Nano-goethite-mediated transformation of anthracene derivatives under low moisture conditions

Abstract

Nano-goethite is widely present in soils and is active in mediating the transformation of soil-borne organic contaminants. However, little is known about the transformation efficacy and mechanisms of polycyclic aromatic hydrocarbons (PAHs) and their derivatives in the presence of nano-goethite under low moisture conditions. This study investigated the transformation of anthracene (ANT) and its amino-, methyl-, chloro- and nitro-derivatives mediated by nano-goethite at gravimetric water content of 2.3–14.5%, in combination with spectroscopic characterization and molecular computation. Nano-goethite exhibited a superior reactivity in the transformation of ANT and ANT derivatives, as a result of the single electron transfer reaction between Fe(III) surface sites and organic reactants. The C9 and C10 atoms of ANT and ANT derivatives are the reactive sites owing to their higher Fukui function and spin density values, which was confirmed by the identification of transformation products (9,10-anthraquinone and its derivatives). The reactivity was positively correlated with the highest occupied molecular orbital (HOMO) energy. Due to their lower HOMO energies, the ANT derivatives with electron withdrawing groups (e.g., –Cl and –NO₂) had slower transformation than ANT and the derivatives with electron donating groups (e.g., –NH₂ and –CH₃). Intriguingly, the transformation was suppressed with increasing water content, probably due to decreased sorption of hydrophobic ANT and ANT derivatives onto the increasingly moisturized nano-goethite surfaces. This study provides insights into the transformation of PAHs mediated by nano-goethite, and thus improves our understanding of the environmental fate of PAHs in soils.

---

Environmental significance

A significant knowledge gap exists in the transformation of organic contaminants by soil nanoparticles, especially in dry soil environments where biological transformation is severely limited. Nano-goethite is reactive in soils and could play an important role in chemical transformation of polycyclic aromatic hydrocarbons (PAHs). This study revealed that nano-goethite has a superior activity in the transformation of anthracene and its derivatives, which is attributed to the single electron transfer reaction between reactants and Fe(III) surface sites. Intriguingly, the transformation was suppressed with increasing water content, probably due to decreased sorption of hydrophobic reactants onto the increasingly moisturized nano-goethite surfaces. Hence chemical transformation of PAHs may be substantial in low-moisture surface layers of goethite-rich soils, and nano-goethite may be added to PAH-contaminated surface soils to augment chemical transformation.

Introduction

Many polycyclic aromatic hydrocarbons (PAHs) are carcinogenic, teratogenic, and mutagenic to humans.¹ Soil contamination by PAHs is pervasive worldwide from natural and anthropogenic pollution sources, such as incomplete fuel combustion, vegetation fires, volcanic exhalations, and leakage or spills of petroleum during exploration, refinery, transportation, and use.^2,3 Concentrations of PAHs in surface soils (e.g., 0–10 cm depth) are often greater than those at lower soil depth.^4,5 In fact, PAHs in surface soils can be more readily released to the atmosphere via volatilization and soil dusts, and transported to surface waters by soil erosion and surface runoff, resulting in their long-range transport and subsequent impact on more human populations farther away from contaminated sites.^6,7 PAHs in surface soils can also leach downward in soil profiles to contaminate groundwater and become available for root uptake by food crops, thus causing additional human exposure via drinking water and food consumption.⁸ Therefore, it is important to investigate the occurrence, fate, and distribution of PAHs in surface soils.

The environmental fate of PAHs in soils is controlled by volatilization, sorption, desorption, biodegradation, and chemical degradation.⁹ Although biodegradation is crucial to natural attenuation of PAHs, chemical degradation of PAHs has received increasing attention recently.^2,10 PAHs in surface soils may experience a greater degree of transformation and degradation due to solar irradiation, greater microbial activities, and greater fluctuation of temperature and moisture.^5,9 Specifically, large fluctuation of soil water content (WC from air dryness to saturation) and relative humidity (RH) occurs in the evaporative dry surface layer (e.g., 0–5 cm).^11,12 In many parts of the world, surface soil moisture is less than 5–10% in the dry season (International Soil Moisture Network, https://www.geo.tuwien.ac.at/insitu/data_viewer/) and the surface soils in many areas of Africa are extremely dry with soil moisture no more than 5%. For example, the gravimetric soil WC could be as low as <2.3% in the surface soils of a costal desert,¹³ and 3.8% in the surface soils of the Loess Plateau.¹⁴ The permafrost in Antarctica also has low gravimetric soil WC (<1–5%) and is mainly contaminated by PAHs from research stations and field camps.¹⁵ Furthermore, many places with predominantly dry surface soils are also major petroleum extraction areas,^16–18 and are thus prone to contamination by PAHs. Intriguingly, petroleum-contaminated soils have high water repellency due to the high hydrophobicity of petroleum, resulting in extremely low water retention.¹⁶ For example, the average gravimetric soil WC of 25 petroleum-contaminated bog soils was merely 7.8%.¹⁹ Interestingly, sorption of PAHs onto soil mineral surfaces is crucial to their chemical degradation and transformation, which may be decreased with increasing soil WC and RH due to adsorption of water film on mineral surfaces that would diminish the sorption of PAHs.^10,20,21 Thus, it is imperative to study the transformation and degradation of PAHs in experimental settings that are more representative of surface soil conditions (e.g., environmentally relevant soil WC and RH conditions).

Additionally, chemical transformation and degradation of PAHs are strongly affected by soil components such as metal oxides (particularly iron oxides) and cation-modified clay minerals (e.g., Fe(III)-modified montmorillonite).^10,20,21 Goethite (hydrous ferric oxide, α-FeOOH) is one of the most common iron oxides in many soils, including loess, desert and Antarctic soils.^22,23 Goethite of nano-, micron-, and mm-size occurs naturally in the environment,^24,25 and goethite nanoparticles (i.e., nano-goethite) may be formed by weathering,²⁴ or transformed from other relatively metastable iron oxyhydroxide nanoparticles. Recently nano-goethite has been collected and identified from soils in many countries (e.g. China, Japan, and England).^26,27 For example, nano-goethite with an equivalent size of 10 nm was found in a podzolic soil.²⁸ Nano-goethite of 5–60 nm was detected in 39 soils from western Australia,²⁹ which had an Fe concentration of 1.2–15.4% and a goethite percentage of 3–73% in the fraction of iron oxides. Goethite has a relatively high surface area and redox activities and can effectively mediate the transformation and degradation of many organic compounds such as phenols,³⁰ amines,³¹ and antibiotics.³² Nano-goethite exhibited even greater reactivity than larger-size goethite due to its larger surface area.^30,33 However, its role in the transformation and degradation of PAHs has rarely been investigated. The lack of research on this topic is probably the artifact of typical aqueous experimental systems rather than its geochemical significance. Indeed, adsorption of organic compounds by goethite is the major step leading to their chemical transformation and degradation.^2,30 In aqueous systems, PAHs and PAH derivatives barely adsorbed onto the goethite surface,³⁴ due to strong hydrogen bonding between the water film and goethite surface and subsequent exclusion of PAHs from being in contact with the goethite surface.^32,35 However, in the absence of water, PAHs and the goethite surface may actually form weak hydrogen bonding and surface complexes, as a result of the polarization of the π-system in PAHs by polar hydroxyl groups on the goethite surface and acceptance of protons by the π-system.^36,37 Interestingly, in water-free or dry systems, PAHs were degraded on manganese oxides and Fe(III)-bearing clay minerals.^2,38 Thus, we hypothesized that the chemical transformation and degradation of PAHs on the goethite surface could be substantial in dry surface soils and are drastically different from those in aqueous systems.

