Stabilizing Reactive Fe(III) Clusters by Freeze-Dry ... - ACS Publications

Apr 22, 2019 - aqueous solution to obtain 10 mL of FS4 solution (with a total of 0.4 ..... (PDF). Accession Codes. CCDC 1885741−1885744 contain the ...
0 downloads 0 Views 2MB Size
Download PDF
Article Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX

pubs.acs.org/IC

Stabilizing Reactive Fe(III) Clusters by Freeze-Dry/Solvent-Exchange To Benchmark Iron Hydrolysis Pathways Wei Wang,*,† Mehran Amiri,‡ Tao Huang,† Lev N. Zakharov,‡ Yining Zhang,*,† and May Nyman*,‡ †

Key Laboratory of Optoelectronic Materials Chemistry and Physics, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, 155 Yangqiao Road West, Fuzhou, Fujian 350002, People’s Republic of China ‡ Department of Chemistry, Oregon State University, 153 Gilbert Hall, Corvallis, Oregon 97331, United States

Inorg. Chem. Downloaded from pubs.acs.org by STOCKHOLM UNIV on 04/22/19. For personal use only.

S Supporting Information *

ABSTRACT: Isolating Fe(III) clusters from water without stabilizing ligands is significantly challenged by the high acidity of Fe3+-bound water, leading to uncontrolled precipitation of iron oxyhydroxides. Here we demonstrate a freeze-drying solventexchange method that enabled the isolation of a metastable Fe(III) sulfate decameric cluster formulated [Fe10O2(SO4)12(OCH3)2]·14CH3OH (Fe10). Without stabilization by solvent-exchange, the aqueous species undergoes rapid conversion to the iron sulfate mineral schwertmannite. Monitoring the hydrolysis process from cluster intermediates to schertmannite by small-angle X-ray scattering, we observe the progression from Fe10 to 37 Å soluble nanoparticles prior to the precipitation process. This condensation behavior of Fe10 is further exploited to develop a simple laboratory synthesis of schwetmannite. In addition, we demonstrate the versatility of the freeze-drying solventexchange method by isolating Al(III), Zn(II), and Cd(II) substituted Fe(III) sulfate clusters. The freeze-drying solvent-exchange method provides a unique opportunity to isolate cluster intermediates and models to aid in our understanding of metal-ion hydrolysis processes in environmental, material science, and geological studies.

■

INTRODUCTION Nucleation and growth of iron oxyhydroxide solids from aqueous solution is a fundamental process of great importance in chemistry, materials, biotechnology, and earth sciences.1−4 In recent years, key iron oxides (e.g., magnetite and ferrihydrite)4−8 have been proposed to grow from water via the attachment of nanoparticle/cluster intermediates with a narrow size distribution,1,9 instead of via the classic atom-byatom growth of monomer Fe 3+-aqua ions. While the thermodynamics of nanoparticle/cluster formation and subsequent aggregation is a focus of these prior studies, clusters intermediate in size between the Fe3+-monomers and nanoparticles (containing hundreds of atoms) remain enigmatic. This is because the rapid hydrolysis−condensation kinetics of Fe3+ species10,11 prevents isolation of intermediates.3,12−14 This challenge is pertinent to understand the growth of metal oxides from water in general, and developing synthetic strategies to isolate and characterize relevant intermediates is critical for investigating nucleation−aggregation processes.12 The key to isolating aqueous Fe3+-clusters is suppressing the rapid Fe3+ hydrolysis and ligand lability (the water exchange rate of hydrated Fe3+ is about 100 times faster than Al3+ aquaion11). This is typically approached by introducing coordinating ligands (often polydentate), which stabilize the iron-oxo core and provide an opportunity to isolate cluster intermediates. However, the presence of coordinating ligands might influence the cluster speciation,15,16 and the result may not necessarily reflect the condensation process. To the best of our © XXXX American Chemical Society

knowledge, there are few examples of iron-oxohydroxide clusters without stabilizing ligands.17 Without the help from coordinating ligands, replacing the bound water with nonaqueous solvent can be an effective approach.11 It is known that the alkoxy-bridged iron clusters are reliable structure models for the corresponding OHbridged clusters existing in aqueous solution.18,19 In this work, we isolate an Fe(III) sulfate cluster intermediate by a freezedrying solvent-exchange method, formulated [Fe10O2(SO4)12(OCH3)2]·14CH3OH (Fe10). It contains two [Fe3O(SO4)6(CH3OH)3]5− units, whose aqueous analogue can be found in nature. The small-angle X-ray scattering (SAXS) analysis reveals that the Fe(III) condensation initiates at pH as low as 2.0. With the Fe10 model cluster (and its constituent Fe3 trimers), we are able to resolve via SAXS the early stage Fe3+hydrolysis and condensation in the Fe3+ sulfate solution, which eventually leads to schwertmannite precipitation. We exploit this condensation characteristics of the Fe10 and develop a scale-up synthesis of schwertmannite. Finally, Al(III), Zn(II), and Cd(II) substituted Fe(III) sulfate clusters are also isolated via the same approach, demonstrating its versatility. This study underlines the importance of identifying molecular clusters as models to study aqueous hydrolysis of metal ions in nature and synthesis. Received: December 12, 2018

