English

• 0.   Introduction

  Single - molecule magnets (SMMs), which exhibits slow relaxation of magnetization at the molecular level, are expected to be the smallest storage units and quantum objects, invoking intense research interest in the field of spintronic devices and quantum computing^[1-3]. Recently, optimizing the crystal field environment of a single Dy(Ⅲ) center based on an electrostatic model allows blocking of magnetization reaching unprecedented liquid nitrogen temperatures^[4]. Whereas, encapsulating two or more spin centers complexes facilitate the generation of more sophisticated architectures and magnetic behavior, giving the new opportunities to design SMMs with augmented performances^[5-7]. The exploitation of new Dy(Ⅲ) polymetallic complexes to drive understanding of influence factors that dictate the slow magnetic relaxation is still being established^[8-12].

  For polymetallic Dy(Ⅲ) - SMMs, the ligand field effect, the geometry, the local symmetry around the Dy(Ⅲ) ions, and the existence of magnetic interactions between the Dy(Ⅲ) sites will result in the significant divergence of magnetic relaxation behavior^[13-16]. Optimizing the electronic structure of molecules and unraveling internal magneto-structural correlations are essential driving forces behind the rapid development of novel nanomagnets. To achieve the objective, the design of polydentate ligands assembling various types of Dy(Ⅲ) complexes and ligand field has become the primary consideration. Carboxylates exhibit excellent performance in the construction of complexes because of their demonstrated coordination motif versatility^[17-19]. And the diversity and variability of substitution groups of carboxylic ligands are conducive to the generation of appropriate steric hindrance fitting well with the coordination geometries of Dy(Ⅲ) ions^[20-21]. Moreover, the origin of the concept of the hard - soft acid - base (HSAB) theory, lanthanide ion prefers to bond with oxygen atoms, and resulting adduct tend to have a more ionic character in its bonding^[22-23]. The carboxylic ligands match the nature of Dy(Ⅲ) ions and induce the orientation of electron density to produce magnetic anisotropy. Also, the incorporation of electron - donating groups (e. g., amino, aryl, alkyl, alkoxyl) on carboxylic ligands can pose positive effects on the electron density of the donor oxygen atoms and intensify coordination field strength-enhancing the magnetic properties of this family. However, carboxylate - based polymetallic Dy(Ⅲ) - SMMs with intriguing magnetic behaviors are still scarce, which is probably attributed to the complexity of carboxylates coordination, resulting in unclear magnetic anisotropy orientation and a significant contribution of the quantum tunneling of magnetization (QTM) to the relaxation dynamics^[24-26].

  Based on the above considerations, to further efficiently control the carboxylic coordination, we selected various carboxylic acid ligands with different steric hindrance and electron donor ability substituent and combined them with polydentate Schiff - base ligand with relatively fixed coordination cavities to construct polymetallic complexes (Scheme 1). Herein, three novel Dy(Ⅲ) - based complexes, [Dy₂(Hhms)₂(C(CH₃)₃ COO)₂(H₂O)₄] (NO₃)₂ (1), [Dy₂(Hhms)₂(C₁₄H₉COO)₂ (C₂H₅OH)₂(CH₃OH)₂] [ZnCl₄] (2), and [Dy₂(Hhms)₂(C₆H₃ (NH₂)₂COO)₂Cl₂] ·2CH₃CN (3) (H₂hms=2 - hydroxy - 3 - methoxybenzylidene)-semicarbazide) were isolated in a centrosymmetric binuclear architecture. The structural characterization and magnetic properties of complexes 1-3 were investigated. Through magneto -structural correlations analysis, the present work highlights the importance of ligand substitution and molecule symmetry in controlling magnetic relaxation and provides a feasible way to develop high-performance carboxylatebased polymetallic SMMs.

  Scheme 1

  

  Scheme 1.  Structures of carboxylic acid ligands and Schiff-base ligand

  下载: 全尺寸图片 幻灯片

  1.   Experimental

  1.1   Materials and measurements

  All chemical reagents and solvents were obtained commercially and used in the reactions without further treatment, except for the ligand of H₂hms, which was synthesized by an analogous method described in the literature^[21]. Powder X -ray diffraction measurements were carried out on a Bruker D8 ADVANCE X-ray powder diffractometer at 40 kV and 40 mA, using Cu Kα radiation (λ =0.154 18 nm) ranging from 5° to 50° (2θ) to verify the phase purity of the polycrystalline samples. The FTIR spectra were recorded with pressed KBr pellets in a range of 4 000-400 cm^-1 on a Tensor 27 spectrometer (Bruker Optics, Ettlingen, Germany). Elemental analyses (C, H, N) were collected on an Elementar Vario EL Ⅲ analyzer. Magnetic susceptibility measurements were conducted on a Quantum Design MPMS-XL7 SQUID magnetometer in a temperature range of 2-300 K with 1 000 Oe applied field. Alternating-current (ac) measurements were performed on the same magnetometer at ac frequencies ranging from 1 to 1 000 Hz using a 2.0 Oe oscillating ac field. The measured susceptibilities were corrected for the diamagnetism of the constituent atoms (Pascal's tables).

