ARTICLE IN PRESS

Biomaterials 25 (2004) 1929–1936

A one-pot synthesis of oleic acid end-capped temperature- and

pH-sensitive amphiphilic polymers
Xue-Ming Liu*, Li-Shan Wang
Institute of Materials Research and Engineering, National University of Singapore, 3 Research Link 117602, Singapore
Received 12 May 2003; accepted 11 August 2003

Abstract

A one-pot synthesis of an amphiphilic oleic acid-end capped random poly(N-isopropylacrylamide-co-N,N-dimethylacrylamide)

(2) is reported. In aqueous media, the solubility of 2 was temperature- and pH-sensitive. Both the lower critical solution temperature
and critical micelle concentration (CMC) of 2 were pH-dependent. The LCST of 2 was around 35.2 C in acidic buffer solutions
(pH=2.00–5.00), and it was increased significantly to around 38.4 C in neutral and alkaline buffer solutions (pH 7.00–11.00).
Polymer 2 exhibited a phase transition pH around 6.7, below which the polymer became significantly less water-soluble. The CMC
of 2 was 40 mg/l in pH 2.0 buffer solution, and it was increased markedly to 60–67 mg/l in pH 7.0 and pH 11.0 buffer solutions.
Micelle solutions of 2 in different pH conditions were prepared by a membrane-dialysis method. In aqueous solution, dynamic light
scattering studies revealed that the size of micelles was 50–90 nm with the particle size being larger in acidic solutions. In solid state,
transmission electron microscope studies showed that micelles were roughly spherical, their sizes were 25–90 nm and it decreased
with the increase of solution pH. The pH-sensitivity of 2 was triggered by the –COOH in the hydrophobic segment. The
temperature- and pH-sensitivity of the novel polymeric micelles would make an interesting drug delivery system.
r 2003 Elsevier Ltd. All rights reserved.

Keywords: Micelle; Thermally responsive material; pH responsive material; Drug delivery; Synthesis

1. Introduction be achieved by the cell affinity of the hydrophobic

Polymeric drug delivery systems can be not only used
to maximize therapeutic activity while minimizing
negative side effects, but also to achieve temporal and
distribution controls in drug therapy [1]. Of biomaterials
studied, stimuli-responsive micelle drug carriers made
from amphiphilic polymers have drawn tremendous
attentions over the past decade [2]. Amphiphilic block
copolymers containing stimuli-responsive hydrophilic
segment(s) and biocompatible hydrophobic segment(s)
have been well established as building blocks for the
preparation of stimuli-responsive micelle drug carriers
[3]. One area in which these properties have been
exploited is targetable delivery of anticancer drugs. The
temporal control for a drug therapy could be achieved
by the stimuli-responsive property of the hydrophilic
segment(s), whereas the distribution control, particu-
larly with respect to the treatment of solid tumor, could
*Corresponding author. Tel.: +65-68748380; fax: +65-68727528.
E-mail address: xm-liu@imre.a-star.edu.sg (X.-M. Liu).
segment(s) via the enhanced vascular permeability of
tumor tissue compared to healthy tissue [4].
Temperature sensitivity is one of the most interesting
properties in stimuli-responsive polymers, which have
been intensively investigated [5]. Thermo-sensitive poly-
mers exhibit a lower critical solution temperature
(LCST) in aqueous solution, below which the polymers
are water-soluble and above which they become water-
insoluble. These polymers are generally composed of
temperature-sensitive hydrophilic segment(s) and suita-
ble hydrophobic segment(s), and are mostly block
copolymers. Up to date, poly(N-isopropylacrylamide)
(PNIPAAm) and its copolymers have been some of the
most extensively studied temperature-sensitive hydro-
philic segments. We have previously employed this
segment and the hydrophobic cholesteryl group to
synthesize thermo-sensitive cholesteryl-end capped
random copolymers [6]. For the synthesis of amphiphilic
block or hydrophobically end-capped copolymers,
one terminal of hydrophilic polyacrylamides are
usually functionalized by the chain-transfer free radical

0142-9612/$ - see front matter r 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biomaterials.2003.08.023
 ARTICLE IN PRESS
1930 X.-M. Liu, L.-S. Wang / Biomaterials 25 (2004) 1929–1936
polymerization method. Several functionalities such
as –OH, –NH2 and –COOH groups could be introduced
into one end of a hydrophilic segment, by using
2-hydroxyethanethiol [5], 2-aminoethanethiol [5] and
3-mercaptopropionic acid [7] as a chain-transfer agent,
respectively. However, this synthetic methodology has
several disadvantages. First, the syntheses of such
copolymers generally involve multi-step reactions and
the use of highly toxic small thiol compounds as chain-
transfer agents. Secondly, the chain-transfer linkage unit
could be a possible toxic source after degradation of the
polymers. Finally, the chain-transfer free radical poly-
merization method can be only used to synthesize
oligomers with the molecular weights being usually less
than 20,000 [8].
