Ab-initio crystal structure of hydroxy adipate of nickel and hydroxy subarate of nickel and cobalt from synchrotron powder diffraction and magnetic properties

Article abstract of DOI:10.1016/j.jssc.2008.06.059

Organic-inorganic hybrid compounds Ni(II)₅(OH)₆(C₆H₈O₄)₂ (1), Ni(II)₅(OH)₆(C₈H₁₂O₄)₂ (2) and Co(II)₅(OH)₆(C₈H₁₂O₄)₂ (3) have a similar layered structure as determined ab initio from synchrotron powder diffraction analysis. The metal sites are octahedrally coordinated by O atoms. The slabs are built from edge-sharing octahedra in such a way that channels with an average size of 4 A are formed. Bis-bidentate and bridging dicarboxylate anions lead to a 3D framework. The compounds (1) and (2) order antiferromagnetically below T_N=26.5 and 19.3 K, respectively, while (3) is ferrimagnetic with T_C=16.2 K. Crystal data for compounds are as follows: (1) a=11.6504(1) A, b=6.8021(3) A, c=6.3603(1) A, α=73.52(1)°, β=99.69(1)°, γ=96.16(1)°, R_B=0.070, 668 reflections; (2) a=13.9325(1) A, b=6.7893(1) A, c=6.3534(4) A, α=73.63(1)°, β=95.14(1)°, γ=91.80(1)°, R_B=0.052, 804 reflections; (3) a=13.9806(1) A, b=6.9588(1) A, c=6.3967(1) A, α=73.05(1)°, β=94.51(1)°, γ=92.19(1)°, R_B=0.048, 410 reflections. The space group is P-1 for the three compounds.

Full text of DOI:10.1016/j.jssc.2008.06.059

ARTICLE IN PRESS
Journal of Solid State Chemistry 181 (2008) 3229–3235
Contents lists available at ScienceDirect
Journal of Solid State Chemistry
journal homepage: www.elsevier.com/locate/jssc
Ab-initio crystal structure of hydroxy adipate of nickel and hydroxy
subarate of nickel and cobalt from synchrotron powder diffraction and
magnetic properties
Ã
Adel Mesbah a, Anne Carton a, Lionel Aranda a, Thomas Mazet a, Florence Porcher b, Michel Franc-ois a,
a Laboratoire de Chimie du Solide Min´eral, UMR 7555, Universite´ Henri Poincare´, F-54506 Vandoeuvre les Nancy, France
b Laboratoire de Cristallographie et de Mode´lisation des Mate´riaux Min´eraux et Biologiques, UMR UHP–CNRS no 7036, Universite´ Henri Poincare´, Nancy I,
54506 Vandoeuvre les Nancy, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Organic–inorganic hybrid compounds Ni(II)5(OH)6(C6H8O4)2 (1), Ni(II)5(OH)6(C8H12O4)2 (2) and
Co(II)5(OH)6(C8H12O4)2 (3) have a similar layered structure as determined ab initio from synchrotron
powder diffraction analysis. The metal sites are octahedrally coordinated by O atoms. The slabs are built
Received 22 February 2008
Received in revised form
27 June 2008
˚
from edge-sharing octahedra in such a way that channels with an average size of 4 A are formed. Bis-
Accepted 29 June 2008
Available online 23 August 2008
bidentate and bridging dicarboxylate anions lead to a 3D framework. The compounds (1) and (2) order
antiferromagnetically below TN ¼ 26.5 and 19.3 K, respectively, while (3) is ferrimagnetic with
Keywords:
˚
˚
TC ¼ 16.2 K. Crystal data for compounds are as follows: (1) a ¼ 11.6504(1) A, b ¼ 6.8021(3) A, c ¼
Hybrid compound
Crystal structure
Rietveld analysis
Magnetic properties
˚
6.3603(1) A,
˚
a
¼ 73.52(1)1,
b
¼ 99.69(1)1,
g
¼ 96.16(1)1, RB ¼ 0.070, 668 reflections; (2) a ¼ 13.9325(1)
˚
˚
A, b ¼ 6.7893(1) A, c ¼ 6.3534(4) A,
a
¼ 73.63(1)1,
b
¼ 95.14(1)1,
g
¼ 91.80(1)1, RB ¼ 0.052, 804 reflec-
¼ 94.51(1)1, ¼ 92.19(1)1,
˚
˚
˚
tions; (3) a ¼ 13.9806(1) A, b ¼ 6.9588(1) A, c ¼ 6.3967(1) A,
a
¼ 73.05(1)1,
b
g
RB ¼ 0.048, 410 reflections. The space group is Pꢀ1 for the three compounds.
