Self-Assembly of Antiferromagnetically-Coupled Copper(II) Supramolecular Architectures with Diverse Structural Complexities

The self-assembly of 2,6-diformyl-4-methylphenol (DFMP) and 1-amino-2-propanol (AP)/2-amino-1,3-propanediol (APD) in the presence of copper(II) ions results in the formation of six new supramolecular architectures containing two versatile double Schiff base ligands (H3L and H5L1) with one-, two-, or three-dimensional structures involving diverse nuclearities: tetranuclear [Cu4(HL2−)2(N3)4]·4CH3OH·56H2O (1) and [Cu4(L3−)2(OH)2(H2O)2] (2), dinuclear [Cu2(H3L12−)(N3)(H2O)(NO3)] (3), polynuclear {[Cu2(H3L12−)(H2O)(BF4)(N3)]·H2O}n (4), heptanuclear [Cu7(H3L12−)2(O)2(C6H5CO2)6]·6CH3OH·44H2O (5), and decanuclear [Cu10(H3L12−)4(O)2(OH)2(C6H5CO2)4] (C6H5CO2)2·20H2O (6). X-ray studies have revealed that the basic building block in 1, 3, and 4 is comprised of two copper centers bridged through one μ-phenolate oxygen atom from HL2− or H3L12−, and one μ-1,1-azido (N3−) ion and in 2, 5, and 6 by μ-phenoxide oxygen of L3− or H3L12− and μ-O2− or μ3-O2− ions. H-bonding involving coordinated/uncoordinated hydroxy groups of the ligands generates fascinating supramolecular architectures with 1D-single chains (1 and 6), 2D-sheets (3), and 3D-structures (4). In 5, benzoate ions display four different coordination modes, which, in our opinion, is unprecedented and constitutes a new discovery. In 1, 3, and 5, Cu(II) ions in [Cu2] units are antiferromagnetically coupled, with J ranging from −177 to −278 cm−1.


Introduction
Self-assembly processes use simple building blocks in biological systems for the construction of symmetrical complex supramolecular biomolecules such as proteins, lipoproteins, DNA, glycoproteins, enzymes etc. [1][2][3]. Inspired by nature, in the last few decades, chemists have used the self-assembly methodology [4][5][6] to successfully generate a variety of supramolecular architectures including organic materials [7][8][9], metalacyclic polygons and polyhedrons [10], and nanoscale systems [11,12] with desirable sizes, shapes, and functions. The self-assembly technique, which utilizes a variety of cooperative and noncovalent interactions such as hydrogen bonding, strong electrostatic, and van der Walls forces, π-π stacking, hydrophobic, hydrophilic, and metal-ligand interactions etc., has many advantages over the stepwise synthesis of large supramolecular assemblies. In these processes, the formation of the desired products from the building blocks occur spontaneously and efficiently in one pot. For the spontaneous self-assembly of supramolecular coordination complexes including 1D-, 2D-, and 3D-network structures and grids, appropriate precursor building blocks are reacted in the presence of metal ions as templates [13][14][15][16][17][18][19][20][21][22][23].
The molecular structure of centrosymmetric complex 1 is shown in Figure 2, together with relevant atomic labeling. Important bond distances and angles are listed in Table S1. In complex 1, H 3 L acts as a tetradentate (N 2 O 2 ) dianionic ligand (HL 2− ), binding through two imine nitrogen atoms, a deprotonated alkoxide oxygen in the side arm of the ligand, and a deprotonated phenoxide oxygen, bridging two copper(II) ions into a dinuclear unit. The alkoxy group on one side of the Schiff-base ligand remains protonated and uncoordinated. In each dinuclear unit, two copper(II) ions are bridged through a phenoxide oxygen and a µ-1,1-N 3 bridge. The link between [Cu 2 ] pairs is established via two end-on (EO) µ-1,1-N 3 bridges that form neutral centrosymmetric tetranuclear units, which are linked though remarkably strong H-bonds (2.688 Å) via protonated uncoordinated hydroxyl group (HO(3)) in the side arm of the ligand forming single chains along the a-axis. A perspective view of the polymeric single chains along the a-axis is presented in Figure 3.
