This article was downloaded by: [University of Arizona]

On: 10 July 2013, At: 00:18

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Drying Technology: An International Journal

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/ldrt20

Modelling and Simulation of a Direct Contact Rotary

Dryer
a b c b
C. Shene , F. Cuhillos , R. Perez & P.I. Alvarez
a
Univenidad de La Fmntera, Chemical Engineering Department, Casilla 54-D, Temuco, Chile
b
Univenidad de Santiago de Chile, Chemical Engineering Department, Casilla, Santiago,
10233, Chile
c
Pontificia Univenidad Católica de Chile, Chemical and Bioprocessing Engineering
Department, Casilla 306 Correo 22, Santiago, Chile
Published online: 07 May 2007.

To cite this article: C. Shene , F. Cuhillos , R. Perez & P.I. Alvarez (1996) Modelling and Simulation of a Direct Contact Rotary
Dryer, Drying Technology: An International Journal, 14:10, 2419-2433, DOI: 10.1080/07373939608917214

To link to this article: http://dx.doi.org/10.1080/07373939608917214

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 DRYING TECHNOLOGY. 14(10). 2419-2433 (1996)

TECHNICAL NOTE

MODELLING AND SIMULATION O F A DIRECT CONTACT ROTARY

DRYER
Downloaded by [University of Arizona] at 00:18 10 July 2013

C. Shene', F. Cuhillos', R. P*rez', P.I. Alvarez'

'Univenidad de L a Fmntera, Chemical Engineering Department, Casilla

54-D, Temuco, Chile
'Univenidad de Santiago de Chile, Chemical Engineering Department,
Casilla 10233, Santiago, Chile
'Pontificia Univenidad Cat6lica de Chile, Chemical and Bioprocessing
Engineering Department, Casilla 306 Correo 22, Santiago, Chile

Key Words and Phrases: heat and mass transfer, meal drying, drying kinetics,
volumetric heat transfer coefficient, residence time.

ABSTRACT

A mathematical model able to predict solid and drying gas temperature and
moisture content axial profiles along a direct contact rotary dryer was developed.
The study was focused on the drying kinetics based on phenomenological models.
Two different drying mechanisms in the decreasing drying rate period were tested:
proponional to the unbound moisture content and moisture diffusion inside the
panicle. Experimental data collected in a pilot-scale direct contact rotary dryer was
used to validate the model. Soya and fish meals were used as drying material.

INTRODUCTION

Direct contact rotary dryers are commonly used in the prccess industry
because they are simple to operate and can handle large volumes of particulate

Copyright 0 1996 by Marcel Dckker. Inc

 2420 SHENE ET AL.

solid materials. The performance of this operation is mainly affected by, the heat
transfer rate from the drying gas to the wet solid panicles and by the retention
time of the solid panicles inside the dryer.

Although the classic results presented by Friedman and Marshall (1949)

and Mc Comick (1962) are useful for designing purposes, they are not suitable
either for process optimization or for temperature or moisture axial profiles
Downloaded by [University of Arizona] at 00:18 10 July 2013

prediction. In order to obtain a more accurate description of the process, heat,

mass and momentum transfer between the drying gas and solid panicles has to be
considered. A proper prediction of moisture wntent and temperature pmfiles is
important for an efficient drying operation. specially when the product quality
depends on the moisture and temperature levels reached by the solid inside the
dryer. In these cases, the inlet drying gas temperature or the retention time of the
solid panicles inside the dryer can be adjusted to keep the product quality between
acceptable levels by means of changes in the feed rate or in other operating
conditions.

In the food industry drying operation is important for the global economy
of the process and for the final quality of the product. Product quality wuld be
affected by the high temperatures used in the rotary drying process, mainly due
to deterioration reactions (i.e. essential aminoacids, lipids). Tkese reactions that are
difficult to predict and prevent. affect the final quality of the product
independently of the raw material freshness.

