Visible-light excitation of infrared lanthanide luminescence via intra-ligand charge-transfer state in 1,3-diketonates containing push-pull chromophores

Article abstract of DOI:10.1002/ejic.200701197

A new 1,3-diketonato ligand containing both an electron-donor [4-(dimethylamino)benzene] and an electron-acceptor (4-nitrobenzene) group and its lanthanide complexes have been prepared. They display an intense intra-ligand charge-transfer absorption transition in the visible region of the spectrum at 400-550 nm which was utilized to achieve visible-light excitation of metal-centred infrared luminescence of Nd^III, Er^III and Yb^III ions. Wiley-VCH Verlag GmbH & Co. KGaA, 2008.

Full text of DOI:10.1002/ejic.200701197

FULL PAPER
DOI: 10.1002/ejic.200701197
Visible-Light Excitation of Infrared Lanthanide Luminescence via Intra-Ligand
Charge-Transfer State in 1,3-Diketonates Containing Push-Pull Chromophores
Nail M. Shavaleev,[a] Rosario Scopelliti,[a] Frédéric Gumy,[a] and Jean-Claude G. Bünzli*[a]
Keywords: Lanthanide / Luminescence / Near infrared / Energy transfer / 1,3-Diketonate / Chromophore
A new 1,3-diketonato ligand containing both an electron-do-
nor [4-(dimethylamino)benzene] and an electron-acceptor (4-
nitrobenzene) group and its lanthanide complexes have been
prepared. They display an intense intra-ligand charge-trans-
fer absorption transition in the visible region of the spectrum
at 400–550 nm which was utilized to achieve visible-light ex-
citation of metal-centred infrared luminescence of NdIII, ErIII
and YbIII ions.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2008)
with maximum absorption at 529 nm) to NdIII, ErIII, and
Introduction
YbIII [16]
.
An ever growing attention is being given to the lantha-
nide ions NdIII, ErIII and YbIII which display a long-lived
metal-centred luminescence in the near-infrared (NIR)
range of the spectrum at 800–2000 nm.[1–3] This interest is
stirred by potent applications in telecommunications, bio-
analyses, and medicine. The forbidden nature of the f-f
transitions in trivalent lanthanide ions results in a very
weak intensity of the metal-centred absorption bands. As a
result one has to coordinate a suitable chromophore to the
lanthanide ion and to rely on energy transfer in order to
achieve efficient excitation of the metal ion.[4] Whereas en-
ergy transfer from the triplet state of the chromophore moi-
ety is often a preferred (and commonly invoked) pathway
for the sensitization of the metal ion luminescence, several
other mechanisms can be operative, which involve the li-
gand singlet state,[5,6] metal-to-ligand charge-transfer
(MLCT) states,[4] or intra-ligand charge-transfer (ILCT)
states.[7–9] One of the challenges in the design of lumines-
cent lanthanide tags, particularly those aimed at bio-
analyses and bioimaging, is to shift the excitation wave-
length from the UV to the visible range, for both an intrin-
sic reason, since biomolecules are usually damaged by UV
light, and a practical one, visible excitation requiring
cheaper optical cells and optics. Several strategies have been
proposed towards this goal, for instance excitation through
d-metal ions,[2,4,10] by multiphoton excitation,[11–14] or via
ILCT states[7–9,15] which may have relatively low energy. For
instance, we recently showed that energy can be transferred
from bodipy (a boradiazaindacene-appended terpyridine
1,3-Diketones form stable lanthanide complexes and dis-
play intense absorption transitions, and thus have been ex-
tensively used to prepare luminescent lanthanide materi-
als.[17] However, most of the 1,3-diketones employed so far
with lanthanides absorb light mostly in the UV range, at
wavelengths Ͻ400 nm.[17–19] Only a few examples of lantha-
nide 1,3-diketonates are reported in which the absorption
spectrum of the donor ligand and the energy of its triplet
state are redshifted by incorporation of a condensed aro-
matic group,[20,21] ferrocene,[22] or a conjugated polyene
chain.[23] Herein we adopt a simple approach for redshifting
the absorption spectrum of the 1,3-diketonato ligand by in-
corporating both an electron-donor and an electron-
[a] École Polytechnique Fédérale de Lausanne (EPFL), Laboratory
of Lanthanide Supramolecular Chemistry
BCH 1402, 1015 Lausanne, Switzerland
Fax: + 41-21-693-9825
E-mail: jean-claude.bunzli@epfl.ch
Supporting information for this article is available on the
WWW under http://www.eurjic.org or from the author.
Scheme 1. Structures of the new 1,3-diketones and the lanthanide
complexes with L–.
Eur. J. Inorg. Chem. 2008, 1523–1529
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1523

N. M. Shavaleev, R. Scopelliti, F. Gumy, J.-C. G. Bünzli
FULL PAPER
acceptor group into the ligand. This creates a low-energy direction, while the electron-rich (dimethylamino)phenyl
absorption having an intra-ligand charge-transfer character. group of one molecule is situated above the electron-de-
To illustrate the concept, two new 1,3-diketones, HL and ficient nitrophenyl group of another molecule resulting in
HL-NMe2 (Scheme 1) and the lanthanide complexes with an aromatic π-π interaction (the molecules in the neigh-
ligand HL are prepared, characterized, and their photo- bouring layers are almost parallel with an average in-
physical properties reported.
terplanar distance of around 3.39 Å).
