Spectroscopic studies of europium(III) and gadolinium(III) tris-β-diketonate complexes with diazabutadiene ligands

Article abstract of DOI:10.1002/ejic.200400191

Adducts of the type Ln(NTA)₃·L [Ln = Eu, Gd; NTA = 1-(2-naphthoyl)-3,3,3-trifluoroacetonate; L = 1,4-diaza-1,3-butadiene RN=CHCH=NR (R = p-tolyl, o-tolyl)] were prepared by treatment of Ln(NTA) ₃·2H₂O with one equivalent of the chelating N-N ligand. The mixed ligand complexes were characterised by elemental analysis, thermogravimetric analysis, IR and Raman spectroscopy, and photoluminescence (PL) spectroscopy. Ab initio calculations were carried out in order to predict coordination geometries and also to interpret the vibrational spectra. For the europium(III) compounds, new Raman-active bands at approximately 440 and 479 cm^-1 are tentatively assigned to (vEu-N)_sym and (vEu-N)_as modes, respectively. A second band at 1185 cm^-1 is associated with the symmetric stretching of the N-C(ring) and C-C (DAB) bonds, and is shifted about 20 cm^-1 to a higher wavenumber relative to the corresponding band for the free ligand. The room-temperature PL spectra of the two Eu complexes are composed of the typical Eu³⁺ red emission, assigned to transitions between the first excited state ( ⁵D₀) and the ground multiplet (⁷F _0-4). Based on the room-temperature emission spectra and lifetime measurements, the quantum efficiencies of the ⁵D₀ Eu ₃₊ excited state were estimated and found to be quite low (e.g. 2-3%). The low efficiencies are attributed to the presence of thermally activated nonradiative channels involving ligand-to-metal charge-transfer (LMCT) states. Indeed, a low-lying LMCT band was detected for the Eu complex containing the ligand p-tolyl-DAB by comparison of the room-temperature diffuse reflectance spectrum for this complex with that of the corresponding Gd analogue. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2004.

Full text of DOI:10.1002/ejic.200400191

FULL PAPER
Spectroscopic Studies of Europium(III) and Gadolinium(III) Tris-β-diketonate
Complexes with Diazabutadiene Ligands
[a]
[b]
[a]
[b]
´
´
´
Jose A. Fernandes, Rute A. Sa Ferreira, Martyn Pillinger, Luıs D. Carlos,*
Isabel S. Goncalves,[a] and Paulo J. A. Ribeiro-Claro*[a]
¸
Keywords: Ab initio calculations / Lanthanides / Luminescence / N ligands / Vibrational spectroscopy
Adducts of the type Ln(NTA)3·L [Ln = Eu, Gd; NTA = 1-(2-
naphthoyl)-3,3,3-trifluoroacetonate; L = 1,4-diaza-1,3-butadi-
ene RN=CHCH=NR (R = p-tolyl, o-tolyl)] were prepared by
treatment of Ln(NTA)3·2H2O with one equivalent of the che-
perature PL spectra of the two Eu complexes are composed
of the typical Eu3+ red emission, assigned to transitions be-
tween the first excited state (5D0) and the ground multiplet
(7F0−4). Based on the room-temperature emission spectra and
lating N−N ligand. The mixed ligand complexes were char- lifetime measurements, the quantum efficiencies of the 5D0
acterised by elemental analysis, thermogravimetric analysis,
IR and Raman spectroscopy, and photoluminescence (PL)
spectroscopy. Ab initio calculations were carried out in order
to predict coordination geometries and also to interpret the
vibrational spectra. For the europium(III) compounds, new
Raman-active bands at approximately 440 and 479 cm−1 are
tentatively assigned to (νEu−N)sym and (νEu−N)as modes, re-
spectively. A second band at 1185 cm−1 is associated with the
symmetric stretching of the N−C(ring) and C−C (DAB) bonds,
and is shifted about 20 cm−1 to a higher wavenumber relative
Eu3+ excited state were estimated and found to be quite low
(e.g. 2−3%). The low efficiencies are attributed to the pres-
ence of thermally activated nonradiative channels involving
ligand-to-metal charge-transfer (LMCT) states. Indeed, a
low-lying LMCT band was detected for the Eu complex con-
taining the ligand p-tolyl-DAB by comparison of the room-
temperature diffuse reflectance spectrum for this complex
with that of the corresponding Gd analogue.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
to the corresponding band for the free ligand. The room-tem- Germany, 2004)
Introduction
the solvent molecules in these complexes by organic ligands
is one way to further modify the luminescence properties of
the parent tris complex. The complex with 2,2Ј-bipyridine
(bipy) has already been reported.[9] In the present work,
Eu(NTA)3 complexes containing 1,4-diaza-1,3-butadienes
(DAB) of the type RNϭCRЈCRЈϭNR (R ϭ p-tolyl, o-tolyl;
RЈ ϭ H) have been prepared and characterised. For com-
parison, the corresponding GdIII complexes have also been
prepared. To the best of our knowledge, very few studies
have been carried out on diazabutadiene compounds of rare
earths,[20Ϫ22] especially with regard to their photophysical
properties. This is surprising since, as α-diimine ligands,
DAB ligands have electronic and structural similarities to
the more extensively studied heteroaromatic chelating
agents bipy and 1,10-phenanthroline.[23] Added flexibility
arises from the presence of the substituents R and RЈ, per-
haps allowing luminescence behaviour to be tuned by varia-
tion of steric and electronic effects.
