Revisiting the Role of Dissolved Silicate in Catalyzing the Hydrogen Peroxide Process with Iron-Bearing Minerals

Abstract

Injecting H₂O₂ into aquifers is a widely used in-situ chemical oxidation (ISCO) technology for groundwater remediation. Dissolved silicate was reported to decrease the reactivity of iron (Ⅲ)-bearing minerals toward H₂O₂. In this study, the effect of naturally occurring levels of dissolved silicate (≤1 mM) on the catalyzing hydrogen peroxide (CHP) with Fe(II) minerals is revaluated, and new observations that contradict with previous studies are reported. Specifically, dissolved silicate enhanced the CHP process by Fe(II) minerals. In the presence of Fe(II) minerals, siderite and ferrous oxide (FeO), which had a stronger dissolution tendency than Fe(III) minerals, dissolved silicate could prevent the dissolved iron species from precipitation through a coordinating effect, therefore reinforcing the homogeneous CHP process and the degradation of 2,4-dichlorophenol. The solution pH decreased due to the generation of degradation intermediates, and the solution acidification in turn promoted further dissolution of Fe(II) minerals. FeO particles exhibited the strongest silicate adsorption among the minerals, therefore a higher initial silicate concentration of 1 mM was needed to observe the enhancing effect. This study redefines the role of dissolved silicate on CHP process and provides clues to the design of efficient H₂O₂-based ISCO system for the remediation of groundwater.

Graphical Abstract

INTRODUCTION

In situ chemical oxidation (ISCO) technology is often used for remediation of groundwater contaminated with organics due to its advantages of ease of operation, low cost and strong oxidation ability.^1,2 From the perspective of engineering application of ISCO, H₂O₂ is one of the popular oxidants because of its low unit cost and applicability to a wide range of organic pollutants.^1,3 With the assistance of ferrous salt catalyst, H₂O₂ develops a homogeneous catalyzing hydrogen peroxide (CHP) process, in which H₂O₂ is activated into free radicals to promote the oxidation of organic pollutants. Therefore, the common operating mode of CHP is injecting H₂O₂ oxidant and ferrous salt catalyst simultaneously (i.e., also known as Fenton reagent) into aquifers to maximize the oxidizing capacity of H₂O₂ oxidants. However, injecting ferrous salts has problems such as pH-sensitivity, poor dispersion, short migration distance, and blocking flow in of groundwater aquifer.^4,5 Considering that various iron-bearing minerals widely exist in aquifers, a CHP process can be implemented by injecting H₂O₂ solution alone. The iron-bearing minerals, working as solid catalysts, could activate H₂O₂ in-situ to oxidize the organics via the heterogeneous CHP process.^6–8 Injecting H₂O₂ alone has some advantages for engineering application. For example, it can affect a larger area due to the decreased decomposition rate of H₂O₂, and mitigate the corresponding safety concerns associated with the exothermic reactions and pressure buildup. ⁹

Dissolved silicate, a common anion with concentrations in the range of 0–1.5 mM in groundwater,^10,11 was found to influence the heterogeneous CHP process with iron mineral catalysts. Pham et al. reported that the presence of dissolved silicate at concentrations comparable to those encountered in natural waters decreased the reactivity of Fe(Ⅲ) minerals toward H₂O₂, mainly because silicate adsorbed onto the surface of iron minerals to cover its active sites. ¹² Under a circumneutral pH, silicate-preequilibrated goethite, amorphous iron oxide, hematite, iron-coated sand, and montmorillonite were significantly less reactive toward H₂O₂ decomposition than their original counterparts, with the H₂O₂ loss rates inversely proportional to silicate concentrations. In the goethite-catalyzed H₂O₂ system, the overall •OH yield, defined as the percentage of decomposed H₂O₂ producing •OH, was almost halved in the presence of 1.5 mM silicate.¹²

In groundwater environment, due to the hypoxic environment, there are a large number of typical Fe(II)-bearing minerals.^13,14 In the present study, we revisit the effect of dissolved silicate on the H₂O₂ catalysis by iron minerals. When the focus shifts to the Fe(II) minerals (i.e., siderite and ferrous oxide), we unexpectedly found that the dissolved silicate can promote the CHP processes, rather than the previously reported inhibitory effect. This study not only allows to redefine the role of dissolved silicate on H₂O₂-based chemical oxidation technology, but also offers clues to developing more efficient ISCO technology for groundwater remediation.

MATERIALS AND METHODS

Chemicals and Materials.

Hematite (α-Fe₂O₃) and ferrous oxide (FeO) were obtained from the Sinopharm Chemical Reagent Ltd. and the Alfa Aesar, respectively. Siderite (FeCO₃), goethite (α-FeOOH) and magnetite (Fe₃O₄) were synthesized following the procedures reported in the literatures.^15–17 The detailed information about preparation method and material characterization can be found in Text S1, Figure S1 and Table S1 in the Supporting Information (SI). All chemical reagents were reagent grade from the Sinopharm Chemical Reagent Ltd., and deionized water was used throughout this study.

