Environ Geochem Health DOI 10.1007/s10653-015-9692-1
ORIGINAL PAPER
Effects of natural organic matter on the coprecipitation of arsenic with iron Eun Jung Kim • Bo-Ram Hwang • Kitae Baek
Received: 24 December 2014 / Accepted: 27 February 2015 Ó Springer Science+Business Media Dordrecht 2015
Abstract Natural organic matter (NOM) can affect arsenic speciation and mobility in the environment. In this study, the effects of NOM on the coprecipitation of arsenic with iron were investigated in order to better understand the fate and transport of arsenic in natural environments. The coprecipitation of arsenic with iron was studied in the presence and absence of NOM under various arsenic-to-iron molar ratios (As/Fe) and pH conditions. The addition of humic acid (HA) hindered the As–Fe coprecipitation under high pH and high As/Fe conditions by forming a soluble As–Fe– HA complex. The X-ray diffraction and Fourier transform infrared studies showed that the As–Fecoprecipitated solid phase was highly affected by pH and As/Fe. The arsenic was coprecipitated with iron as an amorphous ferric arsenate phase at a low pH level or high As/Fe conditions, while the formation of ferrihydrite phase and the arsenic incorporation to the ferrihydrite by adsorption was predominant at high pH levels or low As/Fe conditions. The HA affected the As–Fe-coprecipitated solid phase depending on the
E. J. Kim K. Baek (&) Department of Bioactive Material Sciences, Chonbuk National University, 567 Baekje-daero, Deokjin-gu, Jeonju, Jeollabukdo 561-675, Republic of Korea e-mail: [email protected] B.-R. Hwang K. Baek Department of Environmental Engineering, Chonbuk National University, 567 Baekje-daero, Deokjin-gu, Jeonju, Jeollabukdo 561-675, Republic of Korea
As/Fe molar ratio under neutral and alkaline conditions. Keywords Arsenic Iron Natural organic matter Coprecipitation
Introduction Arsenic is a naturally occurring trace element in the environment that has adverse effects on human health (Smedley and Kinniburgh 2002). In the natural water, mineral–water interactions such as adsorption/desorption, precipitation, and oxidation/reduction play important roles in controlling the mobility of arsenic. Among the many minerals, iron oxide minerals such as hydrous ferric oxide (HFO), goethite, and hematite are considered to be the most important sinks for arsenic because of their affinity to arsenic and their abundance in the environment (Smedley and Kinniburgh 2002; Stollenwerk 2003; Kim et al. 2014). Natural organic matter (NOM) has also been suggested as having effects on arsenic speciation and mobility in the environment (Sharma et al. 2010, 2011; Redman et al. 2002). NOM is a degradation product of biomolecules such as plants and animals, which is ubiquitous in both water and soil environments. Humic and fulvic acids are the two primary components of NOM having complex structures with various functional groups, which include hydroxyl, carboxylic
123
Environ Geochem Health
and phenolic groups (Deng and Dixon 2002). These functional groups are highly reactive with cationic metals and metal oxides and form both soluble and surface NOM–metal complexes (Redman et al. 2002). NOM has been reported to compete with arsenic for the surface sorption sites of iron oxides due to its high affinity to iron oxides (Sharma et al. 2010). In addition, the solubility of metal oxides such as iron oxides and the mobility of organic and inorganic contaminants associated with these metal oxides can be regulated by NOM due to the formation of soluble NOM–metal complexes (Deng and Dixon 2002). The soluble NOM–metal complexes can also bind strongly with dissolved anions, in which cationic metals act as bridging complexes (Lin et al. 2004). The formation of dissolved ternary As–Fe–NOM complexes has been suggested in environments containing arsenic, NOM, and iron ions or iron oxides (Sharma et al. 2010; Lin et al. 2004; Liu et al. 2011). NOM can also participate in redox reactions by serving as an electron shuttle, which may enhance the microbial reductive dissolution of iron oxides and transform As species (Redman et al. 2002; Scott et al. 1998). Therefore, the presence of NOM can greatly affect the speciation, fate, and mobility of arsenic in the environment (Smedley and Kinniburgh 2002). In iron-rich environments, the coprecipitation of arsenic with iron as well as arsenic adsorption on iron oxide minerals has been reported to occur (Smedley and Kinniburgh 2002; Ford 2002). Arsenic may be incorporated into the iron precipitates during the precipitation of soluble iron, which results in an arsenic-rich HFO phase or a ferric arsenate (FeAsO4)like phase depending on the solution conditions, such as pH and As/Fe molar ratio (Ford 2002; Jia et al. 2006; Violante et al. 2007; Tokoro et al. 2010). The presence of NOM can affect the As–Fe coprecipitate phase by forming Fe–NOM or As–Fe–NOM complexation. Furthermore, arsenic speciation can be influenced by the presence of NOM, which can affect the mobility of arsenic. However, exactly how the environmental conditions influence the arsenic mobility and coprecipitation of arsenic with iron in the presence of NOM is not yet fully understood. Therefore, the objective of this study was to investigate the effects of NOM on the arsenic mobility and coprecipitation of arsenic with iron in order to better understand the fate and transport of arsenic in natural environments. In this study, humic acid (HA) was
123
selected as a representative NOM, and the coprecipitation of arsenic with iron was studied in the presence of HA under various arsenic-to-iron molar ratios (As/Fe) and pH conditions. We evaluated the effects of NOM on arsenic coprecipitation with iron via solution-phase experiments and solid-phase characterization by X-ray diffraction (XRD), infrared (IR) spectroscopy, X-ray absorption spectroscopy (XAS), and scanning electron microscopy (SEM).
Materials and methods Materials HA was purchased from Sigma-Aldrich and purified by precipitation in a HCl solution following dissolution in a NaOH solution before the experiments (Hwang et al. 2014). The purchased HA was mixed in 1 M NaOH for 8 h, and the dissolved HA was separated by centrifugation. Then, the pH of the supernatant solution was adjusted to pH 2 with 1 M HCl and stirred for 24 h. The precipitated HA was separated by centrifugation, rinsed twice with 0.01 M HCl, and dried in an oven (60 °C) for 24 h. The elemental composition and functional groups of the purified HA were characterized by an elemental analyzer (Vario EL, Elementar, Germany) and a Fourier transform infrared (FTIR; Spectrum GX, Perkin Elmer, USA) spectrometer, respectively. The elemental analysis of the purified HA showed that it was composed of 44.2 % carbon, 3.75 % hydrogen, 1.07 % nitrogen, and 3.35 % sulfur. The FTIR spectra of the purified HA indicated several strong bands for the commonly observed functional groups in humic compounds (Fig. 1). Each band can be identified as follows: H bound to OH at wavenumber 3440 cm-1, aliphatic CH bonds at 2927 cm-1, C–O of COOH at 1713 cm-1, phenolic OH and aromatic C–C at 1615 cm-1, aliphatic CH at 1438 cm-1, C–O stretching at 1251 cm-1, and Si–O of the silicate impurities at 1050 cm-1 (Pansu and Gautheyrou 2006). The arsenic [As(V)] and iron [Fe(III)] solutions (0.1 M) were prepared by dissolving Na2HAsO47H2O (Sigma-Aldrich) and Fe(NO3)39H2O (Sigma-Aldrich), respectively. Coprecipitation experiments All of the experiments were conducted in a completely mixed batch reactor system. Initially, HA solutions
Transmittance %
Environ Geochem Health
4000
3000
2000
1000
-1
Wavenumber (cm )
Fig. 1 FTIR spectrum of purified humic acid
were prepared in various concentrations at pH 10. The arsenic and iron solutions were simultaneously added to HA solution, and then, pH was adjusted to 4, 7, and 10 with 0.5 M NaOH or 0.5 M HNO3 right after the addition of arsenic and iron solutions. The pH was not controlled throughout the reaction, and the final pH was measured after the reaction. The final HA concentrations in the solutions were 0, 100, 500, and 1000 mg/L. The iron concentration was fixed at 10 mM, and the arsenic concentration was varied in order to adjust the arsenic-to-iron molar ratio (As/Fe) to 0.1, 0.5, and 1. The reaction was continued in a temperature-controlled flatbed rotation incubator for 48 h at 20 °C and at 150 rpm. After the reaction, the samples were centrifuged at 8000 rpm for 20 min and filtered through 0.45-lm membrane filters. The concentrations of arsenic and iron in the filtrates were analyzed via inductively coupled plasma optical emission spectroscopy (ICP-OES; Agilent Tech, USA). Solids characterization The precipitated solid phases were dried and characterized via X-ray diffraction, infrared (IR) spectroscopy, and scanning electron microscopy with energy dispersive X-ray spectroscopy (SEM/EDX). The XRD data were collected on a powder X-ray diffractometer (MAX-2500, Rigaku, Japan) using Cu ˚ ) radiation obtained at 40 kV and Ka (k = 1.54059A 30 mA. The scans were performed between 5 and 70° 2h with 0.02° step sizes. The organic carbon contents
of the solid phases were analyzed using an elemental analyzer (Vario EL, Elementar, Germany). The transmittance spectra of the precipitates in the mid-infrared region (400–4000 cm-1) were obtained using an FTIR (Spectrum GX, Perkin Elmer, USA) spectrometer. In order to investigate arsenic species, selected samples were analyzed by X-ray absorption fine structure spectroscopy (XAFS) at beamline 10C at the Pohang Accelerator Laboratory, Korea. Solid phases were filled into Al sample holders and sealed with Kapton tape. The arsenic K-edge X-ray absorption near-edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) spectra were collected in fluorescence mode at room temperature using a Si(111) double-crystal monochromator. The morphology and particle size of the precipitates were characterized using a Hitachi SU-70 field emission scanning electron microscope (FE-SEM), and the chemical composition was evaluated using energy dispersive X-ray (EDX) spectroscopy.
