Dongliang Bai‡
a,
Dexu Zheng‡c,
Shaoan Yang‡b,
Fengyang Yu*b,
Xuejie Zhua,
Lei Pengc,
Likun Wangb,
Jishuang Liuc,
Dong Yang*b and
Shengzhong (Frank) Liu*ab
aKey Laboratory of Applied Surface and Colloid Chemistry, Ministry of Education, Shaanxi Key Laboratory for Advanced Energy Devices, Shaanxi Engineering Lab for Advanced Energy Technology, School of Materials Science and Engineering, Shaanxi Normal University, Xi'an, 710119, China
bDalian National Laboratory for Clean Energy, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian, 116023, China. E-mail: yufengyang@dicp.ac.cn; dongyang@dicp.ac.cn; szliu@dicp.ac.cn
cChina National Nuclear Power Co., Ltd., Beijing 100097, China
First published on 22nd September 2023
Defects formed by halide ion escape and wettability of the perovskite absorber are essential limiting factors in achieving high performance of perovskite solar cells (PSCs). Herein, a series of ionic organic modulators are designed to contain halide anions to prevent defect formation and improve the surface tension of the perovskite absorber. It was found that the surface modulator containing Br anions is the most effective one due to its capability in bonding with the undercoordinated Pb2+ ions to reduce charge recombination. Moreover, this surface modulator effectively creates a suitable energy level between the perovskite and hole transport layer to promote carrier transfer. In addition, the surface modulator forms a chemisorbed capping layer on the perovskite surface to improve its hydrophobicity. As a result, the efficiency of PSCs based on surface modulators containing Br anion enhances to 23.32% from 21.08% of the control device. The efficiency of unencapsulated PSCs with a surface modulator retains 75.42% of its initial value under about 35% humidity stored in the air for 28 days, while the control device only maintained 44.49% of its initial efficiency. The excellent stability originates from the hydrophobic perovskite surface after capping the surface modulator.
Defects are divided into shallow-level defects and deep-level defects. The charges captured by the shallow-level defects will return to the conduction band/valence band and then be collected by the electrode. Therefore, these shallow-level defects have little influence on carrier recombination. However, the charges trapped by the deep-level defects will inevitably suffer non-radiative recombination, and it has a great effect on charge transfer in PSCs.19,20 In general, deep-level defects are mainly located at grain boundaries and the interface with a high formation energy,21,22 and the density of deep defects at the perovskite surface is around two orders of magnitude higher than that of the bulk.23 The surface defects thus dominate the carrier recombination in PSCs. Therefore, effective surface modification to reduce the defects is essential for improving the performance of PSCs.24,25 To date, post-treatment by coating passivation agents, including Lewis acid-base, alkyl ammonium halide, and organic molecules on the perovskite surface are considered efficient methods to decrease the surface defects. Shi et al.26 added 6,6′-dithiodinicotinic acid and urea as the additives, and the PCE improved from 16.76 to 20.64%. Wang et al.27 introduced thiourea (Lewis acid-base) into the CH3NH3PbI3 precursor, the photoelectric performance of the perovskite layer improved, and optimized efficiency was up to 19.80%. Ye et al.28 compared the passivation capacity of three structurally similar alkyl ammonium salts and found that the water contact angle gradually increased when increasing the number of carbon atoms in the alkane. The PCE was enhanced to 21.22% with a great improvement in humidity stability. Hu et al.29 employed organic molecule 4-(aminomethyl) benzoic acid hydroiodide as an organic cation in the CH3NH3PbI3 precursor solution. The device showed good stability, and the PCE increased to 15.6%. https://pubs.acs.org/action/doSearch?field1=Contrib%26text1=Yevgeny++Rakita et al.30 used alcoholic solutions of alkyl ammonium or halides monovalent alkali metal to treat metal (Pb or Sn) films. In this way, halide perovskite films had been formed without the use of toxic Pb2+ solutions. Cheng et al.31 employed pyridine-terminated conjugated small organic molecules to link the NiOx and perovskite layers, and the efficiency improved from 12.53% to 17.00%.
