Chlorofullerene C60Cl6 Enables Efficient and Stable Tin‐Based Perovskite Solar Cells

Tin‐based perovskite solar cells (TPSCs) have received great attention due to their eco‐friendly properties and high theoretical efficiencies. However, the fast crystallization feature of tin‐based perovskites leads to poor film quality and limits the corresponding device performance. Herein, a chlorofullerene, C60Cl6, with six chlorine attached to the C60 cage, is applied to modulate the crystallization process and passivate grain boundary defects of the perovskite film. The chemical interactions between C60Cl6 and perovskite components retard the transforming process of precursors to perovskite crystals and obtain a high‐quality tin‐based perovskite film. It is also revealed that the C60Cl6 located at the surfaces and grain boundaries can not only passivate the defects but also offer a role in suturing grain boundaries to suppress the detrimental effects of water and oxygen on perovskite films, especially the oxidation of Sn2+ to Sn4+. As a result, the C60Cl6‐based device yields a remarkably improved device efficiency from 10.03% to 13.30% with enhanced stability. This work provides a new strategy to regulate the film quality and stability of TPSCs using functional fullerene materials.


Introduction
[3][4] Nowadays, the highest certified efficiency of lead-based perovskite solar cells has reached 25.7%, [5] comparable to monocrystalline silicon solar cells.However, the toxicity of watersoluble lead ions is one of the important factors hindering its commercialization. [6,7]Therefore, numerous works have been devoted to finding suitable metal ions, such as tin ions (Sn 2+ ), [8,9] germanium ions (Ge 2+ ), [10] bismuth ions (Bi 3+ ), [11] and copper ions (Cu 2+ ), [12] to replace lead ions in perovskites to fabricate lead-free PSCs in recent years.[15] In addition, the tin-based perovskite with an optical bandgap around 1.3 eV is the ideal candidate for high-performance solar cells based on the Shockley-Queisser equation. [15]owever, excessive bulk defects resulting from the fast crystallization process and inherent oxidation of Sn 2+ to Sn 4+ have plagued the development of tin-based perovskite solar cells (TPSCs). [16][23][24][25][26][27] For example, Li et al. [28] reported an enhanced chemical interaction between tribenzylphosphine oxide and perovskite components, which improved the perovskite film quality, passivated the perovskite defects, and raised the device efficiency.Chen et al. [29] demonstrated that the N-(3aminopropyl)-2-pyrrolidinone molecule could coordinate with Sn 2+ via carbonyl group C=O, which effectively changed crystal orientation and improved the stability of the device.Kayesh et al. [30] reported that the carboxyl groups (-COOH) of 5-ammonium valeric acid could form the hydrogen bond with the iodide anions of [SnI 6 ] 4+ , which benefit to enable the compact and homogeneous tin perovskite film.However, most of the commonly used organic molecular additives are electrical insulators, which could hinder charge transport and limit further improvements in device efficiency.Therefore, it is crucial to explore an alternative material that can take into account the perovskite crystallization regulation, defect passivation, and carrier transport.
Herein, a fullerene derivative, C 60 Cl 6 , is employed as an additive to regulate the perovskite film quality and improve the device's performance.First, C 60 cage and chloride can interact with the perovskite components of I − and Sn 2+ , [31,32] respectively, which could tune the perovskite crystallization process and reduce pinholes in the perovskite film.Second, C 60 and its derivatives with semiconducting properties have been proven to be excellent perovskite defect passivation materials, [33][34][35][36] which could effectively decrease the non-radiative recombination within the corresponding device.Lastly, the C 60 Cl 6 existing at the grain boundaries can play the role of stitching the grain boundaries, increasing the resistance of the perovskite film to water and oxygen erosion.As verified by our experiments, the C 60 Cl 6 can not only interact with perovskite components but also significantly improve the quality of perovskite films.As a result, the efficiency of the TPSCs with PEA 0.15- FA 0.85 SnI 3 as an active layer is obviously improved from 10.03% to 13.30%.More importantly, the C 60 Cl 6 -based device shows obviously enhanced stability in the air due to the synergy of both the improved perovskite film quality and the grain boundary suture protection.