Results and discussion Coprecipitation of arsenic with iron Arsenic coprecipitation with iron was studied under various arsenic-to-iron molar concentration ratios (As/ Fe) and pH conditions. In these experiments, the initial iron [Fe(III)] concentration was fixed at 10 mM, and the initial arsenic [As(V)] concentration was 1, 5, and 10 mM in order to adjust the As/Fe molar ratio to 0.1, 0.5, and 1. Figure 2 summarizes the remaining arsenic and iron concentrations in the solutions after As–Fe coprecipitation as a function of HA concentrations conducted at pH 4, 7, and 10. The decrease in arsenic from the solution indicated that the coprecipitation of arsenic with iron had occurred, which included arsenic adsorption to iron oxides and simultaneous arsenic precipitation with iron. Higher arsenic removal has been reported by coprecipitation of As(V) and Fe(III) than by simple As(V) adsorption on ferrihydrite, indicating that As–Fe coprecipitation involves more than adsorption (Tokoro et al. 2010). In addition, our results clearly show that the coprecipitation of arsenic with iron was highly affected by both solution pH and As/Fe molar ratio. The amounts of precipitated arsenic and iron were highest at pH 4 and decreased as pH increased. However, the effects of pH became weaker
123
Environ Geochem Health
(b)
12
10
pH 4 pH 7 pH 10
12
10
Iron Concentration (mmol/L)
Arsenic Concentration (mmol/L)
(a)
8
6
4
2
pH 4 pH 7 pH 10
8
6
4
2
0
0 0
200
400
600
800
1000
1200
0
200
Humic Concentration (mg/L)
(d)
6
5
pH 4 pH 7 pH 10
Iron Concentration (mmol/L)
Arsenic Concentration (mmol/L)
(c)
4
3
2
1
0
600
800
1000
1200
12
10
pH 4 pH 7 pH 10
8
6
4
2
0 0
200
400
600
800
1000
1200
0
200
Humic Concentration (mg/L)
400
600
800
1000
1200
1000
1200
Humic Concentration (mg/L)
(f)
(e) 0.7
0.6
6
pH 4 pH 7 pH 10
Iron Concentration (mmol/L)
Arsenic Concentration (mmol/L)
400
Humic Concentration (mg/L)
0.5
0.4
0.3
0.2
0.1
0.0
5
pH 4 pH 7 pH 10
4
3
2
1
0
0
200
400
600
800
1000
1200
Humic Concentration (mg/L)
0
200
400
600
800
Humic Concentration (mg/L)
Fig. 2 Concentrations of arsenic and iron remained in solutions after coprecipitation in the presence of HA at various As/Fe molar ratios (1, 0.5, and 0.1) and pH conditions (pH 4, 7, and 10). Initial iron concentration was 10 mM, and initial arsenic
concentration was 10, 5, and 1 mM for As/Fe of 1, 0.5, and 1, respectively. Concentrations of a arsenic at As/Fe = 1, b iron at As/Fe = 1, c arsenic at As/Fe = 0.5, d iron at As/Fe = 0.5, e arsenic at As/Fe = 0.1, f iron at As/Fe = 0.1
at lower As/Fe molar ratios. Even though the initial iron concentration was fixed for all of the conditions, greater iron precipitation was observed with a lower
initial arsenic concentration, i.e., lower As/Fe molar ratio. The presence of high concentrations of arsenic seemed to hinder the iron precipitation, particularly in
123
Environ Geochem Health
Effects of humic acid The effects of HA on arsenic coprecipitation with iron were studied by adding HA concentrations up to 1000 mg/L (Fig. 2). The addition of HA increased both the remaining arsenic and iron concentrations, particularly at high pH and high As/Fe conditions. At pH 4, the arsenic and iron precipitation was not
Coprecipitation density (mmol As/mmol Fe)
(a) 1.0 0.8 0.6 0.4 0.2 0.0
pH 4 pH 7 pH 10
-0.2 0
200
400
600
800
1000
1200
Humic Concentration (mg/L)
Coprecipitation density (mmol As/mmol Fe)
(b) 0.6 0.5 0.4 0.3 0.2 0.1
pH 4 pH 7 pH 10
0.0 0
200
400
600
800
1000
1200
Humic Concentration (mg/L)
(c) Coprecipitation density (mmol As/mmol Fe)
high pH conditions. In addition, the ferric arsenate phase, which was considered to be formed in high As/ Fe conditions, was only stable under acidic conditions, and its solubility increased as pH increased (Langmuir et al. 2006; Jia et al. 2006). Therefore, the increased concentrations of both iron and arsenic in high pH conditions could be partly affected by the increased solubility of the ferric arsenate phase. We also studied the coprecipitation density of arsenic, which indicates the molar concentrations of precipitated arsenic per molar concentration of precipitated iron. Figure 3 shows the coprecipitation density of arsenic as a function of HA concentration at various As/Fe molar ratios and pH conditions. The coprecipitation density of arsenic was primarily governed by the arsenic that was initially added to the iron molar concentration ratio and pH conditions. When the As/Fe molar ratio was one, the coprecipitation density of arsenic was 0.80, 0.72 and 0.44 at pH 4, 7, and 10, respectively. A decrease in coprecipitation density was observed with an increase in pH, which might have been due to the low affinity of arsenate to the iron oxides at high pH conditions. The arsenate sorption onto the iron oxides was high in acidic conditions and decreased with an increase in pH due to the increased repulsive electrostatic forces between anionic arsenic and negatively charged surface of iron oxides at high pH conditions (Dixit and Hering 2003). At a lower As/Fe ratio, the coprecipitation density was close to the initial As/Fe molar ratio at pH 4 and 7, indicating that all of the arsenic that was initially added has coprecipitated with the iron. This result might have been due to the presence of sufficient binding sites on the iron oxide at a low As/Fe molar ratio. However, a lower density was observed at pH 10 even with a low As/Fe molar ratio due to the low affinity between arsenic and iron oxides at high pH.
