Three sandwich-type zinc(II)–lanthanide(III) clusters: structures, luminescence and magnetic properties

Abstract

Three new sandwich-type hexanuclear Zn^II–Ln^III clusters based on a methoxy substituted salicylamide Salen-like ligand and o-vanilline, [Zn₂Ln₄(HL)₄(o-vanilline)₂(OH)₄(CH₃OH)₂]·2NO₃·5CH₃OH (Ln = Eu, Tb and Dy, H₃L = 1-(2-hydroxy-4-methoxy-benzamido)-2-(2-hydroxy-3-methoxy-benzylideneamino)-ethane) were synthesised. Single-crystal structure analyses reveal that these Zn^II–Ln^III clusters display a hydroxyl bridged tetranuclear Ln^III₄ core sandwiched in an environment of two Zn^II ions, four doubly deprotonated HL²⁻ and two singly deprotonated o-vanilline. Photophysical determinations indicate that the Zn^II–Eu^III cluster shows mixed emissions both originating from ligands and Eu^III ions, and the other two only display ligand-centered emission in visible region. The magnetic studies reveal that the Zn^II–Dy^III cluster displays rare ferromagnetic behavior with slow magnetic relaxation behavior at zero field.

---

Introduction

High-nuclearity heterometallic clusters containing Zn^II–Ln^III metal ions have blossomed into a promising topic of modern coordination chemistry and attracted increasing attention primarily because of their interesting magnetic and photophysical properties since the first Zn^II–Eu^III/Sm^III heterometallic clusters were reported in 1995 by Brennan et al.¹ This is attributed to: (a) magnetic behaviour that can easily be explained due to the diamagnetic character of the Zn^II ion, and (b) Ln^III ions are desired luminescent and magnetic species because of their characteristic 4f electronic configuration. The magnetic properties of Zn^II–Ln^III heterometallic compounds mainly arise from the intrinsic large spin ground state and the strong uniaxial magnetic anisotropy of some Ln^III ions. It is expected that the incorporation of Zn^II and Ln^III ions into one coordination system may not only enhance the structural diversity, but also integrate the optical and magnetic properties of Ln^III ions which can offer multifunctional properties.^2–7 To our best knowledge, high-nuclearity Zn^II–Ln^III coordination clusters are still scarce because their design and controllable synthesis represent a formidable challenge. Consequently, only a few nanosized Zn^II–Ln^III clusters of varying nuclearities range from kite-like Ln₃Zn,² square-shaped Ln₂Zn₂,³ cyclic Zn₂Ln₂,⁴ cubane-like Ln₂Zn₂,⁵ drum-like Ln₄Zn₈ (ref. 6) and triangular ring-like Ln₂₄Zn₆ (ref. 7) have been reported. Due to the strong easy-axis type magnetic anisotropy and larger J(L + S) value originating from lanthanide ions,⁸ the magnetic exploration for Ln^III-based compounds, especially for single-molecule magnets (SMMs), have stimulated a wide range of research interests. Among which, Dy^III ions are the most appealing mainly because they have an inherent strong spin–orbital coupling effect and very large magnetic anisotropy of the ⁶H_15/2 ground state. It is well known that the coordination environment, the local ligand field and the magnetic interaction strength between metal ions can have impact on SMM behavior of Dy^III containing systems. Undoubtedly, further exploration of Dy^III containing SMMs deserves attention not only for better SMMs but also for deciphering the relaxation dynamics and mechanisms. Although several Cu^II–Dy^III clusters displaying ferromagnetic behavior with slow magnetization relaxation systems have been reported,⁹ it is rare that ferromagnetic interaction with slow magnetic relaxation exists in Zn^II–Dy^III heterometallic clusters.

Recently, we reported a nitrate-bridged Zn^II₆Dy^III₆ dodecanuclear heterometallic cluster of Salen-like salicylamide ligand, 1-(2-hydroxy-benzamido)-2-(2-hydroxy-benzylideneamino)-ethane, which represents an unusual example of Zn^II–Dy^III cluster that exhibits ferromagnetic SMM behavior.¹⁰ However, the blocking temperature is lower than the temperature limit of the magnetometer even an external field of 2 kOe was applied. For the sake of increasing energy barrier and therefore to improve the SMM properties of Zn^II–Dy^III clusters of this kind of ligands, a new methoxy substituted Salen-like salicylamide ligand, 1-(2-hydroxy-4-methoxy-benzamido)-2-(2-hydroxy-3-methoxy-benzylideneamino)-ethane (H₃L, Scheme 1) was synthesised. The basis of the strategy presented herein, relies on the use of a new multidendate Salen-like salicylamide ligand as well as o-vanilline, which are elaborately selected to compartmentalize Ln^III ions through oxygen based coordination pockets free of nitrate coordination. As a result, we succeeded in obtaining three heterometallic hexanuclear clusters, [Zn₂Ln₄(HL)₄(o-vanilline)₂(OH)₄(CH₃OH)₂]·2NO₃·5CH₃OH (Ln = Eu, Tb and Dy). The photophysical properties of compounds Zn^II₂Eu^III₄, Zn^II₂Tb^III₄ and Zn^II₂Dy^III₄ as well as the magnetic properties of Zn^II₂Dy^III₄ have been explored and the results indicate Zn^II₂Eu^III₄ shows mixed emissions both originating from ligands and Eu^III ions, and the other two only display ligand-centered emission in visible region. Interestingly, the corresponding dysprosium containing complex Zn^II₂Dy^III₄ shows typical ferromagnetic single molecule magnetic behavior with slow zero-field relaxation. To our best knowledge, it is rather infrequent to observe ferromagnetic coupling and zero-field slow magnetic relaxation coexisted in Zn^II–Ln^III heterometallic clusters.