A

DOI: 10.1021/acs.inorgchem.8b03446 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

■

H atoms were refined with anisotropic thermal parameters except those in the disordered terminal methyl groups which were refined with isotropic thermal parameters. All H atoms in AlFe, ZnFe, and CdFe were refined in calculated positions in a rigid group model. H atoms at the O atoms in the CH3OH molecules were not found and hence were not taken into consideration. It should be mentioned that in Fe10 there are two O···O contacts of 2.54 and 2.62 Å between OHCH3 bonded to the Fe atoms and the solvent CH3OH molecules. These contacts are relatively short compared to the typical range (2.55−2.96 Å) for such a type of contact.21 Positions of the H atoms involved in these O−H···O H-bonds were found on the residual density maps, and the refinement showed its appropriate thermal parameters. These H atoms were not taken into consideration as the other H atoms on the O atom in the CH3OH molecules. Refinement of AlFe showed that, in the structure, four symmetrically independent positions were pure Fe and one another metal position could not be refined as a pure Fe or Al: the occupation factor for this position was intermediate between that for Fe and Al atoms. Refinement of the structure with the Fe and Al atoms sharing this position gave a ratio of 0.46:0.54, respectively, for Fe and Al atoms. Disordered solvent molecules (CH3OH in ZnFe and AlFe and CH3CH2COOH in Fe10) which were not involved in coordination of the metal atoms of the main clusters were treated by SQUEEZE;22 for the correction of the X-ray data by SQUEEZE, 18, 252, and 180 electrons were the same or close to the required values of 18, 246, and 160 in ZnFe, AlFe, and Fe10, respectively. The solvent molecules in the crystals were positioned between the main clusters, forming H-bonds between them. It should be mentioned that, on the residual density map in AlFe, there was one peak (1.88 e Å−3) that could be related to a possible carbon atom in a terminal methyl group connected to the Al and Fe atoms sharing the same position. Refinement of this position as a possible C atom in a terminal methyl group showed that this position was only partially occupied. Thus, this fragment was refined as (Fe/Al)(H2O)3, but it should be mentioned that a small amount of (CH3OH) instead of a water molecule seems possible at one terminal position. All calculations are performed by the Bruker SHELXL-2014 package.23 Small-Angle X-ray Scattering Characterization. All X-ray scattering data were collected on an Anton Paar SAXSess instrument equipped with line collimated Cu Kα radiation (1.54 Å). A 2D image plate was used for data collection. The sample to image-plate distance was 26.1 cm. Solution samples (including solvent background) were contained in an Anton-Paar liquid cell (with a 0.1 mm thick glass capillary tube) for the 30 min data collection. SAXSQUANT software was used for the data collection and treatment (normalization, primary beam removal, background subtraction, desmearing, and smoothing to remove extra noise created by the desmearing routine). In theory, the 2D image plate contains data in the q = 0.018−2.5 Å−1 range, but generally only data up to q = 1.0 Å−1 is usable because of the imperfect background subtraction introduced by solvent scattering (at high-q). The imperfect background subtraction is a universal problem when studying small clusters. Normalization of scattering curves prior to background subtraction can help with the data interpretation greatly. All analysis and simulations were carried out utilizing the IRENA39 macros within IgorPro 6.3 (Wavemetrics) software. The Gunier− Porod fitting module24 was adopted to analyze the scattering data (a brief description of the Gunier−Porod model is provided in the Supporting Information). TEM Characterization and Sample Preparation. TEM characterization was done on a JEOL 2100 TEM with 200 kV electron beam. A standard copper grid with carbon membrane was used for the characterization. The beam current was controlled between 5 and 20 pA cm−2 for the imaging to avoid sample damage. Solid precipitation from FS4 solution (aged for 1 week) was collected by centrifugation for TEM imaging. Miscellaneous Characterization. TGA measurements were done on an NETZSCH STA449F3 from 30 to 1000 °C in air with a heating rate of 10 °C min−1. ESI-Q-TOF mass spectra were recorded on a Bruker Impact II UHR-TOF mass spectrometry, with