In the structure of the complex, the NdIII ion is octacoor-
dinated by six oxygen atoms from three ligands L– and two
nitrogen atoms from phen. The Nd–O bond lengths are in
the range 2.369(6)–2.420(6) Å (mean 2.386 Å), while the
Nd–N bond lengths are 2.662(7) Å and 2.683(7) Å (Fig-
ure 2). The coordination polyhedron around NdIII can be
best described as a square antiprism with two square faces
comprised of atoms N(1), N(2), O(4), O(3) and O(1), O(2),
O(5), O(6), respectively. The NdIII is situated out of the
plane of the 1,3-diketonato ring, the angles between the pla-
nes defined by Nd–O–O and O–C–C–C–O atoms (where C
and O atoms belong to a given 1,3-diketonato ring) are
12.99°, 20.18° and 20.91° for the three ligands. The NdIII
ion also lies out of the plane of the 1,10-phenanthroline
chelate ring, the angle between the planes defined by Nd–
N(1)–N(2) and N(1)–N(2)–C(56)–C(57) atoms being
13.54°.
Results and Discussion
Synthesis
The ligand HL was obtained as a bright red solid in 33%
yield by Claisen condensation of 4-(dimethylamino)aceto-
phenone with methyl 4-nitrobenzoate. It is present mainly
as enol form (Ͼ95%) in solutions in chloroform and dmso.
The lanthanide complexes [LnL3phen] (phen = 1,10-phen-
anthroline) were obtained as dark red solids in 70–85%
yield by a known procedure.[24] 1H NMR spectroscopy of
the diamagnetic LaIII complex recorded in dmso solution
confirmed the 3:1 (L–/phen) ratio and also indicated that
the ternary ligand phen is not coordinated to metal ions in
this solvent; the complex is therefore present as a solvate,
[LnL3(dmso)n].
The symmetrical ligand HL-NMe2 was obtained as a
bright yellow solid in 70% yield by Claisen condensation of
4-(dimethylamino)acetophenone with ethyl 4-(dimeth-
ylamino)benzoate. It is present as 80% enol and 20%
ketone in solutions in chloroform and dmso. Attempts to
prepare lanthanide complexes [Ln(L-NMe2)3phen] were un-
successful (the synthetic procedure is provided in the Sup-
porting Information). This may be the result of low acidity
of ligand HL-NMe2 due to the presence of electron-rich
dimethylamino groups. Further discussion of HL-NMe2
will be limited to a comparison of its spectroscopic proper-
ties with ligand HL.
Figure 2. Structure of the complex [NdL3phen] as viewed down the
square face of a square-antiprismatic coordination polyhedron of
NdIII centre (50% probability ellipsoids, H atoms, co-crystallized
solvent molecules and phenyl groups of the 1,3-diketonato ligands
omitted). Selected bond lengths [Å] and angles [°]: Nd(1)–O(3)
2.369(6), Nd(1)–O(6) 2.371(6), Nd(1)–O(2) 2.373(6), Nd(1)–O(5)
2.386(6), Nd(1)–O(4) 2.397(6), Nd(1)–O(1) 2.420(6), Nd(1)–N(1)
2.662(7), Nd(1)–N(2) 2.683(7); O(2)–Nd(1)–O(1) 69.3(2), O(3)–
Nd(1)–O(4) 70.06(18), O(6)–Nd(1)–O(5) 70.27(19), N(1)–Nd(1)–
N(2) 61.1(2).
Crystal Structures
The structures of the ligand HL and its NdIII complex
were determined by X-ray crystallography. The ligand HL
is nearly planar (Figure 1). The angles between the planes
defined by the 1,3-diketone central moiety and the benzene
rings of either the nitrophenyl or (dimethylamino)phenyl
groups are 3.58° and 7.64°, respectively. These planar mole-
cules are packed in a column-like structure running along
the crystallographic a-axis. Within the column, the 1,3-dike-
tonato moieties of the molecules are oriented in the same
Only one of the three 1,3-diketonato ligands is relatively
planar. It is involved in an aromatic π-π interaction with
the “planar” 1,3-diketonato ligand of another molecule; the
corresponding plane-to-plane distance is 3.401 Å [the pla-
nes were calculated by including all atoms in the ligand
apart from the NO2 and N(CH3)2 groups]. The interacting
pair of “planar” 1,3-diketonato ligands are related by a cen-
tre of inversion so that an electron-rich (dimethylamino)-
phenyl group of one ligand is situated above an electron-
deficient nitrophenyl group of another, similar to the situa-
tion observed in the structure of ligand HL. The ligand
phen is nearly planar and is involved in aromatic π-π inter-
action with a phen ligand of another molecule; the corre-
sponding plane-to-plane distance is 3.502 Å. The “planar”
Figure 1. Structure of the ligand HL (50% probability ellipsoids,
H atoms omitted).
1524
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 1523–1529

Visible-Light Excitation of Infrared Lanthanide Luminescence
1,3-diketonato ligand is found in “trans” position to the still displayed a low-energy absorption at 436 nm due to a
phen ligand in the coordination sphere of NdIII. The aro- charge transfer from the dimethylamino group to the 1,3-
matic π-π interactions that both of these ligands are in- diketonato moiety (Figure S2, Supporting Information).
volved in, results in the packing of molecules in a chain- One may consider HL-NMe2 as an analogue of Michler’s
like structure running along the crystallographic c-axis ketone (Scheme 2) – a push-pull chromophore, which was
(Figure S1, Supporting Information). The inter-chain inter- shown to be an excellent sensitizer of lanthanide lumines-
action is supported by weaker aromatic π-π interactions be- cence.[27]
tween “non-planar” 1,3-diketonato ligands.
Comparison of the structure of [NdL3phen] with that re-
ported for a model complex [Nd(dbm)3phen] (dbm = diben-
zoylmethane, Scheme 2)[25] reveals somewhat slightly longer
bonds in the case of [Nd(dbm)3phen], which are in the
range 2.379–2.425 Å for Nd–O bonds (mean: 2.393) and
2.692 Å and 2.699 Å for Nd–N bonds. However, in sharp
contrast to [NdL3phen], the chelate rings in [Nd(dbm)3-
phen] are much closer to planarity, the corresponding
angles are 4.99°, 7.13° and 8.91, for 1,3-diketonato chelate
rings, and 8.98° for the phen chelate ring.
Figure 3. Absorption spectra of the ligand HL (2.69ϫ10–4 ) and
the complex [YbL3phen] (1.26ϫ10–4 ) in dmso solution.
In fact, similar 1,3-diketones containing 4-(dialkyl- or
diarylamino)phenyl groups reported in the litera-
ture[15,26,28–32] also display a low-energy absorption band
which was shown by spectroscopic measurements[26,28] and
calculations[31] to be of a charge-transfer nature. In com-
parison, the lowest-energy absorption transition in dibenzo-
ylmethane and bis(p-nitrobenzoyl)methane – model 1,3-di-
ketones which do not contain donor groups – occurs in the
onset of the UV range at 352 nm and 362 nm,[17]8 respec-
tively (Figure S2, Supporting Information).
Upon coordination to a lanthanide ion, the visible ab-
sorption band of the ligand HL is blueshifted by 14–19 nm
Scheme 2. Structures of the known ligands discussed in the paper.
(Figure 3, Figure S3 in the Supporting Information), while
At last, it may be noted that intermolecular interaction
the UV band at 334 nm experiences no spectral shift. The
observed in the crystal structures of ligand HL and its NdIII
intensity of the UV band in the complexes [ε = (5.2–
complex may explain the low solubility of these compounds
in common organic solvents.
5.6)ϫ104 –1 cm–1] is about threefold higher than its inten-
sity in the neutral ligand [ε = 1.8ϫ104 –1 cm–1] as would
be expected in a tris complex. At the same time the intensity
of the visible band in the complexes [ε
= (5.5–
Photophysical Properties
6.5)ϫ104 –1 cm–1] is less than threefold higher than the
The absorption spectrum of ligand HL (Figure 3) dis- neutral ligand [ε = 2.4ϫ104 –1 cm–1]. The ligand-centred
plays two bands with maxima at 339 (logε = 4.25) and visible absorption band gains intensity moving from lighter
442 nm (logε = 4.38). The low-energy band at 442 nm, to heavier lanthanides and is redshifted probably due to the
which determines the intense red colour of the ligand, is increased Lewis acidity of the metal ion, while no signifi-
likely a transition with intra-ligand charge-transfer charac- cant changes occur in the UV band.
ter, wherein the dimethylamino group serves as an electron
The ligand HL and its LaIII and GdIII complexes do not
donor and both 1,3-diketone and nitro groups serve as elec- display any emission in dmso solution at room temperature.
tron acceptors. The charge-transfer absorption band of li- No emission could be detected from the GdIII complex in
gand HL is redshifted by 25 nm compared to the one re- dmso glass at 77 K either, but weak emission is seen in the
ported for the 4-cyanophenyl analogue HL-CN (Scheme 2, solid state at 77 K. It consists of two very broad bands
λmax = 417 nm in polar solvents),[26] probably due to the centred at 526 and 723 nm, which are tentatively assigned
stronger electron-acceptor properties of the nitro substitu- to the fluorescence and phosphorescence of the coordinated
ent relative to the cyano group. The role of the 1,3-diketon- ligand L–, respectively (Figure S4) (the first excited state of
ato moiety as an electron acceptor in a charge-transfer tran- GdIII is located above 32000 cm–1[33] and thus cannot accept
sition in ligand HL was confirmed by comparison with HL- energy from the ligand; as a result, GdIII chelates generally
NMe2. The ligand HL-NMe2, which lacks the nitro group, display ligand-centred photophysical processes only).
Eur. J. Inorg. Chem. 2008, 1523–1529
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1525

N. M. Shavaleev, R. Scopelliti, F. Gumy, J.-C. G. Bünzli
FULL PAPER
Attempts to enforce a time delay (10 µs) in the luminescence
measurements resulted in the complete disappearance of
both bands. Attempts to measure the lifetime of the emis-
sion band at 723 nm failed due to weak intensity and an
apparent poly-exponential decay profile; however, the long
timescale of the decay up to a microsecond range is consis-
tent with a phosphorescence process. The energy of the 0–
0 transition of the triplet state of ligand L–, estimated to be
around 16500 cm–1 from the onset of the phosphorescence
band, indicates that this ligand is potentially suitable for the
sensitization of infrared emission of lanthanide ions.
Indeed, [LnL3phen] complexes (Ln = Nd, Er, and Yb)
display a characteristic line-like infrared emission due to f-
f transitions upon excitation in ligand absorption bands,
both in the solid state and in solutions in dmso (Figure 4).
Figure 4. Corrected and normalized luminescence spectra of
[YbL3phen] in dmso solution (1.8ϫ10–6 ) and in the solid state
at room temperature under excitation at 430 nm.
