- Email: [email protected]
Inductive heating of fluidized beds: Drying of particulate solids
Powder Technology 306 (2017) 26–33
Contents lists available at ScienceDirect
Powder Technology journal homepage: www.elsevier.com/locate/powtec
Inductive heating of fluidized beds: Drying of particulate solids Vesselin V. Idakiev ⁎, Pavleta V. Lazarova, Andreas Bück, Evangelos Tsotsas, Lothar Mörl Thermal Process Engineering, NaWiTec, Otto von Guericke University Magdeburg, Universitätsplatz 2, 39106 Magdeburg, Germany
a r t i c l e
i n f o
Article history: Received 12 April 2016 Received in revised form 5 November 2016 Accepted 10 November 2016 Available online 12 November 2016 Keywords: Fluidized bed Inductive heating Drying Inert particles Porous and non-porous solids
a b s t r a c t The present research work deals with drying of particulate solids in inductively heated fluidized bed. Capillaryporous and non-porous solids are dried by inductive and convective (benchmark experiments) heating. The influence of both heating types and test materials on the drying rate and the product moisture content is discussed. The conducted drying trials present very similar drying rates and product moisture contents under the same drying conditions by both energy input methods in fluidized beds. Moreover, it was demonstrated that this novel method for energy input in fluidized beds allows rapid heating and cooling and therefore enables considerably more controllable temperature profiles. In this way, the costly and time consuming start-up processes in batchwise fluidized bed processing can be shorten. This study therefore clearly points out the enormes potential of inductive energy input in fluidized beds. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Fluidized bed technology is widely used in various industries such as the pharmaceutical, food chemistry and environmental technology sectors. Its advantages lie, among other things, in an intensive heat and mass transfer, a targeted temperature control and a compact construction volume [1,2]. In most cases, the energy input to a fluidized bed is realized by a convective heating of the fluidizing medium or in combination with contact heat transfer. In the present paper, a novel method for energy input into fluidized bed apparatus for drying of solid materials is presented. The inductive energy input offers an alternative for the convective heat transfer in fluidized beds. Here, the heat source are electrically conductive particles (e.g. iron hollow balls IHB) in the fluidized bed to which energy is transmitted contact-free via an alternating electromagnetic field. On the surface of the inert particles (IHB) the heat is released directly into fluidized bed. Since the heat is dissipated to the fluidized material through a large total particle surface, a very high energy density and finally a highly efficient heat transfer can be achieved. In this way, the energy efficiency of fluidized bed processes can be increased. Of particular interest is the fast and precise temperature control during the inductive heating of fluidized beds. This is especially important for batch processes in fluidized bed apparatuses. In previous work, the behavior of fluidized bed with inductive heating was investigated by Idakiev et al. [3]. In this paper, this novel technology for energy input in fluidized beds is used to dry particulate solids. The drying of capillary-porous and nonporous solids with convective (benchmark experiments) and inductive
⁎ Corresponding author. E-mail address: [email protected] (V.V. Idakiev).
http://dx.doi.org/10.1016/j.powtec.2016.11.011 0032-5910/© 2016 Elsevier B.V. All rights reserved.
energy input is studied and compared with each other with respect to moisture content of the product and the drying rates. 2. State of the technology Due to their numerous advantages, the fluidized bed technology has gained significant importance in the last century in many industrial processes. To date, a large number of publications and patents on fluidized bulk materials can be found. To the ever-increasing application areas of fluidized bed technology, belong processes such as drying, molding, granulation, agglomeration or combustion [1,2]. Drying is the oldest known method for removing a liquid from wet bulk materials [4]. Already in the older Stone Age, people have dried food to store it for a long time. In 17th and 18th century various substances were usually dried by hot air or smoke. In the 19th century the drying by vacuum or by spraying was developed. A century later, the drum dryer and vacuum-freeze dryer were invented. But the drying has been greatly developed after the Second World War [5]. Today there are many processes to dry solids, suspensions or solutions. For example, drum drying, spray drying, flash drying, drying by microwave, freeze-drying and other [6]. The individual types of dryers can be divided according to different characteristics (dominant heat transfer mechanism, type of product to be dried, pressure in the apparatus, operation and others). In many cases the required heat is supplied by convection. In these dryers, the solid particles are dried using a hot continuously flowing gas stream. In contact drying, heat is transferred through the wet material from heated surfaces by conduction [7]. Freeze-drying and vacuum-freeze drying are often used for drying of heat-sensitive materials. These methods ensure gentle drying of the goods and increase the quality of the dried product. However, their use is not wide-spread because of
V.V. Idakiev et al. / Powder Technology 306 (2017) 26–33
their equipment and operation cost [8]. Heating by infrared radiation enables to reduce the drying time, since the heat of the heating element is transferred directly to the particle surface without that the ambient air is heated in the apparatus [9]. Another approach is microwave drying, as microwaves can penetrate the material to be dried and in this way the heat is generated directly in the solid [10]. Nevertheless, this type of drying is rarely used in the industry due to the high operating costs, compared to convective or contact drying. Fluidized bed apparatuses are often used for convective drying owing to the excellent mass and heat transfer for instance in drying of solids, granules, pharmaceutical and agricultural materials and foodstuffs such as peas, beans, vegetable and onion dices, coffee and other [11]. Fluidized beds are particularly advantageous due to the simple solids handling through the fluid-like behavior, intensive solids mixing and the resulting uniform temperature distribution, large exchange surface between solid and gas and high heat transfer value between solid and gas [1]. In most cases, the heat required for the drying process in fluidized beds is provided by means of a preheated fluidizing gas. Moreover, additional thermal energy can be inserted to the process by either immersed heating elements, e.g. steam-heated tubes, or by heating of the fluidization chamber wall. Due to contact of particles with heated surfaces additional heat transfer is available leading to process intensification, i.e. higher drying rates due to increased heat transfer area. Lower gas inlet temperature is requested, which is preferable for heat-sensitive products. However, limited number of tubes, influence on flow field and fouling are subject to the limitations of the process intensification by contact heating [12]. An alternative method of providing the required heat to fluidized beds is inductive heating. First studies on the inductive energy input into fluidized beds by Stresing et al. [13] have found that the method of non-contact heating of inert particles by electromagnetic alternating fields in fluidized bed apparatuses is possible. In their work, iron hollow balls (IHB) were used as electrically conductive inert particles in the fluidized bed apparatus submitted to an external electromagnetic field. The heat released by the skin effect on the surface of the iron particles is transferred convectively to the fluidizing medium. Skin effect is a tendency for alternating current to flow mostly near the outer surface of an electrical conductor, such as iron balls. It describes the appearance of the current density and occurs with high frequency alternating current. Due to this effect, the current density is largest near the surface of the conductor and decreases towards the center. This is because electric current flows mainly at the “skin” of the conductor. This effect is greater with increase of the frequency. This is a desired effect by the inductive energy input and it is beneficial, because only the particle surface must be heated for heat transfer [3]. Energy losses were reported. Furthermore, it has been found that very fast heating and cooling can be realized [14]. In Idakiev et al. [15]., the influence of the electrical power supplied to the generator and the bed composition on the heating behavior of an inductively heated fluidized bed has been examined and compared with a convection-heated fluidized bed plant. It has been shown that the magnitude of power has no significant effect on the response time of the inductive heating. A higher proportion of inert particles yields higher gas outlet temperatures, since a larger surface of the iron particles for the heat transfer with the fluidizing medium has been available. The authors also reported on the efficiency of the inductive energy input (percentage of the electric power injected via the generator transferred to the fluidizing medium by the IHB) of about 70%. Furthermore, Roßau [16] has found out that the increase of induction power may yield changes in the fluidization behavior in a bed consisting of inert iron particles due to orientation along field lines. These changes in the fluidization behavior observed by Roßau [16] were prevented by coating of iron hollow balls with kaolin or use of pulsed induction power as described in Idakiev et al. [3]. Higher induction powers also have led to stronger fluctuations of the bed pressure drop. With increasing gas velocity these fluctuations have been smaller. Furthermore, smaller iron hollow balls have influenced the fluidization behavior of the fluidized bed more
