Water Research 37 (2003) 1535–1544
The adsorption of basic dyes from aqueous solution on modi?ed peat–resin particle
gye Sun, Linzhang Yang*
Nanjing Institute of Soil Science, Chinese Academy of Sciences, P. O. Box No. 821, Nanjing Jiangsu Province 210008, People’s Republic of China Received 12 December 2001; received in revised form 30 September 2002; accepted 9 October 2002
Abstract Modi?ed peat was prepared by mixing thoroughly raw peat with sulfuric acid, and modi?ed peat–resin particle was obtained, by mixing modi?ed peat with solutions of polyvinylalcohol (PVA) and formaldehyde. In this paper, the adsorption of Basic Magenta and Basic Brilliant Green onto modi?ed peat–resin particle is examined. The adsorption isotherm showed that the adsorption of basic dyes on modi?ed peat–resin particle deviated from the Langmuir and Freundlich equations. The pseudo-?rst order, pseudo-second order and intraparticle diffusion models were used to ?t the experimental data. By comparing the standard deviation, it was found that the intraparticle diffusion model could be used to well describe the adsorption of two basic dyes on modi?ed peat–resin particle. According to the change of intraparticle diffusion parameter, the adsorption processes could be divided into different stages. The kinetics experiment also indicated that initial dye concentrations, particle dose and particle size could affect the adsorption processes of basic dyes. r 2002 Elsevier Science Ltd. All rights reserved.
Keywords: Modi?ed peat–resin particle; Basic dyes; Adsorption isotherm; Adsorption kinetics; Kinetic models
1. Introduction Wastewater containing dyes from textile industry is very dif?cult to treat using conventional wastewater treatment methods, since the dyes are stable to light and oxidizing agents, and are resistant to aerobic digestion ([1,2]). Adsorption techniques to remove dyes in solution have been widely used. Most commercial systems currently use activated carbon as adsorbent to remove dyes in wastewater. Activated carbon is expensive, which means higher cost is in wastewater treatment. In order to decrease the cost of wastewater treatment, attempts have been made in ?nding inexpensive adsorbents. Studies showed that many materials, such as low rank coals, peat moss, chitosan and wood, could be used as
*Corresponding author. Tel.: +86-25-3360-884; fax: +8625-3353-590. E-mail address: email@example.com (L. Yang).
adsorbents [2–5] which could effectively remove the dyes from solution. Peat, as an adsorbent, is porous and rather complex material, containing lignin and cellulose. Recently, peat has been used to remove some pollutants (such as heavy metals, dyes and oil) from aqueous solution [6–9]. Many studies showed peat could effectively remove the dyes from aqueous solution ([1,10–15]). Poots et al. [1,2] investigated adsorption of acid dye (Acid blue 25) on peat by ?xed bed. And results showed that peat was a good adsorbent for removal of acid blue 25. Using batch adsorption system, Allen et al. [10–12] studied adsorption processes of acid and basic dyes on peat. In the isotherm experiment of basic dyes, Allen et al.  found the isotherm of basic dyes exhibited deviation from the theory (Langmuir isotherm equation). They thought this deviation was attributed to the pressure of adsorbing ions creating a new surface by expanding the adsorbent particle. Allen et al. [11,12] investigated the diffusion of
0043-1354/03/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 3 - 1 3 5 4 ( 0 2 ) 0 0 5 2 0 - 1
Q. Sun, L. Yang / Water Research 37 (2003) 1535–1544
Nomenclature Ct Ce qe C0 D0 F kf dye concentration in solution at time t (mg l?1) equilibrium concentration (mg l?1) adsorption capacity in equilibrium (mg g?1) initial dye concentration in solution (mg .l?1) diffusion coef?cient (m s?1) fractional uptake of solute rate constant of pseudo-?rst-order model (l min?1)
kI ks M qt r t Q V Dq
rate parameter of intraparticle diffusion model, (mg dye g?1 particle min?0.5) rate constant of pseudo-second-order model (g particle mg–1 dye min?1) mass of modi?ed peat–resin particle (g) amount of adsorption at time t, (mg g?1) particle radius (mm, cm, m) time (min) maximum adsorption capacity (mg g?1) volume of solution (l) standard deviation
acid and basic dyes inside a peat particle, also. They used the intraparticle diffusion model to describe the adsorption processes of basic dyes on peat and divided adsorption processes into separate steps in terms of diffusion rate parameter. Peat, as an adsorbent, can effectively remove the pollutants from solution and is inexpensive. However, when raw peat is directly used in wastewater treatment, there are many limitations, such as low chemical stability and mechanical strength, leach of fulvic acid from peat and dif?cult regeneration . In order to overcome these limitations, we prepared the modi?ed peat–resin particle by mixing modi?ed-peat with polyvinylalcohol (PVA) and formaldehyde. The modi?ed peat–resin particle contains polar functional groups, such as alcohols and acids. Both modi?ed peat and resin in particle can adsorb the dyes from solution. In this paper, the adsorption isotherm and kinetics experiments of basic dyes (Basic Magenta and basic Brilliant Green) were conducted and the different kinetic models were used to analyze adsorption processes of basic dyes on modi?ed peat–resin particle.
