Charge Transport and Recombination in TiO2 ... - ACS Publications

Sep 26, 2014 - The photoactivity of the stable nanosized TiO2 polymorphs is challenging for many advanced applications. In the present work, brookite ...
0 downloads 0 Views 2MB Size
Subscriber access provided by TULANE UNIVERSITY

Article 2

Charge Transport and Recombination in TiO Brookite-based Photoelectrodes. Yuly Kusumawati, Mongia Hosni, Mohamad A. Martoprawiro, Sophie Cassaignon, and Thierry Pauporté J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/jp5047479 • Publication Date (Web): 26 Sep 2014 Downloaded from http://pubs.acs.org on October 2, 2014

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.

The Journal of Physical Chemistry C 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

The Journal of Physical Chemistry

J. Phys. Chem. C, Revised version 25/09/2014

Charge Transport and Recombination in TiO2 Brookite-based Photoelectrodes. Y. Kusumawatia,b, M. Hosnia,c , M. A. Martoprawirob, S. Cassaignond,e,f, Th. Pauporté*a a

Institut de Recherche de Chimie Paris, CNRS – Chimie ParisTech- CNRS- PSL Research University, UMR8247, 11 rue Pierre et Marie Curie, 75005 Paris. b Inorganic and Physical Chemistry Division, Faculty of Mathematics and Natural Sciences, Institut Teknologi Bandung (ITB), Jl. Ganesha 10, Bandung 40132, Indonesia. cLaboratoire des Sciences des Procédés et des Matériaux, LSPM UPR 3407, Université Paris 13/ CNRS, Sorbonne Paris Cite, 93430 Villetaneuse, France. dSorbonne Universités, UPMC Univ Paris 06, UMR 7574, Chimie de la Matière Condensée de Paris. eCNRS, UMR 7574, Chimie de la Matière Condensée de Paris. fCollège de France, UMR 7574, Chimie de la Matière Condensée de Paris, F-75005, Paris, France.

Abstract The photoactivity of the stable nano-sized TiO2 polymorphs is challenging for many advanced applications. In the present work, brookite TiO2 nanoparticles with two different shapes have been used as building blocks for the preparation of pure brookite mesoporous layers. The layers have been characterized before and after sensitization. They have been used as photoanodes in dye-sensitized solar cells (DSSCs). The cell functioning coupled processes have been investigated by the impedance spectroscopy (IS) technique at various applied voltages and compared to a reference anatase TiO2 solar cell. The investigations of the chemical capacitance and of the charge transfer resistance, Rct, show that, compared to anatase, the brookite surface is less active for the recombination side reaction. The larger Rct is shown to explain the higher open circuit voltage of the brookite cells. However, the charge transport is much slower in the brookite phase due to a lower electrical conductivity. This parameter has been quantified more than one order of magnitude lower in the brookite layers compared to the anatase one. On the whole, the efficiency of brookite DSSCs is mainly limited by two parameters, the dye loading and the charge collection efficiency. *Author for correspondence: Tel.: (33)1 55 42 63 83; E-mail : [email protected]. Keywords : TiO2, Dye-sensitized solar cells, Brookite, Impedance spectroscopy. 1 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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

1.

Page 2 of 31

Introduction Titanium dioxide is a wide bandgap semiconductor gifted with many advantageous

chemical, structural, optical and electronic properties.1-6 The main applications of TiO2 include pigments,4 functional coatings,7 gas sensors,8 batteries,9 photocatalysis10,11 and the energy field through, among others, hydrogen production12 and solar cells.13-16 Anatase, brookite and rutile are the three TiO2 polymorph phases stable under ambient conditions.17,18 Brookite nanoparticles are stable up to 600°C and the anatase ones up to 836°C19 whereas at higher temperature they transform to the rutile phase. Rutile and anatase are structurally similar, being made of TiO6 octahedra chains. The difference between their structures lies in the higher distortion in anatase compared to in rutile, so anatase has a lower orthorhombic symmetry.20 On the contrary, brookite adopts an orthorhombic structure in which the octahedra share three edges and also corners, and the dominant structural feature is a chain of edge sharing.21 The difference in term of structure between the TiO2 polymorphs leads to different properties. For example, in rutile each octahedron is in contact with 10 neighbor octahedra whereas in anatase, each octahedron is in contact with 8 neighbor octahedra. This lattice structure makes rutile denser compared to anatase, furthermore, it makes their electronic band structures different with different bandgaps.18,20 Some physical properties of brookite lie between those of rutile and anatase. For example, the refractive index of anatase, brookite, and rutile increases in the order 2.52, 2.63, and 2.72, while their theoretical density are 3.84, 4.11, and 4.26 g/cm3, respectively.21 As a wide bandgap semiconductor, TiO2 has an important role in dye-sensitized solar cells (DSSCs). It receives electrons from photoexcited dyes attached on its surface and plays the role of the electron transport material ensuring the electron transfer to the back contact of the solar cell. The injected electrons experience two types of processes: transport and recombination. The

2 ACS Paragon Plus Environment

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

The Journal of Physical Chemistry

transport is a forward movement of electrons to a back contact collector as a result of a gradient of electron concentration. It is a field-free random walk process. In the generally accepted multiple trapping (MT) model,22-24 the electrons are considered to be mostly trapped in localized states below the conduction band edge, from which they can escape by thermal activation. The electron movement is characterized by the transport time (τtr), which is the mean time for the electron to reach the back contact. Recombination is a back flow of electrons to oxidized dyes and tri-iodide ions. To ensure good performances, mesoporous oxide layers are used since they permit the adsorption of a very large amount of dye and can be impregnated with the electrolytic solution. The dye must perfectly cover the TiO2 surface in order to avoid the recombination reaction between electrons in the conduction band of TiO2 and the redox shuttle in the electrolyte. The phenomenon is characterized by the electron lifetime (τn), the mean lifetime before recombination. The phase, shape and size of the TiO2 nanoparticles building blocks have then a great influence on the cell performances.13,18,25,26 Among the three polymorphs of TiO2, anatase is the leading phase used in most of the DSSC litterature because of its good conductivity.13,18 Moreover, the anatase (101) surface has the lowest surface energy, which supports the dye adsorption.27 There are fewer data on DSSCs using the rutile phase. DSSC based on this phase has drawn some attention because of its potentially cheaper production cost and its superior light-scattering characteristics that could be useful for solar light harvesting.28-31 Brookite has been very poorly investigated for DSSC application. A very low conversion efficiency has been reported by Lancelle-Beltran et al. for nano-brookite based solid-state DSSCs.19 Yanagida, et al. have studied the use of a 75% brookite and 25% anatase mixture in DSSC photoelectrodes and they have obtained an efficiency of 4.1%.32 Magne et al. have shown that brookite is a promising photoelectrode material for DSSC

