Applied Clay Science 87 (2014) 298–302

Contents lists available at ScienceDirect

Applied Clay Science

j ourna l homepage: www.e lsev ie r .com/ locate /c lay

Note

Adsorption of anthocyanins using clay–polyethylenenanocomposite particles

Toni J. Lopes a,⁎, Odinei H. Gonçalves b, Mara G.N. Quadri a, Ricardo A.F. Machado a, Marintho B. Quadri a

a Departamento de Engenharia Química e Engenharia de Alimentos, Universidade Federal de Santa Catarina, Santa Catarina, Brazilb Programa de Pós-graduação em Tecnologia de Alimentos—PPGTA, Universidade Tecnológica Federal do Paraná, Campo Mourão, Brazil

⁎ Corresponding author at: Federal University of RioAntônio da Patrulha, Rua Barão do Caí, 125, Cidade Alta,da Patrulha, Rio Grande do Sul, Brazil. Tel.: +55 51 36627

E-mail address: lopes@enq.ufsc.br (T.J. Lopes).

0169-1317/$ – see front matter © 2013 Elsevier B.V. All rihttp://dx.doi.org/10.1016/j.clay.2013.11.038

a b s t r a c t

a r t i c l e i n f o

Article history:Received 8 October 2011Received in revised form 28 November 2013Accepted 29 November 2013Available online 16 December 2013

Keywords:CharacterizationAdsorptionSeparationDyeClay

Clays are often used in adsorption processes due to their remarkable adsorptive characteristics and high specific sur-face area. However, a colloidal system is formed when the clay is dispersed in water, leading to difficulties in clayrecovery and separation at the end of the process since clay sedimentation is hindered by colloidal forces. The objec-tive of thisworkwas to obtain claypolymernanocomposites (CPN) comprised ofmicrometric polyethyleneparticlespresenting clay particles physically adhered to the surface. How the CPN particles interact with the aqueous phasewas also investigated, since the adsorption processes are usually carried out in several washing steps.

© 2013 Elsevier B.V. All rights reserved.

1. Introduction

Adsorption is widely used in industrial processes with adsorbentmaterials generally characterized by their particulate form and highspecific surface area (SSA). Clay is a natural, low cost material thatplays an important role as an alternative adsorbent medium (Gürseset al., 2004; Kelm et al., 2003; Nayak and Singh, 2007). It has beenused to remove contaminants, such as dyes, from industrial effluents(Ghosh and Bhattacharyya, 2002; Harris et al., 2001). When dispersedin water, clays are prone to form a colloidal dispersion mainly due totheir small particle sizes (less than 4 μm according to the Atterbergand Wentworth granulometric scales (Carvalho et al., 2000)), groundtexture and surface charge (Baik and Lee, 2010; Douch et al., 2009).

The plastic behaviour of clay–water dispersions leads to difficultiesin sedimentation and/or separation of the solid adsorbent from the liq-uid phase in static adsorption systems. On the other hand, continuousprocesses, such as bed columns, may undergo structural modifications(cracking, preferential ways formation and permeability loss) due tobed swelling, compaction and/or colmatation, resulting in liquid flowinstability and incomplete adsorption (Lopes et al., 2005, 2007;Yaneva et al., 1995; Yilmaz and Civelekoglua, 2009).

In order to overcome this difficulty, calcination is generally carried outto form pellets and avoid disruption. However, the high temperatures

Grande - FURG, Campus SantoCEP:95500-000, Santo Antônio815.

ghts reserved.

result in a remarkable reduction in the clay adsorption capacity (Ghoshand Bhattacharyya, 2002; Lopes et al., 2005, 2007).

In order to find a solution for the above mentioned problems, thiswork aims to obtain a clay polymer nanocomposite (CPN) composedof clay immobilized on the surface of polymer pellets. The adsorptioncapacity of the CPN particles in the partial purification of anthocyaninsfrom red cabbage was also tested.

2. Experimental procedure

2.1. Material

The adsorbent media used was Tonsil Terrana 580 FF clay (SüdChemie). Pellets of high density polyethylene (HDPE, IpirangaPetroquímica S.A.), HD 7255 LS-L, GE 4960 BR, GD 5150 K, GF 5150 andGM 9450 F were used as support (3 mm average particle diameter,954 ± 3 kg m−3 specific mass and softening temperature from 165 to230 °C). Red cabbage dye was obtained by acidic aqueous extractionfrom crude red cabbage (Cooper-Driver, 2001; Fan et al., 2008; Jackmanand Smith, 1992). Alternatively, an aqueous solution of commercial redcabbage dye (30-WS-P, Christian Hansen Inc.) was prepared.

2.2. Adsorbent characterization

Pastilles of Tonsil Terrana 580 FF claywere analysed by X-ray fluores-cence (Philips, model PW). The zeta potential (Zeta Plus) was measuredevaluating the eletrophoreticmobility of thewater dispersed particles bylaser diffraction. Clay SSAwasmeasured using liquid nitrogen adsorption(Autosorb-1, Quantachrome) (T = 77 K) by the Brunauer, Emmett and

1212

14

299T.J. Lopes et al. / Applied Clay Science 87 (2014) 298–302

Teller (BET)method. SSAwasmeasured for the crude clay and after ther-mal treatment (210 °C for 2 h). X-ray diffraction (XRD) (Philips, modelXpert) was carried out to investigate the crystallinity before and afterthermal treatment (210 °C for 2 h), as well as to observe possible struc-tural modifications. Co Kα radiation (λ = 1.5418 Å) was used at 40 kVand 30 mA. The scanning conditions were 0.05° s−1 from 10 to 70°.

Changes in the claymorphology during the heating processwere ob-served by thermal gravimetry (TG) and differential thermal analysis(DTA) using a simultaneous thermal analyser model Netzsch STA 409.Samples were heated from about 25 °C up to 320 at a 5 °C min−1

heating rate.

2.3. Pellet manufacturing

HDPE pellets were mixed with the clay at a mass ratio of 1:1 (clay/polymer). The system was heated at their nominal softening tempera-tures for 2 h in an oven. The pellets were then washed and dried toremove the clay that did not stick to the pellets' surface.

2.4. Particle morphology

Scanning electronmicroscopy (SEM, PHILIPS, model XL-30 at 20 kV)was used to evaluate the surface of the clay polymer pellets before theadsorption step and after the desorption step. Images from bothbackscattered and secondary electrons were considered. The pelletswere immersed in liquid nitrogen and broken in order to show theinner polymeric matrix and the clay layer was evaluated using anappropriate image analyser software.

2.5. Static adsorption experiments

Static experimentswere carried outwith the clay–polymer particles,mixing 50 mL of a 3 mg·mL−1 aqueous solution of a commercial redcabbage dye under magnetic stirring for 30 min. The clay mass in eachessay was determined. McIlvaine buffer solution (pH 3) was used inthe adsorption, ethanol–water mixtures (30% vol/vol) and acetic aciddilutions (5% vol/vol) (Xavier et al., 2008). Desorption was carried outusing an ethanol/water/citric acid solution (70:30:5 vol/vol/wt) for30 min. Dye concentrations in the liquid samples were measured in aspectrophotometer (Spectronic Unicam, USA, model Genesys 10 Vis)at 550 nm. The calibration curve was obtained using the commercialdye.

3. Results and discussions

3.1. Adsorbent characterization

The clay elementary analysis, the dependence of zeta potential onpH at 22 ± 1 °C and the XRD of the Tonsil Terrana 580 FF clay areshown in Table 1, Figs. 1 and 2, respectively.

Table 1Average chemical composition of Tonsil Terrana 580 FF clay.

Compounds Mass (%)

Silicon dioxide (SiO2) 53.48 ± 2.86Aluminium oxide (Al2O3) 17.86 ± 2.11Iron oxide (Fe2O3) 8.29 ± 1.07Calcium oxide (CaO) 2.60 ± 0.21Sodium oxide (Na2O) 0.12 ± 0.01Potassium oxide (K2O) 3.25 ± 0.44Manganese oxide (MnO) 0.18 ± 0.00Titanium oxide (TiO2) 0.90 ± 0.01Magnesium oxide (MgO) 2.61 ± 0.47Phosphorus oxide (P2O5) 0.34 ± 0.01Weight loss (T = 950 °C) 10.02 ± 0.88

The high content of iron (8.29 ± 1.07%) and potassium(3.25 ± 0.44%) oxides may indicate the presence of illite. Mass lossesat 950 °C are mainly due to water, and hydroxyl or hydroxide groupdecompositions such as Al(OH)3 and Fe(OH)3. Volatile compounds,such as organic matter, sulphides, carbonates and sulphates, are also in-cluded in that determination (Quantachrome Corporation Catálogue,2000; Tateo et al., 2006).

The zeta potential shows the superficial charge as a function of thepH of the colloidal systems (Liu et al., 2002). The existence of negativecharges on the clay surface in an acid medium is very important whenclay is used to adsorb red cabbage anthocyanins from the solution be-cause, in this situation, the dye shows the positively charged flaviliumstructural form (Brouillard, 1983). The pH of zero potential was 9.56,indicating that the clay is a good material to be used in processeswhere positively charged adsorbates are used at a pH of less than 9.56.The pHvalue of clay dispersed inwaterwas 7.8, corresponding to amin-imum potential zeta and a maximum adsorption of the flavilium ion.

Seven crystallographic phaseswere observed inX-ray diffractograms:illite (K(Al Fe)2AlSi3O10(OH)2H2O — JCPDS 15-603), kaolinite(Al2Si2O5(OH)4 — JCPDS 06-221), quartz (SiO2 — JCPDS 05-490),illite–montmorillonite (KAl4(SiAl)8O20 (OH)4 xH2O— JCPDS 07-330),montmorillonite (Na0.3(Al,Mg)2Si4O10(OH)2 xH2O — JCPDS 13-259),muscovite (KAl2Si3AlO10(OH)2 — JCPDS 7-25), and calcium andiron oxides (CaFe4O7 — JCPDS 12-145) (JCPDS, 1981). Similar peakconfigurations were shown by the XRD of crude and heated clay,indicating the stability of the mineralogical phases. Magnitude al-terations of the peak height were observed after heating at 210 °Cfor 2 h, probably due to the loss of water and organic material fromthe structure.

The values found for N2 adsorption at 77 K before and after thermaltreatment at 210 °C for 2 h are shown in Table 2. For heating up to320 °C, themass variation observed in the thermal gravimetric analyses(TG and DTA) of the sample is shown in Fig. 3.

Thewater or organic matter losses (about 4%) can be responsible forthe changes observed in the physical properties of Tonsil Terrana 580 FFclay. During the heating process, mass loss began at approximately60 °C and reached around 80% of the total at 160 °C, indicating that itwas probably due to evaporation of free water found between the claylayers. The DTA showed endothermic effects related to the observedmass loss, showing a peak at 100 °C that corresponds to thewater evap-oration temperature. It was observed that the external surface of theclay darkened after heating, probably due to the oxidation reactions ofthe external organic matter adhering to the clay particles. Accordingto Reed (1998) these reactions are followed by an increase in the activesites available to the adsorption process since an increase in the adsorp-tive capacity is usually observed.

0

1,6

-3,4-2,6

-1,8-1,7

-6

-4

-2

0

2

4

6

8

10

0 2 4 6 8 10 12

pH

Zet

a P

oten

tial

(m

V)

Fig. 1. Zeta potential of Tonsil Terrana 580 FF clay at 22 °C.

Fig. 2. X-ray diffractograms for crude and heated clays (210 °C for 2 h).

0

0,02

0,04

0,06

0,08

0,1

0,12

0,14

0,16

-10

-8

-6

-4

-2

0

2

0 50 100

150

200

250

300

350

DT

A,

V/m

g

mass (%

)

Temperature (oC)

Fig. 3. Mass variation (%) as a function of the temperature (TG) and differential thermalanalysis (DTA) of Tonsil Terrana 580 FF clay.

Table 3HDPE properties and the results of adhered clay.

300 T.J. Lopes et al. / Applied Clay Science 87 (2014) 298–302

3.2. Particle morphology

HDPE was chosen due to its relatively low softening temperatureand because it is used worldwide in food contact applications. Initially,five HDPE grades were evaluated, each one presenting different nomi-nal softening temperatures. Clay fixation was carried out according toTable 3 (the results are the average values of three experiments).

In the evaluated experimental conditions, HDPE 7255 LS-L presentedthe highest amount of adhered clay on the pellet surface showing thatthis HDPE grade can be used as a clay support. Moreover, HDPE gradesGE 4960 BR, GD 5150 K, GF 5150 and GM 9450 F deformed during theclay fixation despite the fact that they were heated at temperaturesnear their nominal softening temperatures. In the following experi-ments, HDPE HD7255 LS-L was used in the adsorption/desorption ofthe red cabbage dye.

Scanning electron micrographs of the surface and inner structure ofa typical clay/HDPE particle before it was used in the adsorption–desorption steps are presented in Fig. 4a and c, respectively. The SEMof the clay/HDPE particle after the desorption step of the commercialred cabbage dye is shown in Fig. 5. SEM images were used to measurethe clay layer thickness before and after the desorption step, as present-ed in Figs. 4c and 5e, respectively. Thirtymeasurements of the clay layerwere carried out for three clay/HDPE particles at 200× amplification.

It can be seen in Fig. 4a that the surface of theHDPE particle is entire-ly covered with a uniform clay layer. In Fig. 4b, the light grey outline ofthe particle corresponds to the clay layer, and the inner darker region tothe polymeric matrix.

Even after 60 min of stirring and the use of three different solutions(buffer pH 3.0, dye solution and acidified 70% ethanol), the clay layerstill covered most of the polymer matrix surface (Fig. 5a). After thedesorption step, less clay can be seen on the particle surface (Fig. 5c)compared to the particle before adsorption (Fig. 4c), as well a reductionin the size of the clay aggregates (Fig. 5b). However, the clay layer on theparticle surface was still uniform (Fig. 5d).

The images in Figs. 4c and 5e show an important amount of clayfixed on the particle surface. The layer was still homogeneous after thedye adsorption–desorption process. The mean value of the layer thick-ness before the process was 55 ± 6 μm adsorption and 28 ± 8 μm

Table 2Physical properties of Tonsil Terrana 580 FF clay.

Crude clay Clay after thermal treatment(210 °C for 2 h)

BET specific superficial area (m2 g−1) 112.3 120.1Micropores volume (cm3 g−1) 0.0439 0.0466Pore diameter (Å) 4.97 7.52

after the dye desorption,meaning that thewhole process led to a reduc-tion of about 50% in the clay layer.

3.3. Static adsorption experiments

The batch essays of dye adsorption were carried out using the clay/HDPE particles. Adsorption was also carried out using only pure clayin order to determine whether the clay fixation procedure affected itsadsorptive capacity. The results for the natural dye adsorption at an ini-tial dye concentration of 1 mg mL−1 are presented in Table 4.

It can be seen that fixation of the clay on the polymer did not hinderdye adsorption on the clay active sites. The amount of dye adsorbed onthe particles was 11.76 mgdye gclay−1 , indicating that the clay adhered onthe polymer had a good adsorption performance. This suggests that itcould be used in either batch or continuous adsorption systems.

4. Conclusion

Clay/polymer pellets were obtained by heating at the polymer soft-ening temperature, resulting in CPN particles suitable to be used inthe adsorption/desorption of red cabbage dye. After the thermal treat-ment, the clay presented negative charges when in aqueous solutionsat a pH lower than9.56, indicating that this clay can beused in processeswhere positively charged adsorbates are involved.

The thermogravimetric analysis and DTA showed that heating up to320 °C promotedwater evaporation and organicmatter oxidationwith-out affecting the adsorptive properties of the clay. Seven crystal phaseswere identified by XRD, namely illite, muscovite, montmorillonite andinterstratification of illite–montmorillonite.

SEM images showed that claywas successfully fixed on the polyeth-ylene pellets and that a small amount of clay was lost during the antho-cyanin adsorption–desorption process. The results strongly suggestedthat the CPN particles could be used as adsorbents and also that they

HDPEgrade

HDPEspecificweight(g/cm3) a

Nominalsofteningtemperaturea

(°C)

Temperature(°C)

Heatingtime (h)

Mass of adheredclay(g clay·gpolymer−1)

GE 4960 BR 1.000 170–175 180 2 0.092 ± 0.005GD 5150 K 0.946 165–180 190 2 0.083 ± 0.012GF 5150 0.948 165–180 190 2 0.100 ± 0.010GM 9450 F 0.952 180–200 210 2 0.119 ± 0.007HD 7255 LS-L 0.954 200–230 210 2 0.135 ± 0.018

a Product Catalog Ipiranga Petroquímica (2005).

a b

Clay Polyethylene

c

Fig. 4. Scanning electronic micrographs of the a) entire (25×), b) cutted (15×) and c) cutted (200×) HDPE particles covered with clay.

a b

dc Clay

Polyethylene

e

Fig. 5.Micrographs of clay–HDPE particles after desorption of the red cabbage commercial dye: a) entire particle (25×); b) entire particle, agglomerate in focus (200×); c) entire particle,absence of clay (25×); d) cutted particle (25×) and e) cutted particle (200×).

301T.J. Lopes et al. / Applied Clay Science 87 (2014) 298–302

Table 4Batch essays of red cabbage anthocyanin adsorption on CPN particles.

Adsorbed dye mass (mgdye·g clay−1)

Clay HDPE nanocomposite 11.76 ± 0.12Pure clay 12.25 ± 0.01

302 T.J. Lopes et al. / Applied Clay Science 87 (2014) 298–302

could solve flow problems in fixed bed continuous adsorptionprocesses.

References

Baik, M.H., Lee, S.Y., 2010. Colloidal stability of bentonite clay considering surface chargeproperties as a function of pH and ionic strength. J. Ind. Eng. Chem. 16 (5), 837–841.

Brouillard, R., 1983. The in vivo expression of anthocyanin colour in plants. Phytochemistry22, 1311–1323.

Carvalho, N.O., Júnior, N.P.F., Santos, P.M.C., Lima, J.E.F.W., 2000. SedimentometricPractices Guide. Brazilian Electricity Regulatory Agency — ANEEL, Hydrologic Studiesand Information Department — SIH, Brazil.

Cooper-Driver, G.A., 2001. Contributions of Jeffrey Harbone and co-workers to the studyof anthocyanins. Phytochemistry 56 (229–236), 2001.

Douch, J., Hamdani, M., Fessi, H., Elaissanri, A., 2009. Acid–base behavior of a colloidalclays fraction extracted from natural quartz sand: effect of permanent surface charge.Colloids Surf. A Physicochem. Eng. Asp. 338 (1–3), 51–60.

Fan, G., Han, Y., Gu, Z., Chen, D., 2008. Optimizing conditions for anthocyanins extractionfrom purple sweet potato using response surface methodology (RSM). LWT Food Sci.Technol. 41, 155–160.

Ghosh, D., Bhattacharyya, K.G., 2002. Adsorption of methylene blue on kaolinite. Appl.Clay Sci. 20, 295–300.

Gürses, A., Karaca, S., Dogar, Ç., Bayrak, R., Acikyldiz, M., Yalçin, M., 2004. Determination ofadsorptive properties of clay/water system: methylene blue sorption. J. Colloid InterfaceSci. 269, 310–314.

Harris, R.G., Wells, J.D., Johnson, B.B., 2001. Selective adsorption of dyes and other organicmolecules to kaolinite and oxide surfaces. Colloids Surf. A Physicochem. Eng. Asp.180, 131–140.

Ipiranga Petroquímica, S.A., 2005. Product Catalog.Jackman, R.L., Smith, J.L., 1992. Anthocyanins and betalains. In: Hendry, G.A.F., Houghton,

J.D. (Eds.), Natural Food Colorants. Blackie Academic, London, pp. 183–241.JCPDS, 1981. Powder Diffraction File Search Manual. International Centre for Diffraction

Data, Pennsylvania.Kelm, U., Sanhueza, V., Guzman, C., 2003. Filtration and retention of mineral processing

slurries with pumice and common clay: low-cost materials for environmentalapplications in the small-scale mining industry. Appl. Clay Sci. 24, 35–42.

Liu, J., Zhou, Z., Xu, Z., Masliyah, J., 2002. Bitumen–clay interactions in aqueous mediastudied by zeta potential distribution measurement. J. Colloid Interface Sci. 252 (2),409–418.

Lopes, T.J., Gonçalves, O.H., Barros, R.B., Capelli, R.B., Machado, R.A.F., Quadri, M.G.N.,Quadri, M.B., 2005. Adsorption of natural dyes on clay fixed on polymers. Braz.Arch. Biol. Technol. 48-Special, 275–280.

Lopes, T.J., Quadri, M.G.N., Quadri, M.B., 2007. Recovery of anthocyanins from red cabbageusing sandy porous medium enriched with clay. Appl. Clay Sci. 37, 97–106.

Nayak, P.S., Singh, B.K., 2007. Removal of phenol from aqueous solutions by sorption onlow cost clay. Desalination 207, 71–79.

Quantachrome Corporation Catálogue, 2000. An Introduction to BET Theory.Reed, J.S., 1998. Principles of Ceramics Processing, 2nd ed. Wiley-Interscience Publication

588–595.Tateo, F., Summa, V., Giannossi, M.L., Ferraro, G., 2006. Healing clays: mineralogical and

geochemical constraints on the preparation of clay–water suspension (“argillicwater”). Appl. Clay Sci. 33 (3–4), 181–194.

Xavier, M.F., Lopes, T.J., Quadri, M.G.N., Quadri, M.B., 2008. Extraction of red cabbageanthocyanins: optimization of the operation conditions of the column process.Braz. Arch. Biol. Technol. 51 (1), 143–152.

Yaneva, M.R., Kostakeva, E., Toshev, D., 1995. Contaminant migration fromsanitary landfill leachate through soil monoliths. Water Sci. Technol. 32 (7),215–219.

Yilmaz, I., Civelekoglua, B., 2009. Gypsum: an additive for stabilization of swelling claysoils. Appl. Clay Sci. 44 (1–2), 166–172.