Structure, Optical and Magnetic Properties of Two Isomeric 2-Bromomethylpyridine Cu(II) Complexes [Cu(C6H9NBr)2(NO3)2] with Very Different Binding Motives

Two isomeric 2-bromomethylpyridine Cu(II) complexes [Cu(C6H9NBr)2(NO3)2] with 2-bromo-5-methylpyridine (L1) and 2-bromo-4-methylpyridine (L2) were synthesized as air-stable blue materials in good yields. The crystal structures were different with [Cu(L1)2(NO3)2] (CuL1) crystallizing in the monoclinic space group P21/c, while the 4-methyl derivative CuL2 was solved and refined in triclinic P1¯. The orientation of the Br substituents in the molecular structure (anti (CuL1) vs. syn (CuL2) conformations) and the geometry around Cu(II) in an overall 4 + 2 distorted coordination was very different with two secondary (axially elongated) Cu–O bonds on each side of the CuN2O2 basal plane in CuL1 or both on one side in CuL2. The two Br substituents in CuL2 come quite close to the Cu(II) centers and to each other (Br⋯Br ~3.7 Å). Regardless of these differences, the thermal behavior (TG/DTA) of both materials is very similar with decomposition starting at around 160 °C and CuO as the final product. In contrast to this, FT-IR and Raman frequencies are markedly different for the two isomers and the UV–vis absorption spectra in solution show marked differences in the π–π* absorptions at 263 (CuL2) or 270 (CuL1) nm and in the ligand-to-metal charge transfer bands at around 320 nm which are pronounced for CuL1 with the higher symmetry at the Cu(II) center, but very weak for CuL2. The T-dependent susceptibility measurements also show very similar results (µeff = 1.98 µB for CuL1 and 2.00 µB for CuL2 and very small Curie–Weiss constants of about −1. The EPR spectra of both complexes show axial symmetry, very similar averaged g values of 2.123 and 2.125, respectively, and no hyper-fine splitting.

Within the vast group of polyfunctional ligands, halogenated pyridines are interesting candidates for the formation of supramolecular materials due to their ability to coordinate to metals using the pyridine function and at the same time potentially form intermolecular halogen bonds to aggregate into supramolecular structures in the solid [14,17,40]. However, only a few Cu(II) complexes or coordination polymers with simple halido-pyridines have previously been reported [43][44][45].
Herein, we report the reaction of Cu(NO3)2•3H2O with 2-bromo-5-methylpyridine and 2-bromo-4-methylpyridine (Scheme 1) forming the two isomeric complexes [Cu(L)2(NO3)2]. The two compounds show very different structural features (in single crystal X-ray studies and Hirshfeld analyses) which also translate into the optical (UV-vis absorption) and magnetic properties (magnetism and EPR). We also report their FT-IR and Raman characterization and thermal properties. Scheme 1. Synthesis and yields of the two Cu(II) complexes of this study.
This stands in contrast to the 2-bromo-4-methylpyridine derivative (CuL 2 ), where two further O atoms of the NO3 − ligands (O2 and O6) are both localized on the same side of the equatorial N2O2 plane. Thus, they form a rather unusual double-tipped square pyramidal coordination around Cu(II). However, whilst this arrangement also agrees with the Jahn-Teller distortion of the Cu(II) d 9 system [47], it is rather exotic. In the case of the 2-bromo-4-methyl ligand, this peculiar orientation of the two NO3 − ligands is in line with The molecular entities of both forms show the same coordination pattern around the Cu(II) centers, which is trans square planar with two pyridine N atoms and two nitrate O atoms contributing to a square planar N 2 O 2 coordination (Figures 2 and 3; essential metrical data in Table S2). For the 2-bromo-5-methylpyridine complex CuL 1 (P2 1 /c), two additional O atoms from the coordinated NO 3 − ligands O1 and O1 i (with (i) = −x + 2, −y, −z + 2) contribute to an overall elongated octahedral (4 + 2) coordination with Cu· · · O axial distances of 2.515(7) Å (Table S2). The Cu-O bonds are much longer than those of the two equatorial O atoms and due to the small bite angle of the NO 3 − ligand, these two O atoms are not in an ideal position for a perfect tetragonally elongated octahedron (D 4h ). The additional deviation of the two axial ligands distorts the coordination sphere towards an overall C 2v symmetry. The tetragonal elongation is frequently found for Cu(II) complexes [46][47][48], this C 2v deviation has been previously observed for derivatives with equatorial-axial chelate ligands [46,47]. The previously reported Cu(II) bis-nitrato bis-pyridine complexes [Cu(L)(NO 3 ) 2 ] (L = 2-chloro-5-methylpyridine [49] and [Cu(4-MTPP)(NO 3 ) 2 ] (4-MTPP = 4-(methylthio)-4-(pyridin-2-yl)pyrimidine [50] and the bis-pyridine-N-oxide complex [Cu(2,6-dmnpn) 2 (NO 3 ) 2 ] (2,6-dmnpn = 4-NO 2 -2,6-Me 2 -pyridine-N-oxide) [51] show the same type of bonding with very similar geometry around Cu(II), while [Cu(Py) 2 (NO 3 ) 2 ] 2 Py has a dimeric structure with the Cu-bound O NO3 atom bridging between the Cu(II) centers [52].     In the crystal structure, both forms show a number of intermolecular hydrogen bonds (  Figure S1).
Powder X-ray diffraction (PXRD) showed an excellent match of the experiment with the patterns calculated from the single crystal XRD for CuL 1 ( Figure S2). The powder diffractogram of CuL 2 also matched with the pattern calculated from the single crystal structure but the materials show a lower degree of crystallinity (broader signals) ( Figure S3). Nevertheless, the PXRD patterns of both compounds confirm that they are phase-pure and thus suitable for the measurement of thermal, spectroscopic, and magnetic properties.

