Improving the performance of solid-state quantum dot-sensitized solar cells based on TiO₂/CuInS₂ photoelectrodes with annealing treatment

Abstract

CuInS₂ quantum dot (CIS QD)-sensitized solar cells (QDSSCs) with spiro-OMeTAD as the solid-state hole transport material were fabricated by using a successive ionic layer adsorption and reaction (SILAR) process. The structural, morphological, optical and photovoltaic characterizations of the composite films indicate the importance of thermal treatment in enhancing the performance of the solar cells. The results reveal that chalcopyrite CIS QDs of around 8 nm in size are distributed homogeneously over the surface of TiO₂ particles and are well separated from each other under the proper annealing conditions. With increasing the temperature, the effect of annealing is to shift the absorption onset to longer wavelengths, thus improving the photocurrent substantially. It is also noteworthy that the annealing is beneficial for the efficient charge transport and the decreased charge recombination. Under simulated illumination (AM 1.5, 100 mW cm⁻²), the solid-state QDSSCs with distinct architectures deliver a maximum efficiency of 1.41% for the solar cell fabricated with a pristine CIS QD-sensitized TiO₂ photoelectrode annealed up to 450 °C.

---

Introduction

Dye-sensitized solar cells (DSSC) pioneered by Grätzel and O'Regan¹ have served as a template for a design of solar cell architecture, giving the ease of modifying critical cell components like the photoanode, the absorbing material, and the hole transport material (HTM). QD-based cell devices have been proposed as a way to realize third-generation solar cells, because QDs possess the possibility of generating multiple exciton generation (MEG) with a single photon via the impact ionization effect.² Moreover, their unique quantum confinement effect and high extinction coefficients suggest that QDs could be considered to be a perspective light absorber for quantum dot-sensitized solar cells (QDSSCs).³ In both DSSCs and QDSSCs, it has been realized that the liquid HTM^4,5 in traditional cell devices should be replaced with solid-state HTM,^6,7 which are desirable to avoid sealing and ameliorate the long-term stability problems associated with liquids. A solid-state device architecture refers to the replacement of traditional liquid electrolyte with solid-state HTMs, such as organic semiconductors spiro-OMeTAD (2,2′,7,7′-tetrakis(N,N-di-p-methoxyphenylamine)-9,9′-spirobifluorene).^8,9

In this paper, we specifically used CuInS₂ (CIS) QD sensitizer in solid-state QDSSC with spiro-OMeTAD as the solid-state hole transport material. Commonly, quantum dot sensitizers for QDSSCs have been extensively focused on CdSe,^10,11 CdS,¹² CdTe,¹³ PbSe,¹⁴ and PbS^15,16 during the past several years. However most of these semiconductor nanocrystals are toxic or air-sensitive. Free of these weaknesses, CuInS₂ is a proper light adsorption material for QDSSCs because of its band gap (E_g ≈ 1.5 eV) and high extinction coefficient (∼10⁵ cm⁻¹) at 500 nm.¹⁷ However, up to now, reports on successful CuInS₂ based QDSSCs are quite rare. Here, CIS QDs were prepared on the surface of TiO₂ mesoporous films as a promising sensitizer by employing a series of successive ionic layer adsorption and reaction (SILAR) processes, which allow each layer to successively grow in a reproducible and controllable manner. We report the effect of annealing on the structural, morphological, optical and photovoltaic properties of solar cells. These TiO₂/CIS films are further demonstrated to be the promising photoelectrodes for solid-state QDSSCs, giving a maximum power conversion efficiency of 1.41%.

Experimental section

Materials

Copper(II) nitrate hemi(pentahydrate) (Cu(NO₃)₂·2.5H₂O, 99.99%), indium(III) nitrate hydrate (In(NO₃)₃·H₂O, 99.99%), sodium sulfide (Na₂S·9H₂O, 99.99%) were purchased from Sigma-Aldrich (Shanghai) Trading Co., Ltd. All the reagents used were of analytical grate and were directly used without further purification. Ultrapure deionized water was used for the preparation of aqueous solutions.

Preparation

To prepare the nanostructured TiO₂ substrates, a compact blocking underlayer of titanium dioxide (ca. 80 nm thick) was first spin coated onto a piece of ultrasonically cleaned F-doped SnO₂ glass. Next, TiO₂ doctor-blading (using Degussa P25 paste) was employed, which was followed by a heat-treatment to achieve a ∼9 μm nanostructured TiO₂ active layer.

