FeLi[PO4]: Dissection of a Crystal Structure - Inorganic 3D Structures - page 84

like Na 2 [SO 4 ] does. A possible explanation is that the transition occurs at high
temperature and high pressure, and that such a second transition to a hexagonal
phase could take place at even higher temperatures and/or pressures. Even if the
Ni 2 In-type structure has not been observed in FeLiP, other related compounds,
such as KZnP, adopt this hexagonal structure [ 36 ] where regular 6 3 graphenes
are formed by both the Zn and P atoms. This feature is consistent with the
Zintl-Klemm concept, making that KZnP can be formulated, either as K + [ C -

GaP] or as K + [

-ZnS]. In any case, the pseudo-formulae correspond to (III)-(V)
and (II)-(VI) compounds, respectively, which normally adopt structures of the
Group 14. The fact that real ZnS forms the adamantane structure and the pseudo-
compound [

C

C

-ZnS] forms graphite-like layers gives support to our previous inter-

pretations.
Although discussed above, a remarkable feature of the HP-HT-phase ( b -FeLi
[PO 4 ]) is that in the P nma-phase (Fig. 10l ) , the P atoms form a much more regular

array than in the mcm phase (Fig. 10o ). In the latter, the P octahedra are almost

identical to the S-octahedra present in real

C

mcm-Fe[SO 4 ][ 33 ] . In our opinion, this
could indicate that at HP the Li atoms produce the stabilization of a real Fe[SO 4 ]
structure, adopting even the same space group (

C

mcm).
Moreover, in both FeLi[PO 4 ] and Fe[SO 4 ], the respective FeP (

C

-FeS) and and
FeS subarrays (NiAs type) coincide with the structure of FeS itself! Recall that in
the olivine-like phase, the FeP arrays (PbO type) were also similar to another phase
of FeS with this type of structure. It is also interesting to remark that in the

C

mcm
phase the PbO-like blocks are formed by both Li and P atoms, and that they are
more flattened than in the

C

nma-phase where the same blocks are formed by the
Fe-P pairs (compare Fig. 10k and n ) .
The impression is that the phase transition, undergone by FeLi[PO 4 ], involves a
clear phase transition of the FeP subarray which seems to control the structure,
adopting in both cases stable structures of FeS. The fact that in the olivine-like
phase the P octahedra are filled by the Li atoms (and not by the Fe atoms) is a clear
indication of the control exerted by the Fe-P pair of atoms, which form, on the other
hand, the PbO-type array.

P

4.3 The Co-existing Structures FeS/Fe[SO 4 ], LiCl/Li[ClO 4 ]
and FeLiP/FeLi[PO 4 ]

The PbO-type arrangements, present in the two phases of FeLi[PO 4 ], have been
widely discussed along this chapter. However, there remain some aspects,
concerning the metric of the structures, which were not discussed in deep, but
which provide additional data to gain an insight into the challenge underlying in this
approach as it was advanced earlier [ 4 ] , i.e. why the substructures are preserved
when more and more atoms are added to form a given compound?
The magnitudes, shown in Table 1 , together with the drawings represented in
Fig. 11 allow us to discover novel and unexpected structural features.