Get Cached PDF (752 KB) - CORE

Copy and paste this link to your website, so they can see this document directly without any plugins.


with, magnetic, were, from, temperature, 6H-BaFeO2F, Figure, that, data, oxidation, between, this, also, iron, moments, parameters, composition, decomposition, fluorine, group, different, Fe3+, which, analysis, high, Chem., anion, oxide, field, structure


Synthesis, structural and magnetic characterisation of
the fully fluorinated compound 6H-BaFeO2F
Oliver Clemens*,a, Adrian J. Wrighta, Frank J. Berrya, Ronald I. Smithb, Peter R.
a School of Chemistry, University of Birmingham, Birmingham B15 2TT, United
b ISIS Facility, Rutherford Appleton Laboratory, Harwell Oxford, Didcot, OX11 0QX,
United Kingdom.
*: corresponding author
Fax +44 (0)121 4144403
The compound 6H-BaFeO2F (P63/mmc) was synthesised by the low temperature
fluorination of 6H-BaFeO3-d using polyvinylidenedifluoride (PVDF) as a fluorination
agent. Structural characterisation by XRD and NPD suggests that the local positions
of the oxygen and fluorine atoms vary with no evidence for ordering on the anion
sites. This compound shows antiferromagnetic ordering at room temperature with
antiparallel alignment of the magnetic moments along the c-axis. The use of PVDF
also allows the possibility of tuning the fluorine content in materials of composition
6H-BaFeO3-dFy to any value of 0 < y ≤ 1. In addition, the oxygen content, and
therefore the iron oxidation state, can be tuned by applying different partial pressures
of oxygen during the reaction.
6H-BaFeO2F, hexagonal perovskite, fluorination, PVDF, magnetic structure
1 Introduction
Perovskite type A[12]B[6]X3 compounds are widely used for a variety of applications,
such as fuel cell cathodes, solid electrolytes, and gas sensors [1-3].
In the perovskite structure, the AX3 subsystem forms a close packed lattice, where
the stacking sequence can differ [4] from purely cubic (c… e. g. for SrFeO2F [5-9]
and BaFeO2F [7, 10-12] (both space group Pm3m)) to purely hexagonal (h… e. g. for
BaCoO3 [13]) and can also form lattices in between those two extremes (e. g.
cchcch… for 6H-BaFeO3-d [14-19] or chcch… for 15R-BaFeO3-dF0.2 [20-22]). For a
fully filled anion lattice X3, all the B cations are octahedrally coordinated and the
connection to the neighbouring octahedrons is determined from the stacking
sequence. For a cation between cc layers, the octahedron is connected by corners to
6 neighbouring octahedra (3 in the layer above, 3 in the layer below) whereas for a
cation between hc layers, it is connected by faces to one octahedron via the h layer
and by corners to 3 octahedra via the c layer. For a cation between hh layers the
connection is to 2 octahedra by faces (1 in the layer above, one in the layer below).
A range of factors can be responsible for the type of structure adopted. In addition to
electronic and electrostatic reasons the average oxidation state of the B cation and
therefore the relative average size of the ions, numerically expressed in the tolerance
factor t [4], strongly influences the structure adopted.
The pure oxide BaFeO3-d can be prepared via a simple high temperature solid state
reaction from the metal-oxides or -carbonates. For this system the oxygen partial
pressure has been shown to be of crucial importance in determining the type of
structure adopted [7, 14, 23, 24] particularly at higher temperatures. These phases
and others can be fluorinated by a variety of fluorination agents [25], and prior
studies have reported either a partially fluorinated BaFeO3-dFy [20-22] in a 15R
perovskite structure (chcch…) or fully fluorinated cubic BaFeO2F [7, 10-12] where the
latter contains only Fe3+ and can be prepared by low temperature fluorination of the
precursor oxide BaFeO2.5 (space group P21/c, vacancy ordered distorted cubic
perovskite) using PVDF.
Low temperature treatments of perovskite phases have, for many years, been known
to be suitable to modify the average transition metal oxidation states and therefore
allow for the synthesis of different modifications of the same product, e.g. cubic and
hexagonal SrMnO3 [26]. Recently, Hayashi et al. showed that cubic BaFeO3 can be
made by low temperature oxidation of BaFeO2.5 (P21/c) [23]. Formation of the
different modifications is supported by the good ionic conductivity of the anions,
whilst the Ba/Fe substructure remains immobile at low temperatures [20]. The PVDF
route [7, 25, 27] has also been shown to be a “chimie douce” route allowing the
formation of kinetically stable products which decompose at higher temperatures to
the thermodynamically most stable products.
The reaction mechanism for the PVDF method depends on the material which is to
be fluorinated [25]. For the fluorination of BaFeO2.5 to cubic BaFeO2F, two F
- ions
replace one O2- ion (and the carbon residues of the polymer are burnt off by the
oxygen from the air) [7, 10, 11], whereas for the fluorination of SrFeO3 to cubic
SrFeO2F one F
- ion replaces one O2- ion (supported by the stability of the trivalent
oxidation state of iron and/or the reductive potential of PVDF) [5, 6]. Hence the filling
of vacancies and/or the substitution of oxygen ions are the possible reaction
mechanisms for this method of fluorination (for nonstoichiometric compounds (e. g.
Sr0.5Ba0.5FeO2.77 [7]) the reaction mechanism would be a combination of vacancy
filling and substitution of oxygen).