& 2008 Elsevier Inc. All rights reserved.
1. Introduction
the covalent bonds between the organic and mineral network, it is
possible to imagine coupling between the properties of the two
sub-networks. For example, it was shown that in Ni(II)-layered
organic–inorganic compounds, the magnetic properties of the
layers of nickel hydroxide influence the optical properties of the
grafted luminescent molecules [9].
These compounds are synthesized in basic media to support
the presence of hydroxides. M is a divalent transition metal
element from the first series of the Periodic Table. The crystal
structures of many such compounds have been accurately
determined. In most cases the metal is located exclusively in an
octahedral site, coordinated by O [10–26]. The oxygen atoms
usually come from the hydroxide and/or carboxylate groups, but
in some cases they come from the water molecules. The octahedra
are connected via edge- or corner-sharing, which often leads to
the formation of 1D chains [10–15] or 2D planes [16–25]. In the
latter case, the planes can leave holes leading to porous materials
[20,25]. Less frequently, the connection of the octahedra may lead
to 3D networks [26].
Over the past two decades, hybrid metal-organic materials
have been intensively investigated, with a particular interest for
the magnetic, optical and electrical properties associated with the
various crystallographic structures. A recent review article surveys
the progress that has been made on this subject [1]. As these
metal-organic compounds often consist of more or less open or
porous networks, these materials also find application as catalysts
or for gas adsorption applications [2–5]. Among these compounds,
those based on carboxylates are of particular interest. Review
papers have been published that discuss the structural properties
and strategies of chemical synthesis [6], magnetic properties, etc.
[7]. Another one, focused especially on cobalt and nickel
carboxylates [8], is also available.
In this paper, we investigated hybrid compounds based on
dicarboxylate linkers for several reasons. These compounds can be
synthesized easily under mild conditions of temperature and
pressure and are thermally stable up to 400 1C. The dicarboxylates
tend to form highly crystalline products. This facilitates the
determination of their crystallographic structures more accurately
than in the corresponding monocarboxylates. Moreover, due to
This work attempts to contribute toward the development of
new compounds in the M(II)-OH/dicarboxylates system. The main
goal was to establish relations between the crystal structure and
magnetic properties in the entitled compounds. The magnetism of
the metal hydroxycarboxylates is relatively well understood [7,27]
for compounds where the metal hydroxide layers form a brucite-
like structure. In metal-hydroxy-dicarboxylates where the mineral
Ã
Corresponding author. Fax: +03 83 68 4611.
E-mail address: Michel.francois@lcsm.uhp-nancy.fr (M. Franc-ois).
0022-4596/$ - see front matter & 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.jssc.2008.06.059

ARTICLE IN PRESS
3230
A. Mesbah et al. / Journal of Solid State Chemistry 181 (2008) 3229–3235
sub-unit differs and does not adopt a brucite-like arrangement,
the magnetic behaviors are less well understood. For example, one
can find ferrimagnetic compounds with ‘chains’ [10,11,13] or
‘planes’ [20,24,25,28], ferromagnetic compounds with ‘planes’
[19], antiferromagnetic compounds with ‘chains’ [14,15] or canted
antiferromagnetics with ‘planes’ [16]. Conversely, the aforemen-
tioned compound with the 3D mineral network [26] remains
paramagnetic at low temperature. In this work, three new
compounds were synthesized by the hydrothermal route:
Ni(II)5(OH)6(C6H8O4)2, Ni(II)5(OH)6(C8H12O4)2 and Co(II)5
(OH)6(C8H12O4)2. Their crystal structures were determined from
synchrotron X-ray powder diffraction (XRPD) measurements and
their magnetic, optical and thermal properties are characterized.
