Supramolecular assemblies from amphiphilic homopolymers: Testing the scope

Article abstract of DOI:10.1021/ja065213o

It has been shown by us in a recent communication that homopolymers, in which each repeat unit contains a hydrophilic and a hydrophobic head group, are capable of forming environment-dependent micellar or inverse micellar assemblies. A systematic structure-property relationship study is carried out here to test the scope of the design. We show here that the molecular design is indeed broadly applicable and that there is a significant gain in the critical aggregation concentrations of these polymers, as compared to the small molecule counterparts. We also show that the design can be tuned to achieve vesicle-type assemblies, which further expands the repertoire of amphiphilic homopolymers in a variety of areas. Characterizations of these assemblies have been carried out using transmission electron microscopy, dynamic light scattering, static light scattering, and dye incorporation experiments.

Full text of DOI:10.1021/ja065213o

Published on Web 11/24/2006
Supramolecular Assemblies from Amphiphilic
Homopolymers: Testing the Scope
Elamprakash N. Savariar, Sivakumar V. Aathimanikandan, and S. Thayumanavan*
Contribution from the Department of Chemistry, UniVersity of Massachusetts,
Amherst, Massachusetts 01003
Received July 20, 2006; E-mail: thai@chem.umass.edu
Abstract: It has been shown by us in a recent communication that homopolymers, in which each repeat
unit contains a hydrophilic and a hydrophobic head group, are capable of forming environment-dependent
micellar or inverse micellar assemblies. A systematic structure-property relationship study is carried out
here to test the scope of the design. We show here that the molecular design is indeed broadly applicable
and that there is a significant gain in the critical aggregation concentrations of these polymers, as compared
to the small molecule counterparts. We also show that the design can be tuned to achieve vesicle-type
assemblies, which further expands the repertoire of amphiphilic homopolymers in a variety of areas.
Characterizations of these assemblies have been carried out using transmission electron microscopy,
dynamic light scattering, static light scattering, and dye incorporation experiments.
Introduction
providing both micelle-like and inverse micelle-like assemblies
depending on the solvent environment, which is an amplified
Molecules based on amphiphilic building blocks are capable
of providing diverse self-assembled structures, because of their
differential interaction energy with the solvent surface.1 Such
amphiphilic assemblies could find use in a variety of applications
ranging from biology to materials.2 There is significant interest
in polymer-based supramolecular assemblies, because of the
enhanced stabilities and lower critical aggregation concentrations
that these macromolecules are capable of providing.2d Block
copolymers have been a popular choice for these types of
assemblies, where one of the blocks is relatively more incom-
patible with the solvent milieu.3 We have recently developed a
molecular design based on homopolymers in which both the
hydrophilic and the hydrophobic moieties are incorporated
within the monomer unit.4 In that preliminary communication,
we had demonstrated that such homopolymers are capable of
consequence of the molecular level conformational changes in
each monomer unit. Since then, we have shown that such
polymer assemblies are indeed unique and therefore could find
interesting applications in a broad range of areas.5-7 For
example, we have reported that the micellar interiors of these
polymers could be used as nanocontainers to carry out organic
photochemical reactions and that the selectivities in these
reactions are much higher than those obtained with micellar
containers based on block copolymers or small molecule
surfactants.5 We have also demonstrated that these environment-
dependent polymer assemblies are unique in that these are
kinetically trapped in the solvent used for initial assembly.6 We
have demonstrated that such a feature could be used as
nanocontainers for separation of organic molecules.6
Considering the possible implications in a variety of areas,
it is important that we fully test the scope of our molecular
design. Here, we describe the details of a systematic structure-
property relationship study. We address the following in this
paper: (i) Could one apply the structural guidelines developed
using our styrene-based polymers for the formation of responsive
micelles to other polymer backbones? In other words, are the
design guidelines applicable broadly? (ii) How does the nature
of the micellar interior vary with the backbone of the polymer
and the hydrophobic functionality? (iii) What are the critical
micelle concentrations (cmc) of these polymers? How does this
vary with the structure of the building blocks of the polymer?
(1) (a) Israelachvili, J. N. Intermolecular and Surface Forces, 2nd ed.;
Academic Press: New York, 1991. (b) Muthukumar, M.; Ober, C. K.;
Thomas, E. L. Science 1997, 277, 1225-1232. (c) Bucknall, D. G.;
Anderson, H. L. Science 2003, 302, 1904-1905. (d) Mckee, M. G.;
Layman, J. M.; Cashion, M. P.; Long, T. E. Science 2006, 311, 353-355.
(e) Chen, W.-H.; Shao, X.-B.; Regen, S. L. J. Am. Chem. Soc. 2005, 127,
12727-12735. (f) Loewik, D. W. P. M.; van Hest, J. C. M. Chem. Soc.
ReV. 2004, 33, 234-245. (g) Menger, F. M.; Keiper, J. S. Angew. Chem.,
Int. Ed. 2000, 39, 1906-1920. (h) Ryu, E.-H.; Zhao, Y. Org. Lett. 2004,
6, 3187-3189. (i) Cornelissen, J. J. L. M.; Fischer, M.; Sommerdijk, N.
