Desalination 267 (2011) 139146

Contents lists available at ScienceDirect

Desalination
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / d e s a l

Adsorption of 2,4,6-trichlorophenol from aqueous solution onto activated carbon derived from loosestrife
Jinlin Fan, Jian Zhang , Chenglu Zhang, Liang Ren, Qianqian Shi
Shandong Provincial Key Laboratory of Water Pollution Control and Resource Reuse, School of Environmental Science and Engineering, Shandong University, Jinan 250100, China

a r t i c l e

i n f o

a b s t r a c t
An activated carbon was prepared from the stalk of the scrap aquatic plant loosestrife, by H3PO4 activation and then was evaluated for its ability to adsorb 2,4,6-trichlorophenol (TCP). The effects of solution pH, agitation time, TCP initial concentration, and temperature on TCP adsorption were investigated. The equilibrium adsorption data of TCP on activated carbon were analyzed by Langmuir, Freundlich and Temkin isotherm models. The Temkin model gave the best correlation with the experimental data. The adsorption was found to follow the pseudo-second-order kinetics. The intraparticle diffusion model was used to determine the mechanism of the adsorption process. Thermodynamic parameters such as standard enthalpy (Ho), standard entropy (So), and standard free energy (Go) were obtained. Adsorption of TCP is exothermic with Ho at 12.87 kJ/mol. 2010 Elsevier B.V. All rights reserved.

Article history: Received 18 April 2010 Received in revised form 9 September 2010 Accepted 10 September 2010 Available online 8 October 2010 Keywords: Loosestrife activated carbon Adsorption Equilibrium isotherm Kinetics 2,4,6-Trichlorophenol

1. Introduction Highly chlorinated phenol derivatives, such as 2,4,6-trichlorophenol (TCP), have been commonly used as pesticides, herbicides, wood preservatives, and defoliants since the early 1930 s [1]. Wastewaters from pesticide, wood, pharmaceuticals, dye manufacturing, paper and pulp industries as well as drinking water treatment are the most important chlorophenol pollution sources [2]. Adverse effects on the human nervous system and many health disorders have been reported to be caused by 2,4,6-TCP, such as respiratory effects, cardiovascular effects, gastrointestinal effects as well as cancer [35]. Evidence of structural chromosomal aberrations in both somatic and germ cells is also available for 2,4,6-TCP [3]. As a result of its high toxicity, carcinogenic properties, and structural stabilization in the environment, the removal of 2,4,6-TCP from the environment is an important remediation goal. Various treatment technologies have been applied to remove chlorophenols from aqueous solutions, including biological treatment [6], photochemical treatment [7], air stripping and incineration [8], adsorption technology using activated clay [4] as well as activated carbons generated from various sources such as oil palm shell [9], polyethylene terephthalate [10], rattan sawdust, and rice straw [11,12]. Among these treatment methods, activated carbons with high surface areas are some of the most effective adsorbents in treating wastewaters containing this pollutant. However, the high cost of the raw materials such as coal used in producing activated

Corresponding author. Tel.: + 86 531 88363015; fax: + 86 531 88364513. E-mail address: zhangjian00@sdu.edu.cn (J. Zhang). 0011-9164/$ see front matter 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.desal.2010.09.016

carbons has limited the extensive use of these adsorbents, which is a major consideration before wide environmental application is feasible [13]. Recently, research interest has been transferred to using low cost activated carbons in wastewater treatment. From the literature, activated carbons prepared from low cost starting materials, mainly agricultural by-products, include coconut shells [1416], peach stones [17], Tamarind wood [18], olive kernels [19], bamboo [20], rice husk [21], coffee endocarp [22,23] and bean pods [24]. Although most of these materials are available for production of activated carbons, two disadvantages still limit the use of these methods: rst, most of these starting materials come from xerophiles (drought-tolerant plants), the inner structure of which is compact and thus activated carbons prepared from these sources usually have relatively low surface areas as well as small pore numbers; second, none of these agricultural by-products are broadly available and many of them have seasonal limitations. Considering all these factors, new plant materials used for producing activated carbons should be hydrophytes, in order to make an adsorbent with a high surface area, and they should be widely available. Loosestrife, which belongs to the Lythraceae, is a perennial plant which can be seen in many tropical and subtropical areas not only in China but also in many countries across the world. A hardy and beautiful plant with varieties such as purple loosestrife and crape myrtle, loosestrife is widely planted as a landscape plant. After use, most of the adult plants are simply discarded or burned, which is a waste of resources and can lead to environmental pollution. Hence, recycling of such solid waste is imperative. As a hydrophyte, the inner structure of the stems of loosestrife are thin and lacunose, which can be a huge advantage when used to produce quality activated carbons with high surface areas and abundant pores.

