Large Differences in the Optical Spectrum Associated with the Same Complex: The Effect of the Anisotropy of the Embedding Lattice

J. A. Aramburu, Pablo García‐Fernández, J. M. García‐Lastra, M. MorenoInorganic 2017 · DOI

TL;DR

This paper challenges the assumption that transition metal complexes behave identically regardless of their host lattice. By studying copper fluoride complexes, the authors demonstrate that the anisotropy of the surrounding crystal lattice, specifically the internal electric field it generates, can drastically alter the complex's optical properties. This finding is crucial for understanding and tuning the behavior of these materials.

What Problem Does This Paper Address?

The prevailing view in solid-state chemistry is that a transition metal complex, once formed, possesses intrinsic properties largely independent of its environment. This "complex as a well-defined unit" concept works well when the host lattices are similar (isomorphous). However, this paper tackles the question of what happens when the host lattices are not isomorphous, and how this difference in the surrounding crystal structure impacts the complex's optical behavior. Specifically, the authors investigate why two systems containing the same square-planar CuF<sub>4</sub><sup>2-</sup> complex exhibit significantly different d-d optical transition energies, even when the copper-fluorine bond lengths are nearly identical. This discrepancy suggests that something beyond the immediate coordination sphere is at play.

Methods

The researchers employed a combination of computational and theoretical approaches:

  • Density Functional Theory (DFT): Periodic DFT calculations were used for geometry optimizations of the CaCuF<sub>4</sub> and Ba<sub>2</sub>ZnF<sub>6</sub>:Cu<sup>2+</sup> systems. This standard computational method allows for the accurate determination of atomic positions and electronic structures.
  • Embedded Cluster Approach: To calculate the d-d electronic transitions, a minimalist embedded cluster model was used. This involved treating the CuF<sub>4</sub><sup>2-</sup> complex as a central unit surrounded by point charges representing the rest of the lattice ions. This approach is designed to isolate and highlight the effect of the internal electric field (ER(r)) generated by the surrounding lattice.
  • Δ-SCF Procedure: The Amsterdam Density Functional (ADF) code was used with a Δ-SCF (state-averaged self-consistent field) procedure to obtain excitation energies for specific electronic configurations. This is a standard method for calculating optical transition energies within DFT.

The novelty lies in the explicit focus on the anisotropic internal electric field (ER(r)) as the primary driver of the observed spectral differences, rather than solely relying on metal-ligand bond lengths or standard crystal field theory.

Datasets

Not discussed in the paper. The study relies on experimental data from existing literature for comparison.

Key Insights and Ideas

  • Lattice Anisotropy is Key: The most significant insight is that the anisotropy of the host lattice's internal electric field (ER(r)) is a critical factor in determining the optical properties of transition metal complexes, even when the complexes themselves are structurally similar.
  • "Isotropic" Lattices Aren't Always Isotropic: Even cubic lattices, often considered isotropic, can exert an anisotropic electric field on embedded complexes, leading to observable effects on their electronic states.
  • Beyond Bond Lengths: The study demonstrates that differences in optical spectra cannot always be explained by subtle variations in metal-ligand bond lengths. The surrounding environment's electric field plays a dominant role.
  • Red vs. Blue Shifts: The anisotropic field can either red-shift (lower energy) or blue-shift (higher energy) the d-d transitions, depending on the specific geometry and charge distribution of the host lattice. CaCuF<sub>4</sub> exhibits red-shifts, while Ba<sub>2</sub>ZnF<sub>6</sub>:Cu<sup>2+</sup> shows blue-shifts.
  • Minimalist Model Power: The study highlights the effectiveness of a simplified model (a CuF<sub>4</sub><sup>2-</sup> cluster with point charges) in capturing complex phenomena, emphasizing the importance of the internal electric field.