A. J. M.; Nolte, R. J. M. Science 1998, 280, 1427-1430.
(2) (a) Hartgerink, J. D.; Beniash, E.; Stupp, S. I. Science 2001, 294, 1684-
1688. (b) Gaucher, G.; Dufresne, M.-H.; Sant, V. P.; Kang, N.; Maysinger,
D.; Leroux, J.-C. J. Controlled Release 2005, 109, 169-188. (c) Riess, J.
G.; Krafft, M. P. Biomaterials 1998, 19, 1529-1539. (d) Torchilin, V. P.
J. Controlled Release 2001, 73, 137-172. (e) Shuai, X.; Ai, H.; Nasongkla,
N.; Kim, S.; Gao, J. J. Controlled Release 2004, 98, 415-426. (f)
Trubetskoy, V. S.; Torchilin, V. P. AdV. Drug DeliVery ReV. 1995, 16,
311-320. (g) Gao, Z.; Lukyanov, A. N.; Singhal, A.; Torchilin, V. P. Nano
Lett. 2002, 2, 979-982. (h) Wang, Y. L.; Banziger, J.; Dubin, P. L.;
Filippelli, G.; Nuraje, N. EnViron. Sci. Technol. 2001, 35, 2608-2611.
(3) For examples, see: (a) Moffitt, M.; Khougaz, K.; Eisenberg, A. Acc. Chem.
Res. 1996, 29, 95-102. (b) Pochan, D. J.; Chen, Z.; Cui, H.; Hales, K.;
Qi, K.; Wooley, K. L. Science 2004, 306, 94-97. (c) Jain, S.; Bates, F. S.
Science 2003, 300, 460-464. (d) Li, Y.; Lokitz, B. S.; McCormick, C. L.
Angew. Chem., Int. Ed. 2006, 45, 5792-5795.
(4) Basu, S.; Vutukuri, D. R.; Shyamroy, S.; Sandanaraj, B. S.; Thayumanavan,
S. J. Am. Chem. Soc. 2004, 126, 9890-9891.
(5) Arumugam, S.; Vutukuri, D. R.; Thayumanavan, S.; Ramamurthy, V. J.
Am. Chem. Soc. 2005, 127, 13200-13206.
(6) Basu, S.; Vutukuri, D. R.; Thayumanavan, S. J. Am. Chem. Soc. 2005,
127, 16794-16795.
(7) Sandanaraj, B. S.; Simard, J.; Vutukuri, D. R.; Hong, R.; Rotello, V. M.;
Thayumanavan, S. J. Am. Chem. Soc. 2005, 127, 10693-10698.
9
16224
J. AM. CHEM. SOC. 2006, 128, 16224-16230
10.1021/ja065213o CCC: $33.50 © 2006 American Chemical Society

Supramolecular Assemblies from Amphiphilic Homopolymers
A R T I C L E S
Scheme 1. General Synthetic Approach to Polymers PA6M,
Chart 1. Structure of the Amphiphilic Homopolymers Studied
PA8M, PA10M, PA5V, PA7V, PA10V, and PA15V
How different are these from the small molecule surfactants?
(iv) Could one use a similar design principle to achieve vesicle-
like assemblies? What are the critical structural requirements
for such assemblies? (v) Are the vesicle-like assemblies
responsive to the solvent environment? We describe here the
design and syntheses of the monomers and the polymers,
followed by the characterization of the assemblies formed by
these structures using transmission electron microscopy (TEM),
dynamic light scattering (DLS), static light scattering (SLS)
studies, dye-based studies to estimate the polarity of the micellar
interior and vesicular membranes, and estimation of critical
aggregation constants for micelles8a and vesicles.8b
Also, to test whether we could use similar design guidelines
to achieve vesicle-type assemblies, polymers PA5V, PA7V,
PA10V, and PA15V containing bolamphiphilic10 moieties were
synthesized. The key structural feature of these polymers is the
presence of two hydrophilic carboxylic functionalities connected
by alkyl chains of varying lengths. Our hypothesis is that such
a presentation of the functionalities in a homopolymer side chain
could provide a more direct pathway to achieve vesicular
assemblies. For example, the only difference between structures
PA10M and PA10V is that the polymer PA10V contains an
additional -CO2H functionality at the end of the long alkyl
chain. Would PA10M form a micelle-like assembly, while
PA10V forms a vesicle-type assembly?
Results and Discussion
Design and Synthesis of Monomers and Polymers. In our
original molecular design, we had stipulated that the hydrophilic
and the hydrophobic functionalities should be on either side of
a rigid scaffold such that the monomer is facially amphiphilic.
As a consequence of this presumed criterion, we had substituted
the hydrophilic and the hydrophobic moieties at the 3- and
5-positions of a benzene ring relative to the polymerizable
double bond, as illustrated by structures PS6M, PS8M, and
PS10M (Chart 1).9 Such a pre-condition in the molecular design
narrows down the scope of these amphiphilic homopolymers.