We herein report a novel one-pot synthesis of oleic
acid-end capped random poly(N-isopropylacrylamide-
co-N,N-dimethylacrylamide) [P(NIPAAm-co-DMAAm)].
The two acrylamide monomers can readily undergo free
radical polymerization [9], whereas oleic acid has shown
much lower polymerization tendency as only low
molecular weight oligomers could be obtained under
harsh reaction conditions [10]. We thus anticipated that
the oleic acid could be a terminating agent to the
copolymerization of NIPAAm and DMAAm due to its
low polymerization reactivity. Furthermore, oleic acid
has been used as a building block for the synthesis of
drug delivery carriers [11], and a pH-sensitive carboxylic
group in the hydrophobic segment could exert much
stronger stimuli-response than that produced by the
stimuli-sensitive units in the hydrophilic segment [12].
There is also speculation that pH-sensitive polymeric
micelles could be interesting therapeutic carriers since
certain pathological conditions (i.e. tumors) and cellular
compartments (i.e. endosome) are associated with a
relatively acidic pH [12]. We found that the copolymer-
ization of NIPAAm and DMAAm using oleic acid as
the chain-terminating agent produced the oleic acid-end
capped copolymer of NIPAAm and DMAAm, which
exhibited temperature- and pH-sensitivities.
2. Experimental section
2.1. Materials and methods
Unless stated otherwise, all reagents and solvents were
used as received. NIPAAm and DMAAm were purified
by crystallization (hexanes) and distillation, respectively.
Oleic acid (95%) was purified by reduced-pressure
distillation. The pH Buffers (Hydriont, pH 2.00–11.00)
were purchased from Aldrich. The solvent p-xylene was
distilled over sodium before use. The reactions were
performed under a purified nitrogen atmosphere.
The 1H NMR spectra of polymers were recorded
on a Bruker AVANCE 400 spectrometer (400 MHz),
chemical shifts were expressed in parts per million (d) using
residual protons in the indicated solvents as the internal
standard. The IR spectra were recorded on a Fourier
transform infrared spectrometer (Perkin-Elmer Spectrum
2000, KBr). The molecular weights were determined by gel
permeation chromatography (GPC, Waters, polystyrene
standards) in THF (elution rate: 1 ml/min) at 25 C.
Differential scanning calorimetry (DSC) experiments were
performed using a TA 2920 Modulated DSC instrument
(CT, USA) with a ramp speed of 10 C/min.
2.2. Synthesis of the oleic acid-end capped random
P(NIPAAm-co-DMAAm)
NIPAAm (4.0 g, 35.3 mmol), DMAAm (0.51 g,
5.1 mmol), oleic acid (0.6–4.3 g, 2.0–15.1 mmol) and
benzoyl peroxide (BPO, 20 mg, 0.08 mmol) were dissolved
in p-xylene (20 ml) in a 100 ml round flask equipped with
a refluxing condenser. The reaction mixture was degassed
by bubbling with nitrogen for 15 min. The reaction
mixture was heated at 120 C for 16 h. Upon completion,
the products were precipitated out by the addition of
diethyl ether, and further purified by re-precipitation
from dichloromethane/diethyl ether (1:5, V/V) using a
slow liquid–liquid diffusion method. The products were
obtained as pale white solids in 37.1–79.0% yield.
2.3. Transmittance measurements, LCST and phase
transition pH determination
The LCST values of polymer 2 in different pH buffer
solutions were determined by monitoring the transmit-
tance change as a function of temperature. Sample
solutions (0.5 wt%) were prepared by the aqueous
Hydriont buffers at pH 2.00, 5.00, 7.00, 9.00 and
11.00, respectively. Optical transmittances of these
polymer buffer solutions were measured at 500 nm with
an UV-VIS spectrometer (Shimadzu, UV-2501PC).