Luminescence spectra of the lanthanide complexes are red- processes in the solid state. The complexes are photo-stable
shifted by 3–11 nm in solution compared to the solid state in dmso solution but undergo some photochemical decom-
(Figure 4, Figures S5, S6), a fact we trace back to the disso- position in the solid state during photophysical experi-
ciation of the ancillary phen ligand in solution (see dis- ments, as evidenced by gradual reduction of the absolute
1
cussion of H NMR spectra above). Energies of the bands quantum yield in a series of consecutive experiments. At the
in the excitation spectra of the NdIII and YbIII complexes same time, the luminescence decays of the solid-state sam-
(Figures S7, S8) match those of the absorption spectra, thus ple are still single exponential functions, and the associated
confirming that the ligand sensitizes lanthanide lumines- lifetimes do not vary in consecutive experiments, indicating
cence. We note, however, that the band intensity in absorp- that the products of photo-decomposition do not contrib-
tion and excitation spectra is similar for the NdIII complex ute to the observed emission.
(Figure S7) but is different in the case of YbIII: excitation
Absolute quantum yields of the metal-centred lumines-
of YbIII within the visible absorption band is less efficient cence upon ligand excitation (QLnL) of the new complexes
compared to the UV band (Figure S8). The reason for this in solution turned out to be an order of magnitude lower
difference between NdIII and YbIII is not clear, but it may than those of other lanthanide complexes with 1,3-diketon-
be traced back to the fact that excitation of YbIII ion in its ates. For example, the emission quantum yield for the
complexes may occur through a double electron transfer[34] model compound [Yb(dbm)3phen] was reported to be
in addition to a common phonon-assisted transfer mecha- 5.9ϫ10–3 in toluene solution,[37] whereas for the new
nism.[35,36] The excitation spectrum of ErIII could not be [YbL3phen] complex it only amounts to 4.0ϫ10–4 in dmso.
recorded due to weak emission intensity.
The QLn in a simplified form can be expressed as QLn
L L
The luminescence decays of the [LnL3phen] complexes = ηϫQLnLn, where η is a ligand-to-metal energy transfer
are in all cases single exponential functions, indicating the (sensitization) efficiency and QLn is an intrinsic quantum
Ln
presence of either only one emitting centre (in solution or yield of a lanthanide ion. A rough estimate of the intrinsic
solid state) or rapid equilibrium between different species quantum yield QLn = τobs/τrad, taking accepted literature
Ln
in solution. The luminescence lifetimes (τ, Table 1) in solu- values[2] for the radiative lifetimes (0.42 ms for NdIII and
tion are about 1.5-fold longer than in the solid state which 2 ms for YbIII), gives QNd
= 2.9ϫ10–3 and QYb
=
Nd
Yb
might indicate the presence of intermolecular quenching 6ϫ10–3, leading to very low ligand-sensitization efficiencies
Table 1. Spectroscopic properties of ligand HL, its lanthanide complexes, and related 1,3-diketones at room temperature.
Compound
Absorption (dmso)[a]
Luminescence[b]
λmax/nm (logε)
λmax/nm (dmso, solid)
τ/µs, QLn (dmso)
τ/µs (solid)[c]
L
[d]
HL
442 (4.38), 339 (4.26)
424 (4.75), 342 (4.74)
424 (4.75), 341 (4.74)
423 (4.74), 340 (4.75)
426 (4.77), 341 (4.73)
428 (4.81), 343 (4.72)
436 (4.76), 352 (1.26)
352 (4.34)
[d]
[LaL3phen]
[NdL3phen]
[GdL3phen]
[ErL3phen]
[YbL3phen]
HL-NMe2
dbm
1062, 1059
1.2, 7.8ϫ10–5
0.71
[d]
1520, 1509
948, 938
1.9[e]
1.4
8.6
12, 4.0ϫ10–4
[a] Spectra were recorded in the spectral range 300–800 nm. Estimated errors are Ϯ1 nm for λmax and Ϯ5% for ε. The concentration of
the samples was in the range (8–27)ϫ10–5 . The complexes were present in dmso solution in the form of solvates [LnL3(dmso)n]. [b] At
room temp. with λexc = 355 (τ) and 430 nm (QLnL). Estimated error on QLn is Ϯ15%. The concentration of the complexes was (0.2–
L
13)ϫ10–5 . The complexes were present in dmso solution in the form of solvates [LnL3(dmso)n]. [c] Photochemical degradation of the
sample (see text); the lower-limit estimation for QYb in solid state is 0.20%. [d] The compounds did not display emission under stated
L
conditions. [e] Luminescence quantum yield could not be determined as a result of a low emission intensity.
1526
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 1523–1529

Visible-Light Excitation of Infrared Lanthanide Luminescence
ples for absorption studies was in the range (8–27)ϫ10–5  and for
luminescence studies in the range (0.2–13)ϫ10–5 .
