Applied Thermal Engineering 73 (2014) 899e913

Contents lists available at ScienceDirect

Applied Thermal Engineering

journal homepage: www.elsevier.com/locate/apthermeng

Simulation of exhaust gas heat recovery from a spray dryer

Boris Golman*, Wittaya Julklang
School of Chemical Engineering, Institute of Engineering, Suranaree University of Technology, 111 University Avenue, Muang District,
Nakhon-Ratchasima 3000, Thailand

h i g h l i g h t s

We model industrial-scale spray drying process with the exhaust gas heat recovery.

We develop an Excel VBA computer program to simulate spray dryer with heat recovery.

We examine effects of process parameters on energy efciency and energy saving.

High energy efciency is obtained during drying of large amount of dilute slurry.

Energy saving is increased using the large amount of hot drying air.

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 10 May 2014
Accepted 20 August 2014
Available online 30 August 2014

This study explored various alternatives in improving the energy utilization of spray drying process
through the exhaust gas heat recovery. Extensible and user-friendly simulation code was written in
Visual Basic for Applications within Microsoft Excel for this purpose. The effects of process parameters
were analyzed on the energy efciency and energy saving in the industrial-scale spray drying system
with exhaust gas heat recovery in an air-to-air heat exchanger and in the system with partial recirculation of exhaust air. The spray dryer is equipped with an indirect heater for heating the drying air. The
maximum gains of 16% in energy efciency and 50% in energy saving were obtained for spray drying
system equipped with heat exchanger for exhaust air heat recovery. In addition, 34% in energy efciency
and 61% in energy saving for system with recirculation of exhaust air in the present range of process
parameters. The high energy efciency was obtained during drying of large amount of dilute slurry. The
energy saving was increased using the large amount of hot drying air.
2014 Elsevier Ltd. All rights reserved.

Keywords:
Spray drying
Energy recovery
Exhaust gas
Simulation
Visual Basic for Applications
Excel

1. Introduction
Spray drying process has been widely used for production of
particulate products on large scale in ceramic, food and other industries [1]. Recently, spray drying has been utilized for
manufacturing various high value particles in the form of agglomerates of nanoparticles [2,3]. These agglomerates are in high
demand as they can be processed in dry phase due to their
enhanced owability. In addition, the materials produced using
these agglomerates possess the superior functional properties due
to their nano-sized constituents.
The spray drying process consists of the spraying of a feed in the
form of liquid or a slurry of nanoparticles into a drying chamber,
drying of small droplets in contact with hot stream of the drying

* Corresponding author.
E-mail addresses: golmanboris@gmail.com, golman@sut.ac.th (B. Golman).
http://dx.doi.org/10.1016/j.applthermaleng.2014.08.045
1359-4311/ 2014 Elsevier Ltd. All rights reserved.

medium and recovering powder product from exhaust stream. In

order to enable the slurry to be sprayed, the solid content of the
slurry is typically limited by its viscosity requirement. Thus, since
large amount of water in the slurry needs to be evaporated, spray
drying is an energy demanding process. The rising energy cost and
growing concerns over greenhouse gas emissions urge companies
to utilize the energy improving measures. Various approaches to
energy conservation in spray drying have been reported in the
literature including optimization of process parameters [4], utilization of exhaust air energy in a heat recovery system [5] and use of
an exhaust air recirculation system [6].
The spray drying is a complex process that involves the heat,
mass and momentum transfer between the droplet and the drying
medium as well as the heat and mass transfer in the partially dried
agglomerates. The models of various complexities have been reported in the literature for modeling of spray drying. Empirical
correlations have been developed for droplet diameter change
during drying of certain products, such as milk powder [7]. The

900

B. Golman, W. Julklang / Applied Thermal Engineering 73 (2014) 899e913

Nomenclature
A
Cp
ES
h
m
Q
RR
T
tol
U
w
X
Y
Z

total surface area of drying chamber (m2)

specic heat (J kg1 K1)
energy saving (%)
specic enthalpy (J kg1)
mass ow rate (kg s1)
amount of heat in unit time (J s1)
recirculation ratio (%)
temperature (K)
tolerance
overall heat transfer coefcient (W m2 K1)
weight percentage of water on wet basis (%)
mass fraction of water on dry basis (kg kg1)
humidity on dry basis (kg kg1)
fractional water balance discrepancy (e)

Greek symbols

heat exchanger overall effectiveness (%)

hR
energy efciency (%)
l
latent heat of vaporization of water (J kg1)
Subscripts
a
dry air

models based on a characteristic drying rate curve have been used

for simulation of industrial spray drying [8]. For example, Patel and
Chen [9] presumed that the drying rate is a linear function of the
particle moisture content. The reaction engineering approach was
developed to describe the droplet drying with an assumption that
evaporation is an activation process similar to the chemical reaction
[10,11]. The equations of models described above are simple, but
these approaches require experimental determination of the
material-specic parameters or relationships such as the activation
energy as a function of the average free moisture content or the
characteristic drying curve. However, such data are not readily
available in the literature. The comprehensive mathematical
models were also developed to describe the heat and mass transfer
in the droplet during the constant and falling drying rate periods
[12e14]. These models involve a numerical solution of the coupled
system of partial and ordinary differential equations with moving
boundaries. The complete mathematical analysis of spray drying
process presents a challenging task and the numerical simulations
in three dimensions tend to take a prohibitively long time. Therefore, Langrish [15] proposed a multi-scale approach to the
modeling of spray dryers. The coarsest scale models are derived on
the basis of overall heat and mass balances in a well-mixed dryer.
These models are useful for preliminary analysis of the effect of
process parameters on the drying system performance and quick
evaluation of various methods for minimization of energy consumption [16]. The ner scale models include momentum, heat and
mass transfer equations for droplets and drying gas assuming that
they are owing in parallel to each other [17,18]. Furthermore, the
detail analysis of ow pattern in the chamber is implemented in the
nest scale models using a computational uid dynamic approach
[19,20]. For the present analysis on effect of process parameters on
efciency of energy utilization in spray drying the coarsest scale
model was chosen as such analysis will require iterative solutions
of mass and energy balances for every set of parameters and the
computation will take long time if we use ne approach.
Computer simulation has proven to be an invaluable tool for the
feasibility analysis of energy saving process alternatives [21].
Currently, a few commercial software packages are available for

