High Efficiency Low-temperature Processed Perovskite Solar Cells

Randi Azmi , Sunbing Hwang , Wenping Yin , Tae-Wook Kim , Tae Kyu Ahn , and Sung-Yeon Jang. ACS Energy Lett. , Just Accepted Manuscript. DOI: 10.1021/...
0 downloads 5 Views 1MB Size
Subscriber access provided by UNIVERSITY OF THE SUNSHINE COAST

Letter

High Efficiency Low-temperature Processed Perovskite Solar Cells Integrated with Alkali Metal Doped ZnO Electron Transport Layers Randi Azmi, Sunbing Hwang, Wenping Yin, Tae-Wook Kim, Tae Kyu Ahn, and Sung-Yeon Jang ACS Energy Lett., Just Accepted Manuscript • DOI: 10.1021/acsenergylett.8b00493 • Publication Date (Web): 01 May 2018 Downloaded from http://pubs.acs.org on May 1, 2018

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 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 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.

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 22 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

ACS Energy Letters

High Efficiency Low-temperature Processed Perovskite Solar Cells Integrated with Alkali Metal Doped ZnO Electron Transport Layers Randi Azmi,† Sunbing Hwang,§ Wenping Yin,‡ Tae-Wook Kim,§ Tae-Kyu Ahn,‡ Sung-Yeon Jang†*



Department of Chemistry, Kookmin University, 77 Jeongneung-ro, Seongbuk-gu, Seoul 136702, Republic of Korea.

§

Applied Quantum Composites Research Center, Institute of Advanced Composite Materials, Korea Institute of Science and Technology, Joellabuk-do 565-905, Republic of Korea ‡

Department of Energy Science, Sungkyunkwan University, 2066 Seobu-Ro, Jangsan-Gu, Suwon 440-746, Republic of Korea.

Email: [email protected]

1 ACS Paragon Plus Environment

ACS Energy Letters 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 Herein, we achieved, air-stable low-temperature processed PSC (L-PSC) using alkalimetal modified ZnO ETLs. Using a simple chemical alkali-metal modification method, the surface defects of the ZnO were effectively passivated. As a result, the interfacial decomposition reactions were suppressed, while raising the Fermi energy level and enhancing electron mobility. The improved interfacial charge transfer and internal electric field in the developed L-PSC using K modified ZnO (ZnO-K) exhibited an improved power conversion efficiency (PCE) of 19.91% with negligible hysteresis, while a pristine ZnO based L-PSC exhibited a PCE of 16.12% with significant hysteresis. The ZnO-K based L-PSC also exhibited remarkably higher long-term airstorage stability (91% retention after 800 h) than pristine ZnO based L-PSCs (36% retention after 800 h) due to the suppressed decomposition reactions. The PCE and air-stability of our L-PSC with the modified ZnO are among the highest reported for PSCs processed ≤ 150 °C.

2 ACS Paragon Plus Environment

Page 2 of 22

Page 3 of 22 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

ACS Energy Letters

TOC GRAPHICS

3 ACS Paragon Plus Environment

ACS Energy Letters 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

Among emerging next-generation solar cells, perovskite based solar cells (PSCs) have attracted a great deal of attention because of their high power-conversion-efficiency (PCE > 22%),1,2 and low fabrication cost.3-5 Their high PCE is largely attributed to high optical absorption in the visible regime and the long charge diffusion length (> 1 µm) of the perovskite photoactive layers.6 Charge extraction losses in such devices are usually more likely to occur at the interfaces between the perovskite layers and the charge transport buffer layers, such as the electron transport layer (ETL) and hole transport layer (HTL). Engineering of these buffer layer materials has been one of the principle targets to improve device charge extraction.7,8 When used in ETLs, mesoporous TiO2 layers have exhibited state-of-the-art performance and have been widely employed in high-efficiency PSCs.

