Study of rhodamine B degradation by persulfate Open Access

Organic colorants are widely used in many industries (textiles, plastics, pharmaceuticals, paint, and food). Unfortunately, in aquatic environments, due to their low biodegradability, they disrupt the ecological balance, resulting in immediate and long-term hazards such as eutrophication, carcinogenicity, and bioaccumulation, and posing a serious threat to aquatic life and human health (Berradi et al. 2019). In addition, removing dyes from the environment remains a difficult task due to their persistence and transport to natural waters. However, pollution caused by dyes can be effectively studied using simulation models integrating physicochemical parameters such as rate constants, temperature, pH, and others (Mehnaz et al. 2023) to predict water quality. Understanding the degradation of dyes is essential for developing effective pollution reduction strategies and protecting water resources.

It has been reported that RB has carcinogenic and teratogenic effects on animals and poses a serious threat to the ecosystem (Elias et al. 2021). To eliminate the risks that RB can cause, various water treatment technologies have been evaluated, including activated sludge biodegradation (Yang et al. 2020), ion exchange (Saruchi & Kumar 2019), coagulation (Adeogun & Balakrishnan 2017), adsorption (Anh Tran et al. 2021), photocatalytic oxidation (Huang et al. 2022), and activated PS (Bing & Wei 2019).

In recent years, hybrid activation systems have been developed to improve the performance of PS in effluent treatment: heat/transition metal activation (Sun et al. 2023), carbocatalytic/heat activation (Duan et al. 2020), and UV/H₂O₂/PS activation (Ding et al. 2020). In addition to SR, activation of the PS can generate a variety of reactive species, including hydroxyl radical (OH^•), hydroperoxide anion (HO₂), and superoxide (⁠⁠) (Furman et al. 2010). Sulfate and hydroxyl radicals are the best known in the PS process for degrading organic pollutants (Ding et al. 2020). In acidic conditions, SRs are the predominant species, whereas in alkaline conditions, hydroxyl radicals are the predominant reactive species (Mora et al. 2009; Furman et al. 2010; Liang & Huang 2012).

The degradation of organic pollutants by free radicals is often called the radical degradation pathway (Yang et al. 2021). In recent years, a nonradical degradation pathway has been proposed, where activated PS by metallic oxides or carbon (Li et al. 2023; Wu et al. 2024) degrades target pollutants through direct electron transfer.

Another reaction pathway has been proposed, in which the PS ion reacts directly with RB and cationic dyes (methylene blue and chloroaniline) (Mcheik & Jamal 2013; Yang et al. 2021; Zhu et al. 2021). This reaction, which is sometimes considered but ignored in other contexts, is poorly understood due to the lack of data and thorough research. For example, Yang et al. (2021) observed a 30% removal of RB using only PS, with a PS/RB molar ratio of 50. In addition, Mcheik & Jamal (2013) reported a 70% removal of RB after 33 min, with a PS/RB molar ratio of 1,000. Conversely, in the study by Lin & Hsiao (2022), which investigates the degradation of RB by heated PS with a molar ratio greater than 1,000, they do not mention this reaction pathway nor any loss of RB, and they state that the kinetic degradation follows first-order kinetics. Furthermore, Crincoli et al. (2020) reported a 4% loss of RB over 75 min with a PS/RB molar ratio of 294.

To our knowledge, for the direct oxidation of RB, no scientific work has detected the fragments resulting from this reaction. Furthermore, there is currently no established mechanism or kinetic data available in the literature. The purpose of this study is to gain a better understanding of the mechanism of this reaction and to supplement the kinetic data on PS available in the literature. We examined the effects of the initial concentration of PS to evaluate the PS/RB ratio on the direct reaction between PS and RB. We also added ethanol to verify the presence of radical species in the reaction and analyzed the impact of temperature on this direct reaction. Finally, we propose a reaction pathway for the direct oxidation of RB based on data obtained from UV-Vis and HPLC-MS² methods.

This model allows us to calculate the rate constant value between SR and RB (Equation (5)), considering the competitive reaction between PS and RB (Equation (7)) as a function of temperature.

The calculation was based on the rate constants used in the RB degradation mechanism (Table 1) as well as the experimental conditions (⁠⁠, [RB]₀, pH, and temperature).

RB (C₂₈H₃₁ClN₂O₃, 99.5%, Merck), potassium persulfate (K₂S₂O₈, 99%, Merck), sulfuric acid (H₂SO₄, 98%, Sigma-Aldrich), ethanol (CH₃CH₂OH, >99.7%, Sigma-Aldrich), and all solutions were prepared with demineralized water. Diluted solutions of RB (1–5 mg L⁻¹) were prepared from an initial solution of 5 mg L⁻¹ to plot the calibration curve and verify that the commercial product followed Beer-Lambert's law. The required quantity of products (PS and ethanol) is added to a flask containing 200 ml of RB solution, which is heated (in some experiments) to the desired temperature with a thermostatic bath, with stirring.

The study was conducted at pH 3 (± 0.2) and monitored with a Metrohm pH/ionometer model 781.

The degradation of RB solutions was monitored by measuring their spectra using a DR 2,800 UV-Vis spectrophotometer (RBT technology, optical path = 2 cm) and a 2,401 PC UV spectrophotometer (Shimadzu, optical path = 1 cm).

An analytical ultra-high performance liquid chromatography-mass spectrometry analysis was conducted using a Thermo Vanquish system coupled with a Q-Exactive mass spectrometer. The mass spectrometer was equipped with a heated electrospray ionization (HESI) source. The chromatographic separation was carried out on an ACQUITY UPLC BEH C18 column (1.7 μm particle size, 100 × 2.1 mm dimensions) with a mobile phase flow rate of 0.4 mL min⁻¹. The column temperature was maintained at 30 °C. The mobile phase consisted of two components: pure water (referred to as eluent A) and acetonitrile (referred to as eluent B), both containing 0.1% formic acid. The sample, injected in a volume of 10 μL, underwent a gradient elution. The initial gradient composition was 90% eluent A, which remained constant for 1 min, followed by a linear decrease to 30% eluent A over 6 min. Eluent A was then reduced to 0% over 2 min before remaining constant for another 2 min. For the mass spectrometry analysis, the HESI source was operated in positive polarity. The operational parameters were as follows: the sheath gas, auxiliary gas, and sweep gas were set to flow rates of 40, 20, and 0 (a.u.), respectively. The auxiliary gas temperature was kept at 400 °C, and the capillary temperature was set to 350 °C. The capillary voltage was set at 4,000 V. The mass range for data acquisition was 100–1,500 m/z.

RB degradation accelerates as the initial PS concentration increases, resulting in more rapid decolorization and potentially higher degrees of mineralization in shorter time intervals. For example, 90% decolorization of RB was achieved in 300 min with a PS/RB molar concentration ratio of 354, whereas only 30% decolorization was achieved with a ratio of 20 and in the same time interval. A similar observation was reported in the work of Wu et al. (2024), where RB degradation was achieved by PS via a nonradical way with PS/RB ratios of 10, 20, 40, and 80. At a ratio of 80, complete mineralization of RB is achieved in 4 h. However, at a ratio of 10, RB decolorization reaches only 50% in 4 h. As mentioned earlier, Mcheik & Jamal (2013) reported a 70% removal of RB after 33 min, with a PS/RB molar ratio of 1000.

As the first step is fast and difficult to monitor, observed rate constants (k_obs) were determined from the slow steps (Figure 2 bottom). The k_obs values were found to be (1.0 ± 0.15) × 10⁻³, (3.0 ± 0.45) × 10⁻³, and (6.0 ± 0.9) × 10⁻³ min⁻¹ for initial PS concentrations of 20, 101, and 354 × [RB]₀, respectively.

The minimal effect of ethanol on RB degradation at 25 °C suggests a limited contribution from SR. This is likely due to the low dissociation rate of PS at 25 °C (1.02 × 10⁻⁷ s⁻¹) (House 1962), which restricts SR generation (Bing & Wei 2019) due to the low dissociation rate of PS at 25 °C. In contrast, at 50 °C, the degradation of RB in the presence of ethanol is significantly reduced, indicating a substantial contribution of SRs to the degradation process at this temperature. This further supports the conclusion that the radical pathway is less prominent at 25 °C. A similar observation was made in the study conducted by Yang et al. (2021), in which the addition of ethanol resulted in a slight inhibition of RB discoloration. As demonstrated in the study conducted by Heidarpour et al. (2020), the efficiency of RB degradation by photocatalytically activated PS significantly reduced with the addition of ethanol, indicating that SRs play a substantial role in RB removal. It can thus be concluded that the degradation of RB at 25 °C occurs primarily through direct oxidation by PS.

The RB degradation kinetics was evaluated at 9, 15, 18, and 25 °C under identical conditions. The measured slopes for the temperatures of 9, 15, 18, and 25 °C are 1.23 × 10⁻⁴, 1.46 × 10⁻⁴, 1.61 × 10⁻⁴, and 1.45 × 10⁻⁴, respectively. These values are very close to each other, giving an average of 0.00014 s⁻¹. This consistency suggests a stable degradation rate across this temperature range and minimal variation in the reaction rate with temperature.

Using Equation (19), the average second-order rate constant k₇ for the direct reaction between PS and RB was calculated to be 0.037 M⁻¹ s⁻¹ for temperatures ranging from 9 to 25 °C. This rate constant can be used to calculate k₅ using Equations (14) and (16).

The second-order rate constant k₅ can be determined at 40, 50, and 60 °C using theoretical calculations (Equations (14) and (16)) based on the initial experimental conditions (⁠⁠, [RB]₀, pH, and T) and rate constants obtained from the literature (Table 1). As illustrated in the section on the effect of ethanol, at 25 °C, the radical pathway (Equation (5)) is negligible compared to the direct oxidation (Equation (7)). Consequently, it is not feasible to determine k₅ at 25 °C using the proposed calculation model. However, following the determination of k₅ at 40, 50, and 60 °C, the ln(k₅) vs 1/T curve is plotted to obtain the Arrhenius equation. Subsequently, k₅ at 25 °C is determined using this equation. Consequently, the reaction rate constants k₅ obtained at 40, 50, and 60 °C are 1.7 × 10⁷, 3.5 × 10⁷, and 7.81 × 10⁷ M⁻¹ s⁻¹ (Table 1).

The Arrhenius plot revealed an E_a of 64.4 (±6.4) kJ mol⁻¹ for the RB degradation reaction at various temperatures (40, 50, and 60 °C). According to several studies (House 1962; Huang et al. 2002; Ahmadi et al. 2016), the reaction rate doubles over a 10 °C temperature range if the activation energy is approximately 50 kJ mol⁻¹ at least. Our results, however, show an activation energy of 64.4 (± 6.4) kJ mol⁻¹ with reaction rate constants k₅ obtained at 40, 50, and 60 °C of 1.7 × 10⁷, 3.5 × 10⁷, and 7.81 × 10⁷ M⁻¹ s⁻¹, respectively. These findings suggest that our data are still consistent with the general rule.

This allows deducing the value of the rate constant k₅ at 25 °C, which is equal to 5.0 × 10⁶ M⁻¹ s⁻¹.

The highest reported value for the reaction rate constant between SR and aromatic molecules is around 10⁹ M⁻¹ s⁻¹ (Wojnárovits & Takács 2019). It should be noted that this value may undergo a slight increase or decrease in the presence of electron-donating or electron-accepting substituents, respectively (Wojnárovits & Takács 2019). The reaction rate constant of SR with methylene blue is 3.4 × 10⁷ M⁻¹ s⁻¹ at 60 °C (Liang & Huang 2012), which is in agreement with the calculated value.

Figure 5 includes the only values of k₅ identified in the existing literature at 25 and 40 °C, which are 8.39 × 10⁶ and 3.02 × 10⁸ M⁻¹ s ⁻¹, respectively (Zeng et al. 2018; Crincoli et al. 2020). It is noteworthy that our findings align with the reported value of k₅ at 25 °C but differ from the observed value at 40 °C. Zeng et al. (2018) measured k₅ at 25 °C using metal-activated peroxymonosulfate (PMS). It should be noted that PMS reacts in a similar manner to PS with cationic dyes (Yang et al. 2018). On the basis of experimental conditions ([PMS]/[RB] = 0.02), Zeng et al. (2018) concluded that RB loss due to reaction with PMS is negligible. This finding is supported by the experiments in Figure 2, which demonstrates that the lower the RB/PS ratio, the lesser the RB reacts with PS.

