Adsorption of food dyes from aqueous solution by glutaraldehyde cross-linked magnetic chitosan nanoparticles

Adsorption of food dyes from aqueous solution by glutaraldehyde cross-linked magnetic chitosan nanoparticles

Journal of Food Engineering 126 (2014) 133–141 Contents lists available at ScienceDirect Journal of Food Engineering journal homepage: www.elsevier...

704KB Sizes 0 Downloads 57 Views

Journal of Food Engineering 126 (2014) 133–141

Contents lists available at ScienceDirect

Journal of Food Engineering journal homepage: www.elsevier.com/locate/jfoodeng

Adsorption of food dyes from aqueous solution by glutaraldehyde cross-linked magnetic chitosan nanoparticles Zhengkun Zhou a,b, Shiqi Lin b, Tianli Yue a,⇑, Tung-Ching Lee b,⇑ a b

College of Food Science and Engineering, Northwest A&F University, Yangling, Shaanxi 712100, PR China Department of Food Science, Rutgers, The State University of New Jersey, 65 Dudley Road, New Brunswick, NJ 08901, USA

a r t i c l e

i n f o

Article history: Received 7 August 2013 Received in revised form 13 November 2013 Accepted 14 November 2013 Available online 22 November 2013 Keywords: Food dye Adsorption Magnetic chitosan nanoparticles Glutaraldehyde Cytotoxicity

a b s t r a c t Glutaraldehyde cross-linked magnetic chitosan nanoparticles (GMCNs) were prepared through crosslinking modification of magnetic chitosan nanoparticles (MCNs) using glutaraldehyde as crosslinker, that not only exhibited excellent food dyes adsorption performance, but also showed low cytotoxicity. Adsorption characteristics of GMCNs in FD&C Blue 1 and D&C Yellow 5 aqueous solutions have been studied and results indicated that the adsorption capacities were affected by initial pH values, initial dye concentrations and temperatures. Food dyes adsorption followed with the pseudo-second-order reaction, and equilibrium experiments were well fitted the Langmuir isotherm model. Maximum adsorption capacities of GMCNs displayed at pH 3.0 and at 25 °C, being up to 475.61 and 292.07 mg/g, for FD&C Blue 1 and D&C Yellow 5 respectively. Thermodynamic results demonstrated that the adsorption processes were spontaneous and exothermic. Furthermore, it was found that the GMCNs can be regenerated and reused through dye desorption in alkaline solution. Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction Food dyes as a member of food additives, are widely used as colorant in food, cosmetic and drug industries, to produce dyeing sweets, chewing gums, puddings, juices, mustard, sodas, drugs and cosmetics (Piccin et al., 2009). However, approximately 10– 20% of the dyes are lost during manufacturing process, result in large amounts of wastewater (Gao et al., 2011a). In addition, food dyes residues in waste effluents are adversely affect aquatic environment by coloring water and impeding light penetration, and are suspected to cause carcinogenic effects, hypersensitivity reactions and genotoxic effects on human health (Gao et al., 2011b; Kobylewski and Jacobson, 2012). Furthermore, that dyes in wastewater are difficult to be treated, because these dyes are recalcitrant molecules and resistant to aerobic digestion. Hence, numerous conventional treatment including coagulation, precipitation, membrane filtration, oxidation, adsorption, and photodegradation have been utilized to remove or eliminate dyes from wastewater (Crini and Badot, 2008; Robinson et al., 2001). Particularly, adsorption process is considered to be an effective and economical procedure to remove food dyes from industrial effluents. A large variety of adsorbent materials have been studied to reduce dye concentration in aqueous solution, such as activated ⇑ Correspondent authors. Tel./fax: +86 29 87092492 (T. Yue). Tel.: +1 848 9325536; fax: +1 732 9326776 (T.-C. Lee). E-mail addresses: [email protected] (T. Yue), [email protected] (T.-C. Lee). 0260-8774/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jfoodeng.2013.11.014

carbon (Ozsoy and van Leeuwen, 2010; Piccin et al., 2012), chitosan (Debrassi et al., 2012; Piccin et al., 2011), cellulose (Tabara et al., 2011), inorganic oxides (Siwinska-Stefanska et al., 2012) and anaerobic sludge (Yu et al., 2011). Chitosan, as an effective dye adsorbent, has received much scientific attention recently and been widely used for food dyes removal from wastewater due to its relatively low cost, high adsorption capacity and rate, the superiorities are ascribe to the presence of large amounts of amino groups (NH2) on chitosan molecules (Chen and Chen, 2009; Dotto and Pinto, 2011b; Kyzas and Lazaridis, 2009). However, chitosan is diffluent in dilute organic acids such as formic acid, acetic acid and the like, because its amino groups are fully protonated at approximately pH 3.0, and the polymer chains with positive charges fall apart in the solution. In this case, cross-linking modification of raw chitosan with epichlorohydrin or glutaraldehyde is a good way to improve the chemical stability of chitosan in acid solution, and the adsorption selectivity and capacity of food dyes from industrial effluents (Chen and Chen, 2009; Rosa et al., 2008). Although crosslinked chitosan has an enhanced stability and adsorption capacity of food dyes, their low separation efficiency in aqueous solution is still affect their industrial application scopes. Meanwhile, magnetic separation technology has attracted much attention for its easy separation procedure and high separation efficiency from aqueous solution by applying an external magnetic field (Rocher et al., 2010). In addition, nano-scaled magnetic chitosan particles, as the adsorbents for food dyes adsorption, have overcome the limitation of micro-scaled particles. The nano-scaled

