Coordination polymers of transition metal diphosphonates: Synthesis, crystal structure and magnetic behaviour

Accepted Manuscript Coordination polymers of transition metal diphosphonates: synthesis, crystal structure and magnetic behaviour Wentao Huang, Hong Zhou PII: DOI: Reference:

S0277-5387(18)30496-0 https://doi.org/10.1016/j.poly.2018.08.030 POLY 13353

To appear in:

Polyhedron

Received Date: Accepted Date:

22 May 2018 10 August 2018

Please cite this article as: W. Huang, H. Zhou, Coordination polymers of transition metal diphosphonates: synthesis, crystal structure and magnetic behaviour, Polyhedron (2018), doi: https://doi.org/10.1016/j.poly.2018.08.030

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Coordination polymers of transition metal diphosphonates: synthesis, crystal structure and magnetic behaviour Wentao Huang & Hong Zhou* School of Chemistry and Environmental Engineering, Wuhan Institute of Technology, Wuhan 430073, China

Abstract

Five new transition metal coordination polymers: [Cu2L(4,4'-bipy)(H2O)0.5] (1), [Cu(H2L)H2O] (2), [Cd2(H2L)2H2O] (3), [Co2L(H2O)2]0.184H2O (4), [Fe3L2(OH)(H2O)3]2H2O (5) (H4L = 4-CH3-C6H4CH2N(CH2PO3H2)2), were obtained through hydrothermal reactions. The complexes have been characterized by IR, elemental analysis, TGA and XRD. Crystal structure analysis reveals that all of 2-5 contain 2D inorganic layer structures, constructed by the coordination interactions between metal ions and phosphonate oxygen atoms. 4,4'-bipy molecules in 1 act as a bridging ligand to connect the adjacent 2D layers into a 3D framework. Moreover, magnetic properties of the complexes have been quantificationally investigated by the classical spin approximation and PHI software program, revealing that 2, 4 and 5 exhibit antiferromagnetism, while 1 shows ferromagnetism.

Keywords: Crystal structure; Diphosphonate; Magnetic studies; PHI

1. Introduction Metal-organic coordination polymers have attracted growing interest in recent years owing to their intriguing architectures and potential applications in catalysis [1, 2], gas storage [3, 4], magnetism [5-7] and so on [8, 9]. Compared with the coordination polymers derived from carboxylic acids, metal organic phosphonates exhibit higher thermal and chemical stability due to their strong interactions between [CPO3] units and metal ions [10, 11]. The known crystallographic structures of the metal phosphonates show that the diphosphonic acid ligands can coordinate with metal ions

to form chain or layer structure coordination polymers [12-19], and the coordination diversities of -CPO3 groups lead to different topologies [13, 20]. However, only limited phosphonate complexes have been crystallographically characterized, illustrating the difficult in cultivating the single crystals of this kind of complexes [12, 15, 16]. It is because that metal phosphonates generally have poor crystallinity in water and organic solvents [21, 22]. Based on 4-CH3-C6H4N(CH2PO3H2)2 (H4L, shown in Scheme 1), which has been reported decades ago, only two complexes, [Cd(H3L)2] and Mn2(H2L)2(H2O), have been fully characterized [12, 14]. For further predicting and exploring the properties of metal phosphonates, it is necessary to enrich their structural information by cultivating crystals. Over the past few decades, much effort has been committed to develop magnetic materials by constructing various complexes with different organic ligands and metal ions. To date, plentiful such materials with 3d transition metals have been synthesized and their magnetic properties have been extensively studied [23]. Zheng and coworkers found that magnetic property can be transmitted by C-PO3 groups, and the magnetic strength depends on the distance between metal ions as well as the bridging groups [24]. For examples, there are stronger antiferromagnetic interactions between two Co(II) ions with distance of 3.750(1) Å, connected only by one phosphonate oxygen atom [13]. This phenomenon was also found in [Cu3[NH2(CH2PO3)2]2 with J = -19 cm-1, which linked by CPO3 groups [25]. Moreover, It was observed that the phosphonate ligand is an effective conduit for magnetic interaction among copper centers with a coupling constant of -86 cm-1 [26]. Although the magnetism of some metal phosphonates has been analyzed, more evidence needs to improve and modify the results from the restricted magnetic-structure reports. To add more information about the structure and magnetism for metal phosphonates, five new transition metal phosphonates: [Cu2L(4,4'-bipy)(H2O)0.5] (1), [Cu(H2L)H2O] (2), [Cd2(H2L)2H2O] (3), [Co2L(H2O)2]0.184H2O (4), [Fe3L2(OH)(H2O)3]2H2O (5), have been synthesized and fully characterized. Moreover, the magnetic properties of 1, 2, 4 and 5 were quantificationally investigated by Curie Weiss law and PHI software program [27-34]. In addition, the luminescence of complex 3 has been measured.

2. Experimental All reagents, Cu(NO3)23H2O, 4,4'-bipy, CuCl22H2O, Cd(NO3)24H2O, FeSO47H2O,

HO O OH

P H3C

N

O P

OH

OH Scheme 1 Structure of the diphosphonate ligand (H4L). Co(OAc)24H2O, were obtained from commercially available sources and used as received. H4L was prepared by Mannish reaction according to the synthesis method reported in literature [12]. 2.1. Synthesis of [Cu2L(4, 4'-bipy) (H2O)0.5] (1) Complex 1 was synthesized by hydrothermal reaction. The procedure was as follows: to a 10 mL vial, Cu(NO3)23H2O (0.0242 g, 0.10 mmol), H4L (0.0310 g, 0.10 mmol), 4,4'-bipy (0.0156 g, 0.10 mmol) and 6 mL water were added. Then the vial was put in a 20 mL Teflon-lined autoclave and kept at 140 0C for 2 days. After cooling to room temperature, blue crystals were collected and washed with deionized water. Yield:15 % (based on Cu). Anal. Calcd for C15H19Cu2N2O7P2: C 34.10, H 3.62, N 5.30. Found: C 34.24, H 3.53, N 5.18. IR (KBr, cm-1):3409(s), 2982(w), 2929(w), 1675(m), 1537(w), 1511(w), 1492(w), 1454(m), 1413(m), 1326(w), 1253(m), 1226(s), 1186(m), 1140(m), 1087(m), 1046(s), 978(m), 849(w), 814(m), 791(w), 763(w), 734(m), 639(w), 594(s), 530(w). 2.2. Synthesis of [Cu(H2L)H2O] (2) Complex 2 was obtained as follows: to a 10 mL vial, CuCl22H2O (0.0256 g, 0.15 mmol), H4L (0.0465 g, 0.15 mmol) and 5 mL water were added and the pH of the mixture was adjusted by (Et)3N until the pH reached 4. Then the vial was put in a 20 mL Teflon-lined autoclave and kept at 140 0C for 3 days in same heating and cooling conditions with 1. Bluish crystals were collected and washed with deionized water. Yield: 40 % (based on Cu). Anal. Calcd for C10H17CuNO7P2: C 30.90, H 4.41, N 3.60. Found: C 30.98, H 4.74, N 3.67. IR (KBr, cm-1): 3352(s), 3237(s), 2969(w), 2927(w), 2858(w), 2726(w), 2019(m), 1616(s), 1516(w), 1438(m), 1351(w), 1268(m), 1164(m), 1130(m), 1060(m), 1020(m), 964(s), 868(w), 819(w), 786(w), 752(m), 597(s), 520(w).

