Ln(iii)-cored complexes based on boron dipyrromethene (Bodipy) ligands for NIR emission

Article abstract of DOI:10.1039/c2nj20786a

Two Bodipy-based chromophores have been synthesized, which bear two methoxyphenyl substituents in the 1,7 positions and either hydrogen (L1) or bromine (L2) atoms in the 2,6 positions, with the aim of testing their ability to sensitize the luminescence of near-infrared emitting Ln^III ions. Despite the presence of the two halogen substituents in L2, the two chromophores display very similar absorption and emission properties, the only difference being a drop in the quantum yield from 31% for L1 to 21% for L2. In addition, no triplet state emission could be evidenced for both chromophores as well as for their tris complexes with Gd^III ions. This is traced back to unfavorable conformation of the molecule, preventing efficient intersystem crossing. Complexes with stoichiometry [Ln(Li)₃(tpy)] have been isolated for Ln = Gd, Er, Yb and fully characterized. Ligand emission is partially quenched (20-40%) in the Er^III and Yb^III complexes and typical metal-centred luminescence is detected at 1530 (Er ^III) and 978 (Yb^III) nm upon excitation in the orange portion of the visible spectrum (583 nm). In these complexes, the main energy transfer pathway involves the singlet state(s) of the ligands: the lowest ¹ππ state emission overlaps the Er(⁴F_9/2 ← ⁴I_15/2) absorption band and the ¹ππ-Yb(²F_5/2) gap amounts to only 5700 cm^-1. This study demonstrates that the Bodipy framework is adequate for sensitizing the luminescence of NIR-emitting lanthanide ions, allowing excitation wavelengths extending into the orange portion of the visible spectrum and yielding complexes with large molar absorption coefficients (log ε ≈ 5.0-5.2). Additionally it also points to the importance of the conformation of the chromophores on the yield of intersystem crossing. The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012.

Full text of DOI:10.1039/c2nj20786a

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
NJC
Cite this: New J. Chem., 2012, 36, 723–731
www.rsc.org/njc
PAPER
Ln(III)-cored complexes based on boron dipyrromethene (Bodipy) ligands
for NIR emissionw
¨
Jung Ho Ryu,a Yu Kyung Eom,a Jean-Claude G. Bunzli*ab and Hwan Kyu Kim*a
Received (in Montpellier, France) 10th September 2011, Accepted 23rd November 2011
DOI: 10.1039/c2nj20786a
Two Bodipy-based chromophores have been synthesized, which bear two methoxyphenyl
substituents in the 1,7 positions and either hydrogen (L1) or bromine (L2) atoms in the 2,6
positions, with the aim of testing their ability to sensitize the luminescence of near-infrared
emitting LnIII ions. Despite the presence of the two halogen substituents in L2, the two
chromophores display very similar absorption and emission properties, the only difference being a
drop in the quantum yield from 31% for L1 to 21% for L2. In addition, no triplet state emission
could be evidenced for both chromophores as well as for their tris complexes with GdIII ions.
This is traced back to unfavorable conformation of the molecule, preventing efficient intersystem
crossing. Complexes with stoichiometry [Ln(Li)3(tpy)] have been isolated for Ln = Gd, Er, Yb
and fully characterized. Ligand emission is partially quenched (20–40%) in the ErIII and YbIII
complexes and typical metal-centred luminescence is detected at 1530 (ErIII) and 978 (YbIII) nm
upon excitation in the orange portion of the visible spectrum (583 nm). In these complexes, the
1
main energy transfer pathway involves the singlet state(s) of the ligands: the lowest pp state
4
1
emission overlaps the Er(4F9/2 ’ I15/2) absorption band and the pp–Yb(2F5/2) gap amounts to
only 5700 cmÀ1. This study demonstrates that the Bodipy framework is adequate for sensitizing
the luminescence of NIR-emitting lanthanide ions, allowing excitation wavelengths extending into
the orange portion of the visible spectrum and yielding complexes with large molar absorption
coefficients (log e E 5.0–5.2). Additionally it also points to the importance of the conformation of
the chromophores on the yield of intersystem crossing.
impede the use of existing materials in real planar waveguide
Introduction
optical amplifiers. At present, ErIII-doped silica materials are being
used widely. However, the forbidden 4fn–4fn electronic transitions
and the poor solubility of ErIII ions in conventional inorganic
media lead to yet unsatisfactory amplification properties.5
Luminescent lanthanide complexes with organic ligands are
attracting much attention because of their fundamental scientific
importance and potential utility in a wide variety of photonic
applications, such as planar waveguide amplifiers, plastic lasers,
light-emitting diodes, luminescent probes and solar cells.1–7 In
particular, the development of planar waveguides integrating
optical amplifiers with wavelength division multiplexing
(WDM) devices based on ErIII emission is recognized to be
essential for successfully realizing high-speed communication
photonic systems. However, there are several problems that still
One way to remedy this problem is to take advantage of the
large absorption coefficients of organic ligands and their
ability to sensitize lanthanide NIR luminescence.8–10 The
sensitization process is much more effective than direct f–f
excitation. Until now, a major factor in this process has often
been identified as being the triplet state(s) of the ligands. As a
consequence many research studies aiming at finding efficient
sensitizers for lanthanide NIR luminescence follow the trends
for visible-luminescence sensitization and are focused on
identifying luminescent ligands with a triplet state matching
the energies of the receiving lanthanide ion levels.11–13
a Department of Advanced Materials Chemistry and WCU Center for
Next Generation Photovoltaic Systems, Korea University, Jochiwon,
Chungnam 339-700, Republic of Korea.
E-mail: hkk777@korea.ac.kr; Fax: +82-41-860-1493;
Tel: +82-41-860-1493
Interestingly, however, it is known for a long time that
energy transfer from excited singlet state(s) to LnIII ions can
also be effective,14 particularly for fluorescent lanthanide ions,
and this mechanism starts now to be fully recognized.15–17
Other important donor states for sensitizing lanthanide lumi-
nescence include intraligand charge transfer states (ILCT)
which are given more attention presently.18–21 In our laboratory,
b
´
Ecole Polytechnique Fe´de´rale de Lausanne, Institute of Chemical
Sciences and Engineering, BCH 1402, CH-1015 Lausanne,
Switzerland. E-mail: jean-claude.bunzli@epfl.ch
w Electronic supplementary information (ESI) available: 1H-NMR
spectrum, 13C-NMR spectrum, 19F-NMR spectrum, EI-mass spectrum,
thermogravimetric analysis and photophysical data (time-resolved
spectroscopy and excitation spectra). See DOI: 10.1039/c2nj20786a
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 723–731 723

View Online
we have long been interested in deciphering energy transfer
pathways in various lanthanide complexes, demonstrating for
example (i) the role of the triplet state of metallo (ZnII, PtII)
porphyrin units in sensitizing ErIII luminescence in macrocyclic
and dendrimeric complexes,22 (ii) singlet state excitation of the
same ion through anthracene units,23,24 and (iii) the atypical
un-substituted and the dibrominated ligands L1 and L2
(Scheme 1) which form neutral tris complexes with LnIII ions
(Ln = Gd, Er, and Yb). Saturation of the coordination sphere is
conveniently achieved with terpyridine (tpy) and the photophy-
sical parameters of both ligands and their ternary complexes are
investigated and discussed.
excitation pathway through ILCT states involving dansyl (NdIII
ErIII 17 and naphthalene (EuIII 20
units. In view of the complexity
,
)
)
Experimental
in the energy transfer processes from the ligand(s) to the
lanthanide ions, which may involve several different mechanisms
and states, it is therefore important to determine which is the
most influential energy transfer path in a given system.
General
Reagents were purchased from Sigma-Aldrich, Fluka Inc., TCI
Co., or Junsei Inc. and were used as received. Tetrahydrofuran
(THF) was freshly distilled from sodium-benzophenone under
N2 atmosphere. Diethyl ether and CH2Cl2 were freshly distilled
from CaH2 under N2 atmosphere.
1H-, 13C- and 19F-NMR spectra were recorded using a
Varian Oxford 300 MHz spectrometer; chemical shifts are
reported in ppm with respect to tetramethylsilane or hexafluoro-
benzene as internal standard. Elemental analysis was performed
by the Center for microanalysis of KAIST (Daejeon) with a
EA1110-FISONS elemental analyzer manufactured by Thermo-
Quest Italia S.P.A (accuracy: 0.3%). Lanthanide content was
determined by atomic emission spectroscopy with a ELAN 6100
ICP-MS spectrometer from Perkin-Elmer-Sciex; experiments
were carried out at Seoul National University. Thermogravimetric
In this paper, we turn our attention to 4,4-difluoro-4-bora-
3a,4a-diaza-s-indacene (boron dipyrromethene or Bodipy)
chromophores which are photostable dyes used in tunable
dye lasers,24 biological labelling,26 singlet oxygen production
for photodynamic treatment (PDT) of cancer,27 and, more
recently, in heterojunction dye sensitized solar cells28 as well as
for solar hydrogen production.29 These dyes feature high
thermal and photochemical stability, large fluorescence quantum
yields and negligible triplet-state formation.25 The yield of inter-
system crossing can however be considerably enhanced upon
bromination of the 2,6 positions,29 which makes this framework
interesting for exploring the relative importance of the triplet
versus singlet state sensitization paths in luminescent lanthanide
complexes. To this end, we have designed and synthesized the
Scheme 1 Schematic diagram for the synthesis of the Bodipy ligands and their complexes.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
724 New J. Chem., 2012, 36, 723–731

View Online
1
analysis was performed on a TGA N-1000 with a heating rate
of 10 1C minÀ1 under N2 atmosphere. Infrared spectra were
measured on KBr pellets using a Jasco FT/IR-4200 Spectrometer.
MALDI-TOF mass spectrometry was performed on a Voyager-
DETM STR Biospectrometry TM workstation from PerSeptive
Biosystems. Absorption spectra were recorded with a Shimadzu
UV-2401PC spectrophotometer and photoluminescence spectra
with a Fluorolog FL-3-22 fluorometer from Horiba-Jobin-Yvon
Ltd. equipped with a 450 W Xe-lamp and two analyzing mono-
chromators. Visible emission spectra were detected with a Hama-
matsu R928 photomultiplier while near-IR emission spectra were
measured with the Peltier-cooled Hamamatsu H9170-75 photo-
multiplier system. Solutions were saturated with N2 to avoid the
quenching effect of dioxygen. Lifetime measurements were
performed with an excitation at 583 nm by using a Horiba Jobin
Yvon iHR320 Spectrometer. The excitation beam had a 100 ps
pulse width at 1 MHz repetition rate. The emission signal was
analyzed in time-correlated single photon counting mode.
Optimized geometry of the ligand have been performed by a
semi-empirical method at AM1 level with Hyperchem 7.5
professional while interpretations of the ligand absorption
spectra are based on calculations with Scigress Explorer Ultra
Version 7.7.
70% yield. H-NMR (300 MHz, CDCl3) = 2.34 (s, 3H), 3.84
(s, 3H), 6.10 (dd, J = 1.8, 3.0 Hz 1H), 6.28 (t, J = 3.4 Hz 1H),
6.81–6.86 (m, 2H), 7.09–7.17 (m, 4H), 7.25–7.26 (m, 2H) and
7.41 (dd, J = 1.8, 3.1 Hz 1H).
2-(4-Methoxyphenyl)-1H-pyrrole (3). A mixture of 0.050 g
(0.15 mmol) of (2) in 0.6 mL of EtOH and 0.2 mL of 15%
ethanolic NaOH was stirred at reflux temperature for 3 h. The
solution was concentrated in vacuo and the residue was
dissolved in CH2Cl2 and washed with H2O. The organic
layer was dried over MgSO4 and concentrated in vacuo to
1
give 0.025 g of (3) as a white solid, in 95% yield. H-NMR
(300 MHz, CDCl3) = 3.81 (s, 3H), 6.27 (dd, J = 2.8, 5.7 Hz
1H), 6.40 (t, J = 3.5 Hz, 1H), 6.79 (dd, J = 2.4, 3.9 Hz 1H),
6.88–6.91 (m, 2H), 7.36–7.39 (m, 2H) and 8.32 (br s, 1H).
