LaNiO2: Synthesis and structural characterization

Article abstract of DOI:10.1016/j.jssc.2005.01.023

The preparation of a single phase LaNiO₂ by controlled H ₂ reduction of the perovskite LaNiO₃ is reported. The different steps of the synthesis are detailed. The structural characterization is made by X-ray diffraction (XRD) and neutron powder diffraction (NPD) and discussed; the derived crystal structure is the infinite-layer CaCuO ₂. Both XRD and NPD patterns are characterized by anisotropic shape broadening originating from size effect linked to the reduction process. From NPD, the LaNiO₂ powder is composed of platelet-like crystallites having an average thickness equal to 20 nm along the [001] direction. A modulation of the neutron diffraction background is observed and discussed in connection to the reduction process.

Full text of DOI:10.1016/j.jssc.2005.01.023

ARTICLE IN PRESS
Journal of Solid State Chemistry 178 (2005) 1326–1334
www.elsevier.com/locate/jssc
LaNiO2: Synthesis and structural characterization
Ã
M. Crespina,1, O. Isnardb,c, F. Duboisa, J. Choisnetd, P. Odierb,
a
`
´
´
´
´
Centre de Recherche sur la Matiere Divisee, UMR CNRS-Universite, 1 B rue de la Ferollerie, 45071 Orleans Cedex 2, France
bLaboratoire de Cristallographie, CNRS UPR 5031, 25 Avenue des Martyrs, BP166, 38042 Grenoble Cedex 09, France
c
´
Institut Universitaire de France, Maison des Universites, 103 boulevard Saint Michel, 75005 Paris, France
d
´
CRISMAT, UMR CNRS-ENSICAEN, 6 Boulevard du Marechal Juin, 14050 Caen Cedex, France
Received 2 September 2004; received in revised form 10 January 2005; accepted 11 January 2005
Abstract
The preparation of a single phase LaNiO2 by controlled H2 reduction of the perovskite LaNiO3 is reported. The different steps of
the synthesis are detailed. The structural characterization is made by X-ray diffraction (XRD) and neutron powder diffraction
(NPD) and discussed; the derived crystal structure is the infinite-layer CaCuO2. Both XRD and NPD patterns are characterized by
anisotropic shape broadening originating from size effect linked to the reduction process. From NPD, the LaNiO2 powder is
composed of platelet-like crystallites having an average thickness equal to 20 nm along the [001] direction. A modulation of the
neutron diffraction background is observed and discussed in connection to the reduction process.
r 2005 Elsevier Inc. All rights reserved.
Keywords: LaNiO2; Nickel monovalent; Controlled reduction
1. Introduction
of the perovskite LaNiO3, with a controlled H2
reduction loop at moderate T (ꢀ250–450 1C). The
authors identified the nickelite LaNiO2 containing
formally Ni(I), as deriving from LaNiO2.5, the phase
issued from the first step of the reduction. Because Ni(I)
3d9 is isoelectronic with Cu(II), the key element for high
Tc superconductivity, some interest was devoted to
synthesize Ni(I) containing oxides with structures
similar to those where superconductivity is observed.
Hence, several new phases involving Ni(I) in K2NiF4
The use of topotactic transformations is an interesting
and energetically economic way to manipulate chemical
species at low temperature. For example, it allows to
synthesize BaTiO3 at a few hundred 1C [1] or to
transform the goethite FeO(OH) oxyhydroxide into
Fe2O3 hematite by mecano-chemical reaction [2]. It is
also efficient to introduce order in solid matter, for
example in Li manganate [3] for electrochemical
applications or in double perovskite manganites with
colossal magneto resistance properties [4]. Oxygen
intercalation–de-intercalation also belongs to topotac-
tic-type reactions; it has been widely used to produce
new phases or metastable phases with novel electronic
and chemical features [5] and mixed-valence [6].
derived structure were obtained: La1.6Sr0.4NiO3.47
,
LaSrNiO3.1 [8,9], La(Ni1ꢁxAlx)O2+x [10]. Manthiran et
al. [11] discussed the factors influencing the stabilization
of Ni(I) in perovskite-related oxide An+1NinO3n+1
.