1,9

However, their high processing termperature

(>450 °C) has been a major shortcoming, leading to high production cost and preventing it use with flexible substrates.10,11 ZnO is one of the strong candidate materials for an advanced ETL because it exhibits proper optoelectronic properties even when prepared by low-temperature solution process.12-14 Recent studies on ZnO ETLs have successfully improved the device PCE up to >18 %,15-17 however to compete with the TiO2 ETLs further improvement is required. The major drawbacks of low-temperature processed ZnO ETLs are the reverse decomposition reaction at the perovskite/ZnO interfaces and the suboptimal energy levels for charge extraction.12,18-20 Minimizing such drawbacks is key to achieve high-efficiency low-temperature PSCs.12,14 The reverse decomposition reaction from perovskite to PbI2 is facilitated by the surface oxygencontaining defect groups on ZnO,19-21 and accordingly, their passivation is essential to reduce the decomposition reactions. At the same time, the manipulation of energy levels is also beneficial for improving interfacial charge extraction.8,10,15,17 4 ACS Paragon Plus Environment

Page 4 of 22

Page 5 of 22 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

ACS Energy Letters

Doping with metal heteroatoms has been an effective strategy for modifying the properties of ZnO.15,16,21 It has been reported that the proper doping of ZnO with metal heteroatoms (such as Al, Mg, Li, Cs) can often reduce the oxygen deficient defects, while simultaneously improving electron mobility.15,16,21 In those previous reports, the partial reduction of interfacial decomposition reactions was achieved, resulting in a device PCE of 16 – 18 %,15,16,21 however relatively high-temperature annealing (200 – 290 °C),15,21 or low-pressure sputtering,16 which are not scalable processing methods, were used. Moreover, the stability of those devices was still insufficient, even though the decomposition reaction was suppressed. Development of a more effective metal doping method can further improve the potential of the ZnO ETL and will be beneficial for the development of high performance low-temperature PSC devices. In this work, we report high-efficiency low-temperature PSCs (L-PSCs) with improved air-stability, using alkali-metal doped ZnO as the ETLs. ZnO was successfully doped by chemical treatment of the ZnO films using aqueous solutions of alkali-metals, with simultaneous passivation of surface defects. These alkali-metal doped ZnO-ETLs demonstrated properties superior to pristine ZnO-ETL. Specifically, 1) the reverse decomposition reaction at the perovskite/ZnO interface was significantly suppressed; 2) the interfacial charge extraction in the devices was improved; and 3) the long-term stability of devices under air was improved. The LPSC using K modified ZnO (ZnO-K) achieved a PCE of 19.91% (VOC of 1.126 V, JSC of 22.95 mA cm−2, FF of 0.771) with negligible photocurrent hysteresis, whereas the L-PSC with pristine ZnO showed a PCE of 16.12% (VOC of 1.068 V, JSC of 21.14 mA cm−2, FF of 0.715) with considerable hysteresis. The long-term storage stability and photostability of L-PSCs under air

5 ACS Paragon Plus Environment

ACS Energy Letters 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

were also improved remarkably: after 800 h of air-storage, the PSCs using ZnO-K retained 91% of their initial PCE, whereas the pristine ZnO based PSCs retained only 36%.

Figure 1. XPS spectra (a) K 2p, (b) the atomic percent profile of K with respect to the depth of ZnO-ETL, and (c) O1s core level. (d) PL emission spectra, and (e) Schematic illustration of energy levels of ZnO-ETLs and perovskite film determined by UPS. Ar beam was used for surface etching. Pristine ZnO-ETLs with a thickness of ≈40 nm were prepared by in situ sol-gel conversion at 130 °C.18,22 The alkali-metal modification was performed by dipping the ZnOETLs in alkali-metal solutions (1.5 mM) in deionized (DI) water for 1 min followed by washing with DI water and drying at 100 °C for 10 min. The doping of alkali-metals, Li, Na, K, on ZnO was revealed by X-ray photoelectron spectroscopy (XPS) analysis (Figure S1). The atomic

6 ACS Paragon Plus Environment

Page 6 of 22

Page 7 of 22 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

ACS Energy Letters

percent of alkali-metals doped on the surface ZnO-ETLs, as determined from the XPS spectra shown in Figure S1, were 5.6%, 4.9%, and 5.8% for the Li doped ZnO (ZnO-Li), Na doped ZnO (ZnO-Na), and ZnO-K, respectively. To investigate the penetration depth of K in the ZnO films, we monitored the K 2P XPS spectra with respect to etching times using an Ar ion beam (Figure 1a). The depth profiling is plotted in the inset of Figure 1a. The atomic percent of K gradually decreased from the initial value, and the penetration depth was ≈20 nm. Figure 1b shows the O 1S core level spectra of the doped ZnO-ETLs (ZnO-Li, ZnO-Na, ZnO-K). Generally, the O 1S spectra can be deconvoluted into two peaks, the lower binding energy peak (≈530.5 eV) that is associated with the oxygen atom in the ZnO crystal lattice, and the higher binding-energy peak (≈532 eV), which is attributable to oxygen-deficient defects (oxygen vacancies (Ov), surface hydroxyl groups (─OH) etc…) of ZnO.12,14 As shown in Figure 1b, the oxygen-deficient defect peak at ≈532 eV was considerably reduced by the alkali-metal modification following the mechanism Zn─OH + AOH → Zn─O─A + H2O (where A represents the alkali-metals).23,24 To confirm the reduction in defect density, we also measured photoluminescence (PL) emission spectra (Figure 1c). The doped ZnO films exhibited a reduction in defect-related PL emissions at ≈550 nm.12,14,18 These defects often act as charge trap sites. These results are in good agreement with the previous literature regarding the reduction of surface hydroxyl groups in metal oxides (ITO, ZnO, etc…) by alkali-metal modification.23,24 To investigate the influence of metal doping on the charge transport properties, the electron mobility of the ZnO-ETLs was determined by space-charge-limited-current (SCLC) analysis. As shown in Figure S2, the doped ZnO-ETLs exhibited enhanced electron mobility compared to pristine ZnO-ETLs. The ZnO-K showed ≈55% higher average electron mobility

7 ACS Paragon Plus Environment

ACS Energy Letters 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

(1.45 × 10-3 cm2/V⋅s) than the pristine ZnO-ETL (0.65 × 10-3 cm2/V⋅s). The electron mobility of the ZnO-ETLs is summarized in Table S1. The effects of metal doping on the energy levels of ZnO-ETLs were investigated by ultraviolet photoelectron spectroscopy (UPS) analysis. As shown in Figure S3a, the onset of the secondary electron cut-off region was shifted to a higher energy by the alkali-metal doping, which indicates the up-shifting of the Fermi energy level. The valence bands were determined by the onset energy of the valence band maxima in the low binding energy region (Figure S3b). The conduction bands were calculated by using valence band values and optical bandgap from the absorbance spectra (Figure S3c). The energy diagrams of the doped ZnO-ETLs are shown in Figure 1d. It is well-known that alkali-metal doping can raise the EF of metal oxides.23,24 The upshift in EF is dependent on the alkali-metals, and the ZnO-K showed the largest shift (≈0.21 eV) from the values of the pristine ZnO (4.39 eV). The shift in EF was consistent with the previous literature,23,24 confirming the efficient doping by our solution treatment. The methylammonium lead triiodide (MAPbI3) based perovskite active layers were deposited on the ZnO-ETLs by a two-step sequential deposition method.17 The detailed deposition process is described in the experimental section in the supporting information (SI). We employed the two-step deposition because it is known to partially mitigate the reverse decomposition reactions.17 The effects of doping on the decomposition of perovskite layers on the ZnO-ETLs were investigated. As shown in Figure S4, the decomposition of perovskite layers after thermal annealing at 100 °C was significantly prevented on the ZnO-K, compared to pristine ZnO. In the samples prepared on the pristine ZnO, yellow colored PbI2 formed substantially after 30 min of annealing. This result was also supported by the absorption spectra (Figure S4c) and XRD analysis results (Figure S4d) after 30 min of annealing at 100 °C. The 8 ACS Paragon Plus Environment

