Latin American applied research
versão ISSN 0327-0793
Lat. Am. appl. res. vol.41 no.2 Bahía Blanca abr. 2011
Moisture migration during a tempering time after the heat treatment step in yerba maté processing
M.E. Schmalko, N.N. Lovera and G.C. Kolomiejec
FCEQyN- Universidad Nacional de Misiones-Felix de Azara 1552 - 3300 - Posadas - Argentina
Abstract - The aim of this research was to study the effect of applying a tempering time to the branches of yerba maté after the heat treatment stage (or sapecado). Assays were carried out in three industrial producers of Misiones Argentina. Branches were obtained from the sapecador outlet and then, they were put in rest in order to form a bed. First, moisture content of leaves and twigs, separately, and then losses of mass of whole branches were measured. When the branches were put in rest in a bed during 30 min, 8.60 kg of water/100 kg of dry matter were transferred from the twigs to the leaves and 5.17 kg of water/ 100 kg of dry matter were lost by evaporation.
Keywords - Yerba Maté; Tempering; Moisture; Migration.
Yerba maté (Ilex paraguarienses Saint Hilaire) processing consists of four stages called sapecado, drying, grinding and seasoning. Generally, the first three stages are carried out in industrial factories located near the plantation.
When the branches are collected, they are cut by hand and quickly carried in order to be processed. At the reception, a wide variation in weights and forms of branches can be observed. These variations are in weight (between 5 and 100 g), shape (with or without sub-ramifications) and length (between 30 and 60 cm) (Crotti et al., 2002). Once received, they are put in a yard and they are introduced into the sapecador.
In the sapecado stage, enzymes producing browning in leaves are inactivated. The equipment consists of a cylinder that rotates at low revolution (about 10 rpm). Branches are fed in the extreme where the wood is burning. They pass through the flame between 1 and 3 times, receiving heat by radiation and convection. According to this, different branches can receive different heat treatments and loss different quantities of water. After that, the burning gases and the branches move to the output extreme in a parallel flux. Temperature of the air in the inlet extreme reaches about 350-400°C, while at the outlet extreme it is above 100°C (Peralta and Schmalko, 2007).
Initial moisture content of leaves and twigs are very similar (about 60%, wet basis), but at the outlet of the sapecador, the differences are high. In leaves, mean value is about 20% (wb) and in twigs, about 55% (wb) (Schmalko and Alzamora, 2001). Ratio between surface area/weight in leaves is 3 or more times greater than twigs. So, heat and mass transfer in leaves is greater than twigs and that is why losses of water are higher in leaves. Moroever, water diffusion coefficient in leaves are higher than in twigs (Schmalko et al., 1996).
Alfalfa is dried as whole branches in rotary dryers, similar to yerba maté. A similar moisture gradient is generated between the stems and leaves (Arinze et al., 2003; Arinze et al., 2007; Moore and Cilanski, 1992; Patil et al., 1992). In order to avoid excess in drying, the dried leaves are separated from the stems by dragging forces in order to have a minor residence time. Consequently, the dryer efficiency becomes higher as the solid flux is increased.
A similar technology could be applied to yerba maté processing. But these changes would modify the types of dryers used now by the industrial producers. These changes could be very expensive. So, other alternatives should be studied.
Intermittent drying (drying with tempering times) is applied to some foods in order to redistribute their internal moisture gradient. An improvement of product quality and energy saves is obtained. It has been used in rice drying (Li et al., 2003; Mabamba and Yabés, 2005; Nishiyama et al., 2006; Shei and Chen, 2002). In this material, when the intermittent drying is applied, cracking of the grain is avoided, because internal moisture gradient is reduced and energy efficiency is erased because time of heat applying is reduced, too.
The aim of this research was to study the effect on moisture content of applying a tempering time to the branches after the sapecado stage. If an appropriate moisture distribution is obtained, energy saves could be obtained in the drying stage.
II. MATERIAL AND METHODS
Branches of yerba maté (Ilex paraguarienses Saint Hilaire) were used as test material. They were obtained at the outlet of the sapecador. Measures were made in three industrial factories between April and August/2008.
Four series of measurements were made:
1. Study of moisture migration in different branches: Branches of different shapes and sizes were used. They were put into an isolated container in order to form a bed. One branch was taken after a certain time (0, 10, 20, 30, 45, 60, 90, 120, 210 y 300 min), then, the leaves were separated from the twigs and put into a hermetic container in order to determine moisture content. Eight series of measures were carried out.
2. Determination of moisture content at different parts of a branch: Branches with at least 4 sub-ramifications were used. In each sub-ramification, leaves were separated from twigs, and put into a hermetic containers in order to determine moisture content. Eight series of measures were carried out.
3. Study of moisture migration in a single branch: Branches, with at least 5 sub-ramifications, were used. They were put into an isolated container and they were taken after certain time (0, 10, 20, 30 and 1000 min, considering time=infinite). In each sub-ramification, leaves were separated from twigs and put into a hermetic container in order to determine moisture content. Thirty series of measures were carried out.
4. Weight losses of branches during the tempering time: Branches of different shapes and sizes were taken and put into a container. The weight of each branch was measured at 0, 5, 10, 15, 20 and 30 min. Sixteen series of measures were carried out with branches with leaves and two without leaves.
C. Moisture content
Moisture content was determined by drying the sample in an oven at 103 ± 2°C until a constant weight was reached. It took about 6 h (IRAM 20503, 1995).
III. RESULTS AND DISCUSSION
A. Study of moisture migration in different branches
A regression analysis was made, considering the variation of moisture content in leaves and twigs during the tempering time. A significative fit level was reached when some data were rejected. The leaves had an intercept value of 21.19 % (wb) and a slope of 0.081 kg/(100 kg.min) (increase of moisture), while in twigs, they were 57.20 % (wb) and -0.0160 kg/(100 kg.min) (loss of moisture) (Fig. 1 and 2). In both cases, the confidence limits were very big.
Figure 1. Moisture content variation of leaves during the tempering
Figure 2. Moisture content variation of twigs during the tempering
The very high variation of data values could have probably appeared for many reasons. First of all, the low control of the equipment, specially the control of the size of the flame. This provokes different heat treatments for different branches. Second, the variability in sizes and shapes of branches makes the residence time vary between 2 and 4 minutes (Crotti et al., 2002). This variation produces different moisture losses in the different branches. Third, the ratio between weight/ surface area is different for different leaves (Coelho et al., 2002) and twigs (i.e. different diameters). This variation produces different heat and mass transfer in different branches.
To reduce the variation in moisture content between branches; the experiences, in a second step, were carried out with the different sub-ramifications of one branch. In this case, similar residence time and heat treatment can be assumed.
B. Moisture distribution at different parts of a branch
In all cases, branches with at least 4 sub-ramifications, were used. Values which varied between 4.05 and 25.33 %(wb) were found. The variation coefficient was very high: 39.13%. Figure 3 shows the frequency distribution of moisture in leaves at the outlet of the sapecador. The experiences were carried out in one day. As can be observed, the distribution is very uniform. The variability between the sub-ramifications of the same branch was very high, too. A variation coefficient of 28.64% was found. Considering that the sub-ramifications of the same branch have the same residence time, this variation can only be found if they have different exposures when the branch crosses the flame. This fact produces a different heat treatment and consequently, different degrees of moisture in the branches.
Figure 3. Moisture content distribution of leaves at the outlet of the sapecador (time=0)
C. Moisture migration in a single branch
In all cases, branches with at least 5 sub-ramifications were used. Significative differences in moisture content of leaves were found at different times (0, 10, 20, 30 and 1000 min) , with P<10-4. (Fig. 4). Mean values were 17.27% (wb) at time= 0 min and 22.75% (wb) at time= 30 min. The mean increase of leaves moisture content was 5.48% (wb) (or 8.60 kg of water/100 kg of dry solid). In twigs, the differences in moisture content were not statistically significative (Fig. 5). The high variability in moisture content is probably the reason of this result. Confidence limits of moisture content, in both materials, were minor than ± 2% (wb). This value is quite different than that found in different branches (± 10%).
Figure 4. Moisture content variation of leaves from one branch during the tempering
Figure 5. Moisture content variation of twigs from one branch during the tempering
D. Weight losses of branches
During the tempering time, weight losses of branches, due to moisture evaporation, were detected. The percentage of the weight losses is shown in Fig. 6. After 30 min, the loss reached 3.3 % (wb) (or 5.17 kg water/100 kg of dry solid). When the experiences with branches without leaves were carried out, the losses were lower (2.4 % after 30 min).
Figure 6. Weight losses of branches during the tempering
Temperature variation in the bed, during the tempering time, was similar in all experiences. It varied between 69°C at time=0 min to 37°C after 30 min (Fig. 7).
Figure 7. Temperature variation in the bed of branches during the tempering
E. Energy aspects
Energy balance was carried out in order to determine the percentage of heat losses during the tempering. It was carried out considering mean values of temperature and a tempering time of 30 min. Energy balance can be expressed as:
The balance was carried out with 100 kg of dry solid. Et was calculated considering the specific heat of the branches and the difference in temperature between 69 and 37°C. Specific heat of the branches was calculated using the equation reported by Schmalko et al. (1996), for leaves and twigs, separately. They were 2.20 kJ/kg°C for leaves and 3.11 kJ/kg°C for twigs. In 100 kg of dry solid there are 66.90 kg of dry leaves and 33.10 kg of dry twigs (Schmalko, 2005). Considering their moisture contents there are 80.87 kg of leaves and 75.16 kg of twigs. Et is calculated with the following equation:
Et= 80.87g*2.20 (kJ/kg°C)*(69-37)°C +
75.16kg*3.11 (kJ/kg°C)*(69-37)°C = 13173 kJ
As it was experimentally determined, in 100 kg of dry solid, 5.17 kg of water was evaporated. Considering that the latent heat at 53°C is 2366 kJ/kg (Schmalko et al., 1998), energy used to evaporate the water is:
Eu= 5.17kg*2380 (kJ/kg) = 12305 kJ ,
and the heat losses are:
El= Et - Eu = 13173 kJ - 12305 kJ= 868 kJ
It represents only the 6.6% of Et, and it can be considered a low value (Kudra, 2004).
When moisture of leaves and twigs were determined when a tempering time was applied to branches of yerba maté after the sapecado stage, an important water migration from twigs to leaves was found. Simultaneously, weight losses, due to moisture evaporation, were found.
Great variability in moisture of leaves and twigs was found after the sapecado stage. This variability could probably be due to different shapes and sizes of the branches and different exposures when the branch crossed the flame. A variation coefficient of 39.13% in leaves moisture was found at the outlet of the sapecador.
A tempering time of 30 min to the branches at the outlet of the sapecador should be applied. In this period the loss of moisture in the bed was 5.17 kg of water/100 kg of dry solid. A migration of 8.60 kg of water/100 kg of dry solid from twigs to leaves was also produced in this tempering time. The migration would reduce the drying time at the drying step, improving the energy efficiency. In this period, bed temperature decreased from 69 to 37°C and most of the energy was used to evaporate the water ( 93.4 %).
The authors would like to thank the Agencia para la Promoción Científica de la República Argentina for funding this research.
1. Arinze, E.A., G.J. Schoenau, S. Sokhansanj and P. Adapa, "Aerodynamic separation and dractional drying of alfalfa leaves and stems-A review and new concept," Drying Technology, 21, 1669-1698 (2003). [ Links ]
2. Arinze, E.A, G.J. Schoenau and P. Adapa, "Modeling the fractional drying and aerodynamic separation of alfalfa into leaves and stems in a rotary dryer," Drying Technology, 25, 785-798 (2007). [ Links ]
3. Coelho, G.C., J.E. De Araujo Mariath and E.P. Schenkel, "Population diversity on leaf morphology of maté (Ilex paraguariensis A. St.-Hil., Aquifoliacae)," Brazilian Archives of Biology and Technolgy, 45, 47-51 (2002). [ Links ]
4. Crotti, L.O., M.E. Schmalko and S. Surkan, "Influencia de las características físicas de las ramas en el tiempo de residencia en secaderos rotatorios," IX Congreso Argentino de Ciencia y Tecnología de Alimentos, Buenos Aires, Argentina (2002). [ Links ]
5. IRAM 20503, Yerba Mate: Determinación de la pérdida de masa a 103° C, Instituto de Racionalización de Materiales (1995). [ Links ]
6. Kudra, T., "Energy aspects in drying," Drying Technology, 22, 917-932 (2004). [ Links ]
7. Li, C., J. Liu and L. Chen, "The moisture distribution of high moisture content rough rice during harvesting, storage and drying," Drying Technology, 21, 1115-1125 (2003). [ Links ]
8. Madamba, P.S. and R.P. Yabes, "Determination of the optimum intermittent drying conditions for rough rice (Oryza sativa, L)," Lebensm-Wiss. U.-Technol., 38, 157-165 (2005). [ Links ]
9. Moore, G.A. and W.K. Cilanski, "Thermal properties of high moisture content alfalfa," Transaction of the ASAE, 34, 61-64 (1992). [ Links ]
10. Nishiyama, Y., W.Cao and B. Li, "Grain intermittent drying characteristic analyzed by a simplified model," Journal of Food Engineering, 76, 272-279 (2006). [ Links ]
11. Patil, R.T., S. Sokhansanj, E.A. Arinze and G.J. Schoneau, "Thin layer drying of components of alfalfa," Transactions of the ASAE, 34, 343-346 (1992). [ Links ]
12. Peralta, J.M. and M.E. Schmalko, "Modeling heat and mass transfer in the heat treatment step of yerba maté processing," Brazilian Journal of Chemical Engineering, 24, 73-82 (2007). [ Links ]
13. Schmalko, M.E., L.A. Ramallo and R.O. Morawicki, "A comparison of the diffusion model and the response surface method (RSM) in estimating the drying time of Ilex paraguariensi," Latin American Applied Research 26, 215-220 (1996). [ Links ]
14. Schmalko, M.E., L.A. Ramallo and R.O. Morawicki, "An application of simultaneous heat and mass transfer in a cylinder using the finite-difference method," Drying Technology, 16, 283-296 (1998). [ Links ]
15. Schmalko, M.E. and S.M. Alzamora, "Color, chlorophyll, caffeine, and water content variation during yerba maté processing," Drying Technology, 19, 599-610 (2001). [ Links ]
16. Schmalko, M.E., Estudio y modelado del procesamiento primario de la yerba mate, Tesis de doctorado de la Universidad Nacional de Buenos Aires, Argentina (2005). [ Links ]
17. Shei, H.J. and Y.L. Chen, "Computer simulation on intermittent drying of rough rice," Drying Technology, 20, 615-636 (2002). [ Links ]
Received: June 29, 2009.
Accepted: July 15, 2010.
Recommended by Subject Editor Ricardo Gómez.