Materials Chemistry C

(1)

Cite this:J. Mater. Chem. C,2019, 7, 3852

Eﬃcient and carbon-based hole transport

layer-free CsPbI

₂

Br planar perovskite solar cells using PMMA modification†

Xiang Zhang,âYang Zhou,âYuzhu Li,âJiawen Sun,âXubing Lu, âXingsen Gao, â Jinwei Gao, âLingling Shui, ^bSujuan Wu *âand Jun-Ming Liu âc

In this work, planar inorganic perovskite solar cells (PSCs) with the simple structure of glass/ITO/SnO2/ CsPbI2Br/C have been fabricated. Solution-processed poly(methyl methacrylate) (PMMA) is selected to modify the CsPbI2Br film. The effects of PMMA modification on the microstructure of the CsPbI2Br film and the photoelectric properties of inorganic PSCs have been systematically investigated. The concentration of the PMMA solution has been optimized. At the optimal concentration, the modified carbon-based hole transport layer (HTL)-free CsPbI2Br PSCs demonstrate an improved open-circuit voltage (Voc) and fill factor (FF), and achieve a champion efficiency of 10.95% with less hysteresis behavior, this being much higher than the 9.14% of the reference PSCs. Moreover, the unencapsulated CsPbI2Br PSCs modified by PMMA demonstrate improved stability and retain about 95% of their initial efficiency after being exposed to air (relative humidity of 25–35%) for 20 days. The enhanced performance is mainly attributed to the reduced trap state density and suppressed charge recombination. This work provides a simple route to fabricate efficient and stable carbon-based HTL-free inorganic PSCs.

Introduction

Organic–inorganic hybrid perovskite materials have attracted lots of attention in the field of perovskite solar cells (PSCs), due to their excellent photoelectric properties such as long carrier diffusion length,^1–3large absorption coefficient,^4–6high carrier mobility^7,8 and so on. Until now, the certified efficiency of organic–inorganic hybrid PSCs has exceeded 23%.⁹ However, the device stability still needs to be improved. It has been reported that the intrinsic stability of perovskite materials can be improved by replacing the organic cation with inorganic Cs⁺.^10,11Although the CsPbI3material has a reasonable bandgap of 1.73 eV, its cubic phase is unstable in the ambient atmosphere.^11,12The CsPbBr3 material shows better stability, but its large bandgap (Eg = 2.3 eV) is not suitable for

photovoltaic applications.^11,12 Fortunately, the mixed-halide CsPbI2Br material has an appropriate bandgap of 1.91 eV and exhibits a better phase stability in ambient air.¹¹The theoretical efficiency of CsPbI₂Br PSCs is predicted to be up to 22.1%.¹³ Obviously the CsPbI₂Br material is a promising candidate for photovoltaic applications. A high-quality CsPbI₂Br film is necessary for efficient charge extraction and transport in PSCs.¹⁴ However, most of the CsPbI₂Br films are fabricated by a simple one-step spin-coating method.^15,16 It is difficult to prepare a pinhole-free and uniform CsPbI2Br film due to its slow crystallization.¹¹Moreover, it is reported that most of the charge recombination and the material degradation occur at the grain boundaries and interfaces.¹⁷ To improve the performance of CsPbI2Br PSCs, a series of studies have been done to control the crystallization process,^18–23 reduce the trap states^24–26 or optimize the interface.10,12,27,28 For example, Liu et al. have synthesized Lewis base adducts as precursors to fabricate CsPbI2Br PSCs with an efficiency of 14.78%.²⁰Tian et al.have achieved an efficiency of 12.34% by reducing the trap states on the CsPbI₂Br film using Pb(NO₃)₂.²⁵Jinet al. have attained an efficiency of 14.45% by modifying the CsPbI₂Br/PTAA interface using CsPbI₃ quantum dots.¹⁰ Recently, the efficiency of CsPbI₂Br PSCs has been enhanced to 16.07%.²³Regrettably, an organic hole transport layer (HTL) and evaporated gold (Au) electrode are used in these efficient CsPbI2Br PSCs. This not only increases the cost, but also reduces the stability of the PSCs.

aInstitute for Advanced Materials and Guangdong Provincial Key Laboratory of Quantum Engineering and Quantum Materials, South China Academy of Advanced Optoelectronics, South China Normal University, Guangzhou 510006, P. R. China.

E-mail: [email protected]

bGuangdong Provincial Key Laboratory of Optical Information Materials and Technology, South China Academy of Advanced Optoelectronics, South China Normal University, Guangzhou 510006, P. R. China

cLaboratory of Solid State Microstructures, Nanjing University, Nanjing 210093, P. R. China

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

c9tc00374f

Received 21st January 2019, Accepted 24th February 2019 DOI: 10.1039/c9tc00374f

rsc.li/materials-c

Materials Chemistry C

PAPER

Published on 01 March 2019. Downloaded by NANJING UNIVERSITY on 4/27/2019 9:48:47 AM.

View Article Online

View Journal | View Issue

(2)

To solve these problems, all-inorganic CsPbI2Br PSCs based on carbon (C) electrodes have been developed.^11,29 But the eﬃciency of carbon-based CsPbI₂Br PSCs is still much lower than that of organic–inorganic hybrid PSCs. Although carbon- based inorganic PSCs have achieved an efficiency over 10%, it is still lower than that of inorganic PSCs based on an evaporated metal electrode and an organic HTL.^29,30It is reported that the lower efficiency in carbon-based inorganic PSCs is closely related to the high trap state density.^11,29,30The trap states can not only cause ion migration, current density–voltage (J–V) hysteresis and device degradation in the ambient environment,³¹ but also generate shunt-leakage paths due to direct contact between the C electrode and the electron transport layer (ETL), resulting in large charge recombination and a reduced open circuit voltage (Voc) and fill factor (FF) of PSCs.^32,33It is necessary to remove or passivate the trap states in order to obtain efficient and stable carbon-based inorganic PSCs. It is found that charge recombination can be suppressed by incorporating polymers with Lewis base groups.³⁴Up to now, various polymers have been used to reduce the trap states in organic–inorganic hybrid PSCs to promote charge extraction and collection.17,32,35–39 It has been identified that PMMA can passivate the trap states at grain boundaries and interfaces, resulting in reduced charge recombination and improved performance in organic–inorganic hybrid PSCs.^40–42However, PMMA has not been used to improve the performance of inorganic PSCs, to the best of our knowledge.

