Exploring 5-fluoronicotinic acid as a versatile building block for the generation of topologically diverse metal–organic and supramolecular Ni, Co, and Cd networks

Abstract

Five new coordination compounds, namely [Ni(5-Fnic)₂(μ₂-H₂O)_0.5]_n (1), [Co(5-Fnic)₂(H₂biim)]_n (2), {[Cd(5-Fnic)₂(phen)]·2H₂O}_n (3), [Cd(5-Fnic)₂(H₂biim)₂] (4), and [Co(5-Fnic)₂(H₂O)₄] (5), were generated by hydrothermal methods from the metal(II) nitrates, 5-fluoronicotinic acid (5-FnicH), and an optional ancillary 1,10-phenanthroline (phen) or 2,2′-biimidazole (H₂biim) ligand. All the products 1–5 were characterized by IR spectroscopy, elemental, thermogravimetric, and single-crystal X-ray diffraction analyses. Their structures range from the intricate 3D metal–organic framework (MOF) 1 to the 1D coordination polymers 2 and 3, and the discrete 0D monomers 4 and 5. The structures of 2–5 are further extended [1D → 2D (2), 1D → 1D (3), 0D → 1D (4), and 0D → 3D (5)] into various H-bonded networks. The topological analysis of the underlying coordination (in 1) and H-bonded (in 2–5) nets revealed a very rare 3,6T10 MOF topology in 1, a parallel 2D + 2D interpenetration of the sql nets in 2, a topologically unique double chain in 3, a simple 2C1 topology in 4, and a pcu (alpha-Po primitive cubic) topology in 5. The magnetic (for 1 and 2) and luminescent (for 3 and 4) properties were also investigated and discussed.

---

1. Introduction

The design and synthesis of coordination polymers, metal–organic frameworks and discrete complexes have been of considerable interest owing to the diversity of intriguing coordination and supramolecular architectures and topological features, as well as various applications in different areas.^1–6 Aromatic carboxylate ligands are recognized building blocks in the construction of compounds with high-dimensional networks and interesting properties, not only due to their ability of adapting diverse coordination modes and showing high stability, but also because of potential function as H-bond donors and acceptors.^7–13 In many cases, the synthesis of coordination compounds is based on self-assembly methods that are highly influenced by a number of factors, including the nature of metal ions and ligands, type of reaction medium, presence of templates and ancillary ligands, stoichiometry and various reaction conditions.^8,9,11 Apart from carboxylate ligands, 1,10-phenanthroline (phen) and 2,2′-biimidazole (H₂biim) have often been used as secondary N,N-donor building blocks to construct and stabilize new structures, on account of their effective π⋯π stacking and/or weak H-bonding interactions.^8,9,12,14

As a continuation of our research in this field,^8,9,12 we have tested the hydrothermal self-assembly reactions of various metal(II) nitrates (M = Cd, Co, Ni) with 5-fluoronicotinic acid (5-FnicH) as a main building block and 1,10-phenanthroline or 2,2′-biimidazole as N,N-donor ancillary ligands in view of the following considerations. (A) 5-Fluoronicotinic acid remains very scarcely explored in the crystal engineering of coordination polymers, as attested by a search of Cambridge Structural Database (CSD) that revealed only four structurally characterized coordination compounds derived from this acid.¹⁵ (B) 5-FnicH possesses one carboxylic group and one pyridyl N atom available for coordination to a metal center, while the presence of tethered fluoride functionality affects its acidity in comparison with unsubstituted nicotinic acid, thus potentially allowing the formation of structurally different coordination compounds. (C) The introduction of phen and H₂biim ligands may facilitate the crystallization of compounds and stabilization of their structures.

Hence, by employing 5-FnicH as a main building block and introducing an optional ancillary N,N-donor ligand, we have produced by hydrothermal self-assembly method a series of novel coordination compounds, namely a 3D metal–organic framework [Ni(5-Fnic)₂(μ₂-H₂O)_0.5]_n (1), 1D coordination polymers [Co(5-Fnic)₂(H₂biim)]_n (2) and {[Cd(5-Fnic)₂(phen)]·2H₂O}_n (3), and discrete 0D complexes [Cd(5-Fnic)₂(H₂biim)₂] (4) and [Co(5-Fnic)₂(H₂O)₄] (5). Their diversity indicates that the type of metal ion and ancillary ligand have an important effect on the formation, structural and topological features of the final products 1–5. All the compounds were characterized by IR spectroscopy, elemental, thermogravimetric, and single-crystal X-ray diffraction analyses. The magnetic behavior of 1 and 2 and the luminescent properties of 3 and 4 were also studied.

2. Experimental

2.1 Materials and methods

All chemicals and solvents were of A.R. grade and used without further purification. Carbon, hydrogen, and nitrogen content in 1–5 was determined using an Elementar Vario EL elemental analyzer. IR spectra were recorded in KBr pellets and a Bruker EQUINOX 55 spectrometer. Thermogravimetric analysis (TGA) was performed under N₂ atmosphere with a heating rate of 10 °C min⁻¹ on a LINSEIS STA PT1600 thermal analyzer. Magnetic susceptibility data were collected in the 2–300 K temperature range with a Quantum Design SQUID Magnetometer MPMS XL-7 with a field of 0.1 T. A correction was made for the diamagnetic contribution prior to data analysis. Excitation and emission spectra were recorded for the solid samples on an Edinburgh FLS920 fluorescence spectrometer at room temperature.

2.2 Synthesis of [Ni(5-Fnic)₂(μ₂-H₂O)_0.5]_n (1)

A mixture of Ni(NO₃)₂·6H₂O (43.6 mg, 0.15 mmol), 5-FnicH (42.3 mg, 0.3 mmol), NaOH (12.0 mg, 0.3 mmol), and H₂O (10 mL) was stirred at room temperature for 15 min, then sealed in a 25 mL Teflon-lined stainless steel vessel, and heated at 160 °C for 3 days, followed by cooling to room temperature at a rate of 10 °C h⁻¹. Green needle-shaped crystals were isolated manually, washed with distilled water and dried to give 1. Yield: 65% (based on 5-FnicH). Calcd for C₁₂H₇F₂NiN₂O_4.5: C 41.43, H 2.03, N 8.05%. Found: C 41.78, H 2.13, N 8.41%. IR (KBr, cm⁻¹): 3286m, 3067w, 1618m, 1593s, 1457w, 1388s, 1300m, 1256m, 1161w, 1141w, 1029w, 954w, 922w, 800s, 691m, 566w, 458w.