2.3. Synthesis of [Cd2(H2L)2H2O] (3) Complex 3 was synthesized as same procedure as 1 except that Cu(NO3)23H2O was replaced by same mol amount of Cd(NO3)24H2O (0.0308 g, 0.10 mmol). Colorless crystals were collected and washed with deionized water. Yield: 50 % (based on Cu). Anal. Calcd for C20H32Cd2N2O13P4: C 28.02, H 3.76, N 3.27. Found: C 28.14, H 3.48, N 3.35. IR (KBr, cm-1): 3350(s), 3010(m), 2970(w), 2840(m), 2790(w), 2610(w), 1610(w), 1550(w), 1520(w), 1460(m), 1400(m), 1320(m), 1290(m), 1190(m), 1110(s), 987(m), 922(w), 895(w), 835(w), 796(w), 766(m), 737(m), 640(w), 607(w), 565(m). 2.4. Synthesis of [Co2L(H2O)2]0.184H2O (4) Complex 4 was synthesized as follows: to a 10 mL vial, Co(OAc)24H2O (0.0373 g, 0.15 mmol), H4L (0.0465g, 0.15 mmol), 4,4'-bipy (0.0156 g, 0.10 mmol) and 5 mL water were added. Then the vial was put in a 20 mL Teflon-lined autoclave and crystallized at 160 0C for 2 days. After cooling to room temperature, pink crystals were collected and washed with deionized water. Yield: 55 % (based on Co). Anal. Calcd for C10H17.35Co2NO8.18P2: C 25.98, H 3.78, N 3.03. Found: C 26.07, H 3.84, N 2.92. IR (KBr, cm-1): 3420(s), 3240(w), 2965(w), 2926(w), 2901(w), 1666(m), 1568(w), 1512(w), 1456(w), 1428(m), 1321(m), 1291(w), 1244(w), 1158(w), 1061(s), 994(w), 976(w), 946(w), 890(w), 786(m), 741(w), 645(w), 632(w), 570(m), 527(w). 2.5. Synthesis of [Fe3L2(OH)(H2O)3]2H2O (5) The preparation of 5 was similar to that of 4 except that Co(OAc)24H2O was replaced by same mol amount of FeSO47H2O (0.0417 g, 0.15 mmol). Brown crystals were collected and washed with deionized water. Yield: 45 % (based on Fe). Anal. Calcd for C10H20Fe1.5NO9P2: C 25.98, H 3.78, N 3.03. Found: C 26.06, H 3.71, N 2.94. IR (KBr, cm-1): 3454(s), 3038(w), 2865(w), 1658(w), 1561(w), 1532(w), 1463(w), 1413(m), 1323(w), 1257(w), 1157(m), 1080(m), 1041(w), 1011(w), 945(w), 908(w), 833(w), 761(m), 606(w), 576(m), 540(m). 2.6. Physical measurements The FT-IR spectra were recorded on a Nicolet Magna 750 FT-IR spectrometer in the range of 4000-400 cm-1 by using KBr pellets. Elemental analyses for C, H and N were measured with a Perkin-Elmer 240 analyzer. Thermogravimetric analyses (TGA) were performed on a Q500 TGA (TA Instruments) under a nitrogen atmosphere at 10 0C min-1 from room temperature to 900 0C. Magnetic susceptibility measurement of metal phosphonates was performed in the temperature

range 2 - 300 K on a MPMS SQUID VSM magnetometer with applied magnetic field of 1000 Oe. Powder X-ray diffraction (PXRD) patterns of the samples were recorded on a bruker D8 diffractometer with Cu Ka radiation (λ = 1.5418 Å). 2.7. Single-crystal structure determination X-ray single crystal diffraction of the obtained complexes 1-5 was performed in sequence on a Super Nova, Dual, Cu at zero, Atlas diffractometer. Diffraction intensity data were collected on a SMART-CCD area-detector diffractometer at 293 K using graphite monochromatic Mo-Ka radiation (l = 0.71073 A˚). Data reduction and cell refinement were performed by SMART and SAINT programs [35], and semiempirical absorption correction was applied to the intensity data using the SADABS program [36]. The structures were solved by direct methods (Bruker SHELXTL) and refined on F2 by full-matrix least squares (Bruker SHELXTL) using all unique data [37]. The non-H atoms in the structure were treated as anisotropic. Hydrogen atoms were located geometrically and refined in riding mode. Crystallographic data for complexes 1-5 is given in Table 1. Selected bond lengths and angles for 1-5 are listed in Table S1. Table 1 Crystallographic Data and structure refinement for the complexes 1-5 1 Formula

C15H19Cu2N2OP2

2 C10H17CuNO7P2

3 C20H32Cd2N2O13P4

4 C10H17.35Co2NO8.18 P2

5 C20H37Fe3N2O18P4

fw

528.34

388.73

857.15

462.28

884.95

cryst.system.

triclinic

monoclinic

monoclinic

triclinic

monoclinic

space group

P1

P 21/c

I2/a

P1

C 2/c

a (Å)

6.0965(6)

15.3140(8)

10.9396(4)

6.3026(5)

32.654(4)

b (Å)

11.2026(11)

10.4888(6)

7.8404(4)

9.4778(7)

9.8496(11)

c (Å)

14.3882(13)

9.1440(6)

32.5361(12)

13.1707(10)

10.5221(11)

β (deg)

91.187(3)°

103.421(2)°

94.350(1)°

83.461(3)°

108.142(4)°

2

4

4

2

8

Dc (g cm 3)

1.919

1.807

2.046

2.037

1.834

F (000)

534.0

796.0

1704.0

468.0

1824.0

T/K

173K

172K

173K

173K

173 K

R1

0.0510

0.0381

0.0233

0.0451

0.0653

wR2

0.0939

0.0837

0.0543

0.0785

0.1043

R(All)

