Modeling of extraction operations for caprolactam requires an accurate description of the phase compositions and the physical properties of all liquid phases. Therefore experimental liquid-liquid equilibrium data were measured for the systems water (1) + caprolactam (2) + ammonium sulfate (3) + organic solvent (4) at 293 K, 313 K, and 333 K, where the organic solvent was either benzene or toluene. Furthermore, the physical properties, density and viscosity of the separate phases and the interfacial tension of the two-phase systems, were determined. The equilibrium phase composition data were correlated using the nonrandom two-liquid model. Density data were fitted via the concentration influence of the solutes and the thermal expansion coefficient. Viscosity data were correlated using the DoleJones equation for describing the influence of the solute concentration and the Guzman-Andrade equation for describing the influence of temperature. Interfacial tension data were described using the Szyzkowski and the Jasper equation for describing the influence of solute concentration and temperature. Excellent agreement was found between the measured data and developed correlations.

No abstract

No abstract

A fully automated workstation has been developed and applied to the measurement of liquid−liquid equilibrium. The measurements, starting from solution preparations through equilibrium establishment and chemical analysis of both phases and ending with the determined solute concentration in both phases, can be done on a 24 h per day basis with no human intervention at all. To evaluate the accuracy of the measurements, the equilibrium distribution ratios of acetone between toluene and water at 20 °C and of caprolactam between toluene and water at 40 °C and between benzene and water at 20 °C are measured and the results are compared with the values reported in the literature and, for the caprolactam systems, also with those obtained in a jacketed equilibrium cell. Furthermore, the distribution ratio of caprolactam between toluene and water has been measured several times for the same initial caprolactam mass fraction in order to evaluate the repeatability. The obtained results are in very good agreement with the literature values for the acetone system and with those measured in the glass cell for the caprolactam systems, while a coefficient of variation of 1.8% for a seven times repeated measurement was found.

Caprolactam is obtained by extraction using organic solvents like benzene, toluene, or chlorinated hydrocarbons. As an alternative solvent, the mixed solvent heptane-heptanol (40 mass %) was selected in previous studies based on a relatively high distribution ratio of caprolactam, a low mutual solvent solubility, beneficial physical properties, and a low distribution ratio of impurities. Now, the hydraulic and mass transfer characteristics of the extraction of caprolactam in a pulsed disc and doughnut column (PDDC) were investigated using the benign solvent. The results were compared to those for toluene.The PDDC showed qualitatively comparable operational characteristics for both solvents. In the hydraulic experiments the mixed solvent showed smaller drop diameters and hold-ups, required lower pulsation intensities for regime transitions, but the operational windows are slightly smaller. For both solvents, mass transfer resulted in increasing drop diameters and pulsation intensities required for regime transition. In the forward extraction the mixed solvent was superior, where HETS/m ¼ 0.26 to 0.37, compared to 0.42 to 0.67 for toluene, while less theoretical stages are required as well. For the back-extraction HETS/m ¼ 0.33 to 0.40 for the mixed solvent compared to 0.30 to 0.37, but toluene requires the lower amount of theoretical stages.The hydraulic characteristics at equilibrium and concentration profiles in both the forward and back-extraction were described accurately using the developed models.

Benign solvents were selected and characterised on labscale as alternative for organic solvents like benzene, toluene, or chlorinated hydrocarbons for the extraction of caprolactam from the Beckmann-rearrangement mixture. The performance of the best candidate solvent was evaluated in a 4 cm diameter pilot scale pulsed disc and doughnut extraction column. Non-polar solvents are unfavourable for caprolactam extraction, because of its low partition ratio. These trends are opposite for polar solvents. As alternative solvents, mixtures containing a polar solvent (40 % heptanol) and an alkane (methylcyclohexane or heptane) were proposed, based on a high capacity and a low mutual solvent solubility. Relative to benzene an equal amount of mixed solvent is needed for the forward and back-extraction for a benchmark industrial process lay-out. The number of theoretical stages NTS for the mixed solvents and benzene is 5 and 9 in the forward and 10 and 5 in the back-extraction, respectively. The solvent mixture heptane/ heptanol was finally selected as best alternative solvent based on its beneficial physical properties (see Fig.).Pilot plant experiments with the mixed solvent were performed to determine the hydrau- Figure. Weight fraction of caprolactam in the raffinate phase as a function of the number of theoretical stages NTS for the mixed solvent (heptane/40 % heptanol) and benzene in forward extraction (a) and back extraction (b) at 40 C.