Methyl-4-(bis(5-(4-methoxyphenyl)-1H-pyrrole-2-yl)methyl)
benzoate (4). A mixture of 1.82 g (10.51 mmol) of (3), 0.69 g
(4.2 mmol) of methyl 4-formylbenzoate, and 20 mL of ethyl
acetate was added to a 250 mL 3-necked round bottomed flask
and degassed with N2 atmosphere for 5 min. Trifluoroacetic
acid (0.03 mL, 0.42 mmol) was then added all at once.
The solution was stirred under N2 atmosphere at room
temperature for 1 h. Then, 50 mL of 0.1 N NaOH was added
to the solution. The solution was extracted with ethyl acetate.
The organic layer was washed with brine and water, dried over
MgSO4, filtered, and concentrated in vacuo. The crude product
was purified by recrystallization from the hexane/CH2Cl2
mixture to give a violet solid in 61% yield. 1H-NMR
(300 MHz, CDCl3) = 3.80 (s, 6H), 3.91 (s, 3H), 5.58
(s, 1H), 5.93 (s, 2H), 6.33 (s, 2H), 6.88 (d, J = 8.4 Hz, 4H),
7.32 (d, J = 8.4 Hz, 4H), 7.38 (d, J = 8.1 Hz, 2H), 8.01
(d, J = 8.4 Hz, 2H) and 8.15 (s, 2H).
Synthesis of Bodipy ligands
2-Bromo-1-tosyl-1H-pyrrole (1). 1,3-Dibromo-5,5-dimethyl-
hydantoin (6.53 g, 22.4 mmol) was added to a solution of
3.1 mL of pyrrole (44.7 mmol) in 120 mL of freshly distilled
THF in a dried flask under N2 atmosphere at À78 1C. The
mixture was stirred for 30 min and allowed to stand for an
additional 3.5 h at À78 1C. To this solution were added 4 mL
of Bu3N and 180 mL of distilled Et2O and the mixture was
stirred for 10 min at À78 1C. The resulting gray precipitate was
removed by filtration and the solution was concentrated to
remove Et2O. p-Toluenesulfonyl chloride (17.04 g, 89.4 mmol)
was then added and the inert atmosphere was reestablished.
The solution was cooled to 0 1C and 5.36 g (134.0 mmol) of
NaH (60% in mineral oil) was added. The mixture was stirred
for 18 h and 60 mL of H2O was added. The mixture was
extracted with Et2O and the ether extracts were washed with
successive portions of 1 M HCl and NaHCO3. This mixture
was combined with an equal volume of 2 M NaOH and was
stirred for 2 h at room temperature The layers were separated
and the organic extract was washed with H2O and 2 M NaOH.
The organic layer was dried over MgSO4 and concentrated
to give a light brown solid. Recrystallization from isopropyl
alcohol gave (1) in 23% yield. 1H-NMR (300 MHz, CDCl3) =
2.43 (s, 3H), 6.23–6.29 (m, 2H), 7.32 (d, J = 8.3 Hz, 2H), 7.46
(dd, J = 2.0, 3.5 Hz, 1H) and 7.81(d, J = 8.4 Hz, 2H).
(Z)-Methyl
4-((5-(4-methoxyphenyl)-1H-pyrrole-2-yl)-
(5-(4-methoxyphenyl)-2H-pyrrole-2-ylidene)methyl)benzoate (5).
2,3-Dichloro-5,6-dicyano-1,4-benzoquinone (0.4 g, 1.76 mmol)
was added to a solution of 0.6 g (1.12 mmol) of (4) in 50 mL
CH2Cl2. The mixture was stirred at room temperature for 1 h.
The solution was concentrated in vacuo. The crude product was
purified by flash column chromatography (silica, hexane/ethyl
acetate). Yield: 23%. 1H-NMR (300 MHz, CDCl3) = 3.90
(s, 6H), 3.98 (s, 3H), 6.61 (d, J = 4.2 Hz, 2H), 6.77 (d, J =
4.2 Hz, 2H), 7.03 (d, J = 9.0 Hz, 4H), 7.62 (d, J = 8.1 Hz, 2H),
7.86 (d, J = 8.4 Hz, 4H) and 8.131 (d, J = 7.8 Hz, 2H).
Boron dipyrromethene (6). A solution of 0.24 g (0.49 mmol)
of (5) and 0.12 g (1.10 mmol) of triethylamine in 30 mL of dried
CH2Cl2 was stirred under N2 atmosphere at room temperature
for 10 min. Then, boron trifluoride diethyl etherate (0.3 mL,
2.47 mmol) was added dropwise over 10 min. The resulting
solution was stirred for 24 h at room temperature and the crude
product was purified by flash column chromatography (silica,
hexane/CH2Cl2). Yield: 42%. 1H-NMR (300 MHz, CDCl3) =
3.85 (s, 6H), 3.99 (s, 3H), 6.62 (d, J = 4.2 Hz, 2H), 6.78 (d, J =
4.5 Hz, 2H), 6.96 (d, J = 9.0 Hz, 4H), 7.65 (d, J = 8.1 Hz, 2H),
7.88 (d, J = 9.0 Hz, 4H) and 8.18 (d, J = 8.4 Hz, 2H).
