Perchloric acid (Panreac) and sodium chloride (Problabo) were used as supporting electrolyte and prepared with distilled water. Hydrogen peroxide 35% (Sharlau) was determined by potassium permanganate (Panreac) for the monitoring of hydroxyl radicals at Scharlau. Sodium iodide (99.5%) and hexane (95% - 97%) were manufactured by Sigma-Aldrich and Ensure respectively. Metronidazole, manufactured by Exphar SA of Belgium, was purchased from a pharmacy in Abidjan in tablet form. The pH was adjusted using H₂SO₄ (Sigma-Aldrich) and NaOH (Panreac). The sodium bromide was manufactured by Panreac.

A volume of 250 mL under magnetic stirring and at laboratory temperature, covered with aluminum foil, was introduced into an electrochemical cell. Boron-doped diamond electrode was used at the anode and zirconium at the cathode. Both electrodes have a surface area of 16 cm². Concentrations of 0.1 M HClO₄, and NaCl were electrolysed individually for the in-situ determination of hydroxyl radicals and chlorine. Subsequently, preparative electrolysis of 0.1 M HClO₄ + 0.1 g/L metronidazole (MNZ) was performed for COD monitoring as well as oxidant detection. The experimental setup for the preparative electrolysis is shown in Figure 1 .

For the detection of hydroxyl radicals, several detection methods, both direct and indirect, can be used. In our work, an indirect and simple method for the detection of ⁻OH was considered: the volumetric determination of hydrogen peroxide by potassium permanganate (KMnO₄). For this purpose, 10 mL of the samples, taken at well-defined times, were dosed with 0.1 M KMnO₄ in the presence of 10 mL 0.1 M H₂SO₄. The turn is reached when the initially colourless solution turns pink. Furthermore, a calibration curve for H₂O₂ concentrations ranging from 0 to 1 mM as a function of the volume of KMnO₄ poured in was established before the preparative electrolysis. The curve obtained is a straight line of equation [H₂O₂] = 0.0473V_KMnO4 − 0.0386, with a correlation coefficient R² of 0.9961 which is close to 1 attesting to the good linearity of the method.

The aim of this work is also to determine the presence of chlorine during the electrolysis of a sodium chloride solution (0.1 M) in order to identify chlorine in an unknown medium. The experiment was carried out at an acidic pH because according to the probaix diagram for chlorine, chlorine exists in the following pH range [0, 4.5].

The initial pH of the sodium chloride solution is 6.833. A 1 M sulphuric acid solution was used to adjust the pH to 2. The electrolysis of the sodium chloride solution is very suffocating due to the release of chlorine which is toxic. Thus, the detection of chlorine was done by taking a certain volume of NaCl samples to which 1 mL of NaI (0.2 M) and 2 to 5 mL of the hexane solution were added. After strong agitation of the mixture, the appearance of a pink coloration of the hexane proves the presence of active chlorine (Cl₂). The same test was performed during the preparative electrolysis of NaI and NaBr.

The COD of the samples was determined in our experiments using HACH COD tubes. To determine it, 2 mL of sample is taken and put into a COD tube and heated in a digester (HACH) at 150˚C for 120 minutes. After cooling, the COD value is read directly with the DR/6000 spectrophotometer (HACH). The COD reduction rate is determined using this formula:

$\Delta \left(\text{COD}\right)=\left(\left({\text{COD}}_0-{\text{COD}}_{\text{t}}\right)/{\text{COD}}_0\right)*100$ (4)

where COD₀ and COD_t are respectively the COD at time 0 and t in mgO₂/L.

In order to demonstrate the in-situ production of hydroxyl radicals, the electrolysis of perchloric acid was carried out under constant stirring for 180 minutes at laboratory temperature (25˚C) and a density of 10 mA/cm². 10 mL of HClO₄ samples taken at defined time intervals t were determined by 10⁻⁴ M KMnO₄ during the electrolysis. The concentration of H₂O₂ at time t is determined from the relation:

${\text{N}}_{{\text{H}}_2{\text{O}}_2}{\text{V}}_{{\text{H}}_2{\text{O}}_2}={\text{N}}_{{\text{KMnO}}_4}{\text{V}}_{{\text{KMnO}}_4}$ (5)

where N is normality and V is volume.

Figure 2 shows the evolution of the concentration of hydroxyl radicals generated in-situ during the electrolysis of 0.1M HClO₄. In Figure 2 , the concentration of hydrogen peroxide increases during HClO₄ electrolysis from 0 mM at t = 0 min to 1.5 mM after 180 min. The electrolysis of HClO₄ thus leads to the production of hydrogen peroxide and thus of hydroxyl radicals.

Figure 3 shows the absorbances of 0.5 M, 1 mM hydrogen peroxide from a 35% commercial hydrogen peroxide solution and a HClO₄ sample after 1 hour of electrolysis. A perfect superposition of the absorption curves in the wavelength range from 190 nm to 1100 nm is shown. This shows that the electrolysis of HClO₄ leads to the formation of hydrogen peroxide and therefore hydroxyl radicals.

