Chemical Engineering Journal 317 (2017) 493–502

Contents lists available at ScienceDirect

Chemical Engineering Journal

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

CO2-activated porous carbon derived from cattail biomass for removalof malachite green dye and application as supercapacitors

http://dx.doi.org/10.1016/j.cej.2017.02.1051385-8947/� 2017 Elsevier B.V. All rights reserved.

⇑ Corresponding author at: College of Material Science and Engineering,Northeast Forestry University, Harbin, PR China.

E-mail address: donglinwlj@163.com (L. Wang).

Miao Yu, Yingying Han, Jian Li, Lijuan Wang ⇑College of Material Science and Engineering, Northeast Forestry University, Harbin, PR ChinaResearch Center of Wood Intelligent Science, Northeast Forestry University, Harbin, PR China

h i g h l i g h t s

� CO2-activated porous carbon (CAC)from cattail was obtained by facilecarbonization and CO2 activation.

� The obtained CAC had excellentmalachite green adsorption capacityof 210.18 mg/g.

� The saturated CAC could be reused bythermal regeneration.

� The CAC exhibited a specificcapacitance of 126.5 F/g in 6 M KOHsolution.

g r a p h i c a l a b s t r a c t

a r t i c l e i n f o

Article history:Received 24 December 2016Received in revised form 16 February 2017Accepted 17 February 2017Available online 21 February 2017

Keywords:CattailActivated carbonMalachite greenAdsorptionSupercapacitors

a b s t r a c t

In this study, we developed a feasible and cost-effective method for obtaining porous carbon from cattailbiomass as a promising material to be applied in various applications. The obtained carbon was furthertreated via CO2 activation without any other chemical reagents. The results showed that the as-preparedactivated carbon had a high specific surface area of 441.12 m2/g after CO2 activation. Besides, theactivated carbon showed excellent capacity to adsorb malachite green dye, and the effects of initialdye concentration, adsorption time, initial pH, and adsorption temperature were studied. Moreover,the activated carbon exhibited exceptional electrochemical performance as supercapacitors, with specificcapacitance reaching 126.5 F/g at a current density of 0.5 A/g within a potential range of �1.0 to 0 V in a6 M KOH solution. The excellent adsorption capacities and electrochemical performance suggested thatthe obtained activated carbon could be a promising candidate as an adsorbent and supercapacitor.

� 2017 Elsevier B.V. All rights reserved.

1. Introduction

Dyes are widely used in various industries such as textiles,paint, tanneries, plastics, rubber, pharmacies, electroplating, andfood processing [1], and discharge of wastewater from these indus-

tries without proper treatment will result in severe environmentalproblems to aquatic animals and human beings [2]. Most of thesynthetic dyes are harmful, thermally stable, and resistant to light,and are difficult to biodegrade owing to their complicated chemicalstructure [3]. As a result, in recent decades, there has been a grow-ing demand for effective removal of these dyes from wastewater.

Malachite green (MG), a basic and cationic triphenylmethanedye, is widely utilized in dyeing of silk, cotton, wool, plastics, andpaper, as well as in the fish industry as a biocide and medicaldisinfectant to control fungal and protozoan infections [2,4,5].

494 M. Yu et al. / Chemical Engineering Journal 317 (2017) 493–502

However, MG has been reported to produce many toxicologicalsymptoms, including carcinogenicity, teratogenicity, mutagenicity,and pleural infections [6,7]. Therefore, effective removal of toxicMG from wastewater before discharging to the environment hasbeen given more attention over the past few years.

Many approaches have been proposed for the removal of dyesfrom wastewater, such as adsorption [8], chemical coagulation[9], oxidation, ozonization [10], membrane separation [11], ion-exchange, and electrochemical treatments [9,12]. Among thesetechniques, adsorption using activated carbon is the most facileand effective method for the removal of noxious dyes fromwastewater [13]. However, the usage of commercially availableactivated carbon from bituminous coal, lignite, and coconut shellshas been limited owing to its high cost [14,15], thus leading to anurgent requirement for low-cost and effective rawmaterials for theproduction of activated carbon. Various materials, especially fromagricultural wastes, ranging from bamboo, orange peel, oil palmfiber, rice bran, and coconut husk, have been employed for the pro-duction of activated carbon [16–20]. The activated carbon obtainedfrom these materials have been noted to show excellent adsorptionproperties towards dyes such as methylene blue, MG, and Direct NBlue-106, which make them promising adsorbents for sewagetreatment. In addition, it has also showed excellent electrochemi-cal performance to be applied to supercapacitors. Activated carbonderived from biomass such as coconut fibers [21], corn stalk core[22] and potato waste residue [23] have been prepared andshowed good electrochemical properties for supercapacitors. How-ever, the chemical activation process was relatively complicatedand the sewage (KOH and ZnCl2) may do harm to the environment.Therefore, it is urgent to find a kind of low-cost raw material andfacile preparation process to produce activated carbon for applica-tion in supercapacitors.

