Temperature effects on arsenate adsorption onto goethite and its preliminary application to arsenate removal from simulative geothermal water

Abstract

Laboratory batch experiments were conducted in order to assess the impacts of temperature on the performance of goethite in removing arsenate from water. All batch experiments were conducted at four temperatures (30, 50, 70 and 90 °C) and pH 4.6. The results showed that both the arsenic uptake rate and capacity were significantly enhanced with increasing temperature from 30 to 90 °C. The adsorption kinetics followed a pseudo-second-order model with coefficients of determination (R²) all above 0.999. The process followed the Langmuir model, and several thermodynamic parameters were calculated. Arsenate adsorption was facilitated more under simulative geothermal water conditions than in RO (reverse osmosis) water. The crystalline structure of goethite was not changed after adsorption at various temperatures. XPS results showed a decrease in the content of iron hydroxyl groups, which demonstrated that arsenate adsorption onto goethite may be realised through the replacement of the iron hydroxyl group to form inner-sphere bidentate/monodentate complexes at pH 4.6.

---

1. Introduction

Geothermal water is a kind of natural resource that has been used by humans for a long time. The importance of geothermal water has increased over the last decade as demand for non-fossil fuel energy sources has expanded.¹ Like a double-edged sword, significant environmental changes such as surface disturbances, thermal effects and emissions of contaminants are also generated by geothermal utilisation.² The main potential pollutants in geothermal discharged waters are hydrogen sulphide, mercury, arsenic and other trace metals.³

Arsenic in geothermal water is detected at elevated concentrations in many places of the world such as the Yangbajain geothermal fields of Tibet, the southeastern coastal area of the Chinese mainland, Taiwan and so on.⁴ Thompson and Demonge reported that geothermal water in Yellowstone National Park contained high concentrations of As 1–7800 μg L⁻¹.⁵ The Rio Loa basin El Tatio geothermal field of Chile was reported to have very high arsenic concentration values of up to 27.0 mg L⁻¹, and the Los Humeros geothermal field of Mexico was even reported to have an arsenic concentration as high as 73.6 mg L⁻¹.⁶ Natural geothermal water (including thermal spring water) is increasingly reported to contain high levels of arsenic, and this phenomenon is frequently found in south-eastern areas of China and Taiwan. Many of the arsenic-containing geothermal waters are discharged directly into the environment without treatment, triggering many environmental problems and potential problems.

The distribution of As(III)/As(V) is influenced by pH and redox conditions. As(III) is more toxic than As(V), and the World Health Organization has lowered the maximum contaminant level of total arsenic in drinking water to 10 μg L⁻¹. The mobility and transformation of As-tainted geothermal water have become significant concerns worldwide in environmental health. The final fate of arsenic in geothermal water involves its rise to the Earth's surface, and there is concern that it may contaminate the related groundwater systems, surface-water systems and soil systems.⁷

Various kinds of iron oxides and oxyhydroxides exist in soils, sediments and aquatic environments such as goethite (α-FeOOH), lepidocrocite (γ-FeOOH), hematite (α-Fe₂O₃) and so on.⁸ A great variety of iron oxides and oxyhydroxides usually have a strong affinity for arsenic species.^9,10 In this study, goethite is selected as the iron oxyhydroxide because it is widespread in soil systems and is a primary component of soil.¹¹ Investigating the reactions between goethite and different arsenic species is important to provide insight into the role of arsenic mobility and transformation in geothermal waters.

So far, many studies have focused on arsenic adsorption onto goethite.^12,13 However, a review of the literature showed that little has been done to determine the impacts of temperature on the adsorption process. Thus, we conduct a detailed study in a batch system in order to gain an understanding of the effects of temperature on arsenate adsorption onto goethite and the adsorption performance under simulative geothermal water conditions.