0.0948

0.0557

0.0312

0.0803

0.121

1562463

1559820

1559818

1583547

1583548

Z -

CCDC number

3. Results and discussion Complexes 1-5 were synthesized under hydrothermal conditions by the reaction of transition metal salts and H4L in aqueous solution. Except for 2, obtained by mixing copper(II) salt and H4L in water solution at pH = 4, the others were acquired in the presence of 4, 4'-bipy. The coordination modes of H4L in the five complexes are summarized in Scheme S1. The coordination modes for 1-5 are 111212, 001110, 012110, 121211 and 111210, respectively. The results indicate that H4L adopts different coordination modes in the complexes, even if the deprotonation number in ligand is the same, the coordination mode is different in the complexes. The changeable character of H4L leads to the structural varieties of the complexes. 3.1 Crystal structure description of 1 Complex 1 crystallizes in a space group P1. The asymmetric unit in 1 contains two Cu(II), one L4- anion, half 4,4'-bipy molecule and one coordination water molecule (Fig. 1a). Cu1 atom has a distorted octahedra coordination geometry, in which equatorial four positions are occupied by one nitrogen atom and three phosphonate oxygen atoms (O1, O3, O4), two apical positions are occupied by O3 and O2 with the distances of O3-Cu1 and O2-Cu1 are 2.802 Å and 2.644 Å, respectively. The longer bond length is due to the Jahn-Teller effect. Cu2 has a tetragonal pyramid coordination geometry as ascertained by  values [38], surrounded by three phosphonate oxygen atoms (O2, O5, O6), one nitrogen atom from 4,4'-bipy and one water molecule (O7) locating in the apical position. Both Cu-O (1.908-2.802Å) and Cu-N (2.037-2.061 Å) in the basal plane are comparable to those reported in literature [39]. The ligand is four deprotonated (O1, O3, O4, O6) to balance charge and P-O bond lengths. L4- anion acts as an hexadentate ligand to chelate six Cu atoms. The two polyhedrons, [CuO5N] and [CuO4N], and two {CPO3} tetrahedra are interconnected into a unit via corner-sharing, forming a dinuclear unit. These units are interlinked by {CPO3} units into a layer (Fig. 1b), which contains 4-, 6-, 8- and 16-member rings. The layer can be formulated as [Cu2(PO3)2] ,where two PO3 units in each L4- involves in it. The layers are further connected by 4,4'-bipy into 3D framework. The distance of the adjacent layers is ca. 11.17 Å, measured between Cu(II) atoms. Phenyl groups are hanging on two sides of the layer, orientating toward the interlayer space. As shown in Fig. 1c, the connections between the layers and 4,4'-bipy molecules lead to the formation of a MOF with a tetragonal array of channels. The channel size is 11.110.8 Å, which is composed of 34-member ring. From the topological point of

Fig. 1 The perspective view of the asymmetric unit of 1 (a), 2D inorganic layer structure with 4-, 6-, 8and 16-membered rings in it (b), 3D packing diagram along the b-axis showing 34-membered ring (c), Topological representation of the network of 1(d).

view, Cu1 center is connected to three L4- units, the L4- anion is connected to three Cu(II) centers, and Cu2 center is connected to three L4- anions and one 4,4'-bipy, they can be regarded as triple three and four connected nodes, respectively. The three dimensional framework can be conveniently described by employing a topological approach [40]. All of the topological studies have been performed by using the software package TOPOS [41]. The overall 3D network can be rationalized as a 3,3,3,4-c tetranodal net with stoichiometry (3-c) (3-c) (3-c) (4-c). It has an extended Schlafli symbol of {4.82.103} {4.82} {4.82} {4.82}, which is assigned to a rare topology, gra (topos&RCSR.ttd) (Fig. 1d). 3.2 Crystal structure description of 2 Complex 2 crystallizes in space group P 21/c. 2 contains one Cu(II), one twofold deprotonated ligand (H2L2-) and one coordination water molecule. As shown in Fig. 2a, Cu(II) is coordinated by four phosphonate oxygen atoms, one nitrogen atom and one water molecule, constructing a distorted octahedron. The bond lengths in the basal plane are Cu1-O4 1.967 Å, Cu1-O1 1.943 Å, Cu1-O6 1.947 Å and Cu1-N1 2.062 Å, respectively. And the axial sites are occupied by one O atom from water molecule and one O atom from phosphonate group with longer distances of

Fig.2 Asymmetric unit of 2 showing the coordination polyhedron 2 (a), 2D inorganic layer structure with 8- and 16-membered rings in it (b), stacking diagram of 2 (c), topological representation of the network of 2 (d)

2.580 Å and 2.424 Å, respectively, due to the Jahn-Teller effect. The bond lengths in basal plane are comparable with those in the other Cu(II) phosphonates [42]. Different from the ligand in 1, the ligand in 2 is two deprotonated (O1 and O6). The coordination interactions between Cu(II) and {CPO3} result in the formation of 2D inorganic layer with alternant 8- and 16-member rings (Fig. 2b), where one Cu(II) connects with adjacent five equivalent Cu(II) atoms through three phosphonate groups. The composition of the layer is [Cu(PO3)], where only one PO3 unit in each ligand involves in the layer. The distances between adjacent Cu atoms are in the range of 5.288(7) - 5.680(7) Å. Phenyl groups are hanging on the two sides of the layer (Fig. 2c). From the topological point of view, each ligand is connected to three Cu(II), they can be regarded as three connected nodes. The overall 2D network can be rationalized as a 3-c uninodal net (Fig. 2d). It has an extended Schlafli symbol of {63}, gra (topos&RCSR.ttd). Although 1 and 2 contain inorganic layers formed by the coordination interactions between Cu(II) and [CPO3], the layer topology is different. 3.3 Crystal structure description of 3

Fig. 3 Asymmetric unit of 3 showing the coordination polyhedron (a), 2D inorganic layer structure showing 8- and 16-membered rings in it (b), stacking diagram of 3 (c), Topological representation of the network of 3 (d).

Complex 3 crystallizes in space group I 2/a. The asymmetric unit of 3 contains two Cd(II), two twofold deprotonated ligands (H2L2-) and one coordination water molecule (Fig. 3a). Cd(II) is coordinated by five phosphonate oxygen atoms and one water molecule, constructing a distorted octahedron. The bond lengths of Cd-O in the basal plane are in the range of 2.235(2) Å - 2.358(2) Å, and the axial sites are occupied by one O atom from water molecule and one O atom from phosphonate group with the distances of 2.402(2) Å and 2.287(2) Å, respectively. These bond lengths are slightly longer than those in Cd(H3L)2 [12]. Besides, the equivalent Cd atoms are doubly bridged by 3-O4 with Cd-O-Cd angle of 96.69(8)0. Similar with the ligand in 2, the ligand in 3 is also two deprotonated (O1 and O5). Each ligand chelates four Cd atoms through four phosphonate oxygen atoms (O1, O3, O4, O6). And each Cd atom connects with five adjacent equivalent Cd atoms through five CPO3 units and one 3-O (H2O), forming an inorganic layer, where the distances of three adjacent Cd···Cd are 3.471(5) Å over the doubly 3-O4 bridges and 4.044(4) Å and 5.323(5) Å across O-P-O units, respectively (Fig. 3b). Similar with complex 2, phenyl groups are hanging on the two sides of the layer (Fig. 3c). The layer can also be regarded as the aggregation of infinite 8- and 18-member rings. It is interesting to compare the structure of 3 [Cd2(H2L)2H2O] with Cd(H3L)2 [12]. Both contain same metal ion and ligand with different metal/ligand ratio, it leads to the difference in both coordination mode of the same ligand (4 -

