Study of the Wastewater Effluent Decontamination of Mercury Production Unit of Azzaba (Algeria)

Study of the Wastewater Effluent Decontamination of Mercury Production Unit of Azzaba (Algeria) by the Adsorption Process.

CHAFIA BOUCHELTA, MOHAMED SALAH MEDJRAM, LAKHDAR TIFOUTI

Laboratory LARMACS, Faculty of Sciences and Engineer Sciences

University of Skikda

Road of El Hadaeik P.O. 26 Skikda 21000

ALGERIA

Abstract: - The mercury production unit of Azzaba, in north-east of Algeria, constitutes one of the principal anthropogenic sources of pollution by mercury in Algeria. The Second worldwide producer of mercury, the mercury complex of Azzaba, exploits a cinnabar ore (sulphide of mercury). This last is decomposed at high temperature into oxides of sulphur and mercury metal vapors in revolving kilns supplied with crushed ore. These vapors are collected in vases after cooling in the form of mercury metallic. From the chimney, a general pollution of the unit and the suburban area is induced by a mercury depot on the ground, water and the vegetation. Slags, waste collected from furnaces, are stored by trucks in free air zones; they constitute a second pollution source. Finally, the settlings tanks which drain the cleaning waters of the ground and rainwater constitute the third source of environment pollution.

Wastewater analysis of mercury production unit, reveal residual mercury contents, in the form of ionic mercury, relatively significant. The obtained concentrations are often 400 times higher than according to the Algerian norm (0,01 mg l-1).

The aim of this work is to evaluate the depollution capacity of a granular activated carbon (GAC) in relation to the westewater discharges of this unit, highly contaminated by mercury and, in corollary, to study some parameters effect in order to determine the best separation

conditions of this pollutant.

Key-words: - pollution, cinnabar, mercury, westewater, toxicity, granular activated carbon, adsorption.

1. Introduction

Heavy metals are one of the most important pollutants in surface and groundwater [1-2]. Mercury(II) [Hg(II)] enters the environment from six major sources : chemical industry; mining and smelting; production and consumption of goods containing Hg(II); fuel burning; final disposal of wastestreams; and natural sources [3-4]. In the waterways, the mercury settles in the sediments and it is transformed into methyl-mercury particularly toxic. Food pollution by mercury causes serious intoxications (foetal malformations, neural troubles which can causes death). The fish, polluted by discharges in the sea, can constitute a potentially significant source of mercury. In all cases, Hg(II) may attain environmentally significant concentrations and pose serious health hazards [5-6]. In the last few years, adsorption has been shown as an alternative method for reducing the heavy metals content in wastewater and water supplies [7]. Adsorption phenomena onto the aquifers’solid matrix also contribute to mitigation of groubwater pollution [8]. Both in the design of adsorption treatment units and in the prediction of reactive contaminants profiles in aquifers, the accuracy in the characterization and modelling of these phenomena plays an important role. The adsorption of heavy metals has been mainly studied on metal oxides while activated carbon has been extensively used for the removal of organic compounds from wastewater. The utility of activated carbon as an effective adsorbant for heavy metals is still being explored, and relatively few data are available [9-11].

The adsorption of inorganic compounds onto solides hydrous is an electrostatic process where specific interactions between the ionic species and the surface of the solid are established . The surface complex-formation

models (SCF) are chemical equilibrium models developed to predict the adsorption of heavy metals by metal oxides [12-15] and later extended to activated carbon [16-18]. The main objective of this present investigation were to(i) determine, the effectiveness degree of the adsorption process onto a commercial

granular activated carbon, in relation to the depollution of a water highly contaminated by mercury, and (ii) to study The effect of important factors such as pH, temperature,

stirring velocity, ionic force, metal concentration (surface loading), onto the process.

2. Materials and methods

2.1. Apparatus

Batch mode adsorption experiments were carried out using 1000 ml (fluid volume) discontinuous reactor perfectly agitated, closed and thermostated as shown in figure 1. The reactor is a glass cylinder of 25 cm height and internal diameter 15 cm. The agitation is carried out using a mechanical agitator. The sliding part used is a stem with four blades inclined with 45 degrees, 5 cm in diameter. The temperature is recorded in various phases. The reactor is plunged in a thermostated bath.

Agitator

Mechanical stem

Thermometer

Reactor

Figure 1: Scheme of the experiment set-up for adsorption in batch mode.

2.2. Analysis of mercury

Hg(II) was proportioned by the atomic absorption spectrometer (M/S Hitachi, Japan, model No. 180-80). The wavelength resolution was 253,7 nm 19. Dilution of the samples prior to analysis was necessary.

