Efficient and Reproducible CH3NH3PbI3 Perovskite Layer Prepared

The surface of annealed film-DMI (film-DMI-A) was smooth, with an average crystal size of 1 μm. Photoluminescence and transient photovoltage measurem...
0 downloads 0 Views 5MB Size
Download PDF
Research Article Cite This: ACS Appl. Mater. Interfaces 2018, 10, 9390−9397

www.acsami.org

Efficient and Reproducible CH3NH3PbI3 Perovskite Layer Prepared Using a Binary Solvent Containing a Cyclic Urea Additive Lin Xie,† An-Na Cho,‡ Nam-Gyu Park,*,‡ and Kyungkon Kim*,† †

Department of Chemistry and Nano Science, Ewha Womans University, Seoul 120-750, South Korea School of Chemical Engineering and Department of Energy Science, Sungkyunkwan University, Suwon 440-746, Korea

‡

S Supporting Information *

ABSTRACT: An efficient CH3NH3PbI3 perovskite solar cell whose performance is reproducible and shows reduced dependence on the processing conditions is fabricated using the cyclic urea compound 1,3-dimethyl-2-imidazolidinone (DMI) as an additive to the precursor solution of CH3NH3PbI3. X-ray diffraction analysis reveals that DMI weakly coordinates with PbI2 and forms a CH3NH3PbI3 film (film-DMI) with no intermediate phase. The surface of annealed film-DMI (film-DMI-A) was smooth, with an average crystal size of 1 μm. Photoluminescence and transient photovoltage measurements show that film-DMI-A exhibits a longer carrier lifetime than a CH3NH3PbI3 film prepared using the strongly coordinating additive dimethyl sulfoxide (DMSO) (film-DMSO-A) because of the reduced number of defect sites in film-DMI-A. A solar cell based on film-DMI-A exhibits a higher power conversion efficiency (17.6%) than that of a cell based on film-DMSO-A (15.8%). Furthermore, the performance of the film-DMI-A solar cell is less sensitive to the ratio between PbI2 and DMI, and film-DMI can be fabricated under a high relative humidity of 55%. KEYWORDS: perovskite, photovoltaics, urea additive, intermediate phase, solar cells

■

INTRODUCTION An organic−inorganic hybrid perovskite material based on methylammonium lead iodide (MAPbI3) was first utilized as a photovoltaic sensitizer for a liquid-electrolyte-based sensitized solar cell in 2009 by Miyasaka et al.1 and delivered a power conversion efficiency (PCE) of 3.8%. Park et al. introduced the perovskite material in a thin-film solar cell, achieving a milestone for the development of a promising next-generation solar cell.2 Over the past 8 years, enormous efforts have been devoted to improving the performance of MAPbI3-based PSCs.3−7 Until now, the PCEs of the PSCs fabricated using a one-step method have increased by over 20%.8,9 The addition of dimethyl sulfoxide (DMSO) into a precursor solution of MAPbI3 used to fabricate PSCs via a one-step method has greatly improved the quality of the MAPbI3 films.10 It was reported that the DMSO formed the intermediate methylammonium iodide (MAI)·PbI2·DMSO speciesthis retards the reaction between PbI2 and MAI because of its strong coordination capability and is a critical consideration factor in the preparation of high-quality MAPbI3 films.9,10 © 2018 American Chemical Society

However, the molar ratio PbI2/DMSO was found to be controlled in quite a narrow stoichiometric ratio around 1:1.11−13 In addition, we found that the intermediate of PbI2· DMSO is very sensitive to humidity because DMSO interacts strongly and quickly with H2O via hydrogen bonding. Furthermore, the as-prepared intermediate phase should be subjected to thermal annealing immediately because of the continuous vaporization of DMSO. Thus, it is required to control the intermediate phase of PbI2·DMSO in a reproducible manner. Recently, Li et al. claimed that a weak coordination additive (acetonitrile) showed a larger crystal size and better photovoltaic performance due to the different crystallization kinetics of MAPbI3.14 Furthermore, Snaith’s group reported that the PSCs processed with a weakly coordinating additive have great potential in the fabrication of large-scale solar cells.15 Because the study of the intermediate phase is still at an early Received: December 9, 2017 Accepted: January 30, 2018 Published: January 30, 2018 9390

