Scholarly article on topic 'Improved efficiency of dye-sensitized solar cells by doping of strontium aluminate phosphor in TiO2 photoelectrode'

Improved efficiency of dye-sensitized solar cells by doping of strontium aluminate phosphor in TiO2 photoelectrode Academic research paper on "Chemical sciences"

Share paper
Academic journal
Materials Science-Poland
OECD Field of science

Academic research paper on topic "Improved efficiency of dye-sensitized solar cells by doping of strontium aluminate phosphor in TiO2 photoelectrode"

Materials Science-Poland, 33(2), 2015, pp. 237-241

DOI: 10.1515/msp-2015-0031


Improved efficiency of dye-sensitized solar cells by doping of strontium aluminate phosphor in TiO2 photoelectrode

Seung Hwangbo1, Jin-Tae Kim2 and Kyu-Seog Hwang3*

1 Department of Electronic & Photonic Engineering, Honam University, 59-1 Seobong-dong, Gwangsan-gu, Gwangju

506-714, Korea

2Department of Photonic Engineering, Chosun University, 309 Pilmun-daero, Dong-gu, Gwangju 501-759, Korea 3Department of Biomedical Engineering, Nambu University, 864-1 Wolgye-dong, Gwangsan-gu, Gwangju 506-824, Korea

SrAl2O4:Eu2+, Dy3+ phosphor was synthesized by chemical solution route to use as a dopant in TiO2 layer employed as a photoelectrode for down conversion of ultraviolet and near-ultraviolet to visible and near-infrared light in a dye-sensitized solar cell. Nano-crystalline structure of the SrAl2O4:Eu2+, Dy3+ powder was confirmed by X-ray diffraction analysis and field emission scanning electron microscopy. Monitored at 520 nm, SrAl2O4:Eu2+, Dy3+ phosphor showed emission peaks at 460 to 610 nm due to 4f6 ^ 4f7 transitions of Eu2+ ions. For the study, SrAl2O4:Eu2+, Dy3+ phosphor-doped TiO2 layer was deposited on fluorine-doped tin oxide coated glass by electrostatic spray deposition. The short circuit current, open circuit voltage, fill factor, and conversion efficiency of the cells were measured. Experimental results revealed that the device efficiency for the SrAl2O4:Eu2+, Dy3+ phosphor-doped TiO2 layer increased to 7.20 %, whereas that of the pure-TiO2 photoelectrode was 4.13 %.

Keywords: phosphor; down conversion; dye-sensitized solar cell © Wroclaw University of Technology.

1. Introduction

Many works have been published on the dye-sensitized solar cells (DSSCs) since the first prototype reported by O'Regan etal. [1]. DSSCs are prepared from low-cost materials and they can be significantly less expensive than conventional solidstate solar cells. Although conversion efficiency of the DSSCs (>13 %) is less than that of the best thin-film cells, its price/performance ratio is high enough to allow them to compete with traditional energy sources based on electrical generation [2]. Generally, the sensitizer of the DSSC contains mainly N3 and N719 dyes, and the DSSC consists of a porous nano-crystalline titania film sensitized by a dye for absorbing incident light [3]. The sensitizers mainly include heteroleptic ruthenium (Ru) complexes that have fairly wide absorption spectra (AA « 350 nm) but low molar extinction coefficients (5,000 to 20,000 M-1cm-1) [4].


In an ideal case, the sensitizers should absorb across a broad range of the solar spectrum to yield the greatest number of photons. However, the efficiencies of the DSSCs are lower than those of silicon solar cells. An efficient method for increasing the efficiency is widening the absorption range of the DSSC. A lot of metal compounds dyes have been produced. However, even the best of them (N719, N749, and YD2-o-C8) only absorb visible light in the wavelength range of 400 to 800 nm, and most of the solar ultraviolet and infrared light is not absorbed [3, 4]. A method to increase light harvesting and the efficiency of the solar cells can be the usage of an inorganic-organic composite material having a perovskite structure (e.g. CH3NH3PbI3) [5] and down-converting phosphor [6, 7]. Such down converter has the ability to convert higher energy photons, below 450 nm, into low energy photons, above 500 nm.

In this study, 1 mol % Eu and 2 mol % Dy-doped strontium aluminate (SrAl2O4:Eu2+, Dy3+) phosphors were prepared by using metal

Seung Hwangbo et al.

carboxylate solution. SrAl2O4:Eu2+, Dy3+ phosphors have a wide ultrafiolet (UV) light absorption band and produce blue to green wavelength, which matches the absorption band of the most typically used dyes N719 [7]. SrAl2O4:Eu2+, Dy3+ as a luminescence source was introduced into the DSSC to increase its efficiency.

