Understanding the mechanism of PEDOT: PSS modification via solvent on the morphology of perovskite films for efficient solar cells

Synthetic Metals 243 (2018) 17–24

Contents lists available at ScienceDirect

Synthetic Metals journal homepage: www.elsevier.com/locate/synmet

Understanding the mechanism of PEDOT: PSS modification via solvent on the morphology of perovskite films for efficient solar cells

T

Qiaoli Niua, Wentao Huanga, Jing Tonga, Hao Lva, Yunkai Denga, Yuhui Maa, Zhenhua Zhaoa, ⁎ ⁎ ⁎ Ruidong Xiaa, , Wenjin Zenga, Yonggang Mina,b, , Wei Huanga, a

Key Laboratory for Organic Electronics and Information Displays & Jiangsu Key Laboratory for Biosensors, Institute of Advanced Materials (IAM), Jiangsu National Synergetic Innovation Center for Advanced Materials (SICAM), Nanjing University of Posts & Telecommunications, 9 Wenyuan Road, Nanjing 210023, PR China b The School of Materials and Energy, Guangdong University of Technology, Panyu, Guangzhou 510006, PR China

A R T I C LE I N FO

A B S T R A C T

Keywords: PEDOT: PSS Modification DMSO Smooth surface Coordination ability

The properties of hole transport layers (HTL) greatly affect the performance of inverted planar heterojunction perovskite solar cells (PSC). To understand the mechanisms involved, solvent treatments to HTL poly(3,4ethylenedioxythiophene) polystyrene sulfonate (PEDOT: PSS) by DMSO, ethanol, ethanol:10 vol% DMSO, DMF and acetonitrile were performed when fabricating the PSC devices. Scanning electron microscopy (SEM) and Xray diffraction (XRD) results show that the crystal size of CH3NH3PbI3 (MAPbI3) increased after solvent treatment to PEDOT: PSS. MAPbI3 film with a uniform crystal size and smooth surface was obtained when being deposited on top of PEDOT: PSS that has been modified with solvent with a strong coordination ability. In the other cases, crystal size homogeneity and surface roughness of the MAPbI3 film was worse. The improved surface roughness and crystal size homogeneity of the MAPbI3 film enhanced the short circuit current (JSC), and therefore the power conversion efficiency (PCE) of PSC. It was found that both the stability of the intermediate phase MAI-DMSO-PbI2, which is related to the coordination ability of the solvent molecules, and the morphology of PEDOT: PSS film, influenced the morphology of the CH3NH3PbI3 (MAPbI3) film. Our findings provide the basis for better controlling the morphology of MAPbI3 films in inverted planar heterojunction PSC.

1. Introduction Perovskite solar cells (PSCs) are a promising photovoltaic technology because of their high power conversion efficiency (PCE) and solution processing technique, which is compatible to future flexible techniques, such as roll-to-roll [1–3]. The PCE has now exceeded 20% [4] narrowing the gap to the PCE of commercialized Si based solar cells [5]. Even though PSCs still suffer from instability and lack of reliable manufacturing techniques [6], they have good prospects for application. Planar heterojunction PSCs evolved from dye-sensitized solar cells, and have become a new direction for PSCs [7]. Compared with the n-i-p structure device, which commonly use TiO2 and spiro-OMeTAD as electron and hole transport layers (ETL and HTL), the device with p-i-n structure (also called inverted planar structure) has a lower fabricating temperature and a much weaker voltage-current density hysteresis [8,9]. Due to the sandwich device structure with perovskite film layered between the ETL and HTL, the device performance is influenced greatly by the properties of the ETL and HTL [10,11]. Poly(3,4-

⁎

ethylenedioxythiophene) polystyrene sulfonate (PEDOT: PSS) is the most commonly used HTL because of its simple solution processing technique, relatively high electrical conductivity, and smooth surface morphology. However, the properties of PEDOT: PSS are far from perfect [12–14]. According to the previous reports, the performance of PSCs could be significantly improved by optimizing the properties of PEDOT: PSS [15–23]. For example, the electrical conductivity of PEDOT: PSS increased after surface modification by methylammonium iodide (MAI) solution, resulting in an increase in short circuit current (JSC) of the PSC [18]. The incorporation of PSSNa with PEDOT: PSS increased its work function, leading to an increase in open circuit voltage (VOC) of the PSC [19–21]. By modifying the pH value of PEDOT: PSS with a mild base, the surface texture and electronic properties of PEDOT: PSS were tuned, which promoted quality and crystallization of the perovskite film deposited on top of it [22]. Huang et al. found that after doping dimethyl sulfoxide (DMSO) into PEDOT: PSS, the aggregation of PEDOT-rich on the surface could induce the crystalline of perovskite on top of it, leading to an increase of perovskite crystal size [23]. Hence, PEDOT: PSS plays an important role in achieving high

Corresponding authors. E-mail addresses: [email protected] (R. Xia), [email protected] (Y. Min), [email protected] (W. Huang).

https://doi.org/10.1016/j.synthmet.2018.05.012 Received 18 December 2017; Received in revised form 25 May 2018; Accepted 27 May 2018 0379-6779/ © 2018 Published by Elsevier B.V.

