- Email: [email protected]
The use of microwave irradiation for zeolite regeneration in a continuous ethanol dewatering process
Chemical Engineering and Processing 95 (2015) 151–158
Contents lists available at ScienceDirect
Chemical Engineering and Processing: Process Intensification journal homepage: www.elsevier.com/locate/cep
The use of microwave irradiation for zeolite regeneration in a continuous ethanol dewatering process Natalia A. Pinchukovaa,* , Alexander Yu. Voloshkoa , Vyacheslav N. Baumera , Oleg V. Shishkina , Valentin A. Chebanova,b a b
SSI "Institute for Single Crystals" of National Academy of Sciences of Ukraine, Lenin Ave., 60, Kharkiv 61001, Ukraine V.N. Karazin Kharkiv National University, Svobody sq., 4, Kharkiv 61022, Ukraine
A R T I C L E I N F O
A B S T R A C T
Article history: Received 15 January 2015 Received in revised form 25 May 2015 Accepted 2 June 2015 Available online 5 June 2015
The processes of zeolite-assisted ethanol dewatering with further microwave (MW) zeolite dehydration has been studied. Both gas-phase and liquid-phase alcohol dewatering methods yielded absolute ethanol. Kinetic study showed that these processes can be performed under dynamic conditions, which is relevant from a scale-up viewpoint. The concept of “in situ MW-assisted zeolite regeneration” was elaborated. Reduction of energy consumption by 1.7 times with 10-fold process time shortening has been shown for lab-scale MW zeolite dehydration, as compared to the corresponding thermal method. Hypothesis explaining the reasons of such a dramatic energy and time saving was proposed based on the diffusion theory. The influence of MW irradiation on the zeolite structure was studied by powder X-ray diffractometry. The results showed that even after 20 regeneration cycles the zeolite remained the same as the initial sample. The outlook for process scale-up has been discussed. A pilot-plant scale model for continuous ethanol dewatering by NaA zeolite with the following zeolite regeneration has been proposed. ã2015 Elsevier B.V. All rights reserved.
Keywords: Microwave irradiation Ethanol dewatering Zeolite regeneration Energy consumption Continuous process
1. Introduction In the recent years zeolite molecular sieves have been widely used for dehydration or dewatering purposes due to their unique properties to selectively adsorb water from water containing solutions, which is especially topical in a view of growing interest to bioethanol production [1–5]. For ethanol dewatering purposes NaA or KA zeolites are most appropriate, being synthetic zeolites, since they have small aperture diameters (4 and 3 Å respectively), therefore organic molecules having two or more C atoms cannot penetrate inside the zeolite pores and remain in a liquid phase, while smaller molecules like water ones are effectively absorbed and retained by zeolites. In spite of a constantly growing interest to zeolites, most of the papers are dedicated to the study of adsorption properties of zeolites; meanwhile, there is much less information in the literature concerning zeolite regeneration. This gap, undoubtedly, should be supplied, since zeolite regeneration step is crucial from commercial viewpoint, and only efficient dehydration of water-saturated zeolites can ensure their durability and successful application at industrial scale.
* Corresponding author. Fax: +38 57 340 93 43. E-mail address: [email protected] (N.A. Pinchukova). http://dx.doi.org/10.1016/j.cep.2015.06.001 0255-2701/ ã 2015 Elsevier B.V. All rights reserved.
The conventional technique for zeolite dehydration is hightemperature calcination [6,7], which is time and energy consuming because of low thermal conductivity of the zeolite materials. On the other hand, the temperature limitation exists, as high temperatures may cause partial or complete damage to zeolite matrix resulting in decrease in adsorption capacity and in zeolite life-time shortening [8]. For efficient zeolite dehydration and reduction of energy consumption, as well as for minimization of the stress on the zeolite structure, process parameters should be carefully selected. Gabrus et al. proposed in-situ thermal regeneration of 3A and 4A zeolites in the fixed bed after liquid-phase dewatering of aliphatic alcohols [8]. Thermal stability of zeolites, water adsorption capacity, adsorption selectivity and thermal desorption efficiency were carefully studied. The optimal conditions of zeolite dehydration were determined which allowed efficient dehydration of the studied zeolites with maintaining their adsorption capacity approximately at the same level over 30 hydration-dehydration cycles. The proposed technique allows continuous alcohol dewatering with zeolites. Long life-time of zeolites is ensured by properly selected regeneration temperature mode. However, it seems reasonable to improve this “in-situ” approach in terms of process acceleration and energy costs reduction by replacing traditional
152
N.A. Pinchukova et al. / Chemical Engineering and Processing 95 (2015) 151–158
heating sources for a novel and more efficient heating method, such as microwave (MW) irradiation, which shows excellent results when applied to chemical or physical processes [9–13]. It is known that ethanol has a good absorbing capacity towards MW irradiation [14] while water has middle one, nevertheless, MW irradiation is widely used for processing materials or reaction mixtures where water is a medium or one of the components [15,16]. Therefore, MW irradiation has proved to be a promising technique for various adsorbents regeneration [17–19] and intensification of separation processes like pervaporation or membrane distillation [20,21]. In view of the above said, there are all the grounds to predict efficient application of MW irradiation to NaA zeolite regeneration resulting in energy cuts and process intensification. In this study the process of liquid phase ethanol dehydration and further zeolite regeneration with the use of MW irradiation were studied with a fixed bed of NaA zeolite. The aim of this study is the development of a continuous process of dewatered ethanol preparation comprising ethanol dewatering and MW-assisted zeolite dehydration; and determination of optimal process parameters ensuring minimization of energy consumption and process duration as compared to corresponding thermal processes. The effect of MW irradiation on zeolite adsorption capacity after multiple uses was also examined. Process scale-up is considered in this study as well. 2. Materials and methods 2.1. Materials NaA zeolite was purchased from Salavat catalyst plant Ltd. (Russian Federation). Ethanol–water solution with ethanol concentration 95 wt.% and NaA zeolite in the form of extruded cylinders 5–7 mm in length and 1.6 0.2 mm in diameter were used in ethanol dewatering and zeolite regeneration experiments. 2.2. Analytical control 2.2.1. Water content determination Water content in ethanol before and after dehydration step was determined by Karl Fisher method [22]. 2.2.2. Zeolite dehydration degree determination Concentration of the water adsorbed by zeolite and zeolite dehydration degree were determined by the gravimetric method. The saturated zeolite was subjected to calcination at 250 C, and the amount of adsorbed water was calculated by the weight difference before and after calcination. The weight loss was calculated from the Eq. (1): W¼
m1 m2 100%; m2
developed by us [23] according to the following procedure. The quarts cell (r = 0.845 cm, l = 10,4 cm) was placed into a cylindrical MW cavity (r = 4.7 cm, l = 10.4 cm) along the cavity axisppopopl. The resonance frequencies and cavity Q factor were measured with and without the tested samples. On the basis of the measurements results the dielectric constant e0 , loss tangent tg d and skin-layer depth were calculated. 2.3. Experimental set-up and procedure 2.3.1. MW set-up MW zeolite dehydration was performed in a multimode MW laboratory system (MARS, CEM Corp., USA) equipped with IR sensor for temperature measurements. A typical distillation set-up was used for zeolite dehydration experiments (Fig. 1), consisting of a 250-mL round-bottom flask containing saturated zeolite and placed inside the microwave cavity, an upward glass tube (l = 30 cm, Ø = 2.5 cm), passing through the aperture in the upper wall of the MW cavity. A standard distillation kit for collecting water condensate, attached to the column, was placed outside the cavity. The process was conducted under slight evacuation provided by membrane vacuum pump for better removal of water vapor. In the experiments on gas-phase ethanol dewatering, the typical distillation set-up was used, similar to that applied for zeolite dehydration (Fig. 1), with some modifications; viz. the zeolite-loaded column was used instead of a hollow tube, and a fiber-optic (FO) probe was applied for temperature measurements (Fig. 2). In this case ethanol–water mixture and zeolite-loaded column were exposed to MW field, while the distillation kit was placed outside of MW cavity. 2.3.2. Ethanol dewatering and NaA zeolite dehydration Ethanol dewatering was performed by applying both gas-phase and liquid-phase approaches. According to the first method the ethanol–water mixture was distilled, with the vapors passing through a column filled with previously dehydrated NaA zeolite (Fig. 2). The following procedure was applied. Ethanol–water mixture was heated by MW irradiation to the boiling point, and the vapors passed through the column with the fixed bed of NaA zeolite. The necessary ratio of zeolite to ethanol–water mixture was calculated on the basis of NaA adsorption capacity determined previously (see Section 3.1), with the 20% surplus of zeolite. Due to high selectivity of NaA zeolite to water, larger molecules of ethanol were not retained and freely passed through the column, condensed and finally were collected in the receiver.
