Lead Halide Perovskites as Charge Generation Layers for Electron

Oct 30, 2017 - Hybrid lead halide perovskites are introduced as charge generation layers (CGLs) for the accurate determination of electron mobilities ...
0 downloads 13 Views 2MB Size
Research Article Cite This: ACS Appl. Mater. Interfaces XXXX, XXX, XXX-XXX

www.acsami.org

Lead Halide Perovskites as Charge Generation Layers for Electron Mobility Measurement in Organic Semiconductors John A. Love,† Markus Feuerstein,† Christian M. Wolff,† Antonio Facchetti,‡ and Dieter Neher*,† †

Institute for Physics and Astronomy, University of Potsdam, Karl-Liebknecht-Straße 24−25, Potsdam-Golm 14476, Germany Department of Chemistry and The Materials Research Center, Northwestern University, 2145 Sheridan Road, Evanston, Illinois 60208, United States



S Supporting Information *

ABSTRACT: Hybrid lead halide perovskites are introduced as charge generation layers (CGLs) for the accurate determination of electron mobilities in thin organic semiconductors. Such hybrid perovskites have become a widely studied photovoltaic material in their own right, for their high efficiencies, ease of processing from solution, strong absorption, and efficient photogeneration of charge. Time-of-flight (ToF) measurements on bilayer samples consisting of the perovskite CGL and an organic semiconductor layer of different thickness are shown to be determined by the carrier motion through the organic material, consistent with the much higher charge carrier mobility in the perovskite. Together with the efficient photon-to-electron conversion in the perovskite, this high mobility imbalance enables electron-only mobility measurement on relatively thin application-relevant organic films, which would not be possible with traditional ToF measurements. This architecture enables electron-selective mobility measurements in single components as well as bulk-heterojunction films as demonstrated in the prototypical polymer/fullerene blends. To further demonstrate the potential of this approach, electron mobilities were measured as a function of electric field and temperature in an only 127 nm thick layer of a prototypical electron-transporting perylene diimide-based polymer, and found to be consistent with an exponential trap distribution of ca. 60 meV. Our study furthermore highlights the importance of high mobility charge transporting layers when designing perovskite solar cells. KEYWORDS: mobility, bulk heterojunction, time of flight, lead halide perovskites, charge generation layers



parameters are experimentally linked.3 Furthermore, mobilities from steady-state electrical measurements might be inappropriate for describing the dynamics of photogenerated charges, particularly in thin layers where charge carriers might exit before thermalizing completely. On the other hand, transient photoconductivity measurements in full working devices can give information about charge transport and recombination,11,12 but the extracted mobility is nonselective with respect to the carrier type, and charge recombination during extraction might falsify the obtained mobility. One of the oldest, most conceptually straightforward mobility measurements employed in disordered semiconductors is the time-of-flight (ToF) technique.13−15 In this methodology, an organic layer is sandwiched between two electrodes, of which at least one is semitransparent. Charges are then photoexcited with a short laser pulse and drift based on the polarity of the internal electric field; the one-dimensional transport of charge to the electrodes is monitored as a function of time.16 In the limit of a thick layer, a sheet of charges is formed selectively near the illuminated electrode and the transit

INTRODUCTION Principal to the operation of organic optoelectronic devices such as bulk heterojunction (BHJ) solar cells and light-emitting diodes is the transport of charges through organic semiconductor (OSC) layers, typically described as a hopping mechanism between localized sites within a broad density of electronic states.1−4 The transport is most often characterized by charge mobility or the speed at which charge carriers drift in response to an electric field. This parameter can depend on field, charge density, and temperature.2 Both experimentally and theoretically, it has been shown that high charge carrier mobilities are essential to achieve efficient charge extraction in high-performance organic solar cells.5−10 Despite the fact that several methodologies have addressed the characterization of charge mobilities within an organic layer, significant challenges remain because each measurement technique has advantages but also limitations when related to solar cell operation. For instance, space charge limited current (SCLC) measurements are well-established for measuring steady-state charge mobilities and are selective for a single carrier type by using appropriate electrical contacts. However, fabricating devices with Ohmic contacts for electron mobility measurements can be very challenging, and discerning between electric field and charge density dependences to mobility is difficult as the two © XXXX American Chemical Society

Received: July 17, 2017 Accepted: October 30, 2017 Published: October 30, 2017 A

DOI: 10.1021/acsami.7b10361 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 1. (a) Schematic diagram of the employed ToF architecture with a perovskite charge generation layer (b) corresponding cross-section SEM image, and (c) energy scheme showing principles of operation.