DOI: 10.1021/acsami.7b18761 ACS Appl. Mater. Interfaces 2018, 10, 9390−9397

Research Article

ACS Applied Materials & Interfaces

Figure 1. XRD patterns of the (a) as-prepared films prepared with DMI and DMSO and (b) thermally annealed perovskite films prepared with DMI and DMSO. 20 s to form mesoporous TiO2 (mp-TiO2), followed by an annealing treatment at 550 °C for 1 h. A 1.9 M MAPbI3 fresh precursor solution was prepared [MAI, PbI2, and DMI were dissolved in dimethylformamide (DMF) at a molar ratio of 1:1:0.5]. The prepared perovskite precursor solution was deposited on the mp-TiO2 at 6000 rpm for 25 s, and then 0.7 mL of diethyl ether was dripped onto the spinning substrate at 10 s after starting the spin-coating process. The asprepared film was preheated at 65 °C on a hot plate for 1 min and 100 °C for 1 h to form a dense perovskite photoactive layer. After that, the hole transport solution consisted of 36 mg of 2,29,7,79-tetrakis(N,Ndi-p-methoxyphenylamine)-9,9-spirobifluorene (spiro-MeOTAD) deposited by spin-coating, 28.8 μL of 4-tert-butylpyridine, and 17.5 μL of 520 mg/mL lithium bis(trifluoromethylsulfonyl)imide acetonitrile solution dissolved in 1 mL of chlorobenzene. Finally, Ag with a thickness of 100 nm was evaporated on top of the prepared film. For the control device, the precursor was prepared with PbI2/MAI/DMSO with an optimum molar ratio of 1:1:1, which was found to be the best ratio for the device. The control devices were prepared in the same manner as for perovskite solar cells (PSCs) processed with DMI. For the humidity tolerance test, the device fabrications were carried out under a fume hood at an ∼55% humidity. The current density−voltage (J−V) curves of the devices were measured using a Keithley 2400 source measure unit. An AM 1.5 G simulator was used (McScience K201 LAB50, Oriel) to simulate the solar spectrum. UV−visible absorption spectra were measured using a UV-2450 (SHIMADZU, Japan). Field emission scanning electron microscopy (SEM) images were obtained using a JSM-6700F (JEOL, Japan). The photoluminescence (PL) spectra of the films were obtained using an LS 55 (PerkinElmer, USA). Atomic force microscopy (AFM) images were obtained using an XE-70 (Park Systems Corp., Korea). The MAPbI3 film properties were characterized by powder X-ray diffraction (XRD) (Rigaku D/Max 2200), using Ni-filtered Cu Kα1 radiation (λ = 1.54184 Å). The external quantum efficiency (EQE) was measured with monochromatic light generated from a 300 W xenon lamp in the range 300−900 nm using a K3100 EQX (McScience, Korea). The steady-state transient photovoltage (TPV) measurements were conducted under continuous illumination from an intensity-adjustable white LED. The resulting voltage transient was acquired by a TDS3054B Tektronix digital oscilloscope with a 1 M input impedance. TPV results were fitted to a monoexponential decay function to find the carrier recombination lifetime.

stage, further studies are required in terms of coordination abilities of the additives. The nucleation and crystallization process strongly influences the morphology, crystallinity, and grain size of the perovskite film. The Lewis acid−base adduct approach is one of the effective methods to control the process.15−18 In the Lewis acid−base adduct approach, PbI2 and polar aprotic solvent, such as DMSO, act as a Lewis acid and a Lewis base, respectively. It was found that the existence of Lewis bases in the annealing process could significantly promote perovskite grain growth.18 Therefore, it is more desirable to use a Lewis base having a high boiling point rather than using volatile DMSO. It was also reported that utilization of a weak coordinative Lewis base effectively controls the nucleation process of a perovskite film.18−23 The urea compound is one of the promising candidates for the Lewis base because it is known to have a high boiling point and form weak Lewis acid−base adducts with PbI2. Recently, utilization of urea and thiourea as a Lewis acid−base adduct has been reported.18−20 However, those urea compounds are solid at room temperature. The residual urea compound after annealing process has negative effects on the performance of PSCs. In this work, we introduce a cyclic urea compound 1,3dimethyl-2-imidazolidinone (DMI), which is liquid at room temperature with a high boiling point and would form a weak Lewis acid−base adduct. We expect that the utilization of DMI as a Lewis base additive could effectively control the growth of perovskite crystal and enhance the performance of PSCs.

■

EXPERIMENTAL SECTION

Device Fabrication. Methylammonium iodide (MAI, CH3NH3I) was synthesized by reacting 27.86 mL of methylamine (40% in methanol, TCI) and 30 mL of hydroiodic acid (57 wt % in water, Aldrich) in a 250 mL round-bottom flask at 0 °C for 2 h with stirring. The precipitate was generated by evaporation at 50 °C for 40 min. The product CH3NH3I was recrystallized three times from a hot saturated ethanol solution and washed two times with diethyl ether. Finally, the resultant white powder was dried in a vacuum oven for 12 h at 65 °C. Lead(II) iodide (PbI2) (99.99%) was purchased from Sigma-Aldrich. Fluorine-doped tin oxide (FTO) glasses were cleaned with detergent, ethanol, and acetone with sonication for 20 min. After cleaning, the FTOs were treated with UV/ozone for 20 min. Then, a blocking TiO2 (bl-TiO2) layer was spin-coated on top of the FTO using a 0.1 M titanium diisopropoxide bis(acetylacetonate) solution (75 wt % in isopropanol, Sigma) in 1-butanol, followed by an annealing treatment at 125 °C for 5 min. A nanocrystalline TiO2 paste (40 nm) diluted in 1-butanol at a concentration of 100 mg/mL was then deposited on the bl-TiO2 layer and spin-coated at 2000 rpm for

■

RESULTS AND DISCUSSION The impact of additives on the intermediate phase composition and microstructure of the films was investigated by XRD. Figure 1a shows the XRD patterns of the as-prepared films processed with DMSO or DMI additives and thermal annealing treatment, and Figure 1b shows those of perovskite films after annealing at 100 °C. As shown in Figure 1a, the film processed 9391

