Langmuir And Freundlich Adsorption Isotherm Models

Published Date: 02 Nov 2017

Equilibrium modelling was applied to adsorption results obtained when C0 was varied from 25−100 mg/L at fixed dosage of 20 g/L and optimum pH of 2.

Langmuir and Freundlich Adsorption Isotherm models

Figures 4.15 shows the Langmuir plots for BSD, SHBSD and SABSD and while Figure 4.16 shows the corresponding Freundlich plots.

Figure 4.15: Langmuir modelling plot for BSD, SABSD and SHBSD for initial dye concentration ranging from 25−100 mg/L

The linear plots of the Langmuir model for all 3 adsorbents resulted in a positive relationship between Ce/qe and Ce with a high correlation coefficients (R2>0.95). The values of constants Qmax and B were calculated based on the slope and the intercept of the resulting plots and are reported in Table 4.5. Benyoucef and Amrani (2011) defined the Qmax value as a practical limiting adsorption capacity corresponding to the surface of adsorbent fully covered by dye ions. It can be observed that the Qmax values of pre−boiled sawdust of 3.63 mg/g were much lower than those of the chemically pre−boiled sawdusts (SHBSD− 4.48 mg/g and SABSD− 5.27 mg/g). These values clearly demonstrated that chemical treatment had enhanced the adsorption properties of BSD.Semerjian (2010) stated that the constant value demonstrated the adsorbent energy. The values for pre−boiled sawdust were higher than those of the chemically pre−boiled sawdusts. According to Ucun et al. (2009), a high value of implies strong bonding. Hence, it can be inferred that stronger bonds were formed between chemically treated preboiled pine sawdusts and the RB 221 dye molecules. Ucun et al. (2009) also added that a large value of is associated with a large coverage of dye onto the adsorbent Hence, given the higher B values for SHBSD and SABSD, there were more coverage of RB 221 dye molecules onto the chemically treated adsorbents as these had higher surface areas exposed and hence more pore structures and binding sites.

A further analysis of the Langmuir model can be made on the basis of a dimensionless equilibrium parameter, known as the separation constant RL ( Mckay, 1982). The calculated RL values for all 3 adsorbents were between 0 and 1 (Table 4.5). According to the interpretations in Ahamad et al. (2009) and Malkoc et al.(2006), the values of RL obtained in this study would clearly imply favourable adsorption. It can also be observed that the higher RL values were noted for the untreated pre−boiled sawdust and the parameter reduced close to zero for the chemically treated pre−boiled sawdusts. This confirmed that chemical treatment enhanced the adsorption properties of BSD. Similar observations and conclusions were made by Batzias and Sidiras (2007b) for the adsorption of basic dye onto untreated and acid treated beech sawdust.

Figure 4.16: Freundlich modelling plot for BSD, SABSD and SHBSD for initial dye concentration ranging from 25−100 mg/L

The experimental data for BSD, SHBSD and SABSD also conformed to the Freundlich model , with high correlation (R2>0.9). Freundlich contants (KF and n) were evaluated from the linear plots and are reported in Table 4.5. The Kf value is indicative of the adsorption capacity of the adsorbent (Vimonses et al., 2009; Mane and Babu, 2011) . The KF values estimated for chemically treated pre−boiled SD (SHBSD −0.525 mg/L and SABSD – 0.605 mg/L) were significantly higher than those of pre−boiled sawdust only ( BSD−0.247 mg/L). The greater the Kf value, the higher the likelihood of the RB 221 dye molecules to bind onto the adsorbent. Batzias and Sidiras (2007b) also noted a higher Kf values for acid treated beech sawdust as compared to untreated ones.

n is a measure of the deviation from linearity of adsorption and is used to verify the types of adsorption (Salleh et al., 2011). For all 3 adsorbents in this study, the values of n were above 1. According to Vimonses et al. (2009), Mohammad et al. (2010), Sen et al. (2011) and Salleh et al.( 2011), the present results for n hence may be associated to favourable adsorption and physical process. However, the values of n did not change significantly after chemical treatment and this observation is shared by Sidiras et al. (2011).

Table : Comparison between Freundlich and Langmuir Model for BSD, SHBSD and SABSD

BSD

SHBSD

SABSD

Langmuir Parameters

, L/mg

0.0348

0.0701

0.085

RL

0.223−0.535

0.125−0.363

0.105−0.320

Qm , mg/g

3.63

4.48

5.27

R2

0.998

0.958

0.997

ARE, %

1.088

5.595

2.217

ARS, %

1.413

7.308

3.202

Freundlich Parameters

Kf, mg/L

0.247

0.517

0.605

n

1.721

1.904

1.756

1/n

0.581

0.525

0.569

R2

0.985

0.972

0.968

ARE, %

4.875

6.198

7.891

ARS, %

5.715

8.151

9.412

The applicability of Langmuir and Freundlich isotherms showed that there are both homogenous and heterogenous binding sites on the surface of BSD, SHBSD and SABSD. These homogenous and heterogenous binding sites would be involved in the monolayer and multilayer adsorption of RB 221 dye molecules, respectively. Tumols et al. (2011) stated that both chemisorption and physorption would be occurring under these conditions. Ucan et al. (2008) and Asadi et al. (2008) also observed that both models were applicable for the adsorption of heavy metals on cone biomass of pine sylvestris and on raw and treated sawdust, respectively. However, due to the lower percentage error observed with the Langmuir model in this study, it has been inferred that more homogenous binding sites rather than the heterogeneous ones were present onto the BSD, SHBSD and SABSD. This in turn must have favoured monolayer rather than multilayer coverage of RB 221 molecules onto these adsorbent. Additionally, the value 1/n was less than unity for all 3 adsorbents, indicating the better suitability of the Langmuir isotherm model (Fytianos et al., 2000). This is because the more heterogeneous the surface of the adsorbent, the closer the value of 1/n is to zero (Tien, 1994: Leechart et al., 2009). Since the values of 1/n were well above 0.5 for SHBSD and SABSD (Table 4.5), it could be deduced that the surfaces of BSD and especially those of SHBSD and SABSD were more of a homogeneous character than being heterogenous.

Sorption energy estimation by Dubinin−Radushkevich and Temkin Isotherm model

The linear plots of the D−R and Temkin models for BSD, SHBSD and SABSD are given by Figure 4.17 and Figure 4.18, respectively.

Figure 4.17: D−R modelling plot for BSD, SABSD and SHBSD for initial dye concentration ranging from 25−100 mg/L

Figure 4.18 : Temkin modelling plot for BSD, SABSD and SHBSD for initial dye concentration ranging from 25−100 mg/L

The application of the experimental data to D−R and Temkin models helped to estimate the adsorption energy. This parametergives an indication of the type of bonding occurring between RB 221 molecules and the adsorbents tested.

For all 3 adsorbents, the D−R plots had high regression coefficients of R2 > 0.9. The saturation capacity, qs and the mean free energy of adsorption, E obtained from this plot aregiven in Table 4.6. A qs value of 2.120 mg/g ,2.797 mg/g and 3.213 mg/g was obtained for BSD , SHBSD and SABSD, respectively. The saturation capacity of chemically treated BSD was higher than untreated BSD. Higher adsorption energy was also achieved with chemically treated BSD. The E values for BSD, SHBSD and SABSD were 0.224,0.354 and 0.5 kJ/mol, repectively. Itwas noted that E values for all 3 adsorbents were lower than 8 KJ/mol. Being less than the threshold of 8 KJ/mol indicated the predominance of physical binding in the sorption of RB 221 molecules onto BSD, SHBSD and SABSD (Zhu et al., 2009, Chakraborty et al., 2011).

The Temkin plots also had high regression coefficient of R2 > 0.9. The Temkin isotherm constant AT and bT deduced from the plots for each adsorbent are reported in Table 4.6. The heat sorption constant B was also calculated. B begin much lower than 20 kJ/mol once more indicated the predominance of ideal physorption (Itodo and Itodo, 2010).

BSD

SHBSD

SABSD

D−R Parameters

BDR

0.00001

0.000004

0.000002

qs (mg/g)

2.120

2.797

3.213

E( KJ/mol)

0.224

0.354

0.500

R2

0.929

0.911

0.937

ARE, %

13.943

12.808

15.273

ARS, %

19.218

15.688

20.205

Temkin

Parameters

B , J/mol

0.839

1.011

0.827

bT

2450.130

2049.274

AT (L/g)

0.308

0.657

0.768

R2

0.997

0.965

0.999

ARE, %

1.825

4.714

0.568

ARS, %

2.398

6.576

0.692

Table : Comparison between D-R and Temkin model for BSD, SHBSD and SABSD