Study a new atmospheric freeze drying system incorporating a vortex tube and multi mode heat input 5

Chapter-5 Experimental Results and Discussion CHAPTER EXPERIMENTAL RESULTS AND DISCUSSION This chapter starts with an investigation of the subzero air temperature distribution produced by the vortex tube. The AF drying kinetics along with quality parameters of the model products tested (color, rehydration properties, scanning electron micrographs (SEM) etc) under fixed bed and multimode heat inputs are presented. Subsequently, effect of osmotical pretreatment on AFD was examined. Also, investigation of a vibro-fluidized bed dryer in AFD with addition of an adsorbent to the model product is presented and discussed. Finally a comparison between vacuum freeze drying (VFD), atmospheric freeze drying (AFD) and heat pump drying (HPD) is made in terms of both drying kinetics and quality aspects. 5.1 Fixed Bed AFD with Multimode Heat Input 5.1.1 Temperature distribution in the AFD chamber The measured temperature distribution generated by the vortex tube as well as inside the drying chamber at inlet air pressure of and bar pressure (absolute) is shown in Figure 5.1. The air temperature inside the drying chamber drop from the ambient temperature to about -10°C and –3°C within 15 minutes of start-up of the experiment, at operating compressed air pressures of bar and bar absolute, respectively. The rate of decrease of the chamber temperature becomes slower with time and stabilizes at about -19°C and –6°C, respectively. The air temperature, immediately after the vortex tube, decreases rapidly and remains constant at -26°C and -17°C, respectively, after only about minutes. Tangential injection of compressed air at room temperature into the vortex tube at high velocity produces a vortex, which spins annularly along the 58 Chapter-5 Experimental Results and Discussion tube inner wall as it moves axially down the tube. A part of this air is adiabatically expanded inward to the centre, according to the explanation of the flow in a vortex tube (Crocker 2003). Figure 5.1 Temperature distributions inside drying chamber at constant pressure at the inlet of the vortex tube The decrease in pressure during expansion causes a drop in temperature, which provides a cooler central column of air directed out of one end of the tube. Following the vortex tube, the air passes through a muffler to reduce the noise level and expands suddenly into the well-insulated drying chamber. As a result, the temperature of the air rises somewhat inside the drying chamber. Figure 5.2 shows the effect of inlet air pressure on the carrier gas temperature inside the drying chamber as well as the vortex 59 Chapter-5 Experimental Results and Discussion 20 Air temperature after vortex tube at variable pressure: bar (0 to 14 min) and bar (after 14 min) 10 Air temperature inside drying chamber at variable pressure: bar(0 to 14 min) and bar (after 14 min) o Temperature, C 30 -10 -20 -30 -40 10 15 Time, 20 25 30 Figure 5.2 Temperature distributions inside drying chamber at variable pressure at the inlet of the vortex tube 40 T1- Air temperature after vortex tube Radiant Heater 30 T2 o Temperature, C 20 T1 T2 T3 T4 T3 T5 10 Conduction Heater T4 T5 -10 -20 -30 -40 20 40 60 80 100 120 Time, Figure 5.3 Temperature distributions inside drying chamber with time at bar absolute pressure 60 Chapter-5 Experimental Results and Discussion tube exit. Initially the pressure was set at bars absolute and the corresponding temperature after the vortex tube and also inside the drying chamber was found to reach about -30oC and -17oC, respectively. Inlet pressure of bars also was set at elapsed time of 14min, which results in an instantaneous change in the air temperature of the cold stream to about –16oC, while the chamber air temperature reached about -11°C. Figure 5.3 shows the temperature variation at different locations inside the drying chamber with time at a fixed operating pressure of bar also. It can be seen from their figure that all points inside the chamber show similar pattern of temperature distribution. After 40 minutes, temperature at all locations approaches asymptotic values between -16°C to -18°C. This indicates a relatively uniform temperature distribution at different locations within the drying chamber in the absence of heat input. 