(36e) Study of Particle Growth and Particle Size Distributions in Batch Fluidised Bed Granulation

Fluidised bed granulation is a technique of size enlargement of powders, and finds application in chemical industry for handling raw materials, intermediates and products as well as for coating of particles in pharmaceutical industry. It is carried out as batch or continuous operation, and is differentiated as solution granulation or binder granulation depending upon whether or not a binder is used for the agglomeration of particles. The present study refers to solution granulation in batch fluidised beds.

 Experimental:

 Experiments are conducted in a fluidised bed of 140 mm i.d. with provision of feeding the fluidising air at the desired temperature, and the granulation solution at the desired concentration and rate. A known quantity of     (-630, +500) mm is batch-fed into the fluidisation column. As fluidisation progresses, solution of known concentration is fed into the column at the desired rate. With granulation in progress, at known time intervals, an amount of solids equivalent to the amount of solute introduced during the time interval is scooped out from the bed for size analysis, and for maintaining solids inventory approximately constant. Table 1 gives details of the experiments.

 Results and Discussion:

Fig. 1 shows typical cumulative weight percent oversize plotted against the granulation time. It is noted from the data that the particles grow steadily in size during granulation with simultaneous depletion of the finer sizes. The distributions when plotted against dp/dpm  where dpm  is the median particle diameter, i.e. corresponding to cumulative 50% weight, the data converge to a single curve (Fig. 2) exhibiting ?Self-preserving' behavior reported earlier for granulation in spouted beds (Uemaki and Mathur, 1976).

(dp/dpm)

FIG.2.VARIATION OF GRANULE SIZE DISTRIBUTION WITH THE RATIO (dp/dpm)

The data further show that granulation is favoured with an increase in fluidising gas temperature, the solute concentration and the flow rate of feed solution, the fluidising gas velocity or a decrease in the initial charge of the solids. A comparison of the data for the different materials indicates that granulation is favoured most by urea and least by sodium sulphate.

 Visual observation of the granulation process and the granule structure suggests that the initial growth of the granules is by agglomeration, while at granulation times exceeding 120 min., the growth is by layering.

 The experimental distributions are compared with normal, log-normal, Weibull, Rosin-Rammler and Mugele-Evans distributions, and found that the data are well represented by log-normal distribution. The volume-surface-mean-diameter, d₃₂ is related to the distribution parameters as (Irani and Callis, 1963).

d₃₂=ln d₅₀ + 2.5 ln² s                                                                                                  ________ (1)

Fig. 3 shows the variation of d₃₂ with the granulation time for the materials covered in the study.

An empirical relation for d₃₂ is developed to give

                                                                                               __________ (2)

where the growth coefficient, k is related to the operating variables as

 
                                           __________ (3)

 where θ depends on the type of material. K values estimated for the experimental data reported in literature    (Bin et al., 1985; Smith and Nienow, 1983) correspond to the values obtained for the data of the present study.

FIG.3: VARIATION OF d₃₂ WITH GRANULATION TIME FOR DIFFERENT MATERIALS.

Table 1: Details of experiments

a) Physical properties of granulating liquids

b) Range of variables Covered

Feed size of material          :               -630, +500mm

Initial charge                        :               500g, 1000g

Fluidising air Velocity       :               0.8 ?1.5 m/s

Granulating liquid rate      :               20, 25, 50 cc/min

Fluidising air temperature :               343, 363, 383, 398 K

Fluidised bed temperature :               306,320,342,355 K                

Notation

 d₃₂           -               Volume-Surface mean diameter, mm

d₅₀           -               Particle size corresponding to 50% oversize in log-normal distribution, mm

d_pm          -               Median particle diameter, mm

k              -               Particle growth coefficient, min^-1

m_b           -               Net input rate of solute, g min^-1

t          -                Granulation time, min

T             -               Air inlet temperature, K

W            -               Weight of the feed, g

x              -               Concentration of granulating liquid (-)

q              -               Factor in equation (3)

s              -               Variance in log-normal distribution

r_l             -               Density of granulating liquid, g/cm³.

d_l              -               Surface tension of granulating liquid, m N/m.

m_l             -               Viscosity of granulating liquid, cp.

Literature Cited

Bin, A., J.Warych and R. Kumorowski, ?The Batch granulation process in a fluidized bed', Powder Technology., 41, 1(1985)

 Irani Riyad, B and C.F. Callis, ?Particle size measurement interpretation and application', John Wiley, N.Y. (1963)

 Smith.P.G. and A.W. Nienow, ?Particle growth mechanisms in fluidised bed granulation I & II', Chem. Engg.Sci., 38, 1223-1232 (1983)

 Uemaki, O.and K.B. Mathur, ?Granulation of ammonium sulphate fertilizer in a spouted bed', Ind. Eng. Chem. Process Des Dev., 15, 504(1976)

^*

Communication Email: kkmurthy@iitm.ac.in

⁺Retired Professor, Department of Chemical Engineering, Indian
Institute of
Technology Madras , Chennai 600 036,
India