3 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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

application. Brookite-based photoelectrodes have been prepared and the best overall conversion efficiency reached 5.97%.25 Brookite cells showed a higher open circuit voltage, Voc, compared to anatase.18,25 In this paper, the functioning of brookite TiO2 DSSCs has been carefully investigated by the impedance spectroscopy (IS) technique in order to discriminate the factors that still limit the performances of brookite DSSCs. We compare brookite cells prepared with two kinds of nanoparticles differing in their shape and size with a reference anatase-based solar cell. IS is a powerful investigation method to extract coupled key cell functioning parameters and to understand the electron dynamics in the photoelectrodes.33,34 The recombination reaction resistance, the electron lifetimes and transport times and electron transport parameters have been determined over a large applied voltage range. Moreover, the origin of the Voc difference between the cells is explained by analyzing the impedance data. The data have been completed by other useful photoelectrode characterizations and linked to the cell performances.

2. Experimental section 2.1- Film preparation. Two different types of brookite nanoparticles, hereafter noted B1 and B2, were prepared. The brookite particles B1 were synthesized by addition of pure TiCl4 in a 3 mol.L-1 HCl solution to form a colorless solution with a Ti concentration of 0.15 mol.L-1.35 The solution was heated and aged at 95°C over 3 days and was then peptized to eliminate the rutile phase. The brookite particles B2 were synthesized by co-hydrolysis of the aqueous precursors TiCl3 and TiCl4 with a total Ti concentration of 0.04 mol L-1.36 The pH of an equimolar solution of Ti3+ and Ti4+ was

4 ACS Paragon Plus Environment

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

The Journal of Physical Chemistry

adjusted to 4.5 and the suspension was aged for one week at 60°C. The pure brookite phase of the resulting nanoparticles was checked by XRD and Raman spectroscopy.25 1 g of B1 or B2 TiO2 brookite nanoparticles, 4.06 g of terpineol, 5 mL of ethanol were mixed together under vigourous agitation. 281 mg of ethyl cellulose (EC) powder (5–15 mPa.s) and 219 mg of EC (30–50 mPa.s) were dissolved in 4.5 mL of ethanol. The two solutions were then mixed and sonicated several times using an ultrasonic horn. Ethanol and water were subsequently removed from the solution in a rotary-evaporator at an initial temperature of 58 °C in order to create a viscous paste.18,37 FTO glass substrates (TEC15, Pilkington) were cleaned with soap and rinsed with distilled water. They were then treated in an ultrasonic bath in acetone for 5 min and in ethanol for 5 min. The substrates were dried and placed in a furnace at 450°C for 30 min. After cooling down, they were immersed in a 40 mmol.L-1 aqueous solution of TiCl4 at 70°C for 30 min and rinsed with water and ethanol. The substrate was then dried and annealed at 500°C for 30 min. This process was repeated once. A layer of brookite TiO2 paste was coated on the conducting glass substrates by the doctor blade technique, relaxed and dried at 125°C for 5 min. The step was repeated several times in order to achieve the accurate film thickness. The reference anatasebased solar cells were prepared in the same manner using the Dyesol DSL18NR-T paste (noted A1). No scattering layer was used for all the investigated cells. The films were then annealed at 500°C for 15 min. A final TiCl4 treatment consisted in immersing the TiO2 films in a 40 mmol.L1

TiCl4 solution at 70°C for 30 min and annealing again at 500°C. The layer thicknesses were measured with a Dektak 6M stylus profiler. The film structure

was determined by a Phillips Xpert high-resolution X-ray diffractometer operated at 40 kV and 45 mA using the CuKα radiation with λ = 1.5406 Å. The mean film porosities were estimated by

5 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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

mass measurements of several layers. They were equal to 0.60 for the B1 brookite layer and equal to 0.65 for both the B2 brookite and A1 anatase layers. The size and the shape of the brookite particles were determined using a JEOL 2011 transmission electron microscope (TEM) operated at 200 kV.

2.2 Solar cell preparation and characterizations The TiO2 layers were immersed upon cooling in a 0.5 mmol.L-1 N719 dye solution. The dye was dissolved in a solvent mixture of acetonitrile and tert-butanol (1:1) plus one equivalent of tetrabutylammonium hydroxide. The sensitization occurred in the dark, at room temperature for 24h. For the counter electrode preparation, FTO glass substrates were cleaned by ultrasound in acetone and ethanol for 5 min each. They were then treated in a furnace for 30 min at 450°C to remove organic contaminants. The Pt catalyst was deposited on the FTO glass by coating with a droplet of H2PtCl6 solution (6 mg in 1 mL ethanol) and subsequently heated at 400°C for 20 min. This step was repeated once. The two electrodes were sealed with a 50 µm hot melt spacer (Surlyn, DuPont) and the internal space was filled with the electrolyte through a hole drilled in the counter electrode, which was subsequently sealed with Surlyn and an aluminum foil. The electrolyte employed was a solution of 0.6 mol.L-1 1,2-dimethyl-3-propylimidazolium iodide (DMPII), 0.1 mol.L-1 LiI, 0.05 mol.L-1 I2, 0.10 mol.L-1 guanidinium thiocyanate and 0.5 mol.L-1 4-tertbutylpyridine in a mixture of acetonitrile and valeronitrile (85/15 volume ratio). The dye loading of the photoelectrode was determined by spectrophotometry, after complete dye desorption in a 0.1 mol.L-1 KOH solution. The I-V curves were recorded by a Keithley 2400 digital sourcemeter, using a 0.01 V.s-1 voltage sweep rate. The solar cells were illuminated with a solar simulator (Abet Technology 6 ACS Paragon Plus Environment

