Evaluation of Kinetic Adsorption Models
Evaluation of Kinetic Adsorption Models
Evaluation of Kinetic Adsorption Models
Engineering
ISSN (Online): 2577-5944
Volume 2, Issue 1, pp: 1-13
Research Article Open Access
Evaluation of Kinetic Adsorption Models in Aquatic
Environment, Models of Lagergren and Ho et al.
Zahra Askari1, Shayan Shamohammadi2, Ali Hasantabar-Amiri3, Kaveh Ostad-Ali-Askari4*
Saeid Eslamian5, Vijay P. Singh6
1
Department of Water Engineering, Shahre-kord University, Shahre-kord, Iran
2
Department of Irrigation Engineering, Shahrekord University, Shahrekord, Iran
3
Department of Civil Engineering, Lenjn Branch, Islamic Azad University, Lenjan, Isfahan, Iran
4*
Department of Civil Engineering, Isfahan (Khorasgan) Branch, Islamic Azad University, Isfahan, Iran
5
Department of Water Engineering, Isfahan University of Technology, Isfahan, Iran
6
Department of Biological and Agricultural Engineering & Zachry Department of Civil Engineering
Texas A and M University, U.S.A
*
Koa.askari@khuisf.ac.ir
Abstract: Heavy metal pollution has become one of the most cucial environmental obstacle today. The
treatment of heavy metals is of special concern due to their refractoriness and resistence in the environment.
In recent years, diferent procedures for heavy metal removal from wastewater have been widely studied.This
paper revises the common procedures that have been applied to treat heavy metal wastewater and considers
these techniques.Presently kinetic adsorption equations are widely used in adsorption processes. The main
objective of the present research is to evaluate the linear and non-linear equations of the Lagergren and Ho
et al. models in the kinetics of nickel adsorption by activated carbon. For this purpose, powdered activated
carbon, the raw material of which is wood, was purchased from the “Merck” company (Germany). The optimum
PH of the adsorption was obtained to be 6. As well, the kinetics of adsorption showed that for the initial nickel
concentration of 2.5, 5, 10, 50 and 125 mg/L, equilibrium time was 40, 75, 120, 150, 50 minutes respectively. The
maximum removal efficiency of nickel in the initial concentration of 2.5 mg/L was obtained to be 89.6%. The
fitting of the four linear kinetic models of Ho et al. (1995), and its non-linear model, as well as the linear model
ofkinetic data of adsorption at diferent concentrations showed that the non-linear model of Ho et al. (1995)
better describes the kinetic data of nickel adsorption by activated carbon thanhis linear models. Comparison
of the non-linear models of Ho et al. and Lagergren (1893) showed that in low concentrations the non-linear
model of Ho et al., and in high concentrations non-linear model of Lagergren better describes the kinetic data
of adsorption.First and second-order rate equations were pragmatic to etude adsorption kinetics.The most
significant advantage of the ofered model is its capability to predicate the balance time of adsorption, which is
very essential to optimize the expense of the adsorption system design.The received consequences depict that
the new model can analyze the experiential information very well.The equation is received clearly by changing
the mass balance equation of a single step batch-type system into integrated first order rate equation. The
proposed new formulation depends on maximum sorption (or removal) rather than equilibrium sorption, as
given by Lagergren equation.The precise value of equilibrium sorption is not easy to describe in many cases, and
it is exhibit that the modified formulation prepares better correlation with sorption data when the procedure
is far from completion.A corrected formulation of the popular Lagergren pseudo first order rate equation is
proposed.The equation is created clearly by changing the mass balance equation of a single step batch-type
system into unified first order rate equation. The proposed new formulation depends on maximum sorption
(or removal) rather than equilibrium sorption, as given by Lagergren equation. The precise value of equilibrium
sorption is not simple to describe in many cases, and it is exhibited that the modified formulation provides
better correlation with sorption data when the process is far from completion.
Keywords: Adsorption, Nickel, Linear and Non-linear Models, Activated Carbon, Kinetic Adsorption
www.arjonline.org Page 1
Evaluation of Kinetic Adsorption Models in Aquatic Environment, Models of Lagergren and Ho et al.
Introduction
Studies on the adsorption of heavy metals from industrial wastewaters have seen a remarkable growth in recent
years due to the increasing use of such pollutants. Based on the objectives of adsorption studies various models
for the analysis of adsorption processes are selected and used. One of the important goals of such studies is
to determine the time required for optimal adsorption in the solid-liquid system (Jonathan et al., 2009:616).
