Microcalorimetric Study of the Temperature-Induced Phase Separation in Aqueous Solutions of Poly(2-isopropyl-2-oxazolines)

Article abstract of DOI:10.1021/ma0358733

The effect of temperature on aqueous solutions of poly(2-isopropyl-2- oxazoline) (PIPOZ) samples of molecular weights ranging from 1900 to 5700 g mol^-1 was monitored by turbidimetry, high-sensitivity microcalorimetry (HS DSC), and pressure perturbation calorimetry (PPC) from 10 to 80°C. The polymers were soluble in cold water and underwent phase separation at T_CP ～ 45-63°C, depending on their molecular weight. The phase transition was endothermic, with an enthalpy change ranging from 0.36 ± 0.01 to 1.16 ± 0.01 kcal mol^-1. The coefficient of thermal expansion of PIPOZ in water (α_pol), determined by PPC, underwent a sharp increase at the temperature corresponding to the onset of phase transition, reaching a maximum value at T ～ T _CP. Microcalorimetric measurements were carried out with solutions of PIPOZ samples in D₂O and in aqueous NaCl solutions. The thermodynamic and volumetric changes associated with the phase transition of aqueous PIPOZ solutions are compared to those of aqueous solutions of two related polymers, poly(vinylcaprolactam) and poly(N-isopropylacrylamide) (PNIPAM), a polymer structurally related to PIPOZ that undergoes a phase transition in water at ～31°C.

Full text of DOI:10.1021/ma0358733

2
556
Macromolecules 2004, 37, 2556-2562
Microcalorimetric Study of the Temperature-Induced Phase Separation
in Aqueous Solutions of Poly(2-isopropyl-2-oxazolines)
†
‡
‡
Ch a r bel Dia b, Yosh itsu gu Ak iya m a , Ka zu n or i Ka ta ok a , a n d
,
†
F r a n c¸ oise M. Win n ik *
Faculty of Pharmacy and Department of Chemistry, Universit e´ de Montr e´ al,
CP 6128 Succursale Centre Ville, Montr e´ al, QC Canada H3C 3J 7, and Department of
Materials Science, Graduate School of Engineering, The University of Tokyo, 7-3-1 Hongo,
Tokyo 113 8656, J apan
Received December 10, 2003; Revised Manuscript Received February 6, 2004
ABSTRACT: The effect of temperature on aqueous solutions of poly(2-isopropyl-2-oxazoline) (PIPOZ)
-
1
samples of molecular weights ranging from 1900 to 5700 g mol was monitored by turbidimetry, high-
sensitivity microcalorimetry (HS DSC), and pressure perturbation calorimetry (PPC) from 10 to 80 °C.
The polymers were soluble in cold water and underwent phase separation at TCP ∼ 45-63 °C, depending
on their molecular weight. The phase transition was endothermic, with an enthalpy change ranging from
-
1
0
.36 ( 0.01 to 1.16 ( 0.01 kcal mol . The coefficient of thermal expansion of PIPOZ in water (Rpol),
determined by PPC, underwent a sharp increase at the temperature corresponding to the onset of phase
transition, reaching a maximum value at T ∼ TCP. Microcalorimetric measurements were carried out
2
with solutions of PIPOZ samples in D O and in aqueous NaCl solutions. The thermodynamic and
volumetric changes associated with the phase transition of aqueous PIPOZ solutions are compared to
those of aqueous solutions of two related polymers, poly(vinylcaprolactam) and poly(N-isopropylacrylamide)
(
∼
PNIPAM), a polymer structurally related to PIPOZ that undergoes a phase transition in water at
31 °C.
In tr od u ction
patibility and interfacial characteristics of poly(2-meth-
yl-2-oxazolines) (PMOZ) and PEOZ have been assessed
A large body of work has been devoted over the past
decades toward the development of drug delivery sys-
1
8-20
in various biotechnological applications.
