Ni ( II ) Dimers of NNO Donor Tridentate Reduced Schiff Base Ligands as Alkali Metal Ion Capturing Agents : Syntheses , Crystal Structures and Magnetic Properties

Three trinuclear Ni(II)-Na(I) complexes, [Ni2(L)2NaCl3(H2O)]·H2O (1), [Ni2(L)2NaCl3 (H2O)] (2), and [Ni2(L)2NaCl3(OC4H10)] (3) have been synthesized using three different NNO donor tridentate reduced Schiff base ligands, HL1 = 2-[(3-methylamino-propylamino)-methyl]-phenol, HL2 = 2-[(3-methylamino-propylamino)-methyl]-4-chloro-phenol, and HL3 = 2-[(3-methylaminopropylamino)-methyl]-6-methoxy-phenol that had been structurally characterized. Among these complexes, 1 and 2 are isostructural in which dinuclearNi(II) units act as metalloligands to bind Na(I) ions via phenoxido and chlorido bridges. The Na(I) atom is five-coordinated, and the Ni(II) atom possesses hexacordinated distorted octahedral geometry. In contrast, in complex 3, two -OMe groups from the dinuclear Ni(II) unit also coordinate to Na(I) to make its geometry heptacordinated pentagonal bipyramidal. The magnetic measurements of complexes 1–3 indicate ferromagnetic interactions between dimeric Ni(II) units with J = 3.97 cm−1, 4.66 cm−1, and 5.50 cm−1 for 1–3, respectively, as is expected from their low phenoxido bridging angles (89.32◦, 89.39◦, and 87.32◦ for 1–3, respectively). The J values have been calculated by broken symmetry DFT method and found to be in good agreement with the experimental values.


Introduction
During the past few decades, synthesis and characterizations of polynuclear transition metal complexes and their magnetic and electronic properties have received much attention from chemists due to their potential applications in magnetic ordering and catalytic and biological mimicking [1][2][3][4][5][6][7][8][9].A common strategy for the synthesis of such polynuclear complexes is to bind the metal centers by a single atom bridge.The Schiff bases derived from diamines and salicylaldehyde derivatives are very useful for this purpose, as the phenoxido oxygen atom can connect two or three metal centers very efficiently.The literature shows that exploiting this property, various N 2 O 2 or N 2 O 4 donor Schiff base ligands have been used widely for the synthesis of 3d-3d , 3d-4f/5f, and 3d-ns heterometallic complexes [10][11][12][13][14][15][16].On the contrary, N 2 O donor tridentate Schiff base ligands are known to form mostly homonuclear complexes [17][18][19].As, for example, with Ni(II), these N 2 O donor ligands form diphenoxido bridged dinuclear complexes [20][21][22].These complexes attracted significant attention due to their importance in the study of magnetic coupling between the metal centers through the phenoxido bridges [23][24][25].It is now well understood that the ferro-to antiferromagnetic crossover angle for diphenoxido bridged Ni(II) is 93-94 • and most of these reported complexes are antiferromagnetically coupled, because they are stable at bond angles higher than the crossover angle [26][27][28][29].It is also known that the ferromagnetically coupled complexes with the lower Ni-O-Ni bridging angle can be obtained by introducing an additional bridging atom between the Ni(II) centers in the complexes of NNO donor ligands [23,[30][31][32][33].However, to date, unlike the mononuclear complexes of N 2 O 2 or N 2 O 4 donor Schiff base ligands, the dinuclear complexes of these NNO donor ligands are not known to act as metalloligands when coordinating with a second metal ion.
Herein, we report the synthesis of three new heterometallic Ni(II)-Na(I) complexes, [Ni 2 (L 1  The complexes are characterized by single crystal X-ray crystallography, electronic spectra, IR spectra and elemental analyses.In 1 and 2, the sodium ion is attached to the dinuclear Ni(II) complex with the help of phenoxido and chlorido bridges, whereas in 3, two -OMe groups form additional bridges between the dinuclear unit and Na(I).The magnetic properties of the complexes have been studied both experimentally and using broken symmetry DFT method.The couplings are found to be ferromagnetic, and consistent with the low phenoxido bridging angles between the NI(II) centers.The experimentally obtained coupling parameters match well with the theoretically calculated values.