Molecules 2020, 25, x 5 of 31 (HO(3)) in the side arm of the ligand forming single chains along the a-axis. A perspective view of the polymeric single chains along the a-axis is presented in Figure 3.  The stereochemistry at Cu(1) in an asymmetric dinuclear unit can best be described as distorted square pyramidal with a phenoxide O, imine N, alkoxide O, and an azido nitrogen atom in the equatorial plane; and an azido nitrogen in the axial position (τ = 0.07); and a square planar geometry at Cu(2) (τ = 0.22) defined by phenoxide O, imine N, and two azido nitrogen atoms in the equatorial plane. (τ is a geometric parameter which is applicable to five coordinate structures as an index of the degree of trigonality). The sum of the angles in the basal plane of Cu(1) and Cu(2) are 359.4(3)° and  (3)) in the side arm of the ligand forming single chains along the a-axis. A perspective view of the polymeric single chains along the a-axis is presented in Figure 3.  The stereochemistry at Cu(1) in an asymmetric dinuclear unit can best be described as distorted square pyramidal with a phenoxide O, imine N, alkoxide O, and an azido nitrogen atom in the equatorial plane; and an azido nitrogen in the axial position (τ = 0.07); and a square planar geometry at Cu(2) (τ = 0.22) defined by phenoxide O, imine N, and two azido nitrogen atoms in the equatorial plane. (τ is a geometric parameter which is applicable to five coordinate structures as an index of the degree of trigonality). The sum of the angles in the basal plane of Cu(1) and Cu(2) are 359.4(3)° and The stereochemistry at Cu(1) in an asymmetric dinuclear unit can best be described as distorted square pyramidal with a phenoxide O, imine N, alkoxide O, and an azido nitrogen atom in the The molecular structure of centrosymmetric complex 2 is comprised of discreate neutral tetranuclear [Cu 4 (L 3− ) 2 (µ 3 -OH) 2 (H 2 O) 2 ] units, and is shown in Figure 4, together with relevant atomic labeling. Important bond distances and angles are listed in Table S2. The coordination mode of H 3 L in 2 is quite different from that in complex 1. In 2, H 3 L utilizes its full coordination potential acting as a pentadentate (N 2 O 3 ) trianionic (L 3− ) ligand by binding through two imine nitrogen, deprotonated phenoxy oxygen, and two deprotonated alkoxy oxygen atoms in the side arms. L 3− holds two Cu(II) ions in close proximity in a dinuclear unit bridged by two single atom bridges: a deprotonated phenoxy oxygen of the ligand, and a hydroxy bridge (µ 3 -OH − ). The two dinuclear units are linked through two µ 3 -OH − ions (Cu-O = 2.361 Å), which in addition to providing an intra-dinuclear bridge also act as an inter-dinuclear bridge, forming neutral tetranuclear units. The molecular structure of centrosymmetric complex 2 is comprised of discreate neutral tetranuclear [Cu4(L 3− )2(μ3-OH)2(H2O)2] units, and is shown in Figure 4, together with relevant atomic labeling. Important bond distances and angles are listed in Table S2. The coordination mode of H3L in 2 is quite different from that in complex 1. In 2, H3L utilizes its full coordination potential acting as a pentadentate (N2O3) trianionic (L 3− ) ligand by binding through two imine nitrogen, deprotonated phenoxy oxygen, and two deprotonated alkoxy oxygen atoms in the side arms. L 3− holds two Cu(II) ions in close proximity in a dinuclear unit bridged by two single atom bridges: a deprotonated phenoxy oxygen of the ligand, and a hydroxy bridge (μ3-OH − ). The two dinuclear units are linked through two μ3-OH − ions (Cu-O = 2.361 Å), which in addition to providing an intra-dinuclear bridge also act as an inter-dinuclear bridge, forming neutral tetranuclear units.   (14) • respectively, indicating a significant distortion from planarity and a strong pyramidal distortion respectively.
H 5 L1 is a potentially heptadentate penta-anionic double Schiff base ligand. Only one tetranuclear Ni 2+ complex of this ligand has been reported [43]. In this publication, we are presenting the results of our investigation on the coordination versatility of this ligand towards copper(II) ions and the effect of the anions on the coordination ability of the ligand and the structural complexity. Reactions of copper(II) ions with H 5 L1 under varied conditions produce complexes of diverse nuclearities including dinuclear (3), heptanuclear (5), decanuclear (6), and polynuclear (4). In the dinuclear compound  Figure 6).
The relevant bond distances and angles are listed in Table S3. The stereochemistry at each copper(II) ion can best be described as a distorted square pyramidal (τ (Cu (1) Figure  6).  The relevant bond distances and angles are listed in Table S3. The stereochemistry at each copper(II) ion can best be described as a distorted square pyramidal (τ (Cu (1) (4) The molecular structure of a dinuclear unit in complex 4 is shown in Figure 7, together with relevant atomic labeling. Important bond distances and angles are listed in Table S4. The structure of

{[Cu
The molecular structure of a dinuclear unit in complex 4 is shown in Figure 7, together with relevant atomic labeling. Important bond distances and angles are listed in Table S4 7)) water molecules generating 2D sheets along bc axis which are further cross linked to produce an interesting 3D supramolecular structure ( Figure 8). In complex 4, the coordination mode of H 5 L1 is identical to that present in 3. H 5 L1 acts as pentadentate (N 2 O 3 ) dianionic ligand (H 3 L 2− ), coordinating via two imine nitrogen atoms, a deprotonated phenoxide oxygen, a deprotonated alkoxide oxygen, and a protonated ethanolic OH group in the side arm of the ligand thereby bridging two copper(II) ions into dinuclear units. The second alkoxy group on either side of the Schiff-base ligand remains protonated and uncoordinated. As in compounds 1 and 3, the two Cu(II) ions in each dinuclear unit are bridged via a phenoxide oxygen and a µ-1,1-N 3 bridge.
Molecules 2020, 25, x 9 of 31 the ethanol OH group in the side arm of the Schiff base ligand/water at the axial position. There is a weak axial contact of Cu(2) with F(4) of BF4 − , thus giving a distorted octahedral geometry at Cu (2). The sum of the angles in the basal plane of Cu(1) and Cu (2)     Molecules 2020, 25, x 9 of 31 the ethanol OH group in the side arm of the Schiff base ligand/water at the axial position. There is a weak axial contact of Cu(2) with F(4) of BF4 − , thus giving a distorted octahedral geometry at Cu (2). The sum of the angles in the basal plane of Cu(1) and Cu (2)     The stereo-chemical environment at Cu(1) and Cu(2) can best be described as distorted square pyramidal (τ = 0.12), and distorted octahedral respectively. The coordination geometry in the basal plane of each copper(II) ion is defined by a phenoxide O-atom, (O(1)), an imine N-atom, (N(1)/N(2)), an alkoxide O-atom, (O(4)/O(3)), and azido nitrogen atoms (N(3)), with oxygen atom (O(5)/O(6)) of the ethanol OH group in the side arm of the Schiff base ligand/water at the axial position. There is a weak axial contact of Cu(2) with F(4) of BF 4 − , thus giving a distorted octahedral geometry at Cu (2).
In complex 5, two dinuclear [Cu 2 H 3 L1 2− ] units are connected to three copper(II) ions which are bonded to benzoate ions in a heptanuclear associated arrangement. In this complex, the benzoate ions exhibit four different types of bridging modes including (µ 4 -1,1,3,3-C 6 H 5 CO 2 ), (µ 3 -1,1,3-C 6 H 5 CO 2 ), (µ-1,3-C 6 H 5 CO 2 ), and (µ-1,1-C 6 H 5 CO 2 ), which is unprecedented. In our opinion, this constitutes the first report of a copper(II) complex in which benzoate ions exhibit four different types of bridging modes. In complex 5, H 5 L1 acts as hexadentate (N 2 O 4 ) dianionic ligand (H 3 L1 2− ) binding through two imine nitrogen atoms, a deprotonated phenoxide oxygen, and a deprotonated alkoxide oxygen, and two protonated ethanol (OH) groups in the side arms of the double Schiff base ligand, bridging two copper(II) ions into a dinuclear unit which is different from that in complexes 3 and 4 (pentadentate (N 2 O 3 )). One alkoxy group in one side arm of the Schiff-base ligand remains protonated and uncoordinated. In each dinuclear unit, two copper(II) ions are bridged through a phenoxide oxygen and µ 3 -O 2− bridges. Two dinuclear [Cu 2 ] units are connected to three Cu 2+ ions which are held in place by µ 3 -O 2− , alkoxide O in the side arm of the ligand, and bridging benzoate ions that produce an interesting heptacopper structure. A perspective view of the 5 is presented in Figure 9. Important distances and angles are listed in Table S5.
