Tb2, Dy2, and Zn2Dy4 Complexes Showing the Unusual Versatility of a Hydrazone Ligand toward Lanthanoid Ions: a Structural and Magnetic Study

1 Tb 2 , Dy 2 and Zn 2 Dy 4 Complexes Showing the Unusual Versatility of a Hydrazone Ligand towards Lanthanoid Ions: a Structural and Magnetic Study Matilde Fondo, §,* Julio Corredoira-Vázquez, § Ana M. García-Deibe, § Jesús Sanmartín- Matalobos, § Juan Manuel Herrera, £ and Enrique Colacio £ § Departamento de Química Inorgánica, Facultade de Química, Universidade de Santiago de Compostela, Campus Vida, 15782 Santiago de Compostela, Spain. £ Departamento de Química Inorgánica, Facultad de Ciencias, Universidad de Granada, Avda. Fuentenueva s/n, 18071 Granada, Spain. KEYWORDS: H 4 daps, hydrazone ligand, dinuclear lanthanoid complex; heteronuclear zinclanthanoid complex, single molecular magnet. ABSTRACT: Three dinuclear complexes: [Tb 2 (H 2 daps) 2 (CH 3 OH) 5.5 (H 2 O) 0.5 ](Cl) 2 (Tb 2 Cl), [Dy 2 (H 2 daps) 2 (Cl)(CH 3 OH) 3 (H 2 O)]Cl (Dy 2 Cl), and [Dy 2 (H 2 daps) 2 (H 2 O) 6 ](CF 3 SO 3 ) 2 (Dy 2 CF 3 SO 3 ), as well as the heterohexanuclear compound [Zn 2 Dy 4 (daps) 2 (Hdaps)(Cl) 3 (OH) 2 (CH 3 OH)(H 2 O) 5 ] (Zn 2 Dy 4 ) (being H 4 daps: 2,6-bis(1- salicyloylhydrazonoethyl)pyridine), which crystallize with different lattice molecules, show their hexa- or heptadentate hydrazone ligands acting with hitherto unknown μ 2 -ț 5 :ț 1 ; μ 3 -ț 5 :ț 4 :ț 1 ; μ 4- ț 5 :ț 4 :ț 1 :ț 1 ; or μ 4- ț 5 :ț 2 :ț 1 :ț 1 bridging modes. The single X-ray crystal structures of the dinuclear complexes show nine-coordinated N 3 O 6 environments for lanthanoid atoms in Tb 2 Cl and Dy 2 CF 3 SO 3 , with distorted geometries, between spherical capped square antiprism and muffin-like, while the dysprosium atoms in Dy 2 Cl are eight- 2 coordinated, with distorted triangular dodecahedron geometries. In the case of Zn 2 Dy 4 , both eight-coordinated, O 8 and O 7 Cl, as well as nine-coordinated N 3 O 6 environments coexist in the crystal structure, with biaugmented trigonal prism, triangular dodecahedron and muffin-like pseudo-geometries. The magnetic study of all the complexes shows that none of the pure samples behaves as SMM, and that the quantum tunnel cannot be removed by dilution of none of the dinuclear complexes, but the diluted [Y 1.904 Dy 0.096 (H 2 daps) 2 (Cl)(H 2 O) 4 ]Cl (Dy 2 Cl@Y) behaves as a weak field induced SMM. The heterohexanuclear Zn 2 Dy 4 complex also lacks slow relaxation of the magnetization. INTRODUCTION The discovery of the first lanthanoid single ion magnet (SIM) by Ishikawa et al. 1 in 2003 ignited broad interest in the study of the organometallic and coordination chemistry of the lanthanoid ions, due to their potentiality as molecular magnets. 2,3 Mononuclear compounds of f-elements contain formally all the ingredients to behave as single molecular magnets (SMMs): large magnetic moments, magnetic anisotropy, and a bistable ground state under the appropriate conditions. However, the anisotropy needs to be axial to trap the magnetic moment in two opposite directions. In this way, many efforts have been made during the last decade in order to control and exploit the single ion anisotropy in the design of lanthanoid SMMs. 4 These studies led to important advances in the field, with growing records of anisotropic energy barriers (U eff ) and blocking temperatures (T B ). Accordingly, the highest U eff described up to now is 1815 K, 5 a value significantly higher than 331 K reported for the first SIM (Bu t4 N)[Tb(Pc) 2 ]. 1 The continuous attainments related to blocking temperatures are even more interesting. Thus, in 2011 J. R. Long et al. 6 achieved a milestone reporting a dinuclear terbium complex with a record T B of 14 K, which was the limit of blocking temperature for SMMs reported 3 until 2016. However, since 2016, this blocking temperature roof was exceeded several times, firstly by the complex [Dy(Cy 3 PO) 2 (H 2 O) 5 ]Br 3, which established the new record T B in 20 K, 7 and finally by the recently reported compound [Dy(Cp ttt ) 2 ][B(C 6 F 5 ) 4 ], with an astounding blocking temperature of 60 K. 8 This T B is tantalizingly close to liquid nitrogen temperature (77 K), so new advancements would make SMMs commercially for storage and processing of digital information in data servers. This latter attainment comes to partially silence those critical voices that claim that it is not possible to use SMMs for real-world applications. Nevertheless, although the field of the molecular magnets seems to be on the right path, there are still some obstacles that must be overcome for using SMMs as commercial high density storage. Apart from the blocking temperature, which should increase even more, another barrier that must also be taken into account is the air-stability of the SMMs reported. In this way, most of the reported molecular magnets with high blocking temperatures are organometallic and/or air-sensitive complexes, what usually involves working under inert conditions. 