aa
d
da
dew
ea
fa
h
ha
i
in
l
loss
ma
oa
out
p
ra
rec
ref
req
s
sl
v
va

atomizing air
dry base
drying air
dew point
exhaust air
feed air
wet base
hot air
ith stream
inlet, input
liquid
loss
mixed air
ambient air
outlet
solid product
recirculating air
recovered
reference
required
solid
slurry
vapor
vent air

design and analysis of drying system [22]. However, these packages

are quite complex as they cover a wide range of drying processes,
which require steep learning curve to operate. In addition to not be
able to be modied due to their closed source code, these packages
are costly. The general commercial simulation packages, such as
Matlab, were also used for spray drying simulation [17]. However,
such software packages are expensive especially for the university
research groups with limiting budget or for small companies in
developing countries. There are also free open-source alternatives
to Matlab, such as GNU Octave [23]. Yet, they are not as userfriendly as the commercial packages and, generally, demand deep
knowledge of programming language to modify the program.
The evaluation of the possibilities of energy improvement and
optimization of spray drying process require development of userfriendly and relatively simple software to be used not only by
university researchers but also by industrial engineers. Microsoft
Excel is the most widely utilized spreadsheet in industry and in
academia. Bilic and Glavac [24] published one of the early articles
on spray drying simulations utilizing spreadsheets. Velic et al. [25]
extended the spreadsheet software to the analysis of possible fuel
saving by recirculation of exhaust air. However, their model
required a-priory knowledge of exhaust air temperature. In
addition, only a few parameters, such as the temperature of drying
air and the recirculation rate, were investigated on energy efciency of spray dryer. Lately, Grasmeijer et al. [26] developed an
Excel program for the simulation of spray drying of pharmaceutical products. Their model is mainly intended for simulating
laboratory scale spray dryers, as it is based on the estimation of
heat losses from experimental data. However, the heat losses in
industrial scale dryers are usually assumed to be negligible in
comparison with the total energy input into the dryer [1]. Therefore, their model and, accordingly, their simulation program
require substantial development in order to be utilized for simulating spray dryers with exhaust gas heat recovery. Furthermore,
their simulations were performed using the spreadsheet commands, which would hamper understanding of complicated expressions and limit further development and extension of the
computer program.

B. Golman, W. Julklang / Applied Thermal Engineering 73 (2014) 899e913

901

Fig. 1. Schematic diagram of spray drying system.

Recently, the combination of Visual Basic for Applications (VBA)

language for simulation with the Excel spreadsheet for input/
output and for graphical visualization of results proved to be
popular for software development of technical applications [27].
Excel-VBA tools have been developed for dryer design [28,29] and
successfully applied for simulation of various drying systems, such
as convective dryers [30], sludge drying-incineration systems [31],
plug-ow spray dryers [32], tunnel dryers [33], etc. The application
of Excel-VBA software allows users to do simulation in the worksheet with the program output as well as to modify or extend the
program according to their needs, as they would have a full access
to the VBA source code.
Therefore, there is a need in comprehensive analysis on effect of
wide variety of process parameters on the efciency of energy recovery in industrial spray dryers. For this purpose, the freely
available and easy to use software was developed using the model
that includes the detailed heat and mass balances for drying air,
solid particles and moisture.
The rst objective of the present research is to describe the
development of user-friendly and extensible software tool in
Excel-VBA, which is intended to aid the researchers and engineers in the selection of proper operational conditions of spray
drying and in the evaluation of various alternatives in the
improvement of process energy utilization. The second objective
is to study energy efciency and investigate the possibilities of
energy saving in spray drying process by means of recovery of
exhaust air energy for heating of feed air and through the recirculation of exhaust air.
2. Mathematical model
Fig. 1 illustrates a schematic diagram of three different congurations of spray dryer system investigated in the present study:
conventional co-current spray dryer, system equipped with a heat
exchanger for energy recovery from dryer exhaust air and system
involving the recirculation of exhaust air. The coarsest scale model
is utilized in the present research for comparative simulation study
on the effect of process parameters on energy efciency of various
congurations of spray drying systems.

2.1. Overall energy balance around spray-drying system

The energy enters the spray drying system through the slurry
feed, atomizing air and feed drying air streams as well as with the
heating medium in a heater for heating the drying air. The energy
leaves the system with the solid product, vent air stream and in the
form of heat loss to the environment.
The energy balance in a spray-drying system at a steady-state
can be stated as

X
 X

mi;d hi 
mi;d hi Q
out

(1)

in

where mi,d is the mass ow rate of ith stream specied on a dry

basis, hi is the specic enthalpy of that stream dened per unit mass
of dry air or dry solids and Q is the heat transfer rate.
The specic enthalpies of slurry feed and solid product streams
are given by

Table 1
Enthalpies of various streams supplied to or withdrawn from spray drying system.
Stream

Equation

Enthalpy of
slurry feed



msl;d hsl msl;d Cps Xsl Cpl Tsl  Tref

Enthalpy of
feed air

mfa;d hfa mfa;d

h

Cpa Yfa Cpv



Tfa  Tref

(12)


Yfa lref

(13)

Enthalpy of

h

i
atomizing air m
aa;d haa maa;d Cpa Yaa Cpv Taa  Tref Yaa lref

(14)

Enthalpy of



solid product m
Tp  Tref
p;d hp mp;d Cps Xp Cpl

(15)