  1.2   Synthesis of [Dy₂(Hhms)₂(C(CH₃)₃COO)₂ (H₂O)₄](NO₃)₂ (1)

  A mixture of Dy(NO₃)₃·6H₂O (0.045 6 g, 0.1 mmol), C(CH₃)₃COOH (0.010 2 g, 0.1 mmol) and H₂hms (0.021 0 g, 0.1 mmol) in methanol (4 mL) and dichloromethane (4 mL) was added sodium azide (0.006 5 g, 0.1 mmol). After stirring for 3 h, the resultant solution was filtered and allowed to stand undisturbed at room temperature. The yellowish block crystals suitable for single - crystal X - ray diffraction were obtained by slow evaporation of the filtrate after 5 d. Yield: ca. 56% (32 mg) based on Dy(Ⅲ). Elemental Anal. Calcd. for C₂₈H₄₆Dy₂N₈O₂₀(%): H, 4.07; C, 29.51; N, 9.83. Found(%): H, 4.43; C, 29.67; N, 9.52. IR (KBr, cm^-1): 3 340 (s), 3 043 (s), 2 584 (w), 2 420 (w), 1 667 (s), 1 568 (s), 1 451 (s), 1 415 (s), 1 385 (m), 1 366 (m), 1 357 (s), 1 225 (m), 1 089 (m), 1 018 (s), 956 (m), 787 (m), 741 (m), 669 (m), 614 (m), 493 (w).

  1.3   Synthesis of [Dy₂(Hhms)₂(C₁₄H₉COO)₂ (C₂H₅OH)₂(CH₃OH)₂][ZnCl₄] (2)

  A mixture of DyCl₃·6H₂O (0.037 7 g, 0.1 mmol), ZnCl₂ (0.027 3 g, 0.2 mmol), C₁₄H₉COOH (0.133 2 g, 0.6 mmol) and, H₂hms (0.021 0 g, 0.1 mmol) in methanol (4 mL) and ethanol (4 mL) was added sodium azide (0.013 0 g, 0.2 mmol). After stirring for 3 h, the resultant solution was filtered and allowed to stand undisturbed at room temperature. The yellowish block crystals suitable for single - crystal X - ray diffraction were obtained by slow evaporation of the filtrate after one week. Yield: ca. 45% (35 mg) based on Dy(Ⅲ). Elemental Anal. Calcd. for C₅₄H₅₈Cl₄Dy₂N₆O₁₄Zn(%): H, 3.78; C, 41.97; N, 5.44. Found(%): H, 3.52; C, 41.69; N, 5.41. IR (KBr, cm^-1): 3 345 (s), 3 294 (s), 3 248 (s), 3 092 (m), 2 932 (m), 2 856 (m), 1 689 (s), 1 602 (m), 1 586 (s), 1 482 (s), 1 454 (s), 1 383 (s), 1 110 (s), 1 093 (m), 854 (w), 823 (w), 784 (w), 744 (m), 669 (m), 615 (m), 494 (w).

  1.4   Synthesis of [Dy₂(Hhms)₂(C₆H₃(NH₂)₂COO)₂ Cl₂]·2CH₃CN (3)

  A mixture of H₂hms (0.021 0 g, 0.1 mmol), C₆H₃ (NH₂)₂COOH (0.015 2 g, 0.1 mmol), and DyCl₃·6H₂O (0.037 7 g, 0.1 mmol) in acetonitrile (4 mL) was placed in a 20 mL Teflon-line autoclave kept for three days at 90 ℃, and then cooled down to room temperature at a rate of 5 ℃ ·h^-1 to form yellowish block crystals suitable for X -ray analysis. Yield: ca. 58% (35 mg) based on Dy(Ⅲ). Elemental Anal. Calcd. for C₃₆H₄₀Cl₂Dy₂N₁₂O₁₀ (%): H, 3.37; C, 36.13; N, 14.05. Found(%): H, 3.55; C, 36.49; N, 14.02. IR (KBr, cm^-1): 3 433 (s), 3 369 (s), 3 293 (s), 3 076 (m), 2 954 (s), 2 867 (w), 2 244 (w), 1 687 (s), 1 607 (s), 1 562 (s), 1 446 (s), 1 416 (s), 1 385 (s), 1 272 (m), 1 113 (s), 1 091 (s), 976 (m), 857 (w), 824 (w), 767 (w), 732 (m), 667 (m), 614 (m), 494 (w).

  1.5   X-ray singlecrystal diffraction analysis

  Single-crystal X-ray data for complexes 1-3 were collected at room temperature on a Bruker Apex Ⅱ CCD diffractometer with graphite monochromated Mo Kα radiation (λ =0.071 073 nm). Cell determination and data reduction were processed with the SAINT processing program. The absorption correction based on multiscan was applied in SADABS. By using Olex2^[27], the structures were solved by direct methods with SHELXT and refined by full-matrix least-squares techniques against F² using SHELXL - 2014 programs^[28]. All non - hydrogen atoms were refined anisotropically. The acidic hydrogen atoms were found from the electron density map and were refined freely. All other hydrogen atoms were placed in calculated geometrically positions and refined using the riding model. The crystallographic data and structure refinement summary are summarized in Table S1 (Supporting information). Selected bond distances and angles are listed in Table S2.

  CCDC: 2092500, 1; 2092502, 2; 2092503, 3.