Therefore, we have been interested in identifying whether a
simple placement of a hydrophilic and hydrophobic moiety with
a flexible linker within a monomer unit is sufficient to achieve
the observed environmentally sensitive assemblies. Demonstra-
tion of such a feature will obviously expand the scope of the
homopolymers in amphiphilic assemblies.
To test this, we have designed and synthesized polymers
PA6M, PA8M, and PA10M, shown in Chart 1. In these
structures, glycine has been derivatized with alkyl chains of
different lengths. The secondary amine is then attached to an
acryl moiety. The resultant acrylamide polymer has simple
design features, that is, a flexible side chain functionality of
the polymer in which each repeat unit contains both a hydro-
philic and a hydrophobic moiety. If our molecular design can
indeed be applied to a wide variety of structures, then polymers
PA6M should form micelles or inverted micelles depending on
the solvent environment, just as with polymers PS6M, PS8M,
and PS10M.
Monomers for the synthesis of acrylamide polymers were
synthesized from the ethyl ester of glycine, haloalkanes (or
methyl ester of haloalkanoic acid), and acryloyl chloride, as
shown in Scheme 1. Synthesis of 11-17 was achieved using
the reversible addition-fragmentation-chain transfer (RAFT)
polymerization method.11 This controlled radical polymerization
technique afforded the polymers 11-17 with a reasonable
control in polydispersity, as illustrated in Scheme 1. Saponifica-
tion of the ester using potassium hydroxide followed by
acidification afforded the polymers. This hydrolysis step
proceeded with >95% yield, as discerned from the complete
1
disappearance of the alkanol part of the ester peaks in the H
NMR spectra of the polymers. Details of the synthesis and
characterization of these polymers are outlined in the Supporting
Information.
Characterization of the Micelle-type Assemblies. First, we
were interested in identifying whether the polymers PA6M,
PA8M, and PA10M would be capable of forming solvent-
dependent micelle and inverse micelle-type assemblies. This will
(8) (a) Micelles are defined here as containers that have a hydrophilic exterior
and hydrophobic interior. These assemblies are capable of encapsulating
hydrophobic guest molecules, but not hydrophilic ones. (b) Vesicles are
containers with a hydrophilic environment in both the exterior and the
interior. These assemblies are capable of encapsulating hydrophilic guests
in the interior and hydrophobic guests within their membranes.
(9) The polymers are here named as PS for polystyrene (PA for polyacryl-
amide), the number following this indicates the length of the alkyl chain,
and M (or V) indicates whether an assembly is expected to provide micelles
or vesicles, based on our molecular design.
(10) For a few small molecule examples of bolamphiphiles, see: (a) Fuhrhop,
J.-H.; Wang, T. Chem. ReV. 2004, 104, 2901-2937. (b) Escamilla, G. H.;
Newkome, G. R. Angew. Chem., Int. Ed. Engl. 1994, 33, 1937-1940. (c)
Fuhrhop, J.-H.; Fritsch, D. Acc. Chem. Res. 1986, 19, 130-137.
(11) (a) Barner-Kowollik, C.; Davis, T. P.; Heuts, J. P. A.; Stenzel, M. H.; Vana,
P.; Whittaker, M. J. Polym. Sci., Part A: Polym. Chem. 2003, 41, 365-
375. (b) Moad, G.; Rizzardo, E.; Thang, S. H. Aust. J. Chem. 2005, 58,
379-410. (c) Chiefari, J.; Chong, Y. K.; Ercole, F.; Kristina, J.; Jeffery,
J.; Le, T. P. T.; Mayadunne, R. T. A.; Meijs, G. F.; Moad, C. L.; Moad,
G.; Rizzardo, E.; Thang, S. H. Macromolecules 1998, 31, 5559-5562.
9
J. AM. CHEM. SOC. VOL. 128, NO. 50, 2006 16225

A R T I C L E S
Savariar et al.
radius of gyration versus the hydrodynamic radius (RG/RH),
known as F, is sensitive to the nature of the assembly in
solution.14 In the case of polymer PA6M, this ratio is about
0.77, which indicates the formation of a micelle-type assembly.
It is to be noted that the sizes we obtain from TEM and DLS
seem rather large to be called micelles. The micelles obtained
from these polymers are expected to be closer to those obtained
from small molecule surfactants, such as SDS. The rather large
sizes obtained here might lead one to suggest that these
assemblies might be akin to vesicles. Note that the size alone
does not qualify these assemblies as vesicles; they are indeed
micellar assemblies, because: (i) the TEM images obtained from
micelles are darker in the interior, while the water-filled interiors
of vesicles are lighter than the periphery. The former was indeed
the case with these polymers and therefore resembles micelles.13
(ii) Micelles are capable of sequestering hydrophobic guest
molecules, but not hydrophilic ones, in water. Vesicles are
capable of sequestering hydrophilic guest molecules in their
interior, and hydrophobic molecules within their membranes.