Enthalpy of
vent air

(16)


h

i
mva;d hva mva;d Cpa Yva Cpv Tva  Tref Yva lref

902

B. Golman, W. Julklang / Applied Thermal Engineering 73 (2014) 899e913

Table 2
Overall mass balances around spray drying system.
Balance

hi hs;i Xi $hl;i

Equation

Dry solids
msl;d mp;d

(20)

mfa;d maa;d mva;d

(21)

h
i
1  Z msl;d Xsl mfa;d Yfa maa;d Yaa mp;d Xp mva;d Yva

(22)

Dry air

Moisture

(2)

where hs,i and hl,i are the specic enthalpies of dry solids and water
in the ith stream, respectively (i sl for slurry and i p for product),
and Xi is the mass fraction of water in the ith stream on dry basis.
The specic enthalpies of drying, atomizing and feed air streams
are dened by

hi ha;i Yi $hv;i

(3)

where ha,i and hv;i are the specic enthalpies of dry air and water
vapor in the ith stream, respectively, and Yi is the humidity of ith
stream.

Fig. 2. Overall owchart of simulation.

B. Golman, W. Julklang / Applied Thermal Engineering 73 (2014) 899e913

The specic enthalpies of dry air, dry solids and water are given
by





ha;i Cpa $ Ti  Tref ; hs;i Cps $ Ti  Tref


hl;i Cpl $ Ti  Tref

and
(4)

903

where Cpa, Cps and Cpl are the specic heat capacities of dry air, dry
solids and liquid, respectively, Ti is the temperature of ith stream
and Tref is the reference temperature, Tref 273.15 K.
The specic enthalpy of water vapor in the ith stream can be
calculated as [34]:



hv;i lref Cpv $ Ti  Tref

Fig. 3. Illustration of computer program with (a) user input and (b) output data cells.

(5)

904

B. Golman, W. Julklang / Applied Thermal Engineering 73 (2014) 899e913

where lref is the latent heat of vaporization of water at Tref and Cpv is
the specic heat capacity of water vapor. Mujumdar [34] recommended to use a more precise equation for calculation of hv.da in the
case of high gas humidity, i.e. Yi > 0.05:



hv;i Cpl $ Tdew  Tref ldew Cpv $Ti  Tdew

(6)

where Tdew is the dew point temperature and ldew is the latent heat
of vaporization of water at Tdew.
Combining Eqs. (4) and (5), Eq. (3) can be rearranged as



hi Cpa Yi $Cpv Ti  Tref Yi $lref :

(7)

The drying air is heated in a heater from the inlet temperature

Tda,in up to the specied temperature Tda,out. The heat delivered to
the drying air by heating media in unit time is given by



Qda mda;d hda;out  hda;in

(8)

Using Eqs. (20)e(22), the humidity of vent air Yva is obtained as

Yva

i
h


msl;d 1  ZXsl  Xp 1  Z mfa;d Yfa maa;d Yaa
mva;d

where Z is the fractional discrepancy of water balance. This

parameter accounts for the moisture that leaks out of the dryer [35].
2.3. Mass and energy balances of heat recovery system of exhaust
air
The energy from exhaust air is assumed to be recovered in a
cross-ow air-to-air heat exchanger. In this study, the energy recovery calculation follows ASHRAE handbook [36] recommendations with the assumption of no condensation of water in the heat
exchanger.
The temperature of the hot air stream leaving the exchanger is
calculated as

!

mmin 
Tea  Tfa
mfa;d

where hda,out and hda,in are the specic enthalpies of drying air per
unit mass of dry air at the outlet and inlet of heater, respectively.
Substituting Eqs. (7) into (8) results in

Tha Tfa



Qda mda;d Cpa Yda Cpv Tda;out  Tda;in :

and the vent stream as

(9)
Tva

The energy balance Eq. (1) can be rewritten as

 

mp;d hp mva;d hva  msl;d hsl mfa;d hfa maa;d haa
Qda  Qloss

(10)

where Qloss is the thermal loss. Here, the thermal loss is expressed
as

Qloss UATea  Toa

(17)

The temperature and humidity of feed air are assumed to be

equal to those of atomizing air as both streams are usually supplied
to the system at ambient conditions:

Tfa Taa Toa ; Yfa Yaa Yoa

(24)

!

mmin 
Tea  Tfa
Tea 
mea;d

(25)

where is the heat exchanger overall effectiveness and mmin is the

smaller of mfa,d and mea,d. Following [36], the normal range of
effectiveness for cross-ow heat exchangers is 50e70%.
2.4. Mass and energy balances of the exhaust air recirculation
system

(11)

where U is the overall heat transfer coefcient and A is the total area
of drying chamber.
The enthalpies of various streams supplied to or withdrawn
from spray drying system are summarized in Table 1.
The energy input in the spray drying system, Ein, is dened as

Ein msl;d hsl mfa;d hfa maa;d haa Qda

(23)

(18)

where Toa and Yoa are the temperature and humidity of ambient air,
respectively.
Thus, using Eqs. (12)e(14) and (18), Eq. (17) can be rewritten as

 
h

Ein msl;d Cps Xsl Cpl Tsl  Tref maa;d mfa;d Cpa

i

Yoa Cpv Toa  Tref Yoa lref mda;d Cpa Yda Cpv


 Tda;out  Tda;in :
(19)

2.2. Overall mass balances around spray drying system

The overall mass balances of solids, air and moisture are summarized in Table 2.