30 o Drying condition-Single stage process: -6 C Product - Potato (Disc type) 25 . Temperature, o C 20 15 Air temperature inside the drying chamber 10 Product temperature -5 -10 -15 100 200 300 Time, 400 500 Figure 5.4 Variation of product and drying air temperature with time 61 Chapter-5 Experimental Results and Discussion Variation of local temperature distribution of potato sample as well as the carrier gas in the drying chamber with time under single and two stage heat input schemes is shown in Figures 5.4. and 5.5, respectively. Figure 5.4 shows that for single stage drying at -6oC a sharp increase of frozen product temperature from -20oC to -9oC within about minutes; placed inside the refrigerator to make the product in freezing state prior to start the experiments. It is likely that the air temperature (-6oC) is comparatively higher than that of the frozen product (-25oC). Also the high value the of convective heat transfer co-efficient due to the high velocity of compressed air contributes to heat transfer from the air to the product. On the other hand, air temperature drops down from ambient to operating temperature (-6oC) within the same time interval. 20 Drying condition-Two stage process: 15 o 10 o Temperature, C o -11 C (0 to hr) and -6 C (after hr) Product - Potato (Disc type) Drying air temperature Product temperature -5 -10 -15 -20 -25 100 200 Time, 300 400 Figure 5.5 Variation of product (potato) and drying air temperature with time for the two stage drying process 62 Chapter-5 Experimental Results and Discussion After about 20 minutes both temperatures approach a stable condition and maintain a nearly constant temperature difference of about -3oC between the carrier gas and the product. This is due to sublimation of ice layer from evaporation of the product. During sublimation only latent heat transfer takes place and keeps the temperature nearly constant through out the course drying. Figure 5.5 shows results for the local temperature distribution for the two stage drying process of the product (-14oC) and the carrier gas temperature (-11oC). Unlike the first stage of drying upto hours of drying time at an operating pressure of 4.4 bar absolute without supplying any additional heat input. In the second stage, an increased and stable air temperature was obtained (-6°C) at the same operating pressure by adding radiation and conduction heat sources. The same temperature difference of about 3oC was also observed between the product and air temperature, which are both well below the melting temperature of the samples. 35 Drying condition-Two stage process: o 15 Air temperature inside the drying chamber Product temperature o Temperature, C o -11 C (0 to hr) and -6 C (after hr) Product - Carrot (Disc type) 25 -5 -15 -25 100 200 Time, 300 400 500 Figure 5.6 Variation of product (carrot) and drying air temperature with time for the two stage drying process 63 Chapter-5 Experimental Results and Discussion This ensures frozen integrity of the potato samples in both stages of drying - an essential requirement for sublimation and hence to maintain product quality. Figure 5.6 shows the variation of the measured local temperature of carrot and the carrier gas in the two-stage process. Almost similar nature of the temperature distribution with somewhat higher temperature differences (4oC) was found for carrot samples which again imply the frozen integrity of the product as well as the consistency of the experimental results. 5.1.2 Drying kinetics Plots of the dimensionless moisture content with drying time for potato samples for the four-heat input schemes are shown in Figure 5.7. It is observed that after hours of drying time, the drying rate gradually dropped under the constant heat input scheme at -11oC. It is likely that supply of energy for the sublimation of evaporation front is not 1.0 Po tato A FD-CT : -11C In itial M as s o f all s amp le = 0.3g 0.9 Po tato A FD-CT : -6 C Dimensionless moisture content 0.8 Po tato -A FD-T wo -6 C Po tato -A FD-T wo -6 C (Rad iatio n ) Po tato A FD-T wo -6 C (Rad iatio n & 0.7 0.6 s tag e: -11C & s tag e: -11C & s tag e: -11C & Co n d u ctio n ) 0.5 0.4 0.3 0.2 0.1 0.0 T ime, h r Figure 5.7 Variation of dimensionless moisture content for potato sample with time 64 Chapter-5 Experimental Results and Discussion enough due to the lower intensity drying condition at this stage. Moreover, as drying progresses the evaporation front recedes deeper into the product and the highly porous structure of the dry layer decreases the thermal conductivity of the product. As a result the evaporation front does not receive enough heat to sustain a higher drying rate. 1.0 Single stage: -11C Dimensionless moisture content 0.9 Two stage: -11C & -6 C 0.8 Two stage: -11C & -6 C ( Radiation ) 0.7 Single stage: -6 C 0.6 Two stage: -11C & -6 C (Rad & Cond) 0.5 0.4 0.3 0.2 Initial mass of all carrot sample = 0.3g 0.1 0.0 Time, hr Figure 5.8 Variation of dimensionless