Page 8 of 22

Page 9 of 22 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

ACS Energy Letters

perovskite film on pristine ZnO exhibited reduced absorption with enhanced PbI2 peak. This result is in a good agreement with the previous literature.20,21 In contrast, the reverse decomposition reaction was effectively suppressed in the perovskite layers on ZnO-K even after 30 min of annealing at 100 °C (Figure S4b). Figure 1e shows scanning electron microscopy images of the perovskite layers on ZnO and ZnO-K. The morphology and crystal grain size of the two samples are nearly identical. The X-ray diffraction (XRD) spectra of the two perovskite samples are also similar with almost the same (001) PbI2 peak intensity (Figure S5). The similar quality of the perovskite layers prepared on the ZnO and ZnO-K indicates that the metal doping does not significantly influence on the perovskite crystal formation, while the reverse decomposition was suppressed.

9 ACS Paragon Plus Environment

ACS Energy Letters 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 10 of 22

Figure 2. (a) Devices structure, (b) J-V characteristics of L-PSCs using various ZnO-ETLs under AM 1.5G one sun illumination (100 mWcm-2). (c) Hysteresis characteristics of L-PSCs using pristine ZnO and ZnO-K under reverse and forward scan. (d) External quantum efficiency (EQE) and integrated JSC of pristine based device and ZnO-K based device. Table 1. Summary of device performance. The values in brackets indicate the averages and standard deviations. The 20 devices for each type of ZnO-ETL are used for the statistics.

ETL

PCE (%)

VOC (V)

JSC (mA cm-2)

10 ACS Paragon Plus Environment

FF

Page 11 of 22 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

ACS Energy Letters

ZnO

ZnO-Li

ZnO-Na

ZnO-K

16.10

1.07

(15.40±0.751) (1.05±0.02) 17.80

1.10

(16.60±0.722) (1.08±0.02) 18.90

1.11

(17.70±0.691) (1.10±0.02) 19.90

1.13

(18.90±0.582) (1.11±0.02)

21.10

0.715

(20.90±0.49) (0.698±0.02) 22.00

0.733

(21.40±0.43) (0.714±0.02) 22.50

0.754

(21.90±0.44) (0.733±0.02) 23.00

0.771

(22.50±0.41) (0.756±0.02)

L-PSC devices with a structure of ITO/ZnO/perovskite/Spiro-OMeTAD/Au were fabricated (Figure 2a). The cross-sectional SEM image of a device is shown in Figure S5. All fabrication steps were carried out under ambient air. The current density-voltage (J-V) characteristics of the L-PSCs with various ZnO-ETLs are shown in Figure 2b. The L-PSCs using pristine ZnO exhibited a PCE of 16.12% under the reverse scan direction, which is comparable to other reported ZnO-based L-PSCs.15-17,21,25 While the ZnO-K based device showed the highest PCE, the ZnO-Na and ZnO-Li based devices also showed superior PCEs compared to pristine ZnO based devices, confirming the effects of the alkali-metal modification. The PCE of the optimized ZnO-K based L-PSC under reverse scan was 19.91% (VOC of 1.126 V, JSC of 22.95 mA cm−2, FF of 0.771) with negligible hysteresis (Figure 2c). The highest VOC of ZnO-K based L-PSC than other alkali-metals (ZnO-Na and ZnO-Li based L-PSC) is due to its boosted EF level (Figure 1e). The FF of ZnO-K based L-PSC was superior to other devices. The charge recombination at the various devices is discussed in the next sections. The JSC value from the J-V analysis was in good agreement with the calculated JSC value from the EQE analysis with