7/26/2019 irvan dahlan

http://slidepdf.com/reader/full/irvan-dahlan 1/5

 Journal of Hazardous Materials 185 (2011) 1609–1613

Contents lists available at ScienceDirect

 Journal of Hazardous Materials

 j o u r n a l h o m e p a g e :   w w w . e l s e v i e r . c o m / l o c a t e / j h a z m a t

Short communication

Sorption of SO2 and NO from simulated flue gas over rice husk ash

(RHA)/CaO/CeO2 sorbent: Evaluation of deactivation kinetic parameters

Irvan Dahlan a, Keat Teong Lee b, Azlina Harun Kamaruddin b, Abdul Rahman Mohamed b,∗

a School of Civil Engineering, Universiti Sains Malaysia, Engineering Campus, Seri Ampangan, 14300 Nibong Tebal, Pulau Pinang, Malaysiab School of Chemical Engineering, Universiti Sains Malaysia, Engineering Campus, Seri Ampangan, 14300 Nibong Tebal, Pulau Pinang, Malaysia

a r t i c l e i n f o

 Article history:

Received 22 June 2010Received in revised form 7 October 2010

Accepted 13 October 2010

Available online 20 October 2010

Keywords:

Rice husk ash (RHA)

Sorbent

SO2/NO sorption

Breakthrough curves

Deactivation kinetic model

a b s t r a c t

In this study, the kinetic parameters of rice husk ash (RHA)/CaO/CeO 2  sorbent for SO2  and NO sorptions

were investigated in a laboratory-scale stainless steelfixed-bed reactor. Dataexperiments were obtained

from our previous results and additional independent experiments were carried out at different condi-

tions. The initial sorption rate constant (k0) and deactivation rate constant (kd) for SO2/NO sorptions

were obtained from the nonlinearregression analysisof the experimental breakthrough datausing deac-

tivation kinetic model. Both the initial sorption rate constants and deactivation rate constants increased

with increasing temperature, except at operating temperature of 170 ◦C. The activation energy and fre-

quencyfactorfor theSO2 sorption were found to be 18.0 kJ/moland 7.37×105 cm3/(g min), respectively.

Whereasthe activation energyand frequencyfactor forthe NOsorption,wereestimated to be 5.64 kJ/mol

and 2.19×104 cm3/(g min), respectively. The deactivation kinetic model was found to give a very good

agreement with the experimental data of the SO2/NO sorptions.

© 2010 Elsevier B.V. All rights reserved.

1. Introduction

Cleaning flue gases from sulfur oxides (SO x) and nitric oxides

(NO x) has become an issue of great importance to governmental

regulatory agencies and general public due to their negative effect

towards the environment and human health. Normally SO x   and

NO x, which consists of more than 98% of sulfur dioxide (SO 2) [1]

andover 90–95% of nitricoxide(NO) [2], are generated mainly from

the combustion of fossil fuels in power stations as well as chemical

plants and metallurgy processes.Attempts havebeen madeto finda

suitable methodfor theremovalof SO2 and NO simultaneously.Dry

sorption method is now considered to be the most attractive way

to treat waste gases containing SO2  and NO due to the drawbacks

of wet sorption methods [3,4]. There are several dry-type sorbents

that have been considered in the previous study for simultaneous

removal of SO2 and NO.RHA, which is produced from the burning of rice husk, has been

chosenin this study as a rawmaterial in thepreparationof dry-type

sorbent since it is available abundantly in rice-producing countries

like Malaysia. RHA also contains high amount of silica. However,

RHA has low sorption capacity when used alone to remove acidic

gases. Therefore, this agricultural waste-siliceous starting material

needs to be activated with other materials and the silica in RHA

∗ Corresponding author. Tel.: +60 4 5996410; fax: +60 4 5941013.

E-mail address:  chrahman@eng.usm.my (A.R. Mohamed).

plays an important role in the formation of reactive species which

is responsible for high sorption capacity [5,6].Previously, we had reported the sorption characteristics of SO2

and NO over rice husk ash (RHA)-based sorbent at low temper-

ature  [5–11].  Nevertheless, our previous reports only dealt with

activity measurement related to sorbent preparation conditions

and effects of reactor operating conditions. Our previous results

alsoshowed thatthe highest sorption capacity for the simultaneous

removal of SO2 andNO wasobtainedusing RHA/CaO/CeO2 sorbent.

Currently,the optimum preparative parameters for this kind of sor-

bent had also been reported [12]. On the other hand, the reaction

between the siliceous/calcium dry-type sorbents and SO2/NO is

very scarcely reported. The reaction between this siliceous/calcium

dry-type sorbents and SO2/NO are very complicated due to the

complex composition of the sorbent. The sorption of these pol-

lutant gases (SO2/NO) on the sorbents is not a simple physicalsorption processes, but also may be described as chemisorption or

as gas–solid non-catalytic reactions.

There are various kinetic models that have been employed to

estimate kinetic parameters in gas–solid reaction, mainly involves

single component sorbent (such as CaO, Ca(OH)2 and CaCO3)andit

wascarried outmainly at high operating temperature. These kinet-

ics models included shrinking unreacted core model [13], changing

grain size model [14] and random pore model [15]. Most of these

models contain large number of adjustable parameters related to

the pore structure, to the product layer and pore diffusion resis-

tances as well as the surface sorption rate parameters. In addition,

0304-3894/$ – see front matter © 2010 Elsevier B.V. All rights reserved.

doi:10.1016/j.jhazmat.2010.10.053

7/26/2019 irvan dahlan

http://slidepdf.com/reader/full/irvan-dahlan 2/5

1610   I. Dahlan et al. / Journal of Hazardous Materials 185 (2011) 1609–1613

it is complicated to incorporate them without having to perform

lengthy computer programs. Therefore in this study, the simplified

deactivation kinetic model was used to estimate kinetic parame-

ters against other models. The breakthrough curves data obtained

from our previous results (SO2 and NO sorptions) [11] was fitted to

deactivation kinetic model. In the present work,kinetic parameters

such as deactivation rate constant, initial sorption rate constant,

activation energy and frequency (pre-exponential) factor of the

SO2

 an NO sorptions were estimated from the breakthrough data

through nonlinear regression analysis. In chemical engineering,the

rate of reaction is a prerequisite to the design and evaluation of 

fixed-bed reactor performance especially under dry-type gas–solid

reaction–sorption processes.

2. Experimental

 2.1. Preparation of sorbent 

RHA-based sorbents (RHA/CaO/CeO2) were prepared from

rice husk ash (RHA), CaO (BDH Laboratories, England) and

Ce(NO3)3·6H2O (Fluka, 98%). The raw RHA was collected directly

without any pretreatment from Kilang Beras & Minyak Sin Guan

Hup Sdn. Bhd., Nibong Tebal, Malaysia. Prior to use, the RHA wassievedto produce less than 200m particle size. Thechemicalcom-

position of raw RHA was 68.0% SiO2, 2.30% K2O, 1.20% P2O5, 0.71%

MgO, 0.59% CaO, 0.32% SO3, 0.32% Cl2O, 0.16% Al2O3, 0.40% others

and 26.0% LOI. The preparation method was based on the optimum

hydration conditions reported in our previous studies  [12].

 2.2. Activity test 

The sorption/activity of the prepared sorbents was tested in a

laboratory-scale stainlesssteel fixed-bed reactor (Swagelok,10 mm

ID, 50cm length) which was vertically fitted in a tube furnace (Lin-

berg/Blue M). The schematic diagram and details of the activity

study is presented elsewhere [7]. The experiments were conducted

at various reactor temperature range of 70–170

◦

C while maintain-ing the simulated flue gas under the fixed condition of 2000 ppm

SO2, 500 ppm NO, 10% O2, 10% RH, balance N2  with total gas flow

rate of 150ml/min. Other operating conditions are given in our

previous study [11].

 2.3. Kinetic parameters estimation of RHA/CaO/CeO 2 sorbent 

using deactivation kinetic model.

The analysis of kinetic parameters was carried out using

breakthrough data of single component gases of SO2   and NO,

respectively. The deactivation kinetic parameters such as initial

sorption rate constants (k0) and deactivation rate constants (kd)

were calculated from breakthrough curve analysis. The outline of 

theanalysisusing deactivationkinetic modeling is given as follows.As in a typical gas–solid reaction, pore structure, active surface

area and activity per unit area of the solid reactant have significant

effects on the reaction rate. In the deactivation model, the effects

of all these factors are combined in an activity term (a) introduced

into the sorption rate expression and is written in Eq.  (1) [16].

−

da

dt   = kdC 

man (1)

where kd is the deactivation rate constants(min−1), C is theconcen-

tration of the reactant gas (kmol/m3), t  is the reaction time (min),

and m and n areexponential coefficients. Assuming thatthe concen-

tration of the reactant gas is independent along the reactor ( m = 0)

and the deactivation of the sorbent is first-order with respect to

the solid active site (n = 1), integration of Eq. (1) gives the following

expression.

a  =  a0  exp(−kdt ) (2)

Furthermore, the following basic assumptions were made in the

derivation of the deactivation model, such as isothermal and

pseudo-steady state conditions, and axial dispersion in the fixed

bedreactor andany mass transfer resistances were neglected. Con-

sidering these assumptions, and the initial activity (a0) of the solid

as unity, the pseudo-steady state species conservation equation forgases in the fixed bed reactor is given by Eq.  (3) [16–18].

−Q   dC 

dW   = k0Ca   (3)

where Q is thevolumetric flowrate(m3/min), W is thesorbentmass

(kg) and  k0  is the initial sorption rate constant (m3 kg−1 min−1).

Combining Eqs. (2) and (3) and solving these equations will yield

Eq. (4)

C 

C 0= exp[−k0B exp(−kdt )] (4)

whereby B  is equal to W /Q  and this kinetic model is known as the

zeroth solution of deactivation model, which predicts the behav-

ior of breakthrough curves for a gas–solid non-catalytic reaction.

This solution assumes a fluid phase concentration that is indepen-

dent of deactivation process along the reactor. However, it would

be reasonable to expect the deactivation rate to be concentration

dependent and axial position dependent in the fixed bed reactor.

In order to find analytical solutions of Eqs. (1) and (2) by con-

sidering concentration and axial position dependents in the fixed

bedreactor(m = n = 1), iterativeprocedurewas applied. The method

used was similar to the method for the estimated solution of non-

linear equations proposed by Dogu [19]. In this procedure, Eq. (4)

was substituted into Eq. (1)  with m = n = 1 and the first estimated

value forthe activity (a) term wasobtained byintegrating theequa-

tion. Then, the estimated value for the activity (a) term expression

was substituted into Eq. (3), and integration of this equation gave

the following corrected solution for the breakthrough curve.

C 

C 0= exp

1 − exp(k0B[1 − exp(−kdt )])

1 − exp(−kdt )  exp(−kdt )

  (5)

This Eq. (5) is also known as the solution of two-parameter deac-

tivation kinetic model. Deactivation rate constant (kd) and initial

sorption rateconstant (k0) wasthen calculated by using a nonlinear

regression technique.

A commercial software, MATHEMATICA ver. 5.2 (Wolfram

Research Inc.), was used for nonlinear regression analysis together

with the experimental/breakthrough datato find the rate constants

for the model. In order to obtain the best fitting results, an error

minimization technique was also applied and included after run-

ning the main program code of MATHEMATICA. MATHEMATICA

software was run under Microsoft Windows XP Professional (ver.

2002) environment.

Based on the analysis of the experimental breakthrough data at

different temperatures, the initial sorption rate constants (k0) can

be obtained by fitting Eq. (5) using nonlinear regression technique.

Then, Arrhenius equation [16] was used for the determination of 

activation energy and frequency (pre-exponential) factor for SO2

and NO sorptions at different temperatures, and is given in Eq. (6).

k0  = A  exp

−E a

RT 

  (6)

where A is a frequency (pre-exponential) factor, E a is the activation

energy, R  is the gas constant (8.314J/(mol K)) and T  is the temper-

ature (K).

7/26/2019 irvan dahlan

http://slidepdf.com/reader/full/irvan-dahlan 3/5

I. Dahlan et al. / Journal of Hazardous Materials 185 (2011) 1609–1613 1611

 Table 1

Rate parameters obtained from the breakthrough data at different temperature.

