Direct monitoring of self-assembly of copolymeric micelles by a luminescent molecular rotor

Article abstract of DOI:10.1039/c3cc44590a

The detailed analysis of the time resolved luminescence of a specifically designed fluorescent molecular rotor enables the direct monitoring of the self-assembly of a poly(N,N-dimethylacrylamide)-block-polystyrene (PDMA-b-PS) copolymer into core-corona nanoparticles. Comparison with bulk PS indicates hard confinement of the micelle core, due to the solvation of the hydrophilic PDMA corona. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c3cc44590a

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Direct monitoring of self-assembly of copolymeric
micelles by a luminescent molecular rotor†
Cite this: Chem. Commun., 2013,
49, 8474
Gianfranco Vaccaro, Alberto Bianchi, Michele Mauri, Simone Bonetti,
Francesco Meinardi, Alessandro Sanguineti, Roberto Simonutti* and
Luca Beverina*
Received 19th June 2013,
Accepted 23rd July 2013
DOI: 10.1039/c3cc44590a
www.rsc.org/chemcomm
The detailed analysis of the time resolved luminescence of a specifically alterations.5 Finally, solution calorimetric techniques are strongly
designed fluorescent molecular rotor enables the direct monitoring of affected by the use of high thermal capacity solvents (like water),
the self-assembly of a poly(N,N-dimethylacrylamide)-block-polystyrene intrinsically slow and insensitive to disorganized chains.6 The direct
(PDMA-b-PS) copolymer into core–corona nanoparticles. Comparison observation of self-assembly dynamics in terms of organized vs.
with bulk PS indicates hard confinement of the micelle core, due to the disorganized chains at any given time and under experimental
solvation of the hydrophilic PDMA corona.
conditions is of crucial importance, particularly for the use of
copolymer micelles in drug delivery. In fact, only if the core-forming
Amphiphilic block copolymers dispersed in water can self- block is below Tg the hydrophobic payload will be trapped by frozen
assemble into a large variety of aggregate structures. Depending polymer chains, eventually allowing selective release.7
on the hydrophilic vs. hydrophobic monomers molar ratio,
Time Resolved Fluorescence Spectroscopy (TRFS) of suitable
overall chain length and polydispersity, the polymers arrange fluorescent probes gives a direct and fast access to the polymer
as spherical micelles, vesicles, polymersomes.1 The applications aggregation states under various experimental conditions. For
of such nanostructures range from micro/nanoelectronics and instance, the decay of excimers in pyrene labelled polymers is a
photonics to bioimaging and drug delivery.2
phenomenon connected to the aggregation state,8 but the need for
In the case of amphiphilic copolymers, the addition of excess covalent functionalization of the polymer limits its scope.9 Molecular
water to a solution of the copolymer in a non-selective solvent rotors, a class of fluorophores whose emission intensity and lifetime
induces the assembly of nanostructures. Established experimental strongly depends on local viscosity, can be directly used – with no
methods are generally effective, but still lack a clear rationalization.3 need for chemical functionalization – to probe the aggregation state
In situ, fast monitoring of the micelle formation cannot be easily in a variety of experimental situations.10 Rotors are for example
accessed through common macromolecular characterization tools. frequently used to study bulk polymerization dynamics11 as well as
Solution Nuclear Magnetic Resonance (NMR) is quite sensitive to aggregation to amyloid fibres in biomedical assays.12
molecular aggregation, mostly evidenced through broadening of
We report here for the first time the use of TRFS of a specifically
resonance line-widths. In a core–corona micelle, however, if the core designed molecular rotor as a simple, fast and versatile method for
is below the glass transition (Tg) temperature, its NMR signals the direct monitoring of the self-assembly of block copolymers in
completely disappear as a consequence of their extreme broadening, solution into core–corona micelles suitable for application in drug
making it thus impossible to quantify for example the molar fraction delivery. We show in particular that the fluorescence lifetime of our
of organized vs. disorganized polymeric chains.4 Conversely, rotor is a sensitive gauge of the molar fraction of disorganized vs.
Dynamic Light Scattering (DLS) provides the micelle hydrodynamic self-assembled copolymeric chains as a function of the experimental
volume, but gives no information about disorganized polymers conditions. We also show that the core of the core–corona micelles is
chains and self-organization dynamics. Transmission Electron below its Tg temperature, irrespective of its nanometric diameter.