DOI: 10.1021/acsami.7b18761 ACS Appl. Mater. Interfaces 2018, 10, 9390−9397

Research Article

ACS Applied Materials & Interfaces with DMSO (film-DMSO) exhibited diffraction peaks at 7°, 7.9°, and 9.6°, which are corresponding to the (002), (021), and (022) planes and originate from the (MA)2Pb3I8·2DMSO intermediate phase, respectively.24−26 The XRD peaks (Figure 1b, film-DMSO-A) at 14.5°, 28.8° and 32.3°, corresponding to the (110), (220), and (310) planes of MAPbI3, respectively, appeared after thermal annealing, implying that the I-deficient structure in the (MA)2Pb3I8·2DMSO intermediate had been replaced by MAI to form MAPbI3 crystals.18,21 The film processed using the DMI additive (film-DMI) did not show diffraction peaks corresponding to the intermediate but showed identical XRD peaks of a pure tetragonal MAPbI3, which indicates that no intermediate phase is formed in the case of the DMI additive. This result implies that the perovskite crystals were formed immediately after spin-coating without any further thermal annealing process. The diffraction peak intensity of MAPbI3 was further enhanced by the thermal annealing treatment of the filmDMI. In addition, the XRD peak intensity was higher than that of the annealed film-DMSO (described as film-DMSO-A). The peak ratios at the (110), (220), and (310) planes of MAPbI3 for annealed film-DMI (described as film-DMI-A) and filmDMSO-A were 1:0.53:0.22 and 1:0:45:0.43, respectively, which suggests that the orientation of the MAPbI3 crystal along the (110) and (220) facets was more prominent in filmDMI-A than film-DMSO-A. Mosconi et al. reported that binding between the adjacent MAPbI3 between undercoordinated Ti(IV) atoms of the mesoporous TiO2 layer would lead the MAPbI3 to grow preferentially in the (110) direction.23,27 On the basis of the XRD results, the final orientation of the MAPbI3 crystal was assumed to be strongly dependent on the crystal orientation of the as-prepared films before annealing. In other words, the fact that the film-DMI and its annealed filmDMI-A have the same MAPbI3 crystal orientation means that the DMI additive is useful for growing the MAPbI3 crystals in a desired direction. We checked the solubility of PbI2 in pure DMI and DMSO solvents to study the nonintermediate phase caused by DMI. The results are shown in Figure S1 in the Supporting Information. PbI2 showed a very poor solubility (28 mg/mL) and dissolved very slowly in pure DMI solvent to give a yellowcolored solution. However, PbI2 dissolved in DMSO very quickly and exhibited a high solubility (922 mg/mL) to form a transparent solution. A solubility test indicated that DMI could form weak coordination with PbI2 and be completely removed during the spin-coating process, resulting in direct crystallization of MAPbI3 in the as-prepared film-DMI. In addition, dielectric constants for DMSO and DMI are 46.7 and 37.5, respectively. The polarity of DMSO is larger than that of DMI. As a result, the lone pair of SO is more activated than that of the CO bonding, which means that the lone pairs of DMSO are more strongly coordinated with Pb(II) than DMI. The addition of DMI to the DMF solvent reduces the vapor pressure of the solution and slows down the growth rate of MAPbI3 without forming an intermediate with precursor materials. The growth mechanism is different from that of the film-DMSO-A because it first becomes the (MA)2Pb3I8· 2DMSO intermediate phase before it is converted into MAPbI3. The surface morphology and film quality were investigated by SEM (Figure 2). Figure 2a,b shows the surfaces of the asprepared film-DMSO and film-DMI samples, respectively. The average grain sizes of the film-DMI and the film-DMSO were

Figure 2. Top-view SEM images of the as-prepared and annealed perovskite films deposited on top of FTO/mp-TiO2 substrates: (a) film-DMSO, (b) film-DMI, (c) film-DMSO-A, and (d) film-DMI-A.

128.2 ± 42.9 nm and 244.3 ± 109.8 nm, respectively. Although the size of the film-DMSO was larger than that of the film-DMI, film-DMI showed a narrow size distribution. The values of film roughness determined by AFM were 5.1 and 5.4 nm for filmDMI and film-DMSO, respectively (Figure S2a,b in the Supporting Information). Figure 2c,d shows the surfaces of film-DMSO-A and film-DMI-A, respectively. The grain size of the film-DMI-A was determined to be ∼1 μm, which is 3−4 times larger than that of the film-DMSO-A showing ∼300 nm grain size. From the SEM images obtained on a large scale (Figure S3a,b in the Supporting Information), both films showed excellent surface coverage. The surface morphology was further investigated by using AFM (Figure 3). The images of surface morphology obtained from the AFM experiments agreed well with the SEM results. In addition, the roughness values of the perovskite films were evaluated by AFM, as shown in the height profile. Zheng et al. reported that film surface roughness is linearly related to the crystal size of the perovskite because of the void between the crystals and grain boundary.28 The surface roughness of the film-DMSO-A is evaluated at 25.1 nm with an average crystal size of 353.4 ± 143.9 nm by AFM. However, the surface roughness of the film-DMI-A is extremely smooth, with a root mean square of 7.3 nm and a large crystal size of 968 ± 247 nm. This result is thought to be due to the introduction of the weakly coordinating DMI as an additive. As shown in the AFM images, a significant change in the surface roughness of the film-DMSO was observed during the thermal annealing. It was reported that the formation of (MA)2Pb3I8·2DMSO intermediate is attributed to the strong coordination between DMSO and PbI2, and the intermediate liberates the DMSO molecule by thermal annealing followed by the reaction with the residual MAI in the thin film, which resulted in the formation of the tetragonal perovskite MAPbI3.24,25 Rong et al. and Cao et al. claimed that the intermediate phase caused by DMSO (denoted as Pb3I8) can be described as triple chains of edge-sharing PbI6 octahedra with all unshared apexes at their sides with disordered orientation, and the reported XRD pattern is consistent with our results shown in Figure 1a.17,21,24 Thus, it is thought that thermal 9392

DOI: 10.1021/acsami.7b18761 ACS Appl. Mater. Interfaces 2018, 10, 9390−9397

Research Article

ACS Applied Materials & Interfaces

Figure 3. Three-dimensional AFM images of (a) film-DMSO-A and (b) film-DMI-A prepared on FTO/mp-TiO2 substrates, along with the corresponding height profiles of each film.

