Trace amount Ca-doped TiO2 films were synthesized by the hydrothermal method and applied as photoanodes of dye-sensitized solar cells (DSSCs). To prepare Ca-doped TiO2 film electrodes, several milliliters of Ca(NO3)2 solution was added in TiO2 solution during the hydrolysis process. The improvements of DSSCs were confirmed by photocurrent density-voltage (J-V) characteristics, electrochemical impedance spectroscopy (EIS) measurements. Owing to the doping effect of Ca, the Ca-doped TiO2 thin film shows power conversion efficiency of 7.45% for 50 ppm Ca-doped TiO2 electrode, which is higher than that of the undoped TiO2 film (6.78%) and the short-circuit photocurrent density (Jsc) increases from 13.68 to 15.42 mA·cm−2. The energy conversion efficiency and short-circuit current density (Jsc) of DSSCs were increased due to the faster electron transport in the Ca-doped TiO2 film. When Ca was incorporated into TiO2 films, the electrons transport faster and the charge collection efficiency ηcc is higher than that in the undoped TiO2 films.
1. Introduction
Dye-sensitized solar cells (DSSCs) based on mesoporous nanocrystalline TiO2 film have achieved photoelectric conversion efficiency (η) up to 13%. In order to develop high performance of DSSCs and commercialize successfully, many nanocrystalline semiconductors such as TiO2 [1], ZnO [2], and SnO2 [3] have been used as photoanode materials. Among them, TiO2 has been proven to be the best semiconductor electrode material due to its high chemical stability [4], excellent charge transport capability, and ideal position of the conduction band edge. It is known to be one of the main components of DSSCs and plays a key role in determining the performance of DSSCs.
In recent years, Doping has been considered as a promising way to improve the properties of TiO2 photoanode. TiO2 films doped with metal and nonmetal have been extensively researched, such as Mg-doping [5], La-doping [6], Nb-doping [7], Ta-doping [8], and N-doping [9], which may increase the photoelectric conversion efficiency. In all the applications mentioned above, the TiO2 films were doped at very high levels (from 0.1% to 10%). However, few studies have been reported about TiO2 doped at parts per million (ppm) level applied as photoanode of DSSCs. Xie et al. [10] have found that trace amount of Cr-doping TiO2 films could improve the efficiency of DSSCs. The improvement was ascribed to Cr additions offers more electrons for TiO2 and increases the property of electron transport for DSSCs. As we all know, doping semiconductors at parts per million (ppm) level is the most common approach for enhancing the Fermi energy level of semiconductors and then increasing their conductivity (for instance, Si).
In this paper, a series of Ca-doped TiO2 films were synthesized by hydrothermal method and were successfully applied as the photoanode materials in DSSCs, and the short-circuit current densities (Jsc) and photoelectric conversion efficiencies of DSSCs were found to be increased by trace amount doping in TiO2. The change in performance of DSSCs employing Ca-doped TiO2 films with different concentrations was obvious. We can conclude that the intrinsic increases in the photocurrent and photoelectric conversion efficiency are primarily related to faster electron transport in the Ca-doped TiO2 film. The effects caused by Ca doping on electron collection, transfer, and recombination of the DSSCs are discussed below.
2. Experimental Section2.1. Preparation of Undoped TiO2 and Ca-Doped TiO2 Pastes
Titanium isopropoxide (TTIP) was used as Ti precursors and calcium nitrate (Ca(NO3)2·4H2O) was the Ca sources. Pure TiO2 and Ca-doped TiO2 pastes were synthesized by a hydrothermal treatment method. The hydrothermal solutions were synthesized as follows [11].
2.1 g acetic acid was added dropwise into 10 mL of TTIP. Subsequently, the mixture was added to 50 mL of deionized water mixed with different amounts of Ca(NO3)2·4H2O (Ca/TiO2 molar ratio: undoped, 20 ppm, 50 ppm, 70 ppm, and 100 ppm) with rapid stirring for 1 h. Then, 0.68 mL of nitric acid was added to the obtained mixture solution. After continuous stirring at 80°C for 2~3 h, a transparent mixture solution was obtained.
