Community Research and Development Information Service - CORDIS

FP7

ALTEREGO Report Summary

Project ID: 309874
Funded under: FP7-NMP
Country: Netherlands

Final Report Summary - ALTEREGO (Alternative Energy Forms for Green Chemistry)

Executive Summary:
The project work started in three research work packages (WP) dealing with development of the three technologies and led to two further ones focused on the application and demonstration for the pharmaceutical synthesis, and green fuels and bulk chemicals respectively.
In WP1, tailored equipment was developed for the ultrasound (US)-assisted processes in the areas of advanced pharmaceutical and green fuels and bulk chemicals synthesis. Three types of multiphase processes were chosen as case studies: reactive solvent extraction, reactive synthesis and cooling crystallization, and reactive distillation (RD) systems. The work involved enabling local positioning of the US energy at the interfaces (liquid/liquid, solid/liquid, gas/liquid); develop mechanistic understanding and models for the US-assisted operation of the above-cited processes; design an efficient US-assisted continuous reactor and demonstrate the developed technology for the specified processes. Regarding reactive solvent extraction, reactors for efficient US transfer were designed and enabled threefold yield increase for a specific solvent extraction reaction. Regarding cooling and reactive crystallization, it was found that US clearly reduces the nucleation induction time and metastable zone width as well as crystal size. US effects on separation of binary systems involving methanol were insignificant. Finally, while US can improve reaction kinetics of enzymatically catalyzed reactions, the combination of the chemical system and the form of enzyme immobilization play a vital role in the feasibility and effect of US application to RD.
In WP2, work was done i) to perform VLE measurements under microwave (MW) for RD processes, ii) to identify suitable reaction-catalyst systems for the RD and an API synthesis and iii) to identify the mechanisms behind the MW effect on different process in order to allow for design of complex equipment. For the investigated RD chemical system (DMC/ethanol transesterification), kinetic enhancement under MW and different homogenous catalysts was seen at T>85°C. No suitable heterogeneous catalysts were found. The influence of MW on VLE was not verified for the system under investigation and simulations revealed that no significant improvement for MW enhanced RD would be expected. Further, the effect of MW on evaporative crystallization was studied and it was shown that MW induce faster solvent evaporation favoring crystal nucleation instead of growth and resulting in smaller crystals with narrower size distribution compared to conventional evaporative crystallization. MW-assisted cooling crystallisation was also pursuit and demonstrated ca. 50% reduction in process time and narrower crystal size distribution due to the fast response of MW heating. WP2 focussed primarily on detailed kinetic investigations for three reaction types, esterification, transesterification and a demethylation reaction; 1.5-5 times higher reaction rates were observed under MW depending on the temperature.
In WP3, the objective was to develop prototypes for methanol synthesis from CO2 and biomass with microwave plasma (MWP) technology. In this context, the reverse water gas shift (RWGS) reaction (first step to methanol production from CO2) was studied. Two setups were developed. Regarding RWGS, a bench-scale microwave plasma reactor was developed. In the case of biomass gasification, a previously built containerized MWP gasifier (10-20 kWth) was further developed within ALTEREGO. Regarding RWGS, plasma treatment enables very high (superequilibrium) conversions, compared to the conventional process, without catalyst and byproduct formation. Regarding gasification, biomass conversions up to 85% were obtained at cold gas efficiencies (CGE) of ~40%. The product composition was close to equilibrium at the outlet temperature; this gives certainty that if the reactor becomes thermally insulated, CGEs>80% will be attained.
In WP4, the objective was the implementation/validation of the US- and MW-based technologies in a pharmaceutical environment. Additionally, the technologies were evaluated economically and a roadmap for implementation was set up. WP5 included the implementation/validation of all energy-based technologies in academic lab/pilot environment. Technical-economic evaluation of the processes, comparison with the current conventional ones and development of a roadmap for industrial implementation were delivered. Based on the results and the decision of the consortium, the prior case study of the MW-RD was adapted to an US-assisted enzymatically catalysed RD for butyl butyrate production.

Project Context and Objectives:
Rationale
Alternative energy sources (microwave, plasma, ultrasound, electric fields, light) are considered novel key methods with high potential for intensification of chemicals syntheses in terms of energy and resource efficiency. Nevertheless, they are not yet adequately applied in the chemical process industry. This is due to existing gaps in scientific and engineering understanding of the mechanisms underlying alternative energy-based syntheses and processes and in the interaction of various process and equipment design parameters determining performance.
General concept
The general aim of ALTEREGO was to develop a hierarchical methodology for targeted supply of three alternative forms of energy (ultrasound, plasma and microwaves) in novel reactors to precisely control chemical transformations and reaction pathways. It was based on fundamental mechanism understanding, advanced modelling, experimentation, and model validation and envisaged creating a reaction environment in which the right type of energy is transferred selectively from the source to the target molecules in the required form, in the required amount, at the required moment, and at the required position.
Unfortunately, current chemical reactors still offer a very limited degree of control of molecular level events. This is due to conductive heating, which is conventionally applied to bring more molecules at the energy levels exceeding the activation energy threshold. However, conductive heating offers only macroscopic control upon the process and is thermodynamically inefficient. It is non-selective in nature, which means that non-reacting (bulk) molecules heat up together with the reacting ones. Also, other elements of the reactor are unnecessarily heated up. Secondly, conductive heating generates temperature gradients, which creates a broad Maxwell-Boltzmann distribution of molecular energy levels.
In this project we demonstrate a number of alternative energy based prototype solutions with process benefits, compared to conventional processes, owing to two general distinct characteristics of alternative energy forms: 1) selective interaction with materials and 2) precise spatial and temporal control of energy dosing. The process benefits that are anticipated to be harnessed are:
Ultrasound
Ultrasound (US) has been investigated as a way to enhance reaction kinetics and mass transfer and to control the formation of primary or secondary nuclei in crystallization processes. Organic synthetic reactions show increased rate (sometimes even from hours to minutes, up to 25 times faster) and/or increased yield (tens of percent, sometimes even starting from 0% yield in non-sonicated conditions). In multiphase systems, gas-liquid and liquid-solid mass transfer has turned out to increase 5 and 20-fold, respectively. Additional benefits in crystallization reactions include more uniform crystal size distribution and the selective synthesis of certain polymorphs. Finally, US energy has proven its potential to manipulate vapour-liquid equilibrium (VLE) in order to break azeotropes and improve separation efficiency. This can be taken advantage of in distillation based processes.

Microwaves
Microwave (MW) energy provides an efficient means of heating of microwave absorbing materials, such as polar molecules and metal nanoparticles; it has proven to significantly accelerate (up to several orders of magnitude) various chemical syntheses and to increase products yield. Furthermore, MWs (in analogy to US) can also shift vapour-liquid-equilibrium composition and to increase separation efficiency in distillation processes.
Non-thermal plasma
Non-thermal plasma (NTP) generated by high-intensity microwave fields will be explored as a technology to intensify gas-phase catalytic reactions. Synergy can occur when non-thermal plasma is combined with a heterogeneous catalyst, as each affects the other physically and chemically. The resulting combined system can exhibit higher energy efficiencies, better conversion to desired product and decreased formation of by-products than either NTP or thermal catalysis alone. NTP can activate the catalyst at temperatures too low for thermal catalysis to occur. Catalysis can lower the energy requirements of a plasma reactor while speeding up the destruction of unwanted or hazardous by-products . Catalytic processes are often more selective than plasma-induced ones, but tend to be more stringent in their requirements for a controlled gas composition and temperature .
Project objectives
The project specific objectives were to develop:
• Novel gas phase catalytic chemical syntheses activated by microwaves and non-thermal plasma with the specific application to be: methanol synthesis from carbon dioxide and hydrogen in micro- or millireactors.
• Novel liquid phase chemical syntheses activated by ultrasound (US) and microwaves (MW) with the specific application to be:
o Ethyl methyl carbonate (EMC) and Diethyl carbonate (DEC) synthesis from methanol activated by microwaves and ultrasound in reactive distillation columns.
o Paracetamol synthesis through reactive and cooling crystallization activated by ultrasound in continuous flow milliscale tubular reactors and Oscillatory Flow Reactors (OFR).
o Active Pharmaceutical Ingredients (API) synthesis (liquid-liquid reactions) activated by ultrasound and microwaves in microreactors and OFRs.
The systems chosen represent challenging real applications in the fields of 1) green fuels and bulk chemicals synthesis (methanol, DEC, EMC) and 2) advanced pharmaceuticals synthesis (paracetamol, APIs) and were proposed by leading industrial companies. Central in the conceptual approach of ALTEREGO is the idea of effective integration of alternative energy forms with intensified reactors into multifunctional units that can enable highly efficient synthesis of both application fields (Figure 1). The novel reactor concepts were to be demonstrated at laboratory and pilot scale in university facilities and/or in the facilities of the industrial end-users involved. Based on the results for the assessment of MW enhanced reactive distillation for the transesterification of DMC with ethanol to yield EMC and DEC, the synthesis of butyl butyrate by transesterification of butanol was investigated as a case study for US-assisted enzymatically catalysed RD, focussing on the development of suitable internals and characterising the effect of US application.

