The difference between atoms and molecules as basic building blocks calls for a very different design of morphology and layout of devices in which these building blocks are used.
The design of molecular devices, in which two (or more) components function interactively, e.g. as an electron donor-acceptor pair, is discussed in terms of the principle of interpenetrating networks leading to a bulk-heterojunction (beta-junction). One of the most intriguing applications of the beta-junction is in molecular PV cells. beta- Junction PV cells show a much higher efficiency than molecular n/p bilayer cells, because in beta-junction cells the whole blend layer is PV active, while in bilayer cells usually only a small area around the n/p interface is active. In the beta-junction, the surface of the n/p interface is enhanced dramatically. It has been shown that spontaneous phase segregation occurs in blends of two conducting polymers and in blends of MEH-PPV with C₆₀, leading to a particle size of ~ 10-20 nm.^1,2 Hence, with regard to the morphology of the PV active layer being chaotic and on the nano-scale, the present beta-junction cells somewhat resemble the nanocrystalline TiO2 /dye cells.
The present status of the plastic beta-junction PV cell³ will be reported briefly.⁴ This virtually all plastic cell, based on the photo- induced donor-acceptor interaction of a conjugated polymer as the dye & donor and a fullerene derivative as the acceptor already shows a remarkable efficiency taking in account the lack of true order in the currently used two component blend (CPC blend). A layout of the cell is shown in Figure I.

Substrate
TCO
CPC blend
Al

From the fact that the fluorescence of MEH-PPV is quenched in the blends with C₆₀ or some of its derivatives by a factor up to ~10³, we conclude that exciton transport is not a limiting factor in the above-mentioned cell. The best cells were made using an active layer of ~100 nm thickness. Devices made with thicker beta-junction layers show relatively much lower I_SC values. This, together with the fact that the cells show a high series resistance, can be best explained by the charge carrier mobilities being low. Together with other research groups in The Netherlands and Europe, we have started a research program in order to make beta-junction PV cells with increased charge carrier mobility. One important goal is to obtain more (supra) molecular order in the beta-junction. A second goal is to obtain control over the scale of the phase segregation. A third goal is to lower the percolation threshold of the fullerene component. In principle, many different types of architectures for beta-junctions can be designed. No studies on theoretical calculations with respect to such architectures have been reported, yet. Introduction of a higher degree of order in the beta-junction will result in higher charge carrier mobility in different ways. First, molecular order inside each phase (each network) will increase the mobility on the donor-acceptor (D-A) unit s scale (the blend particle scale). Second, ordering the networks in such a way that the path for the charge carriers is more or less straight and in line with the electric field, provided by the two electrodes, is expected to increase the mobilities significantly on the scale of the film thickness (100-500 nm). Control over the scale of the phase segregation can be obtained in several manners. One way is through understanding the molecular and engineering factors governing the miscibility of the two (or more) components of the blend. Another way is to construct (supra) molecular building blocks that determine the size of the D-A units. The smallest D-A units are molecular diads (one D and one A). Aligning such diads in a simple linear way, one can envision a linear structure like in Figure 2a-d, depending on the method of construction. Hence, the increasingly larger D-D-A-A, D-D-D-A-A-A, .., (D)_n- (A)_n building blocks, can lead to aligned structures like in Figure 2e and 2f (etc).

. D D D D D D D . ....-D-D-D-D-D-D-D- . D D D D D D D .D D D D D D D
....-A-A-A-A-A-A-A- . D D D D D D D .. .D D D D D D D
.A A A A A A A . A A A A A A A .. .D D D D D D D
a. b. ....A A A A A A A .A A A A A A A .A A A A A A A
D D D D D D D .D-D-D-D-D-D-D .A A A A A A A
.A-A-A-A-A-A-A A A A A A A A
c. d. e. f.

In changing from smaller to larger building blocks, the size of the (supra) molecular networks wires changes in such a way that the electronic properties of the wires are expected to change more and more from molecular to solid phase material in nature.
The percolation threshold of the fullerene component can be lowered in principle by aligning the C₆₀ moieties into one-dimensional poly-structures. In the case of a conjugated polymer / fullerene derivative beta-junction, in which the polymer functions as the dye and donor, we regard asymmetric D-A units, i.e. (D)_n-A with n>1, as very promising low C₆₀ content candidates to serve as building blocks for thin film plastic PV devices.
Focussing on the fullerene part of the donor-acceptor blend, the first synthetic steps as well as future directions for the realization of a (supra) molecular or mesoscopic architecture of the active layer of plastic PV are presented.

[1]	J.J.M. Halls, C.A. Walsh, N.C. Greenham, E.A. Marseglia, R.H. Friend, S.C. Moratti, A.B. Holmes, Nature 376, 498 (1995).
[2]	C.Y. Yang, F. Hide, A.J. Heeger, Y. Cao, Synthetic Metals 84, 895 (1997).
[3]	G. Yu, J. Gao, J.C. Hummelen, F. Wudl, and A.J. Heeger, Science 270, 1789-91 (1995).
[4]	For a detailed report on the present status of the CPC beta-junction cell, see the contribution of N.S. Sariciftci.

[BACK] [MAIN] [FURTHER]