140

J. Fan et al. / Desalination 267 (2011) 139146

The goal of this study was to nd the optimum production conditions for making activated carbon from loosestrife and to investigate the adsorption potential of loosestrife based activated carbon (LMC) for 2,4,6-TCP. The equilibrium and kinetic data of the adsorption studies were analyzed to understand the adsorption kinetics, isotherms, and thermodynamics, as well as the mechanism of 2,4,6-TCP adsorption on the prepared activated carbon. 2. Materials and methods 2.1. 2,4,6-Trichlorophenol 2,4,6-Trichlorophenol (TCP) used as an adsorbate in this study was supplied by Tianjin Guangfu chemical reagents Co.( Tianjin, China), and was used without any purication. The molecular weight of 2,4,6TCP is 197.45 g/mol and it has a chemical formula of C6H3Cl3O. 2.2. Preparation of activated carbon The loosestrife used for preparation of activated carbon in this study was collected from a local hydrophyte research institute in Shandong province, China. The precursor was rst cut into small pieces and washed with distilled water ve times to remove surface adhered particles and water-soluble impurities and then dried. The dried loosestrife was then crushed in a laboratory mill. After crushing, the material was dried again and immersed in 40 wt.% phosphoric acid solution at a ratio of precursor: H3PO4 of 1:2 (w/w). After soaking for 12 h at room temperature, the mass was then transferred to a mufe furnace and carbonized. Since mesopores together with micropores in the activated carbon will enhance its adsorption capacity, especially for large molecules of adsorbates [25], we chose to control the conditions to obtain loosestrife activated carbon (LMC) with a highly mesoporous structure. The formation of the carbon's porous networks was greatly inuenced by process of the activation [14] and the activation temperature and time should be controlled in order to obtain the highest surface area and best pore volume. After a series of experiments to investigate different conditions, in this study the mass was heated slowly from room temperature to 450 C and held for 1 h. Then the activated product was cooled to room temperature and washed repeatedly with hot distilled water and 0.1 mol/L NaOH until the pH of its ltrate reached 67. The resulting activated carbon was then dried at 120 C for 12 h in a vacuum oven. The resulting LMC was dried at 120 C for 2 h and the dried product was then sieved to 160180 mesh by standard sieves (Model 200) before storage in a desiccator for later use. 2.3. Batch adsorption equilibrium studies Batch adsorption equilibrium experiments were performed to investigate the effects of initial concentration, adsorbent dose, solution pH, and contact time on the adsorption uptake. The pH was adjusted with 0.1 mol/L NaOH and 0.1 mol/L HCl solutions. In each batch experiment, a 250 mL stoppered conical ask containing 100 ml 2,4,6-TCP of certain concentration was agitated mechanically at the controlled temperature in a isothermal water bath shaker to reach equilibrium, then the mixture was ltered through a syringe lter to minimize interference between the carbon nes and the analysis. The amount of 2,4,6-TCP adsorbed at equilibrium, qe (mg/g), was calculated by the following equation: qe = C0 Ce V W 1

2.4. Effect of solution temperature To study the effect of temperature on 2,4,6-TCP adsorption, the experiments were studied at 20, 35 and 50 C using different adsorption isotherms, with different concentrations, while other variables such as activated carbon dose, rotation speed remained constant and the solution pH was initial without any adjustment. 2.5. Batch kinetic studies The kinetic experiments were similar to the batch equilibrium tests, however, the samples were taken at certain time intervals, ranging from 1 to 600 min. The amount of 2,4,6-TCP adsorbed (from initial concentrations of 50 mg/L, 100 mg/L and 150 mg/L) at time t, qt (mg/g), was calculated as follows: qt = C0 Ct V W 2

where Ct (mg/L) is the concentration of 2,4,6-TCP at any time t (min). 2.6. Desorption studies Desorption studies were conducted using ethanol desorption technique. After, batch equilibrium tests were performed under the conditions: the adsorbent dose = 0.10 g/100 mL, the concentration of 2,4,6-TCP = 150 mg/L, t = 24 h, the solution pH was kept original, the adsorbent was ltered and gently washed with distilled water to remove any unadsorbed 2,4,6-TCP. Then the sample was immediately mixed with 100 mL 95 vol.% ethanol solution, and agitated at 25 1 C for 24 h. The concentration of adsorbate adsorbed at equilibrium was calculated as the difference between the initial and equilibrium concentration (C0 Ce). After desorption, the concentrations of TCP desorbed, Cde (mg/L) was similarly measured using the UVvis spectrophotometer. The percent desorption was calculated using Eq. (3). Desorption% = C0 Ce 100 Cde 3

2.7. Analytical methods Structural characterization of the loosestrife activated carbon was done by N2 adsorption isotherm at 77 K using an automated surface area and pore size analyzer (Quantachrome Corporation, QUADRASORB SI, USA). The specic surface area (SBET) was calculated by the BrunauerEmmettTeller (BET) method. The BET surface area, total pore volume (VT), average pore radius (D), mesopore volume (Vme) and micropore volume (Vmi) were obtained from the N2 adsorption isotherms. The DFT method was used in the pore size distribution determination and the t-plot method was applied to calculate the pore volumes. The Boehm titration method was used to qualitatively and quantitatively identify the surface functional groups of the activated carbons [26]. First, 0.5 g of LAC was placed in 25 mL of 0.1 mol/L solutions of sodium hydroxide, sodium carbonate, sodium bicarbonate and hydrochloric acid. The sealed vials were shaken at room temperature for 24 h. Then each solution was ltrated and the excess base or acid was back-titrated with HCl or NaOH. The numbers of all acidic sites were calculated under the assumption that NaOH reacts with carboxylic groups, HCl reacts with basic surface functional groups, Na2CO3 reacts with carboxylic and lactonic groups and NaHCO3 reacts only with carboxylic groups. The pH value at the zero point of charge (pHzpc) on the surface of LAC was estimated by mass titration [27]. The pH of each solution was measured with a digital pH meter (Model pHS-3C). The initial and nal concentrations of 2,4,6-TCP

where C0 and Ce are the 2,4,6-TCP concentrations at the initial time and at equilibrium (mg/L), respectively. V is the volume of the solution (L) and W is the mass of adsorbent used (g).

