Phys Chem Minerals DOI 10.1007/s00269-013-0641-1

Original Paper

The structure and transformation of the nanomineral schwertmannite: a synthetic analog representative of field samples Rebecca A. French · Niven Monsegue · Mitsuhiro Murayama · Michael F. Hochella Jr. 

Received: 2 June 2013 / Accepted: 31 October 2013 © Springer-Verlag Berlin Heidelberg 2013

Abstract  The phase transformation of schwertmannite, an iron oxyhydroxide sulfate nanomineral synthesized at room temperature and at 75 °C using H2O2 to drive the precipitation of schwertmannite from ferrous sulfate (Regenspurg et al. in Geochim Cosmochim Acta 68:1185– 1197, 2004), was studied using high-resolution transmission electron microscopy. The results of this study suggest that schwertmannite synthesized using this method should not be described as a single phase with a repeating unit cell, but as a polyphasic nanomineral with crystalline areas spanning less than a few nanometers in diameter, within a characteristic ‘pin-cushion’-like amorphous matrix. The difference in synthesis temperature affected the density of the needles on the schwertmannite surface. The needles on the higher-temperature schwertmannite displayed a dendritic morphology, whereas the needles on the room-temperature schwertmannite were more closely packed. Visible Electronic supplementary material  The online version of this article (doi:10.1007/s00269-013-0641-1) contains supplementary material, which is available to authorized users. R. A. French · M. F. Hochella Jr. Environmental Nanoscience and Technology Laboratory, ICTAS, Virginia Tech, Blacksburg, VA 24061, USA R. A. French (*) · M. F. Hochella Jr. Department of Geosciences, Virginia Tech, 4044 Derring Hall, Blacksburg, VA 24061, USA e-mail: [email protected]

lattice fringes in the schwertmannite samples are consistent with the powder X-ray diffraction (XRD) pattern taken on the bulk schwertmannite and also matched d-spacings for goethite, indicating a close structural relationship between schwertmannite and goethite. The incomplete transformation from schwertmannite to goethite over 24 h at 75 °C was tracked using XRD and TEM. TEM images suggest that the sample collected after 24 h consists of aggregates of goethite nanocrystals. Comparing the synthetic schwertmannite in this study to a study on schwertmannite produced at 85 °C, which used ferric sulfate, reveals that synthesis conditions can result in significant differences in needle crystal structure. The bulk powder XRD patterns for the schwertmannite produced using these two samples were indistinguishable from one another. Future studies using synthetic schwertmannite should account for these differences when determining schwertmannite’s structure, reactivity, and capacity to take up elements like arsenic. The schwertmannite synthesized by the Regenspurg et al. method produces a mineral that is consistent with the structure and morphology of natural schwertmannite observed in our previous study using XRD and TEM, making this an ideal synthetic method for laboratory-based mineralogical and geochemical studies that intend to be environmentally relevant. Keywords  Schwertmannite · Goethite · Iron oxides · Nanomineral · High-resolution transmission electron microscopy · Oriented attachment

N. Monsegue · M. Murayama  Department of Materials Science, Virginia Tech, Blacksburg, VA 24061, USA

Introduction

M. Murayama  Nanoscale Characterization and Fabrication Laboratory, ICTAS, Virginia Tech, Blacksburg, VA 24061, USA

Schwertmannite is a Fe(III)-oxyhydroxysulfate nanomineral with the a proposed chemical formula of Fe8O8(OH)8−x

13



(SO4)x (1< x <1.75) (Bigham et al. 1994). Schwertmannite occurs naturally in acid mine drainage systems and acid sulfate soils (Regenspurg et al. 2004; Espana et al. 2005; Acero et al. 2006; Burton et al. 2007). It has also been found in the waste streams for zinc ore processing (Zinck and Dutrizac 1998; Claassen et al. 2002) and coal processing (Barham 1997) and was suggested as a possible candidate for a raw material for the pigment industry (Barham 1997). Its characteristic ‘pin-cushion’ morphology is made up of masses that are 200–500 nm in diameter that seem to consist of densely packed needles (up to a few tens of nm in width and up to a few hundred nm in length) and that also protrude from the mass surface (Bigham et al. 1994; Bigham and Nordstrom 2000). Therefore, schwertmannite is considered to be a nanomineral because these individual needles are between 1 and 100 nm in at least one of their dimensions (Hochella et al. 2008). Nanominerals are difficult to characterize due to their small size, as well as possible variable crystal structure, which can result from structural disorder, strain, and reconstructed surfaces (relative to the interior structure) (Hochella et al. 2008). Schwertmannite exhibits this structural disorder in its characteristically broad 8-line powder X-ray diffraction (XRD) pattern (Bigham et al. 1994). As a result, determining the structure of schwertmannite remains a challenge. Bigham et al. (1990) proposed a structure based on akaganéite (β-FeO(OH)) with sulfate occupying the tunnels formed by octahedral iron chains rather than chloride, as in akaganéite. The sulfate forms a bidentate complex with the iron in the tunnels, which distorts the structure. More recently, a pair distribution function (PDF) analysis of synchrotron powder diffraction data on synthetic (Schwertmann and Cornell 1991) and natural schwertmannite showed that schwertmannite consists of a ‘highly defective entangled network’ of an iron octahedra similar to akaganéite (FernandezMartinez et al. 2010). In this proposed structure, each unit cell contains two inner-sphere and two outer-sphere sulfate complexes. An electron nanodiffraction study (Loan et al. 2004) on synthetic schwertmannite did not show evidence for an akaganéite-like structure and found similarities between schwertmannite and 2-line ferrihydrite. Using high-resolution transmission electron microscopy (HRTEM), Hockridge et al. (2009) proposed that schwertmannite is composed of a ferrihydrite (Fe10O14(OH)2 (Michel et al. 2007) core with needles grown by oriented attachment of goethite (α-FeOOH) nanocrystals nucleating off the core (Hockridge et al. 2009). In natural systems and in waste streams, schwertmannite is found mixed with other iron oxide phases (Espana et al. 2005), making it difficult to obtain pure samples from field sites. Iron-oxidizing bacteria facilitate the precipitation of schwertmannite at low temperature (Bigham et al. 1996a). The difficulty of synthesizing