Additionally, substituted PAHs (i.e., PAH derivatives such as alkylated, aminated, chlorinated and nitrated PAHs) are also substantial in surface soils,^39,40 and some PAH derivatives are more carcinogenic than the unsubstituted PAHs.¹ For example, alkylated phenanthrenes were more toxic than unsubstituted phenanthrenes due to their higher binding affinity to the aryl hydrocarbon receptor (AhR).⁴¹ Aminated PAHs have higher genotoxicity than the parent PAHs because –NH₂ substitution groups can enhance the binding of PAHs with DNA.⁴² Chlorinated PAHs are structurally similar to dioxins and can induce greater AhR activities than the unsubstituted PAHs.^18,39 Nitrated PAHs have higher mutagenicity and carcinogenicity than their parent compounds.⁴³ Hence it is important to study the transformation of both parent PAHs and PAH derivatives as mediated by nano-goethite.

Therefore, this study aimed to elucidate the transformation efficacy and mechanisms of PAHs and PAH derivatives on the surfaces of nano-goethite (the most reactive goethite phase) in a range of WC and RH values mimicking dry surface soil conditions. Using 3-ring anthracene and its derivatives as model compounds, we first investigated the transformation kinetics, products, and pathways of anthracene (ANT), 9-aminoanthracene (9-NH₂-ANT), 9-methylanthracene (9-CH₃-ANT), 2-methylanthracene (2-CH₃-ANT), 9-chloroanthracene (9-Cl-ANT), 2-chloroanthracene (2-Cl-ANT), 9-nitroanthracene (9-NO₂-ANT) and 9,10-dichloroanthracene (9,10-DiCl-ANT). We further calculated electronic descriptors such as the highest occupied molecular orbital (HOMO) energy to determine the intrinsic reactivity of ANT and its 7 derivatives using a density functional theory (DFT) approach. Their electron spin density and Fukui function values were also calculated to identify reactive sites. The redox reaction on the nano-goethite surface was probed by X-ray photoelectron spectroscopy (XPS), Mössbauer spectroscopy, and electron paramagnetic resonance (EPR) spectroscopy. This study provided new insights into the chemical transformation and degradation of PAHs and PAH derivatives in surface soils, which is key to assessing the transport of PAHs to the atmosphere, surface waters, and groundwater, as well as to food crops and associated health risks to human populations through inhalation, drinking water, and dietary exposure.

Materials and methods

Chemicals

ANT, 9-CH₃-ANT, 2-CH₃-ANT, 9-Cl-ANT, 9-NO₂-ANT and 9,10-DiCl-ANT were purchased from Sigma-Aldrich (St. Louis, MO, USA), and 2-Cl-ANT and 9-NH₂-ANT from Tokyo Chemical Industry Co., Ltd., Tokyo, Japan. Their relevant physicochemical properties are provided in ESI† Table S1. Silica powder was obtained from Sigma-Aldrich (St. Louis, MO, USA), ground with an agate mortar, and passed through a 300-mesh sieve (48 μm opening). Chemicals and solvents used in this study were of analytical grade or high performance liquid chromatography (HPLC) grade.

Nano-goethite synthesis and characterization

Three nano-goethite samples were synthesized according to the methods of Schwertmann and Cornell⁴⁴ (labeled as Goe-01) and Zhang et al.⁴⁵ (labeled as Goe-02 and Goe-03). The synthesis procedure is described in ESI† Text S1. The synthesized samples were characterized by X-ray diffraction, Brunauer–Emmett–Teller (BET) specific surface area analysis with water vapor adsorption, transmission electron microscopy (TEM), and Fourier transform infrared (FTIR) spectroscopy, as detailed in S1.† The XRD spectra of Goe-01, Goe-02, and Goe-03 show the typical diffraction peaks of goethite (standard XRD spectrum PDF # 29-0713), confirming the crystalline structure of goethite (Fig. S1a†). The BET surface area of Goe-01, Goe-02 and Goe-03 was 48.7, 57.8 and 138 m² g⁻¹. These goethite nanoparticles appeared in either a single nanorod or bundles of nanorods (Fig. S1b†), and their particle size was determined using ImageJ software. The average diameter is 87 nm, 23 nm and 11 nm, and the average length is 495 nm, 255 nm and 42 nm for Goe-01, Goe-02 and Goe-03, respectively.

Transformation experiments

Prior to the transformation experiments, Goe-03 samples were first spread on uncovered 5 cm glass Petri dishes, and then dried at 105 °C for 12 h, followed by equilibration for 4 d in desiccators with the RH value controlled at 10% (by LiCl salt), 32% (by MgCl₂ salt), 54% (by MgNO₃ salt), and 80% (by KBr salt), resulting in their gravimetric WC of 4.8%, 6.7%, 8.5% and 14.5%, respectively. Additional Goe-01 and Goe-02 samples were also prepared at 10% RH, and their WC was 2.3% and 3.1%, respectively. Then 1 g nano-goethite samples (dry weight) were dosed with 5 mL of 0.1 mmol L⁻¹ ANT or each ANT derivative in acetone solution in 20 ml amber glass bottles with Teflon caps. The mixtures were continuously oscillated for 40 min to ensure complete mixing before transferring them to Petri dishes, followed by evaporation of acetone in a fume hood, resulting in a nominal initial concentration of 0.50 mmol kg⁻¹ for ANT or each ANT derivative. The nano-goethite samples loaded with ANT or individual ANT derivative were then kept in the RH-controlled desiccators, and placed in a dark artificial climate chamber at room temperature (20 °C). At pre-determined intervals, 0.05 g nano-goethite samples were collected, and the residual ANT or ANT derivatives were extracted immediately as follows. The collected samples were extracted with approximately 10 mL of a 1 : 1 mixture of acetone and dichloromethane in an ultrasonic bath for 30 min, and then centrifuged at 3000 rpm for 5 min. The extraction was repeated twice to ensure the full recovery of the residual ANT or ANT derivatives. The supernatants were pooled and then filtered by a syringe filter (hydrophobic, polytetrafluoroethylene filter unit, 0.22 μm). The filtrates were stored in amber HPLC vials in a refrigerator prior to analysis. The experiments were performed in triplicate.

The extraction efficiency was verified by the recovery experiments, in which the nano-goethite sample was spiked with 0.5 and 0.05 mmol kg⁻¹ of ANT or ANT derivatives, followed by extraction and analysis as abovementioned. The recovery experiments were repeated six times and the extraction efficiency was >96% (Table S2†), demonstrating the validity and efficacy of the extraction method.

Analytical methods

ANT and ANT derivatives in the filtrates were analyzed using a Shimadzu LC-20AT high performance liquid chromatograph (HPLC) fitted with a 250 mm × 4.6 mm reverse-phase C18 column. The mobile phase is a mixture of methanol and water (85 : 15 by volume) at a flow rate of 1 mL min⁻¹ and 40 °C. The sample injection volume is 20 μL and the wavelength of the ultraviolet detector was 254 nm. The limit of detection for ANT and ANT derivatives was in the range of 0.05–2.71 μg L⁻¹ (Table S2†). The transformation products of ANT and ANT derivatives were also analyzed using the above method with a detection wavelength of 323 nm. The extraction solution was concentrated under flowing nitrogen and the transformation products were identified using a GC-MS (450-GC, 320-MS; Bruker) equipped with a DB-5MS capillary column (30 m by 0.25 mm by 0.25 μm). Helium gas was used as the carrier at a flow rate of 1 mL min⁻¹ with splitless injection at 230 °C. The temperature of the column was initially kept at 100 °C for 2 min, then increased to 300 °C at a rate of 5 °C min⁻¹, and finally held at 300 °C for 5 min. The mass spectrometer was operated in the electron ionization mode at 70 eV with a source temperature of 260 °C, with continuous scanning from 50 m/z to 500 m/z. The product identification was performed by comparing the mass spectra with the NIST/EPA/NIH Mass Spectrum Library. Meanwhile, the mass spectra of standard compounds were obtained and compared with the mass fragments of the transformation products.

Furthermore, an additional set of nano-goethite samples reacted with ANT and ANT derivatives for 4 d was used for XPS and Mössbauer measurements. The oxidation state of Fe on the nano-goethite samples was analyzed with XPS (K-Alpha XPS System, Thermo Fisher Scientific Inc.). The Mössbauer measurements were performed with a constant acceleration drive system (WissEL) at 298 K using a 57 Co/Rh γ-radiation source. The Mössbauer spectra were fitted using MossWinn software (version 4.0 Pre) with a Lorentzian line shape. Another set of nano-goethite samples was reacted with ANT for 0.5, 1, and 2 d, and with ANT derivatives for 1 d, respectively. The intermediate organic radicals of ANT and ANT derivatives in these samples were detected by EPR measurements using a Bruker E500 EPR spectrometer. Briefly, 200 mg samples were placed into a 4 mm quartz tube and placed in the EPR resonator. The detailed operation parameters of the EPR measurements are provided in ESI† Text S1.

Finally, Fe(II) was extracted by mixing 0.05 g nano-goethite sample with 4 mL 1 mol L⁻¹ HCl for 8 h with continuous purging of N₂. Then, 2 mL ammonium acetate buffer and 0.2 mL ferrozine were added. The suspensions were agitated for 2 h and filtered through a syringe filter (hydrophilic, polyethersulfone filter unit, 0.45 μm). The concentration of ferrozine-complexed Fe(II) was measured using a UV spectrophotometer (UVmini-1240, Shimadzu) at 562 nm.

Transformation kinetics

The kinetic transformation data were fitted to the first-order kinetic model (eqn (1)):

C = C₀e^−kt (1)

where C₀, C, k and t are the initial concentration (mmol kg⁻¹), the concentration at time t (mmol kg⁻¹), the transformation rate constant (d⁻¹) and reaction time (t), respectively. The k value was estimated by fitting the experimental transformation kinetic data to the first-order kinetic equation.

Theoretical calculations

Theoretical calculations can provide important information on the reactivity of PAHs. It is known that the transformation of PAHs involving electron transfer is related to their electronic properties such as the highest occupied molecular orbital (HOMO) and ionization potential (IP).⁴⁶ The reactivity of PAHs increases with decreasing HOMO energies and PAHs with IP above 7.5–7.6 cannot be degraded by Fe(III)–montmorillonite.¹⁰ Additionally, substitution on the aromatic ring of PAHs can alter their electronic properties.⁴⁷ Electron donating groups such as –CH₃, –OH and –NH₂ groups could increase the electron density of the aromatic ring and increase the HOMO energy. Conversely, electron withdrawing groups such as –NO₂ and –Cl groups can lower the HOMO energy.⁴⁸ A total of 6 electronic descriptors were selected, and are listed in Table 1. Previous studies found that these descriptors impact the transformation kinetics at different molecular regions.² Thus, molecular computation was performed to mechanistically relate the experimental transformation of ANT and its derivatives with their structures and theoretical reactivity. The molecular structures of ANT and ANT derivatives were drawn by ChemBioOffice (CambridgeSoft, Cambridge, USA), and DFT calculations were conducted with Gaussian 09 W. Conformation optimization and calculation of the HOMO, the lowest unoccupied molecular orbital (LUMO), HOMO–LUMO, IP, electron affinity (EA) and molecule hardness were performed at the B3LYP/6-311G** basis. Afterwards the obtained “.pdb” file recording the information of the molecular structures was used to calculate the HOMO, LUMO, HOMO–LUMO, IP, and EA by Multiwfn 3.7. HOMO energies were drawn by VMD 1.9.3.

Table 1 Electrochemical properties of ANT and its derivatives calculated by density functional theory

Compounds	HOMO/eV	LUMO/eV	HOMO–LUMO/eV	IP/eV	EA/eV	Hardness/eV
9-NH2-ANT	−5.0297	−1.8043	3.2254	6.4045	0.3425	6.0620
9-CH3-ANT	−5.4283	−1.9521	3.4762	6.9088	0.5335	6.3753
2-CH3-ANT	−5.4674	−1.884	3.5834	6.9471	0.4377	6.4738
ANT	−5.5449	−1.9778	3.5671	7.0653	0.5232	6.5421
2-Cl-ANT	−5.7176	−2.1777	3.5399	7.1768	0.7679	6.4088
9-Cl-ANT	−5.6595	−2.2104	3.4491	7.1267	0.7902	6.3365
9,10-DiCl-ANT	−5.7651	−2.4356	3.3295	7.1839	1.0469	6.1369
9-NO2-ANT	−6.0599	−2.6742	3.3857	7.5016	0.7376	6.7639

---

To identify the reactivity sites of ANT and ANT derivatives in their transformation facilitated by nano-goethite, their Fukui function values and spin density distributions of compound radicals were also calculated. The Fukui function can show the most electrophilic and nucleophilic regions within a molecule. The Fukui function f⁻ represents the electrophilic attack ability when the molecules lose electrons, and a large value of f⁻ indicates high reactivity.⁴⁹ Thus, f⁻ could be used to identify the locations within a molecule that are most susceptible to attack by Fe(III) of nano-goethite. The distribution of spin densities was analyzed through removing an electron from a molecule by setting the total charge of the molecule as +1 and the spin multiplicity as 2.

Results and discussion

Transformation kinetics

The transformation kinetics of ANT and ANT derivatives in the presence of nano-goethite were well fitted by the first-order kinetic model (Fig. 1, S3 and S5†) with R² ranging between 0.841 and 0.998. The fitted parameters are provided in ESI Tables S3 and S4.† The fitted k values followed the order of 9-NH₂-ANT > 9-CH₃-ANT > 2-CH₃-ANT > ANT > 2-Cl-ANT > 9-Cl-ANT > 9-NO₂-ANT > 9,10-DiCl-ANT (Fig. 1c and Table S3†). However, in the control treatment with the non-reactive silica powder, no significant transformation of ANT and ANT derivatives was observed (Fig. S4†) except for 9-NH₂-ANT. The transformation of 9-NH₂-ANT was much slower with silica powder than with nano-goethite (Fig. 1a and S4†). For example, 9-NH₂-ANT at an initial concentration of 0.5 mmol kg⁻¹ was fully transformed by nano-goethite within 1 d, but only lost 10% with silica powder (probably due to oxidation by oxygen).⁵⁰ The k value of ANT and ANT derivatives fitted from the transformation kinetics (Fig. S5†) increased with decreasing nano-goethite particle size. For example, the k value of ANT was 0.390 d⁻¹ for Goe-01 at a WC of 2.3%, 0.610 d⁻¹ for Goe-02 at a WC of 3.1%, and 1.536 d⁻¹ for Goe-3 at a WC of 4.8%, respectively (Tables S3 and S4†).

Fig. 1 (a and b) Transformation kinetics of 0.5 mmol kg⁻¹ ANT and ANT derivatives on nano-goethite (Goe-03) at 20 °C and water contents of 4.8% (a) and 4.8%, 6.7%, 8.5%, and 14.5% (b). (c) The first-order rate constants as a function of water content.