of 2.7 and 6.7%, respectively. In the case of [Yb(dbm)3phen]
(τ = 10.3 µs) similar calculations give QYbYb = 5ϫ10–3 and
a ligand-sensitization efficiency of 87%. Therefore, the ob-
1,3-Diketone HL: The reaction was performed under nitrogen using
served difference between [YbL3phen] and [Yb(dbm)3phen] degassed and dry solvents. NaH (Fluka; 55–65% suspension in
mineral oil) (110 mg, containing at least 60.5 mg, 2.52 mmol of
NaH) was washed with hexane under nitrogen and was suspended
in dry THF (4 mL) at room temp. 4-(Dimethylamino)acetophenone
(TCI Europe; 250 mg, 1.53 mmol) was added, and the mixture was
stirred or sonicated at room temp. for 10 min. The reaction mixture
was then cooled to 0 °C. This was followed by addition of methyl
4-nitrobenzoate (Aldrich; 300 mg, 1.66 mmol, small excess) in small
portions over 5 min. The colour of the reaction mixture changed
instantly from colourless to orange upon addition of the ester. The
reaction mixture was stirred at 0 °C for 30 min, warmed to room
complexes lies in a less efficient ligand-to-metal energy
transfer in the case of ligand L–, probably as a result of fast
electronic energy loss within the ligand which competes
with the energy-transfer step. The energy transfer is, how-
ever, significantly improved in the solid-state sample of
[YbL3phen] which displays a quantum yield equal to
2.0ϫ10–3 (a lower limit estimate due to a photochemical
decomposition). The corresponding lifetime is 8.6 µs which
results in QYbYb = 4.3ϫ10–3 and a ligand-sensitization effi-
ciency of 47% (again, a lower limit estimate). This differ- temp. and stirred with occasional sonication at room temp. for fur-
ther 2 or 3 hrs. The reaction had an initiation time and was sluggish
for the first 1 hr, the colour of the reaction mixture at that stage
was dark orange, and both of the starting materials were observed
by TLC. The reaction rapidly self-accelerated after about 1 hr and
evolved heat while the colour of the mixture changed to dark
brownish-red, almost black. At that point both starting materials
were completely consumed. The “rapid” stage of the reaction took
about 5–10 min to complete, and one should be careful of that
when planning a synthesis on a larger scale. The mixture was
quenched with methanol, concentrated to dryness and suspended
in water. Acetic acid was added to the water, the product was ex-
tracted with CH2Cl2, and the organic layer was separated and dried
(MgSO4). Purification by column chromatography (17 g of silica,
eluting with CH2Cl2) provided a crude product, containing at least
three by-products, which was dissolved in CH2Cl2; ethanol (10 mL)
was added, and the CH2Cl2 was evaporated to leave a suspension
of pure crystalline product in ethanol. The product was filtered and
ence may indicate that in dmso solution the average metal–
ligand bonds are longer than in the solid state probably
due to a partial dissociation of the 1,3-diketonato ligand[38]
which leads to a diminished energy-transfer efficiency.
Conclusion
We have demonstrated that 1,3-diketonato ligands con-
taining push-pull chromophores are suitable for visible-light
excitation of NIR-emitting lanthanide ions. The main ad-
vantage of the ligand L– is its lowest-energy absorption
transition which extends into the visible range and allows
excitation of lanthanide luminescence with wavelengths up
to 550 nm. For instance, the overall luminescence efficiency
of the YbIII complex, εϫQLnL, is 26 –1 cm–1 at 428 nm
(absorption maximum in the visible range), 5.2 –1 cm–1 at washed with ethanol (5 mL) and diethyl ether (5 mL). The com-
500 nm, and ca. 0.4 –1 cm–1 at 550 nm. However, further
pound can be re-crystallized in the same manner, if required; the
loss of product is minimal due to its very low solubility in ethanol.
improvement of the ligand design ought to be implemented
The compound was soluble in CH2Cl2 up to 3 mg in 1 mL. Bright
to increase the energy-transfer efficiency and photostability.
red solid, yield 157 mg (0.50 mmol, 33%). C17H16N2O4 (312.32):
If successful, these ligands might be used in the design of
calcd. C 65.38, H 5.16, N 8.97; found C 65.05, H 4.98, N 9.11. In
lanthanide complexes with NIR emission and, possibly,
dmso, the compound was present as a mixture of 95% enol and
metal complexes with non-linear optical properties. This as-
pect is being presently investigated in our laboratory.
1
5% ketone. H NMR (400 MHz, [D6]dmso): enol signals: δ = 8.35
(s, 4 H), 8.07 (d, J = 8.8 Hz, 2 H), 7.31 (s, 1 H, methine CH), 6.80
(d, J = 8.8 Hz, 2 H), 3.07 (s, 6 H) ppm; some of the ketone signals:
δ = 8.18 (d, J = 8.8 Hz, 2 H), 7.82 (d, J = 8.8 Hz, 2 H), 6.74 (d, J
= 9.2 Hz, 2 H), 4.80 (s, 2 H, methylene CH2), 3.04 (s, 6 H) ppm.
Experimental Section
Chemicals and General Procedures: Chemicals obtained from com-
mercial suppliers were used without further purification. Elemental
analyses were performed by Dr. E. Solari, Service for Elemental
Analysis, Institute of Chemistry and Chemical Engineering Sci-
ences (EPFL). Absorption spectra were measured with a Perkin–
Elmer Lambda 900 UV/Vis/NIR spectrometer. 1H NMR spectra
(presented as δ in ppm and J in Hz) were recorded with a Bruker
Avance DRX 400 MHz spectrometer. Luminescence emission and
excitation spectra were measured with a Fluorolog FL 3-22 spec-
trometer from Horiba-Jobin Yvon-Spex equipped for both Vis and
NIR measurements. Quantum-yield data for solid samples and for
solutions were determined with the same instrument, through an
absolute method using a home-modified integrating sphere.[39] Lu-
minescence lifetimes were measured with a previously described in-
strumental setup.[16] All luminescence spectra were corrected for
the instrumental function. Spectroscopic studies in solutions were
conducted in dmso (Fluka, Ͼ99.9%, A.C.S. spectrophotometric
grade) using optical cells of either 2 or 10 mm path length and
quartz capillaries with i.d. of 2 mm. The concentration of the sam-
Lanthanide Complexes [LnL3phen] (Ln = La, Nd, Gd, Er, Yb): The
reaction was performed under air. Ligand HL (50 mg, 0.16 mmol),
1,10-phenanthroline (9.6 mg, 0.053 mmol) and NaOH (6.4 mg,
0.16 mmol, dissolved in 1 mL of water) were suspended in a mix-
ture of boiling ethanol (7 mL) and CH3CN (10 mL) and were
stirred until the suspension had dissolved to give a red solution.
The latter was stirred for additional 5 min, followed by dropwise
addition of the appropriate lanthanide chloride hydrate (Aldrich;
0.053 mmol) dissolved in 2 mL of ethanol. This resulted in the for-
mation of a dark red precipitate which was stirred at reflux for
15 min. The suspension was filtered while hot, and the resulting
solid was washed with CH3CN, a 1:1 mixture of ethanol/water, eth-
anol, and finally diethyl ether. The complexes were obtained as
deep-red solids and were soluble in CH2Cl2 and dmso (upon slight
warming) with the solubility significantly decreasing for heavier
lanthanides. [LaL3phen]·H2O: Yield 47 mg (0.037 mmol, 70%).
C63H53LaN8O12·H2O (1271.06): calcd. C 59.53, H 4.36, N 8.82;
1
found C 59.82, H 4.33, N 8.85. H NMR spectrum of the complex
in dmso shows clearly the resolved signals of phen which coincides
Eur. J. Inorg. Chem. 2008, 1523–1529
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1527

N. M. Shavaleev, R. Scopelliti, F. Gumy, J.-C. G. Bünzli
FULL PAPER
with those of “free” phen. This indicates that in dmso, phen is not
coordinated to the lanthanide ion and the complex is present as a
solvate [LaL3(dmso)n]. The 1H NMR spectra of the complex in
dmso at room temp. in the range of the signals of the 1,3-diketon-
ato ligand are independent of the concentration in the range
7.3ϫ10–4 to 2.5ϫ10–3  and show the presence of a major and a
minor species (possibly isomers) in the ratio Ͼ10:1. The signals of
the major and minor species coalesce at higher temperatures in the
range 320–340 K. Integration of signal intensities confirms the 1:3
ratio of phen/1,3-diketone in the complex. 1H NMR (400 MHz,
[D6]dmso): at r.t.: major species: δ = 9.12 (br. s, 2 H, phen), 8.50
(d, J = 7.2 Hz, 2 H, phen), 8.10–8.25 (m, 12 H), 8.00 (s, 2 H, phen),
7.94 (d, J = 8.4 Hz, 6 H), 7.78 (dd, J = 8.0, J = 4.4 Hz, 2 H, phen),
6.72 (s, 3 H, methine CH), 6.64 (d, J = 8.8 Hz, 6 H), 2.98 [s, 18 H,
N(CH3)2] ppm; some signals of minor species: δ = 8.3–8.4 (br.),
8.3–8.10 (d, br.), 6.78 (s, br.), 6.55–6.65 (d, br.) ppm; at 340 K: δ =
9.11 (dd, 2 H, phen), 8.48 (dd, J = 8.0, J = 1.6 Hz, 2 H, phen),
8.05–8.25 (m, 12 H), 7.98 (s, 2 H, phen), 7.93 (d, J = 8.4 Hz, 6 H),
7.76 (dd, J = 8.0, J = 4.4 Hz), 6.55–6.75 (m, 9 H), 2.98 [s, 18 H,
N(CH3)2] ppm. 1H NMR spectra of the complex in CDCl3 are
complicated and concentration-dependent due to the partial disso-
ciation of the phen ligand. [NdL3phen]·H2O: Yield 53 mg
(0.042 mmol, 78%). [C63H53N8NdO12]·H2O (1276.40): calcd. C
59.28, H 4.34, N 8.78; found C 59.32, H 4.15, N 8.85. No water
molecules coordinated to NdIII were observed in the X-ray struc-
ture of the complex. [GdL3phen]: Yield 57 mg (0.045 mmol, 85%).
temperature using Mo-Kα radiation. The used equipment was an
Oxford Diffraction Sapphire/KM4 CCD for [NdL3phen] and a
Bruker APEX II CCD for HL having both kappa geometry goni-
ometers. Data were reduced by CrysAlis PRO[40] in the case of
[NdL3phen] and EvalCCD[41] in the case of HL. Semiempirical ab-
sorption correction[42] was applied to the data sets. Structure solu-
tions and refinements were carried out by SHELXTL.[43] Crystal
structures were refined using full-matrix least squares on F2 with
all non-hydrogen atoms anisotropically refined. Hydrogen atoms
were placed in calculated positions by means of the “riding” model.
Disorder problems dealing with the solvent were found for com-
pound [NdL3phen], and these were treated by applying some re-
straints and constraints. Additional crystal data and structure-re-
finement parameters are provided in Tables S1 and S2. CCDC-
666466 and CCDC-666467 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Supporting Information (see footnote on the first page of this arti-
cle): Description of an attempted synthesis of [La(L–NMe2)3phen],
crystallographic tables (Tables S1, S2), crystal packing of
[NdL3Phen] (Figure S1), additional absorption spectra (Figures S2,
S3), luminescence spectra (Figures S4–S6) and luminescence exci-
tation spectra (Figures S7, S8).
C63H53GdN8O12 (1271.39): calcd. C 59.52, H 4.20, N 8.81; found Acknowledgments
C 59.03, H 4.29, N 8.55. [ErL3phen]: Yield 58 mg (0.045 mmol,
This research is supported by a grant from the Swiss National Sci-
ence Foundation.
85%). C63H53ErN8O12 (1281.40): calcd. C 59.05, H 4.17, N 8.74;
found 58.53, 3.87, 8.49. [YbL3phen]: Yield 56 mg
C
H
N
(0.044 mmol, 82%). C63H53N8O12Yb (1287.18): calcd. C 58.79, H
4.15, N 8.71; found C 58.26, H 4.04, N 8.66.
[1] S. Faulkner, S. J. A. Pope, B. P. Burton-Pye, Appl. Spectrosc.
Rev. 2005, 40, 1–31.
Synthesis of 1,3-Diketone HL-NMe2: The reaction was performed
under nitrogen using degassed and dry solvents. NaH (Fluka; 55–
65% suspension in mineral oil) (110 mg, contains at least 60.5 mg,
2.52 mmol of NaH, excess) was washed with hexane and suspended
in dry THF (4 mL) at room temp. 4-(Dimethylamino)acetophenone
(TCI Europe; 250 mg, 1.53 mmol) was added, and the mixture was
stirred or sonicated at room temp. for 10 min. This was followed
by addition of ethyl 4-(dimethylamino)benzoate (Acros; 340 mg,
1.76 mmol, small excess). The reaction mixture was refluxed for 7 h
to give a viscous yellow suspension. It was quenched by successive
addition of methanol (3 mL), acetic acid (3 mL) and water (50 mL).
The product was extracted from the resulting bright greenish-yel-
low suspension with CH2Cl2. Purification by column chromatog-
raphy (18 g of silica, eluting first with CH2Cl2 to remove impurities
and then with 0.2% CH3OH in CH2Cl2 to recover the product)
provided a crude product. It was dissolved in CH2Cl2, ethanol
(10 mL) was added and the CH2Cl2 was evaporated to leave a sus-
pension of pure product in ethanol which was filtered and washed
with cold ethanol (5 mL) and cold diethyl ether (5 mL). The com-
pound has a reasonable solubility in ethanol, and recrystallization
results in a diminished yield. Yield 332 mg (1.07 mmol, 70%);
bright yellow solid. C19H22N2O2 (310.39): calcd. C 73.52, H 7.14,
N 9.03; found C 73.58, H 7.15, N 9.05. In dmso, the compound is
present as a mixture of 81% enol and 19% ketone. 1H NMR
(400 MHz, [D6]dmso): enol signals: δ = 7.96 (d, J = 9.2 Hz, 4 H),
6.95 (s, 1 H, methine CH), 6.76 (d, J = 8.8 Hz, 4 H), 3.04 (s, 12 H)
ppm; ketone signals: δ = 7.80 (d, J = 9.2 Hz, 4 H), 6.72 (d, J =
8.8 Hz, 4 H), 4.46 (s, 2 H, methylene CH2), 3.02 (s, 12 H) ppm.
[2] Comby, S., Bünzli, J.-C. G., “Lanthanide Near-Infrared Lumi-
nescence in Molecular Probes and Devices”, in Handbook on
the Physics and Chemistry of Rare Earths, Elsevier Science
B. V., Amsterdam, 2007, vol. 37, chapter 235.
[3] T. Gunnlaugsson, F. Stomeo, Org. Biomol. Chem. 2007, 5,
1999–2009.
[4] J.-C. G. Bünzli, C. Piguet, Chem. Soc. Rev. 2005, 34, 1048–
1077.
[5] C. Yang, L. M. Fu, Y. Wang, J. P. Zhang, W. T. Wong, X. C.
Ai, Y. F. Qiao, B. S. Zou, L. L. Gui, Angew. Chem. Int. Ed.
2004, 43, 5010–5013.
[6] N. S. Baek, Y. H. Kim, S. G. Roh, B. K. Kwak, H. K. Kim,
Adv. Funct. Mater. 2006, 16, 1873–1882.
[7] V. F. Zolin, L. N. Puntus, V. I. Tsaryuk, V. A. Kudryashova, J.
Legendziewicz, P. Gawryszewska, R. Szostak, J. Alloys Compd.
2004, 380, 279–284.
[8] A. Vogler, H. Kunkely, Inorg. Chim. Acta 2006, 359, 4130–4138.
[9] L. N. Puntus, A.-S. Chauvin, S. Varbanov, J.-C. G. Bünzli, Eur.
J. Inorg. Chem. 2007, 2315–2326.
[10] a) T. K. Ronson, T. Lazarides, H. Adams, S. J. A. Pope, D.
Sykes, S. Faulkner, S. J. Coles, M. B. Hursthouse, W. Clegg,
R. W. Harrington, M. D. Ward, Chem. Eur. J. 2006, 12, 9299–
9313; b) A. Beeby, R. S. Dickins, S. FitzGerald, L. J. Goven-
lock, D. Parker, J. A. G. Williams, C. L. Maupin, J. P. Riehl, G.
Siligardi, Chem. Commun. 2000, 1183–1184.
[11] J. R. Lakowicz, G. Piszczek, B. P. Maliwal, I. Gryczynski,
ChemPhysChem 2001, 2, 247–252.
[12] M. H. V. Werts, N. Nerambourg, D. Pelegry, Y. Le Grand, M.
Blanchard-Desce, Photochem. Photobiol. Sci. 2005, 4, 531–538.
[13] L. M. Fu, X. F. Wen, X. C. Ai, Y. Sun, Y. S. Wu, J. P. Zhang,
Y. Wang, Angew. Chem. Int. Ed. 2005, 44, 747–750.
[14] A. Picot, F. Malvolti, B. LeGuennic, P. L. Baldeck, J. A. G.
Williams, C. Andraud, O. Maury, Inorg. Chem. 2007, 46, 2659–
2665.