As the market is driven more by the quality, in the fish meal industry
indirect contact rotary dryers are shifting direct contact dryers, with the aim of
improving the fish meal quality. Published research on fish meal drying is scarce
due to the hermeticism of this industry. Myklestad (1963) showed the effects of
high temperatures in the fish meal drying, panicularly those of inlet drying gas on
both moisture and temperature solid axial pmfiles. A maximum solid temperature
 DIRECT CONTACT ROTARY DRYER 242 1

is reached inside the dryer whose position depends on the inlet air temperature. On
the other hand, the controlling mechanism of the moisture movement inside the
fish meal particles during drying is of diffusional kind. Mathematical modelling
of the fish meal drying based in this mechanism shows a good correlation with the
experimental data obtained in a tray dryer (Alvarez et al. 1988).

In the direct contact rotary drying process, mass lmnsfer cannot be isolated,
Downloaded by [University of Arizona] at 00:18 10 July 2013

as the removal of moisture depends on the heat transfer. Therefore, a proper

selection of the drying kinetics could become a critical point in the process
simulation. Myklestad (1963) develops a procedure for predicting the solid
moisture content along the dryer length assuming the drying gas temperature is
linearly related to the solid moisture content.

A more fundamental approach is used by Sharples et al. (1964); this

approach requires the simultaneous solution of four differential equations
corresponding to the energy and mass balances of both solid and drying gas
phases. For solid drying rate estimation empirical coefficients are needed.

Thome (1979) presents a model where the mass transfer is described

through a vapour diffusion model. Drying pumice experiments agree well with the
model predictions.

A simplified model for drying in a direct contact rotary dryer is presented

by Kisakiirek (1982). In this model it is assumed that the solid temperature
remains constant during the process and sensible heat effects are neglected. This
model performs poorly at low solid moisture contents.

The theoretical modelling of direct contact rotary drying process proposed

by Kamke and Wilson (1986) is validated with wood drying experiments, and the
extent of drying estimated by means of an empirical model. Comparisons of
simulation results with experimental drying data obtained in a large-scale direct
 2422 SHENE ET AL.

contact rotary dryer is favourable with 22.2% root mean square error. In this work,
emphasis is put on the drying kinetics and its effects on the solid temperature
profiles and the thermal degradation of the product, as it normally happens in
direct contact rotary dryers operated with high inlet drying gas temperatures (over
loo *C).

Studies of dynamic simulation of a direct contact rotary dryer have been

Downloaded by [University of Arizona] at 00:18 10 July 2013

presented (Douglas et al. 1993. Wang et al. 1993). Douglas et al. (1993) run
simulations to predict the outlet conditions in industrial dryers used in sugar
prwessing. Differences around 10% from industrial data were found.

The aim of the present work is to develop a mathematical model of a direct

contact cocurrent rotary dryer operated in steady state conditions. Comparison
between experimental and simulated temperature and moisture content profiles,
given a drying mechanism and certain operating conditions is made. Soya and fish
meals were used as testing materials.

Mathematical Model

The following assumptions were incorporated in the present direct contact

cocurrent rotary dryer model:
a) heat and mass transfer from the drying gas to the solid panicles takes place
when the panicles fall from the internal dryer flights into the drying gas stream
b) solid panicles form cunains across the dryer cross-section, through all the dryer
length
c) surface area available for heat and mass transfer is that of the falling panicles
d) drying gas temperature is uniform through the dryer cross-section
e) panicle shrinking is negligible
0 thermal properties of solid panicles and drying gas change due to changes in
their temperatures.
 DIRECT CONTACT ROTARY DRYER 2423

g) the process takes place in steady state conditions, i.e. temperatures and moisture
contents at the entrance and exit ends of the dryer remain constant.

To solve the differential equations for mass and energy balances

relationships for: drying rate, heat transfer coefficient between drying gas and solid
particles, and retention time of particles inside the dryer, are needed. 7he
performance of a rotary dryer is strongly affected by the heat transfer rate from
Downloaded by [University of Arizona] at 00:18 10 July 2013

the drying gas to the solid phase and by the retention time of the panicles inside
the dryer. Both phenomena, depending on the solid propenies and the operating
conditions of the dryer fix the drying time.