Lipid vesicles
doped with poly(2-ethyl-2-oxazoline) lipopolymers ex-
hibited enhanced blood circulation times compared to
tems based on polymeric micelles featuring longevity
in blood circulation.1-6 In most studies, these drug
2
1
ordinary phospholipid vesicles. The ability of the
PMOZ and PEOZ chains to elicit beneficial properties
when grafted onto the surface of liposomes has been
explained by the conformational mobility of the poly-
mers and the tendency of their repeat units to form
vehicles consist of a hydrophilic poly(ethylene glycol)
(
PEG) outer corona, since this polymer prevents non-
specific adsorption of the polymeric micelles to proteins
and cells, thus allowing the micelles to evade recognition
by the reticuloendothelial system.3
,7,8
There has been
2
1
hydrogen bonds with water. However, quantitative
information on the interactions of poly(2-alkyl-2-oxazo-
lines) with water is scarce compared to the wealth of
data dealing with the behavior of PEG in water.22
Among the various poly(2-alkyl-2-oxazolines), PEOZ
has attracted the most attention by far. Aqueous solu-
tions of PEOZ exhibit a cloud point around 62-65 °C
depending on molecular weight and concentration.23 The
cloud point temperature is affected by the presence of
cosolvents, such as dioxane, and of salts, decreasing in
the presence of sodium chloride but increasing upon
little investigation into the use of hydrophilic corona-
forming polymers other than PEG to create micellar
drug delivery systems. Charged hydrophilic polymers
have been used for delivery to mucosal surfaces, such
as the respiratory, gastrointestinal, and urogenitary
9
tracts. Polymeric micelles with a corona made up of
natural or semisynthetic polysaccharides are investi-
gated as well as vehicles in the oral administration of
poorly water soluble drugs.1
0,11
Recently, polymeric
micelles based on block copolymers of 2-ethyl-2-oxazo-
line (PEOZ) and ꢀ-caprolactone12 were evaluated as drug
2
4
addition of tetrabutylammonium bromide. Studies of
PMOZ and PEOZ in water by light scattering revealed
that their second virial coefficients decrease with in-
creasing solution temperature, an indication of the
involvement of water/polymer hydrogen bonds in the
mechanism of the temperature-induced phase separa-
carriers and shown to have low cytotoxicity and hemolyt-
ic activity.13
In contrast to PEG or polysaccharide systems, poly-
2-alkyl-2-oxazolines), synthesized by living cationic
(
polymerization, can be tailored through the choice of not
1
4-16
only the end groups but also the side chains.
The
2
5,26
tion.
The phase transition of aqueous PEOZ solu-
length of the alkyl substituents controls to a large extent
the relative hydrophilicity of the poly(2-alkyl-2-oxazo-
lines), and only few oxazolines, such as the 2-methyl-,
tions was investigated over the entire water/polymer
composition domain. Concentrated PEOZ solutions obey
the “classical” Flory-Huggins miscibility behavior: the
cloud point shifts to lower polymer concentration as the
molecular weight of the polymer increases (Mvis 22 000-
2
-ethyl-, and 2-isopropyl-2-oxazolines, lead to polymers
1
7
soluble in water at room temperature. The biocom-
-
1
25,27
24
1
70 000 g mol ).
As pointed out by Lin et al., the
†
Universit e´ de Montr e´ al.
The University of Tokyo.
‡
repeat unit of PEOZ is isomeric to that of poly(N,N-
dimethylacrylamide) (Figure 1), and as such, there are
similarities in the solubility and miscibility character-
*
To whom correspondence should be addressed: Ph (514) 343
6
123; Fax (514) 343 2362; e-mail francoise.winnik@umontreal.ca.
1
0.1021/ma0358733 CCC: $27.50 © 2004 American Chemical Society
Published on Web 03/13/2004

Macromolecules, Vol. 37, No. 7, 2004
Poly(2-isopropyl-2-oxazolines) 2557
niques in order to assess the relative importance of
water/polymer hydrogen bond formation and polymer/
polymer hydrophobic interactions. Light and heavy
water are chemically identical, yet their physical prop-
3
6
erties differ significantly. The dissimilarities between
H2O and D2O are believed to stem from differences in
the intermolecular hydrogen bonds energies. The lengths
of hydrogen bonds in the two liquids are about the same,
but a “hydrogen” bond in D2O is about 5% stronger than
a hydrogen bond in H2O. Overall, D2O is a more
3
7
“structured” solvent than light water.