Temp. (◦C)   k0  (W/Q)   k0 (cm3 /(gmin))   kd (min−1)   R2

SO2  sorption NO sorption SO2 sorption NO sorption SO2  sorption NO sorption SO2 sorption NO sorption

70 4.06 10.22 1.22E+03 3.06E+03 0.12 0.11 0.987 0.989

87 5.84 11.00 1.75E+03 3.30E+03 0.15 0.12 0.975 0.990

100 8.57 11.45 2.57E+03 3.43E+03 0.20 0.126 0.972 0.954

120 10.85 13.38 3.25E+03 4.01E+03 0.21 0.128 0.983 0.976

150 13.12 14.56 3.93E+03 4.36E+03 0.23 0.130 0.991 0.964

170 11.69 15.78 3.50E+03 4.73E+03 0.24 0.135 0.965 0.957

3. Results and discussion

Fig. 1a and b shows the experimental SO2 and NO breakthrough

curves obtained under various operating temperatures, respec-

tively. The initial sorption rate constants (k0) and deactivation rate

constants (kd) values were estimated by nonlinear fitting of Eq. (5)

to the experimental SO2  and NO breakthrough curves at different

temperatures. The results of rate parameters from the regression

analysis of the dataobtained withRHA/CaO/CeO2 sorbents at differ-

enttemperatures aregiven in Table 1. The accuracy of theproposed

deactivation kinetic model was assessed from the coefficient of 

determination (R

2

) which was found to be 0.95 or higher. Otherkind of regression results (including statistical analysis) could be

obtained from the nonlinear regression analysis after running the

main program code of MATHEMATICA.

The initial sorption rate constants and deactivation rate

constants, as expected, increased with increasing temperature

(Table 1).  However, at operating temperature of 170 ◦C, the ini-

tial sorption rate constant for SO2  was decreased. The decrease in

the rate of SO2  sorption at higher temperatures might be due to

Fig. 1.   Effect of operating temperature on the (a) SO 2 and (b) NO sorptions.

water that accumulated and gas dissolving on the RHA/CaO/CeO2

sorbent surface was reduced  [11].  The predictions of the break-

through curves from Eq. (5) at different temperatures using these

rate constants are also shown in  Fig. 1, whereby the deactivation

kinetic model shows good agreement with the experimental data

at different temperatures. As predicted for SO2  sorption at high

temperature (170 ◦C), the breakthrough curves shifted to shorter

time (Fig. 1(a)). For NO sorption, the initial sorption rate con-

stants still increasedat high temperature (170 ◦C) andthe resulting

breakthrough curves shifted to longer time (Fig. 1(b)). This might

be attributed to a lesser amount of water accumulated on the

RHA/CaO/CeO2   sorbent surface thus allowing the metal species(CeO2) present in the sorbent to become more active  [11].

Based on the data obtained in  Table 1, Arrhenius equation (Eq.

(6)) was used for the estimation of activation energy (E a) and fre-

quency (pre-exponential) factor ( A) for SO2  and NO sorptions at

different temperatures. Fig.2(a)and(b)showsln k0 versus1/T plots

for SO2  and NO sorptions, respectively at different temperatures.

The plots were found to yield a straight line indicating that the

SO2 sorption

y = -2165.3x + 13.51

R 2 = 0.9428

7.0

7.2

7.4

7.6

7.8

8.0

8.2

8.4

8.6

 b

a

3.0E-032.8E-032.6E-032.4E-032.2E-032.0E-03

1/T (K -1

)

   l  n   (   k

  o   )

NO sorption

y = -678.48x + 9.9921

R 2 = 0.9844

7.9

8.0

8.1

8.2

8.3

8.4

8.5

3.0E-032.8E-032.6E-032.4E-032.2E-032.0E-03

1/T (K -1

)

   l  n   (   k

  o   )

Fig. 2.   Arrhenius plot of sorption rate constant versus reciprocal of operating tem-

perature for (a) SO2  and (b) NO sorptions.

7/26/2019 irvan dahlan

http://slidepdf.com/reader/full/irvan-dahlan 4/5

1612   I. Dahlan et al. / Journal of Hazardous Materials 185 (2011) 1609–1613

Fig. 3.  Comparison between predicted and experimental breakthrough curves at

two different experimental conditions.

sorption rate constant obtained from deactivation kinetic model do

follow the Arrhenius law as in Eq. (6). Accordingly, the slope of the

plot equal to E a/R and intercept equivalent to ln A, from which acti-

vation energy(E a) andfrequencyfactor ( A)forSO2 andNOsorptionscan be obtained, respectively.

The value of frequency factor ( A) for SO2 and NO sorptions were

calculated to be 7.37 ×105 cm3/(g min) and 2.19×104 cm3/gmin,

respectively. Whereas the activationenergy (E a) values determined

for the SO2   and NO sorptions were 18.0 kJ/mol and 5.64kJ/mol,

respectively. The activation energy of the SO2   sorption at low

temperature using the RHA/CaO/CeO2   sorbent was found to be

slightly higher as compared to sorbent prepared from coal fly

ash/Ca(OH)2   (14.94–15.47 kJ/mol) [20],   activated carbon from oil

palm shell with KOH impregnation (13.2 kJ/mol)  [21]   and acti-

vated carbon from oil palm shell (12.6kJ/mol)   [22].   However,

the activation energy obtained in this study was lower than

the SO2   sorption when Ca(OH)2   (32 kJ/mol)   [23]   and coal fly

ash/CaO/CaSO4  (22.9 kJ/mol)  [24]   were used as the sorbent, andalso much lower than the reported value by Irabien et al.   [25]

and Renedo and Fernandez [26] using Ca(OH)2 (75kJ/mol) and coal

fly ash/Ca(OH)2/CaSO4 (57.7 kJ/mol), respectively. Apart from that,

this activation energyfor SO2 sorption at lowtemperature wasalso

found to be similar as compared to sorbents prepared from vari-

ous type of CaCO3   (15.2–19.5 kJ/mol) [27]. For the case of the NO

sorption, the value of activation energy was also much lower than

previously reported in the literature which include the sorbent pre-

pared from V2O5/NH4Br/TiO2/SiO2  (30.1 kJ/mol) [28], V2O5–Al2O3

(53.56 kJ/mol) [29] and Fe-ZSM-5 (54 kJ/mol) [30]. However, most

ofthe reportedstudiesfor NOsorptionwerecarriedoutat high tem-

perature processes. The relatively small activation energy obtained

in this study suggested an easy sorption process of SO2 and NO by

this kind of sorbent. In other word, the sorption between SO2/NOand the reference sorbent synthesized from RHA/CaO/CeO2 is eas-

ier to occur due to the easier access of SO2 and NO molecules to the

active species in the sorbent.

In order to verify the proposed deactivation kinetic model,

additional independent experiments were carried out at different

conditions using 0.5g RHA/CaO/CeO2 sorbent. The first experiment

was conducted at initial condition of 1500 ppm SO2, 1200ppmNO,

10% O2, 60% RH, balance N2  and 150 ml/min of total flow rate at

a reactor temperature of 80 ◦C. While the second experiment was

conducted at the following conditions of 1800 ppm SO2, 800ppm

NO, 10% O2, 40% RH, balance N2  and 150 ml/min of total flow rate

at a reactor temperature of 110 ◦C. Fig. 3 shows the experimental

versus predicted breakthrough curves of SO2  and NO sorptions at

two different experimental conditions. It was shown that the deac-

SO2 sorption

0.0

0.1

0.20.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0a

 b

1.00.90.80.70.60.50.40.30.20.10.0

Experimental C/Co

   P  r  e   d   i  c   t  e   d   C   /   C  o

NO sorption

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.00.90.80.70.60.50.40.30.20.10.0

Experimental C/Co

   P  r  e

   d   i  c   t  e   d   C   /   C  o

Fig. 4.  Plot of all experimental C/C0 vs predicted C/C0 under various operating con-

ditions for (a) SO2  and (b) NO sorptions.

tivation kinetic model provided a very accurate description of the

experimental data.

For further confirmation, the breakthrough curvesdata fromour

previous results [11] was fitted to the proposeddeactivation kinetic

model. The comparison between predicted breakthrough curves

(obtained with deactivation kinetic model) with the experimen-

tal results was performed for all the SO2/NO sorption experiments

under various operating conditions.   Fig. 4(a) and (b) shows the

comparison between the experimental C /C 0  versus predicted C /C 0ofSO2 and NO sorptionsin all experiments, respectively. The results

indicated that the proposed model prediction agrees reasonably

well with the experimental data of the SO2/NO sorptions within

the range of 10% experimental error.

4. Conclusions

The deactivation model wasapplied successfullyto describe the

experimental breakthrough curves for the sorption of SO2

 and NO

from simulated flue gas in a fixed-bed reactor over RHA/CaO/CeO2

sorbent. Thebreakthrough data obtained forboth SO2 and NO sorp-

tions was fitted to the proposed deactivation kinetic model. Both

the initial sorption rate constants and deactivation rate constants

increased with increasing temperature, except at operating tem-

perature of 170◦C whereby theinitialsorptionrateconstantfor SO2

decreased. The breakthrough curves obtained by using the devel-

oped deactivationkinetic model were found to fit theexperimental

breakthrough curves very well.

 Acknowledgements

Theauthors wish to acknowledge thefinancialsupport from the

Ministry of Science, Technology and Innovation (MOSTI) Malaysia,

7/26/2019 irvan dahlan

http://slidepdf.com/reader/full/irvan-dahlan 5/5

I. Dahlan et al. / Journal of Hazardous Materials 185 (2011) 1609–1613 1613

Yayasan Felda andUniversiti Sains Malaysia (ShortTerm Grant A/C.

6035278 and RU Golden Goose Project Grant A/C. 814004).

References

[1] R.F. Probstein, R.E. Hicks, Synthetic Fuel, Dover Publications, New York,2006.

[2] J.L.Zhu,Y.H.Wang,J.C. Zhang,R.Y. Ma,Experimentalinvestigation ofadsorptionof NO and SO2  on modified activated carbon sorbent from flue gases, EnergyConvers. Manage. 46 (2005) 2173–2184.

[3] T.Ishizuka, H.Tsuchiai, T.Murayama,T. Tanaka,H. Hattori,Preparation ofactiveabsorbent fordry-typeflue gasdesulfurizationfrom calcium oxide, coalfly ash,and gypsum, Ind. Eng. Chem. Res. 39 (2000) 1390–1396.

[4] R.-B. Lin, S.-M. Shih, C.-F. Liu, Characteristics and reactivities of Ca(OH)2 /silicafumesorbents for low-temperature fluegas desulfurization, Chem. Eng.Sci. 58(2003) 3659–3668.

[5] I. Dahlan, K.T. Lee, A.H. Kamaruddin, A.R. Mohamed, Key factor in rice huskash/CaOsorbent for highflue gasdesulfurizationactivity, Environ.Sci. Technol.40 (2006) 6032–6037.

[6] I. Dahlan, K.T. Lee, A.H. Kamaruddin, A.R. Mohamed, Analysis of SO2  sorptioncapacity of rice husk ash (RHA)/CaO/NaOH sorbents using response surfacemethodology (RSM): un-treated andpre-treated RHA,Environ. Sci.Technol.42(2008) 1499–1504.

[7] K.T. Lee,A.M. Mohtar, N.F. Zainudin, S. Bhatia, A.R.Mohamed, Optimum condi-tions for preparation of flue gas desulfurization absorbent from rice husk ash,Fuel 84 (2005) 143–151.

[8] I. Dahlan, G.M. Mei, A.H. Kamaruddin, A.R. Mohamed, K.L. Lee, Removal of SO2  and NO over rice husk ash (RHA)/CaO-supported metal oxides, J. Eng. Sci.

Technol. 3 (2008) 109–116.[9] I.Dahlan, K.T.Lee, A.H.Kamaruddin,A.R. Mohamed,Evaluationof variousaddi-tives on the preparation of rice husk ash (RHA)/CaO-based sorbent for fluegas desulfurization (FGD) at low temperature, J. Hazard. Mater. 161 (2009)570–574.

[10] I. Dahlan, K.T. Lee, A.H. Kamaruddin, A.R. Mohamed, Selection of metal oxidesin the preparation of rice husk ash (RHA)/CaO sorbent for simultaneous SO2

and NO removal, J. Hazard. Mater. 166 (2009) 1556–1559.[11] I. Dahlan, K.T. Lee, A.H. Kamaruddin, A.R. Mohamed, Rice husk ash/calcium

oxide/ceria sorbent for simultaneous removal of sulfur dioxide and nitricoxide from flue gas at low temperature, Environ. Eng. Sci. 26 (2009) 1257–1265.

[12] I. Dahlan,Z. Ahmad, M. Fadly, K.T.Lee, A.H.Kamaruddin,A.R.Mohamed, Param-eters optimization of rice husk ash (RHA)/CaO/CeO2   sorbent for predictingSO2/NO sorption capacity using response surface and neural network models,

 J. Hazard. Mater. 178 (2010) 249–257.

[13] A. Ersoy-Mericboyu, S. Kucukbayrak, Modeling of the isothermal sulphationreactions of natural sorbents, Thermochim. Acta 319 (1998) 163–170.

[14] Y. Yan, X.F. Peng, G.G. Fu, B.X. Wang, Study on kinetics of desulfurization reac-tion by CaO particles, Heat Transfer Asian Res. 32 (2003) 391–401.

[15] Y. Suyadal,H. Oguz, Drydesulfurization of simulatedflue gasin a fluidized-bedreactor fora broad rangeof SO2 concentration and temperature:a comparisonof models, Ind. Eng. Chem. Res. 38 (1999) 2932–2939.

[16] O. Levenspiel, Chemical Reaction Engineering, 3rd ed., John Wiley & Sons, Inc.,New York, 1999.

[17] S. Yasyerli, G. Dogu, I. Ar, Activities of copper oxide and Cu–V and Cu–Momixed oxides for H2S removal in the presence and absence of hydrogen and

predictions of a deactivationmodel, Ind.Eng. Chem. Res.40 (2001) 5206–5214.[18] R.B.Bird,W.E. Stewart,E.N.Lightfoot,TransportPhenomena,JohnWiley & Sons,

Inc., Singapore, 2007.[19] T. Dogu, Extensionof moment analysisto nonlinear systems,AIChEJ. 32 (1986)

849–852.[20] A. Garea, J.R.Viguri,A. Irabien,Kineticsof fluegas desulphurizationat lowtem-

peratures: fly ash/calcium (3/l) sorbent behaviour, Chem. Eng. Sci. 52 (1997)715–732.

[21] J. Guo, A.C. Lua, Adsorption of sulfur dioxide onto activated carbons preparedfromoil-palmshellsimpregnatedwith potassium hydroxide, J. Chem. Technol.Biotechnol. 75 (2000) 971–976.

[22] A.C.Lua, J. Guo,Adsorption of sulfur dioxide on activatedcarbon fromoil-palmwaste, J. Environ. Eng. 127 (2001) 895–901.

[23] J.F. Izquierdo, C. Fite, F. Cunill, M. Iborra, J. Tejero, Kinetic study of the reactionbetween sulfur dioxide and calcium hydroxide at low temperature in a fixed-bed reactor, J. Hazard. Mater. B76 (2000) 113–123.

[24] K.T. Lee, Flue gas desulfurization studies using absorbent prepared from coalfly ash, Ph.D. Thesis, Universiti Sains Malaysia, 2004.

[25] A. Irabien, F. Cortabitarte, M.I. Ortiz, Kinetics of flue gas desulfurization at low

temperatures: nonideal surface adsorption model, Chem. Eng. Sci. 47 (1992)1533–1543.

[26] M.J. Renedo, J. Fernandez, Kinetic modelling of the hydrothermal reaction of fly ash, Ca(OH)2  and CaSO4 in the preparation of desulfurant sorbents, Fuel 83(2004) 525–532.

[27] K. Han, C. Lu, S. Cheng, G. Zhao, W. Wang, J. Zhao, Effect of characteristics of calcium-based sorbents on the sulfation kinetics, Fuel 84 (2005) 1933–1939.

[28] T. Shikada, T. Oba, K. Fujimoto, H. Tominaga, Cleaning of flue gas by use of molten salts. 1. Simultaneous reduction of sulfur oxides (SO x) and nitrogenoxides (NO x) with V2O5   catalyst, Ind. Eng. Chem. Prod. Res. Dev. 23 (1984)417–421.

[29] I.-S.Nam, J.W.Eldridge, J.R.Kittrell,Deactivation of a vanadia–aluminacatalystfor NO reduction by NH3, Ind. Eng. Chem. Prod. Res. Dev. 25 (1986) 192–197.

[30] H.Y. Huang, R.Q. Long, R.T. Yang, Kinetics of selective catalytic reduction of NOwith NH3  on Fe-ZSM-5 catalyst, Appl. Catal. A 235 (2002) 241–251.