1,5,7 However, air-stable coordination compounds with SMM behavior, and high blocking temperatures, and/or energy barriers have been also reported. 6,9-10 Therefore, coordination chemistry seems an appealing alternative in the search for air-stable SMMs with elevated T B . With the aforementioned considerations in mind, we have decided to explore the coordination chemistry of the well-known polydentate, and highly stable, ligand 2,6-bis(1- salicyloylhydrazonoethyl)pyridine (H 4 daps, Scheme 1), 11-15 in this occasion, towards dysprosium(III) and terbium(III) ions. When combined with transition or main group metal ions, this potentially heptadentate ligand typically behaves as diprotic, pentadentate and mononucleating (Scheme 1a), 11 with some remarkable exceptions with daps 4- acting as a μ 2 -ț 5 :ț 4 dinucleating ligand (Scheme 1b), which has been only reported once, 12 and more commonly, acting as a doubly deprotonated μ 2 -ț 3 :ț 3 or μ 2 -ț 3 :ț 2 helicand (Scheme 1c). 13 In 4 addition, the pentadentate N 3 O 2 chelating behavior (Scheme 1a) also prevails for the only three complexes mononuclear with lanthanoids containing this ligand reported in literature. 14,15 By contrast, the results reported in this work show an unforeseen versatility and polynucleating ability of this classic ligand. The static and dynamic magnetic behavior of all the new complexes is also reported and compared. 1 11 11 22 0 1 11 11 22 22 )H )H 22 12FKHODWLQJ 122 EULGJLQJ EULGJLQJ 121 2 RU 1212 GHSHQGLQJ RQ WKH 01S\ GLVWDQFH 1 1 1 2 2+ 0 11 2 +2 0 D E F ++  Scheme 1. Coordination modes reported for complexes derived from H 4 daps. a) doubly deprotonated and pentacoordinated in mononuclear complexes; 11,14,15 b) totally deprotonated and dinucleating; 12 and c) doubly deprotonated and dinucleating helicand. 13 EXPERIMENTAL Materials and general methods. All chemical reagents were purchased from commercial sources, and used as received, without further purification. Elemental analyses of C, H and N were performed on a Carlo Erba EA 1108 analyzer. Infrared spectra were recorded in the ATR mode on a Varian 670 FT/IR spectrophotometer in the range 4000-500 cm -1 . 1 H NMR spectra of H 4 daps and Na 4 daps were recorded on a Bruker DPX-250 spectrometer . Syntheses of the ligands. H 4 daps was obtained by a variation of a previously described method, 11a using methanol instead of benzene as solvent, and satisfactorily characterized by 1 H NMR and IR spectroscopies. 1 H NMR (DMSO-d 6 , δ in ppm): CH 3 signal hidden under a 5 DMSO peak, 6.95-7.02 (m, 4H), 7.41 (t, 2H), 7.90-7.99 (m, 3H), 8.13 (d, 2H), 11.47 (s, 2H), 11.81 (s, br, 2H). %. IR (ATR, Q ෤ /cm -1 ): 3300 (OH), 3180 (NH), 1641 (amide I), 1553 (amide II). Na 4 daps·2H 2 O·CH 3 OH was prepared by reaction in methanol of H 4 daps with NaOH in 1:4 molar ratio, and subsequently characterized as C 23 H 17 N 5 Na 4 O 4 ·2H 2 O·CH 3 OH. Elemental analysis calcd. for C 2 H 25 N 5 Na 4 O 7 (587,44): C,  49,07; N 11.92; H 4.29%. Found: C, 49.04; N, 11.80; H, 4.13%. 1 H NMR spectroscopy (DMSO-d 6 , δ in ppm): 2.31 (s, 6H), 6.38 (t, 2H), 6.50 (d, 2H), 7.04 (t, 2H), 7.69 (d, 2H), 7.80 (d, 2H), 7.93 (t, 1H). IR (ATR, Q ෤ /cm -1 ): 1596 (amide I), 1536 (amide II). The IR spectrum is shown in the supporting information (Figure S1). Syntheses of the complexes. [Tb 2 (H 2 daps) 2 (CH 3 OH) 5.5 (H 2 O) 0.5 ](Cl) 2 (Tb 2 Cl) and [Dy 2 (H 2 daps) 2 (Cl)(CH 3 OH) 3 (H 2 O)]Cl (Dy 2 Cl) were obtained by using a similar method, and which is exemplified by the isolation of Tb 2 Cl: To a suspension of H 4 daps (0.200 g, 0.464 mmol) in methanol (25 mL), (CH 3 ) 4 N(OH)·5H 2 O (0.173 g, 0.928 mmol) was added. This yellow solution was stirred for 10 min, and then TbCl 3 Â6H 2 O (0.173 g, 0.464 mmol) was added. The new mixture was stirred for 4 h at room temperature, with precipitation of a small portion of solid Tb 2 Cl (0.019 g). The solid was filtered and dried in air. The mother liquor was slowly evaporated in air, yielding orangey yellow crystals. Some single crystals were suitable for X-ray diffraction studies, demonstrating that they corresponded to Tb 2 Cl·3.5CH 3 OH·0.5H 2 O. The remaining crystalline sample was filtered and dried in air, with loss of the solvated molecules, also yielding Tb 2 Cl. Yield: 0.120 