Syntheses of the Complexes
Three reduced tridentate Schiff base ligands, HL 1 , HL 2 , and HL 3 , were synthesized by condensation of N-methyl-1,3-propanediamine with salicylaldehyde, 5-chlorosalicylaldehyde, and 2-hydroxy-3-methoxybenzaldehyde in 1:1 molar ratios and subsequent reduction with sodium borohydride in methanol.We obtained complexes (1-3) accidentally in low yield when the methanolic solutions of these as-synthesized ligands were reacted directly with NiCl 2 •6H 2 O after reduction.The sodium ions that were present in the solution (from sodium borohydride) were captured by the dinuclear complexes.Later, in order to improve the yield, we isolated the ligands and reacted them with nickel(II) chloride in 1:1 molar ratios in the presence of sodium chloride in methanol-water solvent mixture (vide experimental section, Scheme 1).
Magnetochemistry 2018, 4, x FOR PEER REVIEW 2 of 14 due to their importance in the study of magnetic coupling between the metal centers through the phenoxido bridges [23][24][25].It is now well understood that the ferro-to antiferromagnetic crossover angle for diphenoxido bridged Ni(II) is 93-94° and most of these reported complexes are antiferromagnetically coupled, because they are stable at bond angles higher than the crossover angle [26][27][28][29].It is also known that the ferromagnetically coupled complexes with the lower Ni-O-Ni bridging angle can be obtained by introducing an additional bridging atom between the Ni(II) centers in the complexes of NNO donor ligands [23,[30][31][32][33].However, to date, unlike the mononuclear complexes of N2O2 or N2O4 donor Schiff base ligands, the dinuclear complexes of these NNO donor ligands are not known to act as metalloligands when coordinating with a second metal ion.Herein, we report the synthesis of three new heterometallic Ni(II)-Na(I) complexes, [Ni2(L 1 )2NaCl3(H2O)]·H2O (1), [Ni2(L 2 )2NaCl3(H2O)] (2), and [Ni2(L 3 )2NaCl3(OC4H10)] (3), by using three different NNO donor tridentate reduced Schiff bases ligands (HL 1 = 2-[(3-methylamino-propylamino)methyl]-phenol, HL 2 = 2-[(3-methylamino-propylamino)-methyl]-4-chloro-phenol, and HL 3 = 2-[(3methylamino-propylamino)-methyl]-6-methoxy-phenol).The complexes are characterized by single crystal X-ray crystallography, electronic spectra, IR spectra and elemental analyses.In 1 and 2, the sodium ion is attached to the dinuclear Ni(II) complex with the help of phenoxido and chlorido bridges, whereas in 3, two -OMe groups form additional bridges between the dinuclear unit and Na(I).The magnetic properties of the complexes have been studied both experimentally and using broken symmetry DFT method.The couplings are found to be ferromagnetic, and consistent with the low phenoxido bridging angles between the NI(II) centers.The experimentally obtained coupling parameters match well with the theoretically calculated values.

Syntheses of the Complexes
Three reduced tridentate Schiff base ligands, HL 1 , HL 2 , and HL 3 , were synthesized by condensation of N-methyl-1,3-propanediamine with salicylaldehyde, 5-chlorosalicylaldehyde, and 2hydroxy-3-methoxybenzaldehyde in 1:1 molar ratios and subsequent reduction with sodium borohydride in methanol.We obtained complexes (1-3) accidentally in low yield when the methanolic solutions of these as-synthesized ligands were reacted directly with NiCl2•6H2O after reduction.The sodium ions that were present in the solution (from sodium borohydride) were captured by the dinuclear complexes.Later, in order to improve the yield, we isolated the ligands and reacted them with nickel(II) chloride in 1:1 molar ratios in the presence of sodium chloride in methanol-water solvent mixture (vide experimental section, Scheme 1).Scheme 1. Synthesis of complexes 1-3.

IR and UV-Vis Spectra of the Complexes
Complexes 1-3 show a moderately strong, sharp peak at 3266 cm −1 , 3270 cm −1 , and 3282 cm −1 , respectively, due to N-H stretching vibration, indicating that the imine group(C=N) of the Schiff base ligands is reduced.Absence of any sharp peak at 1620-1650 cm −1 for stretching vibration of imine group(C=N) also indicates that Schiff base is reduced (Figures S1-S3).The electronic spectra (Figure S4) of complexes 1, 2, and 3 are recorded in methanolic solution.The spectra show bands at 925.2 and 644.7 nm for 1, 933.9 and 641.0 nm for 2, and 973.7 and 638.7 nm for 3, which can be assigned to the spin-allowed transitions 3 T 1g (F)← 3 A 2g and 3 T 2g ← 3 A 2g , respectively.

