Microsoft Word - 476hernandez.docx CHEMICAL ENGINEERING TRANSACTIONS VOL. 43, 2015 A publication of The Italian Association of Chemical Engineering Online at www.aidic.it/cet Chief Editors: Sauro Pierucci, Jiří J. Klemeš Copyright © 2015, AIDIC Servizi S.r.l., ISBN 978-88-95608-34-1; ISSN 2283-9216 Analysis of Dried Onions in a Hybrid Solar Dryer, Freeze Dryer and Tunnel Dryer Alejandro Reyes, Andrea Mahn, Verónica Cares Department of Chemical Engineering, University of Santiago of Chile, Av. L. B. O’Higgins 3363, Santiago, Chile alejandro.reyes@usach.cl Onion (Allium cepa L.) belongs to the lily family, such as garlic and leeks. Dehydrated onion can be presented in powder, chunks, granules or slices. Its applications consider manufacturing of condiments, dehydrated or added to rehydrated meals. We determined the drying curves of onion slices cut into 10 and 5 mm thickness, using a hybrid solar dryer, a tunnel dryer and a freeze-dryer. Drying kinetics adjusted adequately with a phenomenological model and with four empirical models. The best fit was obtained with Ademiluyi empirical model. Diffusivity values obtained from the constant diffusivity model fluctuated between 10-08 and 10 -10 m2 /s. Finally, rehydration curves were determined and their adjustment to the simplified constant diffusivity model resulted in diffusivity values between 10-6 and 10-7 m2/s. 1. Introduction Drying of agro-products corresponds to the most widely used food preservation method, and its purpose is to remove moisture from the food matrix to prevent detriment of nutritional and organoleptic properties (Araya- Farias and Ratti, 2009). Owing to this there is permanent availability of seasonal agro-products, thus facilitating their storage and distribution. The convective drying is one of the most popular techniques due to reduced processing times (Doymaz, 2008, 2013). However, air heating requires large amounts of energy, with two disadvantages: the growing energy deficit and greenhouse effect. Meanwhile, the vacuum freeze-drying process corresponds to dehydration by sublimation which consists of three stages (initial freezing, primary drying or sublimation and desorption or secondary dehydration), with minimal damage to the product (Reyes et al, 2010). However it is a slow process with high economic costs (Ratti, 2001) (estimated at 4 - 6 times that of conventional tunnel drying), and then its use in industry is restricted only to high value-added products. In spite of this, the use of solar radiation as an energy source for pre-heating the drying air is a promising option, considering high energy efficiency (Reyes et al., 2013) and environmental care. The Valencian onion (Allium cepa L.) is a lily bulb which consists of thick, fleshy layers which act as nutrient reserves, and is considered a long-lived onion, and it is one of the most important cultivars in Chile. The nutritional composition of onion varies according to the stage of maturity and variety, highlighting its high water content (close to 90 %) and the presence of important minerals such as potassium, calcium and selenium (Mota et al., 2010; El-Mesery and Mwithiga,2012; Hanif et al., 2013; Pramod et al., 2014). 2. Equipment, materials and methods Drying kinetics curves were obtained in three dryers: hybrid solar dryer (HSD), tunnel dryer (TD) and freeze- dryer (FD). As shown Figure 1, HSD consisted of a 3 m length and 1 m width solar panel, composed by a glass sheet (5 mm thickness) and a black wavy zinc plate. The surface exposed to solar radiation was 10 m2, including 40 zinc fins (3 cm height and 3 m length). Below this zinc plate it was placed a thermal insulating material (50 mm thickness). The air passes through the free space (50 mm height) between the glass and the zinc plate until reaching the mixing point with the recycled air. This air mixture enters the electric heating system composed by 5 kW electric resistances