Alkali-driven disassembly and reassembly of molecular niobium oxide

Subscriber access provided by NEW MEXICO STATE UNIV

Article

Alkali-driven disassembly and reassembly of molecular niobium oxide in water Dylan Sures, Mireia Segado, Carles Bo, and May Nyman J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.8b05015 • Publication Date (Web): 05 Jul 2018 Downloaded from http://pubs.acs.org on July 5, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

Alkali-driven disassembly and reassembly of molecular niobium oxide in water Dylan Sures,†,‡ Mireia Segado,¶,§ Carles Bo,¶ and May Nyman∗,† †Department of Chemistry, Oregon State University, Corvallis, OR 97331-4003, USA ‡Current Affiliation, Department of Chemistry, University of California, Davis, CA 95616-5270, USA ¶Institut Catalá dInvestigació Qu´ımica (ICIQ), The Barcelona Institute of Science and Technology, Av. Pa´ısos Catalans, 17. 43007 Tarragona, Spain §Departament de Qu´ımica F´ısica i Inorg´ anica, Universitat Rovira i Virgili, Tarragona (Spain) E-mail: [email protected]

Abstract Counterions are deemed ‘spectators’ in aqueous solutions of cationic or anionic molecular metal-oxo clusters. While pH and concentration drive aqueous metal speciation as a first approximation, the important effect of counterions is usually overlooked and never considered in standard Pourbaix databases. Alkali counterions for polyoxometalate (POM) clusters control solubility with distinct periodic trends, but evidence for alkali control over speciation is ambiguous. Here we show that a simple Nb-POM, [Nb10 O28 ]6 – ({Nb10 }), converts to oligomers of (Hx Nb24 O72 )(24 – x) – ({Nb24 }) upon adding only alkali chloride salts, even in buffered neutral solutions. Raman and Xray scattering reveal that the rate of {Nb10 } to {Nb24 } conversion increases with alkali cation radius and cation concentration. Cation-bridged oligomers of {Nb24 }y

1

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(y=2,4) are defined by comparing experimental to computed small-angle x-ray scattering spectra. Computational studies and mass spectrometry indicate the alkalis open the compact {Nb10 } cluster in conjunction with protonation of a heptamer {Nb7 } intermediate, where alkali-{Nb10 } association at key locations on the cluster initiates the reaction. Computation also explains the alkali periodic trend for {Nb10 } to {Nb24 } conversion; larger alkalis more effectively destabilize {Nb10 }. This periodic trend asserts the hypothesis that Nb-cluster speciation near neutral pH is driven by the alkali cations in the absence of added base or acid. The extremely high solubility of these 3.5 nm polyoxoanion assemblies – 2M Nb at near neutral pH – is both surprising and exploitable for aqueous synthesis of niobate thin films or nanomaterials used in energy and microelectronics applications.

Introduction The behavior of metal oxides and molecular metal-oxo clusters at an aqueous interface or in solution is dictated by numerous complex and interrelated phenomena. 1–6 Metal oxides and metal-oxo clusters are relevant to industrial, energy-related, and natural processes – specifically but not limited to corrosion, 7,8 catalysis (especially water oxidation), 9–12 mineral weathering and precipitation coupled with contaminant transport, 13–16 and synthesis. 17,18 The most important solution characteristics affecting interfacial metal oxide processes and metal-oxo cluster speciation include pH, concentration, temperature, electrochemical potential (Eh), and electrolytes. The effect of some of these characteristics are well-understood and predictable across the periodic table (i.e., concentration and temperature), while others are confounding, interdependent, and variable, depending on the metal cation(s). While aqueous speciation is important for all metals, polyoxometalates (POMs) have the best-defined and controllable metal-oxo cluster chemistry. POMs are Group 5/6 d0 polynuclear clusters that are surface-stabilized by multiplybonded oxo-ligands. POMs of V, Mo, and W assemble via a bottom-up route. Acidifying

2


Page 2 of 31

Page 3 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


2/3 –

aqueous mononuclear oxoanions MO4

(M = Mo,W,V) drives hydrolysis and condensation;

i.e., 2 [MO3 (OH)] – −−→ [M2 O7 ]2 – + H2 O. 19–27 On the other hand, formation mechanisms of Nb and Ta POMs is not understood, and is best described as a top-down approach. Nb or Ta monomer oxoanions are not isolatable, due to the larger size and lower charge-densities of Ta/Nb5+ , compared to V5+ and W/Mo6+ . 28 Instead, dissolution of niobium or tantalum oxide in highly alkaline solution yields only the Lindqvist ion ([M6 O19 ]8 – ; M = Nb/Ta; {Nb6 }, figure 1a), the most common Nb/Ta POM geometry. 29 The uncertainty of Nb(Ta) POM formation mechanisms is reflected in incomplete speciation diagrams. 30–32 Because Nb/Ta (aq) solutions are difficult to retain except at high pH (>12), formation of other Nb POM geometries cannot be reproducibly achieved by pH change. Successful alternative, yet poorly controlled, strategies have included hydrothermal synthesis 33,34 and use of base-stable, large countercations such as Cu-amine complexes. 35,36

Figure 1: Polyhedral representations of (a) [Nb6 O19 ]8 – ({Nb6 }), (b) [Nb7 O22 ]9 – ({Nb7 } fragment), and (c) [Nb10 O28 ]6 – ({Nb10 }). Green octahedra are Nb-centered and red spheres are oxygen atoms. Decaniobate ([Nb10 O28 ]6 – ; {Nb10 }, figure 1c) is a simple Nb-POM related to {Nb6 } that forms reproducibly in solution, albeit in alcohol rather than water. 37–39 Aqueous {Nb10 } is stable and soluble at neutral pH, but converts to {Nb6 } with pH increase, and precipitates Nb2 O5 via acid addition. 40 While {Nb6 } is a base catalyst, 41,42 Nb2 O5 is an acid catalyst. 43 {Nb10 } plus its derivative products described here represent the neutral intermediate. The