Cattail is an aquatic plant mainly composed of lignocellulose,and is widely used in artificially constructed wetlands for the treat-ment of wastewater and polluted soil [24]. Cattail biomass is abun-dantly available in nature, and cattail fiber possesses a particularbamboo-like structure with a ‘‘p”-shaped gap in the middle part[25]. This unique structure of cattail provides a large specific sur-face area and forms an enclosed space, thus making it a good can-didate for the production of adsorbents. However, to the best of ourknowledge, there are only a few reports on the preparation of acti-vated carbon from cattail for the adsorption of dyes and phenoliccompounds [24,26]. Moreover, the available methods for the pro-duction of activated carbon involve complex chemical activationsuch as potassium hydroxide (KOH) activation or phosphoric acid(H3PO4) activation, after which, a subsequent washing process isrequired to adjust the pH of the product. Thus, a large quantityof water is needed and the generated sewage without proper treat-ment may result in serious environmental pollution, limiting theapplications of these methods. Furthermore, there are only a fewreports on the preparation of activated carbon from cattail fibervia a physical activation process.

Therefore, in the present study, cattail fiber derived activatedcarbon was prepared through a physical activation process usingCO2 as the activation gas without further treatment for the removalof MG from aqueous solution. The effects of contact time, initialdye concentration, solution pH, and temperature on the adsorptionof the prepared activated carbon were studied. In addition, theequilibrium and kinetic data of the adsorption process were mea-sured to investigate the adsorption isotherms, kinetics, and ther-modynamics. Moreover, the potential application of the as-prepared activated carbon in supercapacitors was investigatedusing 6 M KOH solution as the electrolyte. The results obtainedin this study are valuable for the multifunctional applications ofthe as-prepared activated carbon in dye removal and assupercapacitors.

2. Materials and methods

2.1. Materials

The cattail used in this study for the preparation of activatedcarbon was collected from a local pond in Daqing, China. Malachitegreen (MG) and potassium hydroxide (KOH) was purchased fromKermel LLC (Tianjin, China). Polytetrafluoroethylene (PTFE) wasbought from Chenghexiang (Shenzhen, China), acetylene blackwas obtained from Tianyi LLC (Tianjin, China), and nickel foamwas procured from Jiayisheng (Kunshan, China). All chemicalreagents were of analytical grade and used as received without fur-ther purification.

2.2. Preparation of activated carbon

The collected cattails were washed with deionized water toremove soil and dust, and dried at 80 �C for 12 h to a constantweight for the removal of moisture and other volatile impurities.The dried cattails were then cut into pieces (1 cm in height and1.5 cm in diameter) and stored in a desiccator. Carbonization wasperformed in a tube furnace under N2 atmosphere. First, the cat-tails were heated from room temperature to 550 �C at a rate of10 �C/min. Then, the samples were heated from 550 �C to 850 �Cat a rate of 5 �C/min and maintained at 850 �C for 2 h to obtain car-bon monolith. The obtained carbon monolith was further subjectedto CO2 activation. In this process, the carbon monolith was heatedin the tube furnace from room temperature to 200 �C at a rate of5 �C/min under pure CO2 flow and kept at 200 �C for 2 h.Subsequently, the sample was heated from 200 �C to 800 �C at arate of 3 �C/min and maintained at 800 �C for 2 h to obtainCO2-activated carbon, denoted as CAC. The activationmechanism of CO2 could be attributed to the chemical reaction:C(s) + CO2 (g)? 2CO (g).

2.3. Adsorption experiments

MG solutions of various concentrations were prepared usingdistilled water. The pH of the dye solutions was adjusted using0.01 M HCl or NaOH. The adsorption experiments were performedby adding 20 mg of CAC to 50 mL of MG solutions of various con-centrations at specific pH and temperature in a thermostatic waterbath on a rotary shaker at 140 rpm until the adsorption equilib-rium was reached. After centrifugation for 15 min to remove CACparticles, the concentration of MG in the supernatant was deter-mined using visible spectrophotometer at a wavelength of619 nm. The amount of adsorbed MG was calculated based onthe following equation: qe = (C0 � Ce) � V/m, where qe is theamount of the absorbed dye (mg/g) at equilibrium, C0 and Ce arethe initial and residual concentrations of the dye at equilibrium(mg/L), respectively, V is the volume of the dye solution (L), andm is the mass of the adsorbent used (g). Each adsorption experi-ment was repeated for three times.

2.4. Thermal regeneration of CAC

Thermal regeneration of CAC was conducted after three regen-eration cycles. In brief, 20 mg of CAC were added to 50 mL of MGsolution (100 mg/L) in a conical flask and mechanically shakenfor 12 h at 30 �C. Subsequently, the solution was filtered and theMG adsorption capacity was determined. The spent CAC was driedin an oven for 12 h and then placed in a tube furnace, heated fromroom temperature to 300 �C at a rate of 5 �C/min, maintained at300 �C for 2 h, and allowed to cool to room temperature to obtainregenerated CAC.

M. Yu et al. / Chemical Engineering Journal 317 (2017) 493–502 495

2.5. Characterization of CAC

The surface morphology of the CAC samples was observed byscanning electron microscopy (SEM, FEI, Sirion, Japan) aftersputter-coating the samples with a thin layer of gold. The pore-size distribution, mean pore diameter, total pore volume, andspecific surface area of the samples were measured using N2

adsorption analyzer (Maike, ASAP 2020, USA) based on Brunauer–Emmet–Teller (BET) and Barrett–Joyner–Halenda (BJH)methods at �196 �C.