from solution,25,26 and strong absorption.27,28 Importantly, for some of these perovskites, very high mobilities have been recorded, while at the same time, the low exciton binding energy allows for efficient free charge generation.29−31 Also, the conduction and valence bands of many of these perovskites match well with the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) levels of OSCs commonly used in organic devices.32−34 It is well-established that electrons can efficiently be injected into a fullerene or even a BHJ layer35 acting as an electron transport layer to a top electrode; these properties make them ideal candidates as CGLs for ToF electron mobility measurements. Moreover, their broad, strong absorption makes them compatible with a range of OSCs and excitation sources and thus provide advantages over many previous CGLs. Here, as a proof of concept, we have demonstrated electron mobility measurements in a few systems; however, this technique should be extendable to a very broad range of materials systems relevant to organic electronic devices.

time for charges to be dragged through the entire layer and arrive at the counter electrode, τtr, is described by the charge carrier mobility, μ, according to the equation τtr =

d μF

(1)

Here, d is the thickness of the semiconducting layer, and F is the internal electric field created by the superposition of the built-in voltage Vbi and the applied bias Vappl through F = (Vbi + Vappl)/d.17 Despite its seeming simplicity, analysis of the transient relies on the assumption that one charge carrier type is removed significantly faster than the other, and that the slower carrier type begins as a single sheet at the illuminated electrode. Practically, the charge generation profile must be such that the vast majority of charges are formed in only a small fraction of the layer so that one charge type can be quickly removed while the other must traverse through the thickness of the layer. For organic solar cell materials with a typical light penetration depth of around 100−200 nm, this means that the blend films must be microns thick and have appropriate mobilities such that the charges arrive at distinctly different times; these conditions drastically limit the applicability of the ToF technique. One way to circumvent these drawbacks is to insert a charge generation layer (CGL) such that that light is absorbed and charges are formed exclusively in the CGL, and then transport is measured for one type of carrier moving through the material of interest, which is transparent to the incoming radiation. This facilitates a relatively abrupt sheet of charges, improving the technique greatly. In the past, a number of organic dyes, such as perylene derivatives18,19 and pthalocyanine complexes,20,21 as well as organic solar cell blends22,23 among other examples, have effectively been employed as CGLs in combination with OSCs. The CGL must be thin such that the transport of charge proceeds via “hot” thermally excited states. Using a proper combination of CGLs and semiconductor, one can indeed selectively excite the dye, inject charges, and measure mobility. However, the absorption and excitation wavelength of the CGL must lie outside the absorption spectrum of the semiconductor of interest, and the energy levels of the CGL must be appropriate to effectively inject charges into the semiconductor or blend. This limits the applicability of any one particular dye for measuring a large variety of materials or blends. Alternatively, inorganic layers such as silicon24 or selenium21 have been used as CGLs in combination with organic materials, but both have disadvantages with regard to absorption properties and electronic level alignment. Here, we introduce organometallic lead halide perovskites as CGLs for ToF electron mobility measurements. Such hybrid perovskites have become a widely studied photovoltaic material in their own right, for their high efficiencies, ease of processing



EXPERIMENTAL SECTION

Device Fabrication. ToF samples were prepared atop cleaned indium tin oxide (ITO)-patterned glass substrates. After deposition of a thin (40 nm) polyethylenedioxythiophene/polystyrene-sulfonate layer, the lead halide perovskite (MAPbI3) layer was deposited using the single solution antisolvent method described previously.36 Specifically, a dimethylformamide/dimethylsulfoxide solution (4/1, v/v) containing PbI2 (1.5 M) and MAI (methylammonium iodide ,1.5 M) was spun in an N2 atmosphere at 4000 RPMs for 35 s. Approximately 10 s after spinning was initiated, while the layer was still drying, 500 μL of diethyl ether (antisolvent) was dispensed onto the spinning film to form a brown perovskite layer. Spincasting was followed by successive annealing of the sample of 1 min at 80 °C and 2 min at 100 °C. The organic layers were then cast on top according to literature conditions (typically from chlorobenzene or 1,2-orthodichlorobenzene) to obtain layers relevant to solar cells. Electrodes of 8 nm bathocuproine and 100 nm copper were deposited via thermal evaporation at pressures 20 ns. Implicitly, this demonstrates that the charge kinetics in high-efficiency perovskite cells, including charge generation, transfer, and transport, must all occur on the