UC Berkeley

Contained in the following dissertation are detailed investigations of the syntheses, structures, and magnetic properties of a series of paramagnetic molecules. A wide range of magnetic behavior is reported, including interionic magnetic exchange coupling, magnetic anisotropy, molecular magnetic relaxation mechanisms, and ultimately their respective origins on the basis of fundamental concepts in coordination chemistry. Chapter One provides a brief background of the origins of single-molecule magnet behavior with a review of basic molecular magnetism concepts and techniques and is geared toward the comprehension level of a first-year graduate student fluent in the basics of coordination chemistry.

In Chapter Two, the syntheses, structures, and magnetic properties of pentanuclear cyanide-bridged clusters are presented. The complexes are based upon the highly anisotropic building unit [Re(CN)7]4-, which possesses an S = 1/2 ReIV ion with an associated unquenched orbital angular momentum. Further, the clusters produce a variety of magnetic behaviors, including NiII···ReIV ferromagnetic interactions in [(PY5Me2)4Ni4Re(CN)7](PF6)5 and intermolecular CoIII←ReIII charge transfer in [(PY5Me2)4Co4Re(CN)7](PF6)5. Notably, [(PY5Me2)4Ni4Re(CN)7]5+ is a single molecule magnet that is stable at room temperature in contrast to [(PY5Me2)4Mn4Re(CN)7]5+, which is spontaneously reduced in solution at room temperature to [(PY5Me2)4Mn4Re(CN)7]4+, the latter of which contains a diamagnetic ReIII ion. Like this Mn-containing cluster, however, the complex [(PY5Me2)4Cu4Re(CN)7]5+ is temperature sensitive and is reduced in solution at room temperature to provide [(PY5Me2)Cu4Re(CN)7]4+. In the former of these two species, ferromagnetic CuII···ReIV interactions are observed in addition to slow magnetic relaxation, while the magnetic behavior of the latter complex is consistent with four noninteracting CuII centers. The observation of slow magnetic relaxation in [(PY5Me2)Cu4Re(CN)7]4+ thus categorizes this molecule as a redox-switchable single-molecule magnet.

Chapters Three through Seven are shifted in focus to the investigation of the magnetic properties of complexes with only one paramagnetic ion, for reasons detailed in Chapter One. Chapter Three describes the synthesis and characterization of a pseudotetrahedral complex of cobalt(II) and the tri-dentate, nitrogen-donor ligand 1,1,1-tris[2N-(1,1,3,3-tetramethylguanidino)methyl]ethane (3G). Up until this point, most of the literature for single-molecule magnetism focuses on the fact that a negative zero-field splitting value is essential for the observation of slow magnetic relaxation. In this chapter, however, slow magnetic relaxation under applied dc field by the complex [(3G)CoCl]+ is presented, despite being shown by HF-EPR to possess a positive zero-field splitting. The reported results importantly refute the idea that slow magnetic relaxation is only possible in systems with negative axial anisotropy, but further suggest that systems with easy plane magnetic anisotropy need not be omitted from the search for large magnetic relaxation barriers.

Chapter Four details the first observation of slow magnetic relaxation in the mononuclear transition metal complex [Co(SPh)4]2- without the typical required static dc field, providing a Ueff of 21(1) cm-1. The observation of slow magnetic relaxation at zero dc field for this species is attributed to both the large, negative, zero-field splitting, as well as the half-integer spin. All previous accounts of slow magnetic relaxation in mononuclear complexes invoked nonzero transverse anisotropies (E) as the source of fast zero field tunneling, as this parameter leads to mixing of the bistable ±MS levels. However, by Kramers' theorem, E should not mix the bistable MS = ±3/2 levels in [Co(SPh)4]2-, and indeed it does not, as evidenced by the slow magnetic relaxation detected at zero applied dc field. Notably, a lingering influence of tunneling is still detected for this molecule, and dilution studies show that the mechanism of tunneling at low temperature is intermolecular in origin.

The large zero-field splitting (D) in [Co(SPh)4]2- is likely attributed to the nearly degenerate dxy and dx2-y2 orbitals, which provide a low-lying excited state with a large negative contribution to D. As detailed in Chapter Five, the preparation and characterization of the series of complexes [Co(EPh)4]2- (E = O, S, Se) were performed to see how a simple modification of the ligand field would influence D and Ueff. Of note were the isolation and structural characterization of a relatively simple molecule, [Co(OPh)4]2-, a moiety which was hitherto unknown in the literature as a mononuclear species. The D values proved highly influenced by the ligand field, yet Ueff did not, in contrast to the expectation that the barrier to spin reversal should directly scale with D. Also discussed is a detailed analysis of the diffuse reflectance spectra for the series of complexes, which was pursued with the intent of gaining a deeper understanding of the mechanisms by which the ligand field adjusts both the magnetic anisotropy and the low temperature magnetization dynamics.

Chapter Six describes the magnetic investigation of two-coordinate, linear complexes of iron(II). Unlike the [Co(EPh)4]2- species in Chapters Four and Five, the d-orbital splitting diagrams for two-coordinate transition metal ions in linear geometries feature degenerate (or as close as reasonably possible) dxy and dx2-y2 orbitals, which are predominately nonbonding in character. These complexes display magnetic relaxation barriers up to 181 cm-1 under applied dc field and fast tunneling of the magnetization at zero applied dc field. Ab initio studies, initiated to elucidate the influence of the ligands on the slow magnetic relaxation, revealed that though the dxy and dx2-y2 orbitals were not rigorously degenerate (due to molecular symmetries lower than D∞h), the spin-orbit coupling was strong enough to engender electronic ground states corresponding to orbital angular momentum that is essentially unquenched by the ligand field. Thus, these complexes contain transition metal ions that magnetically mimic lanthanide ions, where the spin-orbit coupling is strong enough to override the quenching influence of the very weak ligand-field felt by the spin-bearing 4f orbitals. The calculated barriers are close to the experimental ones for Fe[N(SiMe3)(Dipp)]2 and Fe[C(SiMe3)3]2, but differ strongly for the other investigated complexes, the reasons for which are attributed to a possible vibronic model.

Chapter Seven meshes the ideas presented in Chapters Four through Six while describing the magnetic properties of a two-coordinate, linear complex of iron(I). This species, obtained by a one-electron reduction of the iron(II) complex Fe[C(SiMe3)3]2, possesses both a half-integer spin and unquenched orbital angular momentum. As a result of these two characteristics, [Fe(C(SiMe3)3)2]- displays the largest spin reversal barrier yet observed for all transition-metal-containing species (226(4) cm-1) by an order of magnitude. As a consequence of such a large magnetization reversal barrier, [Fe(C(SiMe3)3)2]- possesses a molecular magnetic moment that is blocked from reversal at low temperature, giving rise to magnetic hysteresis, a phenomenon typically associated with bulk magnetic materials, but here due to an individual transition metal ion.