Acta Polytechnica doi:10.14311/AP.2014.54.0231 Acta Polytechnica 54(3):231–239, 2014 © Czech Technical University in Prague, 2014 available online at http://ojs.cvut.cz/ojs/index.php/ap STIRRED FLUIDIZED-BED DRYER OF REGENERATED ION EXCHANGER PARTICLES Michal Pěnička, Pavel Hoffman∗, Ivan Fořt Czech Technical University in Prague, Department of Process Engineering, Technická 4, 166 07 Prague, Czech Republic ∗ corresponding author: pavel.hoffman@fs.cvut.cz Abstract. This paper describes the intensification of the fluidized-bed drying process for regenerated spherical-shape ion exchanger particles in batch mode, achieved by a mechanical stirrer in the fluidized bed layer of dried particles. The effect of the mechanical stirring system on the drying process is examined. The calculations and also the results of comparative measurements provide evidence of a favourable effect of stirring on the total drying time in comparison with the initial unstirred system. The regenerated ion exchanger particles pass to the fluid state in a shorter time and the ultimate total drying time is thus more than 60 % shorter. Keywords: fluidized-bed drying; stirring; mechanical disruption; drying kinetics; ion exchanger. 1. Introduction This study seeks savings in fluidized-bed drying of par- ticles in batch mode, the particles being highly adhesive due to the surface tension of the liquid adhering to the particles. Drying is an extremely energy-demanding process, and so it is important – especially now that energy prices are becoming so high – to find ways to achieve energy savings. Fluidized-bed drying is a dry- ing process in which intensive heat and mass transfer occur between the particles that are present in the fluid state and the air flowing through the bed. The surface tension of the liquid adhering to the surface of the particles at the beginning of the drying process gives rise to a strong tendency for the particles to stick to each other and also to the walls of the drying cham- ber. The batch mode of this process together with the insufficient amount of material present in a batch does not allow a vibro-fluid dryer to be used, since these dryers are mainly designed for continuous mode. A solution may be found in drying with a fluidized bed layer which is stirred [1–4] and where the particle clusters are disintegrated continuously and the particles adhering to the drying chamber walls are swept off due to the stirring process. The aim of this study is to intensify and improve the drying process of the regenerated ion exchanger particles, i.e. to shorten the process while at the same time improving the homogeneity of the humidity of the particles. 2. Material and methods 2.1. Spherical particles of the regenerated ion exchanger Ion exchanger particles are used in membrane technol- ogy processes for wastewater treatment purposes [5]. They are also widely used for water treatment in the nuclear power sector [6]. Ion exchangers are mostly synthetic high-molecular- weight organic compounds, largely based on styrene, polyacrylate, phenol formaldehyde resins, etc. [7]. The Marathon-A cation exchanger, consisting of spherical particles 450–580 µm in diameter, was selected as the model material [8]. The maximum permitted tem- perature is 120°C to avoid thermal stress resulting in structural degradation of the particles. The drying process consists of three main periods, see Fig. 1. In the initial phase (period 0-1 in Fig. 1), the particles act as a stationary porous layer. This period is characterized by particle surface moisture, resulting in a strong tendency toward adhesion. In their appearance and in their physical properties, the dried particles resemble an amorphous material (Fig. 2, point 0) and stick strongly to the walls and also to each other. The initial moisture of the regenerated ion exchanger is approximately 65 % RM (Fig. 1). In this period the heat supplied by the drying air is used to evaporate the free surface water from the particles (Fig. 2, point 0). Thus the particle diameter remains constant in this period, as is clear from the particle diameter change curve during the drying process (see Fig. 1B). Due to imperfect circumfluence of the drying air around the particles (it flows through the layer only in several channels) the heat energy is not consumed completely, and thus both the drying air temperature and the