Halogen versus Pseudo-Halogen Induced Perovskite for Planar

---

Subscriber access provided by FLORIDA STATE UNIV

Article

Halogen Versus Pseudo-Halogen Induced Perovskite for Planar Heterojunction Solar Cells : Some New Physical Insights Jing Chen, Jia Xu, Shaopeng Zhang, Shijie Zhou, Kaixuan Zhou, Bing Zhang, Xin Xia, Yang Liu, Songyuan Dai, and Jianxi Yao J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.7b10018 • Publication Date (Web): 04 Dec 2017 Downloaded from http://pubs.acs.org on December 4, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry C is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Halogen

Versus

Pseudo-Halogen

Induced

Perovskite for Planar Heterojunction Solar Cells : Some New Physical Insights Jing Chena,b, Jia Xua,b, Shaopeng Zhangc, Shijie Zhoua, Kaixuan Zhoub, Bing Zhangb, Xin Xiaa, Yang Liu c, Songyuan Daia,b, Jianxi Yaoa,b* a

Beijing Key Laboratory of Energy Safety and Clean Utilization, North China Electric

Power University, Beijing 102206, China. b

Beijing Key Laboratory of Novel Film Solar Cell, North China Electric Power University,

Beijing 102206, China. c

School of Nuclear Science and Engineering, North China Electric Power University, Beijing

102206, China.

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT Introducing halogen and pseudo-halogen has been one of the effective methods to improve the quality of perovskite film. Planar heterojunction perovskite solar cells based on three lead sources of PbI2, PbBr2 and Pb(SCN)2 were fabricated through low pressure vapor assisted solution process. The effect of the lead source on the evolution of the crystal structure, film growth kinetics, electron lifetime, defect density and photoelectric performance were systematic studied. Although the ultima products in all the three films were CH3NH3PbI3, the crystal growth kinetics was absolutely different. These differences indeed strongly affected the films morphologies, defect density, electronic lifetime and photovoltaic performance of the devices. All the results indicated that the quality of perovskite film based on Pb(SCN)2 was inferior and the corresponding devices showed obvious deterioration in photovoltaic performance compared to that of PbI2 and PbBr2. Some physical mechanisms have been explained by studying the photoelectric feature of the devices.

ACS Paragon Plus Environment

Page 2 of 44

Page 3 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

INTRODUCTION Recently, organic-inorganic halide perovskites have attracted much attention as novel photovoltaic materials. Perovskite has ABX3 structure where A is organic cation such as CH3NH3+, HC(NH2)2+, B is divalent metal cation (e.g., Pb2+, Sn2+) and X is halide anion (e.g., Cl−, Br−, I−).1-5 To date, the certificated power conversion efficiency (PCE) for the PSC has reached 22.7%.6 Among the perovskite systems, methylammonium lead triiodide (CH3NH3PbI3) is one of the most widely studied perovskite materials, and the PCE has reached up to 19.3%.7 In order to achieve high efficient perovskite solar cells (PSC), the high quality of perovskite film is regarded as one of the most important factors such as perfect morphologies (large grain size and low surface roughness etc), appropriate thickness and excellent crystallinity etc. It has been widely confirmed that the quality of perovskite films could significantly influence the photovoltaic performance of PSCs. 7-11 Currently, introduction of other halogen elements such as Cl or Br into CH3NH3PbI3 to form multi-halide perovskites i.e. CH3NH3PbX3 (X=Cl, Br, I or their mixtures) has become one of the common strategies to improve the quality of perovskite films.12-23 The influence brought by halogen has been successfully adopted in the fabrication of high efficiency PSCs. For both mixed (I, Br) or mixed (I, Cl) lead halide materials, the Cl or Br ion modifies the crystallization pathway by forming the intermediate perovskite phase and changes crystallization kinetics dramatically.13-16 It has been widely reported that the incorporation of Cl or Br ion in perovskite formation could slow down the nucleation process and the crystal growth rate, leading to the preferential growth direction along [110] direction and improve the crystallinity significantly.12-14 So the perovskite films induced formation by Cl or Br possesses larger crystallite sizes and lower surface roughness. As we known, the grain boundaries would act as the recombination centers for photo-generated carriers and deteriorate the performance. The defect density of perovskite materials is significantly