2.3. Reagents and activated carbone

A stock solution of 1000 mg l-1 was prepared by dissolving HgCl2 with Purity greater than 99,5 % in distilled water. A wastewater sample was taken from the unit mercury production of Azzaba. The GAC (Norit) has a BET surface of 700 m2 g-1, a mean granulometry of 2 mm, a density of 375 kg m-1 and pH 9 [20]. The activated carbon was washed and boiling in water for 1h. Lastly, it was dried in an oven at 110 °C to constant weight before use.

2.4. Procedure of mercury elimination

A typical batch mode experiment was carried out as follows: 1000 ml of 100, 200, 400 and 800 mg l-1 Hg(II) at an initial pH 2 and 1g of adsorbent (GAC) were taken in the reactor (fig.1) and agitated at 200 rpm and 18°C.1 ml of the supernatant solution are carried out with intervals of time in order to measure the reduction in the concentration of the solution in metal in the course of time until having a constant metal concentration. The concentration of wastewater sample was also determined under the same practical conditions. The analyses of the supernatant solution were achieved as previously described 19. In order to determine the isotherm of Hg (II) on the GAC, 100 ml of mercury aqueous solutions, at different concentrations (100 to 1000 mg l-1), and a fixed concentration of adsorbent (1g), were continuously stirred at 18°C. The stirring rate was 200 rpm. Equilibrium was reached after a contact time of 5 minutes (determined during preliminary evaluations). The amount of Hg(II) adsorbed q= x/m (mg g-1) can be obtained as follows:

q = (C0 – Ce) V/m (1)

where C0 and Ce are the initial and equilibrium solution concentrations (mg l-1), V is the volume of solution (l), and m is the amount of GAC used (g). The isotherm of metal adsorption considered is obtained by tracing q vs Ce 21. pH effect was studied by adjusting mercury solution pH to different values between 1,5 and 7,0 with dilute NaOH and HCl solutions. The temperature effect on Hg(II) elimination onto GAC was also studied. Hg(II) concentrations in the solution used were from 100 to 1000 mg l-1, pH 2 and 1g of GAC.

Studies were also conducted to examine the change in the adsorption behavior of Hg(II) onto activated carbon. The increase in the ionic force was done by the addition of NaCl solutions to concentrations ranging between 0,1 and 10 g l-1. For the determination of the effect stirring speed on the process, we varies this one between 200 and 1200 rpm. All these tests were carried out as follows: after agitation of mixtures (100 ml of mercury solution at 200 mg l-1 with 0,5 g of GAC) during 5 minutes, Samples of 1ml were withdrawn for analysis using the atomic absorption spectrometer to determine Hg(II) concentration remaining in the fluid phase, and therefore the amounts of Hg(II) adsorbed on the GAC.

For the kinetic studies, 1g of activated carbon was continuously stirred with 1l of aqueous Hg(II) solution (800 mg l-1) at 18°C. The analyses of the supernatant solution were achieved as previously described.

3. Results and discussion

3.1. Effect of shaking time

Figure 2 shows the variation in the percentage adsorption of Hg(II) onto activated carbon with shaking time, using distilled water as the sorptive medium. This figure indicates that whereas Hg(II) adsorption is quite rapid, equilibrium time reached after 1, 2, 3 and 5 minutes for the solution of 100, 200, 400 and 800 mg l-1 respectively.

Figure 2 : Influence of shaking time on the adsorption of Hg(II) ions from activated carbon.

The adsorption speed may be due to the rapid diffusion of the Hg(II) ions into the pores of the GAC. We observed that the adsorption percentage of Hg(II) on to GAC is more important at low concentration, it reaches the 88%, 90% and 93 % whit 400,200 and 100 mg l-1 respectively. The adsorption decreased with increasing Hg(II) concentration, indicating that less favorable sites become involved in adsorption with increasing concentration 22.

wastewater sample analysis reveal that mercury concentration is 4 mg l-1.

3.2. Effect of temperature

The temperature used was varied from 18 to 38 °C. The Comparison between the isotherms Hg(II) on the activated carbon obtained at different temperature (figure 3) reveal that the amount adsorbed increased with increase of the temperature from 23,36, 33 to 50,96 mg g-1 for the temperature of 18, 28 and 38 °C respectively.

Figure 3: Effect of temperature on the adsorption of Hg(II) ions from activated carbon.

The temperature has an appreciably effect on the process of mercury ion adsorption on to activated carbon. Indeed, at 38°C, the elimination rate reaches the 100 % for the Hg(II) concentrations of 100 and 200 mg l-1 (figures 3).

3.3. Effect of pH

An attempt was then made to find the optimal pH range from mercury solution that gives a maximum adsorption capacity. The increase in pH supports the disappearance of metal as shown in figure 4. This disappearance is probably due to the metal precipitation starting from pH 2,4 24 and it is traditionally explained by the changes in distribution of the species with the pH 25-26. Therefore it would be preferable to acidify the medium below the pH of metal precipitation to have a best separation.

