Application of microwave heating
to a polyesterification plant
Application of microwave heating
to a polyesterification plant
Proefschrift
ter verkrijging van de graad van doctor aan de Technische Universiteit Delft,
op gezag van de Rector Magnificus prof. ir. K.C.A.M. Luyben, voorzitter van het College voor Promoties,
in het openbaar te verdedigen op dinsdag 7 juli 2015 om 10:00 uur
door
Magdalena KOMOROWSKA-DURKA
Chemical Engineer, Warsaw University of Technology geboren te Tarnobrzeg, Poland
Prof. dr. ir. G. D. Stefanidis
Samenstelling promotiecommissie:
Rector Magnificus, voorzitter
Prof. dr. ir. A. I.Stankiewicz, Technische Universiteit Delft, promotor Prof. dr. ir. G. D. Stefanidis, Katholieke Universiteit Leuven, promotor
Dr. M. Radoiu, SAIREM SAS
Prof. dr. J. Gascon, Technische Universiteit Delft Prof. dr. ir. E. Molga, Warsaw University of Technology Prof. dr. J. Meuldijk, Technische Universiteit Eindhoven Prof. dr. R. B. Mato, University of Valladolid
Prof. dr. D.J.E.M. Roekaerts, Technische Universiteit Delft, reservelid
This research was founded and supported by SenterNovem (EOS-LT project).
ISBN/EAN: 978-94-6186-491-8
Cover designed by T. Durka
Copyright © 2015 by M. Komorowska-Durka
All rights reserved. No part of the material protected by this copyright notice may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording or by any information storage and retrieval system, without prior permission of the author.
An electronic version of this dissertation is available at http://www.library.tudelft.nl/ Printed by: Gildeprint Drukkerijen - The Netherlands
Summary ... XI Samenvatting ... XV 1 Introduction ... 1
Industrial polyester production processes ... 2 1.1
Process Intensification aspects in a polyesterification plant ... 4 1.2
1.2.1 The polyesterification reaction ... 4 1.2.2. Treatment of polyesterification reaction water ... 5 Microwave energy – process intensification effects and challenges ... 6 1.3
Scope of the thesis and main research questions ... 8 1.4
Outline of the PhD thesis ...10 1.5
2 Literature review on microwave-assisted polycondensation reactions ... 17
Introduction ...19 2.1
The role of polar molecules, high boiling solvents and heterogeneous catalysts .22 2.2
The role of pressure ...28 2.3
Thermal degradation ...31 2.4
Energy aspects ...32 2.5
The role of fillers in polymer composites and hybrid materials ...38 2.6
Conclusions ...43 2.7
3 Comparative study on polyesterification reaction: multi-mode
microwave cavity vs. internal transmission line (INTLI) ... 51
Introduction ...53 3.1
Industrial polyester production process operated in batch mode ...54 3.2
Experimental ...56 3.3
Stainless steel microwave reactor with internal transmission line (INTLI)...56 3.4
Glass reactor in multimode microwave cavity ...58 3.5
INTLI- vs. Multimode cavity-based microwave equipment ...59 3.6
Reaction system...61 3.7
Estimation of power distribution in the INTLI and multimode cavity 3.8
equipment ...61 Comparison of total specific energy consumption between the INTLI and 3.9
multimode cavity equipment ...66 Conclusions ...69 3.10
Introduction ... 75 4.1
Experimental - materials ... 76 4.2
Set-up and chemical systems description ... 76 4.3
Temperature measurement ... 78 4.4
Power control system and power consumption ... 78 4.5
Product analysis ... 79 4.6
Distribution of energy in the microwave unit ... 80 4.7
Non-reactive system... 81 4.8
Reactive system - pre-treatment time ... 85 4.9
Sampling procedure ... 85 4.10
Conversion in non-catalyzed and catalyzed reaction systems... 88 4.11 Collected distillate ... 91 4.12 Efficiency ... 92 4.13 Conclusions ... 94 4.14
5 Microwave swing regeneration (MSR) vs. temperature swing
regeneration (TSR) – comparison of desorption kinetics... 101
Introduction ... 103 5.1
Experimental: Adsorbent and adsorbates ... 108 5.2 Experimental setups ... 108 5.3 GC Calibration curves ... 111 5.4 Adsorption experiments ... 112 5.5
Microwave assisted regeneration in successive cycles and its influence on 5.6
adsorptive capacity ... 113 Desorption experiments... 114 5.7
Desorption rates with 13X beads ... 118 5.8
Desorption rates with 13X pastilles ... 121 5.9 Desorption efficiencies... 125 5.10 Energy consumption ... 126 5.11 Conclusions ... 127 5.12
6 Application of microwave heating to pervaporation ... 131
Introduction ... 133 6.1
Experimental – materials and analysis ... 135 6.2
Equipment description ... 136 6.3
Temperature measurement and power control ... 136 6.4
Experimental procedures ... 137 6.5
Conclusions ... 145 6.7
7 Conclusions, recommendations and outlook ... 151
Conclusions ... 152 7.1 Recommendations ... 155 7.2 Outlook ... 156 7.3 Acknowledgments ... 159 Curriculum Vitae ... 163 Publications, Oral Presentations, Poster Presentations ... 166
Summary
Application of microwave heating
to a polyesterification plant
Utilizing microwave irradiation, a fundamentally different method of the energy transfer, to the chemical process units can potentially be advantageous compared to the conventional heating, inter alia due to the selective nature of interaction of the microwaves with the matter. This doctoral dissertation addresses some of the aspects, benefits and challenges, associated with the use of the microwave technology in a polyesterification plant.
In this context, three different microwave-assisted processes were investigated: a liquid-phase homogeneous chemical reaction, a membrane separation and a desorption process. These two separation techniques were selected, because they can potentially be used in a downstream recovery section in the polyesterification process (for the water purification and recovery of unreacted dihydroxy alcohols). The study was carried out in two types of microwave units suitable for liquid-phase microwave chemistry and selected separation processes: the multimode microwave ovens and a microwave reactor with the internal transmission line (INTLI).
Although the application of microwave energy has been broadly investigated, most of the research efforts have been focused on obtaining a final product of a reaction in shorter time with potentially higher yield and better product characteristics. The approach of this thesis was different; the objective was to look closely at the differences between two types of processes: microwave- and conventionally-heated, while assuring the same processing conditions and the scale of an experiment. By doing this, there is
better understanding how the chosen process conditions influence the microwave-assisted processes.
A comparative study was conducted to investigate the influence of microwave irradiation (multimode microwave) on liquid-phase reaction kinetics, on by-product removal and on the properties of the final polymer/oligomer products (i.e. Acid Value, molecular weight). The results obtained for a model reaction system of adipic acid and neopentyl glycol under conventional and microwave heating in a similar setup did not show significant differences in terms of conversion and end-product properties. After 3-hour reaction experiments (without catalyst and with SnCl2 catalyst) the conversion of
acid groups was similar under microwave heating and conventional heating. The polyesterification reaction kinetics and mass transfer of water out of the reaction mixture to shift the equilibrium was not affected for the studied polyesterification systems and process conditions.