J. Fan et al. / Desalination 267 (2011) 139146 Table 1 Pore characteristics and groups on the surfaces of LAC. Sample LAC Sample LAC SBET (m2/g) 1255.75 Total basicity (mmol g 1) 0 Sext (m2/g) 814.79 Total acidity (mmol g 1) 3.3231 Smic (m2/g) 440.97 Carboxylic (mmol g 1) 1.0800 Vt (cm3/g) 1.20 Phenolic (mmol g 1) 1.5071 Vmic (cm3/g) 0.20 Lactone (mmol g 1) 0.7360

141

Vmeso (cm3/g) 1.00 pHzpc 4.75

solutions were determined on a UVvisible spectrophotometer (UV 754, Shanghai) at its maximum absorbance wavelength of 292 nm. Solutions of known 2,4,6-TCP concentration were used to produce a calibration curve based on standard 2,4,6-TCP solutions. FTIR spectra were recorded in Thermo-Nicolet Avatar FTIR spectrometer (VERTEX70, resolution 0.5 cm 1) within the range of 4004000 cm 1 with the samples prepared as KBr discs. 3. Results and discussion 3.1. Characterization of activated carbons

The FTIR spectrum of raw activated carbon and the activated carbon after adsorption process are shown in Fig. 2. The presence of bands in the 800600 cm 1 can be attributed to the CCl stretching vibrations, another band in the 14701380 cm 1 range might consists of a series of overlapping absorption bands ascribable to the deformation vibration of surface hydroxyl groups and in-plane vibrations of CH in various C = CH structures [29], all are FTIR characteristics of 2,4,6-TCP. The results conrmed that the 2,4,6-TCP has been successfully adsorbed by the activated carbon prepared in this study. There are no signicant changes to surface functional groups. 3.2. Effect of adsorbent dose on TCP adsorption

The adsorption and desorption isotherms of N2 at 77 K for the LAC samples are classied as type IV according to the IUPAC classication [28]. Type IV isotherms are characteristic of the simultaneous presence of micro- and mesopores. Porous structure parameters for the sample studied are listed in Table 1. The derived activated carbon is a high quality adsorbent with surface area of 1255.75 m2/g. The total pore volume is 1.20 cm3/g and mecroporous volume is 1.00 cm3/g, indicating that the activated carbons contain mostly mesopores with only a small contribution of micropores. The presence of mesopores together with micropores in the activated carbon will enhance its adsorption capacity, especially for large adsorbate molecules [25]. According to the IUPAC denition, adsorbent pores are classied into three groups: micropore width ( b 2 nm), mesopore width (2 50 nm) and macropore width (N 50 nm). The pore size distribution of LAC is shown in Fig. 1. The sample exhibits a multimodal distribution in both the micropore and mesopore domains but does not have macropores, which is mainly attributed to successful control of the preparation conditions. According to the results of the Boehm titration (Table 1), the LAC samples have more acidic groups on their surfaces than basic ones and the relative percentages of the acidic groups followed the order phenol N carboxyl N lactone. LAC had a pHzpc of 4.75, at which the net surface charge on carbon was zero. At pH b pHzpc, the carbon surface had a net positive charge, while at pH N pHzpc the surface had a net negative charge.

The adsorbent dose has a great inuence on the adsorptive removal of 2,4,6-TCP by LAC, as shown in Fig. 3. The removal percentage increased rapidly in the beginning with the increase in the adsorbent dose and after the critical dose of 0.1 g/100 mL 2,4,6-TCP solution was reached, the adsorption remained almost constant as adsorption equilibrium was reached. The curve's shape is due to the increased adsorbent surface area and availability of a large number of vacant surface sites during the initial stage [30]. Above the critical dose, the adsorption capacity of 2,4,6-TCP on LAC showed no signicant difference due to saturation at the exterior surface as well as the repulsive forces among the solute molecules [4,7]. Therefore, further equilibrium experiments were carried out at a 0.10 g LAC/100 mL TCP solution in order to maximize the adsorptive capacity of the activated carbon. 3.3. Effect of TCP initial concentration and agitation time on adsorption equilibrium The effect of initial concentration and agitation time on the adsorptive removal of TCP by LAC at 25 C is shown in Fig. 4. From the view of the whole adsorption process, the plot can be divided into three stages: at the initial stages of contact, TCP of different concentrations were adsorbed on the LAC rapidly and then comes to the second stages after certain times, during which the adsorption

0.40 0.35
852 792 728 Raw LAC LAC after adsorption

pore volume (cm3/nm/g)

0.30 0.25 0.20 0.15 0.10 0.05 0.00

Transmittance (%)

1467

4000 0 5 10 15 20 25

3500

3000

2500

2000

1500

1000

500

Pore diameter (nm)

Fig. 1. Pore size distribution of LAC.

Wavenumbers

(cm-1)

Fig. 2. FTIR analysis of raw activated carbon and the activated carbon after adsorption process.