g (36%). [Tb 2 (H 2 daps) 2 (CH 3 OH) 5.5 (H 2 O) 0.5 ](Cl) 2 (Tb 2 Cl). Elemental analysis calcd. for C 51.5 H 61 Cl 2 N 10 O 14 Tb 2 (1432.89): C, 43.12; N 9.77; H 4.26%. Found: C, 43.15; N, 10.03; H, 4.43%. IR (ATR, Q෤/cm -1 ): 1589 (amide I), 1489 (amide II), 3344 (OH) (IR spectrum in Figure S2) 6 [Dy 2 (H 2 daps) 2 (H 2 O) 6 ](CF 3 SO 3 ) 2 ·2THF (Dy 2 CF 3 SO 3 ·2THF): To a suspension of H 4 daps (0.100 g, 0.232 mmol) in THF (25 mL)/acetonitrile (10 mL), (CH 3 ) 4 N(OH)·5H 2 O (0.083 g, 0.463 mmol) was added. After 10 min stirring, a yellow solution is obtained, and then Dy(CF 3 SO 3 ) 3 (0.144 g, 0.232 mmol) was added. The new mixture was stirred for 4 h at room temperature, yielding a yellow solution. This latter one was left to evaporate slowly, and after two days, single crystals of Dy 2 CF 3 SO 3 ·0.3H 2 O·2THF, suitable for X-ray diffraction studies, were obtained. The crystals were filtered and dried in air, with loss of the solvated water on drying, to finally obtain Dy 2 CF 3 SO 3 ·2THF. Yield: 0.055 g (27%). Elemental analysis calcd. for C 56 H 64 Dy 2 F 6 N 10 O 22 S 2 (1732.69): C, 38.79; N 8.08; H, 3.62; S, 3.69 %. Found: C, 38.59; N, 8.27; H, 3.45; S, 3.61%. IR (ATR, Q෤/cm -1 ): 637 (δSO 3 ), 1029, 1270 (SO 3 ), 1161, 1223 (CF 3 ), 1493 (amide II), 1589 (amide I), 3360 (OH). (Spectrum shown in Figure S3) As commented above, the isolation of Dy 2 Cl was analogous, but in this case the slow evaporation of the mother liquor gave rise to yellow single crystals of [Dy 2 (H 2 daps) 2 (Cl)(CH 3 OH) 3 (H 2 O)]Cl·1.75CH 3 OH (Dy 2 Cl·1.75CH 3 OH), suitable for X-ray diffraction studies. Similarly, after drying in air, the crystalline solid lacked lattice molecules. [Dy 2 (H 2 daps) 2 (Cl)(CH 3 OH) 3 (H 2 O)]Cl (Dy 2 Cl): quantity employed of DyCl 3 Â6H 2 O (0.174 g, 0.464 mmol). Yield: 0.083 g (26%). Elemental analysis calcd. for C 49 H 52 Cl 2 Dy 2 N 10 O 12 (1368.98): C, 42.95; N 10.23; H 3.80%. Found: C, 43.30; N, 10.27; H, 3.70%. IR (ATR, Q෤/cm -1 ): 1487 (amide II), 1584 (amide I), 3274 (OH) (IR spectrum in Figure S4). [Zn 2 Dy 4 (daps) 2 (Hdaps)(Cl) 3 (OH) 2 (CH 3 OH)(H 2 O) 5 ]·5H 2 O (Zn 2 Dy 4 ·5H 2 O): DyCl 3 Â6H 2 O (0.053 g, 0.140 mmol) was added to a methanol solution (20 mL) of Na 4 daps (0.050 g, 0.105 mmol), and this mixture was stirred at 35 ºC for 30 min, until obtaining a yellow solution. Likewise, a solution of ZnCl 2 (0.09 g, 0.070 mmol) was basified up to pH = 7.3, with a 0.1 M methanol solution of (CH 3 ) 4 N(OH). This slightly basic solution was then added to the first yellow solution, and the new mixture was refluxed for 2 h. After cooling at room 7 temperature, the resulting solution was slowly evaporated until formation of yellow crystals, corresponding to [Zn 2 Dy 4 (daps) 2 (Hdaps)(Cl) 3 (OH) 2 (CH 3 OH)(H 2 O) 5 ]·0.5CH 3 OH·5H 2 O (Zn 2 Dy 4 ·0.5CH 3 OH·5H 2 O), as single crystal X-ray diffraction studies have demonstrated. The remaining crystalline sample was filtered, and then dried in air, what leaded to the loss of the solvated methanol, and to the final isolation of Zn 2 Dy 4 ·5H 2 O. Yield: 0.028 g (33%). Elemental analysis calcd. for C 70 H 78 Cl 3 Dy 4 N 15 O 25 Zn 2 (2416.58): C, 34.76; N 8.69; H 3.23%. Found: C, 34.90; N, 8.65; H, 2.92%. IR (ATR, Q෤/cm -1 ): 1482, 1497 (amide II), 1564, 1594 (amide I), 3346 (OH). IR spectrum is shown in Figure S5 of the supporting information. Alternatively, another analogue solvate of Zn 2 Dy 4 could be also obtained by means of a similar method, but employing Na 4 daps, DyCl 3 Â6H 2 O and ZnCl 2 in 1:1:1 molar ratios. However, in this case the yield was sensibly lower. [Y 1.904 Tb 0.096 (H 2 daps) 2 (CH 3 OH) 5.5 (H 2 O) 0.5 ]Cl 2 (Tb 2 Cl@Y): To a suspension of H 4 daps (0.200 g, 0.464 mmol) in methanol (25 mL), (CH 3 ) 4 N(OH)·5H 2 O (0.173 g, 0.928 mmol) was added and the mixture was stirred at room temperature until a yellow solution was obtained. Then, YCl 3 Â6H 2 O (0.134 g, 0.442 mmol) and TbCl 3 Â6H 2 O (0.008 g, 0.022 mmol) were added and the new mixture was stirred at room temperature for 12 h and a yellow solid precipitated, which was filtered and dried in air. Yield: 0.254 g (43%). Elemental analysis calcd. for C 51.5 H 61 Cl 2 N 10 O 14 Tb 0.096 Y 1.904 (1299.67): C, 47.55; N 10.77; H 4.69%. Found: C, 47.52; N 10.85; H 4.81%. IR (ATR, Q෤/cm -1 ): 1589 (amide I), 1489 (amide II), 3344 (OH). [Y 1.904 Dy 0.096 (H 2 daps) 2 (Cl)(H 2 O) 4 ]Cl (Dy 2 Cl@Y) was obtained by the same method described for Tb 2 Cl@Y. Yield: 0.157 g (50%). Elemental analysis calcd. for C 46 H 46 Cl 2 Dy 0.096 N 10 O 12 Y 1.904 (1186.69): C, 46.51; N 11.80; H 3.88%. Found: C, 46.49; N, 11.91; H, 