Competitive Removal of Two Contaminants in a Goethite-Catalyzed Fenton Process at Neutral pH

Abstract

Plumes containing more than one contaminant can be found in sites polluted by gasoline or chlorinated solvents. This study evaluated Fenton-like removal efficiencies when two contaminants were coexistent. Perchloroethylene, trichloroethylene, cis-1,2-dichloroethylene, methyl-t-butyl-ether, benzene, and toluene were mixed in pairs and degraded by the goethite-catalyzed Fenton-like reaction at neutral pH and low H₂O₂ doses. Results revealed that the amount of each compound removed in two-contaminant systems was less than that in one-contaminant systems. This decline in removal was related to the reactivity constant (k_HO.), initial concentration (C₀), and number of double bonds of the compounds. In a solution that contained two compounds with similar k_HO. values, the amount of each compound removed was related primarily to the C₀ ratio of two compounds. When the k_HO. values of two compounds differed considerably, the one with the larger k_HO. value or the higher C₀ reduced or inhibited the Fenton-like reaction of the pollutant with the smaller k_HO. or lower C₀. Compounds with few double bonds tended to be less competitive for Fenton-like removal. By adding H₂O₂ repeatedly, the removal of a compound that is less competitive for the Fenton-like reaction can be recovered.

Introduction

The catalysis of H₂O₂ into reactive-free radicals at neutral pH by iron oxides is referred as the Fenton-like reaction (Watts et al., 2000). Soil and aquifer materials contain substantial amounts of natural iron oxides such as goethite (Jackson et al., 1986). Chlorinated ethylenes (Gates and Siegrist, 1995) and gasoline aromatics (Watts et al., 2000) are significantly degraded in soil systems by H₂O₂ without adding an external iron catalyst or pH adjustment. The hydroxyl radical (HO·) is one of the key free radicals in the Fenton reaction (Zepp et al., 1992). In the one-contaminant system, the amount of a compound reacted with HO· is related to its reactivity constant (k_HO.) and the initial concentration (C₀) (Zepp et al., 1992). The reaction efficiency is also influenced by the type and concentrations of the Fe catalyst, the H₂O₂ dosage, the characteristics of the contaminants, and the presence of other organic compounds (Matta et al., 2007).

Oxygenated gasoline contains methyl-t-butyl-ether (MTBE) and aromatic and saturated petroleum hydrocarbons. When an underground gasoline tank leaks, these compounds enter and coexist in the subsurface. Similarly, in spills of chlorinated ethylenes, tetrachloroethylene (PCE), trichloroethylene (TCE), and cis-1,2-dichloroethylene (cis-1,2-DCE) are all found in the groundwater. Earlier studies have performed the Fenton experiment containing only one contaminant. Therefore, this study evaluated the Fenton-like reaction efficiency of each compound in binary-contaminant systems. The results were correlated to the k_HO. value along with C₀, H₂O₂ dose, and goethite concentration.

Materials and Methods

The tested parameters are listed in Table 1. Because of high water solubility and toxicity, MTBE, benzene, and toluene were selected to represent dissolved gasoline hydrocarbons. The amounts of goethite (Fluka Chemical) that were added herein were in similar ranges to those in the soil and aquifer (Jackson et al., 1986; Yeh et al., 2004). Accordingly, the results can simulate the Fenton-like reaction that is catalyzed by natural iron oxide in soil. Low H₂O₂ concentrations were used because the competition of two compounds for the Fenton-like removal is likely to be significant in oxidant-limited situations. The concentrations of contaminants found in groundwater can vary from several μg/L to over hundreds of mg/L (ESTCP, 2006). Therefore, the Fenton-like removal of two compounds at various C₀ ratios was also performed. Buxton et al. (1988) and Zepp et al. (1992) concluded that the k_HO. value of a compound can represent its reactivity in the HO·-dominated Fenton reaction. The k_HO. values of 2.5×10⁹ M⁻¹·s⁻¹ for benzene, 2.8×10⁹ M⁻¹·s⁻¹ for toluene, 2.3×10⁹ M⁻¹·s⁻¹ for PCE, 1.6×10⁹ M⁻¹·s⁻¹ for TCE, 1.4×10⁹ M⁻¹·s⁻¹ for cis-1,2-DCE (Yeh et al., 2008), and 0.19×10⁹ M⁻¹·s⁻¹ for MTBE (Bergendahl and Thies, 2004) were used to represent the reactivity.

