Synthesis and characterization of the Anderson–Evans tungsto­anti­monate [Na₅(H₂O)₁₈{(HOCH₂)₂CHNH₃}₂][SbW₆O₂₄]

The isolation and characterization of the Anderson–Evans-type tungsto­anti­monate [Sb^VW^VI ₆O₂₄]⁷⁻ with a T-shaped sodium–serinol com­plex, [Na₅(H₂O)₁₈{(HOCH₂)₂CHNH₃}₂]⁷⁺, as counter-cation is reported.

Abstract

A novel tungsto­anti­monate, [Na₅(H₂O)₁₈{(HOCH₂)₂CHNH₃}₂][Sb^VW^VI ₆O₂₄] (SbW₆ ), was synthesized from an aqueous solution and structurally characterized by single-crystal X-ray diffraction, which revealed C2/c symmetry. The structure contains two serinol [(HOCH₂)₂CHNH₃]⁺ and five Na⁺ cations, which are octa­hedrally surrounded by 18 water mol­ecules, and one [Sb^VW^VI ₆O₂₄]⁷⁻ anion. The serinol mol­ecules also play a critical role in the synthesis by acting as a mild buffering agent. Each of the W^VI and Sb^V ions is six-coordinated and displays a distorted octa­hedral motif. A three-dimensional supra­molecular framework is formed via hydrogen-bonding inter­actions between the tungsto­anti­monates and cations. Powder X-ray diffraction, elemental analysis, thermogravimetric analy­sis and IR spectroscopy were performed on SbW₆ to prove the purity, to identify the water content and to characterize the vibrational modes of the crystallized phase.

Introduction  

Polyoxometalates (POMs) are known as early transition metal–oxygen clusters (Pope, 1983 ▸; Gumerova & Rompel, 2020 ▸). They are assembled from {MO _x } polyhedra, with x = 4–7 and M being commonly Mo^V/VI, W^V/VI, V^IV/V, Nb^V or Ta^V as addenda ions, through sharing of corners and edges (Pope, 1983 ▸). POMs can be considered as either isopolyanions [M_m O _y ] ^q−, which feature only one metallic M ion (Mo^V/VI, W^V/VI, V^IV/V, Nb^V or Ta^V), or heteropolyanions [X_rM_m O _y ] ^q−, which additionally contain a heteroelement X (Pope, 1983 ▸). POMs display a wide range of crucial applications, ranging from catalysis (Wang & Yang, 2015 ▸), materials science (Cherevan et al., 2020 ▸) and mol­ecular magnetism (Clemente-Juan et al., 2012 ▸), to bio- and nanotechnology (Rhule et al., 1998 ▸; Bijelic et al., 2018 ▸, 2019 ▸), as well as macromolecular crystallography (Bijelic & Rompel, 2015 ▸, 2017 ▸, 2018 ▸).

The Anderson–Evans polyoxoanion has the general for­mula [H _y (XO₆)M ₆O₁₈] ⁿ⁻ , where y = 0–6, n = 2–8, M = addenda ion (Mo^VI or W^VI) and X = central heteroion in oxidation states from +2 to +7 (Blazevic & Rompel, 2016 ▸). Its structure consists of six corner- and edge-shared {MoO₆} or {WO₆} octa­hedra, which surround the {XO}₆ octa­hedron (Evans, 1948 ▸). In the structure, there exist three differently coordinated oxygen ions (Fig. 1 ▸): six triple-bridged oxygen ions (μ₃-O) that connect the heteroion and two addenda ions, six double-bridged oxygen ions (μ₂-O) that connect two addenda ions and lastly two terminal oxygen ions (O_t) per addenda ion (Evans, 1948 ▸; Pope, 1983 ▸). The oxidation state of a heteroion plays a significant role in the protonation mode of the triple-bridged oxygen ions (μ₃-O) in the Anderson–Evans archetype, according to which they can be divided into three groups. The first, i.e. [X^{n +} M ₆O₂₄]^(12–n)− (n = 5–7), referred to as ‘type A’, is a deprotonated structure that exists when it contains hetero­ions with a high oxidation state (e.g. Te^VI or I^VII). The second, i.e. [X^{n +}(OH)₆ M ₆O₁₈]^(6–n)− (n = 2–4), referred to as ‘type B’, is protonated on the six μ₃-O ions, with each side having three protons. The B-type POMs are usually present when the heteroion has a low oxidation state (e.g. Ni^II or Co^II) (Blazevic & Rompel, 2016 ▸). The third group, called ‘mixed type’, is a combination of the two types mentioned above, as it has protonated μ₃-O ions only on one side (Gumerova et al., 2019 ▸). Therefore, it is referred to as one-side protonated with one known polyoxotungstate example, [Cr^III(OH)₃W^VI ₆O₂₁]^6–, so far (Gumerova et al., 2019 ▸). Inter­estingly, there are some platinum-based com­pounds, with the general formula [H _n Pt^IV M ^VI ₆O₂₄]^(8–n)− (M = W or Mo, where 1 < n < 6; Lee et al., 2004 ▸), which exhibit protonation degrees of their μ₃-O ions ranging from 1 to 6 and n is not necessarily an integer.

