Synthetic control to achieve lanthanide(III)/pyrimidine-4,6-dicarboxylate compounds by preventing oxalate formation: Structural, magnetic, and luminescent properties

Article abstract of DOI:10.1021/ic3009392

Figure Persented: Control over the synthetic conditions in many metal/diazinedicarboxylato systems is crucial to prevent oxalate formation, since dicarboxylato ligands easily undergo degradation in the presence of metal salts. We report here an efficient route to obtain oxalato-free compounds for the lanthanide/pyrimidine-4,6-dicarboxylato (pmdc) system on the basis of the reaction temperature and nonacidic pH or oxygen free atmosphere. Two different crystal architectures have been obtained: {[Ln(μ-pmdc)_1.5(H ₂O)₃]·xH₂O}_n (1-Ln) and {[Ln₂(μ₄-pmdc)₂(μ-pmdc)(H ₂O)₂]·H₂O}_n (2-Ln) with Ln(III) = La-Yb, except Pm. Both crystal structures are built from distorted two-dimensional honeycomb networks based on the recurrent double chelating mode established by the pmdc. In compounds 1-Ln, the tricapped trigonal prismatic coordination environment of the lanthanides is completed by three water molecules, precluding a further increase in the dimensionality. Crystallization water molecules are arranged in the interlamellar space, giving rise to highly flexible supramolecular clusters that are responsible for the modulation found in compound 1-Gd. Two of the coordinated water molecules are replaced by nonchelating carboxylate oxygen atoms of pmdc ligands in compounds 2-Ln, joining the metal-organic layers together and thus providing a compact three-dimensional network. The crystal structure of the compounds is governed by the competition between two opposing factors: the ionic size and the reaction temperature. The lanthanide contraction rejects the sterically hindered coordination geometries whereas high-temperature entropy driven desolvation pathway favors the release of solvent molecules leading to more compact frameworks. The characteristic luminescence of the Nd, Eu, and Tb centers is improved when moving from 1-Ln to 2-Ln compounds as a consequence of the decrease of the O-H oscillators. The magnetic properties of the compounds are dominated by the spin-orbit coupling and the ligand field perturbation, the exchange coupling being almost negligible.

Full text of DOI:10.1021/ic3009392

Article
pubs.acs.org/IC
Synthetic Control to Achieve Lanthanide(III)/Pyrimidine-4,6-
dicarboxylate Compounds by Preventing Oxalate Formation:
Structural, Magnetic, and Luminescent Properties
Javier Cepeda,*,† Rolindes Balda,‡ Garikoitz Beobide,† Oscar Castillo,*,† Joaquín Fernandez,‡
́
Antonio Luque,† Sonia Perez-Yanez,† and Pascual Roman†
†Departamento de Química Inorgan
E-48080 Bilbao, Spain
́
́
́
̃
́
ica, Facultad de Ciencia y Tecnología, Universidad del País Vasco, UPV/EHU, Apartado 644,
‡Departamento de Física Aplicada I, Escuela Superior de Ingenieros, Universidad del País Vasco, UPV/EHU, E-48013 Bilbao, Spain
S
* Supporting Information
ABSTRACT: Control over the synthetic conditions in many metal/diazinedicarboxylato systems is crucial to prevent oxalate
formation, since dicarboxylato ligands easily undergo degradation in the presence of metal salts. We report here an efficient route
to obtain oxalato-free compounds for the lanthanide/pyrimidine-4,6-dicarboxylato (pmdc) system on the basis of the reaction
temperature and nonacidic pH or oxygen free atmosphere. Two different crystal architectures have been obtained: {[Ln(μ-
pmdc)1.5(H2O)3]·xH2O}n (1-Ln) and {[Ln2(μ4-pmdc)2(μ-pmdc)(H2O)2]·H2O}n (2-Ln) with Ln(III) = La−Yb, except Pm.
Both crystal structures are built from distorted two-dimensional honeycomb networks based on the recurrent double chelating
mode established by the pmdc. In compounds 1-Ln, the tricapped trigonal prismatic coordination environment of the
lanthanides is completed by three water molecules, precluding a further increase in the dimensionality. Crystallization water
molecules are arranged in the interlamellar space, giving rise to highly flexible supramolecular clusters that are responsible for the
modulation found in compound 1-Gd. Two of the coordinated water molecules are replaced by nonchelating carboxylate oxygen
atoms of pmdc ligands in compounds 2-Ln, joining the metal−organic layers together and thus providing a compact three-
dimensional network. The crystal structure of the compounds is governed by the competition between two opposing factors: the
ionic size and the reaction temperature. The lanthanide contraction rejects the sterically hindered coordination geometries
whereas high-temperature entropy driven desolvation pathway favors the release of solvent molecules leading to more compact
frameworks. The characteristic luminescence of the Nd, Eu, and Tb centers is improved when moving from 1-Ln to 2-Ln
compounds as a consequence of the decrease of the O−H oscillators. The magnetic properties of the compounds are dominated
by the spin−orbit coupling and the ligand field perturbation, the exchange coupling being almost negligible.
the light emission properties with microporosity, magnetism,
sensing, and activity as multimodal imaging contrast agents.3
Compared to first-row transition metal ions, lanthanide ions
possess more flexible and larger coordination geometries, which
makes it difficult to predict the resulting crystal structure.4 In
principle, these characteristics result in more facile routes to
densely packed solids given the trend of the lanthanide atoms
INTRODUCTION
■
The rapidly expanding field of crystal engineering of two- and
three-dimensional coordinated polymers is of great current
interest for both the structural and topological novelty as well
as for their potential application as functional materials.1
Among the diverse metal−organic frameworks, lanthanide
based crystal structures are attractive, owing to their unique
optical and magnetic properties arising from 4f electrons and
characteristic coordination preferences.2 In particular, the
luminescent Ln-MOFs may undertake a dual role by combining
Received: May 8, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic3009392 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
pump beam with an optical fiber by use of two lenses. The
fluorescence was analyzed by using a spectrometer (CVI Spectral
Products SM-240) in the VIS and a Hamamatsu C9913GC in the
near-infrared. Long-pass filters (Semrock LP355 and LP830) were
used to remove light at the pump frequency. Lifetime measurements
were performed by exciting the samples with the frequency triplet
output (355 nm) of a 10 Hz, Q-switched Nd:YAG laser and detecting
the emission with a fast photodiode connected to a digital oscilloscope.
Synthesis of {[Ln(μ-pmdc)1.5(H2O)3]·xH2O}n (x = 5.5, except
6.5 for Gd) [1-Ln; Ln(III) = La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho,
Er, Tm, Yb]. 0.150 mmol of the corresponding metal salt dissolved in
5 mL of water was added dropwise over 25 mL of a water−methanol
(1:2) solution containing H2pmdc (0.0459 g, 0.225 mmol). The
resulting solution, with a pH = 2.0−2.5, was kept under vigorous
stirring for five hours at room temperature. The resulting precipitates,
corresponding to compounds 1-Ln, were collected by filtration,
washed with water and ethanol, and air-dried. Yield: 80−85%. Despite
the low crystallinity shown by compounds 1-La, 1-Ce, and 1-Pr, their
IR spectra contain all the characteristic bands of the isostructural series
(Supporting Information). Colorless square prismatic single crystals of
1-Sm, 1-Eu, 1-Gd, 1-Tb, 1-Dy, and 1-Yb appeared in mother liquid
after one week at room temperature.
Synthesis of {[Ln2(μ4-pmdc)2(μ-pmdc)(H2O)2]·H2O}n [2-Ln;
Ln(III) = La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb].
The compounds were obtained by preparing a water solution (25 mL)
containing 0.1 mmol of the metal salt and 0.2 mmol of H2pmdc
(0.0408 g), setting the pH = 5−6 by the addition of NaOH (1 M),
placing it onto a 45 mL Teflon-lined stainless steel autoclave under
autogenous pressure at 180−200 °C for 3 days (180 °C for La−Nd
and 200 °C for Sm−Yb), and then, slowly cooling to room
temperature (2 °C/h). Yield: 85−90%. Compounds 2-Ln can also
be synthesized performing the synthesis under inert conditions
without prior basification with NaOH, as will be later discussed.