13

Phys Chem Minerals

schwertmannite in a bioreactor, or collecting it in the field, necessitates the use of synthetic schwertmannite for geochemical and mineralogical studies. Current synthetic methods include using elevated temperatures of 60–85 °C to drive hydrolysis of ferric iron. Schwertmann (2000) used ferric nitrate as a starting material, adding an additional step of dialysis to remove extraneous salts. Loan et al. (2004) used ferric sulfate, eliminating the dialysis step. A synthesis method developed by Regenspurg et al. (2004) used peroxide as the oxidizing agent to convert ferrous sulfate to schwertmannite at room temperature, avoiding raising the temperature and adding extraneous salts. The synthetic methods described above were run at temperatures ranging from room temperature to 85 °C, but few studies have focused on the effect of temperature on the produced schwertmannite’s structure and morphology. Studies investigating temperature have largely focused on its influence on the transformation of schwertmannite to other iron oxide phases. Most of these studies found that schwertmannite transforms to goethite at room temperature, but the time of transformation varied significantly from 79 to 543 days (Bigham et al. 1996b; Schwertmann and Carlson 2005; Jonsson and Lovgren 2006; Knorr and Blodau 2007). Knorr and Blodau (2007) found that after 109 days at 10 and 20 °C, schwertmannite transformed to an X-ray amorphous material with somewhat increased crystallinity, compared to schwertmannite, and lower sulfate content. The 10 ° increase in temperature also increased the transformation rate by a factor of 3.8. Davidson et al. (2008) studied the transformation kinetics of schwertmannite from 60 to 240 °C. They found that schwertmannite transforms to goethite at temperatures less than or equal to 80 °C, but at temperatures >80 °C, both goethite and hematite form. At 60 °C, goethite began to form after 200 min of reaction time, while at 80 °C, transformation occurred after only 30 min. In a study by Hockridge et al. (2009) that investigated the growth and transformation of schwertmannite at 85 °C, XRD patterns showed evidence of the presence of goethite after 60 min of reaction time. Whereas elevated temperatures drive the transformation of schwertmannite to goethite, low temperatures stabilize schwertmannite. In a study by Jonsson et al. (2005), schwertmannite remained stable for at least 5 years stored at 4 °C. In field studies of sediment profiles, fresh precipitates of schwertmannite overlie goethite, which suggests that upon aging, schwertmannite transforms to goethite (Schroth and Parnell 2005; Espana et al. 2005; Acero et al. 2006; Burton et al. 2007; Bush et al. 2007). A close structural relationship between schwertmannite and goethite could facilitate the phase transformation observed in the studies described above. Inherited structural motifs during phase transformation have been

Phys Chem Minerals

observed in previous studies of silicate minerals (Hochella and Banfield 1995). This has been seen in iron oxide systems as well. One study (Banfield et al. 2000) found that during the transformation from ferrihydrite to goethite, ferrihydrite nanoparticles are oriented such that there is continuity between ferrihydrite and goethite with respect to the closest packed oxygen planes. The highly defective and strained iron octahedral framework that Fernandez-Martinez et al. (2010) proposed for schwertmannite would also favor transformation to the more stable goethite structure. Fernandez-Martinez et al. (2010) noted that the schwertmannite-to-goethite transformation requires the release of sulfur and just one pair of iron octahedra such that the rectangular channel in goethite forms from schwertmannite’s square channel. In a study on schwertmannite collected in the Iberian Pyrite Belt of Spain, French et al. (2012) found that the majority of the observed lattice fringes in natural schwertmannite overlap with the most intense peak in schwertmannite’s XRD pattern located at 0.255 nm. The second and third most intense reflections in goethite’s XRD pattern—the (111) and (301) XRD reflections, respectively—are also located within the region occupied by the broad schwertmannite peak at 0.255 nm. In this study, the effect of temperature on the structure, morphology, and transformation of schwertmannite, synthesized using the Regenspurg et al. (2004) method at room temperature and at 75 °C, is presented. Despite the use of these schwertmannite samples in geochemical research (Burton et al. 2008, 2010; Burton and Johnston 2012), no studies have used HRTEM to investigate the structure of these synthetic schwertmannites. French et al. (2012) showed that HRTEM could reveal the structure and morphology of schwertmannite that was not apparent in previous studies relying only on XRD and/or pair distribution function data (Fernandez-Martinez et al. 2010). Similarly, the study presented here attempts to fill that gap in knowledge on synthetic schwertmannite using HRTEM.