2-(4-Methoxyphenyl)-1-tosyl-1H-pyrrole (2). A mixture of
(1) (10 g, 33.31 mmol), 4-methoxyphenylboronic acid (6 g,
39.48 mmol), Pd(PPh3)4 (1.8 g, 1.56 mmol), 30 mL of toluene,
30 mL of EtOH, and 30 mL of 2 M Na2CO3 was refluxed at
100 1C for 24 h under N2 atmosphere and was extracted with
ethyl acetate. The organic layer was washed with brine and
water and dried over MgSO4, filtered and concentrated
in vacuo. The crude product was purified by recrystallization
from isopropyl alcohol to give (2) as a white crystalline solid in
Bodipy–CO2H (L1). A solution of ethanolic NaOH (0.85 g,
21 mmol) was added to a stirred solution of 0.2 g (0.37 mmol)
of (6) in THF/EtOH. The solution was stirred at room
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 723–731 725

View Online
temperature overnight. The solution was concentrated in vacuo.
The crude product was purified by flash column chromatography
(silica, CH2Cl2/MeOH), and then the product was acidified to pH
4 with a solution of HCl (1 M). The water layer was extracted
three times with ethyl acetate. The organic phase was dried over
MgSO4, filtered and concentrated in vacuo. The crude product
was purified by recrystallization from the hexane/CH2Cl2 mixture
to give a green solid. Bodipy–CO2H: Yield: 60%. Anal. calcd. for
C30H23BF2N2O4: C (68.72%), H (4.42%), N (5.34%); found:
C (68.41%), H (4.46%), N (5.10%); 1H-NMR (300 MHz,
CDCl3) = 3.86 (s, 6H), 6.64 (d, J = 3.6 Hz, 2H), 6.77 (d, J =
3.5 Hz, 2H), 6.95 (d, J = 9.0 Hz, 4H), 7.70 (d, J = 8.1 Hz, 2H),
7.89 (d, J = 9.0 Hz, 4H) and 8.26 (d, J = 7.8 Hz, 2H); 13C-NMR
(226.4 MHz, DMSO) = 166.82, 160.5, 155.8, 142.5, 136.4, 133.4,
131.8, 131.7, 130.9, 129.5, 121.7, 114.1, 113.4, 109.6, 55.2; 19F-
NMR (282.3 MHz, DMSO) = À137.5 (s, 2F), MS (MALDI-
TOF): m/z 524.06 (M+), calc. 524.17.
H (3.94%), N (6.40%), Er (8.49%); found: C (63.30%),
H (3.72%), N (6.53%), Er (8.53%).
[Yb(L1)3(tpy)]. Yield: 69%. FT-IR (KBr) [cmÀ1]: 2975, 1604,
1558, 1469, 1428, 1263, 1181, 1142, 1066, 835, 797, 736, and 626.
Anal. calcd. for C105H77B3YbF6N9O12: C (63.81%), H (3.93%),
N (6.38%), Yb (8.76%); found: C (64.20%), H (4.02%),
N (6.58%), Yb (8.60%).
[Er(L2)3(tpy)]. Yield 38%. FT-IR (KBr) [cmÀ1]: 2928, 2372,
1667, 1555, 1459, 1427, 1293, 1256, 1225, 1179, 1144, 1067,
833, 734, and 628. Anal. calcd. for C105H71B3Br6ErF6N9O12:
C (51.60%), H (2.93%), N (5.16%), Er (6.84%); found: C
(51.01%), H (2.97%), N (4.98%), Er (6.92%).
[Yb(L2)3(tpy)]. Yield 65%. FT-IR (KBr) [cmÀ1]: 2926, 2375,
1606, 1554, 1459, 1427, 1293, 1256, 1224, 1179, 1144, 1067, 990,
907, 832, 744, 693, and 627. Anal. calcd. for C105H71B3Br6YbF6-
N9O12: C (51.48%), H (2.92%), N (5.15%), Yb (7.06%); found:
C (52.10%), H (2.85%), N (5.25%), Yb (7.12%).
Br-Bodipy-CO2H (L2). A mixture of L1 (0.1 g, 0.19 mmol),
NBS (0.075 g, 0.42 mmol), AIBN (0.004 g, 0.02 mmol) and
20 mL of CH3Cl solution was refluxed at 100 1C for 24 h under
N2 atmosphere. The crude product was purified by flash
column chromatography (silica, CH2Cl2/MeOH), and then
the product was acidified to pH 4 with a solution of
HCl (1 M). The water layer was extracted with ethyl acetate
three times. The organic phase was dried over MgSO4, filtered
and concentrated in vacuo. The crude product was purified by
recrystallization from the hexane/CH2Cl2 mixture to give a
green solid. Br–Bodipy–CO2H: Yield: 61%. Anal. calcd. for
C30H21BBr2F2N2O4: C (52.82%), H (3.10%), N (4.11%); found:
C (52.62%), H (3.50%), N (4.00%). 1H-NMR (300 MHz,
CDCl3) = 3.85 (s, 6H), 6.92 (m, 6H), 7.62 (d, J = 9.0 Hz,
4H), 7.68 (d, J = 8.1 Hz, 2H) and 8.27 (d, J = 7.8 Hz, 2H);
13C-NMR (226.4 MHz, DMSO) 166.8, 160.6, 157.9, 140.8, 137.7,
135.4, 132.3, 131.0, 130.8, 130.7, 129.3, 124.3, 121.4, 113.9, 55.4;
19F-NMR (282.3 MHz, DMSO) = À139.5. MS (MALDI-TOF):
m/z = 681.87 (M+), calc. 681.99.
[Gd(L2)3(tpy)]. This complex was prepared in the same
manner above and a dark blue solid was obtained with a yield
of 42%. FT-IR (KBr) [cmÀ1]: 2961, 2370, 1606, 1554, 1459,
1426, 1292, 1257, 1224, 1178, 1144, 1067, 831, 744, and 627.
Anal. calcd. for C105H71B3Br6GdF6N9O12: C (51.82%), H
(2.94%), N (5.18%), Gd (6.46%); found: C (51.40%), H
(3.01%), N (5.29%), Gd (6.53%).