3



conversion of {Nb10 } to {Nb6 } proceeds via a ‘heptaniobate’ intermediate, {Nb7 } (figure 1b). 38,44 {Nb7 } features three reactive terminal oxo-ligands, leading to condensation into larger clusters such as the {Nb24 } unit, first described by Bontchev and Nyman. 35 These clusters of linked {Nb7 } are synthesized in the presence of Cu amine coordination compounds from K+ –{Nb6 } starting material and form aggregates up to almost 100 Nb-polyhedra. 39,45,46 No alkali cation salt nor reproducible aqueous synthesis of {Nb10 } has ever been reported – only the tetramethylammonium (TMA; [(CH3 )4 N+ ]) salt has been isolated. 37 Additionally, prior-reported base-promoted conversion of {Nb10 } to {Nb6 } was performed in the absence of alkalis, and parallel computational studies did not account for counterions. 38,47 While {Nb7 }-based clusters could offer clues to the potentially rich Nb(aq) speciation between neutral and highly alkaline pH, they have neither been studied in systematically controlled conditions nor without copper amines. Here we show that the conversion of {Nb10 } to {Nb7 } can be controlled by adding alkalis alone, even in solutions buffered near neutral pH. Moreover, both the rate of {Nb10 } to {Nb7 } conversion and the aggregate sizes that assemble follow a periodic trend of Cs>Rb>K>Na>Li. Increasing alkali concentration also increases the reaction rate. With these two variables, we show the cluster growth and aggregation reactions are primarily alkali-promoted. We use small-angle X-ray scattering (SAXS), X-ray total scattering, Raman spectroscopy, and electrospray ionization mass spectrometry (ESI MS) to document these alkali-promoted transitions. Computational studies both provide models for the X-ray scattering data of reaction solutions and explain how the alkalis can promote this species alteration, followed by assembly of large aggregates. Evaporated solutions containing large aggregates (up to ∼5 nm) remain stable at ∼2M Nb concentration, near neutral pH. Moreover, the dry solids obtained by complete solvent evaporation can be redissolved in neat water. Almost all metal oxides are amphoteric in their solubility and aqueous dissolution at neutral pH typically requires capping ligands. Therefore this metal oxide solution behavior is both unusual and can be exploited in aqueous

4


Page 4 of 31

Page 5 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


synthesis of thin film coatings and nanostructured materials from solutions of simple ions. To our knowledge, this is the first documentation of counterions being the primary driving force for metal-oxo cluster assembly and rearrangement in water. The results of this study imply 1) Alkalis in general and even specific alkalis can have profound and differentiating effect on polyanion speciation that should be documented in standard databases, and 2) Counterions and other electrolytes likewise should be considered and therefore exploited to optimize processes at the metal oxide-water interface.

Results & Discussion

Figure 2: Raman spectra showing the decomposition of 20 mM {Nb10 } in the presence 120 mM ACl (A = Li, Na, K, Rb, Cs) after (top left) three hours, (top right) one day, (bottom left) three days, and (bottom right) 1 week. The peak at 755 cm−1 corresponds to tetramethylammonium and serves as an internal standard.

Alkali-promoted {Nb10 } alteration. We monitored aqueous 20 mM aqueous {Nb10 } solutions with added 120 mM ACl (A = Li, Na, K, Rb, Cs) by Raman spectroscopy 28 ◦C over the course of one week (figure 2). Cs+ alters {Nb10 } much more rapidly than the other 5



alkalis, exemplified by the diminished peak intensity of the {Nb10 } symmetric terminaloxo stretch at 936 cm−1 40,48 after only three hours. Within one day, this peak has entirely disappeared in the presence of Cs+ , indicating that all of the {Nb10 } has been altered. The other solutions follow suit, exactly trending up the alkali series. After one week, all alkali cations have fully decomposed {Nb10 }. On the other hand, aqueous TMA{Nb10 } remains almost completely unchanged. We note briefly that ammonium also induces rapid dissociation of {Nb10 }. However, ammonium can serve as a weak acid, so we did not pursue its {Nb10 }-promoted dissociation further in this current study.

Figure 3: pH measurements over time of 20 mM {Nb10 } with 120 mM ACl (A = Li, K, Cs) in neat water (solid diamonds) and in 1M HEPES/TMAOH buffer (pH = 7; open diamonds). In neat water, solutions of {Nb10 } with alkali cations undergo an initial rise in pH from 7 to ∼8.5, which then stabilizes on a time frame commensurate with the Raman studies (figure 3). Specifically, Cs+ promotes the fastest rise in pH, followed by K+ , and Li+ drives the slowest change. However, each solution equilibrates to very nearly the same pH value over time (8.5-9) as the respective reactions go to completion. As expected, the pH of a solution containing only TMA{Nb10 } maintains a constant pH around 7, indicating that {Nb10 } is not prone to protonation in the presence of only TMA+ . Thus, alkali cations are indirectly responsible for inducing the increase in pH. When placed into a 1M HEPES buffer solution (buffered to pH = 7 with TMAOH), the pH change is largely suppressed and does not exceed 7.6 at any time for any alkali cation. We have also monitored the fingerprint {Nb10 } Raman peak at 937 cm−1 (figure 4) for the 6


Page 6 of 31

Page 7 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4: {Nb10 } Raman peak intensity (937 cm−1 ) over time for the decomposition of 20 mM {Nb10 } with 120 mM ACl (A = Li, K, Cs) in both neat water (solid diamonds) and 1 molar HEPES/TMAOH buffer (pH = 7; open diamonds). alkali-HEPES pH-7 buffered solutions. The decomposition of {Nb10 } is significantly slowed in buffered solutions for each alkali cation, confirming that pH has a role in speciation. There is some degree of Raman peak shifting within these series of experiments that is not readily explained. We attribute it to more extensive protonation of the clusters in the lower pH (buffered) solutions. Consistent with this, we observe by SAXS that the resulting species in the buffered solutions are actually larger (yet still discrete) than those of the unbuffered solutions (figure S12). This is noted by the shift in the Guinier region to lower-q (Li, K and Cs solutions) and increasing scattering intensity at qmin (K and Cs). SAXS characterization is discussed in more detail below. We suggest that increased protonation leads to increased size because hydrolysis and condensation is one pathway to linking clusters together, as mentioned in the introduction; i.e., Nb−OH + Nb−OH −−→ Nb−O−Nb + H2 O. Comparing the shapes of the curves of {Nb10 } peak diminution (figure 4) over time reveals insight into the reaction pathway. As exemplified by the shape of the curve of {Nb10 } + LiCl in neat water (the slowest reacting solution), the reaction accelerates when the pH is allowed to increase unbuffered, exemplified by the steepening of the pH (figure 3) and Raman intensity (figure 4) curves after two days. This is further confirmed by the comparatively linear curves for Li+ and K+ in HEPES – since the hydroxide ion concentration in solution is more constant, the reaction does not accelerate as it does in neat water. The alteration of Cs+ –{Nb10 } in HEPES is also significantly slower compared to CsCl solution only, though 7



the apparent difference is less pronounced due to the relatively rapid rate of the reaction driven by Cs+ . It is important to note that despite being significantly slowed, the reaction still occurs when the solution is buffered. The Raman spectra of fully-aged (meaning the peaks in the Raman spectra have stopped changing) HEPES solutions are shown in the Supporting Information (figures S3-S5).