134

Z. Zhou et al. / Journal of Food Engineering 126 (2014) 133–141

Nomenclature bF Ce Cf Ci E k1 k2 K KF KL Q QDR

adsorption intensity defined in Eq. (5) equilibrium concentration of dye (lg/mL) final concentration of dye (mg/mL) initial concentration of dye (mg/mL) mean adsorption energy defined in Eq. (8) (kJ/mol) pseudo-first-order rate constant defined in Eq. (2) (min1) pseudo-second-order rate constant defined in Eq. (3) (mg/g min) Dubinin–Radushkevich constant defined in Eq. (6) (J2/mol2) Freundlich constant (mg/g) Langmuir constant defined in Eq. (4) (L/mg) adsorption capacity (mg/g) Dubinin–Radushkevich maximum adsorption capacity of dye (mg/g)

particles contrarily provide large surface area to increase dyes adsorption capacity and decrease the internal diffusion resistance (Zhou et al., 2011). From the above, magnetic chitosan nanoparticles (MCNs) are chosen for food dyes adsorption. Crosslinkers such as epichlorohydrin (ECH) and glutaraldehyde (GA) can be used to modify MCNs to ensure their high adsorption capacity and good stability, Zhou et al. (2011) employed ethylenediamine to modify MCNs for acid dye adsorption purpose by using ECH as the crosslinker during preparation. However, the potential influence of crosslinker modified MCNs used for food dyes adsorption on environment and human health need to be considered. Therefore, GA was chosen as a modification crosslinker to prepare glutaraldehyde cross-linked magnetic chitosan nanoparticles (GMCNs) for its no carcinogenicity compared with epichlorohydrin. In addition, GA helps chitosan to form quaternary structure on the surface of MCNs, so as to increase the content of protonated amino groups ðNHþ 3 Þ to adsorb the anionic food dyes (Rosa et al., 2008). In the present study, the GMCNs were prepared through crosslinking modification of MCNs using GA as the crosslinker to facilitate the formation of chitosan quaternary structure on GMCNs, the effect of GA concentration on food dyes adsorption capacities and cytotoxicities of the GMCNs was comprehensive evaluated. In addition, the adsorption characteristics of FD&C Blue 1 and D&C Yellow 5 from aqueous solutions on GMCNs were investigated. Furthermore, the influence factors such as pH, dye concentrations and temperature to adsorption were studied; the equilibrium isotherms, thermodynamic parameters and regeneration were determined and discussed. 2. Methods 2.1. Chemicals and regents Ferric chloride (FeCl36H2O), ferrous chloride (FeCl24H2O), ammonia water (NH4OH), sodium hydroxide, acetic acid (99.7+%), cyclohexane, and sodium chloride were obtained from Fisher Scientific (Fair Lawn, NJ, USA). Chitosan (low molecular weight) were purchased from Sigma–Aldrich (St. Louis, MO, USA). Triton X-100, n-hexanol, glutaraldehyde (25%) were purchased from Alfa Aesar (Heysham, Lance, UK). RPMI-1640 media and dimethylsulfoxide (DMSO) were purchased from Thermo Scientific. FD&C Blue 1 and FD&C Yellow 5 were obtained from Emerald Performance Materials (Cincinnati, OH, USA) and their structures are presented in Fig. 1. All regents were analytical grade and used without further

Qe Qe,cal Qt Qm R R2 T V W

e DGo DHo DSo

adsorption capacities of dye at equilibrium (mg/g) calculated equilibrium adsorption capacities (mg/g) adsorption capacities of dye at a given time t (mg/g) Langmuir maximum adsorption capacity of dye (mg/g) universal gas constant (8.314 J/K mol) liner correlation coefficient temperature in Kelvin (K) volume of dye solution (mL) weight of the crosslinked magnetic chitosan nanoparticles used (g) Polanyl potential defined in Eq. (7) Gibbs free energy of the adsorption defined in Eq. (10) (kJ/mol) standard enthalpy changes defined in Eq. (9) (kJ/mol) standard entropy changes defined in Eq. (9) (J/mol K)

purification. Deionized experiments.

water

was

used

throughout

the

2.2. Preparation of the adsorbent The GMCNs were prepared starting from MCNs, and MCNs were prepared by two-step route using Triton X-100 reversed-phase water-in-oil microemulsion system (Zhou et al., 2013). Fe3O4 nanoparticles were prepared by coprecipitating. Cyclohexane, n-hexanol, 10 mg/mL chitosan 2% acetic acid solution and Fe3O4 turbid liquid were mixed (11:6:4:4) in a beaker at 1800 rpm, and then Triton X-100 was added to form a bright black-colored microemulsion. After with quickly addition of 20 mL 5 mol/L NaOH solution, mixture were placed at a 60 °C water bath for 2 h, then MCNs were collected with a magnet and rinsed with ethanol and then deionized water for three times. Finally, 1 mg/mL MCNs were modified by adding different concentrations of GA (0, 0.025, 0.075, 0.125, 0.25, 0.375, 0.5, 0.75 and 1 mol/mL) with a shaking of 150 rpm for 60 min at room temperature to obtain different GMCNs. The GMCNs were intensively washed by deionized water to remove unreacted GA, and then freeze-dried for further experiment. 2.3. Adsorption of food dye The different GMCNs were prepared by adding different concentrations of GA to study their adsorption capacities of two food dyes which are FD&C Blue 1 and FD&C Yellow 5. The adsorption processes were carried out by adding 50 mg GMCNs into 50 mL of 200 lg/mL dye solution at pH 6.6 with a shaking of 150 rpm for 60 min at room temperature. The adsorbents were removed by a 1 T magnet nearby (Sigma–Aldrich, St. Louis, MO, USA). The supernatant of dye solutions was adjusted to a pH level of 6.0 using HCl or NaOH solutions, the dye concentration was determined using spectrophotometry at wavelength of 408 nm for FD&C Blue 1 and 428 for FD&C Yellow 5 by a micro plate reader (Molecular Devices, Sunnyvale, CA). The adsorption capacity (Q) was calculated through following equation:



ðC i  C f ÞV W

ð1Þ

2.4. Cytotoxicity Human hepatocellular carcinoma cell line HepG2 was obtained from American Type Culture Collection (HB-8065, Manassas, VA)

Z. Zhou et al. / Journal of Food Engineering 126 (2014) 133–141

135

Fig. 1. Chemical structures of the dyes: (a) FD&C Blue 1 and (b) FD&C Yellow 5.

and maintained according to pervious procedure (Yu and Huang, 2013). Methyl thiazol tetrazolium bromide (MTT) assay (Mosmann, 1983) was used to evaluate the cytotoxicity of GMCNs. Briefly, 1  104 cells of HepG2 were seeded in each well of a 96well plate. After 24 h, the cells in each well were treated with fresh medium containing 100 lg/mL GMCNs of different glutaraldehyde concentration. The supernatant was removed and cells were washed three times with PBS after 24 h. Subsequently, 100 lL MTT solution (10% 5 mg/mL MTT agent with 90% RPMI-1640) was added to each well. After incubation for 2 h at 37 °C, MTT containing medium was carefully aspirated and the formed formazan crystals were dissolved with 100 lL DMSO per well. Absorbance of the converted dye was measured at a wavelength of 570 nm by a micro plate reader. Cells grown in medium without treatment were taken as negative control, relative cell viability was expressed as the comparison with negative control.

residual dye was measured using spectrophotometry and the amount of adsorption was calculated using Eq. (1). 2.7. Adsorption isotherms Optimum GMCNs were used to study their adsorption isotherms, this part was carried out by placing 50 mg GMCNs into 50 mL initial dye solutions among the concentration range of 50– 1500 lg/mL at the optimum pH values, and shake150 rpm for 60 min at definite temperatures (25, 35 and 45 °C). After adsorption, the GMCNs were separated by a magnet nearby, the supernatant was adjusted to pH 6.0 and the concentration of the residual dye was determined using spectrophotometry through Eq. (1). The dye adsorption capacities on GMCNs were calculated by the mass balance equation. 2.8. Regeneration

2.5. Characterization Optimum GA concentration was chosen based on the results of food dye adsorption performance and cytotoxicities of different GMCNs, and the GMCNs used for characterization evaluations and following adsorption studies were prepared by optimum GA concentration, the size of the GMCNs were observed by transmission electron microscope (TEM) (JEM 1230; JEOL, Tokyo, Japan). Zeta potential measurements were performed using a zeta potential analyzer (Zeta Plus, Brookhaven Instruments, Huntsville, NY) in deionized water. Chitosan content of the GMCNs was determined by a thermal gravimetric analyzer (TA-Q100, TA Instruments, Inc., USA). Fourier transform infrared (FTIR) spectra before and after food dye adsorption were recorded using a Thermo Nicolet 670 FTIR Spectrometer (Thermo Electron Corp., Madison, WI) at ambient temperature. All of the samples were freeze-dried prior to measurement.

2.6. Adsorption kinetics The GMCNs were prepared using optimum GA concentration, and the adsorption kinetics of the two food dyes on GMCNs was conducted in a batch process. Variable parameters including initial pH values, initial dye concentrations and temperatures were studied. In each test, 50 mg of the GMCNs was added into 50 mL dye solutions with a known concentration. The solution pH was maintained at the desired value with HCl or NaOH standardized solutions. Afterwards, 0.5 mL dye solution was taken at different time intervals, and a magnet was used to separate solid phase from aqueous phase so that the supernatant was collected, the solution was adjusted to pH 6.0 before measurement. The concentration of

Regeneration of adsorbents were studied by placing 50 mg optimum GMCNs into 50 mL food dye solutions (800 lg/mL, pH 3.0) with a shaking of 150 rpm for 120 min at 25 °C. To desorb, dye loaded GMCNs were collected and removed from the solution by a magnet nearby, and then dye loaded GMCNs were agitated with 50 mL of NaOH solution (pH 10.0) for 120 min at 25 °C. The reusability of the adsorbents, which is a potential economical parameter was studied by reusing the desorbed GMCNs in adsorption experiments and the process was repeated for three times. The amount of dye adsorption and desorption from the GMCNs were calculated by the absorbance value of dye solutions. 3. Results and discussion 3.1. Adsorption As shown in Fig. 2(a), dye adsorption capacities of the GMCNs increased with GA concentrations, this may be attributed to the cross-linking reaction with chitosan in favor of quaternary structure formation and coating on the surface of the GMCNs (Rosa et al., 2008), followed by increasing dye loading capacities of the GMCNs effectively. Additionally, the growth of dye adsorption capacities was slowed when GA concentration arrived to 0.5 mol/ L. The possible reason was that the GA used for forming chitosan quaternary structure started to be saturate, so that less quaternary structure was formed, thus limited dye adsorption capacities of the GMCNs. Meanwhile, the GMCNs, as a novel reusable adsorbent for food dyes, their potential influence on aqueous environment and human health was considered. Cytotoxicity assays was the initial step to study how they will react in our body, has always

136

Z. Zhou et al. / Journal of Food Engineering 126 (2014) 133–141

Fig. 3. TEM micrograph of the GMCNs.

Fig. 2. Effect of GA concentration on (a) the adsorption of FD&C Blue 1 and FD&C Yellow 5 by GMCNs and (b) the cytotoxicity of GMCNs by using HepG2 cells.

been used for potential toxicity studies (Lewinski et al., 2008). And the surface coating types of samples directly impacted the cell behaviors and morphology under the same exposure levels (Singha et al., 2010). Therefore, in vitro cytotoxicity evaluation by MTT assay using HepG2 cells were employed to assess the cytotoxicity of the GMCNs with different GA concentrations, and the cell viability after 24 h exposure to the GMCNs was shown in Fig. 2(b). Nearly 0.5 mol/mL of GA used for GMCNs cross-linking showed no significant cytotoxicity, and the cytotoxicity grading was qualified and classified into eligible reaction (Cheng et al., 2009). However, the GMCNs exposure exhibited a significant cytotoxicity at higher dose of GA (>1.0 mol/mL). It was due to the fact that GA concentration used for forming chitosan quaternary structure started to be saturate when the concentration were greater than 0.5 mol/mL. So that, at higher concentration, unreacted aldehyde groups of GA were exposed to interact with the cell surface that resulted in the death of the cells (Gough et al., 2002). Above all, 0.5 mol/mL GA was chosen for cross-linking to ensure that the prepared GMCNs had excellent performance on food dye adsorption, low toxicity and no secondary pollution to aqueous environment.

3.2. Characterization The GMCNs were prepared with 0.5 mol/mL GA, their size distribution ranging from 60 to 95 nm based on TEM micrograph (Fig. 3).

Fig. 4. FT-IR analysis: (a) GMCNs, (b) GMCNs adsorbed with FD&C Blue 1, and (c) GMCNs adsorbed with FD&C Yellow 5.

The GMCNs revealed a zeta potential of +22.82 mV, the positive charge of the zeta potential indicated that the GMCNs were stabilized by hydrogen bonds between the amino and hydroxyl groups of chitosan and the hydroxyl group and oxygen atom of water (Hu et al., 2006). Average mass content of chitosan in GMCNs was approximately 50 wt%, as calculated from the data at 700 °C using TGA. The FT-IR spectra of the GMCNs before dye adsorption (Fig. 4a) showed that the characteristic absorption bands appeared at 580.47 and 2719.14 cm1, which corresponded to Fe–O bond of Fe3O4 and aldehyde groups, respectively. The adsorption peaks of the GMCNs at 1035.59 cm1 and 1112.73 cm1 (C–N bond), 1376.93 cm1 (–C–O– stretching of primary alcoholic group), 1571.70 cm1 (–NH2 in amide group), 1641.13 cm1 (amide I band) and 3359.39 cm1 (N–H bond) can also be observed. The results of FT-IR spectra indicated that the MCNs were crosslinked by glutaraldehyde successfully. After the adsorption process of food dyes, it can be observed in Fig. 4(b) and (c) that the absence of sharp peaks at 1035.59 cm1 and 1112.73 cm1 confirmed that the amine group of chitosan changed to become a protonated amine (Sakkayawong et al., 2005). Furthermore, the peaks at 1571.70 cm1 and 1641.13 cm1 were shifted to peaks at 1579.41 cm1 and 1650.77 cm1, respectively. In addition, a new peak appeared at 1342.21 cm1, which indicated that the peak in this area is a sulfonated group in the ring of the dye (Dotto and Pinto, 2011a). These

Z. Zhou et al. / Journal of Food Engineering 126 (2014) 133–141

137

Fig. 5. Effect of initial pH on the adsorption of FD&C Blue 1 and FD&C Yellow 5 on GMCNs.

peaks change and a new peak confirmed the attachment of food dyes sulfonate groups on the chitosan amino groups. 3.3. Effect of pH on dye adsorption The adsorption capacities of both food dyes on GMCNs decreased with the increase of the initial pH values (Fig. 5), changed significantly from pH 3.0 to pH 6.0, and then decreased slightly when pH value exceeded 6.0. The adsorption capacities of FD&C Blue 1 and FD&C Yellow 5 decreased from 375.15 to 53.26 mg/g and from 247.87 to 38.26 mg/g, respectively, with the initial pH values of dye solution increased from 3.0 to 11.0. The sulfonate groups of food dyes (–SO3Na) were dissociated and converted to anionic dye ions ðSO 3 Þ in aqueous solution, the amino groups of the GMCNs (–NH2) became protonated in the presence of H+ at a lower pH solution, which increased the electrostatic attractions between the anionic groups ðSO 3 Þ of food dyes and the protonated amino groups ðNHþ 3 Þ of the GMCNs, thus caused an increase in dye adsorption. In addition, there was a lower dye adsorption at a higher pH solution, which may be ascribed to the number of protonated –NH2 groups decreased, and more hydroxide ions (OH–) were available to the ionic repulsion occurred between negative charged surface of the GMCNs and the anionic dye molecules (Chen and Chen, 2009; Yoshida et al., 1993). 3.4. Dye adsorption kinetic Fig. 6 showed the adsorption capacities of food dyes on GMCNs with initial dye concentrations of 300, 500, 800 and 1500 lg/mL at pH 3.0 and 25 °C. The initial adsorption rate of food dye on GMCNs showed a remarkable increase during the first 20 min, and then the GMCNs gradually approached the adsorption limit after 60 min. The adsorption kinetics revealed the same results, which may be ascribed to more adsorption vacant sites on GMCNs were available at the adsorption beginning, and high concentration gradient existed between adsorbate in solution and adsorbate in the adsorbent. Subsequently, the concentration gradient was reduced due to the fact that most of the vacant sites were occupied by adsorbed dye molecules, which led to a slightly increase of adsorption during the later stages (Vijayaraghavan and Yun, 2008). On the other hand, along with an increase in the initial concentration of food dyes from 300 to 1500 lg/mL, the adsorption capacities of FD&C Blue 1 (Fig. 6a) and FD&C Yellow 5 (Fig. 6b) on GMCNs increased from 285.63 to 475.61 mg/g and from 236.28 to 293.38 mg/g, respectively. However, the dye removal percentage of FD&C Blue

Fig. 6. Adsorption kinetic of (a) FD&C Blue 1 and (b) FD&C Yellow 5 on GMCNs with different initial dye concentrations.

1 and FD&C Yellow 5 decreased from 95.21% to 31.71% and from 78.76% to 19.56%, respectively. The possible reason was that the ratio of the initial number of dye molecules to the available surface area of the GMCNs was low at low initial concentration and subsequently the fractional adsorption became independent of initial concentration. Moreover, the available adsorption sites were fewer than the initial dye molecules present, which resulted in the decrease of removal percentage. Fig. 7 showed the effect of initial temperatures (25, 35 and 45 °C) on adsorption kinetics at 1500 lg/mL initial dye concentration, pH 3.0. It was observed that the adsorption capacity of FD&C Blue 1 on GMCNs (Fig. 7a) was higher than that of FD&C Yellow 5 (Fig. 7b). This results may be ascribed to the vacant sites on GMCNs used for adsorption were constant, and the molecular weight of FD&C Blue 1 was higher than that of FD&C Yellow 5. Meanwhile, with the initial temperatures of dye solution increased from 25 to 45 °C, the adsorption capacities of FD&C Blue 1 and FD&C Yellow 5 decreased from 475.61 to 399.60 mg/g and from 293.38 to 242.90 mg/g, respectively. The increase of initial temperature may causes an increase in the solubility of the dyes (Crini and Badot, 2008), resulted in a stronger interaction forces between dyes and solvent than those between dyes and nanoparticles (Dotto et al., 2012a; Zhou et al., 2011). To measure the rate-controlling and mass transfer mechanism of food dye adsorption on GMCNs, kinetic data were calculated by the pseudo-first-order and pseudo-second-order kinetic models (Rosa et al., 2008).

138

Z. Zhou et al. / Journal of Food Engineering 126 (2014) 133–141

calculated by the pseudo-first-order and pseudo-second-order kinetic models. The R2 of the pseudo-second-order adsorption kinetic model presented higher values than those of the pseudo-first-order one, and their calculated equilibrium adsorption capacities (Qe,cal) fitted the experimental Qe values well. The results indicated that the pseudo-second-order adsorption kinetic model was more appropriate for dye adsorption process than the pseudo-first-order one. 3.5. Dye adsorption isotherms The adsorption isotherms of food dyes on GMCNs at different temperatures (pH 3.0) for 180 min were presented in Fig. 8. The equilibrium adsorption capacities (Qe) increased along with an increase of the dye concentration, and decreased with increasing temperature. The adsorption isotherms were studied using three isotherm models (Chen et al., 2011; Ismail et al., 2013): Langmuir isotherm equation is presented by the following equation

Ce Ce 1 ¼ þ Q e Q m Q mKL

ð4Þ

Freundlich isotherm equation is described through the following equation

ln Q e ¼ bF ln C e þ ln K F

ð5Þ

and Dubinin–Radushkevich isotherm equation is given by the following equation

ln Q e ¼ K e2 þ ln Q DR

ð6Þ

Polanyl potential (e) given as the following equation



e ¼ RT ln 1 þ

Fig. 7. Adsorption kinetic of (a) FD&C Blue 1 and (b) FD&C Yellow 5 on GMCNs with different initial temperatures.

1 Ce

 ð7Þ

and the Dubinin–Radushkevich constant (K) can give the valuable information regarding the mean energy of adsorption by the following equation

E ¼ ð2KÞ1=2 The pseudo-first-order kinetic is presented as the following equation

logðQ e  Q t Þ ¼ log Q e 

k1 t 2:303

ð2Þ

and the pseudo-second-order kinetic is presented as the following equation

t 1 t ¼ þ Q t k2 Q 2e Q e

ð3Þ

Tables 1 and 2 showed the rate constants and correlation coefficients (R2) for both food dyes on GMCNs, respectively, which were

ð8Þ

The parameters of adsorption isotherm models for both food dyes were listed in Table 3. The Langmuir adsorption model was found to fit the experimental data for both dyes sufficiently in accordance with the liner correlation coefficients (R2). It indicates the homogeneity of active sites on the surface of the GMCNs, was provided by the chitosan on GMCNs. Meanwhile, the difference of KL between FD&C Blue 1 and FD&C Yellow 5 refers to the different in binding strength and capacity of the dyes with the surface of the GMCNs (Zhou et al., 2011). In addition, the mean adsorption energy (E) calculated from the Dubinin–Radushkevich isotherm, defined as the free energy for per mol adsorbate in transfer from infinity in solution to the surface.

Table 1 The pseudo-first-order and pseudo-second-order adsorption constants, calculated and experimental Qe values for different initial dye concentrations, pH, and temperatures for FD&C Blue 1 on GMCNs. Parameter

Qe (mg/g)

Pseudo-first-order

Pseudo-second-order Qe,cal (mg/g)

R2

k2 (103) (mg/g min)

Qe,cal (mg/g)

R2

Initial dye concentration (lg/mL) t = 180 min (pH 3.0, 25 °C) 300 285.63 2.5054 500 401.52 2.4324 800 445.59 3.1024 1500 475.61 2.5382

84.85 138.80 180.33 147.46

0.8033 0.8943 0.9358 0.8527

1.18 0.71 0.64 0.75

288.38 404.99 451.42 478.31

0.9996 0.9993 0.9993 0.9994

Temperature (°C) t = 180 min (pH 3.0, 1500 lg/mL) 25 475.61 2.5382 35 449.39 2.0196 45 399.60 1.9412

147.46 139.73 107.10

0.8527 0.8417 0.8017

0.75 0.62 0.83

478.31 450.78 400.06

0.9994 0.9987 0.9990

k1 (102) (min1)

139

Z. Zhou et al. / Journal of Food Engineering 126 (2014) 133–141

Table 2 The pseudo-first-order and pseudo-second-order adsorption constants, calculated and experimental Qe values for different initial dye concentrations, pH, and temperatures for FD&C Yellow 5 on GMCN. Parameter

Qe (mg/g)

Pseudo-first-order

Pseudo-second-order Qe,cal (mg/g)

R2

k2 (103) (mg/g min)

Qe,cal (mg/g)

R2

Initial dye concentration (lg/mL) t = 180 min (pH 3.0, 25 °C) 300 236.28 4.3809 500 259.51 2.2455 800 273.46 3.5586 1500 293.38 2.2122

39.54 74.87 75.03 83.51

0.8308 0.7996 0.8800 0.8190

4.09 1.36 1.80 1.22

237.90 260.69 276.44 294.48

0.9999 0.9995 0.9998 0.9995

Temperature (°C) t = 180 min (pH 3.0, 300 lg/mL) 25 293.38 2.2122 35 272.52 2.0909 45 242.90 1.9126

83.51 108.83 108.80

0.8190 0.8704 0.8899

1.22 0.76 0.68

294.48 274.94 244.82

0.9995 0.9988 0.9977

k1 (102) (min1)

respectively, which indicated that the GMCNs can be used as a potential absorbent to treat food dyes. It was notable that the GMCNs combined by magnetic Fe3O4 nanoparticles and chitosan, while chitosan only accounted for approximately 50 wt%, which was responsible for the food dyes adsorption. High adsorption capacity of food dyes may be ascribe to the specific chitosan structure on the surface of the GMCNs, leading to almost all adsorption vacant sites available. Moreover, the GMCNs adsorbents are economical and can be easily separated from the solution by applying a magnetic field. Therefore, the developed GMCNs in this study facilitated a higher adsorption capacity and offer a more efficient adsorption of food dyes in an aqueous solution compared with other chitosan-based adsorbents.

3.6. Dye adsorption thermodynamics The adsorption thermodynamic parameters for FD&C Blue 1 and FD&C Yellow 5 adsorption process on GMCNs were calculated according to the following van’t Hoff equation (Elwakeel, 2009), as shown by the following equation:

ln K L ¼ 

DH o DSo þ RT R

ð9Þ

The value of DHo and DSo were calculated and listed in Table 5. The negative value of DHo indicated that the adsorption process on GMCNs was exothermic. The positive value of DSo suggested the increased randomness during the adsorption of both food dyes, which may be due to liberation of water molecules from the hydrated shells of the sorbed species (Chen and Chen, 2009). Gibbs free energy of the adsorption (DGo) was obtained by Eq. (10) and also reported in Table 5.

DGo ¼ DHo  T DSo Fig. 8. Equilibrium adsorption of (a) FD&C Blue 1 on GMCNs and (b) FD&C Yellow 5 on GMCNs.

The adsorption behavior can be used to predict the physical adsorption in the range of 1–8 kJ/mol and the chemical adsorption in more than 8 kJ/mol (Chen and Chen, 2009). The free energy E of two food dyes indicated that the adsorption may be due to the dual process of physisorption and chemisorption, and the adsorptions were predominant on the chemisorption process. Table 4 listed the comparison results of the maximum adsorption capacity (Qm) of food dyes on chitosan-based adsorbents. The Qm obtained by Langmuir isotherm for FD&C Blue 1 and FD&C Yellow 5 adsorption was 475.61 and 292.07 mg/g,

ð10Þ

The negative values of DGo for both food dyes demonstrated that the adsorption process on GMCNs was a spontaneous process, and the decrease of DGo values with the increase of temperature indicated that the adsorption became less favorable at higher temperature (Zhou et al., 2011). The increased temperature promoted the mobility of dye molecules and caused the escape of dye molecules from the solid phase to the liquid phase. Therefore, the adsorption capacity of food dyes decreased with the increase of solution temperature. In addition, the values of KL (Table 3) may also reflect that the mobility of dye molecules increased with increasing of temperature. The KL values decreased with the increase of the temperature revealed the lower affinity of the GMCNs to both food dyes at higher temperature.

140

Z. Zhou et al. / Journal of Food Engineering 126 (2014) 133–141

Table 3 Langmuir, Freundlich and Dubinin-Radushkevich isotherms constants for adsorption of FD&C Blue 1 and FD&C Yellow 5 at different temperature. Parameter

Langmuir Qm (mg/g)

Freundlich KL (L/mg)

2

R

Dubinin-Radushikevich QDR (mg/g)

K  108 (J2/mol2)

E (kJ/mol)

R2

0.9933 0.9900 0.9789

313.46 300.34 287.80

0.36 0.48 0.74

11.84 10.17 8.21

0.7710 0.7641 0.8129

0.9852 0.9822 0.9920

223.80 204.71 188.39

0.58 0.53 0.69

9.31 9.69 8.54

0.7988 0.7260 0.7504

KF (mg/g)

bF

R

FD&C Blue 1 Temperature (°C) t = 180 min (pH 3.0) 25 475.61 0.144 0.9993 35 449.15 0.130 0.9992 45 400.79 0.124 0.9996

191.29 169.46 145.40

0.140 0.152 0.160

FD&C Yellow 5 Temperature (°C) t = 180 min (pH 3.0) 25 292.07 0.099 0.9987 35 271.97 0.097 0.9990 45 242.76 0.091 0.9992

128.72 116.56 103.48

0.129 0.131 0.131

2

Table 4 Comparison of the maximum adsorption capacity (Qm) of food dyes on chitosan based adsorbents. Food dye

Adsorbents

Qm (mg/g)

Reference

FD&C FD&C FD&C FD&C FD&C FD&C FD&C FD&C

GA-crosslinked magnetic chitosan nanoparticles GA-crosslinked magnetic chitosan nanoparticles Chitosan film Chitosan film Chitosan powder Chitosan powder Chitosan powder Chitosan powder

475.61 292.07 194.6 154.8 350 210 295 529.0

This work This work Dotto et al. (2013) Dotto et al. (2013) Dotto et al. (2012b) Dotto and Pinto (2011b) Dotto and Pinto (2011b) Piccin et al. (2009)

Blue 1 Yellow 5 Red 2 Blue 2 Yellow 5 Blue 1 Yellow 5 Red 40

Table 5 The thermodynamic parameters and activation energies for adsorption of FD&C Blue 1 and FD&C Yellow 5 on the GMCNs.

FD&C Blue 1

FD&C Yellow 5

Temperature (K)

DGo (kJ/mol)

DHo (kJ/mol)

DSo (J/mol K)

298 308 318

12.31 12.45 12.75

5.76

21.89

298 308 318

11.39 11.71 11.92

3.49

26.56

Fig. 9. Adsorption and desorption of FD&C Blue 1 and FD&C Yellow 5 on GMCNs.

272.16 mg/g, respectively. After the desorption step for 120 min, the adsorbed FD&C Blue 1 and FD&C Yellow 5 were removed by NaOH solution at pH 10.0, and the amounts were approximately 57.15% and 54.29%, respectively. The results may be ascribed to the fact that, in the basic solution, the positively charged amino groups were deprotonated and the electrostatic interaction between chitosan and dye molecules became much weaker (Cestari et al., 2004; Chen et al., 2011). At the same time, the incomplete desorption of food dyes on GMCNs may due to the dual process of physisorption and chemisorption (Chen et al., 2011). The second and the third adsorption step (120 min) revealed a similar dynamical shape of the first adsorption step, and maintained more than 90% of adsorption capacities against the first adsorption step. The adsorption capacities of FD&C Blue 1 and D&C Yellow 5 were maintained 95.30% and 93.77% at the second adsorption step, respectively. And then decreased to 93.31% and 90.60% for both food dyes at the third adsorption step, respectively. Therefore, the GMCNs can be regenerated and reused for further food dye adsorption.

3.7. Regeneration

4. Conclusions

The GMCNs regeneration was shown in Fig. 9. The adsorption condition was 800 lg/mL initial dye concentration with an initial pH of 3.0 at 25 °C, whereas the desorption condition changed to an initial pH of 10.0 of the NaOH solution and at 25 °C. It was observed that after the first adsorption step for 120 min, the adsorption of both food dyes on GMCNs reached to 441.69 and

The GMCNs were prepared through GA cross-linking modification of MCNs, by comprehensively considering the effect of GA concentration on their adsorption capacities and cytotoxicities. Concentration of 0.5 mol/mL GA ensured the as-prepared GMCNs had excellent performance on food dye adsorption in aqueous solution and low cytotoxicity. Food dyes adsorption accurately in

Z. Zhou et al. / Journal of Food Engineering 126 (2014) 133–141

accordance with the second-order adsorption processes and fitted well with the Langmuir isotherm model. The adsorption capacities of the GMCNs for both food dyes increased with decreasing pH values and decreased with increasing temperature, the maximum adsorption of FD&C Blue 1 and D&C Yellow 5 on GMCNs were greater than previously reported chitosan-based adsorbents, up to 475.61 and 292.07 mg/g, respectively. Meanwhile, the mean adsorption energy (E) from the Dubinnin–Radushkevich isotherm revealed that the adsorption process may be the dual process of physisorption and chemisorption, and was predominant on the chemisorption process. In addition, adsorption process was spontaneous and exothermic that can be desorbed efficiently using alkaline solution, thus the GMCNs can be recycled for dye removal. The technique used in this study offered a convenient and efficient method for the preparation of the GMCNs, which facilitated a more safety, efficient and economic adsorption of food dye from aqueous solution and avoided secondary pollution of adsorbents to water environment. Acknowledgements We thank Dr. Qingrong Huang in Department of Food Science, Rutgers University for his insightful discussions and generously providing us access to the instruments in his lab. References Cestari, A.R., Vieira, E.F.S., dos Santos, A.G.P., Mota, J.A., de Almeida, V.P., 2004. Adsorption of anionic dyes on chitosan beads. 1. The influence of the chemical structures of dyes and temperature on the adsorption kinetics. J. Colloid Interface Sci. 280 (2), 380–386. Chen, A.H., Chen, S.M., 2009. Biosorption of azo dyes from aqueous solution by glutaraldehyde-crosslinked chitosans. J. Hazard. Mater. 172 (2–3), 1111–1121. Chen, C.-Y., Chang, J.C., Chen, A.H., 2011. Competitive biosorption of azo dyes from aqueous solution on the templated crosslinked-chitosan nanoparticles. J. Hazard. Mater. 185 (1), 430–441. Cheng, J., Wu, W.W., Chen, B.A., Gao, F., Xu, W.L., Gao, C., Ding, J.H., Sun, Y.Y., Song, H.H., Bao, W., Sun, X.C., Xu, C.R., Chen, W.J., Chen, N.N., Liu, L.J., Xia, G.H., Li, X.M., Wang, X.M., 2009. Effect of magnetic nanoparticles of Fe3O4 and 5bromotetrandrine on reversal of multidrug resistance in K562/A02 leukemic cells. Int. J. Nanomed. 4, 209–216. Crini, G., Badot, P.M., 2008. Application of chitosan, a natural aminopolysaccharide, for dye removal from aqueous solutions by adsorption processes using batch studies: a review of recent literature. Prog. Polym. Sci. 33 (4), 399–447. Debrassi, A., Correa, A.F., Baccarin, T., Nedelko, N., Slawska-Waniewska, A., Sobczak, K., Dluzewski, P., Greneche, J.M., Rodrigues, C.A., 2012. Removal of cationic dyes from aqueous solutions using N-benzyl-O-carboxymethylchitosan magnetic nanoparticles. Chem. Eng. J. 183, 284–293. Dotto, G.L., Pinto, L.A.A., 2011a. Adsorption of food dyes acid blue 9 and food yellow 3 onto chitosan: stirring rate effect in kinetics and mechanism. J. Hazard. Mater. 187 (1–3), 164–170. Dotto, G.L., Pinto, L.A.A., 2011b. Adsorption of food dyes onto chitosan: optimization process and kinetic. Carbohydr. Polym. 84 (1), 231–238. Dotto, G.L., Lima, E.G., Pinto, L.A.A., 2012a. Biosorption of food dyes onto Spirulina platensis nanoparticles: equilibrium isotherm and thermodynamic analysis. Bioresour. Technol. 103 (1), 123–130. Dotto, G.L., Vieira, M.L.G., Pinto, L.A.A., 2012b. Kinetics and Mechanism of Tartrazine Adsorption onto Chitin and Chitosan. Industrial Engineering Chemistry Research 51 (19), 6862–6868. Dotto, G.L., Moura, J.M., et al., 2013. Application of chitosan films for the removal of food dyes from aqueous solutions by adsorption. Chem. Eng. J. 214, 8–16. Elwakeel, K.Z., 2009. Removal of Reactive Black 5 from aqueous solutions using magnetic chitosan resins. J. Hazard. Mater. 167 (1–3), 383–392.

141

Gao, J.F., Zhang, Q., Wang, J.H., Wu, X.L., Wang, S.Y., Peng, Y.Z., 2011a. Contributions of functional groups and extracellular polymeric substances on the biosorption of dyes by aerobic granules. Bioresour. Technol. 102 (2), 805–813. Gao, Y.L., Li, C.M., Shen, J.Y., Yin, H.X., An, X.L., Jin, H.Z., 2011b. Effect of food azo dye tartrazine on learning and memory functions in mice and rats, and the possible mechanisms involved. J. Food Sci. 76 (6), T125–T129. Gough, J.E., Scotchford, C.A., Downes, S., 2002. Cytotoxicity of glutaraldehyde crosslinked collagen/poly(vinyl alcohol) films is by the mechanism of apoptosis. J. Biomed. Mater. Res. 61 (1), 121–130. Hu, Z.G., Zhang, J., Chan, W.L., Szeto, Y.S., 2006. The sorption of acid dye onto chitosan nanoparticles. Polymer 47 (16), 5838–5842. Ismail, B., Hussain, S.T., Akram, S., 2013. Adsorption of methylene blue onto spinel magnesium aluminate nanoparticles: adsorption isotherms, kinetic and thermodynamic studies. Chem. Eng. J. 219, 395–402. Kobylewski, S., Jacobson, M.F., 2012. Toxicology of food dyes. Int. J. Occupational Environ. Health 18 (3), 220–246. Kyzas, G.Z., Lazaridis, N.K., 2009. Reactive and basic dyes removal by sorption onto chitosan derivatives. J. Colloid Interface Sci. 331 (1), 32–39. Lewinski, N., Colvin, V., Drezek, R., 2008. Cytotoxicity of nanoparticles. Small 4 (1), 26–49. Mosmann, T., 1983. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J. Immunol. Methods 65 (1– 2), 55–63. Ozsoy, H.D., van Leeuwen, J., 2010. Removal of color from fruit candy waste by activated carbon adsorption. J. Food Eng. 101 (1), 106–112. Piccin, J.S., Dotto, G.L., Vieira, M.L.G., Pinto, L.A.A., 2011. Kinetics and mechanism of the food dye FD&C Red 40 adsorption onto chitosan. J. Chem. Eng. Data 56 (10), 3759–3765. Piccin, J.S., Gomes, C.S., Feris, L.A., Gutterres, M., 2012. Kinetics and isotherms of leather dye adsorption by tannery solid waste. Chem. Eng. J. 183, 30–38. Piccin, J.S., Vieira, M.L.G., Goncalves, J.O., Dotto, G.L., Pinto, L.A.A., 2009. Adsorption of FD&C Red No. 40 by chitosan: isotherms analysis. J. Food Eng. 95 (1), 16–20. Robinson, T., McMullan, G., Marchant, R., Nigam, P., 2001. Remediation of dyes in textile effluent: a critical review on current treatment technologies with a proposed alternative. Bioresour. Technol. 77 (3), 247–255. Rocher, V., Bee, A., Siaugue, J.M., Cabuil, V., 2010. Dye removal from aqueous solution by magnetic alginate beads crosslinked with epichlorohydrin. J. Hazard. Mater. 178 (1–3), 434–439. Rosa, S., Laranjeira, M.C.M., Riela, H.G., Fávere, V.T., 2008. Cross-linked quaternary chitosan as an adsorbent for the removal of the reactive dye from aqueous solutions. J. Hazard. Mater. 155 (1–2), 253–260. Sakkayawong, N., Thiravetyan, P., Nakbanpote, W., 2005. Adsorption mechanism of synthetic reactive dye wastewater by chitosan. J. Colloid Interface Sci. 286 (1), 36–42. Singha, N.J.S.G., Jenkinsa, Asadib, R., Doaka, S.H., 2010. Potential toxicity of superparamagnetic iron oxide nanoparticles (SPION). Nano Rev. 1, 5358–5372. Siwinska-Stefanska, K., Nowacka, M., Kolodziejczak-Radzimska, A., Jesionowski, T., 2012. Preparation of hybrid pigments via adsorption of selected food dyes onto inorganic oxides based on anatase titanium dioxide. Dyes Pigm. 94 (2), 338– 348. Tabara, A., Yamane, C., Abe, M., Seguchi, M., 2011. Adsorption of xanthene food additive dyes to cellulose granules. Cellulose 18 (1), 45–55. Vijayaraghavan, K., Yun, Y.S., 2008. Biosorption of C.I. Reactive Black 5 from aqueous solution using acid-treated biomass of brown seaweed Laminaria sp. Dyes Pigm. 76 (3), 726–732. Yoshida, H., Okamoto, A., Kataoka, T., 1993. Adsorption of acid dye on cross-linked chitosan fibers: equilibria. Chem. Eng. Sci. 48 (12), 2267–2272. Yu, H., Huang, Q., 2013. Investigation of the cytotoxicity of food-grade nanoemulsions in Caco-2 cell monolayers and HepG2 cells. Food Chem. 141 (1), 29–33. Yu, L., Li, W.W., Lam, M.H.W., Yu, H.Q., 2011. Adsorption and decolorization kinetics of methyl orange by anaerobic sludge. Appl. Microbiol. Biotechnol. 90 (3), 1119–1127. Zhou, L., Jin, J., Liu, Z., Liang, X., Shang, C., 2011. Adsorption of acid dyes from aqueous solutions by the ethylenediamine-modified magnetic chitosan nanoparticles. J. Hazard. Mater. 185 (2–3), 1045–1052. Zhou, Z., Jiang, F., Lee, T.C., Yue, T., 2013. Two-step preparation of nano-scaled magnetic chitosan particles using Triton X-100 reversed-phase water-in-oil microemulsion system. J. Alloy. Compd. 581, 843–848.