Figure 1: Conceptual approach of the ALTEREGO project

In order for a decisive step towards commercial implementation of alternative energy-based processes for chemicals synthesis, a number of fundamental and engineering knowledge gaps were filled. These apply to:
1) Lack of process data and materials properties at local scale inside the process units. Obtaining reliable information locally is particularly challenging due to a) non-uniform spatial distribution of alternative energy fields in process units, b) the extremely short time scales of transport and dissipation of alternative energies and c) limitations in the materials of construction of process analysis tools, as the latter may interact with external fields.
2) Lack of robust and reliable meso- and macroscale models, which account for chemical reaction, fluid flow, mass transfer and alternative energy transport, and would allow for detailed process understanding and optimization.
3) Lack of strategies for effective integration and scale-up of chemical reactor equipment and alternative energies equipment (hardware / engineering limitations)
The ALTEREGO project aspired to enable highly efficient alternative energy-activated chemical syntheses by means of a consistent integrated framework consisting of advanced in-situ and on-line process monitoring, multiphase, multiscale modelling, and novel engineering solutions for effective hardware integration and scale up. The integrated framework will be applied to four industrial processes, namely a pharmaceutical reactive crystallization process, a biphasic mixing process for API synthesis, a reactive distillation process and a gas-solid catalytic process (Figure 1).
The ALTEREGO project aimed at breakthroughs in the key activity areas of the integrated framework. More specifically, breakthroughs in the following scientific and technical domains were to be achieved:

1: In situ and on line process monitoring of alternative-energy enhanced chemical syntheses
Milestone: Implementation of on-line sensors or optical equipment to laboratory scale alternative energy enhanced reactors for real-time monitoring of process conditions and product quality. Compact fiber-optic multi-sensor systems consisting of different probes, such as transmission, reflectance, ATR absorbance, UV and Raman scattering, and network analyzer as well as non-intrusive equipment such as particle image velocimetry (PIV) and thermal cameras were to be utilized for the real-time measurement of dielectric and fluid properties, process conditions and product quality (e.g., electrical permittivity, vapor pressure, viscosity, particle size, flow patterns, chemical composition, temperature etc.). This data is crucial for understanding of the underlying mechanisms, model validation and process development.

2: Multiphase, multiscale modelling framework
Milestone: Development of a computational multi-scale, multi-phase model library for the simulation of the selected fuel and pharmaceutical industrial processes describing with sufficient accuracy the integration of multiple chemical reactions, the phase changes, the heat and multicomponent mass transfer and possible separation steps induced by the alternative energy sources, acting at the same time and in the same place of the multifunctional unit. The library comprises models at different length and time scales (i.e., detailed and reduced reaction kinetics, particle population balances, vapor-liquid equilibrium, bubble dynamics, reactive distillation models, mixing/CFD models and electromagnetics (Maxwell’s equations), etc.). A modular strategy in model development were to be adopted so a wide array of processes can be generated through flexible combination of different models

3: Reactor and alternative energy hardware integration and scale up
Milestone: Development of novel scalable alternative energies-activated reactor concepts that allow transition from bench-scale experimentation to pilot and industrial scale processing. Engineering solutions were to be proposed and developed for the following envisaged multifunctional equipment: 1) integrated ultrasound technology with tubular, micro- and millireactors, oscillatory flow reactors and reactive distillation columns; 2) microwave technology with micro- and millireactors and reactive distillation columns; 3) plasma technology with micro and millireactors. The new multifunctional equipment were to be demonstrated at laboratory scale and pilot scale on the sites of the academic and industrial partners involved in the ALTEREGO project.

4: Techno-/economic feasibility studies
Milestone: Comparative techno-/economic feasibility studies on the applicability of the respective forms of energy to the model processes provided by the end users. The research activity results in the three aforementioned areas provide the necessary information for the technical and economic feasibility studies that were to be carried out for all applications. These studies provide the necessary foundation for the broad introduction of alternative energy-based processes in real production and evince the potential for the targeted average 50% improvement in energy and resource efficiency.
The project was planned for 3.5 years. The members of the interdisciplinary consortium were carefully selected to guarantee the impact by the participation of highly innovative multinational industrial companies (AN, JP) and developers of alternative energy based-technologies (SAIREM (MW, PLASMA) and SM (US)). The relevant scientific foundations were covered by world leaders in the field of alternative energy sources (TUD, KUL), intensified multifunctional reactors and processes (TUDO) and green chemistry and heterogeneous catalysis (YORK) and cover expertise in chemical and mechanical engineering, computer-aided process engineering, applied physics and catalysis.

Project Results:
Most important results by work packages
WP1 Ultrasound
Development of tailored equipment for three types of ultrasound-assisted multiphase processes (reactive synthesis and cooling crystallisation, reactive L/L extraction, VLE of binary systems) enabling local positioning of the ultrasound energy at the interfaces, and experimental determination of flow and temperature profiles. Complete set of comparison between US and non-US and for input to modelling, characterization of the ultrasound field with techniques for the ultrasound field and cavitation bubbles was done for the subsequent experiments to:
Determine the effect of flow on the acoustic field / the cavitation bubbles;
Investigate the distribution of bubble type at various process conditions;
Compare frequencies at equal calorimetric power, sonoluminescence and sonochemiluminescence signal;
Evaluate the effect of standing or travelling waves.
A thermocouple coated with a rubber material was constructed to map the energy of the ultrasonic field. The temperature difference between this sound absorbing material and the bulk liquid quantifies the local acoustic energy. This technique allows identifying the nodes and antinodes within a standing wave. The dimensions of the probe limit the wavelength (or frequency) that can be measured, and the sensitivity strongly depends on the transmitted power. This characterization technique was applied to identify the effect of stirring on the acoustic field.
Measurement of the sonoluminescence signal yields information on the global energy, and addition of quenching products (propanol and acetone) allow estimating the bubble type distribution (transient or stable). Experiments with this technique confirmed the results of the literature, and showed that low frequency produces more transient bubbles.
The global acoustic energy can be evaluated by calorimetric (temperature) or sonoluminescence (radical/light yield within the bubble) measurements. In order to compare frequencies, one should thus maintain one of these signals constant. To investigate the difference between a constant calorimetric or SL-signal, calibration curves were constructed for a single multi-frequency reactor set-up.
The effects of the frequency and intensity on the mixing efficiency and acoustic pressures were characterized. Also, the relation between this mixing efficiency and acoustic pressure was investigated. The effect of US intensity and frequency on micromixing efficiency was studied in a single phase system. It was found that low ultrasonic frequencies and high intensities generate best micromixing (lowest segregation index, Xs).
The effect of US conditions on bubble type (stable or transient) was determined using a sonoluminescence quenching technique with propanol and acetone as described in literature.[1] Results show that, within the reactor geometries investigated, cavitation type depends on frequency, ultrasonic source type, reactor design, and flow regime:
Frequencies below 120 kHz are able to generate both stable and transient cavitation, while only stable cavitation occurs at higher frequencies;
Horn type sources (frequency < 50 kHz) only generate transient cavitation, while piezoelectric transducers of the same frequency domain can yield both cavitation types;
Low frequency transducer suspended within a reactor (plate type) produces stable cavitation, but attaching the same transducer to the outer wall of a flow cell generates transient bubbles;
All bubbles become stable when a turbulent flow regime is established within a flow reactor.

