A comparative study on Co(II) removal capacity from water samples by sorption using limestone and nanolimestone

Powdered nanolimestone (NLS) and limestone (LS) have been investigated as an adsorbent for the removal of cobalt from aqueous solutions. Batch experiments were carried out to investigate the effect of pH. The favorable pH for maximum cobalt adsorption was 6.8. The surface area increased in the case of NLS up to 6.2 m 2 /g, while it was equal to 0.5 m 2 /g in the case of LS. The adsorption capacity calculated by the Langmuir equation was 17.1 mg/g for LS and 60.0 mg/g for NLS at pH 6.8. The adsorption capacity increased with temperature and the kinetics followed a ﬁ rst-order rate equation. The enthalpy change ( Δ H o ) was 20.8 Jmol (cid:1) 1 for LS and 41.6 Jmol (cid:1) 1 for NLS, while entropy change ( Δ S o ) was 33.3 JK (cid:1) 1 mol (cid:1) 1 for LS and 74.8 JK (cid:1) 1 mol (cid:1) 1 for NLS, which substantiates the endothermic and spontaneous nature of the cobalt adsorption process. All of the results suggested that the NLS is very strong and could be an excellent nano-adsorbent for cobalt contaminated water treatment more than limestone.


INTRODUCTION
Cobalt(II) ion is a toxic heavy metal ion in industrial wastewater. A trace amount (μgL À1 ) of cobalt is required and necessary for some organisms as a cofactor for enzymatic activities. However, for most of organisms, concentrations at ppm (5 mg L À1 ) level are known to be toxic because of the irreversible inhibition of some enzymes by heavy metal ions (Leyssens et al. ). In humans, cobalt deficiency has never been reported, but the toxic effects of excess cobalt have been fully described as excess intake will cause acute poisoning by mouth in humans producing gastrointestinal upsets, and chronic absorption may cause abnormalities in skin, heart, blood, or lungs (Payne ).
Due to the mobility and toxicity in natural water ecosystems, the presence of Co(II) ions in surface water and groundwater poses a major inorganic contamination problem. To date, there are various technologies for removing heavy metal ions from solution, including filtration, surface complexation, chemical precipitation, ion exchange, adsorption, electrode position, and membrane processing (Al-Qodah ).
Most of these processes are unacceptable owing to their high cost, low efficiency, disposal of sludge, and inapplicability to a wide range of pollutants (Quintelas et al. ).
Adsorption, on the other hand, is one of the most recommended physico-chemical treatment processes that is commonly used and applied for heavy metals' removal from water samples and aqueous solutions. In addition, the adsorption process is well recognized as one of the most efficient methods for removal of heavy metals from their matrices. Adsorption is mainly based on the utilization of solid adsorbents from either organic, inorganic, biological, or low cost materials (Camel ). Heavy metal removal via adsorption by organic adsorbents is usually accomplished by the application of polymeric ion-exchangers in which the binding and interaction of metal species with these adsorbents is favored via ion-exchange mechanism, or by application of chelating polymers where the target metal ions are directly attached to the adsorbents via chelating or complex formation mechanism (Gode & Pehlivan ). Naturally occurring materials, either modi- Removal and extraction of heavy metals based on applications of biosorption approach are commonly performed owing to the major advantages of various biosorbents, such as economical nature, eco-friendly behavior, regeneration for multiple uses, and high selectivity towards different metals (Quintelas et al. ; Singh et al. ).
Low cost adsorbent materials originating from industrial products or wastes are also known as biosorbents and are widely used for the removal of heavy metals from water samples, and these include components of plants, wood, grasses, compost, peat moss, and carbon materials

INSTRUMENTAL STUDIES
The difference between nanolimestone (NLS) and LS is   ) The remainder of the samples was composed of common minor constituents such as silica, clay, feldspar, pyrite, and sedrite (Bates & Jackson ). The samples were dried for 2 h in an oven at 125 C, packed into stoppered bottles, and stored in a desiccator for future use.
Functional groups of LS were characterized through infrared analysis.
The LS spectrum coincided with pure CaCO 3 . The surface area and porosity of NLS/LS was measured using the Brunauer, Emmett and Teller (BET) method (Quantachrome TouchWin). LS presented no BET porosity and its measured surface area was 0.50 m 2 g À1 for limestone and 6.2 m 2 g À1 for nanolimestone. The pH values of points of zero charge (pHPZC) using a Zetasizer Nano ZSP (Malvern Panalytical) were 9.1 (not aged), 6.2 (aged 60 min) and 8.3 (aged several days), and this agreed with previously reported data (Somasundaran & Goddard ). Stirring the LS and NLS sorbent with distilled water (pH ¼ 6.8) for 1 h decreases the suspension pH to 5.8, confirming the positive charge of the LS and NLS surfaces. Also, the concentration of calcium ion in the solution was measured before and after adsorption in order to confirm that cation exchange was involved.

Reagents
All the solutions were prepared from certified reagent grade chemicals. A cobalt chloride stock solution of 1 molar concentration was prepared and the working solutions were made by diluting the former with doubly distilled water.
Aqueous solutions of HNO 3 and NaOH were used for pH adjustments.
Nanolimestone was prepared by dissolving the powdered limestone in about 150 g of concentrated HCl. The obtained calcium chloride solution was mixed with 1.5 g of chitosan, which was dissolved in 3% acetic acid. This mixture was blended with 70 g Na 2 CO 3 and slightly heated to complete the reaction. The mixture was kept overnight, then filtered and washed several times with water. Finally, it was calcined in a muffle furnace at 650 C for 2 hr (Hariharan et al. ).

Experimental data analysis
The percentage adsorption of Co(II) ion from the solution was calculated from the relationship: where C i corresponds to the initial concentration of Co(II) ion and C f is the residual concentration after equilibration.
The metal uptake q (mg/g) was calculated as: where m is the quantity of sorbent (mg) and V the volume of the suspension (mL).

Effect of pH
The The decrease in the removal efficiency at high pH values may be attributed to the fact that the negative species of cobalt, Co(OH) 3 À and Co(OH) 4 2À , are not capable of a combination with the negative surface of LS, as determined by ZPC (pH ZPC ¼ 6.2 after which the surface is negative).
Moreover, this finding was confirmed by stirring NLS or LS with distilled water; the pH of the suspension was always decreased from 6.8 to 5.8. Therefore, pH 7 was recommended throughout all the other experiments.

Sorption model
Adsorption

).
A standard parameter was used for studying the behavior of the metal adsorption onto the adsorption surface, 50 mg.L À1 of the Co(II) ions adsorbent on LS and NLS surface using the Morris-Weber equation (Dogȃn et al.

)
: where q is the amount of Co(II) ions adsorbed (mg/g).
In Figure 5, the Morris-Weber curve shows that an initial linear portion may be due to the boundary layer effect Log(q e À q) À log q e ¼ ÀK ads t=2:303 (6) where q e is the amount of Co(II) ions adsorbed at equilibrium (mg/g), K ads is the first order rate constant for Co(II) ions adsorption onto sorbent (min À1 ).
The linear plot of Log (q e À q) vs. t (Figure 6) shows the appropriateness of the above equation and, consequently, the first-order nature of the process involved. The value of K ads was calculated to be 0.027 min À1 for LS and 0.04 min À1 for NLS.
The amount of the metal which may be introduced into the pores may be investigated by using Bangham's equation  (Mishra ): where K o is the proportionality constant, and α is Bangham's equation constant.
These results (Figure 7) show that the diffusion of Co(II) ions onto LS pores played a role in the adsorption process (Ali et al. ). The value of α constants deduced were 0.11 for LS and 0.21 for NLS, respectively, favored to be (less than 1).

Langmuir isotherm
The Langmuir model is widely used for modeling equilibrium data to indicate the monolarity of the LS surface from the following equation (El-Sheikh et al. ): where b is the monolayer adsorption capacity and relates to the heat of sorption (L.mg À1 ), Q max is the maximum adsorption capacity (mg.g À1 ).

Freundlich isotherm
The Freundlich expression is an empirical equation describing sorption to a heterogeneous surface (El-Sheikh et al.

). The Freundlich equation is presented as:
ln q e ¼ ln K f þ 1=n ln Ce where K f (mol 1Àn L n g À1 ) represents the sorption capacity when metal ion equilibrium concentration equals 1 and n represents the degree of dependence of sorption with equilibrium concentration. Favorable adsorption was demonstrated by the fact that the value of n is greater than unity.

Dubinin-Radushkevich isotherm
The Dubinin-Radushkevich (D-R) isotherm is more general than the Langmuir, because it does not assume a

):
ln q ¼ ln q (DÀR) À βε 2 (10) where q (D-R) is the theortical adsorption capacity (mg.g À1 ); β is the activity coefficient related to mean sorption energy (mol 2 kJ À2 ); ε is the Polanyi potential; R is the ideal gas constant (0.008314 kJ mol À1 K À1 ) and T is the absolute temperature in Kelvin (K). E (kJ mol À1 ) is defined as the free energy change required to transfer 1 mol of ions from solution to the solid surfaces, which equals: The magnitude of E is useful for estimating the type of sorption reaction. If E is in the range of 8-16 kJ mol À1 , the sorption is governed by chemical ion-exchange. In the case of E < 8 kJ mol À1 , physical forces may affect the sorption.
On the other hand, sorption may be dominated by particle diffusion if E > 16 kJ mol À1 (Sari et al. ).
From the results of D-R model simulation (Table 1), E value is 7.6 kJ mol À1 for LS and 7.8 kJ mol À1 for NLS, indicating that sorption is governed by chemical ion-exchange according to the theory of the D-R model or sorption may be considered as physical-chemical adsorption (Özcan et al. ).

Thermodynamic parameters
In order to investigate the effect of temperature on the adsorption of Co(II) ions onto LS, the distribution coef- The other thermodynamic parameter, Gibbs free energy (ΔG o ) was calculated by: where R is the universal gas constant (8.314 Jmol À1 K À1 ) and T is the temperature (K).
The K d value increases with increasing temperature revealing the adsorption of metals onto LS to be endothermic.  thermodyamically. In addition, the reaction proceeded physically, and these results were in good agreement with that obtained from the D-R isotherm.

CONCLUSION
The experimental results indicate that NLS and LS can be successfully used for the adsorption of Co(II) ions from aqueous solutions. The equilibrium data well followed the Langmuir and Freundlich models. The value of mean sorption energy, E, obtained from the D-R isotherm indicated that the adsorption of the metals on the LS was feasible and spontaneous.
The negative ΔH o value depicted that the adsorption of Co(II) onto LS was an endothermic process, and the increase in K d values with increasing temperature also supported this conception. The positive ΔS o values revealed the randomness of the adsorbed system. On the basis of all results, it can be calculated that limestone (LS) can effectively be used for the removal of cobalt metal cations from different water systems using adsorption method. The nanolimestone presents the major advantage of providing low cost recovery processes making it useful for use in water purification.