Nickel-aluminum layered double hydroxides prepared via inverse micelles formation

Article abstract of DOI:10.1016/j.jssc.2009.03.005

Nickel-aluminum layered double hydroxides have been prepared by conventional coprecipitation and by coprecipitation in the presence of a surfactant. The solids have been characterised by several physicochemical techniques. Calcination leads to formation o

Full text of DOI:10.1016/j.jssc.2009.03.005

ARTICLE IN PRESS
Journal of Solid State Chemistry 182 (2009) 1593–1601
Contents lists available at ScienceDirect
Journal of Solid State Chemistry
journal homepage: www.elsevier.com/locate/jssc
Nickel–aluminum layered double hydroxides prepared via inverse
micelles formation
Ã
Mar ı´ a E. P e´ rez-Bernal, Ricardo J. Ruano-Casero, F a´ tima Benito, Vicente Rives
GIR-QUESCAT, Departamento de Qu ´ı mica Inorg a´ nica, Universidad de Salamanca, 37008 Salamanca, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Nickel–aluminum layered double hydroxides have been prepared by conventional coprecipitation and
by coprecipitation in the presence of a surfactant. The solids have been characterised by several
physicochemical techniques. Calcination leads to formation of homogeneously dispersed mixed oxides,
which have been characterised as well. The colour properties (lightness and chromaticity coordinates)
of both series of solids (layered precursors and calcined ones) have been measured. It has been found
that both the preparation method and the calcination treatment have an important effect on the
luminosity (whiteness/darkness) of the solids, although the effect on the precise chromaticity
coordinates (green/red and blue/yellow) is less marked.
Received 4 November 2008
Received in revised form
3
March 2009
Accepted 7 March 2009
Available online 20 March 2009
Keywords:
Layered double hydroxides
Mixed oxides
&
2009 Elsevier Inc. All rights reserved.
Chromatic parameters
Inverse micelles
2
+
3+
1.
Introduction
cation (the M /M
ratio in stable LDHs is larger than in
stoichiometric spinels). This fact opens new applications for these
compounds when one (or both) of the cations is coloured to
obtain new ceramic pigments [27,28]. We have previously
reported on the preparation and characterisation of different
LDHs with transition metal cations in the brucite-like layers, and
we have characterised their calcination products, usable as
pigments in the ceramic industry [27,29,30]; several colours were
developed in these studies, which properties were correlated to
the molar fractions of the metal cations and the thermal
treatment given. Our current aim is to develop preparation
procedures to obtain these ceramic pigments with pre-stablished
morphological properties, as current development in the ceramic
industry is based on the use of ink-jet ceramic pigments, needing
stable suspensions of the pigments and, as far as possible, a
uniform distribution of particles. Consequently, we here report on
the preparation of a green pigment from a Ni, Al containing LDH
prepared by two alternative routes, namely, conventional copre-
cipitation and coprecipitation in the presence of a surfactant in
order to control aggregation of the elemental particles through
formation of inverse micelles and thus controlling the properties
of the final product. The solids (both the layered precursors and
the calcined ones) have been characterised by several physico-
chemical techniques and the lightness and chromaticity coordi-
nates have been measured.
Layered double hydroxides (LDHs) constitute a family of
layered materials which can be described as brucite layers where
2+
3+
a partial Mg /M substitution has taken place. The positive
charge resulting from this substitution is balanced by hydrated
counteranions between the layers. The best known example is the
6 2 3 2
mineral hydrotalcite, Mg Al (OH)16(CO )4H O, and numerous
compounds with this structure have been prepared with a wide
variety of cations in the layers and of interlayer anions [1–5].
These materials have been widely applied as catalysts and catalyst
precursors [1,2,6–9], drugs carriers [1,10–14], adsorbents [15],
anion scavengers [16–19], composites [20–22], etc., some of these
properties arising from their anion exchange ability and from the
so-called memory effect, i.e., recovering of the layered structure
after being calcined at mild temperatures.
The properties of these compounds are highly dependent on the
specific method used to prepare them, usually by coprecipitation,
although other methods have been reported [23,24]. In addition, post-
preparation treatments (ageing) have been applied usually to improve
their crystallinity, and microwave irradiation has proved to be very
effective to produce deep changes in several properties [25,26].
Calcination of these compounds leads to a mixture of the
corresponding oxides and evolution of the interlayer anion (e.g. as
2
CO from carbonate); if the cations in the original layers are
divalent and trivalent ones, and do not undergo any oxidation
process during calcination, then the mixture usually contains the
corresponding spinel, in addition to the oxide of the divalent
2
. Experimental
2.1. Sample preparation
Ã
All chemicals were from Panreac (PRS quality) and were used
Corresponding author. Fax: +34 923294574.
E-mail address: vrives@usal.es (V. Rives).
without any further purification. Two synthesis routes were used.
0
022-4596/$ - see front matter & 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.jssc.2009.03.005

ARTICLE IN PRESS
1594
M.E. P ´e rez-Bernal et al. / Journal of Solid State Chemistry 182 (2009) 1593–1601
2
+
3+
In method A, an aqueous solution A1 containing Ni and Al
The solids were calcined in an oven at 1200 1C for 1 h at a
heating rate of 5 1C/min. The calcined solids are named as
Ni Al /1200 and Ni Al (MI)/1200 and showed a mass loss of
2 1 2 1
chlorides (molar ratio 2/1; total cation concentration 2 M) was
slowly, dropwise, added into a basic solution A2 (of which volume
was twice that of the metal cations solution) at room temperature.
Solution A1 had been prepared by dissolving the required
38% in both cases with respect to the original solids.
amounts of NiCl
2
ꢀ 6H
2
O and AlCl
3
ꢀ 6H
2
O in distilled water at
room temperature under vigorous stirring; stirring was continued
for 40 min and the solution was filtered to remove any solid
residue. Solution A2 had been prepared by slow addition of the
2.2. Characterisation
Element chemical analysis for metals was carried out at
Servicio General de An a´ lisis Qu ı´ mico Aplicado (University of
Salamanca) by atomic absorption after dissolving the samples in
nitric acid, in a Mark-II ELL-240 instrument.
required amounts of NaHCO
was calculated for a molar Al /CO
amount of NaOH was that needed to neutralise HCO
3
and NaOH. The amount of NaHCO
3
3+
2ꢁ
3
ratio equal to 1, while the
ꢁ
3
species and
to obtain a solid with the hydrotalcite-type structure; the basic
solution was left for 45 min to recover room temperature before
solution A1 was added.