Microscopy (TEM) and cryo-TEM images are extremely detailed but
The block copolymer of choice is poly(N,N-dimethylacrylamide)-
require removal or freezing of the solvent, sometimes with extreme block-polystyrene (PDMA-b-PS) obtained by reversible addition–
fragmentation chain transfer (RAFT) polymerization.13 The fine
`
Dipartimento di Scienza dei Materiali, Universita degli Studi di Milano Bicocca,
tuning of reaction conditions (see ESI† for synthesis and characteriza-
tion of the PDMA-b-PS copolymer) enabled the preparation of the
target polymer PDMA817-b-PS105, a molar ratio suitable for the for-
mation of core–corona micelles (Scheme 1a). The fluorescent probe of
choice is the donor–acceptor derivative AzeNaph1 (Scheme 1b) whose
Via Roberto Cozzi, 53, 20125 Milano, Italy.
E-mail: roberto.simonutti@mater.unimib.it, luca.beverina@unimib.it
† Electronic supplementary information (ESI) available: Synthesis and NMR
characterization of all new materials. Details on DLS and TEM characterization,
details on the luminescence behaviour of rotors. See DOI: 10.1039/c3cc44590a
c
8474 Chem. Commun., 2013, 49, 8474--8476
This journal is The Royal Society of Chemistry 2013

View Article Online
Communication
ChemComm
Fig. 1 (left) DLS spectrum of PDMA817-b-PS105 dispersed in water. (right) TEM
image of the PDMA817-b-PS105 copolymer.
of solvent evaporation. No contrast reagents were employed, the
contrast being connected with the higher density of the closely
packed polystyrene cores with respect to the disorganized PDMA
1
surrounding. Finally, the H NMR spectrum of the copolymer in
Scheme 1 (a) Synthesis of PDMA-b-PS. (b) Synthesis of AzeNaph1.
CDCl3 (Fig. S8, ESI†) shows all the signals of both blocks, whilst in
D2O, only PDMA signals can be distinguished as those of PS are
broadened by their aggregation in the micelle core.4
fluorescence lifetime strongly depends on the rotation of its electron
donating and rigid dibenzoazepine residue with respect to
the electron deficient naphthalene imide residue. Such rotation is
progressively hindered upon increase of the local viscosity and is
eventually almost stopped in a rigid system. We designed this
particular structure in order to provide at the same time negligible
water solubility (common rotors like thioflavine and 2-(4-(dimethyl-
amino)benzylidene)malononitrile12 are at least partially water soluble,
see ESI† for details), sensitivity at the high viscosity limit, high Stokes
shift (to limit re-absorption) and stability.
Photophysics of AzeNaph1 is especially suitable for studying
nanoparticle formation processes. First of all, its fluorescence lifetime
is extremely sensitive to the environment, ranging from hundred ps
for liquids and polymers above Tg to 6–8 ns for glassy polymers, with a
lifetime characteristic of the employed polymer (see ESI† for details).
Just by recording the rotor luminescence lifetime as a function of the
polymerization time, it is possible to easily follow the bulk polymer-
ization kinetics of styrene into polystyrene (see Fig. S2, ESI†).
Even more interestingly, the emission decay of AzeNaph1 behaves
as a perfect single exponential in homogeneous samples. The decay
is fast in liquid samples, slow in solid ones but always well fitting in
a monoexponential law. This is not a common characteristic
of standard molecular rotors. For example, both thioflavine and
2-(4-(dimethylamino)benzylidene)malononitrile show a biexponential
decay under all experimental conditions (see ESI†). This is the reason
why in all previous studies of polymerization kinetics based on the
luminescence of a probe, authors mainly focused on the lumines-
cence intensity instead of lifetime.10,11 As shown in the following,
time resolved luminescence is a more appropriate and informative
technique, particularly in the case of inhomogeneous samples like
core–corona micelles.
All techniques confirm the formation of core–corona
micelles in the presence of excess water. None of them however
gave us hints on the self-assembly dynamics and on the
minimum amount of water required to complete the process.
We repeated the micelles preparation, in this case starting from a
copolymer–AzeNaph1 (99.9 : 0.1 weight ratio) solution in DMF. We
monitored the fluorophore emission lifetime as a function of the
water added from 5% v/v to 500% v/v. After each aliquot the sample
was equilibrated for 2 minutes (as shown in the ESI,† after 2 min
resting time the sample emission behaviour is constant). Fig. 2
shows the AzeNaph1 photoluminescence (PL) decay profile in pure
DMF (blue curve), at 20% v/v (red curve) and 500% v/v (black curve)
water content.
The green curve indicates the AzeNaph1 decay profile in bulk PS.
