Near-IR two photon microscopy imaging of silica nanoparticles functionalized with isolated sensitized Yb(III) centers

Article abstract of DOI:10.1021/cm404140q

Bright nano-objects emitting in the near-infrared with a maximal cross section of 41.4 × 10³ GM (Goppert Mayer) were prepared by implanting ca. 180 4,4′-diethylaminostyryl-2,2′-bipyridine (DEAS) Yb(III) complexes on the surface of 12-nm silica nanoparticles. The surface complexes ([DEAS·Ln@SiO₂], Ln = Y, Yb) were characterized using IR, solid-state NMR, UV-vis, and EXAFS spectroscopies in combination with the preparation and characterization of similar molecular analogues by analytical techniques (IR, solution NMR, UV-vis, X-ray crystallography) as well as DFT calculations. Starting from the partial dehydroxylation of the silica at 700 C under a high vacuum having 0.8 OH·nm^-2, the grafting of Ln(N(SiMe₃)₂)₃ generates ≡SiO- Ln(N(SiMe₃)₂)₂, which upon thermal step and coordination of the DEAS chromophore yields (≡SiO)₃Ln(DEAS). Surface and molecular analogues display similar properties, in terms of DEAS binding constants absorption maxima and luminescence properties (intense emission band assigned to a ligand centered CT fluorescence and lifetime) in the solid state, consistent with the molecular nature of the surface species. The densely functionalized nanoparticles can be dispersed via ultrasonication in small 15-20 nm aggregates (one to six elementary particles) that were detected using two-photon microscopy imaging at 720 nm excitation, making them promising nano-objects for bioimaging.

Full text of DOI:10.1021/cm404140q

Article
pubs.acs.org/cm
Near-IR Two Photon Microscopy Imaging of Silica Nanoparticles
Functionalized with Isolated Sensitized Yb(III) Centers
Giuseppe Lapadula,† Adrien Bourdolle,‡,§ Florian Allouche,† Matthew P. Conley,† Iker del Rosal,∥
Laurent Maron,∥ Wayne W. Lukens,⊥ Yannick Guyot,# Chantal Andraud,‡ Sophie Brasselet,▽
Christophe Coperet,*,†,§,⧫ Olivier Maury,*,‡ and Richard A. Andersen○
́
†ETH Zurich, Department of Chemistry and Applied Biosciences, Vladimir Prelog Weg 2, CH−8093 Zurich, Switzerland
̈
̈
‡Universite
́
de Lyon, CNRS, UMR 5182−CNRS − Ecole Normale Super
́
ieure de Lyon, Universtite
́
́
de Lyon 1, 46 allee d’Italie, 69007
Lyon, France
§Universite
de Lyon, C2P2, UMR 5265 (CNRS−CPE Lyon−Universite
Cedex, France
∥Universite
́
́
Lyon 1), 43 Bd. Du 11 Novembre 1918, 69616 Villeurbanne
́
de Toulouse and CNRS, LPCNO INSA/UPS/CNRS 135, avenue de Rangueil, 31077 Toulouse Cedex 4, France
⊥Chemical Sciences Division, University of California, Lawrence Berkeley National Laboratory, Berkeley, California 94720, United
States
#Universite
France
́
de Lyon, Institut Lumier
̀
e Matier
̀
e, UMR 5306 CNRS−Universite
́
Lyon1, 10 rue Ada Byron, 69622 Villeurbanne Cedex,
▽Institut Fresnel, CNRS UMR 6133, Universite
13397 Marseille Cedex 20, France
́
Aix Marseille III, Ecole Centrale de Marseille, Domaine Universitaire St Jerome,
́
○Department of Chemistry, University of California, Berkeley, California 94720, United States
S
* Supporting Information
ABSTRACT: Bright nano-objects emitting in the near-infrared with a
maximal cross section of 41.4 × 103 GM (Goppert Mayer) were prepared
by implanting ca. 180 4,4′-diethylaminostyryl-2,2′-bipyridine (DEAS)
Yb(III) complexes on the surface of 12-nm silica nanoparticles. The
surface complexes ([DEAS·Ln@SiO2], Ln = Y, Yb) were characterized
using IR, solid-state NMR, UV-vis, and EXAFS spectroscopies in
combination with the preparation and characterization of similar molecular
analogues by analytical techniques (IR, solution NMR, UV−vis, X-ray
crystallography) as well as DFT calculations. Starting from the partial
dehydroxylation of the silica at 700 °C under a high vacuum having 0.8 OH·nm−2, the grafting of Ln(N(SiMe3)2)3 generates 
SiO−Ln(N(SiMe3)2)2, which upon thermal step and coordination of the DEAS chromophore yields (SiO)3Ln(DEAS).
Surface and molecular analogues display similar properties, in terms of DEAS binding constants absorption maxima and
luminescence properties (intense emission band assigned to a ligand centered CT fluorescence and lifetime) in the solid state,
consistent with the molecular nature of the surface species. The densely functionalized nanoparticles can be dispersed via
ultrasonication in small 15−20 nm aggregates (one to six elementary particles) that were detected using two-photon microscopy
imaging at 720 nm excitation, making them promising nano-objects for bioimaging.
transparent in this region, leading to less scattering.5 Though
1. INTRODUCTION
attractive, most chromophores are not fully optimized for this
The need for effective chemical sensors that could be useful for
imaging biological samples in vitro or in vivo has been one of the
major driving forces for the development of luminescent
spectral range (low quantum yield, small Stokes shift).6−8
Biphotonic microscopy would give the benefits of NIR
excitation with desirable luminescent properties.9 Biphotonic
molecules and materials.1,2 While many examples of strongly
microscopy relies on the nonlinear simultaneous absorption of
luminescent molecules for chemical sensing in biological
samples are currently available,3 the excitation wavelength lies
two photons of half energy in the NIR, from an intense
femtosecond Ti:sapphire laser source [700−1050 nm]10−13
in most cases in the visible region of the spectrum, which does
not deeply penetrate biological tissues and can lead to a
which enables the use of chromophores that are excited in the
potential risk for photodamage because of the high energy of
visible light.4 One potential solution to these challenges is the
Received: October 2, 2013
use of near-infrared (NIR) light, located in the 700−1200 nm
spectral range, because biological tissues are effectively
Revised: December 30, 2013
Published: December 31, 2013
© 2013 American Chemical Society
1062
dx.doi.org/10.1021/cm404140q | Chem. Mater. 2014, 26, 1062−1073

Chemistry of Materials
Article
1
1 nm. The H, 13C, and 29Si NMR spectra were obtained on Bruker
DRX 200, DRX 250, DRX 300, DRX 400, or DRX 500 spectrometers.
The solution spectra were recorded in the given solvent at room
visible region with NIR light. However, most chromophores
designed for biphotonic microscopy emit at higher energy
compared to the laser excitation and fall in the visible region.
Only a few notable examples of red emitters have been
temperature. The H and 13C chemical shifts were referenced relative
to the residual solvent peak. For the solid-state spectra, a Bruker DRX
1
reported.14−17 Ytterbium complexes are ideal candidates for
1
400 was used. The MAS frequency was set at 10 kHz for all H and
2
NIR-to-NIR biphotonic microscopy because of the F5/2
→
13C spectra and 5 kHz for 29Si. The samples were introduced in a 4
2F7/2 ytterbium(III) transition around 1000 nm, an emission at
higher wavelength than the incident laser.18,19 This emission
has been used for 3D imaging of a mouse brain vascular
network.19 Though promising, the low luminescent quantum
yield efficiency of molecular ytterbium species limits the
generality of this technique.20
mm zirconia rotor in the glovebox. H, 13C, and 29Si chemical shifts
1
were referenced to external TMS.
X-Ray Crystallography. The crystals were placed in paratone and
mounted in the beam under a flow of nitrogen at 100 K on a Bruker
SMART APEX II diffractometer equipped with a CCD area detector
using Mo Kα radiation. Empirical absorption correction was
performed with SADABS 2008/1 (Bruker). The structures were
solved by direct methods (SHELXS 97) followed by least-square
refinement (SHELXL 97) using the WinGX suite of programs and
OLEX2 1.1. The non-hydrogen atoms were refined anisotropically.
The hydrogen atoms were placed at calculated positions. Additional
crystallographic information is given in the Supporting Information
(Table S1, Figures S1−S4).
EXAFS Spectroscopy. Samples were loaded into an aluminum
holder equipped with aluminized Mylar windows sealed with an
indium gasket in an argon-filled inert atmosphere glovebox. Assembled
holders were sealed in glass jars until just prior to data collection. At
the beamline, the jar was opened, and the sample was quickly
transferred to a helium filled cryostat, which was evacuated then
refilled with helium gas three times. Data were obtained at room
temperature (the cryostat was only used to provide additional oxygen
protection). X-ray absorption data were obtained at beamline 4−1 of
the Stanford Synchrotron Radiation Light Source. The X-ray beam was
monochromatized using a double crystal monochromator with
Si(220), ϕ = 90° crystals. The second crystal was detuned by 50%
to reduce the harmonic content of the beam. Data were obtained in
transmission at the Yb L3 edge using nitrogen-filled ion chambers and
at the Y K edge using argon-filled ion chambers. Data were deglitched
using the EXAFSPAK suite of programs written by Graham George.
Data were treated to remove the pre- and post-edge backgrounds, and
the EXAFS were obtained by subtracting a spline from the absorption
data using the software package Athena. In the case of the Y K edge
data, the EXAFS contained a large, sharp feature at 7 Å−1 presumably
due to a multielectron excitation.
The feature was removed analogously to the process for deglitching
the dataa fourth-order polynomial was fit, though the data on either
side of the feature and the distorted area of the EXAFS spectrum were
replaced by the polynomial. EXAFS data were fit using the software
package Artemis using theoretical scattering curves generated by Feff7.
The number of first shell neighbors was fixed at 3, and the number of
second shell neighbors was varied until the best fit was obtained. In the
case of Yb@SiO2, there appears to be another shell of atoms further
from the Yb center; however, attempts to fit this feature using Si, O, N,
C, or Yb neighbors resulted in an increase in the value of reduced χ2,
so this shell was not fit.
Transmission Electron Microscopy. The sample was ground
mechanically, sonicated for 30 min in isopropanol, and dispersed on a
TEM grid. TEM pictures were obtained on a Phillips CM12
transmission electron microscope. The filament used to generate the
electron beam is a tungsten filament, and the accelerating voltage is
120 kV. Images were required at 100 000 s time magnification.
Luminescence Spectroscopy. The luminescence spectra were
measured using a Horiba−Jobin Yvon Fluorolog-3 spectrofluorimeter,
equipped with a three slit double grating excitation and emission
monochromator with dispersions of 2.1 nm/mm (1200 grooves/mm).
The steady-state luminescence was excited by unpolarized light from a
450 W xenon CW lamp and detected at an angle of 90° for diluted
solution measurements or at 22.5° for solid state measurement (front
face detection) by a Peltier-cooled red-sensitive Hamamatsu R2658P
photomultiplier tube (300−1010 nm). Spectra were reference
corrected for both the excitation source light intensity variation
(lamp and grating) and the emission spectral response (detector and
grating). Uncorrected near infrared spectra were recorded at an angle
To overcome this drawback, the two-photon brightness
parameterdefined as the product of the two-photon cross-
section and the quantum yieldcan be improved using the
local concentration effect, i.e., the confinement of a large number
of chromophores in a nanoscale size object. This strategy has
been successfully used for organic chromophores and europium
based luminescent single nanoparticles.21−30 Though these
nanomaterials offer potentially intriguing luminescence proper-
ties, detrimental luminescent self-quenching may occur in some
doped luminescent nanoparticles prepared by Stober or
̈
ORMOSIL (Organically Modified Silica) emulsion sol−gel
techniques. Here, we describe the preparation and molecular-
level characterization of bright ytterbium-containing lumines-
cent nanoparticles (LSN). We use surface organometallic
chemistry (SOMC)31−36 to regularly position well-defined
chromophoric ytterbium complexes containing DEAS = 4,4′-
diethylaminostyryl-2,2′-bipyridine, a nonlinear optical chromo-
phore−ligand37−41 on the silica surface. These Yb-based LSNs
were detected using NIR-to-NIR biphotonic optical micros-
copy.19
The direct characterization of these materials by spectro-
scopic methods was complemented by characterizing the
corresponding diamagnetic yttrium analogous surface species
and homogeneous molecular siloxide analogues Ln(OSi-
(OtBu)3)3(L) (Ln = Y and Yb with L = 4,4′-Me2bipy and
DEAS). The photophysical properties of the surface complexes
are comparable to the molecular complexes, clearly showing
that the surface species are located in a well-defined molecular
environment. These results allow a direct structure−property
relationship between a structure of the functionalized nano-
particles and the photophysical properties.
2. EXPERIMENTAL SECTION
General Considerations. All of the experiments were performed
under dry, oxygen-free argon using Schlenk and glovebox techniques
for the organometallic synthesis. For the syntheses and the treatment
of the surface species, reactions were carried out using high vacuum
lines (10−5 mbar) and glovebox techniques. Dichloromethane,
pentane, and toluene were purified using a double MBraun SPS
alumina column, degassed before use, and stored over 4 Å molecular
sieves. Hexamethyldisiloxane (HMDSO) was distilled from calcium
hydride and stored on molecular sieves. Deuterated solvents were
directly distilled by vacuum transfer into a J-Young NMR tube.
Toluene d8 and C6D6 were stored on sodium and benzophenone;
CD2Cl2 was stored on P2O5 and degassed before use. The yttrium and
ytterbium amides Ln(N(SiMe3)2)3 were synthesized modifying a
procedure reported in the literature.42 Silica (AEROSIL 200 m2/g)
was dehydroxylated according to a published procedure.43 All infrared
(IR) spectra were recorded using a Bruker α spectrometer placed in
the glovebox equipped with OPUS software. Typically, 32 scans were
accumulated for each spectrum. All the UV−visible spectra were
recorded in CH2Cl2 solution using quartz Schlenk cuvettes with a Cary
5000 UV−vis−NIR spectrometer at 600 nm/min with a resolution of
1063
dx.doi.org/10.1021/cm404140q | Chem. Mater. 2014, 26, 1062−1073

Chemistry of Materials
Article
of 45° using a liquid nitrogen cooled, solid indium/gallium/arsenic
detector (850−1600 nm). The luminescence decay of ytterbium
complexes was performed using a home-made setup: The excitation of
the Yb(III) luminescence decays was performed with an optical
parametric oscillator from EKSPLA NT342, pumped with a pulsed
frequency tripled YAG:Nd laser. The pulse duration was 6 ns at a 10
Hz repetition rate. The detection was performed by a R1767
Hamamatsu photomultiplier through a Jobin−Yvon monochromator
equipped with a 1 μm blazed grating. The signal was visualized and
averaged with a Lecroy digital oscilloscope LT342.
Two-Photon Microscopy Imaging. The sample was ground
mechanically, vortexed, and dispersed on a microscope coverslip. The
two-photon excitation microscope is based on an excitation source
from a Ti:Sa laser (pulse width, 100 fs; repetition rate, 80 MHz) set at
a wavelength of 760 nm, focused on the sample using a water
immersion high numerical aperture objective (NA 1.15, 40) after
reflection on a protected silver mirror. Images are formed by
galvanometric scanning (typically 10 μm × 10 μm regions are
scanned, sampled with 60 nm per pixel, with integration time 100 μs)
in the sample plane. The fluorescence emission is collected by the
objective and passes through in the epi descanned detection path
before being reflected by a dichroic mirror (FF720-SDi01, Semrock).
The remaining laser light is further rejected by interferential filters
(610/70 nm or 1000/50 nm), and the emission is finally focused on an
avalanche photodiode working in the photon-counting mode. The
lateral resolution of the imaging setup is about 300 nm.
Computational Details. All DFT calculations were performed
with Gaussian 03.44 Calculations were carried out at the DFT level of
theory using the hybrid functional B3PW91.45−51 Geometry
optimizations were achieved without any symmetry restriction.
Calculations of vibrational frequencies were systematically done in
order to characterize the nature of stationary points. Stuttgart effective
core potentials and their associated basis set were used for silicon,
yttrium, and ytterbium. The basis sets were augmented by a set of
polarization functions (ζd = 0.284 for Si and ζf = 1.0 for Y and Yb).
Hydrogen, carbon, nitrogen, and oxygen atoms were treated with 6-
31G(d,p) double-ζ basis sets.52
Preparation of [Y(N(SiMe3)2)3@SiO2]. A representative proce-
dure is as follows: A mixture of 2.00 g of dehydroxylated silica
(SiO2−700; 0.52 mmol isolated SiOH, 1 equiv) and 325 mg (0.57
mmol, 1.1 equiv) of Y(N(SiMe3)2)3 was stirred in pentane at room
temperature for 3 h. The reaction mixture was filtered, and the solid
residue was washed five times with pentane (18 mL) and dried under a
high vacuum (10−5mbar) at room temperature for 1 h, yielding a white
solid. 1H MAS NMR (400 MHz): δ/ppm 0.14, −0.17, −0.55. 13C CP
MAS NMR (100 MHz): δ/ppm 1. 29Si CP MAS NMR (80 MHz): δ/
ppm 16, 5, −11, −93,−101 ppm. IR ν(CH3): 2951, 2900 cm−1.
Elemental analysis: Y, 1.83%wt; C, 3.64%wt; N, 0.64%wt.
Preparation of [Yb(N(SiMe3)2)3@SiO2]. This material was
prepared using the procedure described above with Yb(N(SiMe3)2)3
(373 mg, 0.57 mmol, 1.1 equiv). IR ν(CH3): 2951, 2900 cm−1.
Elemental analysis: Yb, 3.95%wt; C, 3.55%wt; N, 0.76%wt.
Preparation of [Y@SiO2]. A representative procedure is as
follows: The white [Y(N(SiMe3)2)3@SiO2] (1 g, 0.21 mmol Y) was
heated at 500 °C at a heating rate of 5 K/min for 12 h at 10−5 mbar.
1H MAS NMR (400 MHz): δ/ppm 0.05 (OSiMe3). 13C CP MAS
NMR (100 MHz): δ/ppm 0 (OSiMe3). IR ν(OSiMe3): 2963, 2905
cm−1. Elemental analysis: Y, 2.10%wt; C, 1.36%wt; N, 0.40%wt.
Preparation of [Yb@SiO2]. The sample was prepared as described
above. IR ν(SiMe3): 2963, 2905 cm−1. Elemental analysis: Yb, 2.10%wt;
C, 1.36%wt; N, 0.40%wt.
Preparation of [DEAS·Y@SiO2]. A mixture of 0.50 g of [Y@SiO2]
(0.10 mmol, 1 equiv) was reacted in dichloromethane with 55 mg
(0.105 mmol, 1 equiv) of DEAS for 4 h. The solid became
immediately orange. After 3 h, the solid was washed 5 times with
CH2Cl2, and the residue was dried at 10−5 mbar. 1H MAS NMR (400
MHz): δ/ppm 7.3, 6.6, 3.1, 1.2. 13C CP MAS NMR (100 MHz) δ/
ppm, note the maxima are reported: 156, 154, 149, 129, 121, 109, 45,
13, 0 (SiMe3). IR: ν(CC) 1591 cm−1. Elemental analysis: Y,
2.05%wt; C, 4.72%wt; N, 1.01%wt.
Preparation of [DEAS·Yb@SiO2]. This material was prepared as
described above. IR: ν(CC) 1591 cm−1. Elemental analysis: Yb,
3.97%wt; C, 4.72%wt; N, 0.99%wt.
Preparation of Y(OSi(OtBu)3)3(η2-HOSi(OtBu)3). To a solution
of 2.00 g of Y(N(SiMe3)2)3 (3.38 mmol, 1.0 equiv) in 20 mL of
CH2Cl2, a solution of 4.0 g of silanol (7.6 mmol, 4 equiv) in 20 mL of
CH2Cl2 was added. After stirring for 8 h, half of the solvent was
evaporated under reduced pressure, and the solution was cooled at
−40 °C, after one day large transparent blocks were obtained. The
crystals were collected by filtration, dried under reduced pressure, and
1
stored under argon. Yield: Y = 75% (2.70 g). H NMR (400 MHz,
toluene d8): δ/ppm 1.55; and minor species (7%), 1.58. 13C NMR
(100 MHz, toluene d8): δ/ppm 73.1 (CtBu), 31.8 (CH3tBu). Anal.
Calcd for C48H109O16Si4Y: C, 50.41%wt; H, 8.81%wt. Found: C,
50.38%wt; H, 8.81%wt.
Preparation of Yb(OSi(OtBu)3)3(η2-HOSi(OtBu)3). The com-
pound was prepared in 72% yield using the procedure above with
1
Yb(N(SiMe3)2)3 (2.00 g, 3.04 mmol, 1 equiv). H NMR (250 MHz,
C6D6): δ/ppm 1.60 (ν1/2 = 472 Hz, tBu); minor species (5%) and at
−7.64 (ν1/2 = 1309 Hz, tBu). Anal. Calcd for C48H109O16Si4Yb: C,
46.96%wt; H, 8.95%wt. Found: C, 46.52%wt; H, 8.85%wt.
Preparation of Y(OSi(OtBu)3)3(DEAS). To a solution of 0.50 g
(0.44 mmol, 1.0 equiv) of Y(OSi(OtBu)3)3(η2-HOSi(OtBu)3) in 20
mL of CH2Cl2, 222 mg of DEAS (0.44 mmol, 1 equiv) in 20 mL of
CH2Cl2 was added. After 8 h, the solvent was removed under reduced
pressure, and the HOSi(OtBu)3 was removed by sublimation at 40 °C
under 10−5 mbar. The resulting material was dissolved in CH2Cl2 and
concentrated until a fine orange precipitate began to form. This
solution was stored at −40 °C overnight, and the solid was collected
by filtration, washed with 10 mL of hexamethyldisiloxane, and dried
under reduced pressure. The yield was 81% (503 mg). 1H NMR (300
MHz, C6D6): δ/ppm 10.05 (d, J = 5.5 Hz, 2H), 7.65 (d, J = 5.8 Hz,
2H), 7.69 (s, 2H), 6.76 (d, J = 15.7 Hz, 2H), 7.37 (d, J = 8.1 Hz, 2H),
6.55 (d, J = 8 Hz, 2H), 7.40 (d, J = 4 Hz, 2H), 2.93 (m, J = 6.7 Hz,
2H), 0.86 (t, J = 5.7 Hz, 2H), 1.61 (s, 81 H). 13C NMR (75 MHz,
C6D6): 153.8, 150.1, 149.5, 148.6, 135.6, 129.2, 122.9, 120.5, 119.6,
117.4, 111.6, 71.5, 44.6, 32.6, 12.8. IR: ν(CC) 1591 cm−1. Anal.
Calcd C70H119N4O12Si3Y: C, 60.84%wt; N, 4.05%wt; H, 8.68%wt.
Found: C, 60.35%wt; N, 4.02%wt; H, 8.65%wt.
Preparation of Yb(OSi(OtBu)3)3(DEAS). The compound was
prepared in 78% yield using the procedure described above starting
from 0.50 g (0.40 mmol) of Yb(OSi(OtBu)3)3(η2-HOSi(OtBu)3) and
1
201 mg (0.40 mmol) of DEAS. H NMR (200 MHz, C6D6): δ/ppm
−0.74 (ν1/2 = 97 Hz, 81 H, tBuO), 57.8 (ν1/2 = 352 Hz, 2 H, ArH),
19.2 (ν1/2 = 250 Hz, 2 H, ArH), −38.8 (ν1/2 = 252 Hz, 2 H, ArH). IR:
ν(CC) 1618 cm−1. Anal. Calcd C70H119N4O12Si3Yb: C, 56.91%wt;
N, 3.82%wt; H, 8.18%wt. Found: C, 57.35%wt; N, 3.82%wt; H, 8.15%wt.
Preparation of Y(OSi(OtBu)3)3(4,4′-Me2bipy). The compound
was prepared in 77% yield using a similar procedure used for the
preparation of the DEAS complex from 1.00 g (0.88 mmol, 1 equiv) of
YOSi(OtBu)3)3(η2-HOSi(OtBu)3) and 4,4′-Me2bipy (162 mg, 0.88
1
mmol, 1 equiv). H NMR (500 MHz, C6D6) δ/ppm: 9.89 (d, J = 5.4
Hz, 2H), 7.14 (d, J = 4.1 Hz, 2H), 7.06 (s, 2H), 1.81 (s, 6H), 1.56 (s,
81H). 13C NMR (125 MHz, C6D6): δ/ppm 154.2, 153.3, 151.4, 126.3,
121.4, 71.5, 32.6, 21.6. IR: ν(CC) 1618 cm−1. Anal. Calcd
C48H93N2O12Si3Y: C, 54.21%wt; N, 2.63%wt; H, 8.81%wt. Found: C,
54.35%wt; N, 2.68%wt; H, 8.81%wt.
Preparation of Yb(OSi(OtBu)3)3(4,4′-Me2bipy). The compound
was prepared in 78% yield (728 mg) using the procedure described
above starting from 1.00 g (0.81 mmol, 1 equiv) of YbOSi(OtBu)3)3(η
1
2-HOSi(OtBu)3) and 4,4′-Me2bipy (149 mg, 0.81 mmol, 1 equiv). H
NMR (250 MHz, C6D6): δ/ppm 57.2 (ν1/2 = 41.4 Hz, 2 H), 9.9 (ν1/2
= 44 Hz, 6H), −0.54 (ν 1/2 = 109 Hz, 27 H), −107.1 (ν 1/2 = 217 Hz,
2 H). IR: ν(CC) 1618 cm−1. Anal. Calcd C48H93N2O12Si3Yb: C,
50.24%wt; N, 2.44%wt; H, 8.17%wt. Found: C, 49.97%wt; N, 2.38%wt; H,
8.15%wt.
1064
dx.doi.org/10.1021/cm404140q | Chem. Mater. 2014, 26, 1062−1073

Chemistry of Materials
Article
The IR spectrum also contains low intensity bands in the
3464 and 3534 cm−1 region that are associated with N−H
vibrations, which can be attributed to strongly adsorbed
compounds such as (Me3Si)3−xN(H)x (0 < x ≤ 3), generated
upon reaction of Y(N(SiMe3)2)3 with surface silanols (eq 1).
The IR spectrum of [Yb(N(SiMe3)2)3@SiO2] shows very
similar features to those of [Y(N(SiMe3)2)3@SiO2] (Figure
S5B).
3. RESULTS AND DISCUSSION
The synthesis of silica particles functionalized with chromo-
phores bound to the surface Ln isolated site is performed in a
three-step procedure as shown in Scheme 1. The silica
Scheme 1. (a) Grafting of Ln(N(SiMe3)2)3 on SiO2−700 to
Give [Ln(N(SiMe3)2)3@SiO2], Thermal Treatment to Give
the Tripodal Species, and Coordination of the bipy Ligand
and (b) DEAS Ligand (left) and 4,4′-Me2bipy (right)
a
[Y(N(SiMe3)2)3@SiO2] contains 1.83%wt Y according to
elemental analysis, which corresponds to 0.21 mmol Y/g of
silica (Table 1, entry 1). This value is slightly lower than
Table 1. Ln, C, and N Elemental Analysis of Y and Yb-Based
Materials after Grafting, Post-Treatment, and Reaction with
DEAS
element loading (wt %)
ratio
entry
1
materials
Ln
C
N
C/Ln
15
N/Ln
[Y(N(SiMe3)3)@
SiO2]
1.83
3.64
0.64
1
2
1
2
[Yb(N(SiMe3)3)
@SiO2]
3.95
3.55
0.76
15
1
2
1
3
4
5
6
[Y@SiO2]
2.10
4.14
2.05
1.36
1.58
4.72
4.72
0.42
0.31
1.01
0.99
5
5
1
1
1
1
0.5
0.5
1.9
1.9
1
1
1
1
[Yb@SiO2]
[DEAS·Y@SiO2]
[DEAS·Yb@SiO2] 3.95
14.5
14.5
expected for quantitative grafting (0.26 mmol SiOH/g for
[SiO2−(700)]), consistent with the competitive reaction of
HN(SiMe3)2 with silanols. Elemental analyses for carbon
(3.64%wt) and nitrogen (0.64%wt) indicate that 15 1 C and
2
1 N are still present per Y atom, which is consistent with
the presence of SiO−Y(N(SiMe3)2)2 (2 N per Y and 12 C
per Y) along with minor amounts of SiO−SiMe3 (ca. 0.05
mmol per gram, 20% of the silanol sites). [Yb(N(SiMe3)2)3@
SiO2] affords similar analytical data: 3.95%wt Yb, 3.55%wt C, and
0.76%wt N (15 C and 2 N per Yb), which corresponds to 0.21
mmol of SiO−Yb(N(SiMe3)2)2 g−1 along with 0.05 mmol·
g−1 of SiO−SiMe3 (Table 1, entry 2). The elemental analysis
data indicate that ca. 360 Ln units are present per 12 nm silica
particle. [Y(N(SiMe3)2)3@SiO2] was also characterized by solid
state NMR. The magic angle spinning (MAS) 1H NMR
spectrum contains a signal centered at 0 ppm and a lower
intensity resonance at 0.44 ppm assigned to the grafted 
SiO−Y(N(SiMe3)2)2 and SiO−SiMe3, respectively, along
with an extra peak at 0.87 ppm that is assigned to the adsorbed
(Me3Si)3−xN(H)x (Figure S6). The carbon-13 cross-polar-
ization magic angle spinning (CPMAS) NMR spectrum shows
only one peak centered at 1 ppm, which can be assigned to the
Me3Si fragment of both SiO−Y(N(SiMe3)2)2 and SiO−
SiMe3 surface species. The 29Si CPMAS spectrum shows the
expected peaks at −92 ppm and −101 ppm corresponding to
Q3 and Q4 sites of silica,55 and three peaks at −11, 5, and 16
ppm, which are assigned to Y−N(SiMe3)2, (Me3Si)3−xN(H)x,
and OSiMe3, respectively.
a
In both species the carbon and nitrogen atoms are labeled.
nanoparticles pretreated at 700 °C under vacuum conditions,
having a low density of surface silanols ([SiO2−700], 0.8
OH.nm−2, 0.26 mmol/g), are brought into contact with
Ln(N(SiMe3)2)3 (Ln = Y and Yb) yielding [Ln(N(SiMe3)3@
SiO2]. The materials were heated at 500 °C for 12 h under high
vacuum conditions to give [Ln@SiO2] and were subsequently
brought into contact with 4,4′-diethylaminostyryl-2,2′-bipyr-
idine (DEAS) to yield [DEAS·Ln@SiO2].
3.1. Synthesis and Characterization of Surface
Species. Silica Functionalization[Ln(N(SiMe3)2)3@SiO2].
The reaction of Y(N(SiMe3)2)3 with [SiO2−700] in pentane
yields the off-white solid [Y(N(SiMe3)2)3@SiO2]. The infrared
spectrum of [Y(N(SiMe3)2)3@SiO2] (Figure S5A) lacks a
strong silanol vibration and contains νCH bands at 2951 and
2900 cm−1 and a shoulder at 2820 cm−1, attributed to the
SiMe3 groups from SiO−Y(N(SiMe3)2)2 and SiO−SiMe3
surface species (eqs 1 and 2). The SiO−SiMe3 surface
species results from the formation of HN(SiMe3)2 (eq 1) and
the subsequent competitive reaction with silanols (eq 2).53,54
Stabilizing Thermal TreatmentGeneration of [Ln@SiO2].
Treatment of [Ln(N(SiMe3)2)3@SiO2] at high temperatures
(500 °C) under vacuum conditions (10−5 mbar) results in a
sharp decrease in intensity of the νCH bands at 2951 and 2900
cm−1 after 8 h. Note that a νOH signal is not regenerated during
this process, but the SiO−SiMe3 surface species are still
present as evidenced by the remaining C−H vibrations at 2963
and 2905 cm−1 (Figures S5A and S7). Elemental analysis of
Ln(N(SiMe3)2)3 +  SiO−H
→ (Me3Si)2NH +  SiO−Ln(N(SiMe3)2
(1)
(Me3Si)2N−H + 2  SiO−H
→ NH3 + 2  SiO−SiMe3
(2)
1065
dx.doi.org/10.1021/cm404140q | Chem. Mater. 2014, 26, 1062−1073

Chemistry of Materials
Article
[Y@SiO2] gives 2.10%wt Y, 1.36%wt C, and 0.42%wt N,
consistent with a constant amount of Y and a loss of most
organic functionalities (Table 1, entry 3). The absolute fate of
the silylamide during this treatment remains unknown, but
elemental analysis shows that most carbon and nitrogen are
eliminated in [Y@SiO2], albeit small quantities of carbon and
nitrogen persist (5 C/Ln and 0.5 N/Ln). Characterization of
[Y@SiO2] by solid-state NMR shows that most surface organic
functionalities have disappeared and that only SiO−SiMe3
groups remain at the surface as evidenced by the presence of a
shell, which are presumably siloxane bridges. Because of the
uncertainty in the coordination numbers, there may be little
actual difference between the coordination environment of
[Yb@SiO2] and [Y@SiO2] other than the M−O bond
distances, which are slightly shorter for Yb than for Y.58
Coordination of DEAS to [Ln@SiO2]Formation of
[DEAS·Ln@SiO2]. Contacting the off-white [Ln@SiO2] with a
yellow solution of DEAS in CH2Cl2 results in a fast
discoloration of the supernatant and yields [DEAS·Ln@SiO2]
as deep red solids (Scheme 1 and eq 5).
1
single peak in the H MAS spectrum at 0.05 ppm, and the
( SiO)3Ln + DEAS → [DEAS·Ln@SiO2]
(5)
resonance at 0.0 ppm in the 13C CPMAS NMR spectrum
(Figure S8). It is likely that during the thermal treatment the
−N(SiMe3)2 fragment is transferred to the surface by opening
an adjacent siloxane bridge, thus forming new Y−O bonds and
Si−N(SiMe3)2 (eqs 3 and 4),56 the latter presumably further
decomposing into SiNHx surface species.57 Similar observations
are obtained for the thermal treatment of the [Yb(N-
(SiMe3)2)3@SiO2] to form [Yb@SiO2], albeit at a slightly
lower decomposition temperature (Figure S7). Elemental
analysis of [Yb@SiO2] gives 4.14%wt Y, 1.58%wt C, and
0.31%wt N (Table 1, entry 4).
The IR spectrum is almost identical for both lanthanides: the
orange-red [DEAS·Y@SiO2] and deep-red [DEAS·Yb@SiO2]
contain new bands associated with DEAS. In particular, the two
bands at 1522 and 1586 cm−1 attributed to ν(CC) are blue-
shifted by 15 cm−1 with respect to free DEAS, evidence for
coordination of DEAS to the metal center (Figure S10).28 The
same shift is observed in model molecular complexes
Ln(OSi(OtBu)3)3(DEAS) (vide infra). Elemental analysis of
[DEAS·Y@SiO2] gives 2.05%wt Y, 4.72%wt C, and 1.01%wt
N
(Table 1, entry 5). With respect to [Y@SiO2], the nitrogen and
carbon contents increase from 0.38 to 1.01%wt and 1.36 to
4.72%wt, respectively, which corresponds to an increase of 1.9

SiO−Ln(N(SiMe3)2)2 +  Si−O−Si
→ ( SiO)2Ln(N(SiMe3)2) +  Si−N(SiMe3)2
1 N and 14.5
1 C per Y atom. For [DEAS·Yb@SiO2],
(3)
elemental analyses (3.95%wt Yb, 0.99%w N, and 4.72%wt C) also
show a similar increase of 2 1 N and 14.5 1 C per Yb
(Table 1, entry 6). Since DEAS contains 28 C and 4 N atoms,
these increases are consistent with the coordination of DEAS to
ca. 50% of the Ln sites. From these data, we can conclude that
the particles contain ca. 180 DEAS·Ln fragments per 12 nm
particle. The non-complete coordination of all the lanthanide
centers (ca. 1 Ln nm−2) is likely due to the size of the DEAS
ligand. The projected area of DEAS is ca. 2 nm2, large enough
that a silica nanoparticle cannot accommodate more than 0.5
DEAS nm−2. Similar behavior was observed for [DEAS·Zn@
SiO2].28
( SiO)2Ln(N(SiMe3) +  Si−O−Si
→ ( SiO)3Ln +  Si−N(SiMe3)2
(4)
We characterized the [Ln@SiO2] materials using extended
X-ray absorption fine structure (EXAFS) analysis (Table 2 and
Table 2. Fitted EXAFS Data for [Y@SiO2] and [Yb@SiO2]
[Y@SiO2]
[Yb@SiO2]
ΔE (eV)
2(1)
0.9
4(2)
0.9
S2
0
The 1H solid-state NMR spectrum of [DEAS·Y@SiO2]
displays additional signals at 7.0 (CHar), 3.2 (NCH2CH3), and
0.94 (NCH2CH3), consistent with the presence of DEAS. The
13C CPMAS solid-state NMR spectrum show typical peaks
associated with the coordination of DEAS to Y as evidenced by
the downfield shift of the methine carbon in alpha of the
pyridine nitrogen (carbon-6 in Scheme 1b, Figure S11) from
155 ppm in free DEAS to 158 ppm in [DEAS·Y@SiO2].28 For
[DEAS·Yb@SiO2], solid-state NMR analyses were not
performed because of the paramagnetic Yb center.
3.2. Molecular and Computational Models. The
synthesis and the characterization of molecular analogues of
the surface species generally results in a deeper understanding
of the surface species.59−63 HOSi(OtBu)3 is a useful model of
an isolated silanol since the silicon is surrounded by four
oxygen atoms, as in silica, three of which are −OtBu groups and
one is the −OH reactive site.64 Additionally, HOSi(OtBu)3 can
also accommodate secondary interactions between the metal
center and the oxygen of the OtBu groups.65 Since the −OtBu
ligands are rather bulky, monomeric lanthanide siloxides are
expected and would serve as good models for the silica surface.
The targeted molecular model complexes are Ln(OSi-
(OtBu)3)3(DEAS). We also synthesized the Ln(OSi-
(OtBu)3)3(4,4′-Me2bipy) for diffraction studies due to the
reluctance of DEAS−metal complexes to form X-ray quality
crystals.
neighbor (#)
O (3)
O (3)
a
a
O
(2)
O
(4)
distance Ln−O (Å)
2.211
2.11
a
distance Ln−O (Å)
2.40
2.30
σ2 Ln−O(Å2)
0.0051(6)
0.0036(9)
<0.001
<0.001
0.009(2)
0.013(3)
0.01
a
σ2 Ln−O(Å2)
b
p
Ln−O
Ln−O
b
a
p
0.002
a
b
Second oxygen neighbor. Probability that the improvement to the fit
due to adding this set of atoms is due to random error. A p < 0.01
means that the improvement to the fit is better than 3 standard
deviations.
Figure S9). For [Y@SiO2], the EXAFS data are consistent with
a yttrium atom surrounded by three oxygen atoms at a short
distance of 2.221(8) Å, and two additional O neighbors at
2.40(1) Å. These values indicate that the yttrium atom is
bonded to three oxygen atoms and also coordinated to nearby
siloxane bridges, (SiO)3Y(SiOSi)2. For [Yb@SiO2], the
Yb ion is also surrounded by three oxygen atoms at short
distances (2.11(2) Å) but interacts with more siloxane bridges
at longer distances, (SiO)3Yb(SiOSi)4. It should be
noted that coordination numbers determined by EXAFS are
strongly correlated with the thermal Debye−Waller factor (σ2),
which results in large uncertainties. In the case of [Ln@SiO2],
this effect is most pronounced for the second oxygen atom
1066
dx.doi.org/10.1021/cm404140q | Chem. Mater. 2014, 26, 1062−1073

Chemistry of Materials
Article
The addition of 4 equiv. of HOSi(OtBu)3 to Ln(N-
(SiMe3)2)3 in CH2Cl2 gives Ln(OSi(OtBu)3)3(η2-HOSi-
(OtBu)3) in good yields (Ln = Y, 75%; Ln = Yb, 72%; Scheme
2). For Y and Yb, the isolated siloxides crystallize with an
(OtBu)3) in toluene d8 solution, we can estimate that the
binding constant of silanol to Y(OSi(OtBu)3)3 is roughly 100.
The 1H NMR spectra of Yb(OSi(OtBu)3)3(η2-HOSi-
(OtBu)3) also show two broadened resonances, associated
with two types of OtBu groups in a ratio of 95:5 at 20 °C at a
50 mM concentration attributed to Yb(OSi(OtBu)3)3(η2-
HOSi(OtBu)3) and Yb(OSi(OtBu)3)3, respectively, as pro-
posed for the corresponding Y complexes (Figures S17−S18).
The addition of bipyridine ligands to Ln(OSi(OtBu)3)3(η2-
HOSi(OtBu)3) in CH2Cl2 results in the displacement of
HOSi(OtBu)3 and coordination of the ligand. The silanol was
removed by vacuum sublimation, and the resulting residue was
crystallized from CH2Cl2. The bipyridine adducts were isolated
in good yields (77−81%). For the diamagnetic yttrium
Scheme 2. Synthesis of Molecular Precursors
Ln(OSi(OtBu)3)3(η2-HOSi(OtBu)3) (Ln = Y or Yb) and
Subsequent Synthesis of the bipy-based Molecular Models,
Where the bipy Ligands Are DEAS and 4,4′-Me2bipy
1
compound, the H and 13C NMR spectra for 4,4′-Me2bipy or
DEAS adducts are consistent with a 1:1 stoichiometry in which
the OtBu groups are equivalent at 20 °C (Figures S19−S22).
At −90 °C, the peak corresponding to the −OtBu groups
broadens (ν1/2 (RT) = 3 Hz to ν1/2 (−90 °C)= 7 Hz), and
lowering the temperature to −120 °C results in further
broadening without resolution into two distinct signals (Figure
S23).
additional molecule of silanol, which accounts for the
stoichiometry of the reaction. When a stoichiometry of 1:3 is
used under the same conditions, only Ln(OSi(OtBu)3)3(η2-
HOSi(OtBu)3) was isolated, though in lower yield. The solid-
state structures for both siloxides were determined by X-ray
crystallography and are shown in Figure 1. The compounds are
isostructural, and the metals are five-coordinate, the silanol
behaving as a bidendate ligand through lone pairs from the
−OH and the −OtBu groups.
1
The H NMR spectra for the ytterbium compounds are
broad, and resonances are paramagnetically shifted (Figures
S24 and S25). The chemical shifts are given in the
Experimental Section. The IR spectra of Ln-
(OSiOtBu)3)3(DEAS) and Ln(OSiOtBu)3)3(4,4′-Me2bipy) in
KBr show a blue shift of the CC band of 14 cm−1 (1591
cm−1) and 27 cm−1 (1616 cm−1), with respect to the free ligand
(1577 cm−1 for DEAS and 1589 cm−1 for 4,4′-Me2bipy), in
good agreement with the values obtained for the surface species
(Figures S10 and S26).
The X-ray structures of the 4,4′-Me2bipy adducts are shown
in Figure 2. The solid-state crystal structures of Ln(OSi-
(OtBu)3)3(η2-HOSi(OtBu)3) and Ln(OSi(OtBu)3)3(4,4′-
Me2bipy) allow us to evaluate the change in geometry around
the Ln(OSi(OtBu)3)3 core as the bidendate ligand is changed
from HOSi(OtBu)3 to 4,4′-Me2bipy. The relevant distances
and angles are listed in Table 3, and further crystallographic
Figure 1. (a) Single crystal structure of Y(OSi(OtBu)3)3(η2-HOSi-
(OtBu)3). (b) Yb(OSi(OtBu)3)3(η2-HOSi(OtBu)) (right), ellipsoids
at 50% of probability. Hydrogen atoms and methyl (tBu) groups are
omitted for clarity.
The 1H NMR spectra of Y(OSi(OtBu)3)3(η2-HOSi(OtBu)3)
in toluene d8 is concentration dependent and gives rise to two
resonances at δH = 1.51 and δH = 1.54 ppm in an area ratio of
20:1 (Figure S12); the minor species is not detected in the 13
C
NMR (Figure S13). At a low concentration (2 mM), the minor
species increases while the signal of the major species decreases
and shifts to a slightly lower field (i.e., toward free silanol,
Figure S14). These results suggest a dissociative equilibrium
between Y(OSi(OtBu)3)3(η2-HOSi(OtBu)3) ⇌ Y(OSi-
(OtBu)3)3 + HOSi(OtBu)3. Consistent with this proposal,
adding an equivalent of silanol to a solution of complex causes
the disappearance of the resonance at δH = 1.54 and an upfield
shift of the δH = 1.51 ppm toward the free ligand (Figure S15).
When the temperature is lowered to −40 °C, a resonance due
to the free silanol appears at δH = 1.45 ppm (Figure S16). From
the concentration dependence of Y(OSi(OtBu)3)3(η2-HOSi-
Figure 2. (a) Single crystal structure of Y(OSiOtBu)3)3(4,4′-Me2bipy).
(b) Yb(OSiOtBu)3)3(4,4′-Me2bipy). Ellipsoids are shown at 50%
probability, and hydrogen atoms and methyl (tBu) groups are omitted
for clarity. (c) Calculated structure of Y(OSiOtBu)3)3(4,4′-Me2bipy)
and (d) Y(OSiOtBu)3)3(DEAS).
1067
dx.doi.org/10.1021/cm404140q | Chem. Mater. 2014, 26, 1062−1073

Chemistry of Materials
Article
Table 3. Summary of the Bond Angles and Distances from X-Ray Crystallography and DFT Calculations
a
b
From the single crystal. From calculations.
information is available in the Supporting Information. The
geometries of the HOSi(OtBu)3 and 4,4′-Me2bipy adducts of
both metals are similar, as expected on the basis of the close
similarity in the metal radii. The geometry is more or less a
square pyramid, where the ligands in the basal plane are
nitrogen from 4,4′-Me2bipy or oxygen from η2-HOSi(OtBu)3.
The other two oxygen atoms of the basal planes and the apical
oxygen are from −OSi(OtBu)3 groups. In all cases, the four
ligands in the basal plane are bent away from the apical ligand.
The dihedral angle would be 0° for an ideal square based
pyramid, where the central atom is in the same plane of the
ligands. In the case of the Ln(OSi(OtBu)3)3(η2-HOSi(OtBu)3),
the dihedral angles between the planes defined by the O1−Ln−
O2 atoms and O4−Ln−O5 are 61° for Y and 67° for Yb,
leading a strong distortion, while in the case of the 4,4′-
Me2bipy adducts, the angle between the planes defined by the
N1−Ln−N2 and O1−Ln−O2 atoms are 42° for Y-
(OSiOtBu)3 )3 (4,4′-Me2 bipy) and 40° for Yb-
(OSiOtBu)3)3(4,4′-Me2bipy). In general, the apical M−O
distance is shorter than those of the basal plane, as usually
observed in square pyramidal geometries.
DFT calculations were carried out on the molecular
analogues, in particular because no suitable crystals could be
grown for the DEAS derivatives. First, the calculated structure
for Ln(OSi(OtBu)3)3(4,4′-Me2bipy) has a similar geometry to
that found experimentally in the solid state and with almost
identical distances and angles of the single crystal structure
(Table 3 and Figure 2). The calculated ΔG298 values of reaction
for Ln(OSi(OtBu)3)3(η2-HOSi(OtBu)3) with 4,4′-Me2-bipy to
give Ln(OSi(OtBu)3)3(4,4′-Me2bipy) are −11.0 kcal/mol and
−11.9 kcal/mol for Yb and Y, respectively (vide infra),
consistent with an exoergic binding of the bipyridine derivatives
to the metal center. The DEAS adduct shows similar behavior
in terms of energetics and structural features with slightly
longer Ln−N bonds (Ln = Y; 2.528 vs 2.514; Ln = Yb; 2.537 vs
2.475 Å, respectively).
These molecular studies show that lanthanide siloxides
favorably bind bipyridines to yield the corresponding
monoadducts that were fully characterized by IR and NMR
spectroscopies, diffraction studies, and their geometries
computed with DFT methods. In view of the close similarities
of IR and NMR data obtained for the molecular and surface
species, it is thus possible to ascertain the structure of the
surface species at a molecular level, with Ln surface sites bound
to a bipyridine ligand.
3.3. Photophysical Properties. DEAS is known to display
a charge transfer transition (CT) that is due to the
diethylamino donation to the pyridyl acceptor, which is
responsible for the absorption, emission, and nonlinear optical
properties of related complexes.37,38,66 This transition is absent
in 4,4′-Me2bipy.
The photophysical studies of the molecular model complexes
Ln(OSi(OtBu)3 )3 (4, 4′-Me2 bipy) and Ln(OSi-
(OtBu)3)3(DEAS) were examined in the solid state and in
dilute pentane, toluene, or dichloromethane solution. These
measurements were compared to the surface species that were
acquired in the solid state.
In the solid state, Y(OSi(OtBu)3)3(DEAS) has an absorption
maximum at 450 nm, while the maximum is red-shifted to 465
nm for Yb, corresponding to a bathochromic shift of 40 and 55
nm for each respective lanthanide (Figure 3). In dichloro-
methane solution, Ln(OSi(OtBu)3)3(DEAS) displays an
absorption maximum and extinction coefficient [λmax; ε] at
[405 nm; 60 600 M−1 cm−1] and [412 nm; 68 000 M−1 cm−1]
for Y and Yb, respectively, corresponding to a much smaller
bathochromic shift of 6 and 13 nm relative to free DEAS than
found in the solid-state Ln(OSi(OtBu))3(4,4′-Me2bipy) (Ln =
Y and Yb): [308 nm; 16 700 M−1 cm−1] and [283 nm; 16 800
M−1 cm−1] for Y and Yb, respectively. The smaller bath-
1068
dx.doi.org/10.1021/cm404140q | Chem. Mater. 2014, 26, 1062−1073

Chemistry of Materials
Article
a binding constant close to that found for the molecular
analogues (1.7 × 103, Table 4, Figures S30 and S31).
The luminescence properties of the model complexes were
studied both in the solid state and in solution and compared to
those of the surface derivatives (Figures S32 and S33).
Irradiation of Y(OSi(OtBu)3)3(DEAS) at 440 nm in the solid
state results in the appearance of an intense emission band
assigned to a ligand CT fluorescence at 585 nm (Figure 4). In
Figure 3. Solid state absorption spectra of Yb(OSi(OtBu)3)3(DEAS)
(red solid), Y(OSi(OtBu)3)3(DEAS) (blue solid, and ) [DEAS·Ln@
SiO2] nanoparticles Ln = Y (blue dashed) and Ln = Yb (red dashed).
ochromic shift observed in solution for the Ln(OSi-
(OtBu)3)3(DEAS) complexes prompted us to determine the
binding constant of bipyridyl ligands to Ln(OSi(OtBu)3)3(η2-
HOSi(OtBu)3).
Titrating a solution of the bipyridyl ligand with a solution of
Ln(OSi(OtBu)3)3(η2-HOSi(OtBu)3) shows clear isosbestic
points independent of ligand and precursor combination
(Figure S27) and indicates that the species involved are related
linearly by stoichiometry. Using the method of continuous
variation (Job’s plots), we found a 1:1 stoichiometry for binding
of a bipyridyl ligand to Ln(OSi(OtBu)3)3(η2-HOSi(OtBu)3).67
The binding constants are calculated using a one-site binding
model (eq 6), where Bmax indicates the saturation point of the
curve and KD, the dissociation constant; the binding constant
(Keq) is the inverse of KD (eq 7). Plots for Keq determination
and Job’s plot are shown in Figures S28 and S29.
Bmax
x
y =
Figure 4. (a) Solid state emission spectra of Ln(OSi(OtBu)3)3(DEAS)
(Ln = Y (blue); Yb (red)) zoom × 200; normalized NIR emission
measured at room temperature (thick red; ∗ hot band) and 77 K (thin
red). (b) Solid state emission spectra of [DEAS·Ln@SiO2] nano-
particles. Ln = Y (blue line), Ln = Yb (red line). λex = 400 nm.
1 + KD
Keq = 1/KD
(6)
(7)
The binding constants of DEAS to “Ln(OSi(OtBu)3)3”
(Table 4) is about an order of magnitude greater than for Me2−
Bipy, and the binding of bipyridyl ligands to Yb is about an
order of magnitude stronger than for Y. These data indicate a
dissociation equilibrium in organic solvents. We also
determined the binding of DEAS to [Y@SiO2] by titration of
DEAS versus a constant quantity of [Y@SiO2].68 The resulting
curve was fitted with the same one-site binding model and gives
pentane, irradiation at 400 nm blue shifts this transition to 465
nm, with a second peak at 430 nm, a sign of the presence of
free DEAS due to the reversible coordination of DEAS as
discussed above. In the case of Yb(OSi(OtBu)3)3(DEAS), solid
state measurements show that a similar transition is observed
but with very weak intensity (Figure 4). This quenching is
explained by energy transfer to the lanthanide excited state
Table 4. Extinction Coefficients and Equilibrium Constant
for the Molecular Models
sensitizing the characteristic Yb(III) luminescence composed of
2
b
four bands partially overlapped and assigned to the F5/2
→
compound
λmax (nm)
ε−1
Keq (103)
2F7/2 transitions. The additional band at higher energy, called a
“hot band” at 940 nm (noted ∗ in Figure 4a), is attributed to
emission from higher crystal-field sublevel of the excited state.69
The thermally populated hot band disappears at 77 K, and the
Yb(III) emission spectrum is better resolved with four
transitions at 982, 1016, 1047, and 1075 nm that correspond
DEAS
403
283
408
436
308
308
440
40000
16300
60600
68000
16700
16800
4,4′-Me2bipy
Y(OSiOtBu)3)3 (DEAS)
Yb(OSiOtBu)3)3 (DEAS)
Y(OSiOtBu)3)3 (4,4′-Me2bipy)
Yb(OSiOtBu)3)3 (4,4′-Me2bipy)
3.6(2)
18(4)
0.4(3)
2.8(9)
1.7
2
a
to the degeneracy of the F7/2 ground state due to crystal field
[DEAS·Y@SiO2]
effects. This result allows us to establish the energy splitting
diagram of the MJ levels and to estimate the total splitting to be
ca. 880 cm−1. This behavior is identical for Yb(OSi-
a
λmax is calculated from recording the UV−vis spectrum in solid state
b
for [DEAS·Y@SiO2]. The binding is given at T = 298 K.
1069
dx.doi.org/10.1021/cm404140q | Chem. Mater. 2014, 26, 1062−1073

Chemistry of Materials
Article
Table 5. Emission and Lifetime of Molecular and Surface Lanthanides in Solution or in Solid State
entry
compound
[DEAS·Y@SiO2]
[DEAS·Yb@SiO2]
measurement
solid-state
λem (vis) [nm]
λem (NIR) [nm]
Δ (NIR) [cm−1
]
τem (NIR) [μs]
1
2
3
632
635
a
solid-state
981
982
983
2.9
b
Yb(OSi(OtBu)3)3(DEAS)
solid-state
toluene (10−5 mM)
582
880
790
5
b
523
6.4
4
5
Y(OSi(OtBu)3)3(DEAS)
solid-state
579
pentane (10−5 mM)
solid-state
464/490
338
b
Yb(OSi(OtBu)3)3(4,4′-Me2bipy)
983
984
860
800
5.8
b
pentane (10−5 mM)
325
6.6
a
b
First moment of time according to eq 12, τ1 = 1.9 μs. According to eq 8.
(OtBu)3)3(DEAS) and for Yb(OSi(OtBu)3)3(4,4′-Me2bipy) in
the solid state (Figures S32−S35).
of lifetimes centered at τ1 which correspond to a distribution of
sites on the surface.30
In solution, a similar behavior is observed, but due to the
relatively low binding, a peak in the free DEAS region is also
visible. It is well established that the splitting of the MJ state is
strongly correlated to the symmetry of the coordination
polyhedron, and therefore an identical splitting indicates a
similar structure for all molecular complexes.61 The total
splitting value (Δ = 880 cm−1 for Yb(OSi(OtBu)3)3(DEAS) in
the solid state) is very high and consistent with the low
symmetry coordination (see Supporting Information for a
representation of the MJ splitting, Figure S36).70−74 Finally, the
lifetime of the Yb(III) excited state was determined for all
molecular complexes in solution and in the solid state (Table
5). In each case, a perfect mono-exponential decay (eq 8) is
observed with a lifetime between 5 and 5.8 μs (Figures S37 and
S38).
The luminescence properties of the [DEAS·Ln@SiO2]
nanoparticles are shown in Figure 4. The CT emission is
centered at 635 nm ( 3 nm) for both compounds, a value
significantly red-shifted compared to the molecular models
(ΔλCT = 50 nm). In the case of [DEAS·Yb@SiO2], additional
characteristic Yb(III) NIR emission is observed in the NIR
spectral range. The spectrum presents the same profile as the
molecular compound, but it remains quite broad even at low
temperatures, suggesting the presence of several emitting
species whose symmetry is close to that of the model
compound. The lifetime of [DEAS·Yb@SiO2] is not well
described by a mono-exponential decay (eq 8), which is likely
due to the presence of several Yb(III) emitting sites at the
surface (Figure 4). A bi-exponential model (eq 9) provides a
better fit, with two different lifetimes, τ1 = 6 μs (40%) and τ2 =
1.3 μs (60%). Though the bi-exponential fit is reasonable, a
more realistic description is to assume a distribution of Yb(III)
emitting sites at the surface of the nanoparticle, and to fit the
distribution of lifetime decays by a stretched exponential model
(eq 10, Figure S39).75,76 This model provides a distribution of
decay times centered around τ1 with a width proportional to 1/
β. The dispersion factor β is dimensionless and varies between
0 and 1. β values close to unity indicate a narrow dispersion of
sites that would resemble to monoexponential decay. The
integration of the function gives the first moment of time ⟨τ⟩,
the average time constant (eq 11), which can also been
expressed as the quotient of the τ1, and the distribution by a
gamma function divided by the distribution (eq 12). The
quantity ⟨τ⟩ is the mean relaxation time. This model is
frequently used in disordered solids such as pyrene dyes grafted
on alumina surfaces,77 and more recently to understand the
TEMPO radical distribution in mesoporous silica materials.78 It
is particularly interesting because it describes a distribution (β)
y = A1 e−x/τ + y
1
(8)
0
y = A1 e−x/τ + A2 e−x/τ + y
1
2
(9)
0
β
y = A1 e(−x/τ ) + y
1
(10)
0
β
⟨τ⟩ = A1 e(−x/τ ) dx
1
∫
(11)
τ
⟨τ⟩ = 1 Γ
β
1
β
(12)
In the present case, an excellent fit was obtained with a τ1 of
1.7 μs, and β = 0.55, which reveals a broad distribution of sites.
The corresponding ⟨τ⟩ is equal to 2.9 μs (Table 5, entry 3),
which is close, albeit lower to that found for the molecular
species, showing that relaxation processes are faster for surface
species, probably because of the close proximity of the DEAS
ligands. The efficiency of the metal centered luminescence
(ηLn) can be estimated using the ηLn = τobs /τ0 relationship,
where τobs is the experimental lifetime of the Yb(III) and τ0 is
the purely radiative lifetime (0.27 ms for Yb(III)). This value
allows a raw estimation of the metal centered luminescence
quantum yield to 1.1%.
Finally, the two-photon cross-section of a [DEAS·Ln@SiO2]
nanoparticle is estimated using an additive model of the two-
photon cross-section of a soluble molecular model.24,25,28 For
practical reasons, this experiment cannot be carried out using
air and moisture sensitive Ln(OSi(OtBu)3)3(DEAS), and
consequently we use the value obtained for the tris-β-
diketonate analogues Ln(TTA)(DEAS), where TTA stands
for 2-thenoyltrifluoroacetonate.44 Both Yb and Y complexes
present identical two-photon absorption spectra with a maximal
cross-section of 230 20 GM at 830−850 nm. Using the
additive approximation for a particle which supports ca. 180
DEAS·Ln complexes, the two-photon cross section of the
[DEAS·Ln@SiO2] nanoparticles can be estimated to ca. 41.4 ×
103 GM, falling in the same range as other doped or grafted
nano-objects and quantum dots.79−81
2.4. Microscopy. The particles are densely functionalized
with luminescent molecular species, and it is possible to
investigate these materials using two-photon microscopy
imaging, using an excitation wavelength of 720 nm. The
images are obtained by dispersing [DEAS·Y@SiO2] or [DEAS·
Yb@SiO2] on a microscope coverslip. We use small amounts of
dispersed material with a sufficient dilution in order to separate
the particles from each other with an interparticle distance
larger than 1 μm. TEM (Transmission Electron Microscopy) of
1070
dx.doi.org/10.1021/cm404140q | Chem. Mater. 2014, 26, 1062−1073

Chemistry of Materials
Article
dispersed material shows small aggregates of 1−2 nanoparticles
(15−20 nm) to 6 nanoparticles (40−50 nm; Figure S40).
Similarly, dynamic light scattering experiments of dispersed
solutions show a maximum at 38.9 nm, consistent with small
aggregates of [DEAS·Ln@SiO2] (Figure S41).
Figure 5a shows two-photon excited images obtained for
[DEAS·Y@SiO2] detected around 610 nm, as well as [DEAS·
W/cm2. The efficiencies measured for these particles are very
close to the ones found in previously developed Zn(DEAS)
particles, in which there is a similar number of sites per
particle.30 Under such low excitation conditions, the particles
are easily detected and do not exhibit any visible photo-
bleaching. For comparison, 20 nm fluorescent nanospheres
(yellow-green (505/515) FluoSpheres, Invitrogen) emit an
average signal of 30 × 104 ph/s for the same incident intensity
conditions and adequate spectral excitation/detection con-
ditions. Finally, at 1000 nm (in the NIR), 33.6 × 104 ph/s were
collected on average for [DEAS·Yb@SiO2] particles. The
signal-to-noise ratio at this wavelength is still good, even though
the detectors are less efficient at this wavelength.
4. CONCLUSIONS
Silica nanoparticles with ca. 180 emitting surface Yb(III) DEAS
species absorb and emit in the NIR range with giant two-
photon cross-sections (40 × 103 GM). This level of brightness
allows for the detection of very small nanoparticle aggregates
(ca. 15−20 nm from TEM) using NIR-to-NIR confocal
biphotonic microscopy. Detailed characterization by EXAFS,
IR, solid-state NMR of the diamagnetic yttrium derivative, and
UV−vis established the structure of [(SiO)3Yb(DEAS)].
Complementary studies of Ln(OSi(OtBu)3)3(DEAS) as
molecular mimics of the surface environment revealed that
these models reproduce the properties found for [(
SiO)3Ln(DEAS)] in terms of binding constants (ca. 104) and
photophysical properties of the surface species, supporting the
structural assignment of the surface species. The luminescence
lifetime in [(SiO)3Yb(DEAS)] is best represented as a
distribution of sites, and the lifetime is slightly shorter than that
found in the molecular equivalents, which is probably due to
the heterogeneity of the amorphous silica surface environment.
The molecular approach presented here provides a “roadmap”
toward the preparation of luminescent nano-objects using
SOMC that allows direct comparison of structure−lumines-
cence properties, thus paving the way for the design of
improved nanoparticles for in-depth imaging of thick tissues.
Further works are currently underway in this direction.
Figure 5. (a) Two-photon fluorescence images of luminescent
nanoparticles excited at 760 nm, detected at 610 nm for [DEAS·Y@
SiO2] (top), [DEAS·Yb@SiO2] (middle), and 1000 nm detection for
[DEAS·Yb@SiO2] (bottom). (b) Corresponding statistics over about
200 particles and zoom on nanoparticles (pixel size: 60 nm). Incident
intensities: 4 × 106 W/cm2 (610 nm detection) and 6 × 106 W/cm2
(1000 nm detection).
Yb@SiO2] detected around 610 or 1000 nm using a
microscopic setup allowing detection at lower energies than
the laser excitation.19 Isolated spots are clearly visible with a
width of about 300 nm, characteristic of the diffraction limit
size of the microscope setup. This is consistent with physical
sizes or particles below ∼50 nm, sizes above this limit leading
indeed to an enlargement of the imaging spot. Occasionally,
larger spots were detected, due to aggregation of particles
during the deposition process; these spots were systematically
rejected for further analysis. To evaluate the average efficiency
of the particles (such as shown in the isolated spots of Figure
5b), the intensities from about 100 to 200 particles (taken in
about 10 images) are collected using an automatic identification
of isolated spots of a diameter below 6 pixels. The resulting
intensity histograms (Figure 5b) show relatively good
homogeneity of the particles’ efficiency and reflect their
reasonable size dispersion, the observed intensities varying by
up to a factor of 2 within the observed population (the intensity
is indeed proportional to the number of molecules in the focal
volume). The presence of various sizes, still below the limit
imaging size below 50 nm, is thus expected in the optical
microscopy samples. The optical measurements are therefore a
reasonable approach to evaluate the averaged efficiency of
deposited particles, for which average size seems to correspond
well to the objects observed in TEM.
ASSOCIATED CONTENT
* Supporting Information
■
S
NMR, IR, and UV−vis spectra, titrations, binding curves,
lifetime measurements, crystallographic information, and the
full list of authors for ref 44 are available. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail: ccoperet@inorg.chem.ethz.ch.
■
Present Address
⧫ETH Zurich, Department of Chemistry, Vladimir Prelog Weg
̈
2, CH-8093 Zurich, Switzerland
̈
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
A.B. and G.L. thank the Ministere de la Recherche et de
l’Education and the Swiss National Foundation
(SNF200021_137691/1), respectively, for pre-doctoral fellow-
ships. Portions of this work were supported by U.S.
Department of Energy, Basic Energy Sciences, Chemical
Averaged efficiencies are 19.3 × 104 photons/s for [DEAS·
Y@SiO2] (610 nm) and 24.4 × 104 ph/s for [DEAS·Yb@SiO2]
(610 nm), for an incident intensity at the focal spot of 6 × 106
1071
dx.doi.org/10.1021/cm404140q | Chem. Mater. 2014, 26, 1062−1073

Chemistry of Materials
Article
(29) Philippot, C.; Bourdolle, A.; Maury, O.; Dubois, F.; Boury, B.;
Brustlein, S.; Brasselet, S.; Andraud, C.; Ibanez, A. J. Mater. Chem.
2011, 21, 18613.
(30) Zagdoun, A.; Rossini, A. J.; Gajan, D.; Bourdolle, A.; Ouari, O.;
Rosay, M.; Maas, W. E.; Tordo, P.; Lelli, M.; Emsley, L.; Lesage, A.;
Sciences, Biosciences, and Geosciences Division and were
performed at Lawrence Berkeley National Laboratory under
Contract No. DE-AC02-05CH11231. Portions of this research
were carried out at the Stanford Synchrotron Radiation
Lightsource, a Directorate of SLAC National Accelerator
Laboratory and an Office of Science User Facility operated
for the U.S. Department of Energy Office of Science by
Stanford University. Electron Microscopy ETH Zurich
(EMEZ) is acknowledged for the TEM measurement.
Coper
(31) Coper
M. Angew. Chem., Int. Ed. 2003, 42, 129.
́
et, C. Chem. Commun. 2012, 48, 654.
et, C.; Chabanas, M.; Petroff Saint-Arroman, R.; Basset, J.-
́
(32) Thomas, J. M.; Raja, R.; Lewis, D. W. Angew. Chem., Int. Ed.
2005, 44, 6456.
(33) Tada, M.; Iwasawa, Y. Coord. Chem. Rev. 2007, 251, 2702.
(34) Marks, T. J. Acc. Chem. Res. 1992, 25, 57.
(35) Wegener, S. L.; Marks, T. J.; Stair, P. C. Acc. Chem. Res. 2011,
45, 206.
(36) Liang, Y.; Anwander, R. Dalton Trans. 2013, 42, 12521.
(37) Hilton, A.; Maury, O.; Le Bozec, H.; Ledoux, I.; Zyss, J. Chem.
Commun. 1999, 2521.
REFERENCES
■
(1) Prasad, P. N. In Introduction to Biophotonics; John Wiley & Sons,
Inc.: New York, 2004, p 203.
(2) Lakowicz, J. In Principles of Fluorescence Spectroscopy; Lakowicz, J.,
Ed.; Springer: New York: 2006, p 675.
(3) Nolan, E. M.; Lippard, S. J. Acc. Chem. Res. 2008, 42, 193.
(4) Kobayashi, H.; Ogawa, M.; Alford, R.; Choyke, P. L.; Urano, Y.
Chem. Rev. 2009, 110, 2620.
́ ̀
(38) Baccouche, A.; Peigne, B.; Ibersiene, F.; Hammoutene, D.;
Boutarfaïa, A.; Boucekkine, A.; Feuvrie, C.; Maury, O.; Ledoux, I.; Le
Bozec, H. J. Phys. Chem. A 2010, 114, 5429.
(5) Frangioni, J. V. Curr. Opin. Chem. Biol. 2003, 7, 626.
(6) Yuan, L.; Lin, W.; Zheng, K.; He, L.; Huang, W. Chem. Soc. Rev.
2013, 42, 622.
(39) Dragonetti, C.; Balordi, M.; Colombo, A.; Roberto, D.; Ugo, R.;
Fortunati, I.; Garbin, E.; Ferrante, C.; Bozio, R.; Abbotto, A.; Le Bozec,
H. Chem. Phys. Lett. 2009, 475, 245.
(7) Kiyose, K.; Kojima, H.; Nagano, T. Chem.Asian J. 2008, 3, 506.
(8) Nyk, M.; Kumar, R.; Ohulchanskyy, T. Y.; Bergey, E. J.; Prasad, P.
N. Nano Lett. 2008, 8, 3834.
(40) Feuvrie, C.; Maury, O.; Le Bozec, H.; Ledoux, I.; Morrall, J. P.;
Dalton, G. T.; Samoc, M.; Humphrey, M. G. J. Phys. Chem. A 2007,
111, 8980.
(9) Denk, W.; Strickler, J.; Webb, W. Science 1990, 248, 73.
(10) Pawlicki, M.; Collins, H. A.; Denning, R. G.; Anderson, H. L.
Angew. Chem., Int. Ed. 2009, 48, 3244.
(11) Kim, H. M.; Cho, B. R. Acc. Chem. Res. 2009, 42, 863.
(12) He, G. S.; Tan, L.-S.; Zheng, Q.; Prasad, P. N. Chem. Rev. 2008,
108, 1245.
(13) Campagnola, P. J.; Loew, L. M. Nat. Biotechnol. 2003, 21, 1356.
́
(14) Picot, A.; D’Aleo, A.; Baldeck, P. L.; Grichine, A.; Duperray, A.;
Andraud, C.; Maury, O. J. Am. Chem. Soc. 2008, 130, 1532.
(15) Zhang, T.; Zhu, X.; Cheng, C. C. W.; Kwok, W. M.; Tam, H.-L.;
Hao, J.; Kwong, D. W. J.; Wong, W. K.; Wong, K. L. J. Am. Chem. Soc.
2011, 133, 20120.
́
(41) Bourdolle, A.; Allali, M.; D’Aleo, A.; Baldeck, P. L.; Kamada, K.;
Williams, J. A. G.; Le Bozec, H.; Andraud, C.; Maury, O.
ChemPhysChem 2013, 14, 3361−3367.
(42) Bradley, D. C.; Ghotra, J. S.; Hart, F. A. J. Chem. Soc. 1973,
1021.
(43) Le Roux, E.; Chabanas, M.; Baudouin, A.; de Mallmann, A.;
Coperet, C.; Quadrelli, E. A.; Thivolle-Cazat, J.; Basset, J.-M.; Lukens,
́
W.; Lesage, A.; Emsley, L.; Sunley, G. J. J. Am. Chem. Soc. 2004, 126,
13391.
(44) Frisch, M. J. Gaussian, revision B.05 (see Supporting
Information for full citation).
(45) Perdew, J. P. Electronic Structure of Solids; Akademie Verlag:
Berlin, 1991.
(46) Perdew, J. P.; Burke, K.; Wang, Y. Phys. Rev. B 1996, 54, 16533.
(47) Burke, K.; Wang, Y.; Perdew, J. P. Electronic Density Functional
Theory: Recent Progress and New Directions; Plenum: New York, 1997;
p 11.
(48) Dobson, J. F.; Vignale, G.; Mas, M. P. Derivation of Generalized
Gradient Approximation: The PW91 Density Functional. Electronic
Density Functional Theory; Springer: New York, 1998; pp 81−111.
(49) Perdew, J. P.; Chevary, J. A.; Vosko, S. H.; Jackson, K. A.;
Pederson, M. R.; Singh, D. J.; Fiolhais, C. Phys. Rev. B 1992, 46, 6671.
(50) Perdew, J. P.; Burke, K.; Wang, Y. Phys. Rev. B 1998, 57, 14999.
(51) Perdew, J. P.; Chevary, J. A.; Vosko, S. H.; Jackson, K. A.;
Pederson, M. R.; Singh, D. J.; Fiolhais, C. Phys. Rev. B 1993, 48, 4978.
(52) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213.
(53) Anwander, R.; Roesky, R. Dalton Trans. 1997, 137.
(54) Gauvin, R. M.; Delevoye, L.; Hassan, R. A.; Keldenich, J.;
Mortreux, A. Inorg. Chem. 2007, 46, 1062.
(55) Burkett, S. L.; Sims, S. D.; Mann, S. Chem. Commun. 1996,
1367.
́
(56) Rataboul, F.; Baudouin, A.; Thieuleux, C.; Veyre, L.; Coperet,
C.; Thivolle-Cazat, J.; Basset, J. M.; Lesage, A.; Emsley, L. J. Am. Chem.
Soc. 2004, 126, 12541.
(57) Asefa, T.; Kruk, M.; Coombs, N.; Grondey, H.; MacLachlan, M.
J.; Jaroniec, M.; Ozin, G. A. J. Am. Chem. Soc. 2003, 125, 11662.
(58) Bauer, J. R.; Braunschweig, H.; Dewhurst, R. D. Chem. Rev.
2012, 112, 4329.
(59) Duchateau, R. Chem. Rev. 2002, 102, 3525.
(60) Quadrelli, E. A.; Basset, J. M. Coord. Chem. Rev. 2010, 254, 707.
(61) Edelmann, F. T.; Freckmann, D. M. M.; Schumann, H. Chem.
Rev. 2002, 102, 1851.
(16) Zhang, T.; Zhu, X.; Wong, W.-K.; Tam, H. L.; Wong, W. Y.
Chem.Eur. J. 2013, 19, 739.
(17) Massin, J.; Charaf-Eddin, A.; Appaix, F.; Bretonniere, Y.;
Jacquemin, D.; van der Sanden, B.; Monnereau, C.; Andraud, C. Chem.
Sci. 2013, 4, 2833.
(18) Piszczek, G.; Gryczynski, I.; Maliwal, B.; Lakowicz, J. J. Fluoresc.
2002, 12, 15.
(19) D Aleo, A.; Bourdolle, A.; Brustlein, S.; Fauquier, T.; Grichine,
́
A.; Duperray, A.; Baldeck, P. L.; Andraud, C.; Brasselet, S.; Maury, O.
Angew. Chem., Int. Ed. 2012, 51, 6622.
(20) Sabbatini, N.; Guardigli, M.; Lehn, J.-M. Coord. Chem. Rev.
1993, 123, 201.
(21) Kim, S.; Pudavar, H. E.; Bonoiu, A.; Prasad, P. N. Adv. Mater.
2007, 19, 3791.
(22) Kim, S.; Pudavar, H. E.; Prasad, P. N. Chem. Commun. 2006,
2071.
(23) Kim, S.; Ohulchanskyy, T. Y.; Pudavar, H. E.; Pandey, R. K.;
Prasad, P. N. J. Am. Chem. Soc. 2007, 129, 2669.
(24) Wen, X.; Li, M.; Wang, Y.; Zhang, J.; Fu, L.; Hao, R.; Ma, Y.; Ai,
X. Langmuir 2008, 24, 6932.
(25) Chelebaeva, E.; Raehm, L.; Durand, J.-O.; Guari, Y.; Larionova,
J.; Guerin, C.; Trifonov, A.; Willinger, M.; Thangavel, K.; Lascialfari,
A.; Mongin, O.; Mir, Y.; Blanchard-Desce, M. J. Mater. Chem. 2010,
20, 1877.
(26) Li, L.; Tian, Y.; Yang, J.; Sun, P.; Kong, L.; Wu, J.; Zhou, H.;
Zhang, S.; Jin, B.; Tao, X.; Jiang, M. Chem. Commun. 2010, 46, 1673.
(27) Shao, G.; Han, R.; Ma, Y.; Tang, M.; Xue, F.; Sha, Y.; Wang, Y.
Chem.Eur. J. 2010, 16, 8647.
́
(28) Rendon, N.; Bourdolle, A.; Baldeck, P. L.; Le Bozec, H.;
Andraud, C.; Brasselet, S.; Coper
23, 3228.
́
et, C.; Maury, O. Chem. Mater. 2011,
(62) Feher, F. J.; Tajima, T. L. J. Am. Chem. Soc. 1994, 116, 2145.
1072
dx.doi.org/10.1021/cm404140q | Chem. Mater. 2014, 26, 1062−1073

Chemistry of Materials
Article
(63) Cordes, D. B.; Lickiss, P. D.; Rataboul, F. Chem. Rev. 2010, 110,
2081.
(64) Fujdala, K. L.; Tilley, T. D. Chem. Mater. 2001, 13, 1817.
(65) Fischbach, A.; E., G.; Scherer, W.; Herdtweck, E.; Anwander, R.
Z. Naturforsch. 2004, 59b, 1353.
(66) Senechal, K.; Maury, O.; Le Bozec, H.; Ledoux, I.; Zyss, J. J. Am.
́ ́
Chem. Soc. 2002, 124, 4560.
(67) Blanda, M. T.; Horner, J. H.; Newcomb, M. J. Org. Chem. 1989,
54, 4626.
(68) Suh, M.; Lee, H.-J.; Park, J.-Y.; Lee, U. H.; Kwon, Y.-U.; Kim, D.
J. ChemPhysChem 2008, 9, 1402.
(69) Ziessel, R. F.; Ulrich, G.; Charbonnier
̀
e, L.; Imbert, D.;
Scopelliti, R.; Bunzli, J.C. G. Chem.Eur. J. 2006, 12, 5060.
̈
(70) Gonca
̧
lves e Silva, F. R.; Malta, O. L.; Reinhard, C.; Gudel, H.-
̈
U.; Piguet, C.; Moser, J. E.; Bunzli, J.-C. G. J. Phys. Chem. A 2002, 106,
̈
1670.
(71) Reinhard, C.; Gudel, H. U. Inorg. Chem. 2002, 41, 1048.
̈
(72) Pointillart, F.; Le Guennic, B.; Golhen, S. p.; Cador, O.; Maury,
O.; Ouahab, L. N. Inorg. Chem. 2013, 52, 1610.
(73) Pointillart, F.; Guennic, B. L.; Golhen, S.; Cador, O.; Maury, O.;
Ouahab, L. Chem. Commun. 2013, 49, 615.
(74) Pointillart, F.; Le Guennic, B.; Cauchy, T.; Golhen, S. p.; Cador,
O.; Maury, O.; Ouahab, L. N. Inorg. Chem. 2013, 52, 5978.
(75) Chen, R. J. Lumin. 2003, 102, 510.
(76) Berberan-Santos, M. N.; Bodunov, E. N.; Valeur, B. Chem. Phys.
2005, 315, 171.
(77) Metivier, R.; Leray, I.; Lefevre, J.-P.; Roy-Auberger, M.; Zanier-
Szydlowski, N.; Valeur, B. Phys. Chem. Chem. Phys. 2003, 5, 758.
(78) Gajan, D.; Schwarzwalder, M.; Conley, M. P.; Gruening, W. R.;
̈
Rossini, A. J.; Zagdoun, A.; Lelli, M.; Yulikov, M.; Jeschke, G.; Sauvee
C.; Ouari, O.; Tordo, P.; Veyre, L.; Lesage, A.; Thieuleux, C.; Emsley,
L.; Coperet, C. J. Am. Chem. Soc. 2013, 135, 15459−15466.
́
,
́
(79) Larson, D. R.; Zipfel, W. R.; Williams, R. M.; Clark, S. W.;
Bruchez, M. P.; Wise, F. W.; Webb, W. W. Science 2003, 300, 1434.
(80) Pu, S.-C.; Yang, M.-J.; Hsu, C.-C.; Lai, C.-W.; Hsieh, C.-C.; Lin,
S. H.; Cheng, Y.-M.; Chou, P. T. Small 2006, 2, 1308.
(81) Clapp, A. R.; Pons, T.; Medintz, I. L.; Delehanty, J. B.; Melinger,
J. S.; Tiefenbrunn, T.; Dawson, P. E.; Fisher, B. R.; O’Rourke, B.;
Mattoussi, H. Adv. Mater. 2007, 19, 1921.
1073
dx.doi.org/10.1021/cm404140q | Chem. Mater. 2014, 26, 1062−1073