Response surface modeling of the removal of methyl orange ...

IP 51031210 On Fri 21 Jan 2022 090109Copyright American Scientific Publishers

Delivered by Ingenta

Materials ExpressArticle

Copyright copy 2018 by American Scientific PublishersAll rights reservedPrinted in the United States of America

2158-584920188234011doi101166mex20181433

wwwaspbscommex

Response surface modeling of the removal ofmethyl orange dye from its aqueous solution usingtwo types of zeolite synthesized from coal fly ashTahani Al-Dahri1 Adnan AbdulJabbar AbdulRazak12lowast Intisar Hussain Khalaf12lowast and Sohrab Rohani1

1Department of Chemical and Biochemical Engineering Western University London ON N6A 5B9 Canada2Department of Chemical Engineering University of Technology 52 AlSinaa Street Baghdad Iraq

ABSTRACT

Coal fly ash (CFA) produced as a waste material in coal power plants was converted into Linde-type A (LTA)and Na-P type (ZP) zeolite using microwave irradiation The prepared zeolites were characterized and modifiedusing a cationic surfactant The removal of methyl orange (MO) dyes from aqueous solution by LTA and ZP asadsorbent was studied by applying the response surface method (RSM) A quadratic model was developed torelate the MO removal rate to the main independent variables namely the solution pH (in the 3ndash9 range) theinitial MO concentration (in the 20ndash100 mgL range) and the adsorbent mass (200ndash1000 mgL) The modifiedMLTA was found to be a more effective adsorbent relative to the modified MZP In addition MLTA mass exertedthe greatest influence on the MO removal process under the studied conditions However when MZP zeolitewas used as an adsorbent the pH of the solution was the most influential factor The two quadratic modelswere found to be statistically significant

Keywords Methyl Orange Coal Fly Ash Zeolite LTA Zeolite Na-P Microwave Synthesis

1 INTRODUCTIONTreatment of wastewater produced in the textile indus-try has been the subject of extensive research due to thelarge volume of water required (up to 100 lkg of fin-ished textile product) These effluents have a significantdetrimental impact on the water resources because theycontain slowly- or non-biodegradable organic materials Atpresent different treatment methods are utilized to removethe dyes from wastewater including those based on biolog-ical treatment floatation coagulation membrane filtrationand adsorption techniques However the economic costeffectiveness and environmental impact of these methodsdiffer significantly12

lowastAuthors to whom correspondence should be addressedEmails adnansss2002yahoocom intisa59yahoocom

Adsorption is the most common dye removal methodbecause it does not produce dangerous byproducts andsome of the adsorbents can be easily regeneratedMethyl orange (MO) is a well-known anionic dye fre-

quently utilized in various industries such as dye man-ufacturing plants and textile industries and as acid andbasic indicator Discharge of effluents containing MO canhave detrimental environmental consequences as it canadversely affect human health and the ecosystem of otherliving organisms Empirical evidence indicates that releaseof MO dyes without proper treatment can cause signif-icant health problems3 In addition presence of even avery low concentration of MO dye in water system mayreduce the light penetration and affect the photosyntheticprocessesRecently various adsorbentsmdashsuch as zeolite3 multi-

wall carbon nanotubes4 graphene oxide5 and calcined

234 Mater Express Vol 8 No 3 2018



Materials ExpressResponse surface modeling of the removal of methyl orange dyeAl-Dahri et al

Article

ZnMgAl hydrotalcite6 have been utilized for MOremovalCoal fly ash (CFA) is a powder produced as a byprod-

uct of burning coal to generate electricity In the EUand the US around 115 million tons of CFA are pro-duced per year7 Presently fly ash is primarily used asadditive in cement and concrete production Due to itshigh silica and alumina content CFA is also convertedto different types of zeolite by hydrothermal treatmentBukhari et al recently reported on this hydrothermal con-ventional method assisted by microwave and ultrasonicenergy8 Microwave energy is employed in the zeolitesynthesis process because this can yield different zeo-lite morphologies which not only leads to different zeo-lite types but also accelerates nucleation shortens thecrystallization time reduces the undesirable phases due tophase selectivity and facilitates control of the particle sizedistribution9 Zeolites are used in various applications asadsorbents ion exchangers catalysts and membranes Theapplications and utilization of synthetic zeolites from CFAwere reviewed by Ahmaruzzaman10

To the best of the authorsrsquo knowledge zeolite extractionfrom CFA by using the microwave assisted method as ameans of removing dyes from a wastewater stream hasrarely been studiedAs zeolites possess a net negative charge their adsorp-

tion of an anionic dye would be relatively low due to therepulsive forces between the zeolite surface and the dyeThus to make zeolites suitable for adsorption of anionicdyes the negative surface charge must be modified byusing cationic surfactants such as hexadecyltrimethylam-monium bromide (HDTMA)The aim of this work was to synthesize Linde-type

A (LTA) and Na-P type (ZP) from CFA by using themicrowave assisted method These zeolites were furthermodified by HDTMA before being utilized as adsorbentto eliminate the MO dye from the aqueous solution

2 EXPERIMENTAL DETAILS21 ChemicalsCoal fly ash (CFA) was supplied by a coal-fired powerplant (Ontario Power Generation Nanticoke Canada)and was stored in a sealed container until requiredNa2SiO35H2O (sodium met silicate) Na2OAl2O33H2O(sodium aluminate anhydrous) and HDTMA bromidewere purchased from Sigma-Aldrich USA while NaOH(sodium hydroxide) was supplied by Alphachem Mis-sissauga Canada Methyl orange with an MW =32734 gmole was purchased from Alfa Aesar USA Thechemical structure of MO is shown in Figure 1 Deionizedwater (DI) was used for the preparation of the solutionsand washing of synthesized zeolite All chemicals (includ-ing those used for characterization tests) were of analyticalgrade and were used as received