The stereochemistry at Cu(1)/Cu (7) ion in each dinuclear unit can best be described as a distorted square pyramidal (τ (Cu(1) = 0.10 and Cu (7)  that in complexes 3 and 4 (pentadentate (N2O3)). One alkoxy group in one side arm of the Schiff-base ligand remains protonated and uncoordinated. In each dinuclear unit, two copper(II) ions are bridged through a phenoxide oxygen and μ3-O 2− bridges. Two dinuclear [Cu2] units are connected to three Cu 2+ ions which are held in place by μ3-O 2− , alkoxide O in the side arm of the ligand, and bridging benzoate ions that produce an interesting heptacopper structure. A perspective view of the 5 is presented in Figure 9. Important distances and angles are listed in Table S5.  ions in another dinuclear unit are bridged by a phenoxide oxygen and a µ 3 -O 2− ion, which acts as an intradinuclear bridge as well as a link between a dinuclear unit and a trinuclear unit forming the decanuclear supramolecular architecture 6. A perspective view of the 6 is presented in Figure 10 and important distances and angles are listed in Table S6.  15)) in the side chains of the ligands (see Figure 11).

Magnetic Properties
In the compounds under investigation, the Cu-Cu distances are quite short. The copper ions in the dinuclear units are bridged via single atom-phenoxy oxygen atom of the ligand and end-on azido μ-1,1-N3/hydroxy (μ-OH − )/oxide (μ-O 2− ) bridges and are likely to result in magnetic exchange interactions between closely placed metal centers. Variable temperature magnetic studies have been carried out on 1, 3, and 5, and the results of our investigations are presented as the best fit curves along with experimental data in the Figures 12-14 respectively. Based on the structural information, we anticipate the presence of strong antiferromagnetic spin exchange interactions via the PhO − and N3 − /hydroxy (μ-OH − )/oxide (μ-O 2− ) bridges within the [Cu2] units in these complexes, which involve all equatorial positions of the metals and bridging units.
Despite the butterfly, cubane type structure of complex 1, the basic arrangement is comprised of two almost planar dinuclear fragments joined axially through long (2.4 Å) and a very long (3.02 Å) axial contacts. Theoretically, these axial contacts are orthogonal and so contribute little to overall antiferromagnetic exchange [114]. Looking at the Cu-O-Cu and Cu-N-Cu angles, one would expect net antiferromagnetic (AF) exchange, as is observed experimentally. The best fit to a dinuclear model is not brilliant, and gives g = 2.03, J = −278 cm −1 , temperature independent magnetism (TIP) = 445 × Figure 11. Perspective view of a portion of 1D-single chain along a-axis in the structure of 6.
For clarity, the structure of 5 showing only the metal centers and the coordinating atoms is shown in Figure 15 [115].
For clarity, the structure of 5 showing only the metal centers and the coordinating atoms is shown in Figure 15  Despite the butterfly, cubane type structure of complex 1, the basic arrangement is comprised of two almost planar dinuclear fragments joined axially through long (2.4 Å) and a very long (3.02 Å) axial contacts. Theoretically, these axial contacts are orthogonal and so contribute little to overall antiferromagnetic exchange [114]. Looking at the Cu-O-Cu and Cu-N-Cu angles, one would expect net antiferromagnetic (AF) exchange, as is observed experimentally. The best fit to a dinuclear model is not brilliant, and gives g = 2.03, J = −278 cm −1 , temperature independent magnetism (TIP) = 445 × 10 −6 cm 3 mol −1 , and fraction paramagnetic impurity (ρ) = 0.003, 10 2 R = 9.  [115], J calc = −332 cm −1 , it is possible that the azide is responsible for a small ferromagnetic contribution, which would agree with our azide correlation (vide supra) [116,117].