Description of the Crystal Structures
The structures of complexes 1 and 2, both with the formula [Ni 2 (L 1−2 ) 2 NaCl 3 (OH 2 )], as shown in Figure 1 (A for 1 and B for 2), contain crystallographic C 2 symmetry and have very similar structures.Each nickel atom is six-coordinated with a distorted octahedral environment.In the equatorial plane are three donor atoms, O (11), N (19), and N(23), of one ligand together with the oxygen atom O(11)$1 ($1=1−x, y, 3/2−z) of a second ligand.In axial positions there are two chlorine atoms, namely, Cl(2) on the two-fold axis, which bridges the two nickel atoms; and Cl (1), which bridges to a sodium ion, also on a two-fold axis.The sodium ion is additionally bonded to two O (11)  Å in 1 and 2, respectively.In both structures, the water oxygen atom O(1) had high thermal parameters and the hydrogen atoms could not be located.In 1, the atom was refined with reduced occupancy, as was an adjacent solvent oxygen atom.For both structures, there is a hydrogen bond from N (19) to Cl(2) ($2 = 1 2 −x,  1).The Ni . . .Ni and Ni . . .Na distances are 2.963(1), 2.966(1) Å and 3.168(2), and 3.186(2) Å in 1 and 2, respectively.In contrast, the structure of 3 that is shown in Figure 2 is different in that the ligand contains an -OMe group that is bonded to the sodium ion.However, the molecule is also located on a two-fold axis, and the arrangement around the two nickel atoms is similar to that found in 1 and 2. The only significant difference in the bond lengths around nickel is to be found in Ni-O(11)$1 ($1 = 1−x, y, 3/2−z), which is considerably longer at 2.3178(11) Å. However the sodium ion, also on a two-fold axis, is now seven-coordinated, with a distorted pentagonal bipyramidal structure being bonded to 2*O( 11 S1.

Magnetic Properties
The variation of the product of molar magnetic susceptibility (χM) and temperature (T) with respect to temperature (T) for Ni II dimers 1, 2, and 3 is presented in Figures 3-5.It can be seen that the experimental χMT values for three Ni II dimers at room temperature (~300 K) lie around 2.24-2.73cm 3 Kmol -1 , which in good agreement with the expected value for two non-interacting Ni II , S = 1 centers (the theoretical spin-only value is 2.0 cm 3 KmoL -1 , g = 2).For all the complexes, variation of χMT values shows similar type of behavior on lowering the temperature: the χMT values gradually increase upon lowering of temperatures and attain a maximum at ca. 10-12 K. On further decreasing the temperature, a steep fall of the values is observed.This type of behavior is typical of a system that shows dominant intramolecular ferromagnetic exchange coupling, and the steep drop in χMT values  Table 1.Hydrogen-Bond Parameters in Complexes (in Å and deg).

Magnetic Properties
The variation of the product of molar magnetic susceptibility (χM) and temperature (T) with respect to temperature (T) for Ni II dimers 1, 2, and 3 is presented in Figures 3-5.It can be seen that the experimental χMT values for three Ni II dimers at room temperature (~300 K) lie around 2.24-2.73cm 3 Kmol -1 , which in good agreement with the expected value for two non-interacting Ni II , S = 1 centers (the theoretical spin-only value is 2.0 cm 3 KmoL -1 , g = 2).For all the complexes, variation of χMT values shows similar type of behavior on lowering the temperature: the χMT values gradually increase upon lowering of temperatures and attain a maximum at ca. 10-12 K. On further decreasing the temperature, a steep fall of the values is observed.This type of behavior is typical of a system that shows dominant intramolecular ferromagnetic exchange coupling, and the steep drop in χMT values