This paper focuses on the effect of temperature on arsenate adsorption onto goethite because the temperature is the first consideration of geothermal water. Previous papers usually investigated the temperature effect between 20 and 60 °C;^14–16 we further consider the comparatively higher temperatures 70 and 90 °C because we measured the temperature of thermal spring water near Xiamen city (Fujian province, China) to be 88 °C. The pH was adjusted to 4.6 with reference to the acidic pH of the Taiwan Datun volcanic region hot springs reported by Chen Bochun et al.;¹⁷ the pH of some hot springs in Taiwan are even as low as 1–2. Usually, the pH values of acidic geothermal discharged waters fall between 4 to 5.

2. Materials and methods

2.1 Materials

The granular goethite (α-FeOOH) used in this study was synthesised in the laboratory. A 1 M solution of Fe(NO₃)₃ was adjusted to pH 11.0 and stirred in a water bath at 70 ± 1 °C for 24 h. The suspension was purged with N₂ to remove CO₂, and the temperature was then adjusted to 90 °C for 72 h followed by repeated rinsing of the solids with deionised water. The solution containing a very high concentration of solids was ultrasonically dispersed for 30 min with the addition of a small amount of absolute ethyl alcohol. Finally, we obtained the granular goethite through a freeze drying technique. The product was stored at 4 °C for subsequent use.

The As(V) stock solutions were prepared by Na₂HAsO₄·7H₂O (AR). All the chemicals used in the experiments were AR grade.

2.2 Batch sorption

Adsorption experiments were performed with a background electrolyte of 0.01 M NaNO₃. Suspensions of goethite were made by adding 0.05 g goethite solids to 100 mL of 0.01 M NaNO₃ and mixing continuously with a magnetic stirring apparatus at 30 °C for 2 h to make the surface of goethite reach equilibrium. The pH of the arsenic stock solutions and goethite solutions were adjusted to 4.6 ± 0.2 using dilute HNO₃ and NaOH solutions.

Adsorption isotherms experiments were conducted in a shaking water bath with a temperature controller. Batch tests were performed in 200 mL bottles containing 0.5 g L⁻¹ goethite equilibrated with 1, 2, 5, 10, 15, 20, 30, 40, 50 mg L⁻¹ As(V) under shaking at 150 rpm for 24 h at 30, 50, 70 and 90 °C. Finally, the suspensions were filtered through a 0.22 μm membrane filter.

The kinetics experiments were conducted in a closed system consisting of a double-layer round glass reactor placed on a magnetic stirring apparatus. The double-layer round glass reactor was connected with a thermostat water bath that could be adjusted to different temperatures; the water flows through the inside of the double-layer glass reactor to keep the temperature constant. A pH electrode combined with a thermometer was inserted below the surface of the solution to detect the pH change in the reactor. The initial As(V) concentration was 1 mg L⁻¹, and the goethite suspensions were 0.2 g L⁻¹. The schematic diagram of the experimental apparatus used for kinetic study is shown in Fig. S1.†

2.3 Characterisations

The morphology of goethite was monitored with SEM (scanning electron microscopy) using a JEOL scanning electron microscope model Hitachi S-4800. XRD (powder X-ray diffraction) data were collected from 10° to 70° 2θ using Cu K_α (λ = 0.15418 nm) incident radiation in a PANalytical X'pert PRO diffractometer. XPS (X-ray photoelectron spectroscopy) data were collected on a PHI QUANTUM 2000 spectrometer (PHI, USA) with monochromatic Al Kα radiation (1486.6 eV).

The arsenic analytical method was hydride generation atomic fluorescence spectroscopy (HG-AFS), which is capable of detecting arsenic as low as 1.0 μg L⁻¹. All the samples were pre-reduced by 5% (w) thiourea – 5% (w) ascorbic acid to ensure that all arsenic species were converted to detectable As(III).

3. Results and discussion

3.1 Granular goethite characterisation

The SEM images are shown in Fig. 1(a); the goethite prepared in this study formed rod-like nanoparticles that are aggregated together. The specific surface area of the goethite samples was determined by the N₂/BET method to be 106.6 ± 1 m² g⁻¹. The XRD structural analysis (Fig. 1(b)) demonstrates that the granular iron oxyhydroxide is goethite by comparison with the standard XRD pattern (JCPDS 29-0713) of pure goethite.

Fig. 1 (a) SEM image and (b) XRD pattern of goethite.