Recent work has shown that perovskite phases with nearly single valent Fe3+ show a
high stability in their magnetic ordering even at elevated temperatures. BaFeO2F
(space group Pm3m) is a G-type antiferromagnet [10] and SrFeO2F [6] also shows
antiferromagnetic ordering at ambient temperature. 15R-BaFeO3-dFy (space group
R3m, 0.15 ≤ y ≤ 0.30) and 6H-Ba0.8Sr0.2FeO3-dFy (space group P63/mmc,
0.15 ≤ y ≤ 0.25) were also shown to be antiferromagnetically ordered, where the
magnetic moments align in different directions with respect to the direction along
which the AX3 layers lie [22]. All these phases show magnetic ordering above room
temperature, whereas fluorine-free mixed valent 6H-BaFeO3-d (space group
P63/mmc, d ~ 0.15) orders below 130 K [16].
In this paper, we report the first synthesis of hexagonal 6H-BaFeO2F (space group
P63/mmc) by low temperature fluorination of 6H-BaFeO3-y using
polyvinylidenedifluoride (PVDF) as a fluorination agent. We report on the
characterisation of the sample by neutron diffraction and magnetic measurements as
well as by high temperature XRD to investigate the thermal stability towards
decomposition. We also describe preliminary results on the synthesis of partly
fluorinated samples 6H-BaFeO3-dFy (0 < y < 1) and use lattice volume relationships to
provide an estimation of their oxygen content (and consequent iron oxidation state).
2 Experimental
2.1 Sample synthesis
The precursor oxide 6H-BaFeO3-d was prepared by a solid state reaction.
Stoichiometric mixtures of high purity BaCO3 and Fe2O3 powders (Sigma Aldrich,
≥ 99.9 %) were ground using a planetary ball mill (Fritsch pulverisette 7, 350 rpm,
1.33 h) and heated to 970°C for 12 h under flowing O2. The samples were slowly
cooled to room temperature (20°C/h) to increase the oxygen uptake and ensure the
formation of the pure 6H-BaFeO3-d phase and the procedure was repeated a second
For the preparation of the oxide fluorides of composition BaFeO3-dFy (y = 0.2, 0.4,
0.8, 1), stoichiometric amounts of the as prepared 6H-BaFeO3-d and
polyvinylidenedifluoride (PVDF) were thoroughly ground in n-pentane (for the
synthesis of BaFeO2F, y = 1, a 4 % excess of PVDF was used). The mixtures were
then slowly heated to 370°C (20°C/h) under air and kept at this temperature for 20 h;
slow heating was found to be beneficial to minimise the amount of BaF2 impurity
formed during the reaction (~ 1 wt-%). The as prepared oxide fluoride materials were
subsequently heated to 370°C for 4h under flowing O2 to allow uptake of oxygen, and
hence maximise the Fe oxidation state.
Structural studies focused on the 6H-BaFeO2F phase, which was shown to be
metastable and its’ thermal decomposition was confirmed by studying the high
temperature decomposition products arising from heating at 1000°C for 5 min in air.
2.2 Diffraction experiments
XRD patterns were recorded with a Bruker D5005 diffractometer with Bragg-Brentano
geometry and a fine focus X-ray tube with Cu anode. No primary beam
monochromator was attached. A PSD detector and a fixed divergence slit were used.
The total scan time was 16 hours for the angular range between 5 and 140° 2θ.
High temperature XRD of 6H-BaFeO2F was performed with a Bruker D8
diffractometer with Bragg-Brentano geometry and a fine focus X-ray tube with Cu
anode in a 2-range from 20 to 60 degrees and at temperatures between 30 and
750°C in steps of 30°C. A primary beam monochromator was attached and a LYNX
eye detector and fixed divergence slit were used. The total scan time was 1 hour for
the angular range between 20 and 60° 2θ for each temperature step.
Time of flight powder neutron diffraction (NPD) data were recorded on the newlyupgraded Polaris medium resolution diffractometer at the ISIS pulsed spallation
source (Rutherford Appleton Laboratory, UK). 4g of 6H-BaFeO2F powder was loaded
into a 8mm diameter thin-walled, cylindrical vanadium sample can and data collected
at ambient temperature for 250µAh proton beam current to the ISIS target
(corresponding to ~1¾ hours beamtime).
Structure refinement of both the XRD and NPD data was performed using the
Rietveld method with the program TOPAS 4.2 (Bruker AXS, Karlsruhe, Germany)
[28]. For the room temperature XRD data the whole 2θ-range was used, while for the
NPD data only those data collected in the highest resolution backscattering detector
bank (bank 5, average 2θ = 146.7º, dmax ~2.65Å) were used. The instrumental
intensity distribution for the X-ray data was determined empirically from a sort of
fundamental parameters set [29] using a reference scan of LaB6, and microstructural
parameters were refined to adjust the peak shapes for the XRD data. For the neutron
diffraction data, a modified pseudo Voigt function plus a Gaussian crystallite size
function was used to model the time-of-flight dependence of the peak width. Lattice
parameters were constrained to be the same for neutron and XRD data and the
same positional parameters were used and refined for both data sets. Independent
thermal displacement parameters were refined for each atom type, but the values for
O and F were constrained to the same value. While these parameters were also
constrained to be the same both for X-ray, and neutron, powder diffraction data, an
additional B overall value was refined for XRD data accounting for further effects
such as absorption or surface roughness. Reflections that showed a large magnetic
scattering contribution were omitted for the initial crystallographic refinement.