012110 for the former and 3 - 001001 for the later, respectively) and space configurations, layer and double chain, respectively. In topology view, the Cd(II) center is connected to four ligands, and the ligand is connected to four Cd(II) centers, they can be regarded as four connected nodes, respectively. The overall 2D network can be rationalized as a 4-c uninodal net (Fig. 3d). It has an extended Schlafli symbol of {43.63}. 3.4 Crystal structure description of 4 Complex 4 crystallizes in space group P1. 4 contains two Co(II), one fourfold deprotonated ligand (L4-) and two coordination water molecules (Fig. 4a). The Co1 atom has an octahedral environment, surrounded by one nitrogen atom, four phosphonate oxygen atoms, and one water molecule [Co1-O 2.054(3) -2.161(3) Å, Co1-N 2.209(4) Å]. The Co2 atom has also an octahedral environment surrounded by five phosphonate oxygen atoms and one water molecule. The bond lengths of Co2-O are in the range of 2.010(3) - 2.260(3) Å, which are comparable with those in the other Co(II) phosphonates [13]. Besides, the equivalent Co1 atoms are doubly bridged by 3-O1 with Co-O-Co angle of 100.59(1)0. Similar with 2，3 and 4 also form inorganic layer structure due to the coordination interactions between metal ions and {CPO3} units. It can be seen in Fig. 4b that there are broken line Co4O8 tetramer formed by doubly 3-O(P) bridge between

Fig. 4 Asymmetric unit of 4 (a), 2D inorganic layer structure showing 8- and 16-membered ring in it (b) and stacking diagram of 2D layer (c), Topological representation of the network of 4 (d).

adjacent Co(II) ions in the layer structure. Each tetramer connects with six ones by {CPO3} units, leading to a 2D layer structure with alternant 4- and 8-member rings in it. Except for the different metal ions, the composition of the layer in 4 is same with that in 1. Phenyl groups are hanging on the two sides of the layer (Fig. 4c). From the topological point of view, the Co(II) center is connected to three ligands, and the ligand is connected to eight Co(II) centers, they can be regarded as 3- and 8- connected nodes, respectively. The overall 2D network can be rationalized as a 3,8-c binodal net with stoichiometry (3-c) (3-c) (8-c) (Fig. 4d). It has an extended Schläfli symbol of {3.42}{3.42}{34.46.56.68.73.8}. 3.5 Crystal structure description of 5 Complex 5 crystallizes in space group C 2/c. The asymmetric unit contains three Fe(III), two L4anions, one OH- anion and three coordination water molecules (Fig. 5a). Fe1 atom has a trigonal bipyramid environment surrounded by four phosphonate oxygen atoms and one water molecule [Fe1-O 2.026(4)-2.149(4) Å], where two equivalent O3 atoms occupied the apical positions and O7 is in disorder. The coordination environment of the two equivalent Fe2 atoms can be described

Fig. 5 Asymmetric unit and coordination polyhedron of Fe(III) in 5 (a), 2D inorganic layer structure with 8 and 16-membered rings in it (b), Stacking diagram of 5 (c), Topological representation of the network of 5 (d).

as an octahedron with the distances of Fe2-O are in the range of 1.962(3)-2.119(4) Å, which are comparable to other Fe(III) organic phosphonate complexes reported in literature [43]. Besides, the equivalent Fe2 atoms are doubly bridged by 3-O(4) with Fe-O-Fe angle of 99.30(2) 0. The L4anion acts as a pentadentate ligand to chelate five Fe atoms, one phosphonate group is tridentate with 3-111 coordination mode, whereas the other one is bidentate with 3-210 coordination mode. Each asymmetric unit interlinks with adjacent four ones through Fe-O-Fe and Fe-O-P-O-Fe coordination interactions, leading to a 2D layer structure with alternant 4- and 8-membered rings in it (Fig. 5b). Phenyl groups are hanging on the two sides of the plane (Fig. 5c), which is similar with those of 2-4. From the topological point of view, the Fe1 center is connected to four ligands, the Fe2 center is connected to three ligands, they can be regarded as three and four connected nodes, respectively. The overall 2D network can be rationalized as a 2,3,4,4-c tetranodal net with stoichiometry (2-c) (2-c) (3-c) (3-c) (4-c) (4-c) (4-c) (Fig. 5d). It has an extended Schlafli symbol of {3.5.7}{3.5.7}{4.5.94} {4.5.94}{34.5.72.9}{4}{4}. The five complexes all exhibit layer structure constructed completely from inorganic components, PO3 groups and metal ions, and the topological structures of the inorganic layers vary with the metal ions.

3. 6. IR spectra In the IR spectra of 1-5, as shown in Fig. S1, the bonds between 900-1200 cm-1 can be ascribed to the stretching vibrations of the tetrahedral {CPO3} groups [12]. The bands located between 2804-2863 cm-1 are likely due to (P-OH) [44]. The broad bands centered at 3405, 3360, 3368, 3431 and 3454 cm-1 for 1-5 are due to the H-OH stretching vibration of the water molecules, respectively [14]. The weak bands around 2921 to 2935 cm-1 are likely assigned to the vibration of CH2 groups [15]. The bands around 1390 to 1603 cm-1 are probably due to the vibration of benzene ring. The metal-ligand (M-O) stretching vibrations are observed at 524, 579, 561, 567, 546 cm-1 for 1-5, respectively [25]. The structural characters are in agreement with the expected ones in 1-5.

3.7. PXRD and TGA study In order to check the phase purity of the bulk samples, powder X-ray diffraction (PXRD) analysis

Complex 1 Complex 2 Complex 3 Complex 4 Complex 5

110 100

Weight %

90 80 70 60 50 40 30

0

200

400

600

800

1000

Temperature C Fig.6 TGA curves of the complexes 1-5

was carried out for complexes 1-5. As shown in Fig. S2, all of the peaks displayed in the experimental patterns closely match those in the simulated patterns generated from single-crystal diffraction data, which indicates the high purity of the bulk samples. To investigate the stability of 1-5, thermal gravimetric analysis (TGA) was performed in the range of 20-900 0C. As shown in Fig. 6, the TGA diagrams of five complexes are different. The pyrolysis of these complexes can be roughly regarded as three main steps of weight losses. The first step is due to the release of water molecules in the complexes, the second step can be ascribed to the pyrolysis of the inorganic and organic components, including the release of water formed by condensation of the protonated phosphonate groups and combustion of organic groups, and the third step can be assigned to the further decomposition of the complexes. The final products are usually assumed to metaphosphate and/or phosphate, confirmed by the comparison between the theoretical and the observed mass. Considering that possible reactions between the residuals and the container (Al2O3), the residuals were not further identified but assigning them only by calculating according to the reports in literature [45] . TGA curves of 1-3 are similar and exhibit two main steps of weight loss. One is corresponding to the loss of the lattice water and coordination water, the observed weight loss of 3.3 %, 4.7 % and 1.9 % are close to the calculated values 3.4 %, 4.6 % and 2.1 % for 1-3, respectively. Another is relative to the combustion of the organic groups with the starting temperature of 240, 240, 2800C, respectively. 5 is not stable, the weight loss of 5 decreases rapidly

along with the increasing temperature. The loss of 8.5% for 4 between 20-185 0C can be attributed to the escape of water molecules and only 3.7 % loss can be observed in the range of 185-432 0C, reflecting its high thermal stability. The residues can be assigned to Cu(PO3)2 for 1, 3/4Cu(PO3)2+1/4CuO for 2, Cd(PO3)2 for 3, Co2P2O7 for 4 and 1/2Fe(PO3)3+1/2Fe2O3 for 5, where the observed and calculated values for 1-5 are 39.6 % (41.9 %), 46.9 % (47.8 %), 61.5 % (63.1 %), 60.4 % (60.6 %) and 42.1 % (40.2 %), respectively. The results show that the residues in high temperature of 1-5 are mainly contributed to the metaphosphate, which are comparable with the components of the layer structure for each complex. In other words, the layer structures in the five complexes have high thermal stability. It will provide useful hints for preparing layer phosphonates with regular arrangement.

3.8. Magnetic Properties

Fit

0.2

0.8

50

100

150

200

250

H = 1000 Oe  T

100

Fit

50

M

M-1

0

Curie-Weiss Fit

Curie-Weiss Fit

0

0.6 0.4

M-1

0.0

150

-3

200

-1

1.0

-1

100

M

250

MT / cm mol K

3

H = 1000 Oe  T

0.4

-3

200

-1

0.6

M / cm mol

300

-1

MT / cm mol K

0.8

1.2

3

400

1.0

M / cm mol

Variable-temperature magnetic susceptibilities for the crystalline samples of 1, 2, 4 and 5 were

0

0.2 0

300

50

100

150

200

250

300

T/K

T/K

1

2 80

8

M

20

0 0

50

100

150

200

250

 T M

2

M-1

Curie-Weiss Fit

0

300

10

Fit

0

0

T/K

4

H = 1000 Oe

-3

-1

M-1

Curie-Weiss Fit

20 4

-1

1

Fit

-1

H = 1000 Oe  T

3

40 2

6

M / cm mol

MT / cm mol K

-3

3

30

M / cm mol

60

3

-1

MT / cm mol K

4

50

100

150

200

250

0

300

T/K

5

Fig.7 Plots of MT and 1/M versus T for the complexes 1, 2, 4 and 5. The purple and red lines represent the best fits by PHI program and Curie-Weiss law, respectively

measured under an applied direct current (dc) field of 1.0 kOe in the temperature range of 2.0-300 K. The magnetic behaviours in the form of χMT and M-1 vs T plots are depicted in Fig. 7. The MT value for 1 is 0.77 cm3 mol-1 K at room-temperature, which is very close to the spin-only value (2

CuT = 0.75 cm3 mol-1 K, S = 1/2) for two magnetically isolated high-spin Cu(II) ion in an octahedral environment [25]. Upon cooling, the MT reaches a maximum value of 1.01 cm3 mol-1 K at 8 K, and then decreases to a minimal value of 0.63 cm3 mol-1 K at 2 K. This sudden decrease may be attributed to various factors as literature analyzed, such as saturation effects due to Zeeman splitting, intermolecular interactions, or zero-field splitting (zfs) within the ground state [46].The fit according to Curie-Weiss law (MT = C/(T-θ)) between 2 -300 K yields solution C = 0.76 emu mol-1 K and θ = 4.66 K. The positive sign of the Curie-Weiss constant indicates ferromagnetic interactions between Cu(II) centers. To estimate the coupling constants between Cu(II) centers in 1 by PHI program, the Hamiltonian H = - J1(S1S2) was adopted and zJ' was also introduced for the magnetic interactions between the units connected by 4, 4'-bipy molecules. The magnetic exchange pathways for the four complexes are shown in Fig. S3. The best fit for 1 in the range of 10-300 K is obtained with g = 1.99, J1 = 14.25 cm-1 and zJ' = 0.04 cm-1. The ferromagnetic property is in agreement with the magnetic behavior shown in the plot of MT vs T as well as the Weiss value. In 1, the dinuclear Cu units are connected by -O(P) and O-P-O unit. Considering that the Cu1-O2(P)-Cu2 bond angle is 106.60C and the O2(P) occupies the axial positions with long Cu1-O2 length (2.644Å), very weak antiferromagnetic interaction could be expected. Therefore, the ferromagnetic property of 1 can be mainly ascribed to the O-P-O connection between the two Cu(II) atoms with shorter Cu-O lengths (1.909-1.959 Å). These results are also comparable with those of (bmim)2[Cu3(μ3-Cl)2(μ-pz)3Cl3] and (Bu4N) [Cu3(μ3-Cl)2(pz)3Cl3] [46, 47]. The small zJ' is reasonable due to the fact that the magnetic exchange through 4, 4'-bipy linker is usually very weak [48]. The room-temperature MT value for 2 is 1.21 cm3 mol-1 K, which is higher than the spin-only value (CuT = 0.43 cm3 mol-1 K) for a magnetically isolated high-spin Cu(II) ion in an octahedral environment (S = 1/2, g = 2.15) [49]. The MT value decreases continuously with decreasing temperature and reaches a minimum value of 0.36 cm3 K mol−1 at 2.0 K, revealing that significant antiferromagnetic interactions between the Cu(II) centers. The temperature dependence of 1/M between 2 and 100 K approximates Curie-Weiss behavior with C = 1.24 emu mol-1 K and θ =

-19.71 K. The negative sign of the Curie-Weiss constant indicates antiferromagnetic interactions between Cu(II) centers. A fit by PHI program was performed using the following Hamiltonian H = -J1(S2S3)-J2(S1S3)-J3(S1S2). The best fit is obtained by introducing the magnetic interactions (zJ'), yielding the parameters g = 1.91, J1 = -1.86 cm-1, J2 = -2.55 cm-1, J3 = 2.80 cm-1 and zJ' = -2.56 cm-1.The negative J2 and positive J3 can be explained by the structural parameters, the torsion angles Cu1-OO-Cu2 56.1760 and Cu1-OO-Cu3 163.580 are expected to have antiferromagnetic and ferromagnetic properties according to the literature reports for the magnetic exchange of metal ions bridged by O-P-O, respectively [24]. The results show that the overall magnetic behavior is contributed to the exchange couplings between the Cu(II) ions marked as J1 and J2 as well as the interactions between the trinuclear units. The room-temperature MT value for 4 is 4.32 cm3 mol-1 K (Fig. 7 (d)), which is higher than the spin-only value (2XCoT = 3.75 cm3 mol-1 K) for two magnetically isolated high-spin Co(II) ions (S = 3/2, g = 2) [50]. The MT value decreases continuously with decreasing temperature and reaches a minimum value of 0.97 cm3 K mol−1 at 9.0 K, revealing that overall antiferromagnetic interactions between the metal centers transmitted by μ-O bridges and O-P-O moieties occur in the nearest neighbors of 4. Below 9 K, MT increases abruptly to reach a maximum at ~6 K of 2.09 emu mol-1 K and finally decreases again at lower temperatures. The behavior has also occurred in other complexes [50]. The susceptibility data above 10 K follow the Curie-Weiss law with C = 5.02 emu mol-1 K and θ = - 56.08 K. The larger negative value of the Curie-Weiss constant indicates strong antiferromagnetic interactions between Co(II) centers (shown in Fig. S3) [13, 25]. The presence of a stronger antiferromagnetic exchange coupling in 4 is reasonable, considering that the shorter CoCo distance of 3.192(1) Å in the trinuclear Co(II) unit. It was reported that modelling Co(II) systems in any situation is difficult due to the strong orbital moment and crystal field effects [51]. To quantify the data, the PHI program was utilized. The best fit was obtained by using the unit (shown in Fig. S3) and considering the interactions between the units. It yields the following parameter, g = 1.90, J1 = -32.41 cm-1, J2 = 14.65 cm-1, J3 = 15.26 cm-1 and zJ' = -1.22 cm-1. The adopted Hamiltonian is H = -J1(S1S3) -J2(S1S2) -J3(S2S3). The small negative zJ' value confirm a weak antiferromagnetic interaction between the units [7]. Considering that the other two J values are positive, the overall antiferromagnetic property of 4 is mainly attributed to the stronger magnetic exchange interactions between Co1 and Co3, where Co1 and Co3 are connected

by double equivalent 3-O(P) with the Co1-O-Co3 angle of 100.59 0. The room-temperature MT values for 5 is 8.63 cm3 mol-1 K smaller than the spin-only value of 13.125 cm3 mol-1 K for three non-coupled high-spin Fe(III) centers in an octahedron environment (S = 5/2, g = 2), which may be contributed to the different coordination polyhedrons, trigonal bipyramid for one Fe(III) and octahedron for two Fe(III) [52]. Similar with 4, the MT value in 5 decreases continuously with decreasing temperature and reaches a minimum value of 1.99 cm3 K mol−1 at 10.0 K, revealing that significant antiferromagnetic interaction in 5. Below 9 K, similar magnetic behavior with 4 was found, MT increases abruptly to reach a maximum at 6 K (6.31 emu mol-1 K) and finally decreases again at lower temperatures. The magnetic data in the range of 10-300 K was fitted by the Curie-Weiss equation to produce C = 10.24 emu mol-1 K and θ = -70.78 K. The negative value of θ indicates antiferromagnetic interactions between Fe(III) centers [25]. (as shown in Fig. S3) The PHI program was utilized to simulate the observed magnetic data over the accessible temperature range (2-300K). The magnetic data of 5 according to the structure (shown in Scheme S1) can be analyzed based on the following Hamiltonian, H = -J1(S2S4) -J2(S1S4) -J1(S1S2)-J4(S1S3). The best fit was obtained by introducing the interaction between the units, giving g = 2.04, J1 = J2 = -27.35 cm-1, J3 = -7.37cm-1, J4 = -13.57 cm-1, J5 = 13.32 cm-1 and zJ' = -0.91 cm-1. The larger negative J as well as θ values confirm a stronger antiferromagnetic interaction in the tetranuclear Fe units and small negative zJ' illustrates weak interactions between the units in the complex [7]. The positive J3 is the synergistic effects from O9 (water) bridge and two O-P-O connections. Meanwhile, the bigger negative J1 and J2 can be owed to the shorter Fe-Fe distance (4.239 Å) with small Fe-OO-Fe torsion angle (47.20), which was reported to contribute to larger negative J value [24].

3.9. Luminescent Properties The solid-state photoluminescent properties for complex 3 and the free ligand have been investigated at room temperature (Fig. 8). The maximal emission peaks of the ligand and complex 3 are 290 nm and 310 nm at the excitation wavelengths of 270 nm, respectively. In comparison with the free ligand, the emission of the complex 3 exhibits obvious shift and lower luminescence intensity, which can be attributed to the coordination interactions of Cd(II) with the ligand. Due to

Fig. 8 The solid-state photoluminescent spectra of the free ligand (red) and complex 3 (black) at room temperature.

d10 closed shell electronic configuration, the photoluminescent behavior of the complex 3 is the intraligand charge transfer of the ligands, caused by the ligand-centered *- transition of the benzene rings [53].

Conclusions In summary, using various metal ions in the synthesized systems of H4L and H2O, four layered coordination polymers were obtained in hydrothermal conditions, when adding 4, 4'-bipy in Cu(II), H4L and H2O system, a layer structure extended into 3D supramolecular network. In the five complexes, neighboring metal ions are interlinked by bridging diphosphonate ligands into inorganic layer structures. The reaction between different metal ions and H4L can lead to layer structures with different topologies, which is completely constructed by inorganic components. These inorganic layers have higher thermal stability. The magnetic investigation for the complexes 1, 2, 4 and 5 show that PO3 or 3-O(P) units can serve as an effective pathway to transmit magnetic interactions between adjacent metal ions. The magnetic nature and magnitude vary with the distance and bridging mode between metal ions.

Conflicts of interest There are no conflicts to declare.

Acknowledgements This work was supported by the National Nature Science Foundation of China (21171135) and the ninth postgraduate education innovation fund of Wuhan Institute of Technology, China (CX2017086)

Notes and references [1] C. Huang, X. Han, Z. Shao, K. Gao, M. Liu, Y. Wang, J. Wu, H. Hou, L. Mi, Solvent-Induced Assembly of Sliver Coordination Polymers (CPs) as Cooperative Catalysts for Synthesizing of Cyclopentenone[b]pyrroles Frameworks, Inorg. Chem., 56 (2017) 4874-4884. [2] G.A.O. Tiago, K.T. Mahmudov, M.F.C. Guedes da Silva, A.P.C. Ribeiro, F.E. Huseynov, L.C. Branco, A.J.L. Pombeiro, Copper(II) coordination polymers of arylhydrazone of 1H-indene-1,3(2H)-dione linked by 4,4′-bipyridineor hexamethylenetetramine: Evaluation of catalytic activity in Henry reaction, Polyhedron, 133 (2017) 33-39. [3] C. Hua, Q.-Y. Yang, M.J. Zaworotko, Construction of a Series of Porous (3,9)-c Coordination Networks Using Dicarboxylate and Tris-pyridyl Ligands and Their Gas Storage Properties, Cryst. Growth Des., 17 (2017) 3475-3481. [4] Y. Liao, Z. Cheng, W. Zuo, A. Thomas, C.F.J. Faul, Nitrogen-Rich Conjugated Microporous Polymers: Facile Synthesis, Efficient Gas Storage, and Heterogeneous Catalysis, ACS Appl. Mater. Interfaces, 9 (2017) 38390-38400. [5] J.-W. Zhang, J.-X. Li, Y. Jiang, B.-Q. Liu, Y.-P. Dong, Syntheses, structures, photoluminescence, and magnetism of a series of lanthanide 1,3-adamantanedicarboxylate coordination polymers, Polyhedron, 138 (2017) 68-73. [6] Y. Ma, L. Du, K. Wang, Q. Zhao, Synthesis, Crystal Structure, Luminescence and Magnetism of Three Novel Coordination Polymers Based on Flexible Multicarboxylate Zwitterionic Ligand, Crystals, 7 (2017) 32. [7] S. Goswami, G. Leitus, B.K. Tripuramallu, I. Goldberg, MnII and CoII Coordination Polymers Showing Field-Dependent Magnetism and Slow Magnetic Relaxation Behavior, Cryst. Growth Des., 17 (2017) 4393-4404. [8] L.H. Jia, R.Y. Li, Z.M. Duan, S.D. Jiang, B.W. Wang, Z.M. Wang, S. Gao, Hydrothermal synthesis, structures and magnetic studies of transition metal sulfates containing hydrazine, Inorg. Chem., 50 (2011) 144-154. [9] Y. Chen, X.-L. Chen, Y. Chen, L. Jia, M.-H. Zeng, Design, structure and magnetic properties of a novel one-dimensional Mn(II) coordination polymer constructed from 4-pyridyl- NH -1,2,3-triazole, Inorg. Chem. Commun., 84 (2017) 182-185. [10] T. Zheng, Z. Yang, D. Gui, Z. Liu, X. Wang, X. Dai, S. Liu, L. Zhang, Y. Gao, L. Chen, D. Sheng, Y. Wang, J. Diwu, J. Wang, R. Zhou, Z. Chai, T.E. Albrecht-Schmitt, S. Wang, Overcoming the crystallization and designability issues in the ultrastable zirconium phosphonate framework system,

