INTRODUCTION

The pore structure and adsorption properties of active carbons otherwise known as activated carbons are strongly influenced by physico-chemical nature of precursor materials and the heat treatment profile (Yun et al., 2001; Jaguaribe et al., 2005). Generally, the quality of active carbons is evaluated in terms of their adsorptive characteristics and superficial area, using different analytical methods for liquid and gas phase adsorption (El-Hendawy et al., 2001; Jaguaribe et al., 2005). For the gas phase, characterization may rely on measurement of the adsorption of nitrogen and application of one of the different procedures available to determine superficial area, such as the BET method (Brunauer et al., 1938) or that of Langmuir (Japanese International Standard, 1992). For the liquid phase, the characterization may be achieved with iodine or methylene blue adsorption (Jaguaribe et al., 2005; Omonmhenle et al., 2006). Determination of the specific surface of active carbon by the low temperature Nitrogen method and to a limited extent by the electron microscope have the disadvantage of being both complex and tedious in operation and therefore the development of a simpler one was desirable (Ekezie and Onwodi, 2005). Consequently, the investigation into the adsorption of iodine by active carbons (Bhatia et al., 2000; Ikhuoria and Omonmhenle, 2006) and subsequent investigation and adoption by ASTM were pursued.

Chemical activation is one of the possible methods for the production of active carbons (Guo et al., 2000; Hayashi et al., 2000, 2002; Baquero et al., 2003; Suarez-Garcia et al., 2004). The carbon precursors are pyrolysed in presence of the chemical activating agent such as phosphoric acid (H₃PO₄), zinc chloride (ZnCl₂), potassium hydroxide (KOH) etc. Phosphoric acid functions as a dehydrating agent and inhibits the formation of tar (Su et al., 2003). The use of phosphoric acid is considered to be an environmentally benign technology, because of the ease of its recovery by washing with water (Baquero et al., 2003). Special emphasis on the preparation of active carbons from agricultural by-products has been given, due to the growing interest in low-cost active carbons from renewable safe copious supplies, especially for applications concerning environmental monitoring. Increased use of agricultural residues is an attractive addition to the utilization of wood and other carbonaceous materials for the production of active carbons. It is expected to help combat global warming and climate change when the burning of the agricultural residues on farm lands are minimized. Active carbons from waste agricultural materials such as bagasse, millet and sorghum straws have been considered in this study, as an attractive solution to mitigate the third world environmental protection needs. Previously, woodbark (Darmstadt et al., 2000; Minkova et al., 2001); bagasse (Minkova et al., 2000, 2001; Garcia-Perez et al., 2002; Jaguaribe et al., 2005); maize cobs (Ekanem, 1996), rice straws (Yun et al., 2001, 2002; Oh and Park, 2002), rice husk (Imagawa et al., 2000), coconut shells (Gimba et al., 2001), oil palm shells, walnut shells (Hayashi et al., 2002), coffee bean husks (Baquero et al., 2003); peach stones (Maroto-Valer et al., 2004) and waste tyres (Mui et al., 2004) have been pyrolysed to produce active carbon, but activation by ortho phosphoric acid impregnation of bagasse, millet and sorghum straws have not been well documented.

In the present study, the effect of phosphoric acid impregnation ratio on the porous texture of the prepared active carbons was investigated with iodine and methylene blue adsorption. The iodine number is a relative indicator of porosity in an activated carbon, although it may not necessarily provide a measure of the carbon`s ability to absorb other species. However iodine number may be used as an approximation of surface area and microporosity of active carbons with good precision but the concentration of the standard iodine solution must be maintained at 0.1 N (ASTM D4607-94, 2006). It is also recommended that the standard iodine solution should have an iodide to iodine weight ratio of 1.5 to 1 (ASTM D4607-94, 2006). Sodium thiosulphate was used as a desorbing reagent because it is not adsorbed by active carbon and its strength does not influence the amount of iodine desorbed (ASTM D4607-94, 2006). According to ASTM, the precision of this test method in the determination of iodine number of active carbons ranging from 600 to 1450 is ±5.6% of the average value measured in milligrams (mg) iodine absorbed per gram of carbon. This range corresponds to two standard deviations (2S) or the 95% confidence limits. The reproducibility of the method is ±10.2% of the average value of iodine number as measured in mg of iodine absorbed/gram carbon. This range also corresponds to 2S or the 95% confidence limits (ASTM D4607-94, 2006).

