The first 3-fold interpenetrating framework containing both azobenzene-3,3^′-dicarboxylicate and 1,2-bis(4-pyridyl)ethylene

Abstract

The reactions of Co ^II or Ni ^II acetate with azobenzene-3,3^′-dicarboxylic acid (3,3′-H₂AZDB) and 1,2-bis(4-pyridyl)ethylene (bpe) afforded two isomorphic compounds [M ₂(3,3′-AZDB)₂(bpe)₂]_n (M=Co (1) and Ni (2)) under hydrothermal conditions. They were characterized by elemental analysis, IR spectra, thermogravimetric analysis and single-crystal X-ray diffraction technique. The structures of compounds 1 and 2 have similar 3-D 3-fold interpenetrating structures in which each 3-D net displayed a 6-connected pcu network consisting of M ²⁺-AZDB²⁻ layers and bpe pillars. Variable-temperature magnetic-susceptibility measurements revealed the occurrence of weak antiferromagnetic interactions between the Co(II) atoms in 1.

Keywords:

• Azobenzene-3,3^′-dicarboxylic acid
• 1,2-Bis(4-pyridyl)ethylene
• α-Polonium cubic network
• Magnetic property

1. Introduction

Entangled systems of coordination polymers (CPs) have been rapidly and continuously evolved not only for their intrinsic aesthetic appeal and topological features [1,2], but also for their potential structure-related applications [3,4]. Currently, many different types of entangled structures, such as interpenetrations, polyrotaxanes, polycatenanes, polyknots, polythreads, self-penetrations and other new entangled motifs, have been extensively investigated and discussed in several distinguished reviews [5–8]. Among them, interpenetrating networks can be regarded as infinite, ordered polycatenanes or polyrotaxanes, and are characterized by the presence of two or more independent networks that are inextricably entangled through the rings belonging to one framework [5,6]. A variety of interpenetrating frameworks have been constructed [9–11], however, predicting and accurately control of the interpenetrating structures still remains a considerable challenge due to the relative unpredictability of the subtle assembly process, including the nature of the organic ligand, metal ion, solvent system, pH value of the solution and the metal-to-ligand ratio, any or all of which may influence the structures of coordination frameworks [12,13]. From the viewpoint of self-assembly, an effective strategy in the design of interpenetrating networks is the use of lengthy rigid/flexible polycarboxylates and N-donor building blocks, as they can promote frame structural patterns, through which other organic linkers can penetrate.

Recent studies have further demonstrated that the entangled coordination systems assembled by mixed ligands, especially those with polycarboxylate and N-donor co-ligands, will show more diverse and interesting polymeric networks with potential properties [14–16]. Following this strategy, we have made a systematic investigation of entangled CPs with aromatic dicarboxylates and N,N′-donor ancillary ligands and different metal ions [17–22]. When azobenzene-3,3^′-dicarboxylic acid (3,3′-H₂AZDB) was used, in the presence of 1,2-bis(4-pyridyl)ethylene (bpe), to react with transition metal ions Co(II), Ni(II) ions, two new compounds have been obtained, [Co ₂(3,3′-AZDB)₂(bpe)₂]_n(1) and [Ni ₂(3,3′-AZDB)₂(bpe)₂]_n (2). Both compounds showed a 3-fold interpenetrating pcu network in the crystal structures.

2. Experimental section

2.1 Materials and general methods

All solvents and reagents for synthesis were commercially available and were used as received. All the products were highly stable in air at ambient conditions. Elemental analyses of C, H and N were performed on a Perkin-Elmer 2400 CHN elemental analyzer. Infrared spectra were recorded using KBr pellets on a Thermo Electron NEXUS IR spectrophotometer in the 400–4000 cm⁻¹ region. Thermogravimetric analyses (TGAs) were performed on a Netzsch STA 449C microanalyzer under air atmosphere at a heating rate of 10°C min⁻¹. Variable-temperature magnetic measurements were carried out on a Quantum Design SQUID MPMS XL-7 instrument (2∼ 300 K) in the magnetic field of 1 KOe, and the diamagnetic corrections were evaluated by using Pascal's constants.