2. Experimental

The crystalline nano- sized SrAl2O4:Eu2+, Dy3+ phosphors were prepared by chemical solution process using stoichiometric quantities of strontium-naphthenate [Sr2(C11H7O2)], europium (III) 2-ethylhexanoate [Eu(C8H15O2)3], dysprosium 2-ethylhexanoate (DyO6C24H45) and hydroxyaluminum bis (2-ethylhexanoate) (C16H31AlO5) and toluene which was used as a solvent. The homogeneous sol was transformed into a yellowish powder by pre-firing at 500 °C for 4 h, and then the precursor was finally annealed at 1100 °C for 240 min in Ar/H2 95 %/5 % atmosphere. For phosphor-doped DSSC study, a commercially available P25 TiO2 powder was used. The TiO2 powder and 5 wt.% of SrAl2O4:Eu2+, Dy3+ phosphor were ball milled for 24 h, adding 0.5 wt.% of ethyl cellulose in 100 mL ethanol. The sol was diluted with ethanol to adjust viscosity.

The precursor was coated on a fluorine doped tin oxide coated glass (FTO) using an electrostatic spray deposition (ESD). The ESD has been commonly used to synthesize homogeneous oxide coatings. The working principles of the ESD have been described by Kim et al. [8]. Briefly, to obtain a stable cone-jet mode of electrostatic atomization, high voltage (20 kV) was applied between the needle tip and the electrode by using high DC power supply. FTO substrates on the ground electrode were heated at 80 °C for 60 min to vaporize the organic compound. The flow rate of precursor solution containing SrAl2O4:Eu2+, Dy3+-doped TiO2 was kept at 0.3 mL/60 min. The as-deposited film was heat-treated at 550 °C for 30 min in air. The annealed film was immersed in ethanol solution containing 0.0005 M N719 dye at 40 °C for 360 min. The counter electrode was prepared by dip-coating a

FTO glass with H2PtCl6 solution (2 mg Pt/1 mL ethanol), followed by heating at 400 °C for 30 min.

The crystallinity and morphology of the samples were examined by X-ray diffraction (XRD, D-Max 1200, Rigaku, Japan) 0 to 20 scan using a CuKa (A = 1.54056 A) radiation and field emission scanning electron microscope (FE-SEM, S-4700, Hitachi, Japan). Photoluminescence (PL) spectra at room temperature were obtained using a fluorescent spectrophotometer (F4500, Hitachi, Japan) equipped with a Xenon lamp source. The UV-Vis absorption spectra of the samples were obtained by using a UV-Vis spectrophotometer (Cary 100, Varian Inc., Australia) attached to an internal diffuse reflectance accessory.

The electrolyte solution consisted of 1-butyl-3-methylimidazolium iodide, iodine, 4-tert-butylpyridine, and guanidinium thiocyanate in acetonitrile/valeronitrile at 85:15 volume ratio. The electrolyte was injected into the sealed cells via predrilled holes on the counter electrode, and the injection holes were hot sealed by a piece of thin cover glass with a hot-melt film underneath as the adhesive. The I-V characteristics of DSSCs were measured under 1 sun conditions (100 mW/cm2) using a solar simulator (Oriel Instruments, U.S.A.) equipped with a 300 w Xenon lamp and a Keithley (Model 2400) source meter, after calibrating with a silicon reference cell. The cell was covered with an aperture mask, while measuring photocurrent and voltage to avoid overestimation of the I-V characteristics.

3. Results and discussion

Fig. 1a shows a typical XRD pattern of SrAl2O4:Eu2+, Dy3+ phosphor annealed at 1100 °C for 240 min in Ar/H2 95 %/5 % atmosphere. It can be seen that the monoclinic SrAl2O4 phase pattern is characterized by three peaks around 20 = 28 to 29° (JCPDS No. 34-0379), and weak peaks at 23.6, 27.8, 30.4 and 34.1° related to Sr4Al14O25 (JCPDS No. 52-1876) as a secondary phase. Sr2+ ions are positioned in the cavities of the framework of corner-sharing AlOij- tetrahedra. Oxygen is shared with two

aluminum ions so that each tetrahedron has one net negative charge. The charge balance is disturbed by Sr2+ ions which occupy interstitial locations within the tetrahedral framework. It is commonly

considered that Eu2+ and Dy3+ ions enter into the Sr ion sites in the SrAl2O4:Eu2+, Dy3+ host. Since the radii of Eu2+ (0.130 nm) and Dy3+ (0.117 nm) ions, are approximately equal to that of the Sr2+ ion (0.127 nm), their replacement does not cause a noteworthy distortion of the lattice parameters [9]. Fig. 1b shows the XRD pattern for a pure-TiO2 sample. The diffraction peaks at 25.3°, 37.9°, 48.1°, and 53.9° are from the (101), (004), (200), and (105) reflection planes of anatase phase, and the peaks at 27.5° and 36.2° are from the (110) and (101) reflection planes of rutile phase, respectively. This suggests that the P25 is composed of a mixed phase of anatase and rutile. The microstructural morphology of the

Fig. 1. XRD patterns of SrAl2O4:Eu2+, Dy3+ phosphor after annealing at 1100 ° C for 240 min in Ar - 5 % H2 gas (a), and pure TO2 (b) FE-SEM image of the phosphor (c).

