See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/233709363
Metal oxide applications in organic-based photovoltaics Article in Materials Science and Technology · September 2011 DOI: 10.1179/026708311X13081465539809
CITATIONS
READS
26
501
1 author: Talia S. Gershon IBM 39 PUBLICATIONS 807 CITATIONS SEE PROFILE
All content following this page was uploaded by Talia S. Gershon on 21 May 2014. The user has requested enhancement of the downloaded file.
LITERATURE REVIEW
Metal oxide applications in organic-based photovoltaics T. Gershon* Organic-based photovoltaics (PV) have attracted increasing attention in recent years and efficiencies exceeding 8% have recently been confirmed. These low cost, lightweight and mechanically flexible devices offer unique advantages and opportunities currently unavailable with crystalline silicon technology. Progress in the field of organic PV has been achieved in part due to the incorporation of transition metal oxides. These offer a wide range of optical and electronic properties, making them applicable in organic-based PV in many capacities. Transparent electrodes can be made from doped metal oxides. The high intrinsic charge carrier mobility of many undoped metal oxides makes them attractive as active materials and charge collectors. Metal oxides can increase the charge selectivity of the electrodes due to the energetic positioning of their valence and conduction bands. Thin films of these materials can manipulate the light distribution inside of organic devices, allowing for improved light harvesting. Metal oxides are stable and can be processed at low temperatures. Consequently, they have been demonstrated as suitable intermediate layer materials in tandem cells. Finally, oxygen-deficient metal oxides can improve the stability of the oxygensensitive organic semiconductors. The present work reviews the various applications of metal oxide layers in organic PV devices and summarises the challenges associated with organic/ oxide interfaces. Keywords: Metal oxide, Organic photovoltaics, Charge blocking layer, Charge collector, Optical spacer, Tandem cell, Stability, Review
This review was the winning review of the 2010 Materials Literature Review Prize of the Institute of Materials, Minerals and Mining, which is administered by the Editorial Board of Materials Science and Technology
Introduction Hybrid photovoltaics (PV) motivation The growing interest in renewable energy of recent decades has rapidly accelerated research into PV technology. Crystalline silicon, one of the most heavily researched PV materials and the current market leader, has yielded power conversion efficiencies of over 20%.1 For large scale power production, however, the energy required to process crystalline silicon makes these devices too costly to compete with inexpensive fossil fuel based electricity,2,3 illustrating the need for low cost PV alternatives. Inexpensive PV devices are attractive not only for large scale power production but also small scale, portable and remote technological applications. A promising and rapidly developing low cost PV system is based on organic semiconducting polymers. These can be dissolved and coated onto many different surfaces via low temperature techniques such as roll-toroll processing.1 These soft materials give way to lightweight and flexible devices, which are currently Department of Materials Science and Metallurgy, University of Cambridge, Pembroke Street, Cambridge CB2 3QZ, UK *Corresponding author, email
[email protected]
ß 2011 Institute of Materials, Minerals and Mining Published by Maney on behalf of the Institute Received 4 May 2010; accepted 9 May 2011 DOI 10.1179/026708311X13081465539809
impossible from crystalline silicon. Organic semiconductors are strong light absorbers and can capture most incident light within 100–200 nm. Their optoelectronic properties can be manipulated by changing their molecular chemistries, allowing for the development of a highly diverse class of materials. Efficiencies exceeding 8% have recently been confirmed and numerous companies have emerged with the aim of commercialising this technology.4–6 Transition metal oxides are another attractive class of semiconducting materials. These have excellent charge transport properties and can be tuned in various ways through the introduction of dopants, the generation of nanostructures, or modification of their surfaces. Owing to the wide range of properties that these offer both optically and electronically, transition metal oxides can play many different roles within a ‘hybrid’ organic/ inorganic PV device. Most of the reviews that have been published on hybrid PV to date have focused on the implementation of metal oxides with polymers as part of the active layer of the device (i.e. the layer responsible for the generation of charge carriers).7–9 The present work will briefly discuss metal oxides as active layer materials but will focus mainly on the other key applications of these materials in organic-based PV.
