The effect of 4-halogenobenzoate ligands on luminescent and structural properties of lanthanide complexes: Experimental and theoretical approaches

Article abstract of DOI:10.1039/c4nj01701c

The ligands 4-fluorobenzoate (4-fba), 4-chlorobenzoate (4-cba), 4-bromobenzoate (4-bba) and 4-iodobenzoate (4-iba) were chosen in order to synthesize europium(iii), gadolinium(iii) and terbium(iii) complexes and compare the effect of halogens on their phy

Full text of DOI:10.1039/c4nj01701c

NJC
View Article Online
View Journal
PAPER
The effect of 4-halogenobenzoate ligands
on luminescent and structural properties of
lanthanide complexes: experimental and
theoretical approaches†
Cite this:DOI: 10.1039/c4nj01701c
Jorge H. S. K. Monteiro,a Ana de Bettencourt-Dias,b Italo O. Mazalia and
Fernando A. Sigoli*a
The ligands 4-fluorobenzoate (4-fba), 4-chlorobenzoate (4-cba), 4-bromobenzoate (4-bba) and
4-iodobenzoate (4-iba) were chosen in order to synthesize europium(III), gadolinium(III) and terbium(III)
complexes and compare the effect of halogens on their physical chemistry and luminescent properties.
The homobimetallic complex [Eu(4-iba)3(H2O)(dmf)]2 crystallizes in the monoclinic P21/c space group
with unit cell parameters a = 8.3987(9) Å, b = 25.314(3) Å, c = 14.1255(17) Å, and b = 105.347(2)1. FTIR
spectroscopy indicates that the bidentate bridging mode of the carboxylato ligand was present in all
complexes while bidentate chelate and a mixture of bidentate bridging and chelate modes were also
found. According to emission spectra profiles and the Judd–Ofelt parameters the halogen of ligand mole-
cules modifies the chemical environment symmetry around the europium(III) ion in their respective
complexes. The complexes [Eu(4-fba)3(H2O)2] and [Eu(4-iba)3(H2O)2] have the highest symmetry around
the europium(III) while the complexes [Eu(4-cba)3]ꢀ2H2O, [Eu(4-bba)3]ꢀ5/2H2O and [Eu(4-iba)3(H2O)(dmf)]2
have the lowest. The different halogens at the para position do not change the covalence degree of Eu–O
bonds significantly, however they play a role in the ligand to metal charge transfer energies. The highest
non-radiative energy transfer rates from ligand to europium(III) were found for the complexes [Eu(4-cba)3]ꢀ
2H2O and [Eu(4-bba)3]ꢀ5/2H2O.
Received (in Victoria, Australia)
30th September 2014,
Accepted 16th December 2014
DOI: 10.1039/c4nj01701c
www.rsc.org/njc
interesting due to their spectroscopic properties, usually high
energy of triplet level and they show several coordination modes,
Introduction
The luminescent properties of lanthanide(III) compounds, due to such as bidentate bridging, bidentate chelate, monodentate, etc.5
the 4f–4f transitions, have been applied in several fields such as Li and co-workers6 obtained a dimeric terbium(III) complex using
illumination, displays, OLEDs, sensors, luminescent markers, the ligand 2-fluorobenzoate (2-fba). Xu and co-workers7 obtained a
etc.1–3 A strategy to increase the emission intensity of the similar dimer of europium(III) using the ligands 4-fluorobenzoate
lanthanides(III) is the coordination of organic chromophores to (4-fba) and phenanthroline. Zhang8 showed an example of a dimer
them. The absorbed energy by the ligand may be transferred from of dysprosium(III) with the ligands 4-chlorobenzoate (4-cba) and
the ligand to the lanthanide via excited singlet, charge transfer phenanthroline. Some examples of dimers of dysprosium(III)
states or triplet levels, the triplet - lanthanide(III) normally being complexes was also shown by Li and co-workers9 using the ligands
the most efficient one due to the energy resonance and selection 2-chlorobenzoate (2-cba) and bipyridine or 2-bromobenzoate
rules imposed by lanthanide ions.2,4 Carboxylate ligands are (2-bba) and phenanthroline. Zhang and co-workers10 showed
an example of a dimer for a samarium(III) complex with the
4-bromobenzoate (4-bba) and the bipyridine as the auxiliary
a Department of Inorganic Chemistry, Institute of Chemistry,
ligand. Li and co-workers11 reported the synthesis, single crystal
˜
University of Campinas, Campinas, Sao Paulo, Brazil.
structure and some photophysical properties of europium(III)
complexes with the ligands 2-iodobenzoate (2-iba) and water or
phenanthroline or 2,20-bipyridine.
In all research results cited above, the syntheses of com-
plexes, single crystal structures and different coordination
modes are highlighted for complexes containing ligand deriva-
tives of benzoic acid.6–11 The presence of different halogens
E-mail: fsigoli@iqm.unicamp.br
b Department of Chemistry, University of Nevada, Reno, NV 89557, USA
† Electronic supplementary information (ESI) available: Elemental analysis, TGA
and DTA curves, FT-IR spectra, UV-Vis absorption spectra, time resolved phos-
phorescence spectra of the gadolinium(III) complexes, emission decay curves,
ground state geometry and charge factors (g) and polarizabilities (a). CCDC
1024660. For ESI and crystallographic data in CIF or other electronic format
see DOI: 10.1039/c4nj01701c
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem.

View Article Online
Paper
NJC
General procedure for [Ln(L)3(H2O)x]ꢀ(H2O)y synthesis14
Sodium hydroxide was added to aqueous suspensions of
ligands in a 1 : 1 (L : OHꢁ) molar ratio and stirred for 30 min
at 80 1C. The aqueous solutions of deprotonated ligands were
added to aqueous LnCl3 solution in a 1 : 3 molar ratio (Ln : L).
