The increase of industrial sites and human activities has negatively impacted the environment, due to the large quantities of toxic species generated and released in the wild in different forms (solid, liquid and gaseous). This constitutes a great concern for public authorities, by the consequences which may occur on living species and their environment.
Currently, industrial wastewater discharges containing relatively large quantities of toxic substances is a growing concern. Their presence in effluents is a threat to any biological organism. Water becomes a vector of pollution. To avoid the toxic effects of these substances and to protect the environment, it is necessary to treat contaminated effluents before their release in the natural environment. Scientists from diverse backgrounds are interested in the identification and removing pollutants directly involved in ecosystem unbalances. Physical or chemical processes have been widely used to remove pollutants from wastewater. These methods include: coagulation-flocculation , biological treatments (biodegradable pollutants) , electrochemical methods  , membrane processes , adsorption  . The use of activated carbon in the adsorption process is in great demand.
In the last decades, activated carbon has attracted a lot of attention because of its versatile application in materials science, meeting various criteria required for new materials . Indeed, activated carbon is a highly porous carbonaceous, hydrophobic, and non-polar adsorbent prepared to obtain a high degree of porosity which confers to it a considerable surface area. Activated carbon is one of the most commonly used adsorbents in many liquid and gas phase applications for its adsorbent properties related to porosity and surface area. The diversity of its applications, its impact on industrial development and health show its importance in our days. It is widely used in environmental technology because of its ease use in the removal of organic and inorganic compounds from aqueous effluents . Activated carbon is one of the most effective material used to remove a wide range of contaminants. It can be used to treat some or all the contaminants, either directly or in combination with other processes. Its utility is no longer to prove. However, because of the high cost of preparation of activated carbon which restricts its frequent use, several scientists have studied this problem in order to find low-cost precursors to prepare activated carbon. Thus, several studies have been conducted in this direction to prepare activated carbons from plant materials such as rice straw , coconut shell   , tamarind wood   and from mineral materials . Their production can be done in two ways, either by physical activation or chemical activation. The physical activation results from the heat treatment and consists of two basic processes: carbonization or pyrolysis and activation. As for the chemical activation, different chemicals such as KOH , NaOH , ZnCl2 , H3PO4  have been used as activation agents. In addition, the efficiency of activated carbons prepared by chemical activation is about 3.7 times greater than those obtained by physical activation .
As a part of the search of natural resources for the local production of cheaper activated carbon, Acacia auriculeaformis wood has attracted our attention because of its abundance. Furthermore, its use is very often limited to carpentry and fuel (firewood or charcoal for cooking). Our previous work  deal with the adsorbent properties of the active carbons of Acacia auriculeaformis and Acacia mangium by methylene blue adsorption. However, using Acacia auriculeaformis as a source of activated carbon production in future applications could be a problem due to deforestation. Therefore, to overcome this eventuality, this species of Acacia could be recommended to farmers as a plantation tree because of its rapid growth even on poor soils. Thus, although this material serves as a precursor of activated carbon, it could also be a source of income for farmers.
The present work focused on the preparation and characteristics comparison of Acacia auriculeaformis activated carbon resulting from chemical activation with three different activation agents. H3PO4, NaOH and NaCl were chosen as activation agents. The objective of this work is to valorize Acacia auriculeaformis charcoal in activated carbon with a view contributing to domestic and industrial waste water treatment.
2. Materials and Methods
2.1. Materials and Carbonization
The precursor of the activated carbons is a charcoal made from the carbonization of Acacia wood (auriculeaformis of 8 to 9 years old) taken from Anguédedou forest located at three kilometers to Abidjan (Côte d’Ivoire). This charcoal was prepared in an experimental butane gas retort at CNRA (National Center for Agronomic Research) Technology Research Station (SRT), Cocody, Abidjan. Acetic acid (99.5%, from Carlo Erba), methylene blue (from Merck Eurolab), phosphoric acid (85%, from Panreac, Spain), sodium hydroxide (95%, from Scharlau, Spain) and sodium chloride (99.6%, from Alpha Chemika, India) all A.R. grade, were purchased from Poly Chimie, Abidjan, Cote d’Ivoire. Distilled water was used for solution preparation.