2. Gelling: distilled water (300 ml) was mixed with modi?ed-peat and heated it to a boil, then add 10% PVA solution (100 ml) and 37% formaldehyde solution (80 ml) were added with swift, agitation until ‘‘modi?ed peat–resin gel’’ was formed. Then the ‘‘modi?ed peat–resin gel’’ was broken with a hammar water was poured into the powdered mixture (200 ml) and heated it to a boil. 3. Coating: 5% PVA solution (100 ml) was added, agitating and heating. This step was repeated for ?ve or six times. 4. Washing and drying: the particle was washed with deionized water until ?ltrate reached pH 5.5—6.5, then it was dried in an oven at 65751C for 24 h. 5. Sieving: the particle was crushed and sieved. With regard to the modi?ed peat–resin particle, the basic structure was that resin adhered modi?edpeat powder into microparticles and, then, many microparticles formed gradually particle with different size. The modi?ed peat–resin particle had a macroreticular structure. The porosity and the ratio of macropore in the modi?ed peat–resin particle changed with the particle size. Both modi?ed-peat and resin could adsorb the basic dyes in the adsorption processes. Fig. 1 shows the pore distribution of modi?ed peat– ( resin particle. Fig. 1, shows that micropore (o 20 A) and ( transitional pore (20—50 A) are predominant in the modi?ed peat–resin particle. Table 1 shows the speci?c surface area of the particle (measured by nitrogen adsorption method, ASAP2000, MICROMERTICS). Table 1 shows that the speci?c surface area of modi?ed peat–resin particle is very low at different size. We think the speci?c surface area presented in Table 1 is lower than the true one. The low value of speci?c surface area is caused by high temperature in measurement. In order to dry the particle, a temperature of 1501C was used in measurement. The temperature of 1501C may seriously cause collapse of pore and/or
2. Materials and methods Raw peat was obtained from Jilin province, which was commercially available for agriculture purpose in China. 2.1. The preparation of modi?ed peat–resin particle Raw peat was dried at room temperature, and then broken using a hammer mill and ?nally screened through a 60 mesh sieve. The process of preparing the modi?ed peat–resin particle is as follows: 1. Oxidizing peat: raw peat (200 g, d. w.) was thoroughly mixed with sulfuric acid (100 ml) in a ?ask (2l) at room temperature, which could prepare modi?edpeat.
Q. Sun, L. Yang / Water Research 37 (2003) 1535–1544 1537
0.003 Pore Volume,cc.g-1
< 0.8mm 0.8_1.3 mm 1.3_2.5 mm 2.5_5.0 mm
Fig. 1. Pore distribution of peat–resin particle.
Table 1 Speci?c surface area of peat–resin particle (m2 g?1) Size of particle (mm) Speci?c surface area 2.5–5.0 9.79 1.3–2.5 12.40 0.8–1.3 10.62 o0.8 10.76
shrinkage of pore diameter. So Table 1 and Fig. 1 only present some information about speci?c surface area and pore distribution of modi?ed peat–resin particle.
Basic Brilliant Green ?C: I: Basic Green 4? 2.3. Experimental procedure Adsorption isotherm: modi?ed peat–resin particle, 0.5000 g was thoroughly mixed into 100 ml of 50— 2400 mg l?1, and then the container was sealed up to prevent change of volume of the solution during the experiments. The size of the particle was 2.5—5.0 mm. Suspensions were played for eight weeks at a 2570.51C constant temperature to ensure that adsorption equilibrium was established. The equilibrium solution was ?ltered using stainless steel sieve (300 mesh, the ?rst ?ltrate of 5—10 ml was discarded). Adsorption kinetics: the particle was thoroughly mixed with 2000 ml dye solution in the reaction vessel at constant temperature (251C). 2 ml samples were drawn at suitable time intervals, and ?ltered through stainlesssteel sieve (300 mesh). The concentration of dyes in solution sample was determined using a spectrophotometer with a 1.0 cm glass light cell at a wavelength corresponding to the maximum absorbance.
2.2. The agents The Basic Magenta (C. I. Basic Violet 14) and Basic Brilliant Green (C. I. Basic Green 4) used was the commercial salt, which were widely used in textile industry. Their structure is as illustrated.
Basic Magenta ?C: I: Basic Violet 14?
Q. Sun, L. Yang / Water Research 37 (2003) 1535–1544
2.4. Calculation The amount of dye adsorbed at time t, qt, was calculated from the mass balance equation qt ? ?C0 ? Ct ?V : m ?1?
In order to quantitatively compare the applicability of different kinetic models in ?tting to data, a normalized standard deviation, Dq, was calculated s?????????????????????????????????????????????????? P ??qt exp ? qt cal ?=qt exp ?2 Dq?%? ? 100 ? ; ?2? n?1 where n is the number of data points; qt exp the experimental values; and qt cal the calculated values by model.
where kf is the rate constant of pseudo-?rst-order model. After de?nite integration by applying the initial conditions qt=0 at t=0 and qt=qt at t=t, the equation becomes : kf log?qe ? qt ? ? log qe ? t: ?4? 2:303 Eq. (4) is a linear form. Plotting the log(qe?qt) against t, a line can be obtained. 3.4. Pseudo-second-order model The pseudo-second-order model can be represented in the following form : dqt ?5? ? ks ?qe ? qt ?2 ; dt where ks is the rate constant of pseudo-second-order model. After integrating Eq. (5) for boundary conditions t=0 to t=t and qt=0 to qt=qt, the following form can be obtained: t 1 1 ? ? t: ?6? qt ks q2 qe e Plotting the t/qt against t, a line can be obtained and the qe also can be calculated. 3.5. Intraparticle diffusion model Theoretically, adsorption rate of the dye onto the particle depend on the rate of mass transport processes of the dye within the modi?ed peat–resin particle. For the porous modi?ed peat–resin particle, the mass transport processes may mainly be conformed in the form of diffusion. In a liquid–solid system, the fractional uptake of the solute on particle, F, will vary with the function : D0 t0:5 : r2 Hence there is a linear relationship between F and t0.5 for much of the adsorption processes. F can be de?ned as F ? ?C0 ? Ct ?=C0 : ?7?