White precipitates were formed and the suspensions were kept
under stirring at 80 1C for 2 h. The white solids were filtered off
and washed with hot water and methanol. Single crystals of the
complex [Eu(4-iba)3(H2O)(dmf)]2 suitable for X-ray diffraction
were obtained by slow diffusion of water into a dimethylforma-
mide (dmf) solution of [Eu(4-iba)3(H2O)2].
Fig. 1 Molecular structures of the ligands: 4-fluorbenzoate (4-fba),
4-chlorobenzoate (4-cba), 4-bromobenzoate (4-bba) and 4-iodobenzoate
(4-iba).
Characterization
(F, Cl, Br or I) may change some physical chemistry properties
of the lanthanides(III) complexes, such as (i) solubility in polar Unless otherwise indicated, all data were collected at a constant
solvents due to the presence of the fluorine atom in the ligand temperature of 298 ꢂ 1 K. The chemical stoichiometries of the
structure; (ii) changes in the Eu–O covalence degree upon complexes were determined by Ln3+ titration using a standard
addition of the chlorine atom in the ligand structure, as 0.01 mol Lꢁ1 EDTA solution and elemental analysis of carbon
reported by Malta and co-workers;12 (iii) increase in the inter- and hydrogen (Perkin Elmer 2400). Thermogravimetric analysis
system crossing (ISC) efficiency due to the presence of the (TA instruments SDTQ600) was carried out using a synthetic air
iodine atom in the ligand structure as reported by Lower and flow (100 mL minꢁ1) at a heating rate of 10 1C minꢁ1. Infrared
El-Sayed;13 (iv) the triplet level energy and (v) the energy of spectra (FT-IR Bomen FTLA 2000) data were obtained in trans-
ligand to metal charge transfer states (LMCT) of europium(III) mission mode using KBr pellets. The diffuse reflectance spectra
complexes.
(DRS) of the europium(III) and gadolinium(III) complexes were
The understanding of the correlation between the ligand obtained in the solid state in a Cary 5000 UV-Vis-NIR using
molecular structure and the energy transfer processes from BaSO4 as the standard, using the absorption mode. The photo-
ligand to lanthanide(III) ions provides a good background for luminescence data were obtained with the samples in the solid
planning future luminophores with high emission quantum state on a Fluorolog-3 spectrofluorometer (Horiba FL3-22-iHR320),
yields. Therefore, the main goal of this work is the evaluation with double-gratings (1200 gr mmꢁ1, 330 nm blazed) in the
of the halogen effect under the structural and photophysical excitation monochromator and double-gratings (1200 gr mmꢁ1
,
properties of a series of 4-halogenobenzoate lanthanide(III) com- 500 nm blazed) in the emission monochromator. An ozone-free
plexes. The ligands chosen for this purpose are shown in Fig. 1. xenon lamp of 450 W (Ushio) was used as a radiation source.
The ligands shown in Fig. 1 allow a good comparison of the The excitation spectra were obtained between 200 and 600 nm
halogen effect on the photophysical properties of the and they were corrected in real time according to the lamp
lanthanide(III) complexes due to a similar coordination unit intensity and the optical system of the excitation monochromator
(the carboxylate group) and molecular/crystalline structures. In using a silicon diode as a reference detector. The emission spectra
order to compare the changes in the molecular and crystalline were obtained between 400 and 750 nm using the front face mode
structures and photophysical properties of the lanthanide(III) at 22.51. All emission spectra were corrected according to the
complexes, results from single crystal X-ray diffraction, TGA, optical system of the emission monochromator and the photo-
FT-IR, DRS, luminescence spectroscopy and ground state geo- multiplier response (Hamamatsu R928P). The time-resolved phos-
metries and energy transfers theoretical calculations will be phorescence emission spectra of the analogous gadolinium(III)
presented in this work.
complexes were obtained at B77 K using a TCSPC system with
successive delay increments, in order to get only the emissions
from triplet levels of the ligands. The energy values of the ligand
triplet level were obtained using two approaches: (i) fitting a
tangent to the highest energy edge of the emission spectra or
(ii) the maximum of the highest energy vibrational-coupled band
Experimental section
General procedure for LnCl3 (Ln = Eu3+, Gd3+ or Tb3+)
All commercially obtained reagents were of analytical grade (0–0 phonon) obtained from the deconvolution of the emission
and were used as received. The lanthanide(III) chlorides, LnCl3 spectra.15 The emission decay curves were obtained with a pulsed
(Ln = Eu3+, Gd3+ or Tb3+), were obtained by the addition of 150 W xenon lamp using a TCSPC system. The absolute quantum
hydrochloric acid into a lanthanide oxide aqueous suspension. yields were measured using a Quanta-j (Horiba F-309) integrating
After complete dissolution of the oxides the pH of the solution sphere equipped with an optical-fibers bundle (NA = 0.22 –
was adjusted to B4 with diluted ammonium hydroxide aqueous Horiba-FL-3000/FM4-3000) using excitation centered in the f–f
solution (0.1 mol Lꢁ1). In the case of terbium oxide a few drops transitions due to the instrumental setup. The Judd–Ofelt (JO)
of hydrogen peroxide (29% wt/vol) were added in order to reduce intensity parameters (O2 and O4) and the efficiency parameters
terbium(IV) to terbium(III).