These polymers do not exhibit the ability to sequester hydro-
philic guest molecules and do sequester hydrophobic guests
(vide infra), consistent with micelle formation. (iii) Micellar
interiors are typically more hydrophobic as compared to the
membranes in typical vesicles. This was indeed the case with
the polymers under study (vide infra). (iv) There is sufficient
literature precedence for micellar assemblies affording sizes
bigger than expected, both with small molecule surfactants and
with amphiphilic block copolymers.15 Although various inves-
tigators call these with different names, the underlying theme
in all of these observations is that these are aggregation of
micelles. It is possible that our polymers also fall under this
category.
If the polymers do indeed form micelle-type and inverse
micelle-type assemblies in water and apolar solvents, respec-
tively, these polymers should be capable of acting as nanocon-
tainers for apolar guest molecules in water and vice versa. To
investigate this possibility, we first used pyrene as the guest
molecule in water. Pyrene also exhibits an emission spectrum,
the fine structure of which is sensitive to the microenvironment
in which it is present.16 Therefore, this guest molecule also
provides the opportunity to investigate the polarity of the
micellar interior formed by these polymers. Pyrene is not very
soluble in water by itself, as could be discerned from the
absorption spectrum of pyrene obtained in water (Figure 1c).
However, in a 10-4 M solution of the polymer PA6M in water,
an absorbance of 0.4 was obtained at 337 nm. This result
suggests that the micelle-type assembly is capable of sequester-
ing pyrene in water. If this apolar guest molecule is indeed
present in the hydrophobic interior of a micelle-type assembly,
Figure 1. TEM pictures from (a) solution of PA6M in water, and (b)
solution of PA6M in toluene. (c) Absorption spectra of pyrene in water
and in aqueous solution of PA6M. (d) Emission spectra of absorption-
matched rhodamine 6G in water and in toluene solution of PA6M.
test whether the design of homopolymers to form micelles or
inverse micelles is more broadly applicable. Transmission
electron microscopy (TEM) has been a popular and effective
tool for characterizing amphiphilic supramolecular assemblies.12
To prepare a micellar solution of the polymer, it had to be
dissolved in water. For this purpose, polymers were converted
to the corresponding metal carboxylates using lithium hydroxide
as the base. The TEM picture obtained from an aqueous solution
of the polymer PA6M is shown in Figure 1a, which suggests
that this polymer does afford a micelle-type assembly in water.
Analysis of these pictures suggests that the average particle size
of the micellar assembly in the aqueous phase is about 70 nm.
Similarly, the polymer PA6M was dissolved in toluene, and
the TEM picture obtained from this solution suggests that an
inverse micelle-type assembly is formed (Figure 1b). The
average size of these assemblies was determined to be about
120 nm. Similar environment-dependent nanoassemblies were
also obtained with polymers PA8M and PA10M.
One could be understandably skeptical of the particle size
measurements obtained from TEM, because the solution-based
sample is dried in the process. Laser light scattering is a
noninvasive method of analyzing the aggregates in solution. To
investigate the correlation between the assemblies in an aqueous
solution as compared to a dried sample in TEM, we measured
the size of the particles obtained from polymer PA6M using
dynamic light scattering (DLS). The hydrodynamic radius (RH),
obtained using DLS, for the aqueous solution of PA6M was
about 53 nm (comparable to RTEM of 35 nm). We also
determined the radius of gyration (RG) of these nanoassemblies
using static light scattering (SLS). The RG of PA6M in the
aqueous phase was found to be about 41 nm.13 The ratio of the
(14) (a) Burchard, W. AdV. Polym. Sci. 1983, 48, 1-167. (b) Rodr´ıguez-
Herna´ndez, J.; Lecommandoux, S. J. Am. Chem. Soc. 2005, 127, 2026
-2027. (c) Zhou, S.; Burger, C.; Chu, B.; Sawamura, M.; Nagahama, N.;
Toganoh, M.; Hackler, U. E.; Isobe, H.; Nakamura, E. Science 2001, 291,
1944-1947. (d) Giacomelli, C.; Men, L. L.; Borsali, R.; Lai-Kee-Him, J.;
Brisson, A.; Armes, S. P.; Lewis, A. L. Biomacromolecules 2006, 7, 817-
828.
(15) (a) Chakraborty, H.; Sarkar, M. Langmuir 2004, 20, 3551-3558. (b) Erhsrdt,
R.; Zhang, M.; Bo¨ker, A.; Zettl, H.; Abetz, C.; Frederic, P.; Krausch, G.;
Abetz, V.; Mu¨ller, A. H. E. J. Am. Chem. Soc. 2003, 125, 3260-3267. (c)
Wang, C.; Ravi, P.; Tam, K. C. Langmuir 2006, 22, 2927-2930. (d)
Kujawa, P.; Tanaka, F.; Winnik, F. M. Macromolecules 2006, 39, 3048-
3055. (e) Petrov, P.; Bozukov, M.; Burkhardt, M.; Muthukrishnan, S.;
Mu¨ller, A. H. E.; Tsvetanov, C. B. J. Mater. Chem. 2006, 16, 2192-2199.
(16) Kalyanasundaram, K.; Thomas, J. K. J. Am. Chem. Soc. 1977, 99, 2039-
2044.