Sample cells were thermostated with a temperature-
controller (Shimadzu, TCC-240A). Heating rate was
0.1 C/min. The LCST values of polymer solutions were
determined at the temperatures showing an optical
transmittance of 50%. The phase transition pHs of the
oleic acid end-capped polymer and a non-end-capped
hydroxyl-terminated P(NIPAAm-co-DMAAm) were
further investigated by monitoring the transmittance
change of 0.5 wt% polymer in pH 2.00–11.00 buffer
solutions as a function of pH at 500 nm and 36 C and
38 C, respectively. The phase transition pH was deter-
mined at the pH showing an optical transmittance of 50%.
2.4. Fluorescence measurements and CMC determination
Fluorescence spectra were recorded on a LS50B
luminescence spectrometer (Perkin-Elmer, USA) at
20 C. The cell-holder temperature was controlled with
 ARTICLE IN PRESS
X.-M. Liu, L.-S. Wang / Biomaterials 25 (2004) 1929–1936
a thermostated circulating bath. The CMC values of the
polymer 2 in pH 2.00, 7.00 and 11.00 buffer solutions
were determined, respectively. Pyrene was used as a
hydrophobic fluorescent probe for all determinations.
Aliquots of pyrene solutions (1.54  10 5 m in acetone,
400 ml) were added to 50 ml containers, and the acetone
was allowed to evaporate. Polymer solutions of a series
of concentrations (1.0  10 5–1.0 g/l) were prepared by
the aqueous Hydriont buffers. Ten milliliters sample
solutions of each concentration were then added to
above containers containing the pyrene residue, respec-
tively. It should be noted that all the aqueous buffer
sample solutions contained excess pyrene content at the
same concentration of 6.17  10 7 m. These sample
solutions were then left to stand at 20 C for 24 h to
dissolve and equilibrate pyrene. Excitation was carried out
at 340 nm, and emission spectra were recorded from 350
to 500 nm. Both excitation and emission bandwidths were
5 nm. A CMC value was taken from the intersection of the
tangent to the curve at the inflection with the horizontal
tangent through the points at low concentrations.
2.5. Preparation of polymeric micelles, dynamic
light-scattering measurements and transmission electron
microscope
Polymeric micelles of the oleic acid-end capped
polymer were prepared by a membrane-dialysis method.
The pH effect on the morphology and size of the
micelles was investigated. The 0.5 wt% polymer solu-
tions in dimethylformamide (DMF) were filtered with a
0.2 mm syringe and then dialyzed against 0.02% HCl, de-
ionized water and 0.02% NaOH for 24 h using a dialysis
tubing with a molecular weight cut-off of 2000 (Sigma
D-7884, benzoylated cellulose tubing) at 20 C, respec-
tively. The resultant micelle solutions were used for DLS
measurements using a N4 Plus instrument (Coulter)
equipped with a He–Ne laser (632 nm). The measure-
ments were repeated five times and were in good
agreements. An average value was finally obtained from
the five measurements. For the TEM observations, the
micelle solutions were stained with phosphotungstic
acid. The final concentrations of phosphotungstic acid
in the micelle solutions were 0.01 wt%. A drop of the
stained micelle solutions was then placed on a copper
grid coated with a polymer film, and self-dried in a dry
box at 20 C. The TEM observations were carried out on
a JEM-2010 with an electron kinetic energy of 200 keV.
3. Results and discussion
3.1. Polymer synthesis and characterization
PNIPAAm is a well-known thermo-sensitive polymer
showing reversible hydration–dehydration changes in
1931
response to small temperature changes [5,6]. The LCSTs
of amphiphilic polymers made from the hydrophilic
PNIPAAm and its copolymers depend on both the
polymer’s composites and their molecular weights.
