Synthesis of a novel Cu^I/Cu^II-containing sandwich-type cluster and its catalytic electron transfer property

Abstract

A novel six-metal sandwich-type heteropolyanion [Na₂Cu^ICu^II(OH₂)Cu₂^II(B-α-SbW₉O₃₃)₂]⁹⁻ (1) has been synthesized in good yield from a one-pot hydrothermal reaction and characterized by IR, XPS, XRD and TG. Single-crystal X-ray analysis reveals that compound H₉[Na₂Cu^ICu^II(OH₂)Cu₂^II(B-α-SbW₉O₃₃)₂]·9H₂O (H₉-1) crystallizes in tetragonal system with space group of P4̄2₁m and consists of a 2D inorganic network based on Cu–O–W linkages among polyanionic clusters. A structural feature is that the six-metal {Na₂Cu₄} central belt of the anion contains the mixed valence states of Cu^I/Cu^II group. This compound shows excellent catalytic activity for effectively promoting the inorganic electron transfer (redox) reaction of ferricyanide to ferrocyanide by thiosulphate with high rate constant value in aqueous solution. In addition, compound (H₉-1) also exhibits an inhibition effect for the organic photodegradation reaction of Rhodamine B (RhB).

---

1. Introduction

The synthesis and exploitation of polyoxometalates (POMs) have attracted considerable attention in recent years, owing to not only their diversities of shape, size, and redox properties but also their potential applications in catalysis, photochemistry, electrochemistry, and magnetism.^1–4 Although POMs has been known for about 200 years, the mechanism of formation of polyanionic clusters is not well understood and is commonly described as self-assembly. Therefore, systematic structural design for novel POMs and derivatization for known POMs are still an attractive, longstanding and challenging topic. In the material world, properties are determined by structures, therefore there is an ongoing search for new POM clusters with special active sites.

Transition-metal substituted POMs (TMSPs) are well known, and many such complexes have been reported.⁵ Among the class of TMSPs, the sandwich-type species represent a relatively larger subfamily.⁶ As is well known, the Hervé-,⁷ Krebs-,⁸ Weakly-,⁹ and Knoth-¹⁰ sandwich-type polyoxoanions had been obtained and reported, and this kind of Sb^III-centered polyoxotungstates have been known for a long time.¹¹ Due to the existence of a lone pair of electrons on the heteroatom, which precludes the formation of the most stable Keggin-type cluster, it is possible to obtain many more novel sandwich-type TMSPs with unprecedented structure and property. The Krebs group has reported a large number of heteropolyanions, such as (C₅₂H₆₀NO₁₂)₁₂[(Mn(H₂O)₃(SbW₉O₃₂)₂], [Mn^II₂(H₂O)₆(WO₂)(SbW₉O₃₂)₂]¹⁰⁻, [M₂(H₂O)₆(WO₂)₂(SbW₉O₃₂)₂]¹⁰⁻ (M = Mn^II, Zn^II) and [Sb₂W₂₂O₇₄(OH)₂]¹²⁻.^12–14 Their work indicated that the manganese(II)-contained polyoxoanion [Mn^II₂(H₂O)₃(Sb-W₉O₃₂)₂]¹²⁻ possesses a highly efficient catalytic feature in the epoxidation of alkenes. The groups of Kortz and Proust respectively reported two compounds (CsNa₂)[{Sn(CH₃)}₃(H₂O)₄(β-SbW₉O₃₃)]·7H₂O and [Sb₂W₂₀O₇₀{ρ-cymene}₂]¹⁰⁻, being the first organotin derivative of β-[SbW₉O₃₃]⁹⁻ and the first obtained organometallic heteropolytungstate related to [Sb₂W₂₂O₇₄(OH)₂]¹²⁻ by self-assembly.^15,16 Very recently, Niu et al. reported the synthesis and magnetic property of five Sb^III-containing polyoxotungstates.⁵ In 2012, our group reported study on the synthesis of a {V=O}₆-containing inorganic–metal–organic sandwich-type tungstoantimonite, which was accompanied by in situ new carbon–carbon bond formation of organic cations.¹⁷