  2.   Results and discussion

  2.1   Description of crystal structures

  Complexes 1-3 were structurally characterized by X-ray crystallography (Table S1). 1 and 2 crystallize in the monoclinic space group C2/c, while 3 crystallizes in space group P2₁/n. 1-3 are all centrosymmetric binuclear structures and the asymmetric unit of the complex contains one - half of the entire molecule. The Dy(Ⅲ) ions in 1 and 2 have similar monocapped squareantiprism geometries with an {O₈N} coordination sphere (Fig. S1 and Table S3). Two antiparallel Hhms^- ligands bind to the Dy(Ⅲ) ion by two phenoxy oxygen atoms (O2, O2ⁱ), one methoxy group (O1), one aldehyde group (O3), and one Schiff - base nitrogen atom (N1), and the remaining coordination sites are filled by two carboxylic oxygen atoms (O4, O5) and two H₂ O molecules for 1 or a superposition of CH₃OH and C₂H₅OH for 2 (O6, O7) (Fig. 1). In 1, the C(CH₃)₃COO^- groups act as chelating ligands, with it coordinating to a single Dy(Ⅲ) ion, while in 2, the C₁₄H₉COO^- groups bridge the two Dy(Ⅲ) centers in a μ₂∶η¹∶η¹ fashion. It is important to note that the different coordination patterns of carboxylate in 1 and 2 do not affect the configurational differences between them. For complex 3, the Dy(Ⅲ) centers are coordinated by Hhms^- ligands and carboxylic oxygen atoms in a manner essentially the same as that in 2 except for two alcohol oxygen atoms substituted by one Cl^- ion. Therefore, the Dy(Ⅲ) ion in 3 has a {DyO₆NCl} coordination sphere, forming a squareantiprism geometry. The two Dy(Ⅲ) centers are bridged by two phenoxy oxygen atoms (O2 and O2ⁱ) of two Hhms^- ligands, besides, the RCOO^- groups with distinct terminal substituent bridge two Dy(Ⅲ) atoms in a μ₂∶η¹∶η¹ fashion in 2 and 3, with the Dy…Dy intramolecular distance of 0.378 1(1), 0.365 6(8) and 0.364 8(7) nm respectively, as well as Dy—O—Dy angle of 107.19°, 101.81°, and 101.23° respectively. The Dy—O bond lengths are in a range of 0.230 5(3)- 0.263 7(2) nm, being the longest related to the methoxy oxygen atom distances. The shortest Dy—O bond lengths in 1-3 involve aldehyde oxygen (for 1) or one of bridged phenoxide oxygen (for 2 and 3) with 0.230 5(3), 0.232 7(2), and 0.231 5(5) nm respectively. The Dy—O distances within O, N, O-tridentate pocket of Hhms^- ligand are shorter than that within O, O-bidentate pocket of Hhms^- ligand. The Dy—C_arboxylate bond lengths of the bidentate bridging coordination mode in 2 (AVG: 0.237 1 nm) and 3 (AVG: 0.233 2 nm) show shorter distance than that of the chelating coordination mode in 1 (AVG: 0.244 2 nm), which may be associated with the different terminal substituents of carboxylic acid ligands. The Dy—N (0.247 8(6) - 0.256 7(3) nm) and Dy—Cl (0.278 3(1) nm) bonds present quite long distances. Furthermore, the shortest inter - dimer Dy…Dy distance is equal to 0.721 9, 1.018 9, and 0.835 6 nm respectively.

  Figure 1

  

  Figure 1.  Molecular structures of complexes 1 (a), 2 (b), and 3 (c)

  Hydrogen atoms and solvent molecules are omitted for clarity; Symmetry codes: ⁱ 2-x, 1-y, 1-z for 1; ⁱ 1-x, 1-y, 1-z for 2; ⁱ 1-x, 1-y, 1-z for 3

  下载: 全尺寸图片 幻灯片

  2.2   Magnetic properties

  The magnetic properties of 1-3 were measured using a SQUID magnetometer. The sample purity was first checked by powder X - ray diffraction experiment comparing experimental diagram to theoretical one obtained from the crystal structure (Fig.S2).

  The variable - temperature magnetic susceptibility for 1-3 was collected between 2 and 300 K under an applied magnetic field of H_dc being 1 kOe. At 300 K, χ_MT products were 28.07, 28.02, 28.16 cm³·K·mol^-1 for 1-3, respectively (Fig. 2), which are slightly lower than the value expected for two ensembles of Dy(Ⅲ) ions in the free - ion approximation. On cooling, the χ_MT products underwent a slow decline and then decreased rapidly at low temperatures to reach a value of 21.00, 20.17, and 16.53 cm³·K·mol^-1 respectively at 2 K. The reduction of χ_MT for all three complexes is accounted for by the gradual depopulation of the Stark sublevels and/or intramolecular antiferromagnetic interactions. The magnetization data of 1-3 at 2 K are depicted in the inset of Fig. 2. The plots of M vs H showed a rapid increase below 1 kOe, and the slight increase reaching the maximum magnetizations of 11.9Nβ, 11.0Nβ, 11.4Nβ at 7 kOe, which were much lower than the theoretical saturation value of 20Nβ for the Dy₂ dimer. The unsaturated magnetization indicates strong magnetic anisotropy of Dy(Ⅲ) ions^[29-30].