humidity at the fluid chamber output are not close to the ideal state, i.e. the drying air humidity approaches the saturation limit and the temperature approaches the wet or moist bulb temperature (Fig. 1D, E). As soon as the attractive forces from the water sur- face tension decrease enough to enable transition of the particles to the fluidized state (period 1–2 in Fig. 1), moisture starts to evaporate from the particles (Fig. 2, point 1). This transition to the fluidized state im- proves the drying air circumfluence of the particles and 231 http://dx.doi.org/10.14311/AP.2014.54.0231 http://ojs.cvut.cz/ojs/index.php/ap M. Pěnička, P. Hoffman, I. Fořt Acta Polytechnica Figure 1. Time dependences of the particle and drying air properties during the whole drying process. consequently accelerates the heat and also the mass transfer. As a result, the drying air temperature at the fluid chamber output starts to approach the wet bulb temperature, and the maximum possible humidity of the drying air is reached (Fig. 1C, D). The dried particles now get to what is referred to as the first dry- ing period, in which the drying rate is constant. This material is characterized by a significant expansion in volume, depending on the moisture content of the material (Fig. 2, point 2). For this reason, the particle diameters are decreased in this period. The rate of moisture evaporation from the particle surface is lower than the rate of moisture transfer from the centre of the particle to the surface. Thus the limiting factor of the drying process is the evaporation from the particle surface. After passing point 2 in Fig. 1, the diameter of the particles remains constant and the particles start to be dried to the required final moisture level. Through this Figure 2. Particle diameter change curve during the drying process. effect, the drying process reaches the second drying period, in which the drying rate decreases. The drying air temperature at the fluid chamber output increases and the humidity decreases. Moisture transfer from the particle centre to the surface is the limiting factor in this period. 2.2. Stirring unit The main issue in this process is that the particles stick to each other or to the dryer walls in the first drying period. This is caused by the surface tension of the liquid, which is present both on the surface and inside the particles. Hence, the properties of the stirrer are an important factor in the technology and in intensifying the process. A suitable stirrer should have the capacity to dis- rupt the ion exchanger particle clusters and sweep the particles off the walls. This ability should be effective enough to minimise the initial drying period in which the particles adhere to each other, but not to cause particle degradation. The effect of the stirrer in the first and second periods, in which the surface of the pre-dried particles is dry enough for the particles to pass to the fluidized state, must not be adverse, i.e. it must not disturb the fluidized bed layer or the contact between the particles and the drying air, which would slow down the heat and mass transfer effects. 2.3. Experimental equipment An experimental equipment assembly for fluidized-bed drying with a stirred layer and various fluid chamber diameters was designed to evaluate the development of the drying process. The layout of the experimental equipment is shown in Fig. 3. Pressurized air at a known temperature and humidity is fed in from the main air distribution line (Fig. 3, item 1). The air flow is controlled by the pressure control valve (Fig. 3, Item 2) and is measured using a rotameter (Fig. 3, item 3). Downstream of the flowmeter the air is heated by a heating device with resistance wires, and the output of the unit is controlled manually by the voltage change using a transformer (Fig. 3, Item 4). The temperature of the heated air is measured upstream of the fluid chamber using a contact thermometer with accuracy of ±0.1°C. The fluidized bed chamber (Fig. 3, item 5) consists of a duct made of a galvanized zinc sheet which is thermally isolated. This main part is 1 m 232 vol. 54 no. 3/2014 Stirred Fluidized-Bed Dryer Figure 3. Experimental equipment layout. long, and then the diameter is expanded to 250 mm. A sensor measuring the humidity of the drying air (accuracy ±0.1 %) and the temperature of the drying air (accuracy ±0.1 %) after passing the fluidized bed layer was positioned always 300 mm above the fluidized- bed chamber grid. This sensor is designed to determine the properties of the drying air at the output (point C in Fig. 4). An adjustable-speed stirrer unit is positioned above the chamber (Fig. 3, item 7). A schematic drawing of the wire stirrer is shown in item 6 in Fig. 3. Fluidized bed chambers with three basic diameters (85 mm, 100 mm and 140 mm) were designed to enable the development of the model drying process to be precisely monitored. 