The synthesis at lower temperatures under the latter conditions
allows obtaining the new compounds: [Ln(pmdc)1.5(H2O)]·xH2O (A,
L a ) , [ L n ( p m d c ) 1 . 5 ( H 2 O ) 2 ] · x H 2 O ( B , S m ) , [ L n -
(pmdc)1. 5(H2O)2]·xH2O (C, Sm, Dy, Yb), and [Ln-
(pmdc)1.5(H2O)2.5]·xH2O (D, Dy, Yb); which have been characterized
by means of elemental analyses, IR spectra, thermogravimetric (TG)
measurements, and X-ray powder diffraction (XRPD) measurements
(see Supporting Information). Their amount of coordinated water
molecules has been estimated as the molecules released at temper-
atures above 90−100 °C in the TG curves. The crystallization
molecules cannot be reliably estimated, since they are lost from room
temperature.
to coordinate to several ligands, although an exhaustive control
of the synthetic conditions may direct the assembling toward
open architectures due to the important role played by water
molecules as terminal ligands.5 In addition, according to the
hard−soft acid−base principle,6 the lanthanide atoms have high
affinity for oxygen donor atoms, which explains the wide variety
of reported compounds with a multycarboxylate ligand.7
Nevertheless, mixed O- and N-donor molecules give the
opportunity to discover new polymeric structures with higher
complexity due to the bonding through both nitrogen or
carboxyl oxygen species.8
The carboxylate groups in the 4- and 6-positions of the
pyrimidinic ring avoids the steric hindrance between them and
confers a planar topology to the pyrimidine-4,6-dicarboxylate
dianion (pmdc). This structural feature favors its bis-chelating
mode, but even higher connectivity is also available through its
uncoordinated carboxylate oxygen atoms; hence, it has a strong
potential to build up highly ordered extended structures.9 On
the other hand, our previous experimental work with this ligand
under hydrothermal conditions has revealed that its in situ
partial decomposition generates oxalate anions in the reaction
media.10 The present work is a new contribution to our
research on trivalent metal centers and diazinedicarboxylato
ligands based systems.8c,10 It reports an exhaustive study of the
synthesis conditions that has allowed us to preclude the
decomposition of pmdc, leading to two families of oxalate-free
extended Ln/pmdc networks ranging from two-dimensional
(2D) sheets (1-Ln) to compact three-dimensional (3D)
networks (2-Ln). All compounds have been structurally and
magnetically characterized, while the luminescence properties
of some of these compounds have been also accomplished.
Furthermore, the use of Fourier-transform infrared (FTIR)
spectroscopy has enabled us to get deeper insights into the
coordination mode of the pmdc ligand, and it even has allowed
distinguishing between the two families of Ln/pmdc com-
pounds. The structural diversity has been rationalized according
to the close relationship encountered between the entropy
driven hydrothermal desolvation pathway and the combination
of ligand steric hindrance and Ln(III) ion size effect.
EXPERIMENTAL PROCEDURES
X-ray Diffraction Data Collection and Structure Determi-
nation. Single crystal diffraction data were collected at 100(2) K on
Oxford Diffraction Xcalibur (λMoKα = 0.71073 Å for 1-Gd and 1-Sm)
and Agilent Technologies Supernova diffractometers (λCuKα = 1.5418
Å for 1-Tb and 2-Ce; λMoKα = 0.71073 Å for 1-Eu, 1-Gd, 1-Dy, 1-Yb,
and 2-La). The data reduction was done with the CrysAlisPro
program.13 Crystal structures of compounds 1-Ln and 2-Ln (except
for that of 1-Gd) were solved by direct methods using the SIR92
program14 and refined by full-matrix least-squares on F2 including all
reflections (WINGX and Jana2006 for 1-Ln and 2-Ln, respec-
tively).15−17 During the data reduction process it became clear that all
the crystal specimens of compounds 1-Ln were pseudomerohedric
■
Chemicals. All the chemicals were of reagent grade and were used
as commercially obtained. The starting material pyrimidine-4,6-
dicarboxylic acid (H2pmdc) was prepared following the previously
reported procedure.11
Physical Measurements. Elemental analyses (C, H, N) were
performed on an Euro EA elemental analyzer, whereas the metal
content, determined by inductively coupled plasma (ICP-AES) was
performed on a Horiba Yobin Yvon Activa spectrometer. The IR
spectra (KBr pellets) were recorded on a FTIR 8400S Shimadzu
spectrometer in the 4000−400 cm−1 spectral region. Magnetic
measurements were performed on polycrystalline samples of the
complexes with a Quantum Design SQUID susceptometer covering
the temperature range 5.0−300 K at a magnetic field of 1000 G. The
susceptibility data were corrected for the diamagnetism estimated from
Pascal’s Tables,12 the temperature-independent paramagnetism, and
the magnetization of the sample holder. Thermal analyses (TG/DTA)
were performed on a TA Instruments SDT 2960 thermal analyzer in a
synthetic air atmosphere (79% N2/21% O2) with a heating rate of 5
°C·min−1. The emission measurements were performed at room
temperature in a backscattering arrangement by using the frequency
triplet output (355 nm) of a 10 Hz, Q-switched Nd:YAG laser (pulse
duration 40 ps) and a Ti-sapphire ring laser in the 770−920 nm
spectral range as exciting sources. The emission from the free sample
surface was collected along the backward direction of the incident
twins with a twin law: (100/010/001). The final results showed a
̅
̅
percentage for the minor domain of 49.1% (1-Sm), 42.7% (1-Eu),
42.3% (1-Tb), 49.6% (1-Dy), and 48.0% (1-Yb). These compounds
present a strong pseudosymmetry, precluding an anisotropic refine-
ment of the nonmetallic atoms of 1-Dy and 1-Yb. In compound 1-Gd,
the CCD images obtained at 100 K showed that, in addition to the
main reflections, the diffraction pattern also contained weak satellite
reflections that could be indexed with four integers as H = ha* + kb*
1
+ lc* + mq with q = (0, /3, 0) (see the Supporting Information).
Similar, but not well-resolved, diffuse scattering was also found for the
remaining compounds of the 1-Ln family. Therefore, the structures of
these compounds were resolved ignoring this difuse scattering, except
for compound 1-Gd, which has been considered to be commensur-
B
dx.doi.org/10.1021/ic3009392 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. Representation of the atomic modulation function at sections through the electron density at the position of: (a) Gd1 and (b) Gd2 atoms
in compound 1-Gd.
ately modulated with an a X 3b X c supercell and solved in the direct
space using the basic cell and the initial structure model provided by
the isostructural compound 1-Sm. Then, it was refined by full-matrix
least-squares on F by using the JANA2006 program, which smoothly
converged it to a stable solution. The structure model was further
improved by adding harmonic positional modulation waves for all the
atoms and a modulation wave for the anisotropic displacement
parameters (ADP) of the metal atoms. Finally, the twin law was added
leading to a percentage of twinned component of 9.30%. After
averaging the electron density according to the superspace symmetry,
we obtained a good estimate of both the basic positions of the atoms
and their modulation functions (Figure 1). During the final steps of
refinement high residual electron density was found near the atom
O8w, so it was considered to be disordered (O8w′) into two different
positions. The disorder was treated introducing a similarity restriction
that forces both atoms to possess the same ADP parameters and their
occupancies to be complementary modulated.
During the refinement of crystal structures of compounds 2-Ln, the
ADP parameters of several atoms became not positive definite due to
the strong pseudosymmetry present. To overcome this problem, a
rigid model was applied to the pmdc molecule in such a way that a
unique molecule describes the three independent ligands, although
independent ADP parameters are employed to refine the thermal
movement of each atom. Details of the structure determination and
refinement of all compounds are summarized in Tables 1−2.
means of the FULLPROF program (pattern-matching analysis)18 on
the basis of the space group and the cell parameters found for
isostructural compounds by single crystal X-ray diffraction. The unit
cell parameters obtained in the final refinement are listed in Tables
1−2. Variable-temperature X-ray powder diffraction measurements of
compounds 1-Eu and 2-Pr were run under ambient atmosphere with
heating rates of 5 °C·min−1 and measuring a complete diffractogram
every 20 °C.
RESULTS AND DISCUSSION
■
Control Over the pmdc Decomposition to Oxalate.
The feasibility to crystallize a wide diversity of compounds
upon reaction between the lanthanides and the pmdc ligand, as
arises from our previous work on Ln/pmdc/ox10 and from the
herein reported Ln/pmdc systems, highlights the importance of
understanding the synthetic conditions that lead to the
formation of each of them. The pmdc ligand tends to undergo
degradation in the presence of lanthanide salts under
hydrothermal conditions rendering oxalate as final product,
which easily incorporates to the solid coordination polymer.
The present work explores the effect of the reaction conditions,
such as the temperature, pretreatment, pH, and atmosphere to
prevent the pmdc decomposition in order to crystallize oxalate-
free Ln/pmdc compounds (Scheme 1).