For the Goe-3 samples, increasing the WC from 4.8% to 14.5% significantly decreased the transformation of ANT and ANT derivatives (Fig. 1c). At a WC of 14.5%, the transformation of ANT, Cl-ANT and NO₂-ANT on goethite surfaces was almost completely suppressed. These results are in agreement with a previous study, reporting that increasing RH (or WC) resulted in a rapid decrease of PAH degradation on Fe(III)–montmorillonite.¹⁰ The polar surface OH groups of goethite interact strongly with water molecules to form hydrogen bonding with an interaction energy ranging from −15.2 to −24.6 kcal mol⁻¹.⁵¹ However, the interaction energies of PAHs and goethite are substantially less negative, such as −10.3 kcal mol⁻¹ (anthracene), −7.6 kcal mol⁻¹ (naphthalene), −5.9 kcal mol⁻¹ (phenanthrene), and −7.5 kcal mol⁻¹ (pyrene).³⁶ Unfortunately, the interaction energies between anthracene derivatives and goethite have not been reported. Nonetheless, the interaction of goethite with water should be more favorable than that with ANT and possibly ANT derivatives. Indeed, Angove et al.³⁴ reported nearly zero adsorption of ANT onto the goethite surface in aqueous solution. As per our calculation the surface of Goe-03 nano-goethite is covered with a monolayer water film at a WC of 4.8%, but with a multi-layer water film at WCs of 6.7%, 8.5% and 14.5% (see detailed calculation in ESI† S1). Therefore, the inhibition effect induced by water molecules may be attributed to the competition between water molecules and PAHs for adsorption onto the nan-goethite surfaces, and ANT and ANT derivatives adsorbed on the nano-goethite surfaces are increasingly displaced at higher WC.^35,36 Consequently, the electron transfer reaction rate and ANT transformation rate on the goethite were decreased with increasing WC. Therefore, WC is the crucial factor controlling the transformation of ANT and ANT derivatives facilitated by nano-goethite.

Reactivity sites of nano-goethite

As shown in Fig. 2a, Fe(II) concentrations initially increased rapidly and then plateaued over time, inversely mirroring the concentrations of ANT and its derivatives (Fig. 1). This phenomenon suggests that the transformation of ANT and its derivatives was accompanied by the reduction of Fe(III) to Fe(II). Previous studies showed that the degradation of organic compounds with goethite could be attributed to the electron transfer from organic molecules to Fe(III).^30,33 Interestingly, the molar concentrations of the produced Fe(II) (0.88–2.18 mmol kg⁻¹ calculated by subtracting 3.41 mmol kg⁻¹ at day 0) are several times greater than the molar concentrations of transformed ANT and its derivatives (0.13–0.53 mmol kg⁻¹), suggesting that multiple electrons are transferred from each molecule of ANT and its derivatives to Fe(III) of nano-goethite. It was noted that the concentration of Fe(II) slightly decreased at the final stage of the reaction, probably due to its oxidation to Fe(III) upon air exposure.³⁰ For example, Lin et al. observed an accumulation of Fe(II) during the degradation of bisphenol A on goethite, but the concentration of Fe(II) slightly decreased after 12 h.⁵² The interactions between PAHs and goethite originate from the formation of a precursor complex by weak hydrogen bonding between the π-system of PAHs and ≡FeOH groups, followed by the electron transfer from PAHs to ≡FeOH groups.^30,36 The FTIR spectra of nano-goethite were determined (Fig. S2†), and the ≡FeOH stretching peaks at 3125, 905 and 795 cm⁻¹ indicate that nano-goethite has abundant reactive sites.⁵³ Previous studies showed the density of ≡FeOH groups in goethite ranged between 0.8 and 18 sites per nm².²⁵ In this study, the BET surface area of Goe-03 nano-goethite is 138 m² g⁻¹, and the density of reactive sites ≡FeOH in Goe-03 can be approximately estimated to be 182–4121 mmol kg⁻¹, which is much higher than the reactive sites of hematite (18.8 mmol kg⁻¹), Fe(III)-bearing illite (37 mmol kg⁻¹) and Fe(III)-bearing pyrophyllite (38 mmol kg⁻¹).^10,38 Therefore, the nano-goethite has significant amounts of reactive sites and exhibits high reactivity in transforming/degrading ANT and its derivatives.

Fig. 2 (a) Fe(II) concentration as a function of reaction time during the transformation experiments of ANT and its derivatives by nano-goethite (Goe-03). (b) The XPS spectra of the original nano-goethite and nano-goethite after the transformation of ANT and its derivatives. (c) The Mössbauer spectra of the original nano-goethite and nano-goethite after ANT and 9-NO₂-ANT transformation. Fitting parameters and relative abundances of Mössbauer spectra are presented in Table S5.†

To further investigate the reduction of Fe(III) during the transformation, the XPS spectra of the nano-goethite samples were obtained before and after the reaction with ANT and its derivatives. As shown in Fig. 2b, the presence of Fe(III) on the nano-goethite could be confirmed by the peaks of Fe 2p_1/2 (725.05) and Fe 2p_3/2 (711.43).⁵⁴ These peaks were shifted by 0.36–0.81 to lower binding energy after the reaction with ANT and its derivatives. The significant shift indicated that Fe(III) accepts electrons from ANT and ANT derivatives, resulting in the reduction of Fe(III).^54,55 In addition, the Mössbauer spectra of the nano-goethite samples show the characteristic six peaks for goethite (Fig. 2c).⁵⁶ The fitting of the Mössbauer spectra of the original nano-goethite indicated that Fe(II) accounted for 2.4% of the total Fe intensity. After the reaction with ANT and 9-NO₂-ANT, the proportion of Fe(II) increased to 5.2% and 4.2%, respectively. These results collectively indicated that Fe(III) induced the transformation of ANT and its derivatives by acting as an electron acceptor.

Reactivity sites of ANT and ANT derivatives

The optimized structure and Fukui function f⁻ values of the parent ANT and 9-NO₂-ANT are shown in Fig. 3a, and the data of the other ANT derivatives are displayed in Fig. S6a.† The f⁻ values of atoms C9 and C10 are far greater than those of other C atoms in ANT and 9-NO₂-ANT. Thus, electron extraction is more likely to occur at C9 and C10 positions and C9 and C10 atoms could lose one electron to Fe(III) with the formation of a radical intermediate. As a result, ANT and 9-NO₂-ANT were degraded via the cleavage of the C9–H/NO₂ bond, C10–H bond and benzene ring. Generally, the spin density represents the probability of the presence of an unpaired electron.^57,58 The calculated spin density of ANT and 9-NO₂-ANT radicals is shown in Fig. 3c and the data for the other ANT derivatives are displayed in Fig. S6b.† The spin density values are at C9 (0.270) and C10 (0.270) in ANT and at C9 (0.261) and C10 (0.254) in 9-NO₂-ANT, where the unpaired electron is most likely distributed. Thus, the calculation of both the Fukui function and spin density demonstrates that C9 and C10 are the reactive sites of ANT and 9-NO₂-ANT. Similarly, the Fukui function and spin densities of C9 and C10 of the other ANT derivatives are greater than those of other atoms, suggesting that C9 and C10 atoms are the reactive sites in these ANT derivatives.

Fig. 3 (a) Spin densities of ANT and 9-NO₂-ANT. (b) Fukui function of ANT and 9-NO₂-ANT. (c) The spin values of the C10 or C9 atom of ANT and its derivatives. (d) Fukui function of the C10 or C9 atom of ANT and its derivatives.

Interestingly, the Fukui function and spin density values of C9 and C10 in the parent ANT are equivalent, but were changed by the introduction of substituent groups. Ortho/para directing groups (–CH₃, –NH₂ and –Cl groups) mainly promote substitution at the ortho or para position. As shown in Fig. 3b, the spin densities and f⁻ values of para position C10 in 9-NH₂-ANT, 9-CH₃-ANT and 9-Cl-ANT were higher than those of C9. Therefore, C10 was first attacked by Fe(III). Meta directing groups such as the –NO₂ group mainly attack at the meta position. However, the f⁻ and spin density values of C10 and C9 in 9-NO₂-ANT are also greater than those of other atoms including the meta position atoms. The f⁻ value of C10 is greater than that of C9 in 9-NO₂-ANT, but the spin density value of C9 is greater. In view of the steric hindrance, C10 is more easily exposed to Fe(III) and loses one electron.^57,59 The f⁻ and spin density values of C9 and C10 in 2-Cl-ANT and 2-CH₃-ANT show a slight change compared with ANT, and both values of C9 are higher than those of C10. In conclusion, C10 may be first attacked in 9-NH₂-ANT, 9-CH₃-ANT, 9-Cl-ANT and 9-NO₂-ANT, but C9 may be the first attack site in 2-Cl-ANT and 2-CH₃-ANT.

Influence of electrochemical properties of ANT and ANT derivatives

The IP values of ANT and seven ANT derivatives are all less than 7.55, which is considered as the threshold of PAH biodegradation and chemical degradation.¹⁰ Thus ANT and ANT derivatives can be degraded by nano-goethite, which is supported by the experimentally observed transformation kinetics (Fig. 1a). Additionally, ANT and its derivatives have different molecular orbital structures and HOMO energies (Fig. 4a). The HOMO energy for 9-NH₂-ANT, 9-CH₃-ANT, 2-CH₃-ANT, ANT, 2-Cl-ANT, 9-Cl-ANT, 9-NO₂-ANT and 9,10-DiCl-ANT was −5.030, −5.428, −5.467, −5.545, −5.030, −5.428, −5.467 and −5.545 eV, respectively. NH₂-ANT and CH₃-ANT have a higher HOMO energy than Cl-ANT and NO₂-ANT, suggesting that electron donating groups such as –NH₂ and –CH₃ could improve the nucleophilicity of the aromatic ring. In contrast, electron withdrawing groups (–Cl and –NO₂) could decrease the electron density of the aromatic ring and make it conducive to electrophilic attack.^48,60 The HOMO energy 9,10-DiCl-ANT with two –Cl groups are more negative than that of 2-Cl-ANT and 9-Cl-ANT, indicating that the number of substituents also influences the electronic properties and reactivity of ANT and its derivatives. The substitution at different atom positions has a minor effect on the HOMO energy of ANT derivatives, but primarily affects the electrophilic attack ability of the C atom at different sites. For example, the HOMO energies of 9-CH₃-ANT (−5.428 eV) and 2-CH₃-ANT (−5.467 eV) only differ by 0.7%, but the spin density values of C10 in 9-CH₃-ANT (0.281) and 2-CH₃-ANT (0.249) differ by 12.9% (Fig. S6b†).

Fig. 4 (a) HOMO energies of ANT and its derivatives. (b) The first-order rate constants as a function of HOMO energy.

There is a positive relationship between the k values and HOMO energies (Fig. 4b), suggesting that the transformation rate depends on the HOMO energies and ANT and its derivatives are probably transformed through electrophilic substitution. Indeed, due to their more positive HOMO energy values (Fig. 4a), ANT and its derivatives with electron donating groups (e.g., –NH₂ and –CH₃) were transformed faster than the ANT derivatives with electron withdrawing groups (e.g., –Cl and –NO₂) (Fig. 1a and Table S3†). Similarly, Tong et al. found that the HOMO energy of phenol, NO₂–phenol and Cl–phenol is −5.01, −5.06 and −5.17 eV, respectively, corresponding to decreasing photocatalytic degradation rate constants in the order of phenol > NO₂–phenol > Cl–phenol.⁶¹ Furthermore, the reactivity of ANT derivatives is also influenced by the position and number of substitution groups, in addition to their electrochemical properties,⁴⁸ as indicated by slight scattering in the linear regression (Fig. 4b, R² = 0.866). Despite the higher HOMO energy of 9-Cl-ANT than that of 2-Cl-ANT, the k value of 2-Cl-ANT is greater than that of 9-Cl-ANT (Fig. 4a and Table S3†). This observation may be because –Cl can decrease the electron density more greatly at C9 than that at C2 (Fig. S6†). Similarly, 9,10-DiCl-ANT also has a lower k value than 9-NO₂-ANT, regardless of its high HOMO value. Due to the influence of both Cl atoms at C9 and C10, the electron density of C9 and C10 in 9,10-DiCl-ANT is lower than that of 9-NO₂-ANT with one substituent group. In fact, when only evaluating the ANT derivatives with the substitution group at C9 their transformation rate constants have a better linear correlation with their HOMO values (R² = 0.975, Fig. S7a†). In addition, Hammett constants of substitution groups are often used to evaluate the difficulty of the electrophilic reaction of aromatic hydrocarbon derivatives,⁶² and the Hammett constant of –NH₂, –CH₃, –Cl, and, –NO₂ is −0.66, −0.17, 0.23, and 0.78, respectively.⁶³ The transformation rate constants of ANT derivatives and the Hammett constants of the four substitution groups are linearly correlated (R² = 0.970) (Fig. S7b†), suggesting that the Hammett constant is a good indicator of the electrophilic reaction.

Transformation products and reaction pathway

Fig. 5a displays the GC-MS spectra of ANT and its derivatives and the corresponding transformation products. The transformation product of ANT is 9,10-anthraquinone (AQ), which was also identified as the major product of ANT transformation by manganese oxides and Fe(III)-bearing clay minerals.^2,10 Additionally, AQ was also detected as the transformation product of 9-NH₂-ANT, 9-CH₃-ANT, 9-Cl-ANT, 9-NO₂-ANT and 9,10-DiCl-ANT, whereas the transformation products of 2-Cl-ANT and 2-CH₃-ANT are 2-Cl-9,10-anthraquinone (2-Cl-AQ) and 2-CH₃-9,10-anthraquinone (2-CH₃-AQ). These results are consistent with the identified reactivity sites of ANT and ANT derivatives by theoretical calculations. Electron extraction more likely occurs at C9 and C10 with higher Fukui functions and spin densities. Fig. S8† shows the GC-MS spectra of ANT, 2-Cl-ANT and 2-CH₃-ANT during the transformation experiment at days 0, 1, 4 and 8, illustrating the decreased peak of the parent compound and the appearance of the product peak. Indeed, the parent compound concentration decreased and the product concentration increased with time (Fig. 5b). The summed concentration of each parent compound and product pair is close to the initial input concentration (Fig. 5b). Also, extending the reaction time to 8 days found no new product peaks and no significant change of observed product concentrations. Thus, AQ, 2-Cl-AQ and 2-CH₃-AQ could be the final products for the transformation of ANT, 2-Cl-ANT and 2-CH₃-ANT by nano-goethite. Since AQ cannot be further degraded, AQ could be also the final product for the transformation of 9-NH₂-ANT, 9-CH₃-ANT, 9-Cl-ANT, 9-NO₂-ANT and 9,10-DiCl-ANT. Previous studies reported that AQ is less toxic than ANT, e.g., the toxicity of ANT for Pimephales promelas fishes is 7300 times higher than that of AQ.² Nonetheless, AQ is more soluble, potentially leading to greater bioavailability to organisms (including PAH-degrading microorganisms), as well as greater transport to receiving water bodies.^2,38 Thus, the transformation of ANT by nano-goethite may be beneficial to remediating anthracene-contaminated soils due to the significantly lower toxicity of AQ than ANT. Nonetheless, the environmental risks and benefits of this process should be further evaluated by taking into account the differences in the persistence, bioaccumulation, and mobility of AQ and ANT in the future.