Furthermore, the intensities of the XRD and NPD patterns were normalised to values
between 0 and 1 to give each pattern similar weight in the Rietveld analysis.
Refinement of the magnetic structure of 6H-BaFeO2F was performed with the
program GSAS [30, 31] using the NPD data collected in one of the Polaris low angle
detector banks (bank 3, average 2θ = 52.2º, dmax ~ 7.02Å). The magnetic contribution
to the diffraction pattern was modelled by introducing a second phase in triclinic
space group P1 containing just Fe atoms (to allow refinement of the magnetic
structure without any symmetry restrictions), and calculating only its magnetic
scattering. Unit cell, atomic positions and thermal vibration parameters in this second
phase were set to the refined values determined above and then fixed to ensure that
the triclinic (P1) cell remained geometrically and symmetrically hexagonal. Different
orientations of the magnetic moments were investigated, including those previously
reported for similar compounds [22].
2.3 Magnetic measurements
DC susceptibility measurements were performed over the temperature range 5-300 K
using a Quantum Design MPMS SQUID magnetometer. The samples were
pre-cooled to 5 K in zero field (ZFC) and also in an applied field of 0.05 T (FC) and
values of χ measured whilst warming in a field of 0.05T. Field-dependent DC
susceptibility measurements were performed with a Quantum Design PPMS system
with the ACMS control system in DC extraction mode. Measurements were
performed at 5 K between 0 and 5 T.
2.4 Mössbauer spectroscopy
The 57Fe Mössbauer spectrum was recorded in constant acceleration mode using a
ca. 25 mCi 57Co/Rh source at 300 K.
3 Results and discussion
3.1 Structural characterisation of 6H-BaFeO2F
Fluorination of the precursor oxide 6H-BaFeO3-d using a 4 mole-% excess of PVDF
under air resulted in the formation of a fine brown powder. The 57Fe Mössbauer
spectrum (see Figure 1) was best fitted to three sextets consistent with a
magnetically ordered material with a Néel temperature greater than 300 K. The
chemical isomer shifts δ 0.41, 0.36, 0.32 ± 0.04 mm/s are consistent [32] with iron
being present only as Fe3+ and the formation of a phase of composition 6H-BaFeO2F
(in combination with the evidence of a completely filled anion lattice from the
evaluation of neutron diffraction data reported later in this article).
Figure 1.
Fe Mössbauer spectrum recorded at 300 K from 6H-BaFeO2F.Highly fluorinated
perovskite phases are known to decompose at high temperatures [7, 9]. This results
from the high thermodynamic stability of the alkaline earth (AE) fluorides (AE)F2 [25]
and is well known for AE, or La, containing oxide fluoride compounds. In two previous
articles [7, 9], it was shown that such decomposition reactions can also be used to
confirm the sample composition. A decomposition experiment at 1000°C for 5 min
showed the formation of only BaF2 (space group Fm3m) and BaFe2O4 (space group
Cmc21). A quantitative Rietveld analysis indicated molar fractions of 49.7(4) and
50.3(4) mole-% of these two phases, giving an overall composition of
Ba0.99Fe1O2F0.99 and thereby independently confirmed the assumed composition of
the sample. The decomposition of 6H-BaFeO2F is therefore very similar compared to
that of cubic BaFeO2F (space group Pm3m) [7].
6H-BaFeO2F  0.5 BaF2 + 0.5 BaFe2O4
A detailed structural characterisation was performed by coupled Rietveld analysis of
XRPD and NPD data (see Figure 2, Figure 3 and Table 1). In this case, the NPD data
allow for a more detailed analysis of the anion sublattice, which was found to have
only a minor influence on the observed intensities in the XRD pattern.
Figure 2. Refined crystal structure of hexagonal 6H-BaFeO2F. Fluoride ions are indicated in grey and
oxide ions in black. The magnetic moments on the different iron sites are indicated by arrows.
Table 1. Structural data for 6H-BaFeO2F (space group P63/mmc) from a coupled Rietveld analysis of XRD
and POLARIS bank 5 NPD data.
x y z occupancy B [Ų]
Ba2+ 2b 0 0 1/4 1 0.50(3)
Ba2+ 4f 1/3
2/3 0.08942(6) 1 0.46(2)
Fe3+ 2a 0 0 0 1 1.54(3)
Fe3+ 4f 1/3
2/3 0.85247(6) 1 1.15(3)
O2- 6h 0.5148(7) 0.4852(7) 1/4
2/3 1.24(4)
O2- 12k 0.1644(5) 0.8356(5) 0.5855(2) 2/3 0.79(2)
F- 6h 0.498(1) 0.502(1) 1/4
1/3 1.24(4)
F- 12k 0.169(1) 0.831(1) 0.5756(4) 1/3 0.79(2)
a [Å] 5.76350(4) c [Å] 14.2119(1) V [ų] 408.842(7)
2.18 RBragg
0.85 (XRD)
2.8 (NPD)
Figure 3. Coupled Rietveld analysis of XRD (a) and POLARIS bank 5 NPD (b) data from the hexagonal
phase 6H-BaFeO2F. The reflection from BaF2 (~1.1 wt-%) with the highest intensity is marked with an
asterisk for the XRD data.