4.22%. (IR included in Figure S4). [Y 1.904 Dy 0.096 (H 2 daps) 2 (H 2 O) 6 ](CF 3 SO 3 ) 2 ·4H 2 O (Dy 2 CF 3 SO 3 @Y): was obtained by the same method described for Tb 2 Cl@Y. Yield: 0.29 g (82%). Elemental analysis calcd. for 8 C 48 H 66 Dy 0.096 F 6 N 10 O 24 S 2 Y 1.904 (1520.87): C, 37.87; N 9.20; H, 4.33; S, 4.21 %. Found: C, 37.79; N, 9.22; H, 4.13; S, 4.65%. IR (ATR, Q෤/cm -1 ): 637 (δSO 3 ), 1029, 1270 (SO 3 ), 1161, 1223 (CF 3 ), 1493 (amide II), 1589 (amide I), 3360 (OH). Crystallographic refinement and structure solution. Main crystal data and some details of refinements are given in the Supporting Information (Table S1). Magnetic measurements. Magnetic susceptibility dc and ac measurements for powder crystalline samples of Tb 2 Cl , Dy 2 Cl, Dy 2 CF 3 SO 3 ·2THF and Zn 2 Dy 4 ·5H 2 O were carried out with a Quantum Design SQUID MPMS-XL-5 susceptometer. The magnetic susceptibility data were recorded under magnetic fields of 1500 Oe in the range 2-300 K. Magnetization measurements at 2.0 K were recorded under magnetic fields ranging from 0 to 50000 Oe. Diamagnetic corrections were estimated from Pascal’s Tables. Alternating current (ac) susceptibility measurements in different applied static fields (H dc = 0 or 1000 Oe) were performed with an oscillating ac field of 3.5 Oe, and ac frequency of 1400 Hz for all the complexes. Similarly, the ac measurements were also recorded for diluted Tb 2 Cl@Y, Dy 2 Cl@Y and Dy 2 CF 3 SO 3 @Y . In the case of Dy 2 Cl@Y, ac susceptibility measurements were additionally recorded under a dc field of 1000 Oe at ac frequencies, ranging from 50 to 1400 Hz. RESULTS AND DISCUSSION Syntheses. Three homodinuclear complexes, Tb 2 Cl and Dy 2 Cl and Tb 2 CF 3 SO 3 ·2THF, were directly obtained, in the presence of (CH 3 ) 4 N(OH)·5H 2 O, by interaction of the H 4 daps ligand with the corresponding metal salt (chloride or triflate) in 1:1 molar ratios, as summarized in Scheme 2. Slow evaporation of their mother liquors allowed obtaining single crystals suitable for X-ray diffraction studies of these species. These crystals lose their lattice methanol and/or 9 water molecules on drying, what leaded to the isolation of Tb 2 Cl, Dy 2 Cl and Tb 2 CF 3 SO 3 ·2THF samples for magnetic studies. Attempts to isolate a heterodinuclear complex of the type ZnDy(daps)Cl, by mixing of the previously prepared Na 4 daps, DyCl 3 Â6H 2 O and ZnCl 2 in 1:1:1 molar ratios, were unsuccessful. However, and surprisingly, this reaction leaded to the obtaining of the heterohexanuclear complex [Zn 2 Dy 4 (daps) 2 (Hdaps)(Cl) 3 (OH) 2 (CH 3 OH)(H 2 O) 5 ]·5H 2 O (Zn 2 Dy 4 ·5H 2 O). The same compound can be also obtained, with a higher yield, by mixing Na 4 daps, DyCl 3 Â6H 2 O and ZnCl 2 in appropriate 3:4:2 molar ratios (Scheme 2). In contrast, similar reactions did not give rise to analogous Zn 2 Tb 4 or Zn 2 Gd 4 complexes, and the powdery samples obtained from these reaction mixtures could not be satisfactorily characterized. Scheme 2. Reaction scheme for the complexes derived from H 4 daps and reported here. All the unequivocally characterized compounds are yellowish solids that are apparently stable, both in solid state and in solution. Their formulations were confirmed by 16 Figure 4. Coordination polyhedron for Dy1 in Dy 2 Cl, illustrating the distorted triangular dodecahedron geometry shown by both dysprosium ions. The crystals structures of Dy 2 CF 3 SO 3 and Dy 2 Cl show a dinuclear nature that strongly contrasts with the other three mononuclear Dy III complexes obtained with the same ligand. 14,15 This is, the two nonacoordinated [Dy(H 4 daps)(H 2 O) 3 (NO 3 )](NO 3 ) 2 (H 2 O) and [Dy(H 3 daps)(H 2 O) 2 (NO 3 )](NO 3 )(CH 3 OH) complexes, 14 and the also mononuclear Et 3 NH[Dy(H 2 daps)Cl 2 ] 15 , which displays a seven-vertexed coordination polyhedron. The H 2 daps 2- ligands present in these three latter mononuclear complexes show a coordinating behavior according with that represented in Scheme 1a, with both phenol residues remaining uncoordinated. Apparently, this brings to light that not only the dysprosium salt, but also the solvents present in the reaction media, can play a crucial role in the structure of these coordination compounds. This is specially outstanding when Dy 2 Cl and Et 3 NH[Dy(H 2 daps)Cl 2 ], 15 are compared. Thus, the mixture of DyCl 3 and H 4 daps in 1:1 molar ratio, in the presence of two equivalents of a base, but in different solvents (methanol or absolute ethanol) leads to these two notably different structures. Zn 2 Dy 4· 0.5CH 3 OH·5H 2 O An ellipsoid diagram for neutral [Zn 2 Dy 4 (daps) 2 (Hdaps)(Cl) 3 (OH) 2 (CH 3 OH)(H 