Experimental procedure Synthesis The room temperature synthesis was based on a method in Regenspurg et al. (2004). Five grams of FeSO4 was dissolved in 500 mL of water (NANOpure®, <18 MΩ-cm, <1 ppb TOC) and placed in a 1,000 mL round-bottom flask with stirring. The reaction clock started when 2.5 mL of 30 % hydrogen peroxide (H2O2) was added to the solution. One synthesis was run at room temperature for 24 h. Another synthesis was run at 75 ± 2 °C for 24 h using a heating mantle and thermocouple (Barnant Co. Temperature Controller) to maintain the temperature over the course

of the experiment. In the experiment performed at 75 °C, the hydrogen peroxide was added after the FeSO4 solution had reached 75 °C. Characterization Samples were collected from the reaction flask for the powder XRD analyses of the 75 °C synthesis at intervals of 1, 3, 6, and 24 h. After 24 h of reaction time, samples were collected from the room-temperature synthesis at 24 h for XRD analysis. At each time interval, 25 mL of the suspension was removed from the reaction flask. The precipitate was separated from the 25 mL of suspension using a 0.2μm syringe filter. The precipitate was washed with 10 mL of water to remove iron sulfate salts. Samples were prepared for XRD analysis by suspending the precipitate in ethanol and pouring it on either a zero-background quartz or a zero-background silicon plate and left, covered, to airdry at room temperature. Samples were run on a PANalytical X’Pert PRO XRD system with Cu K-alpha radiation (45 kV–40 mA) from 5.06° 2θ to 69.916° 2θ with a step size of 0.067°. Samples for TEM were prepared from the same samples collected for the XRD analysis. The dried sample was suspended in ethanol and placed in a sonicating bath for two minutes. A droplet of the suspension was placed on a gold TEM grid with a lacey carbon support film, and the excess liquid was wicked away with a Kimwipe®. The TEM grids were stored in a desiccator under vacuum until analysis. Transmission electron micrographs and selected-area electron diffraction patterns were collected on a Philips EM420 operating at 120 kV. Diffraction patterns of individual needles on the surface of schwertmannite particles were collected using a microbeam electron diffraction technique with a small C2 aperture (small convergence angle). HRTEM images were collected on a FEI Titan operating at 300 kV. HRTEM images were analyzed using fast Fourier transform (FFT) in the Digital Micrograph™ software program with supplemental software [SADP Tools (Wu et al. 2012)] to measure d-spacings from FFT patterns. Where possible, lattice fringes were directly measured using ImageJ (Abramoff et al. 2004), confirming d-spacings determined from FFT analysis. Samples were found to be stable under the beam of the 300-kV FEI Titan during the time needed to collect an image (<1 min). Extended alignment and focusing were performed away from areas where images were collected for analysis to prevent beam damage. Comparing images where samples were deliberately damaged (Online Resource 1; Fig. ESM_1) with images of schwertmannite treated under normal operating procedures used for this study illustrated that all HRTEM images shown in this study represent undamaged schwertmannite needles.

13



Phys Chem Minerals

Results and discussion Comparison of structure and morphology of schwertmannite synthesized at room temperature and 75 °C using XRD and (HR)TEM X-ray diffraction patterns for schwertmannite produced at room temperature after 24 h of reaction time and at 75 °C for 1 h of reaction time display the characteristic schwertmannite pattern of broad peaks (Fig. 1). TEM images of the bulk aggregates of the schwertmannite particles (Fig. 2) show that schwertmannite produced at both temperatures displays the distinctive ‘pin-cushion’ morphology of needles extending from a dense core. However, the individual schwertmannite spheres produced at room temperature are on the order of 2–3 times larger and more electron dense than the spheres produced at 75 °C. The dendritic morphology of the schwertmannite produced at 75 °C and the more compact morphology of the schwertmannite produced at room temperature could be the result of rapid growth at higher temperatures. As reviewed by Waychunas (2001), in two-dimensional classical crystal growth, dendritic patterns arise because the solvent does not have time to flow

0.255

0.486

0.339

0.228 0.195

(a) (b) 10

20

30

40

°2θ

0.166

50

0.151 0.146

60

70

Fig. 1  Powder XRD patterns of precipitates synthesized at a room temperature for 24 h and b at 75 °C for 1 h. Peak locations (black lines) and d-spacings (nm) for schwertmannite (Bigham et al. 1994) are shown for reference

Fig. 2  Schwertmannite aggregate produced in room temperature synthesis a and aggregate of schwertmannite produced at 75 °C b for 1 h reaction time

