Search

Substoichiometric titanium carbide of formula TiC0.6 with different degrees of carbon vacancy ordering has been synthesized first by reactive sintering at high temperature (2100 °C) and then by annealing at low temperature (730 °C) for different durations. The effect of annealing on the structure of the carbide and its capacity to electrochemically insert hydrogen has been investigated. The XRD study reveals two phase transitions in the carbide during annealing. First, annealing of disordered TiC0.6 (space group Fm3¯m) for 40 h leads to a trigonal superstructure of space group R3¯m thanks to ordering of carbon vacancies. After the longest annealing time of 120 h, the structure of the carbide becomes cubic anew, but with a space group of Fd3¯m. The electrochemical hydrogen insertion in these different types of TiC0.6, as studied by cyclic voltammetry, strongly depends on the crystalline structure of the carbide. The maximum storage capacity is obtained for 40 h, corresponding to the R3¯m ordered phase. Because the R3¯m structure consists of alternately empty and full (111) carbon atomic planes, we show that the hydrogen insertion and diffusion in the material is rendered possible by the presence of vacant (111) planes. This behavior is very similar to that of a cation-ordered manganese-nickel spinel vis-à-vis lithium ion insertion.

A new method is described to simultaneously determine the kinetics of surface partitioning and the lateral diffusion constant of redox active amphiphiles. It concerns water-soluble amphiphiles for which the surface adsorption equilibrium constant and the solution diffusion constant are measured independently. The method involves cyclic voltammetric experiments carried out at the air/water interface with microband electrodes aligned with the plane of the water surface. Typically, 100 nm wide, 1.0 cm long microband electrodes are fabricated by the vacuum vapor deposition of gold films on glass. The front face of the electrode substrates are coated with impermeable, dimensionally stable, polymer barrier films with thickness L in the range of approximately 0.1-1.0 microm. Fracturing such gold-coated glass substrates exposes gold microbands. The recorded voltammetric current sensitively depends on the barrier film thickness, the surfactant surface diffusion constant, Dsurf, and its rate constant of desorption, kdes. For a given surfactant, such as the nitroxyl piperidine free radical TEMPO featured in this report, large currents are observed with microband electrodes that do not carry a barrier film (L = 0). This is because the surfactant surface population diffusing along the air/water interface can be directly electro-oxidized at the edge of the microband. Smaller currents are measured in the presence of a barrier film, since, in those instances, the surface population may contribute to the voltammetric current only via a mechanism involving surfactant desorption from the water surface into bulk, where it contributes to the three-dimensional solution diffusion processes. The quantitative interpretation of the voltammetric experiments was made possible with finite element simulations with FEMLAB. These produce a set of calibration curves, Dsurf versus log kdes, for each value of the barrier film thickness. The intersection of the calibration curves determines the unique values of Dsurf and kdes. For TEMPO, Dsurf = 4.4 +/- 1.2 x 10(-5) cm2/s and kdes >/= 2 x 10(4) s(-1). Surfactant desorption rate constants of this magnitude have not been previously experimentally accessible. Since, in our earlier report (Wu, D. G.; Malec, A. D.; Head-Gordon, M.; Majda, M. J. Am. Chem. Soc. 2005, 27, 4490-4496), we showed that TEMPO is not immersed in water and that it diffuses along the interface hydrogen-bonded to just one or two water molecules, its Dsurf value approximates the water diffusion constant in the aqueous liquid-vapor interfacial region.

Surface partitioning and lateral mobility of TEMPO (2,2,6,6-tetramethyl-1-piperidynyloxy free radical) in the aqueous liquid/gas interfacial region were investigated electrochemically with 100 nm wide, 1.0 cm long microband electrodes positioned at the air/water interface. For redox active amphiphiles such as TEMPO, the electrochemical current is the sum of the surface and solution components representing the diffusive transport of TEMPO in both domains as well as the dynamics of equilibration at the air/water interface. Interpretation of the recorded current-voltage curves was aided by a FEMLAB simulation code developed to analyze transport processes in this class of systems. TEMPO and TEMPO(+) partition constants (K(T), K(T)+) and solution diffusivities (D(sol), equal for the two species) were obtained experimentally yielding K(T) = 5.0 +/- 0.7 x 10(2) M(-1), K(T+) = 41 +/- 3 M(-1), and D(sol) = 7.7 +/- 0.35 x 10(-6) cm(2)/s. In view of the low value of K(T)+, TEMPO(+) was assumed not to partition to the air/water interface. We further assumed that the desorption rate constants (k(des)) of both TEMPO and TEMPO(+) were the same. Good fits between the recorded and simulated cyclic voltammograms were obtained using two correlated, adjustable parameters, k(des) and the TEMPO lateral, surface diffusion constant (D(surf)). Detailed analysis of the behavior of this class of systems was obtained for a broad range of D(surf) and k(des) values. In addition, a calibration curve of k(des) versus D(surf) was obtained. Assuming that TEMPO k(des) is in a likely range of 10-100 s(-1), its lateral diffusion constant is in the range of 7.9-3.6 x 10(-5) cm(2)/s. In view of our earlier work (Wu, D. G.; Malec, A. D.; Head-Gordon, M.; Majda, M. J. Am. Chem. Soc. 2005, 127, 4490-4496) showing that at the air/water interface TEMPO is unimmersed, and that its interactions with water are limited to hydrogen bonding with one or two water molecules, D(surf) can be related to the viscosity of the aqueous interfacial region.