X-ray Structural Studies: Single crystals for X-ray analysis were ob-
tained by slow concentration of solutions of HL or [NdL3phen]
in CH2Cl2/heptane or CH2Cl2/hexane. Data were collected at low
1528
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 1523–1529

Visible-Light Excitation of Infrared Lanthanide Luminescence
[15] D. B. Nie, Z. Q. Chen, Z. Q. Bian, J. Q. Zhou, Z. Liu, F. F.
Chen, Y. L. Zhao, C. H. Huang, New J. Chem. 2007, 31, 1639–
1646.
[16] R. Ziessel, G. Ulrich, L. J. Charbonnière, D. Imbert, R. Scopel-
liti, J.-C. G. Bünzli, Chem. Eur. J. 2006, 12, 5060–5067.
[17] K. Binnemans, “Rare earth β-diketonate complexes: function-
alities and applications”, in Handbook on the Physics and
Chemistry of Rare Earths, Elsevier Science B. V., Amsterdam,
2005, vol. 35, chapter 225.
[29] M. R. Robinson, G. C. Bazan, M. B. O’Regan, Chem. Com-
mun. 2000, 1645–1646.
[30] B. Konig, Synth. Commun. 2003, 33, 967–976.
[31] J. K. Yu, Y. M. Cheng, Y. H. Hu, P. T. Chou, Y. L. Chen, S. W.
Lee, Y. Chi, J. Phys. Chem. B 2004, 108, 19908–19911.
[32] N. J. Xiang, L. M. Leung, S. K. So, J. Wang, Q. Su, M. L.
Gong, Mater. Lett. 2006, 60, 2909–2913.
[33] W. T. Carnall, “The Absorption and Fluorescence Spectra of
Rare Earth Ions in Solution”, in Handbook on the Physics and
Chemistry of Rare Earths, North Holland Publ. Co., Amster-
dam, 1979, vol. 3, chapter 24.
[18] W. F. Sager, N. Filipescu, F. A. Serafin, J. Phys. Chem. 1965,
69, 1092–1100.
[19] S. Sato, M. Wada, Bull. Chem. Soc. Jpn. 1970, 43, 1955–1962.
[20] M. K. Nah, H. G. Cho, H. J. Kwon, Y. J. Kim, C. Park, H. K.
Kim, J. G. Kang, J. Phys. Chem. A 2006, 110, 10371–10374.
[21] L. F. Yang, Z. L. Gong, D. B. Nie, B. Lou, Z. Q. Bian, M.
Guan, C. H. Huang, H. J. Lee, W. P. Baik, New J. Chem. 2006,
30, 791–796.
[22] Y. F. Yuan, T. Cardinaels, K. Lunstroot, K. Van Hecke, L.
Van Meervelt, C. Gorller-Walrand, K. Binnemans, P. Nockem-
ann, Inorg. Chem. 2007, 46, 5302–5309.
[23] M. D. Seltzer, S. Fallis, R. A. Hollins, N. Prokopuk, R. N. Bui,
J. Fluoresc. 2005, 15, 597–603.
[24] L. R. Melby, N. J. Rose, E. Abramson, J. C. Caris, J. Am.
Chem. Soc. 1964, 86, 5117–5125.
[34] W. D. Horrocks Jr, J. P. Bolender, W. D. Smith, R. M. Supkow-
ski, J. Am. Chem. Soc. 1997, 119, 5972–5973.
[35] G. A. Crosby, M. Kasha, Spectrochim. Acta 1958, 10, 377–382.
[36] C. Reinhard, H. U. Güdel, Inorg. Chem. 2002, 41, 1048–1055.
[37] M. P. Tsvirko, S. B. Meshkova, V. Y. Venchikov, Z. M. Topi-
lova, D. V. Bol’shoi, Opt. Spectrosc. (Engl. Transl.) 2001, 90,
669–673.
[38] Y. Hasegawa, M. Iwamuro, K. Murakoshi, Y. Wada, R. Arak-
awa, T. Yamanaka, N. Nakashima, S. Yanagida, Bull. Chem.
Soc. Jpn. 1998, 71, 2573–2581.
[39] J. C. de Mello, H. F. Wittmann, R. H. Friend, Adv. Mater.
1997, 9, 230–232.
[40] CrysAlis PRO, Oxford Diffraction Ltd., Abingdon OX14 1
RL, Oxfordshire, UK, 2007.
[41] A. J. M. Duisenberg, L. M. J. Kroon-Batenburg, A. M. M.
Schreurs, J. Appl. Crystallogr. 2003, 36, 220–229.
[25] L. N. Sun, H. J. Zhang, Q. G. Meng, F. Y. Liu, L. S. Fu, C. Y.
Peng, J. B. Yu, G. L. Zheng, S. B. Wang, J. Phys. Chem. B 2005,
109, 6174–6182.
[26] Y. Sato, M. Morimoto, H. Segawa, T. Shimidzu, J. Phys. Chem. [42] R. H. Blessing, Acta Crystallogr., Sect. A 1995, 51, 33–38.
1995, 99, 35–39.
[43] SHELXTL, release 6.1.4, Bruker AXS Inc., Madison, Wiscon-
sin 53719, USA, 2003.
[27] M. H. V. Werts, M. A. Duin, J. W. Hofstraat, J. W. Verhoeven,
Chem. Commun. 1999, 799–800.
Received: November 7, 2007
Published Online: February 1, 2008
[28] K. Gustav, U. Bartsch, Monatsh. Chem. 1991, 122, 269–274.
Eur. J. Inorg. Chem. 2008, 1523–1529
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1529