A greenish precipitate was formed immediately after addition
of the first drops of the cations solution. Addition extended along
The powder X-ray diffraction (PXRD) patterns were obtained in
a Siemens D-500 diffractometer using CuKa radiation (l ¼ 1.5450
˚
A) connected to a DACO-MP microprocessor using Diffract-AT
software. The applied power was 1200 W (40 kV and 30 mA). The
scan speed was 21 (2y)/min, and the crystalline phases were
4
h and the stirring rate was 400710 rpm (Heidolph RXR-Z051
identified using Eva (Graphics Evaluation Program) software, the
ASTM files (Joint Committee on Powder Diffraction Standards,
JCPDS) and literature data.
vertical stirrer). Once addition was complete the suspension was
further stirred at room temperature for 20 h; the solid was then
filtered in a B u¨ chner funnel (20 cm diameter) with a water
vacuum pump. The liquid was colourless with pH ¼ 9.8. The solid
was washed four times with portions of 150 mL of distilled water
and once more with 500 mL. In each washing step the solid was
maintained in contact with water for 30 min before connecting
the water vacuum pump. Washing liquids were colourless and pH
varied in the 9–9.8 range. After the fifth washing step air was
flowed through the cake for 2 h and the solid was spread on a glass
plate to dry at room temperature for 4 days, and then it was
manually ground in an agate mortar. The solid thus obtained has
Differential thermal analyses (DTA) were recorded in a DTA-7
model instrument from Perkin Elmer, and the thermogravimetric
analyses (TG) in a TGA7-HT model thermobalance also from
Perkin Elmer, in oxygen flow (30 mL/min). Both instruments were
connected to a personal computer through TAC 7/DX interfaces,
and the data were analysed using standard software (Pyris, Series
Thermal Analysis system, version 3.0) also provided by Perkin
Elmer. This software also allowed to determine the derivative TG
curves (DTG). The heating rate was 10 1C/min in both cases and
the final temperature reached was 830 1C (TG) and 1280 1C (DTA).
Temperature-programmed reduction (TPR) studies were car-
ried out in a TPR/TPD 2900 model instrument from Micromeritics,
been named as Ni
2 1
Al .
Method B consisted in addition of the aqueous solution of
2+
3+
cations Ni
and Al
to a basic solution containing NaOH,
equipped with
connected to a Olivetti mod. 300-28 acquisition data station.
The reducing agent was a 5% (vol) H /Ar mixture from L’Air
a W–Au thermal conductivity detector and
NaHCO , iso-octane, butanol and sodium dodecylsulfate (NaDDS),
3
which had been prepared by mixing the aqueous and the organic
components previously prepared separately [31]. Under these
conditions, the anionic surfactants formed spheroidal reverse
micelles in the organic solvent with the hydrophobic chains
pointing out into the oil phase. The aqueous phase was prepared
2
Liquide, circulated through the sample at a flow of 60 mL/min; a
heating rate of 10 1C/min and a sample mass of 50 mg were used,
in order to optimize the analysis [32]. The amount of hydrogen
consumed was determined upon integration of the areas of the
peaks, after calibrating the instrument with CuO (from Merck).
The FT-IR spectra were recorded in a M-1700 2B model
spectrometer from Perkin Elmer, using Spectrum V.200 software,
3
by dissolving 6 g (0.072 mol) of NaHCO and 20 g (0.52 mol) of
NaOH in 60 mL of bidistilled water. The organic phase was
prepared by slow addition of 1.4 g (0.005 mol) of NaDDS to
ꢁ1
100 mL of iso-octane also containing 2 mL of butanol. The mixture
in the 4000–400 cm range. The samples were prepared in KBr
discs (ca. 1 mg sample/300 mg KBr). Fifty scans were recorded at a
was vigorously stirred (600710 rpm) to avoid precipitation of
NaDDS in the organic medium.
The compound was prepared at 75 1C and the temperature was
controlled with a linear temperature heater from Selecta and a
ꢁ
1
nominal resolution of 4 cm to improve the signal-to-noise ratio.
The nitrogen adsorption–desorption isotherms of nitrogen at
ꢁ196 1C were recorded in an automatic volumetric instrument
from Micromeritics (model ASAP 2000). A portion of 200 mg of
sample was used, which was degassed in two steps: first at room
temperature for 2–3 h, after which the temperature was raised in
10 1C steps (maintained for 10–15 min) up to 110 1C at a residual
model 2MG Statop temperature controller, using
a K-type
thermocouple. The suspension was also stirred at 600710 rpm.
Once addition of the metal cations solution was started, a greenish
solid was formed, which disappeared on stirring, but finally a
dense green suspension was formed; the suspension was slowly
dropwise added and addition extended along 2 h.
The solid was separated by centrifugation at 3000 rpm for 1 h.
The solid showed a gel-like texture, and was washed to remove
by-products (mainly NaCl) and unreacted NaDDS. Washing was
carried out by heating the solid for 8 h under reflux conditions in a
ꢁ
5
pressure of 10 mmHg.
Porosity measurements were carried out in a model Poresizer
9320 mercury porosimeter also from Micromeritics. A portion of
0.2–0.25 g of sample was used, degassed at room temperature by
applying the following protocol: vacuum was slowly applied up to
a residual pressure of ca. 50 mmHg was reached, and was kept for
30 min; then the pressure was lowered to 0.25 mmHg and finally
below 0.05 mmHg, which was maintained for 4–6 h. For the
mercury intrusion measurements, the pressure was controlled by
two different systems: a low pressure one (from ca. 50 to
1500 mmHg) using synthetic air as pressure transducer, and a
high pressure one using oil, up to 2200 atm.
1
:1 water: ethanol mixture (total volume 200 mL) and then the
suspension was again centrifuged at 3000 rpm for 1 h. A second
washing step was applied, heating the solid under reflux with
100 mL of acetone for 8 h, and then centrifugation at 3000 rpm for
1
h was applied. The whole washing procedure (in water:ethanol
and in acetone) was applied once more. The solid finally obtained
was dried in open air at room temperature and hand ground in an
agate mortar to yield a fine powder. It was stored in a desicator
Colour of the samples was quantified by using a Konica Minolta
colorimeter (Chroma Meter CR 400 model), equipped with a
pulsed Xenon lamp. The data were collected by a personal
with anhydrous CaCl
2 2 1
. The sample was named as Ni Al (MI).

ARTICLE IN PRESS
M.E. P ´e rez-Bernal et al. / Journal of Solid State Chemistry 182 (2009) 1593–1601
1595
computer through a RS-232 interface, and were analysed with
Color Data CM-100 W Spectra Magic NX software. We used the
L*a*b* system (also known as CIELAB) as the colour space, where
L* stands for lightness (ranging from L* ¼ 0 for black to L* ¼ 100
for white), and a* and b* are chromaticity coordinates, where
negative values for a* indicate green and positive values indicate
red, while negative values for b* indicate blue and positive values
indicate yellow. Three measurements (with three twinkles each)
on three different spots were taken for each sample, averaging the
results. Samples were placed between flat colourless glasses on a
black sample holder. The instrument had been previously
calibrated with Minolta white reference CR-A 43 (XYZ parameters
0
.3133, 93.80, and 0.3194, respectively).