Studies related to heating of the individual reagents of the polyesterification process in the multimode microwave oven showed that the heating time is several times shorter compared to conventional heating, at the expense of a higher electric energy consumption. Particular emphasis was given on clarifying the important, though scarcely reported in the microwave literature, issue of energy consumption and efficiency. The latter is primarily attributed to rather low magnetron efficiency. Approximately 20-30 % of the electric energy consumed by the microwave oven is converted to thermal energy in the reactor during (non-reactive) heating of the individual components of the polyesterification process. A vast amount of energy is lost in the magnetron and the multimode cavity. From the operational point of view, higher efficiencies under microwave heating in multimode cavities can be attained when using larger reactor volumes.
Contrary to common cavity-based microwave equipment, where the reactor is exposed to a standing microwave field formed in a confined space, the internal transmission line (INTLI) technology allows for irradiation of liquid phase from the interior of the reactor, thereby enabling better coupling of the microwave energy with liquid mixture. The experiment performed in INTLI reactor was compared with one carried out in a common type multimode cavity in terms of specific energy consumption.
The total specific energy consumption with the INTLI reactor is considerably lower owing to the ability to optimize/adapt the microwave power dissipated in the liquid mixture and to minimize the lost (reflected) power that is scattered off the liquid mixture and sent back toward the magnetron. For the studied reactor system (polyesterification reaction of maleic anhydride, phthalic anhydride and 1,2-propylene glycol) and selected process conditions, the total energy consumption with the INTLI reactor was up to a factor two lower (at the beginning of the reaction) compared to the multimode cavity. Optimized (lower) usage of microwave power during the process and effective scale up possibility are inherent advantages of the INTLI reactor technology over cavity-based microwave equipment.
Comparison of desorption kinetics and desorption efficiencies with two techniques, microwave swing regeneration (MSR) and temperature swing regeneration (TSR), was conducted. MSR (performed in a multimode oven) and TSR (performed with conventional heating) of model polar and non-polar molecules (acetone and toluene, respectively) using the 13X molecular sieves were investigated. It was found that, regardless of the dielectric properties of the adsorbates, the 13X molecular sieves absorbed microwaves producing heat within the adsorbent. Therefore, they form a suitable adsorbent for microwave desorption of polar and non-polar compounds. Furthermore, due to the alternative heating mechanism in MSR, the process runs faster even when the adsorbent temperature is relatively lower than the gas temperature in TSR. Direct and instantaneous heating of the adsorbent in MSR resulted in higher desorption rates and higher desorption efficiencies than in TSR. This observation becomes more pronounced when the adsorbate is polar, or when high heat transfer resistances are present in TSR. Faster desorption processes for both adsorbates were obtained under microwave heating with the effect being more pronounced in the event of the polar compound. More specifically, two times higher acetone desorption rate was observed with microwaves, while the measured final temperature of the adsorbent (spherical beads) was 10 oC lower. This shows significant potential of microwave technology to enable more efficient desorption processes due to less energy waste in the form of heat losses and sensible enthalpy of purge gas stream, since the total gas consumption is considerably decreased. Finally, it was verified that microwaves do not
affect the adsorption capacity of the molecular sieves after several consecutive adsorption-desorption cycles.
Membrane pervaporation for dewatering of water/ethanol mixtures, using a hydrophilic membrane, were conducted under microwave and conventional heating in a multimode microwave oven and a convection oven, respectively. Experiments were conducted at three feed stream temperatures (33.5, 45.5 or 51.5 oC), measured in the liquid film close to the center of the membrane, and two feed compositions (5.5 wt% and 20 wt% water in ethanol). It was observed that at 20 wt% water in the feed, the water flux through the membrane was higher under conventional heating. On the contrary, at 5.5 wt% water in the feed, the opposite trend was found; the water flux through the membrane was higher under microwave heating. In the convection oven, temperature is uniform in the entire volume according to the temperature readings in the feed stream, the feed vessel and in the air outside the membrane volume. Conversely, in the microwave oven, only the liquid binary mixture is heated and the temperature environment in the oven is strongly non-uniform. It is likely that due to heat losses in the microwave set-up, the temperature over a part of the membrane surface is lower than the temperature measured at the center, which would in turn result in lower water driving forces in the case of microwaves, and consequently, in lower water fluxes, as observed in the experiments with 20 wt% water in the feed. Enhancement in the permeate flux under microwaves, compared to conventional heating, at higher ethanol concentrations in the feed can be explained on the basis of stronger coupling of microwaves with ethanol than water. Contrary to water, the dielectric loss of ethanol increases with increasing temperature; therefore, microwave dissipation is preponderant in hot areas and can easily lead to local turbulent heating and spatial temperature gradients. The feed composition and its dielectric properties of the pervaporation system influence the process performance.
Samenvatting
Application of microwave heating
to a polyesterification plant
Het toepassen van microgolfenergie als fundamenteel andere methode van energieoverdracht heeft mogelijk voordelen ten opzichte van conventionele verhitting in chemische-procesapparatuur, onder andere door de selectieve aard van de interacties van microgolven met stoffen. Dit proefschrift handelt over enige aspecten, voordelen en uitdagingen van het gebruik van microgolfenergie in een polyesterificatiefrabriek.
Hiervoor zijn drie verschillende microgolf-geassisteerde processen onderzocht: een homogene chemische reactive in de vloeistoffase, een membraanscheiding en een desorptieproces. Deze twee scheidingstechnieken zijn geselecteerd, omdat ze mogeljik kunnen worden gebruikt in een downstream terugwinningssectie in het polyesterificatieproces (voor waterzuivering en het terugwinnen van ongereageerde dihydroxyalcoholen). De studie is uitgevoerd in twee types microgolf apparaten die geschikt zijn voor microgolfchemie in de vloeistoffase en voor de geselecteerde scheidingsprocessen: de multimode oven en een microgolf reactor met interne transmissielijn (INTLI).
Alhoewel de toepassing van microgolf energie al breed is onderzocht, zijn de meeste onderzoeken gericht op het verkrijgen van een eindproduct van een reactie in een kortere tijd met mogelijk een hogere opbrengst en betere producteigenschappen. De aanpak in dit proefschrift was anders: het doel was om beter te kijken naar de verschillen tussen twee types processen: door microgolf en conventioneel verhitte processen, waarin dezelfde procescondities en schaal van experimenten werden aangehouden. Door dit te doen is een beter begrip ontstaan van hoe de gekozen procescondities microgolf-geassisteerde processen beïnvloeden.
Een vergelijkende studie was uitgevoerd om de invloed van microgolfenergie (multimode oven) op vloeistoffase reactiekinetiek te onderzoeken, op bijproduct verwijdering en op de eigenschappen van de uiteindelijke polymeer/oligomeer producten (i.e. zuurwaarde, molecuulgewicht). De resultaten verkregen voor een model reactiesysteem van adipinezuur en neopentylglycol onder conventionele en microgolf verhitting in een vergelijkbare opstelling liet geen significante verschillen zien in termen van conversie en eindproducteigenschappen. Na experimenten met een reactietijd van 3 uur (zonder katalysator en met SnCl2 katalysator) was de conversie van
zuurgroepen vergelijkbaar onder microgolfverhitting en conventionele verhitting. De reactiekinetiek van de polyesterificatie en massaoverdracht van water uit het reactiemengsel om het evenwicht te verschuiven waren niet beïnvloed voor de bestudeerde polyesterificatiesystemen en procescondities.
