STUDIES ON THE REMOVAL OF SOME ORGANIC AND INORGANIC POLLUTANTS

Abstract

Adsorption process employing activated carbon is commonly used for removing organic and inorganic pollutants from waters and wastes. Although activated carbon has very good adsorbing properties but the high cost involved in its manufacture and difficulties encountered in its regeneration makes it necessary to look for other economically feasible adsorbents. Efforts made in this direction and the results achieved therein are described in this dissertation. 0 A waste product obtained from National Fertilizers Limited Shatinda (India) has been activated and tried for the removal of phenols and metal ions. Since it does not possess significant adsorption of metal ions, other adsorbents viz., montmorillonite, kaolinite and hydrous oxides of aluminium (HAO) and iron (HFO) have been used for the uptake of these adsorbates. The adsorbates chosen for investigations are cadmium and lead, among the heavy metal ions and phenol, 2,4-dinitrophenol (DNP) and p-chlorophenol(PCP) as organic pollutants. In order to make the interpretation of results less complicated, experiments have been carried out with solutions of pure adsorbates while the work with live waste systems is also under progress in the same group. Section-A The waste material, obtained from National Fertilizers Limited Bhatinda (India) was converted in to activated carbon ii by heating at 450°C in the presence of air and used as adsorbent for the adsorption studies of phenol and its derivatives. The product was divided in two fractions (i) powder form, named as PAC and used for batch studies (ii) 250-300 B.S.S. mesh fraction, used in column operations is called as AC. Carbon, aluminium and iron contents in activated carbon sample are 92, 0.4 and 0.6^ respectively. The amount of silica and ash is found to be negligibly small. On the basis of Steenberg's classification, the product is classified as 'L* carbon. The surface area of PAC sample as determined by methylene blue adsorption isotherm is 532 m2/g. The concentration of phenol, p-chlorophenol (PCP) and 2,4-dinitrophenol (DNP) was determined spectrophotometrically at 510, 280 and 360 nm respectively. Kinetic studies of DNP on PAC show that the following rate law is valid for the system under investigation : ^S = Kt.tm where aS is the percent removal of DNP in time »t' (hours); 'm' depicts the adsorption mechanism and the term 'K.' may be taken as rate factor. It is observed that an increase in PAC amount (1.5 to 5.0 g/litre) at fixed DNP concentration (5xlO~4M) results in a decrease in 'm' values with corresponding increase in the constant 'Kt' while the trend of change in 'm' and 'K+' values is reversed with increasing amount of solute at constant PAC concentration. Further the percent removal of adsorbate is found to decrease with increasing DNP concentration. iii According to Weber et al. a value of m= 0.5 reflects •Intrap?rticle diffusion' as the rate determining step and smaller values indicate the involvement of film as well as intraparticle diffusion processes to almost an equal extent. A value of m= 4.2 x 1CT in these investigations indicates the uptake rate to be equally dependent on particle and film diffusion steps. The rate of adsorption of DNP is most rapid at PH<5.0 and constant in between pH 2.0 to 5.0. It decreases in the pH range 5.0 to 7.6 afid becomes constant after wards (Fig.5). The optimum amount of adsorbent and contact time are fixed as 2.5g/litre and 18 hours respectively for the three phenols for sorption studies. The adsorption isotherms of phenol and PCP are regular and concave but in case of DNP it is linear upto 1 x 10" M (Fig.9) concentration. Although the extent of adsorption increases with increasing concentration of adsorbates, the percent removal is quite high when the amount of solute is less. The removal of phenol, PCP and DNP at their lowest concentration is 70, 72 and 94^ respectively. Thus PAC seems to be quite suitable for the removal of phenol and its derivatives even if present in very low concentrations. The adsorption data of phenol and PCP can be fitted in Langmuir and Freundlich models. Q° value is higher for phenol as compared to PCP (expressed as mg/g) (Table 7). The higher value of coefficient 'b' for phenol indicates a greater tendency iv towards adsorption for this solute. Sorption of phenol and PCP does not change with pH while the uptake of DNP remains constant at pH 2.0 and 4.0 but decreases at pH 10.0. This may be attributed to the existence of DNP in anionic form at higher pH. The effect of dissolved salts on the sorption of phenol, PCP and DNP has also been observed. NaCl increases the adsorption of phenol, PCP and