SrAl2O4:Eu2+, Dy3+ phosphor after annealing at 1100 °C for 240 min is shown in Fig. 1c. Normally, the particle size of SrAl2O4:Eu2+, Dy3+ phosphor prepared by the solid-state reaction is about several micrometers, much larger than those of chemical solution-derived powders. The micrograph shows sharp distributions of particles with an average particle size below 100 nm. The morphology is apparently different from the micrometer-scale

particle and/or agglomerations of SrAl2O4 resulting from solid-state reaction method, indicating that organic components in starting materials can efficiently prevent the formation of larger agglomerations [10].

Fig. 2. Excitation (a), emission (b), absorption (c) spectra, and the plot of the transformed Kubelka-Munk function vs. photon energy (d) for SrAl2O4:Eu2+, Dy3+ phosphor after annealing at 1100 °C for 240 min in Ar - 5 % H2 gas.

Fig. 2a shows an excitation spectrum of SrAl2O4:Eu2+, Dy3+ phosphor annealed at 1100 °C, monitored from 300 nm and 500 nm. A wide-ranging band with a maximum at around 360 nm is observed. The excitation spectrum was obtained by monitoring the emission of the Eu2+ 4f ^ 5d transitions, and the peak at about 520 nm resulting in a green emission [11]. Fig. 2b shows the emission spectrum of the SrAl2O4:Eu2+, Dy3+ phosphor monitored for 360 nm excitation. A broad emission peak is observed at around 515 nm, revealing the characteristic emission of Eu2+ activators (4f65d1 ^ 4f7). The slight blue shift in the emission band from 520 nm to 515 nm, similar to that of the phosphors obtained from solid-state reaction, may be attributed to the changes in the crystal field around Eu2+ arising from the nano-sized particles. Zhang et al. [12] reported that the particles with nanometer size make the surface energy increase, which causes variation of the crystal field around the local environment of Eu2+. Since the excited 4f65d1 configuration of Eu2+ ion is greatly sensitive to the lattice environment and 5d electrons can be strongly coupled with the lattice, the mixed states of 4f and 5d are split by the crystal field, which may lead to the shift of the emission

/15 11:

Seung Hwangbo et al.

peak toward shorter wavelength. Fig. 2c shows the absorption spectra of the pure-TiO2 photoelectrode and the TiO2 photoelectrode doped with 5 wt.% of SrAl2O4:Eu2+, Dy3+ phosphor. A comparison of the spectra indicates that phosphor-doped TiO2 can improve light absorption. The absorption spectra display that the pure-TiO2 coating has a lower absorption ability than SrAl2O4:Eu2+, Dy3+ phosphor-doped TiO2 coating, specifically for the wavelength range between 350 to 800 nm. These spectra cover a UV-Vis-IR region. Therefore, the doping of SrAl2O4:Eu2+, Dy3+ phosphor was found to decrease the light loss and increase photon energy. The energy band gap (Eg) of the samples was calculated from the plots of Kubelka-Munk function vs. the energy of exciting light [13]. From Fig. 2d, the Eg of pure-TiO2 and phosphor-doped TiO2 was obtained to be 3.2 eV and 3.02 eV, respectively. Thus, the phosphor doping of TiO2 induced a shift of the absorption edge slightly towards the visible spectral range.

Fig. 3. Cross-sectional FE-SEM image of SrAl2O4: Eu2+, Dy3+ phosphor-doped TiO2 layer (a), and I-V curves of DSSCs with and without nano-phosphor in TiO2 photoelectrode (b).

In order to evaluate the effect of SrAl2O4:Eu2+, Dy3+ phosphor in TiO2 electrode on the efficiency of DSSC, the working electrode was prepared by ESD using a precursor solution containing 5 wt.% of SrAl2O4:Eu2+, Dy3+ phosphor in TiO2 powder. Fig. 3a displays a cross-sectional image of the SrAl2O4:Eu2+, Dy3+ doped-TiO2 layer on a FTO substrate. The thickness of the film is uniform at ^8 |am, as determined by the

cross-sectional image. The film does not contain any aggregations of TiO2 particles, which are dispersed homogeneously.

