For the removal of iron and manganese from aqueous media, the effectiveness of agro-waste (BSAC and RPPAC) was explored through series of adsorption experiments. In the adsorption experiments, a constant mass and variable contact time (from 30 to 240 minutes maximum) were maintained, otherwise, further details are provided in the sub-section.
3.4 Effect Of Contact Time on Removal Efficiency and Adsorption Capacity.
Figure 4 depicts the removal efficiency of BSAC and RPPAC as a function of time. The RPPAC removal efficiency for both adsorbate (Fe2+ and Mn2+) is shown in Fig. 4a. For both the sealed and the unsealed samples, it was discovered that the RPPAC adsorbed nearly all the Mn2+ ions in the solution. This could be because of the surface's abundance of active adsorption sites of the RPPAC which adsorbates rapidly diffuse from the solution to the adsorbent. The unsealed sample showed the highest removal efficiency within the first 30 minutes of contact time and maintained that level until equilibrium was attained at 240-minute. For the plot of the sealed sample, it was observed that there was a drop in adsorption at 120 minutes and there was re-adsorption at 150 minutes. The decline may have occurred due to saturation of sites, indicating that all binding sites were taken up and the adsorbate had no association. Desorption occurred as the agitation proceeded, freeing some of the binding sites and allowing adsorption to continue until it reached its equilibrium point (Gabelman, 2017). Both the sealed and unsealed RPPAC samples had a 99.95% removal efficiency for manganese ion adsorption.
The optimal removal efficiency values for Fe2+ with the sealed and unsealed RPPAC were 99.95% at 60 minutes for the unsealed sample, and 82% at 180 minutes for the sealed sample. Figure 4b shows the BSAC removal efficiency for both the Fe2+ and Mn2+ adsorbate. For the adsorption of the Fe2+ adsorbate, for both sealed and unsealed samples, it was observed that the sealed samples adsorption started from 30 minutes and moved gradually up to 120 minutes having a removal efficiency of 72.55%. It then dropped from 72.55–60% at 150 minutes. Similar removal efficiency plot was observed in the Fe2+ adsorption, asymptotic values are often associated with saturation of the BSAC adsorbent binding sites. All binding sites on the surface of the adsorbent were saturated with the adsorbate and there were no more binding sites on the surface of the adsorbent to aid in adsorption. The agitation of solution results in freeing some sites and commencing of adsorption until it reached its equilibrium point or its saturation point.
The unsealed samples showed superior adsorption of Fe2+ compared to the sealed samples, with a maximum adsorption time of 90 minutes and a 99.1% removal efficiency. Between the 120- and 150-minute contact time, there was minimal desorption, after which the adsorption resumed and was stable. For Mn2+, adsorption worked well for both the sealed and unsealed samples. The unsealed sample's removal efficiency (99.95%) remained consistent throughout the 240 minutes. The sealed sample's removal efficiency ranged from 94.5% to 99,95% and remained constant for 240 minutes.
Figure 4c is a comparison of BSAC and RPPAC for both sealed and unsealed samples for the adsorption of Fe2+ in solution. Comparing the sealed BSAC samples with the sealed RPPAC samples, it was observed that the sealed RPPAC had a better adsorption than the sealed BSAC sample. For sealed RPPAC, the greatest adsorption was 82%, and for sealed BSAC, it was 72.55%. Unsealed adsorption results for BSAC and RPPAC show little variation in their removal efficiency although both had a nearly perfect removal efficiency value. The unsealed BSAC's best removal efficiency was 99.1% at a contact time of 90 minutes, and the unsealed RPPAC's maximum removal efficiency was 99.95% at a contact time of 60 minutes.