EXPERIMENTAL SECTION

Chemicals. Fe(NO3)3·9H2O (AR, Sinopharm Chemical Reagent Co., Ltd.), Fe2(SO4)3·xH2O (AR, Xilong Scientific Co., Ltd.), Al(NO3)3·9H2O (AR, Aladdin), Zn(NO3)2·6H2O (AR, Aladdin), and Cd(NO3)2·4H2O (AR, Aladdin) were used in this work. The hydration number x in Fe2(SO4)3·xH2O was determined as 6.29 from 800 °C calcination in air. Solution and Single-Crystal Synthesis. The Fe(III) sulfate (FS) solution was synthesized by adding hydrous iron oxide/ hydroxide into Fe2(SO4)3 aqueous solution. To synthesize hydrous iron oxide/hydroxide, excess 1 M NH3·H2O was added to Fe(NO3)3· 9H2O aqueous solution under vigorous stirring. As an example, for the synthesis of FS4 solution, 9.0 mL of 1.0 M NH3·H2O was added to 10 mL of 0.1 M Fe(NO3)3·9H2O aqueous solution under vigorous stirring. The precipitation was collected by centrifugation and washed with Millipore water (18.2 MΩ cm) 5 times. The resulting freshly prepared hydrous iron hydroxide was added to a stirring Fe2(SO4)3 aqueous solution to obtain 10 mL of FS4 solution (with a total of 0.4 M Fe3+). Fe oxide/hydroxide and Fe2(SO4)3 were combined in the iron ratios of 1:12, 1:6, and 1:3 for FS2, FS3, and FS4, respectively. Pure Fe2(SO4)3 aqueous solution was used as FS1 in this study. Hence, the final iron to sulfate ratio was 2:3, 13:18, 7:9, and 8:9 for FS1, FS2, FS3, and FS4, respectively. The final Fe3+ concentration was 0.4 M for all FS solutions. Synthesis of Fe10 Crystals. To synthesize Fe10, FS4 aqueous solution was freeze-dried at −51 °C under 11 Pa. The remaining maroon solid was redissolved in methanol. By diffusing ethyl acetate into the filtered methanol solution, orange-yellow crystals were obtained in ∼41% yield. Single-Crystal Synthesis of Aluminum-Containing Iron Sulfate Clusters (AlFe). The procedure was similar to the synthesis of FS4 solution, except hydrous Al(OH)3 was used. The Al:Fe molar ratio was controlled to be 1:3. After overnight stirring, the clear aqueous solution was freeze-dried at −51 °C under 11 Pa. The remaining maroon solid was redissolved in methanol to obtain a clear solution. By diffusing ethyl acetate into the filtered methanol solution, bright yellow crystals were obtained in 65% yield. Synthesis of Zinc-Containing Iron Sulfate Cluster (ZnFe). The procedure is similar to the synthesis of FS4 solution, except hydrous Zn(OH)2 was used. The Zn:Fe molar ratio was controlled to be 1:2. After overnight stirring, the aqueous solution is then freeze-dried at −51 °C under 11 Pa. The remaining maroon solid was redissolved in methanol to make a clear solution. By diffusing butyl acetate into the filtered methanol solution, bright yellow crystals were obtained in 52% yield. Synthesis of Cadmium-Containing Iron Sulfate Nanoclusters (CdFe). The procedure is similar to the synthesis of FS4 solution, except hydrous Cd(OH)2 was used. The Cd:Fe molar ratio was controlled to be 1:2. After overnight stirring, the clear aqueous solution is then freeze-dried at −51 °C under 11 Pa. The remaining maroon solid is redissolved in methanol to obtain a clear solution. By diffusing ethyl acetate into the filtered methanol solution, bright yellow crystals were obtained in the yield of 45%. Synthesis of Schwertmannite. First, 100 mL of FS4 solution was synthesized following the above-described method. The resulting solution was freeze-dried and redissolved in methanol to obtain a 100 mL solution. Excess ethyl acetate was used to crash out the Fe10containing solid from the solution. The solid was washed by ethyl acetate 3 times before it redissolved into 100 mL of methanol again. This methanol solution was then added to 5 L of DI (deionized) water and left to sit for 1 week. The resulting solid schwertmannite was collected via filtration. Single-Crystal X-ray Diffraction. Diffraction intensities for Fe10, AlFe, ZnFe, and CdFe were collected at 133 K on a Rigaku MM007 Saturn724+ CCD diffractometer using Mo Kα radiation, λ = 0.71073 Å. Absorption corrections were applied by CrystalClear Version 1.3.6.20 Space groups were determined on the basis of systematic absences (Fe10) or on intensity statistics (AlFe, ZnFe, and CdFe). Structures were solved by direct methods and Fourier techniques and were refined on F2 using full matrix least-squares procedures. All nonB

DOI: 10.1021/acs.inorgchem.8b03446 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry ESI-L low-concentration tuning mix (from Agilent Technologies) as the internal standard. Data analyses and simulations of ESI-TOF mass spectra were processed on Bruker Data Analysis software (Version 4.3).