DNP by 30, 90 and 66/ respectively The presence of BaCl2 and AlClg do not change the uptake of PCP and DNP but decrease the adsorption of phenol by 5 and 8/ respectively. Anions with common cation (Na+ ion) have no effect on the uptake of three phenols. The enhancement in the uptake of DNP in the presence of NaCl may be attributed to ionpair formation while for phenol and PCP it is due to increase in the vapour phase partitioning which may lead to adsorption of non-ionized compounds on to solid surfaces. The adsorption of phenol has also been studied on oxidised and oxidised-reduced carbon surfaces. The oxidation of the carbon surface reduces its phenol adsorption capacity by 50/. fhe surface properties of the adsorbent are restored back by 91/ on subsequent reduction of the oxidised carbon surface. The time required for 80/ removal (cone. 180 mg/litre) of the three adsorbates is in the following order: Phenol > PCP > DNP These observations are explained by donor-acceptor mechanism as proposed by Mattson et al. The practical applicability of the product in column operations has also been investigated to obtain a factual design model. The shape of the break-through curve is of ' S' type and break-through capacity is more than batch capacity for the three adsorbates. This is due to the difference in the nature of continuous and batch operations. Phenol and its derivatives are desorbed from the loaded columns with 5/ NaOH (Fig.15). An almost complete desorption of both phenol and DNP could be achieved (96/) but only sixty percent elution of PCP could be possible with this eluting agent. The column could be regenerated with 1M HNO~. Finally the effect of pH, flow rate, concentration of DNP and bed-depth on mass transfer zone characteristics viz., HMTZ (Hei9ht of mass transfer zone) $mz (fractional capacity of mass transfor zone) and f^TZ (rate of movement of mass transfer zone) used to design activated carbon contactors designated as the bed-depth-service-time (BQST) method has been studied. With increasing bed-depth (from 2.1 to 10.0 cm) and maintaining a constant concentration as well as the flow rate of DNP, an increase in HMTZ is observed along with a decrease in ^TZ ^•^S'^TZ* however, remains constant. The rate of movement (^TZ) and the height (HMJZ) of the mass transfer zone increase while the value of fractional capacity of mass transfer zone ($MTZ) decreases with increasing flow rate of DNP at constant bed-depth and concentration. The decrease in 4),,^., is TMTZ vi only 7/ -hen the flow rate is increased to about ten times. Lastly with increasing concentration of DNP (at constant flow rate and bed-depth) a decrease in d),,^, is TMTZ observed alongwith an increase in HMTZ and I^.-- values. With the help of the above mentioned data the following relationship has been developed -0.77 V0.02 = 0aiCi and utilized to predict the break through volume (Vn nr>) for a given influent concentration (CA) of the adsorbate, under the specified experimental conditions (flow rate and beddepth) . Section-3 The adsorbents, metal oxides, have been prepared by the methods reported earlier and characterized by spectrophotometry and X-ray analysis. The IR spectra of HFO suggests that the product is a mixed phase of a and (3-FeOOH as revealed by peaks at 1340, 1490 (characteristic of 3-FeOOH) and 575, 450 cm""1 (characteristic of a-FeOOH). X-ray studies indicate the product to be amorphous in nature. The peaks at 1005, 3605, 3510, 3450 (characteristic of gibbsite) and 3300, 1080 cm"1 (characteristic of boehmite) in the IR spectra of HAD may be indentified with the presence of product in the form of gibbsite along with a small amount of boehmite. This is also confirmed by X-ray diffraction pattern of the prepared sample (Table 1). Vll Clay minerals (montmorillonite and kaolinite) were obtained from Ward's National Earth Science Est. New York. The purity was tested by DTA, IR and X-ray studies. Clays were also converted into homoionic form (H ) before use. Cadmium was estimated radiometrically and E.D.T.A. titration method was used for the estimation of lead. The -4 -"? adsorbates concentration, 4x10 to 9x10 M has been chosen on the basis of the metal ion levels likely to be present in waste water. Particles having 100-200 fraction of metal oxides and 250-280 B.S. S. mesh size of clay have been used. The uptake of lead and cadmium is almost complete in 24 hours. As such all adsorption measurements were made after equilibrating the metal ions for 48 hours with various adsorbents the amount of adsorbent being 0.1 g in all the sets prepared for running the isotherms. All the isotherms are concave to initial concentration axis and reveal a relatively rapid initial rate of