Effective Improvement of the Photovoltaic Performance of Carbon-Based Perovskite Solar Cells by Additional Solvents
A solvent-assisted methodology has been developed to synthesize CH3NH3PbI3 perovskite absorber layers. It involved the use of a mixed solvent of CH3NH3I, PbI2, γ-butyrolactone, and dimethyl sulfoxide (DMSO) followed by the addition of chlorobenzene (CB). The method produced ultra-flat and dense perovskite capping layers atop mesoporous TiO2 films, enabling a remarkable improvement in the performance of free hole transport material (HTM) carbon electrode-based perovskite solar cells (PSCs). Toluene (TO) was also studied as an additional solvent for comparison. At the annealing temperature of 100 °C, the fabricated HTM-free PSCs based on drop-casting CB demonstrated power conversion efficiency (PCE) of 9.73 %, which is 36 and 71 % higher than those fabricated from the perovskite films using TO or without adding an extra solvent, respectively. The interaction between the PbI2–DMSO–CH3NH3I intermediate phase and the additional solvent was discussed. Furthermore, the influence of the annealing temperature on the absorber film formation, morphology, and crystalline structure was investigated and correlated with the photovoltaic performance. Highly efficient, simple, and stable HTM-free solar cells with a PCE of 11.44 % were prepared utilizing the optimum perovskite absorbers annealed at 120 °C.
KeywordsHalide perovskite Solar cell Spin-coating Carbon counter electrode Free hole transporting material
During the past 5 years, there has been a surging interest in the study of organic–inorganic hybrid perovskite compounds for applications in photovoltaic devices because of low cost, simple fabrication process, and high-efficiency solar power conversion [1, 2, 3, 4, 5, 6]. Typically, perovskite solar cells (PSCs) employed a mesoporous titania or alumina scaffold, a methylammonium (MA) lead iodide perovskite light absorber, an organic hole transport material (HTM), characteristically spiro-OMeTAD (2,2′,7,7′-tetrakis-(N,N-di-p-methoxyphenylamine)-9,9′-bifluorene), and an Au or Ag electrode. PSCs achieved a power conversion efficiency (PCE) of >10 % in 2012 [2, 3]. TiO2/spiro-OMeTAD is regarded as a successful couple owing to its good optical transparency and perfect band alignment with respect to CH3NH3PbI3. Later, the PCEs were improved to 15 % using a two-step sequential deposition technique, involving spin-coating of a PbI2 followed by exposure to a solution of CH3NH3I to form CH3NH3PbI3, or a dual-source vapor deposition technique to fabricate a planar heterojunction solar cell [4, 5]. Then, a planar structured PSC using a polyethyleneimine ethoxylated modified ITO electrode, yttrium-doped TiO2 layer, mixed halide perovskite CH3NH3PbI3−xClx absorber, and spiro-OMeTAD reached a PCE of 19.3 % . Very recently, triple Cs/MA/formamidinium cation PSCs reached a power output of 21.1 % .
Currently, the most research PSC devices employ gold as a back contact, in conjunction with organic hole conductors acting as electron-blocking or hole transport layers [4, 5, 6, 7]. These cells are expensive because of the high cost of pure hole transportation materials. Besides, hole conductors such as the widely used spiro-OMeTAD are not only expensive but can cause moisture-induced degradation in PSCs, especially with hygroscopic dopants such as lithium bis(trifluoromethanesulfonyl)imide . To overcome the lifetime-limiting problems with the organic hole transporters, HTM-free perovskite photovoltaics were proposed and reported by Etgar et al. . HTM-free PSCs achieved efficiencies of about 10 % with gold as a back electrode [10, 11, 12]. The precious Au electrode also requires a high-vacuum and high-cost evaporation technique, thereby limiting its future application. Nowadays, the efficiency of the best PSCs is competitive with current commercial technologies, and they are potentially much cheaper. However, commercial solar cells must last 20–30 years with minimal degradation. The biggest challenge facing PSCs is long-term stability in a wide range of environments .
Low-cost nano-carbon can be an ideal material to substitute Au as a back contact in PSCs because its function is similar to that of Au. In the past decade, carbon nanomaterials have been demonstrated to be excellent counter-electrode candidates for use in dye-sensitized solar cells (DSSCs) owing to their various fascinating properties, including high electrical conductivity, thermal stability, good optical transparency, unique nanostructure, excellent electrocatalytic activity, low cost, and abundance [13, 14, 15, 16]. Substantial gains have been made in the application of carbon nanomaterials in DSSCs; however, perovskite organic lead iodide is unstable at high temperatures or in some solvents . Thus, the direct preparation of a carbon layer faces some problems. HTM-free PSCs with carbon contacts were first fabricated by infiltrating CH3NH3PbI3 or a mixed-cation 5-ammoniumvaleric (5-AVA) and MA perovskite (5-AVA)x(MA)1−xPbI3 into a high-temperature prefabricated monolithic device, which consists of four layers including TiO2 dense, TiO2 mesoporous, ZrO2 mesoporous, and carbon layers . The (5-AVA)x(MA)1−xPbI3 perovskite device achieved a PCE of 12.8 %. The cells have shown promising stability under long-term light soaking and long-term heat exposure. But these devices employed complex structures and required processing temperatures of up to 400 °C to remove solvents and organic binders in the printed ZrO2 space and carbon black/graphite electrodes. The complicated fabrication and high-temperature processing increased material or energy consumption and limited their mass production and fabrication on a plastic substrate. Very recently, carbon counter electrodes were prepared using low-temperature (LT) processed (70–100 °C) carbon pastes and applied in HTM-free perovskite/TiO2 heterojunction solar cells to substitute noble metallic anodes. Under optimized conditions, a PCE in the range of 8.31–9.00 % has been demonstrated with these carbon counter electrodes [19, 20]. The HTM-free solar cell with the LT carbon counter electrode can have a much simpler structure, thereby reducing the cost and improving the overall stability of PSCs. However, the PSCs based on the LT carbon contact are poor in photovoltaic performance. Therefore, it is worth developing and improving the performance of HTM-free and carbon-based PSCs.
The perovskite layers with a well-defined grain structure, full surface coverage, and small surface roughness allow realization of an efficient solar cell [5, 21]. Therefore, various morphology control protocols including sequential deposition , thermal evaporation deposition , compositional engineering [22, 23], additive-assisted deposition [24, 25], solvent engineering [26, 27, 28], and intramolecular exchange processing  were investigated for high-quality perovskite absorbers. Jeon et al.  reported a solvent engineering method for highly uniform CH3NH3Pb(I1−xBrx)3 (x = 0.1–0.15) perovskite layers and high-efficiency organic HTM-based devices using a mixed solvent of γ-butyrolactone (GBL) and dimethyl sulfoxide (DMSO) followed by toluene (TO) drop-casting (DC). Although DMSO-based solvent engineering [26, 28] is a very potential experimental technique, the related processing involves the anfractuous coupling of fluid rheology, solvent evaporation, and molecular self-assembly, and the formation of high-quality perovskite film is the result of a complex dynamic process still under investigation. Furthermore, the morphology control methods have been mainly developed for organic HTM-based PSCs and rarely explored for HTM-free PSCs. However, the ideal morphology of HTM-free PSC also requires a uniform, highly crystalline, and high-coverage perovskite capping layer on the top of the mesoporous TiO2 .
In this work, we report a chlorobenzene (CB)-based solvent-assisted process toward synthesizing simple, stable, and efficient HTM-free PSCs with carbon counter electrodes. CH3NH3PbI3 absorber layers were synthesized on mesoporous TiO2 films via spin-coating the mixed solution of CH3NH3I, PbI2, GBL, and DMSO, followed by DC CB or TO while spinning. TO was studied as a DC solvent for comparison. The effects of the DC solvent and the annealing temperature on thin film morphology, crystal structure, and the solar cell performance were investigated. Through changing the DC solvent and optimizing the annealing temperature, extremely uniform and dense perovskite capping layers atop mesoporous TiO2 films were obtained, enabling the fabrication of remarkably improved HTM-free PSCs. The efficiency of simple structured HTM-free solar cells was high up to 11.42 %. The obtained carbon-based PSC devices have shown much more promising stability than the HTM devices.
2 Experimental Section
Unless specified, all materials were purchased from either Alfa Aesar or Sigma-Aldrich and used as received. Spiro-MeOTAD was purchased from Merck KGaA and Luminescence Technology Corp. MA iodide (MAI) CH3NH3I was synthesized according to a previous study .
2.2 Device Fabrication
Fluorine-doped tin oxide-coated glass (Pilkington TEC 15) was patterned by etching with Zn powder and 2 M HCl. The etched substrate was then cleaned with surfactant and rinsed with acetone, ethanol, and deionized water. A 50-nm-thick compact TiO2 (c-TiO2) thin layer was synthesized by a procedure reported in our previous work . The porous TiO2 (p-TiO2) layer was deposited by spin-coating at 5000 rpm for 30 s using a commercial TiO2 paste (Dyesol 18NR-T) diluted in ethanol (1:2.5 weight ratio) and consequently heated at 500 °C for 30 min. After cooling to room temperature, the as-prepared nanoporous TiO2 films were then dipped into a 40 mM TiCl4 aqueous solution for 30 min at 70 °C, dried at ambient atmosphere, and then sintered at 500 °C for 30 min.
CH3NH3PbI3 perovskite absorber layers were synthesized by modifying the solvent engineering method reported by Jeon et al. . This was done in a glove box maintaining 10 % RH level. The synthesized CH3NH3I (0.1975 g) powders and lead iodide PbI2 (0.5785 g) were stirred in a mixture of GBL (700 μL) and DMSO (300 μL) at 60 °C for 12 h. The formed precursor solution was deposited onto p-TiO2/c-TiO2/FTO substrate by a successive two-step spin-coating process at 2000 rpm for 50 s and at 3500 rpm for 50 s, respectively. During the second step, anhydrous CB or TO was dripped onto the center of the sample 30 s prior to the end of the program. The perovskite precursor-coated substrate was heated and dried on a hot plate at a temperature of 50–140 °C for 10 min.
The carbon electrodes were prepared by doctor blade coating an LT conductive carbon ink (Shanghai Materwin New Materials Co., Ltd.) on the grown perovskite absorber, followed by drying at 100 °C for 30 min. For comparison, spiro-OMeTAD was deposited onto the perovskite absorber, and the metal cathode around 100 nm was deposited on the spiro-OMeTAD HTM layer by thermal evaporation under the base pressure of 6 × 10−4 Pa [31, 32].
The morphologies of the perovskite absorbers using different post-heating temperatures and various DC solvents were characterized by field emission scanning electron microscope (FESEM, Hitachi S4800). The structures of the formed perovskite absorbers were identified by X-ray diffractometer (XRD, Bruker D8 Davinci instrument, Cu Kα: λ = 0.15406 nm). Photocurrent density–voltage (J–V) measurements were performed using an AM 1.5 solar simulator equipped with a 1000 W xenon lamp (Model No. 91192, Oriel, USA). The solar simulator was calibrated using a standard silicon cell (Newport, USA). The light intensity was 100 mW cm−2 on the surface of the test cell. J–V curves were measured using a computer-controlled digital source meter (Keithley 2440) in the reverse direction. During device photovoltaic performance characterization, a metal aperture mask with an opening of about 0.09 cm2 was used.
3 Results and Discussion
Photovoltaic performance of solar cells with CH3NH3PbI3 layers fabricated without assisted solvent or with chlorobenzene or toluene as an assisted solvent
JSC (mA cm−2)
RS (Ω cm2)
RSH (Ω cm2)
No drop-casting solvent
The morphology and the crystal structure evolution of the perovskite film with CB droplet treatment shown in Figs. 1 and 2 could be attributed to the formation of CH3NH3I–PbI2–DMSO intermediate phase during spin-coating from MAI, PbI2, and DMSO . This process can be regarded as the transformation of PbI2–DMSO–MAI into MAPbI3, similar to the case with a TO DC. Jeon et al.  deposited high-quality CH3NH3Pb(I1−xBrx)3 (x = 0.1–0.15) perovskite layers through the use of a combination of DMSO/GBL followed by a TO drip. They found that the spin-coated PbI2–DMSO–MAI intermediate phase possessed an extremely uniform and flat morphology, the intermediate phase was partly transformed into perovskite phases at 100 °C, and the complete transformation into CH3NH3Pb(I1−xBrx)3 took place at an annealing temperature of 130 °C. But, from Fig. 2, pure perovskite phases were obtained at 100 °C for treatment with both CB and TO in our synthesis. Our results are in accordance with those reported in . High-quality CH3NH3PbI3 films were obtained for a fast deposition–crystallization procedure at 100 °C with the assistance of CB in the latter.
Photovoltaic performance of solar cells with CH3NH3PbI3 layers fabricated via assistance of chlorobenzene at different annealing temperatures of 120–140 °C
Annealing T (°C)
JSC (mA cm−2)
RS (Ω cm2)
RSH (Ω cm2)
Moreover, the series resistance (RS) and the shunt resistance (RSH) were estimated from the J–V characteristics shown in Figs. 3 and 6. The RS and RSH data are summarized in Tables 1 and 2. For solar cells, maintaining the RS as low as possible is vitally important because large RS will decrease JSC, VOC, and FF and consequently PCEs [31, 39, 40]. For the devices with 100 °C, the RS from the PSC with the addition of CB is 73 Ω cm2, which is smaller than 121 or 129 Ω cm2 from the device with the introduction of TO or without DC solvent, respectively. In the case of the CB treatment, when the temperature was increased from 100 to 130 °C, the RS remained at the low values of about 70 Ω cm2. But, when the temperature was increased up to 140 °C, the RS of the device increased to 293 Ω cm2. The low RS from the device with the addition of CB and the annealing temperatures of 100–130 °C is due to small contact resistance and low bulk resistance of the high-quality perovskite layer, indicating that high currents can flow through the cell at low applied voltages . Typically, the shunt resistance (RSH) is due to p–n junction non-idealities and impurities near the junction, which bring about partial shorting of the junction, especially near cell edges. The RSH must be higher to avoid current loss at the junction , dwindling the photocurrent and consequently the solar cell performance. It has been diffusely reported that the pinholes formed in the solution-processed CH3NH3PbI3−xClx absorbers can cause direct contact between the p-type spiro-OMeTAD and the TiO2 compact layer, leading to a shunting path that is probably partially responsible for the low FF and open circuit voltage in devices [5, 17]. In the case of our HTM-free PSCs with 100 °C shown in Table 1, the pinhole or voids in the capping layer on the p-TiO2 layer caused the least RSH, resulting in the lowest FF and VOC in the device with pristine absorber, and induced the second least RSH, leading to the second lowest FF and VOC in the device via dropwise TO application. The smoother perovskite layer due to introducing TO or CB into the spin-coated perovskite precursor layer from the mixed solution of CH3NH3I, PbI2, GBL, and DMSO decreased the series resistance and increased the shunt resistance of cell, resulting in the higher photovoltaic effects (Table 1). In the case of different annealing temperatures with the CB treatment shown in Table 2, as the temperature was increased from 100 to 130 °C, the shunt RSH had the higher and higher resistive value, but, when the temperature was increased up to 140 °C, the pinhole or voids in the capping layer on the p-TiO2 layer caused the least RSH, resulting in the lowest FF and VOC in the device. Therefore, the improved device performance can be associated with the evolved and promoted morphological and crystalline properties of CH3NH3PbI3 film by use of CB or TO droplets and optimized annealing temperature, as exhibited by optical, SEM, and XRD measurements in Figs. 1, 2, 4, and 5. The improved device performance unambiguously verified the significance of CB droplets and annealing temperature on PSC optimization. The possible mechanisms for improving the quality of the perovskite absorber layer with DC solvent and enhancing the resulted PSC performance are further explored below.
The boiling points of the traditional solvents used for growing perovskite absorbers are relatively high, having the values of 153 and 204–205 °C for N,N-dimethylmethanamide (DMF) and GBL, respectively. Such relatively high boiling points can result in the formation of large-area thickness variations or shrinkage/dewetting of the casting precursor solution due to the relatively prolonged drying times of solution-coated films. In our work, the proposed solvent- assisted technology involved spin-coating the mixed solution of CH3NH3I, PbI2, and GBL/DMSO, followed by DC CB or TO while spinning. TO (boiling point 111 °C, vapor pressure 22 mmHg) and CB (boiling point 132 °C, vapor pressure 11.8 mmHg) have much lower boiling point and tremendously higher vapor pressure than DMSO (boiling point 189 °C, vapor pressure 0.6 mmHg) and GBL (boiling point 204–205 °C, vapor pressure 1.5 mmHg) at room temperature. Considering the physical properties of the solvents, at the beginning stage of the spin-coating process, the film was composed of MAI and PbI2 dissolved in the DMSO/GBL solvent mixture, and the precursor film continued to thin because liquid flowed radially owing to the action of centrifugal force, and solvent evaporation was neglected, while in the intermediate stage, the composition of the film was concentrated by the evaporation of GBL due to its higher vapor pressure than DMSO. Then, the introduction of TO or CB droplets with very higher vapor pressure during the second spinning stage induced the immediate freezing of the constituents on spinning and the rapid formation of the MAI–PbI2–DMSO phase, producing a full and even precursor layer. The use of DMSO helped retard the rapid reaction between the inorganic PbI2 and CH3NH3I components of perovskite during the evaporation of solvents in the spin-coating process because DMSO has a stronger coordination ability with PbI2 than that of the usually used DMF . At last, the greatly homogeneous and flat precursor film was converted into a pure crystalline CH3NH3PbI3 perovskite layer after annealing at 100 °C. As displayed in Figs. 1, 2, and 3 and Table 1, CH3NH3PbI3 films via the dropwise CB application showed better morphological, crystalline, and photovoltaic properties than the perovskite layers using TO droplet treatment. For a highly volatile solvent such as TO, one can expect to have very significant evaporation during the spin-off stage. The strong evaporation of the casting TO solvent possibly affects fluid rheology and vice versa [42, 43], tending to give relatively rough precursor films. In contrast, CB has a lower evaporation rate due to its relatively high boiling point and lower vapor pressure compared to those of TO, which should be a benefit to form a flatter, smoother, and more uniform MAI–PbI2–DMSO intermediate phase layer. The higher quality intermediate phase layer resulted in more highly crystalline and more homogeneous CH3NH3PbI3 perovskite capping layer atop the mp-TiO2 after the heat treatment at 100 °C, which caused the lower series resistance, higher shunt resistance, and better photovoltaic performance in the PSC (Fig. 3; Table 1). The fabricated HTM-free PSCs based on CB treatment demonstrated a PCE of 9.73 %, which is 71 and 36 % higher than those of the control device fabricated from the pristine perovskite film or via TO droplet treatment, respectively. The improved device performance unambiguously verified the significance of CB droplets on HTM-free PSC optimization. After further optimizing the annealing temperature, the HTM-free PSCs based on the incorporation of CB achieved a PCE of 11.44 %.
The type of the DC solvent and annealing temperature employed during the preparation of CH3NH3PbI3 films via a solvent-assisted process have a considerable impact on the resulting absorber morphologies, crystalline structures, and device photovoltaic performance. The advantages using CB as an assisted solvent become apparent upon comparing the J–V curves and PCEs of the carbon-based HTM-free CH3NH3PbI3 devices prepared using CB or TO as a DC solvent or without DC solvent. The CH3NH3PbI3 films grown using CB show better morphological, surface coverage, and crystalline properties than the perovskite layers formed by the addition of TO during processing under the same conditions. The pristine perovskite layers without using DC solvent exhibit the even poorer morphological and crystalline properties. The HTM-free devices based on CH3NH3PbI3 fabricated by incorporation of CB show superior PCEs. The effects of the annealing temperature on CH3NH3PbI3 film morphology, crystal structure, and the solar cell performance were also investigated for the CB-assisted process. High-efficiency carbon-based HTM-free solar cells with a PCE of 11.44 % were produced using an optimized annealing temperature of 120 °C. This work provides an effective protocol for fabricating efficient, simple, stable, and low-cost inorganic–organic hybrid heterojunction solar cells.
This work was supported by the National Natural Science Foundation of China (Nos. 11274119, 61275038).
- 27.M. Xiao, F. Huang, W. Huang, Y. Dkhissi, Y. Zhu, J. Etheridge, A.G. Weale, U. Bach, Y.B. Cheng, L. Spiccia, A fast deposition–crystallization procedure for highly efficient lead iodide perovskite thin-film solar cells. Angew. Chem. 126(37), 10056–10061 (2014). doi:10.1002/ange.201405334 CrossRefGoogle Scholar
- 36.T. Baikie, Y. Fang, J.M. Kadro, M. Schreyer, F. Wei, S.G. Mhaisalkar, M. Grätzel, T.J. White, Synthesis and crystal chemistry of the hybrid perovskite (CH3NH3)PbI3 for solid-state sensitized solar cell applications. J. Mater. Chem. A 1(18), 5628–5641 (2013). doi:10.1039/C3TA10518K CrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.