The XRPD patterns were collected on a Phillips X’PERT powder
diffractometer with Cu Kα radiation (λ = 1.5418 Å) over the range 5 <
2θ < 50° with a step size of 0.02° and an acquisition time of 2.5 s per
step at 25 °C. Indexation of the diffraction profiles were made by
The in situ generation of oxalate anion under hydrothermal
conditions is a commonly reported reaction19 for systems
C
dx.doi.org/10.1021/ic3009392 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 1. Crystallographic Data and Structure Refinement Details of Compounds 1-Ln
1-Nd
1-Sm
1-Eu
1-Gd
1-Tb
empirical formula
formula weight
crystal system
space group
a (Å)
C9H26N3NdO14.5
552.55
C9H26N3O14.5Sm
558.67
C9H26EuN3O14.5
560.28
C9H28GdN3O15.5
583.58
C9H26N3O14.5Tb
567.24
monoclinic
P21/c
monoclinic
P21/c
monoclinic
P21/c
monoclinic
P21/c(0\β0)s0
22.577(6)
11.905(2)
14.423(6)
91.35(3)
monoclinic
P21/c
22.753(3)
11.965(2)
14.465(3)
90.67(1)
3937.4(8)
−
22.641(1)
11.933(1)
14.522(1)
91.371(3)
3922.4(5)
1.17
22.601(2)
11.881(1)
14.441(1)
91.293(6)
3876.7(3)
1.12
22.604(1)
11.847(1)
14.375(1)
90.920(4)
3849.0(5)
1.08
b (Å)
c (Å)
β (deg)
V (Å3)
3875.4(2)
a
e
GOF
2.13
Rint
χ2
−
1.02
0.0530
0.0440
0.1101
0.0970
−
−
−
−
final R indices
b
c
d
[I > 2σ(I)] R1 /wR2
0.0543/0.1521
0.0591/0.1491
1-Ho
0.0546/0.1403
0.0650/0.1462
1-Er
0.0871/0.1091
0.1504/0.1206
1-Tm
0.0825/0.2570
0.0978/0.2711
1-Yb
c
d
all data R1 /wR2
1-Dy
empirical formula
formula weight
crystal system
space group
a (Å)
C9H26DyN3O14.5
553.68
C9H26HoN3O14.5
573.25
C9H26ErN3O14.5
575.57
C9H26N3O14.5Tm
577.25
C9H26N3O14.5Yb
581.35
monoclinic
P21/c
monoclinic
P21/c
monoclinic
P21/c
monoclinic
P21/c
monoclinic
P21/c
22.595(2)
11.832(1)
14.343(2)
91.026(1)
3833.9(7)
1.01
22.806(3)
11.975(1)
14.475(2)
90.72(1)
3953.1(8)
−
23.058(3)
12.293(2)
14.535(2)
90.71(2)
4119.7(6)
−
22.913(3)
11.988(1)
14.413(1)
90.74(1)
3958.8(6)
−
22.568(1)
11.817(1)
14.220(1)
90.446(7)
3792.1(4)
1.08
b (Å)
c (Å)
β (deg)
V (Å3)
a
GOF
Rint
χ2
0.0730
−
1.13
−
1.34
−
1.44
0.0790
−
−
final R indices
c
d
[I > 2σ(I)] R1 /wR2
0.0714/0.1855
0.0652/0.1786
c
d
all data R1 /wR2
0.1154/0.2043
0.1024/0.1924
a
b
c
d
2
2
S = [∑w(F02 − Fc2)2/(Nobs − Nparam)]1/2. I > 3σ(I) (1-Gd). R1 = ∑||F0| − |Fc||/∑|F0|. wR2 = [∑w(F02 − Fc2)2/∑wF0 ]1/2; w = 1/[σ2(F0 ) +
(aP)2 + b] where P = (max(F0 ,0) + 2Fc2)/3 with a = 0.0628 (1-Sm), 0.0684 (1-Eu), 0.1686 (1-Tb), 0.1185 (1-Dy), and 0.1025 (1-Yb); b = 56.8404
2
e
(1-Sm), 44.6121 (1-Eu), and 42.5345 (1-Tb). w = 1/[σ2(|F0|) + (0.0001F0)2] and P = ∑w(|Fc| − |F0|)2 (1-Gd).
involving a wide variety of metal cations, anions, organic
ligands, and solvents.20 To date, there have been four
mechanisms proposed to explain the in situ formation of
oxalate: (1) the reductive coupling of dissolved carbon
dioxide,21 (2) the oxidation of ethanol or oxidative coupling
of methanol using nitrate as oxidant,22 (3) the decarboxylation
of organic ligands, followed by reductive coupling of the
resulting carbon dioxide,23 and (4) the oxidation−hydrolysis in
the presence of a metal catalyst implying the general
decomposition of the organic ligand.24 The carboxyl transfer
observed for the pyrazine-2,5-dicarboxylate anion to render the
isomer pyrazine-2,6-dicarboxylate during hydrothermal syn-
thesis of [Ln(2,5-pzdc)0.5(2,6-pzdc)(H2O)3] compounds,8b
following the Raecke process,25 suggests that diazinedicarbox-
ylato ligands can easily undergo a decarboxylative reaction upon
heating. Cahill et al.26 have elucidated that during the
hydrothermal synthesis of the 2D {[Nd(2,3-pzdc)(ox)-
(H2O)2]}n (2,3-pzdc = pyrazine-2,3-dicarboxylato) the oxalato
ligand results from the oxidation of the H2pzdc through several
intermediates by means of the ring-opening that occurs via
cleavage of the C−N bond of the pyrazine ring followed by
hydrolysis−oxidation of the resulting species (mechanism 4).
The proposed mechanism also clarifies that prior to
precipitation, the presence of metal cations, nitrogen-lone
pairs, and carboxylate chelation likely accelerate the in situ
ligand formation and control end products; suggesting that the
CO2 liberated by decarboxylation does not participate in
oxalate formation in contrast to what was previously reported.
The main conclusion of our study, in agreement with
Cahill,26 is that the pmdc decomposition reaction only takes
place if the following three factors are met: relatively high
temperature, oxygen containing atmosphere, and acidic pH.
Performing the synthesis at room temperature, we obtain a
novel family of oxalate-free 1-Ln compounds. The crystallinity
of this phase is highly depending on the pH value, obtaining
crystalline products when the initial pH value of the reagent
mixture is not modified (pH = 2.0−2.5). Additionally, the
crystallinity is also improved when using a water/methanol
mixture as solvent. At higher pH values, the great insolubility of
1-Ln compounds leads to amorphous products. On the other
hand, when the reaction takes place at higher temperatures
(60−160 °C) and pH = 2.0−2.5 Ln/pmdc/ox compounds are
obtained. A significant improvement of the crystallinity is
achieved by pretreatment at 90 °C. This result suggests that the
oxidation of pmdc takes place during the open-atmosphere
pretreatment procedure, probably because of the dissolved
molecular oxygen, and it requires a relatively high temperature
in order to surpass the activation energy of the reaction. When
the reaction temperature is set at 180−200 °C the oxalate-free
2-Ln are obtained as low-crystalline products mixed with
D
dx.doi.org/10.1021/ic3009392 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 2. Crystallographic Data and Structure Refinement Details of Compounds 2-Ln
2-La
2-Ce
2-Pr
2-Nd
2-Sm
2-Eu
empirical formula
formula weight
crystal system
superspace group
a (Å)
C9H6LaN3O7.5
415.07
C9H6CeN3O7.5
416.28
C9H6N3O7.5Pr
417.07
C9H6N3NdO7.5
420.40
C9H6N3O7.5Sm
426.52
C9H6EuN3O7.5
428.12
monoclinic
Pc
monoclinic
Pc
monoclinic
Pc
monoclinic
Pc
monoclinic
Pc
monoclinic
Pc
9.3759(2)
10.4558(2)
12.5447(2)
106.632(2)
1178.34(4)
0.97
9.3148(2)
10.3964(2)
12.4509(3)
106.685(3)
1154.99(5)
1.22
9.316(5)
10.401(6)
12.431(4)
106.96(3)
1152.1(7)
−
9.445(3)
10.424(4)
12.504(2)
107.11(6)
1176.6(3)
−
9.497(4)
10.386(1)
12.521(2)
107.05(3)
1180.7(5)
−
9.543(3)
10.348(4)
12.551(3)
107.16(3)
1184.2(7)
−
b (Å)
c (Å)
β (deg)
V (Å3)
a
GOF
Rint
χ2
0.0264
0.0205
−
1.38
−
2.15
−
1.34
−
1.73
−
−
final R indices
b
c
[I > 2σ(I)] R1 /wR2
0.0268/0.0534
0.0353/0.0589
0.0220/0.0503
0.0232/0.0512
b
c
all data R1 /wR2
2-Gd
2-Tb
2-Dy
2-Ho
2-Er
2-Tm
2-Yb
empirical formula
formula weight
crystal system
superspace group
a (Å)
C9H6GdN3O7.5
433.41
C9H6N3O7.5Tb
435.09
C9H6DyN3O7.5
438.66
C9H6HoN3O7.5
441.09
C9H6ErN3O7.5
443.42
C9H6N3O7.5Tm
445.09
C9H6N3O7.5Yb
449.20
monoclinic
Pc
monoclinic
Pc
monoclinic
Pc
monoclinic
Pc
monoclinic
Pc
monoclinic
Pc
monoclinic
Pc
9.576(2)
10.321(2)
12.594(3)
107.25(3)
1188.7(4)
3.38
9.553(5)
10.287(5)
12.636(3)
107.30(1)
1185.6(9)
9.509(2)
10.356(4)
12.682(4)
107.38(3)
1191.8(5)
9.448(5)
10.398(3)
12.624(4)
107.40(4)
1183.4(8)
1.22
9.487(3)
10.432(3)
12.515(7)
107.25(2)
1182.9(4)
9.389(8)
10.463(6)
12.427(3)
107.19(5)
1166.3(12)
1.37
9.399(3)
10.455(5)
12.392(5)
107.05(2)
1164.1(1)
1.03
b (Å)
c (Å)
β (deg)
V (Å3)
χ2
1.38
2.15
1.88
a
b
c
2
S = [∑w(F02 − Fc2)2/(Nobs − Nparam)]1/2. R1 = ∑||F0| − |Fc||/∑|F0|. wR2 = [∑w(F02 − Fc2)2/∑wF0 ]1/2; w = 1/[σ2(|F0|) + (0.0004F0)2] and P =
∑w(|Fc| − |F0|)2.
