^{1}

^{*}

^{2}

^{2}

^{2}

This study aims at exploring arsenite (As (III)) removal from water using naturally available rocks (laterite, sandstone and shale) in Côte d’Ivoire. The study focused on the adsorbent dose, operating pH, contact time, initial arsenite concentration, and modelisation on the removal of arsenite by performing batch adsorption experiment with well water. The optimal dosage related to an initial As (III) concentration of 5 mg/L was about 50, 75 and 145 g/L for laterite, sandstone and shale respectively. Laterite has a better adsorption capacity in comparison to sandstone and shale. On the other hand, kinetic study reveals that the equilibrium times are 5 h for laterite, 3 h for sandstone and 8 h for shale. Results showed that laterite, sandstone and shale could remove the arsenic in groundwater at initial arsenic concentrations below 5 mg/L, satisfying the World Health Organization (WHO) standard for drinking water. Moreover, kinetics study showed that the overall adsorption rate of arsenite was described by the pseudo-second-order kinetic model.

Groundwater is the primary source of drinking water for many densely-populated countries in the world [^{th} - 21^{st} century calamity. Indeed, according to WHO [

As (III) stock solution (1000 mg/L) was prepared by dissolving reagent grade As (III) oxid of 99.5% purified into deonized water. The volume of the solution was made up to 1L in a standard flask. The working solutions containing arsenic were prepared by dissolving appropriate amount of arsenic from stock solutions in well water. The pH of well water varied from 6.3 to 6.6. The experiments were performed at ambient temperatures up to 25˚C.

Laterite contained goethite, quartz, hematite, gibbsite and kaolinite, while sandstone main components were goethite, quartz and hematite. In this shale, appear mainly, small crystals of quartz, albite, microcline, chlorite, kaolinite and dolomite. The data of elemental composition of laterite highlights that silica represents 20% of the material, while the percentage of potential adsorbents mineral oxides Al_{2}O_{3}, Fe_{2}O, MgO and MnO was about a 63.13%. Concerning sandstone, it contained more silica (57.76%) than potentially adsorbent mineral (36.58%) [_{2} 55.43%) followed by alumina (Al_{2}O_{3} 15.46%) and iron oxide Fe_{2}O_{3} (9.21%). Other oxides (MnO, MgO, CaO, Na_{2} O; K_{2}O; TiO_{2}) in very small proportions, between 0.29% and 3.13%, are also contained in it. The rate of constituent elements potentially adsorbent minerals is 27.98%. This support contains oxide of calcium with a rate of 2.34% [

Batch experiments were performed by adding the sorbent in bottles (500 mL) and aqueous As (III) solution at desired initial pH. For all experiments, initial pH of As (III) solution was controlled with a pH-meter by adding nitric acid (HNO_{3}) and/or sodium hydroxide (NaOH) solution as required. To determine the optimal dose of sorbent, a wide range of adsorbent masses (0.6; 0.8; 1; 2; 3; 4; 5; 6.2; 7; 8 g) were shaken in 40 ml of an arsenic solution (5 mg/L). The samples were agitated with rotary shaker (Retsch, Berlin) at 200 rpm for 24 h. After filtration through a 0.45 μm cellulosic acetate film, the As (III) concentration of the filtered solutions was analyzed with Optical Emission Spectrometer OPTIMA 2100 Dual View (ICP-OES 2100 DV). The As (III) adsorbed percentage was calculated using this relation (1):

% As ( III ) adsorbed = ( C 0 − C f ) C 0 ∗ 100 (1)

The amount of As (III) adsorption at any time t, q_{t} (mg/g), was calculated according to Equation (2):

q t = ( C 0 − C f ) m ∗ V (2)

where:

C_{0} (mg/L) = Initial arsenic concentrations

C_{f} (mg/L) = Equilibrium arsenic concentrations

V (L) = Volume of the As (III) solutions

m (g) = Adsorbent mass

q_{t} (mg/g) = Adsorption capacity.