2.6. Electrochemical measurements

The electrochemical measurements were conducted on an elec-trochemical workstation (CS350, Wuhan, China) in a three-

Fig. 1. (a) SEM images of pristine cattail fibers at low magnification and (b) high magnifimagnification and (e) high magnification. (f) EDS spectrum of carbon. (g) SEM images ofcarbon.

Fig. 2. (a) The N2 adsorption–desorption isoth

electrode system with 6 M KOH solution serving as the aqueouselectrolyte. The working electrode was prepared by loading aslurry containing 80 wt% CAC (about 4 mg), 10 wt% polytetrafluo-roethylene (PTFE), and 10 wt% acetylene carbon black on a nickelfoam substrate, followed by drying in an oven at 100 �C for 12 hand pressing the ensemble together under a pressure of 1 MPa.Ag/AgCl and Pt wire served as the reference and counter electrodes,respectively. The cyclic voltammogram (CV) curves were obtainedin a potential range from �1.0 to 0 V by varying the scan ratefrom 5 to 100 mV/s. Galvanostatic charge/discharge (GCD)measurements were conducted in a potential window from�1.0 to 0 V under different current densities from 0.5 to 10 A/g.The electrochemical impedance spectroscopy (EIS) analysis wasperformed at the frequency range of 0.01 Hz–100 kHz with anamplitude of 5 mV. The specific capacitance (Cs) was calculated

cation. (c) EDS spectrum of pristine cattail fibers. (d) SEM images of carbon at lowCAC at low magnification and (h) high magnification. (i) EDS spectrum of activated

erm and (b) pore-size distribution of CAC.

496 M. Yu et al. / Chemical Engineering Journal 317 (2017) 493–502

from the discharge curves using the equation, Cs = (I � Δt)/(m � ΔV), where I is the constant discharge current (A), Δt is thedischarge time (s), m is the mass of electroactive materials (g),and ΔV is the discharge voltage change (V).

Fig. 3. Effect of contact time and initial MG solution concentration on MGadsorption.

Fig. 4. Effect of solution temperature on MG removal at various initial MGconcentrations.

Fig. 5. Effect of solution pH on MG removal (Conditions: adsorbent dosage, 20 mg;rotation speed, 140 rpm; temperature, 30 �C).

3. Results and discussion

3.1. SEM observation

The structure and morphology of cattail fibers and CAC werecharacterized by SEM. As shown in Fig. 1, the surface of pristinecattail fibers was smooth and neat, barely with a split, and the fiberwidth was around 16.5 lm (Fig. 1a and b). After carbonization andactivation, the strip fiber structure remained intact, but the fibersbecame irregular and wrinkled, and the width got reduced, whichmay be owing to the pyrolysis of biopolymers under high temper-ature after which the carbon framework was formed. The oxygenpeaks in the EDS spectrum of carbon and CAC were reduced, whichfurther indicated the decomposition of oxygen-containing func-tional groups in cattail fibers.

3.2. CAC porous structure characterization

The N2 adsorption–desorption experiment was performed tocharacterize the pore structure of the CAC samples. The N2 adsorp-tion–desorption isotherm and corresponding pore-size distribu-tion curve are presented in Fig. 2. The CA showed a type IVisotherm according to the IUPAC classification [27]. A H4-type hys-teresis loop could be observed at a relative pressure (P/P0) range of0.4–1.0 ascribed to capillary condensation, indicating the existenceof mesopores in the CAC [28]. In addition, the CAC samples exhib-ited narrow BJH adsorption plots as shown in Fig. 2(b). The distri-bution of pore sizes in the range of 2.1–4.0 nm further illustratedthat the CAC contained mesopores. The maximum specific surfacearea and pore volume were calculated as 441.12 m2/g and0.33 cm3/g by the BET method, respectively. And the average porediameter (BET) was less than 2.09 nm, indicating the mesoporecharacter.

3.3. MG adsorption

3.3.1. Effect of contact time and initial MG concentration on MGremoval

The effects of contact time and initial MG concentration on MGremoval at 30 �C are shown in Fig. 3. The MG adsorption rate wasrapid at the initial phase of 120 min, and thereafter became sloweruntil equilibrium was reached and no more MG could be removedfurther. This may be attributed to the fact that at the beginning ofadsorption, numerous surface sites are available for adsorption,which act as a high driving force for the transfer of MG [16]. How-ever, after a lapse of time, the residual surface sites are difficult tobe occupied, which may be owing to the repelling force betweenthe solute molecules on the solid and bulk phases [29]. The adsorp-tion capacity at equilibrium, qe, increased from 93.24 to210.18 mg/g with an increase in the initial MG concentration from40 to 100 mg/L. This is because the initial dye concentration pro-vides a powerful driving force to overcome the resistance to masstransfer of the dye between the aqueous and solid phases [30], andthe increase in initial MG concentration enhances the interactionbetween dye molecules and adsorbents. Therefore, MG adsorptionincreased with the increasing initial concentration of MG. Further-more, it could also be clearly observed from Fig. 3 that the adsorp-tion equilibrium was reached at around 240 min with the initialMG concentrations ranging from 40 to 80 mg/L. For 100 mg/L MGsolutions, about 360 min was needed to reach equilibrium. Thus,

Fig. 6. Langmuir and Freundlich isotherms for the adsorption of MG by CAC at different temperatures.

Table 1Langmuir and Freundlich isotherm parameters for the adsorption of MG onto CAC at different temperatures.