In a first section, we present the preparation of the
oligomers and describe their chemical characteristics
obtained from H NMR spectroscopy, MALDI-TOF
mass spectrometry, and gel permeation chromatography
F igu r e 1. Chemical structure of the repeat units of poly(2-
ethyl-2-oxazoline) (PEOZ), poly(N,N-dimethylacrylamide), poly-
1
(
2-isopropyl-2-oxazoline) (PIPOZ), and poly(N-isopropylacryl-
amide) (PNIPAM).
(GPC). Then, we report the results of the microcalorim-
etry study of PIPOZ samples, placing particular em-
phasis on the effect of the polymer chain length on its
behavior during the solution phase transition. Taken
together, the results suggest a description of the PIPOZ/
water interactions, which will be compared to the
currently accepted view of the phenomena taking place
in aqueous solutions of PNIPAM, the chemical isomer
of PIPOZ.
-
1 27
Exp er im en ta l Section
Ma ter ia ls. Water was deionized with a Millipore Milli-Q
water purification system. Isobutyric acid, 2-aminoethanol,
and methyl p-tosylate were purchased from Wako Pure
Chemical Industries. Deuterium oxide (99.8%) was purchased
from Aldrich-Sigma Chemicals. PNIPAM was prepared by
3
8
RAFT polymerization, using a trithiocarbonate RAFT agent.
In str u m en ta tion . NMR spectra were recorded on a Bruker
ARX-400 400 MHz spectrometer. UV/vis spectra were mea-
sured with a Hewlett-Packard 8452A photodiode array spec-
trometer equipped with a Hewlett-Packard 89090 temperature
controller. Gel permeation chromatography (GPC) measure-
ments were performed with a TOSOH HLC-8220 system
equipped with a TOSOH differential refractometer, two TOSOH
Super HHZ columns (4000 and 3000) eluted with DMF
containing triethylamine (30 mM), and LiCl (10 mM) (flow rate
-1
0
.8 mL min , 40 °C) and calibrated with poly(ethylene glycol)
standards used to determine the molecular weights. MALDI-
TOF mass spectra were recorded with a Reflex III (Bruker).
All spectra were obtained in positive ion mode, and ionization
6
7
was performed with a 337 pulsed (3 ns pulse width, 10 -10
-2
W cm ) nitrogen laser. An external calibration was performed
using poly(ethylene glycol) (MeO-PEG-OH) of M
w
5000 (NOF
Corp.). To prepare the matrix, a solution of the polymer in THF
-
1
(20 µL, 6 mg mL ) was mixed with a solution of 8,9-
-
1
trihydroxyanthracene in THF (20 µL, 10 mg mL ) by soni-
cation for a few minutes. A solution of lithium trifluoroacetate
-
1
in THF (2 µL, 2 mg mL ) was added to the matrix. The
resulting mixture was shaken briefly and applied to the probe
(2 µL).
P r ep a r a tion of 2-Isop r op yl-2-oxa zolin e.39 Isobutyric
acid (132 g, 1.5 mol) was added to 2-aminoethanol (92 g, 1.5
mol) under stirring. The resulting salt was heated slowly to
2
30 °C, a temperature higher than its melting point. The liquid
was refluxed for 48 h, followed by azeotropic distillation at
110 °C. The distillate was diluted in dichloromethane (350 mL)
to separate water from 2-isopropyl-2-oxazoline. The aqueous
layer was extracted repeatedly with dichloromethane. The
combined organic layers were concentrated in vacuo. The
residue was purified by repeated distillations in the presence
of CaH
°
2
, yielding dry 2-isopropyl-2-oxazoline (72 g, b.p. 138
C). H NMR (CDCl , δ): 1.11 (d, 6 H, (CH CH-), 2.49 (m, 1
H, (CH CH-), 3.79 (t, 2 H, NCH CH O), 4.18 ppm (t, 2H,
NCH CH O).
