Ultrabright Lanthanide Nanoparticles

Article abstract of DOI:10.1002/cplu.201600007

Tb-doped La_0.9Tb_0.1F₃ nanoparticles were prepared by a simple and reproducible microwave-assisted synthetic protocol in water. The nanoparticles were characterized by XRD, TEM, dynamic light scattering and inductively coup

Full text of DOI:10.1002/cplu.201600007

DOI: 10.1002/cplu.201600007
Full Papers
Ultrabright Lanthanide Nanoparticles
Joan Goetz,[a] Aline Nonat,[a] Abdoulaye Diallo,[a] Mohamadou Sy,[a] Ildan Sera,[a]
Alexandre Lecointre,[a] Christophe Lefevre,[b] Chi Fai Chan,[c] Ka-Leung Wong,*[c] and
[a]
¨
Loıc J. Charbonni›re*
Tb-doped La0.9Tb0.1F3 nanoparticles were prepared by a simple
and reproducible microwave-assisted synthetic protocol in
water. The nanoparticles were characterized by XRD, TEM, dy-
namic light scattering and inductively coupled plasma atomic
emission spectroscopy elemental analysis. Eleven ligands with
varying coordination and photosensitizing abilities were de-
signed to bind at the surface of the Tb-doped nanoparticles.
The photosensitizing behavior was monitored by electronic ab-
sorption spectroscopy and steady-state and time-resolved
emission spectroscopy. The two most effective photosensitiz-
ing ligands were used to isolate and purify the capped nano-
particles. The composition and spectroscopic properties of
these nanoparticles were measured, which revealed either
2660 and 5240 ligands per nanoparticle, molar absorptivities of
7.6”106 and 1.6”107 mÀ1 cmÀ1 and luminescence quantum
yields of 0.29 and 0.13 in water, respectively. These data corre-
spond to exceptional brightness values of 2.2”106 and 2.1”
106 mÀ1 cmÀ1, respectively. The as-prepared nanoparticles were
imaged in HeLa cells by fluorescence microscopy, which
showed their specific localization in lysosomes.
Introduction
The extreme sensitivity and simplicity of analytical techniques
based on luminescence mean they are currently the most
widespread analytical tools in use. Although the phenomena
of phosphorescence and fluorescence were described as early
as 1565, the major advances towards their understanding are
only a century old,[1] and since the pioneering works on qui-
nine sulfate and fluorite,[1] the pool of luminescent compounds
has dramatically expanded. These range from small organic
and inorganic molecules to natural or genetically modified pro-
teins,[2] semiconducting nanocrystals (quantum dots),[3] and or-
ganic or hybrid luminescent nanoparticles (NPs).[4] Whatever
the source of the luminescence, one of the basic parameters
to quantify its efficacy is its brightness B, which is the product
of the molar absorption coefficient e [mÀ1 cmÀ1] and the lumi-
nescence quantum yield F. Although the brightness of small
inorganic and organic fluorescent compounds ranges from
around 103 to 105 mÀ1 cmÀ1,[5] it can reach values of up to
106 mÀ1 cmÀ1 or more in the case of fluorescent proteins such
as phycoerythrin[6] or for semiconducting luminescent NPs
(quantum dots),[7] and recent studies have demonstrated simi-
lar or slightly higher values for semiconducting polymer NPs.[8]
Unlike such species, lanthanide-based luminophores can be
less effective because of the low absorptivities of their La-
porte-forbidden f–f transitions.[9] Nevertheless, luminescence
from lanthanide ions offers unique properties such as a large
(pseudo) Stokes shift, elemental spectral signatures with
narrow emission bands in the visible and near-infrared (NIR) re-
gions,[10] and extremely long excited-state lifetimes.[9] Due to
indirect photosensitization through chelating ligands,[11]
termed the “antenna effect”, lanthanide-based labels can have
brightness values of the order of 104 mÀ1 cmÀ1 at their high-
est,[12] but the association of long lifetimes and line-like emis-
sion bands offers a broad range of opportunities for ultrasensi-
tive multiplexed analysis in bioassays[13] and time-resolved fluo-
rescence microscopy.[14] Some lanthanide-based NPs,[15] and
more recently molecular systems,[16] also offer an unrivalled op-
portunity with upconverting properties, giving a nearly back-
ground-free emission at higher energy than the excitation
light. However, whatever the targeted application, lanthanide-
based NPs still suffer from low absorption, and there is a great
interest in improving it, either for downshifting applications
(e.g., diagnosis,[17] solar cells[18]) or for upconversion.[19] Coordi-
nating photosensitizing ligands to lanthanide-based NPs is an
attractive way to boost the brightness of lanthanide NPs. First
described in organic solvents,[20] and soon after in aqueous
media,[21] the method has received great interest in recent
years.[22]
[a] J. Goetz, Dr. A. Nonat, A. Diallo, M. Sy, I. Sera, A. Lecointre,
Dr. L. J. Charbonni›re
Laboratoire d’IngØnierie MolØculaire AppliquØe à l’Analyse
Institut Pluridisciplinaire Hubert Curien
École EuropØenne de Chimie, Polym›res et MatØriaux
IPHC, UMR 7178, CNRS/UniversitØ de Strasbourg
25 rue Becquerel, 67087 Strasbourg Cedex (France)
E-mail: l.charbonn@unistra.fr
[b] Dr. C. Lefevre
Institut de Physique et Chimie des MatØriaux de Strasbourg
(UMR 7504 CNRS), and Laboratory of Nanostructures in Interactions with
Their Environment (NIE), UniversitØ de Strasbourg
23 rue du Loess, BP 43, 67034 Strasbourg Cedex 2 (France)
[c] C. F. Chan, Dr. K.-L. Wong
Department of Chemistry
Hong Kong Baptist University
Hong Kong SAR (Hong Kong)
E-mail: klwong@hkbu.edu.hk
Supporting information for this article can be found under http://
dx.doi.org/10.1002/cplu.201600007.