Results and discussion
Synthesis and characterization
The synthesis of the ligands and of the targeted complexes
[Ln(L1)3(tpy)] and [Ln(L2)3(tpy)] (Ln = Gd, Er, Yb) is depicted
in Scheme 1. Intermediate compound (1) was prepared by
treatment of pyrrole with 1,3-dibromo-5,5-dimethylhydantoin
in the presence of AIBN followed by amine protection with
p-toluenesulfonyl chloride. Reaction of (1) with 4-methoxyphenyl-
boronic acid under conditions of the Suzuki cross-coupling
reaction yielded (2). After protection of the tosyl group with
15% ethanolic NaOH in ethanol,30 the resulting intermediate (3)
was reacted with methyl 4-formylbenzoate to obtain (4), the
oxidation of which was achieved by 2,3-dichloro-5,6-dicyano-
1,4-benzoquinone (DDQ), a stronger reactant than 1,4-benzo-
quinone. The reaction proceeded smoothly but separating the
highly colored compound by column chromatography was
difficult, which explains the somewhat low yield of this step.
The oxidized compound (5) was then treated with boron
trifluoride etherate in the presence of triethylamine in dried
CH2Cl2 at room temperature to produce the precursor Bodipy
(6). The ester group was hydrolyzed in the presence of sodium
hydroxide to yield ligand L1. Further bromination at the 2 and 6
positions led to the second ligand L2. The ligands were char-
acterized by elemental analysis, FT-IR, 1H- 13C-, and 19F-NMR
spectra (Fig. S1–S6 in ESIw) MALIDO-TOF mass spectrometry
(Fig. S7 and S8 in ESIw), absorption, and emission spectro-
scopies. Thermogravimetric analyses show the ligands being
devoid of solvation water molecules (Fig. S9 in ESIw).
General procedure for the synthesis of [Ln(Li)(terpy)] (Ln =
Gd, Er, Yb; i = 1, 2). A mixture of Li (i = 1 or 2; 3 equiv.) and
KH (3.3 equiv.) was stirred overnight in freshly distilled THF
at room temperature until no more H2 gas was generated.
After the completion of the reaction, a methanol solution of
anhydrous LnCl3 (Ln = Gd, Er, or Yb; 1 equiv.) and
terpyridine (tpy, 1.1 equiv.) was added to the reaction mixture,
and the resulting solution was stirred for 2 days. Subsequently,
the resulting solution was filtered, the solvents were removed
and the obtained colored solid was washed sequentially with
methanol, yielding a dark blue solid.
[Gd(L1)3(tpy)]. Yield: 70%. FT-IR (KBr) [cmÀ1]: 2932, 1603,
1559, 1469, 1428, 1263, 1217, 1181, 1142.62, 1067, 838, 796, 737,
625, and 546. Anal. calcd. for C105H77B3GdF6N9O12: C
(64.33%), H (3.96%), N (6.43%), Gd (8.02%); found: C
(64.93%), H (3.92%), N (6.70%), Gd (8.10%).
[Er(L1)3(tpy)]. Yield: 25%. FT-IR (KBr) [cmÀ1]: 2927, 1604,
1560, 1469, 1431, 1385, 1262, 1218, 1182, 1144, 1067, 834, 785, 732,
624, and 541. Anal. calcd. for C105H77B3ErF6N9O12: C (64.00%),
Novel lanthanide-cored complexes with Bodipy ligands
were synthesized by a ligand exchange reaction developed in
c
726 New J. Chem., 2012, 36, 723–731
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012

View Online
Table 1 Optical properties of the ligands and their complexes in THF
at room temperature
e/MÀ1 cmÀ1
lem/nm
fF
Iint
a
b
Compound
labs/nm
L1
583
400
583
583
583
584
409
584
584
584
52 400 Æ 3500
11 450 Æ 800
622
0.31
1.00
L1Gd
L1Er
L1Yb
L2
110 300 Æ 3270
121 200 Æ 4650
122 400 Æ 1200
56 100 Æ 2100
13 200 Æ 500
620
621
620
624
0.98
0.74
0.70
1.00
0.21
L2Gd
L2Er
L2Yb
159 800 Æ 5800
166 600 Æ 8400
147 400 Æ 4200
626
625
625
0.92
0.63
0.76
a
Measured with respect to chlorophyll A (fF = 0.32 in dimethyl-
ether);36 estimated accuracy: Æ10%. Relative integrated intensity;
b
estimated error: Æ5%.
spectrum of ligand L1 in THF (Fig. 2) exhibits an intense S0 -
S1 (p–p*) transition (log e = 4.72) centered at lmax = 583 nm
and mainly originating from the central core of the ligand. An
additional, much weaker band (log e = 4.06) is seen at 400 nm; it
contains a contribution from an intra-ligand charge transfer
(ILCT) between the methoxyphenyl groups and the central
aromatic system of the ligand. A third band located at about
310 nm, at the limit of the cutoff wavelength of the solvent,
corresponds to transitions on the phenyl moieties.34 Similar
absorption bands are observed for ligand L2. In line with the
bromination of the methoxyphenyl moieties, the main absorption
remains almost unshifted, while the second one sustains a
bathochromic shift of 9 nm; both transitions display negligible
change in molar absorption coefficients with respect to L1.
Finally, the more energetic absorption is not affected with respect
to its energy, but its intensity is smaller.
Upon excitation at the absorption maximum, the photo-
luminescence spectrum of ligand L1 in THF (Fig. 2) displays a
strong band with maximum at 622 nm and a shoulder around
650–660 nm. A very similar spectrum is seen for the bromi-
nated ligand L2, the only differences being a small red shift to
624 nm and a decrease in intensity. The fluorescence quantum
yields of the ligands were determined in THF by comparison
with chlorophyll A in ether solution. They amount to 31 and
21%, respectively, that is FF decreases by about one third
Fig. 1 FT-IR spectra of Bodipy ligands and the corresponding
[Ln(Li)3(tpy)] complexes. The arrow points to the CQO stretching
vibration of the carboxylic acid function.
our laboratory, starting from anhydrous LnCl3 (Ln = Gd, Er,
Yb), and were obtained in 25B70% yields. The FT-IR spectra
of the free ligands exhibit a band at around 1690–1695 cmÀ1
assigned to the carbonyl stretching of conjugated carboxylic
acid. All of the vibrational spectra of the Ln-cored complexes
are similar and are devoid of this vibration. Instead, the
two characteristic bands of asymmetric (ca. 1560 cmÀ1) and
symmetric (ca. 1430 cmÀ1) stretching vibration modes of the
carboxylate group appear, the latter being mixed with C–C
stretching,31 which explains the presence of a similar, although
less intense band in the spectrum of the ligand (Fig. 1 and
Fig. S10 in ESIw). These data are indicative of the conversion
of the carboxylic acid group into a carboxylate anion as a
result of the formation of the Ln-cored complexes.32,33 The
proposed composition of the [Ln(Li)3(tpy)] complexes is
further substantiated by elemental analysis including metal
content determination.