At the splitting point M1, the exhaust air stream is divided into
two streams, vent air and recirculating air. The temperature and
humidity of all streams are assumed to be the same:

Tva Tra Tea ; Yva Yra Yea

(26)

The mass ow rates of recirculating and vent air streams are

given by


RR
RR
; mva;d mea;d 1 
100
100


mra;d mea;d

(27)

where RR is the recirculation ratio. The validity of chosen value of

recirculation ratio is checked to ensure that the moisture content of
recirculate air is far less than that of saturated air as indicated by the
dew point temperature.
Table 3
Typical values and working ranges of process parameters.
Parameter
Ambient air
Temperature [ C]
Relative humidity [%]
Drying air
Volume ow rate [m3/h WB]
Temperature [ C]
Slurry
Mass ow rate [kg/h WB]
Water content [wt.% WB]
Temperature [ C]
Product particles
Water content [wt.% WB]

Typical value

Working range

20
50

e
e

45,000
240

40,000e50,000
180e300

1800
40
20

1200e2400
20e60
e

B. Golman, W. Julklang / Applied Thermal Engineering 73 (2014) 899e913

905

At the mixing point M2, the recirculate air stream is mixed with
the feed air stream

mma;d mra;d mfa;d

(28)

where mma,d is the mass ow rate of the mixed air, which is equal to
the mass ow rate of drying air, mda,d.
The moisture balance around mixing point M2 is given by

mda;d Yma mfa;d Yfa mra;d Yra

(29)

Using Eq. (23) and taking into account relationships Yra Yva
and Yda Yma, the moisture content of drying air is expressed as

Yda

Fig. 4. Comparison of calculated and experimental temperature and relative humidity

of exhaust gas.

mfa;d Yfa mra;d

mda;d
mda;d
i!
h


msl;d 1  ZXsl  Xp 1  Z mfa;d Yfa maa;d Yaa

:
mva;d
(30)

Fig. 5. Effect of (a) drying air temperature and (b) drying air ow rate on energy efciency in spray drying system with heat recovery using heat exchanger.

Fig. 6. Variations of enthalpy of feed air, heat supplied for heating drying air and
energy input with (a) drying air temperature and (b) drying air ow rate in spray
drying system with heat recovery using heat exchanger.

906

B. Golman, W. Julklang / Applied Thermal Engineering 73 (2014) 899e913

Assuming that humidities of feed and atomizing air streams are

equal to the humidity of ambient air, Yoa, Eq. (30) is further
simplied as

Yda Yoa 



Yoa Zmra;d mra;d msl;d 1  ZXsl  Xp

:
mda;d
mda;d
mva;d

2.5. Denitions of spray drying energy efciency and energy saving

(31)

Substituting Eqs. (27) into (31) results in







msl;d RR 1  ZXsl  Xp
maa;d mda;d Yoa Z RR

:
Yda Yoa 
100
mda;d 100  RR
mda;d

Ereq
 100%
Ein

hR

(33)

where hR is the energy efciency and Ereq is the energy required for
evaporation of specied amount of water, which is expressed as

Using Eqs. (9), (27) and (33), the temperature of mixed air
stream is given by the following equation:

Tma Tref

The energy efciency of spray drying process with exhaust air

heat recovery is dened as the ratio of the energy required for
evaporation of specied amount of water to the energy supplied to
the dryer [37]:

(32)
The heat balance around the mixing point M2 is given by

mma;d hma mfa;d hfa mra;d hra

where hfa and hva are the specic enthalpies of feed and vent air
streams, respectively.




mda;d hfa RR=100 mda;d maa;d hva  hfa
mda;d Cpa Yda Cpv



Ereq Xsl  Xp msl;d lw

(35)

(36)

For spray drying processes with exhaust air heat recovery (S2
and S3), Ein is calculated considering the reduction of heat used in
the heater for heating of drying air by Eq. (9) as the air temperature
at the heater inlet, Tda,in, is high due to the recycling of heat from

Yda lref
Cpa Yda Cpv
(34)

Fig. 7. Effect of (a) slurry feed rate and (b) slurry water content on energy efciency in
spray drying system with heat recovery using heat exchanger.

Fig. 8. Variations of enthalpy of feed air, heat supplied for heating drying air and
energy input with (a) slurry feed rate and (b) slurry water content in spray drying
system with heat recovery using heat exchanger.

B. Golman, W. Julklang / Applied Thermal Engineering 73 (2014) 899e913

exhaust air. It should be noted that Tda,in Tha for the process S2,
while Tda,in Tma for the process S3.
The energy saving is dened as the ratio of energy recovered
from the exhaust air to the energy supplied to the spray dryer
without heat recovery:

ES

Erec
 100%
Ein;S1

(37)

where Erec is the energy recovered from the exhaust air,

Erec Ein,S1  Ein. Here, Ein,S1 is the energy supplied to the dryer
without heat recovery, which is calculated by Eq. (19) assuming
Tda,in Tfa.
2.6. Calculation steps
An overall owchart illustrating the main calculation steps is
shown in Fig. 2. The calculation steps differ depending on the
conguration of spray drying system, i.e. without heat recovery,
with exhaust air heat recovery using an air-to-air heat exchanger
and heat recovery by recirculation of exhaust air.
For systems without heat recovery and with exhaust air heat
recovery using heat exchanger, the humidity of drying air at the
dryer inlet is known, i.e. Yda Yfa Yoa, and the mass balance could
be solved to obtain the exhaust air humidity by Eq. (23). In the case
of the system with heat recovery by recirculation of exhaust air, the
humidity of drying air at the dryer inlet is not known a priory since
it depends on the exhaust air humidity. Thus, Yda is estimated by
iterative solution of mass balance equation. Then, the heat balances
for the spray dryer are solved iteratively for all congurations. The
iterative procedure is based on an estimation of the exhaust air
temperature, Tea, which is adjusted until the heat balance is satised and the converge criteria is achieved, i.e. Error (Tea) < tol, where
tol is the given tolerance. The preliminary simulation conrmed

