Mécanisme de la précipitation de l'acide salicylique par coagulation par le fer ferrique

Abstract

The general objective of our work is to study the precipitation mechanism of organic matter by iron salts. The elimination of aromatic acids by coagulation with iron was found to depend on the number and the position of acid functions on the aromatic cycle. Only compounds which present at least two COOH and/or OH functions in ortho position were significantly removed by iron(III) coagulation (Lefebvre and Legube, 1993). The objective of the present paper is to precise the mechanisms of interaction between salicylic acid (and its isomers) and iron as well as to evaluate the part of complexation reactions. All experiments were carried out with a jar-test apparatus which was described in a previous publication (Lefebvre and Legube, 1990). The coagulation was carried out at 250 r.p.m. for 15 min, the slow agitation (floculation) was carried out at 30 r.p.m. for 30 min, 2 h were chosen for the sedimentation. After filtration on 0.2 μm-membrane, the following analyses were performed: dissolved organic carbon (DOC), analysis of aromatic compounds by HPLC, measurement of u.v.-visible absorbance, analysis of iron by orthophenanthroline-spectrophotometric method (dissolved iron) and by atomic absorption spectrometry (total iron). The main results presented in this paper are: 1. (i) The removal of salicylic acid by coagulation with iron was optimal for pH values of about 5.5 [Fig. 1(B)] where this removal increased with the increasing coagulant dose [Fig. 1(A)]. Furthermore, at pH 3.5 (Table 1), one could notice appearance of an absorbance band at 525 nm and a significant difference between results of DOC and HPLC analyses for weak molar ratios. At pH 5.5 (Table 2), the difference between both determination technics of salicylic acid (DOC and HPLC) was significant only for a salicylic acid/iron molar ratio above 0.8. This difference between the two analytical methods was mainly observed with the samples containing residual iron and which presented a 483 nm-band on the visible spectra. At pH 8.5 (Table 3) the precipitation of iron was predominant, which can explain the weak removal of the compound. 2. (ii) The relative position of hydroxyl and carboxyl functions on the aromatic ring of hydroxy-benzoic acid was found to be very important. Indeed, the removal of 3-hydroxybenzoic (meta position) and 4-hydroxybenzoic (para position) acids were found to be much lower than salicylic acid. Moreover, the organic matter/iron molar ratio for coagulation, at pH 5.5, of the two salicylic acid isomers had no effect either on the elimination of the compounds nor on the solubility of iron (Tables 5 and 6). We conclude from these results and some others (stoichiometry of the complex formation in Fig. 7, effect of pH on the complex stability in Fig. 8) that precipitation of salicylic acid by coagulation with iron is firstly the consequence of the initial formation of a soluble complex with a 1/1 stoichiometry. The formation of this soluble complex was particularly easy to observe at acidic pH and for high salicylic acid/iron molar ratios. The complex stability is due to the bidentate nature of the salicylate ligand, considering the fact that there was evidently no formation of such a complex from meta and para isomers of salicylic acid and only a weak precipitation of these compounds was observed. Secondly, increase of pH leads to hydrolysis of the complex by the partial replacement of salicylic acid with hydroxide ions. During the hydrolysis, the maximum of the absorption band of spectra was found to shift (483 nm at pH 5.5 instead of 525 nm at pH 3.5), then to disappear when pH values are nearly neutral. This hydrolysis leads only to a partial precipitation of the organic compounds, even for optimal pH values (around 5.5).