Complex 3 contains a simple dinuclear unit with two in plane active bridges, both connecting the d x 2y 2 metal magnetic orbitals. The fit is good, indicating overall AF coupling, and gives g = 2.13 (1) [115].
Complex 5 breaks down nicely into two isolated parts both expected to be AF. The fit assumes that all J values are the same, which is not unreasonable given the bridges and the Cu-O-Cu angles. The benzoates are not influencing exchange in any significant way.
For clarity, the structure of 5 showing only the metal centers and the coordinating atoms is shown in Figure 15.

Magneto-Structural Relationships
In doubly bridged [Cu2(μ-OPh)(μ-1,1-N3)] copper(II) complexes, the nature (ferromagnetic/ antiferromagnetic) and the magnitude of the magnetic spin exchange interaction (J) depends primarily on the bridge angles, but other important factors such as the intermetallic distance (d), bond distance in the equatorial plane, stereochemistry, and distortion from planarity in the mean plane of

Magneto-Structural Relationships
In doubly bridged [Cu 2 (µ-OPh)(µ-1,1-N 3 )] copper(II) complexes, the nature (ferromagnetic/ antiferromagnetic) and the magnitude of the magnetic spin exchange interaction (J) depends primarily on the bridge angles, but other important factors such as the intermetallic distance (d), bond distance in the equatorial plane, stereochemistry, and distortion from planarity in the mean plane of dinuclear core can also influence the magnitude of the coupling constant (J) [118]. In order to illustrate the magneto-structural trends, we have compiled the magnetic data of all the copper(II) complexes (Table 1) from the literature that contain endogenous phenoxide bridge and exogenous EO µ-azido bridge along with two new compounds (1 and 3) reported in this study. The relationships between the antiferromagnetic coupling constant (−J) and phenoxide bridge angle (Cu-PhO-Cu), average bridge angles of µ-phenoxide bridges and µ-azido bridges, and the Cu-Cu distance (d) are represented in Figures 16-18 respectively, and summarized in Table 1. large angle (105.75°) with the exception of compound 3 in Table 1 where the average bridge angle is 103.20° and J = −512 cm −1 . While these plots show realistic trends, with dominant linear character, it is necessary to stress that the J values are based on the sum of two counter-complementary exchange contributions, where the individual bridges have linear variations with angle, which are different. This helps to explain why the general appearance of the averaged data plotted in Figure 17 look more linear than those in Figure 16.  Figure 18 summarizes the trend in exchange integral as a function of Cu-Cu distance listed in Table 1. A reasonably linear relationship is evident for majority of the complexes. This agrees with the expected increase in both bridge angles, resulting in an increase of the antiferromagnetic contribution as reported previously [36,41,115,116,129,130]. angle in dinuclear (μ-phenolate/μ-azido bridged) copper(II) complexes. Figure 18 summarizes the trend in exchange integral as a function of Cu-Cu distance listed in Table 1. A reasonably linear relationship is evident for majority of the complexes. This agrees with the expected increase in both bridge angles, resulting in an increase of the antiferromagnetic contribution as reported previously [36,41,115,116,129,130].

Powder X-ray Diffraction Studies
In an attempt to characterize the bulk powder, the XRD patterns of 1, 3 and 5 were collected. (Figures S3-S5). The XRD patterns were collected and compared to the calculated pattern generated from the single-crystal X-ray structure [131]. For the two complexes (1 and 3, in Table 1, 1 and 2) reported in this study, a strong antiferromagnetic interaction (−J = 278, 177.3 cm −1 respectively) occurs within the dinuclear [Cu 2 (µ-OPh)(µ-1,1-N 3 )] core where the Cu-N and Cu-O distances in the equatorial plane fall in the ranges of 1.915-1.980 Å and 1.9341-1.998 Å respectively. These are quite short and are within the plane of the magnetic orbitals of both metals (dx 2 -y 2 ) which are approximately parallel, and are responsible for effective coupling between copper(II) centers in each dinuclear unit. Based on the bridge angles, it is anticipated that the bridging moieties (µ-phenolato and µ-azido) provide counter complementary contributions to the magnetic exchange interaction between the copper(II) centers in [Cu 2 (OPh)(µ-1,1-N 3 )] core [127,128].
