Shuangxue
Wu
a,
Yaomei
Fu
b,
Hongmei
Gan
a,
Liang
Zhao
*a,
Xinlong
Wang
a,
Chao
Qin
*a and
Zhongmin
Su
c
aKey Laboratory of Polyoxometalate and Reticular Material Chemistry of Ministry of Education, Northeast Normal University, Changchun, Jilin 130024, China. E-mail: zhaol352@nenu.edu.cn; qinc703@nenu.edu.cn
bShandong Peninsula Engineering Research Center of Comprehensive Brine Utilization, Weifang University of Science and Technology, Shouguang 262700, China
cKey Laboratory of Advanced Materials of Tropical Island Resources, Hainan University, Haikou 570228, China
First published on 24th December 2024
Only rarely have polyoxometalates been recorded to form core–shell clusters, which presumably relates to the scarcity of appropriate building blocks. Herein, two high-nuclear core–shell polyoxophosphovanadates [M6O19]2−⊂[Na8VV6VIV12O18(PhPO3)24]2+ (M = W 1, Mo 2) are synthesized utilizing Lindqvist polyoxometalates as templates. Both clusters exhibit a nested three-shell architecture characterized by an octahedron@cube@octahedron configuration. The outer shell [VV6VIV12O18(PhPO3)24]6− consists of two simple {VO5} and {PhPO3} building blocks connected alternately and exhibits the same octahedral geometry as the inner core [M6O19]2− due to anionic template effects. To the best of our knowledge, compounds 1 and 2 represent the first core–shell polyoxovanadates templated by Lindqvist polyoxometalates. Furthermore, compound 1 demonstrates efficacy and stability as a catalyst for the oxidation of sulfide. This work provides a new perspective on the construction of novel core–shell polyoxometalates.
![]() | ||
Fig. 1 (a) Representative high-nuclear polyoxophosphovanadates; (b) the {VxPy} building blocks. Color code: red, O; pink, P; green, V. H atoms are omitted for clarity. |
Core–shell POMs demonstrate unique structural characteristics during their assembly, which are influenced by the size, shape, and charge of the anionic templates.3,23,24 These anionic templates, whether pre-introduced or synthesized in situ, serve a crucial role in guiding the structural configuration of the clusters. Typically, most core–shell POVs are constructed with small-size anionic guests (e.g., ClO4−, N3−, halides, etc.) as templates, such as [Cl4⊂V18O25(H2O)2(PhPO3)20]4−, [N3⊂V14O22(OH)4(PhPO3)8]7−,25 and {V16As8}, {V16As10}, {V20As8}, and {V24As8} mediated by {Xx(H2O)6y} (X = Br, Cl; y = 2, 4, 6).26 However, only rarely have polyoxometalates mediated by anionic templates based on large-size POMs been found to form core–shell nanoclusters. Only a few wheel-shaped core–shell polyoxomolybdates have been reported, such as {Mo36⊂Mo150},27 {PW12/PW18/P5W30⊂Mo24Fe12}28 and {P5W30⊂Mo22Fe8}.29 Unfortunately core–shell POV cages utilizing polyanions as templates have not been reported to date. Consequently, we endeavored to employ POM-based anionic templates to modulate the {VxPy} building blocks, facilitating the synthesis of high-nuclear core–shell POVs. Herein, we successfully synthesized two core–shell POVs [M6O19]2−⊂[Na8VV6VIV12O18(PhPO3)24]2+ (M = W 1, Mo 2) with Lindqvist POM template, presenting a new building block {VO(PhPO3)4} (Fig. 1b). The outer shell [VV6VIV12O18(PhPO3)24]6− can be viewed as self-assembled from six {VVO(PhPO3)4} vertices and twelve {VIVO5} edges with the same octahedral geometry as the inner [M6O19]2− template. To our knowledge, compounds 1 and 2 represent the first and largest high-nuclear core–shell POVs encapsulating a POM template. Compound 1 exhibits good thermal and chemical stability alongside good catalytic oxidative activity towards sulfur ethers, efficiently converting substrates such as phenyl methyl sulfide at room temperature.
Furthermore, the octahedral shell [VV6VIV12O18(PhPO3)24]6− can be described from the point of view of the building blocks. It can be viewed as a self-assembly of six {VVO5(PhPO3)4} square pyramids (Fig. 3e) serving as vertices and twelve {VVIO5} units (Fig. 3f) as edges. Each {VIVO5} edge is bordered by two neighboring {VVO5(PhPO3)4} vertices through four V–O–P bonds. All V atoms exhibit two different coordination modes. The VV ion at the apex is connected to the four surrounding {PhPO3} ligands via four μ2-O, resulting in a square pyramid {VVO5(PhPO3)4} centered on {VVO5}, where the P–O–V bond angles vary from 149.27° to 150.23° (Table S4†). Conversely, the VIV ion, which constitutes an edge, is also connected to the surrounding {PhPO3} ligand via four μ2-O, but it forms a nearly planar {VIVO5(PhPO3)4} square building block (Fig. S4†) centered on {VIVO5}, with all V–O–P bonds curved in-plane at an angle of 158.23° to 160.02° (Table S5†). Significantly, the square pyramidal {VVO5(PhPO3)4} and {VIVO5(PhPO3)4} building blocks are observed for the first time in polyoxophosphovanadates.