previously mentioned Ln/pmdc/ox compounds. This behavior
is attributed to an entropy driven effect that will be discussed in
the following section. Shifting the pH to 5−6 or using an inert
atmosphere slows down the pmdc ring-opening stage, bypasses
the formation of Ln/pmdc/ox systems, and leads to crystalline
2-Ln compounds. Thus, it can be concluded that the absence of
oxygen and the nonacidic pH are crucial to obtain oxalate-free
compounds.
Regarding the pmdc decomposition, it deserves to be noted
that the ammonia molecules released after the ring-opening
process26 raise the pH value and inhibit a further decom-
position, explaining why, when the initial pH is not modified,
the Ln/pmdc/ox compounds are obtained instead of simple
lanthanide oxalates.
Synthetic Conditions: Dimensionality Control. The
entropy driven desolvation pathway27 that takes place at higher
temperatures is known to favor the release of solvent molecules.
This effect is clearly observed in the Ln/pmdc system, since the
pmdc ligand raises its denticity to occupy the empty
coordination positions, increasing the connectivity among the
metal nodes. In any case, the temperature required for the
formation of 2-Ln increases on going forward in the lanthanide
series, as it depends on the energetic penalty that must be paid
because of the combined effect of the ligands steric hindrance
and the ionic radius of the lanthanide.28 In this sense, a study by
means of the Solid-G program29 revealed that the greater
hindrance per occupied coordination position is due to the
pmdc ligand coordinated by its nonchelating oxygen atom
(Scheme 2), a coordination mode that is only present in
compounds 2-Ln. Therefore, the smaller the lanthanide ion the
greater the steric hindrance and the higher the required
temperature to obtain 2-Ln. This fact is reinforced by the
increase of the hydration energy on going forward in the
lanthanide series that hinders the release of the coordinated
water molecules. On the other hand, the formation of 2-Ln
compounds in synthetic conditions in which oxalate anions are
present is explained taking into account the lower H2O/Ln
ratio present in 2-Ln in comparison with those of the Ln/
pmdc/ox phases. Therefore, the entropy driven desolvation
reaction favors the formation of 2-Ln at higher temperatures (T
> 180 °C).
To get further detail on the structural trends of the Ln/pmdc
system, the effect of the synthesis temperature was assessed for
−
selected lanthanides where the LnX3/pmdc/H2O (X = NO3
or Cl−) molar ratio was fixed at 1:2:18518 (Scheme 3). The
solutions were prepared by bubbling N2 gas, and a pH value of
5.5 was adjusted. In general, the formation of compounds 1-Ln
occurs at room temperature while compounds 2-Ln are
obtained above 180 °C. However, when applying intermediate
temperatures, four new oxalate-free Ln/pmdc compounds
(named A, B, C, and D) were obtained (Supporting
Information).
In view of these results, it becomes evident again that the
entropic effect with increasing temperature stabilizes more
compact frameworks in which the H2O/Ln ratio experiences a
progressive decrease as the coordinated water molecules are
−
replaced by nonchelating OCOO atoms, which implies an
increase of the connectivity. Compounds A−D present a
general formula of [Ln(pmdc)1.5(H2O)y]·xH2O (y = 1−2.5),
with a coordination water content that is between that of
compounds 1-Ln and 2-Ln. This feature suggests that the
crystal structures of these new compounds are probably
E
dx.doi.org/10.1021/ic3009392 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 1. Synthetic Conditions on the Ln/pmdc System
Scheme 2. Solid Angle (Ω) Occupied by Each Ligand around
the Metal Center, Together with Some Relevant Parameters
of the Compounds
Crystal Structure of 1-Ln. The crystal structure of these
compounds consists of the stacking of neutral [Ln(μ-
pmdc)1.5(H2O)3] layers that accommodates crystallization
water molecules in between. In the layers, the crystallo-
graphically independent metal centers, Ln1 and Ln2, are
surrounded by a N3O3Ow3 donor set (Figure 2), with a
geometry close to a tricapped trigonal prism as revealed the
continuous shape measurements (CShM).30 The S(ttp) results
[0.88 (Yb) and 1.23 (Sm)] show that the less the ionic radius
of the lanthanide(III) atom, the less distortion with regard to
the ideal shape. The coordination Ln−N and Ln−O bond
lengths and angles are within the range observed for analogous
carboxylic azine ligands.31
Each metal center coordinates to three almost planar bis-
bidentate pmdc ligands by means of five member chelate rings,
while three coordinated water molecules complete the
coordination sphere. These Ln(μ-pmdc)3(H2O)3 units are
joined to one another (Ln···Ln distances of ca. 7.2 Å),
generating layers of hexagonal six-membered rings
(Ln···Ln···Ln, ranging from 112.4 to 123.9°) that comprise a
honeycomb pattern. This fact forces the pmdc ligands to
arrange outward the mean plane of the layer established by the
metal centers (47.8° for pmdc1, 15.6° for pmdc2, and 52.28°
for pmdc3), leaving the carboxylate oxygen atoms exposed to
the interlayer space ready to interact with the solvent
molecules. The topological analysis performed by the TOPOS
program package,32 as can be seen in Figure 3, indicates an
midway between the 2D (1-Ln) and the compact 3D (2-Ln)
structures. All these results confirm not only the coordinative
capability and versatility of the pmdc ligand in constructing a
wide variety of crystal structures but also the specific synthetic
conditions that prevent oxalate formation.
F
dx.doi.org/10.1021/ic3009392 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 3. Plot of Temperature vs. Ionic Radius on the
a
Ln(III)/pmdc System
Figure 3. 2D topological network showing the six-membered rings.
a
The room temperature (*) tests have been performed with
continuous stirring and without hydrothermal conditions.
the unit cell volume (Figure 4).33 The hydrogen bonding
interactions established among them generate water clusters in
the form of isolated R7 rings34 anchored to the walls of the Ln-
pmdc framework by means of an extensive network of
hydrogen bonds.
uninodal two-dimensional hcb topological network (3-c net),
the point symbol being (63).
The piling up of the layers takes place along the c axis in an
ABA fashion creating intricate channels that are filled by the
crystallization water molecules, which account for the 34% of
It is worth noting that compound 1-Gd undergoes a phase
transition toward a modulated commensurate structure with a
Figure 2. Coordination environments of the Sm1 and Sm2 atoms in compound 1-Sm.
G
dx.doi.org/10.1021/ic3009392 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 4. Packing of compound 1-Sm showing the chains of solvent
molecules.
1
commensurate q vector (q = 0 /3 0) in the superspace group
P21/c(0\β0)s0 upon cooling down to 100 K. The super-
structure consists of all the atoms of the asymmetric unit but
triply split in agreement with the 3-fold supercell. The atoms
belonging to the [Gd(μ-pmdc)1.5(H2O)3] layers are hardly
affected by the modulation, but the crystallization water
molecules move more sharply introducing remarkable dis-
tortions into the hydrogen bonding clusters. This effect
becomes clear when overlapping the hydrogen bonded rings
of the three superstructure positions in the vicinity of the O7w
atom. The t-plot carried out for the hydrogen bonding
distances involving the neighboring atoms of O7w leads to
the same conclusion. In fact, an additional O18w water
molecule has been included to complete the hydrogen bonding
scheme, which allows the water rings to be joined together and
thus lead to T7(0)A2 chains (Figure 5). These subtle variations
found for the geometry of the water clusters in the
commensurate structure, compared with the periodic struc-
tures, give some insights about how modulation proceeds
throughout the crystal.
Thermal studies of compound 1-Gd reveal that some of the
crystallization water molecules are lost spontaneously when
removing the compound from the mother liquors, as its release
starts from room temperature up to 340 °C, to render a Gd2O3
residue above 650 °C (Supporting Information). The XRPD
profiles of this compound show a rapid crystallinity decrease
above 30 °C due to the release of the crystallization water
molecules. The final anhydrous product at 150 °C is
amorphous.
Figure 5. (a) 3-fold supercell of compound 1-Gd (crystallization water
molecules have been omitted). (b) Distortions in the hydrogen
bonded clusters of the supercell vs the average structure. (c) t-plot
showing variations in the hydrogen bonding distances.
the distortion of the coordination polyhedra with regard to the
ideal shapes (jgsp and ttp), as indicated by the CShM [S(jgsp)
= 2.35−2.14 and S(ttp) = 3.29−3.10, for 2-La and 2-Ce,
respectively].
The coordination of two additional hexadentate μ4-
κ2N,O:κ2N′,O′:κO″:κO‴−pmdc ligands to the Ln(III) centers
imposes shorter Ln···Ln distances than those found through the
bis-bidentate bridge (mean values: 6.6 vs 7.3 Å). As a
consequence, the three chelating pmdc are forced to be more
twisted (dihedral angles: 40.5−56.0° for 2-Ce vs 45.8−78.0° for
1-Ln), exerting a remarkable distortion inside the six-
membered rings (Figure 7). The interlayer coordination of
the O272 atoms to the Ce2 atoms is also responsible for these
distortions, which become more evident when focusing the
junction among the rings, since they are tilted 63.3°. As a result,
the layers are no longer flat, as it can be derived from the mean
deviance of the Ln(III) atoms from the mean plane of the ring
Crystal Structure of 2-Ln. The crystal structure of
compounds 2-Ln consists of a 3D neutral open framework
built up from the junction of the Ln(III)−pmdc layers found in
compounds 1-Ln where two of the coordinated water
molecules of both Ln1 and Ln2 atoms have been replaced by
the noncoordinated carboxylate oxygen atoms from the pmdc
ligands (Figure 6). As a result, the Ln(III) centers are
surrounded by a N3O5Ow donor set that severely increases
H
dx.doi.org/10.1021/ic3009392 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 6. Coordination environments of the Ce1 and Ce2 atoms in compound 2-Ce.
Figure 7. Distorted six-membered rings that form the herringbone
pattern in 2-Ce.
Figure 8. Building up of the 3D framework by the coordination of the
nonchelating oxygen atoms.
and they resemble a herringbone pattern. The piling up of the
corrugated layers takes place along the crystallographic b axis
where the nonchelating carboxylate oxygen atoms (O182,
O272, and O282) of the hexadentate pmdc ligands serve as a
junction between layers by coordinating to the metal centers
belonging to the adjacent layers, leading to the overall 3D open
framework (Figure 8). The crystallization water molecule is
sited in a small isolated cavity of the compact building. The
sharp rise of the connectivity provides a (4,4,5,5-c) network
that possesses a new topology that has been named as jcr2, the
point symbol being (3.4.52.62)(3.4.52.64.72)(3.4.52.64.82)-
(4.5.63.8).
corresponds to the release of the crystallization water
molecules. The loss of the solvent does not promote any
significant structural change until 190 °C, where the loss of the
coordination water molecules takes place. The anhydrous
compound remains stable up to 390 °C, temperature at which
the oxidation of the pmdc ligand occurs to lead to PrO2 as final
product at 550 °C.
FTIR Analyses of the Ln/pmdc System. An exhaustive
analysis of the IR spectra of compounds 1-Ln and 2-Ln has let
us to accurately assign all bands (Figure 9), allowing us to
distinguish between the two different coordination modes of
the pmdc ligand; unlike what happened for the compounds of
the previously published Ln/pmdc/ox system in which the
The thermal analysis performed over compound 2-Pr shows
a first endothermic process between 50 and 120 °C, which
I
dx.doi.org/10.1021/ic3009392 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 9. IR spectra for compounds 1-Ho and 2-La showing the most characteristic differences.
oxalato and pmdc ligands are overlapping.10 All spectra share
the presence of broad and intense bands in frequency region of
3550−3200 cm−1, assigned to the characteristic peaks of OH
vibration of free water; the intense vibrations around 1650
cm−1 and 1360 cm−1 correspond to the asymmetric and
symmetric stretching vibrations of the carboxylate groups, while
at lower frequencies, the remaining bands are attributed to the
distortions originated in the aromatic ring of the pmdc ligand.
However, compounds 2-Ln exhibit a greater number of bands
derived from the simultaneous presence of the bis-chelating μ-
κ2N,O:κ2N′,O′ and the hexadentate μ4-κ2N,O:κ2N′,O′:κO″:κO‴
coordination modes. In particular, a strong splitting is observed
in the asymmetric O−C−O stretching into three other bands
that appear at higher wavenumbers, just as in its symmetric
stretching and its outward ring plane displacement, where the
relative intensity between the major bands is altered. The
change in the coordination also affects the vibrational modes of
the aromatic ring, as it goes from two to four the number of
observable bands around 1460 cm−1, while the intensity of the
displacement of the aromatic hydrogen atoms outward the ring
plane increases notoriously.
All these observations help us in analyzing the non-
structurally characterized A, B, C, and D compounds. As
shown in Supporting Information, compound A exhibits a
strong splitting in the asymmetric and symmetric O−C−O
vibrations, indicative of the presence of both the bis-bidentate
and hexadentate modes; whereas this splitting does not exist in
compound D, with almost the same bands of 1-Ln compounds.
The IR spectra of B and C compounds are representative of an
intermediate situation, where the splitting is perceptible but
weaker than that of A compound. These results are in good
agreement with the trend found for the effect of the
temperature induced entropy driven desolvation process and
the lanthanide contraction that takes place along the series
(Scheme 3). Once again, a larger ionic radius favors the
presence of sterically more hindered coordination modes and
higher temperatures are required for these modes to be present
for shorter ionic radii.
exception of the lanthanum compounds, have been measured.
The room temperature χMT values are close to those expected
for the ground states derived from their strong spin−orbit
coupling. The lanthanide(III) atoms can be classified into three
different groups according to their magnetic behavior. The
group consisting of Ce, Pr, Nd, Tb, Dy, Ho, Er, Tm, and Yb
stands out by the progressive decrease in the χMT product as
lowering the temperature derived from both a selective
depopulation of the excited Stark levels and antiferromagnetic
interaction between lanthanide ions. To obtain a rough
quantitative estimation of the magnetic interaction, an
expression for the χM can be defined by assuming a splitting
of the mj energy levels (H = ΔJz2) in an axial crystal field.35 It
must be noted, the energy separation of Sm(III) and Eu(III)
ions between the ground state and the first excited state is so
small that the first excited state may be thermally populated and
has to be therefore evaluated in the expressions of the magnetic
susceptibility.36 The Gd(III) ions are exceptions among the
8
lanthanide series because their S7/2 ground state allows to
analyze quantitatively their magnetic interactions by applying a
−1
spin-only Hamiltonian. Plots of χM vs T for the gadolinium
compounds (Supporting Information) obey the Curie−Weiss
law, giving C = 8.25(2) cm3 K mol−1, θ = −0.11(2) K and C =
13.35(1) cm3 K mol−1, θ = −1.32(2) K for compounds 1-Gd
and 2-Gd, respectively. Taking into account the 2D crystal
structure of compound 1-Gd, the magnetic data was analyzed
in the whole temperature range (5−300 K) using the
y37 with S = /2 (eq 1).
