Chlorocobalt complexes with pyridylethyl-derived diazacycloalkanes

With cobalt(II) chloride, some piperazine- and homo-piperazine-derived ligands yield tetra- or pentacoordinate complexes. Observed variations in coordination number are ascribed as being related to chloride solvophobicity. Optical spectra are presented, while magnetism measurements indicate governance of the magnetism by zero-field splitting of the cobalt ion.


Structural commentary
The structures are not all entirely what was originally expected, based on previous work with these types of ligands. The Co-N(Pyr) bond lengths (Tables 1-4) range from 2.03 to 2.16 Å , which is within the usual span (Orpen et al., 1989), while the Co-Cl distances average 2.28 AE 0.03 Å , which is again common for cobalt(II) (Orpen et al., 1989). The Co-N amine bond lengths are generally longer than the Co-N pyridine ones, and quite variable (vide infra), with an average of 2.154 Å and covering a 0.153 Å range. The distances are unexceptional for Co II to tertiary amine linkages (Orpen et al., 1989), and indeed tertiary amine nitrogen atoms in tripodal ligands are often notably more distant from the Co II ion (2. 44-3.27 Å ;Brewer, 2020).
For the CoCl 2 -Ppz combination, the dinuclear compound Co 2 (Ppz)Cl 4 was obtained (Fig. 2), rather than the mononuclear Co(Ppz)Cl 2 . The asymmetric unit in this P2 1 /n struc- Molecular structure of Co 2 (Ppz)Cl 4 . Ellipsoids are drawn at the 50% level, and for clarity of presentation, H atoms are omitted. The two halfmolecules in the structure are symmetry equivalent and are related to the other halves via the symmetry operation (1 À x, 1 À y, 2 À z).

Figure 1
Ligands employed in this work. ture is the half-molecule, related to the molecule's other corresponding half by an inversion centre.
The piperazine moiety in Co(Ppz)Cl 2 does not chelate a cobalt ion, but instead bridges between two, so that each tetracoordinate Co is bound by a piperazine-N atom, a pyridyl-N atom and two chloride ions. The two identical coordination cores have ! = 86 (Sakaguchi & Addison, 1979) and ' t = 0.07 (Addison et al., 2004;Yang et al., 2007), so are fairly close to exactly tetrahedral in geometry.
As the same ligand behaves as a straightforward mononucleating quadridentate in the copper and nickel complexes (O'Connor et al., 2012;Muthuramalingam et al., 2017Muthuramalingam et al., , 2019a, this led to the question as to whether the coordination is governed by the ligand bite vs the larger ionic radius of Co 2+ vs Cu 2+ /Ni 2+ . This proposal was approached by synthesising the homopiperazine analogue, Phpz, whose ligand has a larger (N2-N2A) bite. The compound Co(Phpz)Cl 2 was indeed obtained as a mononuclear product (Fig. 3), crystallizing into a P1 lattice. The structure suffers some disorder, but one conformation is dominant, at 91% (the discussion below refers to that major component of the Co(Phpz)Cl 2 crystals). However, anticipatedly quadridentate Phpz is now seen to act as a tridentate ligand, with the cobalt(II) ion being pentacoordinate.
One of the pyridylethyl arms is now in the less-commonly observed dangling mode, pyridine being a consistent protagonist of this phenomenon (Reeves et al., 2014;Ball et al., 1981;Rajendiran et al., 2008;Camerano et al., 2011;Lonnon et al., 2006;Palaniandavar et al., 1996). The core geometry is markedly toward the trigonal-bipyramidal ( = 0.62) (Addison et al., 1984) with Cl2 acting as the erstwhile reference tetragonal axial ligand. The bond from the cobalt ion to Structure of Co(Phpz)Cl 2 , with its dangling pyridine moiety. The dominant conformer is shown. Ellipsoids are drawn at the 50% level, and for clarity of presentation, H atoms are omitted.