27
strongly, because a greater surface area for interaction with the magnetic field is available [3]. Idakiev et al. [17] have presented a model for the calculation of heat transfer in an inductively heated fluidized bed. This model allows the calculation of the particle, gas and wall temperature at a certain introduced energy. The authors performed experiments at various mass of the bed, gas velocity and induction power. The calculated temperatures are in very good agreement with the experimentally measured values. An alternative approach to inductive energy input into a fluidized bed has been shown by Latifi et al. [18]. The authors have used an inductive heated fluidized bed mini-reactor, having a reaction zone with a diameter of 0.025 m and a length of 0.3 m, which has been designed and built for solid-state reactions up to 1500 °C. The fluidized bed chamber of this reactor has consisted of an aluminium oxide tube, which has been externally surrounded by a copper coil with 8 windings. Inside, inductively heated steel bars have been installed. In these tests, up to 1500 °C process temperatures could be obtained a few seconds after switching on the electromagnetic alternating field. Again, the short warm-up and cool-down have been observed using the induction heating. Use of IHB for heat transfer is beneficial in terms of very large surface area for heat transfer, co-fluidization with product (moving transfer area), possibly self-cleaning via particle-particle collisions, combining advantages of heating elements and inductive heating. In view of these benefits, use of IHB offers bright prospects for process intensification by induction heating. First results for drying of a porous material using inductive heating were presented by Roßau [16]. The drying process could be accelerated considerably by increasing the induction power. Intensification of the drying process particularly in the first drying stage has been also reported by increase of the inert particle mass mainly due to the increasing surface area for heat transfer. Based on these first experiences for drying capillary porous bulk materials an in-depth study of drying of a non-porous bulk material is presented in this contribution. The product moisture content and the drying rates for a strongly capillary porous bulk material and a non-porous bulk material are investigated. Variations of the gas mass flow and the process temperature are discussed. In addition, a direct comparison of the drying process with convective and inductive energy input to the fluidized bed is made. 3. Experimental setup This section describes the configuration of the cylindrical fluidized bed apparatus. The properties of the used electrically conductive iron hollow balls and wet solid particles to be dried are presented. In addition, this chapter gives a description of the experimental procedure and the experiments carried out with the corresponding varied parameters. 3.1. Experimental apparatus For the experimental investigations of drying of particulate solids by inductive heating a cylindrical fluidized bed apparatus was used with a bed diameter of 200 mm. This plant was designed and built based on the results of previous work and offers a variety of precise measurement capabilities. The plant makes it possible to heat the fluidizing medium both inductively (by addition of IHB) and convectively (via an upstream electric heater), which leads to a reliable comparison between the two types of heating or drying methods, respectively. The setup of the fluidized bed apparatus is presented schematically in Fig. 1. Air was used as fluidization medium in all experiments. Ambient air can be drawn in either by suction blower, pressure blower or with both of them simultaneously. The drying experiments shown in this paper have been carried out exclusively in suction operation so that the gas
28
V.V. Idakiev et al. / Powder Technology 306 (2017) 26–33
Fig. 1. Scheme of the used cylindrical fluidized bed apparatus with optional inductive or convective heating.
temperature could not be affected by warming the air while it flows through the pressure blower. The main part of the fluidized bed apparatus is a cylinder with a height of 1659 mm. The fluidized bed chamber is constructed of a heat-resistant, electrically nonconductive borosilicate glass, which enables operation without feeding electromagnetic energy into the mantle of the fluidized bed chamber. A special feature of this inductively heated plant are the wood flanges connecting the fluidized bed chamber with the pipes. The screws holding the wood flanges and the distributor plate are also made of an electrically non-conducting glassfiber material. The chamber is surrounded by three sets of induction coils, each with three windings (with a size 5 × 30 mm for each rectangular winding), and operated in parallel. Using these, the fluidized bed can be subjected to an electromagnetic field, by which the conductive inert particles are heated. A photo of the fluidized bed chamber with the induction coil and the wood flanges is shown in Fig. 2. The induction coils are connected to an induction generator with a maximum power of 40 kW (TruHeat MF 3040, Hüttinger Elektronik GmbH). The material (electro-copper) used for the induction coils has good electrical conductivity. In order to avoid heating of the copper, the induction coils are cooled with water. In this way, the electrical conductivity of the material is kept constant.
Fig. 2. Photo of the fluidized bed chamber with the induction coil and wood flanges.
To achieve a uniform fluidization of all particles in the fluidized bed, a perforated plate with a triangle partition was used. It was found in the tests that the ratio of bottom pressure drop to the bed pressure drop caused good fluidization. The diameter of the perforated plate is 200 mm. The aperture diameter is 0.8 mm and the partition is 1.9 mm. After passing through the fluidized bed, the fluidizing air is lead first to the cyclone separator and then into a bag filter. During the drying experiments the most important process parameters, temperature, humidity and pressure were measured at different measuring points within the fluidized bed apparatus. Monitoring and the control of the experiments were performed with Windows Control Center (WinCC, Siemens AG). The temperature measurement was taken with thermocouples. The gas temperature was detected at a total of 14 measurement points (before and after all important plant components such as fluidized bed chamber, cyclone, filter, heater and others). The pressure was measured by capacitive pressure sensors (Kalinsky Sensor Elektronik). The measurement of the air mass flow was carried out using orifice pressure drop measurement. Moreover, the bed, bottom, cyclone and filter pressure drop were measured continuously. The relative ambient humidity and the humidity at the exit from the fluidized bed chamber were measured by moisture measuring sensors from Testo. The data for the humidity were saved with the Testo Comfort software program at a 0.66 Hz clock pulse. 3.2. Experimental material The bulk material used for drying experiments consisted of electrically non-conductive particles and electrically conductive IHB. The availability of IHB with diameters from 2 mm to 11 mm with different wall thickness opens up a wide fluidization regime. The manufacturing method developed by the company hollomet GmbH allows both to generate nearly monodisperse IHB with a defined diameter and to vary the layer thickness of the produced particles [19], allowing the production of conductive particles with similar fluidization behavior. In this work IHB were used with a Sauter mean diameter (average of particle size which is defined as the diameter of a sphere that has the same volume/surface area ratio as a particle of interest) of 6.75 mm and an iron layer thickness of 120 μm (see Fig. 3). The iron particles have a sphericity of 99% and an apparent density of 693 kg/m3. To investigate the drying of wet granulates, porous gamma aluminium oxide spheres (γ-Al2O3) and non-porous alpha aluminium oxide spheres (α-Al2O3) were used as model substances. The determined adsorption isotherms (shown in Fig. 4) of the porous and non-porous material show the difference in the particle porosity. The capillary porous γ-Al2O3 spheres have a Sauter mean diameter of 1.80 mm, an apparent density of 678 kg/m3 and a sphericity of 98.4%. The α-Al2O3 particles produced by sintering process of 1.8 mm
Fig. 3. Iron hollow balls with a Sauter mean diameter of 6.75 mm and an iron layer thickness of 120 μm.
V.V. Idakiev et al. / Powder Technology 306 (2017) 26–33
Fig. 4. Adsorption isotherms of γ-Al2O3 (primary axis, left) and α-Al2O3 (secondary axis, right).