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

The Journal of Physical Chemistry

Sun 2000) filtered to mimic air mass AM 1.5G conditions. The power density was calibrated to 100 mW.cm-2 by the use of a reference silicon solar cell. The illuminated surface was delimited by a black mask. The incident-photon-to-electron-conversion-efficiency (IPCE) spectra were measured at short circuit with a home-made Jobin-Yvon system. The cells, with a surface area of 0.36 cm2, were characterized by impedance spectroscopy (IS). The spectra were measured in the dark (except when mentioned), over a large potential range, by a Solartron FRA1255 frequency response analyser coupled with a PAR273 EGG potentiostat. The AC signal was 10 mV and the frequency range was 100 kHz-0.05 Hz. The reproducibility of the impedance data was checked on several cells prepared from different batches. The presented results are those of typical cells prepared with A1, B1 and B2 initial nanoparticles. The spectra were fitted and analyzed using the Zview modeling software (Scribner). The impedance spectroscopy results are corrected for IR-Drop over the sum of all series resistances. The real potential (Vcor) applied is determined by the subtraction of the voltage drop (VDrop) from the applied potential (Vapplied). The voltage drop is calculated by the integration of the sum of all series resistances, Rseries= Rs+Rce with Rs being the series resistance and Rce the resistance due to the counter electrode (the diffusion resistance was negligible) over the current passed.33,34

3. Results and discussion 3.1 Effect of TiO2 nanoparticle crystal phase on the solar cell characteristics. Figure 1a and 1b are TEM micrographs of the B1 and B2 brookite nanoparticles, respectively, used for the preparation of the mesoporous films. The B1 particles are aggregated. Their size distribution analysis gives a mean diameter of 13 nm (Figure S1 supporting

7 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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

information and Table 1). The B2 particles are more anisotropic and elongated. The size distribution analysis in Figure S1 of the supporting information gives a 17x10 nm mean size (Table 1). The particles have been combined with ethyl cellulose as a blowing agent and mesoporous films have been prepared. In the followed preparation protocole, the maximum layer annealing temperature was 500°C, which is below the reported phase transition temperature of brookite to rutile.19 Figures 1C and S2 show the XRD patterns of B1, B2 and A1 films. All the peaks of the B1 and B2 layers are indexed by the brookite structure. No secondary phase could be found using a logarithmic Y-axis scale (Figure S2). The XRD analysis confirms that no phase change to rutile occurred upon the annealing process. The purity of brookite B1 and B2 final films was also checked by Raman spectroscopy (not shown). The average particle size after the film sintering was estimated from the FWHM of the X-ray diffraction peaks using the Scherrer equation38 at 17 nm and 20 nm for B1 and B2 films, respectively (Table 1). The annealing process leads to the enlargement of the brookite mean particle size, whereas this parameter increases to a less extent in the case of anatase (20 nm).19 We can note that the B2 and A1 films have similar properties, i.e. a porosity, p, of 0.65 and a mean particle diameter of 20 nm.

The dye loadings are gathered in Table 1. The B1 films made of the smallest brookite particles and with the smallest porosity exhibit a slightly lower dye loading compared to the B2 films which can be explained by the surface reduction by necking upon the annealing process.38,39 We note, in good agreement with our previous data,25 that the dye loading for brookite is significantly lower than that of the reference anatase photoelectrode (Table 1) as evidenced by B2 and A1 films which have similar porosity and crystallite size. It suggests a lower density of anchoring sites for the dye on the brookite surface.

8 ACS Paragon Plus Environment

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

The Journal of Physical Chemistry

Sensitized films of similar thickness have been used for the preparation of DSSCs. Typical IV curves are presented in Figure 2 and the photovoltaic curve parameters are gathered in Table 1. The efficiency of brookite cells is lower than that of the reference anatase cell. The fill factors (FF) are similar for the three devices. However, Jsc is significantly smaller for brookite. On the other hand, the brookite cells show higher Voc than the anatase reference one. Jsc of anatase cells is much larger than that of the brookite cells that is in agreement with a much larger dye loading (Table 1). As well, Jsc of B2 cells is slightly larger than B1 cells due to a better dye loading (Table 1). The B2 photoelectrodes have a higher amount of dye, this ensures a better sunlight harvesting which leads to a larger Jsc. However, the B1 cells have a higher Voc due to a slightly lower dark current (Figure 2). Overall, both brookite DCCSs have a similar external conversion efficiency close to 6%. The IPCE curves of the investigated DSSCs are presented in Figure S3. They are in good agreement with the Jsc data with lower IPCE in the case of the brookite DSSCs.

3.2 Analysis of the open circuit voltage. Impedance spectroscopy (IS) is a powerful technique that has been employed in a large extent to investigate the kinetic of electrochemical and photoelectrochemical reactions occurring in many functional systems, including DSSCs, in which coupled processes are involved.40-47 The impedance spectra of brookite and reference anatase cells were measured in the dark over a large applied voltage range in order to extract the kinetic data of the photoelectrodes at variable concentrations of trapped electrons (noted g). Examples of spectra are presented in Figure 3a. They have been analyzed according to the multiple trapping (MT) model described in the introduction.48 They all showed a characteristic low-middle frequency semicircle due to the resistance to charge transfer (recombination) (Rct) across the sensitized oxide-electrolyte

9 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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

interface coupled to the total electrode capacitance, denoted as Cµ, which is a chemical capacitance due to filled trap states localized below the conduction band minimum. At high frequency, a second semicircle was found due to the resistance (RPt) and capacitance (CPt) of the charged counter-electrode/electrolyte interface. In the intermediate frequency range, a ~45° straight-line segment could be clearly observed for both B1 and B2 cells polarized at not too high applied voltage, which is characteristic of the electron transport by diffusion and is modeled by a transport resistance, noted Rtr in Figure 3b. The high frequency series resistance, Rs, is due to the electrical contacts. The full equivalent circuit used to fit the spectra is presented in Figure 3b. We noted that for all the investigated cells based on brookite or anatase, the IS spectra did not exhibit a clear Warburg loop at low frequency, even at a high applied voltage (not shown). This clearly excludes a performance limitation due to the diffusion of the I-/I3- redox shuttle. Cµ is a chemical capacitance due to filled trap states localized below the conduction band minimum (sub-band gap state). In the MT model of the DSSC functioning, these states are usually described as mainly located near the particle surface and/or at the necks between adjacent particles.49-53 Cµ is plotted as a function of Vcor in Figure 4a and it varies more or less exponentially as expected for the model used. The experimental data have been fitted by the relationship:

 qV   qV  Cµ = C 0, µ exp α cor  = C 0, µ exp  cor   k BT   k BT0 

(1)

where kB is the Boltzman constant (1.381 10-23 m2.kg.s-2.K-1), T the absolute temperature and q the elementary charge (1.602 10-19 C). α is a parameter that accounts for the depth of the trap energy distribution below the conduction band. α value for B1 and B2 is equal to 0.18 and 0.22, respectively (Table 2). For the reference anatase TiO2 solar cell this parameter is found at 0.23. α 10 ACS Paragon Plus Environment