2.3 Synthesis of [Co(5-Fnic)₂(H₂biim)]_n (2)

A mixture of Co(NO₃)₂·6H₂O (43.6 mg, 0.15 mmol), 5-FnicH (42.3 mg, 0.3 mmol), H₂biim (20.1 mg, 0.15 mmol), NaOH (12.0 mg, 0.3 mmol), and H₂O (10 mL) was stirred at room temperature for 15 min, then sealed in a 25 mL Teflon-lined stainless steel vessel, and heated at 160 °C for 3 days, followed by cooling to room temperature at a rate of 10 °C h⁻¹. Pink needle-shaped crystals were isolated manually, washed with distilled water and dried to furnish 2. Yield: 60% (based on 5-FnicH). Calcd for C₁₈H₁₂F₂CoN₆O₄: C 45.68, H 2.56, N 17.76%. Found: C 45.29, H 2.65, N 17.37%. IR (KBr, cm⁻¹): 1615s, 1573s, 1461w, 1402s, 1382vs, 1334w, 1287w, 1245w, 1153w, 1125m, 1030w, 990m, 959w, 946w, 924w, 907w, 854w, 819m, 796m, 783m, 768w, 752m, 694m, 574w, 429m.

2.4 Synthesis of {[Cd(5-Fnic)₂(phen)]·2H₂O}_n (3)

A mixture of Cd(NO₃)₂·4H₂O (46.2 mg, 0.15 mmol), 5-FnicH (42.3 mg, 0.3 mmol), phen (30.0 mg, 0.15 mmol), NaOH (12.0 mg, 0.3 mmol), and H₂O (10 mL) was stirred at room temperature for 15 min, then sealed in a 25 mL Teflon-lined stainless steel vessel, and heated at 160 °C for 3 days, followed by cooling to room temperature at a rate of 10 °C h⁻¹. Colorless block-shaped crystals were isolated manually, washed with distilled water and dried to produce 3. Yield: 60% (based on 5-FnicH). Calcd for C₂₄H₁₈F₂CdN₄O₆: C 47.35, H 2.98, N 9.20%. Found: C 47.72, H 3.12, N 8.89%. IR (KBr, cm⁻¹): 3437m, 3055m, 1611s, 1569s, 1516m, 1459m, 1427w, 1399 s, 1293m, 1240m, 1155m, 1139w, 1102w, 1026w, 948w, 916w, 849m, 790s, 727m, 693m, 638w, 565w, 425w.

2.5 Synthesis of [Cd(5-Fnic)₂(H₂biim)₂] (4)

A mixture of Cd(NO₃)₂·4H₂O (46.2 mg, 0.15 mmol), 5-FnicH (42.3 mg, 0.3 mmol), H₂biim (40.2 mg, 0.30 mmol), NaOH (12.0 mg, 0.3 mmol), and H₂O (10 mL) was stirred at room temperature for 15 min, then sealed in a 25 mL Teflon-lined stainless steel vessel, and heated at 160 °C for 3 days, followed by cooling to room temperature at a rate of 10 °C h⁻¹. Yellow block-shaped crystals were isolated manually, washed with distilled water and dried to give 4. Yield: 55% (based on 5-FnicH). Calcd for C₂₄H₁₈F₂CdN₁₀O₄: C 43.62, H 2.74, N 21.19%. Found: C 43.26, H 2.86, N 20.83. IR (KBr, cm⁻¹): 1593s, 1562s, 1455m, 1425m, 1373vs, 1331w, 1240m, 1181w, 1148w, 1133w, 1122w, 1099w, 1025w, 995m, 946m, 892m, 859w, 782s, 760m, 691s, 661w, 580w, 529w, 491w, 431w, 405w.

2.6 Synthesis of [Co(5-Fnic)₂(H₂O)₄] (5)

A mixture of Co(NO₃)₂·6H₂O (43.6 mg, 0.15 mmol), 5-FnicH (42.3 mg, 0.3 mmol), NaOH (12.0 mg, 0.3 mmol), and H₂O (10 mL) was stirred at room temperature for 15 min, then sealed in a 25 mL Teflon-lined stainless steel vessel, and heated at 160 °C for 3 days, followed by cooling to room temperature at a rate of 10 °C h⁻¹. Pink needle-shaped crystals were isolated manually, washed with distilled water and dried to produce 5. Yield: 60% (based on 5-FnicH). Calcd for C₁₂H₁₄F₂CoN₂O₈: C 35.05, H 3.43, N 6.81%. Found: C 34.71, H 3.58, N 7.13%. IR (KBr, cm⁻¹): 3285s, 1617s, 1593s, 1457m, 1386s, 1300s, 1255m, 1159w, 1139w, 1028w, 952w, 921m, 799m, 690w, 564w, 458w.

3. X-ray crystallography

The single-crystal X-ray data collection for 1–5 was performed on a Bruker Smart CCD diffractometer, using graphite-monochromated Mo K_α radiation (λ = 0.71073 Å). Semiempirical absorption corrections were applied using the SADABS program. The structures were solved by direct methods and refined by full-matrix least-squares on F² with the SHELXS-97 and SHELXL-97 programs.¹⁶ All the non-hydrogen atoms were refined anisotropically by full-matrix least-squares methods on F². All the hydrogen atoms (except those bound to water molecules) were placed in calculated positions with fixed isotropic thermal parameters and included in structure factor calculations at the final stage of full-matrix least-squares refinement. The hydrogen atoms of water molecules were located by difference maps and constrained to ride on their parent O atoms. The crystal data for 1–5 are summarized in Table 1 and selected bond lengths are listed in Table 2. Hydrogen bonds in the compounds 2–5 are given in Table 3. The topological analysis of the H-bonded networks in 2–5 was carried out taking into consideration only the conventional (strong) hydrogen bonds D–H⋯A, wherein H⋯A < 2.50 Å, D⋯A < 3.50 Å, ∠(D–H⋯A) > 120°, D and A stand for donor and acceptor atoms.¹⁷

Table 1 Crystallographic data for 1–5

Complex	1	2	3	4	5
Chemical formula	C12H7F2NiN2O4.5	C18H12F2CoN6O4	C24H18F2CdN4O6	C24H18F2CdN10O4	C12H14F2CoN2O8
Formula weight	347.91	473.27	608.82	660.88	411.18
Crystal system	Tetragonal	Orthorhombic	Monoclinic	Triclinic	Monoclinic
Space group	P4/ncc	Pbcn	P21/n	P1̄	I2/m
a/Å	20.5004(7)	7.8197(4)	14.6749(10)	7.3743(4)	8.4112(5)
b/Å	20.5004(7)	17.8112(9)	8.0043(4)	8.4333(5)	6.8339(5)
c/Å	13.9201(6)	26.5942(14)	21.4102(17)	10.3961(5)	13.2136(8)
α/°	90	90	90	92.632(2)	90
β/°	90	90	107.405(7)	94.015(2)	96.514(6)
γ/°	90	90	90	94.544(2)	90
V/Å^3	5850.2(4)	3704.0(3)	2399.8(3)	642.05(6)	754.63(9)
T/K	293(2)	296(2)	293(2)	293(2)	293(2)
Z	16	8	4	1	2
Dc/g cm^−3	1.580	1.697	1.685	1.709	1.810
μ/mm^−1	1.368	0.988	0.974	0.918	1.208
F(000)	2800	1912	1216	330	418
Refl. measured	14382	17808	9817	3262	1376
Unique refl. (Rint)	2611 (0.0581)	3307 (0.0329)	4250(0.0682)	2243 (0.0186)	722 (0.0409)
GOF on F^2	1.041	1.025	1.058	1.062	1.002
R1 [I > 2σ(I)]^a	0.0352	0.0278	0.0613	0.0240	0.0379
wR2 [I > 2σ(I)]^b	0.0891	0.0649	0.1333	0.0622	0.0962