Mass and energy balances applied on both phases, drying gas and solid,
over a dryer length differential element give the following expressions for the
solid moisture content and temperature and, gas humidity and temperature:

Drying Kinetics

One of the objectives of the present work was the adequate description of
the mass transfer rate between the phases. The chosen mechanism has to be related
to the solid nature, moisture content and temperature levels.
 2424 SHENE ET AL.

Most particulate nonhigroscopic systems present a saturated surface drying

kinetics, yielding a mass transfer rate proportional to the gas humidity:

N = kp (Y, - Y) f ( X ) rn
ss

This relationship represents the drying kinetics that has to be used in the
balances given by relationships ( I ) to (4). f(X) is a function of the state of the
Downloaded by [University of Arizona] at 00:18 10 July 2013

moisture in the solid. During the constant drying rate period f(X) is given by:
and in the decreasing drying rate period:

f (X) = I i / X 2 x, (6)

ln the decreasing drying rate period f(X) is a function of the critical moisture
content:

On the other hand, drying of biological source solids where moisture is pan
of their structure, often it has been represented h u g h internal moisture diffusion
models. For this type of solids the moisture diffusivity coefficient inside the solid
depends on the solid average moisture content. Alvarez and Legiies (1986)
presented a model where this coefficient is a function of time because during the
drying process average moisture content is a fun'ction of time. The drying kinetics
given by this model is:

where the Fourier number Fo, is given by:

 DIRECT CONTACT ROTARY DRYER 2425

Do is the moisture d i i s i v i t y coefficient evaluated at the initial solid moisture

content if it is lower than the critical content. In other case. Do represents the
diffusivity coefficient evaluated at the critical content.

Retention Time

Retention time prediction is complex due to the interactions among factors

Downloaded by [University of Arizona] at 00:18 10 July 2013

like solid holdup, amount and geometry of flights, slope, rotation rate, length and
diameter of the dryer, properties of the system (particled diameter and density),
and gas velocity. The effect of each of these variables has been extensively studied
in literature (Friedman and Marshall 1949, Saeman and Mitchell 1954, Sharples
et al. 1964, Kamke and Wilson 1986). Research in this area has been focused in
developing theoretical models based on the cascade cycle concept. However. these
models include several fluiddynamic parameters whose values are difficult to
obtain, limiting their applicability.

In practice, retention time is empirically estimated based on the pioneering

work of Friedman and Marshall (1949):

The drawback of this relationship is that it cannot be used when the dryer
slope is nill. Alvarez and Shene (1994) working with different types of solids
evaluated the retention time from data collected in the same equipment and
presented another empirical relationship:
 2426 SHENE ET AL.

Volumetric Heat Transfer Coefficient

It is a usual practice to represent the heat uansfer rate between the drying
gas and the solid panicles through a volumetric heat transfer coefficient U, based
on the dryer volume. In previous work the authors have wmpared the volumetric
heat transfer coefficient obtained from different approaches with experimental
values. One of them uses the relation:
Downloaded by [University of Arizona] at 00:18 10 July 2013

where A, is the total surface area available for the heat aansfer (Langrish et a].
1988) and h is the heat transfer coefficient evaluated from an empirical
relationship (Kunii and Levenspiel, 1977). However, calculated heat tiansfer
coefficients for different solids do not compare well with the experimental data.

On the other hand, a general and simple empirical relationship wmmonly

accepted is (Friedman and Marshall, 1949):

K is an empirically adjusted parameter depending on the dryer geometry. Values

for parameters m and K of 0.67 and 237 respectively. for the coefficient having
units of W/m3'K have been suggested (Friedman and Marshall. 1949, Saeman and
Mitchell, 1954). Heat transfer experiments in a pilot-scale direct contact r o t a j
dryer gave the values shown in Table 1 for parameters in relationship (13)
(Alvarez and Shene. 1994).

EXPERIMENTAL PROCEDURE

Details of the equipment used in the experimental drying runs are given
elsewhere (Alvarez and Shene, 1994). The same equipment has been used to
 DIRECT CONTACT ROTARY DRYER 2427

TABLE 1. Values of K and m Parameters in the Relationship Between Gas

Mass Density Rate and Volumetric Heat Transfer Coefficient OH/m3" K).

