Defect Passivation of Low-Temperature Processed ZnO Electron

Subscriber access provided by WEBSTER UNIV

Article

Defect Passivation of Low-Temperature Processed ZnO Electron Transport Layer with Polyethylenimine for PbS Quantum Dot Photovoltaics Lei Wang, Yuwen Jia, Yinglin Wang, Shuaipu Zang, Shengsheng Wei, Jinhuan Li, and Xintong Zhang ACS Appl. Energy Mater., Just Accepted Manuscript • DOI: 10.1021/acsaem.8b01756 • Publication Date (Web): 22 Feb 2019 Downloaded from http://pubs.acs.org on February 22, 2019

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 29 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 Applied Energy Materials

Defect Passivation of Low-Temperature Processed ZnO Electron Transport Layer with Polyethylenimine for PbS Quantum Dot Photovoltaics Lei Wang, Yuwen Jia, Yinglin Wang*, Shuaipu Zang, Shengsheng Wei, Jinhuan Li, Xintong Zhang* Center for Advanced Optoelectronic Functional Materials Research, and Key Laboratory of UV Light-Emitting Materials and Technology of Ministry of Education, Northeast Normal University, Changchun 130024, Jilin, P.R. China ABSTRACT: Lead sulfide (PbS) colloidal quantum dot solar cells (CQDSCs) present the distinctive ability to utilize short-wave infrared light, good ambient stability and convenient solution-based fabrication processes, thus attract much attentions in the photovoltaic research field. The performance of CQDSCs has been improved by constructing the ZnO/PbS heterojunction, due to suitable band levels and electron mobility of ZnO electron transfer layer (ETL). However, the huge number of defects in low-temperature processed ZnO cause an unbalanced carrier-related processes, which restrict further performance enhancement and flexible production of CQDSCs. Here, we described a facile method to passivate defects in low-temperature sol-gel ZnO by introducing polyethylenimine (PEI) into the precursor solution. Versus the original ZnO film, the composite ZnO:PEI films exhibit better crystallization because 1

ACS Paragon Plus Environment

ACS Applied Energy Materials 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 2 of 29

of the Zn-N interaction. A series of electronic analyses have shown that the addition of PEI reduces the work function (WF) of ZnO and increases the built-in voltage (Vbi) at the heterojunction interface, suggesting that the carrier separation is improved in the depletion region of solar cells. The carrier transport in ZnO ETL is also optimized by PEI, since the electron mobility of ZnO is maximized when the mass faction of PEI is 5%. In addition, the carrier recombination is effectively suppressed in the ZnO:PEI based solar cells proved by the increased carrier lifetime. Consequently, a power conversion efficiency (PCE) of 7.30% was achieved with the ZnO:PEI 5% film versus 5.84% for the reference cell—this was attributed to the optimized carrier-related processes. KEYWORDS: ZnO, polyethylenimine (PEI), defect passivation, CQDSCs, charge collection.

INTRODUCTION Lead sulfide (PbS) colloidal quantum dots solar cells (CQDSCs) have recently attracted enormous attentions due to their potential for multiple exciton generation and high theoretical efficiency (42%).[1,2] In addition, the large-range size-tunable band gaps of colloidal quantum dots (CQDs) allow CQDSCs to harvest short-wave infrared light.[3-7] The reported external quantum efficiency at 1350 nm could reach 80% for single-junction PbS CQDSCs[6]. And the good air stability and solution processability of CQDs can facilitate further fabrication in large-area devices via a continuous approach.[8-10] Considering the serious surface defect problems of PbS 2