13

away from fast growing crystal faces during rapid crystallization. The solvent can become trapped between growing crystal faces, impeding further growth. Alternatively, the rapid growth does not allow for local heating to decrease, which increases local solubility and impedes supersaturation at those crystal faces. In the case of non-classical crystal growth by aggregation of nanoparticles, dendritic patterns arise under diffusion-limited aggregation (DLA). DLA occurs when the rate of diffusion of the individual nanoparticle to the growing surface controls the rate of formation. Classical diffusion-limited crystal growth can also occur under rapid crystal growth conditions. Determining the mechanism of schwertmannite crystal growth is beyond the scope of our study, but a previous study (Hockridge et al. 2009) proposed that schwertmannite’s morphology was the result of needles made up of the oriented attachment of goethite nanoparticles, nucleating off a ferrihydrite core. The ferrihydrite core was observed in HRTEM images of particles formed at the initial stages of their synthesis that were thin enough to be electron transparent. In our study, no samples were collected at this early stage for comparison. Determining the structure and composition of the core of a fully formed schwertmannite particle remains a continuing challenge because resin embedding and ultramicrotome techniques used to prepare thin sections of the schwertmannite may distort the core and/or structure and can present problems of beam interaction with the resin (French et al. 2012). However, thin sections of natural schwertmannite in French et al. (2012) show that 70 % of the volume of a schwertmannite particle was made up of identifiable needles. If needles make up the majority of (or perhaps all of) a schwertmannite particle’s volume, then determining the structure of the needles at the surface can be used to understand its surface reactivity and to infer the structure and chemical makeup of schwertmannite mass’s internal material. Microbeam electron diffraction collected on needles on the surface of a schwertmannite core, synthesized at room

Phys Chem Minerals

temperature (Online Resource 1; Fig. ESM_2), shows two highly diffuse rings with the brighter ring corresponding to the most intense schwertmannite XRD peak at 0.255-nm d-spacing. The majority of d-spacings measured from lattice fringes observed in the HRTEM images matched with this most intense XRD peaks as well (Fig. 3 and Online Resource 1; Fig. ESM_3 and Fig. ESM_5). The needles average about 5 nm in diameter and exhibit an atomically rough surface with a hill and valley morphology. Lattice fringes, although visible in many areas, do not extend for more than a few nanometers. Portions of the needles that do not display lattice fringes can indicate amorphous areas, but they may also be an artifact of crystals that are not in the correct orientation with respect to the electron beam to give rise to fringes (Janney et al. 2000). In Fig. 3b, the circled area appears to contain a ~3-nm nanoparticle with coherent lattice fringes (see inset). No other aggregated nanoparticles with coherent fringing were observed. However, the hill and valley morphology, which was observed in all HRTEM images collected, is consistent with morphology observed in other studies (Banfield et al. 2000; Burleson and Penn 2006; Penn et al. 2006; Li et al. 2012) on iron oxide growth by aggregation of nanoparticles. Similar to the schwertmannite produced at room temperature, HRTEM images of schwertmannite produced at

75 °C (Fig. 4) also reveal needles with lattice fringes that extend for <5 nm within an amorphous matrix (Fig. 4a, b insets) and which exhibit a surface hill and valley morphology. In the few areas where lattice fringes were visible, the measured d-spacings matched with the XRD peaks at 0.146, 0.166, 0.195, 0.228, and 0.255 nm for the schwertmannite XRD pattern. Goethite also has d-spacings that align with the measured d-spacings on the schwertmannite sample (Online Resource 1; Fig. ESM_4 and Fig. ESM_5). Figure 5 shows needles on schwertmannite produced at 75 °C that are much longer and thicker than the needles described in Fig. 4. For example, the needle in Fig. 5a is ~150  × 20 nm. On average, the surfaces of these larger needles in Fig. 5 are similar to those described in Fig. 4. However, some faceting was observed on these needles (see Fig. 5a circle) that was not observed in needles like those in Fig. 4. The needle in Fig. 5b also exhibits the hill and valley morphology observed in schwertmannite produced at room temperature (Fig. 3) and in the smaller needles of schwertmannite produced at 75 °C (Fig. 4). The ‘hills’ are approximately 2 nm in diameter. The barely visible lattice fringes in Fig. 5 extend for <1 nm and are not measurable using an FFT analysis. A microbeam electron diffraction pattern of a needle of similar size and morphology to the needle in Fig. 5b showed individual spots within

Fig. 3  Needles at the surface of a schwertmannite particle produced in the room-temperature synthesis for 24 h. The arrows point to the corresponding needle shown in the insets in a. The black square marks the area shown in the inset in b. See Online Resource 1, Fig. ESM_3 for d-spacings measured from FFT analysis of these images

Fig. 4  Needles on the surface of schwertmannite particles produced at 75 °C for 1 h of reaction time. White squares mark the areas shown in the insets. See Online Resource 1 for d-spacings measured from FFT analysis of these images. d-spacings measured from a are in Fig. ESM_4 and for b see Fig. ESM_5

13



Phys Chem Minerals

Fig.  5  a, b Needles on the surface of schwertmannite particles formed at 75 °C for 1 h of reaction time. b The tip of the needle seen in the inset