2.2 Synthesis of compound [Co ₂(3,3′-AZDB)₂(bpe)₂]_n (1)

A mixture of Co(Ac)O (0.5 mmol, 0.124 g), H ₂AZDB (0.5 mmol, 0.135 g) and bpe (0.5 mmol, 0.091 g) in 8 mL H ₂O was sealed in a 25-mL Teflon reactor and heated at 160°C for 96 h, and then cooled to room temperature at a rate of 4°C h⁻¹. Pink prism crystals of 1 were collected and washed with distilled water and dried in air. Yield, 47% (based on Co(II)). Anal. Calcd for : C, 61.31%; H, 3.56%; N, 10.99%; Found: C, 61.03%; H, 3.50%; N, 10.89%. IR (cm⁻¹, KBr): 3417.61(s), 2825(s), 1602(w), 1558(w), 1309(vs), 930(s), 769(w), 680(vs).

2.3 Synthesis of complex [Ni₂(3,3′-AZDB)₂(bpe)₂]_n (2)

2 was synthesized in the similar way as that described for 1, except that Co(Ac)₂·6H₂O was replaced by Ni(Ac)₂.4H₂O (0.5 mmol, 0.124 g). Green prism crystals of 2 were collected and washed with distilled water and dried in air to give the product. Yield, 43% (based on Ni(II)). Anal. Calcd for : C, 61.33%; H, 3.56%; N, 11.00%; Found: C, 61.23%; H, 3.60%; N, 10.92%. IR (cm⁻¹, KBr): 3402.61(s), 2718(s), 1606(w), 1549(w), 1304(vs), 934(s), 761(w), 667(vs).

2.4 Crystal structures determination

Single-crystal X-ray diffraction measurements for 1 and 2 were carried out on a Rigaku XtaLAB mini diffractometer equipped with graphite-monochromatized MoK α radiation (). Semi-empirical absorption corrections were applied using SADABS [23]. All structures were solved by direct methods using the SHELXS program of the SHELXTL-97 package [24] and refined with SHELXL-97 [25]. Metal ions were found from E-maps, and other non-hydrogen atoms were located by successive difference Fourier syntheses. The final refinement was performed by full matrix least-squares methods with anisotropic thermal parameters for non-hydrogen atoms on F². The hydrogen atoms of organic ligands were located theoretically and refined with fixed thermal factors. Crystallographic data and experimental details for the structural analyses are summarized in Table 1 and selected bond lengths and angles are listed in Table 2.

Table 1.  Crystal data and structure refinement information for 1 and 2.

Compound	1	2
Empirical formula	O4Co	O4Ni
Formula weight	509.37	509.15
Crystal system, Space group	Triclinic,	Triclinic,
a (Ǻ)	9.101(4)	9.207(2)
b (Ǻ)	11.642(6)	11.559(3)
c (Ǻ)	11.677(6)	11.623(3)
	98.378(6)	99.538(5)
Volume (Ǻ^3)	1175.7(10)	1165.9(5)
Z	2	2
T (K)	296(2)	296(2)
Calculated density(g cm^−3)	1.439	1.450
F(000)	522	524
θ range for data collection (°)	2.33 to 25.5	1.81 to 25
Index range	,	,
	,	,
Reflections collected/unique	8612/4256 []	5677/3905 []
Data/restraints/ parameters	4256 / 0 / 316	3905 / 180 / 316
Goodness-of-fit	1.077	1.048
R1, wR2 [I >2σ(I)]	0.0893, 0.2383	0.0962, 0.2319
R1, wR2 (all data)	0.1023, 0.2473	0.1480, 0.2648

Note: ; wR)²]]^1/2.

Table 2.  Selected bond lengths (Ǻ) and bond angles (deg) for 1 and 2.