ChemPlusChem 2016, 81, 526 – 534
526
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Herein, we present a systematic study of a family of surface-
coordinating ligands with a complete study of the ability of
the ligands for surface coordination and its impact on photo-
sensitization and on the water stability of the as-prepared NPs.
We show that such NPs can be highly stable even in the pres-
ence of large excess of competing anions, they present ex-
tremely large absorption cross-sections and high brightness,
and can be used for luminescence microscopy in living cells.
where x is the Tb content within the structure, and a and c are
the cell parameters. The use of this equation with refined cell
parameters gives a calculated doping rate of 11.7(3)%, which is
slightly higher than the 10.0(5)% introduced during the syn-
thesis. Analysis of the crystallite size was performed using the
broadening of the different peaks, which were corrected of the
instrumental contribution, giving an average size of 21Æ1 nm
for the core of the NPs, in good agreement with TEM observa-
tions. In ultrapure water, the granulometry measurements
showed a narrow distribution with an average hydrodynamic
radius of the NPs centered on 35 nm, which indicates a large
hydration shell. The surface potential of the NPs was deter-
mined to be +34.9 mV, indicative of moderate to good colloi-
dal stability in water.
Results and Discussion
Synthesis of the nanoparticles
La0.9Tb0.1F3 NPs were prepared by adaptation of reported pro-
tocols[23] using a microwave oven in place of conventional
heating. They were characterized by TEM, XRD in the solid
state and dynamic light scattering (DLS) in ultrapure water
(Figure 1). TEM images revealed the presence of NPs with an
average diameter of approximately 20–25 nm, with the small-
est particles having a clearly elongated structure. The X-ray dif-
fractogram of the powdered sample matches that of hexago-
nal LaF3 crystals with respect to peak positions and their inten-
sities (JCPDS standard card 32-0483). Profile matching applied
to the XRD pattern showed cell parameters of the hexagonal
lattice (a=b=7.145(1) Š, c=7.305(1) Š), which are between
those of pure LaF3 (a=b=7.178 Š, c=7.351 Š, JCPDS #00-032-
0483) and that of pure TbF3 (a=b=6.856, c=7.026, JCPDS
#04-006-9969). This result is ascribed to the Tb doping. Assum-
ing Vegard’s law applies, the variation in cell volume V would
be given by Equation (1):
Finally, the elemental composition of the NPs was deter-
mined by inductively coupled plasma atomic emission spec-
troscopy (ICP–AES) analysis, resulting in a doping percentage
of 9.7% of Tb, in excellent agreement with the feeding ratio of
10%.
Synthesis of the photosensitizing capping ligands
Scheme 1 summarizes the structures of the ligands that were
tested for surface photofunctionalization and the synthetic
pathways for the synthesis of the ligands. All ligands are de-
rived from two of the basic coordinating units of lanthanide
coordination chemistry—dipicolinic acid (L1), and 2-hydroxyi-
sophthalic acid (L5). Both units have been shown to be excel-
lent photosensitizing units for Tb,[24,25] but they display distinct
coordination behaviors. Dipicolinic acid acts as a tridentate
chelator, which in complex forms two five-membered chelate
rings,[26] whereas 2-hydroxyisophthalic acid essentially acts in
a bidentate mode, forming a six-membered ring,[27] although
one can easily imagine a m-bridging bis-bidentate mode, form-
ing two six-membered rings with two lanthanide atoms, as ob-
served for different Schiff-base complexes obtained from the
parent 2-hydroxy-1,3-diformylbenzene.[28]
pffiffiffi
À
À
ÁÁ À
À
ÁÁ
3
2
2
ð1Þ
V ¼
aLaF þ x aTbF À aLaF
cLaF þ x cTbF À cLaF
3
3
3
3
3
3
Ligands L2 to L4 are amide derivatives of dipicolinic acid (L1).
The amide carbonyl oxygen is a weaker donor than the car-
boxylate oxygen;[29] therefore, the side arms were functional-
ized with groups that are anionic at neutral pH (phosphonate
for L2, glycinate for L3 and malonate for L4) to ensure the elec-
trostatic enhancement of the coordination. Ligands L6 to L11
are derived from the hydroxyisophthalic moiety. In L6, the in-
troduction of a cyano group in the para position to the central
hydroxyl function is expected to: 1) have an impact on the
photosensitizing efficiency by lowering the ligand-centered
triplet state and 2) decrease the pKa of the phenol moiety due
to its electron-withdrawing properties. L7 to L10 are based on
amide functionalization of the central core. While the replace-
ment of carboxylate by amide groups would not be expected
to strongly affect the photosensitization efficiency, the lower-
ing of charge could weaken the surface coordination, therefore
carboxyalkyl (L7, L8) or ethylene glycol (L9, L10) side chains were
introduced to compensate for any such effect. L11 is a combina-
tion of two probably neutral chelate units with two carboxy-
lates.
Figure 1. a) TEM images of La0.9Tb0.1F3 NPs. b) DLS profile of the NPs in ultra-
pure water. c) X-ray diffraction pattern of the solid NPs (black) and its refine-
ment (red); the positions of the Bragg reflections are marked by green verti-
cal bars.
ChemPlusChem 2016, 81, 526 – 534
www.chempluschem.org
527
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Scheme 1. Synthesis of surface-coordinating ligands for the photosensitization of Tb-doped NPs. i) SOCl2. ii) H2N(CH2)2PO(OEt)2, CH2Cl2, Et3N. iii) Trimethylsilyl
bromide, CH2Cl2. iv) H2NCH2CO2Et, CH2Cl2, Et3N. v) NaOH, H2O. vi) H2NCH(CO2Et)2, Et3N, CH2Cl2. vii) [Pd(PPh3)2Cl2], CO (1 atm.), EtOH, Et3N. viii) KOH, H2O.
ix) KMnO4, H2O. x) HBr, AcOH. xi) BBr3, CH2Cl2. xii) H2N(CH2)3CO2Et, Et3N, CH2Cl2. xiii) H2NCH2(CH2OCH2)2CH2NHCOOtBu, Et3N, CH2Cl2. xiv) Ac2O, pyridine.
xv) H2N(CH2)4NH2, Et3N, CH2Cl2.
