Two New Compounds Based on Bi-Capped Keggin Polyoxoanions and Cu-Bpy Cations Contain Both CuII and CuI Complexes: Synthesis, Characterization and Properties

Two inorganic–organic hybrid complexes based on bi-capped Keggin-type cluster, {([CuII(2,2′-bpy)2]2[PMoVI8VV2VIV2O40(VIVO)2])[CuI(2,2′-bpy)]}∙2H2O (1) and {[CuII(2,2′-bpy)2]2[SiMoVI8.5MoV2.5VIVO40(VIVO)2]}[CuI0.5(2,2′-bpy)(H2O)0.5] (2) (bpy = bipyridine), had been hydrothermally synthesized and structurally characterized by elemental analysis, FT-IR, TGA, PXRD and X-ray single-crystal diffraction analysis. Compound 1 consists of a novel 1-D chain structure constructed from [CuI(2,2′-bpy)]+ unit linking bi-supported POMs anion {[CuII(2,2′-bpy)2]2[PMoVI8VV2VIV2O40(VIVO)2]}−. Compound 2 is a bi-capped Keggin cluster bi-supported Cu-bpy complex. The main highlights of the two compounds are that Cu-bpy cations contain both CuI and CuII complexes. Furthermore, the fluorescence properties, the catalytic properties, and the photocatalytic performance of compounds 1 and 2 have been assessed, and the results show that both compounds are active for styrene epoxidation and degradation and adsorption of Methylene blue (MB), Rhodamine B (RhB) and mixed aqueous solutions.


Synthesis Discussion
Compound 1 was isolated from the reaction of Na 3 [P(Mo 3 O 10 ) 4 ]·xH 2 O, V 2 O 5 , CuCl 2 ·2H 2 O, H 2 C 2 O 4 ·2H 2 O, 2,2 -bpy and distilled water at 180 • C for 3 days by the hydrothermal method with pH adjusted to 4, while compound 2 was separated from the hydrothermal reaction of Na 2 MoO 4 , Na 2 SiO 3 ·9H 2 O, V 2 O 5 , CuCl 2 ·2H 2 O, H 2 C 2 O 4 ·2H 2 O, 2,2 -bpy and distilled water in similar conditions to compound 1. In the synthesis reactions, H 2 C 2 O 4 ·2H 2 O does not appear to be involved in the play and part in the assembly of compounds 1 and 2. To study the effects of H 2 C 2 O 4 ·2H 2 O, we tried to synthesize 1 and 2 under the same conditions without H 2 C 2 O 4 ·2H 2 O, but no desired crystal was found. It means that H 2 C 2 O 4 ·2H 2 O not only influences the pH values of the system but also acts as a reducing agent under hydrothermal conditions. Mo VI , V V and Cu II in the starting materials of 1 and 2 were reduced to Mo V , V IV 4 ] 3− is close to a cubic configuration which is formed from two half-occupied tetrahedral encapsulated {PO 4   [Cu(2)(2,2 -bpy)] + is a linker, in which two half-occupied Cu (2) and Cu(2a) (symmetry codes: 1 − x, −y, 2 − z) receives contributions from four half-occupied N donors of a disordered 2,2 -bpy molecule ( Figure S2 (10) and 2.787(4) Å respectively. A review of the literature revealed that a disordered 2,2 -bpy ligand acting as a linear structural ligand linked into a 1-D chain structure has been rarely reported [46].
A highlighting feature of compound 1 is that two Ow1 form an (H 2 O) 2 water cluster with O . . . O distances of 2.7053(2) Å which is connected to the adjacent 1-D chain to 2-D supramolecular layers. As shown in Figure 2b, Ow1 of (H 2 O) 2 cluster interactions with O8(#1), O9, O13, O20(#2) and O9(#1) from two {([Cu II (2,2 - Table 1.     6− , and in both the central atom and valence are all discrepant. Secondly, the π . . . π interactions are a critical and non-negligible element in the formation of the packing structure in 3, but no strong π . . . π interactions exist in 1. The hydrogen bonds of (H 2 O) 2 water cluster and POMs increase the stability of the crystal of 1. Thirdly, disordered 2,2 -bpy in 3 is C/N co-occupying sites with the ratio of 0.5, and the C and N latter of 2,2 -bpy in 1 is disorderedly occupied over four positions (N5 and C21, N6 and C22) with the occupancy of 0.5 ( Figure S2). Last and most important, all the copper in the Cu-bpy cations of 3 are in the +2 valence, but the copper cations contain both +2 and +1 valence in 1.    (2) ions are very rare. The half-occupancy Cu(2) cation is coordinated with the two nitrogen atoms of one 2,2′-bpy ligand and a semi-water molecule composed of two disorder one-quarter water molecules (O24 and O24′, symmetry codes: 1 − x,y,0.5 − z) whose O…O bond length is 1.600(1) Å, forming a counter cation [Cu(2)0.5(2,2′-bpy)(H2O)0.5] 0.5+ . The bond lengths of Cu(2)-N and Cu(2)-O are 1.976(9) and 2.244(4) Å, respectively. . . π stacking between the 2,2 -bpy aromatic rings, which makes the structure of compound 2 novel and stable. As shown in Figure 4a, there exists only a kind of π . . . π stacking interaction that occurs between identical N1 pyridine rings that come from two adjacent bi-supported POMs to form an infinite 1-D supramolecular chain with the centroid-centroid distance of two rings being 3.708 Å. Two adjacent supramolec-  Figure 4c). The selected hydrogen bonds are also listed in Table 1. ported {[Cu(1)(2,2′-bpy)2]2[SiMo11V3O42]} through C-H…O hydrogen bonds is delineated in Figure 4b. Being a building block, each bi-supported POMs anion linked with eight adjacent POMs anions assembles into a 3-D supramolecular network, and the 3-D supramolecular topological structure can be obtained in 2 if the bi-supported {[Cu(1)(2,2′bpy)2]2[SiMo11V3O42]} 0.5− is perceived as nodes and hydrogen bonds as spacers (shown in Figure 4c). The selected hydrogen bonds are also listed in Table 1.