The mechanism of adsorption process and its time is called adsorption kinetics (Royer et al., 2009:1213;
Lima et al., 2008:536). Adsorption kinetics is afected by various factors, such as concentration of the sorbed
material, amount, size and special surface area of the adsorbent, pH of solution and ambient temperature
(Shamohammadi, Heydari, 1386:126). For the best-fitting of the adsorption kinetic data models of Ho et al.
(1995) and Lagergren (1893) are more efficient (Jonathan et al., 2009:616), where two types of equations,
linear and non-linear, are used. Lagergren’s model does not necessarily show the actual nature of adsorption,
so the capability by this model to analyze the adsorption data is doubtful (Schiewe et al., 2008:1896). Ho et
al. (2006) examined and evaluated the linear and nonlinear behavior of the Ho et al. (1995) model. Their
results showed that the nonlinear model is better fitted with the adsorption data and provides a better ability
to determine the parameters of the model. In the Table 2 the corrections of the Ho et al. model of adsorption
kinetics are showed (Ho, 2006: 119).
At present, more than 40% of the nickel found in the environment is produced by industrial plants and battery
factories (Ajmal et al., 2001:117), that have a great role in contaminating the surface waters(Dursun et al.,
2006:187). Nickel concentration in industrial wastewater ranges from 3.4 to 900 mg/L (Patterson, 1985:217).
Ritchie (1977)
Ho (1995)
Characteristics Amount
Amount of the dissolved ethanol 0.2%
Chloride 0.01%
Sulfate 0.01%
Lead 0.005%
Iron 150 mg/L
Amount of the dissolved hydrochloric acid 1%
Effect of the Optimal Adsorption PH
In eight 100 ml capacity Erlenmeyer flasks 1 g/L of weighted adsorbent was put inside of each of the Erlenmeyer
flasks. Then, a metal solution with a concentration of 10 mg/L (Jamali Paqal’e, 1388:76) was added to each of
the Erlenmeyer flasks. Diferent pH values (ranging from 2 to 8) were examined. The solutions were placed on a
shaker rotating at the speed of 150 rpm (Aydin et al., 2008:37) in a controlled temperature for 12 hours (Babel
et al., 2003:14).
American Research Journal of Civil And Structural Engineering Page 3
Evaluation of Kinetic Adsorption Models in Aquatic Environment, Models of Lagergren and Ho et al.
Effect of Contact Time and Initial Nickel Concentration:
Twelve Erlenmeyer flasks were picked and in each of them nickel solution with a concentration of 2.5 mg/L
was poured. Also, the amount of 0.05 grams of activated carbon adsorbent was added to each of the Erlenmeyer
flasks and they with an optimal pH were put on the rotating shaker at the speed of 150 rpm for 5 to 240 minutes.
This procedure was repeated with the concentrations of 5 mg/L, 10 mg/L, 50 mg/L and 125 mg/L.
Kinetic Adsorption Models
Lagergren’s model(1898): this model is expressed as a equation based on the adsorbent capacity .
By integrating the previous equation and boundary conditions (q=0, t=0 and q=q, t=t), it was changed to
equation , that may be rearranged to the non-linear form . Also its linear form is
written as equation In order to fit the linear models with the experimental
data, qt and is written in front of t. In the linear form line slope and its y-intercept will be
Logqe. Therefore, it is possible to obtain k1 and qe from the slope and y-intercept of linear equation (Ho et al.,
2002:797).
Ho et al. model (1995): this model is widely used in analyzing and examining kinetic adsorption data (Ho et al.,
2001:241; Ho et al., 2000:189). It is possible to write this model in diferent linear forms (Makris et al., 2005:
4250; Duons et al., 2005:438; Wang et al., 2005:336; Janos et al., 2005:19). It has four linear forms (Yuh et al.,
2006:119) (Table 4).
Diferential form of the Ho et al. model (1996) is written as follows: . As well, for the
boundary conditions (t=0 to t and q=0 to q) non-linear equation form is written like this:
Table4. Linear Forms of Ho et al. Model
= +
t 1 1
1 t
qt k q e2 qe
=( 2) +
1 1 1 1
2
qt k qe t qe
qt = q e − (
1 qt
3 )
k qe t
= k q e2 − k q e qt
qt
4
t
The linear equations of Ho and et al. model provide diferent amounts in determining the adsorption parameters
(Ho, 2004:2115; Longhinoti et al., 1998:435; Kinniburgh, 1986:895). First type linear equation (see Table 4)
is more known than the other types (Ho et al., 2006:119; Yuh at al., 2006:119). By fitting the models with the
experimental data, in both linear and nonlinear forms, the coefficients of the Ho et al. model are obtained. Then
qt and t/qt are written against t respectively, and in this way constants k2 and qe are calculated. As well, it is
possible to calculate the qt by using the following equation .