(1)
where m1 is the weight of saturated zeolite (kg); m2 is the weight of calcinated zeolite (kg). 2.2.3. Powder X-ray diffraction Powder X-ray diffraction patterns of the initial zeolite sample and MW irradiated one were recorded with the powder X-ray diffractometer Siemens D500 using Cu irradiation (l = 1.54184 Å, graphitic monochromator), angular range 3–70 2u, step 0.02 . 2.2.4. Dielectric properties measurements Dielectric properties of the saturated and dehydrated zeolite were measured with the use of the apparatus and methodology
Fig. 1. Schematic image of experimental setup used for MW zeolite dehydration.
N.A. Pinchukova et al. / Chemical Engineering and Processing 95 (2015) 151–158
Fig. 2. Schematic image of experimental setup exploited for MW ethanol gas-phase dewatering with the use of NaA zeolite.
Liquid-phase dewatering was carried out by soaking dehydrated zeolite in water–ethanol mixture. For this purpose weighed amount of zeolite was soaked with weighed amount of ethanol– water mixture and allowed to stay for a definite time. Saturated zeolite obtained after ethanol dewatering was separated from ethanol by decanting of the latter, weighed and dried in a draught until the weight constancy was achieved and no smell of alcohol was detected, indicating that all the surface moisture, mostly consisting of ethanol, had volatilized. After such predrying zeolite containing only adsorbed water in its micropores was calcinated in MW field under slight evacuation with the use of MARS MW system. Temperature control was performed by means of IR-sensor mounted into the bottom of MW cavity, for which purpose the flask was placed on the hollow teflon rest right above the IR-sensor. Temperature MW power profiles were continuously displayed on the screen of PC integrated into MARS system. The temperature distribution in the bulk of NaA zeolite was periodically checked by interrupting the dehydration process and measuring temperature in different points of bulk zeolite by electronic thermometer. The measurements taken by this way were close to the temperature readings yielded by the integrated IR-sensor. The amount of water removed was determined from the zeolite mass difference before and after dehydration. 2.3.3. Adsorption kinetics study Adsorption kinetics was studied by keeping dry zeolite in ethanol–water mixture for a definite time, separating zeolite from the liquid, predrying zeolite in a draught and calcinating it in MW field, as described in Section 2.3.2. Adsorption capacity q was determined from the Eq. (2): q¼
m2 m1 100%; m2
(the maximal applied temperature of 230 C is explained by the IRsensor measuring limit). Prior to each run the zeolite was hydrated by keeping in a dessicator over the water layer until the constant weight was attained. The saturated zeolite containing only adsorbed water was then dehydrated by calcination in microwave field, as described in Section 2.3.2. Kinetics studies revealed that the optimal time required for zeolite dehydration was 20 min., since within this time the zeolite reached constant weight irrespective of the temperature applied (Fig. 3). The curves shapes indicate at the molecular dehydration mode, corresponding to reversible processes. This is the ground to suggest that the multiple hydration–dehydration cycles with the use of NaA zeolite should not affect its structure or properties providing durability of the zeolite under production conditions. Fig. 4 illustrates the dependence of zeolite dehydration degree on the temperature. Though it appears to be linear in the whole studied temperature range and does not reach the saturation limit, the weight loss of 15.5% attained at 230 C indicates that the sorption capacity of the zeolite is sufficient for effective alcohol dehydration. The temperature distribution in a bulk of zeolite, controlled by the means of electronic thermometer, was ranging within 2.5 C of the average value, which was close to temperature readings yielded by the MARS-integrated IR sensor within the measurement error limits. The results obtained are in a good accordance with the literature data for thermal 3A and 4A zeolite dehydration, offering the optimal dehydration temperature at the level of 240 C [8]. The dynamic adsorption capacity of these zeolites obtained at this temperature was shown to be 15–15.5%, which constituted about 65 % of the equilibrium adsorption capacity values. At the same time, MW dehydration time obtained in our experiments is much higher than that of the corresponding thermal process at the same temperature, being more than 200 min [8]. MW process acceleration can be explained, firstly, by inertialess and volumetric MW heating resulting in a speedy zeolite heating and dehydration, since the temperature and concentration gradients coincide in this case and are directed from the surface into the bulk of zeolite. Thus, the flux of diffusing water molecules towards zeolite surface increases according to the flow Eq. (3): J ¼ Drv rC DT rv rT kp rP ¼ Drv ðrC þ drTÞ kp rP;
(3)
where D 4 DG are concentration diffusion and thermal diffusion coefficients, kp is a molar transfer coefficient, DT/D = d is the relative thermal diffusion coefficient. Secondly, diffusion acceleration can be supposedly attributed to non-thermal MW effects promoting water diffusion from zeolite micropores, which is put forward in the following section.
(2)
where m1–weight of dry zeolite (kg); m2–weight of watersaturated zeolite (kg). The adsorption kinetic curves were obtained in result of this study. 3. Results and discussion 3.1. Effect of temperature on NaA zeolite adsorption capacity The experiments on NaA zeolite dehydration at different temperatures were carried out in 100–230 C temperature range
153
Fig. 3. Kinetics of zeolite dehydration at different temperatures.
154
N.A. Pinchukova et al. / Chemical Engineering and Processing 95 (2015) 151–158
leading to A growth. It should be stressed that according to this hypothesis no alteration of the real molecular diameter or the bond length occurs. Another explanation of diffusion acceleration can be offered, based on the difference in energy distribution under thermal and MW heating. It is known that the energy distribution in the thermally heated system is equiprobable, whereas in case of MW heating one degree of freedom in energy distribution, determined by electric field direction, prevails. If so, the time required for diffusing molecules to attain DE threshold should be much less than that in case of thermal heating. It is quite probable, that both mechanisms can contribute to MW diffusion acceleration simultaneously. Fig. 4. Effect of temperature on the total zeolite weight loss upon MW dehydration.
3.3. Ethanol dewatering
3.2. The role of non-thermal MW effects in diffusion acceleration
Ethanol zeolite-assisted dewatering was performed both in a gaseous and in a liquid phase. Ethanol–water solution with ethanol concentration of 95 wt.% was taken, and absolute alcohol was achieved in both cases. Prior to dewatering NaA zeolite was calcinated in MW field at the temperature of 230 C. The necessary zeolite-ethanol mixture ratio was calculated on the basis of the zeolite dehydration experiments (Section 3.1). In case of gas-phase dewatering the distillation speed was limited by the column height and cross-section, by the bulk density of zeolite; in our experiments it was ca 2.5 mL/min. This speed was shown to be enough for water vapors to be entrapped by zeolite while coming up through the column, since the moisture content in dewatered ethanol did not exceed 0.2%. Thus, it can be inferred that the ethanol dehydration can be efficiently conducted with NaA zeolite under dynamic conditions. In spite of the satisfactory results obtained with the gas-phase dewatering, even tentative energy estimations reveal that ethanol– water evaporation heat will consume most part of the energy required for the whole ethanol dewatering—zeolite dehydration cycle. Therefore, it makes more sense to perform ethanol dewatering in a liquid phase, e.g., as described in [8], thus minimizing the energy consumption at this stage. However, the ultimate choice of dewatering technique should be made on the grounds of production requirements to the process and to the final product and, therefore, should be left to the manufacturer’s discretion. The experiments on liquid-phase ethanol dewatering kinetics study with the use of NaA zeolite were then carried out, aimed at determination of the process time with the outlook for further scale-up. The series of experiments was carried out as described in Section 2.3.2, wherein the effect of ethanol–water solution – zeolite contact time on the amount of adsorbed water was investigated. The contact time varied from 5 to 120 min.
Diffusion of water molecules from zeolite micropores can be described by diffusion equation similar to Arrhenius’ Eq. (4) [24]: DE (4) ; D ¼ A exp RT where DE is the energy necessary for the diffusing particle to come out from the the potential well and to make a leap from one vacancy to another, and pre-exponential factor A in this case can be interpreted as a free diffusion coefficient [24]. It is unlikely that MW field can affect the activation energy, as the latter is the value depending on the substances nature and is unique for each chemical or physical process. Thus, it can be supposed that the diffusion can be enhanced in MW field due to alteration of the pre-exponential factor A. The following explanation can be proposed. It is known from the Arrhenius collision theory that A designates the average number of collisions per unit of time per moles of reactants (5): sffiffiffiffiffiffiffiffiffiffiffi 8kB T (5) A ¼ NA s
pm
where NA is Avogadro’s constant, s is a cross-section having the dimension of an area and depending on the average molecular diameter d,kB is Boltzmann constant, T is the absolute temperature, m* is the reduced mass of the reacting molecules. For thermal heating processes d is considered invariable, so, according to Eq. (5) A depends only on the temperature, as its increase results in chaotic Brownian movement and molecules multiple collisions. Unlike to the thermal heating, microwave field induces directed motion of the dipole molecules aligned with electric field lines, resulting in the increase of the effective molecular diameter (Fig. 5), and, consequently, in the increase of s
Fig. 5. Increase of molecular effective diameter in result of directed movement under the influence of MW irradiation.
Fig. 6. Adsorption kinetics of NaA zeolite in the liquid phase studies.
N.A. Pinchukova et al. / Chemical Engineering and Processing 95 (2015) 151–158
On the basis of the experimental results the adsorption kinetics curve was obtained (Fig. 6) which showed that the saturation of zeolite was attained approx. after 30 min. contact of zeolite with ethanol–water solution, yielding the adsorption capacity of NaA zeolite under the studied conditions, calculated from Eq. (1), at the level of 18–18.5%. At the same time as many as 85% of all the water were adsorbed within the first 5 min., suggesting that adsorption speed was high enough to perform the experiments on liquidphase ethanol dewatering under dynamic conditions, by passing a feeding solution through a fixed bed of zeolite. The optimal linear flow velocity can be calculated on the basis of these kinetic data by differentiating the integral kinetic curve given in Fig. 6. 3.4. Regeneration of NaA zeolite and effect of the number of cycles on the adsorption capacity Ethanol dewatering and zeolite regeneration were performed according to the procedure described in Section 3.1. Contact time of the ethanol–water mixture with NaA zeolite at the level of 30 min. was set. The adsorption capacity after zeolite dehydration was calculated from Eq. (1). To study the effect of number of cycles on the zeolite adsorption capacity ethanol dewatering-zeolite dehydration procedure was repeated 20 times. The results showed that the average adsorption capacity after 20 cycles maintained at the same level (18 0.7 %), which indicated that the selected MW dehydration mode, as it was supposed, did not affect NaA zeolite structure and properties. This conclusion was supported by powder X-ray diffraction (XRD) measurements of the starting NaA zeolite and the one subjected to MW treatment over 20 times. Phase analysis performed utilizing the software supplied with the diffractometer («Eva» 4 «Search» programs) showed that both treated and
155
untreated samples of NaA zeolite, apart from the zeolite phase, contain albite (NaAlSi3O8), moganite (SiO2), nepheline (NaAlSiO4), Al(OH)3 and other more complex aluminium silicates having water molecules or hydroxyl groups in their structure. Comparison of the XRD patterns before and after MW treatment shows no significant difference (Fig. 7). Basic lines of zeolite phase are not shifted in the MW treated sample as compared to the initial one. Phase indexing carried out with the aid of PowderX program [25] shows that the initial zeolite has a cubic structure with a parameter being 12.2776(16) Å which increases insignificantly after MW treatment up to 12.2908(14) Å. The obtained results suggest that the structure of the NaA zeolite was not affected by MW irradiation under the given experimental conditions. It should be noted that separation of ethanol dewatering and zeolite regeneration steps in the frame of this study was constrained by the laboratory equipment limitations. However, for scale-up purposes “in-situ MW regeneration concept” should be developed, and a tailored approach should be applied for design of a pilot-plant scale MW equipment. 3.5. Energy estimations and process scale-up The total energy consumed for zeolite dehydration is comprised of the energy required for heating zeolite to the set temperature and water desorption heat, and can be calculated from Eq. (6): Q dehydration ¼ mZ C Z DT þ nH2 O DHdesorption ;
(6)
where mZ is the zeolite weight (kg), CZ is zeolite heat capacity (kJ/ kg/ C), DHdesorption is water desorption heat (kJ/mol), nH2 O is amount of adsorbed water (mol). Desorption heat for NaA zeolite varies from 20 to 70 kJ depending on the the zeolite loading [26]. In our calculations
Fig. 7. Powder X-ray diffraction patterns of NaA zeolite before (upper) and after (lower) MW treatment subjected to 20 cycles of MW dehydration.
156
N.A. Pinchukova et al. / Chemical Engineering and Processing 95 (2015) 151–158
Table 1 Specific energy consumption required for thermal and MW dehydration of NaA zeolite and for dewatered ethanol preparation Final product
Specific energy consumption, MJ/kg MW zeolite dehydration
Dry zeolite Dewatered ethanol
Literature sources
Calculated
Experimental (Lab-scale)
Predicted (Plant scale)
0.440 0.145
1.94 0.550
0.600 0.17
0.92 [8] 0.83 3]
the value of DHdesorption was assumed to be 30 kJ/mol, and the calculated value of Qdehydration made up 440 kJ/kg of dry zeolite. Experimental energy value Qexp (J) for zeolite dehydration process was determined from the following Eq. (7): Q exp ¼ P t;
(7)
where P is the average power of MW oscillator (W), t is the process time (s), and under given experimental conditions (temperature 230 C, process time 20 min) made up 1.94 MJ/kg of dry zeolite, or 0.55 MJ/kg of dewatered ethanol. Thus, the energy efficiency factor calculated as the ratio Qdehydration/Qexp is 0.23, which is in a good accordance with our previous results obtained in the investigations of MW-assisted drug substances drying and phosphoric acid dehydration [27,28], wherein the same MW installation was used. Thus, it may be concluded that high factors of MW irradiation utilization can hardly be achieved at a laboratory scale. However, in the same works it was shown that energy
efficiency could be increased up to 70–75% due to MW cavity occupancy optimization. So, the results obtained in this laboratory study should be adjusted for the value of MW cavity occupancy ratio, which in our experiments was about 0.01. At this, taking into consideration potential increase of the energy utilization factor, energy consumption for a plant-scale zeolite dehydration can be predicted at the level of 0.6 MJ/kg of dry zeolite or, respectively, 0.17 MJ/kg of dewatered ethanol, provided that the feeding ethanol concentration is about 95%.The values of the energy consumed for zeolite dehydration with the use of different techniques are demonstrated in Table 1. In our results only the energy used for zeolite dehydration was accounted for, as it made the most part of the total energy expenditures related to dewatered ethanol preparation by the proposed method. Table 1 well illustrates the energy benefits from the use of MW irradiation for zeolite dehydration, especially at a plant scale. Such a drastic energy reduction is the result of specific MW heating, whereby the energy is utilized directly by the zeolite, and heat losses are minimized. Moreover, non-thermal MW effects discussed in Section 3.2 also play a great role in total reduction of energy consumption. In this respect there is no alternative to MW heating, since any other heating method would imply surface heating, wherein thermal conductivity of the processed material is crucial. Taking into account low thermal conductivity of zeolites, which are known to be good thermal insulators, it is obvious that considerable heat losses are inevitable. To realize the concept of “in-situ MW zeolite regeneration” the tailored approach should be applied for design of MW equipment. The flow chart of the plantscale setup is shown in Fig. 8. The 3-module setup is proposed to provide continuous ethanol dewatering process, since the whole cycle of dewatered ethanol
Fig. 8. Flow chart of the pilot-plant scale 3-module setup for ethanol dewatering with further in-situ MW zeolite regeneration: 1, 2, 3–MW oscillators; 4, 5, 6–MW cavities with zeolite-filled columns inside. Bold guide lines indicate ethanol dewatering scheme, thinner solid lines indicate zeolite predrying scheme, and dashed lines indicate MW zeolite dehydration scheme.