3.2 Adsorption kinetics

The effect of time on the arsenate uptake rate at different temperatures is shown in Fig. 2, which shows a rapid initial uptake followed by a slow approach to equilibrium. The initial rapid adsorption rate can be attributed to the more adsorption sites at the initial stage; the arsenic species can interact easily with these sites. The slower adsorption may be due to slower diffusion into the interior of goethite and the decrease of the driving concentration between bulk solution and the goethite surface. The adsorption achieves equilibrium gradually within 100 min at 30, 50, 70 and 90 °C. As temperature increases from 30 °C to 70 °C, the slopes of the kinetic curves (initial rapid stage) gradually become steeper, indicating that higher temperature accelerates the reaction rate. When temperature is increased from 70 °C to 90 °C, an increasing kinetic trend is also observed, although the growth rate is smaller compared to the 30 to 70 °C temperature increase. This may indicate that arsenate adsorption onto goethite became less sensitive to temperature within the 70–90 °C range.

Fig. 2 Kinetics of As(V) adsorption onto goethite at 30, 50, 70, 90 °C (As initial concentration 1 mg L⁻¹, adsorbent dosage 0.2 g L⁻¹, pH 4.6) and the pseudo-second-order fitting curve.

Several kinetic models (i.e., pseudo-second-order, Elovich equation, intraparticle diffusion equations) are used to fit the kinetic data.¹⁸ The calculated parameters of the three kinetic models are listed in Table 1, and the fitting curves of the Elovich equation and intraparticle diffusion model are shown in Fig. S2.†

Table 1 The kinetic model fitting parameters for As(V) adsorption onto goethite at various temperatures and pH 4.6

Kinetic model	Pseudo second-order equation		Intraparticle diffusion function
Temperature	k2 (g mg^−1 min^−1)	R^2	kd (mg L^−1 min^−1/2)	R^2
30 °C	0.033	0.999	0.063	0.576
50 °C	0.043	0.999	0.055	0.461
70 °C	0.156	0.999	0.049	0.330
90 °C	0.290	0.999	0.040	0.273

---

The pseudo-second-order equation can be written as,

and the intraparticle diffusion equation is:

q_t = k_dt^1/2 + C (2)

where t is time, q_t is the adsorption capacity at t, q_e is the equilibrium adsorption capacity, and k₁, k₂, and k_d are the rate constants of the pseudo-first-order, pseudo-second-order and intraparticle diffusion equation, respectively. These parameters are strongly dependent on the applied operating conditions such as the initial solute concentration, pH, temperature and so on.

A simple modified Elovich equation is as follows:

where α and β are constants, t is the time, and q_t is the adsorption capacity at t. The Elovich equation is frequently used to describe the initial time period of a sorption process when the system is relatively far from equilibrium.¹⁹ This model has been proven to be suitable for heterogeneous systems, which might exhibit different activation energies for chemical adsorption on the surface.²⁰

The adsorption process on porous adsorbents is generally described by four stages: bulk diffusion, film diffusion, intraparticle diffusion and adsorption at a special site on the surface. Bulk and film diffusion are generally assumed to be rapid because of the agitation condition. As can be seen from Table 1, the pseudo-second-order could well describe the experimental data with linear regression coefficients (R²) all above 0.999, indicating that arsenate adsorption onto goethite was a second-order chemical adsorption process. To better understand the rate-determining step of adsorption, the kinetic data were tested using the intraparticle diffusion equation. As shown in Fig. S1,† the fitting curves were apparently divided into two stages, which were separately linearly fitted well with the intraparticle diffusion model; this indicates that the intraparticle diffusion process is a key rate-limiting step. According to the research of Barrow,²¹ goethite surfaces are variable and possibly composed of many crystal defects and micropores; the diffusion process may be attributed to these areas.²² The initial rapid stages of the kinetic curves were fitted with the Elovich equation, and the values of the correlation coefficients R² are greater than 0.93 at four temperatures. The good conformation to the Elovich equation suggests monolayer chemical adsorption.

