REVIEW OF LITERATURE
|
Through the various literature that are available, it has been found that little work has been done on drying of liquid food and food pastes and no work has been done on coating of food paste on solids using a spouted bed.
The minimum spouting velocity (Ums) is to spouting as Umf, minimum fluidization velocity, is to fluidization. However, whereas Umf is predictable from the fluid and the particle properties alone, Ums depends on the vessel geometry and bed height. Morgon and Littman (1981)* presented a new general equation for Ums, on the assumption that at minimum spouting superficial velocity at the top of the annulus is equal to the Ums. While other parameters were same as required by Mathur - Gishler equation (1955). The advantage of the former equation is that it gives recognition to fluid viscosity, a variable which is ignored by Mathur-Gishler equation.
Ums = (dp/dc) (di/dc)1/3 (2Hg (r s–r f)/ r f)�
Both Mathur-Gishler and Morgon-Littman* equations predict that for a given ratio of bed height H to bed diameter D, Ums decreases as D increases. However, for units with larger diameter example greater than 0.3m, Ums for the bed showed a linear dependence on bed height, than the square root dependence as given by Mathur-Gishler equation, which greatly underestimated Ums, while measured spout diameter and pressure drops were considerably greater and smaller, respectively than those obtained by correlations for fully round smaller diameter beds.
Bereton (1983)* and Kilkis (1983)* presented numerical analysis of gas flow in conical based and flat -based spouted beds, respectively. In Bereton’s equation quantitative deviations between experiment and mathematical model for gas flow in annulus could be explained by the neglect in the model of solids motion of the venture effect at the gas inlet and of the variation of spout diameter with bed level. In two region mode of Kilkis, spout shape was an additional output rather than an input, but the incorrect initial assumption that the longitudinal pressure profile in the spout is similar to that in a fluidized bed requires iterative correction by pressure distribution in annulus which is actually generated by the analysis. The Kilkis result have yet to be compared quantitatively with the experiment.
2.2 Drying of Granular Materials
2.2.1Irregular material
Ghaly et al. (1974) proposed a simple experimental simulation technique to predict the exit moisture content of dried wheat and the extent of damage done to the functional properties of grains using spouted bed drier. Here monolayer wheat was dried under a constant velocity, isothermal air. A significant increase in germination has been observed as the result of physical changes in the seed coat rather than the thermally induced chemical changes in the grain as a whole.
Zuritz and Singh (1980) developed theoretical model based on mass energy transport equation for a continuous media to describe liquid diffusion controlled drying of cereal grains in batch operated well mixed spouted bed. The model was tested using rough rice and good agreement was obtained between the experimental and theoretical model.
The spouting of paddy was observed by Drona et al. (1982) by recording the air pressure drop with its flow rate. The variation of moisture content of paddy with drying time for three different inlet temperatures (45, 56 and 70�C) of the drying air has been reported for two different bed depths (0.3 and 0.45 m).
2.2.2 Regular shaped material
Nelson and Gay (1971) spouted bed fluid and particle transport process for coarse biological materials (peanuts) were studied to evaluate functional relationships among pressure drop through the bed, flow rate and other relevant bed configuration, fluid and peanut properties. Pressure drops of static bed, transition phase and of fully developed spouting were found, diameter of inlet pipe was found to be the critical factor in performance of the spouted bed. For a given flow rate, pressure is inversely proportional to 0.6 power of inlet pipe diameter and in fully developed spouting it is inversely proportional to inlet diameter squared.
2.3 Particle segregation in spouted bed
Presentation by Osamu Uemaki (1983)* and John Grace (1983)* stressed the importance of foundation development in determining the degree to which segregation by particle mass occurs. An underdeveloped fountain is one which just reaches the wall, and an over developed is one in which the particles strike the wall before bouncing back into the annulus. Thus, segregation of heavier particles occur. Grace (1983) showed that desegregation can be obtained by adding a conical baffle with its apex pointed downwards to deflect the heavier particles outwards from the centre of the fountain.
Uemaki (1983)* modified the coefficients and indices of Mathur-Gishler equation so that it could predict Ums of his binary sand mixtures, but modification on the index on H from 0.5 to 0.324 was in a direction opposite to that what has recently been observed by Ishikura et al. (1982) who found that for several binary particle mixtures that the power on H which best fits all the data was in excess of 0.5. Thus, showing that Ums which depends on H0.5 in Mathur-Gishler equation is an approximation which needs refining. He showed that problem of particle segregation could solved by increasing the discharge rate.
The use of draft tube offers additional degree of freedom for controlling the operation of such beds. Unlike the case of conventional spouting, there is no maximum spoutable bed height when a draft tube is used, and due to the absence of gas leakage from the draft tube to the annulus, the minimum spout nozzle flow to initiate stable solids circulation, and the pressure drop are considerably smaller than for ordinary spouting. Solids circulation and mixing rates are also significantly lower and solid residence time distribution considerably more uniform than in conventional spouted beds due to the draft tube.