Since fluorination leads to the formation of a fully occupied close packed BaO2F
sublattice, no coordination other than octahedral is obviously plausible for the Fe
atoms (also confirmed by Mössbauer spectroscopy). Since no additional reflections
appeared in the diffraction pattern, the same space group as for the precursor oxide
6H-BaFeO3-d (P63/mmc) was assumed. In this space group, the anions occupy two
different crystallographic sites, 6h and 12k, with the refinement indicating significant
anisotropy for the thermal parameters for the anions (rod like (6h) and disk like
(12k)). This led to the assumption that a model involving split positions might be more
appropriate than a situation in which oxygen and fluorine occupy exactly same
crystallographic positions. This is also supported by a simple consideration of their
ionic radii [33], which differ by approximately 0.07 Å and therefore we might expect
the local positioning of these anions to be non-identical. Indeed, initial experimental
evidence for the possible correctness of this assertion was found by Rietveld
analysis, where the occupancies of the split sites were allowed to refine and resulted
in values of approximately 2/3 and
1/3, which are consistent with the occupancies
expected for O and F respectively. In the final refinement, the occupancies of the
anion sites were therefore fixed to the values of 1/3 for F
- and 2/3 for O
2-, as expected
from the composition and the sites assigned accordingly. Nevertheless, anisotropic
parameters might also be a possible model since the difference in Rwp values is small
(2.723 with 90 refined parameters for anisotropic displacement parameters for the
anions vs. 2.707 with 85 refined parameters for split sites).
Although previously reported for the lower F content phase 6H-Ba0.8Sr0.2FeO3-dFy
(0.15 ≤ y ≤ 0.25, prepared by high temperature synthesis) [20, 22] we could not find
clear evidence that the fluorine atoms favour one of the two independent
crystallographic sites (6h and 12k) over the other. This also becomes apparent from
calculations of the anion charges using Pauling’s second rule [4] which show that for
single valence Fe3+ and a composition BaFeO2F both anion sites 6h and 12k should
have the same overall charge of -5/3 (= (-2)*
2/3 + (-1)*
1/3). It should also be noted that
bond valence sums [34] give similar results for an ordered, as compared with a
random distribution of the anions. This is consistent with iron in 6H-BaFeO2F being
present as only Fe3+ (compared to mixed valent Fe3+/Fe4+ in 6H-Ba0.8Sr0.2FeO3-dFy
[20, 22]).
A detailed interpretation of the split anion sites is very difficult, since the closer
approach of each anion to one of the cations would increase the distance to another
cation. However, a tentative assignment of the F and O atoms to these split sites
(according to their occupancies) gives the refined cation-anion distances that are
listed in Table 2. Importantly, the average Fe-O/F distances are 2.028 Å for the 2a
and 4f sites and this compares very well with expectations based on the ionic radii
(2.022 Å) [33] or from the situation in cubic BaFeO2F (2.028 Å) [7]. Furthermore, the
Fe3+ on the 4f site is displaced from the centre of the octahedron as a result from the
cation repulsion in face sharing octahedra.
Table 2. Refined cation-anion distances from a coupled Rietveld analysis of XRD and POLARIS bank 5
(backscattering) NPD data. Assignment of O
and F
to the split positions was made according to the
occupancies for the split sites.
distances [Å]
O2- (6h site) F- (6h site) O2- (12k site) F- (12k site)
Ba2+ (2b site) 6x 2.8856(5) 6x 2.8818(3) 6x 2.857(4) 6x 2.998(7)
Ba2+ (4f site) 3x 2.914(4) 3x 2.814(7)
6x 2.8824(1)
3x 3.003(4)
3x 2.862(7)
6x 2.8885(4)
Fe3+ (2a site) - - 6x 2.042(4) 6x 2.000(9)
Fe3+ (4f site) 3x 2.102 (3) 3x 2.225 (6) 3x 1.903(4) 3x 1.933(7)
3.2 Magnetic characterisation
3.2.1 SQUID measurements
The variation of magnetic susceptibility χ (in an applied field of 0.05 T) with
increasing temperature from 5 to 300 K following pre-cooling in (i) zero applied field
(ZFC) and (ii) an applied field of 0.1 T (FC) is shown in Figure 4.
A divergence in the susceptibility between FC and ZFC measurements is evident
over the whole temperature range and indicates a very weak ferromagnetic
component. The evaluation of the field dependency of the magnetisation at 5 K is
shown in Figure 5. The small magnetic moment per 6H-BaFeO2F formula unit clearly
indicates an antiferromagnetic ordering of the magnetic moments on the iron ions (for
a detailed analysis of the magnetic structure, see section 3.2.2). Furthermore the
material shows a slight hysteresis (Hc ~ 0.2T). The remanent magnetisation per iron
ion is very low and can be estimated to be around 0.0025 µB. Such a magnetic
moment could arise either by the presence of very small amounts of impurity phases
or by a weak canting canting angle of ca. 0.02° of the magnetic moments. Both would
be outside the detection limits of the neutron diffraction experiments. Similar findings
were also reported for the cubic modification of BaFeO2F [10]. It is relevant to note
that the remanent magnetisation is lower by about an order of magnitude than for
6H-Ba0.8Sr0.2FeO3-dFx (0.15 ≤ x ≤ 0.25) [22] and that χmol is reduced by about two
orders of magnitude as compared to the precursor oxide 6H-BaFeO3-d [17, 18].