1
3
3 2
)
3
)
2
2
2
2
2
P olym er iza tion s. Hyd r oxy-Ter m in a ted P oly(2-isop r o-
p yl-2-oxa zolin es). A solution of 2-isopropyl-2-oxazoline (9.74
g, 86 mmol) and methyl p-tosylate (0.30 g, 1.0 mmol) in

2
558 Diab et al.
Macromolecules, Vol. 37, No. 7, 2004
nitromethane (30 mL) was stirred at 40 °C under argon.
Aliquots of the polymerization solution (1.0 mL) were taken
at various time intervals (48, 72, 120, and 168 h). They were
cooled to room temperature and poured into a mixture of
aqueous NaOH (1.5 mL, 1.0 M) and methanol (1.5 mL). The
resulting solution was dialyzed against water for 2 days. The
polymer was isolated by freeze-drying. 1H NMR (CDCl
.1 (br s, [(CH CH-], 2.66 and 2.89 (br m, [(CH CH-], 3.08
s, terminal CH ), 3.54 (br s, -(CH -CH -N)-).
Clou d P oin t Deter m in a tion s. Cloud points were deter-
mined by spectrophotometric detection of the changes in
, δ):
3
1
3
)
2
3 2
)
(
3
2
2
-
1
turbidity (λ ) 600 nm) of aqueous polymer solutions (1.0 g L
)
-1
heated at a constant rate (1 °C min ) in a magnetically stirred
UV cell. The value reported is the temperature corresponding
to a decrease of 20% of the solution transmittance.
F igu r e 2. Reaction scheme for the preparation of poly(2-
isopropyl-2-oxazolines) and structure of the polymer.
Differ en tia l Sca n n in g Ca lor im etr y (DSC). DSC mea-
surements were performed on a VP-DSC microcalorimeter
Ta ble 1. P olym er iza tion Con d ition s a n d Ch a r a cter istics
of th e P oly(2-isop r op yl-2-oxa zolin es)
(
MicroCal Inc.) at an external pressure of ca. 180 kPa. The
-
1
cell volume was 0.52 mL. The heating rate was 1.0 °C min
,
Mnb
Mw/
(g mol-1) Mnb (g mol-1) DPc
Mnc
and the instrument response time was set at 5.6 s. Data were
corrected for instrument response time and analyzed using the
software supplied by the manufacturer. The polymer concen-
[M0]/ time
polymer
[I0]a
(h)
PIPOZ17-OH
PIPOZ21-OH
PIPOZ41-OH
PIPOZ50-OH
PNIPAM
86
86
86
86
48
72
120
168
1700
2400
4300
5700
13000
1.05
1.04
1.03
1.03
1.16
1900
2400
4600
5650
17
21
41
50
-
1
tration ranged from 0.5 to 2.0 g L
.
P r essu r e P er tu r ba tion Ca lor im etr y (P P C). PPC mea-
surements were performed on a VP-DSC microcalorimeter
equipped with a pressure perturbation accessory (MicroCal
Inc.). The pressure applied during the compression cycle was
a
Initial monomer/initiator feed ratio. b From GPC measure-
5
00 kPa. The reference cell and sample cell volumes were
ments. c From 1H NMR spectra in CDCl3.
identical (0.52 mL). The polymer concentration was 5 0.0 g
-
1
L
. Control experiments were performed, namely water
(
sample cell) vs water (reference cell), H O vs D O, D O vs D O,
2
2
2
2
NaCl in water vs NaCl in water, and water vs NaCl in water.
The PPC accessory has been described in detail elsewhere.30
Briefly, it applies a pressure of 500 kPa, then a pressure
release to ambient pressure, to the sample, which is kept at
constant temperature. The temperature of the sample cell is
kept constant by compensation of the heat change caused by
the pressure jump. After equilibration, an upward pressure
jump (500 kPa) is performed. The heat peaks caused by the
compression and decompression are opposite in sign but agree
in absolute value. A large number of compression/decompres-
sion cycles are performed at each temperature.
F igu r e 3. MALDI-TOF mass spectrum of PIPOZ50-OH
obtained after 168 h polymerization. An expanded view of the
spectrum ranging from 6100 to 6400 emu is given in the inset.