---

Table 2 Selected bond lengths [Å] and angles [°] for 1–5^a

a Symmetry transformations used to generate equivalent atoms: #1 y, x, z − 1/2; #2 −x + 1, −y + 1, −z + 1; #3 −y + 1, x + 1/2, −z + 1; #4 y, x, z+1/2; #5 −y + 1, −x + 1, −z + 1/2 for 1; #1 x – 1, y, z for 2; #1 x, y + 1, z for 3; #1 −x + 21, −y + 1, −z for 4; #1 −x + 1, −y + 1, −z + 1; #2 x, −y + 1, z; #3 −x + 1, y, −z + 1 for 5.
1
Ni(1)–O(1)#2	2.075(2)	Ni(1)–O(2)#1	2.0429(19)	Ni(1)–O(4)#3	2.0800(19)
Ni(1)–O(5)	2.1001(15)	Ni(1)–N(1)	2.089(2)	Ni(1)–N(2)	2.109(2)
O(2)#1–Ni(1)–O(1)#2	95.54(7)	O(2)#1–Ni(1)–O(4)#3	178.05(8)	O(1)#2–Ni(1)–O(4)#3	82.53(8)
O(2)#1–Ni(1)–N(1)	90.31(8)	O(1)#2–Ni(1)–N(1)	85.61(8)	O(4)#3–Ni(1)–N(1)	89.81(8)
O(2)#1–Ni(1)–O(5)	88.65(6)	O(1)#2–Ni(1)–O(5)	91.36(7)	O(4)#3–Ni(1)–O(5)	91.13(6)
N(1)–Ni(1)–O(5)	176.69(9)	O(2)#1–Ni(1)–N(2)	90.89(8)	O(1)#2–Ni(1)–N(2)	173.20(8)
O(4)#3–Ni(1)–N(2)	91.05(9)	N(1)–Ni(1)–N(2)	92.16(9)	O(5)–Ni(1)–N(2)	90.99(9)
Ni(1)–O(5)–Ni(1)#5	112.81(12)
2
Co(1)–O(1)	2.1740(14)	Co(1)–O(2)	2.1907(14)	Co(1)–N(1)#1	2.1620(16)
Co(1)–N(2)	2.1363(17)	Co(1)–N(3)	2.0863(16)	Co(1)–N(5)	2.1271(18)
N(3)–Co(1)–N(5)	79.79(7)	N(3)–Co(1)–N(2)	90.92(6)	N(5)–Co(1)–N(2)	170.58(6)
N(3)–Co(1)–N(1)#1	97.22(6)	N(5)–Co(1)–N(1)#1	88.36(6)	N(2)–Co(1)–N(1)#1	94.33(6)
N(3)–Co(1)–O(1)	110.72(6)	N(5)–Co(1)–O(1)	95.64(6)	N(2)–Co(1)–O(1)	86.20(6)
N(1)#1–Co(1)–O(1)	152.05(6)	N(3)–Co(1)–O(2)	168.56(6)	N(5)–Co(1)–O(2)	93.76(6)
N(2)–Co(1)–O(2)	95.16(6)	N(1)#1–Co(1)–O(2)	91.99(5)	O(1)–Co(1)–O(2)	60.19(5)
3
Cd(1)–O(1)	2.335(5)	Cd(1)–O(2)	2.437(5)	Cd(1)–O(3)	2.243(5)
Cd(1)–N(1)#1	2.364(5)	Cd(1)–N(3)	2.417(6)	Cd(1)–N(4)	2.313(7)
O(3)–Cd(1)–N(4)	111.4(3)	O(3)–Cd(1)–O(1)	99.2(2)	N(4)–Cd(1)–O(1)	118.2(2)
O(3)–Cd(1)–N(1)#1	81.3(2)	N(4)–Cd(1)–N(1)#1	96.1(2)	O(1)–Cd(1)–N(1)#1	142.09(19)
O(3)–Cd(1)–N(3)	169.2(2)	N(4)–Cd(1)–N(3)	70.1(3)	O(1)–Cd(1)–N(3)	89.01(19)
N(1)#1–Cd(1)–N(3)	88.0(2)	O(3)–Cd(1)–O(2)	95.0(2)	N(4)–Cd(1)–O(2)	153.6(2)
O(1)–Cd(1)–O(2)	54.89(17)	N(1)#1–Cd(1)–O(2)	87.22(18)	N(3)–Cd(1)–O(2)	83.9(2)
4
Cd(1)–O(1)	2.4503(16)	Cd(1)–O(1)#1	2.4503(16)	Cd(1)–N(2)	2.3031(18)
Cd(1)–N(2)#1	2.3031(18)	Cd(1)–N(4)	2.2915(18)	Cd(1)–N(4)#1	2.2915(18)
N(4)–Cd(1)–N(4)#1	180.0	N(4)–Cd(1)–N(2)	75.02(6)	N(4)#1–Cd(1)–N(2)	104.98(6)
N(4)–Cd(1)–N(2)#1	104.98(6)	N(4)#1–Cd(1)–N(2)#1	75.02(6)	N(2)–Cd(1)–N(2)#1	180.0
N(4)–Cd(1)–O(1)#1	89.29(6)	N(4)#1–Cd(1)–O(1)#1	90.71(6)	N(2)–Cd(1)–O(1)#1	91.44(6)
N(2)#1–Cd(1)–O(1)#1	88.56(6)	N(4)–Cd(1)–O(1)	90.71(6)	N(4)#1–Cd(1)–O(1)	89.29(6)
N(2)–Cd(1)–O(1)	88.56(6)	N(2)#1–Cd(1)–O(1)	91.44(6)	O(1)#1–Cd(1)–O(1)	180.0
5
Co(1)–O(3)	2.102(2)	Co(1)–O(3)#1	2.102(2)	Co(1)–O(3)#2	2.102(2)
Co(1)–O(3)#3	2.102(2)	Co(1)–N(1)	2.157(3)	Co(1)–N(1)#1	2.157(3)
O(3)–Co(1)–O(3)#1	180.0	O(3)–Co(1)–O(3)#2	90.86(14)	O(3)#1–Co(1)–O(3)#2	89.14(14)
O(3)–Co(1)–O(3)#3	89.14(14)	O(3)#1–Co(1)–O(3)#3	90.86(14)	O(3)#2–Co(1)–O(3)#3	180.00(11)
O(3)–Co(1)–N(1)#1	91.18(8)	O(3)#1–Co(1)–N(1)#1	88.82(8)	O(3)#2–Co(1)–N(1)#1	91.18(8)
O(3)#3–Co(1)–N(1)#1	88.82(8)	O(3)–Co(1)–N(1)	88.82(8)	O(3)#1–Co(1)–N(1)	91.18(8)
O(3)#2–Co(1)–N(1)	88.82(8)	O(3)#3–Co(1)–N(1)	91.18(8)	N(1)#1–Co(1)–N(1)	180.0

