A Multiconfigurational Treatment of the |L2,3| XAS in |Ni2+| Compounds
======================================================================

The following tutorial illustrates the limitations of crystal field multiplet
theory in reproducing some of the spectral features in a series of nickel
compounds, and introduces the more elaborate charge transfer multiplet model,
that can be used to overcome these limitations.

Crystal Field Multiplet
-----------------------
1. Calculate the 2p XAS spectrum of |Ni2+| using the default parameters, but
increase the Gaussian broadening to 0.4. Identify the |L2| and |L3| edges in
the calculated spectrum.

2. Change the default scaling of the Slater integrals (κ) to 0.6, 0.4, and 0.2
and rerun the calculation. Overlay the four spectra and describe the evolution
of the main peak and its high energy shoulder when the scaling parameter is
decreased.

3. Compare the previous observation with the changes in the experimental
spectra when going from fluoride to bromide in the series of nickel halides.
How do the reduction factor and covalency of the metal-ligand bond correlate,
knowing that the later increases with an increasing atomic mass of the halide?

.. figure:: assets/laan_fig1.png
    :width: 60 %
    :align: center

    van der Laan et al., J. Phys. Rev. B, 1986, 33 (6), 4253--4263.

4. For what reduction factor do you get the best agreement with the
experimental spectrum of |NiBr2|? Is it reasonable to have to use this scaling
factor?

5. Is there a particular region of the spectrum that doesn't seem to be
reproduced using the previous crystal field multiplet calculations?

Charge Transfer Multiplet (CTM)
-------------------------------
1. Run a calculation using the following parameters: κ = 0.9, U(3d,3d) = 7.5
eV, U(2p,3d) = 8.5 eV (*Atomic*), 10Dq(3d) = 0.7 eV (*Crystal Field*), Δ(3d,Ld)
= 4.3 eV, Veg(3d,Ld) = 2.0 eV, Vt2g(3d,Ld) = 1.0 eV (*3d-Ligands
Hybridization*). How does the calculated spectrum compare with the measured
spectrum of NiO?

2. Repeat the above calculation while varying Δ between 0.0 and 10.0 eV. Notice
the changes in the number of the metal 3d and the ligand electrons (<N_3d> and
<N_Ld> in the logging window). What happens if Δ is negative?

3. Try to get a better agreement with the experimental spectrum of NiO by
varying the crystal field parameters, 10Dq(3d) and 10Dq(Ld), and the hopping
integrals, Veg(3d,Ld) and Vt2g(3d,Ld).

.. |L2,3| replace:: L\ :sub:`2,3`\
.. |Ni2+| replace:: Ni\ :sup:`2+`\
.. |L2| replace:: L\ :sub:`2`\
.. |L3| replace:: L\ :sub:`3`\
.. |NiBr2| replace:: NiBr\ :sub:`2`\