Giok-Koen (1983)* used draft tube insert in a conical cylindrical single nozzle spouted bed for drying of unpeeled rice and concluded that 70% of the air to grain heat transfer typically occurred in gas recirculation zone between the inlet orifice and the draft tube inlet and the remaining 30% within the draft tube itself. While most of the mass transfer (initial moisture migration) occurred in the annulus. They suggested that more air could be directed into the annulus either by using a side gas outlet or by perforating the lower part of the draft tube. The latter recommendation was independently arrived at by Claffin -Fane (1983)*, who studied fluid solid mechanics of a fully porous draft tube, both their experimental and their mathematical model showed that a maximum in the annular gas flow typically occurs part way up. The bed due to permeation of gas from spout to the annulus at the bottom and a reversal further up the draft tube. It was found that pressure drop for the porous tube is less than for a spouted bed without the draft tube using same total gas flow rate. The gas reversal from annulus to spout can be prevented by confining the porous part of the draft tube to the level at which maximum annular gas flow occurs and making the rest of the draft tube non-porous. Such a draft tube composite is yet to be built and tested.
2.5 Drying of suspensions and solutions
Berquin (1961) proposed that drying of pastes suspensions or solution could be obtained in a spouted bed using seed granules as bed materials and injecting liquid phase with hot spouting gas at the base. A thin layer is deposited on the circulating particles as they pass through the liquid spray, which is dried by the action of hot gas as the particles travel up the spout and down the annulus. The granules produced are well rounded and of uniform structure.
Romankov and Rashkovskaya (1967-68) developed an indigenous methods for drying of solution and suspension in a bed of inert particles of spoutable size e.g. glass beads (3-6 mm dia) for applications where dried solids are ultimately required as fine powders. The solution containing 85% water is atomize in the lower region of bed spouted with hot air, deposits as thin film on the glass beads passing through the region and subsequently dries as granulation process. The film becomes fragile on drying and is knocked off the particles surface by inter particle collisions in the spout and is reduced to fine powder. This is then collected in overhead cyclone.
The technology for coating of particles in a spouted bed, whether by deposition from an evaporating atomized liquid solution as in pharmaceutical tablet coating or by high temperature thermochemical deposition as in the coating of nuclear fuel particles (<1 mm dia) with relatively thin layers (<0.2 mm) of either prolitic carbon or silicon carbide is well established. To achieve satisfactory control of such an operation, in which the appreciable volume occupied by the thick coating on both the prosthesis and the particle tends to progressively expand the bed, there must be a continuous feeding of uncoated particles followed by continuous withdrawal of variously coated particles. A cold simulation of such an operation, by feeding smaller particles into a bed composed of two sizes of particles and withdrawing product from any one of the several discharge solids, compositions of which would eventually occur in the case of actual thick coating operation would depend on the location of discharge pipe Epstien- Mathur (1982).
Axel Meison (1983)* developed another type of coating operations in which spouted urea granules were coated with elemental sulphur by injecting atomized liquid sulphur into the inlet spouting air. The object was to produce a slow release fertilizer. It was found that the increase in liquid sulphur flow improved the product quality and the atomizing (as oppose to total) air flow, decrease of which also improved the product quality.
Szentmarjay and Pallai (1989) studied the effects of operational and process conditions on drying output and product quality. Studies were performed using a specially designed laboratory scale spouted bed drier with tangential air inlet and an inner conveyor screw to maintain recirculation of inert particles. Effects of parameters influencing gas solid contact, e.g. inlet air velocity, recirculation of inert particles and chemical and physical properties of inert material were examined. Various test suspensions, including yeast and food products were used in preliminary experiments, which demonstrated the effectiveness of this drying method. Major operational and process parameters were then evaluated using 2 model materials (a suspension of cobalt carbonate and a pulp of zinc carbonate).
Szentmarjay et al. (1991) studied the effects of various factors on the rates of drying and abrasion in a mechanically spouted bed dryer with an inert packing. Glauber’s salt was used as a model material. Drying curves were obtained as a function of drying temperature and volume and revolution speed of inert screw. Abrasion experiments were made to determined loss of the coating as a function of time (can be described by exponential function). Results indicated that of the 3 processes in the dryer i.e., coating of inert particles, drying of coating and abrasion of dry coating abrasion required least time (8% of cycle, vs. 67 and 25% respectively, for coating and drying).
Schneider and Bridgwateh (1993) found that in granulation, coating or drying of slurries on inert particles, a liquid phase is added to a gas-spouted bed. The effect of liquid injection on spouting velocity, foundation height bed pressure drop, and stability of the spouting regime was examined in a 0.15 m dia bed with 4 types of inert particles, 2-5 mm dia size 1140-2880 kg/m3 in true density and 1.3 and 1400 mPa.s in liquid viscosity. Key factors are liquid content and the presence of cohesive forces due to liquid bonds between particles. In a spoutability chart the maximum spoutable liquid content is related to the ratio of internal force of a particle to the viscous or cohesive force exerted onto the particle by the liq. film. The spoutability chart serves to define the region of stable spouting. They also found that when liquid was injected in a bed the spouting behaviour could change dramatically. In the vicinity of maximum spoutable liquid content, the spouting regime became less stable with the fountain fluctuating in height and becoming less dense. In some cases violent eruptions of the bed surface or internal channelling of the spout to the wall of the dryer terminated the controlled spouting operation. When the combination of inert particle and liquid allowed spouting at Ve/Vb in excess of about 0.05, the annulus tended to be flooded and particle circulation almost ceased although a stable fountain could still be maintained.