---

Table 3 Conventional hydrogen bonds in crystal packing [Å, °] of 2–5

Complexes	D–H⋯A	d(D–H)	d(H⋯A)	d(D⋯A)	∠DHA	Symmetry code
2	N(6)–H(2)⋯O(4)	0.97	1.71	2.673	176	x − 1/2, y + 1/2, −z + 1/2
	N(4)–H(1)⋯O(3)	0.87	1.75	2.623	173	x − 1/2, y + 1/2, −z + 1/2
3	O(6)–H(3W) ⋯O(4)	0.85	1.98	2.84	178.0	−x + /2, y − 1/2, −z + 1/2
	O(5)–H(2W)⋯O(6)	0.85	1.75	2.60	177.8
	O(5)–H(1W)⋯O(4)	0.99	2.19	3.04	142.9
4	N(3)–H(1)⋯O(2)	0.92	1.83	2.743	170	x − 1, y, z
	N(5)–H(2)⋯O(1)	0.85	1.87	2.709	173	x − 1, y, z
5	O(3)–H(1W)⋯O(1)	0.82	1.94	2.742	164.8	x − 1/2, y + 1/2, z + 1/2
	O(3)–H(2W)⋯O(2)	0.79	1.99	2.728	155	−x + 2, −y + 1, −z + 1

---

4. Results and discussion

4.1 Crystal structures

4.1.1 [Ni(5-Fnic)₂(μ₂-H₂O)_0.5]_n (1). The crystal structure of 1 reveals that the compound crystallizes in the tetragonal space group P4/ncc and features an intricate 3D metal–organic framework. The asymmetric unit of 1 contains one crystallographically unique Ni(II) atom, two distinct μ₃-and μ₂-5-Fnic⁻ ligands, and a half μ₂-H₂O moiety. As depicted in Fig. 1a, each six-coordinate Ni(II) atom adopts a distorted octahedral {ZnN₂O₄} geometry formed by three carboxylate O atoms of three 5-Fnic⁻ ligands, two N atoms of other two 5-Fnic⁻ moieties, and one O atom of μ₂-H₂O group. The Ni–O bonds range from 2.043(2) to 2.100(2) Å, whereas the Ni–N distances vary from 2.089(2) to 2.109(2) Å, which are comparable to those found in other reported Ni(II) compounds.^9,18 In 1, the 5-Fnic⁻ moieties adopt two different coordination modes (Scheme 1, modes I and II), in which the carboxylate groups are either the η¹:η⁰ monodentate or μ₂-η¹:η¹ bidentate. Two Ni(II) centers are bridged by carboxylate groups from two μ₃-5-Fnic⁻ blocks and one μ₂-H₂O ligand, generating a dinickel(II) unit with the [Ni₂(μ-COO)₂(H₂O)] core (Fig. 1b) and Ni⋯Ni separation of 3.499(3) Å. Further linkage of these Ni₂ units via both the μ₂-and μ₃-5-Fnic⁻ moieties furnishes a complex 3D metal–organic framework (Fig. 1c). To better understand the structure of this 3D MOF, we carried out its topological analysis using the concept of the simplified underlying net.¹⁷ Such a net was generated (Fig. 1d) upon reducing the μ₂- and μ₃-5-Fnic⁻ and μ₂-H₂O moieties to their centroids, thus resulting in a 3D framework assembled from the 6-connected Ni and 3-connected μ₃-5-Fnic⁻ nodes, as well as the 2-connected μ₂-5-Fnic⁻ and μ₂-H₂O linkers. Topological analysis of this framework disclosed a binodal 3,6-connected net with a very rare 3,6T10 topology^17,19 and the point symbol of (3.4.5)(3².4³.5.6³.7⁶), wherein the (3.4.5) and (3².4³.5.6³.7⁶) notations are those of the μ₃-5-Fnic⁻ and Ni nodes, respectively. The rarity of the present topological type was corroborated by a search of various databases (CSD,^15c TOPOS,¹⁷ and RCSR^19c), revealing only a single example of the 3,6T10 framework.²⁰ Interestingly, it concerns a related nickel(II) coordination polymer derived from nicotinic acid, but its topological analysis and classification have not been performed in the original publication.²⁰ Hence, the present work also contributes to the topological identification of unusual metal–organic frameworks, which has been an important research direction in recent years.^17,19

Fig. 1 Structural fragments of 1. (a) Coordination environment of the Ni(II) atom. (b) Dinickel unit. (c) Perspective of the 3D MOF along the ab plane. (d) Topological representation of the underlying binodal 3,6-connected 3D framework with the 3,6T10 topology (view along the c axis). Further details: (a) symmetry codes: i = y, x, z − 1/2; ii = −x + 1, −y + 1, −z + 1; iii = −y + 1, x + 1/2, −z + 1; (b) symmetry code: i = −x + 1, −y + 1, −z + 1/2; (a–c) H atoms (apart from those of H₂O) are omitted for clarity; (d) color codes: 6-connected Ni nodes (pale green), centroids of 3-connected 5-Fnic⁻ nodes (gray), centroids of 2-connected μ-H₂O linkers (red).

Scheme 1 Various coordination modes of 5-Fnic⁻ in compounds 1–5.