Figure 1.

Displacement ellipsoid plot of SbW₆ with the hydrogen-bond inter­actions between the anion and the counter-cation com­plex highlighted (orange dashed line). Displacement ellipsoids are displayed at the 50% probability level. Colour code: W blue, Sb pink, Na green, C grey, N purple, O red and H white.

The majority of Anderson–Evans-type clusters have a planar hexa­gonal configuration with m (D _3d ) point symmetry, which is known as the α-isomer (Anderson, 1937 ▸). Another configuration of the Anderson–Evans-type cluster shows a bent structure of 2mm (C _2v ) point symmetry and is called the β-isomer (Lindqvist, 1959 ▸). An example of the β-isomer was presented by the Ogawa group (Ogawa et al., 1988 ▸), i.e. [Sb^V(OH)₂Mo^VI ₆O₂₂]⁵⁻ and is one of the few reported structures to date (Lee & Sasaki, 1994 ▸; Zhang et al., 2017 ▸; Li & Wei, 2021 ▸). The Anderson–Evans com­pounds can be modified in several ways: (i) by variation of the heteroion in the central position; (ii) by combination with various inorganic and organic cations, and (iii) by covalent attachment of one or two alkoxo ligands.

In recent decades, some unsubstituted tungsto­anti­monates, which have only one heteroion, Sb^III/V, have been synthesized and structurally characterized (Table 1 ▸), but the majority of Sb-containing polyoxotungstates (POTs) contain additional heteroions, such as 3d or 4f metals (Tanuhadi et al., 2018 ▸, 2020 ▸). Only three Sb-POTs of the Anderson–Evans archetype, namely, K₅Na₂[Sb^VW^VI ₆O₂₄]·12H₂O (Lee & Sasaki, 1987 ▸), K_5.5H_1.5[Sb^VW^VI ₆O₂₄]·6H₂O (Naruke & Yamase, 1992 ▸) and Na₇[Sb^VW^VI ₆O₂₄]·24H₂O (Mukhacheva et al., 2017 ▸), have been reported so far. Inter­estingly, the relative luminescence yield from the O→W LCMT transition of K_5.5H_1.5[Sb^VW^VI ₆O₂₄]·6H₂O was higher than that of Ln-containing Na₉[Gd^III(W^VI ₅O₁₈)₂]·18H₂O under the same conditions (Naruke & Yamase, 1992 ▸). The high luminescence yield of [Sb^VW^VI ₆O₂₄]⁷⁻ is attractive for potential photochemical applications in the future.

Table 1. POTs with Sb^III/V as the only heteroion [based on the Inorganic Crystal Structure Database (FIZ, Karlsruhe; http://www.fiz-informationsdienste.de/DB/icsd/www-recherche.htm) and the Cambridge Structural Database (CSD; Groom et al., 2016 ▸) in April 2021].

POT	Type	References
K5Na2[SbVWVI 6O24]	Anderson	Lee & Sasaki (1987 ▸)
K5.5H1.5[SbVWVI 6O24]	Anderson	Naruke & Yamase (1992 ▸)
Na7[SbVWVI 6O24]	Anderson	Mukhacheva et al. (2017 ▸)
K6[H12SbV 6WVI 4O36]	Pseudo-Anderson–Evans dimer	Park et al. (1994 ▸)
(NH4)9[SbVWVI 18O60(OH)2]	Dawson	Zhang et al. (2010 ▸)
Na9[SbIIIWVI 9O33]	Keggin	Bösing et al. (1997 ▸)
K12[SbIII 2WVI 22O74(OH)2]	Krebs	Bösing et al. (1997 ▸)
[N(CH3)4]10Na12[Na2SbIII 8WVI 36O132(H2O)4]	Trimer based on lacunary Keggin	Bösing et al. (1997 ▸)
(H2en)8H6{[SbIII 2(WVIO2)2(B-β-SbIIIWVI 9O33)2][(WVIO2)2(WVIO3)2(B-β-SbIIIWVI 9O33)2]} (en = ethyl­enedi­amine)	Krebs	Xin et al. (2019 ▸)
K11Na16[H2(SbIIIWVI 9O33)(WVI 5O12)(SbIII 2WVI 29O103)]	Trimer based on lacunary Keggin	Tanuhadi et al. (2021 ▸)