Results and Discussion
Determining optimal PH
Environmental acidity, in which the adsorption process takes place, has a great influence on the absorption
rate (Larous at al., 2005:483; Ahmad et al., 2008:127). Part of this efect is due to the adsorption of hydrogen
Fig1. Effect of PH on Nickel Adsorption Efficiency by Powdered Activated Carbon Adsorbent(Initial concentration
of nickel - 10 mg/L, contact time – 5 hours, amount of adsorbent – 1 g/L)
Kinetic Adsorption Experiments
Effect of Contact Time and Initial Concentration
In Figure 2 the changes in the amount of nickel adsorption by the activated carbon adsorbent together with the
time are presented. As can be seen, initially, the adsorption intensity rate is high in all concentrations, but over
time, adsorption intensity decreases until finally it reaches a constant value. The period of time necessary for
the adsorption amount to reach its maximum is called equilibrium time. At this point the maximum amount of
adsorption occurs.
Maximum amount of adsorption in 2.5 mg/L, 5 mg/L, 10 mg/L, 50 mg/L and 125 mg/L concentrations is
respectively 2.24 mg/g, 4.3 mg/g, 7.96 mg/g, 14.26 mg/g and 21.22 mg/g.
Fig2. Effect of Contact Time of Activated Carbon Adsorbent on Removing Nickel (Amount of adsorbent – 10 g/L
and pH=6)
American Research Journal of Civil And Structural Engineering Page 5
Evaluation of Kinetic Adsorption Models in Aquatic Environment, Models of Lagergren and Ho et al.
Therefore, it may be concluded that for a certain amount of adsorbent by increasing nickel ion concentration
from 2.5 to 125 mg/L, the adsorption amount decreased from 89.6% to 21.22%. These results are consistent
with the results of studies performed by Jamali et al. (2008: 65).
Also, the results of the experiments of the efect of initial solution concentration on the equilibrium time of the
nickel adsorption are shown in Figure 3.
Fig3. Effect of Initial Nickel Concentration on Equilibrium Time by Powdered Activated Carbon
As can be seen, equilibrium time of nickel adsorption by activated carbon adsorbent in 2.5 mg/L, 5 mg/L, 10
mg/L, 50 mg/L, 125 mg/L respectively equals to 50, 150, 120, 75, 40 minutes.
According to the above Figure, it can be said that by increasing the concentration of nickel solution from 5 to
125 mg/L, and also by decreasing the concentration of solution from 5 to 2.5 mg/L, the equilibrium time of the
adsorption process decreases. By reducing the initial concentration of the solution, it is expected that the time
to achieve equilibrium will be increased due to a decreased chance of metal contact with adsorbent. However,
given that in low concentrations the amount of metal ions found in the solution is low, this limited amount of ion
is absorbed by the pores of the adsorbent. Therefore, the time to reach the equilibrium decreases by decreasing
initial concentration of nickel. In this case (when concentrations are low), the factor determining equilibrium
time is the concentration of solution. In high concentrations, due to the high accumulation of metal ions around
the adsorbent and increased chance of metal contact with the adsorbent, it takes a little more time for the
adsorbent capacity to reach its saturation and the process to reach equilibrium.
Therefore, in this case also, the factor determining the equilibrium time is the adsorbent’s capacity. These
results are consistent with the results of the studies performed by Jamali (1387:71), Jamali Paqal’e (1388:88)
and (Azizian, 2004:47).
Kinetic Adsorption Equations
The transformation of the non-linear model of Ho et al. to linear equations may lead to an increase in the error
variance and standard least squares (Kinniburgh, 1986:895; Ho, 2004:2115).
Figures 4 and 5 present respectively the fitting of non-linear models of Lagergren and Ho et al., and Figures 6, 7,
8, 9, 10 present respectively the fitting of the linear models of Lagergren and Ho et al. As well kinetic parameters
of the linear and non-linear equations adsorption of the Lagergren and Ho et al. models respectively are shown
in Tables 5 and 6. The results show that the numerical values of the kinetic parameters of adsorption qe, k1 (in
each of the four linear models of Ho et al., Type 1, 2, 3 and 4) difer significantly, which can be explained by
their non-nonlinearregression coefficients. These results correspond to the studies performed by Yuh et al.
(2006:119). Therefore, having a reliable non-linear model (Ho et al., 1995:1327), the usage of diferent linear
equations of this model does not seem necessary.
American Research Journal of Civil And Structural Engineering Page 6
Evaluation of Kinetic Adsorption Models in Aquatic Environment, Models of Lagergren and Ho et al.
As well, the results of the fitting of the non-linear models of Lagergren and Ho et al. (Fig. 6) show that,
although both models describe the data by the accuracy of 95%, the model of Ho et al. is more efficient for all
concentrations used in the test. The reason of this might be due to the hypothesis of Ho et al. regarding the
efect of the adsorbent capacity on the initial rate of adsorption (Jonathan et al., 2009:616). These results do
not correspond to the results of Shamohmmadi and Heydari (1386:126), but is in correspondence with the
results of the studies performed by Jamali at al. (1386:70), Jamali Paqal’e (1388:90), Jonathan et al. (2009:616),
Mohama et al. (2009:12) and Ho (2004:2115).
Comparison of the non-linear models of Ho et al. with the Lagergren’s model in diferent concentrations,presented
in Table 6, shows that in low concentrations the Lagergren’s model and in high concentrationsthe Ho et al.
model has better regression coefficient, and therefore describes the data better. These results correspond to the
results of the study performed by Azizian (2004:47).
Fig4. Fitting of the Lagergren’s non-linear Fig5. Fitting of the Ho et al. non-linear model with
with kinetic adsorption datamodel kinetic adsorption data
Fig6. Fitting of the Lagergren’s linear model with Fig7. Fitting of the Ho et al. Type 1 linear model
kinetic adsorption data with kineticadsorption data
Fig8. Fitting of the Ho et al. Type 2 linear model with Fig9. Fitting of the Ho et al. Type 3 linear model
kineticadsorption data with kineticadsorption data
American Research Journal of Civil And Structural Engineering Page 7
Evaluation of Kinetic Adsorption Models in Aquatic Environment, Models of Lagergren and Ho et al.
Fig10. Fitting of the Ho et al. Type 3 linear model with t kinetic adsorption data
The coefficient of the Ho et al. model (k2) shows surface adsorption rate of the adsorbent before reaching
equilibrium. The higher the coefficient, the greater is the surface adsorption rate of the metal ion by the adsorbent
(Shahmohammadi Heydari, 1386:126). As is shown in Table 6, by increasing the solution concentration from
2.5 to 125 mg/L, the coefficient k2 of the Ho et al. non-linear model is reduced. These results correspond to the
studies of Aksu (2008:112), Mane et al. (2007:390), Han et al. (2009:496) and Azizian et al. (2006:112). Rate of
the coefficient k2 by increasing concentration in the Langergren’s model at first increases and later decreases
(there is no linear relation). As it is seen, this is not a linear relation and therefore in this case the Lagergren’s
model has no good capacity in describing the data (Azizian, 2004:47). Thus, this result confirms the results of
the comparison of the regression coefficients of the the Lagergren models and Hu et al. models.
Table5. Constants of the linear kinetic models for absorbing nickel by powdered activated carbon adsorbent
Concentra-
qe exp
tion mg/l
R2 qe k2 R2 qe k2 R2 qe k2 R2 qe k2 R2 qe k1
0.998 2.503 0.0831 0.1093 1.726 0.997 0.085 2.493 0.999 0.0835 2.501 0.999 0.0833 2.502 0.998 2.5 2.24
0.998 4.4 0.0456 0.0306 1.643 0.969 0.0428 4.430 0.999 0.0428 4.391 0.999 0.0457 4.399 0.998 5 4.3
0.969 8.035 0.0326 0.0414 3.667 0.989 0.025 8.244 0.999 0.0342 7.98 0.986 0.0338 7.999 0.969 10 7.96
0.79 13.861 0.0265 0.0446 6.652 0.969 0.016 14.513 0.994 0.0364 13.368 0.848 0.0345 13.469 0.790 50 14.26
0.981 29.632 0.0072 0.0882 15.649 0.883 0.008 29.06 0.999 0.007 29.85 0.992 0.0074 29.477 0.981 125 26.53
Table6. Constants of the non-linear kinetic models for absorbing nickel by powdered activated carbon adsorbent
Ho et al. Lagergren
Concentration
q e exp
R2 qe k2 R2 qe k1
mg/l