Magnetic Properties
The variation of the product of molar magnetic susceptibility (χ M ) and temperature (T) with respect to temperature (T) for Ni II dimers 1, 2, and 3 is presented in Figures 3-5.It can be seen that the experimental χ M T values for three Ni II dimers at room temperature (~300 K) lie around 2.24-2.73cm 3 K mol −1 , which in good agreement with the expected value for two non-interacting Ni II , S = 1 centers (the theoretical spin-only value is 2.0 cm 3 K mol −1 , g = 2).For all the complexes, variation of χ M T values shows similar type of behavior on lowering the temperature: the χ M T values gradually increase upon lowering of temperatures and attain a maximum at ca. 10-12 K. On further decreasing the temperature, a steep fall of the values is observed.This type of behavior is typical of a system that shows dominant intramolecular ferromagnetic exchange coupling, and the steep drop in χ M T values at lower temperatures can be attributed to zero-field splitting (ZFS) of the ground state (S = 2) and/or intermolecular interactions between the dimers.
The obtained magnetic data of the three compounds were analyzed by using the HDVV Hamiltonian H = −2JS 1 S 2 .We fitted the magnetic data of complexes 1, 2, and 3 for a simple S = 1 dimer model following the equation ( 1), which is derived from the HDVV Hamiltonian [4], in which x = 2J/kT. (1) The Weiss constant was taken (θ) as an additional parameter to include the intermolecular interaction term ( Z J ) as well as the anisotropy term (D).The parameters N, β, and k in Equation ( 1) have their usual meanings, and J is the coupling parameter.
Magnetochemistry 2018, 4, x FOR PEER REVIEW 5 of 14 at lower temperatures can be attributed to zero-field splitting (ZFS) of the ground state (S = 2) and/or intermolecular interactions between the dimers.The obtained magnetic data of the three compounds were analyzed by using the HDVV Hamiltonian H = −2JS1S2.We fitted the magnetic data of complexes 1, 2, and 3 for a simple S = 1 dimer model following the equation (1), which is derived from the HDVV Hamiltonian [4], in which The Weiss constant was taken (θ) as an additional parameter to include the intermolecular interaction term (ZJ′) as well as the anisotropy term (D).The parameters N, β, and k in Equation ( 1) have their usual meanings, and J is the coupling parameter.at lower temperatures can be attributed to zero-field splitting (ZFS) of the ground state (S = 2) and/or intermolecular interactions between the dimers.The obtained magnetic data of the three compounds were analyzed by using the HDVV Hamiltonian H = −2JS1S2.We fitted the magnetic data of complexes 1, 2, and 3 for a simple S = 1 dimer model following the equation (1), which is derived from the HDVV Hamiltonian [4], in which 2 / kT  x J .
Thus, besides intra-dimer coupling, ZFS and inter-dimer coupling are also present in the systems; however, their correct evaluation is not possible due to their close relationship [34,35].
Isothermal magnetization measurements studies (Figures S5-S7) suggest that the ferromagnetic coupling of all complexes leads to S = 2 ground spin states.However, the magnetization values at 5T are smaller than the expected values of ca.4µB, because there are inter-dimer antiferromagnetic interactions and/or zero-field splitting [26].

Theoretical Magnetic Calculation
In this present work, we have calculated the magnetic exchange parameters (J) by broken symmetry DFT calculations using B3LYP functional as described in experimental section.The Hamiltonian for the systems was taken as H = −2JS1S2, in which J is the coupling constants for the pair of Ni II centers having S1and S2 spin vectors for all the dimers.
The magnetic coupling constant (J) between a pair of metallic centers can be estimated by the following relationship [26]: For complex 1, the calculated coupling constant (J) by broken symmetry DFT method is 4.25 cm −1 showing ferromagnetic coupling interaction in agreement with the experimental results.However,  2).Here, it is worth mentioning that consideration of only the anisotropy term (D) in the spin Hamiltonian yielded poor data fitting at low temperature.Thus, besides intra-dimer coupling, ZFS and inter-dimer coupling are also present in the systems; however, their correct evaluation is not possible due to their close relationship [34,35].
Isothermal magnetization measurements studies (Figures S5-S7) suggest that the ferromagnetic coupling of all complexes leads to S = 2 ground spin states.However, the magnetization values at 5T are smaller than the expected values of ca.4µ B , because there are inter-dimer antiferromagnetic interactions and/or zero-field splitting [26].

