ISBN 978-3-8439-2017-9

978-3-8439-2017-9, Reihe Verfahrenstechnik

Jovana Mićović
Hybrid separations combining distillation with organic solvent nanofiltration and melt crystallisation

170 Seiten, Dissertation Technische Universität Dortmund (2014), Softcover, A5

Zusammenfassung / Abstract

Combination of distillation and energy efficient unit operations, e.g. organic solvent nanofiltration (OSN) or melt crystallisation, to integrated hybrid separation processes promises tremendous synergies and potential for increasing resource efficiency and decreasing operating and capital costs. Design of such processes is challenging due to large number of structural and operational degrees of freedom and to limited availability of models to predict membrane or crystallisation performance. The a-priori experimental determination of all parameters is costly and time consuming, without any guarantee that the hybrid process will outperform the conventional separation process such as stand-alone distillation.

To address these challenges, in this thesis a four-step design method for hybrid separations is developed, which can be applied in early process development stage when only little or no experimental data is available. In the first step, different process options are generated based on heuristics and engineering judgement and screened for feasibility. In the second step, options are evaluated based on quantitative metrics using detailed models. For this purpose parametric optimisation is applied to quantify the influence of the unknown parameters on the process performance (e.g energy or investment). In the third step, only if hybrid separations are promising, experiments are performed for model parameterisation and validation. Thus, experimental effort is reduced, as experiments only need to be performed in the operating window, in which the hybrid process outperforms the stand-alone separation. In the last step, an optimisation is performed to identify the best (cost optimal) process.

The applicability of the method and developed tools is demonstrated in two case studies, which are the separation of heavy-boilers from a hydroformylation mixture and the separation of a close boiling, isomeric mixture of long chained aldehydes. The required process models and tools for optimization under uncertainty are developed and presented. In addition, necessary experiments are planned and performed.