First of all, AzeNaph1 PL decay at 500% v/v water is superimposable
with that in bulk PS. This demonstrates that the local environment
experienced by the probe in micelles is indistinguishable from bulk
PS, in agreement with DLS, TEM and NMR characterization
discussed before. This is also a clear indication that in the presence
of excess water all of the rotor molecules are embedded in the
nanoparticle core. In the presence of a large excess of water all of the
polymers assume a core–corona structure. The rotor, completely
insoluble in water, loads in the rigid apolar PS core behaving
During its self-assembly, the block copolymer we prepared goes
from a random coil configuration to a core–corona structure having
a dense, stiff PS core and a swelled PDMA corona. Such behaviour is
clearly evidenced by DLS, TEM and NMR characterization. In short,
we promoted the self-assembly by the slow addition of deionized
water to a copolymer solution in dimethylformamide (DMF). DLS
analysis shows that the micelles had a hydrodynamic radius (Rh) of
62 nm (Fig. 1, left). The TEM image (Fig. 1, right) obtained after
solvent evaporation shows the PS cores as black spots of 8–10 nm
radius whilst the PDMA coronas are disorganized as a consequence
Fig. 2 Time decay profile of AzeNaph1 in DMF solution containing the PDMA817
-
b-PS105 copolymer before addition of water (blue curve) and after addition of 20%
v/v (red curve) and 500% v/v (black curve) water. The decay profile of AzeNaph1
dispersed in a polystyrene bulk sample (green curve) is shown as a reference.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 8474--8476 8475

View Article Online
ChemComm
Communication
accordingly. Conversely, the probe decay in pure DMF is very fast. PDMA external corona of the micelles induces a ‘‘hydrostatic’’
The red curve shows that at intermediate water content (20%) the pressure on the core surface. This pressure brings the core to the
probe decay can be fitted in terms of a biexponential decay having a ‘‘hard confinement’’ limit, discussed by the same authors in the case
fast component identical to that measured in DMF and a slow one of silica capped PS nanoparticles.15 In this case the Tg is no longer
corresponding to that observed in bulk PS. The probe is in this size dependent. This is a very relevant finding as it ensures that a
case partially loaded into the rigid cores of micelles and partially drug loaded in the micelle core will not be able to escape, through
dispersed in solution along with the still disorganized copolymer diffusion in a low viscosity environment, prior to selective delivery to
chains. All of the exponential decays obtained for water contents the appropriate target.
going from 1 to 60% v/v can be fitted in terms of a biexponential
In conclusion, we successfully exploited the time resolved
decay having the same fast and slow components. We fitted all the fluorescence decay of a suitable molecular probe to characterize
normalized data by means of eqn (1),
relevant features of the self-assembly of amphiphilic block
copolymers in water solution: minimum amount of water required
to obtain complete self-assembly, minimum equilibration time of
the dispersion after changing the composition, and aggregation
state of the nanoparticle core. Our analysis is simple, quantitative,
fast and non-destructive. These results may support the prepara-
tion strategies of self-assembly of amphiphilic block copolymers for
application in drug delivery, a research field actually particularly
active and technologically relevant.
t
t
À
Á
PLðtÞ ¼ Ageꢀ
þ 1 ꢀ Ag eꢀ
(1)
tslow
tfast
where PL(t) is the time dependent PL intensity, tfast = 0.8 ns and
tslow = 5.8 ns are the AzeNaph1 lifetime in solution and in bulk PS,
respectively. The pre-exponential factor Ag corresponds to the frac-
tion of slow decaying probes, which, in the case of complete
assembly, is expected to be 1.
Indeed, Fig. 3 shows that the Ag vs. water content plot assumes
the form of a titration curve, giving in real time for any specific
water content the amount of organized vs. disorganized polymer
chains. It should be noted that such analysis greatly profits from
the peculiar emission properties of AzeNaph1. As mentioned
previously, our rotor features simple monoexponential decay in
all homogeneous samples and a biexponential decay only in the
heterogeneous ones. The other commercially available rotors we
tested (see ESI†) follow biexponential decay in both homogeneous
and heterogeneous samples, thus making data interpretation
significantly more complicated.
Our analysis leads to three relevant results. First of all, we clearly
show that a 60% water content is enough to induce the complete
self-organization of all polymer chains. The large excess of water,
prescribed by well-established literature procedures, in our case is
unnecessary. Secondly, the self-assembly is a fast process. In all of
the cases, the PL signal measured was completely constant for
2 minutes after the addition of water (see Fig. S10 of the ESI†).
Finally, our characterization clearly shows that the PS core (whose
radius is below 10 nm) is below its Tg. This finding apparently
disagrees with the work of Priestley and coworkers,14 showing that
the Tg of PS nanoparticles is size dependent below 100 nm of radius.