2.4. Properties of the drying air The flow rate of the heated air downstream of the heating unit was determined using the equation of state of an ideal gas (1), assuming that the amount of substance of the flowing air nA at the heating unit input and output is constant, i.e. that there is no leakage of the drying air from the experimental equipment pAV̇A = nARTA. (1) The values from points A and B in Fig. 4 are inserted in the equation of state, including the thermodynamic temperature of the drying air TA, air pressure pA and the universal gas constant R. The superficial drying air velocity uAB of the drying chamber was determined from the air flow rate V̇A at the point B (Fig. 4) uAB = V̇A Sk , (2) where Sk is the cross-section of the fluidized-bed cham- ber. The density of the wet air depending on the temperature was calculated according to equation (9) %A = 1.316 · 10−3 TA (2.65p + ϕAp′′V ), (3) Figure 4. h-X diagram for the adiabatic drying pro- cess. where ϕA is the relative humidity of the drying air and the pressure of the saturated water vapour p′′V at temperature TA was determined according to rela- tion (4), which is valid for the temperature range 0 to 200°C, where C8 to C13 are coefficients taken from the literature [10] ln p′′V = C8 TA +C9+C10TA+C11T 2A+C12T 3 A+C13 ln TA. (4) The dependence of the dynamic viscosity of the drying air on temperature is expressed by equation (11) µA = 0.00004tA + 0.176, (5) where tA is the temperature of the drying air in degrees Celsius. 2.5. Determination of layer porosity Table 1 presents the values of the non-status initial conditions of the studied drying process of particles of the regenerated ion exchanger, identical for all tested sizes of the fluid drying chambers. The Reynolds num- ber was determined from the known drying air velocity uAB according to the equation uAB = ReµA dp%A , (6) where µA is dynamic viscosity and %A is density of the drying air. These values are determined according to equations (3) and (5). Variable dp is the mean value of the particle diameter in the studied period. The relationship between porosity ε and drying air velocity uAB, or between porosity and the Reynolds number Re is expressed [12] by an equation valid for fine particles: Re = Arε4.75 18 + 0.6 √ Arε4.75 . (7) The Archimedes criterion Ar is determined according to the equation Ar = d3p(%SW −%A)%Ag µ2A , (8) 233 M. Pěnička, P. Hoffman, I. Fořt Acta Polytechnica Column diameter Dk m 0.085 0.100 0.140 Temperature of drying air at point B in Fig. 3 tA °C 120 Density of drying air at point B in Fig. 3 %A kg m−3 0.91 Viscosity of drying air at point B in Fig. 3 µA Pa s 2.24 · 10−5 Density of dry particles %s kg m−3 1440 Superficial drying air velocity uAB m s−1 2.1 Table 1. Input parameters of the drying air. where g is the acceleration of gravity and dp is the mean particle diameter in the studied period. %A is the density and µA is the dynamic viscosity of the drying air determined according to equations (3) and (5). The density of the wet particles %SW for the given interval is determined using the equation %SW = %S ( 1 + mW mS ) , (9) where %S represents the density of the dried material and mW is the mass of the liquid component of the corresponding moisture of the measured material at a given point in the studied process. 2.6. Determining the drying rate The drying rate is thus determined for the given interval according to equation [13] − mS AS ∆XW ∆τ = NW , (10) where ∆XW is the difference in absolute moisture between the initial and final point of the studied interval and ∆τ represents the drying period in the given period. The total particle surface was determined according to the formula AS = VpaS, (11) where the total particle volume VP is calculated from the formula Vp = mS %SW , (12) where mS is the mass of the dry matter of the measured sample and %SW is the density of the wet particles for the given interval (9). The specific particle surface is determined according to the formula valid for monodis- perse systems of spherical particles as = 6(1 −ε) dp , (13) where ε is the porosity in the studied period. 