For the adsorption of Mn2+ in solution, Fig. 4d compares BSAC and RPPAC for sealed and unsealed samples. Comparing the sealed BSAC and sealed RPPAC samples, it was shown that the sealed BSAC sample's adsorption rate increased from 94.5–99.95% at 90 minutes. The sealed RPPAC sample adsorption rate at the first 30 minutes of constant was 98.5% and the at 60 minutes had removal efficiency value of 99.25%. There was some desorption before it gradually began to re-absorb until equilibrium was reached at a removal efficiency of 99.95%. The sealed RPPAC had its highest removal efficiency at 90 minutes (99.85) %. From 90 minutes, a little desorption occurred, and it started adsorbing again from 150 minutes and remained constant till 240 minutes with the highest removal efficiency of 99.95%. It was found that both the unsealed BSAC and RPPAC samples were able to adsorb the Mn2+ from the solution at all contact times (Adsorption was 99.95% from the initial contact time 30 minutes to 240 minutes).
3.5 Adsorption Isotherm Model Studies.
The two activated carbons' (BSAC and RPPAC) adsorption mechanisms were clarified using the Langmuir and Freundlich isotherm models. The model that best describes the adsorption equilibrium process was determined using the correlation coefficient (R2) value and for those with the same R2 values, other characteristics were used to determine the best fit model and also for the understanding of the mechanism of adsorption. Tables 3 to 6 give the isotherm modeling fit plots for the two activated carbons.
Table 3 below shows the adsorption parameters for sealed and unsealed BSAC for the adsorption of Fe2+ in aqueous media. For the Sealed BSAC, it was observed that the R-square (R2) for the Langmuir isotherm model was 0.979 compared with that of Freundlich isotherm which was 0.836. For the unsealed BSAC, the R2 value for Langmuir model was 0.997 and that of the Freundlich was 0.837 respectively. This makes the Langmuir isotherm the best fit for the adsorption data obtained (Goher et al., 2015).
Again, by comparing the qmax and Kf values for the Langmuir and Freundlich models for the unsealed BSAC, the Langmuir isotherm model has the advantage of adsorbing more of the adsorbate than the Freundlich isotherm model. This is because the unsealed BSAC of the Langmuir has a higher qmax value (0.0845) compared to the Kf value (0.077). The separation parameter, RL, for the Langmuir isotherm model, was found to be 0.050, falling within the favorable range of (0 < RL < 1). There is no adsorbate transmigration on the surface plane because the adsorption process is monolayer adsorption on homogenous sites with all sites having identical attraction towards the adsorbate (Kundu and Gupta, 2006).
The sealed BSAC for Fe2+ removal had, R2 values for the Langmuir being greater than Freundlich, hence Langmuir model is the best fit for the adsorption of Fe2+. The R2 values were 0.979 for Langmuir and 0.836 for the Freundlich isotherm model for Fe2+ removal.
Table 3. Isotherm Models Parameters for Sealed and Unsealed BSAC for Fe2+
Adsorbent
|
Metals
|
Langmuir Isotherm
|
Freundlich Isotherm
|
|
|
KL (L/g)
|
qmax (mg/g)
|
R2
|
RL
|
SE (slope)
|
n (L/g)
|
Kf (mg/g)
|
R2
|
SE (slope)
|
Sealed
BSAC
|
Fe2+
|
0.099
|
0.033
|
0.979
|
1.247
|
1.801
|
1.223
|
0.287
|
0.836
|
0.148
|
Unsealed
BSAC
|
Fe2+
|
10.555
|
0.0845
|
0.997
|
0.050
|
0.270
|
22.97
|
0.077
|
0.837
|
0.0092
|
Table 4. Isotherm Models Parameters for Sealed and Unsealed RPPAC for Fe2+
Adsorbent
|
Metals
|
Langmuir Isotherm
|
Freundlich Isotherm
|
|
|
KL (L/g)
|
qmax (mg/g)
|
R2
|
RL
|
SE (slope)
|
n (L/g)
|
Kf (mg/g)
|
R2
|
SE (slope)
|
Sealed
RPPAC
|
Fe2+
|
0.135
|
0.018
|
0.997
|
1.369
|
1.164
|
0.286
|
0.541
|
0.988
|
0.0152
|
Unsealed
RPPAC
|
Fe2+
|
1.369
|
0.023
|
0.993
|
0.576
|
1.449
|
27.03
|
0.449
|
0.459
|
0.0164
|
Table 4 shows isotherm model parameters for both sealed and unsealed RPPAC for the adsorption of Fe2+. The sealed RPPAC was found to have a correlation coefficient R2 value of 0.997 for the Langmuir isotherm and that of Freundlich isotherm was 0.988. This indicates that the adsorption process can be best described by the Langmuir isotherm model. The findings point to uniform mono-layer ion adsorption on homogeneous adsorbing sites on the RPPAC's surface.
