Environ Sci Pollut Res (2014) 21:5086–5097 DOI 10.1007/s11356-013-2452-9
RESEARCH ARTICLE
Predictive modeling of sorption and desorption of a reactive azo dye by pumpkin husk Abuzer Çelekli & Fadime Çelekli & Erdoğan Çiçek & Hüseyin Bozkurt
Received: 14 June 2013 / Accepted: 10 December 2013 / Published online: 28 December 2013 # Springer-Verlag Berlin Heidelberg 2013
Abstract The use of effective disposal of redundant pumpkin husk (PH) to remove pollutants is an important issue for environmental protection and utilization of resource. The aim of this study was to remove a potentially toxic reactive azo dye, Reactive Red (RR) 120, by widespread PH as a lowcost adsorbent. Particle size, adsorbent dose, pH, temperature, initial dye concentration, and contact time affected the sorption process. Amine, amide, hydroxyl, and carboxyl groups of PH played significant roles on the sorption process. Rapid sorption occurred within the first 2 min and equilibrium was reached within 60 min. Sorption kinetic was well represented by logistic equation. Generated secondary logistic model can be used to describe effects of initial dye concentration, contact time, and temperature by a single equation with high R2 value. Monolayer sorption capacity was found as 98.61 mg g-1. Activation energy, thermodynamic, and desorption studies showed that this process was physical, endothermic, and spontaneous. This study indicated that redundant PH as a low-cost adsorbent had a great potential for the removal of RR 120 as an alternative eco-friendly process. Keywords Activation energy . Predictive modeling . Pumpkin husk . Reactive Red 120 . Sorption
Responsible editor: Michael Matthies A. Çelekli (*) Department of Biology, Faculty of Art and Science, University of Gaziantep, 27310 Gaziantep, Turkey e-mail: [email protected] F. Çelekli : E. Çiçek Department of Biology, Faculty of Art and Science, University of Nevşehir Hacı Bektaş Veli, 50300 Nevşehir, Turkey H. Bozkurt Department of Food Engineering, Faculty of Engineering, University of Gaziantep, 27310 Gaziantep, Turkey
Introduction Synthetic dyes are extensively used in textile, paper, pharmaceutical, cosmetics and food industries which generate huge volumes of wastewater. The annual production of the synthetic dyes is about 0.7 Mt and reactive azo dyes are about 50 60 % of them (Phillips 1996; Solís et al. 2012). Reactive azo dyes contain one or more azo bonds (−N=N–), combined with aromatic and heterocyclic groups. They are generally resistant to fading on exposure to light, water and many chemicals due to their chemical structure (Ali 2010; Solís et al. 2012; Wang et al. 2013). Reactive dyes are mainly applied in textile processing, due to the ease and low cost in their use, stability and availability of various colors (Çelekli et al. 2009; Wang et al. 2013). About 15–30 % of these dyes remain in the effluents for coloring of final products in textile industry. These effluents contain not only dyes, but also metals, salts, and other chemicals. In order to color 1 kg of cotton with reactive dyes, 30–60 g dyestuff, 70–150 l water, and 0.6–0.8 kg NaCl are required. Throughout world, 280,000 tons of textile dyes are discharged with industrial effluents every year (Ali 2010; Solís et al. 2012). Discharging of unfixed dyes into receiving water may cause a negative impact on the ecosystems, such as reduction of photosynthetic activity, causing of aesthetic problems, and toxicity to life (Saratale et al. 2011; Daneshvar et al. 2012). Besides, once exposed, reactive azo dyes can cause variety of diseases and disorders in living organisms, even at low concentrations, such as allergy, skin irritation, and cancer (Brookstein 2009; Gupta and Suhas 2009; Salleh et al. 2011). Therefore, it is necessary to remove these recalcitrant compounds from wastewater, prior to discharging into aquatic ecosystems. A wide range of technologies have been developed to remove these dyes from wastewaters (Ali 2010; Srinivasan and Viraraghavan 2010; Salleh et al. 2011; Gupta et al. 2012). Biodegradation of reactive dyes is usually very difficult due to
Environ Sci Pollut Res (2014) 21:5086–5097
