Translate this page into:
Sorption of malachite green from aqueous solution by potato peel: Kinetics and equilibrium modeling using non-linear analysis method
⁎Corresponding author. Tel./fax: +213 38876560. oualid.hamdaoui@univ-annaba.org (Oualid Hamdaoui) ohamdaoui@yahoo.fr (Oualid Hamdaoui)
-
Received: ,
Accepted: ,
This article was originally published by Elsevier and was migrated to Scientific Scholar after the change of Publisher.
Peer review under responsibility of King Saud University.
Abstract
Potato peel (PP) was used as a biosorbent to remove malachite green (MG) from aqueous solution under various operating conditions. The effect of the experimental parameters such as initial dye concentration, biosorbent dose, initial pH, stirring speed, temperature, ionic strength and biosorbent particle size was investigated through a number of batch sorption experiments. The sorption kinetic uptake for MG by PP at various initial dye concentrations was analyzed by non-linear method using pseudo-first, pseudo-second and pseudo-nth order models. It was found that the pseudo-nth order kinetic model was the best applicable model to describe the sorption kinetic data and the order n of sorption reaction was calculated in the range from 0.71 to 2.71. Three sorption isotherms namely the Langmuir, Freundlich and Redlich–Peterson isotherms in their non-linear forms were applied to the biosorption equilibrium data. Both the Langmuir and Redlich–Peterson models were found to fit the sorption isotherm data well, but the Redlich–Peterson model was better. Thermodynamic parameters show that the sorption process of MG is endothermic and more effective process at high temperatures. The results revealed that PP is very effective for the biosorption of MG from aqueous solutions.
Keywords
Sorption
Malachite green
Potato peel
Modeling
Non-linear method
1 Introduction
Synthetic dyes are an important class of recalcitrant organic compounds and are often found in the environment as a result of their wide industrial use. Dyes are used in many industries such as food, paper, plastics, cosmetics, papermaking and textile in order to color their products. These colored effluents can be mixed in surface water and ground water systems, and then they may also be transferred to drinking water. Color is the first contaminant to be recognized in wastewater. The presence of very small amounts of dyes in water (less than 1 ppm for some dyes) is highly visible and aesthetically unpleasant. Due to intense color they reduce sunlight transmission into water hence affecting aquatic plants, which ultimately disturb aquatic ecosystem. Dye bearing effluents are characterized by high chemical oxygen demand, low biodegradability and high salt content. In addition, it has been stressed recently that color compounds such as dyes discharged from several industries are very harmful to aquatic life in rivers and lakes. Therefore, it is necessary to reduce dye concentration in the wastewater before it is released into the environment.
Malachite green (MG), a basic dye, is most widely used for coloring purpose, amongst all other dyes of its category (Crini et al., 2007). This triarylmethane dye is widely used in the aquaculture industry worldwide as a biocide as well as in the silk, wool, cotton, leather, paper and acrylic industries as a dye. However there are several reports describing its hazardous and carcinogenic effects (Srivastava et al., 2004). It is known to be highly cytotoxic and carcinogenic to mammalian cells and acts as a liver tumor promoter. In humans, it may cause irritation to the respiratory tract if inhaled and causes irritation to the gastrointestinal tract upon ingestion. Contact of malachite green with the skin causes irritation with redness and pain; upon contact with eye will lead to permanent injury of human eyes. It also affects the aquatic life and causes detrimental effects in liver, gill, kidney, intestine, gonads and pituitary gonadotrophic cells (Hameed and El-Khaiary, 2008a). Therefore, the treatment of effluent containing such dye is of interest due to its harmful impact on receiving waters.
Among the numerous treatment technologies developed for the removal of dyes from industrial effluents, biosorption is receiving increasing attention in becoming an attractive and promising technology. The study of biosorption is of great importance from an environmental point of view, as it can be considered as an alternative technique for removing toxic pollutants from wastewaters (Vieira and Volesky, 2000; Nouri and Hamdaoui, 2007). A number of low cost biosorbent have been studied in the literature for their capacity to remove MG from aqueous solutions (Baek et al., 2010; Bekçi et al., 2009; Hamdaoui et al., 2008; Hameed and El-Khaiary, 2008a; Khattri and Singh, 2009; Mittal, 2006; Pradeep Sekhar et al., 2009; Sonawane and Shrivastava, 2009). Undoubtedly, agricultural waste biomass is presently one of the most challenging topics, which is gaining stern considerations during the past several decades. In perspective, potato peel has emerged to be an invaluable source.