Serinol (C₃H₉NO₂, 2-amino­propane-1,3-diol) is a very stable, highly water soluble, nontoxic, odourless, biodegradable compound which is widely used as a versatile starting material in organic synthesis and as an additive for materials applications, such as com­posite materials (Barbera et al., 2020 ▸; Andreessen & Steinbüchel, 2011 ▸). In POM synthesis, serinol can be seen as an alk­oxy­lation ligand or a counter-cation or buffering agent due to the presence of an amino group. Considering that the Sb-centred Anderson–Evans POT has not yet been reported with organic counter-cations, we expand the com­pound class by applying serinol, which can coordinate in different ways to metals through the –NH₂ and –HOCH₂ groups and thus significantly affects both the structure and properties, in the synthesis in Sb⁵⁺–WO₄ ²⁻ (with an Sb:W ratio of 1:6) systems. Here we report a novel Anderson–Evans Sb-centred POT, [Na₅(H₂O)₁₈{(HOCH₂)₂CHNH₃}₂][Sb^VW^VI ₆O₂₄] (SbW₆ ), being the first example of [Sb^VW^VI ₆O₂₄]⁷⁻ crystallized with an organic counter-cation, which was synthesized from aqueous solution and has been fully characterized.

Experimental  

Synthesis and crystallization  

The reagents were used as purchased from Merck (Austria) and VWR (Austria) without further purification.

Synthesis of [Na₅(H₂O)₁₈{(HOCH₂)₂CHNH₃}₂][SbW₆O₂₄] (SbW₆)  

Na₂WO₄·2H₂O (0.99 g, 3 mmol) and KSb^V(OH)₆ (0.13 g, 0.5 mmol) were mixed in a 6:1 ratio in H₂O (12 ml), yielding a turbid solution. The solution was then acidified with aqueous HCl (1 M, 4.4 ml) and the pH was set at 4.0. Serinol [(HOCH₂)₂CHNH₂; 0.18 g, 2 mmol] was then added and the pH was altered to 7.1. Under stirring and heating for 1 h, at 75 °C, the precipitate was dissolved, and the final solution was colourless. The pH after the reaction was 7.0. The solution was left for evaporation at room tem­perature, leading to colourless crystals suitable for single-crystal X-ray diffraction within 1 d (yield: 0.4 g, 60%, based on W). The pH of the Sb⁵⁺–WO₄ ²⁻ solution was varied from 3.7 to 5.0; however, after the addition of serinol (0.18 g, 2 mmol), the final pH was in the range from 7.0 to 7.7 and, in all cases, crystals with the same unit cell were obtained. Other synthetic routes, such as reflux reaction and hydro­thermal synthesis at 120 °C, for the same reaction mixture led to the same product. Elemental analysis found (calculated) for [Na₅(H₂O)₁₈{(HOCH₂)₂CHNH₃}₂][Sb^VW^VI ₆O₂₄] (%): C 3.18 (3.23), H 2.56 (2.53), N 1.29 (1.25), O 32.89 (32.97). FT–IR (cm⁻¹): 3357 (s), 2952 (sh), 1610 (s), 1498 (s), 1464 (sh), 1373 (w), 1256 (w), 1099 (w), 1037 (s), 1018 (sh), 927 (s), 850 (s), 703 (sh), 632 (w), 640 (w), 563 (w), 420 (s), 349 (s), 310 (s).

IR spectroscopy  

SbW₆ was characterized by IR spectroscopy on a Bruker Vertex70 IR Spectrometer equipped with a single-reflection diamond-ATR unit in the range 4000–300 cm⁻¹.

TGA measurements  

Thermogravimetric analysis (TGA) was performed on a Mettler SDTA851e Thermogravimetric Analyzer under a nitro­gen flow with a heating rate of 5 K min⁻¹ in the region from 303 to 873 K.

Elemental analysis  

The determination of C/H/N/O was carried out using an ‘EA 1108 CHNS-O’ elemental analyzer by Carlo Erba Instruments at the Mikroanalytisches Laboratorium, Faculty of Chemistry, University of Vienna.

Powder X-ray diffraction (PXRD)  

PXRD was performed on a Bruker D8 Advance diffrac­tometer, with Cu Kα radiation (λ = 1.54056 Å), a Lynxeye silicon strip detector and a SolX energy dispersive detector (variable slit aperture with 12 mm, 10° ≤ 2θ ≤ 50°).