Onderzoek aan het verhitten van individuele reagenten van het polyesterificatieproces in een multimode oven toonde dat de verwarmingstijd verscheidene malen korter is dan met een conventionele oven ten koste van de hogere elektriciteitsconsumptie. Extra aandacht is gegeven aan het verduidelijken van de belangrijke, doch schaars in microgolfliteratuur gerapporteerde, kwestie van energieconsumptie en efficiëntie. De laatste is voornamelijk te wijten aan de nogal lage magnetron efficiëntie. Ongeveer 20-30 % van de elektrische energie gebruikt door de microgolf oven wordt geconverteerd naar thermische energie in de reactor ten tijde van de (non-reactieve) verwarming van de individuele componenten van het polyesterificatieproces. Een grote hoeveelheid energie gaat verloren in de magnetron en in de multimode resonantieholte. Vanuit een operationeel standpunt kunnen hogere efficiëntie worden bereikt onder microgolfverhitting in multimode holtes als grotere reactorvolumes worden gebruikt.
In tegenstelling tot gewone holte-gebaseerde microgolfapparaten, waar de reactor wordt blootgesteld aan een staand microgolfveld, gevormd in een besloten ruimte, is het bij de INTLI technologie mogelijk om de vloeistoffase van binnenin de reactor aan microgolfenergie bloot te stellen, waardoor de microgolfenergie beter in het vloeistofmengsel gekoppeld kan worden. Het experiment dat werd uitgevoerd in de INTLI reactor was vergeleken met een experiment uitgevoerd in een gewone multimode
holte op basis van specifiek energieverbruik. Het totale specifieke energieverbruik met de INTLI reactor is beduidend lager dankzij de mogelijkheid tot optimaliseren/aanpassen van het microgolfvermogen dat wordt gedissipeerd in het vloeibare mengsel en het minimaliseren van het verloren (gereflecteerde) vermogen dat wordt verstrooid door het vloeibare mengsel en wordt teruggezonden naar de magnetron. Voor het bestudeerde reactiesysteem (polyesterificatie van maleïnezuuranhydride, ftaalzuuranhydride en 1,2-propyleenglycol) en geselecteerde procescondities was het totale energieverbruik met de INTLI reactor tot een factor twee lager (aan het begin van de reactie) dan met de multimode holte. Geoptimaliseerd (lager) verbruik van microgolfvermogen tijdens het proces en effectieve opschaalmogelijkheden zijn inherente voordelen van de INTLI reactortechnologie t.o.v. holtegebaseerde microgolfapparaten.
Desorptiekinetiek en -efficiëntie met twee technieken, microgolf swing regeneratie (MSR) en temperatuur swing regeneratie (TSR), zijn vergeleken. MSR (uitgevoerd in een multimode oven) en TSR (uitgevoerd met conventionele verhitting) van polaire en niet-polaire modelmoleculen (respetievelijk aceton en tolueen) werden onderzocht met 13X moleculaire zeven. Er werd bevonden dat, ongeacht de diëlectrische eigenschappen van de adsorbaten, 13X moleculaire zeven de microgolven absorberen, waardoor warmte werd opgewekt in de adsorbent. Derhalve zijn het geschikte adsorbents voor microgolfdesorptie van polaire en niet-polaire stoffen. Bovendien verloopt het proces sneller, dankzij het alternatieve verwarmingsmechanisme in MSR, zelfs wanneer de adsorbenttemperatuur relatief lager is dan de gastemperatuur in TSR. Directe en instantane verwarming van het adsorbent in MSR resulteert in hogere desorptiesnelheden en hogere desorptie-efficiëntie dan in TSR. Deze waarneming wordt duidelijker zichtbaar als de adsorbent polair is, of bij hoge warmteoverdrachtsweerstanden in TSR. Snellere desorptieprocessen voor beide adsorbaten werden verkregen onder microgolfverwarming, waarbij het effect groter was voor de polaire stof. Specifieker werd een twee maal hogere aceton desorptiesnelheid waargenomen met microgolven, terwijl de gemeten eindtemperatuur van de adsorbent (bolvormige deeltjes) 10 °C lager was. Dit toont de enorme potentie van microgolftechnologie in het mogelijk maken van efficiëntere desorptieprocessen,
doordat minder energie verloren gaat in de vorm van warmteverliezen en sensible enthalpie van de purge gasstroom, omdat de totale gasconsumptie aanzienlijk wordt verkleind. Tenslotte is geverifieerd dat microgolven de adsorptie capaciteit van de moleculaire zeven niet beïnvloeden na enkele opeenvolgende adsorptie-desorptie cycli.
Membraanpervaporatie voor het ontwateren van water/ethanolmengsels, gebruikmakend van een hydrofiel membraan, was uitgevoerd onder zowel microgolf- als conventionele verwarming in respectievelijk een multimode oven en een convectieoven. Experimenten zijn gedaan bij drie voedingsstroomtemperaturen (33.5, 45.5 of 51.5 °C), gemeten in de vloeistoffilm dichtbij het midden van het membraan, en bij twee voedingscomposities (5.5 wt% en 20 wt% water in ethanol). Er werd waargenomen dat bij 20 wt% water in de voeding de waterflux door het membraan hoger was onder conventionele verwarming. Omgekeerd werd bij 5.5 wt% water in de voeding de tegengestelde trend gezien; de waterflux door het membraan was hoger onder microgolfverwarming. In de convectieoven is de temperatuur uniform in het gehele volume volgens de temperatuurwaarden in de voedingsstroom, het voedingsvat en in de lucht buiten het membraanvolume. In de microgolf oven, daarentegen, wordt alleen het binaire vloeistofmengsel verhit en is het temperatuurprofiel in de oven zeer ongelijkmatig verdeeld. Het is waarschijnlijk dat, door warmteverliezen in de microgolfopstelling, de temperatuur op een deel van het membraanoppervlak lager is dan de temperatuur gemeten in het centrum, wat op zijn beurt resulteert in een lagere drijvende kracht van het water in het geval dat microgolven worden gebruikt en derhalve in een lagere waterflux, zoals waargenomen in de experimenten met 20 wt% water in de voeding. Bij hogere ethanolconcentraties in de voeding kan de verhoging van de permeaatflux onder microgolven, in vergelijking met conventioneel verhitten, verklaard worden op basis van de sterkere koppeling van microgolven met ethanol dan met water. In tegenstelling tot water neemt de diëlectrische verliesfactor van ethanol toe met de temperatuur; als gevolg hiervan concentreert de microgolfdissipatie zich in de hete gebieden en kan dit snel leiden tot lokale verhitting en temperatuurgradiënten. De voedingscompositie en de diëlectrische eigenschappen van het pervaporatiesysteem hebben invloed op de prestatie van het proces.
1
Introduction
Industrial polyester production processes 1.1
The process considered in this study is a direct polyesterification reaction between dicarboxylic acids with dihydroxy alcohols of industrial relevance. Although the fundamental production technology of polyesters is well established, many efforts have been put into the intensification of these processes, primarily aiming at reduction in energy consumption. For polyesterifications, long reaction times (in the order of hours) are required, while conversion is limited by equilibrium. Therefore, an effective removal of volatile by-product (water) from the reaction zone is essential to ensure high reaction rate and high molecular weight of the product.
Most polyesters are produced in semi-batch reactors with removal of water by means of a distillation column [1]. The total production time in such a batch process can be more than 24 hours for certain saturated polyester polymers. Heat input is required to heat up the materials to the required reaction temperature, to evaporate water and to separate glycol, water and other components in the distillation column. Most of the time, heating is provided by oil-heating systems and during the start-up phase, heat transfer is the rate-limiting step that often leads to batch-to-batch inconsistency. Process intensification in terms of alleviating heat transfer limitations is required to reduce the production time, achieve a better product quality and increase energy efficiency.
