The 2nd Joint International Conference on “Sustainable Energy and Environment (SEE 2006)” 21-23 November 2006, Bangkok, Thailand

E-037 (O)

Drying and Heat Transfer Behavior of Combined Low-pressure Superheated Steam and Far-infrared Radiation Drying of a Food Product Chatchai Nimmol1,∗, Sakamon Devahastin2, Thanit Swasdisevi1 and Somchart Soponronnarit1 1 2

School of Energy and Materials, King Mongkut’s University of Technology Thonburi, Thailand Department of Food Engineering, King Mongkut’s University of Technology Thonburi, Thailand

Abstract: The present study aimed at investigating the use of drying system combining the concept of already proven low-pressure superheated steam drying with far-infrared radiation (LPSSD-FIR) to dry food products. The effects of various operating parameters, i.e., drying medium temperature and pressure, on the drying kinetics of a model material (banana) and the energy consumption of the process were investigated and discussed. Comparison was also made with the similar set of data obtained from the system with combined far-infrared radiation and vacuum drying (VACUUM-FIR) conducted in the same drying chamber. It results showed that LPSSD-FIR drying required more drying time than VACUUM-FIR drying at almost all drying conditions except at the temperature of 90oC. Although LPSSD-FIR drying required more energy consumption than VACUUM-FIR drying at almost all drying conditions, the lowest specific energy consumption was obtained during LPSSD-FIR drying at 90oC and 7 kPa. Keywords: Banana, Convective, Drying Rate, Radiative, Specific Energy Consumption, Vacuum Drying

1. INTRODUCTION Drying is generally performed to preserve the products from being deteriorated and also to reduce the cost of transportation and storage as well as to produce products that would not be able to obtain otherwise. Among the many techniques available hot air drying is one of the most common technique for drying, especially for food and agricultural products. Hot air drying, however, is a very energy-intensive operation and leads to much degradation of product quality. In order to reduce the energy requirement during the dehydration process and also to minimize the quality degradation of the dried products, it is necessary to select an efficient drying system. Far-infrared radiation (FIR) drying which has received much attention is one possible means for the above purpose. In the process of far-infrared radiation drying, the energy in the form of electromagnetic wave is absorbed directly by the product without loss to the environment leading to considerable energy savings, uniform temperature distributions [8,15,14] and keeping good product quality [7]. Several works indeed successfully applied FIR to dry food products such as rice [1], potato [2], barley [3], and welsh onion [11-12]. The results obtained clearly indicated that the drying time could be considerably reduced with the use of FIR. Although the use of FIR provides shorter drying process, hence reduced energy consumption, higher product temperature developed during FIR drying may cause some undesirable change of the products, e.g., surface burning, especially when high radiation intensity (severe drying condition) is applied. This is because most food products are heat-sensitive which are easily damaged by heat. To avoid this problem a means to provide milder dehydration process is therefore desired. One possible way to achi eve this goal is the use of low-pressure (or subtmospheric) superheated steam drying (LPSSD) system [13]. Since superheated steam can be produced at the temperature lower than 100oC due to reduced pressure environment, quality degradation of the products due to elevated temperature are reduced. Therefore, dried products undergoing LPSSD have superior quality (both in terms of physical and nutritional qualities) than those dried by conventional hot air and even vacuum drying [5-6,9]. To attain the advantages and to understand heat transfer behavior of above-mentioned drying methods, the combination of those dr ying techniques should be developed as a novel drying technology. The focus of the present study was to investigate the use of drying system combining the concept of low-pressure superheated steam and far-infrared radiation (LPSSD-FIR) to dry food products. The effects of various operating parameters, i.e., drying medium temperature and pressure, on the drying kinetics of a model material (banana) and the energy consumption of the process were investigated and discussed. Comparison was also made with the similar set of data obtained from the system with combined far-infrared radiation and vacuum drying (VACUUM-FIR) conducted in the same drying chamber

2. EXPERIMENTAL SET-UP, MATERIALS AND METHODS 2.1 Experimental set-up A schematic diagram of the combined low-pressure superheated steam and far-infrared radiation drying system is shown in Fig. 1. 3 The dryer consists of a stainless steel drying chamber, insulated with rock wool, with inner dimensions of 45 × 45 × 45 cm ; a steam reservoir, which received the steam from a boiler; a liquid ring vacuum pump (Nash, ET32030, Trumball, CT), which was used to maintain the vacuum in the drying chamber; a far-infrared radiator (Infrapara, A-2T-500, Malaysia) rated at 500 W with a surface 2 area of 60 × 120 mm , which was used to supply thermal radiation to the drying sample and the drying medium; and electric heater rated at 1500 W, which was used to maintained superheated steam temperature in the case of low-pressure superheated steam drying experiment.

Corresponding author: [email protected]

1

E-037 (O)

The 2nd Joint International Conference on “Sustainable Energy and Environment (SEE 2006)” 21-23 November 2006, Bangkok, Thailand

Fig. 1 A schematic diagram of the combined low-pressure superheated steam and far-infrared radiation drying system: 1) boiler; 2) steam valve; 3) reservoir; 4) pressure gauge; 5) steam trap; 6) pressure reducing valve; 7) drying chamber; 8) vacuum pump; 9) far-infrared radiator; 10) electric fans; 11) steam inlet and distributor; 12) sample holder; 13) thermocouples; 14) load cell; 15); vacuum break-up valve; 16) PID controller; 17) PC with data acquisition card The distance between the far-infrared radiator and the sample holder made of a stainless steel screen with dimensions of 15 × 15 2 cm was set at 165 mm. The operation of the far-infrared radiator was controlled through the temperature of the drying medium (air or superheated steam) at 30 mm above the sample surface via a Proportional-Integral-Differential (PID) controller (Shinko, JCS33A-R/M, Japan) with an accuracy of ±0.1oC. The change of the mass of the sample during drying was detected continuously (at 1 min interval) using a load cell (Minebea, Ucg3 kg, Nagano, Japan) with an accuracy of ±0.2 g. The temperatures of the drying medium and of the drying sample were measured continuously using type K thermocouples. The thermocouple used to measure drying medium temperature was located at the same position as the thermocouple that was used for sending the signal to the PID controller to control the far-infrared radiator. The thermocouple was covered partly with an aluminum foil acting as a radiation shield. The average surface temperature of the farinfrared radiator was also measured using a type K thermocouple. Thermocouple signals were multiplexed to a data acquisition card (Omega Engineering, CIO-DAS16Jr., Stamford, CT) installed in a PC. LABTECH NOTEBOOK software (version 12.1, Laboratory Technologies Corp., MA) was then used to read and record the temperature data. 2.2 Materials and Methods Gros Michel banana (Musa Sapientum L.) with an initial moisture content in the range of 265 to 310% (d.b.) [4] was used as a tested material in this study. Prior to the start of each experiment banana was peeled and sliced by a slicing machine to 3 mm thick. The sliced samples were then cut into 30 mm diameter using a die. To perform a drying experiment approximately 16 pieces of prepared banana slices (about 35 g) was placed on sample holder. To reduce the amount of steam condensation in the drying chamber during a combined low-pressure superheated steam and far-infrared radiation (LPSSD-FIR) experiment the far-infrared radiator was turned on to heat up the sample and maintain the drying chamber temperature to the desired value without the application of steam to the drying chamber during the first 5 min of the drying process. The flow rate of steam into the drying chamber was maintained at about 26 kg/h and the speed of the electric fans was fixed at 2100 rpm. For an experiment using far-infrared radiation under vacuum (VACUUM-FIR) the same experimental set-up was used but without the use of electric fans due to the absence of steam injection to the drying chamber. The same experimental conditions were therefore achievable. The experiments were carried out at the drying medium (air or superheated steam) temperatures of 70o, 80o, and 90oC and absolute pressures of 7 and 10 kPa. The drying experiments were performed until the sample moisture content of 3.5% (d.b.) was obtained. 2.3 Evaluation of energy consumption In this study, energy consumption of dying process measured using kilowatt-hour meter came from the electrical energy consumed by the operation of the vacuum pump and the far-infrared radiator. Since the electric fans consumed a very small of electrical energy due to lower power rating, the energy consumption of the electric fans was neglected. The efficiency of drying process can be evaluated through the specific energy consumption that is the measure of the energy required during the process to remove 1 kg of water in the products being dried. Specific energy consumption was calculated by: SEC = Evacuum + Ethermal

mwater

(1)

where SEC is the specific energy consumption (kWh/kg water), Evacuum is the electrical energy required by vacuum pump (kWh), Ethermal is the electrical energy required by the far-infrared radiator (kWh) and mwater is the amount of water removal (kg) which could be estimated as the different between the initial and final mass of the products.

3. RESULTS AND DISCUSSION 3.1 Dedydration characteristics Fig. 2 shows drying curves of banana slices, expressed in term of moisture ratio, undergoing different drying methods at various operating conditions. In the case of LPSSD-FIR drying it is seen from Fig. 2a that the drying time decreased with an increase in the drying temperature, as expected. This is because the temperature difference between the sample and the superheated steam at a higher drying temperature was greater than that at a lower temperature hence a larger driving force of heat transfer, which is also related to the rate of mass and heat transfer as well. The moisture diffusivity is also higher at a higher temperature. In addition, the drying time

2

E-037 (O)

The 2nd Joint International Conference on “Sustainable Energy and Environment (SEE 2006)” 21-23 November 2006, Bangkok, Thailand

also decreased with a decrease in the drying pressure. This is due to the fact that water within the product evaporated at lower temperature when drying was performed at a lower pressure. It can also be seen that at higher drying temperatures (80o and 90oC) the rates of moisture reduction were more affected by the drying temperature than by the drying pressure. This may probably due to the fact that temperature was the dominant factor influencing the superheated steam thermal properties, especially at higher drying temperatures. Although the drying chamber was preheated by thermal radiation via the use of the far-infrared radiator, small amount of steam condensation still occurred and could be observed over a very short period. Moreover, when drying was performed at 70oC the sample could not reach the required final moisture content during the first 250 min of drying even at the lowest drying pressure tested (7 kPa). This is because of an excessive amount of steam condensation in the drying chamber. This phenomenon was also observed in the case of drying at 10 kPa (drying curve is not shown in Fig. 2a).

Fig. 2 Drying curves of banana slices undergoing different drying methods at various drying conditions In the case of VACUUM-FIR drying (see Fig. 2b) the phenomenon was similar to the case of LPSSD-FIR drying that drying at higher temperatures and lower pressures required shorter drying time. Unlike LPSSD-FIR, however, sample dried at the lowest drying temperature (70oC) could reach the required final moisture content because no steam condensation existed. It was also found that the effect of drying pressure was less clear at all drying temperatures. This may due to the fact that temperature was the main factor influencing the air thermal properties in the operating ranges tested. It was again found from Fig. 2 (see also at Table 1) that the sample dried by LPSSD-FIR and VACUUM-FIR required less drying time than that dried by LPSSD at all drying conditions. This is due to the effect of volumetric heating generated within the sample dried by the system with the far-infrared radiation. In addition, it was observed that the samples dried by VACUUM-FIR required less drying time than that dried by LPSSD-FIR at lower drying temperatures (70o and 80oC). This is due to the initial steam condensation on the sample surface as well as the smaller temperature differences in the case of LPSSD-FIR drying within this range of temperature. However, LPSSD-FIR yielded shorter drying time (higher drying rate) when drying was conducted at 90oC at all drying pressures tested. This is due to the fact that an increased drying temperature led to higher drying rates, resulting from sharply increased the differences between the superheated steam and the samples surface temperature (the saturation temperature corresponding to the drying pressure) in the case of LPSSD-FIR drying. However the differences between the air temperature and the sample surface temperature (wet-bulb temperature) in the case of VACUUM-FIR drying increased only slightly as the drying temperature increased. This suggests that the effective inversion temperature calculated from the overall drying periods was somewhere between 80o and 90oC [17]. 3.2 Temperature evolution Changes in moisture ratio and temperature of banana slices undergoing different drying methods and conditions are shown in Figs. 3 and 4. As revealed by these figures the temperature evolution patterns were affected by both drying methods and conditions. In the case of LPSSD-FIR drying, it can be seen from Fig. 3 that the temperature of the samples, measured at their center, fell suddenly from their initial values within the first 3 min of the process. This is due to the rapid reduction of the chamber pressure, which led to some flash evaporation of surface moisture [16]. After this period the temperature of the samples rose rapidly to the level close to the boiling point of water corresponding to the drying pressure (not at the boiling point since far-infrared radiation was presented) and then remained constant at this level until the surface of the samples started to dry. In addition, it was also observed that the period of constant sample temperature was longer when drying was conducted at lower temperature and higher pressure as can be seen from Figs. 3c and 3d, respectively. Since heat transfer process in the case of LPSSD-FIR drying simultaneously took place by radiation from the far-infrared radiator and by convection from superheated steam, the temperature of the sample rose steadily to the level higher than the pre-determined medium temperatures. This is due to the fact that thermal radiation from the far-infrared radiator, controlled by the pre-determined medium temperature, was the only source of energy supplied to the drying medium and the sample. Therefore, drying at higher temperatures resulted in higher sample temperatures due to higher radiation intensity. After this period the temperature of the sample remained almost constant. This is due to the fact that during the later stage of the process moisture content within the sample was smaller leading to lower far-infrared radiation absorptivity of the sample [15]. It is interesting to noted that, the temperature evolution pattern of LPSSD-FIR samples after the period of constant sample temperature mentioned above was clearly different from that of LPSSD samples reported by Devahastin et al. [6] who found that the sample temperature after the period of constant sample temperature rose steadily and eventually approached to the drying medium temperature.

3

E-037 (O)

The 2nd Joint International Conference on “Sustainable Energy and Environment (SEE 2006)” 21-23 November 2006, Bangkok, Thailand

Fig. 3 Changes in moisture content and temperature of banana slices undergoing LPSSD-FIR at different drying conditions. moisture ratio; drying medium temperature; sample temperature

Fig. 4 Changes in moisture content and temperature of banana slices undergoing VACUUM-FIR at different drying conditions. moisture ratio; drying medium temperature; sample temperature

4

E-037 (O)

The 2nd Joint International Conference on “Sustainable Energy and Environment (SEE 2006)” 21-23 November 2006, Bangkok, Thailand

In the case of VACUUM-FIR drying it was found from Fig. 4 that the temperature of the samples also suddenly dropped during the initial period of the process. However, the temperature of the samples after this period steadily rose to the level higher than the pre-determined medium temperature, as noted earlier in the case of LPSSD-FIR drying, without the period of constant sample temperature. After this period the temperature of the samples also remained almost constant as in the case of LPSSD-FIR drying. To investigate the effects of radiation intensity of far-infrared radiator on the temperature evolution of the sample, the temperature of samples dried by LPSSD-FIR and VACUUM-FIR were again compared. At the same drying conditions, it can be seen from Figs. 3 and 4 that the sample temperature during the later stage of the process in the case of LPSSD-FIR drying was higher than in the case of VACUUM-FIR drying. This is due to the fact that in the case of LPSSD-FIR drying, radiation intensity at the position of the thermocouple used for sending the signal to the PID controller (at 30 mm above the sample surface) was more attenuated due to the absorption of superheated steam, resulting from higher far-infrared radiation absorptivity of superheated steam due to the presence of a large proportion of water vapor [10]. Therefore, the far-infrared radiator during LPSSD-FIR drying was used more often to maintain the desired level of drying medium temperature leading to higher surface temperature of the far-infrared radiator (data is not shown). Consequently, the radiation intensity, depending on the surface temperature of the far-infrared radiator, experienced by LPSSD-FIR sample was greater hence higher level of the sample temperature. It was also observed that in the case of VACUUM-FIR drying at 70oC (see Figs. 4e and 4f) the sample temperature during the later stage of the process was very close to the drying medium temperature (not much higher than the drying medium temperature) compared with those dried at higher temperatures. This may probably due to the fact that when drying at lower temperature (70oC in this case), the radiation intensity was lower indicated by lower surface temperature of the far-infrared radiator. Since the sample temperature was not controlled by the drying medium temperature but by the radiation intensity, very high sample temperatures developed in the cases of drying at higher temperatures resulting in overheating and burning of the product hence degradation of dried product quality, especially in the case of LPSSD-FIR drying at the highest temperature tested (90oC). It should be noted that the effects of drying pressure on the level of sample temperature during the later state of drying mentioned earlier was not significant. 3.3 Energy consumption Table 1 compares energy consumption along with drying time of banana slices undergoing different drying methods and conditions. It can be seen from this table that the energy consumption of the vacuum pump increased with an increase in the drying time. This is because electrical energy consumed by the vacuum pump at all drying conditions were the same, 1.5 kWh in this study. Regarding the electrical energy required to generate thermal energy, it was found that the electrical energy required to generate thermal energy of LPSSD-FIR drying was higher than that of VACUUM-FIR at all drying conditions. This is because the far-infrared radiator was used more often during LPSSD-FIR drying resulting from higher far-infrared radiation absorptivity of superheated steam as mentioned earlier in the case of temperature evolution. Table 1 Energy consumption at different drying methods and conditions Specific Total steam Energy consumption (kWh) Pressure Drying time c energy consumption Consumption (kPa) (min) Vacuum pumpb Thermal Total (kWh/kg water) (kg) 10 N/Aa N/A N/A N/A N/A N/A 70 7 N/A N/A N/A N/A N/A N/A 10 190 4.750 0.530 5.280 193.975 82.333 LPSSD-FIR 80 7 140 3.500 0.490 3.990 146.583 60.667 10 100 2.500 0.420 2.920 107.274 43.333 90 7 90 2.250 0.410 2.660 97.722 39.000 10 255 6.375 0.230 6.605 242.652 – 70 7 185 4.625 0.200 4.825 177.259 – 10 145 3.625 0.270 3.895 143.093 – VACUUM-FIR 80 7 130 3.250 0.250 3.500 128.582 – 10 120 3.000 0.340 3.340 122.704 – 90 7 110 2.750 0.330 3.080 113.152 – a N/A implies that the final moisture content of 3.5% (d.b.) was not obtainable at this condition. b The electrical energy consumption required to operate the vacuum pump was 1.5 kWh. c The flow rate of steam into the drying chamber was maintained at about 26 kg/h. Drying method

Temp (oC)

Considering the total energy consumption that is the sum of the electrical energy required to operate the vacuum pump and to generate thermal energy, it was found that although, at the same drying condition, the electrical energy required to generate thermal energy of LPSSD-FIR drying was higher than that of VACUUM-FIR drying, the total energy consumption of LPSSD-FIR drying was lower when drying was performed at the temperature of 90oC. This is due to the fact that this level of drying temperature (90oC) was above the inversion temperature, which was somewhere between 80o and 90oC as noted earlier, resulted in a faster drying process for LPSSD-FIR drying compared to VACUUM-FIR dying [17]. A similar trend was also observed in the case of specific energy consumption. It is also seen from Table 1 that LPSSD-FIR drying at 90oC and 7 kPa provided the lowest value of the specific energy consumption. Although the dried product quality was generally considered to optimize the operating parameters of the drying process, it was beyond the scope of this paper. Therefore, LPSSD-FIR drying at 90oC and 7 kPa was suggested, based on the energy utilization, as the optimum condition of the process in this study. In the case of drying with the application of steam into the drying chamber (LPSSD-FIR), it was again found from Table 1 that the total steam consumption increased with an increase in the drying time, as expected.

5

E-037 (O)

The 2nd Joint International Conference on “Sustainable Energy and Environment (SEE 2006)” 21-23 November 2006, Bangkok, Thailand

4. CONCLUSION A drying system combining the radiative and convective heat transfer process, so called a combined low-pressure superheated steam and far-infrared radiation drying system (LPSSD-FIR), for drying food products was developed and studied to understand the drying and heat transfer behavior. The results showed that with the use of far-infrared radiation, the temperature of both LPSSD-FIR and VACUUM-FIR samples during the later stage of drying were higher than the pre-determined medium temperatures. It was also found that LPSSD-FIR drying required more drying time than VACUUM-FIR drying at almost all drying conditions except at the temperature of 90oC; this indicated that the inversion temperature calculated from the overall drying rates should be somewhere between 80o and 90oC. Although LPSSD-FIR drying required more energy consumption than VACUUM-FIR drying at almost all drying conditions, the lowest specific energy consumption was obtained during LPSSD-FIR drying at 90oC and 7 kPa. This condition was suggested, based on the energy utilization, as the optimum condition in this study.

5. ACKNOWLEDGMENTS The authors express their sincere appreciation to the Commission on Higher Education, the Thailand Research Fund (TRF), the National Research Council of Thailand and the International Foundation for Science (IFS), Sweden for supporting the study financially.

6. REFERENCES [1] Abe, T. and Afzal, T.M. (1997) Thin-layer infrared radiation drying of rough rice, Journal of Agricultural Engineering Research, 67, pp. 289-297. [2] Abe, T. and Afzal, T.M. (1998) Diffusion in potato during far infrared radiation drying, Journal of Food Engineering, 37, pp. 353-365. [3] Afzal, T.M., Abe, and Hikida, Y. (1999) Energy and quality aspects during combined FIR-convection drying of barley, Journal of Food Engineering, 42, pp. 117-182. [4] AOAC (1984) Official methods of analysis (14th ed.), Washington DC: Association of Official Agricultural Chemists. [5] Barbieri, S., Elustondo, M.P. and Urbicain, M.J. (2004) Retention of aroma compounds in basil dried with low pressure superheated steam, Journal of Food Engineering, 65, pp. 109-115. [6] Devahastin, S., Suvarnakuta, P., Soponronnarit, S. and Mujumdar, A.S. (2004) A comparative study of low-pressure superheated steam and vacuum drying of a heat-sensitive material, Drying Technology, 22, pp 1845-1867. [7] Dostie, M., Seguin, J.N., Maure, D., Ton-That Q.A. and Chatingy, R. (1989) Preliminary measurements on the drying of thick porous materials by combinations of intermittent infrared and continuous convection heating, in: A. S. Mujumdar, M. A. Roques (eds.), Drying ’89, Hemisphere, New York. [8] Ginzburg, A.S. (1969) Application of infrared radiation in food processing, Chemical and Process Engineering Series, Leonard Hill, London. [9] Leeratanarak, N., Devahastin, S. and Chiewchan, N. (2006) Drying kinetics and quality of potato chips undergoing different drying techniques, Journal of Food Engineering, 77, pp. 635-643. [10] Mills, A.F. (1999) Basic heat and mass transfer (2nd ed.), Prentice-Hall, New Jersey. [11] Mongpraneet, S., Abe, T. and Tsurusaki, T. (2002) Accelerated drying of welsh onion by far infrared radiation under vacuum conditions, Journal of Food Engineering, 55, pp. 147-156. [12] Mongpraneet, S., Abe, T. and Tsurusaki, T. (2002) Far infrared-vacuum and convection drying of welsh onion, Transactions of the ASAE, 45, pp. 1529-2535. [13] Mujumdar, A.S. (2000) Superheated steam drying–Technology for the future, in: S. Devahastin, (eds.), Mujumdar’s Practical Guide to Industrial Drying, Brossard, Exergex, pp. 115-138. [14] Ratti, C. and Mujumdar, A.S. (1995) Infrared Drying, in: A.S. Mujumdar, (eds.), Handbook of industrial drying: Volume 1, Marcel Dekker, New York. [15] Sandu, C. (1986) Infrared radiative drying in food engineering: A Process Analysis, Biotechnology Progress, 2, pp. 109-119. [16] Sun, D.W. and Wang, L.J. (2006) Development of a mathematical model for vacuum cooling of cooked meats, Journal of Food Engineering, 77, pp. 379-385. [17] Suvarnakuta, P., Devahastin, S., Soponronnarit, S. and Mujumdar, A.S. (2005) Drying kinetics and inversion temperature in a low-pressure superheated steam-drying system, Industrial & Engineering Chemistry Research, 44, pp. 934-1941.

6