Theoretical Magnetic Calculation
In this present work, we have calculated the magnetic exchange parameters (J) by broken symmetry DFT calculations using B3LYP functional as described in experimental section.The Hamiltonian for the systems was taken as H = −2JS 1 S 2 , in which J is the coupling constants for the pair of Ni II centers having S 1 and S 2 spin vectors for all the dimers.
The magnetic coupling constant (J) between a pair of metallic centers can be estimated by the following relationship [26]: For complex 1, the calculated coupling constant (J) by broken symmetry DFT method is 4.25 cm −1 showing ferromagnetic coupling interaction in agreement with the experimental results.However, the DFT method slightly overestimates the J value.For complex 2, the estimated value is 3.56 cm −1 , which also indicates the presence of ferromagnetic interaction between the Ni II centers, but the theoretical result slightly underestimates the coupling constant.For complex 3, the calculated value of coupling constant from broken symmetry DFT method is found to be 5.50 cm −1 , which is in good agreement with the experimental value.
In order to further investigate the mechanism of exchange pathway, the spin density distribution has been analyzed.Molecular orbital theory suggests that the spin delocalization transfers the spin density from the magnetic metal centers to their corresponding adjacent ligand atoms.The Mulliken spin populations of complexes 1, 2, and 3 in their broken symmetry states are represented in Tables 3-5, respectively (spin density orbital plots are shown in Figures S8-S10).The positive sign of spin density represents the α spin states, and the negative sign represents the β spin states.From the tables, it can be seen that two Ni II centers have same amount of spin populations but the sign is opposite.Moreover, each Ni II center has highest value of spin density among all the atoms present in a compound.Thus, it is obvious that Ni II centers are indeed the magnetic centers.It can be also seen from the tables that about ~17% of total spin density is distributed among the other atoms present in the complexes.
If the crystallographic structure of each compound is examined, one could find two oxygen atoms and one chlorine atom bridging the two magnetic metal centers along with one Cl-Na-Cl moiety.However, we can see from the spin density table that there is nearly zero delocalization on Na and bridging Cl atom for all the complexes, and only the significant amount of spin density is showing on the bridging oxygen atoms from phenoxido bridge.Negligible overall spin population on the bridging chloride atoms is the result of partial α and β spin density overlap due to the spin polarization in broken symmetry state (Figures S8-S10) [36][37][38].Thus, it may be concluded that the ferromagnetic exchange interactions between two Ni II centers are dominated by phenoxido bridges for all the complexes [39].

Comparison of Structural and Magnetic Parameters
The dimeric units of all three Ni II complexes have a C 2 axis of symmetry, and the two Ni II centers are connected through a pair of phenoxido bridges and a chlorido bridge along with a Cl-Na-Cl bridging unit in all the complexes (Figure 6).The core structures are similar for all the complexes, but they have slightly different Ni-O-Ni, Ni-Cl-Ni bond angles and Ni• • • Ni distances.From the structures of these complexes, it is expected that the overall magnetic exchange interaction between the Ni II centers are mediated by two phenoxido and one chlorido bridges.However, from the spin density calculation, no significant overall spin density is observed on the chloride bridge.Thus, only the phenoxido bridges play crucial role in the coupling parameters (J).
If the crystallographic structure of each compound is examined, one could find two oxygen atoms and one chlorine atom bridging the two magnetic metal centers along with one Cl-Na-Cl moiety.However, we can see from the spin density table that there is nearly zero delocalization on Na and bridging Cl atom for all the complexes, and only the significant amount of spin density is showing on the bridging oxygen atoms from phenoxido bridge.Negligible overall spin population on the bridging chloride atoms is the result of partial α and β spin density overlap due to the spin polarization in broken symmetry state (Figures S8-S10) [36][37][38].Thus, it may be concluded that the ferromagnetic exchange interactions between two Ni II centers are dominated by phenoxido bridges for all the complexes [39].