Fig. 5 (a) GC-MS spectra and identified transformation products for ANT and ANT derivatives after 3 h (9-NH₂-ANT, 9-CH₃-ANT, and 2-CH₃-ANT), 1d (ANT, 2-Cl-ANT, and 9-Cl-ANT) or 2d (9-NO₂-ANT and 9,10-DiCl-ANT) on nano-goethite. (b) Concentrations of the parent compound and transformation products for ANT and its derivatives on nano-goethite (Goe-03) with a water content of 4.8% at 20 °C in 4 days. (c) Proposed transformation mechanisms for ANT and ANT derivatives by nano-goethite.

Finally, it has been proposed that the transformation of organic compounds by goethite involves: (i) formation of a precursor complex between organic compounds and goethite; (ii) electron transfer from the organic compounds to goethite, and subsequent formation of organic radicals and Fe(II); (iii) further reactions induced by organic radicals.^33,52 Possible pathways for the transformation of ANT and ANT derivatives are presented in Fig. 5c. In this study, transformation of ANT and ANT derivatives is initiated by the formation of precursors (pathway I), in which one electron is transferred from PAHs to Fe(III), followed by the formation of Fe(II) and an organic radical (radical A). For 2-Cl-ANT and 2-CH₃-ANT, the reactivity of C9 is higher than that of C10. Thus, C9 of radical A is first attacked by H₂O, resulting in the formation of radical I-B. Radical I-B is deprotonated to form 9-hydroxy-2-CH₃/Cl-ANT, which is further tautomerized to the thermodynamically favored 2-CH₃/Cl-anthrone. 2-CH₃/Cl-anthrone then loses one additional electron and forms radical I-C. Afterwards radical I-C was attacked by H₂O to produce 2-CH₃/Cl-AQ (pathway II-i). In contrast, C10 in radical A of the other ANT derivatives first loses one electron and is attacked by H₂O, yielding radical II-B. Radical II-B undergoes a series of reactions similar to the reactions of radical I-B, resulting in radical II-C and other intermediates, which can be further transformed to AQ and AQ derivatives (pathway II-ii). Interestingly, the signals of organic radicals were not detected by the EPR measurements (Fig. S9†). It appears that any formed organic radicals may be immediately hydrolyzed by water molecules and/or oxidized by Fe(III) of nano-goethite and oxygen during the transformation of ANT and its derivatives on nano-goethite surfaces.^38,55,64 Indeed, the much greater molar concentrations of the produced Fe(II) than the transformed ANT and its derivatives confirm the transfer of multiple electrons from each organic molecule to Fe(III) of nano-goethite (Fig. 2a). Additionally, due to the higher oxidation potential of hematite (Fe₂O₃) than other metal oxides (e.g., CuO and ZnO), the yield of organic radicals with hematite was lower.^55,64,65 Considering the higher oxidation potential of goethite (0.727 V) than that of hematite (0.655 V),⁶⁶ the organic radicals could be even of lower yield and less stable with goethite than with hematite. It was also reported that catechol forms mononuclear monodentate and binuclear bidentate complexes with goethite (consequently more directly connected reactive Fe(III) sites), compared to the formation of only a mononuclear bidentate complex with hematite,⁶⁷ again resulting in lower-yield and less stable organic radicals. Thus, despite the reported life-time of PAH radicals (e.g., anthracene and 9-methylanthracene) on the microsecond scale,⁶⁸ it is postulated that the organic radicals during the nano-goethite-mediated transformation of PAHs immediately participate in the transformation reactions or are consumed by oxygen, which renders their detection by EPR unlikely.

Conclusions

This study provided several important findings on the transformation of PAHs by nano-goethite in a range of water contents relevant to surface soils. Nano-goethite has a superior reactivity in the transformation of ANT and its derivatives (as representative PAHs) in the dark without light irradiation. Intriguingly, water content was the crucial factor controlling the transformation of PAHs and the transformation rate decreased substantially with increasing water content. These results fill the knowledge gap on the transformation of PAHs by soil nanoparticles, especially in dry soil environments where biological transformation is severely limited. Dry surface soils are widespread in the environment and nano-goethite occurs naturally in soils. The transformation of PAHs by nano-goethite will most likely occur in the surface layers of goethite-rich soils with low water content (e.g. less than 14.5%), but would be diminished in wetter subsurface soils, groundwater aquifers, and sediments. Thus, this study provides valuable insights into the fate of PAHs in natural soils with important implications for remediation of contaminated soils. Specifically, nano-goethite may be potentially added to PAH-contaminated surface soils as a catalyst for remediation purposes, but it may be necessary to control in situ soil moisture to optimize the transformation efficacy. In order to deepen our understanding of nano-goethite-mediated transformation of PAHs in natural soils and better develop soil remediation strategies, future studies should investigate the transformation of PAHs in natural goethite-rich soils or soils amended with nano-goethite under representative environmental conditions (especially in a range of water contents). Fundamentally, electron transfer between nano-goethite and PAHs is key to the transformation of PAHs. The initial reaction occurs at the C9 and C10 of PAHs with higher f⁻ and spin density values, and the reactivity of these sites is influenced by the introduction of substitution groups in the PAH molecular structure. The transformation rate constants were positively correlated with the HOMO values of PAHs. GC-MS analysis suggested that ANT and ANT derivatives were transformed into AQ and AQ derivatives, which are less toxic but more soluble and bioavailable. This study revealed that both unsubstituted and substituted PAHs are efficiently transformed by nano-goethite following a similar molecular mechanism, and the underlying reactivity is correlated well with electronic descriptors such as HOMO, f⁻ and spin density values, and Hammett constants. Therefore, molecular computation tools may be very useful to assess the transformation potential of a plethora of unsubstituted and substituted PAHs. Future studies should evaluate the overall environmental risks of PAHs and their transformed products by integrating toxicity, persistence, bioaccumulation, and mobility for better development of soil remediation strategies.

Author contributions

Pengfei Cheng: design of study, date acquisition, formal analysis, writing – original draft, manuscript revising. Wei Zhang: manuscript revising, formal analysis. Xuqiang Zhao: investigation, methodology, manuscript revising. Bing Yang: investigation, software. Yanzheng Gao: conceptualization, supervision, funding acquisition, manuscript revising, editing.

Conflicts of interest

The authors declare no competing financial interest.

Acknowledgements

This work was financially supported by the National Science Fund for Distinguished Young Scholars (41925029), the Jiangsu Agricultural Science and Technology Innovation Fund (CX1009), and the National Natural Science Foundation of China (41877125).