Figure 4. (a) PL spectra of perovskite films prepared with glass/film-DMSO-A (black line), glass/film-DMI-A (red line), FTO/mp-TiO2/filmDMSO-A (black dot), and FTO/mp-TiO2/film-DMI-A (red dot) and (b) UV−vis absorption spectra of the film-DMSO-A (black) and film-DMI-A (red) prepared on FTO/mp-TiO2 substrates.

Figure 5. (a) J−V curves and (b) EQE spectrum of the film-DMSO-A and film-DMI-A devices.

coordinative DMI as an additive is more appropriate in terms of MAPbI3 nucleation than using strongly coordinating DMSO as an additive.29−31 Steady-state PL measurements and UV−vis absorption spectra were recorded for film-DMI-A and film-DMSO-A to investigate the trap states and the charge extraction ability of those films (Figure 4a,b). The films were prepared on two different substrates, glass and glass/FTO/TiO2. Generally, the spontaneous radiative recombination between trap states leads to a red-shifted emission peak compared with that from the band edge transition or passivation of these trap states, which can blue-shift the PL peak.32 Comparing the PL spectra of filmDMSO-A and film-DMI-A on the glass substrate, the wavelength of the maximum PL peak of the annealed filmDMI was ∼7 nm shorter than that of the film-DMSO-A, and

annealing initiated a secondary reaction between Pb3I8 and MAI by liberating the strong coordinative DMSO from the intermediate, resulting in a rough surface of the film-DMSO-A. The tetragonal MAPbI3 domains in the film-DMI were formed in the as-prepared samples and grew in size without changing the surface roughness of the film-DMI-A during the thermal annealing because there would be no secondary reaction during the thermal annealing. Considering the XRD, AFM, and SEM results, we concluded that the MAPbI3 crystals were homogeneously grown along the (110)-preferred orientation in the film-DMI. This conclusion accounts for the highly smooth and large MAPbI3 domains in the film-DMI-A. On the other hand, the existence of the strong coordinative DMSO refrained the homogeneous growth of the MAPbI3 domains. This finding reflects that using a weak 9393

DOI: 10.1021/acsami.7b18761 ACS Appl. Mater. Interfaces 2018, 10, 9390−9397

Research Article

ACS Applied Materials & Interfaces Table 1. Photovoltaic Parameters of the PSCs Processed with DMSO and DMIa device film-DMSO-A film-DMI-A a

avg. best avg. best

Jsc (mA/cm2)

calc. Jsc by EQE (mA/cm2)

Voc (V)

FF (%)

PCE (%)

20.4 ± 0.5 20.8 20.3 ± 0.2 20.9

21.5

1.01 ± 0.02 1.00 1.06 ± 0.01 1.09

72.4 ± 5.7 76 73.6 ± 4.3 77.6

14.5 ± 2.3 15.8 16.1 ± 0.9 17.6

20.6

The parameters are average values of 20 devices.

Figure 6. Light intensity-dependent (a) Jsc and (b) Voc of the film-DMSO-A and film-DMI-A devices. Plots of (c) charge density vs light intensity and (d) charge lifetime vs charge density of film-DMSO-A (black square) and film-DMI-A (red circle).

obtained using the film-DMI-A. Figure 5b shows the EQE spectrum and its integrated Jsc over a 100 mW/cm2 AM 1.5 G solar spectrum. The integrated Jsc values of film-DMSO-A and film-DMI-A were 21.5 and 20.6 mA/cm2, respectively, which were in good agreement with the Jsc values obtained from the J−V measurements. To analyze and understand the enhanced Voc, the charge recombination behaviors of film-DMSO-A and film-DMI-A were investigated by analyzing the light intensity dependence of Jsc and Voc. In addition, TPV measurements were conducted for both devices. As shown in Figure 6a,b, a linear dependence of Jsc on the incident light intensity was found for both devices, indicating that the charge collection efficiency was independent of the light intensity. The α values of the devices were obtained by fitting the obtained data with the power law dependence of Jsc with the light intensity (J ∝ Iα). The α value should be close to 1 when there is no recombination by the space charge effects.34−36 The α values of film-DMSO-A and film-DMI-A were obtained to be 0.93 and 0.97, respectively. Although the value of the film-DMI-A is closer to 1 than that of the filmDMSO-A, values of both devices were close to 1, which reflects that there was no significant charge recombination due to the space charge. The trap-assisted recombination can be predicted from the dependence of Voc on the light intensity. A slope of KBT/q could be obtained from the plot of Voc versus Plight, where KB is Boltzmann’s constant, T is the temperature, and q is the elementary charge. When trap-assisted or Shockley−Read−Hall recombination occurs, a stronger dependence of Voc on the light intensity with a slope greater than KBT/q is observed. The