Hirshfeld Surface Analysis and Energy Framework Calculations
Hirshfeld surface analysis was carried out for a more detailed view on the intermolecular interactions. The important intermolecular interactions in the molecule are highlighted by the di and de plotted as fingerprints. The fingerprint plot analysis for CuL 1 showed that intermolecular H … O interactions make the highest contribution with 44.6% This stands in contrast to the 2-bromo-4-methylpyridine derivative (CuL 2 ), where two further O atoms of the NO 3 − ligands (O2 and O6) are both localized on the same side of the equatorial N 2 O 2 plane. Thus, they form a rather unusual double-tipped square pyramidal coordination around Cu(II). However, whilst this arrangement also agrees with the Jahn-Teller distortion of the Cu(II) d 9 system [47], it is rather exotic. In the case of the 2-bromo-4-methyl ligand, this peculiar orientation of the two NO 3 − ligands is in line with the crystal symmetry (Cu does not lie at the center of the inversion) and goes along with the two Br substituents pointing in the same direction which could be called a syn conformation, contrasting with the anti conformation of the two Br atoms in the 2-bromo-5methyl derivative. The Br· · · Br distance of 3.694(3) Å in CuL 2 is slightly shorter than the sum of the van der Waals radii of 3.72 Å [53], while the Cu· · · Br distance of about 3.26 Å is far too long to account for coordination (3.33 Å for CuL 1 ).
In the crystal structure, both forms show a number of intermolecular hydrogen bonds (Table S3). The shortest ones for CuL 1 with~2.58 Å are found between O2 and the H-C4 and H-C6 groups of the pyridine ligand, while for CuL 2 slightly shorter intermolecular distances of about 2.39 (O4· · · H-C12), 2.48 (O6· · · H-C12), and 2.49 Å (O5· · · H-C10) were found. All these hydrogen bonds are considered to be weak and essentially of an electrostatic nature [54]. Intermolecular Br· · · Br halogen bonds were not observed, the two Br substituents in CuL 2 point towards H-C3 and H-C6 functions of neighboring molecules with Br· · · H contacts longer than 3 Å ( Figure S1).
Powder X-ray diffraction (PXRD) showed an excellent match of the experiment with the patterns calculated from the single crystal XRD for CuL 1 ( Figure S2). The powder diffractogram of CuL 2 also matched with the pattern calculated from the single crystal structure but the materials show a lower degree of crystallinity (broader signals) ( Figure S3). Nevertheless, the PXRD patterns of both compounds confirm that they are phase-pure and thus suitable for the measurement of thermal, spectroscopic, and magnetic properties.