Experimental Procedures.

H₂O₂ decomposition and 2,4-dichlorophenol (DCP) degradation experiments were carried out in 50-mL polypropylene vials fixed on a rotary shaker with a constant speed of about 100 rpm at room temperature (20 ± 1 °C). The experimental setup is shown in Figure S2, and the typical experimental procedure is as follows. Predetermined volumes of DCP stock solution (1 g/L) and sodium silicate stock solution (0.1 M) were added to the vials prefilled with deionized water. The solutions were not buffered, H₂SO₄ and NaOH were used to adjust the initial pH. Afterwards, the concentrated H₂O₂ solution and a given weight of iron mineral powders were added into the solution, allowing the commencement of the heterogeneous CHP reaction. The final volume of total solution was set to 50 mL, and the details for all the experimental sequences are given in Table S2. During the reaction, the vial was covered and oscillated. At predetermined time intervals, approximately 1 mL solution was extracted to determine the concentrations of DCP, dissolved iron and residual H₂O₂. The self-decomposition of 5 mM H₂O₂ was negligible over the reaction time of 5 days (Figure S3). p-chlorobenzoic acid (pCBA) was used as a probe compound to evaluate the yield of hydroxyl radicals via its degradation experiment.¹⁸ The silicate adsorption experiments were performed with the same vials. 4 g/L iron minerals were added in the silicate solutions, and the concentrations of silicate and dissolved iron were monitored. Experiments were carried out as well in real groundwater. The groundwater was collected from a well located in the Danjiangkou city of China, and its inorganic content is given in Table S3. More detailed information about the analytical and characterization methods over the experiments is provided in Text S2 of the SI.

RESULTS AND DISSCUSSIONS

Effect of dissolved silicate on CHP for DCP degradation.

Figure 1a shows the degradation of DCP by the CHP process using siderite as catalyst in the presence or absence of silicate. When H₂O₂ or siderite was used alone, the removal efficiency of DCP did not exceed 30% within 5 days, because neither the direct oxidation by H₂O₂ nor the adsorption process could significantly decrease the DCP.¹⁹ The heterogeneous CHP system, using siderite as catalyst, could degrade 46.9% of DCP within 5 days. In contrast, the introduction of dissolved silicate could significantly enhance the removal of DCP: the removal efficiency and removal rate of DCP were proportional to the silicate concentration; and a complete removal could be achieved in the system with 1 mM silicate within 3 days. Noted that in the first two days, the removal of DCP was slightly inhibited in the presence of dissolved silicate, being consistent with the observation that the catalytic decomposition of H₂O₂ was suppressed (see Figure 1b). This observation is ascribed to the adsorption of silicate onto the surface of iron minerals,^20–22 which covered the active sites and hindered the heterogeneous CHP process.¹² From the 3^rd day, an evident increase in dissolved iron species was observed for trials with silicate (Figure 1c). This increased dissolved iron led to the occurrence of homogeneous CHP process, which accounted for the better DCP removal in the presence of silicate. Conversely, for the silicate-free trial, the concentration of dissolved iron was quite low. According to the above observations, it is obvious that the inhibitory effect of silicate on the catalytic decomposition of H₂O₂ still existed in the initial stage of reaction. However, along with the increase in the concentration of soluble iron species, the decomposition of H₂O₂ and the degradation of DCP were instead accelerated. In other words, the dissolved silicate finally showed an enhancing effect, rather than the previously reported inhibitory effect, on the CHP process catalyzed by siderite.

Figure 1.

(a) DCP degradation, (b) H₂O₂ decomposition and (c) variation of dissolved iron in the CHP process using siderite as catalyst. (d) DCP degradation, (e) H₂O₂ decomposition and (f) variation of dissolved iron in the CHP process using ferrous oxide as catalyst. (h) DCP degradation, (i) H₂O₂ decomposition and (j) dissolved iron accumulation in the CHP process using goethite as catalyst. The mineral catalyst was 0.2 g/L, [DCP]_ini = 20 mg/L, [H₂O₂]_ini = 5 mM, and pH_ini = 7.5.

Figure 1d–f shows the results of the CHP process using another Fe(II) mineral ferrous oxide. As can be observed, in the presence of 1 mM silicate, a complete removal of DCP could be achieved in 4 days (see Figure 1d), which is shorter than the 5 days of the silicate-free trial. However, in regard to the 0.2 and 0.5 mM silicate, the removal of DCP was actually inhibited. Meanwhile, the decomposition of H₂O₂ was not significantly affected when 1 mM silicate was introduced, while 0.2 and 0.5 mM silicate could slightly decrease the H₂O₂ consumption (Figure 1e). Figure 1f gives the variation of dissolved iron concentration in different trials. Over the four days after the reaction commenced, dissolved iron species were only detected in the trial with 1mM silicate. On the 5^th day, a significant increase in dissolved iron was observed for the silicate-free trial and one with 1mM silicate. A surge was observed as well in the system with 0.5 mM silicate on the 7^th day. Accordingly, the increased dissolved iron greatly promoted the removal of DCP.

The effect of dissolved silicate on the CHP process with goethite was comparatively studied under the same experimental condition. As shown in Figure 1h–j, the effect of silicate on the goethite-catalyzed system is different from the Fe(Ⅱ)-mineral catalyzed system. Figure 1h shows that the DCP removal efficiency could reach 80% for the silicate-free trial. This observation accords with the result in previous study that a “goethite +H₂O₂” CHP system exhibited good performance for the destruction of organics.²³ However, when dissolved silicate was present, the removal of DCP was significantly inhibited, and the inhibitory effect positively correlated with the concentration of silicate. The suppression of catalytic decomposition of H₂O₂ (Figure 1i) verified the inhibitory effect of silicate on the goethite-catalyzed system. In regard to the concentration of dissolved iron in the goethite system, it can be observed that, different from the Fe(Ⅱ)-mineral system, the dissolved iron was basically negligible (Figure 1j). Besides, as shown in Figure S4a–c, the presence of silicate enhanced the removal of pCBA in the Fe(Ⅱ) mineral system. Conversely, the removal of pCBA was significantly inhibited in the goethite system while silicate was added, confirming that less hydroxyl radicals were produced for the goethite-catalyzed system with silicate. Furthermore, the effect of dissolved silicate on the CHP processes catalyzed by hematite or magnetite was also studied (see Figure S5 in the SI), and the same phenomena as that of goethite were observed. The results of Fe(III) mineral confirm that the inhibitory effect of silicate on the decomposition of H₂O₂ and DCP degradation. Moreover, the negligible soluble iron in the CHP systems with Fe(III) minerals suggested that the heterogeneous catalysis was the dominating mechanism. In contrast, for the CHP systems with Fe(II) minerals, the elevated concentration of soluble iron species in the presence of silicate allowed a homogeneous catalytic mechanism, thereby promoting the degradation of DCP. It has been revealed that dissolved silicate could coordinate with ferrous/ferric irons to keep the soluble state of the iron species.^24,25 Once the ferrous ions dissociated from the crystals, they did not precipitate due to the coordinating effect of the silicate in solution, allowing the occurrence of a homogeneous CHP process. Therefore, the enhancing effect of silicate is associated with its coordinating effect on ferrous/ferric iron.

The pH variation during the reaction was monitored to give an insight into the dissolution behavior of minerals. As shown in Figure S6, the solution pH generally decreased as the reaction progressed in all systems. Moreover, the pH decreasing trend appeared to correlate with the DCP removal rate. For example, for the siderite, a higher silicate concentration led to a higher DCP removal rate and a faster pH dropping; while the goethite-catalyzed CHP process showed the fastest pH decreasing rate without silicate. This observation indicates that the generation of DCP degradation intermediates (such as organic acids with low molecular weights) could led to a pH change towards a more acidic condition.²⁶ This is further confirmed by the observation shown in Figure S7. Comparing with the counterpart without DCP, the mineral-catalyzed CHP systems with DCP all showed a solution pH approaching a more acidic condition. As a result, the dissolution of iron was enhanced by the proton-promoted dissolution process, which partly explained the significant increase in soluble iron species in Figure 1c, f.

Interaction of dissolved silicate and iron minerals.

The above results reveal that silicate could cover the surface of iron minerals via adsorption process, and meanwhile it worked as ligand to prevent ferrous/ferric ions from precipitation. The absorption of silicate on the surface of minerals was confirmed by the FTIR, EDS mapping and XPS (FigureS8, S9, and Text S3 in the SI). It is important to quantitatively analyze the distribution of silicate between the mineral solids and the aqueous solution. The adsorption of dissolved silicate onto iron minerals was further investigated through determining the concentration change of silicate in solution (Figure 2a). Among the five minerals, the siderite particles showed the weakest ability to adsorb silicate (less than 0.5 mg/m² for all initial concentrations) after 5 days of equilibrium; while the FeO particles exhibited the strongest absorption. For the goethite, hematite and magnetite, the absorption amounts of silicate, basically lower than 5 mg/m², increased with the initial concentration of silicate. In regard to the distribution of the silicate, due to the strong adsorption of silicate by FeO, the partition ratio of the silicate in the aqueous solution was all lower than 0.6, and the remaining silicate was down to 0.478 mM even for the trail with 1 mM initial concentration (Figure 2b). This phenomenon could well explain the inhibitory effect of 0.2 and 0.5 mM silicate on the FeO-catalyzed CHP process for DCP removal (Figure 1d). Because of the strong absorption of silicate by FeO, the heterogeneous CHP process was greatly hindered. Moreover, for 0.2 mM and 0.5 mM initial concentration, insufficient free silicate in the solution could not fully complex with the iron species, which was adverse to the homogeneous CHP reactions. Therefore, an enhancing effect of silicate was only observed for the 1 mM initial condition for the FeO-catalyzed CHP process.

Figure 2.

(a) Adsorption capacity of minerals to the dissolved silicate (mg/m²). (b) Distribution of silicate between the solid phase (iron minerals) and the liquid phase (aqueous solution); and the remaining concentration of silicate after 5 days of equilibrium was indicated by the number on the column. (c) Concentration of dissolved iron detected during the silicate adsorption experiments after 5 days. The dosage of various iron minerals was 4 g/L, pH_ini = 7.5, no DCP and H₂O₂ were added.

Figure 2c shows the dissolved iron concentrations after 5 days of adsorption equilibrium. In the absence of DCP and H₂O₂ in the solution, the overall dissolved iron was significantly less than that in the counterpart CHP process (see Figure 1c, f). The dissolved iron was detected at a relatively low level for the three Fe(III) minerals. Fe(II) minerals exhibited a higher level of dissolved iron, and the concentration of dissolved iron increased with the initial concentration of silicate. As shown in Table S4, compared to Fe(III) minerals, Fe(II) minerals had larger k_sp,^27–29 and readily dissolved ferrous ions. For example, hematite has a rhombic symmetry and a corundum structure,³⁰ in which oxygen atoms closely surround Fe³⁺ with the same density. In contrast, for siderite, the bond between Fe(II) and -CO₃ is relatively weaker, so ferrous ions tend to dissolve into solution. In addition, a higher level of dissolved silicate in the solution could complex more ferrous ions, thereby the increase of dissolved iron along with the initial silicate concentration could be observed for the Fe(II) minerals. Because the Fe(II) minerals had an inherent tendency to release more ferrous iron, dissolved silicate could act as a ligand to complex with the released ferrous ions, thus enhancing the homogeneous CHP process. Figure S10 illustrates the role of dissolved silicate on the dissolution of iron and the catalytic decomposition of H₂O₂ by the iron minerals with different iron valence states.

The effect of silicate was further investigated using real groundwater. As shown in Figure S11a–c, the overall tendency of the removal of DCP, the decomposition of H₂O₂ and the variation of dissolved iron concentration in different trials were similar to the counterpart shown in Figure 1a–i. These results confirmed that the above findings were also applied to the real groundwater. Meanwhile, it can be observed that concentration of dissolved iron in the Fe(Ⅱ) mineral system was less than that in the counterpart using deionized water (see Figure 1c, f). This is attributed to the fact that the metallic ions, such as Ca²⁺ and Mg²⁺, could combine with silicate, ²⁵ and thereby resulting in the decrease in silicate for iron complexation.

CONCLUSIONS

The effect of naturally occurring dissolved silicate (≤1 mM) on the CHP processes catalyzed by iron minerals was revisited to improve the understanding of the role of silicate. The effect of dissolved silicate on the CHP process was associated with the valence state of iron in the minerals. In the presence of Fe(II) minerals such as siderite and ferrous oxide, which had a stronger dissolution tendency, dissolved silicate could prevent dissolved iron from precipitation under a circum-neutral pH condition through ligand effect, therefore strengthening the homogeneous catalytic decomposition of H₂O₂ and oxidation of DCP. Especially, ferrous oxide particles exhibited the strongest silicate adsorption among the minerals, therefore the enhancing effect was only observed at a higher initial silicate concentration of 1 mM. In the presence of Fe(III) minerals such as goethite and hematite, the iron dissolution was not significant. Dissolved silicate demonstrated inhibitory effect only via covering the active sites for the heterogeneous CHP process. Similar findings were also observed in the real groundwater. The different effects of dissolved silicate on the catalytic processes with Fe(II) and Fe(III) minerals, suggest that the primary form of iron minerals in aquifers, their valence state and the concentration of dissolved silicate need to be considered when implementing in-situ H₂O₂ injection for the remediation of contaminated groundwater. Meanwhile, the concentration, injection depth and frequency of H₂O₂ should be reasonably selected; if necessary, dissolved silicate can be injected into the aquifer as additives to achieve a higher concentration of dissolved iron species.

Synopsis:

The effects of dissolved silicate upon catalyzing hydrogen peroxide process by Fe(Ⅱ) and Fe(Ⅲ) minerals are radically different.