In this section, an industrial polyesterification production process of DSM (Zwolle, the Netherlands) is briefly described to explain the energy demand discussed below. The process flow diagram is presented in Figure 1.1. In the current batch process, some raw materials are stored at ambient temperature and other materials are stored at elevated temperatures. Hot oil supplied from the hot oil grid is used for maintaining the raw material storage temperature. A gas-powered heater maintains temperature in the hot oil grid. Raw materials are dosed to the batch reactor. Heating of the reactor is maintained by hot oil. Flow of nitrogen is used to assure the product quality. Vacuum is used to withdraw water from the reactor contents. Volatile raw materials that evaporate together with water are separated in the distillation column mounted on the top of the reactor. Vapours leaving the top of the column are condensed. Non-condensable gasses are fed to an incinerator. The condensed liquid is collected in a reflux vessel and part of the liquid is fed back to the separation column. At the end of the batch, the content of the reflux vessel is fed to an incinerator. Excess of heat from the incinerator is used for heating the hot oil grid. When the product specifications in the batch reactor are reached,
the content is cooled down with water on a cooling belt and flaked. Flakes are then sent to product storage tanks, before they are sent to packaging and warehouse. It has been estimated that the total energy demand per kg of polyester product (product acid value (AV=50 mgKOH/g polymer)) in the aforementioned plant is 10 MJ/kg. This value is meant to serve later as reference for the specific energy consumption values when using microwave heating. Unfortunately, the direct comparison between the two cases, the industrial and the microwave laboratory scale one, is not possible as the above value of energy consumption concerns the entire heat-integrated plant. Nonetheless, the above information gives a reasonable feeling regarding the respective energy consumption values using microwave technology, in comparison to industrial practice.
Figure 1.1. Process flow diagram of an industrial polyester production process operated in batch mode
Process Intensification aspects in a polyesterification plant 1.2
1.2.1 The polyesterification reaction
In the initial phase of a direct polyesterification process, the esterification reactions dominate converting monomers into oligomers with rather low molecular weight. At this stage, the viscosity is relatively low and water by-product can be effectively removed with intense agitation. Therefore, as mentioned in the previous section, production of polyesters is traditionally performed in a semi-batch reactor with a distillation column directly coupled to the reactor vessel in order to avoid excessive loss of reactants during the process and to facilitate water removal. Usually, nitrogen or xylene are used as stripping agents in a batch reactor process. However, due to the large reflux ratios required, the energy consumption can be significant [2].
In the later phase of the process, conversion of low-molecular-weight oligomers into polymers occurs. As the molecular weight of the polymer increases, the interfacial mass transfer of volatile products becomes a major controlling factor. For some finishing polymerization stages, the reactors are designed in such a way that the rate of the mass transfer from a highly viscous polymer melt to a vapour phase is maximized. Other types of reactors used in melt polycondensation having high surface-to-volume ratio is the horizontal reactor, also called the wiped-film reactor [3, 4], equipped with double- or single-screw [5] [6], or a rotating disk [7]. Another common type of reactor with high surface area is the vertical falling-film reactor [8], [9].
Replacement of the distillation column system can result in the reduction of energy consumption and of the equipment size. In-situ removal of the water from the reaction mixture by means of a membrane is feasible (by pervaporation or vapour permeation). According to the hydrodynamics study of a pervaporation membrane reactor for resin production, the reduction in concentration and temperature polarization is important in order to obtain high water fluxes [10, 11].
Another attractive alternative for a polyester production, according to a modelling study, may be a reactive distillation column [12]. The modelling study shows that the reaction time to produce unsaturated polyesters in a reactive distillation column (reaction of maleic anhydride with 1,2-propylene glycol) can be reduced by a factor 6 to 8 in comparison with an industrial reactor operating in a batch mode. Furthermore, the
equilibrium conversion in a batch reactor process is ~88-90 %, while in the reactive distillation process; the equilibrium conversion can reach up to 97 % [12].
Conducting a polyesterification reaction in a biphasic reaction system composed of an aqueous reaction phase and an organic extraction phase is a new approach where the shift of equilibrium to higher product yields is achieved by integrated product removal (IPR) [13, 14]. The organic phase extracts the polyester product continuously from the aqueous reaction, enabling higher product concentrations. The long-chain monomers have a higher partition coefficient and therefore higher driving force to pass into the organic phase; the advantage is that the obtained polyester is characterized by a narrow molecular weight distribution [13, 14].
1.2.2. Treatment of polyesterification reaction water
According to the report by California Polymer and Resin Manufacturers [15], the condensate from a polyesterification reaction is routed to distillation column or series of distillation columns where a separation of the dihydroxy alcohols from the water stream takes place. Two aspects are most important; a) recovery of water that could be used in another unit operation, e.g. in cooling towers; b) recovery and recycling of the organics in order to be reused in the polyesterification process. The recovered glycols stream is then recycled to the reaction vessel. Maintaining the stoichiometric ratio of diol to acid needs to be assured in the reactor, in order to meet the final polymer properties. Water content, which could be present in the back-routed glycol stream, results in lower polyesterification reaction rate; therefore, the high purity of re-cycled glycol is favourable [15, 16].
In a typical industrial polyesterification process, different combinations of monomers (dicarboxylic acids and diols) and additives are used, causing a diversified content of the removed aqueous stream. A typical aqueous stream consists of unreacted compounds, partially reacted monomers, derivatives of monomers, additives and water. Water from the resin manufacturing process may contain: glycols, dibasic acids, phthalic acids, phenols, aldehydes, cyclic ethers, styrene, 2-ethyl-1-hexanol end others [17]. Moreover, the process conditions influence the amount and the toxicity rank of the generated hazardous and the ignitable wastewater streams. Due to the complex composition of the process wastewater, no single technology is adequate to treat (clean)
it. Usually, a combination of various treatment steps is needed to effectively remove the organic species present in the wastewater [15].
In case the condensate is not purified properly and remains unrecycled, it becomes a wastewater stream. This non-recovered part of the wastes is usually directed to biological treatment. If the biological treatment is not feasible, it is burned in an incinerator [15, 18].
Performing the process in a closed loop with recycling of recovered glycols, as a first step, and purification of water from VOCs residue, as a second treatment step, before discharging to the wastewater treatment system or reusing in a plant as cooling water, seems to be a better option than incineration [19]. Incineration of wastewater at high water content in the stream has high energy demand [20]. Therefore, pre-concentration of organics in the wastewater stream prior to incineration is motivated by the reduction in the overall waste management process cost through decrease in the mass flow to the incinerator and increase in the caloric value of this stream.
The characteristic feature of the organic compounds that contain more than one – OH group (for example glycols) is that they are miscible in water. As these compounds have high boiling points and strongly hydrophilic nature (high affinity to water) due to the presence of at least two hydroxyl groups, their separation forms a challenging task [21, 22]. Conventional evaporation and distillation techniques for the separation of diols from the aquatic system require many stages and high energy consumption to assure the desired separation purity. As an example, the recovery of ethylene glycol with 70-80% purity [20] from an ethylene glycol – water mixture was ranked as the eighth most energy intensive distillation process in chemical industry [23]. Therefore, attention is paid to alternative separation methods, such as the hybrid membrane separation process presented by Jehle et al. [19]. It was claimed by the authors that a process combining three steps: evaporation, reverse osmosis and pervaporation exhibited 3.5 times lower operation cost than incineration [19]. Another alternative to the common distillation process is the selective recovery of the dihydroxy alcohols by adsorption from the liquid condensate stream of the polyesterification process [24].