Nat. Commun., 8 (2017) 15369. [11] J. Liu, G. Liu, C. Gu, W. Liu, J. Xu, B. Li, W. Wang, Rational synthesis of a novel 3,3,5-c polyhedral metal-organic framework with high thermal stability and hydrogen storage capability, J. Mater. Chem. A, 4 (2016) 11630-11634. [12] Z.-M. Sun, B.-P. Yang, Y.-Q. Sun, J.-G. Mao, A. Clearfield, Hydrothermal syntheses, characterizations and crystal structures of three new cadmium (II) amino-diphosphonates: effects of substitute groups on the structures of metal phosphonates, J. Solid State Chem., 176 (2003) 62-68. [13] Z.S. Cai, M. Ren, S.S. Bao, N. Hoshino, T. Akutagawa, L.M. Zheng, Synthetic-method-dependent magnetic relaxation in a cobalt(II) phosphonate chain compound, Inorg. Chem., 53 (2014) 12546-12552. [14] Z.-M. Sun, J.-G. Mao, Z.-C. Dong, Synthesis and structural characterization of three new manganese(II) diphosphonates with layered and double-chain architectures, Polyhedron, 24 (2005) 571-577. [15] Z.-M. Sun, J.-G. Mao, B.-P. Yang, S.-M. Ying, Two new layered inorganic–organic hybrids: intercalation of 1,3,5-benzenetricarboxylate in lead bisphosphonate, Solid State Sci., 6 (2004) 295-300. [16] Z.-M. Sun, J.-G. Mao, Y.-Q. Sun, H.-Y. Zeng, A. Clearfield, Hydrothermal synthesis, characterization and crystal structures of two new layered lead(ii) diphosphonates, New J. Chem., 27 (2003) 1326. [17] R. Vivani, F. Costantino, M. Nocchetti, G.D. Gatta, Structural homologies in benzylamino-N,N-bis methylphosphonic acid and its layered zirconium derivative, J. Solid State Chem., 177 (2004) 4013-4022. [18] H. Li, G.-s. Zhu, X.-d. Guo, F.-x. Sun, H. Ren, Y. Chen, S.-l. Qiu, Synthesis, Structure, and Magnetism of a Highly Thermally Stable Microporous Metal Organodiphosphonate with Reversible Dehydration/Rehydration Behavior, Eur. J. Inorg. Chem., 2006 (2006) 4123-4128. [19] N. Noshiranzadeh, M. Emami, R. Bikas, K. Ślepokura, Synthesis, crystal structure, spectroscopic studies and magnetic behavior of a new diphosphonate-bridged dinuclear copper(II) complex, Polyhedron, 133 (2017) 155-161. [20] S.K. Langley, M. Helliwell, S.J. Teat, R.E. Winpenny, Synthesis and characterisation of cobalt(II) phosphonate cage complexes utilizing carboxylates and pyridonates as co-ligands, Dalton Trans., 41 (2012) 12807-12817. [21] W.-Q. Kan, J.-M. Xu, S.-Z. Wen, L. Yang, Metal ions directed assembly of two coordination polymers based on an organic phosphonate anion and a multidentate N-donor ligand, J. Mol. Struct., 1128 (2017) 513-519. [22] J. Veliscek-Carolan, A. Rawal, V. Luca, T.L. Hanley, Zirconium phosphonate sorbents with tunable structure and function, Micropor. Mesopor. Mat., 252 (2017) 90-104. [23] Q. Li, C. Tian, H. Zhang, J. Qian, S. Du, An alternative strategy to construct Fe(ii)-based MOFs with multifarious structures and magnetic behaviors, CrystEngComm, 16 (2014) 9208-9215. [24] T.T. Wang, M. Ren, S.S. Bao, Z.S. Cai, B. Liu, Z.H. Zheng, Z.L. Xu, L.M. Zheng, The O-P-O bridged Mn2(salen)2 chains showing coexistence of single chain magnet and metamagnet behaviour, Dalton Trans., 44 (2015) 4271-4279. [25] D. Kong, Y. Li, X. Ouyang, A.V. Prosvirin, H. Zhao, J.H. Ross, K.R. Dunbar, A. Clearfield, Syntheses, Structure, and Magnetic Properties of New Types of Cu(II), Co(II), and Mn(II) Organophosphonate Materials:  Three-Dimensional Frameworks and a One-Dimensional Chain

