Resolving Crystallographic Disorder in Heterobimetallic Clusters via Resonant X-ray Diffraction

Abstract

Probing the chemical makeup of polynuclear reaction centers is central to understanding how composition imparts functionality. Important chemical machinery (e.g., biological inorganic cofactors, catalysts, energy storage materials) often possesses multi-metallic structural elements essential to driving their function. Introducing different metals to the overall composition enables substitutional disorder within the solid-state structure, convoluting the study of structure-function relationships. Resonant X-ray diffraction leverages the large changes to an atomic scattering factor that occur at spectroscopic absorption edges (K-, L-, or M-edge) to differentiate between two different elements, allowing for the resolution of crystallographic disorder stemming from the mixed occupancy of an atomic position within the crystalline lattice. The stepwise synthesis of heterobimetallic trinuclear clusters on the hexaanilide ligand platform, FtbsLH6 (1,3,5-C6H9(NC6H3-4-F-2-NSiMe2tBu)3), generates samples featuring crystallographic disorder that results in mixed-metal occupancies at each site within the trinuclear cores. The compositional homogeneity of these heterobimetallic clusters enables us to test the limitations of metal occupancy refinement via resonant X-ray diffraction. In Chapter 1, we introduce the theory and experimental protocols of a single-crystal resonant X-ray diffraction experiment. The classical derivation of the atomic scattering factor is provided along with key insights from quantum mechanics. Following this explanation, we review the information that single-crystal resonant X-ray diffraction experiments provide in the context of inorganic chemistry. Finally, the experimental design and data treatment procedures required to extract resonant scattering perturbations, f'(ω) and f''(ω), from single-crystal X-ray diffraction data are described. In Chapter 2, we examine the accuracy of determining metal site occupancies in a [Fe2Zn] cluster using resonant X-ray diffraction by directly comparing the result to that obtained from single-crystal neutron diffraction. These studies showcased that resonant X-ray experiments were as accurate at determining metal site occupancies in the heterobimetallic clusters as neutron diffraction. We additionally explored the intrinsic error of single-crystal resonant X-ray diffraction studies when theoretical and experimental f'(ω) references are employed, showcasing that the use of homometallic molecular control samples carries an inherent uncertainty of +/- 5%. In Chapter 3, we expand the synthesis of heterobimetallic clusters on the FtbsLH6 ligand platform towards [Zn2Ni] clusters. Reduction of the heterobimetallic (FtbsL)Zn2Ni(py) (3.3) cluster and subsequent reaction with the nitrene transfer reagent, N3Ad, generates the bridging μ3-nitrenoid adduct [K(THF)3][(FtbsL)Zn2Ni(μ3-NAd)] (3.5). This cluster can be further reduced by one electron, resulting in an isostructural dianionic cluster. The electronic configurations of these clusters were studied by SQUID magnetometry, EPR spectroscopy, and variable wavelength Ni K-edge XRF. These studies suggest a divalent oxidation state for the Ni centers in both the monoanionic and dianionic [Zn2Ni] nitrenoid complexes. In Chapter 4, we examine which factors (i.e., resolution, scaling methods, uncertainty propagation, structural modeling) impact the ability to refine resonant scattering perturbations from synchrotron-based diffraction data collected at energies along the Ni and Zn K-edges utilizing the [Zn2Ni] clusters described in Chapter 3. Although the resonant effect is maximized at an element’s absolute absorption edge, perturbations to X-ray scattering are still present when the incident radiation is off-resonance. We found that the closeness of Cu Kα radiation (8042 eV) to the K-edge absorption energy of Ni and Zn is sufficient to accurately refine resonant scattering perturbations. Remarkably, the resonant scattering terms refined from in-house generated Cu Kα X-ray diffraction data replicated occupancies determined via synchrotron-based resonant diffraction. In Chapter 5, the synthesis and structural characterization of several [Zn2Cr] clusters are explored. Reaction of FtbsLZn2Cr(py) (5.1) with aryl and alkyl azides generates bridging μ3-nitrenoid clusters. Depending on the nitrenoid substituent, these clusters display C1- or C3-molecular symmetry, enabling substitutional disorder within the trimetallic cores. As such, we turned towards two different methods to examine the solid-state Zn and Cr core occupancies: (1) structural refinement via ShelXL using EADP and SUMP constraints, and (2) refinement of the disordered sites’ resonant scattering perturbations when irradiated with Cu Kα radiation. Despite the energetic distance between the Cr K-edge (5990 eV) and Cu Kα X-ray radiation, the refined f'(Cu Kα) values for each core site yielded accurate occupancy estimates when compared against standard structural refinement. In Chapter 6, the electronic and structural characterization of [Fe2M] (where M = Co or Ni) clusters is explored via elastic and inelastic X-ray scattering. The resonant scattering perturbations of these clusters were refined from Cu Kα and variable wavelength synchrotron X-ray diffraction datasets and used to calculate metal-site occupancies. Additionally, experimental resonant scattering factors were derived from the [Fe3], [Co3], and [Ni3], homometallic analogs to provide reference f'(ω) and f''(ω) values for Fe, Co, and Ni in the occupancy calculations. In addition to the structural characterization conducted via elastic X-ray scattering experiments, we were also interested in obtaining information on the electronic configuration of the [Fe2Co] and [Fe2Ni] cores via inelastic spectroscopy. Variable wavelength XRF and 57Fe Mössbauer studies revealed that the nature of the electronic charge distribution within the trinuclear core is dependent on the identity of the metals residing within.