English

• 1.   Introduction

  Titanium dioxide at nanometer scale has been widely studied in the electrochemical and photochemical fields in recent years. Relative topics include photocatalysis [1, 2], sensor [3], perovskite solar cell [4, 5], and other solar cells [6, 7]. Since the report of dyesensitized solar cells (DSSCs) by O'Regan and M. Grätzel in 1991, TiO₂ has been used as the core material for the photoelectrode films of DSSCs [6]. The excited electrons from molecular light absorbers (such as dye sensitizers) were injected into the large band gap of TiO₂ nanoparticles on a transparent conducting glass substrate under light illumination [8]. The effective electron diffusion in the TiO₂ film was obtained by controlling the electron trapping/releasing occurring on the defects, grain boundaries, and surface states [8, 9]. The defect level and the number of grain boundaries must be minimized to suppress the loss of electrons through the recombination or back reaction. To anchor a large number of dye molecules, the TiO₂ nanoparticles need to be optimized, such as decreasing the size of the TiO₂ nanoparticles, increasing the surface area and other improvements. However, more defects and grain boundaries can be generated with the decreasing of the average particle size. Therefore, it has been reported that the particle size in the range 15–25 nm is optimal [10–12].

  As is reported [13, 14], there are three crystal forms for TiO₂: rutile, anatase, and brookite. The anatase TiO₂ with exposed 101 planes is critical for a good bonding interaction between TiO₂ and ruthenium-based dye molecules, which enables the dye molecules to form a chemisorbed monolayer with high density on the particle surface and then promotes the efficient injection of electrons into semiconductor. Different approaches have been developed to synthesize TiO₂ nanoparticles, such as microemulsion, sol-gel, hydrothermal, etc. [15–17]. Hydrothermal reaction is widely used to synthesize different types of TiO₂ nanoparticles with various crystallographic and morphological properties [18]. Previously, the relationship between crystallization dynamics of TiO₂ nanoparticles and hydrothermal reaction conditions has been reported, including reaction time, temperature [19], pH [20], solvent [21], additives [22], etc.

  In our studies, to obtain TiO₂ nanoparticles with high crystallinity, narrow size distribution, large specific surface area, the hydrothermal method under different conditions has been investigated. Diethylene glycol (DEG) was added as a surfactant to modify the morphology of TiO₂ in the hydrothermal reaction (hereafter referred to as TiO₂-DEG particles). The TiO₂-DEG particles were used in fabrication of photoelectrode of DSSCs. XRD, nitrogen adsorption analysis, SEM and HRTEM were employed to characterize the structure and morphology of the synthesized TiO₂ nanoparticles and the resultant photoelectrodes. Then photovoltaic performances of DSSCs were evaluated and discussed via the measurement of the analysis of J-V curves, opencircuit voltage decay (OCVD), and electrical impedance spectra (EIS).

  2.   Results and discussion

  2.1   Characterization of TiO₂ nanoparticles

  The structure of TiO₂-DEG materials synthesized under different conditions was characterized by wide-angle XRD technique, as shown in Fig. 1. The XRD patterns of TiO₂ show peak positions at 25.3, 37.8, 48.1, 53.9 and 55.1°, which are consistent with the standard powder diffraction pattern of anatase-TiO₂ (JCPDS file No. 78-2486), indicating the formation of anatase phase with tetragonal structure. All the diffraction peaks of TiO₂ materials after hydrothermal and sintered treatment are sharp and intense, showing the high crystallinity of the TiO₂. This suggests the synthesized TiO₂ nanoparticles expose more (101) crystal planes, which could facilitate the efficient injection of electrons in dye molecules into the semiconductors.

  图 1

  

  图 1  XRD patterns of the TiO₂ powder with different molar ratios sintered at 500 ℃ for 3 h.

  Figure 1.  XRD patterns of the TiO₂ powder with different molar ratios sintered at 500 ℃ for 3 h.