where the air temperature is adjusted to 60 °C. After that, the DOI: 10.3303/CET1543024 Please cite this article as: Reyes A., Cares V., Mahn A., 2015, Analysis of dried onions in a hybrid solar dryer, freeze dryer and tunnel dryer, Chemical Engineering Transactions, 43, 139-144 DOI: 10.3303/CET1543024 139 air enters the drying chamber (0.5 m*0.5m *1.2 m), where it distributes to pass over 10 perforated plate trays made of stainless steel (0.45m*0.5m), located in two sections of 5 trays each one. The solar energy accumulator, placed beside the solar panel, contained 14 kg of paraffin wax (PCM) distributed in 100 copper pipes (inner diameter of 14 mm) with aluminum fins in order to favor heat transfer to the drying air. The air is sucked through a 3 kW blower and directed to the electric resistances zone, although a 20 – 30 % fraction is removed from the system. The air flow rate through the trays was 0.4 ± 0.1 m/s. An advanced control system based on fuzzy logic permit to control the opening of solar panel valve and energy accumulator valve in function of solar radiation and ambient temperature( Reyes et al., 2014). Figure 1: Hybrid solar dryer. A: Solar panel, B: Solar energy accumulator The tunnel dryer, shown in Fig 2A, is a convective forced-circulation dryer, used with a superficial air velocity in the empty section of the dryer of 2 m/s, that possesses an air heating system composed by electrical resistances. Figure 2B shows the freeze-dryer (LABCONCO) with a 4.5 L capacity and condenser temperature of -50 °C, equipped with a vacuum pump. Figure 2: A) Tunnel dryer; and B) Freeze-dryer The initial moisture content of three onion samples was determined in a vacuum oven until constant weight was attained, according to AOAC 920.151. The initial mean moisture content was 89.12 ± 1.4%.The onions were peeled and cut into 5±0.1 or 10±0.1 mm sliced before dehydration in any of the three dryers. The onion slices formed a single layer in each dryer. In the HSD and TD the temperature was set at 60 °C, and in the HSD the air recycling was set at 70 %. In FD the operating conditions were the maximum given by the equipment (0.080 [m bar] and -50 °C). Drying curves were determined by measuring the samples mass in time until reaching the final moisture content (about 2 %). The rehydration rate was determined by immersing about 2 g of dehydrated onion samples in a water bath at 30 °C. Samples were withdrawn every 2 min, drained, wrapped in absorbent tissue, and weighed with an analytical balance. The moisture content was determined from the sample weight before and after rehydration. The assays were made in duplicate and average values were reported. 3. Results 3.1 Drying kinetics Figures 3 and 4 shown drying kinetics for 5 and 10 [mm] onion slices. The drying curves exhibited a linear decrease of moisture content in the saturated surface period, until reaching the critical moisture content (about 0.45 ± 0.075). Then the moisture content diminished gradually until reach the final moisture content. The slowest process was freeze-drying, reaching the final moisture content in about 1200 min with no significant difference between the two thicknesses. 140 Figure 3: Drying curves for 5 mm onion slices. Figure 4: Drying curves for 10 mm onion slices. Curves obtained in the tunnel dryer and in the hybrid solar were similar, although slightly higher drying times were observe in the HSD, probably because in this dryer we used air recycling, what means that the air enters with higher humidity, reducing the drying potential. In energy terms, there is less electric energy consumption in the HSD because the air is preheated in the solar panel and/or in the accumulator solar panel. The freeze dryer had the highest energy consumption, given the long drying times. 3.2 Adjustment of drying curves The Fick’s second law was integrated, considering that during drying, the air conditions and Deff remain constant, and that shrinkage and external mass transfer resistance are negligible. Additionally, since equilibrium moisture content (Xeq) is low, it can be neglected. With these approximations, for an infinite slab of 2L thickness, the Constant Diffusivity Model (CDM) is expressed by Eq 1 (Crank, 1975). = − ∗− ∗ ≅ = 8 1(2 1) ∙ −(2 1) ∙ ∙ ∙4 (1) The adjustment to experimental data was evaluated through the correlation coefficient (R) and root mean squared error (RMSE). The diffusivity values, presented in Table 1, were lower for freeze-drying, and in all cases they were within the expected range for agro-products (10-8 – 10-11 [m2/s]) (Revaskar et al., 2014). Table 1: Deff estimated by Eq(1) Dryer Thickness [mm] ⁄ RMSE* Hybrid solar 5 6.0283 ∙ 10 0.0894 0.9987 10 6.6824 ∙ 10 0.0794 0.9989 Tunnel 5 6.4689 ∙ 10 0.0996 0.9988 10 6.7268 ∙ 10 0.0916 0.9997 Freeze 5 1.3514 ∙ 10 0.0877 0.9998 10 3.1512 ∙ 10 0.0764 0.9997 * = ∑ 0 0.2 0.4 0.6 0.8 1 0 500 1000 1500 2000 X/ X 0 Time [min] 5 [mm] - Freeze dryer 5 [mm] - Tunnel dryer 5 [mm] - Hybrid solar dryer 0 0.2 0.4 0.6 0.8 1 0 500 1000 1500 2000 X/ X 0 Time[min] 10 [mm] - Freeze dryer 10 [mm] - Tunnel dryer 10 [mm] - Hybrid solar dryer 141 0 2 4 6 8 0 20 40 60 X d. b [K g H 2O /K g d. s] Time [min] 10 [mm] - Freeze dryer 10 [mm] - Tunnel dryer 10 [mm] - Hybrid solar dryer On the other hand, the drying kinetics of agro-products can be described also by empirical models (Bessadok, 2013; Revaskar et al., 2014), some of which are shown in Table 2 Table 2: Empirical models Model Equation Newton = (− ∙ ) (2) Henderson-Pabis = ∙ (− ∙ ) (3) Page = (− ∙ ) (4) Ademiluyi = ∙ (−( ∙ )) (5) In Table 3, the parameters of the empirical models are presented, which were adjusted through minimizing the mean squared error with respect to the experimental values, using the solver of Microsoft Office ®. The Ademiluyi model shown the best adjustment. 3.3 Rehydration curves Figures 5 and 6 show the rehydration curves of dried onion slices (5mm and 10 mm thickness) at 30 ºC. In all cases it was observed a high rehydration rate during the first ten minutes, and after that an asymptotic increase happened during the next 50 min. From this point onwards the moisture content remained constant. The diffusion path through the dry pores of the onion slices allowed a fast re-absorption of water during the first 10 minutes of rehydration, but as the pores became filled, rehydration became slower. The final water content after rehydration was higher for freeze-dried samples, because the particle does not shrink and its pores remain open. In tunnel and hybrid solar dryer, 5mm onion slices showed a higher rehydration level, attributed to a short diffusion water path, compared with 10 mm slices. This owes to a minor structural damage during the drying process (Melquíades et al., 2009). Besides, as in the case of the drying curves, rehydration curves were adjusted to the CDM (Equation 1), resulting in the values presented in Table 4. The values of effective diffusivity, were higher than those obtained in the drying process, because the structure of dehydrated onion is already damaged, facilitating water diffusion (Melquíades et al., 2009). Figure 5: Rehydration curves for 5 mm onion slices. Figure 6: Rehydration curves for 10 mm onion slices. 0 2 4 6 8 0 10 20 30 40 50 60 X d. b. [K g H 2O /K g d. s. ] Time [min] 5 [mm] - Freeze dryer 5 [mm] - Tunnel dryer 5 [mm] - Hybrid solar dryer 142 Table 3: Parameters of empirical models. Model Thickness mm Dryer Parameters RMSE k (s-1)*104 n a Newton 5 Hybrid Solar 1.123 0.0397 0.996 Tunnel 1.121 0.0456 0.995 Freeze 0.258 0.0498 0.999 10 Hybrid Solar 0.523 0.0181 0.999 Tunnel 0.510 0.0323 0.997 Freeze 0.233 0.0475 0.998 Henderson -Pabis 5 Hybrid Solar 1.178 1.05 0.0355 0.995 Tunnel 1.270 1.05 0.0409 0.994 Freeze 0.322 1.22 0.0277 0.995 10 Hybrid Solar 0.537 1.02 0.0167 0.999 Tunnel 0.527 1.03 0.0299 0.997 Freeze 0.301 1.32 0.0238 0.995 Page 5 Hybrid Solar 0.285 1.26 0.0150 0.999 Tunnel 0.300 1.27 0.0197 0.998 Freeze 0.285 1.34 0.0122 0.999 10 Hybrid Solar 0.320 1.08 0.0115 0.999 Tunnel 0.183 1.17 0.0194 0.998 Freeze 0.245 1.33 0.0158 0.998 Ademiluyi 5 Hybrid Solar 0.220 1.31 0.98 0.0136 0.999 Tunnel 0.247 1.31 0.98 0.0189 0.999 Freeze 0.153 1.49 0.92 0.0122 0.999 10 Hybrid Solar 0.268 1.11 0.98 0.0108 0.999 Tunnel 1.338 1.22 0.98 0.0181 0.999 Freeze 0.245 1.64 0.85 0.0129 0.998 143 Table 4: Effective diffusivities in rehydration. Dryer Thickness [mm] ⁄ RMSE Hybrid solar 5 5.3665·10-7 0.0547 0.9808 10 1.3398·10-7 0.0616 0.9730 Tunnel 5 4.3852·10-7 0.0540 0.9859 10 1.5880·10-7 0.0688 0.9648 Freeze 5 6.8012·10-6 0.0574 0.9912 10 1.0760·10-6 0.0824 0.9611 4. Conclusions Drying curves showed similar shape in both hybrid solar dryer and the tunnel dryer, with longer drying times for the 10 mm slices. Drying kinetics adjusted adequately to the Constant Diffusivity Model and Ademiluyi’s empirical model. Freeze-drying times were considerably longer, although resulting in better quality products. The final moisture content after rehydration was higher for freeze-dried samples. In tunnel and hybrid solar dryer, 5mm onion slices showed a higher rehydration due to a shorter diffusion water path, compared with 10 mm slices. The hybrid solar dryer seems promising because of its lower energy consumption. Acknowledgements The authors thank the financial support made by Project FONDECY 1110101 and DICYT/USACH. References AOAC,1990, Official methods of analysis. In: Association of Official Analytical Chemists (15th Ed.); K. Heldrich (Ed.); Washington, DC, USA. 1298. Araya-Farias M., Ratti C.2009, Dehydration of Foods: General Concepts, in Advances in Food Dehydration, Ed. by Cristina Ratti, CRC Press, Boca Raton, USA. Bessadok A., Characterizing The Drying Kinetics Of High Water Content Agro-Food Particles Exhibiting Non- Fickian Mass Transport. Chemical Engineering Transactions (2013), DOI:10.3303/CET1332294. Crank, J., 1975, The mathematical of diffusion (2 ed.). Oxford, UK: Clarendon Press. Doymaz, I. 2008, Convective drying kinetics of strawberry. Chemical Engineering and Processing, 914-919. Doymaz, I. 2013, Experimental study on drying of pear slices in a convective dryer. International Journal of Food Science & Technology 48,1909 -1915. El-Mesery H., Mwithiga G.2012, The drying of onion slices in two types of hot-air convective dryers. African Journal of Agricultural Research7(30), 4284-4296. Hanif M., Rahman M., Amin M., Khattak M., Ramzan M., Din S., Khan K.2013, Convective Drying and Mathematical Modeling of Onions Dried Using a Flat Plate Solar Air Heater. Pakistan Journal of Nutrition 12(11)1003-1007. Melquíades Y., López, C., Rosas, M. 2009, Dehydration kinetics of carrots (Daucus carota L.). Technological Information 65-72 (in Spanish) Mota C., Luciano, C., Dias, A., Barroca, M., Guiné, R.2010, Convective drying of onion: Kinetics and nutricional evaluation. Food and Bioproducts Processing 115-123. Pramod G., Ramachandra C., Nidoni U. 2014, Dehydration of Onions with Different Drying Methods 3(3),210- 216. Ratti, C.2001, Hot air and freeze-drying of high-values foods: a review. Journal Food Engineer 49(4), 311– 319. Reyes A., Pérez N, Mahn A.2011, Theoretical and Experimental Study of Freeze-Drying of ‘‘loco’’ (Concholepas concholepas). Drying Technology 29, 1386–1395. Reyes A., Mahn A., Cubillos F., Huenulaf P. 2013, Mushroom dehydration in a hybrid-solar dryer. Energy Conversion and Management 70, 31–39. Reyes, A., Cubillos, F., Mahn, A, Vásquez, J. 2014, Dehydration of Agro Products in a Hybrid Solar Dryer Controlled through a Fuzzy Logic System. International Journal of Modern Nonlinear Theory and Application 3, 66-76. Revaskar V.A., Pisalkar P.S., Pathare P.B., Sharma G.P.2014, Dehydration kinetics of onion slices in osmotic and air convective drying process Res. Agr. Eng. 60(3) 92-99. 144