Figure 4
Structural representation of [Co(Ppz)Cl]ClO 4 (major component). The perchlorate is disordered by a rocking motion along the O2B-Cl1B-O4B direction, which may be related to weak C-HÁ Á ÁO hydrogen-bonding interactions. Ellipsoids are drawn at the 50% level, and for clarity of presentation, H atoms are omitted.     , it is clear that ligand bite is not the sole factor governing the structural outcome in Co(Phpz)Cl 2 . However, all the complexes herein were prepared in non-aqueous solvents -methanol or THFand we propose that the chloride ion, with its substantial hydration energy, is solvofugic enough to displace a terminal pyridine in a complex involving cobalt(II). We hence prepared the compound of composition [Co(Ppz)Cl]ClO 4 , thus removing a chloride from the binding competition. The resulting structure bears out this hypothesis (Fig. 4).
In a further experimental essay, we eliminated an otherwise dangling pyridyl arm by replacing it with a methyl group, as in the simpler tridentate ligand Pmhpz. The resulting molecule, Co(Pmhpz)Cl 2 (Fig. 5) crystallizes in the P2 1 /n space group.
The coordination core is somewhat trigonal-bipyramidal, with = 0.57 and the reference axis being Co1-Cl1. The sole pyridine nitrogen N3 and the methylated piperazine nitrogen N1 form the pseudo-trigonal axis. Analogously to the [Co(Ppz)Cl)] + situation, the pseudo-equatorial Co-N2 amine bond, at 2.097 (4) Å , is shorter that the Co-N1 amine [2.232 (5) Å ] and Co-N3 pyridine [2.146 (4) Å ] bonds in the trigonal directions. One may note that the same axial vs equatorial Co-N bond-length relationship also holds for Co(Phpz)Cl 2 , above. Molecular structure of the complex Co(Pmhpz)Cl 2 , with the ligand in which a pyridyl arm is replaced by a methyl group. Ellipsoids are drawn at the 50% level, and for clarity of presentation, H atoms are omitted.

Figure 7
Wavefunction density surface maps of MOs involved in several of the visible-NIR transitions in a CoN 2 Cl 2 moiety of Co 2 (Ppz)Cl 4 , modelled with a 2-(dimethylaminoethyl)pyridine ligand. Lower left and right: originating HOMO(À3), HOMO(À4), respectively; upper left and right, the receiving LUMO and LUMO(+1), respectively. Blue indicates highest density. Note the translation of wavefunction density from the CoCl 2 or CoN 2 Cl 2 unit to the pyridine ring in the excitations.
Both might also be compared to [Co(Me 4 en)]Cl 2 , which has maxima at ca 1670, 1380, 1000, 650 and 580 nm, attributed in a crystal-field model to 4 A 2 ! 4 T 1 (F) transitions (the first three) (Lever, 1984), and the latter two to 4 A 2 ! 4 T 1 (P). Though shifted slightly, these maxima are quite similar to the bands for [Co 2 (Ppz)Cl 4 ] and [Co 2 (Pdmpz)Cl 4 ]. The DFT results for a CoN 2 Cl 2 chromophore of Co 2 (Ppz)Cl 4 suggest that even the low-energy transitions involve CT contributions from the CoCl 2 moiety to the pyridine ring (Fig. 7).

4
A 2 (P) and 4 E(P) (Lever, 1984 Figs. 8 and 9 show the solid-state spectra of CoPhpzCl 2 and [Co(Ppz)Cl]ClO 4 , respectively. In comparison with the CoN 2 Cl 2 cores, one should note the rather different pattern of absorption bands in the NIR. Firstly, the band near 1000 nm appears to be supplanted by two bands, one being near 750 nm, the other around 950 nm. More tellingly, the 1100-1500 nm region, which has clear CoN 2 Cl 2 maxima near 1300 and 1700 nm, becomes hollowed out, and broader features appear at 1600-1900 nm. The Vis-NIR spectrum (Fig. S9 in the supporting information) of Co(Pmhpz)Cl 2 is, as expected, similar to that of Co(Phpz)Cl 2 . We do note that the utility of NIR spectroscopy for tetra-and pentacoordinate cobalt(II) complexes, pioneered by Goodgame & Goodgame (1965) has hardly been widely adopted (Table S1).
Magnetism analysis Preliminary data indicated apparently reduced magnetic moments for some samples. However, the structures do not suggest the possibility of any pathway for significant superexchange coupling. Inasmuch as there are pentacoordinate   Solid-state Vis-NIR spectrum of [Co(Ppz)Cl]ClO 4 .

Figure 10
Temperature dependence of T for Co(Pmhpz)Cl 2 . The solid line is the fit using an exact diagonalization method, between 12.5 and 310 K. (Note that the usual units for molar susceptibility have been replaced here by SI units: 1 cm 3 mol À1 = 4 Â10 À6 m 3 mol À1 .)