4.1.2 [Co(5-Fnic)₂(H₂biim)]_n (2). The asymmetric unit of compound 2 contains one crystallographically unique Co(II) atom, two distinct 5-Fnic⁻ ligands, and one H₂biim moiety. As depicted in Fig. 2a, each six-coordinate Co(II) atom exhibits a distorted octahedral {CoN₄O₂} geometry formed by two carboxylate O atoms of one μ₂-5-Fnic⁻ block, two N atoms of two different 5-Fnic⁻ ligands, and two N atoms of one H₂biim moiety. The Co–O [2.174(2)–2.191(2) Å] and Co–N [2.086(2)–2.162(2) Å] bond lengths are in good agreement with those observed in some other Co(II) compounds.^8,9 In 2, the 5-Fnic⁻ ligands exhibit two different η¹- and μ₂-coordination modes (Scheme 1, modes III and IV), in which the carboxylate group either shows a η¹:η¹ bidentate mode or remains uncoordinated. The H₂biim ligand acts in a bidentate chelating mode; the dihedral angle of two imidazole groups is 2.08°. The μ₂-5-Fnic⁻ moieties alternately link the adjacent Co(II) centers to form a linear 1D metal–organic chain with the Co⋯Co separation of 7.820(3) Å (Fig. 2b). The neighboring chains are assembled into 2D supramolecular sheet motifs through strong N–H⋯O hydrogen bonds (Fig. 4c). These sheets are further held together via the π–π packing interactions (the centroid–centroid separation of adjacent imidazole planes of the H₂biim ligands is 3.719(2) Å) (Fig. S1 and S2†). The above-mentioned 2D sheets formed via the conventional H-bonds^17a were classified from the topological viewpoint, following the methodology developed for the topological analysis of H-bonded nets.^17,21 The topological analysis of an underlying network of 2 disclosed an interesting example of parallel 2D + 2D interpenetration,²² wherein each H-bonded layer is composed of 4-connected Co nodes and 2-connected 5-Fnic⁻and H₂biim linkers (Fig. 2d), giving rise to a uninodal 4-connected net with the sql topology and the point symbol of (4⁴.6²).

Fig. 2 Structural fragments of 2. (a) Coordination environment of the Co(II) atom. (b) Perspective of the 1D metal–organic chain along the ab plane. (c) 2D H-bonded sheet motif along the ab plane (blue dashed lines represent the H-bonds). (d) Topological representation (view along the c axis) of two underlying 2D H-bonded layers showing their parallel 2D + 2D interpenetration; each layer is a uninodal 4-connected net with the sql topology. Further details: (a) symmetry code: i = x − 1, y, z; (c) symmetry code: i = x − 1/2, y + 1/2, −z + 1/2; (a–c) H atoms (apart from those participating in H-bonds) are omitted for clarity; (d) color codes: 4-connected Co nodes (magenta), centroids of 2-connected 5-Fnic⁻ (gray) and H₂biim (blue) linkers.

4.1.3 {[Cd(5-Fnic)₂(phen)]·2H₂O}_n (3). The compound 3 crystallizes in the monoclinic space group P2₁/n and also features a linear 1D metal–organic chain structure. The asymmetric unit bears one crystallographically independent Cd(II) atom, one terminal 5-Fnic⁻ and one μ₂-5-Fnic⁻ block, one phen ligand, and two lattice water molecules. As depicted in Fig. 3a, the six-coordinate Cd(II) atom adopts a distorted octahedral {CdN₃O₃} geometry taken by three O atoms from the two distinct 5-Fnic⁻ moieties and three N atoms from one 5-Fnic⁻ and one phen ligand. The Cd–O and Co–N bond distances are in the 2.243(5)–2.437(5) and 2.313(7)–2.417(6) Å ranges, respectively. In 3, the 5-Fnic⁻ ligands show two different coordination modes (Scheme 1, modes III and V), in which the carboxylate groups are either η¹:η⁰ monodentate or η¹:η¹ bidentate. It should be mentioned that the N atom of 5-Fnic⁻ remains uncoordinated in the mode V. The μ₂-5-Fnic⁻ moieties alternately bridge the adjacent Cd(II) centers to form a linear 1D metal–organic chain (Fig. 3b). The neighboring chains are sewed up into 1D double chain motifs though strong O–H⋯O hydrogen bonds involving crystallization water molecules (Fig. 3c). The resulting double chain motifs are further assembled into a 3D supramolecular framework via the weak C–H⋯π interactions (Fig. S3†). After simplification procedure,^17a the topological analysis of the above-mentioned double chain motifs revealed an underlying binodal 3,3-connected net with the unique topology defined by the point symbol of (5².8)(5³), wherein the (5².8) and (5³) notations correspond to the 3-connected Cd and 5-Fnic⁻ nodes, respectively. An unprecedented character of this topological type was confirmed by a search of different databases.^15c,17,19c

Fig. 3 Structural fragments of 3. (a) Coordination environment of the Cd(II) atom. (b) Linear 1D metal–organic chain. (c) Perspective of a double chain along the bc plane (blue lines present the H-bonding). (d) Topological representation (view along the a axis) of an underlying 1D H-bonded double chain classified as a binodal 3,3-connected net with the unique topology. Further details: (a) symmetry code: i = x, y + 1, z; (c) symmetry code: i = −x + 1/2, y − 1/2, −z + 1/2; (a–c) H atoms (apart from those participating in H-bonds) are omitted for clarity; (d) color codes: 3-connected Cd nodes (turquoise), centroids of 3-connected 5-Fnic⁻ nodes (gray) and centroids of 2-connected (H₂O)₂ linkers (red).

4.1.4 [Cd(5-Fnic)₂(H₂biim)₂] (4). Compound 4 is a discrete 0D monomer. As shown in Fig. 4a, the six-coordinate Cd(II) center is bound by two carboxylate O atoms from two symmetry equivalent 5-Fnic⁻ ligands and four N atoms from two symmetry equivalent H₂biim moieties, thus forming a distorted octahedral {CdN₄O₂} geometry. The Cd–O bonds are 2.450(2) Å, while the Cd–N distances vary from 2.292(3) to 2.303(2) Å; all these distances are comparable to those found in the reported Cd(II) compounds.^8,9,23 In 4, the 5-Fnic⁻ acts as a terminal ligand (Scheme 1, mode V), in which the carboxylate group is in the η¹:η⁰ monodentate mode. The discrete monomeric units of 4 are interlinked by the strong N–H⋯O hydrogen bonds to form 1D chains, which are further assembled via the weak C–H⋯N interactions to generate 2D sheet motifs (Fig. 4b, Table 3). From the topological viewpoint, the 1D H-bonded chains in 4 can be classified as an underlying uninodal 2-connected 2C1 topological network composed of the [Cd(5-Fnic)₂(H₂biim)₂] molecular nodes (Fig. 4c).

Fig. 4 Structural fragments of 4. (a) Coordination environment of the Cd(II) atom. (b) Arrangement of mononuclear blocks into a 2D sheet motif via strong and weak H-bonding interactions (blue dashed lines represent the H-bonds; view along the ab plane). (c) Topological representation of the 1D chains formed via strong H-bonds, showing an underlying uninodal 2-connected network with the 2C1 topology. Further details: (a) symmetry code: i = −x + 2, −y + 1, −z + 1; (b) symmetry codes: i = x − 1, y, z; ii = −x + 1, −y + 2, −z; (a and b) H atoms (apart from those participating in H-bonds) are omitted for clarity; (c) centroids of 2-connected [Cd(5-Fnic)₂(H₂biim)₂] molecular nodes (turquoise).