7
expression derived by Curel
́
Ng2β2 S(S + 1)(W + W )
3kT
1
2
χGd
=
(1 − u2)2
(1)
In the expression, u = coth[JS(S + 1)/kT] − kT/JS(S + 1), W1
= (1 + u2)2 + 4u2, W2 = 4u(1 + u2), J is the exchange coupling
parameter between adjacent spins through the pmdc bridge,
and all of the other parameters have their usual meanings. The
absence of an appropriate expression that considers the
magnetic interaction in the 3D framework has prevented us
to evaluate its magnitude in compound 2-Gd. However, the
obtained θ value indicates the greater antiferromagnetic
Magnetic Properties. The temperature-dependent mag-
netic susceptibility data of compounds 1-Ln and 2-Ln, with the
J
dx.doi.org/10.1021/ic3009392 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
character of the latter, although still being almost negligible, in
good agreement with the presence of a shorter second
exchange pathway through the nonchelating oxygen atoms of
the pmdc. The χM expressions derived for all the lanthanide-
(III) atoms can be found in the Supporting Information. The
best fitting results, for the 50−300 K temperature range (5−
300 K for compound 1-Gd), are gathered in Table 3. Figure 10
a
Table 3. Best Least-Squares Fits of the Magnetic Data
compd
g
Δ (cm−1
)
1-Nd
1-Tb
1-Dy
1-Ho
1-Er
0.71
1.47
1.30
1.23
1.23
1.13
1.12
0.86
0.76
0.73
1.47
1.26
1.23
1.23
1.13
1.12
g
2.20
0.18
0.11
0.29
0.40
0.31
3.32
5.18
3.26
2.04
0.21
0.12
0.25
0.38
0.33
3.34
1-Tm
1-Yb
2-Ce
2-Pr
2-Nd
2-Tb
2-Dy
2-Ho
2-Er
2-Tm
2-Yb
compd
λ (cm−1
)
1-Sm
1-Eu
0.28
5.00
0.29
5.01
225.39
373.57
229.74
362.17
2-Sm
2-Eu
compd
g
J (cm−1)/θ (K)
1-Gd
2-Gd
2.02
2.01
−0.01/−0.11
−/−1.32
a
Δ is the zero-field splitting parameter, and λ is the spin−orbit
coupling parameter.
shows the best fitting curve of the χMT product for three
selected compounds. The results show that the magnetic
interaction is almost negligible among the Ln(III) ions.
Luminescence Properties. Lanthanide compounds are
known for their photoluminescent properties, as they have been
used to develop several technological applications that
encompass not only fluorescent tubes and color televisions
but also immunoassays,38 optical amplifiers,39 and in future
perhaps metal−organic light-emitting diodes.40 The solid state
photoluminescence spectra of compounds 1-Nd, 1-Eu, 1-Tb, 2-
Nd, 2-Eu, and 2-Tb were recorded at room temperature, and
the characteristic emission bands for the corresponding Ln(III)
ions are shown in Figure 11.
Figure 10. χMT (o) vs T plot showing the best theoretical fit (−) for
compounds 1-Gd, 2-Ce, and 1-Sm.
site symmetry, whereas the latter is mainly sensitive to the
magnetic dipole effect created by the crystal field environment
and is practically not influenced by the chemical surroundings
5
7
of the ion. The intensity of the D0 → F2 transition, which is
responsible for brilliant red emission of these compounds, is
about 5 times stronger than that of the 5D0 → 7F1 band in both
cases, indicating that the Eu3+ ions occupy low-symmetry sites
without inversion center in agreement with the structural
analyses. Moreover, the presence of the weak symmetry-
forbidden 5D0 → 7F0 transition at 577 nm, which is induced by
crystal field J mixing, is allowed on non-inversion sites. The 5D0
The neodymium compounds 1-Nd and 2-Nd display an
4
emission band at 1061 nm (4F3/2 → I11/2) and another one at
1332 nm (4F3/2 → 4I13/2) in the NIR region under excitation at
→
7F3 transition is very weak according to its forbidden
4
800 nm in the F5/2 level. The first band, which is potentially
character (both in magnetic and induced electric dipole
schemes). Regarding the terbium compounds, both of them
emit green luminescence upon excitation at 355 nm with the
bands peaking at 490, 540, 580, 620, and 650 nm. These are
applicable to laser emission, fluoroimmunoassays, and in vivo
detection,41 is dominant in both cases. Upon excitation at 355
nm, compounds 1-Eu and 2-Eu display intense red
luminescence with five bands occurring at 580, 590, 615, 665,
and 690 nm that can be assigned to D0 → FJ transitions (J =
0, 1, 2, 3, and 4, respectively).42 The 5D0 → 7F2 and 5D0 → 7F1
transitions are ED and MD, respectively, which makes the
former (known as hypersensitivity) to be extremely sensitive to
5
7
5
7
originated from the characteristic D4 → FJ transitions (J = 6,
5, 4, 3, and 2). As expected, in higher Tb3+ concentrated
systems, although the 355 nm excitation light populates level
5
5D3 the observed emission corresponds to D4 level being the
K
dx.doi.org/10.1021/ic3009392 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 11. Room temperature emission spectra obtained under excitation at 800 nm for 1-Nd and 2-Nd; and at 355 nm for compounds 1-Eu, 2-Eu,
5
1-Tb, and 2-Tb. The emissions around 650 nm in compounds 1-Tb and 2-Tb correspond to transitions D4→7FJ (J = 0, 1, 2).
42
5
7
intensity of the D4 → F5 transition the strongest one. This
transition has the largest probability for both electric-dipole and
magnetic-dipole induced transitions.43
decrease of coordination water molecules present in the
chromophores of compounds 2-Ln. Concerning the emission
efficiency of the Ln(III) center, which decreases in the relative
order Tb(III) > Eu(III) > Nd(III), this behavior can be
understood by taking into account that the extent of excited
state deactivation by O−H vibronic coupling from water
molecules is inversely proportional to the energy gap between
the emitting state and the ground state manifold.47 This energy
gap is approximately 5200 cm−1 for Nd3+ (4F3/2→4I15/2), 12200
cm−1 for Eu3+ (5D0→7F6), and 14800 cm−1 for Tb3+
(5D4→7F0), so that an efficient coupling of the Nd3+ excited
state occurs. For Eu3+ excited states the coupling is to the third
The measurement of the experimental emission decay curves
from the F3/2 (Nd3+), D0 (Eu3+), and D4 (Tb3+), excited
states for all compounds allowed us to effectively estimate the
lifetimes in each case (Table 4). The decay curves are shown in
4
5
5
Table 4. Lifetime Values of Compounds
compd
τ (μs)
compd
τ (μs)
1-Nd
1-Eu
1-Tb
19
2-Nd
2-Eu
2-Tb
22
391
783
290
vibrational overtone of proximate OH oscillators (νOH
∼
a
3300−3500 cm−1), and to the fourth harmonic in the case of
Tb3+, consistent with the observed less efficient quenching
because the Franck−Condon overlap factor is less favorable.
593
a
This value corresponds to ⟨τ⟩.
the Supporting Information and can be fitted to a single-
exponential function, to a good approximation, except for
compound 1-Tb. The decay curve in this case can be fitted to a
double-exponential function, [I = A + A1 exp(−t/τ1) + A2
exp(−t/τ2), where τ1 and τ2 are the fast and slow components
of the luminescent lifetimes and A, A1, and A2 are the fitting
parameters]. In this case, an average lifetime can be defined as
⟨τ⟩ =A1τ21 + A2τ22/A1τ1 + A2τ2; however, in order to reduce the
fitting errors in the calculation of the average lifetime for this
sample, we have used in Table 4 the average lifetime defined by
the equivalent expression ⟨τ⟩ = (∫ tI(t)dt)/(∫ I(t)dt) .
In light of these results, a significant increase of the lifetimes
of compounds 2-Ln in comparison with those of compounds 1-
Ln can be inferred. There are several alternative deactivation
pathways of the excited state of a lanthanide(III) center in the
form of nonradiative processes, such as the back-energy transfer
to the sensitizer,44 electron transfer quenching,44c,45 and most
importantly, quenching by matrix vibrations. In particular, the
O−H oscillators are very effective quenchers of luminescence,46
explaining the observed increase in the lifetime according to the
CONCLUSIONS
■
In the present work, the crystal structures of two families of
lanthanide/pyrimidine-4,6-dicarboxylate compounds are re-
ported, both of which are built up from the recurrent
bisbidentate coordination mode of the pmdc ligand. It is
well-known that the N-heteroaromatic-carboxylate ligands
generate oxalate through its partial decomposition in the
presence of lanthanides, so the success in obtaining these
oxalate-free compounds requires a rigorous control over the
synthetic conditions. The exhaustive study carried out to check
the influence of the reaction temperature, pH, atmosphere, and
the lanthanide ionic size has enabled us to access to a wide
unexplored diversity of Ln/pmdc compounds, showing not
only the validity of the method but also that the entropy driven
desolvation pathway is fulfilled. The distorted honeycomb
layers found in compounds 1-Ln are packed generating
interlamellar voids occupied by highly flexible crystallization
water molecules aggregates, which are responsible for the
modulation found for the low temperature phase of the 1-Gd
L
dx.doi.org/10.1021/ic3009392 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(3) (a) Chen, B. L.; Yang, Y.; Zapata, F.; Lin, G. N.; Qian, G. D.;
Lobkovsky, E. B. Adv. Mater. 2007, 19, 1693. (b) Harbuzaru, B. V.;
Corma, A.; Rey, F.; Atienzar, P.; Jorda, J. L.; Garcia, H.; Ananias, D.;
Carlos, L. D.; Rocha, J. Angew. Chem., Int. Ed. 2008, 47, 1080. (c) Xu,
H.; Liu, F.; Cui, Y. J.; Chen, B. L.; Qian, G. D. Chem. Commun. 2011,
47, 3153. (d) Wong, K. L.; Law, G. L.; Yang, Y. Y.; Wong, W. T. Adv.
