False Minima in X-ray Structure Solutions Associated with a “Partial Polar Ambiguity”: Single Crystal X-ray and Neutron Diffraction Studies on the Eight-Coordinate Tungsten Hydride Complexes, W(PMe<sub>3</sub>)<sub>4</sub>H<sub>2</sub>X<sub>2</sub> (X = F, Cl, Br, I) and W(PMe<sub>3</sub>)<sub>4</sub>H<sub>2</sub>F(FHF)
journal contribution
posted on 1998-04-28, 00:00 authored by Vincent J. Murphy, Daniel Rabinovich, Tony Hascall, Wim T. Klooster, Thomas F. Koetzle, Gerard ParkinThe molecular structures of the eight-coordinate tungsten hydride
complexes W(PMe<sub>3</sub>)<sub>4</sub>H<sub>2</sub>X<sub>2</sub>
(X =
F, Cl, Br, I) and
W(PMe<sub>3</sub>)<sub>4</sub>H<sub>2</sub>F(FHF) have been
determined by single-crystal X-ray diffraction;
W(PMe<sub>3</sub>)<sub>4</sub>H<sub>2</sub>Cl<sub>2</sub> and
W(PMe<sub>3</sub>)<sub>4</sub>H<sub>2</sub>F(FHF) have also
been analyzed by single-crystal neutron diffraction, thereby
accurately
locating the positions of the hydride ligands. The structures of
all of these complexes are similar and are
based on a trigonal dodecahedron, with a distorted tetrahedral array of
PMe<sub>3</sub> ligands in which two of the
PMe<sub>3</sub>
ligands are displaced over the halide substituents. However, the
initial structures derived for both
W(PMe<sub>3</sub>)<sub>4</sub>H<sub>2</sub>Cl<sub>2</sub> and
W(PMe<sub>3</sub>)<sub>4</sub>H<sub>2</sub>F(FHF) did
<i>not</i> exhibit the aforementioned geometry, but were based on
an arrangement
in which the two <i>transoid</i>-PMe<sub>3</sub> ligands are
displaced toward the two <i>cis</i>-PMe<sub>3</sub> groups, rather
than tilted
toward the chloride ligands. Interestingly, the unexpected
structures for
W(PMe<sub>3</sub>)<sub>4</sub>H<sub>2</sub>Cl<sub>2</sub> and
W(PMe<sub>3</sub>)<sub>4</sub>H<sub>2</sub>F(FHF) were discovered to be the result of an artifact due to the
presence of a heavy atom in a polar space
group, which allowed the X-ray structure solutions to refine into most
deceptive false minima. Specifically,
for the structures corresponding to the false minima, the
<i>transoid</i>-PMe<sub>3</sub> ligands were incorrectly located
in
positions that are related to their true locations by reflection
perpendicular to the polar axis. In effect, the
incorrect molecular structures are a <i>composite</i> of the two
possible true polar configurations which are related
by a reflection perpendicular to the polar axis, i.e. a “partial
polar ambiguity”. Of most importance, the solutions
corresponding to the false minima are characterized by low <i>R</i>
values and well-behaved displacement parameters,
so that it is not apparent that the derived structures are incorrect.
Thus, for space groups with a polar axis, it
is necessary to establish that all of the atoms in the asymmetric unit
belong to a single <i>true</i> polar configuration.