When a certain range of polymer Mw is desired for
the purpose of controlled release, the LCSTs of the
amphiphilic polymers are usually adjusted by their
composites of the hydrophilic segments [9]. For exam-
ple, DMAAm has been used to increase the LCST of
hydrophilic PNIPAAm [9]. We have previously re-
ported the synthesis of cholesteryl-end capped random
P(NIPAAm-co-DMAAm)s. Both monomers exhibited
similar reactivity, and the LCSTs of the end-capped
polymers were increased to over 38 C by using
DMAAm as a comonomer [6]. In the present study,
DMAAm was employed as a co-monomer to adjust
the LCST of the oleic acid-end capped polymer. The
copolymerization of NIPAAm and DMAAm in the
presence of oleic acid is shown in Scheme 1. A summary
on the synthesis and characterization of the random
copolymers (1–4) is given in Table 1. In these reactions,
the NIPAAm/DMAAm molar ratios were 7:1, a ratio
that would produce copolymers with LCST being higher
than the human body temperature, whereas the oleic
acid feed molar ratios were increased from 0.4 to 3.0. As
the increase of the oleic acid feed ratio, the yield,
molecular weight and glass transition temperature (Tg )
of the polymers decreased, whereas the acidity values of
the polymers increased. Therefore an increase of the feed
ratio of the oleic acid increased the oleic acid content in
the copolymer. The 1H NMR spectra of all the four
polymeric products were similar, with the relative
intensity of the triplet signal of the terminal CH3
protons from the oleic acid moiety being slightly
different. For clarity, only the 1H NMR spectrum
of the polymer 2 in CDCl3 is shown in Fig. 1. In
the 1H NMR spectrum, the signals from the NIPAAm
moieties (a–f for CH2CHCONHCHMe2, CH2CHCO
NHCHMe2, CH2CHCONHCHMe2, CH2CHCONHC
HMe2, CH2CHCONHCHMe2, respectively), the signals
from DMAAm moieties (a, c, g for CH2CHCONMe2,
CH2CHCONMe2, CH2CHCONMe2, respectively), and
the triplet signal from the terminal methyl group of oleic
acid segment (signal h) are observed. From the integra-
tion ratio of the signals c and g to the signal e, the m=n
Scheme 1.
 ARTICLE IN PRESS
1932 X.-M. Liu, L.-S. Wang / Biomaterials 25 (2004) 1929–1936
Table 1
Synthesis and characterization of oleic acid-end capped random P(NIPAAm-co-DMAAm)
Polymer NIPAAm/DMAAm/OA (mole) Yield (%)
1 7.0:1.0:0.4 79.0
2 7.0:1.0:0.8 63.7
3 7.0:1.0:1.5 50.9
4 7.0:1.0:3 37.1
a
Determined by DSC.
b
Determined by GPC.
c
Determined by titration.
d
Estimated from NMR studies.
Fig. 1. The 1H NMR spectrum of the oleic acid-end capped
P(NIPAAm-co-DMAAm) (2) in CDCl3.
ratio is roughly 7:1, similar to the feed ratio of the two
monomers. This result is consistent with the result we
reported previously [6]. The m=l ratios are estimated to
be 7:0.15–0.28 from the integration ratio of the signal e
of the NIPAAm units to the signal h of the oleic acid
unit. In addition, the m=n=l ratios were also calculated
from the acidity and the Mn values of these polymers on
the fact that m=n ratios were 7:1. The m=l ratios
estimated from both methods are in good agreement.
For example, the m=l ratio of polymer 2 is 7:0.20 from
the NMR study, and is 7:0.17 from acidity value. From
the Mn value of the polymer 2 and the m=n=l ratio, the
average composite of the polymer 2 is estimated. Every
polymeric molecule roughly contains 28 NIPAAm
units, four DMAAm units and one oleic acid moiety.
In view of the relatively poorer reactivity of oleic acid,
oleic acid only served as a terminating agent in the
copolymerization reactions, as expected. The FT-IR
spectra of all the four polymers shared similar patterns.
In the IR spectra of the polymers, peaks at 3437.5–
3440.8 cm 1 (nN–H+O–H, s), 1651.4–1655.7 cm 1 (nC=O,
s), 1547.8–1549.4 cm 1 (nC N, m), 1172.2–1172.5 cm 1
(nC O, from COOH, w) were observed. These polymers
showed good solubility in both water and common
organic solvents (CHCl3, CH2Cl2, acetone, THF, etc.).
Taking into consideration of the yield and the content of
oleic acid unit in the amphiphilic polymer product, a
feed ratio of 7:1:0.8 was optimized for the synthesis of
the oleic acid-end capped polymer.
Tg ( C)a Mw, Mn (Mw/Mn)b Acidity (mg NaOH/g)c m/n/ld
132.6 9290, 6400 (1.5) 3.0 7:1:0.15
131.0 8980, 6430 (1.4) 4.7 7:1:0.20
130.4 8020, 5580 (1.4) 5.2 7:1:0.25
129.8 7720, 5490 (1.4) 5.5 7:1:0.28
3.2. LCST determination and effect of pH on LCST
The polymer 2 synthesized in this study contains the
P(NIPAAm-co-DMAAm) as its hydrophilic segment
and the oleic acid moiety as its hydrophobic segment.