Page 11 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

The Journal of Physical Chemistry

of anatase is classically reported in the literature to range between 0.2 and 0.4.54,55 According to Equation (1), the trap state depth can also be expressed as a temperature, T0, found between 1330K and 1600K for the investigated brookite cells (Table 2). In Figure 4a, the B1 curve has a lower slope than the A1 curve but both curves more or less overlap. Therefore it can be concluded that the conduction band edge for these two compounds is located at about the same energy level. On the other hand, Cµ of B2 are higher and shifted by 33 mV (-∆Ec/q) compared to A1 (Figure 4a and Table 3). It suggests that the conduction band edge of B2 is located 33 mV below the conduction band edge of B1 and A1. These results confirm that the better Jsc of anatase cells is due to the higher dye loading and not to a poor efficiency of electron injection between the excited dye and the oxide matrix. In Figure 4b, Rct is presented versus Vcor. Less recombinations are found for the brookite cells compared to the anatase one, this in spite of a lower dye concentration (Table 1). In the same figure, we have also plotted as A2 Rct values for anatase cells sensitized for 4h and that exhibit a dye concentration of 85 mM that is close to that of the brookite cells. As expected, in that case, Rct is lowered with faster recombinations. In Figure 4b we can also note that the cell with the highest Rct values (B1) has the smallest dark current in Figure 2. Rct of the TiO2 cells show an exponential variation, in agreement with the trapping-detrapping model, that follows the relationship:

 qV  Rct = R0, ct exp− β cor  k BT  

(2)

where β can be considered an estimation of the reaction order. β values for B1 and B2 cells are 0.60 and 0.57, respectively (Table 2). β lower than 1 is an empirical way to describe sublinear

11 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 12 of 31

recombination kinetics that takes into account that electrons may be transferred from occupied levels located in the energy gap.56-58 The higher recombination rate in the A1 photoelectrode can explain the lower Voc of these cells. We have quantified the effect of recombination on Voc by analyzing the IS data according to the procedure described by Fabregat-Santiago et al in the Ref.59. In the Figure S4, Rct is plotted versus the applied voltage corrected of the ohmic drop and of the -33 mV shift due to the different conduction band energy level (CBEL) in the case of B2 cells. This potential is noted Vecb with Vecb=Vcor-∆Ec/q. The CBEL of the anatase A1 photoelectrode has been taken has the reference. We can see that after correction, B1 and B2 cells have the same Rct values which are higher than that of A1 cells. The potential shift between the two curves, noted ∆Vk, is 57mV. It is the contribution of Rct on the improvement of the Voc in the brookite DSSCs. ∆Vcal is the voltage shift due to ∆Ec/q and ∆Vk. We found 57 mV for B1 and 24 mV for B2 (Table 3). These values are close to the recorded Voc shifts that are 50 mV and 30 mV for B1 and B2 cells, respectively (Table 3). Therefore, we can specify the origin of the Voc change, already observed in our previous data25 and conclude that the Voc improvement in brookite DSSCs is due to a reduced recombination phenomenon.

3.3. Analysis of charge transport and recombination in the photoelectrodes. The electron lifetime, τn, is the mean time before the charge recombination with the tri-iodine species in solution. It has been calculated according to: τn=RctCµ.48 Due to the different trap state distribution and conduction band level for the various electrodes, these parameters have been compared as a function of the trapped electron concentration, g. From Cµ, g has been calculated according to: 12 ACS Paragon Plus Environment

Page 13 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

The Journal of Physical Chemistry

g (Vcor ) =

Cµ (Vcor ) qAd (1 − p)

(3)

where A is the geometric area of the cell, d the oxide layer thickness and p the film porosity. τn and τtr are displayed in the Figure 5. τn decreases with the applied potential (and then g) because the electron interception increases with the driving force. τn for the two different brookite kind of cells are longer than for the anatase reference cell in spite of a lower dye loading. The transport properties of the films are described by the Rtr parameter (Figure S5 in the supporting information). The transport time is the mean time for the photogenerated electrons to reach the back contact. It has been calculated as τtr=RtrCµ48 and is plotted as a function of g in Figure 5. We observe that the TiO2 crystal phase has a strong influence on this parameter since much shorter transport time is found for anatase compared to brookite. Figure 5 clearly illustrates that the ratio between the electron lifetime and transport time strongly depends on the crystal phase. τn is about two orders of magnitude higher than τtr in the case of anatase. On the other hand, it is only one order of magnitude higher for the two types of investigated brookite cells. This has a consequence on the charge collection efficiency, ηcoll, in the photoelectrodes. ηcoll has been calculated using the following classical relationship:60

ηcoll =

1  τtr  1 +   τn 

(4)

The results are displayed in Figure 6. The collection efficiency is very high, higher than 95%, in the case of anatase cell, whereas significantly lower values are found in the case of the brookite cells. Therefore a cell performance loss is expected for the latter phase due to the poorer

13 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 14 of 31

collection of charge. The lower ηcoll can be explained by the conductivity behavior. The conductivity of TiO2 was determined from the charge transport resistance:

σn =

d A(1 − p) Rtr

(5)

Figure 7 shows the conductivity of brookite and anatase layers as a function of g. σn increases rapidly with this parameter due to the trap filling. Moreover, the conductivity measurements by IS show that brookite is less conductive than anatase by more than one order of magnitude. From τn and τtr measured by IS, we have also estimated the effect of the crystal phase on the chemical diffusion coefficient of electrons in TiO2, noted Dn. Dn = d2/τtr

(6)

In Figure S6 (supporting information), Dn is plotted as a function of g. As expected, the behavior is similar to that found for σn and Dn is much larger for anatase compared to brookite. Using the IS data, we have also determined the electron diffusion length (Ln) defined as the average distance traveled by the electrons in TiO2 before recombination. This parameter is given by:33

Ln =

Dn τ n = d

τn τtr

(7)

The results are displayed in Figure 8. The electron diffusion lengths in the brookite films measured in the dark are found to be larger than the photoelectrode thickness (about 14 µm) in all cases, in agreement with the ηcoll determined above. Ln has also been estimated by measuring the IS spectra of the cells at various applied voltages under one sun illumination. τn and τtr parameters have been extracted and the resulting Ln are plotted as star symbols in Figure 8.