2 O) 5 ] is shown in Figure 5, and main distances and angles are summarized in the supporting information 17 (Tables S4 and S5), and where Figure S9 shows an ellipsoid diagram with the whole labelling scheme for this complex. Figure 5. Ellipsoids diagram (50% probability) for Zn 2 Dy 4 . Only metal and donor atoms have been labelled, for clarity. The asymmetric unit of the heterohexanuclear Zn 2 Dy 4 complex contains [Zn 2 Dy 4 (daps) 2 (Hdaps)(Cl) 3 (OH) 2 (CH 3 OH)(H 2 O) 5 ] neutral molecules, in addition to water and methanol molecules of solvation. Despite the molecules of the complex lack symmetry, some equivalences can help to describe this intricate structure. In fact, Zn 2 Dy 4 can be considered as formed by two [ZnDy(daps)(Cl)(OH)(H 2 O) 2 ] - moieties that are connected through a central [Dy 2 (Hdaps)(Cl)(CH 3 OH)(H 2 O)] 2+ fragment. We will refer to these two external moieties as unit ZnDy1 and unit ZnDy2, as they include the Dy1 or Dy2 atoms (Scheme 3b and 3d), while the central unit containing Dy3 is represented in Scheme 3c. 18 Each ZnDy unit clearly reminds to that tetradeprotonated and dinucleating behavior represented in Scheme 1b, and only reported once in the literature. 12 In our case, the daps 4- ligand accommodates a Zn II ion in its N 3 O 2 internal pocket, with an apical chloride ligand to form a pseudo-pentagonal pyramid with some umbrella-like distortion in its base. In addition, the four external O atoms are coordinated to a Dy atom (Dy1 or Dy2), which also holds an exogenous hydroxide donor (OXH, X = 1 or 2), and two water molecules (OX0W, OX1W, X = 1 or 2), with Zn···Dy distances of ca. 3.73 Å. All the H atoms potentially involved in tautomerism and H bonds were located in Fourier maps. The difference between the ZnDy1 and ZnDy2 units is the additional coordination of one (unit ZnDy1, Scheme 3b), or two dysprosium ions (unit ZnDy2, Scheme 3d) through either one or two of its external phenolate atom(s). These novel μ 3- ț 5 :ț 4 :ț 1 and μ 4- ț 5 :ț 4 :ț 1 :ț 1 coordination modes are represented in the Scheme 3b and 3d as indicated. In the case of the central [Dy 2 (Hdaps)(Cl)(CH 3 OH)(H 2 O)] 2+ fragment, the ligand is strangely triply deprotonated, with a Dy atom (Dy3) in its internal N 3 O 2 compartment, while doubly deprotonated arm are chelating a second Dy atom (Dy4) through O30 and O31 atoms. This latter Dy ion is also coordinated to one water molecule (O40W), one methanol donor (O40S) and a chloride ion (Cl4). In addition, O31 and O32 atoms are acting as additional bridges to the neighboring Dy atoms of the two ZnDy units, while phenol O33 atom remains protonated and uncoordinated. Hence, this unforeseen Hdaps -3 ligand is simultaneously bridging the four dysprosium atoms, with an also novel μ 4 -ț 5 :ț 2 :ț 1 :ț 1 mode (Scheme 3c). The novelty of the varied coordination behavior shown by the three hydrazone ligand units present in Zn 2 Dy 4 (Scheme 3b-d) has been checked in the CSD files. 22 The assembly of these three different moieties, with the presence of their exogenous ligands, fulfills the steric requirements of the six metal ions, and it allows the formation of the whole complex. Worth of mention in this intricate arrangement is also the presence of three 19 μ 3 bridging O atoms: two from hydroxide ions and the alkoxide O31 atom, which are connecting the four Dy III ions. (Figure 6). Figure 6. Left: coordination environments of the four Dy atoms in Zn 2 Dy 4 , and right: a detail of the Dy 4 O 7 core, showing the four μ 2 and the three μ 3 bridges that connect the four ions, indicating the intramolecular Dy···Dy distances. The presence of diverse ligands and coordination behaviors gives rise to varied coordination environments around the lanthanoid atoms; Dy1 and Dy2 display O 8 environments, while Dy4 shows an O 7 Cl one. By contrast, Dy3 is surrounded by an N 3 O 6 coordination sphere (Figure 6). The distortions from the ideal eight or nine-coordinate geometries were calculated by means of SHAPE software. 17 The results indicate that Dy1 is in a biaugmented trigonal prism environment, Dy2 and Dy4 have a triangular dodecahedron geometry, while Dy3 is in a muffin-like environment (Table S3). Magnetic Properties. The direct-current (dc) magnetic susceptibility studies of all the complexes were performed under a field of 1000 Oe in the temperature range 2–300 K. The plots of χ M T versus T for the dinuclear complexes are shown in Figure S10. At 300 K, the χ M T products for the three products (22.60, 31.64 and 30.42 cm 3 K mol −1 ) are compatible 20 with those expected for two uncoupled Tb III (4f 8 , 7 F 6 ) or Dy III (4f 9 , 6 H 15/2 ) ions. In all cases, the experimental χ M T values decrease upon cooling, with a gradual decrease for Dy 2 Cl, whereas the diminishment is much more pronounced at low temperature for Tb 2 Cl and Dy 2 CF 3 SO 3 ·2THF. The minimum values reached at 2 K are 13.49, 20.40 and 15.21 cm 3 K mol −1 for Tb 2 Cl, Dy 2 Cl and Dy 2 CF 3 SO 3 ·2THF, respectively. This behavior could be attributed to the thermal depopulation of the Stark sublevels of the Ln III ions, 23 and/or to weak intramolecular antiferromagnetic interactions between the spin carriers. 