The polymer thus should be temperature-sensitive. In
addition, the carboxylic group in the hydrophobic
segment could influence the hydrophobicity of the oleic
acid moiety to make the polymer also pH-sensitive.
Fig. 2 shows the LCST curves of polymer 2 determined
in different pH buffer solutions. In acidic buffer
solutions (pH 2.00 and 5.00), the LCST values are
35.1 C and 35.4 C, respectively. When the pH of
the aqueous solution is increased to 7.00–11.00, the
LCST value of the polymer increases significantly to
38.2–38.6 C. Thus the oleic acid-end capped polymer is
indeed pH-sensitive. The pH-sensitivity of the present
polymer might be triggered by the hydrophilic segment
due to the variation of the amide–water hydrogen
bonding, or by the hydrophobic segment due to
protonation/de-protonation of the carboxylic group,
or by both in different pH conditions. To investigate the
origin of the pH-sensitivity of the oleic acid-end capped
polymer, a hydroxy-terminated P(NIPAAm-co-
DMAAm) was synthesized (Mw 7970, Mn 5500) [6].
The NIPAAm/DMAAm ratio in the hydroxyl-termi-
nated polymer was similar to that of oleic acid-end
capped polymers. The LCST values of the hydroxyl-
terminated polymer in different buffer solutions were
determined. The LCST value of this hydrophilic
polymer was 40.5 C in pH 2.00 buffer solution, and
were 38.8–39.5 C in pH 7.00–11.00 buffer solutions.
Hence the pH effects on the LCSTs of both hydroxyl-
terminated and oleic acid-end capped polymers were
different. With the increase of pH of the polymer buffer
solutions, the LCST of 2 increased significantly, whereas
the LCST of the hydrophilic hydroxy-terminated poly-
mer decreased slightly. The pH effect on the LCST of
the aqueous hydroxyl-terminated polymer solution is
apparently due to the pH-induced change of the amide–
water hydrogen bonding. As pH decreased, this inter-
molecular hydrogen bonding increased, which in turn
would increase the water solubility of the polymer. That
is, the hydroxyl-terminated polymer is more hydrophilic
 ARTICLE IN PRESS
X.-M. Liu, L.-S. Wang / Biomaterials 25 (2004) 1929–1936
Fig. 2. The LCST curves of polymer 2 in different pH buffer solutions.
in acidic conditions, a feature that would increase the
LCST of the polymer [9]. Unlike the hydroxyl-termi-
nated polymer, two opposite forces influenced the water
solubility of the amphiphilic oleic acid-end capped
polymer 2. In acidic solutions, the amide–water hydro-
gen bonding would increase to make the hydrophilic
segment more hydrophilic, whereas the carboxylic group
in the hydrophobic end would be more protonated
to make the hydrophobic segment more hydrophobic.
The protonation of the carboxylic group would also
invariably decrease its hydrogen bonding with water and
increase the hydrogen bonding among different car-
boxylic groups in the hydrophobic segment, in the
consideration that the hydrophobic core has a relatively
anhydrous environment [13]. Of the two opposite
forces, the first one would increase the LCST, whereas
the second one would decrease the LCST. The LCST
values of polymer 2 in different pH solutions showed
that its pH-sensitivity was predominately triggered
by the carboxylic group in the hydrophobic segment.
It is noteworthy that a pH-sensitive unit in the
hydrophobic core of polymeric micelles would make
the drug carrier more sensitive to the environmental
change.
The phase transition pH of polymer 2 was further
determined (Fig. 3a). The phase transition pH was
estimated to be around 6.7. The acidity of the oleic acid-
end capped polymer is due to the carboxylic group in the
hydrophobic segment. Based on the acidity of oleic acid
(pKa=9.85), the molar ratio of protonated and non-
protonated polymeric molecules was approximated over
1000:1 at the phase transition pH. Hence the carboxylic
groups in polymer 2 were predominately protonated in
the phase transition pH. To further establish the role of
oleic acid in the phase transition, we tested the pH effect
on the phase transition of the above-mentioned hydro-
xyl-terminated P(NIPAAm-co-DMAAm) (Fig. 3b). In
contrast to the oleic acid-end capped polymer, the non-
oleylate capped one does not show a phase transition
pH. This in turn proved indirectly that the pH-
1933
Fig. 3. Plots of transmittance (%) of aqueous solutions of polymers as
a function of pH at 500 nm: (a) the oleic acid-end capped polymer 2 at
36 C; and (b) hydroxyl terminated P(NIPAAm-co-DMAAm) at 38 C.
sensitivity of polymer 2 was triggered by the carboxylic
group of oleic acid moiety.
3.3. CMC determination and pH effect on CMC
The CMC determination and the pH effect on the
CMC of 2 were carried out by fluorescence spectro-
scopic studies using pyrene as a probe [14]. The patterns
of the emission spectra of pyrene/polymer in pH 2.00,
7.00 and 11.00 buffer solutions were similar. For clarity,
only typical emission spectra of polymer 2 in pH 2.00
buffer at various concentrations in the presence of
6.16  10 7 m pyrene are shown in Fig. 4a. An analysis
of peak intensity at l391 as a function of polymer
concentration in all pH conditions is given in Fig. 4b.
With the increase of polymer concentration in an
aqueous solution of pyrene, several significant changes
in the emission spectra are observed (Fig. 4a). Firstly,
there is an increase in the intensity of the fluorescence
spectra. When the polymer concentration is increased
from 10 to 100 mg/l, an abrupt increase in the intensity is
observed. Secondly, the change in the vibrational fine
structure is observed, with the third band (I3 ) at 391 nm
being more sensitive to the polarity of the pyrene
environment. Finally, with the increase of polymer
concentration, the first band (I1 ) and the third band (I3 )
both red shift slightly. From Fig. 4b, it can be clearly
seen that abrupt intensity enhancements are observed in
all conditions at polymer concentration of 10–100 mg/l.
For more accurate CMC determination, the ratio (I3 =I1 )
is shown as a function of log C. Fig. 5 shows the CMC
curves of the polymer in different pH solutions. The
ratio of I3 =I1 increases with the increase of polymer
concentration. The CMC values in pH 2.00, 7.00 and
11.00 solutions are approximately 40, 60 and 67 mg/l,
respectively. Thus the CMC value of 2 in pH 2.00
solution is significantly lower than that in neutral and
alkaline solutions. In acidic solutions, the protonation
of carboxylic group in the hydrophobic segment would
 ARTICLE IN PRESS
1934 X.-M. Liu, L.-S. Wang / Biomaterials 25 (2004) 1929–1936
Fig. 4. (a) The fluorescence emission spectra of pyrene (6.16  10 7 m)
against concentration of oleic acid-end capped P(NIPAAm-co-
DMAAm) (2) at pH 2.00 (excitation wavelength: 339 nm). (b) Plots
of peak intensity at l391 as a function of polymer concentration (log C)
at different pH conditions.
Fig. 5. Plots of the ratio of intensities (I391 =I371 ) from pyrene emission
spectra at different pH solutions as a function of log C.
increase its hydrophobicity thus enhance the ability of
the polymer to form micelles.
3.4. Micelle formation and pH-induced size and
morphology changes: DLS and TEM studies
To investigate the pH effect on the size and
morphology of micelles of 2 in aqueous media and
solid state, DLS and TEM studies were carried out.
Micelle solutions in different pH conditions were
Table 2
Particle size and size distribution of micelle solutions of 2 formed
under: (a) 0.02% HCl solution; (b) deionized water solution; and (c)
0.02% NaOH solution
Micellar Average Standard Polydispersity
solutions diameter (nm) deviation (nm)
a 92 42 1.0
b 60 27.5 1.3
c 51 22.7 0.9
prepared by a membrane dialysis method as described
in the Section 2. The size and size distribution of above
micelles formed in acidic, neutral and alkaline solutions
were determined by DLS studies first (Table 2 and
Fig. 6). In the acidic solution, the average size of micelles
is 92 nm, which is significantly higher than those
in neutral and alkaline solutions (60 and 52 nm,
respectively).
TEM studies showed that the micelles formed in
acidic conditions had significantly larger sizes in solid
state. For the TEM studies, micelle solutions for the
above DLS experiments were stained and used. Fig. 7
shows TEM micrographs of micelles formed under
different conditions. Micelles formed under all condi-
tions are roughly spherical, however, the size of micelles
shows pH-dependence. In 0.02% HCl solution, the
diameters of micelles are 50–90 nm (Fig. 7a, d and e). In
neutral (Fig. 6b) and alkaline (Fig. 7c and f) solutions,
the sizes of the micelles decrease to 30–50 and 25–50 nm,
respectively. A significantly larger size of micelles
formed in the acidic solution indicates that micelles in
the acidic solution contain a higher degree of aggrega-
tion. A higher degree of aggregation of micelles in acidic
conditions is probably resulted from stronger hydro-
phobic–hydrophobic interactions of oleic acid moieties.
