Chemosphere 122 (2015) 88–93

Contents lists available at ScienceDirect

Chemosphere

journal homepage: www.elsevier .com/locate /chemosphere

Coprecipitated arsenate inhibits thermal transformation of 2-lineferrihydrite: Implications for long-term stability of ferrihydrite

http://dx.doi.org/10.1016/j.chemosphere.2014.11.0170045-6535/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author at: Southern Cross GeoScience, Southern Cross Univer-sity, Lismore, NSW 2480, Australia. Tel.: +61 2 6626 9544; fax: +61 2 6626 9499.

E-mail addresses: zhaohui.wang@scu.edu.au, zhaohuiwang@dhu.edu.cn(Z. Wang).

Zhaohui Wang a,b,⇑, Dongxue Xiao b, Richard T. Bush a, Jianshe Liu b

a Southern Cross GeoScience, Southern Cross University, Lismore, NSW 2480, Australiab College of Environmental Science and Engineering, Donghua University, Shanghai 201620, China

h i g h l i g h t s

� Thermal transformation of arsenate-coprecipitated ferrihydrite was examined.� The transformation products were well characterized by a series of techniques.� Arsenate was found to retard the thermal transformation of ferrihydrite.� The results have implications for predicting long-term stability of ferrihydrite.

a r t i c l e i n f o

Article history:Received 5 July 2014Received in revised form 21 October 2014Accepted 9 November 2014Available online 26 November 2014

Handling Editor: X. Cao

Keywords:HematiteElectron spin resonanceCo-precipitation2-line ferrihydriteArsenic

a b s t r a c t

2-line ferrihydrite, a ubiquitous iron oxy-hydroxide found in natural and engineered systems, is an effi-cient sink for the toxic metalloids such as arsenic. While much is known of the excellent capacity of fer-rihydrite to coprecipitate arsenate, there is little information concerning the long-term stability ofarsenate-accumulated ferrihydrite. By thermal treatment methodology, the expedited transformationof ferrihydrite in the presence of coprecipitated arsenate was studied at varying As/Fe ratios (0–0.5)and different heating temperature (40, 300, 450, 600 �C). Pure and transformed minerals were character-ized by thermogravimetry (TG), X-ray diffraction (XRD), Electron Spin Resonance (ESR), Scanning ElectronMicroscopy–Energy Dispersive X-ray Spectroscopy (SEM–EDX) and Fourier Transform Infrared Spectros-copy (FTIR). Arsenate was found to retard the thermal transformation of ferrihydrite. The extents of fer-rihydrite transformation to hematite decreased with increasing As/Fe ratios, but increased at a higherheating temperature. It is predicted that the coprecipitated arsenate can stabilize the amorphous ironoxides against the transformation to more crystalline solids. Arsenate concentration appears to play animportant role in this predicted long-term stability.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Arsenic (As) is a trace metalloid contaminant found in ground-waters and surface waters (Smith et al., 2000; Charlet and Polya,2006). In natural waters, arsenic usually occurs at low concentra-tions (1–10 lg L�1) (Smedley and Kinniburgh, 2002). Elevated lev-els of As are detected in the environments as a consequence ofanthropogenic activities and natural dissolution of As-bearing min-erals (Nriagu, 1994; Henke, 2009). Exceptionally high concentra-tions of arsenic (50–350 mg L�1) were determined in effluentsdischarged from mining and metallurgical industries (Smedley

and Kinniburgh, 2002; Vaughan, 2006). In the light of epidemiolog-ical evidence linking arsenic and cancers, WHO in 1993 reduced itsrecommended maximum acceptable level of arsenic in drinkingwater from 50 to 10 lg L�1 (WHO, 1993).

Adsorption of arsenate (As(V)) onto metal oxide surfaces is akey control on As mobility and fate in the environment, signifi-cantly affecting arsenic bioavailability (Raven et al., 1998; Huanget al., 2006; Anawar et al., 2008; Henke, 2009). 2-line Ferrihydrite(hereafter termed FHD), a common Fe(III) oxyhydroxide at theredox interface in soils and sediments, is a very efficient sink forarsenate, due to a large specific surface area and strong affinitytoward metals or metalloids ions (Jambor and Dutrizac, 1998;Michel et al., 2007; Das et al., 2011a; Mitsunobu et al., 2013). Inengineered systems, FHD is an ideal adsorbent for the treatmentof arsenic-polluted wastewater. Arsenic is generally removed fromacidic mineral processing effluents by neutralization using lime

Z. Wang et al. / Chemosphere 122 (2015) 88–93 89

and coprecipitated with ferric iron, in which FHD is produced dueto the hydrolysis of excess iron (Fe/As > 3).

