lable at ScienceDirect

Journal of Cleaner Production 113 (2016) 995e1004

Contents lists avai

Journal of Cleaner Production

journal homepage: www.elsevier .com/locate/ jc lepro

High surface area mesoporous activated carbon from tomatoprocessing solid waste by zinc chloride activation: processoptimization, characterization and dyes adsorption

Hasan Say�gılı a, Fuat Güzel b, *

a Department of Chemistry, Faculty of Science & Arts, Batman University, 72100 Batman, Turkeyb Department of Chemistry, Faculty of Education, Dicle University, 21280 Diyarbakır, Turkey

a r t i c l e i n f o

Article history:Received 10 September 2015Received in revised form3 December 2015Accepted 15 December 2015Available online 23 December 2015

Keywords:Tomato processing wasteActivated carbonProcess optimizationCharacterizationDyes adsorption

* Corresponding author. Tel.: þ90 (412) 2488377; fE-mail addresses: hasan.saygili33@gmail.com (H.

(F. Güzel).

http://dx.doi.org/10.1016/j.jclepro.2015.12.0550959-6526/© 2016 Elsevier Ltd. All rights reserved.

a b s t r a c t

The aim of this study is to produce the activated carbon at optimum production conditions from thetomato processing waste by ZnCl2 activation. The influences of the production variables, such asimpregnation ratio, carbonization temperature, and carbonization time on the some pore characteristicsof produced carbons have been investigated, and the best production conditions were determined. Theoptimal activated carbon which had the highest surface area and pore volume was obtained by theseconditions as follows: 6:1 impregnation ratio, 600 �C carbonization temperature and 1 h carbonizationtime. The optimum conditions resulted in an activated carbon with a carbon content of 53.92% and ayield of 38.20%, while the surface area of 1093 m2/g, with the total pore volume of 1.569 cm3/g, mes-oporosity of 91.78% and average pore diameter of 5.92 nm. It was characterized by some physical andchemical techniques. Its adsorptive performance was tested using methylene blue and metanil yellowdyes. The adsorption behavior was well described by the Langmuir isothermmodel, showing a maximumadsorption capacity for methylene blue and metanil yellow of 400 mg/g and 385 mg/g, respectively. Theresults revealed the potential use of optimal activated carbon for removal of cationic and anionic dyes.

© 2016 Elsevier Ltd. All rights reserved.

1. Introduction

Activated carbons (ACs) are widely used in various fields such aspurification and separation in many industrial processes includingmedicinal use, gas storage, pollutant, odor removal, gas separation,catalysis, pharmaceuticals, as electrode materials in electro-chemical devices and in the waste water treatment (Ceyhan et al.,2013). Their high surface areas and large pore volumes are causedby heterogeneous porous structures which make them perfectadsorbents. The demand for AC in the world due to these features isincreasing day by day. The world demand for ACs was about4.28 millionmetric tons in 2012, and it is expected to increasemorethan 10% per year over the next 5 years. This increased demand ismainly due to the more stringent pollution controls in the USA andChina. For instance in China the existing plan, named the TwelfthFive-Year Plan (2011e2015), that seeks to improve thewater and air

ax: þ90 (412) 2488257.Say�gılı), fguzel@dicle.edu.tr

quality through the use of more environmentally friendly processeswill significantly increase the AC demand (Valente Nabais et al.,2013).

The ACs can be produced from a variety of natural and syntheticsubstances, lignocellulosic materials being one of the most usedprecursors. The composition of the lignocellulosic materialregarding the cellulose, hemicellulose and lignin content de-termines, to some extent, the porosity development of the pro-duced ACs (Valente Nabais et al., 2013). The production of ACs fromagricultural by-products has received much attention from thescientific community as they are renewable, low-cost and envi-ronmentally friendly materials (Okman et al., 2014). In recent years,there has been increasing interest in the research on the productionof ACs by using renewable and cheaper precursor. A large numberof agricultural by-products, such as coffee grounds (Reffas et al.,2010), cocoa shell (Ahmad et al., 2012), waste tea (Auta andHameed, 2011), coffee husks (Oliveira et al., 2009), pomegranateshell (Ghaedi et al., 2012), mahogany sawdust (Malik, 2003), ricehusk (Malik, 2003; Santra et al., 2008), mehagani sawdust (Santraet al., 2008), coconut shell (Santra et al., 2008), bael shell, grapeindustrial processing waste (Say�gılı et al., 2015), etc. have been

Table 1The experimental runs and conditions for ACs production, and pore characteristics [IR: Impregnation ratio; CT: Carbonization temperature; Ct: Carbonization time].

Run Production conditions Pore characteristics

IR (w/w) CT (�C) Ct (h) SBET (m2/g) VT (cm3/g) Vmic (cm3/g) Vmic (%) Vmes (cm3/g) Vmes (%) DP (nm)

Effect of Impregnation ratio1 1:1 500 1.0 617 0.437 0.313 71 0.124 29 2.642 2:1 '' '' 722 0.457 0.304 66 0.153 34 4.493 4:1 '' '' 760 0.710 0.584 41 0.416 59 4.834 6:1 '' '' 787 1.000 0.277 28 0.723 72 5.785 8:1 '' '' 495 0.452 0.196 43 0.256 57 3.65Effect of carbonization temperature6 6:1 400 1.0 648 0.756 0.086 11 0.670 89 4.567 '' 600 '' 1093 1.569 0.129 8 1.440 92 5.928 '' 800 '' 492 0.106 0.058 55 0.048 45 2.41Effect of carbonization time9 6:1 600 0.5 522 0.662 0.191 29 0.471 71 5.0210 '' '' 2.0 1034 1.451 0.440 30 1.011 70 5.6111 '' '' 4.0 861 0.873 0.494 57 0.379 43 4.06

H. Say�gılı, F. Güzel / Journal of Cleaner Production 113 (2016) 995e1004996

successfully converted into low-cost ACs. As far as we know, nostudy has been done on production of AC from tomato processingsolid waste (TW) which is cheap, accessible and available inabundant quantity. According to the data of the United NationsFood and Agriculture Organization (FAO) in 2012, total freshvegetable grown in the world is 1.1 billion t. Tomato is the mostwidely grown fresh vegetable products in the world with produc-tion of about 162 million t. In the production of tomatoes in theworld, China (50 million tons), India (17.5 million t), US(13.2 million t), Turkey (11.3 million t) and Egypt (8.6 million t) arethe leading countries. Tomato income is almost at the beginning ofthe most consumed fresh vegetable. It is often used in the cookingand paste production. After the tomato paste production, theyreveal waste up 10e18% of the weight depending on their type(Sarısaçlı, 2007). In Turkey, a great quantity of TW is generated assolid waste in the tomato paste production.