Potato peel (PP), agricultural and easily available waste, could be an alternative for more costly wastewater treatment processes. Losses caused by potato peeling range from 15% to 40% their amount depending on the procedure applied, i.e. steam, abrasion or lye peeling (Scieber et al., 2001). Plants peel the potatoes as part of the production of French fries, crisps, puree, instant potatoes and similar products. The produced waste is 90 kg per Mg of influent potatoes and is apportioned to 50 kg of potato skins, 30 kg starch and 10 kg inert material. The problem of the management of PP causes considerable concern to the potato industries, thus implying the need to identify an integrated, environmentally-friendly solution. PP is a zero value waste from potato processing plants. In the interest of the environment, we propose this agricultural waste as a low-cost sorbent to remove malachite green from aqueous solution.
The objective of this work was to investigate the potential of PP as a novel sorbent in the removal of the basic dye, malachite green, from aqueous solutions. Batch sorption studies were conducted to investigate the effects of various parameters such initial dye concentration, biosorbent dose, pH, temperature, stirring speed, ionic strength and particle size on MG biosorption. Equilibrium isotherm and kinetic data are analyzed and modeled using different models by non-linear regression technique.
2 Materials and methods
2.1 Biosorbent
The PP used in the present study was obtained from the university canteen. It was washed, dried, crushed and sieved to desired mesh size (0.5–1.25 mm). Finally, the obtained material was then dried in an air circulating oven at 50 °C for 7 days and stored in a desiccator until use.
2.2 Sorbate
The cationic basic dye (C.I. 42000; Basic Green 4), malachite green oxalate salt, (molecular formula C52H56N4O12, FW 929), was obtained from Merck and used without further purification. The structure of this dye is displayed in Fig. 1. Five hundred milligram per liter stock solution was prepared by dissolving the required amount of dye in distilled water. Working solutions of the desired concentrations were obtained by successive dilutions.
Chemical structure of malachite green (oxalate salt).
2.3 Analytical method
A well-known procedure for determining MG concentrations, based on Beer’s law calibration plots, was applied using a UV–visible spectrophotometer (Jenway 6405). The wavelength resolution and the bandwidth were, respectively, 1 and 0.5 nm. The length of the optical path in glass cell was 1 cm. The maximum absorption wavelength was determined as equal to 618 nm. Then, the calibration plot was constructed. The calibration was repeated five times during the period of measurements. The linearization of this plot usually provided determination coefficient close to 99.99%. These data were used to calculate the sorption capacity of the biosorbent. In most cases, a proper dilution was necessary to obtain a well measurable absorption.
2.4 Procedures
The initial concentration of MG solution was 50 mg L−1 for all experiments, except for those carried out to examine the effect of the initial dye concentration. For dye removal kinetic experiments, the batch method was used because of its simplicity: 0.25 g of biosorbent was contacted with 100 mL of dye solution in a sealed flask agitated vigorously by a magnetic stirrer using a water bath maintained at a constant temperature. The stirring speed was kept constant at 400 rpm, except for experiments carried out to investigate the effect of stirring speed. At predetermined intervals of time, samples of the mixture was withdrawn at suitable time intervals and analyzed by a UV–visible spectrophotometer for the concentration of MG.
The experiments were performed at the pH that resulted from solving the dye in water (around 4) without further adjustment, except for those conducted to examine the effect of solution pH.
To study the effect of solution pH on MG sorption, 0.25 g of PP was agitated with 100 mL of MG solution of dye concentration 50 mg L−1 at 25 °C. The experiment was conducted at different pH values ranging from 2 to 8. The solution pH was adjusted using 0.1 N HCl or NaOH aqueous solutions. Agitation was provided for a period which is sufficient to reach equilibrium with a constant agitation speed of 400 rpm.
The effect of temperature on the sorption of MG was studied by contacting 0.25 g of sorbent with 100 mL of dye solution of 50 mg L−1 initial concentration at different temperatures (25–45 °C).
The influence of ionic strength on the sorption of MG by PP was studied with a constant initial concentration of 50 mg L−1, sorbent mass of 0.25 g, solution volume of 100 mL, and temperature of 25 °C. The ionic strength of the dye solution was modified using different dosages of NaCl (0.25–2 g (100 mL)−1).
