Nicholas F. Chilton

Large crystal-field anisotropy and strong exchange interactions are key ingredients in the design of next-generation molecule-based magnetic materials. Radical-bridged lanthanide complexes are particularly promising in this regard owing to their highly anisotropic ground states and strong exchange interactions mediated by the radical bridge. Here, we present a comprehensive ab initio investigation of the magnetic and exchange anisotropy in the {[((Me3Si)2N)2(THF)Ln]2(N2)}− complexes, where Ln = Gd3+, Tb3+, Dy3+, Ho3+, and Er3+. For the isotropic gadolinium (Ln = Gd3+) analogue, the microscopic origins of the exchange interaction are disentangled using density functional theory calculations, revealing significant contributions from both direct and kinetic exchange mechanisms, with the latter dominating due to efficient virtual Gd-to-radical electron transfer involving out-of-plane Gd 4f orbitals and the radical π* orbital. For the anisotropic complexes (Ln = Tb3+, Dy3+, Ho3+ and Er3+), crystal-field and exchange interactions are described using irreducible tensor-operator model Hamiltonians parametrized from multiconfigurational ab initio calculations. The resulting framework employs a reduced set of physically transparent parameters, facilitating systematic comparison across the lanthanide series. While the crystal field potential is largely constant across the series, the exchange interactions vary significantly.

The success of lanthanide-based single-molecule magnets (SMMs) has not yet been matched by their actinide counterparts, despite large calculated barriers to magnetic relaxation. Existing studies often conclude that standard multiconfigurational methods cannot accurately predict the magnetic relaxation barrier, but these models represent only a static picture of magnetic relaxation. Here, we perform the first ab initio calculations of spin-phonon coupling and magnetic relaxation in actinide SMMs, studying [UV(TrenTIPS)(E)] (TrenTIPS = N(CH2CH2NSiPr3i)3), with the terminal atom systematically varied across E = N, O, F, P, S, Cl, Br. We find that changing E strongly modifies U-E bonding, crystal field splitting, and wave function composition; however, these static descriptors do not predict low-temperature magnetic relaxation. All complexes relax through the two-phonon Raman mechanism, with low-energy phonon pathways revealed through the Raman rate integrand. The [UV(TrenTIPS)(F)] complex relaxes faster despite a large excited state energy gap, while [UV(TrenTIPS)(P)] shows that the overall relaxation rate reflects the combined contributions from the Raman rate integrand. Our results demonstrate that Ueff is not an appropriate metric for assessing these systems and that spin-phonon coupling and phonon availability, rather than static electronic structure descriptors alone, are central to understanding magnetic relaxation in this family of actinide complexes.

Controlling terahertz (THz) waves is important for next‐generation communications, sensing, and imaging technologies. However, it is still unclear precisely how THz signals can be manipulated by various materials. Herein, we report the design and synthesis of four coordination networks, {[Fe ₂ (Xpz) ₂ ][Hg(SCN) ₄ ] ₂ [Hg(SCN) ₂ ]} _n (Xpz = fluo140ropyrazine, 1‐F ; Xpz = chloropyrazine, 1‐Cl ; Xpz = bromopyrazine, 1‐Br ) and {[Fe(Ipz)][Hg(SCN) ₄ ]} _n (Ipz = iodopyrazine, 1‐I ) as tunable THz absorbers. 1‐Cl and 1‐Br form isostructural three‐dimensional frameworks closely related to 1‐F , differentiated in the Xpz ligand orientation. In contrast, 1‐I adopts a distinct two‐dimensional framework due to the more sterically demanding halogen. Clear differences are observed in the sub‐THz (< 1 THz) absorption features between the four materials, with the lowest features shifting to lower frequencies from 1‐F to 1‐Br . As these materials contain open‐shell high‐spin Fe(II) ions, we show that the application of a magnetic field can precisely modulate low‐frequency (LF) Raman bands by ca. 0.06 THz/T at 4 K in the THz range. We show using Raman magnetospectroscopy (RAMS) that this tunability arises from the zero‐field splitting (ZFS) of Fe(II).