This study has focused on the production of AC from TW bychemical activation with ZnCl2. The effects of impregnation ratio(ZnCl2/TW, w/w), carbonization temperature and time on the somepore characteristics of the produced ACs were investigated. More-over, the AC produced at optimized conditions was characterizedwith various physicochemical techniques. Also, the AC was testedto remove cationic and anionic dyes from aqueous solutions. Forthis purpose, methylene blue (3, 7-bis(dimethylamino)phenaza-thionium chloride trihydrate) (MB) and metanil yellow (3-(4-Anilinophenylazo)benzenesulfonic acid sodium salt) (MY) wasused asmodel cationic and anionic dyes, respectively. They are usedin many industries such as paper, textile, plastics and leather aswell as cosmetic and pharmaceutical industries use dyes to colortheir products. MB and MY dyes like most of the dyes are carci-nogenic, mutagenic, and can cause a severe health hazard to humanbeings. In addition, they impede light penetration into the water,retard photosynthetic activity, and inhibit the growth of biota(Hejazifar and Azizian, 2012).

2. Materials and methods

2.1. Materials

The precursor, TW was collected from a local tomato pastefactory in Adana, Turkey. It was washed with water to take offimpurities and subsequently dried naturally. Later, the solid wastewas crushed and ground to pass through 80-mesh sieves prior to itschemical activation. ZnCl2 of purity 99.9% was used as chemicalactivator in the AC production.

The basic dye, MB (Type: cationic, chemical formula:C16H18ClN3S, Mw: 319.85 g/mol, color index: basic blue 9, lmax:665 nm) and acidic dye, MY (Type: anionic, chemical formula:C18H14N3NaO3S, color index: acid yellow 36, lmax: 433 nm, Mw:375.38 g/mol) were used as sorbates. The used all chemicals werepurchased from SigmaeAldrich Company in Ankara, Turkey.

2.2. Optimization studies of production conditions

The optimal conditions for AC preparation from TW weredetermined by examining effects of the impregnation ratio (1:1, 2:1,4:1, 6:1 and 8:1; ZnCl2/TW, w/w), carbonization temperature (400,500, 600 and 800 �C) and carbonization time (0.5, 1.0, 2.0 and 4 h).The experimental run of the production and conditions for ACsfrom TW are given in Table 1. These processes were performed byheating in a horizontal stainless-steel tubular reactor(7.0 cm diameter � 100 cm length). At every turn, twenty-fivegrams of the dry sample impregnated with ZnCl2 was placed intothe reactor and N2 was passed through at a flow rate of 100 mL/min; the systemwas heated at a rate of 10 �C/min. the produced ACswere cooled down to room temperature under nitrogen flow, and0.2 N HCl was added on to ACs. They were washed sequentiallyseveral times with hot distillated water to remove residual chem-ical until it did not give Cl� reactionwith AgNO3, and dried at 105 �Cfor 12 h and then sieved between 177 and 400 mm. They werereferred to as AC followed by the impregnation ratio, carbonizationtemperature and time; for instance, AC11405 represented impreg-nation ratio 1:1 for the first two number of 11, activation temper-ature of 400 �C for the third number of 4, and activation for 0.5 h forlast 05. The yield of optimal ACwas calculated as the ratio of the dryweight of produced AC to the weight of the air-dried of theprecursor.

2.3. Characterization of tomato waste and its activated carbon

TW was characterized by thermogravimetry/differential ther-mogravimetry (TG/DTG) analysis, proximate, ultimate andbiochemical component analysis, scanning electron microscopic(SEM), X-ray diffraction (XRD), X-ray photoelectron spectroscopic(XPS) and Fourier transform infrared spectroscopic (FT-IR) whilethe optimal AC produced was characterized by proximate, ultimate,pore structure, SEM, XRD, XPS, FT-IR analysis, Boehm's titration,pHpzc and CEC.

The thermal behavior of TW was measured with a thermogra-vimetric analyzer (Shimadzu, TG/DTA-50). About 10 mg of samplematerial was heated from 25 �C to 1000 �C at a ramping rate of

Fig. 1. TG/DTA curves for TW.

H. Say�gılı, F. Güzel / Journal of Cleaner Production 113 (2016) 995e1004 997

10 �C/min under nitrogen gas atmosphere with the low rate of200 mL/min, and constantly weighed. Percent weight loss versustemperature plots were taken for the thermogravimetric (TG)analysis and the derivative weight loss against temperature wastaken for the derivative thermogravimetric (DTG) analysis.

The advantage of the TW for AC production was first checkedusing various methods such as ultimate, proximate and biochem-ical component analysis. The elemental analysis (C, H, N, and S) wasdetermined with an elemental analyzer (Leco CHNS 932). Theproximate analysis was conducted according to ASTM D 3172-3175test standards and the results were given as moisture, volatilematter, ash, and fixed carbon contents. Biochemical componentanalysis including extractives, lignin, hemicelluloses, and celluloseof TW was done according to Technical Association of the Pulp andPaper Industry, (TAPPI) standards; T264 om-88, T222 om-88 andT203 om-83, respectively.

Crystallographic characterization of TWand TACwere examinedby XRD measurements performed on a Bruker D8 Discovery EVAdiffractometer using monochromatic copper radiation (Cu Ka,l ¼ 0.1541 nm) at 40 kV and 40 mA over the 2q range 0-80�.

The surface physical morphologies of TW and TAC were iden-tified by using SEM technique (Jeol/jsm- 6335F, USA).

The SBET, VT, Vmic, and Dp of the produced ACs were determinedby using a surface area and pore size distribution analyzer(Micromeritics ASAP 2020). The Vmes was determined by sub-tracting the Vmic from the VT while the Vmic(%) and Vmes(%) werebased on the VT.

In order to determine quantitatively and qualitatively surfacefunctional groups of the optimal AC produced, FT-IR analysis andBoehm's titration method (Boehm, 2002) were applied, respec-tively. Surface functional groups were detected using the pressedpotassium bromide (KBr) pellets containing 5% of carbon sample byFT-IR spectrometer (Perkin Elmer spectrum 100) in the scanningrange of 4000e650 cm�1. In addition, the surface properties of theTW and TAC were measured by XPS. These measurements wereperformed on THERMO K-ALPHA spectrometer with a mono-chromic Al Ka source at 1486.7 eV, with a voltage of 15 kV and anemission current of 10 mA. The amounts of surface acidic and basicfunctional groups were determined by the Boehm's titrationmethod. This method is one of the most widely used methods toquantify acidic groups with different strengths on ACs. The variousfree acidic groups were determined using the assumption thatNaOH neutralizes carboxyl, lactone and phenolic groups; Na2CO3neutralizes carboxyl, lactone; NaHCO3 neutralizes only carboxylgroups, respectively (Laszlo and Szucs, 2001). The total acidity andbasicity were determined from the amount of NaOH and HCl,respectively. pHpzc was determined using the pH drift method. ThepHpzc value is reflecting the combined influence of all the surfacefunctional groups of samples. Net charge of the adsorbent underpHpzc is positive, while it is negative above pHpzc (Smiciklas et al.,2000). pH was measured in a distilled water suspension(12.5 mL) of 0.5 g of sample after heating at 90 �C and then coolingto room temperature (Reffas et al., 2010). CEC which is a scale toreversibly adsorb positively charged species of optimal AC wasdetermined by the method described by Puziy et al. (2002).