Page 3 of 29 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


CQDs, researchers focused on surface modification of PbS CQDs for efficient carrier separation and transport.[11-14] Recently, a solution-phase ligand-exchange strategy, combined with advanced device engineering, has boosted the power conversion efficiency (PCE) of PbS CQDSCs to 12.01%.[14] Nowadays, ZnO is a common electron transport layer (ETL) in PbS CQDSCS due to its high transmittance in the visible and infrared regions, low cost, suitable band levels for fast electron extraction, low-temperature operation, and good solution processability.[8,10,11,15] Nevertheless, ZnO materials usually suffer from the huge amount of defect states derived from oxygen vacancies, zinc vacancies, zinc interstitials and/or oxygen interstitials.[16-18] On one hand, these defects can reduce the electron mobility (μe) of ZnO and hinder the electron transport of solar cells.[19-21] On the other hand, defects at the PbS/ZnO heterojunction aggravate the trap-assisted interfacial carrier recombination and influence the quasi-Fermi level splitting of solar cell.[22-24] Especially, the low-temperature ZnO ETLs can process more defects because of the lack of good crystallization procedure, which largely restricts the future exploitation of flexible PbS CQDSCs. The use of additives in the ZnO film is a common approach to reduce the adverse effects on solar cells.[25-28] Polyethylenimine (PEI, amine-functionalized polymer, Figure S1) and its derivative polyethylenimine ethoxylated (PEIE) can reduce the work functions of ZnO and other metal oxide layer due to the strong electrostatic or dipole interaction between metal and nitrogen atoms—these have been used as a modification layer and an additive in different optoelectronic devices.[29-31] In contrast to the dramatic effects of PEI on the 3



Page 4 of 29

performance enhancement of optoelectronic devices, the chemical, optical, electrical, and morphology variation of ZnO induced by PEI remains unclear. Furthermore, the repercussions of PEI-based ZnO on different carrier-related processes and the photovoltaic parameters of PbS CQDSCs remain unknown. Here, we prepared organic-inorganic composite ZnO:PEI films with PEI in the precursor solution of sol-gel ZnO, and for the first time employed these composite ZnO:PEI ETLs in PbS CQDSCs. We systematically investigated the influence of mass content of PEI on the properties of ZnO ETL and PbS CQDSCs. The addition of PEI could effectively passivate defects and improve crystalline grains of ZnO film. A series of the material and devices were characterized for carrier separation, transport, and recombination to understand the origin of the variation in photovoltaic performance. The appropriate amount of PEI in the ZnO film can balance different carrier-related processes, thus improves the carrier collection and photovoltaic performance of solar cell. Consequently, superior device appeared when the mass content of PEI was 5%; it exhibited a PCE of 7.30% versus 5.84% for a reference cell. EXPERIMENTAL SECTION Synthesis of PbS CQDs: PbS CQDs were synthesized by hot injection method according to the literature with some modifications

[32.33].

The as-synthesized PbS

CQDs had an exciton peak at 908 nm and were purified once by acetone and methanol, respectively. Finally, CQDs were dissolved in octane to 40 mg/mL for device fabrication. 4


Page 5 of 29 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


Preparation of Sol-gel ZnO:PEI films: Sol-gel ZnO:PEI films were prepared according to Heeger with some modifications[34]. Zinc acetate dehydrate (1.00 g), ethanolamine (0.27 mL), 2-methoxyethanol (10 mL), and different mass ratios of PEI (0%, 1%, 5%, and 9%) were blended and vigorously stirred for 12 hours in the dark. The precursor was spin-coated on the cleaned FTO or quartz substrate at 2000 rpm for 30 seconds, and the film was then annealed at 200°C for 30 minutes. Device Fabrication and Characterization: PbS CQDSCs were fabricated with a device structure of FTO/ZnO:PEI/PbS-TBAI/PbS-EDT/Au. The PbS-TBAI and PbS-EDT represented tetrabutylammonium iodide (TBAI) and ethanedithiol (EDT) capped PbS CQDs, respectively. PbS CQDs were deposited via the solid-state ligand exchange process[8]. All of the fabrication processes were operated under ambient conditions. Finally, Au electrodes were deposited by thermal evaporation. The active area (3.14 mm2) of the devices was defined by the shadow masks with a diameter of 2 mm. X-ray diffraction patterns (XRD) data were measured with Rigaku D/max-2500 X-ray diffractometer. UV-vis absorption spectra were recorded with a UH4150 spectrophotometer. Photoluminescence spectra (PL) were obtained with a 325 nm He-Cd laser from a J-Y Horiba UV-lamb micro-Raman spectrometer. X-ray photoelectron spectroscopy (XPS) measurements were performed on a Kratos AXIS ULTRA X-ray photoelectron spectroscope equipped with Al Kα radiation. Atomic force microscopy (AFM) characterizations were measured using Bruker Dimension 5



Page 6 of 29

Icon with ScanAsyst. Scanning Kelvin probe microscopy (SKPM) was taken with KP-6500 Digital Kelvin Probe System. The current density-voltage plots of the devices were acquired from Keithley 2400 sourcemeter equipped with a 3A Steady-State Solar Simulator (Enli Technology Co., Ltd.). External quantum efficiency data (EQE) were acquired using a Zolix Instruments system equipped with monochromatic illumination and standard photodetector. Capacitance-voltage (C-V), open-circuit voltage decay (OCVD), and charge extraction characteristics were collected by ModuLab XM Photoelectrochemical Test System. RESULTS AND DISCUSSION A series of sol-gel ZnO films with different mass contents of PEI were prepared and named as ZnO:PEI 0%, ZnO:PEI 1%, ZnO:PEI 5%, and ZnO:PEI 9% according to the mass ratio of PEI (mPEI/mZnO) in the films. Atomic force microscope (AFM) and scanning electron microscopy (SEM) measurements were used to observe surface morphologies of composite films. Figure 1 shows that the addition of PEI do not eliminate the wrinkled structure of the original sol-gel ZnO films or induce obvious pinholes and cracks. The PEI additive decreases the root-mean-square (RMS) of ZnO films from 10.00 nm (ZnO:PEI 0%) to 8.07 nm (ZnO:PEI 5%), which indicates that a reduced interfacial area of ZnO/PbS heterojunction may result in less interfacial carrier recombination. Figure 2a shows that the optical transmittance of these composite films on the quartz substrates are almost over 90% from 400 to 800 nm. Almost the same bandgap 6


Page 7 of 29 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


of these films are derived via the Tauc plots of (αhυ)2-hυ from the optical transmittance spectra, as displayed in the inset of Figure 2a.[18] Interestingly, the slope gradually increases with the increase of PEI, which implies a more ordered crystal lattice in the ZnO:PEI films. The diffraction peak could not be observed by X-ray diffraction (XRD) due to the low heating temperature (200°C) and thin film thickness (about 60 nm, Figure S2). Thus, we performed high-resolution transmission electron microscopy (HRTEM, Figure S3) to survey the influence of PEI on the crystallinity of ZnO films. An indistinct crystal lattice and dim diffraction ring are captured in the ZnO:PEI 0% film. In contrast, larger crystalline grains and brighter diffraction rings emerges in the HRTEM results of ZnO:PEI 1% - 9% samples. The lattice spacing values are 0.281 and 0.260 nm, which are corresponding to the (100) and (002) faces of ZnO, respectively. These results imply that PEI can improve the crystallization of sol-gel ZnO. Photoluminescence spectrum (PL) and X-ray photoelectron spectroscopy (XPS) measurements were employed to systematically investigate the defect variation of ZnO films caused by PEI. Figure 2b shows that all of the ZnO films have a strong emission peak centered at 365 nm and a broad peak around 500–600 nm. These represent the band-edge emission and the trap-assisted emission, respectively. The intensity of the trap-assisted emission peaks gradually decreases with increasing PEI, suggesting effective defect-passivation effect of PEI. Furthermore, the XPS results of the four ZnO films display Zn2p peaks with a binding energy of 1044.44 eV and 1021.03 eV and N1s peaks at 399.68 eV (Figure 2d and 2e). The O1s peaks of the 7



Page 8 of 29

four ZnO films could be disintegrated into three Gaussian peaks (Figure 2c). The peaks at 530.2eV (O 1) are associated to O atoms with their full complement of the neighboring Zn atom in the ZnO lattice array (Zn-O bond). The peaks at 521.5 (O 2) and 532.5 eV (O 3) are related with the structural defects of ZnO.[35] The former is assigned to the O atom of the zinc oxyhydroxide complexes in the oxygen-deficient regions of ZnO films, and the latter one is derived from the chemisorbed oxygen species (CO32-, H2O). The relative atomic ratio of O1/Otatol and (O 2+O 3)/Otatol (Otatol= O1+ O2+ O3) was calculated from the area of different peaks (Table S1). The proportion of oxygen related to the surface defects of ZnO decreases with PEI addition, which implies fewer defects in the ZnO films. This result is consistent with the changes of trap-assisted emission in the ZnO:PEI films. XPS depth profiles were employed to depict the PEI distribution in the ZnO films (Figure S4). The N atomic ratio of PEI-based ZnO film keeps almost unchanged when the etch time increases, which declares that the PEI disperses uniformly along the longitudinal depth of ZnO films. Consequently, the addition of PEI in the sol-gel ZnO film can improve the crystallinity and passivate the traps with negligible variation in the film transmittance and morphology. We fabricated CQDSCs via a conventional solid-state ligand-exchange method to evaluate the effect of PEI on the photovoltaic performance of the solar cells.[8] PbS CQDs (the exciton peak at 908 nm, Figure S5) were synthesized via a hot injection method. The device structure of FTO/ZnO/PbS-TBAI/PbS-EDT/Au were displayed in the schematic and cross-sectional scanning electron microscopy (SEM) images of 8


Page 9 of 29 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


CQDSCs in Figure 3a and 3b, respectively. We systematically compared the photovoltaic parameters of the four cells (Table 1) that extracted from the current density-voltage (J-V) curves (Figure 3c) measured under simulated AM1.5G illumination (100 mW cm-2). The open-circuit voltage (Voc) of solar cells increases from 0.48 to 0.52 V as the amount of PEI increases from 0% to 9%. However, the change of short-circuit current density (Jsc) is more complicated. The Jsc obtained from the reference cell with ZnO:PEI 0% film is 21.20 mA/cm2. This improves to 24.01 mA/cm2 in ZnO:PEI 5% but decreases to 19.12 mA/cm2 with ZnO:PEI 9%. The corresponding external quantum efficiency (EQE) of these cells and integrated photocurrents are shown in Figure 3d and Figure S6, respectively. The trend of fill factor (FF) is similar to that of Jsc. Consequently, a power-conversion efficiency (PCE) of 7.30% is obtained by the ZnO:PEI 5% film. This is higher than that of reference cell by 25%. The variation in the photovoltaic parameters along with the addition of PEI was more obvious in the statistical analysis of 30 cells in Figure S7a-c. These devices without any encapsulation remained durably stable in the experimental condition for more than 100 days (Figure S7d). Next, we systematically analyzed the photovoltaic performance variation through carrier transport and carrier recombination processes of solar cells. The work functions (WF) of four ZnO-based films were first measured by scanning Kelvin probe microscopy (SKPM). The WF values (Figure 4a) of reference ZnO was 4.28 eV. This decreased to 4.22 eV for ZnO:PEI 1%, 4.16 eV for ZnO:PEI 5%, and 4.14 eV for ZnO:PEI 9%, respectively. This decrease in WF for semiconductors is induced by PEI 9



Page 10 of 29

and is consistent with prior literatures.[29-31] This may be derived from the trap-passivation of ZnO by PEI. We further calculated the built-in potential (Vbi) at the ZnO/PbS interfaces by the Mott–Schottky (Figure 4b) plots obtained from the capacitance-voltage (C-V) analysis. The smaller WF of ETL and the larger Vbi values at the heterojunction interface could enhance the carrier separation in the depletion region of solar cell; thus, it is reasonable that the cells with ZnO:PEI 1% and 5% have a higher Jsc than the reference cell with an untreated ZnO film. However, the cell with ZnO:PEI 9% has an abnormal Jsc decrease. To

further

understand

this

phenomenon,

the

electron-only

devices

(FTO/ETLs/Al) were fabricated to evaluate the electron mobility (μe) change of different ZnO films by space-charge-limited current measurements (SCLC). The detailed values of μe were provided in Table S2. The μe increases from 1.06 × 10-4 cm2 V-1 S-1 to 4.46 × 10-4 cm2 V-1 S-1 as the PEI ratio increases from 0% to 5% (Figure 4c). Nevertheless, it decreases to 0.08 × 10-4 cm2 V-1 S-1 when PEI ratio increases to 9% probably due to the insulation from excess PEI.[36] The combined variation of Vbi and μe explains why the highest Jsc is obtained with the cell with ZnO:PEI 5% with a Jsc drop in the 9% PEI-based cells. The μe change of ZnO:PEI can also explain the FF difference in the devices. Open circuit voltage decay (OCVD) technology was applied to evaluate the charge recombination behavior of devices,[24,37,38] and the voltage decay profiles were shown in Figure 4d. The voltage decay rates gradually decrease as the PEI content 10


Page 11 of 29 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


increases in the sol-gel ZnO films. The electron recombination rate could be directly embodied by the electron lifetime (τn) based on the following equation:

n  

kBT dVoc 1 ( ) . q dt

Here, kB, T, and q are the Boltzmann constant, Kelvin temperature, and elementary charge, respectively. Obviously, the τn of CQDSCs increases as the PEI contents increases in the ZnO ETLs at the identical voltage. This suggests a reduction in the carrier recombination of PEI-based CQDSCs. Furthermore, the ideality factor (n) was acquired via linear fitting of Voc -logarithmic light intensity (Figure S8). The smaller n of the ZnO:PEI 5%-based device (1.43) versus that of a reference device with untreated ZnO (1.67) further suggests that the carrier recombination is inhibited in the PEI-based CQDSCs. Moreover, the density of trap states (DOS) of solar cells was measured with respect to Voc via charge extraction technology (Figure 4f).[39] The DOS gradually decrease with PEI increase. Combining the PL and XPS results discussed above, we can confirm that the addition of PEI in the sol-gel ZnO films not only passivates the ZnO traps but also suppresses the trap-assisted carrier recombination of CQDSCs. The origin of photovoltaic performance variation caused by PEI could be clarified via the properties of the different ZnO:PEI films and corresponding CQDSCs. The addition of PEI can effectively passivate the traps of sol-gel ZnO films to decrease the work function of the ZnO films and increase the Vbi at the ZnO/PbS heterojunction.

11



Page 12 of 29

Thus, the Voc increases as PEI increases due to the improved carrier separation and suppressed recombination. In addition, the μe of the ZnO films monotonically increases as the mass content of PEI increases (when lower than 5%). PEI leads to a slight up-shift of valence band edge (Ev) and conduction band edge (Ec) of ZnO (Table S3). Thus, we attribute the Jsc rise of ZnO: PEI 1% and 5% cells to the enhanced carrier separation and transport. The Jsc recession of ZnO:PEI 9% cells is mainly caused by the stagnant electron mobility of ZnO:PEI 9% ETL (μZnO:PEI 9% is an order of magnitude lower than μZnO:PEI 5%). The decreased electron mobility results in deterioration of carrier transport in the device with ZnO:PEI 9% ETL. It is worthwhile to note that ZnO film with 9% PEI lead to a lowered EQE, especially at longer wavelength. We ascribe the reduction of EQE to the structural variation of composite ZnO:PEI film caused by the excess PEI. Previous literatures proved that the excess organic polymers could generate an aggregate domain in ZnO film.[30,40] This aggregation of PEI may exist in both inside and on the surface of ZnO film. While the former can decrease the μe of ZnO film as demonstrated and discussed above, the latter might affect the packing density of PbS CQD film due to the interaction between PEI and the solvents involved in the film deposition process.[40,41] To verify this point of view, we spin-coated PEI solution onto the untreated ZnO film to ensure the aggregation of PEI on the top of ZnO film, and fabricated CQDSCs thereafter (Figure S9). The corresponding cell with PEI surface layer also showed a significant decrease of EQE at the long wavelength. Therefore, we concluded that the reduction of EQE in the cell with ZnO:PEI 9% ETL may be resulted from the 12


Page 13 of 29 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


deteriorate carrier transport in the PbS layer caused by the aggregation of excess PEI. Incidentally, the addition of PEI slightly increased the Voc, suggesting the charge recombination within CQDs layer is still the major factor for Voc deficiency.[11,42] As previous literatures, the low Voc of CQDSCs usually were caused by the traps of CQDs which accelerated the carrier recombination.[6,12,42] Consequently, series of strategies to modify the processes of CQDs synthesis and ligand exchange were employed to obtain high-quality PbS CQDs.[11,12,33] We noticed that CQDs in our work exhibited a broader full width at half-maximum (FWHM) of the absorption peak (106 nm) compared with that of PbS CQDs reported in the literature (

Defect Passivation of Low-Temperature Processed ZnO Electron

Recommend Documents