Microsoft word - 21-31 kuye.doc

Journal of Engineering Science and Technology Vol. 2, No. 1 (2007) 21- 31 School of Engineering, Taylor’s University College SIMULATION OF THE DRYING CHARACTERISTICS OF GROUND NEEM
SEEDS IN A FLUIDISED BED
A. KUYE1*, C. O. C. OKO2, S. N. NNAMCHI1 University of Port Harcourt, PMB 5323, Port Harcourt, NIGERIA. *Corresponding Author: aokuye@enguniport.org Abstract
The neem seed is a good source of neem oil as well as insecticides and pesticides. The oil and insecticides can be extracted by two consecutive leaching of neem seed kernels with hexane and ethanol. This work presents a model for simulating the drying of neem seeds in a fluidized bed. Experimental values obtained from literature were used to validate the model prediction. The drying simulation results show that there was a good agreement between the experimental values and the corresponding model predictions. Keywords: Simulation, Neem seeds, Fluidized bed dryer. 1. Introduction

Neem seeds are extracted from neem tree (Azadirachta indica A. Juss) and are rich in neem oil as well as insecticides and pesticides. The seed kernels contain 40 to 50% by mass of non-edible oil and 20% of biodegradable substances that are active against a wide variety of pests [1]. In a typical manufacturing process, the oil and insecticides are removed by two consecutive leaching of neem seed kernels with organic solvents. Hexane is first used to remove oil from the seed, thereafter; ethanol is used to extract the insecticide from the de-oiled seeds [2]. Nomenclatures

Mean logarithmic molar concentration of the dry air kmol/ m 3 Drying rate per surface of solid kg/m 2 sec Initial drying rate per surface of solid kg/m 2 sec Drying rate during constant rate period kg/m 2 sec
Journal of Engineering Science and Technology APRIL 2007, Vol. 2(1)
Drying of Neem Seeds in Fluidised Bed 23 Drying of ground solids in a fluidized bed dryer has three characteristic zones namely: the warm-up, constant and falling drying rate zones [3]. In the warm-up zone, the solids immediately in contact with drying medium tend to approach steady state. Thus, resulting in either decrease or increase of drying rate [4]. At constant rate drying there is enough of the solvent to saturate the solids [5]. Towards the termination of the constant rate period is the critical solvent and transition to third characteristic zone (falling rate period). Drying takes longer time and less solvent is transferred due to internal resistances [4, 6]. Numerous models are available for predicting the drying rates in fluidized beds [7, 8]. Drying kinetics is crucial for effective process modelling and design, but the available data in literature are rarely sufficient for process design [8]. Espinosa et al
[2] presented the experimental data for a fluidized bed dryer that was used to recover
the solvents at three different temperatures of 40, 60 and 80oC. In their work,
Espinosa et al [2] used two types of batch fluidized bed dryer to experimentally study
the drying of hexane and ethanol from the neem seeds. The ground neem seeds were
wetted separately with hexane and ethanol. The main objective of this work is to
present a mathematical model for simulating the drying characteristics and the
temperature distribution in a fluidized bed dryer. The proposed model is then
validated using the experimental data from Espinosa et al [2].
2. Mathematical Model

Assuming that the solid particles are regarded as spherical, the heat loss to the surroundings is negligible and the gradients of concentration, temperature and pressure are ignored, the mathematical models, which describe the interchange of mass and energy between the solid and gas phases, are as presented below: Unsteady state mass balance for the solid and gas phases are:
Journal of Engineering Science and Technology APRIL 2007, Vol. 2(1)
Unsteady state energy balance for the solid and gas phases are: Integrating Eq. (1) at initial condition, X = X0, gives the drying time, where f(X)is the normalized drying rate describing the drying characteristics [9]. Also, integrating Equation (2) at initial condition, NV = NV0 gives the drying rate in terms of time. Substituting Equation (10) in Equation (6) and integrating, at the initial condition, T = Ts0, gives the temperature distribution in the solid, t + T ⎜1 − EXP − t − EXP(− k′t)⎤⎥ A close look at Equations (9) and (11) will reveal that f(X), hC, hD, NV0 and k′ as well as the physical properties of air and solvent must be specified. Generally, normalized drying rate [9-11] is defined as follows: f ( X ) = 1.0 − EXP(− N where, NtG is the number of gas transfer units in the bed. Hodges [10] approximates it as, where a and b are constants.
Journal of Engineering Science and Technology APRIL 2007, Vol. 2(1)
Drying of Neem Seeds in Fluidised Bed 25 Alternatively, NtG could be approximated by a polynomial function, = c + c X + c X + c X + . + c X where c’s and n are coefficients and order of the polynomial respectively. The heat transfer coefficient, hC, is calculated from the equation: where Nu is the Nusselt number, Re is the Reynolds number, Pr is the Prandtl number, p and r are the correlation constant and exponent respectively. The mass transfer coefficient, hD, is calculated using the Chilton-Colburn analogy ρ ⎠⎝ C BLM ⎠⎝ Sc ⎠ The gas bulk mean concentration, CBLM, is expressed as [12], The vapour fraction at solid and gas phases are defined as, where VP is the vapour pressure. Since the solvent is not present in the drying medium, yg = 0 The vapour pressure of the solvents at the solid surfaces are expressed by The latent heat of vaporisation of solvent (ethanol and hexane) is expressed as linear function of their corresponding temperature at melting and boiling points [14].
Journal of Engineering Science and Technology APRIL 2007, Vol. 2(1)
Also the physical properties of air (heat capacity, density, viscosity and thermal conductivity) as a function of temperature were obtained from Welty et al [12] in tabular form. The physical properties data were fitted with a polynomial using Microsoft Excel. The order of the polynomial was varied from 1 to 5 and the corresponding correlation coefficient (R2) obtained. The polynomial with the highest R2 is used for the computations. As mentioned in the introduction the experimental data were obtained from Espinosa et al [2]. These data and other relevant inputs are summarized in Tables 1, 2 and 3. Table 1. Physical Properties of Neem Seeds.
De-oiled seed
Exhausted seed
Description
Large Large
Table 2. Data Used for Drying of Neem Seed Treated with Hexane.
Tg,in (oC) Ts0 (K) ms (kg) ug (m/s) X0 (kg/kg)
Table 3. Data Used for Drying of Neem Seed Treated with Ethanol.
Tg,in (oC) Ts0 (K) ms (kg) ug (m/s) X0 (kg/kg)
A Fortran 77 programme was written to solve the model Eqs. (9) and (11). 3. Results and Discussion
The best polynomials that fitted the physical properties of air are shown in Table 4. These correlations are valid for the temperature range 290 to 370 K. Except for the heat capacity data with a correlation coefficient of 0.9999, the correlation coefficients
Journal of Engineering Science and Technology APRIL 2007, Vol. 2(1)
Drying of Neem Seeds in Fluidised Bed 27 for other properties are 1.0000. This means that the polynomials fit the physical property data for air very well. Table 4. Correlations for Heat Capacity, Density, Viscosity and Thermal
Conductivity of Air.
Property Correlation
. − 0.1447T + 0.0003T The experimental data was correlated assuming the Hodges’ Equation, Eq. (13), and a polynomial function, Eq. (14). The results are shown respectively in Tables 5 and 6. These Tables indicate that, for a given solvent and temperature, the polynomial function gave a higher correlation coefficient than power law equation by Hodges [10]. Consequently equation 14 was used for the computation of drying rate. Table 5. Correlation constants for Hodges Equation.
Solvent Temperature
Table 6. Correlation constants for Polynomial Function- Eq. (14)
The values of the initial drying rate (Nv0) and the drying rate constant (k′) are presented in Table 7. From this Table it can be seen that the drying constant increases
Journal of Engineering Science and Technology APRIL 2007, Vol. 2(1)
with temperature. This is in agreement with Arrhenius theory on the variation of rate constant with temperature [15, 16]. Table 7. Drying Kinetics (Drying Characteristics)
Temperature
(kg/m2sec)
Figures 1 and 2 depict the drying rate curve and the temperature distribution for Hexane treated solids whilst those for Ethanol treated solids are shown in Figures 3 and 4. In these figures the experimental values are shown as symbols while the model predictions are shown as solid lines. At low solvent content (X < 0.4kg/kg) the model prediction is not very accurate especially for the ethanol treated solids. Otherwise the model fairly predicts the experimental data adequately. Except for hexane treated solid at 60oC, the model under predicts the solid temperature for time less than 500 seconds. At higher values of time, the model prediction is fairly accurate; the level of accuracy increasing as the time increases. The temperature curves (Figs. 2 and 4) also showed an initial drop. According to Espinosa et al [2] this is an indication of fast rate of drying. Within this region the model prediction is not very accurate. Fig.1. Drying Rate Curve of Hexane Treated Solids.