Fig 1 MO chemical structure

22 Synthesis ProcedureThe CFA zeolitization process of CFA commenced withdissolving 2 g of CFA in 01 M of HCl and heating themixture for 3 hours at 90 C as acid treatment enhancesCFA purity by removing impurities11 Subsequently themixture was filtered and the solid phase was washed withdeionized water and was left to try overnight at room tem-perature Next 2 g of acid treated CFA 15 M of NaOHgranules 04 g of NaAlO3 and 007ndash1 g of NaSiO4 wereadded to 20 mL of distilled water to produce the reactionsolution of Linde-type A (LTA) and Na-P type (ZP) zeo-lite respectively91213 The obtained samples were sub-jected to microwave irradiation for a certain period oftime and a fixed power required for crystallization A self-adjusting microwave (single-mode 25 GHz CEM coop-eration Discover USA) was used to conduct experimentsat atmospheric pressure A reflux condenser was attachedto a cylindrical batch PTFE vessel (28 mm IDtimes108 mm)and was placed in the microwave chamber for 30 min at250 W microwave power The samples were subsequentlyfiltered and the solid phase was collected washed withdeionized water and dried at room temperature overnight

23 CharacterizationsMicroscopic images revealing phase purity and morphol-ogy of the zeolitized CFA were obtained by using scan-ning electron microscope (SEM) Hitachi S 2600N SEM(Tokyo Japan) operating at 5 kV accelerating voltagePowder X-ray diffraction (XRD) images of the synthe-sized ZP and LTA zeolites were obtained by RigakundashMiniflex powder diffractometer (Japan) with CuK ( forK = 154059 Aring) over the 5 lt 2 lt 40 range in 2

increments The obtained crystalline phase was verified bycomparing the characteristic peaks of ZP and LTA withthe standard peaks reported in the pertinent literature14

The thermogravimetric analysis (TGA) of the sampleswas performed using a Mettler Toledo TGASDTA 851emodel (Switzerland) coupled with Stare software version61 The samples were heated from 25 C to 600 C in10 Cmin steps under nitrogen purge Specific surfacearea and pore size of the synthesized zeolite (ZP and LTA)were measured using a Micromeritics ASAP 2020 Sur-face Analyzer while a Zetasizer Nano ZS 3000 HSA(Malvern Worcestershire UK) was used for measuring thezeta potential of the product particles

24 Modification of as-Synthesized ZeolitesFor the modification of LTA and ZP 500 mL of the solu-tion of the cationic surfactant HDTMA bromide at a

Mater Express Vol 8 2018 235



Materials ExpressResponse surface modeling of the removal of methyl orange dye

Al-Dahri et al

Article

Table I Process-independent variables and their levels

Levels and range

Variables (+1) (0) (minus1)

pH of the solution (X1 9 6 3Initial MO dye concentration (X2 (mgL) 100 60 20Mass of the adsorbent (X3 (mgL) 1000 600 200

concentration of 55 mmolL was strongly mixed with 10 gof LTA at 50 C for 4 h The reaction settings were cho-sen to assure proper surfactant loading After being cooledto room temperature the product was centrifuged washedwith doubly distilled water dried in an oven at 60 C andground to pass through an 80-mesh sieve

25 Experimental DesignResponse Surface Method (RSM) was employed in thiswork as it is a well-known statistical technique uti-lized for designing experiments in various engineeringand research disciplines1516 To investigate the influenceof the independent variables the solution pH the initialdye concentration and the adsorbent mass on the adsorp-tion process a three-factorial three-level central compositedesign (CCD) was adopted Seventeen experimental runswere required including five replicates at the center pointsThe experiments were designed using Design-Expert 7software Table I shows the low center and high levels ofthe independent variables investigated in this study

26 Removal ProcessBatch mode experiments were carried out randomly tostudy the influence of the three independent variables onthe MO removal by the MLTA and MZP

Table II CCD for three independent variables and the obtained dyeremoval

MO dye removal

Run no X1 X2 X3 (MZP) (MLTA)

1 600 6000 60000 2709 5352 600 6000 60000 278 5353 900 6000 20000 347 13414 600 10000 20000 1872 20525 900 10000 60000 4935 23416 600 2000 20000 3094 4017 600 6000 60000 276 53668 900 6000 100000 2174 86289 600 6000 60000 271 539610 300 6000 100000 6059 923111 600 6000 60000 272 528612 600 2000 100000 91 99913 600 10000 100000 3199 562414 300 6000 20000 3613 733315 300 2000 60000 8273 824416 900 2000 60000 5864 90217 300 10000 60000 6318 902

In the dye removal process two adsorbents were utilizedwith different MLTA and MZP masses (200ndash1000 mgL)at a fixed pH (3 6 and 9) and an initial concentrationof (20ndash100 mgL) after the equilibrium was reached theadsorbents separating from the solution using a centrifugeThe residual MO concentration in the solution was mea-

sured using the UV-visible (Cary 60 Agilent TechnologyGermany) at 464 nm (max of MO)The dye removal percentage was calculated by applying

the following equation

MO dye removal= Co minusCf

Co

lowast100 (1)

where Co and Cf are the initial and final MO concentra-tion (mgL) The details of the experiments are shown inTable II

3 RESULTS AND DISCUSSIONMicrowave heating (MW) is a volumetric heating processdue to ionic conduction and dipolar nature of H2O There-fore MW heating is preferable to conventional heatingsince its two aforementioned characteristics shorten thereaction time while increasing the yield and selectivity17

CFA is converted to zeolite via three main steps18 Dur-ing the first step Si and Al dissolve from CFA particlesinto the solution In the second step alumina-silicate gel isformed in the solution In the last step the zeolite nucleithat formed in previous stage are grown and crystalized tozeolite particles on the CFA surface In the present studyexposing the mixture to MW irradiation assisted with dis-solving the Si and Al from the CFA particles into the solu-tion and enhanced the crystal growth Thus zeolite ZP andLTA were deposited on the CFA surface9

31 Characterization of LTA and ZPFigure 2 presents the nitrogen adsorptiondesorptionisotherms of the ZP and LTA Both show isotherms of