2. Experimental
2.1. Synthesis, thermal and chemical analyses
The nickel hydroxy adipate Ni(II)5(OH)6(C6H8O4)2 (1) was
synthesized by the hydrothermal route from a mixture (2:3) of
Ni(NO3)2 ꢁ 2H2O (Aldrich, 98%) and adipic acid C6H8O4H2 (Aldrich,
98%) in aqueous solution, typically Ni(NO3)2 ꢁ 2H2O (1.50 g,
6.8 mmol), C6H8O4H2 (1.50 g, 10.3 mmol,). The pH of the solution
was increased up to 8 by the addition of NaOH (0.1 M). About
20 ml of the starting mixture was homogenized and transferred
into a 25 ml teflon-walled acid digestion bomb, and then heated
under autogenous pressure for 72 h at 150 1C. The reaction product
was collected by filtration, washed twice with a mixture of
distilled water and anhydrous ethanol (1/1) and then dried at
room temperature. The same procedure was applied for
Ni(II)5(OH)6(C8H12O4)2 (2) with subaric acid C8H12O4H2 (Aldrich,
98%) (1.79 g, 10.3 mmol), Ni(NO3)2 ꢁ 2H2O (1.50 g, 6.8 mmol) and
for Co(II)5(OH)6(C8H12O4)2 (3) with subaric acid C8H12O4H2
(1.79 g, 10.3 mmol), Co(NO3)2 ꢁ 2H2O (Aldrich, 98%) (1.50 g,
6.85 mmol).
Fig. 1. TGA curve of (1), (2) and (3).
frequency of 100 Hz and alternative magnetic field HAC ¼ 5 Oe. No
correction for diamagnetism was applied.
2.4. XRPD and ab-initio structure determination
XRPD data were collected at 100 K using synchrotron radiation
(ESRF, ID 31, the transmission Debye Scherrer geometry). The
diffractometer was equipped with a primary Si(111) double-
crystal monochromator and nine sensitive linear position detec-
tors with crystal analyzers [30]. The sample of a fine powder form
was introduced in a Lindeman tube (
recorded using a wave length of 0.85124 A, in the 2y range 4–601
F ¼ 0.8 mm). Data were
˚
with an interval of 0.0031 and a total counting time of 2 h. Crystal
Thermogravimetric (TG) measurements were performed with
a ‘TG/ATD 92–16.18’ SETARAM instrument between 20 and 600 1C
in air and using a heating rate of 11/min. The thermal curves for
(1), (2) and (3) are reported in Fig. 1. For each compound the
weight loss occurs in a single stage, at 320 1C for (1), and at 300 1C
for (2) and (3). It is ascribed to the transformation from (1) or (2)
to NiO (as determined by X-ray diffraction, PDF: 44–1159) (43.5%
obs; 45.5% calc (1), 48.5% obs; 47.5% calc (2)) and from (3) to CoO
(PDF: 75–0533) (45.3% obs; 45.3% calc (3)).
data and structure refinement parameters are reported in Table 1.
2.4.1. Indexing
Standard peak search methods with Reflex program from
Material Studio (MS) system software (Accelrys) [31] were used to
locate the diffraction maxima; indexing was performed with the
Xcell [32] program. For the three compounds, the solutions were
found in the triclinic system aP. The lattice parameters, presented
in Table 1, were refined by the Rietveld method.
Chemical analysis: (3) C (obs: 26.20%; calc 25.92%), Co (obs:
36.02%; calc 30.24%); H (obs: 4.08%; calc 4.05%).
2.4.2. Resolution
The three structures were solved in the space group Pꢀ1
applying optimization methods (parallel tempering) in the direct
space, using the FOX program [33]. On the basis of thermal and
chemical analyzes and expected density as well as UV–visible
results (see hereafter), the asymmetric unit was filled for each
model with three independent M(II)O6 octahedra and one linear
dicarboxylate molecule C6H8O4 for (1); C8H12O4 for (2) and
(3)—introduced as rigid bodies in the starting models without
their H atoms. The optimization led to initial models in agreement
with the chemical formulae Ni(II)5(OH)6(C6H8O4)2 for (1),
Ni(II)5(OH)6(C8H12O4)2 for (2) and Co(II)5(OH)6(C8H12O4)2 for (3).
All three structures contain three metallic sites (M1, M2 and M3)
and seven O sites (4 Ocarb and 3 OOH); (1) contains six, whereas (2)
and (3) contain eight C-sites, respectively. The M1 site is located in
the inversion center.
2.2. IR and UV spectroscopy
The IR spectrum was recorded with a ‘Spectrum one FT-IR’
spectrometer (Perking Elmer Instrument) in the ATR mode using
the ‘Universal Sampling Accessory’.
UV–visible spectra were performed for compounds (1), (2) and
(3) with a CARY 4000 spectrometer operating in the 175–900 nm
range.
