Projects per year
Abstract
In industrial practice, process systems usually consist of multiple units and can be much more complex than a single process unit. The process system, therefore, needs to be designed carefully such that all operational units work synergistically to achieve an overall objective. Designing a process system is a complicated work, which until now is mostly done intuitively. The design is usually developed via a trial-and-error process based on heuristic knowledge. The resulting design may be sufficient and acceptable, but the design may not the optimum design. A more rational and objective approach is needed to design a process system to end up with an objective design that is truly optimal.
This thesis aims to develop such a rational procedure for designing process systems, represented by inhomogeneous nanofiltration cascades for fractionation of fructooligosaccharides (FOS). The design process is approached by modelling within four levels of a process system: (1) single stage nanofiltration, (2) nanofiltration cascades, (3) optimization and (4) process design. This multi-level modelling is elaborated in chapter 2 – 7. Based on these findings, a guideline for designing a process system is given in chapter 8.
Chapter 2 starts off with the multi-level modelling by developing a model for a single stage nanofiltration for FOS. The model was developed based on the steric pore model (SPM), which was then extended to application for FOS by putting more attention on the effect of temperature. Temperature affects the nanofiltration process by following three mechanisms: (1) expanding the solutes, (2) reducing solution viscosity and (3) expanding the membrane pore size.
Chapter 3 discusses the experimental finding that fructose and glucose can be separated during the fractionation of FOS by nanofiltration, even though they have the same molecular weight. Models for nanofiltration do not predict this separation since it considers both sugars as one lumped component: monosaccharides. The finding of this separation enriches the fractionation spectrum in chapter 2, where separation is only based on molecular weights.
Chapter 4 discusses modelling nanofiltration cascades, more specifically systems that produce three products simultaneously. The chapter starts with 3-stage cascade designs that are further improved by addition of extra stages, towards 4- and 5-stage cascades. Addition of stages indeed increased the separation performance. The best performance, however, was found with a 4-stage cascade and not with a 5-stage cascade. This chapter concludes that in addition to adding extra stages, the direction of the expansion is also important. Expansion towards the top section improves the purification of small products (mono-and disaccharides) while expansion towards the bottom section improves the concentration of large product (larger oligosaccharides).
Chapter 5 describes another approach to improve the nanofiltration cascades: adaptation of the stream configurations while keeping the stage number at 3. This chapter confirms the result in chapter 4 related to direction of expansion of the cascade. A model is developed to handle multiple objectives in a design. Based on this model, further analysis was performed to identify the critical parameters for a design.
Chapter 6 describes the development of an optimization model based on mixed integer non-linear programming (MINLP). This model can automatically select the optimum combination of operating parameters in a 3-stage cascade, which could not be done in previous chapters. As the outcome of this model, a frontier curve could be drawn to map the window of operation for optimum performance.
Chapter 7 discusses a method to design inhomogeneous nanofiltration cascades based on the McCabe-Thiele approach, which is a classical design procedure for distillation processes. This method allows us to design an inhomogeneous nanofiltration cascade that can achieve a certain purity target. The models developed before this chapter were only able to predict and optimize the outcome of a given nanofiltration cascade designs, and the outcomes of these models often do not reach a satisfactory target. With the method described in chapter 7, this problem was solved.
Chapter 8 concludes this thesis by highlighting the main findings of each chapter. Extrapolating from the result in nanofiltration cascades, a general design guideline was constructed into a seven-phase design procedures. The chapter concludes with a discussion related to the intuitive choices that are still important in the design procedure, in spite of the great reduction in the subjectivity in the design that follows after the intuitive design choices have been made.
Original language | English |
---|---|
Qualification | Doctor of Philosophy |
Awarding Institution |
|
Supervisors/Advisors |
|
Award date | 3 Nov 2020 |
Place of Publication | Wageningen |
Publisher | |
Print ISBNs | 9789463955362 |
DOIs | |
Publication status | Published - 3 Nov 2020 |
Fingerprint
Dive into the research topics of 'Rational design of cascaded nanofiltration systems'. Together they form a unique fingerprint.Projects
- 1 Finished
-
Design of membrane cascades for fractionation of agro-materials at elevated concentration
Rizki, Z. (PhD candidate), Boom, R. (Promotor), van der Padt, A. (Promotor) & Janssen (FPE), A. (Co-promotor)
15/11/16 → 3/11/20
Project: PhD