907

that the value of tolerance equal to 103 introduces negligible error

in simulation results. Calculation of spray drying system with
exhaust air recirculation requires an additional outer iteration loop
for evaluation of the temperature of drying air at the heater inlet,
Tda,in.
3. Simulation study of spray drying process with heat
recovery
3.1. Simulation software
The software code is written in VBA using EXCEL as a graphical
front-end to input data, control simulation process and output results of simulation. Typically users interact with software at Excel
level. However, users can adapt the code to their needs by changing
VBA code or by adding additional modules. A computationally
intensive task could be implemented in a high-level language, such
as Fortran or C, in the form of a dynamic link library and connected
to the VBA program through the standard interface.
The software code was organized in a modular way such as
modules realizing models of spray dryer and heat exchanger can be
chosen from the simple to more complicated ones depending on
the objectives of simulation. The thermodynamic water properties
were implemented using the empirical correlations [38] as well as
the more precise models based on the IAPWS Industrial Formulation 1997 [39]. The moist properties were calculated using the
experimental-based correlations [40,41]. The advanced thermodynamic models of moist air [42] will be connected to the program
in a near future.
The front-end of the computer program is illustrated in Fig. 3.
The user should select a conguration of spray dryer system and
input the process parameters. The intermediate results informing
user about the progress of simulation as well as the nal results are
displayed in the main sheet and the supplementary results in

Fig. 9. Effect of process parameters on energy efciency in spray drying system with heat recovery by recirculation of exhaust air: (a) drying air temperature, (b) drying air ow rate,
(c) slurry feed rate and (d) slurry water content.

908

B. Golman, W. Julklang / Applied Thermal Engineering 73 (2014) 899e913

additional sheets. The basic information for each process stream

and equipment will be shown on the process ow diagram together
with the resulting energy efciency and energy saving facilitating
the user analysis. The user can also output additional information
for detail analysis by modifying the output data in VBA module.
3.2. Process studied
The heat recovery of exhaust gas was investigated in an
industrial-scale spray drying system using the developed software.
The dryer throughput is around 1000 kg/h of product agglomerates
[43]. The co-current ow dryer that is equipped with indirect
heater for heating the drying air and the rotary atomizer for
spraying of alumina slurry was studied.
The effects of process parameters were investigated on the energy efciency and energy saving of the spray drying system with
exhaust gas heat recovery at four values of the total efciency of the
air-to-air heat exchanger (0, 50, 60 and 70%) and also the system
with recirculation of exhaust air at four recirculation rates (0, 20, 50
and 70%). The typical values and working ranges of process parameters are summarized in Table 3.
This study was carried out by varying the values of the volumetric ow rate and temperature of the drying air, and the mass

Fig. 10. Variations of heat supplied for heating drying air and energy input with (a)
drying air temperature and (b) drying air ow rate in spray drying system with heat
recovery by recirculation of exhaust air.

ow rate and water content of the slurry individually within the

working range while keeping other parameters constant. However,
the temperature and relative humidity of the ambient air were not
included in the present study as they represent the uncontrolled
parameters. The effects of slurry temperature and product water
content were also excluded from the study because of their limited
ranges of variation.
4. Results and discussion
4.1. Model validation
The validity of the spray drying model was conrmed by
comparing simulation results with experimental data of Ozmen
and Langrish [35] for drying of skimmed milk in the pilot-scale
spray dryer. The studied co-current ow dryer was equipped with
two-uid nozzle. The electrical heater was used for heating the feed
air. The dryer has a diameter 0.8 m, the height of upper cylindrical
part of 1.3 m and the lower conical part of 0.63 m. The slurry mass
ow rate, water content and temperature was set at 1.8 kg/h, 91.8%
and 22  C, respectively. The ambient temperature was 25  C and the
relative humidity of feed and atomizing air were 52%. The drying air
ow rate was set at 81.6 m3/h, while the atomizing air ow rate at
1.22 m3/h.

Fig. 11. Variations of heat supplied for heating drying air and energy input with (a)
slurry feed rate and (b) slurry water content in spray drying system with heat recovery
by recirculation of exhaust air.

B. Golman, W. Julklang / Applied Thermal Engineering 73 (2014) 899e913

The spray drying process of skimmed milk was simulated using

our developed model described previously. As reported in [35], the
value of parameter U$A in the heat loss Eq. (11) was set as
0.039 kW/K. The parameter Z in the moisture balance Eq. (23) was
set as 0.18. Fig. 4 shows the calculated temperature and relative
humidity of exhaust air in addition to the measured ones for
various values of the inlet drying air temperature. It was found that
the calculated temperature and relative humidity of exhaust gas are
in a good agreement with experimental data both at low and high
inlet drying air temperature. The maximum absolute difference in
calculated and experimental exhaust air temperatures was 1.4  C
and 2.3% in relative humidity. These results conrm the validity of
our model that we have developed for analyzing the energy recovery in spray drying system.
4.2. Energy efciency
4.2.1. Spray drying system with heat recovery using air-to-air heat
exchanger
Fig. 5 illustrates the effects of drying air temperature and ow
rate on the energy efciency of spray drying system with heat
recovery of exhaust air using the air-to-air heat exchanger. It is
observed that as the drying air ow rate increases and temperature rises, the energy efciency decreases because both parameters contribute to the gain in the energy input to the spray
drying system, while the energy required for evaporation of
specied amount of water remains unchanged causing a decline
in energy efciency.
As the drying air temperature is increased, the mass ow rate of
drying air declines even at constant volumetric ow rate due to a
decrease in air density. As a result, the enthalpy of feed air (mfa,dhfa)
declines with temperature, as shown in Fig. 6(a) for 70%.
However, the larger amount of heat should be supplied to the
heater for heating the drying air up to the higher temperature as
the gain in the temperature difference between the specied

909

drying air temperature and the exhaust air temperature compensates for the drop in the air mass ow rate. Thus, the energy input to
the system increases with drying air temperature, resulting in the
reduction of energy efciency.
With a rise in the drying air ow rate, both the enthalpy of
drying air and the heat supplied to the heater increase, despite the
fact that the temperature difference, Tda,out  Tea, declines at constant Tda,out. This leads to the enhancement of energy input and
corresponding decline of energy efciency, as illustrated in
Fig. 6(b).
However, the increase in the slurry mass ow rate and water
content results in an enhancement of energy efciency, as shown in
Fig. 7(a) and (b).
As the slurry feed rate and water content are increased, the
temperature of exhaust air declines and the humidity increases as
more energy is consumed on evaporation of large amount of water.
These lead to a rise in the temperature difference, Tda,out  Tea, and
the heat supplied to the heater for heating the drying air, Qda, yet
the contribution of enthalpy of slurry feed is not signicant, as
illustrated in Fig. 8(a) and (b). Nevertheless, the energy required for
water evaporation, Ereq, grows more drastically than the energy
input, Ein, with the increase of slurry feed rate and water content,
thus bringing about the increase of energy efciency.
An increase in the total effectiveness factor of heat exchanger
leads to enhancement of energy efciency for all process parameters as heat can be recovered in greater amounts from the exhaust
air in the air-to-air heat exchanger. Therefore, the feed air will be
preheated to the higher temperature in the heat exchanger and less
heat will be required for heating air in the heater up to the specied
drying temperature, thus decreasing the heat input to spray dryer.
As a result, the utilization of air-to-air heat exchanger for
recovering exhaust air heat allows us to enhance the energy efciency of spray drying system on 16% in comparison with that of
system without heat recovery in the present range of process
parameters.

Fig. 12. Effect of process parameters on energy saving in spray drying system with heat recovery using heat exchanger: (a) drying air temperature, (b) drying air ow rate, (c) slurry
feed rate and (d) slurry water content.

910

B. Golman, W. Julklang / Applied Thermal Engineering 73 (2014) 899e913

4.2.2. Spray drying system with heat recovery by recirculation of

exhaust air
Fig. 9 conrms that the inuences of process parameters on the
energy efciency of spray drying system with recirculation of
exhaust air are similar to those of system with heat recovery by airto-air heat exchanger.
The variations of energy input with temperature and ow rate of
drying air follows mainly with the alteration of the heat delivered
to the drying air in the heater, Qda, as shown in Fig. 10(a) and (b) for
the recirculation ratio equal to 70%. The contributions of enthalpies
of slurry feed and feed air streams to the energy input were minor,
because the feed air accounts for only 30% of total volume of drying
air used in the spray dryer. Higher drying air temperature increases
the difference between the drying air at the dryer inlet and mixed
air temperatures, Tda,out  Tda,in, as well as the humidity of the
mixed air, and accordingly gives rise to higher Qda. The enhancement of ow rate of drying air results in an increase of the heat
supplied to the heater for heating drying air as Qda is proportional
to the mass ow rate of drying air according to Eq. (8). Thus, the
energy efciency of the spray drying system with heat recovery by
recirculation of exhaust air decreases with the increase of drying air
temperature and ow rate to follow the gain in energy input at a

constant energy required for evaporation, as conrmed in Fig. 9(a)

and (b). These results are in agreement with recent results by Toneli
et al. [6].
The higher the ow rate and water content of the slurry, the
lower the temperature and higher the humidity of exhaust air.
These leads to an increase in the heat used for heating the drying air
up to the specied temperature, Tda,out, and, correspondingly, to rise
of the input energy, as shown in Fig. 11 for RR 70%. At the same
time, the evaporation of increasing amount of water requires the
growing consumption of energy, Ereq. Nevertheless, Ereq increases
more rapidly than Ein for increasing slurry feed rate or increasing
slurry water content. Thus, as energy efciency is represented as
the ratio of Ereq to Ein, increasing slurry feed rate or increasing slurry
water content will lead to increase in energy efciency. This is
conformed in Fig. 9(c) and (d).
The results in Fig. 9 also show that the energy efciency tends to
increase with increasing the recirculate rate of exhaust air as less
amount of heat should be supplied to the heater for heating air.
The maximum gain of 34% in the energy efciency was obtained
for the spray drying system with heat recovery by recirculation of
exhaust air relative to the efciency of the system without
recirculation.

Fig. 13. Variations of energy input without heat recovery, Ein,S1, energy input with heat
recovery system, Ein, and energy recovered, Erec, with (a) drying air temperature and
(b) drying air ow rate in spray drying system with heat recovery using heat
exchanger.

Fig. 14. Variations of energy input without heat recovery, Ein,S1, energy input with heat
recovery system, Ein, and energy recovered, Erec, with (a) slurry feed rate and (b) slurry
water content in spray drying system with heat recovery using heat exchanger.

B. Golman, W. Julklang / Applied Thermal Engineering 73 (2014) 899e913

4.3. Energy saving

4.3.1. Spray drying system with heat recovery using air-to-air heat
exchanger
Comparison of Figs. 5 and 7 with Fig. 12 indicates opposite
trends in the variations of energy saving and energy recovery
with process parameters. The energy saving tends to rise with
increasing the drying air mass ow rate and temperature
(Fig. 12(a) and (b)) due to an increase in the amount of energy of
exhaust air that could be recovered in the air-to-air heat
exchanger. As a result, more energy can be saved in the heater
for heating air up to the drying temperature, as illustrated in
Fig. 13.
On the other hand, the energy saving declines with increasing
the slurry mass ow rate and water content (Fig. 12(c) and (d)) as a
large amount of drying air energy is consumed for evaporation of
large quantity of water resulting in low exhaust air temperature.
Thus, less heat will be available for pre-heating feed air in the airto-air heat exchanger and more heat will be consumed for heating air in the heater up to the drying temperature (Qda in Fig. 11(a)
and (b)) resulting in increase of energy input, Ein, as illustrated in
Fig. 14(a) and (b). The energy input to the system without heat
recovery, Ein,S1, remains almost constant with the change in the
slurry feed rate or concentration as the contribution of the enthalpy
of slurry feed to the energy input is very low. Accordingly, the energy recovered from the exhaust air decreases with increasing
slurry feed rate and slurry water content resulting in the decline of
energy saving.
As shown in Fig. 12, more energy can be saved through the
utilization of the heat exchanger with high value of the total
effectiveness factor. The employment of the air-to-air heat
exchanger for exhaust air heat recovery offers an energy saving of
about 50% in the present range of process parameters.