their complex molecular structure. Therefore, they cannot be easily remove from wastewater by conventional coagulation or activated sludge methods. Adsorption is an attractive, cheap, and effective method for the treatment of dye-bearing effluents. Researchers try to optimize adsorption process and develop novel alternative adsorbents with high adsorptive capacity and low cost (Ali et al. 2012; Çelekli et al. 2012a; Prola et al. 2013; Reddy and Lee 2013). For this purpose, agricultural products and other by-products have been investigated such as pineapple (Hameed et al. 2009), pistachio husk (Çelekli et al. 2010a), peanut husk (Zhong et al. 2012), walnut husk (Çelekli et al. 2012b), Jatropha curcas shells (Prola et al. 2013), rice husk (Katal et al. 2012), and lentil straw (Çelekli et al. 2012c). In the literature, pumpkin husk (PH) has not been previously used in sorption process to remove a reactive azo dye. Pumpkin, a gourd-like squash of Cucurbitaceae's family, is one of the widely cultivated plant species for its fruit. The fruit of pumpkin is one of the most important vegetables in traditional agricultural systems in the world. The fruit represents rich sources of pectin-type dietary fiber, antioxidants (carotenoids), vitamins (C, E, B6, K, thiamine, and riboflavin), and minerals (potassium, phosphorus, magnesium, iron, and selenium; Rakcejeva et al. 2011). Throughout the world, production of pumpkin species is about 21.2 million tons/year in 2009 (FAO 2010). Main pumpkin crop producers are China (6,309,623 ton), India (3,500,000 ton), Russia (1,318,150 ton), and USA (861,870 ton). From that point, Turkey (337,882 ton) is at 12th place around the world (Balkaya et al. 2010). Sorption on biomaterials has been mainly attributed by the functional groups on the cell wall, which consist of various chemical groups, such as carboxyl, hydroxyl, amino, phosphate (Çelekli et al. 2009). Each agricultural product has distinctive cell wall properties, has been proven to be effective adsorbent for the treatment of wastewaters. Çelekli and Bozkurt (2013) reported that the surface of PH has various functional groups such as amino, hydroxyl, and carboxyl groups. Logistic, a sigmoidal model has been developed for describing the whole microbial growth curve, biovolume, and biomass productions (Zwietering et al. 1990; Çelekli and Yavuzatmaca 2009). Logistic model has been fitted to experimental data not only to describe whole sorption process (kinetics and equilibrium) but also to reveal more information about the sorption behavior, such as sorption rate (μ; min-1) and maximum dye uptake (A; mg g-1). Recently, it has been applied (Çelekli et al. 2012a; Çelekli and Bozkurt 2013) to take more information and describe all sorption process. Reactive Red (RR) 120 is one of the most used reactive azo dyes in textile industries for coloring of cotton and cellulose fibers because of their simple dying procedures and good stability during washing process (Çelekli et al. 2009, 2012a; Paul et al. 2013). This dye is a potential threat to the aquatic environment because of its high solubility in water and poor biodegradability.
5087
Although there are many studies on the dye sorption by use of agricultural waste, PH has not been previously used for the sorption of RR 120. Thus, PH was selected because it is a relatively abundant and low-cost material. The use of waste materials for sorption purposes can play a significant role in helping solve disposal problems, to protect environment, and to improve the balance of trade by reducing the dependence on imported materials. The major objectives of the present study were (i) to investigate the potential of PH for removing a potentially toxic RR 120 as functions of particle size, adsorbent dose, pH, temperature, ionic strength, initial dye concentration, and contact time in the batch system; (ii) to predict kinetic data by use of recently proposed logistic and generated modified logistic models to get more information about the sorption of RR 120; and (iii) to evaluate desorption behavior by use of logistic model. Furthermore, sorption kinetic behavior, activation energy, thermodynamic, and desorption studies were also investigated.
Materials and methods 2.1 Preparation and characterization of adsorbent PH (Cucurbita moschata Duchesne ex Poir.) was obtained from a field crop of Anatolia (Turkey). Collected sample was washed twice with tap water. Dried adsorbent was ground in a mortar, sieved by use of different mesh sizes of sieves (43, 65, 125, 250, and 500 μm), and stored in air tight polyethylene bottle for further studies. No chemical treatment was applied prior to adsorption experiments. Fourier transform infrared (FTIR) equipped with an attenuated total reflection spectrometer (Perkin-Elmer Spectrum 100 FTIR–ATR Spectrometer) was used to characterize PH surface before and after the sorption of RR 120. Zero point charge (pHzpc) of PH was determined by use of powder addition method. A series of mixture solution (0.5 g adsorbent and 50 ml 0.1 M NaCl) in 100 ml conical flask were prepared at various initial pH values (ranging from pH 1 to 10), adjusted with 0.1 M HCl and/or 1.0 M NaOH solutions. Batches were agitated on orbital shaker at 150 rpm for 24 h and the final pH (pHf) was measured at equilibrium. Value of pHzpc was determined from the plot of pHf against pHi. Adsorbate RR 120, reactive azo dye, was obtained from Sigma (Sigma– Aldrich Chemical, St. Louis, MO, USA). The chemical structure and properties of this dye are given on Table 1. A stock dye solution (1 g l-1) was prepared with distilled water. Experimental dye solutions were prepared by diluting stock dye solution with distilled water.
5088
Environ Sci Pollut Res (2014) 21:5086–5097
Table 1 General characteristics of Reactive Red 120
Name of dye
Procion Red HE-3B
Chemical formula Molar mass Color index name CAS number λmax
C44H24Cl2N14O20S6Na6 1,469.98 g mol-1 Reactive Red-120 61951-82-4 515 nm
Chemical structure
Sorption studies
Sorption modeling
The pH of solution was adjusted to desired values with 0.1 M HCl and/or 1.0 M NaOH. Experiments were performed with 100 ml of sorption solutions with desired dye concentration, pH, and adsorbent dose in 250 ml conical flasks. The flasks were agitated on an orbital shaker at 150 rpm for 90 min. Effects of particle size (43–65, 125– 250, 250–500, and >500 μm), adsorbent dose (0.5, 1.0, 2.0, and 4.0 g l-1), initial pH value (pH 1–9), temperature (298, 308, 318, and 328 K), ionic strength (0.0, 0.001, 0.010, and 0.100 M NaCl), initial dye concentration (40, 80, 120, 160, 200, and 240 mg l-1), and contact time (0–90 min) were studied under the aspects of sorption kinetics, activation energy, thermodynamic, and desorption studies. A sample solution containing PH was prepared without dye (blank) at the same conditions and its giving color into solution was determined by using spectrophotometer at 515 nm. It was found that it gave very little (negligible) color into solution. Samples were withdrawn at 0, 2, 4, 6, 8, 10, 15, 30, 45, 60, and 90 min and centrifuged at 1,790×g for 5 min. The residual RR 120 concentration in the supernatant was measured with a spectrophotometer (Jenway 6305) at 515 nm. Each data point was the mean of two independent sorption studies. Value of qt is amount of RR 120 adsorbed on PH at time t (mg g-1), calculated by use of Eq. 1:
In the present study, various theoretical equations were applied to experimental data in order to find the best model(s) which adequately predict kinetic or equilibrium data. Besides, secondary logistic equation was developed and applied to the kinetic data.
qt ¼
ðC 0 −C t Þ V m
ð1Þ
where C0 and Ct represent remain dye concentrations (mg l-1) at initial and at t time, respectively. V is the volume of solution (l), and m is adsorbent mass (g l-1).
Validation of models The fitting procedure was performed by use of commercial computer software SigmaPlot version 11 (Systat Sofware, CA, USA) via the Marquardt–Levenberg algorithm. The validity of models was evaluated using the coefficient of determination (R2) and the sum of squares errors (SSE). SSE is expressed as:
SSE ¼
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 u u t ∑ qexp −qpredict N
ð2Þ
where N is the number of data point, qpredict is the predicted sorption data from kinetic models, qexp is the observed experimental sorption data. Thermodynamic study Thermodynamic experiments were performed by agitating dye solutions of different concentrations (ranging from 40 to 240 mg l-1) on PH at various temperatures (298, 308, 318, and 328 K).
Environ Sci Pollut Res (2014) 21:5086–5097
5089
Desorption study RR 120 solution (1 g l-1) was mixed with PH at pH 1 for 90 min. Residual dye concentration was measured in order to calculate adsorbed dye on the adsorbent. After then, obtained adsorbent was dried in the vacuum oven at 70 °C for 24 h. RR 120 loaded PH was allowed to contact with 50 ml of distilled water at alkaline pH values (pH 8, 9, 10, and 11) in 100 ml flask, stirred on the orbital shaker at 150 rpm for 30 min. Effects of pH values (pH 8, 9, 10, and 11) on the desorption behavior was described by logistic and pseudo second-order kinetic models, has not been previously applied. Four cycles of sorption–desorption for RR 120 were studied for the reuse of PH at optimum pH value. Amount of desorbed dye was determined by using spectrophotometer (Jenway 6305) at 515 nm. The percentage of desorbed dye from the adsorbent was calculated: Desorption ð%Þ ¼
mass of desorbed 100 mass of adsorbed
ð3Þ
various functional groups such as amine, hydroxyl, and carboxyl groups. After the sorption of RR 120, several peaks at 3,286, 2,918, 2,850, 1,727, 1,648, 1,556, 1,515, 1,434, 1,304, 1,268, 1,229, 1,157, 1,028, and 819 cm-1 were found on the spectrum (Fig. 1b). Comparison of RR 120 loaded PH with its unloaded biomass, FTIR spectra displayed significant changes in some of the peaks. A few bands (2,918, 2,850, 1,651, and 1,229 cm-1) on the PH were not shifted. This indicated that they could not participate in the sorption process. Several peaks shifted to 3,304, 1,731, 1,543, 1,331, 1,208, and 1,157 cm-1 (Fig. 1b). Also, some new peaks such as at 1,463, 1,438, and 1,394 cm-1 were observed in the spectra of RR 120 loaded PH. Both new and shifted peaks could be due to formation of bonds between PH and RR 120 molecules. Results of FTIR analyses indicated that amine, amide, hydroxyl, and carboxyl groups on PH had a significant role on the sorption of RR 120. Similar results were also found for sorption of Congo Red on cashew nut shell (Kumar et al. 2010), RR 120 on Spirogyra majuscula (Çelekli et al. 2009) and on Hydrilla verticillata (Naveen et al. 2011). Effects of particle size and adsorbent dose
Results and discussion Characterization of adsorbent Each adsorbent has different binding capacity for each dye molecules. Adsorption capacity is not only affected by the textural or porous structure of adsorbents but also strongly influenced by the chemical structures of the surface. The cell surface consist of various polysaccharides, proteins, and lipid containing various functional groups such as amino, hydroxyl, carboxyl, carbonyl, sulphonate, sulphydryl, and phosphate, which can act as binding sites for dye molecules. FTIR technique was used to discover changes in the surface of adsorbent and FTIR spectra of PH before and after the sorption of RR 120 are shown in Fig. 1a and b, respectively. Various major peaks at 3,394, 2,918, 2,850, 1,729, 1,652, 1,535, 1,368, 1,317, 1,229, 1,151, 1,015, and 831 cm-1 were found on the spectrum of untreated PH powder (Fig. 1a). Peaks at 3,394 and 2,918 cm-1 could be corresponded to the presence of –OH and –NH2 groups 1,323 (Çelekli et al. 2012a; Mahmoodi et al. 2012;) and –CH stretching vibrations (Cardoso et al. 2011), respectively. Other peaks could be attributed as follows: 2,850 cm -1 (−CH 2 symmetric stretching), 1,729 cm-1 (−C=O stretching), 1,652 cm-1 (C=O bending), 1,535 cm-1 (−C NH bending), 1,368 cm-1 (carboxyl group), 1,317 cm-1 (−C–O stretching of COOH), 1,229 cm-1 (−C–N stretching), 1,151 cm-1 (P=O stretching), 1,015 cm-1 (−C–O stretching), and 831 cm−1 (−C–O and – OH stretching vibrations; Kumar et al. 2010; Prola et al. 2013). FTIR experiment revealed that the surface of PH has
In order to evaluate function of particle size on the sorption process, four particle sizes (43–65, 125–250, 250–500, and >500 μm) of PH were conducted with 100 mg l-1 RR 120. Relationship between particle size of PH and adsorbed RR 120 value is given in Fig. 2a. The sorption capacity significantly decreased with increasing particle size (p
RESEARCH ARTICLE
Predictive modeling of sorption and desorption of a reactive azo dye by pumpkin husk Abuzer Çelekli & Fadime Çelekli & Erdoğan Çiçek & Hüseyin Bozkurt
Received: 14 June 2013 / Accepted: 10 December 2013 / Published online: 28 December 2013 # Springer-Verlag Berlin Heidelberg 2013
Abstract The use of effective disposal of redundant pumpkin husk (PH) to remove pollutants is an important issue for environmental protection and utilization of resource. The aim of this study was to remove a potentially toxic reactive azo dye, Reactive Red (RR) 120, by widespread PH as a lowcost adsorbent. Particle size, adsorbent dose, pH, temperature, initial dye concentration, and contact time affected the sorption process. Amine, amide, hydroxyl, and carboxyl groups of PH played significant roles on the sorption process. Rapid sorption occurred within the first 2 min and equilibrium was reached within 60 min. Sorption kinetic was well represented by logistic equation. Generated secondary logistic model can be used to describe effects of initial dye concentration, contact time, and temperature by a single equation with high R2 value. Monolayer sorption capacity was found as 98.61 mg g-1. Activation energy, thermodynamic, and desorption studies showed that this process was physical, endothermic, and spontaneous. This study indicated that redundant PH as a low-cost adsorbent had a great potential for the removal of RR 120 as an alternative eco-friendly process. Keywords Activation energy . Predictive modeling . Pumpkin husk . Reactive Red 120 . Sorption
Responsible editor: Michael Matthies A. Çelekli (*) Department of Biology, Faculty of Art and Science, University of Gaziantep, 27310 Gaziantep, Turkey e-mail: [email protected] F. Çelekli : E. Çiçek Department of Biology, Faculty of Art and Science, University of Nevşehir Hacı Bektaş Veli, 50300 Nevşehir, Turkey H. Bozkurt Department of Food Engineering, Faculty of Engineering, University of Gaziantep, 27310 Gaziantep, Turkey
Introduction Synthetic dyes are extensively used in textile, paper, pharmaceutical, cosmetics and food industries which generate huge volumes of wastewater. The annual production of the synthetic dyes is about 0.7 Mt and reactive azo dyes are about 50 60 % of them (Phillips 1996; Solís et al. 2012). Reactive azo dyes contain one or more azo bonds (−N=N–), combined with aromatic and heterocyclic groups. They are generally resistant to fading on exposure to light, water and many chemicals due to their chemical structure (Ali 2010; Solís et al. 2012; Wang et al. 2013). Reactive dyes are mainly applied in textile processing, due to the ease and low cost in their use, stability and availability of various colors (Çelekli et al. 2009; Wang et al. 2013). About 15–30 % of these dyes remain in the effluents for coloring of final products in textile industry. These effluents contain not only dyes, but also metals, salts, and other chemicals. In order to color 1 kg of cotton with reactive dyes, 30–60 g dyestuff, 70–150 l water, and 0.6–0.8 kg NaCl are required. Throughout world, 280,000 tons of textile dyes are discharged with industrial effluents every year (Ali 2010; Solís et al. 2012). Discharging of unfixed dyes into receiving water may cause a negative impact on the ecosystems, such as reduction of photosynthetic activity, causing of aesthetic problems, and toxicity to life (Saratale et al. 2011; Daneshvar et al. 2012). Besides, once exposed, reactive azo dyes can cause variety of diseases and disorders in living organisms, even at low concentrations, such as allergy, skin irritation, and cancer (Brookstein 2009; Gupta and Suhas 2009; Salleh et al. 2011). Therefore, it is necessary to remove these recalcitrant compounds from wastewater, prior to discharging into aquatic ecosystems. A wide range of technologies have been developed to remove these dyes from wastewaters (Ali 2010; Srinivasan and Viraraghavan 2010; Salleh et al. 2011; Gupta et al. 2012). Biodegradation of reactive dyes is usually very difficult due to
Environ Sci Pollut Res (2014) 21:5086–5097
their complex molecular structure. Therefore, they cannot be easily remove from wastewater by conventional coagulation or activated sludge methods. Adsorption is an attractive, cheap, and effective method for the treatment of dye-bearing effluents. Researchers try to optimize adsorption process and develop novel alternative adsorbents with high adsorptive capacity and low cost (Ali et al. 2012; Çelekli et al. 2012a; Prola et al. 2013; Reddy and Lee 2013). For this purpose, agricultural products and other by-products have been investigated such as pineapple (Hameed et al. 2009), pistachio husk (Çelekli et al. 2010a), peanut husk (Zhong et al. 2012), walnut husk (Çelekli et al. 2012b), Jatropha curcas shells (Prola et al. 2013), rice husk (Katal et al. 2012), and lentil straw (Çelekli et al. 2012c). In the literature, pumpkin husk (PH) has not been previously used in sorption process to remove a reactive azo dye. Pumpkin, a gourd-like squash of Cucurbitaceae's family, is one of the widely cultivated plant species for its fruit. The fruit of pumpkin is one of the most important vegetables in traditional agricultural systems in the world. The fruit represents rich sources of pectin-type dietary fiber, antioxidants (carotenoids), vitamins (C, E, B6, K, thiamine, and riboflavin), and minerals (potassium, phosphorus, magnesium, iron, and selenium; Rakcejeva et al. 2011). Throughout the world, production of pumpkin species is about 21.2 million tons/year in 2009 (FAO 2010). Main pumpkin crop producers are China (6,309,623 ton), India (3,500,000 ton), Russia (1,318,150 ton), and USA (861,870 ton). From that point, Turkey (337,882 ton) is at 12th place around the world (Balkaya et al. 2010). Sorption on biomaterials has been mainly attributed by the functional groups on the cell wall, which consist of various chemical groups, such as carboxyl, hydroxyl, amino, phosphate (Çelekli et al. 2009). Each agricultural product has distinctive cell wall properties, has been proven to be effective adsorbent for the treatment of wastewaters. Çelekli and Bozkurt (2013) reported that the surface of PH has various functional groups such as amino, hydroxyl, and carboxyl groups. Logistic, a sigmoidal model has been developed for describing the whole microbial growth curve, biovolume, and biomass productions (Zwietering et al. 1990; Çelekli and Yavuzatmaca 2009). Logistic model has been fitted to experimental data not only to describe whole sorption process (kinetics and equilibrium) but also to reveal more information about the sorption behavior, such as sorption rate (μ; min-1) and maximum dye uptake (A; mg g-1). Recently, it has been applied (Çelekli et al. 2012a; Çelekli and Bozkurt 2013) to take more information and describe all sorption process. Reactive Red (RR) 120 is one of the most used reactive azo dyes in textile industries for coloring of cotton and cellulose fibers because of their simple dying procedures and good stability during washing process (Çelekli et al. 2009, 2012a; Paul et al. 2013). This dye is a potential threat to the aquatic environment because of its high solubility in water and poor biodegradability.
5087
Although there are many studies on the dye sorption by use of agricultural waste, PH has not been previously used for the sorption of RR 120. Thus, PH was selected because it is a relatively abundant and low-cost material. The use of waste materials for sorption purposes can play a significant role in helping solve disposal problems, to protect environment, and to improve the balance of trade by reducing the dependence on imported materials. The major objectives of the present study were (i) to investigate the potential of PH for removing a potentially toxic RR 120 as functions of particle size, adsorbent dose, pH, temperature, ionic strength, initial dye concentration, and contact time in the batch system; (ii) to predict kinetic data by use of recently proposed logistic and generated modified logistic models to get more information about the sorption of RR 120; and (iii) to evaluate desorption behavior by use of logistic model. Furthermore, sorption kinetic behavior, activation energy, thermodynamic, and desorption studies were also investigated.
Materials and methods 2.1 Preparation and characterization of adsorbent PH (Cucurbita moschata Duchesne ex Poir.) was obtained from a field crop of Anatolia (Turkey). Collected sample was washed twice with tap water. Dried adsorbent was ground in a mortar, sieved by use of different mesh sizes of sieves (43, 65, 125, 250, and 500 μm), and stored in air tight polyethylene bottle for further studies. No chemical treatment was applied prior to adsorption experiments. Fourier transform infrared (FTIR) equipped with an attenuated total reflection spectrometer (Perkin-Elmer Spectrum 100 FTIR–ATR Spectrometer) was used to characterize PH surface before and after the sorption of RR 120. Zero point charge (pHzpc) of PH was determined by use of powder addition method. A series of mixture solution (0.5 g adsorbent and 50 ml 0.1 M NaCl) in 100 ml conical flask were prepared at various initial pH values (ranging from pH 1 to 10), adjusted with 0.1 M HCl and/or 1.0 M NaOH solutions. Batches were agitated on orbital shaker at 150 rpm for 24 h and the final pH (pHf) was measured at equilibrium. Value of pHzpc was determined from the plot of pHf against pHi. Adsorbate RR 120, reactive azo dye, was obtained from Sigma (Sigma– Aldrich Chemical, St. Louis, MO, USA). The chemical structure and properties of this dye are given on Table 1. A stock dye solution (1 g l-1) was prepared with distilled water. Experimental dye solutions were prepared by diluting stock dye solution with distilled water.
5088
Environ Sci Pollut Res (2014) 21:5086–5097
Table 1 General characteristics of Reactive Red 120
Name of dye
Procion Red HE-3B
Chemical formula Molar mass Color index name CAS number λmax
C44H24Cl2N14O20S6Na6 1,469.98 g mol-1 Reactive Red-120 61951-82-4 515 nm
Chemical structure
Sorption studies
Sorption modeling
The pH of solution was adjusted to desired values with 0.1 M HCl and/or 1.0 M NaOH. Experiments were performed with 100 ml of sorption solutions with desired dye concentration, pH, and adsorbent dose in 250 ml conical flasks. The flasks were agitated on an orbital shaker at 150 rpm for 90 min. Effects of particle size (43–65, 125– 250, 250–500, and >500 μm), adsorbent dose (0.5, 1.0, 2.0, and 4.0 g l-1), initial pH value (pH 1–9), temperature (298, 308, 318, and 328 K), ionic strength (0.0, 0.001, 0.010, and 0.100 M NaCl), initial dye concentration (40, 80, 120, 160, 200, and 240 mg l-1), and contact time (0–90 min) were studied under the aspects of sorption kinetics, activation energy, thermodynamic, and desorption studies. A sample solution containing PH was prepared without dye (blank) at the same conditions and its giving color into solution was determined by using spectrophotometer at 515 nm. It was found that it gave very little (negligible) color into solution. Samples were withdrawn at 0, 2, 4, 6, 8, 10, 15, 30, 45, 60, and 90 min and centrifuged at 1,790×g for 5 min. The residual RR 120 concentration in the supernatant was measured with a spectrophotometer (Jenway 6305) at 515 nm. Each data point was the mean of two independent sorption studies. Value of qt is amount of RR 120 adsorbed on PH at time t (mg g-1), calculated by use of Eq. 1:
In the present study, various theoretical equations were applied to experimental data in order to find the best model(s) which adequately predict kinetic or equilibrium data. Besides, secondary logistic equation was developed and applied to the kinetic data.
qt ¼
ðC 0 −C t Þ V m
ð1Þ
where C0 and Ct represent remain dye concentrations (mg l-1) at initial and at t time, respectively. V is the volume of solution (l), and m is adsorbent mass (g l-1).
Validation of models The fitting procedure was performed by use of commercial computer software SigmaPlot version 11 (Systat Sofware, CA, USA) via the Marquardt–Levenberg algorithm. The validity of models was evaluated using the coefficient of determination (R2) and the sum of squares errors (SSE). SSE is expressed as:
SSE ¼
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 u u t ∑ qexp −qpredict N
ð2Þ
where N is the number of data point, qpredict is the predicted sorption data from kinetic models, qexp is the observed experimental sorption data. Thermodynamic study Thermodynamic experiments were performed by agitating dye solutions of different concentrations (ranging from 40 to 240 mg l-1) on PH at various temperatures (298, 308, 318, and 328 K).
Environ Sci Pollut Res (2014) 21:5086–5097
5089
Desorption study RR 120 solution (1 g l-1) was mixed with PH at pH 1 for 90 min. Residual dye concentration was measured in order to calculate adsorbed dye on the adsorbent. After then, obtained adsorbent was dried in the vacuum oven at 70 °C for 24 h. RR 120 loaded PH was allowed to contact with 50 ml of distilled water at alkaline pH values (pH 8, 9, 10, and 11) in 100 ml flask, stirred on the orbital shaker at 150 rpm for 30 min. Effects of pH values (pH 8, 9, 10, and 11) on the desorption behavior was described by logistic and pseudo second-order kinetic models, has not been previously applied. Four cycles of sorption–desorption for RR 120 were studied for the reuse of PH at optimum pH value. Amount of desorbed dye was determined by using spectrophotometer (Jenway 6305) at 515 nm. The percentage of desorbed dye from the adsorbent was calculated: Desorption ð%Þ ¼
mass of desorbed 100 mass of adsorbed
ð3Þ
various functional groups such as amine, hydroxyl, and carboxyl groups. After the sorption of RR 120, several peaks at 3,286, 2,918, 2,850, 1,727, 1,648, 1,556, 1,515, 1,434, 1,304, 1,268, 1,229, 1,157, 1,028, and 819 cm-1 were found on the spectrum (Fig. 1b). Comparison of RR 120 loaded PH with its unloaded biomass, FTIR spectra displayed significant changes in some of the peaks. A few bands (2,918, 2,850, 1,651, and 1,229 cm-1) on the PH were not shifted. This indicated that they could not participate in the sorption process. Several peaks shifted to 3,304, 1,731, 1,543, 1,331, 1,208, and 1,157 cm-1 (Fig. 1b). Also, some new peaks such as at 1,463, 1,438, and 1,394 cm-1 were observed in the spectra of RR 120 loaded PH. Both new and shifted peaks could be due to formation of bonds between PH and RR 120 molecules. Results of FTIR analyses indicated that amine, amide, hydroxyl, and carboxyl groups on PH had a significant role on the sorption of RR 120. Similar results were also found for sorption of Congo Red on cashew nut shell (Kumar et al. 2010), RR 120 on Spirogyra majuscula (Çelekli et al. 2009) and on Hydrilla verticillata (Naveen et al. 2011). Effects of particle size and adsorbent dose
Results and discussion Characterization of adsorbent Each adsorbent has different binding capacity for each dye molecules. Adsorption capacity is not only affected by the textural or porous structure of adsorbents but also strongly influenced by the chemical structures of the surface. The cell surface consist of various polysaccharides, proteins, and lipid containing various functional groups such as amino, hydroxyl, carboxyl, carbonyl, sulphonate, sulphydryl, and phosphate, which can act as binding sites for dye molecules. FTIR technique was used to discover changes in the surface of adsorbent and FTIR spectra of PH before and after the sorption of RR 120 are shown in Fig. 1a and b, respectively. Various major peaks at 3,394, 2,918, 2,850, 1,729, 1,652, 1,535, 1,368, 1,317, 1,229, 1,151, 1,015, and 831 cm-1 were found on the spectrum of untreated PH powder (Fig. 1a). Peaks at 3,394 and 2,918 cm-1 could be corresponded to the presence of –OH and –NH2 groups 1,323 (Çelekli et al. 2012a; Mahmoodi et al. 2012;) and –CH stretching vibrations (Cardoso et al. 2011), respectively. Other peaks could be attributed as follows: 2,850 cm -1 (−CH 2 symmetric stretching), 1,729 cm-1 (−C=O stretching), 1,652 cm-1 (C=O bending), 1,535 cm-1 (−C NH bending), 1,368 cm-1 (carboxyl group), 1,317 cm-1 (−C–O stretching of COOH), 1,229 cm-1 (−C–N stretching), 1,151 cm-1 (P=O stretching), 1,015 cm-1 (−C–O stretching), and 831 cm−1 (−C–O and – OH stretching vibrations; Kumar et al. 2010; Prola et al. 2013). FTIR experiment revealed that the surface of PH has
In order to evaluate function of particle size on the sorption process, four particle sizes (43–65, 125–250, 250–500, and >500 μm) of PH were conducted with 100 mg l-1 RR 120. Relationship between particle size of PH and adsorbed RR 120 value is given in Fig. 2a. The sorption capacity significantly decreased with increasing particle size (p