Controlling terahertz (THz) waves is important for next‐generation communications, sensing, and imaging technologies. However, it is still unclear precisely how THz signals can be manipulated by various materials. Herein, we report the design and synthesis of four coordination networks, {[Fe ₂ (Xpz) ₂ ][Hg(SCN) ₄ ] ₂ [Hg(SCN) ₂ ]} _n (Xpz = fluo140ropyrazine, 1‐F ; Xpz = chloropyrazine, 1‐Cl ; Xpz = bromopyrazine, 1‐Br ) and {[Fe(Ipz)][Hg(SCN) ₄ ]} _n (Ipz = iodopyrazine, 1‐I ) as tunable THz absorbers. 1‐Cl and 1‐Br form isostructural three‐dimensional frameworks closely related to 1‐F , differentiated in the Xpz ligand orientation. In contrast, 1‐I adopts a distinct two‐dimensional framework due to the more sterically demanding halogen. Clear differences are observed in the sub‐THz (< 1 THz) absorption features between the four materials, with the lowest features shifting to lower frequencies from 1‐F to 1‐Br . As these materials contain open‐shell high‐spin Fe(II) ions, we show that the application of a magnetic field can precisely modulate low‐frequency (LF) Raman bands by ca. 0.06 THz/T at 4 K in the THz range. We show using Raman magnetospectroscopy (RAMS) that this tunability arises from the zero‐field splitting (ZFS) of Fe(II).

The most fundamental and widely studied form of metal-ion complexation is hydration. Nevertheless, the hydration of the heaviest elements with experimentally accessible bulk properties, namely americium, curium, berkelium, and californium, remains a subject of ongoing debate. Herein, the solid-state series of nona-aqua complexes from uranium to californium is completed through the single-crystal structure characterization of [An(H2O)9](CF3SO3)3 (An3+ = Bk3+ and Cf3+) at 100 K. In addition, the largely unexplored solid-state octa-aqua series, [An(H2O)8](Hdtp)(dtp)·H2O (An3+ = Am3+, Cm3+, Bk3+ and Cf3+, H2dtp = 2,3-di(tetrazol-5-yl)pyrazine) is synthesized and comprehensively characterized to elucidate how variations in hydration manifest in electronic spectra and structural metrics. By combining crystallographic and spectroscopic data with ab initio multiconfigurational wave function calculations, we demonstrate that differences between the two hydration motifs give rise to distinct trends in bond lengths, spectral features, and crystal-field strengths. Although the average An–OH2 bond lengths decrease with similar slopes in both [An(H2O)8]3+ and [An(H2O)9]3+ from Am3+ to Cf3+, the trends in individual An–OH2 bond length contractions differ between the two series. Remarkably, the crystal-field splitting in the octa-aqua complexes can be up to twice that observed in the nona-aqua complexes.

Understanding vibronic coupling in molecular materials is crucial in the development of high performance molecule-based optical and quantum technologies. Vibronic coupling governs spin state lifetimes in these systems, and is additionally a source of decoherence. Hence, gaining detailed understanding of vibronic coupling will enable improvement of candidate materials. In this regard, magneto-spectroscopic techniques provide a direct probe of the vibronic coupling, which can be performed on polycrystalline and more recently single crystal samples. Single crystal magneto-spectroscopy offers a clear advantage due to its improved resolution and its ability to probe magnetic anisotropy, however, simulating the resulting spectra has proven difficult due to the simultaneous requirements of solid-state phonon calculations, embedded correlated electronic structure theory methods, and inclusion of large numbers of vibrational modes in the Hamiltonian to recreate experimental spectra. In this work we present single crystal far-infrared magneto-spectroscopy (FIRMS) and photo-luminescence magneto-spectroscopy (PLUMS) measurements on [Yb(trensal)] . We can experimentally verify previous assertions that most of the vibronic transitions in the FIRMS map are intra-Kramers doublet transitions and additionally elucidate the origin of the transitions observed in the PLUMS map. Crucially, to achieve this we have developed a new computational framework to perform robust and efficient simulations of magneto-spectroscopy incorporating hundreds of vibrational modes based on ab initio simulations in the solid state.

Molecules showing magnetic memory effects at high temperatures could be used in high-density data storage devices. A magnetic hysteresis temperature ( T _H ) of 100 K has been achieved for a near-linear dysprosium bis(amide)-alkene single-molecule magnet (SMM), and a bent dysprosium cyclopentadienyl-amide SMM has shown T _H = 73 K. However, both these complexes exhibit fast magnetic relaxation at lower temperatures compared to dysprosium bis(cyclopentadienyl) SMMs. Here we report the synthesis of two highly axial dysprosium cyclopentadienyl-amide SMMs, [Dy{N(Si ⁱ Pr ₃ ) ₂ }(Cp ^R )][Al{OC(CF ₃ ) ₃ } ₄ ] ( 1-Dy , Cp ^R = Cp ^ttt , C ₅ H ₂ ^t Bu ₃ −1,2,4; 2-Dy , Cp ^R = Cpʹʹʹ, C ₅ H ₂ (SiMe ₃ ) ₃ −1,2,4). We find T _H = 91 K for 1-Dy and 92 K for 2-Dy ; ab initio calculations on 1-Dy show that magnetic relaxation is controlled by a combination of short dysprosium-ligand distances, a near-linear N–Dy–Cpʹʹʹ _centroid angle, and constrained vibrational modes. Together, this work demonstrates that dysprosium SMMs with large T _H values can be achieved with a judicious combination of π-aromatic and σ-donor ligands.

Here, we investigate the magnetic exchange coupling and magnetic anisotropy of a Co(II)–Cu(II) complex using ab initio multiconfigurational calculations. This is one of the simplest cases of magnetic exchange coupling involving an orbitally degenerate metal center: a pseudo-octahedral 3d7 Co(II) ion and a spin-only 3d9 Cu(II) ion. Electronic structure calculations were performed using CASSCF and CASPT2/NEVPT2 methods, and the resulting electronic states were further analyzed via model Hamiltonians to extract physically meaningful parameters, including crystal field splitting, spin–orbit coupling, and exchange coupling. The CASPT2/NEVPT2 calculations improved the agreement with experiment compared to that of CASSCF by correcting the overestimation of ferromagnetic coupling. Our calculations highlight that spin–orbit exchange coupling (i.e., spin-other-orbit coupling) is just as important as spin–spin exchange coupling for an accurate description of exchange coupling.

Magnetic exchange coupling between high-anisotropy spin centers is a key requirement for single-molecule magnets with high operating temperatures. Here, we report the synthesis and characterization of the dinuclear iron(II) and cobalt(II) complexes [(iPr3Si)(Dipp)NFe]2(m-(NSiMe3)2C6Me3H) (m-Fe2), [(iPr3Si)(Dipp)NFe]2(p-(NSiMe3)2C6Me4) (p-Fe2), and [(iPr3Si)(Dipp)NCo]2(p-(NSiMe3)2C6Me4) (p-Co2), with each metal center residing in a linear, two-coordinate geometry. In these complexes, the two high-magnetic-anisotropy metal centers are linked through a meta- or para-bis(amido)arene, which mediates ferro- or antiferromagnetic superexchange, respectively. Variable-temperature magnetic susceptibility data reveal antiferromagnetic coupling in the para-substituted complexes, giving S = 0 ground states with exchange constants of −3.56(5) cm–1 for p-Fe2 and −9.82(15) cm–1 for p-Co2. In stark contrast, the meta-linker in m-Fe2 mediates weak ferromagnetic exchange (+0.44(1) cm–1), leading to an S = 4 ground state. Here, the coexistence of ferromagnetic coupling and strong single-ion anisotropy gives rise to single-molecule-magnet behavior with a relaxation barrier of 139 cm–1, the highest yet reported for a multinuclear transition metal complex, and the coupling markedly suppresses the fast through-barrier quantum relaxation that plagues related mononuclear species. To gain a microscopic understanding of the magnetic exchange in this family of complexes, we employed a combination of far-infrared magnetospectroscopy and computation. While a spin-only model reproduces the spectroscopic energy levels well, the substantial unquenched orbital angular momentum at the FeII and CoIIcenters inspired a fully ab initio treatment of the electronic structure, crystal-field splitting, and exchange constants. These calculations indicate that both the sign and the magnitude of the exchange constant depend strongly on the linker geometry.

Divalent europium formamidinates [Eu(DFForm) ₂ (py) ₃ ] and [Eu(DippForm) ₂ (py) ₃ ]·3py were synthesized by the direct reaction of europium metal with N , N ′-bis(2,6-difluorophenyl)formamidine or N , N ′-bis(2,6-diisopropylphenyl)formamidine in pyridine.