This paper presents a solvothermal strategy for chemical modification of TiO(2) nanoparticles with carboxylic acids. Solvothermal reaction between the TiO(2) nanoparticles and carboxylic acid molecules in an autoclave at 100 degrees C provides carboxylic acid-modified TiO(2) particles with a modification efficiency much higher than the conventional immersion method. TiO(2) nanoparticles were prepared by hydrolysis of titanium isopropoxide in nitric acid solution; the modified nanoparticles were characterized by powder X-ray diffraction pattern, scanning electron microscopy, absorption and Fourier transform infrared spectra, and thermogravimetric analysis. Results show that the binding form of the modifier molecules on TiO(2) surface is in a bidentate chelating mode, the crystalline phase composition and morphological structure of the preformed TiO(2) nanoparticles are not affected by the solvothermal reaction, and the surface coverage of the modifier molecules can be adjusted by the weight ratio of modifier/TiO(2) during feeding. It is evident that the reaction processes in the solvothermal strategy involve the formation of double hydrogen bondings between carboxylic acid molecule and TiO(2) at the same Ti site and the coordination at solvothermal temperature by dehydration from the hydrogen bondings. The solvothermal strategy for modifying TiO(2) nanoparticles is expected to find potential applications in many fields; for example, our results demonstrate that the photovoltaic performance of the TiO(2) nanoparticles can be improved by the solvothermal modification even with an insulating modifier and controlled by the modifier coverage.

For characterization of polymer-based solar cells with vertically aligned ZnO nanorod arrays (ZnO-NAs) by intensity modulated photocurrent spectroscopy (IMPS), a dynamic IMPS model is developed, where the structure-related charge generation and transport dynamics are considered. The model describes the IMPS responses affected by the phase shift φ n (ω) due to the exciton diffusion property (ω 0 ) and the structurerelated device ideality factor N, the electron diffusion coefficient D e , the exciton dissociation rate S, and the device structure (e.g., nanorod length d and interspacing l). The main expectations of the model are confirmed by the experimental data of the polymer/ZnO-NA cells with d ) 180-650 nm, offering mechanistic information on the structure-related charge generation, charge transport, and device performance. The presence of the φ n (ω) makes IMPS response not spiral into the origin and the phase angle in its Bode plot not tend to 90°; the d-dependent direct diffusion (DD) and diffusion-reflection (DF) transport processes are normally involved in the travel of injected electrons to the collection electrode; the incident photon-to-current conversion efficiency (IPCE) and the transit time (τ D ) for DD transport under the influence of DF process reach their peak values at d ≈ 500 nm, and the φ n (ω) effect on electron transport is affected by ω 0 , D e , and S. Satisfactory fittings of measured IMPS responses to the model further reveal that the d dependence of the IPCE or the photocurrent actually originates from the S value governed by d-dependent exciton generation and dissociation; when changing d, a larger number of electrons for DD transport causes a smaller N or a more remarkable φ n (ω) effect; a longer τ D is accompanied by a larger RC effect of the ZnO electrode. Those results clearly suggest that a highly efficient polymer/ZnO-NA device requires d ≈ 500 nm and l ) 5-10 nm, along with a high interfacial exciton dissociation efficiency.