N.A. Pinchukova et al. / Chemical Engineering and Processing 95 (2015) 151–158
preparation, including zeolite regeneration, is comprised of three constituents (8):
t ¼ t ethanol þ t zeolite þ t cooling ;
(8)
where t ethanol is time required for ethanol dewatering, t zeolite is time required for zeolite dehydration, t cooling is time required for zeolite cooling. According to our experimental results, t ethanol is 30 min, t zeolite is 20 min, and t cooling was approx. 90 min. However, the last step’s duration can be markedly reduced due to optimization of a cooling mode, e.g., compressed gas cooling can be applied. Thus, in the first approximation durations of these three steps can be considered equal or commensurable. If so, the operation mode of MW plantscale setup would be as follows: one of the three modules would operate in the ethanol dewatering mode, at the same time another one, having been previously used for ethanol dewatering, would undergo zeolite dehydration, and the third module, having passed the two above steps, would be in a cooling regime. Thus, 3-module apparatus should provide continuous process of ethanol dewatering and zeolite regeneration. It should be noted that the concept of the module design has been borrowed from our previous investigations on MW-assisted drying of alkali halides [28–30]. In result of this study the tailored MW drying equipment was designed [29], comprising an ampoule loaded with the salt to be dried, and placed in alignment into MW cylindrical cavity. To ensure homogeneous temperature of the bulk product, ampoule rotation was used provided by vacuum-sealed rotating junction. In case of zeolite regeneration, MW setup can be simplified and zeolite-loaded column rotation can be avoided by virtue of applying the improved MW mode pattern [30] ensuring more homogeneous, as compared to conventional MW systems, electromagnetic field distribution throughout MW cavity, and consequently, homogeneous temperature distribution throughout the bulk of the material. The penetration depth for NaA zeolite estimated on the basis of the dielectric properties measurements as described in Section 2.2.4, changes from 3.7 cm for the saturated zeolite to 10.5 cm for the dry one. These values allow effective scale-up of the process, taking into account that the zeolite loaded into a cylindrical column is irradiated by microwaves all round, so the column diameter will be equal to a doubled penetration depth. Moreover, a special custom-made MW inlet was designed in order to increase the efficiency of MW irradiation utilization up to 85% and to decrease a specific metal content of the MW setup [31]. 4. Conclusions The study revealed that regeneration of NaA zeolite used for ethanol dewatering is more efficient with application of microwave irradiation instead of conventional heating in terms of process time shortening and energy cuts. Powder X-ray diffraction measurements showed that even after 20 microwave regeneration cycles the zeolite structure remains the same as in the initial sample. The decrease in energy consumption by the factor of 1.7 as compared to the thermal zeolite dehydration has been obtained at laboratory scale, with 10-fold shortening of MW process time. The reasons of process intensification resulting in considerable energy saving were analyzed. Firstly, it was shown that drastic process acceleration under MW conditions was the result of minimized heating time and energy losses due to the incident properties of microwave heating. Secondly, the hypothesis about non-thermal MW effects resulting from the specific interactions of MW irradiation with matter, contributing to the process acceleration and the model illustrating the influence of MW irradiation on the
157
molecular collision cross-section, based on Arrhenius collision theory, were proposed. The outlook for the process scale up has been considered, and a pilot-plant scale model for continuous ethanol dewatering by NaA zeolite comprising MW-assisted zeolite regeneration stage has been proposed. Taking into account the potential for improving energy efficiency upon scale-up, the energy consumption at the level of 0.17 MJ/kg of dewatered ethanol has been predicted for a MW large scale zeolite dehydration process, which constitutes only 1/5 of the energy consumed in the corresponding thermal process. References [1] M. Boldi^s, K. Melzoch, J. Pokorny, M. Ko^ cirík, Dynamics of sorption columns in dewatering of bioethanol using zeolites, Stud. Surf. Sci. Catal. 142 (2002) 1663– 1670. [2] Zeolite dewatering membranes to be commercialised for biofuels, Filtr. Separat. 45 (2008) 9, www.hyflux.com. [3] E. Nagy, S. Boldyryev, Energy demand of biofuel production applying distillation and/or pervaporation, Chem. Eng. Trans. 35 (2013) 265–270. [4] J.F.G. Oliveira, I.L. Lucena, R.M.A. Saboya, M.L. Rodrigues, A.E.B. Torres, F.A.N. Fernandes, C.L. Cavalcante Jr., E.J.S. Parente Jr., Biodiesel production from waste coconut oil by esterification with ethanol: the effect of water removal by adsorption, Renew. Energ. 35 (2010) 2581–2584. [5] S.K. Wahono, A. Hernawan Kristiani, S. Tursiloadi, H. Abimanyu, Characterization and utilization of gunungkidul natural zeolite for bioethanol dehydration, Energ. Proced. 47 (2014) 263–267. [6] A.A. Lukanov, I.I. Kondratiev, M.Y. Levinter, Zeolite adsorbent, SU Patent 1558, 870, 1990. [7] B.V. Krasiy, G.N. Khomenko, Y.A. Shavandin, Method for preparation of granulated zeolite, SU Patent 1655, 901, 1991. [8] E. Gabrus’, J. Nastaj, P. Tabero, T. Aleksandrzak, Experimental studies on 3A and 4A zeolite molecular sieves regeneration in TSA process: Aliphatic alcohols dewatering–water desorption, Chem. Eng. J. 259 (2015) 232–242. [9] J.M.R. Bélanger, J.R.J. Paré, O. Poon, C. Fairbridge, S. Ng, S. Mutyala, R. Hawkins, Remarks on various applications of microwave energy, J. Microwave Power EE 42 (2008) 24–44. [10] P.M. Bendale, C.-M. Sun, Rapid microwave-assisted liquid-phase combinatorial synthesis of 2-(arylamino) benzimidazoles, J. Comb. Chem. 4 (2002) 359–361. [11] F. Bergamelli, M. Iannelli, J.A. Marafie, J.D. Moseley, A commercial continuous flow microwave reactor evaluated for scale-up, Org. Process Res. Dev. 14 (2010) 926–930. [12] S. Caddick, R. Fitzmaurice, Microwave enhanced synthesis, Tetrahedron 65 (2009) 3325–3355. [13] C.O. Kappe, D. Dallinger, S.S. Murphree, Practical Microwave Synthesis For Organic Chemists, WILEY-VCH, Weinheim, 2009. [14] A. Loupy, Microwaves in Organic Synthesis, 2nd ed., Wiley-VCH, Weinheim, 2006. [15] D. Dallinger, C.O. Kappe, Microwave-assisted synthesis in water as solvent, Chem. Rev. (2007) 2563–2591. [16] B.A. Roberts, C.R. Strauss, Toward rapid, green, predictable microwave-assisted synthesis, Acc. Chem. Res. 38 (2005) 653–661. [17] I. Polaert, L. Estel, R. Huyghe, M. Thomas, Adsorbents regeneration under microwave irradiation for dehydration and volatile organic compounds gas treatment, Chem. Eng. J. 162 (2010) 941. [18] R. Han, Y. Wang, Q. Sun, L. Wang, J. Song, X. He, C. Dou, Malachite green adsorption onto natural zeolite and reuse by microwave irradiation, J. Hazard. Mater. 175 (2010) 1056. ski, M. Komorowska-Durka, G.D. Stefanidis, A.I. Stankiewicz, [19] R. Cherban Microwave swing regeneration vs. temperature swing regeneration— comparison of desorption kinetics, Ind. Eng. Chem. 50 (2011) 8632–8644. [20] M. Komorowska-Durka, R. Van Houten, G.D. Stefanidis, Application of microwave heating to pervaporation: a case study for separation of ethanol–water mixtures, Chem. Eng. Process. 81 (2014) 35–40. [21] Z. Ji, J. Wang, D. Hou, Z. Yin, Z. Luan, Effect of microwave irradiation on vacuum membrane distillation, J. Membrane Sci. 429 (2013) 473–479. [22] P. Bruttel, R. Schlink, Water determination by Karl Fischer Titration, Metrohm Ltd., Herisau, 2006. [23] A.Y. Voloshko, G.D. Kramskoy, Y.D. Kramskoy, V.L. Samoilov, V.P. Seminozhenko, A.A. Chepkiy, O.V. Shishkin, Method for measuring dielectric constant and loss tangent of the material, UA Patent 34, 146, 2008. [24] C. Antonio, R.T. Deam, Can “microwave effects” be explained by enhanced diffusion? Phys. Chem. Chem. Phys. 9 (2007) 2976–2982. [25] J. Rodriguez-Carvajal, T. Roisnel, FullProf.98 and WinPLOTR: New Windows 95/ NT Applications for Diffraction. Commission for Powder Diffraction, International Union of crystallography, Newsletter No. 20 (May–August) Summer 1998. [26] K.F. Loughlin, Water isoterm models for 4A (NaA) zeolite, Adsorption 15 (2009) 337–353.
158
N.A. Pinchukova et al. / Chemical Engineering and Processing 95 (2015) 151–158
[27] N.A. Pinchukova, A.Y. Voloshko, O.V. Shyshkin, V.A. Chebanov, B.H.P. Van de Kruijs, J.C.L. Arts, M.H.C.L. Dressen, J. Meuldijk, J.A.J.M. Vekemans, L.A. Hulshof, A facile microwave-mediated drying process of thermally unstable/labile products, Org. Process Res. Dev. 14 (2010) 1130–1139. [28] 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, Chem. Eng. Process. 50 (2011) 1193–1197.
[29] V.D. Bal'nikh, A.Y. Voloshko, B.V. Grinyov, Y.D. Kramskoy, A.I. Ovchinnikov, N.N. Smirnov, O.V. Shishkin, Method and device for drying free-flowing dielectric materials, UA Patent 56 629, 2005. [30] A.Y. Voloshko, Y.D. Kramskoy, V.L. Samoilov, V.P. Seminozhenko, A.A. Chepkiy, O.V. Shishkin, Method for dielectric materials heating, UA Patent 28 893, 2007. [31] A.Y. Voloshko, V.B. Moiseev, N.A. Pinchukova, V.L. Samoilov, V.P. Seminozhenko, I.B. Stel'mah, O.V. Shishkin, Device for drying free-flowing dielectric materials, UA Patent 84354, 2008.
Contents lists available at ScienceDirect
Chemical Engineering and Processing: Process Intensification journal homepage: www.elsevier.com/locate/cep
The use of microwave irradiation for zeolite regeneration in a continuous ethanol dewatering process Natalia A. Pinchukovaa,* , Alexander Yu. Voloshkoa , Vyacheslav N. Baumera , Oleg V. Shishkina , Valentin A. Chebanova,b a b
SSI "Institute for Single Crystals" of National Academy of Sciences of Ukraine, Lenin Ave., 60, Kharkiv 61001, Ukraine V.N. Karazin Kharkiv National University, Svobody sq., 4, Kharkiv 61022, Ukraine
A R T I C L E I N F O
A B S T R A C T
Article history: Received 15 January 2015 Received in revised form 25 May 2015 Accepted 2 June 2015 Available online 5 June 2015
The processes of zeolite-assisted ethanol dewatering with further microwave (MW) zeolite dehydration has been studied. Both gas-phase and liquid-phase alcohol dewatering methods yielded absolute ethanol. Kinetic study showed that these processes can be performed under dynamic conditions, which is relevant from a scale-up viewpoint. The concept of “in situ MW-assisted zeolite regeneration” was elaborated. Reduction of energy consumption by 1.7 times with 10-fold process time shortening has been shown for lab-scale MW zeolite dehydration, as compared to the corresponding thermal method. Hypothesis explaining the reasons of such a dramatic energy and time saving was proposed based on the diffusion theory. The influence of MW irradiation on the zeolite structure was studied by powder X-ray diffractometry. The results showed that even after 20 regeneration cycles the zeolite remained the same as the initial sample. The outlook for process scale-up has been discussed. A pilot-plant scale model for continuous ethanol dewatering by NaA zeolite with the following zeolite regeneration has been proposed. ã2015 Elsevier B.V. All rights reserved.
Keywords: Microwave irradiation Ethanol dewatering Zeolite regeneration Energy consumption Continuous process
1. Introduction In the recent years zeolite molecular sieves have been widely used for dehydration or dewatering purposes due to their unique properties to selectively adsorb water from water containing solutions, which is especially topical in a view of growing interest to bioethanol production [1–5]. For ethanol dewatering purposes NaA or KA zeolites are most appropriate, being synthetic zeolites, since they have small aperture diameters (4 and 3 Å respectively), therefore organic molecules having two or more C atoms cannot penetrate inside the zeolite pores and remain in a liquid phase, while smaller molecules like water ones are effectively absorbed and retained by zeolites. In spite of a constantly growing interest to zeolites, most of the papers are dedicated to the study of adsorption properties of zeolites; meanwhile, there is much less information in the literature concerning zeolite regeneration. This gap, undoubtedly, should be supplied, since zeolite regeneration step is crucial from commercial viewpoint, and only efficient dehydration of water-saturated zeolites can ensure their durability and successful application at industrial scale.
* Corresponding author. Fax: +38 57 340 93 43. E-mail address: [email protected] (N.A. Pinchukova). http://dx.doi.org/10.1016/j.cep.2015.06.001 0255-2701/ ã 2015 Elsevier B.V. All rights reserved.
The conventional technique for zeolite dehydration is hightemperature calcination [6,7], which is time and energy consuming because of low thermal conductivity of the zeolite materials. On the other hand, the temperature limitation exists, as high temperatures may cause partial or complete damage to zeolite matrix resulting in decrease in adsorption capacity and in zeolite life-time shortening [8]. For efficient zeolite dehydration and reduction of energy consumption, as well as for minimization of the stress on the zeolite structure, process parameters should be carefully selected. Gabrus et al. proposed in-situ thermal regeneration of 3A and 4A zeolites in the fixed bed after liquid-phase dewatering of aliphatic alcohols [8]. Thermal stability of zeolites, water adsorption capacity, adsorption selectivity and thermal desorption efficiency were carefully studied. The optimal conditions of zeolite dehydration were determined which allowed efficient dehydration of the studied zeolites with maintaining their adsorption capacity approximately at the same level over 30 hydration-dehydration cycles. The proposed technique allows continuous alcohol dewatering with zeolites. Long life-time of zeolites is ensured by properly selected regeneration temperature mode. However, it seems reasonable to improve this “in-situ” approach in terms of process acceleration and energy costs reduction by replacing traditional
152
N.A. Pinchukova et al. / Chemical Engineering and Processing 95 (2015) 151–158
heating sources for a novel and more efficient heating method, such as microwave (MW) irradiation, which shows excellent results when applied to chemical or physical processes [9–13]. It is known that ethanol has a good absorbing capacity towards MW irradiation [14] while water has middle one, nevertheless, MW irradiation is widely used for processing materials or reaction mixtures where water is a medium or one of the components [15,16]. Therefore, MW irradiation has proved to be a promising technique for various adsorbents regeneration [17–19] and intensification of separation processes like pervaporation or membrane distillation [20,21]. In view of the above said, there are all the grounds to predict efficient application of MW irradiation to NaA zeolite regeneration resulting in energy cuts and process intensification. In this study the process of liquid phase ethanol dehydration and further zeolite regeneration with the use of MW irradiation were studied with a fixed bed of NaA zeolite. The aim of this study is the development of a continuous process of dewatered ethanol preparation comprising ethanol dewatering and MW-assisted zeolite dehydration; and determination of optimal process parameters ensuring minimization of energy consumption and process duration as compared to corresponding thermal processes. The effect of MW irradiation on zeolite adsorption capacity after multiple uses was also examined. Process scale-up is considered in this study as well. 2. Materials and methods 2.1. Materials NaA zeolite was purchased from Salavat catalyst plant Ltd. (Russian Federation). Ethanol–water solution with ethanol concentration 95 wt.% and NaA zeolite in the form of extruded cylinders 5–7 mm in length and 1.6 0.2 mm in diameter were used in ethanol dewatering and zeolite regeneration experiments. 2.2. Analytical control 2.2.1. Water content determination Water content in ethanol before and after dehydration step was determined by Karl Fisher method [22]. 2.2.2. Zeolite dehydration degree determination Concentration of the water adsorbed by zeolite and zeolite dehydration degree were determined by the gravimetric method. The saturated zeolite was subjected to calcination at 250 C, and the amount of adsorbed water was calculated by the weight difference before and after calcination. The weight loss was calculated from the Eq. (1): W¼
m1 m2 100%; m2
developed by us [23] according to the following procedure. The quarts cell (r = 0.845 cm, l = 10,4 cm) was placed into a cylindrical MW cavity (r = 4.7 cm, l = 10.4 cm) along the cavity axisppopopl. The resonance frequencies and cavity Q factor were measured with and without the tested samples. On the basis of the measurements results the dielectric constant e0 , loss tangent tg d and skin-layer depth were calculated. 2.3. Experimental set-up and procedure 2.3.1. MW set-up MW zeolite dehydration was performed in a multimode MW laboratory system (MARS, CEM Corp., USA) equipped with IR sensor for temperature measurements. A typical distillation set-up was used for zeolite dehydration experiments (Fig. 1), consisting of a 250-mL round-bottom flask containing saturated zeolite and placed inside the microwave cavity, an upward glass tube (l = 30 cm, Ø = 2.5 cm), passing through the aperture in the upper wall of the MW cavity. A standard distillation kit for collecting water condensate, attached to the column, was placed outside the cavity. The process was conducted under slight evacuation provided by membrane vacuum pump for better removal of water vapor. In the experiments on gas-phase ethanol dewatering, the typical distillation set-up was used, similar to that applied for zeolite dehydration (Fig. 1), with some modifications; viz. the zeolite-loaded column was used instead of a hollow tube, and a fiber-optic (FO) probe was applied for temperature measurements (Fig. 2). In this case ethanol–water mixture and zeolite-loaded column were exposed to MW field, while the distillation kit was placed outside of MW cavity. 2.3.2. Ethanol dewatering and NaA zeolite dehydration Ethanol dewatering was performed by applying both gas-phase and liquid-phase approaches. According to the first method the ethanol–water mixture was distilled, with the vapors passing through a column filled with previously dehydrated NaA zeolite (Fig. 2). The following procedure was applied. Ethanol–water mixture was heated by MW irradiation to the boiling point, and the vapors passed through the column with the fixed bed of NaA zeolite. The necessary ratio of zeolite to ethanol–water mixture was calculated on the basis of NaA adsorption capacity determined previously (see Section 3.1), with the 20% surplus of zeolite. Due to high selectivity of NaA zeolite to water, larger molecules of ethanol were not retained and freely passed through the column, condensed and finally were collected in the receiver.
