STUDIES ON THE APPLICATION OF ACTIVATED CARBON DEVELOPED FROM THE FERTILIZER WASTE FOR THE REMOVAL OF PHENOLS

Abstract

The presence of small concentrations of organics in drinking water can manifest themselves in many ways. The most frequently reported concern about the presence of organic matter in water is the objectionable taste and odour these compounds can impart. However, the long term physiological hazards from the consumption of water contain ing refractory and other organics have recently received wide publicity in the media as well as in technical journals. The removal of refractory organics viz., phenols, anionic detergents and aromatic polynuclear hydrocarbons etc., from water is a difficult task because of the diverse number of compounds typically present and the extremely low concentrations normally encountered. Adsorption process using activated carbon, for the removal of organic and inorganic substances, has received a widespread attention because of its ability to remove a broad spectrum of pollutants. The process seems to be of special application where drinking water sources are subjected to a wide variety and varying load of pollutants. Although activated carbon has very good adsorption properties, the high cost involved in its manufacture and difficulties encountered in its regeneration makes it necessary to look for other economically feasible adsorbent materials. ii The foremost purpose of the present research is to prepare a low cost carbonaceous adsorbent from the waste slurry generated by the fertilizer plants of this country and to evaluate the physical and chemical parameters associ ated with the adsorption of some phenols on this adsorbent material. In India fertilizer plants produce a waste, which is generated after liquid fuel combustion and washed away from the scrubbers after ferrous sulphate treatment in a slurry form. This material causes a disposal problem and is currently being used as fillers in synthetic rubber plants. Efforts have been made to convert the waste slurry into a carbonaceous adsorbent and see its utility in the removal of some organic contaminants. Phenols impart objec tionable taste and odour to drinking water even when present in traces and are quite toxic for health. As such investi gations were planned to use the carbon generated from ferti lizer waste for the adsorption of some phenolic compounds, viz. phenol, 4-chlorophenol, 4-nitrophenol, 2,4 ,6-trinitrophenol, 1,3-dihydroxybenzene and 1,3,5-trihydroxybenzene . The waste slurry was in the form of small, spherical, black, greasy granules. This was first of all treated with hydrogen peroxide to oxidise the adhering organic material and then heated at 20U°C till the evolution of black soot stopped. The heated product was cooled and then activated in air. Activation, in presence of air, Ill was performed in an ordinary furnace at JI50°C for one hour. The temperature and time were optimised by observing the surface properties of the carbon obtained by activating the raw material at different temperatures. Products obtained at temperature higher than 450 C possess poor adsorption capacity. Activation was also tried in different environment viz. nitrogen and steam but the sample generated in air was found to be the best and hence used in all subse quent studies. The sample was characterized by analysis and termed as 'L' carbon. The following investigations have been included in this dissertation. Adsorption isotherms of the above mentioned phenols on this material under varying physical conditions have been obtained. The plots indicate a faster removal at lower concentrations (extent of removal is forty to hundred per cent) while at higher adsorbate concentrations the removal is comparatively slow (twenty-five to eighty per cent). The parameters evaluated include hydronium ion concentration, temperature and adsorbent dose and competi tive influence of other phenols plus some detergents. It is found that the adsorption of phenols is an endothermic process and the uptake increases with tempera ture. All the phenols are strongly adsorbed at pH below their dissociation constant value (pKa). A direct relation ship is observed between pK value and percentage a IV adsorption. The various substituents on the phenyl ring of the phenols also influence the adsorption of these compounds. The interaction of the aromatic ring of phenol with the surface of active carbon has been considered to be the major influence on the adsorption process. The charge-transfer interaction as postulated by various workers is supported on the basis of the electron densities in the sorbate molecules. As the positive