The transparent mixture solution was filtered to remove insoluble impurities and transferred into an autoclave at 220°C for 12 h. After cooling to room temperature, 0.4 mL of nitric acid was added into the colloid, using ultrasonicator to disperse.
The colloid was concentrated to 20 mL by rotary evaporator. Finally, PEG and Triton X-100 were added to form TiO2 paste. The obtained pastes were denoted as undoped TiO2, Ca-1: 20 ppm Ca-doped TiO2, Ca-2: 50 ppm Ca-doped TiO2, Ca-3: 70 ppm Ca-doped TiO2, and Ca-4: 100 ppm Ca-doped TiO2.
Following the procedure described, a total of five pastes with different concentrations of Ca were prepared.
2.2. Fabrication of Photoelectrodes and DSSCs
The FTO glass was used as substrate after careful cleaning. The pure TiO2 and Ca-doped TiO2 pastes were coated onto FTO substrates using a doctor-blade method, respectively. Next, the photoelectrodes were sintered at 500°C for 30 min to obtain the mesoporous TiO2 film photoelectrodes. The thicknesses of these films were around 8 μm measured with a TalyForm S4C-3D profilometer. They were dipped into a 0.5 mM N719 dye solution for 24 h at room temperature, and the excessive dye was washed away by using ethanol, followed by drying at 60°C. The platinum-coated FTO was used as the counter. A drop of electrolyte solution was injected into the photoelectrode and then the counter was clamped onto the photoelectrode; the electrolyte solution consisted of 0.05 mM LiI, 0.03 M I2, 0.1 M PMII (1-methyl-3-propyl imidazolium iodide), 0.1 M GNCS, and 0.5 M TBP in mixed solvent of acetonitrile and PC (volume ratio: 1/1). A sandwich-type DSSC configuration was fabricated.
2.3. Measurements
Photovoltaic measurements were performed by a CHI660C electrochemical workstation (CH Instruments, Shanghai, China) at room temperature. The irradiated area of each cell was kept at 0.25 cm2 by using a light-tight metal mask. Electrochemical impedance spectroscopy (EIS) technique [12] was employed to investigate the electron transport in DSSCs.
3. Results and Discussion3.1. J-V Characteristics
The photocurrent density-voltage (J-V) characteristics of the DSSCs based on the pure TiO2 film photoelectrodes and Ca-doped TiO2 film photoelectrodes are shown in Figure 1. The average performance characteristics obtained from multiple cells with the same Ca content were summarized in Table 1, which shows the correlation between the photovoltaic performance parameters and the Ca content in the TiO2. The best photovoltaic performance was obtained from 50 ppm Ca-doped TiO2. Obviously, as can be seen from the graph, the energy conversion efficiency (η) went up with the increase of Ca content, which was attributed to the enhancement of the short-circuit current density (Jsc). The Jsc of DSSCs based on 50 ppm Ca-doped TiO2 was 15.42 mA·cm−2, which was 12.7% higher than that of undoped cells. The energy conversion efficiency of 7.58% was achieved for cells based on 50 ppm Ca-doped TiO2 electrode, which accounts for 9.88% higher than that of undoped cells. The effect on open-circuit voltage (Voc) and fill factor as a result of such a little amount of Ca-doping was negligible. The energy conversion efficiency (η) increases gradually with the increase of Ca content and reaches an optimum value coinciding with Ca quantity of 50 ppm. However, the higher Ca amounts (>50 ppm) cause electron scattering and trap electrons which increase dark current. As a result, the energy conversion efficiency of DSSCs begin to fall.
Performance of DSSCs based on undoped and Ca-doped TiO2 photoanodes.
DSSCs
Doping/ppm
Voc/mV
Jsc/mAcm2
FF
η/%
Rct/Ω
ηcc
Undoped
0
695
13.68
0.71
6.78
7.34
0.75
Ca-1
20
686
14.78
0.69
7.00
7.32
0.73
Ca-2
50
691
15.42
0.70
7.45
5.24
0.80
Ca-3
70
684
14.82
0.70
7.00
8.09
0.72
Ca-4
100
686
14.32
0.69
6.84
7.30
0.73
J-V characteristics of DSSCs based on the undoped and Ca-doped TiO2 photoanodes.