Figure 11
Temperature dependence of T for Co(Phpz)Cl 2 . The solid line is the fit using an exact diagonalization method, between 5 and 310 K. cobalt(II) complexes that have recently been discovered to act as single-ion/single-molecule magnets (SIM/SMM) at reduced temperature (Rechkemmer et al., 2016;Ś witlicka et al., 2018), we studied the temperature dependence of the magnetic behaviour of powdered samples of Co(Pmhpz)Cl 2 and Co(Phpz)Cl 2 (Figs. 10 and 11).
The magnetism as a function of temperature and applied field showed no evidence for SMM behaviour. In situations like this, the temperature dependence of the moments has been recognized as being due to zero-field splitting (Nemec et al., 2016;Cruz et al., 2018;Boča et al., 1999;Papá nková et al., 2010;Rajná k et al., 2013;Ż urowska et al., 2008) (see the supporting information for further discussion). We were able to fit the data through most of the temperature regime and the extracted D, g ave , Á, a and b which are listed in Table 6, via: where x and z are the longitudinal and transverse modes of the anisotropic responses (Á = S x /S z ), a is the TIP and b the total diamagnetic correction.
Both compounds have a positive axial single-ion anisotropy (SIA) term, and the anisotropy values also confirm the findings as self-consistent (e.g. Á > 1 for positive D and Á < 1 for negative D, and larger D leads to larger Á). The D and g ave values appear to be in the normal ranges; D values for Co II do cover a wide range, from ca À50 to +100 cm À1 (Cruz et al., 2018;Nemec et al., 2016). While Co II g values intrinsically also cover a wide range, applicable values for fitting ZFS data have been observed to be about 2.0-2.4 (Voronkova et al., 1974;Baum et al., 2016;Banci et al., 1980;Martinelli et al., 1989). Both compounds here show a faster drop in T and a distinct kink at temperatures below ca 15 K. These features have been seen in several other Co II systems (Ż urowska et al., 2008;Papá nková et al., 2010;Boča et al., 1999;Rajná k et al., 2013); however, no definitive accounting for this has been advanced as yet, apart from the not infrequently employed addition of a weak antiferromagnetism mean field term.

Supramolecular features
There are no true supramolecular structures formed by the compounds, whose crystal lattices containing individual mol-ecules are defined mainly by weak, non-bonding interactions. Along with the absence of any solvation of these crystals, the only hydrogen-bonding interactions observed are in [Co(Ppz)Cl]ClO 4 , which has weak C-HÁ Á ÁO hydrogen-bonds (numerical values are given in the CIF), likely of little energetic consequence.
Some lattice views of the compounds are displayed in the supporting information (Figs. S1-S8).

Methods
Chemical ionization mass spectra were obtained on a Thermo-Electron LTQ-FT 7T FT-ICR instrument. UVvisible-near infrared spectra were obtained using Perki-nElmer Lambda-35 or Shimadzu UV3600Plus spectrophotometers equipped with integrating spheres for solid-state spectroscopy. Magnetic susceptibility data between 1.8 and 310 K in an applied field of 1 kOe were collected using a Quantum Design MPMS-XL SQUID magnetometer. Crystals were powdered and packed into #3 gel capsules that were placed inside drinking straws attached to the sample rod. The magnetization was measured at 1.8 K as a function of increasing field from zero to five tesla and at selected fields returning to zero. The data were corrected for the contributions from the sample holders (measured independently) and the diamagnetism of the constituent atoms, as estimated using Pascal's constants (Carlin, 1986). DFT calculations were performed using the !B97X-D/6-31G* method on an iMac16,2 with Spartan-18 software (Wavefunction Inc., Irvine CA, version 1.4.4), while structural diagrams were generated using the CrystalMaker-10 software and Preview-10. Reagents were used as received from TCI America, Sigma-Aldrich, MCB and Fisher Scientific. Elemental microanalyses were by Robertson Microlit Laboratories (Ledgewood, NJ).

