Comparison of a homemade cocoa shell activated carbon with commercial activated carbon for the removal of reactive violet 5 dye from aqueous solutions

Accepted Manuscript Comparison of a homemade cocoa shell activated carbon with commercial activated carbon for the removal of reactive violet 5 dye from aqueous solutions Marielen C. Ribas, Matthew A. Adebayo, Lizie D.T. Prola, Eder C. Lima, Renato Cataluña, Liliana A. Feris, M.J. Puchana-Rosero, Fernando M. Machado, Flávio A. Pavan, Tatiana Calvete PII: DOI: Reference:

S1385-8947(14)00332-5 http://dx.doi.org/10.1016/j.cej.2014.03.054 CEJ 11913

To appear in:

Chemical Engineering Journal

Received Date: Revised Date: Accepted Date:

23 January 2014 12 March 2014 15 March 2014

Please cite this article as: M.C. Ribas, M.A. Adebayo, L.D.T. Prola, E.C. Lima, R. Cataluña, L.A. Feris, M.J. Puchana-Rosero, F.M. Machado, F.A. Pavan, T. Calvete, Comparison of a homemade cocoa shell activated carbon with commercial activated carbon for the removal of reactive violet 5 dye from aqueous solutions, Chemical Engineering Journal (2014), doi: http://dx.doi.org/10.1016/j.cej.2014.03.054

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

Comparison of a homemade cocoa shell activated carbon with commercial activated carbon for the removal of reactive violet 5 dye from aqueous solutions.

Marielen C. Ribas1, Matthew A. Adebayo2,3, Lizie D.T. Prola2, Eder C. Lima2*, Renato Cataluña2, Liliana A. Feris1, M.J. Puchana-Rosero2, Fernando M. Machado4, Flávio A. Pavan5, Tatiana Calvete1,6.

1-

Department of Chemical Engineering, Federal University of Rio Grande do Sul (UFRGS),

Rua Engenheiro Luiz Englert, Building 12204, ZIP 90040-040, Porto Alegre, RS, Brazil. 2-

Institute of Chemistry, Federal University of Rio Grande do Sul (UFRGS), Av. Bento

Gonçalves 9500, Postal Box 15003, ZIP 91501-970, Porto Alegre, RS, Brazil. 3-

Department of Chemical Sciences, Ajayi Crowther University, PMB 1066, Oyo, Oyo State,

Nigeria. 4-

Universitary Center Franciscano (UNIFRA), R. dos Andradas 1614, ZIP 97010-032, Santa

Maria, RS, Brazil. 5-

Federal University of Pampa, UNIPAMPA, Bagé, RS, Brazil.

6-

Universitary Center La Salle (UNILASALLE), Av. Victor Barreto 2288, 92010-000, Canoas-

RS, Brazil.

*

Corresponding author: FAX + 55 (51) 3308 7304; Phone: +55 (51) 3308 7175; e-mail:

[email protected] or [email protected]

2

Abstract A novel homemade furnace setup for preparation of chemically activated carbon was proposed in a quest for industrial wastewater treatment. Cocoa shell was initially mixed with inorganic components (red mud + lime + KOH + Al(NO3)3 + Na2SO4) and water to form a paste. The paste was placed in a mould cylinder, dried at a room temperature, and then the material was placed in a stainless steel reactor and heated up to 1073 K under inert atmosphere. Three carbon adsorbents with inorganic : organic ratio of 1.0 (CC-1.0), 1.5 (CC1.5) and 2.0 (CC-2.0) were prepared. The adsorbents were acidified with a 6 mol L-1 HCl under reflux (24 h) to obtain corresponding ACC-1.0, ACC-1.5 and ACC-2.0. The chemical activation process was completed by leaching the inorganic components from the carbonaceous matrix through acidification. ACC-1.0 exhibited highest sorption capacity compared with the other two adsorbents. CC-1.0 and ACC-1.0 were characterised using FTIR, SEM, N2 adsorption/desorption curves and X-ray diffraction. A well-known commercially activated carbon (CAC) was used to compare the sorption capacity of ACC1.0. The ACC-1.0 and CAC adsorbents were used for adsorption of reactive violet 5 (RV-5) textile dye from aqueous solutions. The equilibrium times of 45 and 150 minutes were observed for ACC-1.0 and CAC, respectively, at optimum pH 2.0. General order kinetic model best described the adsorption process than pseudo first-order and pseudo-second order kinetic models. Liu isotherm model gave the best fit of the equilibrium data at all experimental temperatures. The maximum amounts of RV-5 dye adsorbed at 298 K were 603.3 (ACC-1.0) and 517.1 mg g-1 (CAC). The adsorbents were tested on two simulated dyehouse effluents. ACC-1.0 is effectively capable of decolourising industrial textile effluents.

Keywords: adsorption; industrial effluents; activated carbons; general order kinetic; nonlinear isotherms.

3

1. Introduction

Synthetic dyes are used by nearly all the textile industries as colourants. These industries use reactive dyes, which constitute about 30% of dyes used in the textile industries [1,2]. Coloured wastewater emanates from textile industries because dyes (10 to 60%) are lost during dyeing process [3]. The effluent from textile industries is hazardous to the living organisms; it lowers the rate of photosynthesis of aqueous flora by impeding light penetration [4-6]. Some of these dyes are toxic, mutagenic and cancer inducing [7-9], which invariably have negative impacts on the environment [4,10]. Reactive dyes are characterised with complex aromatic molecular structure, which makes it difficult to treat the effluents containing reactive dyes [11,12]. Reactive dyes are, therefore, stable and non-biodegradable [13]. As a result of imposition strict regulations [3], effluents emanating from the fabric industries must be treated before being discharged into the environment [14,15]. This has led to quest for nature friendly techniques for removal of dyes from wastewater [16,17]. To remove synthetic dyes from wastewaters, adsorption is a useful one technology [17,18]. Adsorption is now a predominant technique for effluent treatment because of its simplicity, effectiveness and availability of low cost adsorbents [19-24]. It is a process that transfers pollutants from the effluent to a solid phase thereby reducing the bioavailability of toxic species to living organisms [25]. The treated effluent can be regarded not harmful and then discharged into the environment [25,26]. For industrial processes that requires water of low or moderate purity, the treated water could also be used. Adsorbents can be regenerated after adsorption process, stored and reused [25,26]. Activated carbons possess pore structures that enhance high adsorption. This characteristic feature makes activated carbons one of the materials employed for removal of organic compounds from wastewaters [27-30]. Ability of activated carbons to remove pollutants from aqueous solutions depends on the nature of organic material used for the preparation of the activated carbon [27,28] and experimental conditions in the activation processes [28,29]. Among the reagents widely used for chemical activation of activated

4

carbons are H3PO4, ZnCl2, KOH, NaOH, H2SO4, and K2CO3 [28-30]. In a chemical activation process, the raw materials and inorganic activating agents are mixed in aqueous medium. The resulting material is subsequently dried in an oven and carbonised (673 – 1073 K). Inorganic material (activating agent) can be excluded using water or solutions of acid. However, this procedure did not provide a homogeneous distribution of the inorganics in all the organic carbonaceous materials because of the problems related with the drying of impregnated organic material [28-30]. Cocoa shell is a residue that corresponds to 80% of the dry weight of all cocoa fruit [31]. The annual production of cocoa in Brazil is about 250,000 ton/year that would correspond to approximately 200,000 ton cocoa shell per year. Therefore, the Brazilian industry of chocolate generate large quantities of cocoa shell, most of which are discarded in aquifers and soil environments. The decomposition of these residues leads generating various chemicals and microorganisms that can contaminate the environment in an uncontrolled manner. Therefore it is necessary to find a use for this waste in order to avoid environmental problems, with sustainable solution. To the best of our knowledge, this is the first paper that reports the chemical activation of an activated carbon using a mixture of inorganic components (28.6% red mud, 28.6% lime, 14.3% KOH, 14.3% Al(NO3)3 and 14.2% Na2SO4) and powdered cocoa shell with inorganic : organic proportions of 1.0, 1.5 and 2.0. In the preparation, water was added to allow formation of a paste. The paste was dried at room temperature for 24 h and thereafter placed in a stainless steel reactor, where it was heated up to 1073 K under inert atmosphere. The carbonised materials (CC1.0, CC-1.5, CC-2.0) were acidified with a 6.0 mol L-1 of HCl under reflux to obtain chemically activated cocoa shell carbons (ACC1.0, ACC-1.5, ACC2.0). Preliminary experiments in this study revealed that ACC-1.0 has best adsorption capacity for reactive violet 5 (RV-5) textile dye from aqueous solutions than ACC-1.5 and ACC-2.0. Commercially available activated carbon (CAC), an established adsorbent for removal of toxic species from aqueous solutions, was used to verify the sorption capacity of ACC-1.0. ACC-1.0 and CAC were utilised for the adsorption of RV-5 dye from aqueous

5

solutions. Influence of pH of dye solutions, time and temperature on adsorption capacity were investigated. The adsorbents were used for treatment of different simulated dye-house effluents.