Spectroscopic titrations of the nanoparticles by the ligands
ciency. For all titrations, the optimum excitation wavelength
The coordinating ability of the ligands towards the NPs was in-
vestigated by means of titration experiments in which the ab-
sorption and luminescence spectra of a solution containing
Tb-doped NPs were monitored upon addition of increasing
amounts of the ligand. In parallel, the excitation spectra and
the Tb-centered excited-state lifetimes were also monitored.
The concentration of Tb in the mother solution was deter-
mined by ICP–AES analysis after mineralization of the samples.
Assuming a spherical morphology for the NPs and knowing
the average size of the NPs as determined by XRD (21 nm, see
above), it was possible to determine the concentration of NPs
in the solution. By keeping this concentration as well as the
parameters of the spectrometers (excitation and emission slits,
corrected excitation intensity of the lamp) constant, the differ-
ent titrations can be directly compared to obtain data on both
the strength of the coordination and the photosensitizing effi-
was chosen as the maximum of the excitation spectra of Tb at
545 nm in the presence of ligand.
A typical example is shown for ligand L6 in Figure 2. The ad-
dition of ligand resulted in increased absorption across the UV
range, with the initial spectrum providing a sloping baseline as
a result of the presence of the NPs. Upon excitation into the
ligand bands, the luminescence titration showed a gradual in-
crease of the emission of Tb, with the narrow bands at 485,
7
545, 584 and 621 nm typically associated with the 5D4! FJ
transitions of Tb (J=6 to 3, respectively).[9] Minor emission
bands could also be detected between 650 and 680 nm, asso-
ciated to J=2 to 0. In the absence of ligand, the Tb-centered
emission was only faintly observed in the background as
a result of a weak excitation. Figure 2b also shows the appear-
ance of a ligand-centered emission band with a maximum at
461 nm, the intensity of which is weak at the beginning of the
ChemPlusChem 2016, 81, 526 – 534
www.chempluschem.org
528
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Figure 3. Evolution of the Tb-centered emission intensity at 545 nm upon
addition of increasing amounts of ligands L1 (blue, lexc =270 nm), L2 (violet,
lexc =276 nm), L6 (red, lexc =307 nm) and L10 (orange, lexc =330 nm) to an
aqueous solution of La0.9Tb0.1F3 NPs [8.2 nm, Tris-HCl (0.1m), pH 7.0].
in emission intensity, and the photosensitization was modest,
indicating that the weaker coordinating ability of the amide
oxygen atom relative to the carboxylate oxygen was not com-
pensated for by the side-chain anionic groups. In the case of
L1, there was both strong coordination and excellent photo-
sensitization, but the emission intensity was reduced by the
addition of excess ligand. Close inspection of the absorption
spectra revealed that the scattering effects due to the NPs
were reduced in the presence of excess ligand (Figure S2), indi-
cating that some leaching of the TbIII might have occurred to
give soluble solution species. Interestingly, dipicolinate was
previously reported to be a good photosensitizer for Eu-doped
NPs,[30] however, leaching was not reported, and the concentra-
tion of ligand was kept low. Figure 4 represents the value of
the Tb-centered emission intensity obtained with the different
ligands at a 5000-fold excess of ligand per NP.
Figure 2. Titration of La0.9Tb0.1F3 NPs [8.2 nm, Tris-HCl (0.1m), pH 7.0] with L6
(5”10À4 m) monitored by: a) UV/Vis absorption and b) fluorescence spectros-
copy (lexc =329 nm); inset: evolution of the emission intensity at 545 nm as
a function of the concentration of added ligand.
titration, but increases strongly for larger amounts of added
ligand. As shown in Figure 2b (inset), the Tb-centered emission
was greatly enhanced at the beginning of the titration, where-
as above a certain volume of added L6, the emission became
almost constant. These forms of behavior are typical of effi-
cient surface capping and the intercept of the two straight
lines related to the growth and plateau regions was defined as
an arbitrary “equivalent volume”, which was further used for
the preparation of isolated NPs. Figure 3 shows the evolution
of the Tb-centered emission intensity at 545 nm, during the ti-
trations with ligands L1, L2, L6 and L10, as a function of the ratio
of the concentrations of ligand and NPs. These cases represent
the four different forms of behavior observed during the titra-
tions. The evolution for all the ligands studied can be found in
Figure S1 in the Supporting Information.
The best result was obtained with L6, for which the emitted
intensity grew rapidly, indicating strong surface chelation, and
the photosensitization was efficient, as evident from the high
emission intensity. Similar behavior was observed for L5 and
L11. For L2, the surface coordination was strong (rapid intensity
increase), but the photosensitization was inefficient—the emis-
sion intensities being an order of magnitude weaker, possibly
as a result of ligand binding to the surface through the phos-
phonate group only, thus leaving the picolinamide group far
from the surface. A third case was observed for L10, for which
the surface coordination is weak, resulting in a smooth in-
crease of the Tb emission intensity, and the photosensitization
was modest. Ligands L3, L4, L7, L9 and L10 all displayed relatively
weak surface coordination, resulting in only a gradual increase
Figure 4. Values of Tb-centered emission intensity of solution of La0.9Tb0.1F3
NPs [8.2 nm, Tris-HCl (0.1m), pH 7.0] upon addition of a 5000-fold excess of
the different ligands.