shown in Figures S1−S3, and the structure is stabilized by hydrogen bonding between clusters. It is well-known that the freeze-drying method can isolate proteins and biomolecules from water without destroying their structures. In this process, noncoordinated water ligands are removed in vacuo, and any existing cluster structures are kinetically trapped in the remaining solid. In particular, excess water (nwater:nFe ⩾ 6:1) molecules, including those which bind to Fe3+, are found in the solid by thermogravimetric analysis (Figure S4). After redissolution in methanol, the remaining water ligands are exchanged. This suppresses the hydrolytic polymerization and affords the opportunity to crystallize potential reaction intermediates. Indeed, the aqueous analogue of the Fe3 subunit ([Fe3O(SO4)6(H2O)3]5−) exists both in natural mineral (e.g., carlsonite26 (Figure S5a) and metavoltine27,28 (Figure S5b)) and synthetic structures (Maus’ salt29 (Figure S6)). This provides strong evidence that the relevant Fe(III) clusters do form in aqueous environments, and that the MeOH substitution for the bound waters does not dramatically bias this polynuclear assembly. The Fe10 crystal is dissolved in methanol and characterized by small-angle X-ray scattering (SAXS). The spectrum is fit with a two-level Guinier−Porod (GP) model24 (Figure S7, Table S3; for details on the model, see Supporting Information). Essentially, the GP model interprets the scattering curve by the superposition of multiple levels of particle scattering. The first GP level represents a cylindrical particle, which is 23 Å in height and 7 Å in diameter, consistent with the intact Fe10 cluster; the second level corresponds to a sphere that is 7 Å in diameter, which could be attributed to the Fe3 subunit. Our interpretation is that Fe10 goes through partial dissociation and equilibrates with Fe3 anions. No condensation into larger clusters is observed, as inferred by the flat low-q range. Electrospray ionization mass spectroscopy (ESI-MS, Figure S8 and Table S4) of the Fe10 crystal in methanol confirms the existence of iron trimers, tetrameters, pentamers, and hexamers, generally formulated as [FexO(SO4)y(OCH3, CH3OH, H2O)w]+ (x = 3−6; y = 2−6), and is consistent with Fe10 and its fragments. This confirms our interpretation on the SAXS data. By employing the structural model we obtained from Figure S7, SAXS spectra of all aqueous solutions (FS1−FS4) are collected and interpreted. As we can see, from FS1 to FS4, the solution color changes from orange to maroon with the decreasing sulfate content and increasing pH (Figure 2a), indicating the condensation of iron species.30 From the SAXS spectra (Figure 2b), we notice the increasing deviation in low-q range with increasing Fe3+:SO42− ratio. For solutions with lower Fe3+:SO42− ratio (FS1 and FS2, pH < 1.50), the two-level GP simulation is adequate to reproduce the experimental data (Table S5). When the Fe3+:SO42− ratio increases to 7:9 (FS3, pH = 1.65), the twolevel model still fits for q above 0.1 Å−1 but deviates at lower qvalues. When the Fe3+:SO42− ratio increases to 8:9 (FS4, pH = 2.00), significant deviations in the low-q region indicate further condensation. Therefore, a third GP level is introduced to greatly improve the fitting. As a result, the low-q region for both FS3 (Table S5) and FS4 (Table S5 and Figure 2c) can be fit with scattering from spherical particles (level 3) of size about 37 Å. The existence of these nanoparticles indicated the precipitation of Fe(III) oxide/hydroxide has already initiated at a pH of 2.0. Indeed, after aging for 1 week, precipitations form slowly from the FS4 solution. The HRTEM image reveals the

■

RESULTS AND DISCUSSION We first prepare a series of Fe(III) sulfate (FS) aqueous solutions by adding hydrated iron oxide/hydroxide to an Fe2(SO4)3 aqueous solution. The ratios between Fe3+ and SO42− include 2:3 (FS1), 13:18 (FS2), 7:9 (FS3), and 8:9 (FS4). Hydrated iron oxide/hydroxide both increases the pH and provides an iron source. The addition of metal hydroxide solid instead of a strong base (i.e., NaOH) avoids the steep pH gradients in the system that could lead to either extremely rapid hydrolysis or species that are not representative of those found in natural settings.25 Next, by freeze-drying the FS4 solution and exchanging the remaining water with methanol, we obtain crystals of [Fe10O2(SO4)12(OCH3)2]·14CH3OH (Fe10) (see detailed discussion in the Supporting Information). Fe10 crystallizes in the monoclinic space group P21/c (see Supporting Information). In this structure, all iron is trivalent, determined by the bond valence sum (BVS, Table S1) and in octahedral coordination. The Fe10 cluster (Figure 1a) consists of a

Figure 1. Fe10 cluster structure: (a) side view of the decametric iron sulfate cluster (Fe10), (b) top view of the pentamer unit, and (c) top view of the trimeric subunit Fe3.