Figure 5: {Nb10 } Raman peak intensity (937 cm−1 ) over time for the decomposition of 20 mM {Nb10 } with 20 mM, 60 mM, and 120 mM ACl (A = Li, K, Cs) in neat water. In addition to the noted periodic trend of alkali promoted {Nb10 } alteration (figure 4) and the effect of the buffer, the reaction rate also increases with alkali concentration. With decreased equivalents of alkali cation per {Nb10 } from six to three, the diminution of the decaniobate Raman peak is slowed significantly – by a factor of approximately four for each alkali cation (figure 5). Decreasing the alkali equivalents to one per {Nb10 } slowed the reaction rate by a factor of approximately nine. This further illustrates the role of alkali cations in dissociating {Nb10 } in water without additional reactants, and indicates an approximate second-order reaction with respect to alkali cation concentration. The Raman peak that appears at 904 cm−1 simultaneously with the disappearance of the {Nb10 } peak is very close to those from reported spectra that were ascribed to symmetric terminal-oxo stretches of {Nb6 }. 40 In the absence of other measurements, we might conclude that alkali cations decompose {Nb10 } into hexaniobate. However, the solutions’ pH of 8.5-9 is below hexaniobates window of aqueous stability, 28 so this conclusion is unsatisfactory. SAXS analyses and ESI MS provided insight into the process and stable species in mildly 8


Page 8 of 31

Page 9 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


alkaline pH.

Figure 6: SAXS spectra of 20 mM {Nb10 } in neat water and with 120 mM ACl (A = Li, Na, K, Rb, Cs). Solutions were aged for two weeks to assure full transition from {Nb10 } to {Nb24 } aggregates. After the alkali-promoted {Nb10 } alteration was complete as evidenced by Raman, the solutions were measured by SAXS to ascertain the size and, eventually, identify the species observed by Raman (figure 6). The TMA{Nb10 } solution retained small, approximately spherical particles (Rg = 3.4 ˚ A) indicative of stable {Nb10 }. However, the alkali cationcontaining solutions exhibit Guinier regions that are shifted to lower q and an order of magnitude increase in I0 values, indicating the presence of much larger species. These scattering curves also exhibit 0-slope at low q, suggesting low polydispersity. The similar Raman spectra and the unique shape of these scattering curves with a second plateau beginning at q > 0.3 ˚ A

−1

(figure 6; S7-S11) suggest each assembly is likely made of similar subunits, but

with differing oligomerization of the subunits. The radius of gyration (Rg ) and approximate diameter (maximum linear extent) of the cluster species for each alkali-promoted alteration were determined from a pair distance distribution function analysis (PDDF; probability histogram of scattering vectors through the cluster) and are summarized in Table 1. Following the same trend of the rate of alkali-promoted alteration of {Nb10 }, the aggregates scale in size as Cs>Rb>K>Na>Li. The {Nb7 } unit ([Nb7 O22 ]9 – , figure 1b), has been reported as 9



the building block of a multitude of larger polyniobates, 45 and even suggested as an intermediate between {Nb10 } and {Nb6 } conversion in base. 38 Because the Raman stretching from an {Nb7 } unit is very likely to be similar to that of a Lindqvist ion, these clusters provide a good starting point to predict the identity of the alkali-restructured niobium-oxo clusters. We simulated the SAXS curves for {Nb24 O72 }, {Nb32 O96 }, and {K12 Nb96 O288 } 35,45 and compared them to the experimental data of the TMA{Nb10 } and aged {Nb10 } + KCl solutions (figure S16). The K-{Nb10 } solution was chosen on account of its median size species out of the alkali-promoted alteration products, the lack of structure factor in its SAXS curve, and its prominence in the synthesis of the other reported clusters. The relative position of the Guinier elbow of the experimental aged {Nb10 } + KCl scattering curve is between that of {Nb32 } and {Nb96 }, bracketing the approximately nuclearity of these species between 32 and 96 Nb-polyhedra. The shape of the experimental curves’ plateau and second “dip” in the ˚−1 range most closely matches those of {Nb24 } and {Nb32 }. This feature has been 0.4-0.9 A noted in clusters with a less electron-dense ‘core’, 49 also consistent with these crown-like assemblies.

Figure 7: Optimized simulated structures for {Nb24 }2 (dimer; top) and {Nb24 }2 (tetramer; bottom). Green polyhedra are [NbO6 ], purple spheres are alkalis; Li+ for the dimer (top) and Cs+ for the tetramer (bottom).

10


Page 10 of 31

Page 11 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 8: Experimental spectra of 20 mM TMA{Nb10 } in neat water and aged with 120 mM LiCl, along with a simulated scattering curve of an {Nb24 }2 dimer bridged by Li+ (see figure 7).

Table 1: Radii of gyration (Rg ) and maximum linear extent of experimental spectra of {Nb10 } + ACl (A = Li, Na, K, Rb, Cs) and simulated spectra of cation-bridged oligomers of {Nb24 }. The comparison of the simulated and experimental scattering data are shown in figure 8 for Li, and figures S17-S18 for all other alkali cations. Alkali Simulated Cation Oligomer Li+ dimer Li+ (solid redissolved) Na+ equilibrium1 + K equilibrium2 + Rb tetramer + Cs tetramer Cs+ (solid redissolved) 1 2

Experimental Rg (˚ A) 8.8 9.4 9.7 9.9 11.8 12.9 17.3

Simulated Rg (˚ A) 8.8 N/A 10.6 12.1 13.6 13.4 N/A

Experimental Simulated Max Extent (˚ A) Max Extent (˚ A) 23.0 23.0 24.1 N/A 26.9 28.6 28.7 31.1 31.3 35.4 34.4 35.5 46.5 N/A

25% tetramer, 75% dimer 50% tetramer, 50% dimer

With this information as a starting point, we simulated and geometry-optimized a dimer ({Nb24 }2 ) and a tetramer ({Nb24 }2 ) that are linked by alkalis (figure 7). The Li-{Nb10 } solution has a scattering curve that matches very closely the simulated scattering for {Nb24 }2 . (figure 8). The Rg values of the experimental and simulated curves are identical (both −1 8.8 ˚ A, Table 1). Furthermore, the plateau and second feature (q = 0.3 − 0.8 ˚ A ) align

almost perfectly between the experimental and simulated curves, suggesting {Nb24 }2 is a 11



likely species that dominates this solution. The mismatch between the experimental and ˚−1 is owed to a structure factor in the experimental solution. simulated curves at q < 0.1 A The larger alkalis, Rb and Cs, produce scattering curves that are more similar to that of a hypothetical tetramer, {Nb24 }4 . Visual comparison of experimental and simulated scattering curves is depicted in figures S17-S18 and the metrical dimensions as determined by the PDDF analyses are summarized in Table 1. On the other hand, the Na and K altered {Nb10 } solutions better match an average of the simulated scattering curves for the dimer and tetramer of {Nb24 } (figure S16 and Table 1). With the exception of the Li-{Nb24 }2 dimer, the matches between simulation and experiment are not ideal, but they match well in the Guinier region, indicated visually and by the calculated radius of gyration (Rg , Table 1). We attribute the differences in geometry between simulation and experiment to several factors including 1) The contribution of the heavier alkalis is more significant, and more greatly affects the scattering data; and 2) Change in the orientation of the {Nb24 } units within the dimers and tetramers relative to each other also changes the scattering curves. Future molecular dynamics simulations may yield more accurate depiction of these data. Many repeated attempts and methods to crystallize these {Nb24 } oligomers were unsuccessful. Complete evaporation and redissolution of the solid followed by SAXS analysis yielded even larger aggregates for the Cs-altered {Nb10 } solution, while the Li-altered {Nb10 } solution remained essentially unchanged (figure S19 and Table 1). Since these solids can be redissolved, this suggests that the discrete {Nb24 } units remain upon solution condensation, without significant further hydrolysis. This also gives indication of the robust (stable) nature of the {Nb24 } unit. On the other hand, the tendency of these solutions to form gels at very high concentration (>2M Nb) has led to studies of thin film deposition of niobia (NH4+ promoted alteration), various alkali niobate perovskites, and intercalated layered materials used in microelectronics and energy applications. The assertion that each alkali cation promotes and assembles {Nb24 } oligomers is also corroborated by the close match of the experimental Raman spectra at the completion of

