Influence de la longue mémoire sur le comportement asymptotique d'estimateurs fonctionnelsInfluence of long memory on the asymptotic behaviour of functional estimators

This Note gives asymptotic results for kernel estimators of several functionals (density, failure rate and distribution functions) of a long-memory process. The results are presented in such a way as to highlight the influence of the long dependence on the rates of convergence. To cite this article: G. Estevez-Perez, P. Vieu, C. R. Acad. Sci. Paris, Ser. I 335 (2002) 271–274.     

Effects of processing on the structure of zein/oleic Acid films investigated by x-ray diffraction

Zein films plasticized with oleic acid were formed by solution casting, by the stretching of moldable resins, and by blown film extrusion. The effects of the forming process on film structure were investigated by X-ray diffraction. Wide-angle X-ray scattering (WAXS) patterns showed d-spacings at 4.5 and 10 A, which were attributed to the zein alpha-helix backbone and inter-helix packing, respectively. The 4.5 A d-spacing remained stable under processing while the 10 A d-spacing varied with processing treatment. Small-angle X-ray scattering (SAXS) detected a long-range periodicity for the formed films but not for unprocessed zein, which suggests that the forming process-promoted film structure development is possibly aided by oleic acid. The SAXS d-spacing varied among the samples (130-238 A) according to zein origin and film-forming method. X-ray scattering data suggest that the zein molecular structure resists processing but the zein supramolecular arrangements in the formed films are dependent on processing methods. Structural model for a zein molecular aggregate (based on Matsushima et al.10). Rectangular prisms of individual zein molecules are hexagonally aligned parallel to each other.     

High-performance liquid chromatographic determination of Levetiracetam in human plasma: comparison of different sample clean-up procedures

An accurate and precise high-performance liquid chromatographic method using diode array detection for the determination of the novel antiepileptic, Levetiracetam, has been developed. Three clean-up procedures for the analysis of Levetiracetam in human plasma were implemented and evaluated, namely solid-phase extraction, deproteinization by addition of organic solvents and formation of insoluble salts. Adenosine was used as the internal standard for all three sample pretreatment procedures. Among the several cartridges used for solid-phase extraction, the hydrophilic-lypophilic balance (Oasis) HLB) phase provides the best extraction yield of Levetiracetam, together with high precision. With the two other clean-up procedures involving plasma deproteinization by addition of methanol or zinc sulphate, lower sensitivity and precision of the assays were obtained. However, they are cheaper and faster when compared with the solid-phase extraction procedure.     

Electron–Electron interaction energy in homonuclear diatomic molecules

The derivative of the electron–electron potential energy U_ee with respect to internuclear separation R is studied for light homonuclear diatomic molecules at equilibrium. It is readily related to nuclear–nuclear potential energy U_nn, the force constant K, and the electron–nuclear potential energy U_en. An approximate expression, based on the simplest form of density functional theory, is then used to eliminate dU_en/dR|_Re. The result thus obtained for dU_ee/dR|_Re transcends an earlier proposal of Kryachko by including a term 2/3R_eK, with K the force constant. Numerical tests at SCF–RHF level are presented for nine homonuclear diatomic molecules.     

A Protein‐Resistant Matrix for Electrochemical Based Recognition Assays

In this work, two well known polymers, carboxymethyl dextran and poly(ethyleneglycol), are used for easily producing a platform for electrochemical affinity assays, avoiding nonspecific adsorption of proteins. In this way, modified electrode surfaces able to bind a recognition agent are constructed through simple and reliable reactions.     

First- and second-order multivariate calibration applied to biological samples: determination of anti-inflammatories in serum and urine

First- and second-order multivariate calibration of fluorescence data have been compared as regards the determination of anti-inflammatories and metabolites in the biological fluids serum and urine. The simultaneous resolution of naproxen-salicylic acid mixtures in serum and naproxen-salicylic acid-salicyluric acid mixtures in urine was accomplished and employed for a discussion of the relative advantages of the applied chemometric tools. The analysis of second-order fluorescence excitation-emission matrices was performed using iteratively reweighted generalized rank annihilation method (IRGRAM), parallel factor analysis (PARAFAC), and self-weighted alternating trilinear decomposition (SWATLD). The results were compared with first-order fluorescence emission data analyzed with partial least-squares regression (PLS). In all cases, the performance of the methods was improved through the formation of inclusion complexes of the analytes with beta-cyclodextrin. The concentration ranges in which the analytes could be determined were as follows: naproxen, 0-250 ng mL(-1) in serum and 0-200 ng mL(-1) in urine; salicylic acid, 0-500 ng mL(-1) in serum and 0-300 ng mL(-1) in urine, and salicyluric acid, 0-300 ng mL(-1) in urine.     

The crystal structure of eudesma-4(15),7(11)-dien-8α, 12-olide: a sesquiterpene lactone fromTrattinickia rhoifolia Willd

The title compound, formula C₁₅H₂₀O₂, is orthorhombic, P2₁2₁2₁ witha=8.747(2),b=12.025(3),c=12.554(3)Å,Z=4, andD _m =1.32(2)g/ml. The structural analysis shows that the compound corresponds to eudesma-4(15),7(11)-dien-8,12-olide, a sesquiterpene lactone previously isolated fromAster umbellatus but whose crystal structure was unknown.     

Flow-through optical fiber sensor for automatic sulfide determination in waters by multisyringe flow injection analysis using solid-phase reflectometry

A software-controlled flow-through optical fiber diffuse reflectance sensor capitalized on the implementation of disk-based solid-phase pre-concentration schemes in a multisyringe flow injection analysis (MSFIA) set-up is proposed for the trace determination of sulfide in environmental waters and wastewaters. The fully automated flowing methodology is based on Fischer's coupling reaction of sulfide with N,N-dimethyl-p-phenylenediamine (DMPD) in the presence of Fe(iii) as oxidizing reagent in a 0.5 M HCl medium. The on-line generated methylene blue dye is subsequently delivered downstream to a dedicated optode cell furnished with an octadecyl-chemically modified (C(18)) disk, while continuously recording the diffuse reflectance spectrum of the pre-concentrated compound. A double regeneration protocol is finally executed to warrant minimum background noise and negligible baseline. Under the optimized chemical and hydrodynamic conditions, the optosensing MSFIA method features coefficients of variation better than 0.7%(n= 10) at 50 microg l(-1) concentration, a linear working range of 20-200 microg l(-1) sulfide, a 3sigma(blank) detection limit of 2.9 microg l(-1) sulfide and an injection throughput of 8 h(-1) for a pre-concentration sample volume of 2.9 ml. The interfacing of the robust and versatile multisyringe flow injection-based optode with a plug-in spectrophotometer furnished with a light emitting diode assures the miniaturization of the overall flow analyzer, which is, thus, readily adaptable to real-time monitoring schemes. The potential of the multisyringe flow method was assessed via the determination of sulfide traces in water samples of different complexity (namely, freshwater, seawater and wastewater).