|
|
Spectroscopies and simulations to understand materials
properties
|
|
|
People involved in the subject:
Florent BOUCHER (CR), Nicolas DUPRE (CR), Christophe PAYEN
(Pr), Michaël PARIS (IR), Lionel TRUFLANDIER (Thèse
2007)
|
|
|
Overview
|
In the field of materials science,
it is of paramount importance to achieve a chemical and
structural characterization as complete as possible before
studying the evolution of physico-chemical properties
as functions of various parameters such as composition,
temperature, pressure, oxidation state. XRD is the reference
tool that comes to mind for structural characterization.
However, when crystalline domains shrink (nanomaterials)
or in the case of disordered materials (batteries materials),
limitations are reached as far as this technique is concerned.
It becomes then very interesting to use local or selective
probes such as XAS, EELS, RAMAN, IR or NMR. In order to
widen its range of characterization methods, IMN acquired
few years back a 500 MHz Bruker NMR spectrometer and MAS
probes. This spectroscopy is an extremely selective probe,
very sensitive to the evolutions of atomic local environments:
site distortions, changes in bonds length, evolution of
the iono-covalent character of interatomic forces. Moreover,
thanks to its local character, solid state NMR can be
applied to crystallized materials as well as glasses or
amorphous compounds.
Still, NMR spectra interpretation is a delicate step and,
in numerous cases, the assignments of resonances cannot
be done without the help of simulation methods. By setting
tools allowing the simulation of NMR parameters, our aim
is to contribute to the development of this technique
and to shed light on the complementarity simulation/NMR
experiment.
Prior to the work done by Pickard and Mauri [1] who introduced
the GIPAW (Gauge-Including Projector Augmented-Wave) approach,
NMR parameters simulation on periodic systems was performed
using "clusters" approach, inappropriate for
an meaningful description of solids. The GIPAW concept,
now implemented in the NMR CASTEP code, permits this type
of simulations for complex compounds with several hundreds
of atoms per lattice. Nevertheless, up to now, no GIPAW
simulation had been reported for transition metals of
the first period, such as titanium or vanadium, interesting
for various applications: pigments, catalysis, photovoltaic,
lithium batteries electrode...Thus it seemed important
to validate the GIPAW approach for transition metals.
Within the Ph.D research project carried by Lionel Truflandier
(2004/2007), we focused on the simulation of 51V NMR parameters,
namely chemical shift anisotropy (CSA) and electrical
field gradient (EFG) tensors.
- (1) C. J. Pickard and F. Mauri Phys. Rev. B (2001), 63, 245101
|
| Results |
|
The first studies introducing
chemical shift calculations for 3d metals using the GIPAW
method have just been published on the AlVO4 compound, giving
new insights on its structure. Recent EFG calculations performed
on this compound allowed the assignments of the NMR resonances
to the three distinct vanadium sites found in the structure.
Moreover, our work, yielding CSA tensor parameters, permitted
first, to confirm this attribution and second, to obtain
accurate values of Euler angles describing the relative
orientation of the two CSA and EFG tensors.
L. Truflandier, M. Paris, C. Payen and F. Boucher J. Phys. Chem. B (2006), 110, 21403-21407. |
Figure 1: Experimental and theoretical
eigenvalues obtained by the GIPAW method for the CSA tensor
of the three vanadium sites in AlVO4
|
Figure 2 Correlation between isotropic
values of the shielding tensor and the chemical shift for
different 51V sites found in numerous molecular extended
systems.
|
Before the finalization of these first
studies, a comprehensive work was however, necessary from
the building of an efficient pseudo-potential for the vanadium
atom to its validation in the case of chemical shift calculation
and finally to its application on numerous molecular extended
systems. This body of work can be found in PRB (ref http://link.aps.org/abstract/PRB/v76/e035102).
We show that it is possible to predict the isotropic chemical
shift for 51V with an error margin within 30 ppm for a chemical
shift range wider than 800 ppm. Concerning the assignment
of vanadium resonances in a particular structure, acceptable
error margins would have to be narrowed, with an order of
magnitude smaller than 10 ppm.
Based on these results, we started a collaboration with
Nathalie Steunou et Christian Bonhomme (Université
Pierre et Marie Curie, Paris) aiming at guiding the analysis
of complex NMR spectra acquired for some of their compounds.
We focused in particular, on Cs4(H2V10O28),4H2O displaying
5 distinct vanadium sites with very similar chemical shifts.
Since this compound is used as a precursor in sol-gel syntheses
of functionalized materials, it was of paramount importance
to gather information on the alkalinity and then the reactivity
of the different anionic sites within the structure. The
attribution of the 51V resonances permitted to make progress
in this direction. By coupling geometry optimization based
on DFT and chemical shifts calculations based on the GIPAW
method, we are able to propose a complete and accurate crystallographic
structure including well defined protons positions. It is
now also possible to assign, without any doubt, the five
observed NMR resonances to the five non-equivalent vanadium
crystallographic sites, thanks to the accurate calculations
of the EFG and CSA tensors parameters. These results will
be soon submitted for publication.
|
|
Our attention was also held by the family
of the VOPO4 compounds, studied during the Ph.D work of
Maxence Launay, in which several phases display structural
ambiguities: polymorphism problems, metastability, stacking
faults. MAS NMR coupled with simulation should allow to
shed light on some of these unsolved problems. This is a
work in progress.
|
Publications |
1- L. Truflandier, M. Paris, C. Payen and F. Boucher J. Phys. Chem. B (2006), 110, 21403-21407
2- L. Truflandier, M. Paris, and
F. Boucher :Phys.
Rev. B 76, 035102 (2007)
|
|
| |
© IMN
Téléphone : +33 2 40 37 39 39
Télécopie : +33 2 40 37 39 95
|
|
Updated on spetember, 20th 2007
|
|