  下载: 全尺寸图片 幻灯片

  Nitrogen adsorption experiments were conducted to characterize the surface area (S_BET), the pore size (D_a) distribution and the pore volume (V_T) of TiO₂ nanoparticles, and the results are listed in Table 1. The Brunauer-Emmett-Teller (BET) measurement showed that the S_BET of TiO₂-DEG (1:2), TiO₂ (DEG free) and P25 are 96.74, 78.94 and 54.64 m²/g, respectively, which indicated that TiO₂ nanoparticles prepared by hydrothermal method possessed larger surface area than P25 nanoparticles, and the S_BET values of the resultant TiO₂-DEG nanoparticles are larger than the other two. The pore size distribution and the pore volume could be derived from the Barrett-Joyner-Halenda (BJH) model resulting in a D_a value of 9.05, 9.07 and 10.52 nm and a V_T value of 0.347, 0.351 and 0.250 cm³/g, for TiO₂-DEG (1:2), TiO₂ (DEG free) and P25, respectively. The average pore sizes (D_a) were found to be ~10 nm which was much bigger than the ionic radius of I^-/I₃^-(0.2/0.5 nm) and would allow effective diffusion of electrolyte. The combined results from the S_BET and D_a indicated that the physical properties of TiO₂-DEG nanoparticles were more suitable for photoelectrode materials of DSSCs.

  表 1

  

  表 1  Specific surface area (S_BET), average pore size (D_a) and total pore volume (V_T) of the different TiO₂ determined from gas adsorption analysis.

  Table 1.  Specific surface area (S_BET), average pore size (D_a) and total pore volume (V_T) of the different TiO₂ determined from gas adsorption analysis.

  下载: 导出CSV

  SEM and HRTEM were performed to investigate the influence of different ratios of DEG added in hydrothermal reaction on the particles and the corresponding TiO₂ nanofilms. Images of TiO₂ nanoparticles are shown in Fig. 2(a–c) and TiO₂ namofilms are shown in Fig. 2(d–f), representing TiO₂-DEG (1:2), TiO₂ (DEG free) and P25, respectively. The SEM image reveals the good connectivity between the adjacent nanoparticles for TiO₂-DEG nanoparticles compared with the TiO₂ (DEG free) and the P25 nanoparticles. It can be seen that the TiO₂-DEG nanoparticles are, as is wellestablished, in 15–25 nm size and uniformly distributed, and the nanoparticles for TiO₂ (DEG free) and P25 are not in uniform sizes. It has been shown that TiO₂ (DEG free) nanoparticles are closely connected and possess smaller sizes, compared to TiO₂-DEG nanoparticles, but they easily agglomerate into larger conglobation as shown in Fig. 2b, which is unsuitable for dye adsorption. To yield uniform and porous layers, the hydrothermal reaction procedure was modified by the addition of DEG with abundant ether bonds and hydroxyl groups, which effectively decrease the inter molecular interactions among the TiO₂ nanoparticles and the chance of forming bulky agglomeration. The hydroxyl and ether groups of DEG change the distribution of hydroxyl on the surface of TiO₂ nanoparticles, and contribute to the dye molecules adsorption. The photoelectrode films prepared directly on FTO by screen printing are shown in Fig. 2(d–f). It can be seen that the TiO₂-DEG films are smoother compared to the TiO₂ (DEG free) and P25 films. This structure of photoelectrode films are expected to enhance the diffusion of charge carriers and the electrolyte penetration in DSSCs. Fig. 3 shows the HRTEM of the sample TiO₂-DEG (1:2). The low power TEM image (Fig. 3a) shows that this sample is monodispersed TiO₂ particles and no agglomeration occurs. The HRTEM image (Fig. 3b) shows clearly that the (101) plane is exposed and the lattice is distributed in order in the surface of TiO₂, which makes it easy for light harvesting. Combined results from XRD, BET, SEM and HRTEM analysis suggested that the photoelectrodes involving TiO₂-DEG material with high crystallinity of anatase phase and larger S_BET, have smooth and uniform nanofilms, which are expected to adsorb more dye molecules and have better performance in DSSCs compared to those based on the TiO₂ (DEG free) and P25.