charge on the benzene ring increases the formation of donar acceptor complex between phenol and carbon is facilitated. Hence the phenols in which an electron withdrawing group (-N0„) is the substituent, are more adsorbed in comparison to one in which an electron donating group is present. Addition of surfactant to phenol solution slightly affects its adsorption. Competitive adsorption from a mixture of phenols has also been studied. The uptake of 2,4,6-trinitrophenol, 4-nitrophenol, 4-chlorophenol and 1,3-dihydroxybenzene in presence of each other presents some interesting results. Langmuir parameters have also been evaluated as the data for the adsorption of phenols fits well into the Langmuir model. Kinetic studies have been undertaken, in order, to have an understanding of the mechanistic aspects of the process. Preliminary investigations on the rate of adsorption of 4-chlorophenol, 4-nitrophenol, 2 ,4 ,6-trinitrophenol, 1, 3-dihydroxybenzene and 1, 3,5-trihydroxybenzene on activated carbon indicate an initial rapid adsorption which slows down as it approaches equilibrium. The rate of adsorption is found to be dependent on the initial adsorbate concentration, amount of adsorbent, temperature and particle size of the adsorbent. Film and particle diffusion both influence the diffusion process. Relevant parameters such as diffusion coefficient, activation energy and entropy of activation of the process have also been evaluated. It is observed that the diffusion coefficient (D.) values for five substituted phenols are in the sequence 2,4,6-trinitrophenol > 4-nitrophenol> 4-chlorophenol > 1,3 -dihydroxybenzene > 1,3,5-trihydroxybenzene and follow the order in which these get adsorbed on the carbonaceous material. The trend of energy of activation (E ) is also 3. similar. The positive entropy of activation, obtained for 2,4,6-trinitrophenol and 4-nitrophenol is consistent with the observed adsorption data, while the negative values of entropies of activation (ASff) for 'l-chlorophenol, 1,3-dihydroxybenzene and 1,3,5-trihydroxybenzene are not uncommon and similar observations have also been reported by earlier workers (Chapter IV, 176-179). Regeneration of the spent carbon and the recovery of loaded pollutant is quite important in batch or column operations. Used carbon can either be thrown away as waste or regenerated after dismantling the reactors and this VI would need standby or auxiliary columns. As such studies have been undertaken for the 'in situ' regeneration of adsorbent in batch and column reactors. Efforts in this direction include the removal of adsorbates, 4-chlorophenol, 4-nitrophenol, 2,4,6-trinitrophenol and 1,3-dihydroxybenzene loaded on carbon generated from fertilizer waste slurry. The samples of exhausted adsorbent were treated with various regenerants viz., boiling water, 6M NaOH, 5$ HNCK, 10% HC1, carboxylic acids (formic acid, acetic acid, propionic acid &. n-butyric acid), alcohols (methanol, ethanol, propanol-1, propanol-2) acetone and benzene. Per cent regeneration efficiency, for all the regenerants mentioned above, and a correlation between the molecular size, structure and properties of regenerant to the molecular size, structure and properties of adsorbate has also been tried. In the case of boiling water, it is observed that maximum Regeneration efficiencies (REs) are achieved after approximately 5.0 hours for all the four adsorbates. On the basis of maximum KEs achieved, the ease with which adsorbates could be removed decreases in the order NO > Cl> OH. In the case of carboxylic acids, the RE values are greater than 90%. The trend for decreasing RE values is similar as in the case of boiling water. As the molecular Vll weight of the acids increases, beyond formic acid, the RE values decrease in general i.e., from acetic to butyric acid. Formic acid, however, exhibits a poor regeneration efficiency in spite of low molecular mass. The decrease in RE with increasing molecular size of the acids is very true, in particular, for 1,3-dihydroxybenzene. Thus it can be concluded that the RE value of a regenerant normally depends on its molecular weight. In the case of acetone and benzene, the superiority of acetone as a regenerant compared with benzene has been observed. The effect arises from the smaller molecular size and weight of acetone, thereby giving a better pene trating and adsorbate displacement power. The difference in water solubilities of the two regenerants is more likely to be the reason for the difference in results achieved with acetone and benzene. The Regeneration efficiency values achieved by the alcohols show no obvious correlation with their molecular