Dinuclear lanthanide ( III ) / zinc ( II ) complexes with methyl 2-pyridyl ketone oxime †

Dinuclear and polynuclear M or /Ln coordination cluster complexes, where M or III are paramagnetic 3d-metal ions and Ln is a trivalent lanthanide ion, occupy a unique place among mixed-metal molecular materials as a result of the interaction between 3d and 4f electron systems giving rise, for example, to alternatives to homometallic 3d-metal SingleMolecule Magnets (SMMs) and magnetic refrigerants. An important feature here is the fact that 4f metal ions can contribute a large spin and, for most 4f ions, also the magnetic anisotropy needed for SMM behaviour. Furthermore, the coupling between 3d and 4f metal ions can be relatively strong in terms of superexchange interaction. However, in contrast to the plethora of studies concerning discrete 3d/4f-metal complexes, where both metal ions are paramagnetic, there is rather limited information on complexes containing Zn (a diamagnetic 3d metal ion) and paramagnetic Ln centres. Such complexes are extremely useful because (i) they can help scientists to elucidate the Ln⋯Ln magnetic exchange interactions in a series of isostructural MxLn III y coordination clusters (M = Mn, Fe, Co, Ni, Cu; y ≥ 2) and (ii) they often exhibit interesting photoluminescence properties and phenomena distinctly different from those of analogous complexes containing only Ln ions. From a synthetic inorganic chemistry viewpoint, methods must be devised to combine 3dand 4f-metal ions within a dinuclear or polynuclear molecule. One of our preferred routes is a “one-pot” procedure involving a mixture of 3dand 4f-metal starting materials and an organic ligand possessing distinct functionalities, or “pockets”, for preferential bonding of the 3d and 4f ions. For example, the various anionic 2-pyridylmonoximes (Scheme 1) have been widely used to date in the synthesis of structurally and magnetically interesting 3d-, 3d/3d′9 and 3d/4f-metal complexes [M = paramagnetic 3d-metal ion]. However, there are no reports of their use in Zn/Ln chemistry. These ligands are, in fact, attractive for Zn/Ln chemistry because the hard, deprotonated O atom will favour binding to strongly oxophilic Ln ions, whereas the softer 2-pyridyl and oximate N atoms will favour coordination to the Zn centre. We have been recently involved in a new research programme aiming to prepare, characterize and study discrete (i.e. non-polymeric), mixed Zn/Ln coordination cluster complexes with 2-pyridylmonoximate (Scheme 1) and 2,6-pyridylbisoximate bridging ligands. Our short-term goal is to

Dinuclear and polynuclear M II or III /Ln III coordination cluster complexes, where M II or III are paramagnetic 3d-metal ions and Ln III is a trivalent lanthanide ion, 1 occupy a unique place among mixed-metal molecular materials as a result of the interaction between 3d and 4f electron systems giving rise, for example, to alternatives 2 to homometallic 3d-metal Single-Molecule Magnets (SMMs) 3 and magnetic refrigerants. 4An important feature here is the fact that 4f metal ions can contribute a large spin and, for most 4f ions, also the magnetic anisotropy needed for SMM behaviour.Furthermore, the coupling between 3d and 4f metal ions can be relatively strong in terms of superexchange interaction.
However, in contrast to the plethora of studies concerning discrete 3d/4f-metal complexes, where both metal ions are paramagnetic, there is rather limited information on complexes containing Zn II (a diamagnetic 3d 10 metal ion) and paramagnetic Ln III centres. 5Such complexes are extremely useful because (i) they can help scientists to elucidate the Ln III ⋯Ln III magnetic exchange interactions in a series of isostructural M II x Ln III y coordination clusters (M = Mn, Fe, Co, Ni, Cu; y ≥ 2) 1c,6 and (ii) they often exhibit interesting photoluminescence properties and phenomena 7 distinctly different from those of analogous complexes containing only Ln III ions.
From a synthetic inorganic chemistry viewpoint, methods must be devised to combine 3d-and 4f-metal ions within a dinuclear or polynuclear molecule.One of our preferred routes is a "one-pot" procedure involving a mixture of 3d-and 4f-metal starting materials and an organic ligand possessing distinct functionalities, or "pockets", for preferential bonding of the 3d and 4f ions.For example, the various anionic 2-pyridylmonoximes (Scheme 1) have been widely used to date in the synthesis of structurally and magnetically interesting 3d-, 8 3d/3d′-9 and 3d/4f-metal 10 complexes [M II = paramagnetic 3d-metal ion].However, there are no reports of their use in Zn II /Ln III chemistry.These ligands are, in fact, attractive for Zn II /Ln III chemistry because the hard, deprotonated O atom will favour binding to strongly oxophilic Ln III ions, whereas the softer 2-pyridyl and oximate N atoms will favour coordination to the Zn II centre.
We have been recently involved in a new research programme aiming to prepare, characterize and study discrete (i.e.non-polymeric), mixed Zn II /Ln III coordination cluster complexes with 2-pyridylmonoximate (Scheme 1) and 2,6-pyridylbisoximate bridging ligands.Our short-term goal is to establish routes for such complexes and to isolate the maximum number of products from a given ligand.Our longerterm goal is to force the carefully designed Zn II -ligand moiety of the heterometallic complex to act as an efficient sensitizer, and one that is more efficient than the organic-only chromophore in exciting Ln III ions (mainly Eu III , Tb III and Dy III ) for emission in the visible region of the spectrum. 11It is also well known 12 that highly luminescent Ln III complexes are of interest for a wide variety of photonic applications such as planar waveguide amplifiers, light-emitting diodes and bio-inspired luminescent probes.Most of the electronic transitions of the Ln III ions involve a redistribution of electrons within the 4f sub-shell.Electric dipole selection rules forbid such transitions but these rules are relaxed by several mechanisms, such as coupling with vibrational states, J-mixing and mixing with opposite-parity wave functions (5d orbitals, ligand orbitals or charge transfer states).The coupling between these vibrational and electronic states and the 4f wavefunctions depends on the strength of the interaction between the 4f orbitals and the surrounding ligands; in view of the shielding of the 4f orbitals, the degree of mixing remains small, and so are the oscillator strengths of the f-f transitions.12b As a consequence, even if many Ln III compounds display a good quantum yield, direct excitation of the Ln III ions rarely yields highly luminescent materials.This disadvantage may be overcome by employing suitable organic 12,13 or d-block 14 (making use of fully-allowed, low-energy charge transfer transitions from p-character systems to d-character systems, e.g.luminescent complexes of d 6 and d 8 metal ions) chromophores as antenna groups to generate sensitised emission from Ln III ions.In this communication we describe our preliminary efforts towards the realisation of the short-term goal mentioned above, i.e. to establish the chemistry of the Zn II /Ln III /2-pyridylmonoxime system.
Reactions of Zn(ClO The metal ions are bridged by the three oximate groups of the η 1 :η 1 :η 1 :μ mpko − ligands.The Zn II ion is coordinated by six nitrogen atoms belonging to the "chelate" part of these ligands.This metal ion has a facial distorted octahedral geometry, the trans coordination angles being in the range 155.2(3)-158.4(3)°.The Eu III centre is bound to an O 3 N 6 set of donor atoms.The oxygen atoms (O4, O5, O6) belong to the deprotonated oximate groups of the {Zn(mpko) 3 } − unit, while the six nitrogen atoms belong to the three bidentate chelating (η 1 :η 1 ) mpkoH ligands.Using the Continuous Shape Measure (CShM) approach, 15 the coordination geometry of Eu III can be best described as spherical capped square antiprismatic (Fig. S1 †), the capping atom being N5; since the CShM value for this geometry (1.80735) is very close to that for the spherical tricapped trigonal prismatic geometry (1.80967), the Eu III ion can also be considered as having the latter.Most 9-coordinate metal ions possessing three 5-membered chelating rings and three monodentate ligands (the three monodentate "ligands" for Eu are the terminally coordinated oximate oxygen atoms O4, O5 and O6) form polyhedra that appear very close to the interconversion path between the capped square antiprism and tricapped trigonal prism. 15here are three strong intramolecular (intracationic) hydrogen bonds with uncoordinated oxime oxygens (O1, O2, O3) as donors and the coordinated oximate oxygens (O6, O4 and O5, respectively) as acceptors.There are also H-bonded parallelograms of the type HOH⋯O(ClO 2 )O⋯HOH⋯O(ClO 2 )O (Fig. S2 †) and single HOH⋯O(ClO 3 ) H bonds in the crystal structure of the complex, which is further stabilized by weak intermolecular π-π stacking interactions and C-H⋯π interactions to form 2D honeycomb layers parallel to the plane formed by c and the bisection line of the a0b.
Complex 4 crystallizes in the monoclinic space group Cc.Its structure consists of dinuclear molecules [ZnDy(NO 3 ) 2 -(mpko) 3 (mpkoH)].The molecular structure of 4 is similar to that of the cation [ZnEu(mpko) 3 (mpkoH) 3 ] 2+ , the main difference being the replacement of two N,N′-bidentate chelating mpkoH ligands of the latter with two bidentate chelating nitrato groups in the former (Fig. 2).This replacement gives neutral molecules (and not cations) in 4 and the Dy III centre is thus bound to an O 7 N 2 donor set.The coordination polyhedron of Dy III can be described as a spherical capped square antiprism (Fig. S3 †) with the oxime nitrogen N8 as the capping atom, although descriptions as a spherical tricapped trigonal prism and muffin are equally acceptable.Again there is evidence for a strong intramolecular O4-H(O4)⋯O3 H bond; two  Upon maximum excitation at 400 nm, solid 4 displays photoluminescence at 448, 483 and 540 nm at room temperature (Fig. 3).The most probable origin of emission is ligandbased as the excitation and emission spectra of both mpkoH and 4 are observed in the same region.Thus, somewhat to our disappointment, no significant Dy III emission was detected.Almost identical excitation and emission spectra are observed for solid samples of 1•2H 2 O and 3•2H 2 O (Fig. S5 and S6 †), supporting our view that no Ln III emission appears.Although not desirable, this is not an unusual situation for lanthanide(III) complexes with certain types of organic ligands. 16In our case, this means that the energy transfer from the organic ligands to Ln III is not efficient, probably because the energy levels of the excited states of these ions lie higher than that of the excited state of the {Zn-mpko − }moiety; this requires a change of the ligand in the Zn II moiety of the complex.
The room temperature χ M T product for 4 (14.30cm 3 K mol −1 ) under an applied dc field of 1000 Oe is consistent with the expected value of 14.17 cm 3 K mol −1 for one isolated Dy III ion ( 6 H 15/2 free ion, S = 5/2, L = 5, g J = 4/3).The product decreases slowly to a value of 12.58 cm 3 K mol −1 at 2.2 K, before a small upturn at 2.0 K (Fig. 4, left).The decrease in χ M T is due to the progressive depopulation of the Dy III excitedstate Stark sublevels. 16The small upturn below 2 K could be due to a weak ferromagnetic interaction between the complexes. 17The field dependence of magnetization shows that the magnetization reaches 5.4 μ B at 70 kOe after a rapid increase at low fields (Fig. 4, right).The observed non-saturated magnetization at the highest field is much lower than the expected 10 μ B for the Dy III ion indicating some anisotropy in the system. 17,18he dynamic magnetic properties of 4 were probed using ac susceptometry.Practically no signals for the out-of-phase component of the ac susceptibility were observed in the absence of a dc field at 1.8 K (Fig. S7 †).However, an intense signal is observed with the application of an external dc field of 1000 Oe.Thus, on application of this field the positions of the maxima of the out-of-phase signals become strongly frequency-dependent (Fig. 5) as expected for a Single-Ion Magnet (SIM).The application of this field suppresses fast zero-field tunnelling of the magnetization, which is a well-documented behaviour for Ln III -based SIMs and SMMs. 17To calculate the characteristic time and the barrier to relaxation of 4, the relaxation times were fitted with the frequencies occurring at the χ″ max of the frequency-dependent ac susceptibility data (Fig. 5, left) by using the Orbach thermally activated relaxation law τ = τ 0 exp(U eff /k B T ).Linear data following this law were only obtained between 4 K and 5.5 K (solid line, Fig. 5, right), with an effective energy barrier of U eff = 33.3K and τ 0 = 2.0 × 10 −7 s.This suggests that the relaxation might follow a quantum regime below 4 K or that there is more than one thermally activated relaxation process in the complex.Thus, complex 4 can   be termed as an "emissive field-induced SMM/SIM".1c, 19 The observation of more than one well-resolved pathway for magnetic relaxation remains a rarity among mononuclear SMMs. 20

Conclusions
In conclusion, the members of the first two families of Zn II / Ln III /2-pyridylmonoximate complexes have been prepared.The complexes exhibit ligand-based blue-green photoluminescence, while Zn II /Dy III is considered to be a fieldinduced single-ion magnet.We are currently trying to investigate the possible presence of members of a third family of products in the Ln(NO
weak C-H⋯π interactions (Fig. S4 †) extend the structure into two dimensions forming layers parallel to the ab plane and giving a sql topology.The lanthanide(III)-N(mpkoH) and -O(oximato) bond lengths around Eu III and Dy III in 1•2H 2 O and 4 show the expected tendency towards shorter values in the latter in line with the lanthanide(III) contraction.Compounds 1•2H 2 O and 4 are the first structurally characterized heterometallic Zn II /Ln III complexes with 2-pyridyloxime/oximate ligation.

Fig. 4
Fig. 4 Temperature dependence of χ M T for 4 (left) and molar magnetization vs. field at indicated temperatures (right, the solid lines are guides for the eyes).
3 ) 3 •xH 2 O/Zn(ClO 4 ) 2 •xH 2 O/mpkoH general reaction system (we have strong evidence for this).Work is also in progress to enhance the aromatic content of the dinuclear Zn II Ln III complexes by employing anionic phenyl 2-pyridine ketone oxime (R = Ph in Scheme 1) as a bridging ligand and PhCO 2 − (instead of NO 3 − or mpkoH) as terminal groups, with the hope of achieving efficient energy transfer from the Zn IIligand moiety to the lanthanide and "switch on" Ln III -based emission.Achieving Dy III -based emission in a Zn II Dy III SIM would enable us to correlate luminescence and magnetism 20a,21 since, in theory, the highest energy f-f transitions in the emission spectra of mononuclear Dy III SIMs can be modelled to provide a direct picture of the splitting of the ground J multiplet.

Fig. 5
Fig. 5 Frequency dependence of the out-of-phase component of the ac susceptibility under an external dc field of 1000 Oe at the indicated temperatures (left) and the relaxation time of 4 as a function of temperature (dots) plotted against a thermally activated Arrhenius law (solid line) (right).