3. Results and discussion 3.1. Design of the optimized stirrer for minimizing the total drying time The Wire Stirrer prototype stirrer as shown schemati- cally in Fig. 3, was designed on the basis of the results of preliminary experiments with standardized stirrers. The stirrer is made of stainless steel wire 0.4 to 0.6 mm in diameter. It was designed for a centric arrangement. The diameter of the stirrer is 0.95–0.97dm/Dk and its height is 2.8 to 5 times that of the material layer at rest. The expected rotation speed is approximately 50 rpm for all fluidized bed dryer sizes tested here. The peripheral speed of its blades is in the range of 0.225 to 0.26 ms−1 for the column sizes tested. The preliminary experiments clearly showed that when the stirrer mesh hO is too small the wet ion ex- changer particle clusters are insufficiently disintegrated and the clusters move along with it. If the stirrer mesh is too large, the particle clusters are disintegrated but the resulting clusters are not small enough. There was no significant effect of mesh orientation on the cluster disintegration process. The optimum mesh size was found to be 10 mm×10 mm. 3.2. The course of changes in ion exchanger particle diameters in the test periods During individual experiments, samples of dried ion exchanger were taken every 5 minutes from the same point in the drying column. During the drying process, over 30 samples were taken for the unstirred version and over 13 samples for the stirred version. In a part of this sample, the absolute moisture of the sample was determined using drying balances. In this way, the moisture of the dried particles in a particular point of the drying process was determined. The rest of the sample was immediately subjected to microscopic analysis and photos of at least six particles were taken. The mean diameter of these particles for one particular point in the whole process of drying (and thus for the particular moisture of these particles) was then determined using an optical method. The average mean particle diameter value for all diameters of the columns, for the stirred and unstirred variants, was for the porous layer period dP01 = 590·10−6 m±2 %, for the first period of the fluidized-bed layer dP12 = 530 · 10−6 m ± 3 %, and for the second period of the fluidized-bed layer dP23 = 460·10−6 m±3 %. The dependence of the mean particle diameter value on the absolute moisture of the particle is shown in Fig. 5, where the transition points between individual periods in Fig. 1 are highlighted. 234 vol. 54 no. 3/2014 Stirred Fluidized-Bed Dryer Figure 5. The dependence of mean particle diameter on particle absolute humidity. 3.3. The drying rate in the porous layer period The porous layer period is defined for section 0–1 in Fig. 1. The porosity of the dried particles layer was determined according to the formula for a porous layer [14] ε01 = ε0 + 0.33 dP01 Dk , (14) where ε0, as the probable mean porosity value, is within the interval ε0 = 0.38 to 0.39 [11]. The mean ion exchanger particle diameter in the studied area is dP01 , and Dk is the diameter of the fluid dryer column. If you introduce porosity ε01 and mean particle diameter dP01 into the equation (13), you will obtain the specific surface of particles as01 in this studied period. The total particle volume VP01 is determined using equation (12), where ms is the mass of the dry particles and %SW1 is the density of the wet particles at the final point of the studied area, i.e., at point 1 in Fig. 1 and Fig. 2. This value is determined using equation (9), where %S is the density and ms is the mass of the dry particles. The mass of the moisture or of the dried water in this area is expressed by mW01 . If we introduce the specific surface of particles as01 and the total particle volume VP01 into the equation (11), we will obtain AS01 , i.e., the total surface of the dried particles in this period. The drying rate in the porous layer period NW01 is determined using equation (10), where ms is the mass of the dry particles and AS01 is the total surface of the dried particles. Quantity ∆XW01 is the difference in absolute moisture of the dried particles between the initial point and the final point of the studied period and ∆τ01 is the length of this period. Table 2 summarizes the determined values of characteristic quantities for this period of the drying process of particles of the regenerated ion exchanger for individual diameters of drying columns. 3.4. The drying rate in the first period of fluidized-bed drying In the region of the first period of fluidized-bed drying (period 1–2 in Fig. 1) the transition of the particles to the fluidized state begins and moisture from the particle Figure 6. Resulting drying curves. centre starts to evaporate. This causes a change in the particle diameter, which depends on the moisture of the particle (see Fig. 5). There are important val- ues for determining the properties of the fluidized-bed layer, namely the velocity at the fluidization threshold uA,mf and the porosity ε12 at superficial drying air velocity uAB. The rate at the fluidization threshold is determined using the following formula: uA,mf = ReP,mf µAB dP12%AB , (15) where µAB is the dynamic viscosity and %AB is the density of the drying air. Variable dP12 represents the mean particle diameter in this studied period. To determine the velocity at the fluidization threshold uA,mf , quantity ReP,mf must be determined in advance [15] according to the formula in the period valid for fine particles ReP,mf = (33.72 + 0.0408Ar )0.5 − 33.7, (16) where the Archimedes number Ar is defined by equation (8). This is determined by the dynamic viscosity µAB, the density of the drying air %AB, the acceleration of gravity g, the mean particle diameter dP12 and the final particle density %SW12 in the calculated period 1–2. This value is determined according to equation (9), where mS is the dry mass of the particles, %S is the density of the dry particles and value mW12 is given by the average mass of the particle moisture for the studied period. The porosity for the first period of fluidized-bed drying ε12 is also determined using formula (7), where Ar is the Archimedes number for this period and Re is the Reynolds number determined at the superficial drying air velocity uAB, the dynamic viscosity µAB and the density %AB of drying air in the period B in Fig. 3. The drying rate for the first period of fluidized-bed drying is determined according to formula (10), where mS represents dry material mass and As12 is the total area of dried particles determined using formulas (11), (12), (13) and (9) for the studied periods. An overview of the values of characteristic quantities for this period 235 M. Pěnička, P. Hoffman, I. Fořt Acta Polytechnica Column diameter Dk m 0.085 0.100 0.140 Stirred version Mean value of particle size dP01 m 5.9 · 10−4 5.9 · 10−4 5.9 · 10−4 Period time ∆τ01 s 1634 1625 1860 Difference in humidity ∆XW01 kgW kg −1 S 0.962 1.125 1.030 Water mass at point 0 mW0 kg 0.227 0.306 0.533 Water mass at point 1 mW1 kg 0.122 0.165 0.287 Water mass at period 0–1 mW01 kg 0.105 0.141 0.246 Density of wet particles at point 1 %s1 kg m−3 2880 2880 2880 Mean value of porosity ε0 – 0.39 0.39 0.39 Porosity ε01 – 0.39 0.39 0.39 Specific surface of particles as1 m−1 6207 6210 6216 Total particle volume Vp1 m3 4.54 · 10−5 5.61 · 10−5 9.79 · 10−5 Total surface of particles As1 m2 0.28 0.35 0.61 Drying rate Nw01 kg m−2 s−1 2.55 · 10−4 3.27 · 10−4 2.61 · 10−4 Unstirred version Mean value of particle size dP01 m 5.9 · 10−4 5.9 · 10−4 5.9 · 10−4 Period time ∆τ01 s 5687 5880 5765 Difference in humidity ∆XW01 kgW kg −1 S 1.035 1.018 1.003 Water mass at point 0 mW0 kg 0.220 0.327 0.571 Water mass at point 1 mW1 kg 0.097 0.144 0.251 Water mass at period 0–1 mW01 kg 0.123 0.183 0.319 Density of wet particles at point 1 %s1 kg m−3 2618 2618 2618 Mean value of porosity ε0 – 0.39 0.39 0.39 Porosity ε01 – 0.39 0.39 0.39 Specific surface of particles as1 m−1 6207 6210 6216 Total particle volume Vp1 m3 3.85 · 10−5 5.87 · 10−5 1.03 · 10−4 Total surface of particles As1 m2 0.24 0.36 0.64 Drying rate Nw01 kg m−2 s−1 9.02 · 10−5 8.35 · 10−5 8.38 · 10−5 Table 2. Overview of values for determination of drying rate in the porous layer period. of the drying process of particles of the regenerated ion exchanger is summarized in Table 3. 3.5. Effect of the stirrer on the total drying time A comparison of the drying curves for the stirred and unstirred fluidized bed layers is shown in Fig. 6. A comparison of these curves shows the effect of the stirrer on the porous layer area, where the drying time is significantly shorter. The rotating stirrer also helps the particles pass to the fluid state already at 50 % RH instead of the initial 45 % RH. There is also a positive effect of the stirrer on the first period of fluidized-bed drying, because it sweeps the wet par- ticles stuck on the walls back into the fluidized-bed layer, where the transfer phenomena are more inten- sive. 4. Conclusions It has been demonstrated that the stirrer has a positive effect on the porous layer period and during the first period of the drying process in the fluidized bed layer. The stirrer regularly disturbs the stationary porous layer that has formed, and this intensifies the rate of the transport phenomena. The total drying time to reach the required moisture content of the material is thus 60 minutes, which corresponds to a 63 % shorter drying time. 236 vol. 54 no. 3/2014 Stirred Fluidized-Bed Dryer Column diameter Dk m 0.085 0.100 0.140 Stirred version Mean value of particle size dP12 m 5.30 · 10−4 5.30 · 10−4 5.30 · 10−4 Period time ∆τ12 s 1434 1262 1587 Difference in humidity ∆XW12 kgW kg −1 S 0.588 0.588 0.588 Water mass at point 1 mW1 kg 0.122 0.165 0.287 Water mass at point 2 mW2 kg 0.022 0.029 0.051 Water mass at period 1–2 mW12 kg 0.100 0.136 0.236 Density of wet particles at point 2 %SW2 kg m−3 2290 2070 1802 Archimedes number Ar – 6046 5465 4756 Reynolds number of fluidization Rep,mf – 3.480 3.160 2.766 threshold Rate at fluidization threshold uA,mf m s−1 0.16 0.15 0.13 Porosity at fluidization start εmf – 0.40 0.40 0.40 Reynolds number at uAB Re – 43 47 44 Porosity at working conditions ε12 – 0.79 0.82 0.83 Specific surface of particles as12 m−1 2429 2016 1930 Total particle volume Vp12 m3 8.47 · 10−5 1.14 · 10−4 1.99 · 10−4 Total surface of particles As12 m2 0.21 0.23 0.38 Drying rate Nw12 kg m−2 s−1 2.43 · 10−4 3.33 · 10−4 2.76 · 10−4 Unstirred version Mean value of particle size dP12 m 5.30 · 10−4 5.30 · 10−4 5.30 · 10−4 Period time ∆τ12 s 1406 1411 1691 Difference in humidity ∆XW12 kgW kg −1 S 0.497 0.497 0.497 Water mass at point 1 mW1 kg 0.097 0.144 0.251 Water mass at point 2 mW2 kg 0.021 0.031 0.054 Water mass at period 1-2 mW12 kg 0.076 0.113 0.197 Density of wet particles at point 2 %SW2 kgm−3 2315 2030 1778 Archimedes number Ar – 6114 5359 4693 Reynolds number of fluidization Rep,mf – 3.517 3.101 2.730 threshold Rate at fluidization threshold uA,mf m s−1 0.16 0.14 0.13 Porosity at fluidization start εmf – 0.40 0.40 0.40 Reynolds number at uAB Re – 43 47 44 Porosity at working conditions ε12 – 0.78 0.83 0.83 Specific surface of particles as12 m−1 2450 1977 1904 Total particle volume Vp12 m3 8.22 · 10−5 1.22 · 10−4 2.13 · 10−4 Total surface of particles As12 m2 0.20 0.24 0.41 Drying rate Nw12 kg m−2 s−1 2.08 · 10−4 2.57 · 10−4 2.22 · 10−4 Table 3. Overview of values for determination of drying rate in the first period of fluidized-bed drying 237 M. Pěnička, P. Hoffman, I. Fořt Acta Polytechnica It follows from the experiments and the results that the drying rate in the period of the porous stationary layer is 8.58 · 10−5 kg m−2 s−1 ± 5 %, and after introducing the stirrer the rate increases to 2.81 · 10−4 kg m−2 s−1 ± 16 %, which represents an in- crease of 220 %. In the first and second periods of fluidized-bed drying, the stirrer does not disturb the fluidized bed layer that has formed. Then intensification of the process helps, and the stuck wet particles are swept off the walls of the drying chamber into the fluidized bed. This results in an increase in the drying rate in the first period of fluidized-bed drying from 2.29 ·10−4 kg m−2 s−1 ±12 % to 2.84 · 10−4 kg m−2 s−1 ± 17 %, i.e. an increase of 17%. The results presented here are based on more than one hundred long-term drying experiments resulting in figures that provide a description of the course of investigated drying process. Acknowledgements The authors appreciate the support provided by internal grant SGS12/057/OHK2/1T/12 of the Czech Technical Uni- versity in Prague, grant INGO LG 13036 of the Czech Min- istry of Education, and research project J04/98: 212200008 of the Czech Ministry of Education. List of symbols Dk Column diameter [m] Ar Archimedes number [–] as Specific surface of