Fig 2 Adsorptiondesorption of ZP and LTA synthesized from CFA

236 Mater Express Vol 8 2018




Article

type IV based on the International Union of Pure AppliedChemistry (IUPAC) isotherm classification1920 typical ofthe mesoporous materials This finding could be due tothe presence of impurities trapped inside the pores andchannels of the synthesized zeolite which limit nitrogenadsorption inside the pores Table III summarizes the tex-tural properties of ZP LTA and raw CFA The Brunauer-Emmett-Teller (BET) surface area of ZP was found tobe 37807 m2g higher than that of the raw CFA samplewhich had a surface area of 1547 m2g The surface areaof LTA was 4847 m2g which is less than the ZP sur-face area The pore size of ZP and LTA was similar as1189336 Aring and 1106834 Aring respectively were measuredOn the other hand the micro pore volume of LTA was0134135 cm3g greater than that of ZP (0015059 cm3g)The higher pore volume may be one of the main reasonsfor the higher removal efficiency of LTA compared to ZPIn both cases however the HCl acid treatment of CFAprior to the microwave-assisted synthesis step resulted inproducing ZP and LTA with greater surface area and poresizeObviously the pore volume of zeolite ZP is much lower

than that of LTA while its surface area is much higherthan that of LTA These characteristics could be attributedto the porosity of both zeolites Based on their crys-tal structuresmdashcubic structure for LTA and fiber struc-ture for ZPmdashzeolites have various connective pores21

In zeolites pores grow from the radical crystalline struc-ture which follows configurational diffusion22 Zeolitescomprise of primary tetrahedral units whereby Si and Alatoms are located at the center of the tetrahedral and aresurrounded by four oxygen atoms located at the cornersAs SindashO and AlndashO cannot fill the space perfectly cavitiesare produced23 These cavities could result in porositywhich increases the zeolite surface area Owing to thischaracteristic the surface area of ZP is much higher thanthat of LTA while the pore volume of ZP is much lowerthan that of LTAThe samples were analyzed by XRD after being sub-

jected to microwave irradiation to determine the zeolitephases present Figure 3 shows that CFA was success-fully converted into zeolite ZP and LTA and the maincharacteristic peaks were observed at 125 17 21328 318 and 33 2-theta for ZP and 725 10261256 162 218 24 272 301 and 343 2-thetafor LTA Moreover a major zeolitic phase was identi-fied namely zeolite Na-P (ZP) and Linde-Type A (LTA)

Table III BET surface area of CFA and synthesized ZP and LTA

Surface area Pore volume Pore size(m2g) (cm3g) (Aring)

CFA 1547 ndash ndashLTA 4847 0134135 1106834ZP 37807 0015059 1189336

Fig 3 The XRD patterns of CFA ZP and LTA (A = zeolite Na-AQ = quartz M = mullite H = hematite M = magnetite P = zeoliteNa-P H= hydroxy-sodalite and C= cristobalite)

The average crystalline size of ZP and LTA as calculatedusing Scherrerrsquos equation was found to be 31 and 43 nmrespectivelySEM images obtained for raw CFA and synthesized LTA

and ZP zeolites are shown in Figures 4(a)ndash(c) Figure 4(b)shows one major structure observed namely a fiber struc-ture associated with the Na-P zeolite2425 Figure 4(c)shows another major cubic structure related to the Linde-Type A (LTA) framework26 These observations confirmthe BET and XRD results In addition they concur withthe findings reported in pertinent literature12 confirmingthat ZP and LTA structures cover the undissolved CFAsurfaceTGA curves of the synthesized ZP and LTA zeolites

are shown in Figure 5 Both samples exhibit a significantweight loss at approximately 105ndash128 C correspondingto sim14 and sim3 of the initial ZP and LTA weightrespectively while 5 was measured for the raw CFAThese findings demonstrate the high water holding capac-ity of synthesized ZP and LTA zeolite compared to the rawCFA and confirm the BET resultsThe zeta potential of MLTA and MZP versus the pH of

the solution is shown in Table IV The surface charges for

(b) (c)(a)

Fig 4 SEM of (a) CFA and synthesized (b) ZP and(c) LTA zeolites





Al-Dahri et al

Article

Fig 5 TGA of CFA and synthesized ZP and LTA zeolite

both types of adsorbent were positive and had the highestvalue at pH 6

32 Dye Removal Regression EquationsRegression equations of MO dye removal using LTA andZP as adsorbent are shown below The quadratic modelequations were obtained by applying the Design ExpertSoftware

MO dye removal (ZP)

=2736minus1923lowastX1minus1806lowastX2+1451lowastX3

minus854lowastX1lowastX2minus155lowastX1lowastX3minus1170lowastX2lowastX3

+617lowastX21+1885lowastX2

2minus304lowastX23 (2)

MO dye removal (LTA)


minus1864lowastX1lowastX2+1348lowastX1lowastX3minus602lowastX2lowastX3



where X1 is the pH of the solution X2 denotes the ini-tial MO dye concentration (mgL) and X3 pertains to theadsorbent mass (mgL)Positive sign in the above equations reveal an increase

in the value of the variable results in an increase in the dye removalTo study the adequacy and the significance of the cur-

rent models ANOVA test was performed and the resultsare shown in Table V The p value was set at lt005 and

Table IV Zeta potential of modified zeolite (MLTA) and MZP versuspH

pH MLTA (mV) MZP (mV)

3 24 206 327 258 158 89 10 4

the F -value was high at 5759 and 84710 for MZP andMLTA respectively These results confirm that the modelsare statistically significantAccuracy of the models fit was checked by calculat-

ing the correlation coefficient R2 which had the value of9860 and 9991 for MZP and MLTA respectivelyThese values indicate that only 14 and 009 of thetotal data variations were not captured by the current mod-els Moreover the high values of adjusted determinationcoefficients (Adj = 9695) and (Adj = 9979) whichwere close to R2 confirm the high statistical significanceof these modelsThe predicted and actual values of the dye removal

are plotted in Figures 6(a) and (b) These plots show thatthe predicted values provide a good fit for the experimentalresults pertaining to the MO removal using LTA and ZPSimilar results were reported by Singh et al16

Influence of the independent variables and theirjoint effects are presented in Pareto charts shown inFigures 7(a) and (b) In these graphs the influence of thevariables is denoted by the bar length As can be seenin Figure 7(a) when MLTA was employed the amountof the adsorbent (X3 had the greatest and positive effectin the MO dye removal followed by X1 X2 X1ndashX2

interaction (X12 X1ndashX3 interaction X2ndashX3 interaction

(X22 and (X3

2 The terms X1 X2 X1ndashX2 X2ndashX3 and(X3

2 had a negative effect while the remaining termshad a positive effect For ZP the Pareto chart shown inFigure 7(b) suggests that the pH of the solution (X1 hadthe greatest effect on the MO removal process followedby the initial MO dye concentration (X2 the amountof the adsorbent (X3 (X2