Microwave energy – process intensification effects and challenges 1.3
Heat transfer to the reaction zone is one of the bottlenecks clearly identified in industrial chemical processes [25]. In view of rising energy costs, the development of more energy-efficient processes has become a popular trend in chemical engineering.
Thus, application of the process intensification (PI) principles through the use of alternative sources and forms of heat, among others acoustic waves and electromagnetic energy, might contribute to the development of sustainable industrial processes [26-28].
In the last couple of years, new interesting insights have been obtained concerning the use of microwave irradiation for intensification of reaction and separation systems. Examples of fields where Process Intensification effects by application of microwaves were reported include:
• Organic chemistry • Medicinal chemistry • Polymer chemistry • Gas-solid chemistry • Biomass/waste pyrolysis • Combustion synthesis • Plasma chemistry • Adsorbent/catalyst regeneration • Crystallization • Extraction • Pervaporation
The advantages of microwave heating arise from the specific interactions of particular materials with electromagnetic fields and observed effects are attributed to different interactions (some of the reported microwave effects are summarized in Table 1.1.). The transformation of electromagnetic energy into heat takes place within the whole irradiated volume of microwave-absorbing materials/substances [29] and can address heat transfer limitations [30]. Materials such as water, alcohols, ionic liquids, and carbon readily absorb microwave energy, whereas other materials such as plastics, oils and non-polar solvents are effectively transparent. The magnitude of energy-matter interaction is determined exclusively by the dielectric properties of the materials [31]. Nonetheless, one of the main limitations of microwaves is the short penetration depth, which is one of the main bottlenecks of microwave equipment preventing effective scale up of microwave technology for chemical reaction applications [32-41]. Therefore, the majority of research works is still focussed on small-scale applications. Depending on the reaction mixture and its dielectric properties, the microwave penetration depth, at the most common microwave frequency of 2.45 GHz, is usually in the order of just a
few centimetres [42]. This can be partly overcome by applying lower ISM (industrial, scientific and medical) frequencies (e.g. 0.915 GHz) [43, 44].
Table 1.1. Summary of reported microwave effects (in comparison to thermal heating) and explanation
Reported microwave effects (in comparison to thermal heating)
Faster chemistry Improved selectivity
Potential for energy savings due to reduction in processing time Improved materials properties:
- Structure and molecular weight of polymers - Composition of co-crystals
- Mechanical properties
Explanation
- Selective (catalyst) heating - Superheating of polar solvents - High heating rates
- Hot spot at microscale
- Direct interaction of the electric component of the EM field with specific molecules/bonds
- Microdischarges - Electrochemical effects
Scope of the thesis and main research questions 1.4
This doctoral dissertation addresses selected aspects of the use of microwaves, instead of conventional heating, in a polyesterification plant. One of the aspects investigated concerns the polyestrerification kinetics and equilibrium shift due to water removal from the reaction mixture. Furthermore, application of microwave irradiation to pervaporation and adsorption-desorption, which can potentially be used downstream for water purification and recovery of unreacted dihydroxy alcohols, is explored. The microwave-activated processes were compared with conventional ones. The main challenges in a polyesterification plant along with the relevant research questions are included in Table 1.2.
Table 1.2. Challenges in a polyesterification process, potential solutions and research question
# Challenges in a polyesterification process
Potential solution Research question
1)
Batch to batch inconsistency due to heat transfer limitation
Faster melting and/or more homogeneous temperature distribution of the reaction mixture in the reactor.
Q1: Can microwave heating intensify melting of solids and consequently provide more uniform temperature distribution due to selective microwave - liquid (and solids) interactions? 2) Long processing time Microwave activation (thermal/non-thermal effects).
Q2: Can microwaves intensify the polycondensation
(polyesterification) reaction kinetics? If so, what is the mechanism (thermal vs. non-thermal effects) 3) Equilibrium limited reaction Shift equilibrium to higher conversions by microwave activation that can possibly enable more effective water removal from the reactor.
Q3: Can microwave heating enable more effective water removal from a viscous polymer mixture?
Q4: What is the most appropriate type of microwave equipment in terms of scalability and energy efficiency to perform a polycondensation reaction? 4) Post-treatment of the condensed aqueous-organic mixture produced from the polycondensation reactor is done by means of energy intensive distillation Potentially less energy-expensive separation techniques are the adsorption-desorption and/or pervaporation. We study them under microwaves to explore the intensification potential of these alternative separation methods.
Q5: Can microwaves intensify the organics desorption step and what is the difference of desorption rate and efficiency for polar/non-polar adsorbates? Q6: Can microwaves enhance pervaporation, which can potentially be used for water removal and concentration of the organics downstream of the polycondensation reactor?
Outline of the PhD thesis 1.5
The dissertation consists of seven chapters reporting on application of microwave irradiation to different steps of polyester production and downstream processing. In Table 1.3., the thesis structure is summarized. Figure 1.2. in turn presents a scheme of the process flow diagram of an industrial polyester production process with a schematic representation of the PhD thesis. In the introductory section (Chapter 1) of this thesis, the fundamental aspects of a conventional polyester production process together with the challenges associated with PI, and an introduction to the field of microwave-assisted processes are discussed.
A concise review on application of microwaves to polycondensation reactions is presented in Chapter 2, while the two following chapters present experimental work. In Chapter 3, the focus is on energy distribution and energy efficiency in the event of microwave-assisted polyesterification experiments conducted in two different microwave reactor types. In this context, a novel microwave applicator with an internal transmission line (INTLI) is presented and applied to a reactive process for first time. In this type of equipment, an antenna is immersed in the reaction mixture and the electromagnetic energy is converted to heat inside the reactor with about 90 % efficiency, which is higher than in conventional multimode applicators. In Chapter 4, the influence of microwave irradiation on the kinetics of a homogeneous polycondensation system, on by-product removal and on the properties of the final polymer/oligomer products (i.e. Acid Value, molecular weight, molecular weight distribution, and structure) is investigated. The results obtained under microwave heating are compared to a conventionally heated process carried out at the same scale in similar equipment.
Chapter 5 and Chapter 6 focus on experimental feasibility studies on microwave activated-separation processes, which can potentially be used for water purification and the recovery of unreacted dihydroxy alcohols downstream, as mentioned earlier. The separation processes are pervaporation and adsorption-desorption and were investigated using model compounds. In Chapter 5, the results of the study on microwave-assisted desorption are presented. Microwave swing regeneration and temperature swing regeneration of polar and non-polar molecules (acetone and toluene) from the 13X molecular sieves were investigated and compared with each other in terms of desorption kinetics and desorption efficiencies. An effect of microwaves on the adsorption capacity
of the 13X molecular sieves after several consecutive adsorption-desorption cycles were investigated. Chapter 6 presents the experimental study on microwave-assisted pervaporation for dewatering of water/ethanol mixtures, using a hydrophilic membrane. Experiments were conducted at different water/ethanol ratios in the feed stream. The fluxes of water and ethanol obtained under microwave heating were compared to the results obtained under conventional heating in a convective oven.
