Maps of pollutants concentration are usually generated by means of interpolation and extrapolation methods. The quality of the results depends mainly of the number of permanent or temporary measuring stations. This paper deals with a method for the virtual densification of the network of stations. The method creates "virtual measuring stations". It aims at improving the quality of interpolation by increasing the number of data on pollutant concentration. The virtual stations are determined by the means of a classification method applied to each pixel of the area under concern. Discriminating elements are pollutants emission classes, land cover types, urban morphological indicators created to this purpose and distance to major roads. A first implementation was made for particulate matter (PM) for the city of Strasbourg (France) using thin-plates spline interpolation method in Arcview 9 GIS. The relative Root Mean Square Error decreases from 49% for five input stations down to 15% for the virtual stations.

In the present study, are reported investigations obtained with the room temperature molten salt (RTMS) ethyl-methyl-imidazolium bis-(trifluoromethanesulfonyl)-imide (EMI-TFSI) in order to use it as solvent in lithium battery. The thermal stability, viscosity, conductivity and electrochemical properties are presented. A solution of 1m lithium bis-(trifluoromethanesulfonyl)-imide (LiTFSI) in EMI-TFSI has been used to test the electrolyte in a battery with LiCoO2 and Li4Ti5O12 as respectively cathode and anode materials. Cycling and power measurements have been obtained. The results have been compared with those obtained with a molten salt formulated with a different anion, BF4− and with a conventional liquid organic solvent EC/DMC containing LiTFSI. The 1m LiTFSI/EMI-TFSI electrolyte provides the best cycling performance: a capacity up to 106mAhg−1 is still delivered after 200 cycles, with 1C rate at 25°C.

L’objectif de ce travail était de comprendre le comportement des mélanges de solvants sur la résistance des matériaux protecteurs et de développer un modèle de prédiction de la résistance des matériaux aux solvants industriels. Dans ce but, deux techniques ont été développées, l'une consistant à mesurer le gonflement dynamique à partir des mesures d'allongement d’une membrane polymérique et l'autre pour mesurer la perméation par volumétrie. La technique d'allongement dynamique s’est avérée très utile pour la mesure des coefficients de diffusion et pour obtenir les paramètres tridimensionnels de solubilité de Hansen. La technique volumétrique, comme celle de l’allongement dynamique, est applicable aux solvants peu volatiles. Les résultats obtenus avec ces deux méthodes se comparent très bien à ceux obtenus par les tests standards de perméation, ASTM F 739 et F 1407. Une étude systématique de la perméation de membranes de néoprène, de nitrile et de butyle aux mélanges de solvants a été faite. En général, on observe que (1) le signe de la déviation de l’additivité, sur une échelle de fraction molaire, est le même pour les taux de perméation et pour les taux de gonflement et de signe opposé au temps de claquage, (2) les déviations de l’additivité sont en grande partie indépendantes de la nature des trois membranes étudiées, pourvu que le matériau ne soit pas dégradé par les solvants, (3) pour les mélanges qui ont une enthalpie de mélange exothermique élevée, la protection de la membrane au mélange est plus grande que prévue par le principe d’additivité, alors que pour les mélanges ayant une enthalpie de mélange endothermique importante, la protection est plus faible que prévue. À partir de ces observations, un nouveau modèle thermodynamique a été développé pour prédire la perméation aux mélanges de solvants basés sur l’utilisation de paramètres enthalpiques. Considérant l’incertitude dans la détermination des paramètres de perméation, le modèle prédit généralement assez bien le signe et souvent l’ordre de grandeur des déviations de l’additivité, lorsque les paramètres enthalpiques sont élevés. Pour les systèmes où les paramètres sont relativement petits, le modèle prédit correctement que les déviations de l’additivité pour les paramètres de perméation seront faibles. Toutefois, dans plusieurs cas où l’un des solvants est très soluble dans la membrane et l’autre peu soluble, le modèle prédit le mauvais signe de la déviation pour les temps de claquage lorsque l’enthalpie de mélange est endothermique et pour les taux de perméation et les taux de gonflement lorsque les enthalpies sont exothermiques. Les facteurs responsables de ces anomalies n’ont pas été explorés dans ce travail. Néanmoins, bien que dans l'interaction solvant pur - polymère la perméation soit reliée fondamentalement à la cinétique de diffusion, la déviation de l'idéalité dans le cas des mélanges des solvants est interprétée par le modèle thermodynamique. En effet, les interactions solvant - solvant vont affecter la cinétique de diffusion du mélange. L’approche du modèle thermodynamique a été extensionnée à des mélanges ternaires et à des solvants industriels complexes. Dans les cas où ces mélanges contiennent des solvants polaires et des solvants peu polaires (enthalpies de mélange endothermiques), il suffit qu’un des composants soit soluble dans le polymère pour que le matériau offre peu de résistance au mélange. Il est aussi possible d'utiliser des informations sur les paramètres de solubilité des polymères et des solvants, plutôt que d'utiliser des données de perméation, comme point de départ dans le calcul de prédiction de la résistance des matériaux aux mélanges de solvants. Dans cette étude, nous avons obtenu les paramètres de solubilité tridimensionnels de Hansen pour les matériaux des gants, à partir des données d'élongation dynamique. Un progiciel a été élaboré pour la prévision de la résistance des gants contre les mélanges de solvants. Cependant, le travail doit être complété en mettant au point une base des données avec des informations pour les solvants les plus utilisés en milieu de travail.