Solid
Sawdust

General
Downloaded by [University of Arizona] at 00:18 10 July 2013

obtained data related to the heat transfer coefficient and the retention time. After
steady state conditions were reached. outlet solid rate and inlout flow air
temperature, solid samples were taken at nine different positions along the dryer
to measure the moisture content. In the same positions solid and drying gas
temperature were recorded. Soya and fish meals were used as drying materials.
Physical properties of the solids and operating conditions used in the drying runs
are given in Table 2.

RESULTS

To run a simulation any needed parameter (K,m, b, Do) has to be

estimated from different drying experiments under conditions (temperature and
moisture content) similar to those found in the direct contact rotary drying. Drying
experiments in fluidized bed dryers are suitable to estimate these values because
the contact between the drying gas and solid panicles resembles that of a direct
contact rotary dryer.

Drying kinetiw chosen for the simulations was a saturated surface kinetics
followed by a decreasing drying rate period depending on the inlet solid moisture
content. In the decreasing drying rate period the drying kinetics wuld be
proportional to the unbound moisture content or diffusional.
 2428 SHENE ET AL.

Table 2. Solids Physical Roperties and Operdting Conditions uied in the

Experimental Drying Runs in a Pilot-Scale Rotary Dryer.

Soya meal F i b meal

p,: 620 kglm3 p,: 560 kglrn'
4: O.2lxlO" m 4 : 0.77~10'm
G : 0.075-0.120 kgls G : 0.013-0.020 kgls
Downloaded by [University of Arizona] at 00:18 10 July 2013

S.: 0.020-0.035 kgls S,: 0.005-0.014 kgls

RPM: 9 RPM: 9
s: 0.066 m/m s: 0.035 d m

Parameter values in relationship for the volumetric heat transfer coefficient

(equation 13) used in the simulations were those shown in Table I, for the
considered solids. Retention time was evaluated using relationship in equation
(1 1).

DISCUSSION AND CONCLUSIONS

Saturated surface kinetics (equations 5 to 7) need no any extra parameter

specification as it is needed when diffusional kinetics is chosen. Ln the latter case,
one parameter value (parameter b) is needed if the inlet solid moisture content is
higher than the critical content, and two parameter values (Do and b) are needed
if this moisture content is lower than the critical value. This is so because if a
shih between two different drying rate periods takes place inside the dryer a
continuity in the drying rate curve has to be met. This continuity is achieved by
making equal the drying rate expressions in both periods. Do value is estimated
from this relation.

Different drying mechanisms give rise to different axial profiles shapes,

because drying kinetics controls the dynamic equilibrium between heat and mass
 DIRECTCONTACTROTARYDRYER
Downloaded by [University of Arizona] at 00:18 10 July 2013

Figure I. Comparison between experimental and simulated profiles (continuous

line) in a pilot-scale direct contact rotary dryer for soya meal. In the decreasing
drying rate period: (a-d) proponional to the solid unbound moisture content. ( 0
drying air temperature, solid moisture content,. solid temperature).
 SHENE ET AL.
Downloaded by [University of Arizona] at 00:18 10 July 2013

0 1 2 3
PoslUon (m)

Figure 2: Comparison between experimental and simulated profiles (continuous

line) in a pilot-scale direct contact rotary dryer for fish meal. In the decreasing
drying rate period: (a-d) proportional to the solid unbound moisture content and
@-c) diffusional kinetics. ( 0 drying air tempemture, A solid moisture content,.
solid temperature).
 DIRECTCONTACTROTARYDRYER 243 1

transfer rates. This equilibrium which is closely related to the temperature and
moisture content profiles along the dryer. strongly influences the solid temperature
due to the opposite thermal effect between the gas-solid heat uansfer rate and the
latent heat needed to evaporate the solid moisture. A proper specification of U,
and N must ensure a thermal equilibrium in the system, otherwise t w low solid
temperatures or drying levels can be obtained.
Downloaded by [University of Arizona] at 00:18 10 July 2013

Comparisons between experimental and simulated profiles for soya and fish
meals are shown in Figures I and 2 respectively. From the Figures it is shown that
a bener prediction is given for soya meal. For fish meal, the experimental solid
temperature profiles present a maximum near the dryer enuance which is not
predicted by the model. The square r w t of the average square differences between
experimental and calculated values for drying air temperature. solid temperature
and moisture content: 17.3.5.5 and 0.03 for soya meal and 21.2, 38.8 and 0.07 for
fish meal.