In all the experiments in which the effect of stirring speed was studied, the initial MG concentration, sorbent mass and solution temperature were 50 mg L−1, 0.25 g and 25 °C, respectively. The stirring speed was varied from 0 (without stirring) to 800 rpm.
Sorption equilibrium experiments were carried out by adding a fixed amount of PP (0.25 g) into a number of sealed glass flasks containing a definite volume (100 mL in each case) of different initial concentrations (50–500 mg L−1) of MG solution without changing pH. The flasks were placed in a thermostatic water bath in order to maintain a constant temperature (25, 35 or 45 °C) and stirring was provided at 400 rpm to ensure equilibrium was reached. Samples of solutions were analyzed for the remaining dye concentration with a UV–vis spectrophotometer. The amount of sorption at equilibrium, qe (mg g−1), was calculated by:
Each run of the experiments was replicated at least two times and the mean values were reported. The maximum standard deviation obtained for duplicate or triplicate measurements of the sorbed amount was ±2%.
3 Results and discussion
3.1 Effect of operational parameters
The influence of several operational parameters such as initial dye concentration, sorbent dosage, pH, stirring speed, temperature, ionic strength and particle size on the sorption of MG by PP was investigated.
3.1.1 Effect of initial dye concentration
The effect of initial dye concentration on the sorption of MG by PP at 25 °C is shown in Fig. 2. It can be observed that the sorption capacity increased with time and, at some point in time, reached a constant value where no more dye was removed from the solution. At this point, the amount of MG being sorbed by the sorbent was in a state of dynamic equilibrium with the amount of MG desorbed from the sorbent. It was observed that an increase in initial dye concentration leads to an increase in the sorption of MG by PP. It was observed that the MG removal varied with varying initial dye concentration. It was noticed that an increase in initial dye concentration leads to an increase in the sorption capacity of MG by PP. Equilibrium uptake increased with the increase of initial dye concentration at the range of experimental concentration. The amount of MG sorbed at equilibrium increased from 1.64 to 10.29 mg g−1 as the concentration was increased from 5 to 50 mg L−1. The initial rate of sorption was greater for higher initial MG concentration, because the resistance to the dye uptake decreased as the mass transfer driving force increased. It is also noticed that as the initial dye concentration increased, the equilibrium removal of MG decreased. This effect can be explained on the basis of the dye/sorbent ratio. At low dye/sorbent ratios, there are number of sorption sites in PP structure. As the dye/sorbent ratio increases, sorption sites are saturated, resulting in a decrease in the sorption efficiency.
Kinetics of MG uptake by PP for various initial dye concentrations (conditions: sorbent dosage = 0.25 g (100 mL)−1; stirring speed = 400 rpm; T = 25 °C; pH 4).
It is also shown in Fig. 2 that the contact time needed for MG solutions with initial concentrations of 50 and 40 mg L−1 to reach equilibrium was 210 and 180 min, respectively. For MG solution with initial concentrations in the range of 5–30 mg L−1, equilibrium time of 150 min was required. This is due to the fact that sorption sites took up the available dye molecules more quickly at low concentration, but dye needed to diffuse to the inner sites of the sorbent for high concentration. The kinetic results also showed that the curves of contact time are single, smooth and continuous leading to equilibrium. These curves indicate the possible monolayer coverage of dye on the surface of PP.
3.1.2 Effect of biosorbent dose
The mass of sorbent was varied in the range of 0.25–1 g for the removal of MG from aqueous solution by PP. In these series of experiments, the concentration of MG in solution was fixed at 50 mg L−1. The effect of sorbent dose on the sorption kinetics of MG by the biosorbent is shown in Fig. 3. The amount of dye sorbed per unit mass of sorbent decreases with an increase in sorbent dose. The increase in sorbent dose at constant dye concentration and volume will lead to unsaturation of sorption sites through the sorption process. At higher PP to solute concentration ratios, there is a fast superficial sorption onto the sorbent surface that produces a lower solute concentration in the solution than when the sorbent to solute concentration ratio is lower. This is because a fixed mass of PP can only sorb a certain amount of dye. Therefore, the higher the sorbent dosage is, the larger the volume of effluent that a fixed mass of sorbent can purify. The decrease in the amount of MG sorbed with increasing sorbent mass is due to the split in the flux or the concentration gradient between solute concentrations in the solution and on the sorbent surface. Thus, with increasing sorbent dosage, the amount of dye sorbed by unit weight of sorbent gets reduced, thus causing a decrease in sorption capacity with increasing sorbent dosage. Additionally, this decrease may be attributed to overlapping or aggregation of sorption sites resulting in decrease in total sorbent surface area available to dye molecules and an increase in diffusion path length.