0.12 0.10 0.08 0.06 0.04 0.02
pH 4 pH 7 pH 10
0.00 0
200
400
600
800
1000
1200
Humic Concentration (mg/L)
Fig. 3 Coprecipitation density of arsenic calculated by precipitated arsenic concentrations per precipitated iron concentrations at arsenic-to-iron ratios of a As/Fe = 1, b As/Fe = 0.5, c As/Fe = 0.1 at various pH conditions
123
Environ Geochem Health
123
(a)
7 As/Fe=0.1 As/Fe=0.5 As/Fe=1
Fraction of Carbon (%)
6 5 4 3 2 1 0
4
7
10
pH
(b) 14 As/Fe=0.1 As/Fe=0.5 As/Fe=1
Fraction of Carbon (%)
12 10 8 6 4 2 0
4
7
10
pH
(c) 20 As/Fe=0.1 As/Fe=0.5 As/Fe=1
18
Fraction of Carbon (%)
affected by the addition of HA. However, at pH 7 and 10, the dissolved iron and arsenic concentrations increased as the HA concentration increased, and almost all of the arsenic and iron that were initially added remained in the solutions with addition of 1000 mg/L HA at pH 10. These results clearly indicated that HA inhibited the precipitation of both As and Fe at higher pH conditions. HA primarily influenced the arsenic mobility by increasing the solubility of the iron oxides and the mobility of arsenic by forming soluble HA–Fe complexes or As–Fe–HA (Sharma et al. 2010; Lin et al. 2004; Liu et al. 2011; Chen et al. 2006) and by inhibiting arsenic adsorption onto iron oxides due to its high affinity with iron oxides (Redman et al. 2002; Sharma et al. 2010). When we considered the coprecipitation density of arsenic (Fig. 3), even though the inhibition of arsenic and iron precipitation by HA was observed at pH 7, the coprecipitation density of arsenic was not affected by the addition of HA. This result indicates that the addition of HA inhibited arsenic and iron precipitation at the same rate in neutral pH conditions. HA might form soluble As–Fe–HA complex and increase solution arsenic and iron concentrations proportionally, which seemed to result in no change of coprecipitation density with HA. On the other hand, the coprecipitation density of arsenic decreased as the HA concentration increased at pH 10, indicating that less arsenic was associated with iron in the presence of HA. Under alkaline conditions, arsenic seems to be mobilized by competition with HA for sorption sites of iron oxide phase in addition to the soluble As–Fe–HA complex formation. In order to evaluate HA incorporation into the As– Fe coprecipitate, the organic carbon (OC) contents of the precipitates were further evaluated. Figure 4 shows the OC contents in the solid phase as a function of pH and As/Fe molar ratio. A decrease in OC fraction was observed with an increase in pH in most cases, which was consistent with the previously reported pH-dependent NOM adsorption on iron oxides. NOM adsorption on iron oxides is reported to be high in acidic conditions and decreases as pH increases (Gu et al. 1995). In addition, higher OC contents were observed at lower As/Fe molar ratios, probably due to the additional available complexation sites of the iron. However, no clear correlation was observed between HA content in the solid phase and As–Fe coprecipitation. High HA contents in the solid
16 14 12 10 8 6 4 2 0
4
7
10
pH
Fig. 4 Organic carbon contents of the As–Fe coprecipitates with various HA concentrations. a HA = 100 mg/L, b HA = 500 mg/L, c HA = 1000 mg/L
phase did not inhibit the As–Fe coprecipitation, which indicates that the competitive adsorption of HA with arsenic onto the iron oxides did not greatly affect the arsenic mobilization. HA seemed to affect the iron and
Environ Geochem Health
arsenic mobilization primarily through As–Fe–HA complexation, preventing solid-phase formation in high pH conditions. XRD study
Fig. 5 XRD patterns of the As–Fe-coprecipitated solid phases as a function of HA concentrations and initial As/Fe molar ratio reacted at a pH 4, b pH 7, and c pH 10
The XRD patterns of the As–Fe-coprecipitated solid phases are summarized in Fig. 5 as a function of initial As/Fe molar ratio and HA concentrations at pH 4, 7, and 10. The XRD patterns indicate the formation of poorly crystalline phases having broad bands near 30° 2h. Previous studies have reported the distinctions of the XRD spectra between two-line ferrihydrite and poorly crystalline ferric arsenate as follows: Two-line ferrihydrite shows two broad bands at *34° and *61° 2h, while poorly crystalline ferric arsenate has broad bands at *28° and *58° 2h (Jia et al. 2006; Tokoro et al. 2010). At pH 4, the major bands of the XRD spectra were located between the characteristic bands of the poorly crystalline ferric arsenate and the ferrihydrite depending on the As/Fe value (Fig. 4a). The major band of the XRD spectra was close to the poorly crystalline ferric arsenate band located at *28° 2h at a As/Fe molar ratio of 1. However, the major peak of the XRD spectrum was shifted to higher degree 2h with a decrease in the As/Fe molar ratio, and the XRD spectrum was close to that of the ferrihydrite when the As/Fe molar ratio was 0.1. This result suggests that arsenic was coprecipitated with iron as an amorphous ferric arsenate phase at high As/Fe molar ratios, while formation of ferrihydrite phase was predominant at low As/Fe molar ratios, and arsenic seems to be incorporated into the ferrihydrite phase by adsorption. The addition of HA did not show any clear effects on the XRD observed phase at low pH conditions as observed in the solution-phase experiments. At pH 7, the major XRD peak was located at *31° 2h, between 28° and 34° 2h, indicating that both the poorly crystalline ferric arsenate and the ferrihydrite phases were present. However, the presence of HA shifted the major band to higher degree 2h at a As/Fe molar ratio of 0.1. The HA seemed to promote the formation of the ferrihydrite phase at low As/Fe molar ratios. At pH 10, the major XRD band was located near 30° 2h at a As/Fe molar ratio of 1, while the main peak was shifted to higher degree 2h close to the ferrihydrite phase as the As/Fe decreases. Under alkaline conditions, both the ferrihydrite and the poorly crystalline
123
Environ Geochem Health
ferric arsenate phases seemed to be present at a As/Fe molar ratio of 1, while the formation of the ferrihydrite phase was predominant at a lower As/Fe molar ratio, as observed in lower pH conditions. However, addition of HA resulted in a broadened XRD band shifted to lower degree 2h at As/Fe molar ratios of 1 and 0.5, which indicates that the amorphous ferric arsenate phase was primarily present in the presence of HA. The solution-phase experiments showed that the As– Fe coprecipitation was highly inhibited by the addition of HA at pH 10, and only very small amounts of the solid phase were precipitated at a high concentration of HA. The HA seemed to form soluble Fe–HA and/or As–Fe–HA complexes and prevented ferrihydrite phase formation, which resulted in the presence of only poorly crystalline ferric arsenate phases in the solid precipitates as observed by XRD. However, the effects of HA were not as prominent at low As/Fe molar ratios. The results of the XRD study suggest that the addition of HA affected the As–Fe coprecipitation phase depending on the As/Fe molar ratio under neutral and alkaline conditions.
Fig. 6 FTIR spectra of As–Fe coprecipitates with As/Fe molar ratio of 1 at (a) HA = 0 mg/L and pH 4, (b) HA = 1000 mg/L and pH 4, (c) HA = 0 mg/L and pH 7, (d) HA = 1000 mg/L and pH 7, (e) HA = 0 mg/L and pH 10, and (f) HA = 1000 mg/ L and pH 10
123
FTIR study The As–Fe-coprecipitated solid phases were further characterized using FTIR. Figure 6 summarizes the infrared spectra of the As–Fe-coprecipitated solid phases formed in the presence or absence of HA at a As/Fe molar ratio of 1 and pH 4, 7, and 10. The infrared spectra of the As–Fe precipitates exhibited strong bands at 828–806 cm-1 wavenumber, which were attributed to the As–O stretching vibration of the As–O–Fe bonds (Jia et al. 2007; Gomez et al. 2009). The As–O–Fe stretching vibration band of the poorly crystalline ferric arsenate phase has been reported to be positioned at approximately 833–838 cm-1, whereas the As–O–Fe stretching vibration band of the surface complexed to the ferrihydrite phase was reported to be at 806–808 cm-1 (Jia et al. 2007; Goldberg and Johnston 2001). The As–O–Fe stretching vibration band of the precipitates that formed at pH 4 was located at 827 cm-1, which was similar to that of the amorphous ferric arsenate. However, as the pH increased, the As–O–Fe stretching vibration band gradually shifted to lower wavenumbers. The As–O– Fe bands of the precipitates were located at 822 and
Fig. 7 XANES spectra of As–Fe coprecipitates with As/Fe molar ratio of 1 at (a) HA = 0 mg/L and pH 4, (b) HA = 1000 mg/L and pH 4, (c) HA = 0 mg/L and pH 7, (d) HA = 1000 mg/L and pH 7, (e) HA = 0 mg/L and pH 10, and (f) HA = 1000 mg/L and pH 10
Environ Geochem Health
Fig. 8 FE-SEM images of As–Fe coprecipitates with As/Fe molar ratio of 1 at a HA = 0 mg/L and pH 4, b HA = 1000 mg/L and pH 4, c HA = 0 mg/L and pH 7, d HA = 1000 mg/L and pH 7, e HA = 0 mg/L and pH 10, and f HA = 1000 mg/L and pH 10
810 cm-1 at pH 7 and 10, respectively. In addition, a new band was observed at approximately 870 cm-1 at pH 10, which was attributed to the uncomplexed As–O bonds of the adsorbed arsenate (Goldberg and Johnston 2001). This result indicates that arsenic was coprecipitated with iron as an amorphous ferric arsenate phase at lower pH, but arsenic was removed by adsorption onto the ferrihydrite phase as pH increased. This result was consistent with the XRD characterization noted in the previous section. In addition, the infrared spectra of the precipitates exhibited a strong band at 1622–1640 cm-1 and a broad band located near 3400 cm-1, which were assigned to the water O–H bending mode and the
stretching vibration band of the O–H, respectively. The sharp band at 1385 cm-1 was due to NO32- ions, which were incorporated into the precipitates from the iron solution [Fe(NO3)39H2O]. When HA was added, new bands were observed at approximately 1100 and 1700 cm-1, which were the functional group bands for the HA such as C–O of COOH, phenolic OH, and aromatic C–C (Fig. 1). This result indicates that HA was incorporated into the As– Fe precipitates probably by adsorption, as observed in Fig. 4. In addition to the bands for HA functional group, no changes were observed for the As–O–Fe band at pH 4. On the other hand, the As–O–Fe band at pH 7 was shifted to a slightly lower wavenumber and
123
Environ Geochem Health
an additional band emerged at *870 cm-1, indicating that HA promoted formation of ferrihydrite phase and arsenic adsorption. However, at pH 10, the intensity of the As–O–Fe band was decreased in comparison with the band without HA, which was consistent with the observed decrease in coprecipitation density. In addition, the As–O–Fe band near 830 and 806 cm-1 was broadened, indicating that the addition of HA promotes the formation of ferric arsenate phase under alkaline conditions. XAS study Figure 7 shows XANES spectra for As–Fe-coprecipitated solid phases formed in the presence or absence of HA at a As/Fe molar ratio of 1 and pH 4, 7, and 10. The arsenic absorption edge energy positions provided by XANES provide an arsenic oxidation state. The absorption edge energies of the reference As(III) (NaAsO2) and As(V) (Na2HAsO47H2O) salts were positioned at 11,869 and 11,873 eV, respectively, which were indicated as vertical dotted lines in Fig. 7. The XANES spectra of the all As–Fe coprecipitates exhibited the arsenic absorption edge energy at *11,873 eV, suggesting that As(V) is the predominant valance state in the As–Fe coprecipitates regardless of the presence of HA. In the As–Fe coprecipitation experiments, As(V) was originally reacted with Fe(III). Thus, XANES study clearly showed that arsenic speciation was not transformed during the As–Fe coprecipitation and HA did not show any effect on the arsenic speciation. SEM study The SEM images of the As–Fe-coprecipitated solid phases indicate that the As–Fe coprecipitates were present in highly aggregated forms with nanoparticles of approximate diameters of 10–30 nm (Fig. 8). The particle sizes of the precipitates slightly decreased as pH increased, indicating that the ferrihydrite phase had smaller particles than the amorphous ferric arsenate phase. However, morphological differences among the precipitates formed at different pH conditions were not clearly observed. In the presence of HA, the As–Fe coprecipitates were present in more aggregated forms with smaller particles than those formed in the absence of HA. In addition, aggregation of the particles as rod shapes was observed, particularly at low pH conditions.
123
The SEM image of the HA solid phase, which was prepared by precipitation in acidic conditions, showed the aggregation of nanoparticles and the presence of some rod-shaped particles. In the presence of HA, the As–Fe coprecipitate particles seemed to form and grow on the surface of the HA. Conclusions In this study, the effects of NOM on arsenic coprecipitation with iron were studied under various As/Fe molar ratios and pH conditions. The presence of HA hindered the As–Fe coprecipitation at high pH and high As/Fe conditions probably due to the formation of soluble Fe–HA or As–Fe–HA complexes. The results of the XRD and FTIR studies showed that the As–Fe-coprecipitated solid phase was highly affected by pH and As/Fe molar ratio. Arsenic was coprecipitated with iron as an amorphous ferric arsenate phase at low pH and at high As/Fe molar ratios, while the formation of ferrihydrite and the subsequent association of arsenic with the ferrihydrite phase by adsorption were predominant at high pH and at low As/Fe molar ratios. The addition of HA also affected the As–Fe-coprecipitated solid phase depending on the As/Fe molar ratio under neutral and alkaline conditions. Iron oxides have been considered to be the most important sinks to attenuate the mobility of arsenic in the natural environment. (Dixit and Hering 2003; Smedley and Kinniburgh 2002). The results of our study clearly suggest that NOM could increase the mobility of arsenic even in the presence of iron oxides depending on the solution composition, such as pH and As/Fe molar ratio. Therefore, the bioavailability and risks of arsenic could be increased in NOM-rich environments. In addition, the application of iron oxides for arsenic removal in the presence of NOM should be more carefully evaluated. Acknowledgments This study was supported by Korea Environment Industry and Technology Institute (KEITI) through Geo-Advanced Innovation Action Program (2013000550006).
References Chen, Z., Cai, Y., Solo-Gabriele, H., Snyder, G. H., & Cisar, J. L. (2006). Interactions of arsenic and the dissolved substances derived from turf soils. Envionmental Science & Technology, 40, 4659–4665.
Environ Geochem Health Deng, Y., & Dixon, J. B. (2002). Soil organic matter and organic-mineral interactions. In J. B. Dixon & D. G. Schulze (Eds.), Soil mineralogy with environmental applications (Vol. 3). Madison, WI: Soil Science Society of America Inc. Dixit, S., & Hering, J. G. (2003). Comparison of arsenic(V) and arsenic(III) sorption onto iron oxide minerals: Implications for arsenic mobility. Envionmental Science & Technology, 37, 4182–4189. Ford, R. G. (2002). Rates of hydrous ferric oxide crystallization and the influence on coprecipitated arsenate. Envionmental Science & Technology, 36, 2459–2463. Goldberg, S., & Johnston, C. T. (2001). Mechanisms of arsenic adsorption on amorphous oxides evaluated using macroscopic measurements, vibrational spectroscopy, and surface complexation modeling. Journal of Colloid and Interface Science, 234, 204–216. Gomez, M. A., Assaaoudi, H., Becze, L., Cutler, J. N., & Demopoulos, G. P. (2009). Vibrational spectroscopy study of hydrothermally produced scorodite (FeAsO42H2O), ferric arsenate sub-hydrate (FAsH; FeAsO40.75H2O) and basic ferric arsenate sulfate (BFAS; Fe[(AsO4)1 - x(SO4)x(OH)x]wH2O). Journal of Raman Spectroscopy, 41, 212–221. Gu, B., Schmitt, J., Chen, Z., Liang, L., & Mccarthy, J. F. (1995). Adsorption and desorption of different organic matter fractions on iron oxide. Geochimica et Cosmochimica Acta, 59, 219–229. Hwang, B.-R., Kim, E. J., Yang, J.-S., & Baek, K. (2014). Extractive and oxidative removal of copper bound to humic acid in soil. Envionmental Science and Pollution Research,. doi:10.1007/s11356-014-3810-y. Jia, Y., Xu, L., Fang, Z., & Demopoulos, G. P. (2006). Observation of surface precipitation of arsenate on ferrihydrite. Envionmental Science & Technology, 40, 3248–3253. Jia, Y., Xu, L., Wang, X., & Demopoulos, G. P. (2007). Infrared spectroscopic and X-ray diffraction characterization of the nature of adsorbed arsenate on ferrihydrite. Geochimica et Cosmochimica Acta, 71, 1643–1654. Kim, E. J., Yoo, J.-C., & Baek, K. (2014). Arsenic speciation and bioaccessibility in arsenic-contaminated soils: Sequential extraction and mineralogical investigation. Environmental Pollution, 186, 29–35. Langmuir, D., Mahoney, J., & Rowson, J. (2006). Solubility products of amorphous ferric arsenate and crystalline scorodite (FeAsO42H2O) and their application to arsenic
behavior in buried mine tailings. Geochimica et Cosmochimica Acta, 70(12), 2942–2956. Lin, H.-T., Wang, M. C., & Li, G.-C. (2004). Complexation of arsenate with humic substance in water extract of compost. Chemosphere, 56, 1105–1112. Liu, G., Fernandez, A., & Cai, Y. (2011). Complexation of arsenite with humic acid in the presence of ferric iron. Envionmental Science & Technology, 45, 3210–3216. Pansu, M., & Gautheyrou, J. (2006). Handbook of soil analysis: Mineralogical, organic and inorganic methods. Heidelberg: Springer. Redman, A. D., Macalady, D. L., & Ahmann, D. (2002). Natural organic matter affects arsenic speciation and sorption onto hematite. Envionmental Science & Technology, 36, 2889–2896. Scott, D. T., Mcknight, D. M., Blunt-Harris, E. L., Kolesar, S. E., & Lovley, D. R. (1998). Quinone moieties act as electron acceptors in the reduction of humic substances by humics-reducing microorganisms. Envionmental Science & Technology, 32, 2984–2989. Sharma, P., Ofner, J., & Kappler, A. (2010). Formation of binary and ternary colloids and dissolved complexes of organic matter, Fe and As. Envionmental Science & Technology, 44, 4479–4485. Sharma, P., Rolle, M., Kocar, B., Fendorf, S., & Kappler, A. (2011). Influence of natural organic matter on As transport and retention. Envionmental Science & Technology, 45, 546–553. Smedley, P. L., & Kinniburgh, D. G. (2002). A review of the source, behavior and distribution of arsenic in natural waters. Applied Geochemistry, 17, 517–568. Stollenwerk, K. G. (2003). Geochemical processes controlling transport of arsenic in groundwater: A review of adsorption (Arsenic in ground water: Geochemistry and occurrence) (Vol. 3). Boston, Dordrecht, London: Kluwer Academic Publishers. Tokoro, C., Yatsugi, Y., Koga, H., & Owada, S. (2010). Sorption mechanisms of arsenate during coprecipitation with ferrihydrite in aqueous solution. Envionmental Science & Technology, 44, 638–643. Violante, A., Gaudio, S. D., Pigna, M., Ricciardella, M., & Banerjee, D. (2007). Coprecipitation of arsenate with metal oxides. 2. Nature, mineralogy and reactivity of iron(III) precipitates. Envionmental Science & Technology, 41, 8275–8280.