142

J. Fan et al. / Desalination 267 (2011) 139146

110 100

study are comparable with previously studied activated carbons, but the adsorption performance was much better. 3.4. Effect of solution pH on TCP adsorption

Removal efficiency (%)

90 80 70 60 50 0.00
The effect of solution pH on the adsorptive removal of TCP was studied by varying the initial pH of the solution from 2 to 12 at an initial concentration of 150 mg/L (Fig. 5.) The highest adsorption of TCP on this carbon occurred at pHs from 2 to the initial pH of the solution, where the removal percentage remained almost constant. The removal decreased remarkably as the pH increased from the initial pH of the solution to 12, which is comparable to the results obtained by Tan et al. [30]. TCP is a weak acid. When the pH of the solution is below the pKa of TCP, dispersion interactions with the LAC dominate [30,31]. At high pH, TCP is partially ionized and the negatively charged TCP ions are repelled by the negatively charged LAC surface. The low TCP removal at high pH may also be a result of the competition between the OH ions and the TCP anions. From the results of the experiments, we chose to leave the pH of the TCP solution unadjusted in further experiments. 3.5. Effect of solution temperature on TCP adsorption As shown in Fig. 6, the TCP adsorption uptake, qe (mg/g) on LAC decreased with increasing solution temperature from 20 C to 50 C for all initial concentrations (25200 mg/L), indicating the adsorption was a exothermic reaction. A similar trend was also observed by Bilgili [32] with 4-chlorophenol adsorbed onto XAD-4 resin. The decrease in adsorption uptake with the rise of temperature may be attributed to the weakening of adsorptive forces between the active sites of the adsorbent and adsorbate species as well as between the adjacent molecules of the adsorbed phase [33]. In this study, all further experiments were carried out at 25 C in an isothermal water bath shaker. 3.6. Adsorption isotherms Equilibrium data which was commonly known as adsorption isotherms indicate how the adsorption molecules distribute between the liquid phase and the solid phase in the whole adsorption process, and give a comprehensive understanding of the nature of adsorption. The equilibrium data obtained from different models provides important information on the properties of the adsorbent surface. In this study, three isotherm models (Langmuir, Freundlich and Temkin) were applied to describe the equilibrium characteristics of adsorption.

0.04

0.08

0.12

0.16

Adsorbent dose (g/100ml)

Fig. 3. Effect of adsorbent dose on the removal of 2,4,6-TCP onto LAC. 2,4,6-TCP concentration 150 mg/L, temperature 25 1 C, contact time 360 min.

gradually decreases, nally, the adsorption reached a dynamic equilibrium where the amount of TCP being adsorbed on the activated carbon was approximate to the amount of TCP desorbing from the activated carbon. The reason for the rapid adsorption at the initial contact time was mainly due to the availability of a large number of vacant surface sites, and the slow rate of TCP removal after certain times can be attributed to the repulsive forces between the solute molecules and the decrease of vacant surface sites, besides, the adsorption at the exterior surface reached a saturation at this time then the solute molecules began to enter the pores of the carbon and were adsorbed by the interior surface of the particles [4,7]. Increasing the initial TCP concentration led to an increase in TCP uptake. With an increase in initial concentration from 50 to 200 mg/L, the adsorption capacity at equilibrium, qe, increased from 48.93 to 191.05 mg/g. There is probably a higher mass transfer driving force when the initial TCP concentration is increased, due to the increase in the number of ions competing for the available binding sites on sorbent [30]. Higher initial concentrations of TCP solutions required longer contact times to reach equilibrium, but for TCP solutions with initial concentrations of 50150 mg/L, there was a high adsorption rate, and the contact times required to reach dynamic equilibrium were less than 20 min, much better than the results obtained in other studies [30]. For TCP solutions with initial concentrations of 200 mg/L, the contact times were about 56 h to reach dynamic equilibrium. The adsorption characteristics of the activated carbon prepared in this

300
50 mg/L 100 mg/L 200 mg/L

250 200

150 mg/L

100

Removal efficiency (%)

0 100 200 300 400 500 600

80 60 40 20 0

qt (mg/g)

150 100 50 0

t/min
Fig. 4. Effect of initial concentrations (50200 mg/L) and agitation time on 2,4,6-TCP adsorption on LAC at 25 1 C. Agitation speed: 130 rpm; adsorbent dose: 0.1 g/ 100 ml; pH of solution: natural.

10

12

pH
Fig. 5. Effect of solution pH on 2,4,6-TCP removal at 251 C. Initial concentration= 150 mg/L; agitation speed: 130 rpm; adsorbent dose: 0.1 g/100 ml.

J. Fan et al. / Desalination 267 (2011) 139146

143

250
25 mg/L 150 mg/L 50 mg/L 175 mg/L 100 mg/L 200 mg/L

linearly with coverage due to these interactions [37]. The linear form of the Temkin isotherm is expressed as:     RT RT ln A + ln Ce bT bT

200

qe =

t (mg/g)

150

100

50

293

308

323

solution temperature (K)

Fig. 6. Effect of solution temperature on 2,4,6-TCP uptake at various initial concentrations.

The Langmuir isotherm [34] is based on the assumption that adsorption takes place at specic homogeneous sites within the adsorbent and once a TCP molecule occupies a site, no further adsorption takes place at that site. The linear form of the Langmuir isotherm equation is given as: Ce 1 C = + e Qm KL qe Qm 4

where Ce is the concentration of the TCP solution (mg/L) at equilibrium, qe is the amount of adsorbate adsorbed per unit mass of adsorbent at equilibrium (mg/g), Qm and KL are the Langmuir constants related to adsorption capacity (mg/g) and free energy of adsorption (L/mg), respectively. In order to determine if the adsorption process is favorable or unfavorable, a dimensionless constant separation factor or equilibrium parameter RL, is dened according to the following equation [35]: RL = 1 1 + KL C0 5

where KL is the Langmuir constant (L/mg) and C0 is the initial TCP concentration (mg/L). The value of RL indicates the type of isotherm to be favorable (0 b RL b 1), linear (RL =1), unfavorable (RL N 1), or irreversible RL =0. The Freundlich isotherm is an empirical equation assuming that the adsorption process takes place on a heterogeneous surface through a multilayer adsorption mechanism, stronger binding sites are occupied rst and that the binding strength is related to the concentration of TCP at equilibrium [36]. The linear form of which can be given by the equation: ln qe = ln KF + 1 ln Ce n 6