the full width at half-maximum was narrower. These results indicate a less sharp band edge, which might be attributed to the reduced shallow trap density or defect state in the filmDMI-A.33 The extraction efficiency of photogenerated charges at the TiO2/MAPbI3 interface was investigated by measuring the PL quenching efficiency of the films (Figure 4a). Because the absorption intensities of film-DMI-A and film-DMSO-A were almost the same, the degree of PL quenching mainly depended on the charge extraction efficiency at the TiO2/MAPbI3 interface. The charge extraction efficiency of FTO/TiO2/filmDMI-A (quenching efficiency 73.7%) was significantly higher than that of FTO/TiO2/film-DMSO-A (quenching efficiency 65.9%), which implies that photogenerated charges were extracted more efficiently at the TiO2/film-DMI-A interface than at the TiO2/film-DMSO-A interface. The efficient charge extraction in the film-DMI-A was possibly due to the presence of fewer MAPbI3 defects in the film-DMI-A and the better interaction between film-DMI-A and TiO2. As a result, the bimolecular recombination at the interface between the photoactive layer and the TiO2 layer was suppressed. Solar cell devices were fabricated using film-DMI-A and filmDMSO-A, and the J−V curves and EQE spectra of the devices are presented in the Figure 5a,b. Table 1 summarizes the extracted parameters of open-circuit voltage (Voc), short-circuit current (Jsc), fill factor (FF), and PCE. The film-DMSO-A achieved a PCE of 15.8% with a Voc, Jsc, and FF of 1.00 V, 20.8 mA/cm2, and 76.0%, respectively, as measured under a reverse voltage scan. A significantly improved Voc, Jsc, FF, and PCE of 1.09 V, 20.9 mA/cm2, 77.6%, and 17.6%, respectively, were 9394

DOI: 10.1021/acsami.7b18761 ACS Appl. Mater. Interfaces 2018, 10, 9390−9397

Research Article

ACS Applied Materials & Interfaces

Figure 7. J−V curves of perovskite devices processed using different molar ratios between PbI2 and (a) DMSO and (b) DMI; J−V curves of perovskite devices fabricated at ∼55% humidity with (c) DMSO and (d) DMI.

Table 2. Summary of the Photovoltaic Parameters of PSCs Fabricated with Different Molar Ratios of DMI and DMSO (Relative to PbI2)a device

molar ratio (PbI2/DMSO or DMI)

film-DMSO-A

1:0.5 1:1 1:1.5 1:0.5 1:1 1:1.5

film-DMI-A

a

Jsc (mA/cm2) 14.2 18.8 17.6 20.3 20.8 20.1

± ± ± ± ± ±

0.38 0.73 0.53 0.21 0.33 0.29

Voc (V) 0.99 1.03 0.97 1.07 1.06 1.08

± ± ± ± ± ±

0.01 0.02 0.02 0.01 0.01 0.02

FF (%) 69.3 71.4 74.3 73.4 76.5 76.9

± ± ± ± ± ±

5.4 4.9 2.4 2.5 3.1 2.2

PCE (%) 9.0 13.5 11.2 16.7 16.4 16.8

± ± ± ± ± ±

3.4 2.8 2.9 0.6 0.4 1.1

The parameters are average values of 20 devices.

photovoltage parameters are summarized in Table 2. The film-DMI-A prepared with a molar ratio of PbI2/DMI from 1:0.5 to 1:1.5 shows similar PCEs above 17%. As shown in Figure S4a,b in the Supporting Information, the peak intensity at 2θ = 12.1° was increased with increasing concentration of DMI. Thus, we concluded that the peaks at 12.1° belong to the intermediate phases that existed in a trace amount in the prepared films, and those peaks disappeared after annealing. On the basis of the XRD results, the amount of intermediate phase in the film-DMI is smaller than that in the film-DMSO. Therefore, it is thought that CH3NH3PbI3 is the primary phase existing in the film-DMI. It is indicated that DMI is weakly coordinated with PbI2, and DMI is able to accelerate the crystallization of CH3NH3PbI3. Thus, the performance of the film-DMI is less sensitive to the concentration of DMI, whereas the PCE of the film-DMSO-A showed a strong dependence on the ratio between PbI2 and DMSO. This could be attributed to the fact that different amounts of PbI2/DMSO lead to different molecular structures of the intermediate phase in the asprepared film-DMSO.24 Furthermore, the reproducibility of the device performance under humid conditions was investigated. Devices were fabricated under a high relative humidity (∼55%), and the results are shown in Figure 7c,d. The inset photographs show the perovskite films after thermal annealing. The film-DMSO-A always exhibited white color when it was fabricated under

obtained slopes of the devices were between 1 and 2, which indicates the influence of trap-assisted recombination on the device performance.37 The slope of the film-DMSO-A was higher than that of the film-DMI-A, implying that trap-assisted recombination was more significant in the film-DMSO-A than in the film-DMI-A. The recombination mechanism was also proven using TPV measurements. As shown in Figure 6c, the charge collection efficiency of the film-DMSO-A was similar to that of the film-DMI-A, which was consistent with the result of the Jsc dependence on the light intensity. The charge carrier lifetime (τ) was extracted by fitting a monoexponential decay function (Figure 6d). As revealed in Table S1 (Supporting Information), τ of the film-DMI-A was longer than that of the film-DMI-A at different light intensities. This finding implies that the film-DMI-A had fewer charge recombination sites than the film-DMSO-A, which is attributed to the smoother film surface and reduced number of defects in the MAPbI3 crystals induced by DMI. We showed that the weak coordination capability of DMI to PbI2 allows DMI and DMF to be easily removed during the spin-coating process. Therefore, it is expected that the fabrication of perovskite solar cells with the DMI additive would be less sensitive to the stoichiometric ratio between DMI and PbI2. To prove our postulation, devices with different molar ratios between PbI2 and DMSO/DMI were fabricated; the results are shown in Figure 7a,b, and the averaged 9395