Although moisture content levels used in the experimental runs were not
too high, the saturated surface kinetics gave a bener prediction. This was
explained because the experiments were made with dry solid moistened and
homogenized to the initial moisture content. The drying is actually removing water
added to the solid and not its intrinsic moisture, which might display a different
drying mechanism.

7he model can predict solid and drying gas moisture content and
temperature axial profiles in a direct contact cocurrent rotary dryer operated in
steady state conditions. With this model the process optimization can be a
straightforward task.

ACKNOWLEDGEMENTS

The authors wish thank to Fondo Nacional de Ciencia y Tecnologia, for the
economical assistance through projects No 930344 and 1950489.
 SHENE ET AL.

NOMENCLATURE

: dryer cross section area

: variable diffusivity model parameter
: specific heat
: dryer diameter
: particle diameter
: moisture diffusivity coefficient
: Fourier number
Downloaded by [University of Arizona] at 00:18 10 July 2013

: function given by equation (7) and (8)

: total gas mass rate
: total gas mass density rate
: dry gas mass rate
: parameter in equation (13)
: volumetric mass transfer coefficient
: parameter in equation (1 3)
: drying rate
: rotation rate
: panicle radii
: dryer slope
: dry solid mass rate
: time
: temperature
: volumeuic heat transfer coefficient
: solid velocity
: rotation rate
: solid moisture, dry basis
: gas humidity, dry basis
: dryer length
: axial coordinate
: retention time
: latent heat of vaporization
: density
index
a : vapour
c : critical
G : gas
h :wet
o : initial
s : solid
sat : saturated
* : equilibrium
 DIRECT CONTACT ROTARY DRYER

REFERENCES

Alvarez P.I., Legiies P., Drying Technology 4, 1986,l

Alvarez P.I.,M. Rutman&. GuerreroJ988, Contribuciones 81, 5.
Alvarez P.1, Shene C., 1994, Drying Technology 12(7), 1605.
Alvarez P.1, Shene C., 1994, Drying Technology 12(7), 1629.
Douglas P.L., A. Kwade, P.L. Lee, S.K. Malliek, 1993, Drying Technology
11(1).129
Friedman S.J., W.R. Marshall, Chem. Eng. hog. 45, 1949. 573.
Downloaded by [University of Arizona] at 00:18 10 July 2013

Kamke F., Wilson J., AIChEJ 32(2), 1986, 263.

Kisakiirek B., 1982, Proc. 3rd h t . Drying Symp., ~ ( U N VBirmingham).
.
J.C. Ashwonh, ed. Drying Research Ltd., Wolverhampton, England.
Kunii D., 0.Levenspiel, 1977. "Fluidization Engineering". Kreiger, New
York.
Langrish T.A.G., D. Raey, R.E. Bahu, 1988, J. Separ. Proc. Technol. 9.15-
20.
Mc. Cormick P., Chem Eng. Prog. 58, 1962.57.
Myklestad O., Chem Eng. Prog. Sym. Ser. 59, 1963, 126.
Saeman W.C. Mitchel T.R., Chem. Eng. hog. 50, 1954,467.
Sharples K., Glikin P.G., Wame R.,1964, Trans. Instn. Chem. Engrs. 42,
T257.
Thome B.. 1979, Ph.D. Thesis, UNV. Coll., Dublin, Ireland.
Wang F.Y., I.T. Cameron, J.D. Litster, P.L. Douglas, 1993, Drying
Technology 11(7), 1641.