Atomic compositions and combination modes of the central belt have important influence on the physicochemical property of the last sandwich-type assemblies. Introduction of the active metal sites into the polyanionic skeleton might adjust some parameters of physical properties. In the current work, when Cu cations with the evident Jahn–Teller effect and flexible coordination geometries are utilized to hydrothermally react with Na₂WO₄, Sb₂O₃ and ethylenediamine(en), one distinctive inorganic sandwich-type tungstoantimonate H₉[Na₂Cu^ICu^II(OH₂)Cu^II₂(B-α-SbW₉O₃₃)₂]·9H₂O (H₉-1) has been successfully obtained and characterized. A structural feature is that the six-metal central belt of anion contains mixed metal {Na₂Cu₄} centers, and the copper atoms present mixed valence states of Cu^I/Cu^II. Its photodegradation behaviors for Rhodamine-B (RhB), as well the catalytic electron transfer (redox) for ferricyanide to ferrocyanide by thiosulphate have been investigated during our experiments. It is found that this compound shows excellent catalytic activity for inorganic reduction reaction of ferricyanide to ferrocyanide by thiosulphate with high rate constant value in aqueous solution. But on the contrary, compound (H₉-1) exhibits an inhibition effect for the organic photodegradation reaction of RhB. A feature of POM catalyst is in their structural stability and integrality with nano-sized anionic cluster, which can be separated easily after reaction and reused.

2. Results and discussion

2.1. Structure description of compound (H₉-1)

Single-crystal X-ray diffraction study reveals that polyoxoanion in 1 consists of two [B-α-SbW₉O₃₃]⁹⁻ moieties with one mixed six-membered metal {Cu₄Na₂} belt, resulting in the formation of sandwich-type cluster (see Fig. 1a). The trivacant B-α-[SbW₉O₃₃]⁹⁻ unit could be seen as a derivative from the parent α-Keggin structure by taking off three edge-sharing {WO₆} octahedra. The {SbO₃} tetrahedron located at the center is surrounded by three vertex-sharing {W₃O₁₃} trimers. The central belt of polyanion is composed of three adjacent oxygen-shared copper ions (Cu(1), Cu(2) and Cu(2i)) and an isolated copper ion (Cu(3)), which is separated from the copper triad by two sodium ions, namely the six-membered center belt consists of one Cu⁺(1), three Cu²⁺(2/3) and two Na⁺(1) ions within a plan (Fig. 1b). Interestingly, the central four copper ions are not completely equivalent to each other because of their different coordinated environments, which could be grouped as three types: the first is two {Cu(2)O₅} tetragonal pyramids, which are coordinated by four interior oxygen atoms from two [α-SbW₉O₃₃]⁹⁻ units, and the vertex position is occupied by O_b shared with W atom (Cu(2)–O(20)–W(5)) from one adjacent polyanion; the second is {Cu(3)O₄(OH₂)} with a similar polyhedral environment to Cu(2) except the vertex position occupied by one water molecule (Cu(3)–O(8) = 2.23(2) Å); the last type is that the {Cu(1)O₄} group is merely coordinated by four interior oxygen atoms from two [α-SbW₉O₃₃]⁹⁻ subunits.

Fig. 1 (a) Polyhedral and ball-and-stick representation of the sandwich-type structural unit of 1. (b) Ball-and-stick representation of the central six-metal belt {Cu₄Na₂O₁₅} in 1. The {WO₆} octahedra are shown in pink, and the balls represent Sb (teal), Na (gray), Cu (turquoise), O (red) atoms, respectively. Solvent water molecules are omitted for clarity.

