Constraining and Tuning the Coordination Geometry of a Lanthanide

Oct 14, 2015 - coordination geometry around metal centers in a high- dimensional molecular system, namely, metal−organic frame- works (MOFs), which ...
0 downloads 0 Views 4MB Size
Download PDF
Article pubs.acs.org/IC

Constraining and Tuning the Coordination Geometry of a Lanthanide Ion in Metal−Organic Frameworks: Approach toward a Single-Molecule Magnet Ke Liu, Huanhuan Li, Xuejing Zhang, Wei Shi,* and Peng Cheng Department of Chemistry, Key Laboratory of Advanced Energy Materials Chemistry (MOE), Collaborative Innovation Center of Chemical Science and Engineering (Tianjin), Nankai University, Tianjin 300071, P. R. China S Supporting Information *

ABSTRACT: It is available to constrain and tune the coordination geometries around lanthanide ions in metal−organic frameworks (MOFs) for the study of single-molecule-magnet (SMM) behavior. A series of DyIII-MOFs are synthesized via a solvothermal method by using furan-2,5-dicarboxylic acid (H2FDA) as the ligand. {[Dy2(FDA)3(DMF)2]·1.5DMF}n (1) and [Dy2(FDA)3(DMF)2(CH3OH)]n (2) show similar threedimensional structures, but the coordination geometries around the dysprosium(III) ions in 1 and 2 exhibit different deviations from ideal square antiprism (D4d symmetry) because of the coordinated solvent molecules. Slow relaxation of the magnetization can be observed for both complexes, indicative of SMM behavior. The effective energy barriers for 1 and 2 can be obtained from alternating-current susceptibility measurements by applying an external 2000 Oe direct-current field. MOF 2 possesses a less distorted D4d coordination sphere and gives a higher effective energy barrier (Ueff) than that of MOF 1. Their diamagnetic YIII-diluted samples 1@Y and 2@Y exhibit similar relationships between the geometries and Ueff values, demonstrating that the magnetization relaxation is mainly from the symmetry-related single-ion behavior.

■

INTRODUCTION Lanthanide ions are widely used as excellent magnetic building blocks for single-molecule magnets (SMMs) recently because of the large magnetic moment and significant single-ion anisotropy derived from spin−orbit coupling and the crystalfield effect, which are not easily satisfied in transition-metal complexes.1 For Kramers’ dysprosium(III) ion, its ground state will always be bistable irrespective of the ligand-field symmetry. Thus, it is a great benefit for dysprosium(III) complexes to behave as SMMs.2 Nevertheless, the strength and symmetry of the crystal field is still worthy of more attention3 because the interaction between f electrons and the crystal field will greatly influence the magnetic anisotropy barrier that blocks reorientation of the magentizations.4 Strategies to enhance the energy barriers for magnetization reversal have been proposed. Rinehart and Long promoted that magnetic anisotropy of lanthanide ions can be efficiently improved by the appropriate ligand field whose shape matches the electronic density of free metal ions.5 On the other hand, quantum tunneling of magnetization (QTM), which is often observed in lanthanide complexes and always lowers the effective energy barriers,6 originates from the overlap of wave functions with different mJ states7 and depends mainly on the influence of the crystal field, intra/intermolecular interactions, and hyperfine interactions.8 It is known that several particular © XXXX American Chemical Society

local symmetries would efficiently preclude QTM effects, such as D5h and D4d, and generate high energy barriers for magnetization reversal.9,10 Thus, fine-tuning the crystal field and local symmetry around an individual lanthanide ion has become a successful strategy to get enhanced SMM behaviors, which was well demonstrated in a zero-dimensional molecular system.3,9,11 However, it is still quite challenging to fine-tune the coordination geometry of a lanthanide ion because of its large ionic radius and high coordination number. In this contribution, we present a new strategy to constrain the coordination geometry around metal centers in a highdimensional molecular system, namely, metal−organic frameworks (MOFs), which benefits from the structural features of the rigid frameworks.12 Although employing SMMs as building blocks to construct MOFs has been documented recently,13,14 from the viewpoint of constraining and tuning local geometries around lanthanide ions for tuning the SMM behaviors, examples obtained using this strategy are rather rare.14 Two DyIII-MOFs with furan-2,5-dicarboxylic acid (H2FDA) have been obtained via the solvothermal method, namely, {[Dy2(FDA)3(DMF)2]·1.5DMF}n (1) and [Dy2(FDA)3(DMF)2(CH3OH)]n (2). As shown in Scheme 1, Received: June 17, 2015

A

DOI: 10.1021/acs.inorgchem.5b01356 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Scheme 1. Constraining and Tuning the Local Geometries of Lanthanide Ions in MOFs

because of the different mixed solvents employed in the synthesis process, the dysprosium(III) ions in the latter one possess less distorted coordination spheres from ideal D4d than the former one, resulting in considerable improvements of the SMM behavior.