There have been numerous studies on the efficiency and mech-anism of arsenate removal by FHD, however, to date, the long-termstability of the coprecipitated sludge is not fully assessed. Freshlyprecipitated iron oxides are metastable and can over time undergotransformation into more crystalline and stable minerals, such ashematite and goethite. This transformation is likely to increasethe risk of arsenic release due to the diminished capacity of thetransformed crystalline solids to bind arsenate. On the other hand,The presence of coprecipitated As(V) is known to inhibit the rate ofFHD transformation under either highly acidic (pH = 1.3, 25 �C)(Paige et al., 1997a) or alkaline (pH = 10 or 12, 70 �C) conditions(Paige et al., 1996; Das et al., 2011a). Das et al. (2011a) reportedthat the rate of FHD aqueous transformation decreased by 2 ordersof magnitude when the As/Fe ratio increased from 0.010 to 0.018.Similar to arsenate, other inorganic anions like silicate (Campbellet al., 2002) and phosphate (Paige et al., 1997b) can form surfacecomplexes that retard the dissolution and phase transformationof FHD. Therefore, determining the long-term stability of arse-nate-coprecipitated FHD is challenging because the expectedtransformation would proceed very slowly (up to 10–12 years) atambient temperature (Schwertmann et al., 2004). In order to studythe possible transformation within a manageable time frame, theprocesses have to be expedited considerably. By heating the sam-ples, a fast transformation of the amorphous iron oxides isexpected. The transformation of an amorphous iron oxide structureinduced by thermal treatment, as an useful experimental tech-nique, can be assumed to reflect the processes that may occur nat-urally over a long period of time at ambient temperature. Usingthis technique, Sørensen et al. (2000) studied long-term changesanticipated to occur in a disposal site receiving amorphous iron(hydr)oxides. To our best knowledge, no detailed information isavailable on the prediction of the fate of arsenic coprecipitatedFHD by dry heating methodology.

The objectives of this study were to investigate the thermaltransformation of FHD over a range of coprecipitated arsenate con-centrations. A series of heating experiments of As(V)-coprecipitat-ed FHD were performed over a temperature range from 40 to600 �C to investigate FHD transformation under a range of temper-ature gradients. Since transformation of pure ferrihydrite tohematite is expected to occur at T > 350 �C, we selected two tem-peratures below (40 �C and 300 �C) and above this transformationtemperature (450 �C and 600 �C) to assess the effect of arsenate onphase conversion. Transformation products of minerals were char-acterized by multiple techniques such as X-ray diffraction (XRD),Electron Spin Resonance (ESR) and Fourier Transform InfraredSpectroscopy (FTIR).

2. Materials and methods

2.1. Chemicals

Fe(NO3)3�9H2O, Na2HAsO4�7H2O, KI, MgCl2�6H2O, K2HPO4,NaBH4, NH2OH�HCl, NaOH, HCl (32%), ammonium acetate, ascorbicacid, methanol and phenanthroline were of analytical grade andused without further purification. MilliQ UltraPure water(18.2 MX cm) was used for all experiments.

2.2. Preparation of As(V)-coprecipitated FHD

Synthetic 2-line ferrihydrite was prepared according to themethods of Cornell and Schwertmann (2003). Briefly, Fe(NO3)3-

�9H2O and Na2HAsO4�7H2O were dissolved in 700 mL of MilliQwater at a fixed total molar ratio (As/Fe molar ratios = 0, 0.005,

0.012, 0.02, 0.05, 0.1 and 0.5). The solutions were titrated with1 M NaOH to pH of 7.5. The solid products were collected by cen-trifugation, washed four times by suspending in 50 mL of MilliQwater, followed by centrifugation. The obtained solids were driedat 40 �C for 24 h and then ground using an agate mortar. Dryingsamples at 40 �C does not apparently change the level of crystallin-ity. To examine the thermal transformation of As(V)-coprecipitated2-line ferrihydrite, samples in porcelain crucibles with lids wereheated to 300 �C, 450 �C and 600 �C in a muffle furnace (KSL1700X,MTI corporation) for 2 h. The cooled samples were stored at roomtemperature.