3. Results and discussion
3.1. Element chemical analysis
2 1 2 1
Fig. 1. Powder X-ray diffraction patterns of samples Ni Al and Ni Al (MI).
Element chemical analysis results for Ni and Al for both
compounds are given in Table 1; the Ni/Al molar ratios are also
given. The amount of sodium was below the detection level, and
only an almost negligible amount of sulphur (0.02% weight) was
to diffraction by planes (110) and (113), is well resolved and even a
new peak at higher diffraction angle is perfectly defined. The
found for sample Ni
2
Al
1
(MI). The Ni/Al molar ratios were similar
Al (MI),
in both cases, although ca. 10% larger for sample Ni
2
1
2 1
larger sharpness of the diffraction maxima for sample Ni Al (MI)
indicating a lower positive formal charge of the layers in this
last case.
can be roughly attributed to a larger crystallinity (larger crystallite
size) of this sample.
Lattice parameter c has been calculated from the position of
the first diffraction maximum at low diffraction angles. Assuming
a 3R packing of the layers, this maximum is due to diffraction by
planes (003), and its spacing corresponds to one third of lattice
3
.2. Powder X-ray diffraction
The PXRD patterns for both samples are included in Fig. 1. The
2 1
parameter c. To diminish uncertainity (specially for sample Ni Al ,
overall shape of these patterns corresponds to the expected nature
of the crystalline LDHs with the hydrotalcite-type structure, to
which all the signals recorded could be attributed [33]. The
maxima were recorded in almost coincident positions for both
samples. In both cases the signals recorded at low diffraction
angles, due to diffraction by basal planes, are symmetric and
rather sharp, while in the medium range of the diagram the
signals are broader and asymmetric. This is due to turbostratic
disorder in layers stacking, and these peaks are due to diffraction
by non-basal planes.
with broader diffraction maxima), however, the value of c has
been averaged from the values determined from spacings
measured for planes (003) and (006). This spacing depends on
˚
the width of the brucite-like layers, 4.8 A [34], interlayer
hydration, size of the interlayer anion and, for non-spherical
anions, its orientation within the interlayer space. Lattice para-
meter a has been determined from the spacing for the diffraction
maximum due to planes (110), recorded at ca.
2y ¼ 621,
a ¼ 2d(110) [2,35–37]. From these lattice parameters and assum-
ing a hexagonal cell, all diffraction maxima have been indexed, as
shown in Fig. 1 [38].
The peaks are sharper and more intense in the diagram of
2 1
sample Ni Al (MI), for which the doublet close to 2y 60–651, due
Lattice parameters c and a (included in Table 1) are similar to
the values reported in the literature for LDHs containing other
3
+
divalent cations and Al in the brucite-like layers and with
interlayer carbonate anions [39]. The values for parameter c are
coincident within experimental error, and the slight difference in
the values for parameter a can be rationalised on the basis of the
ionic radii of the layer cations; this parameter is related to the
average cation–cation distance in the layer, which depends on
their ionic radii. In an octahedral coordination the ionic radius of
Table 1
Element chemical analysis (weight percentage), lattice parameters a and c ( A˚ ),
crystallite size D ( A˚ ), TG data (%), molecular formula, TPR results and specific
surface area (mercury intrusion, SHg, and nitrogen adsorption, SBET) for the layered
precursors.
Parameter/sample
Ni
2
Al
1
2 1
Ni Al (MI)
2
+
3+
2+
˚
˚
Ni is 0.83 A, while for Al it is 0.675 A [40]; the larger Ni
content (larger Ni/Al molar ratio) for sample Ni Al (MI) would
Ni
Al
31.9
7.8
35.4
7.1
2
1
II
III
Ni /Al molar ratio
solid)
1.90
2.10
account for the larger value of lattice parameter a for this sample.
The crystallite sizes, calculated by applying the Scherrer formula
(
a
c
3.01
3.03
[
41,42] to diffractions due to planes (003) and (006), are also
included in Table 1. As expected from the sharper peaks for sample
Ni Al (MI), the crystallite size (measured in the direction
perpendicular to the layers) is 4.5 times larger in this sample
22.91
22.87
179
D
41
Mass loss 70–290 1C
Mass loss 290–830 1C
Mass loss 70–830 1C
Molecular formula
Water content per
formula
12.6
13.4
22.1
36.5
2
1
23.5
36.1
˚
than for sample Ni Al . From these values and taking 7.6 A as the
2 1
[Ni0,65Al0,35(OH)
0.74
2 3
](CO )
0,173 [Ni0,68Al0,32(OH)
0.71
2 3 0,16
](CO )
thickness of a brucite-like layer and an interlayer (actually, the
spacing for diffraction due to planes (003), one third of parameter
c), we can conclude that the average crystallite is formed by
TPR[H
2
2
]
exp/[H
]
2 theo
1.09
13
0.96
107
119
SHg (m /g)
stacking ca. 6 layers in sample Ni
Ni Al (MI).
2 1
Al , and 24 layers for sample
2
SBET (m /g)
o10
2
1

ARTICLE IN PRESS
1596
M.E. P ´e rez-Bernal et al. / Journal of Solid State Chemistry 182 (2009) 1593–1601
3.3. Thermal decomposition and solids formed thereof
almost negligible mass loss is observed. This has been attributed
[45] to removal of carbon dioxide from residual carbonate species.
As one of the aims of this study is the preparation of mixed
The DTA curve (inset in Fig. 2) also shows two minima,
oxides through thermal decomposition of these LDHs, we have
studied their thermal decomposition up to ca. 830 1C under
dynamic oxygen or nitrogen atmospheres. No significant differ-
ences were found between the results obtained under both
atmospheres, and so only the results obtained in oxygen will be
shown. The results from the TG analysis (mass loss percentages in
the three temperature ranges corresponding to the different
decomposition steps) are summarised in Table 1. The TG curves
for both samples are shown in Fig. 2. As the decomposition steps
are somewhat overlapped, the DTG curves (also shown in this
figure) permit an easier comparison of the processes for both
samples.
corresponding to two endothermic processes. The first minimum
is very broad, and should correspond to the two first mass losses
(physisorbed and interlayer water removal), while the second one,
with a rather sharp shape, should be associated to the dehydrox-
ylation and decarbonation processes.