14 ACS Paragon Plus Environment

Page 15 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

The Journal of Physical Chemistry

Similar Ln are found either in the dark or under illumination once plotted as a function of the electron concentration. Therefore, ηcoll can be estimated using any of these measurement conditions.

Conclusions We have investigated DSSCs prepared using (i) brookite nanoparticles with two different sizes and shapes and (ii) a reference anatase commercial paste. The photoelectrodes with the same thickness have been thoroughly characterized, especially for their dye loading. Moreover, the cell performances have been measured and their functioning has been examined by impedance spectroscopy. We show that layers with the same porosity and particle size have a significantly higher dye loading in the case of anatase TiO2 compared to brookite TiO2. This is due to a lower density of anchoring sites for the dye on the brookite surface. The analysis of Cµ has shown that the conduction band edge in brookite is similar or below that of the anatase photoelectrode. We have also found higher Rct for brookite cells due to a lower recombination reaction rate and therefore to the lower reactivity of the brookite surface. We have explained quantitatively the higher Voc of brookite cells by the reduced recombination side reaction. The transport time of electrons is much slower in brookite films compared to the anatase one. Therefore, in spite of larger electron lifetimes in brookite cells, the collection efficiency is significantly lower compared to anatase. The transport is slowed by the poor conductivity of brookite since we have measured σn lower by much than one order of magnitude compared to anatase. In summary, the brookite cell efficiency is mainly limited by two parameters, the dye loading and the charge collection efficiency. The latter limits the optimized photoelectrode thickness and makes it not

15 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 16 of 31

possible to compensate the lower dye loading by the use of thicker films. Further improvements should then require the doping of brookite TiO2 in order to circumvent the electrical limitations.

Acknowledgments Gilles Lancel (Collège de France and ENSCP, France) is acknowledged for the brookite nanoparticle preparation. Yuly Kusumawati acknowledges Campus France and the Higher Education Ministry of Indonesia (DIKTI) government for financial support in the framework of the DDIP collaboration program. The DIM C-nano Ile-de-France is acknowledged for financial support through the Hybrid-PV project.

Supporting Information Particle size distribution, XRD patterns, IPCE curves, Rct versus Vecb, Rtr versus Vcor, Dn versus the density of states. This information is available free of charge via the Internet at http://pubs.acs.org.

16 ACS Paragon Plus Environment

Page 17 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

The Journal of Physical Chemistry

References (1) Fenoglio, I.; Greco, G.; Livraghi, S.; Fubini, B. Non-UV-Induced Radical Reactions at the Surface of TiO2 Nanoparticles That May Trigger Toxic Responses. Chem. – Eur. J. 2009, 15, 4614–4621. (2) Khare, P.; Sonane, M.; Pandey, R.; Ali, S.; Gipta, K.C.; Satish, A.J Adverse Effects of TiO2 and ZnO Nanoparticles in Soil Nematode, Caenorhabditis Elegans. J. Biomed. Nanotechnol. 2011, 7, 116–117. (3) Facci, T.; Huguenin, F. Spectroelectrochemical Properties and Lithium Ion Storage in SelfAssembled Nanocomposites from TiO2. Langmuir 2010, 26, 4489–4496. (4) Liao, Y. ; Que, W.; Jia, Q.; He, Y.; Zhang J.; Zhong, P. Controllable Synthesis of Brookite/Anatase/Rutile TiO2 Nanocomposites and Single-Crystalline Rutile Nanorods Array. J. Mater. Chem. 2012, 22, 7937–7944. (5) Koziej, D.; Fischer, F.; Kränzlin, N.; Caseri, W. R.; Niederberger, M. Nonaqueous TiO2 Nanoparticle Synthesis: a Versatile Basis for the Fabrication of Self-Supporting, Transparent, and UV-Absorbing Composite Films. ACS Appl. Mater. Interfaces 2009, 1, 1097–1104. (6) Nussbaumer, R. J.; Caseri, W. R.; Smith, P.; Tervoort, T. Polymer-TiO2 Nanocomposites: A Route Towards Visually Transparent Broadband UV Filters and High Refractive Index Materials. Macromol. Mater. Eng. 2003, 288, 44-49. (7) Parkin, I. P.; Palgrave, R. G. Self-Cleaning Coatings. J. Mater. Chem. 2005, 15, 1689-1695. (8) Karunagaran, B.; Uthirakumar, P.; Chung, S. J.; Velumani, S.; Suh, E.-K. TiO2 Thin Film Gas Sensor for Monitoring Ammonia. Mater. Charact. 2007, 58, 680–684. (9) Wang, D.H.; Choi, D.W.; Li, J. ; Yang, Z.G.; Nie, Z.M.; Kou, R.; Hu, D.H.; Wang, C.M. ; Saraf, L.V.; Zhang, J.G. Self-Assembled TiO2-Graphene Hybrid Nanostructures for Enhanced Li-Ion Insertion , ACS Nano, 2009, 3, 907-914. (10) Fujishima, A.; Zhang, X.; Tryk, D. A. TiO2 Photocatalysis and Related Surface Phenomena. Surf. Sci. Rep. 2008, 63, 515–582. (11) Linsebigler, A. L.; Lu, G.; Yates, J. T. Photocatalysis on TiO2 Surfaces: Principles, Mechanisms, and Selected Results. Chem. Rev. 1995, 95, 735–758. (12) Ni, M.; Leung, M. K. H.; Leung, D. Y. C. Sumathy, K. A Review and Recent Developments in Photocatalytic Water-Splitting Using TiO2 for Hydrogen Production. Renew. Sustain. Energy Rev. 2007, 11, 401–425. (13) Hagfeldt, A.; Boschloo, G.; Sun, L.; Kloo, L.; Pettersson, H. Dye-Sensitized Solar Cells. Chem. Rev. 2010, 110, 6595–6663. 17 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 18 of 31