2.3. Magnetic measurements
DC and AC magnetic susceptibility measurements were carried
out with a PPMS Quantum Design [29], between 5 and 300 K for
(1), (2) and (3). The wDC curves were recorded under a field of
10 kOe to determine the molar Curie constant (CM), the para-
magnetic Curie temperature (
The wAC dependences with temperature were collected using a
Structural models were refined by the Rietveld method using
the FULLPROF program [34]. A total of 46 and 52 intensity-
dependent parameters for (1) and (2), (3), respectively, including
yp) and the effective moment (meff).

ARTICLE IN PRESS
A. Mesbah et al. / Journal of Solid State Chemistry 181 (2008) 3229–3235
3231
Table 1
Crystal data and structure refinement parameters
Compound
(1) Ni5(OH)6(C6H8O4)2
(2) Ni5(OH)6(C8H12O4)2
(3) Co5(OH)6(C8H12O4)2
Fw (g/mol)
System
683.75
Triclinic Pꢀ1
11.6504(1)
6.8021(3)
6.3603(1)
73.52(1)
99.69(1)
96.16(1)
475.491(4)
1
739.84
Triclinic Pꢀ1
13.9325(1)
6.7893(1)
6.3534(4)
73.63(1)
95.14(1)
91.80(1)
574.288(6)
1
741.05
Triclinic Pꢀ1
13.9806(1)
6.9588(1)
6.3967(1)
73.05(1)
94.51(1)
92.19(1)
593.375(10)
1
˚
a (A)
˚
b (A)
˚
c (A)
a
b
g
(1)
(1)
(1)
3
˚
V (A )
Z
Colour
Dx (g/cm3)
Green
Green
2.137
Pink
2.387
2.0738
0.85124
2.53
˚
Wave length (A)
0.85124
3.87
0.85124
3.21
Absorption coefficient (
mꢃ r)
1.59–42.96
13789
1.71–42.96
13751
804
1.81–32.96
10384
2y range (deg)
N obs of points
Nref
668
410
Rp
0.069
0.065
0.045
Rwp
w2
0.114
0.094
0.073
5.003
5.97
5.248
RBragg
0.070
0.052
0.048
RF
0.091
0.097
0.083
N profile parameters
32
19
32
N intensity dependent parameters
N constraints
46
52
52
31
35
35
Excluded regions (1)
10.0–10.78
–
4.5–6.00
10.74–10.92
4
the atoms coordinates and an overall thermal displacement were
refined. The following soft geometrical constrains have been
applied on the organic molecules during the refinement:
500 nm corresponds to the transition T1g(F)-4A2g(F) in case
high-spin Co2+ is located in octahedral geometry.
˚
˚
C–C ¼ 1.54(1) A, C–O ¼ 1.26(1) A; C–C–C ¼ 110.0(2)1 (13 and 17
soft constrains for adipate and subarate molecules, respectively).
Additionally, for each structure, soft constrains on the three MO6
3.2. Structure description
˚
As the structures of (1), (2) and (3) are very close, only that of
(2) is presented in Fig. 5. All the structures are formed by layers
connected by dicarboxylates anions. They differ mainly either by
the nature of the metal cation or by the inter-layer distance d0
octahedra were also applied: Ni–O ¼ 2.10(1) A for (1) and (2), and
˚
Co–O ¼ 2.15(2) A for (3). Consequently, 31 (13+18) constraints
were used in (1) and 35 (17+18) in (2) and (3). The 2y ranges,
10.0–10.781 corresponding to the main line of the impurity
identified as Ni(OH)2 (PDF 14–117) in (1), 4.50–6.001 and
10.74–10.921 corresponding to an unidentified impurity in (3)
were excluded from the refinements. The observed, calculated and
difference patterns are compared in Fig. 2 and the fractional
atomic coordinates are reported in Supplementary material in
Tables SI–III for (1), (2) and (3). Interatomic distances were
calculated using the Fullprof program.
˚
(d0 ¼ 11.43, 13.93 and 13.86 A for (1), (2) and (3), respectively),
which is related to cell parameter a (see Table 1). The Ni2+ and
Co2+ cations are coordinated by six O atoms (four O atoms from
hydroxide and two
O atoms from the bidentate bridging
carboxylate) in an octahedral arrangement, in agreement with
the UV spectra.