centrosymmetric pair of iron sulfate pentamers (Figure 1b), joined by two μ-OCH3 bridges. Each pentamer contains one trimeric [Fe3O(SO4)6(CH3OH)3]5− (Fe3) unit (Figure 1c). In the Fe3 unit, the three Fe(III) octahedra are joined by one μ3O ligand on the axial position and bridged by four sulfates in the equatorial sites of the octahedron, with CH3OH occupying the axial position trans to the μ3-O. Two additional Fe(III) octahedra (site A and site B) are connected to Fe3 via three bridging sulfate groups: for site A, the three remaining vertices are occupied by CH3OH groups; on the other hand, OCH3 groups occupy these positions of site B. Assignment of the CH3OH ligands instead of OCH3 is supported by both BVS calculations (Table S2) and the ligand geometry. Importantly, Fe3+ mostly does not deprotonate the CH3OH ligand. Rather, this “soft” ligation combined with the freeze-drying process leaves the cluster unaltered from their state in the aqueous solutions. The packing of zero-charged Fe10 in the crystal is C

DOI: 10.1021/acs.inorgchem.8b03446 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

at room temperature for weeks. This demonstrates the freezedrying solvent-exchange method is capable of suppressing the precipitation effectively. Meanwhile, our study might also provide insight into formation processes of iron oxide/ hydroxide form aqueous solutions. For example, diluting the FS4 solution by 50× yields precipitation within minutes (Figure S10a). We collect HRTEM images at different time intervals. Two minutes after dilution, 10 nm particles form (Figure S10b), and 200 nm agglomerates form after 20 min (Figure S10c). After 30 min, spheres larger than 300 nm are evident in the images (Figure S10d), leading us to speculate that the 300 nm spheres serve as nuclei for developing the fibrous schwertmannite morphology. A high-magnification view (Figure S10e,f) of the dashed area in Figure S10c reveals small particles attaching to the edge of the 200 nm agglomeration, providing circumstantial evidence that schwertmannite might grow via the aggregation of these particles (Figure S10e). These poorly crystalline/amorphous particles are 37−42 Å in diameter (Figure S10f), consistent with the nanoparticle (Figure 2c, level 3) identified by SAXS. Fast Fourier transformation (FFT) reveals the 2.3 Å lattice spacing of the (302) plane of schwertmannite. In combination with the SAXS results, these observations suggest that the conversion of the Fe10 cluster to ∼37 Å nanoparticles might be an important step toward the formation of schwertmannite. Schwertmannite (with a generalized formula Fe16O16(OH)y(SO4)z·xH2O (16 − y = 2z and ∼2.0 < z < ∼ 3.5)) is known for large surface area and its effectiveness in removing arsenic from the aqueous environment.33,34 Hence, we exploit the condensation characteristics of Fe10 and develop a new synthetic method of schwertmannite. Without optimization, the synthesis can readily scale up to 100 mL (∼5000 mL after dilution) and produces more than 10 g of schwertmannite in a batch (Figure 3c). Finally, the polymerization pathway of the iron sulfate clusters may change when heterometals are introduced into the solution, yielding related clusters that consist of these heterometals. By replacing the hydrous iron oxyhydroxide in the synthesis with hydrous Al(OH)3, Zn(OH)2, or Cd(OH)2, we isolate three additional cluster structures via X-ray crystallography (Figure 4). These structures contain Fe3 with Al3+ (AlFe, Figure S11), Zn2+ (ZnFe, Figure S12), and Cd2+ (CdFe, Figure S13) located at site A/B, respectively. With trivalent Al3+, the Fe10 analogue forms in solution, containing Al3+ at a terminal position. On the other hand, with divalent cations, the condensation deviates from the “Fe10 route”, and the two “pentamers” are connected via metal sulfate linkage instead of hydroxo/methoxo bridges. Notably, corner-linking

Figure 2. (a) Appearance of the FS1, FS2, FS3, and FS4 aqueous solution; (b) the SAXS spectra of the above four solutions with twolevel GP fitting; and (c) the three-level GP fitting of the FS4 solution.

formation of fibrous particles larger than 500 nm (Figure 3a). The electron-diffraction pattern (inset of Figure 3a) of the precipitation contains two diffraction rings, corresponding to the (004) and (212) lattice planes of schwertmannite (JCPDS 47-1775), a common iron oxyhydroxide sulfate mineral. The fibrous morphology (Figure 3b) is consistent with the reported schwertmannite.31,32 We collect the powder X-ray diffraction pattern of the aged solid (Figure S9). The results indicate the formation of poorly crystalline schwertmannite (JCPDS 471775). The sample exhibits broad and weak diffraction peaks (e.g., from the (310), (201), and (004) lattice plane) on top of the background from the glass sample holder. It is also likely that a significant portion of the material is amorphous, leading to the broad hump in the spectrum. On the contrary, the redissolved methanol FS4 solution after freeze-drying is stable

Figure 3. (a) TEM image of the solid precipitation formed in the aged FS4 solution (inset: the electron-diffraction pattern), (b) HRTEM image of the solid precipitation, and (c) a scale-up synthesis of schwertmannite (in a 5000 mL beaker). D

DOI: 10.1021/acs.inorgchem.8b03446 Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

■

Article

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. (W.W.) *E-mail: [email protected]. (Y.Z.) *E-mail: [email protected]. (M.N.) ORCID

Wei Wang: 0000-0003-3178-236X May Nyman: 0000-0002-1787-0518 Author Contributions

The solution and crystal synthesis, SAXS, and HRTEM were performed by W.W. and T.H. M.A., L.N.Z., and M.N. performed the mass spectral and single-crystal analyses. The manuscript was written through contributions of all authors.