2.4. Methylene blue and metanil yellow dye adsorption

A constant mass of produced optimal AC (30 mg) was weighedinto 100 mL Erlenmeyer flasks and contacted with 50 mL of eachdye solutions of different initial concentration from 200 to900 mg/L at natural pH (6.33 for MB; 6.23 for MY) of MB/MY-ACadsorption systems. The flasks were shaken mechanically at120 rpm in a water bath shaker (Daihan-WSB-30) at 30 �C for 24 h,which is enough for guaranteeing adsorption equilibrium for both.

dye onto surface. After each adsorption process, the sampleswere filtrated for solideliquid separation and the concentrations ofMB andMY in the supernatant solutionswere analyzed by a UVevisspectrophotometer (Perkin Elmer-Lambda 25) at 665 and 433 nmof lmax, respectively. Each dye adsorption experiment was carriedout in duplicate, and the average values are given.

A number of isotherms have been developed to describe equi-librium relationships. For the isotherm modeling, two famousisothermmodels, namely the Freundlich (Freundlich, 1906) and theLangmuir (Langmuir, 1918) isotherm models were used for theequilibrium data obtained. The used linearized isotherm equationswere:

Freundlich isotherm ln qe ¼ ln KF þ1nF

ln Ce (1)

Langmuir isothermCeqe

¼ 1qmb

þ 1qm

Ce (2)

The higher value of R2 the more accurate is the model thatpredicts the experimental data.

Further, the SMB and SMY related to the MB and MY for porecharacterization from aqueous phase were calculated from thefollowing equation:

S ¼ 10�20qmaDNA

M(3)

In this study, aD values taken for MB and MY dyes were 1.20 and1.02 nm2, respectively (Güzel, 1996).

3. Results and discussion

In this part, the properties of TW, the effect of conditions in theproduction of optimal AC and also the pore, surface and dyesadsorption property of the AC produced in optimal conditions wereinvestigated.

3.1. Thermal analysis of the tomato waste

Thermal analysis has been widely used to get knowledge of thethermal behaviors of agricultural by-products. Thermograms ob-tained from the TG/DTG analysis of the TW as a function of tem-perature is shown in Fig. 1. As seen in this figure, the decompositionof TW takes place in four steps. The first step, which occurs attemperatures ranging from 28 to 150 �C, involves the loss ofmoisture present in the sample with approximately the weight lossof 4.45%. The second weight loss step with approximately the

H. Say�gılı, F. Güzel / Journal of Cleaner Production 113 (2016) 995e1004998

weight loss of 5.25% was occurred at 150e280 �C. This step isrelated to the release of volatiles resulting from the decompositionof hemicellulose. In the third step the occurring at 280e480 �C ischaracterized by the decomposition of cellulose and lignin. Themaximum rate of weight loss occurred at the third stage approxi-mately with the weight loss of 52%. Nevertheless, in the fourth step(above 480 �C), no significant weight loss (0.95%) is observed.Therefore, this temperature can be said to be the lowest carbon-ization temperature for AC production from TW. In the DTG spec-trum in Fig. 1, the endothermic peak around 98 �C, belongs to theremoval of water. The weak endothermic peak around 470 �C be-longs to hydroxyl deformation (Yang et al., 2007).

3.2. Selection of optimum process parameters

In order to produce optimal AC from the TW, the effects ofexperimental process variables represented by impregnation ratio(ZnCl2/TW), carbonization temperature, and carbonization time onthe pore characteristics such as SBET, VT, Vmes (%) and Dp determinedfrom the nitrogen adsorptionedesorption isotherms and pore sizedistribution graphs (Fig. 2aec) of the ACs produced were demon-strated in Table 1. The obtained results are presented below underthe sub-heads.

3.2.1. Effect of the impregnation ratioFirstly, to see the effect of the chemical activation, TW was

carbonized at 500 �C without ZnCl2 activation. It had the BET

Fig. 2. Nitrogen sorption (filled symbols)-desorption (empty symbols) isotherms and pore s(a), activation temperatures (b) and activation times (c).

surface area of 12.79 m2/g and VT of 0.012 cm3/g. The effects ofimpregnation ratio (ZnCl2/TW, w/w) on the SBET, VT, Vmes% and Dp ofACs produced for a carbonization temperature of 500 �C andcarbonization time of 1 h are shown in Table 1. As seen in this table,while the impregnation ratio was increased from 1 to 6, the SBET, VT,Vmes (%) and Dp values of ACs produced increased from 617 to787m2/g, from0.437 to 0.973 cm3/g, from 29% to 72% and from 2.64to 5.78 nm, respectively, and above 6:1 these values decreased.From these results, it is understood that ZnCl2 impregnation ratiosignificantly affect to porosity. With low impregnation ratio, due tothe effect of ZnCl2, the formation of tar is inhibited and the releaseof volatiles is promoted, preparing more micropores. But at higherimpregnation ratio, the more swelling into impregnated precursormaterial and stronger release of volatiles in the carbonizationprocess will lead to the widening of pores; micropores formed aresubsequently converted to mesopores (Yang and Qiu, 2011).Consequently, the ratio of 6 was chosen as optimum impregnationratio.

3.2.2. Effect of carbonization temperatureThe production of ACs was carried out at different temperatures

to be carbonized for the time of 1 h by keeping the impregnationratio at 6:1. As seen from the Table 1, while temperature increasedfrom 400 to 600 �C, the SBET, VT, Vmes% and Dp values increased from648 to 1093 m2/g, from 0.756 to 1.569 cm3/g, from 89% to 92% andfrom 4.56 to 5.91 nm, respectively, and above 6:1 these valuesdecreased. This increase is due to the formation of new pores by

ize distribution (inset) for ACs produced from TW at the different impregnation ratios

Table 2Physical and chemical characteristics of TAC.