Reactive crystallisation
The focus was set on pulsed ultrasound and the scale up of this technology in batch crystallization reactors. The reactor volume was systematically increased from 250 mL to 2 L, while the same ultrasonic probe of 30 kHz and a power of 10 W was used to provide sonication. Figure 1a summarizes the main results of the experiments regarding the effect on nucleation. It is shown that ultrasound increases the nucleation temperature independently of the reactor volume. In addition, pulsed ultrasound shows a clear benefit as it consumes less energy and yields roughly the same nucleation temperature as continuous sonication. Tests on the effect on particle size, shown in Figure 2b indicate that application of the same ultrasonic power in a larger volume results in the formation of larger particles. Due to the lower power density (W.L-1), less seed material is created, resulting in larger particles.

Figure 2: Effect of pulsed ultrasound on the crystallization of paracetamol in different reactor volumes. (a) Effect on nucleation temperature. (b) Effect on particle size.

A setup was developed to measure the nucleation kinetics in a flow reactor under sonication. Besides the construction of the setup, some testing of the heat exchanger was done and adaptations on the design were performed. This setup can be used to investigate nucleation rates under sonication in a flow reactor, which delivers input for population balance modeling.
Further the sonofragmentation model using Kapur function analysis was developed. With this model, it could be concluded that transducers are more efficient to break the largest particles, while probes are more efficient to break finer particles. There is relatively more abrasion with a probe as compared to a transducer.
The crystallisation in the oscillatory flow baffle crystallizer (OFBC) was further investigated. Poor solids handling and mixing of solids was encountered at the optimum mixing conditions found during the residence time distribution characterization in the OFBC, using a homogenous tracer (methylene blue in water). In order to process solids; much higher mixing intensity was required in order to prevent settling and segregation of solids in the introduced slurry. The residence-time-distribution (RTD) study has been repeated using solid crystals as tracer. The new optimum condition operates at much higher amplitudes (32 mm compared to 4 mm) for crystals to achieve plug flow behaviour (Figure 3). The quantified dispersion is much less as compared to the predicted dispersion based on homogenous tracer studies popularly being carried out in OFBC literature.
Based on the RTD response curve, a classical tank-in-series (TiS) model has been implemented. The TiS model predicts plug flow behaviour when a large number of tanks are connected in series. This analogy has been extended to a MSMPR-like crystallization model in which the number of crystallizers in series estimated from the tracer experiments is used to model RTD of the crystals, using kinetic equations and parameters for growth and secondary nucleation from literature. A sensitivity study has been performed to predict the effect of seed distribution, seed loading and cooling profile on the final crystal size distribution.

Figure 3: Residence time distributions of the crystals after implementation of the new optimized mixing parameters (2 Hz and 32 mm)

Reactive extraction
On comparing the three modes (direct, intermittent & indirect) of application of ultrasound to the microchannel designed, the best design, which gave the maximum yield to the hydrolysis reaction studied, was the intermittent contact type. This particular design was seen to be accompanied by a disadvantage of the inability to control the temperature of the system. Hence a hybridization between the intermittent and the indirect contact design was explored. Two designs were constructed: the open-interval and the closed-interval type. These two designs were seen to give similar results, with yields comparable to the intermittent design at the lower residence times. Out of the two the closed interval was chosen to be explored further as the variation in the yield was minimal. For use in industrial applications the possibility of scale up of the closed interval was also studied. The reactor was scaled up from a volume of 0.27 ml (0.8 mm tubing) to 2.22 ml (2 mm tubing). Relative to the yields obtained in the silent condition the scaled up version showed better increase in the yield at similar residence times (75% increase for 2.22 ml in comparison to 60% for the 0.24 ml, both at 87 s) and the volumetric mass transfer coefficient values (80% increase for 2.22 ml in comparison to 70% for the 0.24 ml, both at 87 s).

Enzymatic reactive distillation
Experimental investigations of the enzymatically-catalyzed transesterification of ethyl butyrate with Candida antarctica lipase B were performed with different immobilizates and packing materials in lab-scale. These are for once coated packings, which can directly be applied as packing material for the application in distillation columns, and enzyme beads, which can be introduced in corrugated sheets made from wire gauze inside of hybrid structured packings like Katapak SP from Sulzer. Both options present suitable packing materials for an implementation in reactive distillation. Besides the determination of a suitable kinetic model for the different catalytic materials, the influence of ultrasound on the different immobilizates regarding enhancement of the reaction rate as well as its chemical and mechanical stability were part of the investigation. While a kinetic model was successfully identified based on the kinetic experiments for both immobilizates, significantly different results were obtained when applying ultrasound irradiation to the different immobilizates. By means of ultrasound irradiation it was possible to significantly enhance the reaction rate for the coated packing. However, when enzyme beads are used as immobilizate, there was no influence on the reaction rate compared to mechanical stirring and ultrasound irradiation was causing disintegration of the enzyme beads. Therefore, the feasibility and potential benefit of ultrasound application depends significantly on the type of immobilizate for enzyme application. The enhancement of the reaction kinetics for ultrasound application in combination with the coated packing is most likely explained by the increase of the internal mass transfer rate. To investigate the effect of ultrasound in reactive distillation, a kinetic model was established as well, based on the experimental results. Model validation for the reactive distillation has been performed in combination with the pilot plant experiments performed in the context of WP 5.
Ultrasound equipment design
Work was performed on design improvements for the ultrasound transducers and up scaling of design solutions which have been identified as properly working for the first tubular milli- or microreactors. According to this, 2 new design approaches with direct coupled bulk transducers as well as a draft idea for a so-called clamp-on transducer have been developed and tested. Two reactors have been developed on this basis: a 37 kHz and a 67 kHz microreactor.
Regarding the necessary ultrasound power amplifier, a first principal prototype of a so-called Royer-Converter tuned to 600 kHz was designed in accordance to the already presented 600 kHz tubular reactor. This prototype was tested successfully. In addition a compact Class C unit for a 220 kHz reactor was developed and tested. As the frequency of these types of amplifier is fixed, a broadband amplifier solution was developed. Experimental support was provided regarding the evaluation of the reactor design. Using a terephthalic acid dosimetry method reported in literature, the hydroxyl radical yield, and thus the cavitation activity, of all reactors could be measured and appeared satisfactorily for application in chemical processes.
WP2 Microwaves
As to the testing of different acids other than HBr no other suitable non-halogenated acid that could do the demethylation reaction could be identified. In this respect, we have tested sulphuric acid, para-toluenesulfonic acid (PTSA), methanesulfonic acid (MSA) and also trifluoroacetic acid. Other acids of lower acid strength were not considered, nor were acids, which may have promoted alternative chemistries (e.g. HI: radical chemistry, carboxylic acids: esterification/acylation). The potential use of triflic acid (TFA) has been forsaken, as this would have created even more challenges in terms of safe operation and reactor development than HBr. Also, the potential use of TFA was considered as a no go. Additionally also the use of KBr was tested as it would avoid the strong acidity while still delivering the required nucleophilic agent – yet in vain. Ideally the sole use of water at higher reaction temperatures, easily attainable with microwaves, would have been a most interesting way forward. Past experiences showed that this was possible on 2-methoxynaphtalene but applied to 3-methoxybenzylammonium bromide it was found impossible.
With respect to reactor development an opportunity presented itself when working under reflux conditions, an operational regime coinciding essentially with a constant MW power input. In this zone extensive amounts of gas/vapour bubbles are created, notably H2O and MeBr, and in rigorously checking different models we were able to explain the conventional and low T microwave operation by mechanism 1a, which can be reduced to a two-step equation denoted 1b:
█(A+H^(+ ) □(□(↔┴(k_1,k_(-1) ) )) P+MeBr(L)@MeBr(L) □(→┴k_2 ) MeBr(G)@P+ H^+ □(↔┴(k_3,k_(-3) ) )〖PH〗^+ )
(1a)
█(A+H^(+ ) □(→┴k_1 ) P+MeBr(g)@P+ H^+ □(↔┴(k_3,k_(-3) ) )〖PH〗^+ )
(1b)
█(A+H^(+ ) □(↔┴K_1 ) P+M_L@M_L+S□(↔┴K_2 ) M_L S@M_L S□(→┴k_3 )M_G+S@P+H^+ □(□(↔┴K_4 )) PH^+ )
(2)
Conversely, at the high microwave reaction temperature, the occurrence of zero order behaviour was observed as a very specific 'microwave effect'. This could be explained by the introduction of a surface concept at which MeBr (liquid) is converted to MeBr (gas) (equation set 2). It is exactly the intense reflux/gas production conditions, which presented an opportunity to the development of a novel microwave reactor concept – following in a way the principles of gas-lift but to the best of our knowledge this has never been demonstrated when employing microwaves specifically.
In Figure 4 an advanced working set up is depicted, which was designed for the MiniFlow TM cavity, and a real live video of the demethylation reaction of 3-methoxybenzylammonium bromide can be observed at: https://drive.google.com/file/d/0B79Ou6B6H9hFdDY2VUZfZV9jV3M/view?usp=sharing.

Figure 4. the eventual reactor concept designed for use with the TM cavity

Additionally, Figure 5 displays two schematics illustrating the working operation of said reactor and how it could be, easily, converted to a true continuous microwave reactor. Continuous operation provides an efficient way of dealing with the methyl bromide side product. In addition, the reactor in Figure 4 can be adapted to a configuration in which the flow enters the reactor from below and the reaction product (or better the partially converted mixture) exits from the top. Such microwave reactor set-up was made/explored on the MiniFlow U-form device and a more advanced version is shown in Figure 5.

Figure 4. left: schematic of the 'circular' flow continuous microwave reactor & right: alternative flow configuration as to account for a truly continuous microwave flow reactor

Figure 5: alternative continuous MW flow set up fit for use on the U-form MW device.

Additional temperature simulations and measurements of the dielectric properties of the reaction mixture were completed. Temperature variations due to a non-uniform heating rate have caused inaccurate temperature readings and given rise to faulty conclusions on microwave kinetics. Multiple examples of this can be found in the literature where they are often attributed to special 'microwave effects'. To alleviate this concern in the present API demethylation reaction, the temperature distribution within the reaction mixture was carefully evaluated by means of a coupled heat transfer, fluid dynamics and electromagnetics simulation. In Figure 6 the simulated temperature distribution in the reactor is shown, illustrating that the applied vigorous stirring does indeed reduce the temperature gradients in the reactant mixture to negligible levels. This is so despite strong fluctuations in the heat generation, which occur both spatially and temporally. The following link shows an animation on the distribution of electromagnetic dissipation over the reactant volume over the course of one stirrer bar revolution: https://surfdrive.surf.nl/files/index.php/s/1nLqMbAF7ZMxrCk. The rotation of the stirrer bar causes large fluctuations in the electromagnetic field around it. This demonstrates the large degree of parametric interdependence in resonant electromagnetic fields, also reaffirming that the evaluation of temperature distribution, whether via simulation or otherwise, is a necessity before conclusions can be drawn.

Figure 6. Simulation of the temperature distribution in the reactant mixture. The scale range is limited to only show the temperature gradient in the reactant mixture.

The connecting factors between the electromagnetic field and the heat generation are the dielectric properties. These need to be determined to enable an even more accurate simulation. Figure 7a shows the set-up used to measure the dielectric properties at relevant temperature; essentially a measurement probe connected to an Agilent ENA-series network analyzer. Figure 7b shows the 0.5 – 10 GHz spectrum for the dielectric properties at 107 °C; the loss factor is low relative to the permittivity at 2.45 GHz, which corresponds to relatively deep propagation of microwave energy into the reactant mixture.

Figure 7a. Setup for dielectric properties measurement.

Figure 7b. Dielectric spectrum from 0.5 to 10 GHz at 107 °C.

WP 3 Non-thermal plasma
The chemistry under consideration is the reverse water gas shift reaction (CO2 + H2 -> CO + H2O), as a first step to convert CO2 to methanol. Only gas phase chemistry is considered, as thermal catalysis has turned out not to be beneficial. A reduced plasma kinetic model for CO2 dissociation has been developed and validated against the detailed kinetic model from which it was derived. Afterwards, a multiphysics plasma reactor model was developed that combines fluid dynamics, Maxwell’s electromagnetic equations in the wave equation representation, and drift diffusion physics to account for electron mobility. Further, the ionization and recombination reactions for argon were implemented to represent the plasma interactions of this gas. The model was constructed in a 2-dimensional axisymmetric domain. Because the microwave resonant cavity of the Surfatron is not axisymmetric, it was verified that the geometrical and excitation port adjustments do not significantly affect the microwave field in the critical zone where it interacts with the plasma. It was found that the variations caused by the geometric simplification are indeed negligible. The main insight is that the electromagnetic field exits the microwave resonant cavity where the plasma was initially generated, and that it travels along the plasma flame in a wave pattern known as a surface wave. It thus reheats the plasma as it travels along the reactor tube. This mechanism extends the plasma flame beyond the vicinity of the zone of initial plasma generator in the resonant cavity; it therefore leads to an increased process – or reactor – volume.
Further, several modifications have been implemented, in an existing atmospheric plasma reactor (co-funded by the Bill & Melinda Gates Foundation) in order to refine thermal management. The modifications are outlined as follows:
Redesign and construction of a modified downstream reactor assembly;
Application of ceramic reactor lining to provide adjustable thermal insulation;
Implementation of a redesigned microwave power controller
As for improving the energy efficiency, the bench scale microwave plasma reactor was modified by adding an extended waveguide with the objective of utilizing a larger fraction of the input microwave energy. An extended waveguide was added to bench scale microwave plasma reactor. In comparison to the previous version, this system is nearly fully automated, meaning that most of the operating parameters can be tuned by the control interface (input flow rates, pressure and input microwave power). This approach represents a step ahead toward the implementation of this technology to commercial scale, as this level of controllability will be required to run large microwave plasma reactors due to the extremely fast dynamics of the system. The reactor (middle), the gas supply unit (left) and the control interface (right) are shown in Figure 8.

Figure 8: New and automated bench scale microwave plasma reactor.

Besides exploring controllability issues, the new reactor configuration was designed and built, as presented in Figure 9. The purpose of this novel configuration is to optimize the utilization of input microwave energy by enabling larger plasma volumes and thus longer residence time.

Figure 9: Novel reactor configuration including the extended waveguide (self-customize) in combination with the plasma generator (Surfatron, SAIREM).

As mentioned earlier, the focus was set on the CO2 hydrogenation to CO, which is part of the two-step methanol synthesis process, so-called CAMERE process. This reaction is an equilibrium endothermic reaction, meaning that higher temperatures lead to higher conversion of CO2. An increase in the gas temperature in the plasma reactor when using the extended waveguide was noted by means of thermal imaging, i.e. measuring the incoming radiation from the plasma, as shown in Figure 10.

Figure 10. Gas temperature measured by Thermal Camera (FLIR series) with and without using the extended waveguide.

In Figure 11, the conversion of CO2 for both configurations, without waveguide (case 1) and with waveguide (case 2), is presented. It is noted that at feed H2:CO2 ratios equal to 1 and 2, there is a noticeable improvement in the CO2 conversion under the same operating conditions. Remarkably, the conversion of CO2 is slightly lower at H2:CO2 = 3, as opposed to the lower ratios. In order to investigate this fact, the analysis of the emission spectrum for both configurations is carried out through optical emission spectroscopy. In this regard, two different spectrometers were used, HR2000CG and Maya 2000Pro (Figure 12). The Maya 200Pro offers a better resolution in the UV-VIS range, whereas the HR2000CG shows higher sensitivity in the NIR range. As seen in Figure 12, there is a significant change in the chemistry of the reaction. For the case 1, when the waveguide is not used, the concentration of intermediate species such as OH radical and Hβ atom in the plasma is much larger (red line) as compared to the case 2 (blue line), when the waveguide is installed. The reduction in the intermediate species concentrations leads the process to a lower conversion of CO2, and thus lower energy efficiency.