12


Page 12 of 31

Page 13 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


each aging process. Furthermore, comparing the simulated Raman spectra of {Nb10 } and {Nb24 } reveals a shift of 30 cm−1 between the two, perfectly matching the experimental shift (figure S20).

Figure 9: Experimental scattering curve for (left) 100 mM TMA{Nb10 } and (right) aged and evaporated-down Nb10 + LiCl solution, along with their simulated pair distribution function (PDF) spectra for {Nb10 } and {Nb24 }, respectively. Peak assignments: W = Nb-O, X = cis Nb-Nb, Y = trans Nb-Nb, Z = distal Nb-Nb. A-D are labeled accordingly on the pictured {Nb24 } unit. Pair distribution function (PDF) of X-ray total scattering is a powerful tool to characterize cluster species in solution and amorphous solids. Comparing an experimental PDF spectrum of 100 mM TMA{Nb10 } solution to a simulation from its crystal structure (figure 9), we see good agreement in the peak positions (with notable Nb-Nb distances marked X, Y Z), as well as agreement in the relative peak intensities. The Li-aged {Nb10 } solution is ideal for comparing with simulated {Nb24 } since associated lithium ions contribute minimally to the X-ray scattering, unlike the larger alkali cations. The experimental PDF of Li-aged solution obtained by evaporation of most of the water matches well with the simulated PDF for {Nb24 } (figure 10). The shoulder (marked ‘A’) on the cis Nb-Nb peak (marked ‘X’) at 3.6 ˚ A is ascribable to Nb-Nb distances from a corner-sharing [NbO6 ] octahedron to the nearest metal center on an {Nb7 } unit, which is present in {Nb24 } and absent in {Nb10 }. Peaks W, X and Y are observed in both {Nb10 } and {Nb7 } units and are also present in both experimental and simulated {Nb24 } PDFs.

13



Furthermore, the relatively intense peak Z of {Nb10 } is significantly suppressed in {Nb24 } and also overlaps with peak B. This decrease in intensity arises from fewer cis-Nb-O-Nb-O-Nb distances present in relatively open structure of {Nb24 }, compared to {Nb10 }. Electrospray ionization mass spectrometry of the Li, K, and Cs solutions further confirm that {Nb24 } is the niobate building block that dominates all of these solutions. Spectra and identified peaks are summarized in the Supporting Information (Figure S21-S24, Tables S1-S4). Identified peaks range from the {Nb6 } unit to the {Nb24 } unit. We attribute observation of some {Nb6 } to the ionization process, which likely fragments the metastable {Nb7 }. Some formulae include both the alkali and chloride. In fact, it has been recently shown that small alkali halide clusters can serve as counterions to metal-oxide clusters. 50 Notably, there is a trend for the Li, K and Cs solutions in the amount of alkali present in the assigned peak formulae. A semi-quantitative assessment of the ESI-MS data reveals that Cs is most abundant in the observed species, Li is the least abundant, and K is intermediate. We attribute this to the well-known periodic trend of ion-association for polyniobates, where ion-association in solution increases with alkali radius. 28,51

Mechanistic proposal. We can assume the first step to alkali-promoted conversion of {Nb10 } to {Nb24 } units should involve a direct contact association with the alkali. A single alkali cation in an A–{Nb10 } pair can associate to one of three primary sites (figure 10). In sites P1 and P3, the alkali cation associates to the face of one of the fused superoctahedra in the {Nb10 } structure in a µ3 -oxo coordination environment, regardless of cation radius. However, in site P2, the smaller alkali cations (Li+ , Na+ and K+ ) associate to only the two terminal oxo ligands at the “top” of the cluster, while larger Rb+ and Cs+ generally associates to a third (bridging) oxo as well. The Gibbs Free Energy for the interaction of any alkali cation with any of the three {Nb10 } association sites (∆Gassoc,n , n = P1, P2, P3; figure 10) can be ascertained with DFT calculations upon accounting for differences in solvation between the solvent-separated A+

14


Page 14 of 31

Page 15 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 10: (top): Possible coordination positions (P) for single alkali cations to {Nb10 }, with oxo types labeled (top right). (bottom): Two possible conformers (C) for six alkali cation coordination to {Nb10 }. Key: Blue spheres = Nb, red spheres = O, purple spheres = Cs.

15



and {Nb10 }, and the ion-paired assembly (see table 2). In general ∆Gassoc decreases with increasing alkali radius. While ∆Gassoc,P 1 and ∆Gassoc,P 3 are comparable for each alkali cation, ∆Gassoc,P 2 is more negative in all cases. However, the magnitude of the difference between them decreases from around 10 kcal mol−1 in the case of Li+ to 0.5 kcal mol−1 for Cs+ . Larger cations are thus less selective about their coordination environment. Ion-pair strength alone could not directly explain reaction rates of formation of large polyniobates, since Cs+ shows lower association energy with the cluster than Li+ at all three of these positions. In the conversion of {Nb10 } to {Nb24 }, {Nb7 } subunits will form by {Nb10 } dissociation into {Nb7 } and {Nb3 }. In a previous oxygen-exchange study by Rustad and Casey, Nb-O bond strengths within the cluster did not correlate with cluster-water oxygen exchange rates, suggesting that rearrangement of {Nb10 }’s structure requires multi-step pathways. Reaction pathways of successive water molecule additions leading to {Nb10 }’s dissociation into {Nb7 } and {Nb3 } were explored by choosing structure-opening pathways based on metastable intermediates predicted by Rustad and Casey (figure S25) using the same designations for oxo types. {Nb10 } has fourteen alkali association sites: eight P1 association sites, two P2 association sites, and four P3 association sites. In the C1 conformation, the 6 alkali are associated in four P1 sites and two P3 sites, whereas there are no alkali associated in site P2. For C2, four P1 sites are occupied along with the two P2 sites (Figure 10). For Cs+ , the energy difference between the two alkali association conformers of {Nb10 } is 4 kcal mol−1 , with C2 being the most stable, but in the case of Li+ this difference is around 28 kcal mol−1 (driving the configuration of Li6 {Nb10 } mainly to the C2 conformer). Accordingly Li+ and Cs+ show different alkali association modes. We determined the energy of formation of metastable intermediates (figures S25-S27) by adding one water molecule to {Nb10 }, without alkalis and with 6 alkali ions associated to {Nb10 } (C1 and C2 conformations; figure 10, bottom). These results show the alkali association significantly affects the reaction energy, where the alkalis lower (more negative,