Parameter Value Parameter Value

Proximate analysis (dry basis, wt%)

Ultimate analysis (dry basis,wt %)

Moisture 9.22 Carbon 79.38Ash 25.56 Hydrogen 1.76Volatile matter 11.30 Nitrogen 1.70Fixed carbon 53.92 Oxygena 17.06Yield 38.20 Sulfur 0.10Burn-off 62.80Pore characteristics Surface functional groups

(meq/g)SBET (m2/g) 1093 Carboxylic 0.52VT (cm3/g) 1.569 Phenolic 0.13Vmic (cm3/g) 0.129 Lactonic 0.52Vmes (cm3/g) 1.440 Total acidity 1.17Vmic (%) 8.22 Total basicity 1.04Vmes (%) 91.78 CEC 0.95Dp (nm) 5.92 pHpzc 6.17

a By difference.

H. Say�gılı, F. Güzel / Journal of Cleaner Production 113 (2016) 995e1004 999

continuation of the evaporating substances resulting from thedecomposition of primary compounds of the TWwith increasing oftemperature. However, decrease in pore characteristics above600 �C might be due to the sintering effect at high temperature,followed by shrinkage of the char, and realignment of the carbonstructure which resulted in reduced pore areas as well as volume(Mohanty et al., 2006). Consequently, 600 �C was chosen as theoptimum carbonization temperature for subsequent experiments.

3.2.3. Effect of carbonization timeThe production of ACs was carried out at different carbonization

time (0.5e4 h) to be activated for 600 �C by keeping the impreg-nation ratio at 6:1. As seen from the Table 1, while the timeincreased from 0.5 h to 1 h, the SBET, VT, Vmes(%) and Dp valuesincreased from 522 to 1093 m2/g, from 0.662 to 1.569 cm3/g, from71% to 92% and from 5.02 to 5.92 nm, respectively. However, whilethe carbonization time increased from 1 h to 4 h, they weredecreased. This decrease was possibly due to some of the poresbeing sealed off as a result of sintering for a prolonged time(Mohanty et al., 2006). Therefore, the time of 1 hminwas chosen asthe best carbonization time.

3.3. Process optimization

From the pore characteristics in Table 1, the optimal AC whichhad the highest SBET, (1063m2/g), VT (1.569 cm3/g), Vmes(%) (91.78%)and Dp (5.916 nm) was obtained by these conditions as follows:6:1 ZnCl2 to TW impregnation ratio (w/w), 600 �C carbonizationtemperature and 1 h carbonization time (Run 7 in Table 1). OptimalAC is labeled as TAC. It was used in the characterization analysesand removal performance of MB and MY dyes from the aqueousphase.

3.4. Characterization

3.4.1. Proximate, elemental and biochemical component analysisTheproximate, ultimate andcomponent analysis results revealed

that TW contains 56.96% holocellulose (33.17% cellulose þ 23.79%hemicellulose), 35.70% lignin, 7.34% extractives, 2.95% moisture,82.67% volatilematter,12.80%fixed carbon,1.58%ash, 58.95% carbon,8.21%hydrogen,0.12%sulfurand29.77%oxygenasexpected formostlignocellulosic material. The high carbon, low ash and low sulfurcontent of the precursor are positive factors and therefore, thesewastes can be potential precursormaterials for the production of AC(Stavropoulos and Zabaniotou, 2005; Açıkyıldız et al., 2014).Although TW has very high volatile content, low ash content and areasonable amount of carbon makes it a suitable precursor for thecarbon sorbent production having large porous.

The proximate and ultimate analysis results related to the TACare given in Table 2. It can be seen that the contents of carbon andash increased, in the TAC according to the TW, whereas the con-tents of hydrogen, nitrogen, oxygen and sulfur decreased. This isdue to the release of volatiles during carbonization, which canresult in the elimination of the non-carbon species (Chen et al.,2013). The TAC also possessed a low concentration of ash(25.56%) indicating a high-quality carbon sorbent. Low ash contenttypically leads to superior removal of organic substances fromaqueous phase due to the hydrophobicity of the material (Zuimet al., 2011).

3.4.2. Pore analysisThe nitrogen adsorptionedesorption isotherms (inset) as well as

the pore size distributions for the TAC are shown in Fig. 2c. Theshape of its N2 adsorptionedesorption isotherm was a mixture oftypes I and IV in BDDT (Brunauer, Deming, Deming and Teller)

classification, as defined by the IUPAC (Sing et al., 1985), with awider hysteresis loop (H4) at high relative pressures, suggestingthat it had a mixed microporous and mesoporous structure (Wanget al., 2012). The main pore characteristics of TAC obtained from thenitrogen adsorptionedesorption isotherms were summarized inTable 2, indicating that it has a large SBET (1093 m2/g) and VT

(1.569 cm3/g). Pore size distribution is a very important property ofporous sorbents because the difference in the pore size affects theadsorption capacity for molecules of different sizes and shapes, andalso this is one of the criteria by which carbon sorbents are selectedfor a particular application. According to the classification adoptedby IUPAC, sorbent pores are classified into three groups:micropores(<2 nm), mesopores (2e50 nm), and macropores (>50 nm) (Xiaoet al., 2012). Fig. 2c presents the pore size distribution for theTAC. As seen this figure, the dominant pore of TAC is in the range of4e95 nm, and its average pore diameter is 5.92 nm. Therefore, it istypically a mesoporous carbon sorbent. This can be also concludedfrom the results in Table 2, which show a Vmes of 1.440 cm3/gcompared to a Vmic of 0.129 cm3/g. These results indicate that asignificant amount of Vmes(%) (91.78%) and a certain amount ofVmic(%) (8.22%) were formed in the TAC. These results indicate thatTW is efficient precursor for production of mesoporous AC havinghigh surface area and large pore volume. Table 3 shows a com-parison of the SBET of the TAC with other reported values for agrowaste-based ACs in literatures.

3.4.3. SEM analysisFig. 3(a) and (b) show the SEM micrographs of the TW and TAC,

respectively. Moreover, TAC has some cavities, not full of cavities.The cavities are caused by chemical activation with ZnCl2 duringcarbonization. Namely, ZnCl2 catalyzes the polymerization re-actions (Scholl condensation reactions) between tar-formingcompounds and aromatic hydrocarbons in the structure of pre-cursor (Ahmadpour and Do, 1997). The polycyclic compounds withlarge molecules consisting of these reactions move away from thestructure with the effect of temperature during the carbonization,and leave a porous structure. According to this micrograph in-dicates that it would be a good sorbent in adsorption applicationfrom solution. This observation can be supported by pore charac-teristics of TAC as illustrated in Table 2.

3.4.4. XRD analysisFig. 4 shows the XRD patterns of the TW and TAC. In the TW,

a broad peak appearing at 2q ¼ 21.8� was attributed to the

Table 3BET surface area comparison of TAC with other agricultural waste-based ACs reported in literatures [IR: Impregnation ratio; CT: Carbonization temperature; Ct: Carbonizationtime].