24 The field dependence of the magnetization at 2 K for the all the dinuclear complexes (Figure S10) shows their similar behaviors, with a relatively rapid increase of the reduced magnetization at low field without reaching saturation at 5 T. The magnetization value at the maximum applied field of 5 T (in the 9.36–10.63 Ms/Nμ B range) is significantly lower than that expected for two isolated Tb III or Dy III ions (Ms/Nμ B = g j J), which can be attributed to the crystal-field effects that lead to a substantial magnetic anisotropy. Alternating current (ac) magnetic susceptibility measurements were recorded for the three dinuclear complexes, in order to know the low-temperature magnetic relaxation behavior of the samples. None of these compounds shows in-phase (Ȥ’) nor out-of-phase (ȤƎ) ac susceptibilities peaks at a frequency of 1400 Hz in the absence (0 Oe), or in the presence of a magnetic dc field of 1000 Oe, as exemplified for Dy 2 Cl in Figure S11 of the supplementary information. This fact indicates that either there is a fast quantum tunnelling (QTM) relaxation process, which cannot be suppressed by the application of a small static magnetic field, or the thermal energy barrier is so small that no maxima are observed above 2 K for none of the compounds. In a second attempt to remove the QTM, the Tb 2 Cl, Dy 2 Cl and Dy 2 CF 3 SO 3 ·2THF samples were diluted in order to eliminate dipolar interactions. 25 Accordingly, in-phase and out-of-phase susceptibility measurements at 1400 Hz were carried out on the diluted samples 21 [Y 1.904 Tb 0.096 (H 2 daps) 2 (CH 3 OH) 5.5 (H 2 O) 0.5 ]Cl 2 (Tb 2 Cl@Y), [Y 1.904 Dy 0.096 (H 2 daps) 2 (Cl)(H 2 O) 4 ]Cl (Dy 2 Cl@Y) and [Y 1.904 Dy 0.096 (H 2 daps) 2 (H 2 O) 6 ](CF 3 SO 3 ) 2 ·4H 2 O (Dy 2 CF 3 SO 3 @Y), which were prepared using a Ln/Y 1:20 molar ratio (see experimental section). These measurements were recorded both in the absence and in the presence of a dc field of 1000 Oe. In the absence of a magnetic field, none of the diluted samples show a net peak. This is also true for Tb 2 Cl@Y and Dy 2 CF 3 SO 3 @Y under a magnetic field of 1000 Oe. By contrast, when Dy 2 Cl@Y is subjected to the presence of an external dc field of 1000 Oe, two clear peaks at ca. 12 and 4 K are observed (Figure 7). Figure 7. Temperature dependence of χ M ’’ for Dy 2 Cl@Y in a dc applied field of 1000 Oe at different frequencies. In view of these facts, it can be concluded that the intermolecular forces and dipolar interactions play an important role in the quantum tunnelling (QTM) for Dy 2 Cl@Y, given that its suppression enables observation of the slow relaxation of the magnetization. Interestingly, this complex exhibits two thermally activated relaxation processes, probably associated with the existence of two distinct anisotropic centers Dy1 and Dy2. 26 Nevertheless, when the data are recorded at frequencies below 800 Hz, the response is too weak and, 22 therefore, the relaxation times and U eff could not be calculated. As a result, Dy 2 Cl@Y can be considered as a weak field induced SMM. It could appear surprising that Dy 2 CF 3 SO 3 @Y does not show slow relaxation of the magnetization, but the crystal structures of Dy 2 Cl and Dy 2 CF 3 SO 3 ·2THF described above reveal a notable difference in the dysprosium environments: eight-coordinated in Dy 2 Cl and nine-coordinated in Dy 2 CF 3 SO 3 . In addition, the Dy III centers in Dy 2 Cl are bridged by a single OCCCO fragment while in Dy 2 CF 3 SO 3 two bridging OCCCO moieties connect the Dy atoms. This could promote a more intense, although weak, antiferromagnetic interaction in Dy 2 CF 3 SO 3 , which could give rise to fast relaxation of the magnetization, and damp the SMM behavior. 