911

4.3.2. Spray drying system with heat recovery by recirculation of

exhaust air
The effects of process parameters on energy saving in spray
drying system with heat recovery by recirculation of exhaust air
are illustrated in Fig. 15. The high percentage of energy saving
was obtained at the high ow rate and temperature of drying air,
and at the low feed rate and water content of slurry similar to
the system with exhaust air heat recovery using the heat
exchanger.
The energy input in the spray drying system without heat recovery signicantly increases with the temperature and ow rate of
drying air, as a growing amount of heat should be supplied to the
heater for heating the drying air from the ambient temperature up
to higher temperature of drying air at the dryer inlet, Tda,out, as
illustrated in Fig. 16(a) and (b) for RR 70%. However, the energy
input in the spray drying system with recirculation of exhaust air
slightly increases with Tda,out and Vda as the drying air is heated
from the temperature of mixed air, which is much higher than the
ambient one, to the specied temperature of drying air at the dryer
inlet. Thus, the energy recovered from the exhaust air also increases
with temperature and ow rate of drying air resulting in corresponding minor gain in energy saving.
In contrast, the energy recovered from the exhaust air
signicantly declines with an increase in the slurry feed rate and
water content, due to the enhancement of the energy input to the
system with recirculation of exhaust air as more energy is
consumed to heat the mixed air of lower temperature and higher
humidity up to the specied drying air temperature at the dryer
inlet, as illustrated in Fig. 17(a) and (b). As a result, the low energy
saving is attained at the high feed rate of slurry of low solid
content.
The maximum percentage of energy saving in the system with
recirculation of exhaust air was obtained around 61%.

Fig. 15. Effect of process parameters on energy saving in spray drying system with heat recovery by recirculation of exhaust air: (a) drying air temperature, (b) drying air ow rate,
(c) slurry feed rate and (d) slurry water content.

912

B. Golman, W. Julklang / Applied Thermal Engineering 73 (2014) 899e913

Fig. 16. Variations of energy input without heat recovery, Ein,S1, energy input with heat
recovery system, Ein, and energy recovered, Erec, with (a) drying air temperature and
(b) drying air ow rate in spray drying system with heat recovery by recirculation of
exhaust air.

5. Conclusions
In the present study, the heat recovery systems of exhaust
drying gas were analyzed in an attempt to decrease the energy
consumption in spray drying process. The simulation code was
developed in VBA within Excel to provide the simple and userfriendly tool for evaluation of process parameters effects on the
performance of spray drying system with exhaust gas heat recovery. Furthermore, the modular and extensible software design
presents a means for more comprehensive or special-purpose analyses by modication of VBA code.
This study conrmed that the energy consumption in industrialscale spray drying process can be improved by utilization of the
high-performance heat exchanger for recovering exhaust air energy with the maximum gains of 16% in energy efciency and 50%
in energy saving and by recirculation of exhaust air with the high
recirculation ratio resulting in 34% increase in energy efciency and
61% in energy saving in the present range of process parameters.
The energy efciency was improved by supplying the dilute slurry
at high ow rate. The energy saving was increased using the large
amount of hot drying air.

Fig. 17. Variations of energy input without heat recovery, Ein,S1, energy input with heat
recovery system, Ein, and energy recovered, Erec, with (a) slurry feed rate and (b) slurry
water content in spray drying system with heat recovery by recirculation of exhaust
air.

As the large number of process parameters inuences on energy

performance of spray dryer, the combinations of parameters should
be analyzed in the future for optimization of energy consumption in
spray drying.
Acknowledgements
The authors would like to thank the National Research Council
of Thailand for their funding of this research.
References
[1] K. Masters, Spray Drying Handbook, Longman Scientic and Technical, Harlow, 1985.
[2] K. Sollohub, K. Cal, Spray drying technique: II. Current applications in pharmaceutical technology, J. Pharm. Sci. 99 (2010) 587e597.
[3] A.B. Nandiyanto, K. Okuyama, Progress in developing spray-drying methods
for the production of controlled morphology particles: from the nanometer to
submicrometer size ranges, Adv. Powder Technol. 22 (2011) 1e19.
[4] N.V. Menshutina, M.G. Gordienko, A.A. Voinovskiy, I. Zbicinski, Spray drying of
probiotics: process development and scale-up, Dry. Technol. 28 (2010)
1170e1177.
[5] M.J. Atkins, M.R.W. Walmsley, J.R. Neale, Integrating heat recovery from milk
powder spray dryer exhausts in the dairy industry, Appl. Therm. Eng. 31
(2011) 2101e2106.