adsorption with a slow approach to saturation condition (Figs. 2-5). The uptake of lead is maximum on HFO and minimum on kaolinite while cadmium adsorption is maximum on montmorillonite and minimum on HAD (Tables 3-6). HFO has better adsorbing capacity as compared to HAD. Out of the two clays, montmor illonite is a better adsorbent than kaolinite for both the metal ions. The uptake of the two metal ions, lead and cadmium increases with increasing pH. The investigations Vll Clay minerals (montmorillonite and kaolinite) were obtained from Ward's National Earth Science Est. New York. The purity was tested by DTA, IR and X-ray studies. Clays were also converted into homoionic form (H+) before use. Cadmium was estimated radiometrically and E.D.T.A. titration method was used for the estimation of lead. The adsorbates concentration, 4xl0~4 to 9xlO~3 Mhas been chosen on the basis of the metal ion levels likely to be present in waste water. Particles having 100-200 fraction of metal oxides and 250-280 B.S. S. mesh size of clay have been used. The uptake of lead and cadmium is almost complete in 24 hours. As such all adsorption measurements were made after equilibrating the metal ions for 48 hours with various adsorbents the amount of adsorbent being 0.1 g in all the sets prepared for running the isotherms. All the isotherms are concave to initial concentration axis and reveal a relatively rapid initial rate of adsorption with a slow approach to saturation condition (Figs. 2-5). The uptake of lead is maximum on HFO and minimum on kaolinite while cadmium adsorption is maximum on montmorillonite and minimum on HAD (Tables 3-6). HFO has better adsorbing capacity as compared to HAD. Out of the two clays, montmor illonite is a better adsorbent than kaolinite for both the metal ions. The uptake of the two metal ions, lead and cadmium increases with increasing pH. The investigations viii could not be extended in a wider pH range due to precipit ation tendency of the adsorbate ions at higher pH. In case of clays this increase in adsorption is possibly due to reduced competition with protons and increase in the negative adsorption sites as the positive double layer at edges of clay particles changes its polarity in alkaline medium to a negative double layer. Hydrous metal oxides also change their polarity from positive to negative with increasing pH. Adsorption data in the entire concentration range is neither represented by Langmuir nor Freundlich models except in case of montmorillonite and HFO. Adsorption data for these two adsorbents fits well in Langmuir model. The value of the coefficient 'b' is more for lead as compared to cadmium on the two adsorbents montmorillonite and HFO (Table 7) The displacement of two protons per bivalent metal ion in adsorption stoichiometric studies, indicate that metal ions are taken up on these adsorbents by ion-exchange mechanism. The uptake of lead and cadmium decreases with increasing concentration of cetyltrimethyl ammonium bromide (CTAB) except for kaolinite-cadmium system where the amount adsorbed does not change in presence of CTAB. Decrease in adsorption may be attributed to competitive adsorption for surface sites by adsorbate and detergent ions. In kaolinitecadmium system organic cations do not successfully compete against metal ion. ix The presence of ABS increases the adsorption of metal ions on clays and hydrous oxides. The increase in the uptake of lead and cadmium is maximum on kaolinite and minimum on montmorillonite. Out of the two adsorbates the enhancement in the uptake is more for cadmium on all the solids. The increase in the adsorption of metals on these adsorbents in the presence of ABS may be due to the formation of sparingly soluble metal-ABS species, either as a separate phase or on the surface of adsorbents. 2+ -4-0 Studies on the adsorption kinetics of Pb and Cd on montmorillonite and HFO reveal that the sorption of Pb 2+ and Cd is particle diffusion controlled at and above 6x10 M concentration on montmorillonite while in case of HFO it is particle diffusion controlled at 5xlO"3 and 6xl0"3M for Pb + and Cd2+ respectively (Figs. 12 and J3). The effective diffusion coefficient (D. ) values for the two metal ions are higher for HFO than montmorillonite (Table 13). Smaller D. value for clay may be due to the compact structure of mineral. The value for cadmium is more on montmorillonite and for lead it is higher on HFO. The activation energy (E ) is higher for cadmium as compared to lead and among the two adsorbents it follows the order : Montmorillonite > HFO The negative entropy ( As/) values indicate that as a result of exchange, no significant change occurs in the internal structure of the two adsorbents.