Synthetic Metals 243 (2018) 17–24

Q. Niu et al.

than 1 ppm. Following this, the solvent treatment was carried out by spin-coating polar solvents of DMSO, ethanol, ethanol/ 10 vol% DMSO, DMF or acetonitrile at 2000 rpm for 1 min, respectively. Then, MAPbI3 precursor solution was spin-coated. Specifically, the spin-coating process was composed of two stages: 1000 rpm for 15 s and then 4000 rpm for 25 s. For toluene washing during the spin-coating of MAPbI3 precursor solution, at delay time of 16 s from the beginning of the second stage, 600 μL of toluene was dripped. After being thermally annealed at 100 °C for 10 min., PCBM solution in chlorobenzene (40 mg/mL) was spin-coated at 3000 rpm for 20 s and then 6000 rpm for 20 s. After that, the substrates were annealed at 70 °C for 40 min. Finally, 10 nm-thick BCP and 150 nm-thick Ag were sequentially thermally evaporated at a basic pressure of 3 × 10−4 Pa. The active area defined by a shadow mask was 0.096 cm2.

performance PSCs. As is well-known, PEDOT: PSS has also been widely used in organic photovoltaic and organic light-emitting diodes (OLEDs) [24,25]. Various approaches have been tried to improve the properties of PEDOT: PSS. Solvent treatment on the surface of PEDOT: PSS was a simple and effective method. After being treated by methanol, ethanol, H2O, DMF, DMSO or ethylene glycol, the electrical conductivity of PEDOT: PSS can be increased [20,21,26,27]. However, the PCE of PSC decreased after treating PEDOT: PSS by H2O or H2O: ethanol [21]. Recently, we also noted that the performance of PSC based on solvent treated PEDOT: PSS by ethanol, DMF, DMSO and acetonitrile were very different from each other despite the similar properties of the treated PEDOT: PSS layers. That is to say, apart from the electrical conductivity of PEDOT: PSS, there must exist other factors that influence the performance of PSCs after solvent treatment. However, the precise mechanism was not well understood because the previous reports were mainly focused on the properties of PEDOT: PSS, without considering the interaction between solvent molecules and perovskite precursor solution. Hence, it is necessary to understand the influence of solvent treatment to PEDOT: PSS on the performance of PSCs in detail. In this work, we designed a series of experiments to investigate the effect of solvent treatment to PEDOT: PSS on the morphology of CH3NH3PbI3 (MAPbI3) and the performance of PSCs. Polar solvents such as DMSO, ethanol, ethanol/10 vol% DMSO, DMF and acetonitrile were used to perform solvent treatment. After being treated by different solvents, PEDOT: PSS films exhibited similar properties including surface morphology, optical absorption and electrical conductivity. However, the morphology of MAPbI3 films deposited on top of the solvent treated PEDOT: PSS were different, resulting in significant differences in solar cell performance. It was found that MAPbI3 film with a smooth surface and homogeneous crystals can be obtained if deposited on PEDOT: PSS treated by solvent with a strong coordination ability, leading to an improved PCE of the PSC. In other cases, PEDOT: PSS modified by a solvent with a weak coordination ability increased the surface roughness and grain inhomogeneity of MAPbI3 film, which deteriorated the performance of the PSC. Detailed study suggested that both of the stability of the intermediate phase, which depends on the coordination ability of solvent molecules used for PEDOT: PSS treatment, and the morphology of PEDOT: PSS were important in controlling the morphology of MAPbI3 film and the performance of the PSC.

2.3. Device characterizations The current density-voltage (J–V) curves were collected by a Keithley 2602 source meter under illumination of 1 sun (100 mW/cm2 AM 1.5 G) provided by a solar simulator (Oriel/Newport, model 94,043 A, USA), which was calibrated with a Si photodiode. The scan voltage step size was 0.01715 V. The scan directions from 0 to 1.2 V and from 1.2 to 0 V were referred to as forward scan and reverse scan, respectively. The surface morphology of MAPbI3 films was investigated using a Hitachi S4800 microscope (SEM). The microscopic surface morphology measurements were conducted by Bruker icon Dimension with scan Asyst atomic force microscopy (AFM). X-ray diffraction (XRD) experiments were conducted by a Bruker D8 ADVANCE X-ray diffractometer. The absorption spectra were measured by a UV–vis spectrophotometer (UV3600). Incident photon-to-current efficiency (IPCE) data were collected by using a QTest Station1000. The steadystate PL spectra were collected by a steady-state spectrometer (Edinburgh Instruments F900) under excitation by a Xenon lamp (Xe900, 520 nm). The contact angles were examined by a contact angle measurement (KRUSS DSA20). All the above measurements were carried out in atmosphere, and the devices were not encapsulated. 3. Results PSC with configuration of ITO/PEDOT: PSS (20 nm)/MAPbI3 (380 nm)/PCBM (65 nm)/BCP (10 nm)/Ag (150 nm) was fabricated. Polar solvents of DMSO, ethanol, ethanol/ 10 vol% DMSO, DMF and acetonitrile were spin-coated on the thermal annealed PEDOT: PSS, respectively, to perform solvent treatment. The photovoltaic characteristics of PSCs with or without solvent treatment to PEDOT: PSS were compared. Fig. 1a showed the J–V plots of PSCs as a function of the solvent used for PEDOT: PSS treatment. The champion control device exhibited a PCE of 11.6%, with a VOC of 0.98 V, a JSC of 16.73 mA/cm2 and a fill factor (FF) of 70.6%. The PCE value of the control cell is comparable to the reported results from similar device structures [23,28–30]. After PEDOT: PSS was treated by using DMSO, ethanol or ethanol/10 vol% DMSO, the JSC values of the corresponding champion devices increased to 17.9, 17.59 and 19.61 mA/cm2, respectively. Meanwhile, the FF increased slightly to 73.6, 73.1 and 73.2%, while, the VOC values slightly reduced to 0.896, 0.937 and 0.90 V, respectively. As a result, the PCE increased to 11.77, 12.04 and 12.97%, respectively. That is to say, the highest PCE value was achieved by treating PEDOT: PSS using ethanol/10 vol% DMSO. The performance values (with both forward scan and reverse scan) of the device based on ethanol/10 vol% DMSO treated PEDOT: PSS were also shown in Table 1 and Fig. S1a. The very small deviation between forward scan and reverse scan demonstrated a negligible voltage-current density hysteresis. The significant JSC enhancement was further demonstrated by the IPCE data shown in Fig. 1b. The calculated JSC values of control device and devices based on solvent treated PEDOT: PSS by DMSO, ethanol and ethanol/10 vol% DMSO were 14.79, 17.34,

