Numerical Modelling and Experimental Validation of Mass Transfer in a Structured Packed Distillation Column Using Nigerian Crude-Derived Fractions

The design optimisation of packed distillation columns for Nigerian crude oil fractionation is constrained by a limited experimental database on mass transfer efficiency for locally produced crude-derived feed mixtures. This study develops a numerical model for mass transfer in a structured packed column using binary and ternary mixtures representing light and medium Nigerian crude fractions, validated against bench-scale experimental data from a glass distillation column. The model employs the rate-based approach incorporating Maxwell-Stefan diffusion equations, vapour-liquid equilibrium data from the NRTL activity coefficient model parameterised from Nigerian crude compositional data, and structured packing correlations for effective interfacial area and mass transfer coefficients. Experiments were conducted at atmospheric pressure for C7/C8 hydrocarbon binary systems and a representative kerosene-range ternary system, measuring concentration profiles by gas chromatography. Model predictions of separation efficiency and height equivalent to a theoretical plate agreed with experimental data within 6.3% average relative error across the range of vapour and liquid loads tested. Sensitivity analysis identified the effective interfacial area correlation as the dominant source of model uncertainty. Parametric studies using the validated model showed that increasing the vapour load factor from 1.5 to 2.5 improved column efficiency by 14% for the ternary system. The validated model provides a reliable tool for packed column design in Nigerian petroleum refinery applications. Keywords: mass transfer, structured packing, distillation modelling, Nigerian crude, rate-based model