Precursor Atmosphere Activating agents IR (w/w) CT (�C) Ct (h) SBET (m2/g) References

Grape stalk CO2 ZnCl2 2.0 700 2.0 1411 €Ozdemir et al., 2014Pistachio nutshell N2 ZnCl2 1.0 400 1.0 1635 Lua and Yang, 2005Coffee husk N2 ZnCl2 1.0 550 3.0 1522 Oliveira et al., 2009Waste potato residue e ZnCl2 1.5 600 1.0 1357 Zhang et al., 2015Fox nutshell N2 ZnCl2 2.0 800 1.0 2869 Kumar and Jena, 2015Grape waste N2 ZnCl2 6.0 600 1.0 1455 Say�gılı et al., 2015Coffee husks N2 FeCl3 1.0 280 1.0 965 Oliveira et al., 2009Spartina alterniflora N2 KOH 3.0 450 1.5 2825 Liu et al., 2013Plum kernels N2 NaOH 4.0 780 1.0 1887 Tseng, 2007Corncob N2 H3PO4 1.0 400 1.0 2081 Sych et al., 2012Rice straw Air (NH4)2HPO4 5.0 200 2.0 1154 Gao et al., 2011Sunflower oil cake N2 H2SO4 0.85 600 e 240 Karag€oz et al., 2008Waste tea N2 KCH3COO 1.4 800 2.0 854 Auta and Hameed, 2011Tomato waste N2 ZnCl2 6.0 600 1.0 1093 This study

Fig. 3. SEM micrographs of TW and TAC.

Fig. 4. XRD spectra of TW and TAC.

H. Say�gılı, F. Güzel / Journal of Cleaner Production 113 (2016) 995e10041000

characteristic crystal structure of cellulose (Regmi et al., 2012). Theintensity of the peak largely decreased after carbonization due tothe decomposition of the cellulose. Two broad peaks appeared at

approximately 2q ¼ 25� and 43� after activation with ZnCl2 in TAC.They corresponded to the formation of hexagonal graphite struc-tures (0 0 2) and (10 0) (overlapped 10 0 and 10 1). In addition, the

H. Say�gılı, F. Güzel / Journal of Cleaner Production 113 (2016) 995e1004 1001

XRD pattern of TAC contains sharp peak at 2q ¼ 35.3� which wasdue to the presence of zinc oxide and zinc carbides on the washedTAC surface. The sharpness of the peak indicates that the zincpresented is comparatively large, though still in the micro range(Danish et al., 2013).

3.4.5. FT-IR analysisThe FT-IR spectra of TWand TAC are shown in Fig. 5. The spectra

exhibit a number of absorption peaks, indicating the complex na-ture of the materials. In the spectra of TW in Fig. 5, the presence ofthe broad and intense absorption peak at 3323.17 cm�1 indicatesthe OeH stretching vibrations of cellulose, pectin, absorbed water,hemicellulose, and lignin (Gündo�gdu et al., 2013). The peakobserved at 3003.96 cm�1 can be attributed to the aliphatic satu-rated CeH stretching vibrations of in lignin polysaccharidesincluding cellulose and hemicellulose. The two notable bands arelocated at 2923 cm�1 and 2853 cm�1 due to the symmetric andasymmetric stretching vibration of methyl groups (Gündo�gdu et al.,2013; Nguyen et al., 2013). The presence of the peak at1743.27 cm�1 indicates the carbonyl (C]O) stretching vibration ofthe carboxyl groups of pectin, hemicellulose and lignin (Njokuet al., 2013; Hameed and Daud, 2008). The peak at 1635.10 cm�1

corresponds to absorbed water HeOeH bend. The vibrations at1453.96 cm�1 and 1375.75 cm�1 could be due to aliphatic andaromatic (CeH) groups in the plane deformation vibrations ofmethyl, methylene and methoxy groups (Hameed and Daud, 2008;Huang et al., 2015; Angın, 2013). The bands in the range of1250e1000 cm�1 can be assigned to the CeO stretching vibration ofcarboxylic acids and alcohols. The band at 721.64 cm�1 is related toaromatic, out of plane CeH bending with different degrees ofsubstitution (Angın, 2013; Mastalerz and Bustin, 1995). Aftercarbonization, the FT-IR spectrum of TAC was markedly differentfrom TW. The broad band located in the region of 3100e3400 cm�1

related to OeH stretching vibrations existed but shifted to the highwavenumber. The band located at about 1581.71 cm�1, which couldbe attributed to (C]C) vibration in aromatic rings (Lua and Yang,2005; Li et al., 2015). The band at 1228.21 cm�1 is due to the CeOstretching vibration of phenol group (Hameed and Daud, 2008;Angın, 2013). The many absorption bands disappeared when the

Fig. 5. FT-IR spectra

FT-IR spectrum for the TAC are compared to that for TW. This islikely to be due to the vaporization of organic matter at hightemperatures.

3.4.6. XPS analysisIn order to further investigate the surface chemistry compo-

sition of the materials, XPS analysis was performed for the priorto and after carbonization. Fig. 6 shows the XPS spectra of TW andTAC. For the TW (Fig. 6a), the high resolution C1s spectrum can befitted to three components corresponding to the C]C (284.37 eV),CeOeC (285.9 eV) and eCOOR (288 eV) with the relative per-centages of 41.75%, 30.25% and 5.98%. After carbonization, the C1sspectrum of TAC (Fig. 6c) was resolved into two individualcomponent peaks: the groups of C]C (284.68 eV) and the groupsof CeOeC (285.91 eV). No carboxylic acid and/or ester groups(eCOOR) are detected. The absence of eCOOR groups could beattributed to evaporate from surface in the carbonization stage.For the carbon species, a substantial increase in C]C (72.83%) anda decrease in CeOeC (17.21%) groups were also found. This meansthat the pyrolysis tends to increase the carbon fraction mainly dueto the volatilization of the oxygenated compounds (Ncibi et al.,2009). On the other hand, Fig. 6b shows that the O1s region inthe XPS spectrum of TW exhibits two deconvoluted peaks at530.75 and 532.24 eV, representing carboxylic acid and/or estergroups (O]CeO) and oxygen singly bonded to carbon in phenolsand ethers groups (CeO), respectively. Fig. 6d shows the typicalhigh resolution XPS spectra of the O1s region of TAC. As can beseen from this figure, O1s XPS spectrum of TAC was fitted to twocomponents: (I) metal oxides (531.11 eV) and (II) oxygen singlybonded to carbon in phenols and ethers (532.8 eV) (Biniak et al.,2013; Burg et al., 2002). The XPS analysis results indicate thatsignificant differences existed on the percentage amounts ofcarbon and oxygen fractions after carbonization procedure, whichis in agreement with the above FT-IR and Boehm's titrationresults.

3.4.7. Quantitative surface analysisAccording to the Boehm titration results in the Table 2, it

exhibited a weak acidic property, with the surface acidity of

of TW and TAC.