where B (J/mol) = RT/bT is the Temkin constant which related to heat of adsorption while A (L/g) is the equilibrium binding constant. R (8.314 J/mol K) is the universal gas constant and T (K) is the absolute solution temperature. All the correlation coefcients, the constants and the R2 values obtained from the three isotherm models applied for adsorption of 2,4,6-TCP at 20, 35 and 50 C are summarized in Table 2. The calculated RL values versus the initial concentration of TCP at three different temperatures are shown in Fig. 7. All the RL values were between 0 and 1, indicating that the adsorption of TCP on the activated carbon was favorable in this study and as the initial concentration increased the values of RL decreased which showed that the adsorption was more favorable at higher initial concentration. The Temkin isotherm model gave the highest R values which were greater than 0.98 at all three temperatures studied, showing that the adsorption of TCP on the activated carbon was best described by this model. The constant B showed that the heat of TCP adsorption decreased with temperature, indicating exothermic adsorption. This is consistent with the results of the adsorption thermodynamics. Parameter uctuations were observed at different temperatures in the Langmuir and Freundlich isotherm models, which restricted the general application of these two isotherm models to our system, however, certain valuable information can also be observed from the parameters of the models. All the 1/n values obtained from the Freundlich model were below 1 at all solution temperatures, representing that adsorption was favorable, and the Freundlich model gave a better t than the Langmuir model. The results was compatible with previous work on the adsorption of chlorophenols [30,38,41]. The maximum monolayer adsorption capacity, qm, which has been widely used to compare the efciency of an absorbent, can also be obtained from Table 2. In Table 3 there is a comparison of the maximum monolayer adsorption capacity of some phenols on various adsorbents. The activated carbon prepared in this work had a relatively large TCP adsorption capacity of 367.65 mg/g, as compared to some previous studies reported in the literature. The large adsorption capacity of this activated carbon could be a result of its relatively high surface area (1255.75 m2/g) and its mesoporous structure.
Table 2 Langmuir, Freundlich, Temkin isotherm model constants and correlation coefcients for adsorption of 2,4,6-TCP onto LAC. Isotherms Solution temperature (K) Constants Qm (mg/g) Langmuir 293 308 323 367.6471 284.0909 214.5923 KL (L/mg) 0.15120 0.18945 0.09109 0.5222 0.5305 0.6905 R2
1/n

R2

where KF ((mg/g)(L/mg)1/n) and n are the Freundlich constants. KF can be dened as the distribution coefcient and indicative of the relative adsorption capacity of the adsorbent, commonly the adsorption capacity of adsorbent increases as the KF value increases. The value of n is indicative of the intensity of the adsorption. If the slope 1/n is below 1, it indicates a normal Langmuir isotherm while 1/n above 1 indicates cooperative adsorption. Log qe versus log Ce gave a straight line with slope 1/n and intercept log KF. The Temkin isotherm takes the effects of indirect adsorbate adsorbate interactions on adsorption into account, and suggests that the heat of adsorption of all the molecules in the layer would decrease

Isotherms

Solution temperature (K) Constants KF ((mg/g)(L/mg) ) 1/n 0.8756 0.9597 0.9905

Freundlich 293 308 323 Isotherms

24.0634 21.2193 16.6011

0.9009 0.9514 0.9535 R2

Solution temperature (K) Constants A (L/mg) B 92.1234 79.7628 76.5266

Temkin

293 308 323

0.9987 0.9051 0.7542

0.9981 0.9989 0.9880

144

J. Fan et al. / Desalination 267 (2011) 139146

0.20
293 K 308 K 323 K

4 2 0

0.16

RL

0.12

ln (qe-qt)

-2 -4
50 mg/L 100 mg/L 200 mg/L

0.08 -6 0.04 50 100 150 200 -8 -10 0 100

150 mg/L

200

300

400

500

600

C0 (mg/L)
Fig. 7. Separation factor RL of 2,4,6-TCP adsorbed onto LAC.

t (min)
Fig. 8. Pseudo-rst-order kinetic plots for the adsorption of 2,4,6-TCP adsorbed onto LAC at 25 1 C.

3.7. Adsorption kinetic studies The adsorption kinetics describes the efciency and the rate of the process and it controls the equilibrium time. In order to investigate the mechanism of adsorption of TCP on this activated carbon, The pseudo-rst-order, pseudo-second-order kinetic models and Elovich equation were studied and then the intraparticle diffusion model was further tested to determine the diffusion mechanism. 3.7.1. Pseudo-rst-order kinetic model The pseudo-rst-order equation given by Langergren and Svenska [42] can be expressed as: lnqe qt = ln qe k1 t 8

where k2 is the rate constant of pseudo-second-order adsorption (g/(mg min)). The linear plot of t/qt versus t determined the slope 1/qe and the intercept 1/k2q2e (Fig. 9). From Table 4, the values of the correlation coefcient for the second-order kinetic model were found to be in the range 0.99991, indicating excellent applicability of the pseudo-second-order kinetic model to describe the adsorption process of TCP on LMC. The experimental qe values were close to the calculated data at different initial concentrations, which also shows good agreement. We conclude that the pseudo-second-order adsorption is predominant and that chemisorption mainly controls the rate of the TCP adsorption process [44,45]. 3.7.3. Elovich equation The Elovich equation is one of the most useful models for describing chemisorption. The equation is given as [46]: qt =   1 1 lnab + ln t b b 10