Ligand-centred photophysical properties
Fig. 2 Absorption (left) and photoluminescence (right; lexc =
583 nm) spectra of the Bodipy ligands with the concentration of
1.0 Â 10À5 M in THF.
The photophysical parameters for the electronic spectra of the
ligands and their complexes are listed in Table 1. The absorption
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 723–731 727

View Online
is detected around 1.25 mm. At liquid nitrogen temperature,
the corresponding data are similar, 621, 683 (L1) and 624, 687
(L2) nm, but the low energy component is much more intense.
When a time delay is enforced the signals disappear completely
at both temperatures (Fig. S11 in ESIw). At room temperature,
decays of the fluorescence maxima can be fitted with single
exponential functions, yielding lifetimes of 1.0 and 0.6 ns for
L1 and L2, respectively. At 77 K, lifetimes associated with the
decays monitored at 626 and 711 nm are in the range 5–6 ns
for L1 and 3.9–4.3 ns for L2, comparable to the lifetimes
reported for L4-H (2.9 ns) and L4-Br (1.3 ns).29 Therefore
both components are fluorescence bands, the second one being
probably a vibronic component since the energy difference
between the two maxima is 1240–1300 cmÀ1 at room temperature
and 1400–1450 cmÀ1 at 77 K, corresponding to a ring-breathing
vibrational mode.
A similar situation is met for the GdIII complexes but the
emission spectra are broadened rendering Gaussian decom-
position difficult (Fig. 3, bottom). The main difference is a very
large enhancement of the low energy component at 77 K, red
shifted at 1300–1550 cmÀ1 with respect to the high energy
component, as well as the appearance of a weak shoulder at
around 750 nm. However, enforcing a time delay results again
in a total loss of the signal (Fig. S12 in ESIw). The lumines-
cence decays of both components are bi-exponential at room
temperature with a short component (B0.6–1 ns) and a longer
one (2.4–3.1 ns) corresponding to a population of B10%. At
77 K, the decays become monoexponential with a lifetime of
5–6 ns (GdL1) and 3.7–4.0 ns (GdL2). Finally, closer scrutiny
of all spectra in the range 700–850 nm did not reveal any
Scheme 2 Halogenated Bodipy compounds from the literature:
L3-X27 and L4-X.29
upon bromination of L1. Preparation of halogenated Bodipy
derivatives L3-X and L4-X (Scheme 2, X = H, Br, I) has been
previously reported.27,29 In both cases, halogenation of the 2,6
positions led to a red-shift of the fluorescence band up to
30–40 nm, reduction in fluorescence lifetime, and decrease in
the fluorescence quantum yield. This has been explained by an
increase in the isc rate constant attributable to the heavy atom
effect. For L3-X ligands (X = H, Br), which were designed for
PDT of cancer, the consequence is an increase in the yield of
1O2,27 for L4-X compounds (X = H, Br, I), very weak bands
at around 800 nm were detected at room temperature, with
lifetimes in the ms range, assigned to triplet states.29
To determine whether the bromo-substituted ligand L2
displays phosphorescence or not, we have measured the
emission of both ligands in N2-saturated 2-methyl-THF at
room temperature and 77 K under fluorescence (Fig. 3, top)
and phosphorescence (Fig. S11 in ESIw) conditions. An
intense band is seen at room temperature for each ligand.
Gaussian decomposition reveals main maxima at 623 (L1) and
627 (L2) nm and additional components at 678 (L1) and 680
(L2) nm, respectively. When solutions of the ligands are
saturated with di-oxygen, no luminescence from singlet oxygen
Fig. 3 Fluorescence spectra of the ligands (top) and GdIII complexes (bottom) in 2-methyl THF at room temperature (left) and 77 K (right).
c
728 New J. Chem., 2012, 36, 723–731
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012

View Online
unbound ligands (Fig. 5 and Table 1). Upon photo-excitation
at the wavelength corresponding to the absorption maximum
of the ligands, the intensity of the ligand emission remains
unchanged for GdL1 (within experimental errors) and is
slightly smaller for GdL2 (À8%) while it is noticeably reduced
for the ErIII and YbIII complexes. The decrease is the same
(BÀ30%) for the complexes with L1 while being differen-
tiated for the complexes with L2: À40% for ErIII and À20%
for YbIII. Lifetime measurements in m-THF confirm data
reported above for GdLi: decays are biexponential at room
temperature with the longest component (2.4–3.5 ns) accounting
for about 5–15% of the population while they become mono-
exponential at 77 K with lifetime comparable to those of the
GdIII complexes. In parallel, typical, although weak, metal-
centered emission of the ErIII and YbIII ions is observed with
maxima about 1530 and 978 nm, respectively (Fig. 6). The
emission band of ErIII is ascribed to the 4f–4f electronic transi-
tion originating from the first excited state (4I13/2) to the ground
state (4I15/2) and the emission of the YbIII complexes to the
Fig. 4 Snapshots of the calculated structures of L1 (left) and L2
(right) at the AM1 level.
genuine emission band that could be assigned to triplet state
emission, even for the GdIII complex with brominated ligand
L2, pointing to no or small heavy atom effect and/or fast
de-activation of the triplet state. We note, however, that triplet
state emission can be very weak in Bodipy ligands29 and that it
occurs in a spectral range in which our detector is not very
sensitive. As a comparison, very weak triplet state emission
from Bodipy has been evidenced at 77 K around 730 nm in
an IrIII complex with an ethynyl derivatized Bodipy. The
associated lifetime of the 3IL state is 25 ms.35 In the latter
compound, the Bodipy moiety adopts a planar conformation,
which is not the case for ligands L1 and L2 (Fig. 4).