In addition, the octahedral shell [VV6VIV12O18(PhPO3)24]6− can also be viewed as consisting of two larger square pyramidal building blocks {VVVIV4(PhPO3)4} (Fig. 3h) encircling a square {VV4VIV4(PhPO3)16} ring (Fig. 3i) with fourfold symmetry. In particular, the square {VV4VIV4(PhPO3)16} ring cluster is interconnected by four {VVO5(PhPO3)4} square pyramids and four {VIVO5(PhPO3)4} square shared {PhPO3} groups with all {PhPO3} groups and VO bonds are orientated outward. It is intriguing to note that the {VVVIV4(PhPO3)4} square pyramid presented here is different from the reported {V5P4} pyramid (Fig. S5†).25 At first glance, exchanging the positions of VIV and PO3− in the bottom squares of the pyramid {VVVIV4(PhPO3)4} evolves into {V5P4}. The distinction between the two structures primarily arises from the sequential arrangement in which VIV/PO3− is connected to the center VV cation. In the {VVVIV4(PhPO3)4} unit, a VV center coordinates with four {PhPO3} groups through four μ2-O bonds, forming a convex square unit that subsequently connects to the surrounding four VIV ions via eight μ2-O bonds to form a square pyramid. Conversely, in the {V5P4} unit, a VV center is first linked to the four surrounding VIV ions through four μ3-O bonds, resulting in a {V5} convex square unit that is further linked to the four {PhPO3} groups via eight μ2-O bonds, thereby forming a {V5P4} square pyramid. Additionally, the square {VV4VIV4(PhPO3)16} ring cluster is distinct from the {V5P5} ring,22 mainly in the valence state of vanadium and the coordination mode of vanadium with PO3− (Fig. S6†). The {V5P5} building block is an expanded ring with 5-fold symmetry consisting of five {VIVO5} units alternately linked with five {PhPO3} ligands, where only two oxygen atoms on the diagonal of the chassis in the {VIVO5} pyramid are linked to neighboring {PhPO3} ligands. In contrast, the square {VV4VIV4(PhPO3)16} ring cluster comprises four {VVO5 (PhPO3)4} pyramids alternately linked with four {VIVO5} units, with all V–O bonds connected to {PhPO3} ligands. Notably, this planar square {VV4VIV4(PhPO3)16} ring, composed of vanadium centers with varying valences, is presented in POVs for the first time.
Powder X-ray diffraction analysis (Fig. S7†) revealed that the experimental peaks of 1 and 2 closely matched the simulated curves, confirming its good phase purity. The FTIR spectrum (Fig. S8†) of 1 exhibited characteristic absorption peaks at 1076 cm−1, 725 cm−1, and 450 cm−1, which are attributed to the telescopic vibration of the terminal oxygen WO, the asymmetric telescopic vibration and the bending vibration of the W–O–W, respectively. The V
O bond stretches occurred at 1025 cm−1 and 992 cm−1, while the V–O bond stretches were observed at 696 cm−1 and 564 cm−1. The V–O–P bond stretches at 1148 cm−1 and 1121 cm−1, and the monosubstituted absorption peak of the phenyl groups at 756 cm−1; C
C bond stretches occurred at 1430 cm−1, 1484 cm−1, and 1669 cm−1; and the absorption band at 3054 cm−1 is caused by the C–H stretching vibration. The X-ray photoelectron spectra (XPS) (Fig. S9 and S10†) of W in compound 1 showed four peaks fitted at 38.8 eV, 37.5 eV, 35.1 eV, and 33.6 eV corresponding to the binding energies of the W(VI) 4f5/2 satellite peak, the W(VI) 4f7/2 satellite peak, and the binding energies of W(VI) 4f5/2 and W(VI) 4f7/2, thus proving that the W atoms of compound 1 have the valence state +VI. The XPS spectrum of V 2p shows two peaks at 513.2 eV and 520.6 eV corresponding to the binding energies of VIV2p3/2 and VIV2p1/2, respectively, and peaks at 514.8 eV and 521.9 eV corresponding to the binding energies of VV2p3/2 and VV2p1/2, respectively, which confirms that the atomic valence states of V are +IV and +V.30 In addition, charge balance analysis and bond valence calculations (BVS) indicate that the oxidation state of the W atoms in compound 1 is +VI. The oxidation states of the V atoms are +IV and +V, which is consistent with the X-ray photoelectron spectra (Tables S6 and S7†).