  图 2

  

  图 2  SEM images (a) TiO₂-DEG (1:2) nanoparticles; (b) TiO₂ (DEG free) nanoparticles; (c) P25; (d) photoelectrode film based on TiO₂-DEG (1:2) nanoparticles; (e) photoelectrode film based on TiO₂ (DEG free) nanoparticles; (f) photoelectrode film based on P25.

  Figure 2.  SEM images (a) TiO₂-DEG (1:2) nanoparticles; (b) TiO₂ (DEG free) nanoparticles; (c) P25; (d) photoelectrode film based on TiO₂-DEG (1:2) nanoparticles; (e) photoelectrode film based on TiO₂ (DEG free) nanoparticles; (f) photoelectrode film based on P25.

  下载: 全尺寸图片 幻灯片

  图 3

  

  图 3  HRTEM images of TiO₂-DEG (1:2) nanoparticles (a) monodispersed TiO₂ particles (b) lattice fringes of anatase.

  Figure 3.  HRTEM images of TiO₂-DEG (1:2) nanoparticles (a) monodispersed TiO₂ particles (b) lattice fringes of anatase.

  下载: 全尺寸图片 幻灯片

  2.2   UV–vis spectroscopy

  Fig. 4 shows the UV–vis absorption spectra of the desorbed dye using 0.5 mol/L NaOH solution and thus the amounts of dye adsorbed on the surfaces of the photoelectrodes were calculated based on the UV–vis absorption spectra and shown in Table 2. The absorption peaks for N719 are observed in the range of 375–515 nm [27]. Dye adsorption amount reached the highest for the TiO₂-DEG (1:2) film. The second highest dye adsorption amount was obtained for the TiO₂-DEG (1:3) film, followed by TiO₂ (DEG free) and TiO₂-DEG (1:1). The lowest dye adsorption amount was for P25 film. The results are in well agreement with those obtained from above analysis.

  图 4

  

  图 4  UV–vis absorption spectra of the N719 dye desorbed from TiO₂ film electrodes made from TiO₂ synthesized by different additive amounts of DEG.

  Figure 4.  UV–vis absorption spectra of the N719 dye desorbed from TiO₂ film electrodes made from TiO₂ synthesized by different additive amounts of DEG.

  下载: 全尺寸图片 幻灯片

  表 2

  

  表 2  Photovoltaic parameters with different TiO₂ photoelectrodes.

  Table 2.  Photovoltaic parameters with different TiO₂ photoelectrodes.

  下载: 导出CSV

  2.3   Photovoltaic performance

  The N719 sensitized solar cells involving the prepared TiO₂-DEG nanoparticles with different molar ratios and P25 (all in the similar thickness of 11 ± 2 μm and active area of 0.16 cm²) as the photoelectrode films were fabricated. Fig. 5 shows the photocurrent density-voltage (J-V) characteristics of the DSSCs based on different photoelectrodes. Photovoltaic properties, such as open circuit voltage (V_oc), short circuit current density (J_sc), fill factor (FF), power conversion efficiency (η) base on Fig. 5, and for a comparison, the calculated dye loading amounts based on Fig. 4 are all collected in Table 2.

  图 5

  

  图 5  J-V curves for DSSCs fabricated from different TiO₂ photoelectrodes.

  Figure 5.  J-V curves for DSSCs fabricated from different TiO₂ photoelectrodes.

  下载: 全尺寸图片 幻灯片

  It can be seen from Table 2 that the solar cell based on TiO₂-DEG photoelectrodes exhibited better performance than those involving the TiO₂ (DEG free) and P25 and the best performance was realized for the TiO₂-DEG cell, of which the photoelectrode was fabricated with the TiO₂ nanoparticles prepared by using the 1:2 molar ratio of TiO₂ to DEG. The η value increased while the DEG molar ratio was increased from 1:3 to 1:2, and declined at 1:1. However, these values were all greater than the cells based on P25. These results confirmed the influence of DEG on the properties of synthesized TiO₂ photoelectrodes and the performance of DSSCs. A combination of the results obtained above suggested that the increase of the surface area and the uniform porous photoelectrode films could enhance the ability of dye adsorption and lead to higher photocurrent for DSSCs, which are important for the optimization of the performance of DSSCs.