16


Page 16 of 31

Page 17 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


see Table S8, figures S25-S27) the energy of water addition to different oxo sites of {Nb10 }. Hence when alkalis are in solution, metastable intermediates of {Nb10 } prevail, in contrast to only “intact” {Nb10 } when there are no alkalis in solution. Finally, we determine the effect of multiple water additions on ring openings via sequential Nb-O bond breaking in E, B, E, C, and C oxo sites (figure 10, upper right). While the first two water additions (resulting in protonation of oxos) on {Nb10 } without alkalis were found to be energetically favorable, a third water molecule did not directly associate to the cluster (Table S8). Thus, we did not find any pathway that leads to the formation of {Nb7 } from {Nb10 } in neat water, in concordance with experimental results.

Figure 11: Energetic profile of the multi-step pathway converting {Nb10 } to {Nb7 } and {Nb3 } for Cs of conformer C1 figure 10, bottom left). Energy values are in kcal mol−1 . The sequence of Nb-O bond breaking is denoted by the oxo letter designations shown in figure 10. The changes at each step from each additional water molecule are highlighted in green/yellow. As alternative reaction pathways, Cs in C2 conformation and Li in C2 conformation are shown in figures S28 and S29, along with the associated reaction energies. 17



Alkali association has a strong effect on the successive water additions. The sequence of five water additions to form {Nb7 } and {Nb3 } from {Nb10 } with Cs+ ions in conformation C1 is energetically favorable, with a total energy value of −34.5 kcal mol−1 (figure 11). For conformer C2, Cs+ associated to P2 site moves to the P1 site upon the second water addition (Figure S28), allowing the system to be pulled apart to incorporate the next water molecule, demonstrating the lability of Cs. The total energy of this reaction pathway is −34.5 kcal mol−1 . This is in contrast to the behavior of Li in conformation C2. The terminal oxygens are held together by Li associated in P2, which does not permit the units to be pulled apart, resulting only in metastable structures with a maximum of three water additions (figure S29; total energy of reaction pathway = −6.59 kcal mol−1 ). The significantly more negative ∆Gassoc,P 2 of Li+ (Table 2) indicates that Li+ disproportionately favors associating to the terminal oxo ligands of {Nb10 }. K+ also favors P2, but to less of a degree than Li+ . Rb+ and especially Cs+ have a ∆Gassoc,P 2 much closer to ∆Gassoc,P 1 and ∆Gassoc,P 3 , so they are less likely to favor site P2, and will also often associate to sites P1 and P3. Additional information regarding the energetics of alkali cation association to {Nb10 } can be found in Table S5. In summary, lower association energies within increasing alkali radius plays an important role in the dissociation of {Nb10 } by facilitating successive addition of water molecules. More negative association energies and preferred occupation of P2 sites by small alkali cations (such as Li+ ) hinder the water-addition process, inhibiting dissociation, explaining the experimentally-observed trend of faster {Nb10 } to {Nb24 } n conversion with increasing alkali cation size. Table 2: Gibbs Free Energies of alkali cation association to {Nb10 } in each of the three sites. Energy values are in kcal mol−1 . ∗

Cation Li+ K+ Rb+ Cs+ ∗

∆Gassoc,P 1 -26.0 -23.6 -22.0 -20.0

∆Gassoc,P 2 -35.6 -27.5 -23.1 -20.6

∆Gassoc,P 3 -25.4 -23.0 -21.7 -19.9

Na+ not computationally studied

18


Page 18 of 31

Page 19 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Hydroxide ions are generated by protonation of {Nb7 } and {Nb24 }. Protonation occurs to stabilize the terminal oxo sites in both the intermediate {Nb7 } structure and {Nb24 } unit – specifically, the three [NbO6 ] octahedra that link the three {Nb7 } units in {Nb24 }. Protonation increases the solution pH as the reaction progresses, further accelerating the conversion of {Nb10 } to {Nb24 }. The energetics of hydroxide addition to {Nb10 } were also considered, using the open metastable state B (figure S25). 52 The association energy of OH – addition was found to be at least one order of magnitude greater than that of water addition for each position/alkali (Table 3), confirming the marked acceleration of the reaction rate with increasing solution pH that was experimentally observed by Raman (Figure 3-4). Table 3: Difference Energies between A–{Nb10 } and {Nb10 } in open metastable state B upon water and OH- addition. All energies are in kcal mol−1 . Alkali Cation, Conformation Water Addition OH – Additon Cs, C1 -1.3 -60.2 Cs, C2 0.1 -34.3 Li, C2 -8.5 -79.4 To account for the increase in solution pH without added base, we must write balanced chemical equations that yield {Nb24 } plus hydroxide as products. This requires niobate species with a positive charge in the reactants. While cationic species are generally not considered in POM chemistry, ESI MS analysis of K+ –{Nb10 } in the positive mode did indeed yield peaks (figure S24), dominated by m/z = 192.16, which is consistent with {Nb3 }+ with a range of O2 – /OH – /H2 O ligands (Table S4). We do not need to consider the alkalis because they are invariable between the reactants and products. An initial approximation is: 3 [Nb10 O28 ]6 – + 15 H2 O −−→ [Nb24 O72 H5 ]19 – + 2 [Nb3 O5 (OH)4 (H2 O)4 ]+ + OH – However, with 20 mM {Nb10 } concentration, this would yield 6.7 mM hydroxide (pH = 11.8) which is higher than the observed ∼9. Moreover, this pH would drive the decomposition of {Nb7 } to {Nb6 }, but this is not indicated by SAXS. Since {Nb6 } is a compact and stable cluster, it has no means of linking to other clusters to grow into the large species that are 19



observed by SAXS. A modified general reaction equation to account for a lower pH is: 3 [Nb10 O28 ]6 – + (14+x)H2 O −−→ [Nb24 O72 H8 ]16 – + 1.07 [Nb3 O6 (OH)3 (H2 O)4 ]2+ + 0.93 [Nb3 O6 (OH)5 (H2 O)2 ]2 – + xOH – The experimentally obtained pH of 9 is computed with x = 0.14. {Nb24 } was first structurally characterized with nine associated protons, 35 close to the eight in equation 2. Of course there are many possible variations on O2 – /OH – /H2 O ligation of the niobate clusters and fragments that is probably dynamic, as well as possible {Nb3 } ←−→ {Nb1 } equilibria. This exercise shows that we can decompose {Nb10 } to the observed, reconstructed species ({Nb24 } aggregates) with hydroxide as a byproduct and without adding acid or base to the reaction solution.

Conclusion Decaniobate ({Nb10 }, [Nb10 O28 ]6 – ) is unique amongst metal-oxo clusters of both polyanions and polycations, in that is neither acidic nor basic, while the related {Nb6 } POM and Nb2 O5 are respectively base and acid catalysts. {Nb10 } has no self-buffering capacity in water, is not readily synthesized in water, and cannot persist in water with the simplest aqueous counterions (alkalis and NH4+ ). This characteristic is also unprecedented for POMs, which are often isolated as alkali salts. We demonstrated that {Nb10 } converts to {Nb7 } building blocks, leading to large {Nb24 } oligomers, simply by addition of alkali cations, even in solution that is buffered near neutral pH. Moreover, the distinct periodic trend (increasing reaction rate: Cs>Rb>K>Na>Li) and alkali concentration dependence confirms that the alkalis are primarily responsible for driving the reaction. This, to our knowledge, is the first documented example of alkali-cation promoted cluster reassembly and growth amongst polyoxometalates or other metal-oxo clusters. This result emphasizes the importance of alkalis in not just stabilizing and crystallizing metal-oxo cluster forms in water, but also

20


Page 20 of 31

Page 21 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


driving their self-assembly. The high solubility (∼2M Nb) of these oligomers at neutral pH and their tendency to form amorphous gel networks instead of crystalline material makes them ideal precursor solutions for thin film coatings. Aqueous deposition of niobate perovskites with precise control over alkali composition is one future application-driven study conceived in this investigation. Inclusion of oxide-forming heterometals in the reaction solution is also being explored to promote {Nb10 } disassembly and reassembly, targeting functional mixed-metal polyniobates and derivative metal oxides, employing these intriguingly simple procedures.

Experimental General Methods and Materials Milli-Q water (Millipore, 18.2MΩ cm at 25 ◦C) was used for all aqueous solutions. Isopropyl alcohol (ACS grade) and ethanol (ACS grade) were purchased from Macron Fine Chemicals. The pH of the reaction mixtures was measured using an OrionT M VERSA START M pH/ISE Benchtop Multiparameter Meter. The instrument was calibrated using three standard solutions of pH 4, 7, and 10.

Synthesis and Solution Preparation We developed our own synthesis of decaniobate via hydrothermal treatment of hexaniobate. One gram of [(CH3 )4 N]5 H3 [Nb6 O19 ] 53 was added to 10 mL of ethanol and stirred for 20 minutes, resulting in a white suspension. This suspension was loaded into a Teflon cup in a Parr Reactor and heated at 140 ◦C for 18 hours. The brown supernatant was discarded and the resulting white powder was washed under vacuum with 50 mL of ethanol and allowed to dry in air. Yield: 638 mg (87.9%). Supplementary characterization data (infrared spectrum, figure S1) is identical to other published syntheses. 1 The 20 mM solutions of {Nb10 } were prepared by dissolving 579 milligrams {Nb10 } in 15 21



mL water. Six equivalents (to prepare a 120 mM alkali solution) of Li (15.3 milligrams), Na (21.0 milligrams), K (24.8 milligrams), Rb (43.5 milligrams) or Cs (60.6 milligrams) chloride were added to 3 mL aliquots of the {Nb10 } solution. The three alkali equivalent solutions were prepared by adding one-half of the above masses of alkali chloride to 3 mL of {Nb10 } solution and the one alkali equivalent solutions were prepared by adding one-sixth of the above masses.

Infrared Spectroscopy Infrared spectra (400-3500 cm−1 ) were collected on a Thermo Scientific Nicolet iS10 with a Smart Orbit Diamond ATR accessory and are reported in the Supplemental Information.

Raman Spectroscopy Raman spectra were collected on a Thermo Scientific DXR spectrometer with a 780 nm laser source, 400 lines per mm grating, and 50 µm slit with eight scans at eight seconds each.

Small & Wide Angle X-ray Scattering (SWAXS) Small and wide angle X-ray scattering was collected on an Anton Paar SAXSess with Cu-Kα −1 radiation (1.54 ˚ A) and line collimation with a q-range of 0.0182.5 ˚ A . The instrument is

equipped with a 2-dimensional image plate detector with a sample to image plate distance of 26.1 cm. Reaction solutions as well as neat water were sealed in 1.5 mm borosilicate glass capillaries. Data collection time was 30 minutes. SAXSquant software was used for data collection and initial processing. Igor Pro software utilizing Irena macros was used for the data analysis. 54 SolX software was used for creating simulated scattering curves.cite Pair distance distribution function (PDDF) analysis to determine Rg and maximum linear extent (Table 1) used the method of Moore within the Irena macros, fitting the curve up to the −1 second plateau at approximately q = 0.3 ˚ A .

22


Page 22 of 31

Page 23 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


PDF Analysis of X-Ray Total Scattering Raw X-ray scattering data were collected with a Rigaku Smartlab X-ray diffractometer with a Mo-K source (λ= 0.71073 ˚ A). For these solution X-ray scattering measurements, an aliquot of the solution was injected in a Kapton 1.5 mm capillary, sealed and positioned in the goniometer. Transmission mode of the data collection was applied, where 2θ range of 3.0118.61◦ ˚−1 . The data collection time was used. Therefore, the maximum available Q-value is 15.2 A was 0.2◦ min−1 using a 0.01 degree resolution. In order to eliminate the contribution of the solvent and the sample holder, Milli-Q water was also measured for background subtraction applying identical experimental parameters. The solution scattering curves were transformed to the reduced structure functions, then they were Fourier transformed to obtain the reduced atomic pair distribution functions (PDF, denoted as G(r) on the graph). For the mathematical transformations and background subtractions we used the PDFgetX3 software. 55 Simulated PDF data were obtained with the solX software 56 using the appropriate parameters.

Electrospray Ionization Mass Spectrometry ESI MS was carried out using an Agilent 6230 ESI MS system comprised of a Time-of-Flight (TOF) mass spectrometer coupled to an electrospray ionizer. The aged solutions were diluted to 1 mM Nb and infused into the ESI MS system at a flow rate of 0.4 mL min−1 using a syringe pump. The solutions were nebulized with the aid of heated N2 (325 ◦C) flowing at 8 L min−1 and a pressure of 35 psig (241 kPa). The voltages of the capillary, skimmer, and RT octopole were set at 3500, 65, and 750 V respectively, while the voltage of the fragmenter was set at 100 V. The data were collected in the negative and positive ionization modes.