As shown in our previous work¹⁷ on six-metal cations in sandwich-type POMs, 1 represents a new member in this family. It is interesting to discuss their bond lengths and angles within the central mixed-metal section of the polyanion. The 3 five-coordinated Cu²⁺ ions are arranged in an approximately equilateral triangle with distances of Cu(2)⋯Cu(3) = 4.8175(37) Å, Cu(2)⋯Cu(2i) = 4.8608(23) Å and angles of Cu⋯Cu⋯Cu = 59.70° and 60.60° (Fig. 1b). Moreover, the square-pyramidal coordination spheres of Cu(2) (Cu(2)–O_eq. = 1.933(9) Å, Cu(2)–O_ax = 2.263(13) Å) and Cu(3) (Cu(3)–O_eq. = 1.937(10) Å, Cu(3)–O_ax = 2.23(2) Å) are fairly regular and exhibit the expected Jahn–Teller distortion (axial elongation). However, the Cu(1) (Cu(1)–O_eq. = 2.149(9) Å) located between two Cu(2) atoms exhibits a distorted square-planar coordination. In addition to the four copper ions, the central belt of polyanion 1 also incorporates two Na⁺ ions, which are located at two sides of the equilateral triangle of copper ions and are bound to four oxygen atoms from two {SbW₉O₃₃} fragments with the expected bond length range of (Na(1)⋯O = 2.161(12)–2.352(14) Å). To the best of our knowledge, the structure⁵ of previously reported [Na₂Cu₄Cl(B-α-SbW₉O₃₃)₂]⁹⁻ is similar to that in 1, but mainly arising from the highly disordered positions of Cu and Na sites.

Furthermore, the results of bond valence calculations¹⁸ for 1 show +1 for Cu(1), +2 for Cu(2/3) and +6 for W centers, which have also been confirmed by XPS measures. These oxidation states of atoms are consistent with their coordination environments. It should be noted that this kind of Cu^I/Cu^II mixed oxidation state should be ascribed to the reduction activity of the organic amine during hydrothermal reaction. So en here plays a double role of structure-template and reducing agent during the reaction.

The 2D structure of (H₉-1) is composed of individual polyoxoanions connected to four neighbors via four Cu(2)–O(20)–W(5) bridges. The two equivalent copper centers Cu(2) and Cu(2i) share the terminal oxygen donors with tungsten centers W(5) of the two adjacent polyanions. Simultaneously, the two symmetrically equivalent tungsten centers W(5) and W(5i) in 1 are linked to the two equivalent copper atoms Cu(2) from the two adjacent polyanions (Fig. 2a). This solid-state structure distribution is a clear result of the unsymmetrical composition of four copper ions within the central ring of each polyanion 1. Three of the four copper ions (Cu(2), Cu(2i), Cu(1)) are close on the copper-rich side of the ring, whereas the Cu(3) is located on the copper-poor side of the ring (separated from each other by the two equivalent sodium ions Na(1) and Na(1i) (Fig. 1b). The lattice of (H₉-1) can be described as double layers of 1 where the copper-rich sections of the central belt point at each other (Fig. 2b). Polyanions in one layer are arranged vertically to each other, whereas the other layers are arranged horizontally, and all polyanions in the “vertical layer” are completely parallel, leading to an attractive 2D (4,4′) network Fig. 2c). As we know, the extended 2D-netlike-structure based on sandwich type tungstoantimonate containing mixed valence copper ions in the central belt has not been reported.

Fig. 2 (a) Polyhedral/ball-and-stick representation showing the connectivity sheet of neighbouring polyanions by Cu(2)–O(20)–W(5) bridges in the lattice of 1; (b) a side view of the sheet seen along the b axis; (c) a simplified stacking scheme showing the arrangement of different inorganic sheets. Each node represents that of anion 1. The color codes are the same as that in Fig. 1.

2.2. IR spectroscopy and XPS of Cu in H₉-1

In IR spectrum of compound (H₉-1), the peaks at 947, 893, 773 and 734 cm⁻¹ are attributed to the characteristic vibrations of ν(Sb–O_a), ν(W–O_d), ν(W–O_b–W) and v(W–O_c–W), respectively. The strong peak at 1610 cm⁻¹ is assigned to ν(H–O–H). In addition, the wide bands between 2800 and 3450 cm⁻¹ can be assigned to ν(O–H) (Fig. 3a).¹⁴ The existence of Cu^I and Cu^II in the (H₉-1) is confirmed by XPS measurement (Fig. 3b). The XPS spectrum of the pure sample displays the peaks of Cu2P_3/2 and Cu2P_1/2 at 933.6 and 952.8 eV attributable to Cu^I, respectively, and the peaks at 942.6 and 961.4 eV indicate the presence of Cu^II ion.¹⁹ XPS spectra also confirm the existence of W^VI and Sb^III in 1 (Fig. S1 and S2†).

Fig. 3 (a) IR Spectroscopy, (b) XPS of Cu in compound 1.