Figure 11. Conversion of CO2 for two different reactor configurations, case 1 (without waveguide) and case 2 (with waveguide).

Figure 12. Emission spectrum for two different reactor configurations, (a) using the Maya 200Pro spectrometer (better resolution in the UV-VIS range) and (b) the HR2000GC (higher sensitivity in the NIR range).

Finally, a larger atmospheric microwave plasma reactor (6 kW) was upgraded for hydrogenation of CO2 in the context of process scale up. Various modifications were required in order to perform the hydrogenation of CO2 as the setup was previously used for biomass gasification experiments. Among the most important modifications: 1) installation of a new gas supply unit to feed gases such as CO2 and H2, 2) implementation of additional safety measures due to the large amount of CO produced during the reaction, 3) addition of a new connection to analyze gas product composition in-line by mass spectroscopy, 4) design and construction of a condenser to collect the large amount of water generated during the reaction. Figure 13 shows the modified layout of the atmospheric microwave plasma reactor. Preliminary experiments with diluted CO2/H2 mixtures in N2 indicate stable operation and complete conversion of the reactants. This experimental campaign is currently ongoing.

Figure 13. Upgraded atmospheric microwave plasma reactor (6 kW) for hydrogenation of CO2.

WP4 implementation/validation of the alternative energy based technologies in a pharmaceutical environment.

Ultrasound assisted crystallization
Two set-ups have been made operational (Figure 14). The Easymax 102 Reactor is equipped with temperature control and an ultrasound horn (30 kHz, 50 W), which irradiates a reactor volume of 100 ml. The other reactor is a Recirculation Reactor of 1000 ml, equipped with a 20 kHz sonication cell of 100 ml, in-line FBRM and IR analysis as well as temperature control.

Figure 14: Set-up of the Easymax 102 Reactor (left) and the Recirculation Reactor (right).

These set-ups were tested on 3 API's. For API 1 and API 2, the effect of pulsed ultrasound was tested on agglomeration and the production of monocrystals. For API 3, the effect of ultrasound on fragmentation (particle size reduction) was investigated with the purpose to improve the morphology from star shape crystals to mono crystals. It was shown that wet seeding with the application of ultrasound improves the morphology. Mono crystals are obtained if the solution is exposed for a sufficient amount of time to the ultrasonic field at the seeding temperature. As an alternative, ultrasound can be applied during the entire cooling crystallization in continuous or pulsed mode in order to reduce the formation of agglomerate. Similar results were obtained for API2 (see Figure 15)

Figure 15: Left (top and bottom): standard seeded crystallization, middle (top and bottom) additional temperature cycling and high shear milling, right (top and bottom) pulsed US during crystallization.
With API3 continuous milling using ultrasound was tested in the loop reactor. The results show that effective particle size reduction can be obtained using ultrasound. The ultrasound intensity has a limited effect on the resulting particle size. Particles having a chord length < 100 micron do not seem to be broken by ultrasound.

Reactive extraction
The setups developed were tested for the following reactions:
a Suzuki Miyura reaction
a selective extraction of a diol impurity
the removal of acetic acid anhydride

Suzuki-Miyura Reaction
The original plant process consist of mixing two phases in a batch reactor at room temperature and slowly heating to 89°C at 1 K/min and then refluxing the reaction at the same temperature for 30 min. The experiments were conducted in a 50 ml Easy max reactor and even before the system is refluxed the conversion started. This does not yield an accurate understanding into the kinetics of the reaction. Hence the process was decided to be carried out differently to understand the kinetics better and to translate this batch process into a continuous manner. Each of the phases will be heated to the required temperature individually and then contacted maintaining the same temperature throughout the reaction. A visual indication of the end of the reaction is the formation of the palladium black, which turns the reaction mass black. The reaction was carried out at different temperature in the manner explained above in a batch reactor and the conversion at the different temperatures is plotted below. It could be shown that the system is temperature sensitive and below 80°C a lowering of the temperature even by 5°C has a drastic effect on the rate of the reaction.
Similarly the reaction was carried out in a continuous manner with 2 mm tubing as large residence time is needed for the reaction at 75°C. To maintain the temperature of the feeds the inlet lines are coiled through a water batch maintained at the required temperature as shown in Figure 16.

Figure 16: Modification of the flow experimental setup to include preheating of the streams.

The results show no difference between the silent and sonicated condition and also with the batch process. The points look like they are in line with the batch conditions. The same results were obtained in increasing the input power to 30 W. Assuming the inefficiency of the sonication to the lack of temperature, the experiment where repeated for 80°C and the results obtained are negative on sonication.
To understand this behaviour better and not a single point the reaction was repeated at 75°C in a batch sonicated condition; the results obtained are plotted in Figure 17. The results obtained confirm that worse yields were obtained on sonication.

Figure 17: Yield in silent versus sonicated in batch at 75°C.

To understand what happens on sonication the system was first qualitatively studied, the reaction mass was sonicated and the two-phase system was observed in the batch and continuous system. Two phases were obtained. Red precipitate was observed at the interface and the aqueous layer of the reaction mass for both the two modes of operation. The only compound that was added which has a similar characteristic was the palladium acetate catalyst. An experiment was conducted with purely the palladium catalyst dissolved in toluene and the system was sonicated with the basic aqueous solution. From the Figure 18(a) it is evident that the sonication resulted in the movement of the Pd catalyst from the organic to the aqueous layer, with the movement more intensive with increase in temperature. Similar behaviour was observed when the experiments were conducted with a weaker base potassium carbonate or at different frequency at the same temperatures (Figure 18(b)).
(a)
Figure 18: Movement of the palladium catalyst from the organic to the aqueous phase.

Selective extraction of Diol Impurity
The original process of the impurity removal involves adding water to the organic phase and phosphoric acid is added till the pH of the reaction mass is in the range of 4.5-4.6. The extraction is very much dependent on the pH of the reaction mass. Lower the pH the higher the impurity and product extracted, hence and optimal value of 4.5 to 4.6 is selected for the reaction. Also it is evident that a number of extraction steps have to be carried out to reach the desired level of impurity. Each of these extractions takes at least 1.5 hrs (pH Adjustment + 20 mins mixing + 60 mins settling + transfers). The reaction was carried out in a flow-sonicated setup with the aqueous already set to a pH of ≈ 2.9 to obtain a pH of 4.9 and the results obtained are shown in Figure 19.

Figure 19: Reaction in a in a flow-sonicated setup

An improvement was detected in the amount of the diol extracted, but the values remain the same for the batch silent, flow silent and sonicated conditions. To understand whether this is a mass transfer limited or not the experiments were repeated in a 2mm tube flow reactor at the same residence times. A similar behaviour between the silent and the sonicated conditions were obtained. The process was carried out in the 0.8 mm tubing channel again at very low residence times and the results were similar again. These results confirm that the process is not mass transfer controlled but kinetic controlled and thus the availability of the protons determined by the pH of the aqueous phase defines the extraction efficiency of the process.
An important observation in this experiment is that when carried out in a batch manner the reaction mass tends to be cloudy, which warrants a settling time of 60 mins but when carried out in a continuous manner without ultrasounds yields the same results without the cloudiness of reactions mass. This might be a good system to be converted to a continuous process as it can reduce the time for the entire process.

The removal of acetic anhydride
The hydrolysis of acetic anhydride was first studied in the batch setup with acetic anhydride dissolved in toluene at a concentration of 0.3 M and distilled water as the aqueous phase. When the experiments were conducted for the silent and sonicated experiments at 25°C the separation of the batch in silent condition happened well but on sonication a stable like emulsion was formed which did not settle quickly. A sample kept for 24 hrs provided to have not separated completely. For sonication an additional separating device was required and the centrifuge proved to be fruitful, the operating parameters for the centrifuge are at 2700 rpm and 2 mins. From these results obtained it is clearly evident that sonication helps in a quicker acetic anhydride removal.
The experiments were repeated in a flow condition at 25°C in 2 mm ID tubing and the results obtained are as shown in Figure 20.