2. Experimental methods 2.1. Materials and solvents PbI2 (99.99%), PC61BM (99.5%), MAI, γ-Butyrolactone (GBL), DMSO and toluene (98%) were all purchased from Sigma Aldrich and used as received. PEDOT: PSS (P4083) was purchased from Bayer. The MAPbI3 precursor solution was composed of 1.4 M PbI2 and CH3NH3I (1:1/n:n) in DMSO and GBL (7:3/v:v). Before spin-coating, the MAPbI3 precursor solution was stirred at 66 °C for 24 h in N2 atmosphere glove box. 2.2. Solar cell fabrication Perovskite solar cells with device configurations of ITO/PEDOT: PSS (20 nm)/ MAPbI3 (380 nm)/PCBM (65 nm)/ bathocuproine (BCP) (10 nm)/Ag (150 nm) were fabricated. Before using, the ITO substrates underwent a succession of ultrasonic cleaning in deionized water, acetone and ethanol, for 20 min. for each. After being dried by the N2 flow, 5 min. of ozone plasma was applied to remove any organic residues. Immediately after, 20 nm of PEDOT: PSS film was spin-coated onto the pre-cleaned ITO-coated glass at 3000 rpm for 50 s and baked at 150 °C for 20 min. The thickness was determined by profilometry (Bruker DektakXT). After that, the ITO substrates were quickly transferred into a N2 protected glovebox with water and oxygen content less 18

Synthetic Metals 243 (2018) 17–24

Q. Niu et al.

Fig. 1. J–V curves (a) and IPCE data together with calculated JSC values (b) of the best PSCs based on PEDOT: PSS with or without solvent treatment.

general after treating PEDOT: PSS, the crystals size distribution of MAPbI3 film varied dramatically with different solvents. MAPbI3 films with the narrowest and widest crystal size distribution were formed for ethanol/ 10 vol% DMSO and DMF treatment (Table 2), respectively. These results suggested that the device with the most homogeneous MAPbI3 crystal had the highest PCE, while MAPbI3 with the largest but non-uniform crystal size destroyed device performance. This phenomenon was in accordance with the previous reports that a homogeneous small crystal was better than a non-uniform large one in terms of achieving efficient PSCs [31,32]. It was clear that solvent treatment on PEDOT: PSS enlarged the crystal size of MAPbI3 films, which can also be confirmed by the XRD patterns (Fig. S2a). Two intense peaks at 13.97° and 28.3° originated from the (110) and (220) planes of MAPbI3 crystal were observed from all MAPbI3 films, indicating the unchanged crystal growth orientation along (110) planes. However, the full width at half maximum (FWHM) values of the most intense peak at 13.98° changes with the solvent used for PEDOT: PSS treatment. As shown in Fig. S2b, the FWHM values for the MAPbI3 films deposited on the pristine and treated PEDOT: PSS film by DMSO, ethanol, ethanol/10 vol% DMSO, DMF and acetonitrile were 0.119°, 0.100°, 0.101°, 0.103°, 0.081°, and 0.081°, respectively. According to the Scherer Formula, large crystal size results in narrow FWHM values [33]. Therefore, the narrow FWHM after solvent treatment to PEDOT: PSS confirmed the increase of MAPbI3 crystal size, which was in agreement with SEM results (Fig. 2). Even though the crystal size of MAPbI3 film was enlarged after DMF or acetonitrile treatment, the PCE values of the PSCs significantly reduced. To understand this unexpected PCE reduction, atomic force microscopy (AFM) was used to examine the morphology of MAPbI3 films as shown in Fig. S3. The root mean square (rms) roughness value of MAPbI3 film on pristine PEDOT: PSS was 10.6 ± 0.4 nm. For the MAPbI3 film on PEDOT: PSS treated by DMSO, ethanol or ethanol/ 10 vol% DMSO, the rms roughness values reduced to 8.8 ± 0.4, 8.4 ± 0.3 and 8.1 ± 0.2 nm, respectively. In contrast, the rms roughness values increased to 10.8 ± 0.2 and 12.2 ± 0.1 nm for the MAPbI3 film deposited on PEDOT: PSS treated by DMF or acetonitrile. According to the previous report, the reduction of surface roughness was in favor of the enhancement of solar cell performance [34]. In this work, the optimal PSC was obtained from the device based on ethanol/ 10 vol% DMSO treated PEDOT: PSS, which had the smoothest MAPbI3 film. MAPbI3 film with rough surface has lots of defects, which act as recombination centers, leading to large leakage current, and therefore, low JSC [34]. This could be one of the reasons for the low PCE of PSCs based on DMF or acetonitrile treated PEDOT: PSS. To inspect the morphology change of MAPbI3 films deposited on the solvent treated PEDOT: PSS, the morphology and optical absorption properties of PEDOT: PSS were characterized. Fig. 3 shows the AFM

Table 1 The performance parameters of PSCs based on PEDOT: PSS with or without solvent treatment. treatment of PEDOT: PSS

VOC (V)

JSC (mA/ cm2)

FFa

PCE (%)

Rsb

untreated

0.98 0.93 0.896 0.91 0.937 0.926 0.90 0.89 0.89 0.88 0.926 0.93 0.936 0.932

16.73 16.77 17.9 17.53 17.59 16.56 19.61 19.3 20.4 18.9 15.19 13.44 12.9 12.46

70.6 73.6 73.6 73.58 73.1 73.37 73.2 75 72.6 72.8 66.3 60.23 71.4 59.8

92.56 – 43.4 – 55.5 – 49.2 42.8 42.3 – 71.58 – 54.8 –

11.6 11.17 11.77 11.68 12.04 11.26 12.97 12.88 13.19 12.2 9.33 7.55 8.63 6.92

DMSO ethanol ethanol/10 vol% DMSO

acetonitrile DMF

a

champion average champion average champion average champion forward reverse average champion average champion average