Figure 4. Variation of susceptibility χ of 6H-BaFeO2F between 5 and 300 K. The data were recorded at
increasing temperature in a measuring field of 0.05 T. Separate plots shows field cooled (FC) and zero
field cooled (ZFC) data.
Figure 5. Field dependent magnetisation of 6H-BaFeO2F measured at 5 K.
3.2.2 Determination of the magnetic structure
As discussed in section 3.2.1, temperature dependent measurements of the
susceptibility and field dependent measurements of the magnetisation indicate an
antiferromagnetic ordering of the magnetic moments. The magnetic moments are
antiferromagnetically aligned in adjacent layers along the c-axis (ferromagnetically
ordered in a single layer with no simple superexchange pathways between the
cations in the layer). This is often found in perovskite-related materials. Strong
superexchange interactions between the cations in the cc and the ch layers (~180°
Fe-(O/F)-Fe angle) induces antiferromagnetism between these sites. For face shared
octahedral cations in two ch layers, the Fe-(O/F)-Fe angle is close to 90°. Such
interactions favour weak ferromagnetic interactions, but direct exchange between
high-spin Fe3+ atoms is necessarily antiferromagnetic, and the latter might therefore
be the case for 6H-BaFeO2F (d(Fe4f-Fe4f) = 2.91 Å).
In the literature [22, 35], two models for the orientation of the magnetic moments are
generally discussed: along the a- or along c-axis. A detailed analysis of bank 3 (low
angle) POLARIS data reveals that, for 6H-BaFeO2F, the moments align along the
a-axis (resp. lie in the ab-plane) (see Figure 6a). This becomes especially evident by
analysis of the reflections at ca. 4.7 Å (see inlays Figure 6a+b). Two reflections may
contribute to the observed intensity of this reflection, (0 1 1) and (0 0 3), and the latter
would have to carry the scattered magnetic intensity to yield a good fit. For an
alignment of the magnetic moments only along c, scattered magnetic intensity would
be zero for (0 0 3) [36]. The difference between these two models is also expressed
in the goodness of fit for the refinements, which were significantly better when the
magnetic moments were oriented along the a-axis than along the c-axis. The
magnetic space group according to an alignment of the magnetic moments along the
a-axis is Cmcm (BNS 63.457) and a subsequent refinement using this group results
in an identical fit. The magnetic structure is shown in Figure 2.
The magnetic moments for the Fe3+ ions on the 2a and the 4f site are nearly identical
(3.65(4) and 3.32(3) µB) being in good agreement with values observed for
6H-Ba0.8Sr0.2FeO3-dFy (0.15 ≤ y≤ 0.25) which were slightly lower (3.56(4) and
2.72(7) µB) [22]. From the similarity of the magnetic moments to those of
6H-Ba0.8Sr0.2FeO3-dFy, one might expect the Néel Temperature to be similar to - or
even higher than - that of 6H-Ba0.8Sr0.2FeO3-dFy (~ around 700 K [22]). 6H-BaFeO2F
therefore shows very robust antiferromagnetic ordering and TN is expected to be of
the same magnitude as reported for 6H-Ba0.8Sr0.2FeO3-dFy (0.15 ≤ y≤ 0.25) [22],
15R-BaFeO3-dFy’ (0.15 ≤ y’ ≤ 0.30) [22], cubic BaFeO2F [11] and Sr2Fe2O5 [37].
We suggest that the exchange interaction might be different from that observed
previously in the lower fluorine content 6H-Ba0.8Sr0.2FeO3-dFy (0.15 ≤ y ≤ 0.25) [22],
where a tetrahedral corner shared coordination occurs around the Fe3+ ions on the 4f
site (caused by a shift of the anions on their positions, not by one of the Fe3+ ions),
leading to a superexchange angle of nearly 180°. This was reported to be an
additional cause of the observed robust antiferromagnetism of Ba0.8Sr0.2FeO3-dFy
(0.15 ≤ y ≤ 0.25) next to the reduction of the average iron oxidation state [22]. Since
the anion lattice is completely filled for 6H-BaFeO2F only octahedral coordination can
occur in this material.
It is probable that the lowering of the oxidation state helps to stabilise the
antiferromagnetic ordering, as mentioned previously by Sturza et al. [22]. The Néel
temperature of 6H-BaFeO3-d (130 K [18]) is significantly lower than that of the
fluorinated compounds.
Figure 6. Rietveld analysis of POLARIS bank 3 (low angle) NPD data for different models of magnetic
structures. Magnetic moments aligned parallel to the a-axis and antiferromagnetic interactions between
neighbouring layers (a) and magnetic moments aligned parallel to the c-axis and antiferromagnetic
interactions between neighbouring layers (b).