2.3. Voltammetric behaviour

To study the redox property of compound (H₉-1), 1-modified CPE (1-CPE) was fabricated as the working electrode, according to the method described in Section 2. The cyclic voltammetric behavior of 1-CPE in 1 mol L⁻¹ H₂SO₄ solution in the range from +1.00 V to −0.85 V at a scan rate of 5 mV s⁻¹ is shown in Fig. 4. In the potential domain explored, it consists of three evident reduction waves with peak potentials located at −0.055 V (II), −0.170 V (III) and −0.704 V (IV) vs. SCE. In the voltammogram, the first oxidation peak (I) for 1-CPE is attributed to oxidation processes of the Cu⁰ → Cu²⁺, and the second pair of quasi-reversible redox peaks (II–II′) is attributed to the redox process of the Cu⁺ → Cu²⁺,^20,21 whereas the last two waves (III and IV) for 1-CPE are assigned to the redox processes of W atoms in the polyoxoanion framework; the domain where the waves are located was also observed in the other tungsten-containing POMs.^22–24

Fig. 4 Cyclic voltammogram of the 1-CPE in the 1 mol L⁻¹ H₂SO₄ solution, scan rate: 5 mV s⁻¹, vs. SCE.

2.4. Catalytic study

2.4.1 Photocatalysis. POM-based photocatalytically active materials have received considerable attention, because a variety of organic substrates can be oxidized photocatalytically, even mineralized, by POMs under UV-visible irradiation.^25–27 In this experiment, we examined the degradation process of RhB by 500 W Hg lamp irradiation in the presence of (H₉-1) as the photocatalyst. As is well-known, the RhB substrate, consisting of four N-ethyl groups at either side of the xanthene ring, is relatively stable in aqueous solution in darkness or upon visible light irradiation.²⁵ When RhB solution is kept in darkness either in the presence of or in the absence of (H₉-1), the degradation reactions of RhB hardly occurred (Fig. S3 and S4†). It can also be found that crystals do not have clear adsorption performance for RhB. However, the RhB substrate in the absence of (H₉-1) undergoes pronounced photodegradation in aqueous solution upon 500 W Hg lamp irradiation, and the UV-visible spectral maximum absorbance of the degradation solution decreases from 1.71 to 0.54 after 140 min (Fig. 5a). The UV-visible spectral change processes during the photodegradation of RhB at various times in the presence of (H₉-1) is shown in Fig. 5b. Interestingly, compared with that in the absence of (H₉-1), the UV-visible spectral maximum absorbance of the degradation solution in the presence of (H₉-1) slowly decreases from 1.71 to 1.36, after 140 min, proving that (H₉-1) can to some extent inhibit the photodegradation of RhB. The phenomenon is different from those POM-based compounds that can promote the photodegradation of RhB.^25,26 The characteristic absorption peaks of RhB (around 553 nm) are chosen to study the conversion of RhB. The plots of the conversion of RhB (K) varying with reaction time (t) are shown in the insets of Fig. 5. The conversion of K can be expressed as K = (I₀ − I_t)/I₀, where I₀ represents the UV-visible absorption intensity of RhB at the initial time (t = 0), and I_t is the UV-visible absorption intensity at a given time (t).²⁸ The conversion of RhB in the absence of (H₉-1) upon irradiation for 140 min is ca. 69.84%, whereas the conversion of RhB in the presence of (H₉-1) is only 19.36%, which also reveals the inhibiting effect for RhB. the main reasons might be as follows: (i) 1 can work as absorbers of the Hg lamp irradiation; (ii) it is found in experiments that the crystal of 1 is only sparingly soluble, resulting in a very small amount of polyanions dissolved in solution. The hydrogen-bonding interactions between donors of RhB substrates (N(C₂H₅)₂, COOH) and acceptors of (H₉-1) (surface oxygen atoms of POMs) enhance the chemical stability of RhB substrate in solution, which leads to the slow photodegradation of RhB substrates.² In addition, IR spectrum of (H₉-1) after the photocatalytic degradations was also measured, which proves that the structure of (H₉-1) is still retained after the photocatalytic experiment (Fig. S5†). The recycle experiments show that compound (H₉-1) maintained high inhibition performance after 2 repeated experiments (Fig. S6 and S7†). Evidently, the inhibition of the photodegradation experiment of RhB is interesting to explore many functionalized TMSP compounds as color protection agents.