3.3 Adsorption isotherms

The adsorption isotherms of arsenate at pH 4.6 at an ionic strength of 0.01 mol L⁻¹ NaNO₃ are presented in Fig. 3. The adsorption capacity increased with increasing initial arsenic concentration. Adsorption capacity also increased from 19.84 mg g⁻¹ to 25.97 mg g⁻¹ with a rise in temperature from 30 to 70 °C. However, a smaller change in adsorption capacity (25.97 to 26.60 mg g⁻¹) was observed as temperature increased from 70 to 90 °C, in agreement with the kinetic results. At pH 4.6, As(V) exists as the negatively charged H₂AsO₄⁻, while the surface of goethite is positively charged with the key functional group –FeOH₂⁺.²³ Thus, anionic arsenate adsorption is probably enhanced by coulombic attractions.²⁴

Fig. 3 Adsorption isotherm for As(V) adsorption onto goethite at 30, 50, 70, 90 °C (initial As concentration of 1, 2, 5, 10, 15, 20, 30, 40, 50 mg L⁻¹, respectively, adsorbent dosage 0.5 g L⁻¹, pH 4.6).

The isotherms are fitted with the Freundlich and Langmuir equations, and the parameters are summarised in Table 2. The linear regression coefficients for the Langmuir model are all above 0.995, suggesting identical adsorption sites of the goethite surface and monolayer adsorption. The isotherms were also well described by the Freundlich model. According to C.-H. Yang,²⁵ the Freundlich model was created with emphasis on two factors: the lateral interaction among the adsorbed molecules and the heterogeneity of the energetic surface. In addition, the Freundlich model is often applied to situations where the initial concentration of adsorbate is relatively low.²⁶

Table 2 Adsorption isotherm parameters for As(V) adsorption onto goethite at pH 4.6

Langmuir isotherm				Freundlich isotherm
Temperature (°C)	Qmax (mg g^−1)	kL (L mg^−1)	R^2	n	KF (mg g^−1)	R^2
30	19.84	0.446	0.995	2.958	6.208	0.974
50	22.32	0.598	0.995	3.094	7.740	0.984
70	25.97	0.992	0.997	3.188	10.15	0.967
90	26.60	1.001	0.997	4.301	13.31	0.963

---

The Langmuir isotherm equation is given by:

where q_e is the quantity of the species adsorbed at equilibrium (mg g⁻¹), K_L is a constant representing the virtual bonding strength between the target species and adsorber, C_e is the equilibrium concentration of adsorbate in the solution, and q_max is the maximum loading of the adsorbate onto the adsorbent.

The Freundlich isotherm equation was expressed as follows:

ln q_e = ln K_F + 1/n ln C_e (5)

where q_e is the quantity of the species adsorbed at equilibrium (mg g⁻¹), K_F is a constant that is a measure of sorption capacity, 1/n is a measure of adsorption density, and C_e is the equilibrium concentration of adsorbate in the solution.

As shown in Table 2, the values of K_L for arsenate adsorption increased from 0.446 to 1.001 as temperature increased from 30 to 90 °C, which is in good agreement with the observation that adsorption was promoted by increasing temperature. The values of 1/n (0.233–0.338) between 0 and 1 represent the favourable adsorption of arsenate onto goethite.

3.4 Calculation of thermodynamic parameters

The temperature dependence of arsenic adsorption is associated with changes in several thermodynamic parameters such as ΔG^o (the standard Gibbs free energy change), ΔH^o (enthalpy change), and ΔS^o (entropy change); these parameters are calculated using the following equations:

ln(K₀) = ΔS°/R − ΔH°/RT (6)

ΔG° = −RT ln(K₀) (7)

where R is the universal gas constant, T is temperature (K), and K₀ is the thermodynamic equilibrium constant; K₀ is determined using the method of Karthikeyan²⁷ by plotting ln(q_e/C_e) versus q_e and extrapolating ln(q_e/C_e) to zero (Fig. S3†).