3.2. Electrochemical Impedance Spectroscopy Analysis of DSSCs
To investigate the difference of the charge transport properties between pure DSSCs and Ca-doped DSSCs, we performed electrochemical impedance spectroscopy (EIS) analysis. EIS has been widely employed to investigate the electron transport in DSSCs, for example, measuring the respective time constants for charge combination and for the combined processes of charge collection. From the measured spectra of EIS, we can get reliable value of the parameter. Figure 2 shows the EIS spectra of pure DSSCs and Ca-doped DSSCs, the impedance spectra of DSSCs based on the pure TiO2 and Ca-doped TiO2 were measured from 0.1 to 105 Hz in the illumination at the applied bias of Voc. The spectra are composed of two semicircles: the small semicircle in the high frequency range of 103 to 105 Hz fitted to a charge transfer resistance (Rct) at the interfaces of the redox electrolyte/Pt counter electrode and FTO/TiO2 and the large semicircle in the frequency range of 1 to 103 Hz fitted to a transport resistance (Rw), which is related to the charge transport resistance of the accumulation/transport [13] of the injected electrons within TiO2 film and the charge transport resistance at the TiO2/redox electrolyte interfaces. This large semicircle is the major concern here. As shown in Figure 2, the large semicircle got smaller with the increase of Ca in TiO2 films. This change reflected the acceleration of electron transport process in TiO2 photoanode. The modeled internal resistances of the DSSCs based on five different electrodes are exhibited in Table 2, in which tr is the electron transport time, tc is the electron lifetime, Rw is charge transport resistance, and ηcc is the charge collection efficiency of DSSCs. The apparent value of ηcc can be estimated on the basis of the tr and tc data from the following [14]:
(1)ηcc=1-trtc.
The mean electron life time (tr), the mean electron transit time (tc), and the charge collection efficiency (ηcc) of DSSCs based on undoped and Ca-doped TiO2 photoanodes.
DSSCs
Doping/ppm
Rw/Ω
tr/ms
tc/ms
ηcc
Undoped
0
7.36
0.0350
0.0088
0.75
Ca-1
20
7.32
0.0290
0.0077
0.73
Ca-2
50
5.24
0.0294
0.0059
0.80
Ca-3
70
7.27
0.0290
0.0076
0.72
Ca-4
100
8.09
0.0290
0.0080
0.73
EIS of DSSCs based on the undoped and Ca-doped TiO2 photoanodes measured in the illumination at the applied bias of Voc (a) Nyquist plots and (b) Bode phase plots.
The electron transport time constants for Ca-doped TiO2 films decrease, which indicates the electrons transport faster in the Ca-doped TiO2 films than the undoped TiO2 films. This enhanced the charge collection efficiency ηcc and led to higher current density (Jsc) of DSSCs. The electron life time constants for the Ca-doped TiO2 films also slightly decrease. The electron lifetime in DSSCs is determined by the characteristic frequency peak in the low frequency (fmax) according to the following equation [15]:
(2)τe=12πfmax.
The shorter electron life time indicates the faster recombination rate in the Ca-doped TiO2 films, and that could be attributed to impurities of Ca-doping, which acts as a charge trapping site for the electron-hole recombination. The electron lifetime decreases slightly, so we concluded that less photogenerated electrons are captured by empty trap states in the Ca-doped TiO2 films [16], and this result favors the electron transport. The improvement of electron transport ability was helpful to increase the short-circuit current density (Jsc), resulting in higher conversion efficiency.
4. Conclusion
In summary, the Ca-doping TiO2 nanoparticles were successfully applied as the photoanode material in DSSCs. By comparing the Ca-doping TiO2 with undoping TiO2, a faster electron transport and shorter lifetime existed for the Ca-doping DSSCs. Moreover, the higher electron transport rates of Ca-doped TiO2 photoanode can improve the charge collection efficiency and thus lead to higher short-circuit photocurrent density of DSSCs. The best photovoltaic performance was obtained from 50 ppm Ca-doping with the conversion efficiency of 7.45%. This value was 9.88% higher than that of the undoped device. The short-circuit current density (Jsc) was increased due to the faster electron transport in the Ca-doped TiO2 film. We can conclude that Ca-doped TiO2 is a better photoanode material and a more promising alternative for high efficient DSSCs than pure TiO2.