(1)
where m1 is the weight of saturated zeolite (kg); m2 is the weight of calcinated zeolite (kg). 2.2.3. Powder X-ray diffraction Powder X-ray diffraction patterns of the initial zeolite sample and MW irradiated one were recorded with the powder X-ray diffractometer Siemens D500 using Cu irradiation (l = 1.54184 Å, graphitic monochromator), angular range 3–70 2u, step 0.02 . 2.2.4. Dielectric properties measurements Dielectric properties of the saturated and dehydrated zeolite were measured with the use of the apparatus and methodology
Fig. 1. Schematic image of experimental setup used for MW zeolite dehydration.
N.A. Pinchukova et al. / Chemical Engineering and Processing 95 (2015) 151–158
Fig. 2. Schematic image of experimental setup exploited for MW ethanol gas-phase dewatering with the use of NaA zeolite.
Liquid-phase dewatering was carried out by soaking dehydrated zeolite in water–ethanol mixture. For this purpose weighed amount of zeolite was soaked with weighed amount of ethanol– water mixture and allowed to stay for a definite time. Saturated zeolite obtained after ethanol dewatering was separated from ethanol by decanting of the latter, weighed and dried in a draught until the weight constancy was achieved and no smell of alcohol was detected, indicating that all the surface moisture, mostly consisting of ethanol, had volatilized. After such predrying zeolite containing only adsorbed water in its micropores was calcinated in MW field under slight evacuation with the use of MARS MW system. Temperature control was performed by means of IR-sensor mounted into the bottom of MW cavity, for which purpose the flask was placed on the hollow teflon rest right above the IR-sensor. Temperature MW power profiles were continuously displayed on the screen of PC integrated into MARS system. The temperature distribution in the bulk of NaA zeolite was periodically checked by interrupting the dehydration process and measuring temperature in different points of bulk zeolite by electronic thermometer. The measurements taken by this way were close to the temperature readings yielded by the integrated IR-sensor. The amount of water removed was determined from the zeolite mass difference before and after dehydration. 2.3.3. Adsorption kinetics study Adsorption kinetics was studied by keeping dry zeolite in ethanol–water mixture for a definite time, separating zeolite from the liquid, predrying zeolite in a draught and calcinating it in MW field, as described in Section 2.3.2. Adsorption capacity q was determined from the Eq. (2): q¼
m2 m1 100%; m2
(the maximal applied temperature of 230 C is explained by the IRsensor measuring limit). Prior to each run the zeolite was hydrated by keeping in a dessicator over the water layer until the constant weight was attained. The saturated zeolite containing only adsorbed water was then dehydrated by calcination in microwave field, as described in Section 2.3.2. Kinetics studies revealed that the optimal time required for zeolite dehydration was 20 min., since within this time the zeolite reached constant weight irrespective of the temperature applied (Fig. 3). The curves shapes indicate at the molecular dehydration mode, corresponding to reversible processes. This is the ground to suggest that the multiple hydration–dehydration cycles with the use of NaA zeolite should not affect its structure or properties providing durability of the zeolite under production conditions. Fig. 4 illustrates the dependence of zeolite dehydration degree on the temperature. Though it appears to be linear in the whole studied temperature range and does not reach the saturation limit, the weight loss of 15.5% attained at 230 C indicates that the sorption capacity of the zeolite is sufficient for effective alcohol dehydration. The temperature distribution in a bulk of zeolite, controlled by the means of electronic thermometer, was ranging within 2.5 C of the average value, which was close to temperature readings yielded by the MARS-integrated IR sensor within the measurement error limits. The results obtained are in a good accordance with the literature data for thermal 3A and 4A zeolite dehydration, offering the optimal dehydration temperature at the level of 240 C [8]. The dynamic adsorption capacity of these zeolites obtained at this temperature was shown to be 15–15.5%, which constituted about 65 % of the equilibrium adsorption capacity values. At the same time, MW dehydration time obtained in our experiments is much higher than that of the corresponding thermal process at the same temperature, being more than 200 min [8]. MW process acceleration can be explained, firstly, by inertialess and volumetric MW heating resulting in a speedy zeolite heating and dehydration, since the temperature and concentration gradients coincide in this case and are directed from the surface into the bulk of zeolite. Thus, the flux of diffusing water molecules towards zeolite surface increases according to the flow Eq. (3): J ¼ Drv rC DT rv rT kp rP ¼ Drv ðrC þ drTÞ kp rP;
(3)
where D 4 DG are concentration diffusion and thermal diffusion coefficients, kp is a molar transfer coefficient, DT/D = d is the relative thermal diffusion coefficient. Secondly, diffusion acceleration can be supposedly attributed to non-thermal MW effects promoting water diffusion from zeolite micropores, which is put forward in the following section.
(2)
where m1–weight of dry zeolite (kg); m2–weight of watersaturated zeolite (kg). The adsorption kinetic curves were obtained in result of this study. 3. Results and discussion 3.1. Effect of temperature on NaA zeolite adsorption capacity The experiments on NaA zeolite dehydration at different temperatures were carried out in 100–230 C temperature range
153
Fig. 3. Kinetics of zeolite dehydration at different temperatures.
154
N.A. Pinchukova et al. / Chemical Engineering and Processing 95 (2015) 151–158
leading to A growth. It should be stressed that according to this hypothesis no alteration of the real molecular diameter or the bond length occurs. Another explanation of diffusion acceleration can be offered, based on the difference in energy distribution under thermal and MW heating. It is known that the energy distribution in the thermally heated system is equiprobable, whereas in case of MW heating one degree of freedom in energy distribution, determined by electric field direction, prevails. If so, the time required for diffusing molecules to attain DE threshold should be much less than that in case of thermal heating. It is quite probable, that both mechanisms can contribute to MW diffusion acceleration simultaneously. Fig. 4. Effect of temperature on the total zeolite weight loss upon MW dehydration.
3.3. Ethanol dewatering
3.2. The role of non-thermal MW effects in diffusion acceleration
Ethanol zeolite-assisted dewatering was performed both in a gaseous and in a liquid phase. Ethanol–water solution with ethanol concentration of 95 wt.% was taken, and absolute alcohol was achieved in both cases. Prior to dewatering NaA zeolite was calcinated in MW field at the temperature of 230 C. The necessary zeolite-ethanol mixture ratio was calculated on the basis of the zeolite dehydration experiments (Section 3.1). In case of gas-phase dewatering the distillation speed was limited by the column height and cross-section, by the bulk density of zeolite; in our experiments it was ca 2.5 mL/min. This speed was shown to be enough for water vapors to be entrapped by zeolite while coming up through the column, since the moisture content in dewatered ethanol did not exceed 0.2%. Thus, it can be inferred that the ethanol dehydration can be efficiently conducted with NaA zeolite under dynamic conditions. In spite of the satisfactory results obtained with the gas-phase dewatering, even tentative energy estimations reveal that ethanol– water evaporation heat will consume most part of the energy required for the whole ethanol dewatering—zeolite dehydration cycle. Therefore, it makes more sense to perform ethanol dewatering in a liquid phase, e.g., as described in [8], thus minimizing the energy consumption at this stage. However, the ultimate choice of dewatering technique should be made on the grounds of production requirements to the process and to the final product and, therefore, should be left to the manufacturer’s discretion. The experiments on liquid-phase ethanol dewatering kinetics study with the use of NaA zeolite were then carried out, aimed at determination of the process time with the outlook for further scale-up. The series of experiments was carried out as described in Section 2.3.2, wherein the effect of ethanol–water solution – zeolite contact time on the amount of adsorbed water was investigated. The contact time varied from 5 to 120 min.
Diffusion of water molecules from zeolite micropores can be described by diffusion equation similar to Arrhenius’ Eq. (4) [24]: DE (4) ; D ¼ A exp RT where DE is the energy necessary for the diffusing particle to come out from the the potential well and to make a leap from one vacancy to another, and pre-exponential factor A in this case can be interpreted as a free diffusion coefficient [24]. It is unlikely that MW field can affect the activation energy, as the latter is the value depending on the substances nature and is unique for each chemical or physical process. Thus, it can be supposed that the diffusion can be enhanced in MW field due to alteration of the pre-exponential factor A. The following explanation can be proposed. It is known from the Arrhenius collision theory that A designates the average number of collisions per unit of time per moles of reactants (5): sffiffiffiffiffiffiffiffiffiffiffi 8kB T (5) A ¼ NA s
pm
where NA is Avogadro’s constant, s is a cross-section having the dimension of an area and depending on the average molecular diameter d,kB is Boltzmann constant, T is the absolute temperature, m* is the reduced mass of the reacting molecules. For thermal heating processes d is considered invariable, so, according to Eq. (5) A depends only on the temperature, as its increase results in chaotic Brownian movement and molecules multiple collisions. Unlike to the thermal heating, microwave field induces directed motion of the dipole molecules aligned with electric field lines, resulting in the increase of the effective molecular diameter (Fig. 5), and, consequently, in the increase of s
Fig. 5. Increase of molecular effective diameter in result of directed movement under the influence of MW irradiation.
