Mater. 4 (2005) 455−459.

L. L. Li, C. W. Chang, H. H. Wu, J. W. Shiu, P. T. Wu and W. G. Diau, J. Mater. Chem. 22

SC

[21]

(2012) 6267−6273.

B. L. He, X. Meng, Q. W. Tang, ACS Appl. Mater. Interfaces 6 (2014) 4812−4818.

[23]

T. Daeneke, A. J. Mozer, T. H. Kwon, N. W. Duffy, A. B. Holmes, U. Bach, L. Spiccia,

M AN U

[22]

Energy Environ. Sci. 5 (2012) 7090−7099. [24]

Z. B. Lv, J. F. Yu, H. W. Wu, J. Shang, D. Wang, S. C. Hou, Y. P. Fu, K. Wu, D. C. Zou,

[25]

TE D

Nanoscale 4 (2012) 1248−1253.

M. Wu, X. Lin, Y. Wang, L. Wang, W. Guo, D. Qi, X. Peng, A. Hagfeldt, M. Grätzel, J. Am.

[26]

EP

Chem. Soc. 134 (2012) 3419−3428.

M. K. Wang, A. M. Anghel, B. Marsan, N. L. C. Ha, N. Pootrakulchote, S. M. Zakeeruddin,

[27]

AC C

M. Grätzel, J. Am. Chem. Soc. 131 (2009) 15976−15977. M. H. Yeh, L. Y. Lin, C. P. Lee, C. Y. Chou, K. W. Tsai, J. T. Lin, K. C. Ho, J. Power Sources 237 (2013) 141−148.

[28]

J. R. Mann, M. K. Gannon, T. C. Fitzgibbons, M. R. Detty, D. F. Watson, J. Phys. Chem. C 112 (2008) 13057−13061.

[29]

C. Laberty-Robert, K. Vallé, F. Pereira, C. Sanchez, Chem. Soc. Rev. 40 (2011) 961−1005.

[30]

L. Li, X. C. Yang, J. J. Gao, H. N. Tian, J. Z. Zhao, A. Hagfeldt, L. C. Sun, J. Am. Chem. 13

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Soc. 133 (2011) 8458−8460.

14

ACCEPTED MANUSCRIPT Table and figure captions Table 1. Output photovoltaic characteristics of the DSSCs employing binary Fe−Se alloy CEs for front and rear irradiations. η: power conversion efficiency; Jsc: short−circuit current density; Voc:

RI PT

open−voltage; FF: fill factor.

Fig. 1. (a) & (b) Top−view SEM photographs of FeSe alloy CE. (c) XRD patterns of various CE materials.

SC

Fig. 2. (a) Optical transmittance spectra of various alloys and pristine Fe, Se, and Pt (FTO glass as a baseline) as well as characteristic J−V curves of the DSSCs employing various electrodes for (b)

M AN U

front and (c) rear irradiations, respectively.

Fig. 3. (a) CV curves of various CEs, and (b) Nyquist EIS spectra, and (c) Tafel polarization curves of symmetric dummy cells from double CEs. (d) IPCE spectral action responses of the DSSCs from

50 mV s−1.

TE D

various CEs. The inset displays an equivalent circuit. The CV curves are recorded at a scan rate of

Fig. 4. The on−off switches with irradiation from front, rear and both were achieved by alternately

EP

illuminating (100 mW cm−2) and darkening (0 mW cm−2) the DSSC devices from (a) FeSe and (c) Pt electrodes at an interval of 25 s and 0 V. Photocurrent stabilities of the DSSC devices with (c)

AC C

FeSe and (d) Pt electrodes under durable irradiations at 100 mW cm−2.

15

ACCEPTED MANUSCRIPT

Table 1

Fe1.2Se Fe1.4Se Se Fe

AC C

EP

TE D

Pt

Voc (V) 0.668 0.601 0.748 0.691 0.738 0.733 0.690 0.667 0.678 0.652 0.681 0.522 0.707 0.627 0.675 0.652

Jsc (mA cm−2) 11.59 9.36 12.89 5.96 17.15 8.96 14.65 6.85 12.08 7.48 14.06 8.95 12.51 9.39 17.15 9.52

16

FF (%) 54.4 49.2 59.5 59.5 60.4 75.4 56.4 54.1 59.7 55.2 60.1 56.7 31.0 36.2 60.2 59.3

RI PT

FeSe

η (%) 4.21 2.77 5.74 2.45 7.64 4.95 5.70 2.47 4.89 2.69 5.75 2.65 2.74 2.13 6.97 3.56

SC

Fe0.8Se

irradiation front rear front rear front rear front rear front rear front rear front rear front rear

M AN U

CEs Fe0.6Se

(100)

Se Intensity (a.u.)

