Lanthanide single-molecule magnets with high anisotropy barrier: where to from here?

Abstract

High-temperature lanthanide single-molecule magnets with large coercive fields are being pursued.

Lanthanide complexes with magnetic hysteresis known as lanthanide single-molecule magnets (Ln-SMMs) are promising candidates for high-density information storage and quantum technologies. Due to the predominantly electrostatic nature of 4f–ligand interactions, the coordination geometry of lanthanide complexes is contingent upon the steric constraints. The oblate–prolate model suggests that axial coordination environment with linear arrangement of more repulsive ligands is suitable for oblate ions, e.g. Dy³⁺, while equatorially coordinating geometry is preferable for prolate ions, e.g. Er³⁺ [1]. In terms of synthesis, by contrast to the placement of more repulsive ligands along a single axis, it is more challenging to achieve the arrangement along a plane. Therefore, few ligands have been discovered to produce effective equatorial crystal field (CF) and the most popular one in constructing Er-based SMMs is dianion of substituted cyclooctatetraenyl (Fig. 1A and C). In contrast, various N-/O-donor ligands rooted in Werner-type coordination chemistry, e.g. Schiff-base and phenolate/alkoxide, are employed to enhance the magnetic anisotropy of Dy³⁺ complexes (Fig. 1A). In addition, for powerful Tb-based SMMs, high molecular symmetry is necessary to guarantee ±M_J degeneracy, which is indeed difficult in synthesis due to the large radii and variable coordination patterns of Ln³⁺ ions. Fortunately, the Kramers nature of Dy³⁺ ensures a bistable ground state regardless of the coordination geometry. Furthermore, the coordination geometry does not need to be perfectly axial to stabilize the magnetic ground state of M_J = ±15/2 of Dy³⁺ ions, i.e. allowing the existence of weak transverse CF. Therefore, some low-symmetry Dy-based SMMs with high anisotropy barrier (U_eff, the energy required to convert an SMM back into a simple paramagnet) were constructed by the predominant bond that usually refers to the shortest Dy–X chemical bond (X is the closest bonding atom) (Fig. 1A and B). To sum up, the design of Ln-SMMs with high U_eff is almost exclusively driven by static CF considerations.

Figure 1.

(A) Timeline of the development of high-performance Ln-SMMs [numbers in blue and red are U_eff (cm⁻¹) and T_B (K), respectively. Asterisk indicates the two independent studies of the same compound.]. (B) Relationship between U_eff and T_B of the compounds in Fig. 1A. (C) Solid structures of [Er(COT)₂]⁻ (left) [2] and [(Cp^iPr5)Dy(Cp*)]⁺ (right) [3]. (D) Singly occupied Ln–Ln molecular orbital (left) and orientation of Ln magnetic moments in Ln₂@C₈₀(CH₂Ph) (right) [4]. (E) Solid structure of (Cp^iPr5)₂Ln₂I₃ (left) and molecular orbital diagram for (Cp^iPr5)₂Gd₂I₃ (right). Adapted from [5] with permission from American Association for the Advancement of Science. (F) A possible synthesis route for half-sandwich dysprosium metallocene.

However, U_eff can only provide information on magnetic relaxation in the high-temperature regime corresponding to the Orbach mechanism, while the blocking temperature (T_B, the temperature at which the relaxation time of the magnetization equals the characteristic time of the experiment) is a more informative parameter involving all sources of demagnetization. As we all know, the Raman and quantum tunneling processes together prevent T_B landing in the exponential relaxation regime. Over the last several years, air-sensitive dysprosium metallocene cations, [Dy(Cp^R)₂]⁺, have emerged as clear front-runners in the search for high-temperature Ln-SMMs. They possessed both large U_eff and high T_B benefitting from the near-perfect axiality and the constrained metal–ligand vibrational modes, among which compound [(Cp^iPr5)Dy(Cp*)][B(C₆F₅)₄] (Cp^iPr5 = penta-iso-propylcyclopentadienyl, Cp* = pentamethylcyclopentadienyl), exhibits T_B up to 80 K (Fig. 1C), exceeding the boiling point of liquid nitrogen [3]. In addition, it is noteworthy that Dy³⁺ corannulene complexes are expected to have both improved air stability and excellent SMM properties [6].

The most notable exceptions beyond the above design criteria are radical-bridged Ln-SMMs, which usually display a modest U_eff but a high T_B and/or a huge coercive field due to the strong exchange coupling. The ultimate realization of such exchange coupling might be completed by a single-electron Ln–Ln bond. The confined space in fullerene provides a unique opportunity for building such a metal–metal bond. For example, Tb₂@C₈₀(CH₂Ph) showed U_eff and T_B of 555 cm⁻¹ and 28 K, benefitting the combination of strong single-ion magnetic anisotropy and exchange coupling from a single-electron Tb–Tb bond (Fig. 1D) [4]. Furthermore, ab initio calculations revealed a higher U_eff and a stronger exchange coupling in Tb₂@C₅₉N [7]. However, the coulombic repulsion between the lanthanide spins in fullerene is stronger than the bonding interaction, so it is necessary to find a more conventional way to construct a single-electron Ln–Ln bond, which was recently completed in mixed-valence dilanthanide complexes, (Cp^iPr5)₂Ln₂I₃ (Fig. 1E) [5]. The singly occupied Ln–Ln σ-bonding orbital of 5d_z² parentage enables the well-isolated high-spin ground state even at room temperature. The 4f-σ coupling constant in (Cp^iPr5)₂Gd₂I₃ reaches 387(4) cm⁻¹. Compound (Cp^iPr5)₂Dy₂I₃ exhibited U_eff and 100 s-T_B of 1631(25) cm⁻¹ and 72 K, respectively, as well as an enormous coercive magnetic field with a lower bound of 14 T at 60 K. Note that all of them are the largest values for any reported Ln-SMMs.

The above advances demonstrate that the most important figure of merit for SMMs has evolved from U_eff to T_B. In particular, the dominance of the Cp ligand in enhancing T_B has garnered considerable attention, although ab initio spin dynamics methodology revealed that [(Cp^iPr5)Dy(Cp*)][B(C₆F₅)₄] has reached the upper limit to the U_eff of [Dy(Cp^R)₂]⁺, which means the predominant limiting mechanism for a further increase in T_B is the Orbach process [8]. Given the strong interaction between Dy³⁺ and phenolate, we proposed a half-sandwich dysprosium metallocene (Fig. 1F), which is expected to break this limit by virtue of the combination of large CF splitting from phenolate and unique vibrational modes of the Cp ligand. Besides, the molecular vibrations also need to be taken into account as they play a prominent role in under-barrier magnetic relaxations by coupling to spin states or an electrostatic polarization effect [9]. In general, detrimental vibrational modes can be blocked by decoupling spin from vibration in the chemical approach or choosing a substrate with a low phonon density of state, e.g. MgO [10]. It is noteworthy that in addition to the requirements in physical properties of SMMs, their stability under ambient conditions is also crucial to realize the proposed applications, so a complementary stream of research has focused on improving SMM performance in air-stable lanthanide macrocycles.

FUNDING

This work was supported by the Key Research Program of Frontier Sciences, Chinese Academy of Sciences (ZDBSLY-SLH023), the National Natural Science Foundation of China (21871247) and the National Science and Technology Major Project (2020YFE0204500).

Conflict of interest statement

None declared.