Comparison of Structural and Magnetic Parameters
The dimeric units of all three Ni II complexes have a C2 axis of symmetry, and the two Ni II centers are connected through a pair of phenoxido bridges and a chlorido bridge along with a Cl-Na-Cl bridging unit in all the complexes (Figure 6).The core structures are similar for all the complexes, but they have slightly different Ni-O-Ni, Ni-Cl-Ni bond angles and Ni⋯Ni distances.From the structures of these complexes, it is expected that the overall magnetic exchange interaction between the Ni II centers are mediated by two phenoxido and one chlorido bridges.However, from the spin density calculation, no significant overall spin density is observed on the chloride bridge.Thus, only the phenoxido bridges play crucial role in the coupling parameters (J).The sign and magnitude of the magnetic coupling constants of binuclear oxido/phenoxido bridging Ni(II) complexes have been investigated both experimentally and theoretically and correlated with the structural parameters by various groups [24,[40][41][42][43].It is now understood that ferromagnetic coupling is observed when the bridging angle is around 90° as a result of the orthogonality of magnetic orbitals.As the Ni-O-Ni angle increases, the magnitude of coupling constant decreases and tends to become antiferromagnetic at around 95°-96°.For the present complexes, the two phenoxido bridging angles are close to 90°, (89.32(8)°, 89.39(6)°, and 87.32(4)° for 1-3, respectively).Hence, the coupling is ferromagnetic.It is to be noted here that several ferromagnetically coupled diphenoxido bridged dinuclear Ni(II) were synthesized by introducing an additional water bridge between the Ni(II) centers.In the present complexes, this additional bridging atom is chloride, and it has a similar effect on magnetic coupling as the water bridge, i.e., it stabilizes the molecule below the crossover angle, and consequently the coupling becomes ferromagnetic.The sign and magnitude of the magnetic coupling constants of binuclear oxido/phenoxido bridging Ni(II) complexes have been investigated both experimentally and theoretically and correlated with the structural parameters by various groups [24,[40][41][42][43].It is now understood that ferromagnetic coupling is observed when the bridging angle is around 90 • as a result of the orthogonality of magnetic orbitals.As the Ni-O-Ni angle increases, the magnitude of coupling constant decreases and tends to become antiferromagnetic at around 95 • -96 • .For the present complexes, the two phenoxido bridging angles are close to 90 • , (89.32(8) • , 89.39(6) • , and 87.32(4) • for 1-3, respectively).Hence, the coupling is ferromagnetic.It is to be noted here that several ferromagnetically coupled diphenoxido bridged dinuclear Ni(II) were synthesized by introducing an additional water bridge between the Ni(II) centers.In the present complexes, this additional bridging atom is chloride, and it has a similar effect on magnetic coupling as the water bridge, i.e., it stabilizes the molecule below the crossover angle, and consequently the coupling becomes ferromagnetic.

Starting Materials
Salicylaldehyde, 5-chlorosalicylaldehyde, 2-hydroxy-3-methoxybenzaldehyde and N-methyl-1,3-propanediamine, and sodium borohydride were purchased from Spectrochem, India and were of reagent grade.They were used without further purification.The other reagents and solvents were of commercially available reagent quality, unless otherwise stated.The ligand, HL 1 , was synthesized [44] by refluxing a solution of salicylaldehyde (0.52 mL, 5 mmoL) and N-methyl-1,3-propanediamine (0.52 mL, 5 mmoL) in methanol (30 mL) for 1 h.The methanolic solution was subsequently cooled to 0 • C and solid sodium borohydride (210 mg, 6 mmoL) was added slowly with constant stirring.After completion of the addition, the resulting reaction mixture was acidified with concentrated HCl (5 mL) and then evaporated to dryness.The reduced Schiff-base ligand (HL 1 ) was extracted from the solid residue with methanol.This methanolic solution was used for further reaction.HL 2 and HL 3 were synthesized, maintaining a similar procedure by using 5-chlorosalicylaldehyde (782.85 mg, 5 mmoL) and 2-hydroxy-3-methoxybenzaldehyde (760.75 mg, 5 mmoL), respectively, instead of salicylaldehyde.

Synthesis of the Complexes
NiCl 2 •6H 2 O (647.95 mg, 5 mmoL), dissolved in 20 mL methanol, was added to methanolic solution of previously prepared HL 1 as described above with constant stirring.An aqueous solution of (2 mL) of NaCl (146.1 mg, 2.5 mmoL) was slowly added with constant stirring after 15 min.The resulting green solution was kept in air for overnight.On slow evaporation of the resulting filtrate, green single crystals of complex 1 were obtained for X-ray diffraction.Complexes 2 and 3 were synthesized following the same procedure, using HL 2 and HL 3 , respectively.Green single crystals of 2 were obtained by the same procedure as stated above.For complex 3, crystals suitable for X-ray diffraction were obtained by layer diffusion of diethylether to the acetonitrile solution of 3.