Effect of sorbent dosage on the sorption of MG by PP (conditions: initial dye concentration = 50 mg L−1; stirring speed = 400 rpm; T = 25 °C; pH 4).
3.1.3 Effect of initial pH
Fig. 4 shows the effect of pH on the sorption of MG by PP. It can be seen that the decrease in the sorption with decrease in pH. As pH of the system decreased, the number of negatively charged sorption sites decreased and the number of positively charged surface sites increased, which did not favor the sorption of positively charged dye cations due to electrostatic repulsion. Additionally, lower sorption of MG at acidic pH is due to the presence of excess H+ ions competing with dye cations for the sorption sites of PP (Hamdaoui, 2006). Similar result was reported for the sorption of MG by rice straw-derived char (Hameed and El-Khaiary, 2008b).
Effect of initial pH on the sorption of MG by PP (conditions: initial dye concentration = 50 mg L−1; sorbent dosage = 0.25 g (100 mL)−1; stirring speed = 400 rpm; T = 25 °C).
3.1.4 Effect of stirring speed
Agitation is a classic parameter in sorption phenomena, influencing the distribution of the solute in the bulk solution and the formation of the external boundary film. Fig. 5 shows the sorption of MG by PP at different stirring speeds ranging from 0 (without stirring) to 800 rpm. The obtained results show that the high sorption rate and capacity were obtained with a stirring speed of 400 rpm. This agitation speed assures a good diffusion of dye cations toward biosorbent particles. For the experiment conducted without agitation, it is noticed a significant reduction of the sorption. The amount of dye sorption increases with the increase of the stirring speed from 0 to 400 rpm. When increasing the agitation speed, the diffusion rate of dye molecules from the bulk liquid to the liquid boundary layer surrounding particles became higher because of an enhancement of turbulence and a decrease of the thickness of the liquid boundary layer. Finally, the boundary layer became very thin and approached the laminar sublayer at high agitation speeds. On the other hand, for a high stirring speed of 800 rpm, a significant reduction of sorption was observed.
Effect of stirring speed on the sorption of MG by PP (conditions: initial dye concentration = 50 mg L−1; sorbent dosage = 0.25 g (100 mL)−1; T = 25 °C; pH 4).
3.1.5 Effect of temperature
Fig. 6 presents the sorptive removal of MG as a function of time at different temperatures (25, 35 and 45 °C). Similar shape of the kinetic curves is observed for all the temperatures. It was found that the sorption kinetics increased with the increase in temperature. This indicates that the sorption process is endothermic in nature. As is known, the rate of diffusion of the sorbate molecules is increased by increasing the temperature, owing to the decrease in the viscosity of the solution. This enhancement is felt to be due to the acceleration of the sorption process by the increased movement of dye molecules from the bulk solution to the surface of the solid particles at higher temperatures. On the other hand, there is no significant effect of temperature on the equilibrium sorption capacity.
Effect of temperature on the sorption of MG by PP (conditions: initial dye concentration = 50 mg L−1; sorbent dosage = 0.25 g (100 mL)−1; stirring speed = 400 rpm; pH 4).
3.1.6 Effect of ionic strength
Fig. 7 shows the effect of salt concentration (ionic strength) on the amount of MG sorbed by PP. It was observed that the sorption potential decreased with increased concentration of NaCl in the medium. This behavior could be attributed to the competitive effect between dyes cations and ions from the salt (Na+) for the sites available for the sorption process. Additionally, salt screens the electrostatic interaction between sorbent and sorbate and the great ionic strength influences on the activity coefficient of MG, which should decrease the sorbed amount with increase of salt concentration.
Effect of salt (NaCl) concentration on the sorption of MG by PP (conditions: initial dye concentration = 50 mg L−1; sorbent dosage = 0.25 g (100 mL)−1; stirring speed = 400 rpm; T = 25 °C; pH 4).