Scheme 1 The synthesis route of the ligand H₃L.

Experimental section

Materials

Methyl 2-hydroxy-4-methoxybenzoate, ethane-1,2-diamine and 3-methoxy-4-hydroxybenzaldehyde were obtained from Aladain. Solvents and all other chemicals were analytical grade, available commercially, and used as received.

Synthesis of the ligand

1-(2-Hydroxy-4-methoxy-benzamido)-2-(2-hydroxy-3-methoxy-benzylideneamino)-ethane (H₃L) is shown in Scheme 1. N-(2-Aminoethyl)-2-hydroxy-4-methoxybenzamide was prepared according to literature using methyl 2-hydroxy-4-methoxybenzoate instead.¹¹ 10 mmol (1.52 g) o-vanilline was added to a 30 mL ethanol solution of N-(2-aminoethyl)-2-hydroxy-4-methoxybenzamide (2.10 g, 10 mmol) under stirring. The mixture was refluxed for 4 h to make a clear solution. Then the solution was rotary-evaporated to near dryness to obtain a pale-yellow solid. The yellow crystalline product obtained by ethanol recrystallization was washed with ethanol and dried in air. H₃L: 2.68 g, yield 78.0%. Mp 166–167 °C. Analytical data, calc. for C₁₈H₂₀N₂O₅: C, 62.78; H, 5.85; N, 8.13; found: C, 62.81, H, 5.85, N, 8.17; IR (KBr, ν, cm⁻¹): 3393 (w), 2938 (w), 1649 (s), 1598 (s), 1544 (m), 1467 (m), 1389 (m), 1251 (s), 1204 (m), 1160 (m), 1076 (m), 1022 (m), 958 (m), 831 (m), 736 (m). ¹H NMR (CDCl₃, 400 MHz) δ: 3.77 (m, 3H, CH₃), 3.88 (m, 3H, CH₃), 3.80 (m, 2H, CH₂), 3.85 (m, 2H, CH₂), 6.71 (s, 1H, ArH), 6.80 (m, 1H, ArH), 6.88 (m, 1H, ArH), 6.97 (m, 1H, ArH), 7.23 (m, 1H, ArH), 7.35 (m, 3H, CH=N), 8.41 (t, 1H, NH, J = 4 Hz), 12.21 (s, 2H, OH).

Preparation of complexes

General procedure. 27 μL (0.2 mmol) triethylamine was added to a 15 mL acetonitrile solution containing 0.1 mmol (0.028 g) H₃L and 0.1 mmol (0.015 g) o-vanilline. Then 0.1 mmol (0.021 g) Zn(NO₃)₂·2H₂O was added and the solution was stirred for 4 h to obtain a suspension. 0.1 mmol Ln(NO₃)₃·6H₂O in 5 mL of methanol was added to this epinephelus solution and after another 4 h a clear solution was obtained. The mixture was stirred overnight and filtered into a sealed glass vial for crystallization at room temperature. After about three weeks pale yellow single crystals suitable for crystal analysis were obtained which were collected by filtration, washed with cold methanol, and dried in the air.

[Zn₂Eu₄(HL)₄(o-vanilline)₂(OH)₄(CH₃OH)₂]·2NO₃·5CH₃OH (Zn^II₂Eu^III₄). The empirical formula and the molecular weight is C₉₅H₁₁₈Eu₄N₁₀O₄₃Zn₂ and 2826.64 respectively. Yield: 36.7 mg, 52% based on Eu(NO₃)₃·6H₂O. Analytical data (%), calcd: C, 40.37; H, 4.21; N, 4.96; found: C, 40.22; H, 4.20; N, 4.98; IR (KBr, ν, cm⁻¹): 3456 (w), 1618 (s), 1541 (s), 1470 (s), 1446 (s), 1383 (m), 1217 (s), 1168 (s), 1130 (m), 1074 (m), 734 (m), 594 (m).

[Zn₂Tb₄(HL)₄(o-vanilline)₂(OH)₄(CH₃OH)₂]·2NO₃·5CH₃OH (Zn^II₂Tb^III₄). The empirical formula and the molecular weight is C₉₅H₁₁₈Tb₄N₁₀O₄₃Zn₂ and 2854.48 respectively. Yield: 34.2 mg, 48% based on Tb(NO₃)₃·6H₂O. Analytical data (%), calcd: C, 39.97; H, 4.17; N, 4.91; found: C, 40.16; H, 4.19; N, 4.94; IR (KBr, ν, cm⁻¹): 3435 (w), 1618 (s), 1543 (s), 1448 (s), 1445 (s), 1385 (m), 1217 (s), 1168 (s), 1130 (m), 1074 (m), 737 (m), 592 (m).