The coefficient of thermal expansion of a polymer solute in
3
3
a solvent is obtained from eq 1
∆
Qrev
Rpol ) Rsolv
-
(1)
polymerization, initiated with methyl p-tosylate, of
T∆Pmpol Vh pol
2
-isopropyl-2-oxazoline, followed by termination with a
NaOH/methanol solution yielding hydroxyl-terminated
chains (Figure 2). Polymerization conditions were se-
lected to achieve the narrow molecular weight polydis-
persity and low polymer molecular weight required by
the ultimate application of the PIPOZ chains as drug
delivery vehicles. One set of samples (PIPOZn-OH, n )
17, 21, 41, and 50; Table 1) was obtained by sampling a
single polymerization mixture at various time intervals.
This method provided oligo(2-isopropyl-2-oxazolines) of
narrow polydispersity index, as determined by GPC and
MALDI-TOF mass spectrometry (Table 1).
where Rpol and Rsolv are the thermal expansion coefficients of
the polymer and the solvent, respectively, ∆Qrev corresponds
to the heat consumed or released upon the small pressure
changes, ∆P is the change in pressure, T is the temperature,
Vh pol is the partial specific volume of the polymer, and mpol is
the mass of polymer in solution. The partial specific volumes
of the polymers were determined by an increment method
based on the group contribution theory developed to estimate
Vh of aqueous systems and estimated to be accurate within
4
0
3
-1
2
%. Thus, for PIPOZ Vh ) 0.888 cm
g
and for PNIPAM
3
-1
Vh ) 0.894 cm
g .
The change in volume of the solvation layer of the polymer
The mass spectrum of PIPOZ-OH (Mn 6190) presented
in Figure 3 is dominated by one main ion population.
The spacings of the ∆m/z signals, 113.192, are in good
agreement with the expected mass of one monomer unit.
The major population of signals was attributed to the
(∆V) during a phase transition is obtained by integration of
the curve of the changes in the coefficient of thermal expansion
with temperature (eq 2), assuming that ∆V is small compared
to Vh , the intrinsic volume of the polymer. A complete derivation
3
0,33
of the equations has been reported previously.
Data were
analyzed using the software supplied by the manufacturer.
+
Li adduct and the minor population of signals to the
+
Na adduct. On the basis of the mass spectra, it can be
∆
V
V
assumed that all oligomers bear the desired end-
functionalities, namely a methyl group at one chain end
and either a hydroxyl or a primary amine group at the
other chain end.
)
∫R dT
(2)
Resu lts
P r ep a r a tion a n d Ch a r a cter iza tion of th e P oly-
m er s. The oligomers were synthesized by living cationic
The degree of polymerization of the oligomers was
calculated from the H NMR spectra of the samples in
1

Macromolecules, Vol. 37, No. 7, 2004
Poly(2-isopropyl-2-oxazolines) 2559
F igu r e 4. 1H NMR spectrum of PIPOZ21-OH in CDCl
.