(12) (a) Disher, D. E.; Eisenberg, A. Science 2002, 297, 967-973. (b) Discher,
B. M.; Won, Y.-Y.; Ege, D. S.; Lee, J. C.-M.; Bates, F. S.; Discher, D. E.;
Hammer, D. A. Science 1999, 284, 1143-1146. (c) Xu, J.; Zubarev, E. R.
Angew. Chem., Int. Ed. 2004, 43, 5491-5496. (d) Kim, B. S.; Hong, D.
J.; Bae, J.; Lee, M. J. Am. Chem. Soc. 2005, 127, 16333-16337.
(13) See the Supporting Information for details.
9
16226 J. AM. CHEM. SOC. VOL. 128, NO. 50, 2006

Supramolecular Assemblies from Amphiphilic Homopolymers
A R T I C L E S
Scheme 2. Synthesis of the Building Block Surfactant
Table 1. CAC and I1/I3 Values for Polymers and Building Blocks
CAC
I1/I3 above
compound
mg/L
M
the CAC
PS6M
PS8M
PS10M
PA6M
PA8M
PA10M
PA5V
PA7V
PA10V
PA15V
18
7
6
5
5
5
5.8 × 10-7
3.9 × 10-7
3.3 × 10-7
2.5 × 10-7
2.4 × 10-7
3.3 × 10-8
NA
1.10
1.10
0.98
1.20
1.16
1.14
1.85
1.75
1.52
1.22
1.00
0.85
1.10
0.90
0.78
the ratio of the intensity of the fluorescence peaks (I1/I3) should
be much less than is obtained for just water. Indeed, the value
obtained for the polymer PA6M is about 1.2. This value
compares well with the micellar interior formed from small
molecule surfactants such as SDS, which exhibits a value of
1.2.17
0.6
NA
NA
50
NA
5.8 × 10-6
7.4 × 10-8
2.1 × 10-3
7.9 × 10-4
5 × 10-4
5.7 × 10-4
3.6 × 10-4
2
450
193
136
180
140
19
20
21
22
Similarly, to investigate the formation of inverse micelles,
rhodamine 6G (R6G) that is not soluble in toluene is used as
the probe.18 In the presence of the polymer PA6M, it is clear
that R6G can be dissolved in the solution from its color and
the linear absorption spectra. Because this dye molecule exhibits
self-quenching properties, it is also possible to investigate
whether it is indeed sequestered inside a nanoscale inverse
micelle-type particle. If a certain amount of R6G is uniformly
distributed in a solvent that it is soluble, such as water, it could
fluoresce without any self-quenching effects (Figure 1d).
However, when the same amount of the dye molecule is
sequestered inside a 120 nm particle, the local concentration of
R6G within the inverse micelle increases dramatically, although
the overall solution concentration of R6G is the same as the
aqueous solution. This increase in local concentration would
result in self-quenching. This is indeed the observed result with
this dye molecule and the inverse micelle formed by polymer
PA6M, as shown in Figure 1d. Similar behavior was also
observed with polymers PA8M and PA10M.
The sensitivity of pyrene emission to its microenvironment
has been utilized to determine the CMC of the micelle.19
Because these polymers can be considered to be polymerized
surfactants, it is useful to compare the CMC of these polymers
with the corresponding building block units. For this purpose,
we have synthesized the corresponding building blocks 18-20
by reducing the double bond of the acrylamide monomers, as
shown in Scheme 2. When comparing the small molecules with
the corresponding polymers, comparison of the CMCs using
molar concentration can be misleading, because the molecular
weights of the polymers are high and these macromolecules
contain several copies of the surfactants. Therefore, comparison
of the CMCs based on weight is more appropriate. For example,
comparison of PA6M and 18 based on molar concentration
suggests an increase in the CMC for the polymer by 4 orders
of magnitude (Table 1). However, the comparison using weight
suggests that this increase corresponds to only about 2 orders
of magnitude. It is gratifying to note a significant enhancement
in the CMC for the polymer PA6M. The CMC understandably
decreases in the small molecule as well as in the polymer, when
the length of the hydrophobic chain is increased. Yet, the
magnitude of the gain in CMCs upon going from the monomer
to a polymer seems to vary with the nature of the alkyl group;
the CMC decreases about 39 times for the octyl and 226 times
for the decyl chain. This difference in stage at which a
significant difference in CMC is observed for the polymer and
the small molecule is attributed to the relative packing of the
hydrophobic chains in the small molecule versus the polymer.
In the polymer, the relative distance between an amphiphilic
functionality and its neighbor is dictated by the covalent bonds
in the polymer backbone. On the other hand, this packing is
more adjustable in the small molecule.
It is also interesting to note that the I1/I3 values for the small
molecules above CMC are lower than those for the correspond-
ing polymers; that is, small molecules afford a better hydro-
phobic environment as compared to the polymers. This can be
explained by the same argument above. That is, the packing
mode in the polymer is dictated by the repeat units, whereas
small molecules can better accommodate themselves relative
to their neighbors to optimize the packing. This could afford a
more hydrophobic environment in the latter case. While this is
a reasonable explanation, there is also an alternate possibility.
