A hard permanent magnet through molecular design

Permanent magnets constructed from metal ions and organic linkers using molecular design principles could bring transformative advances in areas such as energy conversion, transportation, and information storage. This comment highlights the recent discovery of a metal–organic magnet ordering at 242 °C, and discusses future research directions and possible applications involving such materials.

open structures comprising metal ions or clusters that are connected together by anionic organic linkers via strong bonding interactions ( Fig. 1) 11 . The expansive chemical space associated with framework design is unparalleled in extended solids, and it permits the design of isoreticular metal-and linker-substituted series. In conjunction with this remarkable degree of synthetic control, the lightweight, highly porous structures of frameworks enable their use in a wide range of applications, from gas storage and separations to catalysis 12 . Moreover, these materials often accommodate post-synthetic modification, such as reductive insertions analogous to the soft chemistry developed for layered solid-state compounds 13 , and their kinetically controlled structures can further allow exchange of metal and linker components 14 . For these reasons, metal-organic frameworks are also versatile materials for the directed assembly of designer, high-temperature magnets 15 .
Recently, Oyarzabal, Clérac, and coworkers reported a remarkable ordering temperature of 242°C for the crystalline framework material Cr(pz) 2 ·0.7LiCl (pz = pyrazine), surpassing V(TCNE)~2·~0.5CH 2 Cl 2 and representing a new record among all molecule-based magnets 16 . This compound was synthesized via chemical reduction of the parent framework trans-CrCl 2 (pz) 2 ( Fig. 2), which was obtained through the reaction of CrCl 2 with pyrazine, conceptionally similar to formation of the molecular complex trans-CrCl 2 (py) 4 (py = pyridine) 17 , and features chloroterminated Cr III ions bridged by an equal number of formally neutral and radical anionic pz linkers to form 2D sheets 18 . An array of physical methods was deployed on Cr(pz) 2 ·0.7LiCl to unambiguously determine that reduction in the Cr(pz) 2 ·0.7LiCl product occurs at both Cr and pz, giving a material containing 2D sheets of square planar S = 2 Cr II centers linked by pz radical anions. The reduction is accompanied by the loss of 1.3 equivalents of LiCl, giving 2D sheets of Cr(pz) 2 intercalated with 0.7 equivalents of LiCl. Strong magnetic coupling between spins in energetically similar orbitals of Cr II and pz engenders the unprecedented ordering temperature, and magnetic anisotropy gives rise to magnetic hysteresis with large coercivities, including 0.75 T at 27°C (Fig. 3). This value constitutes a new record among molecule-based magnets, surpassing that of 60 Oe (0.006 T) for V(TCNE)~2·~0.5CH 2 Cl 2 by over two orders of magnitude. The coercivity-the strength of magnetic field needed to demagnetize a material-in part governs the energy that can be stored or converted by a permanent magnet, and is therefore a key metric to quantify the strength, or "hardness", of a permanent magnet 1 . Importantly, while the square planar Cr II ion does not formally feature first-order orbital angular momentum, transfer of spin density onto the pz linker may lead to a spin-orbitcoupled magnetic moment of Cr, thereby imparting angular momentum 18 . Such a scenario is consistent with the experimentally observed low magnetization values at 7 T, and may be the primary source of the large coercivity (Fig. 3c).
Relative to commercial magnets, the coercivity of 0.75 T obtained for Cr(pz) 2 ·0.7LiCl at 27°C compares reasonably well to the room-temperature values of 1.9 T and 4.4 T for Nd 2 Fe 14 B and SmCo 5 , respectively, and is higher than the rare-earth-free hard ferrites (0.40 T), AlNiCo 5 (0.080 T) and AlNiCo 8 (0.20 T) 16 . These comparisons are particularly promising considering the lack of formal first-order angular momentum in Cr(pz) 2 ·0.7 LiCl. Nevertheless, the corresponding remanence, which quantifies the magnetization of a material remaining upon removal of an applied field and is thus another important parameter in determining magnet performance, of 0.52 μ B per formula unit for Cr (pz) 2 ·0.7LiCl, is dwarfed by the analogous values for commercial magnets, such as 32.5 and 8 μ B for Nd 2 Fe 14 B and SmCo 5 , respectively 19 . While this disparity illustrates the need for further advances, the discovery of high-temperature, strong magnetism in a molecule-based material represents a monumental advance, and it underscores the merit of pursuing metal-organic magnets with design features such as covalent bonding between paramagnetic metals and organic linkers.

Outlook
The modular nature of Cr(pz) 2 ·0.7LiCl offers the possibility of making chemical adjustments to target specific structures and properties. Related frameworks with different metal ions, reduced aromatic linkers, terminal axial ligands, or pillaring axial ligands to link the 2D sheets along the c axis, may all be accessible through direct synthesis or post-synthetic substitution chemistry. In particular, the incorporation of metal ions with larger magnetic anisotropy could give rise to a new generation of ultrahard magnets. For instance, the spin-orbit coupling associated with lanthanide and low-coordinate transition metal ions imbues them with immense magnetic anisotropy and large magnetic moments, which are directly correlated to coercivity and remanence, respectively. Indeed, over the past two decades, research in the field of single-molecule magnetism has uncovered how to precisely manipulate the ligand field of certain lanthanide ions to give complexes with record molecular coercivities 20 , yet such molecules have not been chemically linked to give the strong coupling necessary for long-range magnetic order.
The discovery of room-temperature hard magnetism in a lowdensity material could promote several technological advances. For instance, lightweight hard magnets could replace dense rareearth magnets to improve energy efficiency in automotive and power conversion applications 3 . In addition, these magnets could find use in sensing media or even in magnetic gas separations, such as the separation of paramagnetic O 2 (S = 1) from diamagnetic N 2 (S = 0) 21 . Along these lines, upon reduction of CrCl 2 (pz) 2 to Cr(pz) 2 ·0.7LiCl, the 2D sheets shift from a staggered to an eclipsed conformation along the c axis, giving rise to tetragonal channels (Fig. 2). Moreover, heating a crystalline sample of THF-solvated Cr(pz) 2 ·0.7LiCl led to partial desolvation of THF with no loss of crystallinity and an increase in coercivity. These observations suggest that the partially or fully desolvated sample may show permanent porosity, wherein the framework exhibits measurable surface area upon desolvation, and it highlights the possibility of connecting the 2D sheets at fixed distances with pillaring ligands. The presence of stacked, neutral 2D sheets in Cr(pz) 2 ·0.7LiCl suggests that exfoliation may be readily accessible. Simple chemical methods, such as soaking Cr(pz) 2 ·0.7LiCl in a solution of a Li + -sequestering crown ether or cryptand ligand, may effect deintercalation of the LiCl sheets. Alternatively, chemical reduction of CrCl 2 (pz) 2 in the presence of a halide-abstracting agent may provide a route to directly access Cr(pz) 2 . Such exfoliation could be used to cast thin layers of the material onto substrates for numerous studies, potentially allowing for magnetism to be studied as a function of thickness and twist angle down to the monolayer limit using magneto-optical spectroscopy 22 . In general, the development of solutionprocessable layered magnets synthesized from the bottom up would represent an enormous scientific advancement, with direct utility in nanoscale information storage and other spintronics applications 23 .
Since the discovery of electron delocalization in the archetypal Creutz-Taube ion 24 , [(NH 3 ) 5 Ru(pz)Ru[(NH 3 ) 5 ] 5+ , the capacity of a bridging pyrazine ligand to mediate strong electronic coupling and delocalization within mixed-valence metal complexes has been widely studied 25 . In tetragonal MX 2 (pz) 2 frameworks, certain mixed-valence combinations of metal and linker could lead to both a high electronic conductivity and a high magnetic ordering temperature. Further, electron delocalization in a mixedvalence species with more than one unpaired spin can enforce exceptionally strong ferromagnetic alignment of spins via a double-exchange mechanism, analogous to conventional permanent magnets 26 . Indeed, such itinerant ferromagnetism was recently observed to promote magnetic order up to -48°C in a chromium(II/III) triazolate framework 27 . The parent material CrCl 2 (pz) 2 exhibits a high room-temperature conductivity of 32 mS cm -1 , which is attributed to electron delocalization  stemming from linker-based mixed valency. Subsequent reduction to Cr(pz) 2 ·0.7LiCl is associated with an approximately hundred-million-fold decrease in conductivity owing to the univalent metal and linker constituents in the latter, suggesting that this and related materials may serve as redox-or light-actuated magnetic semiconductors.