Chemistry Reference
In-Depth Information
underlying symmetry of a particular space group,
regardless of the chemical nature
of the occupancy
. At issue in this context is the question whether or not any
particular, derived substructure makes chemical sense. The pure crystallographer
might take the view that this question is irrelevant.
3 The Gamma Phase of Li
7
V
V
N
4
-Li
7
V
V
N
4
, space group
The lattice sites occupied in the cubic compound
g
P
43
n
, are
listed below (see also Table
2
):
Li(1) in 12
f
at (
x
, ½, ½): Li(2) in 6
d
at (¼, 0, ½): Li(3) in 6
b
at (0, 0, ½).
Li(4) in 8
e
at (
x, x, x
): Li(5) in24
i
at (
x
,
y, z
): V(2) in 6
c
at (¼, ½, 0).
V(1) in 2
a
at (0,0,0): N(1) in 8
e
at (
x, x, x
) and N(2) in 24
i
at (
x, y, z
).
Hence, the question then arises as to whether there are some compounds in
43
n
or in its immediate supergroups whose atoms occupy
some
of the above sites, and if
so, whether there is a rational chemical argument as to why there should be this
correspondence.
P
3.1 Substructures in the Space Group Pm
3
n
3.1.1 The
A
15-Type Structure
We begin this discussion by recalling the comment by Schnick and Luecke [
11
]
on the structure of the isomorphic compound, Li
7
PN
4
, as determined by them:
“
...
distribution of P atoms in the cubic unit cell of Li
7
PN
4
(
a
¼
936.48(3) pm)
resembles a
” The
archetypal
A
15 structure is that of the alloy Cr
3
Si, and indeed the arrangements of
P atoms in Li
7
PN
4
and of V atoms in
b
-tungsten analogous arrangement (A15-type) as illustrated
...
-Li
7
V
V
N
4
conform to
th
e A15 type. This
compound, Cr
3
S
i
(and many like it), crystallizes in the S.G.
g
m
3
n
, which immedi-
P
(¼, 0, ½) and the Si atoms
ar
e in site 2
a
at (0, 0, 0). However, these sites correspond
to the sites 2
a
and 6
d
of
ately precedes
P
43
n
, occupied by V(1) and Li(2), respectively,
not V(1)
V(1) and Li(2) arrangement is complementary to that of V(1) and V(2). The
Thus, the pattern of the Cr
3
Si structure is present in the structures of both
P
g
-Li
7
VN
4
and Li
7
PN
4
, although these latter patterns are considerably expanded
with respect to the actual Cr
3
Si structure: this derives from the fact that the cell edge
of Cr
3
Si is approximately half that of
-Li
7
VN4. Nevertheless, the patterns are the
same, despite the bond lengths being very different, as is seen in identical views
icosahedra in the same orientation. If we now depict the V(2) linkages in
g
g
-Li
7
VN
4