Fe0.6Se

(101)

( 012) (201) (110)(111)

M AN U

c

SC

RI PT

ACCEPTED MANUSCRIPT

(111) (110) (012)

Fe0.8Se FeSe Fe1.2Se Fe1.4Se

(112)

Fe-Se, PDF-#74-0247

TE D

Fe

10

20

30

40

50

60

2theta (degree)

AC C

EP

Figure 1

17

70

80

90

ACCEPTED MANUSCRIPT

18

b

100

15

-2

FeSe Fe1.2Se Fe1.4Se

40

Se Fe Pt

20

12 Fe0.6Se

9

Fe0.8Se FeSe Fe1.2Se

6

Fe1.4Se

3

Se Fe Pt

0

0 400

500

600

700

800

900

0.0

1000

-2

8

6 Fe0.6Se Fe0.8Se FeSe Fe1.2Se Fe1.4Se

2

TE D

Se Fe Pt

0

0.4

M AN U

Current density (mA cm )

10

4

0.2

Voltage (V)

Wavelength (nm)

c

RI PT

Fe0.8Se

60

SC

Fe0.6Se

0.0

0.2

0.4

Voltage (V)

EP

Figure 2

AC C

Transparency (%)

80

Current density (mA cm )

a

18

0.6

0.8

0.6

0.8

ACCEPTED MANUSCRIPT

8

80

6

Fe0.8Se

4

FeSe Fe1.2Se

b

-

Ox2, 2I3 -2e=3I2 -

Ox1, 3I -2e=I3

-

Rs

60

Fe0.6Se

2

-Z'' (ohm cm )

Se Fe

0

Red2, 3I2+2e=2I3

-

-2

Fe0.8Se

40

FeSe Fe1.2Se Fe1.4Se

20 -

Red1, I3 +2e=3I

Se Fe

-

-6

0 -0.6

-0.3

0.0

0.3

0.6

0.9

1.2

0

Potential (V vs Ag/AgCl)

d

100

IPCE (%)

Jlim

Fe0.6Se Fe0.8Se

-2

FeSe Fe1.2Se

TE D

Se Fe

0.0

80

0.5

Fe0.6Se

Fe1.4Se

Fe0.8Se

Se

FeSe Fe1.2Se

Fe Pt

60

40

20

Fe1.4Se

-0.5

60 2

80

0

-1.0

40

M AN U

J0

-4

20

Z' (ohm cm )

2

-2

Log (Current density, mA cm )

c

SC

-4

W

R ct

Fe1.4Se

2

CPE

RI PT

Fe0.6Se

-2

Current density (mA cm )

a

1.0

Potential (V)

AC C

EP

Figure 3

19

0 400

500

600

Wavelength (nm)

700

800

ACCEPTED MANUSCRIPT

20

b

15

light off

-2

15

on front

10

rear 5

10

RI PT

light on

Current density (mA cm )

-2

Current density (mA cm )

a

front

5

rear

0

0 0

50

100

150

200

250

0

Time (s)

200

400

600

d

16

-2

Current density (mA cm )

M AN U

16

12

front 8

rear

0

50

100

TE D

4

0

150

800

1000

Time (s)

200

12

250

Time (s)

EP

Figure 4

AC C

-2

Current density (mA cm )

c

SC

off

20

front

8

rear

4

0 0

200

400

600

Time (s)

800

1000

1200

ACCEPTED MANUSCRIPT

● Fe−Se alloys are synthesized on FTO glass by a mild solution strategy ● Fe−Se alloys are employed as transparent CEs for bifacial DSSCs

RI PT

● The bifacial DSSC with FeSe CE shows front and rear efficiencies of 7.64%, and 4.95%, respectively

● The high optical transparency of FeSe CE is favorable to rear efficiency

SC

enhancement

AC C

EP

TE D

M AN U

● The transparent CEs are stable and robust for efficient bifacial DSSC application

ACCEPTED MANUSCRIPT

Cost−effective bifacial dye−sensitized solar cells with transparent iron selenide counter electrodes. An avenue of enhancing rear−side

RI PT

electricity generation capability

Juan Liu,a,b Qunwei Tang,*a,b Benlin Heb, Liangmin Yu*a a

Key Laboratory of Marine Chemistry Theory and Technology, Ministry of Education, Ocean

Institute of Materials Science and Engineering, Ocean University of China, Qingdao 266100, P.R.

China;

M AN U

b

SC

University of China, Qingdao 266100, P.R. China;

E-mail: [email protected]; [email protected]; Tel/Fax: 86 532 66781690;

Supporting table and figures

TE D

Table S1. Output EIS parameters of the DSSCs employing various CEs as well as diffusion coefficients from CV curves. Rs: series resistance; Rct: charge−transfer resistance; CPE:

Rs (Ω cm2) 17.76 15.62 20.23 10.54 17.74 9.81 16.03

AC C

CEs Fe0.6Se Fe0.8Se FeSe Fe1.2Se Fe1.4Se Se Fe

EP

corresponding chemical capacitance; W: Nerst diffusion resistance; Dn: diffusion coefficient. CPE (mF cm−2) 0.852 0.892 0.871 0.891 0.921 0.856 0.947

