Semiconductor Quantum Dot Sensitized Solar Cells Based on

May 12, 2016 - The optimized ferricyanide/ferrocyanide species concentrations (0.01/0.2 M) result in solar energy conversion efficiency of 2% with VOC...
0 downloads 0 Views 1MB Size
Subscriber access provided by Karolinska Institutet, University Library

Article

Semiconductor Quantum Dot Sensitized Solar Cells Based on Ferricyanide/Ferrocyanide Redox Electrolyte Reaching an Open Circuit Photovoltage of 0.8 V Rosemarie M Evangelista, Satoshi Makuta, Shota Yonezu, John Andrews, and Yasuhiro Tachibana ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.6b03633 • Publication Date (Web): 12 May 2016 Downloaded from http://pubs.acs.org on May 17, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Semiconductor Quantum Dot Sensitized Solar Cells Based on Ferricyanide/Ferrocyanide Redox Electrolyte Reaching an Open Circuit Photovoltage of 0.8 V Rosemarie M. Evangelista,† Satoshi Makuta,† Shota Yonezu,† John Andrews† and Yasuhiro Tachibana*,†,‡,# †

School of Engineering, RMIT University, Bundoora, VIC 3083, Australia; ‡Office for

University-Industry Collaboration, Osaka University, 2-1 Yamada-oka, Suita, Osaka 565-0871, Japan; #Japan Science and Technology Agency (JST), PRESTO, 4-1-8 Honcho Kawaguchi, Saitama 332-0012, Japan

Corresponding Author *E-mail: [email protected]; [email protected]; Tel: +61 (0)3 9925 6127; Fax: +61 (0)3 9925 6139 Keywords: Semiconductor quantum dots sensitized solar cells, ferrocyanide/ferricyanide redox couple, CdS, photovoltage decay, transient photocurrent

ACS Paragon Plus Environment

1

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 31

ABSTRACT Semiconductor quantum dot sensitized solar cells (QDSSCs) have rapidly been developed, and their efficiency has recently exceeded 9 %. Their performances have mainly been achieved by focusing on improving short circuit photocurrent employing polysulfide electrolytes. However, increase of open circuit photovoltage (VOC) cannot be expected with QDSSCs based on the polysulfide electrolytes owing to their relatively negative redox potential (around -0.65 V vs. Ag/AgCl). Here, we demonstrate enhancement of the open circuit voltage by employing an alternative electrolyte, ferricyanide/ferrocyanide redox couple. The solar cell performance was optimized by investigating influence of ferricyanide and ferrocyanide concentration on their interfacial charge transfer and transport kinetics. The optimized ferricyanide/ferrocyanide species concentrations (0.01/0.2 M) result in solar energy conversion efficiency of 2 % with VOC of 0.8 V. Since the potential difference between the TiO2 conduction band edge at pH 7 and the electrolyte redox potential is about 0.79 V, although the conduction band edge shifts negatively under the negative bias application into the TiO2 electrode, the solar cell with the optimized electrolyte composition has reached nearly the theoretical maximum voltage. This study suggests a promising method to optimize an electrolyte composition for maximizing solar energy conversion efficiency.

ACS Paragon Plus Environment

2

Page 3 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

INTRODUCTION Semiconductor quantum dot sensitized solar cells (QDSSCs) have been recognized as one of the most attractive solar cells owing to QD’s unique optical characteristics. QDs have a wide tuning range of light absorption wavelengths or band gap by selecting constituting elements and their size.1 Additionally, a QD can absorb a higher number of photons per unit volume as a result of its high extinction coefficient.2-4 Over the last 25 years, the development of a metal chalcogenide QDSSC has reached an overall solar energy conversion efficiency of greater than 5 %,5-6 and to 9 %.7 However, in most cases, these improvements were achieved by focusing on increasing their short circuit photocurrent (JSC).7-8 To increase JSC, increasing a light absorption wavelength range, i.e. increase of light harvesting efficiency or employing relatively lower band gap QD, is a common target in the development. In contrast, studies focusing on increasing open circuit photovoltage (VOC) are relatively limited.9 Typically, QDSSC employs a polysulfide electrolyte, generally resulting in relatively low open circuit photovoltage (VOC) of up to 0.6 V and fill factor (FF) of up to 0.55, owing to a relatively negative polysulfide redox potential (around -0.65 V vs. Ag/AgCl, see Scheme 1).10-14 To improve VOC, it is inevitable to consider an alternative electrolyte whose redox potential is significantly more positive compared to the polysulfide redox potential. Interestingly, studies to employ an alternative electrolyte are also limited. A metal chalcogenide QDSSC working mechanism can be the same as dye sensitized solar cells (DSSCs), however the electrolyte may not necessarily function in the same way as those employed for DSSC, i.e. iodine-iodide redox couple (I3-/I-) or cobalt complex, since the QD sensitizer is composed of inorganic nanomaterials, unlike an organic sensitizer in DSSCs. An iodine-iodide redox couple is not suitable to be employed in QDSSCs, since the reaction of

ACS Paragon Plus Environment

3

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 31

iodine with QDs occurs as soon as the solar cell is assembled.15-16 Metal chalcogenide QDs are relatively stable in the presence of a polysulfide electrolyte. However, the improvement of VOC cannot be expected, if a TiO2 nanocrystalline is employed as an electron acceptor (conduction band edge: -0.62 V vs. Ag/AgCl at pH 7).15, 17 We have previously employed a ferricyanide/ferrocyanide (Fe(CN)63-/4-) redox couple to improve a VOC and FF,18 indicating relatively high VOC of 0.7 V. A Fe(CN)63-/4- redox couple is known to have a high charge transfer rate to the fluorine doped tin oxide electrode (FTO), and thus the introduction of the hole blocking layer on the surface of the FTO was a key to prevent leakage of electrons from the FTO, and to improve the solar cell performance.18 The ferricyanide/ferrocyanide redox potential is approximately 0.9 V more positive, compared to the typical polysulfide redox potential (see Scheme 1). Therefore, further improvement, in particular a higher VOC, can be expected by employing the ferricyanide/ferrocyanide electrolyte, if their interfacial electrolyte reaction mechanisms are clarified inside a solar cell, and their charge recombination processes are minimised. In this paper, we demonstrate optimisation of QDSSC performance by investigating the charge transfer and transport kinetics of the ferricyanide/ferrocyanide electrolyte. CdS QD was selected as a sensitizer, since the solar cell performance is not limited by charge transfer reaction at CdS QD/TiO2 interface.19-20 We have identified that an electron injection from CdS QD to TiO2 occurs with 1~9 ps,19 while the charge recombination between an electron in TiO2 and a hole in CdS occurs on ns~µs time scale.20 We selected a polysulfide redox couple as a control electrolyte, since it is the most common to be employed for QDSSCs.10-14 Photocurrent generation from a CdS QD sensitized TiO2 electrode was assessed by altering concentration of the oxidized or reduced species, and applied voltage.

ACS Paragon Plus Environment

4

Page 5 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

EXPERIMENTAL Chemicals. Titanium di-isopropoxide bis(acetylacetonate) (75 wt% in isopropanol), Na2S.9H2O (>98%), Cd(ClO4)2.xH2O, Zn(O2CCH3)2.2H2O (>99%), potassium hexacyanoferrate (III) (>99%), potassium hexacyanoferrate (II).3H2O (reagent grade), sulfur powder (99.98%), and lithium perchlorate (99.99%) were purchased from Sigma Aldrich, Australia, and ethanol (99.5%) from ChemSupply, Australia. All chemicals were used without further purification. Ferrocyanide

and

ferricyanide

solutions

were

prepared

by

dissolving

potassium

hexacyanoferrate (II) and (III), respectively, in ultrapure water (Milli-Q purification system) at 40~50 oC. A polysulfide electrolyte was prepared using 3 M S, 2 M Na2S.9H2O and 0.1 M LiClO4 in an ethanol:water mixture at a ratio of 1:4, and purged with N2 gas. Preparation of TiO2 dense layer and nanocrystalline TiO2 films. A fluorine doped tin oxide (FTO) glass plate (15 × 25 mm2, 9 - 11 Ω/sq., Asahi Glass, type-U) was sonicated for 15 minutes in acetone. A dense TiO2 layer was prepared based on the method described previously.18 An isopropanol solution containing 0.38 M titanium di-isopropoxide bis(acetylacetonate) was prepared and spray-pyrolyzed onto the cleaned FTO substrate at 450 oC for 1 s. The coating was repeated 10 cycles. The substrate was then sintered at 450 oC for 15 minutes in air. A TiO2 nanocrystalline film was prepared on top of the TiO2 dense layer by screen-printing a TiO2 paste, PST-30NRD (JGC-CCIC Co., Ltd., Japan), and then sintered at 500 oC for 1 hour in an air flow oven (printed area: 0.25 cm2, film thickness: 6 µm).18 Preparation of CdS QD sensitized TiO2 films. A TiO2 film was sensitized by CdS QD with ZnS coating by using the SILAR method described previously.15, 18 One cycle was completed by dipping the film in 0.1 M Cd(ClO4)2 aqueous solution for 1 minute, followed by rinsing in water and drying, and afterwards dipping in 0.1 M Na2S aqueous solution for 1 minute, followed by

ACS Paragon Plus Environment

5

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 31

rinsing in water and drying. This coating procedure was repeated 10 cycles. Subsequently, to form ZnS shell structure, the CdS QD sensitized film was dipped in 0.1 M Zn(O2CCH3)2 aqueous solution, followed by rinsing in water and drying, and afterwards dipping in 0.1 M Na2S aqueous solution for 1 min, followed by rinsing in water and drying. This procedure was repeated 2 cycles. The ZnS coating was employed to passivate trap states in the CdS QDs.18, 21-23 Solar cell fabrication. A CdS QD sensitized TiO2 film was employed as a working electrode and a platinum coated FTO substrate as a counter-electrode. A sandwich-type solar cell was fabricated by binding the ferricyanide/ferrocyanide redox electrolyte using these two electrodes with 25 µm thick polyethylene spacer (PANAC Ltd., Japan).15, 24 All the steps involved in the solar cell fabrication were conducted in an ambient condition. Note that we noticed that the solar cell fabricated under the identical condition indicated slightly different short circuit photocurrent density, probably owing to an experimental error, and thus to minimise this difference, a set of experiments were conducted at the same time when we investigated influence of one parameter, e.g. oxidized or reduced species concentration. Photovoltaic measurements. The photovoltaic performance was assessed by observing incident photon-to-current conversion efficiency (IPCE) spectra and current density – voltage (J – V) curves under the light irradiation with an irradiation aperture area of 4.8 × 4.8 mm2 using a photo-mask. IPCE spectra were obtained by using a monochromatic light source (Bunko-Keiki, SM-100) and a source meter (Keithley Instruments Inc.: 2400). A 0.5 cm2 silicon photodiode was used to calibrate the monochromatic light intensity. J – V curves were measured under AM1.5G solar simulated light (one sun condition, 100 mW/cm2 at 25 ± 2 °C) from a solar simulator (HAL-320, Asahi Spectra Co. Ltd.) and using a source meter (Keithley Instruments Inc.: 2400). The simulated light power density was calibrated by a reference Si photodiode. The

ACS Paragon Plus Environment

6

Page 7 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

measurements were performed by sweeping voltage between two electrodes with a step of 5 mV and delay time of 0.1 s. Light intensity dependence data were collected with the light intensity from 9 to 130 mW.cm-2 by adjusting the power with stainless screen mesh sheets. Transient photocurrent measurements. Transient photocurrent measurement was conducted for a CdS QD sensitized solar cells using AM1.5G solar simulated light (one sun condition, 100 mW.cm-2) at the short circuit condition. The transient photocurrent was recorded after the illumination was turned on, and the light was continuously irradiated during the measurements. Transient open circuit voltage decay measurements. Transient VOC decay measurement was conducted using the method described previously25 for the sensitized solar cells. The cell was illuminated at an open circuit condition under AM1.5G solar simulated light (one sun condition, 100 mW/cm2 at 25 ± 2 °C) from a solar simulator (Asahi Spectra Co., Ltd., HAL-320). After the VOC indicates a steady value, the illumination was turned off with a shutter, and a VOC decay was observed using a potentiostat (Ivium Technologies B.V., Compact Stat) with the measurement step of 0.1~0.2 s. The observed VOC decay was analyzed, and the lifetimes of the electrons in the TiO2 conduction band, as a function of VOC, was calculated based on the following equation (1).25

 = −

  







(1)

where τn is lifetime of an electron in the TiO2 conduction band, k is Boltzmann constant, T is temperature, e is an elementary charge, and t is time after turning off the solar simulated light.

ACS Paragon Plus Environment

7

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 31

RESULTS AND DISCUSSION Potential level diagram. We first identified potential level relationships of a TiO2 conduction band edge, CdS QD conduction and valence band edges, and an electrolyte redox potential in a different electrolyte composition. A polysulfide electrolyte and a ferricyanide/ferrocyanide electrolyte were compared, as shown in Scheme 1. Scheme 1a shows potential level diagrams when a polysulfide electrolyte (3 M S, 2 M Na2S and 0.1 M LiClO4) is employed in a solar cell. The electrolyte solution is alkaline with pH 12.5,26-27 and thus conduction band edges of TiO2 and CdS shift negatively.28-30 A TiO2 conduction band edge at pH 12.5 was determined to be 0.95 V vs. Ag/AgCl, by using -0.2-(0.06 × pH) V vs. Ag/AgCl.28-29 The band gap of SILAR coated CdS QDs (coating cycles of 10~15) with an average diameter of 4 nm20 was calculated to be 2.83 eV, using the quantum chemical calculation with a finite depth potential well model.20 A CdS QD conduction band edge was estimated to be -1.22 V vs. Ag/AgCl, considering potential shifts owing to the quantum size effect20 and pH 12.5 (negative shift with 0.035 V per pH for pH of >10).30 An electrolyte redox potential is -0.65 V vs. Ag/AgCl, invariant to the composition and pH.26, 31

ACS Paragon Plus Environment

8

Page 9 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

(a)

(b)

TiO2 CB CdS QD

-1.0

-0.95 V

-1.22 V S 2-/S2X

TiO2 CB CdS QD

-1.0

-0.62 V

-1.13 V

-0.65 V 0

Fe(CN)63-/4-

0

+0.25 V +1.0

+1.0 +1.61 V +2.0 V vs. Ag/AgCl

VB

+1.70 V +2.0

VB

V vs. Ag/AgCl

Scheme 1. Potential level diagram of (a) CdS QD sensitized TiO2 in aqueous polysulfide electrolyte (Sx2-/S2-) at pH 12.5, and (b) CdS QD sensitized TiO2 in aqueous Fe(CN)63-/4electrolyte at pH 7. A TiO2 conduction band edge was calculated, using -0.2-(0.06 × pH) V vs. Ag/AgCl.28-29 An exponential tail of density of trap states below the TiO2 conduction band is also illustrated. CdS QD band gap and conduction band edge were estimated using the finite depth potential well model20 and with the pH dependence.30 The redox potentials of Sx2-/S2- and Fe(CN)63-/4- electrolytes are invariant to the solution pH.26, 31-32

Scheme 1b shows a potential level diagram when a ferricyanide/ferrocyanide electrolyte is employed in a solar cell. Following the same methods above, a TiO2 and CdS QD conduction band edge at pH 7 was calculated to be -0.62 V and -1.13 V vs. Ag/AgCl, respectively. A ferricyanide/ferrocyanide redox potential is +0.25 V vs. Ag/AgCl (see below for the determination), invariant to the solution pH.32 This scheme suggests that the potential difference between the TiO2 conduction band edge and the polysulfide redox potential, 0.3 eV, is smaller

ACS Paragon Plus Environment

9

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 31

than that between the TiO2 conduction band edge and the ferricyanide/ferrocyanide redox potential, 0.87 eV, and thus the larger VOC can be expected for a solar cell based on the ferricyanide/ferrocyanide electrolyte. Comparison of electrolytes between ferricyanide/ferrocyanide and polysulfide redox couple on solar cell performance. We have first investigated performance of solar cells with ferricyanide/ferrocyanide in comparison with a polysulfide redox couple. Their performances were shown in Figures 1a and 1b. The composition of a ferricyanide/ferrocyanide aqueous redox electrolyte is 0.3 M/0.015 M, respectively. The J-V curve obtained for the solar cell based on the ferricyanide/ferrocyanide redox couple has reproduced our previous results.18 The resultant J-V curves clearly indicate that VOC and FF (0.67V and 0.74) have increased for the solar cell based on the ferricyanide/ferrocyanide electrolyte, compared to 0.32 V and 0.35 with the polysulfide electrolyte, while the short circuit photocurrent, JSC, of these solar cells are similar. Note that both solar cells show no sign of degradation over several hours, supporting that ZnS coating on CdS QDs protects corrosion reaction with the electrolyte.32 Previously, the rapid performance degradation of the dye sensitized solar cells with the ferricyanide/ferrocyanide (40/400 mM) redox couple was reported.33 This performance degradation was initiated by light excitation of ferricyanide that absorbs light at