Refinement  

In Table 2 ▸, the crystallographic characteristics of SbW₆ and the experimental conditions of data collection and refinement are reported. The positions of the H atoms of the water mol­ecules were obtained by difference Fourier techniques and were refined with free isotropic displacement parameters and O—H distances restrained to 0.95 (2) Å. The disordered water mol­ecule in the coordination sphere of atom Na1 was refined with two positions (O23 and O24), with free occupancy factors to a total of 100%. The H atoms of this disordered group had U _iso(H) values set to 1.5U _eq(O) of the parent atom. H atoms bound to N or C atoms were placed in idealized positions (N—H = 0.91 Å and C—H = 0.99 or 1.00 Å for CH₂ and CH groups, respectively) and refined in riding modes, with U _iso(H) values set to 1.5U _eq(N) or to 1.2U _eq(C).

Table 2. Experimental details.

Crystal data
Chemical formula	[Na5(H2O)18(C3H10NO2)2][SbW6O24]
M r	2232.32
Crystal system, space group	Monoclinic, C2/c
Temperature (K)	200
a, b, c (Å)	21.9761 (14), 13.9179 (9), 16.209 (1)
β (°)	111.189 (2)
V (Å3)	4622.5 (5)
Z	4
Radiation type	Mo Kα
μ (mm−1)	15.61
Crystal size (mm)	0.1 × 0.08 × 0.05
Data collection
Diffractometer	Bruker APEXII CCD
Absorption correction	Multi-scan (SADABS; Bruker, 2016 ▸)
T min, T max	0.542, 0.747
No. of measured, independent and observed [I > 2σ(I)] reflections	100214, 8841, 8213
R int	0.030
(sin θ/λ)max (Å−1)	0.770
Refinement
R[F 2 > 2σ(F 2)], wR(F 2), S	0.013, 0.025, 1.15
No. of reflections	8841
No. of parameters	390
No. of restraints	23
H-atom treatment	H atoms treated by a mixture of independent and constrained refinement
Δρmax, Δρmin (e Å−3)	0.60, −0.69

Computer programs: APEX3 (Bruker, 2015 ▸), SAINT (Bruker, 2016 ▸), SHELXT2018 (Sheldrick, 2015a ▸), shelXle (Hübschle et al., 2011 ▸), SHELXL2018 (Sheldrick, 2015b ▸), OLEX2 (Dolomanov et al., 2009 ▸), DIAMOND (Brandenburg, 2006 ▸), OLEX2 (Dolomanov et al., 2009 ▸) and PLATON (Spek, 2020 ▸).

Results and discussion  

The preparation of SbW₆ was carried out at a W^VI to Sb^V ratio of 6:1 and at a pH of 7.1. In the absence of serinol, at pH 7.5, protonated K_5.5H_1.5[Sb^VW^VI ₆O₂₄]·6H₂O (Naruke & Yamase, 1992 ▸), and at pH 4.5, unprotonated K₅Na₂[Sb^VW^VI ₆O₂₄]·12H₂O (Lee & Sasaki, 1987 ▸), were obtained.

The main structural elements of SbW₆ are the Anderson–Evans [Sb^VW^VI ₆O₂₄]⁷⁻ anion and the com­plex [Na₅(H₂O)₁₈{(HOCH₂)₂CHNH₃}₂]⁷⁺ cation, which are connected via hydrogen bonds between terminal (O_t) and bridging O atoms (μ₂-O) of the polyanion and protons from the cationic com­plex (Fig. 1 ▸). Crystallographically centrosymmetric [Sb^VW^VI ₆O₂₄]⁷⁻ shows the characteristic Anderson–Evans A-type structure with a central {SbO₆} octa­hedron surrounded by six edge-shared {WO₆} octa­hedra that form a planar array of distorted octa­hedra (Fig. 1 ▸). The average W—Sb bond length is 3.26 Å. As is typical for all Anderson–Evans A-type structures, three different coordination modes of the O atoms are present in the structure: six triple-bridged oxygen ions (μ₃-O) connect the heteroion and two W ions, six double-bridged oxygen ions (μ₂-O) connect two W ions and two terminal oxygen (O_t) ions are connected to each of the six W ions (Fig. 1 ▸). The average distance for Sb—μ₃-O is 1.98 Å, for W—μ₃-O is 2.27 Å, for W—μ₂-O is 1.94 Å and for W—O_t is 1.74 Å. The values are com­parable with those of K_5.5H_1.5[Sb^VW^VI ₆O₂₄] (Naruke & Yamase, 1992 ▸). For instance, the Sb—μ₃-O bond length (1.98 Å) differs by only 0.03 Å from the others reported by Naruke & Yamase (1992 ▸) (2.01 Å). Applying bond valence sum (BVS) calculations (Brown & Altermatt, 1985 ▸), all the W ions in [Sb^VW^VI ₆O₂₄]⁷⁻ exhibit the +VI oxidation state (average calculated value of 6.01) and Sb shows the +V oxidation state (5.37). Based on BVS analysis and the number of counter-cations, it was concluded that the Anderson–Evans anion is not protonated and belongs to type A.