They stressed that ‘‘bond length matching between the
stretched Ni+–O and A–O bond to minimize the
internal stress and coordination preference and size of
the A cations are found to play an important role in
accessing Ni+’’. As another example, the strongly
reduced nickelite Nd4Ni3O8 is the triple-[NiO2] layer,
intermediate between the T’-type and the infinite layer
type where all the Nd atoms are in a cubic coordination
Ni(I) has been discovered in solid state, 20 years ago
by Crespin et al. [7] by operating a topotactic reduction
Ã
Corresponding author. Fax: +33 4 76 88 10 38.
E-mail address: philippe.odier@grenoble.cnsr.fr (P. Odier).
1Retired from CNRS.
0022-4596/$ - see front matter r 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.jssc.2005.01.023

ARTICLE IN PRESS
M. Crespin et al. / Journal of Solid State Chemistry 178 (2005) 1326–1334
1327
nitrogen trap (4). The reaction cell can be heated and
isolated from the circulated loop. The system can be
pumped (primary and secondary vacuum, (5)) and filled
with He, H2, O2, N2 (6) or other gases.
The volumetric measurement is based on the use of a
gas burette system containing five spheres of different
known volumes. A reservoir of mercury is used to
modify the free volume of the burette; the pressure is
measured by a mercury manometer. The determination
of a volume of the loop (a fraction of it or its totality)
simply uses the Boyle–Mariotte’s law and is given by
Pnbn
Vx ¼
,
(1)
P0 ꢁ Pn
where Vx is the unknown volume of glass in standard
conditions (stp). P0 is the pressure of gas when the
burette is full of mercury (volume b0), Pn is the pressure
when the nth sphere is filled with gas and bn is
the volume of the nth sphere (b0 ¼ b1 þ b2 þ Á Á Á bn). In
the actual system b1 ¼ 3:650 cm3; b2 ¼ 13:890 cm3; b3 ¼
33:900 cm3; b4 ¼ 80:730 cm3; and b5 ¼ 188:950 cm3; all
being measured with a high precision leading to an error
in the range of 71 Â 10ꢁ3 cm3.
Fig. 1. Structure of LaNiO2 initially proposed by Crespin et al. [7] and
isotype to the ‘‘N layer’’.
[12,13]. Recently, Ni(I) and Cu(II) where shown to
coexist in LaNi0.5Cu0.5O2.25 [14].
In the first report [7], LaNiO2 was assumed to have
the very simple structure of (Ba,Ca)CuO2, the so called
‘‘N layer’’ discovered later [15], see Fig. 1. Several
unsuccessful attempts to reproduce the experiment of
Crespin et al. have blown, however, some doubt on the
existence of the LaNiO2 phase. Hayward et al. [16] using
NaH, the most powerful reducing agent, succeeded
some years ago to reduce LaNiO3 at a sufficiently low
temperature to avoid the decomposition of LaNiO2. By
neutron powder diffraction (NPD), they confirmed the
structure predicted initially, but due to a rather
imperfect sample, i.e., it had a composition LaNiO2+x
and therefore was not fully reduced, they were forced to
work a complicated procedure to fit the data. Despite
the progress obtained by Hayward et al., the problem of
the synthesis of pure LaNiO2 still remains questionable.
This motivated us to have a better insight on the first
method of synthesis and to obtain a sample as pure
as possible.
The gas volume at room temperature is
Vgas ¼ VxðPx273:16Þ=ð760TÞ,
(2)
where Px is the pressure of gas for the volume Vx; and
TðKÞ is the ambient temperature.