3.3 High temperature XRD investigations of 6H-BaFeO2F
The high temperature behaviour of 6H-BaFeO2F was investigated to examine the
decomposition reaction in more detail. Figure 7a shows the variation of the lattice
parameters (normalised to those from the lowest measurement temperature) with
temperature. An almost linear relationship is observed up to ~510°C. Thereafter, the
c-lattice parameter increases greater than the a-lattice parameter and small amounts
of BaF2 begin to appear as a first decomposition product (Figure 7b). Around 700°C,
the decomposition reaction seems to be complete and BaF2 and BaFe2O4 are found
in the same relative amounts as observed for full decomposition at 1000°C (see
section 3.1). The fact that BaF2 can be observed at lower temperatures than BaFe2O4
might reflect the decomposition of BaFeO2F creating a Ba/F deficient perovskite
phase with formula Ba1-cFeO2F1-2c. This could also explain why the lattice parameters
a and c behave differently at temperatures higher than 510°C. The simultaneous
formation of (e. g. amorphous) BaFe2O4 and BaF2 would be expected to leave the
overall composition, and therefore the nearly linear increase in lattice parameters of
the hexagonal perovskite phase, largely unaffected. This is also in agreement with an
attempt to refine the Ba occupancy of the partly decomposed product at ~620°C,
which indicates 15 % of vacancies on the 2b site (c layer) and 5 % vacancies on the
4f site (h layer). The refined formula of Ba0.92FeO2F0.84 is in excellent agreement with
the refined weight fraction of BaF2, again commensurate with an overall total sample
composition of approximately Ba1Fe1.01O2.03F0.99. Clearly, further work will be
necessary to fully understand this complex decomposition behaviour and it is planned
to use this decomposition method as a synthesis route for novel A site deficient
perovskites in the future.
Figure 7. The high temperature behaviour of 6H-BaFeO2F. Dependency of lattice parameters and cell
volume (normalised to the lowest measurement temperature ~ 35°C) (a) and weight fractions of
6H-BaFeO2F and its decomposition products (b).
3.4 XRD study of partly fluorinated 6H-BaFeO3-dFy
One of the key advantages in the use of PVDF as a fluorinating agent is that by
varying the amount of PVDF used the fluorine content of a material can be controlled.
This was illustrated in our recent fluorination studies of Sr3Fe2O7-x, where a range of
different phases were prepared with different fluorine contents [27]. The PVDF
method also allows partial fluorination of 6H-BaFeO3-d leading to materials of
composition 6H-BaFeO3-dFy (see Figure 8 for XRD powder patterns). In contrast, high
temperature preparations of 6H-Ba0.8Sr0.2FeO3-dFy from BaF2, BaCO3, SrCO3 and
Fe2O3 are only suitable for a narrow range of fluorine content [20-22], due to the low
thermal stability of phases with higher F contents.
Figure 8. XRD diffraction data for samples of composition 6H-BaFeO3-dFy (after treatment under O2
atmosphere). The most intense reflection of the impurity phase BaF2 (~ 1 wt-%) found for 6H-BaFeO2F is
marked with an asterisk.
In previous studies [7, 9], we showed that the oxidation of Fe3+ has the main
influence on the observed cell volume. Therefore, the cell volume (see Figure 9) can
be used to roughly estimate the average oxidation state of the iron species. From
this, the value of d in 6H-BaFeO3-dFy (resp. the composition of the anion lattice) can
be approximated, assuming complete incorporation of F- from the PVDF (which is
likely since the material can be fully fluorinated using only a very slight excess of F- in
the polymer). This furthermore assumes a composition of 6H-BaFeO2.81 for the pure
oxide, as determined from our earlier titration experiments [7] and which is also in
good agreement with other values from the literature [22] and a composition of
6H-BaFeO2F for the fully fluorinated material, as shown by the combination of
neutron diffraction, Mössbauer studies and decomposition experiments. It is also
relevant to note that the cell volume of cubic BaFeO3 (62.621 ų, Fe
4+ only [23]) is in
good agreement with that (61.899 ų) from the volumes and average iron oxidation
states of 6H-BaFeO2.81 and 6H-BaFeO2F and therefore independently confirms the
approximate validity of these assumptions [23].
Figure 9. Dependency of the volume per BaFeO3-dFy unit on the value of y (assuming complete F
Values for the average iron oxidation state as well as the calculated overall
composition for samples prepared under air and subsequently heated under O2 are
shown in Table 3. The incorporation of even low amounts of fluorine (y = 0.2)
drastically lowers the iron oxidation state, with the change estimated from the cell
volume being from 3.62  3.32. This is in agreement with the work by Sturza et al.
[22], who showed that their lower F content materials 6H-Ba0.8Sr0.2FeO3-dFy
(0.15 ≤ y ≤ 0.25) also displayed a significantly lower oxidation state than the pure
oxide (although it should be recognised that their samples were quenched from
higher temperatures, which is beneficial for stabilising lower Fe oxidation states).
Heating in oxygen at 370°C can only establish a marginally higher oxidation state for
the Fe ions (Δq ~ 0.01 - 0.06), while the low fluoride-containing samples can be
oxidised to higher degrees (as illustrated by the greater difference between air/O2
points at low F contents in Figure 9). For fully fluorinated 6H-BaFeO2F no significant
change in the refined lattice parameters was observed and this is in agreement with a
completely filled anion sublattice. Although it should be emphasised that this is an
approximate approach, the results presented here suggest that low temperature Fincorporation helps stabilise the iron oxidation states closer to 3+ in the 6H perovskite
structure. Such low oxidation states cannot be stabilised without destabilising the
structure for the pure oxide 6H-BaFeO3-d (6H-BaFeO3-d began to decompose when
heated under air at 370°C). Further investigations by neutron diffraction are planned
to investigate the influence of the fluorine content (and total anion content) on the
structure and magnetic ordering in this system.