2 X2ndashX3 interaction X1ndashX2

interaction and (X12 while (X3

2 and X1ndashX3 interac-tion had no effect The negative sign of the terms inPareto charts indicates that the increase in the term causesa decreases in MO dye removal (antagonistic effect)which was the case for X1 X2 X2ndashX3 interaction andX1ndashX2 interaction Conversely X3 (X2

2 and (X12 had a

synergistic effect as these terms increased the MO dyeremovalFrom ANOVA results the percentage contributions (PC)

of each independent variable in the MO dye removal werecalculated and the results are shown in Table VI where SSis the sum of the squares of these terms calculated usingthe equation below27

PC = SSsum

SStimes100 (4)

As shown in Table VI using ZP as the adsorbent the pHof the solution (X1 showed the highest effect on MO dyeremoval as its contribution was 302 while for LTDthe adsorbent mass exerted the highest effect at 3819These results confirm the findings reported in the Paretocharts





Article

Table V Results of ANOVA analysis for MZP and MLTA

Sum of squares Mean squares F -value p-value prop gt F

Source MZP MLTA df MZP MLTA MZP MLTA MZP MLTA

Model 98245 114968 9 109161 127743 5759 84710 lt00001 lt00001 SignificantX1 29585 195209 1 295852 195210 15607 12945 lt00001 lt00001X2 261015 186836 1 261015 186836 13770 12389 lt00001 lt00001X3 16837 438797 1 168369 438797 8882 29098 lt00001 lt00001X1X2 2918 138975 1 29183 138975 1540 92159 00057 lt00001X1X3 96 72634 1 959 72634 051 48166 05000 lt00001X2X3 5474 14508 1 54740 14508 2888 9621 00010 lt00001X2

1 16016 96089 1 16016 96090 845 63720 00228 lt00001X2

2 14953 3696 1 149532 3696 7888 2451 lt00001 00017X2

3 3898 21663 1 3898 2166 206 1437 01947 00068Residual 13269 1056 7 1896 151 SignificantLack of fit 13228 990 3 4409 330 42821 2027 lt00001 00070Pure error 041 065 4 010 016Cor total 99572 115074 16

33 3D Response Surface and Contour PlotsIn three-dimensional (3D) response surface plots MOremoval as the dependent variable was plotted against twoindependent variables while keeping the third independent

(a)

(b)

Fig 6 Actual versus predicted values of MO dye removal(a) MLTA and (b) MZP

variable constant In addition 3D response surfaces andtheir analogous 2D contour plots which were con-structed using the model equation can assist in eluci-dating the influence of the independent variables on theresponse28

Based on the findings reported in extant literature theremoval of MO dye is highly dependent on pH of thesolution29 Figures 8(a) and (b) show the interactive effect

10788

ndash7196

ndash7049

ndash6072

5049

4389

ndash1961

99

ndash758

0 20 40 60 80 100 120

X3

X1

X2

X1X2

X1^2

X1X3

X2X3

X2^2

X3^2

Standardized effect estimate

Qua

drat

ic m

odel

term

s

(a)

ndash25

ndash2346

1886

178

ndash1075

ndash78

582

ndash29

14

0 5 10 15 20 25 30

X1

X2

X3

X2^2

X2X3

X1X2

X1^2

X3^2

X1X3


Qua

drat

ic m

odel

term

s

(b)

Fig 7 Pareto chart for (a) MLTA and (b) MZP





Al-Dahri et al

Article

Table VI The percentage contributions (PC) for MZP and MLTA

PC

Source X1 X2 X3 X1ndashX2 X1ndashX3 X2ndashX3 X21 X2

2 X23

MZP 3020 2665 1719 298 0098 559 164 1527 0398MLTA 1699 16262 3819 12096 6322 1263 8364 0322 019

of the pH of the solution and the initial MO dye con-centration on the dye removal at a fixed mass of theadsorbent It is obvious that the dye removal decreaseswith the increase in the solution pH as well as the ini-tial MO dye concentration A maximum dye removal isobserved at pH = 6 which may be attributed to the factthat the MO dye consists of the (ndashSO3Na) group Thusin aqueous solution it dissociates whereby anionic form

(a)

(b)

Fig 8 Response surface for MO removal at constant mass of adsorbent (1000 mgL) by (a) MLTA and (b) MZP respectively

of the dye can emerge Hence because of the electrostaticattraction between positive charges of the modified LTAand ZP and MO dye can be efficiently removed As shownin Table IV for both zeolites the zeta potential increasesas the pH increase until pH 6 when it starts to decreaseHence the dye removal increases as the number of pos-itive charge sites on the surface increases Moreover themodified LTA zeolite was found to be more effective when





Article

(b)

(a)


compared with modified ZP zeolite because its modifiedsurface is more positive than that of ZP yielding morepositive active sites In addition the combined effect ofinitial dye concentration and the solution pH is shown inFigures 8(a) and (b) where it can be seen that maximum dye removal as 100 and 91 is attained at pH 617 and515 and initial MO dye concentration 20 and 2024 mgLfor MLTA and MZP respectively adsorbentsIn addition it is clearly demonstrated that the initial

MO dye concentration had a negative effect on the dyeremoval under these conditions these is due to at lowdye concentrations there is a sufficient number of activesites on the surface of the adsorbed material for a reducednumber of dye molecules However when the initial dyeconcentration is increased the number of active sites isinsufficient to adsorb the dye molecules The contour curve

placed down in the 3D plots gives an indication on thetype of the interaction between the independent variablesIf the shape is circular or a parallel plane it suggestsabsence of interaction whether the shape is elliptical orcurved3031 Figures 8(a) and (b) depict an elliptical curveof contour plots indicating a good interaction between theinitial MO dye concentration and the pH of the solutionfor both adsorbentsFigures 9(a) and 9(b) depict the influence of the adsor-

bent mass and the solution pH at a constant initial MO dyeconcentration It can be seen from the plots that the adsor-bent mass had a positive effect on the dye removal Forboth modified adsorbents in the studied solution pH rangethe dye removal increased with the increase in mass Thisbehavior can be attributed to the fact that when adsorbentmass is high (1000 mgL) the available active sites at the





Al-Dahri et al

Article

adsorbent surface are sufficient to interact with the MOions ensuring high dye removal rate Conversely at lowadsorbent mass (200 mgL) certain percentage of activesites will be occupied and the saturation state will beachieved whereby the adsorbent is unable to adsorb moreMO molecules Similar findings were reported by Noori-motlagh et al for the uptake of acid orange by activatedcarbon32 As can be seen in Figures 9(a) and (b) for LTAthere is interaction between the mass of the adsorbent andthe pH of the solution due to the elliptical nature of thecontour plots However for the ZP there is no interac-tion in the effect of these variables due to the contour plotshape which is a parallel straight line These results arein good agreement with the Pareto chartsFigures 10(a) and (b) show the joint effect of the MO

initial dye concentration and the amount of the adsorbenton the MO removal when the pH of the solution is keptconstant For both adsorbents the positive effect of theamount of the adsorbent and the negative effect of the MOinitial dye concentration on the MO dye removal is inagreement with the results yielded by Eqs (2) and (3)In the entire range of the adsorbent mass examined inthis study the MO removal decreases as the initial dyeconcentration increases This trend can be explained by

(a)

(b)

Fig 10 Response surface for MO removal at constant pH 6 by (a) MLTA and (b) MZP respectively

the fact that at a low initial concentration (20 mgL) theMO ions present in the solution can interact with acces-sible binding sites However at a high dye concentration(100 mgL) due to the limited number of the energeticadsorption sites the adsorbent cannot interact with all MOmolecules in the solution33

Based on the above discussion the trends inFigures 10(a) and (b) indicate that the interaction betweenthe initial MO dye concentration and the mass of theadsorbent is in good agreement with the previous conclu-sion and the findings reported in Pareto charts for bothadsorbents

34 Selection of Optimum ConditionsThe DESIGN EXPERT software was applied to optimizethe MO removal The input variables were ldquowithin therangerdquo and the response ( dye removal) for both adsor-bents was chosen as ldquomaximumrdquo The desirability val-ues of 0975 and 0973 were obtained for LTA and ZPrespectively The maximumMO dye removal was obtainedas 100 and 91 at pH 617 and 515 the initial MOdye concentration of 20 and 2024 mgL and adsorbentmass of 1000 and 1000 mgL respectively To assess thevalidity of the quadratic models developed in this work





Article

the corresponding experimental MO dye removal val-ues were determined at the optimum conditions as 100and 91 which was in good agreement with the predictedvalues Therefore the two models presented here may beused to optimize the dye removal process and predict theMO removal

4 CONCLUSIONIn this study two types of zeolite namely Na-P andLinde type-A were prepared by microwave-assisted syn-thesis of waste CFA Acid pretreatment of raw CFAresulted in high-purity ZP and LTA zeolite with highpore volume and surface area The synthesized zeoliteswere modified by cationic surfactant (HDTMA) bromideThe dependency of the MO removal process on the pHof the solution the initial dye concentration and the massof the modified zeolites was studied utilizing the RSMThe CCD method was applied to develop a quadraticmodel in each case The predicated values calculatedusing the quadratic models were in good agreement withthe response ( dye removal) for LTA and ZP (R2 =09991 and 0986 respectively) The pH of the solutionhad the greatest effect when using ZP as an adsorbentwhereas the adsorbent amount had the greatest influenceon the dye removal rate when LTA was employedANOVA results indicate presence of interaction between

all the variables except pH and the initial dye concentra-tion when using ZP as an adsorbentThe results obtained from the optimization of the MO

removal process (100 at pH 617 the initial MO con-centration of 20 mgL and 1000 mgL adsorbent mass)and for LTA as adsorbent 91 at pH 515 with the ini-tial MO dye concentration of 2024 mgL and adsorbentmass of 1000 mgL for ZP as an adsorbent Finally usingwaste CFA as a source for LTA and ZP production fordye removal from aqueous solutions is economically veryattractive

Acknowledgments T H Aldahri is grateful to theTaibah University and Ministry of Higher Education inSaudi Arabia and the Department of Physics Faculty ofScience Taibah University Saudi Arabia for a scholarshipand the Natural Sciences and Engineering Research Coun-cil of Canada (NSERC-CRD grant) for financial supportto carry out this work Adnan A AbdulRazak and IntisarH Khalaf wish to express their gratitude to the Depart-ment of Chemical Engineering University of TechnologyBaghdadIraq for their support

References and Notes1 S Mondal Methods of dye removal from dye house effluent An

overview Environ Eng Sci 25 383 (2008)2 S Yadav D K Tyagi and O P Yadav An overview of efflu-

ent treatment for the removal of pollutant dyes Asian Journal ofResearch in Chemistry 5 1 (2012)

3 T R Das and S Barman Removal of methyl orange and mytheleneblue dyes from aqueous solution using low cost adsorbent zeolitesynthesized from fly ash J Env Sci Eng 54 472 (2012)

4 M Ghaedi S Hajati M Zaree Y Shajaripour A Asfaram andM K Purkait Removal of methyl orange by multiwall carbonnanotube accelerated by ultrasound devise Optimized experimentaldesign Adv Powder Technol 26 1087 (2015)

5 D Robati B Mirza M Rajabi O Moradi I Tyagi S Agarwa andV K Gupta Removal of hazardous dyes-BR 12 and methyl orangeusing graphene oxide as an adsorbent from aqueous phase ChemEng J 284 687 (2016)

6 D Yuan L Zhou and D Fu Adsorption of methyl orange fromaqueous solutions by calcined ZnMgAl hydrotalcite Appl Phys A123 146 (2017)

7 X Querol N Moreno J C Umantildea A Alastuey E HernaacutendezA Loacutepez-Soler and F Plana Synthesis of zeolites from coal fly ashAn overview Int J Coal Geol 50 413 (2002)

8 S S Bukhari J Behin H Kazemian and S Rohani Conversion ofcoal fly ash to zeolite utilizing microwave and ultrasound energiesA review Fuel 140 250 (2015)

9 T Aldahri J Behin H Kazemian and S Rohani Effect ofmicrowave irradiation on crystal growth of zeolitized coal fly ashwith different solidliquid ratios Adv Powder Technol 28 2865(2017)

10 M Ahmaruzzaman A review on the utilization of fly ash ProgEnerg Combust 36 327 (2010)

11 A M Doyle Z T Alismaeel T M Albayati and A S AbbasHigh purity FAU-type zeolite catalysts from shale rock for biodieselproduction Fuel 199 394 (2017)

12 K S Hui and C Y H Chao Effects of step-change of synthesistemperature on synthesis of zeolite 4A from coal fly ash MicroporMesopor Mat 88 145 (2006)

13 X Querol A Alastuey A Loacutepez-Soler F Plana J M AndreacutesR Juan P Ferrer and C R Ruiz A fast method for recyclingfly ash Microwave-assisted zeolite synthesis Environ Sci Technol31 2527 (1997)

14 M M J Treacy and J B Higgins Collection of Simulated XRDPowder Patterns for Zeolites Fourth Revised Edition ElsevierAmsterdam (2001)

15 G R Daham A A AbdulRazak A S Hamadi and A AMohammed Re-refining of used lubricant oil by solvent extrac-tion using central composite design method Korean J Chem Eng34 2435 (2017)

16 K P Singh A K Singh S Gupta and S Sinha Optimizing adsorp-tion of crystal violet dye from water by magnetic nanocompos-ite using response surface modeling approach J Hazard Mater186 1462 (2011)

17 K Fukui M Katoh T Yamamoto and H Yoshida Utilization ofNaCl for phillipsite synthesis from fly ash by hydrothermal treatmentwith microwave heating Adv Powder Technol 20 35 (2009)

18 M Inada H Tsujimoto Y Eguchi N Enomoto and J HojoMicrowave-assisted zeolite synthesis from coal fly ash in hydrother-mal process Fuel 84 1482 (2005)

19 F Rouquerol J Rouquerol K S W Sing P L Llewellyn andG Maurin Adsorption by Powders and Porous Solids PrinciplesMethodology and Applications Elsevier Amsterdam (2014)

20 K S W Sing D H Everett R A W Haul L Moscou R APierotti J Rouquerol and T Siemieniewska Reporting Physisorp-tion Data for GasSolid Systems Handbook of Heterogeneous Catal-ysis Wiley-VCH Verlag GmbH amp Co KGaA Germany (2008)

21 R M Barrer Zeolites and Clay Minerals as Sorbents and MolecularSieves Academic Press London (1978)

22 T Masuda Diffusion mechanisms in zeolite catalysts Catalysis Sur-veys from Asia 7 133 (2003)

23 K Ramesh K Sammi Reddy I Rashmi and A K Biswas Porositydistribution surface area and morphology of synthetic potassium





Al-Dahri et al

Article

Zeolites A SEM and N2 adsorption study Commun Soil Sci Plan45 2878 (2014)

24 T Aldahri J Behin H Kazemian and S Rohani Synthesis of zeo-lite Na-P from coal fly ash by thermo-sonochemical treatment Fuel182 494 (2016)

25 P Pal J K Das N Das and S Bandyopadhyay Synthe-sis of NaP zeolite at room temperature and short crystalliza-tion time by sonochemical method Ultrason Sonochem 20 314(2013)

26 S S Bukhari J Behin H Kazemian and S Rohani Synthesisof zeolite NA-A using single mode microwave irradiation at atmo-spheric pressure The effect of microwave power Can J ChemEng 93 1081 (2015)

27 K Yetilmezsoy S Demirel and R J Vanderbei Response sur-face modeling of Pb(II) removal from aqueous solution by Pista-cia vera L BoxndashBehnken experimental design J Hazard Mater171 551 (2009)

28 D Wu J Zhou and Y Li Effect of the sulfidation process onthe mechanical properties of a CoMoPAl2O3 hydrotreating catalystChem Eng Sci 64 198 (2009)

29 A Mittal A Malviya D Kaur J Mittal and L Kurup Studies onthe adsorption kinetics and isotherms for the removal and recovery ofMethyl Orange from wastewaters using waste materials J HazardMater 148 229 (2007)

30 M Rajasimman and P Karthic Application of response surfacemethodology for the extraction of chromium (VI) by emulsion liquidmembrane Journal of the Taiwan Institute of Chemical Engineers41 105 (2010)

31 K Anupam S Dutta C Bhattacharjee and S Datta Adsorptiveremoval of chromium (VI) from aqueous solution over powderedactivated carbon Optimisation through response surface methodol-ogy Chem Eng J 173 135 (2011)

32 Z Noorimotlagh R Darvishi Cheshmeh Soltani A R KhataeeS Shahriyar and H Nourmoradi Adsorption of a textile dye inaqueous phase using mesoporous activated carbon prepared fromIranian milk vetch J Taiwan Inst Chem E 45 1783 (2014)

33 M I Mohammed A A Abdul Razak and D A HusseinAl-Timimi Modified multiwalled carbon nanotubes for treatment ofsome organic dyes in wastewater Adv Mater Sci Eng 2014 1(2014)

Received 24 March 2018 RevisedAccepted 30 April 2018





Article

ZnMgAl hydrotalcite6 have been utilized for MOremovalCoal fly ash (CFA) is a powder produced as a byprod-

uct of burning coal to generate electricity In the EUand the US around 115 million tons of CFA are pro-duced per year7 Presently fly ash is primarily used asadditive in cement and concrete production Due to itshigh silica and alumina content CFA is also convertedto different types of zeolite by hydrothermal treatmentBukhari et al recently reported on this hydrothermal con-ventional method assisted by microwave and ultrasonicenergy8 Microwave energy is employed in the zeolitesynthesis process because this can yield different zeo-lite morphologies which not only leads to different zeo-lite types but also accelerates nucleation shortens thecrystallization time reduces the undesirable phases due tophase selectivity and facilitates control of the particle sizedistribution9 Zeolites are used in various applications asadsorbents ion exchangers catalysts and membranes Theapplications and utilization of synthetic zeolites from CFAwere reviewed by Ahmaruzzaman10

To the best of the authorsrsquo knowledge zeolite extractionfrom CFA by using the microwave assisted method as ameans of removing dyes from a wastewater stream hasrarely been studiedAs zeolites possess a net negative charge their adsorp-

tion of an anionic dye would be relatively low due to therepulsive forces between the zeolite surface and the dyeThus to make zeolites suitable for adsorption of anionicdyes the negative surface charge must be modified byusing cationic surfactants such as hexadecyltrimethylam-monium bromide (HDTMA)The aim of this work was to synthesize Linde-type

A (LTA) and Na-P type (ZP) from CFA by using themicrowave assisted method These zeolites were furthermodified by HDTMA before being utilized as adsorbentto eliminate the MO dye from the aqueous solution

2 EXPERIMENTAL DETAILS21 ChemicalsCoal fly ash (CFA) was supplied by a coal-fired powerplant (Ontario Power Generation Nanticoke Canada)and was stored in a sealed container until requiredNa2SiO35H2O (sodium met silicate) Na2OAl2O33H2O(sodium aluminate anhydrous) and HDTMA bromidewere purchased from Sigma-Aldrich USA while NaOH(sodium hydroxide) was supplied by Alphachem Mis-sissauga Canada Methyl orange with an MW =32734 gmole was purchased from Alfa Aesar USA Thechemical structure of MO is shown in Figure 1 Deionizedwater (DI) was used for the preparation of the solutionsand washing of synthesized zeolite All chemicals (includ-ing those used for characterization tests) were of analyticalgrade and were used as received

Fig 1 MO chemical structure

22 Synthesis ProcedureThe CFA zeolitization process of CFA commenced withdissolving 2 g of CFA in 01 M of HCl and heating themixture for 3 hours at 90 C as acid treatment enhancesCFA purity by removing impurities11 Subsequently themixture was filtered and the solid phase was washed withdeionized water and was left to try overnight at room tem-perature Next 2 g of acid treated CFA 15 M of NaOHgranules 04 g of NaAlO3 and 007ndash1 g of NaSiO4 wereadded to 20 mL of distilled water to produce the reactionsolution of Linde-type A (LTA) and Na-P type (ZP) zeo-lite respectively91213 The obtained samples were sub-jected to microwave irradiation for a certain period oftime and a fixed power required for crystallization A self-adjusting microwave (single-mode 25 GHz CEM coop-eration Discover USA) was used to conduct experimentsat atmospheric pressure A reflux condenser was attachedto a cylindrical batch PTFE vessel (28 mm IDtimes108 mm)and was placed in the microwave chamber for 30 min at250 W microwave power The samples were subsequentlyfiltered and the solid phase was collected washed withdeionized water and dried at room temperature overnight

23 CharacterizationsMicroscopic images revealing phase purity and morphol-ogy of the zeolitized CFA were obtained by using scan-ning electron microscope (SEM) Hitachi S 2600N SEM(Tokyo Japan) operating at 5 kV accelerating voltagePowder X-ray diffraction (XRD) images of the synthe-sized ZP and LTA zeolites were obtained by RigakundashMiniflex powder diffractometer (Japan) with CuK ( forK = 154059 Aring) over the 5 lt 2 lt 40 range in 2

increments The obtained crystalline phase was verified bycomparing the characteristic peaks of ZP and LTA withthe standard peaks reported in the pertinent literature14

The thermogravimetric analysis (TGA) of the sampleswas performed using a Mettler Toledo TGASDTA 851emodel (Switzerland) coupled with Stare software version61 The samples were heated from 25 C to 600 C in10 Cmin steps under nitrogen purge Specific surfacearea and pore size of the synthesized zeolite (ZP and LTA)were measured using a Micromeritics ASAP 2020 Sur-face Analyzer while a Zetasizer Nano ZS 3000 HSA(Malvern Worcestershire UK) was used for measuring thezeta potential of the product particles

24 Modification of as-Synthesized ZeolitesFor the modification of LTA and ZP 500 mL of the solu-tion of the cationic surfactant HDTMA bromide at a





Al-Dahri et al

Article


Levels and range







MO dye removal


1 600 6000 60000 2709 5352 600 6000 60000 278 5353 900 6000 20000 347 13414 600 10000 20000 1872 20525 900 10000 60000 4935 23416 600 2000 20000 3094 4017 600 6000 60000 276 53668 900 6000 100000 2174 86289 600 6000 60000 271 539610 300 6000 100000 6059 923111 600 6000 60000 272 528612 600 2000 100000 91 99913 600 10000 100000 3199 562414 300 6000 20000 3613 733315 300 2000 60000 8273 824416 900 2000 60000 5864 90217 300 10000 60000 6318 902





Co

lowast100 (1)










Article













(b) (c)(a)






Al-Dahri et al

Article




MO dye removal (ZP)















3 24 206 327 258 158 89 10 4






(X22 and (X3






2 and (X12 had a



PC = SSsum

SStimes100 (4)






Article





1 16016 96089 1 16016 96090 845 63720 00228 lt00001X2

2 14953 3696 1 149532 3696 7888 2451 lt00001 00017X2



(a)

(b)




10788

ndash7196

ndash7049

ndash6072

5049

4389

ndash1961

99

ndash758

0 20 40 60 80 100 120

X3

X1

X2

X1X2

X1^2

X1X3

X2X3

X2^2

X3^2


Qua

drat

ic m

odel

term

s

(a)

ndash25

ndash2346

1886

178

ndash1075

ndash78

582

ndash29

14

0 5 10 15 20 25 30

X1

X2

X3

X2^2

X2X3

X1X2

X1^2

X3^2

X1X3


Qua

drat

ic m

odel

term

s

(b)






Al-Dahri et al

Article


PC


2 X23

MZP 3020 2665 1719 298 0098 559 164 1527 0398MLTA 1699 16262 3819 12096 6322 1263 8364 0322 019


(a)

(b)







Article

(b)

(a)










Al-Dahri et al

Article



(a)

(b)









Article


































Al-Dahri et al

Article

















Al-Dahri et al

Article


Levels and range







MO dye removal


1 600 6000 60000 2709 5352 600 6000 60000 278 5353 900 6000 20000 347 13414 600 10000 20000 1872 20525 900 10000 60000 4935 23416 600 2000 20000 3094 4017 600 6000 60000 276 53668 900 6000 100000 2174 86289 600 6000 60000 271 539610 300 6000 100000 6059 923111 600 6000 60000 272 528612 600 2000 100000 91 99913 600 10000 100000 3199 562414 300 6000 20000 3613 733315 300 2000 60000 8273 824416 900 2000 60000 5864 90217 300 10000 60000 6318 902





Co

lowast100 (1)










Article













(b) (c)(a)






Al-Dahri et al

Article




MO dye removal (ZP)















3 24 206 327 258 158 89 10 4






(X22 and (X3






2 and (X12 had a



PC = SSsum

SStimes100 (4)






Article





1 16016 96089 1 16016 96090 845 63720 00228 lt00001X2

2 14953 3696 1 149532 3696 7888 2451 lt00001 00017X2



(a)

(b)




10788

ndash7196

ndash7049

ndash6072

5049

4389

ndash1961

99

ndash758

0 20 40 60 80 100 120

X3

X1

X2

X1X2

X1^2

X1X3

X2X3

X2^2

X3^2


Qua

drat

ic m

odel

term

s

(a)

ndash25

ndash2346

1886

178

ndash1075

ndash78

582

ndash29

14

0 5 10 15 20 25 30

X1

X2

X3

X2^2

X2X3

X1X2

X1^2

X3^2

X1X3


Qua

drat

ic m

odel

term

s

(b)






Al-Dahri et al

Article


PC


2 X23

MZP 3020 2665 1719 298 0098 559 164 1527 0398MLTA 1699 16262 3819 12096 6322 1263 8364 0322 019


(a)

(b)







Article

(b)

(a)










Al-Dahri et al

Article



(a)

(b)









Article


































Al-Dahri et al

Article

















Article













(b) (c)(a)






Al-Dahri et al

Article




MO dye removal (ZP)















3 24 206 327 258 158 89 10 4






(X22 and (X3






2 and (X12 had a



PC = SSsum

SStimes100 (4)






Article





1 16016 96089 1 16016 96090 845 63720 00228 lt00001X2

2 14953 3696 1 149532 3696 7888 2451 lt00001 00017X2



(a)

(b)




10788

ndash7196

ndash7049

ndash6072

5049

4389

ndash1961

99

ndash758

0 20 40 60 80 100 120

X3

X1

X2

X1X2

X1^2

X1X3

X2X3

X2^2

X3^2


Qua

drat

ic m

odel

term

s

(a)

ndash25

ndash2346

1886

178

ndash1075

ndash78

582

ndash29

14

0 5 10 15 20 25 30

X1

X2

X3

X2^2

X2X3

X1X2

X1^2

X3^2

X1X3


Qua

drat

ic m

odel

term

s

(b)






Al-Dahri et al

Article


PC


2 X23

MZP 3020 2665 1719 298 0098 559 164 1527 0398MLTA 1699 16262 3819 12096 6322 1263 8364 0322 019


(a)

(b)







Article

(b)

(a)










Al-Dahri et al

Article



(a)

(b)









Article


































Al-Dahri et al

Article

















Al-Dahri et al

Article




MO dye removal (ZP)















3 24 206 327 258 158 89 10 4






(X22 and (X3






2 and (X12 had a



PC = SSsum

SStimes100 (4)






Article





1 16016 96089 1 16016 96090 845 63720 00228 lt00001X2

2 14953 3696 1 149532 3696 7888 2451 lt00001 00017X2



(a)

(b)




10788

ndash7196

ndash7049

ndash6072

5049

4389

ndash1961

99

ndash758

0 20 40 60 80 100 120

X3

X1

X2

X1X2

X1^2

X1X3

X2X3

X2^2

X3^2


Qua

drat

ic m

odel

term

s

(a)

ndash25

ndash2346

1886

178

ndash1075

ndash78

582

ndash29

14

0 5 10 15 20 25 30

X1

X2

X3

X2^2

X2X3

X1X2

X1^2

X3^2

X1X3


Qua

drat

ic m

odel

term

s

(b)






Al-Dahri et al

Article


PC


2 X23

MZP 3020 2665 1719 298 0098 559 164 1527 0398MLTA 1699 16262 3819 12096 6322 1263 8364 0322 019


(a)

(b)







Article

(b)

(a)










Al-Dahri et al

Article



(a)

(b)









Article


































Al-Dahri et al

Article

















Article





1 16016 96089 1 16016 96090 845 63720 00228 lt00001X2

2 14953 3696 1 149532 3696 7888 2451 lt00001 00017X2



(a)

(b)




10788

ndash7196

ndash7049

ndash6072

5049

4389

ndash1961

99

ndash758

0 20 40 60 80 100 120

X3

X1

X2

X1X2

X1^2

X1X3

X2X3

X2^2

X3^2


Qua

drat

ic m

odel

term

s

(a)

ndash25

ndash2346

1886

178

ndash1075

ndash78

582

ndash29

14

0 5 10 15 20 25 30

X1

X2

X3

X2^2

X2X3

X1X2

X1^2

X3^2

X1X3


Qua

drat

ic m

odel

term

s

(b)






Al-Dahri et al

Article


PC


2 X23

MZP 3020 2665 1719 298 0098 559 164 1527 0398MLTA 1699 16262 3819 12096 6322 1263 8364 0322 019


(a)

(b)







Article

(b)

(a)










Al-Dahri et al

Article



(a)

(b)









Article


































Al-Dahri et al

Article

















Al-Dahri et al

Article


PC


2 X23

MZP 3020 2665 1719 298 0098 559 164 1527 0398MLTA 1699 16262 3819 12096 6322 1263 8364 0322 019


(a)

(b)







Article

(b)

(a)










Al-Dahri et al

Article



(a)

(b)









Article


































Al-Dahri et al

Article

















Article

(b)

(a)










Al-Dahri et al

Article



(a)

(b)









Article


































Al-Dahri et al

Article

















Al-Dahri et al

Article



(a)

(b)









Article


































Al-Dahri et al

Article

















Article


































Al-Dahri et al

Article

















Al-Dahri et al

Article














Response surface modeling of the removal of methyl orange ...

Documents

Transcript of Response surface modeling of the removal of methyl orange ...