diffuse rings (Online Resource 1, Fig. ESM_6). This suggests that the larger needles in Fig. 5 contain a greater percentage of single-crystal material or unidirectional oriented nanocrystals than the smaller needles like those in Fig.  3 from schwertmannite formed at room-temperature reaction conditions. A needle like the one shown in Fig. 3 exhibited only diffuse rings (Online Resource 1, Fig. ESM_2), indicative of polycrystalline and/or randomly oriented nanoparticles. The polyphasic needles, with an atomically rough surface and hill and valley surface morphology, observed in our study are significantly different than the schwertmannite needles observed in the Hockridge et al. (2009) study. In that study, the authors observed schwertmannite needles consisting of highly faceted goethite nanoparticles with coherent lattice fringes corresponding to the (101) plane of goethite. In our study, there were no observable lattice fringes with d-spacing corresponding with the (101) plane of goethite (Online Resource 1; Fig. ESM_3, Fig. ESM_4 and Fig. ESM_5). There is no peak in the schwertmannite XRD pattern that corresponds with the (101) goethite reflection; therefore, it is not surprising that this spacing was not observed in our HRTEM images. Hockridge et al. (2009) proposed that the absence of this peak was due to goethite only being present in the outer needles of the schwertmannite particles and not present in the core. They proposed that the goethite was too small a percentage of the total material to be observable in an XRD pattern. Most of the d-spacings measured on HRTEM images in our study correspond with the most intense peak in the schwertmannite XRD pattern at 0.255 nm (Online Resource 1; Fig. ESM_3 and Fig. ESM_5). Goethite also has diffraction peaks in this area, namely the (111), (400), (011), (210), and (301) planes, and our measured d-spacings are within error of matching these lines.

13

Transformation from schwertmannite to goethite at 75 °C In our study, the transformation from schwertmannite to goethite is observed in XRD patterns collected over 24 h (Fig. 6; Online Resource 1, Fig. ESM_7, and Fig. ESM_8). Within 3 h, characteristic goethite peaks appear (Fig.  6b). The (101) reflection at 21.4 °2θ is particularly noteworthy as it is absent in the schwertmannite XRD pattern (Fig. 6a). The goethite peaks increased in intensity as reaction time increased from 6 to 24 h. At 24 h, the (101) reflection for goethite is only slightly more intense that the (111) reflection for goethite. Goethite’s most intense peak is the (101) reflection with an intensity more than twice that of the (111) reflection (Fig. 6). This indicates that the sample has not completely converted to highly crystalline goethite and that schwertmannite may still be present in the sample and contributing to the XRD signal.

(101)

(111)

(d) (c) (b) (a) 10

20

30

40

50

60

70

°2θ Fig. 6  Powder XRD patterns showing the transformation from schwertmannite to goethite at 75 °C. Diffraction pattern from precipitate collected after a 1 h, b 3 h, c 6 h, and d 24 h of reaction time. Schwertmannite literature XRD reflection locations (Bigham et al. 1994) (dashed gray lines) and goethite standard peak locations (solid gray lines) are shown for reference (Cornell and Schwertmann 2003). The goethite (111) and (101) XRD reflections are labeled

Phys Chem Minerals

Fig. 8  TEM image of precipitate collected after 24 h of reaction time at 75 °C

to goethite. In images of precipitate collected after 6 h of reaction time (Fig. 7b), the characteristic spherical shape of schwertmannite is less visible and the sample is composed of aggregates of needles with little to no observable electrondense core. Although thicker than average needles are present at 1 and 3 h, these appear to have increased significantly after 6 h (Fig. 7b). After 24 h, the characteristic spherical shape of schwertmannite observed in Fig. 7a, b has disappeared entirely (Fig. 7c). At higher magnification, Fig. 8 shows that the sample collected after 24 h is composed of aggregated nanoparticles with varying contrast and no faceting. A Scherrer analysis of the (101) reflection of the peak in the XRD pattern collected after 24 h of reaction time (Fig. 6d) gave an approximate crystallite size of 20 nm, which is much smaller than the aggregates, further supporting the idea that these aggregates are made up of much smaller particles. Twenty nanometers is 5–10 times larger than the 2–4 nm diameter of the crystal circled in Fig. 8, but peak broadening only gives an average crystallite size and it is influenced by both structural strain and crystallite size. As nanominerals are subject to higher than average structural strain (Hochella et al. 2008), it is not unexpected that the Scherrer analysis may have significant error associated with it. Although there are limitations to using Scherrer analysis, the calculated peak width combined with the XRD pattern (Fig. 6d) and an SAED pattern (Online Resource 1, Fig. ESM_9), taken on samples collected after 24 h of reaction time, shows that goethite nanoparticles form during the transformation of schwertmannite to goethite under these reaction conditions. Fig. 7  TEM images of precipitate collected after a 3 h, b 6 h, and c 24 h of reaction time in the synthesis performed at 75 °C

Conclusions In TEM images of precipitate collected after 1 h (Fig. 2b) and 3 h (Fig. 7a) of reaction time, there are no readily observable changes in morphology as the schwertmannite transforms

Using HRTEM, our study reveals that the synthesis method used by Regenspurg et al. (2004), which uses ferrous

13



sulfate with hydrogen peroxide as an oxidant, at room temperature and at 75 °C, produces schwertmannite with the characteristics of a polyphasic nanomineral with crystalline areas spanning only a few nanometers within an amorphous matrix. This description is consistent with the study by French et al. (2012) on natural schwertmannite samples. In contrast, the Hockridge et al. (2009) method, which uses ferric sulfate heated to 85 °C, produces a material with needles made up of faceted goethite nanocrystals on the surfaces of what are otherwise apparently schwertmannite cores. The differences in structure and morphology between the schwertmannite produced by Regenspurg et al. (2004) and Hockridge et al. (2009) methods could directly affect the reactivity of these materials and therefore affect the conclusions of geochemical studies using these abiotic synthetic materials. The schwertmannite synthesized by the Regenspurg et al. (2004) method produces a mineral that is consistent with the structure and morphology of natural schwertmannite observed in our previous study using XRD and TEM, making this an ideal synthetic method for laboratory-based mineralogical and geochemical studies that intend to be environmentally relevant. Using an appropriate synthetic analog is of particular importance in the study of arsenic in acid mine drainage systems containing schwertmannite. Schwertmannite is known to take up arsenic in the field (Acero et al. 2006; Asta et al. 2010) and in the laboratory (Majzlan and Myneni 2005; Regenspurg and Peiffer 2005; Carlson et al. 2002). It is an ongoing area of study to determine the mechanism by which this occurs (Maillot et al. 2012). It is essential that future studies take into account the differences in structure and reactivity between schwertmannite produced by different synthetic methods when comparing results across studies and to field conditions. There is precedent for taking extra care when working with nanominerals even as a function of particle size and/or shape which can affect the catalytic oxidation of manganese by nanohematite (Madden and Hochella 2005), the hydroquinone-driven reductive dissolution of nanogoethite (Anschutz and Penn 2005), nanohematite’s sorption capacity for copper (Madden et al. 2006), and the dissolution behavior of nanohematite (Echigo et al. 2012) and nanogalena (Liu et al. 2008, 2009). Our study underscores the importance of using HRTEM to characterize nanominerals in addition to XRD in order to fully understand variations in atomic structure, defects and size and shape, and their resulting influences on reactivity. The rapid transformation from schwertmannite to goethite, driven by only a 50 °C increase in temperature in this study, demonstrates the highly metastable nature of schwertmannite. Although it is beyond the scope of this research to determine the mechanism by which schwertmannite transforms to goethite, this is the first study to track that transformation at 75 °C for 24 h using TEM and XRD.

13

Phys Chem Minerals

Clearly, there is evidence that goethite nanoparticles are an interim step in the transformation from schwertmannite to goethite, but how and why they form is not possible to deduce using the techniques employed in this study. However, these initial observations will serve as a useful baseline for future studies of the crystal growth and transformation of schwertmannite. As reviewed and explained in French et al. (2012) structural inheritance (Banfield et al. 2000; Hochella and Banfield 1995), the Ostwald step rule as modified for nanomaterials (Navrotsky 2004), kinetic factors (Majzlan and Myneni 2005), and oriented attachment (Banfield et al. 2000) may all play a role in the formation and transformation of schwertmannite. A recent landmark paper (Li et al. 2012) confirmed heretofore ex situ inferences of iron oxide crystal growth mechanisms by recording the in situ growth of iron oxide from the oriented attachment of ferrihydrite particles using a liquid cell mounted within a HRTEM. This study revealed that oriented attachment is driven by direction-specific interactions in combination with atomby-atom addition postoriented attachment. Ostwald ripening also played a role in the recrystallization of misaligned particles and the dissolution of small particles when they were near larger particles. The presence of sulfate in schwertmannite, which likely leads to its polyphasic makeup, makes it challenging to make inferences of crystal growth and phase transformation pathways from ex situ HRTEM studies like the one presented in this paper. Future work in this area may take advantage of the in situ technique used by Li et al. (2012) to finally reveal these pathways and, more importantly, how elements like arsenic are affected by that process. Acknowledgments  We thank the ICTAS Nanocharacterization and Fabrication Laboratory (NCFL) at Virginia Tech and the Materials Science and Engineering Nanoscale Materials Characterization Facility (MSE-NMCF) at the University of Virginia for the use of their Titan microscopes. Grants from the US Department of Energy (DE-FG02-06ER15786) and the Institute for Critical Technology and Applied Sciences at Virginia Tech provided major financial support for this project. We are also appreciative of the support from the National Science Foundation (NSF) and the Environmental Protection Agency through the Center for Environmental Implications of NanoTechnology (CEINT) funded under NSF Cooperative Agreement EF0830093. Fellowship support for this research was provided by the National Science Foundation (NSF IGERT grant DGE-0504196). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of NSF, EPA, or DOE.

References Abramoff MD, Magalhaes PJ, Ram SJ (2004) Image processing with ImageJ. Biophotonics Int 11(7):36–42 Acero P, Ayora C, Torrento C, Nieto JM (2006) The behavior of trace elements during schwertmannite precipitation and subsequent

