Citation

Peng Song, Ouyang Xiaoming, Wang Yihan, Teng Qinghu, Li Yan, Zhang Xiuqing, Hu Zhaobo, Wang Kai, Liang Fupei. In situ synthesized homochiral dysprosium-oxo clusters with threonine Schiff bases[J]. Chin. Chem. Lett., 2023, 34(8): 108044  

Song Peng^a,1, Xiaoming Ouyang^a,1, Yihan Wang^a,1, Qinghu Teng^a,*, Yan Li^a, Xiuqing Zhang^a, Zhaobo Hu^b, Kai Wang^a,*, Fupei Liang^a,*     

Polymetallic coordination clusters have attracted great interest over the past few years due to their applications in various fields [1–3]. One of the most intriguing species among their family is those featuring chirality. Apart from possessing general properties of achiral systems, chiral clusters can also display other interesting functions brought by chiral topologies, such as ferroelectricity [4,5], multiferroic [6], magnetic circular dichroism [7], circularly polarized luminescence (CPL) [8,9], chiral separation and catalysis [10,11]. Among them, chiral high-nuclearity 4f clusters are paid much more attentions probably due to their impressive topologies and fantastic properties resulted from the integration of functional homochiral ligands with the metal centers featuring 4f electron shell. For examples, the reported triple helix {La₈} could fix CO₂ efficiently [12], circular {Eu₈} helicates exhibit remarkable visible CPL activity with a high luminescence dissymmetry factor |g_lum| of 1.25 [13], and {La₁₂} shows an unprecedented homochiral octahedral "cage in cage" topology and remarkable enantioselective separation abilities [14]. In comparison with those based on 3d and 3d-4f metal centers, however, the reported chiral 4f clusters with the nuclearity higher than 10 are still limited at present (Table S2 in Supporting information) [14–18]. Thus, the exploration of novel chiral high-nuclearity 4f clusters is of great importance, whether from the view of their potential application or from the present scarcity.

As a kind of versatile molecules, Schiff base compounds have shown broad applications in different fields such as biology and medicine [19,20]. It is also noteworthy that they have demonstrated extraordinary capabilities in the construction of high-nuclearity metal clusters. In recent years, a series of novel Schiff base ligands have been designed and synthesized by our group and have been put into the exploration of high-nuclearity 4f clusters. N,N′-Bis(o-vanillidene)pyridine-2,6-dicarbohydrazide N-oxide is just one of them (H₄ovpho, Scheme 1, top). Its structural specificity is reflected in the O_N-Oxide, which situates at centered parent and forms closely situated chelating pockets with two Schiff base arms synergistically. Its reactions with high equivalent lanthanide salts gave a series of novel high-nuclearity 4f clusters successfully [21,22]. Inspired by these works, we believed that the ligand featuring such special chelating structure would also be competent to assemble chiral high-nuclearity 4f clusters if it is endowed with the chirality. A novel Schiff base ligand of H₅btp was thus designed (Scheme 1, top). It is composed of a phenol parent and two homochiral threonine Schiff base arms, displaying a similar chelating structure as that of H₄ovpho ligand. In light of the effectiveness of free threonine in controlling the hydrolysis of 4f ions, which would favor the formation of high nuclearity lanthanide-oxo products [15,23,24], the designed H₅btp ligand was introduced through in situ synthetic approach starting from threonine and 2,6-diformyl-4-methylphenol.

Scheme 1

	Download: JPG larger image
Scheme 1. The designed and achieved ligands in this work.

As expected, two pairs of pure enantiomers of high-nuclearity 4f clusters with novel topologies, four-blade propeller shaped {Dy₁₈} (1R and 1S) and sandglass-like {Dy₉} (2R and 2S), were achieved successfully. They are all homochiral Dy-oxo clusters based on two kinds of in situ formed threonine Schiff base ligands. Strikingly, 1R and 1S are the second largest species in the chiral 4f cluster family with their nuclearity being next only to the reported {Er₆₀} [15].

All clusters were synthesized from the reactions of Dy^Ⅲ salts, 2,6-diformyl-4-methylphenol and threonine in a molar ratio of 8:1:2 in the presence of Et₃N. Their single crystals were obtained through the slow evaporation of organic-aqueous mixed solvents at room temperature (ESI). It revealed from all synthetic exploration that the in situ reaction condition is crucial for successful cultivation of single crystal samples. Rich oxo-bridges were generated in all clusters. It means that the hydrolysis of the Dy^Ⅲ ions was controlled successfully in reaction process. The designed H₅btp ligand was formed in situ in all clusters as we expected. Unexpectedly, the H₃ftp ligand featuring single Schiff base arm was also formed in all clusters (Scheme 1, bottom), although the feeding molar ratio of 2,6-diformyl-4-methylphenol and threonine was controlled at 1:2 strictly. IR spectra analysis further confirmed the formation of these Schiff base ligands [25,26]. Charge balance calculation suggested that two kinds of ligands were deprotonated incompletely with the coordination forms of H₂btp³⁻ and Hftp²⁻, respectively.

Clusters 1R and 1S are enantiomer with each other, so as 2R and 2S. Thus, only the structures of 1R and 2R are descripted in detail here. Cluster 1R crystallizes in orthorhombic chiral space group of I222. It is made up of eighteen Dy^Ⅲ ions, four ^DHftp²⁻ and four ^DH₂btp³⁻ ligands, eight μ-OH⁻ ions, twenty μ₃-OH⁻ ions, a μ₆-O²⁻ ion, four NO₃⁻ ions, eight H₂O molecules, and several free solvent molecules. In the center of its molecule, six Dy^Ⅲ ions are linked together by a μ₆-O²⁻ and eight μ₃-OH⁻ ions, forming an octahedral [Dy₆(μ₃-OH)₈(μ₆-O)]⁸⁺ unit (Fig. 1a). Four triangular [Dy₃(μ₃-OH)]⁸⁺ units are located around this octahedral unit. Each of them is connected to the Dy4 ion locating at the equatorial plane of the octahedron unit through the bridging of two μ-OH⁻ and two μ₃-OH⁻ ions. An unprecedent four-blade propeller shaped [Dy₁₈(μ-OH)₈(μ₃-OH)₂₀(μ₆-O)]²⁴⁺ core with a D₂-symmetry was thus assembled.

Figure 1

	Download: JPG larger image
Fig. 1. (a) [Dy18(μ-OH)8(μ3-OH)20(μ6-O)]24+ core. (b) Coordination mode of DHftp2−. (c) Coordination mode of DH2btp3−. (d) Ligation of the ligands. (e) Whole cluster structure of 1R. Symmetric codes: a, 1-x, 1-y, z; b, x, 1-y, 1-z; c, 1-x, y, 1-z.

There are one ^DHftp²⁻ and one ^DH₂btp³⁻ in each triangular [Dy₃(μ₃-OH)]⁸⁺ unit. The former bridges Dy1 and Dy2 with its O_phenol, displaying a μ-ƞ²: ƞ⁴ coordination mode (Fig. 1b). The latter shows a mode of μ₃-ƞ¹: ƞ⁴: ƞ⁴ (Fig. 1c). It links the Dy1 and Dy3 within the same [[Dy₃(μ₃-OH)]⁸⁺ unit, and another Dy3 from the adjacent triangular [Dy₃(μ₃-OH)]⁸⁺ unit (Fig. 1c). The [Dy₁₈(μ₂-OH)₈(μ₃-OH)₂₀(μ₆-O)]²⁴⁺ core is thus protected by the surrounding of four ^DHftp²⁻ and four ^DH₂btp³⁻ ligands (Fig. 1d). The charge of whole cluster is further balanced by four NO₃⁻ ions, each of which coordinates to a Dy2 ion with a monodentate mode (Fig. 1e). In addition, there are twelve H₂O molecules coordinating to Dy2 and Dy5 ions, respectively, which further complete the ligation of whole cluster (Fig. 1e). Continuous symmetry measure (CShM) method was employed to evaluate exact coordination geometries of the Dy^Ⅲ ion using SHAPE 2.1 software [27,28]. According to the CShM calculation (Tables S3-S5 in Supporting information), the Dy^Ⅲ ions in 1R exhibit three coordination geometries (Fig. S4 in Supporting information), namely nine-coordinated spherical capped square antiprism for Dy1 and Dy2, eight-coordinated triangular dodecahedron for Dy3, and nine-coordinated capped square antiprism for Dy4 and Dy5.

Although a considerable number of high-nuclearity 4f clusters have sprung up in recently years, such as {Gd₄₈}, {Er₄₈}, {Gd₆₀}, {Dy₇₂}, {Dy₇₆}, {Ln₁₀₄}, {Ln₁₄₀} [29–35], reported chiral high-nuclearity 4f clusters were still sparse. A literature survey indicated that the existing chiral lanthanide clusters whose nuclearity are over 10 are only {La₁₂}, {Ln₁₄} and {Er₆₀} (Table S2) [14–18], with the {Er₆₀} being the biggest one up to now. It follows that 1R and 1S could be the second largest members in chiral 4f cluster family so far.

Cluster 2R crystallizes in monoclinic P2₁ space group. It consists of nine Dy^Ⅲ ions, two ^DHftp²⁻ and two ^DH₂btp³⁻ ligands, six OAc⁻ ions, ten μ₃-OH⁻ ions, six H₂O molecules, as well as an OAc⁻ counterion and several free solvent molecules. As shown in Fig. 2a, its sandglass-like [Dy₉(μ₃-OH)₁₀(OAc)₂]¹⁵⁺ skeleton contains two {Dy₅} square pyramid units sharing a vertex ion (Dy5). Each triangular face of square pyramid unit is capped by a μ₃-OH⁻group, by which eight Dy^Ⅲ ions are linked to the Dy5 vertex. For four Dy^Ⅲ ions on the base of each pyramid unit, three of them (Dy1, Dy2 and Dy3 or Dy6, Dy7 and Dy8) are linked together by a μ₃-OH⁻ ion, and the remaining one (Dy4 or Dy9) is bridged to one of its neighboring Dy^Ⅲ ions (Dy1 or Dy6) by a μ-ƞ¹: ƞ² OAc⁻ ion. This constitutes the most remarkable feature that distinguish 2R from other sandglass-like {Dy₉} clusters [36–40]. In the previous reported sandglass-like {Dy₉} clusters, four Dy^Ⅲ ions on pyramid base were usually linked together by μ₄-OH⁻ and μ₅-O²⁻ ion (Table S5). Such unique bridging fashion in 2R also resulted in considerable discrepancy for the distances between the adjacent Dy^Ⅲ ions of pyramid base (Fig. 2a). However, the four Dy^Ⅲ ions belonging to each pyramid base are still coplanar, with the RMSD value of 0.002 and 0.068, respectively.

Figure 2

	Download: JPG larger image
Fig. 2. (a) [Dy9(μ3-OH)10(OAc)2]15+ skeleton. (b) Coordination mode of DHftp2−. (c) Coordination mode of DH2btp3−. (d) Ligation of the ligands. (e) [Dy9(DHftp)2(DH2btp)2(OAc)6(μ3-OH)10(H2O)6]+ core of 2R.

The base of each square pyramid unit is further sheathed by a ^DHftp²⁻ and a ^DH₂btp³⁻ ligands (Fig. 2d). The ^DHftp^2- ligand adopts a μ-ƞ²: ƞ⁴ mode to link Dy1 with Dy2 (or Dy6 with Dy7) using its O_phenol (Fig. 2b). While the ^DH₂btp³⁻ ligand affords its O_phenol and one of O_carboxylate to bridge Dy2, Dy3 and Dy4 (or Dy7, Dy8 and Dy9), displaying a μ₃-ƞ¹: ƞ⁴: ƞ⁴ mode (Fig. 2c). Furthermore, there are six H₂O molecules and four OAc⁻ ions acting as the terminal ligands (Fig. 2e). They are one H₂O and one monodentate-coordinated OAc⁻ ion on Dy3, two H₂O and one bidentate-chelated OAc⁻ ion on Dy4, one bidentate-chelated OAc⁻ ion on Dy8, and three H₂O and one monodentate-coordinated OAc⁻ ion on Dy9, respectively. All these ligands construct the cation core of [Dy₉(^DHftp)₂(^DH₂btp)₂(OAc)₆(μ₃-OH)₁₀(H₂O)₆]⁺ of 2R on the basis of [Dy₉(μ₃-OH)₁₀(OAc)₂]¹⁵⁺ skeleton. As suggested by calculated CShM values (Tables S6-S8 in Supporting information), the Dy^Ⅲ ions in 2R present three types of coordination geometries (Fig. S5 in Supporting information). The Dy1, Dy3 and Dy6 ions have spherical capped square antiprism geometries, while Dy2, Dy7 and Dy8 ions adopt spherical tricapped trigonal prism one, all of which are nine-coordinated. The Dy4, Dy5 and Dy9 ions were eight-coordinated, exhibiting square antiprism configurations.

Clusters 1R and 1S or 2R and 2S show strict mirror symmetry judged whether from the Dy-oxo skeletons and the binding modes of ligands, or from the whole molecular structures (Figs. S6 and S7 in Supporting information). Solid-state circular dichroism (CD) tests on their single crystal sample further support these results. As shown in Fig. 3, the CD spectra show obvious mirror symmetry effect whether 1R versus 1S or 2R versus 2S. Taking 2R and 2S as the examples, cluster 2R exhibits two peaks with negative Cotton effect near 270 and 412 nm, and two peaks with positive Cotton effect near 318 and 366 nm. Correspondingly, four peaks can be observed in CD spectra of 2S, which appear as the opposite Cotton effect in relative to those of 2R. All these results are indicative of the homochirality of each cluster, and confirm that 1R versus 1S and 2R versus 2S are pairs of enantiomers, respectively.

Figure 3

	Download: JPG larger image
Fig. 3. Solid state CD spectra of all clusters.

The temperature dependence of magnetic susceptibilities was measured in the range of 300–2 K with an applied dc field of 1 kOe (Fig. S8 in Supporting information). At room temperature, the χ_MT values of 1R, 1S, 2R and 2S are 254.81, 254.12, 128.25 and 128.09 cm³ K/mol, respectively, being close to their corresponding theoretical values for eighteen or nine isolated Dy^Ⅲ ions (J = 5/2, g = 4/3). Upon cooling, the χ_MT-T curves all show gradually decreasing tendency, finally reaching the minimums of 153.36, 164.54, 73.74 and 74.66 cm³ K/mol at 2 K, respectively. This could be ascribed to the contributions of thermal depopulation effect of Stark sublevels of Dy^Ⅲ ions and also possible antiferromagnetic couplings between the Dy^Ⅲ ions within cluster [41]. Furthermore, the χ_MT-T curve of each cluster was almost coincided with that of its enantiomer. It's no surprise that the chiral isomerization did not induce substantial change of magnetic structures. At different temperatures, the field dependence of the magnetization (M) of all clusters were collected in the field (H) range of 0–70 kOe (Fig. S9 in Supporting information). The M-H curves of all clusters revealed a slow rise of M along with the increasing of H. The maximum of M values appeared at 70 kOe and 2 K, but didn't reach saturation. And the M-H curves recorded at different temperature are not overlapped. These behaviors suggest the presence of low-lying excited states and/or magnetic anisotropy of the Dy^Ⅲ ions in all clusters [42].

In conclusion, we have succeeded in the synthesis of two pairs of chiral high nuclearity 4f clusters based on threonine Schiff base ligands. The hydrolysis of Dy^Ⅲ ions were successfully controlled in introducing the chiral source through in situ synthesis method. This contributes to the formation of rich oxo-bridges, which construct unprecedent four-blade propeller shaped {Dy₁₈} and new sandglass-like {Dy₉} Dy-oxo skeletons together with chiral threonine Schiff bases ligands. Magnetic investigation revealed possible antiferromagnetic interactions between the Dy^Ⅲ centers of each cluster. This work thus affords novel and rare members of chiral high nucleartiy 4f clusters, as well as an efficient synthesis strategy towards these clusters.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

This work was financially supported by the National Natural Science Foundation of China (Nos. 21961008, 22271068, 22075058 and 22261012), Guangxi Science and Technology Base and Talents Program (No. AD21220105), and Guangxi Natural Science Foundation (No. 2022GXNSFBA035472).

Supplementary material associated with this article can be found, in the online version, at https://doi.org/10.1016/j.cclet.2022.108044.