Efficient and Stable Perovskite Solar Cell Achieved with Bifunctional

Efficient and Stable Perovskite Solar Cell Achieved with Bifunctional. Interfacial Layers. Fuhua Hou,†,‡,§,║ Biao Shi,†,‡,§,║ Tiantian L...
0 downloads 0 Views 4MB Size
Research Article www.acsami.org

Cite This: ACS Appl. Mater. Interfaces 2019, 11, 25218−25226

Efficient and Stable Perovskite Solar Cell Achieved with Bifunctional Interfacial Layers Fuhua Hou,†,‡,§,∥ Biao Shi,†,‡,§,∥ Tiantian Li,⊥ Chenguang Xin,†,‡,§,∥ Yi Ding,†,‡,§,∥ Changchun Wei,†,‡,§,∥ Guangcai Wang,†,‡,§,∥ Yuelong Li,*,†,‡,§,∥ Ying Zhao,†,‡,§,∥ and Xiaodan Zhang*,†,‡,§,∥ †

Institute of Photoelectronic Thin Film Devices and Technology of Nankai University, Tianjin 300350, P. R. China Key Laboratory of Photoelectronic Thin Film Devices and Technology of Tianjin, Tianjin 300350, P. R. China § Collaborative Innovation Center of Chemical Science and Engineering (Tianjin), Tianjin 300072, P. R. China ∥ Renewable Energy Conversion and Storage Center of Nankai University, Tianjin 300072, P. R. China ⊥ School of Physical Science and Technology, Key Laboratory of Semiconductor Photovoltaic Technology at Universities of Inner Mongolia Autonomous Region, Inner Mongolia University, Hohhot 010021, P. R. China

Downloaded via UNIV OF SOUTHERN INDIANA on July 21, 2019 at 08:21:21 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.



S Supporting Information *

ABSTRACT: The elaborate control of the surface morphologies and trap states of solution-processed perovskite films significantly governs the photovoltaic performance and moisture resistance of perovskite solar cells (PSCs). Herein, a thin layer of poly(triaryl amine) (PTAA) was unprecedentedly devised on top of perovskite quasi-film by spin-coating PTAA/chlorobenzene solution before annealing the perovskite film. This treatment induced a smooth and compact perovskite layer with passivated surface defects and grain boundaries, which result in a significantly reduced charge recombination. Besides, the time-resolved photoluminescence spectra of the PTAA-treated perovskite films confirmed a faster charge transfer and a much longer lifetime compared to the control cells without the PTAA treatment. Moreover, such a hydrophobic polymer atop the perovskite layer could effectively protect the perovskite against humidity and retain 83% of its initial efficiency in contrast to 56% of control cells stored for 1 month in ambient conditions (25 °C, 35 RH%). As a result, the PTAA-treated PSCs displayed an average efficiency of 17.77% (with a peak efficiency of 18.75%), in contrast to 16.15% of the control cells, and enhanced stability. These results demonstrate that PTAA and the method thereof constitute a promising passivation strategy for constructing stable and efficient PSCs. KEYWORDS: perovskite solar cell, surface defects, PTAA, interfacial passivation, moisture resistance



INTRODUCTION The record power conversion efficiency (PCE) of the solutionprocessed organic−inorganic hybrid perovskite solar cells (PSCs) has reached 24.2%1 within a few years because of several unique properties of high absorption coefficient,2 tunable band gap3 and excellent electronic merits such as good charge carrier mobility,4 long carrier lifetime,5 and diffusion length.6 The control of both morphology and electronic properties of the perovskite absorber films are highly important to PSCs. The rate of nucleation and crystal growth mainly govern the morphology of the solution-processed perovskite films. To this end, many methods have been reported to obtain highly crystalline perovskite films, such as incorporation of additives into precursor solutions,7−10 solvent engineering,11 and interfacial modification12−16 to precisely control the amount of nuclei and manipulate the growth rates of perovskite grains. At the same time, elaborative interfacial modifications of perovskite layers with adjacent layers have © 2019 American Chemical Society

been considered as effective tactics to further boost the PCE. Bi et al. reported a method that spinning poly(methyl methacrylate) (PMMA) in antisolvent enables heterogeneous nucleation and orients perovskite nuclei to form highly crystalline perovskite films.17 Moreover, interfacial carrier recombination is another key factor limiting further improvement of device performance. For example, both Wang et al.12 and Han et al.18 reported that during the annealing processes, low boiling point organic components constituting perovskite, such as methylammonium iodide (MAI) and formamidinium iodide (FAI), vaporize from the perovskite films, leading to under-coordinated ions or vacancies at the surfaces and between grain boundaries, thus generating surface and bulk defects that retard the charge Received: April 12, 2019 Accepted: June 20, 2019 Published: June 20, 2019 25218

DOI: 10.1021/acsami.9b06424 ACS Appl. Mater. Interfaces 2019, 11, 25218−25226

Research Article

ACS Applied Materials & Interfaces

Figure 1. Schematics of the deposition methods of the control device and the PTAA-capped device, and the corresponding complete device configurations. (a) Control device. (b) PTAA-capped device.

surface defects of mixed-halide perovskite absorbers, (FAI)0.81(PbI2)0.85(MAPbBr3)0.15, and inducing heterogeneous nucleation to form perovskite crystals with a large grain size. It is well known that PTAA, as an excellent hole-transport material (HTM), was employed commonly in conventional and inverted PSCs. Moreover, PTAA can provide the amine functional group with the lone pair of electrons to passivate the defects and also coordinate with the under-coordinated Pb2+ of the perovskite films. The PSCs show an average PCE of 17.77% with the PTAA treatment, superior to that of the reference devices (16.15%). In addition, the stability of the PSCs is greatly improved from 56 to 83% after the treatment with PTAA when stored in ambient condition (25 °C, 30−40 RH%) for 1 month.

transport or cause serious recombination. Consequently, the under-coordinated Pb atoms results in the accumulation of photogenerated positive carriers within the hole-transporter phase at the heterojunction,19,20 which may be one of the reasons of the hysteresis phenomenon, and also may result in the low performance of short-circuit current density (JSC) and fill factor (FF). Therefore, to further improve the performance, rational surface passivation techniques by reducing surface trap states and charge carrier recombination occurring at the heterojunction are critical. Normally, the surface defects of the undercoordinated Pb2+ of the perovskite film can be passivated by some amine-functional materials.21 The amine-functional materials could supply the lone pair of electrons to the under-coordinated Pb2+ to form a dative-covalent or a coordinate bond at the surface of the perovskite film. For example, pyridine,20,22 thiophene,20 C8-diammonium iodide,23 amine-functionalized organic molecules,21−25 and [N-9″-heptadecanyl-2,7-carbazole-alt-5,5-(4′,7′-di-2-thienyl-2′,1′,3′-benzothiadiaz-ole)] (PCDTBT)24 were used to passivate the surface of the perovskites in PSCs. Meanwhile, the stability of the perovskite has been improved to some extent. Although the surface treatments of perovskite films with a hydrophobic PMMA26 or polystyrene15 have been evidenced to endow the films with excellent moisture tolerance, while the nonconducting interfacial layers adversely increase the series resistance and thus deteriorate the FF of the PSCs. To avoid this phenomenon, Peng et al. added the highly conductive PCBM into PMMA to maximize its efficiency by adjusting the trade-off between passivation and conductivity.27 Ideally, the passivation layers should combine the amine-functional molecule with hydrophobic properties to minimize the surface defects and improve the moisture resistance of the perovskite films simultaneously. In this work, a hydrophobic poly(triaryl amine) (PTAA) was introduced through a simple antisolvent process, i.e., spin-casting PTAA in the chlorobenzene solution during the second process of perovskite deposition. We employed the polymeric PTAA for the purpose of passivating



RESULTS AND DISCUSSION Figure 1 illustrates the preparation of the PTAA-modified perovskite film. Perovskite films were prepared by one-step, solution-processed method. The perovskite precursor solution composed of formamidinium iodide (FAI), lead(II) iodide (PbI2), methylammonium bromide (MABr), and lead(II) bromide (PbBr2) dissolved in a mixed solvent (DMF/DMSO = 4:1) with a chemical formula of (FAI)0.81(PbI2)0.85(MAPbBr3)0.15. The device structure is Au/ Sprio-OMeTAD/perovskite/SnO2/FTO, with successive spincoating hydrosol SnO2 as the electron-transport layer, perovskite as the absorber layer, Spiro-OMeTAD as the hole-transport layer (HTL). The major difference here is the use of PTAA as the interfacial layer to passivate the perovskite film to induce a larger crystal size and coordinate Pb2+ to reduce surface defects. The molecular structure of PTAA and the energy diagram of the PSCs are displayed in Figure S1. In comparison to the conventional antisolvent methods reported by Seok’s group, who prepared perovskite with pure antisolvent,28 we dissolved a certain amount of PTAA in chlorobenzene with varied concentrations of CPTAA of 0, 2, 4, 6, and 8 mg/mL (marked as P0, P2, P4, P6, and P8, respectively) to extract organic solvent from perovskite precursor films and 25219

DOI: 10.1021/acsami.9b06424 ACS Appl. Mater. Interfaces 2019, 11, 25218−25226

Research Article

ACS Applied Materials & Interfaces

device structures were fabricated (Table S2). In addition, device with PTAA treatment but without Spiro-OMeTAD as HTL displayed a lower performance. Meanwhile, if we simply made the device with ITO/SnO2/perovskite/Au, the efficiency was almost zero. This indicates that a thin PTAA introduced by dripping antisolvent-deposited method cannot fully act as HTL alone. Moreover, PTAA introduced by postdeposited can partly improve the device performance but not as efficient as that PTAA incorporated through antisolvent process, as demonstrated in Table S2. When the PTAA concentration increases up to 8 mg/mL (P8), the hydrophobic nature of PTAA and fast extracting organic solvents from perovskite precursor film generates an internal stress and triggers crack formation within the perovskite films, which potentially deteriorates JSC and FF, as listed in Table S1. We need to note that the PTAA layer can be partially dissolved during the spin-coating of Spiro-OMeTAD because the same solvent, chlorobenzene, used; however, a thin layer is still obviously observed when filling the grain boundaries or covering the surface of perovskites, as testified by spin-casting 100 μL of chlorobenzene (nearly three times the volume of 35 μL of Spiro-OMeTAD/chlorobenzene solution) directly onto annealed PTAA-capped perovskite film, as seen in the SEM images of Figure S3c,f. Furthermore, the hydrophobicity of PTAA is also tested by water contact angle measurement in which the gradually improved PTAA concentrations at the surface of the perovskite displays increased water contact angles (Figure S3) of 52.13° for control device and 87.19° for P6-modified films (insets in Figure 2a,b), which are desirable for better moisture resistance of the PSCs. It is worthy noting that a thin PTAA layer also prevents the perovskite layer from the harmful polar solvent vapor during the spin-coating process and confines MAI or FAI within the perovskite film to minimize the surface vacancy defects as well. Meanwhile, the PTAA polymer molecule with lone pair of electrons can anchor the molecule to the Pb−I octahedron through coordination with the Pb ions or hydrogen bonding with the iodide ions,30 and thus reduce the PbI2 at the surface of the perovskite, which is the reason of the lower X-ray diffraction (XRD) signal intensity of PbI2 (12.7°) as described in Figure 3a. Moreover, the peak intensity of the PTAAmodified perovskite crystalline (14.0°) is higher than that the control and the crystal structure is similar to that reported by other groups.19,31 To further characterize the surface properties of the perovskite films, X-ray photoelectron spectroscopy (XPS) was conducted. Figure 3b exhibits the binding energies relative to Pb valence electrons, the XPS peaks of the PTAA-capped perovskite film are located at 142.4 and 137.6 eV, which can be assigned to Pb 4f5/2 and 4f7/2, respectively. The peak positions shift by 0.7 eV toward lower binding energy, indicating the occurrence of some molecular interactions between PTAA and (FAPbI3)0.85(MAPbBr3)0.15. Generally, when an element acquires electrons, the binding energy is decreased. This phenomenon confirms that the lone pair of electrons of the PTAA molecule was successfully delocalized through interactions with under-coordinated Pb ions.20 Figure S4a displays the hysteresis phenomenon in the J−V curves of the PSCs fabricated by two different methods. It is obvious that the device treated by PTAA exhibits almost identical J−V curves under two scanning directions, whereas notable hysteresis is demonstrated for the reference device. The higher JSC of the PTAA-modified device indicates a

simultaneously grow a thin PTAA layer atop the perovskite to intentionally manipulate the formation process and passivate surface trap states. After that, thin films were transferred immediately to the hot plate at 130 °C for 20 min to form perovskite. The photoelectric parameters of the PTAAmodified PSCs are displayed in Table S1, and the P6-modified PSCs show the best performance averaged from 15 samples. Although PTAA has been widely used as HTL, it has been rarely investigated to have an effect on the perovskite film formation during the deposition process. In the experiment, with the dripping solvent mixing and extracting the major solvent (DMF/DMSO = 4:1), the carried PTAA can insert the perovskite surface and cap it. The PTAA can slow down the perovskite nucleation rate and induce the crystal growth. As a result, the PTAA-modified perovskite can induce highly crystalline perovskite films. The surface morphologies of the P6-modified films are shown in Figure 2b,d, which are different

Figure 2. Morphology of the perovskite films with and without PTAA treatment. (a, c) Cross-sectional SEM images (insets: water contact angle images). (b, d) Top-view SEM (insets: AFM images). Scale bars, 300 nm.

from those of the pristine perovskite films (Figure 2a,c). The average grain sizes of the P6-modified perovskite films are approximately 500 nm compared to those of the pristine perovskite film of 300 nm. As the concentration of PTAA increases to 6 mg/mL (P6), a continuous and smooth PTAA layer is gradually formed on the perovskite film, as presented in Figure S2. Moreover, the continuous PTAA films formed on the surface of the perovskite films and between the grain boundaries thus levels the roughness of the perovskite film, as confirmed in the AFM images (inset in Figure 2c,d), where the roughness values of 37.7 and 26.3 nm were obtained from the control and the P6-modified perovskite film, respectively. These results are consistent with the cross-sectional SEM images (Figure 2a,b). At the same time, a uniform and thin PTAA layer fully covering the perovskite film builds a direct formation of a perovskite−PTAA heterojunction to accelerate hole extraction from the perovskite film to HTL due to the gradient band energy formed between perovskite−PTAA and Spiro-OMeTAD, as depicted in Figure S1b. The incorporation of PTAA can influence the work function of the perovskite film and the interfacial band alignment; the charge-transfer dynamics are similar to those reported by Wu et al.29 To better explain the PTAA-modified perovskite films, different 25220

DOI: 10.1021/acsami.9b06424 ACS Appl. Mater. Interfaces 2019, 11, 25218−25226

Research Article

ACS Applied Materials & Interfaces

Figure 3. Surface passivation of the perovskite layer by PTAA. (a) XRD patterns; (b) Pb 4f XPS spectra; (c) steady photoluminescence (PL) spectra; and (d) time-resolved PL spectra.

the bulk recombination.39 The control perovskite without P6 modification gives a fast and slow lifetime of τ1 = 9.4 ns and τ2 = 182.9 ns with A1 = 1.57 and A2 = 98.43, respectively. However, the sample with P6-treated film shows τ1 = 5.0 ns and τ2 = 225.6 ns with A1 = 3.44 and A2 = 96.56, respectively. The average PL decay lifetimes can also be calculated by eq 2