In addition, the capping layer above the perovskite absorber effectively inhibits ion migration and diffusion at the surface and efficiently protects the perovskite degradation from the influence of water and oxygen in the atmosphere. Zhong et al.32 added a polymer mixture of polyvinylpyrrolidone (PVP) and polyethylene glycol into the PbI2 precursor solution, and there was an efficiency enhancement of 30% over the pristine one. Polymer mixture-modified PSC also represents better air stability. Sandoval-Torrientes et al.33 used a variety of chemically modified fullerenes to prepare perovskite/fullerene blends, and the PCE of ETL-free perovskite devices improved to 14.3%. Chen et al.34 prepare 2D PEA2PbI4 capping layers on top of a 3D perovskite film, and a high efficiency of 18.51% with drastically improved stability was obtained.
In this study, we developed an ionic organic modulator, 1-hexadecyl-3-methylimidazolium hexafluorophosphate (1-h-3-MIHFP), 1-hexadecyl-3-methylimidazolium chloride (1-h-3-MICl) and 1-hexadecyl-3-methylimidazolium bromide (1-h-3-MIBr) as the surface modulator layers and compared their effect on the performance of PSCs. Experimental results revealed that all these surface modulator layers effectively enhanced the carrier lifetime and reduced the surface defects of the perovskite absorber. The perovskite devices with 1-h-3-MIBr exhibited the best performance. The efficiency of PSCs based on 1-h-3-MIBr increased to 23.32% from 21.08% of the control device. Meanwhile, the long-term environmental stability of the device based on 1-h-3-MIBr is better than that of the control perovskite device. The efficiency of the unencapsulated perovskite device with 1-h-3-MIBr was maintained at 75.42% of its initial value under about 35% humidity stored in the air for 28 days, while the control device only retained 44.49% of its initial efficiency. The excellent stability contributed to the hydrophobic surface after modifying 1-h-3-MIBr, which effectively prohibited moisture permeation. Halogen atoms have a great influence on passivation and hydrophobicity ability. Therefore, it is meaningful to choose a passivant containing suitable halogen, which would provide an insightful strategy for high-efficiency and stable perovskite devices.
Fig. 1i–l shows the water contact angles of the perovskite film without and with surface modulator layers. The same concentration (2 mg mL−1) of the passivant was spun onto the perovskite film. The static water contact angles of control and optimized perovskite films were measured. The contact angle of the pristine perovskite film was 57.22°, and it dramatically increased to 81.55°, 83.67°, and 91.25° for the 1-h-3-MIHFP, 1-h-3-MICl, and 1-h-3-MIBr, respectively. The significantly increased contact angles are due to the hydrophobic organic moieties in surface modulators. The perovskite covered by 1-h-3-MIBr showed the largest water contact angle compared to other surface modulators, likely due to the stronger molecules symmetry caused by a larger ionic radius of Br. The large water contact angle would enhance the barrier for moisture permeation to improve the stability of perovskite.35
The perovskite film with 1-h-3-MIBr was selected as the representation, and positive secondary ion depth profiles were measured to investigate the distribution of the elements using a surface modulator. Fig. 2a and b show the time-of-flight secondary-ion mass spectrometry (TOF-SIMS) results of the perovskite without and with 1-h-3-MIBr deposited on FTO substrates. Fig. S1† shows the photos of the samples before and after sputter etching. Approximately, the positions of various ions can be detected by the depth profiles during the sputtering process. The traces for ions directly related to FA0.95MA0.05PbI3 (Pb+), FTO (Sn+), and 1-h-3-MIBr (Br+) when sputtering depth of the films. The trace of the Pb+ position appears in the plateau part and then decreases with increasing sputtering time. The Pb+ and Sn+ positions are simultaneously observed, which suggests the ion migration into FTO from perovskite. It cannot detect the Br+ position in pristine perovskite (Fig. 2a), and this position clearly identifies in perovskite with 1-h-3-MIBr (Fig. 2b). The signal of the Br+ position quickly disappears with increasing sputtering time, indicating that1-h-3-MIBr forms an ultrathin film on perovskite. Fig. 2c, d, and f, g show the two-dimensional and three-dimensional depth profiles of Br elements for perovskite without and with 1-h-3-MIBr, respectively. For the pristine sample, negligible Br+ was detected likely due to the cross-contamination during the fabrication/measurement process. For a sample of perovskite with 1 h-3MIBr, the Br elements are mostly focused on the perovskite surface. TOF-SIMS results revealed that the surface modulator forms a uniform thin film on the perovskite surface.