Selected interatomic distances are reported in Table 2. Average
˚
M–O distances in the three compounds are M(1)–O ¼ 2.091 A in
˚
˚
˚
(1), 2.062 A in (2) and 2.135 A in (3); M(2)–O ¼ 2.062 A in (1),
3. Results and discussion
˚
˚
˚
˚
2.060 A in (2) and 2.145 A in (3) M(3)–O ¼ 2.071 A in (1), 2.034 A
˚
in (2) and 2.102 A in (3), in line with the sum of the ionic radii
given by the Shannon table (rNi2+ ¼ 0.69 A, rCo ¼ 0.745 A (HS),
2+
˚
˚
3.1. Spectroscopy results
rO2ꢀ ¼ 1.40 A). The C–C(E1.55 A), C–O(E1.25 A) distances and
C–C–C(E1101), C–O–C (E1221) angles are close to the expected
values.
˚
˚
˚
The FT-IR spectra shown in Fig. 3 for (1), (2) and (3) are very
similar. Two intense bands at 1564 and 1407 cmꢀ1 were assigned
to stretching uas(–COO–) and us(–COO–), respectively. The differ-
The layers are built from edge-sharing octahedra and exhibit
pores that can be seen as apertures of channels running along
[111]. Their average diameter estimated for the O–O distance is
ence between these two bands (
Dn
¼ 155 cmꢀ1) agrees with a
bridging character of the –COOꢀ groups, with two oxygen atoms
linked to NiII or CoII. Three bands around 2930 cmꢀ1 correspond to
different stretching modes of the H–C–H group. The bands
observed in the 3600–3500 cmꢀ1 range correspond to the
stretching of the OH groups.
˚
ꢂ4.5 A, which is probably too small to generate absorption
properties. The microporous aspect of the slabs is represented in
Fig. 6. The condensation of the ‘MO6’ octahedra observed in (1),
(2) or (3) is quite uncommon, to our knowledge. It is presented in
Fig. 7. The connection of five octahedra of the unit formula
‘(M2)2(M1)(M3)2’ forms a cross-like unit: the central M1 octahe-
dron sits on an inversion center while the horizontal and vertical
branches are formed by the M3 and M2 octahedra, respectively.
M3 and M2 are connected to M1 through their four equatorial
The UV spectra are reported in Fig. 4. For the nickel-based
compounds (1) and (2), the bands observed towards 700 and
390 nm correspond to electronic transitions from the fundamental
state A2g(F) towards the excited states 3T1g (u2) and 3T1g(P) (u3) of a
nickel atom in octahedral coordination. For (3), the band at about

ARTICLE IN PRESS
3232
A. Mesbah et al. / Journal of Solid State Chemistry 181 (2008) 3229–3235
Fig. 2. Observed, calculated and difference XRPD pattern for (1), (2) and (3).
Fig. 4. UV–visible spectra of (1), (2) and (3).
sharing octahedra. This contrasts with some open structures
observed for other layered hydroxy–dicarboxylate materials that
exhibit pores with a size up to tens of Angstro¨ems [20,25].
Fig. 3. FT-IR spectrum of (1), (2) and (3).
3.3. Magnetic properties
edges. The connection of two cross-like units by edges, which are
out of the equatorial plane of the central M1 octahedron, leaves a
free space or hole as can be seen in Fig. 7. It is different from the
connection of the four equatorial edges of octahedra leading to the
brucite layer. The brucite layers do not present any holes. Due to
this connection mode of the octahedra in (1), (2) and (3), holes are
formed, which are the smallest that can be generated by edge-
Compounds (1) and (2) have similar magnetic behaviors as can
be seen in Figs. 8 and 9, respectively. In both cases, the wT product
continuously decreases upon cooling (from 4.50 cm3 K/mol at 300 K
down to 1.97 cm3 K/mol at 27 K for (1) and from 3.29 cm3 K/mol at
300 K down to 0.62 cm3 K/mol at 19 K for (2)), which is a clear

ARTICLE IN PRESS
A. Mesbah et al. / Journal of Solid State Chemistry 181 (2008) 3229–3235
3233
Fig. 6. Representation of one layer showing the pores.
Fig. 5. General view of the layered structure of (2) and (3) in projection along the
c-axis. The structure of (1) is similar but with six carbon atoms for the aliphatic
chains.
Table 2
˚
Selected interatomic distances (A) and angles (1) in (1), (2) and (3)
Ni5(OH)6(C6H8O4)2 (1)
Ni5(OH)6(C8H12O4)2 (2)
Co5(OH)6(C8H12O4)2 (3)
Fig. 7. View showing the edge-sharing octahedra into and between two cross-like
units M5(OH)6(O2CCnH2nCO2)2. Carbon and H atoms are not represented.