Flow densimetry and flow microcalorimetry have been used to study the effect of electrolytes, surfactants, and alcohols on positively and negatively charged polystyrene latexes. The thermodynamic volumes and heat capacities of transfer of the latex from water to the aqueous solutions and of the additives from water to the latex dispersions were measured as a function of the additive concentration. With some systems, the adsorption isotherms were also determined. These thermodynamic properties are largely insensitive to electrostatic interactions but depend significantly on changes in hydrophobic interactions resulting from adsorption of the additive on the latex. Negligible changes in the quantities of transfer are thus observed for electrolytes, methoxyethanol, and methanol, but ionic and nonionic surfactants are adsorbed on both types of latexes. The adsorption of medium-chain-length alcohols and alkoxyethanols increases with their hydrophobic character. A simple model based on the Langmuir isotherm, the thermodynamic properties of the additives in water, and the properties of the adsorbed additives correctly predicts the general trends of the transfer functions of the additive and of the latex. However, a detailed analysis of the data indicates that the adsorption isotherm is more complex than expected from the Langmuir equation and that the thermodynamic properties of the adsorbed species depend on the extent of coverage. At high concentrations, the volumes and heat capacities of adsorbed surfactants are of the same order of magnitude as the corresponding values of the surfactants in micellar form.

Lithium bis(trifluoromethylsulfone)imide (LiTFSI), a promising electrolyte for high energy lithium batteries, forms several stable solvates having low melting points in aprotic solvents. In a previous study (D. Brouillette, G. Perron and J. E. Desnoyers, J. Solution Chem., 1998, 27, 151), it was suggested, based on thermodynamic studies, that such stable solvates may persist in solution and influence their properties. To verify this hypothesis, phase diagrams and Raman spectra have been measured for solutions of LiTFSI in acetonitrile, propylene carbonate and glymes (n(ethyleneglycol) dimethyl ether or Gn), which have the chemical structure CH3-O(CH2-CH2-O)(n)-CH3 for n = 1 to 4 and 10. The relative intensities of the LiTFSI and solvent Raman bands are proportional to the concentration for systems without solvates. The systems for which stable solvates were identified in the phase diagram show important changes in the relative intensities for both the LiTFSI and the solvent Raman bands at concentrations corresponding to particular stoichiometries and support the conclusion that stable solvates are present in the solutions. The structure of the crystalline G1:LiTFSI solvate was determined by X-ray crystallography. Structures for (G2)(2):LiTFSI and (G1)(3):LiTFSI solvates are proposed.