Based on these considerations, inorganic PSCs with the simple structure of glass/ITO/SnO2/CsPbI2Br/C were fabricated. The PMMA solution was spin-coated on the CsPbI2Br film. For con- venience of description, two types of carbon-based CsPbI2Br PSCs are discussed in this work: the reference-film/PSC which refers to the CsPbI2Br film/PSC without PMMA modification, and the PMMA-film/PSC which refers to the CsPbI2Br film/PSC with PMMA modification. The eﬀects of PMMA modification on the microstructure of the CsPbI₂Br film and the photoelectric properties of carbon-based CsPbI₂Br PSCs have been investigated.

The results indicate that the PMMA modification can eﬀectively improve the performance of carbon-based CsPbI₂Br PSCs. At the optimal PMMA solution concentration, the champion eﬃciency of the PMMA-PSC is improved to 10.95% from 9.14% of the reference-PSC. Moreover, the PMMA-PSC shows lessJ–Vhysteresis behavior and improved stability. A series of measurements have been carried out to investigate the related mechanisms for the enhanced performance. It is found that the PMMA modification can effectively suppress the non-radiative recombination induced by the trap states in the CsPbI2Br film and the recombination at the CsPbI2Br/C interface, resulting in promoted charge extraction.

This work provides a facile and feasible strategy to prepare efficient and stable carbon-based HTL-free inorganic PSCs.

Experimental

Materials

Cesium iodide (CsI) (99.999%, metals basis), cesium bromide (CsBr) (99.999%, metals basis), PMMA (molecular weight: 43982),

chlorobenzene (99.9%, spectrophotometric grade) and tin(IV) oxide (SnO2) solution (15 wt% in H2O colloidal dispersion) were bought from Alfa Aesar. Lead(II) iodide (PbI₂) (99.99%) and lead(II) bromide (PbBr₂) (99.99%) were purchased from Xi’an Polymer Light Technology Corp. Dimethyl sulfoxide (DMSO) (Z99.9%) and indium-doped tin oxide (ITO) glass substrates (15 Ohm per square) were obtained from Sigma-Aldrich and Nippon Sheet Glass (NSG) Company Limited of Japan, respectively. Conductive carbon paste (printing ink) (CH-8 (MOD2)) was purchased from Jujo Printing Supplies & Technology (Pinghu) Co., Ltd, and its related properties are as follows: surface resistivity: 10.0Ocm² (measured by the four-probe method); volume resistivity: 1.0 10²Ocm (measured by the four-probe method); specific gravity:

1.15 (201C); adhesion: 100/100; heat resistance: 1501C/72 hour with the resistance changing within 10%; moisture resistance:

90% relative humidity/240 hour, 40 1C with the resistance changing within 10%. All the materials were directly used without any further purification.

Device fabrication

The glass/ITO substrate was patterned by etching with diluted hydrochloric acid solution and zinc powder, then successively cleaned with detergent, deionized water, acetone, isopropanol and ethanolviaan ultrasonic process for 15 min, respectively.

Fig. S1 (ESI†) shows the schematic drawing of the fabrication of the PMMA-PSC. After the glass/ITO substrate was dried with a nitrogen (N2) flow and UV-ozone treatment for 5 min, the SnO2

film was prepared by spin-coating a SnO2solution (3 wt%) on the glass/ITO substrate at 4000 rpm for 30 s and then annealed at 150 1C for 30 min in air. After cooling down to room temperature, the glass/ITO/SnO2 substrate was treated with UV-ozone for 15 min, and then the mixture CsPbI2Br precursor solution of CsBr : CsI : PbBr2: PbI2(molar ratio = 0.5 : 1 : 0.5 : 1) in DMSO at 0.8 M (filtered with a 0.22mm nylon filter) was spin- coated on the glass/ITO/SnO₂substrate at 4000 rpm for 30 s and annealed at 2601C for 10 min to get the CsPbI₂Br film in air (a relative humidity of 25–35%). For the PMMA-PSC, PMMA chlorobenzene solutions with different concentrations were respectively spin-coated on the CsPbI2Br film at a speed of 3000 rpm for 30 s and annealed at 100 1C for 10 min. Sub- sequently, the commercial carbon paste was doctor-bladed on the sample surface. Finally, the sample was annealed at 1201C for 20 min. The size of the carbon-based CsPbI2Br PSCs was 0.085 cm².

Characterization

TheJ–Vcharacteristics of PSCs were measured using a source meter (Keithley 2420) under illumination of 100 mW cm² (Newport 91160, AM 1.5G) calibrated by a standard silicon solar cell (certified by NREL) as the reference. ForJ–Vmeasure- ments, the scan voltage was from0.2 V to 1.5 V. The scan rate and delay time were 100 mV s¹and 50 ms, respectively. The crystallinity of the perovskite film was investigated using an X-ray diffractometer (XRD) (PANalytical X’Pert PRO) using a Cu Karadiation source. The morphology and contact angle of the perovskite film were characterized by scanning electron

(3)

microscopy (SEM, ZEISS ULTRA 55) and a video-based optical contact angle measurement instrument (Dataphysics OCA Pro 15, Germany), respectively. The UV-vis absorption spectra were obtained using a SHIMADZU UV-2550 spectrophotometer. A standard EQE system (Newport 66902) was used to measure the external quantum efficiency (EQE). Atomic Force Microscopy (AFM) (Asylum Research, Cypher) was employed to investigate the surface morphology and local contact potential difference (CPD) of the perovskite films. The current–voltage (I–V) curves were acquired using a Keithley 2420 source meter with a scan rate of 100 mV s¹in the dark. The steady-state photoluminescence (PL) spectra were measured using a fluorescence spectrophotometer (HITACHIF-5000), exciting at 322 nm. The time- resolved photoluminescence (TRPL) decay spectra were recorded with a time-correlated single photon counting method using a fluorescence spectrophotometer (Fluorolog-3, Horiba), exciting at 473 nm. The electrochemical impedance spectroscopy (EIS) measurement was carried out with an electrochemical workstation (Zahner Zennium, Germany) under 10 mW cm² white LED light, applying a 10 mV AC signal with the frequency ranging from 1 Hz to 1 MHz under a voltage bias of1.1 V.