CuInS₂ QDs were deposited onto TiO₂ film by the SILAR process, which was similar to the procedure described by Chang et al.¹⁸ Briefly, TiO₂ substrate was dipped alternately in aqueous solutions of 0.1 M In(NO₃)₃ for 2 min, and 0.1 M Cu(NO₃)₂ for 30 s, following in 0.2 M Na₂S aqueous solutions for 2 min. Between each dip, the films were rinsed with deionized water to remove excess precursors. Such the three-step procedure is termed as one SILAR cycle for incorporating CuInS₂ QDs, which was repeated ten times. To improve the crystallinity of the SILAR-deposited CuInS₂ QDs, the samples were annealed in furnace under nitrogen atmosphere at 250–450 °C for 30 min after SILAR deposition. Afterward, the organic hole conductor spiro-OMeTAD [(2,2′,7,7′-tetrakis-(N,N-di-p-methoxyphenylamine)-9,9′-spirobifluorene)] was introduced into the mesopores of TiO₂/CIS film by spin coating a 96 mg mL⁻¹ M⁻¹ chlorobenzene solution of spiro-OMeTAD with two additives: tert-butylpyridine (tBP) at a ratio of 1 : 9.6 μL mg⁻¹ of spiro-OMeTAD, Li salt (Li[CF₃SO₂]₂N, 170 mg mL⁻¹ in acetonitrile) at a ratio of 1 : 3.2 μL mg⁻¹ of spiro-OMeTAD. Finally, silver was deposited by thermal evaporation as a counter electrode and to define the active area of 0.09 cm².

Characterization

The as-prepared samples were characterized by X-ray diffraction (XRD) using a Bruker F8 diffractometer with Cu Kα radiation. High-resolution transmission electron microscopy (TEM) images and scanning electron microscopy (SEM) and were carried out with a Tecnai G2 F20S-TWIN TEM and a JSM-7500 SEM, respectively. Energy dispersion X-ray (EDX) spectroscopy was obtained by EDX (Oxford 80T) integrated in TEM. Absorption spectra were recorded with a UV-vis spectrophotometer (PerkinElmer, Lambda 950). Photovoltaic performances (J–V curves) of cell devices were recorded on a Keithley 2400 source meter under illumination by an AM 1.5 G solar simulator (SAN-EI Electric, XES-200S1, equipped with a 300 W xenon lamp).

Results and discussion

The structural phases of TiO₂/CIS photoelectrodes were determined through X-ray diffraction (XRD) characterizations, as shown in Fig. 1a. The bottom line of Fig. 1a is the XRD pattern of plain TiO₂ film without incorporating CIS QDs. The peak observed at 2θ = 25.4°, 37.8°, 48.2°, 53.5°, and 55.0° correspond to the (101), (004), (200), (105), and (211) planes of anatase TiO₂ phase, and the peaks at 26.6°, 33.8°, and 51.6° are identified as the characteristic diffraction peaks of FTO. As a comparison, the XRD patterns for the samples annealed at 250 and 300 °C show no discernible peaks corresponding to crystalline CIS phase except for TiO₂ and FTO phase, indicating that the QDs is essentially amorphous. At 350 °C, only some weak characteristic peaks of CIS chalcopyrite phase appear. The broad CIS (112) peak with a low intensity indicates the small nanocrystal size due to the lack of thermodynamic driving force at a relatively low temperature (≤400 °C).¹⁹ Increasing the temperature up to 450 °C, it is found that the characteristic peaks of CIS become more stronger, indicating that thermal treatment improve the crystallinity of CIS QDs up to 450 °C. XRD patterns are tentatively identified to report the chalcopyrite phase. Fig. 1b additionally confirms the well-matched XRD pattern with reference CIS phase (JCPDS-270159) for the sample deposited on a glass substrate with annealing at 450 °C. The first three diffraction peaks (around 28°, 46° and 55°) can be indexed to the (112), (204)/(220) and (116)/(312) reflections of the tetragonal crystal structure, respectively.

Fig. 1 (a) XRD patterns of plain TiO₂ film and TiO₂/CIS photoelectrodes annealed at different temperatures (250–450 °C); (b) the formation of CIS chalcopyrite phase on a glass substrate at 450 °C.

Fig. 2a presents an overview TEM image of pure TiO₂ nanoparticles. It can be seen that TiO₂ shows a better dispersity with the average particle size of ∼20 nm, while the distinctive lattice planes could be observed. To provide the convincing proof of the formation of CIS chalcopyrite phase by current SILAR method, subsequent TEM image of TiO₂/CuInS₂ annealed at 450 °C is shown in Fig. 2b, displaying that the surface of bare TiO₂ nanoparticles appears to be covered by a large amount of smaller QDs. It demonstrates that the as-prepared CIS QDs exhibit a relatively uniform spherical shape, and a narrow particle size distribution with the average size of 8 ± 0.07 nm (seeing Fig. S1 and Table S1 in the ESI†). HRTEM observations were also employed to provide the detailed microscopic information. Fig. 2c shows the HRTEM image at the edge side of TiO₂, indicating the high crystallinity of TiO₂ and CuInS₂. The larger crystallite appearing in the left region of the image is identified to be TiO₂, and the observed lattice spacing of 0.352 nm corresponds to the (101) plane of anatase TiO₂. The observed lattice fringes with a distance of 0.195 and 0.320 nm on the edge of TiO₂ can be identified to the (220) and (112) planes of the tetragonal CIS QDs, respectively. In addition, EDX analysis shows that the ratio of Cu/In/S is 0.92 : 1.00 : 1.97, corresponding to a slightly Cu-deficient CuInS₂.