Temperatures

30 �C 40 �C 50 �C 60 �C

Langmuir isothermqm (mg/g) 276.24 284.09 288.18 295.86KL (L/mg) 0.1877 0.1859 0.1935 0.1946R2 0.9909 0.9941 0.9956 0.9971

Freundlich isothermKF (mg/g(L/mg)1/n) 62.94 62.81 64.56 65.28n 2.259 2.191 2.179 2.130R2 0.9678 0.9737 0.9830 0.9896

M. Yu et al. / Chemical Engineering Journal 317 (2017) 493–502 497

the experimental data was measured at 720 min to ensure thatequilibrium was reached.

3.3.2. Effect of solution temperature on MG removalThe effect of solution temperature on MG removal by CAC is

presented in Fig. 4. At the same initial MG concentration, the MGadsorption capacity of CAC slightly increased with the increasingtemperature from 30 �C to 60 �C. The MG adsorption capacityslightly increased from 93.24 to 94.36 mg/g with the solution tem-perature increase from 30 to 60 �C at MG concentration of 40 mg/g.But the MG adsorption capacity increased from 210.18 to216.19 mg/g at MG concentration of 100 mg/g. A similar resulton MG adsorption by activated carbon was reported by Onalet al. [9]. This effect of temperature on MG adsorption may beattributed to the endothermic nature of the adsorption reaction.At higher temperature, the dye molecules have more energy tointeract with the active sites on the CAC surface. Moreover, the dif-fusion rate of the adsorbate molecules across the external bound-

Table 2Comparison of maximum adsorption capacity of MG dye using activated carbonderived from various waste biomasses.

Adsorbents Adsorptioncapacity(mg/g)

Activationmethod

Ref.

Arundo donax root based carbon 8.69 – [32]Borassus aethiopum flower biomass based

activated carbon48.48 H2SO4 [33]

Ricinus communis based activatedcarbon

27.78 H2SO4 [34]

Jute fiber based carbon 196.08 H3PO4 [35]CAC 210.18 CO2 This

work

ary layer increases with the increase in solution temperaturebecause of the decrease in the viscosity of the adsorbate solution[31].

3.3.3. Effect of solution pH on MG removalThe effects of initial pH of external solutions on the MG removal

percentage are shown in Fig. 5. The MG removal percentage wasfound to significantly increase from 35.55% to 93.24% with theincrease in solution pH from 2 to 8, which may be attributed tothe increase in positive charges on the surface sites of the CACand the decrease in adsorption at lower pH owing to electrostaticrepulsive force. In addition, MG is a cationic dye, and the presenceof considerable amount of H+ at lower pH may result in competi-tion with dye cations for the adsorption sites of CAC, thus leadingto a decrease in dye adsorption. However, no obvious change in thedye adsorption was observed at pH 8–12, which may be attributedto the increasing negative charge of the CAC surface at this pHleading to enhanced electrostatic force of the CAC surface towardspositively charged dye cations [31].

3.3.4. Adsorption isothermsThe adsorption isotherms could illustrate the interactive behav-

ior between the adsorbate in the solution and adsorbent at con-stant temperature under equilibrium condition. In the presentstudy, the equilibrium data under different conditions were fittedwith Langmuir and Freundlich isotherms. The linear form of Lang-muir isotherm equation can be expressed as [24]:

Ce

qe¼ 1

KLqmþ Ce

qmð1Þ

where qe (mg/g) is the amount of dye adsorbed at equilibrium time,Ce (mg/L) is the equilibrium concentration of the dye solution, and

Fig. 7. (a) Lagergren-first-order kinetic model, (b) pseudo-second-order kineticmodel, and (c) intra-particle diffusion model for adsorption of MG onto CAC at30 �C.

498 M. Yu et al. / Chemical Engineering Journal 317 (2017) 493–502

qm (mg/g) and KL (L/mg) are the maximum adsorption capacity andequilibrium adsorption constant related to the free energy ofadsorption, respectively.

The logarithmic form of Freundlich adsorption isotherm can begiven as [24]:

log qe ¼ logKF þ 1nlogCe ð2Þ

where KF (L/mg) and n are the isotherm constants that represent thecapacity and intensity of the adsorption, respectively.

The fitted Langmuir and Freundlich plots and the correspondingisotherm parameters and correlation coefficients are shown inFig. 6 and Table 1, respectively. It can be noted that the fitting toLangmuir isotherms was better, when compared with that to Fre-undlich isotherms. The higher values of R2 from Langmuir modelalso confirmed that the adsorption was better represented by theLangmuir isotherm model, indicating that the process of adsorp-tion was mainly monolayer. The values of n > 1 suggested a favor-able adsorption on CAC [9]. Moreover, the adsorption capacitiesincreased with increasing temperature, confirming that theadsorption process was endothermic. Table 2 shows the compar-ison of the maximum adsorption capacity of MG dye on activatedcarbon derived from various agricultural by-products. The as-prepared CAC in this study shows relatively higher MG dye adsorp-tion than that reported in the previous works.

3.3.5. Adsorption kineticsThe Lagergren pseudo-first-order kinetics model, pseudo-

second-order kinetics model, and intraparticle diffusion modelwere employed to investigate the adsorption kinetics. The Lager-gren pseudo-first-order kinetics model was developed by Lager-gren and Svenska [36] to predict the adsorption kinetics, and itslinear form can be expressed as follows:

logðqe � qtÞ ¼ log qe � k1t ð3Þwhere qe and qt (mg/g) are the amounts of MG adsorbed at equilib-rium and given time, respectively, and k1 (min�1) is the rate con-stant of pseudo-first-order adsorption.