  Figure 2

  

  Figure 2.  Temperature dependence of χ_MT for complexes 1-3 under 1 000 Oe field

  Inset: variable-field magnetization data for 1-3 at 2 K where lines are guides to the eyes

  下载: 全尺寸图片 幻灯片

  The magnetization relaxation dynamics were investigated by studying the temperature (T) and frequency (ν) dependence of ac magnetic susceptibilities. Complex 2 displayed no out-of-phase signal in ac susceptibility measurements in zero-field (Fig.S3). As a result of a strong QTM of Dy(Ⅲ) ions, the temperature-dependent out - of - phase signals of 1 only showed a small tail below 4 K (without peak maximum) under zero dc field (Fig. 3).

  Figure 3

  

  Figure 3.  Temperature-dependent outof-phase χ″ ac susceptibility signals (a) and frequencydependent out-of-phase χ″ ac susceptibility signals (b) for 1 under zero dc field

  下载: 全尺寸图片 幻灯片

  For complex 3, the outofphase (χ″) component of the temperature - dependent ac susceptibility exhibited one maximum in the 3.5 K (100 Hz) - 8 K (1 000 Hz) range, which indicates the slow relaxation of magnetization of 3 (Fig. 4). But a tail of the peak was observed below 3 K, and this observation suggests a temperature-independent QTM regime at lower temperatures, as is often observed for Ln(Ⅲ) - containing systems. Consistently, the frequency dependence of the maximum in association with single relaxation mode appeared on the plot of the χ″ vs the frequency between 1 and 1 000 Hz. The ColeCole plot of χ″ vs χ' clearly identified single relaxation process and were fitted using a generalized Debye model^[31] with CCFIT package^[32] to extract the relaxation times (τ) (Fig.S4). The fitting gave a relatively narrow distribution of τ with a small coefficient α from 0.19 to 0.34. Insight into the dynamic magnetization under zero dc field of 3 was obtained through plots of ln τ vs T^-1 (Fig. 5). An Arrhenius fitting of the hightemperature region gave unphysical parameters of U_eff= 68 K and τ₀=6.30×10^-8 s. In the whole temperature range, a combination of multiple relaxation pathways including Orbach, Raman, and QTM processes can be considered here. The data for 3 were fitted using Eq.1: τ^-1= τ_QTM^-1+CTn+ τ₀^-1exp[-U_eff/(k_BT)], where U_eff is the Orbach parameters, C and n are the Raman parameters, and τ_QTM^-1 is the QTM rate. The best fit parameters were found to be U_eff=96 K, τ₀=3.1 ×10^-10 s, C=0.122 s^-1·K-4.77, n=4.77 and τ_QTM=5.7 ms. To get a better understanding of the magnetic relaxation, we also tried to fit the relaxation time of zero-field data with only QTM and Raman processes, however, no better results were obtained by leaving out the Orbach term (Fig.S5).

  Figure 4

  

  Figure 4.  Temperature-dependent in-phase χ' (a) and out-of-phase χ″ (b) ac susceptibility signals for 3 under 0 Oe dc field; Frequency-dependent in-phase χ' (c) and out-of-phase χ″ (d) ac susceptibility signals for 3 under 0 Oe dc field

  Symmetry codes: ⁱ 2-x, 1-y, 1-z for 1; ⁱ 1-x, 1-y, 1-z for 2; ⁱ 1-x, 1-y, 1-z for 3

  下载: 全尺寸图片 幻灯片

  Figure 5

  

  Figure 5.  Relaxation data plotted as ln τ vs T^-1 for 3

  Solid lines represent the best fit

  下载: 全尺寸图片 幻灯片

  The ac magnetic susceptibility was further investigated under an applied static field. Applying a dc field of 1 000 Oe induces slower relaxation with shifting peak maxima toward lower frequency (Fig.S6), indicating a significant suppression of QTM. Cole - Cole plots in a range of 4 - 9 K were obtained (Fig. S7) and fitted using the generalized Debye model^[31] to give the relaxation time τ (Fig. 5) and distribution coefficient α (0.15- 0.27). The thermal energy barrier was estimated from the high-temperature data of ln τ vs T^-1 plot to be 83 K with τ₀ =2.1×10^-8 s (Fig. 5). An additional term of Raman processes was required to fit the relaxation time behavior in the whole temperature range. The obtained satisfied parameters were U_eff=98 K, τ₀=7.4×10^-9 s, C= 0.133 s^-1 ·K^-4.32, n=4.32. Furthermore, if we only considered the Raman process for τ without the Orbach process, the fitting line did not fit well the ln τ vs T^-1 plots (Fig. S5), also suggesting the existence of a thermally activated relaxation in 3.

  From the ac susceptibilities, we can see that complex 3 has the highest energy barrier. The structures of 2 and 3 are analogous, with the major difference being in R substituent of carboxylate ligand and the coordinated small molecules with four superposed CH₃OH/C₂H₅OH in 2 and two Cl^- in 3. It should be noted that there is no change in the coordination modes of Hhms^- and RCOO^- between them. In addition, complexes 1 and 2 have a similar coordination configuration, with the main distinction only in the bonding mode of carboxylate which leads to the relative differences in the spatial arrangement of ligands. However, their magnetic properties are quite different: 1 exhibits a frequencydependent behavior under zero dc field though without a clear peak, while 2 does not show magnetic relaxation behavior.