4.1.5 [Co(5-Fnic)₂(H₂O)₄] (5). Compound 5 also possesses a discrete monomeric structure. As represented in Fig. 5a, the six-coordinate Co(II) atom features a distorted octahedral {CoN₂O₄} geometry, filled by four symmetry equivalent O atoms of terminal H₂O ligands and two N atoms of the symmetry equivalent 5-Fnic⁻ moieties. The Co–O [2.102(2) Å] and Co–N [2.157(3) Å] bonds are in good agreement with those distances observed in 2 and other related Co(II) compounds.^8,9 The 5-Fnic⁻ blocks adopt the coordination mode V with the carboxylate groups remaining uncoordinated. The discrete monomeric units of 5 are interlinked by the repeating O–H⋯O hydrogen bonds to generate a symmetric 3D H-bonded framework (Fig. 5b and Table 3). From the topological perspective,^17,19 this framework is composed of the [Co(5-Fnic)₂(H₂O)₄] molecular nodes and can be classified as an underlying uninodal 6-connected net (Fig. 5c) with the pcu (alpha-Po primitive cubic) topology and the point symbol of (4¹².6³).

Fig. 5 Structural fragments of 5. (a) Coordination environment of the Co(II) atom. (b) Perspective of the 3D H-bonded framework along the ac plane (blue dashed lines represent the H-bonds). (c) Topological representation (view along the a axis) of the underlying uninodal 6-connected 3D H-bonded framework with the pcu topology. Further details: (a) symmetry codes: i = −x + 1, −y + 1, −z + 1; ii = x, −y + 1, z; iii = −x + 1, y, −z + 1; (a and b) H atoms (apart from those participating in H-bonds) are omitted for clarity; (c) centroids of 6-connected [Co(5-Fnic)₂(H₂O)₄] molecular nodes (magenta).

4.1.6 Coordination modes of 5-fluoronicotinate blocks and structural comparison. As depicted in Scheme 1, the 5-Fnic⁻ blocks in 1–5 behave as versatile N,O,O- (modes I, III), N,O- (mode II), N- (mode IV), or O-donors (type V), acting either as μ₂- and μ₃-brdiging or terminal ligands. In particular, the carboxylate group of 5-Fnic⁻ can be in a η¹:η⁰ monodentate (Scheme 1, modes II, V), η¹:η¹ bidentate (mode III), and μ₂-η¹:η¹ bidentate (mode I) fashion, or remains uncoordinated (mode IV). Hence, the structures of 1–5 vary from the 3D MOF (in 1) to the linear 1D chains (in 2 and 3) and the 0D discrete monomers (in 4 and 5), which give rise to the topologically distinct metal–organic (in 1) or H-bonded (in 2–5) networks. The observed structural differences indicate that 5-fluoronicotinate moiety acts as a versatile building block for the construction of coordination compounds, the dimensionality, topology, and supramolecular features of which depend on the type of central metal ion and the presence (optional) of the N,N-ancillary ligand.

4.2 Thermal analysis

The thermal stability of compounds 1–5 was studied under nitrogen atmosphere by thermogravimetric analysis (TGA) and the obtained plots are given in Fig. 6. The compound 1 releases its half water ligand (exptl, 2.57%; calcd, 2.59%) in the 167–188 °C range, followed by the decomposition at 257 °C. The TGA curves of 2 and 4 indicate that these derivatives are stable up to 258 and 215 °C, respectively, and then decompose upon further heating. Superior thermal stability of 2 can be explained by the 2D + 2D interpenetration of the H-bonded layers in comparison with the non-interpenetrating 1D H-bonded chains in 4. For 3, the weight loss associated with the removal of two crystallization water molecules is observed in the 62–100 °C interval (exptl, 6.15%; calcd, 5.91%), and the decomposition of the remaining solid begins at 302 °C. For 5, there are two distinct thermal effects in the 116–281 °C range that correspond to the removal of four coordinated H₂O molecules (exptl, 17.20%; calcd, 17.51%). Further heating up to 343 °C leads to the decomposition of a dehydrated sample. In general, the thermal behavior of compounds 1–5 is comparable to that of other related coordination compounds bearing nicotinate derivatives.²⁴

Fig. 6 Thermogravimetric analysis (TGA) curves of compounds 1–5.

4.3 Luminescent properties

The emission spectra of 5-fluoronicotinic acid (5-FnicH) and its cadmium(II) derivatives 3 and 4 were measured in the solid state at room temperature (Fig. 7). The uncoordinated 5-FnicH shows a weak photoluminescence with an emission maximum at 455 nm, if excited at 392 nm. In contrast, the compounds 3 and 4 display the significantly more intense emission bands with the maxima at 456 nm (λ_ex = 380 nm) and 535 nm (λ_ex = 450 nm), respectively. These bands can be due to the intraligand (π* → n or π* → π) emission.²⁵ The enhancement of luminescence in the coordination compounds can be associated with the type of ligand binding to metal centers,²⁵ and can also be influenced by the H-bonding interactions.^26,27 A somewhat stronger emission intensity in 3 over 4 can potentially be explained by their structural differences in terms of dimensionality, coordination modes of 5-Fnic⁻ blocks, and presence of different N,N-donor ligands.

Fig. 7 Solid state emission spectra of 5-FnicH, 3 and 4 (λ_ex of 392, 380 and 450 nm, respectively).

4.4 Magnetic properties

Variable-temperature magnetic susceptibility studies were carried out on powder samples of 1 and 2 in the 2–300 K temperature range. For the Ni(II) MOF 1, the χ_MT value at 300 K is 1.12 cm³ mol⁻¹ K, which is higher than the spin only value of 1.00 cm³ mol⁻¹ K for one magnetically isolated Ni(II) center (S_Ni = 1, g = 2.0). Upon cooling, the χ_MT value drops down very slowly from 1.12 cm³ mol⁻¹ K at 300 K to 1.08 cm³ mol⁻¹ K at 22 K, and then decreases steeply to 0.50 cm³ mol⁻¹ K at 2 K (Fig. 8). In the 50–300 K interval, the χ_M⁻¹ vs. T plot for 1 obeys the Curie–Weiss law with a Weiss constant θ of −10.12 K and a Curie constant C of 1.11 cm³ mol⁻¹ K, suggesting a weak antiferromagnetic interaction between the Ni(II) ions. Because of the long separation between the adjacent Ni₂ units, only the coupling interactions within the dinickel(II) blocks were considered.

Fig. 8 Temperature dependence of χ_MT (○) and 1/χ_M(□) vs. T for compound 1. The blue line represents the best fit to the equations in the text. The red line shows the Curie–Weiss fitting.