Figure 20: Extraction of acetic anhydride

The flow experiments proved to have similar performance as the batch silent experiments. When comparing to the sonicated system there seems to be no improvement in the anhydride extracted. As mentioned in the flow experimental setup, the separation is carried out with a membrane separator. The membrane is hydrophobic and allows the organic layer to pass through it but the emulsion is in the aqueous layer and hence we believe an effective separation is not occurring. The experiments were repeated with centrifugal separation and the results are also plotted in Figure 20, which showed no or little different for the silent condition, but a good difference for the sonicated. Hence for this system effective separation of the two phases is really important. The batch-sonicated results show better performance in comparison with the flow-sonicated ones for the same residence times as additional mixing is also provided by the stirrers in the batch setup.

WP5 Demonstration in green fuels and bulk chemical synthesis
The work package includes the implementation/validation of the energy-based technologies from WP 1-3 in an academic lab/pilot environment. A technical-economic evaluation of the processes and comparison with the conventional processes currently applied in industry and the development of a roadmap for industrial implementation of the proposed technologies
Ultrasound & Microwaves
The application investigated for implementation is the microwave- and/or ultrasound-assisted reactive distillation process for the transesterification of DMC with ethanol for DEC and EMC synthesis. The effect of various operational parameters (frequency, power intensity, system pressure and temperature, heat duty, flow rate and reflux ratio) was studied in relation to up scaling properties. Since experimental investigations of the effect of MW irradiation on VLE for the chemical system of the EMC and DEC synthesis and subsequent simulation studies did not indicate a significant improvement by MW-enhanced RD (Werth et al. 2015), the investigations for RD were instead focussed on the added US-enhanced enzymatic reactive distillation case study.
The performed work was focused on the demonstration of enzymatic reactive distillation (ERD) in pilot scale equipment and model development and validation to yield for the techno-economic evaluation of ERD with and without ultrasound. Thus in the last period pilot-scale experiments were performed for the enzymatic reactive distillation in a DN50 reactive distillation column equipped with different internals and different enzyme immobilizates: enzyme beads and coated packings. Both concepts of enzymatic reactive distillation, using enzyme beads and coated packings, were shown to be viable options without ultrasound. In order to investigate the potential benefits for ultrasound irradiation to enzymatic reactive distillation the process is modelled, making use of the reaction kinetic models with and without ultrasound irradiation developed within WP 1. For the enzyme beads and the coated packing a model of the ERD was successfully validated against the pilot scale experiments. A validated model is the basis for further evaluation of the Ultrasound-assisted Enzymatic Reactive Distillation (US-ERD), considering the experimental investigations from WP 1, which indicated a considerable improvement of the reaction rates for the coated packings.
Plasma
Evaluation of the novel prototype equipment for microwave (non-thermal) plasma-assisted CO2 hydrogenation to methanol. The effect of various operating parameters (flow rate, feed composition, pressure, power input), was studied in relation to up scaling properties. A new bench scale microwave plasma generator has been built, see Figure 8, section WP3. Besides exploring controllability issues related to MW plasma, a new reactor configuration has been designed and built, as presented in Figure 9, section WP3. This figure shows an extended waveguide coupled to the previously used plasma generator Surfatron. The purpose of this novel configuration is not only to measure more accurately spatial changes in the emission spectra by means of optical emission spectroscopy but also, and more importantly, extend the plasma column enabling a larger plasma volume to achieve a better reactor performance. The experiments indeed show better conversions with the extended waveguide for H2:CO2 ratios equal to 1 and 2. Overall 60-80% conversions per reactor pass are attained. These are higher compared to conventional thermal catalytic reactors.
Technical-economic evaluation of the processes and comparison with the conventional processes currently applied in industry.
It covers aspects including process description, the interface between the new process/equipment and the plant, the operating behaviour and efficiencies, the control system and personnel demand, maintenance demand, ecological aspects, feed characterization and handling and any relevant weak points. The economical evaluation focuses on two key economic figures: (i) required capital investment including total process unit/plant capital, permits, royalties, the initial charge of catalysts and chemicals and startup costs. (ii) cash margin defined as the annual revenue from expected sales less chemicals cost, freight and operating expenses. Cash margin covers capital recovery and profit.
US-ERD: The framework of AlterEgo included a techno-economic evaluation of the Enzymatic Reactive Distillation (ERD) with and without ultrasound assistance. Therefore the ERD model was used and the kinetic model for coated packing (with and without ultrasound assistance) was implemented. The economic potential of US-ERD and ERD was evaluated on the basis of a case study, in which the minimum total annual costs of the ERD and Ultrasound-assisted Enzymatic Reactive Distillation (US-ERD) column for a production capacity of 10 kilotons per year with purity specifications of BuBu (99% pure) were determined based on an annual operation of 8000 hours. Cost models for equipment installation costs as well as the operating costs were implemented into the process model and an optimization with respect to the total annual costs was done for both, ERD with and without ultrasound assistance. Comparing the total annualized costs of ERD and US-ERD they are nearly the same with approximately 18.45 Mio. €. Assuming that the reaction rate enhancement by US could further be enhanced a sensitivity analysis was performed, which showed that installation costs could be reduced significantly by further improvement of the reaction rate. Based on the results of the techno-economic evaluation a roadmap for the ultrasound application in reactive distillation in an industrial setting was created.
Development of a roadmap for industrial implementation of the proposed technologies.
This roadmap covers scalability aspects, the requirements in terms of associated enabling technologies such as measurements and analysis methods and process control systems as well as necessary knowledge dissemination for know-how transfer to industrial process technologists (e.g. equipment and process training programs). A roadmap on the implementation of ultrasound (US) in reactive distillation and US-assisted Enzymatic Reactive Distillation (US-ERD) processes was created. It examines the necessary steps before using US in an industrial reactive distillation set-up and addresses the current state-of-the-art implementations as well as design and scale-up considerations. The basis of the roadmap are the findings in reaction rate enhancement concerning investigations of the enzymatic reaction kinetics of the transesterification of ethyl butyrate and investigations of the (US-) ERD. Furthermore, a literature survey about the scalability, equipment design, field uniformity, and penetration depths showed opportunities to establish a US-RD process. The improvement of reaction rates by US is the key element for a successful implementation in US-RD. The process of gaining knowledge of the mechanisms behind this improvement was described. Finally, a timeline for the process implementation was created, showing the necessary steps to design an industrial US-ERD set-up. Furthermore, a roadmap for the implementation of MW plasma technology at industrial scale was developed.

Highlights of most significant results
Successful implementation and demonstration of ERD on pilot-scale equipment as well as provision of a validated model for ERD, which can be used to evaluate the potential of US-ERD by taking into account reaction rate improvements on the basis of reaction kinetic models.
Comparison of the total annualized costs of ERD and US-ERD showed that both technologies give similar results regarding the costs.
An implementation of US in RD technology can take several years until generation of knowledge of the mechanisms behind reaction rate improvement and equipment design is carried out.
MW plasma reactors are currently being used at lab/pilot scale. It has not yet been implemented at commercial scale. Two particular applications, microwave plasma gasification and microwave plasma-assisted carbon fibre production represent the first attempts to overcome the transition from lab to industrial scale.
MW plasma offers outstanding benefits for processes in which high temperatures or high energy densities are required (gasification, endothermic reactions among others).
The main technology limitations were identified such as 1) microwave generator capacity, 2) plasma stability, 3) equipment design, control and safety, 4) material of construction and 5) scale up among others.
The cost of the kW microwave energy was assessed for various commercially available MW generators. The best-case scenario is about ~1500 €/kW.
A twelve years timeline (2016-2028) where the most relevant scientific and technological challenges as for the further development of the technology was included.