ChemPlusChem 2016, 81, 526 – 534
www.chempluschem.org
529
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Synthesis and characterization of ligand-coated nanoparti-
cles
minescence decay for emission at 545 nm was slow and bi-ex-
ponential, with one component corresponding to a half-life of
more than 2 ms, the other nearer to 1 ms. That two processes
are involved might be because photosensitization can involve
both TbIII ions on the surface[30] and those within the NPs, or
there might be Tb-to-Tb energy transfer within the particle or
ligand-mediated energy migration on the surface.[31,32] The
overall luminescence quantum yields are good for both NPs
and, in combination with the strong absorption, provide
brightness values which exceed those of QDs[7] and semicon-
ducting nanopolymers.[8] These values can also be compared
to a series of mononuclear Tb complexes developed by Ray-
mond and co-workers, which contains four 2-hydroxyisophtha-
lamide moieties (N=4).[12a] The measured lifetimes were typi-
cally mono-exponential (2.45–2.67 ms) in these cases, and the
brightness of the complexes ranged from 10820 to
15818mÀ1 cmÀ1, revealing the power of the NP approach.
The number of ligands bound to the surface of the NPs
varied from L6 to L11, probably reflecting the size of the li-
gands. A crude estimation of the average surface occupied by
a single ligand for a NP of diameter 21 nm gave a value of
Considering the performance of the different ligands, L6 and
L11 were chosen as the best candidates for the preparation and
isolation of surface-capped NPs. The capped NPs were synthe-
sized by adding a quantity corresponding to 1.5 times the
volume at the equivalence defined above to a solution of the
Tb-doped NPs. Purification of the NPs was achieved by size-ex-
clusion chromatography on Sephadex G75 (L11), or with centri-
fugation filters (L6) and the purified NPs were characterized by
electronic absorption spectroscopy and luminescence spectros-
copy, while the lanthanide contents were determined by ICP–
AES. Table 1 summarizes the most important spectroscopic
properties of the purified NPs and Figure 5 shows the absorp-
tion, emission and excitation spectra of the NPs capped with
L11. Similar data can be found for NPs capped with ligand L6 in
Figure S3. Deconvolution of the observed spectra as linear
combinations of the spectra of the NPs and the added ligand
enabled estimation of the component concentrations, and
thus of the number of ligand molecules per NP and the NP ab-
sorptivity values.
2
2
208 Š for L11 and 106 Š for L6. The doubled surface occupied
by L11 compared to L6 is in excellent agreement with the struc-
ture of the ligands, and might indicate full surface coverage of
the NPs.
Interestingly, the luminescence spectra displayed almost no
ligand-centered emission (previously observed at 461 nm, Fig-
ure 2b), indicating an efficient ligand-to-NP energy transfer.
The emission is essentially composed of the Tb bands. The lu-
Nanoparticle stability in the presence of citrate anions
In order to estimate the stability of the capping layer of the
NPs in solution, the luminescence of the NPs was recorded in
the presence of 0, 1, 10, 100 and 1000 equivalents of citrate
anions (equivalents calculated relative to the number of li-
gands) at pH 7.0. Citrate was chosen as a strongly competing
anion as this highly charged species is often used as a surface
capping ligand for the preparation of water-soluble lanthanide
NPs.[33] Figure 6 shows the evolution of the emission spectra of
the NPs capped with L11.
Table 1. Main spectroscopic properties of the surface-coated Tb-doped
NPs.
Absorption
lmax [nm]
(e [mÀ1 cmÀ1])
Emission
t545nm [ms] (%)[a] FH2O N[b]
B [mÀ1 cmÀ1
]
L6–Tb NPs 307 (1.6”107) 1.58 (48)
0.13 5240 2.1”106
3.76 (52)
L11–Tb NPs 337 (7.6”106) 2.59 (53)
0.29 2660 2.2”106
1.11 (47)
For a small amount of citrate added, the emitted intensity
was only slightly decreased (by 9%, for 10 equivalents) and
even with a large thousand-fold excess the signal remained in-
tense at 66% of the original value. Similar results were ob-
tained with L6-capped NPs (Figure S4), although in that case
the intensity drop is more significant with 43% of the original
intensity remaining in the presence of 1000 equivalents. In
both cases, the NPs remain bright, even with a large excess of
competing capping anions.
[a] % Population. [b] N=number of ligands per NP.
Cell staining experiments with Tb-doped nanoparticles
To validate the potential of the capped NPs for in vivo imaging,
cell-staining experiments were performed by incubation of the
NPs in a HeLa cell culture for 24 h, followed by washing of the
excess NPs and then confocal microscopy imaging. Figure 7
shows two regions of the cell culture observed by transmission
or confocal fluorescence microscopy. Upon UV excitation at
330 nm, the green Tb emission of the NPs can be readily ob-
served in the cytosol of the cells, with no trace in the nuclei.
Figure 5. Absorption (violet), excitation (blue, lem =545 nm) and emission
(green, lexc =337 nm) spectra of L11-capped Tb-doped NPs (0.89 nm) after
purification.
ChemPlusChem 2016, 81, 526 – 534
www.chempluschem.org
530
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Conclusion
As the result of a systematic study of the surface-capping abili-
ty and of the photosensitization efficiency of a family of eleven
ligands, we have been able to select and optimize the surface-
capping photosensitization of Tb-doped LaF3 NPs. For the two
best ligands, the photophysical stability of the surface-capped
NPs was tested by competition with citrate anion, showing the
capping ligands to be firmly anchored at the surface of the
NPs, even in the presence of a 1000-fold excess of citrate per
ligand, with only approximately 50% loss of the luminescence
intensity of the NPs. Characterization of the NPs showed them
to contain a few thousand ligands anchored at the surface, the
number being inversely proportional to the size of the ligands,
pointing to a full surface coverage. With such a composition,
the average absorption and brightness per NP is extremely
large, exceeding a few million mÀ1 cmÀ1 units, positioning
these NPs as excellent dyes compared to QDs[7] or semicon-
ducting polymer NPs.[8] The size of the lanthanide NPs is com-
parable to that of QDs coated with hydrophilic layers,[34,35] but
they benefit from two important advantages: a large energy
gap between excitation and emission (the Stokes shift, typical-
ly 9500 cmÀ1 for excitation at 350 nm and the first emission
band of Tb at 480 nm), and long excited-state lifetimes exceed-
ing a millisecond. Both criteria are particularly appealing for
highly sensitive time-resolved luminescence applications[36] and
time-resolved luminescence microscopy,[14] allowing for large
improvement of the signal-to-noise ratio and thus of the sensi-
tivity of the analysis. Finally, the use of Tb-doped NPs was
demonstrated to be efficient for luminescence staining of HeLa
cells, with specific localization into the lysosomes of the cells.
