Adsorption of Ni ( II ) , Cu ( II ) and Fe ( III ) from Aqueous Solutions Using Activated Carbon

An activated carbon was tested for its ability to remove transition metal ions from aqueous solutions. Physical, chemical and liquid-phase adsorption characterizations of the carbon were done following standard procedures. Studies on the removal of Ni(II), Cu(II) and Fe(III) ions were attempted by varying adsorbate dose, pH of the metal ion solution and time in batch mode. The equilibrium adsorption data were fitted with Freundlich, Langmuir and Redlich-Peterson isotherms and the isotherm constants were evaluated. Time variation studies indicate that adsorptions follow pseudosecond order kinetics. pH was found to have a significant role to play in the adsorption. The processes were endothermic and the thermodynamic parameters were evaluated. Desorption studies indicate that ion-exchange mechanism is operating.


Introduction
Wastewater containing heavy metal pollutants cause direct toxicity, both to human and other living organisms due to their presence beyond specified limits.The main goal today is to adopt appropriate methods and to develop suitable techniques either to prevent metal pollution or to reduce it to very low levels.
Nickel is largely present in the wastewaters of electroplating, motor vehicle and aircraft industries 1 .Acute nickel(II) poisoning causes dizziness, head ache, nausea and vomiting, chest pain, dry cough and shortness of breathe, rapid respiration, cyanosis and extreme weakness 2 .World Health Organization, WHO has suggested that the maximum amount of nickel in drinking water should be 0.1 mg/L 3 .But in many electroplating effluent water it is as high 4 as 50 mg/L.
Human body requires about 2 mg/L of copper but prolonged oral administration of excessive quantities of copper may result in lever damage and chronic poisoning and gastro-intestinal catarrh 5 .Copper is also toxic to aquatic organisms even at very small concentrations 6 .It is found in sizeable amounts in the liquid effluent streams of printed circuit board plants 7 and other effluents.Iron(III) is also toxic at higher concentrations.It is desirable, therefore, to undertake investigations on the removal of these metal ions from water.
Among the various cleanup methods available for metal ions removal, namely, electrolytic reduction, precipitation, oxidation, ultra filtration, ion exchange, adsorption, etc, adsorption (especially, activated carbon adsorption) appears to have the least adverse effects.It includes a broad range of carbonaceous materials at a high degree of porosity and large surface area 8 and finds use for the removal of toxic, biodegradable and non-biodegradable substances from wastewaters.It is attractive as it can treat wastewater to acceptable quality suitable for reuse.
The removal efficiency of a commercial activated carbon (M/s.Loba Company, Mumbai) towards nickel(II), copper(II) and iron(III) was attempted in the present study.The effects of adsorbate dose, pH, time and temperature were studied.

Adsorbent
The granular activated carbon was ground and the portion retained between 150 and 250µm sieves was used for study.The adsorbent was named as CC.

Characterization of CC
Physicochemical characteristics such as, moisture content, density, volatile matter content, total ash content, water solubles, acid extractable content and pH were determined following standard procedures 9 .Boehm titrations were performed to find out the amounts of surface functional groups 10 .Liquid phase adsorption characterizations were done following the method of Maria J. Martin et al. 11 The infrared spectrum was recorded in an Impart -420, Nicolet spectrometer.BET surface area of the sample was measured using nitrogen adsorption isotherms.

Analysis of metal ions
All the metals were estimated following a suitable colorimetric method.Nickel(II) was estimated by the dimethylglyoxime method 12 , copper(II) and iron(III) by the thiocyanate methods 13,14 .

Isotherm procedure
Prior to isotherm studies, minimum contact times for adsorption equilibria to become established were estimated.Each experiment comprised three replicate 100mL glassstoppered bottles containing appropriate amount of adsorbent and 50mL of adsorbate solutions of selected concentrations.Control flasks without the adsorbents are also prepared simultaneously.Mixtures were maintained in a rotary shaker (orbitek) at constant temperature (30, 45 or 60 °C).After the attainment of equilibrium the contents of each flask were filtered through Whatmann No.41 filter paper, with the first 10mL discarded.The filtered samples were then analyzed for unadsorbed metal ions.The equilibrium adsorption data were then fitted to Freundlich, Langmuir and Redlich-Peterson isotherm equations: where q e is the adsorption capacity in mg/g; C e is the equilibrium concentration of adsorbate (mg/L); K F and n are Freundlich constants; K L and b are Langmuir constants; q m is the Langmuir monolayer adsorption capacity and K R , b R and β are Redlich-Peterson isotherm constants.