Results and Discussion
The molecular structure and electrostatic potential (ESP) of C 60 Cl 6 are shown in Figure 1a, and the electricity enrichment of chlorine may contribute to the chemical interaction with Sn 2+ .And the synthetic route of C 60 Cl 6 is displayed in Figure S1, Supporting Information.In addition, C 60 Cl 6 presents enhanced solubility in the mixture solvent (DMF:DMSO = 4:1) than that of pristine C 60 (Figure S2, Supporting Information), which enables additive engineering to be implemented.Interestingly, the prepared perovskite precursor solutions with and without C 60 Cl 6 exhibit obvious different color evolutions after being stored in the air.As shown in Figure 1b, the control sample presents a significant color change from yellow to brown.The apparent change in color of the control solution can be attributed to the oxidation of Sn 2+ to Sn 4+ . [37]While the aged perovskite precursor solution with C 60 Cl 6 only shows a slightly different color compared to the fresh one (Figure 1c), indicating that the C 60 Cl 6 has a chemical interaction with Sn 2+ and inhibits the oxidation process of Sn 2+ to Sn 4+ .The suppressed oxidation process may be closely related to the high conduction band position (−3.74 eV) of C 60 Cl 6 (Figure S3, Supporting Information).
The possible interaction between C 60 Cl 6 and Sn 2+ is further illustrated by X-ray photoelectron spectroscopy (XPS).As shown in Figure 1d, the binding energy of Sn 3d 5/2 shifted −0.5 eV in the perovskite films with C 60 Cl 6 additives compared with the pristine perovskite films, indicating the existence of the coordination interaction between the chlorine of C 60 Cl 6 and Sn 2+ ions. [38,39]The result is also confirmed by XPS spectra of Cl 2p (Figure S4, Supporting Information).A significant shift in the peak position of Cl 2p 1/2 and Cl 2p 3/2 in the perovskite films with C 60 Cl 6 additives was found.
Therefore, we propose a C 60 Cl 6 -assisted strategy to enhance the device performance of TPSCs (Figure 1e).First, the interaction between the chlorines of C 60 Cl 6 and Sn 2+ ions could regulate the crystallization process and improve the quality of the perovskite film.Meanwhile, the iodide ions could interact with the electrondeficient C 60 cage, which can not only passivate negatively charged defects such as Sn-I antisite defects but also suppress the ion migration. [35]urthermore, the chemical interaction also could increase the formation energy of tin vacancy, decreasing Sn-induced deep traps. [39]ast but not least, the chemical interactions between C 60 Cl 6 and perovskite at the grain boundaries may enable in situ stitching of the grain boundaries, which can effectively inhibit the erosion of water and oxygen on the perovskite film, thereby improving the stability of the perovskite film and corresponding TPSCs.
The surface morphology of the perovskite films with and without C 60 Cl 6 is investigated by the scanning electron microscope (SEM).As shown in Figure 2a, there are abundant pinholes in the control sample, which can be attributed to the uncontrolled crystallization during the film formation process. [40]In contrast, the C 60 Cl 6 -based perovskite film presents larger grains and significantly reduced pinholes due to the efficient regulation of the crystallization process (Figure 2b).The improved perovskite film could suppress carrier recombination and inhibit the corrosion of O 2 and H 2 O.Meanwhile, the distribution of C 60 Cl 6 in the perovskite films is also characterized by the energy-dispersive spectrum (EDS).As shown in Figure S5, Supporting Information, the same distribution of chlorine and tin indicates that C 60 Cl 6 is uniformly distributed in the tin-based perovskite film.Considering that fullerenes are unlikely to enter the perovskite lattice due to their large size, C 60 Cl 6 should be uniformly located at the perovskite grain boundaries.We then measure the steady-state photoluminescence (PL) and timeresolved photoluminescence (TRPL) of the corresponding perovskite films.In Figure 2c, the PL intensity of C 60 Cl 6 -based perovskite film is noteworthily stronger than the control sample, which suggests that C 60 Cl 6 suppresses trap-assisted non-radiative recombination.The slightly blue-shifted PL peak wavelength of the control sample may be related to the formation of a more low-dimensional structure as observed in previous works. [41,42]Moreover, the narrower full width at half maximum (FWHM) of the C 60 Cl 6 (58.77nm) implies fewer deep-level defects and higher perovskite crystallinity.The improvement of perovskite crystallinity is in good agreement with the calculation of Urbach energy (Figure S6, Supporting Information).[45] As shown in Figure 2d and Table S1, Supporting Information, the optimal perovskite film with C 60 Cl 6 displays a nearly threefold longer lifetime (1.77 ns) than that of the control sample film (0.67 ns), which is in good agreement with the PL results.