Fig. 5 (a) UV-visible absorption spectra of RhB solutions at various UV irradiation times as a blank experiment; (b) UV-visible absorption spectra of RhB solutions at various UV irradiation times with the existence of compound (H₉-1). Insets: the conversion of RhB (K) with reaction time (t).

2.4.2 Catalytic studies for reduction of Fe(CN)₆³⁻ by S₂O₃^2-. The catalytic electron transfer property of compound (H₉-1) is tested in case of the inorganic redox reaction between Fe(CN)₆³⁻ and S₂O₃²⁻ at room temperature (25 °C) and 55 °C, respectively. This reaction could hardly occur in absence of catalyst at these temperatures.²⁹ It has been recently reported that the noble Au or Pt nanoparticles (NPs) are considerably more catalytically active for this redox reaction.^30–32 However, the rate of reaction was slow at 25 °C as observed by other workers using Au and Pt NPs as catalysts. Earlier work also confirmed that the catalysis reaction occurred through the electron transfer at the noble metal Au or Pt surface. Sandwich-type polyanion (H₉-1) as a nano-sized cluster has a large metal delocalized π bond located at the surface, which might be more conducive to electron transfer. The reduction process is monitored with respect to time using UV-visible spectrometer. The depletion of the Fe(CN)₆³⁻ peak at 420 nm has been used to study the rate of such catalyzed reaction. Fortunately, the experimental result reveals that the (H₉-1) is catalytically active for this inorganic redox reaction. We found that the electron transfer reaction proceeds with difficulty using compound (H₉-1) as catalyst at room temperature (Fig. 6a). However, the rate of reaction and reduction degree increases a lot at higher temperature (55 °C) (Fig. 6b). The apparent rate constant values calculated from the slope of the plot of −ln A₄₂₀ against time are: k = 1.23 × 10⁻³ min⁻¹ at 25 °C and k = 9.52 × 10⁻³ min⁻¹ at 55 °C (Fig. 6c). The k parameter is even higher than that observed in the presence of Au doped mesoporous boehmite film material as catalyst (the k = 8.43 × 10⁻³ min⁻¹ at 55 °C).²⁹

Fig. 6 Successive UV-visible absorption spectra of the reduction of Fe(CN)₆³⁻ in the presence of (H₉-1) at two different temperatures: (a) 25 °C; (b) 55 °C; (c) the pseudo first order plots of −ln A (absorbance intensity at 420 nm) versus time for the above reactions.

With the consideration that compound (H₉-1) may be partially dissolved in this case, the simple catalytic redox comparative experiments were also completed in the similar fashion with the use of WO₃ (abbreviated as W), Sb₂O₃+WO₃ (abbreviated as Sb + W), CuO + WO₃ (abbreviated as Cu + W) and CuO + Sb₂O₃+WO₃ (abbreviated as Cu + Sb + W). The ratio of metal oxides is calculated according to the corresponding value in 1. As illustrated in Fig. 7, changes in the concentration of Fe(CN)₆³⁻ solution are versus reaction time. It can be seen that the catalytic activity increases from 10% for Sb + W, 24% for W, 35% for Cu + Sb + W and 39% for Cu + W to 78% for compound (H₉-1) after 2.5 h of reaction time. With a rate of 1.664 mg L⁻¹ min⁻¹ for the reduction of Fe(CN)₆³⁻ solution catalyzed by (H₉-1), approximately 50% of Fe(CN)₆³⁻ had been reduced within an hour, which indicates that the probable synergistic effect between the transition metals assembled in the polyanionic cluster produce a higher catalytic activity of compound (H₉-1) for the reduction of Fe(CN)₆³⁻. In addition, when compared to mixed metal oxides with WO₃ alone, it is easy to find that the presence of Sb₂O₃ and CuO may play a role in inhibiting and promoting, respectively, for the reduction reaction of Fe(CN)₆³⁻. Through these abovementioned results, it can also be concluded that compound (H₉-1) does not decompose but just partially dissolves in the solution during the reaction process, which is probably affected by high temperature. In contrast to the previously reported use of precious metals Au and Pt NPs as catalysts, the compound (H₉-1) is inexpensive and easier to synthesize. Furthermore, the repeated experiment had also been done, but the result displayed that the catalytic performance of (H₉-1) declined a lot (Fig. S8†). A possible reason is that the surface of (H₉-1) adsorbs a lot of Fe(CN)₆⁴⁻, which could impede the contact (H₉-1) and Fe(CN)₆³⁻, reducing the catalytic activity. The IR spectrum of (H₉-1) after the catalytic reduction may be used to prove it (Fig. S9†). Considerable work needs to done for addressing more problems.