3
Ta ble 2. Th er m od yn a m ic P a r a m eter s of th e P oly(2-isop r op yl-oxa zolin es)
Tonset (°C)b
(DSC)
Tmax (°C)b
(DSC)
∆H (kcal mol-1)b
Tmax (°C)c
(PPC)
polymer
solvent
H2O
TCP (°C)a
(DSC)
∆V/V (%)c
PIPOZ17-OH
72.5 ( 0.2
73.1 ( 0.2
68.9 ( 0.2
61.4 ( 0.2
58.9 ( 0.2
51.3 ( 0.2
45.2 ( 0.2
46.5 ( 0.2
46.9 ( 0.2
37.1 ( 0.2
33.5 ( 0.2
24.5 ( 0.2
76.5 ( 0.3
75.2 ( 0.3
64.0 ( 0.3
62.1 ( 0.3
54.0 ( 0.3
51.3 ( 0.3
48.0 ( 0.3
47.0 ( 0.3
40.6 ( 0.3
35.0 ( 0.3
32.6 ( 0.3
0.36 ( 0.01
0.39 ( 0.01
0.72 ( 0.01
0.85 ( 0.01
0.90 ( 0.01
1.16 ( 0.01
1.35 ( 0.01
1.43 ( 0.01
1.61 ( 0.01
1.76 ( 0.01
0.80 ( 0.01
65.0 ( 0.3
63.0 ( 0.3
52.5 ( 0.3
53.0 ( 0.3
45.2 ( 0.3
47.5 ( 0.3
45.0 ( 0.3
43.0 ( 0.3
37.0 ( 0.3
32.0 ( 0.3
32.0 ( 0.3
0.43
0.70
0.64
1.00
0.84
1.30
1.10
1.80
1.30
1.40
0.90
D2O
H2O
D2O
H2O
D2O
H2O
D2O
NaCl (0.5 M)
NaCl (1.0 M)
H2O
PIPOZ21-OH
PIPOZ41-OH
PIPOZ50-OH
62.8 ( 0.2
60.3 ( 0.2
51.3 ( 0.2
48.1 ( 0.2
PNIPAM
D2O
NaCl (0.5 M)
NaCl (1.0 M)
H2O
18.4( 0.2
13.6 ( 0.2
25.6 ( 0.3
20.1 ( 0.3
31.7d
1.20 ( 0.01
1.37 ( 0.01
0.45d
26.0 ( 0.3
21.0 ( 0.3
1.00
1.20
PNIPAM-COOHd
32.7d
(
Mn ∼ 2700)
a
From turbidity measurements (polymer concentration: 1.0 g L-1). b From DSC measurements (polymer concentration: 1.0 g L-1).
From PPC measurements (polymer concentration: 5.0 g L ). From ref 44.
c
-1
d
chloroform-D, using the broad singlet at δ 1.11 ppm,
ascribed to the resonance of the methyl protons of the
isopropyl group of the repeat unit, together with the
singlet at δ 3.08 ppm, attributed to the resonance of the
terminal methyl protons (Figure 4). The number-aver-
age molecular weights derived from integrations of these
H NMR signals are in excellent agreement with the
molecular weights determined by GPC (Table 1) and
MALDI-TOF mass spectra (Figure 3).
P h a se Tr a n sit ion of t h e Oligom er s in Wa t er .
Tu r bid ity Mea su r em en ts. The fastest method to
determine the cloud point of a polymer solution consists
of measuring the changes in turbidity as the solution
is heated at a constant rate. The temperature of
turbidity onset, defined here as the cloud point temper-
ature, TCP, decreases with increasing PIPOZ molecular
weight (Table 2). The turbidity increased sharply as the
temperature exceeded this onset value. To ascertain the
reversibility of the phase transition, polymer solutions
heated to 70 °C were kept at this temperature for 30
min and then cooled to room temperature at constant
cooling rate. The solutions became clear in all cases, but
the temperature for which 80% transmission was re-
covered was lower by ∼2 °C than TCP recorded during
a heating scan. The cloud point temperature of a
1
-
1
solution of PIPOZ21-OH (1.0 g L ) was determined in
light and heavy water: it was lower by ∼1.5 °C in D2O
compared to H2O (Table 2). Note that this trend is in
agreement with an observation made by Chen et al. in
their study of a polydisperse sample of PEOZ (Mw
116 000). They reported that use of D2O in place of H2O
2
5
yielded a cloud point some 3 °C lower.

2
560 Diab et al.
Macromolecules, Vol. 37, No. 7, 2004
F igu r e 5. Microcalorimetric endotherms for aqueous solutions
of PIPOZ-OH samples in water. Polymer concentration: 1.0 g
-
1
-1
L
. Heating rate: 60 °C h
.
High -Sen sitivity Differ en tia l Ca lor im etr y. The
changes with temperature of the partial excess heat
capacity Cp of aqueous solutions of several PIPOZ-OH
samples (1 g L- ) are presented in Figure 5. The
thermograms are endothermic, broad, and markedly
asymmetric, with a sharp increase in heat capacity on
the low-temperature side (onset of the transition, Tonset)
and a gradual decrease of the heat capacity for temper-
atures higher than a maximum temperature Tmax. For
all polymers the values of Cp were the same, within
experimental error, before and after the transition.
