Instant Beans Obtained from Simultaneous Processes of Drying and Size Reduction

XXII IACChE (CIIQ) 2006 / V CAIQ AAIQ Asociación Argentina de Ingenieros Químicos IACCHE - Interamerican Confederation of Chemical Engineering INSTANT BEANS OBTAINED FROM SIMULTANEOUS PROCESSES OF DRYING AND SIZE REDUCTION M. Villanueva1, J. Martínez2 and R. Gamero1∗ 1Faculty of Chemical Engineering, National University of Engineering (UNI) Avenida Universitaria, PO BOX 5595, Managua, Nicaragua C.A. E-mail: r[email protected].se 2Department of Chemical Engineering and Technology, Royal Institute of Technology Teknikringen 26, 100 44 Stockholm, Sweden E-mail: [email protected] Abstract. The kinetics of the drying of beans in simultaneous processes of contact drying and particle size reduction has been experimentally studied, and the particle size distribution of the final dried solid has been analyzed. The work is focused on some operating parameters, such as drying temperature, stirring speed and bean load mass, that influence on the dehydration of beans during those simultaneous processes. Drying rate curves were developed from the experimental data of moisture content changing with time, and particle size distribution of the final dried solid was analyzed. The results reveals that drying temperature, stirring speed and mass load are crucial for both the drying processing and the quality of the final product. The particle size reduction itself improves the drying rate as well since the heat and mass transfer area increases with the grain destruction. The final product contains values between 40-50% of suitable size for commercial proposes. The simultaneous process of two operations shows advantages as energy saving and fast processing that would be greatly significant once brought to industrial scale since the milling process needed afterwards would be reduced. ∗ To whom all correspondence should be addressed XXII IACChE (CIIQ) 2006 / V CAIQ AAIQ Asociación Argentina de Ingenieros Químicos IACCHE - Interamerican Confederation of Chemical Engineering Keywords: Dehydration, Particle size distribution, Contact drying. 1. Introduction Drying of grains is a critical post harvest operation to reduce the moisture content for preservation and storage purposes. Besides drying as a post harvest operation, dehydration is also an important processes applied to food products to allow long-time storage, minimize packing and reduce shipping. Dehydration also allows flavor enhancement and ease of preparation for final consumption. Dehydration of beans is a process found frequently in food industry since beans are one of the main foodstuff consumed by the population. Beans (phaseolus vulgaris) are highly rich in proteins, carbohydrates and minerals and low in lipids content. Serrano and Goñi (2004) present a detailed data of beans nutrients compositions. Beans are found in the market in different presentations, e.g. canned beans, milled beans, and even fresh boiled beans. However, dehydrated beans provide particular safety and health warranties since the microorganism activity is significantly diminished with the moisture reduction. Dehydration process frequently is followed by a milling process; for example, in the production of powdered dry products as instant corn mass, instant beans and diverse cereals. The combination of dehydrated beans with different types of spices can provide a preservative-free product with enhanced quality, a great nutritional content, and with diverse consume options due to its powdered presentation like flour or milled cereal. From the production point of view, dehydration of beans provides a great opportunity to improve quality, diversify the variety of products and even innovate in both the product and process. This work is focused on obtaining dehydrated beans in simultaneous contact drying and size reduction resulting from the mechanical agitation associated to this type of drying processes. XXII IACChE (CIIQ) 2006 / V CAIQ AAIQ Asociación Argentina de Ingenieros Químicos IACCHE - Interamerican Confederation of Chemical Engineering 2. Contact Drying The most common drying mechanism is convection, in which heat and mass transfer take place between the solid surface and a hot inert gas. In the particular case of contact drying, heat transfer primarily occurs by conduction; therefore, air is needed only to remove the vaporized moisture (Richard and Raghavan, 1984). This makes contact drying much more efficient than convective drying. From different works developed on contact drying, it is known that commonly heat is supplied from the dryer walls. Thin layer of grains can be dried over a hot plate faster than using hot air. However, the efficiency can decrease as the layer becomes thicker because of the over heating of grains in the lowest layer and the poor distribution of heat to the upper layers, which revels a drawback in heat transfer throughout a grain fixed bed. Chancellor (1975) added mechanical stirring to overcome the heat distribution problem along the entire grain bed. Several works have examined the subject making special emphasis in agitation aided heat transfer; for instance, Schlünder and Mollekoft (1984) introduced the influence of mechanical agitation in vacuum contact drying of particulate materials; Tsotsas and Schlünder (1986) developed a mathematical model that describes the mechanical agitated drying of particulate material in presence of air; Tórrez and Martínez (1994) extended the model to grains immerse in a particulate medium. Although the mechanical agitation improves the contact drying process, the selection of suitable agitation elements is important to avoid grain destruction. Ortega (2001) recommends either low agitation velocities or special propeller design for non-spherical grains such as corn and beans. 3. Drying Kinetics Basic drying theory illustrates the drying kinetics with the drying and the rate drying curves shown in Figure 1 (Strumillo and Kudra, 1986; Okos et al, 1992). The drying curve results from data of moisture changing with time and is generally divided into two periods: A constant rate period (B-C), in which the evaporation occurs freely without resistance of the solid, and a falling rate period (C-E), in which the free moisture content is removed completely (C-D) followed by the removal of internal moisture XXII IACChE (CIIQ) 2006 / V CAIQ AAIQ Asociación Argentina de Ingenieros Químicos IACCHE - Interamerican Confederation of Chemical Engineering against the solid resistance (D-E). Toledo (1991) presents methods to predict drying times from drying rate data for both constant and falling periods displayed in Figure 1. (a) (b) Fig.1. Typical graphic representation of drying kinetics: (a) Drying curve, (b) drying rate curve. The moisture content as a function of time is given by: ! x(t)=m(t)"md md (1) where x is the moisture content, t is the time, and md is the mass of the dry solid. The drying rate plotted in Figure 1(b) is calculated from the derivative of Figure 1(a) as: ! NA="md As dx dt (2) where NA is the evaporation flux or drying rate and As is the free surface exposed to the gas stream. For through drying in fixed beds and contact drying in agitated beds, it is referred in terms of volume units: ! NV="md V dx dt (3) where V is the total volume of the granular bed. 4. Experimental Work The experiments of contact drying were performed in the set-up shown in Figure 2. The apparatus consists of a cylindrical open dryer. XXII IACChE (CIIQ) 2006 / V CAIQ AAIQ Asociación Argentina de Ingenieros Químicos IACCHE - Interamerican Confederation of Chemical Engineering Heat is supplied from a hot plate located in the bottom of the cylinder. A chamber below the heating plate contains hot oil, which flows continuously pumped from a tank. The oil is heated in the tank by three electrical resistances of 1 kW automatically controlled to keep a constant temperature in the dryer hot plate. 1. Hot oil tank 8. Hot oil chamber 2. Electrical resistances 9. Hot oil feed pipe 3. Pump 10. Oil return 4. Dryer 11. Frequency inverters 5. Mechanical stirrer 12. Temperature control 6. Stirring blades 13. Energy supply 7. Hot plate T Thermo-resistance Fig. 2. Experimental set-up for contact drying of grains Two frequency inverters govern the pump and the mechanical stirrer with an electrical motor of 0.75 kW. The stirring velocity is controlled setting the rpm in its respective frequency inverter. 4.1. Experiment Description After starting up the experimental set-up, the beans were boiled slowly. The oil was pumped from the storage tank to the oil chamber. Afterwards, the temperature was set XXII IACChE (CIIQ) 2006 / V CAIQ AAIQ Asociación Argentina de Ingenieros Químicos IACCHE - Interamerican Confederation of Chemical Engineering in the automatic temperature control, and the electrical resistances were connected to the power supply. After reaching the set point temperature in the hot plate, the mechanical stirrer was turned on, and the boiled beans were loaded in the cylindrical dryer. Small samples were taken every 5 minutes until the presence of dry powder was observed as an indication that the entire beans bulk has been completely dried. Each sample removed from the bulk was weighed in an analytical balance, completely dried in an oven and weighed again. To assure a complete drying, all the collected and weighed samples were stored in the oven for 24 hour before being weighted the second time. The moisture content for each elapsed time was calculated according to: ! x=mw"md md (4) were mw is the mass of wet solid (removed from the bulk), and md is the mass of the dry solid (after drying in the oven). The difference of mass is the amount of water removed from the sample. To compute the drying rate curve, Eq. (3) is approached by discretization as: ! NV="md V xi+1"xi #t $ % & ' ( ) (5) 