(Arad, Atot and Z) were calculated using the equations widely
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
NJC
Paper
described in the literature.4 The number of coordinated water
molecules (q) was determined using eqn (1), proposed by Horrocks
et al. considering that the O–H oscillator is the only quenching
route of the emitting level.16
Table 2 Eu–O bond distances [Å] in [Eu(4-iba)3(H2O)(dmf)]2
Bond
d/Å
Bond
d/Å
Eu1–O50
Eu1–O7
Eu1–O8
Eu1–O1
2.332(2)
2.355(2)
2.377(2)
2.399(2)
Eu1–O6
Eu1–O3
Eu1–O2
Eu1–O4
2.407(2)
2.449(2)
2.525(2)
2.545(2)
ꢀ
ꢁ
1
1
q ¼ A ꢀ
ꢁ
ꢁ a
(1)
t
t
H2O
D2O
Symmetry code: 2 ꢁ x, 1 ꢁ y, 1 ꢁ z.
where A = 1.1, a = 0.31 and tD O = trad for Eu(III).17
2
X-ray crystallographic characterization
T = 1D. The theoretical JO intensity parameters were calculated
using the adequate equations and adjusting, in the physical
acceptable range,4 the polarizability (a) and the charge factors (g)
of the ligands in order to fit the theoretical JO intensity parameters
with the experimental ones. The excited state calculations were
performed using the ORCA software25 using the INDO/S-CIS26 with
the lanthanide replaced by a +3e point charge.4,23 The transfer and
Crystal data, data collection and refinement details for
[Eu(4-iba)3(H2O)(dmf)]2 are given in Table 1. Eu–O bond
lengths for this complex are shown in Table 2. A suitable crystal
was mounted on a glass fiber and placed in the low-
temperature nitrogen stream (Oxford Cryosystems 700 series).
Data were collected on a Bruker APEX II DUO area detector
diffractometer equipped with a low-temperature device, using
graphite-monochromated Mo-Ka radiation (l = 0.71073 Å). Data
were measured using a strategy combining o and j scans of
0.31 per frame and an acquisition time of 10 s per frame. Multi-
scan absorption corrections were applied. Cell parameters were
retrieved using the SMART18 software and refined using SAINT-
Plus19 on all observed reflections. Data reduction and correction
for Lp and decay were performed using the SAINTPlus soft-
ware.19 Absorption corrections were applied using SADABS.20
The structures were solved by direct methods and refined by
5
back transfer energy rates from ligand triplet levels to D1,0
europium(III) levels as well the theoretical quantum efficiency
and quantum yield were calculated using the adequate kinetics
equations described by Malta and co-workers27–29 implemented in
the LUMPAC software.30
Results and discussion
X-ray quality single crystals of the complex [Eu(4-iba)3(H2O)(dmf)]2
were obtained by slow diffusion of water into a dmf solution of
least-squares methods on F2 using the SHELXTL program pack- [Eu(4-iba)3(H2O)2]. It crystallizes in the monoclinic P21/c space
age.21 All non-hydrogen atoms were refined anisotropically. group, with two homobimetallic dimers in the unit cell. Details
Almost all hydrogen atoms were added geometrically, and their of the crystallographic characterization are summarized in Table 1
parameters constrained to the parent site. The hydrogen atoms and selected bond distances are given in Table 2. The asymmetric
of the coordinated water molecule were located on the difference unit contains one half of the molecule, with the other half being
map. CCDC 1024660 contains the supplementary crystallo-
graphic data for [Eu(4-iba)3(H2O)(dmf)]2 in this paper.
generated by inversion. Each of the two Eu centers binds to two
bidentate 4-iba ligands, as well as one water and one dmf molecule;
two additional 4-iba ligands which bridge the two metal centers, as
shown in Fig. 2a, complete the coordination sphere with a coordi-
Ground state geometries and theoretical calculations
The Sparkle/PM3 model was used to determine the ground state nation number of 8. The coordination polyhedron around the ions
geometries of the complexes.22 In this model the lanthanide ion is best described by a bicapped trigonal prism, as shown in Fig. 2b,
is replaced by a +3e point charge.23 The RHF wave functions were with O2 and O4 of the bidentate carboxylates as the capping atoms.
optimized using the Broyden–Fletcher–Goldfarb–Shanno (BFGS)
The Eu–O bond lengths (Table 2) are in the range 2.332(2)–
procedure with a convergence criterion of 0.15 kcal molꢁ1 Åꢁ1 2.545(2) Å, within that expected for this type of ligands,6,31,32
and the semi empirical PM3 with a convergence criterion of with the shortest bond to a bridging oxygen atom. While
10ꢁ6 kcal molꢁ1 for the SCF. In the Mopac2012 package24 the bridging carboxylates are often found in a bridging bidentate
following keywords were used: PM3, SPARKLE, XYZ, SCFCRT = mode, in which two oxygen atoms bind to one metal and one of
1D-10, GEO-OK, BFGS, CHARGE = X, PRECISE, GNORM = 0.15 and them bridges between two metals, the oxygen atom O5 binds
Table 1 Details of the X-ray crystallographic characterization of [Eu(4-iba)3(H2O)(dmf)]2
Formula
C48H42Eu2I6N4O16
T/K
100(2)
M/g molꢁ1
Crystal system
Space group
1968.15
Monoclinic
P21/c
Z
2
Dc/g cmꢁ3
2.258
5.411
m(Mo-Ka)/mmꢁ1
a/Å
b/Å
c/Å
a/1
8.3983(8)
25.310(3)
14.1228(16)
90
105.357(2)
90
Independent reflections, Rint [Fo Z 4s(Fo)]
Reflections collected
R1, wR2 [I 4 2s(I)]
Data/restrains/parameters
GooF
10 938; 0.0534
42 747
0.0331; 0.0492
10 938/0/344
1.010
b/1
g/1
Largest diff. peak and hole/e Åꢁ3
1.229–0.900
V/Å3
2894.8(5)
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem.

View Article Online
Paper
NJC
coordination modes for each ligand (Table S3) are shown in
the ESI.†
The triplet energy level (T) of the ligands was determined
using the two approaches described in the Experimental sec-
tion. The time resolved phosphorescence emission spectra of
analogous gadolinium(III) complexes (Fig. S7) and the deconvo-
lution of the phosphorescence bands (Fig. S8) are shown in the
ESI.† The energy values of the triplet energy level obtained by
the two approaches are shown in Table 3.
