Intercalation of anionic oxalato complexes into layered double hydroxides

Article abstract of DOI:10.1006/jssc.2000.8769

Intercalation compounds of layered double hydroxide (LDH), M(1-x)/(II)M(x)/(III)(OH)₂A(x/y) · nH₂O (with M(II)=Zn, Cu and M(III)=Al, Cr, Ga), with oxalato complexes of aluminium, gallium, chromium, copper, and beryllium, were obtaine

Full text of DOI:10.1006/jssc.2000.8769

KANNAN BALA CHRIS
DISC USED YES ! NO
JSSC 8769
DATE..............................
Journal of Solid State Chemistry 153, 301}309 (2000)
doi:10.1006/jssc.2000.8769, available online at http://www.idealibrary.com on
Intercalation of Anionic Oxalato Complexes into Layered
Double Hydroxides
V. Prevot, C. Forano, and J. P. Besse
Universite Blaise Pascal, Laboratoire des Materiaux Inorganiques, UPRES-A 6002, 63177 Aubiere Cedex, France
E-mail: forano@chimtp.univ-bpclermont.fr
Received February 10, 2000; in revised form May 5, 2000; accepted May 11, 2000; published online August 3, 2000
E
the metal composition of the intermediate and "nal mixed
Intercalation compounds of layered double hydroxide (LDH),
metallic oxides obtained under calcination;
II
M
MxIII(OH)2Ax/y nH O (with MII؍Zn, Cu and MIII؍Al, Cr,
E
'
2
1؊x
the reactivity of the metal, in its both intra- and interlayer
Ga), with oxalato complexes of aluminium, gallium, chromium,
copper, and beryllium, were obtained via anion-exchange pro-
cesses. Powder X-ray di4raction indicated that the intercalation
reactions were successful. The basal spacings measured after
intercalation are near 0.98 ؎ 0.02 nm, whatever the host matrix
composition. Studies by FTIR spectroscopy con5rmed the inter-
calation of the oxalato complex, too. In order to study the
thermal decomposition of the exchanged products, TGA-coupled
mass spectrometry was performed. ( 2000 Academic Press
forms; and
E
the crystallization of the mixed oxides, depending on the
nature of the intercalated complexes.
The complexation of metallic cations by the oxalato
ꢀ\
ligand (C O ) forms one of the larger series of anionic
ꢀ ꢃ
monomolecular complexes, either with a charged equal to
ꢄ\
ꢀ\
3-, M'''(C O ) , or 2-, M''(C O ) . Such complexes
ꢀ ꢃ
ꢀ ꢃ
ꢀ
ꢄ
Key Words: layered double hydroxydes; oxalato complexes;
intercalation chemistry; thermal decomposition; spinel oxides.
are structurally and chemically stable under a large range
of pH in aqueous solutions. This allows us to envisage their
intercalation in the basic LDHs host structure.
Moreover, the complexation properties of the oxalic acid
ꢀ\
and its anion C O toward metallic cations are used in the
so-called oxalate process. Metal oxalates are used as precur-
INTRODUCTION
ꢀ
ꢃ
The layered double hydroxides (LDHs) (or the so-called sors for the preparation of metallic oxide nanoparticles,
anionic clays) are lamellar compounds with a structure using controlled thermal decompositions. Such complexa-
derived from that of the brucite-like structure M''(OH) , tion prevents, at medium-range temperatures, the crystalli-
ꢀ
where trivalent metallic cations partially replace divalent zation of oxide particles. We recently extended this
ones with the consequence that a net positive charge ap- approach to the preparation of mixed oxides derived from
''
'''
pears on the layers, [M
M (OH) ]V>. Overall elec- oxalates containing LDHs (2) as did Traversa and
ꢀ
V
ꢁ\V
troneutrality is obtained by intercalation between the layers co-workers (3).