Table 1.

Experimental Parameters Used in Single- and Two-Contaminant Experiments

Compound(s)	k_HO. ratio	C₀ (mM)	C₀ ratio	H₂O₂ (%)	Goethite (g/mL)
Single-contaminant experiment
Toluene, benzene, MTBE, PCE, TCE, cis-1,2-DCE	—	0.1–0.9	—	0.01, 0.5	0.011, 0.11
Two-contaminant experiment
PCE/cis-1,2-DCE	1.6	0.1–0.9	0.11, 1, 4, 6, 9	0.01, 0.5	0.011, 0.11
Toluene/benzene	1.1
Benzene/PCE	1.1
PCE/toluene	1.4	0.1–0.9	0.11, 1, 4, 6, 9	0.5	0.11
Benzene/cis-1,2-DCE	1.8
Benzene/MTBE	13.2

MTBE, methyl-t-butyl-ether; PCE, tetrachloroethylene; TCE, trichloroethylene; DCE, dichloroethylene.

Goethite-catalyzed Fenton-like experiments

The Fenton-like reactor and procedures are similar as those used in our earlier studies (Yeh et al., 2004, 2008). In brief, the goethite slurry in the reactor was stirred for 15 min to ensure complete wetting of goethite powder. Then, the solution that contained one or two tested organic compounds was added to the reactor, and H₂O₂ was subsequently injected. Following H₂O₂ injection, the Fenton-like reaction continued for 5 min, and then methanol was added to fill the reactor to trap any unreacted radicals and to desorb organic compounds sorbed on goethite. The volatilized organic compounds by O₂ produced from the Fenton-like reaction were trapped in a gas impinger. The reaction pH remained around 6.8. All experiments were performed in triplicate at 22±3°C. The reaction time of 5 min was used, based on earlier studies that the HO·-dominated Fenton-like reaction finished within minutes after H₂O₂ addition (Yeh et al., 2004; Zimbron and Reardon, 2005). In the experiment with repeated H₂O₂ doses, the reaction time following each H₂O₂ application was 10 min; methanol was added only after the final H₂O₂ dose. The difference in removal efficiency between single- and two-contaminant systems was calculated as follows: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}\hbox{Difference in removal efficiency} \, ({\rm \%}) = (M_2 - M_1)/M_1 \times 100{\rm \%}\end{align*} \end{document}

where M₁ is mass of removal for compound A when only A is present, and M₂ is mass of removal for compound A when both compounds A and B were present.

Analyses

The aliquot in the reactor and the impinger were combined, centrifuged (4000 g, 20 min), and analyzed by an Agilent 6890 GC equipped with a Techmar purge-and-trap system. The detector was an FID and the capillary column was an HP-5 (30 m×0.32 mm). The flow rate of the carrier gas (N₂) was 1.5 mL/min. The injector and detector were heated at 200°C, and the oven temperature program was 40°C for 5 min, increasing to 250°C at 15°C/min.

Statistical analyses

All analyses were performed using S-Plus V 6.2 statistical software. Differences in removal efficiency of two and three groups were analyzed with the Standard Two-Sample t-Test and ANOVA, respectively. Statistical significance was accepted at p<0.05 for all comparisons.