Motif, Chem. Mater., 16 (2004) 3020-3031. [26] X.-m. Meng, L.-s. Cui, X.-p. Wang, X.-y. Zhang, X. Zhang, S.-y. Bi, Syntheses, structural diversity, magnetic properties and dye absorption of various Co(ii) MOFs based on a semi-flexible 4-(3,5-dicarboxylatobenzyloxy)benzoic acid, CrystEngComm, 19 (2017) 6630-6643. [27] D. Pinkowicz, H. Southerland, X.Y. Wang, K.R. Dunbar, Record antiferromagnetic coupling for a 3d/4d cyanide-bridged compound, J. Am. Chem. Soc., 136 (2014) 9922-9924. [28] D. Shao, Y. Zhou, Q. Pi, F.-X. Shen, S.-R. Yang, S.-L. Zhang, X.-Y. Wang, Two-dimensional frameworks formed by pentagonal bipyramidal cobalt(ii) ions and hexacyanometallates: antiferromagnetic ordering, metamagnetism and slow magnetic relaxation, Dalton Trans., 46 (2017) 9088-9096. [29] T. Hamaguchi, M.D. Doud, J. Hilgar, J.D. Rinehart, C.P. Kubiak, Competing ferro- and antiferromagnetic interactions in a hexagonal bipyramidal nickel thiolate cluster, Dalton Trans., 45 (2016) 2374-2377. [30] J. Nishijo, T. Yoshida, M. Enomoto, Structures and intra-dimer ferromagnetic interactions of [MnX2Saloph(NCS)] (X=Cl, Br, I), Polyhedron, 87 (2015) 233-236. [31] M.N. Akhtar, M.A. AlDamen, Y.-C. Chen, M. Sadakiyo, J. Khan, M.-L. Tong, Alkoxo- and carboxylato-bridged hexanuclear copper(II) complex: Synthesis, structure and magnetic properties, Inorg. Chem. Commun., 83 (2017) 49-51. [32] M.R. Saber, K.R. Dunbar, Trigonal bipyramidal 5d-4f molecules with SMM behavior, Chem. Commun. (Camb.), 50 (2014) 2177-2179. [33] W. Yang, H. Yang, Q. Yao, S. Zeng, D. Li, J. Dou, Azametallacrowns with aza18-MC-6 and aza12-MC-4 motifs constructed by phenolicpyrazoles : Syntheses, structures, and magnetic properties, Inorg. Chem. Commun., 82 (2017) 16-19. [34] N.F. Chilton, R.P. Anderson, L.D. Turner, A. Soncini, K.S. Murray, PHI: a powerful new program for the analysis of anisotropic monomeric and exchange-coupled polynuclear d- and f-block complexes, J. Comput. Chem., 34 (2013) 1164-1175. [35] M. Er, R. Ustabas, U. Coruh, K. Sancak, E. Vazquez-Lopez, Synthesis and characterization of some new tetraaldehyde and tetraketone derivatives and X-ray structure of 1,1'-(4,4'-(2-(1,3-bis(4-acetylphenoxy)propan-2-ylidene)propane-1,3-di-yl)bis(oxy )bis(4,1-phenyle ne))diethanone, Int. J. Mol. Sci., 9 (2008) 1000-1007. [36] G.M. Sheldrick, Program for Empirical Absorption Correction of Area Detector Data, SADABS, (1996). [37] H.S. Wang, F.J. Yang, Q.Q. Long, Z.Y. Huang, W. Chen, Z.Q. Pan, Y. Song, Two unprecedented decanuclear heterometallic [MnMnLn] (Ln = Dy, Tb) complexes displaying relaxation of magnetization, Dalton Trans., 45 (2016) 18221-18228. [38] F.A. Mautner, M.S. El Fallah, O. Roubeau, S. Speed, S.J. Teat, R. Vicente, Molecular Copper(II) Complexes Derived from Phosphonoacetic Acid: Crystal Structures and Magnetic Behavior, Eur. J. Inorg. Chem., 2013 (2013) 3483-3490. [39] M. Biswas, G. Pilet, J. Tercero, M. Salah El Fallah, S. Mitra, Synthesis, crystal structure, and magnetic properties of a new tetranuclear Cu(II) Schiff base compound, Inorg. Chim. Acta, 362 (2009) 2915-2920. [40] I.A. Baburin, V.A. Blatov, L. Carlucci, G. Ciani, D.M. Proserpio, Interpenetrating metal-organic and inorganic 3D networks: a computer-aided systematic investigation. Part II [1]. Analysis of the Inorganic Crystal Structure Database (ICSD), J. Solid State Chem., 178 (2005) 2452-2474.

[41] V.A. Blatov, Nanocluster analysis of intermetallic structures with the program package TOPOS, Struct. Chem., 23 (2012) 955-963. [42] S.-F. Tang, L.-J. Li, X.-X. Lv, C. Wang, X.-B. Zhao, Tuning the structure of metal phosphonates using uncoordinating methyl group: syntheses, structures and properties of a series of metal diphosphonates, CrystEngComm, 16 (2014) 7043. [43] S. Khanra, M. Helliwell, F. Tuna, E.J. McInnes, R.E. Winpenny, Synthesis, structural characterisation and magnetic studies of polymetallic iron phosphonate cages, Dalton Trans., (2009) 6166-6174. [44] Y.Y. Zhu, Z.G. Sun, F. Tong, Z.M. Liu, C.Y. Huang, W.N. Wang, C.Q. Jiao, C.L. Wang, C. Li, K. Chen, Hydrothermal synthesis, crystal structures, and luminescent properties of a series of new lanthanide oxalatophosphonates with a layer architecture, Dalton Trans., 40 (2011) 5584-5590. [45] Y.-Q. Guo, S.-F. Tang, B.-P. Yang, J.-G. Mao, Synthesis, crystal structures, and luminescent properties of two types of lanthanide phosphonates, J. Solid State Chem., 181 (2008) 2713-2718. [46] A.K. Boudalis, G. Rogez, B. Heinrich, R.G. Raptis, P. Turek, Towards ionic liquids with tailored magnetic properties: bmim+ salts of ferro- and antiferromagnetic CuII3 triangles, Dalton Trans., 46 (2017) 12263-12273. [47] R. Boča, L.u. Dlháň, G. Mezei, T. Ortiz-Pérez, R.G. Raptis, J. Telser, Triangular, Ferromagnetically-Coupled CuII3−Pyrazolato Complexes as Possible Models of Particulate Methane Monooxygenase (pMMO), Inorg. Chem., 42 (2003) 5801-5803. [48] V. Chandrasekhar, T. Senapati, E.C. Sanudo, R. Clerac, Tri-, tetra-, and hexanuclear copper(II) phosphonates containing N-donor chelating ligands: synthesis, structure, magnetic properties, and nuclease activity, Inorg. Chem., 48 (2009) 6192-6204. [49] Z.G. Lada, C.P. Raptopoulou, V. Psycharis, N. Klouras, A. Escuer, S.P. Perlepes, Bis(di-2-pyridyl ketoximate- O , N , N ′)bis(di-2-pyridyl ketoxime- N , N ′)dicopper(II) diperchlorate: A plausible, weakly ferromagnetically-coupled intermediate in the formation of the neutral, strongly antiferromagnetically-coupled neutral dimer bearing only deprotonated ligands, Inorg. Chem. Commun., 70 (2016) 95-98. [50] Z.Y. Liu, H.A. Zou, Z.J. Hou, E.C. Yang, X.J. Zhao, Structural motif-dependent magnetic diversity observed in three-dimensional tetrazolyl-based MMOFs: synthesis, structures and magnetism, Dalton Trans., 42 (2013) 15716-15725. [51] M.E. Russell, C.S. Hawes, A. Ferguson, M.I.J. Polson, N.F. Chilton, B. Moubaraki, K.S. Murray, P.E. Kruger, Synthesis, structural and magnetic characterisation of iron(ii/iii), cobalt(ii) and copper(ii) cluster complexes of the polytopic ligand: N-(2-pyridyl)-3-carboxypropanamide, Dalton Trans., 42 (2013) 13576-13583. [52] K.L. Harriman, I.A. Kuhne, A.A. Leitch, I. Korobkov, R. Clerac, M. Murugesu, J.L. Brusso, Halide Influence on Molecular and Supramolecular Arrangements of Iron Complexes with a 3,5-Bis(2-Pyridyl)-1,2,4,6-Thiatriazine Ligand, Inorg. Chem., 55 (2016) 5375-5383. [53] J.-X. Yang, Y.-Y. Qin, R.-P. Ye, X. Zhang, Y.-G. Yao, Employing mixed-ligand strategy to construct a series of luminescent Cd(ii) compounds with structural diversities, CrystEngComm, 18 (2016) 8309-8320.

Five new transition metal phosphonates have been synthesized and fully characterized. And their magnetic properties have been quantificationally investigated