  To further better understand the SMM properties of 1-3, Fig. 6 depicts the simulated principal magnetic axes of Dy(Ⅲ) ions in each molecule defined by the Magellan program^[33]. The directions of the calculated anisotropy axis of Dy(Ⅲ) in 1 are found to be tilted towards the O4 atom belonging to the bidentate—C(CH₃)₃COO^- anion and bridged phenoxy oxygen atoms (O2). Both of them make a small angle of 17.4° and 34° respectively. While the anisotropy axes of complex 2 show parallel arrangement through the middle of each C₁₄H₉COO^- ligands, almost perpendicular to the bridged phenoxy oxygen atoms (O2 and O2ⁱ). For dinuclear complex 3, the anisotropy of the Dy(Ⅲ) ion is nearly oriented along with the Dy—O_carboxylate (O4) and Dy—Cl bonds, which deviation angles are in a range of 17.5° - 18.9°. These illustrate the fact that the bonding mode of carboxylate and the coordination of chloride anions affect the magnetic anisotropy of the Dy(Ⅲ) ion and the SMM properties.

  Figure 6

  

  Figure 6.  Magnetic anisotropy axes (green lines) predicted by Magellan program for Dy(Ⅲ) centers in complexes 1-3

  下载: 全尺寸图片 幻灯片

  From the analysis of the anisotropic axis above, the most magnetic axes in the ground states are closer along with the carboxylic oxygen moieties. It is well - known that the magnetic anisotropy of an Ln(Ⅲ) ion is extremely sensitive to its coordination environment^[34-39]. Even changes in the second coordination sphere ligated atoms could also lead to a significant variation on the single - ion anisotropy of the Ln(Ⅲ) ions, as demonstrated by Sessoli and co - workers^[40-42]. In 3, the two strongly electron - donating groups —NH₂ through inductive effects increase the charge of axially shortest carboxylic oxygen donor thus strengthening the axial component of crystal field relative to 1 and 2. Careful inspection of the bond distance of complexes 1-3 reveals that the different electron-donating effects induced by the R substituent are necessarily felt by the carboxylic oxygen ligands, and consequently has an effect on the strength of the chemical bond to the Dy(Ⅲ) ions. This is indeed suggested by the shortest Dy—O_carboxylate bond for 3 (Dy—O4 0.232 6(4) nm, Dy—O5 0.233 7(5) nm). Whilst simultaneously, coordination of a chloride ion, in place of two solvent molecules (H₂O or CH₃OH/ EtOH), reduces the coordination number, resulting in the Dy(Ⅲ) ions in square antiprismatic environments with D_4d symmetry, which are much different from the monocapped square - antiprism in 1 and 2. Moreover, among complexes 1-3, the shortest intramolecular Dy… Dy distances (0.364 8(7) nm) and the smallest Dy—O—Dy angles (101.23°) are found for 3, which may have a positive influence on the magnetic exchange coupling between two Dy(Ⅲ) ions. In contrast to 3, complex 2 did not show magnetic relaxation behavior. It might be associated with the relatively high charge distribution of the O_phenoxo atoms (O2 and O2ⁱ) lying on the transverse plane (perpendicular to the magnetic anisotropy easy axis) which contributes to transverse ligand field, and thus is disadvantageous to obtain the high magnetic anisotropy of oblate Dy(Ⅲ) ion. In total, the different magnetic properties of 3 and 1-2 mainly result from the differences of ligand substitution by the non-coordinating electron-donating group, coordination geometry, and magnetic interaction between two Dy(Ⅲ) ions. The results emphasize that the utility of the ligand substitution can give rise to good modulation in molecule symmetry, as well as the magnetic anisotropy of Ln(Ⅲ) ion, thus improving SMM performance.

  3.   Conclusions

  In summary, a series of dinuclear Dy(Ⅲ) complexes based on different substituted carboxylate ligand and Schiff-base H₂hms ligands have been reported. In complexes 1 and 2, donor atom sets around the Dy(Ⅲ) ions are the nearly same as a {DyO₈N} coordination sphere, whereas, in 3, the Dy(Ⅲ) ions exhibit a {DyO₆ NCl} coordination environment. Complex 3 exhibited zero - field SMM behavior with a high thermal energy barrier for the reversal of the magnetization (96 K), revealing a significant improvement in performance across the series. The large U_eff value observed for 3 is a consequence of the enhanced magnetic anisotropy and local symmetry promoted by the presence of the significant electron - donating groups —NH₂ on the carboxylic oxygen atoms and the coordinated chloride ligands reducing the coordination number of the Dy(Ⅲ) ions, resulting in a higher symmetry environment (D_4d symmetry) of the Ln(Ⅲ) ions. The results demonstrate that the non-coordinating ligand substitution and coordination small molecule play essential roles in charge distributions of the ligand-field environments and molecular symmetry thus an SMM performance, which should be considered seriously and utilized efficiently during the rational design of new SMMs.