of anions solvated with water molecules in variable
A limited number of papers about intercalation of anionic
amounts, [A ) nH O]V\. The LDH generic formula is complexes in LDHs have appeared in the literature (4). Only
VꢂW
ꢀ
''
'''
then M
M (OH) A ) nH O abbreviated as M''M'''-A. one work (5) reported on oxalato complexes containing
ꢀ VꢂW
ꢀ
ꢁ\V
V
The main interest in such inorganic compounds arises from LDHs, concerning the thermal decomposition study of the
the basic or redox catalytic activities of their calcined deriv- intercalation of tris-oxalato ferrate into MgAl-. In this pa-
atives for a large series of organic reactions (low alcohol per, we investigate the possibility of producing "ne metal
synthesis, etheri"cation, condensation reaction, polymeriz- oxide particles from oxalato complexes containing LDHs.
ation) (1). According to their chemical formula, a great We prepared and characterized a series of LDH host layers,
variety of intralayer metals (M''M''') or interlayer species M''M'''-, with M''"Zn, Cu and M'''"Al, Cr, Ga. We
A can be combined.
then intercalated oxalato complexes of the metals
The intercalation of metallic complexes can be seen as (M'''"Al, Cr, Ga, Fe and M''"Cu, Be) in these host
a means to prepare precursors for mixed oxide catalysts and structures. Structural and spectroscopic characterizations
nanophase composites. It is possible under this process to as well as thermal analyses of the di!erent oxalato com-
tune
plexes containing LDHs are presented here.
301
0022-4596/00 $35.00
Copyright ( 2000 by Academic Press
All rights of reproduction in any form reserved.

302
PREVOT, FORANO, AND BESSE
TABLE 1
Chemical Compositions of the LDH Precursors
EXPERIMENTAL
Synthesis
LDH
Mꢀ>/Mꢄ> Cl\/Mꢄ> H O/Mꢄ>
Formula
ꢀ
The LDH precursors ZnAl-Cl, ZnGa-Cl, and ZnCr-Cl
with the ideal formulae [Zn Al(OH) ]Cl ) 2H O,
[Zn Ga(OH) ]Cl ) 2H O, and [Zn Cr(OH) ]Cl ) 2H O
ZnAl-Cl
ZnGa-Cl
ZnCr-Cl
CuCr-Cl
2.93
2.95
1.98
2.20
1.13
1.20
1.0
1.9
2.5
1.9
2.0
Zn Al(OH) Cl ) 1.9H O
ꢄ
ꢈ
ꢀ
ꢀꢅꢆ
ꢇꢅꢈ ꢁꢅꢁ
ꢀ
Zn Ga(OH) Cl ) 2.5H O
ꢀꢅꢆ ꢇꢅꢈ ꢁꢅꢀ
Zn Cr(OH) Cl ) 1.9H O
ꢀꢅꢉ ꢇꢅꢉ ꢁꢅꢉ
Cu Cr(OH) Cl ) 2.0H O
ꢀꢅꢀ ꢊꢅꢃ ꢁꢅꢁ
ꢀ
ꢄ
ꢈ
ꢀ
ꢀ
ꢇ
ꢀ
were prepared by coprecipitation at controlled pH as de-
scribed in the literature (6, 7). For the ZnGa-Cl LDHs, some
modi"cations were made to the synthesis proposed by Fuda
et al. (8). Mixed solutions of ZnCl and Ga(NO ) in the
ꢀ
1.1
ꢀ
ꢀ
ꢄ ꢄ
expected Znꢀ>/Gaꢄ> molar ratio, having total cation con-
centration of 1 M, were used, and the pH was maintained analyses (Zn, Al, Ga, Cr, Cu, Cl, H, C) were performed in the
during the coprecipitation at pH 8.0 by simultaneous addi- Vernaison Analysis Center of the CNRS.
tion of 1 M NaOH. The precursor CuCr-Cl
([Cu Cr(OH) ]Cl ) 2H O) was obtained by the hydroly-
sis/precipitation method based on the slow addition of 1 M
RESULTS AND DISCUSSION
ꢀ
ꢇ
ꢀ
CrCl in an aqueous suspension of CuO (9). The slow
dissolution of divalent oxide induces a controlled release of
Intercalation Study
ꢄ
divalent metal species and the formation of the LDH phase.
Chemical analyses and empirical formulae of the four pre-
cursors are given in Table 1.
Characterization of the LDH Precursors
Di!ractograms of the four LDH precursors are displayed
Potassium tris-oxalato aluminate, ferriate, chromiate, in Fig. 1. All di!raction lines were identi"ed in the hexag-
galliate [K Al(C O ) ] ) 3H O, [K Fe(C O ) ] ) 3H O, onal lattice with the rhombohedral symmetry R3m. The
ꢄ
ꢀ ꢃ ꢄ
ꢀ
ꢄ
ꢀ ꢃ ꢄ
ꢀ
[K Cr(C O ) ] ) 3H O, [K Ga(C O ) ] ) 3H O, and potas- re"ned cell parameters are listed in Table 2. Whatever the
ꢀ ꢃ ꢄ ꢀ ꢃ ꢄ
sium di-oxalato cuprate and beryllate were prepared follow- preparation methods used, either coprecipitation for ZnAl-
ꢄ
ꢀ
ꢄ
ꢀ
ing a procedure from the literature (10). Cl, ZnGa-Cl, and ZnCr-Cl or hydrolysis/precipitation for
Intercalation compounds were obtained by anion ex- CuCr-Cl, well-crystallized materials were obtained. The
change on the M''M'''-Cl precursors. They were performed chemical analyses of the prepared samples are in good
in an aqueous solution containing the sodium salt of the agreement with that expected from the conditions "xed in
anionic complexes, in three molar excess over the anion- the synthesis.
exchange capacity. Two methods were used for the anion
The a cell parameter is equal to the shorter metal}metal
exchange. The standard exchange was carried out at room intralayer distance and is directly related to the size of the
temperature, for 1 day, under nitrogen to prevent contami-
nation by carbonate from atmospheric CO . The second
ꢀ
method was performed in an autoclave at 1203C, under an
autogenous pressure of 2.4 bar, for 24 h. The powdered
products were recovered, in all cases, after three washings,
with carbonate-free deionized water and a drying at room
temperature. The compounds exchanged are henceforth
noted as M''M'''-MꢀOx , while those obtained under hy-
%ꢀ#
drothermal treatment are labeled as M''M'''-MꢀOx .
!ꢁ
Physical Measurements
Powder X-ray di!raction (PXRD) patterns were obtained
with a Siemens D501 X-ray di!ractometer using CuKa
radiation and "tted with a graphite back-end mono-
chromator. Fourier transform infrared (FTIR) spectra were
recorded on a Perkin}Elmer 16PC spectrophotometer on
pressed KBr pellets. Thermogravimetric analyses (TGA)
were performed with a Setaram TG-DTA92 thermo-
gravimetric analyzer at a typical rate of 53C/min under an
air atmosphere. The mass spectrometer used in this study is
the Thermostar 300 from Balzers Instruments. Chemical
FIG. 1. X-ray di!ractograms of the LDH precursors.

INTERCALATION COMPOUNDS OF LDH
303
TABLE 2
the oxalato anions (11) on the FTIR spectra (Fig. 2). The
X-ray patterns of the various oxalato complexes containing
LDHs are displayed in Fig. 3.
Re5ned Cell Parameters of LDH Precursors
LDHs
a (nm)
c (nm)
d (nm)
ꢄ\
In the cases of the stable tris-oxalato complexes AlOx
,
ꢄ
ꢄ\
ꢄ\
CrOx , and FeOx , the pure LDH phases are prepared.
