Elsevier

Water Research

Volume 34, Issue 11, 1 August 2000, Pages 2971-2978
Water Research

Mercury removal from water by ion exchange resins adsorption

https://doi.org/10.1016/S0043-1354(00)00044-0Get rights and content

Abstract

In this paper a detailed experimental and theoretical analysis of the adsorption process of mercury by ion exchange resins is presented. Experiments have been performed to study adsorption efficiency, the effect of pH on the uptake of mercury and the adsorption kinetics. The experimental apparatus was a batch mechanically stirred reactor (volume 1.5 dm3), under almost isothermal conditions (±0.1°C). The resin used in this study is Duolite GT-73, a chelating resin, macroreticular with thiol (S-H) functional groups. The studied resin has a very high adsorption efficiency, reaching 30–40% in weight and the efficiency decreases, decreasing pH, due to competition between ions H+ and Hg2+. Such a high efficiency confirms previous results and justifies the great interest for the application of ion exchange resins in water treatment plants. As in previous works, measurements of process kinetics show that the adsorption rate decreases as the initial mercury concentration is increased. This fact suggests that intraparticle diffusion rate can be the controlling step for the adsorption process. To verify this, a simplified mathematical model has been identified, accounting for a diffusional resistance inside solid particles and where the equilibrium relationship between Hg concentration in the liquid and in the solid is described by the Freundlich isotherm, neglecting H+ competition: this model is very effective in the prediction of the change in the adsorption kinetics with the initial Hg concentration. Hence this preliminary approach can be held as the reference starting point for the adsorption model: further developments will concern the equilibrium thermodynamics (H+ competition).

Introduction

Many authors, in the past, studied the best method, both from the economical and technological point of view, to remove mercury and heavy metals in general from superficial waters, with the aim of the preservation of the entire ecosystem from the damages due to their accumulation (see e.g. Humenick and Schoor, 1974, Logsdon and Symons, 1975, Knocke and Hemphill, 1981, Deshkar et al., 1990, Barkley, 1991, Ritter and Bibler, 1992, Mazidji et al., 1992, Zouboulis et al., 1993, Couillard and Mercier, 1994, Pazirandeh et al., 1998).

Mercury is commonly transferred from water by the following methods:

  • conventional coagulation;

  • chemical precipitation;

  • adsorption (both chemical and physical);

  • reverse osmosis.

Adsorption processes have been and actually are the most applied in the industries, and consequently the most studied. Logsdon and Symons (1975) studied the removal of mercury during the conventional water treatments, showing that mercury is adsorbed on clay particles during coagulation. Humenick and Schoor (1974) analysed the mercury removal from water through chemical and physical adsorption on activated carbons, obtaining good results. Zouboulis et al. (1993) studied removal of toxic metal ions from solutions using industrial solid by-products. Deshkar et al. (1990) used bark as the adsorbing material, observing good adsorption capacity with respect to mercury, also considering its low cost.

All the above mentioned materials present good adsorption capacity, but in general they are not able to reduce mercury concentration below discharge limits established in environment legislation. Discharge limits are, e.g. 10 ppb in USA (Ritter and Bibler, 1992), 5 ppb (Italian legislation), whereas for drinking water the limit is 1 ppb for the Italian legislation.

Such discharge requirements can be matched using specific ion exchange resins, characterised by a very high selectivity with respect to mercury: see e.g. Ritter and Bibler, 1992, Mazidji et al., 1992.

In particular, Ritter and Bibler (1992) studied the performance of Duolite GT-73 ion exchange resin in a large-scale process, discussing some operational problems encountered and their remedy. This work aims to perform an experimental and theoretical study of the removal of mercury from water by ion applying the same exchange resin as Ritter and Bibler (1992) process and to evaluate:

  • adsorption efficiency;

  • effect of pH on the uptake of mercury;

  • adsorption kinetics.

Furthermore a suitable mathematical model is identified, able to describe the main phenomena observed.

Section snippets

Experimental

All experimental analysis has been carried out in a batch mechanically stirred reactor (volume 1.5 dm3). The resin used in this study was Duolite GT-73, a polystyrene/divinylbenzene resin with thiol (S-H) functional groups, which has been already studied by other authors (e.g. see Ritter and Bibler, 1992). Before each experiment, the resin was dried by air, weighted and wetted by adding distilled, deionised water as described in the manual provided by Duolite supplier. Since the resin was

The model

As in previous works (Deshkar et al., 1990, Orhan and Büyükgüngör, 1993, Ho et al., 1995), experiments show, that sorption kinetics changes with the initial mercury concentration in the liquid. In such previous studies, the mathematical description of the system was addressed to the identification of a first order reaction rate constant, in order to fit the experimental data of the removal extent vs time. This kind of description was justified by the chelating properties of such resins, by the

Conclusion

In this work a detailed experimental and theoretical analysis of mercury sorption on ion exchange resins is presented. On the basis of the present study, the following conclusions can be drawn:

  • Sorption capacity of mercury on this resin is very high, with the possibility to obtain extremely purified water.

  • The degree of removal strongly dependent on the initial pH of the solution: it decreases as the pH increases. Furthermore pH decreases with time during adsorption. This is due to the exchange

References (16)

There are more references available in the full text version of this article.

Cited by (291)

View all citing articles on Scopus
View full text