Ochoa et al. (1993) carried out experiments to determine the effect of factors such as air and feed flow rates, inlet air temperature and bed height on overall heat, transfer coefficient, ha, using deionized water in a spouted bed of inert particles. The spouted bed had a conical bottom (semi vertical angle of 30�C) on which was mounted a cylindrical top (0.3 mm dia) spherical polypropylene beads (3.9 mm dia) were used as inert particles. Effects of adding a surfactant and a thickening agent to the aqueous feed on ha values were also reported. Values of ha for all the experiments at various operation conditions ranged from 1.2 to 5.4 kwm–2k–1.
Ochoa-Martinez et al. (1993) dried concentrated skim milk and non-homogenized whole milk in a spouted bed dryer, which consisted of spherical polypropylene beads (3.9 mm diameter) spouted with hot air. Release of dry skim milk from the particles did not take place; it adhered to the particles. Whole milk was successfully dried in this device and the powder released easily from the spouted bed. Additional experiments using casein and lactose solutions, separately and in different combinations, were carried out. When casein and lactose solutions were individually fed to the dryer, release of powder took place. It was observed that casein released more readily than lactose. When mixture of casein and lactose solutions were fed to the dryer, the resulting powder failed to release. The structure of the deposit formed on the surface of the polypropylene beads was examined using SEM. Since dried whole milk releases readily from the surface of the inert particles, it appears that fat determined the structure of the solid deposit left on the particle. Important functional properties of dried whole milk obtained by spouted bed drying were compared with those of spray-dried whole milk.
Tia et al. (1995) studied the performance of a laboratory scale jet spouted bed (JSB) for drying rice flow slurry was studied. The bed consisted of ceramic balls (5.28 mm dia) and the rice flow slurry was sprayed onto the moving particle surface near the air inlet part. All the experiments were carried out at the jet spouting regime. This regime has high bed void fraction and violent movement and collision of bed particles. As a result, the dried product layer is attired from the particle surface as a fine powder and entrained from the bed by the spouting air. The experimental results showed the effects of static bed height, inlet air-flow rate, temperature, feed concentration and flow rate on the outlet air temperature, thermal efficiency, mean particle size and moisture content of the product. A simple mathematical model, which is based on the conservation of mass and energy equations was developed, predicted results agreed well with those obtained from the experiment.
Szentmarjay et al. (1995) found that suspension, slurries and paste like materials can be dried in the mechanical spouted bed (MSB) dryer with inert packing. The circulation of the inert particles, characteristics of classical spouted beds, is provided with a houseless, screw conveyor mounted in the vertical axis of the bed. Radioactive isotopic tracer technique was used for the measurement of the cycle time distribution (CTD) of the spherical inert particles as a function of the operational parameters of drying. The variances (s 2) of the CTDs and the particle velocity in the various zones of the MSB dryer were calculated. The circulation of the inert particles can be characterised by nearly plug flow. According to the physical model of the drying on inert packing, the heat and mass transfer, coefficient were calculated. Due to relatively uniform film-like wet coating formed on the surface of the spherical inert particles, the drying process may be characterised with constant rate of drying. A method was elaborated for calculation of drying time, hereby the partial processes of drying on inert particles can be synchronised.
Liu and Chen (1996) studied the spouted bed pneumatic drying with glass beads as inert particles and the effects of temperature and velocity of the air blast on the residence time, the water content, and the gas solid heat transfer coefficient of the drying material were determined. An empirical relationship between Nu (Nusselt no.) and Re (Reynold no.) was obtained by least square regression data fitting.
Szentmarjay et al. (1996) have found that the continuously operated mechanical spouted bed (MSB) dryer of high evaporative capacity can be advantageously used to produce fine powder from paste like materials slurries, suspensions and sludges. Due to the thin layer formed on the surface of the spherical inert particles intensive heat and mass transfer occur and the drying process takes place in the constant rate period. Steady state drying conditions can be achieved when the total operational time of partial processes of inert bed drying does not exceed the cycle time of the inert particles. They have developed a calculation method to control the drying process.
Passos et al. (1997) have stated that drying of paste like materials (such as animal blood, banana mash, umbu pulp, tomato pulp, skim milk suspensions, alumina suspensions) can be performed well in spouted beds of inert particles. In their work the drying performance of pastes in conical spouted beds is analysed as a function of column dimensions, fluid flow characteristics and paste properties. Simulated data on fluid flow together with the experimental results in drying of different paste like materials are presented and discussed to provide criteria for the design of conical spouted bed dryer for suspensions.
Introduction Review material and methods results and discussion summary