3. Theoretical The adsorption of dye onto modi?ed peat–resin particle can be divided into three consecutive stages. First, dye migrates through the solution to the exterior surface of modi?ed peat–resin particle. Second, the dye moves within the particle. Then, third, the dye is adsorbed at sites on the interior surface of the modi?ed peat–resin particle. Many factors can affect the adsorption rate of dye on modi?ed peat–resin particle, such as the initial dye concentration, particle size, particle dose and turbulence state of solution. Generally the third stage is very rapid and does not form a rate-limiting stage in the adsorption. 3.1. Adsorption isotherm In adsorption isotherm study, adsorption isotherm equations usually used include Langmuir equation, Freundlich equation, Redlich–Peterson equation and BET equation. For the liquid–solid system, the Langmuir and Freundlich isotherm equations usually are employed. 3.2. Kinetic models In order to investigate the adsorption processes of two basic dyes on modi?ed peat–resin particle, three kinetic models are used, including pseudo-?rst-order model, pseudo-second-order model and intraparticle diffusion model. 3.3. Pseudo-?rst-order model The pseudo-?rst-order equation is dqt ? kf ?qe ? qt ?; dt ?3?
The diffusion rate of the dye molecule in particle, ki, can be calculated by the following equation: ki ? qt =t0:5 : Its linear form is qt ? ki0:5 ?9? ?8?
where, ki is the intraparticle diffusion rate. According to previous studies, the plot of qt against t0.5 may present a multi-linearity , which indicates that two or more steps occur in the adsorption processes. The ?rst sharper portion is the external
surface adsorption or instantaneous adsorption stage. The second portion is the gradual adsorption stage, where the intraparticle diffusion is rate-controlled. The third portion is the ?nal equilibrium stage, where the intraparticle diffusion starts to slow down due to the extremely low solute concentration in solution [5,19].
Q. Sun, L. Yang / Water Research 37 (2003) 1535–1544 1539
4. Results and discussion 4.1. Adsorption isotherm
The experiments of the adsorption isotherm showed that modi?ed peat–resin particle could effectively remove both Basic Magenta and Brilliant Green from the aqueous solution (especially, at a high dye concentration). The modi?ed peat contained an amount of negatively charged groups and exchangeable ions (such as H+) associated with humic acid. The basic dyes could be adsorbed by electrostatic interactions and ion exchange. Of course, the resin, which contained an amount of hydroxide (?OH), could also adsorb the basic dyes by chemical reaction. 4.2. Kinetics of adsorption
Fig. 2 is the adsorption isotherm of Basic Magenta and Basic Brilliant Green. With the initial dye concentration increase, the amount of adsorption of dye on modi?ed peat–resin particle increased swiftly. No maximum adsorption value was observed as shown in Fig. 2. The data of adsorption isotherm of Basic Magenta and Basic Brilliant Green could not be described by isotherm equations of Langmuir and Freundlich. Plots of Ce/qe vs. Ce and ln qe vs. ln Ce were not line. The ?tness between experimental data and theoretical models was not good (r2 o0:95). This result is similar to that of Allen et al. . Allen et al.  studied adsorption isotherm of basic dyes using sphagnum peat, the results show that the adsorption isotherm of basic dyes deviate from Langmuir and Freundlich adsorption isotherm equations. They thought that the deviation could be attributed to the pressure of adsorbing ions creating new surface by expanding the adsorbent particle. The adsorption isotherms indicated that the adsorption of Basic Magenta and Basic Brilliant Green on modi?ed peat–resin particle might be the ‘‘multimolecular layer in thickness’’.
4.2.1. Effect of experimental conditions on adsorption processes Figs. 3–6 show the effect of the initial dye concentration, particle size, agitation speed and particle mass on adsorption of Basic Magenta. Figs. 3–6 indicate that adsorption rate of the dye was very fast at the beginning stage and, then, slow down gradually. The initial dye concentration, particle size and mass could signi?cantly affect the adsorption rate. In Fig. 5, the adsorption rate of less than 0.8 mm was between that of 2.5–5.0 and 1.3– 2.5 mm, the reason needs to be investigated further. The difference of adsorption rate at different agitation speed was insigni?cant. Similar phenomena were observed in the kinetic experiment of Basic Brilliant Green. 4.3. Test of kinetic models In order to explore the kinetic processes of adsorption, it is a good method that using different models to ?t the experimental data and comparing the normalized standard deviation. Table 2 presents the result of ?tting experimental data with pseudo-?rst and pseudo-secondorder models and intraparticle diffusion model.