The stability of POMs is crucial for both scientific research and industrial applications. Therefore, we further investigated the thermal and solvent stability of compound 1. Thermogravimetric analysis (Fig. S11†) of compound 1 showed a stable plateau between 240 °C and 350 °C, indicating that compound 1 maintains its structural integrity at this temperature and has good thermal stability. Furthermore, after immersing 1 mg of compound 1 in 2 ml of solvents with different polarities (e.g., MeCN, MeOH, EtOH, H2O, DMF, CH2Cl2) for 72 h, 1 can retain its structural integrity, which was consistent with the morphology before immersion, indicating that compound 1 exhibits high solvent stability (Fig. S12†).
In the presence of a catalyst, sulfides can be converted to sulfone and sulfoxide, which are essential intermediates in organic synthesis, thereby reducing the emission of hazardous substances sulfur oxides (SOx). Therefore, the exploration of catalysts with good catalytic activity for sulfur ether compounds is of great significance for industrial applications and environmental protection.31 Polyoxovanadates exhibit advantageous electron transfer capabilities and redox activity, making them suitable as environmentally friendly molecular catalysts.32–34 It is essential to investigate the oxidative desulfurization catalytic performance of compound 1 to further elucidate the relationship between molecular structure and its functional characteristics. Firstly, the oxidative desulfurization was carried out in different solvents at room temperature with methyl phenyl sulfide (MBT) (0.2 mmol) as a substrate, compound 1 (0.00125 mmol) as oxidative desulfurization catalyst, 30% hydrogen peroxide (2 mmol) as oxidant, and biphenyl as counterpart. The experimental outcomes were monitored using liquid chromatography (LC). The results showed that optimal catalytic performance was achieved when the reaction solvent was MeOH (2 mL) with 99% conversion of MBT within 10 min (Table S8†). Under this solvent condition, the oxidizing agent was replaced with an equivalent amount of tert-butyl hydroperoxide, which showed a slightly poorer oxidation effect and 78.63% MBT conversion at the same time. Furthermore, in the absence of compound 1, the conversion of MBT was only 19.25%, indicating that compound 1 is the key substance facilitating the catalytic oxidation process.
Next, we performed desulfurization experiments on thioether compounds with different electron-absorbing (electron-donating) effect substituents under suitable catalytic conditions to investigate the catalytic activity of compound 1 (Table 1). The experimental results showed that under the optimal conditions, ethyl phenyl sulfide and 4-methoxy thioanisole were almost completely converted in 10 min, 4-chlorothioanisole was completely converted in 20 min, diphenyl sulfide was completely converted in 1 h. In the case of dibenzothiophene, the reaction was almost non-responsive at room temperature, and the conversion rate could reach 52.64% after warming up to 50 °C and reacting for 3 h. These results indicate that compound 1 can catalyze compounds with different electron-withdrawing (electron-donating) effect substituents. Furthermore, substituents with electron-withdrawing effects and high space resistance are detrimental to the reaction. Meanwhile, increasing the reaction temperature can accelerate the catalytic oxidation.
Entry | Substrate | Temperature (°C) | Time (min) | Conv.b (%) |
---|---|---|---|---|
a Reaction conditions: substrate (0.2 mmol), compound 1 (0.00125 mmol), H2O2 (2 mmol), and MeOH (2 mL). b Isolated conversions were calculated by LC. | ||||
1 |
![]() |
25 | 10 | >99 |
2 |
![]() |
25 | 10 | >99 |
3 |
![]() |
25 | 10 | >99 |
4 |
![]() |
25 | 10 | 89.64 |
20 | >99 | |||
5 |
![]() |
25 | 10 | 45.38 |
60 | >99 | |||
6 |
![]() |
25 | 10 | Trace |
50 | 180 | 52.64 |
The catalytic stability and recoverability of compound 1 were further investigated using MBT as a substrate. After one round of catalytic experiments, the reaction mixture was centrifuged, the precipitate was washed with methanol several times and observed under a microscope, revealing the complete morphology of crystal 1. Furthermore, the crystals were centrifuged and filtered after a period of catalytic reaction, and the remaining liquid continued to be reacted. The results showed that the system without crystals had no catalytic activity for MBT, which further confirmed that compound 1 was a stable non-homogeneous catalyst (Fig. S13†). Subsequently, the precipitate was dried and reintroduced into a new reaction for further catalysis. After five rounds of cyclic experiments, the conversion of MBT was still as high as 98.55% (Fig. S14 and Table S9†). The PXRD patterns before and after catalysis are in high agreement, indicating that compound 1 has good structural stability and recoverability in the catalytic process (Fig. S15†). Furthermore, a potential mechanism for the catalytic oxidation of sulfides by Compound 1 is detailed in the ESI (Fig. S16†).
Footnote |
† Electronic supplementary information (ESI) available: Additional characterization data, including PXRD patterns, IR and XPS spectra, TG curve and single-crystal X-ray diffraction data. CCDC 2401866 and 2401867. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d4qi02884h |
This journal is © the Partner Organisations 2025 |