The main conclusions from the work are summarized in Chapter 7. In addition, recommendations for further research are provided, while the chapter, and the dissertation itself, is concluded with a personal view on the valorisation potential of microwave technology in chemical process industry.
Figure 1.2. Process flow diagram of an industrial polyester production process operated in batch mode with a schematic representation of the PhD thesis
Table 1.3. PhD thesis composition
Introduction: conventional polyester production process and the challenges
associated with process intensification via microwaves
Chapter 1
Polyesterification reaction Separation Part I: Part II:
Literature review on microwave-assisted polycondensation reactions Chapter 2 Microwave swing regeneration vs. temperature swing regeneration: comparison of desorption kinetics Chapter 5 Comparative study on polyesterification reaction: multi-mode microwave cavity vs. internal transmission line (INTLI)
Chapter 3
Comparative experimental study: microwave vs. conventional heating of a polyesterification reaction Chapter 4 Application of microwave heating to pervaporation Chapter 6 Chapter 7
Bibliography
[1] T. Salmi, E. Paatero, J. Lehtonen, P. Nyholm, T. Harju, K. Immonen, H. Haario, Polyesterification kinetics of complex mixtures in semibatch reactors, Chemical Engineering Science 56 (2001) 1293-1298.
[2] J.T.F. Keurentjes, G.H.R. Janssen, J.J. Gorissen, The esterification of tartaric acid with ethanol: Kinetics and shifting the equilibrium by means of pervaporation, Chemical Engineering Science 49 (1994) 4681-4689.
[3] A. Nisoli, M.F. Doherty, M.F. Malone, Feasible Regions for Step-Growth Melt Polycondensation Systems, Industrial & Engineering Chemistry Research 43 (2004) 428-440.
[4] M. Amon, C.D. Denson, Simplified Analysis of the Performance of Wiped-Film Polycondensation Reactors, Industrial & Engineering Chemistry Fundamentals 19 (1980) 415-420.
[5] R. Giudici, C.A.O. do Nascimento, I.C. Beiler, N. Scherbakoff, Transient Experiments and Mathematical Modeling of an Industrial Twin-Screw Extruder Reactor for Nylon-6,6 Polymerization†, Industrial & Engineering Chemistry Research 36 (1997) 3513-3519.
[6] C. Laubriet, B. LeCorre, K.Y. Choi, Two-phase model for continuous final stage melt polycondensation of poly(ethylene terephthalate). 1. Steady-state analysis, Industrial & Engineering Chemistry Research 30 (1991) 2-12.
[7] S.I. Cheong, K.Y. Choi, Melt polycondensation of poly(ethylene terephthalate) in a rotating disk reactor, Journal of Applied Polymer Science 58 (1995) 1473-1483. [8] G.A. Campbell, E.F. Elton, E.G. Bobalek, The kinetics of thin film
polyesterification, Wiley Subscription Services, Inc., A Wiley Company, 1970, pp. 1025-1035.
[9] T.C. Moore, S.S. Griffth, D.E. Thompson, N.R. Kluesener, J.R. Honeycutt, Methods for making polyester resins in falling film melt polycondensation reactors, WO2007127786 (A2) ― 2007-11-08.
[10] S. van der Gulik, Design and Scale-up of Polycondensation Reactors, PhD dissertation 2002.
[11] G.J.S. van der Gulik, R.E.G. Janssen, J.G. Wijers, J.T.F. Keurentjes, Hydrodynamics in a ceramic pervaporation membrane reactor for resin production, Chemical Engineering Science 56 (2001) 371-379.
[12] M. Shah, E. Zondervan, A.B. de Haan, Modelling and Simulation of an Unsaturated Polyester Process, Journal of Applied Science 10 (2010) 2551-2557. [13] J. Duwensee, S. Wenda, W. Ruth, U. Kragl, Lipase-Catalyzed Polycondensation in
Water: A New Approach for Polyester Synthesis, Organic Process Research & Development 14 (2010) 48-57.
[14] C. Korupp, R. Weberskirch, J.J. Müller, A. Liese, L. Hilterhaus, Scaleup of Lipase -Catalyzed Polyester Synthesis, Organic Process Research & Development 14 (2010) 1118-1124.
[15] D. Weightman, Assessment of California Polymer and Resin Manufacturer's Hazardous Waste Source Reduction Planning Efforts, Office of Pollution prevention and Technology Development, State of California, 1996.
[17] J. Noblet, L. Schweitzer, E. Ibrahim, K.D. Stolzenbach, L. Zhou, I.H. Suffet, Evaluation of a Taste and Odor Incident on the Ohio River Water Science and Technology 40 (1999) 185-194
[18] C. Rincón, J.M. Ortiz de Zárate, J.I. Mengual, Separation of water and glycols by direct contact membrane distillation, Journal of Membrane Science 158 (1999) 155-165.
[19] W. Jehle, Staneff, J. Steinwandel, Separation of Glycol and Water from Coolant Liquids by Evaporation, Reverse Osmosis, and Pervaporation, J. Membr. Sci. 102 (1995) 9.
[20] R.Y.M. Huang, P. Shao, X. Feng, W.A. Anderson, Separation of Ethylene Glycol-Water Mixtures Using Sulfonated Poly(ether ether ketone) Pervaporation Membranes; Membrane Relaxation and Separation Performance Analysis, Industrial & Engineering Chemistry Research 41 (2002) 2957-2965.
[21] R.R. Broekhuis, S. Lynn, C.J. King, Recovery of Propylene Glycol from Dilute Aqueous Solutions via Reversible Reaction with Aldehydes, Ind. Eng. Chem. Res. 33 (1994) 3230.
[22] A.D. Dhale, L.K. Myrant, S.P. Chopade, J.E. Jackson, D.J. Miller, Propylene glycol and ethylene glycol recovery from aqueous solution via reactive distillation, Chemical Engineering Science 59 (2004) 2881-2890.
[23] J.L. Bravo, J.R. Fair, J.L. Humphrey, C.L. Martin, A.F. Seibert, S. Joshi, Fluid mixture separation technologies for cost reduction and process improvement, 1986. [24] B. Debruin, Roger, Adsorber system to replace water column in a polyester process,
US, 2000.
[25] Technology Outlook (based on 42 PI Scans supported by AGNL), http://www.traxxys.com/Download%20default.html.
[26] J.A. Moulijn, A. Stankiewicz, J. Grievink, A. Górak, Process intensification and process systems engineering: A friendly symbiosis, Computers & Chemical Engineering 32 (2008) 3-11.
[27] A. Stankiewicz, Energy Matters: Alternative Sources and Forms of Energy for Intensification of Chemical and Biochemical Processes, Chemical Engineering Research and Design 84 (2006) 511-521.
[28] T. Van Gerven, A. Stankiewicz, Structure, Energy, Synergy, Time - The Fundamentals of Process Intensification, Industrial & Engineering Chemistry Research 48 (2009) 2465-2474.
[29] B.L. Hayes, Microwave Synthesis - Chemistry at the Speed of Light, CEM Publishing, Metthews, 2002.
[30] D.M.P. Mingos, D.R. Baghurst, Tilden Lecture. Applications of microwave dielectric heating effects to synthetic problems in chemistry, Chemical Society Reviews 20 (1991) 1-47.
