Finished research topics

Membrane processes provide a tremendous potential for improving separation performance and energy efficiency compared to thermal separation processes. However, the benefits of membrane processes, as e.g. the ability to overcome limitations of other separation techniques introduced by azeotropes or eutectic points, are exploited best when integrated in hybrid processes. In hybrid separations, two or more unit operations, based on different separation phenomena, are combined that lead to more efficient and sustainable processes. Nevertheless, industrial applications are still limited due to a lack of reliable models that allow for an accurate description of the separation performance of the membrane. Process design approaches oftentimes neglect limiting effects, such as pressure drop, or concentration and temperature polarization.

In order to overcome those limitations, a systematic five-step design method evaluating the potential of membrane-assisted hybrid processes is developed at the chair of fluid separations. Current investigations are focused on pervaporation-assisted distillation and organic solvent nanofiltration. Therefore, the developed design approach includes driving force reducing effects as well as means for process intensification, such as energy integration, in order to evaluate the overall potential of the process. Additionally, the correct consideration of the flow pattern, such as co- and counter-current as well as cross-flow, can have a significant impact on the accuracy of the model that will also be considered within this work. The aim is to evaluate the potential of membrane-assisted processes prior to any experimental effort.

Project: SFB/TR 63 Inprompt – Transferprojekt "Hybride Trennprozesse: Modellierung und Entwurf von membrangestützter Rektifikation"

Contact Person :  Bettina Scharzec, Mirko Skiborowski

Process design in general requires accurate models for dimensioning of separation units, i.e. for the design of distillation columns. At present process engineers commonly use equilibrium (EQ) models for column dimensioning. However, for some separation tasks using the EQ models leads to considerable uncertainties and possible design errors. In these cases so called non-equilibrium (NEQ) models may be used to decrease the uncertainties. NEQ models consider the actual mass and energy transfers within the column and thus, allow correct design and dimensioning of columns even for the separation of highly non-ideal mixtures. However, currently no reliable criteria are available for the selection of the necessary modeling depth – EQ or NEQ model – for a specific separation task. At the laboratory of fluid separations, the need for NEQ modeling for distillation processes is systematically investigated.

Moreover, optimization-based methods for the design of energy efficient distillation processes are developed. Conventional distillation processes suffer from low energy efficiency. The energy efficiency can be significantly improved by using alternative process variants such as heat-pump assisted and thermally coupled distillation columns. The developed design methods are based on rigorous models and can consider different process configurations, e.g. dividing wall columns. These methods are extended step by step and are applied for the design of difficult separation tasks such as extractive or azeotropic distillation processes.

Project: SFB/TR 63 Inprompt – Transferprojekt "Hybride Trennprozesse: Modellierung und Entwurf von membrangestützter Rektifikation"

Contact Person :  Thomas Waltermann, Mirko Skiborowski

Membrane processes have been attracting more and more attention in industry. Because of different interactions between components and the membrane material some components can permeate through the membrane preferably which leads to a selective separation. A new but promising technology is organic solvent nanofiltration which can be operated in liquid state without phase transitions. Organic solvent Nanofiltration can be employed for the retention of different molecules like catalysts or pharmaceutical products, since the operating temperatures are not necessarily high.

Due to complex interactions between membrane, solvents and solutes varying for each membrane type and chemical system, a high number of experiments needs to be performed in order to identify suitable membranes for a given separation task or even for feasibility studies. In order to reduce the experimental effort, a new approach for predictions of the separation behavior of organic solvent nanofiltration membranes is developed. Based on a smaller number of experimental measurements a new shortcut model is developed following the procedure in the graphic below. For identifying the new model structure a data driven approach is used.

Project: ESIMEM "Energy efficient Separation in the chemical and pharmaceutical Industry using Membrane processes"

Contact Person :  Rebecca Goebel, Mirko Skiborowski

This project aims on the cost-effective purification of biosurfactants on the example of sophorose lipids. These can be produced by microorganisms, based on renewable resources, such as sugar and oil. Apart from the non-petroleum feedstock, sophorose lipids show good surfactant properties and are biodegradable.

An early consideration of necessary purification steps and their interactions with the fermentation during process synthesis contributes to a cost-efficient and therefore competitive production. Due to the complexity of biological mixtures and unpredictable behaviour, experiments are essential for feasability studies and measurement of necessary model parameters. Additionally the sustainablity of different process options should be considered when selecting the optimal process.

Contact Person :  Andreas Weber, Tim Zeiner

Over the past thirty years biotechnological products from the white and red biotechnology are gaining continuously increasing importance. Nevertheless, the widespread application of these products – mostly proteins – is often hampered by the lack of effective and economic purification processes. Chromatography as state of the art purification technology faces certain bottlenecks regarding capacity and scale-up. Aqueous two-phase systems (ATPS) have shown promising selectivity for the purification of proteins while extraction is known to offer large capacity and easy scale-up. ATPS form two liquid phases when two hydrophilic solutes (e.g. a polymer and a salt) are mixed at adequate concentrations. Both liquid phases consist mainly of water and therefore offer mild extraction conditions. Due to the multi component character of ATPS for protein purification high experimental efforts to design an optimal process result in scarce industrial application. From chemical engineering it is well known that process modeling significantly reduces process design efforts.

Purification of antibodies

In this research project an ATPS comprised of Polyethylene glycol, a phosphate salt and water is applied for the extraction of antibody (AK1). Multi-stage extraction experiments are performed in a Mixer-Settler unit and the results are applied for the validation of an equilibrium stage extraction model. The simulation of a multi-stage extraction is a powerful tool to simplify the design of extraction of antibodies in aqueous two-phase systems mainly reducing experimental efforts.

Contact Person: Jan Mündges, Tim Zeiner

Purification of enzymes

Within this project a process model for protein purification using ATPS is developed and experimentally validated. The enzyme Laccase from a Pleurotus sapidus supernatant is used as a reference system for the protein purification model. The model is validated experimentally in a multi-stage mixer settler extraction unit.

Contact Person :  Axel Prinz, Tim Zeiner

The aim of the research project is to develop and to verify a simple and reliable design methodology for membrane-assisted reactive separation processes. Those processes have the potential to increase ecological and economical sustainability but there is still a lack in process design tools and process know how which is responsible for the limited industrial use of these processes. The supposed design methodology consists of the three steps (1) flowsheet generation, (2) flowsheet evaluation as well as flowsheet choice and (3) final flowsheet optimization.

The process design methodology is applied to the chemical-equilibrium transesterification of propylene carbonate with methanol producing dimethyl carbonate and 1,2-propanediol. The optimal flowsheet that is determined in step (2) consists of reactive distillation, vapour permeation and conventional distillation for the final purification of dimethyl carbonate. Experimental validation for the unit-operations reactive distillation and vapour permeation as well as for the whole process will be performed. Furthermore the final optimization of the membrane-assisted reactive separation process and economic comparison to the benchmark process consisting of one continuous reactor and four distillation columns will be done.

Contact Person :  Johannes Holtbrügge, Philip Lutze

Due to the scarcity of fossil raw materials, the development of processes using bio-based resources is currently investigated. A main challenge of using bio-based feedstocks is the complexity of the impurity profiles. To account for this issue, design approaches are needed or the production of bio-based chemicals to allow an integrated design of upstream and downstream processes. By the help of these approaches, critical impurities affecting the downstream processing can be identified. Based on these studies, first predictions can be made to define, which impurities should be avoided or reduced in the upstream processing or the purification of the reactants. Hence, this thesis presents a design approach for using bio-based feedstocks, exemplarily applied to reactive distillation processes. The developed approach is used to identify the most critical impurities with respect to the purity of the final product. Based on thermodynamic and physical property data analyses and reactive distillation process simulations, critical impurities are identified among all possible bio-based impurities. Subsequently, a process analysis is performed to determine an operating window of the reactive distillation process that still allows a production with the desired specification for the product. Furthermore, possible process alterations are studied. A workflow of the developed design approach. As it can be seen in the figure, the number of investigated components strongly varies throughout the single steps.

Contact Person :  Alexander Niesbach, Philip Lutze

The productivity of fermentative biobutanol production is limited due to the high toxicity of n-butanol to the production strains. Pervaporation offers the opportunity for energy-efficient and selective product removal facilitating continuous fermentation leading to an increase in productivity. Because of low product concentrations up to 2 wt.-% in fermentation broths, the driving force for pervaporation is rather small, resulting in either low permeate fluxes or low selectivity for n-butanol. To overcome this limitation, a promising new class of membranes called supported ionic liquid membranes (SILMs) are examined. These membranes combine the advantages of membrane processes with characteristics of ionic liquids suitable for extraction of n butanol. The permeation properties of a SILM can be adjusted by changing the immobilised ionic liquid and can thus be adapted to the requirements of the separation problem. The corresponding process will be modelled and simulated based on the experimental results.  According to the results of simulation studies, the properties of the investigated membranes can be adapted to the requirements of the n-butanol recovery process. The goal is to identify the combination of ionic liquid and membrane which is best suited for the separation by pervaporation of n-butanol.

Contact Person :  Sebastian Heitmann, Philip Lutze

The goal of this research is to develop and optimise new, energy-efficient hybrid processes combining organic solvent nanofiltration (OSN), melt crystallisation and distillation. In contrary to distillation, the design methods for OSN and melt crystallisation are not well established, demanding great experimental effort for model parameterisation. In order to minimize this effort, a four step design method is developed. In the first step, different process variants are generated. In the second step, the variants are evaluated using rigorous models, wherein the unknown model parameters are varied to quantify their influence on the process performance. If hybrid separations are promising, experiments are performed to determine the unknown parameters in the third step. In the last step, an optimisation is performed to find the optimal process. The developed tools and the feasibility of the approach are illustrated in case study of separation of close and wide boiling mixtures from hydroformylation.

Contact Person :  Jovana Micovic, Philip Lutze

The research project aims at a rigorous model-based process design methodology for Organic Solvent Nanofiltration (OSN) membrane module cascades. In contrast to most of the existing design methodologies mainly applied for gas separation, a detailed mathematical modelling of the permeation mechanism through polymeric and ceramic OSN membranes is conducted, followed by a rigorous economic optimisation of the process.

The process design methodology is applied to the recycling of homogeneous catalysts during hydroformylation of C5-C8 olefins. Apart from the membrane separation, also the reaction and product purification are taken into account (see Figure 1). Experiments are conducted at project partners within the F3 Factory project. Supporting information on the membrane permeation mechanism through polymeric membranes is gathered by experiments in a cross-flow OSN plant, focusing on rejection of different chemical compounds within single solvents or solvent mixtures.

Contact Person: Patrick Schmidt, Philip Lutze

Instead of hydroformylation of crude-oil based propylene, 1-butanol can be produced from biomass by acetone-ethanol-butanol fermentation. It can be used as a chemical intermediate or as an end product, e.g. biofuel. With its mild conditions liquid-liquid extraction is a unit operation with high potential to compete against distillation. However, conventional organic extracting agents exhibit unfavourable properties, e.g. they are volatile, hazardous or show insufficient extraction performance. A possibility to substitute these organic solvents is the use of ionic liquids. Since ionic liquids have a negligible vapour pressure their recovery can be achieved by stripping or distilling off the solutes. Therefore, in this research project the extraction of 1-butanol from dilute aqueous solution using ionic liquids is investigated.

Contact person :  Martin Stoffers

Post combustion capture of CO₂ using amines is one of the most attractive methods to reduce the CO₂ emissions from coal-fired power plants and petroleum refineries. To design industrial scale absorption columns, rigorous and reliable mathematical models as well as extensive experimental study are necessary. The reaction kinetics and thermodynamics for the CO₂ absorption with amines has been extensively studied over the last decade.

In the current PhD project, an experimental investigation of mass transfer kinetics of reactive absorption process and validation of rate-based model has been planned. For the experimental study, a new in-house absorption mini-plant comprising of three absorption columns with variable diameters (DN25, DN50, DN100) and heights (1m to 4 m) will be used. The mini-plant will be initially benchmarked using hydraulics and physical desorption experiments. Consequently, the reactive absorption of CO₂ will be studied in the mini-plant. A generic rate-based model will be validated using the experimental results from the pilot scale absorption column. Furthermore, an optimization study using the rate-based model will be carried out by applying an evolutionary algorithm to design reactive absorption column.

Contact Person :  Chinmay Kale, Philip Lutze

1,3-Propanediol (PDO) is a bulk chemical whose demand is annually increasing. It is mainly used as monomer for the manufacturing of the polymer polytrimethylene terephthalate which possesses unique properties as a result of its odd number of carbon atoms. The application of high temperatures and high pressures during the chemical production of PDO makes the biotechnological route with its moderate conditions an attractive alternative. However, the downstreaming of the very polar PDO is still challenging. Thus far, extraction has not been considered as potential separation method because of the extremely low solubility of 1,3-propanediol in organic solvents. Hence, this research project focuses on the application of ionic liquids for the extraction of 1,3-propanediol to overcome this limitation.

Contact person :  Anja Müller

Reactive distillation (RD), in which chemical reaction and distillation are integrated into one single apparatus, is a well known alternative technology to conventional process with sequential reaction and distillation steps. Currently, the main application of RD at commercial scale is the production of high-octane gasoline components such as MTBE. However, due to environmental concerns, MTBE has been banned in several countries and a growing interest exists to produce alternative components to MTBE.

The aim of my PhD is the modeling and simulation in both steady and dynamic state as well as the design of RD columns with focus in the production of fuel additives such as tertiary ethers and dialkyl carbonates.

Contact Person :  Carlos Rugerio, Philip Lutze

The separation of azeotropic multicomponent mixtures is an inherent part of biochemical and chemical processes and still remains a challenging issue in terms of process design and identification of selective separations. A promising process concept for these separations is combination of different unit operations to create a hybrid process. By coupling, e.g., distillation with membrane separations, the advantages of both processes – the high throughput associated with distillation and the high selectivity of membrane separations – are combined and the limitations of at least one of the unit operations involved can be overcome. This classic example of using synergies for process intensification can lead to significant energy savings and cleaner processes due to the avoidance of entrainer. Nevertheless, this significant economic and ecological potential is hardly exploited in industry due to a lack of general process design methods and missing detailed process know-how. To promote the industrial application of membrane-assisted separation processes, an optimisation-based design method for membrane-assisted separation processes comprising distillation, pervaporation and vapour permeation is developed in this study.

Contact Person :  Katharina Koch

The focus of this research is the application of Membrane Adsorbers (MA) for downstream processing of biological active compounds. With an automated HPLC-system (Äkta) membrane parameters - such as membrane capacity - are investigated. With a validated detailed simulation tool the performance of MA unit operations over complete purification cycles, consisting of loading, washing and elution over wide ranges of operating conditions is evaluated. Process analysis, scale-up and optimization are performed to identify optimal MA configurations and process parameters. Furthermore the focus lies on improving downstream processes by the use of membrane chromatography additionally to the established unit operations.

Contact Person :  Mayuratheepan Puthirasigamany, Tim Zeiner