Fig. 7 UV-visible absorption spectra of the reduction of Fe(CN)₆³⁻ with the use of five groups of catalysts; black curve represents the control experiment without any catalyst. (The ratio of metal oxides is calculated according to the corresponding value in 1; the reaction temperature is at 55 °C).

3. Experimental section

3.1. Synthesis and characterization

All chemicals are commercially purchased and used without further purification. FTIR spectrum is recorded in a KBr pellet with a FTIR-8900 IR spectrometer in the range of 400–4000 cm⁻¹ region. Thermogravimetric analysis (TG) is carried out using a Perkin-Elmer Pyris Diamond TG/DTA instrument in flowing N₂ with a heating rate of 10 °C min⁻¹. Powder X-ray diffraction (XRD) is determined by a Bruker AXS D8 Advance diffractometer. Cyclic voltammogram (CV) is obtained with a CHI 660B electrochemical workstation at room temperature. UV spectra are obtained by a U3010 UV-visible spectrophotometer (Shimadzu).

3.1.1 Synthesis of compound (H₉-1). A mixture of Na₂WO₄·2H₂O (400 mg, 1.2 mmol), Sb₂O₃ (50 mg, 0.17 mmol), Cu(CH₃COO)₂·H₂O (0.25 mg, 1.2 mmol), en (0.4 mL, 5.92 mmol) was dissolved in 10 mL distilled water at room temperature. The pH value of the mixture was adjusted to ca. 3.7 with 4 mol L⁻¹ HCl. Then the suspension was put into a Teflon-lined autoclave and maintained under autogenous pressure at 180 °C for 5 days. After slow cooling to room temperature at a rate of 10 °C h⁻¹, green tetragonal crystals were filtered and washed with distilled water (35% yield based on W).

The XRD and TG curves for compound (H₉-1) are provided in Supporting materials as Fig. S10 and S11,† respectively.

3.2. X-ray crystallography

Single crystal of compound (H₉-1) was selected for data collection performed on a Smart Apex CCD diffractometer at 296(2) K with Mo Kα monochromated radiation (λ = 0.71073 Å). The structure was solved by direct methods and refined by the full-matrix least-squares method on F² using the SHELXTL-97 program package.³³ Anisotropic thermal parameters were used to refine all non-hydrogen atoms. The limiting indices were (−17 ≤ h ≤ 19, −15 ≤ k ≤ 19, −16 ≤ l ≤ 15) for (H₉-1). All crystal data and structure refinement details for the compound (H₉-1) are given in Table 1.

Table 1 Crystal data and structure refinement details for the compound (H₉-1)

Empirical formula	Cu4H23Na2O76Sb2W18
Formula weight	5068.94
Crystal system	Tetragonal
Space group	P4̄2(1)m
a, b, c/Å	16.466(4),16.466(4), 14.115(6)
α, β, γ/°	90
Volume/Å^3, Z	3827(2), 2
Density (calculated)/(Mg m^−3)	4.399
Absorption coefficient	28.820
F(000)	4360
Crystal size/mm^3	0.17 × 0.15 × 0.13
Theta range	1.90–25.01°
Reflections collected	18 926
Independent reflections/R(int)	3552 (0.0393)
Completeness	99.6
Absorption correction (%)	Empirical
Max. and min. transmission	0.024 and 0.010
Data/restraints/parameters	3552/0/259
Goodness-of-fit on F^2	1.045
Final R indices [I > 2σ(I)]	R1 = 0.0265, wR2 = 0.0651
R Indices (all data)	R1 = 0.0302,wR2 = 0.0674

---

3.3. Methods and materials

3.3.1 Electrochemical property. Cyclic voltammetry measurements were carried out on a CHI 660 electrochemical workstation at room temperature. Platinum gauze was used as a counter electrode, and a saturated calomel electrode (SCE) was used as reference electrode. Chemically bulk-modified carbon paste electrodes (CPEs) were used as the working electrodes.