The pseudo-second-order kinetic model was developed by Hoand Mckay [36] to predict the entire adsorption behavior, and itslinear form can be given as follows:

tqt

¼ 1k2q2

eþ tqe

ð4Þ

where k2 (g/mg/min) is the rate constant of second-orderadsorption.

As shown in Fig. 7(a) and (b), the pseudo-second-order modelshowed good agreement with the experimental kinetics data,which could also be obtained from higher R2 values as shown inTable 3. In addition, the adsorption capacities (qe) calculated fromthe pseudo-second-order model were closer to the experimentalvalues (qe,exp), indicating that the pseudo-second-order kineticmodel is more likely to describe the adsorption behavior of MGonto CAC.

However, as the two above-mentioned models could not suffi-ciently illustrate the diffusion mechanism during the adsorptionprocess, the intraparticle diffusion model was employed to furtheranalyze the adsorption process. The intraparticle diffusion can beexpressed as follows:

qt ¼ kpit1=2 þ Ci ð5Þwhere Ci represents the intercept of stage i and kpi is the intraparti-cle diffusion rate constant (mg/g min1/2), which can be obtainedfrom the slope of the linear plots of qt versus t1/2. It can be notedfrom the plots in Fig. 7(c) that the adsorption process comprised

two stages and the plot was linear within a certain range, but non-linear over the entire time range, revealing that more than onemodel contributed to the adsorption mechanism with respect tothe removal of MG by CAC. The initial linear portion might be owingto the electrostatic attraction between the dye and external surfaceof the adsorbent [31], and the second linear portion may be attrib-

Table 3Kinetic parameters for the adsorption of MG onto CAC.

C0 (mg/L) qe,exp (mg/g) Lagergren-first-order kinetic model Pseudo-second-order kinetic model

qe1 k1 � 102 (1/min) R2 qe2 k2 � 105 (g/mg min) R2

40 93.24 59.65 11.15 0.9417 104.93 18.68 0.988760 137.57 110.12 11.49 0.9702 159.51 9.22 0.988580 174.38 95.73 9.63 0.8551 191.74 9.39 0.9877100 210.18 134.72 9.93 0.9721 230.95 9.61 0.9967

Fig. 8. Recycling of CAC for adsorption of MG by adsorption–regenerationprocesses.

Table 4Thermodynamic parameters for the adsorption of MG onto CAC.

Concentration (mg/L) Temperature (�C) Kc 4G� (kJ/mol) 4H� (kJ/mol) 4S� (J/mol K)

40 30 13.79 �6.61 4.93 38.0040 14.26 �6.9250 15.43 �7.3560 16.35 �7.74

60 30 11.07 �6.06 2.27 27.4940 11.37 �6.3350 11.76 �6.6260 11.98 �6.88

80 30 6.81 �4.83 6.66 37.9040 7.35 �5.1950 7.92 �5.5660 8.66 �5.98

100 30 5.28 �4.19 5.28 31.2340 5.61 �4.4950 5.97 �4.8060 6.38 �5.13

M. Yu et al. / Chemical Engineering Journal 317 (2017) 493–502 499

uted to the gradual adsorption because of intraparticle diffusion[29]. As shown in Fig. 7(c), the linear lines in the second stage didnot pass through the origin, which may be ascribed to the differencein the mass transfer rate between the initial and final stages ofadsorption. Furthermore, the values of C increased with the initialMG concentration, indicating that boundary layer effect may beinvolved in the adsorption process [29].

3.3.6. Adsorption thermodynamicsThe three thermodynamic parameters that must be considered

during evaluation of the adsorption process are the free energychange (4G�), enthalpy (4H�), and entropy (4S�). The value of4H� and 4S� can be obtained by van’t Hoff equation as follows:

lnKc ¼ DS�

R� DH�

RTð6Þ

where R (8.314 J/mol K) is the universal gas constant, T (K) is theabsolute temperature, and Kc is the equilibrium constant. The Kc

value can be calculated as:

Kc ¼ CSe

CAeð7Þ

where CAe and CSe are the equilibrium concentration of the dye onadsorbent (mg/L) and in solution (mg/L), respectively.

The values of 4H� and 4S� can be obtained from the slope andintercept of ln Kc versus 1/T, respectively, and 4G� can be obtainedfrom the following relation:

DG� ¼ �RT lnKc ð8ÞThe 4G�, 4H�, and 4S� values of adsorption are shown in

Table 4. The positive value of 4H� (4.93, 2.27, 6.66, 5.28 kJ/mol)confirmed that the adsorption process was endothermic in nature,which was consistent with the result obtained earlierdemonstrating that the MG removal increased with increasingsolution temperature. The positive value of 4S� (38.00, 27.49,37.90, 31.23 J/mol K) indicated increased randomness during the

adsorption process [30]. In addition, the negative values of 4G�(–4.19 to –7.74 kJ/mol) suggested the spontaneity of the adsorp-tion process. Table 4 shows that the negative values of Gibbs freeenergy decreased with increasing the temperature, indicating thespontaneous nature and feasibility of the adsorption via physicalforces as well as high tendency of adsorbent to the adsorbate[10]. Thus, higher temperature could promote the adsorption ofMG, which is in accordance with the results of the adsorptionisotherms.

3.4. Regeneration

Thermal regeneration of the spent activated carbon was per-formed after three cycles, and the value of adsorption capacity

Fig. 9. Electrochemical performance of the CAC electrodes: (a) CV curves of the CAC electrodes at different scan rates. (b) Charge/discharge curves of the CAC electrodes atdifferent constant current densities. (c) Calculated specific capacitances of the CAC electrodes from the discharge curves at different current densities. (d) Nyquist plots of theCAC electrodes in 6 M KOH solution.

500 M. Yu et al. / Chemical Engineering Journal 317 (2017) 493–502

was investigated. It can be seen from Fig. 8 that the adsorptioncapacity of CAC decreased from 210.18 to 148.51 mg/g after threeregeneration cycles and the regeneration efficiency was 70.66%.This finding illustrated the efficacy of thermal treatment on theregeneration of dye-adsorbed CAC, suggesting that this methodcould also be applied for the regeneration of CAC used for theremoval of other organic pollutants.

Table 5Comparison of specific capacitance of activated carbon derived from various wastebiomasses.

Electrode materials Specificcapacitance(F/g)

Activationmethod

Ref.

Banana fibers based activated carbon 74 KOH/ZnCl2

[39]

Coconut fiber based activated carbon 155 KOH [21]Potato waste residue based activated

carbon255 ZnCl2 [23]

Corn stalk core based activated carbon 323 KOH [22]CAC 126.5 CO2 This

work

3.5. Electrochemical performance of CAC

To investigate the electrochemical performance of CAC, CV andGCD measurements were performed within the potential windowof measurement in a 6 M KOH solution with a three-electrode sys-tem. The CV curves of the CAC samples were tested at scan ratesfrom 5 to 100 mV/s, as shown in Fig. 9 (a). It can be observed thatthe CV curves of the electrode displayed a nearly rectangular shapein a potential range from �1.0 to 0 V, and maintained a quasi-rectangular shape even at a high potential scan rate of 100 mV/s,implying the formation of a perfect electrical double layer acrossthe carbon network [37]. Furthermore, all of the discharge curveswere almost symmetrical to the corresponding charge curves,revealing good electrochemical reversibility of the electrode. TheGCD curves of the CAC electrodes at current densities ranging from0.5 to 10 A/g are illustrated in Fig. 9(b). The five charge/dischargecurves exhibited a quasi-triangular shape without obvious IR drop,which suggested the good capacitive behavior and electrochemicalreversibility of the CAC samples [38]. The specific capacitance wascalculated to be 126.5 F/g at a current density of 0.5 A/g. In addi-tion, the charge/discharge time decreased with the current densityranging from 0.5 to 10 A/g, which may be ascribed to the fact that

the electrolyte ions have sufficient time to enter and diffuse intothe pores of the electrodes at lower current densities, when com-pared with that at higher current densities [23]. Table 5 showsthe comparison of the specific capacitance of activated carbonderived from different biomass. Although the as-prepared CACexhibits a relatively lower specific capacitance, it has a cost-effective and facile preparation compared with those in theliteratures.

To further investigate the behavior of the as-prepared CAC elec-trode as a supercapacitor, the EIS analysis was performed in a 6 MKOH aqueous solution at a frequency range of 0.01 Hz–100 kHz.The Nyquist plots of the sample exhibited a linear trend at low fre-quency as shown in Fig. 9(d), indicating an ideal capacitive behav-ior of the electrodes [40]. At intermediate frequency, the inclinedpart of the curve (�45�) was associated with Warburg impedancecaused by the frequency dependence of ion diffusion/transport

M. Yu et al. / Chemical Engineering Journal 317 (2017) 493–502 501

from the electrolyte to the electrode surface [41]. At high fre-quency, series resistance (Rs) was obtained from the intercept atthe real axis (Z0) in the Nyquist plot as shown in the inset ofFig. 9. The Rs indicates the internal resistance of the electrodesand contact resistance at the interface of the active material/cur-rent collector [23], and the Rs value of the CAC electrode was1.53 X, which revealed good ion transfer performance and excel-lent electronic conductivity of the sample.

4. Conclusions

In the present study, activated carbon derived from cattail fiberwas prepared using a facile and cost-effective method and CO2

activation process. The activated carbon exhibited a high specificsurface area of 441.12 m2/g after CO2 activation for 2 h. Further-more, the as-prepared activated carbon showed high MG adsorp-tion capacity of 210.18 mg/g. Subsequently, the effects of initialdye concentration, solution pH, contact time, and temperature onMG adsorption were investigated. The adsorption process fittedwell with the Langmuir model (R2 > 0.99) as well as pseudo-second-order model (R2 > 0.98). Further thermodynamic analysisindicated that the removal of MG from aqueous solution was aspontaneous and endothermic process. The activated carbon couldbe recycled using a thermal treatment method. Besides, the acti-vated carbon showed good electrochemical performance as asupercapacitor, with specific capacitance reaching 126.5 F/g at acurrent density of 0.5 A/g within a potential range from �1.0 to0 V. These results suggested that the activated carbon derived fromcattail fiber could be a potential low-cost material for cationic dyeremoval and energy storage.

Acknowledgments

This work was supported by Heilongjiang Province outstandingyouth science fund (JC201301).

References

[1] M. Ghasemi, S. Mashhadi, M. Asif, I. Tyagi, S. Agarwal, V.K. Gupta, Microwave-assisted synthesis of tetraethylenepentamine functionalized activated carbonwith high adsorption capacity for Malachite green dye, J. Mol. Liq. 213 (2016)317–325.

[2] Y. Yin, C. Li, C. Song, P. Tao, M. Sun, Z. Pan, et al., The design of coal-basedcarbon membrane coupled with the electric field and its application on thetreatment of malachite green (MG) aqueous solution, Colloids Surf., A 506(2016) 629–636.

[3] M. Ghaedi, F.N. Azad, K. Dashtian, S. Hajati, A. Goudarzi, M. Soylak, Centralcomposite design and genetic algorithm applied for the optimization ofultrasonic-assisted removal of malachite green by ZnO Nanorod-loadedactivated carbon, Spectrochim. Acta Part A: Mol. Biomol. Spectrosc. 167(2016) 157–164.

[4] Y. Tian, P. Liu, X. Wang, H. Lin, Adsorption of malachite green from aqueoussolutions onto ordered mesoporous carbons, Chem. Eng. J. 171 (2011) 1263–1269.

[5] P. Saha, S. Chowdhury, S. Gupta, I. Kumar, Insight into adsorption equilibrium,kinetics and thermodynamics of Malachite Green onto clayey soil of Indianorigin, Chem. Eng. J. 165 (2010) 874–882.

[6] S. Agarwal, I. Tyagi, V.K. Gupta, S. Mashhadi, M. Ghasemi, Kinetics andthermodynamics of Malachite Green dye removal from aqueous phase usingiron nanoparticles loaded on ash, J. Mol. Liq. 223 (2016) 1340–1347.

[7] Y. Kan, Q. Yue, J. Kong, B. Gao, Q. Li, The application of activated carbonproduced from waste printed circuit boards (PCBs) by H3PO4 and steamactivation for the removal of malachite green, Chem. Eng. J. 260 (2015) 541–549.

[8] I.A. Tan, A.L. Ahmad, B.H. Hameed, Adsorption of basic dye on high-surface-area activated carbon prepared from coconut husk: equilibrium, kinetic andthermodynamic studies, J. Hazard. Mater. 154 (2008) 337–346.

[9] M.A. Ahmad, R. Alrozi, Removal of malachite green dye from aqueous solutionusing rambutan peel-based activated carbon: equilibrium, kinetic andthermodynamic studies, Chem. Eng. J. 171 (2011) 510–516.

[10] F. Nekouei, H. Kargarzadeh, S. Nekouei, I. Tyagi, S. Agarwal, V. Kumar Gupta,Preparation of Nickel hydroxide nanoplates modified activated carbon forMalachite Green removal from solutions: kinetic, thermodynamic, isothermand antibacterial studies, Process Saf. Environ. Prot. 102 (2016) 85–97.

[11] S. Sachdeva, A. Kumar, Preparation of nanoporous composite carbonmembrane for separation of rhodamine B dye, J. Membr. Sci. 329 (2009)2–10.

[12] Y. Sun, J. Lin, Y. Zhan, Adsorption of congo red from aqueous solution onsurfactant-modified zeolites with different coverage types: behavior andmechanism, Sep. Sci. Technol. 48 (2013) 2036–2046.

[13] L. Ai, H. Huang, Z. Chen, X. Wei, J. Jiang, Activated carbon/CoFe2O4composites: facile synthesis, magnetic performance and their potentialapplication for the removal of malachite green from water, Chem. Eng. J.156 (2010) 243–249.

[14] V.K. Gupta, Suhas, Application of low-cost adsorbents for dye removal – areview, J. Environ. Manage. 90 (2009) 2313–2342.

[15] A.A. Attia, B.S. Girgis, N.A. Fathy, Removal of methylene blue by carbonsderived from peach stones by H3PO4 activation: batch and column studies,Dyes Pigm. 76 (2008) 282–289.

[16] B.H. Hameed, M.I. El-Khaiary, Equilibrium, kinetics and mechanism ofmalachite green adsorption on activated carbon prepared from bamboo byK2CO3 activation and subsequent gasification with CO2, J. Hazard. Mater. 157(2008) 344–351.

[17] A. Khaled, A. El Nemr, A. El-Sikaily, O. Abdelwahab, Removal of Direct N Blue-106 from artificial textile dye effluent using activated carbon from orangepeel: adsorption isotherm and kinetic studies, J. Hazard. Mater. 165 (2009)100–110.

[18] I.A.W. Tan, B.H. Hameed, A.L. Ahmad, Equilibrium and kinetic studies on basicdye adsorption by oil palm fibre activated carbon, Chem. Eng. J. 127 (2007)111–119.

[19] R.M. Suzuki, A.D. Andrade, J.C. Sousa, M.C. Rollemberg, Preparation andcharacterization of activated carbon from rice bran, Bioresour. Technol. 98(2007) 1985–1991.

[20] I. Tan, A. Ahmad, B. Hameed, Optimization of preparation conditions foractivated carbons from coconut husk using response surface methodology,Chem. Eng. J. 137 (2008) 462–470.

[21] L. Yin, Y. Chen, D. Li, X. Zhao, B. Hou, B. Cao, 3-Dimensionalhierarchical porous activated carbon derived from coconut fibers with high-rate performance for symmetric supercapacitors, Mater. Des. 111 (2016)44–50.

[22] Y. Cao, K. Wang, X. Wang, Z. Gu, Q. Fan, W. Gibbons, J.D. Hoefelmeyer, P.R.Kharel, M. Shrestha, Hierarchical porous activated carbon for supercapacitorderived from corn stalk core by potassium hydroxide activation, Electrochim.Acta 212 (2016) 839–847.

[23] G. Ma, Q. Yang, K. Sun, H. Peng, F. Ran, X. Zhao, et al., Nitrogen-doped porouscarbon derived from biomass waste for high-performance supercapacitor,Bioresour. Technol. 197 (2015) 137–142.

[24] L. Ren, J. Zhang, Y. Li, C. Zhang, Preparation and evaluation of cattail fiber-based activated carbon for 2,4-dichlorophenol and 2,4,6-trichlorophenolremoval, Chem. Eng. J. 168 (2011) 553–561.

[25] Q. Shi, J. Zhang, C. Zhang, C. Li, B. Zhang, W. Hu, et al., Preparation of activatedcarbon from cattail and its application for dyes removal, J. Environ. Sci. 22(2010) 91–97.

[26] X. Chen, F. Liu, B. Liu, L. Tian, W. Hu, Q. Xia, A novel route to graphite-likecarbon supporting SnO2 with high electron transfer and photocatalyticactivity, J. Hazard. Mater. 287 (2015) 126–132.

[27] Z. Fan, D. Qi, Y. Xiao, J. Yan, T. Wei, One-step synthesis of biomass-derivedporous carbon foam for high performance supercapacitors, Mater. Lett. 101(2013) 29–32.

[28] I.A. Tan, A.L. Ahmad, B.H. Hameed, Adsorption isotherms, kinetics,thermodynamics and desorption studies of 2,4,6-trichlorophenol on oil palmempty fruit bunch-based activated carbon, J. Hazard. Mater. 164 (2009) 473–482.

[29] P. Wang, Q. Ma, D. Hu, L. Wang, Removal of Reactive Blue 21 onto magneticchitosan microparticles functionalized with polyamidoamine dendrimers,React. Funct. Polym. 91–92 (2015) 43–50.

[30] Y. Onal, C. Akmil-Basar, D. Eren, C. Sarici-Ozdemir, T. Depci, Adsorptionkinetics of malachite green onto activated carbon prepared from Tuncbileklignite, J. Hazard. Mater. 128 (2006) 150–157.

[31] X. Sun, P. Cheng, H. Wang, H. Xu, L. Dang, Z. Liu, et al., Activation of grapheneaerogel with phosphoric acid for enhanced electrocapacitive performance,Carbon 92 (2015) 1–10.

[32] J. Zhang, Y. Li, C. Zhang, Y. Jing, Adsorption of malachite green from aqueoussolution onto carbon prepared from Arundo donax root, J. Hazard. Mater. 150(2008) 774–782.

[33] S. Nethaji, A. Sivasamy, G. Thennarasu, S. Saravanan, Adsorption of MalachiteGreen dye onto activated carbon derived from Borassus aethiopum flowerbiomass, J. Hazard. Mater. 181 (2010) 271–280.

[34] T. Santhi, S. Manonmani, T. Smitha, Removal of malachite green from aqueoussolution by activated carbon prepared from the epicarp of Ricinus communisby adsorption, J. Hazard. Mater. 179 (2010) 178–186.

[35] K. Porkodi, K. Vasanth Kumar, Equilibrium, kinetics and mechanism modelingand simulation of basic and acid dyes sorption onto jute fiber carbon: eosinyellow, malachite green and crystal violet single component systems, J.Hazard. Mater. 143 (2007) 311–327.

[36] Y.S. Ho, G. McKay, Sorption of dye from aqueous solution by peat, Chem. Eng. J.70 (1998) 115–124.

[37] H. Zhuo, Y. Hu, X. Tong, L. Zhong, X. Peng, R. Sun, Sustainable hierarchicalporous carbon aerogel from cellulose for high-performance supercapacitor andCO2 capture, Ind. Crops Prod. 87 (2016) 229–235.

502 M. Yu et al. / Chemical Engineering Journal 317 (2017) 493–502

[38] F. Ran, K. Shen, Y. Tan, B. Peng, S. Chen, W. Zhang, et al., Activated hierarchicalporous carbon as electrode membrane accommodated with triblockcopolymer for supercapacitors, J. Membr. Sci. 514 (2016) 366–375.

[39] V. Subramanian, C. Luo, A.M. Stephan, K.S. Nahm, S. Thomas, B. Wei,Supercapacitors from activated carbon derived from banana fibers, J. Phys.Chem. C 111 (2007) 7527–7531.

[40] W.H. Qu, Y.Y. Xu, A.H. Lu, X.Q. Zhang, W.C. Li, Converting biowaste corncobresidue into high value added porous carbon for supercapacitor electrodes,Bioresour. Technol. 189 (2015) 285–291.

[41] X.L. Wu, T. Wen, H.L. Guo, et al., Biomass-derived sponge-like carbonaceoushydrogels and aerogels for supercapacitors, ACS Nano 7 (2013) 3589–3597.