Journal of Engineering Science and Technology APRIL 2007, Vol. 2(1)
Drying of Neem Seeds in Fluidised Bed 29 Fig. 2. Temperature Curve of Hexane Treated Solids.
Fig. 3. Drying Rate Curve of Ethanol Treated Solids.

Journal of Engineering Science and Technology APRIL 2007, Vol. 2(1)
TEMP (EXP. AIR AT 80 oC)TEMP (MOD. AIR AT 80 oC) TEMP (MOD. AIR AT 60 oC)TEMP (MOD. AIR AT 40 oC) Fig.4. Temperature Curve of Ethanol Treated Solids
4. Conclusions
A model for simulating the drying of neem seeds in a fluidised bed has been presented. Before solving the model equations, the physical property data for air (specific heat, viscosity, density and thermal conductivity) as well as the number of gas transfer in the bed were fitted with a polynomial. The drying simulation results show that there was a good agreement between the experimental values and the corresponding model predictions except during the initial period. References
1. Schmutterer, H. (1995). The neem tree azadirachta indica a. juss. and other meliaceous plants: Source of unique natural products for integrated pest management, medicine, industry and other purposes. VCH Verlagsgesellschaft, Weinheim, 326-365. 2. Espinosa, R., Lagerstedt, J., Nyman, T. & Martinez, J. (2002). Drying of ground neem seeds in a fluidized bed dryer. Proceedings of the 13th International Drying Symposium (IDS’ 2002), Beijing, China, vol. B, 1360. 3. Traub, D.A. (2003). The drying curve. Process Heating Magazine. Business
Journal of Engineering Science and Technology APRIL 2007, Vol. 2(1)
Drying of Neem Seeds in Fluidised Bed 31 4. Foust, S.A., Wenzel, L.A., Llump, C.W.& Anderson, L.B. (1980). Principles of Unit Operations (2nd ed.). New York: John Wily & Sons. 5. Butler, W. H. & Clark, D. (2000). Cascade vertical drying. Proceedings of 12th International Drying Symposium ( IDS 2000), No. 41, 1 – 10. 6. Coulson, J.M., Richardson, J.F., Backhurst J.R. & Harker, J.H. (1991). Chemical Engineering, vol. 2 (4th ed.). Oxford: Pergamon Press. 7. Peglow M., S. Heinrich, E. Tsotsas & L. Mörl (2004). Fluidized bed drying: influence of dispersion and transport phenomena. Proceedings of the 14th International Drying Symposium (IDS 2004), São Paulo, Brazil, 22-25 August, vol. A, 129-136 8. Maroulis, Z. B. & Saravacos, G. D. (2002). Modelling, simulation and design of drying processes. Proceedings of the 13th International Drying Symposium, vol. A, 38. 9. Stenzel, M., Motta Lima, O.C., Pereira, N.C. & Mendes, E.S. (2003). Generalization of drying curves in conductive/convective drying of cellulose. Braz. J. Chem. Eng., 20 (1), 81 - 86. 10. Hodges, C.R. (1982). Laboratory drying study and applications for paper machine 11. Treybal R.E. (1980). Mass-Transfer Operations (3rd ed.). Auckland: McGraw- 12. Welty, J. R., Wicks, C. E. & Wilson, R. E. (1976). Fundamentals of Momentum, Heat and Mass Transfer (2nd ed.). New York: John Wily & Sons. Isidoro, M. (2003). Vapour Pressure of Solvents, http://imartinez.etsin.upm.es/dat1/ePV.htm Isidoro, M. (2003). Liquid Properties. http://imartinez.etsin.upm.es/dat1/eLIQ.htm 15. Octave L. (1988). Chemical Reaction Engineering. New Delhi: Wiley Eastern 16. Susu, A. A. (1997). Chemical Kinetics and Heterogeneous Catalysis. Nigeria:
Journal of Engineering Science and Technology APRIL 2007, Vol. 2(1)

Source: http://jestec.taylors.edu.my/Vol%202%20Issue%201%20April%2007/21-31%20Kuye.pdf