123
ORIGINAL PAPER
Effects of natural organic matter on the coprecipitation of arsenic with iron Eun Jung Kim • Bo-Ram Hwang • Kitae Baek
Received: 24 December 2014 / Accepted: 27 February 2015 Ó Springer Science+Business Media Dordrecht 2015
Abstract Natural organic matter (NOM) can affect arsenic speciation and mobility in the environment. In this study, the effects of NOM on the coprecipitation of arsenic with iron were investigated in order to better understand the fate and transport of arsenic in natural environments. The coprecipitation of arsenic with iron was studied in the presence and absence of NOM under various arsenic-to-iron molar ratios (As/Fe) and pH conditions. The addition of humic acid (HA) hindered the As–Fe coprecipitation under high pH and high As/Fe conditions by forming a soluble As–Fe– HA complex. The X-ray diffraction and Fourier transform infrared studies showed that the As–Fecoprecipitated solid phase was highly affected by pH and As/Fe. The arsenic was coprecipitated with iron as an amorphous ferric arsenate phase at a low pH level or high As/Fe conditions, while the formation of ferrihydrite phase and the arsenic incorporation to the ferrihydrite by adsorption was predominant at high pH levels or low As/Fe conditions. The HA affected the As–Fe-coprecipitated solid phase depending on the
E. J. Kim K. Baek (&) Department of Bioactive Material Sciences, Chonbuk National University, 567 Baekje-daero, Deokjin-gu, Jeonju, Jeollabukdo 561-675, Republic of Korea e-mail: [email protected] B.-R. Hwang K. Baek Department of Environmental Engineering, Chonbuk National University, 567 Baekje-daero, Deokjin-gu, Jeonju, Jeollabukdo 561-675, Republic of Korea
As/Fe molar ratio under neutral and alkaline conditions. Keywords Arsenic Iron Natural organic matter Coprecipitation
Introduction Arsenic is a naturally occurring trace element in the environment that has adverse effects on human health (Smedley and Kinniburgh 2002). In the natural water, mineral–water interactions such as adsorption/desorption, precipitation, and oxidation/reduction play important roles in controlling the mobility of arsenic. Among the many minerals, iron oxide minerals such as hydrous ferric oxide (HFO), goethite, and hematite are considered to be the most important sinks for arsenic because of their affinity to arsenic and their abundance in the environment (Smedley and Kinniburgh 2002; Stollenwerk 2003; Kim et al. 2014). Natural organic matter (NOM) has also been suggested as having effects on arsenic speciation and mobility in the environment (Sharma et al. 2010, 2011; Redman et al. 2002). NOM is a degradation product of biomolecules such as plants and animals, which is ubiquitous in both water and soil environments. Humic and fulvic acids are the two primary components of NOM having complex structures with various functional groups, which include hydroxyl, carboxylic
123
Environ Geochem Health
and phenolic groups (Deng and Dixon 2002). These functional groups are highly reactive with cationic metals and metal oxides and form both soluble and surface NOM–metal complexes (Redman et al. 2002). NOM has been reported to compete with arsenic for the surface sorption sites of iron oxides due to its high affinity to iron oxides (Sharma et al. 2010). In addition, the solubility of metal oxides such as iron oxides and the mobility of organic and inorganic contaminants associated with these metal oxides can be regulated by NOM due to the formation of soluble NOM–metal complexes (Deng and Dixon 2002). The soluble NOM–metal complexes can also bind strongly with dissolved anions, in which cationic metals act as bridging complexes (Lin et al. 2004). The formation of dissolved ternary As–Fe–NOM complexes has been suggested in environments containing arsenic, NOM, and iron ions or iron oxides (Sharma et al. 2010; Lin et al. 2004; Liu et al. 2011). NOM can also participate in redox reactions by serving as an electron shuttle, which may enhance the microbial reductive dissolution of iron oxides and transform As species (Redman et al. 2002; Scott et al. 1998). Therefore, the presence of NOM can greatly affect the speciation, fate, and mobility of arsenic in the environment (Smedley and Kinniburgh 2002). In iron-rich environments, the coprecipitation of arsenic with iron as well as arsenic adsorption on iron oxide minerals has been reported to occur (Smedley and Kinniburgh 2002; Ford 2002). Arsenic may be incorporated into the iron precipitates during the precipitation of soluble iron, which results in an arsenic-rich HFO phase or a ferric arsenate (FeAsO4)like phase depending on the solution conditions, such as pH and As/Fe molar ratio (Ford 2002; Jia et al. 2006; Violante et al. 2007; Tokoro et al. 2010). The presence of NOM can affect the As–Fe coprecipitate phase by forming Fe–NOM or As–Fe–NOM complexation. Furthermore, arsenic speciation can be influenced by the presence of NOM, which can affect the mobility of arsenic. However, exactly how the environmental conditions influence the arsenic mobility and coprecipitation of arsenic with iron in the presence of NOM is not yet fully understood. Therefore, the objective of this study was to investigate the effects of NOM on the arsenic mobility and coprecipitation of arsenic with iron in order to better understand the fate and transport of arsenic in natural environments. In this study, humic acid (HA) was
123
selected as a representative NOM, and the coprecipitation of arsenic with iron was studied in the presence of HA under various arsenic-to-iron molar ratios (As/Fe) and pH conditions. We evaluated the effects of NOM on arsenic coprecipitation with iron via solution-phase experiments and solid-phase characterization by X-ray diffraction (XRD), infrared (IR) spectroscopy, X-ray absorption spectroscopy (XAS), and scanning electron microscopy (SEM).
Materials and methods Materials HA was purchased from Sigma-Aldrich and purified by precipitation in a HCl solution following dissolution in a NaOH solution before the experiments (Hwang et al. 2014). The purchased HA was mixed in 1 M NaOH for 8 h, and the dissolved HA was separated by centrifugation. Then, the pH of the supernatant solution was adjusted to pH 2 with 1 M HCl and stirred for 24 h. The precipitated HA was separated by centrifugation, rinsed twice with 0.01 M HCl, and dried in an oven (60 °C) for 24 h. The elemental composition and functional groups of the purified HA were characterized by an elemental analyzer (Vario EL, Elementar, Germany) and a Fourier transform infrared (FTIR; Spectrum GX, Perkin Elmer, USA) spectrometer, respectively. The elemental analysis of the purified HA showed that it was composed of 44.2 % carbon, 3.75 % hydrogen, 1.07 % nitrogen, and 3.35 % sulfur. The FTIR spectra of the purified HA indicated several strong bands for the commonly observed functional groups in humic compounds (Fig. 1). Each band can be identified as follows: H bound to OH at wavenumber 3440 cm-1, aliphatic CH bonds at 2927 cm-1, C–O of COOH at 1713 cm-1, phenolic OH and aromatic C–C at 1615 cm-1, aliphatic CH at 1438 cm-1, C–O stretching at 1251 cm-1, and Si–O of the silicate impurities at 1050 cm-1 (Pansu and Gautheyrou 2006). The arsenic [As(V)] and iron [Fe(III)] solutions (0.1 M) were prepared by dissolving Na2HAsO47H2O (Sigma-Aldrich) and Fe(NO3)39H2O (Sigma-Aldrich), respectively. Coprecipitation experiments All of the experiments were conducted in a completely mixed batch reactor system. Initially, HA solutions
Transmittance %
Environ Geochem Health
4000
3000
2000
1000
-1
Wavenumber (cm )
Fig. 1 FTIR spectrum of purified humic acid
were prepared in various concentrations at pH 10. The arsenic and iron solutions were simultaneously added to HA solution, and then, pH was adjusted to 4, 7, and 10 with 0.5 M NaOH or 0.5 M HNO3 right after the addition of arsenic and iron solutions. The pH was not controlled throughout the reaction, and the final pH was measured after the reaction. The final HA concentrations in the solutions were 0, 100, 500, and 1000 mg/L. The iron concentration was fixed at 10 mM, and the arsenic concentration was varied in order to adjust the arsenic-to-iron molar ratio (As/Fe) to 0.1, 0.5, and 1. The reaction was continued in a temperature-controlled flatbed rotation incubator for 48 h at 20 °C and at 150 rpm. After the reaction, the samples were centrifuged at 8000 rpm for 20 min and filtered through 0.45-lm membrane filters. The concentrations of arsenic and iron in the filtrates were analyzed via inductively coupled plasma optical emission spectroscopy (ICP-OES; Agilent Tech, USA). Solids characterization The precipitated solid phases were dried and characterized via X-ray diffraction, infrared (IR) spectroscopy, and scanning electron microscopy with energy dispersive X-ray spectroscopy (SEM/EDX). The XRD data were collected on a powder X-ray diffractometer (MAX-2500, Rigaku, Japan) using Cu ˚ ) radiation obtained at 40 kV and Ka (k = 1.54059A 30 mA. The scans were performed between 5 and 70° 2h with 0.02° step sizes. The organic carbon contents
of the solid phases were analyzed using an elemental analyzer (Vario EL, Elementar, Germany). The transmittance spectra of the precipitates in the mid-infrared region (400–4000 cm-1) were obtained using an FTIR (Spectrum GX, Perkin Elmer, USA) spectrometer. In order to investigate arsenic species, selected samples were analyzed by X-ray absorption fine structure spectroscopy (XAFS) at beamline 10C at the Pohang Accelerator Laboratory, Korea. Solid phases were filled into Al sample holders and sealed with Kapton tape. The arsenic K-edge X-ray absorption near-edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) spectra were collected in fluorescence mode at room temperature using a Si(111) double-crystal monochromator. The morphology and particle size of the precipitates were characterized using a Hitachi SU-70 field emission scanning electron microscope (FE-SEM), and the chemical composition was evaluated using energy dispersive X-ray (EDX) spectroscopy.
Results and discussion Coprecipitation of arsenic with iron Arsenic coprecipitation with iron was studied under various arsenic-to-iron molar concentration ratios (As/ Fe) and pH conditions. In these experiments, the initial iron [Fe(III)] concentration was fixed at 10 mM, and the initial arsenic [As(V)] concentration was 1, 5, and 10 mM in order to adjust the As/Fe molar ratio to 0.1, 0.5, and 1. Figure 2 summarizes the remaining arsenic and iron concentrations in the solutions after As–Fe coprecipitation as a function of HA concentrations conducted at pH 4, 7, and 10. The decrease in arsenic from the solution indicated that the coprecipitation of arsenic with iron had occurred, which included arsenic adsorption to iron oxides and simultaneous arsenic precipitation with iron. Higher arsenic removal has been reported by coprecipitation of As(V) and Fe(III) than by simple As(V) adsorption on ferrihydrite, indicating that As–Fe coprecipitation involves more than adsorption (Tokoro et al. 2010). In addition, our results clearly show that the coprecipitation of arsenic with iron was highly affected by both solution pH and As/Fe molar ratio. The amounts of precipitated arsenic and iron were highest at pH 4 and decreased as pH increased. However, the effects of pH became weaker
123
Environ Geochem Health
(b)
12
10
pH 4 pH 7 pH 10
12
10
Iron Concentration (mmol/L)
Arsenic Concentration (mmol/L)
(a)
8
6
4
2
pH 4 pH 7 pH 10
8
6
4
2
0
0 0
200
400
600
800
1000
1200
0
200
Humic Concentration (mg/L)
(d)
6
5
pH 4 pH 7 pH 10
Iron Concentration (mmol/L)
Arsenic Concentration (mmol/L)
(c)
4
3
2
1
0
600
800
1000
1200
12
10
pH 4 pH 7 pH 10
8
6
4
2
0 0
200
400
600
800
1000
1200
0
200
Humic Concentration (mg/L)
400
600
800
1000
1200
1000
1200
Humic Concentration (mg/L)
(f)
(e) 0.7
0.6
6
pH 4 pH 7 pH 10
Iron Concentration (mmol/L)
Arsenic Concentration (mmol/L)
400
Humic Concentration (mg/L)
0.5
0.4
0.3
0.2
0.1
0.0
5
pH 4 pH 7 pH 10
4
3
2
1
0
0
200
400
600
800
1000
1200
Humic Concentration (mg/L)
0
200
400
600
800
Humic Concentration (mg/L)
Fig. 2 Concentrations of arsenic and iron remained in solutions after coprecipitation in the presence of HA at various As/Fe molar ratios (1, 0.5, and 0.1) and pH conditions (pH 4, 7, and 10). Initial iron concentration was 10 mM, and initial arsenic
concentration was 10, 5, and 1 mM for As/Fe of 1, 0.5, and 1, respectively. Concentrations of a arsenic at As/Fe = 1, b iron at As/Fe = 1, c arsenic at As/Fe = 0.5, d iron at As/Fe = 0.5, e arsenic at As/Fe = 0.1, f iron at As/Fe = 0.1
at lower As/Fe molar ratios. Even though the initial iron concentration was fixed for all of the conditions, greater iron precipitation was observed with a lower
initial arsenic concentration, i.e., lower As/Fe molar ratio. The presence of high concentrations of arsenic seemed to hinder the iron precipitation, particularly in
123
Environ Geochem Health
Effects of humic acid The effects of HA on arsenic coprecipitation with iron were studied by adding HA concentrations up to 1000 mg/L (Fig. 2). The addition of HA increased both the remaining arsenic and iron concentrations, particularly at high pH and high As/Fe conditions. At pH 4, the arsenic and iron precipitation was not
Coprecipitation density (mmol As/mmol Fe)
(a) 1.0 0.8 0.6 0.4 0.2 0.0
pH 4 pH 7 pH 10
-0.2 0
200
400
600
800
1000
1200
Humic Concentration (mg/L)
Coprecipitation density (mmol As/mmol Fe)
(b) 0.6 0.5 0.4 0.3 0.2 0.1
pH 4 pH 7 pH 10
0.0 0
200
400
600
800
1000
1200
Humic Concentration (mg/L)
(c) Coprecipitation density (mmol As/mmol Fe)
high pH conditions. In addition, the ferric arsenate phase, which was considered to be formed in high As/ Fe conditions, was only stable under acidic conditions, and its solubility increased as pH increased (Langmuir et al. 2006; Jia et al. 2006). Therefore, the increased concentrations of both iron and arsenic in high pH conditions could be partly affected by the increased solubility of the ferric arsenate phase. We also studied the coprecipitation density of arsenic, which indicates the molar concentrations of precipitated arsenic per molar concentration of precipitated iron. Figure 3 shows the coprecipitation density of arsenic as a function of HA concentration at various As/Fe molar ratios and pH conditions. The coprecipitation density of arsenic was primarily governed by the arsenic that was initially added to the iron molar concentration ratio and pH conditions. When the As/Fe molar ratio was one, the coprecipitation density of arsenic was 0.80, 0.72 and 0.44 at pH 4, 7, and 10, respectively. A decrease in coprecipitation density was observed with an increase in pH, which might have been due to the low affinity of arsenate to the iron oxides at high pH conditions. The arsenate sorption onto the iron oxides was high in acidic conditions and decreased with an increase in pH due to the increased repulsive electrostatic forces between anionic arsenic and negatively charged surface of iron oxides at high pH conditions (Dixit and Hering 2003). At a lower As/Fe ratio, the coprecipitation density was close to the initial As/Fe molar ratio at pH 4 and 7, indicating that all of the arsenic that was initially added has coprecipitated with the iron. This result might have been due to the presence of sufficient binding sites on the iron oxide at a low As/Fe molar ratio. However, a lower density was observed at pH 10 even with a low As/Fe molar ratio due to the low affinity between arsenic and iron oxides at high pH.