2.3. Solid-phase characterization

2.3.1. Thermogravity analysisTo determine the thermal stability and transformation temper-

ature of all synthetic solids, thermogravity analysis was conducted.Samples weighing between 18 and 22 mg were heated at a temper-ature ramp of 10 �C min�1 in static air, from room temperature to1000 �C, on a TGA/DSC simultaneous thermal analyzer (DTU-2A,Beijing BoYuan Precision Machinery Technology Co., Ltd.).

2.3.2. XRDXRD analyses were performed on the synthetic FHD to confirm

its purity, and on the heated samples for all As/Fe molar ratiosexperiments to identify the transformation products. A smallamount of methanol (�200 lL) was mixed with the dried mineralsample. The slurry was carefully dropped onto the specimen holderand evenly distributed. Samples were allowed to dry for at least15 min before XRD analyses were conducted. The powder XRD pat-terns were recorded using a Bruker D4 Endeavour XRD with aLynxeye position sensitive detector. Cobalt radiation was used at40 kV and 40 mA over a range of 5� and 80� 2h, a step size of0.03572� 2h and 1.65 s step�1. Crystalline materials were identifiedusing Bruker ‘‘Diffracplus EVA’’ Search/Match software and the ICDDPDF-2 database. The average crystallite size (L) of iron mineralswas calculated by Scherrer equation (Eq. (1)) (Scherrer, 1918).

L ¼ Kk=b cos h ð1Þ

where K is a constant related to crystallite shape, normally taken as0.9; k is the X-ray wavelength in nanometer (nm); b is the full widthat half maximum of diffraction peak in radians.

2.3.3. ESRESR spectra of all the heated samples were recorded, to identify

the Fe oxides according to their unique characteristic peaks, on aMS5000 ESR spectrometer (Magnettech, Freiberg Instruments)operating at X-band frequency with 100 kHz field modulation.The ESR signals were registered at microwave power 6.36 mW,sweep time 60 s and modulation amplitude 0.1 mT in the fieldrange of 48–600 mT.

2.3.4. Scanning Electron Microscopy (SEM)The morphology and elemental composition of the selected

As(V)-coprecipitated minerals were examined by Scanning Elec-tron Microscopy–Energy Dispersive X-ray Spectroscopy (SEM–EDX) using a Zeiss EVO/LS15 Scanning Electron Microscope withEDX. Samples for SEM examination were mounted onto pin-typealuminum stubs using double-sided carbon tape and coated withcarbon. Quantitive Energy Dispersive X-ray analysis including spotanalysis and mapping were undertaken at probe current of 500 pA,voltage (EHT) of 10 kV, working distance of 7.5 mm. EDX mappingwas used to relate element composition and spatial distribution tomicromorphological properties. Hereafter, the same samples weregold-coated for high resolution imaging.

90 Z. Wang et al. / Chemosphere 122 (2015) 88–93

2.3.5. FTIRFTIR was used to characterize changes of surface groups of sam-

ples with As/Fe = 0, 0.02, 0.1 and 0.5 and their thermal transforma-tion products. The experiments were performed on a Nicolet 6700FTIR spectrometer (Thermo Fisher) at lower than 0.09 cm�1 resolu-tion in transmission mode in the 4000–400 cm�1 range. The KBrpellets (1 cm in diameter) were prepared by mixing 1 mg of samplewith 100 mg of KBr.

2.3.6. Surface area measurementTo characterize changes of the specific surface area before and

after thermal treatment, Brunauer–Emmett–Teller (BET) surfacearea of samples with As/Fe = 0, 0.02 and their heated samples at600 �C were measured by a nitrogen adsorption/desorptionmethod on a TriStar II 3020 BET surface analyzer system at 77 K.Prior to measurement, the tested samples were degassed by N2

at 110 �C for 5 h.

3. Results and discussion

3.1. Thermogravity (TG)/differential thermal analysis (DTA)

Fig. S1 shows DTA curves for pure-ferrihydrite and As(V)-copre-cipitated ferrihydrite with As/Fe = 0.02, 0.05, which were heated to1000 �C. The low temperature (<300 �C) endotherm corresponds toa weight loss of 18%, indicating the expulsion of surface-adsorbedwater without disrupting the ferrihydrite structure (Xu et al.,2011). The exothermic deviation at 386 �C for FHD without anapparent weight loss is typical for the transformation of ferrihy-drite to hematite or the recrystallization of hematite (Towe andBradley, 1967; Chukhrov et al., 1973).

TG/DTA is very important to provide information on the ther-mal stability and transformation temperature Tt, which cannot beprovided by XRD or FTIR. Fig. 1 illustrates the dependence of Tt

on initial As/Fe ratios. Tt increased almost linearly from 386 �C to487 �C as As/Fe ratio increased from 0 to 0.02. Tt was up to703 �C when As/Fe ratio reached to 0.1. The presence of arsenateretarded the thermal transformation of ferrihydrite to hematite.To investigate the ferrihydrite ? hematite transformation in detail,dried samples with different As/Fe ratios were heated to 300 �C,450 �C and 600 �C. The prepared samples were characterized byXRD, ESR, FTIR and SEM.

0.00 0.02 0.04 0.06 0.08 0.10

400

450

500

550

600

650

700

Y=384.2+5197X

Y=423.7+2740.6X

Tran

sfor

mat

ion

tem

pera

ture

(°C

)

As/Fe ratio

Fig. 1. Effect of As/Fe ratio on the ferrihydrite to hematite transformationtemperature.

3.2. XRD

Fig. 2 and Table S1 show a comparison of the XRD patterns ofthe co-precipitated As(V) with FHD as a function of the initialmolar ratio of As/Fe. FHD has two characteristic broad peaks at2h of 41� (d � 2.6 Å) and 75� (d � 1.5 Å) (Fig. 2a). The XRD patternsof the co-precipitated products with an initial As/Fe ratio of 0.005–0.1 were almost identical to those of FHD, but the two peaks corre-sponding to FHD broadened. A larger As/Fe ratio (P0.05) resultedin a systematic decrease in the estimated crystallite size of theseminerals (Fig. S2), indicating the formation of FHD with lower crys-tallinity in the presence of an increasing As(V) levels. As the initialAs/Fe ratio increased to 0.5, two broad XRD bands but located atlower 2h position were observed: one at 2h 35 � (d � 3.0 Å) and asecond small peak 2h 71� (d � 1.5 Å). According to the literature(Jia et al., 2006), these two peaks are indicative of poorly crystallineferric arsenate (PCFA), and have been observed in As(V)-FHDcoprecipitation experiment at pH 7. Tokoro et al. (2009) proposedthat the mechanism of coprecipitation of As(V) changes fromadsorption to precipitation at a As/Fe ratio of 0.4. Therefore, atAs/Fe = 0.5, the product to form was PCFA, not As(V)-adsorbedFHD.

When heated at 300 �C for 2 h, the As(V)-FHD was transformedto hematite (HMT) at As/Fe = 0–0.012 (Fig. 2b), however, no peaksindicative of HMT were observed for samples with a higher As/Feratio (P0.02). This nicely demonstrates that a transformation ofFHD to HMT is inhibited by As(V) adsorption/coprecipitation. Thelack of a goethite intermediate is most likely because of the lackof FHD dissolution under dry thermal heating, mitigating the crit-ical solution phase for FHD transformation to goethite (Fischer andSchwertmann, 1975; Johnston and Lewis, 1983; Das et al., 2011b).The further temperature increase resulted in HMT formation at As/Fe = 0.02 and 0.05 for 450 �C, 0.1 for 600 �C, respectively. The crys-tallite sizes of FHD slightly increased with the increasing heatingtemperature (Fig. S3). At a lower As/Fe ratio (60.012), the crystal-lite sizes of newly formed HMT increased from 130 nm to 370 nm,in good agreement with reported values of 150–400 nm for HMT(Fig. S4) (Cornell and Schwertmann, 2003). The changes in crystal-lite sizes became insignificant at a larger As/Fe ratio, suggestingthat crystallinity of HMT was still subject to change in the copre-cipitated As(V).

The iron mineral transformations were accompanied by a sharpand significant decrease in surface area. For example, the surfacearea of coprecipitated samples with As/Fe = 0.02 decreased by94%, from 310 m2 g�1 at 40 �C to 17.5 m2 g�1 at 600 �C, which arein keeping with literature values of 200–400 m2 g�1 for FHD (Hsiand Langmuir, 1985; Cornell and Schwertmann, 2000) and<30 m2 g�1 for HMT (Cornell and Schwertmann, 2000; Song et al.,2009), respectively. For all heating treatments, XRD patterns andSEM morphology (data not shown) of samples with As/Fe = 0.5remained the same. The distribution of arsenate was not alteredby heating, remaining uniform before and after heat treatments.This is predicted by TG/DTA curve (data not shown) in which nodistinct exothermic peak appeared within the temperature range(30–1000 �C).

3.3. ESR

ESR is a very sensitive analytical method to characterize themineralogy of iron oxides according to their different g factors(Wang et al., 2009). FHD has a broad signal near g = 2 whileHMT is ESR-silent at the room temperature due to its purely anti-ferromagnetic nature (Martin, 1967; Gehring and Hofmeister,1994). Therefore, the overall conversion of FHD to HMT can bemonitored by ESR (Fig. 3). All of the samples prepared at 40 �Cexhibited a characteristic signal of FHD near g = 2. When heated

Fig. 2. XRD patterns for pure FHD, As(V)-coprecipitated FHD with varying As/Fe ratios and their thermal transformation products at (a) 40 �C, (b) 300 �C, (c) 450 �C, (d) 600 �C.The peaks of FHD are highlighted (dotted line). Two characteristic broad peaks at 2h of 41� (d � 2.6 Å) and 75� (d � 1.5 Å) were assigned to FHD; two peaks at 2h 35�(d � 3.0 Å) and 71� (d � 1.5 Å) were assigned to the poorly crystalline ferric arsenate; the three peaks at 2h of 28� (d � 3.7 Å), 39� (d � 2.7 Å) and 42� (d � 2.5 Å) were assignedto hematite (Cornell and Schwertmann, 2003).

Z. Wang et al. / Chemosphere 122 (2015) 88–93 91

to 300 �C, the g = 2 signal of the pure FHD disappeared, indicatinga complete conversion of FHD to HMT. Further heating to 450 �Cand 600 �C, samples at As/Fe = 0.005–0.02 and As/Fe = 0.05 weretransformed to HMT, respectively. A newly formed signal with ag value of 2 superimposed the broader spectrum of heated sam-ple with As/Fe = 0.5 at 450 �C. This new signal disappeared onheating to 600 �C. This observation indicates the distinct struc-tural changes of PCFA, which was not reflected by XRD measure-ment. ESR spectra, supplemented by XRD data, gave completeinformation about the structural changes of heated minerals.For example, as illustrated in Fig. 2b, the characteristic peaks ofHMT appeared for samples with As/Fe = 0.005 and 0.02, but theircorresponding ESR spectra still displayed the broader FHD signals.This implies the incomplete conversion of FHD to HMT underthese specific conditions. Once a full transformation was com-pleted, ESR signal of FHD completely disappeared (Fig. 3b–d)and well-shaped characteristic XRD peaks of HMT were observed(Fig. 2b–d).

Fig. 3. ESR spectra of pure FHD, As(V)-coprecipitated FHD with varying As/Fe ratios andThe field corresponding to g = 2 is highlighted (dotted line).

3.4. Mechanism consideration

It is established that dry heating of FHD results in the transfor-mation of FHD ? HMT (Jambor and Dutrizac, 1998), and the phasetransformation rate depends on surface water content and temper-ature. H2O tends to adsorb onto the coordination-unsaturated(CUS) sites that leads to particle aggregation. FTIR results inFig. 4 indicate the evolution of H2O at elevated temperatures,and reflected by the decrease in characteristic peak (1640,3340 cm�1) intensities of chemisorbed water (Jambor andDutrizac, 1998). However, at temperature >300 �C, the phase trans-formation is facilitated by dehydroxylation which occurs in theinterior of the particles. Continuous removal of OH from FHD struc-ture necessarily minimizes the average coordination number ofoxygen and OH around Fe (Zhao et al., 1994). It is evident thatthe adsorption band corresponding to Fe–OH complex(1350 cm�1) (Colombo et al., 2012) was absent after thermaltreatment (Fig. 4). The resulting charge imbalance and structural

their thermal transformation products at (a) 40 �C, (b) 300 �C, (c) 450 �C, (d) 600 �C.

Fig. 4. FTIR spectra for (a) pure FHD, As(V)-coprecipitated FHD with varying As/Fe ratios and their thermal transformation products. (b) As/Fe = 0.02, (c) As/Fe = 0.1, (d) As/Fe = 0.5. The adsorption bands at 1640 and 3340 cm�1 are assigned to molecular water (Jambor and Dutrizac, 1998). The band at 1350 cm�1 is assigned to a surface ferric iron-hydroxyl complex (Colombo et al., 2012). The peaks at 809, 811 and 819 cm�1 are assigned to adsorbed arsenate (Pena et al., 2006).

92 Z. Wang et al. / Chemosphere 122 (2015) 88–93

distortion favour the phase transformation. These dehydroxylatedparticles rapidly agglomerate to form larger hematite particles, asindicated in results of crystallite size (Fig. S4).

When FHD is being formed by precipitation in the presence ofarsenate, the polymerization process is prone to chain crosslinkingby arsenate complexes and the potential formation of Fe–O–M lay-ers (Zhao et al., 1994). The dehydroxylation may therefore beinhibited by the surface coverage of coprecipitated arsenate orFe–OAs layer. The coprecipitated arsenate groups are likely to filltetrahedral holes in the FHD structure (Waychunas et al., 1993.The incorporated arsenate further prevents the particle agglomer-ation and phase transformation. As a result, the formation of HMTmust take place at higher temperature (Figs. 1 and 2).

Das et al. (2014) assessed arsenate partitioning during thetransformation by aging synthetic FHD with adsorbed arsenateunder highly alkaline conditions (pH � 10). Their XAS data indi-cated adsorbed arsenate might be incorporated into the HMTstructure via both a bidentate-mononuclear complex and a biden-tate binuclear corner-sharing complex. Another X-ray absorptionspectroscopic (XAS) study by Bolanz et al. (2013) shows that arse-nate is not directly incorporated in the HMT, but is loosely linked inangelellite-like clusters within the hematite crystals. Elucidatingarsenate partitioning behaviour during thermal transformation ofFHD to HMT is a fascinating topic but beyond the scope of thepresent study. As a progression, the structural configuration ofincorporated arsenate during mineral transformation will befurther explored using XAS techniques.

4. Conclusion

The effects of coprecipitated arsenate on the thermal transfor-mation of ferrihydrite were investigated at varying As/Fe ratios(0–0.5) and different heating temperature (40, 300, 450, 600 �C).A variety of techniques such as TG, XRD, ESR, SEM–EDX and FTIRwere applied to characterize the mineral phases before and afterthermal treatment. It was found that arsenate is able to retardthe thermal transformation of ferrihydrite to hematite. The extentsof ferrihydrite transformation increased at an elevated tempera-ture, but decreased with increasing As/Fe ratios. The findings ofthis study are valuable for predicting the long-term changes ofarsenic-associated ferrihydrite wastes. The solubility of the formedcrystalline phase is orders of magnitude lower than that of amor-phous FHD (Stanjek and Weidler, 1992). The vulnerability of thesetransformed minerals to microbial iron reduction is expected to bealso considerably diminished. In addition, the capacity of the crys-talline solids to bind heavy metals or metalloids would be largelyreduced. In contrast, arsenate would retard the phase transforma-tion of FHD, thus significantly influencing the iron bioavailabilityand fate of trace metals-associated with FHD.

Acknowledgments

This work was financially supported by the Australian ResearchCouncil Linkage Grant Scheme (Grant No. LP110100732 ‘‘Electron

Z. Wang et al. / Chemosphere 122 (2015) 88–93 93

flow in iron hyper-enriched acidifying coastal environments’’), CRCCARE Project 6.6.01.06/07, National Natural Science Foundation ofChina (NSFC) (Grant No. 41273108) and Fundamental Researchfunds for Central Universities Central (13D111312). Z.H.W wishesto thank Dr. Michelle Bush for her kind help on XRD measurementand Dr. Maxine Dawes for her SEM operation.

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.chemosphere.2014.11.017.

References

Anawar, H.M., Garcia-Sanchez, A., Santa Regina, I., 2008. Evaluation of variouschemical extraction methods to estimate plant-available arsenic in mine soils.Chemosphere 70, 1459–1467.

Bolanz, R.M., Wierzbicka-Wieczorek, M., Caplovicová, M., Uhlík, P., Göttlicher, J.,Steininger, R., Majzlan, J., 2013. Structural incorporation of As5+ into hematite.Environ. Sci. Technol. 47, 9140–9147.