Figure 4. Various Fe3-containing structures formed in ferric sulfate solution with different metal ions, including Al3+, Zn2+, and Cd2+. All structures are determined by X-ray crystallography.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge the financial support from the National Natural Science Foundation of China (No. 51602310; W.W., T.H., and Y.Z.) and the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Material Sciences and Engineering (Award DE SC0010802; M.A., L.N.Z., and M.N.). We thank Professor William H. Casey (University of California, Davis) for valuable comments, Qing Ren (FJIRSM, CAS) for single-crystal X-ray diffraction data collection, Qi Peng (FJIRSM, CAS) for mass spectroscopy measurement, and Binbin Xu (Xiamen University) for TEM sample preparation and imaging.

of MO6-octahedra dominates in all these heterometallic clusters, which means they remain labile and subject to rearrangement as the nanoclusters evolve to iron oxide nanoparticles. The successful isolation of these heterometal clusters also demonstrates the versatility of the freeze-drying solvent-exchange approach.

■

CONCLUSION In conclusion, we develop a freeze-drying solvent-exchange method to isolate metastable aqueous metal-hydrolysis intermediates, Fe(III) clusters in particular. This method yields a new Fe(III) cluster, [Fe10O2(SO4)12(OCH3)2]· 14CH3OH (Fe10), with three derivatives that contain Al(III), Zn(II), or Cd(II) substituted on the Fe sites. We follow the aqueous hydrolysis of iron sulfate solutions from the cluster intermediates to nanoparticles, prior to precipitating schwertmannite. These results show that Fe10 clusters (and its Fe3 fragment) are instrumental to interpret the reaction pathway and species involved. It also underlines the importance of deriving synthetic strategies to isolate and characterize cluster intermediates without influencing the cluster structures. Therefore, the freeze-drying solvent-exchange method provides a potential platform to identify molecular condensation intermediates or to synthesize solid oxides of the fast exchanging ions. These isolated cluster forms are potentially relevant intermediates in environments that are rich in Fe3+ and SO42−, including acid mine drainage35,36 and Martian soils.37,38

■

■

REFERENCES

(1) Banfield, J. F.; Welch, S. A.; Zhang, H. Z.; Ebert, T. T.; Penn, R. L. Aggregation-based crystal growth and microstructure development in natural iron oxyhydroxide biomineralization products. Science 2000, 289 (5480), 751−754. (2) Cornell, R. M.; Schwertmann, U. The Iron Oxides: Structure, Properties, Reactions, Occurrences and Uses, 2nd, Completely Revised and Extended, ed.; Wiley-VCH, 2006. (3) Jolivet, J.-P.; Chaneac, C.; Tronc, E. Iron oxide chemistry. From molecular clusters to extended solid networks. Chem. Commun. 2004, No. 5, 477−483. (4) Zhu, M.; Legg, B.; Zhang, H.; Gilbert, B.; Ren, Y.; Banfield, J. F.; Waychunas, G. A. Early Stage Formation of Iron Oxyhydroxides during Neutralization of Simulated Acid Mine Drainage Solutions. Environ. Sci. Technol. 2012, 46 (15), 8140−8147. (5) Baumgartner, J.; Dey, A.; Bomans, P. H. H.; Le Coadou, C.; Fratzl, P.; Sommerdijk, N.; Faivre, D. Nucleation and growth of magnetite from solution. Nat. Mater. 2013, 12 (4), 310−314. (6) Sun, S.; Gebauer, D.; Cölfen, H. Alignment of Amorphous Iron Oxide Clusters: A Non-Classical Mechanism for Magnetite Formation. Angew. Chem., Int. Ed. 2017, 56 (14), 4042−4046. (7) Baumgartner, J.; Faivre, D. Iron solubility, colloids and their impact on iron (oxyhydr)oxide formation from solution. Earth-Sci. Rev. 2015, 150, 520−530. (8) Zhu, M.; Frandsen, C.; Wallace, A. F.; Legg, B.; Khalid, S.; Zhang, H.; Mørup, S.; Banfield, J. F.; Waychunas, G. A. Precipitation pathways for ferrihydrite formation in acidic solutions. Geochim. Cosmochim. Acta 2016, 172, 247−264. (9) De Yoreo, J. J.; Gilbert, P. U. P. A.; Sommerdijk, N. A. J. M.; Penn, R. L.; Whitelam, S.; Joester, D.; Zhang, H.; Rimer, J. D.; Navrotsky, A.; Banfield, J. F.; Wallace, A. F.; Michel, F. M.; Meldrum, F. C.; Cölfen, H.; Dove, P. M. Crystallization by particle attachment in synthetic, biogenic, and geologic environments. Science 2015, 349 (6247), DOI: 10.1126/science.aaa6760. (10) Scheck, J.; Wu, B.; Drechsler, M.; Rosenberg, R.; Van Driessche, A. E. S.; Stawski, T. M.; Gebauer, D. The Molecular