Neither changing the concentration of the solution, from
1
F igu r e 6. (top) Plots of the changes as a function of polymer
concentration of the transition temperature, T
m
(full circle),
and of the cloud point, CP (full triangle), for aqueous solutions
of PIPOZ41-OH. (bottom) Plots of the changes as a function of
polymer molecular weight of the enthalpy of the transition,
∆H (open circle), the temperature corresponding to the maxi-
-
1
0
.5 to 5 g L , nor varying the scanning rate, from 10
mum in DSC enthalpograms, T
m
(full circle), and the cloud
to 90 °C/h, affected the shape of the thermograms.
Therefore, the characteristic times of the transitions are
shorter than those of the thermal equilibration of the
cells. The reproducibility of the thermograms was
demonstrated during the second and third heating at
the same heating rate. The transition observed is
completely reversible.
point, CP (full triangle), for aqueous solutions of poly(2-
-1
isopropyl-2-oxazolines). Polymer concentration: 1.0 g L
.
-
1
Heating rate: 60 °C min
.
Uyama and Kobayashi in their study of a PIPOZ sample
2
9
of higher molecular weight and of Lin et al. in their
2
4
investigation of PEOZ.
From plots of the partial heat capacity of PIPOZ
solutions vs temperature, one can extract three ther-
modynamic parameters: Tonset, Tmax, and ∆H, the en-
thalpy of the transition (Table 2). The values of Tonset
and Tmax follow the same trends as the values of TCP:
for a given polymer, they decrease with increasing
concentration, and for solutions of identical concentra-
tion, they decrease with increasing polymer molecular
weight (Figure 6). The enthalpy of transition, as well,
exhibits a significant dependence on molecular weight:
it increases from 0.36 ( 0.01 kcal (mol of monomer
Thermograms recorded for polymer solutions in D O
exhibit the same skewed shape as the thermograms of
2
polymer solutions in H O, as exemplified in Figure 6
2
(top), where we show DSC scans for solutions of PI-
POZ -OH in heavy and light water. In all cases, the
50
Tonset and Tmax values recorded for solutions in D O are
2
lower than those obtained for solutions in H O, and the
2
differences between solutions in the two solvents in-
crease as the molecular weight of the polymer decreases.
P r essu r e P er tu r ba tion Ca lor im etr y. PPC scans
were carried out next with solutions of PIPOZ -OH
n
-
1
unit) for the shortest oligomer to 1.40 ( 0.01 kcal
(n ) 17, 21, 41, and 50) in H O and in D O, yielding
2
2
-
1
mol
for the longest (Figure 6). The latter value
the changes with temperature of the thermal expansion
coefficient, Rpol, shown in Figure 7 (bottom), in the case
of PIPOZ -OH. The plot can be divided in three tem-
corresponds approximately to the energy required to
4
1
break one hydrogen bond per repeat unit. It would
appear, therefore, that the availability of the carbonyl
groups in the PIPOZ chain to form hydrogen bonds with
water increases with molecular weight. A similar ob-
servation was made by Chen et al.26 in their light
scattering studies of PEOZ in water: they report that,
for solutions of this polymer in water, the excess
enthalpy of dilution becomes more negative with in-
creasing polymer molecular weight.
5
0
perature ranges. Below Tonset, for 15 °C < T < 42 °C,
R
pol
remains constant. It undergoes a sharp increase,
reaches a maximum for T
decreases as the temperature further increases. PPC
scans recorded with solutions of the same polymer in
peak
) 45.0 °C, and gradually
D O presented the same features (Figure 7, bottom),
2
with slight differences in the temperatures of the
maximum of Rpol, that correspond to the slight shifts of
Tonset and Tmax already noted in the DSC scan (Figure
7, top). The changes in volume of the solvation layer of
the polymer (∆V), corresponding to the collapse of the
chain and chain aggregation during the phase transi-
tion, can be extracted from PPC scans by integration of
the changes in Rpol as a function of temperature,
The phase transition of PIPOZ oligomers was affected
by the presence of NaCl: increased electrolyte concen-
tration decreased the phase transition temperature and
increased the transition enthalpy (Table 2). Both phe-
nomena are consistent with the well-known “salting-
out” effect of NaCl and corroborate observations of

Macromolecules, Vol. 37, No. 7, 2004
Poly(2-isopropyl-2-oxazolines) 2561
2
), reflecting salt-induced changes in the hydration layer
and in the polymer chain stiffness in solutions of
temperature lower than the transition temperature.