2 1
Decomposition of sample Ni Al (MI) follows a similar pattern
at medium and high temperatures, although some differences are
evident in the low temperature region. So, the DTG curve clearly
shows a weak minimum below 110 1C which could be tentatively
attributed to removal of traces of acetone (boiling point 56 1C)
and/or isooctane (b.p. 99 1C) which could remain physisorbed on
the external surface of the crystallites. The mass loss responsible
for the DTG minimum below 300 1C should be due to removal of
interlayer water molecules, while the last loss (DTA minimum at
2 1
Decomposition of sample Ni Al takes place in two well
defined steps. According to data in the literature [43,44], the first
one should be due to removal of water weakly adsorbed
ca. 400 1C, same position as for sample Ni
dehydroxylation and decarbonation processes. Similar comments
as for sample Ni Al can be applied to the DTA curve (inset in Fig.
2). It should be noticed, however, that dehydration takes place at a
higher temperature for sample Ni Al (MI) than for sample Ni Al
2 1
Al ) should be due to
(physisorbed) on the external surface of the crystallites; a small
mass loss is observed up to 140–150 1C; such a process seems to
be enhanced at 275–290 1C, corresponding to removal of water
molecules from the interlayer, where it was bonded via hydrogen
bonds to layer hydroxyl groups and interlayer carbonate anions
2
1
2
1
2
1
,
probably because of the larger crystallinity of the former, which
would permit to stablish more and stronger hydrogen bonds
between water molecules and layer hydroxyl groups and inter-
layer carbonate anions.
The crystalline phases in the solid residue formed at 830 1C
(TG) or 1280 1C (DTA) were determined by PXRD, Fig. 3. The
[
33]. Unfortunately the gases evolved during thermal decomposi-
tion could not be analysed. Such a two steps process for water
removal can be easily identified in the DTG curve, with two
minima roughly corresponding with the two sections of the TG
curve showing slightly different slopes. The second decomposition
step, identified by the DTG minimum at ca. 400 1C, corresponds to
decomposition of the layers, i.e., water removal through con-
2 1
pattern for the 830 1C residue of sample Ni Al (MI) shows only
three rather weak diffraction maxima which positions coincide
with those for the rock salt structure of nickel oxide (JCPDS file
22–1189). For the DTA residue (1280 1C) of the same sample the
crystallinity of this oxide is enhanced, while other diffraction
2
densation of layer hydroxyl groups, and CO release from
interlayer carbonate anions. After this second effect no major
mass loss takes place even up to 830 1C, and only a continuous,
2
.5
.0
DTG NI AI (MI)
2
1
DTG Ni AI
2
1
100
0
0
DTA Ni AI1 (MI)
2
-
4
8
DTA Ni AI1
2
-
80
-2.5
-
12
200
400
600
Temperature (°C)
TG Ni AI (MI)
2
1
TG Ni AI
2
1
60
-5.0
200
400
Temperature (°C)
600
800
Fig. 2. Thermogravimetric (bottom, left y-axis) and differential thermogravimetric (top, right y-axis) diagrams of the samples. Inset: differential thermal analysis curves. All
analysis performed in oxygen.

ARTICLE IN PRESS
M.E. P ´e rez-Bernal et al. / Journal of Solid State Chemistry 182 (2009) 1593–1601
1597
maxima develop. Two maxima coinciding with those of NiO and
the new ones correspond to the spinel structure, NiAl (JCPDS
3.4. Temperature-programmed reduction
2 4
O
3+
file 10-0339). It is not clear, however, where the Al cations were
located in the residue formed at 830 1C, well dissolved in the NiO
structure, or forming an amorphous phase, undetectable by PXRD.
In the former case the signals due to NiO would have been
recorded at larger diffraction angles (because of the smaller ionic
The TPR curves for both samples are shown in Fig. 4. It should
be recalled that decomposition along the TPR experiments takes
place in a reducing atmosphere, while the TG and DTA analyses
were carried out in oxygen, so the processes taking place along
both sets of experiments should not necessarily occur at
coincident temperatures.
3+
2+
radius of Al
broadness of the diffraction signals for the sample calcined at
30 1C does not permit locating the maxima with enough accuracy
to conclude formation (or not) of the solid solution.
A similar behaviour is observed for sample Ni Al
with respect to Ni ); however, the relative
According to the literature [43,48], under the experimental
8
conditions here used carbonate is removed as CO before being
2
3
+
reduced to coke or methane [49] and Al cations are not reduced,
2
+
2
1
, with broad
while Ni is quantitatively reduced to the zero-valent state [50].
Hydrogen consumption stoichiometrically corresponds to one
peaks when calcined at 830 1C, sharpening when the calcination
temperature was raised to 1280 1C. However, in this case some
2
+
hydrogen molecule per Ni cation in the solid.
extremely weak maxima due to the NiAl
already observed for the sample calcined at 830 1C, and the overall
sharpness of the maxima recorded for the sample calcined at
2
O
4
spinel phase are
Upward deviation of the curve means hydrogen consumption
(i.e., lower hydrogen concentration in the gas exiting from the
sample chamber than flowing through the reference branch of the
TPR instrument), while a downwards deviation would eventually
stand for hydrogen enrichment by hydrogen released from the
sample. The curves have been normalised by referring them per
unit mass of nickel in the samples. The curve for sample Ni Al ,
1280 1C (and so the crystallinity of the sample) is larger than for
the calcined sample Ni Al (MI).
2
1
We should then conclude that calcination of these two samples
at 830 1C leads to formation of a solid with the rock salt structure
2
1
3+
of NiO, where Al cations could be dissolved forming a solid
solution, or forming an amorphous aluminum oxide phase
(
although the presence of an amorphous phase in a solid calcined
at 830 1C should be hardly expected). Calcination at 1280 1C leads
to formation of the NiAl spinel, in addition to NiO. The NiAl
spinel has [46] a partially inverse structure with
¼ 0.38, a value
very close to that (
2
O
4
2 4
O
l
l
¼ 0.33) corresponding to a random distribu-
tion of the cations in octahedral and tetrahedral holes.
The molecular formula of the samples, as determined from
element chemical analysis (to determine the composition of the
hydrotalcite-type layers) and assuming carbonate as the only
interlayer anion, and from the TG results (to determine the
interlayer water content) are also included in Table 1. The
maximum amount of interlayer water content can be calculated
assuming a close packing of the interlayer by water molecules and
carbonate anions [47]; the values are 1.48 and 1.52 water
2 1 2 1
molecules per molecular formula for samples Ni Al and Ni Al
(MI), respectively. The experimental values are ca. 50% of the
maximum ones, indicating that a more open structure is adopted,
probably to facilitate hydrogen bonding between the water
molecules and the layer hydroxyl groups and the interlayer
carbonate anions.