ZnAl-Cl
ZnGa-Cl
ZnCr-Cl
CuCr-Cl
0.3086
0.3121
0.3111
0.3111
2.357
2.357
2.363
2.309
0.786
0.786
0.788
0.770
ꢄ
ꢄ
ꢄ\
The presence of all the vibrations bands of FeOx on the
ꢄ\
FTIR of the intercalated compounds con"rms that FeOx
ꢄ
ꢄ
has not undergone photodissociation, as is the case in aque-
ous solution. It is noteworthy that the exchange reactions of
ꢀ\
ꢄ\
CuOx and GaOx do not lead to pure compounds. In
ꢀ
ꢄ
ꢀ\
the case of the CuOx intercalation the di!raction lines of
cations and the chemical composition of the sheets. With either moolooite, Cu(C O ), or Cu(OH) appear always
ꢀ
ꢀ ꢃ
ꢀ
a nearly identical Mꢀ>/Mꢄ> ratio for ZnCr-Cl (1.98) and beside the LDH diagram, whatever the exchange conditions
CuCr-Cl (2.20 (Table 1) and a similar ionic radius for Znꢀ> used. This is explained by the lowest stability of the copper
(0.74 A) and Cuꢀ> (0.72 A), both phases displayed the same oxalato anion which tends to lose one oxalate ligand. The
a value (0.311 nm). The metal}metal distance in ZnM-Cl formation of the stable copper oxalate byproduct with
(M"Alꢄ>, Gaꢄ>), for an equal Mꢀ>/Mꢄ> ratio (2.93 and CuCr-AlOx and CuCr-CrOx shows the high reactivity of
2.95), increases with the ionic radius of the trivalent metal the CuCr- matrix toward the various oxalato complexes
(respectively 0.56 A and 0.62 A for Alꢄ> and Gaꢄ>). In all involving decomplexation of the anion. This high reactivity
cases, the basal spacing is typical of LDH containing inter- of the CuCr- layers has already been reported in a paper (12)
layered chloride anions with slight variations of the on the grafting of organic anions in LDHs and is explained
hydration state. Absence of contamination by carbonate, by the presence of two OH labile ligands around the dis-
usually observed for the MgAl- system, is evidenced on the torted dꢆ Cuꢀ> ions. Surprisingly, the ZnAl- matrix, which
FTIR spectra of all precursors.
is generally stable under standard exchange reactions,
undergoes partial hydrolysis by the oxalato gallium com-
plex, even in mild exchange conditions; the undesired
Intercalation of the Oxalato Complexes
Zn(C O ) phase is still formed. No explanation has been
ꢀ ꢃ
ꢄ\
The exchange of the anionic oxalato complexes, AlOx
,
proposed so far.
ꢄ
ꢄ\
ꢄ\
ꢄ\
ꢀ\
CrOx , GaOx , FeOx , and CuOx , was obtained
The X-ray patterns of the various oxalato complexes
ꢄ
ꢄ
ꢄ
ꢀ
for the four LDH matrices. The intercalation of the com- containing LDH display strong similarities. Unfortunately,
plexes is con"rmed by the presence of the vibration bands of the low number of observed di!raction lines prevents the
FIG. 2. FTIR spectra of (a) AlOx containing LDHs and (b) ZnAl LDH intercalated by oxalato complexes.

304
PREVOT, FORANO, AND BESSE
FIG. 3. X-ray di!ractograms of LDHs containing (a) AlOx, (b) CrOx, (c) CuOx, and (d) GaOx and FeOx.
cell parameters re"nement which was only realistic for
The basal spacing for the ion-exchange reaction products
ZnAl-AlOx (a"0.3081(2) nm and c"2.973(3) nm). The ba- is directly related to the orientation of the intercalated
ꢄ\
sal spacing for the other phases was determined from the complexes. The four oxalato complexes MOx are nearly
ꢄ
mean value of the "rst two (001) peaks (Table 3). The the same size and display the pseudo-octahedral symmetry
interlayer distances are short and in the same order range (D3d) (Fig. 4a). The projection of the molecules in the plane
(0.967}0.997 nm). Values are lower than the one measured perpendicular to its C3 axis shows a building block with six
by Carlino and co-workers (6) for MgAl-FeOx (1.052 nm), oxygen atoms in the front plane and six other in the back
but the authors reported for this last compound a co-inter- plane. Such a geometry can easily "t in the interlayer oc-
ꢀ\
calation with CO anion which explains, in our opinion, tahedral sites formed by 3#3 OH of two adjacent layers
ꢄ
the higher basal spacing.