It can be seen that the sizes of micelles in solution and in
solid state are not significantly different, indicating that
further aggregation might not occur during the process
of self-drying prior to TEM observation. Hence the size
and pH effect on size determined by both the TEM and
DLS methods are consistent. It can be concluded that
the oleic acid-end capped polymer is indeed pH-
responsive. In acidic solutions, the polymer exhibited
lower CMC and higher degree of aggregation. Conse-
quently larger micelles were formed in acidic solutions
(Fig. 7).
4. Conclusion
We present in this paper a novel one-pot method that
could be generally used for the synthesis of hydro-
phobically end capped amphiphilic polymers by using
naturally occurring fatty acids as chain-terminating
 ARTICLE IN PRESS
X.-M. Liu, L.-S. Wang / Biomaterials 25 (2004) 1929–1936 1935

Fig. 6. Particle size distribution of micelle solutions prepared by dialyzing a 0.5% initial solution of 2 in DMF against: (a) 0.02% HCl; (b) de-ionized
water; and (c) 0.02% NaOH.

Fig. 7. TEM pictures showing micelles formed by dialyzing a 0.5% solution of 2 in DMF against: (a), (d) and (e) 0.02% HCl; (b) de-ionized water,
and (c) and (f) 0.02% NaOH.

agents. The new method avoided the use of highly Acknowledgements

toxic and extremely smelly thiol compounds, which have
been generally used as chain-transfer agents for the
synthesis of thermally responsive biomaterials. There-
fore the new method is more environmentally and
biologically friendly. The oleic acid end-capped random
P(NIPAAm-co-DMAAm) was temperature- as well
as pH-sensitive. The LCST was significantly higher
in neutral and alkaline conditions and its CMC
decreased significantly in acidic solutions. The polymer
exhibited a phase transition around pH 6.7. The pH-
sensitivity of the polymer was further established by
TEM and DLS studies. The facile one-pot synthesis,
temperature- and pH-sensitive of the oleic acid-end
capped polymer would make an interesting drug
delivery system. We are currently investigating the
controlled release of cyclosporin by this novel polymeric
micellar drug carrier.
We are grateful to the Agency for Science, Technol-
ogy and Research, Singapore for their financial support.
References
[1] Uhrich KE, Cannizzaro SM, Langer RS, Shakesheff KM.
Polymeric systems for controlled drug release. Chem Rev
1999;99:3181–98.
[2] Kost J, Langer R. Responsive polymeric delivery systems. Adv
Drug Deliv Rev 1991;6:19–50;
Yokoyama M, Okano T. Targetable drug carriers: present status
and future perspective-introduction. Adv Drug Deliv Rev
1996;21:77–80;
Langer R. Drug delivery and targeting. Nature 1998;392:5–10;
Langer R. Perspectives: drug delivery-drugs on target. Science
2001;293:58–9.
 ARTICLE IN PRESS
1936 X.-M. Liu, L.-S. Wang / Biomaterials 25 (2004) 1929–1936
.
[3] Rosler A, Vandermeulen GWM, Klok H-A. Advanced drug
delivery devices via self-assembly of amphiphilic block copoly-
mers. Adv Drug Deliv Rev 2001;53:95–108;
Kumar N, Ravikumar MNV, Domb AJ. Polyanhydrides: an
overview. Adv Drug Deliv Rev 2001;53:23–44.
[4] Yokoyama M, Okano T, Sakurai Y, Fukushima S, Okamoto K,
Kataoka K. Selective delivery of adiramycin to a solid tumor using
a polymeric micelle carrier system. J Drug Target 1999;7:171–86;
Maeda H, Wu J, Sawa T, Matsumura Y, Hori K. Tumor vascular
permeability and the EPR effect in macromolecular therapeutics: a
review. J Controlled Release 2000;65:271–84.
[5] Kohori F, Sakai K, Aoyagi T, Yokoyama M, Sakurai Y, Okano
T. Preparation and characterization of thermally responsive block
copolymer micelles comprising poly(N-isopropylacrylamide-b-dl-
lactide). J Controlled Release 1998;55:87–98;
Rapoport N. Stabilization and activation of Pluronic micelles for
targeted drug delivery. Colloids Surf B 1999;16:93–111;
Chung JE, Yokoyama M, Yamato M, Aoyagi T, Sakurai Y,
Okano T. Thermo-responsive drug delivery from polymeric
micelles constructed using block copolymer of poly(N-isopropy-
lacrylamide) and poly(butylmethacrylate). J Controlled Release
1999;62:115–27;
Meyer DE, Shin BC, Kong GA, Dewhirst MW, Chilkoti A. Drug
targeting using thermally responsive polymers and local hy-
perthermia. J Controlled Release 2001;74:213–24.
[6] Liu X-M, Yang Y-Y, Leong KW. Thermally responsive
polymeric micellar Nanoparticles self-assembled from cholesteryl
end-capped random poly(N-isopropylacrylamide-co-N,N-di-
methylacrylamide): synthesis, temperature-sensitivity and mor-
phologies. J Colloids Interface Sci, in press.
[7] Lu ZR, Kopeckova P, Wu ZC, Kopecek J. Functionalized
semitelechelic poly(N-2-hydroxypropyl)methacrylamide] for pro-
tein modification. Bioconjugate Chem 1998;9:793–804.
[8] Odian G. Principles of polymerization, 2nd ed. New York: Wiley;
1981.
[9] Kim I-S, Jeong Y-I, Cho C-S, Kim S-H. Core-shell type polymeric
nanoparticles composed of poly(l-lactic acid) and poly(N-
isopropylacrylamide). Int J Pharm 2000;211:1–8;
Kim I-S, Jeong Y-I, Cho C-S, Kim S-H. Thermo-responsive
self-assembled polymeric micelles for drug delivery in vivo.
Int J Pharm 2000;205:165–72;
Chung JE, Yokoyama M, Suzuki K, Aoyagi T, Sakurai Y, Okano
T. Reversibly thermo-responsive alkyl-terminated poly(N-isopro-
pylacrylamide) core-shell micellar structures. Colloids Surf B
1997;9:37–48;
Kohori F, Sakai K, Aoyagi T, Yokoyama M, Yamato M, Sakurai
Y, Okano T. Control of adriamycin cytotoxic activity using
thermally responsive micelles composed of poly(N-isopropylacry-
lamide-co-N,N-dimethylacrylamide)-b-poly(d,l-lactide). Colloids
Surf B 1999;16:195–205.
[10] Erhan SM, Kleiman R, Isbell TA. Estolides from meadowfoam
oil fatty-acids and other monounsaturated fatty acids. J Am Oil
Chem Soc 1993;70:461–5;
Li F, Zhang ZQ, Friberg SE, Aikens PA. Polymeric sur-
factants based on oleic acid. II. Copolymerization of sodium
acrylamidostearate and 10-undecen-1-ol in a lamellar liquid
crystal.
J Polymer Sci: Part A Polym Chem 1999;37:2863–972.
[11] Leroux J-C, Roux E, Garrec DL, Hong K, Drummond DC. N-
isopropylacrylamide copolymers for the preparation of pH-
responsive liposomes and polymeric micelles. J Controlled
Release 2001;72:71–84.
[12] Taillefer J, Jones M-C, Brasseur N, Van Lier JE, Leroux J-C.
Preparation and characterization of pH-responsive polymeric
micelles for the delivery of photosensitizing anticancer drugs.
J Pharm Sci 2000;89:52–62.
[13] Torchilin VP. Structure and design of polymeric surfactant-based
drug delivery systems. J Controlled Release 2001;73:137–72;
Allen C, Maysinger D, Eisenberg A. Nano-engineering block
copolymer aggregates for drug delivery. Colloids Surf B 1999;
16:3–27;
Rosler A, Vandermeulen GWM, Klok HA. Advanced drug
delivery devices via self-assembly of amphiphilic block copoly-
mers. Adv Drug Deliv Rev 2001;53:95–108.
[14] Wilhelm M, Zhao C-L, Wang Y, Xu R, Winnik MA, Mura J–L,
Riess G, Croucher MD. Polymer micelle formation. 3.
Poly(styrene-ethylene oxide) block copolymer micelle formation
in water—a fluorescence probe study. Macromolecules 1991;
24:1033–40;
Astafieva I, Fu X, Eisenberg A. Critical micellization phenomena
in block polyelectrolyte solutions. Macromolecules 1993;26:
7339–52.