29
achieve a constant outlet gas temperature of 60 °C compared to the convective energy input [15]. Once a constant process temperature of 60 °C has been achieved at the outlet of the fluidized bed chamber, the wet material was fed into the fluidized bed, starting the drying process. The time from product inlet to turning off the heating was set to 40 min. After this phase, the dried material was further fluidized and cooled for additional 40 min by ambient air. During the drying experiments samples were regularly taken with a mass of approximately 10 g. From the beginning of the drying process (after the addition of the wet material) until a constant gas outlet temperature has been reached, samples from the fluidized bed were taken every second minute. After reaching a constant gas temperature, samples were taken every tenth minute. Immediately after withdrawal, the samples were measured for their moisture using a Sartorius Moisture Infrared Analyzer MA100. Control measurements using of drying oven have confirmed the values measured with the moisture analyzer Sartorius MA100. 4. Results and discussion
γ-Al2O3 spheres have a smaller Sauter mean diameter of 1.46 mm, a higher apparent density of 1497 kg/m3 and also a sphericity of 98.4%. The IHB and the wet granulates were fluidized and subjected to the alternating electromagnetic field together. Despite the difference between the diameters and densities of both types of particles a very good mixing and no separation of the particles in the fluidized bed were observed. 3.3. Experimental design In order to compare both drying methods (inductive and convective), firstly, the drying of wet granulates was realized with convective energy input. The conducted experiments were then repeated whereby the wet material was dried by an inductive heating of co-fluidized IHB. During the trials the main process parameters such as gas temperature and moistures were monitored electronically at one-second intervals and recorded. The experiments were realized with a total bed mass of 3.0 kg, consisting of 0.9 kg IHB and 2.1 kg wet material in all cases. The bed material was fluidized at threefold and fourfold minimum fluidization velocity of IHB (3vmf ≈ 3.3 m/s, 4vmf ≈ 4.4 m/s). By both types of heating the initial temperature in the fluidized bed chamber was set to values of 60 °C and 80 °C (prior to addition of the wet material, measured at the outlet of the fluidized bed chamber). The material to be dried was moistened to values from 24 to 34 mass-% moisture. A tabular presentation of the experiments reported in this paper is given in the Table 1. In order to ensure better comparability, all drying tests were carried out following the same procedure. First, the IHB were fluidized at constant gas velocity. Then the convective or inductive bed heating was turned on. By the inductive heating over 70% less time is required to
In this section, the results of the drying experiments are presented and discussed. Here the temperature profiles of the outlet gas temperature are shown for both types of heating, both test materials and variation of the air velocity and air temperature. In addition, the moisture content of the product as a function of experimental time and the drying rate as a function of product moisture content are compared. Table 1 gives an overview of the conducted experiments. In the first 4 trials, the drying of a capillary porous (γ-Al2O3) and of a non-porous bulk material (α-Al2O3) was examined, whereby the type of the heating of the fluidized bed was varied. In experiments 5 to 12, γ-Al2O3 was selected as test material and the gas velocity and gas temperature were varied. 4.1. Drying of porous and non-porous material Fig. 5 shows the curves for the gas outlet temperature over experimental time for drying of wet γ-Al2O3 and wet α-Al2O3 granules. The cylindrical fluidized bed was heated by convection via the fluidizing gas or by inductive heating of electrically conductive IHB. The gas outlet temperature was about 60 °C prior to the addition of the wet material in both energy input methods. In both types of heating, electric powers was set at the level securing comparable gas outlet temperatures in the initial state. The efficiencies of the heating types were taken into account. After adding the wet granulates to the bed the gas outlet temperature dropped immediately. In the four trials, two drying stages can be well differentiated from each other. The gas outlet temperature in the
Table 1 Experimental parameters. Trial number
Bed material
1 2 3 4 5 6 7 8 9 10 11 12
IHB + γ-Al2O3
d32,
d32, Gran.
Energy input
ϕGran. vgas
IHB
mm
mm
–
%
m/s
6.75
1.80
Conv. Ind. Conv. Ind. Conv. Ind. Conv. Ind. Conv. Ind. Conv. Ind.
34.40 34.00 29.58 29.57 23.99 23.71 23.78 24.09 23.99 23.71 23.72 22.90
3.30 63.00 20.70 63.70 20.10 3.24 63.90 20.15 4.32 63.70 19.95 3.24 64.10 20.55 85.00 20.22
IHB + α-Al2O3
1.46
IHB + γ-Al2O3
1.80
ϑgas, inlet
Fig. 5. Gas outlet temperature profile during drying of γ-Al2O3 and α-Al2O3 particles by convective and inductive heating of fluidized bed.
30
V.V. Idakiev et al. / Powder Technology 306 (2017) 26–33
Fig. 6. Comparison of the measured product moisture content in drying of γ-Al2O3 particles by convective and inductive heating of fluidized bed.
Fig. 8. Drying rate during drying of γ-Al2O3 and α-Al2O3 particles by convective and inductive heating of fluidized bed.
first drying stage by the convective energy input was about 30 °C. By the inductive heating the first drying stage took place at a gas outlet temperature of about 32 °C. After addition of the wet granulates in case of inductive heating, the bed height was influenced, and thereby a better fluidization or better distribution of the IHB over the whole height of the induction coil was achieved. The reason for the temperature difference between the both heating types in the first drying stage is then attributed to the better use of the IHB surface area in an alternating electromagnetic field. The same tendencies of temperature profiles can be observed also during the drying of non-porous α-Al2O3 particles. The energy fed in was transferred to the fluidization medium, which increases the gas outlet temperature leading to dry the wet material. On the other hand, the energy was also transferred partly by the direct contact of the IHB with the wet granulates. Through this effect, the drying process can be intensified at the same time. After evaporation of surface moisture, the first drying stage moves into the second drying stage. After the 40 min of the heating phase, the heating was turned off, whereby by the inductive energy input the gas outlet temperature decreased immediately. This can be justified by the direct use of cold air into the fluidized bed chamber whereas in convective heating the heated pipes provide further heat for a certain time. In the case of drying of wet non-porous α-Al2O3 granulates, the first drying stage lasts about 400 s in both types of heating. Due to the use of a non-porous material, the duration of the second drying stages is significantly shorter. Afterwards the gas outlet temperature increased rapidly. Reaching ambient temperature in the fluidized bed chamber after
turning off the relevant heating took less than 1 min by inductive heating and more than 25 min by convective heating. The slight fluctuations (every 2 min) in the curves in Fig. 5 are due to sampling. The samples were taken to measure the moisture of the dried material. A decisive feature of the drying experiments was the evaporated amount of water and the final moisture content of the product. In Figs. 6 and 7 the measured moisture content of both experimental materials are shown for both types of heating. Both figures present the change in the moisture content with the experimental time and the evaporated mass of water roughly calculated from the gas outlet temperatures. Figs. 6 and 7 show clearly similar trends and similar moisture contents under the same drying conditions with respect to the energy input method. At the end of the drying process, the water content in dry material obtained by both types of heating is approaching zero. In Fig. 8, the drying rates related to the particle surface are shown as a function of the product moisture content of both bulk materials attained by both types of heating. The drying rate has its maximum value in the first drying stage. After the transition from the first drying stage to the second drying stage, the drying rate decreases rapidly for both materials. The slight fluctuations in the curves of the drying rates in Fig. 8 show the periodic sampling. The strong fluctuations in the drying rate at the beginning of drying experiments of α-Al2O3 particles can be accounted for by a non-optimal addition of the wet granulates. At the end of the
Fig. 7. Comparison of the measured product moisture content in drying of α-Al2O3 particles by convective and inductive heating of fluidized bed.
Fig. 9. Gas outlet temperature profile during drying of γ-Al2O3 particles by convective and inductive heating of fluidized bed and by variation of the air velocity.
V.V. Idakiev et al. / Powder Technology 306 (2017) 26–33
Fig. 10. Comparison of the measured product moisture content in drying of γ-Al2O3 particles by convective and inductive heating of fluidized bed and by variation of the air velocity (Exp. 5 and 6).
drying process, the drying rate by both energy input methods is approaching zero. Figs. 6, 7 and 8 show that the type of the energy input to the cylindrical fluidized bed has no significant impact on the moisture content and the drying rate. 4.2. Variation of the gas velocity In order to investigate the influence of the gas velocity on the drying process, four experiments were carried out with convective and inductive heating under variation of the gas velocity. The experiments were realized by a threefold or fourfold minimum fluidization velocity of IHB (about 3.24 and 4.32 m/s). During these experiments, the other process parameters (such as bed mass and gas temperature, and initial moisture content of the granulates) were kept constant. The temperature profiles of the drying trials with the variation of the gas velocity are presented in Fig. 9. With the increase in gas velocity it is to be observed that the first drying stage is shorter in the both convective and inductive heating. With the increase in gas velocity a more intensive fluidization and a more intensive heat transfer were observed, which justify the intensified drying of wet granulates. The slight fluctuations in the curves of the gas outlet temperatures in Fig. 9 show the periodic sampling.
Fig. 11. Comparison of the measured product moisture content in drying of γ-Al2O3 particles by convective and inductive heating of fluidized bed and by variation of the air velocity (Exp. 7 and 8).