Key Results

  • CaCuF<sub>4</sub>: The d-d transitions in CaCuF<sub>4</sub> are red-shifted by the anisotropic internal electric field, bringing the calculated energies closer to experimental values (e.g., the a<sub>1g</sub>(3z<sup>2</sup>-r<sup>2</sup>) → b<sub>1g</sub>(x<sup>2</sup>-y<sup>2</sup>) transition shifts from 2.58 eV for an isolated complex to 1.84 eV when the field is included, compared to an experimental value of 1.71 eV).
  • Ba<sub>2</sub>ZnF<sub>6</sub>:Cu<sup>2+</sup>: In contrast, the internal electric field in Ba<sub>2</sub>ZnF<sub>6</sub>:Cu<sup>2+</sup> blue-shifts the d-d transitions. The calculated energies for the isolated complex are lower than experimental values, but including the internal field brings them closer (e.g., the a<sub>1g</sub>(3z<sup>2</sup>-r<sup>2</sup>) → b<sub>1g</sub>(x<sup>2</sup>-y<sup>2</sup>) transition shifts from 2.61 eV to 2.73 eV, compared to an experimental shoulder around 2.98 eV).
  • Magnitude of Difference: The spectral differences between CaCuF<sub>4</sub> and Ba<sub>2</sub>ZnF<sub>6</sub>:Cu<sup>2+</sup> can be as large as 1.27 eV, which is substantial for d-d transitions.
  • Charge Transfer vs. d-d: The study confirms that the observed transitions in Ba<sub>2</sub>ZnF<sub>6</sub>:Cu<sup>2+</sup> are indeed d-d transitions, not charge transfer transitions as previously suggested.

Limitations and Future Work

  • Model Simplification: While the minimalist model is effective, it relies on point charges to represent the lattice ions. Future work could explore more sophisticated embedding schemes.
  • Broader Applicability: The authors suggest that this understanding of lattice anisotropy can be applied to a wide range of transition metal complexes, particularly those in non-cubic host lattices, and could help explain phenomena currently attributed solely to Jahn-Teller effects or requiring empirical parameters.
  • Tuning Properties: The knowledge gained opens avenues for chemically tuning the optical properties of these materials by designing host lattices with specific anisotropic electric fields.
  • d<sup>0</sup> Cations: Further work is underway to explore the electronic and structural properties of compounds containing d<sup>0</sup> cations.

Summary updated

Abstract

Transition-metal complexes with a well-defined geometry are usually considered to display almost the same properties independently of the system where they are embedded. Here we show that the above statement is not true depending on the anisotropy of the host lattice, which is revealed in the form of the electric field created by the rest of lattice ions over the complex. To illustrate this concept we analyze the origin of the surprisingly large differences in the d-d optical transitions of two systems containing square-planar CuF<sub>4</sub><sup>2-</sup> complexes, CaCuF<sub>4</sub>, and center II in Cu<sup>2+</sup>-doped Ba<sub>2</sub>ZnF<sub>6</sub>, even though the Cu<sup>2+</sup>-F<sup>-</sup>distance difference is just found to be 1%. Using a minimalist first-principles model we show that the different morphology of the host lattices creates an anisotropic field that red-shifts the in vacuo complex transitions to the 1.25-1.70 eV range in CaCuF<sub>4</sub>, while it blue-shifts them to the 1.70-3.0 eV region in Ba<sub>2</sub>ZnF<sub>6</sub>:Cu<sup>2+</sup>. This particular example shows how the lattice anisotropy strongly alters the optical properties of a given transition-metal complex. This knowledge opens a new path to tune the spectra of this large family of systems.
BibTeX
@inproceedings{aramburu2017large,
  author = {J. A. Aramburu and Pablo García‐Fernández and J. M. García‐Lastra and M. Moreno},
  title = {{Large Differences in the Optical Spectrum Associated with the Same Complex: The Effect of the Anisotropy of the Embedding Lattice}},
  booktitle = {Inorganic},
  year = {2017},
  doi = {10.1021/acs.inorgchem.7b00932},
}
Figure