1
Co(1)–O(2)#1	2.028(5)	Co(1)–O(4)	2.191(5)
Co(1) –O(1)	2.045(4)	Co(1)–O(3)	2.216(6)
Co(1)–N(3)	2.152(5)	Co(1)–N(4)#2	2.185(5)
O(2)#1–Co(1)–O(1)	119.2(2)	N(3)–Co(1)–O(4)	94.0(2)
O(2)#1–Co(1)–N(3)	87.6(2)	O(1)–Co(1)–O(3)	92.89(19)
O(1)–Co(1)–N(3)	88.6(2)	O(4)–Co(1)–O(3)	59.44(19)
O(1)–Co(1)–O(4)	151.7(2)	C(7)–O(1)–Co(1)	130.3(4)
O(2)#1–Co(1)–N(4)#2	90.7(2)	O(1)–Co(1)–N(4)#2	93.0(2)
N(3)–Co(1)–N(4)#2	178.1(2)	O(2)#1–Co(1)–O(4)	141.51(6)
O(2)#1–Co(1)–O(3)	146.7(2)	N(3)–Co(1)–O(3)	84.1(2)
N(4)#2–Co(1)–O(3)	96.8(2)
2
Ni(1)–O(3)	2.009(6)	Ni(1)–O(4)#1	2.023(6)
Ni(1)–N(1)	2.124(6)	Ni(1)–O(2)	2.135(6)
Ni(1)–N(2)#2	2.128(7)	Ni(1)–O(1)	2.199(6)
O(3)–Ni(1)–O(4)#1	113.8(3)	O(3)–Ni(1)–N(1)	87.3(3)
O(4)#1–Ni(1)–N(1)	87.1(3)	O(3)–Ni(1)–O(2)	90.5(2)
O(4)#1–Ni(1)–O(2)	155.5(3)	N(1)–Ni(1)–O(2)	91.3(2)
O(3)–Ni(1)–N(2)#2	91.9(3)	O(4)#1–Ni(1)–N(2)#2	94.1(3)
N(1)–Ni(1)–N(2)#2	178.7(3)	O(2)–Ni(1)–N(2)#2	87.5(3)
O(3)–Ni(1)–O(1)	150.4(2)	O(4)#1–Ni(1)–O(1)	94.6(2)
N(1)–Ni(1)–O(1)	85.5(3)	O(2)–Ni(1)–O(1)	60.9(2)
N(2)#2–Ni(1)–O(1)	94.3(3)

Note: Symmetry operations: ; for ; ; for 2.

3. Results and discussions

3.1 Crystal structure

Complexes 1 and 2 crystallize in triclinic space group Pī, and the crystal structure analysis reveals that they have the same structure. Thus only the results of 1 are given in the ensuing discussion. As shown in Figure 1, the asymmetric unit contains a Co(II) center, a 3,3′-AZDB²⁻ anion and one bpe ligand. Each Co(II) center is six-coordinated with a distorted octahedron coordination geometry, by four oxygen atoms (O1–Co˭, O3, O4, O2#1; , , , , symmetry code: ) from three individual 3,3′-AZDB²⁻ in the equatorial plane and two N atoms (Co1–, ) from two bpe in the axial positions. And these bond distances and the bond angles around Co(II) atom [59.44(2)−178.1(2)^o] are falling into the normal range [16,18,20].

Figure 1. Coordination environment of the Co(II) atom in complex 1.The hydrogen atoms are omitted for clarity. (Symmetry codes: # 1,−x+1,−y+1,−z+2; # 2, x+1, y, z+1; # 3,−x+1,−y,−−z+1; # 4,−x+2,−y+2,−z+1).

In 1, the 3,3′-H₂AZDB ligand is completely deprotonated and adopts two different coordination modes, namely, μ₄-η¹:η¹:η¹:η¹ (two μ₂-η¹:η¹-C7O1O2) and μ₂-η¹:η¹:η¹:η¹(two μ₁-η¹:η¹-C15O3O4). In such linking mode, adjacent Co(II) atoms are connected by 3,3′-AZDB²⁻ anions to form 2-D layers based on dinuclear [Co ₂(COO)₄] units (Co Co separation is 4.02 Å for 1 and NiNi separation is 4.12 Å for 2) (Figure 2). These 2-D layers are further pillared by rigid bpe ligands to generate a 3-D network with 1-D channel of for 1 and for 2 (Figure 3). Topologically, if we consider the dinuclear [Co ₂(COO)₄] units as the secondary building units (SBUs), each dinuclear unit is linked by six other SBUs through four 3,3′-AZDB²⁻ and two bpe ligands. Thus, the whole 3-D network can be viewed as a uniform 6-connected pcu network (Figure 4). In order to minimize the big void cavities and stabilize the framework, the potential voids formed by a single 3-D network show incorporation of other two identical ones, thus affording a 3-fold parallel interpenetrating pcu nets (Figure 4). An analysis of the interpenetration fashion with the TOPOS program [26] reveals that it belongs to class Ia with the translating interpenetration vector being equal to the a axis. To date, many poly-fold parallel interpenetrating pcu frameworks have been reported [27,28], however, 3-fold parallel interpenetrating pcu nets based on isomeric H ₂ AZDB have rarely been observed.

