Info

2000 4000 6000 BOOO

Wave number (cm1)

in the orthorhombic phase remains independent of temperature down to 34 K, where the transmission again increases, indicative of another phase transition, as is also suggested by EPR measurements [13.162], A fit to the IR transmission spectrum for the slow-cooled sample in Fig. 13.33 suggests a dc conductivity of ~10 mil-cm for that phase. When the rapidly quenched state is prepared by rapid cooling from 225°C to room temperature, the optical transmission spectrum is similar to that shown for the — 25° C slow-cooled spectrum in Fig. 13.33. However, rapid quenching of the RbjC60 film from 225°C to lower temperatures ('—25°C) [13.161] produces a phase with higher optical transmission at low frequencies, corresponding to a more insulating phase associated with a lower carrier density (see Fig. 13.32). In this phase two adjacent C60 anions may become linked in a shortened intermolecular distance, which has been identified with dimer formation in structural studies [13.162]. However, since the dimers are formed in random directions, it is not believed that the rapidly quenched dimer phase is a precursor for the chain-like structures of the slow-cooled more conductive phase [13.160]. More detailed analysis of the transmission spectra for the K, Cft0 samples that were rapidly quenched to —50°C indicate a semiconducting phase with an energy gap of ~5000 cm-1 (0.63 eV) in the electronic system [13.160,161]. The cutoff in transmission seen above 8000 cm"1 for all the transmission spectra in Fig. 13.33 is attributed to the tXu -> tXg interband transition, also seen in other alkali metal-doped fullerides at about 1 eV [13.143,144],

13.5. Optical Properties of C60-Polymer Composites

The first report of very high photoconductivity from a fullerene-polymer composite was made by Ying Wang at the duPont Research Laboratories [13.163]. This discovery may lead someday to applications for fullerenes in xerography, photovoltaic cells, and photorefractive devices as discussed in §20.1.2, §20.1.3, §20.2.3, and §20.2.4. Wang's work was followed by a series of papers from researchers at University of California (UC) Santa Barbara [13.164—170] who explored the underlying physical mechanism behind the enhanced photoconductivity, which led them to investigate the enhanced photoconductivity phenomenon in a variety of C60-doped polymers. Concurrently, optical absorption and luminescence studies on C60-doped poly-hexylthiophene (PHT) were reported by Morita and co-workers [13.171], who showed that PHT can be doped with acceptor molecular dopants to produce a conducting polymer [13.172].

Wang's discovery was made in polyvinylcarbazole (PVK) films doped with a few (~2.7) weight % of fullerenes C60 and C70 in the ratio 85:15 [13.163]. The PVK and fullerenes were both dissolved in toluene, and the solution was then spin-coated onto an aluminum substrate to form a thin solid film after subsequent solvent evaporation. The resulting films, 1-28 jtim thick, were found to be air-stable and optically clear. Photoconductivity in these films was studied by a photoinduced discharge method in which the surface of the film is corona-charged, either positively or negatively. This surface charge induces a voltage difference between the front surface of the film and the substrate. When the film is exposed to light, photogenerated electron-hole pairs are produced in the film. If the exciton (electron-hole) binding energy is sufficiently small, mobile electrons or holes, depending on the sign of the surface charge, drift to the surface of the film, thereby neutralizing the surface charge and reducing the voltage across the film. The rate at which the film voltage V decays to zero is a measure of the charge generation efficiency </> of the film

where e and L are, respectively, the dielectric constant and thickness of the film, e is the electronic charge, I is the incident photon flux, and the initial discharge rate of the surface potential dV/dt is calculated at the moment when the light is first incident on the film. A discharge process of this type is the basis for xerography [13.163].

In Fig. 13.34, we display the photoinduced discharge curves for a pure PVK film (upper curve) and a PVK film doped with a few weight percent of fullerenes (lower curve) [13.163]. Both samples were exposed to broadband light from a tungsten lamp beginning at time t = 0. From the data in

Fig. 13.34. A qualitative comparison of the photoinduced discharge curves for pure PVK and fullerene-doped PVK under the same experimental conditions. A tungsten lamp (50 mW/cm2) is used as the light source [13.163].

Fig. 13.34 at times t < 0, the finite slope in dV/dt is due to the dark conductivity of the respective films, and it can be seen that the fullerene-doped PVK film has a low dark conductivity, comparable to that of the pristine PVK. However, under illumination (f > 0) the two films behave very differently. As shown in the figure, the discharge rate of the surface potential for the fullerene-doped PVK is much larger than that for the pristine PVK; i.e., the charge generation efficiency $ for PVK has been greatly enhanced by the addition of a few (e.g., 2.7%) weight percent of fullerenes.

To elucidate the mechanism for enhanced photoconductivity, Wang studied the wavelength (A) dependence of the charge generation efficiency (j> and observed that the long-wavelength threshold for cf> was very near to the threshold for optical absorption in molecular C60. This result suggests that the enhanced photoconductivity in the polymer composite involves the photoexcitation of Qq, and through this process an electron is exchanged between the fullerene and the nearby carbazole unit. Thus mobile carriers are generated in the PVK polymer chains by the incident light and these carriers can drift to the film surface, neutralizing the applied surface charge. In the UV, it was noted [13.163] that the charge generation efficiency <f> was much larger for positive, than for negative, surface charge neutralization.

Fig. 13.35. Schematic illustration of the photoinduced electron transfer from semiconducting polymers to form Cj,, anions [13.168].

In a series of experiments on a variety of C60-polymer composites, the UC Santa Barbara researchers obtained results indicating that a photoinduced electron transfer from a semiconducting polymer to a nearby C60 molecule occurs, forming metastable Q,, anions (consistent with the high electron affinity of Qg) and mobile holes in the polymer [13.164,166,168, 169,173]. A schematic illustration of this process is presented in Fig. 13.35, where it is shown that the photoexcitation occurs on the polymer (rather than on the C60 molecule), followed by electron transfer to the C^q. For example, using 2.4 eV excitation, the intense luminescence from the photo-excited polymer [e.g., poly(2-methoxy, 5-12'-ethyl-hexyloxyl)-/>-phenylene vinylene, or MEH-PPV] was quenched by nearly three orders of magnitude (see Fig. 13.36) when C60 was introduced into MEH-PPV at 1:1 by weight [13.164], This observation indicates a strong excited state interaction between MEH-PPV and C50. It was also shown that at low light intensity the absorption spectrum of a 1:1 C60:MEH-PPV sample appears to be a simple superposition of spectra from C«, and from MEH-PPV, indicating that in the ground state, the C50-polymer composite is a simple, weakly interacting mixture [13.164]. A schematic band diagram useful for understanding

Was this article helpful?

0 0

Post a comment