31
Fig. 12. Drying rate during drying of γ-Al2O3 particles by convective and inductive heating of fluidized bed and by variation of the air velocity.
In Figs. 10 and 11 the measured moisture content is shown for both types of heating and for varied air velocity. Both figures present the change in moisture content with experimental time and the evaporated water mass roughly calculated from the gas outlet temperatures. By comparison of Figs. 10 and 11, it can be seen that the curves describing the evaporation of water are steeper, which corresponds to an intensified evaporation of water with increasing the gas velocity. The course of the drying rates in the experiments with a variation of the gas velocity and the type of heating of fluidized bed is shown graphically in Fig. 12. 4.3. Variation of the initial gas temperature The influence of initial gas temperature on the drying process was investigated by means of four experiments with variation of the type of heating of fluidized bed. Fig. 13 shows the gas outlet temperature during the drying process. The fluidized bed was initially heated to a target temperature of 60 or 80 °C by induction or convection. After adding the wet granulates to the bed the gas outlet temperature dropped immediately. The first drying stage lasts about 300 s for both initial gas temperatures. In both trials, the two drying stages can be well differentiated from each other. After the 40 min of the inductive or convective heating phase, the heating was turned off, whereby in the inductive
Fig. 13. Gas outlet temperature profile during drying of γ-Al2O3 particles by convective and inductive heating of fluidized bed and by variation of initial gas temperature.
32
V.V. Idakiev et al. / Powder Technology 306 (2017) 26–33
Fig. 14. Comparison of the measured product moisture content in drying of γ-Al2O3 particles by convective and inductive heating of fluidized bed and by variation of the initial gas temperature (Exp. 9 and 10).
energy input the gas outlet temperature again decreased immediately (see the grey lines in Fig. 13). In Figs. 14 and 15, the measured moisture content is shown by variation of the air velocity in both types of heating. These two figures show similar trends and similar moisture contents under the same drying conditions in both energy input methods in fluidized beds. When comparing the calculated mass of water evaporated from the wet granulates in the experiments with varying the gas temperature, it can be clearly seen that the deviation between the measured mass of evaporated water and product moisture content is smaller in Fig. 15. The reason for this is the intensified drying of the bulk material at a higher temperature (see the discussion in Section 4.1). Fig. 16 shows the drying rate in experiments with variation of the initial gas temperature and the type of heating of fluidized bed. Here it can be seen that a higher initial gas temperature leads to a slight acceleration of the drying process. This graph demonstrates once again that the energy input method in fluidized bed has no significant impact on the drying process with respect to the product quality (product moisture content).
Fig. 16. Drying rate during drying of γ-Al2O3 particles by convective and inductive heating of fluidized bed and by variation of the initial gas temperature.
moisture content were presented. Within the scope of the performed drying experiments it was demonstrated that the induction technology enables massively reduced heating and cooling times and therefore considerably more controllable temperature profiles. This allows to shorten the costly and time consuming start-up processes in batchwise fluidized bed processing. Using IHB, very large surface area for heat transfer is available. Cofluidization of electrically conductive IHB with the product to be treated, provides movable heat transfer area. Particle-particle collisions cater for self-cleaning of IHB. Furthermore, use of IHB combines advantages of conventional heating methods and inductive heating. Considering these benefits, induction heating is a promising method for process intensification. Conducted drying experiments show very similar drying rates and product moisture contents under the same drying conditions in the both energy input methods in fluidized beds. The present study confirms the great potential of inductive energy input in fluidized beds and thus offers new possibilities for product treatment under a precise temperature control and a constant product quality.
5. Summary Latin letters In this research work a new method for energy input in a cylindrical fluidized bed for drying particulate solids was tested. Effects of different types of heating and test materials on the drying rate and the product
Symbol
Unit
Meaning
d32 v
[mm] [m/s]
Sauter mean diameter Velocity
Greek letters
Symbol
Unit
Meaning
ϕ ϑ
[%] [°C]
Moisture Temperature
Indexes
Fig. 15. Comparison of the measured product moisture content in drying of γ-Al2O3 particles by convective and inductive heating of fluidized bed and by variation of the initial gas temperature (Exp. 11 and 12).
Symbol
Meaning
gas IHB Gran. mf w
Gas Iron hollow balls Granulate Minimum fluidization Water
V.V. Idakiev et al. / Powder Technology 306 (2017) 26–33
Abbreviations
Symbol
Meaning
Conv. DM Exp. IHB Ind. m RH t W
Convective Dry matter Experiment Iron hollow balls Inductive Mass Relative humidity Temperature Water
Acknowledgments The authors gratefully acknowledge the funding of this work by the German Federal Ministry of Science and Education (BMBF) as part of the InnoProfile-Transfer project NaWiTec (grant numbers: 03IPT701A). References [1] H. Uhlemann, L. Mörl, Wirbelschichtsprühgranulation, Springer-Verlag, Berlin, Heidelberg, New York, 2000. [2] L. Mörl, S. Heinrich, M. Peglow, Fluidized bed spray granulation, in: A.D. Salman, M.J. Hounslow, J.P.K. Seville (Eds.), Handbook of Powder Technology, 11, Elsevier, Amsterdam, 2007. [3] V.V. Idakiev, S. Marx, A. Roßau, A. Bück, E. Tsotsas, L. Mörl, Inductive heating of fluidized beds: influence on fluidization behavior, Powder Technol. 286 (2015) 90–97. [4] O. Krischer, W. Kast, Die wissenschaftlichen Grundlagen der Trocknungstechnik, Springer-Verlag, Berlin/Heidelberg/New York, 1978.
33
[5] H. Hayashi, Drying technologies of foods - their history and future, Dry. Technol. 7 (2) (1989) 315–369. [6] M.R. Okos, Food dehydration, in: D.R. Heldmann (Ed.), Handbook for Food Engineering, 2 ed.Marcel Dekker, New York 1992, pp. 601–740. [7] C.G.J. Baker, Industrial drying of foods, 309, Blackie Academic & Professional, London, UK, 1997. [8] C. Ratti, Hot air and freeze-drying of high-value foods- a review, J. Food Eng. 49 (2001) 311–319. [9] P. Jones, Electromagnetic wave energy in drying processes, in: A.S. Mujumdar (Ed.), Drying’92, Elsevier Science, Amsterdam 1992, pp. 114–136. [10] X.C. Jia, Study of heat pump assisted microwave drying, Dry. Technol. 11 (7) (1993) 1583–1616. [11] K.S. Jayaraman, D.K. Das Gupta, Drying of fruits and vegetables, in: A.S. Mujumdar (Ed.), Developments in Drying: Food Dehydration, Kasetsart University Press, Bangkok 1995, pp. 179–206. [12] E. Tsotsas, A.S. Mujumdar, Modern drying technology, Volume 5: Process Intensification, WILEY-VCH Verlag GmbH & Co. KGaA, 2014. [13] A. Stresing, L. Mörl, J. Neum, M. Jacob, K. Walther, Non-Contact Energy Transfer to a Fluidized Bed, in Proceedings of European Drying Conference, Palma de Mallorca, Spain, 2011. [14] A. Stresing, L. Mörl, A. Khaidurova, M. Jacob, K. Walther, Investigation of the time response of a fluidized bed with inductive heating and the influencing variables, Chem. Ing. Tech. 85 (3) (2013) 308–312. [15] V.V. Idakiev, A. Bück, E. Tsotsas, L. Mörl, Inductive energy input in fluidized beds, Proceedings of Conference Machines, Technologies, Materials, Varna. Bulgaria 2015, pp. 29–32. [16] A. Roßau, Induktiver Energieeintrag in eine fluidisierte Schüttung(Dissertation) Otto von Guericke University Magdeburg, 2013. [17] V.V. Idakiev, A. Bück, E. Tsotsas, L. Mörl, Model-based studies on the heat transfer in an inductively heated fluidized bed, Chem. Ing. Tech. 88 (5) (2016) 656–665. [18] M. Latifi, J. Chaouki, A novel induction heating fluidized bed reactor: its design and applications in high temperature screening tests with solid feedstocks and prediction of defluidization state, AICHE J. 61 (5) (2015) 1507–1523. [19] hollomet®, hollomet Brochure, 2014.