All samples without any encapsulation were stored and tested in air (relative humidity of 25–35%).

Results and discussion

Fig. 1a–c show the structure and the cross-sectional SEM image of the PMMA-PSC and the molecular structure of PMMA, respectively. It can be seen that the cross-sectional SEM image demonstrates a well-defined layer-by-layer structure. The thick- nesses of different components in the PMMA-PSC have been provided in the details. To more clearly distinguish the thickness of the PMMA layer and the C electrode, the cross-sectional SEM images for the PMMA-film modified by the PMMA solution with a concentration of 0.3 mg mL¹ and the C electrode are shown in Fig. S2 and S3 (ESI†). The thickness of the PMMA layer is about 9.2 nm, which is similar to previous reports.^32,41–43The effect of the PMMA solution concentration on the J–V performance of the PSCs has been investigated.

Fig. S4 (ESI†) shows theJ–Vcurves and the detailed photovoltaic parameters including the short-circuit current density (Jsc),Voc, FF and power conversion efficiency (PCE) of PMMA-PSCs as a function of PMMA solution concentration. Moreover, the specific values of the photovoltaic parameters are summarized in Table S1 (ESI†). It can be seen that the PCE of PMMA-PSCs increases with the increase of the PMMA solution concentration at the beginning. When the PMMA solution concentration is increased to 0.3 mg mL¹, the PMMA-PSC demonstrates the highest PCE.

It is noted that theJscin PMMA-PSCs showed a slightly down- ward tendency with the increase of the PMMA concentration.

This may be attributed to the inherently insulating property of PMMA, which increases the series resistance (Rs) of the device, leading to the reduced Jscin PMMA-PSCs.32,37,40,44,45Then the device performance becomes worse with the increase of the PMMA solution concentration when it is over 0.3 mg mL¹. Therefore, the optimal concentration for the PMMA solution is 0.3 mg mL¹, which is the condition for PMMA-PSCs in all of our other experiments from Fig. 2–9, as well as Fig. S2 and S5 (ESI†).

It is well-known that the morphology and crystallinity of the perovskite film have a significant impact on the photovoltaic performance of PSCs.¹¹ Regrettably, cracks/pinholes in the perovskite film can cause trap states, which will damage the eﬃciency and stability of PSCs.³²Fig. 2a–d show the top-view SEM images of the reference-film and PMMA-film with different scale bars, respectively. As shown in Fig. 2a–d, it is noted that the reference-film exhibits relatively obvious cracks/pinholes at the grain boundary, while the cracks/pinholes become unclear in the PMMA-film. This indicates that the solution-processed PMMA fills the cracks/pinholes at the grain boundary of the CsPbI2Br film, which can alleviate the surface trap states and block the direct contact between the SnO2 ETL and the C electrode.^32,43,46 The contact angles of water directly dropped on the reference-film and PMMA-film are inserted in Fig. 2a and c, respectively. It can be seen that the contact angle on the PMMA-film is much larger. The increased contact angle suggests that the hydrophobicity in the PMMA-film is improved, which will facilitate prevention of moisture intrusion and improve the device stability.²⁷ It will be further discussed in the following.

The XRD patterns of the reference-film and PMMA-film deposited

Fig. 1 (a) The schematic illustration of the PMMA-PSC structure. (b) The cross-sectional SEM image of the PMMA-PSC. (c) The molecular structure of PMMA.

(4)

on the glass/ITO/SnO2 substrate are shown in Fig. 2e. The diffraction peaks at 14.6 and 29.51are assigned to the (100) and (200) crystal planes of CsPbI2Br, respectively. This is consistent with previous reports.18,19,22,25The XRD pattern of the PMMA-film exhibits a similar peak intensity to the reference-film and no other phases appear, indicating that the PMMA modification does not change the structure of CsPbI2Br.

In order to investigate the eﬀect of PMMA modification on the photovoltaic performance, the reference-PSC and PMMA-PSC have been fabricated. Fig. 3a shows the J–V curves of the reference-PSC and PMMA-PSC measured in the reverse scan direction. It is noticed that the PMMA-PSC achieves a PCE of 10.95% with aV_ocof 1.202 V, aJ_scof 12.64 mA cm²and a FF of 0.71. While the reference-PSC shows a PCE of 9.14% with aV_oc of 1.145 V, aJscof 12.88 mA cm²and a FF of 0.62. Compared to the reference-PSC, the PMMA-PSC exhibits an increasedVocand FF. The FF is related to the ratio of the shunt resistance (Rsh)

toRs(Rsh/Rs).³²The higher FF in the PMMA-PSC is attributed to the largerRsh/Rs.³⁷In order to verify the reproducibility of the reference-PSC and PMMA-PSC, the distributions of the PCE,Jsc, Voc, FF andRsh/Rsfor the two types of PSCs extracted from 35 devices are shown in Fig. 3b–f, respectively. It can be seen that the dispersions of all parameters in the PMMA-PSCs are nar- rower than those of the reference-PSCs, which implies better reproducibility in the PMMA-PSC.^10,17It is well-known thatJ–V hysteresis behavior is a widely observed phenomenon in PSCs.^47,48There is a seriousJ–Vhysteresis behavior in mixed- halide inorganic PSCs owing to the iodide and bromide phase segregation under illumination.²⁰ To illustrate the effect of PMMA modification on the hysteresis behavior of the device, the J–V curves of the reference-PSC and PMMA-PSC were measured in both the reverse and forward scan directions.

