Abstract: Pure and Nb-doped TiO2 thick-films were prepared by screen-printing, starting from nanosized powders. Grain growth and crystalline phase modification occurred as consequence of firing at high temperature. It has been shown that niobium addition inhibits grain coarsening and hinders anatase-to-rutile phase transition. These semiconducting films exhibited n-type behavior, while Nb acted as donor-dopant. Gas measurements demonstrated that the films are suitable for CO or NO2 sensing. Microstructural characterization by electron microscopy and differential thermal analysis (DTA) highlights the dependence of gas-sensing behavior on film's properties. (C) 2000 Elsevier Science S.A. All rights reserved.

Abstract: Catalytic methods for the production of carbon nanotubules have been developed based on the decomposition of acetylene on well-dispersed metal particles strongly adsorbed on a support. Cobalt on silica was found to be the best catalyst-support combination for the production of graphitic tubules. The method for the catalyst preparation and the reaction conditions were optimized. Straight and coiled carbon tubules were obtained with inner and outer diameter of 3-7 and 15-20 nm, respectively, and up to 30 mum in length. These nanotubules were not coated by amorphous carbon. Traces of amorphous carbon could be removed by hydrogen. High resolution electron microscopy images and electron diffraction patterns of the straight nanotubules were similar to those obtained by the arc-discharge method. Coiled nanotubules were revealed by TEM to be regular polygonized helices where the bends are caused by pairs of pentagon-heptagon carbon rings among the hexagonal network.

Abstract: FT-IR spectroscopy and impedance measurements of tin dioxide sensor materials at working temperatures up to 450 °C in atmospheres with varying O2/N2 ratio are used as an in situ probe to study the interactions at the surface of the semiconducting oxide. Every diminution in the oxygen content above the sample induces a broad IR absorption band (X-band) between 2300700 cm−1 with a few small peaks in the 1400850 cm−1 region of the spectrum superimposed on it. The X-band results from the enchanced electron concentration in the bulk of the tin dioxide domain. The fine structure is due to the absorption of several kinds of surface oxygen species associated vibration modes. The porous tin dioxide consists of domains were the outward shell is depleted of electrons by the formation of adsorbed O− species on oxygen surface sites, SO(O− species. In our proposed model for the impedance data this gives rise to a parallel RpCp circuit for the domain boundary characteristics and to an Rs parameter for the intradomain resistance. The evolution of these IR and impedance spectroscopic effects with temperature and oxygen content is used to set up, to confirm and refine a physicochemical operation model of tin dioxide gas sensor. This model consists of a sensitizing reaction sequence in the presence of oxygen and a gas-detection reaction sequence when a reducing gas is present. Based on this model, the principal disadvantages of this type of gas sensor become clear. Every factor that influences the concentration of SO(O−) species, causes a conductance modification. If we can control and direct the nature, the number and the arrangement of the tin dioxide domains, a directed development and improvement of the sensor characteristics is possible.

Abstract: Adding 20mg of 1-(2-pyridylazo)-2-naphthol (PAN) to a water sample at 70°, and filtering off the precipitate after cooling, gives efficient preconcentration prior to X-ray fluorescence analysis of water. Up to the capacity of about 100 μeq of PAN used, the trace metal recoveries are around 90% or higher for Cr3+, Mn2+, Ni2+, Cu2+, Zn2+, Hg2+ and Eu3+, and above 70% for many other ions. The recovery yields usually do not vary critically with pH in the neutral pH-range, and are practically independent of the sample salinity, sample volume and trace-metal concentration. Enrichment factors as high as 2 × 105 can be achieved. Counting statistics would then allow detection limits of 0.03 ppM. The blank levels in commercial PAN, however, lead to typical detection limits of about 1 ppm. The coefficient of variation is typically in the 510% range at the 10-ppM level. The accuracy and applicability of the procedure are illustrated by comparative analyses on samples of synthetic solutions, river and drinking water.

Abstract: For trace analyses of environmental waters, spark-source mass-spectrometry has been combined with a preconcentration procedure involving chelation of the dissolved trace elements with oxine and subsequent adsorption of the oxinates and naturally occurring organic and colloidal metal species onto activated carbon. The activated carbon is filtered off and ashed at low temperature. The residue is dissolved, an internal standard and pure graphite are added and, after drying, the electrodes are prepared. The photographically recorded mass spectrum is evaluated by a suitable computer routine. The error of the procedure is around 30%. While this preconcentration and analysis procedure is capable of measuring about 40 elements quantitatively, in practice 1025 trace elements are determined simultaneously above the 0.1-μg/l. detection limit, as is illustrated by analyses of drinking water, surface and ground water samples. Although a sophisticated technique, SSMS can be considered for regular panoramic survey analyses.