Fig. 2 (a) TEM image of pure TiO₂ nanoparticles. The inset is the corresponding lattice plane; (b) TEM and (c) HRTEM images of TiO₂/CIS QDs annealed at 450 °C. The inset in (c) shows the EDX spectrum of TiO₂/CIS film.

SEM images in Fig. 3 show the morphologies of a bare TiO₂ film, TiO₂/CIS photoelectrodes without and with annealing, respectively. The bare TiO₂ film is a uniform network of TiO₂ nanoparticles with an average size of 20 nm. As shown in Fig. 3b, TiO₂ nanoparticles after CIS adsorption show slightly enlarged size (∼25 nm) with near round shape and clear boundary. After the annealing at 450 °C (Fig. 3c), the size of the constituting nanoparticles further increases to ∼32 nm, and the interspaces between TiO₂ and CuInS₂ become shrink after annealing. Thus, there is a formation of an enforced interfaces connection between CuInS₂ and TiO₂, as well as a relatively uniform and dense CIS layer, which is potentially more favorable for transport of photogenerated electrons in QDSSCs,^20,21 and is advantageous to reduce direct contact areas between bare TiO₂ surface and electrolyte, consequently decreasing the probability of charge recombination at the TiO₂/HTM interface.

Fig. 3 SEM images of (a) bare TiO₂ film, CuInS₂ absorbed TiO₂ film (b) without and (c) with 450 °C annealing.

Fig. 4 tracks the progression in UV-vis spectra with increasing the temperature of thermal treatments. It is noteworthy that the increase of temperature leads to the absorption onsets at longer wavelengths. This demonstrates the quantum effect that increasing the QD size decreases the band gap, until the bulk conditions are approached. Subsequently, photovoltaic devices with TiO₂/CIS photoelectrodes were fabricated and characterized. Fig. 5 shows the current density–voltage (J–V) characteristics of the solar cells, and the key performance parameters of various devices are summarized in Table 1. The device with annealing at 250 °C generates an open circuit voltage (V_OC) of 0.29 V. The V_OC are increased to 0.39, 0.47, 0.58 and 0.64 V for photoelectrodes annealed at 300, 350, 400 and 450 °C, respectively, which provides a clear evidence to support that the annealing treatment can suppress the charge recombination. The second photovoltaic parameter, short circuit current density (J_SC), initially is 4.39 mA cm⁻², while that with annealing at 450 °C induces 5.34 mA cm⁻². The enhanced J_SC results from the beneficial light-harvesting enhancement over the UV-vis spectrum from photoelectrodes with thermal treatment at a higher temperature. Consequently, it is found that with increasing the annealing temperature, the power conversion efficiency (η) for solar cell based on TiO₂/CIS photoelectrode drastically increases from 0.52 to 1.41%. Therefore, it implies that the annealing process has a significant contribution to the performance of solar cell, and it could be widely used to the production of highly efficient QDSSCs. These results demonstrate a robust method of improving QD property, identify the specific mechanisms by which thermal treatment impacts device performance, and provide a framework for our future efforts optimizing the device architecture of solid-state QDSSCs.

Fig. 4 UV-vis absorption spectra of CIS QD-sensitized TiO₂ films annealed at different temperatures.

Fig. 5 Current density–voltage (J–V) characteristics of the solar cells using CIS QD-sensitized TiO₂ photoanodes with annealing at different temperatures.

Table 1 Photovoltaic parameters obtained from the J–V curves using TiO₂/CIS films annealed at different temperatures as photoelectrodes

Temperature of annealing (°C)	VOC (V)	JSC (mA cm^−2)	FF	Efficiency (%)
250	0.29	4.39	0.41	0.52
300	0.39	4.92	0.41	0.79
350	0.47	4.86	0.37	0.85
400	0.58	4.79	0.47	1.31
450	0.64	5.34	0.41	1.41

---

Conclusions

In this work, we developed a solid-state QDSSCs with using CIS QD-sensitized TiO₂ photoanodes via SILAR process. The optimized thermal treatment for solar cells would enhance the contact between QD sensitizer and TiO₂ photoanode, thus improving the photovoltaic performance, especially V_OC value of the devices. Meanwhile, the TiO₂/CIS photoelectrode annealed at a relatively higher temperature presents a broad-range absorption due to the lager QD size, rendering the enhancement of J_SC. The device using TiO₂/CIS film annealed at 450 °C as the photoelectrode generates a V_OC of 0.64 V, a J_SC of 5.34 mA cm⁻², and a FF of 0.41, yielding a η of 1.41%. The power conversion efficiency of QDSSCs might be further improved with the use of CIS QD sensitizer with the higher loading amount onto TiO₂ film and the longer absorption onsets by the modification of the synthetic method. It would be expected to enhance the photovoltaic performance of QDSSCs significantly.

Acknowledgements

This research was supported by Scientific and Technological Research Project of Hebei Province (No. QN2015008), the Preferentially Supported Project for People Studying Abroad of Hebei Province (No. C2015005010), National Natural Science Foundation of China (No. 11604072), and the Midwest Universities Comprehensive Strength Promotion Project.

Notes and references

1. B. O'Regan and M. Grätzel, Nature, 1991, 353, 737–740.
2. V. González-Pedro, X. Xu, I. Mora-Seró and J. Bisquert, ACS Nano, 2010, 4, 5783–5790.
3. Z. Pan, K. Zhao, J. Wang, H. Zhang, Y. Feng and X. Zhong, ACS Nano, 2013, 7, 5215–5222.
4. F. Ren, S. Li and C. He, Sci. China Mater., 2015, 58, 490–495.
5. L. Li, X. Yang, J. Gao, H. Tian, J. Zhao, A. Hagfeldt and L. Sun, J. Am. Chem. Soc., 2011, 133, 8458–8460.
6. Z. Huo, L. Tao, S. Dai, J. Zhu, C. Zhang, S. Chen and B. Zhang, Sci. China Mater., 2015, 58, 447–454.
7. H. Lee, H. C. Leventis, S.-J. Moon, P. Chen, S. Ito, S. A. Haque, T. Torres, F. Nüesch, T. Geiger, S. M. Zakeeruddin, M. Grätzel and M. K. Nazeeruddin, Adv. Funct. Mater., 2009, 19, 2735–2742.
8. T. P. Brennan, O. Trejo, K. E. Roelofs, J. Xu, F. B. Prinz and S. F. Bent, J. Mater. Chem. A, 2013, 1, 7566–7571.
9. E. L. Unger, A. Morandeira, M. Persson, B. Zietz, E. Ripaud, P. Leriche, J. Roncali, A. Hagfeldt and G. Boschloo, Phys. Chem. Chem. Phys., 2011, 13, 20172–20177.
10. K. N. Hui, K. S. Hui, X. L. Zhang, R. S. Mane and M. Naushad, Sol. Energy, 2016, 125, 125–134.
11. R. Ahmed, L. Zhao, A. J. Mozer, G. Will, J. Bell and H. Wang, J. Phys. Chem. C, 2015, 119, 2297–2307.
12. P. K. Santra and P. V. Kamat, J. Am. Chem. Soc., 2012, 134, 2508–2511.
13. M. G. Panthani, J. M. Kurley, R. W. Crisp, T. C. Dietz, T. Ezzyat, J. M. Luther and D. V. Talapin, Nano Lett., 2014, 14, 670–675.
14. W. K. Bae, J. Joo, L. A. Padilha, J. Won, D. C. Lee, Q. Lin, W.-k. Koh, H. Luo, V. I. Klimov and J. M. Pietryga, J. Am. Chem. Soc., 2012, 134, 20160–20168.
15. D. Vankhade, A. Kothari and T. K. Chaudhuri, J. Electron. Mater., 2016, 45, 2789–2795.
16. H. Wang, T. Kubo, J. Nakazaki, T. Kinoshita and H. Segawa, J. Phys. Chem. Lett., 2013, 4, 2455–2460.
17. M. Nanu, J. Schoonman and A. Goossens, Adv. Funct. Mater., 2005, 15, 95–100.
18. J.-Y. Chang, J.-M. Lin, L.-F. Su and C.-F. Chang, ACS Appl. Mater. Interfaces, 2013, 5, 8740–8752.
19. D. Lee and K. Yong, ACS Appl. Mater. Interfaces, 2012, 4, 6758–6765.
20. S.-Q. Fan, D. Kim, J.-J. Kim, D. W. Jung, S. O. Kang and J. Ko, Electrochem. Commun., 2009, 11, 1337–1339.
21. Z. Pan, H. Zhang, K. Cheng, Y. Hou, J. Hua and X. Zhong, ACS Nano, 2012, 6, 3982–3991.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra18528b

---