(14) Kim, H.S.; Lee, C.R.; Im, J.H.; Lee, K-B.; Moehl, T.; Marchioro, A.; Moon, S.J.; HumphryBaker, R.; Yum, J.H.; Moser, J.E.; et al. Lead Iodide Perovskite Sensitized All-Solid-State Submicron Thin Film Mesoscopic Solar Cell with Efficiency Exceeding 9%. Scientific reports, 2012, 2, 591-599. (15) Burschka, J.; Pellet N.; Moon S.J.; Humphry-Baker, R.; Gao, P.; Nazeeruddin, M.K.; Grätzel, M. Sequential Deposition as a Route to High-Performance Perovskite-Sensitized Solar Cells. Nature, 2013, 499, 316-319. (16) Liu, D.; Kelly, K. Perovskite Solar Cells with a Planar Heterojunction Structure Prepared using Room-Temperature Solution Processing Techniques. Nature Photonics, 2014, 8, 133-138. (17) Reyes-Coronado, D.; Rodriguez-Gattorno, G.; Espinosa-Pesqueira M.E.; Cab, C.; de Coss, C. ; Oskam, G. Phase-pure TiO2 Nanoparticles : Anatase, Brookite and Rutile. Nanotechnology 2008, 19, 145605. (18) Magne, C.; Dufour, F.; Labat, F.; Lancel, G.; Durupthy, O.; Cassaignon, S.; Pauporté, T. Effects of TiO2 Nanoparticle Polymorphism on Dye-Sensitized Solar Cell Photovoltaic Properties. J. Photochem. Photobiol. Chem. 2012, 232, 22–31. (19) Lancelle-Beltran, E.; Prene, P. ; Boscher, C .; Belleville, P. ; Buvat, P. ; Lambert, S. ; Guillet, F. ; Marcel, C. ; Sanchez, C. Solid-State Organic/Inorganic Hybrid Solar Cells Based on Poly(octylthiophene) and Dye-Sensitized Nanobrookite and Nanoanatase TiO2 Electrodes. Eur. J. Inorg. Chem. 2008, 6, 903 -910. (20) Chen, X.; Mao, S.S. Titanium Dioxide Nanomaterials : Synthesis, Properties, Modifications and Applications. Chem. Rev. 2007, 107, 2891-2959. (21) Li, J.-G.; Ishigaki, T.; Sun, X. Anatase, Brookite, and Rutile Nanocrystals via Redox Reactions under Mild Hydrothermal Conditions: Phase-Selective Synthesis and Physicochemical Properties. J. Phys. Chem. C 2007, 111, 4969–4976. (22) Bisquert, J.; Vikhrenko, V. S. Interpretation of the Time Constants Measured by Kinetic Techniques in Nanostructured Semiconductor Electrodes and Dye-Sensitized Solar Cells. J. Phys. Chem. B 2004, 108, 2313-2322. (23) Fisher, A. C.; Peter, L. M.; Ponomarev, E. A.; Walker, A. B.; Wijayantha, K. G. U. Intensity Dependence of the Back Reaction and Transport of Electrons in Dye-Sensitized Nanocrystalline TiO2 Solar Cells. J. Phys. Chem. B 2000, 104, 949–958. (24) Bisquert, J. Chemical Diffusion Coefficient of Electrons in Nanostructured Semiconductor Electrodes and Dye-Sensitized Solar Cells. J. Phys. Chem. B 2004, 108, 2323-2332. (25) Magne, C.; Cassaignon, S.; Lancel, G.; Pauporté, T. Brookite TiO2 Nanoparticle Films for Dye-Sensitized Solar Cells. ChemPhysChem 2011, 12, 2461–2467.

18 ACS Paragon Plus Environment

Page 19 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

The Journal of Physical Chemistry

(26) O’Regan, B.; Grätzel, M.A. Low-Cost, High-Efficiency Solar Cell Based on Dye-Sensitized Colloidal TiO2 Films. Nature 1991, 353, 737–740. (27) Thomas, A. G.; Syres, K. L. Adsorption of Organic Molecules on Rutile TiO2 and Anatase TiO2 Single Crystal Surfaces. Chem. Soc. Rev. 2012, 41, 4207-4217. (28) Park, N.-G.; Van de Lagemaat, J.; Frank, A. J. Comparison of Dye-Sensitized Rutile- and Anatase-Based TiO2 Solar Cells. J. Phys. Chem. B 2000, 104, 8989–8994. (29) Yang, W.; Wan, F.; Wang, Y.; Jiang, C. Achievement of 6.03% Conversion Efficiency of Dye-Sensitized Solar Cells with Single-Crystalline Rutile TiO2 Nanorod Photoanode. Appl. Phys. Lett. 2009, 95, 133121. (30) Byun, H.Y.; Vittal, R.; Kim, D.Y.; Kim K.J. Beneficial Role of Cetyltrimethylammonium Bromide in the Enhancement of Photovoltaic Properties of Dye-Sensitized Rutile TiO2 Solar Cells. Langmuir 2004, 20, 6853-685. (31) Ye, M.; Zheng, D.; Wang, M.; Chen, C.; Liao, W.; Lin C.; Lin Z. et al. Hierarchically Structured Microspheres for High-Efficiency Rutile TiO2-Based Dye-Sensitized Solar Cells. ACS Appl. Mater. Interfaces 2014, 6, 2893–2901. (32) Jiang, K.-J.; Kitamura, T.; Yin, H.; Ito, S.; Yanagida, S. Dye-sensitized Solar Cells Using Brookite Nanoparticle TiO2 Films as Electrodes. Chem. Lett. 2002, 31,872–873. (33) Magne, C.; Moehl, T.; Urien, M.; Grätzel, M.; Pauporté, T. Effects of ZnO Film Growth Route and Nanostructure on Electron Transport and Recombination in Dye-Sensitized Solar Cells. J. Mater. Chem. A 2013, 1, 2079–2088. (34) Pauporté, T.; Magne, C. Impedance Spectroscopy Study of N719-Sensitized ZnO-Based Solar Cells. Thin Solid Films 2014, 560, 20-26. (35) Pottier, A.; Chaneac, C.; Tronc, E.; Mazerolles, L.; Jolivet, J.P. Synthesis of Brookite TiO2 Nanoparticles by Thermolysis of TiCl4 in Strongly Acidic Aqueous Media. J. Mater. Chem. 2001, 11, 1116–1121. (36) Perego, C.; Clemencon¸ I.; Rebours, B.; Revel, R.; Durupthy, O. ; Cassaignon,S. ; Jolivet, J.P. Mater. Res. Soc. Symp. Proc. 2009, 1146E. (37) Guérin, V.M.; Magne, C.; Pauporté, T.; Le Bahers, T.; Rathousky, J.; Nanoporous Electrodeposited versus Nanoparticulate ZnO Porous Films of Similar Roughness for Dye Sensitized Solar Cell Application. ACS Appl. Mater. Interfaces 2010, 2, 3677–3685. (38) Chou, T.P.; Zhang, Q.; Russo, B. ; Fryxell, G.E. ; Cao, G. Titania Particle Size Effect on the Overall Performance of Dye-Sensitized Solar Cells. J. Phys. Chem. C. 2007, 111, 6296-6302.