The effect of solution pH was carried out by adding the optimal dose of sorbent in 40 mL of As (III) solution at 5 mg/L as initial concentration at different pH values (4.0 - 10.0). These pH values were obtained by adding into each solution the required amounts of dilute nitric acid (HNO_{3}) or sodium hydroxide (NaOH). The mixture was agitated with a rotary shaker (Retsch, Berlin) for 12 hours at 25˚C. The As (III) adsorbed percentage was calculated according to Equation (1).

The adsorption kinetic study was performed for As (III) in aqueous solution at pH 7 and room temperature (25˚C). Several glass vials were used to hold 40 mL As (III) aqueous solution of known initial concentration (1, 5 and 10 mg/L) and optimal dose of different adsorbents (Laterite, Sandstone and Shale), and shaken at 200 rpm for 24 hours. Samples were taken at a definite time interval and filtered through a 0.45 μm cellulosic acetate film. Filtrates were analyzed to determine residual As (III) concentration.

In the present investigation, the adsorption data were analyzed using three kinetic models: the pseudo-ﬁrst-order, pseudo-second-order kinetic and the intraparticle diffusion models.

The first-order Lagergren’s equation is used to determine the rate of the reaction. The equation is:

log ( q e − q t ) = log q e − K 1 2.303 ∗ t (3)

where K_{1} = constant rate of adsorption, q_{e} = amount of solute adsorbed (mg/g) at equilibrium, q_{t} = amount of solute adsorbed (mg/g) at any time t and t = time (min). When log(q_{e} − q_{t}) is plotted against t, and K_{1} could be obtained from the slope of the straight line.

The pseudo-second-order reaction is greatly inﬂuenced by the amount of pollutant adsorbed on the material’s surface and the amount of equilibrium adsorbed pollutant. The pseudo-second-order kinetics may be expressed in a linear form as

t q t = 1 k 2 q e 2 + t q e (4)

where the equilibrium adsorption capacity (q_{e}), and the second order constants k_{2} (g/mg h) can be determined experimentally from the slope and intercept of plot t/q versus t.

The kinetic experimental results were also be ﬁtted to the Weber’s intraparticle diffusion model [_{d}) can be calculated from the Weber Morris equation. The equation is:

q ( t ) = K d t 0.5 + C (5)

where, q(t)= amount of As (III) adsorbed in mg/g, t = time in minute.

A batch test was performed to determine the best sorbent concentration.

The effect of contact time on the amount of arsenic adsorption by laterite, sandstone and shale was studied using optimal mass of the adsorbents, at pH 7.0 with initial concentration of As (III) at 5 mg/L (

Figures 3-5 present respectively the effect of initial arsenic concentration on laterite, sandstone and shale adsorption capacity. The As (III) adsorption capacity increased with increasing initial arsenic concentration (

Arsenic adsorption efficiency by laterite, sandstone and shale versus pH is reported in

Materials | Experimental parameters | Pseudo-first-order kinetic model | Pseudo-second-order kinetic model | Intra-particle diffusion model | |||||||
---|---|---|---|---|---|---|---|---|---|---|---|

C_{o} (mg/L) | q_{e}_{ }_{exp} (mg/g) | q_{e} _{cal} (mg/g) | K_{1} (h^{−1}) | R | q_{e} _{cal} (mg/g) | K_{2} | R | K_{d} (mg/h^{1/2}g) | C | R | |

Laterite | 1 | 0.034 | 0.001 | 0.107 | 0.422 | 0.031 | 1809.901 | 0.998 | 0.001 | 0.028 | 0.725 |

5 | 0.076 | 0.005 | 0.059 | 0.322 | 0.073 | 262.424 | 0.998 | 0.003 | 0.060 | 0.673 | |

10 | 0.219 | 0.028 | 0.039 | 0.219 | 0.199 | 172.932 | 0.998 | 0.013 | 0.151 | 0.757 | |

Sandstone | 1 | 0.022 | 0.001 | 0.049 | 0.277 | 0.022 | 1331.689 | 0.999 | 0.0006 | 0.020 | 0.635 |

5 | 0.050 | 0.003 | 0.103 | 0.534 | 0.050 | 203.402 | 0.999 | 0.002 | 0.040 | 0.769 | |

10 | 0.144 | 0.019 | 0.034 | 0.243 | 0.133 | 47.420 | 0.998 | 0.009 | 0.098 | 0.673 | |

Shale | 1 | 0.010 | 0.002 | 0.092 | 0.592 | 0.009 | 355.900 | 0.997 | 0.0008 | 0.006 | 0.837 |

5 | 0.020 | 0.002 | 0.073 | 0.277 | 0.017 | 240.955 | 0.992 | 0.001 | 0.014 | 0.672 | |

10 | 0.056 | 0016 | 0.092 | 0.643 | 0.053 | 40.896 | 0.998 | 0.005 | 0.030 | 0.901 |

decreased from 98.01% to 91.62% and from 95.32% to 86.66% on laterite and sandstone respectively. The adsorbents surfaces are highly protonated and As (III) mainly exists neutral form H_{3}AsO_{3}. The Arsenic adsorption of hydrous iron and/or aluminium oxide of adsorbents surface is mainly by ligand exchange. The ligand exchange is envisaged like Stumm [

MOH ( s ) + H 3 AsO 3 ( aq ) → MH 2 AsO 3 + H 2 O

where, M as iron or aluminium.

The Point of Zero Charge (PZC) of laterite is 6.8 and the PZC of sandstone is 4.3 - 5.6 [

Kinetic models including the pseudo-first-order model of Lagergren, the pseudo-second-order model of Richie and intra-particle diffusion models were tested for experimental results simulation.

The fact that the plot would be found to be linear with a week correlation coefﬁcient (Figures 7-9), would indicate that Lagergren’s equation is not appropriate to describe the As (III) adsorption. However, it was observed that the Lagergren pseudo-ﬁrst-order model did not ﬁt well, since the calculated q_{e} values do not agree with the experimental q_{e} values (

The fraction of arsenic adsorbed using the pseudo-second-order model is presented in Figures 10-12. The calculated values of K_{2}, experimental values of q_{e}

and the corresponding linear regression correlation coefficients R are presented in _{e} (cal) obtained with the pseudo-second kinetic model, are in agreement with experimental adsorption capacity q_{e} (exp). The pseudo-second-order model describes better the effect of arsenic adsorption by the adsorbents, and suggests that chemisorption could be the dominant mechanism in the As (III) adsorption by laterite, sandstone and shale.

The graphs are plotted between q(t) and t^{0.5} and are shown as Figures 13-15. K d , the constant rates for intra-particle diffusion are determined from the slopes of the linear portion of the respective plots and are shown in

The Figures 13-15 shows that all initial concentration presented two stages. The first stage could correspond to the mass transfer of the absorbed ions from the bulk solution to the adsorbents surface or instantaneous reactions and the second stage is the intra-particle diffusion on absorbents. It is appears that those intra-particles rate constant values (K_{d}) increased with initial As concentration. The increase of K_{d} with the increase of initial As concentration could be explained by the growing effect of driving force which will reduce the diffusion of As species in boundary layer and enhance to diffusion in the solid. Otherwise, the high K_{d} values of laterite could be related to its high porosity and specific area in relation to the sandstone and shale [

Laterite, sandstone and shale were successful in removing arsenic from groundwater. About 88%, 83% and 76% arsenic was removed respectively by laterite, sandstone and shale using dose of 50, 75 and 145 g/L, for an initial arsenic concentration of 5.0 mg/L. Studies revealed that for optimal operation, the pH should be set between 6 and 7. From kinetic study, it is observed that maximum adsorption occurs in five hours for the laterite, three hours for the sandstone and eight hours for the shale. The pH, contact time and initial concentration, affect significantly the As (III) absorption capacity. Water satisfying the World Health Organization (WHO) standard for drinking water for concentrations below 5 mg/L. The pseudo-second-order model better describes the adsorption of As (III) on the laterite, sandstone and shale. The adsorption process is dominated by the chemisorption.

Koua-Koffi, N.A.A., Coulibaly, L.S., Sangare, D. and Coulibaly, L. (2018) Laterite, Sandstone and Shale as Adsorbents for the Removal of Arsenic from Water. American Journal of Analytical Chemistry, 9, 340-352. https://doi.org/10.4236/ajac.2018.97027