Densities and Viscosities of MTBE + Heptane or Octane at p ) 0.1 MPa from (273.15 to 363.15) K Diana C. Landaverde-Cortes, Alejandro Estrada-Baltazar, and Gustavo A. Iglesias-Silva* Departamento de Ingenierı ´a Quı ´mica, Instituto Tecnolo ´gico de Celaya, Celaya, Guanajuato, CP 38010, Mexico Kenneth R. Hall Chemical Engineering Department, Texas A&M University, College Station, Texas 77843-3122 This work presents atmospheric density and viscosity values for methyl tert-butyl ether + heptane or octane over the whole range of compositions from (273.15 to 363.15) K. A vibrating tube densimeter is used for the density measurements, and a Cannon-Fenske viscosimeter is used for the viscosity measurements. Excess molar volumes have been calculated from the density measurements and represented with a Redlich-Kister equation. The root- mean-square deviation of the excess molar volumes from the literature values is 0.018 cm 3 ‚mol -1 . We also have represented our kinematic viscosity values with a three-body McAllister equation. Introduction Knowledge of densities and viscosities of pure substances and mixtures is necessary for the design of chemical processes. Also, a need always exists to have accurate experimental densities and viscosities of liquids to develop new theories that probe the interactions between molecules of dissimilar size and polarity. With these data, we can develop accurate predictive models useful in the chemical industry. Methyl tert-butyl ether (MTBE) is a volatile, flammable, and colorless liquid that is relatively soluble in water. It also is used medically to dissolve gallstones. MTBE has been a gasoline additive at low levels since 1979, replacing tetraethyl lead to increase octane rating and reduce engine knocking. Since 1992, MTBE has been used at higher concentrations in some gasolines to fulfill oxygenate dictates, but MTBE has begun to be phased out because of groundwater contamination. Accurate experi- mental densities and viscosities for MTBE + hydrocarbons can guide correlations of oxygenates + gasoline. In 1992, Marsh et al. 1 reviewed the thermophysical property measurements (excess volume, vapor-liquid equilibrium (VLE), excess enthalpy, and infinite dilution activity coefficient) of MTBE with different hydrocarbons. It is surprising that only two authors had measured the density of MTBE with heptane and that no data had appeared for MTBE with octane. Pinnick et al. 2 report excess volumes for MTBE + heptane from (243.15 to 333.14) K at pressures of (0.34, 1.72, and 4.85) MPa. Doman ´ska 3 calculates the excess volumes of heptane with MTBE at (298.15 and 308.15) K at atmospheric pressure. Rodrı ´guez et al. 4 has measured densities, the refractive index, and speed of sound of binary mixtures of MTBE + hexane, heptane, octane, and nonane at (288.15, 293.15, and 298.15) K. To the best of our knowledge, experimental viscosity data do not exist for the systems in this paper. In this work, densities have been measured with a vibrating densimeter for the binary mixtures of MTBE + heptane and MTBE + octane over the entire composition range from (273.15 to 343.15) K and from (273.15 to 363.15) K, respectively. Also, the kinematic viscosity of these mixtures has been measured over the whole composition range using a Cannon-Fenske viscosimeter from (273.15 to 313.15) K for the heptane mixture and from (273.15 to 333.15) K for the octane mixture. Excess molar volumes are calculated using Redlich-Kister-type equa- tions, and we have compared our results with predictions of densities from Peng-Robinson (PR) EOS 5 and the Prigogine- Flory-Patterson 6-8 (PFP) theory. The kinematic viscosity has been correlated to a three-body McAllister equation 9 only for mixtures and conditions where component viscosities were measured. We have generated dynamic viscosity values from our kinematic viscosity results using our density measurements. Experimental Apparatus and Procedures. We have described our vibrating tube densimeter (Anton Paar, model DMA 5000) earlier. 10 The repeatability in the density and temperature measurements provided by the manufacturer is ( 1‚10 -6 g‚cm -3 and ( 0.001 K, respectively. The uncertainty of the thermometer and the density measurements is ( 0.01 K on ITS-90 and ( 5‚10 -6 g‚cm -3 , respectively. We believe the uncertainty in the density measurements is less than ( 3‚10 -5 g‚cm -3 . Using a propagation error formula, 11 the uncertainty in the excess volume is less than 0.008 cm 3 ‚mol -1 . The kinematic viscosity is measured using a Cannon-Fenske viscosimeter, size 25, with flow ranges of (0.5‚10 -6 to 2‚10 -6 ) m 2 ‚s -1 . The measurements follow ASTM 445. The viscosimeter resides in a Polyscience constant-temperature water bath controlled within ( 0.01 K. A digital thermometer is used to measure the temperature with an accuracy of 0.01 K. The efflux time was measured manually using a digital stopwatch within an accuracy of 0.01 s. Each datum is an average of at least five runs with a maximum deviation in the kinematic viscosity of ( 0.1 %. The viscosity resulted from multiplying the time by the calibration constant of the viscosimeter and by the density * Corresponding author. Tel: 011 52 461 611 7575. Fax: 011 52 461 611 7744. E-mail address: gais@iqcelaya.itc.mx. 1226 J. Chem. Eng. Data 2007, 52, 1226-1232 10.1021/je600554h CCC: $37.00 © 2007 American Chemical Society Published on Web 05/27/2007