Drying 2004 – Proceedings of the 14th International Drying Symposium (IDS 2004) São Paulo, Brazil, 22-25 August 2004, vol. C, pp. 1922-1929
APPLICATION OF HEAT PUMP IN DRYING OF APPLE CYLINDERS
Ana L. Gabas1, Marina Bernardi, Javier Telis-Romero2 and Vânia R. N. Telis2 1. Department of Food Engineering, FZEA/USP, P.O. Box 23, 13630-900 – Pirassununga, SP, Brazil, E-mail:
[email protected] 2. Department of Food Engineering and Technology, UNESP- Campus de São José do Rio Preto, 15054-000 – São José do Rio Preto, SP, Brazil Keywords: heat pump dryer, apple, drying kinetics, COP, rheological properties ABSTRACT Drying kinetics of apple (var. Gala) was studied by using heat pump dryer (HPD) and electric resistance dryers with parallel airflow. The use of HPD showed to be adequate in the drying process of apples, mainly in relation to the rate conversion of electric energy into thermal energy. The heat pump effective coefficient of performance (COPHT,EF) was between 2.48 and 2.58, with an energy economy of about 40% when compared to the drying system with electric resistance. The rehydration ratio and shrinkage coefficient of the apples were similar for the both type of dryers. The relaxation times obtained from the two elements Maxwell model are mainly affected by the viscous modulus in relation of its elastic character for both dryers. The analyses of vitamin C realized during drying showed that a smaller loss of ascorbic acid occurred for the pump dryer, supposing due to the air drying conditions. INTRODUCTION Drying processes are energy intensive and knowledge about their efficiency and optimum operating conditions is vital for the economical operation of dryers. For heat sensitive materials such as food, the loss of quality relating to color, nutrients, taste and texture is another important factor to be considered simultaneously with energy conservation. Much work has been done to increase the drying efficiency of convection drying, particularly by the application of heat pump dryers (HPD). HPD are finding increasing applications in food industry for drying nuts, fruit, vegetables, herbs and fish products in many countries (Alves-Filho and Strommen, 1996; Ho et al., 2001; Chua et al., 2001). A heat pump dryer is made up of five major components, namely a compressor, a condenser, an evaporator, an expansion valve and a dryer with a fan to provide air movement. The humid air from the dryer is passed over the evaporator of the heat pump, which acts as a dehumidifier. In this section the
1922
humid air loses a part of its moisture by giving up heat to the working fluid. The dry air passes then over the condenser, where it is heated by the condensing refrigerant. The HPD shares the advantage of lowtemperature drying and manifests the same desirable characteristics of resistance heat (reliability and easy control), whereas it is considerably more energy efficient (Hogan et al., 1983). The HPD application is suitable for high value products. Meyer and Greyvenstein (1992) studied economic feasibility of the HPD as applied to grain drying and found a reduction in the operating period that made the HPD more economical than other dryers. Rossi et al. (1992) reported that onion slices dried by an HPD consumed less energy in comparison to a conventional hot air dryer, presenting higher quality. The objective of this work was to compare the performance of heat pump and electrical resistance dryers (with parallel airflow) on apple drying. The influence of dryer type on drying kinetics, rehydration ratio, rheological properties, shrinkage coefficient and vitamin C were also investigated. Rheology The rheological parameters of some fruits and vegetables are useful information for the textural characterization of the materials. For a small deformation (strain) many foods can be assumed to behave as linear viscoelastic materials. The simplest linear viscoelastic model is the Maxwell model, consisting of an elastic (spring) and a viscous (dashpot) bodies in series. For a simple Maxwell model, at constant strain, the applied stress (σ) decays from σo to σ (t), after time t (Mohsenin 1986): σ(t ) = ε o E1 . exp −
t + Ee λ1
(1)
where σ (t) denotes stress at any time “t”; εo is a constant strain; E1 and Ee are the moduli of the spring representing the ideal elastic body and equilibrium moduli, respectively and λ1 (= η /E1) is the relaxation time. According to Mohsenin (1986), the physical significance of relaxation time can best be appreciated considering that the relaxation time is the time required for the stress to decay to 1/e times or to 36.8 percent of its original value. Then, at constant strain (ε), a Maxwell element relaxes its stress exponentially at a rate determined by relaxation time (λ). Most viscoelastic foods do not follow the simple Maxwell model (Eq. 1) and it is necessary more complex models to describe their stress relaxation curves. The generalized Maxwell model, consisting of a finite number of parallel simple elements, has been applied in several food systems. The stress relaxation curve of this model for two elements can be described by the equation: σ(t ) = ε o E1 . exp −
t t + E 2 . exp − + Ee λ1 λ2
(2)
where εo is a constant strain; λ1 and λ2 and E1 and E2 are relaxation times and spring moduli for the first and second elements, respectively. On the other hand, Peleg and Normand (1983) suggest that stress relaxation data can be calculated as a normalized stress (or normalized force term) and fitted to the following linear equation (Steffe, 1996): σo t (3) = k1 + k 2 t σo − σ Bulk Shrinkage Coefficient An interesting potential application of data on volume change as a function of moisture content results from the drying theory of Whitaker (1980). Cellular tissue making up the solid foodstuff may be regarded as a multiphase system and, by making use of the transport theorem and the average theorem, the macroscopic transport equations were developed. In this approach, the bulk shrinkage coefficient is necessary, and it is given by (Lozano et al., 1980):
1923
sb =
Vb (x ) Vb,o
(4)
Heat Pump Dryer Efficiency The drying efficiency is a measure of the quantity of energy used in removing a unit mass of water from a product. The efficiency of a heat pump can be expressed as its effective coefficient of performance (COPHT,EF) defined as the ratio of the heat recovered at the condenser to the work required by the compressor: COPHT, EF = Q HT WCP (5)
where WCP is the power input to the compressor and Q HT is the heat delivered in the condenser (capacity of air heating) estimated as: (6) Q HT = m AIR C pAIR (T3 − T2 ) (7) m AIR = ρ AIR VAIR where m AIR is the mass flow rate of air, CpAIR is the specific heat of air, VAIR is the volumetric flow rate of air, ρAIR is the density of air, T2 and T3 are the average air temperatures entering and leaving the condenser, respectively. The increase of the drying potential, r, related to ambient air is given by: ( WHP,SAT − WHP ) ∆WHP r= = (8) ∆WAMB ( WAMB,SAT − WAMB ) where WAMB and WHP are the absolute humidities of the ambient air and of the air at the exit of heat pump evaporator, respectively; WAMB,SAT and WHP,SAT and are the absolute humidities of the saturated air with the same enthalpies of the ambient air and of the air at the exit of heat pump evaporator, respectively (Rossi, 1993). The estimated energy consumption for a hypothetical electrical resistance system supplying drying air with the same drying potential as the HPD, was calculated by: WER =
m AIR C pAIR (T3∗ − T1 )
(9) η ER is the electrical efficiency, taken as 80% (Rossi, 1993), T1 is the ambient air
In equation (9) ηER temperature and T3∗ is the required air drying temperature in order to provide the same drying potential as
the HPD. Temperature T3∗ is estimated by using the psychometric chart, considering that the change in absolute humidity of the air conditioned by electrical resistance and submitted to an adiabatic saturation process, ∆WER (equation 10), should have the same value that ∆WHP (see equation 8). ∆WER = (WER ,SAT − WAMB ) = ∆WHP (10) In the above equation, WER,SAT is the absolute humidity of the saturated air with the same enthalpy as the air with humidity WAMB at temperature T3∗ .
MATERIAL AND METHODS Material
Apples of variety Gala (7.21 kg/kg dry basis) were acquired, namely in the fresh market apple and kept in a cold room at 7oC prior to their use. Drying curves were obtained for ripe, fresh apples, which were
1924
sorted visually for color, size and physical damage. Fruits were rinsed in fresh water, peeled and cut in small cylinders (diameter = 0.024 m and length = 0.012 m). Drying Equipment
Two dryer configurations – heat pump (HPD) with air flowing parallel to drying product and electrical resistance with parallel (PR) airflow – were simultaneously tested to avoid the influence of different environmental conditions. A schematic illustration of both dryers can be seen in the work of Queiroz et al. (2004). Drying temperatures were fixed at 50º, 40o and 30oC with air velocities of 1.5 m/s. The refrigerant R22 was chosen as the working fluid. Dryer consisted of three basic sections: heating, air flowing and drying. The drying chamber was horizontal with parallel airflow in relation to the drying material. Air is forced through an axial blower with air velocity being measured by means of an anemometer. Dry bulb temperatures were measured by copper-constantan thermocouples. The relative humidities of the ambient and just before drying chamber were measured by thermo-hygrometers connected to a portable meteorological station (Oregon Scientific, model WMR918). Power consumption was determined by periodically measuring the current of each component of the system with a multimeter (Minipa, model ET2700). Rehydration Ratio
The rehydration ratio of the dried apple cylinders was determined by mixing 10 g de apple cylinders with 100 ml of distilled water in a 500 ml beaker. The beaker was left for 4 h at 30oC. The rehydration (rehydrated apple cylinders mass/dried apple cylinders mass) was calculated according to Gabas (1998). Rheological Properties
After the drying processes samples were conditioned at 25oC in a known relative humidity in order to attain known water activities. Conditioning was carried on desiccators containing K2CO3 saturated salt solutions, corresponding to water activities of 0.432. The samples were kept into desiccators until they reached constant weight. The rheological behavior of rehydrated samples was evaluated by compressionrelaxation tests performed at room temperature (25oC) in a Universal texturometer (TA-XT2i Texture Analyzer, Stable Micro Systems, Surrey, UK). Samples were individually compressed using an acrylic 35 mm diameter plate probe lubricated by glycerin. The initial strain rate was 1 mm/s and total strain was limited to 10% in order to ensure that changes in the transversal section area would be minimum during compression tests. Three samples obtained at each drying condition and rehydrated at water activity were subjected to compression-relaxation tests, leading to average relaxation curves. Shrinkage during drying
The dimensions and shrinkage of individual apple particles were measured as follows: dimensions were determined by averaging a number of measurements with calipers and volume by a method based on the buoyant forces which act on a body submerged in a liquid (Gabas, 1998). Different curves of shrinkage were obtained during drying by using pump and tray dryers. Each of these curves was the average of three replicates. It may be pointed out that the difference between three replications was less than 9% in all cases.
1925
Vitamin C
The analyses of vitamin C were realized in a duplicate form according to Gabas et al. (2003). The final filtrated aliquots were taken for titration with 2,6-dichlorophenolindophenol 0.01% and the titration end point was detected visually. RESULTS AND DISCUSSION Performance of the Heat Pump Dryer
The use of the heat pump systems can significantly enhance the energy efficiency. This could be observed by the magnitude of the effective coefficient of performance (COPHT,EF), which shows the efficiency of the equipment in converting mechanical into thermal energy that was later supplied to drying air when flowing through the condenser. The COP of a heat pump dryer is a direct function of the refrigerant used, dryer configuration and drying conditions. Table 1 presents COPHT,EF (Eq. 6) of the heat pump dryer system for drying chamber temperatures of 50o, 40o and 30oC. In spite of heat losses in the condenser and in the high-pressure line and inefficiencies inherent to the compressor processes, the conversion rate of mechanical to thermal energy was satisfactory. Rossi (1993) found values in the range of 1.3 to 1.7 in a similar system. Table 1: Effective coefficients of performance.
T (ºC) 30 40 50
Power (kW) 2.222 2.283 2.552
RHEXT (%) 19 14 14
RHINT (%) 16 11 12
COPHT,EF
COPHT,EF*
2.54 2.58 2.48
2.08 2.12 2.03
RHEXT: Relative humidity of the atmospheric air heated to the corresponding drying temperature without dehumidification. RHINT: Relative humidity inside the drying chamber. *Considering the energy consumption of the fan responsible for the air movement.
The advantageous energetic performance of the HPD is also demonstrated on Table 2, which presents the electric energy savings obtained by using the HPD instead of an electrical resistance dryer supplying air with the same drying potential. Comparison between WCP , the power input to the compressor in the real HPD, and WER estimated by eq. (10), indicated a reduction of about 38 to 47% in the electric energy consumption. Table 2 also shows the calculated values of R, the increase in drying potential of air conditioned by the HPD in relation to the ambient air, which were in the range of 2.1 to 3.3. Relative humidity of the atmosphere is also of fundamental importance for the process efficiency. Clements et al. (1993) concluded that a high relative humidity of the air entering the evaporator is very important to the successful design of a heat pump assisted dryer. Their results showed that when ambient relative humidity was increased from 30 to 80%, the SMER (Specific Moisture Extraction Rate) of the dryer increased nearly two fold. The present work was carried out in a region of hot and dry climate, in such a way that the performance of the system was limited by meteorological conditions. Nevertheless, in spite of the relatively adverse environmental conditions, the energetic performance of the used heat pump dryer was satisfactory.
1926
Table 2: Increase of drying potential and economy of electric energy for HPD.
Atmospheric Drying Chamber Conditions Conditions T (ºC) RH (%) T (ºC) RH (%)
28 28 29
41 49 44
30 40 50
16 11 12
•
•
r
W ER (kW)
W HP (kW)
Reduction (%)
2.11 3.31 2.53
3.56 4.33 4.11
2.14 2.29 2.55
39.9 47.1 38.0
Drying Kinetics
The HPD showed the highest drying rates due to the lower relative humidity inside drying chamber. At the considered drying conditions, a total drying time of 6.7 hours was necessary to reach the adimensional moisture content of 0.24 in the HPD, whilst about 7.8 hours were required in the PR dryer, what means a reduction of approximately 14.1% in the drying time. Rossi (1993), also comparing an HPD with an electric resistance dryer, obtained reductions in the range of 26 to 32%. All the tested drying conditions resulted in drying rates with the same behavior. The period of constant drying rate was not observed, indicating that a film of water did not exist at the surface of the apples and whatever water reached the surface from within the body evaporated almost immediately. This indicated that the apple drying processes were controlled by moisture diffusion inside the solid. Rehydration Ratio
The rehydration ratio of the dried samples (44% dry basis and air velocity of 1.5 ± 0.12 m/s) was 1.661 and 1.591 at heat pump and tray dryer, respectively. It was not observed significant differences between sample dried in the pump heat and tray dryers. Rheological Properties
The Single Maxwell, Two Elements Maxwell and Peleg & Normand models, given by equations (1), (2) and (3), respectively, were selected to fit the experimental data. The two elements Maxwell model exhibited the smallest RMS values. The effect of dryer type on the rheological behavior (Eq. 3) of rehydrated apple is clearly demonstrated on Table 3. Each experimental point represents the average of four measurements because of the heterogeneity of the natural products (Gabas et al., 2002). There was a small increase in the values of both viscous constants, ηi when the drying conditions changed of tray dryer to heat pump dryer. On the other hand, the elastic moduli Ei were not very affected by the pretreatment and different dryers. It shows that relaxation times are mainly affected by the viscous modulus in relation to its elastic character for both dryers. The results suggested that the dried apples obtained by tray dryer or heat pump dryer were very similar. Table 3: Parameters of the two elements Maxwell model for rehydrated apple cylinders after different drying conditions and water activity of 0.432.
Drying conditions T = 40oC Pump Heat Dryer Tray dryer
E1 (kPa) 88.4 79.8
λ1 (s) 16.9 15.3
η1 (kPa.s) 1493.9 1220.9
E2 (kPa) 44.0 40.9
λ2 (s) 2.1 2.0
Ee RMS* R** η2 (kPa.s) (kPa) (%) 92.4 33.3 8.63 0.983 81.8 30.2 7.88 0.989
* Root Mean Square ** Coefficient of Determination
1927
Shrinkage during drying
The average results (triplicate) of shrinkage coefficients showed that there was not a difference of the shrinkage coefficient during drying of apples in the both type of dryers. The equations of shrinkage coefficient as a function of moisture content of the apples are shown as following: for the heat pump dryer (Equation 11) and for the tray dryer (Equation 12): X X Sb = 1.001 + 0.872 + 0.885 exp − 4.5 × 10− 8 (11) XO XO Sb = 0.998 + 0.861
X X + 0.888 exp − 1.8 × 10− 7 XO XO
(12)
Vitamin C
Table 3 shows the ascorbic acid content of fresh and dried apple with the two methods of drying. A marked loss of ascorbic acid occurred during drying in both dryers, however a less ascorbic acid loss was observed for the pump dryer, supposing due to the air drying conditions. Table 3: Ascorbic acid content of fresh and dried apple during drying in heat pump and tray dryer.
Moisture (d.b.) 7.601 5.185 3.363 2.221 1.521 1.249 1.054 0.636 0.439 0.349
Heat pump dryer Time of Ascorbic Ascorbic drying acid content Acid loss (min) (mg/100gTS) (%) 0 4.02 0 60 3.66 8.96 180 3.44 14.43 300 3.11 22.64 420 3.01 25.12 480 2.86 28.86 540 2.77 31.09 720 2.45 39.05 900 2.05 49.00 1020 1.88 53.23
Tray dryer Moisture Time of Ascorbic acid Ascorbic (d.b.) drying content Acid loss (min) (mg/100gTS) (%) 7.601 0 4.02 0 5.574 60 3.45 14.18 3.418 240 2.99 25.62 2.241 420 2.78 30.85 1.525 600 2.57 36.07 1.277 660 2.22 44.78 1.124 720 1.98 50.75 0.641 1020 1.64 59.20 0.469 1200 1.28 68.16 0.341 1440 0.87 78.36
CONCLUSIONS In spite of the adverse environmental conditions, the energetic performance of the used HPD dryer was satisfactory, indicating a reduction of about 38 to 47% in the electric energy consumption. It was obtained a total drying time of 6.7 hours to reach the adimensional moisture content of 0.24 in the HPD and about 7.8 hours in the tray dryer. Similar values of rehydration ratio were observed for both dryers. The rheological tests of the rehydrated apples were well performed by the two elements Maxwell model. The relaxation times were mainly affected by the viscous modulus in relation to its elastic character for both dryers. Equations of shrinkage coefficient as a function of moisture content of the apples were obtained. A less ascorbic acid loss was observed for the pump dryer, supposing due to the air drying conditions. REFERENCES Alves-Filho, O., Strommen, I. (1996), The Application of Heat Pump in Drying of Biomaterials. Drying Technology, 14 (9), pp. 2061-2090.
1928
Chua, K.J., Mujumdar, A.S., Hawlader, M.N.A., Chou, S.K., Ho, J.C. (2001), Batch Drying of Banana Pieces – Effect of Stepwise Change in Drying Air Temperature on Drying Kinetics and Product Colour. Food Research International, 34, pp. 721-731. Clements, S., Jia, X., Jolly, P. (1993), Experimental Verification of a Heat Pump Assisted Continuous Dryer Simulation Model. International Journal of Energy Research, 17, pp. 19-28. Gabas, A.L., Telis-Romero, J., Menegalli, F.C. (2003), Cinética de Degradação do Ácido Ascórbico em Ameixas Liofilizadas. Ciência e Tecnologia de Alimentos, 23(Supl), pp. 66-70. Gabas, A.L. (1998), Secagem de Uva Itália em Leito Fixo. MS Thesis, UNICAMP, Campinas – SP, Brasil, 137 p. Gabas, A.L., Menegalli, F.C., Ferrari, F., Telis-Romero, J. (2002), Influence of drying conditions on the rheological properties of prunes. Drying Technology, 20(7), pp. 1485-1502. Ho, J.C., Chou, S.K., Mujumdar, A.S., Hawlader, M.N.A., Chua, K.J. (2001), An Optimization Framework for Drying of Heat-sensitive Products. Applied Thermal Engineering, 21, pp. 1779-1798. Hogan, M.R., Ayers, D.L., Muller Jr., R.E., Rall, E.C., Doering, O.C. (1983), Heat Pump for Lowtemperature Grain Drying. Transactions of the ASAE, 26 (4), pp. 1234-1238. Lozano J.E., Rotstein E., Urbicain M.J. (1980), Total porosity and open-pore porosity in the drying of fruits. Journal of Food Science, 45, 1403-1407. Meyer, J.P., Greyvenstein, G.P. (1992), The Drying of Grain with Heat Pumps in South Africa: a TechnoEconomic Analysis. International Journal of Energy Research, 16, pp. 13-20. Mohsenin, N. (1986), Physical Properties of Plant and Animal Materials, 2th ed. Gordon and Breach, London. Peleg, M. and Normand, M.D. (1983), Comparison of two methods for stress relaxation data presentation of solid foods. Rheological Acta, 22, pp. 108-113. Queiroz, R., Gabas, A.L., Telis, V.R.N. (2004), Drying Kinetics of Tomato by using Electric Resistance and Heat Pump Dryer. Drying Technology, 22(6). Rossi, S.J. (1993), Desenvolvimento e Avaliação de uma Bomba de Calor usada no Condicionamento de Ar para Secagem de Alimentos. Doctor Thesis, UNICAMP, 214p. Rossi, S.J., Neves-Filho, L.C., Kieckbush, T.G. (1992), Thermodynamic and Energetic Evaluation of a Heat Pump Applied to the Drying of Vegetables, Drying’92, Ed. Mujumdar, A.S., Elsevier Science, pp. 1475-1478. Steffe, J.F. (1996), Rheological Methods in Food Process Engineering, 2nd Ed., Freeman Press, USA. Whitaker, S. (1980), Heat and Mass Transfer in granular porous media. In “Advances in Drying”, Ed., Arum Mujumdar, vol 1. Hemisphere-Mc Graw-Hill, New York.
1929