4.2. Particle Size Analysis Particle size distribution was performed applying a sieve analysis to the dried beans removed from the contact dryer through a set of sieves with decreasing openings. After the shacking was done, the mass of dried beans retained in each sieve was weighed. Particle size distribution curves were created in such a way that accumulative percentage of passing (finer) particles was plotted against the particle diameter in a semi-log scale. The percentage retained in each sieve is given by: ! pj=100 mj mt " # $ % & ' with: ! j=1,2,...n (6) where n is the total number of sieves. The accumulative percentage of fine particles is then: ! pacc =100 "pj j=1 n # (7) XXII IACChE (CIIQ) 2006 / V CAIQ AAIQ Asociación Argentina de Ingenieros Químicos IACCHE - Interamerican Confederation of Chemical Engineering 5. Results and Discussion 5.1. Drying Kinetics To examine the drying kinetics of dehydration of beans, four full experiments were performed. Drying curves were created by plotting the moisture content calculated by Eq. (4) against the corresponding sampling time. As observed in Figure 3, the curves show a clear descending trend. At a first sight, it can be observed that the curves provide valuable qualitative and quantitative information. Drying rate curves were created computing the evaporation rates in Eq (5). To avoid spread points, all the drying curves were smoothed making a polynomial adjustment of 4th degree with values of R2 almost equal to 1. (a) (b) (c) (d) Fig. 3. Drying curves of dehydrating beans XXII IACChE (CIIQ) 2006 / V CAIQ AAIQ Asociación Argentina de Ingenieros Químicos IACCHE - Interamerican Confederation of Chemical Engineering Drying kinetic need to be analyzed in both Figure 3 and Figure 4. Influence of agitation velocity (rpm) is shown in Figures 3(a) and 3(b). Temperature is more influencing that the rpm; therefore, Figures 3(a) and 3(b) barely show a small difference between them in the last period. Note that at 9 rpm the curve keeps slowly falling to suddenly fall near zero moisture content as shown in Figure 4(b). Figure 3(a) shows at the beginning a tendency of constant rate drying period. The maximum value of drying rate is higher with 9 rpm. The higher the agitation velocity, the faster the moisture release from the inner bulk porosity and the internal grain moisture due to the grain destruction, which is an advantage in contrast to the recommendation given by Ortega (2001). Influence of temperature can be observed in Figures 3(c) and 3(d). In this case, it is quite clear that at higher temperatures the solid is dried faster. Figures 4(c) and 4(d) also display the influence of temperature. Note that at higher temperature, the drying rate is much higher. This Figures also shows the particular behavior of the drying rates curves. Although the curve in Figure 4(d) is more pronounced, both curves present a drying rate reduction followed by a sudden increase until reaching a critical point. The explanation of this phenomenon is that at the beginning, temperature provokes a drastic reduction of free moisture content, and then, when the grain destruction begins, internal moisture starts being released; therefore, the evaporation flux increases. At high temperatures (see Figure 4(d)) the phenomenon is more noticeable. Influence of load mass is displayed comparing Figure 3(a) and 3(d). The drying rate is faster at smaller amounts of load mass. Figures 4(a) and 4(d) also illustrate this fact. The behaviors of the drying rate curves are also different. With an amount of load mass of 1 kg, the evaporation flux decreases gradually since agitation releases the free and internal moisture. On the other hand, with a load mass of 2 kg, the agitation is not able to destroy the grains from the beginning; therefore, the free moisture is released, and the evaporation flux tends to decrease until the grain destruction begins later, and the evaporation flux eventually improves. XXII IACChE (CIIQ) 2006 / V CAIQ AAIQ Asociación Argentina de Ingenieros Químicos IACCHE - Interamerican Confederation of Chemical Engineering (a) (b) (c) (d) Fig. 4. Drying curves of dehydrating beans 5.2. Particle Analysis The accumulative percentage of fine particles is shown in Figure 5. The experimental work demonstrated that in all cases a particle size smaller than 0.2 mm was achieved between 40 and 50%. During the experimental work with conditions reported in Figure 5(a) and 5(b) the drying occurred faster because both experiments correspond to the smallest amount of load mass (1 kg). Once dried, the propeller threw away an important amount of dusty dry material. Therefore, probably in all case the percentage reached would have been near 50%. This finding indicates that the simultaneous drying and size reduction could save energy if 50% of the material already reached a suitable size, and only half of the total mass would need extra milling process.