Potential Impact:
Strategic impact
Improved energy efficiency and raw material savings in advanced pharmaceuticals syntheses
The era of highly profitable blockbuster products in Pharma industry has passed. Declining R&D productivity, rising costs of commercialization, increasing payer influence and shorter exclusivity periods have driven up the average cost per successful launch to $1.7 billion and reduced average expected returns on new investment to the unsustainable level of 5% . Cost-competiveness can be achieved through :
• Reducing the lead time of the entire production process from the delivery of raw materials until the completion of the product.
• Increasing the selectivity of the reactions and thus the material yield and sustainability.
• Switching from a batch wise to a continuous production, making the production more economic viable, sustainable and energy efficient
The first two factors are linked with the drug discovery phase, which accounts for >30% of the total investment required for one successful drug launch (discovery through launch)7. Lead compound optimization and medicinal chemistry are the major bottlenecks in the drug discovery process , and so there is a constant need for technologies for rapid synthesis of chemical compounds. In this field, microwave technology can play a major role due its known capability of causing multifold increase in reaction times. More specifically, the time required from the design of a compound library to the production stage is 15-22 weeks.9 Microwaves can increase productivity during the proof-of-principle and validation phases with a factor 18.9 This can thus reduce the overall lead time (including design, proof-of-principle, validation and production) with a factor of 2.5 to 3. This can be translated into substantial capital savings. At cost of about 10000 €/g of API using standard technologies the total value can accumulate to 10 bln. €. Even if the project brings about only a moderate cost reduction by a factor of two, the overall savings will be 5 bln €/year which will be split between suppliers and health organisations depending on permissible market prices once the technologies are applied.
Besides, while continuous manufacturing has been the norm in almost all manufacturing industries, the production of pharmaceuticals has remained batch wise even where the production processes of active pharmaceutical ingredients (APIs) aren’t all that different from those of fine chemicals. It has been recognized however that to date, major efficiency gains have already been implemented within the drug manufacturing arena and “additional quantum gains” in batch processing are limited. Thus a shift towards continuous manufacturing will make it economically viable and sustainable and at the same time cleaner, leaner and more energy efficient. The integration of the reaction and crystallization in intensified continuous processes as developed in the ALTEREGO project, has shown to have great benefits . Studies at the MIT Novartis center have shown that by the implementation of a “fully integrated continuous manufacturing” of a drug substance, 40 % of the unit operations can be eliminated, the production cycles can be reduced by a factor of ten and cost reduction achieved between 7 and 40% depending on the level of optimization. In addition, the continuous production allows for a considerable reduction of the development times and a reduction in the manufacturing footprint and waste and material flows.
Further, substantial increase in selectivity and thus material yield can be achieved both with microwaves and ultrasound. For example, Dhumal et al. reports an increase in crystallization yield of salbutamol sulphate by ultrasound from 61% to 92-96% that is, 50% increase in material efficiency. The liquid-liquid case study for API synthesis, proposed by JP, represents many similar synthesis reactions of APIs that are performed in liquid-liquid systems, in particular organic-water or organic-organic biphasic systems. In such cases efficient mixing between the two phases becomes essential for the reaction to proceed. Bad mixing potentially leads to low process rates, conversion and/or selectivity. Ultrasound can here be of benefit due to its ability to improve micromixing at the interface between the two immiscible phases. Microwaves can improve the reaction yield similar to previous studies. One of the two case studies investigated in this project is the reaction of an organic component in toluene with hydrazine in water to form an API. In conventional reactors, high excess of hydrazine (up to 5 to 1) and a residence time of up to 1 hour (depending on the temperature) are required, resulting in a conversion of maximum 60%. The ultrasound is expected to decrease the excess required, shorten the residence time and increase the conversion. By integrating the synthesis with the crystallization in a continuous reactor it is expected that it will be possible to increase the yield to 90% with ultrasound/microwave, thus 50% increase in material efficiency.
The other case study concerns paracetamol synthesis through reactive and cooling crystallization. Paracetamol (N-(4-hydroxyphenyl)acetamide) is a widely used over-the-counter analgesic (pain reliever) and antipyretic (fever reducer). It is commonly used for the relief of headaches, other minor aches and pains, and is a major ingredient in numerous cold and flu remedies. The total world market for Paracetamol bulk drug is estimated to be about 170,000 tonnes per year with Europe and North America accounting for nearly 30% of the total market. The world market for Paracetamol is growing at an average growth rate of 5-6% per annum. Increasing global demand for Paracetamol is primarily attributed to the growing third world market. The growth of Paracetamol market in countries like USA and Japan is due to incorporation of Paracetamol in cough and cold drug formulations. World over, Paracetamol continues to effect the market share of Aspirin. However, Paracetamol is, in turn, facing competition from newer drugs.
Paracetamol has a low solubility in water and has a tendency to form agglomerates during crystallization. Its low solubility makes especially challenging the on-line measurement of the particle size distribution (due to the low number of crystals formed during crystallization), and the solution concentration with high enough accuracy for characterizing the crystal growth kinetics for on-line control of batch crystallization processes. The goal is to operate the crystallizer in such a way that agglomeration is avoided, and large paracetamol single crystals are formed.
In this project an improvement of the energy efficiency of the synthesis and separation process was demonstrated at the facilities of JP, showing not only the benefits of the application of ultrasound on the efficiency, throughput and product quality of the pharmaceutical product, but will also enlighten the advantages of the integrated continuous production process to the pharmaceutical industry.
Improved energy and resources efficiency in green fuel and bulk chemical syntheses
The focus on this part of ALTEREGO is on the synthesis of methanol, ethyl methyl carbonate (EMC) and diethyl carbonate (DEC). Methanol is an industrial chemical product, but it can be produced also from biomass http://www.syntecbiofuel.com/methanol.php. As a basic product, it is used in synthesis of MTBE (ensures cleaner combustion of petrol), formaldehyde, acetic acid, solvents, chloromethanes, methyl methacrylate and diverse other products. It has been used almost exclusively as a basic material in the chemical industry so far. However, this is changing rapidly as methanol also has excellent combustion properties, making it a suitable and proven fuel for the internal combustion engine.
EMC and DEC are also industrially relevant products. The asymmetric carbonic ester EMC has been found to be a suitable cosolvent for incorporation into nonaqueous electrolytes to enhance the low-temperature performance of rechargeable alkali metal-ion batteries with respect to characteristics such as energy density, discharge and capacity. EMC has been used in the methylation of primary aromatic amines. DEC represents an attractive alternative for hazardous ethyl halides and phosgene as an ethylization and carbonylation reagent in organic synthesis and is also used as an intermediate for various pharmaceuticals such as antibiotics and narcotics, notably phenobarbital. Additionally, DEC can be used as a raw material for the manufacture of polycarbonates. Like EMC, DEC is widely used as a co-solvent in alkali metalion batteries. Most importantly, DEC is considered to be one of the best alternatives for methyl tert-butyl ether (MTBE) as an oxygen-containing fuel additive , . DEC is miscible with fuel without any phase separation and has 40.6% oxygen. The further investigated synthesis of butyl butyrate also presents a potential fuel additive (Dwidar et al. 2012). However, its potential use as natural food flavour presents an additional application in the food industry, in which its resemblance of the smell of pineapple can be exploited. Besides the chemocatalytical synthesis from butyric acid, from direct phytoextraction or concentrated fermentation broth, bio-based production with the help of enzymes as natural catalyst and synthesis via transesterification allows for direct utilisation in the food industry (Fayolle et al. 1991).
In this project we studied two novel processes: 1) EMC and DEC synthesis from transesterification of DMC with ethanol in reactive distillation units enhanced mainly by microwaves, and synthesis of butyl butytrate by transesterification of ethyl butytrate in an enzymatic reactive distillation, using the enzyme lipase B form Candida Antarctica (CalB), investigating the effect of US enhancement. 2) Direct methanol synthesis from CO2 hydrogenation using plasma in combination with catalysts. As explained below, significant impact can be expected not only with respect to the particular syntheses under study in the relevant fuel market, but also due to the development of new technologies that will form breakthrough in reactive separations and gas phase catalytic processes in general.
Microwave enhanced reactive distillation as novel reactor concept for EMC and DEC synthesis from DMC
The European chemical processing industry is responsible for about 15% of the total energy consumption in Europe. The majority (ca. 80%) of energy in the chemical industry is consumed in the form of heat. Separation operations are responsible for ca. 40% of the total energy consumption in the chemical industry. The combination of reaction and separation presents an attractive class of novel technologies that should lead to significant reductions in energy use and waste generation. Besides, the use of alternative sources and forms of energy are known to have significant potency for intensification of chemical processes and improvement of their sustainability performance. This new integrated reactor concept unifies three driving forces (chemical reaction, physical separation and targeted energy input) in one place and at the same time in one apparatus.
The new applications of alternative energies mean fundamental technological improvement for various types of reactive separation processes.
Specifically, the new microwave technology enables
• Improvement in reaction yield or selectivity due to in-situ product removal, equilibrium shift, and microwave activation. It has been shown that microwaves can drastically speed up several catalytic reactions up to several orders of magnitude depending on the operating conditions . Shorter reaction times imply lower total heat input and thus energy savings. As an example, Table 5 shows a comparison of the energy requirements to perform three types of reactions under conventional and microwave heating . Significant energy savings (up to two orders of magnitude) could be obtained under microwave heating.
• Improvement in separation efficiency. Separation is facilitated by reaction and by microwave input in operating windows that are not attainable via conventional operation; this can be translated into lower reflux ratios for a given production capacity, resulting in lower heat duty and energy savings. However, super equilibrium conditions, which were previously reported for MW heating (17), have not been verified for the current chemical system in the synthesis of EMC and DEC (Werth at al. 2015).
• Improvement in resource efficiency. It has been shown that reactions compatible with reactive distillation processes can be performed with significantly lower excess of the non-limiting reactant under microwaves. As an example, it has been shown that transesterification of soybean oil with methanol was performed with 5:1 methanol oil under microwave heating vs. 9:1 under conventional heating, an 80% increase in resource efficiency .
• Reduction of investment costs by up to 25% due to smaller equipment via more compact design and shorter reaction times and improved separation efficiency implying a lower number of contact stages
Table 5: Energy savings in chemical reactions performed with MW15
Reaction Thermal yield (%) / energy (kW/mol) Microwave yield (%) / energy (kW/mol)
Heterogeneous Suzuki 28.6/120700 77.4/1448
Friedel-Crafts acylation 100/13.6 100/9.7
Knoevenagel 68.2/132 18.2/27

Direct methanol synthesis from methane partial oxidation with non-thermal plasma
The most common method for production of methanol (and other synthetic fuels) is via the combination of methane steam reforming with Fischer-Tropsch (FT) synthesis. Overall, this is a multistep energy consuming method. Direct conversion of methane to methanol has been a research quest for decades. However, past efforts have largely faltered because of their inability to prevent the initial methanol oxidation product from being degraded into further, unwanted, derivative oxidation products (such as formaldehyde, carbon dioxide, and water) mainly on transition metal oxides, basic oxides and iron complexes in zeolites. , On the other hand, the process of reductive conversion of carbon dioxide with hydrogen (CO2 + 3H2 → CH3OH + H2O), which should ideally be produced from water by electrolysis with a net consumption of wind, solar, hydroelectric, geothermal, or atomic energy, is a very attractive alternative to provide a long term sustainable and environmentally benign solution for fuel and energy production (concept of “methanol” economy).
The bottleneck in this process is that reduction of the highly stable CO2 molecule to CO is limited by both the thermodynamic equilibrium and the kinetic barrier. In Ref , it is reported that even at process temperature as high as 900o C only 16% of CO2 conversion on Pd/SCFZ catalyst is attained with good selectivity to CO (>90%). Of course such extreme process conditions make the process energetically and economically unfavourable. Another process route for methanol production utilizing CO2 is the CAMERE process , which combines in series the high temperature reverse water gas shift process (CO2 + H2 → CO + H2O, 773 K, 10Atm) with the high pressure methanol synthesis process from syngas (CO + 2H2 → CH3OH, 573 K, 30Atm). Simple calculations show that approximately equal amounts of energy (~ 8 kJ/mol) are required to pressurize syngas to 30 atm and to raise its temperature to 573 K in the methanol synthesis reactor. It follows that 50% of the energy input can be saved only by enabling the process at ambient pressure. This implies significant economic benefits. According to Gradassi and Green , the operating cost of a direct methanol synthesis plant is 36 mln $ per year. It follows that 50% reduction in energy requirements entails savings of over 10 mln $ per year.
In this project, we employed a novel combined non-thermal plasma–thermal catalysis approach to address the aforementioned bottleneck. Using non-thermal plasma, CO2 and H2 can be easily excited . The excited CO2 and H2 have sufficient energy to break down the limitation of high activation energy of direct hydrogenation of carbon dioxide at mild conditions (moderate temperature and ambient pressure). As mentioned above, only pressure reduction can bring about 50% reduction in the energy input as mentioned above. This is in fact a conservative estimate, as the elimitation of the energy needs for the reverse water gas shift process are not taken into account. Furthermore, capital cost reduction by merging the two sequential processes into one unit operation will be achieved as well. However, the achievement of high product selectivity is known to be a challenge in plasma-assisted processes. To this end, focus will be given on the development of tailored catalysts that optimize the synergy with plasma at mild process conditions resulting in a system that can exhibit both substantially improved energy efficiency and high product yield compared to the thermally activated process. Finally, recent relevant works on non-thermal plasma-assisted catalytic methane reforming support our expectations. For example, Jasiński et al. and O. Mutaf-Yardimci et al. report a 4-fold and 10-fold increase, respectively, in the energy efficiency of hydrogen production with the use of non-thermal plasma compared to the conventional reforming processes (steam and dry reforming, respectively).
Technological impact and risks
Despite the long known capabilities of ultrasound, microwaves and plasma to boost the selectivity, conversions and speed of reaction and separation, widespread application has been hampered by the lack of technological know-how on the reactor design and the optimization and control of the application of these external fields in industrial processes. The research in this project is focused on the discovery and description of the mechanisms involved in the effects of the external fields on the reactions, mass transfer and separation phenomena and will allow the development of the technologies needed for the efficient design and optimization of the reactors and processes in the manufacturing of pharmaceutical and green fuels syntheses. The ALTEREGO consortium forms an ideal mix of technology providers, able to translate the acquired technological knowledge into commercial end products and end users which, will directly benefit from the increased knowledge on the improvement of the manufacturing processes.
The intensification of the pharmaceutical production and green fuel synthesis by the application of the alternative energy sources is advantageous for the European industry and can enhance their competitiveness enormously. In addition, it generates new business opportunities, and will provide the chemical and biochemical industry with novel approaches regarding production, optimization and control of multi-phase, particulate processes.
European Transnational Approach
In order to develop and implement the intensified material- and energy-efficient sustainable process for the manufacturing of chemical or pharmaceutical products we need to strengthen the cooperation between European industry and research centers, which are nowadays threatened in the highly competitive global economic market. As many different technological aspects play a role in the realization of the intensification of reactors and processes, technological expertise is widely spread over Europe. Therefore, collaboration and sharing of knowledge goes beyond borders of individual European countries. The ALTEREGO consortium consists of four academic research centers from Belgium, Germany, United Kingdom and the Netherlands, two technology providers from Germany and France and two end users from Belgium and the Netherlands that form altogether an ideal mix to enable an efficient research strategy for the desired technological breakthroughs.
The realization of this project brought an improvement of the fundamental knowledge on process intensification and integration. This knowledge accumulation in the involved research centers will eventually lead to new activities improving the employment in different sectors and will therefore, with the time, have a socially positive impact on the European Community. In addition, the industries and SMEs involved in this project already benefit from the novel developments and related markets that improve their short and long-term profits.

List of Websites:
www.alterego-project.eu
coord.alterego@bci.tu-dortmund.de

Coordinator: Andrzej Stankiewicz
a.i.stankiewicz@tudelft.nl

Technical Manager: Georgios Stefanidis
georgios.stefanidis@kuleuven.be

Administrative Manager: Dorota Pawlucka
dorota.pawlucka@tu-dortmund.de

Contact

Andrzej Stankiewicz, (Head of the Process Intensification section)
Tel.: +31 15 2786678
E-mail
Record Number: 193445 / Last updated on: 2017-01-12