where qe and qt are the amounts of TCP adsorbed at equilibrium and at time t in mg/g, respectively, and k1 is the pseudo-rst-order rate constant (min 1). The values of qe and k1 for the pseudo-rst-order kinetic model are given by the intercepts and the slopes of the plots of ln(qe qt) versus t, respectively (Fig. 8). From Table 4, for the rstorder kinetic model, the obtained R2 values were surprisingly low and the calculated qe(cal) were much lower than the experimental qe(exp), which might attributed to the adsorption rate and as a result the values of ln(qe qt) varied considerably. Therefore, the adsorption of TCP on the activated carbon is not a rst-order reaction, especially compared to the pseudo-second-order kinetic model which agrees well with the reaction process. 3.7.2. Pseudo-second-order kinetic model The pseudo-second-order kinetic model can be given in the following form [43]: t 1 1 = + t qt qe k2 q2 e 9

where a (mg/(g min)) is the initial sorption rate and b (g/mg) is the desorption constant related to the extent of surface coverage and activation energy for chemisorption. As shown in Fig. 10, The R2 values obtained from Elovich equation were in the range of 0.7620.974 for TCP initial concentrations of 50200 mg/L (Table 4). The parameters (1/b) and (1/b)ln(ab) were given by the slope and intercept of the linear plot of qt versus ln t. The parameter 1/b is related to the number of sites available for adsorption while the (1/b)ln(ab) is the adsorption quantity when ln t is equal to zero. This value can help to explain the adsorption behavior of the rst step [47]. 3.7.4. Intraparticle diffusion study As the above kinetic models were not able to identify the diffusion mechanisms and rate controlling steps in the adsorption process, the intraparticle diffusion model based on the theory proposed by Weber

Table 3 Comparison of maximum monolayer adsorption capacity of various chlorophenols on various adsorbents. Adsorbent Loosestrife-based activated carbon Activated clay EFB-based activated carbon Commercial grade coconut shell-based activated carbon Paper mill sludge rattan sawdust activated carbon Adsorbate 2,4,6-Trichlorophenol 2,4,6-Trichlorophenol 2,4,6-Trichlorophenol 2,4,6-Trichlorophenol 2,4-Dichlorophenol 4-Chlorophenol Maximum monolayer adsorption capacity (mg/g) 367.65 123.46 500.00 112.35 4.49 188.68 References This work 4 26 38 39 40

J. Fan et al. / Desalination 267 (2011) 139146 Table 4 Pseudo-rst-order model, pseudo-second-order model and Elovich equation constants and correlation coefcients for adsorption of 2,4,6-TCP onto LAC at 25 1 C. Initial TCP concentration (mg/L) 50 100 150 200 qe,exp (mg/g) 48.9316 98.5568 146.7474 193.3286 Pseudo-rst-order kinetic model q e,
cal

145

Pseudo-second-order kinetic model q e,

cal

Elovich equation (1/b)ln(ab) (mg/g) 45.7977 92.0654 135.8090 158.3380 1/b (mg/g) 0.5158 1.0898 1.9189 5.6031 R2 0.6354 0.5515 0.6545 0.9018

(mg/g)

k1 (1/min) 0.00198 0.00134 0.00354 0.00301

R2 0.1805 0.0655 0.4330 0.3868

(mg/g)

k2 (g/mg min) 0.1261 0.03523 0.02793 0.01819

R2 0.9999 0.9999 1 0.9999

0.9397 1.5200 3.2435 14.8988

48.5201 98.4252 145.9854 188.3239

and Morris [48] was tested. It is an empirical functional relationship, common to most adsorption processes, where uptake varies almost proportionally with t1/2. According to this theory, the intraparticle diffusion equation can be expressed as: qt = kpi t
1=2

the only rate controlling step, some other rate controlling steps might be involved and may affect the adsorption. 3.8. Adsorption thermodynamics Entropy change is the driving force of adsorption processes. Changes in standard entropy (So), standard enthalpy (Ho), standard free energy (Go) were obtained in order to determine the thermodynamics of the whole adsorption process. The values of Ho and So can be calculated from the following equation: So Ho R RT 12

+C

11

where kpi (mg/g min1/2) is the intraparticle diffusion rate constant of stage i, and C is the intercept, gives an idea about the thickness of boundary layer, i.e., the larger the intercept, the greater the boundary layer effect. If the adsorption process follows the intraparticle diffusion model, then qt versus t1/2 will be linear and if the plot passes through the origin, then intraparticle diffusion is the sole ratelimiting step. Otherwise, some other mechanism along with intraparticle diffusion is also involved. Fig. 11 shows a plot of the amount of TCP adsorbed versus t1/2 for different initial concentrations. The regression was not linear over the whole time range, which suggests that more than one mode of sorption along with intraparticle diffusion is involved in the TCP adsorption. The values of kpi, Ci as well as the correlation coefcients, R2 obtained from the plots are given in Table 5. The values of kpi generally increased as the initial TCP concentration increased, which can be attributed to the greater driving force. As mentioned in the literature [30], the adsorption mechanism of an adsorption process may be assumed to involve three steps, and the three sequential steps in the adsorption are: i. lm diffusion, ii. pore diffusion or intraparticle diffusion, iii. adsorption of the adsorbate ions on the interior surface of the adsorbent. As shown in Fig. 11, the plots in this experiment presented that two steps occurred in the process. The rst, sharper portion was the external surface adsorption or the instantaneous adsorption. The second portion was the gradual adsorption stage where intraparticle diffusion was rate-limiting. In Fig. 10, there was a linear relationship over a period of time, but the lines did not pass through the origin, which suggested that intraparticle diffusion was present, but was not

ln KL =

where KL (L/mg) is the Langmuir isotherm constant, R (8.314 J/mol K) is the universal gas constant and T (K) is the absolute temperature of the solution. The values of Ho and So can be given by the slope and intercept of the van't Hoff plot of lnKL versus 1/T, respectively. Go can then be calculated by the equation: G = H T S
o o o

13

The thermodynamic parameters such as Ho, So, Go of the adsorption process are listed in Table 6. The negative value of So shows the entropy at the solidsolution interface was decreased during the adsorption process. The negative Ho value shows that the adsorption is exothermic, which is consistent with the results of the adsorption isotherms, and the decreased TCP uptake with increased solution temperature. The free energy changes (Go) were positive, reecting the nonspontaneous nature of the adsorption process over the range of temperatures in the experiment. Kara et al. [49] suggested that the Ho of physisorption is smaller than 40 kJ/mol. Generally, Go for physisorption is less than that for chemisorption. The former is between 20 and 0 kJ/mol and the latter is between 80 and 400 kJ/mol [50]. Therefore, Ho, Go all suggested the

14
50 mg/L 100 mg/L 200 mg/L

300
50 mg/L 100 mg/L 200 mg/L

12
150 mg/L

250 200 150 100

150 mg/L

t/qt (min g/mg)

10 8 6 4 2 0 50 0 100 200 300 400 500 600 0 1 2 3 4 5 6 7

t (min)
Fig. 9. Pseudo-second-order kinetic plots for the adsorption of 2,4,6-TCP adsorbed onto LAC at 25 1 C.

qt (mg/g)

ln t
Fig. 10. Plot of Elovich equation for adsorption of 2,4,6-TCP adsorbed onto LAC at 25 1 C.

146

J. Fan et al. / Desalination 267 (2011) 139146

300
50 mg/L 100 mg/L 200 mg/L

Table 6 Thermodynamic parameters for the adsorption of adsorption of 2,4,6-TCP onto LAC. H(kJ/mol) 12.87 S(J/mol K) 58.35 G-(kJ/mol) 293 K 4.23 308 K 5.15 323 K 5.98

250 200 150 100 50 0

150 mg/L

qt (mg/g)