The larger size of methylene blue makes it suitable for estimating mesoporosity. The suitability of iodine and dyestuffs adsorption for characterizing active carbon have been implicated in several studies by Bhatia et al. (2000), Gimba et al. (2001), Ochonogor (2005), Ochonogor and Ejikeme (2005) and Omonmhenle et al. (2006).

Most important, in adsorption process is a porous solid medium having high adsorptive capacity. A large surface area or large micro-pore volume may be achieved due to the porous structure of the solid. The success of the adsorption process depends on the performance of adsorbents in both equilibria and kinetics (Das et al., 2004). A solid exhibiting favourable adsorption isotherm as well as faster kinetics is supposed to be a good adsorbent. Therefore in order to be a good adsorbent, a solid must have a reasonably larger surface area and relatively larger pore network (Barata-Rodrigues et al., 2003; Das et al., 2004). Transient response of the adsorbent bed to a step-change in the influent concentration is reflective of the adsorbents performance under dynamic conditions (Das et al., 2004).

For the applications involving removal of colour, taste, odour and toxicants from aqueous matrices, adsorption capacity of active carbon may be determined by aqueous phase isotherm technique (ASTM D3860-98, 2003). The amount of constituents removed and the adsorptive capacity is calculated from a Freundlich isotherm (ASTM D3860-98, 2003). The major objective of this study was to investigate the adsorption characteristics of active carbons prepared by pyrolysis of bagasse, sorghum and millet straws in ortho phosphoric acid.

MATERIALS AND METHODS

Pyrolysis and Activation
Dry straws of sorghum and millet were collected after harvest, from Rigasa farm center (non industrial area) in Kaduna state (Northern Nigeria), in March 2006. Sugarcanes were obtained from the same area and processed to bagasse. The straws and bagasse were air-dried for 3 months on plastic mesh, to ensure adequate drying for easy milling and avoid the loss of carbon residue due to oven-drying. The straws were cut into pieces of approximately 3 cm to obtain the samples for milling. The bagasse and chopped straw samples were milled with Christy and Morns miller at the National Animal Production Research Institute (NAPRI), Zaria. The milled bagasse and straw samples were sieved into 1180 particle size with Endecotts laboratory test sieve on Omron No. 17748 (manual timer) sieve shaker. In porcelain crucibles, 0.5 g of the 1180 μm grains of bagasse, sorghum or millet straws was impregnated with H₃PO₄, by adding drop wise, (while stirring the solid to facilitate homogeneous absorption of the acid) the amount of the aqueous acid (2.5 cm³) necessary to produce swelling until incipient wetness (Baquero et al., 2003). Different concentrations of the H₃PO₄ corresponding to 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5 and 7.0 cm³ H₃PO₄ in aqueous solutions were used to vary the content of impregnation agent. After impregnation, the samples were dried for 1 h at 110°C in an oven. The dried samples were transferred to digital thermo control, Gallenkamp electric furnace and heated at the various pyrolysis temperatures (100-450°C) for the various heating periods (residence time) (16-280 min). The solid pyrolysis residues were cooled at room temperature and later transferred to desiccators containing silica gel as desiccant. The carbonized samples were washed sequentially, several times with hot deionised water and finally with cold deionised water until the solutions were neutral to litmus paper. The washed samples were dried at 110°C to prepare the active carbons.

Iodine Adsorption
American standard test method (ASTM D4607-94, 2006) for determination of iodine number of activated carbon was used to assess the porosity of the active carbons prepared. The test is based on a three-point adsorption isotherm. The amount of iodine adsorbed (mg g^-1 carbon) at a residual iodine concentration of 0.02 N was reported as the iodine number. Granular forms of the active carbons were used throughout.

The carbon dosages used were estimated as follows:

Three carbon dosages (2.2594, 1.9801 and 1.7009 g) were calculated using three standard values of C (0.01, 0.02 and 0.03, respectively) and estimated iodine number of 500 mg g^-1 (minimum value recommended for active carbon to be used in removing compounds of low molecular weight (Jaguaribe et al., 2005)). Each weighed sample of carbon was transferred to a clean, dry 250 cm³ Erlenmeyer flask equipped with a ground glass stopper. Ten cubic centimeter of 5 wt.% HCl solution was added to each flask containing carbon. Each flask was stoppered and swirled gently until the carbon was completely wetted. The stoppers were loosened to vent the flasks and they were heated to bring the contents to boil for 30 sec. The flasks were removed and cooled to room temperature (29°C). One hundred cubic centimeter of 0.1 N iodine solution was pipetted into each flask. The addition of iodine solution to the three flasks was staggered to minimize delay in handling. The flasks were immediately stoppered and shaken vigorously for 30 sec. Each mixture was quickly filtered by gravity through one sheet of folded filter paper (Whatman No. 2v). Clean beakers were used to collect the filtrates after discarding the first 30 cm³ portions of the filtrates. Each filtrate was swirled and 50 cm³ of it pipetted into a clean 250 cm³ Erlenmeyer flask. Each filtrate was titrated with 0.1 N sodium thiosulphate solution until the solution turned yellow. Two qubic centimeter of freshly prepared starch indicator solution was added and the titration continued with sodium thiosulphate until one drop produced a colourless solution. The volume of sodium thiosulphate used was noted.

Iodine absorbed (X/M) per gram of carbon (mg g^-1) was calculated as follows:

Concentration of residual filtrate (C) expressed in normal (N) was calculated as follows:

Methylene Blue Adsorption
0.1 g each of the active carbons from bagasse, sorghum and millet straws (obtained using 13.6 acid impregnation ratio) was dispensed into 250 cm³ Erlenmeyer flasks. Fifty cubic centimeter of different solutions of methylene blue containing 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5 and 5.0 mg of methylene blue, respectively, were added and stoppered. In each case, the flask was shaken on Griffin flask shaker at room temperature of 29°C, for various times of 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55 and 60 min, respectively. The mixture in each flask was filtered by suction through a sintered glass crucible and the different filtrates were collected and analysed using Cole 7506 UV-VIS spectrophotometer. Standard solutions of methylene blue were used for calibration. To determine the adsorptive capacity of the active carbons produced, 1.0, 2.5, 5.0, 7.5, 10.0, 15.0, 25.0, 35.0, 45.0 and 50.0 mg of active carbons were used for the adsorption of methylene blue from solutions containing 5 mg methylene blue in 500 cm³ (ASTM D3860-98, 2003).

RESULTS AND DISCUSSION

Starting Materials
From the proximate studies reported by Lori et al. (2007) the effects of particle size on weight loss characteristics, rates of dehydration and volatile matter of the carbon precursors were used to adjudge particle size of 1180 μm as appropriate for the carbonization of the bagasse, sorghum and millet straws. High contents of volatile matter (64.52±1.18-66.65±3.07), fixed carbon (21.93±2.74-24.16±1.94) and low ash (2.27±0.15-3.80±0.17) are favourable properties of the precursors for the production of active carbon (Lori et al., 2007).

Pyrolysis and Activation
Carbon precursors with particle size of 1180 μm were used, in order to overcome difficulties caused by low density and high ash content. The impregnation ration, which was controlled by varying the proportion of H₃PO₄ used for activation, had a strong influence on the yield of active carbon. The yields of active carbons indicate that the temperature domain in which carbonization occurs depends on the interaction between the furnace residence condition and chemical activation. Equilibrium yields of the active carbons were highest at 450°C. At 450°C productions of active carbons from bagasse, sorghum and millet straws were completed in 45 min. The percentage yields of active carbons from bagasse, sorghum and millet straws at 450°C, were 53.6, 30.6 and 29.6% with residence times of 28, 36 and 32 min, respectively. The yields were obtained with optimum impregnation ratio of 13.6. Attainment of steady equilibrium during pyrolysis was the criteria for selecting the best temperature for the production of active carbons from the precursors. Consequently, only carbons produced at 450°C were studied, as there were no steady equilibria attained at 100-350°C.

Adsorption Characteristics
Pyrolysis of bagasse, sorghum and millet straws impregnated with ortho phosphoric acid produced carbon with well developed pore structure and high adsorption capacities. The porous texture of the active carbons was characterised by aqueous phase adsorption of iodine and methylene blue.

Fig. 1:	Effect of phosphoric acid impregnation ratio on evolution of micropores in active carbons from bagasse, sorghum and millet straws

Fig. 2:	Effect of contact time on the adsorption of different initial concentrations of methylene blue [Cx (mg/50 cm3)] on active carbons from sorghum straws

From Fig. 1, it was deduced that the active carbons obtained from low impregnation ratios are microporous as indicated by the steady increase in iodine number below impregnation ratio of 13.6. Higher impregnation ratio did not substantially benefit the evolution of micropores in the carbonised samples as shown in Fig. 1. Except active carbons from millet straws with iodine number of 593, carbons from bagasse and sorghum straw gave iodine number within the reported values (600-647) for carbons from pelletised straws and bagasse, that were activated physically during carbonisation (Minkova et al., 2001; Zanzi, 2001). The results from the present study may be considered superior to those reported by Minkova et al. (2001) and Zanzi (2001) with regard to energy demand of the two processes.

The adsorption of different concentrations of methylene blue on active carbons from bagasse, sorghum and millet straws appeared to depend heavily on the contact time and initial adsorbate concentration (Fig. 2-4). Similar effect of adsorption contact time and initial adsorbate concentration was earlier reported for indigo carmine removal by activated carbons from Terminalia catappa and Cinnarium schweinfurthi nut shell (Ochonogor, 2005). Fast attainment of sorption equilibrium was skewed towards high initial concentrations of methylene blue. With active carbons from sorghum straws, initial methylene blue concentrations of 1.5- 4.5 mg in 50 cm³ aqueous solutions produced adsorption that tends to equilibria after 30 min. The Fastest equilibrium was attained in 25 min with the optimum initial adsorbate load of 0.1 mg cm^-3 methylene blue solution (5 mg/50 cm³ solution). However, adsorption was nearly all time linear below initial methylene blue concentration of 1.5 mg/50 cm³ solution. The adsorption profile of carbons from bagasse and millet straw were similar to the sorption profile obtained for carbons from sorghum straw with the various initial concentrations, except the variability in their adsorption capacities (Fig. 2-4). The maximum amount of methylene blue adsorbed by active carbons from the different precursors in 5 mg/50 cm³ solutions were in the order, sorghum straw (4.65 mg) > bagasse (4.545 mg) > millet straw (4.435 mg) for sorption equilibrium time of 25 min. High impregnation ratio that resulted in increased amount of impregnated phosphoric acid, may have developed pores in the carbon, with wide range of sizes that may be predominantly mesoporous as indicated by the high percentage of methylene blue adsorbed (Fig. 5). This may have resulted from the increased amount of the intercalated phosphoric acid and various polyphosphates, which may produce larger pore volume and size when washed away from the carbons.

Fig. 3:	Effect of contact time on the adsorption of different initial concentrations of methylene blue [Cx (mg/50 cm3)] on active carbons from bagasse

Fig. 4:	Effect of contact time on the adsorption of different initial concentrations of methylene blue [Cx (mg/50 cm3)] on active carbons from millet straw

Similar effects of the phosphoric acid impregnation on porosity development have been reported for other cellulosic precursors (Baquero et al., 2003). The evolution of wide micropores and mesopores for methylene blue adsorption on carbons from bagasse, sorghum and millet straws follow the sequence, sorghum (93%) > bagasse (90.9%) > millet straw (88.7 %) (Fig. 5). General decrease in evolution of the pores with impregnation ratios greater than 13.6 was also showed in Fig. 5.

Fig. 5:	Effect of phosphoric acid impregnation ratio on evolution of wide micropores and mesopores in active carbons from bagasse, sorghum and millet straws

Fig. 6:	Adsorption isotherm of methylene blue on active carbons from sorghum straw

The adsorption isotherm for iodine on active carbons has been reported to follow Langmurian isotherm (Bhatia et al., 2000). Figure 6, 7 and 8 show the methylene blue, Freundlich adsorption isotherms at room temperature (29°C), on active carbons from bagasse, sorghum and millet straws impregnated with phosphoric acid at 450°C with null soaking time. The amounts of the adsorbed methylene blue were calculated from the experimental data using material balance (ASTM D3860-98, 2003). The amount of methylene blue (x) adsorbed per unit weight of carbon (m) was expressed as the difference between the amount of methylene blue before carbon treatment and amount of methylene blue after carbon treatment, per unit weight of carbon. The isotherms were extrapolated and the equations of the lines were used to calculate the x/m that corresponds to the initial amount of

Fig. 7:	Adsorption isotherm of methylene blue on active carbons from bagasse

Fig. 8:	Adsorption isotherm of methylene blue on active carbons from millet straw

methylene blue (C_o) before carbon treatment. This value termed [x/m]_Co represents the amount of methylene blue adsorbed when the carbon is in equilibrium with the influent concentration. The [x/m]_Co is the ultimate capacity of the active carbon for the methylene blue (ASTM D3860-98, 2003). From Fig. 6-8, the adsorptive capacities of the active carbons from bagasse, sorghum and millet straws, for large molecular weight compounds such as methylene blue, were 502, 662 and 390 mg g^-1, respectively. The iodine numbers which indicate the adsorptive capacities for low molecular weight substances were 626, 667 and 593 mg g^-1 for bagasse, sorghum and millet straws, respectively. These iodine numbers are more than 18% higher than the value recommended for low molecular weight compounds by American Water Works Association.

CONCLUSIONS

The adsorptive capacities of the granular active carbons from bagasse, sorghum and millet straws, for large molecular weight compounds such as methylene blue, were 502, 662 and 390 mg g^-1, respectively. The iodine numbers which indicate the adsorptive capacities for low molecular weight substances were 626, 667 and 593 mg g^-1 for bagasse, sorghum and millet straws, respectively. These iodine numbers are more than 18% higher than the value recommended for low molecular weight compounds by American Water works Association. The active carbons from the various cellulosic precursors had similar capacities to adsorb iodine and methylene blue. They are thus, recommended for removal of waste dye from textile effluents and trace metals from wastewater.

ACKNOWLEDGMENTS

We acknowledge the assistance of Gbenga Falode (National Animal Production Research Institute (NAPRI), Zaria, Nigeria) for milling the samples, the Head of Department, Mineral Resources Engineering and staff of the Mining laboratory (Kaduna Polytechnic, Nigeria) for sieving the samples, Yunusa Umar (Department of Chemistry, Jubail college, Saudi Arabia), Dr. Monica Melillo (Brighton University, UK) and Dr. Geoff Morggridge (Cambridge University, UK) for the library resources.