Conflict of Interests
The authors do not have any conflict of interests in their submitted paper.
Acknowledgments
We acknowledge the supported of the National Natural Science Foundation of China (Grant no. 11364014), the National Science Foundation of Hainan Province, China (Grant no. 113005), and Physics Graduate Project of Hainan Normal University (20140092102).
YuH.ZhangS. Q.ZhaoH. J.XueB.LiuP.WillG.High-performance TiO2 photoanode with an efficient electron transport network for dye-sensitized solar cells200911336162771628210.1021/jp90419742-s2.0-70149121043LiF. R.WangG. C.JiaoY.Efficiency enhancement of ZnO-based dye-sensitized solar cell by hollow TiO2 nanofibers201461151923WangH. K.RogachA. L.Hierarchical SnO2 nanostructures: recent advances in design, synthesis, and applications201426112313310.1021/cm40182482-s2.0-84892590368QuintanaM.EdvinssonT.HagfeldtA.BoschlooG.Comparison of dye-sensitized ZnO and TiO2 solar cells: studies of charge transport and carrier lifetime200711121035104110.1021/jp065948f2-s2.0-33847652173ZhangC. N.ChenS. H.MoL.-E.HuangY.TianH.HuL.HuoZ.DaiS.KongF.PanX.Charge recombination and band-edge shift in the dye-sensitized Mg2+-doped TiO2 solar cells201111533164181642410.1021/jp20243182-s2.0-80051946437ZhangJ. Y.ZhaoZ. Y.WangX. Y.YuT.GuanJ.YuZ.LiZ.ZouZ.Increasing the oxygen vacancy density on the TiO2 surface by La-doping for dye-sensitized solar cells201011443183961840010.1021/jp106648c2-s2.0-78650390179YaoQ. H.LiuJ. F.PengQ.WangX.LiY.Nd-doped TiO2 nanorods: preparation and application in dye-sensitized solar cells20061573774110.1002/asia.2006001952-s2.0-33845243673LiuJ.YangH. T.TanW. W.ZhouX.LinY.Photovoltaic performance improvement of dye-sensitized solar cells based on tantalum-doped TiO2 thin films201056139640010.1016/j.electacta.2010.08.0632-s2.0-78649983191GuoW.WuL. Q.ChenZ.BoschlooG.HagfeldtA.MaT.Highly efficient dye-sensitized solar cells based on nitrogen-doped titania with excellent stability20112192-318018710.1016/j.jphotochem.2011.01.0042-s2.0-79954665487XieY. N.HuangN.YouS. J.LiuY.SeboB.LiangL.FangX.LiuW.GuoS.ZhaoX.-Z.Improved performance of dye-sensitized solar cells by trace amount Cr-doped TiO2 photoelectrodes20132241016817310.1016/j.jpowsour.2012.09.0782-s2.0-84867748829TongZ. F.PengT.SunW. W.Introducing an intermediate band into dye-sensitized solar cells by W6+ doping into TiO2 nanocrystalline photoanodes2014245AdachiM.SakamotoM.JiuJ.OgataY.IsodaS.Determination of parameters of electron transport in dye-sensitized solar cells using electrochemical impedance spectroscopy200611028138721388010.1021/jp061693u2-s2.0-33746922353WangQ.MoserJ.-E.GrätzelM.Electrochemical impedance spectroscopic analysis of dye-sensitized solar cells200510931149451495310.1021/jp052768h2-s2.0-23844443970WangK.-P.TengH.Zinc-doping in TiO2 films to enhance electron transport in dye-sensitized solar cells under low-intensity illumination200911419489949610.1039/b912672d2-s2.0-70350055054ParkN.-G.van de LagemaatJ.FrankA. J.Comparison of dye-sensitized rutile- and anatase-based TiO2 solar cells200010438898989942-s2.0-0033634006DuanY. D.FuN. Q.LiuQ. P.FangY.ZhouX.ZhangJ.LinY.Sn-doped TiO2 photoanode for dye-sensitized solar cells2012116168888889310.1021/jp212517k2-s2.0-84860355898