The case of core–corona nanoparticles is different. The swollen
Financial support from the Fondazione Cariplo grant 2010-
0564 ‘‘LumiPhoto’’ is gratefully acknowledged. M.M. thanks the
‘‘Dote Ricercatore’’ FSE project of Regione Lombardia.
Notes and references
1 E. B. Zhulina and O. V. Borisov, Macromolecules, 2012, 45, 4429–4440.
2 H.-C. Kim, S.-M. Park and W. D. Hinsberg, Chem. Rev., 2010, 110,
146–177.
3 H. Huang, B. Chung, J. Jung, H. W. Park and T. Chang, Angew.
Chem., Int. Ed., 2009, 48, 4594–4597.
4 (a) J. S. Hrkach, M. T. Peracchia, A. Domb, N. Lotan and R. Langer,
Biomaterials, 1997, 18, 27–30; (b) C. R. Heald, S. Stolnik,
K. S. Kujawinski, C. De Matteis, M. C. Garnett, L. Illum, S. S. Davis,
S. C. Purkiss, R. J. Barlow and P. R. Gellert, Langmuir, 2002, 18,
3669–3675; (c) G. M. Wilmes, D. J. Arnold and K. S. Kawchak, J. Polym.
Res., 2011, 18, 1787–1797; (d) K. Vijayakrishna, D. Mecerreyes,
Y. Gnanou and D. Taton, Macromolecules, 2009, 42, 5167–5174.
5 T. Azzam and A. Eisenberg, Langmuir, 2010, 26, 10513–10523.
6 (a) C. J. Lai, W. B. Russel and R. A. Register, Macromolecules, 2002,
35, 841–849; (b) J. M. Virgili, A. Hexemer, J. A. Pople, N. P. Balsara
and R. A. Segalman, Macromolecules, 2009, 42, 4604–4613.
7 (a) K. T. Kim, J. Cornelissen, R. J. M. Nolte and J. C. M. van Hest, Adv.
Mater., 2009, 21, 2787–2791; (b) W. Scarano, H. T. T. Duong,
H. X. Lu, P. L. De Souza and M. H. Stenzel, Biomacromolecules,
2013, 14, 962–975; (c) S. Hocine, D. Cui, M. N. Rager, A. Di Cicco,
J. M. Liu, J. Wdzieczak-Bakala, A. Brulet and M. H. Li, Langmuir,
2013, 29, 1356–1369.
8 F. M. Winnik, Chem. Rev., 1993, 93, 587–614.
9 (a) M. Wilhelm, C. L. Zhao, Y. C. Wang, R. L. Xu, M. A. Winnik,
J. L. Mura, G. Riess and M. D. Croucher, Macromolecules, 1991, 24,
1033–1040; (b) S. C. Lee, Y. K. Chang, J. S. Yoon, C. H. Kim,
I. C. Kwon, Y. H. Kim and S. Y. Jeong, Macromolecules, 1999, 32,
1847–1852; (c) R. Rutkaite, L. Swanson, Y. Li and S. P. Armes,
Polymer, 2008, 49, 1800–1811; (d) M. M. Mok and T. P. Lodge,
J. Polym. Sci., Part B: Polym. Phys., 2012, 50, 500–515.
10 (a) M. A. Haidekker and E. A. Theodorakis, J. Biol. Eng., 2010, 4;
(b) K. Y. Law, Chem. Phys. Lett., 1980, 75, 545–549; (c) Y. Shiraishi,
T. Inoue and T. Hirai, Langmuir, 2010, 26, 17505–17512.
11 R. O. Loufty, Pure Appl. Chem., 1986, 58, 1239–1248.
12 N. Amdursky, Y. Erez and D. Huppert, Acc. Chem. Res., 2012, 45,
1548–1557.
13 (a) G. Moad, E. Rizzardo and S. H. Thang, Polymer, 2008, 49,
1079–1131; (b) K. H. Wong, T. P. Davis, C. Barner-Kowollik and
M. H. Stenzel, Polymer, 2007, 48, 4950–4965.
14 R. D. Priestley, C. J. Ellison, L. J. Broadbelt and J. M. Torkelson,
Science, 2005, 309, 456–459.
15 C. Zhang, Y. L. Guo and R. D. Priestley, Macromolecules, 2011, 44,
4001–4006.
Fig. 3 Relative population with tslow, obtained by fitting the time decay profiles
of AzeNaph1 to eqn (1), as a function of the H2O added.
c
8476 Chem. Commun., 2013, 49, 8474--8476
This journal is The Royal Society of Chemistry 2013