Indeed, during the electrolysis of the supporting electrolyte, water is oxidised. The oxidation of H₂O leads to the formation of hydroxyl radicals (Equation (6)). In the absence of an organic compound, the hydroxyl radicals can only react with each other and form hydrogen peroxide (Equation (7)).

$2{\text{H}}_2\text{O}\to \cdot \text{OH}+{\text{OH}}^{-}+2{\text{H}}^{+}+{\text{e}}^{-}$ (6)

$\cdot \text{OH}+\cdot \text{OH}\to {\text{H}}_2{\text{O}}_2$ (7)

The amount of hydroxyl radicals generated can be influenced by various parameters including the applied current density. The influence of the applied current (current density 3.125, 5 and 10 mA/cm² respectively) on the production of hydroxyl radicals was investigated by monitoring the concentration of hydrogen peroxide during the electrolysis of 0.1 M HClO₄. The results obtained are presented in Figure 4 .

It can be seen from this figure that the production of hydroxyl radicals becomes important with the applied current. For an applied current, the production of hydrogen peroxide, and therefore of hydroxyl radicals, increases with the duration of the electrolysis. Indeed, Gnamba et al showed that the rate of degradation of amoxicillin on the boron-doped diamond electrode (BDD) increased with the imposed current density. This is related to the hydroxyl radicals formed, the amount of which depends on the imposed current.

The generation of hydroxyl radicals on the BDD occurs as follows:

$\text{BDD}+{\text{H}}_2\text{O}\to \text{BDD}\left(\cdot \text{OH}\right)+{\text{H}}^{+}+{\text{e}}^{-}$ (8)

Electrolysis of several concentrations (0.1 M, 0.2 M, 0.4 M) of perchloric acid was performed on the BDD at a current density of 5 mA/cm² for 180 min. Figure 5 shows the hydrogen peroxide concentration versus time curves. In Figure 5 , it can be seen that the concentration of H₂O₂, initially zero for all investigated concentrations, increases to 0.723, 1.10 and 1.915 mM after 180 min of electrolysis for 0.1, 0.2 and 0.4 M 0.1 M HClO₄ respectively. This shows that the initial concentration of HClO₄ plays an important role in the production of hydroxyl radicals.

Chlorine detection was investigated by degradation of a sodium chloride solution (0.1 M) under a current density of 10 mA/cm². In order to promote chlorine generation, the initial pH (6.833) was reduced and maintained at 2 using 1 M HClO₄. The electrode used at the anode for NaCl electrolysis was boron-doped diamond (BDD). After 15 minutes of electrolysis, a suffocating and very unpleasant odour characteristic of chlorine release was noted. After 30 min of electrolysis, NaCl samples were taken and tested for Cl₂. NaI and hexane were added to the NaCl electrolysis samples. The photo in Figure 6 shows the photo of the NaCl sample after 30 minutes of electrolysis (tube A), the tube containing the NaCl (30 minutes) + NaI sample (tube B). The third tube C contains a mixture of NaI + NaCl (30 minutes) + 1 mL of hexane and tube D, NaI+ NaCl (30 minutes) + 2 - 5 mL of hexane.

According to this picture, the addition of NaI to the NaCl sample gives a yellow coloration indicating the presence of an oxidant. The addition of hexane (2 - 5 mL) to the NaCl + NaI mixture gives a pinkish-purple colouration. This would indicate the presence of the oxidant Cl₂. It should be noted that a very suffocating odour is given off during the electrolysis of NaCl. To confirm that chlorine is responsible for the pinkish-purple colouration observed, the same test was carried out during the electrolysis of some solutions containing halogens (NaI and NaBr).

Electrolysis of NaBr concentration (0.1 M) under a current density of 10 mA/cm² was performed. The electrode used was boron-doped diamond (BDD). During this electrolysis, a 5 mL sample was taken after 30 minutes of electrolysis. In order to highlight the purplish-pink coloration of the hexane, NaI (0.2 M) and hexane were added to the samples of the NaBr electrolysis. The result obtained is shown in Figure 7 .

The colour of the NaBr electrolyte solution is pale green ( Figure 7(a) ). The addition of NaI (0.2 M) ( Figure 7(b) ) to this solution transforms this green colour (bromine water colour) into yellow, indicating the presence of an oxidant, in this case Br₂. The addition of hexane to the mixture (NaBr + NaI) ( Figure 7(c) ) leads to a very pale pink colouration on the surface of the mixture ( Figure 7(c) ). This could be characteristic of the presence of a weak oxidant (Br₂) in the reaction medium.