We tried to fit the magnetic data of 1 using the following expression²⁸ for a dinuclear Ni(II) unit:

H = −JS₁S₂

where ρ is a paramagnetic impurity fraction and TIP is temperature independent paramagnetism. Using this model, the susceptibility for 1 above 2.0 K was simulated, leading to the values of J = −2.21 cm⁻¹, g = 2.11, ρ = 0.011, and TIP = 378 × 10⁻⁶ cm³ mol⁻¹, with the agreement factor R = 4.84 × 10⁻⁴ (R = ∑(_obsT − _calcT)²/∑(_obsT)²). The negative J parameter confirms that a weak antiferromagnetic exchange coupling exists between the adjacent Ni(II) centers, which is in agreement with a negative θ value. In the structure of 1 (Fig. 1b), there are two types of the magnetic exchange pathways within the dinuclear units, namely via the μ₂-H₂O and μ₂-η¹:η¹-carboxylate (syn–syn) bridges. According to previous studies, the magnetic interaction is highly sensitive to the value of the Ni–O–Ni bridging angle, showing the domination of the Ni–Ni ferromagnetic coupling when the Ni–O–Ni angles are 90 ± 14°.²⁹ The larger Ni–O–Ni angles in the Ni₂ unit [112.81(12)°] might suggest that the O bridges could be responsible for an antiferromagnetic exchange component. Meanwhile, the syn–syn carboxylate group bridging is likely to have a dominant antiferromagnetic effect. Hence, the small J value observed for 1 (−2.21 cm⁻¹) could well be a result of two compatible exchange effects.

The temperature-dependent magnetic properties of 2 are shown in Fig. 9 in the form of χ_MT vs. T curve. The χ_MT value of 2.76 cm³ mol⁻¹ K at 300 K is much larger than the value (1.87 cm³ mol⁻¹ K) expected for one magnetically isolated high-spin Co(II) ion (S = 3/2, g = 2.0). This is a common phenomenon for Co(II) ions due to their strong spin-orbital coupling interactions.^8,9 The χ_MT values steadily decrease on lowering the temperature and reach the minimum of 1.48 cm³ mol⁻¹ K at 1.99 K. Between 100 and 300 K, the magnetic susceptibility can be fitted to the Curie–Weiss law with C_M = 2.90 cm³ mol⁻¹ K and θ = −14.7 K. These results indicate an antiferromagnetic interaction between the adjacent Co(II) ions.

Fig. 9 Temperature dependence of χ_MT (○) and 1/χ_M(□) vs. T for compound 2. The blue line represents the best fit to the equations in the text. The red line shows the Curie–Weiss fitting.

We tried to fit the magnetic data of 2 using the following expression for a 1D Co(II) chain:³⁰

χ_chain = (Ng²β²/kT)[2.0 + 0.0194x + 0.777x²][3.0 + 4.346x + 3.232x² + 5.834x³]⁻¹x = |J|/kT

Using this rough model, the susceptibility for 2 was simulated, leading to the values of J = −6.18 cm⁻¹ and g = 2.36, and the agreement factor R = 4.47 × 10⁻⁵. The magnetic behavior of compound 2 is comparable to that of a related cobalt(II) nicotinate derivative.^24b

5. Conclusions

In the present study, we applied 5-fluoronicotinic acid as a versatile but still unexplored building block in crystal engineering, resulting in the synthesis and structural characterization of five new coordination compounds 1–5. Depending on the type of metal(II) ion (Ni, Co, Cd) and the presence of optional N,N-donor ancillary ligand, the obtained structures vary from the intricate 3D metal–organic framework 1 to 1D coordination polymers 2 and 3 or discrete 0D monomers 4 and 5. The structures of 2–5 are further extended [1D → 2D (2), 1D → 1D (3), 0D → 1D (4), and 0D → 3D (5)] into various H-bonded networks by means of conventional hydrogen bonds. These H-bonded nets in 2–5 along with the coordination network in 1 were topologically classified, revealing (i) a very rare 3,6T10 MOF topology in 1, (ii) an interesting example of parallel 2D + 2D interpenetration of the sql nets in 2, (iii) a binodal 3,3-connected 1D network with undocumented topology in 3, and (iv) the 2C1 and pcu (alpha-Po primitive cubic) topologies in 4 and 5, respectively.

Besides, by reporting some advanced topological features (namely, very rare topology in 1, interpenetrated topology in 2, or even unprecedented topology in 3), the current work also contributes to the classification, identification, and prediction of various topological motifs in compounds that form either coordination or H-bonded networks.^17,19,21,22 Although a vast number of topological types has been theoretically predicted or experimentally determined, the identification of novel topologies is a constantly growing research direction.¹⁷ In many cases, certain types of properties of coordination compounds can be rationalized by applying the topological analysis methods.^19,21,22 For example, an enhanced thermal stability of derivative 2 over 4 can be explained by the observed 2D + 2D interpenetration of the sql nets in 2.

Apart from topological characteristics, some of the obtained compounds reveal advanced luminescent properties. In fact, the types of ancillary ligands in cadmium(II) derivatives 3 and 4 appear to affect not only their basic structural features, dimensionality, and H-bonding interactions, but also luminescent characteristics with the maxima of strong emission bands ranging from 456 to 535 nm in 3 and 4, respectively. These luminescence features suggest that such compounds can potentially be applied as photoactive materials.

In summary, the disclosed herein structural, H-bonding, and topological rarity of the compounds 1–5, as well as their rather interesting advanced properties (e.g., luminescence of 3 and 4) open up the exploration of 5-fluoronicotinic acid as a promising building block in crystal engineering of metal–organic and supramolecular networks.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (Project 21201091) and the Fundamental Research Funds for the Central Universities (Project lzujbky-2013-59). AMK acknowledges the Foundation for Science and Technology (FCT), Portugal (PTDC/QUI-QUI/121526/2010, PEst-OE/QUI/UI0100/2013).

References

1. (a) Y. Ma, A. L. Cheng, B. Tang and E. Q. Gao, Dalton Trans., 2014, 43, 13957–13964; (b) B. Xu, J. Li, N. N. Kong and C. C. Li, Inorg. Chem. Commun., 2014, 47, 119–122; (c) G. M. Zhuang, X. B. Li and E. Q. Gao, Inorg. Chem. Commun., 2014, 47, 134–137; (d) J. Z. Gu, D. Y. Lv, Z. Q. Gao, J. Z. Liu and W. Dou, Transition Met. Chem., 2011, 36, 53–58.
2. X. F. Zheng and L. G. Zhu, J. Mol. Struct., 2013, 1039, 1–7.
3. (a) L. Y. Yang, L. L. Xin, W. Gu, J. L. Tian, S. Y. Liao, P. Y. Du, Y. Z. Tong, Y. P. Zhang, R. Lv, J. Y. Wang and X. Liu, J. Solid State Chem., 2014, 218, 64–70; (b) Y. X. Hu, Y. T. Qian, W. W. Zhang, Y. Z. Li and J. F. Bai, Inorg. Chem. Commun., 2014, 47, 102–107; (c) B. Van de Voorde, B. Bueken, J. Denayer and D. D. Vos, Chem. Soc. Rev., 2014, 43, 5766–5788.
4. (a) G. Akiyama, R. Matsuda, H. Sato and S. Kitagawa, Chem.–Asian J., 2014, 9, 2772–2777; (b) L. Hashemi, A. Morsali, V. T. Yilmaz, O. Büyükgüngor, H. R. Khavasi, F. Ashouri and M. Bagherzadeh, J. Mol. Struct., 2014, 1072, 260–266; (c) J. Gas, A. Corma, F. Kapteijn and F. X. L. Xamena, ACS Catal., 2014, 4, 361–378.
5. (a) Y. P. He, Y. X. Tan and J. Zhang, Inorg. Chem., 2013, 52, 12758–12762; (b) X. Y. Duan, Q. Q. Meng, Y. Su, Y. Z. Li, C. Y. Duan, X. M. Ren and C. S. Lu, Chem.–Eur. J., 2011, 17, 9936–9943.
6. (a) H. Wang, X. Y. Yang, Y. Q. Ma, W. B. Cui, Y. H. Li, W. G. Tian, S. Yao, Y. Gao, S. Dang and W. Zhu, Inorg. Chim. Acta, 2014, 416, 63–68; (b) F. A. L. Porta, P. H. Ramos, E. C. d. Resende, M. C. Guerreiro, J. O. S. Giacoppo, T. C. Ramalho, J. R. Sambrano, J. Andrés and E. Longo, Inorg. Chim. Acta, 2014, 416, 200–206; (c) W. G. Lu, L. Jiang, X. L. Feng and T. B. Lu, Inorg. Chem., 2009, 48, 6997–6999.
7. W. G. Lu, C. Y. Su, T. B. Lu, L. Jiang and J. M. Chen, J. Am. Chem. Soc., 2006, 128, 34–35.
8. J. Z. Gu, Z. Q. Gao and Y. Tang, Cryst. Growth Des., 2012, 12, 3312–3323.
9. J. Z. Gu, A. M. Kirillov, J. Wu, D. Y. Lv, Y. Tang and J. C. Wu, CrystEngComm, 2013, 15, 10287–10303.
10. T. Liu, S. N. Wang, J. Lu, J. M. Dou, D. C. Li and J. F. Bai, CrystEngComm, 2013, 15, 5476–5489.
11. L. Chen, S. H. Gou and J. Q. Wang, J. Mol. Struct., 2011, 991, 149–157.
12. J. Z. Gu, D. Y. Lv, Z. Q. Gao, J. Z. Liu, W. Dou and Y. Tang, J. Solid State Chem., 2011, 184, 675–683.
13. Y. Zhang, B. B. Guo, S. F. Liu and G. Li, Cryst. Growth Des., 2013, 13, 367–376.
14. (a) B. B. Ding, Y. Q. Weng, Z. W. Mao, C. K. Lan, X. M. Chen and B. H. Ye, Inorg. Chem., 2005, 44, 8836–8845; (b) L. Chen, S. H. Gou and J. Q. Wang, J. Mol. Struct., 2011, 991, 149–157.
15. (a) C. P. Li, J. Chen, P. W. Liu and M. Du, CrystEngComm, 2013, 15, 9713–9721; (b) X.-M. Wang, W. P. Wu, Y. H. Jiang, G. P. Yang and Z. P. Xi, Chin. J. Inorg. Chem., 2014, 30, 192–203; (c) See the Cambridge Structural Database (CSD, version 5.35, May 2014): F. H. Allen, Acta Crystallogr., Sect. B: Struct. Sci., 2002, 58, 380–388.
16. (a) G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Adv., 1990, 46, 467–473; (b) G. M. Sheldrick, SHELXS-97, A Program for X-ray Crystal Structure Solution, and SHELXL-97, A Program for X-ray Structure Refinement, Göttingen University, Germany, 1997.
17. (a) V. A. Blatov, IUCr CompComm Newsletter, 2006, 7, 4–38; (b) V. A. Blatov, A. P. Shevchenko and D. M. Proserpio, Cryst. Growth Des., 2014, 14, 3576–3586.
18. D. Y. Lv, Z. Q. Gao, J. Z. Gu, R. Ren and W. Dou, Transition Met. Chem., 2011, 36, 313–318.
19. (a) M. O'Keeffe and O. M. Yaghi, Chem. Rev., 2012, 112, 675–702; (b) M. Li, D. Li, M. O'Keeffe and O. M. Yaghi, Chem. Rev., 2014, 114, 1343–1370; (c) The Reticular Chemistry Structure Resource (RCSR) Database; M. O'Keeffe, M. A. Peskov, S. J. Ramsden and O. M. Yaghi, Acc. Chem. Res., 2008, 30, 1782–1789.
20. P. Ayyappan, O. R. Evans and W. Lin, Inorg. Chem., 2001, 40, 4627–4632.
21. P. N. Zolotarev, M. N. Arshad, A. M. Asiri, Z. M. Al-amshany and V. A. Blatov, Cryst. Growth Des., 2014, 14, 1938–1949.
22. L. Carlucci, G. Ciani, D. M. Proserpio, T. G. Mitina and V. A. Blatov, Chem. Rev., 2014, 114, 7557–7580.
23. D. Sun, L. L. Han, S. Yuan, Y. K. Deng, M. Z. Xu and D. F. Sun, Cryst. Growth Des., 2013, 13, 377–385.
24. (a) J. Z. Gu, J. Wu, A. M. Kirillov, D. Y. Lv, Y. Tang and J. C. Wu, J. Solid State Chem., 2014, 213, 256–267; (b) H. J. Li, Z. Q. Gao and J. Z. Gu, Chin. J. Struct. Chem., 2014, 33, 1713–1721.
25. Y. Zhang, B. B. Guo, L. Li, S. F. Liu and G. Li, Cryst. Growth Des., 2013, 13, 367–376.
26. Q. Wang, X. H. Yan, H. R. Zhang, W. S. Liu, Y. Tang and M. Y. Tan, J. Solid State Chem., 2011, 184, 164–170.
27. Y. L. Guo, W. Dou, X. Y. Zhou, W. S. Liu, W. W. Qin, Z. P. Zang, H. R. Zhang and D. Q. Wang, Inorg. Chem., 2009, 48, 3581–3590.
28. L. K. Thompson, V. Niel, H. Grove, D. O. Miller, M. J. Newlands, P. H. Bird, W. A. Wickramasinghe and A. B. P. Lever, Polyhedron, 2004, 23, 1175–1184.
29. P. Mahata, S. Natarajan, P. Panissod and M. Drillon, J. Am. Chem. Soc., 2009, 131, 10140–10150.
30. O. Kahn, Molecular Magnetism, VCH Publishers Inc., New York, 1993, p. 258.

---

Footnote

† Electronic supplementary information (ESI) available: Additional structural fragments (Fig. S1–S3†). CCDC 1030268–1030272. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c4ra14534h

---