The compound (H₉-1) modified CPEs (1-CPE) was fabricated as follows: 0.1 g of graphite powder and 0.01 g of (H₉-1) were mixed and ground together by an agate mortar and pestle to achieve a uniform mixture and then 0.1 mL paraffin oil was added with stirring. The homogenized mixture was packed into a plastic tube with a 2 mm inner diameter. Electrical contact was established with a copper rod through the back of electrode.

3.3.2 Photocatalytic reaction. Photocatalytic reactions were carried out in a light reaction tube of 50 mL capacity attached to an inner radiation type. 25 mg crystal (H₉-1) was dispersed into aqueous solution containing RhB (10 mg L⁻¹). Then the mixture was stirred continuously under ultraviolet (UV) irradiation from a 500 W high pressure mercury vapour lamp. A sample was taken every 20 min and then centrifuged to remove the particles of the catalyst, but the first 10 min were not considered to rule out the effect of its adsorption to the particle surfaces. Changes in the value of the maximum absorbance at 553 nm were used to measure the degradation rate of RhB.

3.3.3 Catalytic reduction of Fe(CN)₆³⁻. In this experiment, 35 mg compound (H₉-1) was mixed together with 16 mg K₃Fe(CN)₆ and 108 mg Na₂S₂O₃·5H₂O dissolving in 50 mL distilled water at 25 or 55 °C. Then the mixture was stirred continuously. At every 20 min interval, a 5 mL aliquot was sampled and then centrifuged to remove the particles of the catalyst. The Fe(CN)₆³⁻ concentration (C) was determined by measuring the maximum absorbance at 420 nm as a function of irradiation time using a U3010 UV-visible spectrophotometer (Shimadzu). The amount of catalysts in the reference tests were calculated according to the corresponding metal oxides mass fraction in compound (H₉-1).

4. Conclusions

In summary, a novel sandwich-type transition-metal-substituted tungstoantimonate has been hydrothermally synthesized and structurally characterized. The sandwich-type skeleton [Na₂Cu^ICu^II(OH₂)Cu^II₂(B-α-SbW₉O₃₃)₂]⁹⁻ is composed of two trivacant Keggin [B-α-SbW₉O₃₃]⁹⁻ moieties linked by a belt of six-membered ring cluster {Na₂Cu₄} displaying an intriguing 2D-netlike structure in a paralleled fashion. Voltammetric behavior of (H₉-1) has been investigated. Furthermore, the catalytic activity of (H₉-1) for the photocatalytic degradation of RhB and the reduction of ferricyanide to ferrocyanide by thiosulphate has been explored. Especially, in contrast to the previously reported use of precious metals Au and Pt NPs as catalysts for the electron transfer reaction, compound (H₉-1) is inexpensive and efficient. This work provides a new idea for POM-based materials' application. Currently, we are trying to synthesize many more different sandwich type transition-metal substituted POMs with novel structure and special features.

Acknowledgements

This work was financially supported by the Natural Science Foundation of China (21341003, 21272054), and the Hebei Natural Science Foundation of China (no. B2011205035).