particles [m−1] AS Total surface of particles [m2] C8–C11 Calculation constants [–] d Diameter [m] g Acceleration of gravity [m s−2] h Height [m] m Mass [kg] n Amount of substance [mol] Nw Drying rate [kg m−2 s−1] p Pressure [Pa] p′′ Pressure of saturated water vapours [Pa] R Universal gas constant [J K−1mol−1] Re Reynolds number [–] S Cross-section [m2] t Temperature [°C] T Thermodynamic temperature [K] u Velocity [m s−1] V Total particle volume [m3] V̄ Volume flow [m3 s−1] X Absolute humidity (moisture) [kgW kg−1S ] ε Porosity [–] ϕ Relative humidity (moisture) [% RH (RM)] µ Viscosity [Pa s] % Density [kg m−3] τ Time period [s] List of subscripts A Air W Water (humidity, moisture) SW Wet material S Dry matter P Particle m Stirrer K Column 0 Initial point 1 Transition point between the periods of the porous layer and the first period of fluidized-bed drying 2 Transition point between the periods of the first and second period of fluidized-bed drying A Zone of cold air B Zone of heated air C Zone of chilled wet air WALL Wall References [1] Kim J, Han G.Y.: Effect of agitation on fluidation characteristics of fine particles in a fluidized bed. Powder Technol. 2006; 166:113-122 doi: 10.1016/j.powtec.2006.06.001 [2] Daud W.R.W.: Fluidized bed dryers - recent advances. Adv. Powder Technol. 2008; 19:403-418 doi: 10.1163/156855208X336675 [3] Ying Han, Jia-Jun Way, Xue-Ping Gu, Lian-Fang Feng, Guo-Hua Hu.: Homogeneous Fluidization of Geldart D particles in a Gas-Solid Fluidized Bed with a Frame Impeller. Industrial & Engineering Chemistry Research. 2012; 51(50) 16482-16487 doi: 10.1021/ie301574q [4] Ying Han, Jia-Jun Way, Xue-Ping G.U., Lian-Fang Feng, Guo-Hua Hu.: Effect of Agitation on the Fluidization Behavior of a Gas-Solid Fluidized Bed with a Frame Impeller. AIChE Journal. 2013;59: 1066-1074. doi: 10.1002/aic.13893 [5] Mega.cz, RALEX heterogenous ionex membranes (in Czech). QARTIN s.r.o., [Online]. Available: http://www.mega.cz/ heterogenni-iontomenicove-membrany-ralex.html [Approach obtained 20 11 2012]. [6] M. J. Hudson, P. A. Williams.: Recent Developments in Ion Exchangers, Ion Exchangers in the Nuclear Industry. Springer Netherlands. 1990. [7] L. Jelínek.: Desalination and separation methods in water treatment (in Czech). Prague Institute of Chemical Technology. 2009. [8] Dow, DoweX marathon A, 26 11 2010. [Online]. Available: http://www.dowwaterandprocess.com/products/ix/ dx_mar_a.htm [Approach obtained 22 05 2011]. [9] J. Chyský.: Moist air (in Czech). Publishing House of the Czech Technical University in Prague. 1977. [10] J. Šesták, J. Bukovský and M. Houška.: Thermal Processes: Transport and thermodynamic data (in Czech). Publishing House of the Czech Technical University in Prague. 1993. [11] S. H. Touloukian.: Thermophysical properties of matter, sv. 11. IFI/Plenum Press, 1975. 238 http://dx.doi.org/10.1016/j.powtec.2006.06.001 http://dx.doi.org/10.1163/156855208X336675 http://dx.doi.org/10.1021/ie301574q http://dx.doi.org/10.1002/aic.13893 http://www.mega.cz/heterogenni-iontomenicove-membrany-ralex.html http://www.mega.cz/heterogenni-iontomenicove-membrany-ralex.html http://www.dowwaterandprocess.com/products/ix/dx_mar_a.htm http://www.dowwaterandprocess.com/products/ix/dx_mar_a.htm vol. 54 no. 3/2014 Stirred Fluidized-Bed Dryer [12] R. T. Goroško. Izv. Vuzov. Neft i gaz.: 1958;1:125. [13] A. S. Mujumdar.: Handbook of Industrial Drying. Dekker. 1995. [14] V. Hlavačka.: Thermal processes in technical systems of gas-solid particles (in Czech). Publishing House of Technical Literature. 1980. [15] C. Wen and Y. Yu.: A generalized method for predicting the minimum fluidization velocity. AIChE Journal. 1966; 12:610-612. 239 Acta Polytechnica 54(3):231–239, 2014 1 Introduction 2 Material and methods 2.1 Spherical particles of the regenerated ion exchanger 2.2 Stirring unit 2.3 Experimental equipment 2.4 Properties of the drying air 2.5 Determination of layer porosity 2.6 Determining the drying rate 3 Results and discussion 3.1 Design of the optimized stirrer for minimizing the total drying time 3.2 The course of changes in ion exchanger particle diameters in the test periods 3.3 The drying rate in the porous layer period 3.4 The drying rate in the first period of fluidized-bed drying 3.5 Effect of the stirrer on the total drying time 4 Conclusions Acknowledgements List of symbols List of symbols References