Inverted polymer solar cells generate electricity without UV light

Polymer solar cells have attracted considerable attention because they are low-cost, lightweight, flexible and environmentally friendly. New organic precursors have pushed the power conversion efficiency (PCE) of polymer solar cells to almost 10%. The device structures have two groups: conventional and inverted types. The active layer of the conventional structure is generally sandwiched between transparent indium tin oxide (ITO) and aluminum (Al) electrodes. The Al electrode collects electrons because of its low work function, but this also causes it to be easily oxidized, making it sensitive to oxygen and humidity. The only solution to this problem has been to seal it from the atmosphere under an ultrahigh vacuum.

However, in an inverted arrangement, the organic active layer can be sandwiched between an ITO electrode coated with a chemically stable n-type metal oxide and an electrode made of a non-corrosive metal such as gold or silver. The photo-generated electrons in the active layer are, in this arrangement, collected by the modified ITO, while the photo-generated holes are collected by the high-work function metal. These polymer solar cells are described as inverted because the electrons flow in the opposite direction to those in conventional solar cells. This structure has the advantage of long-term stability, because the electrode compositions are robust to both oxygen and humidity.

Figure 1. Arrangement of our inverted bulk-heterojunction polymer solar cell, with structures of monomers used. PCBM: [6,6]-phenyl C₆₁ butyric acid methyl ester. P3HT: regioregular poly(3-hexylthiophene). PEDOT: poly(3,4-ethylenedioxylenethiophene). PSS: poly(4-styrene sulfonic acid).

We previously reported electron collection layers such as titanium oxide (TiOx and TiO2),^1–3 zinc oxide (ZnO),^{4, 5} and zinc sulfide⁶ for inverted polymer solar cells. More recently, we have examined the relationship between the ZnO layer preparation temperature and device performance in simulated sunlight while filtering out UV light.⁷

Figure 1 shows the inverted organic bulk-heterojunction solar cells we made, with the structures of the monomers used. We made the ZnO electron collection layer by spin-coating ZnO precursor films and subsequently heating them to one of 150° C, 250° C, 350° C, or 450° C for 1h in an electric furnace. We then analyzed the ZnO films using x-ray diffraction and observed three diffraction peaks at 2θ = 31.7°, 34.4°, and 36.2°. We interpreted these as being orientations along the (100), (002) and (101) planes, respectively. Heating the films to the higher temperatures caused both the sharpness and intensity of the diffraction peaks to increase, indicating greater crystallinity within the ZnO films.

We subsequently measured photo I-V curves, under full-spectrum simulated sunlight, for inverted solar cells containing ZnO layers from precursor films heated to each of the different temperatures. We found that the maximum PCE achieved for each sample was 3.08–3.37%, irrespective of the ZnO heating temperature. This suggests that, in full-spectrum sunlight, increased crystallinity does not affect the electron collection efficiency of the ZnO layers.

Figure 2. Illustration of the electron transfer process between PCBM: P3HT and ZnO layers irradiated with both UV-filtered and unfiltered simulated sunlight. ZnO films were prepared by heating at (a) 250°C, (b) 350°C, and 450°C, respectively. Carriers: (red dots) electrons photo-generated from PCBM:P3HT; (blue dots) electrons photo-generated from ZnO; (red rings) holes photo-generated from PCBM:P3HT; and (blue rings) holes photo-generated from ZnO. The solid and broken arrows show the major and minor carrier motion directions, respectively. C.B., V.B., and ET are the lower end of a conduction band, the upper end of a valence band, and the recombination center formed in a forbidden band, respectively.

However, when we inserted a UV-cut filter to eliminate light at wavelengths <440nm, the cells exhibited a PCE that varied according to film preparation temperature. Cells containing ZnO layers from films heated to 250° C, for example, yield a PCE of 2.7%. However, for the cells with ZnO layers heated to 350° C and 450° C, the shapes of the I-V curves changed with the duration of irradiation, accompanying an increase in their series resistance. Overall, after irradiation with UV-filtered light for 1h, the PCE of the cell containing a ZnO layer prepared by heating to 350° C decreased to 1.80%, while that of the cell with ZnO prepared by heating to 450° C fell to 1.35%. Alternating current impedance spectroscopy investigations showed that this performance change was due to the formation of charge-trapping sites at the ZnO/PCBM:P3HT interface, which act as recombination centers for photo-produced charges in the PCBM:P3HT layer: see Figure 2. This suggests that recombination centers were formed at the ZnO/PCBM:P3HT interface by ZnO micro-crystallization.

To summarize, heating ZnO films to higher temperatures increases their crystallinity. When such films are used as electron collection layers in inverted polymer solar cells, the power conversion efficiency of the cell decreases. In contrast, in cells containing less crystalline ZnO layers prepared at lower temperatures, electron transfer proceeds more readily, even in the absence of UV light. Our investigations indicate that this improvement is due to surface sites, which act as charge recombination centers, not being formed in lower-crystallinity ZnO. We conclude that preparing ZnO films at lower temperatures is a key condition for the fabrication of inverted polymer solar cells that generate electricity efficiently without UV light. We next plan to make bifacial polymer solar cells using this approach.