FT-IR Spectrophotometry
The FT-IR spectra of compounds 1 and 2 were recorded in the regions between 4000 and 400 cm −1 and exhibit the characteristic peaks of Keggin POMs ( Figure S3

Thermogravimetric Analyses
The thermogravimetric analyses (TGA) curves for 1 and 2 between 40 • C and 900 • C under the N 2 atmosphere are presented in Figure S6. The TG curve for 1 exhibits a twostage weight loss. The first stage lost 1.27% of weight in 40-210 • C (calcd. 1.29%), which is consistent with the vaporization of two water molecules. The next stage loss is 27.98% (calcd. 28.01%) from 338 to 675 • C, which can be attributed to the removal of five 2,2 -bpy ligands. The TG curve for 2 exhibiting one-stage weight loss of 27.67% (calcd. 28.23%) from 248 to 854 • C can be caused by the removal of five 2,2 -bpy ligands and one-half of a water molecule. The total weight loss for compound 1 is 29.25%, which is consistent with the calculated value of about 29.30%, attributed to the release of five 2,2 -bpy and two water molecules. On the other hand, the total weight loss for compound 2 is 27.67%, which is consistent with the calculated value of about 28.23%, attributed to the release of five 2,2 -bpy and a semi-water molecule.

Powder X-ray Diffraction
The powder X-ray diffractions (PXRD) were studied at an angle range of 0-50 • to analyze the purity of the sample of the two title compounds. The experimental PXRD patterns of compounds 1 and 2 are in great agreement with the simulated ones, and the intensity of the peaks are slightly different, indicating the crystal phase purity, as shown in Figure S7. The preferred orientations of crystalline samples for compounds 1 and 2 eventually led to the distinction in reflection intensity.