Because the CMCs of the small molecules are inherently higher,
the I1/I3 values for these are calculated at a higher concentration.
This also means that the number of micellar containers (or more
appropriately the relative volume of the micellar containers) is
larger in the small molecules. The I1/I3 values are really a
measure of the distribution coefficient of pyrene between the
bulk solvent and the micellar container. When there is a greater
volume of the hydrophobic containers available, the relative I1/
I3 is likely to be lower. To distinguish these two possibilities,
we measured the I1/I3 for pyrene in the presence of polymers
PA6M and PA10M at concentrations similar to those above
the CMC of 18 and 20. We found that the I1/I3 value did not
change with this huge increase in concentration of the polymer.
This suggests that the former mechanism (differential packing
for the polymer vs the small molecule) is operational.
Previously, we had reported the TEM behavior and the
environment-dependent container properties of the amphiphilic
polystyrenes.4 However, the CMCs of these polymers have not
been reported. It is interesting to compare the CMCs obtained
from these polymers as compared to the corresponding acryl-
amide polymers PA6M, PA8M, and PA10M. The CMCs of
the polystyrenes are similar to those of polyacrylamides, when
the hydrophobic functionalities are based on hexyl and the octyl
groups. However, this difference increases to about an order of
(17) (a) Goddard, E. D.; Turro, N. J.; Kuo, P. L. Langmuir 1985, 1, 352-355.
(b) Lee, J.; Moroi, Y. J. Colloid Interface Sci. 2004, 273, 645-650.
(18) Jung, H. M.; Price, K. E.; McQuade, D. T. J. Am. Chem. Soc. 2003, 125,
5351-5355.
(19) (a) Wilhelm, M.; Zhao, C.-L.; Wang, Y.; Xu, R.; Winnik, M. A.; Mura,
J.-L.; Riess, G.; Croucher, M. D. Macromolecules 1991, 24, 1033-1040.
(b) Astafieva, I.; Zhong, X. F.; Eisenberg, A. Macromolecules 1993, 26,
7339-7352. (c) Kwon, G.; Naito, M.; Yokoyama, M.; Okano, T.; Sakurai,
Y.; Kataoka, K. Langmuir 1993, 9, 945-949.
9
J. AM. CHEM. SOC. VOL. 128, NO. 50, 2006 16227

A R T I C L E S
Savariar et al.
magnitude in the case of decyl moiety. This is attributed to the
relatively flexible nature of the acrylamide backbone to accom-
modate the formation of a micellar assembly, relative to the
polystyrene backbone. It is also interesting to note that the
hydrophobicity of the interior in the case of the polystyrenes is
higher than the corresponding polyacrylamides. This is under-
standable, because the polymer backbone itself is more hydro-
phobic in the case of polystyrenes relative to the polyacryl-
amides.
The observed CMCs for the polymers PS6M, PS8M, and
PS10M could seem contradictory to our preliminary com-
munication, at first glance.4 In our previous report, we had
reported that we have observed the micellar assemblies using
TEM even at nanomolar concentrations. However, here we
report that the CMCs of these types of polymer are in the range
of 10-7 M. This apparent inconsistency can be easily understood,
when the underlying themes in both of these measurements are
understood. The TEM measurement represents the picture of a
dried sample of an amphiphilic polymer assembly at various
concentrations. On the other hand, the above-mentioned CMC
measurements are based on the distribution ratio of a guest
molecule between the micellar interior formed by the polymer
and the bulk solvent. This distribution ratio will obviously
change with the concentration of the polymer. Therefore, it is
very possible that at lower concentrations the polymer still forms
a spherical assembly. However, this assembly does not ef-
fectively compete for the guest molecule with the bulk solvent
at low concentrations. Note that the inherent solubility of pyrene
in water is about 10-7 M.19a
Characterization of the Vesicle-type Assemblies. It is
interesting to ask whether the structures of the building blocks
within the homopolymers could be tuned to achieve vesicular
assemblies. We hypothesized that incorporating two hydrophilic
head groups on either side of an alkyl chain should be able to
afford vesicle structures. Polymer vesicles based on amphiphilic
polymers have been previously achieved using block copolymers
or polymers containing two hydrophobic tails for every hydro-
philic head group.20 In our case, we ask whether polymerized
bolamphiphiles are capable of providing vesicle-type assemblies.
First, we studied the lithium carboxylate of polymer, PA10V,
in water. It was gratifying to note that the TEM picture of this
solution exhibited a vesicular morphology, as shown in Figure
2a. Our hypothesis is that a vesicle-like assembly is obtained
from the polymer PA10V, because one of the carboxylate
moieties is presented in the exterior while the other carboxylate
is presented in the water-filled interior of the vesicle. It is
reasonable to expect the length of the alkyl chain between the
two carboxylate moieties to impact the nature of the amphiphilic
assembly formed from the polymers. Accordingly, we studied
polymers PA5V, PA7V, and PA15V, where the nitrogen of the
acrylamide functionality and the non-glycine carboxylate func-
tionality are separated by 5, 7, and 15 methylene units,
respectively. TEM images obtained from an aqueous solution
of PA15V clearly exhibit a vesicle-type assembly, while those
Figure 2. TEM pictures from (a) PA10V in water, (b) PA5V in water,
and (c) PA10V in toluene.