Hirshfeld Surface Analysis and Energy Framework Calculations
Hirshfeld surface analysis was carried out for a more detailed view on the intermolecular interactions. The important intermolecular interactions in the molecule are highlighted by the d i and d e plotted as fingerprints. The fingerprint plot analysis for CuL 1 showed that intermolecular H· · · O interactions make the highest contribution with 44.6% of the total, while others are markedly smaller 14.6% (H· · · Br), 12.5% (H· · · H), 12.4% (C· · · H), and 5% (N· · · H), respectively ( Figure S4). For CuL 2 , the O· · · H interactions account for 41.2% which is slightly lower than the 44.6% for the CuL 1 derivative and stands somewhat in contrast to the shorter hydrogen bonds found in the crystal structures for the CuL 2 complex. However, this analysis summarizes the interactions and does not weight them. The Br· · · H (15.9%), H· · · H (13.4%), C· · · H (9.4%), and N· · · H (4.8%) interactions are also very similar.
An energy framework analysis [55] was carried out for the two compounds ( Figure S5, Table S4). The interaction energies include electrostatic (E ele(L When comparing the individual contributions, it becomes clear that all of them are higher for CuL 2 (or lower for the repulsion) due to the lower symmetry. However, concerning the resulting marked difference in total energy of −122.7 kJ mol −1 for CuL 1 vs. −340.4 kJ mol −1 for CuL 2 , we are skeptical, because the thermal properties and the IR and Raman spectroscopy of the two complexes are not very different (see later). In future work, we will calculate the stability of the two complexes in the crystal lattice in more detail using density functional theory (DFT) methods.

Infrared and Raman Spectroscopy
The asymmetric and symmetric H-C vibrations of the CH 3 and CH groups of both complexes were observed in the FT-IR spectra in the range from 3150 to 2850 cm −1 followed by aromatic H-C overtones in the 2800-1700 cm −1 range ( Figure S6) [56][57][58][59][60]. When comparing the two complexes CuL 1 and CuL 2 (L 1 = 2-bromo-5-methylpyridine, L 2 = 2-bromo-4-methylpyridine), similar resonances were found in the range from 1600 to 120 cm −1 (Figure 4). The underlying pyridine C=C and C=N and NO 3 stretching frequencies were generally at a higher energy for the CuL 1 complex and the relative intensities differ between the two stereoisomers with their different coordination pattern. The same is true for the umbrella-like H-C vibration modes of the CH 3 substituents, the H-C arene oscillations, the H-C wagging, and the C-Br stretches [56][57][58] in the range from 1150 to 900 cm −1 . Even more different are the Raman spectra of the two complexes ( Figure S7). However, here all the characteristic bands are shifted to a lower energy for the CuL 1 complex compared with the CuL 2 derivative especially the C-Br stretches at around 800 cm −1 and the NO 3 stretches at around 700 cm −1 which are also found in the FT-IR. Additionally, the Cu-N and Cu-O stretches in the range between 400 and 200 cm −1 are shifted to lower energy for the CuL 1 derivative which would mean slightly weaker Cu-ligand bonds. However, the thermal analyses (TG/DTA) do not show marked differences in their behavior (see below) and the different T of the single crystal X-ray diffraction data precludes comparison.