UV-vis Spectrophotometry
The UV-vis spectra for compounds 1 and 2 were measured in the range of 200-800 nm and are revealed in Figure S8

Fluorescence Properties
Aromatic organic ligands with conjugated structures have special fluorescence properties, so organic-inorganic hybrid materials with aromatic ligands in their structures will also have salutary fluorescence properties, which have been practical applications in numerous fields, such as chemical sensing and photoluminescence. The fluorescent properties of the free 2,2 -bpy and two title compounds were examined in the solid state at room temperature, and the emission spectra are revealed in Figure S9. The fluorescent spectrum of 2,2 -bpy displays an emission peak at 415 nm (λ ex = 366 nm). The fluorescent spectra of compounds 1 and 2 exhibit similar emission peaks at 423 nm (λ ex = 372 nm) and 421 nm (λ ex = 376 nm), respectively. It is clear that the emission bands of compounds 1 and 2 are similar to a free 2,2 -bpy ligand in terms of the position and band shape, and these bands should be assigned to the intra-ligand charge transition of 2,2 -bpy.

Catalytic Properties
Using an aqueous solution of tert-butyl hydrogen peroxide (TBHP) as a strong oxidant and compounds 1 and 2 as the catalyst, the epoxidation of styrene to styrene oxide with TBHP was carried in a batch reactor which was probed by the catalytic performances of 1 and 2. 2 mg (0.72 µmol) of finely ground compounds 1 and 2; 0.114 mL (1 µmol) of styrene and 2 mL CH 3 CN were dropped to a 10 mL double-necked flask equipped with a reflux condenser and stirrer. The mixture solution was heated to 80 • C in an oil bath and then 2 mL of TBHP was injected to start the reaction which lasted 8 h. A gas chromatograph (Shimadzu, GC-8A) with a flame detector and an HP-5 capillary column were used to quantify the organic composition of the reaction system, and the catalytic activity of compounds 1 and 2 was evaluated based on the conversion of styrene and product selectivity. The same conditions were tested for zero styrene conversion in the absence of the catalyst.
The results of the catalytic reactions for styrene oxidation by TBHP using various compounds as catalysts are presented in Figure 5 and more data information is in Table S1. It is of note that both 1 and 2 presented activity for styrene oxidation. As a catalyst, the total conversions of compounds 1 and 2 were 91.7% and 86.7% after 8 h, and the selectivity of the main target products, phenylethylene oxide: yard, was 65.9% and 89.4%, respectively. The excellent catalytic activity of compounds 1 and 2 is attributed to the novel structure including Keggin Mo-V polyoxoanions and Cu-bpy cations containing both Cu II     The recyclability and reusability of compounds 1 and 2 were also studied, including the conversion and catalyst recovery in three cycles. The results are shown in Table 2. The same experimental conditions were employed except that 10 mg of compounds 1 and 2 were used as catalysts. Because of the increase in the dose of compounds 1 and 2, the conversion increased to 93.0% for 1 and 87.3% for 2, and the selectivity increased to 72.3% for 1 and 91.1% for 2. The catalysts were recovered by filtration and washed with acetonitrile when every cycle was over. After they were dried at room temperature, the recovered compounds 1 and 2 were directly reused. The catalytic activity of compounds 1 and 2 did not exhibit a significant decrease even over three cycles, and the conversions of compounds 1 and 2 from the first to three cycles are 93.0%, 90.5% and 92.6% for 1, and 87.3%, 84.9%, 83.8% for 2, respectively. The selectivity of compounds 1 and 2 from the first to three cycles are 72.3%, 60.8% and 75.2% for 1, and 91.1%, 85.8% and 91.5% for 2, respectively. The residual catalysts of compounds 1 and 2 were recorded in the power XRD patterns and IR spectra to compare whether the structures changed. The IR spectra and the power XRD patterns of compounds 1 and 2 after three cycles are shown in Figures S10 and S11, and the main characteristic peaks of suspended solids after three runs of repeated experiments of compounds 1 and 2 have little change, which are still consistent with experimental ones. It means that the samples have good stability in catalytic experiments.