19 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 20 of 31

(39) Hosni, M.; Kusumawati, Y.; Farhat, S.; Jouini, N.; Pauporté, T. Effects of Oxide Nanoparticle Size and Shape on Electronic Structure, Charge Transport and Recombination in Dye-Sensitized Solar Cell Photoelectrodes. J. Phys. Chem. C, 2014, 118, 16791–16798. (40) Pauporté, T.; Durand, R. Impedance Spectroscopy Study of Electrochromism in Sputtered Iridium Oxide Films. J. Appl. Electrochem., 2000, 30, 35-41. (41) Pauporté, T.; Andolfatto, F.; Durand, R. Some Electrocatalytic Properties of Anodic Iridium Oxide Nanoparticles in Acidic Solution, Electrochim. Acta 1999, 45, 431-439. (42) Pauporté, T.; Finne, J.; Kahn-Harari, A.; Lincot, D. Growth by Plasma Electrolysis of Zirconium Oxide films in the Micrometer Range. Surf. Coat. Technol. 2005, 199, 213-219. (43) Bisquert, J. Theory of the Impedance of Electron Diffusion and Recombination in a Thin Layer. J. Phys. Chem. B, 2002, 106, 325-333. (44) Hauch, A.; Georg, A. Diffusion in the Electrolyte and Charge-Transfer Reaction at the Platinum Electrode in Dye-Sensitized Solar Cells. Electrochim. Acta, 2001, 46, 3457-3466. (45) Wang, Q.; Moser, J.E.; Grätzel, M. Electrochemical Impedance Spectroscopic Analysis of Dye-Sensitized Solar Cells. J. Phys. Chem. B. 2005, 109, 14945-14953. (46) Adachi, M.; Sakamoto, M.; Jiu, J.; Ogata, Y.; Isoda, S. Determination of Parameters of Electron Transport in Dye-Sensitized Solar Cells using Electrochemical Impedance Spectroscopy. J. Phys. Chem. B. 2006, 110, 13872-13880. (47) Fabregat-Santiago, F.; Bisquert, J.; Palomares, E.; Otero, E.; Kuang, D.; Zakeeruddin, S.M.; M. Grätzel, M. Correlation Between Photovoltaic Performance and Impedance Spectroscopy of Dye-Sensitized Solar Cells Based on Ionic Liquids, J. Phys. Chem. C, 2007, 111, 6550-6560. (48) Bisquert, J.; Fabregat-Santiago, F. In Dye-Sensitized Solar Cells; Kalyanasundaram, K., Ed.; EPFL Press: Lausanne, Switzerland, 2010; Chapter 12, pp 457−554. (49) Zhu, K.; Kopidakis, N.; Neale, N. R.; van de Lagemaat, J.; Frank, A. J. Influence of Surface Area on Charge Transport and Recombination in Dye-Sensitized TiO2 Solar Cells. J. Phys. Chem. B 2006, 110, 25174–25180. (50) Westermark, K.; Henningsson, A.; Rensmo, H.; Södergren, S.; Siegbahn, H.; Hagfeldt, A. Determination of the Electronic Density of States at a Nanostructured TiO2/Ru-Dye/Electrolyte Interface by Means of Photoelectron Spectroscopy. Chem. Phys. 2002, 285, 157 -165. (51) Peter, L. M. “Sticky Electrons” Transport and Interfacial Transfer of Electrons in the DyeSensitized Solar Cell. Acc. Chem. Res. 2009, 42, 1839 –1847.

20 ACS Paragon Plus Environment

Page 21 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

The Journal of Physical Chemistry

(52) Galoppini, E. R., J.; Chen, H.; Saraf, G.; Lu, Y.; Hagfeldt, A.; Boschloo, G. Fast Electron Transport in Metal Organic Vapor Deposition Grown Dye-sensitized ZnO Nanorod Solar Cells. J. Phys. Chem. B 2006, 110, 16159–16161. (53) Park, K. ; Zhang, Q. ; Myers D. ; Cao, G. Charge Transport Properties in TiO2 Network with Different Particle Sizes for Dye Sensitized Solar Cells. ACS Appl. Mater. Interfaces 2013, 5, 1044 –1052. (54) Peter, L. M. Characterization and Modeling of Dye-Sensitized Solar Cells. J. Phys. Chem. C. 2007, 111, 6601–6612. (55) Wang, Q.; Ito, S.; Grätzel, M.; Fabregat-Santiago, F.; Mora-Sero, I.; Bisquert, J.; Bessho, T.; Hachiro, I. Characteristics of High Efficiency Dye-Sensitized Solar Cells. J. Phys. Chem. B 2006, 110, 25210–25221. (56) Bisquert, J.; Mora-Sero, I., Simulation of Steady-State Characteristics of Dye-Sensitized Solar Cells and the Interpretation of the Diffusion Length. J. Phys. Chem. Lett. 2010, 1, 450– 456. (57) Villanueva-Cab, J.; H. Wang, H.; G. Oskam, G.; L.M. Peter, L.M. Electron Diffusion and Back Reaction in Dye-Sensitized Solar Cells: The Effect of Nonlinear Recombination Kinetics. J. Phys. Chem. Lett. 2010, 1, 748–751. (58) Gonzalez-Vazquez, J.P.; Oskam, G.; Anta, J.A. Origin of Nonlinear Recombination in DyeSensitized Solar Cells: Interplay Between Charge Transport and Charge Transfer. J. Phys. Chem. C 2012, 116, 22687-22697. (59) Raga, S.R.; Barea, E.M.; Fabregat-Santiago, F. Analysis of the Origin of the Open Circuit Voltage in Dye Solar Cells. J. Phys. Chem. Lett. 2013, 3, 1629-1634. (60) Hagfeldt, A.; Boschloo, G.; Sun, L.; Kloo, L.; Pettersson, H. Dye-Sensitized Solar Cells. Chem. Rev. 2010, 110, 6595–6663.