The phenoxide bridge angle of 101.88 • (1)/100.80 • (3) and the µ-azido bridge angles of 102.4 • (1)/100.07 • (3) in these complexes are expected to mediate antiferromagnetic and ferromagnetic spin coupling respectively between the copper centers, based on previous studies [111,118]. This is consistent with the reported data for µ-phenolato/µ-azido bridged copper(II) complexes presented in Figures 16 and 17. It is a well-established fact that for bis(µ-phenolato) bridged copper(II) complexes if the bridge angle is less than the critical angle of~97 • /98 • , the spin coupling constant (J) between the copper centers is dominantly ferromagnetic, and for larger angles an antiferromagnetic interaction is expected [36,129]. For dinuclear complexes involving a µ-azido bridge, it has been established that the nature of the spin coupling constant (J) is dependent on the Cu-(µ-N 3 )-Cu bridge angle, and that the magnitude of ferromagnetic coupling (J ferro ) decreases with increasing bridge angle (critical angle 104º (according to theoretical calculation) [130] or 108 • (experimental studies) [115,116]).
The antiferromagnetic spin coupling constant (−J) plotted against the phenolate bridge angle and the averaged bridge angle of phenolate and azido for all the reported copper(II) complexes is shown in Figures 16 and 17 respectively. Figure 16 shows a relationship between −J and phenolate bridge angles with reasonable linear character. A graph of −J vs. averaged bond angles ( Figure 17) Table 1 where the average bridge angle is 103.20 • and J = −512 cm −1 . While these plots show realistic trends, with dominant linear character, it is necessary to stress that the J values are based on the sum of two counter-complementary exchange contributions, where the individual bridges have linear variations with angle, which are different. This helps to explain why the general appearance of the averaged data plotted in Figure 17 look more linear than those in Figure 16. Figure 18 summarizes the trend in exchange integral as a function of Cu-Cu distance listed in Table 1. A reasonably linear relationship is evident for majority of the complexes. This agrees with the expected increase in both bridge angles, resulting in an increase of the antiferromagnetic contribution as reported previously [36,41,115,116,129,130].

Powder X-ray Diffraction Studies
In an attempt to characterize the bulk powder, the XRD patterns of 1, 3 and 5 were collected. (Figures S3-S5). The XRD patterns were collected and compared to the calculated pattern generated from the single-crystal X-ray structure [131]. For 3 and 5, the form of the diffraction curve for the observed pattern was similar to that of the calculated pattern. There were minor differences (i.e., intensity variations, changes in peak full-width, and peak position) between the calculated and observed peaks. Peak shifts are an artifact, given that the powder data were collected at room temperature, while the calculated pattern was based on structural data from −100 • C. This difference may change the unit cell dimensions and shift peak positions along the 2θ axis. For 1, peak differences may result from the sample being grinded prior to characterizing. Mechanical grinding could alter the crystallite structure, possibly through the loss of solvent in the lattice. Although the XRD powder pattern shows consistency to that of the calculated powder pattern, the measurement also does not reveal amorphous content that may be present within the sample. Hence, the XRD powder data are not useful for commenting on the purity of the crystalline phase(s) present.

Physical Measurements
Infrared spectra were recorded as Nujol mulls using a Perkin Elmer FT-IR instrument, and Uv/Vis spectra of the powdered compounds were obtained as Nujol mulls or in solution using a Cary 5E spectrometer. Micro-analyses were carried out using a Leco CHNS-Analyzer. Variable temperature magnetic data (2-300 K) were obtained using a Quantum Design MPMS5S SQUID magnetometer with a field strength 0.1 T. Background corrections for the sample holder assembly and diamagnetic components of the complexes were applied. X-ray powder patterns were collected using a Rigaku Miniflex 600 X-ray Diffractometer. The radiation used was Cu Kα radiation (λ = 1.54059 Å).

Synthesis of the Coordination Complexes
Caution: Azide and perchlorate complexes of metal ions involving organic ligands are potentially explosive. Only small quantities of the complexes should be prepared, and these should be handled with care.
In some cases, there is a difference between the most reasonable formula based on the elemental analysis (analytical formula) and that obtained from X-ray crystallography. In these compounds the CHN analysis showed a different number of solvent molecules (methanol and water) compared with the X-ray formulae, as the analysis was carried out on air dried samples due to their potential explosive nature. For consistency, the X-ray formulae will be used in the discussion. For compound 1, the X-ray formula is [Cu 4 (HL 2− ) 2  First, 1-Amino-2-propanol (AP) (0.08 g, 1.0 mmol) dissolved in 3 mL of methanol was added dropwise to a solution of 2,6-diformyl-4-methylphenol (DFMP, 0.09 g, 0.50 mmol) in hot methanol (10 mL) while stirring under reflux. The yellow solution formed was refluxed for 30 min and a solution of Cu(BF 4 ) 2 ·6H 2 O (1.0 mmol, 0.35 g) in methanol (5 mL) was added dropwise. The reaction mixture (green) was refluxed for 10 min, and a solution of NaN 3 (0.07 g, 1.0 mmol) in a hot methanol (10 mL) was added dropwise. The color of the reaction mixture changed to dark green and it was refluxed further for 2.0 h. The green solution was filtered hot, and the filtrate was kept unperturbed at room temperature for slow evaporation. After two weeks, dark green crystals suitable for X-ray studies were obtained and some were kept in the mother liquor for X-ray analysis. The bulk sample was separated from the mother liquor and washed with methanol (2 × 2 mL) and air dried at ambient temperature. IR spectrum: 3423 cm −1 (υ(OH) H 2 O and CH 3 OH), 2093, 2037 cm −1 (υas (N 3 )), 1637 cm −1 (υ(C=N)). UV-Vis Spectrum: 330 nm (s), 370 nm (sh) (Cu-azide and Cu-ligand charge transfer transitions respectively), and 625 nm (d-d transition). Yield: 0.14 g, 48%, based on DFMP. First, 2,6-Diformyl-4-methylphenol (DFMP, 0.09 g, 0.50 mmol) dissolved in hot methanol (10 mL) was added to a solution of 2-amino-1,3-propanediol (APD) (0.09 g, 1.0 mmol) in the same solvent (5 mL). The yellow solution of the Schiff-base ligand (H 5 L1) formed was stirred under reflux for 30 min, and a solution of Cu(NO 3 ) 2 ·3H 2 O (0.24 g, 1.0 mmol) in methanol (5 mL) was added to it dropwise. The solution changed from brown to green in about 5 min. The resulting green solution was refluxed for 10 min and a solution of NaN 3 (0.070 g, 1.0 mmol) in hot methanol (10 mL) was added dropwise. The color of the reaction mixture changed to dark green and was refluxed further for 1.5 h. A clear green solution was filtered hot and the filtrate was left undisturbed at ambient temperature for slow evaporation. After four weeks, dark green crystals suitable for x-ray analysis were formed, separated from the mother liquor, and washed with methanol (2 × 2 mL). IR spectrum: 3392, 3322 cm   (4) Compound 4 was obtained in a similar manner as compound 3. In this case after adding the NaN 3 solution to the reaction mixture of DFMP (0.50 mmol), 2-amino-1,3-propanediol (1.0 mmol), and Cu(BF 4 ) 2 ·6H 2 O (0.35 g, 1.0 mmol), the mixture was further refluxed for 2.0 h and left at room temperature undisturbed for slow evaporation. After one week, very nice crystals suitable for X-ray studies separated from the dark green solution. The crystals used for X-ray studies were kept in the mother liquor. The remaining crystals were separated and washed with methanol (2 × 2 mL). First, 2,6-Diformyl-4-methylphenol (0.17 g, 1.0 mmol) dissolved in hot methanol (15 mL) was added to a solution of 2-amino-1,3-propanediol (APD) (0.18 g, 2.0 mmol) in methanol (10 mL). The yellow solution of the Schiff-base ligand (H 5 L1) was then refluxed for 30 min and Cu(ClO 4 ) 2 ·6H 2 O (0.92 g, 0.25 mmol) dissolved in hot methanol (10 mL) was added to it dropwise with stirring under reflux. The bright green solution formed was refluxed further for 10 min and a solution of sodium benzoate (C 6 H 5 CO 2 Na) (0.30 g, 2.0 mmol) in hot methanol (15 mL) was added dropwise. After refluxing the reaction mixture for 10 min, a solution of triethylamine (0.20 g, 2.0 mmol) dissolved in 5 mL of methanol was