Thermal Analysis
Thermogravimetric (TG) analyses shows that the two compounds are thermally stable up to about 160 °C ( Figure S8). The differential thermal analysis (DTA) shows an endothermic peak at around 200 °C for both complexes (slightly higher for CuL 2 ), followed by one or two intense exothermic signals in the range 220 to 300 °C. We assume the decomposition of the nitrate ligands to NO, NO2, and O2 being responsible for these exothermic peaks. The final mass loss after heating to 300° C is 83% for CuL 1 and 86% for CuL 2 , which is in line with the formation of CuO (calculated loss of 85%). The same behavior was observed for comparable Cu(II) complexes [60-63].

UV-vis Absorption Spectroscopy
The two complexes show very similar absorption spectra in EtOH solution. An in-

Thermal Analysis
Thermogravimetric (TG) analyses shows that the two compounds are thermally stable up to about 160 • C ( Figure S8). The differential thermal analysis (DTA) shows an endothermic peak at around 200 • C for both complexes (slightly higher for CuL 2 ), followed by one or two intense exothermic signals in the range 220 to 300 • C. We assume the decomposition of the nitrate ligands to NO, NO 2 , and O 2 being responsible for these exothermic peaks. The final mass loss after heating to 300 • C is 83% for CuL 1 and 86% for CuL 2 , which is in line with the formation of CuO (calculated loss of 85%). The same behavior was observed for comparable Cu(II) complexes [60][61][62][63].

UV-Vis Absorption Spectroscopy
The two complexes show very similar absorption spectra in EtOH solution. An intense band around the 220 nm is identical for both complexes ( Figure 5). Marked differences were found in the UV-vis range where the CuL 1 complex shows a maximum at 270 nm, while the same band for the CuL 2 derivative is found at 263 nm. These bands can be assigned to π-π* transition as they also occur for the ligands ( Figure S9). For the CuL 1 complex, we observed a broad shoulder in the range from 300 to 350 nm which we assign to a ligand (NO3)to-metal-charge transfer excitation. For the CuL 2 derivative, this is far less pronounced which we ascribe to the unusual orientation of the two NO 3 − ligands compared to CuL 1 . The long-wavelength absorptions at around 750 nm are quite similar for both complexes (743 nm for CuL 1 , 753 nm for CuL 2 ). They were assigned to the 2 B 2g → 2 E g and 2 A 1g → 2 E g electronic transitions in Cu(II), d 9 systems [29,32,46,59,63]. They are responsible for the typical blue color of the two Cu(II) complexes. While the similar NIR bands might point to a decomposition of the complexes in EtOH solution, forming the non-coordinated ligands and a [Cu(II)(EtOH)n] 2+ solvate, the bands in the 300-350 nm range which do not occur for the ligands, support that the complexes were stable in EtOH solution. If that is the case, we conclude that the NIR absorptions of the complexes are either not affected by the different coordination of the two isomers, or CuL 2 forms the same anti structure as CuL 1 in solution. In future work, we will study this possibility in more detail.

Magnetization and EPR Studies
Temperature-dependent magnetic susceptibility data ( Figure 6) reveal a nearly perfect paramagnetic behavior in the entire temperature range from 5 to 350 K, in line with the isolated Cu(II) centers. A Curie-Weiss fit of the inverse magnetic susceptibility yields nearly identical effective magnetic moments of μeff = 1.98 μB for CuL 1 and μeff = 2.00 μB for CuL 2 . Both are constant over a wide temperature range and within the range from 1.7 to 2.2 μB experimentally observed for similar Cu(II) complexes at room temperature [25,[64][65][66][67][68][69][70]. The effective moments being larger than the spin-only value of 1.73 μB expected for a 3d 9 configuration (S = 1/2) indicate spin-orbit coupling [  While the similar NIR bands might point to a decomposition of the complexes in EtOH solution, forming the non-coordinated ligands and a [Cu(II)(EtOH) n ] 2+ solvate, the bands in the 300-350 nm range which do not occur for the ligands, support that the complexes were stable in EtOH solution. If that is the case, we conclude that the NIR absorptions of the complexes are either not affected by the different coordination of the two isomers, or CuL 2 forms the same anti structure as CuL 1 in solution. In future work, we will study this possibility in more detail.

Magnetization and EPR Studies
Temperature-dependent magnetic susceptibility data ( Figure 6) reveal a nearly perfect paramagnetic behavior in the entire temperature range from 5 to 350 K, in line with the isolated Cu(II) centers. A Curie-Weiss fit of the inverse magnetic susceptibility yields nearly identical effective magnetic moments of µ eff = 1.98 µ B for CuL 1 and µ eff = 2.00 µ B for CuL 2 . Both are constant over a wide temperature range and within the range from 1.7 to 2.2 µ B experimentally observed for similar Cu(II) complexes at room temperature [25,[64][65][66][67][68][69][70]. The effective moments being larger than the spin-only value of 1.73 µ B expected for a 3d 9 configuration (S = 1/2) indicate spin-orbit coupling [23,37,39,64,65].
The X-band EPR spectra of a powder sample of CuL 1 showed an axial spectrum (Figure 7) with no hyperfine splitting (HFS). The spectrum was simulated ( Figure S10) using the g values g = 2.230 and g ⊥ = 2.070, which average to g av = 2.123, the g anisotropy ∆g calculates to 0.160 (Table 1).  In contrast to this, we observed two signals for CuL 2 (L 2 = 2-bromo-4-methylpyridine) (Figure 7; Figure S11). Simulations allowed to identify one species with similar parameters as the CuL 1 complex, g|| = 2.260 and g = 2.057, which averages to gav = 2.125. The g anisotropy ∆g for this species calculates to 0.203 (species I in Table 1). We conclude that this spectrum represents CuL 2 , which is supported by the missing HFS in both spectra. The spectrum of the second species was simulated using g|| = 2.135 and g = 2.057, which averages to gav = 2.083 The g anisotropy ∆g calculates to 0.078 (species II in Table 1). Here, pronounced HFS of 63/65 Cu (I = 3/2) and 14 N (I = 1) was observed. At the moment, we have no idea about the nature of the second Cu(II) species and how the PXRD failed to detect it in the CuL 2 product. It may be that the EPR sample of CuL 2 contained molecules of both syn and anti conformations in the same crystal structure. Such syn-to-anti conversion was already discussed for the UV-vis absorption spectroscopy in solution. To observe this in the solid would be even more fascinating and we will study this in detail in the near future.
The g values derived from the magnetic measurements of 2.29 for CuL 1 and 2.32 for CuL 2 both lie higher than the averaged g values from EPR spectroscopy but both values are higher for CuL 2 while the values from the magnetic measurements roughly coincide  a Powdered samples at 298 K, L 1 = 2-bromo-5-methylpyridine, L 2 = 2-bromo-4-methylpyridine. Averaged g value g av = (g + 2g ⊥ )/3; g anisotropy ∆g = g − g ⊥ . b Hyperfine splitting (HFS) in G. c Simulated line width in G.
In contrast to this, we observed two signals for CuL 2 (L 2 = 2-bromo-4-methylpyridine) (Figure 7; Figure S11). Simulations allowed to identify one species with similar parameters as the CuL 1 complex, g = 2.260 and g ⊥ = 2.057, which averages to g av = 2.125. The g anisotropy ∆g for this species calculates to 0.203 (species I in Table 1). We conclude that this spectrum represents CuL 2 , which is supported by the missing HFS in both spectra. The spectrum of the second species was simulated using g = 2.135 and g ⊥ = 2.057, which averages to g av = 2.083 The g anisotropy ∆g calculates to 0.078 (species II in Table 1). Here, pronounced HFS of 63/65 Cu (I = 3/2) and 14 N (I = 1) was observed. At the moment, we have no idea about the nature of the second Cu(II) species and how the PXRD failed to detect it in the CuL 2 product. It may be that the EPR sample of CuL 2 contained molecules of both syn and anti conformations in the same crystal structure. Such syn-to-anti conversion was already discussed for the UV-vis absorption spectroscopy in solution. To observe this in the solid would be even more fascinating and we will study this in detail in the near future. The g values derived from the magnetic measurements of 2.29 for CuL 1 and 2.32 for CuL 2 both lie higher than the averaged g values from EPR spectroscopy but both values are higher for CuL 2 while the values from the magnetic measurements roughly coincide with the g values from EPR. We attribute these differences to the specific geometry and orientation of crystallites in the EPR experiment.
For the geometrically very similar complex [Cu(2,6-dmnpn) 2 (NO 3 ) 2 ] (2,6-dmnpn = 4-NO 2 -2,6-Me 2 -pyridine-N-oxide) [51], three g values g z = 2.295, g y = 2.086, g x = 2.066 were reported. Overall, the spectrum looked very similar to those of CuL 1 and CuL 2 with a slightly different g av = 2.149 which is probably due to the O-Cu bonding of the pyridine-N-oxide ligands, but a ∆g = 0.229 which is very similar to our values. We have recently shown that EPR spectra of such pseudo-axial geometry (g y and g x very similar) are frequently observed for Cu(II) complexes containing pyridine-amide ligands. Very similar g av values of over 2.1, ∆g values of around 0.2 and no HFS are in line with various tetragonally elongated octahedral structures [46]. Although the local symmetry around Cu(II) is very different for CuL 1 and CuL 2 , EPR spectroscopy does not allow for tracing of the two very different distortion variants from the ideal octahedral geometry.

Instrumentation
Elemental analyses were carried out using a Hekatech CHNS EuroEA 3000 Analyzer (Hekatech, Wegberg, Germany). Raman measurement was carried out using the Renishaw inVia Reflex Raman (Renishaw, Kingswood, UK). TG measurements were performed using a simultaneous Mettler Toledo TGA/DSC 2/1600 thermal analyzer (Mettler Toledo, Barcelona, Spain) in an N 2 atmosphere at a heating rate of 10 K min −1 . Fourier-transformed infrared (FT-IR) spectra on KBr pellets were obtained using a PerkinElmer UATR II spectrometer (Perkin Elmer, Waltham, MA, USA) in the range from 4000 to 400 cm −1 at room temperature. UV-vis absorption spectra were recorded in EtOH using a SCINCO S-3100 spectrophotometer (Scinco, Seoul, Republic of Korea) or a Varian Cary 05E spectrophotometer (Varian, Darmstadt, Germany). EPR spectroscopy was measured on solid samples in the X-band using a Bruker EMXNano spectrometer (Bruker, Rheinhausen, Germany) at 298 K; g values were calibrated using dpph.

Magnetization Measurements
The magnetic properties were studied using a Quantum Design PPMS Evercool II (Quantum Design GmbH, Darmstadt, Germany) with vibrating sample magnetometer (VSM) option. Then, 12.8 mg of CuL 1 and 10.1 mg of CuL 2 of powdered samples were enclosed in commercially available polypropylene powder capsules and fixed in a brass sample holder. The temperature-dependent magnetization was measured after cooling in an applied field of 10 mT in a temperature range from 5 to 350 K with a heating rate of 1 K min −1 . A small background signal resulting from the sample holder was determined from reference measurements and subtracted from the measured sample data. The obtained susceptibility data were corrected for a diamagnetic susceptibility of χ dia = −271.88 × 10 −11 m 3 mol −1 [64]. A fit of the inverse molar susceptibility data according to the Curie-Weiss law was carried out in the temperature range from 20 to 220 K.

Hirshfeld Surface Analysis
Hirshfeld surface analysis, their relative 2D fingerprint plots, and the energy framework calculations are an essential technique for showing packing modes and intermolecular interactions of crystals' structure using a CIF file in the Crystal Explorer 17 software [74]. The quantifying and decoding of the inter molecular contacts in the crystal packing are visualized using d norm (normalized contact distance) and 2D fingerprint plots, respectively. The dark-red spots on the d norm surface arise as a result of short interatomic contacts, while the other inter molecular interactions appear as light-red spots. d i (inside) and d e (outside) represent the distances to the Hirshfeld surface from the nuclei, with respect to the relative van der Waals radii. The proportional contribution of the contacts over the surface is visualized by the color gradient (blue to red) in the fingerprint plots. The energy framework calculations are estimated from a single point molecular wave function at HF/3-21G. A cluster of 3.8 Å radius was generated around the molecule and the energy calculation was performed. The neighboring molecules (density matrices) are generated within this shell by applying crystallographic symmetry operations with respect to the central molecule (density matrix). The interaction energy is broken down as E tot = k ele E ele + k pol E pol + k dis E dis + k rep E rep where the k values are scale factors, E ele represents the electrostatic component, E pol the polarization energy, E dis the dispersion energy, and E rep the exchange-repulsion energy [55,75,76].

Conclusions
Reacting Cu(NO 3 ) 2 ·3H 2 O with the two isomeric 2-bromo-5-methylpyridine (L 1 ) and 2-bromo-4-methylpyridine (L 2 ) ligands led to two isomeric 2-bromomethylpyridine Cu(II) complexes [Cu(C 6 H 9 NBr) 2 (NO 3 ) 2 ] as air-stable blue materials in good yields. The x-ray single crystal diffraction showed different crystal structures ([Cu(L 1 ) 2 (NO 3 ) 2 ] (CuL 1 ) in P2 1 /c, [Cu(L 2 ) 2 (NO 3 ) 2 ] (CuL 2 ) in P1). The molecular structure showed a primary trans-N 2 (Py)O 2 (NO 3 ) coordination of the Cu(II) centers for both complexes but a very different orientation of the Br substituents. The expected structure, which we described as anti was found for CuL 1 while CuL 2 , showed an unusual syn conformation with neighboring Br atoms and rather short Br· · · Br contacts of about 3.7 Å. Moreover, CuL 1 showed the expected 4+2 distorted coordination with the two secondary (axially elongated) Cu-O bonds from the NO 3 ligands on each side of the CuN 2 O 2 basal plane, while for CuL 2 both secondary O atoms lie on the side which is not occupied by the Br substituents. While the general Hirshfeld surface analysis gave very similar results for both structures, an energy framework analysis gave superior stability for CuL 2 due to the lower symmetry. In contrast to this, the thermal stability (TG/DTA) of both complexes is almost the same with decomposition starting at around 160 • C. The FT-IR and Raman frequencies are markedly different for the two isomers. The T-dependent susceptibilities again show very similar results (µ eff = 1.98 µ B for CuL 1 and 2.00 µ B for CuL 2 and very small Curie-Weiss constants of about −1 K. The EPR spectra of both complexes show axial symmetry, very similar averaged g values of 2.123 and 2.125, respectively, and no hyper-fine splitting. The only difference here lies in the markedly higher g anisotropy ∆g of 0.203 for CuL 2 compared with 0.160 for CuL 1 . In the fluid EtOH solution, the two complexes seem to be stable and UV-vis absorption spectra show marked differences in the π-π* absorptions at 263 for CuL 2 and 270 for CuL 1 nm and in the ligand-to-metal charge transfer bands at around 320 nm which are pronounced for CuL 1 but very weak for CuL 2 . We assign this difference to the higher symmetry at the Cu(II) center for CuL 1 . In contrast to this, the long-wavelength d-d* bands at around 750 nm are very similar for both complexes. At the same time, we cannot rule out that CuL 2 is present in solution in the same isomeric anti form as CuL 1 and that only lattice energies are responsible for the unexpected structure of CuL 2 in the solid. In future work, we will gauge the stability of the two isomeric forms by sophisticated DFT calculations as well as by UV-vis absorption and EPR spectroscopy.