[Zn₂Dy₄(HL)₄(o-vanilline)₂(OH)₄(CH₃OH)₂]·2NO₃·5CH₃OH (Zn^II₂Dy^III₂). The empirical formula and the molecular weight is C₉₅H₁₁₈Dy₄N₁₀O₄₃Zn₂ and 2868.68 respectively. Yield: 30.1 mg, 42% based on Dy(NO₃)₃·6H₂O. Analytical data (%), calcd: C, 39.77; H, 4.15; N, 4.88; found: C, 39.94; H, 4.13; N, 4.90. IR (KBr, ν, cm⁻¹): 3440 (w), 1618 (s), 1544 (s), 1470 (m), 1446 (m), 1386 (m), 1211 (m), 1164 (m), 1130 (m), 1093 (m), 740 (m), 594 (m).

Physical measurements

Elemental analyses were performed on a Perkin-Elmer 2400 Series II CHNO elemental analyzer. IR spectra were recorded in the range 400–4000 cm⁻¹ on a Perkin-Elmer FTIR spectrometer using KBr pellets. X-ray powder diffraction (XRPD) patterns of the samples were recorded on a X-ray diffractometer (Rigaku D/Max 2200PC) with a graphite monochromator and Cu Kα radiation (λ = 1.5418 Å) at room temperature with a scan speed of 0.2 s per step and a step size of 0.02 (2θ), while the voltage and electric current were held at 40 kV and 20 mA. Thermogravimetric analysis experiments were performed using a TGA/NETZSCH STA449C instrument heated from 25–800 °C (heating rate of 10 °C min⁻¹, nitrogen stream). Emission and excitation spectra were recorded with a Hitachi F-7000 spectrophotometer equipped with quartz cuvettes of 2.5 cm path length. The luminescence decays were recorded using a pumped dye laser (Lambda Physics model FL2002) as the excitation source. The magnetic susceptibility measurements were obtained on a Quantum Design SQUID magnetometer MPMS-XL. Measurements were performed on a polycrystalline sample of 30.09 mg for Zn^II₂Dy^III₄. Direct-current magnetic susceptibility measurements were carried out at 2–300 K for dc applied fields at 1000 Oe. Field dependence of the magnetization magnetic susceptibility measurements on the polycrystalline samples were performed with the same magnetometer. Alternating-current susceptibility measurements were carried out utilizing an oscillating ac field of 3.0 Oe and frequencies ranging from 20 to 1600 Hz under 0 and 2000 Oe dc field respectively. All crystalline samples for photoluminescence and magnetic studies were obtained from the same batch and characterized by PXRD, TG, EA and IR spectra.

X-ray crystallography

Suitable pale yellow block crystals of Zn^II₂Eu^III₄ (0.24 × 0.18 × 0.14 mm³), Zn^II₂Tb^III₄ (0.32 × 0.26 × 0.14 mm³) and Zn^II₂Dy^III₄ (0.33 × 0.31 × 0.23 mm³) were coated with perfluoropolyether oil before mounting. Intensity data of all the complexes were recorded at 293(2) K employing a Bruker SMART APEX II CCD diffractometer equipped with a monochromatized Mo Kα radiation (λ = 0.71073 Å) source. No crystal decay was observed during the data collections. In all cases, absorption corrections based on multiscans using the SADABS software¹² were applied. The structures were solved by direct methods¹³ and refined on F² by a full-matrix least-squares procedure. SHELXL was used for both structure solutions and refinements.¹⁴ All non-hydrogen atoms were refined anisotropically. The positions of hydrogen atoms were calculated and isotropically fixed in the final refinement. The SMART and SAINT software packages¹⁵ were used for data collection and reduction respectively. Crystallographic diagrams were drawn using the DIAMOND software package.¹⁶ Also severely disordered methanol molecules in Zn^II₂Eu^III₄, Zn^II₂Tb^III₄ and Zn^II₂Dy^III₄ were removed by SQUEEZE during the structural refinements.¹⁷ For details about the squeezed material, see CIF data in ESI.† Therefore, five methanol molecules which were determined on the basis of TGA and elemental microanalysis, and the data treated with the SQUEEZE routine within PLATON were added to the molecular formula of Zn^II₂Eu^III₄, Zn^II₂Tb^III₄ and Zn^II₂Dy^III₄ respectively.

Result and discussion

Synthesis and characterization

A stepwise synthetic protocol was employed to prepare the three heterometallic hexanuclear Zn^II–Ln^III clusters. A suspension solution of zinc containing solution was obtained from the treatment of Zn(NO₃)₂·2H₂O with mixture of H₃L, o-vanilline and triethylamine in CH₃CN in a 1 : 1 : 2 molar ratio. Addition of a solution of Ln(NO₃)₃·6H₂O in MeOH followed by magnetic stirring gave a clear pale yellow solution from which yellow crystals suitable for crystal analysis of Zn^II₂Eu^III₄, Zn^II₂Tb^III₄ and Zn^II₂Dy^II₄ were obtained during the evaporation of the solvent after three weeks. The series of compounds were soluble in methanol, DMSO and DMF. Their chemical formulas were confirmed by elemental analysis. Consistent with the successful formation of the desired compounds, the characteristic bands of carbonyl and imine of the H₃L (1648 and 1598 cm⁻¹) hypochromatic shifted about 30 and 49 cm⁻¹, with two new bands presented at IR spectra ca. 1618 cm⁻¹ and 1544 cm⁻¹ respectively. The none but ν₃ (1380 cm⁻¹) of the free nitrate groups' appearance in the IR spectra of the three compounds clearly indicates the exclusive existence of free nitrate groups as further confirmed by X-ray crystallography analysis as follows.

Thermal gravimetric analysis (TGA) were carried out under N₂ atmosphere to show their thermal stability. As shown in Fig. S1,† the three hexanuclear complexes Zn^II₂Eu^III₄, Zn^II₂Tb^III₄ and Zn^II₂Dy^III₄ showed very similar TG curves, and Zn^II₂Dy^III₄ is selected to describe as a representative. The thermal decomposition of Zn^II₂Dy^III₄ occurs in a three-step process from 25 to 800 °C. The first stage takes place in the range of 27–112 °C with a weight loss of 3.36% which was caused by the release of three crystalline methanol molecules free of hydrogen bonding. Upon further heating to about 168 °C, the two crystalline hydrogen bonded nitrates and methanol molecules were lost and the weight loss was 6.54%. Above 172 °C, the entire architecture began to collapse. PXRD experiments were also carried out for Zn^II₂Eu^III₄, Zn^II₂Tb^III₄ and Zn^II₂Dy^III₄ to verify whether the crystal structures are truly representative of the bulk materials. As we can see in Fig. S2,† the experimental patterns are in good agreement with that of the simulated ones indicating the as-synthesized samples are pure enough for spectroscopic and magnetic properties' study.

Description of structures

Signal-crystal X-ray diffraction analysis revealed that the as-synthesized Zn^II–Ln^III clusters are all sandwich-type hexanuclear compounds with a formula of [Zn₂Ln₄(HL)₄(o-vanilline)₂(OH)₄(CH₃OH)₂]·2NO₃·5CH₃OH. For they are isostructural, here we select Zn^II₂Dy^III₄ to describe their structure features in detail. The crystal structure of Zn^II₂Dy^III₄ consists of a divalent cationic entity [Zn₂Dy₄(HL)₄(o-vanilline)₂(OH)₄(CH₃OH)₂]²⁺, two uncoordinated nitrate anions for charge balance, and five crystalline methanol molecules. A cursory glance at the structure reveals that a μ₃-OH⁻-bridged Dy^III₄ core is sandwiched between two layers composed of one Zn^II, two HL²⁻ and one deprotonated o-vanilline (Fig. 1a) with the asymmetric unit shows half of the molecule. As depicted in Fig. 1b, Zn1 is in an tetragonal pyramidal geometry coordinated by two phenolate oxygen atoms from two different HL²⁻ ligands, one oxygen atom from μ₃-hydroxyl group, and two imine nitrogen atoms from two different HL²⁻ ligands. Meanwhile, two crystallographically independent Dy^III ions (Dy1 and Dy2) are merely coordinated to eight oxygen atoms with Dy–O bond lengths range between 2.189 and 2.645 Å. Exact geometry analysis by SHAPE 2.1 software shows that the inner coordination sphere of eight-coordinated Dy^III ions are residing in distorted biaugmented trigonal prisms with a deviation of 6.174 for Dy1 and 8.035 for Dy2 from the ideal C_2v symmetry (Fig. S3†).^9k,18 Owing to the two oxygen atoms from two hydroxyl groups (O11 and O15) and one phenoxo oxygen atom (O12) of singly deprotonated o-vanilline, Dy1 and Dy2 are bridged to give a trigonal pyramid [Dy₂O₃] unit. Furthermore, Zn1 is appended at Dy2 site with two oxygen bridge (O4 and O15) giving a Zn⋯Dy distance of 3.609 (3) Å. Noticeably, the oxygen atoms both from two hydroxyl groups (O11) and four phenoxo oxygen atom of doubly deprotonated HL²⁻ anion (O7) and singly deprotonated o-vanilline (O12) further bridged the crystallographically Dy1 and Dy2 forming two plane-sharing cubic units [Dy₄O₆] short of one vertex (Fig. 1c). The Dy⋯Dy distances of the four precisely coplanar Dy^III ions are 3.567(7) Å and 3.889(5) Å. Such tetranuclear subunits with four μ₃-OH⁻ displaced above and below the Dy₄ plane by 0.869(2) Å and 0.490(4) Å can be viewed as a parallelogram and the shortest edge, Dy1⋯Dy2A, is that which comprises three oxygen bridges. Noticeably, this structural characteristic is quite different from reported OH⁻ bridged Dy₄ clusters.¹⁹

Fig. 1 (a) Structure of the cation [Zn₂Dy₄(HL)₄(o-vanilline)₂(OH)₄(CH₃OH)₂]²⁺ in Zn^II₂Dy^III₄ (b) asymmetric unit with atomic labels in Zn^II₂Dy^III₄ (H atoms as well as crystalline nitrate anion and methanol molecules are omitted for clarity); (c) the phenoxy/hydroxo-bridged metal framework [Zn₂Dy₄O₈] in Zn^II₂Dy^III₄.

Luminescent properties

Chromogenic Zn^II components have been used as sensitizers for lanthanide luminescence following ligand → f and d → f energy-transfers.²⁰ The photophysical properties of H₃L, o-vanilline, Zn^II₂Eu^III₄, Zn^II₂Tb^III₄ and Zn^II₂Dy^III₄ in solid state were studied and their emission spectra is shown in Fig. 2.

Fig. 2 Emission spectra of H₃L, o-vanilline, Zn^II₂Eu^III₄, Zn^II₂Tb^III₄ and Zn^II₂Dy^III₄ in solid state (excitation and emission passes = 2.5 nm).

Upon UV irradiation at 340 nm, both H₃L and o-vanilline present very broad bands ranging from 435 nm to 600 nm, which could be attributed to π* → π and intramolecular charge transitions of H₃L and o-vanilline. The excitation spectra of Zn^II₂Eu^III₄ exhibits broad bands ranging from 325 nm to 425 nm upon Eu^III-centered emission (Fig. S4†). Excited at ca. 365 nm, Zn^II₂Eu^III₄ shows reddish luminescence and exhibits both ligand-centered (λ_max = 456 nm) and the typical f → f transitions of the Eu^III ion with the former stronger than the latter. As for the characteristic emission of Eu^III ion, the strong emission peak at 599 and 621 nm is assigned to magnetic dipolar ⁵D₀ → ⁷F₁ transition and the electric dipolar ⁵D₀ → ⁷F₂ transition respectively, the weak one at 582 nm can be attributed to the ⁵D₀ → ⁷F₀ transition. The intensity of the ⁵D₀ → ⁷F₂ transition is about 2.26 times stronger than that of the ⁵D₀ → ⁷F₁ transitions, higher than the value (0.67) for a centrosymmetric Eu^III compound.²¹ The symmetry-forbidden emission ⁵D₀ → ⁷F₀ also appears in the emission spectra of Zn^II₂Eu^III₄. All these indicate that Eu^III ions in Zn^II₂Eu^III₄ occupy sites with low symmetry and have no inversion center which is in good agreement with the result of the single crystal structure analyses. The luminescence decay of Zn^II₂Eu^III₄ is best described by a double-exponential process with significantly shorter lifetimes of τ₁ = 0.245 ms, τ₂ = 0.072 ms observed likely because of the presence of two distinct emitting species (Fig. S5†). When the same system were adopted to populate the excited state of the Tb^III ion, the ligands-centered emission rather than Tb^III luminescence is observed, indicating this mixed system can transfer some excitation energy to Eu^III than to Tb^III ions. Compared to that of Zn^II₂Eu^III₄, the luminescence lifetime of Zn^II₂Tb^III₄ is quite shorter with τ = 0.097 ms (Fig. S6†). In the case of Zn^II₂Dy^III₄, very strong blue emission with shortest lifetime of 9 ns (Fig. S7†) attributed to Zn–ligands system presented. It is well-known that the presence of OH oscillators in the lanthanide first coordination sphere provides an efficient non-radiative path,²² so we can suggest that the observed weak Eu^III luminescence in Zn^II₂Eu^III₄ and the absence of Tb^III/Dy^III-centered emission are mainly related to increases in non-radiative transitions due to the presence of OH oscillators in the first coordination shell together with the mismatch between resonance energy levels of Ln^III and excited states of ligands in these clusters. Notably, the emission spectra of the three clusters showed negligible changes when excited at different wavelengths as shown in Fig. S8–S10.†

Magnetic properties

The static magnetic behaviour of the Zn^II₂Dy^III₄ complex through direct current (dc) magnetic measurements between 2 and 300 K with an applied dc field of 1000 Oe was performed on microcrystalline samples. As observed in Fig. 3a, the χ_MT value almost remains unchanged from 300 to 12 K with a value of 57.44 cm³ K mol⁻¹ at 300 K which is fairly close to the expected value of 56.68 cm³ K mol⁻¹ for four uncoupled Dy^III ions (⁶H_15/2, g = 4/3). Below 10 K, the χ_MT value sharply goes up upon cooling and reaches 83.11 cm³ K mol⁻¹ at 2 K. This behavior indicates that spin–orbital coupling leads the 4fⁿ configuration of Ln^III to split into ^2S+1L_J states, and further into Stark components under the crystal-field perturbation, and the effect of depopulation of the Stark components of Dy^III is nearly equal of the ferromagnetic coupling between Dy^III ions in the higher temperature range. The increase of χ_MT in the low-temperature range obviously implies the presence of ferromagnetic coupling between Dy^III ions, and it is strong enough to overcome the effect of depopulation of the Stark components of Dy^III.⁸ Such strong ferromagnetic interactions are rare for both Zn^II–Ln^III clusters and Ln^III complexes.²³ Its occurrence may result from the connecting of two [Dy₂Zn] building blocks through two μ₃-OH⁻-bridge. In Zn^II₂Dy^III₄, the [Dy₄O₆] core was closely wrapped and the Dy–OH⁻–Dy angles with shortest distance of 3.567(7) Å are 96.51 and 98.76°, which therefore may lead to the strong uniaxial anisotropy of the Dy^III ion.^23d By comparison, a similar Dy₄ clusters with the Dy–OH⁻–Dy angle above 103.95° exhibits antiferromagnetic interactions.^19a The M vs. H data below 8 K show a rapid increase in the magnetization at low magnetic fields (Fig. 2b) which is expected for materials having ferromagnetically coupled spins. at higher fields, M increases slowly reaching a value of 36 μ_B which almost close to the theoretical value of 40 μ_B of four Dy^III ions. The nonsuperposition of M vs. HT⁻¹ in different fields (Fig. 2b inset) suggests the presence of significant magnetic anisotropy and/or low-lying excited states in Zn^II₂Dy^III₄.

Fig. 3 (a) Temperature dependence of χ_MT for Zn^II₂Dy^III₄ at 1000 Oe over the temperature of 2–300 K; (b) M vs. H plots and M vs. HT⁻¹ plots (inset) for Zn^II₂Dy^III₄ at different temperatures.

In addition, it is worth mentioning that the M vs. H data do not exhibit a hysteresis effect above 2 K with sweep rates used (100–300 Oe min⁻¹). The absence of the M vs. H hysteresis loop at 2 K may be caused by the presence of a relatively fast zero-field relaxation as ascertained as follows.²⁴

To further explore the dynamics of magnetization, we performed alternating current (ac) magnetic measurements on Zn^II₂Dy^III₄. Expectedly, strong frequency dependences for the real χ′_M–T and the imaginary χ′′_M–T parts of the ac susceptibilities were observed below 20 K as shown in Fig. 4. In plots of both χ′_M–T and χ′′_M–T, the peaks are observed above 20 Hz and shift to high temperature with increasing frequencies, which is the nature of slow magnetic relaxation. Noticeably, both χ′_M and χ′′_M below about 6 K increase with decreasing temperature, indicating the intervention of QTM. For Kramer's ions, such as Dy^III, dipole–dipole and hyperfine interactions allow the mixing of the two Kramer's ground states at zero field, leading to the quantum tunnelling dynamics of the magnetization. Therefore, the temperature and frequency dependent ac susceptibility were measured under an applied 2000 Oe field which is also shown in Fig. 4. The diminishing of the upward χ′_M and χ′′_M at low temperatures can be taken as a clear indication of the efficient suppression of zero-field tunneling of magnetization occurring in Zn^II₂Dy^III₄. In addition, compared with magnetization at zero-field, the out-of-phase component χ′′_M for higher frequencies exhibits a series of frequency-dependent peaks around 16.5 K, with a second set around 6 K, which indicates two relaxation processes dominate for Zn₂^IIDy₄^III under an applied 2 kOe field. The low temperature one (RLT) and high temperature one (RHT) could be ascribed to the two different Dy^III ions centers in Zn^II₂Dy^III₄.²⁵

Fig. 4 Ac magnetic measurements for Zn^II₂Dy^III₄ at H_ac = 3 Oe and H_dc = 0 Oe/2000 Oe respectively.

Soncini et al. used an electrostatic model to determine the magnetic anisotropy in dysprosium complexes and their results indicated the ground state of Dy^III is doublet quantized along the anisotropy axis with an angular momentum quantum number m_J = ±15/2 in absence of high symmetry.²⁶ Theoretical research has presented that the large magnetic anisotropy is obtained when only the low-lying ground state, |±15/2〉 Kramers doublet, is significantly populated.²⁷ In case of Zn^II₂Dy^III₄, the combination of long and short Dy–O bond length in the ligand field can facilitate the stability of the oblate shape of |±15/2〉 Kramers doublet, which in turn leads to strong magnetic anisotropy.²⁹ The peak temperatures, T_p, obtained by the Lorentzian peak function fitting from plots of χ′′_M–T deduce a linear plot of 1/T_p vs. ln(2πf) and obey the Arrhénius law 1/T_p = −k_B/ΔE[ln(2πf) + ln τ₀] (Fig. 5), where f is the frequency. The best fitting of out-of-phase component χ′′_M under zero dc field yields the energy barrier ΔE/k_B = 56.7 K and the relaxation time τ₀ = 4.8 × 10⁻⁷ s, which are comparable to those of reported Dy^III-based SMMS.²⁸

Fig. 5 The least-squares fit of the experimental data at H_dc = 0 Oe to the Arrhenius equation.

Plotting the relaxation time τ of out-of-phase component χ′′_M under 2000 Oe dc field versus the temperature T_p⁻¹ and a linear fitting of the thermally activated points to the Arrhenius law afford ΔE₁/k_B = 6.30 K and the relaxation time τ₁ = 9.55 × 10⁻³ s for RLT and ΔE₂/k_B = 67.1 K and τ₂ = 1.66 × 10⁻⁶ s for RHT (Fig. 6). As a comparison, the energy barriers extracted from the high temperature regions is only a slight increase of the energy barrier, indicating that the quantum tunneling effect in this compound is not very pronounced.²⁹

Fig. 6 The least-squares fit of the experimental data at H_dc = 2000 Oe to the Arrhenius equation.

The frequency dependence of ac susceptibilities at zero field leads to the semicircle Cole–Cole plots of χ′′_M vs. χ′_M (Fig. 7) at 8, 10, 12 and 14 K, respectively. The least-squares fitting results of the data are in good agreement with a distribution of single relaxation processes with α parameters of 0.07–0.15 (α = 0 corresponding to an infinitely narrow distribution of relaxation times), which is also compatible with the value reported.³⁰ The shift of the peak temperature (T_p) of χ′′_M is measured by a parameter ϕ = (ΔT_p/T_p)/Δ(log f) = 0.26, which falls in the range of a normal value for a superparamagnet.⁸

Fig. 7 Plots of χ′′_M vs. χ′_M of Zn₂Dy₄ at 8 K (○), 10 K (◇), 12 K (□) and 14 K ([hexagon open, flat-side down]) respectively. Red solid lines are the least-squares fitting results to a distribution of single relaxation processes with the small α value.

Conclusions

The self-assembly of a methoxy substituted salicylamide Salen-like ligand and o-vanilline in Zn^II–Ln^III chemistry led to three sandwich-like heterometallic clusters with hydroxyl bridged tetranuclear Ln^III₄ core swaddled by Zn^II ion and deprotonated ligands. Photophysical determination indicate that Zn^II₂Eu^III₄ shows mixed emission both originating from ligands and Eu^III ions, and ligand-centered emission in visible region is exclusively found for Zn^II₂Tb^III₄ and Zn^II₂Dy^III₄. Interestingly, Zn^II₂Dy^III₄ exhibits strong ferromagnetic magnetic behavior with a slow zero-field relaxation with τ₀ = 4.8 × 10⁻⁷ s and ΔE/k_B = 56.7 K. To our best knowledge, it is quite unusual that strong ferromagnetic coupling and slow magnetic relaxation coexist in Zn^II–Ln^III heterometallic clusters. Our current studies on construction of high-nuclearity heterometallic clusters are helpful in achieving further insights into the rational design and preparation of novel multifunctional Zn^II–Ln^III heterometallic clusters. Further studies focused on the construction of d–f nanoclusters of Salen-like salicylamide ligands with other auxiliary ligands, different substitution and different d-metal ions are in progress in our group.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grant: 21661019) and Gansu Natural Science Foundation of China. (Grant: 1212RJZA038).

Notes and references

1. (a) M. Brewer, J. Lee and J. G. Brennan, Inorg. Chem., 1995, 34, 5919; (b) M. Berardini, T. J. Emge and J. G. Brennan, Inorg. Chem., 1995, 34, 5327; (c) J. Ruiz, G. Lorusso, M. Evangelisti, E. K. Brechin, S. J. A. Pope and E. Colacio, Inorg. Chem., 2014, 53, 3586; (d) T. D. Pasatoiu, C. Tiseanu, A. M. Madalan, B. Jurca, C. Duhayon, J. P. Sutter and M. Andruh, Inorg. Chem., 2011, 50, 5879.
2. K. Z. Su, F. L. Jiang, J. J. Qian, M. Y. Wu, K. C. Xiong, Y. L. Gai and M. C. Hong, Inorg. Chem., 2013, 52, 3780.
3. H.-B. Xu, H.-M. Wen, Z.-H. Chen, J. Li, L.-X. Shi and Z.-N. Chen, Dalton Trans., 2010, 39, 1948.
4. X.-Q. Lü, W.-X. Feng, Y.-N. Hui, T. Wei, J.-R. Song, S.-S. Zhao, W.-Y. Wong, W.-K. Wong and R. A. Jones, Eur. J. Inorg. Chem., 2010, 2010, 2714.
5. Z.-S. Meng, F.-S. Guo, J.-L. Liu, J.-D. Leng and M.-L. Tong, Dalton Trans., 2012, 41, 2320.
6. X. P. Yang, Z. P. Li, S. Q. Wang, S. M. Huang, D. Schipperc and R. A. Jones, Chem. Commun., 2014, 50, 15569.
7. L. Zhang, L. Zhao, P. Zhang, C. Wang, S.-W. Yuan and J. K. Tang, Inorg. Chem., 2015, 54, 11535.
8. Z. Chen, B. Zhao, P. Cheng, X.-Q. Zhao, W. Shi and Y. Song, Inorg. Chem., 2009, 48, 3493.
9. (a) H. Zhang, G.-L. Zhuang, X.-J. Kong, Y.-P. Ren, L.-S. Long, R.-B. Huang and L.-S. Zheng, Cryst. Growth Des., 2013, 13, 2493; (b) R. Modak, Y. Sikdar, G. Cosquer, S. Hatterjee, M. Yamashita and S. Goswami, Inorg. Chem., 2016, 55, 691; (c) S. Osa, T. Kido, N. Matsumoto, N. Re, A. Pochaba and J. Mrozinski, J. Am. Chem. Soc., 2004, 126, 420; (d) S. F. Xue, Y.-N. Guo, L. Zhao, H. X. Zhang and J. K. Tang, Inorg. Chem., 2014, 53, 8165; (e) I. A. Kühne, G. E. Kostakis, C. E. Anson and A. K. Powell, Inorg. Chem., 2016, 55, 4072; (f) J. Q. Zhang, C. Li, J. J. Wang, M. Zhu and L. C. Li, Eur. J. Inorg. Chem., 2016, 2016, 1383; (g) X. F. Wang, P. Hu and J.-P. Sutter, Inorg. Chem., 2015, 54, 9664; (h) H.-R. Wen, J. Bao, S.-J. Liu, C. M. Liu, C.-W. Zhang and Y.-Z. Tang, Dalton Trans., 2015, 44, 11191; (i) W.-W. Kuang, L. L. Zhu, L. C. Li and P.-P. Yang, Eur. J. Inorg. Chem., 2015, 2015, 2245; (j) Y.-A. Liu, C.-Y. Wang, M. Zhang and X.-Q. Song, Polyhedron, 2017, 127, 278; (k) F. J. Kettles, V. A. Milway, F. Una, R. Valiente, L. H. Thomas, W. Wernsdorfer, S. T. Ochsenbein and M. Murrie, Inorg. Chem., 2014, 53, 8970.
10. X.-Q. Song, P.-P. Liu, Y.-A. Liu, J.-J. Zhou and X. L. Wang, Dalton Trans., 2016, 45, 8154.
11. X.-Q. Song, P.-P. Liu, Z.-R. Xiao, X. Li and Y.-A. Liu, Inorg. Chim. Acta, 2015, 438, 232.
12. D. C. Santra, K. B. Manas, K. S. Pradip and M. Sudip, Chem.–Eur. J., 2016, 22, 2012.
13. G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 1990, 46, 467.
14. G. M. Sheldrick, SHELXL-2014, University of Gottingen, Gottingen, Germany, 2014.
15. (a) SAINT, version 6.02, Bruker AXS Inc, Madison, WI, 2002; (b) K. Bhattacharya, M. Maity, D. Mondal, A. Endo and M. Chaudhury, Inorg. Chem., 2012, 51, 7454.
16. (a) DIAMOND, Visual Crystal Structure Information System, version 3.1, Crystal Impact, Bonn, Germany, 2004; (b) X.-Q. Song, Y.-K. Lei, X.-R. Wang, M.-M. Zhao, Y.-Q. Peng and G.-Q. Cheng, J. Solid State Chem., 2014, 218, 202.
17. A. L. Spek, J. Appl. Crystallogr., 2003, 36, 7.
18. A. K. Mondal, H. S. Jena, A. Malviya and S. Konar, Inorg. Chem., 2016, 55, 5237.
19. (a) P.-H. Guo, J.-L. Liu, Z.-M. Zhang, L. Ungur, L. F. Chibotaru, J. D. Leng, F.-S. Guo and M.-L. Tong, Inorg. Chem., 2012, 51, 1233; (b) H. S. Ke, P. Gamez, L. Zhao, G.-F. Xu, S. F. Xue and J. K. Tang, Inorg. Chem., 2010, 49, 7549.
20. X. P. Yang, D. Schipper, R. A. Jones, L. A. Lytwak, B. J. Holliday and S. M. Huang, J. Am. Chem. Soc., 2013, 135, 8468.
21. M. Latva, H. Takalo, V.-M. Mukkala, C. Matachescu, J. C. Rodríguez-Ubis and J. Kankare, J. Lumin., 1997, 75, 149.
22. (a) W. D. Horrocks Jr, J. Am. Chem. Soc., 1979, 101, 334; (b) X. Q. Song, X. Y. Zhou, W. S. Liu, W. Dou, J. X. Ma, X. L. Tang and J. R. Zheng, Inorg. Chem., 2008, 47, 11501.
23. (a) M. Chen, E. C. Sañudo, E. Jiménez, S.-M. Fang, C.-S. Liu and M. Du, Inorg. Chem., 2014, 53, 6708; (b) R.-X. Yao, X. Xu and X.-M. Zhang, RSC Adv., 2014, 4, 53954; (c) S.-J. Liu, J.-P. Zhao, W.-C. Song, S.-D. Han, Z.-Y. Liu and X.-H. Bu, Inorg. Chem., 2013, 52, 2103; (d) P. Zhang, L. Zhang, S.-Y. Lin and J. K. Tang, Inorg. Chem., 2013, 52, 6595.
24. P.-H. Lin, T. J. Burchell, R. Clérac and M. Murugesu, Angew. Chem., Int. Ed., 2008, 47, 8848.
25. W.-B. Sun, B. Yan, L.-H. Jia, B.-W. Wang, Q. Yang, X. Cheng, H.-F. Li, P. Chen, Z.-M. Wang and S. Gao, Dalton Trans., 2016, 45, 8790.
26. N. F. Chilton, D. Collison, E. J. L. McInnes, R. E. P. Winpenny and A. Soncini, Nat. Commun., 2013, 4, 2551.
27. I. Oyarzabal, J. Ruiz, J. M. Eco, M. Evangelisti, A. Camón, E. Ruiz, D. Aravena and E. Colacio, Chem.–Eur. J., 2014, 20, 14262.
28. S. Liu, K. Lang, Y. Zhang, Q. Yang, B. Wang and S. A Gao, Dalton Trans., 2016, 45, 8149.
29. G. Abbas, Y. Lan, G. E. Kostakis, W. Wernsdorfer, C. E. Anson and A. K. Powell, Inorg. Chem., 2010, 49, 8067.
30. (a) W.-M. Wang, W.-Z. Qiao, H.-X. Zhang, S.-Y. Wang, Y.-Y. Nie, H.-M. Chen, Z. Liu, H.-L. Gao, J.-Z. Cui and B. Zhao, Dalton Trans., 2016, 45, 8182; (b) G. Xiong, X. Y. Qin, P. F. Shi, Y. L. Hou, J. Z. Cui and B. Zhao, Chem. Commun., 2014, 50, 4255; (c) R. Modak, Y. Sikdar, G. Cosquer, S. Chatterjee, M. Yamashita and S. Goswami, Inorg. Chem., 2016, 55, 691.

---

Footnote

† Electronic supplementary information (ESI) available: A brief statement in CIF, figures of TGA curves (Fig. S1), figures of PXRD curves (Fig. S2), coordination polyhedron of Dy^III in Zn^II₂Dy^III₄ (Fig. S3), excitation spectra of Zn^II₂Eu^III₄ (Fig. S4), emission decay curves of Zn^II₂Eu^III₄, Zn^II₂Eu^III₄, and Zn^II₂Dy^III₄ (Fig. S5–S7), emission spectra of different wavelength excitation (Fig. S8–S10). A summary of the relevant crystallographic data and the final refinement details are given in Table S1, important bond lengths are listed in Table S2. CCDC 482428–1482430. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c7ra01469d

---