21 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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 22 of 31

Table 1: Crystal phase, particle size, dye loading and cell characteristics under 100 mW.cm-2, AM 1.5G filtered illumination. Sample

Crystal

Size

Size

phase

TEMa XRDb ess

pc

Thickn

/ nm

/ nm

µm

/

Dye

Voc

Jsc

concentration

/V

/mA.cm-

FF

η /%

2

/mM

B1

Brookite

13

17

13.4

0.60

80

0.74

9.76

0.79

5.71

B2

Brookite

17x1

20

13.5

0.65

86

0.72

10.2

0.79

5.79

20

13.4

0.65

122

0.69

14.5

0.76

7.42

0 A1

Anatase a

18

b

c

Initial nanoparticle size; in the film after annealing. Film porosity.

Table 2: α, T0 and β parameters extracted from the Cµ(Vcor) and Rct(Vcor) curve analysis. Sample

α

β

T0 (K)

B1

0.18

1600

0.60

B2

0.22

1330

0.57

A1

0.23

1280

0.61

22 ACS Paragon Plus Environment

Page 23 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

The Journal of Physical Chemistry

Table 3: Analysis of the open circuit voltage difference between anatase (taken as a reference) and brookite cells. ∆Vcal is a calculated value from the impedance data analysis and ∆Voc are the experimental data.

a

Sample

A1

B1

B2

Voc (mV)

690

740

720

∆Ec/q (mV)a

/

0

-33

∆Vk (mV)b

/

57

57

∆Vcal (mV)c

/

57

24

∆Voc (mV)d

/

50

30

∆Ec/q is the voltage shift needed to compare all the cells at the same conduction band level obtained after displacing

the capacitances in Figure 4a. b∆Vk is the voltage difference in Rct due to the differences in recombination rates. c

∆Vcal is the voltage shift due to ∆Ec/q and ∆Vk (sum of both). d∆Voc is the experimental open circuit voltage shift

compared to the A1 cell used as a reference.

23 ACS Paragon Plus Environment

The Journal of Physical Chemistry

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

(a)

Page 24 of 31

(b)

(c)

Figure 1 : (a) TEM view of B1 nanoparticles; (b) TEM view of B2 nanoparticles; (c) XRD patterns of (a) B1; (b) B2 and (c) A1 TiO2 mesoporous films. A denote the anatase reflection planes, B denotes the brookite reflection planes and the red stars are the FTO diffraction contributions.

24 ACS Paragon Plus Environment

Page 25 of 31

16

c

14 12

Jsc (mA/cm¯¹)

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

The Journal of Physical Chemistry

b

10 8

a

6 4 2 0 -2 0

0.2

0.4

0.6

0.8

-4

Voltage (V) Figure 2: I-V curves of (a) B1, (b) B2 and (c) A1 under 100 mW.cm-2, AM 1.5G filtered illumination. The dashed lines are the dark currents.

25 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1000

(a)

0.60V 800

Z" (Ohm)

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 26 of 31

600

0.65V

400

200

0.70V 0 0

200

400

600

800

1000

Z' (Ohm)

(b)

Figure 3 : (a) Typical impedance spectra of a B1 brookite cell; (b) Equivalent electrical circuit used to fit the impedance spectra.

26 ACS Paragon Plus Environment

Page 27 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

The Journal of Physical Chemistry

(a)

(b)

Figure 4 : Variation of (a) Cµ and (b) Rct with Vcor for B1 and B2 brookite solar cells and A1 anatase solar cells. A2 denotes an anatase cell sensitized for 4h instead of 24h.

27 ACS Paragon Plus Environment

The Journal of Physical Chemistry

10

1 B2

τ (s)

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 28 of 31

B1

0.1 A1

B2 B1

0.01

0.001 1E+18

1E+19

g

1E+20

(cm-3)

Figure 5 : Effect of the phase and particle size on the charge carrier lifetime τn (diamond symbol) and τtr (dot symbols) versus g.

28 ACS Paragon Plus Environment

Page 29 of 31

1

A1 0.8

B2

B1

ηcoll

0.6

0.4

0.2

0 0.45

0.5

0.55

0.6

0.65

0.7

0.75

0.8

Vcor (V) Figure 6: Effect of photoelectrode TiO2 crystal phase on the charge collection efficiency.

1.E-02

1.E-03

σn (S.cm-1)

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

The Journal of Physical Chemistry

B2 1.E-04

A1 1.E-05

B1 1.E-06 1E+18

1E+19

1E+20

g (cm-3) Figure 7: Effect of photoelectrode TiO2 crystal phase on the electronic conductivity as a function of the electron concentration, g. 29 ACS Paragon Plus Environment

The Journal of Physical Chemistry

60 50 40

Ln (µ µm)

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 30 of 31

30 20 10 0 0

5E+19

g (cm-3)

1E+20

1.5E+20

Figure 8 : Mean diffusion length of electrons, Ln, in mesoporous brookite TiO2 films measured by IS. Dot symbols : measurements in the dark. Star symbols : measurements under 1 sun illumination. Red : B1 cells; Blue: B2 cells. The dashed line is the photoelectrode thickness.

30 ACS Paragon Plus Environment

Page 31 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

The Journal of Physical Chemistry

Table of Contents Image

31 ACS Paragon Plus Environment