The optical and charge transfer of the perovskite covered by different surface modulator layers were measured. Fig. 3a shows the normalized absorption and photoluminescence (PL) spectra of perovskite without and with different surface modulator layers. It is clear that the absorb edges and PL peaks of the perovskite show a blue shift, which should be caused by the reduced defects by surface modulator layers,36 as discussed below. Fig. 3b shows the steady-state PL spectra of the perovskite with and without surface modulators. All surface modulators could improve the peak intensity, and the perovskite with 1-h-3-MIBr gives the strongest PL intensity, demonstrating the fewest charge recombination.
The time-resolved PL (TRPL) measurements were tested to investigate the carrier transfer dynamics. Fig. 3c shows the TRPL spectra of the perovskite without and with surface modulator layers. The carrier decay lifetime (τ) can be fitted by an exponential equation.37,38 As a result, the carrier decay lifetimes are 328.33 ns, 646.71 ns, 498.88 ns, and 1265.60 ns for pristine perovskite, perovskite with 1-h-3-MIHFP, 1-h-3-MICl, and 1-h-3-MIBr, respectively. The perovskite with 1-h-3-MIBr showed the longest carrier lifetime, indicating efficient charge carrier dissociation and suppressed non-radiative recombination.
The perovskite devices were fabricated to evaluate the influence of different surface modulators. The cross-sectional SEM image of the completed device is shown in Fig. S2.† FTO and Au have been used as the anode and cathode, respectively. SnO2 and 2,2′,7,7′-tetrakis(N,N-di-p-methoxyphenylamine)-9,9′-spirobifluorene (spiro-OMeTAD) were employed as the electron transport layer (ETL) and hole transport layer (HTL), respectively. FA0.95MA0.05PbI3 was adopted as an absorber, and different ionic organic modulators are developed as surface modulator layers. Fig. 4a shows the current density–voltage (J–V) curves of PSCs without and with various surface modulator layers, and the key photovoltaic parameters including short-circuit current density (Jsc), open-circuit voltage (Voc), fill factor (FF) and PCE are summarized in Table 1. The PCE of the best control perovskite device was 21.08% with Jsc of 25.67 mA cm−2, Voc of 1.08 V, and FF of 75.87%. When the perovskite is covered by surface modulators, both Voc and FF are enhanced. The perovskite device with 1-h-3-MIBr exhibits the largest Voc of 1.12 V and FF of 80.95%, yielding the highest efficiency of 23.32%. The high Voc and FF are attributed to the long carrier lifetime (Fig. 3b and c) and the decreased carrier recombination (as discussed below) when using 1-h-3-MIBr surface modulator. In addition, the performance of our device with passivant has been compared with data from previous reports (Table S1†), and it is clear that our device shows high efficiency. Fig. 4b shows the external quantum efficiency (EQE) measurements of the perovskite devices without and with 1-h-3-MIBr. The integrated photocurrents are 24.41 mA cm−2 and 24.42 mA cm−2, in agreement with J–V results.