Fig. 6. XPS C1s and O1s spectra of TW (aeb) and TAC (ced).

H. Say�gılı, F. Güzel / Journal of Cleaner Production 113 (2016) 995e10041002

1.16 meq/g with the maximum composition of phenolic group(0.13 meq/g) with traces of lactonic (0.51 meq/g) and carboxylic(0.52 meq/g) groups, and 1.04 meq/g as surface basicity. The con-centration of lactonic and carboxyl groups was significantly highercompared with the phenolic group, indicating that carboxyl groupwas more prevalent on the surface. Furthermore, the surface of theTAC has substantially equal concentration of the basic and acidicgroup referring to an adsorptive, which is effective in the removalof anionic and cationic dyestuffs. The pH and pHpzc values for theTAC were found to be 5.80 and 6.17, respectively. The pH value fallsin the slightly acidic region and value of pHpzc < 7 shows dominantof acidic groups over basic groups.

100 250 400 550 700 850180

216

252

288

324

360

Methylene Blue

Metanil Yellow

Ce (mg/L)

q e(mg/g)

Fig. 7. Sorption isotherms of MB and MY dyes onto TAC at 30 �C.

3.4.8. Methylene blue and metanil yellow adsorption analysisFig. 7 shows equilibrium isotherms of MB and MY at 30 �C onto

TAC. The experimental equilibrium data in this figure are fittedwithFreundlich and Langmuir linear isotherm equations (Eqs. (1)e(2)).The isotherm constants related to these models were obtainedaccording to the intercept and slope from the plots between lnqe vs.lnCe and Ce/qe vs. Ce (not shown). Isotherm parameters and R2

values obtained from the two isotherm models are summarized inTable 3. As seen from this table, R2 values obtained for bothisotherm models revealed that the equilibrium data for both dyesfitted well to the Langmuir isotherm under the studied conditions.The maximum sorption capacities (qm) calculated by the Langmuirequation of the MB and MY onto TAC are 400 and 385 mg/g,respectively (Table 4). A comparison of the pore characteristics andmaximum sorption capacities of the TACwith other reported valuesfor some commercial and vegetable-based ACs are listed in Table 5.The results in this table showed that the TAC can be effectively usedfor the removal of cationic and anionic dyes from aqueous solution,in which its sorption capacity is higher than many commercial andothers agricultural waste-based carbonaceous sorbents. Further-more, a value for 1/nF for both dyes was found to range betweenzero and one, indicating that the sorption condition was favorable.Besides, the 1/nF values obtained for MB sorption were lower incomparison to MY sorption, leading to conclusion that TAC is moreefficient sorbent for the removal of cationic dye than anionic dye.

Table 4Freundlich and Langmuir isotherm parameters for MB and MY dyes sorption ontoTAC at 30 �C.

Dyes Freundlich Langmuir

KF (mg/g) (L/mg)1/n 1/n R2 qm (mg/g) b (L/g) R2

MB 65 0.259 0.925 400 0.010 0.997MY 47 0.301 0.894 385 0.008 0.994

Table 5A comparison of adsorptive characteristics of TAC and some commercial and agricultural waste-based ACs reported in literatures.

ACs qm,MB (mg/g) References ACs qm,MY (mg/g) References

Filtrasorb-300 240 Stavropoulos and Zabaniotou, 2005 Pomegranate 19 Ghaedi et al., 2012Filtrasorb-400 455 Qada et al., 2013 Mahogany sawdust 184 Malik, 2003Coffee grounds 182 Reffas et al., 2010 Rice husk 87 Malik, 2003Cocoa shell 213 Ahmad et al., 2012 Mehagani sawdust 118 Santra et al., 2008Waste tea 446 Auta and Hameed, 2011 Coconut shell 90 Santra et al., 2008Coffee husk 263 Oliveira et al., 2009 Rice husk 55 Santra et al., 2008Grape waste 417 Say�gılı et al., 2015 Grape waste 386 Say�gılı et al., 2015Tomato waste 400 This study Tomato waste 385 This study

H. Say�gılı, F. Güzel / Journal of Cleaner Production 113 (2016) 995e1004 1003

Otherwise, by evaluating in the Eq. (3) of maximum adsorptioncapacity values obtained from the Langmuir linear equation, thesurface area values (SMB and SMY) according to the MB and MY dyeswere calculated as 903 and 630 m2/g, respectively. Furthermore, bytaking N2 as basis, the percent portions (SMB/SN2 and SMY/SN2) of theMB and MY for the reaching to the surface relative to N2 weredetermined as 83 and 58%, respectively. These results also confirmthat TAC has a mesoporous pore structure and a perfect adsorptivefor anionic and cationic dyes removal from aqueous phase.

4. Conclusion

The influence of operating parameters on the pore propertiesof the produced carbons followed a sequence of carbonizationtemperature > carbonization time > impregnation ratio. Theoptimal AC produced by chemical activation with ZnCl2 attainedmaximum value of SBET as 1093 m2/g, a VT as 1.569 cm3/g, a highVmes (%) as 91.78%, a DP as 5.92 nm and a pHpzc as 6.17 at carbon-ization temperature of 600 �C for 1 h with a impregnation ratio of 6.These results showed that the TW seemed to be an alternativeprecursor for the commercial AC production. The equilibrium dataof both dye adsorptions onto TAC followed the Langmuir isothermmodel, showing the maximum monolayer adsorption capacities of400 mg/g for MB and 385 mg/g for MY. Consequently, TAC can beused as a cleaner sorbent for solving many current-day environ-mental pollution problems and other applications.

Acknowledgments

The authors acknowledge the Scientific Research Fund of DicleUniversity for financial support (Project No: 12-ZEF-95).

List of abbreviations

aD Cross section area of one molecule of dye (nm2)b Langmuir equilibrium constant related to the energy of

sorption (L/mg)C0 Initial concentration of dye (mg/L)Ce Equilibrium concentration of dye (mg/L)CEC Cation exchange capacity (meq/g)Dp Average pore diameter (nm)KF Freundlich constant representing adsorption capacity

(mg/g) (L/mg) 1/n

M Molecular weight of dye (g/mole)nF Adsorption intensitypHpzc pH at the point of zero charge of surfaceqm Maximum adsorption capacity of sorbate per unit mass of

sorbent (mg/g)qm,MB Maximum adsorption capacity for Methylene Blue (mg/g)qm,MY Maximum adsorption capacity for Metanil Yellow (mg/g)R2 Correlation coefficientSBET BET (Brunauer-Emmett-Teller) surface area (m2/g)

SMB Surface area determined from Methylene Blue sorption(m2/g)

SMY Surface area determined from Metanil Yellow sorption(m2/g)

Vmes Mesopore volume (cm3/g)Vmes(%) MesoporosityVmic Micropore volume (cm3/g)Vmic(%) MicroporosityVT Total pore volume (cm3/g)lmax Maximum wavelength (nm)

References

Ahmad, F., Daud, W.M.A.W., Ahmad, M.A., Radzi, R., 2012. Cocoa (Theobroma cacao)shell-based activated carbon by CO2 activation in removing of cationic dye fromaqueous solution: kinetics and equilibrium studies. Chem. Eng. Res. Des. 90(10), 1480e1490.