As shown in Table 3, the values of ΔG° are calculated from eqn (7), while the values of ΔH° and ΔS° are calculated from the slope and intercept of the Van't Hoff plot, respectively. The positive value of ΔH° (11.29 kJ mol⁻¹) and negative values of ΔG° (−3.31 to −6.31 kJ mol⁻¹) confirm the spontaneous and endothermic nature of the adsorption process, and the decrease in ΔG° with increasing temperature suggests a stronger affinity at higher temperatures. The positive value of ΔS° implies an increase in randomness at the solid/solution interface.

Table 3 Thermodynamic parameters for As(V) adsorption onto goethite at different temperatures and pH 4.6

Temperature (°C)	K0	ΔG° (kJ mol^−1)	ΔS° (kJ mol^−1 K)	ΔH° (kJ mol^−1)
30	3.72	−3.31	0.045	11.29
50	4.57	−4.08
70	5.40	−4.81
90	8.09	−6.31

---

3.5 Adsorption under simulative geothermal water conditions

The components of various types of geothermal waters differ from each other; thus, modelling of such systems is very challenging. Simulative geothermal water was prepared with reference to the components of the Datun volcanic region hot springs in Taiwan reported by Chen Bochun et al.¹⁷ The detailed components of the simulative geothermal water are listed in Table 4. As can be seen from Fig. 4, the simulative geothermal water with multiple co-existing components promoted the adsorption of As(V), indicating that certain ionic strengths were beneficial for arsenate adsorption onto goethite. The maximum adsorption capacity increased from 21.7 to 27.1 mg g⁻¹, likely due to the compression of the double charged layer, allowing the arsenic species to get closer to the goethite surface. Multivalent cations such as Ca²⁺, Mg²⁺, and Fe³⁺ likely co-precipitate (CaHAsO₄, MgHAsO₄, FeAsO₄·2H₂O) with arsenate, thereby improving arsenate removal efficiency. Thus, adsorption experiments in simulated geothermal water lacking Ca²⁺, Mg²⁺, Fe³⁺, and Al³⁺ were carried out in this study. The results show that arsenate adsorption in simulative geothermal water is still enhanced compared with that in simulative geothermal water lacking Ca²⁺, Mg²⁺, Fe³⁺, Al³⁺, indicating that multivalent cations such as Ca²⁺, Mg²⁺, and Al³⁺ promote the As(V) adsorption process. This result is similar to the results of M. Stachowicz²⁸ suggesting that both Ca²⁺and Mg²⁺ promote PO₄³⁻ adsorption onto goethite.

Table 4 The components of simulative geothermal water

Components	Concentration (mg L^−1)
Mn^2+	1.0
Mg^2+	1.0
Al^3+	1.3
Ca^2+	2.3
Fe^2+	0.5
F^−	2.0
PO4^3−	1.0
SiO3^2−	2.6
NO3^−	6.2
Cl^−	4.4
SO4^2−	17.5
K^+	0.5
Na^+	4.8
NH4^+	0.3

---

Fig. 4 As(V) adsorption onto goethite under simulative geothermal water conditions at 30, 50, 70, and 90 °C (initial As concentration 20 mg L⁻¹, goethite concentration 0.5 g L⁻¹, pH 4.6).

Many competitive anions are reported to have adverse effects on arsenic adsorption onto goethite. Phosphate and fluoride, the main single interfering ions, were studied using concentrations of the competing anions equal to the concentration of arsenic. The results showed that phosphate had a profound competing effect on arsenic adsorption. This is reasonable because phosphorus and arsenic are in the same main group, and PO₄³⁻, AsO₄³⁻ have identical chemical structures; both molecules are tetrahedral oxyanions with similar pK_a values.²⁹ Fluoride is frequently detected at high concentrations in geothermal waters. Our results showed that fluoride had little interfering effect on arsenate adsorption, as shown in Fig. S4.†

3.6 Analysis of XRD and XPS spectra

The crystalline structures of goethite after arsenate adsorption (initial arsenic solution of 10 mg L⁻¹) at four different temperatures were investigated. Fig. 5 shows the XRD patterns of goethite at four temperatures. All XRD patterns are consistent with that of the standard XRD card (JCPDS 29-0713), demonstrating that goethite existed stably even at temperatures as high as 90 °C, and that the crystalline structure of goethite was not changed after arsenic adsorption.

Fig. 5 XRD patterns of goethite before and after arsenic adsorption at different temperatures.

To prove that arsenic adsorbed onto goethite, XPS analysis was conducted on a sample of goethite reacted with arsenate. Fig. 6 illustrates the wide scan XPS spectrum of goethite after arsenate adsorption. The occurrence of new arsenate peaks for As 1 mm and As3d were observed, confirming the adsorption of arsenate onto goethite. Fig. S5† shows that the binding energies of the As3d core level in arsenic oxides are ca. 45.5 eV for As(V). Therefore, it can be demonstrated that the valence state of As(V) was not changed during the adsorption process.

Fig. 6 The wide scan XPS spectra of As-loaded goethite.

To further investigate the surface hydroxyl group of goethite, O(1s) narrow scans of goethite before and after As(V) adsorption were analysed. The O(1s) spectrum is composed of overlapped peaks of oxide oxygen (O²⁻), hydroxyl (OH⁻), and sorbed water (H₂O).³⁰ The O(1s) peaks are fitted with two components (O²⁻ at 529.6 eV and OH⁻ at 530.9 eV) for pure α-FeOOH, and an additional peak at 531.3 eV can be attributed to the absorbed H₂O.³¹

All of the spectra were fitted using a 50 : 50 Gaussian : Lorentzian peak shape, and the obtained fitting results are shown in Fig. 7 and 8. The key reactive group of OH⁻ occupies 37.4% of the surface oxygen in goethite, as shown in Fig. 7. A significant decrease in the OH⁻ species was observed after arsenate adsorption at different temperatures, occupying 25.6%, 24.5%, 23.9% and 23.6% at 30 °C, 50 °C, 70 °C and 90 °C, respectively (Fig. 8). This result might indicate that singly coordinated iron hydroxyl (OH⁻) was involved and replaced arsenate oxyanion; the reaction was probably carried out through the formation of the inner-sphere monodentate complex FeOAsO₂OH and the inner-sphere bidentate complex (FeO)₂AsO₂, in accordance with the possible reactions shown below:²⁸

≡FeOH^−1/2 + 2H⁺ + AsO₄³⁻ → FeO^{−1/2+⊿Z₀}AsO₂OH^⊿Z₁ + H₂O (8)

≡FeOH^−1/2 + 2H⁺ + AsO₄³⁻ → (FeO)₂^{−1/2+⊿Z₀}AsO₂^⊿Z₁ + 2H₂O (9)

Fig. 7 O(1s) spectra of goethite.

Fig. 8 O(1s) spectra of goethite after As(V) adsorption.

In these reactions, ⊿z₀ and ⊿z₁ are the interfacial charge distribution (CD) model coefficients, where ⊿z₀ + ⊿z₁ is equal to the charge introduced by arsenate adsorption.

4. Conclusions

The present study showed that the synthesised goethite has a strong affinity for inorganic arsenate in aqueous systems, temperature plays an important role in the adsorption of arsenic species, and adsorption capacities increase with increasing temperature from 30–90 °C. Adsorption isotherms were well fitted to the Langmuir model, and the kinetic data were best fit by the pseudo-second equation. The adsorption of arsenate exhibited a better performance under simulative geothermal water conditions, and co-existing multivalent cations such as Ca²⁺, Mg²⁺, and Al³⁺ facilitated arsenate adsorption. XRD analysis revealed that the crystalline structure of goethite after arsenate adsorption was not changed within the temperature range of 30–90 °C. The content of iron hydroxyl groups decreased from 35.4% in raw goethite to 25.6% at 30 °C and 23.2% at 90 °C after adsorption. This indicated that the singly coordinated iron hydroxyl (OH⁻) was involved, and that higher temperature (50–90 °C) geothermal water facilitated arsenate to form inner-sphere bidentate or monodentate complexes at pH 4.6.

Acknowledgements

This research is financially supported by the National Natural Science Foundation of China (Contract no. 21077086). The authors would like to thank all the reviewers for their helpful comments and valuable suggestions.

References

1. G. Axelsson, Geothermics, 2010, 39, 283–291.
2. L. Paoli and S. Loppi, Environ. Pollut., 2008, 155, 383–388.
3. S. Loppi, L. Paoli and C. Gaggi, J. Atmos. Chem., 2006, 53, 93–105.
4. K. Tyrovola, N. P. Nikolaidis, N. Veranis, N. Kallithrakas-Kontos and P. E. Koulouridakis, Water Res., 2006, 40, 2375–2386.
5. B. Planer-Friedrich, J. London, R. B. McCleskey, D. K. Nordstrom and D. Wallschläger, Environ. Sci. Technol., 2007, 41, 5245–5251.
6. P. Smedley and D. Kinniburgh, Appl. Geochem., 2002, 17, 517–568.
7. À. Piqué, F. Grandia and À. Canals, Water Res., 2010, 44, 5618–5630.
8. R. V. Morris, H. V. Lauer, C. A. Lawson, E. K. Gibson, G. A. Nace and C. Stewart, J. Geophys. Res.: Solid Earth, 1985, 90, 3126–3144.
9. Y. Mamindy-Pajany, C. Hurel, N. Marmier and M. Roméo, Desalination, 2011, 281, 93–99.
10. S. K. Maji, Y.-H. Kao and C.-W. Liu, Desalination, 2011, 280, 72–79.
11. W. Hartley, R. Edwards and N. W. Lepp, Environ. Pollut., 2004, 131, 495–504.
12. Y. Mamindy-Pajany, C. Hurel and N. Marmier, C. R. Chim., 2009, 12, 876–881.
13. T. S. Choong, T. Chuah, Y. Robiah, F. Gregory Koay and I. Azni, Desalination, 2007, 217, 139–166.
14. K. Banerjee, G. L. Amy, M. Prevost, S. Nour, M. Jekel, P. M. Gallagher and C. D. Blumenschein, Water Res., 2008, 42, 3371–3378.
15. F. Partey, D. Norman, S. Ndur and R. Nartey, J. Colloid Interface Sci., 2008, 321, 493–500.
16. S. Kundu and A. K. Gupta, Colloids Surf., A, 2006, 273, 121–128.
17. B. C. Chen, Taiwan joint academic symposium on earth science[C], 2007, pp. 125–128.
18. D. Bulgariu and L. Bulgariu, Bioresour. Technol., 2012, 103, 489–493.
19. W. Plazinski, W. Rudzinski and A. Plazinska, Adv. Colloid Interface Sci., 2009, 152, 2–13.
20. S. Fendorf, M. J. Eick, P. Grossl and D. L. Sparks, Environ. Sci. Technol., 1997, 31, 315–320.
21. N. Barrow, G. Brümmer and L. Fischer, Eur. J. Soil Sci., 2012, 63, 389–398.
22. H. Liu, T. Chen and R. L. Frost, Chemosensors, 2014, 103, 1–11.
23. P. Lakshmipathiraj, B. Narasimhan, S. Prabhakar and G. Bhaskar Raju, J. Hazard. Mater., 2006, 136, 281–287.
24. C. Su and R. W. Puls, Environ. Sci. Technol., 2001, 35, 1487–1492.
25. C. H. Yang, J. Colloid Interface Sci., 1998, 208, 379–387.
26. J. Zhang and R. Stanforth, Langmuir, 2005, 21, 2895–2901.
27. G. Karthikeyan, K. Anbalagan and N. M. Andal, J. Chem. Sci., 2004, 116, 119–127.
28. M. Stachowicz, T. Hiemstra and W. H. van Riemsdijk, J. Colloid Interface Sci., 2008, 320, 400–414.
29. F. Kolbe, H. Weiss, P. Morgenstern, R. Wennrich, W. Lorenz, K. Schurk, H. Stanjek and B. Daus, J. Colloid Interface Sci., 2011, 45, 565–571.
30. Y. Zhang, M. Yang, X. M. Dou, H. He and D. S. Wang, Environ. Sci. Technol., 2005, 39, 7246–7253.
31. Z. Sun, X. Feng and W. Hou, Nanotech., 2007, 18, 455–607.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra08318k

---