Over the past 40 years there has been considerable inter-
est in the luminescence and lasing ability of certain diketon-
ates of lanthanides, particularly europium() and
terbium().[1Ϫ4] Two classes of complexes may be dis-
tinguished depending on whether the ratio of diketonate ion
to metal ion in the product is 3:1 (tris complex) or 4:1
(tetrakis complex).[5,6] The tris-β-diketonates are usually co-
ordinatively unsaturated and, therefore, form adducts with
a variety of Lewis base ligands, to give seven-, eight- or
even nine-coordinate complexes.[7Ϫ19] Some of us recently
reported on the luminescence features and the absolute
quantum yield of europium() tris[1-(2-naphthoyl)-3,3,3-
trifluoroacetonate] complexes, Eu(NTA)3·2L (NTA
ϭ
1-(2-naphthoyl)-3,3,3-trifluoroacetonate, L ϭ H2O and
DMSO).[19] The experimental quantum yield measured for
Eu(NTA)3·2DMSO (0.75) is one of the highest so far re-
ported for solid-state europium complexes. Replacement of
Results and Discussion
[a]
Department of Chemistry, CICECO, University of Aveiro,
3810-193 Aveiro, Portugal
Fax: (internat.) ϩ 351-234-370084
E-mail: pclaro@dq.ua.pt
Synthesis and Thermogravimetric Analysis
The mononuclear complexes Ln(NTA)3·p-tolylϪDAB
[Ln ϭ Eu (1), Gd (3)] and Ln(NTA)3·o-tolylϪDAB [Ln ϭ
[b]
Department of Physics, CICECO, University of Aveiro,
3810-193 Aveiro, Portugal
Eur. J. Inorg. Chem. 2004, 3913Ϫ3919
DOI: 10.1002/ejic.200400191
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3913

L. D. Carlos, P. J. A. Ribeiro-Claro et al.
FULL PAPER
Eu (2), Gd (4)] were prepared in reasonable yields by the
treatment of solutions of Ln(NTA)3·2H2O in chloroform
with one equivalent of the respective diazabutadiene ligand.
Figure 1 shows the TGA results for o-tolylϪDAB,
Eu(NTA)3·2H2O and Eu(NTA)3·o-tolylϪDAB (2). Similar
traces were obtained for the compounds p-tolylϪDAB,
Gd(NTA)3·2H2O and Ln(NTA)3·L (1,3,4), respectively. The
compounds o-tolylϪDAB and p-tolylϪDAB sublime in the
temperature range 100Ϫ230 °C. The precursors
Ln(NTA)3·2H2O undergo a mass loss up to 100 °C, which
corresponds to the removal of coordinated water. Further
decomposition takes place in two steps; the first of these
occurs between 100 and 350 °C (54% mass loss for Eu, 63%
for Gd), and the second between 350 and 600 °C (21% mass
loss for Eu, 19% for Gd). Finally, a small mass loss of 2%
occurs between 600 and 700 °C for both compounds, leav-
ing a residual mass of 17.7% for Eu and 13.8% for Gd. The
TGA curves of the adducts 1Ϫ4 are very similar to each
other and to the reagents Ln(NTA)3·2H2O, and all steps for
these six compounds take place at the same temperatures.
There is no step characteristic of ‘‘free’’ ligand, indicating
that the ligands are coordinated to the metal centres. The
first mass loss (3.0% for 1) corresponds to the removal of
water content. Following this, the mass loss for the step
between 100 and 350 °C is smaller than that observed for
the reagents (47.2% for compound 1). Conversely, the step
between 350 and 600 °C corresponds to a larger amount
(34.3% for compound 1). Finally, the TGA shows a mass
loss of about 2% between 575 and 720 °C, leaving a residual
mass of about 16% for compounds 1 and 2, and 14% for
compounds 3 and 4.
Figure 1. TGA of Eu(NTA)3·o-tolylϪDAB (2) (Ϫ), o-tolylϪDAB
(···) and Eu(NTA)3·2H2O (Ϫ Ϫ Ϫ)
Figure 2. Lowest energy structure calculated for the model
Eu[O(CH)3O]3·p-tolylϪDAB; the (NTA)3 framework was
obtained from the X-ray single crystal structure for
Eu(NTA)3·phenanthroline[24]
(2), with the exception of the twist angles of the phenyl
ring, which are about 10° larger than those for 1 due to
the presence of the ortho methyl groups. Calculated EuϪN
distances are about 233 pm, and the NϪEuϪN angle is 72°
in both 1 and 2. The EuϪN bond lengths are shorter than
those calculated at the same level for the 2,2Ј-bipyrimidine
Ab Initio Calculations and Vibrational Spectra
and 1,10-phenanthroline derivatives,[24] suggesting
a
Figure 2 shows the calculated coordination geometry of stronger EuϪN interaction in the case of the present li-
DAB in 1 (NTA ligands are shown in wireframe). The com- gands. As a result, the calculated frequencies for the EuϪN
plexed tolylϪDAB adopts a propeller-like configuration; stretching modes are higher than those usually reported. In
the phenyl rings are rotated 33° and 42° from the Nϭ fact, ab initio calculations predict the νEuϪN stretching
CϪCϭN plane, in opposite directions. In the free ligand, modes at about 460 cmϪ1 (after scaling), with a moderately
the lowest energy form presents a similar rotation angle (ap- strong Raman intensity and weak IR intensity.