DOI: 10.1021/acsami.7b18761 ACS Appl. Mater. Interfaces 2018, 10, 9390−9397

ACS Applied Materials & Interfaces

Research Article

■

humidity conditions above ∼30%, whereas the film-DMI-A film was brown under the same condition. It is known that DMSO is highly hygroscopic because of the hydrophilic SO group, which quickly forms strong hydrogen bonds with H2O in air.38 It is predicted that DMSO or DMSO-containing intermediates would interact with H2O under a high humidity, which resulted in the unreproducible and poor device performance of the filmDMSO-A with a PCE of 11.05%. In contrast, the cyclic urea additive of DMI containing CO bonding has a lower possibility of forming a hydrogen bond with H2O. We suggest that the high humidity tolerance of the film-DMI during the fabrication process was due to the termination of the perovskite crystal with PbI2 and I atoms at the early stage of the fabrication. For the as-prepared film-DMSO, its intermediate phase terminated with DMSO is very sensitive to the fabrication condition. On the basis of the XRD results in Figure 1, it is implied that CH3NH3PbI3 films processed with DMI can be prepared at room temperature; thus, the performances of film-DMSO and film-DMI were investigated, as shown in Figure 8. The film-

CONCLUSIONS MAPbI3 films with a smooth and regular crystal size were obtained using the cyclic urea compound DMI as an additive in the precursor solution of MAPbI3. The DMI additive weakly coordinated with PbI2, which nucleated MAPbI3 through a mechanism that was different from that involving the strongly coordinating DMSO additive. A solar cell fabricated with DMI exhibited an efficient and reproducible performance even under conditions of high relative humidity. It is expected that the perovskite solar cells processed with weakly coordinating additives have a great potential for large-scale fabrication at room temperature.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.7b18761. Solubility of PbI2 in pure DMI and DMSO solvents, AFM images and height profiles of the as-prepared filmDMSO and film-DMI, SEM images of film-DMSO-A and film-DMI-A, various light intensity-dependent charge carrier lifetimes of film-DMSO-A and film-DMI-A, XRD patterns of the as-prepared perovskite films and thermally annealed perovskite films, device performance of film-DMSO and film-DMI, photovoltaic parameters of the devices fabricated from film-DMSO and film-DMI, and photographs of the as-prepared films (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected] (N.-G.P.). *E-mail: [email protected] (K.K.).

Figure 8. Device performance of film-DMSO and film-DMI under room temperature.

ORCID

Nam-Gyu Park: 0000-0003-2368-6300 Kyungkon Kim: 0000-0002-0115-8112

DMI device has a Jsc, Voc, FF, and PCE of 18.2 mA/cm2, 1.05 V, 0.74, and 14.2%, respectively, which are significantly higher than those of the film-DMSO device with a Jsc, Voc, FF, and PCE of 10.2 mA/cm2, 0.92 V, 54.9, and 5.2%, respectively. The poor performance of the film-DMSO is attributed to the DMSO-induced intermediate phases existing inside the film. Compared to the film-DMI-A, the film-DMI showed low Jsc, which was mainly due to the smaller size of perovskite grains in the film-DMI (Figure 2b). We believed that the device performance of the film-DMI will be further enhanced after sophisticated optimization of the process. It is indicated that the utilization of cyclic urea has a potential to fabricate efficient PSCs at room temperature. A photograph of the as-prepared films confirmed the above explanation (Figure S5 in the Supporting Information). The prepared sample of the film-DMSO was transparent yellow because of the existence of the intermediate of Pb3I8. In contrast, the as-prepared film of the film-DMI was dark brown, which was similar to the perovskite film after thermal annealing. These results agreed well with the XRD results. The asprepared film-DMSO requires immediate heating to prevent contact with H2O, which may be a hindrance to commercialization. However, the as-prepared film-DMI was very stable in air because of the fast formation of the perovskite crystals without intermediate steps that may come with H2O or other uncertain factors.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research was supported by NRF under the program number NRF-2015M1A2A2057506 and 2016M1A2A2940914. This research was also supported by the New & Renewable Energy Core Technology Program of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) and by a financial grant from the Ministry of Trade, Industry & Energy, Republic of Korea (no. 20163030013900). A.-N.C. and N.-G.P. acknowledge financial support from the National Research Foundation of Korea (NRF) grants funded by the Ministry of Science, ICT & Future Planning (MSIP) of Korea under contract nos. NRF-2012M3A6A7054861 (Global Frontier R&D Program on Center for Multiscale Energy System) and NRF-2015M1A2A2053004 (Climate Change Management Program).

■

REFERENCES

(1) Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T. Organometal Halide Perovskites as Visible-Light Sensitizers for Photovoltaic Cells. J. Am. Chem. Soc. 2009, 131, 6050−6051. (2) Kim, H.-S.; Lee, C.-R.; Im, J.-H.; Lee, K.-B.; Moehl, T.; Marchioro, A.; Moon, S.-J.; Humphry-Baker, R.; Yum, J.-H.; Moser, J. E.; Grätzel, M.; Park, N.-G. Lead Iodide Perovskite Sensitized All-