In order to explore the influence of the halogen at the para
position of the benzoic acid on the luminescent properties
of the europium(III) and terbium(III) complexes, the excitation
(Fig. 4(A) and 5(A)) and the emission (Fig. 4(B) and 5(B)) spectra
were obtained.
Fig. 2 (a) Plot of [Eu(4-iba)3(H2O)(dmf)]2 with thermal ellipsoids at 70%
probability. Hydrogen atoms were omitted for clarity. (b) Bicapped trigonal
prismatic coordination around the Eu(III).
The excitation spectra of the complexes are composed of
broad (assigned to the ligand transitions) and narrow bands
(assigned to the 4f–4f transitions). For the majority of the
complexes the intensities of the broad bands are higher than
that of the narrow ones that might indicate energy transfer
from the ligand to the lanthanide ion. The only exception of
this case is for the complexes containing the 4-iba ligand. The
absence of a broad band or a very low intensity one might be
indicative of inefficient energy transfer from the 4-iba ligand to
the lanthanide. The lower energy edge of the broad bands in
the excitation spectra of the europium(III) complexes (Fig. 4(A))
seems to go towards lower energies from the 4-fba to the 4-iba
ligands. That shift might be related to the presence of high
energy LMCT states localized between the excited singlet and
the triplet levels. The presence of LMCT states was investigated
by the arithmetic difference between the diffuse reflectance
spectra in the solid state of the europium(III) and gadolinium(III)
complexes as performed by Carlos and co-workers.34 The
europium(III) and gadolinium(III) diffuse reflectance spectra of
the complexes are shown in Fig. S9 (ESI†) and the resulting
arithmetic difference between them is shown in Fig. 6.
The maxima of the LMCT bands are shifted towards lower
energies from the 4-fba to 4-iba ligands and are in agreement with
Fig. 3 Packing diagram of [Eu(4-iba)3(H2O)(dmf)]2.
only to Eu1’, as its distance to Eu1 is 3.016(2) Å, which is longer
than would be expected for a bonding interaction.33
Two dimers are present in the unit cell (see packing diagram the redshift of the excitation bands edges. All LMCT bands have
in Fig. 3) with van der Waals interactions between the mole- higher energy than the triplet state and they may transfer energy to
cules and edge-to-face arrangement of the aromatic rings the lanthanide ions, once they appear in the excitation spectra, or
within the dimers and between adjacent molecules.
behave as intermediate states and transfer the energy to triplet
To date, no other X-ray quality single crystals were isolated levels.35 The LMCT states can also work as deactivation channels if
and the remaining precipitated complexes were characterized they have energy levels close or lower than the emission states as
in the powder form by TGA as well as FT-IR. The amount of reported by Carlos and co-workers.34 For terbium(III) compounds
non-coordinated (weight losses between 27 and 75 1C) and the presence of the LMCT states is more unlikely which can be
coordinated (weight losses between 75 and 200 1C) water observed by the narrowing of the excitation bands edges when
molecules were determined by TGA analysis (Fig. S1–S4 and compared with the europium(III) complexes (Fig. 5(A)) and also
Table S2, ESI†). The coordination modes of the carboxylate because of the lower Tb3+ - Tb2+ potential of reduction.35
group were determined by the calculation of Dn, where Dn =
The emission spectra of europium(III) complexes provide a
na(COOꢁ) ꢁ ns(COOꢁ). The FT-IR spectra (Fig. S6) and the lot of information about the point symmetry around the
Table 3 Energy of triplet levels of the ligands obtained by method (i) and (ii), described in the experimental section
Complexes
[Gd(4-fba)3(H2O)2]ꢀ1/2H2O
26 765
26 035
[Gd(4-cba)3(H2O)]
[Gd(4-bba)3(H2O)2]
[Gd(4-iba)3(H2O)2]
T
T
(i)/cmꢁ1
(ii)/cmꢁ1
24 704
23 299
25 485
23 811
21 789
20 736
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
NJC
Paper
Fig. 4 (A) Excitation spectra of the europium(III) complexes obtained in the solid state. (B) Emission spectra of the europium(III) complexes obtained in
the solid state. (a) [Eu(4-fba)3(H2O)2]. (b) [Eu(4-cba)3]ꢀ2H2O. (c) [Eu(4-bba)3]ꢀ5/2H2O. (d) [Eu(4-iba)3(H2O)2]. (e) [Eu(4-iba)3(H2O)(dmf)]2.
Fig. 5 (A) Excitation spectra of the terbium(III) complexes obtained in the solid state. (B) Emission spectra of the terbium(III) complexes obtained in the
solid state. (a) [Tb(4-fba)3(H2O)2]ꢀ1/2H2O. (b) [Tb(4-cba)3(H2O)2]. (c) [Tb(4-bba)3(H2O)]. (d) [Tb(4-iba)3(H2O)2].
europium(III) ion36 and also about the covalence between Eu–L complexes, Fig. 4B(a and d). Therefore, the chemical environ-
bonds.37–39 For all emission spectra of europium(III) complexes, ments around europium(III) ions in the [Eu(4-cba)3]ꢀ2H2O,
Fig. 4(B), the expected transitions 5D0 - 7FJ, ( J = 0; 1; 2; 3 and [Eu(4-bba)3]ꢀ5/2H2O and [Eu2(4-iba)6(H2O)2(dmf)2] complexes
4) are present. In addition, for all of them, the 5D0
transition intensity is higher than D0 - F1 ones, indicating
that the forced electric dipole and the dynamic coupling with the covalence degree of Eu–L bonds.37–39 In order to build
-
7F2 are less symmetric than that of the [Eu(4-fba)3(H2O)2] complex.
5
7
The energy of the 5D0 7F0 transition can be correlated
-
5
7
mechanisms are predominant in relation to magnetic dipole a comparative scale, high-resolution spectra for the D0 - F0
ones because the europium(III) ion is located in a low symmetry transition were obtained, Fig. 7. The centroid energy of this
site. The number of lines (splitting) of the europium(III) emis- transition was determined and the values are shown in Table 4.
sion spectrum may be correlated with the point symmetry.36
The values of the D0 - F0 energy are virtually the same.