Figure 3. Schematic representation of vesicle and reverse micelle formation.
from PA5V do not exhibit anything that resembles a vesicle,
as exemplified in Figure 2b.
We were intrigued by the fact that these polymers are also
soluble in apolar organic solvents. To investigate the nature of
the assembly in these solvents, TEM images were obtained from
the toluene solution of polymer PA10V (Figure 2c). These
images indicate the formation of large inverted micelle-like
assemblies. We hypothesize that the long alkyl chains could
bend to form a hydrophobic corona and a hydrophilic core.
However, we note that the size of the assembly is much larger
than even the vesicles from aqueous solutions. These assemblies
are thus reminiscent of the so-called compound micelles.21
Compound micelles are considered to be larger aggregates of
smaller inverted-micellar assemblies. The presence of the inverse
micelle-type assembly was further confirmed by a dye encap-
sulation study (vide infra). We propose a simple model in that,
in the case of polymers PA10V and PA15V, the chain could
fold, thereby presenting a hydrophobic surface while tucking
in the hydrophilic carboxylic acid moieties. The proposed model
for these polymers is schematically represented in Figure 3.
We measured the size of the particles obtained from polymer
PA10V using dynamic light scattering. The hydrodynamic
radius (RH) of the particle was found to be about 28 nm. Note
that the size of the particles obtained from TEM is the diameter
of the particle and therefore the particle sizes obtained from
the two measurements are very similar. To obtain additional
evidence for the vesicle formation in this solution, we also
carried out static light scattering measurements that afforded a
radius of gyration (RG) of about 29 nm. As mentioned earlier,
the RG/RH (F) is sensitive to a particular morphology. The fact
that this ratio is 1.04 further supports the formation of vesicle-
like assemblies in solution. The TEM pictures of polymers
(20) (a) Schille´n, K.; Bryskhe, K.; Melnikova, Y. S. Macromolecules 1999, 32,
6885-6888. (b) Liu, F.; Eisenberg, A. J. Am. Chem. Soc. 2003, 125,
15059-15064. (c) Nardin, C.; Thoeni, S.; Widmer, J.; Winterhalter, M.;
Meier, W. Chem. Commun. 2000, 1433-1434. (d) Du, J.; Armes, S. P. J.
Am. Chem. Soc. 2005, 127, 12800-12801. (e) Ringsdorf, H.; Schlarb, B.;
Venzmer, J. Angew. Chem., Int. Ed. 1988, 27, 113-158. (f) Tsuchida, E.;
Nishide, H.; Yuasa, M.; Babe, T.; Fukuzumi, M. Macromolecules 1989,
22, 66-72.
(21) (a) Zhang, L.; Eisenberg, A. J. Am. Chem. Soc. 1996, 118, 3168-3181.
(b) Yu, Y.; Zhang, L.; Eisenberg, A. Langmuir 1997, 9, 2578-2581.
9
16228 J. AM. CHEM. SOC. VOL. 128, NO. 50, 2006

Supramolecular Assemblies from Amphiphilic Homopolymers
A R T I C L E S
Figure 5. Plot of I338/I333 in the excitation spectra versus concentrations
of PA15V and 22.
with the self-quenching behavior suggests the formation of
inverse micelle-type assemblies in toluene.13
As mentioned above, hydrophobic guest molecules can also
be incorporated within the vesicle membranes in water. When
pyrene was incorporated within the aqueous polymer assemblies
of PA5V, PA7V, PA10V, and PA15V in water, I1/I3 ratios were
found to be 1.85, 1.75, 1.52, and 1.22, respectively.13 It is
immediately obvious that polymers PA5V and PA7V do not
provide environments that are sufficiently different from bulk
water for pyrene. This supports the absence of vesicle or micelle-
type structures for the polymers, as discerned from our TEM
studies above. Both polymers PA10V and PA15V provide
sufficiently different environments as compared to water. It is
more so in the case of polymer PA15V, which is understandable
due to the longer alkyl chain length and therefore an opportunity
for better isolation of pyrene from the bulk solvent (water). It
is remarkable that among the micelle-forming polymers includ-
ing PA6M, which contains only a hexyl hydrophobic chain,
PA6M exhibits a much less polar environment as compared to
polymer PA15V containing a pentadecyl alkyl chain linker. We
attribute this to the fact that micellar interiors are more capable
of excluding water from its interior as compared to the vesicular
membranes. In the latter case, the guest molecule is in an
environment that is surrounded by water on both sides, while
the interior does not contain water in the former scenario.8
Figure 4. Absorbance matched emission spectra of (a) rhodamine 6G and
(b) calcein in water and in PA10V.
PA10V and PA15V also suggest that these assemblies are
unilamellar in nature.
The formation of these assemblies can be further ascertained
by studying the container properties. Interestingly, the vesicle
assemblies should be able to sequester both hydrophilic dye
molecules as well as hydrophobic dye molecules. The hydro-
philic dye molecules can be sequestered within the polar, water-
containing interiors of the vesicles; the apolar guest molecules
can be incorporated within the apolar membranes of the vesicles.
We utilized the self-quenching nature of rhodamine 6G to probe
the nature of the vesicle assemblies. R6G was incorporated into
the polymer assembly from PA10V in water and dialyzed the
solution to remove any unbound dye molecules. The fluores-
cence from this solution was compared to an absorbance-
matched solution of R6G in water without polymer PA10V.
The enhanced quenching in the presence of the polymer supports
the presence of vesicle-type structures, as shown in Figure 4a.
Because the polymer is negatively charged and R6G is positively
charged, the observed quenching could be attributed to a simple
surface binding of R6G and thus increase in local concentration
and self-quenching. To address this issue, we carried out these
experiments with a negatively charged polar dye (calcein) that
also exhibits self-quenching at higher concentrations. Similar
results were observed in this system, as shown in Figure 4b,
further supporting the presence of vesicle structures. The amount
of R6G inside the vesicular interiors is greater than that of
calcein, as discerned from the absorbance of the solutions and
extent of quenching in these two dyes. This is understandable,
considering the electrostatic complementarity of R6G with the
carboxylates of the polymer.
We utilized changes in the excitation spectra of pyrene to
evaluate the critical aggregation concentration of polymers
PA5V, PA7V, PA10V, and PA15V. No change in the excitation
spectra was observed with increasing concentration of polymers
PA5V and PA7V. This is understandable, because we have
already noted from our TEM studies that these polymers do
not form well-defined amphiphilic assemblies in water. The
CAC of the polymers PA10V and PA15V is 50 and 2 mg/L,
respectively. We have also synthesized the corresponding small
molecule building block units 21 and 22 for comparison. The
CACs for these molecules were found to be 180 and 140 mg/
L, respectively (Table 1). Note that the difference between the
decyl and hexadecyl linkers in the case of small molecules is
much smaller than that in the case of polymer. We attribute
this to the possibility that preorganization of a large number of
amphiphiles in a polymer chain lends itself to making the critical
jump in the CAC upon increasing the alkyl chain lengths earlier
(even between decyl and hexadecyl), as compared to the small
Similarly, the self-quenching behavior of R6G was also used
to investigate the formation of inverse micelles in toluene with
polymers PA10V and PA15V. R6G is insoluble in toluene by
itself. The fact that a significant amount of the dye is solubilized
in toluene in the presence of the polymer PA10V combined
9
J. AM. CHEM. SOC. VOL. 128, NO. 50, 2006 16229

A R T I C L E S
Savariar et al.
molecules. Figure 5 shows the significant difference in CAC
between monomer and polymer having hexadecyl linker. Here
too, the hydrophobicity of the aggregates formed by the small
molecules is much higher than the corresponding polymers,
which can be explained by the differential packing of the
amphiphiles in these two classes of molecules, as explained
above.
however, less hydrophobic as compared to the building blocks,
which has been attributed to the relative packing abilities in
the assemblies. (v) The micellar, inverse micellar, and the
vesicular assemblies are all capable of acting as nanocontainers
for the corresponding small organic molecules. However, we
do not yet understand the exact reason for the rather large sizes
of the micellar assemblies formed from the amphiphilic ho-
mopolymers. (vi) Finally, it is important to note that all of these
assemblies are obtained by merely dissolving the polymer in
the solvents mentioned. The fact that we do not need extensive
processing to achieve these amphiphilic supramolecular as-
semblies is likely to render them broadly applicable in a variety
of applications.
Summary
We have synthesized a series of amphiphilic homopolymers
and the corresponding building block units as model compounds
to develop a systematic structure-property relationship. We
show here that: (i) the amphiphilic homopolymer design is
indeed applicable broadly. We have shown that polymers
containing hydrophilic and hydrophobic groups within a flexible
building block do indeed provide environment-dependent am-
phiphilic assemblies. (ii) When two hydrophilic head groups
linked by a long hydrophobic alkyl chain are incorporated onto
the repeat units of the homopolymer, vesicular assemblies are
obtained. A critical chain length is necessary to obtain such
assemblies. (iii) The vesicle-type assemblies are also environ-
ment-dependent; that is, these polymers also form inverted
micelle-type assemblies in apolar solvents such as toluene. (iv)
As compared to the small molecule building blocks, the critical
aggregation concentrations for these polymers are much lower.
The polarity of the interior of the polymer-based assemblies is,
Acknowledgment. We thank the NIGMS of the National
Institutes of Health for support. We also acknowledge the NSF
for a CAREER award and a MRSEC facility at the University
of Massachusetts for the infrastructure used in these character-
izations.
Supporting Information Available: Experimental details
including synthesis and characterization of monomers and
polymers. This material is available free of charge via the
Internet at http://pubs.acs.org.
JA065213O
9
16230 J. AM. CHEM. SOC. VOL. 128, NO. 50, 2006