2
2F5/2 - F7/2 transition. Excitation spectra clearly match the
absorption spectra (Fig. S13 in ESIw). All these data point to
energy transfer operating between the ligands and the NIR-
emitting LnIII ions. From data reported in Table 1 one can
estimate the fraction (1 À ILn/IGd) of energy transferred onto the
Metal-centered NIR luminescence
The LnIII-cored Bodipy complexes, LnLi, display UV-Vis
absorption and ligand emission behaviors similar to the
Fig. 5 Absorption (up) and emission (bottom) spectra of Li (i = 1, 2)
and LnLi complexes in THF at room temperature; measured at
constant absorbance, [Li] E 3.0 Â 10À6 M and [LnLi] E 1.0 Â 10À6 M.
Fig. 6 Metal-centred NIR emission of ErLi (up) and YbLi (bottom)
with the concentration of 1 Â 10À5 M in THF at room temperature;
lex = 583 nm.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 723–731 729

View Online
Table 2 Emission characteristics of the LnLi complexes measured at
constant absorbance in THF at room temperature
the former one and the p-extended dipyrromethene unit for two
latter ones Thus, phosphorescence is not detected because the
restricted rotation between the phenyl and dipyrromethene
units may prevent heavy atom effect (Fig. 4). Therefore we
postulate that the energy-transfer pathway for the sensitization
of the NIR-emitting LnLi complexes mainly involves singlet
excited states. It is not clear though why L2 leads to sensitizaton
efficiencies which are much less good (À40 to À100%) than
those induced by L1 whereas its energy levels are very compar-
able and quantum yield is only B30% smaller. A detailed study
for elucidating this trend would involve fast time-resolved
spectroscopy to which we have presently no access.
Compound
l
em/nm
Iint
1237
565
54 335
39 110
Iint(rel)
ErL1
ErL2
YbL1
YbL2
1530
1530
978
1.0
0.46
1.0
978
0.72
metal ion as being 25–29% for complexes with L1, 32% for
ErL2, but only 17% for YbL2. In reality, both complexes with
L2 are less luminescent than those with L1. Comparing solutions
with the same complex concentration (1 Â 10À5 M) shows that
the NIR emission intensity of ErL1 is enhanced about 2-fold
compared to the fluorescence of ErL2 while the enhancement is
a factor 1.4 for YbL1 with respect to YbL2 (Table 2).
Conclusions
The efficiency of energy transfer between ligands and metal
ions is primarily controlled by two parameters: (i) the position
(distance and orientation) of the energy-donor chromophores
with respect to the acceptor ion and (ii) the spectral overlap
between the emission spectrum of the donor and the absorp-
tion spectrum of the acceptor. This overlap is shown in Fig. 7
for L1 and ErIII: the emission band of the ligand occurs at the
same energy as the 4F9/2 ’ 4I15/2 transition so that sensitization
may be expected even if the overlap integral is small. Possible
energy-transfer pathways for the sensitized emission in lumi-
nescent lanthanide complexes involve different ligand and
metal-ion states. A common path is energy transfer through
the triplet state of the photoexcited luminescent ligand after
suitable conversion from the excited singlet state via intersystem
crossing. This path is usually reinforced when a heavy-atom
effect is operating. In our case, we have been unable to locate
the triplet state of the ligands even in the GdIII complexes
despite the presence of two bromine substituents on L2. We
trace back the absence of phosphorescence to the dihedral angle
between the phenyl groups and the dipyrromethene unit which
amounts to B551, leading to less conjugation between the
phenyl moiety and the p-extended dipyrromethene unit. It is
worth pointing out that the calculated dihedral angle of the
ligands is consistent with the absorption spectra which exhibit
three maxima in THF at 300–310 nm, 400 nm and 583–584 nm,
corresponding to transitions located on the phenyl groups for
Although initially designed for testing the singlet versus triplet
state pathway for the sensitization of NIR-emitting LnIII ions,
ligands L1 and L2 proved to behave similarly in that inter-
system crossing is negligible for both of them as well as for the
corresponding LnIII complexes. Therefore sensitization of the
ErIII- and YbIII-cored complexes is postulated to be achieved
through the lowest 1pp state which is in resonance with the
4F9/2 level of the former ion while the energy difference with
the Yb(2F5/2) excited level amounts to 5700 cmÀ1, a not too
large gap for energy transfer. The experimental evidences
reported here point to the Bodipy framework being adequate
for transferring energy onto NIR-emitting LnIII ions, with
excitation in the orange region of the visible spectrum; in
addition the complexes have high molar absorption coeffi-
cients (log e E 5.0–5.2). While remaining modest, the amount
of energy transferred from the ligands (20–40%) is sizeable;
the weak metal-centered luminescence observed at room tempera-
ture is mainly due to deactivation through high-energy vibrations,
which explains why ErIII complexes are less luminescent than their
YbIII counterparts, as observed for complexes with a 8-hydroxy-
quinoline-derivatized Bodipy ligand.37
For a more efficient sensitization, the present framework
will have to be modified in order to render it more planar so
that intersystem crossing can be induced by the heavy atom
effect; moreover, non-radiative de-activations will have to be
minimized by halogenation. Work is in progress in our laboratory
toward these goals.
Acknowledgements
This research was supported by a grant from the Fundamental
R&D Program for Core Technology of Materials funded by
the Ministry of Knowledge Economy, WCU Center for Next
Generation Photovoltaic System (The Ministry of Education
and Science, grant R31-2008-000-10035-0), and by a grant
from Korea University (2010).
References
1 L. D. Carlos, R. A. S. Ferreira, V. de Zea Bermudez, B. Julian-
Lopez and P. Escribano, Chem. Soc. Rev., 2011, 40, 536.
2 S. V. Eliseeva and J.-C. G. Bunzli, New J. Chem., 2011, 35, 1165.
¨
3 J.-C. G. Bunzli and S. V. Eliseeva, J. Rare Earths, 2010, 28, 824.
´
and L. D. Carlos, Opt. Mater.
¨
4 R. A. S. Ferreira, P. S. Andre
Fig. 7 Spectral overlap between the absorption spectrum of ErCl3 in
THF and the emission spectrum of L1 in THF.
(Amsterdam), 2010, 32, 1397.
c
730 New J. Chem., 2012, 36, 723–731
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012

View Online
5 H. K. Kim, J. B. Oh, N. S. Baek, S. G. Roh, M. K. Nah and
Y. H. Kim, Bull. Korean Chem. Soc., 2005, 26, 201.
22 J. B. Oh, Y. H. Kim, M. K. Nah and H. K. Kim, J. Lumin., 2005,
111, 255.
6 H. K. Kim, S. G. Roh, K. S. Hong, J. W. Ka, N. S. Baek, J. B. Oh,
M. K. Nah, Y. H. Cha and J. Ko, Macromol. Res., 2003, 11, 133.
7 J. H. Oh, H. M. Song, Y. K. Eom, J. H. Ryu, M. J. Ju and
H. K. Kim, Bull. Korean Chem. Soc., 2011, 32, 2743.
8 M. P. O. Wolbers, F. C. J. M. Van Veggel, F. G. A. Peters, E. S.
E. Van Beelen, J. W. Hofstraat, F. A. J. Geurts and
D. N. Reinhoudt, Chem.–Eur. J., 1998, 4, 772.
23 N. S. Baek, Y. H. Kim, S. G. Roh, B. K. Kwak and H. K. Kim,
Adv. Funct. Mater., 2006, 16, 1873.
24 M. K. Nah, H. G. Cho, H. J. Kwon, Y. J. Kim, C. Park,
H. K. Kim and J. G. Kang, J. Phys. Chem. A, 2006, 110, 10371.
25 G. Ulrich, R. Ziessel and A. Harriman, Angew. Chem., Int. Ed.,
2008, 47, 1184.
26 S. L. Niu, C. Massif, G. Ulrich, R. Ziessel, P. Y. Renard and
A. Romieu, Org. Biomol. Chem., 2011, 9, 66.
27 T. Yogo, Y. Urano, Y. Ishitsuka, F. Maniwa and T. Nagano,
J. Am. Chem. Soc., 2005, 127, 12162.
9 J. Kido and Y. Okamoto, Chem. Rev., 2002, 102, 2357.
10 S. V. Eliseeva and J.-C. G. Bunzli, Chem. Soc. Rev., 2010, 39, 189.
¨
11 E. S. Andreiadis, R. Demadrille, D. Imbert, J. Pecaut and
M. Mazzanti, Chem.–Eur. J., 2009, 15, 9458.
12 J. E. Jones and S. J. A. Pope, Dalton Trans., 2009, 8421.
28 T. Rousseau, A. Cravino, T. Bura, G. Ulrich, R. Ziessel and
J. Roncali, Chem. Commun., 2009, 1673.
13 N. M. Shavaleev, R. Scopelliti, F. Gumy and J.-C. G. Bunzli,
¨
Inorg. Chem., 2009, 48, 6178.
14 M. Kleinerman, J. Chem. Phys., 1969, 51, 2370.
15 A. P. Souza, L. C. V. Rodrigues, H. F. Brito, S. Alves and
O. L. Malta, J. Lumin., 2010, 130, 181.
16 M. H. Ha-Thi, J. A. Delaire, V. Michelet and I. Leray, J. Phys.
Chem. A, 2009, 114, 3264.
17 S. G. Roh, N. S. Baek, Y. H. Kim and H. K. Kim, Bull. Korean
Chem. Soc., 2007, 28, 1249.
29 R. P. Sabatini, T. M. McCormick, T. Lazarides, K. C. Wilson,
R. Eisenberg and D. W. McCamant, J. Phys. Chem. Lett., 2011,
2, 223.
30 L. W. Knight, J. W. Huffman and M. L. Isherwood, Synlett, 2003,
1993.
31 E. Spinner, J. Chem. Soc. B, 1967, 874.
32 S.-G. Roh, N. S. Baek, K.-S. Hong and H. K. Kim, Bull. Korean
Chem. Soc., 2004, 25, 343.
33 A. Dupraz, P. Guy and C. Dupuy, Tetrahedron Lett., 1996,
37, 1237.
18 V. F. Zolin, L. N. Puntus, V. I. Tsaryuk, V. A. Kudryashova,
J. Legendziewicz, P. Gawryszewska and R. Szostak, J. Alloys
Compd., 2004, 380, 279.
34 J. Karolin, L. B. A. Johansson, L. Strandberg and T. Ny, J. Am.
Chem. Soc., 1994, 116, 7801.
19 A. Vogler and H. Kunkely, Inorg. Chim. Acta, 2006, 359, 4130.
20 Y. H. Kim, N. S. Baek and H. K. Kim, ChemPhysChem, 2006,
7, 213.
21 N. M. Shavaleev, R. Scopelliti, F. Gumy and J.-C. G. Bunzli,
¨
Inorg. Chem., 2009, 48, 7937.
35 A. A. Rachford, R. Ziessel, T. Bura, P. Retailleau and
F. N. Castellano, Inorg. Chem., 2010, 49, 3730.
36 G. Weber and F. W. J. Teale, Trans. Faraday Soc., 1957, 53, 646.
37 R. Ziessel, G. Ulrich, L. J. Charbonniere, D. Imbert, R. Scopelliti
and J.-C. G. Bunzli, Chem.–Eur. J., 2006, 12, 5060.
¨
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 723–731 731