French physicist Edmond Becquerel discovered the photovoltaic (PV) effect in 1839. Attempts at commercialization did not begin until a century later; Bell Labs developed the first crystalline silicon PV device in 1954. Now silicon solar cells have achieved efficiencies of more than 24%, but their high cost and complicated technology limit further household applications. As such, researchers have begun to examine the potential of organic/polymer solar cells, which are competitive with inorganic PV technologies for several reasons. First, organic and polymer materials are inherently inexpensive and can have very high optical absorption coefficients (are highly effective at absorbing sunlight). Furthermore, these materials can be fabricated as flexible devices using well-established printing techniques in a roll-to-roll process. Organic/polymer solar cells with energy conversion efficiencies (n_e) of ~5% have been reported.¹ Since several companies and research institutions are focused on this field, n_e of 8–10% is highly likely in the near future.

Like other solar cells, the performance of organic/polymer devices depends on absorbing as much light as possible, converting the photon energy into free electrons, removing the electrons, and minimizing resistance. Figure 1 illustrates the typical configuration of a bulk heterojunction photovoltaic device. To absorb more sunlight, the band gap (E_g) of the polymer must be as low as possible without sacrificing electron transfer. At present, substituted poly(p-phenylene vinylene)s (PPVs) and polythiophenes (PThs) are typically used as donors in polymer photovoltaic devices. The optical E_g of these conjugated polymers (E_g=2.0–2.2eV) is not optimized with respect to the solar emission, which has a maximum photon flux around 1.8eV. Furthermore, organic electron acceptors—other than the fullerene derivatives and perylene diimides in common use—are also needed. Our group has addressed these issues by synthesizing both small band gap polymer donors as well as novel organic acceptors.

We synthesized the novel alternating conjugated copolymer PPV-BT (see Scheme 1) consisting of electron-rich dioctyloxyphenylene vinylene (PPV) and electron-deficient 2,1,3-benzothiadiazole (BT) units via a Pd-catalyzed Heck cross-coupling polycondensation. PPV-BT has a band gap of (E_g^EC=1.77eV, E_g^OPT=1.94eV), much closer to the sun's photon flux. Photovoltaic devices of indium tin oxide/ polyethylenedioxythiophene–poly(styrene sulfonic acid)/PPV-BT + [6,6]-phenyl C₆₀ butric acid methyl ester (ITO/PEDOT-PSS/PPV-BT+PCBM) (1/4, w/w)/Ba/Al were fabricated. At 451nm, these exhibited an n_e of 0.335% (see Figure 3) and external quantum efficiency (EQE) of 11.7%.

We also synthesized the copolymer PPE-BT, a poly(phenylene ethynylene) (PPE) derivative that has alternating triple bonds and an E_g^EC of 1.94eV (see Scheme 1). PPV-BT and PPE-BT have similar structures except that the double bonds of PPV-BT are triple bonds in PPE-BT. As shown in Figure 2, a device based on PPV-BT is more than an order of magnitude more efficient than a device based on PPE-BT. This is probably because the former has a greater ability to remove electrons and a broader optical absorption range than the latter.^2,3

Another small-bandgap polymer synthesized by our group is a poly(heteroarylene methines) derivative called PTBBQ (see Figure 3), which contains alternating aromatic and quinoid segments in the main chain. The E_g^OPT of PTBBQ is 1.88eV according to its absorption edge at 660nm. Figure 3 presents the I/V characteristics of a PTBBQ and PCBM device. The efficiency of the device is only 0.055%, lower than the MEH-PPV and PCBM devices, probably because the low molecular weight of PTBBQ resulted in rather poor film-forming properties.⁴

Similar to perylene diimides, perylenetetracarboxylate (see Scheme 2) has four electron-deficient ester groups, and the proximity of the p-orbitals of adjacent molecules indicates an ordered structure that makes it suitable for electron transport. We found that perylenetetracarboxylate can be used as an electron acceptor in organic photovoltaic devices. Table 1 presents the photovoltaic parameters of devices based on poly(3-hexylthophene) P3HT and perylenetetracarboxylates with different ester groups. The highest efficiency of a device based on tetra-methyl perylene-3,4,9,10-tetracarboxylate (TMEP) with P3HT is almost one order of magnitude higher than that of P3HT alone and six-fold higher than a diimide acceptor (PV) with P3HT composite device.⁵

Recently we synthesized another new perylenetetracarboxylate, tetra-benzyl perylene-3,4,9,10-tetracarboxylate (TBEP). We found that the annealing process resulted in the formation of a TBEP crystal network (see Figure 4). This property increases the EQE and n_e of photovoltaic devices based on TBEP and the poly[2-methoxy-5-(2'-ethyl-hexyloxy)-1,4-phenylene vinylene] (MEH-PPV) composite by a factor of five as compared to non-annealing devices.⁶

In conclusion, the state of the art in polymer PVs is only just beginning to achieve the minimum level of performance necessary for broad market entry. Reaching 8–10% efficiency will require better designed organic PV cell architectures, but new conjugated polymers with smaller band gaps and higher carrier mobility, as well as new organic acceptors, are also needed. These innovations will be the key to plastic solar cells becoming a viable energy source for the 21st century.