Figure 4: Effect of pH on the adsorption of Hg(II) ions from activated carbon.

3.4. Effect of the ionic force

The addition of NaCl (Na+, Cl-), in the mercury solution, affected the adsorption of Hg(II) on GAC. The increase in the ionic force causes a reduction in the adsorption capacity. However, this effect begin only from 0,1 g l-1 in NaCl, as shown in figure 5. This decreasing is due to the competition between Hg2+ and Na+1 to occupy the GAC pores, i.e. the increasing in Na+1 concentration constrain the mass transport in the pores.

Figure 5: Effect of ionic force on the adsorption of Hg(II) ions from activated carbon.

3.5. Effect of the speed velocity

Figure 6 presents the effect of speed velocity on the removal of Hg(II). The removal (mg g-1) increased meanly with the increases of speed velocity until 500 rpm. Speed velocity has not a notable effect on the elimination of Hg(II), because of kinetics rapidity of the process wich obstruct the mercury diffusion.

Figure 6: Effect of the speed velocity on the adsorption of Hg(II) ions from activated carbon.

4. Adsorption kinetics.

The adsorption kinetics can be described by the Adam-Thomas relation

(dq / dt) = k1C(qm -q) – k2q (2)

where q is the adsorption capacity (mg g-1), C is the solution concentration (mg l-1), k1 is the adsorption kinetic constant (l mg-1 s-1), k2 is the desorption kinetic constant (s-1), qm is the maximal surface concentration (mg g-1) and t the time (s).

At the initial stage of the adsorption reaction, when t  0, then q  0 and C  C0. Then, equation (2) could be rewritten:

(dq / dt) t0 = k1C0 qm=(v/m)d(C-C0)/dt t0 (3)

where C0 is the initial concentration (mg l-1), v the volume of solution (l) and m the weight of activated carbon (g). It is then possible to calculate the initial adsorption kinetic coefficient γ 23  :

(4)

The value of γ (l mg-1 min-1), calculated taking into account the initial slopes of the kinetic curves, given in Table 1. (dC/dt) t  0 is the initial adsorption rate. In general, adsorption may be described as a series of steps: mass transfer from the fluid to the particle surface, diffusion within the porous particle and adorption itself onto the surface. In order to have more information concerning this process we have calculate many coefficient and constant.

The Morris-Weber equation:

q = Kd (t) 0.5 (5)

was tested with respect to the experimental data depicted in figure 2. Figure 7 indicate that two distinct regions were observed: an initial linear portion which is due to the boundary layer diffusion effect 27 and a second linear portion which is due to the intraparticle diffusion . The values of the intraparticule diffusion coefficient is tabulated in Table 1.

Figure 7: The amount of Hg(II) adsorbed on to activated carbon as a function of time.

The Lagergren equation 23 was also employed for studying the rate constant for the mercury ion adsorption process in the activated carbon in the form:

Ln[(qe – qt )/qe] = - Kt (6)

Where qe is the adsorption capacity at equilibrium (mg g-1), K is the first-order rate constant of adsorption (min –1) , t (min) and qt are the time and the adsorption capacity (mg g-1). The linear plot of Ln[(qe – qt )/qe] versus t obtained shows the appropriateness of the above equation and concequently the first-order nature of process involved. The value of the rate constant was calculated and given in Table 1. Finally, D (cm2 s-1) is the extraparticle diffusion coefficient 23.

In addition to the very short equilibrium time of adsorption (1 to 5 min.), the initial speed of elimination of Hg 2+ shows that the process has very fast initial kinetics; this rapidity can be explained by the great surface of the adsorbent and the great porosity of the material which helps the diffusion of the metal ions inside the surface of the grains, which supports the depollution of metal in question.

Table 1: Results of the kinetic study.

Kinetic parameters

5. Conclusion

The possibility of depollution of water contaminated by Hg 2+ by using the process of adsorption on GAC were studied.The different test proved the effectiveness of this process. In effect, the residual concentration of Hg 2+ in the water initially polluted by 100 and 200 mg l-1 reaches traces at 38 °C. Mercury elimination increases with pH. This rise is probably due to the appearance of the phenomenon of metal precipitation in hydroxides form: Hg(OH)2 wich starting from pH 2,4. On the other hand, the increase of ionic force reduces the capacity of adsorption,because there is a competition between Hg+2 and Na+. Mercury removal is not affected by speed velocity.

Study of adsorption kinetics revealed that this process supports the elimination of mercury. It emerges that the process of adsorption on GAC studied can be efficiently used for the depollution of the wastewater of mercury production unit of Azzaba, wich is contaminated by 4 mg l-1.

References

1 Edline F, Elimination of Heavy Metals in waste Water, Platform of Water, Studies and Memories N° 565/5, October seven (1993)p. 7.