For carbonization, acacia woods were exposed to ambient air for sun drying for 3 months in order to obtain humidity level of 20% to 25% recommended for good mass yield during the carbonization process . The sun-dried acacia woods were cut into almost parallelepiped shaped pieces with 10 cm long, 8 cm wide and 5 cm thick, then baked and charred in a furnace heated by an external furnace fed with butane gas. The temperature was read each 15 minutes using the thermocouple in order to control the carbonization process and to determine the carbonization time. Thereby, the carbonization temperature reached a plateau of 350˚C after 3.5 hours of carbonization. The carbon was removed after a period of 4 hours of the plateau.
Chemical activation was used according to the methods described by Kra et al.  and Kaledia et al.  using chemical agents such as H3PO4, NaOH and NaCl. The raw charcoal was ground. One of the first two steps of the methods described consists in impregnation of the raw precursor in an aqueous solution of the activating agent and the second one in the heat treatment of the impregnated carbons.
Activation in acidic medium was done following the protocol reported . Thus, in 1 L beaker, 30 g of ground charcoal and 400 ml of 10% phosphoric acid solution were mixed. For NaOH activation, the method proposed by Kaledia et al.  was followed. Therefore, 30 g of crushed raw charcoal were dispersed in 400 ml of 1M solution of sodium hydroxide. As for NaCl salt activation, a saturated saline solution was prepared by dissolving approximatively 14.6 g of NaCl in 400 ml distilled water. Then, 30 g of ground raw material are impregnated in the prepared solution. In each case, after stirring for 30 min at 350 rpm, the mixture was kept standing for 24 hours. Finally, the slurry obtained was filtered and dried in an oven (MEMMERT) at 105˚C for 24 hours. The dry residue was crushed and only the fraction of particles with the diameter between 125 µm and 2 mm were recovered using an AFNOR sieve.
2.2.2. Heat Treatment
The dry sample obtained after impregnation was placed in a muffle furnace (NABERTHERM mark 30˚C - 3000˚C) at 800˚C for 6 hours for heat treatment. The choice of the duration and the temperature of the heat treatment were inspired by the work of Kra et al. . The charcoal, once taken out of the oven at room temperature, was thoroughly washed with hot distilled water until the filtered water pH was between 6.5 and 7. The sample was then dried in oven at 105˚C for 24 hours. Activated carbon is stored in hermetically sealed jars prior to testing after cooling in a desiccator. In this work, the modified carbons were called CAA (activation with H3PO4, acid), CAB (activation with NaOH, base) and CAS (activation with NaCl, salt). The same process was followed for each sample.
2.3. Characterization of Activated Carbons
2.3.1. Activation Efficiency
The activation efficiency R (%) is defined as the mass ratio of activated carbon produced after activation to the mass of the raw material used before activation. The yield values of the activated carbons product are estimated by application of the following Formula (1):
2.3.2. Ash Content
The ash content C (%) was determined according to ASTM (American Standards Technology Method) described as below. In a dry crucible of mass , 1 g of carbon sample is placed and weighed as . The crucible is then placed in the muffle furnace at 650˚C for 7 hours until the ash is obtained. After cooling it is weighed as . The ash content is calculated from the Formula (2):
2.3.3. Determination of Iodine Number
Iodine value is the most fundamental parameter used to measure the performance of activated carbon in terms of activity. The higher the value of iodine value, the higher the degree of activated carbon activity. Iodine number (mg/g) is the amount of iodine (I2) adsorbed (mg) per gram of charcoal in aqueous solution 0.02N iodine. It characterizes the zones accessible to any particle of size less or equal to that of the iodine molecule . The iodine value was determined based the method which is an adaptation of the 1989 CEFIC method and the AWWA B600-76 standard applied by Mamane et al.  during the preparation and characterization of activated carbon based on hulls of Balanites Eagyptiaca and Zizyphus Mauritiana. Therefore, 0.2 g of activated carbon was placed in 100 ml beaker containing 20 ml 0.02N iodine solution and stirred for 4 to 5 minutes. Then, filtered and 10 ml of the filtrate were collected and tested with 0.1N sodium thiosulfate solution in the presence of two drops of starch paste until the solution completely discolored. From the reaction of I2 on sodium thiosulfate whose equation-balance (3), the remaining concentration in I2 ( ) was calculated and the iodine number ( ) was determined from Formula (4).
where (ml) is the volume of sodium thiosulfate poured at the equivalence, is sodium thiosulfate concentration, is I2 initial concentration, is I2 dosed Volume, is iodine Molar mass (253.81 g/mol), is adsorption volume (20 ml) and m (g) is activated carbon mass.
2.3.4. Chemical Characterization
1) Surface Functional Groups
The measurements were done according to Boehm method  . This method was used to determine the composition of activated carbons in acidic or basic surface functional groups. To neutralize the acid groups, 0.1N solution of strong bases such as NaHCO3, Na2CO3 and NaOH were prepared. As far as 0.1N solution of HCl was used to neutralize the basic groups. For each sample, 1 g was suspended in 50 ml of the prepared solution and maintained in constant agitation during 72 hours. After filtration, a back titration with NaOH (0.1N) or HCl (0.1N) solution, help to determine the number of sites of the acidic or basic functional groups through the Formula (5) below:
in which, represents the number of gram equivalent before the reaction, is the number of gram equivalent after the reaction and is the number of gram equivalent reacted.
2) Determination of point zero-charge pH (pHZPC)
The pHZPC or pH of zero point charge corresponds to the pH value for which the net charge at the activated carbon surface is zero. Then, 100mg of activated carbon and 50ml of 0.01M NaCl solution of pH ranging from 2 to 12, pH adjusted by addition of 0.01M HCl or NaOH solution , were kept in constant stirring at room temperature for 48 hours and the final pH was measured. The pHZPC is the point where the pHfinal = f (initial pH) curve intercepts the first pHfinal bisector = f (initial pH).
2.3.5. Surface Characterization
1) Surface morphology of activated carbon
Scanning electron microscopy (SEM) was used to determine the morphology structure of the prepared samples. Images were collected using JSM-840 model.
2) Specific surface area
The determination of the specific surface area of activated carbons was carried out on the one hand, according to the Langmuir model based on the adsorption method of acetic acid (AC) and methylene blue (BM)   and on the other hand by the method developed by Brunauer-Emmet-Teller, commonly called BET method . Regarding the specific surface area related to AC and BM adsorption, it was necessary to determine the equilibrium time, then the adsorption isotherms according to Langmuir model and finally the maximum adsorption capacity Qm. Knowing Qm, the specific surface could be calculated with the following Formula (6):
with SL the specific surface area (m2/g), Qm the maximum adsorption capacity (mol/g) or (mg/g), area occupied by acetic acid or methylene blue (m2) (acetic acid area = 21 Å2 and methylene blue area = 175 Å2) and N Avogadro Number (mol−1).
For this purpose, 1 g of activated carbon was suspended into 50 ml of acetic acid or methylene blue (C0) and stirred. At different interval times of 5, 10, 30, 45 and 60 min samples were collected and the residual concentration was calculated. The process of the isotherm was almost similar to that of the adsorption kinetic. 50 ml of acetic acid or methylene blue at various initial concentrations with 1 g activated carbon under constant agitation during the equilibrium time calculated during the kinetic.To determine the residual concentration at equilibrium Ce, the filtrate was titrated volumetrically in presence of phenolphthalein for acetic acid and spectrophotometrically with a Helios OMEGA type spectrophotometer at a wavelength of 665 nm for methylene blue. The amount of adsorbed adsorbate was determined from Equation (7):
BET surface area, pore volumes and pore dimensions were calculated from nitrogen adsorption isotherms at 77 K. The samples were analyzed using a Quantachrom Autosorb 1 device (Chemical Engineering Department, Laval University). The total pore volume is determined by the method of the adsorbed amount of nitrogen at . The microporous volume is determined by applying the Dubinin-Radushkevich model to the adsorption isotherm.