Table 5. Isotherm Models Parameters for Sealed and Unsealed BSAC for Mn2+
Adsorbent
|
Metals
|
Langmuir Isotherm
|
Freundlich Isotherm
|
|
|
KL
(L/g)
|
qmax (mg/g)
|
R2
|
RL
|
SE (slope)
|
n (L/g)
|
Kf (mg/g)
|
R2
|
SE (slope)
|
Sealed
BSAC
|
Mn2+
|
115.1
|
0.094
|
0.999
|
4.4x10-3
|
0.063
|
101.6
|
0.024
|
0.587
|
0.0034
|
Unsealed
BSAC
|
Mn2+
|
71.5
|
0.042
|
0.000
|
7.0x10-3
|
2.7x10-15
|
5.999
|
0.69
|
1.00
|
1.4x10-17
|
Table 5 gives the obtained parameters for isotherm models for sealed and unsealed BSAC for Mn2+ adsorption. The R2 value for the Langmuir isotherm of the sealed BSAC was 0.999 and that of the Freundlich isotherm was 0.587. Looking at their qmax and Kf values (0.094 and 0.024), respectively, it shows that the Langmuir isotherm model has an advantage of adsorbing more of the Mn2+. We can confidently say that for unsealed BSAC on the adsorption of Mn2+, the adsorption data is best described by Freundlich isotherm model. The qmax and Kf Freundlich parameters were 0.69 and 0.042 respectively and indicates greater adsorption. The result indicates that there is a multilayer surface, and its adsorption affinity is not uniformly distributed on the heterogeneous surface [29]. It also describes the reversible and non-linear adsorption process. The constant ‘n’ in the Freundlich model identifies the adsorption intensity. It is classified as: n > 1 (physical process); n < 1 (chemical process); n = 1 (linear). In adsorption, n > 1 is the most common condition and an n value between 1 and 10 indicates the desired adsorption (Desta, 2013). In this study, it was observed that the value of n for unsealed BSAC was 5.99, which exhibits the most common condition for adsorption physical process and indicate a good adsorption of Mn2+ because it shows high heterogeneous surface and high ion exchange intensity.
Table 6. Isotherm Models Parameters for Sealed and Unsealed RPPAC for Mn2+
Adsorbent
|
Metals
|
Langmuir Isotherm
|
Freundlich Isotherm
|
|
|
KL(L/g)
|
qmax (mg/g)
|
R2
|
RL
|
SE (slope)
|
n (L/g)
|
Kf (mg/g)
|
R2
|
SE (slope)
|
Sealed
RPPAC
|
Mn2+
|
71.480
|
0.097
|
0.999
|
7.0x10-3
|
0.044
|
159
|
0.017
|
0.710
|
0.002
|
Unsealed
RPPAC
|
Mn2+
|
0.499
|
0.099
|
1.00
|
500
|
2.9x10-15
|
926
|
3.5x10-3
|
1.00
|
8.7x10-17
|
Table 6 shows isotherm model parameters of unsealed RPPAC on the adsorption of Mn2+. The R2 for the Freundlich and Langmuir isotherms both have 1.00, so the ‘sips’ isotherm model was used to determine which of the models is the best fit for the adsorption process. The Freundlich and Langmuir isotherm models were combined to create the sips isotherm model, which was developed to forecast the heterogeneity of adsorption systems and get beyond the Freundlich model's restriction caused by higher concentrations of the adsorbate [30]. Regarding the Sips model, the adsorption process will fit Langmuir, when the “n” (heterogeneity factor) value is equal to “1” and for the adsorption process to fit to Freundlich isotherm the Ks constant value must approach “0”. The ‘n’ value obtained for the sips model value was 0.622, which indicates that the adsorption process best fits the Freundlich isotherm model.
The correlation coefficient R2 value for the sealed RPPAC was 0.999 for the Langmuir isotherm and 0.710 for the Freundlich isotherm. The Langmuir isotherm model was then the best fit for the adsorption process for the sealed RPPAC.
3.6 Adsorption Kinetics Studies
The various parameters for the pseudo-first order (PFO) and pseudo-second order (PSO) for the two activated carbon (BSAC and RPPAC) are given in Tables 7 - 10. The PSO model describes most adsorbent active site occupations, while the PFO model only describes a few of these activities. Few active sites are occupied during the early stages of adsorption, which corresponds to the PFO model, whereas most active sites are occupied during the late stages of adsorption, which corresponds to the PSO model (Guo and Wang, 2019).
Table 7. Kinetic Models Parameters for Sealed and Unsealed BSAC for Fe2+
Adsorbent
|
Metal
|
Pseudo first order
|
|
Pseudo second order
|
|
|
R2
|
qe
(mg/g)
|
K1
(min-1)
|
Expt.qe
(mg/g)
|
R2
|
qe
(mg/g)
|
K2
(g/mg min)
|
Sealed
BSAC
|
Fe2+
|
0.305
|
0.067
|
5.6x10-6
|
0.065
|
0.978
|
1.118
|
72.85
|
Unsealed
BSAC
|
Fe2+
|
0.210
|
0.079
|
7.8x10-7
|
0.091
|
0.994
|
0.088
|
2.635
|
The PSO model is the best fit for the unsealed BSAC adsorption of Fe2+, according to the parameters calculated in Table 7. The PSO for the unsealed BSAC has a higher correlation coefficient (R2) value of 0.994 compared with the R2 value of the PFO (0.210) for the unsealed BSAC. Considering their theoretical adsorption capacity at equilibrium, qe values and experimental qe value, it can be observed that the theoretical qe and the experimental qe values for the PSO are in close agreement with each other (0.088 and 0.091) respectively. The theoretical and experimental qe values for the PFO were not in close agreement with each other (0.079 and 0.091). Their adsorption rate constant K2 was higher than K1, which indicates that the adsorption of Fe2+ was faster in the PSO model than the PFO model, making PSO model the best fit for describing the adsorption of Fe2+ by the Unsealed BSAC and implies that the Fe2+ was adsorbed through chemisorption on the surface of the unsealed BSAC. Considering the sealed BSAC, it was observed that the R2 value for the PSO was higher than that of the PFO (0.978 and 0.305), respectively, making the PSO the best fit for the adsorption process.
Table 8. Kinetic Models Parameters for Sealed and Unsealed RPPAC for Fe2+
Adsorbent
|
Metal
|
Pseudo first order
|
|
Pseudo second order
|
|
|
R2
|
qe
(mg/g)
|
K1
(min-1)
|
Expt.qe
(mg/g)
|
R2
|
qe
(mg/g)
|
K2
(g/mg min)
|
Sealed
RPPAC
|
Fe2+
|
0.009
|
0.067
|
1.19x10-5
|
0.021
|
0.948
|
0.023
|
3.777
|
Unsealed
RPPAC
|
Fe2+
|
0.086
|
4.429
|
1.18x10-9
|
0.033
|
0.961
|
0.034
|
2.505
|
The various parameters of PFO and PSO for sealed and unsealed RPPAC for the adsorption of Fe2+ are computed in Table 8. It was observed that the PSO model for Unsealed RPPAC fitted well for Fe2+ adsorption as the correlation coefficient R2 value was (0.961) which was higher than that of the PFO model R2 value (0.086) of the unsealed RPPAC. The theoretical qe value was in close agreement with the experimental qe value of 0.034 and 0.033, respectively. Comparing the adsorption rate constant, K2 value was higher than K1 indicating that, the Fe2+ removal by the unsealed RPPAC was faster in PSO compared to PFO (K2 = 2.505 g/mg.min) and (K1 =1.182x10-9 /min) respectively. The PSO is the best fit implying that the rate limiting step of Fe2+ onto the unsealed RPPAC may be chemisorption.