Fig. 4. Temperature-programmed reduction curves for sample Ni
2 1
Al (a) and
Ni Al (MI) (b). The deconvolution curves and the calculated curve are also shown.
2
1
Fig. 3. Powder X-ray diffraction patterns of the residues of the TG (830 1C) and DTA
(Dotted line: component curves; solid line: experimental data; dashed line: sum of
component lines).
(1280 1C) analyses.

ARTICLE IN PRESS
1598
M.E. P ´e rez-Bernal et al. / Journal of Solid State Chemistry 182 (2009) 1593–1601
Fig. 4(A), shows a maximum at 398 1C with a shoulder at 496 1C.
However, for sample Ni Al (MI), two weaker minima are recorded
2
1
at 267 and 398 1C in addition to a maximum at 460 1C and a
shoulder close to 515 1C. The origin of the minima remains
uncertain, although they can be tentatively attributed to release of
hydrogen from decomposition of physisorbed acetone or iso-
octane residues, a process also detected in the thermal analysis
studies (Fig. 2); the presence of these species can be also
concluded from the FT-IR spectrum of this sample (see below).
These two hydrogen release processes will not be further
considered.
A Gaussian-type model was applied for an approximate
deconvolution of the TPR curves. Two components were found
in both cases. The maxima are recorded at 398 and 496 1C for
sample Ni
area under the curve) are 45:55. Regarding sample Ni
also two components were obtained, with maxima at 460 and
12 1C (i.e., 60 1C higher than for sample Ni Al ), with relative
2
1
Al , and their relative intensities (by integrating the
2
Al (MI),
1
5
2
1
intensities 60:40. The shift in the positions of the maxima may
have a kinetics origin, due to the restricted flow of hydrogen (or
2 1 2 1
Fig. 5. FT-IR spectra of samples Ni Al and Ni Al (MI).
2+
water vapour formed upon Ni reduction) in the larger crystal-
lites of sample Ni Al (MI) than in the more ‘‘amorphous’’ sample
Ni Al . However, it should be recalled that only Ni cations are
2
1
ꢁ1
3
000 cm is due to the O–H stretching mode of interlayer water
molecules hydrogen-bonded to interlayer carbonate anions. Water
2+
2
1
able to be reduced along these experiments. It should be also
noticed that while the first process takes place in a rather narrow
temperature range (the width of the maximum at half height is
molecules are also responsible for the medium-weak band at
ꢁ1
1
640 cm , due to their angular deformation mode [33]. The
ꢁ1
negative doublet around 2200 cm is due to a miscancelation of
4
5 1C), it results much broader (width at half height 180 1C) for the
2
the mode due to atmospheric CO .
second one. The presence of two reduction processes probably
suggests that in both samples there are two structurally different
ꢁ1
3
The band at 1370 cm is due to the antisymmetric n mode of
ꢁ1
carbonate anions; the shoulder close to 1470 cm is due to the
2+
Ni species. In the original LDH structure the metal cations are
located in octahedral holes of the brucite-like layers, but the
sample is decomposed (and structurally modified) along the TPR
experiments, and so it would be hardly accepted that at 400 1C
decreased symmetry of this anion from D3h to C3v (or even C2v
)
upon hydrogen bonding with other species (water molecules and
hydroxyl groups from the brucite-type layers) in the interlayer.
This decreased symmetry is also responsible for IR-activation of
2+
under
octahedral holes. On the other hand, formation of the NiO and
NiAl spinel phases (which presence was concluded from PXRD
a
reducing atmosphere all Ni
species are still in
ꢁ1
the
n
1
mode, recorded as a weak shoulder at 1040 cm [33,51,52].
ꢁ1
The extremely weak bands just below 3000 cm are due to
2 4
O
the C–H stretching modes of organic species existing in the
laboratory atmosphere. However, in the spectrum of sample
analysis of the residues formed after calcination above 1000 1C)
cannot be discarded (or accepted) during the TPR analysis, despite
the heating conditions (atmosphere surrounding the sample)
were absolutely different. Although further insight in the actual
nature of the species being reduced would be desirable, we can
Ni
sample Ni
shows a band, absent in the spectrum of the latter, at 1220 cm
2
Al
1
(MI) they are markedly stronger than in the spectrum of
2
Al . Moreover, the spectrum of the former sample
1
ꢁ1
which can be ascribed to the antisymmetric S–O stretching mode
of sulphate species, in agreement with the low sulphur content
2
+
tentatively ascribe this behaviour well to the reduction of Ni
cations existing in the borders of the crystallites or further inside
(
0.02%) in this sample.
The bands at 600, 570 and 425 cm
2+
them, or to reduction of Ni cations existing in two structurally
different environments, octahedral and tetrahedral holes. More-
over, the relative concentration of the low temperature reducible
ꢁ1
are due to Ni–O–Al
deformation mode and Ni–O and Al–O stretching modes [33] and,
as the other bands, are much better defined in the spectrum of
species is larger for sample Ni
If the total hydrogen consumption is calculated, the Ni/H
molar ratio corresponds to 1.09 and 0.96, respectively, for samples
Ni Al and Ni Al (MI). These values are acceptably close to the
2 1 2 1
Al (MI) than for sample Ni Al .
2 1 2 1
sample Ni Al (MI) than in that for sample Ni Al .
2
3.6. Surface area and porosity
2
1
2
1
stoichiometric one (1.00), taking into account the accuracy usually
accepted for this technique.
Results from surface area and porosity analysis, both through
nitrogen adsorption and mercury intrusion (inset), for sample
2 1
Ni Al (MI), are shown in Fig. 6. The nitrogen adsorption capacity
3.5. FT-IR spectroscopy
of sample Ni Al was very low. The nitrogen adsorption curve
2
1
corresponds to type IV in the IUPAC classification [53] with a
narrow type H3 hysteresis loop [54], suggesting a mesoporous,
uniform structure of the pores. It should be noticed that the height
of the interlayer (populated by carbonate anions and water
molecules) does not permit inclusion of nitrogen molecules and
consequently the interlayer space (actually corresponding to
micropores) is not measured by this technique. The specific
surface area values, measured by the BET method from the
The FT-IR spectra of these two samples are shown in Fig. 5.
They are typical of the LDH structure with interlayer carbonate
anions. The noticeable larger sharpness of the low wavenumber
2 1
bands for sample Ni Al (MI) should be related to the larger
crystallinity of this sample, but the bands are roughly recorded in
the same positions in the spectra of both samples. The broad band
ꢁ1
at 3500–3450 cm is due to the O–H stretching modes of layer
hydroxyl groups and water molecules. The broadness arises from
the very different strengths of hydrogen bonding between these
species. The band is asymmetric and a broad shoulder around
2
nitrogen adsorption isotherms, were 119 and less than 10 m /g for
2 1 2 1
samples Ni Al (MI) and Ni Al , respectively. The values calculated
from the mercury intrusion experiments were roughly coincident,

ARTICLE IN PRESS
M.E. P ´e rez-Bernal et al. / Journal of Solid State Chemistry 182 (2009) 1593–1601
Table 2
1599
Lattice parameter a ( A˚ ), crystallite size D ( A˚ ), TPR results and specific surface area
(nitrogen adsorption) for the solids calcined at 1200 1C.
Parameter/sample
Ni
2
Al
1
/1200
Ni
2
Al
1
(MI)/1200
2 4
NiAl O
NiO
NiAl
2
O
4
NiO
a
4.189
350
8.040
330
4.189
450
8.061
340
D
TPR [H
2
2
]
exp/[H
2
]
theo
1.08
1.05
16.0
SBET (m /g)
o6
Fig. 6. Nitrogen adsorption–desorption isotherms for uncalcined and calcined
Ni Al (MI) sample. Inset: Pore size distribution curve for the uncalcined samples
2
1
as measured by mercury intrusion.
2
1
07 and 13 m /g, respectively. The difference between the values
for both samples indicates that the preparation method has an
outstanding influence not only on the crystallinity of the particles
and ordering of the interlayer species, but also on the texture
properties.
The pore size distribution curves, as measured from the
mercury intrusion experiments, are also included in Fig. 6; only
a sharp maximum (ca. 15 nm) is obtained for sample Ni
confirming the ordered pore structure of this sample.
2 1
Al (MI),
If we are dealing with non porous particles, we would expect
that as the size of the crystallites increase, the specific surface area
would decrease. However, the results here obtained show just the
2 1
opposite behaviour, i.e., sample Ni Al (MI), formed by larger,
˚
ordered crystallites (D ¼ 180 A) show a larger specific surface area
than sample Ni
to 15 nm in sample Ni
larger surface area development.
2
1
Al . The presence of mesopores with a radius close
2
1
Al (MI) would, however, account for the
3.7. Characterisation of the calcined solids
The PXRD patterns of the calcined solids are essentially
coincident with those reported in Fig. 3 for the DTA residues,
with diffraction maxima corresponding to NiO (JCPDS file
2
2 4
2–1189) and NiAl O (file 10–0339). The maxima are sharper
for the Ni Al precursor, indicating slightly larger crystallites in
2
1
this sample. These PXRD patterns have been indexed for a cubic
system and the values for lattice parameters a for both crystalline
phases are included in Table 2 [46]. Maxima due to diffraction by
planes (200) and (220) were used to determine the parameter for
the NiO crystallites, (111) being ignored because of its rather low
intensity and coincidence with the most intense diffraction of the
spinel phase. Maxima due to diffraction by planes (111), (311),
Fig. 7. Temperature-programmed reduction curves for calcined (1200 1C) sample
Ni Al (a) and Ni Al (MI) (b). The deconvolution curves and the calculated curve
2
1
2
1
are also shown. (Dotted line: component curves; solid line: experimental data;
dashed line: sum of component lines).
(400), (511), and (440) were used for the spinel phase, all being
very sharp and rather intense. The values reported in Table 2 for
the spinel phase are very close for both samples and also close to
correspond to the organic impurities adsorbed on the crystallites
surface. While for the uncalcined solids two maxima could be
detected (although the second one appeared as a broad shoulder
of the most intense one at lower temperature, Fig. 4), in this case
the curves recorded could be deconvoluted into three maxima, at
different positions (temperatures) and with different relative
that reported in the literature [55] for the partially inverse
˚
(
l
¼ 0.375) NiAl
2 4
O spinel, 8.046 A. Table 2 also includes the
average crystallite size of the calcined solids. This parameter
2 1
markedly increases for NiO formed after calcinations of Ni Al (MI)
at 1200 1C.
The TPR curves for the calcined samples are shown in Fig. 7.
The lack of ‘‘negative’’ effects for the calcined Ni Al (MI) sample
confirms that those recorded for the uncalcined samples did
intensities (10:70:20 for sample Ni
sample Ni Al (MI)/1200). Of course, of all them correspond to
reduction of Ni cations in different structural locations, and in
2 1
Al /1200, and 40:45:15 for
2
1
2
1
2
+

ARTICLE IN PRESS
1600
M.E. P ´e rez-Bernal et al. / Journal of Solid State Chemistry 182 (2009) 1593–1601
this case reduction occurs on the well stabilised structures of NiO
rock salt type) and NiAl (spinel). According to the PXRD data
above commented, the spinel shows a partially inverse structure,
Table 3
Colour parameters of the samples prepared.
(
2 4
O
Sample
L*
a*
b*
D
L*
D
a*
D
b*
DE
a*b*
2+
and so Ni occupy (probably in an almost random distribution,
for which
¼ 0.333) both octahedral and tetrahedral sites, while
in NiO only octahedral sites are occupied. Consequently, we might
l
Ni Al
56.76
65.96
45.13
54.56
ꢁ21.86
ꢁ14.29
ꢁ9.21
ꢁ1.1
ꢁ5.15
1.28
17.17
26.37
5.54
ꢁ21.61
ꢁ14.04
ꢁ8.96
0.76
3.29
3.14
9.33
27.61
30.05
10.99
24.70
2
2
1
1
Ni
Al
(MI)
2
+
Ni Al /1200
2
2
1
1
have up to three different structural environments for Ni
Ni
Al
(MI)/1200
ꢁ17.54
7.47
14.97
ꢁ17.29
cations: (i) in octahedral sites in the NiO structure, (ii) in
octahedral sites in the spinel structure, and (iii) in tetrahedral
sites in the spinel structure. Probably, the easiness with which
L* (luminosity), a* (red 4 0 4 green), b* (yellow 4 0 4 blue).
2+
Ni
species are reduced (in other words, the reduction
temperature) depends on their precise structural location, thus
accounting for the three reduction components determined.
From these curves it is evident that the preparation procedure
has a marked effect on the nature of the calcination products and
their reducibility. The maxima are recorded at higher tempera-
tures for the sample prepared in an organic medium. The slight
difference in the value of lattice parameter a for the spinel phase
could be due to a different degree of inversion of the spinel
after calcination, if the original sample had been prepared in an
organic medium.
It can be also observed that the preparation procedure has a
regular effect on the luminosity of the solids, this parameter
showing larger values for samples prepared in an organic medium
than for those prepared by conventional precipitation. A similar
behaviour is observed for parameter b* (negative for blue and
positive for yellow), the solids prepared in an organic medium
2
+
crystallites in both samples, thus changing the population of Ni
cations in octahedral and tetrahedral sites, with different ionic
showing larger
Db* values.