Physical Measurements
Elemental analyses (C, H, and N) were performed using a Perkin-Elmer 2400 series II elemental analyzer.IR spectra in KBr pellets (4500-500 cm −1 ) were recorded using a Perkin-Elmer RXI FT-IR spectrophotometer.Electronic spectra (1500-250 nm) were recorded in a Hitachi U-3501 spectro-photometer.The magnetic measurements of all the three Ni II dimers (1, 2, and 3) were investigated with a Quantum Design superconducting quantum interference device vibrating sample magnetometer (SQUID-VSM).Powdered polycrystalline samples were used for all measurements under a DC magnetic field.The molar paramagnetic susceptibilities (χ M ) were measured at a constant magnetic field at 0.075 T under a decreasing temperature range of 300 to 2.5 K. Isothermal magnetizations measurements were performed at 2 K up to 5 Tesla for all the complexes.The measured susceptibilities were corrected according to the literature values of Pascal's table due to the diamagnetic contributions [45].For complex 3, all the magnetic calculations were carried out by deducting the molecular weight of readily volatile diethyl ether molecule from the actual molecular weight of the sample.

X-ray Crystallographic Data Collection and Refinement
Collected diffractable single crystals of complexes 1, 2, and 3 were mounted on a Bruker-AXS SMART APEX II diffractometer equipped with a graphite monochromator and Mo-Kα (λ = 0.71073 Å) radiation.The crystals were positioned at 60 mm from the CCD.360 frames were measured with a counting time of 5 s.The structures were solved using direct methods with the Shelxs97 program [46].The non-hydrogen atoms were refined with anisotropic thermal parameters.The hydrogen atoms bonded to carbon were included in geometric positions and given thermal parameters equivalent to 1.2 times those of the atom to which they were attached.The structures were refined on F 2 using Shelxl16/6 on F 2 [47].In 1, there were two oxygen atoms refined with reduced occupancy but the attached hydrogen atoms could not be located.For 2, the squeeze option in Platon was used [48].Details of the crystallographic data are summarized in Table 6.CCDC-1874587 (1), CCDC-1874588 (2), and CCDC-1874589 (3) contain the supplementary crystallographic data for this paper.These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.

Computational Methodology
The Ni II • • • Ni II coupling constants (J) of these complexes were calculated theoretically by broken symmetry DFT as proposed by Ruiz et al. [49][50][51].All the coordinates of atoms were obtained from experimental X-ray structures and used in calculations without further optimization of the structures.The hybrid B3LYP functional [52][53][54] along with Ahlrichs type triple zeta with polarization function def2-tzvp basis set [55] has been employed in all calculations as implemented in the ORCA package (version 3.0.3)[56].The zero'th-order regular approximation (ZORA) has been also incorporated to describe scalar relativistic effect [57].To speed up the calculations with desired accuracy, RIJCOSX approximation with auxiliary def2-TZVP/J coulomb fitting basis set and tight SCF convergence criteria (Grid 4) have also been incorporated [58].

Figure 3 .
Figure 3. Plot of χMT vs. T for complex 1.The circles are the experimental data, and the solid line is generated from the fitted curve.

Figure 4 .
Figure 4. Plot of χMT vs. T for complex 2. The circles are the experimental data, and the solid line is generated from the fitted curve.

Figure 3 .
Figure 3. Plot of χ M T vs. T for complex 1.The circles are the experimental data, and the solid line is generated from the fitted curve.

Figure 3 .
Figure 3. Plot of χMT vs. T for complex 1.The circles are the experimental data, and the solid line is generated from the fitted curve.

Figure 4 .
Figure 4. Plot of χMT vs. T for complex 2. The circles are the experimental data, and the solid line is generated from the fitted curve.

Figure 4 .
Figure 4. Plot of χ M T vs. T for complex 2. The circles are the experimental data, and the solid line is generated from the fitted curve.

Figure 5 .
Figure 5. Plot of χMT vs. T for complex 3.The circles are the experimental data, and the solid line is generated from the fitted curve.

Table 1 .
Hydrogen-Bond Parameters in Complexes (in Å and deg).

Table 1 .
Hydrogen-Bond Parameters in Complexes (in Å and deg).

Table 2 .
Magnetic coupling constants (J) and selective structural parameters for complexes 1

Table 2 .
Magnetic coupling constants (J) and selective structural parameters for complexes 1

Table 3 .
The spin densities on the selected atoms for complex 1.

Table 4 .
The spin densities on the selected atoms for complex 2.

Table 5 .
The spin densities on the selected atoms for complex 3.