Table 3. Refined cell volumes per BaFeX3-z unit and approximated average iron oxidation states and
material compositions for fluorinated samples 6H-BaFeO3-dFy (from an analysis of the cell volumes
assuming complete incorporation of the F
ions from the PVDF fluorinating agent). * The sample started to
transform/decompose when heated at 370°C under air).
y Vf.u. [ų]
V (O2)
air O2
1 68.125 3.00 BaFeO2F 68.123 3.00 BaFeO2F
0.8 67.455 3.11 BaFeO2.15F0.8 67.349 3.12 BaFeO2.16F0.8
0.6 66.983 3.18 BaFeO2.29F0.6 66.773 3.22 BaFeO2.31F0.6
0.4 66.661 3.24 BaFeO2.42F0.4 66.238 3.30 BaFeO2.45F0.4
0.2 66.157 3.32 BaFeO2.56F0.2 65.753 3.38 BaFeO2.59F0.2
0 65.704* -* -* 64.264 3.62 BaFeO2.81
4 Conclusions
We report here the first synthesis of 6H-BaFeO2F by low temperature fluorination of
the precursor oxide 6H-BaFeO3-d. In contrast to high temperature syntheses routes
[20-22], this allows the production of a material with the highest possible fluorine
content. The material decomposes at ~510°C into BaF2 and BaFe2O4, which explains
why the synthesis of this phase is not possible by high temperature reactions.
6H-BaFeO2F shows robust antiferromagnetism at room temperature. Oxygen and
fluorine ions are randomly distributed across the lattice, but evidence for slightly
different anion positions linked to the local presence of oxygen and fluorine was
found, providing different local bond distances for these anions. The PVDF method is
confirmed as an excellent route to controlled fluorine contents, here allowing the
preparation of 6H-BaFeO3-dFy (0 < y ≤ 1) phases with different fluorine contents by
simple control of the amount of PVDF used.
5 Acknowledgements
Oliver Clemens wants to thank the German Academic Exchange Service (DAAD) for
being given a Postdoctoral Research Fellowship. The Bruker D8 diffractometer used
in this research was obtained through the Science City Advanced Materials project:
Creating and Characterising Next generation Advanced Materials project, with
support from Advantage West Midlands (AWM) and part funded by the European
Regional Development Fund (ERDF). Neutron diffraction beamtime at ISIS was
provided by the Science and Technology Facilities Council (STFC).
6 References
[1] G. Deng, Y. Chen, M. Tao, C. Wu, X. Shen, H. Yang, Electrochim. Acta 54
(2009) 3910-3914.
[2] P.A. Murade, V.S. Sangawar, G.N. Chaudhari, V.D. Kapse, A.U. Bajpeyee,
Curr. Appl. Phys. 11 (2011) 451-456.
[3] L. Nalbandian, A. Evdou, V. Zaspalis, Int. J. Hydrogen Energy 34 (2009)
[4] U. Müller, Anorganische Strukturchemie. B. G. Teubner Verlag, Wiesbaden,
[5] F.J. Berry, X. Ren, R. Heap, P. Slater, M.F. Thomas, Solid State Commun.
134 (2005) 621-624.
[6] F.J. Berry, et al., J. Phys.: Condens. Matter 20 (2008) 215207.
[7] O. Clemens, R. Haberkorn, P.R. Slater, H.P. Beck, Solid State Sci. 12 (2010)
[8] E. Sullivan, C. Greaves, Mater. Res. Bull. 47 (2012) 2541-2546.
[9] O. Clemens, M. Kuhn, R. Haberkorn, J. Solid State Chem. 184 (2011) 28702876.
[10] F.J. Berry, F.C. Coomer, C. Hancock, Ö. Helgason, E.A. Moore, P.R. Slater,
A.J. Wright, M.F. Thomas, J. Solid State Chem. 184 (2011) 1361-1366.
[11] R. Heap, P.R. Slater, F.J. Berry, O. Helgason, A.J. Wright, Solid State
Commun. 141 (2007) 467-470.
[12] I.O. Troyanchuk, N.V. Kasper, O.S. Mantytskaya, E.F. Shapovalova, Mater.
Res. Bull. 30 (1995) 421-425.
[13] K. Yamaura, H.W. Zandbergen, K. Abe, R.J. Cava, J. Solid State Chem. 146
(1999) 96-102.
[14] J.C. Grenier, A. Wattiaux, M. Pouchard, P. Hagenmuller, M. Parras, M. Vallet,
J. Calbet, M.A. Alario-Franco, J. Solid State Chem. 80 (1989) 6-11.
[15] F. Iga, Y. Nishihara, G. Kido, Y. Takeda, J. Magn. Magn. Mater. 104-107
(1992) 1969-1972.
[16] F. Iga, Y. Nishihara, T. Katayama, K. Murata, Y. Takeda, J. Magn. Magn.
Mater. 104–107, Part 3 (1992) 1973-1975.
[17] K. Mori, T. Kamiyama, H. Kobayashi, K. Itoh, T. Otomo, S. Ikeda, Physica B:
Condensed Matter 329–333, Part 2 (2003) 807-808.
[18] K. Mori, T. Kamiyama, H. Kobayashi, T. Otomo, K. Nishiyama, M. Sugiyama,
K. Itoh, T. Fukunaga, S. Ikeda, J. Appl. Crystallogr. 40 (2007) s501-s505.
[19] M. Parras, M. Vallet-Regi, J.M. González-Calbet, J.C. Grenier, J. Solid State
Chem. 83 (1989) 121-131.