Finally, the parameters are closer for the precursor solids and
decrease upon calcination, although such a decrease is less
marked for samples prepared in an organic medium.
2+
3+
radii. The ionic radii of tetrahedrally coordinated Ni and Al are
˚
0
0
.69 and 0.53 A, while in octahedral coordination these values are
˚
.83 and 0.68 A, respectively [56].
2+
The larger the location of Ni cations in tetrahedral sites, the
larger the inversion degree of the structure and the larger the
value of lattice parameter a. The third component reduction peak
has roughly the same relative intensity in both cases, and then
could be tentatively ascribed to reduction of Ni in the octahedral
holes of the NiO phase. The intensity of the second reduction
4. Conclusions
Two different routes have been used to prepare Ni,Al LDHs
with almost coincident chemical compositions. Larger and more
ordered crystals were obtained following the inverse micelles
route, probably because within the hydrophylic core of the
micelles a slow—and thus better ordered—growing of the crystals
2+
maximum decreases on passing from sample Ni
sample Ni Al (MI)/1200, and could be then tentatively ascribed to
reduction of Ni cations in octahedral sites of the spinel phase.
For both phases the H /Ni ratio, Table 2, is equal to the expected
value (1.0) within experimental error, indicating total reduction of
2 1
Al /1200 to
2
1
2+
may take place. Moreover, if the micelles have
a narrow
distribution size, a homogeneous distribution of the inorganic
particles would exist, thus accounting for the homogeneous
interparticle mesopores determined by mercury intrusion.
Thermal decomposition leads to distorted NiO at 830 1C, finally
2
2+
the Ni species in the temperature range studied.
The calcined samples show an extremely low ability for
nitrogen adsorption in all the relative pressure range studied.
The specific surface areas values determined by the BET method
are given in Table 2. Although the specific surface area for sample
2 4
leading to a mixture of NiO ad partially inverse NiAl O spinel at
2
+
1280 1C; the existence of Ni cations in three different structural
environments is evidenced from the TPR measurements.
Ni
uncalcined sample, it is still larger than the value measured for
the calcined Ni Al sample. The isotherm for the former sample,
Fig. 6, belongs to type IV in the IUPAC classification with a type H3
hysteresis loop extending in the relative pressure range 0.45–1.0,
suggesting the presence of pores only in the mesoporosity range.
2
Al
1
(MI)/1200 is reduced to ca. 13% of the value for the
Regarding the colour coordinates of these solids, an enhance-
ment of the green colour is observed for the sample prepared in an
organic medium upon calcination; this sample also shows a larger
luminosity than that prepared by conventional coprecipitation.
2
1
Acknowledgments
Authors thank financial support from MEC (grant MAT2006-
0800-C02-01) and Junta de Castilla y Le o´ n (grant SA024A06). The
3.8. Colour study
1
assistance of Mr. A. Montero to obtain some of the experimental
results is also acknowledged.
Colour coordinate values (L*, a* and b*) for all samples (original
and calcined) studied are given in Table 3. The corrected
parameters after substracting the values corresponding to the
sample holder used as a reference, and the values for colour
difference calculated using the equation
References
[
1] V. Rives (Ed.), Layered Double Hydroxides: Present and Future, Nova Science
Publishers, Inc., New York, 2001.
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Eꢂ ¼
ꢂ
2
ꢂ
2
ꢂ
2
D
ð
D
L Þ þ ð
D
a Þ þ ð
D
b Þ
[
[
2] F. Cavani, F. Trifir o´ , A. Vaccari, Catal. Today 11 (1991) 173.
3] X. Duan, D.G. Evans (Eds.), Layered double hydroxides, Struct. Bonding 119
[
57,58] are also included in this table.
(2005).
Luminosity of the solids decreases after calcination (i.e., they
become darker). More positive values of Db* indicate that calcined
samples are more yellowish than the precursor solids. Regarding
parameter a* (red–green), the values measured are always
negative, without any evident regular change upon calcination.
So, the greener sample leads to a sample with a lesser negative
[4] F. Wypych, K.G. Satyanarayana (Eds.), Clay Surfaces: Fundamentals and
Applications, Elsevier, Amsterdam, 2004.
[
[
5] V. Rives, M.A. Ulibarri, Coord. Chem. Rev 181 (1999) 61.
6] A. Vaccari, Catal. Today 41 (1998) 53.
[7] D. Carriazo, C. Mart ı´ n, V. Rives, Eur. J. Inorg. Chem. (2006) 4608.
[
[
8] G. Centi, S. Perathoner, Microp. Mesopor. Mater. 170 (2008) 3.
9] S. Albertazi, F. Basile, A. Vaccari, Catalytic properties of hydrotalcite-type
anionic clays, in: F. Wypych, K.G. Satyanarayana (Eds.), Clay Surfaces:
Fundamentals and Applications, Elsevier, Amsterdam, 2004 (Chapter 17).
D
a* value after calcination, while the green colour is enhanced

ARTICLE IN PRESS
M.E. P ´e rez-Bernal et al. / Journal of Solid State Chemistry 182 (2009) 1593–1601
1601
[
[
[
[
10] J.H. Yang, Y.S. Han, M. Park, T. Park, S.J. Hwang, J.H. Choy, Chem. Mater. 19
2007) 2679.
11] M. del Arco, E. Cebadera, S. Guti e´ rrez, C. Mart ı´ n, M.J. Montero, V. Rives,
J. Rocha, M.A. Sevilla, J. Pharm. Sci. 93 (2004) 1649.
[36] M.A. Ulibarri, F.M. Labajos, V. Rives, R. Trujillano, W. Kagunya, W. Jones, Inorg.
Chem. 33 (1994) 2592.
[37] M.A. Ulibarri, F.M. Labajos, V. Rives, R. Trujillano, W. Kagunya, W. Jones, Mol.
Cryst. Liq. Cryst. 167 (1994) 244.
[38] P. Gay, The Crystalline State. An Introduction,, third ed, Oliver & Boyd,
Edinburgh, 1972.
[39] V.A. Drits, A.S. Bookin, Crystal structure and X-ray identification of layered
double hydroxides, in: V. Rives (Ed.), Layered Double Hydroxides: Present and
Future, Nova Science Publishers, Inc., New York, 2001 (Chapter 2).
(
12] J.H. Yang, S.Y. Lee, Y.S. Han, K.C. Park, J.H. Choy, Bull. Korean Chem. Soc. 24
(2003) 499.
13] L. Peroli, V. Ambrogi, B. Bertini, M. Ricci, M. Nochetti, L. Latterini, C. Rossi, Eur.
J. Pharm. Biopharm. 62 (2006) 185.
[
[
14] S.J. Choy, J.M. Oh, T. Park, J.H. Choy, J. Nanosci. Nanotech. 7 (2007) 4017.
15] N.N. Das, J. Konar, M.K. Mohanta, S.C. Srivastava, J. Colloid Interface Sci. 270
[40] W.W. Porterfield, Inorganic Chemistry.
Academic Press Inc., London, 1993, p. 82.
A Unified Approach, second ed,
(
2004) 1.
16] F. Kovanda, E. Kovacsova, D. Kolousek, Collect. Czech. Chem. Commun. 64
1999) 1517.
[
[41] H.P. Klug, L.E. Alexander, X-ray Diffraction Procedures, John Wiley and Sons,
Inc., New York, 1974.
(
[
[
[
17] N.K. Lazaridis, T.A. Pandi, K.A. Matis, Ind. Eng. Chem. Res. 43 (2004) 2209.
18] E. A´ lvarez-Ayuso, H.W. Nugteren, Water Res. 39 (2005) 2535.
[42] R. Jenkins, J.L. De Vries, in: Worked examples of X-ray Analysis, second ed.,
Phillips Technical Library, Macmillan Press Ltd., London, 1978.
[43] V. Rives, Study of layered double hydroxides by thermal methods, in: V. Rives
(Ed.), Layered Double Hydroxides: Present and Future, Nova Science Publish-
er, Inc., New York, 2001 (Chapter. 4).
19] C. Barriga, M. Gait a´ n, I. Pavlovic, M.A. Ulibarri, M.C. Hermos ı´ n, J. Cornejo,
J. Mater. Chem. 12 (2002) 1027.
20] F. Leroux, J.-P. Besse, Layered double hydroxide/polymer nanocomposites, in:
F. Wypych, K.G. Satyanarayana (Eds.), Clay Surfaces: Fundamentals and
Applications, Elsevier, Amsterdam, 2004 (Chapter. 16).
[
[44] Ts. Stanimirova, I. Vergilov, G. Kirov, N. Petrova, J. Mater. Sci. 34 (1999)
4153.
[
[
[
21] X. Wang, Q. Zhang, Polym. Int. 53 (2004) 698.
[45] T. Hibino, Y. Yamashita, K. Kosuge, A. Tsunashima, Clays Clay Miner. 43 (1995)
427.
[46] A.W. Wells, Qu ı´ mica Inorg a´ nica Estructural, Editorial Revert e´ , Barcelona,
1978.
[47] S.K. Yun, T.J. Pinnavaia, Chem. Mater. 7 (1995) 348.
[48] V. Rives, M.A. Ulibarri, A. Montero, Appl. Clay Sci. 10 (1995) 83.
[49] V. Rives, O. Prieto, A. Dubey, S. Kannan, J. Catal. 220 (2003) 161.
[50] M. Jitianu, M. Balasoiu, R. Marchidan, M. Zaharescu, D. Crisan, M. Craiu, Int.
J. Inorg. Mater. 2 (2000) 287.
22] R. Kaluskova, M. Novota, Z. Vymazal, Polym. Degrad. Stabl. 85 (2004) 903.
23] A. Roy, C. Forano, J.P. Besse, Layered double hydroxides: synthesis and post-
synthesis modifications, in: V. Rives (Ed.), Layered Double Hydroxides:
Present and Future, Nova Science Publishers, Inc., New York, 2001 (Chapter 1).
24] J. He, M. Wei, B. Li, Y. Kang, D.G. Evans, X. Duan, Preparation of layered double
hydroxides in: X. Duan, D.G. Evans, (Eds.), Layered Double Hydroxides,
Structure and Bonding, vol. 119, (2005) pp. 89–119.
[
[
25] P. Benito, I. Guinea, M. Herrero, F.M. Labajos, V. Rives, Z. Anorg. Allg. Chem.
6
33 (2007) 1815.
[51] F.M. Labajos, V. Rives, Inorg. Chem. 35 (1986) 5313.
[52] J.T. Kloprogge, R.L. Frost, Infrared and Raman spectroscopic studies of layered
double hydroxides (LDHs), in: V. Rives (Ed.), Layered Double Hydroxides:
Present and Future, Nova Science Publisher, Inc., New York, 2001 (Chapter 5).
[53] A.R. Paniego, An. Quim. 85 (1986) 386.
[
[
26] M. Herrero, P. Benito, F.M. Labajos, V. Rives, Catal. Today 128 (2007) 129.
27] J.M. Garc ı´ a-Garc ı´ a, M.E. P e´ rez-Bernal, R.J. Ruano-Casero, V. Rives, Solid State
Sci. 9 (2007) 1115.
28] I. Nebot-Diaz, M. Marchal, M. Ir u´ n, J.B. Carda, Nuevas tecnologias para el
sector ceramico/New Technologies for Ceramic Industry, Universitat Jaume I,
Castell o´ , 2000.
[
[54] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscou, R. Pierotti, J. Rouquerol,
T. Sieminiewska, Pure Appl. Chem. 57 (1985) 603.
[
29] M.E. P e´ rez-Bernal, R.J. Ruano-Casero, V. Rives, Ceramics Silikaty 48 (2004)
[55] C.N.R. Rao, J. Gopalakrishnan, New Directions in Solid State Chemistry, second
ed, Cambridge University Press, Cambridge, 1997.
[56] J.E. Huheey, E.A. Keiter, R.L. Keiter, Inorganic Chemistry. Principles of
Structure and Reactivity, fourth ed, Harper Collins College Publishers,
New York, 1993, p. 114.
[57] P. Escribano L o´ pez, J.B. Carda Castell o´ , E. Cordoncillo Cordoncillo, Enciclopedia
Cer a´ mica, Tomo I: Esmaltes y pigmentos cer a´ micos, Universidad Jaime I,
Castell o´ n, 2001.
1
45.
30] R.J. Ruano-Casero, M.E. P e´ rez-Bernal, V. Rives, Z. Anorg. Allg. Chem. 631
2005) 2142.
[
(
[
[
[
[
[
31] G. Hu, N. Wang, D. O’Hare, J. Davis, J. Mater. Chem 17 (2007) 2257.
32] P. Malet, A. Caballero, J. Chem. Soc. Faraday Trans. I 84 (1988) 2369.
33] S. Kannan, A. Narayanan, C.S. Swamy, J. Mater. Sci. 31 (1996) 2353.
34] M.A. Drezdon, Inorg. Chem. 27 (1988) 4628.
35] A.S. Bookin, V.I. Cherkashin, V.A. Drits, Clays Clay Miner. 41 (1993) 558.
[58] /http://www.konicaminolta.com/instruments/knowledge/index.htmlS.