Mater. 2006, 18, 1051.
compound. The replacement of two of the three coordination
water molecules by two μ4-κ2N,O:κ2N′,O′:κO″:κO‴−pmdc
ligands in the chromophores of compounds 2-Ln allow the
layers to be joined together, rendering a compact 3D
framework with the new topology jcr2. The variability of the
coordination modes of the pmdc, including that of the
nonstructurally characterized A−D compounds, has been
monitored by the FTIR spectroscopy.
(4) (a) Kiritsis, V.; Michaelides, A.; Skoulika, S.; Golhen, S.; Ouahab,
L. Inorg. Chem. 1998, 37, 3407. (b) Long, D. L.; Blake, A. J.;
The magnetic behavior of the compounds is dominated by
the spin−orbit coupling and the ligand field perturbation with
an almost negligible contribution of the exchange coupling
between the lanthanide atoms that is practically independent of
the coordination mode. Luminescence measurements per-
formed over the neodymium, europium, and terbium
compounds show the characteristic near-IR, red, and green
emissions, respectively. Furthermore, the lifetimes of these
compounds estimated from the experimental emission decay
curves from the excited states show a marked increase for
compounds 2-Ln compared to 1-Ln, which is attributed to the
decrease of coordination water molecules present in the
chromophores of the former as they are replaced by carboxylic
oxygen atoms of the pmdc ligands.
Champness, N. R.; Schroder, M. Chem. Commun. 2000, 1369.
̈
(c) Long, D. L.; Blake, A. J.; Champness, N. R.; Wilson, C.;
Schroder, M. J. Am. Chem. Soc. 2001, 123, 3401. (d) Wang, Z.; Jin, C.
M.; Shao, T.; Li, Y. Z.; Zhang, K. L.; Zhang, H. T.; You, X. Z. Inorg.
Chem. Commun. 2002, 5, 642.
(5) (a) Fer
Dutour, J.; Surble,
(b) Kesanli, B.; Cui, Y.; Smith, M. R.; Bittner, E. W.; Bockrath, B.
C.; Lin, W. B. Angew. Chem., Int. Ed. 2005, 44, 72.
́
ey, G.; Mellot-Draznieks, C.; Serre, C.; Millange, F.;
́
S.; Margiolaki, I. Science 2005, 309, 2040.
(6) Pearson, R. G. J. Am. Chem. Soc. 1963, 85, 3533.
(7) (a) Serre, C.; Fer
(b) Serpaggi, F.; Ferey, G. J. Mater. Chem. 1998, 8, 2749. (c) Serpaggi,
F.; Ferey, G. Inorg. Chem. 1999, 38, 4741. (d) Yang, A.−H.; Gao, H.−
́
ey, G. J. Mater. Chem. 2002, 12, 3053.
́
́
L.; Cui, J.−Z.; Zhao, B. CrystEngComm 2011, 13, 1870. (e) Yan, X. H.;
Li, Y. F.; Wang, Q.; Huang, X. G.; Zhang, Y.; Gao, C. J.; Liu, W. S.;
Tang, Y.; Zhang, H. R.; Shao, Y. L. Cryst. Growth Des. 2011, 11, 4205.
(f) Zhuang, G. L.; Kong, X. J.; Long, L. S.; Huang, R. B.; Zheng, L. S.
CrystEngComm 2010, 12, 2691.
ASSOCIATED CONTENT
■
S
* Supporting Information
́
(8) (a) Henry, N.; Costenoble, S.; Lagrenee, M.; Loiseau, T.;
Abraham, F. CrystEngComm 2011, 13, 251. (b) Cai, B.; Yang, P.; Dai,
J.-W.; Wu, J.-Z. CrystEngComm 2011, 13, 985. (c) Cepeda, J.; Beobide,
Elemental analyses, IR data, X-ray analysis, thermogravimetric
measurements, magnetic susceptibility data, emission decay
curves, and cif files. This material is available free of charge via
the Internet at http://pubs.acs.org.
G.; Castillo, O.; Luque, A.; Per
Des. 2012, 12, 1501.
(9) (a) Beobide, G.; Wang, W. G.; Castillo, O.; Luque, A.; Roman
́
ez-Yanez, S.; Roman, P. Cryst. Growth
́
́
̃
́
, P.;
AUTHOR INFORMATION
Tagliabue, G.; Galli, S.; Navarro, J. A. R. Inorg. Chem. 2008, 47, 5267.
■
(b) Beobide, G.; Castillo, O.; Luque, A.; García-Couceiro, U.; García-
Corresponding Author
Teran
Castillo, O.; Cepeda, J.; Luque, A.; Per
Vallejo-Sanchez, D. Eur. J. Inorg. Chem. 2011, 68.
(10) Cepeda, J.; Balda, R.; Beobide, G.; Castillo, O.; Fernan
Luque, A.; Perez-Yanez, S.; Roman, P.; Vallejo-Sanchez, D. Inorg.
Chem. 2011, 50, 8437.
(11) Hunt, R. R.; McOmie, J. F. W.; Sayer, E. R. J. Chem. Soc. 1959,
525.
́
, J. P.; Roman
́
, P. Dalton Trans. 2007, 2669. (c) Beobide, G.;
*Email: oscar.castillo@ehu.es, javier.cepeda@ehu.es. Fax: +34-
94601-3500.
́
ez-Yanez, S.; Roman, P.;
́
́
̃
́
Notes
́
dez, J.;
The authors declare no competing financial interest.
́
́
́
́
̃
ACKNOWLEDGMENTS
■
Financial support from the Gobierno Vasco (IT477-10 and
IT331-07) is gratefully acknowledged. We also thank
(12) Earnshaw, A. Introduction to Magnetochemistry; Academic Press:
London, 1968.
(13) CrysAlisPro, version 1.171.35.15; Agilent Technologies: Yarnton,
U.K., 2011.
(14) Altomare, A.; Cascarano, M.; Giacovazzo, C.; Guagliardi, A. J.
́
Universidad del Pais Vasco/Euskal Herriko Unibertsitatea for
UFI11/53 and predoctoral fellowships (PIFA01/2007/021).
Technical and human support provided by SGIker (UPV/
EHU, MICINN, GV/EJ, ESF) is also acknowledged.
Appl. Crystallogr. 1993, 26, 343.
(15) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837.
(16) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112.
(17) Petricek, V.; Dusek, M.; Palatinus, L. Jana2006, The Crystallo-
graphic Computing System; Institute of Physics: Praha, Czech Republic,
2006.
(18) (a) Rodríguez-Carvajal, J. FULLPROF, Program Rietveld for
Pattern Matching Analysis of Powder Patterns; Abstacts of the Satellite
Meeting on Powder Diffraction of the XV Congress of the IUCr:
Toulouse, France, 1990, 127. (b) Rodríguez-Carvajal, J. FULLPROF
REFERENCES
■
(1) (a) Kitagawa, S.; Kondo, M. Bull. Chem. Soc. Jpn. 1998, 71, 1739.
(b) Batten, S. R.; Robson, R. Angew. Chem., Int. Ed. 1998, 37, 1460.
(c) Blake, A. J.; Champness, N. R.; Hubberstey, P.; Li, W. S.;
Withersby, M. A.; Schroder, M. Coord. Chem. Rev. 1999, 183, 117.
̈
(d) Moulton, B.; Zaworotko, M. Chem. Rev. 2001, 101, 1629.
(e) Moulton, B.; Lu, J.; Mondal, A.; Zaworotko, M. J. Chem. Commun.
2001, 863.
́
2000, version 2.5d; Laboratoire Leon Brillouin (CEA-CNRS), Centre
(2) (a) Rowsell, J. L. C.; Yaghi, O. M. J. Am. Chem. Soc. 2006, 128,
́
d’Etudes de Saclay, Gif sur Yvette Cedex: France, 2003.
(19) Zhang, X. M. Coord. Chem. Rev. 2005, 249, 1201.