For the sealed RPPAC, the PSO model has the highest R2 value, which was 0.948, indicating that it is the best fit for the adsorption process. The theoretical qe value for the PSO was very close to the experimental qe value (0.023 and 0.021), respectively, when compared. K2 has a higher adsorption rate than K1 when their respective adsorption rate constants, K1 and K2, are considered (1.19x10-5 and 3.777) respectively. Considering all, the PSO model is the best fit for the adsorption of Fe2+ onto the surface of the sealed RPPAC and its chemisorption process.
Table 9. Kinetic Models Parameters for Sealed and Unsealed BSAC for Mn2+
Adsorbent
|
Metal
|
Pseudo first order
|
|
Pseudo second order
|
|
|
R2
|
qe
(mg/g)
|
K1
(min-1)
|
Expt.qe
(mg/g)
|
R2
|
qe
(mg/g)
|
K2
(g/mg min)
|
Sealed
BSAC
|
Mn2+
|
0.318
|
0.079
|
2.09X10-5
|
0.099
|
0.999
|
0.099
|
9.783
|
Unsealed
BSAC
|
Mn2+
|
1.000
|
7.133
|
0.000
|
0.099
|
1.000
|
0.099
|
8.8X1013
|
Table 9 shows kinetic models for the sealed and unsealed BSAC on the adsorption of Mn2+, the PFO and PSO models. The table indicates that, for the unsealed BSAC, both the PFO and PSO model have the same R2 value (1.00) but their theoretical and experimented qe values are not the same. The theoretical qe value for the PSO agreed with the experimental qe value (0.099 and 0.099), respectively. The theoretical qe value for the PFO was not in close agreement with its experimental qe value (7.133 and 0.099), respectively. Based on the adsorption capacity values, PSO model is the best fit for the adsorption data. When considering the sealed BSAC, it was found that the PFO and PSO models' R2 values were 0.416 and 0.999, respectively. The theoretical qe value for the PFO was lower (0.079) and it was not in close agreement with the experimental qe value (0.099). The PSO theoretical qe value (0.099) was in close agreement with the experimental qe value (0.099). This result supports the PSO's position as the optimum fit for the adsorption process. These findings demonstrate that the Mn2+ diffusion onto the pores of the sealed BSAC and the subsequent chemisorption process determine the adsorption process.
Table 10. Kinetic Models Parameters for Sealed and Unsealed RPPAC for Mn2+
|
Adsorbent
|
Metal
|
Pseudo first order
|
|
Pseudo second order
|
|
|
R2
|
qe
(mg/g)
|
K1
(min-1)
|
Expt.qe
(mg/g)
|
R2
|
qe
(mg/g)
|
K2
(g/mg/min)
|
Sealed
RPPAC
|
Mn2+
|
0.102
|
0.044
|
1.5x10-6
|
0.099
|
0.999
|
0.099
|
10.747
|
Unsealed
RPPAC
|
Mn2+
|
0.062
|
4.424
|
1.1x10-11
|
0.099
|
1.000
|
0.099
|
259.130
|
Table 10 indicates that PSO model of the sealed RPPAC has the highest R2 value of 0.999. It has a higher adsorption rate constant K2 (10.74) compared to PFO of the sealed RPPAC. Analyzing all these parameters shows that the PSO model can best be used to describe the adsorption of Mn2+ ions, thereby adsorption process is a chemisorption. Looking at the unsealed RPPAC, it was observed that the R2 value for the PSO is the highest (1.000) compared with the PFO R2 value which is 0.062. The theoretical qe value for the PSO is the same as the experimental qe value (0.099).