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.8b03446. Crystal information, introduction of Guinier−Porod (GP) model, supporting figures, and supporting tables (PDF) Accession Codes

CCDC 1885741−1885744 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing [email protected], or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033. E

DOI: 10.1021/acs.inorgchem.8b03446 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Mechanism of Iron(III) Oxide Nucleation. J. Phys. Chem. Lett. 2016, 7 (16), 3123−3130. (11) Dunand, F. A.; Helm, L.; Merbach, A. E. Solvent exchange on metal ions. Adv. Inorg. Chem. 2003, 54, 1−69. (12) Casey, W. H.; Swaddle, T. W. Why small? The use of small inorganic clusters to understand mineral surface and dissolution reactions in geochemistry. Rev. Geophys. 2003, 41 (2), 1008. (13) Casey, W. H.; Rustad, J. R.; Spiccia, L. Minerals as MoleculesUse of Aqueous Oxide and Hydroxide Clusters to Understand Geochemical Reactions. Chem. - Eur. J. 2009, 15 (18), 4496−4515. (14) Jolivet, J.-P.; Tronc, E.; Chanéac, C. Iron oxides: From molecular clusters to solid. A nice example of chemical versatility. C. R. Geosci. 2006, 338 (6−7), 488−497. (15) Faponle, A. S.; Quesne, M. G.; Sastri, C. V.; Banse, F.; de Visser, S. P. Differences and Comparisons of the Properties and Reactivities of Iron(III)-hydroperoxo Complexes with Saturated Coordination Sphere. Chem. - Eur. J. 2015, 21 (3), 1221−1236. (16) Hutchins, D. A.; Witter, A. E.; Butler, A.; Luther, G. W., III Competition among marine phytoplankton for different chelated iron species. Nature 1999, 400, 858. (17) Bino, A.; Ardon, M.; Lee, D.; Spingler, B.; Lippard, S. J. Synthesis and structure of Fe13O4F24(OMe)(12) (5-): The first open-shelf Keggin ion. J. Am. Chem. Soc. 2002, 124 (17), 4578−4579. (18) Caneschi, A.; Cornia, A.; Fabretti, A. C.; Gatteschi, D.; Malavasi, W. Polyiron(III)-Alkoxo Clusters: a Novel Trinuclear Complex and Its Relevance to the Extended Lattices of Iron Oxides and Hydroxides. Inorg. Chem. 1995, 34 (18), 4660−4668. (19) Caneschi, A.; Cornia, A.; Fabretti, A. C.; Gatteschi, D. Structure and magnetic properties of a decanuclear oxoiron(III) cluster: A further step to understanding iron aggregation processes. Angew. Chem., Int. Ed. Engl. 1996, 34 (23−24), 2716−2718. (20) CrystalClear Version 1.3.6; Rigaku/MSC: The Woodlands, TX, 2004. (21) Jeffrey, G. A.; Saenger, W. Hydrogen Bonding in Biological Structures; Springer: Berlin, 1991. (22) van der Sluis, P.; Spek, A. L. BYPASS: an effective method for the refinement of crystal structures containing disordered solvent regions. Acta Crystallogr., Sect. A: Found. Crystallogr. 1990, 46 (3), 194−201. (23) Sheldrick, G. A short history of SHELX. Acta Crystallogr., Sect. A: Found. Crystallogr. 2008, 64 (1), 112−122. (24) Hammouda, B. A new Guinier-Porod model. J. Appl. Crystallogr. 2010, 43, 716−719. (25) Wang, W.; Wentz, K. M.; Hayes, S. E.; Johnson, D. W.; Keszler, D. A. Synthesis of the Hydroxide Cluster [Al-13(mu(3)-OH)(6)(muOH)(18)(H2O)(24)](15+) from an Aqueous Solution. Inorg. Chem. 2011, 50 (11), 4683−4685. (26) Kampf, A. R.; Richards, R. P.; Nash, B. P.; Murowchick, J. B.; Rakovan, J. F. Carlsonite, (NH4)(5)Fe33+O(SO4)(6)center dot 7H(2)O, and huizingite-(Al), (NH4)(9)Al-3(SO4)(8)(OH)(2)center dot 4H(2)O, two new minerals from a natural fire in an oilbearing shale near Milan, Ohio. Am. Mineral. 2016, 101 (9−10), 2095−2107. (27) Giacovazzo, G.; Scordari, F.; Todisco, A.; Menchetti, S. Crystalstructure model for metavoltine from Sierra Gorda. TMPM, Tschermaks Mineral. Petrogr. Mitt. 1976, 23 (3), 155−166. (28) Scordari, F.; Vurro, F.; Menchetti, S. The metavoltine problem: Relationships between metavoltine and Maus’ salt. TMPM, Tschermaks Mineral. Petrogr. Mitt. 1975, 22 (1), 88−97. (29) Giacovazzo, C.; Scordari, F.; Menchetti, S. Hydrous potassium and ferric iron sulphate (Maus’s salt). Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem. 1975, 31 (8), 2171−2173. (30) Schwertmann, U.; Cornell, R. M. Iron Oxides in the Laboratory: Preparation and Characterization; Wiley-VCH: Weinheim, New York, 2000. (31) Bigham, J. M.; Carlson, L.; Murad, E. Schwertmannite, A New Iron Oxyhydroxysulphate from Pyhasalmi, Finland, and Other Localities. Mineral. Mag. 1994, 58 (393), 641−648.

(32) French, R. A.; Caraballo, M. A.; Kim, B.; Rimstidt, J. D.; Murayama, M.; Hochella, M. F., Jr. The enigmatic iron oxyhydroxysulfate nanomineral schwertmannite: Morphology, structure, and composition. Am. Mineral. 2012, 97 (8−9), 1469−1482. (33) Waychunas, G. A.; Kim, C. S.; Banfield, J. F. Nanoparticulate iron oxide minerals in soils and sediments: unique properties and contaminant scavenging mechanisms. J. Nanopart. Res. 2005, 7 (4−5), 409−433. (34) Carlson, L.; Bigham, J. M.; Schwertmann, U.; Kyek, A.; Wagner, F. Scavenging of as from acid mine drainage by schwertmannite and ferrihydrite: A comparison with synthetic analogues. Environ. Sci. Technol. 2002, 36 (8), 1712−1719. (35) Akcil, A.; Koldas, S. Acid Mine Drainage (AMD): causes, treatment and case studies. J. Cleaner Prod. 2006, 14 (12−13), 1139− 1145. (36) Johnson, D. B.; Hallberg, K. B. Acid mine drainage remediation options: a review. Sci. Total Environ. 2005, 338 (1−2), 3−14. (37) Bish, D. L.; Blake, D. F.; Vaniman, D. T.; Chipera, S. J.; Morris, R. V.; Ming, D. W.; Treiman, A. H.; Sarrazin, P.; Morrison, S. M.; Downs, R. T.; Achilles, C. N.; Yen, A. S.; Bristow, T. F.; Crisp, J. A.; Morookian, J. M.; Farmer, J. D.; Rampe, E. B.; Stolper, E. M.; Spanovich, N.; MSL Science Team. X-ray Diffraction Results from Mars Science Laboratory: Mineralogy of Rocknest at Gale Crater. Science 2013, 341 (6153), DOI: 10.1126/science.1238932. (38) Blake, D. F.; Morris, R. V.; Kocurek, G.; Morrison, S. M.; Downs, R. T.; Bish, D.; Ming, D. W.; Edgett, K. S.; Rubin, D.; Goetz, W.; Madsen, M. B.; Sullivan, R.; Gellert, R.; Campbell, I.; Treiman, A. H.; McLennan, S. M.; Yen, A. S.; Grotzinger, J.; Vaniman, D. T.; Chipera, S. J.; Achilles, C. N.; Rampe, E. B.; Sumner, D.; Meslin, P. Y.; Maurice, S.; Forni, O.; Gasnault, O.; Fisk, M.; Schmidt, M.; Mahaffy, P.; Leshin, L. A.; Glavin, D.; Steele, A.; Freissinet, C.; NavarroGonzalez, R.; Yingst, R. A.; Kah, L. C.; Bridges, N.; Lewis, K. W.; Bristow, T. F.; Farmer, J. D.; Crisp, J. A.; Stolper, E. M.; Marais, D. J. D.; Sarrazin, P.; MSL Science Team. Curiosity at Gale Crater, Mars: Characterization and Analysis of the Rocknest Sand Shadow. Science 2013, 341 (6153), DOI: 10.1126/science.1239505. (39) Ilavsky, J.; Jemian, P. R. Irena: tool suite for modeling and analysis of small-angle scattering. J. Appl. Crystallogr. 2009, 42, 347− 353.

F

DOI: 10.1021/acs.inorgchem.8b03446 Inorg. Chem. XXXX, XXX, XXX−XXX