Efficient organic solar cells using polycrystalline pentacene films

Organic solar cells constitute the most recent class of photovoltaic devices for solar power conversion and are gaining a lot of attention due to their potential to be fabricated at low cost onto lightweight and flexible substrates. Recent studies have led to improvements in power conversion efficiencies with values exceeding 5 %. These developments can be considered an important step towards the demonstration of their practical viability.^1,2

In contrast to their inorganic counterparts like Si, organic semiconductors are often referred to as ‘excitonic’ semiconductors due to the large binding energy of excitons - bound electron-hole pairs - that are being formed in these materials³ following absorption of light. One important consequence is that these excitons are not likely to dissociate into uncorrelated electron-hole pairs by means of thermal processes at temperatures close to room temperature as they generally do in inorganic semiconductors. Instead, those excitons have to migrate to a donor-acceptor heterojunction with an energy offset that is large enough to drive their dissociation. Because of the limited distance over which these excitons can migrate without recombining before they reach the heterojunction, the thickness of multilayer-type organic solar cells is limited and typically of the order of the exciton diffusion length L. However, thin photoactive materials have limited absorption and do not harvest the sun light efficiently, limiting the overall power conversion efficiency. That interplay between efficient harvesting and efficient exciton dissociation can be overcome by creating bulk heterojunctions by mixing donor and acceptor molecules. However, this mixing generally reduces the charge mobility of holes and electrons within the mixed layers and can become a limiting factor. This bottleneck could be alleviated if films could be developed in which the product αL of the absorption coefficient α and exciton diffusion length L can be maximized. This is illustrated in Figure 1, which shows the calculated external quantum efficiency (EQE) as a function of a ratio of active layer thickness d to L for various values of the product αL. These calculations are based on a model that describes the steady-state concentration of excitons within a single photoactive donor layer and a very thin acceptor layer. The model describes generation, diffusion and recombination. For the generation, light absorption is described by Beer’s law and considers reflection on the back electrode. Optical interference effects are neglected. It is easily illustrated that the product αL is a figure of merit determining maximum EQE values.

While the exciton diffusion length can depend on many different factors, one way for its improvement is to increase the molecular ordering in the organic films. In this case, however, the challenge is to retain easy processibility and scalability. Solution-processible columnar discotic liquid crystals were proposed for such a purpose⁴, but it appears that more studies are needed to control their processing in device geometries. An alternative approach we followed⁵, is to use polycrystalline thin films prepared by vacuum sublimation. Such films were typically used in organic field-effect transistors, in which carrier mobility is among the most important parameters for its performance. Field-effect mobilities (1 cm²/Vs) competing with that of amorphous Si have been demonstrated in pentacene thin films having a high degree of crystalline order. In addition, pentacene is known to have good light absorbing properties in the red part of the visible spectrum, which makes it ideal to use with a common electron acceptor material like fullerenes having absorption dominantly in the blue part of the spectrum.

Figure 2(a) shows the device structure of the solar cells based on pentacene and C₆₀ heterojunctions with a schematic cartoon showing ordered arrangement of pentacene molecules. To minimize impurity-driven recombination of excitons and carriers, all the organic materials used are first purified by a three-zone thermal gradient sublimation technique in which same species of molecules condense in a well-defined zone and others condense in different zones. Then, these purified molecules are deposited successively on ITO substrates in vacuum by thermal evaporation using shadow masking. Al electrodes are then deposited as top electrodes. A bathocuproine (BCP) layer is used as a passivation layer to prevent excitons generated in the C₆₀ layer from being quenched at the organic/metal interface, and to protect the acceptor layer during metal deposition⁶. A typical device area is ∼0.1 cm². With a careful optimization of layer thicknesses and deposition conditions, we have achieved a peak external quantum efficiency as high as 69% at α = 668 nm (1.86 eV), which is among the highest at this wavelength range. [see Fig. 2 (b)]

Upon analysis of the EQE spectrum using the exciton diffusion model involving a full optical analysis⁶, ⁷ and the characterization by ellipsometry of the optical properties of the materials that compose the cell, one can show that the exciton diffusion length in the pentacene film is as large as ∼ 60 nm, leading to a peak (αL) product to be ∼ 0.5, a value that is significantly higher compared to other organic films commonly used for solar cells. Power conversion efficiencies of these optimized cells are currently ∼ 3.6 % when illuminated with a solar simulator based on a filtered Xe lamp. Based on the measured EQE, an efficiency of approximately 2 % can be predicted for the true AM 1.5G illumination condition.

These results indicate that the use of highly ordered organic polycrystalline thin films is a promising approach to the development of future organic solar cells since they can be conducive to large exciton diffusion length and thus a high external quantum efficiency. Further improvements can be expected in multi-junction type geometries¹ utilizing combinations of materials with optimized spectral properties. However, before these new materials reach their true commercial potential, scalability to larger areas and long-term stability have to be demonstrated. For manufacturing on flexible substrates, breakthroughs in packaging technologies will be required.

This work was supported in part by the STC Program of the National Science Foundation under Agreement Number DMR-0120967, by the Office of Naval Research, and by an NSF CAREER program (B.K.).