Browsing by Author "Momoh, Chris Favour Ojonugua"
Now showing 1 - 1 of 1
Results Per Page
Sort Options
- ItemPhase behaviour and thermodynamic modelling of carbon dioxide + (1-octanol or 1-decanol) + large n-alkanes(Stellenbosch : Stellenbosch University, 2023-03) Momoh, Chris Favour Ojonugua; Schwarz, Cara Elsbeth; Stellenbosch University. Faculty of Engineering. Dept. of Process Engineering.ENGLISH ABSTRACT: Large alcohols are useful in the synthesis of detergents, surfactants and plasticisers. However, the product stream of these processes often contains an impurity of large n-alkanes. Azeotropic distillation has been the mainly used process to separate the 1-alcohol product from the n-alkanes. However, due to the process requirements, health and safety risks and extreme operating conditions, investigations have been carried out into utilising an alternative process for this separation. Supercritical CO2 fractionation has been proven to be a feasible method for the separation of detergent range n-alkanes and alcohols because it serves as a cheaper, less toxic method to azeotropic distillation. However, there are notable solute-solute interactions exhibited by ternary systems containing CO2, n-alkanes, and 1-alcohol. The main aim of this work was to experimentally identify and analyse the solute-solute interactions that occur in ternary systems containing supercritical CO2 as a solvent and a solute mixture of 1-alcohols (C8 or C10) and n-alkanes (C10 – C20). The secondary aim of the study was to evaluate the capability of the RK-Aspen thermodynamic model to predict the phase transition data. The solute-solute interactions exhibited in these systems were identified and characterised by analysing their high-pressure bubble- and dew-point (HPBDP) data. Literature revealed that there is, however, a lack of data for the systems relevant to this study. HPBDP data for the ternary systems containing CO2 + 1-alcohols + n-alkanes were thus measured at temperatures ranging from 308 K to 358 K. The phase behaviour data were measured at solute mass fractions ranging from 0.015 to 0.65, with pressures up to 30 MPa, using the synthetic visual phase detection method. The experimental HPBDP data revealed that co-solvency was present in all the ternary systems measured. The systems exhibiting the most significant co-solvency effects were CO2 + 1-octanol + n-hexadecane and CO2 + 1-decanol + n-octadecane. The measured data highlighted that the influence of the n-alkanes on the solute mixtures decreased with an increase in molecular mass. The experimental data revealed that co-solvency occurs in alkane-rich mixtures for systems containing C10 – C14 alkanes while the co-solvency effects for systems containing larger alkanes are more prominent for alcohol-richer mixtures. The range of solute mixture compositions and temperatures investigated did not reveal the effects of other complex phase behaviour phenomena such as miscibility windows and l-l-g holes; these may be present at other conditions for these systems. The evaluation of the RK-Aspen model revealed that the model correlated well with the HPBDP data at higher temperatures (T > 328 K). Furthermore, the addition of experimentally determined solute-solute binary interaction parameters improved the ability of the RK- Aspen model to predict the data. The %AADP for the ternary systems investigated improved from a range of 3.95 % - 17.41 % to a range of 1.65 % - 5.99 % when the solute-solute BIPs were included. The RK-Aspen method also struggled to predict the phase behaviour for systems containing larger n-alkanes as these systems exhibited significant temperature inversions.