5
7
The complexes [Eu(4-cba)3]ꢀ2H2O, [Eu(4-bba)3]ꢀ5/2H2O and That means that the changes of the halogen at the para position
[Eu2(4-iba)6(H2O)2(dmf)2] show more splittings at B77 K do not cause significant changes in the covalence between
(Fig. 4B(b, c and e)), for each 5D0
-
7FJ transition when Eu–O (carboxylate) bonds. It is important to point out that
compared with the [Eu(4-fba)3(H2O)2] and [Eu(4-iba)3(H2O)2] the 5D0
-
7F0 emission spectrum of the [Eu(4-iba)3(H2O)2]
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem.

View Article Online
Paper
NJC
Table 5 Judd–Ofelt intensity parameters (O2 and O4), radiative rate (Arad),
and quantum efficiency (Z) for europium(III) complexes
Complexes
O2/10ꢁ20 cm2 O4/10ꢁ20 cm2 Arad/sꢁ1 Z/%
[Eu(4-fba)3(H2O)2]
[Eu(4-cba)3]ꢀ2H2O
[Eu(4-bba)3]ꢀ5/2H2O
[Eu(4-iba)3(H2O)2]
10.4
10.0
11.5
8.9
9.0
5.4
7.6
8.4
5.5
496
461
506
441
532
23
46
52
19
24
[Eu(4-iba)3(H2O)(dmf)]2 13.4
not cause significant changes in the covalence degree of the
Eu–O bonds (Table 4).
In order to get a better insight about the point symmetry
around the europium(III) ion the Judd–Ofelt intensity para-
meters41,42 were obtained, Table 5. The equations used to obtain
these parameters are widely discussed in the literature.4,43,44
The JO intensity parameters O2 and O4 are strongly corre-
lated with the point symmetry around the europium(III) ion.
A high value for the O2 intensity parameter means a low
point symmetry around the europium(III) ion. For this series,
the complexes [Eu(4-cba)3]ꢀ2H2O, [Eu(4-bba)3]ꢀ5/2H2O and
[Eu(4-iba)3(H2O)(dmf)]2 show the highest values for the O2
intensity parameter that means the europium ion lies in a
lower symmetry site in these complexes when compared with
the [Eu(4-fba)3(H2O)2] and [Eu(4-iba)3(H2O)(dmf)]2 complexes.
These conclusions match well with the observations done for the
emission spectra where the [Eu(4-cba)3]ꢀ2H2O, [Eu(4-bba)3]ꢀ5/2H2O
and [Eu(4-iba)3(H2O)(dmf)]2 complexes show a similar emission
profile regarding the splitting of the 5D0 - 7FJ transitions.
The Horrock’s equation16 (eqn (1)) can be used for the
determination of coordinated water molecules when the O–H
oscillators are the only non-radiative route for de-activation of
the emitting level of the lanthanide(III). The values obtained by
the Horrock’s equation16 (qa) (Table 6) are very close to the ones
determined by TGA (qb) (Table S2, ESI†). The good agreement
between the values obtained (Table 6) confirms the absence of
additional de-activation routes, such as low energy LMCT states.
Fig. 6 Arithmetic difference between the europium(III) and gadolinium(III)
diffuse reflectance spectra showing the LMCT bands; (a) 4-fba. (b) 4-cba.
(c) 4-bba. (d) 4-iba.
5
5
Fig. 7 High-resolution spectra of the 5D0
-
7F0 transition of the
All emission decay curves of the D0 (europium(III)) and D4
(terbium(III)) emitting levels were fitted as a mono exponential
decay (Fig. S10, ESI†) and the values obtained for the emission
lifetime (t) are shown in Table 7. It is well known that for
europium(III) ions there is a strong correlation between the
lifetime and the non-radiative processes such as O–H and N–H
oscillators.2 As expected the highest lifetime values were obtained
for the [Eu(4-cba)3]ꢀ2H2O and [Eu(4-bba)3]ꢀ5/2H2O complexes due
to the absence of water in the first coordination sphere.
europium(III) complexes; black dot: [Eu(4-fba)3(H2O)2], red down triangle:
[Eu(4-cba)3]ꢀ2H2O, blue up triangle: [Eu(4-bba)3]ꢀ5/2H2O, purple square:
[Eu(4-iba)3(H2O)(dmf)]2 and light magenta rhomb: [Eu(4-iba)3(H2O)2].
Table 4 Energy of the 5D0 - 7F0 transition of the europium(III) complexes
Complexes
E00/cmꢁ1
[Eu(bza)3(H2O)]40
[Eu(4-fba)3(H2O)2]
[Eu(4-cba)3(H2O)2]ꢀ2H2O
[Eu(4-bba)3]ꢀ5/2H2O
[Eu(4-iba)3(H2O)2]
17 270
17 270
17 266
17 269
17 278
17 275
Table
6
Number of coordinated water molecules obtained by the
Horrocks equation16 (qa) and by TGA (qb)
[Eu(4-iba)3(H2O)(dmf)]2
Complexes
qa
qb
[Eu(4-fba)3(H2O)2]
[Eu(4-cba)3]ꢀ2H2O
[Eu(4-bba)3]ꢀ5/2H2O
[Eu(4-iba)3(H2O)2]
[Eu(4-iba)3(H2O)(dmf)]2
2.4
2
0
0
(powder) complex is broader than that of the
[Eu(4-iba)3(H2O)(dmf)]2 (single crystal), probably due to the
less homogeneity of the europium(III) sites in the powder
([Eu(4-iba)3(H2O)2]) when compared with the single crystal
([Eu(4-iba)3(H2O)(dmf)]2). However, this broadening or the
0.24
0.18
1.8
2.0
1.5
2.0a
a
Coordinated water molecules obtained from crystal data of homo-
insertion of the dmf ligand in the coordination sphere does bimetallic (Fig. 2a).