Campbell, A.S., Schwertmann, U., Stanjek, H., Friedl, J., Kyek, A., Campbell, P.A., 2002.Si incorporation into hematite by heating Si–ferrihydrite. Langmuir 18, 7804–7809.

Charlet, L., Polya, D.A., 2006. Arsenic in shallow, reducing ground-waters insouthern Asia: an environmental health disaster. Elements 2, 91–96.

Chukhrov, F.V., Zvyagin, B.B., Gorshkov, A.I., Yermilova, L.P., Balashova, V.V., 1973.Ferrihydrite. Int. Geol. Rev. 16, 1131–1143.

Colombo, C., Palumbo, G., Sellitto, V.M., Rizzardo, C., Tomasi, N., Pinton, R., Cesco, S.,2012. Characteristics of insoluble, high molecular weight iron–humicsubstances used as plant iron sources. Soil Sci. Soc. Am. J. 76, 1246–1256.

Cornell, R.M., Schwertmann, U., 2000. Iron Oxides in the Laboratory: Preparationand Characterization, second ed. Wiley-VCH, Hoboken, NJ.

Cornell, R.M., Schwertmann, U., 2003. The Iron Oxides, Properties, Reactions,Occurrences, and Uses. Wiley-VCH Verlag, Weinheim.

Das, S., Hendry, M.J., Essilfie-Dughan, J., 2011a. Effects of adsorbed arsenate on therate of transformation of 2-line ferrihydrite at pH 10. Environ. Sci. Technol. 45,5557–5563.

Das, S., Hendry, M.J., Essilfie-Dughan, J., 2011b. Transformation of two-lineferrihydrite to goethite and hematite as a function of pH and temperature.Environ. Sci. Technol. 45, 268–275.

Das, S., Essilfie-Dughan, J., Hendry, M.J., 2014. Arsenate partitioning fromferrihydrite to hematite: spectroscopic evidence. Am. Mineral. 99, 749–754.

Fischer, W., Schwertmann, U., 1975. The formation of hematite from amorphousiron(III) hydroxide. Clays Clay Miner. 23, 33–37.

Gehring, A.U., Hofmeister, A.M., 1994. The transformation of lepidocrocite duringheating: a magnetic and spectroscopic study. Clays Clay Miner. 42 (4), 409–415.

Henke, K., 2009. Arsenic�Environmental Chemistry, Health Threats and WasteTreatment. John Wiley & sons Ltd., Chichester, UK.

Hsi, C.D., Langmuir, D., 1985. Adsorption of uranyl onto ferric oxyhydroxide:application of the surface complexation site-binding model. Geochim.Cosmochim. Acta 49, 1931–1941.

Huang, R.Q., Gao, S.F., Wang, W.L., Staunton, S., Wang, G., 2006. Soil arsenicavailability and the transfer of soil arsenic to crops in suburban areas in FujianProvince, southeast China. Sci. Total Environ. 368, 531–541.

Jambor, J.L., Dutrizac, J.E., 1998. Occurrence and constitution of natural andsynthetic ferrihydrite, a widespread iron oxyhydroxide. Chem. Rev. 98, 2549–2585.

Jia, Y., Xu, L., Fang, Z., Demopoulos, G.P., 2006. Observation of surface precipitationof arsenate on ferrihydrite. Environ. Sci. Technol. 40, 3248–3253.

Johnston, J.H., Lewis, D.G., 1983. A detailed study of the transformation offerrihydrite to hematite in an aqueous medium at 92 �C. Geochim.Cosmochim. Acta 47, 1823–1831.

Martin, D.H., 1967. Magnetism in solids. MIT Press, Cambridge, Mass. p. 462.Michel, M.F., Ehm, L., Antao, S.M., Lee, P.L., Chupas, P.J., Liu, G., Strongin, D.,

Schoonen, M.A.A., Phillips, B.L., Parise, J.B., 2007. The structure of ferrihydrite, ananocrystalline material. Science 316, 1726–1729.

Mitsunobu, S., Muramatsu, C., Watanabe, K., Sakata, M., 2013. Behavior ofantimony(V) during the transformation of ferrihydrite and its environmentalimplications. Environ. Sci. Technol. 47, 9660–9667.

Nriagu, J.O., 1994. Arsenic in the Environment: Part I, Advances in EnvironmentalScience and Technology, vol. 26. John Wiley & Sons, New York.

Paige, C.R., Snodgrass, W.J., Nicholson, R.V., Scharer, J.M., 1996. The crystallization ofarsenate-contaminated iron hydroxide solids at high pH. Water Environ. Res.68, 981–987.

Paige, C.R., Snodgrass, W.J., Nicholson, R.V., Scharer, J.M., 1997a. An arsenate effecton ferrihydrite dissolution kinetics under acidic oxic condition. Water Res. 31,2370–2382.

Paige, C.R., Snodgrass, W.J., Nicholson, R.V., Scharer, J.M., He, Q.H., 1997b. The effectof phosphate on the transformation of ferrihydrite into crystalline products inalkaline media. Water, Air, Soil Pollut. 97, 397–412.

Pena, M., Meng, X., Korfiatis, G.P., Jing, C., 2006. Adsorption mechanism of arsenic onnanocrystalline titanium dioxide. Environ. Sci. Technol. 40, 1257–1262.

Raven, K.P., Jain, A., Loeppert, R.H., 1998. Arsenite and arsenate adsorption onferrihydrite: kinetics, equilibrium, and adsorption envelopes. Environ. Sci.Technol. 32, 344–349.

Scherrer, P., 1918. Bestimmung der grösse und der inneren struktur vonkolloidteilchen mittels röntgenstrahlen. Nachrichten von der Gesellschaft derWissenschaften zu Göttingen, mathematisch-physikalische Klasse 1918, 98–100.

Schwertmann, U., Stanjek, H., Becher, H.-H., 2004. Long-term in vitrotransformation of 2-line ferrihydrite to goethite/hematite at 4, 10, 15 and25 �C. Clay Miner. 39, 433–438.

Smedley, P.L., Kinniburgh, D.G., 2002. A review of the source, behaviour anddistribution of arsenic in natural waters. Appl. Geochem. 17, 517–568.

Smith, A.H., Lingas, E.O., Rahman, M., 2000. Contamination of drinking-water byarsenic in Bangladesh: a public health emergency. Bull. World Health Org. 78,1093–1103.

Song, F., Guan, J., Fan, X., Yan, G., 2009. Single-crystal star-like arrayed particles ofhematite: synthesis, formation mechanism and magnetic properties. J. AlloysCompd. 485, 753–758.

Sørensen, M.A., Stackpoole, M.M., Frenkel, A.I., Bordia, R.K., Korshin, G.V.,Christensen, T.H., 2000. Aging of iron (hydr) oxides by heat treatment andeffects on heavy metal binding. Environ. Sci. Technol. 34, 3991–4000.

Stanjek, H., Weidler, P.G., 1992. The effect of dry heating on the chemistry, surfacearea, and oxalate solubility of synthetic 2-line and 6-line ferrihydrites. ClayMiner. 27, 397–412.

Tokoro, C., Yatsugi, Y., Koga, H., Owada, S., 2009. Sorption mechanisms of arsenateduring coprecipitation with ferrihydrite in aqueous solution. Environ. Sci.Technol. 44, 638–643.

Towe, K.M., Bradley, W.F., 1967. Mineralogical constitution of colloidal hydrousferric oxides. J. Colloid Interface Sci. 24, 384–392.

Vaughan, D.J., 2006. Arsenic. Elements 2, 71–75.Wang, Z., Ma, W., Chen, C., Zhao, J., 2009. Light-assisted decomposition of dyes over

iron-bearing soil clays in the presence of H2O2. J. Hazard. Mater. 168, 1246–1252.

Waychunas, G.A., Rea, B.A., Fuller, C.C., Davis, J.A., 1993. Surface chemistry offerrihydrite: Part 1. EXAFS studies of the geometry of coprecipitated andadsorbed arsenate. Geochim. Cosmochim. Acta 57, 2251–2269.

WHO, 1993. Guidelines for Drinking Water Quality, vol. 1: Recommendations, 2nded. WHO, Geneva.

Xu, W., Hausner, D.B., Harrington, R., Lee, P.L., Strongin, D.R., Parise, J.B., 2011.Structural water in ferrihydrite and constraints this provides on possiblestructure models. Am. Mineral. 96, 513–520.

Zhao, J., Huggrns, F.E., Feng, Z., Huffman, G.P., 1994. Ferrihydrite: surface structureand its effects on phase transformation. Clays Clay Miner. 42, 737–746.