reduced recombination, resulting in high charge separation yields.32 We also test the effect of varying scan rates on the P6modified PSCs. As shown in Figure S4b, the PSCs show negligible hysteresis in the J−V curves. The photoelectric parameters extracted from the J−V curves are listed in Table S3. It has been reported that the photocurrent hysteresis in the PSCs can be partly due to the trap states at the interface of the perovskite/HTL.12 The lone pair of electrons of N atoms of the PTAA molecule can coordinate with Pb2+, reducing the carrier recombination centers. Therefore, the PTAA-modified perovskite films can reduce the defect states and benefit to eliminate the hysteresis phenomenon. Moreover, to study the hole-quenching ability of the perovskite films with PTAA treatment, the PL spectra were recorded, as illustrated in Figure 3c. The control perovskite film shows a high intensity of the steady PL emission peak; after Spiro-OMeTAD deposition, a higher quenching ability with PTAA-capped perovskite is easily observed. The HOMO of PTAA is −5.3 eV,33,34 which matches well with those of Spiro-OMeTAD (−5.2 eV)35 and (FAI)0.81(PbI2)0.85(MAPbBr3)0.15 (−5.71 eV).31,36 Such a conspicuous PL quenching explained that the PTAA-modified interface of perovskite-PTAA/Spiro-OMeTAD can efficiently reinforce the hole extraction and injection from the absorber layer to Spiro-OMeTAD HTL. To illustrate the carrier lifetimes of the perovskite films with and without PTAA treatment, the time-resolved PL measurement was performed, as demonstrated in Figure 3d. Biexponential function is employed to fit the PL decay curves (eq 1)37,38 ij t yz ij t yz Y = A1 expjjj− zzz + A 2 expjjj− zzz + y0 j τz j τ z k 1{ k 2{

τave =

∑ Ai τi2 ∑ Ai τi

(2)

τave for the reference and PTAA-treated perovskite films are 182.76 and 225.43 ns, respectively. These results represent a superior perovskite film quality for PTAA treatment with a lower defect concentration, in great agreement with the higher FF of the corresponding PSCs. To monitor the interfacial change in photovoltaic devices, electrochemical impedance spectroscopy (EIS) measurement was carried out to investigate the role of the PTAA treatment at the perovskite/HTL interface. The Nyquist plots were recorded under dark conditions with the potential bias of 0.9 V and the alternating current amplitude of 10 mV. The frequency ranged from 1 Hz to 100 kHz (Figure S5). Normally, the diameter of the main semicircle in the low-frequency region shows the recombination resistance (Rrec) of the electrons from the active layer/SnO2 with Spiro-OMeTAD.15 As can be seen in the image, the increased radius of the semicircle of the P6-modified perovskite films indicates an improved recombination resistance at the interface of the perovskite/HTM, which is beneficial in reducing the charge recombination at the interface. Compared with the pristine devices, the recombination resistance of the PTAA-modified devices is obviously improved, indicating the efficient blocking of the charge recombination. To further scrutinize the optoelectronic properties of the two different perovskite layers, the space charge limited current (SCLC) was measured. Figure S6 shows the J−V curves of the hole-only devices recorded under dark conditions and plotted

(1)

where τ1 and τ2 are the lifetimes for the fast and slow recombinations; and A1 and A2 are the corresponding amplitudes. τ1 is due to the recombination occurring in the surface of the perovskite film, whereas τ2 could be attributed to 25221

DOI: 10.1021/acsami.9b06424 ACS Appl. Mater. Interfaces 2019, 11, 25218−25226

Research Article

ACS Applied Materials & Interfaces

Figure 4. Statistics of photovoltaic parameters distribution of the PSCs fabricated with and without P6 modification. (a) JSC; (b) VOC; (c) FF; and (d) PCE.

Figure 5. Optimized performance of the PTAA-capped perovskite device. (a) A colored cross-sectional SEM image. Scale bars, 300 nm. (b) J−V curves of the champion PSC prepared with optimized PTAA treatment in both reverse and forward scan directions, showing negligible hysteresis. (c) EQE and the corresponding integrated JSC of the device. (d) Steady-state photocurrent measured at a bias voltage (0.92 V) over 300 s of a maximum power point tracking.

on a double logarithmic graph. There are three regions in the dark J−V curves. At low bias voltage (