weights except for the two adsorbates viz., 2,4,6-trinitrophenol, and 4-nitrophenol where a decrease in the same is observed with increasing molecular mass. The molecular weights of all the alcohols were considerably smaller than that of even the smallest adsorbate. This would mean that a very high order of regeneration efficiency was expected with these regenerants. Vlll In the case of 10% hydrochloric acid, 5% nitric acid and 6 M sodium hydroxide, the maximum regeneration efficiencies are observed with sodium hydroxide and minimum with hydrochloric acid. Those organic compounds which possess acidic and basic properties in solution, adsorption is at its strongest in the pH region which yields the higher proportion of undissociated molecules. Thus the acidic conditions created by the application of hydrochloric acid would definitely encourage the adsorption of phenols, thereby discouraging subsequent desorption and so the low values obtained with HC1 are thus quite expected. Higher RE values recorded with nitric acid may be attributed to its oxidising nature. In highly alkaline environment a chemical reaction between the substituted phenols and NaOH may take place. Phenols are converted into their salts by aqueous hydroxide solutions. The formation of the water soluble sodium salt of phenol means that desorption is facilitated, resulting thereby in higher RE values for this regenerant. At different concentration of regenerants the cycle of exhaustion, regeneration and re-exhaustion has been studied. It is found that organic chemical regenerants with solubilising powers are generally much more effective than inorganic chemical regenerants with oxidising powers. In this work each carbon sample, loaded with adsorbates, has been subjected to over 23-28 cycles of regeneration IX and exhaustion using acetone, acetic acid, methanol, nitric acid and sodium hydroxide. The results for all the regene rants exhibit similar characteristics. After 22-27 cycles of repeated exhaustion and chemical regeneration, 100% acetone could achieve a respectable regeneration efficiency for all the adsorbates and the same for 2,4,6-trinitrophenol, 4-nitrophenol, 4-chlorophenol and 1,3-dihydroxyben zene is 90, 82, 70 and 60 per cent respectively. These results reflect that it is desirable to use a regenerant of molecular weight smaller than that of the adsorbate to facilitate the physical displacement of the adsorbate molecule by the regenerant molecule in the confined space of the micropores predominantly responsible for adsorption. RE values achieved by aqueous solutions of acetone dropped very rapidly and negative RE values did result after only 7-9 cycles or so after application of fresh regenerant. Negative RE values indicate that the carbon not only failed to adsorb the substituted phenols from the solution in the re-exhaustion phase, but desorption from carbon took place giving a higher equilibrium solute concentration at the end of the phase than that in the beginning and this results in negative regeneration efficiency. The trend of all the regenerants is approximately similar in nature. With acetic acid values for 2,4,6-trini trophenol, 4-nitrophenol, 4-chlorophenol and 1,3-dihydroxy benzene are 96, 90, 60 and 44 per cent respectively and the RE values for methanol are lesser in comparison to acetone and acetic acid and the same are 90, 82, 58 and 52 per cent respectively. In case of inorganic regenerant i.e., 5% HNO the RE values are 70, 66, 38 and 36 per cent for 2,4,6-trinitrophenol, 4-nitrophenol, 4-chlorophenol and 1,3-dihydroxybenzene respectively. Almost similar pattern of RE values i.e. 70, 69, 44 and 37 per cent for 2,4,6-trinitrophenol, 4-nitrophenol, 4-chlorophenol and 1,3-dihydroxybenzene are observed with NaOH. The practical applicability of the product in column operations has also been investigated to design a fixed bed adsorber. The value of the break-through capacity is more than the batch capacity. This is due to the diffe rence in the nature of the continuous and batch operations. The total time (t ) involved in the primary zone x to establish itself, move down the length of the column of carbon and out of the bed, the time (t , required for o the movement of the zone down its own length in the column after it has been established, the fractional capacity, f, of the carbon in the adsorption zone at breakpoint to continue to remove solute from solutions, the length of the adsorption zone, 6 , and the percentage saturation at break point have been evaluated for carbon columns for adsorbates under investigation viz., i|-chlorophenol , 4-nitrophenol, 2,4,6-trinitrophenol, 1,3-dihydroxybenzene.