Figure 3. View of the 3-D network of 1.

Figure 4. The 3-fold interpenetrating pcu network of 1.

Figure 2. View of the 2-D layer of Co(II) atoms bridged by AZDB²⁻ linkers with two different coordination modes.

3.2 Thermogravimetric analysis

To examine the thermal stability of complexes 1 and 2, TGAs were carried out between 20°C and 800°C. The TGA curves of the two compounds are shown in Figure 5. The samples were heated up in a static air atmosphere with a heating rate of 10°C min⁻¹. The TGA curve indicated that 1 lost weight gradually from 130°C to 409°C, then the sample suffers incessant weight loss until it reaches a temperature of 464°C. The total mass loss of 85.5% corresponds to the removal of the organic species (calc. 85.3%). While for complex 2, the sample was stable up to ca.110°C. The gradual weight loss in the temperature ranges 110–407°C, then a rapid weight loss was detected until 480°C, which is attributed to the decomposition of complex 2. The total mass loss of 85.9% corresponds to the removal of the organic species (calc. 85.33%).

Figure 5. The TGA curves of 1 and 2.

3.3 Magnetic properties

The variable-temperature magnetic susceptibilities of 1 were measured in the 2∼ 300 K region at 2000 G. The plots of χ_MT and χ_M versus T are shown in Figure 6. The χ_MT value at room temperature is 5.15 cm³ mol⁻¹ K, which is greater and larger than two isolated spin-only Co ²⁺ cations (3.75 cm³ mol^.−1 K). This larger value is the result of contributions to the susceptibility from orbital angular momentum at high temperature [29]. As the temperature is lowered to 2 K, the χ_MT value continuously decrease, which suggests that antiferromagnetic interactions are operative in 1. In the whole temperature region, a typical paramagnetic Curie–Weiss behavior is observed, with a Weiss constant  K, Curie constant C=5.25 cm³ K mol⁻¹ and , respectively. The negative θ value and the decrease of the χ_MT are all indicative of antiferromagnetic interactions in 1.

Figure 6. Temperature dependence of χ_MT and χ_M vs. T for 1. Open circles are the experimental data, and the solid line represents the best fit.

To quantitatively evaluate the magnetic interactions in 1, for similar binuclear Co(II) complexes, the following equation is induced by an isotropic dimmer mode of spin S=3/2 [30].

The least-squares analysis of magnetic susceptibility data led to J=−0.98 cm⁻¹, g=2.36 and (the agreement factor ).

4. Conclusions

In summary, two new CPs, [M ₂(3,3′-AZDB)₂(bpe)₂]_n ( (1) and Ni (2)), have been successfully synthesized under hydrothermal conditions. Structure showed a 3-fold interpenetrated pcu network. In addition, compound 1 displayed weakly antiferromagnetic interactions. This observation prompts us to further develop the rational synthetic strategy to obtain new entangled CPs with azobenzene-dicarboxylicate and the N-donor mixed-ligands system.

Supporting material

Crystallographic data for the structural analysis have been deposited with the Cambridge Crystallographic Data Centre (CCDC) No.941169-941170 for compound 1 and 2. Copies of this information may be obtained free of charge on application to CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax: [phone]; E-mail: [email protected] or http://www.ccdc.cam.ac.uk).

Funding

This work was financially supported by the NSF of China [grant number 21373122], [grant number 21301106]; the NSF of Hubei Province of China [grant number 2011CDA118]; the Project of Hubei Provincial Education Office [grant number Q20131304].

Acknowledgements

JYL acknowledges the support from the Welch Foundation.