  
  Supporting information is available at http://www.wjhxxb.cn
  
• 1. [1]

     Woodruff D N, Winpenny R E, Layfield R A. Lanthanide Single-Molecule Magnets[J]. Chem. Rev., 2013, 113(7):  5110-5148. doi: 10.1021/cr400018q

  2. [2]

     Bogani L, Wernsdorfer W. Molecular Spintronics Using Single-Molecule Magnets[J]. Nat. Mater., 2008, 7(3):  179-186. doi: 10.1038/nmat2133

  3. [3]

     Candini A, Klyatskaya S, Ruben M, Wernsdorfer W, Affronte M. Graphene Spintronic Devices with Molecular Nanomagnets[J]. Nano Lett., 2011, 11(7):  2634-2639. doi: 10.1021/nl2006142

  4. [4]

     Guo F S, Day B M, Chen Y C, Tong M L, Mansikkamäki A, Layfield R A. Magnetic Hysteresis up to 80 Kelvin in a Dysprosium Metallocene Single-Molecule Magnet[J]. Science, 2018, 362(6421):  1400-1403. doi: 10.1126/science.aav0652

  5. [5]

     Blagg R J, Ungur L, Tuna F, Speak J, Comar P, Collison D, Wernsdorfer W, McInnes E J, Chibotaru L F, Winpenny R E. Magnetic Relaxation Pathways in Lanthanide Single-Molecule Magnets[J]. Nat. Chem., 2013, 5(8):  673-678. doi: 10.1038/nchem.1707

  6. [6]

     Guo Y N, Xu G F, Wernsdorfer W, Ungur L, Guo Y, Tang J K, Zhang H J, Chibotaru L F, Powell A K. Strong Axiality And Ising Exchange Interaction Suppress Zero - Field Tunneling of Magnetization of an Asymmetric Dy₂ Single - Molecule Magnet[J]. J. Am. Chem. Soc., 2011, 133(31):  11948-11951. doi: 10.1021/ja205035g

  7. [7]

     Habib F, Lin P H, Long J, Korobkov I, Wernsdorfer W, Murugesu M. The Use of Magnetic Dilution to Elucidate the Slow Magnetic Relaxation Effects of a Dy₂ Single - Molecule Magnet[J]. J. Am. Chem. Soc., 2011, 133(23):  8830-8833. doi: 10.1021/ja2017009

  8. [8]

     Demir S, Gonzalez M I, Darago L E, Evans W J, Long J R. Giant Coercivity and High Magnetic Blocking Temperatures for N₂^3- Radical-Bridged Dilanthanide Complexes upon Ligand Dissociation[J]. Nat. Commun., 2017, 8(1):  2144. doi: 10.1038/s41467-017-01553-w

  9. [9]

     Pineda E M, Chilton N F, Marx R, Dorfel M, Sells D O, Neugebauer P, Jiang S D. Direct Measurement of Dysprosium (Ⅲ)…Dysprosium(Ⅲ) Interactions in a Single-Molecule Magnet[J]. Nat. Commun., 2014, 5(1):  5243. doi: 10.1038/ncomms6243

  10. [10]

      Fernandez Garcia G, Guettas D, Montigaud V, Larini P, Sessoli R, Totti F, Cador O, Pilet G, Le Guennic B. A Dy₄ Cubane: A New Member in the Single -Molecule Toroics Family[J]. Angew. Chem. Int. Ed., 2018, 57(52):  17089-17093. doi: 10.1002/anie.201810156

  11. [11]

      Morita T, Damjanovic M, Katoh K, Kitagawa Y, Yasuda N, Lan Y. Comparison of the Magnetic Anisotropy and Spin Relaxation Phenomenon of Dinuclear Terbium(Ⅲ) Phthalocyaninato Single-Molecule Magnets Using the Geometric Spin Arrangement[J]. J. Am. Chem. Soc., 2018, 140(8):  2995-3007. doi: 10.1021/jacs.7b12667

  12. [12]

      Wu H P, Li M, Xia Z Q, Montigaud V, Cador O, Le Guennic B, Ke H S, Wang W Y, Xie G, Chen S P, Gao S L. High Temperature Quantum Tunnelling of Magnetization and Thousand Kelvin Anisotropy Barrier in a Dy₂ Single-Molecule Magnet[J]. Chem. Commun., 2021, 57(3):  371-374. doi: 10.1039/D0CC06993K

  13. [13]

      Zhu Z H, Guo M, Li X L, Tang J K. Recent Advance on Single Molecule Magnets[J]. Sci. China Chem., 2018, 48(8):  790-803.

  14. [14]

      Kong M, Feng X, Wang J, Zhang Y Q, Song Y. Tuning Magnetic Anisotropy via Terminal Ligands along the Dy…Dy Orientation in Novel Centrosymmetric [Dy₂] Single Molecule Magnets[J]. Dalton Trans., 2021, 50(2):  568-577. doi: 10.1039/D0DT03854G

  15. [15]

      Jin C Y, Li X L, Liu Z L, Mansikkamäki A, Tang J K. An Investigation into the Magnetic Interactions in a Series of Dy₂ Single-Molecule Magnets[J]. Dalton Trans., 2020, 49(30):  10477-10485. doi: 10.1039/D0DT01926G

  16. [16]

      Liu J L, Chen Y C, Tong M L. Symmetry Strategies for High Performance Lanthanide - Based Single - Molecule Magnets[J]. Chem. Soc. Rev., 2018, 47(7):  2431-2453. doi: 10.1039/C7CS00266A

  17. [17]

      Su J, Yuan S, Li J, Wang H Y, Ge J Y, Drake H F, Leong C F, Yu F, D'Alessandro D M, Kurmoo M, Zuo J L, Zhou H C. Rare-Earth Metal Tetrathiafulvalene Carboxylate Frameworks as Redox - Switchable Single-Molecule Magnets[J]. Chem. Eur. J., 2021, 27(2):  622-627. doi: 10.1002/chem.202004883

  18. [18]