˚
˚
˚
Atom–atom Distances (A) Atom–atom Distances (A) Atom–atom Distances (A)
Ni1–O12
Ni1–OH1
Ni1–OH2
Ni2–O11
Ni2–O62
Ni2–OH1
Ni2–OH1
Ni2–OH2
Ni2–OH3
Ni3–O11
Ni3–O61
Ni3–OH1
Ni3–OH2
Ni3–OH3
Ni3–O3
2 ꢃ 2.079(9) Ni1–O12
2 ꢃ 2.100(5) Ni1–OH1
2 ꢃ 2.065(9) Ni1–OH2
2 ꢃ 2.034(9) Co1–O12
2 ꢃ 2.067(6) Co1–OH2
2 ꢃ 2.045(8) Co1–OH1
2 ꢃ 2.107(8)
2 ꢃ 2.115(8)
2 ꢃ 2.071(4)
2.089(8)
2.119(8)
2.065(9)
2.039(9)
2.181(8)
2.063(8)
2.076(9)
2.133(9)
2.13(1)
Ni2–O11
Ni2–O82
Ni2–OH1
Ni2–OH1
Ni2–OH2
Ni2–OH3
Ni3–O11
Ni3–O81
Ni3–OH1
Ni3–OH2
Ni3–OH3
Ni3–OH3
2.049(9)
2.064(9)
2.182(9)
2.065(7)
2.034(8)
2.039(9)
2.11(1)
Co2–O11
Co2–O82
Co2–OH1
Co2–OH1
Co2–OH2
Co2–OH3
Co3–O11
Co3–O81
Co3–OH1
Co3–OH2
Co3–OH3
Co3–OH3
2.191(6)
2.081(6)
2.111(8)
2.128(8)
2.110(9)
2.08(1)
2.09(1)
2.161(9)
2.176(8)
2.066(7)
2.048(8)
2.080(8)
2.188(9)
2.052(6)
2.028(8)
1.993(9)
2.234(7)
2.069(6)
2.041(8)
2.120(7)
C1–C2
C2–C3
C3–C4
C4–C5
C5–C6
C1–O11
C1–O12
C6–O61
C6–O62
1.57(2)
1.55(2)
1.55(2)
1.55(2)
1.58(2)
1.25(2)
1.26(2)
1.27(2)
1.26(2)
C1–C2
C2–C3
C3–C4
C4–C5
C5–C6
C6–C7
C7–C8
C1–O11
C1–O12
C8–O81
C8–O82
1.59(1)
1.53(2)
1.52(1)
1.54(2)
1.55(1)
1.56(2)
1.58(1)
1.33(2)
1.28(2)
1.28(2)
1.30(2)
C1–C2
C2–C3
C3–C4
C4–C5
C5–C6
C6–C7
C7–C8
C1–O11
C1–O12
C8–O81
C8–O82
1.569(9)
1.55(2)
1.55(1)
1.55(2)
1.55(1)
1.56(2)
1.565(8)
1.30(2)
1.30(1)
1.28(2)
1.29(1)
Fig. 8.
wT product and wAC (lower inset) vs. temperature and magnetization vs.
field (upper inset) in (1).

ARTICLE IN PRESS
3234
A. Mesbah et al. / Journal of Solid State Chemistry 181 (2008) 3229–3235
Fig. 9.
wT product and wAC (lower inset) vs. temperature and magnetization vs.
Fig. 11. Field-dependent magnetization in (3) at 5 K.
field (upper inset) in (2).
spacing. The main difference between the magnetic properties of (1)
and (2) is found in the field dependence of their magnetization at
5K (inset of Figs. 8 and 9): that of (2) is a straight line, as expected
for a simple antiferromagnet, whereas the M(H) curve of (1) shows a
weak and almost anhysteretic field-induced transition at a thresh-
old field of 50 kOe.
The magnetic behavior of (3) is quite different. The wT product
first decreases upon cooling (Fig. 10) from 12.3 cm3 K/mol at room
temperature down to a minimum of 9.6 cm3 K/mol at 74 K and
then goes to a sharp maximum at 20.8 K before decreasing rapidly
at lower temperatures. The decrease of wT upon cooling between
300 and 74 K points to 2D antiferromagnetic interactions.
Spin–orbit effects might also partially contribute. The fit of the
reciprocal susceptibility above 90 K gives yp ¼ ꢀ15 K and an
effective Co(II) magnetic moment value of 4.42
m
B, close to that
expected for a high-spin Co2+ ion. The sharp increase in
wT at
lower temperatures is characteristic of a magnetized 3D magnetic
order, in agreement with the existence of an out-of-phase signal
in the temperature dependence of AC susceptibility (inset of
Fig. 10). Compound (3) is thus a ferrimagnet (or equivalently a
canted uncompensated antiferromagnet) below TC ¼ 16.3 K.
The field dependence of the magnetization M(H) recorded at 5 K
(Fig. 11) does not reveal any significant hysteresis. It shows that
above about 20 kOe the magnetization still increases significantly
linearly, likely due to the decrease of the canting angle between
the Co(II) moments, and is far from saturation even at 90 kOe,
where the magnetization M per formula unit (i.e. for five Co
Fig. 10.
wT product and wC (inset) vs. temperature in (3).
signature of 2D antiferromagnetic interactions. A fit of the reciprocal
susceptibility 1/ in the high-temperature range (ca. 120–300 K)
w
using the Curie–Weiss law yields the following paramagnetic Curie
temperatures and Ni(II) effective moments: yp ¼ ꢀ228 K and
meff. ¼ 3.34 mB for (1) and yp ¼ ꢀ249 K and meff. ¼ 3.12 mB for (2).
Such yp values are indicative of strong in-plane antiferromagnetic
interactions while the meff. values are close to theꢂ3.2 mB expected
for the Ni2+ ion. At TN ¼ 27 K for (1) and TN ¼ 19 K for (2), the rate of
atoms) reaches M ¼ 4.27
yB/f.u.
4. Conclusions
decrease of wT with temperature significantly increases, in a more
pronounced manner for (1), suggesting a 3D antiferromagnetic
ordering of the Ni moments. This is corroborated by the AC
susceptibility measurements (inset of Figs. 8 and 9) characterized
by a peak in w0 at these temperatures without the corresponding
peak in the out-of-phase signal. Note that the anomalies observed at
18 K in the thermal dependence of the AC susceptibility of (1) is
unambiguously due to a small amount of the Ni(OH)2 ferromagnetic
impurity detected in the powder X-ray diffraction patterns. The
peak in w0 of (1) observed at ꢂ60 K is an arte fact as it was not
reproduced on other magnetic curves (not shown here). The higher
ordering temperature of (1) likely results from its lower inter-layer
Three new hybrid compounds in the system M(II)–OH-linear
dicarboxylates synthesized by the hydrothermal route in basic
media base have been found (M ¼ Ni, Co). Their structures have
been solved ab initio from powder synchrotron data. All the
structures are layered, with bidendate carboxylate groups cova-
lently grafted to the M–OH layers and acting as a spacer. The
layers are built from edge-sharing M(II)O6 octahedra, with an
original connection scheme that leads to the formation of very
˚
small channels with about 4 A of aperture running along [111]. The
two Ni-based compounds ((1) and (2)) are antiferromagnets while
the Co-based one (3) is ferrimagnetic. Since the crystal structures

ARTICLE IN PRESS
A. Mesbah et al. / Journal of Solid State Chemistry 181 (2008) 3229–3235
3235
are very close, this difference is likely linked to the nature of the
[6] C.N.R. Rao, S. Natarajan, R. Vaidhyanathan, Angew. Chem. Int. Ed. 43 (2004)
1466.
[7] P. Rabu, M. Drillon, Adv. Eng. Mater. 5 (4) (2003) 189.
[8] N. Guillou, C. Livage, G. Fe´rey, Eur. J. Inorg. Chem. (2006) 4963.
[9] J.M. Rueff, J.F. Nierengarten, P. Gilliot, A. Demessence, O. Cregut, M. Drillon,
P. Rabu, Chem. Mater. 16 (2004) 2933.
magnetic orbitals (eg for Ni2+; eg and t2g for Co2+).
Supporting material
[10] N. Guillou, S. Pastre, C. Livage, G. Fe´rey, Chem. Commun. (2002) 2358.
[11] S. Konar, P.S. Mukherjee, E. Zangrando, F. Lloret, N.R. Chaudhuri, Angew. Chem.
Int. Ed. 41 (9) (2002) 1561.
[12] Y.-Q. Zheng, H.-Z. Xie, J. Solid State Chem. 177 (2004) 1352.
[13] N. Guillou, C. Livage, W. Van Beek, M. Nogue`s, G. Fe´rey, Angew. Chem. Int. Ed.
42 (6) (2003) 643.
Crystallographic data (excluding structure factors) for the
structure(s) reported in this paper have been deposited with the
Cambridge Crystallographic Data Centre as supplementary pub-
lication no. CCDC_652775–652777.
`
¨
´
[14] C. Serre, F. Millange, C. Thouvenot, M. Nogues, G. Marsolier, D. Louer, G. Ferey,
Copies of the data can be obtained free of charge on application
to CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax:
+441223 336 033; e-mail: deposit@ccdc.cam.ac.uk).
J. Am. Chem. Soc. 124 (2002) 13519.
[15] D.T. Tran, X. Fan, D.P. Brennan, P.Y. Zavalij, S.R.J. Olivier, Inorg. Chem. 44 (18)
(2005) 6192.
[16] Z.L. Huang, M. Drillon, M. Masciocchi, A. Sirani, J.T. Zhao, P. Rabu, P. Panissod,
Chem. Mater 12 (2000) 2805.
[17] M. Kurmoo, H. Kumagai, M.A. Greeen, B.W. Lovett, S.J. Blundell, A. Ardavan,
J. Singleton, J. Solid State Chem. 159 (2001) 343.
Acknowledgment
[18] R. Feyerherm, A. Loose, P. Rabu, M. Drillon, Solid State Sci. 5 (2003) 321.
[19] S. Abdelouhab, M. Franc-ois, E. Elkaim, P. Rabu, Solid State Sci. 7 (2005)
227–232.
[20] C. Livage, C. Egger, M. Nogues, G. Fe´rey, J. Mater. Chem. 8 (1998) 2743.
[21] Y.J. Kim, D.-Y. Jung, Bull. Korean Chem. Soc. 20 (7) (1999) 827.
[22] Y.J. Kim, D.-Y. Jung, K.-P. Hong, G. Demazeau, Solid State Sci. 3 (2001) 837.
[23] Y.-Q. Zheng, J. Sun, J. Solid State Chem. 172 (2003) 288.
[24] M. Kurmoo, J. Mater. Chem. 9 (1999) 2595.
The authors would like to thank Irena Margiolaki (European
Synchrotron Radiation Facilities (ESRF)’ in Grenoble, France) for
her help during the experiments.
Appendix A. Supplementary materials
[25] C. Livage, C. Egger, G. Fe´rey, Chem. Mater. 11 (1999) 1546.
[26] P.M. Forster, A.K. Cheetham, Angew. Chem. Int. Ed. 41 (3) (2002) 457.
[27] M. Drillon, P. Panissod, J. Magn. Magn. Mater. 188 (1998) 93.
[28] M. Kurmoo, H. Kumagai, S.M. Hughes, C.J. Kepert, Inorg. Chem. 42 (2003)
6709.
Supplementary data associated with this article can be found
in the online version at doi:10.1016/j.jssc.2008.06.059.
[29] /http://www.qdusa.comS.
[30] A.N. Fitch, J. Res. Natl. Inst. Stand. Technol. 109 (2004) 33–142.
[31] /http://www.accelrys.com/products/mstudio/modeling/crystallization/reflex.
htmlS.
[32] A.M.J. Neumann, Appl. Crystallogr. 36 (2003) 356.
[33] V. Favre-Nicolin, R. Cerny, Z. Kristallogr. 219 (2004) 847.
[34] J. Rodriguez-Carvajal, M.T. Fernandez-Diaz, J.L. Martinez, J. Phys.: Condens.
Matter 3 (1991) 3215.
References
[1] D. Maspoch, D. Ruiz-Molina, J. Veciana, Chem. Soc. Rev. 36 (2007) 770.
[2] S. Kitagawa, R. Kitaura, S.-I. Noro, Angew. Chem. Int. Ed. 43 (2004) 2334.
[3] C. Sanchez, G.J. de, A.A. Soler-Illia, F. Ribot, T. Lalot, C.R. Mayer, V. Cabuil,
Chem. Mater. 13 (2001) 3061.
´
[4] G. Ferey, Chem. Mater. 13 (10) (2001) 3084.
[5] J.L.C. Rowsell, O.M. Yaghi, Angew. Chem. Int. Ed. 44 (2005) 4670.