proximately 37°), but to the same side of the NϭCϪCϭ
An important geometrical change upon complexation is
N plane (therefore, the phenyl rings are co-planar). Similar revealed in the bond lengths of the diazabutadiene frag-
results were obtained for the simulation of o-tolylϪDAB ment. From the free DAB ligand to complex 1, the
3914
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 3913Ϫ3919

Spectroscopic Studies of Europium(III) and Gadolinium(III) Tris-β-diketonate Complexes
FULL PAPER
C
(ring)ϪN, CϭN and CϪC bond lengths are reduced by the free ligands and products due to their strong Raman
approximately 2.2, 1.8 and 3.3 pm, respectively. From the intensity. As can be seen, the strong band at 1165 cmϪ1 for
free ligand to complex 2, the corresponding values are 1.7, the pure DAB ligand [trace (a)] moves to about 1185 cmϪ1
1.8 and 2.9 pm, respectively. The effect is residual in the in 1 [trace (c)]. The ca. 20 cmϪ1 upwards shift upon com-
aromatic CϪC bonds (Ͻ 1 pm) and is absent in the CϪCH3 plexation is in agreement with the contraction of the CϪC
bonds. This contraction affects the vibrational modes in- and NϪC bonds from the free ligand to the complexes.
volving the diazabutadiene fragment, which are predicted
Photoluminescence Studies
to be generally shifted to higher wavenumbers upon com-
plexation. The effect is particularly evident in the mode
combining the symmetric stretching of the NϪC(ring) and
CϪC (DAB) bonds, with a strong Raman intensity, for
which a shift of 20 cmϪ1 is predicted. The calculated scaled
values are 1164 cmϪ1 for the free DAB ligands and 1184
cmϪ1 for the complex model. These predictions find exper-
imental support from the vibrational spectra, as discussed
below.
The vibrational spectra of 1 and 2 are well-described by
the sum of the spectra of their reagents, with some ad-
ditional details ascribed to the complexation process. Fig-
ure 3 shows two relevant regions in the Raman spectra of
compound 1 [trace (c)], the reagents Eu(NTA)3·2H2O [trace
(b)] and p-tolylϪDAB [trace (a)].
Figure 4 shows the room-temperature PL spectra of com-
pounds 1 and 2 obtained under the excitation wavelength
that maximises the emission intensity. Both spectra are
composed of the typical Eu3ϩ red emission, assigned to
transitions between the first excited state (5D0) and the
ground multiplet (7F0Ϫ4). It is possible to identify 3 and 5
Stark components for the D0 Ǟ 7F1Ϫ2 transitions, which
5
indicates that the Eu3ϩ ions are located in a low site sym-
metry group, without an inversion centre according to the
higher intensity of the 5D0 Ǟ 7F2 transition. Changes in the
excitation wavelength did not produce significant alter-
ations either in the number of emission components or in
their energy. The presence of only one component for the
5D0 Ǟ 7F0 transition suggests only one Eu3ϩ local environ-
ment. However, the high value found for the full width at
half maximum (fwhm) in both compounds( 26 cmϪ1) indi-
cates a large distribution of local coordination sites,[25]
probably induced by the amorphous nature of the com-
pounds. The similarity between the two spectra is also ap-
parent in the energy and fwhm of the D0 Ǟ 7F0Ϫ2 tran-
5
sitions. This fact, in particular the comparable energies for
the 5D0 Ǟ 7F0 transition, indicates that the Eu3ϩ first coor-
dination sphere might be very similar for the two com-
pounds. This is based on the knowledge that the energy of
this transition is usually related to the so-called nephelaux-
etic effect induced by the ionЈs first neighbours.[26] Such
conclusions are also supported by the PL spectra measured
at 14 K. Apart from small changes in the relative intensity
5
of the D0 Ǟ 7F2 transition Stark components, the energy,
fwhm and number of emission components of the intra-
4f6 transitions are essentially the same as observed at room
Figure 3. Raman spectra of (a) p-tolylϪDAB, (b) Eu(NTA)3·2H2O
and (c) Eu(NTA)3·p-tolylϪDAB (1); the arrows point to the bands
assigned to the νEuϪN modes (about 460 cmϪ1) and symmetric
stretching of the NϪC(ring) and CϪC (DAB) bonds (1185 cmϪ1
)
The vibrational modes involving the EuϪN bonds were
searched for in the low wavenumber region. The two bands
observed at approximately 440 and 479 cmϪ1 are not pre-
sent in the spectra of the reagents and fit the intensity and
wavenumber predicted from ab initio calculations. In this
way, they have been tentatively assigned to (νEuϪN)sym and
(νEuϪN)as, respectively. Identical bands are observed at
449 cmϪ1 and 473 cmϪ1 in the Raman spectrum of 2, sup-
porting this assignment. Analysis of the IR spectrum in this
region was inconclusive, due to band overlap.