2. Materials and Methods

2.1 Solutions and reagents Deionised water was utilised for preparation of solutions. The textile dye, C.I. Reactive Violet 5 dye (RV-5, C.I. 18097; CAS: 12226-38-9; C20H16N3O15S4Na3, 735.58 g mol-1, λmax = 545 nm see Supplementary Fig. 1) at 85% purity, was purchased from SigmaAldrich (St. Louis, M.O. USA) and was used without purification. The RV-5 dye contains one sulphato-ethyl-sulphone and two sulphonate groups. The functional groups of the dye are negatively charged. Their pKas are lower than zero [32]. RV-5 dye was used in this study, because it is largely employed in the Brazilian industries for dye clothing. A stock solution (5.00 g L-1) of RV-5 dye was prepared by weighing a calculated amount of the dye and dissolving in deionised water. The stock solution was diluted to obtain various working solutions. The pH of the solutions was adjusted using the Schott Lab 850 set pH meter with a 0.10 mol L-1 NaOH and/or a 0.10 mol L-1 HCl.

2.2 Adsorbents

Commercial activated carbon (CAC) supplied by Merck (325-400 mesh) was used to compare the sorption performance of cocoa shell activated carbon used in this work. The activated carbon adsorbent was prepared using the following procedures (Fig. 1): a 70.0 g of powdered cocoa shell, a 70.0 g of inorganic components (28.6% red mud, 28.6% lime, 14.3% KOH, 14.3% Al(NO3)3 and 14.2% Na2SO4) and 50.0 mL of water were thoroughly mixed to obtain a homogeneous paste. The resulting paste was placed in a 4.8 by

6

14.0 cm mould cylinder (253.34 cm3), wet-shaped and dried at room temperature for 24 h. The dried cylinder (without the mould) was subsequently placed in a stainless steel reactor (Fig. 1). The reactor allowed uniform gas distribution and homogeneous gas exchange rate (argon at 100 mL min-1) to produce homogeneous carbon adsorbents. The reactor was thereafter heated in the tubular furnace at 20 K min-1 up to 1073 K for 30 minutes. The adsorbent was later cooled down to room temperature under argon (25 mL min-1), milled, sieved to a particle size ≤ 150 µm and stored in an airtight container until use. This carbon adsorbent was named CC-1.0. Adsorbents with inorganic : organic proportion of 1.5 (CC-1.5) and 2.0 (CC-2.0) were also prepared using the similar procedure. For comparison purpose, a carbon adsorbent (CC-0) without inorganic components was prepared as well. To complete the chemical activation of the carbon adsorbent, a 10.0 g of CC-1.0 and 200 mL of 6 mol L-1 HCl were placed in a 500 mL boiling flask, the mixture was stirred on a magnetic stirrer and reflux for 2 h (343 K). The slurry was cooled down and filtered under vacuum using 0.45 µm membrane in a polycarbonate Sartorius system. The solid material, after extensive washing with aqueous solution (pH 2.0), was oven dried at 383 K for 5 h, milled to particle sizes ≤ 150 µm and stored properly until use. The adsorbent was named acid treated activated carbon (ACC-1.0). Acidification removed the inorganics from the carbon adsorbent.

Insert Fig 1. 2.3 Characterization of CC-1.0 and ACC-1.0

To verify the effects of the chemical activation on the cocoa shell activated carbon, CC-1.0 precursor was used to check the improvements in the textural properties, thermal stability and morphology. Surface morphologies of CC-1.0 and ACC-1.0 were performed using Scanning electron microscopy (SEM) (Hitachi, Tabletop microscope model TM3000, Tokyo, Japan) [33].

7

The CC-1.0 and ACC-1.0 were characterised using Fourier Transform Infra-Red Spectroscopy (FTIR) (Shimadzu Spectrometer, IR Prestige 21, Kyoto, Japan). For this analysis, adsorbents and KBr were oven dried at 393 K for 8 h, stored in capped flasks and kept in a desiccator prior analysis. The spectra were obtained with a resolution of 4 cm-1 with 100 cumulative scans [34]. The N2 adsorption-desorption isotherms of CC-1.0, CC-1.5, CC-2.0, ACC-0, ACC-1.0, ACC-1.5 and ACC-2.0 were done at liquid nitrogen boiling point (77 K) using a surface analyser (Micrometrics Instrument, TriStar II 3020). Before the analysis, the adsorbents were degassed at 453 K for 12 h under vacuum. The specific surface areas and pore size distribution of the adsorbents were evaluated using BET (Brunauer, Emmett and Teller) multipoint technique [35] and BJH (Barret, Joyner and Halenda) [33], respectively. Thermogravimetric (TGA) and derivative thermogravimetric (DTG) curves of CC-1.0 and ACC-1.0 were obtained on a TA Instruments model SDT Q600 (New Castle, USA) with a heating rate of 283 K min-1 at 100 mL min-1 of synthetic air flow (White Martins, Canoas, RS, Brazil). Temperature was varied the from 293 K to 1273 K (acquisition time of 1 point per 5 s) with a mass of 10.00 – 15.00 mg of solid [27]. X-ray diffractions (XRD) (Philips X’pert MPD diffractometer, Netherlands) at 40 kV and 40 mA with Cu Kα radiation (λ = 1.5406 Å) of CC-1.0 and ACC-1.0 were determined. Measurements were done with scanning step width of 0.01o and time of 1 s, over the 2θ range of 10 – 70o [10]. X-ray fluorescence (Shimadzu XRF1800 X-ray Fluorescence Spectrometer, Japan) was used to determine the chemical composition of CC-1.0 and ACC1.0. Procedure for determination of the point of zero charge (pHpzc): a 20.00 mL of 0.050 mol L-1 NaCl solution was added to various 50 mL Falcon tubes containing 50.0 mg of the adsorbent, and were covered immediately. The pH (pHi) values of the solutions were adjusted from 1.0 to 9.0 using a 0.10 mol L-1 of HCl and 0.10 mol L-1 NaOH. The suspensions were agitated and equilibrated in a thermostatic shaker at 298 K for 48 h. The suspensions were centrifuged at 15,000 rpm for 10 min. The pHi of the solutions without adsorbent and

8

pHf of the supernatant after contact with the adsorbents were recorded. The value of pHpzc is the point where the plot of ∆pH (pHf - pHi) versus pHi crosses a line equal to zero [26].

2.4 Batch adsorption studies

The adsorption capacity of proposed ACC-1.0 was compared with a well-established CAC. A 50.0 mg of adsorbent and a 20.0 mL of RV-5 dye solution (300.00 to 1500.0 mg L-1) were placed in various 50.0 mL Falcon tubes at different pH values (2.0 to 10.0). The mixtures were agitated between 5.0 to 480 min inside a thermostatic shaker (Oxylab, São Leopoldo, Brazil) from 298 to 323 K. The mixtures were centrifuged (Unicen M Herolab centrifuge, Stuttgart, Germany) for 5 min at 15,000 rpm after the batch adsorption experiments to separate the adsorbents from the dye solutions. When necessary, aliquots of the supernatant were diluted with deionised water at pH 2.0 before spectroscopic measurement. UV/visible spectrophotometer (T90+ UV-VIS spectrophotometer, PG Instruments, London, United Kingdom) was used to quantify the residual of RV-5 dye in solution after adsorption at a maximum wavelength of 545 nm.

2.5. Quality assurance and statistical evaluation of models

All the experiments were triplicated to ensure reproducibility, reliability, and accuracy of the experimental data. The relative standard deviations of all measurements were < 5% [36]. Blanks were run in parallel and corrected when needed [37]. The solutions of RV-5 dye were stored in glass bottles, which were precleaned by immersing in 1.4 mol L-1 HNO3 for 24 h [38], rinsing with deionised water, drying and storing them in a flow-hood. Concentrations ranges from 10.00 to 200.0 mg L-1 standard dye solutions were used for calibration, in parallel with a blank (aqueous solution, a function of pH of the dye solution

9

being measured). The linear analytical calibration of the curve was performed on the UVWin software of the T90+ PG Instruments spectrophotometer. All the analytical measurements were carried out in triplicate, and the precision of the standards were better than 3 % (n=3). The RV-5 dye has detection limit of 0.17 mg L-1 with a signal/noise ratio of 3 [39]. A 60.0 mg L-1 standard dye solution was used as a quality control after every five measurements to ensure the accuracies of the RV-5 dye solutions [35]. Kinetic and equilibrium data were fitted using nonlinear methods with successive interactions calculated by the Levenberg–Marquardt method. Interactions were also evaluated with the aid of the Simplex method, based on the nonlinear fitting facilities of the Microcal Origin 9.0 software. A determination coefficient (R2), an adjusted determination coefficient (R2adj) and an error function (Ferror) were jointly used to evaluate the suitability of the models [40]. A Ferror is defined as a measure of the differences between the theoretical and experimental amounts of dye adsorbed. The R2, R2adj and Ferror are depicted in Equations 3, 4 and 5, respectively.

n i

(q

∑

∑

2

i,exp

-qi,exp ) -

∑

 R2=   

n i

(q

i,exp

n i

(q

i, exp

-qi,exp )

2

2 -qi, model )    

 n-1  2 Radj =1- (1-R 2 ) .   n-p-1  2  1  n Ferror =   .∑ ( qi, exp -qi, model )  n-p  i

(3)

(4)

(5)

In these equations, qi, model represents individual theoretical q value predicted by the model, qi,exp represents individual experimental q value, qexp is the average of experimental q, n represents the number of experiments while p represents the number of parameters in the fitting model [40].

2.6. Kinetic models

10

Pseudo-first order, Pseudo second-order, General order kinetic model and intraparticle diffusion model, as shown in Equations 6, 7, 8 and 9, respectively, were used to explain the kinetic studies.

q t = q e [1-exp(-k 1 ⋅ t) ]

(6)

qe [ k2 (qe ) ⋅ t+1]

(7)

qt =qe -

qt =qe -

(8)

qe  k N (qe )

n-1

⋅ t ⋅ (n-1)+1 

1 1-n

qt = kid t + C

(9)

For details of these models [41-45], see the Supplementary material†

2.7 Equilibrium models

Langmuir, Freundlich and Liu models as shown in Equations 10, 11 and 12, respectively, were used in this work.

qe =

Qmax ⋅ K L ⋅ Ce 1 + K L ⋅ Ce

qe = KF ⋅ Ce1 nF qe =

Qmax ⋅ ( K g ⋅ Ce )nL

1 + ( K g ⋅ Ce )nL

Supplementary material [46-48] contains the details of these models

2.8. Simulated dye-house effluents †

Supplementary Material should be downloaded

(10)

(11) (12)

11

Two simulated dye-house effluents that mimic industrial textile effluents were prepared at pH 2.0. The synthetic dye-house effluents contained five typical dyes commonly used as fabric colourants together with auxiliary chemicals. It has been established that between 10-60% [3] of synthetic dyes and virtually all the auxiliaries remain in the spent dye bath, and their compositions undergo a 5-30-fold dilution during the subsequent washing and rinsing stages [2,5,21,25,26,40,43]. The representative dyes and auxiliaries as well as their concentrations are presented in Table 1

Insert Table 1

3. Results and discussions

3.1. Preparation of cocoa shell activated carbon and preliminary experiments of adsorption

In the preparation of carbon adsorbents, lime (CaCO3 + Ca(OH)2 + CaO) is used as one of the inorganic components. Inclusion of lime is to avoid impregnation of the material with aqueous solution [28-30]. Lime holds all the paste-forming solid components together (see Fig. 1). Red mud as well as lime, KOH, Al(NO3)3, and Na2SO4 were the inorganics used as activating agents [49,50]. Na2SO4 [28], KOH [49] and aluminium salts [50] have previously been employed as activating agents for preparation of activated carbons [49] and red mud as adsorbent [51]. The dried cylinders were inserted in the stainless reactor and pyrolysed at 1073 K under inert atmosphere (see Fig. 1). The reactor was vertically fixed to ensure delivering of the tar (by gravity) into the recipient vessel. In this arrangement, the pores of carbon adsorbent were prevented from being blocked by the tar as reported [28]. The carbon material produced at this step was labelled as CC-1.0, CC-1.5, CC-2.0 for inorganic : organic ratios of 1.0, 1.5, and 2.0, respectively. A carbon material without the inorganic components was also prepared (CC-0) to investigate the effects of inorganic components in the carbon adsorbent.

12

The CC-0, CC-1.0, CC-1.5 and CC-2.0 were treated with a 6 mol L-1 HCl to leach the inorganic components from the carbonised adsorbents. The inorganic component of the carbon material can hydrolyse the organic material precursor and mediate the release of some organic components during the pyrolysis, thereby weakening the carbon material particles, which expands. The inorganic component, on the contrary, takes a volume that prevents the contraction of the particle during the pyrolysis. This phenomenon improves the porosity of the carbon material as the inorganic components are sequestrated during acidification [28]. The CC-1.0, CC-1,5, CC-2.0, ACC-0, ACC-1.0, ACC-1.5 and ACC-2.0 were used to remove 400 mg L-1 of RV-5 dye from aqueous solutions with 1.0 h agitation. The percentages of removal are furnished in Supplementary Table 1. The ACC-0 has the lowest percentage of removal among all adsorbents tested. The non-inclusion of inorganic components in the carbonisation of the powdered cocoa shell generated large quantity of tar-oil, which blocks the pores of the carbon material. It was observed that is necessary to leach-out the inorganic components from the carbonaceous material, in order to increase the percentage of removal of dye, since all ACC adsorbents present higher sorption capacity when compared with the correspondent CC adsorbent. ACC-1.0 was used for the rest of the study since it gave highest RV-5 dye adsorption during preliminary study. For comparative purpose, a commercial activated carbon (CAC) from Merck, in addition to ACC-1.0, was used for RV-5 dye adsorption.

3.2 Characterisation of CC-1.0 and ACC1.0

In order to verify the effects of the chemical activation of the cocoa shell activated carbon the CC-1.0 precursor was used to check the improvements in the textural properties, in the thermal stability and morphology. Taking into account that ACC-1.0 presented the best adsorption capacity for removal of RV-5 from aqueous solutions, it was just chosen its precursor of the activated carbon (CC-

13

1.0) and the activated carbon (ACC-1.0) for characterization purposes. However, the unique observation that could be relevant to show would be the surface area, the average pore radius and total pore volume of all tested adsorbents. These are presented on Supplementary Table 1. It can be observed, that the higher surface area and higher total pore volume of the adsorbent leads to a higher percentage of removal of RV-5 dye from aqueous solution. Comparison of the textural properties of non-leached carbon materials (CC-1.0, CC-1.5, and CC-2.0) with the leached carbon adsorbents (ACC-1.0, ACC-1.5 and ACC-2.0) showed that acidification produced adsorbents with better textural characteristics for adsorption of dye, as already reported in the literature [52]. The surface area and total pore volume of ACC-1.0, ACC-1.5 and ACC-2.0 were increased by 25.71 and 11.86-fold (ACC-1.0), 27.16 and 11.88-fold (ACC-1.5) and 27.05 and 12.55-fold (ACC-2.0) after acidification of CC-1.0, CC-1.5 and CC-2.0, respectively. Supplementary Figs. 2A and 2B show the respective XRD patterns of CC-1.0 and ACC-1.0. The peaks of CC-1.0 (Supplementary Fig. 2A) are assigned as follow: potassium sodium aluminium oxide carbonate hydrate (KNa1.2Al2O2(CO3)2. (H2O)2.16; JCPDS Card 00034-0544); sodium aluminium carbonate nitrite silicate (Na8[AlSiO4]6(NO2)(CO3)0.5; JCPDS Card 00-048-0443); iron (Fe; JCPDS Card 00-006-0696); magnetite (Fe3O4; JCPDS Card 00001-1111); titanium oxide (TiO2; JCPDS Card 00-049-1433); silicon oxide (SiO2; JCPDS Card 00-033-1161); sodium nitrate (NaNO3; JCPDS Card 00-001-0840), calcium sulphate hydrate (CaSO4.(H2O)0.67; JCPDS Card 00-036-0617). This XRD pattern is compatible with the inorganic components present in the red mud [53], and other activating agents in CC-1.0. The XRD pattern of ACC-1.0 (Supplementary Fig. 2B), however, is a typical of amorphous material with a wide band between 13° to 35° that corresponds to amorphous carbon, and three crystalline phases that remained in the ACC-1.0 after the leaching process with 6.0 mol L-1 HCl. These mineralogical phases are : aluminium titanium silicate (Al 4Ti2SiO12; JCPDS Card 00-022-0502); titanium oxide (TiO2; JCPDS Card 00-004-0551); and silicon oxide (SiO2; JCPDS Card 00-033-1161). As shown by XRD, the mineralogical phases conform to the chemical analysis of X-ray fluorescence of CC-1.0 and ACC-1.0 (Supplementary Table 2).

14

All the peaks in ACC-1.0 have smaller intensity unlike those of CC-1.0. This is an indication that acidification was efficient to sequestrate the inorganics from ACC-1.0. Similarly, XRD data agreed with the textural properties previously discussed. The leaching of inorganic components led to a carbon material with higher pores structure and superficial area, and low crystallinity. The thermo gravimetric profiles of CC-1.0 and ACC-1.0 are shown in Supplementary Fig. 3. There exist many peaks in the DTG curve of CC-1.0: 52°C and 88°C, the elimination of solvent; 254°, 352° and 431°C, decomposition of the organic contents of CC-1.0; and a very small peak at 669°C. Using the DTG, the TG curves can be divided into three portions. The first part from 20°C to 140°C, a mass loss of 2.5% was assigned to the loss of water. In the second part, a higher mass loss (22.7%) that could be attributed to decomposition of carbonaceous matrix was observed from 243°C to 472°C [27]. In the last part, a mass loss that could be attributed to residual carbon skeleton decomposition from 472°C to 1000°C (3.8%) was observed. However, at 1000°C, a residual mass of about 65.0% was also observed. This mass was assigned to the inorganics of the precursor of activated carbon (CC-1.0) and undecomposed organic compounds. The ACC-1.0 showed a thermo gravimetric profile different from that of CC-1.0 (see Supplementary Fig. 3). The final residual mass is 23.5%, an indication that the most of inorganics have been eliminated during acidification, which is consistent with the XRD results discussed earlier. Two decomposition regions (43°, and 459°C) were observed on DTG curves of ACC-1.0. Using DTG curves, there are four regions in the TG curves. There was a mass loss of 14.8% from 20° to 90°C, which could be linked to water molecules. ACC-1.0 has higher humidity than CC-1.0 because of the treatment with HCl solution. Similarly, a mass loss of 5.6% from 90°C to 378°C, which could be assigned to water molecules that were captured in the carbonaceous matrix, was observed. A mass loss of 53.8% from 378°C to 527°C, which could be attributed to the decomposition of carbon skeleton, was also observed. Finally, there was an additional mass loss of 2.3% from 527°C to 1000°C.

15

Supplementary Fig. 4 shows the SEM images of CC-1.0 and ACC-1.0. The two adsorbents lost their fibrous properties during pyrolysis [19,23,26,34]. The roughness of the carbon materials are visible (see Supplementary Fig. 4). The major difference is that the granules of CC-1.0 are more aggregated, which result in particles of lower dimension. The granules of ACC-1.0, however, are of higher dimension with higher void space among the particles compatible with the leaching of inorganics during the acidification, which is consistent with the results of TGA, XRD, BET and BJH results. For identification of functional groups in CC-1.0 and ACC-1.0 and the groups responsible for adsorption of RV-5 dye, FTIR technique was used. The FTIR spectra of the adsorbents were recorded in the range 4000-400 cm-1 (Table 2). Band assignments of the functional groups in CC-1.0 and ACC-1.0 are presented in Table 2 [16,21,23,26,40]. The FTIR spectra of CC-1.0 is slightly different from that of ACC-1.0 (see Supplementary Fig. 5). The CC-1.0 has more vibrational bands than ACC-1.0. The Si-O stretch of silicates is conspicuous in CC-1.0 at 995 cm-1 unlike in ACC-1.0 (at 1094 cm-1). Other noticeable groups in CC-1.0 are O-H (alcohols, phenols), aromatic rings, C-O (alcohols), Si-O (silicates) and CH (aromatics, aliphatic chains). The prominent groups in ACC-1.0 are O-H (alcohols, phenols), aromatic rings, C-O (phenols, alcohols), Si-O (silicates) and C-H (aromatics, aliphatic chains).

Insert Table 2 3.2. pH dependence studies

The pH of the adsorbate solution is one of the contributory factors affecting adsorption of dye on adsorbent [5,10,14]. Dyes have different optimum pHs depending on the type of adsorbent under investigation. Because of this assertion, pH dependence studies of RV-5 dye removal from aqueous solutions (400 mg L-1) using ACC-1.0 and CAC were carried out in the pH range of 2 and 10 (see Fig. 2). The percentage of RV-5 dye removal decreased from 96.2% (pH 2.0) to 60.2% (pH 10.0) for CAC. Similar pattern was observed

16

for ACC-1.0, percentage removal decreased from 97.7% (pH 2.0) to 20.0% (pH 10.0). The two adsorbents behaved similar within the pH range of 2.0 to 3.0. However, from pH 3.5 to pH 10.0, ACC-1.0 exhibited a steeper decrease in the percentage of dye removal unlike that of CAC. The striking difference of the CAC and ACC-1.0 with respect to the pH of the dye solutions is attributed to the pHpzc of the adsorbents. The pHpzc values of CAC and ACC-1.0 are 7.31 and 2.35, respectively (see Fig 2). Adsorbents with lower pH values than pHpzc will possess positive surface charges [26]. RV-5 dye is negatively charged in solutions [32], therefore, adsorption of RV-5 dye will take place when adsorbents have positive surface charge. The electrostatic interactions occur at pH < 7.31 and < 2.35 for CAC and ACC-1.0, respectively. The lower the pH value from the pHpzc, the more positive the surface charge of the adsorbent [23,26]. For continuation of our experiments, initial pH of adsorbate solution was fixed at 2.0. The final pH values of adsorbate solutions after adsorption were quantified. These values varied from 5.0 to 5.8 and 2.0 to 2.3 for CAC and ACC-1.0, respectively. These observations concurred with the pHpzc values of the adsorbents. The ∆pH (pHfinal – pHinitial) for CAC with a neutral solid surface ranges (pHpzc 7.31) from 3.0 to 3.8 while this value is only 0.3 for ACC-1.0 with acidic solid surface (pHpzc 2.35). The closer the initial pH of adsorbate solution to the pHpzc of the adsorbent, lower the ∆pH during the adsorption process.

Insert Fig 2.

3.3. Kinetic studies

Nonlinear pseudo-first order, pseudo-second order and general order kinetic models were used to probe the kinetic of adsorption of RV-5 dye onto ACC-1.0 and CAC. Fig. 3 shows the kinetic plots of while Table 3 shows the fitting parameters of the kinetic models. To ascertain the correctness of these fits using nonlinear kinetic models, the Ferror was used. The lower the Ferror, the smaller the difference between the theoretical q and experimental q

17

values [2,5,19,21,40] (see Equation 5). It has been reported [54] that the best fit of the data is a function of the number of parameters in nonlinear equations. Due to this fact, the number of parameters in nonlinear models should be taking into consideration when evaluating the Ferror. This prompted us to take the number of fitting parameters (p term of Equation 5) into consideration when evaluating the Ferror. The Ferror of the minimum value was used to divide Ferror of each model (Ferror ratio) for comparison of the different kinetic models. General order kinetic model has the lowest Ferror values. Lowest Ferror values of this model is an indication that the values of experimental q and theoretical q are closer. Ferror ratios of the pseudo-first order kinetic model varied from 3.60 to 4.33 (ACC-1.0) and 2.86 to 2.96 (CAC). Ferror ratios of the pseudo-second order model varied from 4.37 to 6.28 (ACC-1.0) and 6.12 to 6.45 (CAC). The general order kinetic model, with lowest Ferror ratio values, suitably explained the adsorption process of RV-5 dye onto ACC-1.0 and CAC. Insert Fig 3 Insert Table 3

The general order kinetic model assumed that the order of reaction must follow the same pattern of a chemical reaction, where the order of reaction is experimentally determined [21,25,26,43] and not being confined by a model. ACC-1.0 has faster kinetics of adsorption of RV-5 dye than CAC. The half-life (t1/2), the time taken to attain 50% of qe (amount adsorbed at the equilibrium), was determined by interpolation in the fitted curves. Table 3 shows these values. Taking into account that the general order kinetic model best fit our experimental data, only t1/2 values of this kinetic isotherm model have physical meaning. The t1/2 values of RV-5 dye adsorption on ACC-1.0 were 4.2-fold lower than those of CAC. To investigate the influence of mass transfer resistance on the binding of RV-5 dye onto the ACC-1.0 and CAC, the intra-particle diffusion model [45] was used (Table 3 and Supplementary Fig. 6). The slope of plots of

qt vs t (Supplementary Fig. 6) gave kid (mg

18

g-1 min-0.5), intra-particle diffusion constant. The plots have three linear sections for RV-5 dye adsorption onto ACC-1.0 and CAC, an indication that the adsorption processes required more than one adsorption rate [2,5,10,14]. There are three stages of adsorption process for both adsorbents, each stage is assigned to each linear portion of the plots in Supplementary Fig. 6. The fastest sorption stage, which is the first linear segment, is attributed to the process in which dye molecules diffuse to the surface of the adsorbents [2,5,10,14]. The second segment, which is a delayed process, is attributed to intra-particle diffusion [27]. The third segment, which is followed by the attainment of equilibrium, is regarded as diffusion through smaller pores [27]. A close examination of the first point of the third portion, minimum equilibrium contact-time for RV-5 dye adsorption is 45 min for ACC1.0 and 150 min for CAC. The difference in the minimum equilibrium time is consistent with the t1/2 values and initial sorption rates presented in Table 3. The contact time of 60 min (ACC-1.0) and 180 min (CAC) were used for adsorption of RV-5 dye. The contact time was increased to ensure that equilibrium is attained by the dye (even at higher concentrations) on the carbon adsorbents [25,26].

3.4. Equilibrium studies

At a constant temperature, an adsorption isotherm describes the relationship between the amount of adsorbate adsorbed by the adsorbent (qe) and the adsorbate concentration remaining in solution after equilibrium is reached (Ce). The parameters from the adsorption equilibrium models provide useful pieces of information on the surface properties, adsorption mechanism and interaction between the adsorbent and adsorbate. Langmuir [46], Freundlich [47] and Liu [48] isotherms were used to evaluate and interpret our isothermal data. The isotherms of adsorption were performed between 298 and 323 K using optimum experimental conditions previously emphasised (see Table 4 and Fig. 4). Fig. 4 shows the adsorption isotherms of RV-5 dye onto ACC-1.0 and CAC at 298 K. The Liu model suitably described adsorption of RV-5 dye onto ACC-1.0 and CAC at all temperatures going by the

19

Ferror values presented in Table 4. Computation and usefulness of Ferror ratio values have been discussed earlier. Insert Table 4 Insert Fig 4 Ferror ratio values of the Freundlich model varied from 11.33 to 16.53 (ACC-1.0) and 5.11 to 12.84 (CAC) while those of Langmuir model varied from 1.74 to 19.75 (ACC-1.0) and from 13.03 to 34.19 (CAC). Based on the Ferror ratio values, the equilibrium of adsorption of RV-5 dye on ACC-1.0 and CAC at the temperature range of 298 to 323 K best fit into Liu isotherm model. Maximum amounts of RV-5 dye removed (Qmax values) were 603.3 and 517.1 mg g-1 for ACC-1.0 and CAC, respectively. The Qmax values of ACC-1.0 were ca 17% higher than those of CAC at all the experimental temperatures. Moreover, the t1/2 of RV-5 dye adsorption on ACC-1.0 was 4.2-fold shorter than that of CAC, but ho of adsorption of RV-5 dye adsorption on ACC-1.0 was 5.9-fold higher than that of CAC, indicating that the kinetics of adsorption on ACC-1.0 are faster than those of CAC. These results show that ACC-1.0 adsorbent has higher sorption capacity and faster kinetics of adsorption of RV-5 dye than the widely known commercial activated carbon (CAC). Comparison of the adsorption capacities of the dyes with different adsorbents is difficult because dyes reported in the literature are commercial dyes. Similarly, a dye could have different commercial trade names depending on the manufacturer. Table 5 [1,2,15,17,18,20,21,23,24,25,27,55,56] shows a comparison of sorption capacities of different dyes using different adsorbents. The values were obtained using the best experimental conditions of each study. The ACC-1.0 adsorbent has an excellent sorption capacity compared to the adsorption capacities of different dyes on different adsorbents as shown in Table 5. Therefore, ACC-1.0 can be used as an alternative adsorbent for the removal of reactive dyes from aqueous solutions and effluents. Considering twenty-two adsorbents/adsorbates, the proposed ACC-1.0 has highest adsorption capacity, which is a confirmation that the acidified cocoa shell activated carbon is a good adsorbent. It is

20

noteworthy to emphasised that the second ranked adsorbent, single wall carbon nanotube (SWCNT) [56], has a maximum sorption capacity of 6.29% lower than ACC-1.0. This result ranks the maximum sorption capacity of ACC-1.0 with the same level of carbon nanotubes, however, carbon nanotubes [2,25,56] are much more expensive than ACC-1.0. Similarly, ACC-1.0 can be extensively applied in real wastewater treatment because of its cheapness, unlike carbon nanotubes. Insert Table 5.

3.5 Thermodynamics studies

The Gibb’s free energy change (∆G°, kJ mol-1), enthalpy change (∆H°, kJ mol-1) and entropy change (∆S°, J mol-1K-1) were evaluated with the aid of Equations 13, 14 and 15, respectively.

∆G 0 = ∆H 0 − T ∆S 0

(13)

∆G ° = - RTLn( K )

(14)

Equation 15 is obtained from Equation 13 and 14.

Ln(K)=

∆S° ∆H° 1 x R R T

(15)

where R is the universal gas constant (8.314 J K-1 mol-1); T is the absolute temperature (Kelvin); K represents that adsorption constants of the isotherm fits (Kg – Liu equilibrium constant, which must be converted to SI units, by using the molecular mass of the dye) obtained from the isotherm plots. It is known that different adsorption equilibrium constants (K) can be obtained from different isotherm models [2,16,19,21,25,26,27,40,42,43,56,57,58]. Thermodynamic parameters can also be calculated from the Liu equilibrium constant, Kg as already reported in the literature [2,19,21,26,56]. The ∆H° and ∆S° values were obtained from the respective slope and intercept of the plot of Ln(K) versus T-1 . The R2adj values of the linear plots are ca 0.99, indicating that the calculated ∆H° and ∆S° values are accurate. Table 6 shows the thermodynamic parameters.

21

The magnitude of enthalpy of RV-5 dyes on ACC-1.0 and CAC corresponds to physical adsorption process [59]. This can be used to specify the type of interaction between adsorbate and adsorbent. The physical sorption is <40 kJ mol-1 [59]. The negative values of enthalpy change (∆H°) indicate that interaction between RV-5 dye and adsorbents is exothermic. Negative values of ∆G° signify that the adsorption of RV-5 dye onto ACC-1.0 and CAC is spontaneous, and a favourable process. The positive values of ∆S° indicate that randomness at the solid/liquid interface increases. The water coordinated molecules are displaced by dye molecules hereby gain more translational entropy than what is lost by dye molecules leading to a more randomness in the interaction between the dye and adsorbents [60,61].

Insert Table 6 3.6 Simulated dye-house effluents

To evaluate application of ACC-1.0 and CAC for treatment of industrial textile effluents, two simulated dye-house effluents were utilised (see Table 1). The spectra of the treated and untreated effluents were recorded from 300 to 800 nm on UV–VIS spectrophotometer (Fig. 5). The percentage of the dyes mixture removed from the simulated dye effluents depends on the areas under the absorption bands. The ACC-1.0 performed more effectively in the treatment of simulated effluents than CAC. The percentages of removal for ACC-1.0 are 99.45% and 95.62 for effluents A and B, respectively. For CAC, the percentages 90.58% (effluent A) and 87.83% (effluent B). The ACC-1.0 can be successfully used for the treatment of simulated dye effluents compared with other adsorbents [2,5,21,25,26,40,43,56]. However, commercially available activated carbon (CAC) showed acceptable performance for treatment of simulated dye-house effluents.

Insert Fig 5

22

4. Conclusion

A mixture of powdered cocoa shell and inorganic components (red mud, lime, KOH, Al(NO3)3 and Na2SO4) were pyrolysed at 1073 K. The ratio of inorganics to organic component was varied from 1.0 to 2.0. The resulting material was treated a 6 mol L-1 HCl, under reflux (24 h), to obtain ACC-1.0 to ACC-2.0. The inorganic constituents was leached from the carbonaceous matrix during acidification, thereby, complement the chemical activation process. ACC-1.0 exhibited the best adsorption capacity for removal of RV-5 dye from aqueous solutions. For comparison purposes, the precursor (CC-1.0) was used as a proof of an improvement in the textural properties of the acidified cocoa shell activated carbon. FTIR spectroscopy, SEM, nitrogen adsorption/desorption curves, TGA/DTA thermal analysis and XRD were used for characterisation of CC-1.0 and ACC-1.0. A commercial activated carbon (CAC) was used to compare the sorption capacity of the proposed ACC1.0. The minimum contact time between the RV-5 dye and the adsorbents to attain the equilibrium were 45 min and 150 min for ACC-1.0 and CAC, respectively, at optimum pH 2.0. The adsorption process was well described by the general order kinetic model. The intraparticle diffusion model gave multiple linear portions, an indication that the adsorption followed multiple adsorption rates. Liu isotherm model gave the best fit of isothermal data. The maximum amounts of RV-5 dye removed at 298 K were 603.3 (ACC-1.0) and 517.1 mg g-1 (CAC). Thermodynamic adsorption parameters were computed. There was a physical interaction between RV-5 dye and the two adsorbents based on the magnitude of enthalpy of adsorption. The ACC-1.0 excellently decolourised simulated industrial textile effluents. It effectively removed ca 95.62% of mixture of different dyes in concentrated saline media. The sorption capacity of ACC-1.0 was compared with several adsorbents, and the acidified cocoa shell activated carbon presents sorption capacity that is comparable with those of carbon nanotubes.

Acknowledgements

23

The authors are grateful to The National Council for Scientific and Technological Development (CNPq, Brazil), The Coordination of Improvement of Higher Education Personnel (CAPES, Brazil) and to The Academy of Sciences for Developing World (TWAS, Italy) for financial support and fellowships.

References [1] H.Y. Zhu, Y.Q. Fu, R. Jiang, J. Yao, L. Liu, Y.W. Chen, L. Xiao, G.M. Zeng, Preparation, characterization and adsorption properties of chitosan modified magnetic graphitized multi-walled carbon nanotubes for highly effective removal of a carcinogenic dye from aqueous solution, Appl. Surf. Sci. 285 (2013) 865-873. [2] M.A. Adebayo, L.D.T. Prola, E.C. Lima, M.J. Puchana-Rosero, R. Cataluña, C. Saucier, C.S. Umpierres, J.C.P. Vaghetti, L.G. da Silva, R. Ruggiero, Adsorption of Procion Blue MX-R dye from aqueous solutions by lignin chemically modified with aluminium and manganese, J. Hazard. Mater., 268 (2014) 43-50. [3] C. Hessel, C. Allegre, M. Maisseu, F. Charbit, P. Moulin, Guidelines and legislation for dye house effluents. J. Environ. Manage. 83 (2007) 171-180. [4] C. Allègre, P. Moulin, M. Maisseu, F. Charbit, Treatment and reuse of reactive dyeing effluents. J. Membr. Sci. 269 (2006) 15-34. [5] L.D.T. Prola, E. Acayanka, E.C. Lima, C. Bestetti, W.O. Santos, F.A. Pavan, S.L.P. Dias, C.R.T. Tarley, Application of aqai stalks as biosorbent for the removal of Evans Blue and Vilmafix Red RR-2B dyes from aqueous solutions. Desalin. Water Treat. 51 (2013) 4582-4592. [6] S.E. Moradi, Microwave assisted preparation of sodium dodecyl sulphate (SDS) modified ordered nanoporous carbon and its adsorption for MB dye, J. Ind. Eng. Chem. 20 (2014) 208-215. [7] R.O.A. de Lima, A.P. Bazo, D.M.F. Salvadori, C.M. Rech, D.P. Oliveira, G.A. Umbuzeiro, Mutagenic and carcinogenic potential of a textile azo dye processing plant

24

effluent that impacts a drinking water source. Mutat. Res., Genet. Toxicol. Environ. Mutagen. 626 (2007) 53-60. [8] P.A. Carneiro, G.A. Umbuzeiro, D.P. Oliveira, M.V.B. Zanoni, Assessment of water contamination caused by a mutagenic textile effluent/dyehouse effluent bearing disperse dyes. J. Hazard. Mater. 174 (2010) 694-699. [9] M. Oplatowska, R.F. Donnelly, R.J. Majithiya, D.G. Kennedy, C.T. Elliott, The potential for human exposure, direct and indirect, to the suspected carcinogenic triphenylmethane dye Brilliant Green from green paper towels. Food Chem. Toxicol. 49 (2011) 1870-1876. [10] B.

Royer,

N.F.

Cardoso,

E.C.

Lima,

T.R.

Macedo,

C.

Airoldi,

A

useful

organofunctionalized layered silicate for textile dye removal. J. Hazard. Mater. 181 (2010) 366-374. [11] H. Balavi, S. Samadanian-Isfahani, M. Mehrabani-Zeinabad, M. Edrissi, Preparation and optimization of CeO2 nanoparticles and its application in photocatalytic degradation of Reactive Orange 16 dye, Powder Technol. 249 (2013) 549-555. [12] S. Singh, V.C. Srivastava, I.D. Mall, Mechanism of Dye Degradation during Electrochemical Treatment, J. Phys. Chem. C. 117 (2013) 15229–15240. [13] D. Kalpana, N. Velmurugan, J.H. Shim, B.T. Oh, K. Senthil, Y.S. Lee, Biodecolorization and biodegradation of reactive Levafix Blue E-RA granulate dye by the white rot fungus Irpex lacteus, J. Environ. Manage. 111 (2012) 142-149. [14] B. Royer, N.F. Cardoso, E.C. Lima, V.S.O. Ruiz, T.R. Macedo, C. Airoldi, Organofunctionalized kenyaite for dye removal from aqueous solution. J. Colloid Interface Sci. 336 (2009) 398-405. [15] T. Akar, S. Arslan, S.T. Akar, Utilization of Thamnidium elegans fungal culture in environmental clean-up: A reactive dye biosorption study, Ecol. Eng. 58 (2013) 363-370. [16] G.L. Dotto, E.C. Lima, L.A.A. Pinto, Biosorption of food dyes onto Spirulina platensis nanoparticles: Equilibrium isotherm and thermodynamic analysis, Bioresour. Technol. 103 (2012) 123–130.

25

[17] C.A. Demarchi, M. Campos, C.A. Rodrigues, Adsorption of textile dye Reactive Red 120 by the chitosan–Fe(III)-crosslinked: Batch and fixed-bed studies, J. Environ. Chem. Eng. 1 (2013) 1350-1358. [18] Y.N. Patel, M.P. Patel, Adsorption of azo dyes from water by new poly (3acrylamidopropyl)-trimethylammoniumchloride-co-N,N-dimethylacrylamide superabsorbent hydrogel – Equilibrium and kinetic studies, J. Environ. Chem. Eng. 1 (2013) 1368-1374. [19] W.S. Alencar, E. Acayanka, E.C. Lima, B. Royer, F.E. de Souza, J. Lameira, C.N. Alves, Application of Mangifera indica (mango) seeds as a biosorbent for removal of Victazol Orange 3R dye from aqueous solution and study of the biosorption mechanism. Chem. Eng. J., 209 (2012) 577-588. [20] S. Noreen, H.N. Bhatti, S. Nausheen, S. Sadaf, M. Ashfaq, Batch and fixed bed adsorption study for the removal of Drimarine Black CL-B dye from aqueous solution using a lignocellulosic waste: A cost affective adsorbent, Ind. Crops Prod. 50 (2013) 568-579. [21] N.F. Cardoso, E.C. Lima, B. Royer, M.V. Bach, G.L. Dotto, L.A.A. Pinto, T. Calvete, Comparison of Spirulina platensis microalgae and commercial activated carbon as adsorbents for the removal of Reactive Red 120 dye from aqueous effluents, J. Hazard. Mater. 241-242 (2012) 146-153. [22] R. D. C. Soltani, A.R. Khataee, M. Safari, S.W. Joo, Preparation of bio-silica/chitosan nanocomposite for adsorption of a textile dye in aqueous solutions, Int. Biodeter. Biodegr., 85 (2013) 383-391. [23] N.F. Cardoso, E.C. Lima, I.S. Pinto, C.V. Amavisca, B. Royer, R.B. Pinto, W.S. Alencar, S.F.P. Pereira, Application of cupuassu shell as biosorbent for the removal of textile dyes from aqueous solution. J. Environ. Manage. 92 (2011) 1237-1247. [24] T. Akar, E. Ozkara, S. Celik, S. Turkyilmaz, S. T. Akar, Chemical modification of a plant origin biomass using cationic surfactant ABDAC and the biosorptive decolorization of RR45 containing solutions Colloid Surf. B 101 (2013) 307-314.

26

[25] L.D.T. Prola, F.M. Machado, C.P. Bergmann, F.E. de Souza, C.R. Gally, E.C. Lima, M.A. Adebayo, S.L.P. Dias, T. Calvete, Adsorption of Direct Blue 53 dye from aqueous solutions by multi-walled carbon nanotubes and activated carbon, J. Environ. Manage. 130 (2013) 166-175. [26] L.D.T. Prola, E. Acayanka, E.C. Lima, C.S. Umpierres, J.C.P. Vaghetti, W.O. Santos, S. Laminsi, P.T. Njifon, Comparison of Jatropha curcas shells in natural form and treated by non-thermal plasma as biosorbents for removal of Reactive Red 120 textile dye from aqueous solution. Ind. Crop. Prod. 46 (2013) 328-340. [27] T. Calvete, E.C. Lima, N.F. Cardoso, J.C.P. Vaghetti, S.L.P. Dias, F.A. Pavan, Application of carbon adsorbents prepared from Brazilian-pine fruit shell for the removal of reactive orange 16 from aqueous solution: Kinetic, equilibrium, and thermodynamic studies, J. Environ. Manage. 91 (2010) 1695-1706. [28] F.R. Reinoso, A.S. Escribano, Porous carbons in adsorption and catalysis, in: Nalwa, H.S. (Eds), Handbook of Surfaces and Interfaces of Materials: Volume 5: Biomolecules, Biointerfaces, and Applications. Academic Press, Oxford, 2001. [29] H. Marsh, F.R. Reinoso, Activated Carbon, Amsterdam: Elsevier, 2006. [30] L. Lin, S.R. Zhai, Z.Y. Xiao, Y. Song, Q.D. An, X.W. Song, Dye adsorption of mesoporous activated carbons produced from NaOH-pretreated rice husks, Bioresou. Technol. 136 (2013) 437-443 [31] O Estado do Pará e a Produção Brasileira de Cacau, Ministério da Agricultura, Pecuária e

Abastecimento,

Comissão

Executiva

do

Plano

da

Lavoura

Cacaueira,

Superintendência no Estado do Pará, http://www.ceplacpa.gov.br/site/?p=3009, web site visited March, 10th, 2014 [32] P.S. Tully, Sulfonic acids. Kirk-Othmer Encyclopedia of Chemical Technology, John Wiley and Sons, 2000, DOI: 10.1002/0471238961.1921120620211212.a01 [33] R.A. Jacques, R. Bernardi, M. Caovila, E.C. Lima, F.A. Pavan, J.C.P. Vaghetti, C. Airoldi, Removal of Cu(II), Fe(III) and Cr(III) from aqueous solution by aniline grafted silica gel, Sep. Sci. Technol. 42 (2007) 591-609.

27

[34] R.A. Jacques, E.C. Lima, S.L.P. Dias, A.C. Mazzocato, F.A. Pavan, Yellow passion-fruit shell as biosorbent to remove Cr(III) and Pb(II) from aqueous solution, Sep. Purif. Technol. 57 (2007) 193-198. [35] J.C.P. Vaghetti, M. Zat, K.R.S. Bentes, L.S. Ferreira, E.V. Benvenutti, E.C. Lima, 4Phenylenediaminepropylsilica xerogel as a sorbent for copper determination in waters by slurry-sampling ETAAS, J Anal Atom Spectrom. 18 (2003) 376-380. [36] F. Barbosa Jr., E.C. Lima, F.J. Krug, Determination of arsenic in sediment and soil slurries by electrothermal atomic absorption spectrometry using W-Rh permanent modifier, Analyst 125 (2000) 2079-2083. [37] E.C. Lima, F. Barbosa-Jr, F.J. Krug, A. Tavares, Copper determination in biological materials by ETAAS using W-Rh permanent modifier, Talanta 57 (2002) 177-186. [38] E.C. Lima, F.J. Krug, J.A. Nóbrega, A.R.A. Nogueira, Determination of ytterbium in animal faeces by tungsten coil electrothermal atomic absorption spectrometry, Talanta 47 (1998) 613-623. [39] E.C. Lima, P.G. Fenga, J.R. Romero, W.F. de Giovani, Electrochemical behaviour of [Ru(4,4'-Me2bpy)2(PPh3)(H2O)](ClO4)2 in homogeneous solution and incorporated into carbon paste electrodes. Application to oxidation of benzylic compounds, Polyhedron 17 (1998) 313-318. [40] N.F. Cardoso, R.B. Pinto, E.C. Lima, T. Calvete, C.V. Amavisca, B. Royer, M.L. Cunha, T.H.M. Fernandes, I.S. Pinto, Removal of remazol black B textile dye from aqueous solution by adsorption, Desalination 269 (2011) 92-103. [41] Y. Liu, L. Shen, A general rate law equation for biosorption, Biochem. Eng. J. 38 (2008) 390-394. [42] Y. Liu, Y.J. Liu, Review- Biosorption isotherms, kinetics and thermodynamics, Sep. Purif. Technol. 61 (2008) 229-242. [43] W.S. Alencar, E.C. Lima, B. Royer, B.D. dos Santos, T. Calvete, E.A. da Silva, C.N. Alves, Application of aqai stalks as biosorbents for the removal of the dye Procion Blue MX-R from aqueous solution, Sep. Sci. Technol. 47 (2012) 513-526.

