naar top
Menu
Logo Print
31/07/2019 - PROF. DR. IR. ANTON A. KISS

NEW HORIZONS IN REACTIVE DISTILLATION

Taking reactive distillation to the next level of process intensification

Reactive distillation (RD) reduces capital investment and saves energy because it can get passed equilibrium limitations, simplify complex processes, increase product selectivity and improve separation efficiency. This article gives a brief overview into novel integrated technologies that combine RD principles with new intensified distillation - e.g. dividing-wall column, cyclic, high gravity and heat-integrated distillation. This leads to new processes and applications.

INTRODUCTION

Process intensification technologies already have a strong positive impact in the chemical process industries [1] and reactive distillation (RD) is one of its biggest successes. It relies on the synergy generated when catalyzed reactions are combined with distillation into a single unit. Today it is considered an established industrial unit operation and the front-runner in the PI field with many applications such as: MTBE, ETBE, TAME, MIBK, acetates, fatty esters, etc. [1-3] For RD to be feasible, there must be a good match between the operating parameters required for reaction and distillation, but this may be limited by properties of the components, catalytic activity and selectivity, and the equipment design. Other parameters that play a role such as mass transfer, residence time and reaction rate can be altered by using different equipment or operating modes. [4]

Despite remarkable developments during the past decades, the potential of RD technology has not been fully realized yet and research is still ongoing to improve it further by various intensification means: e.g. use of dividing walls, cyclic operation, internal heat integration, high-gravity fields, the combination of RD with other operations (membrane separations), or use of alternative energy (ultrasounds or microwaves). [4] This paper focuses only on recent novel RD configurations.

Figure 1: Configurations of reactive distillation (RD) processes: classical RD column (left), azeotropic RD (middle), RD with optional (dashed line) pre- and/or side-reactors (right)
Figure 1: Configurations of reactive distillation (RD) processes: classical RD column (left), azeotropic RD (center), RD with optional (dashed line) pre- and/or side-reactors (right)

 

CLASSICAL REACTIVE DISTILLATION

Figure 1 shows the main configurations of classic RD processes, [2-3] ranging from a conventional setup (for quaternary reactive systems) to azeotropic reactive distillation (a column coupled with a decanter to separate the heterogeneous azeotrope) and RD with pre-reactors and/or side-reactors to increase the residence time. A catalyst may be homogeneous, heterogeneous or absent altogether. A homogeneous catalyst is allowed in practice to leave the RD column in one of the product streams, or it is neutralized and separated as salt waste or separated and recycled, or no catalyst separation or recovery is required.

A special case of catalysis is the use of immobilized enzymes that allow new possibilities for the application of biocatalysts, for example in the production of chiral compounds.

ADVANCED REACTIVE DISTILLATION

Taking reactive distillation to the next level of process intensification requires more advanced configurations, which extend the range (and overlap) of operating conditions beyond those applicable to classical RD. Other variations of RD may include for example integration with other PI technologies or the use of different operating modes, for further energy savings, increased efficiency and environmental friendliness.

Reactive dividing-wall column

RD in a dividing wall column (DWC) was a logical next step developed in academia and applied in industry (Figure 2). [4]

Reports in literature range from rate-based modeling and simulation of R-DWC processes to broad analysis of the R-DWC, its minimum energy demand and potential for energy savings and a comprehensive review on R-DWC. Compared to conventional processes R-DWC can add significant value to new chemical processes with a strong potential to improve production yields, save 15-75% in energy use and over 20% in capital cost.

The design and control of R-DWC can draw on the extensive experience of RD and DWC, respectively.

The key limitations of R-DWC relate to catalyst formulation, hold-up & residence time, pressure drop & flooding, and the need for equal pressure drop on the two sides of the dividing wall. The main potential applications are reversible reactions where the components have suitable boiling point characteristics. Potential R-DWC applications have been reported for the production of various chemicals e.g. methyl acetate, ethyl acetate, dimethyl ether, fatty acid methyl esters (FAME), biodiesel, diethyl carbonate, n-propyl propionate, and others. [5]

Catalytic cyclic distillation

Cyclic distillation uses separate movement of the liquid and the vapor phase. This can be achieved with technology-specific internals and a periodic operation mode (Figure 2). [4] One operating cycle consists of two key parts: a vapor flow period (when the thrust of rising vapor prevents liquid downflow) followed by a liquid flow period (when the liquid flows down the column, dropping by gravity from one tray to the tray below).