      Zhang C C, Ma X F, Cen P P, Jin X Y, Yang J H, Zhang Y Q, Ferrando-Soria J, Pardo E, Liu X Y. A Series of Lanthanide(Ⅲ) Metal-Organic Frameworks Derived From a Pyridyl-Dicarboxylate Ligand: Single-Molecule Magnet Behaviour and Luminescence Properties[J]. Dalton Trans., 2020, 49(40):  14123-14132. doi: 10.1039/D0DT02736G

  19. [19]

      Biswas S, Mandal L, Shen Y B, Yamashita M. Explorationof SMM Behavior of Ln₂ Complexes Derived from Thianaphthene-2-carboxylic Acid[J]. Dalton Trans., 2019, 48(37):  14096-1410. doi: 10.1039/C9DT01984G

  20. [20]

      Li Y, Shang Q, Zhang Y Q, Yang E C, Zhao X J. Fine Tuning of the Anisotropy Barrier by Ligand Substitution Observed in Linear {Dy₂Ni₂} Clusters[J]. Chem. Eur. J., 2016, 22(52):  18840-18849. doi: 10.1002/chem.201603800

  21. [21]

      Li M, Wu H P, Wei Q, Ke H S, Yin B, Zhang S, Lv X Q, Xie G, Chen S P. Two {Zn₂^ⅡDy^Ⅲ} Complexes Supported by Monophenoxido/Dicarboxylate Bridges with Multiple Relaxation Processes: Carboxylato Ancillary Ligand-Controlled Magnetic Anisotropy in Square Antiprismatic Dy^Ⅲ Species[J]. Dalton Trans., 2018, 47(28):  9482-9491. doi: 10.1039/C8DT01842A

  22. [22]

      Davies K, Bourne S A, Oliver C L. Solvent - and Vapor - Mediated Solid - State Transformations in 1, 3, 5 - Benzenetricarboxylate Metal-Organic Frameworks[J]. Cryst. Growth Des., 2012, 12(4):  1999-2003. doi: 10.1021/cg201707e

  23. [23]

      Cai S L, Zheng S R, Wen Z Z, Fan J, Zhang W G. Construction of Luminescent Three-Dimensional Ln(Ⅲ)-Zn(Ⅱ) Heterometallic Coordination Polymers Based on 2-Pyridyl Imidazole Dicarboxylate[J]. CrystEngComm, 2012, 14(23):  8236-8243. doi: 10.1039/c2ce25803j

  24. [24]

      Peng Y, Mereacre V, Anson C, Powell A. Tuning of Hula-Hoop Coordination Geometry in a Dy Dimer[J]. Inorganics, 2016, 4(1):  2. doi: 10.3390/inorganics4010002

  25. [25]

      Yang J W, Yang Z F, Chen P, Tian Y M, Sun W B. A Dy₂ Dimer Embedded in One Salen-Type Ligand with Different Local Symmetries Behaves as Zero-Field Single-Molecule Magnet[J]. Z. Anorg. Allg. Chem., 2018, 644(8/9):  443-448.

  26. [26]

      Wu H P, Li M, Zhang S, Ke H S, Zhang Y Q, Zhuang G L, Wang W Y, Wei Q, Xie G, Chen S P. Magnetic Interaction Affecting the Zero-Field Single-Molecule Magnet Behaviors in Isomorphic {Ni₂^ⅡDy₂^Ⅲ} and {Co₂^ⅡDy₂^Ⅲ} Tetranuclear Complexes[J]. Inorg. Chem., 2017, 56(18):  11387-11397. doi: 10.1021/acs.inorgchem.7b01840

  27. [27]

      Dolomanov O V, Bourhis L J, Gildea R J, Howard J A K, Puschmann H. OLEX2: A Complete Structure Solution, Refinement and Analysis Program[J]. J. Appl. Crystallogr., 2009, 42(2):  339-341. doi: 10.1107/S0021889808042726

  28. [28]

      Sheldrick G M A. A Short History of SHELX[J]. Acta Crystallogr. Sect. A, 2008, 64(1):  112-122. doi: 10.1107/S0108767307043930

  29. [29]

      Osa S, Kido T, Matsumoto N, Re N, Pochaba A, Mrozinski J A. A Tetranuclear 3d - 4f Single Molecule Magnet: [Cu^ⅡLTb^Ⅲ (hfac)₂]₂[J]. J. Am. Chem. Soc., 2004, 126(2):  420-421. doi: 10.1021/ja037365e

  30. [30]

      Tang J K, Hewitt I, Madhu N T, Chastanet G, Wernsdorfer W, Anson C E, Benelli C, Sessoli R, Powell A K. Dysprosium Triangles Showing Single - Molecule Magnet Behavior of Thermally Excited Spin States[J]. Angew. Chem. Int. Ed., 2006, 45(11):  1729-1733. doi: 10.1002/anie.200503564

  31. [31]

      Guo Y N, Xu G F, Guo Y, Tang J K. Relaxation Dynamics of Dysprosium(Ⅲ) Single Molecule Magnets[J]. Dalton Trans., 2011, 40(39):  9953-9963. doi: 10.1039/c1dt10474h

  32. [32]

      Reta D, Chilton N F. Uncertainty Estimates For Magnetic Relaxation Times and Magnetic Relaxation Parameters[J]. Phys. Chem. Chem. Phys., 2019, 21(42):  23567-23575. doi: 10.1039/C9CP04301B

  33. [33]

      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[J]. Nat. Commun., 2013, 4(1):  2551. doi: 10.1038/ncomms3551

  34. [34]

      Liu S S, Xu L, Jiang S D, Zhang Y Q, Meng Y S, Wang Z, Wang B W, Zhang W X, Xi Z, Gao S. Half-Sandwich Complexes of Dy^Ⅲ: A Janus - Motif with Facile Tunability of Magnetism[J]. Inorg. Chem., 2015, 54(11):  5162-5168. doi: 10.1021/ic502734z

  35. [35]

      Habib F, Brunet G, Vieru V, Korobkov I, Chibotaru L F, Murugesu M. Significant Enhancement of Energy Barriers in Dinuclear Dysprosium Single - Molecule Magnets through Electron-Withdrawing Effects[J]. J. Am. Chem. Soc., 2013, 135(36):  13242-13245. doi: 10.1021/ja404846s

  36. [36]

      Chen Y X, Ma F, Chen X, Zhang Y H, Wang H L, Wang K, Qi D D, Sun H L, Jiang J Z. Bis(1, 8, 15, 22-tetrakis(3-pentyloxy)phthalocyaninato)terbium Double - Decker Single - Ion Magnets[J]. Inorg. Chem., 2019, 58(4):  2422-2429. doi: 10.1021/acs.inorgchem.8b02949

  37. [37]

      Zhang W Y, Zhang Y Q, Jiang S D, Sun W B, Li H F, Wang B W, Chen P, Yan P F, Gao S. Dramatic Impact of the Lattice Solvent on the Dynamic Magnetic Relaxation of Dinuclear Dysprosium Single-Molecule Magnets[J]. Inorg. Chem. Front., 2018, 5(7):  1575-1586. doi: 10.1039/C8QI00266E

  38. [38]

      Li X L, Li H, Chen D M, Wang C, Wu J F, Tang J K, Shi W, Cheng P. Planar Dy³⁺ Dy₃ Clusters: Design, Structure and Axial Ligand Perturbed Magnetic Dynamics[J]. Dalton Trans., 2015, 44(47):  20316-20320. doi: 10.1039/C5DT03931B

  39. [39]

      Wu H P, Li M, Yin B, Xia Z Q, Ke H S, Wei Q, Xie G, Chen S P, Gao S L. Fine-Tuning the Type of Equatorial Donor Atom in Pentagonal Bipyramidal Dy(Ⅲ) Complexes to Enhance Single-Molecule Magnet Properties[J]. Dalton Trans., 2019, 48(43):  16384-16394. doi: 10.1039/C9DT03292D

  40. [40]

      Wang Y L, Han C B, Zhang Y Q, Liu Q Y, Liu C M, Yin S G. Giant Field Dependence of the Low Temperature Relaxation of the Magnetization in a Dysprosium(Ⅲ)-DOTA Complex[J]. Inorg. Chem., 2016, 55(11):  5578-5584. doi: 10.1021/acs.inorgchem.6b00653

  41. [41]

      Car P E, Perfetti M, Mannini M, Favre A, Caneschi A, Sessoli R. Magnetic Anisotropy in a Dysprosium/DOTA Single-Molecule Magnet: Beyond Simple Magneto-Structural Correlations[J]. Chem. Commun., 2011, 47(13):  3751-3753. doi: 10.1039/c0cc05850e

  42. [42]

      Cucinotta G, Perfetti M, Luzon J, Etienne M, Car P E, Caneschi A, Calvez G, Bernot K, Sessoli R. Magnetic Anisotropy and Spin-Parity Effect along the Series of Lanthanide Complexes with DOTA[J]. Angew. Chem. Int. Ed., 2012, 51(7):  1606-1610. doi: 10.1002/anie.201107453

• 

  Scheme 1  Structures of carboxylic acid ligands and Schiff-base ligand

  下载: 全尺寸图片 幻灯片
  

  Figure 1  Molecular structures of complexes 1 (a), 2 (b), and 3 (c)

  Hydrogen atoms and solvent molecules are omitted for clarity; Symmetry codes: ⁱ 2-x, 1-y, 1-z for 1; ⁱ 1-x, 1-y, 1-z for 2; ⁱ 1-x, 1-y, 1-z for 3

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Temperature dependence of χ_MT for complexes 1-3 under 1 000 Oe field

  Inset: variable-field magnetization data for 1-3 at 2 K where lines are guides to the eyes

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Temperature-dependent outof-phase χ″ ac susceptibility signals (a) and frequencydependent out-of-phase χ″ ac susceptibility signals (b) for 1 under zero dc field

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Temperature-dependent in-phase χ' (a) and out-of-phase χ″ (b) ac susceptibility signals for 3 under 0 Oe dc field; Frequency-dependent in-phase χ' (c) and out-of-phase χ″ (d) ac susceptibility signals for 3 under 0 Oe dc field

  Symmetry codes: ⁱ 2-x, 1-y, 1-z for 1; ⁱ 1-x, 1-y, 1-z for 2; ⁱ 1-x, 1-y, 1-z for 3

  下载: 全尺寸图片 幻灯片
  

  Figure 5  Relaxation data plotted as ln τ vs T^-1 for 3

  Solid lines represent the best fit

  下载: 全尺寸图片 幻灯片
  

  Figure 6  Magnetic anisotropy axes (green lines) predicted by Magellan program for Dy(Ⅲ) centers in complexes 1-3

  下载: 全尺寸图片 幻灯片
•