(20) (a) Knope, K. E.; Cahill, C. L. Inorg. Chem. 2007, 46, 6607.
1304. (b) Costes, J. P.; Nicodeme, F. Chem.Eur. J. 2002, 8, 3442.
́
(c) Xu, G. F.; Wang, Q. L.; Gamez, P.; Ma, Y.; Clerac, R..; Tang, J.;
Yan, S. P.; Cheng, P.; Liao, D. Z. Chem. Commun. 2010, 46, 1506.
(d) Hernandez-Molina, M.; Ruiz-Perez, C.; Lopez, T.; Lloret, F.; Julve,
(b) Rodríguez-Dieg
(c) Weng, D. F.; Mu, W. H.; Zheng, X. J.; Fang, D. C.; Jin, L. P. Inorg.
Chem. 2008, 47, 1249. (d) Thuery, P. CrystEngComm 2010, 12, 1905.
́
uez, A.; Colacio, E. Chem. Commun. 2006, 4140.
́
́
́
M. Inorg. Chem. 2003, 42, 5456. (e) Hou, H. W.; Li, G.; Li, L. K.; Zhu,
Y.; Meng, X. R.; Fan, Y. T. Inorg. Chem. 2003, 42, 428. (f) Lee, W. R.;
Ryu, D. W.; Lee, J. W.; Yoon, J. H.; Koh, E. K.; Hong, C. S. Inorg.
Chem. 2010, 49, 4723. (g) Zhang, Z. H.; Song, Y.; Okamura, T.;
Hasegawa, Y.; Sun, W. Y.; Ueyama, N. Inorg. Chem. 2006, 45, 2896.
́
(21) Barrett Adams, D. M. Y.; Kahwa, I. A.; Mague, J. T. New J. Chem.
1998, 22, 919.
(22) Evans, O. R.; Lin, W. Cryst. Growth Des. 2001, 1, 9.
M
dx.doi.org/10.1021/ic3009392 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(23) (a) Zhang, J. P.; Lin, Y. Y.; Huang, X. C.; Chen, X. M. Eur. J.
Inorg. Chem. 2006, 3407. (b) Huh, H. S.; Lee, S. W. Bull. Korean Chem.
Soc. 2006, 27, 1839.
(47) Beeby, A.; Clarkson, I. M.; Dickins, R. S.; Faulkner, S.; Parker,
D.; Royle, L.; de Sousa, A. S.; Williams, J. A. G.; Woods, M. J. Chem.
Soc., Perkin Trans. 2 1999, 493.
(24) (a) Li, X.; Cao, R.; Sun, D.; Shi, Q.; Bi, W.; Hong, M. Inorg.
Chem. Commun. 2003, 6, 203. (b) Rowland, C. E.; Cahill, C. L. Inorg.
Chem. 2010, 49, 6716. (c) Andrews, M. B.; Cahill, C. L.
CrystEngComm 2011, 13, 7068.
(25) (a) Fujishiro, K.; Mitamura, S. Bull. Chem. Soc. Jpn. 1989, 62,
786. (b) McNelis, E. J. Org. Chem. 1965, 30, 1209.
(26) Knope, K. E.; Kimura, H.; Yasaka, Y.; Nakahara, M.; Andrews,
M. B.; Cahill, C. L. Inorg. Chem. 2012, 51, 3883.
(27) (a) Mahata, P.; Prabu, M.; Natarajan, S. Inorg. Chem. 2008, 47,
́
8451. (b) Forster, P. M.; Burbank, A. R.; Livage, C.; Ferey, G.;
Cheetam, A. K. Chem. Commun. 2004, 368.
(28) Shannon, R. D. Acta Crystallogr. 1976, A32, 751.
(29) Guzei, I. A.; Wendt, M. Dalton Trans. 2006, 3991.
(30) (a) Llunell, M.; Casanova, D.; Cirera, J.; Bofill, J. M.; Alemany,
P.; Alvarez, S.; Pinsky, M.; Avnir, D. SHAPE (1.7); University of
Barcelona: Barcelona, 2010. (b) Ruiz-Martínez, A.; Casanova, D.;
Alvarez, S. Chem.Eur. J. 2008, 14, 1291.
(31) (a) Shi, F.−N.; Cunha−Silva, L.; Trindade, T; Paz, F. A. A.;
Rocha, J. Cryst. Growth Des. 2009, 9, 2098. (b) Li, S.; Chen, Y.; He,
H.−M.; Ma, Y.−F. Acta Crystallogr. 2009, E65, m411. (c) Huang, Y.−
G.; Wu, B.−I.; Yuan, D.−Q.; Xu, Y.−Q.; Jiang, F.−I.; Hong, M.−C.
Inorg. Chem. 2007, 46, 1171.
(32) (a) TOPOS Main Page. http://www.topos.ssu.samara.ru
(accessed Nov. 2010). (b) Blatov, V. A. IUCR CompComm Newsletter;
2006, p 4.
(33) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7.
(34) Infantes, L.; Chisholm, J.; Motherwell, S. CrystEngComm 2003,
5, 480.
(35) (a) Kahwa, I. A.; Selbin, J.; O’Connor, C. J.; Foise, J. W.;
McPherson, G. L. Inorg. Chim. Acta 1988, 148, 265. (b) Ke, H.; Zhao,
L.; Xu, G.−F.; Guo, Y.−N.; Tang, J.; Zhang, X.−Y.; Zhang, H.−J.
Dalton Trans. 2009, 10609. (c) Li, B.; Gu, W.; Zhang, L. Z.; Qu, J.;
Ma, Z. P.; Liu, X.; Liao, D. Z. Inorg. Chem. 2006, 45, 10425.
(d) Ouyang, Y.; Zhang, W.; Xu, N.; Xu, G. F.; Liao, D. Z.; Yoshimura,
K.; Yan, S. P.; Cheng, P. Inorg. Chem. 2007, 46, 8454. (e) O’Connor,
C. J. Prog. Inorg. Chem. 1982, 29, 203.
(36) (a) Li, Y.; Zheng, F. K.; Liu, X.; Zou, W. Q.; Guo, G. C.; Lu, C.
Z.; Huang, J. S. Inorg. Chem. 2006, 45, 6308. (b) Andruh, M.;
Bakalbassis, E.; Kahn, O.; Trombe, J. C.; Porcher, P. Inorg. Chem.
1993, 32, 1616. (c) Wang, Y.; Li, X. L.; Wang, T. W.; Song, Y.; You, X.
Z. Inorg. Chem. 2010, 49, 969.
(37) (a) Curel
B 1998, 245, 263. (c) Delgado, F. S.; Kerbellec, N.; Ruiz-Per
Cano, J.; Lloret, F.; Julve, M. Inorg. Chem. 2006, 45, 1012.
́
y, J. Europhys. Lett. 1995, 32, 529. (b) Curel
́
y, J. Physica
́
ez, C.;
(38) (a) Hemmila, I. J. Alloys. Compd 1995, 225, 480. (b) Moynagh,
̈
J.; Schimmel, H. Nature 1999, 400, 105.
(39) Desurvire, E. Phys. Today 1994, 47 (1), 20.
(40) Baldo, A. M.; Thompson, M. E.; Forrest, S. R. Pure Appl. Chem.
1999, 71, 2095.
(41) (a) Janiak, C. Dalton Trans. 2003, 2781. (b) Bunzli, J. C. G.;
̈
Eliseeva, S. V. J. Rare Earths 2010, 28, 824.
(42) Bunzli, J. C. G. Lanthanide Probes in Life, Chemical, and Earth
̈
Sciences: Theory and Practice; Elsevier Science Publications: Am-
sterdam, 1989.
(43) Yen, W. M.; Shionoya, S.; Yamamoto, H. Phosphor Handbook;
CRC Press: Athens, 2007.
(44) (a) Bhaumik, M. L. J. Chem. Phys. 1964, 40, 3711. (b) Sabbatini,
N.; Guardigli, M.; Manet, I. Adv. Photochem. 1997, 23, 213. (c) Prodi,
L.; Maestri, M.; Balzani, V.; Lehn, J. M.; Roth, C. Chem. Phys. Lett.
1991, 180, 45.
(45) Sabbatini, N.; Perathoner, S.; Lattanzi, G.; Dellonte, S.; Balzani,
V. J. Phys. Chem. 1987, 91, 6136.
(46) (a) Kropp, J. L.; Windsor, M. W. J. Chem. Phys. 1965, 42, 1599.
(b) Horrocks, W. D.; Sudnick, D. R. Acc. Chem. Res. 1981, 14, 384.
N
dx.doi.org/10.1021/ic3009392 | Inorg. Chem. XXXX, XXX, XXX−XXX