This cyclic mode of operation provides key advantages: high throughput and equipment productivity, high separation efficiencies (140-200% Murphree efficiency), reduced energy requirements (20-35% savings), and increased quality of the products. [6] Adding a catalyst on the trays leads to catalytic cyclic distillation (CCD) that is a novel PI approach in reactive separations. [7]

 

Figure 2: Reactive dividing-wall column (left) and internals for catalytic cyclic distillation (right)
Figure 2: Reactive dividing-wall column (left) and internals for catalytic cyclic distillation (right)

 

Figure 2 illustrates the typical internals used for CCD. [4] As the liquid holdup and the amount of catalyst per tray can be significantly greater than in conventional RD systems, applications to slower reactions become feasible as well, thus extending the range of applicability of RD. Key challenges include moving parts (especially in the presence of catalyst particles) and obtaining good mixing and turbulence in the liquid phase to enhance the reaction without damaging the catalyst particles. Adding structured catalytic packing or catalytic internals on the tray may overcome this challenge. CCD is limited to liquid-phase reactions (hence reactions taking place at mild temperatures), but it enables the space in the column to be used efficiently as there is a high ratio of catalyst volume and liquid hold-up to the column space. Potentially CCD has a good range of application (even for slower reactions), but there is a need for development and validation of reliable process modeling approaches for design, control and optimization.

Reactive heat-integrated distillation column

The internally heat-integrated distillation column (HIDiC) is a radical approach for applying a heat pump to assist distillation.

The technology makes use of internal heat integration: the whole rectifying section forms the heat source, while the stripping section acts as a heat sink. The heat required for evaporation in the stripping section is thus obtained from the rectifying section, reducing the heat duty of the reboiler (but shaft work is needed for compression). Overall, HIDiC can achieve primary energy savings (and associated CO2 emissions reduction) of 70-90%. [8-9]

Figure 3: Reactive heat integrated distillation column (left) and Rotating Zig-zag Bed for reactive HiGee distillation (right)
Figure 3: Reactive heat integrated distillation column (left) and Rotating Zig-zag Bed for reactive HiGee distillation (right)

 

The industrial implementation of HIDiC (SuperHIDiC by Toyo Engineering Corporation) is a valuable demonstrator that increases the confidence of industry in the HIDiC technology. [9] Combining HIDiC with RD in a single unit leads to a reactive system (R-HIDiC) with potential for industrial applications (Figure 3).[4]

Reductions in capital and energy costs can be expected. R-HIDiC has potential to be applied to azeotrope-forming mixtures where the reaction can consume the azeotropes.

The simulation of this new technology was reported for the production of tert-amyl methyl ether, ethyl acetate synthesis, and EO hydration to ethylene glycol. Only simulation studies have been reported so far and experimental validation of the process concept is lacking.[10] Design and control can benefit in practice from the 'simplicity' of the SuperHIDiC concept using two discrete heat exchangers, which is a good compromise that provides high energy savings at low investment costs in an operable configuration. The outlook for R-HIDiC is positive as the high energy savings of HIDiC provide strong incentives - although the initial R-HIDiC studies showed only up to 22% savings, compared to classical RD processes. However, the range of applicability (beneficial to equilibrium-limited reactions) is likely to be narrower than that for conventional RD, i.e. close-boiling mixtures (to avoid a high-pressure difference between the rectifying and stripping sections). Additionally, the difference in boiling points and the compression ratio are key limitations on the applicability of R-HIDiC to various reactive systems.

Reactive higee distillation

High-gravity (HiGee) technology replaces the usual gravitational field by a centrifugal field achieved in a specially built rotating device. The high-gravity field (100-1000 g) shifts the flooding limit and allows the use of dense packing materials with a high interfacial area. For rotating packed beds (RPB), HETP values as low as 2 to 8 cm have been reported, while for rotating zig-zag beds (RZB), a volume reduction of a factor of 4 to 7 is claimed for certain distillations (compared to conventional columns). [11]
One could also add a catalyst to convert the system into a reactive HiGee distillation (Figure 3).[4]
The two-stage counter-current rotating packed bed combines the benefits of the RZB with the capability to use packing. It has been assessed (together with conventional RPB) to be the most appropriate equipment to perform RD among all HiGee contractors available.

A model of a novel reactive HiGee distillation process has been developed and then applied for analysis and optimization of a reactive HiGee stripper-membrane process for methyl lactate hydrolysis. [12] Solid-catalyzed reactive stripping has also been reported in a simulation study of the production of octyl-hexanoate with simultaneous water removal from the reaction zone of an RPB. While few studies have included solid catalysts in the (wire mesh) packing, published results showed the potential of HiGee for solid-catalyzed gas-liquid reactions and even for reactive distillation. This potential mainly results from the intensified gas-liquid mass transfer rates and the good catalyst wetting at high gravity.

But prolonged operation at high gravity also results in erosion of the catalyst trapped in the internals (e.g. wire mesh). An important limitation is the very low liquid holdup of RPBs, and the correspondingly low residence times that limit application of RPBs to reactive systems with very high reaction rates. Therefore, reactive HiGee technologies need more research to enhance packing design in order to achieve higher liquid-phase residence times and thus to extend the range of applicability of RPBs. The range of applicability is limited to liquid-phase reactions only and includes fast reactions with competing serial reactions, to achieve enhanced selectivity and ensure low yield losses.

 

"Novel RD technologies that combine the principles of RD with other intensified distillation technologies can take RD to the next level of intensification, leading to promising new processes for the efficient production of chemicals in terms of resources, energy and capital"

 

SUMMING UP

Novel RD technologies that combine the principles of RD with other intensified distillation technologies can take RD to the next level of intensification, leading to promising new processes for the efficient production of chemicals in terms of resources, energy and capital. The potential benefit offered by these novel intensified technologies is significant and motivates accelerated research efforts, especially experimental and pilot-scale studies, as well as advanced dynamic and/or rate-based modeling techniques and methodologies to optimize process design and control.

While all these new RD technologies are promising, they are in different stages of development, with various niche applications typically driving the research. Especially R-DWC (building on the success of DWC) and CCD (pushing cyclic distillation further) have high potential for industrial implementation.

R-HIDiC is also promising but has yet to build on the success of SuperHIDiC for industrial applications. Reactive HiGee could drastically reduce equipment sizes, but it does not offer significant energy savings (on the contrary, it uses additional electricity to drive the rotating equipment) and its range of applicability is limited to fast reactions.

In addition, alternative energies could be also used but there is a need for research to understand the fundamental phenomena by which microwave and/or ultrasound affect the chemicals undergoing reaction and/or distillation.

 

AFFILIATIONS

  • The University of Manchester, School of Chemical Engineering and Analytical Science, Centre for Process Integration, Sackville Street, Manchester M13 9PL, United Kingdom
  • University of Twente, Sustainable Process Technology, Enschede, The NetherlandsTonyKiss@gmail.com, Tel: +44-161-306.8759

 

REFERENCES

  • [1] Harmsen, G. J.; Reactive distillation: The front-runner of industrial process intensification: A full review of commercial applications, research, scale-up, design and operation, Chem. Eng. Process., 2007, 46, 774-780.
  • [2] Kiss, A. A.; Process intensification: Industrial applications, in Segovia-Hernandez, J. G.; Bonilla-Petriciolet A. (Eds); Process intensification in chemical engineering: Design, optimization and control, Springer International Publishing, 2016.
  • [3] Kiss, A. A.; Process intensification for reactive distillation, in Rong, B-G. (Ed); Process synthesis and process intensification: Methodological approaches, de Gruyter, 2017.
  • [4] Kiss, A. A.; Jobson, M.; Gao, X.; Reactive distillation: Stepping up to the next level of process intensification, Ind. Eng. Chem. Res., 2019, 58, 5909-5918.
  • [5] Weinfeld, J. A.; Owens, S. A.; Eldridge, R. B.; Reactive dividing wall columns: A comprehensive review, Chem. Eng. Process., 2018, 123, 20-33.
  • [6] Bildea, C. S.; Patrut, C.; Jorgensen, S. B.; Abildskov, J.; Kiss, A. A.; Cyclic distillation technology - A mini-review, J. Chem. Technol. Biot., 2016, 91, 1215-1223.
  • [7] Patrut, C.; Bildea, C. S.; Kiss, A. A.; Catalytic cyclic distillation - A novel process intensification approach in reactive separations, Chem. Eng. Process., 2014, 81, 1-12.
  • [8] Olujic, Z.; Fakhri, F.; De Rijke, A.; De Graauw, J.; Jansens, P. J.; Internal heat integration - The key to an energy-conserving distillation column, J. Chem. Technol. Biot., 2003, 78, 241-248.
  • [9] Kiss, A. A.; Olujic, Z.; A review on process intensification in internally heat-integrated distillation columns, Chem. Eng. Process., 2014, 86, 125-144.
  • [10] Vanaki, A.; Eslamloueyan, R.; Steady-state simulation of a reactive internally heat integrated distillation column (R-HIDiC) for synthesis of tertiary-amyl methyl ether (TAME), Chem. Eng. Process., 2012, 52, 21-27.
  • [11] Cortes Garcia, G. E.; van der Schaaf, J.; Kiss, A. A.; A review on process intensification in HiGee distillation, J. Chem. Technol. Biot., 2017, 92, 1136-1156.
  • [12] Gudena, K.; Min, T. H.; Rangaiah, G. P.; Modeling and analysis of novel reactive HiGee distillation, Comput. Aided Chem. Eng., 2012, 31, 1201-1205.