■

Table 1. Crystal Data and Structural Refinement Details for 1, 2, and 1′ formula fw temp (K) cryst syst space group a (Å) b (Å) c (Å) α (deg) β (deg) γ (deg) V (Å3) Z Dc (g/cm3) μ (mm−1) Rint GOF R1 [I > 2σ(I)] wR2 [I > 2σ(I)] R1 (all data) wR2 (all data) Δρmax (e/ Å3) Δρmin (e/Å3)

EXPERIMENTAL SECTION

General Procedures. All chemicals were obtained commercially and used without further purification. Analyses for carbon, hydrogen, and nitrogen were obtained on a PerkinElmer 240 CHN elemental analyzer. Fourier transform infrared spectroscopy was carried in the range 400−4000 cm−1 on a Bruker TENOR 27 spectrophotometer using KBr pellets. Powder X-ray diffraction (PXRD) measurements were performed on a Rigaku D/Max-2500 X-ray diffractometer using Cu Kα radiation. Inductively coupled plasma atomic emission spectroscopy (ICP-AES) was measured by ICP-9000(N+M) (Thermo Jarrell-Ash Corp., USA). Thermogravimetric analyses (TGA; N2 atmosphere) were carried out in a Labsys NETZSCH TG 09 Setaram apparatus with a heating rate of 1.5 °C/min. Magnetic susceptibilities were measured on a Quantum Design SQUID MPMS VSM magnetometer. Microcrystalline samples were put into sample holders and fixed with eicosane. Diamagnetic corrections were made with Pascal’s constants for all of the constituent atoms and sample holders. X-ray Crystallography. Crystallographic data and refinement details are given in Table 1. Selected bond and angle parameters are listed in Table S1. Single crystals of 1, 2, and 1′ available for X-ray diffraction were measured at 123(2) K to determine their structures. The measurements were performed on a Agilent Technologies SuperNova single-crystal diffractometer with graphite-monochromatic Mo Kα radiation (λ = 0.71073 Å) by ω-scan mode. All of the structures were solved by direct methods with SHELXS, and all nonhydrogen atoms are refined by full-matrix least-squares techniques with anisotropic thermal factors using SHELXL. The hydrogen atoms are fixed by using riding mode and treated isotropically with the program package Olex2. CCDC 1041095 for 1, 1041096 for 2, and 1418234 for 1′ contain the supplementary crystallographic data for

1

2

1′

C24H20Dy2N2O17 933.42 123(2) monoclinic C2/c 18.3085(13) 9.0788(8) 19.6464(19) 90 97.068(8) 90 3240.8(5) 4 1.913 4.651 0.0488 0.996 0.0409

C26H28Dy2N2O19 997.50 123(2) monoclinic C2/c 18.465(5) 9.2442(14) 19.699(5) 90 102.18(3) 90 3286.7(13) 4 2.016 4.597 0.0612 1.145 0.0700

C24H20Dy2N2O17 933.42 123(2) orthorhombic Pnma 8.6326(3) 28.1206(10) 15.3462(5) 90 90 90 3725.3(2) 4 1.664 4.046 0.0498 1.019 0.0442

0.0867

0.1832

0.1404

0.0422 0.0920

0.0705 0.1965

0.0568 0.1507

1.439

3.034

1.829

−0.870

−2.709

−2.461

this paper (Supporting Information). These data can also be obtained free of charge from the Cambridge Crystallographic Data Centre via www.ccdc.cam.uk/data-request/cif. B

DOI: 10.1021/acs.inorgchem.5b01356 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 1. Asymmetric unit of 1 (a), the coordination environment (b), and a binuclear building block (c). Synthesis of 1. A mixture of H2FDA (0.20 mmol), Dy(NO3)3· 6H2O (0.10 mmol), 2 mL of N,N-dimethylformamide (DMF), and 6 mL of CH2Cl2 was sealed in a 25 mL Teflon-lined autoclave, heated at 100 °C for 3 days, and then slowly cooled to room temperature over an additional 3 days. Colorless block crystals were collected after filtration and dried in air. The mass of the crystals obtained was about 25 mg (50% yield based on dysprosium). Elem anal. Found (calcd) for C28.5H30.5Dy2N3.5O18.5: C, 32.91 (32.82); H, 2.50 (2.95); N, 4.80 (4.70). IR (KBr, ν/cm−1): 3357 (s), 1569 (s), 1417 (s), 1108 (w), 1019 (m), 967 (m), 832 (w), 782 (s), 677 (m), 621 (m), 505 (s). Synthesis of 2. The synthetic process is similar to that of 1, except CH2Cl2 was replaced by CH3OH. Light-brown block crystals were collected after filtration and dried in air. About 40 mg of crystals were obtained (80% yield based on dysprosium). Elem anal. Found (calcd) for C26H28Dy2N2O19: C, 30.86 (31.31); H, 2.66 (2.83); N, 2.99 (2.81). IR (KBr, ν/cm−1): 3350 (br), 1665 (br), 1414 (s), 1222 (w), 1107 (w), 1019 (s), 968 (m), 780 (s), 620 (m), 507 (s). Synthesis of 1′. A mixture of H2FDA (0.20 mmol), Dy(NO3)3· 6H2O (0.10 mmol), and 2 mL of DMF was sealed in a 25 mL Teflonlined autoclave, heated at 120 °C for 24 h, and then cooled to room temperature. Colorless block crystals were collected after filtration and dried in air. The mass of the crystals obtained was about 25 mg (50% yield based on dysprosium). Elem anal. Found (calcd) for C30.6H35.4Dy2N4.2O19.2: C, 33.64 (33.59); H, 3.45 (3.26); N, 5.45 (5.38). IR (KBr, ν/cm−1): 3357 (s), 1569 (s), 1417 (s), 1108 (w), 1019 (m), 967 (m), 832 (w), 782 (s), 677 (m), 621 (m), 505 (s). Synthesis of 1@Y. The synthetic process is similar to that of 1, except Dy(NO3)3·6H2O was replaced by a mixture of Y(NO3)3·6H2O and Dy(NO3)3·6H2O in a methanol solution in a molar ratio of 50:1. Colorless block crystals were collected after filtration and dried in air. The mass of the crystals obtained was about 30 mg (70% yield based on yttrium). Elem anal. Found (calcd) for C28.5H30.5Dy0.041Y1.959N3.5O18.5: C, 38.22 (38.08); H, 3.53 (3.42); N, 5.49 (5.45). ICP-AES anal. Found: Dy, 0.84; Y, 22.06. IR (KBr, ν/ cm−1): 3357 (s), 1569 (s), 1417 (s), 1108 (w), 1019 (m), 967 (m), 832 (w), 782 (s), 677 (m), 621 (m), 505 (s). Synthesis of 2@Y. The synthetic process is similar to that of 2, except Dy(NO3)3·6H2O was replaced by a mixture of Y(NO3)3·6H2O

and Dy(NO3)3·6H2O in a methanol solution in a molar ratio of 50:1. Light-brown block crystals were collected after filtration and dried in air. The mass of the crystals obtained was about 35 mg (80% yield based on yttrium). Elem anal. Found (calcd) for C26H28Dy0.047Y1.953N2O19: C, 36.86 (36.57); H, 3.46 (3.31); N, 3.29 (3.28). ICP-AES anal. Found: Dy, 0.89; Y, 20.34. IR (KBr, ν/cm−1): 3350 (br), 1666 (br), 1414 (s), 1224 (w), 1106 (w), 1019 (s), 969 (m), 781 (s), 620 (m), 506 (s).

■

RESULTS AND DISCUSSION Crystal Structures. Single-crystal X-ray measurements reveal that complexes 1 and 2 crystallize in the monoclinic C2/c space group. As shown in Figures 1a,b and S1, peripheral positions of the Dy1 center of 1 are occupied by two oxygen atoms from one μ2-η1:η2-carboxylate group (O6 and O8), four oxygen atoms from four separate μ2-η1:η1-carboxylate groups (O1, O2, O4, and O5), one μ2-bridging carboxyl oxygen atom (O6A), and one oxygen of the coordinated DMF molecule (O7). The Dy−O bond lengths are in the range of 2.258(2)− 2.754(2) Å. The neighboring dysprosium(III) ions are connected in a binuclear secondary building unit (SBU; Figure 1c) via two μ2-η1:η1-carboxylate groups and two μ2-η1:η2carboxylate groups. The SBUs are further linked to each other through pairs of μ2-η1:η1-carboxylate groups to form a onedimensional carboxyl-bridged dysprosium chain (Figure 2). The distances between adjacent intrachain dysprosium(III) ions are 3.990(4) and 5.326(5) Å, respectively. The nearest Dy···Dy distance between different dysprosium chains is 8.151(1) Å. The chains are further connected via the ligand to construct a (3,8)-connected three-dimensional tfz-d topological framework with a point symbol of {43}2{46·618·84} (Figure S2).15 MOF 2 shows a slightly different coordination environment compared with MOF 1 because of its coordinated methanol molecules. The dysprosium(III) ion in 2 is also eightC

DOI: 10.1021/acs.inorgchem.5b01356 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 3. Asymmetric unit of 1′ (a), the coordination environment (b), and a binuclear building block (c). Figure 2. One-dimensional carboxyl-bridged dysprosium chain and the three-dimensional framework of 1.

PXRD and TGA. The PXRD patterns of 1, 2, and their corresponding diamagnetic yttrium(III) diluted samples are collected at room temperature to confirm the purity and isomorphism (Figures S8 and S9). The PXRD patterns of 1′ are collected at room temperature to confirm the purity (Figure S10). The experimental patterns are in good agreement with the simulated peaks based on the single-crystal X-ray data. TGA of 1, 1′, and 2 are performed in the temperature range of 30− 900 °C (Figure S11). The weight of 1 begins to decrease above 200 °C, indicating that DMF molecules start to escape from the frameworks. The framework of 1 may decompose during DMF’s exit. MOF 2 exhibits a rather different thermal behavior: the coordinated methanol molecules leave at about 100 °C, and the coordinated DMF molecules leave at about 230 °C. MOF 1′ does not lose weight until 250 °C, and the abrupt decrease of weight above 250 °C implies decomposition of the framework. Magnetic Properties. The direct-current (dc) magnetic susceptibilities of MOFs 1 and 2 are collected in the temperature range of 300−2 K under an applied field of 1000 Oe (Figure 4). The χMT products of 1 and 2 at room temperature are 27.43 and 27.71 cm3 K mol−1, respectively, which are close to the expected value of 28.34 cm3 K mol−1 for two free dysprosium(III) ions (6H15/2 and g = 4/3). Upon cooling, the χMT values of both MOFs decrease slowly until 100 K and then decrease quickly in the range of 50−20 K to reach the minimum at 14 K for 1 and at 12 K for 2. These behaviors are mainly attributed to thermal depopulation of the MJ states of the dysprosium(III) ions and the crystal-field effect.17 The χMT values increase sharply when the temperature is further decreased to 2 K, suggesting the presence of ferromagnetic interactions.18 The magnitude of the ferromagnetic interactions in MOFs 1 and 2 can be estimated by the noncritical scaling theory using the sum of two exponential functions: χT = C1 exp(E1/T) + C2 exp(E2/T), in which C1 + C2 is the extrapolated Curie constant at room temperature while E1 and E2 represent the magnitudes of intrachain magnetic interactions and the crystal-field contributions.19 The small positive E1 values of 0.56(9) cm−1 for 1 and 0.50(6) cm−1 for 2 indicate the weak but nonnegligible

coordinate, completed by six carboxylic oxygen atoms, one DMF, and one methanol molecule (Figures S3 and S4). The Dy−O bond lengths are in the range of 2.256(9)−2.594(1) Å. The neighboring dysprosium(III) ions in 2 are bridged by four μ2-η1:η1-carboxylate groups into binuclear units (Figure S5), and its topology is similar to that of MOF 1. The distances between intrachain adjacent dysprosium(III) ions are 4.122(9) and 5.271(0) Å, respectively. The nearest interchain Dy···Dy distance is 8.447(2) Å. The geometries around the dysprosium(III) ions in 1 and 2 are deviated from the ideal eight-coordinate polyhedron, which can be evaluated through a continuous-symmetry-measure method using the program SHAPE (Tables S2 and S3).16 The calculations demonstrate that dysprosium(III) ions in MOFs 1 and 2 exhibit biaugmented trigonal-prismatic geometries (C2v) and the estimated deviation parameters are 1.579 and 0.671 for 1 and 2, respectively. It is worth noting that the calculated deviation parameter from SAPR-8 is 1.786 for 2, suggesting a moderate deviation from an ideal D4d symmetry. The deviation parameter from SAPR-8 is as large as 4.076 for 1. Because of the occupancy of the two coordinated solvents, all FDA2− ligands in MOF 2 adopt a μ2-η1:η1-bridging mode and less distorted DyO8 polyhedra from D4d symmetry are isolated. In addition, a different MOF with a similar formula of MOF 1, {[Dy2(FDA)3(DMF)2]·2.2DMF} (1′), has been obtained by varying the synthetic conditions. MOF 1′ possesses a sevencoordinate dysprosium(III) ion whose coordination sphere is completed by six carboxylic oxygen atoms and one DMF molecule (Figures 3 and S6). The geometry around dysprosium(III) can be described as monocapped triangularprismatic (C 2v ; Tables S4 and S5). The neighboring dysprosium(III) ions are bridged by carboxylic groups into one-dimensional chains, which are further connected via FDA2− ligands to construct a three-dimensional stp topological framework (Figure S7). D

DOI: 10.1021/acs.inorgchem.5b01356 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 4. Temperature-dependent magnetic susceptibilities of 1 (a) and 2 (b). The fitting lines are obtained from the scaling model (red line), the ferromagnet (dashed line), and the crystal-field component (dash-dotted line).

temperature-dependent ac susceptibilities. The parameter φ [=(ΔTp/Tp)/Δ(logv)] is 0.23, lying well in the range of 0.1 < φ < 0.3 and excluding the possibility of spin-glass behavior.24 The relaxation time obeys the Arrhenius law, and the best fit gives the effective energy barrier of 41.8 K and τ0 = 1.08 × 10−8 s (Figure S19). Similar relaxation behaviors can be observed for 2 when a dc field of 2000 Oe is applied (Figure S20). The shift parameter φ is 0.14, and the effective energy barrier is 67.5 K with τ0 of 2.26 × 10−11 s for 2 (Figure S21). To investigate whether there are Ising-type chain dynamics, plots of ln(χMT) versus 1/T are subtracted in the temperature range of 2−10 K. As shown in Figures S22 and S23, the plots reveal a clear linearity in the temperature range, but the correlation lengths Δζ for 1 and 2 are very low (0.81 K for 1 and 0.75 K for 2), probably because of the very weak intrachain magnetic interactions. Therefore, relaxation of the magnetizations for the two MOFs is best described as SMM behavior and mainly derived from the single-ion property of the dysprosium(III) ion. To better understand the single-ion contribution to the magnetization dynamics, diluted samples of 1 and 2 in their yttrium(III) analogues in a molar ratio of 1:48 for 1@Y and 1:42 for 2@Y are obtained to preclude the ferromagnetic interactions and the QTM effects originating from dipolar magnetic interactions between dysprosium(III) ions. Because the local geometries around metal centers can be constrained by the frameworks, it is expected that single-ion anisotropies of dysprosium(III) ions remain the same as those in MOFs 1 and 2. The dc magnetic susceptibilities for 1@Y and 2@Y continuously decrease upon cooling, in agreement with the single-ion behaviors (Figures S24 and S25). The field dependence of the magnetizations for 1@Y and 2@Y reveals the presence of magnetic anisotropy, the crystal-field effect, and/or low-lying excited states (Figures S26 and S27). ac susceptibilities of 1@Y and 2@Y are collected at zero dc field (Figures S28−S31). However, the maxima of out-of-phase signals are still not observed in the range of 1−1000 Hz above 1.8 K. To further suppress QTM, an external dc field is applied in the measurements. It is found that the field of 300 Oe is the optimum field for both MOFs that can minimize the quantum relaxation pathway of the magnetization relaxation (Figures S32 and S33). Therefore, the variable-temperature (Figures S34 and

ferromagnetic interactions between dysprosium(III) ions. The values of C1 and C2 are 22.36(4) and 5.87(4) cm3 K mol−1 for MOF 1 and 21.43(5) and 7.03(6) cm3·K/mol for MOF 2, respectively. The field dependence of the magnetizations for 1 and 2 has been determined at 2.0 K in the range of 0−70 kOe (Figures S12 and S13). The magnetization values rise abruptly at low fields and reach 10.7 Nβ for 1 and 10.0 Nβ for 2 at 70 kOe, much lower than the theoretical saturation value of 20 Nβ due to the presence of magnetic anisotropy, crystal-field effect, and/ or low-lying excited states.20 The alternating-current (ac) magnetic susceptibilities of 1 and 2 are measured at zero dc field in the temperature range of 2−12 K and the frequency range of 1−1000 Hz (Figures S14− S17). Both of them exhibit a significant frequency dependence of out-of-phase signals, strongly suggesting the presence of slow relaxation of the magnetization. However, the χ″ peaks are not observed in the temperature ranges available; thus, the energy barriers for magnetization reversal cannot be determined. The χ″ peaks would possibly be observed above 1000 Hz (Figures S16 and S17), suggesting the presence of a fast relaxation that results from QTM. As a result of a spin-parity effect,21 the mixing of the degenerate sublevels through transverse anisotropy is forbidden in the half-integer spin system.22 However, in a real system, QTM can be observed even in Kramers’ systems induced by several factors such as environmental distortions as well as dipolar coupling via transverse field components. Because the coordination geometries are deviating from the ideal D4d for MOFs 1 and 2, there should be the possibility of relaxation via the xy plane, or the so-called transverse anisotropy. Thus, a magnetization reversal mechanism is allowed via QTM within the lowestenergy doublet as well as thermally activated QTM (TA-QTM) within the excited doublet. Such factors can significantly reduce the effective energy barriers for the magnetization reversal.7a,b QTM can be removed by applying an external dc field that removes the mixing of the degenerate doublets to avoid a crossing region23 or by diluting the paramagnetic centers in diamagnetic yttrium samples, which subtracts the dipolar magnetic interactions between dysprosium(III) ions. A dc field of 2000 Oe is applied to 1 to suppress QTM. As shown in Figure S18, broad peaks of χ′ and χ″ for 1 are observed in the E

DOI: 10.1021/acs.inorgchem.5b01356 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 5. Cole−Cole plots for 1@Y (a) and 2@Y (b). The lines represent the best-fit calculated values with the Debye model. Plots of ln(τ) versus T−1 of 1@Y (c) and 2@Y (d) with fitting results.

Table 2. Parameters of Fitting ln(τ) versus T−1 Plots for MOFs 1@Y and 2@Y Using Eqiation 1 and Deviation Parameters from Ideal Geometries deviation parameter Aa (s−1 K−1) 1@Y 2@Y a

τ0 (s) −8

8.24 × 10 1.46 × 10−7

Ueff/kBT (K)

Cb (s−1 K−5)

τQTM (s−1)

D4d

C2v

37.0 46.0

0.0786 0.0660

11.1 13.4

4.076 1.786

1.579 0.671

Parameter for the direct process n = 1. bParameter for the Raman process m = 5.

lattice energy exchange.26 The significant deviations from linearity of the plots of ln(τ) versus T−1 for 1@Y and 2@Y also support the presence of multiple relaxation processes. By considering all of the possible relaxation mechanisms, we successed in fitting the entire range of ln(τ) versus T−1 plots by employing the equation

S35) and -frequency (Figure S36) ac susceptibilities are collected under a dc field of 300 Oe. The frequency shift parameter φ is 0.27 for both 1@Y and 2@Y, suggesting that they are superparamagnets rather than spin glass.24 Additionally, the best fits for Cole−Cole semicircles (Figure 5a,b) by using the Debye model give α values ranging from 0.22 to 0.60 for 1@Y and from 0.19 to 0.55 for 2@Y (Tables S6 and S7), indicative of the presence of multiple relaxation processes. Multiple relaxation processes have been observed in many reported f-element-based SMMs because of the existence of different anisotropic centers or isomers and conformers in the crystal.25 In the case of mononuclear complexes, multiple relaxations may be composed of the Orbach, direct, and Raman processes, together with the QTM process. The former three relaxation processes arise from the energy exchanges between the paramagnetic ions and phonon radiations, while QTM is caused by the transitions between different states without spin−

τ −1 = AT n + τ0−1 exp( −Ueff /kBT ) + CT m + τQTM

(1)

in which the four terms represent the contributions of the direct (n = 1 or 2), Orbach, and Raman (m = 5 or 9 for Kramers’ ion) processes and QTM.6a The magnetization relaxation parameters are summarized in Table 2. The bestfitting results suggest that the relaxations for the two MOFs in the high-temperature range are clearly dominated by the Orbach process, with the energy barriers of 37.0 and 46.0 K with τ0 values of 8.24 × 10−8 and 1.46 × 10−7 s for 1@Y and 2@Y, respectively (Figure 5c,d). The improvement of energy F

DOI: 10.1021/acs.inorgchem.5b01356 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Notes

barrier values for 2@Y compared with 1@Y, as well as that for 2 compared with 1, is related to the minor change of their coordination environments: less deviation from ideal geometry and higher energy barriers. Besides, it is rare that the undiluted complexes possess higher energy barriers than the diluted ones. Considering the ferromagnetic interactions between dysprosium(III) ions, it is obvious that magnetic couplings make a positive contribution to the energy barriers in this case. It is noted that the magnetization relaxations of the two MOFs are also related to the fast QTM and Raman processes besides the Orbach process. The parameter τQTM is 11.1 s−1 for 1@Y and 13.4 s−1 for 2@Y, indicative of the obvious presence of fast QTM. The parameter for the direct process obtained by fitting is rather small and can be neglected. In the two diluted samples, dysprosium(III) ions should be thoroughly isolated by diamagnetic yttrium(III) ions. Thus, the magnetic interactions between dysprosium(III) ions should be absolutely eliminated. However, there is still obvious QTM relaxation, which can be ascribed to the deviation from the ideal D4d crystal field around dysprosium(III) ions. Finally, we studied the magnetic properties when coordinated methanol was removed from 2. We have successfully obtained a new complex in which the dysprosium(III) ion is seven-coordinate (1′). No out-of-phase ac signals of 1′ are observed (Figure S39), suggesting the absence of slow relaxation of magnetization. The magnetic behavior of 1′ further confirms the significant role of the coordination geometries around lanthanide ions when constructing SMMs.

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors thank the “973” program (Grant 2012CB821702), NSFC (Grants 21331003, 21373115, and 21421001), and MOE (Grants NCET-13-0305 and IRT-13R30).

■

■

CONCLUSION The crystal-field geometry around the lanthanide ions plays a significant role when pursuing SMMs with remarkable energy barriers for magnetization reversal. Although excellent SMMs have been well demonstrated in discrete molecular systems with specific coordination symmetry, it is still quite challenging to rationally control the local geometries around the lanthanide centers because of the flexible and high coordination number of the lanthanide ion. By the introduction of an MOF as a platform, the coordination geometry of the SMM node can be constrained by the rigid framework and chemically tuned via variation of the synthetic conditions. The coordination geometries in the two MOFs mainly depend on the coordinated solvent molecules, which resulted in different SMM behaviors. This work illustrated a new strategy to obtain SMM by combining the research fields of both molecular nanomagnets and MOFs. Further researches based on this strategy are underway in our laboratory.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.5b01356. Crystallographic data, additional crystal structure diagrams, magnetic diagrams, fitting parameters, and PXRD and TGA patterns (PDF) Crystallographic data in CIF format (CIF)

■

REFERENCES

(1) Ruiz, E.; Cirera, J.; Cano, J.; Alvarez, S.; Loose, C.; Kortus, J. Chem. Commun. 2008, 44, 52−54. (2) (a) Gatteschi, D. Nat. Chem. 2011, 3, 830. (b) Sorace, L.; Benelli, C.; Gatteschi, D. Chem. Soc. Rev. 2011, 40, 3092−3104. (c) Woodruff, D. N.; Winpenny, R. E. P.; Layfield, R. A. Chem. Rev. 2013, 113, 5110−5148. (d) Liu, K.; Shi, W.; Cheng, P. Coord. Chem. Rev. 2015, 289−290, 74−122. (3) (a) Jiang, S. D.; Wang, B. W.; Su, G.; Wang, Z. M.; Gao, S. Angew. Chem., Int. Ed. 2010, 49, 7448−7451. (b) Ungur, L.; Le Roy, J. J.; Korobkov, I.; Murugesu, M.; Chibotaru, L. F. Angew. Chem., Int. Ed. 2014, 53, 4413−4417. (4) (a) Becquere, J. Z. Phys. 1929, 58, 205−208. (b) Stark, J. Nationalsozialismus and Wissenschaft; Zentralverlag der NSDAP, Franz Eher Nachfolger: Munich, Germany1934 (c) Leone, M.; Paoletti, A.; Robotti, N. Phys. Perspect. 2004, 6, 271−294. (5) Rinehart, J. D.; Long, J. R. Chem. Sci. 2011, 2, 2078−2085. (6) (a) Bi, Y.; Guo, Y. N.; Zhao, L.; Guo, L.; Lin, S. Y.; Jiang, S. D.; Tang, J.; Wang, B. W.; Gao, S. Chem. - Eur. J. 2011, 17, 12476−12481. (b) Rinehart, J. D.; Fang, M.; Evans, W. J.; Long, J. R. Nat. Chem. 2011, 3, 538−542. (c) Meihaus, K. R.; Minasian, S. G.; Lukens, W. W.; Kozimor, S. A., Jr.; Shuh, D. K.; Tyliszczak, T.; Long, J. R. J. Am. Chem. Soc. 2014, 136, 6056−6068. (7) (a) Wernsdorfer, W.; Sessoli, R. Science 1999, 284, 133−135. (b) Leuenberger, M. N.; Loss, D. Nature 2001, 410, 789−793. (c) Gatteschi, D.; Sessoli, R. Angew. Chem., Int. Ed. 2003, 42, 268−297. (d) Dei, D.; Gatteschi, D. Angew. Chem., Int. Ed. 2011, 50, 11852− 11858. (8) Gatteschi, D.; Sessoli, R.; Villain, J. Molecular Nanomagnets; Oxford University Press: Oxford, U.K., 2006. (9) (a) Liu, J. L.; Chen, Y. C.; Zheng, Y. Z.; Lin, W. Q.; Ungur, L.; Wernsdorfer, W.; Chibotaru, L. F.; Tong, M. L. Chem. Sci. 2013, 4, 3310−3316. (b) Liu, J. L.; Wu, J. Y.; Chen, Y. C.; Mereacre, V.; Powell, A. K.; Ungur, L.; Chibotaru, L. F.; Chen, X. M.; Tong, M. L. Angew. Chem., Int. Ed. 2014, 53, 12966−12970. (10) (a) Wang, H.; Wang, K.; Tao, J.; Jiang, J. Chem. Commun. 2012, 48, 2973−2975. (b) Chen, G.-J.; Guo, Y.-N.; Tian, J.-L.; Tang, J.; Gu, W.; Liu, X.; Yan, S.-P.; Cheng, P.; Liao, D.-Z. Chem. - Eur. J. 2012, 18, 2484−2487. (11) (a) Dey, M.; Gogoi, N. Angew. Chem., Int. Ed. 2013, 52, 12780− 12782. (b) Zhang, P.; Zhang, L.; Wang, C.; Xue, S.; Lin, S.-Y.; Tang, J. J. Am. Chem. Soc. 2014, 136, 4484−4487. (c) Coutinho, J. T.; Antunes, M. A.; Pereira, L. C. J.; Marcalo, J.; Almeida, M. Chem. Commun. 2014, 50, 10262−10264. (d) Singh, S. K.; Gupta, T.; Shanmugam, M.; Rajaraman, G. Chem. Commun. 2014, 50, 15513−15516. (12) Na, B.; Zhang, X.-J.; Shi, W.; Zhang, Y.-Q.; Wang, B.-W.; Gao, C.; Gao, S.; Cheng, P. Chem. - Eur. J. 2014, 20, 15975−15980. (13) (a) Shi, P.-F.; Zhao, B.; Xiong, G.; Hou, Y.-L.; Cheng, P. Chem. Commun. 2012, 48, 8231−8233. (b) Shi, P.-F.; Xiong, G.; Zhao, B.; Zhang, Z.-Y.; Cheng, P. Chem. Commun. 2013, 49, 2338−2340. (c) Fang, M.; Li, J.-J.; Shi, P.-F.; Zhao, B.; Cheng, P. Dalton Trans. 2013, 42, 6553−6563. (d) Baldoví, J. J.; Coronado, E.; Gaita-Ariño, A.; Gamer, C.; Giménez-Marqués, M.; Espallargas, G. M. Chem. - Eur. J. 2014, 20, 10695−10702. (e) Yi, X.; Calvez, G.; Daiguebonne, C.; Guillou, O.; Bernot, K. Inorg. Chem. 2015, 54, 5213−5219. (f) Aulakh, D.; Pyser, J. B.; Zhang, X.; Yakovenko, A. A.; Dunbar, K. R.; Wriedt, M. J. Am. Chem. Soc. 2015, 137, 9254−9257. (14) Zhang, X.; Vieru, V.; Feng, X.; Liu, J.-L.; Zhang, Z.; Na, B.; Shi, W.; Wang, B.-W.; Powell, A. K.; Chibotaru, L. F.; Gao, S.; Cheng, P.; Long, J. R. Angew. Chem., Int. Ed. 2015, 54, 9861−9865.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. G

DOI: 10.1021/acs.inorgchem.5b01356 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry (15) Blatov, V. A. IUCr Comput. Commission Newsl.. 2006, 7, 4. The program package TOPOS available at http://incrcomputing.ccp14.ac. uk/incrtop/comm/ccom/newsletters/2006nov. (16) (a) Zabrodsky, H.; Peleg, S.; Avnir, D. J. Am. Chem. Soc. 1992, 114, 7843−7851. (b) Pinsky, M.; Avnir, D. Inorg. Chem. 1998, 37, 5575−5582. (17) (a) Bernot, K.; Bogani, L.; Caneschi, A.; Gatteschi, D.; Sessoli, R. J. Am. Chem. Soc. 2006, 128, 7947−7956. (b) Rinck, J.; Novitchi, G.; Van den Heuvel, W.; Ungur, L.; Lan, Y.; Wernsdorfer, W.; Anson, C. E.; Chibotaru, L. F.; Powell, A. K. Angew. Chem., Int. Ed. 2010, 49, 7583−7587. (18) (a) Chen, Z.; Zhao, B.; Cheng, P.; Zhao, X.-Q.; Shi, W.; Song, Y. Inorg. Chem. 2009, 48, 3493−3495. (b) Jia, L.; Chen, Q.; Meng, Y. S.; Sun, H. L.; Gao, S. Chem. Commun. 2014, 50, 6052−6055. (19) (a) Souletie, J.; Rabu, P.; Drillon, M. Phys. Rev. B: Condens. Matter Mater. Phys. 2005, 72, 214427. (b) Zheng, Y.-Z.; Lan, Y.; Wernsdorfer, W.; Anson, C. E.; Powell, A. K. Chem. - Eur. J. 2009, 15, 12566−12570. (20) (a) Rinehart, J. D.; Meihaus, K. R.; Long, J. R. J. Am. Chem. Soc. 2010, 132, 7572−7573. (b) Mondal, K. C.; Sundt, A.; Lan, Y.; Kostakis, G. E.; Waldmann, O.; Ungur, L.; Chibotaru, L. F.; Anson, C. E.; Powell, A. K. Angew. Chem., Int. Ed. 2012, 51, 7550−7554. (21) Boulon, M.-E.; Cucinotta, G.; Luzon, J.; Degl'Innocenti, C.; Perfetti, M.; Bernot, K.; Calvez, G.; Caneschi, A.; Sessoli, R. Angew. Chem., Int. Ed. 2013, 52, 350−354. (22) Kramers, H. A. Proc. R. Acad. Sci. Amsterdam 1930, 33, 959. (23) (a) Ruiz, J.; Mota, A. J.; Rodríguez-Diéguez, A.; Titos, S.; Herrera, J. M.; Ruiz, E.; Cremades, E.; Costes, J. P.; Colacio, E. Chem. Commun. 2012, 48, 7916−7918. (b) Yang, F.; Zhou, Q.; Zhang, Y.; Zeng, G.; Li, G.; Shi, Z.; Wang, B.; Feng, S. Chem. Commun. 2013, 49, 5289−5291. (24) Mydosh, J. A. Spin Glasses: An Experimental Introduction; Taylor & Francis: London, 1993. (25) (a) Guo, Y.-N.; Xu, G.-F.; Gamez, P.; Zhao, L.; Lin, S.-Y.; Deng, R.; Tang, J.; Zhang, H.-J. J. Am. Chem. Soc. 2010, 132, 8538−8539. (b) Chandrasekhar, V.; Das, S.; Dey, A.; Hossain, S.; Sutter, J. P. Inorg. Chem. 2013, 52, 11956−11965. (26) Orbach, R. Proc. R. Soc. London, Ser. A 1961, 264, 458−484.

H

DOI: 10.1021/acs.inorgchem.5b01356 Inorg. Chem. XXXX, XXX, XXX−XXX