Phys Chem Minerals transformation into goethite and jarosite. Geochim Cosmochim Acta 70(16):4130–4139 Anschutz AJ, Penn RL (2005) Reduction of crystalline iron(III) oxyhydroxides using hydroquinone: influence of phase and particle size. Geochem Trans 6(3):60–66 Asta MP, Ayora C, Roman-Ross G, Cama J, Acero P, Gault AG, Charnock JM, Bardelli F (2010) Natural attenuation of arsenic in the Tinto Santa Rosa acid stream (Iberian Pyritic Belt, SW Spain): the role of iron precipitates. Chem Geol 271(1–2):1–12. doi:10.1016/j.chemgeo.2009.12.005 Banfield JF, Welch SA, Zhang HZ, Ebert TT, Penn RL (2000) Aggregation-based crystal growth and microstructure development in natural iron oxyhydroxide biomineralization products. Science 289(5480):751–754 Barham RJ (1997) Schwertmannite: a unique mineral, contains a replaceable ligand, transforms to jarosites, hematites, and/or basic iron sulfate. J Mater Res 12(10):2751–2758 Bigham JM, Nordstrom DK (2000) Iron and aluminum hydroxysulfates from acid sulfate waters. In: Alpers CN, Jambor JL, Nordstrom DK (eds) Sulfate minerals-crystallography, geochemistry and environmental significance, vol 40, Reviews in mineralogy and geochemistryMineralogical Society of America, Washington, pp 351–403 Bigham JM, Schwertmann U, Carlson L, Murad E (1990) A poorly crystallized oxyhydroxysulfate of iron formed by bacterial oxidation of Fe(II) in acid-mine waters. Geochim Cosmochim Acta 54(10):2743–2758 Bigham JM, Carlson L, Murad E (1994) Schwertmannite, a new iron oxyhydroxysulphate from Pyhasalmi, Finland, and other localities. Mineral Mag 58(393):641–648 Bigham JM, Schwertmann U, Pfab G (1996a) Influence of pH on mineral speciation in a bioreactor simulating acid mine drainage. Appl Geochem 11(6):845–849 Bigham JM, Schwertmann U, Traina SJ, Winland RL, Wolf M (1996b) Schwertmannite and the chemical modeling of iron in acid sulfate waters. Geochim Cosmochim Acta 60(12):2111–2121 Burleson DJ, Penn RL (2006) Two-step growth of goethite from ferrihydrite. Langmuir 22(1):402–409 Burton ED, Johnston SG (2012) Impact of silica on the reductive transformation of schwertmannite and the mobilization of arsenic. Geochimica Et Cosmochimica Acta 96:134–153 Burton ED, Bush RT, Sullivan LA, Mitchell DRG (2007) Reductive transformation of iron and sulfur in schwertmannite-rich accumulations associated with acidified coastal lowlands. Geochim Cosmochim Acta 71(18):4456–4473 Burton ED, Bush RT, Sullivan LA, Mitchell DRG (2008) Schwertmannite transformation to goethite via the Fe(II) pathway: reaction rates and implications for iron-sulfide formation. Geochim Cosmochim Acta 72(18):4551–4564. doi:10.1016/ j.gca.2008.06.019 Burton ED, Johnston SG, Watling K, Bush RT, Keene AF, Sullivan LA (2010) Arsenic effects and behavior in association with the Fe(II)-catalyzed transformation of schwertmannite. Environ Sci Technol 44(6):2016–2021. doi:10.1021/es903424h Bush R, Burton E, Sullivan L (2007) Catalytic action of aqueous Fe(II) and S(II) on the transformation of schwertmannite to goethite. Geochim Cosmochim Acta 71(15):A137–A137 Carlson L, Bigham JM, Schwertmann U, Kyek A, Wagner F (2002) Scavenging of As from acid mine drainage by schwertmannite and ferrihydrite: a comparison with synthetic analogues. Environ Sci Technol 36(8):1712–1719 Claassen JO, Meyer EHO, Rennie J, Sandenbergh RF (2002) Iron precipitation from zinc-rich solutions: defining the Zincor Process. Hydrometallurgy 67(1–3):87–108 Cornell RM, Schwertmann U (2003) The iron oxides, 2nd edn. WileyVCH, Weinheim

Davidson LE, Shaw S, Benning LG (2008) The kinetics and mechanisms of schwertmannite transformation to goethite and hematite under alkaline conditions. Am Mineral 93(8–9):1326–1337. doi:10.2138/am2008.2761 Echigo T, Aruguetea DM, Murayama M, Hochella MF (2012) Influence of size, morphology, surface structure, and aggregation state on reductive dissolution of hematite nanoparticles with ascorbic acid. Geochimica Et Cosmochimica Acta 90:149–162 Espana JS, Pamo EL, Santofimia E, Aduvire O, Reyes J, Barettino D (2005) Acid mine drainage in the Iberian Pyrite Belt (Odiel river watershed, Huelva, SW Spain): geochemistry, mineralogy and environmental implications. Appl Geochem 20(7):1320–1356. doi:10.1016/j.apgeochem.2005.01.011 Fernandez-Martinez A, Timon V, Roman-Ross G, Cuello GJ, Daniels JE, Ayora C (2010) The structure of schwertmannite, a nanocrystalline iron oxyhydroxysulfate. Am Mineral 95:1312–1322 French RA, Caraballo MA, Kim B, Rimstidt J, Murayama M, Hochella MF (2012) The enigmatic iron oxyhydroxysulfate nanomineral schwertmannite: morphology, structure, and composition. Am Mineral 97(8–9):1469–1482 Hochella MF, Banfield JF (1995) Chemical weathering of silicates in nature: a microscopic perspective with theoretical considerations. In: White AF, Brantley SL (eds) Chemical weathering rates of silicate minerals, vol 31., vol 31The Mineralogical Society of America, Washington, pp 353–406 Hochella MF, Lower SK, Maurice PA, Penn RL, Sahai N, Sparks DL, Twining BS (2008) Nanominerals, mineral nanoparticles, and Earth systems. Science 319(5870):1631–1635 Hockridge JG, Jones F, Loan M, Richmond WR (2009) An electron microscopy study of the crystal growth of schwertmannite needles through oriented aggregation of goethite nanocrystals. J Cryst Growth 311(15):3876–3882. doi:10.1016/j.jcrysgro.2009.06.023 Janney DE, Cowley JM, Buseck PR (2000) Structure of synthetic 2-line ferrihydrite by electron nanodiffraction. Am Mineral 85(9):1180–1187 Jonsson J, Lovgren L (2006) Precipitation of secondary Fe(III) minerals from acid mine drainage. Appl Geochem 21(3):437–445 Jonsson J, Persson P, Sjoberg S, Lovgren L (2005) Schwertmannite precipitated from acid mine drainage: phase transformation, sulphate release and surface properties. Appl Geochem 20(1):179–191 Knorr KH, Blodau C (2007) Controls on schwertmannite transformation rates and products. Appl Geochem 22(9):2006–2015 Li D, Nielsen MH, Lee JRI, Frandsen C, Banfield JF, De Yoreo JJ (2012) Direction-specific interactions control crystal growth by oriented attachment. Science 336:1014–1018 Liu J, Aruguete DA, Jinschek JR, Rimstidt JD, Hochella MF (2008) The non-oxidative dissolution of galena nanocrystals: insights into mineral dissolution rates as a function of grain size, shape, and aggregation state. Geochim Cosmochim Acta 72(24):5984– 5996. doi:10.1016/j.gca.2008.10.010 Liu J, Aruguete DM, Murayama M, Hochella MF (2009) Influence of size and aggregation on the reactivity of an environmentally and industrially relevant manomaterial (PbS). Environ Sci Technol 43(21):8178–8183. doi:10.1021/es902121r Loan M, Cowley JM, Hart R, Parkinson GM (2004) Evidence on the structure of synthetic schwertmannite. Am Mineral 89(11–12):1735–1742 Madden AS, Hochella MF (2005) A test of geochemical reactivity as a function of mineral size: manganese oxidation promoted by hematite nanoparticles. Geochim Cosmochim Acta 69(2): 389–398 Madden AS, Hochella MF, Luxton TP (2006) Insights for sizedependent reactivity of hematite nanomineral surfaces through Cu2+ sorption. Geochim Cosmochim Acta 70(16):4095–4104. doi:10.1016/j.gca.2006.06.1366

13

Maillot F, Morin G, Juillot F, Bruneel O, Casiot C, Ona-Nguema G, Wang Y, Lebrun S, Aubry E, Vlaic G (2012) Structure and reactivity of as (III)-and As (V)-rich schwertmannites and amorphous ferric arsenate sulfate from the Carnoulès Acid Mine Drainage. Comparison with biotic and abiotic model compounds and implications for As remediation. Geochimica Et Cosmochimica Acta, France Majzlan J, Myneni SCB (2005) Speciation of iron and sulfate in acid waters: aqueous clusters to mineral precipitates. Environ Sci Technol 39(1):188–194 Michel FM, Ehm L, Antao SM, Lee PL, Chupas PJ, Liu G, Strongin DR, Schoonen MAA, Phillips BL, Parise JB (2007) The structure of ferrihydrite, a nanocrystalline material. Science 316(5832):1726–1729. doi:10.1126/science.1142525 Navrotsky A (2004) Energetic clues to pathways to biomineralization: precursors, clusters, and nanoparticles. Proc Natl Acad Sci USA 101(33):12096–12101 Penn RL, Erbs JJ, Gulliver DM (2006) Controlled growth of alphaFeOOH nanorods by exploiting-oriented aggregation. J Cryst Growth 293(1):1–4. doi:10.1016/j.jcrysgro.2006.05.005 Regenspurg S, Peiffer S (2005) Arsenate and chromate incorporation in schwertmannite. Appl Geochem 20(6):1226–1239 Regenspurg S, Brand A, Peiffer S (2004) Formation and stability of schwertmannite in acidic mining lakes. Geochim Cosmochim Acta 68(6):1185–1197

13

Phys Chem Minerals Schroth AW, Parnell RA (2005) Trace metal retention through the schwertmannite to goethite transformation as observed in a field setting, Alta Mine, MT. Appl Geochem 20(5):907–917. doi:10.1016/j.apgeochem.2004.09.020 Schwertmann U (2000) Iron oxides in the laboratory: preparation and characterization, 2nd edn. Wiley-VCH, Weinheim, New York Schwertmann U, Carlson L (2005) The pH-dependent transformation of schwertmannite to goethite at 25 °C. Clay Miner 40(1):63–66 Schwertmann U, Cornell RM (1991) Iron oxides in the laboratory: preparation and characterization. Wiley-VCH, New York Waychunas GA (2001) Structure, aggregation and characterization of nanoparticles. In: Banfield JF, Navrotsky A (eds) Nanoparticles and the environment, vol 44., Reviews in mineralogy and geochemistry, vol 44The Mineralogical Society of America, Washington, pp 105–166 Wu CH, Reynolds WT Jr, Murayama M (2012) A software tool for automatic analysis of selected area diffraction patterns within Digital Micrograph™. Ultramicroscopy 112:10–14 Zinck JM, Dutrizac JE (1998) Environment—The behaviour of zinc, cadmium, thallium, tin and selenium during ferrihydrite precipitation from sulphate media. CIM Bull 91(1019):94–101