Results and Discussion

Degradation in one-contaminant systems

Figure 1 plots the amount of six compounds removed against their k_HO. and C₀ values. At C₀ of 0.5 and 0.9 mM, the amount removed is linearly proportional to the k_HO. value (R²: 0.83–0.88). However, the flat line for C₀=0.1 mM suggests that at low C₀, the reaction efficiency is low and irrelevant to the k_HO. The amount of MTBE removed is significantly less than those of the other five compounds. The k_HO. value of MTBE is one order of magnitude lower than those of the other compounds. The HO· is electrophilic and preferentially attacks the unsaturated bonds rather than abstracting an H-atom (Buxton et al., 1988). Therefore, the aromatic hydrocarbons and chlorinated ethylenes, which have double bonds, are more reactive than MTBE, which contains mainly –CH– bonds.

FIG. 1.

Correlation of compound removal with reactivity constant (k_HO.) value at different initial concentrations (C₀) in one-contaminant system. Goethite=0.11 g/mL; H₂O₂=0.5%. MTBE, methyl-t-butyl-ether; PCE, tetrachloroethylene; TCE, trichloroethylene; DCE, dichloroethylene.

Degradation in two-contaminant systems

The degradation of PCE and cis-1,2-DCE in the PCE/cis-1,2-DCE solution still increased with its C₀ (Fig. 2a). However, the destruction of each compound is less than that in the one-contaminant system. Moreover, cis-1,2-DCE degradation dropped more than PCE. The benzene/MTBE pair exhibits a similar result (Fig. 2b). In particular, in the presence of benzene, there is no MTBE degradation at MTBE C₀ of <0.5 mM. Miller et al. (2003) also found that dinitrotoluene degradation was significantly decreased by acetone in the dinitrotoluene/acetone solution. Figures 3 and 4 present the difference in removal efficiency for compound pairs that contain the same and different number of double bonds, respectively. A negative value for the difference in removal efficiency indicates a decrease in degradation efficiency when two contaminants coexist. Of all tested compound pairs, the compound with the smaller k_HO. exhibited a greater decrease in degradation efficiency. When the k_HO. values of the compound pairs differed more, the degradation of the compound with the smaller k_HO. was more inhibited. As shown in Fig. 3a, a 20% suppression of benzene degradation occurred at toluene/benzene C₀ ratios >1. At C₀ ratios ≥1.0, no statistical significance was observed (p=0.513) for PCE in the PCE/TCE pair (Fig. 3b) and in the PCE/cis-1,2-DCE pair (Fig. 3c). On the other hand, the p-value of 0.014 indicates a statistical significance for differences in removal efficiency of TCE in the PCE/TCE pair (Fig. 3b) and cis-1,2-DCE in the PCE/cis-1,2-DCE pair (Fig. 3c). During the Fenton reaction of chlorinated organic compounds, the breakage of C–Cl bonds is one of the major pathways (Poulopoulos et al., 2008). At C₀ ratios of 4–9, TCE degradation by PCE decreased 20%–30% (Fig. 3b), while cis 1,2-DCE degradation by PCE dropped 30%–40% (Fig. 3c). Therefore, the amount of Cl substitution is another important factor on competition.

FIG. 2.

Comparison of Fenton-like removal in singular- and two-contaminant systems: (a) PCE, cis-1,2-DCE, and paired PCE/cis-1,2-DCE compound; (b) benzene, MTBE, and paired benzene/MTBE compound. Goethite=0.11 g/mL; H₂O₂=0.5%.

FIG. 3.

Changes in removal of compound pairs with same numbers of double bonds: (a) toluene/benzene (k_HO· ratio=1.1); (b) PCE/TCE (k_HO· ratio=1.4); (c) PCE/cis-1,2-DCE (k_HO· ratio=1.6). Goethite=0.11 g/mL; H₂O₂=0.5%.

FIG. 4.

Changes in removal of compound pairs with different numbers of double bonds: (a) benzene/PCE (k_HO· ratio=1.1); (b) benzene/cis-1,2-DCE (k_HO· ratio=1.8); (c) benzene/MTBE (k_HO· ratio=13.2). Goethite=0.11 g/mL; H₂O₂=0.5%.