  2.4   Open-circuit voltage decay (OCVD) measurement

  OCVD technique has been employed as a convincing method to provide valuable information of the electron recombination velocity in DSSCs [24, 25]. The simulated solar light illuminated at DSSCs, resulting in a steady-state voltage. It indicated that electron injection and electron recombination reached equilibrium under the simulated solar light, and then the decay of photovoltage was detected by the potentiost after the removal of the light illumination. The rate of the photovoltage decay reflected the decrease of the electron concentration which is mainly caused by the charge recombination [26]. Fig. 6 shows the OCVD decay curves of the DSSCs based on the TiO₂-DEG (1:2), TiO₂ (DEG free), and P25 photoelectrodes. It is observed that the OCVD response of the DSSC based on the TiO₂-DEG was much slower than those of the cells fabricated using TiO₂ (DEG free) and P25, especially in the short time domain (within 8 s). Slower decay means higher electron concentration, less charge recombination, and longer electron lifetime in DSSCs. The slower OCVD for the cell fabricated with TiO₂-DEG nanoparticles can be well explained based on the previous results, that the TiO₂-DEG nanoparticles have more dye molecules absorbed, and more photo-generated electrons in the nanofilms.

  图 6

  

  图 6  Open-circuit voltage decay curves of the DSSCs fabricated by different TiO₂ photoelectrodes.

  Figure 6.  Open-circuit voltage decay curves of the DSSCs fabricated by different TiO₂ photoelectrodes.

  下载: 全尺寸图片 幻灯片

  2.5   Electrical impedance spectroscopy (EIS)

  Fig. 7 shows the Nyquist plots obtained by electrical impedance spectroscopy (EIS), and the according values of resistances are shown in Table 3. In general, three semicircles were obtained from the Nyquist plot. The semicircles are assigned to electrochemical reaction at the Pt counter electrode/electrolyte or FTO/TiO₂ (R₁), charge transfer reaction at TiO₂/dye/electrolyte (R₂) and the Nernst diffusion of the electrolyte (R₃) [28, 29]. In the high frequency region (10³–10⁵ Hz), and a relatively smaller semicircle was observed for TiO₂-DEG photoelectrode, which indicated that a lower resistance (0.85Ω) was obtained for the TiO₂-DEG photoelectrode than that for the other two (1.26Ω and 1.69Ω). The middle semicircle (1–10³ Hz) indicates the charge transfer resistance in the photoelectrode [30]. The smaller value (2.00Ω) for the TiO₂-DEG photoelectrode suggested that TiO₂-DEG may contribute to the reduction of charge transfer resistance at the TiO₂/dye/electrolyte interface [31]. Similar behaviors are also observed in the low-frequency region (0.1–1 Hz), implying the facile diffusion of electrolyte through TiO₂-DEG photoelectrode. Eventually, the impedance spectroscopy studies suggested the photoelectrode based on the prepared TiO₂-DEG material would allow fast electron transport and diffusion as expected, and also a well diffusion of electrolyte, which may arise from the smooth and uniform nanofilms.

  图 7

  

  图 7  Nyquist plots of DSSCs constructed using different TiO₂ photoelectrodes.

  Figure 7.  Nyquist plots of DSSCs constructed using different TiO₂ photoelectrodes.

  下载: 全尺寸图片 幻灯片

  表 3

  

  表 3  The values of resistances of R₁, R₂, R₃ of different DSSC samples.

  Table 3.  The values of resistances of R₁, R₂, R₃ of different DSSC samples.