23



DFT studies All structures were optimized using PBE0 hybrid exchange-correlation functional 57 and mixed Stuttgard-Dresden basis set SDD, whereas oxygen were treated using D95 ( T. Dunning Jr and P. J. Hay, in Modern Theoretical Chemistry, ed. H. F. Schaefer III, 1977, vol. 3) basis set, MWB28 58 were used for Niobium, SDF2 59 for lithium, MWB10 60 for potassium and MWB28 60 for rubidium and cesium. The nature of all stationary points was verified by analytical computation of vibrational frequencies, which were also used to compute 298.15 K thermochemical quantities and Raman spectra. To take into account water effects polarizable Continium model (PCM) 61 was used. After testing, Allinger radius was used for. Li, Rb and Cs and Bragg radius for K. Calculations were performed with Gaussian09. 62 For the study of the reaction mechanism, our goal is to understand intrinsic bond-breaking intermediates, though we focus on electronic energies on the energetic discussion, as the corresponding free energies will depend on the treatment of entropic effects. In addition, ADF2017 63 program system was used to perform Energy Decomposition Analysis, and structural optimization and calculation of Raman spectra of large hypothetical clusters ({Nb24 }, {Nb32 }, {Nb48 }). The PBE functional and a slater triple-zeta plus polarization basis set was used. Solvent effects were introduced by COSMO model 64 and relativistic effects are considered by means of including scalar relativistic ZORA 65,66 approach. A data set collection of computational results is available in the ioChem-BD repository[ref paper] and can be accessed via https://iochem-bd.iciq.es/browse/review-collection/ 100/9366/089e869bc6000c43427729b9. 67

Acknowledgements Both the experiments performed at OSU and computational studies performed at ICIQ were supported by the U.S. Department of Energy, Office of Basic Energy Sciences, Divi-

24


Page 24 of 31

Page 25 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


sions of Materials Sciences and Engineering, under award DE-SC0010802. MS and CB also thank CERCA Program/Generalitat de Catalunya, and the Spanish Ministerio de Econom´ıa Competitividad (MINECO) through projects CTQ-2014-52824-R and the Severo Ochoa Excellence Accreditation 2014-2018 SEV-2013-0319.

Supporting Information Available Additional FTIR, SAXS, ESI-MS, and Computational Data is available in the Supplementary Information. This material is available free of charge via the Internet at http://pubs.acs.org/.

References (1) Ohlin, C. A.; Villa, E. M.; Rustad, J. R.; Casey, W. H. Nat. Mater. 2010, 9, 11–19. (2) Zhang, G.; Liu, C.; Long, D.-L.; Cronin, L.; Tung, C.-H.; Wang, Y. J. Am. Chem. Soc. 2016, 138, 11097–11100. (3) Wills, L. A.; Qu, X.; Chang, I.-Y.; Mustard, T. J.; Keszler, D. A.; Persson, K. A.; Cheong, P. H.-Y. Nat. Commun. 2017, 8, 15852. (4) Ju, M.; Lv, J.; Kuang, X.-Y.; Ding, L.-P.; Lu, C.; Wang, J.-J.; Jin, Y.-Y.; Maroulis, G. RSC Adv. 2015, 5, 6560–6570. (5) Bera, M. K.; Antonio, M. R. J. Am. Chem. Soc. 2016, 138, 7282–7288. (6) Schreiber, R. E.; Houben, L.; Wolf, S. G.; Leitus, G.; Lang, Z.-L.; Carbó, J. J.; Poblet, J. M.; Neumann, R. Nat. Chem. 2017, 9, 369–373. (7) Talbot, D. E.; Talbot, J. D. Corrosion Science and Technology, 2nd Ed.; CRC press: New York, 2018.

25



(8) Binninger, T.; Mohamed, R.; Waltar, K.; Fabbri, E.; Levecque, P.; Kötz, R.; Schmidt, T. J. Sci. Rep. 2015, 5, 12167. (9) Hutchings, G. S.; Zhang, Y.; Li, J.; Yonemoto, B. T.; Zhou, X.; Zhu, K.; Jiao, F. 2015, 137, 4223–4229. (10) Burke, M. S.; Enman, L. J.; Batchellor, A. S.; Zou, S.; Boettcher, S. W. Chem. Mater. 2015, 27, 7549–7558. (11) Blasco-Ahicart, M.; Soriano-López, J.; Carbó, J. J.; Poblet, J. M.; Galan-Mascaros, J. Nat. Chem. 2018, 10, 24–30. (12) Fabbri, E.; Nachtegaal, M.; Binninger, T.; Cheng, X.; Kim, B.-J.; Durst, J.; Bozza, F.; Graule, T.; Schäublin, R.; Wiles, L. Nat. Mater. 2017, 16, 925–931. (13) Hochella, M. F.; Lower, S. K.; Maurice, P. A.; Penn, R. L.; Sahai, N.; Sparks, D. L.; Twining, B. S. Science 2008, 319, 1631–1635. (14) Waychunas, G. A.; Kim, C. S.; Banfield, J. F. J. Nanopart. Res. 2005, 7, 409–433. (15) Majzlan, J.; Myneni, S. C. Environ. Sci. Technol. 2005, 39, 188–194. (16) Casey, W. H.; Rustad, J. R.; Spiccia, L. Chem. - Eur. J. 2009, 15, 4496–4515. (17) Eilertsen, E. A.; Haouas, M.; Pinar, A. B.; Hould, N. D.; Lobo, R. F.; Lillerud, K. P.; Taulelle, F. Chem. Mater. 2012, 24, 571–578. (18) Jalava, J.-P.; Hiltunen, E.; Kähkönen, H.; Erkkilä, H.; Härmä, H.; Taavitsainen, V.-M. Ind. Eng. Chem. Res. 2000, 39, 349–361. (19) Borrás-Almenar, J. J.; Coronado, E.; M¨ uller, A.; Pope, M. Polyoxometalate Molecular Science; Kluwer Academic Publishers: Dordrecht, The Netherlands, 2003; Vol. 98. (20) S. Fischer, J. M. K. T., D. Kurad In Bonding and Charge Distribution in Polyoxometalates; Mingos, D., Ed.; Springer: Berlin, 1999; Vol. 93. 26


Page 26 of 31

Page 27 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(21) M¨ uller, A.; Beckmann, E.; Bögge, H.; Schmidtmann, M.; Dress, A. Angew. Chem., Int. Ed. 2002, 41, 1162–1167. (22) Zhan, C.-H.; Winter, R. S.; Zheng, Q.; Yan, J.; Cameron, J. M.; Long, D.-L.; Cronin, L. Angew. Chem. 2015, 127, 14516–14520. (23) Long, D.-L.; Tsunashima, R.; Cronin, L. Angew. Chem., Int. Ed. 2010, 49, 1736–1758. (24) Streb, C. Polyoxometalate-Based Assemblies and Functional Materials 31–47. (25) M¨ uller, A.; Sessoli, R.; Krickemeyer, E.; Bögge, H.; Meyer, J.; Gatteschi, D.; Pardi, L.; Westphal, J.; Hovemeier, K.; Rohling, R.; Döring, J.; Hellweg, F.; Beugholdt, C.; Schmidtmann, M. Inorg, Chem. 1997, 36, 5239–5250. (26) Forster, J.; Rösner, B.; Fink, R. H.; Nye, L. C.; Ivanovic-Burmazovic, I.; Kastner, K.; Tucher, J.; Streb, C. Chem. Sci. 2013, 4, 418–424. (27) Wutkowski, A.; Niefind, F.; Näther, C.; Bensch, W. Z. Anorg. Allg. Chem. 2011, 637, 2198–2204. (28) Nyman, M. Dalton Trans. 2011, 40, 8049–8058. (29) Nyman, M.; Alam, T. M.; Bonhomme, F.; Rodriguez, M. A.; Frazer, C. S.; Welk, M. E. J. Cluster Sci. 2006, 17, 197–219. (30) Livage, J. Materials 2010, 3, 4175–4195. (31) Osseo-Asare, K. Metall. Trans. B 1982, 13, 555–564. (32) Pope, M. Heteropoly and Isopoly Oxometalates (Springer-Verlag, Berlin, 1983); MT Pope and A. Miller ; Springer: Berlin, 1983. (33) Nyman, M.; Bonhomme, F.; Alam, T. M.; Rodriguez, M. A.; Cherry, B. R.; Krumhansl, J. L.; Nenoff, T. M.; Sattler, A. M. Science 2002, 297, 996–998.