The compound with more unsaturated bonds causes more inhibition on the destruction of another compound. A comparison of the results in Figs. 3a and 4a reveals that although the k_HO. ratios of toluene/benzene and benzene/PCE are the same (1.1), the PCE degradation in the benzene/PCE solution was suppressed more than the benzene degradation in the toluene/benzene solution (at C₀ ratios ≥1.0, p=0.0004 for benzene [Fig. 3a] and PCE [Fig. 4a]). A similar result was found for PCE/cis-1,2-DCE (Fig. 3c) and benzene/cis-1,2-DCE (Fig. 4b). The k_HO. ratio of PCE/cis-1,2-DCE is 11% less compared with benzene/cis-1,2-DCE; however, cis-1,2-DCE degradation was completely inhibited (−100%) by benzene in the benzene/cis-1,2-DCE solution, but was decreased 70% by PCE in the PCE/cis-1,2-DCE solution (at C₀ ratios ≥1.0, p=0.002 for PCE [Fig. 3c] and cis-1,2-DCE [Fig. 4b]). Similar to cis-1,2-DCE, there was no MTBE degradation in the benzene/MTBE solution at C₀ ratios >1 (Fig. 4c). The MTBE contains no double bond in the chemical structure. The cis-1,2-DCE contains one double bond, but its k_HO. is the second smallest among the tested compounds. Therefore, at C₀ ratios ≥4.0, the degradation of benzene is not affected (p=0.794 for benzene in Fig. 4a–c), but the degradation of cis-1,2-DCE and MTBE is completely inhibited.

At relatively high C₀, a compound with lower k_HO. and fewer double bonds can maintain its competitive potential. As shown in Fig. 3b, although the k_HO. ratio of PCE/TCE is 1.4, at a PCE/TCE C₀ ratio of 0.11 (i.e., the C₀ of TCE was nine times that of PCE), there is no difference in mass removal of TCE between the TCE-only and PCE/TCE systems. On the other hand, the degradation of PCE was suppressed by the presence of high C₀ of TCE. For the pairs with different numbers of double bonds (Fig. 4), both compounds exhibit the same extent of decline in degradation efficiency at a C₀ ratio of 0.11. Similar results were found for all of the tested pairs.

Effects of H₂O₂ and goethite doses

Increasing H₂O₂ or goethite dose recovered the degradation efficiency of each compound in the two-contaminant system (Table 2). However, the extent of recovery for the compound is related to its k_HO. value and C₀ ratios. When the PCE/cis-1,2-DCE C₀ ratio is 0.11 and the H₂O₂ dose was increased 50-fold, PCE degradation recovered 21% (from −78.4% to −56.8%); while cis-1,2-DCE degradation decreased slightly (from −13.7% to −17.4%). At a high PCE/cis-1,2-DCE C₀ ratio (9), an increasing H₂O₂ dose recovered cis-1,2-DCE degradation from −86.7% to −73.7%, but PCE degradation efficiency decreased from −19.6% to −22.7%. Thus, increasing H₂O₂ dose helps compounds with a lower concentration to compete for Fenton-like removal. Such an improvement is more significant for compounds with higher k_HO. values. A similar result was found in the goethite trials.

Table 2.

Difference in Reaction Efficiency Between Single and Dual Contaminants at Different H₂O₂ and Goethite Doses

			Difference in removal (%)
			Goethite=0.011 g/mL				H₂O₂=0.5%
C₀ (mM)			H₂O₂=0.01%		H₂O₂=0.5%		Goethite=0.011 g/mL		Goethite=0.11 g/mL
PCE	Cis-1,2-DCE	C₀ ratio	Cis-1,2-DCE	PCE	Cis-1,2-DCE	PCE	Cis-1,2-DCE	PCE	Cis-1,2-DCE	PCE
0.1	0.9	0.11	−13.7	−78.4	−17.4	−56.8	−43.2	−78.4	−46.7	−43.5
0.5	0.5	1.0	−62.2	−42.2	−45.7	−37.9	−67.0	−55.3	−57.5	−24.0
0.9	0.1	9	−86.7	−19.6	−73.7	−22.7	−89.5	−25.8	−84.2	−20.4

Repeated injection of H₂O₂ is commonly applied in the field to replenish the oxidant. The repeated addition of H₂O₂ restores the Fenton-like removal of a compound with less competitive potential. As shown in Fig. 5a, benzene degradation was suppressed by toluene only at the first H₂O₂ dose, but was recovered in the subsequent H₂O₂ addition. This is because each H₂O₂ dose reduced the toluene/benzene C₀ ratio and favored benzene degradation upon the subsequent H₂O₂ doses. When the k_HO. values of two compounds differ more, more repeated H₂O₂ injection is required to recover the removal of a compound with lower competitive potential. For the PCE/TCE pair with a C₀ ratio=9.0 (Fig. 5b), TCE degradation recovered by 20% at the fifth H₂O₂ dose, while the PCE removal dropped by ∼20%. Due to the very high benzene/MTBE k_HO. ratio and the effect of double bonds, little MTBE in the benzene/MTBE mixture was destroyed even after five doses of H₂O₂ (Fig. 5c). However, MTBE degradation is possible after benzene has been removed by persistent H₂O₂ injection.

FIG. 5.

Change in destruction after multiple doses of H₂O₂: (a) toluene/benzene (k_HO· ratio=1.1); (b) PCE/TCE (k_HO· ratio=1.4); (c) benzene/MTBE (k_HO· ratio=13.2). C₀ ratio=9; Goethite=0.11 g/mL; H₂O₂=+0.5% each time.

Impacts of competition on remediation

The above results imply that in two-contaminant systems, proper adjustments during site remediation are recommended due to the competition for Fenton-like removal. Take the benzene/MTBE plume as an example. MTBE has a larger weight percentage in gasoline, a higher solubility in water (1780 mg/L), and a lower sorption in soils. The benzene and MTBE concentrations are usually high in the region near the leakage point (inner zone), while MTBE is the dominant contaminant in the outer region of plume. Benzene and MTBE concentrations in sites in San Diego County, CA, ranged from 0.01–10² mg/L (Happel et al., 1998). The MTBE plume at the Naval Base, Ventura County, California was 1.5-fold in length of the benzene/toluene/ethylbenzene/xylenes plume (ESTCP, 2006). Therefore, persistent H₂O₂ injections and a long treatment period are necessary to overcome the inhibition of MTBE removal by benzene in the inner zone. In the outer region, because of the reduced benzene/MTBE C₀ ratios, less H₂O₂ injection and a shorter treatment time are expected. Similar phenomena are expected in plumes of other binary contaminants.

Conclusion

The amounts of compounds removed increase with the C₀ and the k_HO. value of the tested components during in-situ Fenton-like remediation of two-contaminant systems catalyzed by natural iron oxides in aquifers at neutral pH. However, in two-contaminant systems, competition for radicals causes a reduction in degradation efficiency of each compound. The competitive potential of a compound is related to its k_HO., C₀, and number of double bonds of the compound. For two compounds that have similar k_HO. values, the decline in the degradation is primarily related to the C₀ ratio. When the k_HO. values of two compounds differ greatly, the compound with the larger k_HO. value can suppress the Fenton-like reaction of the smaller k_HO. pollutant. The competition tended to more strongly affect the compound with fewer double bonds. In benzene/cis-1,2-DCE and benzene/MTBE systems, no degradation of cis-1,2-DCE or MTBE occurred at benzene/cis-1,2-DCE and benzene/MTBE C₀ ratios >1. The lower k_HO. compound retained its competitive potential only when it was present at relatively high C₀. Repeated addition of H₂O₂ gradually restored the Fenton-like removal of the compound with less competitive potential.

The contaminated plumes can also contain multiple organic compounds, metals, ions, and humic matter. Furthermore, H₂O₂, O₂, Fe(II), Fe(III), substrates and their radicals and other HO radicals are involved in the complicated branching pathways of Fenton-like degradation. Future studies addressing the competitive effects on the degradation subpathways among these species should help the design of in-situ Fenton-like remediation efficiency.

Footnotes

Acknowledgment

The study was supported by the National Science Council of Taiwan, R.O.C. (NSC 95-2211-E-020-006).

Author Disclosure Statement

No competing financial interests exist.

References

Bergendahl

J.A.

, Thies

T.P.

2004. Fenton's oxidation of MTBE with zero valent iron. Water Res., 38:327.

Buxton

G.V.

, Greenstock

C.L.

, Helman

W.P.

, Ross

A.B.

1988. Critical review of rate constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals (·OH/·O⁻) in aqueous solution. J. Phys. Chem. Ref. Data, 17:513.

Environmental Security Technology Certification Program (ESTCP). 2006. Field Demonstration and Validation of a New Device for Measuring Water and Solute Fluxes at Naval Base Ventura County (NBVC), Port Hueneme, CA. Washington, DC: U.S. Department of Defense.

Gates

D.D.

, Siegrist

R.L.

1995. In-situ chemical oxidation of trichloroethylene using hydrogen peroxide. J. Environ. Eng., 121:639.

Happel

A.M.

, Beckenbach

E.H.

, Halden

R.U.

1998. An Evaluation of MTBE Impacts to California Groundwater Resources. Report submitted to the California State Water Resources Control Board, Underground Storage Tank Program.

Jackson

M.L.

, Lim

C.H.

, Zelazny

L.W.

1986. Oxides, hydroxides and aluminosilicates. Klute

Methods of Soil Analysis: Part I. Madison, WI: American Society of Agronomy, 113.

Matta

, Hanna

, Chiron

2007. Fenton-like oxidation of 2,4,6-trinitrotoluene using different iron minerals. Sci. Total Environ., 385:242.

Miller

C.M.

, Bower

K.C.

, Ries

2003. Fenton's reagent degradation of 2,4-dinitrotoluene in water–acetone mixtures. Environ. Eng. Sci., 20:65.

Poulopoulos

S.G.

, Nikolaki

, Karampetsos

, Philippopoulos

C.J.

2008. Photochemical treatment of 2-chlorophenol aqueous solutions using ultraviolet radiation, hydrogen peroxide and photo-Fenton reaction. J. Hazard. Mater., 153:582.

10.

Watts

R.J.

, Haller

D.R.

, Jones

A.P.

2000. A foundation for the risk-based treatment of gasoline-contaminated soils using modified Fenton's reagent. J. Hazard. Mater., 76:73.

11.

Yeh

C.K.J.

, Chen

W.S.

, Chen

W.Y.

2004. Production of hydroxyl radicals from the decomposition of hydrogen peroxide catalyzed by various iron oxides at pH 7. Pract. Period. Hazard. Tox. Radioact. Waste Manage, 8:161.

12.

Yeh

C.K.J.

, Hsu

C.Y.

, Chiu

C.H.

, Huang

K.L.

2008. Reaction efficiencies and rate constants for the goethite-catalyzed Fenton-like reaction of NAPL-form aromatic hydrocarbons and chloroethylenes. J. Hazard. Mater., 151:562.

13.

Zepp

R.G.

, Faust

B.C.

, Hoigne

1992. Hydroxyl radical formation in aqueous reaction (pH 3–8) of iron (II) with hydrogen peroxide: the photo-Fenton reaction. Environ. Sci. Technol., 26:313.

14.

Zimbron

J.A.

, Reardon

K.F.

2005. Hydroxyl free radical reactivity toward aqueous chlorinated phenols. Water Res., 39:865.

Competitive Removal of Two Contaminants in a Goethite-Catalyzed Fenton Process at Neutral pH

Abstract

Abstract

Introduction

Materials and Methods

Goethite-catalyzed Fenton-like experiments

Analyses

Statistical analyses

Results and Discussion

Degradation in one-contaminant systems

Degradation in two-contaminant systems

Effects of H2O2 and goethite doses

Impacts of competition on remediation

Conclusion

Footnotes

Acknowledgment

Author Disclosure Statement

References

Effects of H₂O₂ and goethite doses