In the study conducted by Crincoli et al. (2020), the value of k₅ was determined at 40 °C through the use of thermally activated PS. For a PS/RB ratio of 294 (2.94/0.01) and a 75-min reaction follow-up, the authors estimate that the RB loss is no more than 4%. These findings are inconsistent with our own observations, which indicated a 25% loss for a ratio of 101 and the same reaction time. Furthermore, these findings differ from those reported by Yang et al. (2021) and Mcheik & Jamal (2013), who observed a 30 and 65% loss of RB, respectively, for a PS/RB ratio of 50 and 950 after 80 and 33 min of reaction time.

Several pathways (radical and nonradical) for RB degradation by PS have been proposed, including the removal of ethyl groups (Hu et al. 2006; Zheng et al. 2012), the removal of the carboxylate ring (Ma et al. 2003), the rupture of the oxy bridge in the heterocycle (Zheng et al. 2012), and the chromophore cleavage (Wu et al. 2024).

In light of the experimental evidence, we will assess these pathways with a view to proposing a reaction scheme for nonradical RB degradation at 25 °C.

The results of the UV-Vis spectral monitoring at 25 °C are consistent with the masses detected and the corresponding chemical formulas proposed in Table 2. However, the detection limits of our MS prevented the detection of small fragments with m/z values below 150. It has been previously reported by numerous authors that the disappearance of ethyl fragments during the RB is accompanied by the appearance of intermediates (Huang et al. 2022). The disappearance of each ethyl fragment results in the loss of two carbon atoms and four hydrogen atoms from the molecular structure. However, in line with our spectrophotometric findings, the elimination of ethyl fragments from RB is uncommon under our conditions. Indeed, Compound 3 detected by the mass spectrum (Table 2), which is the result of the liberation of two ethyl groups from the RB structure, is to be considered with caution due to a low signal-to-noise ratio and a very high polarity observed during its retention time in the column.

Figure 7 depicts possible RB degradation pathways based on the aforementioned results and previous works (Wu et al. 1998; He et al. 2009; Jiang et al. 2018; Zhou et al. 2020; Huang et al. 2022). The predominant reaction pathways under these conditions are RB hydroxylation and chromophore disruption. Mass spectra were employed to identify 13 degradation products, and chemical structures were proposed accordingly. The compound with an m/z value of 443 was identified as RB. Our findings indicate that RB undergoes mono-hydroxylation, resulting in an intermediate (Compound 2). In parallel, RB may eliminate two ethyls (Compound 3). Compound 2 reacts in two ways: (1) the removal of an ethyl moiety produces two isomers (Compounds 4 and 5), and (2) hydroxylation produces a tri-hydroxylated compound (Compound 6), followed by tetra-hydroxylated isomers (Compounds 8 and 9). The hydrolysis of Compound 6 results in the formation of Compound 7, which is characterized by oxo-bridge disruption or, alternatively, the disruption of heterocycles, leading to the generation of benzene derivatives, namely, compounds 10 and 11. The latter is oxidized with the addition of an oxygen atom, resulting in the formation of two isomers (Compounds 12 and 13).

The direct oxidation of RB by PS was investigated at room temperature. The reaction is dependent on the initial concentration of PS, exhibits no temperature dependence between 9 and 25 °C, does not follow a radical pathway, and does not follow the first-order kinetics. This reaction is characterized by two distinct kinetic steps: an initial rapid phase, followed by a subsequent slower phase. The results of the experiments at higher temperatures indicated that the degradation of RB occurs via two parallel competing reactions, with one reaction dominating at low temperatures and the other at high temperatures. It was demonstrated that the exclusion of the direct reaction between PS and RB in kinetic studies leads to the generation of erroneous rate constants. Ultimately, the analysis of the reaction mixture by UV-Vis and MS revealed the absence of ethyl group elimination in the proposed reaction mechanism, which contrasts with the mechanisms previously postulated in the literature for RB degradation by activated persulfate. A scheme for the direct oxidation of RB by PS was put forth based on kinetic monitoring using UV-Vis spectrophotometry and HPLC-MS².

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.