Photocatalytic Activities
Safely and effectively disposing of industrial wastewater is one of the most challenging problems in environmental governance [64][65][66][67]. Typical organic dyes such as RhB and MB were chosen to simulate organic pollutants, and the reaction process of the system was monitored by the change of absorbance intensity at the characteristic absorption wavelength of different organic dyes to judge its ability to decontaminate industrial wastewater and further evaluate the photocatalytic activity of cluster-based hybrid materials 1 and 2. The experimental procedure is as follows: (a) 5 mg of fine powder of compounds 1 and 2 were sufficiently ground in an agate mortar and dissolved in 200 mL aqueous solutions of RhB (1.0 × 10 -5 mol·L -1 ), the reaction system was adjusted to pH = 1, 3 and 10, and the catalyst was dispersed uniformly by ultrasonic shaking to form a suspension. (b) The suspension was stirred on a magnetic stirrer about 20 min in the dark environment, to achieve a adsorption-desorption balance between the catalyst and RhB aqueous solution. A total of 5 mL suspension with solids was removed by centrifugal force at 5000 rpm, and the liquid supernatant was tested by UV-vis spectroscopy and the absorbance value was measured and recorded. (c) The reaction system was irradiated with a 400 W Xe lamp, and the liquid was centrifuged at 10,000 rpm at 5 cm from the lamp. A 5 mL sample was analyzed every 20 min by UV-vis spectroscopy. As a photocatalyst of compounds 1 and 2, the UV-vis absorption curves of 1 and 2 for photocatalytic degradation RhB solution with different reaction time and pH values were presented in Figures S12a-c and S13a, and the concentration of RhB solution (C t /C 0 ) versus irradiation time was plotted in Figures S12d and S13d. After 120 min reaction, the degradation rates of the RhB solution at pH = 1, 3 and 10 were 74.1%, 85.8%, 79.5% for 1, and 93.9%, 94.0%, 83.8% for 2, respectively. Compounds 1 and 2 present better activity for the degradation of RhB at pH = 3.
The same operation method was used and the pH of the reaction system was adjusted to about 3 to test the photocatalytic performances for MB and RhB + MB aqueous solution. The UV-vis absorption curves of compounds 1 and 2 with reaction time for the photocatalytic degradation MB and MB + RhB solution were revealed in Figure 6, and changes in the concentration of MB solution (C t /C 0 ) versus irradiation time were shown in Figure S14. After a 120 min reaction, the degradation rates of MB in only one MB dye solution reached 93.9% for 1 and 95.1% for 2 (Figure 6a,b). The degradation rates of RhB and MB were 66.6% and 92.1% for 1 (Figure 6c), 90.4% and 98.0% for 2 ( Figure 6d) in a mixed solution containing MB and RhB. The results showed that compounds 1 and 2 both had a superior photocatalytic performance for the degradation of RhB, MB in MB + RhB solution. It is noteworthy that both compounds 1 and 2 showed slightly smaller catalytic degradation of RhB and MB in mixed solution than in pure solution, and the reasons for these phenomena need to be further investigated.
at 5000 rpm, and the liquid supernatant was tested by UV-vis spectroscopy and the ab-sorbance value was measured and recorded. (c) The reaction system was irradiated with a 400 W Xe lamp, and the liquid was centrifuged at 10,000 rpm at 5 cm from the lamp. A 5 mL sample was analyzed every 20 min by UV-vis spectroscopy. As a photocatalyst of compounds 1 and 2, the UV-vis absorption curves of 1 and 2 for photocatalytic degradation RhB solution with different reaction time and pH values were presented in Figures S12a-c and S13a-c, and the concentration of RhB solution (Ct/C0) versus irradiation time was plotted in Figures S12d and S13d. After 120 min reaction, the degradation rates of the RhB solution at pH = 1, 3 and 10 were 74.1%, 85.8%, 79.5% for 1, and 93.9%, 94.0%, 83.8% for 2, respectively. Compounds 1 and 2 present better activity for the degradation of RhB at pH = 3.
The same operation method was used and the pH of the reaction system was adjusted to about 3 to test the photocatalytic performances for MB and RhB + MB aqueous solution. The UV-vis absorption curves of compounds 1 and 2 with reaction time for the photocatalytic degradation MB and MB + RhB solution were revealed in Figure 6, and changes in the concentration of MB solution (Ct/C0) versus irradiation time were shown in Figure S14. After a 120 min reaction, the degradation rates of MB in only one MB dye solution reached 93.9% for 1 and 95.1% for 2 (Figure 6a,b). The degradation rates of RhB and MB were 66.6% and 92.1% for 1 (Figure 6c), 90.4% and 98.0% for 2 ( Figure 6d) in a mixed solution containing MB and RhB. The results showed that compounds 1 and 2 both had a superior photocatalytic performance for the degradation of RhB, MB in MB + RhB solution. It is noteworthy that both compounds 1 and 2 showed slightly smaller catalytic degradation of RhB and MB in mixed solution than in pure solution, and the reasons for these phenomena need to be further investigated.