Device | Voc (V) | Jsc (mA cm−2) | FF (%) | PCE (%) |
---|---|---|---|---|
Control | 1.08 | 25.68 | 75.87 | 21.08 |
1-h-3-MIHFP | 1.11 | 25.89 | 79.34 | 22.81 |
1-h-3-MICl | 1.09 | 25.79 | 79.95 | 22.50 |
1-h-3-MIBr | 1.12 | 25.84 | 80.95 | 23.32 |
In addition, 13 individual devices for each surface modulator layer have been fabricated and tested to accurately quantify the repeatability of the perovskite devices. The parameter distribution histograms are shown in Fig. S3,† and the statistics are listed in Tables S2–S5.† All the devices exhibit good repeatability with a small standard deviation, indicating that these ionic organic modulators are excellent surface modulators in the PSCs. The above results revealed that 1-h-3-MIBr is the most effective surface modulator, and the below discussions thereby focus on perovskite with 1-h-3-MIBr.
Environmental stability is a key characteristic of PSCs. Fig. 4c shows the long-term environmental stability of the perovskite devices without and with 1-h-3-MIBr. The PCE of the unencapsulated device with 1-h-3-MIBr maintains 75.42% of its initial efficiency under about 35% humidity stored in the air for 28 days. However, the efficiency of the control device without 1-h-3-MIBr only holds 44.49% of its initial value. The excellent environmental stability originates from the hydrophobic perovskite surface covered by 1-h-3-MIBr (Fig. 1l), which successfully blocks the moisture permeation into the device.
Fig. 5a shows the ultraviolet photoelectron spectrometer (UPS) results of the perovskite without and with 1-h-3-MIBr. The Fermi level (EF) is calculated by equation EF = Ecut − 21.22 eV, where Ecut is the cut-off binding energy, and 21.22 eV is the emission energy of the helium irradiation. The Ecut of pristine perovskite and perovskite with 1-h-3-MIBr are 16.76 eV and 17.01 eV, respectively. Thus, the EF of the pristine perovskite and perovskite with 1-h-3-MIBr are −4.46 eV and −4.21 eV, respectively. The Fermi edge (Eedge) of the pristine perovskite and perovskite with 1-h-3-MIBr are 1.62 eV and 1.44 eV (Fig. 5a), respectively. Therefore, the valence band (EVB) of the pristine perovskite and the perovskite with 1-h-3-MIBr are −6.08 eV and −5.65 eV according to the equation of EVB = EF − Eedge. The energy–level diagram is described in Fig. 5b. The EVB of perovskite based on 1-h-3-MIBr is closer to spiro-OMeTAD (HTL) than that of pristine perovskite, leading to larger Voc due to less energy loss during charge transfer.39
X-ray photoelectron spectra (XPS) were collected to identify the interaction between the perovskite and 1-h-3-MIBr. Fig. 5c shows the XPS results of Pb 4f for the pristine perovskite and perovskite with 1-h-3-MIBr. The binding energy scale for XPS was calibrated according to the C 1 s line at 284.8eV. There is a small binding energy at 141.0 eV, corresponding to the Pb 0 core level for the pristine perovskite. This binding energy disappears after covering 1-h-3-MIBr thin film. Meanwhile, the Pb 4f peaks of the perovskite with 1-h-3-MIBr shift to a lower binding energy compared to the pristine perovskite. These results indicate that the 1-h-3-MIBr reacts with uncoordinated Pb2+ ions, which would reduce the defects due to the decreased ion vacancy.
Fig. 5d shows the dark current–voltage (I–V) curves of the single carrier devices. The defect density (Nt) can be calculated by the equation of Nt = 2ε0εrVTFL/eL2.40 Where the ε0 is the vacuum permittivity, εr is the relative dielectric constant of perovskite, e is the element charge, and L is the thickness of the perovskite film. The VTFL is the trap-filled limit voltage, which can be obtained from the dark I–V curves. The VTFL values are 0.13 V and 0.18 V for the pristine perovskite and perovskite with 1-h-3-MIBr, respectively. Therefore, the defect densities are 3.62 × 1014 and 2.58 × 1014 cm3 for the control perovskite and perovskite with 1-h-3-MIBr, respectively. The defect density calculation results are in good agreement with the above analysis. The low defect density significantly decreases the charge recombination, resulting in high Voc and FF.
Footnotes |
† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3ra00809f |
‡ These three authors contributed equally to this study and share the first authorship. |
This journal is © The Royal Society of Chemistry 2023 |