The effect of complexation can also be observed in the
higher wavenumber region of Figure 3. The bands corre-
sponding to the symmetric stretching of the NϪC(ring) and
Figure 4. Room-temperature PL spectra of compounds 1 and 2
excited at 345 nm; The dotted line corresponds to the PL spectrum
CϪC (DAB) bonds are easily identified in the spectra of of 1 detected at 14 K and excited at 365 nm
Eur. J. Inorg. Chem. 2004, 3913Ϫ3919
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3915

L. D. Carlos, P. J. A. Ribeiro-Claro et al.
FULL PAPER
temperature. However, the integrated intensity of the emis- ical experimental conditions is different from those plotted
sion arising from the 5D0 level strongly depends on the in Figure 6.[19]
5
selected temperature, presenting the behaviour shown in
The D0 lifetime, τ, was measured around the most in-
Figure 5 for 1. For both compounds, no emission from the tense PL line (not shown). For both compounds the data
ligands could be detected, suggesting an effective interac- are well described by a single exponential function, re-
tion between them and the Eu3ϩ ions.
vealing lifetime values of 0.038 Ϯ 0.001 and 0.030 Ϯ
0.001 ms for compounds 1 and 2, respectively. These values
are smaller than the lifetime value previously found for the
precursor complex.[19]
Based on the room-temperature emission spectra and on
the lifetime measurements we can estimate the efficiency, q,
5
of the D0 Eu3ϩ excited state, considering that only non-
radiative and radiative processes are essentially involved in
5
the depopulation of the D0 state,[3,19,27,28] where kr and knr
are the radiative and nonradiative transition probabilities,
respectively, (kexp ϭ 1/τexp ϭ kr ϩ knr).
(1)
The emission intensity, I, taken as the integrated inten-
5
7
Figure 5. Temperature-dependence (14Ϫ300 K) of the Eu3ϩ 5D0
emission for compound 1 (open circles); the dotted line is the best
fit to the data using Equation (4)
sity, S, of the emission curves, for the D0 Ǟ F0Ϫ4 tran-
sitions, is expressed by Equation (2): where i and j represent
the initial (5D0) and final (7F0Ϫ6) levels, respectively, h wiǞj
-
is the transition energy, AiǞj corresponds to Einstein’s coef-
ficient of spontaneous emission and Ni is the population of
The excitation spectra (PLE), monitored around the most
intense line of the cation (Figure 6), enable us to draw the
same conclusion. The spectra exhibit a large broad band
between 280 and 420 nm, with a maximum at around
345 nm, and no lines associated with the direct excitation
of the intra-4f6 levels are detected. This strongly suggests
an efficient sensitization process between the ligands and
the metal ion. Lowering the temperature to 14 K induces a
red-shift to 365 nm, and a low-intensity line associated with
the excitation of the 5D2 level is observed, as shown in Fig-
ure 6 for 1. This large band may originate from the ligand
levels involving the NTA and diimine components, since the
PLE spectrum acquired for Eu(NTA)3·2H2O under ident-
5
the D0 emitting level.[3,19,27,28]
(2)
The radiative contribution may be calculated from the
relative intensities of the 5D0 Ǟ 7F0Ϫ6 transitions. Since the
7
5D0 Ǟ F1 transition can be considered as a reference due
to its dipolar magnetic nature, kr can be calculated accord-
ing to Equation (3), where A0Ǟ1 is the Einstein’s coefficient
of spontaneous emission between the 5D0 and the 7F1
Stark levels.
(3)
5
7
The D0 Ǟ F5,6 transitions were not measured exper-
imentally and so the branching ratio for these transitions,
which is usually very small, can be neglected. Therefore, we
5
can ignore their influence in the depopulation of the D0
5
excited state. The D0 Ǟ 7F1 transition does not depend on
the local ligand field seen by Eu3ϩ ions and, thus, may be
used as a reference for the whole spectrum; in vacuo
A(5D0 Ǟ 7F1) ϭ 14.65 sϪ1 [28]
The values found for q, kr
.
and knr for compounds 1 and 2 are gathered in Table 1. An
average index of refraction equal to 1.5 was considered for
Figure 6. room-temperature PLE spectra of compounds 1 and 2
monitored around 612 nm; the dotted line is the PLE spectrum of
1 detected at 14 K
both
compounds
leading
to
an
approximate
A(5D0 Ǟ 7F1) value of 50 sϪ1 [29]
.
3916
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 3913Ϫ3919

Spectroscopic Studies of Europium(III) and Gadolinium(III) Tris-β-diketonate Complexes
FULL PAPER
17800 cmϪ1). As far as we know, this is one of the clearest
examples where the arithmetic difference between the dif-
fuse reflectance spectra of isostructural Eu3ϩ and Gd3ϩ
compounds allows the explicit identification of ligand-to-
metal charge transfer states by direct means. Unfortunately,
and in agreement with the mentioned difficulties, these
states cannot be identified so visibly in the reflectance spec-
tra of compounds 2 and 4.
Table 1. Experimental 5D0 lifetimes (τ) and decay rates (kexp), calcu-
lated radiative (kr) and nonradiative (knr) 5D0 decay rates and quan-
tum efficiency (q) for compounds 1 and 2
τ (ms)
kexp (msϪ1
)
kr (msϪ1
)
knr (msϪ1
)
q (%)
1
2
0.038 Ϯ 0.001 26.32
0.030 Ϯ 0.001 33.33
0.82
0.72
25.50
32.61
3.1
2.2
Compound 1 has the highest q value, induced by higher
kr (12%) and lower knr (22%) values. However, both com-
pounds present lower values for the 5D0 quantum efficiency
compared with that found for the precursor complex
Eu(NTA)3·2H2O (q ϭ 29%)[19] that may be induced by the
higher knr contribution found in 1 and 2. In fact, the tem-
perature-dependence of the Eu3ϩ emission-integrated inten-
sity, Iint (Figure 5), suggests a thermally activated nonradi-
ative mechanism. Such nonradiative de-excitation is well-
described by the following expression:
(4)
where A is related to the ratio between the nonradiative
(at T ϭ 0) and radiative rates, KB is the Boltzmann constant
and Ea is the energy of activation of the de-excitation chan-
Figure 7. room-temperature diffuse reflectance spectra of com-
pounds 1 and 3; the arithmetic difference between the two curves
is also represented
5
nel with respect to the energy of the D0 level. The energy
activation obtained is 550 Ϯ 136 cmϪ1, which places the
nonradiative level in a position resonant with the Eu3ϩ
emitting level.
The charge-transfer band observed in the reflectance
In Eu3ϩ complexes such thermally activated nonradiative spectrum of compound 1 (Figure 7) does not appear at all
channels may involve ligand-to-metal charge-transfer as an excitation band (see the PLE spectra of Figure 6),
(LMCT) states, formed by the orbital overlap of the ligand indicating that the LMCT band in these compounds is pre-
and Eu3ϩ energy levels.[30Ϫ32] Quenching of the Eu3ϩ PL by dominantly a quenching channel. This is exactly the con-
low-lying thermally activated LMCT transitions is a well- clusion arrived at previously from the temperature-depen-
known phenomenon in molecular solids and complexes in dence of the Eu3ϩ emission integrated intensity of com-
solution.[31Ϫ34] These states are usually difficult to detect pound 1 (Figure 5). Actually, the energy of the thermally
by absorption and emission spectroscopy, and indeed their activated nonradiative level derived from Equation (4),
5
unequivocal assignment and detection have only been per- 550 Ϯ 136 cmϪ1 above the D0 level (about 17256.3 cmϪ1
,
formed for a few lanthanide compounds.[3,31Ϫ34] Due to the Figure 4), is in excellent agreement with the maximum of
difficulties in detecting LMCT states, they are usually the LMCT band discerned from the arithmetic difference
claimed to exist in order to explain low PL quantum yields between the reflectance spectra of 1 and 3 (approximately
and/or the strong dependence of the emission intensity with 17800 cmϪ1, Figure 7). Therefore, the low D0 quantum ef-
5
temperature. The comparison of Eu3ϩ and Gd3ϩ reflec- ficiency observed for compounds 1 and 2 can be reliably
tance spectra is a useful tool to perceive and ascribe the assigned to a nonradiative decay through the LMCT state
LMCT transitions, since the energetic difference between of Eu3ϩ, which is at rather low energies in these diimine
the Gd3ϩ first excited state and the fundamental levels is compounds. This is in agreement with previously reported
5
too high to allow the detection of a charge-transfer state in experimental results that indicate that the D0 quantum ef-
the UV/Vis range. Therefore, in the Gd3ϩ spectrum we ex- ficiency decreases as the energy of the LMCT states de-
pect to measure only the contributions that are not related creases.[31,32] Moreover, recent theoretical results on energy
with the LMCT processes.
transfer mechanisms between these LMCT states and the
Figure 7 shows the room-temperature diffuse reflectance ligand and Eu3ϩ levels demonstrate that the quantum yield
5
spectra of compounds 1 and 3. The arithmetic difference of the complex (and thus the D0 quantum efficiency) de-
between the two curves is also represented, showing the creases drastically as the LMCT barycentre becomes res-
5
presence of a low-lying LMCT band (centred at about onant with the D0 level.[35]
Eur. J. Inorg. Chem. 2004, 3913Ϫ3919
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3917

L. D. Carlos, P. J. A. Ribeiro-Claro et al.
FULL PAPER
1460 (m), 1384 (w), 1353 (w), 1298 (s), 1251 (m), 1198 (m), 1135
(m), 1074 (w), 959 (w), 865 (w), 792 (m), 749 (w), 684 (m), 568 (w),
472 (w) cmϪ1. Raman: 3058 (m), 1629 (s), 1596 (m), 1468 (s), 1432
(w), 1386 (s), 1298 (m), 1228 (w), 1199 (w), 1019 (w), 771 (m), 517
Concluding Remarks
New europium and gadolinium tris-β-diketonate com-
plexes have been prepared, containing diazabutadiene
(DAB) ligands coordinated to the metal centres. Accord-
ingly, the complexes with the general formula Ln(NTA)3·nL
(m) cmϪ1
.
Eu(NTA)3·o-tolyl؊DAB
(2):
Yield:
0.441 g
(82%).
(Ln ϭ Eu, Gd) are now known with L ϭ H2O, DMSO, C58H40EuF9N2O6 (1183.9): calcd. C 58.84, H 3.41, N 2.37; found
C 58.78, H 3.33, N 2.28. IR (KBr, selected): ν˜max. ϭ 3060 (w), 2959
(w), 2924 (w), 1613 (vs), 1593 (s), 1570 (s), 1532 (s), 1509 (s), 1460
(m), 1384 (w), 1353 (w), 1298 (vs), 1251 (m), 1198 (s), 1136 (s),
1074 (w), 958 (w), 864 (w), 793 (m), 749 (m), 684 (m), 569 (w), 519
(w), 472 (w) cmϪ1. Raman: 3056 (m), 1628 (s), 1596 (m), 1532 (m),
1468 (s), 1432 (w), 1386 (s), 1297 (m), 1228 (w), 1198 (w), 1019 (w),
2,2Ј-bipyridine, 1,10-phenanthroline and DAB. Binuclear
complexes of the type [Ln(NTA)3]2·bpym (bpym ϭ 2,2Ј-
bipyrimidine) have also been reported. The motivation for
the preparation of these derivatives stems from the fact that
the experimental quantum yield measured for
Eu(NTA)3·2DMSO (0.75) is one of the highest so far re-
ported for solid-state europium complexes. It has been
found that the 5D0 quantum efficiencies for these complexes
vary considerably depending on the nature of L, decreasing
in the order L ϭ DMSO (62%), phen (40%), bpym (39%),
771 (m), 517 (m) cmϪ1
.
Gd(NTA)3·p-tolyl؊DAB
(3):
Yield:
0.255 g
(46%).
C58H40F9GdN2O6 (1189.2): calcd. C 58.58, H 3.39, N 2.36; found
C 58.43, H 3.21, N 2.29. IR (KBr, selected): ν˜max. ϭ 3058 (w), 2960
(w), 2958 (w), 1615 (s), 1570 (m), 1532 (m), 1510 (m), 1384 (vs),
1299 (m), 1198 (m), 1135 (m), 959 (w), 792 (m), 684 (m), 570 (w),
471(w) cmϪ1. Raman: 3057 (m), 1629 (s), 1596 (m), 1468 (s), 1432
(w), 1386 (s), 1299 (m), 1228 (w), 1198 (w), 1019 (w), 770 (m), 517
5
H2O (29%) and DAB (2Ϫ3%). The low D0 quantum ef-
ficiencies for the diazabutadiene adducts can be reliably as-
signed to a nonradiative decay through the LMCT state
of Eu3ϩ, which is at rather low energies in these diimine
compounds. The use of photoluminescence spectroscopy,
together with vibrational spectroscopy and ab initio calcu-
lations, therefore allows us to build up quite a detailed pic-
ture about the nature of the metalϪligand interactions in
these tris-β-diketonate complexes.
(m) cmϪ1
.
Gd(NTA)3·o-tolyl؊DAB
(4):
Yield:
0.168 g
(39%).
C58H40F9GdN2O6 (1189.2): calcd. C 58.58, H 3.39, N 2.36; found
C 58.75, H 3.33, N 2.38. IR (KBr, selected): ν˜max. ϭ 3058 (w), 2960
(w), 2926 (w), 1614 (vs), 1594 (s), 1570 (s), 1531 (s), 1509 (s), 1460
(m), 1384 (m), 1353 (m), 1298 (s), 1251 (m), 1187 (s), 1135 (s), 1074
(w), 958 (w), 865 (w), 792 (m), 748 (m), 684 (m), 568 (w), 472 (w)
cmϪ1. Raman: 3057 (m), 1628 (s), 1596 (m), 1532 (m), 1467 (s),
1432 (w), 1386 (s), 1297 (m), 1227 (w), 1197 (w), 1019 (w), 770 (m),
Experimental Section
517 (m) cmϪ1
.
Materials and Methods: Eu(NTA)3·2H2O,[36] Gd(NTA)3·2H2O,[36]
p-tolylϪDAB,[37] and o-tolylϪDAB[37] were prepared by literature
methods.
Ab Initio Calculations: Ab initio calculations were performed using
the G03w program package,[38] running on a personal computer.
The fully optimised geometry, the harmonic vibrational frequenc-
ies, and the infrared and Raman intensities were obtained at the
B3LYP level, using the Effective Core Potentials (ECP) of Pacios
and Christiansen[39] and f function (zeta ϭ 0.591) for the Eu atom,
and the Stevens/Basch/Krauss[40,41] ECP minimal basis set (CEP-
4G option of G03) for the remaining atoms. Due to the large size
of the systems, the NTA ligands were replaced by the coordination
fragment [O(CH)3O]Ϫ. This model consistently yields EuϪO and
EuϪN bond lengths shorter than the reported X-ray values for the
phenanthroline,[24] bipyridine,[24] and diaquo[42] analogues, by 16%
and 7%, respectively. The calculated wavenumbers were scaled by
a factor of 0.90 before comparison with the experimental values.
´
Microanalyses were performed at the Instituto de Tecnologia Quı-
´
mica e Biologica (ITQB), Lisbon. Thermogravimetric analyses
(TGA) were performed using a Mettler TA3000 system at a heating
rate of 5 K minϪ1 under a static atmosphere of air. IR spectra were
obtained as KBr pellets with a FTIR Mattson-7000 infrared spec-
trophotometer. Raman spectra were recorded with a Bruker
RFS100/S FT instrument (Nd:YAG laser, 1064 nm excitation, In-
GaAs detector). The photoluminescence spectra (PL) were re-
corded on a Jobin Yvon-Spex spectrometer (HR, 460) coupled to
a R928 Hamamatsu photomultiplier (14Ϫ300 K). A 300 W Xe arc
lamp coupled to a Jobin Yvon-Spex (TRIAX 180) monochromator
was used as the excitation source. All the spectra were corrected
for the response of the detector. The lifetime measurements were
carried out at room temperature (r.t.) on a modular double grating
excitation spectrofluorimeter with a TRIAX 320 emission mono-
chromator (Fluorolog-3, Jobin Yvon-Spex) in front face mode cou-
pled to a R928 Hamamatsu photomultiplier.
Acknowledgments
The authors are grateful to FCT, POCTI and FEDER for financial
support (Project POCTI/CTM/46780/2002). J. F. thanks the Uni-
versity of Aveiro and the FCT for research grants.
Preparation of Complexes: A solution of the diimine (0.102 g,
0.432 mmol) in CHCl3 (5 mL) was added to a solution of
Ln(NTA)3·2H2O (0.432 mmol) in CHCl3 (15 mL). The reaction
mixture was stirred for 3 h at room temperature. The solvent was
then removed, and the resultant brown solid was washed with hex-
ane and dried in vacuo.
[1]
L. C. Thompson, in Handbook on the Physics and Chemistry
of Rare Earths (Eds.: K. A. Gscheidner, Jr. and L. Eyring),
North-Holland, Amsterdam, 1979, vol. 3, chapter 25.
W. DeW. Horrocks, M. Albin, Prog. Inorg. Chem. 1984, 31, 1.
[2]
[3]
´
´
G. F. de Sa, O. L. Malta, C. de Mello Donega, A. M. Simas,
R. L. Longo, P. A. Santa-Cruz, E. F. da Silva, Coord. Chem.
Rev. 2000, 196, 165.
Eu(NTA)3·p-tolyl؊DAB
(1):
Yield:
0.455 g
(89%).
[4]
[5]
C58H40EuF9N2O6 (1183.9): calcd. C 58.84, H 3.41, N 2.37; found
C 58.89, H 3.46, N 2.37. IR (KBr, selected): ν˜max. ϭ 3065 (w), 2961
(w), 2927 (w), 1614 (s), 1593 (m), 1570 (m), 1531 (m), 1510 (m),
K. C. Joshi, V. N. Pathak, Coord. Chem. Rev. 1977, 22, 37.
L. R. Melby, N. J. Rose, E. Abramson, J. C. Caris, J. Am.
Chem. Soc. 1964, 86, 5117.
3918
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 3913Ϫ3919

Spectroscopic Studies of Europium(III) and Gadolinium(III) Tris-β-diketonate Complexes
FULL PAPER
[6]
[7]
[8]
[27] [27a]
H. Bauer, J. Blanc, D. L. Ross, J. Am. Chem. Soc. 1964, 86,
5125.
O. L. Malta, M. A. Couto dos Santos, L. C. Thompson,
N. K. Ito, J. Lumin. 1996, 69, 77. [27b] O. L. Malta, H. F. Brito,
J. F. S. Menezes, F. R. Goncalves e Silva, S. Alves Jr., F. S.
A. K. Trikha, L. B. Zinner, K. Zinner, P. C. Isolani, Polyhedron
1996, 15, 1651.
¸
Farias Jr., A. V. M. de Andrade, J. Lumin. 1997, 75, 255.
M. H. V. Werts, R. T. F. Jukes, J. W. Verhoeven, Phys. Chem.
Chem. Phys. 2002, 4, 1542.
[28]
H. J. Batista, A. V. M. de Andrade, R. L. Longo, A. M. Simas,
´
G. F. de Sa, N. K. Ito, L. C. Thompson, Inorg. Chem. 1998,
[29]
[30]
37, 3542.
M. F. Hazenkamp, G. Blasse, Chem. Mater. 1990, 2, 105.
C. W. Struck, W. H. Fonger, Understanding Luminescence Spec-
tra and Efficiency Using Wp and Related Functions, Springer-
Verlag, Berlin, 1978.
[9]
L. C. Thompson, F. W. Atchison, V. G. Young, J. Alloys
Compd. 1998, 275Ϫ277, 765.
K. Iftikhar, M. Sayeed, N. Ahmed, Inorg. Chem. 1982, 21, 80.
D. F. Moser, L. C. Thompson, V. G. Young Jr., J. Alloys
Compd. 2000, 303Ϫ304, 121.
O. L. Malta, M. A. Couto dos Santos, L. C. Thompson, N. K.
Ito, J. Lumin. 1996, 69, 77.
V. Tsaryuk, J. Legendziewicz, L. Puntus, V. Zolin, J. Sokolnicki,
J. Alloys Compd. 2000, 300Ϫ301, 464.
[10]
[11]
[31]
´
C. de Mello Donega, S. J. L. Ribeiro, R. R. Goncalves, G.
¸
Blasse, J. Phys. Chem. Solids 1996, 11, 1727.
G. Blasse, N. Sabbatini, Mater. Chem. Phys. 1987, 16, 237.
J.-C. G. Bünzli, P. Froidevaux, J. M. Harrowfield, Inorg. Chem.
1993, 32, 3306.
[12]
[13]
[32]
[33]
[34]
R. Longo, F. R. Goncalves e Silva, O. L. Malta, Chem. Phys.
¸
[14]
[15]
R. C. Holz, L. C. Thompson, Inorg. Chem. 1993, 32, 5251.
L. C. Thompson, S. Berry, J. Alloys Compd. 2001, 323Ϫ324,
177.
Lett. 2000, 328, 67.
[35]
[36]
[37]
[38]
O. L. Malta, personal communication.
R. G. Charles, A. Perrotto, J. Inorg. Nucl. Chem. 1964, 26, 373.
R. Colton, I. B. Tomkins, Aust. J. Chem. 1965, 18, 447.
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M.
A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgom-
ery, R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam,
A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi,
V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C.
Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala,
Q. Cui, K. Morokuma, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz, B. B.
Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.
Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham,
C. Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe,
P. M. W. Gill, B. G. Johnson, W. Chen, M. W. Wong, J. L.
Andres, M. Head-Gordon, E. S. Replogle, J. A. Pople, -
 98 (Revision A.1), Gaussian, Inc., Pittsburgh, PA, 1998.
L. F. Pacios, P. A. Christiansen, J. Chem. Phys. 1985, 82, 2664.
W. Stevens, H. Basch, J. Krauss, J. Chem. Phys. 1984, 81, 6026.
W. J. Stevens, M. Krauss, H. Basch, P. G. Jasien, Can. J. Chem.
1992, 70, 612.
[16]
[17]
R. C. Holz, L. C. Thompson, Inorg. Chem. 1988, 27, 4640.
O. L. Malta, H. F. Brito, J. F. S. Menezes, F. R. Goncalves e
¸
´
Silva, C. de Mello Donega, S. Alves Jr., Chem. Phys. Lett.
1998, 282, 233.
[18]
[19]
O. L. Malta, H. F. Brito, J. F. S. Menezes, F. R. Goncalves e
¸
Silva, S. Alves Jr., F. S. Farias Jr., A. V. M. de Andrade, J.
Lumin. 1997, 75, 255.
´
L. D. Carlos, C. de Mello Donega, R. Q. Albuquerque, S. Alves
Jr., J. F. S. Menezes, O. L. Malta, Mol. Phys. 2003, 101, 1037.
G. van Koten, K. Vrieze, Adv. Organomet. Chem. 1982, 21, 151.
F. G. N. Cloke, H. C. Delemos, A. A. Sameh, J. Chem. Soc.,
Chem. Commun. 1986, 1344.
[20]
[21]
[22]
[23]
F. G. N. Cloke, Chem. Soc. Rev. 1993, 22, 17.
M. N. Bochkarev, A. A. Trifonov, F. G. N. Cloke, C. I. Dalby,
P. T. Matsunaga, R. A. Andersen, H. Schumann, J. Loebel, H.
Hemling, J. Organomet. Chem. 1995, 486, 177.
[39]
[40]
[41]
[24]
[25]
´
J. A. Fernandes, R. A. Sa Ferreira, M. Pillinger, L. D. Carlos,
J. Jepsen, A. Hazell, P. Ribeiro-Claro, I. S. Goncalves, J. Lu-
¸
min., in press.
[42]
J.-C. G. Bünzli, in Lanthanide Probes in Life, Chemical, and
Earth Sciences (Eds.: J.-C. G. Bünzli, G. R. Choppin), Elsevier,
Amsterdam, 1989, p. 219.
L. VanMeervelt, A. Froyen, W. D’Olieslager, C. Wal-
randGorller, I. Drisque, G. S. D. King, S. Maes, A. T. H. Len-
stra, Bull. Soc. Chim. Bel. 1996, 105, 377.
[26] [26a]
´
L. D. Carlos, R. A. Sa Ferreira, V. de Zea Bermudez, C.
Molina, L. A. Bueno, S. J. L. Ribeiro, Phys. Rev. B 1999, 60,
Received March 9, 2004
Early View Article
Published Online July 29, 2004
[26b]
10042.
O. L. Malta, H. J. Batista, L. D. Carlos, Chem.
Phys. 2002, 282, 21.
Eur. J. Inorg. Chem. 2004, 3913Ϫ3919
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3919