  下载: 导出CSV

  3.   Conclusion

  TiO₂ nanoparticles synthesized by hydrothermal reaction assisted by the addition of DEG (TiO₂-DEG nanoparticles), were employed for the DSSC fabrication. A detailed investigation of the influence of the different molar ratios of TiO₂: DEG on the resultant cell performance suggested that the best photoelectric performance was achieved when the molar ratio of TiO₂ to DEG is 1:2, with a η of 7.90%. The η values of 7.53% and 6.59% were obtained for cells based on the TiO₂ (DEG free) and P25 photoelectrodes, respectively. Compared with TiO₂ (DEG free) and P25, the TiO₂-DEG nanoparticles and nanofilms showed a significant improvement in the morphology and crystallization properties contributing to the resultant better cell performance. The lager surface area, mono-dispersed particles, uniform porous layers, etc. for the TiO₂-DEG photoelectrode, resulted in more dye adsorption amount, better electrolyte penetration, less trapping of carriers in grain boundaries and surface defects. Furthermore, the OCVD and the EIS results showed longer lifetime of electrons and lower resistance in the fabricated DSSCs based on TiO₂-DEG nanoparticles. Therefore, in this study, an efficient method for the DEG assisted preparation of TiO₂ nanoparticles for use in the fabrication of photoelectrodes of high performance dye-sensitized solar cells has been developed. Further studies on the photophysical process occurring in the TiO₂-DEG photoelectrode are still in progress.

  4.   Experimental

  4.1   Materials

  The chemicals used in the synthesis, including tetrabutyl titanate, isopropanol, absolute ethanol, acetonitrile, ethyl cellulose (EC), terpineol, valeronitrile, and di-tetrabutyl ammonium were of analytical grade and purchased from Aladdin Chemical Regent Co. (Shanghai, China). Other chemicals used include cis-bis(isothiocyanato)bis(2, 2'-bipyridyl-4, 4'-dicarboxylato)ruthenium(Ⅱ), (N719 dye, Solaronix, Switzerland), LiI, I₂, 4-tert-butylpyridine(4-TBP), GuSCN (Heptachroma Solar Tech Co., Ltd., China), and P25 (Degussa AG, Germany).

  4.2   Preparation of TiO₂ nanoparticles

  TiO₂ nanoparticles were prepared as follows: (1) 2 mL HNO₃(65 wt%) was added into a 500 mL three 3-neck flask containing 250 mL distilled water in ice water bath. (2) The mixed solution of 40 mL butyl titanate and 10 mL isopropanol was slowly dropping into the 3-neck flask and then stirred for 1 h, and the aqueous solution slowly changed from milky to clear after stirring. (3) Aqueous ammonia was used to adjust the pH value of the mixture to 11, and the solution turned back to milky. (4) Different amounts of DEG were added to the milky solution (34 mL) in a 100 mL flask with stirring for 1 h, and then the hydrothermal precursor was obtained. The resultant solution was placed in an autoclave and aged at 220 ℃ for 16 h to obtain the uniform TiO₂ nanoparticles. A hierarchical product was obtained after the hydrothermal reaction. The supernatant liquor was discarded and the bottom precipitates were washed and centrifuged with distilled water (three times) and ethanol (two times), respectively, to remove residual organic compounds on nanoparticles surface. The nanoparticles were then calcined at 500 ℃ for 3 h. The molar ratios of TiO₂ to DEG were 1:0, 1:1, 1:2, and 1:3. The obtained samples were denoted as TiO₂-DEG (1:x). And TiO₂ nanoparticles prepared as above without the process 4 was denoted as TiO₂ (DEG free).

  4.3   TiO₂ films and fabrication of DSSCs

  The TiO₂ paste is prepared by Graitzel's method [23]. A 15–16 μm thick TiO₂ nanoparticle film was prepared by screen printing on a clean F-doped SnO₂ (FTO) glass. The films were then heated at 325 ℃ for 5 min, 375 ℃ for 5 min, 450 ℃ for 15 min, and 500 ℃ for 15 min. Finally, the TiO₂ films were treated with an aqueous TiCl₄ solution (50 mmol/L) at 70 ℃ for 30 min, rinsed with ethanol, and heated at 500 ℃ for 30 min. At 110 ℃, the TiO₂ electrodes were immersed in N719 dye solution (0.5 mmol/L) for 24 h. The electrolyte consisted of DMPⅡ, I₂, 4-TBP and GuSCN in acetonitrile and valeronitrile (in the volume ratio of 85/15) was injected into the prepared TiO₂ electrode and then the treated TiO₂ electrodes were assembled with Pt counter electrode into DSSCs.