[20] M. Sturza, S. Daviero-Minaud, H. Kabbour, O. Gardoll, O. Mentré, Chem.
Mater. 22 (2010) 6726-6735.
[21] M. Sturza, S. Daviero-Minaud, M. Huvé, N. Renaut, N. Tiercelin, O. Mentré,
Inorg. Chem. 50 (2011) 12499-12507.
[22] M. Sturza, H. Kabbour, S. Daviero-Minaud, D. Filimonov, K. Pokholok, N.
Tiercelin, F. Porcher, L. Aldon, O. Mentre, J. Am. Chem. Soc. 133 (2011) 1090110909.
[23] N. Hayashi, T. Yamamoto, H. Kageyama, M. Nishi, Y. Watanabe, T.
Kawakami, Y. Matsushita, A. Fujimori, M. Takano, Angewandte Chemie International
Edition 50 (2011) 12547-12550.
[24] J.M. Gonzalez-Calbet, M. Parras, M. Vallet-Regi, J.C. Grenier, J. Solid State
Chem. 86 (1990) 149-159.
[25] P.R. Slater, J. Fluorine Chem. 117 (2002) 43-45.
[26] T. Negas, R.S. Roth, J. Solid State Chem. 1 (1970) 409-418.
[27] C.A. Hancock, T. Herranz, J.F. Marco, F.J. Berry, P.R. Slater, J. Solid State
Chem. 186 (2012) 195-203.
[28] Topas V4.2, General profile and structure analysis software for powder
diffraction data, User's Manual. Bruker AXS, Karlsruhe, 2008.
[29] R.W. Cheary, A.A. Coelho, J.P. Cline, J. Res. Nat. Inst. Stand. Technol. 109
(2004) 1-25.
[30] A.C. Larson, R.B. Von Dreele, Los Alamos National Laboratory Report LAUR
(1994) 86-748.
[31] B.H. Toby, J. Appl. Cryst. 34 (2001) 210-213.
[32] F. Menil, J. Phys. Chem. Solids 46 (1985) 763-789.
[33] R.D. Shannon, Acta Crystallogr. A32 (1976) 751-767.
[34] I.D. Brown, The chemical bond in inorganic chemistry: the bond valence
model. Oxford University Press Inc., New York, 2002.
[35] S.A. Ivanov, S.G. Eriksson, J. Erikssen, R. Tellgren, H. Rundlof, Mater. Res.
Bull. 39 (2004) 615-628.
[36] J.M.D. Coey, Magnetism and Magnetic Materials. Cambridge University Press,
Cambridge, 2009.
[37] M. Schmidt, S.J. Campbell, J. Solid State Chem. 156 (2001) 292-304.
Figure Captions
Figure 1. 57Fe Mössbauer spectrum recorded at 300 K from 6H BaFeO2F.
Figure 2. Refined crystal structure of hexagonal 6H BaFeO2F. Fluoride ions are
indicated in grey and oxide ions in black. The magnetic moments on the different iron
sites are indicated by arrows.
Figure 3. Coupled Rietveld analysis of XRD (a) and POLARIS bank 5 NPD (b) data
from the hexagonal phase 6H-BaFeO2F. The reflection from BaF2 (~1.1 wt-%) with
the highest intensity is marked with an asterisk for the XRD data.
Figure 4. Variation of susceptibility χ of 6H-BaFeO2F between 5 and 300 K. The data
were recorded at increasing temperature in a measuring field of 0.05 T. Separate
plots shows field cooled (FC) and zero field cooled (ZFC) data.
Figure 5. Field dependent magnetisation of 6H-BaFeO2F measured at 5 K.
Figure 6. Rietveld analysis of POLARIS bank 3 (low angle) NPD data for different
models of magnetic structures. Magnetic moments aligned parallel to the a-axis and
antiferromagnetic interactions between neighbouring layers (a) and magnetic
moments aligned parallel to the c-axis and antiferromagnetic interactions between
neighbouring layers (b).
Figure 7. Results of the high temperature behaviour of 6H-BaFeO2F. Dependency of
lattice parameters and cell volume (normalized to the lowest measurement
temperature ~ 35°C) (a) and weight fractions of 6H-BaFeO2F and its decomposition
products (b).
Figure 8. XRD diffraction data for samples of composition 6H BaFeO3-dFy (after
treatment under O2 atmosphere). The most intense reflection of the impurity phase
BaF2 (~ 1 wt %) found for 6H-BaFeO2F is marked with an asterisk.
Figure 9. Dependency of the volume per BaFeO3-dFy unit on the value of y (assuming
complete F incorporation).

PDF Document reader online

This website is focused on providing document in readable format, online without need to install any type of software on your computer. If you are using thin client, or are not allowed to install document reader of particular type, this application may come in hand for you. Simply upload your document, and will transform it into readable format in a few seconds. Why choose

  1. Unlimited sharing - you can upload document of any size. If we are able to convert it into readable format, you have it here - saved for later or immediate reading
  2. Cross-platform - no compromised when reading your document. We support most of modern browers without the need of installing any of external plugins. If your device can oper a browser - then you can read any document on it
  3. Simple uploading - no need to register. Just enter your email, title of document and select the file, we do the rest. Once the document is ready for you, you will receive automatic email from us.

Previous 10

Next 10