Materials and Methods
All the chemical reagents are analytical pure from commercially purchased sources and have not been further purified for the experiment. The elemental analyses (C, H and N) were recorded on a Multi EA 5100 elemental analyzer (Jena, Germany) and Cu, Mo, V, P and Si elemental analyses were performed on a Perkin-Elmer Optima 7300 V spectrophotometer (Waltham, MA, USA). Fourier-transform infrared (FT-IR) spectra for the solid samples were measured (Bruker VERTEX 70v, Billerica, MA, USA) in the range of 400-4000 cm −1 by pressing KBr pellets. X-ray photoelectron spectroscopy (XPS) was conducted on single crystals with Thermo ESCALAS 250 spectrometer, using the Mg Kα (1253.6 eV) achromatic X-ray source. Thermogravimetric analyses (TGA) were recorded with a Setaram Themys HP thermal analyzer (Lyon, France) under a nitrogen flow with a temperature rate of 10 • C·min −1 . The powder X-ray diffraction (PXRD) patterns were determined on a Rigaku X-Smartlab SE X-ray diffractometer (Tokyo, Japan) using Cu-Kα radiation (λ = 1.541 Å) to study the crystalline phase of the samples. UV-vis spectra were measured using a Shimadzu UV-3100 spectrophotometer (Kyoto, Japan) with DMSO as a solvent in room atmosphere. Photoluminescence (PL) properties were conducted on a Hitachi F-7000 fluorescence analyzer (Kyoto, Japan) in the range of 450-700 nm.

X-ray Crystallography Data Collection and Refinement Study
The crystallographic data of compounds 1 and 2 were determined by a Bruker SMART CCD x-ray diffraction in ψ-ω scanning mode with graphite monochromated Mo-Kα radia-tion (λ = 0.71073 Å) at normal temperature. No sign of crystal decay during single crystal data collections. Data restoration was accomplished by the SAINT procedure, and direct methods were used to solve the crystal structure by SHELXTL-2018/3 and the full-matrix least-squares method on F 2 was used for crystal refinement and correction via SHELXTL-2018/3 crystallographic software package [72,73]. All atoms were corrected for anisotropy except Cu 2 atom, O 24 , N 5 and C 21 -C 25 in the half a [Cu 0.5 (2,2 -bpy)(H 2 O) 0.5 ] + cation in 2. The majority of hydrogen atoms were included in their geometrically calculated positions, but the hydrogen atoms of water molecular and disordered 2,2 -bpy in 1 and half 2,2 -bpy in 2 were not added. Supplementary crystal data and more detailed information for compounds 1 and 2 were stored at the Cambridge Crystal Data Center, and the CCDC numbers re 2,221,021 and 2,221,022. A summary of the crystal data and structure refinements for 1 and 2 is given in Table 3. Table 3. Crystal data and structure refinements for compounds 1 and 2.

Conclusions
In this work, two new hybrids based on different bi-capped Keggin POMs and Cu-bpy cations containing both Cu II and Cu I complexes were synthesized and characterized. The most striking feature of 1 is the (H 2 O) 2 water cluster linking 1-D chains to 2-D supramolecular layer, and 2 contains a rare half-occupancy bi-coordinated [Cu 0.5 (2,2 -bpy)] 0.5+ cation. Compounds 1 and 2 both show excellent catalytic activities for the styrene epoxidation to styrene oxide with aqueous TBHP and honorable photocatalytic degradation properties for RhB, MB and MB + RhB mixed solutions. The excellent catalytic activities of compounds 1 and 2 also prove that the modification of classical metal oxygen clusters by introducing appropriate transition metal ions is one of the effective means to obtaining high-activity cluster catalytic materials.