Ahmadpour, A., Do, D.D., 1997. The Preparation of activated carbon frommacadamiaNutshell by chemical activation. Carbon 35, 1723e1732.

Angın, D., 2013. Production and characterization of activated carbon from sourcherry stones by zinc chloride. Fuel 115, 804e811.

Auta, M., Hameed, B.H., 2011. Optimized waste tea activated carbon for adsorptionof Methylene Blue and Acid Blue 29 dyes using response surface methodology.Chem. Eng. J. 175, 233e243.

Açıkyıldız, M., Gürses, A., Karaca, S., 2014. Agricultural wastes preparation andcharacterization of activated carbon from plant wastes with chemical activa-tion. Microporous Mesoporous Mater. 198, 45e49.

Biniak, S., �Swiatkowski, A., Pakuła, M., Sankowska, M., Ku�smierek, K., Trykowski, G.,2013. Cyclic voltammetric and FTIR studies of powdered carbon electrodes inthe electrosorption of 4-chlorophenols from aqueous electrolytes. Carbon 51,301e312.

Boehm, H.P., 2002. Surface oxides on carbon and their analysis: a critical assess-ment. Carbon 40 (2), 145e149.

Burg, P., Fydrych, P., Cagniant, D., Nanse, G., Bimer, J., Jankowska, A., 2002. Thecharacterization of nitrogen-enriched activated carbons by IR, XPS and LSERmethods. Carbon 40 (9), 1521e1531.

Ceyhan, A.A., Sahin, O., Saka, C., Yalçın, A., 2013. A novel thermal process for acti-vated carbon production from the vetch biomass with air at low temperature bytwo-stage procedure. J. Anal. Appl. Pyrol. 104, 170e175.

Chen, W., Liu, X., He, R.L., Lin, T., Zeng, Q.F., Wang, X.G., 2013. Activated carbonpowders from wool fibers. Powder Technol. 234, 76e83.

Danish, M., Hashim, R., Ibrahim, M.N.M., Sulaiman, O., 2013. Effect of acidic acti-vating agents on surface area and surface functional groups of activated carbonsproduced from Acacia mangium. J. Anal. Appl. Pyrol. 104, 418e425.

Freundlich, H., 1906. On the adsorption in solutions (in German). Z. Phys. Chem. 57(A), 385e470.

Gao, P., Liu, Z.H., Xue, G., Han, B., Zhou, M.H., 2011. Preparation and characterizationof activated carbon produced from rice straw by (NH4)2PO4 activation. Bio-resour. Technol. 102 (3), 3645e3648.

Ghaedi, M., Tavallali, H., Sharifi, M., Kokhdan, S.N., Asghari, A., 2012. Preparation oflow cost activated carbon from Myrtus communis and pomegranate and theirefficient application for removal of Congo red from aqueous solution. Spec-trochim. Acta A 86, 107e114.

Gündo�gdu, A., Duran, C., Sentürk, H.B., Soylak, M., _Imamo�glu, M., €Onal, Y., 2013.Physicochemical characteristics of a novel activated carbon produced from teaindustry waste. J. Anal. Appl. Pyrol. 104, 249e259.

Güzel, F., 1996. The effect of surface acidity upon the adsorption capacities ofactivated carbons. Separ. Sci. Technol. 31 (2), 283e290.

Hameed, B.H., Daud, F.B.M., 2008. Adsorption studies of basic dye on activatedcarbon derived from agricultural waste: Hevea brasiliensis seed coat. Chem. Eng.J. 139 (1), 48e55.

Hejazifar, M., Azizian, S., 2012. Adsorption of cationic and anionic dyes onto theactivated carbon prepared from Grapevine Rhytidome. J. Disper. Sci. Technol. 33(6), 846e853.

H. Say�gılı, F. Güzel / Journal of Cleaner Production 113 (2016) 995e10041004

Huang, Y., Ma, E., Zhao, G., 2015. Thermal and structure analysis on reactionmechanisms during the preparation of activated carbon fibers by KOH activa-tion from liquefied wood-based fibers. Ind. Crop. Prod. 69, 447e455.

Karag€oz, S., Tay, T., Ucar, S., Erdem, M., 2008. Activated carbons fromwaste biomassby sulfuric acid activation and their use on methylene blue adsorption. Bio-resour. Technol. 99 (14), 6214e6222.

Kumar, A., Jena, H.M., 2015. High surface area microporous activated carbons pre-pared from Fox nut (Euryale ferox) shell by zinc chloride activation. Appl. Surf.Sci. 356, 753e761.

Langmuir, I., 1918. The adsorption of gases on plane surfaces of glass, mica andplatinum. J. Am. Chem. Soc. 40 (9), 1361e1403.

Laszlo, K., Szucs, A., 2001. Surface characterization of polyethyleneterephthalate (PET)based activated carbon and the effect of pH on its adsorption capacity fromaqueous phenol and 2, 3, 4-trichlorophenol solutions. Carbon 39 (13),1945e1953.

Li, J., Ng, D.H.L., Song, P., Kong, C., Song, Y., Yang, P., 2015. Preparation and charac-terization of high-surface area activated carbon fibers from silkworm cocoonwaste for congo red adsorption. Biomass Bioenerg. 75, 189e200.

Liu, J., Li, Y., Li, K., 2013. Optimization of preparation of microporous activatedcarbon with high surface area from Spartina alterniflora and its p-nitroanilineadsorption characteristics. J. Environ. Chem. Eng. 1 (3), 389e397.

Lua, A.C., Yang, T., 2005. Characteristics of activated carbon prepared frompistachio-nut shell by zinc chloride activation under nitrogen and vacuumconditions. J. Colloid Interf. 290 (2), 505e513.

Malik, P.K., 2003. Use of activated carbons prepared from sawdust and rice-husk foradsorption of acid dyes: a case study of Acid Yellow 36. Dyes Pigments 56 (3),239e249.

Mastalerz, M., Bustin, R.M., 1995. Application of reflectance micro-transforminfrared spectrometry in studying coal macerals. Fuel 74 (4), 536e542.

Mohanty, K., Das, D., Biswas, M.N., 2006. Preparation and characterization of acti-vated carbons from Sterculia alata nutshell by chemical activation with zincchloride to remove phenol from wastewater. Adsorption 12, 119e132.