9396

DOI: 10.1021/acsami.7b18761 ACS Appl. Mater. Interfaces 2018, 10, 9390−9397

Research Article

ACS Applied Materials & Interfaces Solid-State Submicron Thin Film Mesoscopic Solar Cell with Efficiency Exceeding 9%. Sci. Rep. 2012, 2, 591. (3) Park, N.-G. Organometal Perovskite Light Absorbers Toward A 20% Efficiency Low-Cost Solid-State Mesoscopic Solar Cell. J. Phys. Chem. Lett. 2013, 4, 2423−2429. (4) Eperon, G. E.; Burlakov, V. M.; Docampo, P.; Goriely, A.; Snaith, H. J. Morphological Control for High Performance, SolutionProcessed Planar Heterojunction Perovskite Solar Cells. Adv. Funct. Mater. 2014, 24, 151−157. (5) Im, J.-H.; Jang, I.-H.; Pellet, N.; Grätzel, M.; Park, N.-G. Growth of CH3NH3PbI3 Cuboids with Controlled Size for High-Efficiency Perovskite Solar Cells. Nat. Nanotechnol. 2014, 9, 927−932. (6) Saliba, M.; Matsui, T.; Seo, J.-Y.; Domanski, K.; Correa-Baena, J.P.; Nazeeruddin, M. K.; Zakeeruddin, S. M.; Tress, W.; Abate, A.; Hagfeldt, A. Cesium-Containing Triple Cation Perovskite Solar Cells: Improved Stability, Reproducibility and High Efficiency. Energy Environ. Sci. 2016, 9, 1989−1997. (7) Correa-Baena, J.-P.; Abate, A.; Saliba, M.; Tress, W.; Jacobsson, T. J.; Grätzel, M.; Hagfeldt, A. The Rapid Evolution of Highly Efficient Perovskite Solar Cells. Energy Environ. Sci. 2017, 10, 710−727. (8) Tan, H.; Jain, A.; Voznyy, O.; Lan, X.; de Arquer, F. P. G.; Fan, J. Z.; Quintero-Bermudez, R.; Yuan, M.; Zhang, B.; Zhao, Y. Efficient and Stable Solution-Processed Planar Perovskite Solar Cells via Contact Passivation. Science 2017, 355, 722−726. (9) Chiang, C.-H.; Nazeeruddin, M. K.; Grätzel, M.; Wu, C.-G. The Synergistic Effect of H2O and DMF Towards Stable and 20% Efficiency Inverted Perovskite Solar Cells. Energy Environ. Sci. 2017, 10, 808−817. (10) Jeon, N. J.; Noh, J. H.; Kim, Y. C.; Yang, W. S.; Ryu, S.; Seok, S. I. Solvent Engineering for High-Performance Inorganic−Organic Hybrid Perovskite Solar Cells. Nat. Mater. 2014, 13, 897−903. (11) Ahn, N.; Son, D.-Y.; Jang, I.-H.; Kang, S. M.; Choi, M.; Park, N.G. Highly Reproducible Perovskite Solar Cells with Average Efficiency of 18.3% and Best Efficiency of 19.7% Fabricated via Lewis Base Adduct of Lead(II) Iodide. J. Am. Chem. Soc. 2015, 137, 8696−8699. (12) Li, W.; Fan, J.; Li, J.; Mai, Y.; Wang, L. Controllable Grain Morphology of Perovskite Absorber Film by Molecular Self-Assembly Toward Efficient Solar Cell Exceeding 17%. J. Am. Chem. Soc. 2015, 137, 10399−10405. (13) Guo, Y.; Shoyama, K.; Sato, W.; Matsuo, Y.; Inoue, K.; Harano, K.; Liu, C.; Tanaka, H.; Nakamura, E. Chemical Pathways Connecting Lead(II) Iodide and Perovskite via Polymeric Plumbate(II) Fiber. J. Am. Chem. Soc. 2015, 137, 15907−15914. (14) Li, L.; Chen, Y.; Liu, Z.; Chen, Q.; Wang, X.; Zhou, H. The Additive Coordination Effect on Hybrids Perovskite Crystallization and High-Performance Solar Cell. Adv. Mater. 2016, 28, 9862−9868. (15) Noel, N. K.; Habisreutinger, S. N.; Wenger, B.; Klug, M. T.; Hörantner, M. T.; Johnston, M. B.; Nicholas, R. J.; Moore, D. T.; Snaith, H. J. A Low Viscosity, Low Boiling Point, Clean Solvent System for the Rapid Crystallisation of Highly Specular Perovskite Films. Energy Environ. Sci. 2017, 10, 145−152. (16) Zhang, Y.; Gao, P.; Oveisi, E.; Lee, Y.; Jeangros, Q.; Grancini, G.; Paek, S.; Feng, Y.; Nazeeruddin, M. K. PbI2−HMPA Complex Pretreatment for Highly Reproducible and Efficient CH3NH3PbI3 Perovskite Solar Cells. J. Am. Chem. Soc. 2016, 138, 14380−14387. (17) Cao, X.; Li, C.; Li, Y.; Fang, F.; Cui, X.; Yao, Y.; Wei, J. Enhanced Performance of Perovskite Solar Cells by Modulating the Lewis Acid-Base Reaction. Nanoscale 2016, 8, 19804−19810. (18) Zhu, L.; Xu, Y.; Zhang, P.; Shi, J.; Zhao, Y.; Zhang, H.; Wu, J.; Luo, Y.; Li, D.; Meng, Q. Investigation on the Role of Lewis Bases in the Ripening Process of Perovskite Films for Highly Efficient Perovskite Solar Cells. J. Mater. Chem. A 2017, 5, 20874−20881. (19) Xiang, W.; Chen, Q.; Wang, Y.; Liu, M.; Huang, F.; Bu, T.; Wang, T.; Cheng, Y.-B.; Gong, X.; Zhong, J.; Liu, P.; Yao, X.; Zhao, X. Improved Air Stability of Perovskite Hybrid Solar Cells via Blending Poly(dimethylsiloxane)-urea Copolymers. J. Mater. Chem. A 2017, 5, 5486−5494.