Materials Science and Technology
2011
VOL
27
NO
9
1357
Gershon
Metal oxide applications in organic-based photovoltaics
1 Schematic of charge photogeneration and transport in excitonic PV. The transition between the polymer’s highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) represents the polymer’s ‘band gap’ energy (Eg). Incident light (1) is absorbed by the polymer and an exciton is generated (2). The exciton diffuses inside the material (3). If it reaches an interface with an appropriate acceptor, the exciton dissociates (4). Holes are transported through the p-type material and electrons through the n-type material (5)
It should be noted that one specific metal oxide application has been intentionally excluded from this review. ‘Dye-sensitised’ solar cells utilise dye-coated metal oxide nanoparticles and oxidation/reduction reactions. The breadth of this topic made its inclusion in the present work impossible and the reader is therefore referred to a review by Gra¨tzel.10
Excitonic PV Unlike crystalline inorganic semiconductors, organic semiconducting polymers do not have continuous bands of energy levels, and thus no ‘valence’ and ‘conduction’ bands. Instead, these molecules contain a series of filled and unfilled electron orbitals. Delocalised p molecular orbitals are formed when atomic p orbitals in a polymer’s backbone overlap. The resulting ‘highest occupied molecular orbital’, or HOMO, and the ‘lowest unoccupied molecular orbital’, or LUMO, determine the optical and electronic properties of the macromolecule, as discussed in the review by Kippelen and Bre´das.11 The energy of transition between these two is often referred to as the polymer’s ‘band gap’ energy (Eg) because of its similarity to that of inorganic semiconductors. Like inorganic solar cells, contemporary organic PV operate via the implementation of a quasi-p–n junction. Some organic materials naturally conduct positive charges (‘p-type’) and others negative charges (‘n-type’). Unlike the case in inorganic PV, however, the electron–hole pairs, known as excitons, generated in organic semiconductors are not as mobile at room temperature.12 These tightly bound excitons must overcome an energy barrier of approximately 0?3–0?4 eV before independent charge carrier motion can take place.1 Therefore, there must be a sufficient mismatch between the energy levels of the donor (p-type) and acceptor (n-type) materials to induce exciton dissociation, or separation of the electron from the hole.12 This mismatch makes it energetically preferable for the exciton’s electron to occupy the material with the lower LUMO level while the hole preferentially occupies the material with the higher HOMO level. A schematic of how independent charges are generated and transported in organic materials is shown in Fig. 1. A more comprehensive explanation of organic PV device physics can be found in other reviews.11,13 Unless otherwise noted, it should be assumed that the organic polymers are p-type and act as the light
1358
Materials Science and Technology
2011
VOL
27
NO
9
absorbers, as this is true of the majority of hybrid PV devices. The most successful and widely implemented p-type polymers include poly(3-hexylthiophene) (P3HT), poly(cyclopenta-dithiophenebenzothiadiazole) (PCPD TBT) and derivatives of poly(phenylene-vinylene) (PPV).14–16 In organic PV devices, the n-type acceptor material is traditionally fullerene-based and is commonly [6,6]-phenyl-C61 butyric acid methyl ester (PCBM) or similar molecules with larger cage structures (e.g. PC70BM). In some hybrid devices, the acceptor is an n-type inorganic material. The highest-efficiency solar cells certified in industry contain proprietary polymers whose structures are not commonly known.4 In the remainder of the present work, the terms ‘oxide’, ‘metal oxide’ and ‘transition metal oxide’ will be used interchangeably. It should be assumed, unless otherwise noted, that notations like ZnO and TiO2 represent pure metal oxide materials. In general, these materials contain an equilibrium concentration of vacancies and thus are not stoichiometric. The deviation from stoichiometry depends on the oxide’s processing history and has a significant impact on the material’s properties. This fact is often overlooked and stoichiometries are rarely reported. References in the present work to ‘ZnO’ and ‘TiO2’ are made without providing additional stoichiometric information because this information is unavailable. The reader should remember that metal oxide processing influences fundamental material properties including defect densities and surface states. Finally, the reported devices were tested under standard conditions unless otherwise noted. This includes exposure to a standardised solar spectrum, known as AM 1?5G, with an irradiance intensity of 100 mW cm22. Power conversion efficiency (g) is calculated from equation (1), where Voc is the open-circuit voltage, Jsc is the short-circuit current density and FF is the fill factor, which is a unitless measurement, out of 1, that represents the diode quality. An index of the different devices described in the present work, along with their performance characteristics, can be found at the end of the work in Table 2
g~
Pout Pincident
~
Voc |Jsc |FF Pincident
(1)
Gershon
Roles of metal oxide materials Metal oxides can play a number of different roles in organic-based PV. The ones addressed further in this review are introduced here: Active materials. The active materials are responsible for the absorption of light and photogeneration of charge carriers as illustrated in Fig. 1. Light is absorbed by the polymer and excitons are dissociated at the interfaces with the metal oxide. In the absence of an exciton dissociation interface (the oxide), free carriers are not generated. Holes are then transported through the polymer, and electrons through the oxide, until they are collected at the electrodes. Some common hybrid active layer geometries are shown in Fig. 2. Transparent electrodes. Organic semiconductors have relatively low conductivities and thus photogenerated charge carriers can only diffuse several hundred nanometres before they recombine. Transparent electrodes are thus needed, which can collect charges over the entire device area while still allowing incident light to reach the active layer. Wide band gap transition metal oxides are suitable materials for this application due to their excellent optical properties and the ability to tune their electrical properties through doping. Charge blocking layers. One major source of inefficiency in organic PV stems from ‘dark current’, or ‘leakage current’. Charges injected at the electrodes recombine with photogenerated charges, thus reducing device performance. Organic PV suffers from this problem in particular because of the blended nature of the active layer; both p- and n-type materials directly contact both electrodes. One way to reduce leakage current is to insert metal oxide ‘blocking’ layers on either side of the active layer. Owing to the differences in energy level positioning, one layer favours the transport of holes and the other favours the transport of electrons (Fig. 3). This improves electrode selectivity and reduces the dark current. Charge collectors. Some metal oxides have relatively high bulk electron mobilities (hundreds of cm2 V21 s21, 3–4 orders of magnitude higher than that of many organics). As such, they can act as charge collectors in traditional organic PV devices. Organic or hybrid blends are intercalated into an array of oxide nanostructures (Fig. 4), where the high-mobility nanowires extend into the active layer and conduct photogenerated charge carriers to an electrode. Optical spacers. As light travels through a device, it passes through multiple films with different optical properties (i.e. glass slide, transparent electrode, electron/hole blocking layer, active layer, etc.) and can experience reflections at each interface. When the light reaches the electrode, it is reflected back through all of the layers. Based on the thicknesses of these films and their refractive indices, the intensity of each wavelength of light is maximised in different regions of the device due to constructive and destructive interference. Optical spacers can be used to ensure that the wavelengths of light most efficiently absorbed by the polymer have their highest intensities in the active region (Fig. 5). Intermediate layers in tandem cells. The highestefficiency inorganic PV devices to date contain multiple layers, with different band gaps, stacked in a tandem architecture (e.g. GaInP/GaInAs/Ge).18 To accomplish
Metal oxide applications in organic-based photovoltaics
this in an organic-based device, an intermediate layer must be present for several reasons. First, this layer protects the bottom cell and prevents it from dissolving upon deposition of subsequent cells. Second, it makes ohmic contact between the different cells. Third, it allows unwanted electrons from one layer and unwanted holes from another to recombine; otherwise these recombine with the charges intended for extraction. This intermediate layer can be comprised of a few nanometres of a metal oxide (Fig. 6). Stability enhancers. Semiconducting polymers are generally unstable in the presence of light and oxygen. It has been shown, however, that oxygen-deficient oxides like TiOx (x
