Modern trends in large-scale industrial systems for mass-transfer operations have resulted in an evolution from tray and random packing mass-transfer devices towards the use of ordered packing mass-transfer devices. This is true throughout the petroleum refining, petrochemical, chemical and gas processing industries. One of the most important determinants of successful application of ordered packing devices is the quality of liquid and vapor distribution to the packing. Industrial practice has been to use an arbitrary measure of flow quality for equipment design, fabrication and installation. The commonly used value is up to ~5% flow variation from point-to-point across a liquid distribution system. No analytical work is available in the literature that quantifies the impact of flow maldistribution on mass-transfer systems based on the fundamentals of distillation. Some experimental work on small diameter systems has been published. The assumed distribution quality required is critical for setting manufacturing and installation tolerances for distributors.
Work executed to quantify the flow maldistribution impact on medium volatility (alpha) distillation systems is presented. Simulation techniques are shown for modeling the impact of maldistribution on mass-transfer in the system. The basic techniques involve modifying a standard flowsheet simulator's rigorous distillation unit operation module. The approach taken splits the fractionator into a series of linked rectifying-stripping sections. Each section uses from one to N parallel modules. One module implies perfect distribution. Multiple modules, each with discrete vapor to liquid ratios, are used to emulate the impact of flow maldistribution on distillation. The number of parallel modules, N, at each stage is a function of the expected maldistribution profile across the fractionator cross-section. A graphical flowsheet is provided of the modules used.
Application of the technique to a C3/C4 splitter (cumene feed stock) is shown. For the system, the maximum tolerable flow distribution is evaluated as a function of unit design parameters: equilibrium stages, reflux ratio and maldistribution to the rectifying or stripping section.
Electronic version available in Adobe PDF format file 029.PDF 1902k.
Request paper 029.
* Current affiliation.