Fig. 6. Adsorption kinetics of NaA zeolite in the liquid phase studies.
N.A. Pinchukova et al. / Chemical Engineering and Processing 95 (2015) 151–158
On the basis of the experimental results the adsorption kinetics curve was obtained (Fig. 6) which showed that the saturation of zeolite was attained approx. after 30 min. contact of zeolite with ethanol–water solution, yielding the adsorption capacity of NaA zeolite under the studied conditions, calculated from Eq. (1), at the level of 18–18.5%. At the same time as many as 85% of all the water were adsorbed within the first 5 min., suggesting that adsorption speed was high enough to perform the experiments on liquidphase ethanol dewatering under dynamic conditions, by passing a feeding solution through a fixed bed of zeolite. The optimal linear flow velocity can be calculated on the basis of these kinetic data by differentiating the integral kinetic curve given in Fig. 6. 3.4. Regeneration of NaA zeolite and effect of the number of cycles on the adsorption capacity Ethanol dewatering and zeolite regeneration were performed according to the procedure described in Section 3.1. Contact time of the ethanol–water mixture with NaA zeolite at the level of 30 min. was set. The adsorption capacity after zeolite dehydration was calculated from Eq. (1). To study the effect of number of cycles on the zeolite adsorption capacity ethanol dewatering-zeolite dehydration procedure was repeated 20 times. The results showed that the average adsorption capacity after 20 cycles maintained at the same level (18 0.7 %), which indicated that the selected MW dehydration mode, as it was supposed, did not affect NaA zeolite structure and properties. This conclusion was supported by powder X-ray diffraction (XRD) measurements of the starting NaA zeolite and the one subjected to MW treatment over 20 times. Phase analysis performed utilizing the software supplied with the diffractometer («Eva» 4 «Search» programs) showed that both treated and
155
untreated samples of NaA zeolite, apart from the zeolite phase, contain albite (NaAlSi3O8), moganite (SiO2), nepheline (NaAlSiO4), Al(OH)3 and other more complex aluminium silicates having water molecules or hydroxyl groups in their structure. Comparison of the XRD patterns before and after MW treatment shows no significant difference (Fig. 7). Basic lines of zeolite phase are not shifted in the MW treated sample as compared to the initial one. Phase indexing carried out with the aid of PowderX program [25] shows that the initial zeolite has a cubic structure with a parameter being 12.2776(16) Å which increases insignificantly after MW treatment up to 12.2908(14) Å. The obtained results suggest that the structure of the NaA zeolite was not affected by MW irradiation under the given experimental conditions. It should be noted that separation of ethanol dewatering and zeolite regeneration steps in the frame of this study was constrained by the laboratory equipment limitations. However, for scale-up purposes “in-situ MW regeneration concept” should be developed, and a tailored approach should be applied for design of a pilot-plant scale MW equipment. 3.5. Energy estimations and process scale-up The total energy consumed for zeolite dehydration is comprised of the energy required for heating zeolite to the set temperature and water desorption heat, and can be calculated from Eq. (6): Q dehydration ¼ mZ C Z DT þ nH2 O DHdesorption ;
(6)
where mZ is the zeolite weight (kg), CZ is zeolite heat capacity (kJ/ kg/ C), DHdesorption is water desorption heat (kJ/mol), nH2 O is amount of adsorbed water (mol). Desorption heat for NaA zeolite varies from 20 to 70 kJ depending on the the zeolite loading [26]. In our calculations
Fig. 7. Powder X-ray diffraction patterns of NaA zeolite before (upper) and after (lower) MW treatment subjected to 20 cycles of MW dehydration.
156
N.A. Pinchukova et al. / Chemical Engineering and Processing 95 (2015) 151–158
Table 1 Specific energy consumption required for thermal and MW dehydration of NaA zeolite and for dewatered ethanol preparation Final product
Specific energy consumption, MJ/kg MW zeolite dehydration
Dry zeolite Dewatered ethanol
Literature sources
Calculated
Experimental (Lab-scale)
Predicted (Plant scale)
0.440 0.145
1.94 0.550
0.600 0.17
0.92 [8] 0.83 3]
the value of DHdesorption was assumed to be 30 kJ/mol, and the calculated value of Qdehydration made up 440 kJ/kg of dry zeolite. Experimental energy value Qexp (J) for zeolite dehydration process was determined from the following Eq. (7): Q exp ¼ P t;
(7)
where P is the average power of MW oscillator (W), t is the process time (s), and under given experimental conditions (temperature 230 C, process time 20 min) made up 1.94 MJ/kg of dry zeolite, or 0.55 MJ/kg of dewatered ethanol. Thus, the energy efficiency factor calculated as the ratio Qdehydration/Qexp is 0.23, which is in a good accordance with our previous results obtained in the investigations of MW-assisted drug substances drying and phosphoric acid dehydration [27,28], wherein the same MW installation was used. Thus, it may be concluded that high factors of MW irradiation utilization can hardly be achieved at a laboratory scale. However, in the same works it was shown that energy
efficiency could be increased up to 70–75% due to MW cavity occupancy optimization. So, the results obtained in this laboratory study should be adjusted for the value of MW cavity occupancy ratio, which in our experiments was about 0.01. At this, taking into consideration potential increase of the energy utilization factor, energy consumption for a plant-scale zeolite dehydration can be predicted at the level of 0.6 MJ/kg of dry zeolite or, respectively, 0.17 MJ/kg of dewatered ethanol, provided that the feeding ethanol concentration is about 95%.The values of the energy consumed for zeolite dehydration with the use of different techniques are demonstrated in Table 1. In our results only the energy used for zeolite dehydration was accounted for, as it made the most part of the total energy expenditures related to dewatered ethanol preparation by the proposed method. Table 1 well illustrates the energy benefits from the use of MW irradiation for zeolite dehydration, especially at a plant scale. Such a drastic energy reduction is the result of specific MW heating, whereby the energy is utilized directly by the zeolite, and heat losses are minimized. Moreover, non-thermal MW effects discussed in Section 3.2 also play a great role in total reduction of energy consumption. In this respect there is no alternative to MW heating, since any other heating method would imply surface heating, wherein thermal conductivity of the processed material is crucial. Taking into account low thermal conductivity of zeolites, which are known to be good thermal insulators, it is obvious that considerable heat losses are inevitable. To realize the concept of “in-situ MW zeolite regeneration” the tailored approach should be applied for design of MW equipment. The flow chart of the plantscale setup is shown in Fig. 8. The 3-module setup is proposed to provide continuous ethanol dewatering process, since the whole cycle of dewatered ethanol
Fig. 8. Flow chart of the pilot-plant scale 3-module setup for ethanol dewatering with further in-situ MW zeolite regeneration: 1, 2, 3–MW oscillators; 4, 5, 6–MW cavities with zeolite-filled columns inside. Bold guide lines indicate ethanol dewatering scheme, thinner solid lines indicate zeolite predrying scheme, and dashed lines indicate MW zeolite dehydration scheme.
N.A. Pinchukova et al. / Chemical Engineering and Processing 95 (2015) 151–158
preparation, including zeolite regeneration, is comprised of three constituents (8):
t ¼ t ethanol þ t zeolite þ t cooling ;
(8)
where t ethanol is time required for ethanol dewatering, t zeolite is time required for zeolite dehydration, t cooling is time required for zeolite cooling. According to our experimental results, t ethanol is 30 min, t zeolite is 20 min, and t cooling was approx. 90 min. However, the last step’s duration can be markedly reduced due to optimization of a cooling mode, e.g., compressed gas cooling can be applied. Thus, in the first approximation durations of these three steps can be considered equal or commensurable. If so, the operation mode of MW plantscale setup would be as follows: one of the three modules would operate in the ethanol dewatering mode, at the same time another one, having been previously used for ethanol dewatering, would undergo zeolite dehydration, and the third module, having passed the two above steps, would be in a cooling regime. Thus, 3-module apparatus should provide continuous process of ethanol dewatering and zeolite regeneration. It should be noted that the concept of the module design has been borrowed from our previous investigations on MW-assisted drying of alkali halides [28–30]. In result of this study the tailored MW drying equipment was designed [29], comprising an ampoule loaded with the salt to be dried, and placed in alignment into MW cylindrical cavity. To ensure homogeneous temperature of the bulk product, ampoule rotation was used provided by vacuum-sealed rotating junction. In case of zeolite regeneration, MW setup can be simplified and zeolite-loaded column rotation can be avoided by virtue of applying the improved MW mode pattern [30] ensuring more homogeneous, as compared to conventional MW systems, electromagnetic field distribution throughout MW cavity, and consequently, homogeneous temperature distribution throughout the bulk of the material. The penetration depth for NaA zeolite estimated on the basis of the dielectric properties measurements as described in Section 2.2.4, changes from 3.7 cm for the saturated zeolite to 10.5 cm for the dry one. These values allow effective scale-up of the process, taking into account that the zeolite loaded into a cylindrical column is irradiated by microwaves all round, so the column diameter will be equal to a doubled penetration depth. Moreover, a special custom-made MW inlet was designed in order to increase the efficiency of MW irradiation utilization up to 85% and to decrease a specific metal content of the MW setup [31]. 4. Conclusions The study revealed that regeneration of NaA zeolite used for ethanol dewatering is more efficient with application of microwave irradiation instead of conventional heating in terms of process time shortening and energy cuts. Powder X-ray diffraction measurements showed that even after 20 microwave regeneration cycles the zeolite structure remains the same as in the initial sample. The decrease in energy consumption by the factor of 1.7 as compared to the thermal zeolite dehydration has been obtained at laboratory scale, with 10-fold shortening of MW process time. The reasons of process intensification resulting in considerable energy saving were analyzed. Firstly, it was shown that drastic process acceleration under MW conditions was the result of minimized heating time and energy losses due to the incident properties of microwave heating. Secondly, the hypothesis about non-thermal MW effects resulting from the specific interactions of MW irradiation with matter, contributing to the process acceleration and the model illustrating the influence of MW irradiation on the
157
molecular collision cross-section, based on Arrhenius collision theory, were proposed. The outlook for the process scale up has been considered, and a pilot-plant scale model for continuous ethanol dewatering by NaA zeolite comprising MW-assisted zeolite regeneration stage has been proposed. Taking into account the potential for improving energy efficiency upon scale-up, the energy consumption at the level of 0.17 MJ/kg of dewatered ethanol has been predicted for a MW large scale zeolite dehydration process, which constitutes only 1/5 of the energy consumed in the corresponding thermal process. References [1] M. Boldi^s, K. Melzoch, J. Pokorny, M. Ko^ cirík, Dynamics of sorption columns in dewatering of bioethanol using zeolites, Stud. Surf. Sci. Catal. 142 (2002) 1663– 1670. [2] Zeolite dewatering membranes to be commercialised for biofuels, Filtr. Separat. 45 (2008) 9, www.hyflux.com. [3] E. Nagy, S. Boldyryev, Energy demand of biofuel production applying distillation and/or pervaporation, Chem. Eng. Trans. 35 (2013) 265–270. [4] J.F.G. Oliveira, I.L. Lucena, R.M.A. Saboya, M.L. Rodrigues, A.E.B. Torres, F.A.N. Fernandes, C.L. Cavalcante Jr., E.J.S. Parente Jr., Biodiesel production from waste coconut oil by esterification with ethanol: the effect of water removal by adsorption, Renew. Energ. 35 (2010) 2581–2584. [5] S.K. Wahono, A. Hernawan Kristiani, S. Tursiloadi, H. Abimanyu, Characterization and utilization of gunungkidul natural zeolite for bioethanol dehydration, Energ. Proced. 47 (2014) 263–267. [6] A.A. Lukanov, I.I. Kondratiev, M.Y. Levinter, Zeolite adsorbent, SU Patent 1558, 870, 1990. [7] B.V. Krasiy, G.N. Khomenko, Y.A. Shavandin, Method for preparation of granulated zeolite, SU Patent 1655, 901, 1991. [8] E. Gabrus’, J. Nastaj, P. Tabero, T. Aleksandrzak, Experimental studies on 3A and 4A zeolite molecular sieves regeneration in TSA process: Aliphatic alcohols dewatering–water desorption, Chem. Eng. J. 259 (2015) 232–242. [9] J.M.R. Bélanger, J.R.J. Paré, O. Poon, C. Fairbridge, S. Ng, S. Mutyala, R. Hawkins, Remarks on various applications of microwave energy, J. Microwave Power EE 42 (2008) 24–44. [10] P.M. Bendale, C.-M. Sun, Rapid microwave-assisted liquid-phase combinatorial synthesis of 2-(arylamino) benzimidazoles, J. Comb. Chem. 4 (2002) 359–361. [11] F. Bergamelli, M. Iannelli, J.A. Marafie, J.D. Moseley, A commercial continuous flow microwave reactor evaluated for scale-up, Org. Process Res. Dev. 14 (2010) 926–930. [12] S. Caddick, R. Fitzmaurice, Microwave enhanced synthesis, Tetrahedron 65 (2009) 3325–3355. [13] C.O. Kappe, D. Dallinger, S.S. Murphree, Practical Microwave Synthesis For Organic Chemists, WILEY-VCH, Weinheim, 2009. [14] A. Loupy, Microwaves in Organic Synthesis, 2nd ed., Wiley-VCH, Weinheim, 2006. [15] D. Dallinger, C.O. Kappe, Microwave-assisted synthesis in water as solvent, Chem. Rev. (2007) 2563–2591. [16] B.A. Roberts, C.R. Strauss, Toward rapid, green, predictable microwave-assisted synthesis, Acc. Chem. Res. 38 (2005) 653–661. [17] I. Polaert, L. Estel, R. Huyghe, M. Thomas, Adsorbents regeneration under microwave irradiation for dehydration and volatile organic compounds gas treatment, Chem. Eng. J. 162 (2010) 941. [18] R. Han, Y. Wang, Q. Sun, L. Wang, J. Song, X. He, C. Dou, Malachite green adsorption onto natural zeolite and reuse by microwave irradiation, J. Hazard. Mater. 175 (2010) 1056. ski, M. Komorowska-Durka, G.D. Stefanidis, A.I. Stankiewicz, [19] R. Cherban Microwave swing regeneration vs. temperature swing regeneration— comparison of desorption kinetics, Ind. Eng. Chem. 50 (2011) 8632–8644. [20] M. Komorowska-Durka, R. Van Houten, G.D. Stefanidis, Application of microwave heating to pervaporation: a case study for separation of ethanol–water mixtures, Chem. Eng. Process. 81 (2014) 35–40. [21] Z. Ji, J. Wang, D. Hou, Z. Yin, Z. Luan, Effect of microwave irradiation on vacuum membrane distillation, J. Membrane Sci. 429 (2013) 473–479. [22] P. Bruttel, R. Schlink, Water determination by Karl Fischer Titration, Metrohm Ltd., Herisau, 2006. [23] A.Y. Voloshko, G.D. Kramskoy, Y.D. Kramskoy, V.L. Samoilov, V.P. Seminozhenko, A.A. Chepkiy, O.V. Shishkin, Method for measuring dielectric constant and loss tangent of the material, UA Patent 34, 146, 2008. [24] C. Antonio, R.T. Deam, Can “microwave effects” be explained by enhanced diffusion? Phys. Chem. Chem. Phys. 9 (2007) 2976–2982. [25] J. Rodriguez-Carvajal, T. Roisnel, FullProf.98 and WinPLOTR: New Windows 95/ NT Applications for Diffraction. Commission for Powder Diffraction, International Union of crystallography, Newsletter No. 20 (May–August) Summer 1998. [26] K.F. Loughlin, Water isoterm models for 4A (NaA) zeolite, Adsorption 15 (2009) 337–353.
158
N.A. Pinchukova et al. / Chemical Engineering and Processing 95 (2015) 151–158
[27] N.A. Pinchukova, A.Y. Voloshko, O.V. Shyshkin, V.A. Chebanov, B.H.P. Van de Kruijs, J.C.L. Arts, M.H.C.L. Dressen, J. Meuldijk, J.A.J.M. Vekemans, L.A. Hulshof, A facile microwave-mediated drying process of thermally unstable/labile products, Org. Process Res. Dev. 14 (2010) 1130–1139. [28] 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, Chem. Eng. Process. 50 (2011) 1193–1197.
[29] V.D. Bal'nikh, A.Y. Voloshko, B.V. Grinyov, Y.D. Kramskoy, A.I. Ovchinnikov, N.N. Smirnov, O.V. Shishkin, Method and device for drying free-flowing dielectric materials, UA Patent 56 629, 2005. [30] A.Y. Voloshko, Y.D. Kramskoy, V.L. Samoilov, V.P. Seminozhenko, A.A. Chepkiy, O.V. Shishkin, Method for dielectric materials heating, UA Patent 28 893, 2007. [31] A.Y. Voloshko, V.B. Moiseev, N.A. Pinchukova, V.L. Samoilov, V.P. Seminozhenko, I.B. Stel'mah, O.V. Shishkin, Device for drying free-flowing dielectric materials, UA Patent 84354, 2008.