(20) Lee, J.-W.; Bae, S.-H.; Hsieh, Y.-T.; De Marco, N.; Wang, M.; Sun, P.; Yang, Y. A Bifunctional Lewis Base Additive for Microscopic Homogeneity in Perovskite Solar Cells. Chem 2017, 3, 290−302. (21) Rong, Y.; Tang, Z.; Zhao, Y.; Zhong, X.; Venkatesan, S.; Graham, H.; Patton, M.; Jing, Y.; Guloy, A. M.; Yao, Y. Solvent Engineering Towards Controlled Grain Growth in Perovskite Planar Heterojunction Solar Cells. Nanoscale 2015, 7, 10595−10599. (22) Foley, B. J.; Girard, J.; Sorenson, B. A.; Chen, A. Z.; Niezgoda, J. S.; Alpert, M. R.; Harper, A. F.; Smilgies, D.-M.; Clancy, P.; Saidi, W. A. Controlling Nucleation, Growth, and Orientation of Metal Halide Perovskite Thin Films with Rationally Selected Additives. J. Mater. Chem. A 2017, 5, 113−123. (23) Mosconi, E.; Ronca, E.; De Angelis, F. First-Principles Investigation of the TiO2/Organohalide Perovskites Interface: The Role of Interfacial Chlorine. J. Phys. Chem. Lett. 2014, 5, 2619−2625. (24) Cao, J.; Jing, X.; Yan, J.; Hu, C.; Chen, R.; Yin, J.; Li, J.; Zheng, N. Identifying the Molecular Structures of Intermediates for Optimizing the Fabrication of High-Quality Perovskite Films. J. Am. Chem. Soc. 2016, 138, 9919−9926. (25) Jo, Y.; Oh, K. S.; Kim, M.; Kim, K.-H.; Lee, H.; Lee, C.-W.; Kim, D. S. High Performance of Planar Perovskite Solar Cells Produced from PbI2(DMSO) and PbI2(NMP) Complexes by Intramolecular Exchange. Adv. Mater. Interfaces 2016, 3, 1500768. (26) Chen, J.; Xiong, Y.; Rong, Y.; Mei, A.; Sheng, Y.; Jiang, P.; Hu, Y.; Li, X.; Han, H. Solvent Effect on the Hole-Conductor-Free Fully Printable Perovskite Solar Cells. Nano Energy 2016, 27, 130−137. (27) Wang, Q.; Lyu, M.; Zhang, M.; Yun, J.-H.; Chen, H.; Wang, L. Transition from the Tetragonal to Cubic Phase of Organohalide Perovskite: The Role of Chlorine in Crystal Formation of CH3NH3PbI3 on TiO2 Substrates. J. Phys. Chem. Lett. 2015, 6, 4379−4384. (28) Zheng, L.; Ma, Y.; Chu, S.; Wang, S.; Qu, B.; Xiao, L.; Chen, Z.; Gong, Q.; Wu, Z.; Hou, X. Improved Light Absorption and Charge Transport for Perovskite Solar Cells with Rough Interfaces by Sequential Deposition. Nanoscale 2014, 6, 8171−8176. (29) Song, X.; Wang, W.; Sun, P.; Ma, W.; Chen, Z.-K. Additive to Regulate the Perovskite Crystal Film Growth in Planar Heterojunction Solar Cells. Appl. Phys. Lett. 2015, 106, 033901. (30) Liang, P.-W.; Liao, C.-Y.; Chueh, C.-C.; Zuo, F.; Williams, S. T.; Xin, X.-K.; Lin, J.; Jen, A. K.-Y. Additive Enhanced Crystallization of Solution-Processed Perovskite for Highly Efficient Planar-Heterojunction Solar Cells. Adv. Mater. 2014, 26, 3748−3754. (31) Xie, L.; Hwang, H.; Kim, M.; Kim, K. Ternary Solvent for CH3NH3PbI3 Perovskite Films with Uniform Domain Size. Phys. Chem. Chem. Phys. 2017, 19, 1143−1150. (32) Shao, Y.; Xiao, Z.; Bi, C.; Yuan, Y.; Huang, J. Origin and Elimination of Photocurrent Hysteresis by Fullerene Passivation in CH3NH3PbI3 Planar Heterojunction Solar Cells. Nat. Commun. 2014, 5, 5784. (33) de Quilettes, D. W.; Vorpahl, S. M.; Stranks, S. D.; Nagaoka, H.; Eperon, G. E.; Ziffer, M. E.; Snaith, H. J.; Ginger, D. S. Impact of Microstructure on Local Carrier Lifetime in Perovskite Solar Cells. Science 2015, 348, 683−686. (34) Koster, L. J. A.; Mihailetchi, V. D.; Xie, H.; Blom, P. W. M. Origin of the Light Intensity Dependence of the Short-Circuit Current of Polymer/Fullerene Solar Cells. Appl. Phys. Lett. 2005, 87, 203502. (35) Schilinsky, P.; Waldauf, C.; Brabec, C. J. Recombination and Loss Analysis in Polythiophene Based Bulk Heterojunction Photodetectors. Appl. Phys. Lett. 2002, 81, 3885−3887. (36) Jeon, I.; Matsuo, Y. Vertical Phase Separation and Light-Soaking Effect Improvements by Photoactive Layer Spin Coating Initiation Time Control in Air-Processed Inverted Organic Solar Cells. Sol. Energy Mater. Sol. Cells 2015, 140, 335−343. (37) Koster, L. J. A.; Mihailetchi, V. D.; Ramaker, R.; Blom, P. W. M. Light Intensity Dependence of Open-Circuit Voltage of Polymer:Fullerene Solar Cells. Appl. Phys. Lett. 2005, 86, 123509. (38) LeBel, R. G.; Goring, D. A. I. Density, Viscosity, Refractive Index, and Hygroscopicity of Mixtures of Water and Dimethyl Sulfoxide. J. Chem. Eng. Data 1962, 7, 100−101. 9397

DOI: 10.1021/acsami.7b18761 ACS Appl. Mater. Interfaces 2018, 10, 9390−9397