The experiment is conducted as follows. Typically,
500 mg of powder is poured in a dense alumina boat that
was previously outgassed. After introducing the sample
in the reaction cell, it is treated for 12 h at 400 1C under
dry oxygen flow to remove all traces of organic residue
and fully oxidize the sample, and then the system is out
gazed under vacuum at 150 1C max. The volume of the
loop is then measured with He as indicated above; it
includes the sample cell heated at the temperature of the
reducing experiment, the cold trap and the circulating
loop. The sample cell is then isolated and a given volume
of H2 introduced in the vessel and measured in stp (the
amount of introduced H2 at starting is 3 to 4 times more
than required for the total reduction). H2 is further
introduced in the circulating loop to react with the
sample; the water vapor formed being trapped in the
cold trap. Pressure is measured versus time; it is a
measure of the H2 consumption and gives access to the
progress of the reduction process and to its kinetics.
Both volume and pressure can be measured with a very
low uncertainty, allowing a good precision even for
small volumes. Typically, a volume of 0.1 cm3 can be
measured with a precision of 5%. This ensures a rather
accurate follow-up of the de-oxygenation kinetics and of
the oxygen stoichiometry, as well. As a result, the error
in the oxygen stoichiometry factor is estimated to be
70.015 oxygen.
This paper reports our successful attempt to repro-
duce the initial experiment of synthesis of LaNiO2, but
this time with a sufficiently large amount of powder to
allow an NPD study.
2. Experimental
2.1. Reduction vessel
The apparatus was described in Ref. [7] but no details
were given at that time on the operational mode, which
is described now. To help the reader, we show on Fig. 2
a draw of the vessel. It is composed of a circulated glass
loop (1) (volume close to 380 cm3), a volumetric
measurement vessel (2), a reaction cell (3) and a liquid

ARTICLE IN PRESS
M. Crespin et al. / Journal of Solid State Chemistry 178 (2005) 1326–1334
1328
Fig. 2. Reactor for reduction experiments.
2.2. Sample preparation
cylindrical vanadium sample holder of 6 mm inner
diameter is used. The room temperature diffraction
patterns are recorded over an angular range of 801 (2y),
using a multidetector with a step of 0.21 between each of
A batch (10 g) of LaNiO3 powder is prepared by a
sol–gel method using acrylamid gelification. This
method has been already reported for LaNi0.5Cu0.5O3
phase [17]. It allows to obtain fine particles (o100 nm)
with the exact stoichiometry. The LaNiO3 phase is
obtained pure at 800 1C under pure oxygen flow with the
typical unit cell. To complete its crystallization, it is
heated to 950 1C under oxygen flow and slowly furnace
cooled, no line broadening can be detected on such
sample with our D5000 apparatus (line width ꢀ0.151).
The initial oxygen stoichiometry of the sample is
always measured as a preliminary step by using the
previous procedure; it is close to O3 in the present case
(within the experiment error) asserting the quality of
both sample and experimental procedure.
3
the 400 He detection cells. In this configuration, the
D1B instrument is operating with a wavelength of
(
l ¼ 1:28 A selected by the (311) Bragg reflection of a Ge
monochromator, the take off angle being 44.21 2y at this
wavelength. The data are analysed with the Rietveld
structure refinement FULLPROF program [19]. Guide-
lines of the Rietveld refinement can be found elsewhere
[20]. The neutron scattering lengths bNi ¼ 1:030 Â
10ꢁ14 m; bLa ¼ 0:824 Â 10ꢁ14 m; bO ¼ 0:580 Â 10ꢁ14 m
values are taken from Ref. [21].
X-ray powder diffractions have been recorded either
in the reflection or transmission mode in a D8 (Brucker)
and D5000 (Siemens) apparatus, respectively, the
wavelength was that of CuKa1 (Ni filtered).