Notes and references

1. D. L. Long, R. Tsunashima and L. Cronin, Angew. Chem., Int. Ed., 2010, 49, 1736.
2. Z. G. Han, X. Q. Chang, J. S. Yan, K. N. Gong, C. Zhao and X. L. Zhai, Inorg. Chem., 2014, 53, 670–672.
3. H. C. Li, X. Yu, H. W. Zheng, Y. M. Li, X. H. Wang and M. X. Huo, RSC Adv., 2014, 4, 7266–7274.
4. D. M. Fernandes, L. C. Silva, R. A. S. Ferreira, S. S. Balula, L. D. Carlos, B. D. Castro and C. Freire, RSC Adv., 2013, 3, 16697.
5. J. P. Wang, P. T. Ma, J. Li, H. Y. Niu and J. Y. Niu, Chem.–Asian. J., 2008, 3, 822.
6. M. T. Pope, Compr. Coord. Chem. II, 2003, 4, 635.
7. U. Kortz, S. Isber, M. H. Dickman and D. Ravot, Inorg. Chem., 2000, 39, 2915.
8. B. Botar, T. Yamase and E. Ishikawa, Inorg. Chem. Commun., 2001, 4, 551.
9. D. Drewes, E. M. Limanski, M. Piepnbink and B. Krebs, Z. Anorg. Allg. Chem., 2004, 630, 58.
10. F. B. Xin and M. T. Pope, J. Am. Chem. Soc., 1996, 118, 7731.
11. J. Fischer, L. Richard and R. Weiss, J. Am. Chem. Soc., 1976, 98, 3050.
12. D. Volkmer, B. Bredenkötter, J. Tellenbröker, P. Kögerler, D. G. Kurth, P. Lehmann, H. Schnablegger, D. Schwahn, M. Piepenbrink and B. Krebs, J. Am. Chem. Soc., 2002, 124, 10489.
13. M. Piepenbrink, E. M. Limanskiu and B. Krebs, Z. Anorg. Allg. Chem., 2002, 628, 1187.
14. M. Bösing, I. Loose, H. Pohlmann and B. Krebs, Chem.–Eur. J., 1997, 3, 1232.
15. F. Hussain, M. Reicke and U. Kortz, Eur. J. Inorg. Chem., 2004, 2733.
16. D. Laurencin, R. Villanneau, P. Herson, R. Thouvenot, Y. Jeannin and A. Proust, Chem. Commun., 2005, 5524.
17. Z. G. Han, Q. X. Zhang, Y. Z. Gao, J. J. Wu and X. L. Zhai, Dalton Trans., 2012, 1332–1337.
18. I. D. Brown and D. Altermatt, Acta Crystallogr., Sect. B: Struct. Sci., 1985, 41, 244–247.
19. L. M. Zheng, P. Yin and X. Q. Xin, Inorg. Chem., 2002, 41, 4084–4086.
20. B. Keita, E. Abdeljalil, L. Nadjo, R. Contant and R. Belgiche, Electrochem. Commun., 2001, 3, 56.
21. A. X. Tian, J. Ying, J. Peng and J. Q. Sha, Inorg. Chem., 2008, 47, 3274.
22. B. S. Bassil, U. Kortz, A. S. Tigan, J. M. Clemente-Juan, B. Keita, P. Oliveira and L. Nadjo, Inorg. Chem., 2005, 44, 9360.
23. B. S. Bassil, S. Nellutla, U. Kortz, A. C. Stowe, J. van Tol, N. S. Dalal, B. Keuta and L. Nadjo, Inorg. Chem., 2005, 44, 2659–2665.
24. I. M. Mbomekalle, B. Keita, M. Nierlich, U. Kortz, P. Berthet and L. Nadjo, Inorg. Chem., 2003, 42, 5143–5152.
25. C. C. Chen, W. Zhao, P. O. Lei, J. C.Zhao and N. Serpone, Chem.–Eur. J., 2004, 10, 1956.
26. Q. Wu, W. L. Chen, D. Liu, C. Liang, Y. G. Li, S. W. Lin and E. B. Wang, Dalton Trans., 2011, 56.
27. I. A. Weinstock, Chem. Rev., 1998, 98, 113.
28. J. W. Zhao, D. Y. Shi, L. J. Chen, X. M. Cai, Z. Q. Wang, P. T. Ma, J. P. Wang and J. Y. Niu, CrystEngComm, 2012, 14, 2797.
29. D. Jana, A. Dandapat and G. De, Langmuir, 2010, 26, 12177.
30. K. H. Su, Q. H. Wei, X. Zhang, J. J. Mock, D. R. Smith and S. Schultz, Nano Lett., 2003, 3, 1087.
31. S. K. Medda, S. De and G. J. De, J. Mater. Chem., 2005, 15, 3278.
32. A. Pöppl, T. Rudolf and D. Michel, J. Am. Chem. Soc., 1998, 120, 4879.
33. G. M. Sheldrick, SHELXTL-97, Programs for Crystal Structure Refinement, University of Göttingen, Germany, 1997.

---

Footnote

† Electronic supplementary information (ESI) available: The additional XPS, UV, IR, XRD, and TG curves. CCDC 1017979. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c4ra08145e

---