The permeation of pure organic liquids and mixtures of organic liquids through commercial butyl, neoprene, and nitrile membranes was studied using dynamic material deformation (swelling) and permeation techniques. The derived parameters, the breakthrough time (t(BT)), steady-state permeation rate (SSPR), and initial swelling rate (SR), show deviations from additivity for the mixtures, based on the parameters of the pure liquids on a mol fraction basis. In the majority of cases for the three membranes examined, the deviations are independent of the nature of the membranes, and the signs of the deviations for t(BT) are opposite to those for SSPR or SR, provided that the membranes are not degraded by one of the solvents. An approach that considers only solvent-solvent interactions based on the enthalpy of mixing was used to predict deviations for mixtures. For mixtures where the enthalpy of mixing is large and exothermic, the permeation of the mixture is less than expected, while for systems where the enthalpy of mixing is large and endothermic, the permeation is larger than expected. A simple semiempirical model predicts the sign and magnitude of the permeation of 73% of the system-permeation property combinations investigated, which show significant deviations from ideality. It is interesting to note that the wrong predictions are for systems where the predictions are positive, that is, for SSPR and SR rates with endothermic systems and for t(BT) with exothermic systems. The exceptions also seem to be for systems that correspond to materials having a high resistance to one of the solvents and a very low resistance to the other solvent. Examples of ternary-mixture permeation data are also given and show that, even if two of the pure components do not permeate through a membrane, the membrane will offer little protection if the third component shows a high affinity for the membrane and if the enthalpies of mixing of this component with the other liquids are endothermic. (C) 2002 Wiley Periodicals, Inc.

Atmospheric pollution becomes a critical factor of anticipated deaths, which concerns ambient air quality as well as air quality in houses and places of work. For the well being of human and for the population information, we need to evaluate the actual exposure of persons to ambient pollution. One way to perform it is to evaluate the space time budget of air pollution exposition. Hence an information on the spatial distribution of pollutant concentrations is required. Several tools exist; most of them provide maps of pollutant concentrations but over a regional scale with a grid of I km, which is insufficient. Measuring stations, scarcely distributed in the city, provide a complete surveillance but their costs limit the knowledge of pollutant concentration to specific points of the city. To overcome this problem, several notions are defined: "identity card of a measuring station", "pseudostation" and "virtual station". Based on a multi-sources approach, this paper presents a methodology using remotely sensed data for the mapping of pollutant concentrations over a city.

As part of a study on the optimization of the electrolyte for high energy lithium batteries, the conductivity, viscosity and density of LiAsF6, LiBr, and LiClO4 were measured in aprotic solvent mixtures. The conductivity of lithium bis(trifluoromethylsulfone)imide (LiTFSI) was also obtained in a large number of mixed aprotic solvents. The solvents were chosen to verify the effect of various parameters such as viscosity, permittivity, volume, acceptor number and donor number on the conductivity. These results were used to develop a simple model for excess conductivities based on the viscosity and volume of the pure solvents. Without adjustable parameters, this model predicts the correct sign of the excess conductivities in ≈90% of the cases and the magnitude of the conductivity of the ternary mixtures within an average of 15%. Deviations from the predictions are mostly observed with solvents of low permittivity and this supports the hypothesis that a different conduction mechanism is in operation at high concentration in these solvents, and the solvating power of these solvents plays an important role in this mechanism.

The Bjerrum association model, developed in 1926, is now incorporated in many conductance theories of electrolyte systems to extract Λ 0 and K A from experimental data. The Bjerrum concept is simply a convenient way of taking into account short-range electrostatic interactions between ions. The equations of the Bjerrum model can be applied to the prediction of the temperature and pressure dependencies of K A from the value of K A at a reference T and P and from the dielectric properties of the solvent. This feature will be essential when the relaxation effect is taken into account when applying the model to heat capacities and compressibilities. These equations were tested against literature K A values (obtained from treatment of conductance data by equations that incorporate the Bjerrum concept) in aqueous electrolyte solutions at high temperatures and pressures and in some electrolyte systems in acetonitrile, 2-butanone, propylene carbonate, γ-butyrolactone, and propanol. In the absence of specific interactions in solution, the agreement between experimental and predicted K A are generally quite good. Notable exceptions are acids and bases in water, lithium perchlorate in most solvents, and the majority of electrolytes in propylene carbonate, suggesting that specific interactions in these systems may cause the model to fail.