moisture content for carrot sample with time The change in gradient of moisture content after hours suggests that the drying air temperature should be elevated at this time to enhance the moisture removal gradient. This phenomenon is used in the two-stage process using multimode heat input; it was shown to yield a significant improvement in the atmospheric freeze drying kinetics. Final dimensionless moisture contents for Case2, Case3 and Case4 were 0.0775, 0.0491, and 0.03, respectively. Case4 showed better drying performance than case2 and case3. This can be explained by the fact that a higher process temperature provides 65 Chapter-5 Experimental Results and Discussion greater energy to the product surface due to increased heat transfer inside the drying chamber. It is worthwhile to note that radiation heats the product superficially without heating the surrounding and penetrates gradually with time into the product (Ratti and Mujumdar, 1995). Conduction heat from the bottom of the product also provides energy for the sublimation front as drying progresses. Note that the increased heat transfer must be achieved without melting of the frozen product. Figure 5.8 shows variation of the dimensionless moisture content with drying time for carrot samples for the various input schemes tested. Similar phenomena were observed for carrot samples as well. Unlike, potato, the two-stage process coupled with conduction and radiation heat input showed the higher drying rates for the carrot samples. Variation of mass flux for potato with dimensionless moisture content is shown for all four cases in Figure 5.9. It can be seen from this figure that the initial drying rate for case-1, case-2, case-3, and case-4 were 0.228 kg/m 2h, 0.149 kg/m 2h, 0.158 kg/m 2h, 0.164 kg/m 2h and 0.176 kg/m 2h, respectively. Results show that, except for the drying condition -6oC; for all other cases the drying rate was nearly same in the first four hours of drying because of identicle operating conditions during this period. The application of higher constant heat input at -6oC increases the drying rate. However, a quality analysis performed in parallel showed that melting occurs due to the higher initial temperature causing some product degradation in terms of the internal structure, which effects rehydration rate when compared with other heat input schemes studied in this work. Figure 5.9 also shows linear behaviour of the mass flux in the falling rate period for all four cases for carrot. No constant drying rate period was observed. A rapid increase in drying rate was observed in all tests of the two-stage process when the dimensionless 66 Chapter-5 Experimental Results and Discussion 0.3 Single stage: -11C Two stage: -11C & -6C Two stage: -11C & -6C ( Radiation ) Single stage: -6C Two stage: -11C & -6C (Rad & Cond) Drying rate, kg/m h 0.2 0.2 0.1 0.1 0.0 0.0 0.2 0.4 0.6 0.8 Dimensionless moisture content 1.0 1.2 Figure 5.9 Variation of drying rate for potato with dimensionless moisture content 0.30 Single stage: -11C Two stage: -11C & -6C Two stage: -11C & -6C (Radiation ) Single stage: -6C Two stage: -11C & -6C (Rad & Cond) Drying rate, kg/m h 0.25 0.20 0.15 0.10 0.05 0.00 0.0 0.2 0.4 0.6 0.8 1.0 Dimensionless moisture content, kg/lkg db 1.2 Figure 5.10 Variation of drying rate for carrot with dimensionless moisture content 67 Chapter-5 Experimental Results and Discussion capacity, which in turn maintains a higher driving force for migration of moisture through the dry layer to the surface of the product and thus increases the drying rate. In this study only a small quantity of frozen product was used for the tests with a fixed adsorbent-product mass ratio. That is why adsorbent even without refreshing also retains its adsorptive capacity over whole drying process and shows good drying performance. However, adsorbent refreshing technique may yet be effective if a lower adsorbent/ drying material ratio are used. However, this procedure is complex probably not useful on larger scale. Similar experiments were carried out at vibration factor of 4.8; the results are shown in Figure 5.42. Almost similar output was obtained but with slower drying kinetics in all cases; this is expected due to reduced mixing at the lower vibrating factor. 