(Fig. 4b). In such an orientation the complex with its C3 axis

INTERCALATION COMPOUNDS OF LDH
305
TABLE 3
Basal Spacing of the Oxalato Complexes Containing LDHs
LDH
AlOxꢄ\ CrOxꢄ\ GaOxꢄ\ FeOxꢄ\ CuOxꢀ\
ꢄ
ꢄ
ꢄ
ꢄ
ꢀ
ZnAl-Cl
ZnGa-Cl
ZnCr-Cl
CuCr-Cl
0.991
}
0.984
0.991
0.989
}
0.991
0.967
0.983
0.969
}
}
}
0.991
}
0.997
0.972
0.976
0.973
}
FIG. 5. Ideal distribution of MOxꢄ\ pillars in the LDH interlayer
ꢄ
domains.
parallel to the c direction of the layered structure can
interact with the host structure with a maximum number of
hydrogen bonds. A theoretical interlayer distance based on les. The orientation of the complex between the plane is
the formula d"d #2d #l can be calculated probably analogous to the former cases. A structural di!er-
ꢂ!ꢃ ꢂ!ꢃꢂ!ꢄ*ꢅꢄ !ꢄ*ꢅꢄ
(13, 14), where d is the thickness of the octahedral layers ence is observed for the tetrahedral BeOx
ꢂ!ꢃ
ꢀ\
anion. The
ꢀ\
is the distance between the adjacent oxy- shorter observed basal spacing (0.907 nm) for ZnAl-BeOx
ꢀ
(0.21 nm), d
ꢂ!ꢃꢂ!ꢄ*ꢅꢄ
gen planes of both the layer and the oxalate, and l
ꢀ
can only be explained by orientation of the tetrahedral
(0.21 nm) is the size of the pseudo-octahedral anion along anion with its C2 axis parallel to the c direction.
the C3 axis. With a mean experimental d value of 0.99 nm, A common feature on all the X-ray patterns is that the
!ꢄ*ꢅꢄ
we calculate a guest}host distance of 0.285 nm characteristic second (001) di!raction line becomes more intense than the
of strong hydrogen interactions. The structure can then be "rst one. This is typical of an increase of the electron density
ꢄ\
described as a pillared layer structure (Fig. 5) where MOx
in the midplane of the interlayers due to the presence of the
pillars are distributed in the interlayer approximately heavy metals. As expected, the inversion is not observed for
ꢀ\
ꢄ
0.9 nm apart from each other and strongly attached to the the BeOx -containing phase.
ꢀ
layers, even if they are not covalently bonded.
The vibration bands of the oxalato anions are very sensi-
The LDH phases containing the copper oxalato anion tive to the interactions with the host layers, as shown on the
display similar basal spacings as the ones observed for the IR spectra (Figs. 2 and 3) and as mentioned in a previous
ꢄ\
M''M'''-MOx
ꢄ\
systems (Table 3). Edwards and co- study on intercalation of AlOx
in ZnAl LDH (2). The
ꢄ
ꢄ
workers (15) reported, on the basis of IR and Raman data, asymmetric C}O vibrations of the ligand appears parti-
that the structure of this anion is based on the formula cularly a!ected, with values between 1600 and 1800 cm\ꢁ.
[CuOx (H O) ]ꢀ\. It has a square-planar structure of two We observed a change in the relative intensities of the
ꢀ
ꢀ
ꢀ
oxalate ligands around Cuꢀ> plus two axial water molecu- asymmetric C"O vibrations, l and l . The l band is the
ꢋ
ꢁ
ꢁ
FIG. 4. Structural models of (a) the oxalato complexes (black and gray circles correspond respectively to front and back atoms) and (b) the
exchanged LDH.

306
PREVOT, FORANO, AND BESSE
FIG. 6. DTG curves for LDH precursors.
most intense for the intercalated anions and shifts from C6 axes parallel to the stacking direction. The oxalato
1680 cm\ꢁ for the tris-oxalato aluminium salt to 1762, 1715, complexes containing LDHs prepared in this study display
and 1710 cm\ꢁ, respectively, for CuCr-, ZnCr-, and ZnAl- a very similar structure.
host layers. The stronger the hydrogen interactions between
oxygen of C O and OH planes are, the higher the shift to
ꢀ ꢃ
a low-energy value is. The symmetric stretching band,
Thermal Behavior of the Precursors and Exchanged Phases
Thermogravimetry Analysis of the Precursors
l (C"O), remains at nearly the same position. On the other
hand, the intercalation leads also to a great broadening of
ꢆ
the lattice vibrations (400}700 cm\ꢁ) of the LDH matrix
The thermal decompositions of the four LDH precursors
compared to the spectra of the precursors. This is due to the are plotted as DTG curves in Fig. 6a and 6b. Quantitative
compact stacking of the layers induced by the strong inter- data processing of these curves is listed in Table 4. In all
actions with the interlamellar species.
In the cyanide complexes (Fe(CN)
cases, two steps of dehydration are observed: the dehydra-
tion of water molecules, physisorbed at the external surface
ꢃ\
ꢄ\
,
,
Fe(CN)
ꢇ
ꢇ
ꢄ\
ꢄ
Co(CN) , Mo(CN) ) the metallic cations displays an oc- of the crystallites, and the dehydration of intercalated water
ꢇ
ꢇ
tahedral symmetry, the complexes have a larger size com- molecules more strongly attached to the network via
pared to the oxalato complexes. Their intercalation in stronger hydrogen bonds. The temperature ranges and the
MgAl- (16}18) results in an increase of the basal spacing temperatures of maximal decomposition rate of these two
from 0.78 nm for the LDH precursor to 1.07 nm for the events are quite similar for all precursors except for ZnCr-Cl
intercalated products. Kikkawa and co-workers (15) con- which allows de-intercalation of bulk water molecules at
cluded that the anionic complexes were oriented with their very low temperature (1173C). The dehydration processes
TABLE 4
Thermal Decomposition Steps of the LDH Precursors
Dehydration
Dehydroxylation
LDH
Step 1?
Step 2?
Step 1?
Step 2?
ZnAl-Cl
ZnGa-Cl
ZnCr-Cl
CuCr-Cl
20-85/50
20-80/48
20-97/77
20-84/55
85-175/144
80-187/145
97-175/117
84-157/135
175-300/217
187-230/216
237-350/304
157-222/194
}
230-310/256
}
222-345/253
?T1-T2/T3 with T1-T2, decomposition thermal range; T3, temperature of maximal decomposition rate.

INTERCALATION COMPOUNDS OF LDH
307
TABLE 5
about 803C from the least stable to the highest stable
compound.
Temperature of Crystallization of Derived Oxides
At temperatures higher than 4003C, the chloride anion is
evolved following various mechanisms according the
matrix, involving either evaporation of HCl or sublimation
of metal chloride. However, the decomposition of CuCr-Cl
involves an additional step related to the formation of the
intermediate Cu OCl , while for ZnCr-Cl, ZnAl-Cl, and
ZnGa-Cl the continuous loss arises from the contribution of
the slow sublimation of zinc chloride. In the temperature
range of 400 to 6003C, a mixture of amorphous but reactive
Crystallization temperature (3C)
LDHs
MO
M''M'''O
ꢃ
ꢀ
ZnAl-Cl
ZnGa-Cl
ZnCr-Cl
CuCr-Cl
240
}
580
}
ꢀ
ꢀ
340
250
340
300
depend greatly on the nature of the metallic cations. The divalent and prespinel metal oxides of catalytic interest (19,
thermal stability of the phases increases with the temper- 20) is formed. Crystallization of the derived oxides, MO and
'''
ature at which the dehydroxylation starts in the order M''M O spinel phases, begins above 6003C at di!erent
ꢃ
ꢀ
CuCr-Cl(ZnAl-Cl(ZnGa-Cl(ZnCr-Cl, with a shift of temperatures (Table 5) according the matrix.
FIG. 7. TG/SM curves of (a and b) precursor salts, (c and d) AlOx, and (e and f) CrOx-containing LDHs.

308
PREVOT, FORANO, AND BESSE
TABLE 6
Spinel Phases Obtained from Calcination of Oxalato Complexes Containing LDHs
LDH
AlOxꢄ\
CrOxꢄ\
GaOxꢄ\
FeOxꢄ\
CuOxꢀ\
ꢄ
ꢄ
ꢄ
ꢄ
ꢀ
ZnAl-
ZnAl O
ZnAl Cr
O
ZnAl Ga
O
}
}
ZnAl O #CuAl O
ꢀ
ꢃ
ꢁꢅꢊ ꢉꢅꢊ
}
ꢃ
ꢁꢅꢊ ꢉꢅꢊ
ZnGa O
ꢃ
ꢀ
ꢃ
ꢀ
ꢃ
ZnGa-
}
}
ꢀ
ꢃ
ZnCr-
CuCr-
ZnCr Al
ꢁꢅꢊ ꢉꢅꢊ
CuCr Al
ꢁꢅꢊ ꢉꢅꢊ
O
O
ZnCr O
}
}
ZnCr Fe
CuCr Fe
O
O
ZnGa O #CuAlO
ꢀ ꢃ
ꢃ
ꢀ
ꢃ
ꢁꢅꢊ ꢉꢅꢊ
ꢁꢅꢊ ꢉꢅꢊ
ꢃ
ꢃ
CuCr O
CuCr O
ꢃ
ꢀ
ꢃ
ꢃ
ꢀ ꢃ
X-ray di!raction shows that a single spinel phase is ob-
tained, corresponding to solid solution between
''' ꢀ'''
composition leads to 33% more spinel in the oxide mixture.
Solid solutions of spinel in combination with two divalent
or two trivalent metals are then obtained.
Thermogravimetry Analysis Coupled with Mass
Spectrometry of the Oxalato Complexes Containing LDHs
a
M''M O ) and M''M O ). When Mꢀ'''"M''', the de-
ꢃ
ꢃ
Under air atmosphere, the hydrated potassium oxalato
metal salts decompose in three steps following Eqs. [1], [2],
and [3]:
ꢀ
ꢀ
If one compares the thermograms of the oxalato ex-
change products and the precursors, the major di!erence
arises in the temperature range from 200 to 4003C. While for
the precursor the dehydroxylation and the loss of chloride
are two well-separated thermal events, in the exchange
phases both decomposition processes occur jointly. For the
ZnAl- matrix, the decomposition occurs at 603C higher than
the dehydroxylation step in the ZnAl-Cl precursor. Because
K M'''(C O ) ) 3H OPK M'''(C O ) #3H O [1]
ꢄ
ꢀ ꢃ ꢄ
ꢀ
ꢄ
ꢀ ꢃ ꢄ
ꢀ
ꢄ
ꢄ
ꢁ
K M'''(C O ) # O P K CO # M (CO ) #3CO
ꢄ
ꢀ ꢃ ꢄ
ꢀ
ꢀ
ꢄ
ꢀ
ꢄ ꢄ
ꢀ
ꢀ
ꢀ
ꢀ
[2]
ꢁ
ꢁ
ꢄ
M (CO ) P M O # CO
[3]
ꢀ
ꢄ ꢄ ꢀ ꢄ
ꢀ
ꢀ
ꢀ
ꢀ
Figure 7a and 7b display the thermograms of of the higher stability of the ZnCr- network (3003C), the
K AlOx ) 3H O and K CrOx ) 3H O coupled with the dehydroxylation meets the complex decomposition. In the
ꢄ
ꢄ
ꢀ
ꢄ
ꢄ
ꢀ
mass spectrometry signals corresponding to masses 18 and case of ZnAl- this delay of dehydroxylation is due to the
44, respectively, H O and CO . The decomposition of the stabilization of the matrix by the oxalato pillars through
ꢀ
ꢀ
oxalate ligands occurs in two separate events for the former strong hydrogen interactions.
compound, between 350 and 5003C, while for the latter
compound the decomposition begins at lower temperature
(3203C) following a single continuous thermal phenomenon.
When [AlOx ]ꢄ\ and [CrOx ]ꢄ\ are intercalated in
CONCLUSION
The intercalation of a large series of tris-oxalato and bis-
ꢄ
ꢄ
ZnAl- and ZnCr- LDHs, the decomposition of both com- oxalato metallates in four-layered double hydroxydes
plexes occurs simultaneously with the dehydroxylation of (ZnAl-, ZnCr-, ZnGa-, and CuCr-) has been obtained by
the host structures. When the mineral matrix is ZnCr-, anion-exchange reactions. The strong interactions between
evolution of both CO and H O occurs at the same temper- the anionic complexes and the host structures lead to pil-
ꢀ
ꢀ
ature (3003C), while for ZnAl-dehydroxylation begins at lared-like layered structures with basal spacings lower than
503C lower than the combustion of the oxalate ligands. In 1 nm and internal galleries with limited sizes. As shown by
this case the anion undergoes a decomposition in two steps, the TGA/SM analysis, the thermal decomposition of the
as observed for K AlOx ) 3H O. The decomposition is host structure is induced by the pyrolysis of the anionic
ꢄ
ꢄ
ꢀ
probably an oxidative pyrolysis induced by the water mol- complexes, at lower temperatures than for the chloride
ecules evolved under dehydroxylation. In the case of ZnCr- precursors. The chemical combination of the guest and host
the decomposition of the oxalato complexes is probably through the intercalation of oxalato complexes in LDHs is
assisted and increased by the oxidative action of the in- a very promising way to prepare precursors to obtain mix-
tralayer chromium cations. It then results in a one-step tures of "nely divided oxides with a desired composition of
reaction and does not involve intermediate carbonate the spinel phase. This study evidences a new means to
compounds.
prepare catalysts with tunable metal content.
The total decomposition of LDHs leads to a mixture of
divalent metal oxide and spinel phases. If the starting LDH
REFERENCES
''
'''
is the formula [M
M (OH) ][Mꢀ'''Ox ] ) nH O, then
ꢀ
ꢄ Vꢂꢄ
ꢀ
ꢁ\V
V
the composition of the mixed oxides obtained after calcina-
tion is (1!5x/3)M''O#(2x/3)M''(M
1. F. Tri"ro and A. Vacari, Comp. Supramol. Chem. 7, 251 (1996).
2. V. Prevot, C. Forano, and J. P. Besse, J. Mater. Chem. 8, 1 (1998).
''' ꢀ'''
M
)O (Table 6).
ꢃ
ꢄꢂꢀ ꢁꢂꢀ

INTERCALATION COMPOUNDS OF LDH
309
3. E. Traversa, P. Nunziante, and G. Chiozzini, ¹hermochim. Acta 199, 20 10. J. C. Bailar and E. M. Jones, &&Inorganic Synthesis'' (H. S. Booth, Ed.),
(1992).
vol. 1, p. 35. 1939.
4. (a) E. P. Giannelis, D. G. Nocera, and T. J. Pinnavaia, Inorg. Chem. 26,
203 (1987). (b) H. C. B. Hansen and C. B. Koch, Clays Clay Miner. 42,
11. K. Nakamoto, in &&Infrared and Raman Spectra of Inorganic and
Coordination Compounds.'' Wiley, New York, 1986.
170 (1994). (c) K. Okada, F. Matsushita, and S. Hayashi, Clay Miner. 12. V. Prevot, C. Forano, and J. P. Besse, Appl. Clay Sci. (in press).
32, 299 (1997).
13. V. Prevot, C. Forano, and J. P. Besse, Inorg. Chem. 37, 4293 (1998).
14. S. Bonnet, C. Forano, and J. P. Besse, Mater. Res. Bull. 33, 783
(1998).
5. S. Carlino and M. J. Hudson, Solid State Ionics 110, 153 (1998).
6. W. T. Reichle, Solid State Ionics 22, 135 (1986).
7. A. de Roy, C. Forano, K. El Malki, and J. P. Besse, in 00Synthesis of 15. H. G. M. Edwards, D. W. Farwell, S. J. Rose, and D. N. Smith, J. Mol.
Microporous Materials'' (L. Ocelli and H. Robson, Eds.), vol. 2, p. 108.
Struct. 249, 233 (1991).
Van Nostrand Reinhold, New York, 1992.
16. S. Kikkawa and M. Koizumi, Mater. Res. Bull. 17, 191 (1982).
8. K. Fuda, N. Kudo, S. Kawai, and T. Matsunaga, Chem. ¸ett. 777 17. S. Idemura, E. Suzuki, and Y. Ono, Clays Clay Miner. 37, 553 (1989).
(1993). 18. K.Itaya and H. C. Chang, Inorg. Chem. 26, 624 (1987).
9. K. El Malki, A. De Roy, and J. P. Besse, Eur. J. Solid State Inorg. Chem. 19. F. Cavani, F. Tri"ro, and A. Vaccari, Catal. ¹oday 11, 173 (1991).
26, 339 (1989). 20. A. Vaccari (Ed.), Appl. Clay Sci. 10 (1995).