0.12 0.10 0.08 0.06 0.04 0.02
pH 4 pH 7 pH 10
0.00 0
200
400
600
800
1000
1200
Humic Concentration (mg/L)
Fig. 3 Coprecipitation density of arsenic calculated by precipitated arsenic concentrations per precipitated iron concentrations at arsenic-to-iron ratios of a As/Fe = 1, b As/Fe = 0.5, c As/Fe = 0.1 at various pH conditions
123
Environ Geochem Health
123
(a)
7 As/Fe=0.1 As/Fe=0.5 As/Fe=1
Fraction of Carbon (%)
6 5 4 3 2 1 0
4
7
10
pH
(b) 14 As/Fe=0.1 As/Fe=0.5 As/Fe=1
Fraction of Carbon (%)
12 10 8 6 4 2 0
4
7
10
pH
(c) 20 As/Fe=0.1 As/Fe=0.5 As/Fe=1
18
Fraction of Carbon (%)
affected by the addition of HA. However, at pH 7 and 10, the dissolved iron and arsenic concentrations increased as the HA concentration increased, and almost all of the arsenic and iron that were initially added remained in the solutions with addition of 1000 mg/L HA at pH 10. These results clearly indicated that HA inhibited the precipitation of both As and Fe at higher pH conditions. HA primarily influenced the arsenic mobility by increasing the solubility of the iron oxides and the mobility of arsenic by forming soluble HA–Fe complexes or As–Fe–HA (Sharma et al. 2010; Lin et al. 2004; Liu et al. 2011; Chen et al. 2006) and by inhibiting arsenic adsorption onto iron oxides due to its high affinity with iron oxides (Redman et al. 2002; Sharma et al. 2010). When we considered the coprecipitation density of arsenic (Fig. 3), even though the inhibition of arsenic and iron precipitation by HA was observed at pH 7, the coprecipitation density of arsenic was not affected by the addition of HA. This result indicates that the addition of HA inhibited arsenic and iron precipitation at the same rate in neutral pH conditions. HA might form soluble As–Fe–HA complex and increase solution arsenic and iron concentrations proportionally, which seemed to result in no change of coprecipitation density with HA. On the other hand, the coprecipitation density of arsenic decreased as the HA concentration increased at pH 10, indicating that less arsenic was associated with iron in the presence of HA. Under alkaline conditions, arsenic seems to be mobilized by competition with HA for sorption sites of iron oxide phase in addition to the soluble As–Fe–HA complex formation. In order to evaluate HA incorporation into the As– Fe coprecipitate, the organic carbon (OC) contents of the precipitates were further evaluated. Figure 4 shows the OC contents in the solid phase as a function of pH and As/Fe molar ratio. A decrease in OC fraction was observed with an increase in pH in most cases, which was consistent with the previously reported pH-dependent NOM adsorption on iron oxides. NOM adsorption on iron oxides is reported to be high in acidic conditions and decreases as pH increases (Gu et al. 1995). In addition, higher OC contents were observed at lower As/Fe molar ratios, probably due to the additional available complexation sites of the iron. However, no clear correlation was observed between HA content in the solid phase and As–Fe coprecipitation. High HA contents in the solid
16 14 12 10 8 6 4 2 0
4
7
10
pH
Fig. 4 Organic carbon contents of the As–Fe coprecipitates with various HA concentrations. a HA = 100 mg/L, b HA = 500 mg/L, c HA = 1000 mg/L
phase did not inhibit the As–Fe coprecipitation, which indicates that the competitive adsorption of HA with arsenic onto the iron oxides did not greatly affect the arsenic mobilization. HA seemed to affect the iron and
Environ Geochem Health
arsenic mobilization primarily through As–Fe–HA complexation, preventing solid-phase formation in high pH conditions. XRD study
Fig. 5 XRD patterns of the As–Fe-coprecipitated solid phases as a function of HA concentrations and initial As/Fe molar ratio reacted at a pH 4, b pH 7, and c pH 10
The XRD patterns of the As–Fe-coprecipitated solid phases are summarized in Fig. 5 as a function of initial As/Fe molar ratio and HA concentrations at pH 4, 7, and 10. The XRD patterns indicate the formation of poorly crystalline phases having broad bands near 30° 2h. Previous studies have reported the distinctions of the XRD spectra between two-line ferrihydrite and poorly crystalline ferric arsenate as follows: Two-line ferrihydrite shows two broad bands at *34° and *61° 2h, while poorly crystalline ferric arsenate has broad bands at *28° and *58° 2h (Jia et al. 2006; Tokoro et al. 2010). At pH 4, the major bands of the XRD spectra were located between the characteristic bands of the poorly crystalline ferric arsenate and the ferrihydrite depending on the As/Fe value (Fig. 4a). The major band of the XRD spectra was close to the poorly crystalline ferric arsenate band located at *28° 2h at a As/Fe molar ratio of 1. However, the major peak of the XRD spectrum was shifted to higher degree 2h with a decrease in the As/Fe molar ratio, and the XRD spectrum was close to that of the ferrihydrite when the As/Fe molar ratio was 0.1. This result suggests that arsenic was coprecipitated with iron as an amorphous ferric arsenate phase at high As/Fe molar ratios, while formation of ferrihydrite phase was predominant at low As/Fe molar ratios, and arsenic seems to be incorporated into the ferrihydrite phase by adsorption. The addition of HA did not show any clear effects on the XRD observed phase at low pH conditions as observed in the solution-phase experiments. At pH 7, the major XRD peak was located at *31° 2h, between 28° and 34° 2h, indicating that both the poorly crystalline ferric arsenate and the ferrihydrite phases were present. However, the presence of HA shifted the major band to higher degree 2h at a As/Fe molar ratio of 0.1. The HA seemed to promote the formation of the ferrihydrite phase at low As/Fe molar ratios. At pH 10, the major XRD band was located near 30° 2h at a As/Fe molar ratio of 1, while the main peak was shifted to higher degree 2h close to the ferrihydrite phase as the As/Fe decreases. Under alkaline conditions, both the ferrihydrite and the poorly crystalline
123
Environ Geochem Health
ferric arsenate phases seemed to be present at a As/Fe molar ratio of 1, while the formation of the ferrihydrite phase was predominant at a lower As/Fe molar ratio, as observed in lower pH conditions. However, addition of HA resulted in a broadened XRD band shifted to lower degree 2h at As/Fe molar ratios of 1 and 0.5, which indicates that the amorphous ferric arsenate phase was primarily present in the presence of HA. The solution-phase experiments showed that the As– Fe coprecipitation was highly inhibited by the addition of HA at pH 10, and only very small amounts of the solid phase were precipitated at a high concentration of HA. The HA seemed to form soluble Fe–HA and/or As–Fe–HA complexes and prevented ferrihydrite phase formation, which resulted in the presence of only poorly crystalline ferric arsenate phases in the solid precipitates as observed by XRD. However, the effects of HA were not as prominent at low As/Fe molar ratios. The results of the XRD study suggest that the addition of HA affected the As–Fe coprecipitation phase depending on the As/Fe molar ratio under neutral and alkaline conditions.
Fig. 6 FTIR spectra of As–Fe coprecipitates with As/Fe molar ratio of 1 at (a) HA = 0 mg/L and pH 4, (b) HA = 1000 mg/L and pH 4, (c) HA = 0 mg/L and pH 7, (d) HA = 1000 mg/L and pH 7, (e) HA = 0 mg/L and pH 10, and (f) HA = 1000 mg/ L and pH 10
123
FTIR study The As–Fe-coprecipitated solid phases were further characterized using FTIR. Figure 6 summarizes the infrared spectra of the As–Fe-coprecipitated solid phases formed in the presence or absence of HA at a As/Fe molar ratio of 1 and pH 4, 7, and 10. The infrared spectra of the As–Fe precipitates exhibited strong bands at 828–806 cm-1 wavenumber, which were attributed to the As–O stretching vibration of the As–O–Fe bonds (Jia et al. 2007; Gomez et al. 2009). The As–O–Fe stretching vibration band of the poorly crystalline ferric arsenate phase has been reported to be positioned at approximately 833–838 cm-1, whereas the As–O–Fe stretching vibration band of the surface complexed to the ferrihydrite phase was reported to be at 806–808 cm-1 (Jia et al. 2007; Goldberg and Johnston 2001). The As–O–Fe stretching vibration band of the precipitates that formed at pH 4 was located at 827 cm-1, which was similar to that of the amorphous ferric arsenate. However, as the pH increased, the As–O–Fe stretching vibration band gradually shifted to lower wavenumbers. The As–O– Fe bands of the precipitates were located at 822 and
Fig. 7 XANES spectra of As–Fe coprecipitates with As/Fe molar ratio of 1 at (a) HA = 0 mg/L and pH 4, (b) HA = 1000 mg/L and pH 4, (c) HA = 0 mg/L and pH 7, (d) HA = 1000 mg/L and pH 7, (e) HA = 0 mg/L and pH 10, and (f) HA = 1000 mg/L and pH 10
Environ Geochem Health
Fig. 8 FE-SEM images of As–Fe coprecipitates with As/Fe molar ratio of 1 at a HA = 0 mg/L and pH 4, b HA = 1000 mg/L and pH 4, c HA = 0 mg/L and pH 7, d HA = 1000 mg/L and pH 7, e HA = 0 mg/L and pH 10, and f HA = 1000 mg/L and pH 10
810 cm-1 at pH 7 and 10, respectively. In addition, a new band was observed at approximately 870 cm-1 at pH 10, which was attributed to the uncomplexed As–O bonds of the adsorbed arsenate (Goldberg and Johnston 2001). This result indicates that arsenic was coprecipitated with iron as an amorphous ferric arsenate phase at lower pH, but arsenic was removed by adsorption onto the ferrihydrite phase as pH increased. This result was consistent with the XRD characterization noted in the previous section. In addition, the infrared spectra of the precipitates exhibited a strong band at 1622–1640 cm-1 and a broad band located near 3400 cm-1, which were assigned to the water O–H bending mode and the
stretching vibration band of the O–H, respectively. The sharp band at 1385 cm-1 was due to NO32- ions, which were incorporated into the precipitates from the iron solution [Fe(NO3)39H2O]. When HA was added, new bands were observed at approximately 1100 and 1700 cm-1, which were the functional group bands for the HA such as C–O of COOH, phenolic OH, and aromatic C–C (Fig. 1). This result indicates that HA was incorporated into the As– Fe precipitates probably by adsorption, as observed in Fig. 4. In addition to the bands for HA functional group, no changes were observed for the As–O–Fe band at pH 4. On the other hand, the As–O–Fe band at pH 7 was shifted to a slightly lower wavenumber and
123
Environ Geochem Health
an additional band emerged at *870 cm-1, indicating that HA promoted formation of ferrihydrite phase and arsenic adsorption. However, at pH 10, the intensity of the As–O–Fe band was decreased in comparison with the band without HA, which was consistent with the observed decrease in coprecipitation density. In addition, the As–O–Fe band near 830 and 806 cm-1 was broadened, indicating that the addition of HA promotes the formation of ferric arsenate phase under alkaline conditions. XAS study Figure 7 shows XANES spectra for As–Fe-coprecipitated solid phases formed in the presence or absence of HA at a As/Fe molar ratio of 1 and pH 4, 7, and 10. The arsenic absorption edge energy positions provided by XANES provide an arsenic oxidation state. The absorption edge energies of the reference As(III) (NaAsO2) and As(V) (Na2HAsO47H2O) salts were positioned at 11,869 and 11,873 eV, respectively, which were indicated as vertical dotted lines in Fig. 7. The XANES spectra of the all As–Fe coprecipitates exhibited the arsenic absorption edge energy at *11,873 eV, suggesting that As(V) is the predominant valance state in the As–Fe coprecipitates regardless of the presence of HA. In the As–Fe coprecipitation experiments, As(V) was originally reacted with Fe(III). Thus, XANES study clearly showed that arsenic speciation was not transformed during the As–Fe coprecipitation and HA did not show any effect on the arsenic speciation. SEM study The SEM images of the As–Fe-coprecipitated solid phases indicate that the As–Fe coprecipitates were present in highly aggregated forms with nanoparticles of approximate diameters of 10–30 nm (Fig. 8). The particle sizes of the precipitates slightly decreased as pH increased, indicating that the ferrihydrite phase had smaller particles than the amorphous ferric arsenate phase. However, morphological differences among the precipitates formed at different pH conditions were not clearly observed. In the presence of HA, the As–Fe coprecipitates were present in more aggregated forms with smaller particles than those formed in the absence of HA. In addition, aggregation of the particles as rod shapes was observed, particularly at low pH conditions.
123
The SEM image of the HA solid phase, which was prepared by precipitation in acidic conditions, showed the aggregation of nanoparticles and the presence of some rod-shaped particles. In the presence of HA, the As–Fe coprecipitate particles seemed to form and grow on the surface of the HA. Conclusions In this study, the effects of NOM on arsenic coprecipitation with iron were studied under various As/Fe molar ratios and pH conditions. The presence of HA hindered the As–Fe coprecipitation at high pH and high As/Fe conditions probably due to the formation of soluble Fe–HA or As–Fe–HA complexes. The results of the XRD and FTIR studies showed that the As–Fe-coprecipitated solid phase was highly affected by pH and As/Fe molar ratio. Arsenic was coprecipitated with iron as an amorphous ferric arsenate phase at low pH and at high As/Fe molar ratios, while the formation of ferrihydrite and the subsequent association of arsenic with the ferrihydrite phase by adsorption were predominant at high pH and at low As/Fe molar ratios. The addition of HA also affected the As–Fe-coprecipitated solid phase depending on the As/Fe molar ratio under neutral and alkaline conditions. Iron oxides have been considered to be the most important sinks to attenuate the mobility of arsenic in the natural environment. (Dixit and Hering 2003; Smedley and Kinniburgh 2002). The results of our study clearly suggest that NOM could increase the mobility of arsenic even in the presence of iron oxides depending on the solution composition, such as pH and As/Fe molar ratio. Therefore, the bioavailability and risks of arsenic could be increased in NOM-rich environments. In addition, the application of iron oxides for arsenic removal in the presence of NOM should be more carefully evaluated. Acknowledgments This study was supported by Korea Environment Industry and Technology Institute (KEITI) through Geo-Advanced Innovation Action Program (2013000550006).
References Chen, Z., Cai, Y., Solo-Gabriele, H., Snyder, G. H., & Cisar, J. L. (2006). Interactions of arsenic and the dissolved substances derived from turf soils. Envionmental Science & Technology, 40, 4659–4665.
Environ Geochem Health Deng, Y., & Dixon, J. B. (2002). Soil organic matter and organic-mineral interactions. In J. B. Dixon & D. G. Schulze (Eds.), Soil mineralogy with environmental applications (Vol. 3). Madison, WI: Soil Science Society of America Inc. Dixit, S., & Hering, J. G. (2003). Comparison of arsenic(V) and arsenic(III) sorption onto iron oxide minerals: Implications for arsenic mobility. Envionmental Science & Technology, 37, 4182–4189. Ford, R. G. (2002). Rates of hydrous ferric oxide crystallization and the influence on coprecipitated arsenate. Envionmental Science & Technology, 36, 2459–2463. Goldberg, S., & Johnston, C. T. (2001). Mechanisms of arsenic adsorption on amorphous oxides evaluated using macroscopic measurements, vibrational spectroscopy, and surface complexation modeling. Journal of Colloid and Interface Science, 234, 204–216. Gomez, M. A., Assaaoudi, H., Becze, L., Cutler, J. N., & Demopoulos, G. P. (2009). Vibrational spectroscopy study of hydrothermally produced scorodite (FeAsO42H2O), ferric arsenate sub-hydrate (FAsH; FeAsO40.75H2O) and basic ferric arsenate sulfate (BFAS; Fe[(AsO4)1 - x(SO4)x(OH)x]wH2O). Journal of Raman Spectroscopy, 41, 212–221. Gu, B., Schmitt, J., Chen, Z., Liang, L., & Mccarthy, J. F. (1995). Adsorption and desorption of different organic matter fractions on iron oxide. Geochimica et Cosmochimica Acta, 59, 219–229. Hwang, B.-R., Kim, E. J., Yang, J.-S., & Baek, K. (2014). Extractive and oxidative removal of copper bound to humic acid in soil. Envionmental Science and Pollution Research,. doi:10.1007/s11356-014-3810-y. Jia, Y., Xu, L., Fang, Z., & Demopoulos, G. P. (2006). Observation of surface precipitation of arsenate on ferrihydrite. Envionmental Science & Technology, 40, 3248–3253. Jia, Y., Xu, L., Wang, X., & Demopoulos, G. P. (2007). Infrared spectroscopic and X-ray diffraction characterization of the nature of adsorbed arsenate on ferrihydrite. Geochimica et Cosmochimica Acta, 71, 1643–1654. Kim, E. J., Yoo, J.-C., & Baek, K. (2014). Arsenic speciation and bioaccessibility in arsenic-contaminated soils: Sequential extraction and mineralogical investigation. Environmental Pollution, 186, 29–35. Langmuir, D., Mahoney, J., & Rowson, J. (2006). Solubility products of amorphous ferric arsenate and crystalline scorodite (FeAsO42H2O) and their application to arsenic
behavior in buried mine tailings. Geochimica et Cosmochimica Acta, 70(12), 2942–2956. Lin, H.-T., Wang, M. C., & Li, G.-C. (2004). Complexation of arsenate with humic substance in water extract of compost. Chemosphere, 56, 1105–1112. Liu, G., Fernandez, A., & Cai, Y. (2011). Complexation of arsenite with humic acid in the presence of ferric iron. Envionmental Science & Technology, 45, 3210–3216. Pansu, M., & Gautheyrou, J. (2006). Handbook of soil analysis: Mineralogical, organic and inorganic methods. Heidelberg: Springer. Redman, A. D., Macalady, D. L., & Ahmann, D. (2002). Natural organic matter affects arsenic speciation and sorption onto hematite. Envionmental Science & Technology, 36, 2889–2896. Scott, D. T., Mcknight, D. M., Blunt-Harris, E. L., Kolesar, S. E., & Lovley, D. R. (1998). Quinone moieties act as electron acceptors in the reduction of humic substances by humics-reducing microorganisms. Envionmental Science & Technology, 32, 2984–2989. Sharma, P., Ofner, J., & Kappler, A. (2010). Formation of binary and ternary colloids and dissolved complexes of organic matter, Fe and As. Envionmental Science & Technology, 44, 4479–4485. Sharma, P., Rolle, M., Kocar, B., Fendorf, S., & Kappler, A. (2011). Influence of natural organic matter on As transport and retention. Envionmental Science & Technology, 45, 546–553. Smedley, P. L., & Kinniburgh, D. G. (2002). A review of the source, behavior and distribution of arsenic in natural waters. Applied Geochemistry, 17, 517–568. Stollenwerk, K. G. (2003). Geochemical processes controlling transport of arsenic in groundwater: A review of adsorption (Arsenic in ground water: Geochemistry and occurrence) (Vol. 3). Boston, Dordrecht, London: Kluwer Academic Publishers. Tokoro, C., Yatsugi, Y., Koga, H., & Owada, S. (2010). Sorption mechanisms of arsenate during coprecipitation with ferrihydrite in aqueous solution. Envionmental Science & Technology, 44, 638–643. Violante, A., Gaudio, S. D., Pigna, M., Ricciardella, M., & Banerjee, D. (2007). Coprecipitation of arsenate with metal oxides. 2. Nature, mineralogy and reactivity of iron(III) precipitates. Envionmental Science & Technology, 41, 8275–8280.
123