Fig. 3b shows the photovoltaic I-V characteristics of the pure-TiO2-based DSSC and SrAl2O4:Eu2+, Dy3+ phosphor-doped TiO2-based DSSC. The light-to-electric-energy conversion efficiency (n) of the DSSC systems with the pure-TiO2 photoelectrode as well as SrAl2O4:Eu2+, Dy3+ phosphor-doped TiO2 photoelectrode were calculated using equation 1, with the values of short-circuit photocurrent density (Jsc, mA/cm2), the open circuit voltage (Voc, V), the fill factor (FF) taken from the I-V characteristics:

n (%) = JSc x Voc x

where I is the intensity of the incident light. The FF is defined by equation 2:

Jmax ((Vmax )

Jsc (Voc )

where Jmax and Vmax are the photocurrent density and the photovoltage, respectively, in the I-V curve at the point of maximum power output. Fig. 3b and Table 1 show that the DSSC with undoped TiO2 have the following photovolatic parameters: Jsc = 12.3 mA/cm2, Voc = 0.65 V, Jmax = 8.6 mA/cm2, Vmax = 0.48 V, FF = 51.6 % and n = 4.13 %. The DSSC containing the SrAl2O4:Eu2+, Dy3+ phosphor as a dopant has the improved photovoltaic parameters: Jsc = 14.1 mA/cm2, Voc = 0.78 V, Jmax = 12 mA/cm2, Vmax = 0.6 V, FF = 65.7 % and n = 7.20 %. The pure-TiO2 device efficiency of 4.13 %, is improved to 7.20 % for the SrAl2O4:Eu2+, Dy3+ phosphor-doped-TiO2 layer. when the phosphor powder was inserted, the number of photons increased, hence, the possibility of photon and dye molecule interactions increased, whereas the DSSC with the pure-TiO2 layer had lower Jsc and Voc because of lower number of excitations [6]. The improvement of the photovoltaic performances shows that the SrAl2O4:Eu2+, Dy3+ phosphor transforms UV to Vis and near-IR light which the N-719 dye can absorb effectively, increasing the harvested sunlight and improving the efficiency of the DSSC.

Table 1. Performance of DSSCs based on different photoelectrodes.

Photoelectrode Jsc Voc Jmax Vmax FF n

_(mA/cm2) (V) (mA/cm2) (V) (%) (%)

Undoped TÍO2 12.3 0.65 8.6 0.48 51.6 4.13

Phosphor-doped TÍO2 14.1 0.78 12 0.6 65.7 7.20

4. Conclusions

Down-conversion SrAl2O4 phosphor was fabricated by using chemical solution method to improve the power transformation efficiency of DSSCs by means of solar spectral transformation. 5 wt.% of SrAl2O4:Eu2+, Dy3+ phosphor-doped-TiO2 photoelectrode on FTO substrate was prepared by the ESD process. The pure-TiO2 device efficiency of 4.13 %, was increased to 7.20 % for the SrAl2O4:Eu2+, Dy3+ phosphor-doped-TiO2 layer. The photovoltaic performances show that the SrAl2O4:Eu2+, Dy3+ phosphor converts UV to Vis and near-IR light which the N719 dye can absorb effectively, increasing the harvested sunlight and improving the efficiency of the DSSC.


This work was supported in part by research funds from Nambu University, 2013.


[3] Gratzel M., Nature, 414 (2001), 338.

[4] Hagfeldt A., Boschloo G., Sun L., Kloo L., Pettersson H., Chem. Rev., 110 (2010), 6595.

[5] Gratzel M., Nat. Mater., 13 (2014), 838.

[6] Hong C., Ko H., Han E., Yun J., Park K., Nanoscale Res. Lett., 8 (2013), 219.

[7] He W., Timur A.S.H., Kim H.K., Hwang Y.H., J. Phys. Chem. C, 117 (2013), 17894.

[8] Kim S.G., Kim J.Y., Kim H.J., Thin Solid Films, 376 (2000), 110.

[9] Haranath D., Shanker V., Chander H., Sharma P., J. Phys. DAppl. Phys., 36 (2003), 2244.

[10] Han C.H., lee H.S., Han S.D., B. Korean Chem. Soc., 29 (2008), 1495.

[11] Escribano P., Marchal M., Sanjuan M.L., Alonso-Gutierrez P., Julian B., Cordoncillo E., J. Solid State Chem., 178 (2005), 1978.

[12] Zhang R., Han G., Zhang L., Yang B., Mater. Chem. Phys., 113 (2009), 255.

[13] Hreniak A., Sikora A., Iwan A., Int. J. Mater. Chem., 4 (2014), 15.

[1] O'Regan B., Gratzel M., Nature, 353 (1991), 737.

[2] He W., Timur A.S.H., Kim H.K., Hwang Y.H., IOP

Conf.-Mater. Sci. Eng., 54 (2014), 012025.

Received 2014-07-12 Accepted 2014-12-15