[31] A.C. Metaxas, R.J. Meredith, Industrial Microwave Heating, IEEE Power Engineering series 4, (1983).
[32] D. Dallinger, H.r. Lehmann, J.D. Moseley, A. Stadler, C.O. Kappe, Scale-Up of Microwave-Assisted Reactions in a Multimode Bench-Top Reactor, Organic Process Research & Development 15 (2011) 841-854.
[33] T. Durka, G.D. Stefanidis, T. Van Gerven, A.I. Stankiewicz, Microwave-activated methanol steam reforming for hydrogen production, Int. J. Hydrog. Energy 36 (2011) 12843-12852.
[34] J.D. Moseley, C.O. Kappe, A critical assessment of the greenness and energy efficiency of microwave-assisted organic synthesis, Green Chemistry 13 (2011) 794-806.
[35] N.A. Pinchukova, V.A. Chebanov, N.Y. Gorobets, L.V. Gudzenko, K.S. Ostras, O.V. Shishkin, L.A. Hulshof, A.Y. Voloshko, Beneficial energy-efficiencies in the microwave-assisted vacuum preparation of polyphosphoric acid, Chemical Engineering and Processing: Process Intensification 50 (2011) 1193-1197.
[36] F. Benaskar, V. Engels, N. Patil, E.V. Rebrov, J. Meuldijk, V. Hessel, L.A. Hulshof, D.A. Jefferson, J.C. Schouten, A.E.H. Wheatley, Copper(0) in the Ullmann heterocycle-aryl ether synthesis of 4-phenoxypyridine using multimode microwave heating, Tetrahedron Letters 51 (2010) 248-251.
[37] A. Navarrete, R.B. Mato, M.J. Cocero, A predictive approach in modeling and simulation of heat and mass transfer during microwave heating. Application to SFME of essential oil of Lavandin Super, Chemical Engineering Science 68 (2012) 192-201.
[38] E. Altman, G.D. Stefanidis, T. Van Gerven, A. Stankiewicz, Microwave-promoted synthesis of n-propyl propionate using homogeneous zinc triflate catalyst, Industrial and Engineering Chemistry Research 51 (2012) 1612-1619.
[39] E. Altman, G.D. Stefanidis, T. Van Gerven, A.I. Stankiewicz, Process intensification of reactive distillation for the synthesis of n -propyl propionate: The effects of microwave radiation on molecular separation and esterification reaction, Industrial and Engineering Chemistry Research 49 (2010) 10287-10296.
[40] V.L. Budarin, J.H. Clark, B.A. Lanigan, P. Shuttleworth, S.W. Breeden, A.J. Wilson, D.J. Macquarrie, K. Milkowski, J. Jones, T. Bridgeman, A. Ross, The preparation of high-grade bio-oils through the controlled, low temperature microwave activation of wheat straw, Bioresource Technology 100 (2009) 6064-6068.
[41] M.J. Gronnow, V.L. Budarin, J. Fan, P.S. Shuttleworth, D.J. Macquarrie, J.H. Clark, Microwave pyrolysis of biomass, 2011, pp. 679-682.
[42] S.G. Camelia Gabriel, Edward H. Grant, Edward H. Grant, Ben S. J. Halstead and D. Michael P. Mingos, Dielectric parameters relevant to microwave dielectric heating, Chem. Soc. Rev 27 (1998) 213 - 224.
[43] M. Mehdizadeh, Engineering and scale-up considerations for microwave induced reactions, Research on Chemical Intermediates 20 (1994) 79-84.
[44] T. Durka, G.D. Stefanidis, T.V. Gerven, A. Stankiewicz, On the accuracy and reproducibility of fiber optic (FO) and infrared (IR) temperature measurements of solid materials in microwave applications, Measurement Science and Technology 21 (2010) 045108.
2
Literature review on
microwave-assisted
polycondensation reactions
i
i
This chapter was published as:
Komorowska-Durka M., Dimitrakis G., Stankiewicz A.I., Bogdał D., Stefanidis G.D. “A concise review on microwave-assisted polycondensation reactions and curing of polycondensation polymers with focus on the effect of process conditions” Chemical
Engineering Journal, Volume 264, 15 March 2015, Pages 633-644, ISSN 1385-8947, http://dx.doi.org/10.1016/j.cej.2014.11.087.
Abstract
During the past 15 years, increasing application of microwave heating to polycondensation reactions has been witnessed. Experiments have been carried out at laboratory scale using widely different experimental procedures. The use of microwaves has often led to significant benefits compared to conventional heating experiments in terms of multi-fold decrease in reaction times and energy consumption and production of polymers with increased molecular weight and improved mechanical properties. In other cases, microwaves do not appear to produce any significant benefits compared to conventional heating. At present, guidelines to experimentalist as to the process conditions and experimental design that should be applied are missing and experimentation seems to be based on an empirical trial-and-error approach. In view of the very different experimental protocols that have been applied and the contradictory trends that are frequently reported, we aim in this review to shed light on the role of important process parameters, such as the presence and type of solvent, the dielectric properties of the mixture and the individual phases, the use of heterogeneous catalysts, pressure, stirring, reflux conditions, temperature measurement method and microwave absorbing fillers, which all seem to determine the occurrence and magnitude of the benefits enabled by microwaves during polycondensation reactions.
Introduction 2.1
Polycondensation reactions are a form of step-growth polymerization reactions widely used in the polymer industry to form proteins, polyesters and polyamides and certain resins and silicones. Effective removal of water and volatile by-products from the reaction zone is essential to ensure high reaction rate, conversion and molecular weight of the final product. One example of step-growth polymerization is the formation of polyesters from diols and diacids according to the general scheme:
HO-R-OH + HOOC-R’-COOH H-[O-R-OOC-R’-CO]n-OH + (2n-1) H2O
Activation of polycondensation reactions by means of microwave irradiation has been reported to be a fast and facile synthesis method. Previous reviews on the subject were given in books [1, 2] and the key review papers [3-10]. Reviews on microwave-assisted polycondensation of polylactic acid, polyesters and polyamides were presented by Hirao and Nakamura [11-13]. Compared to past works, the approach in this review is different. We do not aim at an extensive review of the results from the application of microwaves to different polymer systems and chemistries, but rather, to shed light on the role of important process parameters in microwave-assisted polycondensation processes through selected literature examples; inter alia syntheses of polyamides, polyimides and polyesters. In view of the very different experimental protocols that have been applied and the contradictory trends that are frequently reported, it is important to understand how the choice of process conditions may affect process performance (e.g., process time, energy consumption, conversion and selectivity) and physical and mechanical product properties. Most of the works in which a domestic oven was used without temperature measurement are not included in this review.
In Table 2.1, a summary of key publications focused on polycondensation reactions arranged in reverse-chronological order is presented. For each reviewed paper, the author’s explanation of the observed effect and the experimental conditions are listed, including the type of microwave oven, the temperature measurement technique, the scale of reactor/process, the pressure, the kind of solvent used, and information whether a reference experiment under conventional heating was carried out. In the final section,
the effects of various microwave absorbing fillers on the end-properties of polymer composite and hybrid materials prepared under microwave heating are reviewed.
The advantages of microwave heating arise from the specific interactions of particular materials with high frequency electromagnetic waves. Materials such as water, alcohols, ionic liquids, and carbon readily absorb microwave energy, whereas other materials such as plastics, oils and non-polar solvents are effectively transparent. This is determined exclusively by the dielectric properties of materials [14]. Microwaves can therefore penetrate into the depth of a mixture of components, interacting in different ways with various phases. The materials susceptible to microwave energy will heat selectively and rapidly compared to the non-microwave susceptible materials, quickly generating large temperature gradients [15]. Energy can hence be concentrated selectively within these microwave susceptible components in a mixture, with little or no energy expended on direct heating of the surrounding. Another important aspect of microwave heating is its volumetric nature, which can help overcome heat transfer limitations often present in processes heated by means of heat conduction or convection through the solid boundaries of the system. Consequently, heating times can potentially be several orders of magnitude shorter with microwave heating than with conventional heating on the basis of equal energy input. This feature has been exploited in the field of chemistry and numerous reports in the literature have shown the potential of microwave energy to accelerate chemical reactions [16]. According to the Arrhenius law, the rate constants vary exponentially with the reciprocal of temperature and hence small changes in temperature can significantly affect the speed of chemistry. However, the inability to accurately probe the temperature excursions in chemical phases inside a reaction medium at microscopic level has generated a lot of controversy regarding the actual wave-material interactions [17]. Nevertheless, it is generally accepted that “local” temperatures in the reaction mixture can be significantly higher than the bulk temperature and this may be the reason of acceleration of microwave-assisted reactions in many cases. [18, 19]
The principle and mechanisms that underpin microwave heating are well understood and have been extensively discussed in the literature [14, 20, 21]. In this review, we will provide only a brief description of the mechanisms taking place.
Microwave heating in non-magnetic materials takes places via the interaction of the electric field with charged species (free charge, induced dipoles) in a material. This interaction depends upon the polarisation of the species. There are mainly two mechanisms associated with the heating of materials at radiofrequency and microwave frequencies. The first one is dipolar reorientation (polarisation) and is the dominant mechanism in the GHz region of frequencies. The electric field will interact with the induced dipoles and will force them to rotate until the dipoles are balanced by electrostatic interactions [20]. The molecules’ orientation changes but their speed of motion is limited due to dipole-dipole interactions and molecules are unable to completely relax, i.e., they cannot reorient themselves completely. Movement of species creates friction and leads to thermal effect - heat generation. A perturbed system does not return into equilibrium stage before a next portion of energy is delivered because the relaxation times corresponding to rotation of the whole molecule and to rotation of the polar group are longer than the energy transfer time. The second one is due to the presence of ionic species in solid materials, or ionic solutions, and is prominent at lower frequencies. The applied electric field induces motion to cations and anions in opposite directions and further causes a net dipole moment [22]. In heterogeneous systems, a heating mechanism that relates to the interfacial or Maxwell-Wagner polarisation could also be observed. It originates from a charge build-up in the contact areas or interfaces between different components. This polarisation is due to differences in the conductivity and dielectric properties of the different substances at their interface. As a result, accumulation of charges can occur and lead to field distortions and dielectric heating effects [22].
The extent to which a material heats up when subjected to electromagnetic radiation (e.g., microwave radiation) is mainly determined by its dielectric properties, which can be expressed by the complex permittivity (ε*) that has two components: a
real part, the dielectric constant (ε’) and an imaginary part, the dielectric loss factor (ε”)
and is described by the following equation.
∗= − " Equation 2.1
The dielectric constant (ε’) effectively denotes the ability of a material to be
whereas the loss factor quantifies the efficiency of the material to convert the stored electromagnetic energy into heat [23]. The ratio of loss factor to dielectric constant is termed the dielectric loss tangent of a material:
= ”
Equation 2.2 The dielectric loss tangent of a material determines its ability to absorb and convert the electromagnetic energy into thermal energy at a given temperature and frequency [24] when the E-field is applied to a sample. A material with a high loss tangent ( 10) can be considered as a good microwave absorber and is capable of converting electromagnetic energy into thermal energy. On the other hand, a material with a low loss tangent ( 10) is recognised as a low microwave absorbing material because it cannot be heated sufficiently under microwave irradiation [25].
The role of polar molecules, high boiling solvents and heterogeneous 2.2
catalysts
Polar solvents play an important role as primary microwave energy absorbers in microwave-assisted solution polycondensations and may lead to significant temperature increase in very short times [26, 27]. Furthermore, high boiling point solvents facilitate high reaction temperatures without solvent loss [28]. When non-polar solvents were used in combination with microwaves, lower molecular weight products were obtained due to insufficient heating [29, 30].
For solution polycondensation of aromatic dianhydrides and diisocyanate [27] a polar organic medium was necessary to enable effective homogeneous heating of the monomers in order to produce polyimides. Screening of different high boiling point solvents revealed that the solvent with the highest boiling point from the studied list of solvents (N-methyl pyrrolidinone) gave a polymer with the highest inherent viscosity at conditions of maximum microwave power 900 W and 15 minutes reaction time.
Polycondensation of Poly(ArylEtherKetone)s under microwave heating with different high boiling point polar and non-polar solvents was studied by Brunel et al. [26]. A comparative study with different type of solvents, namely chlorobenzene, N-methyl pyrrolidinone and diN-methyl sulfoxide, which have different dielectric properties, was conducted. Use of dimethyl sulfoxide resulted in the highest molecular weight
polymer. When the weak microwave absorbing solvent, by comparison to the others, chlorobenzene was used as a solvent, only oligomers were produced. This forms an indication that use of solvents with high dielectric loss factor under microwave heating results in more effective energy transfer to monomers and high molecular weight polymers.
Lewis et al. [31] claimed that a non-uniform temperature profile on molecular scale was responsible for the enhancement of microwave-assisted imidization kinetics. Data obtained at 160 °C showed a 34-fold enhancement in the reaction rate constant; at 170
o
C, the enhancement observed was 20-fold in comparison to the conventionally heated process. The authors explained that in the event of conventional heating, the assumption of uniform temperature at macroscopic and molecular scales is true; in the presence of a microwave field however, the direct coupling of microwaves with polar substrates (forced oscillation of the dipoles) may result in local hot spots at molecular scale (in the vicinity of dipole moment). On the basis of this assumption, the authors reported that the actual reaction temperature was 50 oC higher in the vicinity of the dipole moment than the average temperature measured by the fiber optic thermometer.
Solution polycondensation of aromatic polyamides from aromatic diamines with aromatic dicarboxylic acids (terephthalic or isophthalic acid) and a mixture of triphenyl phosphite and pyridine as a condensing agent was carried out under microwaves by Park et al. [32]. The reaction mixtures were heated from 30 to 230 oC in air atmosphere without stirring. The results were compared to data from conventionally heated polycondensations at 220 oC in nitrogen atmosphere with stirring. Somewhat shorter reaction times are reported for the microwave reactions to obtain polymer products with the same range of inherent viscosities and molecular weights.
Synthesis of polyamides from nylon salts with and without solvents was carried out by Watanabe et al. [28]. Temperature was monitored by a teflon-shielded thermocouple. The presence of a high boiling point solvent was necessary to obtain a polymer. In the presence of ethanol, the maximum reaction temperature was 85 oC and nylon-6,6 was not produced. In contrast, in the presence of a high boiling solvent, the reaction temperature exceeded 200 oC within 2 min. Continuous microwave irradiation was applied and nylon-6,6 with inherent viscosity in the range 0,14 - 0,26 dL/g was