Fig. 4a and b show theJ–Vcurves and the detailed photovoltaic parameters are listed. Here the hysteresis index (HI) is used to Fig. 2 The top-view SEM images with diﬀerent scale bars and contact angles: (a and b) the reference-film; (c and d) the PMMA-film. Inset: Contact angles for the reference-film and PMMA-film, respectively. (e) The XRD patterns of the reference-film and PMMA-film deposited on the glass/ITO/SnO2

substrate.

Fig. 3 (a) TheJ–Vcurves of the reference-PSC and PMMA-PSC measured in the reverse scan direction. (b) Efficiency statistics histograms of the reference-PSC and PMMA-PSC. (c)J_sc; (d)V_oc; (e) FF and (f)R_sh/R_sdistributions of the reference-PSC and PMMA-PSC from 35 devices.

(5)

characterize the J–V hysteresis behavior. The HI is defined according to eqn (1):¹⁷

HI¼PCE_reversePCE_forward

PCE_reverse (1)

Fig. 4c demonstrates the distributions of the HI for the reference-PSCs and PMMA-PSCs from 15 devices. The average HI for the reference-PSCs and PMMA-PSCs is 0.191 and 0.114, respectively. Obviously the PMMA-PSC shows the smaller HI.

The reduced HI in the PMMA-PSC indicates that the J–V hysteresis behavior is suppressed compared to the reference- PSC. It is reported that theJ–Vhysteresis behavior is related to the steady-state current density as a function of time.⁴⁸Fig. S5 (ESI†) shows the steady-state current density of the reference- PSC and PMMA-PSC as a function of time. It can be seen that the current density of the PMMA-PSC quickly saturates at about 40 s while it saturates at about 80 s for the reference-PSC.

The faster stabilization of the current density can partly explain the smaller HI in the PMMA-PSC.^49,50 The detailed reasons for the reduced HI in the PMMA-PSC will be further discussed in the following.

To confirm the reliability of theJ–Vmeasurements, the plots of steady-state current density and PCE as functions of time for Fig. 4 TheJ–Vcurves of the best-performing devices in both the reverse and forward scan directions: (a) the reference-PSC; (b) the PMMA-PSC.

(c) Hysteresis index (HI) distribution of the reference-PSC and PMMA-PSC from 15 devices. The steady-state PCE and current density as a function of time measured at their maximum power output point: (d) the reference-PSC; (e) the PMMA-PSC. (f) Evolution of the PCE over time measured at a relative humidity of about 25–35%.

Fig. 5 (a) The UV-vis absorption spectra of the reference-film and PMMA-film deposited on the glass/ITO/SnO2substrate; (b) the EQE and integrated current density curves of the reference-PSC and PMMA-PSC.

Fig. 6 AFM images: (a) the reference-film; (b) the PMMA-film. KPFM images:

(c) the reference-film; (d) the PMMA-film. The scan area was 88mm.

The scale bar is 1mm.

(6)

the reference-PSC and PMMA-PSC have been recorded, as shown in Fig. 4d and e. The reference-PSC and PMMA-PSC

were biased at their respective optimal bias voltage (Vopt) at the maximum power output point. The values of Vopt for the Fig. 7 The darkI–Vcurves of electron-only devices withV_TFLkink point behavior: (a) the reference device, (b) the device modified by PMMA. Inset:

The corresponding schematic illustration of the electron-only device structure.

Fig. 8 (a) Steady-state PL spectra; (b) TRPL decay spectra of the reference-film and PMMA-film deposited on the quartz glass substrate. (c) Nyquist plots of the reference-PSC and PMMA-PSC measured under light illumination. The solid symbol represents the experimental data and the solid line is the fitting results. Inset: The equivalent circuit diagram used to fit the data of the Nyquist plots. (d)R_traandR_recfrom the fitting results of the Nyquist plots.

Fig. 9 (a)V_oc; (b)J_sc; and (c) FF as a function of the light intensity for the reference-PSC and PMMA-PSC.

(7)

reference-PSC and PMMA-PSC are 0.851 and 0.963 V, respectively. Under standard AM 1.5G continuous illumination for 130 s, the PCE of the reference-PSC and PMMA-PSC stabilizes at 9.05% and 10.89% with a current density of 10.64 and 11.31 mA cm², respectively. These values are close to the corresponding values obtained from the J–V measurements.

This confirms the reliability of theJ–Vresults. Fig. 4f shows the normalized PCE of the reference-PSC and PMMA-PSC as a function of storage time. After being exposed to air with 25–35% relative humidity at 20–251C for 20 days, the PCE for the reference-PSC and PMMA-PSC remains at about 78% and 95% of its initial value, respectively. Obviously, the PMMA modification enhances the device stability.

To study the eﬀect of PMMA modification on the optical absorption of the CsPbI2Br film, the UV-vis absorption spectra of the reference-film and PMMA-film deposited on the glass/

ITO/SnO2substrate were measured, as shown in Fig. 5a. It can be seen that the two films demonstrate similar absorption.

This result confirms that the PMMA modification does not change the optical absorption of the CsPbI₂Br film. Fig. 5b displays the EQE and integrated current density curves of the reference-PSC and PMMA-PSC. It is noted that the PMMA-PSC shows a slightly reduced EQE. The integrated current density for the reference-PSC and PMMA-PSC is 12.52 and 12.26 mA cm², respectively. These values are quite comparable to the values ofJsc

derived from theJ–Vcurves (errors less than 4%).

Fig. 6a and b display the AFM images of the reference-film and PMMA-film, respectively. The values of the root mean square roughness (RMS) for the reference-film and PMMA-film are 14.78 and 11.90 nm, respectively. Obviously the PMMA-film is smoother compared to the reference-film. KPFM is an eﬀective way to investigate the contact potential distribution on local grain boundaries and the grain bulk.^51–54It is reported that the change of the obtained average CPD from KPFM at the nano- meter scale is consistent with the bulk measurement, which is correlated with the device performance as well.⁵³To investigate the effect of PMMA modification on the local electrical properties of the perovskite film at the nanoscale level, the average CPD was measured by KPFM. Fig. 6c and d display the KPFM images and the obtained average CPD of the reference-film and PMMA-film, respectively. It is noted that the PMMA-film exhibits a higher average CPD than that of the reference-film. The increased CPD is related to the higher Voc of the PMMA-PSC, which corresponds to effective photo-generated charge separation.⁵³ This is in agreement with theJ–Vresults.

In order to further explore the reason for the enhanced performance in the PMMA-PSC, the electron trap state density (Ntrap) of the reference-film and PMMA-film was characterized.

The values ofN_trapcan be obtained from the darkI–Vcurves of electron-only devices with the structure of glass/ITO/SnO₂/ CsPbI₂Br/PMMA(with or without)/PCBM/Ag. As shown in Fig. 7a and b, the current is linearly dependent on voltage in the range of low bias voltage up to a kink point, which indicates an ohmic response.⁵⁵ When the voltage increases over the kink point, the current demonstrates a faster exponential rise. This can be attributed to the completely occupied trap states by the

injected charge.⁵⁶The voltage at the kink point is defined as the trap-filled limit voltage (VTFL), which can be obtained from the dark I–V curves. N_trap can be calculated by the following eqn (2):²⁰

N_trap¼2ee₀V_TFL

eL² (2)

whereeis the elementary charge of an electron (e= 1.610¹⁹C), Lis the thickness of the CsPbI2Br film (LB316 nm, as shown in Fig. 1b and Fig. S6, ESI†), ande₀andeare the vacuum permittivity (e₀ = 8.854 10¹⁴ F cm¹) and relative dielectric constant (e B 16.4) of the CsPbI₂Br film,²⁰ respectively. As shown in Fig. 7a and b, theV_TFLfor the reference-film and PMMA-film is 1.980 and 1.263 V, respectively. According to eqn (2), the calculated N_trap is 3.6 10¹⁶ and 2.3 10¹⁶ cm³ for the reference-film and PMMA-film, respectively. This indicates that the trap states in the CsPbI2Br film are effectively passivated by PMMA. The reduced trap state density in the PMMA-film is beneficial to suppress charge recombination, which also con- tributes to decreasing the HI and enhance theVocand FF in the PMMA-PSC.^11,57,58This can explain the improved performance in the PMMA-PSC.

It is necessary for PSCs to reduce the charge recombination in order to achieve excellent photovoltaic performance.⁵⁷ PL and EIS spectra were measured to investigate the eﬀect of PMMA modification on charge recombination. Fig. 8a and b show the steady-state PL spectra and TRPL decay spectra of the reference-film and PMMA-film deposited on the quartz glass substrate. Steady-state PL spectra are generally used to estimate charge non-radiative recombination induced by the trap states in perovskite films.²³ As seen in Fig. 8a, the steady-state PL emission peaks of the reference-film and PMMA-film are around 650 nm, which is consistent with previous reports.^11,14,27More- over, the intensity of steady-state PL spectra for the PMMA-film shows a significant increase compared with the reference-film.

The enhanced PL intensity indicates fewer trap states in the PMMA-film, which also illustrates that charge non-radiative recombination is suppressed.11,23,39,57This can be further con- firmed by TRPL decay spectra. Fig. 8b shows the TRPL decay spectra of the reference-film and PMMA-film, which can be fitted with the following bi-exponential decay eqn (3):¹⁴

y(t) =A1e^t/t¹+A2e^t/t²+y0 (3) whereA1andA2are the relative amplitudes,t₁is the fast decay component, related to non-radiative recombination, t₂is the slow decay component, related to radiative recombination, and y₀ is a constant for the base-line oﬀset.⁵⁷ The relevant key parameters resulting from the fitted results are listed in Table S2 (ESI†). The average carrier lifetimes (t_ave) can be calculated by the following formula (4):⁵⁹

t_ave¼ A₁t₁ A1t1þA2t2

t₁þ A₂t₂ A1t1þA2t2

t₂ (4)

The obtainedt_avefor the reference-film and PMMA-film is 14.74 and 16.91 ns, respectively. The larger t_ave in the PMMA-film demonstrates that the PMMA passivates the trap states in the

(8)

CsPbI2Br film to suppress charge non-radiative recombination.^20,35 The suppressed charge recombination will contribute to enhance the V_oc and FF in the PMMA-PSC.⁶⁰ Therefore, the PMMA-PSC demonstrates improved photovoltaic performance. EIS measurements are used to investigate the interfacial charge transfer and recombination process.^24,50Fig. 8c shows the Nyquist plots of the reference-PSC and PMMA-PSC measured at a bias of1.1 V under illumination conditions, where the solid lines are the fitting results using the inserted model. In the fitted Nyquist plots, the value of the starting point at the real part (Z⁰) corresponds toRs, and the transfer resistance (Rtra) is ascribed to the high-frequency arc, which is associated with the charge transfer at the SnO2/CsPbI2Br/C interface; the recombination resistance (Rrec) is assigned to the low-frequency arc.⁶¹For more reliable fitting, the ideal capacitance is replaced by a constant phase angle element (CPEtraand CPErec) to illustrate the spatial inhomogeneity induced by defects and impurities at the interface.⁶² These fitted parameters are summarized in Table S3 (ESI†). This result shows that the PMMA modification slightly increase theR_s of the PSCs due to the inherently insulating property of PMMA.^32,37,40Since the SnO₂/CsPbI₂Br interface in the reference-PSC and PMMA-PSC is under the same conditions,R_traandR_recare only related to the charge transfer and recombination process at the CsPbI₂Br/C interface. As shown in Fig. 8d, the values of R_tra for the reference-PSC and PMMA-PSC are 9.22 and 7.65Ocm², respectively. While the values of Rrec for the reference-PSC and PMMA-PSC are 24.83 and 56.12Ocm², respectively. Obviously the PMMA-PSC shows the lowerRtraand the higherRrecvalue compared to those of the reference-PSC. This indicates less interfacial recombination loss between the PMMA-film and the C electrode, which is responsible for the increasedVocand FF.^50,57 Therefore, the EIS measurement results confirm that the PMMA modification can suppress the charge recombination, resulting in promoted charge extraction and improved device performance.⁵⁰

TheJ–Vcharacteristics (V_oc,J_scand FF) as a function of light intensity (P_light) have been measured to further investigate the charge recombination mechanisms in the PSCs. The linear relationship of Vocand the natural logarithm of Plightcan be plotted according to the following eqn (5):⁶³

V_oc¼nkT

q ln P_light

þconstant (5) wheren,k,Tandqare the ideal factor, Boltzmann constant, absolute temperature and elementary charge, respectively.

There is no charge extraction under open-circuit conditions, indicating that all photo-generated charge carriers recombine within the PSCs.⁶⁴The slope of Vocversus Plightwill be equal to kT/qwhen trap-assisted recombination is absent or bimolecular recombination determines the performance of the PSCs.^23,64,65 In contrast, if trap-assisted recombination is involved in the device operation, the slope would be larger than kT/q.⁶⁴ As shown in Fig. 9a, the reference-PSC and PMMA-PSC show a slope of 1.75kT/qand 1.57kT/q, respectively. Moreover, theVoc

of the reference-PSC demonstrate a stronger dependence on Plight.⁶⁴ This result indicates that the PMMA modification

reduces trap-assisted recombination in the PSCs.^12,20,64 The dependence ofJsconPlightcould be described by the power law relationship (6):⁶³

J_scpP^a_light (ar1) (6)

whereais an exponential factor. Here, bimolecular recombination is expected to be a minimum for the maximum carrier sweep out (aE1) at the short circuit.⁶³As displayed in Fig. 9b, the values ofafor both the reference-PSC (a= 0.96) and PMMA- PSC (a= 0.98) are close to 1, indicating negligible bimolecular recombination in the devices.⁶⁶The dependence of the FF on Plightis presented in Fig. 9c. At relatively lowPlight( just under 0.5 sun), the FF of the reference-PSC and PMMA-PSC appears steady around a certain constant and almost remains unchanged.

AsPlightcontinues to increase, the FF of the reference-PSC appears more obviously decreased than that of the PMMA-PSC. This can be attributed to trap-assisted recombination.⁶⁷All of these char- acterizations indicate that the PMMA modification at the CsPbI₂Br/C interface can reduce the charge recombination and improve the photovoltaic performance of PSCs.

Conclusions

In summary, we have successfully demonstrated that PMMA modification is an eﬀective and facile method to improve the photovoltaic performance of inorganic PSCs. At the optimal PMMA concentration, the PMMA-PSC achieves a highest PCE of 10.95% and a steady-state PCE of 10.89%, as well as less J–V hysteresis behavior and improved device stability. The improved performance can be attributed to the reduced non- radiative recombination induced by the trap states in the CsPbI2Br film and recombination at the CsPbI2Br/C interface, resulting in promoted charge extraction. This work provides a new and simple route to fabricate efficient and stable carbon- based HTL-free inorganic PSCs.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We acknowledge the financial support of the National Key R & D Program of China (2016YFB0401502, 2016YFA0201002), the National Natural Science Foundation of China (Grant No.

51431006, 51472093, 61574065, 51571094), the Project for Guangdong Province Universities and Colleges Pearl River Scholar Funded Scheme (2016), the Characteristic Innovation Project of Guangdong Provincial Department of Education (Science 2016, 22), the Natural Science Foundation of Guangdong Province (No. 2016A030313421, 2016A030308019), the Program for Changjiang Scholars and Innovative Research Team in Uni- versity (IRT_17R70), the Guangdong Innovative Research Team Program (No. 2011D039), the MOE International Laboratory for Optical Information Technologies, the Science and Technology

(9)

Planning Project of Guangdong Province (Grant No.

2016B090906004, 2015B090927006), Guangdong Provincial Engineering Technology Research Center for Transparent Con- ductive Materials and Guangdong Provincial Key Laboratory of Optical Information Materials and Technology (2017B030301007).

Notes and references

1 S. D. Stranks, G. E. Eperon, G. Grancini, C. Menelaou, M. J. P. Alcocer, T. Leijtens, L. M. Herz, A. Petrozza and H. J. Snaith,Science, 2013,342, 341–344.

2 G. Xing, N. Mathews, S. Sun, S. S. Lim, Y. M. Lam, M. Gra¨tzel, S. Mhaisalkar and T. C. Sum, Science, 2013, 342, 344–347.

3 Q. Dong, Y. Fang, Y. Shao, P. Mulligan, J. Qiu, L. Cao and J. Huang,Science, 2015,347, 967–970.

4 S. De Wolf, J. Holovsky, S.-J. Moon, P. Lo¨per, B. Niesen, M. Ledinsky, F.-J. Haug, J.-H. Yum and C. Ballif, J. Phys.

Chem. Lett., 2014,5, 1035–1039.

5 Z. Xiao, Q. Dong, C. Bi, Y. Shao, Y. Yuan and J. Huang, Adv. Mater., 2014,26, 6503–6509.

6 F. Hao, C. C. Stoumpos, D. H. Cao, R. P. H. Chang and M. G. Kanatzidis,Nat. Photonics, 2014,8, 489–494.

7 C. Wehrenfennig, G. E. Eperon, M. B. Johnston, H. J. Snaith and L. M. Herz,Adv. Mater., 2014,26, 1584–1589.

8 L. M. Herz,ACS Energy Lett., 2017,2, 1539–1548.

9 N. J. Jeon, H. Na, E. H. Jung, T.-Y. Yang, Y. G. Lee, G. Kim, H.-W. Shin, S. Il Seok, J. Lee and J. Seo,Nat. Energy, 2018,3, 682–689.

10 H. Bian, D. Bai, Z. Jin, K. Wang, L. Liang, H. Wang, J. Zhang, Q. Wang and S. Liu,Joule, 2018,2, 1500–1510.

11 C. Dong, X. Han, Y. Zhao, J. Li, L. Chang and W. Zhao, Sol. RRL, 2018,2, 1800139.

12 L. Yan, Q. Xue, M. Liu, Z. Zhu, J. Tian, Z. Li, Z. Chen, Z. Chen, H. Yan, H.-L. Yip and Y. Cao,Adv. Mater., 2018, 30, 1802509.

13 Q. Zeng, X. Zhang, C. Liu, T. Feng, Z. Chen, W. Zhang, W. Zheng, H. Zhang and B. Yang,Sol. RRL, 2018,3, 1800239.

14 Y. Gao, Y. Dong, K. Huang, C. Zhang, B. Liu, S. Wang, J. Shi, H. Xie, H. Huang, S. Xiao, J. He, Y. Gao, R. A. Hatton and J. Yang,ACS Photonics, 2018,5, 4104–4110.

15 J. K. Nam, M. S. Jung, S. U. Chai, Y. J. Choi, D. Kim and J. H. Park,J. Phys. Chem. Lett., 2017,8, 2936–2940.

16 J. K. Nam, S. U. Chai, W. Cha, Y. J. Choi, W. Kim, M. S. Jung, J. Kwon, D. Kim and J. H. Park, Nano Lett., 2017, 17, 2028–2033.

17 J. Jiang, Q. Wang, Z. Jin, X. Zhang, J. Lei, H. Bin, Z.-G.

Zhang, Y. Li and S. Liu,Adv. Energy Mater., 2018,8, 1701757.

18 L. Zhang, B. Li, J. Yuan, M. Wang, T. Shen, F. Huang, W. Wen, G. Cao and J. Tian,J. Phys. Chem. Lett., 2018,9, 3646–3653.

19 D. Bai, H. Bian, Z. Jin, H. Wang, L. Meng, Q. Wang and S. Liu,Nano Energy, 2018,52, 408–415.

20 G. Yin, H. Zhao, H. Jiang, S. Yuan, T. Niu, K. Zhao, Z. Liu and S. Liu,Adv. Funct. Mater., 2018,28, 1803269.

21 Y. Wang, T. Zhang, F. Xu, Y. Li and Y. Zhao,Sol. RRL, 2018, 2, 1700180.

22 H. Rao, S. Ye, F. Gu, Z. Zhao, Z. Liu, Z. Bian and C. Huang, Adv. Energy Mater., 2018,8, 1800758.

23 W. Chen, H. Chen, G. Xu, R. Xue, S. Wang, Y. Li and Y. Li, Joule, 2019,3, 191–204.

24 Z. Zeng, J. Zhang, X. Gan, H. Sun, M. Shang, D. Hou, C. Lu, R. Chen, Y. Zhu and L. Han, Adv. Energy Mater., 2018, 8, 1801050.

25 J. Yuan, L. Zhang, C. Bi, M. Wang and J. Tian,Sol. RRL, 2018, 2, 1800188.

26 Q. Zeng, X. Zhang, X. Feng, S. Lu, Z. Chen, X. Yong, S. A. T. Redfern, H. Wei, H. Wang, H. Shen, W. Zhang, W. Zheng, H. Zhang, J. S. Tse and B. Yang,Adv. Mater., 2018, 30, 1705393.

27 J. Zhang, D. Bai, Z. Jin, H. Bian, K. Wang, J. Sun, Q. Wang and S. Liu,Adv. Energy Mater., 2018,8, 1703246.

28 C. Liu, W. Li, C. Zhang, Y. Ma, J. Fan and Y. Mai, J. Am.

Chem. Soc., 2018,140, 3825–3828.

29 Y. Guo, F. Zhao, J. Tao, J. Jiang, J. Zhang, J. Yang, Z. Hu and J. Chu,ChemSusChem, 2019, DOI: 10.1002/cssc.201802690.

30 X. Meng, Z. Wang, W. Qian, Z. Zhu, T. Zhang, Y. Bai, C. Hu, S. Xiao, Y. Yang and S. Yang,J. Phys. Chem. Lett., 2019,10, 194–199.

31 Y. Guo, J. Ma, H. Lei, F. Yao, B. Li, L. Xiong and G. Fang, J. Mater. Chem. A, 2018,6, 5919–5925.

32 F. Wang, A. Shimazaki, F. Yang, K. Kanahashi, K. Matsuki, Y. Miyauchi, T. Takenobu, A. Wakamiya, Y. Murata and K. Matsuda,J. Phys. Chem. C, 2017,121, 1562–1568.

33 Y. Xiong, X. Zhu, A. Mei, F. Qin, S. Liu, S. Zhang, Y. Jiang, Y. Zhou and H. Han,Sol. RRL, 2018,2, 1800002.

34 H. Li, K. Zhu, K. Zhang, P. Huang, D. Li, L. Yuan, T. Cao, Z. Sun, Z. Li, Q. Chen, B. Song, H. Zhu and Y. Zhou, Org. Electron., 2019,66, 47–52.

35 M. Kim, S. G. Motti, R. Sorrentino and A. Petrozza,Energy Environ. Sci., 2018,11, 2609–2619.

36 S.-H. Turren-Cruz, A. Hagfeldt and M. Saliba,Science, 2018, 362, 449–453.

37 Y. Cai, Z. Zhang, Y. Zhou, H. Liu, Q. Qin, X. Lu, X. Gao, L. Shui, S. Wu and J. Liu, Electrochim. Acta, 2018, 261, 445–453.

38 L. Zuo, H. Guo, D. W. deQuilettes, S. Jariwala, N. De Marco, S. Dong, R. DeBlock, D. S. Ginger, B. Dunn, M. Wang and Y. Yang,Sci. Adv., 2017,3, e1700106.

39 B. Chaudhary, A. Kulkarni, A. K. Jena, M. Ikegami, Y. Udagawa, H. Kunugita, K. Ema and T. Miyasaka,Chem- SusChem, 2017,10, 2473–2479.

40 J. Peng, J. I. Khan, W. Liu, E. Ugur, T. Duong, Y. Wu, H. Shen, K. Wang, H. Dang, E. Aydin, X. Yang, Y. Wan, K. J. Weber, K. R. Catchpole, F. Laquai, S. De Wolf and T. P. White,Adv. Energy Mater., 2018,8, 1801208.

41 C. Wu, H. Li, Y. Yan, B. Chi, K. M. Felice, R. B. Moore, B. A. Magill, R. R. H. H. Mudiyanselage, G. A. Khodaparast, M. Sanghadasa and S. Priya,Sol. RRL, 2018,2, 1800052.

42 F. Zhang, J. Song, R. Hu, Y. Xiang, J. He, Y. Hao, J. Lian, B. Zhang, P. Zeng and J. Qu,Small, 2018,14, 1704007.

(10)

43 F. Yang, H. E. Lim, F. Wang, M. Ozaki, A. Shimazaki, J. Liu, N. B. Mohamed, K. Shinokita, Y. Miyauchi, A. Wakamiya, Y. Murata and K. Matsuda,Adv. Mater. Interfaces, 2018,5, 1701256.

44 D. Bi, C. Yi, J. Luo, J.-D. De´coppet, F. Zhang, S. M.

Zakeeruddin, X. Li, A. Hagfeldt and M. Gra¨tzel,Nat. Energy, 2016,1, 16142.

45 L. Deng, K. Wang, H. Yang, H. Yu and B. Hu,J. Phys. D: Appl.

Phys., 2018,51, 475102.

46 F. Li, J. Yuan, X. Ling, Y. Zhang, Y. Yang, S. H. Cheung, C. H. Y. Ho, X. Gao and W. Ma,Adv. Funct. Mater., 2018, 28, 1706377.

47 H. J. Snaith, A. Abate, J. M. Ball, G. E. Eperon, T. Leijtens, N. K. Noel, S. D. Stranks, J. T.-W. Wang, K. Wojciechowski and W. Zhang,J. Phys. Chem. Lett., 2014,5, 1511–1515.

48 H.-S. Kim, I.-H. Jang, N. Ahn, M. Choi, A. Guerrero, J. Bisquert and N.-G. Park,J. Phys. Chem. Lett., 2015,6, 4633–4639.

49 K.-H. Jung, J.-Y. Seo, S. Lee, H. Shin and N.-G. Park,J. Mater.

Chem. A, 2017,5, 24790–24803.

50 D. Yang, X. Zhou, R. Yang, Z. Yang, W. Yu, X. Wang, C. Li, S. Liu and R. P. H. Chang,Energy Environ. Sci., 2016, 9, 3071–3078.

51 P. Cui, P. Fu, D. Wei, M. Li, D. Song, X. Yue, Y. Li, Z. Zhang, Y. Li and J. M. Mbebgue,RSC Adv., 2015,5, 75622–75629.

52 J. Li, B. Huang, E. Nasr Esfahani, L. Wei, J. Yao, J. Zhao and W. Chen,npj Quantum Mater., 2017,2, 56.

53 M. Salado, R. K. Kokal, L. Calio, S. Kazim, M. Deepa and S. Ahmad,Phys. Chem. Chem. Phys., 2017,19, 22905–22914.

54 W. Melitz, J. Shen, A. C. Kummel and S. Lee,Surf. Sci. Rep., 2011,66, 1–27.

55 C. Chen, Y. Xu, S. Wu, S. Zhang, Z. Yang, W. Zhang, H. Zhu, Z. Xiong, W. Chen and W. Chen,J. Mater. Chem. A, 2018,6, 7903–7912.

56 W. Li, C. Zhang, Y. Ma, C. Liu, J. Fan, Y. Mai and R. E. I. Schropp,Energy Environ. Sci., 2018,11, 286–293.

57 J. Chen and N.-G. Park,Adv. Mater., 2018, 1803019.

58 J. Chen, S.-G. Kim and N.-G. Park, Adv. Mater., 2018, 30, 1801948.

59 R. Li, H. Zhang, M. Zhang and M. Guo,Appl. Surf. Sci., 2018, 458, 172–182.

60 S.-G. Kim, J. Chen, J.-Y. Seo, D.-H. Kang and N.-G. Park, ACS Appl. Mater. Interfaces, 2018,10, 25372–25383.

61 X. Liu, Z. Liu, B. Sun, X. Tan, H. Ye, Y. Tu, T. Shi, Z. Tang and G. Liao,Electrochim. Acta, 2018,283, 1115–1124.

62 W. Wang, Z. Zhang, Y. Cai, J. Chen, J. Wang, R. Huang, X. Lu, X. Gao, L. Shui, S. Wu and J.-M. Liu,Nanoscale Res.

Lett., 2016,11, 316.

63 J. Duan, Y. Zhao, B. He and Q. Tang,Angew. Chem., Int. Ed., 2018,130, 3849–3853.

64 Q. Xue, G. Chen, M. Liu, J. Xiao, Z. Chen, Z. Hu, X.-F. Jiang, B. Zhang, F. Huang, W. Yang, H.-L. Yip and Y. Cao, Adv. Energy Mater., 2016,6, 1502021.

65 T. Singh and T. Miyasaka, Adv. Energy Mater., 2018, 8, 1700677.

66 J. Ding, Y. Zhao, J. Duan, B. He and Q. Tang,ChemSusChem, 2018,11, 1432–1437.

67 J. Zhang, H. Luo, W. Xie, X. Lin, X. Hou, J. Zhou, S. Huang, W. Ou-Yang, Z. Sun and X. Chen, Nanoscale, 2018, 10, 5617–5625.