pH variation studies
In order to find out the optimum pH for maximum removal of adsorbate, experiments were carried out with solutions of same metal ion concentration but adjusted to different initial pH values (with HNO 3 or NaOH).Measurements were carried out below which chemical precipitation of metal hydroxides do not occur.These values have been estimated to be 7.8 for Ni(OH) 2 ; 2.5 for Fe(OH) 3 ; and 7.5 for Cu(OH) 2 .

Desorption Studies
Some desorption experiments were also conducted in order to explore the feasibility of recovering both the adsorbed species and the adsorbent and to elucidate the nature of adsorption processes.They were carried out as follows.After adsorption experiments using the selected adsorbent and adsorbate doses, the adsorbate loaded adsorbents were separated and washed gently with several portions of distilled water to remove any unadsorbed species.The samples were then air-dried and agitated with 0.1M solutions of HCl, AcOH or water for a period of 10 hours and the amounts of desorbed species were determined in the usual way.

Characterization of the adsorbent
The parameters evaluated are listed in Table 1.
Table 1 1 reveals that the carbon under study is of high surface area and that it has quite good amounts of micro and mesopores as well.Apart from having the desired textural characteristics of a good adsorbent, the carbon was also found to have quite good amounts of surface groups like carboxyl, phenolic, lactonic and basic groups.The presence of these groups is also evident from the FT-IR spectrum of the carbon (Figure 1).

Adsorption Isotherms: Adsorption Models
The assumptions associated with the Langmuir isotherm are well known 15 ; Adsorption can not proceed beyond a monolayer coverage and all adsorption sites are equivalent.The Freundlich model, on the other hand, assumes a heterogeneous adsorption surface with sites that have different energies of adsorption and are not equally available.The Redlich-Peterson model is described as a combination of both the other models and is often used to describe equilibrium data over a wide range of concentration.For each and every individual adsorption system, the data were fitted to the three isotherm equations and the results are given in Table 2 [Conditions: adsorbent dose 0.1g/50mL of adsorbate solution; initial concentration of metal ion solutions, C i = 15-100mg/L for Ni; 20-100mg/L for Cu and 15-80mg/L for Fe(III)].All the curves were of L type under Giles classification 16 .
According to Treybal 17 it has been shown using mathematical equations that n values between 0 and 1 represent beneficial adsorption.Indeed, the n values found for all the adsorption systems fall in this range.The Langmuir constant b is a measure of adsorption intensity and the parameter q m is a measure of adsorption capacity.Adsorption capacity of the adsorbents toward metal ions decrease in the order Ni(II), Cu(II), Fe(III) [values in Table 2], whereas the adsorption intensity decrease in the opposite order.The b values found indicate stronger interaction forces between carbon surface and Fe(III) ions compared to Ni(II) and Cu(II), in agreement with the higher ionic potential of Fe(III).The b values determined are further used to calculate the dimensionless separation factor, R L 18, 19 , defined as where C i is the initial solute concentration.The magnitude of R L value gives an idea about the nature of adsorption equilibrium: the process is non-spontaneous when R L is greater than one; favourable when R L lies between 0 and 1; and irreversible when R L is zero.
In all the systems studied, R L values were comprised between 0 and 1 (values not listed) indicating favourable adsorption of all the metal ions on the activated carbon.

Effect of pH
The effects of pH in external solutions on adsorption extent are presented in Figure 2 (adsorbent doses: 0.1g/50mL, C i = 50mg/L).

Figure 2. Adsorption of metals on CC: Effect of pH
For the three metal cations (Ni 2+ , Cu 2+ and Fe 3+ ) increase in solution pH throughout the range studied resulted in increased adsorption.This trend is in agreement with the fact that the metal ions are adsorbed by ion-exchange mechanism.As the solution pH is lowered, concentration of H + ions will increase proportionally which will effectively compete with metal cations for active adsorption sites on carbon surface, thereby reducing the available sites for metal cations.

Kinetics
To describe the adsorption kinetics, a pseudo-second order rate model reported in the literature was applied [20][21][22] in the following form where, h is the initial sorption rate (mg g -1 min -1 ); k 2 is the over all rate constant for the sorption process (g mg -1 min -1 ); q e(2) is the amount of metal ion adsorbed at equilibrium (mg/g); and q t is the adsorbed at time t (mg/g).
The second order kinetic parameters evaluated are presented in Table 3. Good agreement between the predicted and experimental results was found.Hence, the applied pseudo-second order rate model succeeded in representing properly the experimental data for the adsorption of metal ions with high correlation coefficients (r 2 > 0.99).It is also to be noted that the experimental adsorption capacities [q e(exp) ] are very close to the adsorption capacities predicted by the second order kinetic model [q e (2) ], these values are also furnished in Table 3.

Effect of Temperature
The equilibrium studies for all the systems were conducted at two more temperatures in addition to room temperature (30 °C), namely 45 and 60 °C.For all the systems increase in temperature resulted in greater adsorption.A representative graph for showing the effect of temperature is shown in Figure 3 for the Ni (II)-CC system.
The increased adsorption at higher temperatures can be due to one or more of the following reasons.Acceleration of some originally slow step(s); 23 creation of some new activation sites on the adsorbent surface 24 and decrease in the size of the adsorbing species. 25his could well occur due to progressive desolvation of the adsorbing ion as the solution temperature increases.(1) Where K C is the equilibrium constant for the distribution of metal ions between the liquid and solid phases; C ae is the solid phase metal ion concentration, mg/L; C e is the liquid phase metal ion concentration, mg/L; T is absolute temperature, °K and R the gas constant.Equation ( 1) was used to construct Van't Hoff plots and ∆H and ∆S were calculated from the slope and intercept of the Van't Hoff plot, respectively.Thermodynamic parameters evaluated for varied metal ion concentrations are listed in Table 4.
The ∆G values obtained for most CC-metal ion systems were positive indicating that CC is less effective in removing metal ions from aqueous solutions.The positive values of ∆H obtained for all the processes are further conformations that they are endothermic, which is an indication of strong interaction between the adsorbate and the adsorbent.The positive values of ∆S suggest increased randomness at the solid-liquid interface during the adsorption of metal ions.The adsorbed solvent (water) molecules, which are displaced by the adsorbated species, gain more translational entropy than is lost by the adsorbate ions.Furthermore, before the adsorption process takes place the adsorbate ions are heavily solvated (the system is more ordered) and this order is lost when the ions are adsorbed on the surface, due to the release of solvated water molecules.

Desorption Studies
Attempts were made to regenerate the adsorbed metal ions with water, 0.1 M acetic acid and 0.1 M hydrochloric acid as regenerating agents.The results are presented in Table 5.These results indicate that the metal ions are adsorbed by ion-exchange mechanism by the surface groups present on the carbon surface.

Conclusion
The work described has shown that the activated carbon under study can be successfully used for the adsorptive removal of metal ions from solution.The carbon possessed good textural and chemical properties.The three parameter Redlich-Peterson model can be used to represent the equilibrium adsorption of Ni(II), Cu(II) and Fe(III) and was found to be superior than Freundlich and Langmuir models.Adsorptions followed pseudo second order kinetics.Increase in solution pH result in greater retention of metal ions on CC.
The adsorption processes were found to be endothermic and the thermodynamic parameters were evaluated.Dilute hydrochloric acid can be used for desorption of these metal ions.

Figure 3 .
Figure 3.Effect of temperature: Adsorption of Ni(II) on CC Thermodynamic parameters such as Gibbs's free energy change (∆G), enthalpy change (∆H) and entropy change (∆S) were calculated using the following expressions:

.
Characteristics of Activated Carbon

Table 2 .
Isotherm parameters for the adsorption of metal ions on CC

Table 3 .
Pseudo-second order parameters for CC-metal ion systems

Table 4 .
Thermodynamic parameters for the adsorption of metal ions on CC ∆G, kJ mol−1

Table 5 .
Results of desorption