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
NJC
Paper
Table 7 Values of lifetimes (t) and absolute quantum yield obtained for the complexes (all the values of the absolute quantum yield were obtained with
5
excitation centered in 5D2
’
7F0, for europium(III), or L10
’
7F6, for terbium(III), transitions)
Complexes
FEEuu/%
t/ms
Complexes
FTTbb/%
t/ms
[Eu(4-fba)3(H2O)2]
[Eu(4-cba)3]ꢀ2H2O
[Eu(4-bba)3]ꢀ5/2H2O
[Eu(4-iba)3(H2O)2]
[Eu(4-iba)3(H2O)(dmf)]2
4 ꢂ 1
19 ꢂ 1
18 ꢂ 1
4 ꢂ 1
0.421 ꢂ 0.002
1.006 ꢂ 0.007
1.02 ꢂ 0.02
[Tb(4-fba)3(H2O)]ꢀ1/2(H2O)
[Tb(4-cba)3(H2O)2]
[Tb(4-bba)3(H2O)]
[Tb(4-iba)3(H2O)2]
—
4.5 ꢂ 0.2
14 ꢂ 1
2.4 ꢂ 0.3
o1
0.966 ꢂ 0.003
0.888 ꢂ 0.004
0.623 ꢂ 0.006
0.80 ꢂ 0.01
—
0.503 ꢂ 0.001
0.458 ꢂ 0.002
8 ꢂ 1
—
The emission lifetime values for the [Eu(4-fba)3(H2O)2],
[Eu(4-iba)3(H2O)(dmf)]2 and [Eu(4-iba)3(H2O)2] complexes are
similar to each other and lower than the values obtained for the
[Eu(4-cba)3]ꢀ2H2O and [Eu(4-bba)3]ꢀ5/2H2O complexes due to the
presence of water in the first coordination sphere. For terbium(III)
compounds there is no strong correlation between the lifetime and
the non-radiative processes due to the higher energy gap
between the fundamental and excited levels, when compared
with europium(III) ions. One can see, in Table 7, that the
terbium(III) complexes show nearly the same emission lifetime
values, the one for [Tb(4-bba)3(H2O)] being the smallest.
The absolute quantum yield values of the complexes (Table 7)
were obtained in the solid state with the excitation centered in the
7
5D2 ’ F0 (B396 nm) transition of the europium(III) ion or in the
Fig. 8 Strategy used to simulate the structure of the complexes.
5L10 ’ 7F6 (B368 nm) transition of the terbium(III) ion. In this case
the absolute quantum yields cannot be correlated with the ligand
properties because they were obtained with excitation centered in the
metal bands due to the instrumental limitations. Once the excitation
was centered in the 4f–4f transitions one should expect a quantum
yield strongly correlated with the non-radiative processes, such as
vibronic coupling with O–H oscillators or back energy transfer from
europium(III) or terbium(III) to ligand levels. The highest values
were obtained for the [Eu(4-cba)3]ꢀ2H2O and [Eu(4-bba)3]ꢀ5/2H2O
complexes probably because of the absence of coordinated water
molecules, once the energy value of the triplet levels of the ligands
are similar (Table 3) and the back energy transfer rates do not play an
important role in these series of complexes (Table 9). As expected,
the presence of coordinated water molecules plays an important role
in the quantum yield in this series.
Fig. 9 Division of the oxygen atoms into groups for calculation of the
polarizability and charge factors.
To get a better insight about the energy transfer rates, the
ground state geometries of the europium(III) complexes were
determined using the Sparkle/PM322 model implemented in match the experimental JO intensity parameter values. The
the MOPAC2012 software24 and are shown in Fig. S11 (ESI†). oxygen atoms were divided into 4 groups: (i) oxygens that bond
In order to keep rigid the central polyhedron, the strategy to exclusively by the bidentate bridging mode (group 1); (ii) oxygens
calculate the ground state geometries was based on the that simultaneously bond by the bidentate bridging and chelate
methodology used by Freire and co-workers45 and Sigoli and mode (group 2A and 2B); (iii) oxygens that bond exclusively by
co-workers40 for dimers or polymers: three metallic centers the bidentate chelate mode (group 3) and (iv) oxygens from water
were drawn, the central metal is coordinated by different molecules (group 4). The values obtained for polarizability (a)
modes: bidentate bridge and/or bidenate chelate and/or and charge factor (g) are shown in the ESI† (Table S3).
bidentate bridging and chelate simultaneously, the two other
metallic centers have only the function to keep the structure are shown in Table 8 and compared with the experimental ones.
rigid and the coordination sphere was filled with water There is good agreement between the experimental and
The values obtained for the theoretical JO intensity parameters
molecules, Fig. 8. The oxygen atoms, directly bonded to the calculated JO intensity parameters except for the case of
europium(III) ion, were also divided into groups in order to the O4 intensity parameter in the [Eu(4-bba)3]ꢀ5/2H2O complex.
obtain the polarizability (a) and charge factor (g) values, Fig. 9. The poor agreement between the calculated and experimental
The theoretical JO intensity parameters were obtained by JO intensity parameters might be due to the approximated
adjusting the polarizability (a) and charge factor (g) in order to propositions of the coordination polyhedron of carboxylate
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem.

View Article Online
Paper
NJC
Table 8 Calculated and experimental JO intensity parameters. In parentheses the data obtained using single crystals
Complexes
(O2)theo/10ꢁ20 cm2
(O2)exp/10ꢁ20 cm2
(O4)theo/10ꢁ20 cm2
(O4)exp/10ꢁ20 cm2
(O6)theo/10ꢁ20 cm2
[Eu(4-fba)3(H2O)2]
[Eu(4-cba)3]ꢀ2(H2O)
[Eu(4-bba)3]ꢀ5/2H2O
[Eu(4-iba)3(H2O)2]
[Eu(4-iba)3(H2O)(dmf)]2
10.4
11.7
12.1
8.9
10.4
11.0
11.5
8.9
9.0
4.0
1.4
8.4
8.4
5.4
7.6
8.4
5.5
0.26
0.42
0.53
0.26
13.4 (13.4)
13.4
5.5 (5.5)
0.13 (0.17)
Table 9 Energy transfer and back transfer rates between the ligand triplet level and 5D1 and 5D0 europium(III) levels, calculated from the ground state
geometries provided by Sparkle/PM3
Complexes
T - 5D1
5D1 - T
T - 5D0
5D0 - T
RL/Å
[Eu(4-fba)3(H2O)2]
[Eu(4-cba)3]ꢀ2(H2O)
[Eu(4-bba)3]ꢀ5/2H2O
1.5 ꢃ 104
1.4 ꢃ 106
2.1 ꢃ 105
1.8 ꢃ 10ꢁ10
6.3 ꢃ 10ꢁ3
2.6 ꢃ 10ꢁ8
5.0 ꢃ 103
8.6 ꢃ 105
7.9 ꢃ 104
1.2 ꢃ 10ꢁ14
7.9 ꢃ 10ꢁ7
2.0 ꢃ 10ꢁ12
6.2057
5.6366
5.7297
Table 10 Theoretical and experimental quantum efficiencies (Z) and that there is a similar symmetry occupied by the europium(III)
quantum yields (F)
ion in the complexes [Eu(4-cba)3]ꢀ2H2O, [Eu(4-bba)3]ꢀ5/2H2O and
[Eu(4-iba)3(H2O)(dmf)]2. A comparative covalence degree of the
Complexes
Ztheo/%
Zexp/%
Ftheo/%
Eu–O bonds was obtained by the energy of the 5D0 - 7F0 transition.
In this series the different halogens (F, Cl, Br and I) at the para
position and the solvent molecule do not change the degree of
covalence for the Eu–O bonds significantly. The presence of LMCT
states was confirmed by the arithmetic subtraction between the
diffuse reflectance spectra of europium(III) and gadolinium(III) and
the energy of the LMCT state decreases from the 4-fba to the 4-iba
ligand. The highest emission lifetimes and quantum yields were
obtained for the [Eu(4-cba)3]ꢀ2H2O and [Eu(4-bba)3]ꢀ5/2H2O com-
plexes due to the absence of coordinated water molecules.
Therefore, the halogen change at the para position of the
aromatic ring of the benzoic acid does change both molecular
structure and photophysical properties. The experimental observa-
tions done in this work will help the design of new luminescent
compounds with desirable properties and also as data for under-
standing the influence of the modifications in the ligand structure
and the changes in the photophysical properties.
[Eu(4-fba)3(H2O)2]
[Eu(4-cba)3]ꢀ2(H2O)
[Eu(4-bba)3]ꢀ5/2H2O
21
35
44
23
46
52
1
24
10
groups coordinated. Also, in this case intermolecular interac-
tions might play an important role for the structure.
The energy transfer rates (Table 9) were calculated for
all complexes except for the [Eu2(4-iba)6(H2O)2] and the
[Eu(4-iba)3(H2O)(dmf)]2 complexes due to the absence of parameters
for the iodine atom in the ORCA software.25 For all complexes the
energy transfer, from the ligand to the lanthanide, is higher than the
back transfer ones. The highest energy transfer rates, from the ligand
to the lanthanide, were obtained for the [Eu(4-cba)3]ꢀ2H2O complex
which might be due to the smaller value of the RL parameter. The
values obtained for the transfer rates are smaller than the ones
obtained for other systems like tris b-diketonates.46
The quantum efficiency and quantum yield values were
obtained using adequate kinetics equations27–29 and are shown
in Table 10. The values obtained using the Sparkle/PM322 ground
state geometries are in good agreement with the experimental ones.
As expected the anhydrous complex shows higher values for,
both, theoretical quantum efficiency and quantum yield. The
high quantum yield of the [Eu(4-cba)3]ꢀ2H2O complex is due to
the combination of absence of coordinated solvent molecules
and also the low value of the back transfer energy from the 5D1,0
to the ligand triplet levels, Table 10.
Acknowledgements
JHSKM is indebted to CNPq for a PhD fellowship. AdBD
acknowledges NSF (grant CHE 1363325) for financial support. FAS
and IOM are indebted to CNPq, CAPES and FAPESP (Grant no. 2008/
53868-0) for financial support. The authors would like to thank the
Laboratory of Advanced Optical Spectroscopy (LMEOA/IQ-UNICAMP/
FAPESP Grant no. 2009/54066-7). This work is a contribution of the
National Institute of Science and Technology in Complex Functional
Materials (CNPq-MCT/FAPESP).
Conclusions
References
The rich coordination modes of the ligands containing the
carboxylate group provide a very interesting system for photophysical
studies. The influence of the halogen on the photophysical
properties was verified by photoluminescence spectroscopy. The
point symmetries of the complexes were compared using the split-
tings of each 5D0 - 7FJ and the JO intensity parameters. It was found
1 K. Binnemans, Chem. Rev., 2007, 107, 2592–2614.
2 J.-C. G. Bu¨nzli and S. V. Eliseeva, in Lanthanide Luminescence:
Photophysical, Analytical and Biological Aspects, ed. P. Hanninen
¨
¨
¨
and H. Harma, Springer, Berlin, 2011, ch. 1, pp. 1–46.
3 J.-C. G. Bu¨nzli, Chem. Rev., 2010, 110, 2729–2755.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
NJC
Paper
´
´
4 G. F. de Sa, O. L. Malta, C. de Mello Donega, A. M. Simas, 25 F. Neese, Wiley Interdiscip. Rev.: Comput. Mol. Sci., 2012, 2,
R. L. Longo, P. A. Santa-Cruz and E. F. da Silva Jr, Coord.
Chem. Rev., 2000, 196, 165–195.