Refinement
Crystal data, data collection and structure refinement details are summarized in Table 7. X-ray diffraction data were collected on a Rigaku Oxford Diffraction Gemini diffractometer via !-scans using an Atlas CCD detector using Cu K radiation or a Bruker AXS D8 Quest diffractometer with a PhotonII charge-integrating pixel array detector (CPAD). Data for those structures were collected, scaled and corrected for absorption using the CrysAlis PRO 2015 software suite program package (Rigaku OD, 2015) or APEX4 and SAINT (Bruker, 2021) and SADABS (Krause et al., 2015). Crystal structures were solved using SHELXT (Sheldrick, 2015a), and refined using SHELXL (Sheldrick, 2015b) and ShelXle (Hü bschle et al., 2011), with refinement by full-matrix leastsquares on F 2 . Further processing for the Ppz and Pmhpz complexes utilized the OLEX software (Dolomanov et al., 2009).
The structure of Co(Phpz)Cl 2 contains an additional 121 Å 3 of solvent-accessible voids filled by extensively disordered nitromethane recrystallization solvent. The residual electron density peaks are not arranged in an interpretable pattern. The structure factors were instead augmented via reverse-Fourier-transform methods using the SQUEEZE routine (van der Sluis & Spek, 1990;Spek, 2015) as implemented in PLATON. The resultant FAB file containing the structurefactor contribution from the electron content of the void space was used together with the original hkl file in the further refinement. (The FAB file with details of the SQUEEZE results is included in the CIF in the supporting information). The SQUEEZE procedure corrected for 69 electrons within the solvent-accessible voids, or around two nitromethane molecules. The central part of the metal complex (two of the Co-coordinated nitrogen atoms and the C atoms bridging between them) are disordered by a pseudo-mirror operation. Additional disorder that is vaguely recognizable (largest difference peak 0.78 electrons) was ignored. The two disordered moieties were restrained to have similar geometries. U ij components of ADPs for disordered atoms closer to each other than 2.0 Å were restrained to be similar. Subject to these conditions, the occupancy ratio refined to 0.914 (3)

{1,4-Bis[2-(pyridin-2-yl)ethyl]-1,4-diazacycloheptane}dichloridocobalt(II) (CoPhpzCl2_sq)
Crystal data [CoCl 2 (C 19  where P = (F o 2 + 2F c 2 )/3 (Δ/σ) max = 0.002 Δρ max = 0.81 e Å −3 Δρ min = −0.35 e Å −3 Special details Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes. Refinement. The central part of the metal complex (two of the Co-coordinated nitrogen atoms and the C atoms bridging between them) are disordered by a pseudo-mirror operation. Addtional disorder that is vaguely recognizable (largest difference peak 0.78 electrons) was ignored. The two disordered moieties were restrained to have similar geometries. Uij components of ADPs for disordered atoms closer to each other than 2.0 Angstrom were restrained to be similar. Subject to these conditions the occupancy ratio refined to 0.914 (3) to 0.086 (3). The structure contains additional 121 Ang3 of solvent accessible voids filled by extensively disordered solvate molecules (presumably nitromethane, the solvate of crystallization). The residual electron density peaks are not arranged in an interpretable pattern. The structure factors were instead augmented via reverse Fourier transform methods using the SQUEEZE routine (P. van der Sluis & A.L. Spek (1990). Acta Cryst. A46, 194-201) as implemented in the program Platon. The resultant FAB file containing the structure factor contribution from the electron content of the void space was used in together with the original hkl file in the further refinement. (The FAB file with details of the Squeeze results is appended to this cif file). The Squeeze procedure corrected for 69 electrons within the solvent accessible voids, or around two nitromethane molecules.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å 2 )
x y z U iso */U eq Occ. (<1) 0.0410 (10) 0.0349 (9) 0.0153 (7) −0.0006 (8) 0.0008 (6) −0.0020 (6) C4 0.0342 (9) 0.0331 (9) 0.0182 (7) 0.0033 (7  where P = (F o 2 + 2F c 2 )/3 (Δ/σ) max = 0.002 Δρ max = 0.77 e Å −3 Δρ min = −0.40 e Å −3 Absolute structure: Classical Flack method preferred over Parsons because s.u. lower Absolute structure parameter: −0.021 (7) Special details Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes. Refinement. The perchlorate ion was refined as disordered by a slight rotation. The two disordered moieties were restrained to have similar geometries. Uij components of ADPs for disordered atoms closer to each other than 2.0 Angstrom were restrained to be similar. Subject to these conditions the occupancy ratio refined to 0.540 (19)  0.069 (7) 0.078 (9) 0.099 (9) 0.024 (7) 0.016 (7) 0.012 (7)  where P = (F o 2 + 2F c 2 )/3 (Δ/σ) max < 0.001 Δρ max = 0.54 e Å −3 Δρ min = −0.33 e Å −3 Special details Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å 2 )
x y z U iso */U eq