The electrolysis of NaI concentration (0.1 M) was carried out under a current density of 10 mA/cm². The electrode used was boron-doped diamond (BDD). A sample of the NaI solution was taken after 30 minutes of electrolysis. To highlight the purplish-pink coloration of the hexane; NaI (0.2 M) and hexane were added to the sample taken from the NaI electrolysis. The result obtained is shown in Figure 8 .

The coloration of the NaI solution after 30 minutes of electrolysis is orange-yellow ( Figure 8(a) ) characteristic of the presence of I₂. This colouration is maintained with the addition of NaI ( Figure 8(b) ), attesting to the presence of an oxidant, I₂. The addition of hexane leads to a very pale pink coloration on the surface of the mixture ( Figure 8(c) ). This could be characteristic of the presence of a weak oxidant (I₂) in the reaction medium.

The chlorine detection test was carried out on the electrolytic solutions of NaCl, NaBr and NaI. From these investigations, it appears that the test is positive with all these solutions indicating the formation of oxidants during preparative electrolysis. However, the vivid purplish-pink colouration obtained in the NaCl case, highlighting the electrogeneration of chlorine, in contrast to the very pale pink colouration obtained in the case of NaBr and NaI would show a selectivity of this method. This technique can therefore be applied to the detection of chlorine.

The electrolysis of 0.1 M HClO₄ was performed on the DDB electrode at a current density of 10 mA/cm².

In order to demonstrate the presence of chlorine during the electrolysis of perchloric acid, 10 mL of the reaction mixture was taken and 1 mL of NaI (0.2 M) and 2 mL of hexane were added. The mixture was stirred vigorously and then left to stand to allow decantation. The photo in Figure 9 shows the mixture of the electrolyte solution, NaI and hexane.

In Figure 9(a) , mixing the 0.1 M HClO₄ and 0.2 M NaI electrolyte solution gives a yellow coloration. The addition of hexane leads to a purplish-pink coloration on the surface of the mixture ( Figure 9(b) ). This is characteristic of the presence of Cl₂ in the mixture. Indeed, the initial pH of the reaction medium was 0.865 and was maintained below 2.5 throughout the electrolysis. This confirms the formation of Cl₂ during the electrolysis of 0.1 M HClO₄. Compared to NaCl electrolysis, 0.1 M HClO₄ electrolysis does not cause a suffocating odour to be released.

The formation of chlorine can be summarised as follows (Equation (9)):

$4{\text{HClO}}_4\to 2{\text{Cl}}_2+7{\text{O}}_2+2{\text{H}}_2\text{O}$ (9)

In-situ detection of hydroxyl radicals and chlorine was investigated during electrolysis of HClO4 (0.1 M) containing 0.1 g/L metronidazole (MNZ) under a current density of 10 mA/cm².

To detect the hydroxyl radicals, 10 mL of the sample mixture (KClO₄ + MNZ) was assayed at different times with KMnO₄ of 10⁻⁴ M concentration. The results are shown in Figure 10 .

According to this figure, the concentration of hydrogen peroxide increases from 0 to 0.13 mM after 120 minutes of electrolysis and then remains almost constant until 150 minutes (inset in Figure 10 ). This shows that hydroxyl radicals are formed during the degradation of MNZ on boron-doped diamond. However, the amount of hydroxyl radicals detected during the degradation of MNZ is much lower than that detected during the electrolysis of HClO₄.

For example, for 60 minutes of electrolysis, the amount of radicals detected in HClO₄ alone is almost 9 times that detected in the case (MNZ + HClO₄). This difference could be explained by the fact that the hydroxyl radicals generated during the electrolysis of MNZ participate significantly in the degradation of the organic compound (metronidazole).

The degradation of MNZ (0.1 g/L) in a 0.1 M HClO₄ solution on the DDB at a density of 10 mA/cm² was monitored by measuring the COD. The results obtained are shown in Figure 11 .

The COD evolution curve obtained decreases exponentially. The decrease in COD during electrolysis implies that the MNZ is degraded on the BDD below 10 mA/cm². This degradation leads to a COD removal rate of 83.48% after 180 minutes of electrolysis.

The limiting current density, determined by Equation (6), for the degradation of MNZ is 0.473 mA/cm², which is lower than 10 mA/cm². The degradation of MNZ was essentially controlled by mass transfer, i.e. limited by diffusion. This can explain the exponential curve of the COD.

J_lim = 4Fk_dCOD (10)

where F is faraday constant (96458 C/mol); k_d is mass transfer coefficient (cm/s) and COD in mol O₂/L.

In Figure 12 , mixing the solution of electrolysed MNZ in 0.1 M HClO₄ and 0.2 M NaI gives a yellow coloration. The addition of hexane leads to a purplish-pink coloration on the surface of the mixture. This is characteristic of the presence of Cl₂ in the mixture. Indeed, the initial pH of the reaction medium was kept below 2.5 throughout the electrolysis. The purplish-pink colouration is characteristic of the presence of Cl₂ in the reaction medium during the electrolysis of MNZ in 0.1 M HClO₄.