FF is refer to fill factor. Rs is refer to series resistance.

b

16.39 and 19.01 mA/cm2, respectively. These values were in good agreement with the measured data. The average values of device performance parameters from about ten individual devices were also demonstrated in Table 1, which showed slightly decreased VOC and almost unchanged FF values (around 73%) after solvent treatments to PEDOT: PSS. Hence, the boost in PCE was mainly attributed to the considerable enhancement of JSC. In addition, the concentration of DMSO in ethanol was optimized. Fig. S1b shows the J–V curves of PSCs based on PEDOT: PSS treated by ethanol/ x vol% DMSO (x = 5, 10 and 20). All devices based on the ethanol/DMSO treated PEDOT: PSS demonstrated increased PCE values compared with the control device. The best performance was obtained from a device based on ethanol/10 vol% DMSO treatment. On the other hand, after DMF or acetonitrile treatment to PEDOT: PSS, both the JSC and FF of the devices decreased significantly, resulting in very low PCE values of 8.63% and 9.33%, respectively. The morphology of MAPbI3 films plays a vital role in solar cell performance. Therefore, MAPbI3 films deposited on PEDOT: PSS with or without solvent treatment were examined via SEM as shown in Fig. 2. Full coverage of MAPbI3 crystal on PEDOT: PSS substrates were observed for all MAPbI3 films. However, the size and uniformity of the crystals were very different. The crystal sizes were analyzed by using Nano measurer and the statistical values were summarized in Table 2. The mean crystal size values of MAPbI3 films on PEDOT: PSS without solvent treatment and with DMSO, ethanol, ethanol/ 10 vol% DMSO, DMF or acetonitrile treatment were 150, 190, 180, 160, 210 and 220 nm, respectively. Although, MAPbI3 crystal size increased in 19

Synthetic Metals 243 (2018) 17–24

Q. Niu et al.

Fig. 2. Top and cross-sectional SEM images of perovskite films on top of PEDOT: PSS without (a, a-1) or with solvent treatment by DMSO (b, b-1), ethanol (c, c-1), ethanol: 10 vol% DMSO (d, d-1), DMF (e) and acetonitrile (f).

the absorption of PSS. It indicated that PSS was partially removed from PEDOT: PSS by solvent [26,27]. The increased surface roughness was caused by the removal of PSS and the aggregation of PEDOT [26,27]. According to the seed-mediated growth mechanism of MAPbI3 crystal, the aggregation of PEDOT at the surface was in favor of obtaining MAPbI3 film with large grains [35,36]. Therefore, the aggregation of PEDOT caused by DMSO, ethanol or ethanol/10 vol% DMSO treatments induced large MAPbI3 crystal grain size. DMSO treated PEDOT: PSS demonstrated a rougher surface than the ethanol treated one, and therefore led to larger grain size of MAPbI3 crystal as shown in Fig. 2(b) and (c). However, although the rms roughness values of DMF or acetonitrile treated PEDOT: PSS were much smaller compared with that of DMSO treated PEDOT: PSS film (Fig. S4), the mean crystal size values of MAPbI3 films on them were much larger. Therefore, we believe that there must be other reasons responsible for the morphology change of MAPbI3 film by polar solvent treatment to PEDOT: PSS. To further understand the mechanism of PEDOT: PSS modification, XRD measurements were carried on the as-spin-coated MAPbI3 films. As shown in Fig. 4a, peaks at 6.43°, 7.08° and 9.08° were the characteristic peaks of intermediate phase, MAI-DMSO-PbI2, as proven in the previous reports [37,38]. Although, the locations of peaks were unchanged with or without solvent treatment to PEDOT: PSS, the peak intensity (height) changed significantly. The detailed height ratios (with set the 9.08° peak intensity as 100) of peaks at 6.43°, 7.08° and 9.08°, FWHM and intensity values of the peak at 9.08° were summarized in Table S1. The ethanol or ethanol/ 10 vol% DMSO treatment to PEDOT: PSS led to a

Table 2 The detailed crystal size values of perovskite films. treatment of PEDOT: PSS

untreated DMSO ethanol ethanol/10 vol% DMSO DMF acetonitrile

crystal size minimum (nm)

maximum (nm)

mean (nm)

variance (nm2)

40 70 60 60

350 400 360 360

150 190 180 160

4525 4968 4338 3583

50 58

430 576

210 220

7512 7210

height and phase images and absorption spectra of PEDOT: PSS films with or without solvent treatment. The rms roughness values of pristine and treated PEDOT: PSS films by DMSO, ethanol, ethanol/10 vol% DMSO, DMF and acetonitrile were 1.7 ± 0.1, 2.66 ± 0.01, 1.79 ± 0.09, 2.13 ± 0.04, 1.88 ± 0.03 and 1.69 ± 0.08 nm, respectively. Obviously, PEDOT: PSS films with rougher surfaces were formed after being treated by a solvent, except for acetonitrile. In addition, the bright and dark regions in Fig. 3 corresponded to PEDOTrich and PSS-rich regions respectively [26]. The phase separation between PEDOT and PSS can be observed from the AFM phase images in Fig. 3. The dark area decreased after solvent treatment, indicating the selective removal of PSS by solvent. UV–vis absorption (Fig. 3e) at 200∼300 nm decreased after solvent treatment, which corresponded to 20

Synthetic Metals 243 (2018) 17–24

Q. Niu et al.

Fig. 3. AFM height images and phase images across 2 × 2 μm in 2D views of pristine PEDOT: PSS (a, a-1) and PEDOT: PSS treated by DMSO (b, b-1), ethanol (c, c-1), ethanol/10 vol% DMSO (d, d-1), and absorption spectra of PEDOT: PSS films with or without solvent treatment (e); insets are the corresponding AFM height images in 3D views.

decrease in height ratio values at peaks of 6.43° and 7.08°, suggesting stronger crystal growth orientation along the (022) plane. In contrast, DMF or acetonitrile treatment to PEDOT: PSS caused crystal plane stacking along (002) and (021) to increase, but decrease along (022). Meanwhile, the film deposited on the DMSO, ethanol or ethanol/ 10 vol % DMSO treated PEDOT: PSS displayed stronger intensity of peak at 9.08°, indicating better crystalline of intermediate phase. The FWHM values of the most intense peak of 9.08° were examined, which was 0.248° for the untreated sample. After solvent treatment to PEDOT: PSS by DMSO, ethanol or ethanol/10 vol% DMSO, the FWHM values decreased to 0.199°, 0.17° and 0.19°, respectively. While the values for DMF or acetonitrile treatment increased to 0.30° and 0.264°, respectively. Narrower FWHM reflected the larger grain size of the intermediate phase caused by DMSO, ethanol or ethanol/ 10 vol% DMSO treatments. According to Lewis acid-base theory, the Lewis acids PbI2

and MAI can form a complexes with a Lewis base [39,40]. Ethanol, DMSO, DMF and acetonitrile are all Lewis bases, which can provide lone pair electrons via the oxygen or nitrogen atoms. The sequence of the Lewis bases in order of the coordination ability is DMSO, ethanol, DMF and acetonitrile [40]. Therefore, the relatively strong coordination capability of DMSO and ethanol is favorable for the formation of intermediate phases with larger crystal sizes because of the strong coordination effect at the interface. In contrast, the crystal size of the intermediate phase was much smaller after DMF or acetonitrile treatment because of the weak coordination effect. The stability of the intermediate phase plays an important role in the morphology of MAPbI3 film [41]. It had been well established that MAI and PbI2 will react with each other and form MAPbI3 after the evaporation of DMSO in the intermediate phase [23]. Therefore, the stability of the intermediate phase was evaluated by examining the XRD

Fig. 4. XRD patterns of perovskite films, (a) immediately after spin-coating of perovskite precursor solution without thermal annealing, and (b) after 60 min. storage in air atmosphere at room temperature. 21

Synthetic Metals 243 (2018) 17–24

Q. Niu et al.

patterns of as-spin-coated perovskite films after 60 min. storage at room temperature and pressure, as shown in Fig. 4b. The intense peaks at 13.97° represented the formation of MAPbI3. As for films on pristine, DMF or acetonitrile treated PEDOT: PSS, the intensity of peak at 13.97° was strong. Whilst the films on DMSO, ethanol and ethanol/10 vol% DMSO treated PEDOT: PSS, had a weak intensity peak at 13.97°, indicating the formation of stable intermediate phase. In contrast, the stability of intermediate phase was poor after DMF or acetonitrile treatment, indicating the quick decomposition of the intermediate phase, which will induce fast crystallization of MAPbI3. According to the classic nucleation rate and grain growth theory, large grain growth rate leads to the formation of crystals with larger sizes [40]. Therefore, DMF or acetonitrile treatment to PEDOT: PSS greatly enlarged the crystal size of MAPbI3. On the other hand, although a stable intermediate phase was formed after DMSO, ethanol or ethanol/ 10 vol% DMSO treatments, the grain size of the MAPbI3 crystal was also enlarged. That can be ascribed to the rough surface of PEDOT: PSS caused by the PEDOT-rich aggregation. On the base of the surface roughness of the MAPbI3 film and the stability of the intermediate phase, it can be found that stable precursor film will result in a smooth MAPbI3 film. The precursor film deposited on the DMSO, ethanol or ethanol/ 10 vol% DMSO treated PEDOT: PSS displayed a stable intermediate phase, which led to smooth MAPbI3 films. However, the fast decomposition of the intermediate phase caused by the quick evaporation of DMSO led to the rapid growth of MAPbI3 and induced large, but rough crystals. Uniformly distributed large size grain of MAPbI3 is favorable for efficient solar cells [31]. Hence, even though DMF or acetonitrile treatment resulted in large crystal size, the non-uniform crystal and rough surface reduced the PCE of PSC, as shown in Fig. 2. Since MAPbI3 films were deposited on top of PEDOT: PSS, the wettability of the surface of PEDOT: PSS was also important to the filmforming of MAPbI3. Therefore, the contact angles of the solvents of perovskite precursor solution on PEDOT: PSS with or without solvent treatments were measured, as shown in Fig. 5. The contact angles on pristine PEDOT: PSS and DMF or acetonitrile treated PEDOT: PSS were 26°, 11° and 11°, respectively. However, the angles on DMSO, ethanol or ethanol/ 10 vol% DMSO treated PEDOT: PSS were about 3°, indicating good wettability, which facilitated deposition of MAPbI3 with a smooth surface. The modification of PEDOT: PSS at the surface may influence the charge carrier recombination at the PEDOT: PSS/MAPbI3 interface [42]. Therefore, the steady-state photoluminescence intensity of MAPbI3 films deposited on pristine and solvent treated PEDOT: PSS were examined (Fig. 6). The PL peak at about 778 nm was originated from MAPbI3 [43]. Obviously, the peak intensity decreased when MAPbI3 was deposited on PEDOT: PSS treated by DMSO, ethanol or

Fig. 6. Photoluminescence spectra of MAPbI3 films deposited on pristine and solvent treated PEDOT: PSS.

ethanol: 10 vol% DMSO. In contrast, it increased in case of DMF or acetonitrile treatment. The enhancement of PL peak intensity of MAPbI3/PEDOT: PSS indicates the increase of charge carrier recombination at the interface, which was unfavorable for the collection of charge carriers [42,43]. Therefore, the JSC of PSC based on DMF or acetonitrile treated PEDOT: PSS reduced. To ensure the accuracy of the results, PL spectra in Fig. 6 were retested and the results shown in Fig.S5. It demonstrates the same change in PL intensities with the variation of PEDOT: PSS films shown in Fig. 6. PSS is an insulator. The partial removal of PSS will increase the electrical conductivity of the PEDOT: PSS film. Therefore, the conductivity of PEDOT: PSS was examined by fabricating devices with configuration of ITO/PEDOT: PSS (20 nm)/Ag (150 nm). The currentvoltage curves of the above devices are shown in Fig. S6. The conductivities of PEDOT: PSS films along vertical direction were 0.0019, 0.00258, 0.00253, 0.00256, 0.0022 and 0.00216 mS/cm for the pristine film and films treated by DMSO, ethanol, ethanol/10 vol% DMSO, DMF and acetonitrile, respectively. Low resistance of PEDOT: PSS film led to a decrease in series resistance (RS) (Table 1) of PSC after solvent treatment. Though the decrease of Rs was favorable for the improvement of FF, the increase of rms value influenced the FF of PSC in an unfavorable way. Finally, the FF values of PSCs were almost unchanged after solvent treatment. We also noticed that the VOC values of the PSCs decreased after treating PEDOT: PSS using DMSO, ethanol and ethanol: 10 vol% DMSO. It was probably caused by the reduced work function of PEDOT: PSS. In a previous report [26], it has been shown that after treating PEDOT: PSS using polar solvent, a dipole layer was formed at the interface of PEDOT: PSS/active layer, which led to a work function reduction of

Fig. 5. the contact angles of the solvents of perovskite precursor solution on pristine PEDOT: PSS (a) and PEDOT: PSS treated by DMSO (b), ethanol (c), ethanol/ 10 vol% DMSO (d), DMF (e), acetonitrile (f). 22

Synthetic Metals 243 (2018) 17–24

Q. Niu et al.

PEDOT: PSS. The dipole layer was formed because of the polarity of the solvent itself [24,44,45]. Hence, polar solvent treatment will lead to a lower work function of PEDOT: PSS, and therefore a lower VOC of perovskite solar cells.

[11] S.Z. Yue, S.D. Lu, K.K. Ren, K. Liu, M. Azam, D.W. Cao, Z.J. Wang, Y. Lei, S.C. Qu, Z.G. Wang, Insights into the influence of work functions of cathodes on efficiencies of perovskite solar cells, Small 13 (2017) 1700007. [12] Q. Wang, M.R. Ahmadian-Yazdi, M. Eslamian, Investigation of morphology and physical properties of modified PEDOT: PSS films made via in-situ grafting method, Synth. Met. 209 (2015) 521–527. [13] Q. Chena, F. Zabihia, M. Eslamian, Improved functionality of PEDOT: PSS thin films via graphene doping, fabricated by ultrasonic substrate vibration-assisted spray coating, Synth. Met. 222 (2016) 309–317. [14] Q. Wang, M. Eslamian, Improving uniformity and nanostructure of solution-processed thin films using ultrasonic substrate vibration post treatment (SVPT), Ultrasonics 67 (2016) 55–64. [15] W. Yu, K.X. Wang, B. Guo, X.Q. Qiu, Y. Hao, J.J. Chang, Y. Li, Effect of ultraviolet absorptivity and waterproofness of poly(3,4-ethylenedioxythiophene) with extremely weak acidity, high conductivity on enhanced stability of perovskite solar cells, J. Power Sour. 358 (2017) 29–38. [16] W.H. Sun, Y.L. Li, Y. Xiao, Z.R. Zhao, S.Y. Ye, H.X. Rao, H.K. Ting, Z.Q. Bian, L.X. Xiao, C.H. Huang, Z.J. Chen, An ammonia modified PEDOT: PSS for interfacial engineering in inverted planar perovskite solar cells, Org. Electron. 46 (2017) 22–27. [17] D.Y. Liu, Y. Li, J.Y. Yuan, Q.M. Hong, G.Z. Shi, D.X. Yuan, J. Wei, C.C. Huang, J.X. Tang, M.K. Fung, Improved performance of inverted planar perovskite solar cells with F4-TCNQ doped PEDOT: PSS hole transport layers, J. Mater. Chem. A 5 (2017) 5701–5708. [18] Y.J. Xia, K. Sun, J.J. Chang, J.Y. Ouyang, Effects of organic inorganic hybrid perovskite materials on the electronic properties and morphology of poly(3,4-ethylenedioxythiophene): poly(styrenesulfonate) and the photovoltaic performance of planar perovskite solar cells, J. Mater. Chem. A 3 (2015) 15897–15904. [19] C.T. Zuo, L.M. Ding, Modified PEDOT layer makes a 1.52 v voc for Perovskite/ PCBM solar cells, Adv. Energy Mater. 7 (2017) 1601193. [20] H.M. Liu, X.Y. Li, L.P. Zhang, Q.M. Hong, J.X. Tang, A.P. Zhang, C.Q. Ma, Influence of the surface treatment of PEDOT: PSS layer with high boiling point solvent on the performance of inverted planar perovskite solar cells, Org. Electron. 47 (2017) 220–227. [21] X.Y. Li, L.P. Zhang, F. Tang, Z.M. Bao, J. Lin, Y.Q. Li, L. Chen, C.Q. Ma, The solvent treatment effect of the PEDOT: PSS anode interlayer in inverted planar perovskite solar cells, Rsc Adv. 6 (2016) 24501–24507. [22] Q. Wang, C.C. Chueh, M. Eslamian, A.K.-Y. Jen, Modulation of PEDOT: PSS pH for efficient inverted perovskite solar cells with reduced potential loss and enhanced stability, ACS Appl. Mater. Interfaces 8 (2016) 32068–32076. [23] D. Huang, T. Goh, J. Kong, Y.F. Zheng, S.L. Zhao, Z. Xu, A.D. Taylor, Perovskite solar cells with a DMSO-treated PEDOT: PSS hole transport layer exhibit higher photovoltaic performance and enhanced durability, Nanoscale 9 (2017) 4236–4243. [24] Q. Niu, H. Wu, W. Huang, J. Tong, Y. Ye, Y. Zhang, W. Zeng, R. Xia, Y. Min, Highly promoting the performances of polymer light-emitting diodes via controlling the residue of polar solvent on emissive layer, ACS Appl. Mater. Interfaces 9 (2017) 18399–18404. [25] H. Cha, C.E. Park, S.K. Kwon, T.K. An, Ternary blends to achieve well-developed nanoscale morphology in organic bulk heterojunction solar cells, Org. Electron. 45 (2017) 263–272. [26] Q. Wang, Y.W. Chen, Y.N. Zheng, N. Ai, S.H. Han, W. Xu, Z.X. Jiang, Y.H. Meng, D.G. Hu, J.B. Peng, J. Wang, Y. Cao, Solvent treatment as an efficient anode modification method to improve device performance of polymer light-emitting diodes, Org. Electron. 14 (2013) 548–553. [27] C. Gong, H.B. Yang, Q.L. Song, Z.S. Lu, C.M. Li, Mechanism for dimethylformamidetreatment of poly(3,4-ethylenedioxythiophene): poly(styrene sulfonate) layer to enhance short circuit current of polymer solar cells, Sol. Energy Mater. Sol. Cells 100 (2012) 115–119. [28] Q. Wang, Q.F. Dong, T. Li, A. Gruverman, J.S. Huang, Thin insulating tunneling contacts for efficient and water-resistant perovskite solar cells, Adv. Mater. 28 (2016) 6734. [29] Y. Li, Z. Xu, S.L. Zhao, B. Qiao, D. Huang, L. Zhao, J. Zhao, P. Wang, Y.Q. Zhu, X.G. Li, X.C. Liu, X.R. Xu, Highly efficient p-i-n perovskite solar cells utilizing novel low-temperature solution-processed hole transport materials with linear pi-conjugated structure, Small 12 (2016) 4902–4908. [30] X. Huang, K. Wang, C. Yi, T.Y. Meng, X. Gong, Efficient perovskite hybrid solar cells by highly electrical conductive PEDOT: PSS Hole transport layer, Adv. Energy Mater. 6 (2016) 1501773. [31] L. Xie, H. Hwang, M. Kim, K. Kim, Ternary solvent for CH3NH3PbI3 perovskite films with uniform domain size, Phys. Chem. Chem. Phys. 19 (2017) 1143–1150. [32] M.R. Ahmadian-Yazdi, M. Eslamian, Toward scale-up of perovskite solar cells: annealing-free perovskite layer by low-cost ultrasonic substrate vibration of wet films, Mater. Today Commun. 14 (2018) 151–159. [33] Y.Q. Zhai, Z. Zhang, G.Y. Huo, M.H. Ren, Microwave synthesis, characterization, magnetic and electrical properties of double perovskite Sr2FeMoO6, Rare Metal Mat. Eng. 40 (2011) 906–910. [34] M.A. Mahmud, N.K. Elumalai, M.B. Upama, D. Wang, B. Puthen-Veettil, F. Haque, M. Wright, C. Xu, A. Pivrikas, A. Uddin, Controlled ostwald ripening mediated grain growth for smooth perovskite morphology and enhanced device performance, Sol. Energy Mater. Sol. Cells 167 (2017) 87–101. [35] J.P. Thomas, L.Y. Zhao, D. McGillivray, K.T. Leung, High-efficiency hybrid solar cells by nanostructural modification in PEDOT: PSS with co-solvent addition, J. Mater. Chem. A 2 (2014) 2383–2389. [36] G. Adam, M. Kaltenbrunner, E.D. Glowacki, D.H. Apaydin, M.S. White, H. Heilbrunner, S. Tombe, P. Stadler, B. Ernecker, C.W. Klampfl, N.S. Sariciftci,

4. Conclusions Treatment using strong coordination molecules to PEDOT: PSS enlarged the crystal size of intermediate phase and improved its stability because of the strong coordination effect with MAI and PbI2 at the interface. This resulted in homogeneous MAPbI3 crystals, and thus enhanced JSC and PCE of PSCs. The enlarged crystal size of MAPbI3 was ascribed to the aggregation of PEDOT at the surface, which serves as the seed for nucleation. In contrast, large, but non-uniform MAPbI3 crystals were formed after weak coordination molecule treatment because of the formation of volatile intermediate phase caused by the weak coordination effect to MAI and PbI2, which led to the PCE values decreasing significantly. Acknowledgements We express our gratitude to the National Key Basic Research Program of China (973 Program, 2015CB932203), the National Natural Science Foundation of China (Grants 61376023 and 61504066), A Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD), Natural Science Foundation of Jiangsu Higher Education Institutions of China (15KJB430024), and Natural Science Foundation of Jiangsu Province (BK20150838), and the Natural Science Foundation of Nanjing University of Posts and Telecommunications (NUPTSF Grants NY212013 and NY213044). Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.synthmet.2018.05. 012. References [1] D.Y. Luo, L.C. Zhao, J. Wu, Q. Hu, Y.F. Zhang, Z.J. Xu, Y. Liu, T.H. Liu, K. Chen, W.Q. Yang, W. Zhang, R. Zhu, Q.H. Gong, Dual-source precursor approach for highly efficient inverted planar heterojunction perovskite solar cells, Adv. Mater. 29 (2017) 1604758. [2] W.Q. Wu, D. Chen, R.A. Caruso, Y.B. Cheng, Recent progress in hybrid perovskite solar cells based on n-type materials, J. Mater. Chem. A 5 (2017) 10092–10109. [3] H. Tan, A. Jain, O. Voznyy, X. Lan, G.D.A. Fp, J.Z. Fan, R. Quintero-Bermudez, M. Yuan, B. Zhang, Y. Zhao, F. Fan, P. Li, L.N. Quan, Y. Zhao, Z.H. Lu, Z. Yang, S. Hoogland, E.H. Sargent, Efficient and stable solution-processed planar perovskite solar cells via contact passivation, Science 355 (2017) 722–726. [4] F.L. Cai, L.Y. Yang, Y. Yan, J.H. Zhang, F. Qin, D. Liu, Y.B. Cheng, Y.H. Zhou, T. Wang, Eliminated hysteresis and stabilized power output over 20% in planar heterojunction perovskite solar cells by compositional and surface modifications to the low-temperature-processed TiO2 layer, J. Mater. Chem. A 5 (2017) 9402–9411. [5] K. Masuko, M. Shigematsu, T. Hashiguchi, D. Fujishima, M. Kai, N. Yoshimura, T. Yamaguchi, Y. Ichihashi, T. Mishima, N. Matsubara, T. Yamanishi, T. Takahama, M. Taguchi, E. Maruyama, S. Okamoto, Achievement of more than 25% conversion efficiency with crystalline silicon heterojunction solar cell, IEEE J. Photovolt. 4 (2014) 1433–1435. [6] M. Habibi, F. Zabihi, M.R. Ahmadian-Yazdi, M. Eslamian, Progress in emerging solution-processed thin film solar cells-part II: perovskite solar cells, Renew. Sustain. Energy Rev. 62 (2016) 1012–1031. [7] C. Mu, J.L. Pan, S.Q. Feng, Q. Li, D.S. Xu, Quantitative doping of chlorine in formamidinium lead trihalide (FAPbI(3-x)Cl(x)) for planar heterojunction perovskite solar cells, Adv. Energy Mater. 7 (2017) 1601297. [8] R.A. Kerner, B.P. Rand, Linking chemistry at the TiO2/CH3NH3PbI3 interface to current-voltage hysteresis, J. Phys. Chem. Lett. 8 (2017) 2298–2303. [9] J. Huang, K.X. Wang, J.J. Chang, Y.Y. Jiang, Q.S. Xiao, Y. Li, Improving the efficiency and stability of inverted perovskite solar cells with dopamine-copolymerized PEDOT: PSS as a hole extraction layer, J. Mater. Chem. A 5 (2017) 13817–13822. [10] F. Galatopoulos, A. Savva, I.T. Papadas, S.A. Choulis, The effect of hole transporting layer in charge accumulation properties of p-i-n perovskite solar cells, Appl. Mater. 5 (2017) 076102.

23

Synthetic Metals 243 (2018) 17–24

Q. Niu et al.

[37]

[38]

[39] [40]

[41]

cells with high processibility and stability, Nano Energy 34 (2017) 58–68. [42] N.D. Pham, V.T. Tiong, P. Chen, L.Z. Wang, G.J. Wilson, J. Bell, H.X. Wang, Enhanced perovskite electronic properties via a modified lead(II) chloride lewis acid–base adduct and their effect in high-efficiency perovskite solar cells, J. Mater. Chem. A 5 (2017) 5195–5203. [43] C.X. Wang, H. Hao, S.F. Chen, K. Cao, H.T. Yu, Q. Zhang, G.J. Wan, W.J. Shang, W. Huang, Inverse-architecture perovskite solar cells with 5,6,11,12-tetraphenylnaphthacene as a hole conductor, RSC Adv. 7 (2017) 29944. [44] H.Q. Zhou, Y. Zhang, J. Seifter, S.D. Collins, C. Luo, G.C. Bazan, T.Q. Nguyen, A.J. Heeger, High-efficiency polymer solar cells enhanced by solvent treatment, Adv. Mater. 25 (2013) 1646–1652. [45] Q. Wang, Y. Zhou, H. Zheng, J. Shi, C.Z. Li, C.M.Q. Su, L. Wang, C. Luo, D.G. Hu, J. Pei, J. Wang, J.B. Peng, Y. Cao, Modifying organic/metal interface via solvent treatment to improve electron injection in organic light emitting diodes, Org. Electron. 12 (2011) 1858–1863.

M.C. Scharber, Solution processed perovskite solar cells using highly conductive PEDOT: PSS interfacial layer, Sol. Energy Mater. Sol. Cells 157 (2016) 318–325. Y.G. Rong, Z.J. Tang, Y.F. Zhao, X. Zhong, S. Venkatesan, H. Graham, M. Patton, Y. Jing, A.M. Guloy, Y. Yao, Solvent engineering towards controlled grain growth in perovskite planar heterojunction solar cells, Nanoscale 7 (2015) 10595–10599. N.J. Jeon, J.H. Noh, Y.C. Kim, W.S. Yang, S. Ryu, S. Il Seol, Solvent engineering for high-performance inorganic-organic hybrid perovskite solar cells, Nat. Mater. 13 (2014) 897–903. J.W. Lee, H.S. Kim, N.G. Park, Lewis acid-Base adduct approach for High efficiency perovskite solar cells, Acc. Chem. Res. 49 (2016) 311–319. L. Li, Y.H. Chen, Z.H. Liu, Q. Chen, X.D. Wang, H.P. Zhou, The additive coordination effect on hybrids perovskite crystallization and High-performance solar cell, Adv. Mater. 28 (2016) 9862. Y. Bai, S. Xiao, C. Hu, T. Zhang, X.Y. Meng, Q. Li, Y.L. Yang, K.S. Wong, H.N. Chen, S.H. Yang, A pure and stable intermediate phase is key to growing aligned and vertically monolithic perovskite crystals for efficient PIN planar perovskite solar

24