28

[44] Y.S. Ho, Review of second-order models for adsorption systems, J. Hazard. Mater. 136 (2006) 681-689. [45] W.J. Weber-Jr, J.C. Morris, Kinetics of adsorption on carbon from solution, J. Sanit. Eng. Div. Am. Soc. Civil Eng. 89 (1963) 31-59. [46] I. Langmuir, The adsorption of gases on plane surfaces of glass, mica and platinum, J. Am. Chem. Soc. 40 (1918) 1361-1403. [47] H. Freundlich, Adsorption in solution. Phys. Chem. Soc. 40 (1906) 1361-1368. [48] Y. Liu, H. Xu, S.F. Yang, J.H. Tay, A general model for biosorption of Cd2+, Cu2+ and Zn2+ by aerobic granules, J. Biotechnol. 102 (2003) 233-239. [49] T. Mitome, Y. Uchida, Y. Egashira, K. Hayashi, A. Nishiura, N. Nishiyama, Adsorption of indole on KOH-activated mesoporous carbon. Colloid Surf. A 424 (2013) 89-95. [50] J.A.M. Agulló, B.C. Moore, D.C. Amorós, A.L. Solano, Activation of coal tar pitch carbon fibres: Physical activation vs. chemical activation, Carbon 42 (2004) 1367-1370. [51] E.B. da Silva-Filho, M.C.M. Alves, M. da Motta, E.H.C. Oliveira, W. Brander-Jr, Study on the use of red mud for removal of dyes from textile effluents. Quim. Nova 31 (2008) 985989. [52] E.W. de Menezes, E.C. Lima, B. Royer, F.E. de Souza, B.D. dos Santos, J.R. Gregório, T.M.H. Costa, Y. Gushikem, E.V. Benvenutti, Ionic silica based hybrid material containing the pyridinium group used as adsorbent for textile dye. J. Colloid Interface Sci. 378 (2012) 10-20. [53] Maria L.P. Antunes, S.J. Couperthwaite, F.T. da Conceição, C. P.C. de Jesus, P.K. Kiyohara, A.C.V. Coelho, R.L. Frost, Red Mud from Brazil: Thermal Behavior and Physical Properties, Ind. Eng. Chem. Res. 51 (2012) 775-779. [54] M.I. El-Khaiary, G.F. Malash, Common data analysis errors in batch adsorption studies, Hydrometallurgy, 105 (2011) 314-320. [55] M. Ghaedi, A. Ansari, R. Sahraei, ZnS:Cu nanoparticles loaded on activated carbon as novel adsorbent for kinetic, thermodynamic and isotherm studies of Reactive Orange 12 and Direct yellow 12 adsorption Spectrochim. Acta Part A, 114 (2013) 687-694.

29

[56] F.M. Machado, C.P. Bergmann, E.C. Lima, B. Royer, F.E. de Souza, I.M. Jauris, T. Calvete, S.B. Fagan, Adsorption of Reactive Blue 4 dye from water solutions by carbon nanotubes: experiment and theory. Phys. Chem. Chem. Phys. 14 (2012) 11139-11153. [57] V.K. Gupta, A. Nayak, Cadmium removal and recovery from aqueous solutions by novel adsorbents prepared from orange peel and Fe2O3 nanoparticles, Chem. Eng. J. 180 (2012) 81-90. [58] P. Suksabye, P. Thiravetyan, Cr(VI) adsorption from electroplating plating wastewater by chemically modified coir pith, J. Environ. Manage., 102 (2012) 1-8. [59] C.L. Sun, C.S. Wang, Estimation on the intramolecular hydrogen-bonding energies in proteins and peptides by the analytic potential energy function, J. Mol. Struct. 956 (2010) 38-43. [60] G.Z. Kyzas, N.K. Lazaridis, D.N. Bikiaris, Optimization of chitosan and β-cyclodextrin molecularly imprinted polymer synthesis for dye adsorption, Carbohyd. Polym. 91 (2013) 198-208. [61] N.A. Travlou, G.Z. Kyzas, N.K. Lazaridis, E.A. Deliyanni, Graphite oxide/chitosan composite for reactive dye removal. Chem. Eng. J., 217 (2013) 256-265.

30

Table 1. Chemical composition of the simulated dyehouse effluents. Concentration (mg L-1)

Dyes

Effluent A

Effluent B

Reactive Violet 5 (λmax 545 nm)

100

200

Reactive Orange 16 (λmax 489 nm)

20

40

Direct Blue 53 (λmax 607 nm)

20

40

Cibacron Brilliant Yellow 3G-P (λmax 402 nm)

20

40

Reactive Red M-2BE (λmax 505 nm)

20

40

Na2SO4

150

300

NaCl

150

300

Na2CO3

100

200

CH3COONa

100

200

CH3COOH

300

300

Sodium dodecyl sulphate (SDS)

106

106

2

2

Auxiliary chemicals

pH*

-1

*pH of the solution was adjusted with 0.10 mol L HCl and/or 0.10 mol L-1 NaOH.

31

Table 2: FTIR vibrational bands of CC-1.0 and ACC-1.0. Assignments are based on literature [16,21,23,26,40]. CC-1.0

Band (cm-1)

Assignments

3344

O-H stretch

1545

Rings mode of aromatic

1406

CH2 bending

995

C-O stretch of alcohols or Si-O stretch of silicates

665, 615

CH out of plane bends of aromatic rings

ACC-1.0 3345

O-H stretch

1587

Rings mode of aromatic

1094

C-O stretch of alcohols or Si-O stretch of silicates

966, 799

CH out of plane bends of aromatic rings

32

Table 3. Kinetic parameters RV-5 dye adsorption onto ACC-1.0 and CAC. Conditions: temperature, 298 K; pH, 2.0; mass of adsorbent, 50.0 mg.

Pseudo-first-order kf (min-1) qe (mg g-1) ho (mg g-1 min-1) t1/2 (min) R2 adj Ferror (mg g-1) Pseudo-second-order ks (g mg-1 min-1) qe (mg g-1) ho (mg g-1 min-1) t1/2 (min) R2 adj Ferror (mg g-1) General order kN [h-1.(g mg-1)n-1] qe (mg g-1) n h0 (mg g-1 min-1) t1/2 (min) R2 adj Ferror (mg g-1)

Intra-particle diffusion kid,2 (mg g-1 min-0.5)a a second stage

ACC-1.0 400.0 mg L-1 1000.0 mg L-1

400.0 mg L-1

CAC 1000.0 mg L-1

0.1898 157.4 29.87 3.65 0.9958 2.437

0.1894 388.5 73.58 3.66 0.9948 6.721

0.05038 156.0 7.860 13.76 0.9979 2.205

0.05042 229.7 11.58 13.75 0.9976 3.437

2.620.10-3 161.8 68.61 2.36 0.9912 3.536

1.052.10-3 399.7 167.98 2.38 0.9923 8.165

4.498.10-4 167.1 12.56 13.30 0.9891 4.978

3.049.10-4 246.1 18.47 13.32 0.9897 7.116

0.04389 158.5 1.339 38.68 3.20 0.9998 0.5628

0.02623 391.8 1.378 98.22 3.16 0.9996 1.869

0.01780 157.4 1.228 8.881 13.31 0.9997 0.7715

0.01491 231.9 1.246 13.20 13.28 0.9997 1.163

6.869

16.34

4.354

6.359

33

Table 4. Isotherm parameters for RV-5 dye adsorption using ACC-1.0 and CAC. Conditions: pH, 2.0; adsorbent mass, 50.0 mg; contact time, 60 min for ACC-1.0 and 180 min for CAC. 298 K

303 K

308 K

313 K

318 K

323 K

470.4 0.7034 0.9389 37.67

474.0 0.5199 0.9594 31.59

481.1 0.3502 0.9746 25.20

486.64 0.2503 0.9894 16.72

488.0 0.1918 0.9970 9.177

480.4 0.1526 0.9991 5.326

252.0 7.089 0.9779 22.64

232.5 6.382 0.9690 27.58

215.9 5.807 0.9603 31.53

196.79 5.237 0.9442 38.32

175.1 4.701 0.9305 44.47

153.8 4.283 0.9168 50.63

603.3 0.3237 0.3873 0.9998 1.907

577.7 0.2756 0.4704 0.9998 2.4351

554.1 0.2384 0.5738 0.9997 2.5876

530.4 0.2049 0.7017 0.9996 3.316

510.8 0.1734 0.8259 0.9997 2.947

491.0 0.1471 0.9099 0.9997 3.0632

383.6 0.8658 0.9368 29.14

369.2 0.6080 0.9382 26.99

349.6 0.6033 0.9437 24.40

340.6 0.3994 0.9591 19.77

320.1 0.4761 0.9440 23.36

301.3 0.5068 0.9374 23.75

222.3 8.261 0.9903 11.42

211.0 8.179 0.9913 10.14

198.0 8.056 0.9916 9.428

184.2 7.697 0.9925 8.444

163.0 6.951 0.9887 10.49

150.7 6.715 0.9887 10.08

517.1 0.2836 0.3339 0.9996 2.236

496.3 0.2402 0.3454 0.9999 0.7896

476.3 0.2121 0.3418 0.9998 1.302

453.9 0.1792 0.3677 0.9998 1.468

430.4 0.1609 0.3775 0.9997 1.588

416.5 0.1351 0.3699 0.9998 1.323

ACC-1.0

Langmuir Qmax (mg g-1) KL (L mg-1) 2 R adj -1 Ferror (mg g ) Freudlich -1 -1 -1/n KF (mg g (mg L ) F) nF 2 R adj -1 Ferror (mg g ) Liu -1 Qmax (mg g ) -1 Kg (L mg ) nL R2 adj -1 Ferror (mg g ) CAC Langmuir Qmax (mg g-1) KL (L mg-1) 2 R adj -1 Ferror (mg g ) Freudlich -1 -1 -1/n KF (mg g (mg L ) F) nF 2 R adj -1 Ferror (mg g ) Liu -1 Qmax (mg g ) -1 Kg (L mg ) nL R2 adj Ferror (mg g-1)

34

Table 5. Maximum sorption capacities of different adsorbents used for removal of various dyes.

Adsorbent

Dye Adsorbate

chitosan-modified magnetic graphitized

Congo Red

multi-walled carbon nanotubes Multi wall carbon nanotube (MWCNT)

Qm ax (mg g-1) 263.3

Ref. 1

Reactive Red M-2BE

335.7

2

Fungi: Thamnidium elegans

Reactive Red 198

234.24

15

chitosan–Fe(III)-crosslinked

Reactive Red 120

336.77

17

poly (3-acrylamidopropyl)trimethylammonium chloride-co-N,Ndimethylacrylamide hydrogel (SAH) SAH

Acid Red 18

249.87

18

Acid Yellow 36

199.96

18

Acid Red 73

198.16

18

Peanut husk treated with HCl

Drimarine Black CL-B

51.02

20

Spirulina platensis microalgae

Reactive red 120

482.2

21

Commercial activated carbon

Reactive red 120

267.2

21

Cupuassu shell

Reactive Red 194

64.1

23

Cupuassu shell

Direct Blue 53

37.5

23

Reactive Red 45

152.49

24

Direct Blue 53

409.4

25

Direct Blue 53

135.2

25

Reactive orange 16

472

27

SAH

Pyracantha coccinea treated with quaternarium ammonium salt Multi wall carbon nanotube (MWCNT) Commercial activated carbon Brazilian-pine-fruit shell ZnS:Cu

nanoparticles

loaded

on

Reactive Orange 12

activated carbono (ZnS–Cu–NP–AC) ZnS–Cu–NP–AC Single-walled

Direct Yellow 12 carbon

nanotubes

Reactive Blue 4

(SWCNT)

382.8 325.3 567.7

55 55 56

Multi wall carbon nanotube (MWCNT)

Reactive Blue 4

502.5

56

Commercial Activated Carbon (CAC)

Reactive Violet 5

517.1

This work

Acidified cocoa shell activated carbon

Reactive Violet 5

(ACC-1.0)

603.3

This work

35

Table 6. Thermodynamic parameters of RV-5 dye adsorption onto ACC-1.0 and CAC. Conditions: mass of adsorbent, 50.0 mg; pH, 2.0; contact time, 60 min for ACC-1.0 and 180 min for CAC. Temperature (K) 298 303

ACC-1.0 Kg (L mol-1)

308

313

318

323

2.381·105

2.027·105

1.753·105

1.507·105

1.275·105

1.082·105

∆G (kJ mol-1)

-30.67

-30.78

-30.92

-31.03

-31.08

-31.13

-1

-25.07 18.91 0.9976

-

-

-

-

-

2.086·105

1.767·105

1.560· 105

1.318·105

1.183·105

9.937·104

∆G (kJ mol-1)

-30.35

-30.44

-30.62

-30.68

-30.88

-30.90

-1

-23.22 23.90 0.9960

-

-

-

-

-

∆H° (kJ mol ) ∆S° (J K-1 mol-1)

R

2

adj

CAC Kg (L mol-1) ∆H° (kJ mol ) ∆S° (J K-1 mol-1) R

2

adj

36

Figures Captions

Fig. 1. Production of cocoa shell activated carbon.

Fig. 2A) Dependence of pH on the sorption capacity of RV-5 dye on CAC and ACC-1.0. Conditions: temperature, 298 K; adsorbent mass, 50.0 mg; dye concentration, 400.0 mg L-1. B) pHpzc CAC; C) pHpzc CAC-1.0. Fig. 3. Kinetic isotherm curves. A) CAC and RV-5 dye at 400.0 mg L-1; B) CAC and RV-5 dye at 1000.0 mg L-1; C) ACC-1.0 and RV-5 at 400.0 mg L-1; D) ACC-1.0 and RV-5 at 1000.0 mg L-1. Conditions: initial pH, 2.0; temperature, 298 K; adsorbent mass, 50.0 mg.

Fig. 4. Isotherm curves of RV-5 dye on (A) CAC and (B) ACC-1.0 at 298 K. Conditions: initial pH, 2.0; adsorbent mass, 50.0 mg; contact time, 60 min for ACC-1.0 and 180 min for CAC.

Fig. 5. UV/Vis spectra of simulated dye effluents before and after treatment with ACC-1.0 and CAC. A) Effluent A; B) Effluent B. See Table 1 for constituents of effluents.

Cylinder for preparation of activated carbon

Oil extracted from biomass

Carbon material cylinder dried

carbon material removed in cylinder form out of reactor

Cylinder introduced into reactor

Activation step: Lixiviation of inorganics and filtration

Furnace setup

Activated Carbon

Fig. 1

pHPZC=2.35

100

CAC ACC-1.0

A

% Removal

80

60

pHPZC=7.31

40

20 0

2

4

6

8

10

initial pH

B

2.5 CAC

2.0 1.5

DpH

1.0 0.5 pHpcz= 7.31

0.0 -0.5 0

2

4

6

8

10

Initial pH

2 1

ACC-1.0

pHPZC= 2.35

C

0

DpH

-1 -2 -3 -4 -5 -6 0

2

4

6

Initial pH

8

10

Fig. 2

A

B

250

150

Experimental points Pseudo-first order Pseudo-second order General-order

100

qt (mg g-1)

qt (mg g-1)

200

50

150

Experimental points Pseudo-first order Pseudo-second order General-order

100 50 0

0 0

60

120 180 240 300 360 420 480 540

0

60

Time (min)

Time (min)

C

160

D

400

120

300 Experimental points Pseudo-first order Pseudo-second order General-order

80

qt (mg g-1)

qt (mg g-1)

120 180 240 300 360 420 480 540

40

100

0

0 0

60

120 180 240 300 360 420 480 540

Time (min)

Experimental points Pseudo-first order Pseudo-second order General-order

200

0

60

120 180 240 300 360 420 480 540

Time (min)

Fig 3

A

400

qe (mg g-1)

300 Experimental points Langmuir Freundlich Liu

200

100

0 0

40

80

120

160

200

Ce (mg L-1)

B

500

qe (mg g-1)

400 300 Experimental points Langmuir Freundlich Liu

200 100 0 0

25

50

75

100

125

150

175

200

Ce (mg L-1)

Fig 4

2.5

Effluent A before treatment (diluted 2X before spectroscopic measurements)

A

Effluent A undiluted after adsorption with ACC-1.0 Effluent A undiluted after adsorption with CAC

Absorbance

2.0 1.5 1.0 0.5 0.0 300

400

500

600

700

800

Wavelength (nm)

2.5

Effluent B before treatment (diluted 4X before spectroscopic measurements)

B

Effluent B undiluted after adsorption with ACC-1.0 Effluent B undiluted after adsorption with CAC

Absorbance

2.0 1.5 1.0 0.5 0.0 300

400

500

600

700

800

Wavelength (nm)

Fig 5

2.5

Effluent before treatment (diluted 4X before spectroscopic measurements)

2.5 Effluent undiluted after adsorption with ACC-1.0

2.0

1.5 Cocoa activated carbon

1.0 0.5

Absorbance

Absorbance

2.0

1.5 1.0 0.5

0.0

0.0 300

400

500

600

Wavelength (nm)

700

800

300

400

500

600

Wavelength (nm)

700

800



Acidified cocoa activated carbon (ACC-1.0) was prepared and characterised



Adsorption of RV-5 dye was studied using ACC-1.0 and CAC



Adsorption maximum values were 603.3 mg g (ACC-1.0) and 517.1 mg g (CAC)



General order kinetic model suitably described the adsorption processes



ACC-1.0 effectively decolourised simulated industrial effluents

-1

-1