References

1. H. D. Yu, B. Yang, M. G. Waigi, F. Peng, Z. K. Li and X. J. Hu, The effects of functional groups on the sorption of naphthalene on microplastics, Chemosphere, 2020, 261, 127592.
2. M. Villalobos, M. Carrillo-Cardenas, R. Gibson, N. Ruth Lopez-Santiago and J. A. Morales, The influence of particle size and structure on the sorption and oxidation behaviour of birnessite: II. Adsorption and oxidation of four polycyclic aromatic hydrocarbons, Environ. Chem., 2014, 11, 279–288.
3. J. Nam, G. Thomas, F. Jaward, E. Steinnes, O. Gustafsson and K. Jones, PAHs in background soils from Western Europe: Influence of atmospheric deposition and soil organic matter, Chemosphere, 2008, 70, 1596–1602.
4. X. Liu, Z. Bai, Q. Yu, Y. Cao and Z. Wei, Polycyclic aromatic hydrocarbons in the soil profiles (0–100 cm) from the industrial district of a large open-pit coal mine, China, RSC Adv., 2017, 7, 28029–28037.
5. L. Han, J. H. Bai, Z. Q. Gao, W. Wang, D. W. Wang, B. S. Cui and X. H. Liu, Polycyclic aromatic hydrocarbons (PAHs) in surface soils from reclaimed and ditch wetlands along a 100-year chronosequence of reclamation in a Chinese estuary: Occurrence, sources, and risk assessment, Agric., Ecosyst. Environ., 2019, 286, 106648.
6. Z. Taihui, Y. Ran and L. Chen, Polycyclic aromatic hydrocarbons (PAHs) in rural soils of Dongjiang River Basin: Occurrence, source apportionment, and potential human health risk, J. Soils Sediments, 2014, 14, 110–120.
7. I. C. Yadav and N. L. Devi, Nitrated- and oxygenated-polycyclic aromatic hydrocarbon in urban soil from Nepal: Source assessment, air-soil exchange, and soil-air partitioning, Ecotoxicol. Environ. Saf., 2021, 211, 111951.
8. Y. Li, L. Long, J. Ge, L. X. Yang, J. J. Cheng, L. X. Sun, C. Y. Lu and X. Y. Yu, Presence, distribution and risk assessment of polycyclic aromatic hydrocarbons in rice-wheat continuous cropping soils close to five industrial parks of Suzhou, China, Chemosphere, 2017, 184, 753–761.
9. M. Marques, D. Cervello, M. Mari, J. Sierra, M. Schuhmacher, J. L. Domingo and M. Nadal, The role of iron oxide on the photodegradation of polycyclic aromatic hydrocarbons: Characterization and toxicity, Polycyclic Aromat. Compd., 2018, 1–11.
10. H. Jia, J. Zhao, L. Li, X. Li and C. Wang, Transformation of polycyclic aromatic hydrocarbons (PAHs) on Fe(III)-modified clay minerals: Role of molecular chemistry and clay surface properties, Appl. Catal., B, 2014, 154, 238–245.
11. K.-U. Goss and M. Madliger, Estimation of water transport based on in situ measurements of relative humidity and temperature in a dry Tanzanian soil, Water Resour. Res., 2007, 43, W05433.
12. A. F. Davila, B. Gomez-Silva, A. de los Rios, C. Ascaso, H. Olivares, C. P. McKay and J. Wierzchos, Facilitation of endolithic microbial survival in the hyperarid core of the Atacama Desert by mineral deliquescence, J. Geophys. Res. Biogeosci, 2008, 113, G01028.
13. N. Agam and P. R. Berliner, Diurnal water content changes in the bare soil of a coastal desert, J. Hydrometeorol., 2004, 5, 922–933.
14. X. Yao, B. Fu, Y. Lu, R. Chang, S. Wang, Y. Wang and C. Su, The multi-scale spatial variance of soil moisture in the semi-arid Loess Plateau of China, J. Soils Sediments, 2012, 12, 694–703.
15. J. M. Aislabie, M. R. Balks, J. M. Foght and E. J. Waterhouse, Hydrocarbon spills on Antarctic soils: Effects and management, Environ. Sci. Technol., 2004, 38, 1265–1274.
16. R. M. Abousnina, A. Manalo, J. Shiau and W. Lokuge, Effects of light crude oil contamination on the physical and mechanical properties of fine sand, Soil Sediment Contam., 2015, 24, 833–845.
17. X. P. Wang, Y. X. Pan, Y. F. Zhang, D. Q. Dou, R. Hu and H. Zhang, Temporal stability analysis of surface and subsurface soil moisture for a transect in artificial revegetation desert area, China, J. Hydrol., 2013, 507, 100–109.
18. C. Z. Jia, M. Zheng and Y. F. Zhang, Unconventional hydrocarbon resources in China and the prospect of exploration and development, Pet. Explor. Dev., 2012, 39, 139–146.
19. Y. Wang, J. Feng, Q. Lin, X. Lyu, X. Wang and G. Wang, Effects of crude oil contamination on soil physical and chemical properties in momoge wetland of china, Chin. Geogr. Sci., 2013, 23, 708–715.
20. C. Clarke, J. Tourney and K. Johnson, Oxidation of anthracene using waste Mn oxide minerals: The importance of wetting and drying sequences, J. Hazard. Mater., 2012, 205, 126–130.
21. S. C. Marie-Rose, T. Belin, J. Mijoin, E. Fiani, M. Taralunga, F. Nicol, X. Chaucherie and P. Magnoux, Catalytic combustion of polycyclic aromatic hydrocarbons (PAHs) over zeolite type catalysts: Effect of water and PAHs concentration, Appl. Catal., B, 2009, 90, 489–496.
22. M. Abd El Monsef, Ore controls and metallogenesis of au-ag deposits at atalla mine, central eastern desert of Egypt, Acta Geol. Sin., 2020, 94, 1451–1470.
23. D. N. Spinola, R. D. C. Portes, P. Srivastava, J. Torrent, V. Barron and P. Kuehn, Diagenetic reddening of Early Eocene paleosols on King George Island, Antarctica, Geoderma, 2018, 315, 149–159.
24. H. Guo and A. Barnard, Naturally occurring iron oxide nanoparticles: Morphology, surface chemistry and environmental stability, J. Mater. Chem. A, 2013, 1, 27–42.
25. R. M. Cornell and U. Schwertmann, The iron oxides: Structure, properties reactions occurrence and uses, VCH, Germany, 2003.
26. M. Nakai and N. Yoshinaga, Fibrous goethite in some soils from Japan and Scotland, Geoderma, 1980, 24, 143–158.
27. T. M. Tsao, Y. M. Chen and M. K. Wang, Origin, separation and identification of environmental nanoparticles: a review, J. Environ. Monit., 2011, 13, 1156–1163.
28. J. T. Cornelis, B. Delvaux, E. Van Ranst and P. G. Rouxhet, Sub-micrometer distribution of Fe oxides and organic matter in Podzol horizons, Geoderma, 2018, 323, 126–135.
29. B. Singh and R. J. Gilkes, Properties and distribution of iron-oxides and their association with minor elements in the soils of south-western australia, Eur. J. Soil Sci., 1992, 43, 77–98.
30. K. Lin, J. Ding, H. Wang, X. Huang and J. Gan, Goethite-mediated transformation of bisphenol A, Chemosphere, 2012, 89, 789–795.
31. H. Zhang and C.-H. Huang, Adsorption and oxidation of fluoroquinolone antibacterial agents and structurally related amines with goethite, Chemosphere, 2007, 66, 1502–1512.
32. D. Wu, S. Huang, X. Zhang, H. Ren, X. Jin and C. Gu, Iron minerals mediated interfacial hydrolysis of chloramphenicol antibiotic under limited moisture conditions, Environ. Sci. Technol., 2021, 55, 9569–9578.
33. J. Ding, M. Su, C. Wu and K. Lin, Transformation of triclosan to 2,8-dichlorodibenzo-p-dioxin by iron and manganese oxides under near dry conditions, Chemosphere, 2015, 133, 41–46.
34. M. J. Angove, M. B. Fernandes and J. Ikhsan, The sorption of anthracene onto goethite and kaolinite in the presence of some benzene carboxylic acids, J. Colloid Interface Sci., 2002, 247, 282–289.
35. X. Song and J.-F. Boily, Water vapor adsorption on goethite, Environ. Sci. Technol., 2013, 47, 7171–7177.
36. D. Tunega, M. H. Gerzabek, G. Haberhauer, K. U. Totsche and H. Lischka, Model study on sorption of polycyclic aromatic hydrocarbons to goethite, J. Colloid Interface Sci., 2009, 330, 244–249.
37. D. Tunega and A. Zaoui, Adsorption of polycyclic aromatic hydrocarbons on FeOOH polymorphs: A theoretical study, Surf. Sci., 2021, 706, 121795.
38. L. Li, H. Jia, X. Li and C. Wang, Transformation of anthracene on various cation-modified clay minerals, Environ. Sci. Pollut. Res., 2015, 22, 1261–1269.
39. J. Ma, Y. Horii, J. Cheng, W. Wang, Q. Wu, T. Ohura and K. Kannan, Chlorinated and parent polycyclic aromatic hydrocarbons in environmental samples from an electronic waste recycling facility and a chemical industrial complex in china, Environ. Sci. Technol., 2009, 43, 643–649.
40. I. C. Yadav and N. L. Devi, Nitrated- and oxygenated-polycyclic aromatic hydrocarbon in urban soil from Nepal: Source assessment, air-soil exchange, and soil-air partitioning, Ecotoxicol. Environ. Saf., 2021, 211, 111951.
41. D. Turcotte, P. Akhtar, M. Bowerman, Y. Kiparissis, R. S. Brown and P. V. Hodson, Measuring the toxicity of alkyl-phenanthrenes to early life stages of medaka (Oryzias latipes) using partition-controlled delivery, Environ. Toxicol. Chem., 2011, 30, 487–495.
42. C. Qin, X. Hu, M. G. Waigi, B. Yang and Y. Gao, Amino and hydroxy substitution in fluences pyrene-DNA binding, Sci. Total Environ., 2020, 725, 138542.
43. G. Shen, S. Tao, S. Wei, Y. Zhang, R. Wang, B. Wang, W. Li, H. Shen, Y. Huang and Y. Chen, Emissions of parent, nitro, and oxygenated polycyclic aromatic hydrocarbons from residential wood combustion in rural china, Environ. Sci. Technol., 2012, 46, 8123–8130.
44. U. Schwertmann and R. M. Cornell, Iron Oxides in the Laboratory: Preparation and Characterization, VCH, Germany, 2000.
45. H. Zhang, M. Bayne, S. Fernando, B. Legg, M. Zhu, R. L. Penn and J. F. Banfield, Size-dependent bandgap of nanogoethite, J. Phys. Chem. C, 2011, 115, 17704–17710.
46. X. Xu, R. Xiao, D. D. Dionysiou, R. Spinney, T. Fu, Q. Li, Z. Wang, D. Wang and Z. Wei, Kinetics and mechanisms of the formation of chlorinated and oxygenated polycyclic aromatic hydrocarbons during chlorination, Chem. Eng. J., 2018, 351, 248–257.
47. K. Umavathy and P. Sankar, Ontology based conceptual models for predicting fundamental organic reactivity, J. Mol. Graphics Modell., 2020, 100, 107691.
48. S. B. Liu, Where does the electron go? The nature of ortho/para and meta group directing in electrophilic aromatic substitution, J. Chem. Phys., 2014, 141, 194109.
49. H. Chi, J. Wan, Y. Ma, Y. Wang, M. Huang, X. Li and M. Pu, ZSM-5-(C@Fe) activated peroxymonosulfate for effectively degrading ciprofloxacin: In-depth analysis of degradation mode and degradation path, J. Hazard. Mater., 2020, 398, 123024.
50. H. Přichystalová, N. Almonasy, M. Nepraš, F. Bureš, M. Dvořák, M. Michl, J. Cermák and L. Burgert, N-Triazinyl Derivatives of 1- and 9-aminoanthracene: Synthesis and Photo-Physical Properties, J. Fluoresc., 2013, 23, 425–437.
51. A. Aquino, D. Tunega, G. Haberhauer, M. Gerzabek and H. Lischka, Quantum chemical adsorption studies on the (110) surface of the mineral goethite, J. Phys. Chem. C, 2006, 111, 877–885.
52. K. Lin, W. Liu and J. Gan, Oxidative removal of bisphenol a by manganese dioxide: Efficacy, products, and pathways, Environ. Sci. Technol., 2009, 43, 3860–3864.
53. X. B. Wang, N. Chen and L. Z. Zhang, Enhanced Cr(VI) immobilization on goethite derived from an extremely acidic environment, Environ. Sci.: Nano, 2019, 6, 2185–2194.
54. Y. Turu and H. Peter, Analysis of XPS spectra of Fe²⁺ and Fe³⁺ ions in oxide materials, Appl. Surf. Sci., 2008, 254, 2441–2449.
55. H. Jia, S. Zhao, Y. Shi, L. Zhu, C. Wang and V. K. Sharma, Transformation of polycyclic aromatic hydrocarbons and formation of environmentally persistent free radicals on modified montmorillonite: The role of surface metal ions and polycyclic aromatic hydrocarbon molecular properties, Environ. Sci. Technol., 2018, 52, 5725–5733.
56. J. Hua, Y.-H. Fei, C. Feng, C. Liu, S. Liang, S.-L. Wang and F. Wu, Anoxic oxidation of As(III) during Fe(II)-induced goethite recrystallization: Evidence and importance of Fe(IV) intermediate, J. Hazard. Mater., 2022, 421, 126806.
57. J. Lu, J. Shao, H. Liu, Z. Wang and Q. Huang, Formation of halogenated polyaromatic compounds by laccase catalyzed transformation of halophenols, Environ. Sci. Technol., 2015, 49, 8550–8557.
58. J. Chen, N. Wu, X. Xu, R. Qu, C. Li, X. Pan, Z. Wei and Z. Wang, Fe(VI)-mediated single-electron coupling processes for the removal of chlorophene: A combined experimental and computational study, Environ. Sci. Technol., 2018, 52, 12592–12601.
59. G. Crisponi, V. M. Nurchi, T. Pivetta, J. Gałęzowska, E. Gumienna-Kontecka, T. Bailly, R. Burgada and H. Kozłowski, Towards a new attenuating compound: A potentiometric, spectrophotometric and NMR equilibrium study on Fe(III), Al(III) and a new tetradentate mixed bisphosphonate–hydroxypyridinonate ligand, J. Inorg. Biochem., 2008, 102, 1486–1494.
60. R. C. Martins and R. M. Quinta-Ferreira, Catalytic ozonation of phenolic acids over a Mn-Ce-O catalyst, Appl. Catal., B, 2009, 90, 268–277.
61. H. Tong, X. Zhan, X. Tian, J. Li, D. Qian and D. Wu, Understanding the energy level matching relationships between semiconductor photocatalysts and organic pollutants for effective photocatalytic degradations, J. Colloid Interface Sci., 2018, 526, 384–391.
62. S. Luo, Z. Wei, R. Spinney, Z. Yang, L. Chai and R. Xiao, A novel model to predict gas phase hydroxyl radical oxidation kinetics of polychlorinated compounds, Chemosphere, 2017, 172, 333–340.
63. C. Hansch, A. Leo and R. W. Taft, A survey of Hammett substituent constants and resonance and field parameters, Chem. Rev., 1991, 91, 165–195.
64. G. Fang, C. Liu, J. Gao, D. D. Dionysiou and D. Zhou, Manipulation of persistent free radicals in biochar to activate persulfate for contaminant degradation, Environ. Sci. Technol., 2015, 49, 5645–5653.
65. E. Vejerano, S. Lomnicki and B. Dellinger, Formation and stabilization of combustion-generated environmentally persistent free radicals on an Fe(III)₂O₃/Silica Surface, Environ. Sci. Technol., 2011, 45, 589–594.
66. Z. Shi, J. M. Zachara, Z. Wang, L. Shi and J. K. Fredrickson, Reductive dissolution of goethite and hematite by reduced flavins, Geochim. Cosmochim. Acta, 2013, 121, 139–154.
67. P. Yi, Q. Chen, H. Li, D. Lang, Q. Zhao, B. Pan and B. Xing, A comparative study on the formation of environmentally persistent free radicals (EPFRs) on hematite and goethite: Contribution of various catechol degradation byproducts, Environ. Sci. Technol., 2019, 53, 13713–13719.
68. M. P. Fasnacht and N. V. Blough, Mechanisms of the aqueous photodegradation of polycyclic aromatic hydrocarbons, Environ. Sci. Technol., 2003, 37, 5767–5772.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1en00570g

---