450 400 Adsorption capacity,mg.g-1 350 300 250 200 150 100 50 0 0 50 100 150
Basic Magenta Basic Brilliant Green 200 250
Fig. 2. Adsorption Isotherm of Basic Magenta and Basic Brilliant Green.
Q. Sun, L. Yang / Water Research 37 (2003) 1535–1544
450 400 350 100 mg/l 300 Ct, mg.g-1 250 200 150 100 50 200 mg/l 400 mg/l
400 Time, min
Fig. 3. Plot of Ct vs. time at different initial concentrations (251C, 300 rpm).
200 180 160 140 Ct , mg.l-1 120 100 80 60 40 20 100 200 300 400 Time, min
Fig. 4. Plot of Ct vs. time at different particle mass (Ci =200 mg l?1, 251C, 300 rpm).
4.0000g 6.0000g 8.0000g 12.0000g
From Table 2, the order of Dq?%? was pseudo?rst>pseudo-second> intraparticle diffusion model in all of the experimental conditions, which indicated that the intraparticle diffusion model was the best one in describing the adsorption kinetics of basic dyes on modi?ed peat–resin particle. Fig. 7 typically illustrates the comparison between the calculated and measured results for adsorption of Basic Magenta. From Fig. 7, it was found that the pseudo-?rst and pseudo-secondmodels underestimate at the initial stage of adsorption.
The measured results indicated that the adsorption rate was very fast at the beginning stage of adsorption. 4.4. The intraparticle diffusion of basic dye in particle As mentioned above, the intraparticle diffusion model could describe the adsorption kinetics of basic dyes well. Fig. 8 is the plot of qt vs. t0.5 at different initial dye concentration, which indicated that the plot of qt vs. t0.5 was multi-linear. The intraparticle diffusion constants
200 180 160 140 Ct, mg.l-1 120 100 80 60 40 20 0 100 200 300 400 Time, min 500
Q. Sun, L. Yang / Water Research 37 (2003) 1535–1544 1541
2.5--5.0mm 1.3--2.5mm 0.8--1.3mm <0.8mm
Fig. 5. Plot of Ct vs. time at different particle sizes (251C, Ci =200 mg l?1, 300 rpm)
200rpm 150 300rpm 500rpm
Ct , mg.l-1
400 Time, min
Fig. 6. Plot of Ct vs. time at different agitation speed (Ci =200 mg l?1, 251C)
could be calculated using Eq. (8) or Eq. (9). Table 2 presents intraparticle diffusion constants (ki1, ki2 and ki3). The ki1, ki2 and ki3 express diffusion rates of the different stages in the adsorption.
From Table 2, it is seen that the order of adsorption rate was the ?rst stage (ki1)>second stage (ki2)>third stage (ki3). At the beginning, the dye was adsorbed by the exterior surface of the modi?ed peat–resin particle,
Q. Sun, L. Yang / Water Research 37 (2003) 1535–1544
Table 2 Kinetic parameters and normalized standard deviation for adsorption of basic dyes on modi?ed peat–resin particle (251C) Solute Pseudo-?rst order kf ( ? 10 ) Basic Magenta particle mass (g) 4.0000 1.84 6.0000 3.22 8.0000 4.14 12.0000 5.99 Initial concentration (mg l?1) 100 8.75 200 6.45 400 2.99 Particle sizes (mm) 2.5–5.0 6.45 1.3–2.5 2.53 0.8–1.3 2.07 o0.8 3.22 Agitation speed (rpm) 200 5.30 300 5.76 500 6.45 Basic Brilliant Green Initial concentration 100 200 400 Particle mass (g) 5.0000 10.0000 15.0000 20.0000
Pseudo-second order ks( ? 10 )
Intraparticle diffusion Dq(%) ki.1 ki.2 ki.3 Dq (%)
53.62 48.72 44.79 40.46 33.33 39.29 51.65 39.29 53.00 52.81 46.85 44.65 44.16 39.60
0.30 1.00 3.00 12.00 13.60 5.82 2.94 5.82 11.60 10.20 7.79 5.12 5.82 5.18
90.90 62.50 47.62 33.22 19.95 39.29 67.56 39.29 29.33 31.35 35.84 39.53 39.29 39.78
48.38 39.31 22.86 22.20 9.86 23.19 25.97 21.70 28.81 27.30 22.83 23.86 21.70 17.86
6.11 5.47 3.16 2.42 1.74 3.99 5.78 3.99 3.06 1.64 2.06 3.26 3.99 3.29
1.76 1.61 1.56 1.33 0.97 1.29 1.80 1.29 0.86 0.89 1.17 1.11 1.29 1.48
1.54 1.23 0.39 0.31 0.12 0.42 0.99 0.42 0.54 0.32 0.49 0.43 0.42 0.40
5.78 6.05 3.54 4.52 4.84 3.19 5.07 3.19 4.43 3.46 2.91 2.02 3.19 3.28
(mg l?1) 6.68 5.76 5.76 4.61 5.30 6.22 7.14
47.76 45.81 47.68 42.08 43.04 45.51 46.52
25.20 6.50 2.10 3.39 6.30 13.84 24.41
19.92 39.22 68.97 58.35 39.84 26.88 20.24
15.79 27.00 39.32 22.74 16.06 6.15 6.78
1.47 2.90 5.91 4.20 2.90 2.75 2.23
0.80 1.26 1.65 2.39 1.26 1.09 0.73
0.10 0.38 0.82 1.12 0.38 0.17 0.11
5.18 17.21 19.34 9.89 10.23 4.38 5.70
both the modi?ed peat and resin could adsorb the dye, so the adsorption rate was very fast. When the adsorption of the exterior surface reached saturation, the molecular dye entered into the modi?ed peat–resin particle by the pore within the particle and was adsorbed by the interior surface of the particle. When the molecular dye diffused in the pore of the particle, the diffusion resistance increased, which caused the diffusion rate to decrease. With decrease of the dye concentration in the solution, the diffusion rate became lower and lower, the diffusion processes reached the ?nal equilibrium stage. Therefore the changes of ki1, ki2 and ki3 could be attributed to the adsorption stages of the exterior surface, interior surface and equilibrium, respectively. Allen et al.  thought that there were four separate regions depicting the mass transfer onto peat, i.e. external mass transfer effect, macropore diffusion, transitional pore diffusion and micropore diffusion. Table 2 shows that all of ki1, ki2 and ki3 increased with initial dye concentration. The driving force of diffusion
was very important for adsorption processes. Generally the driving force changes with the dye concentration in bulk solution. The increases of dye concentration result in increase of the driving force, which will increase the diffusion rate of the molecular dye in pore. ki1, ki2 and ki3 decreased with increase of particle mass. The increase of particle mass usually results in increase of the exterior surface, which causes swift decrease of the dye concentration in bulk solution and driving force. Generally, the diffusion rate increases with decreasing particle size . From Table 2, it is found that the effect of diffusion rate on particle size was irregular, which could be related to structure and chemical components of modi?ed peat–resin particle. Table 2 also indicates that the effect of diffusion rate on agitation speed was not signi?cant. With increase of agitation speed from 200 to 500 rpm, the change of diffusion rates, ki1, ki2 and ki3, was little, which means that the resistance of particle surface was very low for molecular dye from bulk solution entering into a
40 35 30 25 qt, mg.g-1 20 15 10 5
Q. Sun, L. Yang / Water Research 37 (2003) 1535–1544 1543
experimental data second order model first order model intraparticle diffusion model
400 time, min
Fig. 7. Plot of between the measured and modeled time pro?les for adsorption of Basic Magenta at initial concentration 200 mg l?1 (251C)
70 60 50 qt, mg.g-1 40 30 20 10
C0 = 100mg.1-1 C0 = 200mg.1-1 C0 = 400mg.1-1
ki.2 ki.1 5 10 15 20 min0.5
0 e 0
Fig. 8. Plot of qt vs. t0:5 at different concentrations.
Basic Magenta and Basic Brilliant Green on a modi?ed peat–resin particle was very high. 2. The adsorption processes of Basic Magenta and Basic Brilliant Green on a modi?ed peat–resin particle could be well described by intraparticle diffusion model, and the adsorption rate of two basic dyes on a modi?ed peat–resin particle was mainly controlled by the diffusion rate of the molecular dye within a particle. 3. The initial dye concentration could signi?cantly affect the diffusion rate of molecular dye in a particle. The diffusion rate of basic dyes increased with initial dye concentration, and decreased with increase of particle mass. 4. The change of agitation speed did not cause signi?cant difference of intraparticle diffusion parameter in experimental conditions. Acknowledgements
particle. This result was different from that obtained by Allen et al. .
Support for this work by the Chinese Academy of Sciences (No. KZCX2-311 and KZCX2-413) and National Key Project (G1999011802) is gratefully acknowledged. References
 Poots VJP, McKay G, Healy JJ. The removal of acid dye from ef?uent using natural adsorbents—I peat. Water Res 1976;10:1061–6.  Poots VJP, McKay G, Healy JJ. The removal of acid dye from ef?uent using natural adsorbents—II wood. Water Res 1976;10:1067–70.
5. Conclusions The adsorption isotherm and kinetics of two basic dyes were studied in batch experimental system. The following results were obtained: 1. The adsorption isotherm of Basic Magenta and Basic Brilliant Green deviated from the Langmuir and Freundlich equations. The adsorption capacity of
1544  Cullen GV, Siviour NG. Removing metals from waste solutions with low rank coals and related materials. Water Res 1982;16:1357–66.  Bailey SE, Olin TJ, Bricka RM, Adrian DD. A review of potentially low-cost sorbents for heavy metals. Water Res 1999;33:2469–79.  Wu FC, Tseng RL, Juang RS. Kinetic modeling of liquidphase adsorption of reactive dyes and metal ions on chitosan. Water Res 2001;35:613–8.  Coupal B, Lalancette JM. The treatment of waste waters with peat moss. Water Res 1976;10:1071–6.  McKay G. Peat—an adsorbent/?ltration medium for wastewater treatment. Water Services 1980;84:357–9.  Couillard D. Appropriate wastewater management technologies using peat. J Environ Syst 1991;21:1–20.  Couillard D. The use of peat in wastewater treatment. Water Res 1994;28:1261–74.  Allen SJ, McKay G, Khader KYH. Multi-component sorption isotherms of basic dyes onto peat. Environ Pollut 1988;52:39–53.  Allen SJ, McKay G, Khader KYH. The adsorption of acid dye onto peat from aqueous solution-solid diffusion model. J Colloid Interface Sci 1988;126:517–24.
Q. Sun, L. Yang / Water Research 37 (2003) 1535–1544  Allen SJ, McKay G, Khader KYH. Intraparticle diffusion of a basic dye during adsorption onto sphagnum peat. Environ Pollut 1989;56:39–50.  Allen SJ, Murray M, Brown P, Flynn O. Peat as an adsorbent for dyestuffs and metals in wastewater. Resour Conserv Recycl 1994;11:25–39.  Ramakrishina KR, Viraraghavan T. Dye removal using low cost adsorbents. Water Sci Technol 1997;36:189–96.  Ho YS, McKay G. Sorption of dye from aqueous solution by peat. Chem Eng J 1998;70:115–24.  Smith EF, MacCarthy P, Yu TC, Mark Jr, HB. Sulfuric acid treatment of peat for cation exchange. J Water Pollut Contr Fed 1977;49:633–8.  Ho YS, McKay G. The sorption of Lead (II) ions on peat. Water Res 1999;33(2):578–84.  Ho YS, McKay G. The kinetics of sorption of divalent metal ions onto sphagnum moss peat. Water Res 2000;34:735–42.  Zhan XM, Wang JL, Wen XH, et al. Adsorption of aniline on NKA—II macroporous adsorption resin. Environ Chem 1999;18:359–65.