Ncibi, M.C., Jeanne-Rose, V., Mahjoub, B., Jean-Marius, C., Lambert, J., Ehrhardt, J.J.,Bercion, Y., Seffen, M., Gaspard, S., 2009. Preparation and characterisation ofraw chars and physically activated carbons derived from marine Posidoniaoceanica (L.) fibres, 2009. J. Hazard Mater. 165 (1e3), 240e249.

Nguyen, H.D., Mai, T.T.T., Nguyen, N.B., Dang, T.D., Le, M.L.P., Dang, T.T., Tran, V.M.,2013. Novel method for preparing micro fibrilated cellulose from bamboo fi-bers. Adv. Nat. Sci. Nanosci. Nanotechnol. 4, 9e19.

Njoku, V.O., Foo, K.Y., Hameed, B.H., 2013. Microwave-assisted preparation ofpumpkin seed hull activated carbon and its application for the adsorptiveremoval of 2,4- dichlorophenoxyacetic acid. Chem. Eng. J. 215e216, 383e388.

Okman, I., Karag€oz, S., Tay, T., Erdem, M., 2014. Activated carbons from grape seedsby chemical activation with potassium carbonate and potassium hydroxide.Appl. Surf. Sci. 293, 138e142.

Oliveira, L.C., Pereira, E., Guimaraes, I.R., Vallone, A., Pereira, M., Mesquita, J.P.,Sapag, K., 2009. Preparation of activated carbons from coffee husks utilizingFeCl3 and ZnCl2 as activating agents. J. Hazard. Mater. 165 (1e3), 87e94.

€Ozdemir, I., Sahin, M., Orhan, R., Erdem, M., 2014. Preparation and characterizationof activated carbon from grape stalk by zinc chloride activation. Fuel Process.Technol. 125, 200e206.

Puziy, A.M., Poddubnaya, O.I., Martınez-Alonso, A., Suarez-Garcıa, F., Tascon, J.M.D.,2002. Synthetic carbons activated with phosphoric acid I. Surface chemistry andion binding properties. Carbon 40 (9), 1493e1505.

Qada, E.N.E., Allen, S.J., Walker, G.M., 2013. Adsorption of basic dyes from aqueoussolution onto activated carbons. Chem. Eng. J. 135 (3), 174e184.

Reffas, A., Bernardet, V., David, B., Reinert, L., Lehocine, M.B., Dubois, M., Batisse, N.,Duclaux, L., 2010. Carbons prepared from coffee grounds by H3PO4 activation:characterization and adsorption of methylene blue and Nylosan red N-2RBL.J. Hazard. Mater. 175 (1e3), 779e788.

Regmi, P., Moscoso, J.L.G., Kumar, S., Cao, X.Y., Mao, J.D., Schafran, G., 2012. Removal ofcopper and cadmium from aqueous solutions using switchgrass biochar producedvia hydrothermal carbonization process. J. Environ. Manage. 109, 61e69.

Santra, A.K., Pal, T.K., Datta, S., 2008. Removal of metanil yellow from its aqueoussolution by fly ash and activated carbon produced from different sources. Separ.Sci. Technol. 43 (6), 1434e1458.

Sarısaçlı, E., 2007. Tomato Paste, the Prime Ministry Foreign Trade Undersecretariatof the Turkey Republic (Salça, Türkiye Cumhuriyeti Basbakanlık Dıs TicaretMüstesarlı�gı). IGEME Publisher, Ankara. www.igeme.gov.tr.

Say�gılı, H., Güzel, F., €Onal, Y., 2015. Conversion of grape industrial processing wasteto activated carbon sorbent and its performance in cationic and anionic dyesadsorption. J. Clean. Prod. 93, 84e93.

Sing, K.S.W., Everett, D.H., Haul, R.A.W., Moscou, L., Pierotti, R.A., Rouquerol, J.,Siemieniewska, T., 1985. Reporting physisorption data for gas/solid systemswith special reference to the determination of surface area and porosity. PureAppl. Chem. 57 (4), 603e619.

Smiciklas, I.D., Milonjic, S.K., Pfendt, P., Raicevic, S., 2000. The point of zero chargeand sorption of cadmium (II) and strontium (II) ions on synthetic hydroxyap-atite. Sep. Purif. Technol. 18 (3), 185e194.

Stavropoulos, G.G., Zabaniotou, A.A., 2005. Production and characterization ofactivated carbons from olive-seed waste residue. Microporous MesoporousMater. 82 (1e2), 79e85.

Sych, N.V., Trofymenko, S.I., Poddubnaya, O.I., Tsyba, M.M., Sapsay, V.I.,Klymchuk, D.O., 2012. Porous structure and surface chemistry of phosphoricacid activated carbon from corncob. Appl. Surf. Sci. 261, 75e82.

Tseng, R.L., 2007. Physical and chemical properties and adsorption type of activatedcarbon prepared from plum kernels by NaOH activation. J. Hazard. Mater. 147(3), 1020e1027.

Valente Nabais, J., Laginhas, C., Ribeiro Carrott, M.M.L., Carrott, P.J.M., CrespoAmor�os, J.E., Nadal Gisbert, A.V., 2013. Surface and porous characterisation ofactivated carbons made from a novel biomass precursor, the esparto grass.Appl. Surf. Sci. 265, 919e924.

Wang, Y., Wang, X., Wang, X., Liu, M., Yang, L., Wu, Z., Xia, S., Zhao, J., 2012.Adsorption of Pb(II) in aqueous solutions by bamboo charcoal modified withKMnO4 via microwave irradiation. Colloid. Surf. A 414, 1e8.

Xiao, H., Peng, H., Deng, S., Yang, X., Zhang, Y., Li, Y., 2012. Preparation of activatedcarbon from edible fungi residue by microwave assisted K2CO3 activation-Application in reactive black 5 adsorption from aqueous solution. Bioresour.Technol. 111, 127e133.

Yang, J., Qiu, K., 2011. Development of high surface area mesoporous activatedcarbons from herb residues. Chem. Eng. J. 167 (1), 148e154.

Yang, H., Yan, R., Chen, H., Lee, D.H., Zheng, C., 2007. Characteristics of hemicellu-lose, cellulose and lignin pyrolysis. Fuel 86 (12e13), 1781e1788.

Zhang, Z., Luo, X., Liu, Y., Zhou, P., Ma, G., Lei, Z., Lei, L., 2015. A low cost and highlyefficient adsorbent (activated carbon) prepared from waste potato residue.J. Taiwan Inst. Chem. Eng. 49, 206e211.

Zuim, D.R., Carpin�e, D., Distler, G.A.R., Scheer, A.P., Igarashi-Mafra, L., Mafra, M.R.,2011. Adsorption of two coffee aromas from synthetic aqueous solution ontogranular activated carbon derived from coconut husks. J. Food Eng. 104 (2),284e292.