We further quantify the trap densities by measuring the spacecharge-limited current on the electron-only devices for control and C 60 Cl 6 -based samples with the structure of ITO/SnO 2 /perovskite/ PCBM/BCP/Ag.The values of the onset voltage of the trap-filled limit region (V TFL ) of the devices without and with C 60 Cl 6 are determined to be 0.314 and 0.223 V, respectively (Figures 2e,f).And the trap-state density (N trap ) is estimated by the following formula: [45,46] where ε r is the relative permittivity, ε 0 is the permittivity of vacuum, q is the charge of an electron, and L means the thickness of the perovskite film.The N trap is estimated to be 1.47 × 10 16 cm −3 with the incorporation of the C 60 Cl 6 .In contrast, the pristine film shows a relatively high N trap (2.07 × 10 16 cm −3 ).
Above all results have confirmed that C 60 Cl 6 -based perovskite may display better photovoltaic performance.Then, we fabricate the devices based on the structure of ITO/PEDOT: PSS/perovskite/PCBM/BCP/Ag (Figure 3a).We first optimize the concentration of C 60 Cl 6 .The best device is achieved at a concentration of 0.5 mg mL −1 (Figure S7 and Table S2, Supporting Information), which is selected for further study.Figure 3b shows the current density-voltage (J-V) curves of the devices without and with C 60 Cl 6 .The C 60 Cl 6 -based device yields a PCE of 13.30%, along with the short-circuit photocurrent density (J SC ) of 20.31 mA cm −2 , open-circuit voltage (V OC ) of 0.86 V, and fill factor (FF) of 0.76.In comparison, the control device presents a PCE of 10.03%, along with the J SC of 19.43 mA cm −2 , V OC of 0.72 V, and FF of 0.72 (Table S3, Supporting Information).The reproducibility of the device is also verified by 30 independent devices (Figure S8, Supporting Information).The average J SC , V OC , and FF of the devices with C 60 Cl 6 are substantially higher than those of the control devices.We attribute the improvement in device performance to the high-quality perovskite films, which are induced by the interaction between C 60 Cl 6 and perovskite components.We also test the steady photocurrent outputs and corresponding power outputs of the TPSCs.As shown in Figures 3d,e, the control device yields a steady PCE of 9.64%, while the devices with C 60 Cl 6 demonstrate a higher steady PCE of 13.10%, both agree well with their PCE values from the J-V measurements.
To further elaborate carrier recombination in the devices, we conduct transient photovoltage (TPV) experiments. [47]As shown in Figure 3f, the carrier recombination lifetime (τ r ) of the C 60 Cl 6 -based device is as high as 65.42 ns, while the τ r of the control device is only 39.99 ns (Table S4, Supporting Information).Meanwhile, the dark J-V curves are recorded to obtain the magnitude of leakage current.As shown in Figure 3g, the C 60 Cl 6 -based device showed a smaller leakage current than that of the control devices, which is consistent with the data of TPV.The decreased carrier recombination in C 60 Cl 6 -based devices is beneficial for reducing their V OC losses.As shown in Figure 3h, the C 60 Cl 6 -based device yields an obviously improved built-in potential (V bi ; 0.64 V) compared with the control device (0.49 V).The improved V bi could promote carrier separation and decrease V OC loss.
To accurately evaluate the V OC loss due to carrier recombination in the devices, the electroluminance (EL) spectra of the devices are recorded under 1.5 V.As shown in Figure 3i, the C 60 Cl 6 -based device shows a stronger EL intensity at 885 nm, which implies that the incorporation of C 60 Cl 6 benefits reducing non-radiative recombination.
The external radiative efficiency (ERE) [9,48] (Figure S9, Supporting Information) from the EL measurement is employed to evaluate the V OC loss (ΔV nrad oc ) of non-radiative recombination based on the following equation: [47]