We believe that surface photofunctionalized lanthanide-
doped NPs represent an interesting alternative to the current
luminescent NPs and that there is place for numerous further
improvements of their capabilities, such as the incorporation
of activated functions into the ligand skeleton for biomolecule
labeling and targeted delivery, the possibility of multiplexed
fluorescence resonance energy transfer analysis,[13] and the
possibility of using different lanthanide cations for dual-wave-
length imaging (visible and NIR),[37] or dual-mode imaging such
as magnetic resonance imaging and optical spectroscopy.[38]
Figure 6. Evolution of the emission spectra of L11-capped La0.9Tb0.1F3 NPs
(lexc =341 nm) in the presence of 0 (green), 1 (orange), 10 (red), 100 (violet)
and 1000 (blue) equivalents of citrate anions added to the solution (water,
pH 7.0); inset: evolution of the intensity at 545 nm.
Experimental Section
Figure 7. a) Transmission, b) QD fluorescence (lexc =330 nm, lem =545 nm),
c) LysoTracker fluorescence (lexc =577 nm, lem =590 nm), and d) merged
confocal microscopy images of two regions (left and right column) of HeLa
cells incubated for 24 h with a solution of L6-capped La0.9Tb0.1F3 NPs
(1.33 nm); scale bar=20 mm.
General methods
ICP–AES analysis of samples in water were performed on a Varian
720 spectrometer equipped with a quartz Meinhard nebulizer and
a cyclone spray chamber. In a typical experiment, the mother solu-
tion of NPs was sonicated for 10 min in an ultrasound bath, and
was strongly agitated on a vortex before a sample (1 mL) was re-
moved by pipette. The sample was diluted with high-purity nitric
acid (3 mL) and heated in a microwave oven at 2008C for 45 min.
The concentration was then determined by ICP–AES spectrometry
by comparison with commercial standard samples.
Co-localization experiments were performed with a LysoTracker
dye, showing an excellent overlay of the dye with the green
Tb emission, and clearly indicating localization of the Tb NPs
into the lysosomes of the HeLa cells.
The XRD pattern was recorded at room temperature using a Bruck-
er D8 Advance diffractometer equipped with a monochromatic
copper radiation source (Ka =1.54056 Š) and a Sol-X detector in
ChemPlusChem 2016, 81, 526 – 534
www.chempluschem.org
531
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
the 20–608 (2q) range with a scan step of 0.028. Profile-matching
refinement was performed with the Fullprof[39] software using Le
Bail’s[40] method with the modified Thompson–Cox–Hasting profile
function. Instrumental broadening has been previously determined
by measuring the scattering from corundum (NIST standard SRM
1976b). Using such a process enabled us to calculate cell parame-
ters and the size of the diffracting domain. TEM was performed
with a JEOL 2100F electron microscope operating at 200 kV
equipped with a GATAN GIF 200 electron imaging filter. Granulom-
etry measurements based on DLS were performed on a suspension
of NPs in distilled water using a Malvern Nano-Zs Zetasizer appara-
tus.
citrate ions. A solution of NPs (5.25”10À9 m) in Tris-HCl buffer
(0.1m, pH 7.0) was mixed with 2Veq of ligand stock solution and in-
creasing amounts of a citrate stock solution (pH 7) were added to
reach the following citrate/ligand ratios: 1:1, 10:1, 100:1, and
1000:1.
Cell cultures
Human cervical carcinoma (HeLa) cells were purchased from the
American Type Culture Collection (Manassas, VA ; #CCL-2). The
HeLa cells were grown in Dulbecco’s modified Eagle’s medium
(DMEM) supplemented with 10% fetal bovine serum, 1% penicillin
and streptomycin at 378C and 5% CO2.
TEM images were recorded with a TOPCON model 002B transmis-
sion electron microscope coupled with energy dispersive X-ray
(EDX) spectroscopy, operating at 200 kV, with a point-to-point reso-
lution of 0.18 nm. Powder samples were dispersed in ethanol and
a drop of this suspension was deposited on TEM grids coated by
holey amorphous carbon. In order to avoid disturbing random sig-
nals coming from the amorphous carbon, the detected LaxTbyF3
particles were those which lies on strand of these holes.
In vitro imaging
To test the suitability of the Tb NPs as bioprobes, in vitro imaging
of HeLa cells incubated with Tb NPs was performed on a Zeiss
Axio Observer Z1 fluorescence microscope. HeLa cells were incu-
bated in DMEM containing Tb NPs (0.5 nm; prepared by dilution of
a 96 nm stock solution with DMEM) at 378C for 24 h under 5%
CO2, and then washed with phosphate-buffered saline (PBS) to
completely remove excess Tb NPs before imaging. The samples
were excited at 330 nm and their emission detected at 547 nm.
Co-localization experiments were performed by adding LysoTracker
Red DND-99 (Life Technologies; excitation: 577 nm; emission:
590 nm) into the incubation medium.
Photophysical measurements
UV/Vis absorption spectra were recorded on
a PerkinElmer
Lambda 950 spectrometer. Steady-state emission spectra were re-
corded on an Edinburgh Instruments FLP920 spectrometer with
a continuous 450 W Xe lamp and a red-sensitive photomultiplier in
a Peltier housing. All spectra were corrected for the instrumental
functions. If necessary, a 399 nm cutoff filter was used to eliminate
second-order artifacts. Phosphorescence lifetimes were measured
on the same instrument operating in the multichannel spectrosco-
py mode, using a Xe flash lamp as the excitation source. Errors in
luminescence lifetimes were estimated to Æ10%. Luminescence
quantum yields were measured according to conventional proce-
dures,[1] with optically dilute solutions (optical density <0.05),
using rhodamine 6G in water (F=0.76)[41] and a Tb complex pre-
pared in the laboratory ([TbL(H2O)Na], F=0.31)[42] as references.
Errors in absolute quantum yields were estimated to Æ15%. For
the calculation of the extinction coefficients of the capped-NPs,
the measured UV/Vis absorption spectrum was deconvoluted into
the sum of two contributions, that is, the absorption of the cap-
ping ligands and the diffraction of the NP core, and only the
ligand absorption was taken into account.
Synthesis of the ligands
Solvents and starting materials were purchased from Sigma–Al-
drich, Acros and Alfa Aesar and used without further purification.
IR spectra were recorded on a PerkinElmer Spectrum One spectro-
photometer as solid samples and only the most significant absorp-
tion bands are reported [cmÀ1]. Elemental analyses and MS analysis
were performed by the Service Commun d’Analyses of the Univer-
sity of Strasbourg. 13C NMR spectra and 2D COSY and NOESY ex-
periments were measured on Bruker Avance 300 and Avance 400
spectrometers operating at 300 and 400 MHz, respectively. Chemi-
cal shifts are reported in ppm, with residual protonated solvent as
internal reference.[43] Ligand L1, ethyl glycinate, ethyl 4-aminobuty-
rate, ethyl aminomalonate, 2,6-dimethylanisole, and 3,5-diiodo-4-
hydroxybenzonitrile were commercially sourced. 2,6-Pyridinedicar-
bonyl dichloride,[44] 5,[45] 6,[46] L5,[45] diethyl (2-aminoethyl)phospho-
nate,[47] and N-butyloxycarbonyl-3,6-dioxaoctane-1,8-diamine[48]
were prepared according to literature procedures. Compound 2
was prepared according to a new protocol and its analysis corre-
spond to those reported in the literature.[49] Full experimental de-
tails for the synthesis of ligands L2–L4 and L6–L11 can be found in
the Supporting Information.
Spectroscopic titrations of the NPs with the ligands were per-
formed by monitoring the changes in the UV/Vis absorption, emis-
sion and TbIII excitation (lem =545 nm) spectra, as well as the TbIII
luminescence lifetime of the NPs in the presence of increasing
amounts of the ligand. In a typical experiment, a solution of
La0.9Tb0.1F3 NPs in water (16 mL, 1.03”10À6 m for L1–L4 and 6.57”
10À7 m for ligands L5–L11) was diluted with a Tris-HCl-buffered solu-
tion (0.1m, pH 7.0, 1984 mL). These solutions were titrated at room
temperature with solutions of the ligands in the same buffer (5”
10À4 m). For each titration, the excitation wavelength was adjusted
to the maximum of the TbIII excitation spectrum (lem =545 nm)
and the excitation and emission slits were kept constant in order
to enable comparison of the different datasets. Plotting the
Synthesis of the nanoparticles
A solution of NH4F (0.72m, 3.51 mL) in water was added dropwise
to a stirred aqueous mixture of LaCl3 (0.05m, 14.4 mL,) and TbCl3
(0.05m, 1.6 mL,) at room temperature, resulting in the formation of
a slightly turbid solution. The mixture was heated in a microwave
oven at 1508C for 12 min. After cooling, the precipitate was col-
lected by centrifugation at 9000 rpm for 25 min. The isolated solid
was dispersed in milliQ water (30 mL) with sonication at 608C for
1 h. The Tb and La content of the obtained solution was deter-
mined by ICP–AES.
7
changes in Tb emission intensity (as the integral of the 5D4! F5
transition) as a function of added volume of ligand solution al-
lowed the determination of the “equivalent volume” Veq (see the
Results and Discussion section).
The stability in water of the NPs capped with the ligands L6 and
L11 was assessed by competition experiments in the presence of
ChemPlusChem 2016, 81, 526 – 534
www.chempluschem.org
532
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
debrandt, Inorg. Chem. 2014, 53, 1824; c) L. J. Charbonni›re, N. Weibel,
C. Estournes, C. Leuvrey, R. Ziessel, New J. Chem. 2004, 28, 777.
[15] G. Chen, H. Qiu, P. N. Prasad, X. Chen, Chem. Rev. 2014, 114, 5161.
[16] a) L. Aboshyan-Sorgho, C. Besnard, P. Pattison, K. R. Kittilstved, A. Ae-
bischer, J.-C. G. Bünzli, A. Hauser, C. Piguet, Angew. Chem. Int. Ed. 2011,
50, 4108; Angew. Chem. 2011, 123, 4194; b) O. A. Blackburn, M. Tropia-
no, T. Just Sørensen, J. Thom, A. Beeby, L. M. Bushby, D. Parker, L. S. Na-
trajan, S. Faulkner, Phys. Chem. Chem. Phys. 2012, 14, 13378.
[17] W. Zheng, D. Tu, P. Huang, S. Zhou, Z. Chen, X. Chen, Chem. Commun.
2015, 51, 4129.
[18] X. Huang, S. Han, W. Huang, X. Liu, Chem. Soc. Rev. 2013, 42, 173.
[19] W. Zou, C. Visser, J. A. Maduro, M. S. Pshenichnikov, J. C. Hummelen,
Nat. Photonics 2012, 6, 560.
[20] J. Zhang, C. M. Shade, D. A. Chengelis, S. Petoud, J. Am. Chem. Soc.
2007, 129, 14834.