The counter-cation is com­posed of five octa­hedrally coordinated Na⁺ ions, which assemble in an elevated T-shape form, and two protonated serinol mol­ecules that are coordinated to two Na⁺ ions via –HOCH₂ groups. This group has crystallographically imposed twofold symmetry. One serinol ligand inter­acts through the –NH₃ group with the terminal O atom of the POT anion, and the second inter­acts with the adjacent O atom in {NaO₆} (Fig. 1 ▸).

The three-dimensional (3D) structure of SbW₆ consists of two-dimensional (2D) sheets formed of [Sb^VW^VI ₆O₂₄]⁷⁻ anions and com­plex [Na₅(H₂O)₁₈{(HOCH₂)₂CHNH₃}₂]⁷⁺, cations connected via hydrogen bonds (Figs. 1 ▸ and 2 ▸). The distances between 2D layers are approximately 2.79 Å, which allows the formation of hydrogen bonds between the layers and creates cavities along the b axis (Fig. 2 ▸ a). This packing is different to that observed in K_5.5H_1.5[Sb^VW^VI ₆O₂₄]·6H₂O and K₅Na₂[Sb^VW^VI ₆O₂₄]·12H₂O, where the layers of anions alternate with layers or single polyhedra of counter-cations.

Figure 2.

The crystal packing of SbW₆ , (a) viewed along the b axis and (b) viewed along the z axis. Colour code: {WO₆} turquoise, {SbO₆} orange, {Na(H₂O)₆} grey, C grey, N blue, O red and H black.

The IR spectrum of SbW₆ (Fig. 3 ▸) is characteristic for Anderson–Evans POMs (Liu et al., 2015 ▸; Qu et al., 2012 ▸). The broad bands in the region between 2300 to 3750 cm⁻¹ represent the vibrations of the –OH groups of H₂O and the N—H bonds of the amine groups of serinol. In the area between 1610 and 1018 cm⁻¹, the bands are attributed to the vibrations of C—H, C—O and again N—H in serinol. The bands at about 930 and 880 cm⁻¹ are attributed to anti­symmetric stretching vibrations of the terminal W=O bonds and Sb—O—W bridges (O_b), respectively. The bands at 640 and 563 cm⁻¹ are associated with the asymmetric stretching of W—O—W bridges (O_b) and the bending vibrations of W—O—W, respectively. Lastly, the bands between 750 and 300 cm⁻¹ are contributed by Sb—O—W vibrations (Liu et al., 2015 ▸).

Figure 3.

IR spectrum of SbW₆ in the region from 4000 to 300 cm⁻¹.

The powder XRD pattern of SbW₆ (Fig. 4 ▸) was investigated at room tem­perature. The simulated powder diffraction pattern was based on the single-crystal structural data. The observed peak positions are in good alignment with the simulated patterns, which confirms that the POT structure had been solved accurately and that SbW₆ consists of a single phase.

Figure 4.

Experimental (blue colour) and simulated (black colour) powder XRD pattern of SbW₆ .

The exact number of water mol­ecules was determined using TGA. The curve (Fig. 5 ▸) shows three weight-loss steps during the heating process from 30 to 600 °C. The first weight loss of 14.7% in the tem­perature range 30–200 °C corresponds to all water mol­ecules from the Na⁺ coordinating spheres. The second and third step correspond in total to 8.4% and the loss of two serinol mol­ecules.

Figure 5.

Thermogravimetric curve of SbW₆ .

The com­position of the counter-cation has remarkable effects on the crystal packing and thus on the physical properties of Anderson–Evans POMs (Blazevic & Rompel, 2016 ▸). The success in synthesizing SbW₆ shows that the Sb-centred Anderson–Evans POT is a versatile building block, which can be modified by organic counter-cations into high-dimensional architectures. SbW₆ is the first reported K⁺-free salt with an organic counter-cation, and it has much higher water solubility and can expand the areas of its application in aqueous solution.

Supplementary Material

CCDC reference: 2079597

Funding Statement

This work was funded by Austrian Science Fund grants P33927 and P33089 to Nadiia I. Gumerova and Annette Rompel; Erasmus+ grant 1016/2020 to Kleanthi Sifaki.