The Smallest Polyoxotungstate Retained by TRIS-Stabilization

A polycondensation reaction of the orthotungstate anion WO42–, buffered at pH 7.5 in a TRIS-HCl (0.15 M) solution, results in the first example of a discrete polyoxotungstate anion, with just two W ions stabilized with TRIS ligands. It was isolated and characterized as Na2[WVI2O6(C4O3NH10)2]·6H2O by single-crystal and powder X-ray diffraction, FT-IR spectroscopy, thermogravimetrical analysis (TGA), and elemental analysis in solid state and by electro-spray ionization mass spectrometry (ESI-MS), 13C, and 183W NMR, as well as Raman spectroscopy in solution. This synthesis demonstrates the crucial and new role of the added tris-alkoxy ligand in the development of a new hybrid TRIS-isopolytungstate with the lowest known nuclearity (so far) and the terminal oxygens substituted with two nitrogen atoms arising from amines of the TRIS ligands.

P olyoxometalates (POMs) are discrete anionic molecular metal-oxide clusters that are usually composed of group V and VI transition metals in their highest oxidation states and exist at the unique interface between monomeric oxometalates and polymeric metal oxides 1 exhibiting a wide range of applications. 2 There are three main criteria by which a metal oxide can be called a POM: (1) addenda ions have quasioctahedral-coordination and form d π -p π bonds with oxygen atoms, 1 (2) two octahedra are connected via sharing the edge, and (3) each octahedral unit has just two terminal O atoms. 3 Oxygen atoms are the primary ligands for the addenda metals in POMs; however, replacing them with other elements while maintaining the structure is possible. So far, many synthesis procedures of oxo-replaced POM structures require water-free organic solvents and the overwhelming majority of POMs with nitrogen atoms bound to the addenda ion are polyoxomolybdates (POMos) 4 or polyoxovanadates (POVs). 5 Attempts to directly functionalize [W 6 O 19 ] 2− polyoxotungstate (POT) have failed when applying imido approaches suitable for POMos and POVs, since [W 6 O 19 ] 2− does not react with phosphinimines, isocyanates, or primary amines. 4 In solution, small and stable POMs are interesting as useful building blocks for constructing huge metal−oxo clusters. 6 In many cases, polycondensation reactions of [MO x ] n-(M = addenda ion) immediately lead to larger structures, and POMs with low nuclearity are quite elusive. 1  = hexadentate ligand, e.g., ethylenediaminetetraacetate) in octahedral coordination are known; 11 however, they have not been obtained by acidification of orthotungstate. The strategy of using organic ligands to inhibit the formation of large clusters has previously been applied to POMos 12 and POVs 13 while it has not been reported for POTs. Tris(hydroxymethyl)aminomethane ((HOCH 2 ) 3 CNH 2 , TRIS) has been recently utilized to stabilize and isolate elusive heptavanadate. 13 While in biochemistry, TRIS (pK a 8.06) is used as pH buffer between 7.0 and 9.2, 14 in POM chemistry, TRIS is usually used for covalent organic functionalization via alkoxo-groups −CH 2 OH attachment to the POM, 15−17 and −NH 2 plays the key role for postfunctionalization through amide bond formation. 18−20 Recently, we expanded the role of TRIS by showing that TRIS, as part of a buffer solution that is normally considered unimportant, plays a defining role in the formation of a new member of the Keggin family. 21 In POM chemistry, TRIS has never acted as a primary amine by replacing oxo-ligands and thus as a protective ligand to prevent the formation of POTs with a higher nuclearity.
A discrete small anion, which is additionally organically functionalized for better stability, could be an ideal candidate for its use as a building block. To synthesize such an anion, we investigated an IPOT formation in a TRIS buffered solution (pH 7.5, 0.15 M) of WO 4 2− . Herein, we report for the first t i m e t h e s u c c e s s f u l s y n t h e s i s of t h e d i s c r e t e [W VI 2 O 6 (C 4 O 3 NH 10 ) 2 ] 2− (W 2 ), which meets all the requirements to be called a POM 1 (W VI has quasi-octahedralcoordination; two octahedra are connected via sharing the edge; W VI has just two terminal O atoms) and is the smallest POT hybridized with TRIS known so far. W 2 is a representative of a POM family with replaced oxygen ions by other nonmetals and the first POT functionalized directly by a primary amine in aqueous solution.
Initially, 12 mL of an aqueous solution of WO 4 2− (0.188 M) was acidified with HCl (1 M) to pH 4.4 followed by the addition of TRIS (0.3 g, 2.5 mmol, 0.15 M) that led to an increase in pH to 7.5 ( Figure 1A). After the final mixture was heated for 1 h at 90°C and kept at room temperature, colorless crystals of Na 2 [W VI 2 O 6 (C 4 O 3 NH 10 ) 2 ]·6H 2 O (Na 2 W 2 ) were formed. Single crystal structure analysis of W 2 revealed that TRIS acts not only as a buffer component but also as a shielding ligand, preventing formation of IPOTs with a higher nuclearity in unbuffered WO 4 10 Figure 2). The W−N bond lengths in Na 2 W 2 are slightly longer than those previously reported in POTs (2.13 to 2.17 Å), 22,23 indicating the weaker π contribution of the bonds. 24 The W−W bond length is 3.194 Å, which is shorter than that in classical IPOTs. 9,21 The stretching vibrations of the W=O units are present at 919 cm −1 in the IR spectrum of Na 2 W 2 ( Figure S1). The bands at 894 cm −1 and in the region from 470 to 750 cm −1 correspond to the antisymmetric and symmetric deformation vibrations of W−O−W. The three bands at 1084, 1059, and 1040 cm −1 are assigned to C−O stretching vibrations, indicating the successful grafting of TRIS. TGA was used to examine the weight loss and thermal stability of the synthesized hybrid POT. The TG curve shows four weightloss regions up to 700°C due to dehydration followed by disintegration of TRIS ( Figure S2). The experimental and simulated X-ray diffraction patterns of Na 2 W 2 ( Figure S3) fit perfectly in the range from 10 to 50°2θ confirming its homogeneity.
The stability of W 2 in H 2 O at pH 6.8 was investigated by ESI-MS (Figures 3, S4, S5, and Table S2). The ESI-MS spectrum recorded in positive mode exhibits three series of peaks' envelopes at m/z between 400 and 850, which can be unambiguously assigned to the singly charged cations ( Figure  3, Table S2). The first group of signals (m/z = 439.1, 461.1, 479.1, 501.0) corresponds to the monomeric complex [W VI (C 4 O 3 NH 8 ) 2 ] 0 in which two TRIS are coordinated to W VI via −CH 2 O-groups (blue in Figure 3). The signals at 711.0, 733.0, and 755.0 m/z correspond to the W 2 (purple in Figure 3) in which one more −CH 2 OH fragment is attached to one of the equivalent W VI . The W 2 with the symmetrical attachment of two additional −CH 2 OH (green in Figure 3) gives signals at 792.1, 814.1, 832.1, and 854.1 m/z. The attachment of TRIS ligands through three functional groups to the trimeric face of W VI ions is a known strategy in POM chemistry for elusive anion stabilization 13 and may be the reason that these compounds are detected in solution. The ESI-MS spectrum of Na 2 W 2 recorded in the negative mode ( Figure S4) demonstrates the absence of signals from any IPOTs.
To further examine the solution behavior of W 2 , 13 C and 183 W NMR spectroscopic studies were performed in D 2 O at pH 7.5 ( Figures S6 and S7). The 183 W NMR spectrum of Na 2 W 2 (0.058 M) in D 2 O shows one intense signal with a chemical shift at −3.4 ppm and one minor signal at −91.9 ppm ( Figure S6A). Since there exist no reference spectra for POTs with two equivalent W atoms and the chemical shift in Na 2 W 2 spectrum is very close to 0 ppm (reference Na 2 WO 4 ), a 183 W NMR spectrum for the equimolar mixture of W 2 and WO 4 2−   Figure S6D). The spectrum of W 2 recorded in NaOAc/50% D 2 O at pH 6 ( Figure S6C), where no signal for WO 4 2− should be observed (Figure 1B), shows a very intense signal at −2.1 ppm. DFT calculations have been performed to assign the peaks to tungsten species (Table S3). A comparison of calculated and experimental values for [W VI 6 O 19 ] 2− , WX 6 (X = F − , Cl − , CO) and for W 2 shows that the calculated shift of −17 ppm can be attributed to the signal around 0 ppm if assigned to W 2 . The 13 C NMR spectrum of Na 2 W 2 in D 2 O (pH 7.5) shows two signals at 59.8 and 60.8 ppm ( Figure S7B), which do not correspond to the W 2 structure observed in the solid state, where three types of carbons are present ( Figure 2). However, the signals at 59.8 and 60.8 ppm also cannot be attributed to the free TRIS with signals at 56.4 and 63.0 ppm ( Figure S7A). In the Na 2 W 2 solution at pH 6, the 13 C NMR spectrum shows three signals at 59.3, 61.4, and 63.4 ppm, which indicate potential ligand binding according to the W 2 structure (Figure 2). Raman spectroscopy was performed in the solid state and in solution to support ESI-MS and NMR spectroscopic data. The Raman spectrum of Na 2 W 2 in H 2 O (pH 7.5) points to its dissociation to orthotungstate WO 4 2− (Figure 4). On the basis of DFT calculations, ESI-MS, and NMR and Raman spectroscopy, we argue that W 2 is unstable in an aqueous solution at pH 7.5; however, some intermediates detected by ESI-MS (Figure 3) with another type of TRIS attachment are at least partially present in slightly acidic solutions.
Considering the facile and reproducible synthesis of W 2 , as well as its small size, there are two scenarios for the use of W 2 in the future for the formation of novel metal-oxide-based materials. One possible route is based on the rigid W 2 nature, which is similar to that of the dinuclear [M V 2 S 2 O 2 (H 2 O) 6 ] 2+ (M = Mo or W), 25 Figure S5 shows the spectrum in the range from 100 to 1000 m/z, and Table S2 provides all species with experimental and theoretical m/z values.  The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.