added dropwise, which caused a color change of the reaction mixture to brownish green. It was stirred under reflux for 2.0 h and filtered hot. The filtrate was left unperturbed at ambient temperature for slow evaporation. After three weeks some colorless crystals, which were possibly of sodium benzoate, separated and were filtered off. 5 mL of ethanol and 2 mL of water was added to the filtrate and left at room temperature for slow evaporation. After two weeks green crystals suitable for X-ray analysis formed and were kept in the mother liquor. The crystals of the bulk sample were separated from the mother liquor and washed with methanol (2 × 2 mL). IR spectrum:  4 (O) 2 (OH) 2 (C 6 H 5 CO 2 ) 4 ](C 6 H 5 CO 2 H) 2 ·20H 2 O (6) Complex 6 was prepared by using the same method as used for 5 by reacting Cu(NO 3 ) 2 ·3H 2 O (0.58 g, 3 mmol) with the Schiff base prepared by reacting 2,6-diformyl-4-methyphenol (0.17 g, 1 mmol) and 2-amino-1,3-propanediol (0.18 g, 2.0 mmol), C 6 H 5 CO 2 Na (0.30 g, 2.0 mmol), and triethylamine (0.

X-ray Crystallography
Suitable single crystals for X-ray diffraction studies were obtained for 1-6. Crystal data for the compounds were collected by the same method by mounting a crystal onto a thin glass fiber from a pool of Fluorolube TM and immediately placing it under a liquid N 2 cooled stream on a Bruker AXS diffractometer upgraded with an APEX II CCD detector. The radiation used was graphite monochromatized Mo Kα radiation (λ = 0.7107 Å). The lattice parameters were optimized from a least-squares calculation on carefully centered reflections. Lattice determination, data collection, structure refinement, scaling, and data reduction were carried out using APEX3 Version 2018.11 software package [133,134]. The data were corrected for absorption using the SCALE program within the APEX3 software package [133,134]. The structures were solved using SHELXT [135]. This procedure yielded a number of the C, N, Cu, O, F and B atoms. Subsequent Fourier synthesis yielded the remaining atom positions. The hydrogen atoms were fixed in positions of ideal geometry (riding model) and refined within the XSHELL software package [136]. The final refinement of each compound with anisotropic thermal parameters on all nonhydrogen atoms was performed using OLEX2-1.2 [137]. The crystal data for compounds 1-6 are given in Table 2. Crystallographic data for the structures has been deposited with the Cambridge Crystallographic Data Centre as supplementary publication nos: CCDC 1944278-1944283. Copies of the data can be obtained, free of charge on application to CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax, +44-(0)1223-336033; or e-mail, deposit@ccdc.cam.ac.uk).

Conclusions
We reported the coordination versality of two double Schiff base ligands H 3 L (potentially pentadentate (N 2 O 3 ) tri-anionic) and H 5 L1 (potentially heptadentate (N 2 O 5 ) penta-anionic) with a high degree of conformational flexibility, having one or two ethanoate hydroxy groups in the side arms and a potential to coordinate in a convergent and a divergent fashion with Cu(II) ions. Based on the reaction conditions, the nature of the anion, and the stereochemical requirements of the metal, these ligands were shown to exhibit diverse coordination versatility. In complex 1, H 3 L acts as tetradentate (N 2 O 2 ) dianionic ligand (HL 2− ), whereas in 2, it acts as a pentadentate (N 2 O 3 ) tri-anionic (L 3− ) ligand, holding two Cu(II) ions in close proximity for magnetic exchange interaction. Reactions of copper(II) ions with H 5 L1 under varied conditions resulted in the formation of dinuclear (3), polynuclear (4), heptanuclear (5), and decanuclear (6) complexes, depending upon the anions. This clearly demonstrated the significant effect of the nature of an anion on the nuclearity of the complex produced. In dinuclear complex 3 and polynuclear complex 4, which grew into very fascinating 3D-network structures through H-bonding, and in decanuclear complex 6, H 5 L1 acted as a pentadentate (N 2 O 3 ) dianionic dinculeating ligand