2.1. Synthesis
Rhodamine 6G-type ligands have attracted significant attention in the realm of fluorescent sensor. Previous studies have primarily focused on the synthesis of mononuclear rare earth or transition metal complexes [
46,
47,
48,
49,
50,
51,
52]. However, these ligands have yet to be fully unearthed. We speculate that the reaction between Ln
III and the rhodamine 6G ligands might lead to high-nuclearity lanthanide complexes under high pH value. On the basis of the hydrolysis strategy, we have successfully synthesized two new Gd
III clusters
1 and
2 (
Figure 1). Additionally, our research has unveiled the influential factors that impact the synthesis process, including the ratio of central ions to ligands, solvent, alkaline, and reaction temperature.
Figure 1.
Transformation of ring-opened and ring-closed form of Rhodamine 6G-type ligands and its reactions.
Figure 1.
Transformation of ring-opened and ring-closed form of Rhodamine 6G-type ligands and its reactions.
The synthetic procedure of the two clusters is similar with the exception of the molar ratio of Gd:L, i.e. an excess amount of gadolinium nitrate was used in the synthesis of {Gd
14}. Both Complexes were prepared by the reaction of the ligand HL with gadolinium nitrate in a mixed solution of methanol and acetonitrile. A quantity of triethylamine was used to induce the ring close of HL and hydrolysis. The resulting mixture is left undisturbed at room temperature for one week, facilitating formation of yellow plate-like single crystals for {Gd
7}. The reactant mixture was heated at 60°C in an oven for three days, resulting in the formation of yellow, cubic-shaped samples of {Gd
14}. These two Complexes tend to lose solvents in the air. When the crystals are taken out of the solution, they turn from yellow to red at room temperature, and they lose the crystallinity. The powder XRD pattern of Gd
7 illustrates that the main strong peaks show disagreement with that simulated, indicating the desolvation of the crystals (
Figure S1). The peaks at low 2θ angles of PXRD data for Gd
14 are approximately consistent with that simulated.
2.2. Structure
A yellow single crystal of complex 1 was selected for single crystal X-ray diffraction at 100 K. Complex
1 crystallizes in the monoclinic system with the
P2
1/n space group. The volume of the unit cell is very large with a monoclinic system, and therefore the diffraction data are not so good. The command MASK was used during the structural refinement, and 452 electrons were masked per formula unit which account for the missing NO
3- anion, 10 acetonitrile, 10 methanol and 2 H
2O molecules with the total electrons of 451. The crystallographic data and selected bond distances and angles are given in
Tables S1 and S2. The asymmetric unit cell contains three crystallographically independent {Gd
III7} moieties. Because they have similar molecular structures, only one of them is described in detail as a representative. The coordination number of each Gd
III ion in {Gd
7} (
Figure 2) is between 7 and 9. Their coordination patterns are shown in
Figures S2 and S3. Using SHAPE software for calculation, their coordination patterns were obtained as shown in
Tables S3–S5. The central Gd1 ions are coordinated by eight oxygen atoms and have a square antiprism structure
D4d. Among the eight oxygen atoms, four are μ
3-OH
- and another four are μ
3-CH
3O
-. The remaining six Gd
III ions are evenly distributed around the central Gd
III ion, forming a saddle-shaped structure, which is relatively rare in rare earth complexes [
53] (
Figure S4). One ring-closed ligands L
- is coordinated to one peripheral Gd
III ion, and adjacent Gd
III ions are bridged by CH
3O
-/μ
3-OH
- and/or phenoxy oxygen atoms as shown in
Figure 3.
As for complex
2, a yellow cube single crystal was selected for single crystal X-ray diffraction at 100 K. A solvent mask was used and 183.4 electrons were found in a volume of 3448.9 Å
3 in 9 voids per unit cell, which is consistent with the presence of 2 CH
3OH, 2 H
2O and 1.5 CH
3CN per formula unit with 178 electrons. The crystalographic data and selected bond distances and angles are given in
Tables S1 and S6. Complex
2 crystallizes in the tetragonal system with space group of
P4/n and has a
D4h symmetry. The asymmetric unit contains 1/4 of the tetradecanuclear molecule and there are five different kinds of Gd
III ions (Gd1-Gd5,
Figure 4b). They have three different coordination modes (
Table S7), and their coordination patterns are shown in the
Figures S5 and S6. The tetradecanuclear cluster core is neutral, and has a highly symmetrical three-layer double sandwich structure (
Figure 4b). In the structure, four nine-coordinated Gd
III ions form a square plane layer, and a nine-coordinated Gd
III ion is located between the layers. The distance between the two layers is 5.792 Å. The center of the middle layer is six-coordinated μ
6-O
2-, while the outer layers on both sides are μ
4-O
2-. The sandwiched Gd
III ions and the square-shaped layer are connected by μ
3-OH
- ions. The Gd ions are linked together through hydrophilic hydroxo bridges, forming a [Gd
14(μ
6-O)(μ
4-O)
2(μ
3-OH)
16] core. This core contains one octahedral [Gd
6(μ
6-O)(μ
3-OH)
8] unit that shares two apexes with two [Gd
5(μ
4-O)(μ
3-OH)
4] square pyramid moieties. The cluster core is surrounded by eight ring-closed L
- ligands. Additionally, the Gd
III of middle layer are coordinated with two nitrate ions, and Gd
III of outer layers on both sides are coordinated with a nitrate ion and a L
- ligand. Square plane layers are bridged by phenolic oxygen on the ligand as shown in the
Figure 3b.
Although there are many reports on tetranuclear clusters of planar quadrilateral [
54,
55] and nine-nuclear molecules of double-layer sandwich type [
56,
57], the molecular structure of rare earth in the form of a three-layer double sandwich is rare. Similar molecules have been reported before, as shown in the
Figures S7–S10. The tetradecanuclear hydroxo–lanthanide acetylacetonato complexes formulated as Ln
14(μ
4-OH)
2(μ
3-OH)
16(μ-η
2-acac)
8(η
2-acac)
16 (Ln = Tb and Eu, acac
- = acetylacetonato) [
58] and chiral tetradecanuclear hydroxo-lanthanide clusters Ln
14(μ
4-OH)
2(μ
3-OH)
16(μ-η
2-acac)
8(η
2-acac)
16·6H
2O (Ln = Dy and Tb) [
59] have been reported. The ligands used in these two works are both based on acetylacetonato, but the present tetradecanuclear {Gd
III14} are completely different ligands, i.e. ring-closed rhodamine L
-. The use of ortho-nitrophenolate afforded the tetradecanuclear H
18[Ln
14(μ-η
2-o-O
2N-C
6H
4O)
8(η
2-o-O
2N-C
6H
4O)
16(μ
4-O)
2(μ
3-O)
16] (Ln = Dy and Tm; o-O
2N-C
6H
4O
- = o-nitrophenolate) [
60]. Despite the above similar Ln
14 complexes, the μ
6-O
2- in
2 is unique among them.
It is worth mentioning that similar hexadecanuclear molecules [Eu
III16(tfac)
20(CH
3OH)
8(μ
3-OH)
24(μ
6-O)
2] have also been reported based on trifluoroacetylacetone (tfac
-) [
22]. In addition to the tetradecanuclear {Eu
14}, there is another two Eu ions on opposite sides. The study provides insights into the formation, evolution, and assembly of lanthanide hydroxide clusters. The formation of {Gd
7} and {Gd
14} in this work further verifies that the hydrolysis under high pH values is an effective way of constructing high-nuclearity Ln
III species.
2.3. Magnetic Measurements
The temperature dependence of the magnetic susceptibility of Complexes
1 and
2 are measured under a 1000 Oe magnetic field in the range of 2–300 K (
Figure 5). At room temperature, the
χMT value of 54.1 cm
3 K mol
-1 for {Gd
7} and 109.0 cm
3 K mol
-1 for {Gd
14} is close to the theoretical value of 55.09 cm
3 K mol
-1 for heptanuclear and 110.18 cm
3 K mol
-1 for tetradecanuclear uncoupled Gd
III (
S = 7/2,
g = 2,
C = 7.87 cm
3 K mol
-1 per Gd), respectively. For
1, upon lowering the temperature, the
χMT value slightly decreases to 49.76 cm
3 K mol
-1 at 20 K and then rapidly falls to 30.22 cm
3 K mol
-1 at 2 K. 2 exhibits a similar behavior: when lowering the temperature, the
χMT value slightly decreases to 95.4 cm
3 K mol
-1 at 30 K and then rapidly falls to 33.0 cm
3 K mol
-1 at 2 K. These changes indicate the presence of dominant antiferromagnetic interactions between the Gd
III ions in the clusters. The data can be perfectly fitted to the Curie-Weiss law, giving
C = 54.44 cm
3 K mol
-1 and
θ = -1.712 K for {Gd
7} and
C = 110.50 cm
3 K mol
-1 and
θ = -4.629 K for {Gd
14}. Larger absolute
θ value in {Gd
14} suggests that the antiferromagnetic interaction is stronger than that for {Gd
7} (
Figure S11).
The field dependence of the magnetizations (
M) for Complexes
1 and
2 was measured in the temperature range of 2 - 10 K (
Figure S12). It can be seen that the magnetization has not reached to saturation at 5 T and 2 K. At 2 K, the experimental maximum magnetization value of 47.03 Nβ for
1 and 89.35 Nβ for
2 is lower than the theoretical saturation value of Gd
III (49 Nβ for
1 and 98 Nβ for
2, respectively), which may be owing to the antiferromagnetic coupling and higher magnetic field is needed to suppress the magnetic coupling effect. For complexes
1 and
2, the experimental
M-
H curves at 2 K lie below the calculated Brillouin curve for non-interacting
SGd spins (
Figure S12), also suggesting the presence of intermetallic antiferromagnetic coupling. The difference between the experimental and the calculated ones for Gd
14 is obviously larger than that for Gd
7, indicating that the former shows stronger antiferromagnetic coupling than that of the latter. The presence of μ
6-O
2--bridged Gd
6O moiety may be responsible for this.
The half-filled 4f electronic configuration in a Gd
III ion makes it magnetically isotropic. This makes gadolinium a valuable material in various applications, especially magnetic refrigeration. Thus, the magnetocaloric effect (MCE) of Complexes
1 and
2 was studied using the Maxwell equation:
At 3 K and ∆
H = 5 T, the value of ∆
Sm is 17.44 J kg
−1 K
−1 for
1(
Figure 6a), which is slightly lower than the expected value of 14.56
R (25.37 J kg
−1 K
−1) calculated for 7 uncorrelated Gd
III using the equation ∆
Sm = n
Rln(2
S+1) (
R ≈ 8.314 J mol
-1 K
-1). The value of ∆
Sm for
2 is 22.30 J kg
−1 K
−1 at 2 K and ∆
H = 5 T (
Figure 6b), which is close to the expected value of 29.11
R (28.72 J kg
−1 K
−1) calculated for 14 uncorrelated Gd
III. To improve the magnetic refrigeration effect of gadolinium clusters [
37], several approaches can be considered. Firstly, the experimental conditions can be optimized. For instance, the temperature and magnetic field can be carefully controlled to ensure the most efficient operation of the gadolinium clusters. The thermal conductivity of the environment and the pressure during the refrigeration cycle can also be adjusted to minimize energy loss. Secondly, the chemical composition of the clusters can be varied. Gadolinium can be alloyed with other metals to create compounds with different magnetic properties. Thirdly, the size of the clusters can be optimized. The optimal size will depend on the specific setup and application, but generally, smaller clusters have a higher surface-to-volume ratio, which leads to more efficient heat exchange and therefore a stronger refrigeration effect. However, too small clusters may also suffer from higher energy barriers between spin states, which can decrease the magnetic entropy change. Overall, a combination of these strategies can be used to improve the magnetic refrigeration effect of gadolinium clusters for various applications, such as cryogenic cooling of scientific instruments, temperature control in electronics, and energy-efficient refrigeration in households and industries.