27,28 The lower χ M T value for Dy 2 Cl at 2 K compared with that of Dy 2 CF 3 SO 3 ·2THF (20.40 and 15.21 cm 3 K mol −1 , respectively) is also compatible with this supposition. Besides, it also should be remembered that Dy 2 Cl is composed of two different blocks while both units are equal and symmetry related by an inversion center in Dy 2 CF 3 SO 3 . This may be a fundamental difference because the anisotropy axes in both Dy III ions in Dy 2 Cl, calculated with the MAGELLAN software, 29 are not parallel (Figure 8), and do not cancel each other while in Dy 2 CF 3 SO 3 both axes must have the same direction but different signs, given that they are related by an inversion center, and this could favor the cancelling of the magnetic anisotropy. 30 Figure 8. Calculated orientations of the local main magnetic axes (blue) on Dy III ions for Dy 2 Cl using MAGELLAN software. 29 23 For Zn 2 Dy 4 ·5H 2 O, the χ M T value at room temperature is 56.75 cm 3 K mol −1 (Figure S10d, supporting information), in accordance which the expected value of 56.68 cm 3 K mol −1 for four uncoupled Dy III ions ( 6 H 15/2 , g = 4/3). As the temperature decreases, the χ M T value gradually diminishes to reach a minimum of 37.03 cm 3 K mol −1 at 2 K, behavior that can be mainly ascribed to the depopulation of the Stark sublevels. 23 Weak antiferromagnetic coupling between metal centers can also contribute to this behavior but cannot be easily quantified because of the strong orbital momentum of the Dy III ions. Nevertheless, the rather large χ M T observed at 2 K (37.04 cm 3 K mol −1 ) allows suggesting that exchange interactions between these Dy III centers, if any, are weak. Besides, the magnetization value at the maximum applied field of 5 T is 18.83 Ms/Nμ B , which is significantly lower than that expected for four isolated Dy III ions (Ms/Nμ B = g j J), could be in agreement with a significant magnetic anisotropy. To probe the dynamic magnetic behavior of Zn 2 Dy 4 ·5H 2 O, the temperature dependence of in-phase and out-of-phase ac magnetic susceptibility was examined at a frequency of 1400 Hz. The results reveal that Zn 2 Dy 4 ·5H 2 O lacks any in-phase or out-of-phase ac peak under a zero or a 1000 Oe dc field (Figure S12). Attempts to dilute the sample with yttrium were unsuccessful and, accordingly, no further magnetic data could be obtained for this complex. The absence of slow magnetic relaxation for Zn 2 Dy 4 ·5H 2 O clearly contrasts with the magnetic behavior observed for many non-linear Dy 4 complexes with oxygen bridges, which usually behave as SMMs. 31-35 This fact is especially remarkable, especially, if we take into account that it has been demonstrated, both theoretically and experimentally, that the introduction of Zn II typically increases the anisotropic energy barrier of lanthanoid complexes. 36-38 The absence of magnetic relaxation for this complex maybe due to a possible 24 exchange coupling between Dy III ions. Unfortunately, we have not been able to obtain a similar Gd 3+ derivative, which could bring some light about this issue. In a final attempt to contribute to the comprehension of the absence of SMM behavior, the anisotropy axes of these dysprosium ions were calculated with MAGELLAN 29 (Figure 9). Despite some parallelism and stacking of the three hydrazone ligands, the divergent directions of the four magnetic axes could also contribute to explain the non-existence of slow magnetic relaxation. Figure 9. Calculated orientations using MAGELLAN software 29 of the local main magnetic axes (blue) on Dy III centers for Zn 2 Dy 4 CONCLUSIONS Three new dinuclear complexes (Tb 2 Cl, Dy 2 Cl and Dy 2 CF 3 SO 3 ·2THF), and a heterohexanuclear compound (Zn 2 Dy 4 ·5H 2 O), all of them derived from the polydentate H 4 daps ligand, have been obtained, and structurally and magnetically characterized. The crystallographic structures were solved of all these complexes, showing a versatile 25 polynucleating behavior of the hydrazone ligand. Thus, in the three dinuclear compounds, the doubly deprotonated ligand displays a novel hexadentate μ 2 -ț 5 :ț 1 fashion. Likewise, three unprecedented bridging coordination modes (hexadentate μ 4 -ț 5 :ț 2 :ț 1 :ț 1 , heptadentate μ 3 -ț 5 :ț 4 :ț 1 and heptadentate μ 4 -ț 5 :ț 4 :ț 1 :ț 1 ) are displayed by two tetradeprotonated and a triply deprotonated ligand units present in Zn 2 Dy 4 . Hence, this work contributes four hitherto unknown coordination modes for this hydrazone donor. All of them are related to the scarcely reported lanthanoid coordination chemistry of H 4 daps, a very well-known chelating ligand for transition metal or main group ions. This work demonstrates an unknown versatility and bridging character of this classic ligand, which appears favored by lanthanoid ions. Furthermore, the comparison of the complexes described herein with the three previously reported mononuclear lanthanoid compounds containing H 2 daps 2- , appears to indicate that both the solvent and the salt employed play a crucial role in the structure of the final complexes. We hope that these contributions allow increasing the use and knowledge of this type of classic hydrazone ligands, so commonly used with transition metal or main group ions, with new f-block ions, as the size of its internal compartment seems suitable for all of them, and it allows varied and close connections between neighboring metal ions of varied nature. The magnetic study of all the complexes shows that none of the pure samples behaves as SMM, and that the quantum tunnel cannot be removed by dilution of none of the dinuclear complexes, except for Dy 2 Cl, as the diluted Dy 2 Cl@Y behaves as a weak field induced SMM. The absence of slow relaxation of the magnetization for Zn 2 Dy 4 is unexpected in view of previous results, and it could be due to the possible existence of weak magnetic exchange among the oxygen-bridged dysprosium ions, and to the non-parallel alignment of the magnetic anisotropy axes. 32 (28) Colacio, E.; Ruiz, J.; Mota, A. J.; Palacios, M. A.; Cremades, E.; Ruiz, E.; White, F. J.; Brechin, E. K. Family of carboxylate- and nitrate-diphenoxo triply bridged dinuclear NiII-LnIII complexes (Ln = Eu, Gd, Tb, Ho, Er, Y): Synthesis, experimental and theoretical magneto-structural studies, and single-molecule magnet behavior. Inorg. Chem. 2012, 51, 5857-5868. (29) Chilton, N. F.; Collison, D.; McInnes, E. J. L.; Winpenny, R. E. P.; Soncini, A. An electrostatic model for the determination of magnetic anisotropy in dysprosium complexes. Nat. Commun. , 2013, 4 , 2551 (30) Lin, W-Q.; Liao, X-F.; Jia, J-H.; Leng, J-D.; Liu, J-L.; Guo, F-S.; Tong, M-L. Lanthanide oxide clusters: from tetrahedral [Dy 4 (μ 4 -O)] 10+ to supertetrahedral [Ln 20 ((μ 4 -O) 11 ] 38+ (Ln=Tb, Dy, Ho, Er). Chem.-Eur. J . 2013 , 19 , 12254–12258 (31) Xue, S.; Guo, Y-N.; Zhao, L.; Zhang, P.; Tang, J. Unique Y-shaped lanthanide aggregates and single molecule magnet behaviour for the Dy 4 analogue. Dalton Trans. 2014, 43, 1564-1570. (32) Luan, F.; Liu, T.; Yan, P.; Zou, X.; Li, Y.; Li, G. Single-molecule magnet of a tetranuclear dysprosium complex disturbed by a salen-type ligand and chloride counterions. Inorg. Chem. 2015, 54, 3485Ѹ3490 (33) Gusev, A.; Herchel, R.; Nemec, I.; Shul’gin, V.; Eremenko, I. L.; Lyssenko, K.; Linert, W.; Trávnícek, Z. Tetranuclear lanthanide complexes containing a hydrazone-type ligand. Dysprosium [2 × 2] gridlike single-molecule magnet and toroic. Inorg. Chem. 2016, 55, 12470Ѹ12476. (34) Das, C.; Vaidya, S.; Gupta, T.; Frost, J. M.; Righi, M.; Brechin, E. K.; Affronte, M.; Rajaraman, G.; Shanmugam, M. Single-molecule magnetism, enhanced magnetocaloric effect, and toroidal magnetic moments in a family of Ln4 squares. Chem. Eur. J. 2015, 21, 15639–15650. 33 (35) Gao, H-L.; Zhou, X-P.; Bi, Y-X.; Shen, H-Y.; Wang, W-M.; Wang, N-N.; Chang,Y-X.; Zhang, R-X.; Cui, J-Z.; A Dy4 single-molecule magnet and its Gd(III), Tb(III), Ho(III), and Er(III) analogues encapsulated by an 8-hydroxyquinoline Schiff base derivative and ȕ-diketonate coligand. Dalton Trans. 2017, 46, 4669–4677. (36) Das, S.; Bejoymohandas, K. S.; Dey, A.; Biswas, S.; Reddy, M. L. P.; Morales, R.; Ruiz, E.; Titos-Padilla, S.; Colacio, E.; Chandrasekhar, V. Amending the anisotropy barrier and luminescence behavior of heterometallic trinuclear linear [M II Ln III M II ] (Ln III =Gd, Tb, Dy; M II =Mg/Zn) complexes by change from divalent paramagnetic to diamagnetic metal ions. Chem. Eur. J. 2015, 21, 6449–6464. (37) (a) Upadhyay, A.; Singh, S. K.; Das, C.; Mondol, R.; Langley, S. K.; Murray, K. S.; Rajaraman, G.; Shanmugam, M. Enhancing the effective energy barrier of a Dy(III) SMM using a bridged diamagnetic Zn(II) ion. Chem. Commun. 2014, 50, 8838–8841. (b) Upadhyay, A.; Das, C.; Vaidya, S.; Singh, S. K.; Gupta, T.; Mondol, R.; Langley, S. K.; Murray, K. S.; Rajaraman, G.; Shanmugam, M. Role of the diamagnetic zinc(II) ion in determining the electronic structure of lanthanide single-ion magnets. Chem. Eur. J. 2017, 23, 4903-4916. (38) Amjad, A.; Madalan, A. M.; Andruh, M.; Caneschi, A.; Sorace, L. Slow relaxation of magnetization in an isostructural series of zinc–lanthanide complexes: an integrated EPR and AC susceptibility study. Chem. Eur. J. 2016, 22, 12849–12858. 34 Table of Contents Graphic and Synopsis The coordination chemistry of a well-known polydentate ligand towards lanthanoids shows a high versatility. In fact, four hitherto unknown coordination modes are presented. This rich structural chemistry contrasts with a surprisingly poor magnetic response, with only a diluted sample behaving as a field-induced SMM. View publication stats