UC Berkeley

This dissertation describes the research that led to the discovery of single-molecule magnetism in the actinides. Chapter One is an introduction to the concepts that lead to single-molecule magnet behavior with an emphasis on the specific qualities of the f-elements that make them interesting for such studies. A simple model for predicting ligand field environments that should be amenable to single-molecule magnet behavior is presented along with several examples of its application to lanthanide and actinide systems. The study of magnetic exchange coupling in uranium-containing multinuclear complexes is discussed and the literature on the subject is reviewed.

Chapter Two describes how the homoleptic dimer complex [U(Me2Pz)4]2 (Me2Pz- = 3,5-dimethylpyrazolate) can be cleaved via insertion of terminal chloride ligands, such that reactions with (cyclam)MCl2 (M = Ni, Cu, Zn; cyclam = 1,4,8,11-tetraazacyclotetradecane) in dichloromethane generate the linear, chloride-bridged clusters (cyclam)M[(μ-Cl)U(Me2Pz)4]2. Variable-temperature magnetic susceptometry is used to reveal the presence of weak ferromagnetic coupling between the Ni(II) (S = 1) and U(IV) centers and no coupling between the Cu(II) (S = 1/2) and U(IV) centers. Consistent with a simple superexchange mechanism for the coupling, density functional theory calculations performed on a [(Me2Pz)4UCl]⁻ fragment of the cluster show the spin resides in 5fxyz and 5fz(x2-y2) orbitals, exhibiting delta symmetry with respect to the U-Cl bond.

Chapter Three extends the analysis of exchange coupling in Chapter Two to include the (cyclam)Co[(μ-Cl)U(Me2Pz)4]2 cluster. As in the Cu(II) case, Co(II) has a single unpaired electron (S = 1/2), however this unpaired electron resides in a dz2 orbital and is therefore oriented directly along the superexchange pathway. This provides a significantly better magnetic exchange pathway leading to the strongest magnetic coupling of the series.

Chapter Four deviates briefly from the pursuit of molecular magnets to study a series of multinuclear clusters formed from the activation of the 3,5-dimethylpyrazolate anion by uranium(III) via two-electron reductive cleavage of the N-N bond to form 4-ketimidopent-2-ene-2-imido (kipi3-) units, as isolated in three related tetranuclear uranium cluster compounds, two of which are mixed valent. The kipi3- ligand represents an exotic latecomer to the acetylacetonato (acac-) ligand family. Unlike the related and widely-utilized β-diketimido (nacnac-) ligands, kipi3- can be represented as containing both imido and ketimido functionalities. Thus, it provides a true nitrogen-based, isoelectronic analogue of acac-, a ligand that has played a long and vital role in coordination chemistry.

Chapter Five turns from the synthesis of exchange coupled clusters to mononuclear species. Drawing on the model of f-element anisotropy presented in Chapter One, the trigonal prismatic complex U(Ph2BPz2)3 was chosen for study. Ac magnetic susceptibility measurements performed on it demonstrate the presence of slow magnetic relaxation under zero applied dc field. Analysis of both the temperature and frequency dependence of the ac susceptibility indicate a temperature regime (T > ~3 K) where Arrhenius behavior dominates the relaxation processes, leading to a spin relaxation barrier of Ueff = 20 cm−1. The dc field dependence of the relaxation time is studied to reveal evidence of quantum tunneling processes occurring at lower temperatures. The results represent the first example of an actinide complex displaying single-molecule magnet behavior and confirm the general strategy for identifying further uranium(III)-based single-molecule magnets by concentrating ligand field contributions above and below the equatorial plane of an axially-symmetric coordination complex.

Chapter Six builds on the results presented in Chapter Five to characterize the related complex the trigonal prismatic complex U(H2BPz2)3. This tricapped trigonal prismatic complex is characterized by single crystal x-ray diffraction and ac magnetic susceptibility measurements. The ac susceptibility data demonstrate the presence of multiple processes responsible for slow magnetic relaxation. Out-of-phase signals observed at ac switching frequencies between 1 and 1500 Hz in dc fields of 500-5000 Oe indicate a thermal relaxation barrier of ca. 8 cm-1 for the molecule, with a temperature-independent process taking over at the lowest temperatures probed. Significantly, an unprecedented, slower relaxation process becomes apparent for ac switching frequencies between 0.06 and 1 Hz, for which a monotonic increase of the relaxation time with applied dc field suggests a direct relaxation pathway.