Rct (Ω cm2) 154.9 18.0 4.9 31.8 41.4 619.0 766.0

1

W (Ω cm2) 48.0 1.10 1.59 1.53 2.30 4.90 399.2

Dn ( cm−2 s−1) 1.81 ×10−7 3.76 × 10−7 4.20 ×10−7 3.46 ×10−7 2.79 ×10−7 5.21 ×10−8 8.36 ×10−10

ACCEPTED MANUSCRIPT

b

1000

SC

Se3d

Se3p

Counts

RI PT

C1s

O1s

Fe2p

FeSe alloy

800

600

400

200

0

Bonding energy (eV)

AC C

EP

TE D

M AN U

Fig. S1. (a) HRTEM image and (b) XPS spectrum of FeSe alloy.

Fig. S2. Schemes representing incident light descent and electron density on conduction band of TiO2 with irradiation from front or rear side. The incident light intensity is controlled at 100 mW cm−2 (calibrated by a standard silicon solar cell) in either side.

2

20

b

15

12

10 7.64% 7.56% 7.63% 8.01% 7.83%

5

0

9

5.05% 4.96% 4.95% 4.37% 4.31%

6

3

RI PT

-2

-2

Current density (mA cm )

a

Current density (mA cm )

ACCEPTED MANUSCRIPT

0

0.0

0.2

0.4

0.6

0.8

0.0

0.2

0.4

Voltage (V)

0.6

0.8

SC

Voltage (V)

Fig. S3. Repeated characteristic J−V curves of the DSSCs employing FeSe alloy CE for (a) front

EP

3

b

0

AC C

-3 -6

0.0

0.3

6

Oxi

3

ti o n O x id a

0

R ed uc tio n

ak at

p eak

0 .3 8

at 0 .9

0.6

0.9

V

1V

pe ak at 0. 57 V

-3 R e du

-6

cti on

peak

at -0

.0 61

V

-9

-9

-0.3

n pe d atio

-1

25, 50, 75, 100, 125, 150, 175, and 200 mV s

-0.6

9

-2

6

Current density (mA cm )

TE D

9

-2

Current density (mA cm )

a

M AN U

and (b) rear irradiations.

4

1.2

Potential (V vs Ag/AgCl)

6

8

10

Scan rate

1/2

12

14

16

-1

(mV s )

Fig. S4. (a) CV curves of FeSe CE for I−/I3− redox species at varied scan rates (from inner to outer: 25, 50, 75, 100, 125, 150, 175, and 200 mV s−1), and (b) relationship between peak current density and square root of scan rate.

3

ACCEPTED MANUSCRIPT

80

RI PT

Fe0.6Se, τ =61.9 µ s Fe0.8Se, τ =37.3 µ s FeSe, τ =24.1 µ s Fe1.2Se, τ =44.9 µ s

-Phase (degree)

60

Fe1.4Se, τ =55.2 µ s Se, τ =98.1 µ s Fe, τ =711.0 µ s

40

0 -2

-1

0

1

2

3

4

5

M AN U

Log(frequency, Hz)

SC

20

EP

4 2 0 -2 -4

-0.6

-0.3

0.0

0.3

0.6

b

4

-2

6

8

Current density (mA cm )

8

AC C

-2

Current density (mA cm )

a

TE D

Fig. S5. Bode EIS spectra of symmetric dummy cells from various CEs.

st

1 cycle th 10 cycle th 20 cycle th 30 cycle th 40 cycle th 50 cycle

0.9

Red1 Ox1

0

-4

-8

1.2

0

10

Potential (V vs Ag/AgCl)

20

30

40

50

Cycle number

Fig. S6. (a) A total of 50 CV curves of FeSe CE for I−/I3− redox species at a scan rate of 50 mV s−1, and (b) relationship between peak current density and cycle number.

4

ACCEPTED MANUSCRIPT

0

b

-1

80 Fe0.6Se Fe0.8Se

60

RI PT

FeSe Fe1.2Se

2

-Z'' (ohm cm )

-2

Current density (mA cm )

a

-2 Fe0.6Se

-3

Fe0.8Se

-4

Fe1.4Se

FeSe Fe1.2Se

CPE1

40

Se Fe

Rs

Rct1

20

Se Fe

0.0

0.2

0.4

0.6

0

0.8

Rct2

SC

0

-5

20

Voltage (V)

Fe1.4Se

CPE2

40 2 Z' (ohm cm )

W

60

80

M AN U

Fig. S7. (a) Characteristic J−V curves and (b) EIS plots of the DSSCs recorded in the dark. The

TE D

inset shows an equivalent circuit.

70

AC C

-Phase (degree)

EP

60

Fe0.6Se

50

Fe0.8Se

40

FeSe Fe1.2Se

30

Fe1.4Se Se Fe

20 10 0 -1

0

1

2

3

Log(frequency, Hz)

Fig. S8. Bode EIS spectra of the DSSCs from various CEs.

5

4

5