[2] C Tiffreau, Sorption of mercury (II) with the interface water/solid, experimental study and modeling, thesis of doctorate of the university Louis Pasteur, Strasbourg, June 1996.

[3] Waston, Jr., W.D., Economic considerations in controlling mercury pollution In J.O.Nriagu (ed). The biogeochimistry of mercury in the environment, Elsevier/North-Holland Biomedical Press 1979,pp.231-276.

[4] Barb. N. R. Proust, the mercury and its compounds about speciation to toxicity, chemical topicality, 1998, pp. 16.

[5] Mr. Gerard Miquel, Report/ratio on the effects of heavy metals on the environment and health, SUBDUES Principal industrial wastes in France - Assessment of the year 1998, February 2000.

[6] Jean Luc Potelon, The guide of the analysis of drinking water, ed. bets, 1998, pp 67-143-149.

[7] Matsumoto M. R., Weber A. S. and Kyles J. H., Use of metal adsorbing compounds (MAC) to mitigate adverse effects of heavy metals in biological unit processes, Eng. Chem. Comm. N° 86(1), 1989, pp 1-16.

[8] Weber, W.J. Jr., McGilney, P. M. and Katz, L.E., Sorption phenomena in subsurface systems; concepts models and effects on contaminant fate and tran., Water Res. N° 25, 1991, pp 499-528.

[9] Aktar, S., R. Kadeer Active carbon as an adsorbent for lead ions, Science and technology, volume 15, N° 10 1997, pp. 815-825.

[10] A. Hamdpour, A. Bradley, Comparison of équilibria and cinetics of high surface area activated carbon produced from different precursors and by different chemical treatments, Ind. Eng. Chem. LMBO, volume 37, 1998, pp 1329- 1334.

[11] A. Christian, E Ligwe, B Nereus, Adsorption thermodynamics and kinetics of Mercury (II), Cadmium (II) and lead (II) one lignite, chem. Ing. Technol. N° 22, 1999, pp.22.

[12] Davis J. A. and Leckie J. O., Surface ionization and complexations at the oxide/water interface: II. Surface properties of amorphous iron oxyhidroxide and adsorption of metal ions., J.Colloid Interface Sci., N°67(1), 1978, pp 90-97.

[13] Kent D. B., Tripathi V. S., Ball N. B. And Leckie J. O., Surface-complexation modeling of radio nuclide adsroption in sub-surface environments, Tech. Report. No. 294 Department of civil Engineering. Standford. University 1986.

[14] Hayes K. F. and Leckie J. O., Modeling ionic strength effects on cation adsorption at hydrous oxyde/solution interfaces, J. Colloid Interface Sci. N° 115, 1987, pp 564-572.

[15] Dzombak D. A. And Morel F. M. M., Surface Complexation Modeling. Hydrous Ferric Oxide, John Wiley and Sons Inc., New York 1990.

[16] Corapcioglu M. O.and Huang C. P., The adsorption of heavy metals onto hydrous activated carbon, Water Res. N° 9(9), 1987, pp1031-1044.

[17] Reed B. E., and Novavinakere S. K., Heavy Metal adsorption by activated carbon, effect of complexing ligends, comp. Adsorbates- ionic strength, Sci. and Tech. N° 27(14), 1992, pp 1985-2000.

[18] Reed B. E. and Matsumoto M. R., Modeling cadmium adsorption in single and binary adsorbent (PAC) systems, J. Envir. Eng. Div. ASCE N° 119(2), 1993, pp 332-348.

19 Skoog. West. Uller, analytical Chemistry, translation and scientific revision of the 7éme edition American by Claudine Buess – Herman, J Sette Dauchot Weyneers And Freddy, 1997, pp 631.

[20] Card-index technical " NORIT ", purification of gases and liquids with granulated activated carbon ' NORIT'.

21 C. E. Chitour, Chemistry of Surfaces, introd. at the catalysis, 2 3rd ed., Algiers, 1981.

[22] G. S.Gupta and S. P. Shukla, Adsorp. Sci. Technol., 1996, pp 13-15

23 F Kaouh, H. ait Amar, Adsorption on activated carbon and mineral clays of some compounds organics present in water, TSM – water, 1998.

24 B Yavorski, A. Detlaf, memory assistance of physics, 5 3rd edition, MIR, Moscow, 1986.

25 F Chinwu, R. Tseng, R Juang. Role of pH in metal adsorption from aqueous solution containing chelating agents on chitosan, ind. Eng. Chem. Res. N° 38, 1999, pp 270-275.

26 E. Guibal, I Saucedo, J. Roussy, P. Lecloirec, Uptake of uranyl ions by new sorbing polymers: discussion of adsorption isotherme and pH effect, React. Polym, N° 23, 1994, pp147-156.

27 J. Crank, The Mathematics of Diffusion, Clarendon Press, Oxford, UK., 1965.