Synthesis and purification of ligand-capped nanoparticles
L6-capped nanoparticles: stock solution of La0.9Tb0.1F3 NPs
A
(0.321 mM, 660 mL) and a solution of L6 (0.5 mm, 990 mL) were
mixed and stirred at room temperature for 5 h at pH 7.40. The mix-
ture was filtered in an Amicon 50 kDa eppendorf tube by centrifu-
gation at 14000 rpm for 10 min. The filter was washed with milliQ
water (400 mL) by centrifugation. MilliQ water (200 mL) was added
into the filter, which was centrifuged at 5000 rpm for 5 min to col-
lect the ligand-coated NPs.
L11-capped NPs: Solutions of La0.9Tb0.1F3 NPs (0.66 mm, 1.67 mL) and
L11 (0.5 mm, 12 mL) were mixed and stirred at room temperature
for 5 h at pH 7.40. The mixture was purified by size-exclusion chro-
matography (Sephadex G75, milliQ water) and green luminescent
fractions containing the capped NPs were separated and recov-
ered.
[21] L. J. Charbonni›re, J. L. Rehspringer, R. Ziessel, Y. Zimmermann, New J.
Chem. 2008, 32, 1055.
[22] For examples, see: a) M. Irfanullah, D. Kumar Sharma, R. Chulliyil, A.
Chowdhury, Dalton Trans. 2015, 44, 3082; b) N. Gauthier, O. Raccurt, D.
Imbert, M. Mazzanti, J. Nanopart. Res. 2013, 15, 1723; c) S. Li, X. Li, Y.
Jiang, Z. Hou, Z. Cheng, P. Ma, C. Li, J. Lin, RSC Adv. 2014, 4, 55100.
[23] F. Wang, Y. Zhang, X. Fan, M. Wang, J. Mater. Chem. 2006, 16, 1031.
[24] For dipicolinic acid and its derivatives, see: a) A. Aebischer, F. Gumy,
J. C. Bünzli, Phys. Chem. Chem. Phys. 2009, 11, 1346; b) A.-S. Chauvin, F.
Gumy, D. Imbert, J.-C. G. Bünzli, Spectrosc. Lett. 2004, 37, 517; c) M. Reg-
ueiro-Figueroa, B. Bensenane, E. Ruscsak, D. Esteban-Gomez, L. J. Char-
bonni›re, G. Tircsco, I. Toth, A. de Blas, T. Rodríguez-Blas, C. Platas-Igle-
sias, Inorg. Chem. 2011, 50, 4125.
Acknowledgements
J.G. thanks the RØgion Alsace and Hong Kong Baptist University
for funding. The French Minist›re des Affaires Étrang›res is grate-
fully acknowledged for financial support (PHC Procore no.
30707NM). K.L.W. thanks the PROCORE France/Hong Kong joint
research scheme (F-HKBU201/13). Dr. Anne Boos and Pascale
Ronot are gratefully acknowledged for having performed the
ICP–AES analysis.
[25] For 2-hydroxyisophthalic acid, see : a) L. Benisvy, P. Gamez, W. Tian Fu,
H. Kooijman, A. L. Spek, A. Meijerink, J. Reedijk, Dalton Trans. 2008,
3147; b) K. L. Peterson, M. J. Margherio, P. Doan, K. T. Wilke, V. C. Pierre,
Inorg. Chem. 2013, 52, 9390.
[26] P. A. Brayshaw, J.-C. G. Bünzli, P. Froidevaux, J. M. Harrowfield, Y. Kim,
A. N. Sobolev, Inorg. Chem. 1995, 34, 2068.
[27] G.-L. Law, T. A. Pham, J. Xu, K. N. Raymond, Angew. Chem. Int. Ed. 2012,
51, 2371; Angew. Chem. 2012, 124, 2421.
Keywords: antenna effects · lanthanides · luminescence ·
microscopy · nanoparticles
[28] a) F. Avecilla, C. Platas-Iglesias, R. Rodríguez-CortiÇas, G. Guillemot, J.-
C. G. Bünzli, C. D. Brondino, C. F. G. C. Geraldes, A. de Blas, T. Rodríguez-
Blas, Dalton Trans. 2002, 4658; b) F. Avecilla, A. de Blas, R. Bastida, D. E.
Fenton, J. Mahía, A. Macías, C. Platas-Iglesias, A. Rodríguez, T. Rodrí-
guez-Blas, Chem. Commun. 1999, 125.
[29] F. Renaud, C. Piguet, G. Bernardinelli, J.-C. G. Bünzli, G. Hopfgartner,
Chem. Eur. J. 1997, 3, 1646.
[30] A. M. Cross, P. S. May, F. C. J. M. van Veggel, M. T. Berry, J. Phys. Chem. C
2010, 114, 14740.
[31] A. Nonat, M. Regueiro-Figueroa, D. Esteban-Gomez, A. de Blas, T. Rodri-
guez-Blas, C. Platas-Iglesias, L. J. Charbonni›re, Chem. Eur. J. 2012, 18,
8163.
[32] E. A. Mikhalyova, A. V. Yakovenko, M. Zeller, M. A. Kiskin, Y. V. Kolomzar-
ov, I. L. Eremenko, A. W. Addison, V. V. Pavlishchuk, Inorg. Chem. 2015,
54, 3125.
[33] V. Sudarsan, F. C. J. M. van Veggel, R. A. Herring, M. Raudsepp, J. Mater.
Chem. 2005, 15, 1332.
[34] T. Pons, H. T. Uyeda, I. L. Medintz, H. Mattoussi, J. Phys. Chem. B 2006,
110, 20308.
[35] T. Pellegrino, L. Manna, S. Kudera, T. Liedl, D. Koktysh, A. L. Rogach, S.
Keller, J. Radler, G. Natile, W. J. Parak, Nano Lett. 2004, 4, 703.
[36] N. Hildebrandt, L. J. Charbonniere, H. G. Lçhmannsrçben, J. Biomed. Bio-
technol. 2007, 2007, 79169.
[37] T. Zhang, X. Zhu, C. C. W. Cheng, W.-M. Kwok, H.-L. Tam, J. Hao, D. W. J.