1.0 Potato: Without vibration Dimensionless moisture content 0.9 Potato: Vibration without adsorbent 0.8 Potato: Vibration with adsorbent Potato: Vibration with adsorbent refreshing 0.7 0.6 0.5 ` 0.4 0.3 0.2 Vibration factor: 4.8 Adsorbent: Silica gel-D: 3mm Ratio to product: 1:1 0.1 0.0 Time, hr Figure 5.42 Variation of dimensionless moisture content with time for different drying conditions 106 Chapter-5 Experimental Results and Discussion Comparison of the variation of mass flux under different drying conditions, namely csae1- drying without vibration; case2- drying with vibration; case3- vibration with adsorbent and csae4- vibration with adsorbent refreshing, is shown in Figure 5.43. Different trends were observed under different drying conditions. The initial value of the mass flux was 0.17 kg/m2hr, 0.24 kg/m2hr, 0.32 kg/m2hr and 0.36 kg/m2hr for case1, case2, case3 and case4, respectively. Results show that the mass flux is higher for case2 than for case1 due to incorporation of vibration. A significant improvement of mass flux, specially at the initial stage of drying, was found for case3 in comparison with case2. This is attributed to the fact that adsorbent addition plays an important role in increasing the driving potential for vapor transfer from the ice-vapor interface to the 0.40 Potato: Without vibration Potato: Vibration without adsorbent 0.35 Potato: Vibration with adsorbent Potato:Vibration with adsorbent refreshing Drying rate, kg/m h 0.30 0.25 0.20 0.15 0.10 Vibration factor:4.8 (Freq: 20hz, Amp: 3mm) Adsorbent: Silica gel -D: mm Ratio to product: 1:1 0.05 0.00 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Dimensionless moisture content Figure 5.43 Variation of drying rate with dimensionless moisture content for different drying conditions 107 Chapter-5 Experimental Results and Discussion surface of product. It was also observed that adsorbent refreshment at regular intervals (case4) also enhances the mass flux. For case1, the mass flux curve starts with a short falling rate period up to dimensionless moisture content of about 0.8, followed by a rapid increase. However, a constant drying rate curve for case2 and a linear falling rate period was observed for case and case4 until dimensionless moisture content of 0.65. It is interesting to note that for all cases an increase in mass flux was observed in between dimensionless moisture content values of 0.7 to 0.4. This is the results of the high intensity drying condition achieved in the two-stage process. At the end of drying, the mass flux decreases rapidly for case2, csae3 and case4, where the drying rate is mainly controlled by internal transport rate. Moreover, bound moisture is also responsible for the slower drying rate, which needs extra energy to break up the watersolid bond. 1.0 No adsorbent Product: Potato cubes Amp: 3mm Dimensionless moisture content 0.9 0.8 0.7 0.6 0.5 0.4 0.3 20hz, 3mm, (VF 4.8) 22hz, 3mm, (VF 5.8) 17hz, 3mm, (VF 3.5) 0.2 0.1 0.0 Time, hr Figure 5.44 Effect of frequency on dimensionless moisture content with time 108 Chapter-5 Experimental Results and Discussion Figure 5.44 shows the effect of frequency on the variation of dimensionless moisture content with time. Final dimensionless moisture content at frequencies of 17 Hz and 20 Hz is about 0.18 and 0.09, respectively, after eight hours of drying time. It can be seen from this figure that the drying rate increases with increase of frequency. This is due to the fact that at the higher frequency, the product contacts the heated plate more frequently and hence receives more thermal energy by conduction. However, as the frequency increases to 22 Hz, the final dimensionless moisture falls to 0.13. During this experiment, it was observed that the product starts to mix well at 17Hz with vibration amplitude of 3mm. As the frequency increases, the bed of particles vibrates more intensely at 20Hz with higher degree of mixing. Vibration of the bed subsequently decreases as the frequency is increased beyond 20Hz; in fact it was observed to drop significantly beyond 22Hz. Mixing also decreases beyond 22Hz. It is 1.0 1.0 Freq-20; Amp-3mm Freq-20; Amp-3mm 0.9 0.9 Freq-20; Amp-4mm Freq-20; Amp-4mm Dimensionless moisture content Dimensionless moisture content 0.8 0.8 0.7 0.7 0.6 0.6 0.5 0.5 0.4 0.4 0.3 0.3 Two stage: -11C & -6 C ( Rad and Cond) Without adsorbent Two stage:Silica -11C gel-D: & -6 C 4mm ( Rad and Cond) Adsorbent: Adsorbent: Silica gel-D: 4mm 0.2 0.2 0.1 0.1 0.0 0.0 2 3 5 Time, hr Time, hr 6 7 8 9 Figure 5.45 Effect of amplitude of vibration at fixed frequency of 20 Hz 109 Chapter-5 Experimental Results and Discussion Therefore, it is concluded that at 20Hz the tray is vibrated at its resonant frequency, which in turn plays an important role in delivering the maximum amount of heat to the product through conduction. The lower and upper threshold frequency for the start and rapid drop of mixing were found to be 17Hz and 22Hz, respectively. These values are specific to the apparatus used, however, and hence cannot be generalized. 1.0 Without vibration 0.9 Vibration (VF- 6.4) with adsorbent refreshing VFD Dimensionless moisture content 0.8 0.7 Material: mm cubic potato sample W-1.3 gm; MC- 7.47 kg/kgdb. Adsorbent-Silicagel (D-4mm) Weight ratio-1:1 0.6 0.5 0.4 0.3 0.2 0.1 0.0 Time, hr Figure 5.46 Comparison of different methods on dimensionless moisture content with time Effect of vibration amplitude on freeze-drying kinetics at a given frequency was also investigated and the corresponding results are shown in Figure 5.45. Increase of drying rate is seen to be insignificant with increases of amplitude. These phenomena reveal that vibration amplitude has little influence on the migration of moisture to the surface 110 Chapter-5 Experimental Results and Discussion from the interior of the particles for the specific test conditions used. These results agree with the findings of some of the previous work by Dong et al. 1991. Plots of comparison of freeze drying kinetics between a vibro-fluidized bed dryer, a fixed bed dryer and a vacuum freeze dryer to investigate the viability of the proposed novel AFD drying system are shown in Figure 5.46. It was observed that adsorbent refreshment coupled action increased the drying performance significantly in comparison with that of the fixed bed of drying particles for the same material. Final dimensionless moisture content of 0.05 was obtained after and 5.5 hours of drying time for VFD and AFD with vibro-fluidized bed couple with adsorbent refreshment, Dimensionless moisture conte 1.0 0.9 Two stage: -11C & -6 C ( Rad and Cond) Vibration: Freq-20Hz; Amp-3mm Adsorbent: Silica gel-D: 4mm 0.8 Adsorbent without freezing (25 C) o o Adsorbent below freezing temperature (-22 C) 0.7 0.6 Adsorbent without freezing Adsorbent with freezing 0.5 0.4 Drying condition: 0.3 Two stage process (-11oC & -6oC) with vibration. mm cubic potato sample. Initial weight : 1.3 gm. M oisture content: 7.47 kg/kgdb. 0.2 0.1 0.0 Time, hr Figure 5.47 Effect of adsorbents on drying kinetic with time respectively. However, it took hours of drying time to reach the moisture content of about 0.13 for the AFD with fixed bed. Theses results show the significant 111 Chapter-5 Experimental Results and Discussion improvement of using vibro-fluidized bed dryer with adsorbent in AFD reduce drying time of about 2.5 hours compared with fixed bed dryer without adsorbent for the same final moisture content. Effect of adsorbent addition below the freezing temperature and at room temperature on FD kinetics is shown in Figure 5.47. No improvement in drying kinetics was observed using frozen adsorbents. This is probably due to lower thermal conductivity of the adsorbent and also heat capacity. There is therefore no concern about melting of the frozen product by addition of room temperature adsorbent. 5.2.2 Quality analysis Variation of the rehydration ratio with time for several cases involving vibration and addition of adsorbent particles (silica gel) is shown in Figure 5.48. A minor improvement in rehydration quality was observed for vibro-fluidized bed without adsorbent when compared with the fixed bed dryer. However, mixing with adsorbent in a vibro-fluidized bed dryer showed higher rehydration ratio compare to without adsorbent and using a fixed bed. Rehydration ratios for vibro-fluidized bed with adsorbent, without adsorbent and fixed bed dryer were approximately equal to 4.2, 4.5 and 6.5 after minutes. Higher rehydration quality generally implies a more porous structure of the dried product. In terms of rehydration quality drying with adsorbent in a vibro-fluidized dryer shows best results. A fully dried product of porous structure is obtained after hours of drying time (Fig 5.46). A product with a highly porous structure absorbs water more readily and hence shows higher rehydration quality. However, in the case of a fixed bad and vibration without adsorbent only obtained a partially dried product is obtained at the same 112 Chapter-5 Experimental Results and Discussion drying time. It had poor rehydration quality. This result shows that gentle vibration causes minimal damage to the internal structure of the product. Effect of vibration factor and adsorbent refreshing on rehydration quality can also been seen in Fig 5.48. No significant impact of these parameters was found on rehydration quality of the AFD product. Rehydration ratio VF-6.4; Adsorbent refreshing VF-6.4; Adsorbent VF-6.4; No adsorbent VF-4.8; Adsorbent refreshing Without vibration Time, Figure 5.48 Variation of rehydration ratio with time for different drying conditions Figures 5.49A and 5.49B show scanning electron microscope (SEM) images of the cross section of potato samples subjected to VFD and vibro-fluidized bed AFD with adsorbent, respectively. Some minor damage of the internal structure is noted for AFD with vibro-fluidized bed, which causes a less porous structure inside the product. Due to vibration, the product impacts the bed as well as the adsorbent particles causing this phenomenon. 113 Chapter-5 Experimental Results and Discussion Bed Product Porosity Cell Figure 5.49A Microscopic photograph of cross section (x 40 500μm) of dried potato under VFD. (Drying condition-Two stage: -11°C (convection) and -6°C (convection, radiation and conduction)) Bed Product Porosity Cell Figure 5.49B Microscopic photograph of cross section (x 40 500μm) of dried potato under AFD with fix bed dryer (Drying condition-Two stage: -11°C (convection) and -6°C (convection, radiation and conduction)) 114 Chapter-5 Experimental Results and Discussion Bed Product Cell Porosity Figure 5.49C Microscopic photograph of cross section (x 40 500μm) of dried potato under AFD with vibro-fluidized bed dryer with coating (Drying condition-Two stage: -11°C (convection) and -6°C (convection, radiation and conduction)) Bed Product Porosity Cell Figure 5.49 D Microscopic photograph of cross section (x 40 500μm) of dried potato under AFD with vibro-fluidized bed dryer without coating (Drying condition-Two stage: -11°C (convection) and -6°C (convection, radiation and conduction)) 115 Chapter-5 Experimental Results and Discussion Usually all types of foods and vegetable products need to perform a coating operation of thin film of gold over the sample due to non conductive properties prior to a SEM test. Therefore it was decided to carry out an investigation of the effect of the coating itself on the internal structure of the product. Figures 5.49C and 5.49D show a comparison of the internal structure between SEM images with coating and without coating of the product. It was observed that the product without coating showed more porous structure than the product with coating. The product was placed in a vacuum chamber for about 1.5 -2 minutes and mild heating was employed during coating; this process which causes the observed minor damage to the internal structure of the product. However, the internal structure was more clearly visible only when the usual coating was applied. Therefore, neglecting the minor damage of the internal structure of the product, coating of the test specimens was employed during the SEM tests. Table 5.5 C shows the measured changes in color after drying using various AFD techniques. Overall color difference (ΔE) for AFD couple with vibro-fluidized bed dryer and adsorbent, AFD with fixed bed and VFD are found to be 29.19, 11.16 and 32.34, respectively. Results illustrate that for the case of VFD and AFD with vibroTable 5.5 Change in color for potato samples subject to AFD drying Potato L a b c Original 42.58 -0.25 10.15 10.15 AFD: -11oC (0-4 hrs) & -6oC (4-8 hrs) Vibrated AFD: Feq-20hz; A-3mm; -11oC (0-4 hrs) & -6oC (4-8 hrs); with adsorbent 53.49 -2.39 9.2 9.51 71.2 -3.01 15.22 16.01 VFD 74.87 -1.83 10.9 11.06 116 Chapter-5 Experimental Results and Discussion Table 5.5 B Change in color for potato sample Carrot L a b c Original 49.72 26.3 34.77 43.6 AFD: -11oC (0-4 hrs) & -6oC (4-8 hrs) Vibrated AFD: Feq-20hz; A- 3mm; -11oC (0-4 hrs) & -6oC (4-8 hrs); with adsorbent 40.32 19.42 30.82 36.43 41.32 17.98 24.89 31.21 VFD 47.8 23.62 33.11 40.67 Table 5.5 C Change in color before and after drying for potato Potato ∆L ∆a ∆b ∆c ∆E*ab AFD: -11oC (0-4 hrs) & -6oC (4-8 hrs) -10.91 Vibrated AFD: f-20hz; A-3mm; -11oC (0-4 hrs) & -6oC (4-8 hrs); with adsorbent -28.62 2.14 0.95 0.64 11.16 2.76 -5.07 -5.86 29.196 VFD 1.58 -0.75 -0.91 32.34 -32.29 Table 5.5 D Change in color before and after drying for carrot Carrot AFD: -11oC (0-4 hrs) & -6oC (4-8 hrs); with adsorbent Vibrated AFD: f-20hz; A-3mm; -11oC (0-4 hrs) & -6oC (4-8 hrs); with adsorbent VFD ∆L ∆a ∆b ∆c ∆E*ab 9.4 6.88 3.95 7.17 12.30 8.4 8.32 9.88 12.39 15.41 1.92 2.68 1.66 2.93 3.69 fluidized bed with adsorbent showed brighter color of the dried potato cubes. The value of L was found even higher than the original product for the above two case is as shown in Table 5.5 A. However, for AFD with fixed bed, a lower value of L (less brightness) was noted in the final dried product. On the other hand, carrot shows slight color degradation in both AFD as shown in Tables 5.5 B and 5.5 D. 117 Chapter-5 Experimental Results and Discussion Figure 5.50 Shape and size potato and carrot samples before drying Figure 5.51 Shape and size of potato and carrot samples after AFD using vibrofluidized bed and adsorbent Figure 5.50 and 5.51 shows the dimensions and shape of the potato and carrot samples before and after drying. Besides minor shrinkage and deformation due to gentle vibration and physical contact with the adsorbent, it was observed that potato sample was almost same as original product, while a slightly more deformation and shrinkage 118 Chapter-5 Experimental Results and Discussion was noted in carrot sample. This can be explained with the help of SEM micrographs and rehydration quality measurements for the dried carrot samples. To investigate viability of the proposed AFD system (Vibro-fluidized bed dryer with multimode heat input and adsorbent), a comparison was made between the proposed AFD process with similar sets of data available in the literature (Eikevik, et al. 2005) for the other types of drying i.e. of AFD using heat pump assisted fluidized bed dryer; this is shown in Figure 5.52. 1.0 Drying Conditions-Two stage: o Dimensionless moisture conten o -8 C and +20 C Product: Cod fish cube Adsorbent: Silica gel (dia 4mm) Dimension: 5x5x5mm 0.9 0.8 0.7 0.6 0.5 Literature data (Eikevik et al.2005)-convection heat input using heat pump 0.4 Multimode heat input without vibration ( convection +radiation+conduction) 0.3 Multimode heat input with vibration (convection+radiation+coduction) 0.2 Multimode with vibration and adsorbent (convection+radiation+conduction) 0.1 Time, hr Figure 5.52 Comparison of different drying conditions on dimensionless moisture content with time Comparison was carried out under the drying conditions for the two-stage process at -8oC and 20oC. Cubic-shaped (5mm) cod fish product was used for this comparison. It can be seen from Figure 5-52 that the proposed system displays better drying performance than heat-pump based system. The final dimensionless moisture content 119 Chapter-5 Experimental Results and Discussion for the heat pump-assisted fluidized bed dryer after hours of drying time was about 0.38. However, for the vibro-fluidized bed dryer with multimode heat input and vibrofluidized bed dryer with multimode heat input and adsorbent, it was about 0.19 and 0.16, respectively, for the same drying time. Supply of required amount of energy for sublimation through multimode heat input (Rahman et al 2007) plays an important role in achieving this improvement 5.3 Summary A new two-stage atmospheric freeze drying (AFD) scheme is proposed and tested in a laboratory scale apparatus utilizing a vortex tube to supply cryogenic air for drying. It was shown that a multi-mode AFD process using conduction and radiation coupled with convection gives faster drying kinetics without compromising product quality which compared favourably with vacuum freeze dried products. The two-stage process uses a lower temperature air for an initial period followed by an increased temperature when the drying rates decreases sharply. Furthermore, a comparative study was carried out to investigate the drying kinetics as well as quality analysis on an AFD system using a vibro-fluidized bed with adsorbent and multimode heat input. Experimental results showed that the proposed system presents a significant improvement in terms of the freeze drying kinetics as compared to an existing commercial AFD systems using heat-pump assisted fluidized bed, with analysis of product quality demonstrated that the proposed system compares well with VFD and AFD with fixed bed in terms of rehydration, color and internal structure of the dried product. Results revealed that SEM tests without coating reflect better the 120 Chapter-5 Experimental Results and Discussion true porous structure of the dried product; on the other hand, coating (with gold film) helps obtain clear visualization of the internal structure with minor damage. Results of osmotic dehydration as a pretreatment method prior to AFD show that although osmotic pretreatment does reduce the initial moisture content prior to start of drying, and also can enhance the drying kinetics, it has an adverse effect on product quality for both atmospheric and vacuum freeze drying. On the basis of extensive parametric studies is concluded that the proposed AFD system (Vibro-fluidized bed dryer with a vortex tube, multimode heat input and adsorbent) is an attractive alternatives to overcome some of the drawbacks of existing AFD as well as VFD systems. However, this based on laboratory scale studies. Pilot and full scale tests are recommended since this process shows promise of success. Also, technoeconomical studies are needed prior to commercialization. 121 [...]... A similar size and shape of finally dried product was observed for VFD and AFD Figure 5. 24, which shows the puffed and undeformed shape of 82 Chapter -5 Experimental Results and Discussion Vacuum freeze dried potato Atmospheric freeze dried potato Heat pump dried potato Figure 5. 24 Photographs of disc and rectangle-shaped dried potato sample using VFD, AFD and HPD 83 Chapter -5 Experimental Results and. .. rehydration quality was observed for case-1 at -11oC 0.20 Absorbed water, gm Carrot-Two stage -11C & -6C (Rad & Cond) Potato- Two stage: -11C & -6C ( Rad and Cond) 0. 15 0.10 0. 05 0.00 1.0 1 .5 2.0 2 .5 3.0 3 .5 4.0 4 .5 5.0 Time, min Figure 5. 15 Comparison of rehydration quality of potato and carrot samples with time 72 Chapter -5 Experimental Results and Discussion Figure 5. 15 shows the comparison of the absorbed... 3 .5 3 2 .5 2 Banana-Untreated Banana-Osmotic in sugar solution 1 .5 Potato- Untreated Potato- Osmotic in sugar solution 1 0 1 2 3 4 Time, min 5 6 7 8 Figure 5. 33 Variation of rehydration ratio of osmotically treated (concentrated sugar solution) and untreated banana slices with time 91 Chapter -5 Experimental Results and Discussion approaches the equilibrium moisture content; the moisture content was about... the “skinning” effect noted for the treated samples It is also obvious this figure that untreated dried biological products 92 Chapter -5 Experimental Results and Discussion Pre-dried banana Untreated dried banana Dried banana-Treated with sugar solution Dried banana- Treated with salt solution Rehydrated banana-Treated with sugar Rehydrated banana- Treated with salt solution solution 93 ... -6C (Rad & Cond) 2 Potato-HPD: 43C and 1. 45 m/s 1 0 1 2 3 4 5 Time, min Figure 5. 21 Comparison of relative mass index with time for different drying processes for potato Figures 5. 21 and 5. 22 show the relative mass indices for potato and carrot slices dried using VFD, AFD and HPD The relative mass index was found to be about 6.24, 5. 62 and 4.82, respectively, for potato, while 2 .55 , 7.27 and 6 .55 , respectively,... 0.366 kg/m2h and 0.1 95 kg/m2h, respectively Results show a comparable drying rate between VFD and AFD for both samples, while a higher drying rate was observed for HPD relative to VFD and AFD However, quality tests showed that HPD caused significant product degradation in terms of the internal structure, rehydration rate and colour 7 6 Relative mass index 5 4 3 Potato VFD: -57 C Potato-AFD:Two stage: -11C... using carrot is shown in Figure 5. 18 Similar results are noted for carrot, which gives support for viability of the proposed AFD system Although a full technoeconomic evaluations was not made Figure 5. 19 and Figure 5. 20 show the comparison of drying rate data for potato and carrot between AFD, VFD and HPD At the beginning of the drying mass flux for potato was about 0.649 kg/m2h, 0.416 kg/m2h and 0.176... potato and carrot Cell Porosity Cell Figure 5. 16 (b) SEM photographs picture of horizontal surface (x 80 200μm) of dried potato and carrot (Drying condition-Two stage: -11°C (convection) and -6°C (convection, radiation and conduction)) 73 Chapter -5 Experimental Results and Discussion respectively Figures 5. 16 (a) and 5. 16(b) shows the microscopic picture of cross sectional area and horizontal surface area...Chapter -5 Experimental Results and Discussion moisture content reached 0 .5 and the corresponding drying time was about 4 hours This can be attributed to the fact that the higher heat transfer rates attained in the two stage process accelerate the rate of sublimation at the ice vapor interface It causes a higher pressure gradient to develop between the interface and the drying medium This... shows higher drying rate when the sample is pretreated in 85 Chapter -5 Experimental Results and Discussion 2 .5 Banana-Untreated Banana-Osmotic in sugar solution Banana-Osmotic in salt solution Moisture content, kg/kg db 2 Atmospheric freeze drying Two stage process: 0 0 -11 C (0 to 4 hr) and -6 C (after 4 hr) Disc type : D-16 cm and Thickness-1 mm 1 .5 1 0 .5 0 0 1 2 3 4 5 6 7 8 9 Time, hr Figure 5. 26 Effect . annularly along the Chapter -5 Experimental Results and Discussion 59 tube inner wall as it moves axially down the tube. A part of this air is adiabatically expanded inward to the centre, according. Figure 5. 5 Variation of product (potato) and drying air temperature with time for the two stage drying process Chapter -5 Experimental Results and Discussion 63 - 25 - 15 -5 5 15 25 35 0. 10 15 20 25 30 Time, min Temperature, C o Air temperature after vortex tube at variable pressure: 6 bar (0 to 14 min) and 4 bar (after 14 min) Air temperature inside drying chamber at variable pressure: