Recent advances in dysprosium-based single molecule magnets: Structural overview and synthetic strategies

Abstract

The last few years have seen a huge renaissance in the study of the magnetism of lanthanide coordination complexes, especially in the field of single molecule magnets (SMMs) due to the large inherent anisotropy of lanthanide metal ions. It has led to intense activity on the part of synthetic chemists to produce systems suitable for detailed study by physicists and materials scientists, thus synthetic development has been playing a major role in the advancement of this field. In this review, we demonstrate the research developed in the few years in the fascinating and challenging field of Dy-based SMMs with particular focus on how recent studies tend to address the issue of relaxation dynamics in these systems from synthetic point of view. In addition, the assembly of multinuclear Dy SMMs using various ligands is summarized, showing that several typical motifs are favorable structural units which could be exploited in the formation of new Dy-based SMMs and supramolecular architectures.

Introduction

The interest in nanoscale magnetic materials has been driven by the rapid growth in high-speed computers and high-density magnetic storage devices with the promise of a revolution in information technology [1], [2]. The discovery of single molecule magnet (SMM), where slow relaxation and quantum tunneling of the magnetization result from a molecular-based blocking anisotropy [3], is recognized as an important breakthrough in the field of nanomagnetism [4]. It opens up a popular avenue to nanoscale electronic devices, sensors and high-density data storage media at the molecular level (the ultimate size limit) [5], [6], [7]. Further, SMMs provide unique opportunities to observe quantum effects (quantum tunneling of magnetization (QTM) and quantum phase interference) [8], [9], [10], [11], because they straddle the interface between classical and quantum mechanical behavior and all magnetic particles, based on tailor-made molecules, are identical and monodisperse [11]. Magnetic properties of SMMs depend strongly on their intrinsic magnetic characteristics, such as spin ground state (S) and magneto-crystalline anisotropy (D), which lead to a spin-reversal barrier (U_eff) for the slow relaxation of magnetization, U_eff = |D|S² and |D|(S² − 1/4) for a transition metal single molecule magnet (TM-SMM) with integer and half integer spins, respectively [3], [8]. D must be negative in order to give rise to this spin-reversal barrier. A negative D is indicative of Ising type magnetic anisotropy.

The initial study was just confined to the realm of coordination complexes based on 3d metals, including SMMs (the first wave of SMMs, Scheme 1) [3], [12], [13], [14], [15], [16], [17], [18], [19] and single chain magnets (SCMs) [20], [21], [22]. In particular, the discoveries of Co^II-organic radical and Mn^III–Ni^II chains in 2001 and 2002 provide the experimental confirmation of Glauber's prediction for SCMs [23]. However, recently particular emphasis has been placed on the design of new SMMs applying 4f metal ions [24], [25], as a result of their significant magnetic anisotropy arising from the large, unquenched orbital angular momentum. As shown in Scheme 1, the discovery that lanthanide mononuclear complexes can exhibit slow relaxation of the magnetization has initiated intensive interest in the SMMs containing lanthanide metals (4f/3d–4f), leading to the second wave of SMMs [5], [26], [27], [28], [29]. Herein, the Dy^III ion seems to be especially useful in this respect. Dy^III-radical family (4f–2p) was considered and used over the last years as a bench for understanding the magnetism of the lanthanide ions and has given rise to many groundbreaking results in SMMs [5], [30] and SCMs in recent years. Owing to the strong Ising type anisotropy of Dy^III ion, Dy^III-radical chains are very appealing candidates for constructing SCMs [31], [32]. Research into heterometallic dysprosium complexes has also led to a flood of intriguing SMMs, including Cr–Dy^III [33], Mn–Dy^III [34], [35], [36], [37], [38], Fe–Dy^III [39], [40], [41], Co–Dy^III [42], [43], [44], Ni–Dy^III [45], [46], [47] and Cu–Dy^III [48], [49], [50], even a 5d–Dy^III complex [51].

Since a triangular Dy₃ cluster with toroidal arrangement of magnetic moments on the dysprosium sites (spin chirality on the ground states) shows SMM behavior of thermally excited spin states [27], [52], [53], pure Dy^III-based SMMs with different topologies have really been the center of attention for chemical, physical, and material scientists and yielded a flood of remarkable results [26], [54], [55] such as the highest relaxation energy barriers for multinuclear clusters (U_eff = 528 K for Dy₅) [29] and the highest temperature at which hysteresis has been observed for any single molecule magnet (T = 8.3 and 14 K for Dy₂ and Tb₂, respectively) [5], [56]. Our recent efforts have been considerably dedicated to this field [28], [57], [58], [59]. Herein, we provided a critical discussion on the most up-to-date achievements associated with Dy^III-based SMMs, regarding the synthesis, structural motifs (Scheme 2) and magnetic studies of Dy single ion magnet to complicated Dy₂₆ cluster, with an emphasis focusing on the synthetic strategies to Dy^III-based compounds with the aim to shed light on the design of to Dy^III-based compounds with specific magnetic properties.

The applications of dysprosium to some advanced materials, such as MRI (magnetic resonance imaging), MS (effect of the magnetostriction), and SMMs, suggest its great potential in the field of magnetism [1], [60], [61], [62]. In MRI, Dy^III-based complexes are considered to be ideal negative contrast agents at high magnetic fields compared with Gd^III-based complexes [63]. In MS, metal dysprosium has a huge saturation MS λ_s value about 100 times larger than for Fe, Ni and Co [62]. Especially for SMMs, Dy^III ion represents a very ideal ion, because, as a number of the late lanthanides (4fⁿ, n > 7), it can provide stronger magnetic anisotropy than the early lanthanides (n < 7) despite the weak exchange coupling [64], [65]. In particular, the Dy^III ion possesses an odd number of electrons (n = 9), thus insuring the Kramers doublet ground state [66], a critical factor in the presence of typical SMM properties. This has been indicated in a series of mononuclear lanthanide compounds, discovered by Ishikawa, with a trigonal (D₃) symmetry where Dy^III, Er^III, and Yb^III complexes show field-induced SMM behavior, while Tb^III, Ho^III and Tm^III complexes are non-SMMs due to fast quantum tunneling [67]. All those suggest that dysprosium will play a crucial role in the exploitation of molecular magnetic materials.

The free Dy^III ion is characterized by f⁹ configuration, which, in combination with spin–orbit coupling effects, give rise to ⁶H_15/2 (^2S+1L_J) multiplets with 16-fold (2J + 1) degeneracy [65]. Nevertheless, the surrounding crystal field can remove those degenerate multiplets into new sublevel structure characterized by m_J = ±15/2, ±13/2, ±11/2, ±9/2, ±7/2, ±5/2, ±3/2, ±1/2 (high symmetry). The doubly degenerate ±m_J state cannot be separated by the crystal field because of the feature of spin-parity effect for a Kramers ion (odd numbers of 4f electrons) [64]. This is due to time reversal symmetry, and the pairs of degenerate levels are called Kramers doublets [68]. From the point view of magnetism, the doublets demonstrate the decisive role for the occurrence of SMM behavior. Furthermore, a suitable crystal field can place the higher m_J states (±15/2 or ±13/2) as the ground state, and make a large separation between the ground state with other m_J states, which defines the energy required to relax the spin and will further afford a high relaxation barrier [26], [69], [70]. In result, two underlying conditions for SMM behavior, i.e. doubly-degenerate ground states with a high ±m_J value and a large separation between ground states and excited states, are easy to be achieved for a Dy-based compound [64].

Persistent axiality of Dy^III ions in lanthanide compounds is an important feature enabling them functioning as SMMs and responsible for the unusual magnetic behavior such as multiple relaxation processes and spin chirality due to the noncollinearity of their single-ion anisotropic axes [28], [52], [55]. For highly symmetric mononuclear systems, such as phthalocyanine double-decker Dy complexes ([DyPc₂]) [26] with the local symmetry of D_4d, the unaxial anisotropy (C₄) seems to be clear. In particular, Long and coworkers have developed a simple model to explain and predict the presence of significant single-ion anisotropy, based on the shape variation of the f-electron charge cloud [64]. However, in polynuclear Dy^III systems or Dy^III monomers with low-symmetry environments, the determination of single-ion anisotropy should be rather tricky, where the simple analysis of anisotropy may be misled by oversimplification associated with the idealized symmetry, as seen in Dy-DOTA (1, [Na{Dy(DOTA)(H₂O)}]·4H₂O, H₄DOTA = 1,4,7,10-tetraazacyclododecane N,N′,N″,N‴-tetraacetic acid) [71]. Therefore, complicated accurate measurements should be taken into consideration for a deep insight into the magnetism of Dy^III ion. The Electron Paramagnetic Resonance (EPR) technique suitable for transition-metal ions is not available for characterization of its magnetic anisotropy, because anisotropic rare earths present very fast electronic relaxation that broadens the EPR signal, hampering a precise determination of the g tensor [30]. Recently, post-Hartree–Fock ab initio calculations have proved to be an invaluable method for predicting the magnetic anisotropy of Dy^III ions in some multinuclear Dy^III systems (triangular Dy₃ and planar Dy₄). Besides energies of the multiplets, directions of the anisotropy axes and the g tensors for the lowest Kramers doublets of each dysprosium site can be obtained, providing a theoretical prediction of the single-ion anisotropy [52], [55]. To explore the nature of single dysprosium anisotropy in low-symmetry environments, Sessoli and coworkers initiated the single-crystal magnetic measurements in molecular magnetism, which straightforwardly confirmed the presence of local anisotropy in mononuclear 4f systems, such as Dy-DOTA (1) [71] and [Dy(hfac)₃-(NIT-R)₂] (2, hfac⁻ = hexafluoroacetylacetonate; NIT-R = 2-R-4,4,5,5-tetramethylimidazolidine-3-oxide-1-oxyl) [30]. In fact, single-crystal susceptibility work has been performed by R. Mason, S. Mitra and others before SMMs were known [72], [73]. Moreover, an interesting toroidal arrangement of the spins has been simultaneously confirmed by single-crystal magnetic measurements and ab initio calculations [52].

The coordination of a H₂O molecule breaks the square antiprism coordinated environment in the first coordination sphere, and the tetragonal symmetry is also suppressed in a larger coordination sphere due to the coordinated Na⁺ ion [74]. Thus, simple analysis based on the idealized symmetry of the lanthanide complex can be misleading. Here the triclinic symmetry is present in Dy^III-DOTA (1), which is suited for single-crystal magnetic measurements [71].

The single-crystal magnetic measurements on Dy-DOTA (1) reveal the high angular dependence of the magnetization, suggesting a strong magnetic anisotropy of Dy^III ion. The diagonalization of susceptibility tensor (χ) can provide the principal values of the susceptibility and their orientation: g_// = 17.0 for an effective S_eff = 1/2, which confirms the Ising anisotropy of the Dy^III ion in the low-symmetry environment. The orientation of the easy axis (violet rod) is shown in Fig. 1 left. To ensure the accuracy, ab initio calculations including all the atoms of the complex have been performed. The effective g tensor of the ground-state doublet of the ⁶H_15/2 has also been estimated at 18.6, approaching the Ising limit. The orientation of anisotropy (pale blue rod) is close (10°, within experimental uncertainty) to the experimental one, and is almost perpendicular to the idealized symmetry axis. In addition, the calculations were performed on several reduced models, revealing that the hydrogen bonds on coordinated water molecule play a key role in the unexpected orientation of the easy axis.

In addition, the single-ion anisotropy of Dy^III ion in Dy^III-radical complex [Dy(hfac)₃-(NIT-R)₂] (2) [30] was investigated as early as 2009 (Fig. 2). The complex crystallizes in the triclinic space group with only one Dy ion located in a distorted square-antiprismatic environment. The analytic outcomes of angle-resolved magnetic measurements show a strong Ising-type anisotropy for the complex. The largest g component of ground-state doublet obtained from ab initio calculations is about 18, and its orientation is very close (7°) to the experimental one.

The above examples provide experimental and theoretical evidence of the local magnetic anisotropy of the Dy^III ion in a low-symmetry environment. It is obvious that the determination of magnetic anisotropy of lanthanide ion in a low-symmetry environment is fraught with difficulties and simple magneto-structural correlations based on the coordination environment are not enough. Detailed information on the type of anisotropy and on the orientation of the principal axes of the magnetic tensors in the molecular structure can only be obtained by a joint effort, combining single-crystal magnetic studies with ab initio calculations.