National Science and Technology Project (No. 2009ZX07210-009), Special Science Innovation Foundation of Shandong University (2009JQ009) and National Natural Science Foundation (21007032). The authors thank Dr. Pamela Holt for editing the manuscript. References
[1] C. Tai, G. Jiang, Chemosphere 59 (2005) 321326. [2] R. Gao, J. Wang, J. Hazard. Mater. 145 (2007) 398403. [3] Agency for toxic substances disease registry (ATSDR), Toxicological Prole for 2,4,6-Trichlorophenol, U.S. Public Health Service, U.S. Department of Health and Human Services, Atlanta, GA, 1990 [4] B.H. Hameed, Coll. Surf. A. 307 (2007) 4552. [5] Ph. Howard, Handbook of Environmental Degradation Rates, Lewis Publishers, MI, 1991. [6] Mehmet Dilaver, Fikret Kargi, Biores. Tech. 100 (2009) 14591462. [7] S.G. Poulopoulos, M. Nikolaki, D. Karampetsos, C.J. Philippopoulos, J. Hazard. Mater. 153 (2008) 582587. [8] S. Rengaraj, X.Z. Li, J. Mol. Catal. A. 243 (2006) 6067. [9] I.A.W. Tan, A.L. Ahmad, B.H. Hameed, Biores. Tech. 100 (2009) 14941496. [10] K. Laszlo, A. Szcs, Carbon 39 (2001) 19451953. [11] B.H. Hameed, A.A. Rahman, J. Hazard. Mater. 160 (2008) 576581. [12] S.L. Wang, Y.M. Tzou, Y.H. Lu, G. Sheng, J. Hazard. Mater. 147 (2007) 313318. [13] F.J. Guymont, in: I.H. Suffet, M.J. McGuire (Eds.), Ann Arbor Science, vol. 2, Ann Arbor,MI, 1984, (Chapter 23). [14] M.K.B. Gratuito, T. Panyathanmaporn, R.-A. Chumnanklang, N. Sirinuntawittaya, A. Dutta, Biores. Tech. 99 (2008) 48874895. [15] I.A.W. Tan, B.H. Hameed, A.L. Ahmad, Chem. Engn. J. 137 (2008) 462470. [16] I.A.W. Tan, A.L. Ahmad, B.H. Hameed, J. Hazard. Mater. 153 (2008) 709717. [17] Ayrton F. Martins, Andr de.L. Cardoso, Joo A. Stahl, Juraci Diniz, Biores. Tech. 98 (2007) 10951100. [18] J.N. Sahu, Jyotikusum Achary, B.C. Meikap, Biores. Tech. 101 (2010) 19741982. [19] A. Zabaniotou, G. Stavropoulos, V. Skoulou, Biores. Tech. 99 (2008) 320326. [20] B.H. Hameed, A.T.M. Din, A.L. Ahmad, J. Hazard. Mater. 141 (2007) 819825. [21] D. Mohan, K.P. Singh, V.K. Singh, J. Hazard. Mater. 152 (2008) 10451053. [22] M. Joo, Valente Nabaisa, Pedro Nunesa, Peter J.M. Carrotta, M. Manuela, L. Ribeiro Carrotta, A. Macas Garcab, M.A. Daz-Dezb, Fuel Proc. T. 89 (2008) 262268. [23] Joo Valente Nabais, Peter Carrott, M.M.L. Ribeiro Carrott, Vnia Luz, Angel L. Ortiz, Biores. Tech. 99 (2008) 72247231. [24] B. Cabal, T. Budinova, C.O. Ania, B. Tsyntsarski, J.B. Parra, B. Petrova, J. Hazard. Mater. 161 (2009) 11501156. [25] S.B. Hartono, S. Ismadji, Y. Sudaryanto, W. Irawaty, J. Ind. Eng. C. 6 (2005) 864. [26] H.P. Boehm, Adv. Catal. 16 (1966) 179274. [27] S.C.R. Santos, V.J.P. Vilar, R.A.R. Boaventura, J. Hazard. Mater. 153 (2008) 9991008. [28] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscou, R.A. Pierotti, J. Rouquerol, T. Siemieniewska, Pur. A. Chem. 57 (1985) 603619. [29] M. Pakua, M. Walczyk, S. Biniak, Chemosphere 69 (2007) 209219. [30] I.A.W. Tan, A.L. Ahmad, B.H. Hameed, J. Hazard. Mater. 164 (2009) 473482. [31] O. Hamdaoui, E. Naffrechoux, J. Hazard. Mater. 147 (2007) 381394. [32] M.S. Bilgili, J. Hazard. Mater. 137 (2006) 157164. [33] K.K. Panday, G. Prasad, V.N. Singh, Water A. S. P. 27 (1986) 287296. [34] I. Langmuir, J. Am. Chem. S. 40 (1918) 1361. [35] T.W. Weber, R.K. Chakkravorti, Alche. J. 20 (1974) 228238. [36] H.M.F. Freundlich, J. Phys. Chem. 57 (1906) 385470. [37] M.I. Temkin, V. Pyzhev, Acta PhysC. 12 (1940) 327356 URSS. [38] A. Kuleyin, J. Hazard. Mater. 144 (2007) 307315. [39] M. Radhika, K. Palanivelu, J. Hazard. Mater. B138 (2006) 116124. [40] N. Calace, E. Nardi, B.M. Petronio, M. Pietroletti, Envir. Pollu. 118 (2002) 315319. [41] B.H. Hameed, L.H. Chin, S. Rengaraj, Desalination 225 (2008) 185198. [42] S. Langergren, B.K. Svenska, Veternskapsakad Handlingar 24 (4) (1898) 139. [43] Y.S. Ho, G. McKay, Can. J. Ch. En. 76 (1998) 822826. [44] Y.S. Ho, G. McKay, Process Bio. 34 (1999) 451465. [45] Suhong Chen, Jian Zhang, et al., Desalination 252 (1) (2010) 149156. [46] M. Ozacar, I.A. Sengil, Process Bio. 40 (2005) 565572. [47] R.L. Tseng, J. Coll. I. Sc. 303 (2006) 494502. [48] W.J. Weber, J.C. Morris, Proc. Int. Conf. Water Pollution Symposium, vol. 2, Pergamon, Oxford, 1962, pp. 231266. [49] M. Kara, H. Yuzer, E. Sabah, M.S. Celik, Water Res. 37 (2003) 224232. [50] M.J. Jaycock, G.D. Partt, J. Coll. I. Sc. (1981) 1213.

10

15

t1/2 (min1/2)
Fig. 11. Plot of intraparticle diffusion model for adsorption of 2,4,6-TCP adsorbed onto LAC at 25 1 C.

same fact: the adsorption of 2,4,6-TCP onto the activated carbon prepared in this study was mainly a physisorption process. 3.9. Desorption studies The feasibility of regenerating the spent activated carbon saturated with TCP was evaluated using ethanol desorption technique. The results of the TCP percent desorption was as high as 99.4%, which was in accord with some previous work [30]. Thus, as nearly all the adsorption sites were recovered from the desorption process, ethanol desorption technique was observed to be a promising way to regenerate the spent activated carbon in this study. 4. Conclusion Activated carbon prepared from loosestrife is a promising low cost adsorbent for the removal of TCP from aqueous solutions. The surface area of the activated carbon prepared by phosphoric acidactivation was relatively high (1275.5 m2/g) and the structure was found to be a well-developed mesoporous array. Adsorption of TCP increased with increases in agitation time and initial TCP concentration at a certain limit. The original solution pH was favorable for TCP adsorption on the activated carbon. The equilibrium data were best described by the Temkin isotherm model and the adsorption kinetics by pseudosecond-order kinetics. The results of the intraparticle diffusion model suggest that intraparticle diffusion was not the only rate controlling step. The loosestrife used in this work is freely and abundantly available as a perennial plant, which offers greater potential for wastewater treatment than other activated carbons prepared from waste materials, suggesting that it may be a good alternative to more costly adsorbent materials for removal of TCP from aqueous solutions. Acknowledgements This work was supported by the National Key Technology R&D Program for the Eleventh Five-year Plan (No. 2006BAC10B03), Major
Table 5 Intraparticle diffusion model constants and correlation coefcients for adsorption of 2,4,6-TCP onto LAC at 25 1 C. C0 (mg/L) 50 100 150 200 kpi (mg/g min1/2) 0.0893 0.1775 0.3399 1.2394 C 46.9561 94.6463 140.0518 166.5859 R2 0.3156 0.2427 0.3463 0.5792