B. Golman, W. Julklang / Applied Thermal Engineering 73 (2014) 899e913

[6] J.T.C.L. Toneli, L.B. Monteiro, M.A.J. Briso, D. Moraes Jr., Effect of the outlet air
reuse on thermal efciency of a pilot plant spray dryer with rotary atomizer,
Chem. Eng. Trans. 32 (2013) 241e246.
[7] S. Rahman, Food Properties Handbook, CRC Press, New York, 1997.
[8] T.A.G. Langrish, T.K. Kockel, The assessment of a characteristic drying curve for
milk powder for use in computational uid dynamics modeling, Chem. Eng. J.
84 (2001) 69e74.
[9] K.C. Patel, X.D. Chen, Prediction of spray-dried product quality using two
simple drying kinetics models, J. Food Process Eng. 28 (2005) 567e594.
[10] X.D. Chen, S.X.Q. Lin, Air drying of milk droplet under constant and timedependent conditions, AIChE J. 51 (2005) 1790e1799.
n, J.E. Julia
, L. Herna
ndez, J.C. Jarque, Inuence of particle size on
[11] R. Mondrago
the drying kinetics of single droplets containing mixtures of nanoparticles and
microparticles: modeling and pilot-scale validation, Dry. Technol. 31 (2013)
759e768.
[12] N. Dalmaz, H.O. Ozbelge, A.N. Eraslan, Y. Uludag, Heat and mass transfer
mechanisms in drying of a suspension droplet: a new computational model,
Dry. Technol. 25 (2007) 391e400.
[13] M. Mezhericher, A. Levy, I. Borde, Theoretical drying model of single droplet
containing insoluble or dissolved solids, Dry. Technol. 25 (2007) 1025e1032.
[14] M. Mezhericher, A. Levy, I. Borde, Multi-scale multiphase modeling of transport phenomena in spray drying processes, Dry. Technol. (2014) in press.
http://dx.doi.org/10.1080/07373937.2014.941110.
[15] T.A.G. Langrish, Multi-scale mathematical modelling of spray dryers, J. Food
Eng. 93 (2009) 218e228.
[16] D.E. Dobry, D.M. Settell, J.M. Baumann, R.J. Ray, L.J. Graham, R.A. Beyerinck,
A model-based methodology for spray-drying process development, J. Pharm.
Innov. 4 (2009) 133e142.
[17] M. Ali, T. Mahmud, P.J. Heggs, M. Ghadiri, D. Djurdjevic, H. Ahmadian, L.M. de
Juan, C. Amador, A. Bayly, A one-dimensional plug-ow model of a countercurrent spray drying tower, Chem. Eng. Res. Des. 92 (2014) 826e841.
[18] W. Julklang, B. Golman, Analysis of slurry drying in a spray dryer, Int. J. Eng.
Technol. 5 (2013) 5178e5189.
[19] R. Kuriakose, Computational uid dynamics (CFD) applications in spray drying
of food products, Trends Food Sci. Technol. 21 (2010) 383e398.
[20] C. Anandharamakrishnan, Computational Fluid Dynamics Applications in Food
Processing, Springer, New York, 2013, pp. 11e25.
[21] F. Friedler, Process integration, modelling and optimisation for energy saving
and pollution reduction, Appl. Therm. Eng. 30 (2010) 2270e2280.
[22] Z.-X. Gong, A.S. Mujumdar, Software for design and analysis of drying systems, Dry. Technol. 26 (2008) 884e894.
[23] J.W. Eaton, D. Bateman, S. Hauberg, GNU Octave Version 3.0.1 Manual: A Highlevel Interactive Language for Numerical Computations, Create Space Independent Publishing Platform, 2009.
[24] M. Bilic, G. Glavac, Spray-drying simulation in spreadsheet, Dry. Technol. 10
(1992) 509e519.

913

[25] D. Velic, M. Bilic, S. Tomas, M. Planinic, Simulation, calculation and possibilities of energy saving in spray drying process, Appl. Therm. Eng. 23 (2003)
2119e2131.
[26] N. Grasmeijer, H. de Waard, W.L.J. Hinrichs, H.W. Frijlink, A user-friendly
model for spray drying to aid pharmaceutical product development, PLoS ONE
8 (2013) e74403.
[27] K.W.W. Wong, J.P. Barford, Teaching Excel VBA as a problem solving tool for
chemical engineering core courses, Educ. Chem. Eng. 5 (2010) 72e77.
[28] Z.B. Maroulis, G.D. Saravacos, A.S. Mujumdar, Spreadsheet-aided dryer design,
in: A.S. Mujumdar (Ed.), Handbook of Industrial Drying, third ed., CRC Press,
New York, 2007 (Chapter 5).
[29] T.T.-E. Boroze, J.-M. Meot, Y.O. Azouma, H. Desmorieux, K. Napo, Decision
making in designing and choice of dryer for tropical agricultural crop, Indian J.
Sci. Technol. 6 (2013) 5495e5506.
[30] T. Kudra, R. Platon, P. Navarri, Excel-based tool to analyze the energy performance of convective dryers, Dry. Technol. 27 (2009) 1302e1308.
[31] J. Yan, C.C. Zhou, R. Liao, J. Wang, Thermal balance calculation for sludge
drying-incineration system based on spreadsheet software, Adv. Mat. Res.
516-517 (2012) 124e128.
[32] K.C. Patel, X.D. Chen, Mathematical modelling for plug-ow spray dryer, in:
CD-ROM of Chemeca 2004 Conference, 26e28 September 2004, Sydney,
Australia, 2004.
[33] I.C. Claussena, T. Andresena, T.M. Eikevika, I. Strommena, Atmospheric freeze
drying e modeling and simulation of a tunnel dryer, Dry. Technol. 25 (2007)
1959e1965.
[34] A.S. Mujumdar, Handbook of Industrial Drying, third ed., CRC Press, New York,
2012.
[35] L. Ozmen, T.A.G. Langrish, A study of the limitation to spray dryer outlet
performance, Dry. Technol. 21 (2003) 895e917.
[36] ASHRAE Handbook: Heating, Ventilating, and Air-conditioning: Systems and
Equipment, ASHRAE Inc., 2000, pp. 44.1e44.19.
[37] T. Kudra, Energy aspects in drying, Dry. Technol. 22 (2004) 917e932.
[38] D.R. Lide (Ed.), CRC Handbook of Chemistry and Physics, 85th ed., CRC Press,
New York, 2005.
[39] Revised Release on the IAPWS Industrial Formulations 1997 for the Thermodynamic Properties of Water and Steam, IAPWS, Switzerland, 2007.
[40] K. Raznjevic, Handbook of Thermodynamic Tables, second ed., Begell House,
1995.
[41] N.B. Vargaftik, Y.S. Touloukian, Handbook of Physical Properties of Liquids and
Gases: Pure Substances and Mixtures, second ed., Springer, 1975.
[42] S. Herrmann, H.-J. Kretzschmar, D.P. Gatley, Thermodynamic Properties of
Real Moist Air, Dry Air, Steam, Water, and Ice, ASHRAE RP-1485, ASHRAE Inc.,
2011.
[43] K. Masters, Spray Drying Handbook, Longman Scientic and Technical, Harlow, 1985, p. 109.