Kwong, W.-K. Wong, K.-L. Wong, J. Am. Chem. Soc. 2011, 133, 20120.
[38] M. Regueiro-Figueroa, A. Nonat, G. A. Rolla, D. Esteban-Gomez, A.
de Blas, T. Rodriguez-Blas, L. J. Charbonni›re, M. Botta, C. Platas-Iglesias,
Chem. Eur. J. 2013, 19, 11696.
[1] B. Valeur, Molecular Fluorescence, Wiley-VCH, Weinheim, 2002, Chap-
ter 1, pp 3–19.
[2] H.-W. Ai, Anal. Bioanal. Chem. 2015, 407, 9.
[3] a) I. L. Medintz, H. T. Uyeda, E. R. Goldman, H. Mattoussi, Nat. Mater.
2005, 4, 435; b) R. Gill, M. Zayats, I. Willner, Angew. Chem. Int. Ed. 2008,
47, 7602; Angew. Chem. 2008, 120, 7714.
[4] a) P. D. Howes, R. Chandrawati, M. M. Stevens, Science 2014, 346, 53;
b) R. Nishiyabu, N. Hashimoto, T. Cho, K. Watanabe, T. Yasunaga, A.
Endo, K. Kaneko, T. Niidome, M. Murata, C. Adachi, Y. Katayama, M. Ha-
shizume, N. Kimizuka, J. Am. Chem. Soc. 2009, 131, 2151.
[5] L. D. Lavis, R. T. Raines, ACS Chem. Biol. 2008, 3, 142.
[6] P. K. Chattopadhyay, B. Gaylord, A. Palmer, N. Jiang, M. Raven, G. Lewis,
M. A. Reuter, A. K. M. N. Rahman, D. A. Price, M. R. Betts, M. Roederer, Cy-
tometry A 2012, 81A, 456.
[7] P. Kukura, M. Celebrano, A. Renn, V. Sandoghdar, Nano Lett. 2009, 9,
926.
[8] L. Wei, P. Zhou, Q. Yang, Q. Yang, M. Ma, B. Chen, L. Xiao, Nanoscale
2014, 6, 11351.
[9] S. V. Eliseeva, J.-C. G. Bünzli, Chem. Soc. Rev. 2010, 39, 189.
[10] S. Quici, M. Cavazzini, G. Marzanni, G. Accorsi, N. Armaroli, B. Ventura, F.
Barigelletti, Inorg. Chem. 2005, 44, 529.
[11] S. I. Weissmann, J. Chem. Phys. 1942, 10, 214.
[12] a) J. Xu, T. M. Corneillie, E. G. Moore, G.-L. Law, N. G. Butlin, K. N. Ray-
mond, J. Am. Chem. Soc. 2011, 133, 19900; b) M. Delbianco, V. Sadovni-
kova, E. Bourrier, G. Mathis, L. Lamarque, J. M. Zwier, D. Parker, Angew.
Chem. Int. Ed. 2014, 53, 10718; Angew. Chem. 2014, 126, 10894; c) M.
Starck, P. Kadjane, E. Bois, B. Darbouret, A. Incamps, R. Ziessel, L. J. Char-
bonni›re, Chem. Eur. J. 2011, 17, 9164.
[39] J. Rodríguez-Carvajal, J. Phys. B: Condens. Matter 1993, 192, 55.
[40] A. Le Bail, H. Duroy, J. L. Fourquet, Mater. Res. Bull. 1988, 23, 447.
[41] J. Olmsted, J. Phys. Chem. 1979, 83, 2581–2584.
[13] D. Geißler, L. J. Charbonni›re, R. Ziessel, N. G. Butlin, H. G. Lçhmannsrç-
ben, N. Hildebrandt, Angew. Chem. Int. Ed. 2010, 49, 1396; Angew.
Chem. 2010, 122, 1438.
[42] N. Weibel, L. J. Charbonni›re, M. Guardigli, A. Roda, R. Ziessel, J. Am.
Chem. Soc. 2004, 126, 4888.
[43] H. E. Gottlieb, K. Kotlyar, A. Nudelman, J. Org. Chem. 1997, 62, 7512.
[14] a) O. Faklaris, M. Cottet, A. Falco, B. Villier, M. Laget, J. M. Zwier, E. Trin-
quet, B. Mouillac, J. P. Pin, T. Durroux, FASEB J. 2015, 29, 2235; b) D.
Geißler, S. Linden, K. Liermann, K. D. Wegner, L. J. Charbonni›re, N. Hil-
ChemPlusChem 2016, 81, 526 – 534
www.chempluschem.org
533
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[44] M. Mateescu, I. Nuss, A. Southan, H. Messenger, S. V. Wegner, J. Kupka,
M. Bach, G. E. M. Tovar, H. Boehm, S. Laschat, Synthesis 2014, 46, 1243.
[45] K. Wang, H.-H. Zou, Z.-L. Chen, Z. Zhang, W.-Y. Sun, F.-P. Liang, Dalton
Trans. 2014, 43, 12989.
[49] H. Svobodovµ, Nonappa, M. Lahtinen, Z. Wimmer, E. Kolehmainen, Soft
Mater. 2012, 8, 7840.
[46] P. Ghosh, G. Federwisch, M. Kogej, C. A. Schalley, D. Haase, W. Saak, A.
Lützen, R. M. Gschwind, Org. Biomol. Chem. 2005, 3, 2691.
[47] A. Graillot, S. Monge, C. Faur, D. Bouyer, J. J. Robin, Polym. Chem. 2013,
4, 795.
Manuscript received: January 6, 2016
Revised: January 27, 2016
Accepted Article published: January 28, 2016
Final Article published: February 16, 2016
[48] M. Trester-Zedlitz, K. Kamada, S. K. Burley, D. Fenyo, B. T. Chait, T. W.
Muir, J. Am. Chem. Soc. 2003, 125, 2416.
ChemPlusChem 2016, 81, 526 – 534
www.chempluschem.org
534
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim