Ultrabright fluorescent polymeric nanoparticles made from a new family of BODIPY monomers

Article abstract of DOI:10.1021/ma400590q

Four novel BODIPY derivatives (π-) functionalized by different polymerizable groups, styrene (S), phenyl acrylate (PhA), ethyl methacrylate (EtMA) and ethyl acrylate (EtA) have been synthesized. Following a formerly established one-pot RAFT miniemulsion p

Full text of DOI:10.1021/ma400590q

Article
pubs.acs.org/Macromolecules
Ultrabright Fluorescent Polymeric Nanoparticles Made from a New
Family of BODIPY Monomers
Chloe Grazon,† Jutta Rieger,‡,* Rachel Meallet-Renault,† Bernadette Charleux,§ and Gilles Clavier†,*
́
́
†PPSM, ENS Cachan, CNRS, 61 av Pres
́
ident Wilson, F-94230 Cachan, France
‡UPMC Univ Paris 06, Laboratoire de Chimie des Polymer
Chimie des Polymer
§Universite
̀
es (LCP), UMR 7610 94200 IVRY, France and CNRS, Laboratoire de
es (LCP), UMR 7610, 94200 IVRY, France
Lyon 1, CPE Lyon, CNRS UMR 5265, Universite de Lyon, C2P2, Team LCPP Bat 308F, 43 Bd du 11 novembre 1918,
̀
́
́
69616 Villeurbanne, France
S
* Supporting Information
ABSTRACT: Four novel BODIPY derivatives (π-) functionalized by
different polymerizable groups, styrene (S), phenyl acrylate (PhA), ethyl
methacrylate (EtMA) and ethyl acrylate (EtA) have been synthesized.
Following a formerly established one-pot RAFT miniemulsion polymer-
ization process (Grazon et al. Macromol. Rapid Commun. 2011, 32,
699−705), the fluorophores were copolymerized in a controlled way at 2.6
mol % with styrene in water. On the basis of the polymerization-induced
self-assembly (PISA) principle, the copolymers assembled during their
formation into fluorescent nanoparticles. The distribution of the
fluorescent monomers along the polymer backbone was monitored by
kinetic studies of the copolymerization reaction. Fluorescent stationary and
time-resolved spectroscopy was then performed on both the monomers
and the nanoparticles (NPs) and the observed differences are discussed in
view of the distribution of the fluorescent monomers in the polymer chain. With two of the novel fluorescent monomers (πS and
πPhA), the brightness of the NPs could be significantly improved (by a factor 2) compared to particles comprising the other
BODIPY monomers. The obtained particles were 200 to 2000 times brighter than usual quantum dots and 40 to 300 times
brighter than most of the fluorescent polymeric nanoparticles reported in the literature.
parameter in the design of fluorescent nanoparticles since it
takes into account both absorption and fluorescence parameters
and will determine how the fluorescent nano-objects will be
detected in single particle fluorescence imaging.
INTRODUCTION
■
Fluorescent molecules and nano-objects receive nowadays
increasing interest for their high potential in sensing, imaging
and biomedical applications.1 Most of the organic fluorophores
are hydrophobic compounds that are not soluble in water-based
biological media. They can be modified with water solubilizing
groups but generally at the expense of the fluorescence
quantum yield which drops dramatically.2 Furthermore, the
toxicity of these compounds is not well-known. An appealing
alternative is to incorporate them in organic or inorganic
(nano)particles that are water-dispersible. However, one of the
main problems of this approach is that the fluorophores can
leak out of the particles with time. In order to avoid this
shortcoming, the best solution is to covalently link the
fluorophore to the polymer backbone. This can be achieved
either by postmodifying the polymer3 with reactive fluoro-
phores or by copolymerizing fluorescent monomers with a
comonomer. As such, rhodamine,4 fluorescein5 or BODIPY-
derived monomers2b have been successfully used to prepare
fluorescent nano-objects. Nevertheless, when fluorophores are
concentrated in a confined space such as a particle,
luminescence quenching may appear.6 This usually leads to a
decrease in the fluorescence lifetime, quantum yield and
therefore the brightness. The latter is a particularly important
Recent research in controlled radical polymerization (CRP)
has shown that atom transfer radical polymerization (ATRP)7
and reversible addition−fragmentation transfer (RAFT)8,9
polymerization are efficient tools to achieve polymer chains
that are functionalized by fluorescent dyes either at their α-end
(using functional initiators)10 or along the polymer backbone
by copolymerization with fluorescent monomers.11 Indeed,
CRP methods are especially appealing compared to conven-
tional radical polymerization since they allow the formation of
copolymers which are homogeneous both in molar mass and
monomer composition.12 The microstructure of these copoly-
mers, i.e., the distribution of monomer units within the chain, is
governed by the relative reactivity of the monomers M1 and
M2, expressed by the reactivity ratios rM1 and rM2. Depending
on these values as well as on the molar fractions of the
Received: March 20, 2013
Revised: May 28, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ma400590q | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
monomers in the feed, “random”, “gradient”, “alternated”, or
“blocky” structures are obtained.
the polymerization of different organic fluorophores with the
same polymerizable function.13,19
Hitherto, only few works dealing with the study of the
microstructure of fluorescent copolymers have been reported.
Winnik et al. have studied the incorporation of fluorescent
benzothioxanthene-based monomers possessing a methacrylate
function in polymer particles synthesized by emulsion or
miniemulsion polymerizations.13 When copolymerized with
styrene or butyl methacrylate as a comonomer, they
demonstrated that the fluorescent monomer was homoge-
neously distributed along the polymer chains. Nevertheless, the
fluorescence properties of the obtained nanoparticles were not
studied.
EXPERIMENTAL SECTION
■
Instrumentation. 1H, 13C, 11B, and 19F NMR spectra were
recorded in CDCl3 on a JEOL ECS (400 MHz) spectrometer. All
chemical shifts are in ppm and referenced to tetramethylsilane (TMS).
Coupling constants (J) values are given in Hz. For numbering of
protons and carbons in the NMR spectra, see Figure SI-1, Supporting
Information.
The number-average molar mass (Mn), the weight-average molar
mass (Mw), and the molar mass distribution (polydispersity index Mw/
Mn) were determined by size exclusion chromatography (SEC, Mn,SEC
)
using THF as an eluent at a flow rate of 1 mL·min−1. For analytical
purposes, the acidic functions of the block or alternated copolymers
were turned into methyl esters. Therefore, the copolymers were
recovered by drying of the aqueous suspensions. After dissolution in a
THF/H2O mixture and acidification of the medium with a 1 M HCl
solution, they were methylated using an excess of trimethylsilyldiazo-
methane.20 Polymers were analyzed at a concentration of 5 mg·mL−1
in THF after filtration through 0.45 μm pore size membrane. The SEC
apparatus was equipped with a Viskotek VE 2100 automatic injector
and two columns thermostated at 40 °C (PLgel Mixed, 7.5 mm × 300
mm, bead diameter = 5 μm). Detection was made with a differential
refractive index detector (Viscotek VE 3580 RI detector) and a UV−
vis detector (Waters 486 Tunable Absorbance Detector). The
Viscotek OmniSEC software (v 4.6.2) was used for data analysis and
the relative Mn and Mw/Mn were calculated with a calibration curve
based on polystyrene standards (from Polymer Laboratories).
The z-average particle diameter (named Dz) and the particle size
distribution (dispersity factor, named σ), were determined by dynamic
light scattering (DLS) of the diluted aqueous dispersions, at an angle
of 90° at 20 °C, with a Zetasizer Nano S90 from Malvern, using a 4
mW He−Ne laser at 633 nm. A value of σ below 0.1 is characteristic of
a narrow particle size distribution. All calculations were performed
using the Nano DTS software.
We recently reported the synthesis of bright fluorescent
nanoparticles with a core made of styrene copolymerized with a
BODIPY monomer.14 BODIPY had been chosen as a
hydrophobic fluorophore,15 as it exhibits attractive spectro-
scopic characteristics such as an emission spectrum tunable
from green to red and high fluorescence quantum yields. The
synthesis of the nanoparticles was achieved directly in water by
copolymerizing styrene with a few mol % of a phenyl
methacrylate BODIPY (πPhMA) derivative in a one-pot
RAFT miniemulsion polymerization. This process is very
attractive since neither ultrahydrophobic agents nor low molar
mass surfactantswhich may be detrimental to the targeted
biological applicationswere used. The stability of the particles
and the control of the chain growth were achieved by the use of
amphiphilic poly(ethylene oxide)-block-poly(acrylic acid)-block-
polystyrene (PEO-b-PAA-b-PS) copolymers terminated by a
reactive trithiocarbonate RAFT agent, which are chain extended
during the miniemulsion polymerization. In a kinetic study of
the copolymerization of the BODIPY monomer with styrene,
we found that the BODIPY monomer was very rapidly
incorporated into the polymer backbone, leading to the
formation of a composition gradient of fluorophores.
Fluorescence studies showed that the quantum yield of the
fluorophore (0.69 in toluene) decreased to 0.20 in the
nanoparticles. Parallel research showed that this decrease of
the quantum yield was due to poorly fluorescent aggregates of
BODIPY, maybe resulting from an inhomogeneous distribution
of the fluorescent monomer in the polymer backbone.16,17
Indeed in the literature, the reactivity ratios rS and rPhMA for the
copolymerization of styrene (S) with phenyl methacrylate
(PhMA)a model monomer for πPhMAhave been reported
to be 0.25 and 0.5,18 respectively, corroborating the initial
enrichment of the chains in fluorescent comonomer units.
In this work, we report the synthesis of BODIPY derivatives
possessing various polymerizable functions (ethyl acrylate or
methacrylate, phenyl acrylate or methacrylate and styrene) and
their copolymerization with styrene using the formerly
established one-pot miniemulsion polymerization process.14
The spectroscopic signature of the new BODIPY derivatives
was analyzed and compared to that of the particles obtained.
The individual consumption of the different monomers was
monitored with respect to styrene allowing us to conclude on
the microstructure of the copolymers formed. We finally
discuss the possibility to establish a relationship between their
distribution along the polymer chain and the fluorescence
properties of the particles. To the best of our knowledge, this is
the first time that such study, namely the attempt to control the
distribution of a fluorophore along a copolymer backbone by
systematically varying the nature of the polymerizable function,
is reported. Previous comparable reports have mainly dealt with
UV−visible spectra were recorded on a Varian Cary (Palo Alto, CA)
double beam spectrometer using a 10 mm path quartz cell from Thuet
(Bodelsheim, France). Molar extinction coefficients (ε) are given at
the maximal absorption for each monomer, with an error of 5%.
Excitation and emission spectra were measured on a SPEX Fluoromax-
3 (Horiba Jobin-Yvon). A right-angle configuration was used. Optical
density of the samples was checked to be less than 0.1 to avoid
reabsorption artifacts. The fluorescence quantum yields ΦF were
determined using Rhodamine 590 (ΦF = 0.95 in ethanol) as a
reference (error of 15%).21 The fluorescence decay curves were
obtained with a time-correlated single-photon-counting method using
a titanium-sapphire laser (82 MHz, repetition rate lowered to 4 MHz
thanks to a pulse-peaker, 1 ps pulse width, a doubling crystals is used
to reach 495 nm excitation) pumped by an argon ion laser from
Spectra Physics (Mountain View, CA). For monomers, the
Levenberg−Marquardt algorithm was used for nonlinear least-squares
fit as implemented in the Globals software (Globals Unlimited, Villa
Grove, USA). Lifetimes are given with an error of 0.05 ns. In order
to estimate the quality of the fit, the weighted residuals were
calculated. In the case of single photon counting, they are defined as
the residuals, i.e., the difference between the measured value and the
fit, divided by the square root of the fit. χ2 is equal to the variance of
the weighted residuals. A fit was said appropriate for χ2 values between
0.8 and 1.2.
For multiexponential fluorescent decays (nanoparticles), no fit was
attempted and the average fluorescence lifetimes were calculated by
integrating the area below the decay curve as22
∞ tI(t) dt
∫
∫
0
⟨τ⟩ ≡
∞ I(t) dt
(1)
0
Radiative decay rates (kr) and non radiative decay rates (knr) are
calculated as:
B
dx.doi.org/10.1021/ma400590q | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
ϕF
τ
1 − ϕF
The mixture was concentrated under vacuum and purified by
chromatography on silica gel (dichloromethane, DCM). 318 mg of a
pink powder were obtained (yield: 72%).
kr =
;
knr =
(2)
τ
1H NMR (400 MHz, CDCl3): δ = 7.18 (d, 2H, JH−H = 6.4 Hz, H10),
7.03 (d, 2H, JH−H = 6.9 Hz, H11), 4.16 (t, 2H, JH−H = 4.6 Hz, H13),
4.03 (t, 2H, JH−H = 4.4 Hz, H14), 2.53 (s, 6H, H3′, H5′), 2.30 (q, 4H,
JH−H = 7.6 Hz, H2′, H6′), 1.33 (s, 6H, H1′, H7′), 0.98 (t, 6H, JH−H = 7.6
Hz, H2″, H6″) ppm. 13C NMR (100 MHz, CDCl3): δ = 159.2 (C12),
153.7, 140.2, 138.4, 132.8, 131.3, 129.7 (C10), 128.5, 115.1 (C11), 69.3
(C13), 61.6 (C14), 17.2 (C2′, C6′), 14.8 (C2″, C6″), 12.6 (C3′, C5′), 12.0
Materials. 2,4-Dimethyl-3-ethylpyrrole (97%, Aldrich, kryptopyr-
role), boron trifluoride diethyl etherate (2 M in diethyl ether, Aldrich),
4-hydroxybenzaldehyde (98%, Aldrich), tetrachloro-1,4-benzoquinone
(99%, Aldrich, Chloranil), N,N-diisopropylethylamine (99.5%, Sigma-
Aldrich, DIPEA), 1,8-diazobicyclo[5,4,0]undec-7-ene (≥98%, Fluka,
DBU), trifluoroacetic acid (99%, Sigma-Aldrich, TFA), acryloyl
chloride (97%, Aldrich), triethylamine (≥99%, Sigma-Aldrich), thionyl
chloride (99%, Fluka), 4,4′-azobis(4-cyanopentanoic acid) (Aldrich,
ACPA), (trimethylsilyl)diazomethane (2 M solution in diethyl ether,
Aldrich) were used as received. Solvents (Carlo Erba) were of
synthetic grade and purified according to standard procedures.
Methacryloyl chloride (97%, Fluka) and styrene were distilled under
reduced pressure. 2,2′-Azobis(2-methylpropionitrile) (98%, Sigma,
AIBN) was recrystallized from chloroform and few drops of petroleum
ether. Silica gel 60 Å (70−200 mm porosity) was bought from SDS.
Synthesis of BODIPY phenol 6: 2,6-Diethyl-4,4-difluoro-8-(4-
hydroxyphenyl)-1,3,5,7-tetramethyl-4-bora-3a,4a-diaza-s-in-
dacene. BODIPY phenol 6 was obtained as described elsewhere23
(C1′, C7′) ppm. 19F NMR (376 MHz, CDCl3): δ = −145.7 (q, JF−B
32.3 Hz) ppm. 11B NMR (128 MHz, CDCl3): δ = −0.15 (t, JB−F
=
=
34.5 Hz) ppm. HRMS (ESI) m/z [M + Na]+ calculated for
C25H31N2O2F2BNa, 463.2339; found, 463.2342 (100%).
Synthesis of Monomer πEtMA 3: 2,6-Diethyl-4,4-difluoro-8-
(4-(2-(methacryloyloxy)ethoxy)phenyl)-1,3,5,7-tetramethyl-4-
bora-3a,4a-diaza-s-indacene. BODIPY derivative πEtOH 7 (0.43
mmol, 190 mg) was dissolved in 4 mL of DCM in a round-bottom
flask at 0 °C. Once the dye dissolved, triethylamine (7 equiv., 3.0
mmol, 0.41 mL) was added. At last, methacryloyl chloride (1.5 equiv.,
65 mmol, 60 μL) diluted in 1 mL of DCM was slowly added to the
reaction with a syringe. The reaction was stirred overnight, until
disappearance of the πEtOH 7 on a TLC plate. The mixture was then
washed twice with ∼50 mL of water and ∼50 mL of brine. The organic
phase was dried over MgSO4 and concentrated under vacuum. The
residue was purified by chromatography on silica gel (dichloro-
methane/petroleum ether: 2/1). 100 mg of a pink powder were
obtained (yield: 50%).
1
(6.40 g, yield 80%; H NMR (400 MHz, CDCl3) δ = 7.12 (d, 2H,
H10), 6.94 (d, 2H, H11), 5.23 (s, 1H, −OH), 2.53 (s, 6H, H3′, H5′),
2.30 (q, 4H, H2′, H6′), 1.35 (s, 6H, H1′, H7′), 0.98 (t, 6H, H2″, H6″);
13C NMR (100 MHz, CDCl3) δ = 156.2, 153.7, 138.6, 132.9, 131.3,
129.9, 128.3, 116.2, 17.2, 14.8, 12.6, 12.0).
Synthesis of Monomer πPhMA 1: 2,6-Diethyl-4,4-difluoro-8-
(4-(methacryloyloxy)phenyl)-1,3,5,7-tetramethyl-4-bora-
3a,4a-diaza-s-indacene. The synthesis of monomer πPhMA 1 was
performed as previously described,14 by an esterification of BODIPY
1H NMR (400 MHz, CDCl3): δ = 7.17 (d, 2H, JH−H = 7.3 Hz, H10),
7.02 (d, 2H, JH−H = 7.3 Hz, H11), 6.19 (s, 1H, H18), 5.62 (s, 1H, H18′),
4.54 (t, 2H, JH−H = 4.1 Hz, H14), 4.29 (t, 2H, JH−H = 3.7 Hz, H13), 2.52
(s, 6H, H3′, H5′), 2.30 (q, 4H, JH−H = 7.3 Hz, H2′, H6′), 1.98 (s, 3H,
H17), 1.33 (s, 6H, H1′, H7′), 0.98 (t, 6H, JH−H = 7.3 Hz, H2″, H6″) ppm.
13C NMR (100 MHz, CDCl3): δ = 167.5 (C15), 159.1 (C12), 153.7,
140.2, 138.5, 136.1, 132.8, 131.3, 129.7 (C10), 128.6, 126.3 (C18),
115.3 (C11), 66.1 (C13), 63.1 (C14), 18.5 (C17), 17.2 (C2′, C6′), 14.8
(C2″, C6″), 12.6 (C3′, C5′), 12.0 (C1′, C7′) ppm. 19F NMR (376 MHz,
CDCl3): δ = −145.7 (q, JF−B = 32.3 Hz) ppm. 11B NMR (128 MHz,
CDCl3): δ = −0.17 (t, JB−F = 32.0 Hz) ppm. HRMS (ESI) m/z [M +
Na]+ calculated for C29H35N2O3F2BNa, 531.2607; found, 531.2604
(100%). Mp =169 °C.
1
phenol 6 with methacryloyl chloride (432 mg, yield 78%; H NMR
(400 MHz, CDCl3) δ = 7.32 (d, 2H, H10), 7.27 (d, 2H, H11), 6.39 (s,
1H, H16′), 5.80 (s, 1H, H16), 2.52 (s, 6H, H3′, H5′), 2.29 (q, 4H, H2′,
H6′), 2.08 (s, 3H, H15), 1.33 (s, 6H, H1′, H7′), 0.97 (t, 6H, H2″, H6″)
ppm; 13C NMR (100 MHz, CDCl3) δ = 165.69, 154.01, 151.44,
139.22, 138.45, 135.78, 133.32, 133.00, 130.90, 129.53, 127.68, 122.58,
18.46, 17.15, 14.68, 12.60, 11.92 ppm).
Synthesis of Monomer πPhA 2: 2,6-Diethyl-4,4-difluoro-8-
(4-(acryloyloxy)phenyl)-1,3,5,7-tetramethyl-4-bora-3a,4a-
diaza-s-indacene. The reaction was carried out using the same
protocol as for πPhMA 1 using acryloyl chloride to give 565 mg of a
pink powder (yield 50%).
1H NMR (400 MHz, CDCl3): δ = 7.33−7.27 (m, 4H, H10, H11),
Synthesis of Monomer πEtA 4: 2,6-Diethyl-4,4-difluoro-8-(4-
(2-acryloyloxy) ethoxy)phenyl)-1,3,5,7-tetramethyl-4-bora-
3a,4a-diaza-s-indacene. The reaction was carried out using the
same protocol as for πEtMA 3 using acryloyl chloride for the
esterification. So, 43 mg of a pink powder were obtained from 88 mg
of BODIPY πEtOH 7 (yield: 44%).
6.65 (dd, JH−H = 1.2 Hz, JH−H = 17.4 Hz, 1H, H15), 6.36 (dd, JH−H
17.2 Hz, JH−H = 10.3 Hz, 1H, H14), 6.06 (dd, JH−H =1.2 Hz, JH−H
=
=
10.5 Hz, 1H, H15), 2.53 (s, 6H, H3′, H5′), 2.30 (q, JH−H = 7.8 Hz, 4H,
H2′, H6′), 1.34 (s, 6H, H1′, H7′), 0.98 (t, JH−H = 7.3 Hz, 6H, H2″, H6″)
ppm. 13C NMR (100 MHz, CDCl3): δ = 164.3 (C13), 154.0, 151.1,
139.2, 138.4, 133.4, 133.1 (C15), 130.9, 129.5 (C10), 127.8 (C14), 122.4
(C11), 17.2 (C2′, C6′), 14.7 (C2″, C6″), 12.6 (C3′, C5′), 11.9 (C1′, C7′)
ppm. 19F NMR (376 MHz, CDCl3): δ = −145.7 (q, JF−B = 32.3 Hz)
ppm. 11B NMR (128 MHz, CDCl3): δ = −0.15 (t, JB−F = 34.5 Hz)
ppm. HRMS (ESI) m/z [M + H]+ calculated for C26H29BF2N2O2H,
451.2368; found, 451.2359 (100%). Mp =175 °C.
1H NMR (400 MHz, CDCl3): δ = 7.18 (d, 2H, JH−H = 8.7 Hz, H10),
7.02 (d, 2H, JH−H = 9.1 Hz, H11), 6.50 (d, 1H, JH−H = 17.4 Hz, H17),
6.21 (dd, 1H, JH−H = 17.2, JH−H = 10.3 Hz, H16), 5.90 (d, 1H, JH−H
=
10.5 Hz, H17), 2.53 (s, 6H, H3′, H5′), 2,30 (q, 4H, JH−H = 7.6 Hz, H2′,
H6′), 1.33 (s, 6H, H1′, H7′), 0.98 (t, 6H, JH−H = 7.6 Hz, H2″, H6″) ppm.
13C NMR (100 MHz, CDCl3): δ = 166.3 (C15), 159.0 (C12), 153.7,
138.5, 132.8, 131.7, 131.3, 129.7 (C10), 128.6, 128.2, 115.3 (C11), 66.1
(C13), 63.0 (C14), 17.2 (C2′, C6′), 14.8 (C2″, C6″), 12.6 (C3′, C5′), 12.0
Synthesis of BODIPY πEtOH 7: 2,6-Diethyl-4,4-difluoro-8-(4-
(2-hydroxyethoxy)phenyl)-1,3,5,7-tetramethyl-4-bora-3a,4a-
diaza-s-indacene. 4-(2-Hydroxyethoxy)benzaldehyde used in this
procedure was synthesized according to the literature24 (325 mg, yield
(C1′, C7′) ppm. 19F NMR (376 MHz, CDCl3): δ = −145.7 (q, JF−B
32.3 Hz) ppm. 11B NMR (128 MHz, CDCl3): δ = −0.15 (t, JB−F
=
=
1
33.2 Hz) ppm. HRMS (ESI) m/z [M + Na]+ calculated for
C28H33N2O3F2BNa: 517.2445, found: 517.2447 (100%). mp =172 °C.
Synthesis of BODIPY πS 5: 2,6-Diethyl-4,4-difluoro-8-(4-
vinylphenyl)-1,3,5,7-tetramethyl-4-bora-3a,4a-diaza-s-inda-
cene. First, the 4-vinylbenzoyl chloride was synthesized in four steps
following literature procedures. 4-Bromomethylbenzoic acid was
obtained starting from the 4-methylbenzoic acid (7.90 g, yield:
quantitative; 1H NMR (400 MHz, CDCl3): δ = 10.62 (bs, 1H,
COOH), 8.08 (d, 2H, JH−H = 8.2 Hz, HAr), 7.49 (d, 2H, JH−H = 8.2 Hz,
HAr), 4.51 (s, 2H, CH2-Br) ppm).25 (4-Carboxybenzyl)-
triphenylphosphonium bromide was obtained starting from 4-
66%; H NMR (400 MHz, CDCl3) δ = 9.90 (s, 1H, CHO), 7.85 (d,
2H, JH−H = 8.7 Hz, HAr), 7.03 (d, 2H, JH−H = 8.7 Hz, HAr), 4.18 (t, 2H,
JH−H = 4.6 Hz, CH2), 4.02 (t, 2H, JH−H = 4.6 Hz, CH2) ppm).
4-(2-Hydroxyethoxy)benzaldehyde (1 mmol, 0.17 g) and krypto-
pyrrole (2 equiv., 2 mmol, 0.25 g) dissolved in 25 mL of anhydrous
dichloromethane were introduced in a round-bottom flask flushed with
argon and equipped with a CaCl2 moisture trap. Then four drops of
TFA were added to the reaction. The dark reaction mixture was stirred
at room temperature until total disappearance of the aldehyde
(determined by TLC analysis). Chloranil (1 equiv, 1 mmol, 0.24 g)
was added and the reaction stirred for 2 min. Then DIPEA (7 equiv, 7
mmol, 0.9 g) and boron trifluoride diethyl etherate (11 equiv, 11
mmol, 1.56 g) were introduced. After 2h the reaction was stopped.
1
bromomethylbenzoic acid (6.49 g, yield: 45%; H NMR (400 MHz,
DMSO-d6): δ = 12.88 (s, 1H, COOH), 7.71−8.01 (m, 17H, HAr), 7.08
C
dx.doi.org/10.1021/ma400590q | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 1. Molecular structure of the BODIPY monomers possessing different polymerizable functions.
(m, 2H, HAr), 5.25 (d, 2H, JH−P = 12.01 Hz, CH2−P) ppm).26 4-
Vinylbenzoic acid was obtained starting from (4-carboxybenzyl)-
triphenylphosphonium (2.14 g, 79%; 1H NMR (400 MHz, CDCl3): δ
= 10.60 (bs, 1H, COOH), 8.07 (d, 2H, JH−H = 8.2 Hz, HAr), 7.50 (d,
2H, JH−H = 8.2 Hz, HAr), 6.77 (dd, 1H, JH−H = 10.9, 17.5 Hz, CH
with nitrogen for 30 min in an ice bath, then placed in an oil bath
thermostated at 80 °C to initiate the polymerization. After 90 min, the
reaction was stopped by immersion of the flask in iced water. The
1
monomer conversion was determined by H NMR in CDCl3. The
copolymer was dried under reduced pressure in order to remove the
residual acrylic acid monomer. PEO45-b-PAA18-C12 macro-RAFT agent
(Mn,th = 3.7, Mn,SECPS = 4.6, Mw/Mn = 1.14) composed of a PEO block
of Mn = 2.0 kg/mol and a PAA block of Mn = 1.7 kg/mol was
obtained.
CH2), 5.90 (d, 1H, JH−H = 17.4 Hz, CHCH2), 5.42 (d, 1H, JH−H
=
11.0 Hz, CHCH2) ppm). To obtain the 4-vinylbenzoyl chloride, 4-
vinylbenzoic acid (3.38 mmol, 0.5 g) was dissolved in 12 mL of
chloroform. Then, thionyl chloride (10 equiv., 33.8 mmol, 4.02 g) was
quickly added followed by one drop of DMF. The solution was heated
under reflux for 5h. The mixture was purified by a short
chromatography on silica gel (DCM). A yellow oil was obtained
(yield: quantitative).
Then, in a typical experiment of the nanoparticles synthesis, 140 mg
of PEO-b-PAA-C12 macro-RAFT (4.0 × 10−5 mol, Mn = 3.7 kg/mol)
was dissolved in a mixture of 650 mg of styrene (6.3 × 10−3 mol), 2.2
mg of AIBN (1.3 × 10−5 mol) and 58 mg of monomer πPhA 2 (1.3 ×
10−4 mol), in a septum-sealed flask. The mixture was purged with
argon for 30 min in an ice bath, and then placed in an oil bath
thermostatically controlled at 80 °C to initiate polymerization. After
70 min, the reaction was stopped by immersion of the flask in iced
water. The conversion of the monomers (styrene and BODIPY
derivative) was determined by gravimetry and SEC, respectively (for
details see below). To the cold organic mixture, 5 mL of basic water
(pH = 12.5) was added. An ultrasonic horn (Bandelin electronics,
Sonoplus HD 2200) is then placed in the biphasic mixture cooled
down in an ice bath and powered at 130 W for 10 min.
1H NMR (400 MHz, CDCl3): δ = 8.08 (d, 2H, JH−H = 8.7 Hz, H10),
7.52 (d, 2H, JH−H = 8.7 Hz, H11), 6.77 (dd, 1H, JH−H = 11.0, 17.4 Hz,
H13), 5.95 (d, 1H, JH−H = 17.4 Hz, H14), 5.49 (d, 1H, JH−H = 11.0 Hz,
H
14′) ppm. 13C NMR (100 MHz, CDCl3): δ = 168.0 (C8), 144.5
(C12), 135.5 (C13), 132.0 (C9), 131.9 (C10), 126.7 (C11), 118.6 (C14)
ppm.
The 4-vinylbenzoyl chloride (2.0 mmol, 332 mg) was dissolved in
50 mL of dry dichloromethane and the solution placed in a round-
bottom flask under argon. Kryptopyrrole (2.1 equiv., 4.2 mmol, 570
μL) was then added and the mixture heated under reflux for 2 h. Once
the temperature decreased to room temperature, DIPEA (8 equiv.,
16.0 mmol, 2.07 g) was added and 15 min later boron trifluoride
diethyl etherate (11 equiv., 22 mmol, 3.12 g). The solvent was
evaporated under vacuum and the residue purified by chromatography
on silica gel (DCM). 60 mg of a pink powder were obtained (yield:
7%).
After the miniemulsion formation, the pH decreased to 11. The
miniemulsion is purged with argon for 30 min in an ice bath, and then
placed in an oil bath thermostatically controlled at 80 °C to reinitiate
the polymerization. Sampling is performed at regular time intervals
and monomer conversions are determined by gravimetric analysis for
styrene corrected from the styrene loss by evaporation during the
1H NMR (400 MHz, CDCl3): δ = 7.53 (d, 2H, JH−H = 7.8 Hz, H11),
7.24 (d, 2H, JH−H = 8.2 Hz, H10), 6.79 (dd, 1H, JH−H = 10.8, 17.4 Hz,
H13), 5.86 (d, 1H, JH−H = 17.4 Hz, H14), 5.35 (d, 1H JH−H = 11.0,
1
sonication process (25 wt % calculated by comparison of H NMR
spectra in CDCl3 and gravimetric analysis (see Figure SI-9, Supporting
Information); considering this the molar percentage of BODIPY
monomer was therefore corrected to 2.6 0.1 mol %30) and by SEC
using the UV−visible detection for BODIPY monomers (Table 2).
H
14′), 2.53 (s, 6H, H3′, H5′), 2.30 (q, 4H, JH−H = 7.56, H2′, H6′), 1.32
(s, 6H, H1′, H7′), 0.98 (t, 6H, JH−H = 7.6, H2″, H6″) ppm. 13C NMR
(100 MHz, CDCl3): δ = 153.8, 140.1, 138.5, 138.1, 136.4 (C13), 135.4,
132.9, 130.9, 128.7 (C10), 126.9 (C11), 114.9 (C14), 17.2 (C2′, C6′),
14.8 (C2″, C6″), 12.6 (C3′, C5′), 12.0 (C1′, C7′) ppm. 19F NMR (376
MHz, CDCl3): δ = −145.7 (q, JF−B = 32.9 Hz) ppm. 11B NMR (128
MHz, CDCl3): δ = −0.15 (t, JB−F = 33.2 Hz) ppm. HRMS (ESI) m/z
[M + Na]+ calculated for C25H29BF2N2Na, 429.2284; found, 429.2289
(100%). Mp = 250 °C.
RESULTS AND DISCUSSION
■
Synthesis and Characterization of BODIPY Mono-
mers. Four novel fluorescent BODIPY derivatives possessing
different polymerizable functions have been synthesized (Figure
1): BODIPY phenyl acrylate (πPhA, 2), BODIPY ethyl
methacrylate (πEtMA, 3), BODIPY ethyl acrylate (πEtA, 4)
and BODIPY styrene (πS, 5) in order to compare them with
the previously reported phenyl methacrylate BODIPY
(πPhMA, 1).14
Synthesis of Fluorescent Nanoparticles by RAFT Miniemul-
sion Polymerization. The RAFT copolymerization of styrene and
BODIPY monomers (initial feed 2 mol %) in the presence of PEO-b-
PAA-C12 macro-RAFT agent in a one-pot phase inversion process was
performed as described elsewhere.14
First, the PEO-b-PAA-C12 macro-RAFT agent was synthesized
according to reference14 in 1,4-dioxane at 80 °C under argon
atmosphere: In a typical experiment, the PEO-based trithiocarbonate
macro-RAFT agent,14 PEO-C12, (0.5 mmol, 1.21 g, Mn = 2420 g/mol),
acrylic acid (10 mmol, 720 mg), and DMF (as an internal reference for
the 1H NMR determination of the monomer consumption in
deuterated chloroform) (2 mmol, 146 mg) were dissolved in 4.9 mL
of 1,4-dioxane at room temperature. Then, 0.1 mL of a 0.33 M
solution of ACPA in 1,4-dioxane was added. The mixture is purged
Monomers πPhMA 1 and πPhA 2 were synthesized starting
from a BODIPY bearing a phenol function (6) on the meso
position (Scheme SI-1, Supporting Information). The latter was
synthesized via a conventional one-pot three steps approach by
reacting 2 equiv of kryptopyrrole with one equivalent of 4-
hydroxybenzaldehyde (yield: 80%).14 To obtain the monomers,
a classical esterification of the BODIPY phenol 6 was
performed in presence of either methacryloyl chloride to
D
dx.doi.org/10.1021/ma400590q | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
obtain the monomer πPhMA 1 (yield: 78%), or acryloyl
chloride to obtain the monomer πPhA2 (yield: 50%).
Syntheses of monomers πEtMA 3 and πEtA 4 were first
attempted from the same BODIPY phenol 6. However, all trials
to effect a Williamson reaction on the phenol with 2-
bromoethanol or 2-bromoethyl methacrylate failed and led to
the degradation of compound 6. An alternative reaction scheme
was therefore established (Scheme SI-2, Supporting Informa-
tion), in which 4-(2-hydroxyethoxy)benzaldehyde was synthe-
sized first by a Williamson reaction of 4-hydroxybenzaldehyde
with 2-bromoethanol. The BODIPY framework 7 was then
formed from this new aldehyde and kryptopyrrole (yield: 72%).
Then, monomers πEtMA 3 and πEtA 4 were successfully
obtained by esterification of the alcohol with methacryloyl
chloride (yield: 50%) and acryloyl chloride (yield: 44%),
respectively.
Figure 3. Fluorescence decays of the different BODIPY monomers in
toluene. Monomers πPhMA 1 (green line), πPhA 2 (gray line),
πEtMA 3 (yellow line), πEtA 4 (pink line) and πS 5 (blue line). The
instrument response function (IRF) is presented in light gray (λexc
495 nm, λF = 543 nm).
=
At last, monomer πS 5 was synthesized via another
conventional route for BODIPY synthesis (Scheme SI-3),27
using 4-vinylbenzoyl chloride which was synthesized in four
steps and 21% overall yield starting from 4-methylbenzoic acid
using a Wittig reaction as the key step. Monomer πS 5 was then
obtained by reacting 4-vinylbenzoyl chloride with kryptopyrrole
in low (7%) but not optimized yield.
rate was of the same order (1.73 × 108 s−1 vs. ∼1.5 × 108 s−1
for the others). Since all monomers exhibit the same
fluorescent framework (same substituents on the pyrroles and
a phenyl group on the meso position), it is not surprising that
they display similar spectroscopic properties. This will allow for
an easier comparison of the spectroscopic properties of the
nanoparticles.
Absorption and fluorescence spectra of the BODIPY-based
monomers recorded in toluene are shown in Figure 2 and their
Synthesis of the Fluorescent Nanoparticles. We
previously reported14 the synthesis of fluorescent polymer
nanoparticles via controlled radical RAFT polymerization in
miniemulsion, using a surfactant-free one-pot phase inversion
process (Scheme 1). The selected synthetic approach relied on
the polymerization-induced self-assembly (PISA29) principle
where amphiphilic copolymers form during polymerization and
assemble simultaneously into core−shell particles. In more
details, a fluorescent BODIPY monomer (π) was copolymer-
ized with styrene (S) in the presence of a poly(ethylene oxide)-
b-poly(acrylic acid) macro-RAFT agent, PEO-b-PAA-C12, first
in bulk andafter phase inversionin miniemulsion. PEO-b-
PAA-b-P(S-co-π) triblock copolymers formed and self-
assembled in situ during the miniemulsion step affording well-
defined core−shell nanoparticles with a fluorescent core and a
biocompatible and reactive shell (see cartoons in Figure 4,
schematic representation of the NP, and Scheme SI-4,
Supporting Information for the synthetic approach). In this
former study (NP1 in Table 2), the core was made of styrene
copolymerized with 2.6 mol %30 of monomer πPhMA 1, and
the shell was a PEO-b-PAA diblock copolymer with a number-
average molar mass, Mn, of PEO of 2 kg/mol and that of PAA
was 1 kg/mol. The final particles had a hydrodynamic diameter
about 65 nm (σ ∼ 0.1) and an aggregation number (i.e., an
average number of copolymer chains per particle) of 1750
( 250).17
Figure 2. Absorption (full lines) and fluorescence spectra (dotted
lines, λexc = 495 nm) of the BODIPY monomer derivatives recorded in
toluene. Monomers πPhMA 1 (green line), πPhA 2 (gray line),
πEtMA 3 (yellow line), πEtA 4 (pink line), and πS 5 (blue line).
spectroscopic properties are given in Table 1. All monomers
had a maximum of absorption around 527 nm and a maximum
of fluorescence emission between 538 and 540 nm. They all
showed a quantum yield (ΦF) around 70%, a usual value for
that kind of BODIPY.28
In addition, time-resolved fluorescence measurements were
performed in toluene (Figure 3). Fluorescence decays could be
fitted by a monoexponential function (Figures SI-3−SI-7,
Supporting Information) and lifetimes were found to be
between 4.1 and 4.9 ns. Monomer πS 5 had a lifetime shorter
than the other monomeric derivatives but its radiative decay
Using the same synthetic strategy (Scheme 1, Scheme SI-4,
Supporting Information), in this work four novel types of
fluorescent nanoparticles (Table 2) were synthesized by
Table 1. Spectroscopic Properties of the BODIPY Monomer Derivatives Recorded in Toluene at 20 °C
a
b
monomer
λabs/nm
λF/nm
ΦF
ε (×10‑3) L·mol‑1·cm‑1
τ /ns
kr/10‑8 s‑1
knr/10‑7 s‑1
B
(×10‑3) L·mol‑1·cm‑1
πPhMA 1
πPhA 2
πEtMA 3
πEtA 4
πS 5
528
528
527
527
527
540
540
538
538
539
0.69
0.74
0.75
0.71
0.71
73
79
70
74
68
4.9
4.8
4.8
4.7
4.1
1.41
1.54
1.56
1.49
1.73
6.33
5.42
5.21
6.38
7.07
50.4
58.5
52.5
52.5
48.3
a
b
Decay fitted with a monoexponential function (λexc = 495 nm, λF = 543 nm). Molecular brightness, B = ε × ΦF.
E
dx.doi.org/10.1021/ma400590q | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Scheme 1. Synthetic Pathway for the Synthesis of the Fluorescent Nanoparticles
hydrodynamic diameter Dz ranging from 60 to 90 nm. Size
exclusion chromatography analyses on unpurified crude
samples taken from the reaction medium in regular time
intervals allowed not only following the consumption of the
PEO-b-PAA-C12 macro-RAFT agent during the polymerization
with the formation of well-defined growing PEO-b-PAA-b-P(S-
co-π) triblock copolymers (Figure 5 left, RI response), but also
confirmed the successful incorporation of the fluorescent
comonomer in the triblock copolymer. Indeed, using SEC
with an in-line UV−vis. detection set at the maximum
absorption wavelength of the monomers (λabs = 528 nm)
made possible the separation of the fluorescent monomer,
eluted at around 18 mL, from the fluorescent triblock
copolymer eluted between 13.5 and 16 mL (Figure 5, right,
UV−vis response at λ = 528 nm). The absorption of the
formed triblock copolymers at 528 nm undoubtedly confirms
the covalent incorporation of the fluorescent monomer in the
polymer chain. In addition, this detection mode also allows
following the fluorescent monomer consumption and the
calculation of the fluorescent monomer’s individual conversion
by comparison of both the monomer’s and the polymer’s signal
areas.
Figure 4. Schematic representation of the fluorescent core−shell
nanoparticles.
copolymerizing 2.6 mol %30 of the different BODIPY
monomers 2−5 with styrene. The target degree of polymer-
ization, DPn, was about 120 in all cases (initial monomer to
macro-RAFT agent molar ratio of 160, which leads to 120
considering the 25% of styrene evaporation).30 The final
particle size distributions were all monomodal with an average
Table 2. Experimental Results of the Different Fluorescent Nanoparticles (NP) Having Different BODIPY Monomers in their
Cores
a
b
c
d
e
e
f
NP
BODIPY monomer
nπ
convnS
convnπ
Mn,th /kg/mol
Mn,SEC /kg/mol
Mw/Mn
Dz (σ) /nm
NP1
NP2
NP3
NP4
NP5
πPhMA 1
πPhA 2
πEtMA 3
πEtA 4
πS 5
3.0
2.9
3.2
3.2
3.1
0.99
0.95
1.00
0.9
0.98
0.95
0.98
0.95
0.92
16.4
16.1
17.3
15.3
12.6
21.0
16.1
19.0
16.8
13.9
1.25
1.50
1.40
1.26
1.46
65 (0.08)
60 (0.13)
85 (0.21)
90 (0.05)
75 (0.20)
0.94
a
b
Average number of BODIPY monomers per polymer chain (with a DPn,total about 120; Figure SI-8, Supporting Information). Styrene conversion
c
determined by gravimetry. BODIPY conversion determined by SEC by comparison of the calculated area of the polymer and monomer peaks from
SEC equipped with a UV−vis detector.14 Theoretical number-average molar mass (Mn,th) (Mn,th = Mn CTA + 1/nCTA × (convnS × mS30 + convnπ ×
d
mπ), where CTA stands for chain transfer agent, convnS30 and convnπ the individual conversion of styrene and BODIPY monomer and m the mass of
e
monomer used in the synthesis). Number-average molar mass (Mn,SEC) and polydispersity index (Mw/Mn) determined by SEC using a polystyrene
calibration. z-average diameter (Dz) and dispersity factor (σ) determined by DLS.
f
F
dx.doi.org/10.1021/ma400590q | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 5. Monitoring of the synthesis of nanoparticles 4 (with πEtA) by size exclusion chromatography in THF. Left: RI detection. Right: UV−vis
detection at the maximal absorption of the monomer πEtA 4 (λ = 528 nm). (macro-RAFT agents: black line, PEO-C12; yellow line, PEO-b-PAA-C12;
PEO-b-PAA-b-P(S-co-π)-C12: gray dots, end of bulk polymerization; gray lines, during miniemulsion polymerization; green line, end of miniemulsion
polymerization.
Figure 6 shows the final SECs for the different NPs samples
at the end of the polymerization, without purification (Table 2,
Figure 7. Evolution of the average molar fraction of BODIPY
monomer in the copolymer with the overall molar monomer
conversion (symbols, experimental values; lines, calculated with the
reactivity ratios; initial molar fraction of BODIPY monomer = 2.6 mol
%). Monomers πPhMA 1 (green and line, rS = 0.25, rPhMA = 0.5018),
■
◆
πPhA 2 (gray
and line, r = 0.38 determined by the Kelen-Tudos
̈
̈
S
method, rPhA estimated at 0.50),31 πEtMA 3 (yellow ∗ and line, rS =
Figure 6. Size exclusion chromatograms (UV−vis. detection, λ = 528
nm) in THF for polymer chains of nanoparticles with different
BODIPY monomers (πPhMA 1 (green line), πPhA 2 (gray line),
πEtMA 3 (yellow line), πEtA 4 (pink line), and πS 5 (blue line)).
0.46, rEtMA = 0.38 ), πEtA 4 (pink and line, rS = 0.8, rEtA = 0.233),
πS 5 (blue and line, rS = 1, rS = 1).
32
▲
●
monomer (determined from the areas of the signal of the
polymer and the monomer in the size exclusion chromatogram
using UV−vis. detection, Figure 5, right), and nS, nπ, and ntotal
are respectively the mole number of styrene and BODIPY
monomer and the total mole number of styrene and BODIPY
used in the synthesis.
The symbols in Figure 7 are the experimentally determined
average molar fraction of the different BODIPY monomers
incorporated in the copolymers, plotted as a function of the
overall molar monomer conversion. Two tendencies can mainly
be observed: the BODIPY monomer is either quickly
incorporated in the polymer chain (πPhMA 1, πPhA 2 and
πEtMA 3) leading to a gradient of composition in the polymer
chain, or the BODIPY monomer is homogeneously and quasi
randomly distributed in the polymer backbone (πS 5 and πEtA
4). Unlike the other monomers, for BODIPY monomers πEtA
4 and πS 5, given the low amount of BODIPY monomer in the
copolymer (∼2.6 mol %), one may assume that the fluorescent
monomer units are well separated by large polystyrene
segments, which could have an impact on the fluorescence
properties.
NP1 to NP5 made with the different fluorescent monomers).
For all samples, the great majority of the fluorescent monomer
molecules was incorporated into the polymer chains (Vret ∼ 15
mL) and only a negligible fraction was left over (Vret ∼ 18 mL).
From the relative integration of the peaks it was calculated that
the individual monomer conversion of all BODIPY monomers
was high (conv. > 0.90), meaning that the new fluorescent
monomers 2−5 can be copolymerized with styrene under the
conditions established before for monomer 1. For all
fluorescent monomers the calculated (Mn,th) and experimental
(Mn,SEC) number-average molar mass values were close, and
Mw/Mn remained below 1.5, meaning that the polymerization
was under RAFT control.
In order to determine the distribution of the BODIPY
monomers along the polymer chains the individual conversions
of both styrene and of the BODIPY monomers were monitored
in the course of the nanoparticles synthesis. For each
copolymerization system, the molar fraction of BODIPY
incorporated into the copolymer could then be calculated
with respect to the overall molar conversion (Figure 7).
The overall monomer molar conversion was then determined
as follows:
As mentioned above, in a controlled radical polymerization,
all polymer chains possess approximately the same composition
and their microstructure is determined by the reactivity ratios
rM1 and rM2 in the terminal model. Reactivity ratios of simple
monomers possessing the same polymerizable functions as the
new fluorescent monomers have been extensively tabulated.18
The fluorophore framework of the BODIPY monomers is not
nS
ntotal
nπ
ntotal
convntotal (mol) = convnS
+ convnπ
(3)
Where convnS and convnπ are the individual conversion of
styrene (determined by gravimetry) and of the BODIPY
G
dx.doi.org/10.1021/ma400590q | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 8. Spectra of nanoparticles with different BODIPY monomers (πPhMA (green line), πPhA (gray line), πEtMA (yellow line), πEtA (pink
line), and πS (blue line) recorded in water. Left: absorption (full lines) and emission (dotted lines, λexc = 495 nm) spectra. Right: fluorescence decays
(light gray line is the instrument response function, λexc = 495 nm, λF = 543 nm).
Table 3. Fluorescence Properties of Different Nanoparticles Recorded in Water
a
b
c
d
NP
monomer
nπ
λabs max/nm
λF/nm
ΦF
⟨τ⟩ /ns
B
(×10‑7)/L·cm‑1·mol‑1
NP1
NP2
NP3
NP4
NP5
πPhMA 1
πPhA 2
πEtMA 3
πEtA 4
πS 5
3.0
2.9
3.2
3.2
3.1
529
529
528
528
527
544
544
542
544
542
0.20 ( 0.02)
0.35 ( 0.03)
0.28 ( 0.02)
0.23 ( 0.03)
0.34 ( 0.02)
2.8
3.6
2.7
2.6
3.1
7.7
14.0
11.0
9.5
13.0
a
b
Average number of BODIPY monomers per chain (nπ = [π]0/[RAFT]0). Average fluorescence quantum yield calculated for two different
c
nanoparticles synthesis and, in parentheses, deviation from the average. Average fluorescence lifetime (λexc = 495 nm, λF = 543 nm) calculated as
∞tI(t) dt
∞I(t) dt
∫
d
0
⟨τ⟩ ≡
(eq 1). Nanoparticles brightness, calculated using eq 4 with ε taken at λabs max (527 or 528 nm) and Nagg = 1750 (see Supporting
∫
0
Information for calculation).
conjugated to the polymerizable vinyl groups, and their
reactivity should therefore not significantly differ from that of
the corresponding simple monomer structures in their radical
copolymerization with styrene (even if steric effects may
arise34). The continuous lines in Figure 7 represent the average
molar fractions calculated by the Skeist equation35 knowing the
initial molar fraction of the BODPY monomer and using the
reactivity ratios tabulated for the corresponding model
monomers, i.e., styrene (S) for πS 5, phenyl methacrylate
(PhMA) for πPhMA 1 etc.).18,31−33 In all cases the plots
matched well the experimental data points. It can thus be
concluded that the tabulated reactivity ratios for model
monomers are valid to estimate the microstructure in the
fluorescent copolymer.
nanoparticles. Gradient compositions lead to an increased
proximity of the fluorescent monomer which might lead to
quenching caused by poorly fluorescent intrachain aggregates.
In addition, a favorable incorporation of the fluorescent
monomer from the beginning or at the end of the
polymerization might lead to particles where the fluorophores
are concentrated either in the inner core or close to the
hydrophilic shell.
According to our kinetic study, monomer πS 5 is
homogeneously spread along the polymer chain and the
resulting particles possess high quantum yields and brightness.
On the other hand, πPhMA 1 is heterogeneously spread along
the polymer chain and leads to nanoparticles with much poorer
fluorescence properties. For these two examples, it seems that a
correlation between the microstructure of the polymer chains
and fluorescence properties exists.
In the next part, the fluorescence properties of the different
fluorescent nanoparticles are studied.
Spectroscopic Characterization of the Fluorescent
Nanoparticles. Absorption and fluorescence emission spectra
of the nanoparticles are reported in Figure 8. It appears that the
nanoparticles in water have the same absorption and emission
spectra as their monomers recorded in toluene. In order to
evaluate the formation of poorly fluorescent aggregates, the
quantum yield and average fluorescence lifetimes (see eq 1) of
the nanoparticles were compared (Table 3). Particles with the
lowest quantum yield and lifetimes were particles with a core
made of the monomers πPhMA 1 and πEtA 4 (ΦF < 0.25 and
⟨τ⟩ < 3 ns). In contrast, particles with the highest quantum
yield were particles with a core made of πS 5 and πPhA 2 (ΦF >
0.30 and ⟨τ⟩ > 3 ns). Particles with a core of πEtMA 3 showed
intermediate properties.
However, regarding monomer 4 (πEtA), which possesses
also a quasi-random distribution of the fluorophores in the
polymer chain, a poor quantum yield and the lowest average
fluorescence lifetime of the series were determined. A quasi-
random distribution of the BODIPY monomer in the polymer
chain constituting the nanoparticle (case of monomer πS 5 and
πEtA 4) does not necessarily lead to the most fluorescent
particles, and vice versa, BODIPY monomers which are not
evenly spread along the polymer backbone (for instance πPhA)
do not unavoidably lead to the less fluorescent particles. It must
therefore be concluded that the fluorescence properties
(quantum yield and lifetime) do not exclusively depend on
the distribution of the fluorescent monomer in the polymer
chain, but that other parameters must also be considered, such
as the organization of the polymer chains in the nanoparticle
during the assembly step occurring during synthesis, the
formation of interchain aggregates etc. Indeed, because of the
confined space of the polymeric chains in the nanoparticles,
How can the differences in the fluorescence properties be
explained? One possible explanation which might be considered
relies on the differences of the fluorescent monomer
distribution in the copolymer chain constituting the fluorescent
H
dx.doi.org/10.1021/ma400590q | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
aggregates of fluorescent monomers between different polymer
chains may form. Their formation might depend on the
BODIPY monomers’ orientation in the polymer chain or on
differences in solubility of the monomers in the organic phase
(made of styrene and PEO-b-PAA-C12 macro-RAFT agent),
possibly leading to a local phase separation during the
polymerization process.
The nanoparticle brightness (B), which is a function of the
quantum yield and in practice one of the most important
features for biological imaging, was also evaluated from eq 4:
CONCLUSIONS
■
Five BODIPY derivatives with different polymerizable functions
were successfully synthesized and their fluorescence properties
were studied in toluene. All fluorophores could be successfully
copolymerized at 2.6 mol % with styrene in a controlled
manner via our formerly established one-pot RAFT mini-
emulsion polymerization process in the absence of surfactant.
This route to form fluorescent polymeric core shell nano-
particles is thus a robust method and should be applicable to a
large variety of fluorescent molecules. We have further
demonstrated that the microstructure of the resulting
copolymers is governed by the nature of the polymerizable
function on the BODIPY, and that the fluorescent monomer
distribution can be predicted by using reactivity ratios tabulated
for the corresponding model monomers.
The fluorescence features of the nanoparticles were studied
by stationary and time-resolved spectroscopy. Surprisingly, the
most fluorescent nanoparticles (in terms of lifetime, quantum
yield and thus brightness) were not necessarily obtained with
the quasi-random polymer structure. It must thus be concluded
that not only intrachain aggregates of BODIPY monomers
must be considered, but that fluorophores’ interchain
aggregates and maybe the solubility of the different fluorescent
monomers in the nanoparticle core matrix are of crucial
importance for the outcome of the fluorescence properties.
Nevertheless, simply by changing the nature of the polymer-
izable function on the BODIPY (to a phenyl acrylate or styrene
function), the brightness of the nanoparticles could be
improved by a factor of 2 compared to the formerly studied
BODIPY monomer bearing a phenyl methacrylate function.
Those ultrabright nanoparticles were found to be 200 to 2000
times brighter than usual quantum dots and 40−300 times
brighter than other polymeric nanoparticles reported in the
literature.
B = (επ × Nπ) × ϕF
(4)
where επ is the molar coefficient extinction of the BODIPY
monomer at the excitation wavelength, Nπ the number of
BODIPY per particle (equation SI-3, Supporting Information)
and ΦF the quantum yield of the fluorescent nanoparticle.
For all systems, brightness values higher than 7 × 107
cm−1·mol−1·L were obtained (Table 3). They were all in the
same order of magnitude, but NP containing monomer πPhA 2
and πS 5, possessing the best quantum yields, were the
brightest ones (respectively 14 × 107 and 13 × 107
cm−1·mol−1·L). For comparison, the NP prepared in our
former study with monomer πPhMA 1 had a brightness of 7.7
× 107 cm−1·mol−1·L. It is thus possible to increase the
brightness of the fluorescent nanoparticles by a factor of 1.8,
only by changing the nature of the fluorescent monomer in the
core.
Quantum dots are famous for their extremely high brightness
making them ideal candidates for biological imaging. Generally,
quantum dots with a core made of CdSe, CdS, or CdTe
emitting between 370 and 750 nm have a brightness between 6
× 104 and 6 × 105 cm−1·mol−1·L.36 The fluorescent
nanoparticles NP2 with a core of monomer πPhA 2 are thus
200 to 2000 times brighter than those quantum dots.
Compared to most of other systems based on fluorescent
organic polymeric nanoparticles, NP2 are on average 40 to 300
times brighter. For instance, Sun et al. have synthesized
nanoparticles (diameter: 20−25 nm) made of PEO113-b-
PVBA46-C12 (PVBA: poly(4-vinylbenzaldehyde)) function-
alized with a fluorescein dye. The best nanoparticles have a
ASSOCIATED CONTENT
■
S
* Supporting Information
Synthetic schemes for BODIPY monomers 1−5 and the
fluorescent nanoparticles; numbering of protons and carbons of
BODIPY monomers 2−5; determination of the reactivity ratio
rs for the copolymerization of styrene with πPhA 2 by the
Kelen-Tudos method; fluorescence decays, fit, instrument
̈
̈
1
brightness of 5.1 × 105 cm−1.mol−1.L.3b Mea
́
llet-Renault et al.
response function (IRF), and residuals for all monomers; H
NMR spectra of dried and crude NP4; and determination of
the aggregation number and the number of BOPIDY units per
NP. This material is available free of charge via the Internet at
http://pubs.acs.org.
have loaded 16 nm diameter polystyrene nanoparticles with a
BODIPY derivative dye by the solvent displacement method.37
With approximately 76 BODIPY molecules per particle and a
quantum yield of 0.77, the nanoparticles had a brightness of 3.4
× 106 cm−1.mol−1.L.
AUTHOR INFORMATION
■
In conclusion, it is possible to modulate the fluorescence
efficiency simply by changing the nature of the fluorescent
BODIPY monomer, and thus its distribution within the
polymer backbone. Nevertheless, it is difficult to predict the
formation of BODIPY aggregates in the fluorescent polymer
nanoparticles, by considering exclusively aggregates along the
polymer backbone (intrachain aggregates) and disregarding
aggregate formation between fluorescent polymer chains
(interchain aggregates).
Finally, it should be emphasized that the BODIPY monomer
with a phenyl acrylate polymerizable function (πPhA 2) is the
best candidate for future NP design since it is the easiest to
synthesize leading to ultrabright NP.
Corresponding Author
*E-mail: jutta.rieger@upmc.fr (J.R.); gilles.clavier@ppsm.ens-
cachan.fr (G.C.).
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank Gaelle Pembouong (LCP) for technical
̈
support on SEC analyses and Arnaud Brosseau (PPSM) for
technical support on fluorescence spectroscopy measurements.
B. C. acknowledges the Institut Universitaire de France for her
nomination as a senior member.
I
dx.doi.org/10.1021/ma400590q | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
1
comparing gravimetric results and H NMR spectra (CDCl3, residual
vinylic protons; see Figure SI-9, Supporting Information).
REFERENCES
■
(1) (a) Rosi, N. L.; Mirkin, C. A. Chem. Rev. 2005, 105, 1547−1562.
(b) Louie, A. Chem. Rev. 2010, 110, 3146−3195. (c) Hu, J.; Liu, S.
Macromolecules 2010, 43, 8315−8330.
(2) (a) Qin, W.; Rohand, T.; Dehaen, W.; Clifford, J. N.; Driesen, K.;
Beljonne, D.; Van Averbeke, B.; Van der Auweraer, M.; Boens, N. J
Phys. Chem. A 2007, 111, 8588−8597. (b) Sauer, R.; Turshatov, A.;
Baluschev, S.; Landfester, K. Macromolecules 2012, 45, 3787−3796.
(3) (a) Palma, A.; Tasior, M.; Frimannsson, D. O.; Vu, T. T.;
(31) To the best of our knowledge, the reactivity ratio for the
copolymerization of styrene with phenyl acrylate have not been
tabulated. We therefore determined r with the Kelen-Tudos method
̈
̈
S
for the copolymerization of styrene with πPhA 2 (see Figure SI-2,
Supporting Information). The value for rS was 0.37 0.03.
(32) Tacx, J. C. J. R.; Ammerdorffer, J. L.; German, A. L. Polym. Bull.
1984, 12, 343−348.
(33) Otsu, T.; Ito, T.; Fukumizu, T.; Imoto, M. Bull. Chem. Soc. Jpn.
1966, 39, 2257−2260.
́
Meallet-Renault, R.; O’Shea, D. F. Org. Lett. 2009, 11, 3638−3641.
(b) Sun, G.; Berezin, M. Y.; Fan, J.; Lee, H.; Ma, J.; Zhang, K.; Wooley,
K. L.; Achilefu, S. Nanoscale 2010, 2, 548−558.
(34) Bisht, H. S.; Ray, S. S.; Chatterjee, A. K. J. Polym. Sci.Part A:
Polym. Chem. 2003, 41, 1864−1866.
(4) Hu, J.; Zhang, X.; Wang, D.; Hu, X.; Liu, T.; Zhang, G.; Liu, S. J.
Mater. Chem. 2011, 21, 19030−19038.
(35) Odian, G. (ed.), Principles of polymerization, 4th ed.; Wiley-
Interscience: New York, 2004; p 476.
(5) Sun, H.; Scharff-Poulsen, A. M.; Gu, H.; Almdal, K. Chem. Mater.
2006, 18, 3381−3384.
(36) Resch-Genger, U.; Grabolle, M.; Cavaliere-Jaricot, S.; Nitschke,
R.; Nann, T. Nat. Meth. 2008, 5, 763−775.
(6) Montalti, M.; Prodi, L.; Zaccheroni, N.; Zattoni, A.; Reschiglian,
P.; Falini, G. Langmuir 2004, 20, 2989−2991.
́
(37) Meallet-Renault, R.; Herault, A.; Vachon, J.-J.; Pansu, R. B.;
Amigoni-Gerbier, S.; Larpent, C. Photochem. Photobiol. Sci. 2006, 5,
300−310.
(7) Matyjaszewski, K. Macromolecules 2012, 45, 4015−4039.
(8) Moad, G.; Rizzardo, E.; Thang, S. H. Polymer 2008, 49, 1079−
1131.
(9) C. Barner-Kowollik (ed.), Handbook of RAFT Polymerization;
Wiley-VCH: Weinheim, Germmany, 2008.
(10) Nagai, A.; Yoshii, R.; Otsuka, T.; Kokado, K.; Chujo, Y.
Langmuir 2010, 26, 15644−15649.
(11) (a) You, J.; Yoon, J. A.; Kim, J.; Huang, C.-F.; Matyjaszewski, K.;
Kim, E. Chem. Mater. 2010, 22, 4426−4434. (b) Li, C.; Zhang, Y.; Hu,
J.; Cheng, J.; Liu, S. Angew. Chem., Int. Ed. 2010, 49, 5120−5124.
(12) Matyaszewski, K., Gnanou, Y., Leibler, L., Eds. Macromolecular
Engineering; Wiley-VCH: Weinheim, Germany, 2007.
(13) Tronc, F.; Li, M.; Lu, J.; Winnik, M. A.; Kaul, B. L.; Graciet, J.-C.
J. Polym. Sci., Part A: Polym. Chem. 2003, 41, 766−778.
́
(14) Grazon, C.; Rieger, J.; Meallet-Renault, R.; Clavier, G.;
Charleux, B. Macromol. Rapid Commun. 2011, 32, 699−705.
(15) Loudet, A.; Burgess, K. Chem. Rev. 2007, 107, 4891−4932.
́
(16) Grazon, C.; Rieger, J.; Charleux, B.; Clavier, G.; Meallet-Renault,
R. To be published.
(17) Grazon, C. Ph.D. Dissertation. ENS Cachan: Cachan, France,
2012.
(18) Brandrup, J.; Immergut, E. H.; Grulke, E. A.; Abe, A.; Bloch, D.
R., Polymer Handbook; John Wiley & Sons, Inc: New York, 1999; pp
II-181−II-308.
(19) (a) Yan, Y.; Lu, N.; Cui, J.; Zhang, J. J. Appl. Polym. Sci. 2012,
125, 2867−2873. (b) Morishima, Y.; Sang Lim, H.; Nozakura, S.-i.
Macromolecules 1989, 22, 1148−1154.
(20) Couvreur, L.; Lefay, C.; Belleney, J.; Charleux, B.; Guerret, O.;
Magnet, S. Macromolecules 2003, 36, 8260−8267.
(21) Kubin, R. F.; Fletcher, A. N. J. Lumin. 1982, 27, 455−462.
(22) Valeur, B. Molecular Fluorescence: Principles and Applications;
Wiley-VCH: Weinheim, Germmany, 2002.
(23) Dumas-Verdes, C.; Miomandre, F.; Lep
Truc Vu, T.; Clavier, G.; Meallet-Renault, R.; Audebert, P. Eur. J. Org.
Chem. 2010, 2525−2535.
́
icier, E.; Galangau, O.;
́
(24) Roncucci, G.; Dei, D.; Giuntini, F.; Chiti, G.; Nistri, D.; Fantetti,
L.; Paschetta, V.; Cocchi, A. WO/2004/035590 patent, 2004.
(25) Kikuchi, D.; Sakaguchi, S.; Ishii, Y. J. Org. Chem. 1998, 63,
6023−6026.
(26) Wang, S.; Li, X.; Xun, S.; Wan, X.; Wang, Z. Y. Macromolecules
2006, 39, 7502−7507.
(27) Boyer, J. H.; Haag, A. M.; Sathyamoorthi, G.; Soong, M.-L.;
Thangaraj, K.; Pavlopoulos, T. G. Heteroatom Chem. 1993, 4, 39−49.
(28) Garcia-Moreno, A. C. I.; Campo, L.; Sastre, R.; Amat-Guerri, F.;
Liras, M.; Lopez Arbeloa, F.; Banuelos Prieto, J.; Lopez Arbeloa, I. J.
Phys. Chem. A 2004, 108, 3315−3323.
(29) Charleux, B.; Delaittre, G.; Rieger, J.; D’Agosto, F. Macro-
molecules 2012, 45, 6753−6765.
(30) Annotation by the author: mass of styrene, corrected by the loss
of styrene by evaporation during the sonication step (25%). For each
experiment, a loss of 25% of styrene has been determined by
J
dx.doi.org/10.1021/ma400590q | Macromolecules XXXX, XXX, XXX−XXX