3.1.7 Effect of biosorbent particle size
The effect of particle size of PP on the MG removal was studied using three particle size ranges: 0.18–0.5, 0.5–1.25 and 1.25–2 mm. Fig. 8 shows the sorption kinetics of the dye at three different particle sizes. The results show the role played by the sorbent particle size on the sorptive properties of PP. As can be seen from Fig. 8, the kinetic curves obtained have an identical shape, and that the removal is improved as the particle size decreased. This is because the smaller particles have more surface area and access to the particle pores is facilitated when their size is small. It is also believed that the breaking up of large particles to form smaller ones opens some tiny sealed channels, which might then become available for sorption, and so the sorption by smaller particles is higher than that by larger particles. The relationship between the effective surface area of the sorbent particles and their sizes is that the effective surface area increases as the particle size decreases and as a consequence, the sorption capacity per unit mass of the sorbent increased. So the smaller PP particle sizes for a given mass of biosorbent have more surface area and therefore the number of available sites is more.
Effect of biosorbent particle size on the sorption of MG by PP (conditions: initial dye concentration = 50 mg L−1; sorbent dosage = 0.25 g (100 mL)−1; stirring speed = 400 rpm; T = 25 °C; pH 4).
3.2 Modeling of sorption kinetics
Various sorption kinetic models have been used to describe the uptake of sorbate depending upon the time. Understanding of biosorption kinetics is much less than theoretical description of biosorption equilibrium. The pseudo-first order kinetic equation or the so-called Lagergren equation has the following formulation (Lagergren and Sven, 1898):
The experimental values of the amount of sorbate sorbed (q) and time were analyzed by non-linear curve fitting analysis in order to determine the models parameters and the curves were reconstituted using the determined values. The obtained curves showed the superposition of experimental results and the theoretical calculated points. The determination coefficients (R2) showed the fit between experimental data and models while the average percentage errors (APE) calculated according to Eq. (7) indicated the fit between the experimental and predicted values of the sorbed amount used for plotting the curves showing the sorption amount as a function of time.
Comparison of experimental and predicted kinetics for the biosorption of MG by PP at various initial dye concentrations.
5 mg L−1
10 mg L−1
20 mg L−1
30 mg L−1
40 mg L−1
50 mg L−1
Pseudo-first order model
k1 (min−1) × 103
50.62
43.34
36.33
25.95
27.70
22.61
qe (mg g−1)
1.64
3.27
6.10
8.02
8.72
10.20
R2
0.998
0.998
0.995
0.994
0.996
0.988
APE (%)
2.05
2.14
3.74
4.73
3.77
7.41
Pseudo-second order model
k2 (g mg−1 min−1) × 103
37.82
15.69
6.74
3.04
3.17
2.04
qe (mg g−1)
1.82
3.68
6.94
9.54
10.24
12.28
R2
0.985
0.992
0.986
0.980
0.995
0.996
APE (%)
6.73
4.30
5.25
8.12
3.18
3.43
Pseudo-nth order model
kn (min−1) (mg g−1)1−n × 103
50.62
37.67
34.25
41.03
17.26
0.26
n
1.07
1.19
1.04
0.71
1.25
2.71
qe (mg g−1)
1.64
3.32
6.12
7.90
8.99
14.13
R2
0.998
0.999
0.995
0.995
0.997
0.996
APE (%)
1.86
1.34
3.67
3.38
3.22
2.49
3.3 Equilibrium isotherms
The analysis of the isotherm data by fitting them to different isotherm models is an important step to find the suitable model that can be used for design purpose. In this work, three sorption isotherms namely the Langmuir, Freundlich and Redlich–Peterson isotherms in their non-linear forms were applied to the equilibrium data of sorption of MG by PP.
Langmuir isotherm assumes monolayer sorption onto a surface containing a finite number of sorption sites of uniform strategies of sorption with no transmigration of sorbate in the plane of surface (Weber and Chakkravorti, 1974):
Fig. 10 shows the experimental data fitted to non-linear forms of the three isotherms, using MicrocalTM Origin® software, for MG sorption by PP. The isotherms constants related to Langmuir, Freundlich and Redlich–Peterson models determined from the plots shown in Fig. 10 are listed in Table 2. The sorption equilibrium data were fitted well to the Redlich–Peterson model which combines the features of Langmuir and Freundlich models. Determination coefficients (R2) and average percentage errors (APE) for Redlich–Peterson model were determined in the range 0.984–0.990 and 3.54–4.35, respectively, for all temperatures studied. As can be clearly seen from Table 2, the values of β tend to unity, which shows that the isotherms are approaching Langmuir form. At 35 °C, the Langmuir and Redlich–Peterson models gave the highest determination coefficient value showing that the sorption isotherms of MG by PP were best described by these two models. The suitability of the Langmuir isotherm to fit the data at 35 °C was confirmed by the exponent value of the Redlich–Peterson model, β, which was equal to one. At 25 and 45 °C, taking into account the average percentage errors and determination coefficients, it was observed that both the Langmuir and Redlich–Peterson isotherms could well represent the experimental sorption data, but the Redlich–Peterson model was better. The monolayer saturation capacity, qm, was found to be 32.39, 34.62 and 35.61 mg g−1 at 25, 35 and 45 °C, respectively.In the sense of the sorption thermodynamics, change in free energy (ΔG°) of the removal of MG by PP can be calculated in a way such that
Comparison between the experimental and predicted isotherms for the biosorption of MG by PP (conditions: initial dye concentration = 50–500 mg L−1; sorbent dosage = 0.25 g (100 mL)−1; stirring speed = 400 rpm; pH 4).
Model
Parameters
qm (mg g−1)
b (L mg−1) × 103
R2
APE (%)
Langmuir (°C)
25
32.39
17.89
0.983
4.55
35
34.62
18.06
0.985
4.32
45
35.61
18.60
0.988
3.85
KF (
)
n
R2
APE (%)
Freundlich (°C)
25
4.47
3.17
0.950
6.91
35
4.77
3.16
0.939
8.02
45
4.94
3.16
0.958
6.57
A (L g−1)
B (L mg−1)β × 103
β
R2
APE (%)
Redlich–Peterson (°C)
25
0.66
28.15
0.95
0.984
4.35
35
0.62
16.96
1.00
0.985
4.32
45
0.78
33.02
0.93
0.990
3.54
Plot of Gibbs free energy change versus temperature.
4 Conclusion
The potential of potato peel for the biosorption of MG from aqueous solution was investigated. The effects of experimental parameters such as initial dye concentration, biosorbent dose, initial pH, stirring speed, temperature, ionic strength, biosorbent particle size on dye biosorption were studied. Initial dye concentration, sorbent dose, initial pH, stirring speed, ionic strength and biosorbent particle size were found to have an influence on the biosorption efficiency. However, temperature showed a restricted effect on the removal kinetics. The sorption kinetic uptake for MG by PP at various initial dye concentrations was analyzed by non-linear curve fitting analysis method to fit the pseudo-first, pseudo-second and pseudo-nth order equations. The obtained results showed that the pseudo-nth order model fit the experimental data well and was best than the pseudo-first and pseudo-second order equations. The equilibrium data were analyzed using non-linear method by fitting them to the Langmuir, Freundlich and Redlich–Peterson model equations. Both the Langmuir and Redlich–Peterson isotherms represent well the experimental sorption data, but the Redlich–Peterson model was better. The maximum biosorption capacity was found to be 32.39, 34.62 and 35.61 mg g−1 at 25, 35 and 45 °C, respectively. It can be concluded that PP can be an alternative economic biosorbent to more costly adsorbents used for dye removal in wastewater treatment processes.
Acknowledgements
The authors acknowledge the research grant provided by The General Directorate for Scientific Research and Technological Development (PNR project) and the Ministry of Higher Education and Scientific Research of Algeria (Project No. J0101120090018).
References
- J. Hazard. Mater.. 2010;176:820-828.
- J. Hazard. Mater.. 2009;161:1454-1460.
- Water Res.. 1984;18:1501-1507.
- Sep. Purif. Technol.. 2007;53:97-110.
- J. Phys. Chem.. 1906;57:385-470.
- J. Hazrad. Mater.. 2006;B135:264-273.
- Ultrason. Sonochem.. 2008;15:799-807.
- J. Hazard. Mater.. 2008;154:237-244.
- J. Hazrad. Mater.. 2008;153:701-708.
- J. Hazard. Mater.. 2009;167:1089-1094.
- Handl.. 1898;24:1-39.
- Biochem. Eng. J.. 2008;38:390-394.
- J. Hazard. Mater.. 2006;133:196-202.
- J. Phys. Chem. A. 2007;111:8456-8463.
- J. Hazrad. Mater.. 2007;141:753-761.
- Chemosphere. 2009;77:842-847.
- J. Phys. Chem.. 1959;63:1024-1029.
- Trends Food Sci. Tech.. 2001;1:401-413.
- Desalination. 2009;247:430-441.
- Aquat. Toxicol.. 2004;66:319-329.
- Int. Microbiol.. 2000;3:17-24.
- AIChE J.. 1974;20:228-238.
