Effects of Cd Diffusion and Doping in High ... - ACS Publications

Subscriber access provided by University of Sussex Library

Article

Effects of Cd Diffusion and Doping in High-Performance Perovskite Solar Cells Using CdS as Electron Transport Layer Wiley A. Dunlap-Shohl, Robert Younts, Bhoj Raj Gautam, Kenan Gundogdu, and David B. Mitzi J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.6b05406 • Publication Date (Web): 11 Jul 2016 Downloaded from http://pubs.acs.org on July 13, 2016

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

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

Page 1 of 27

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

The Journal of Physical Chemistry

Effects of Cd Diffusion and Doping in High-Performance Perovskite Solar Cells Using CdS as Electron Transport Layer

Wiley A. Dunlap-Shohl†, Robert Younts‡, Bhoj Gautam‡, Kenan Gundogdu‡ and David B. Mitzi†§*

†

Department of Mechanical Engineering and Materials Science, Duke University, Durham, NC, 27708 ‡

Department of Physics, North Carolina State University, Raleigh, NC 27695 §

Department of Chemistry, Duke University, Durham, NC 27708

*Corresponding Author Contact Information: Email: [email protected] Tel: (919)-660-5356

1 ACS Paragon Plus Environment


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 27

Abstract Perovskite solar cells with stabilized power conversion efficiency exceeding 15% have been achieved, using a methylammonium lead iodide (MAPbI3) absorber and CdS as the electron transport layer (ETL). X-ray photoemission spectroscopy (XPS) reveals a small presence of Cd at the surface of most perovskite films fabricated on CdS. To understand the possible impacts of Cd diffusion into the perovskite absorber layer, perovskite films were deliberately doped with Cd. Doping substantially increases the grain size of the perovskite films, but also reduces device performance through the formation of an electrical barrier, as inferred by the S-shape of their J-V curves. Photoluminescence decay measurements of the doped films do not indicate substantial non-radiative recombination due to bulk defects, but a secondary phase is evident in these films, which experiments have revealed to be the organic-inorganic hybrid material methylammonium cadmium iodide, (CH3NH3)2CdI4. It is further demonstrated that this compound can form via the reaction of CdS with methylammonium iodide and may form as a competing phase during deposition of the perovskite. Buildup of this insulating compound may act as an electrical barrier at perovskite interfaces, accounting for the drop in device performance.


Page 3 of 27

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


Introduction Perovskite solar cells based on CH3NH3PbI3 (or MAPbI3, where MA = CH3NH3+ = methylammonium) have attracted considerable interest since their introduction as a photovoltaic material by Kojima et al in 2009, achieving 20% power conversion efficiency (PCE) after scarcely more than half a decade of research.1,2,3,4,5 Perovskites owe their superior photovoltaic properties to a wide range of favorable optical and electronic properties, such as high light absorption coefficient, long carrier lifetimes/diffusion lengths, and benign grain boundaries.6,7,8 Moreover, high quality perovskite thin films may be easily synthesized via simple and rapid solution-processing approaches that are compatible with established industrial processes such as blade coating,9,10,11,12 slit-casting,13 and spray-coating.14,15 The combination of excellent optoelectronic properties and ease of processing have made perovskite materials an extremely attractive field within photovoltaic research. Though perovskite solar cells possess many desirable properties, three major obstacles impede their pathway towards commercialization. First, the toxic heavy metal lead is an integral and not easily replaceable component of all high-performance perovskite solar cells. Second, current-voltage characteristics of perovskite solar cells frequently exhibit hysteresis,16,17,18,19 which leads to inaccuracy in determination of device efficiency, and inconsistent performance when operating under realistic power generation conditions due to large natural fluctuations of light intensity. This problem may be mitigated through careful design of the solar cell, as certain device architectures (e.g., planar cells using TiO2 as the electron transport layer, or ETL) often show a large propensity towards hysteresis, while it is virtually negligible in others (e.g., those using poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT:PSS) and phenyl-C61butyric acid methyl ester (PCBM) as hole and electron transport layers, respectively).17,20 Third,



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 4 of 27

the hybrid perovskites are quite unstable, particularly under moderate to highly humid atmospheres. Like hysteresis, instability of the perovskites can depend significantly on the other materials used to build the device.21,22,23 Hence, careful design and optimization of these interface materials is critical to the development of solar cells that are not only efficient, but also reliable under the full range of illumination, temperature, and atmospheric conditions entailed by commercial solar power generation. From a commercial perspective, it is also important that these interface materials be inexpensive and easy to process. The most common electron transport layer materials used in perovskite solar cells, TiO2 and PCBM, offer high performance and relatively low hysteresis, but entail significant fabrication challenges. The low-hysteresis TiO2-based ETL generally relies on a mesoporous nanoparticle framework, which must be calcined at high temperature2,4,5,24,25 in order to fuse the component particles and remove the organic binder. This difficult to control and energy-intensive process extends the energy payback time of devices fabricated using this architecture. PCBM, by contrast, is an organic semiconductor that is fully compatible with relatively gentle, low-temperature solution processing approaches. Unfortunately, it is also a fullerene-based material, and it is thus questionable whether it can be synthesized costeffectively on an industrial scale due to the high cost of these molecules. Other n-type oxides deposited by sol-gel processes, such as In2O326 and SnO227, have been used successfully in the planar architecture and do not require the high-temperature calcination step. However, these films still require a moderate anneal (180oC or higher), resulting in non-trivial energy input requirements. Thus, there is a need for an electron transport material that can provide not only high performance, but simple and economical processing.


Page 5 of 27

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


One promising material for this application is CdS. Despite concerns about toxicity, its superior electrical properties have allowed it to play a similar role in commercially successful thin-film photovoltaic technologies based on copper indium gallium selenide/sulfide (CIGS)28,29,30 and CdTe31,32, and it can be deposited via a rapid low-temperature chemical bath deposition process. CdS is fully compatible with planar solar cell architectures, further simplifying processing and reducing the interface area, which may reduce surface recombination. Despite its advantages, relatively little attention has been focused on its use in perovskite solar cells. An initial report indicated that, although device performance was quite low, higher Voc could be obtained when using CdS in place of TiO2 or ZnO.33 More recent attempts have seen efficiency increase beyond 16%, a value still far below the record efficiency of 22.1%.3,34,35,36,37 Furthermore, none of these reports demonstrate devices with stabilized efficiency, a more robust figure of merit for the efficiency than that obtained from the J-V curves, above 13%. In this work, we report the fabrication of CdS-based photovoltaic devices through careful optimization of the CdS layer thickness and processing conditions, and a champion cell exceeding 15% stabilized PCE. Through detailed study of the device structure, we found Cd on the surface of our perovskite films, indicating some slight migration of the electron transport material into the overlying perovskite film. To understand possible impacts of this Cd migration, perovskite films were fabricated in which varying amounts of CdI2 were used to partially replace the PbI2 precursor. Devices using these films showed a dramatic reduction in performance, along with significant distortion of the current density-voltage (J-V) curves, indicative of the formation of an electrical barrier. Previous experiments have shown that replacement of Pb with Cd should not introduce deep trap states into the perovskite band structure.38,39 Time-resolved photoluminescence measurements on our present films confirm that incorporation of Cd does not



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 6 of 27

reduce the carrier lifetimes, but scanning electron microscopy (SEM) / energy-dispersive X-ray (EDX) images and X-ray diffraction (XRD) patterns of the Cd-doped films indicate the presence of an insulating Cd-rich secondary phase, (CH3NH3)2PbI4. Based on these results, the barrier in Cd-doped perovskite device performance may be ascribed to the presence of this phase, which may form not only by direct incorporation of Cd in the perovskite precursor but also by reaction of the CdS substrate with the perovskite precursor methylammonium iodide (MAI). These results demonstrate that, although CdS is a promising ETL for use in high-performance perovskite solar cells, careful passivation and/or annealing strategies are required to prevent barrier formation in these devices.

Experimental Methods Perovskite Solar Cell Fabrication. Fluorine-doped tin oxide (FTO)-coated glass substrates (Sigma Aldrich, 8 Ω/□, 2.3 mm thick) were patterned using zinc powder and hydrochloric acid, rinsed in water, sonicated in Alconox, deionized water, acetone, and isopropanol for 10 minutes each, then baked on a hot plate in air for 1 hour at 540oC. Immediately prior to deposition of CdS, they were sonicated again in Alconox and deionized water for 10 minutes each. The CdS electron transport layer was deposited by immersing the cleaned FTO substrates in an aqueous solution containing 1.5 mM CdSO4 (99.99%, Sigma Aldrich), 0.94 M NH4OH (28-30 wt%, J. T. Baker), and 75 mM thiourea (99%, Sigma Aldrich), maintained at 72oC by a jacketed beaker, using a process analogous to that described for producing high-performance CIGS cells.28 Immediately after the CdS deposition, the substrates were sonicated in deionized water for 10 minutes, then plasma cleaned in O2 for 10 minutes. The CdS-coated substrates were then transferred to a nitrogen-filled glovebox (O2 and H2O

Effects of Cd Diffusion and Doping in High ... - ACS Publications

Recommend Documents