3. Results and Discussion
3.1. Physico-Chemical Analysis: Yield, Ash Content and Iodine Value
Table 1 shows activation efficiency, ash content and iodine value of carbons activated with three different chemical agents. From Table 1, it can be seen that the activation efficiencies of the prepared activated carbons are less than 50% indicating the formation of micropore and mesopore . In the three cases of activation, the best yields are obtained with the acidic agent and salt activating agent. However, the mass yield related to the salt is greater than that of the acid. This could be explained by combustion reactions that are very pronounced with H3PO4 because of its low stability temperature (boiling temperature 213˚C against 1465˚C for NaCl). Similar observations were reported by Haimour and Emeisch . The weak mass yield obtained with NaOH (26.25%), could be due to the fact that NaOH is a strong base, corrosive base and stable product that can melt at 318˚C. At temperatures above 360˚C, this molten hydroxide reacts easily with most of the raw materials containing carbon (carbon, lignocellulosic materials, ...)  and certainly accelerates the combustion. Since the yield is an indicator of the performance of the preparation method of the material especially on an industrial scale, one can note that NaOH activating agent does not have a major advantage in terms of activated carbon gain because of its low yield. The difference in yields observed suggests that the nature of the chemical activation agent used during the impregnation could be involved.
Table 1. Physico-chimical parameters of the prepared activated carbon.
Ash levels (5% for CAA, 6.6% for CAS and 14.9% for CAB) are relatively low. One of the parameters influencing the adsorption properties of coal is its ash content. This parameter has a significant effect on the quality of activated carbon. It appears that a high ash content decreases the specific surface area. Therefore, the ash content of a good adsorbent must not be too high, i.e. less than 20% . An excessively high ash content (>20%) reduces coal activity, its potential reactivation and can generate leakage of impurities (mineral salts). The ash value reported in this study is an indication for good adsorption capacity of the synthesized activated carbons. In addition, this study shows that activation with phosphoric acid shows low ash content compared to the two other activating agents. This could be assigned to the corrosive action of phosphoric acid which eliminates impurities on precursors . CAB ash rate slightly high compared to the two others suggests that this coal is still rich in the mineral matter after activation. The nature of the chemical agent, therefore, has a considerable influence on the ash content.
Activated carbons with different agents contain pores accessible to iodine molecules. The iodine number values was 304.47 mg/g, 190.35 mg/g and 380.71 mg/g respectively for CAA, CAB and CAS. Similar iodine number (319.67 and 286.26 mg/g) were reported during the characterization of acacia activated carbons prepared with basic agents . The determination of porosity with iodine index helps to define, on the one hand, activated carbons containing many macropores and mesopores (low iodine index), and on the other hand, activated carbons containing mainly micropores (high iodine value). The order of iodine value CAB < CAA < CAS indicates that chemical activation leads to the creation of new pores as well as to the enlargement of available pores. From the three prepared activated carbons, those activated with H3PO4 and NaCl have greater I2 adsorption capacity . It has been found from the literature that activated carbon with iodine value higher than 950 mg/g are mainly used to measure micropores (0 - 20 Å) present in the inner surface of the coal . The obtained values are below this value, suggesting that out of micropores created, the presence of mesopores and macropores is expected.
A comparative study of the three characteristics studied reveals that CAA and CAS show high I2 adsorption capacity, low ash content and high yield contrary to CAB.
3.2. Chemical Analysis: Surface Functional Groups and Zero Point Charge pH
The amounts of functional groups of the prepared carbons are given in Table 2. The results of the chemical analysis revealed that CAA has an acidic functional group on the surface (total acid = 1.85 meq/g), particularly carboxylic and phenolic functions. This is due to the chemical activation with phosphoric acid. However, CAB possesses basic surface functional groups (total basic = 1.18 meq/g) attributed to the basic impregnating agent. The basicity of the sample is partly due to the free oxygen of the basic Lewis sites, which is often attributed to
Table 2. Surface functional groups and zero point charge pH of the prepared activated carbons.
the π electrons of the graphitic planes . With regard to the acid functions for CAB, the lactone group is more. As for CAS, the surface functions are almost neutral (slightly acidic and among the acids, the phenolic groups dominate). Therefore, it can be conclued that it behaves like an acid, then as a base (amphoteric behavior). According to the literature , as adsorption is a surface phenomenon, carbons with acidic surface groups (CAA) will have cation exchange properties, whereas those with a basic character (CAB) will have anion exchange properties.
The pHZPC value helps to determine the acidic or basic character of activated carbon and to know, according to the pH of the solution, its net surface charge. If both, carbon and pollutant molecule, are charged, electrostatic type interactions may exist and explain (in part) preferential or disadvantageous adsorption . The above values (Table 2) were obtained using the curves in Figure 1. CAA has acidic pHZPC while CAB is basic indicating high oxygen content in CAA compared to CAB. However, the pHZPC of CAS is very close to neutral medium. This value confirms the amphoteric nature of the CAS. It should be noted that there are some consistencies between the pHZPC values and the proportions in acidic and basic functional groups  .
3.3. Surface Characterization: Surface Area, Porosity and SEM Analysis
3.3.1. Surfaces Area Related to Acetic Acid and Methylene Blue Adsorption
The kinetic curves for acetic acid and methylene blue adsorption on the various carbons given in Figure 2 show that the residual concentration decreases with time. After 30 min for acetic acid and 45 min for methylene blue, the residual concentration does not seem to change at all. These two times were used as equilibrium time for the isotherm studies.
Figure 3 shows the adsorption isotherms of acetic acid and methylene blueon various activated carbons (CAA, CAB and CAS). IUPAC classifies isotherms into six categories. All the curves in Figure 3 are similar to Type I isotherm of this classification . This type of isotherms is observed in the case of adsorption of microporous solids having pores whose diameter is less than 2 nm. It is also observed in the case of monolayer adsorption. These isotherms of adsorption can therefore be described by Langmuir model.
Figure 1. Zero point charge pH for the different activated carbons.
Figure 2. Equilibrium time for (a) acetic acid adsorption and (b) methylene blue adsorption on different activated carbons.
Figure 3. Adsorption isotherms for (a) acetic acid and (b) methylene blue on different activated carbons.
Figure 4 shows the linear representation of the adsorption isotherms according to Langmuir model. Table 3 indicates that the correlation coefficients (R2) are close to 1. The adsorption of acetic acid and methylene blue on all the carbons can, therefore, be described by Langmuir model.
The maximum adsorption capacity Qm of each carbon was determined to apply Langmuir isotherm model. High Qm value implies good adsorption. The surface specific area offered by each carbon was calculated using Qm and summarized in Table 3. The surface areas related to acetic acid and methylene blue adsorptions decrease in both case in the following order: CAA > CAS > CAB. It seems clear that CAA is the best one of the prepared activated carbon because of its high surface area related to acetic acid, suggesting significant microporous and some mesoporous adsorption. In addition, the large surface area related to methylene blue, indicating mesoporous adsorption. BET method is very accurate method for the determination of the surface area, therefore, it was used to verify the results obtained previously.
3.3.2. BET Specific Surface Area and Porosity
N2 adsorption data were obtained from the Department of Chemical Engineering, Laval University, Quebec. However, N2 adsorption isotherms at 77 K of the prepared activated carbons with different activating agents are shown in Figure 5 and pore properties are given in Table 4. These isotherms are type I according to IUPAC classification. The type I isotherm represents monomolecular
Table 3. Summary of the different surface areas obtained.
Figure 4. Langmuir isotherm model of (a) acetic acid and (b) methylene blue adsorptions.
Figure 5. N2 adsorption isotherm on CAA, CAB and CAS.
Table 4. BET and BJH surface properties of CAA, CAB and CAS.
adsorption and applies only to microporous adsorbents with small pore sizes. Therefore, it can be concluded that activated carbons prepared from the above methods are mainly microporous.
Specific surface areas are calculated according to BET method . From the basic hypotheses of BET's theory, BET equation follows Equation (8):
where V is the volume adsorbed at the pressure P, Vm is the volume of gas necessary to cover 1 g of adsorbent with a single layer of gas and C is BET constant.
BET equation is valid for low relative pressures . By plotting of versus a straight line is obtained whose slope and intercept allow us to calculate Vm and C.
Figure 6 shows the linear representation of BET equation for the three prepared carbons. The linear fitting curves obtained seem to confirm our previous results (R2 close to unit). The knowledge of Vm helps in the determination of Qm and consequently BET surface area (SBET) which is calculated using the following Equation (9):
with σ area occupied by a molecule of N2 vapor, N Avogadro Number (6.025 × 1023 mol−1) and at T = 77 K, the surface of a nitrogen molecule is σ = 16.2 Å2.
Table 4 indicates that CAA sample has a specific surface area higher than that of CAS and CAB samples and the same order is followed for N2 amount adsorbed. This would be assigned to the nature of the activating agent used during
Figure 6. Linear representation of BET equation for CAA, CAB and CAS.
the chemical activation. The decreasing order of the BET specific surfaces area of the different carbons (CAA > CAS > CAB) was in good accordance with those found in acetic acid and methylene blue adsorption studies. In view of the results obtained for surface area determination, BET method seems to be the best one indicated for the determination of the specific surface of a porous solid. These results once again confirm the iodine indices obtained (Figure 7).
The nitrogen adsorption isotherms also give information about pore volume and pore diameter (Table 4). The pore size distribution is calculated according to Barrett, Joyner and Halenda method  also called BJH method. Figure 8 shows pore diameter distribution of each prepared activated carbon. Pore size can be described by its width (for slit-like pores), that means the average distance between two opposite walls, or by its diameter. According to the classification recommended by IUPAC , there are three groups of pores according to their size: micropores (size < 20 Å), mesopores (size between 20 Å and 500 Å), macropores, (size > 500 Å). However, the thresholds between the different categories are not really so well defined, as their determination depends particularly on the shape of the pore. BJH pore size distribution curves show peaks ranging from 10 to 50 Å, with a maximum at around 11.26 Å for CAA, 11.78 Å for CAB and 11.77 Å for CAS. This result indicates the presence of micropores and mesopores in the three adsorbents. However, micropores appear to be predominant in activated carbons independently to the activating agent. Since the adsorption capacity of activated carbons is the most important property directly related to the specific surface area and pore volume  and according to BET analysis, CAA seems better among the three carbons prepared.
SEM images were obtained from the Department of Chemical Engineering, Laval University, Quebec, Canada. The sample of activated carbons was observed under three different magnifications which are displayed in Figure 9. The surface morphologies of the prepared materials are quite different, despite
Figure 7. Effect of activation agent on (a) surface area and (b) iodine number.
Figure 8. Pore diameter distribution of CAA, CAB and CAS.
Figure 9. SEM images of CAA, CAB and CAS at different magnifications (10 μm, 20 μm and 100 μm).
the fact that all are from the same biomass. The difference in texture between CAA, CAB and CAS could be assigned to the type of activating agent used during their synthesis. These observations once more indicate that the activating agent has some effects during the activation and thus on pore creation.
A set of activated carbons from local material (Acacia auriculeaformis) chemical activated with H3PO4 acid, NaOH base and NaCl salt were prepared and characterized. As compared to the three methods (AC, BM and BET) used for surface area determination, it was found that BET method gives the best results. Thus, the prepared activated carbons have specific surface areas ranging from 265.00 to 561.60 m2/g and pore volumes from 0.13 to 0.26 cm3/g. These specific surface areas confirm the orders of adsorption capacity of iodine ranging from 190.35 to 380.71 mg/g. The chemical activation of this material has provided active microporous sites (isothermal type I). Boehm chemical analysis and the determination of zero point charge pH showed the presence of various functional groups on the surface of the activated carbons and some uniformities between the pHZPC values and the proportions in acidic and basic functional groups. At the end of this work, it is clear that carbon from Acacia auriculeaformis activated with phosphoric acid seems better than the two others as it provides high yield (41.81%), low ash rate (5%), high specific surface area (561.60 m2/g) and large pore volume (0.26 cm3/g). However, the next work will be devoted to the study of the adsorptive capacity of this carbon in the removal of various metallic pollutants in order to evaluate its effectiveness in water treatment.
The authors express their heartfelt thanks to Prof. Serge Kaliaguine, Chemical Engineering Division, University Laval, Canada, for BET analysis facility.