5 G. B. Deacon and R. J. Phillips, Coord. Chem. Rev., 1980, 33,
227–250.
73–78.
26 J. E. Ridley and M. C. Zerner, Theor. Chim. Acta, 1976, 42,
223–236.
27 O. L. Malta, J. Lumin., 1997, 71, 229–236.
6 X. Li, Z.-Y. Zhang and Y.-Q. Zou, Eur. J. Inorg. Chem., 2005, 28 F. R. G. Silva and O. L. Malta, J. Alloys Compd., 1997, 250,
2909–2918. 427–430.
7 L.-J. Xu, S.-P. Wang, R.-F. Wang and J.-J. Zhang, J. Coord. 29 O. L. Malta, J. Non-Cryst. Solids, 2008, 354, 4770–4776.
Chem., 2008, 61, 237–250.
8 B. S. Zhang, Acta Crystallogr., Sect. E: Struct. Rep. Online,
2006, 62, M2645–M2647.
9 X. Li, T.-T. Zhang, Y.-L. Ju, C.-Y. Wang, Y.-Q. Li, L. Zhang
and Q. Zhang, J. Coord. Chem., 2007, 60, 2121–2132.
10 H.-Y. Zhang, J.-J. Zhang, N. Ren, S.-L. Xu, L. Tian and J.-H.
Bai, J. Alloys Compd., 2008, 464, 277–281.
11 X. Li, Y.-L. Ju and Y.-Q. Zou, J. Coord. Chem., 2007, 60,
1513–1526.
30 J. D. Dutra, T. D. Bispo and R. O. Freire, J. Comput. Chem.,
2014, 35, 772–775.
31 F. Cagnin, M. R. Davolos and E. E. Castellano, Polyhedron,
2014, 67, 65–72.
32 E. R. Souza, I. O. Mazali and F. A. Sigoli, J. Fluoresc., 2014,
24, 203–211.
33 S. Viswanathan and A. de Bettencourt-Dias, Inorg. Chem.,
2006, 45, 10138–10146.
34 J. A. Fernandes, R. A. S. Ferreira, M. Pillinger, L. D. Carlos,
I. S. Goncalves and P. J. A. Ribeiro-Claro, Eur. J. Inorg. Chem.,
2004, 3913–3919, DOI: 10.1002/ejic.200400191.
12 A. P. Souza, L. C. V. Rodrigues, H. F. Brito, S. Alves Jr and
O. L. Malta, J. Lumin., 2010, 130, 181–189.
13 S. K. Lower and M. A. El-Sayed, Chem. Rev., 1966, 66, 35 J.-C. G. Bu¨nzli, Coord. Chem. Rev., DOI: 10.1016/j.ccr.2014.10.013.
199–241.
36 P. A. Tanner, in Lanthanide Luminescence: Photophysical,
¨
Analytical and Biological Aspects, ed. P. Hanninen and
14 M. Hilder, P. C. Junk, U. H. Kynast and M. M. Lezhnina,
J. Photochem. Photobiol., A, 2009, 202, 10–20.
¨
¨
H. Harma, Springer, Berlin, 2011, ch. 7, pp. 183–233.
15 G. A. Crosby, R. E. Whan and R. M. Alire, J. Chem. Phys., 37 S. T. Frey and W. D. Horrocks, Inorg. Chim. Acta, 1995, 229,
1961, 34, 743. 383–390.
16 R. M. Supkowski and W. D. Horrocks, Inorg. Chim. Acta, 38 L. D. Carlos, O. L. Malta and R. Q. Albuquerque, Chem. Phys.
2002, 340, 44–48. Lett., 2005, 415, 238–242.
17 L. D. Carlos, R. A. Ferreira, Z. Bermudez Vde and S. J. 39 O. L. Malta, H. J. Batista and L. D. Carlos, Chem. Phys., 2002,
Ribeiro, Adv. Mater., 2009, 21, 509–534. 282, 21–30.
18 SMART: Bruker Molecular Analysis Research Tool, Bruker, 40 J. H. S. K. Monteiro, A. L. B. Formiga and F. A. Sigoli,
v.5.626 edn, 2002. J. Lumin., 2014, 154, 22–31.
19 SAINTPlus: Data Reduction and Correction Program, v.6.36a 41 B. R. Judd, Phys. Rev., 1962, 127, 750–761.
edn, 2001. 42 G. S. Ofelt, J. Chem. Phys., 1962, 37, 511–520.
20 SADABS: an Empirical Absorption Correction Program, v. 2.01 43 O. L. Malta, Chem. Phys. Lett., 1982, 87, 27–29.
edn, 2001.
44 O. L. Malta, M. A. C. dos Santos, L. C. Thompson and
21 G. M. Sheldrick, v.6.10 edn, 2001.
N. K. Ito, J. Lumin., 1996, 69, 77–84.
22 R. O. Freire, G. B. Rocha and A. M. Simas, J. Braz. Chem. Soc., 45 M. O. Rodrigues, N. B. da Costa Jr., C. A. de Simone,
2009, 20, 1638–1645.
23 A. V. M. Andrade, N. B. Costa, A. M. Simas and G. F. de Sa,
A. A. S. Araujo, A. M. Brito-Silva, F. A. A. Paz, M. E. de
Mesquita, S. A. Junior and R. O. Freire, J. Phys. Chem. B,
2008, 112, 4204–4212.
´
Chem. Phys. Lett., 1994, 227, 349–353.
24 J. J. P. Stewart, Stewart Computational Chemistry, Colorado 46 J. H. S. K. Monteiro, R. D. Adati, M. R. Davolos, J. R. M. Vicenti
Springs, 2012. and R. A. Burrow, New J. Chem., 2011, 35, 1234.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem.