Toxic and refractory organic pollutants in industrial waste water can be degraded by advanced oxidation processes (AOPs) alone, or in combination with physico-chemical and biological processes. Of these oxidation methods, Fenton's reagent (Fe(II)/H2O2) and Fenton-like reagents (Fe(III)/H2O2, Mn+ or Mn+1/H2O2) are effective oxidants of large variety of organic pollutants. The mechanism of decomposition of H2O2 and of oxidation of organic solutes by Fenton's and Fenton-like reactions has been the subject of numerous studies. However, there are still many uncertainties as to the nature of the oxidant species formed and the rate constants of elementary reactions (table 1). Our recent studies carried out in HClO4/NaClO4 solutions and in the presence of very low concentrations of organic solutes (atrazine, 1,2,4-trichlorobenzene; concentration < 3 μM) have shown that the reaction of Fe(II) with H2O2 leads to the formation of two intermediates and that the overall initiation step (reaction 1, table 1) at pH < 3.5 leads to the formation of OH radical (Gallard et al., 1998a). Other work with different organic compounds and higher concentrations of organic solutes indicates that the intermediates (Fe(II)-hydroperoxy complexes, ferrous ion) might also oxidize organic compounds. Ferric ion can also catalyze the decomposition of H2O2. The mechanism is initiated by the formation of two Fe(III)-peroxy complexes at pH < 3.5 (reaction 2a, table 1) followed by their slow decomposition into Fe(II) and HO.2/O.-2 (reaction 2b, table 1) (Gallard et al, 1999; De Laat and Gallard, 1999; Gallard and De Laat, 2000). The formation of intermediates (complexes, cupryl ion) has also been postulated for the catalytic decomposition of H2O2 by Cu(II). Depending on the experimental conditions (nature and concentrations of organic solutes, pH...), the degradation of organic compounds might be attributed to the hydroxyl radical (reaction 1, table 1) or to other species like the cupryl ion (Cu(III)). Production of Cu(III) by reaction of OH. with Cu(II) has also been demonstrated by pulse radiolysis experiments. Kinetic data indicate that the rate of decomposition of H2O2 and the rate of oxidation of organic compounds are faster with Fe(III)/H2O2 than with Cu(II)/H2O2 and that Cu(II) can improve the efficiency of the Fe(III)/H2O2 process. The present study has been undertaken in order to compare the rates of decomposition of H2O2 and the rates of oxidation of atrazine by Fe(III)/H2O2, Cu(II)/H2O2 and Fe(III)/Cu(II)/H2O2 under identical conditions. These conditions (pH 3.0, I = 0.1 M, [Atrazine]o < 1 μM) were the same as those used in previous studies of the Fe(II)/H2O2 and Fe(III)/H2O2 systems. Experiments were carried out in MilliQ water, in the dark, at 25.0 (± 0.2)°C, pH 3.0, ionic strength (I) of 0.1 M, in the presence and in the absence of dissolved oxygen. pH and I were adjusted with perchloric acid and sodium perchlorate. The concentrations of hydrogen peroxide ([H2O2]o ≤ 10 mM) and of atrazine ([atrazine]o ≤ 1 μM) were determined iodometrically and by HPLC, respectively. In the absence of organic solutes, experimental results have shown that the rate of decomposition of H2O2 is faster with Fe(III) than with Cu(II) (figure 2). In agreement with previous data (De Laat and Gallard, 1999), the initial rate of decomposition of H2O2 by Fe(III) can be described by a pseudo first-order kinetic law with respect to H2O2, and dissolved oxygen (0-1 mM) has no effect on the rate of decomposition. For the Cu(II)/H2O2 system, our spectrophotometric data (figure 1) gave evidence that the decomposition of H2O2 by Cu(II) goes through the formation of an intermediate which might be a Cu(H)-hydroperoxy complex and which absorbs in the region 350-600 nm. Furthermore, the rate of decomposition of H2O2 by Cu(II) does not follow a first-order kinetic law and is affected by the concentration of dissolved oxygen (figures 2 et 3). As far as the oxidation of atrazine is concerned, a preliminary study of the oxidation of solutions containing atrazine, ,2,4 trichlorobenzene and 2,5 dichloronitrobenzene in very dilute aqueous solutions ([organic solutes]o < 3 μM) has been conducted at pH 3.0. Experimental results showed that the relative rates of decomposition of organic solutes by Fe(III)/H2O2, Fe(II)/H2O2 and Cu(II)/H2O2 were identical and could be described by the competitive kinetic expression (figure 4). These data suggest that the oxidation of the organic solutes by the three systems of oxidation tested can be attributed to a unique oxidant species, the hydroxyl radical, under our experimental conditions. The rate of oxidation of atrazine by Cu(II)/H2O2 was found to be much slower than by Fe(III)/H2O2 (figure 5), to be dependent on the concentrations of reactants ([Cu(II)]o, [H2O2J]o figure 6) and to decrease in the presence of dissolved oxygen (figure 7). These data confirm that the rate of decomposition of H2O2 by Cu(II), and as a consequence, the rate of production of OH radicals by Cu(II)/H2O2, are much slower than by Fe(III)/H2O2. In addition, a fraction of Cu(I) may be oxidized by dissolved oxygen and this reaction, which competes with the reaction of Cu(I) with H2O2, may also decrease the rate of formation of OH radical. For the Fe(III)/Cu(II)/H2O2 system, experimental data have demonstrated that the addition of Cu(II) increases the rate of decomposition of H2O2 (figure 8a) and atrazine (figure 8b) by Fe(III)/H2O2 and that these increases in reaction rates depend on the concentration of dissolved oxygen. This catalytic effect of Cu(II) has been attributed to a fast regeneration of Fe(II) (which is the major source of OH radical) by the reaction of Cu(I) with Fe(III). Since this reaction competes with oxidation of Cu(I) by O2 and H2O2, the catalytic properties of Fe(III) and Cu(II) mixtures will depend on the experimental conditions, such as the relative concentrations of reactants. In conclusion, this comparative study has confirmed that the rates of decomposition of H2O2 and atrazine, in dilute aqueous solution, by Fe(III)/Cu(II)/H2O2 are faster than by Fe(III)/H2O2 and Cu(II)/H2O2. This study has also demonstrated that dissolved oxygen has a significant effect on the reaction rates in the Cu(II)/H2O2 and Fe(III)/Cu(II)/H2O2.oxidation systems. The effects of dissolved oxygen and of the addition of Cu(II) on the efficiency of the Fe(III)/H2O2 system could be explained by assuming that the OH radical is the major oxidant species under our experimental conditions. However, additional research is needed in order to better understand the mechanism of decomposition of H2O2 by Cu(II) and Cu(I) and to determine the rate constants of individual reactions involved in the Cu(II)/H2O2 and Cu(I)/H2O2 systems.