27



(34) Tsunashima, R.; Long, D.-L.; Miras, H. N.; Gabb, D.; Pradeep, C. P.; Cronin, L. Angew. Chem., Int. Ed. 2010, 49, 113–116. (35) Bontchev, R. P.; Nyman, M. Angew. Chem., Int. Ed. 2006, 45, 6670–6672. (36) Bontchev, R. P.; Venturini, E. L.; Nyman, M. Inorg. Chem. 2007, 46, 4483–4491. (37) Graeber, E. J.; Morosin, B. Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem. 1977, 33, 2137–2143. (38) Villa, E. M.; Ohlin, C. A.; Balogh, E.; Anderson, T. M.; Nyman, M. D.; Casey, W. H. Angew. Chem., Int. Ed. 2008, 47, 4844–4846. (39) Ohlin, C. A.; Villa, E. M.; Casey, W. H. Inorg. Chim. Acta 2009, 362, 1391–1392. (40) Aureliano, M.; Ohlin, C. A.; Vieira, M. O.; Marques, M. P. M.; Casey, W. H.; de Carvalho, L. A. B. Dalton Trans. 2016, 45, 7391–7399. (41) Kinnan, M. K.; Creasy, W. R.; Fullmer, L. B.; Schreuder-Gibson, H. L.; Nyman, M. Eur. J. of Inorg. Chem. 2014, 2014, 2361–2367. (42) Guo, W.; Lv, H.; Sullivan, K. P.; Gordon, W. O.; Balboa, A.; Wagner, G. W.; Musaev, D. G.; Bacsa, J.; Hill, C. L. Angew. Chem., Int. Ed. 2016, 55, 7403–7407. (43) Murayama, T.; Chen, J.; Hirata, J.; Matsumoto, K.; Ueda, W. Catal. Sci. Technol. 2014, 4, 4250–4257. (44) Klemperer, W. G.; Marek, K. A. Eur. J. Inorg. Chem. 2013, 2013, 1762–1771. (45) Huang, P.; Qin, C.; Su, Z.-M.; Xing, Y.; Wang, X.-L.; Shao, K.-Z.; Lan, Y.-Q.; Wang, E.-B. J. Am. Chem. Soc. 2012, 134, 14004–14010. (46) Niu, J.; Ma, P.; Niu, H.; Li, J.; Zhao, J.; Song, Y.; Wang, J. Chem. - Eur. J. 2007, 13, 8739–8748.

28


Page 28 of 31

Page 29 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(47) Kojima, T.; Antonio, M. R.; Ozeki, T. J. Am. Chem. Soc. 2011, 133, 7248–7251. (48) Fullmer, L. B.; Malmberg, C. E.; Fast, D. B.; Wills, L. A.; Cheong, P. H.-Y.; Dolgos, M. R.; Nyman, M. Dalton Trans. 2017, 46, 8486–8493. (49) Goberna-Ferrón, S.; Soriano-López, J.; Galán-Mascarós, J. R.; Nyman, M. Eur. J. Inorg. Chem. 2015, 2015, 2833–2840. (50) Sun, Q.; Zhang, G.; Tung, C.-H.; Wang, Y. Inorg. Chem. 2016, 55, 11125–11130. (51) Antonio, M. R.; Nyman, M.; Anderson, T. M. Angew. Chem. 2009, 121, 6252–6256. (52) Rustad, J. R.; Casey, W. H. Nat. Mater. 2012, 11, 223–226. (53) Fullmer, L. B.; Mansergh, R. H.; Zakharov, L. N.; Keszler, D. A.; Nyman, M. Cryst. Growth Des. 2015, 15, 3885–3892. (54) Zhang, R.; Thiyagarajan, P.; Tiede, D. M. J. Appl. Crystallogr. 2000, 33, 565–568. (55) Juhás, P.; Davis, T.; Farrow, C. L.; Billinge, S. J. J. Appl. Crystallogr. 2013, 46, 560–566. (56) Zuo, X.; Cui, G.; Merz, K. M.; Zhang, L.; Lewis, F. D.; Tiede, D. M. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 3534–3539. (57) Adamo, C.; Barone, V. J. Chem. Phys. 1999, 110, 6158–6170. (58) Andrae, D.; Häußermann, U.; Dolg, M.; Stoll, H.; Preuß, H. Theor. Chim. Acta 1990, 77, 123–141. (59) Fuentealba, P.; Preuss, H.; Stoll, H.; Szentply, L. V. Chem. Phys. Lett. 1982, 89, 418 – 422. (60) Leininger, T.; Nicklass, A.; Kchle, W.; Stoll, H.; Dolg, M.; Bergner, A. Chem. Phys. Lett. 1996, 255, 274 – 280. 29



(61) Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev. 2005, 105, 2999–3094, PMID: 16092826. (62) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, .; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian09 Revision E.01. Gaussian Inc. Wallingford CT 2009. (63) te Velde, G.; Bickelhaupt, F. M.; Baerends, E. J.; Fonseca Guerra, C.; van Gisbergen, S. J. A.; Snijders, J. G.; Ziegler, T. J. Comput. Chem. 22, 931–967. (64) Klamt, A.; Schuurmann, G. J. Chem. Soc., Perkin Trans. 2 1993, 799–805. (65) van Lenthe, E.; Baerends, E. J.; Snijders, J. G. J. Chem. Phys. 1994, 101, 9783–9792. (66) van Lenthe, E.; Ehlers, A.; Baerends, E.-J. J. Chem. Phys. 1999, 110, 8943–8953. ´ (67) Alvarez-Moreno, M.; De Graaf, C.; Lopez, N.; Maseras, F.; Poblet, J. M.; Bo, C. J. Chem. Inf. Model. 2014, 55, 95–103.

30


Page 30 of 31

Page 31 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Graphical TOC Entry

31


Alkali-driven disassembly and reassembly of molecular niobium oxide

Recommend Documents