  4.4   Characterization

  X-ray diffraction (XRD, nickel-filtered by Cu Kα ray) measurement and transmission electron micrograph (HRTEM, Tecnai G2 F20) were performed in order to analyze the phase of samples. The nitrogen adsorption isotherms of TiO₂ materials were measured on a surface area analyser (BET, BEL-Mini). The morphology of TiO₂ was characterized by field-emission scanning electron microscope (SEM, Hitashi, S-4800), operated at an accelerating voltage of 15 kV. To study the loading capacity, the dye attached on the TiO₂ electrode was dissolved in 0.1 mol/L NaOH solution, and its adsorption capability was characterized by using a UV–vis spectrometer (SHIMADATSU). Photocurrent density-voltage (J-V) measurements were obtained by a solar simulator (Zolix) source measurement unit. A Xenon lamp was used as the light source and its light intensity was adjusted to approximating AM 1.5G one sun light intensity by using an NREL-calibrated Si solar cell (National Institute of Metrology), and then the J-V curves were recorded with a computer-controlled digital source meter (Kethily 2400). To show DSSCs' electrochemical properties, electrochemical impedance spectra (EIS) and open circuit voltage decay (OCVD) were obtained by using CHI660 (Chenhua, China).

  Acknowledgments

  This work is supported by National Natural Science Foundation of China (No. 21476162), and China International Science and Technology Project (Nos. 2012DFG41980, S2016G3413).

• 1. [1]

     J. Tian, Y. H. Leng, Z. H. Zhao, et al. , Carbon quantum dots/hydrogenated TiO₂ nanobelt heterostructures and their broad spectrum photocatalytic properties under UV, visible, and near-infrared irradiation, Nano Energy 11(2015)419-427.

  2. [2]

     Momeni M.M., Ghayeb Y.. Photochemical deposition of platinum on titanium dioxide-tungsten trioxide nanocomposites:an efficient photocatalyst under visible light irradiation[J]. J.Mater.Sci.:Mater.Electron., 2016, 27:  1062-1069. doi: 10.1007/s10854-015-3852-z

  3. [3]

     Yan Y.T., Liu Q., Du X.J.. Visible light photoelectrochemical sensor for ultrasensitive determination of dopamine based on synergistic effect of graphene quantum dots and TiO₂ nanoparticles[J]. Anal.Chim.Acta, 2015, 853:  258-264. doi: 10.1016/j.aca.2014.10.021

  4. [4]

     Heo J.H., Han H.J., Kim D., Ahn T.K., Im S.H.. Hysteresis-less inverted CH₃NH₃PbI₃ planar perovskite hybrid solar cells with 18.1%power conversion efficiency[J]. Energy Environ.Sci., 2015, 8:  1602-1608. doi: 10.1039/C5EE00120J

  5. [5]

     Giordano F., Abate A., Baena J.P.C.. Enhanced electronic properties in mesoporous TiO₂ via lithium doping for high-efficiency perovskite solar cells[J]. Nat.Commun., 2016, 7:  10379. doi: 10.1038/ncomms10379

  6. [6]

     O'Regan B., Grätzel M.. A low-cost, high-efficiency solar cell based on dye-sensitized colloidal TiO₂ films[J]. Nature, 1991, 353:  737-740. doi: 10.1038/353737a0

  7. [7]

     Xu J., Wang G.X., Fan J.J.. g-C₃N₄ modified TiO₂ nanosheets with enhanced photoelectric conversion efficiency in dye-sensitized solar cells[J]. J.Power Sources, 2015, 274:  77-84. doi: 10.1016/j.jpowsour.2014.10.033

  8. [8]

     Martsinovich N., Troisi A.. Theoretical studies of dye-sensitised solar cells: from electronic structure to elementary processes[J]. Energy Environ.Sci., 2011, 4:  4473-4495. doi: 10.1039/c1ee01906f

  9. [9]

     Qian J.F., Liu P., Xiao Y.. TiO₂-coated multilayered SnO₂ hollow microspheres for dye-sensitized solar cells[J]. Adv.Mater., 2009, 21:  3663-3667. doi: 10.1002/adma.v21:36

  10. [10]

      Wang Z.B., Tang Q.W., He B.L.. Titanium dioxide/calcium fluoride nanocrystallite for efficient dye-sensitized solar cell.A strategy of enhancing light harvest[J]. J.Power Sources, 2015, 275:  175-180. doi: 10.1016/j.jpowsour.2014.11.006

  11. [11]

      Wang M.K., Chamberland N., Breau L.. An organic redox electrolyte to rival triiodide/iodide in dye-sensitized solar cells[J]. Nat.Chem., 2010, 2:  385-389. doi: 10.1038/nchem.610

  12. [12]

      Lim S.P., Pandikumar A., Lim H.N.. Boosting photovoltaic performance of dye-sensitized solar cells using silver nanoparticle-decorated N S-Co-doped-TiO₂ photoanode[J]. Sci.Rep., 2015, 5:  11922. doi: 10.1038/srep11922

  13. [13]

      Polo A.S., Itokazu M.K., Iha N.Y.M.. Metal complex sensitizers in dye-sensitized solar cells[J]. Coord.Chem.Rev., 2004, 248:  1343-1361. doi: 10.1016/j.ccr.2004.04.013

  14. [14]

      Park N.G., van de Lagemaat J., Frank A.J.. Comparison of dye-sensitized rutile-and anatase-based TiO₂ solar cells[J]. J.Phys.Chem.B, 2000, 104:  8989-8994.

  15. [15]

      Kumar S.G., Rao K.S.R.K.. Polymorphic phase transition among the titania crystal structures using a solution-based approach:from precursor chemistry to nucleation process[J]. Nanoscale, 2014, 6:  11574-11632. doi: 10.1039/C4NR01657B

  16. [16]

      Liu H.Y., Joo J.B., Dahl M.. Crystallinity control of TiO₂ hollow shells through resin-protected calcination for enhanced photocatalytic activity[J]. Energy Environ.Sci., 2015, 8:  286-296. doi: 10.1039/C4EE02618G

  17. [17]

      Qiu B.C., Xing M.Y., Zhang J.L.. Mesoporous TiO₂ nanocrystals grown in situ on graphene aerogels for high photocatalysis and lithium-ion batteries[J]. J.Am. Chem.Soc., 2014, 136:  5852-5855. doi: 10.1021/ja500873u

  18. [18]

      Ong W.J., Tan L.L., Chai S.P.. Highly reactive{001}facets of TiO₂-based composites:synthesis, formation mechanism and characterization[J]. Nanoscale, 2014, 6:  1946-2008. doi: 10.1039/c3nr04655a

  19. [19]

      Liu N., Chen X.Y., Zhang J.L., Schwank J.W.. A review on TiO₂-based nanotubes synthesized via hydrothermal method:formation mechanism, structure modification, and photocatalytic applications[J]. Catal.Today, 2014, 225:  34-51. doi: 10.1016/j.cattod.2013.10.090

  20. [20]

      Barnard A.S., Curtiss L.A.. Prediction of TiO₂ nanoparticle phase and shape transitions controlled by surface chemistry[J]. Nano Lett., 2005, 5:  1261-1266. doi: 10.1021/nl050355m

  21. [21]

      Mosconi E., Selloni A., Angelis F.D.. Solvent effects on the adsorption geometry and electronic structure of dye-sensitized TiO₂:a first-principles investigation[J]. J.Phys.Chem.C, 2012, 116:  5932-5940.

  22. [22]

      Jung H.S., Lee J.K., Lee S., Hong K.S., Shin H.. Acid adsorption on TiO₂ nanoparticles-an electrochemical properties study[J]. J.Phys.Chem.C, 2008, 112:  8476-8480. doi: 10.1021/jp711689u

  23. [23]

      Ito S., Chen P., Comte P.. Fabrication of screen-printing pastes from TiO₂ powders for dye-sensitised solar cells[J]. Prog.Photovolt.:Res.Appl., 2007, 15:  603-612. doi: 10.1002/(ISSN)1099-159X

  24. [24]

      Sun X.H., Liu Y.M., Tai Q.D.. High efficiency dye-sensitized solar cells based on a bi-layered photoanode made of TiO₂ nanocrystallites and microspheres with high thermal stability[J]. J.Phys.Chem.C, 2012, 116:  11859-11866. doi: 10.1021/jp211838g

  25. [25]

      Abdullaha M.H., Rusop M.. Improved performance of dye-sensitized solar cell with a specially tailored TiO₂ compact layer prepared by RF magnetron sputtering[J]. J.Alloys Compd., 2014, 600:  60-66. doi: 10.1016/j.jallcom.2014.01.139

  26. [26]

      Yu H., Zhang S.Q., Zhao H.J., Will G., Liu P.R.. An efficient and low-cost TiO₂ compact layer for performance improvement of dye-sensitized solar cells[J]. Electrochim.Acta, 2009, 54:  1319-1324. doi: 10.1016/j.electacta.2008.09.025

  27. [27]

      Karthick S.N., Hemalatha K.V., Raj C.J., Subramania A., Kim H.J.. Preparation of TiO₂ paste using poly(vinylpyrrolidone)for dye sensitized solar cells[J]. Thin Solid Films, 2012, 520:  7018-7021. doi: 10.1016/j.tsf.2012.07.050

  28. [28]

      Yuan S.J., Li Y.G., Zhang Q.H., Wang H.Z.. Anatase TiO₂ sol as a low reactive precursor to form the photoanodes with compact films of dye-sensitized solar cells[J]. Electrochim.Acta, 2012, 79:  182-188. doi: 10.1016/j.electacta.2012.06.104

  29. [29]

      Fang X.L., Li M.Y., Guo K.M.. Improved properties of dye-sensitized solar cells by incorporation of graphene into the photoelectrodes[J]. Electrochim.Acta, 2012, 65:  174-178. doi: 10.1016/j.electacta.2012.01.038

  30. [30]

      Cameron P.J., Peter L.M.. How does back-reaction at the conducting glass substrate influence the dynamic photovoltage response of nanocrystalline dye-sensitized solar cells[J]. J.Phys.Chem.B, 2005, 109:  7392-7398. doi: 10.1021/jp0407270

  31. [31]

      Wang Q., Maser J.E., Grätzel M.. Electrochemical impedance spectroscopic analysis of dye-sensitized solar cells[J]. J.Phys.Chem.B, 2005, 109:  14945-14953. doi: 10.1021/jp052768h

• 

  Figure 1  XRD patterns of the TiO₂ powder with different molar ratios sintered at 500 ℃ for 3 h.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  SEM images (a) TiO₂-DEG (1:2) nanoparticles; (b) TiO₂ (DEG free) nanoparticles; (c) P25; (d) photoelectrode film based on TiO₂-DEG (1:2) nanoparticles; (e) photoelectrode film based on TiO₂ (DEG free) nanoparticles; (f) photoelectrode film based on P25.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  HRTEM images of TiO₂-DEG (1:2) nanoparticles (a) monodispersed TiO₂ particles (b) lattice fringes of anatase.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  UV–vis absorption spectra of the N719 dye desorbed from TiO₂ film electrodes made from TiO₂ synthesized by different additive amounts of DEG.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  J-V curves for DSSCs fabricated from different TiO₂ photoelectrodes.

  下载: 全尺寸图片 幻灯片
  

  Figure 6  Open-circuit voltage decay curves of the DSSCs fabricated by different TiO₂ photoelectrodes.

  下载: 全尺寸图片 幻灯片
  

  Figure 7  Nyquist plots of DSSCs constructed using different TiO₂ photoelectrodes.

  下载: 全尺寸图片 幻灯片

  Table 1.  Specific surface area (S_BET), average pore size (D_a) and total pore volume (V_T) of the different TiO₂ determined from gas adsorption analysis.

  下载: 导出CSV

  Table 2.  Photovoltaic parameters with different TiO₂ photoelectrodes.

  下载: 导出CSV

  Table 3.  The values of resistances of R₁, R₂, R₃ of different DSSC samples.

  下载: 导出CSV
•