2.3. Structural characterization
The neutron diffraction experiments is performed at
the Institut Laue Langevin (I.L.L.), Grenoble, France,
on the D1B instruments operated by the CNRS whose
detailed description can be found elsewhere [18]. A
3. Results and discussion
Typical results of de-oxygenation versus time
are shown onto Fig 3. It shows plots of oxygen

ARTICLE IN PRESS
M. Crespin et al. / Journal of Solid State Chemistry 178 (2005) 1326–1334
1329
exothermic and then if a too large batch is reduced, the
sample overheats, leading to an uncontrolled product of
reaction. Probably, a rate-controlled synthesis should be a
useful method for further experiments. In any case, when
using a small mass of sample (ꢀ500 mg), the right
temperature (310 1C), a low heating rate (o50 1C hꢁ1), a
low pressure of gas (not higher than 0.6 atm H2), and a
sufficiently large depression (2.5 Â 10ꢁ3 atm) at the
circulating pump, reproducible pure samples of LaNiO2
phase can be produced. To get a sufficient amount of
product for NPD, we mixed six samples of smaller
amount but of quasi-identical quality. The X-ray dif-
fractogram of Fig. 4 is obtained from such a large batch
(3 g). It is clear that all lines index within the LaNiO2
tetragonal cell, as reported initially in Ref. [7]. N PD is
performed on that batch of powder and reported Fig. 5.
From that diffractogram, it is detected some Ni impurities
but not accompanied with La2O3.
3.0
2.8
2.6
2.4
2.2
2.0
1.8
LaNiO3 -> LaNiO2
310°C. 0.814 atm H2
-200
0
200 400 600 800 1000 1200 1400 1600 1800
t (min)
Fig. 3. Kinetics of de-oxygenation of LaNiO3 at 310 1C under an
initial H2 pressure of 0.5 atm.
stoichiometry calculated from the H2 consumption
versus time for two samples operated under identical
3.1. Structural analysis
conditions, i.e., T ¼ 310 1C—PH ꢀ 0:500 atm H2—in-
2
itial weight ꢀ400 mg. The de-oxygenation occurs in two
steps of very different kinetics. The first one is very
rapid, it lowers the oxygen stoichiometry down to O2.5
and give rises to LaNiO2.5 [22]. This has been confirmed
in a preliminary experiment of a reduction in situ
followed by NPD. The second de-oxygenation step is
lengthier and tends progressively to the formation of
LaNiO2 after 1500 min (25 h) at 310 1C. At this level, the
hydrogen pressure reaches ꢀ0.400 atm. Note that in the
course of the reduction we could maintain a constant
pressure of H2 by increasing the level of mercury in the
reservoir V (Fig. 2) and then perform the reduction at
constant pressure. However, after each measurement of
pressure, the value of b0 must be measured again and
then, the circulation in the loop has to be stopped for
some time. This makes more difficult to follow-up the
reduction kinetics, especially in its fast regime. Accord-
ing to our experiments, this does not change the kinetics
if the reduction of pressure is small, that is the case here.
The structure of the reduced phase is checked by X-ray
diffraction. For small samples, typically p500 mg, a pure
LaNiO2 phase is obtained and this result is reproducible.
Attempts to produce larger batches of pure LaNiO2
powders (1–3 g) were undertaken but failed. We have tried
to find the optimal temperature. When the temperature is
too low, i.e., a few degrees below T ¼ 310 1C, the kinetics
is considerably lowered and even after a week, the
reaction is not finished, the product being composed of
LaNiO2.5 and other phases. On the other hand, if the
temperature is raised even a few degrees above 3101C, the
reduction invariably yields to uncontrolled and significant
amounts of Ni and La2O3 in addition to LaNiO2. The
crucial point to access the pure phase is the first step of the
reaction, i.e., the formation of LaNiO2.5. From several
experiments, we have observed that the first step is
It seems to be clear that this diffraction pattern is
featured by rather broad reflections, certainly much
broader than that observed for a well-crystallized oxide
such as Al2O3 (Fig. 6). Even more, the line shape
broadening is not constant over the reciprocal space but
depends upon the (h,k,l) values. This feature has been
also noticed by X-ray diffraction (transmission mode).
Fig. 7 shows the full width at half maximum (FWHM)
of the different reflections pointing to a differentiation
between (hk0) and the other (hkl) indexes. Note that the
error in the experiment bar is in the range of 0.11.
