In the present work, the well-known friction theory (FT) based on friction concepts of classical mechanics and the van der Waals theory of fluids has been modified to accurately represent the dynamic viscosity of pure ionic liquids in a rather simplified manner. Unlike previous FT applications to pure ionic liquids, the dilute gas limit for viscosity was excluded from the present model since it is negligible for ionic liquids which exhibit extremely low vapor pressures; the friction term thus prevailed for viscosity calculations. The latter was expressed in terms of new temperature-dependent friction coefficients whereas the repulsive and attractive pressure terms were in turn estimated from two simple cubic equations of state (Soave or Peng–Robinson) rather than using sophisticated multiparameter equations of state such as SAFT- or CPA-based expressions previously used by other authors. The resulting model was successfully validated during the representation of experimental dynamic viscosities of three families of imidazolium-based ionic liquids ([CXmim]­[BF4], [CXmim]­[PF6], and [CXmim]­[Tf2N]) within a temperature range varying from 0 to 80 °C and at pressures from 1 up to 3000 bar.

No abstract

iscosity is an important transport property in process design and development. Knowledge of viscosity of pure liquids and of the liquid mixtures plays a key role in solving chemical engineering problems dealing with heat-transfer and masstransfer operations, and fluid-flow processes. Much effort has been directed towards the development of reliable models for correlating and predicting viscosities of pure nonpolar and polar liquids. However, there is still a lack of a sound theory for the accurate prediction of liquid viscosities within a wide Ž . range of temperatures and pressures Reid et al., 1987 . Furthermore, viscosity models based on similarities either with gas-like or solid-like behavior have not successfully been proven to yield values of liquid viscosity for pure liquids and Ž . mixtures as functions of temperature Chhabra, 1992 . Consequently, most of the viscosity models reported in the literature have been developed under semi-empirical and empirical considerations. Pertinent literature on viscosity models for pure liquids has been reviewed and evaluated by Przezdziecki Ž . Ž . Ž . and Sridhar 1985, Reid et al. 1987, Mehrotra 1991 , Mon-Ž . Ž . nery et al. 1995, Mehrotra et al. 1996 , and Poling et al. Ž . 2001 . A thorough review of the literature reveals that numerous empirical and semiempirical models have been published for viscosities of pure liquids with exponential viscosity-temperature functions receiving most of the attention. The corresponding states principle along with the Eyring's absolute rate theory are the most widely used approaches to represent the Ž . viscosity of pure liquids. Recently, Lei et al. 1997 presented a two-parameter model based on the Eyring's absolute rate theory for the correlation of the viscosity of pure liquids under saturated conditions. Their correlating results yielded an overall ADD of 1.51% in the representation of experimental liquid viscosities of 106 pure nonpolar and polar fluids. An important feature of the model proposed by Lei et al. is that, at a given temperature, the liquid viscosity can directly be computed from equilibrium properties such as vapor presCorrespondence concerning this article should be addressed to R. Macıas-Salinas. sure, heat of vaporization, and compressibility factors of the phases at equilibrium. Based on the two-parameter model of Ž . Lei et al. 1997 , the purpose of the present study is three fold. The first purpose is to make use of a well-known cubic equation of state to calculate all of the thermodynamic properties required in the model, thus enabling the simultaneous calculation of vapor-liquid equilibria and saturated liquid viscosity. The second purpose is to improve the correlation of liquid viscosity over a wider temperature range by modifying the functionality between the activation energy and the internal energy of vaporization, and, the third purpose is to extend the use of the model to the calculation of compressed Ž liquid viscosities at pressures larger than the saturation . point .

No abstract

Axial mixing measurements of air and water under two-phase flow conditions were ( ) made in a large-scale packed column 0.43 m diameter using tracer experiments. Part I of this article dealt with single-phase mixing in the same column, with the same internals. Four packings were studied: 25.4-mm ceramic Raschig rings, 25.4-mm metal Pall rings, Sulzer BX structured packing, and Flexipac 2 structured packing. Air and water flowed countercurrently through the column at atmospheric pressure and at gas rates ®arying from 0.25 kgrm 2 ؒ s up to the flooding point, and liquid rates from 3.25 to 8.5 kgrm 2 ؒ s. A diffusion-type model ser®ed to reproduce the experimental response cur®es obtained for both phases. The results confirmed pre®ious obser®ations for first-generation packings: axial mixing in the gas increases with both gas and liquid rates, whereas liquid-phase axial mixing is a decreasing function of liquid rate and is insensiti®e to gas rate up to the flooding point. It was also found that the BX packing produces the least mixing in both phases. The largest mixing effects in the gas phase are found for the Raschig rings, and the largest mixing effects for the liquid phase are found for Flexipac 2. Correlations were de®eloped to reproduce the results, yielding an a®erage " 22% difference between experimental and correlated data.