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

reduced by Cl or Br which can suppress non-radiative recombination channels.20, 23 Therefore, the introduction of halogen atom into the perovskite system is favorable for the prolongation of charge carrier lifetime and enhancement of carrier diffusion length. Generally, the as-fabricated devices have superior photovoltaic properties with a lower hysteresis effects comparing with the devices prepared by pure CH3NH3PbI3.20-22 Apart from halogen, there are many studies about pseudo-halogen have been used for the improvement of perovskite film quality.24-32 Perovskite films in most of these investigations were prepared by solution process. Results have demonstrated that adding a small amount of Pb(SCN)2 in perovskite precursor solution can significantly improve the crystalline quality of perovskite films accompanying by enlarged grain size and effectively enhance the moisture stability of PSCs.24-26 Among them, Yan24 group prepared CH3NH3PbIx(SCN)3-x PSCs through one-step solution process. The as-prepared perovskite films possessed enlarged grain size and improved crystalline quality. And the optimized efficiency was 19.45% and the device was fabricated by the precursor solution which was added 5 wt% Pb(SCN)2. Vapor deposition as another simple fabrication process has been adopted to prepare perovskite film widely. However, the study on introduction of SCN based on vapor deposition is in its infancy. As far as we know, there was only one report about the fabrication of perovskite films induced by SCN through vapor process by Chen et al36 group. They prepared mesoporous PSCs via low pressure vapor assisted process (LP-VASP) adopting Pb(SCN)2 as lead source. In their studies, the grain size of perovskite films was enlarged and the corresponding devices exhibited a relative higher PCE of 12.72% compared to that of CH3NH3PbI3 prepared by PbI2 which was just 11.32%. There was just a small amount of residual S in the final mesoporous based perovskite film. As known, the growth kinetics of perovskite on planar or mesoporous substrate is different.37-38 Whether the results of Pb(SCN)2 on planar substrate would be consistent with the Chen et al results deserves

ACS Paragon Plus Environment

Page 4 of 44

Page 5 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

exploring. Moreover, it was known that the vapor deposition can reduce the over-rapid intercalating reaction rate compared to solution process,33-35, 39 so the vapor deposition start a window for understanding the intermediate reaction process deeply. In this study, three kinds of planar heterojunction perovskite solar cells were fabricated through LP-VASP using PbI2, PbBr2, and Pb(SCN)2 as lead sources respectively. The effect of the lead source on the evolution of the crystal structure, film growth kinetics, electron lifetime, defect density and photoelectric performance were systematic studied. The crystal structure in all ultima films was all CH3NH3PbI3, even if three lead sources were used. It was interesting that the crystal growth processes were absolutely discrepant in the three films. There was no intermediate perovskite phase during the whole vapor reaction process in the Pb(SCN)2 case, which was different from the PbBr2 case. These discrepancies caused the increasing defect densities and shorter electronic lifetime, and then decayed the photovoltaic performance of Pb(SCN)2 based devices. The highest PCE of devices based on PbI2 and PbBr2 reached 16.15% and 16.28% respectively. But for the device based on Pb(SCN)2, the PCE was just 10.24%. The study can help the researches to get the insight into the effect of Br and SCN on the photovoltaic properties of planar heterojunction PSCs. EXPERIMENTAL SECTION Materials CH3NH3I,

2,2',7,7'-tetrakis-(N,N-di-p-methoxyphenylamine)-9,9'-spirobifluorene

(Spiro-OMeTAD), FK209-cobalt(III)-TFSI, bis (trifluoromethane) sulfonimide lithium salt (Li-TFSI) were purchased from Borun Chemicals (Ningbo, China). PbI2 (99%), Titanium (IV) isopropoxide (98+%), chlorobenzene and 1,2-dichlorobenzene (98%) were both purchased from Acros. PbBr2, C60 and N,N-dimethylformamide (DMF) were both purchased from Alfa Aesar. Anhydrous isopropanol was purchased from J&K Scientific Co., Ltd. Pb(SCN)2 and 4-tert-butylpyridine (tBP) were purchased from Sigma-Aldrich. All chemicals were used

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

directly used without further purification. Glass substrates with a transparent fluorine-doped tin oxide (FTO, thickness 2.2 mm, sheet resistance 15 Ω/square) layer were used for the PSCs. Device Fabrication The devices with an FTO/compact-TiO2/C60/CH3NH3PbI3/spiro-OMeTAD/Au structure were prepared as follows. First, the FTO glass substrates were etched using Zn powder and hydrochloric acid. Then, the FTO glass substrates were cleaned sequentially using alkaline liquor, liquid detergent, deionized water, and ethanol. After dying under clean dry air, the FTO was sintered at 500℃ for 30 min in air to remove residual organic matter. Subsequently, a TiO2 blocking layer was deposited on the cleaned FTO by spray pyrolysis, using dry air as carrier gas, at 460 °C from a precursor solution of 0.6 ml of titanium diisopropoxide and 0.4 ml of bis(acetylacetonate) in 7 ml of anhydrous isopropanol. Then, the C60 layer was spin-coated on the c-TiO2 layer at 1500rpm for 30 s and annealed at 60℃ for 2 min. After cooling to room temperature, the PbI2/PbBr2/Pb(SCN)2 solution in DMF (kept at 70℃) was spin-coated onto the FTO/c-TiO2/C60 layer at 3000 rpm for 30 s and then heated on a hot plate at 70 °C for 30 min in a nitrogen-filled glovebox. CH3NH3I powder was uniformly spread around the PbI2/PbBr2/Pb(SCN)2 coated substrates in a petri dish covered with a lid. And the perovskite films were marked as P-I, P-Br and P-SCN in later discussions. The petridish was placed in a vacuum oven (10 kPa) set at 150 °C for 30 min. As-prepared perovskite films were first washed with isopropanol to remove the MAI residue and subsequently heated at 150 °C for 5 min in anitrogen-filled glovebox. Lastly, the Spiro-OMeTAD, which served as the hole transport material layer (HTL), was spin-coated on the perovskite layers at 4000rpm for 30 s. Then Au was thermally evaporated on top to form the back electrode (60 nm) at an atmospheric pressure of 4×10−4 Pa. Characterization

ACS Paragon Plus Environment

Page 6 of 44

Page 7 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

The current density–voltage (J–V) characteristic was measured with a Keithley 2400 source-meter together with a sunlight simulator (XES-300T1, SAN-EI Electric, AM 1.5G 100 mW cm-2), which was calibrated using a standard silicon reference cell. The solar cells were masked with a black aperture to define an active area of 0.09 cm2. Curves under both forward and reverse voltage scanning directions were recorded. External quantum efficiency (EQE) was measured in air using a QE-R measurement system (Enli Technology). Scanning electron microscopy (SEM), images were taken with a SU8010SEM (Hitachi). X-ray diffraction (XRD), UV-vis and was measured with a Bruker X-ray diffractometer with a Cu-Kα radiation source. The diffraction angle was scanned from 10° to 80° at a scanning speed of 3.5° per min. The absorption spectra were measured with a UV-2450 spectrophotometer (Shimadzu) from 300 nm to 900 nm. The X-ray photoemission spectra (XPS) was measured with ESCALAB 250Xi, Thermo Fisher Scientific. Time-resolved photoluminescence (TR-PL) spectra were collected using a transient state spectrophotometer (F900, Edinburgh Instruments). Samples were excited with a 660 nm pulsed diode laser with a repetition rate of 2.5 MHz and an excitation intensity of ~14 nJ/cm2. Impedance spectroscopy (IS) measurements were performed both in dark and 1 sun illumination (AM 1.5) conditions, using a 20 mV perturbation with the applied potential. A Zahner electrochemical workstation was used as a frequency response analyzer, and impedance measurements were performed in the frequency range of 10 mHz to 1 MHz. Impedance data were analyzed using Zview equivalent circuit modeling software. Voc decay of the cells was measured using a step function illumination of the cells (on the period of 20 s). The voltage decay after switching the light off was measured using Zahner electrochemical workstation. RESULTS AND DISCUSSION SEM images of perovskite films using different lead sources (PbI2, PbBr2 and Pb(SCN)2) prepared by LP-VASP have been shown in Figure 1. It could be obviously seen that all the

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 44

perovskite films were compact and uniform. The mean grain size of P-I (Figure 1a), P-Br (Figure 1b) and P-SCN (Figure 1c) were respectively about 500 nm, 1µm and 1µm measured from the insert figures. From these results, it could be concluded that the incorporation of both halogen Br and pseudo-halogen SCN was really favorable for the formation of large grain size particles of the perovskite films. Thirty devices were fabricated using P-I, P-Br and P-SCN, respectively. The best efficiency among them under illumination (AM 1.5 100W/cm2) were shown in Figure 2a. The device based on P-I obtained the highest PCE of 16.15%, with a Voc of 0.96 V, a Jsc of 22.73 mA/cm2, and an FF 74.00% and the best PCE of P-Br reached 16.28%, with a Voc of 1.02 V, a Jsc of 22.04 mA/cm2, and an FF 72.69%. However, for PSC based on P-SCN, the PCE was just 10.24%, with a Voc of 0.91 V, a Jsc of 20.32 mA/cm2, and an FF 55.63%. Corresponding photovoltaic performances were listed in Table 1. From the results it can be seen that, the device based on P-SCN mainly showed lowered parameters in the FF and Voc compared to that of P-I and P-Br. The FF and Voc would be affected by the carrier’s lifetime and defect density which has been analyzed in the discussion of TR-PL results in detail. Besides, hysteresis is a typical feature of PSCs, which have been reported widely. So far, hysteresis index (HI) has been recognized as a reasonable parameter to quantify the hysteresis effect.40-42 HI was defined by Equation 1:
 = 



     



  



(1)



    and     stand for the area under J-V curves measured from forward and reverse scans, respectively. OC and SC are abbreviations of open circuit and short circuit, respectively. HI reflected the integrated power output with different scan direction. All the changes in hysteresis reflected from all bias voltage regions have been taken in consideration in the Equation 1. The calculated HIs results were 0.077 for P-I device, 0.060 for P-Br device and 0.085 for P-SCN device, respectively. The results were consistent

ACS Paragon Plus Environment

Page 9 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

with the trend of PCE. The small changed HI of P-Br device was attributed to the improvement of photovoltaic performance. For P-SCN devices, the HI was a bigger one than that of P-I. (Thirty minutes of reaction time between CH3NH3I vapor and Pb(SCN)2 was the optimized condition, as shown in Figure S1). Besides, the EQE spectra of the best-performing PSCs based on different lead sources are shown in Figure 2b. The integrated photocurrent obtained from the IPCE curves of P-I, P-Br and P-SCN was 20.60 mA/cm2, 20.26 mA/cm2 and 18.61 mA/cm2 respectively. The histograms of PCEs with thirty devices for each case were shown in Figure 2c. The average PCE of PSCs based on P-I, P-Br and P-SCN were 15.71%, 15.60% and 7.44% respectively. It was clearly seen that PSCs based on P-I and P-Br has superior photovoltaic performance than that of PSCs based on P-SCN. The existence of Br and SCN indeed increased the grain size of perovskite film. However, it was so unexpected that the large grain size brought out by the introduction of pseudo-halogen SCN really did not cause more excellent photovoltaic performances compared to the devices based on P-I. It have been widely reported that the large grain size usually could decrease the defect density and improve the transportation and collections of carriers.43-45 However, from Figure 1 and Figure 2, distinct-different results were observed. Using Pb(SCN)2 as lead source, even ~1µm grain size was obtained, but the solar cell still showed poor performance. That is to say, some ambiguous factors might greatly affect the nature of devices based on P-SCN. In order to reveal the intrinsic effect mechanism of the SCN on the deteriorative performance in our system, different ways were employed to characterize the feature of the solar cell fabricated by the three perovskite films. Firstly, the final crystal structure of the perovskite films prepared by different lead sources was investigated. XRD results were shown in Figure3a. From the results, it can be seen that the peaks at 14.11°, 19.87°, 23.50°, 24.50°, 28.44°, 31.90°, 35.00°, 40.57°, 42.63°, and 43.22° marked by black solid diamonds are indexed to the

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(110), (112), (211), (202), (220), (310), (312), (224), (314) and (330) crystallographic planes of pure CH3NH3PbI3, respectively. And the insert figure in Figure 3a is an enlarged view of XRD patterns for all the samples with the 14.11° indicating that the value of 2θ remained coincident for the film of P-I, P-Br and P-SCN, which means that all final films were CH3NH3PbI3. Figure 3b was the UV-vis absorption spectrum of P-I, P-Br and P-SCN films. The absorption onset of P-I, P-Br and P-SCN films did not shift obviously which indicated that the bandgap of perovskite films prepared by different lead sources was almost the same. The bandgap was obtained on the basis of the Tauc plot as the intercept value of the plot of (αhν)2 against light energy (hν). By extrapolating the linear region of the (αhν)2 to the energy-axis (hν) intercept as shown in insert figure of Figure 3b. The value of Eg is established which was about 1.59eV. Besides, sulfur element (S) was not found in the P-SCN films by XPS measurement (Figure S2) reacted with CH3NH3I vapor for 30 min. Therefore, these results can be concluded that the final perovskite films prepared by halogen Br and pseudo-halogen SCN were pure CH3NH3PbI3. The growth process of the perovskite films based on different lead sources was monitored by XRD. The time evolution (0-30 min) of XRD patterns from 10° to 17° region for the PbBr2, and Pb(SCN)2 films exposed to CH3NH3I vapor are showed in Figure 4a and 4b, respectively. The corresponding XRD patterns of Pb(SCN)2 films over the entire range of angles are shown in Figure S3a. After PbBr2 film was exposed in CH3NH3I vapor for three minutes, a diffraction peak at 14.9° appeared which is assigned to the characteristics peak of CH3NH3PbIxBry intermediates phase.46 With the reaction time prolonging, a diffraction peak at 14.11° emerged within ten minutes which is attributed to the characteristics peak of CH3NH3PbI3. After twenty minutes of reaction time the diffraction peak at 14.9° was completely disappeared, the diffraction peak at 14.11° became stronger until the reaction ended. Besides, a diffraction peak at 12.7° appeared belonging to the characteristics peak of

ACS Paragon Plus Environment

Page 10 of 44

Page 11 of 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

PbI2 (001) takes place within five minutes. With the progress of reaction process the peak at 12.7° disappeared until the end of the reaction. Therefore the growth process of P-Br was that, in the whole reaction progress, an intermediates perovskite phase CH3NH3PbIxBry was firstly formed. Then with the increasing of reaction time, the PbI2 was formed and at the same time CH3NH3PbIxBry transformed to CH3NH3PbI3 until the termination of reaction. The intermediate perovskite phase is curial for the photovoltaic performance. It has been reported that the main impact of intermediate perovskite phase on final perovskite morphologies and crystallinities is the way that it alters in the nucleation dynamics of the film.47-49 But for the P-SCN XRD results, it indicated absolute different crystallinity evolution process. During the progress of reaction process, the diffraction peak at 16.4° belonging to the characteristic peak of Pb(SCN)2 disappeared gradually and a diffraction peak at 12.7° assigning to PbI2 emerged within three minutes. After forming a good crystallized PbI2 film a diffraction peak at 14.11° start to appear which is attributed to the characteristics peak of CH3NH3PbI3. Subsequently, the diffraction peak at 12.7° disappeared gradually and the diffraction peak at 14.11° became stronger until the end of reaction process. Therefore the growth process of P-SCN can be summarized as that, Pb(SCN)2 was firstly converted into PbI2. After fully conversion, the PbI2 films reacted with CH3NH3I vapor gradually. That is to say, in the formation process of P-SCN, there was no existence of intermediate perovskite phase. Reaction mechanism of formation of perovskite films induced by halogen Br and pseudo-halogen SCN was different. Corresponding UV-vis absorption with the evolution of P-SCN was shown in Figure S3b.The absorption onset of PbI2 appeared when the Pb(SCN)2 film exposed in CH3NH3I vapor three minutes as indicated by purple line and the absorption onset of CH3NH3PbI3 was emerged when the exposure time was fifteen minutes which was pointed out by green line. It can be concluded that the reaction process of P-SCN was that after Pb(SCN)2 converted into PbI2 completely, the formation of perovskite would start. There was no existence of

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 44

intermediates perovskite phase during the formation which was consistent with the results of Figure 4b. Surface morphology evolution of P-SCN formation process was shown in Figure S4. With the progress of reaction process the morphology changed continuously which was consistent with the results of XRD analysis. In summary, the formation process of CH3NH3PbI3 induced by different lead sources in LP-VASP can be described by the reaction (1)-(3) in Scheme 1. To further clarify the photovoltaic characteristics, equivalent circuit analysis was carried out to extract more information about the cells from their J–V curves at illumination (AM 1.5) and in dark conditions as shown in Figure 5a. Solar cells are roughly equivalent to a parallel circuit consisting of a current source and a diode. The output current density (J) of the cells can be described as50:  =  

!"#$  %&'

− 1* +

$ $,

− -.

(2)

where J0 is the reverse saturated current density, Jsc is the measured short-circuit current density, Rs is the series resistance, A is the ideality factor of a heterojunction, K is the Boltzmann constant, T is the absolute temperature, q is the elementary charge, and V is the direct-circuit bias voltage applied to the cell. When Rs