Contents lists available at ScienceDirect
Powder Technology journal homepage: www.elsevier.com/locate/powtec
Inductive heating of fluidized beds: Drying of particulate solids Vesselin V. Idakiev ⁎, Pavleta V. Lazarova, Andreas Bück, Evangelos Tsotsas, Lothar Mörl Thermal Process Engineering, NaWiTec, Otto von Guericke University Magdeburg, Universitätsplatz 2, 39106 Magdeburg, Germany
a r t i c l e
i n f o
Article history: Received 12 April 2016 Received in revised form 5 November 2016 Accepted 10 November 2016 Available online 12 November 2016 Keywords: Fluidized bed Inductive heating Drying Inert particles Porous and non-porous solids
a b s t r a c t The present research work deals with drying of particulate solids in inductively heated fluidized bed. Capillaryporous and non-porous solids are dried by inductive and convective (benchmark experiments) heating. The influence of both heating types and test materials on the drying rate and the product moisture content is discussed. The conducted drying trials present very similar drying rates and product moisture contents under the same drying conditions by both energy input methods in fluidized beds. Moreover, it was demonstrated that this novel method for energy input in fluidized beds allows rapid heating and cooling and therefore enables considerably more controllable temperature profiles. In this way, the costly and time consuming start-up processes in batchwise fluidized bed processing can be shorten. This study therefore clearly points out the enormes potential of inductive energy input in fluidized beds. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Fluidized bed technology is widely used in various industries such as the pharmaceutical, food chemistry and environmental technology sectors. Its advantages lie, among other things, in an intensive heat and mass transfer, a targeted temperature control and a compact construction volume [1,2]. In most cases, the energy input to a fluidized bed is realized by a convective heating of the fluidizing medium or in combination with contact heat transfer. In the present paper, a novel method for energy input into fluidized bed apparatus for drying of solid materials is presented. The inductive energy input offers an alternative for the convective heat transfer in fluidized beds. Here, the heat source are electrically conductive particles (e.g. iron hollow balls IHB) in the fluidized bed to which energy is transmitted contact-free via an alternating electromagnetic field. On the surface of the inert particles (IHB) the heat is released directly into fluidized bed. Since the heat is dissipated to the fluidized material through a large total particle surface, a very high energy density and finally a highly efficient heat transfer can be achieved. In this way, the energy efficiency of fluidized bed processes can be increased. Of particular interest is the fast and precise temperature control during the inductive heating of fluidized beds. This is especially important for batch processes in fluidized bed apparatuses. In previous work, the behavior of fluidized bed with inductive heating was investigated by Idakiev et al. [3]. In this paper, this novel technology for energy input in fluidized beds is used to dry particulate solids. The drying of capillary-porous and nonporous solids with convective (benchmark experiments) and inductive
⁎ Corresponding author. E-mail address: [email protected] (V.V. Idakiev).
http://dx.doi.org/10.1016/j.powtec.2016.11.011 0032-5910/© 2016 Elsevier B.V. All rights reserved.
energy input is studied and compared with each other with respect to moisture content of the product and the drying rates. 2. State of the technology Due to their numerous advantages, the fluidized bed technology has gained significant importance in the last century in many industrial processes. To date, a large number of publications and patents on fluidized bulk materials can be found. To the ever-increasing application areas of fluidized bed technology, belong processes such as drying, molding, granulation, agglomeration or combustion [1,2]. Drying is the oldest known method for removing a liquid from wet bulk materials [4]. Already in the older Stone Age, people have dried food to store it for a long time. In 17th and 18th century various substances were usually dried by hot air or smoke. In the 19th century the drying by vacuum or by spraying was developed. A century later, the drum dryer and vacuum-freeze dryer were invented. But the drying has been greatly developed after the Second World War [5]. Today there are many processes to dry solids, suspensions or solutions. For example, drum drying, spray drying, flash drying, drying by microwave, freeze-drying and other [6]. The individual types of dryers can be divided according to different characteristics (dominant heat transfer mechanism, type of product to be dried, pressure in the apparatus, operation and others). In many cases the required heat is supplied by convection. In these dryers, the solid particles are dried using a hot continuously flowing gas stream. In contact drying, heat is transferred through the wet material from heated surfaces by conduction [7]. Freeze-drying and vacuum-freeze drying are often used for drying of heat-sensitive materials. These methods ensure gentle drying of the goods and increase the quality of the dried product. However, their use is not wide-spread because of
V.V. Idakiev et al. / Powder Technology 306 (2017) 26–33
their equipment and operation cost [8]. Heating by infrared radiation enables to reduce the drying time, since the heat of the heating element is transferred directly to the particle surface without that the ambient air is heated in the apparatus [9]. Another approach is microwave drying, as microwaves can penetrate the material to be dried and in this way the heat is generated directly in the solid [10]. Nevertheless, this type of drying is rarely used in the industry due to the high operating costs, compared to convective or contact drying. Fluidized bed apparatuses are often used for convective drying owing to the excellent mass and heat transfer for instance in drying of solids, granules, pharmaceutical and agricultural materials and foodstuffs such as peas, beans, vegetable and onion dices, coffee and other [11]. Fluidized beds are particularly advantageous due to the simple solids handling through the fluid-like behavior, intensive solids mixing and the resulting uniform temperature distribution, large exchange surface between solid and gas and high heat transfer value between solid and gas [1]. In most cases, the heat required for the drying process in fluidized beds is provided by means of a preheated fluidizing gas. Moreover, additional thermal energy can be inserted to the process by either immersed heating elements, e.g. steam-heated tubes, or by heating of the fluidization chamber wall. Due to contact of particles with heated surfaces additional heat transfer is available leading to process intensification, i.e. higher drying rates due to increased heat transfer area. Lower gas inlet temperature is requested, which is preferable for heat-sensitive products. However, limited number of tubes, influence on flow field and fouling are subject to the limitations of the process intensification by contact heating [12]. An alternative method of providing the required heat to fluidized beds is inductive heating. First studies on the inductive energy input into fluidized beds by Stresing et al. [13] have found that the method of non-contact heating of inert particles by electromagnetic alternating fields in fluidized bed apparatuses is possible. In their work, iron hollow balls (IHB) were used as electrically conductive inert particles in the fluidized bed apparatus submitted to an external electromagnetic field. The heat released by the skin effect on the surface of the iron particles is transferred convectively to the fluidizing medium. Skin effect is a tendency for alternating current to flow mostly near the outer surface of an electrical conductor, such as iron balls. It describes the appearance of the current density and occurs with high frequency alternating current. Due to this effect, the current density is largest near the surface of the conductor and decreases towards the center. This is because electric current flows mainly at the “skin” of the conductor. This effect is greater with increase of the frequency. This is a desired effect by the inductive energy input and it is beneficial, because only the particle surface must be heated for heat transfer [3]. Energy losses were reported. Furthermore, it has been found that very fast heating and cooling can be realized [14]. In Idakiev et al. [15]., the influence of the electrical power supplied to the generator and the bed composition on the heating behavior of an inductively heated fluidized bed has been examined and compared with a convection-heated fluidized bed plant. It has been shown that the magnitude of power has no significant effect on the response time of the inductive heating. A higher proportion of inert particles yields higher gas outlet temperatures, since a larger surface of the iron particles for the heat transfer with the fluidizing medium has been available. The authors also reported on the efficiency of the inductive energy input (percentage of the electric power injected via the generator transferred to the fluidizing medium by the IHB) of about 70%. Furthermore, Roßau [16] has found out that the increase of induction power may yield changes in the fluidization behavior in a bed consisting of inert iron particles due to orientation along field lines. These changes in the fluidization behavior observed by Roßau [16] were prevented by coating of iron hollow balls with kaolin or use of pulsed induction power as described in Idakiev et al. [3]. Higher induction powers also have led to stronger fluctuations of the bed pressure drop. With increasing gas velocity these fluctuations have been smaller. Furthermore, smaller iron hollow balls have influenced the fluidization behavior of the fluidized bed more
27
strongly, because a greater surface area for interaction with the magnetic field is available [3]. Idakiev et al. [17] have presented a model for the calculation of heat transfer in an inductively heated fluidized bed. This model allows the calculation of the particle, gas and wall temperature at a certain introduced energy. The authors performed experiments at various mass of the bed, gas velocity and induction power. The calculated temperatures are in very good agreement with the experimentally measured values. An alternative approach to inductive energy input into a fluidized bed has been shown by Latifi et al. [18]. The authors have used an inductive heated fluidized bed mini-reactor, having a reaction zone with a diameter of 0.025 m and a length of 0.3 m, which has been designed and built for solid-state reactions up to 1500 °C. The fluidized bed chamber of this reactor has consisted of an aluminium oxide tube, which has been externally surrounded by a copper coil with 8 windings. Inside, inductively heated steel bars have been installed. In these tests, up to 1500 °C process temperatures could be obtained a few seconds after switching on the electromagnetic alternating field. Again, the short warm-up and cool-down have been observed using the induction heating. Use of IHB for heat transfer is beneficial in terms of very large surface area for heat transfer, co-fluidization with product (moving transfer area), possibly self-cleaning via particle-particle collisions, combining advantages of heating elements and inductive heating. In view of these benefits, use of IHB offers bright prospects for process intensification by induction heating. First results for drying of a porous material using inductive heating were presented by Roßau [16]. The drying process could be accelerated considerably by increasing the induction power. Intensification of the drying process particularly in the first drying stage has been also reported by increase of the inert particle mass mainly due to the increasing surface area for heat transfer. Based on these first experiences for drying capillary porous bulk materials an in-depth study of drying of a non-porous bulk material is presented in this contribution. The product moisture content and the drying rates for a strongly capillary porous bulk material and a non-porous bulk material are investigated. Variations of the gas mass flow and the process temperature are discussed. In addition, a direct comparison of the drying process with convective and inductive energy input to the fluidized bed is made. 3. Experimental setup This section describes the configuration of the cylindrical fluidized bed apparatus. The properties of the used electrically conductive iron hollow balls and wet solid particles to be dried are presented. In addition, this chapter gives a description of the experimental procedure and the experiments carried out with the corresponding varied parameters. 3.1. Experimental apparatus For the experimental investigations of drying of particulate solids by inductive heating a cylindrical fluidized bed apparatus was used with a bed diameter of 200 mm. This plant was designed and built based on the results of previous work and offers a variety of precise measurement capabilities. The plant makes it possible to heat the fluidizing medium both inductively (by addition of IHB) and convectively (via an upstream electric heater), which leads to a reliable comparison between the two types of heating or drying methods, respectively. The setup of the fluidized bed apparatus is presented schematically in Fig. 1. Air was used as fluidization medium in all experiments. Ambient air can be drawn in either by suction blower, pressure blower or with both of them simultaneously. The drying experiments shown in this paper have been carried out exclusively in suction operation so that the gas
28
V.V. Idakiev et al. / Powder Technology 306 (2017) 26–33
Fig. 1. Scheme of the used cylindrical fluidized bed apparatus with optional inductive or convective heating.
temperature could not be affected by warming the air while it flows through the pressure blower. The main part of the fluidized bed apparatus is a cylinder with a height of 1659 mm. The fluidized bed chamber is constructed of a heat-resistant, electrically nonconductive borosilicate glass, which enables operation without feeding electromagnetic energy into the mantle of the fluidized bed chamber. A special feature of this inductively heated plant are the wood flanges connecting the fluidized bed chamber with the pipes. The screws holding the wood flanges and the distributor plate are also made of an electrically non-conducting glassfiber material. The chamber is surrounded by three sets of induction coils, each with three windings (with a size 5 × 30 mm for each rectangular winding), and operated in parallel. Using these, the fluidized bed can be subjected to an electromagnetic field, by which the conductive inert particles are heated. A photo of the fluidized bed chamber with the induction coil and the wood flanges is shown in Fig. 2. The induction coils are connected to an induction generator with a maximum power of 40 kW (TruHeat MF 3040, Hüttinger Elektronik GmbH). The material (electro-copper) used for the induction coils has good electrical conductivity. In order to avoid heating of the copper, the induction coils are cooled with water. In this way, the electrical conductivity of the material is kept constant.
Fig. 2. Photo of the fluidized bed chamber with the induction coil and wood flanges.
To achieve a uniform fluidization of all particles in the fluidized bed, a perforated plate with a triangle partition was used. It was found in the tests that the ratio of bottom pressure drop to the bed pressure drop caused good fluidization. The diameter of the perforated plate is 200 mm. The aperture diameter is 0.8 mm and the partition is 1.9 mm. After passing through the fluidized bed, the fluidizing air is lead first to the cyclone separator and then into a bag filter. During the drying experiments the most important process parameters, temperature, humidity and pressure were measured at different measuring points within the fluidized bed apparatus. Monitoring and the control of the experiments were performed with Windows Control Center (WinCC, Siemens AG). The temperature measurement was taken with thermocouples. The gas temperature was detected at a total of 14 measurement points (before and after all important plant components such as fluidized bed chamber, cyclone, filter, heater and others). The pressure was measured by capacitive pressure sensors (Kalinsky Sensor Elektronik). The measurement of the air mass flow was carried out using orifice pressure drop measurement. Moreover, the bed, bottom, cyclone and filter pressure drop were measured continuously. The relative ambient humidity and the humidity at the exit from the fluidized bed chamber were measured by moisture measuring sensors from Testo. The data for the humidity were saved with the Testo Comfort software program at a 0.66 Hz clock pulse. 3.2. Experimental material The bulk material used for drying experiments consisted of electrically non-conductive particles and electrically conductive IHB. The availability of IHB with diameters from 2 mm to 11 mm with different wall thickness opens up a wide fluidization regime. The manufacturing method developed by the company hollomet GmbH allows both to generate nearly monodisperse IHB with a defined diameter and to vary the layer thickness of the produced particles [19], allowing the production of conductive particles with similar fluidization behavior. In this work IHB were used with a Sauter mean diameter (average of particle size which is defined as the diameter of a sphere that has the same volume/surface area ratio as a particle of interest) of 6.75 mm and an iron layer thickness of 120 μm (see Fig. 3). The iron particles have a sphericity of 99% and an apparent density of 693 kg/m3. To investigate the drying of wet granulates, porous gamma aluminium oxide spheres (γ-Al2O3) and non-porous alpha aluminium oxide spheres (α-Al2O3) were used as model substances. The determined adsorption isotherms (shown in Fig. 4) of the porous and non-porous material show the difference in the particle porosity. The capillary porous γ-Al2O3 spheres have a Sauter mean diameter of 1.80 mm, an apparent density of 678 kg/m3 and a sphericity of 98.4%. The α-Al2O3 particles produced by sintering process of 1.8 mm
Fig. 3. Iron hollow balls with a Sauter mean diameter of 6.75 mm and an iron layer thickness of 120 μm.
V.V. Idakiev et al. / Powder Technology 306 (2017) 26–33
Fig. 4. Adsorption isotherms of γ-Al2O3 (primary axis, left) and α-Al2O3 (secondary axis, right).
29
achieve a constant outlet gas temperature of 60 °C compared to the convective energy input [15]. Once a constant process temperature of 60 °C has been achieved at the outlet of the fluidized bed chamber, the wet material was fed into the fluidized bed, starting the drying process. The time from product inlet to turning off the heating was set to 40 min. After this phase, the dried material was further fluidized and cooled for additional 40 min by ambient air. During the drying experiments samples were regularly taken with a mass of approximately 10 g. From the beginning of the drying process (after the addition of the wet material) until a constant gas outlet temperature has been reached, samples from the fluidized bed were taken every second minute. After reaching a constant gas temperature, samples were taken every tenth minute. Immediately after withdrawal, the samples were measured for their moisture using a Sartorius Moisture Infrared Analyzer MA100. Control measurements using of drying oven have confirmed the values measured with the moisture analyzer Sartorius MA100. 4. Results and discussion
γ-Al2O3 spheres have a smaller Sauter mean diameter of 1.46 mm, a higher apparent density of 1497 kg/m3 and also a sphericity of 98.4%. The IHB and the wet granulates were fluidized and subjected to the alternating electromagnetic field together. Despite the difference between the diameters and densities of both types of particles a very good mixing and no separation of the particles in the fluidized bed were observed. 3.3. Experimental design In order to compare both drying methods (inductive and convective), firstly, the drying of wet granulates was realized with convective energy input. The conducted experiments were then repeated whereby the wet material was dried by an inductive heating of co-fluidized IHB. During the trials the main process parameters such as gas temperature and moistures were monitored electronically at one-second intervals and recorded. The experiments were realized with a total bed mass of 3.0 kg, consisting of 0.9 kg IHB and 2.1 kg wet material in all cases. The bed material was fluidized at threefold and fourfold minimum fluidization velocity of IHB (3vmf ≈ 3.3 m/s, 4vmf ≈ 4.4 m/s). By both types of heating the initial temperature in the fluidized bed chamber was set to values of 60 °C and 80 °C (prior to addition of the wet material, measured at the outlet of the fluidized bed chamber). The material to be dried was moistened to values from 24 to 34 mass-% moisture. A tabular presentation of the experiments reported in this paper is given in the Table 1. In order to ensure better comparability, all drying tests were carried out following the same procedure. First, the IHB were fluidized at constant gas velocity. Then the convective or inductive bed heating was turned on. By the inductive heating over 70% less time is required to
In this section, the results of the drying experiments are presented and discussed. Here the temperature profiles of the outlet gas temperature are shown for both types of heating, both test materials and variation of the air velocity and air temperature. In addition, the moisture content of the product as a function of experimental time and the drying rate as a function of product moisture content are compared. Table 1 gives an overview of the conducted experiments. In the first 4 trials, the drying of a capillary porous (γ-Al2O3) and of a non-porous bulk material (α-Al2O3) was examined, whereby the type of the heating of the fluidized bed was varied. In experiments 5 to 12, γ-Al2O3 was selected as test material and the gas velocity and gas temperature were varied. 4.1. Drying of porous and non-porous material Fig. 5 shows the curves for the gas outlet temperature over experimental time for drying of wet γ-Al2O3 and wet α-Al2O3 granules. The cylindrical fluidized bed was heated by convection via the fluidizing gas or by inductive heating of electrically conductive IHB. The gas outlet temperature was about 60 °C prior to the addition of the wet material in both energy input methods. In both types of heating, electric powers was set at the level securing comparable gas outlet temperatures in the initial state. The efficiencies of the heating types were taken into account. After adding the wet granulates to the bed the gas outlet temperature dropped immediately. In the four trials, two drying stages can be well differentiated from each other. The gas outlet temperature in the
Table 1 Experimental parameters. Trial number
Bed material
1 2 3 4 5 6 7 8 9 10 11 12
IHB + γ-Al2O3
d32,
d32, Gran.
Energy input
ϕGran. vgas
IHB
mm
mm
–
%
m/s
6.75
1.80
Conv. Ind. Conv. Ind. Conv. Ind. Conv. Ind. Conv. Ind. Conv. Ind.
34.40 34.00 29.58 29.57 23.99 23.71 23.78 24.09 23.99 23.71 23.72 22.90
3.30 63.00 20.70 63.70 20.10 3.24 63.90 20.15 4.32 63.70 19.95 3.24 64.10 20.55 85.00 20.22
IHB + α-Al2O3
1.46
IHB + γ-Al2O3
1.80
ϑgas, inlet
Fig. 5. Gas outlet temperature profile during drying of γ-Al2O3 and α-Al2O3 particles by convective and inductive heating of fluidized bed.
30
V.V. Idakiev et al. / Powder Technology 306 (2017) 26–33
Fig. 6. Comparison of the measured product moisture content in drying of γ-Al2O3 particles by convective and inductive heating of fluidized bed.
Fig. 8. Drying rate during drying of γ-Al2O3 and α-Al2O3 particles by convective and inductive heating of fluidized bed.
first drying stage by the convective energy input was about 30 °C. By the inductive heating the first drying stage took place at a gas outlet temperature of about 32 °C. After addition of the wet granulates in case of inductive heating, the bed height was influenced, and thereby a better fluidization or better distribution of the IHB over the whole height of the induction coil was achieved. The reason for the temperature difference between the both heating types in the first drying stage is then attributed to the better use of the IHB surface area in an alternating electromagnetic field. The same tendencies of temperature profiles can be observed also during the drying of non-porous α-Al2O3 particles. The energy fed in was transferred to the fluidization medium, which increases the gas outlet temperature leading to dry the wet material. On the other hand, the energy was also transferred partly by the direct contact of the IHB with the wet granulates. Through this effect, the drying process can be intensified at the same time. After evaporation of surface moisture, the first drying stage moves into the second drying stage. After the 40 min of the heating phase, the heating was turned off, whereby by the inductive energy input the gas outlet temperature decreased immediately. This can be justified by the direct use of cold air into the fluidized bed chamber whereas in convective heating the heated pipes provide further heat for a certain time. In the case of drying of wet non-porous α-Al2O3 granulates, the first drying stage lasts about 400 s in both types of heating. Due to the use of a non-porous material, the duration of the second drying stages is significantly shorter. Afterwards the gas outlet temperature increased rapidly. Reaching ambient temperature in the fluidized bed chamber after
turning off the relevant heating took less than 1 min by inductive heating and more than 25 min by convective heating. The slight fluctuations (every 2 min) in the curves in Fig. 5 are due to sampling. The samples were taken to measure the moisture of the dried material. A decisive feature of the drying experiments was the evaporated amount of water and the final moisture content of the product. In Figs. 6 and 7 the measured moisture content of both experimental materials are shown for both types of heating. Both figures present the change in the moisture content with the experimental time and the evaporated mass of water roughly calculated from the gas outlet temperatures. Figs. 6 and 7 show clearly similar trends and similar moisture contents under the same drying conditions with respect to the energy input method. At the end of the drying process, the water content in dry material obtained by both types of heating is approaching zero. In Fig. 8, the drying rates related to the particle surface are shown as a function of the product moisture content of both bulk materials attained by both types of heating. The drying rate has its maximum value in the first drying stage. After the transition from the first drying stage to the second drying stage, the drying rate decreases rapidly for both materials. The slight fluctuations in the curves of the drying rates in Fig. 8 show the periodic sampling. The strong fluctuations in the drying rate at the beginning of drying experiments of α-Al2O3 particles can be accounted for by a non-optimal addition of the wet granulates. At the end of the
Fig. 7. Comparison of the measured product moisture content in drying of α-Al2O3 particles by convective and inductive heating of fluidized bed.
Fig. 9. Gas outlet temperature profile during drying of γ-Al2O3 particles by convective and inductive heating of fluidized bed and by variation of the air velocity.
V.V. Idakiev et al. / Powder Technology 306 (2017) 26–33
Fig. 10. Comparison of the measured product moisture content in drying of γ-Al2O3 particles by convective and inductive heating of fluidized bed and by variation of the air velocity (Exp. 5 and 6).
drying process, the drying rate by both energy input methods is approaching zero. Figs. 6, 7 and 8 show that the type of the energy input to the cylindrical fluidized bed has no significant impact on the moisture content and the drying rate. 4.2. Variation of the gas velocity In order to investigate the influence of the gas velocity on the drying process, four experiments were carried out with convective and inductive heating under variation of the gas velocity. The experiments were realized by a threefold or fourfold minimum fluidization velocity of IHB (about 3.24 and 4.32 m/s). During these experiments, the other process parameters (such as bed mass and gas temperature, and initial moisture content of the granulates) were kept constant. The temperature profiles of the drying trials with the variation of the gas velocity are presented in Fig. 9. With the increase in gas velocity it is to be observed that the first drying stage is shorter in the both convective and inductive heating. With the increase in gas velocity a more intensive fluidization and a more intensive heat transfer were observed, which justify the intensified drying of wet granulates. The slight fluctuations in the curves of the gas outlet temperatures in Fig. 9 show the periodic sampling.
Fig. 11. Comparison of the measured product moisture content in drying of γ-Al2O3 particles by convective and inductive heating of fluidized bed and by variation of the air velocity (Exp. 7 and 8).
31
Fig. 12. Drying rate during drying of γ-Al2O3 particles by convective and inductive heating of fluidized bed and by variation of the air velocity.
In Figs. 10 and 11 the measured moisture content is shown for both types of heating and for varied air velocity. Both figures present the change in moisture content with experimental time and the evaporated water mass roughly calculated from the gas outlet temperatures. By comparison of Figs. 10 and 11, it can be seen that the curves describing the evaporation of water are steeper, which corresponds to an intensified evaporation of water with increasing the gas velocity. The course of the drying rates in the experiments with a variation of the gas velocity and the type of heating of fluidized bed is shown graphically in Fig. 12. 4.3. Variation of the initial gas temperature The influence of initial gas temperature on the drying process was investigated by means of four experiments with variation of the type of heating of fluidized bed. Fig. 13 shows the gas outlet temperature during the drying process. The fluidized bed was initially heated to a target temperature of 60 or 80 °C by induction or convection. After adding the wet granulates to the bed the gas outlet temperature dropped immediately. The first drying stage lasts about 300 s for both initial gas temperatures. In both trials, the two drying stages can be well differentiated from each other. After the 40 min of the inductive or convective heating phase, the heating was turned off, whereby in the inductive
Fig. 13. Gas outlet temperature profile during drying of γ-Al2O3 particles by convective and inductive heating of fluidized bed and by variation of initial gas temperature.
32
V.V. Idakiev et al. / Powder Technology 306 (2017) 26–33
Fig. 14. Comparison of the measured product moisture content in drying of γ-Al2O3 particles by convective and inductive heating of fluidized bed and by variation of the initial gas temperature (Exp. 9 and 10).
energy input the gas outlet temperature again decreased immediately (see the grey lines in Fig. 13). In Figs. 14 and 15, the measured moisture content is shown by variation of the air velocity in both types of heating. These two figures show similar trends and similar moisture contents under the same drying conditions in both energy input methods in fluidized beds. When comparing the calculated mass of water evaporated from the wet granulates in the experiments with varying the gas temperature, it can be clearly seen that the deviation between the measured mass of evaporated water and product moisture content is smaller in Fig. 15. The reason for this is the intensified drying of the bulk material at a higher temperature (see the discussion in Section 4.1). Fig. 16 shows the drying rate in experiments with variation of the initial gas temperature and the type of heating of fluidized bed. Here it can be seen that a higher initial gas temperature leads to a slight acceleration of the drying process. This graph demonstrates once again that the energy input method in fluidized bed has no significant impact on the drying process with respect to the product quality (product moisture content).
Fig. 16. Drying rate during drying of γ-Al2O3 particles by convective and inductive heating of fluidized bed and by variation of the initial gas temperature.
moisture content were presented. Within the scope of the performed drying experiments it was demonstrated that the induction technology enables massively reduced heating and cooling times and therefore considerably more controllable temperature profiles. This allows to shorten the costly and time consuming start-up processes in batchwise fluidized bed processing. Using IHB, very large surface area for heat transfer is available. Cofluidization of electrically conductive IHB with the product to be treated, provides movable heat transfer area. Particle-particle collisions cater for self-cleaning of IHB. Furthermore, use of IHB combines advantages of conventional heating methods and inductive heating. Considering these benefits, induction heating is a promising method for process intensification. Conducted drying experiments show very similar drying rates and product moisture contents under the same drying conditions in the both energy input methods in fluidized beds. The present study confirms the great potential of inductive energy input in fluidized beds and thus offers new possibilities for product treatment under a precise temperature control and a constant product quality.
5. Summary Latin letters In this research work a new method for energy input in a cylindrical fluidized bed for drying particulate solids was tested. Effects of different types of heating and test materials on the drying rate and the product
Symbol
Unit
Meaning
d32 v
[mm] [m/s]
Sauter mean diameter Velocity
Greek letters
Symbol
Unit
Meaning
ϕ ϑ
[%] [°C]
Moisture Temperature
Indexes
Fig. 15. Comparison of the measured product moisture content in drying of γ-Al2O3 particles by convective and inductive heating of fluidized bed and by variation of the initial gas temperature (Exp. 11 and 12).
Symbol
Meaning
gas IHB Gran. mf w
Gas Iron hollow balls Granulate Minimum fluidization Water
V.V. Idakiev et al. / Powder Technology 306 (2017) 26–33
Abbreviations
Symbol
Meaning
Conv. DM Exp. IHB Ind. m RH t W
Convective Dry matter Experiment Iron hollow balls Inductive Mass Relative humidity Temperature Water
Acknowledgments The authors gratefully acknowledge the funding of this work by the German Federal Ministry of Science and Education (BMBF) as part of the InnoProfile-Transfer project NaWiTec (grant numbers: 03IPT701A). References [1] H. Uhlemann, L. Mörl, Wirbelschichtsprühgranulation, Springer-Verlag, Berlin, Heidelberg, New York, 2000. [2] L. Mörl, S. Heinrich, M. Peglow, Fluidized bed spray granulation, in: A.D. Salman, M.J. Hounslow, J.P.K. Seville (Eds.), Handbook of Powder Technology, 11, Elsevier, Amsterdam, 2007. [3] V.V. Idakiev, S. Marx, A. Roßau, A. Bück, E. Tsotsas, L. Mörl, Inductive heating of fluidized beds: influence on fluidization behavior, Powder Technol. 286 (2015) 90–97. [4] O. Krischer, W. Kast, Die wissenschaftlichen Grundlagen der Trocknungstechnik, Springer-Verlag, Berlin/Heidelberg/New York, 1978.
33
[5] H. Hayashi, Drying technologies of foods - their history and future, Dry. Technol. 7 (2) (1989) 315–369. [6] M.R. Okos, Food dehydration, in: D.R. Heldmann (Ed.), Handbook for Food Engineering, 2 ed.Marcel Dekker, New York 1992, pp. 601–740. [7] C.G.J. Baker, Industrial drying of foods, 309, Blackie Academic & Professional, London, UK, 1997. [8] C. Ratti, Hot air and freeze-drying of high-value foods- a review, J. Food Eng. 49 (2001) 311–319. [9] P. Jones, Electromagnetic wave energy in drying processes, in: A.S. Mujumdar (Ed.), Drying’92, Elsevier Science, Amsterdam 1992, pp. 114–136. [10] X.C. Jia, Study of heat pump assisted microwave drying, Dry. Technol. 11 (7) (1993) 1583–1616. [11] K.S. Jayaraman, D.K. Das Gupta, Drying of fruits and vegetables, in: A.S. Mujumdar (Ed.), Developments in Drying: Food Dehydration, Kasetsart University Press, Bangkok 1995, pp. 179–206. [12] E. Tsotsas, A.S. Mujumdar, Modern drying technology, Volume 5: Process Intensification, WILEY-VCH Verlag GmbH & Co. KGaA, 2014. [13] A. Stresing, L. Mörl, J. Neum, M. Jacob, K. Walther, Non-Contact Energy Transfer to a Fluidized Bed, in Proceedings of European Drying Conference, Palma de Mallorca, Spain, 2011. [14] A. Stresing, L. Mörl, A. Khaidurova, M. Jacob, K. Walther, Investigation of the time response of a fluidized bed with inductive heating and the influencing variables, Chem. Ing. Tech. 85 (3) (2013) 308–312. [15] V.V. Idakiev, A. Bück, E. Tsotsas, L. Mörl, Inductive energy input in fluidized beds, Proceedings of Conference Machines, Technologies, Materials, Varna. Bulgaria 2015, pp. 29–32. [16] A. Roßau, Induktiver Energieeintrag in eine fluidisierte Schüttung(Dissertation) Otto von Guericke University Magdeburg, 2013. [17] V.V. Idakiev, A. Bück, E. Tsotsas, L. Mörl, Model-based studies on the heat transfer in an inductively heated fluidized bed, Chem. Ing. Tech. 88 (5) (2016) 656–665. [18] M. Latifi, J. Chaouki, A novel induction heating fluidized bed reactor: its design and applications in high temperature screening tests with solid feedstocks and prediction of defluidization state, AICHE J. 61 (5) (2015) 1507–1523. [19] hollomet®, hollomet Brochure, 2014.