Such an anisotropic line shape broadening led to
difficulties in the Rietveld analysis. A detailed analysis
of the systematic broadening as function of (hkl) led us
to conclude that this is due to a size effect. The (00l)
reflections are exhibiting the largest FWHM as can be
seen from XRD (Fig. 7) and from NPD (Fig. 5).
Modelling the anisotropic broadening involves the
hypothesis that the correlation length for coherent
crystallites is finite along the c-axis and infinite within
the (001) plane. The integral width (bhkl; corrected for
instrumental resolution) of a Bragg reflection for a small
crystallite size is given by the Scherrer formula:
bhkl ¼ l=Dhkl cos y,
(3)
where l is the wavelength, y is the Bragg angle and Dhkl
is the average diameter (for the all volume of the sample)
in the direction normal to the (hkl) planes as defined in
Ref. [23]. The instrument resolution is obtained from the
fit of the Al2O3 diffraction pattern given in Fig. 6. If D is
the average thickness (over the volume of the sample) of
the platelets, then D ¼ Dhkl cos ahkl; where ahkl is the
acute angle formed by the normal to the (hkl) planes and
the normal to the platelets. Assuming that the broad-
ening is mainly of Lorentzian character, then the

ARTICLE IN PRESS
M. Crespin et al. / Journal of Solid State Chemistry 178 (2005) 1326–1334
1330
3500
3000
2500
2000
1500
1000
500
0
10
20
30
40
50
60
70
80
90 100 110 120
2θ
Fig. 4. High resolution X-ray diffractograms of a batch of 3 g of LaNiO2 obtained by addition of 6 smaller samples.
(
Fig. 5. Neutron powder diffraction pattern for LaNiO2 at 300 K (l ¼ 1:28 A). The drawn pattern refers to the experimental data. The difference
between the experimental data and the calculated fit is plotted in the lower part of the figure. Vertical bars indicate calculated Bragg peak positions
for (hkl) reflections. The second row of bars corresponds to the Bragg peak positions of the minority impurity Ni phase.
FWHM is given (in degrees) by
where Sz is the parameter refined by FULLPROF from
which the average thickness D can be obtained. This
Lorentzian broadening when added to the instrumental
Lorentzian component reproduces the experimental
FWHML ¼ 360l cos ahkl=ðp2D cos yÞ
¼ Sz cos ahkl= cos y,
ð4Þ

ARTICLE IN PRESS
M. Crespin et al. / Journal of Solid State Chemistry 178 (2005) 1326–1334
1331
(
Fig. 6. Neutron powder diffraction pattern for Al2O3 at 300 K (l ¼ 1:28 A). This reference diffraction pattern has been recorded in the same
conditions as the one plotted on Fig. 4.
used to convert LaNiO3 into LaNiO2. In this respect, it
1.2
1.0
0.8
0.6
0.4
0.2
0.0
LaNiO2
must be emphasized that the X-ray pattern of the
starting LaNiO3 phase did not exhibited such peak
broadening. One can perhaps speculate that due to the
stronger La–O bonds than Ni–O bonds, one can expect
that going from LaNiO3 to LaNiO2, the oxygen is
removed from the Ni planes and then the structure
brakes up along the [001] direction, then contributing to
smaller coherent domains along [001] than along [100].
It is worth to note that attempts to refine the
occupancy of all sites were performed but that no
significant deviation from LaNiO2 stoichiometry has
been obtained. The occupancy for Ni was found to be
0.96(5) and that of oxygen 1.08(6) that is not significant
deviations. In the same way, the possibility of a
coexistence of a perfect infinite layer model NiO2–La–
NiO2 with defective NiO2–LaO–NiO2 regions leading to
an overall excess stoichiometry, as reported by Hayward
et al. [16], has been herein checked. Unlike their results,
a good fit of the diffraction pattern has been obtained
without addition of a more complex description of the
LaNiO2 phase. Consequently, the final calculation is
based on a full occupancy of the La, Ni and O sites. In
the refinement procedure, Ni metal was introduced as a
secondary phase; in any case, the calculated amount is
0.370.02 weight percent. One could think that the
removal of some Ni from the [yLa/NiO2/Lay]
sequence along the c axis must be accompanied by a
20
30
40
50
60
70
80
90
100
2θ
Fig. 7. Anisotropic X-ray line broadening in LaNiO2.
Voigt peak shape (approximated by the Pseudo–Voigt
function). The value obtained for the volume averaged
thickness is D ¼ 200ð17Þ A. The precise nature of the
defects given rise to the lack of long-range coherence of
crystallites along the [001] direction is not known at
present. An anisotropic broadening of the Bragg
reflections has already been observed in several com-
pounds prepared by low temperature de-oxygenation
and particularly in nickelates and cobaltites
[13,15,24–26]. This anisotropic line shape broadening
most probably originates from the topotactic transfor-
mation involved in the oxygen de-intercalation process

ARTICLE IN PRESS
M. Crespin et al. / Journal of Solid State Chemistry 178 (2005) 1326–1334
1332
redistribution of the oxygen atoms toward the defective
La planes and would induces the formation of
corresponding pairs of LaO/LaO planes, typical of the
intergrowth structure of La2NiO4. Even if the presence
of an La2NiO4 extra-phase was not detected, the
formation of some local La2NiO4-like defects is not
precluded in agreement with the disproportionation of
LaNiO2 to Ni and La2NiO4 reported by Hayward et al.
[16]. The experimental and calculated diffractograms of
the final calculation are shown in Fig. 5 and the
corresponding parameters and agreement factors are
reported in Table 1.
The main M–O, M–M and O–O distances calculated
in the structure of LaNiO2, are reported in Table 2. The
four Ni–O bond lengths (1.98 A) are the same as found
previously in LaNiO2 (16). Of interest is the significant
anisotropy of the La–La and Ni–Ni distances. In both
cases, the two apical one are much shorter (3.37 A) than
the four in plane one (3.96 A). The apical La–La
distances are significantly shorter than the mean
corresponding value in La2O3: 3.94 A (27). It can be
reasonably assumed that there will be a decrease of the
La3+–La3+ repulsive forces in the LaNiO2 justifying
the La–La shortening. As previously emphasized in the
intergrowth oxides (27), this will play an important role
in the bidimensionality of the structure and further in
the metastability of LaNiO2.
A careful examination of the neutron diffraction
background (Fig. 8) reveals the existence of some
modulation in the range 2y ¼ 281: Such feature is also
visible in the X-ray diagram in the corresponding range
2y ¼ 331: The background intensity of neutron diffrac-
tograms includes a contribution of elastic diffuse
scattering, coming from a local static structural order
[28]. Note that in the neutron diffraction pattern here
recorded at room temperature in a regular vanadium
container, the pattern is free from any diffuse scattering
from the container, thus allowing a careful examination
of the contribution due to the sample. It is well known
that such a phenomenon gives evidence for the
occurrence of short range ordering suggesting a possible
second phase. This is not so surprising in the structure of
an oxide, which is synthesized by a low temperature
procedure. Such modulation effects have been identified
in the neutron diffraction background of highly oxygen-
conductive oxides and in Al2O3 treated at high
temperature [29–32] and in the X-ray scattering back-
ground as well, for example in perovskite-like inter-
growth type aluminate [33].
The Debye formula [28] can be used to express the
contribution of a static structural order to the diffuse
elastic scattering. It exhibits a first pronounced max-
Table 1
Structural parameters of LaNiO2 from NPD refinement
imum for Qmax ¼ ð2p  1:23Þ=dm; where dm is
a
Temperature
Instrument
290 K
D1B
preferred pair distance. In our case, the modulation
has a maximum at Q] 2.41 Aꢁ1, then it corresponds to a
preferred pair distance close to 3.22 A. Looking for
metals able to pair at this distance points to two groups:
(La–Ni) 8 distances of 3.26 A in LaNiO2 (see Table 2)
and (La–La) one distance between LaO layers (3.25 A)
in the c direction of La2NiO4 [27]. Clearly, the former
case cannot be considered resulting in a local ordering,
since the La–Ni pairs are eaqually spread over the whole
structure of LaNiO2. On the contrary, the later case, i.e.,
La–La pairing along c corresponds to a local ordering
induced by the removal of Ni atoms, as mentioned
hearabove to account for the presence of the extra Ni
phase.
Wavelength (A)
La 1d (x; y; z)
N i 1a(x; y; z)
O 2f (x; y; z)
1.28
(0.5,0.5,0.5)
(0,0,0)
(0,0.5,0)
BLa (A2)
BNi (A2)
0.4 (2)
0.5 (1)
BO (A2)
a (A)
c (A)
1.1 (1)
3.959 (1)
3.375 (1)
RBragg (%)
Rwp (%)
Rp (%)
6.7
5.11
3.97
0.43
Rexp (%)
Preliminary magnetization measurements with an
SQUID magnetometer show a behaviour related to the
strong magnetic contribution from Ni impurity
(0.3 wt%) and complex MðTÞ under low field. Attempts
to subtract the contribution of Ni to that of the sample
was not done.
SG(123), P4/mmm.
Table 2
Main interatomic distances at room temperature in LaNiO2
Distances (A)
La
La
N
i
O
4 Â 3.959(1)
2 Â 3.375(1)
i  3.268(3)
8 Â 3.268(3)
8 Â 2.600(3)
4. Conclusion
N
O
84 Â 3.959(1)
2 Â 3.375(1)
4 Â 1.980(1)
4 Â 1.980(1)
A single LaNiO2 phase sample is synthesized by
controlled reduction under H2 at 310 1C. Its structure is
refined from NPD data taking into account an
8 Â 2.600(3)
2 Â 3.375(1)
4 Â 2.800(1)

ARTICLE IN PRESS
M. Crespin et al. / Journal of Solid State Chemistry 178 (2005) 1326–1334
1333
(
Fig. 8. Details of the neutron powder diffraction pattern LaNiO2 at room temperature versus Q ¼ 4p sin y=l (l ¼ 1:28 A) showing a large diffuse
peak in the background around 2.41 Aꢁ1 (due to short range order with pair distances around 3.22 A). Note that the container contribution is
negligible (vanadium container).
anisotropic line broadening that is clearly evidenced. It
is ascribed to the existence of coherent domains with
different size according to the direction, i.e., 20 nm along
[001], larger than 100 nm in the perpendicular direction.
These domains are apparently linked with the reduction
process giving rise to LaNiO2. A modulation of the
neutron diffraction background is observed, in connec-
tion with a cationic local ordering. It is suggested to
correspond to small La2NiO4 particles expected to form
because of an exsolution of Ni during the reduction
process. Unfortunately, it probably results in a meta-
stable phase that degrades at room temperature after
several months.
[2] G. Gonzales, A. Sarzazu, R. Villalba, Mater. Res. Bull. 35 (14–15)
(2000) 2295.
[3] A. Ibarra-Pales, M. Anne, P. Strobel, J. Solid State Chem. 160 (1)
(2001) 108.
[4] S.V. Trukhanov, I.O. Troyanchuk, M. Hervieu, H. Szymczak, K.
Barner, Phys. Rev. B 66 (18) (2002) 184421–184424.
[5] A. Stein, S.W. Keller, T.E. Mallok, Science 259 (1993) 1558.
[6] J.N. Lalena, R.A. McIntyre, B.L. Cushing, K.A. Thomas, J.L.
Heintz, C.T. Seip, C.J. O’Connor, J.B. Wiley, Solid State Chem.
Inorg. Mater. II-Symp. 99 (1999) 404.
[7] M. Crespin, P. Levitz, L. Gatineau, J. Chem. Soc. Fraday Trans.
79 (1983) 1181.
[8] M. Crespin, J.M. Bassat, P. Odier, P. Mouron, J. Choisnet,
J. Solid State Chem. 84 (1990) 165.
[9] M. Crespin, C. Landron, J.P. Odier, M. Bassat, P. Mouron,
J. Choisnet, J. Solid State Chem. 100 (1992) 281.
[10] M.J. Martinez-Lope, M.T. Casais, J.A. Alonso, J. Alloys
Compds. 275–277 (1998) 109.
Acknowledgments
[11] A. Manthiran, J.P. Tang, V. Manivannan, J. Solid State Chem.
148 (1999) 499.
[12] P. Lacorre, J. Solid State Chem. 97 (1992) 495.
[13] R. Retoux, J. Rodriguez-Carvajal, P. Lacorre, J. Solid State
Chem. 140 (1998) 307.
The Institut Laue Langevin is warmly acknowledged
for providing the neutron facilities as well as the CNRS
for the use of the CRG–D1B instrument. Dr. A. Sulpice
is thanked for preliminary measurements of magnetisa-
tion and fruitful discussions.
[14] F. Dubois, P. Odier, J. Choisnet, J. Mater. Chem. 13 (2003) 1.
[15] T. Siegrist, S.M. Zahurak, D.W. Murphy, R.S. Roth, Nature 334
(1988) 231.
[16] M.A. Hayward, M.A. Green, M.J. Rossdeinsky, J. Sloan, J. Am.
Chem. Soc. 121 (1999) 8843.
References
[17] P. Odier, M. Municken, M. Crespin, F. Dubois, P. Mouron,
J. Choisnet, J. Mater. Chem. 12 (5) (2002) 1370.
[18] http://www.ill.fr, Yellow Book.
[1] Q. Feng, M. Hirasawa, K. Yanagisawa, Chem. Mater. 13 (2)
(2001) 290.
[19] J. Rodriguez Carjaval, Physica B 192 (1993) 55.

ARTICLE IN PRESS
M. Crespin et al. / Journal of Solid State Chemistry 178 (2005) 1326–1334
1334
[20] L.B. McCusker, R.B. Von Dreele, D.E. Cox, D. Louer, P. Scardi,
J. Appl. Crystallogr. 32 (1999) 36.
[21] V.F. Sears, Neutron News 3 (3) (1992) 26.
[28] A. Guinier, Theorie et Techniques de la Radiocristallographie,
´
Editions Dunod, Paris, 1964.
[29] R.N. Vannier, F. Abraham, G. Nowogrocki, G. Mairesse, J. Solid
State Chem. 142 (1999) 294–304.
[22] F. Gotor-Martinez, P. Odier, O. Isnard, unpublished data.
[23] E.F. Bertaut, Acta Crystallogr. 5 (1952) 117.
[30] F. Goutenoire, O. Isnard, R. Retoux, P. Lacorre, Chem. Mater.
12 (2000) 2575.
[24] J.A. Alonso, M.J. Martinez-Lope, J.L. Garcia-Munoz, M.T.
Fernandez-Diaz, J. Phys.: Condens. Mater. 9 (1997) 6417.
[25] M.A. Hayward, M.J. Rosseinsky, Chem. Mater. 12 (2000) 2182.
[26] J. Rodriguez-Carvajal, M.T. Fernandez-Diaz, J.L. Martinez,
J. Phys. Condens. Matter 3 (1991) 3215.
[31] F. Goutenoire, O. Isnard, E. Suard, O. Bohnke, Y. Laligant, R.
Retoux, P. Lacorre, J. Mater. Chem. 11 (2001) 119.
[32] P. Aldebert, A.J. Dianoux, J.P. Traverse, J. Phys. 40 (1979) 1005.
[33] I. Zvereva, Yr. Smirnov, J. Choisnet, Mater. Chem. Phys. 60
(1999) 63.
[27] J. Choisnet, J. Solid State Chem. 147 (1999) 379.