(ΔV
where the K B is Boltzmann's constant and T is temperature.By calculation, the ΔV nrad oc are determined to be 231.30and 128.79 mV for the control and C 60 Cl 6 -based device, respectively.The reduced V OC loss of C 60 Cl 6 -based devices is reasonably attributed to their improved V OC . Perovskite films with good environmental tolerance are critical for their corresponding device stability.First, we record the X-ray diffraction (XRD) patterns of the control and C 60 Cl 6 -based perovskite films before and after exposure to air conditions.As shown in Figure 4a, for the control perovskite film, two distinct diffraction peaks appear near the peaks of ( 101) and (202) crystal planes as the increase in exposure time.The two new peaks may be attributed to the crystal planes of the water-perovskite complex formed by the interaction of water and perovskite. [49,50]In contrast, the C 60 Cl 6 -based perovskite films show significantly enhanced stability after 10 h of storage under the same conditions (Figure 4b).This result is consistent with the results of the water contact angle measurement, the color, UV-Vis spectra, and The oxidation of Sn 2+ in the corresponding perovskite film is also detected by X-ray photoelectron spectroscopy (XPS).As shown in Fig- ures 4c,h, the ratios of Sn 4+ /Sn 2+ of the perovskite films with and without C 60 Cl 6 gradually increase with a prolonging exposure time to air.After the perovskite films are placed in air for 600 min, the Sn 4+ / Sn 2+ value increases from 0.27 to 6.09 in the control sample, while the Sn 4+ /Sn 2+ value increases from 0.14 to 3.11 in the C 60 Cl 6 -based one.The results are summarized in Figure S14, Supporting Information.The lower Sn 4+ /Sn 2+ ratio benefits from the higher-quality C 60 Cl 6based perovskite film, which can suppress the corrosion of the water and oxygen in the air.The water and oxygen resistance of the C 60 Cl 6based perovskite layer lays the foundation for the stability of its device.
The long-term stability of the devices is also an important indicator for practical applications.While ion migration is one of the important factors affecting device stability, [35] we performed the EL measurements using different applied voltages to explore the ion migration.As shown in Figure 5a, the peak of the control device presents a significant blueshifted from 885 to 861 nm with an increasing bias voltage from 1.5 to 3.5 V, which may be attributed to the migration of iodine ions. [51- 53]While the C 60 Cl 6 -based device shows slightly blue-shifted from 885 to 875 nm (Figure 5b), which can be attributed to the suppressed ion migration caused by the chemical interactions between C 60 Cl 6 and perovskite components.Therefore, the suppressed ion migration can be expected to increase the thermal and illumination stability of the C 60 Cl 6 -based device. [35]e then systematically evaluate the long-term stability of the unencapsulated TPSCs without and with C 60 Cl 6 .As shown in Figure 5c, the C 60 Cl 6 -based device retains 90% of its initial efficiency after 600 min of thermal aging (85 °C).By contrast, the control device presents poor thermal stability, which only retains 40% under the same condition.Meanwhile, we further test the stability at open-circuit (OC) conditions, and the C 60 Cl 6 -based device also displays enhanced stability under light illumination (Figure S15, Supporting Information).
Furthermore, for the TPSCs, the oxidation of the Sn 2+ can cause serious degradation of device performance.We evaluate the stability of unencapsulated TPSCs in ambient conditions with a relative humidity of 20 AE 5%.As shown in Figure 5d, after 10 h of aging, the control devices present a rapid PCE decay, maintaining about 60% of the initial value.In contrast, the C 60 Cl 6 -based devices exhibit excellent air stability and can maintain 95% of the initial efficiency.The improved stability of C 60 Cl 6based devices not only can be attributed to higher-quality perovskite films induced by the chemical interactions between C 60 Cl 6 and perovskite components but also due to the inhibition of ion migration of C 60 Cl 6 .

Conclusion
In this work, a fullerene derivative, C 60 Cl 6 , is synthesized and used as an additive to regulate the crystallization process of perovskite film and improve the device's performance.The quality of perovskite films is significantly improved by the induction of the chemical interactions between C 60 Cl 6 and perovskite components.Combined with the passivation effect of fullerene, the C 60 Cl 6 -based films exhibit significantly reduced defect density.As a result, the C 60 Cl 6 -based device presents an efficiency of 13.30%, which is much higher than that of the control device (10.03%).Moreover, the C 60 Cl 6 -based devices exhibit significantly improved air, light soaking, and thermal stability due to the functions of grain boundary stitching and ion migration suppression of C 60 Cl 6 .This work provides an effective strategy to regulate the tinbased perovskite film quality, which could guide the design of functional materials for high-performance TPSCs in the future.

Figure 1 .
Figure 1.a) Molecular structure and electrostatic potential of C 60 Cl 6 .Photographs of perovskite solution b) without and c) with C 60 Cl 6 exposed to the air at different times.d) X-ray photoelectron spectroscopy spectra of Sn 3d 5/2 in control and doping C 60 Cl 6 -based perovskite film.e) Schematic diagram of the C 60 Cl 6 -assisted strategy for device performance enhancement.
Moreover, the external quantum efficiency (EQE) measurement of the control and C 60 Cl 6 -based devices in Figure 3c produces 19.35 and 20.32 mA cm −2 in integrated J SC , which shows a minor difference with the J-V measurement.The improved J SC of C 60 Cl 6 -based device can be reasonably attributed to the stronger absorption of the C 60 Cl 6 -based perovskite film between 630 and 770 nm.

Figure 2 .
Figure 2. Scanning electron microscope images of perovskite film a) without and b) with C 60 Cl 6 treatment.c) Steady-state PL and d) time-resolved photoluminescence spectra of perovskite film without and with C 60 Cl 6 .Dark I-V curve of the electron-only device e) without and f) with C 60 Cl 6 additive.

Figure 3 .
Figure 3. a) Schematic diagram of the device structure.b) J-V curves of the control and C 60 Cl 6 -based devices.c) External quantum efficiency spectra and the integrated current density curves of the control and C 60 Cl 6 -based TPSCs.Stabilized photocurrent densities and the steady power output at the maximum power point of d) control and e) C 60 Cl 6 -based TPSCs.f) Charge carrier lifetime from normalized TPV decay for control and C 60 Cl 6 -based devices.g) J-V curves of the control and C 60 Cl 6 -based TPSCs under dark conditions.h) Mott-Schottky analysis at 10 kHz.i) EL spectra of the devices under a bias voltage of 1.5 V.

Figure 4 .
Figure 4. X-ray diffraction patterns of perovskite films a) without and b) with C 60 Cl 6 additive exposed to the air for different times.X-ray photoelectron spectroscopy spectra of Sn 3d 5/2 of the control sample, c) fresh, d) aged in the air for 360 min, and e) 600 min.XPS spectra of Sn 3d 5/2 of C 60 Cl 6 -based perovskite film, f) fresh and g) aged in the air for 360 min and h) 600 min.

Figure 5 .
Figure 5. Normalized electroluminance spectra of the a) control and b) C 60 Cl 6 -based perovskite devices under different applied voltages.c) Thermal stability measurement at 85 °C in a filled with N 2 glovebox.d) Stability curves of the control and C 60 Cl 6 -based devices in the air under dark conditions.