Con clu sion s
As noted in the Introduction, the repeat unit of poly-
2-isopropyl-2-oxazoline) is a structural isomer of the
(
repeat unit of poly(N-isopropylacrylamide). These two
polymers share a number of characteristics when dis-
solved in water: they both undergo a reversible heat-
induced phase transition when brought to a tempera-
ture beyond a critical value. This temperature, detectable
by changes in the transmission of the solution, or more
accurately measured by DSC, depends on salt concen-
tration and is different for solutions of the polymers in
H2O and D2O. The phase transition of PNIPAM has
4
2
been studied in great detail. As the PIPOZ samples
investigated here are monodisperse and of low molec-
ular weight, we chose to compare their properties to
those of PNIPAM oligomers, taking values from pub-
lished reports. Microcalorimetric studies carried out
independently with NIPAM oligomers bearing a car-
boxylic acid end group revealed that solutions of PNIPAM
(Mn 1600, 2200, 2400, and 3500) have a transition
temperature slightly higher than solutions of higher
molecular weight PNIPAM (33.3 vs 31.5 °C) with a
-
1
-1 43,44
transition enthalpy of 0.45 kcal mol
L
.
These
studies concur to indicate that, although the transition
temperature, Tm, is only slightly higher for NIPAM
oligomeric solutions, compared to Tm of solutions of
PNIPAM samples of higher molecular weight and wider
3
3,45
polydispersity,
the transition peaks are wider and
assuming that the intrinsic volume occupied by a
polymer chain remains constant. The volume change,
expressed as ∆V/V in percent of the partial volume of
the polymer (eq 2), is taken as the area defined by the
peak of the PPC scan and a progress baseline (dashed
line, Figure 6, bottom) drawn from projections of the
baselines in the pretransition and posttransition re-
gions. Values of ∆V/V for PIPOZ-OH samples in H2O,
D2O, and aqueous NaCl are listed in Table 2. Comparing
samples of PIPOZ-OH of increasing molecular weight,
we observe that the solvation volume change during the
phase transition is extremely sensitive to the molecular
weight of the oligomer, nearly doubling in value for
solutions of PIPOZ50-OH, compared to the solutions of
the shortest oligomer, PIPOZ17-OH. This significant
effect corroborates the sensitivity to molecular weight
of the enthalpy of transition (vide supra) and may be
taken as added evidence that the number of hydrogen
bond sites along the chain increases with molecular
weight.
Considering next the values of ∆V/V recorded in H2O
and in D2O, we note that for all samples the volume
change in D2O is larger by ∼50% than the value
recovered from measurements in H2O. The enhanced
change in hydration volume in D2O, compared to H2O,
results from a combination of two effects: (1) the PIPOZ
coils may adopt a more extended conformation in D2O,
resulting in a higher level of ordering of the D2O
molecules bound to the polymer chain and hence a
larger volume of bound water, and (2) since D2O is a
more “structured” solvent than H2O, the water mol-
ecules expelled from the hydration layer occupy a larger
volume of bulk solvent. The changes in the volume of
hydration of the polymer during the phase transition
are also affected by the presence of NaCl: the values of
the transition enthalpies weaker in the case of NIPAM
oligomers. The latter two features are observed in our
study of PIPOZ, suggesting that they are related to the
molecular weights of the samples rather than to their
chemical constitution.
There are however several differences in the thermo-
dynamic traits of aqueous solutions of PNIPAM and
PIPOZ oligomers (Table 2). First, the molecular weight
sensitivity of Tm is pronounced in the case of PIPOZ and
not PNIPAM. Second, the transition temperature of
aqueous PNIPAM is higher in D2O, compared to H2O,
whereas aqueous PIPOZ solutions exhibit the opposite
effect. The latter situation seems to be a more common
occurrence: it has been reported for solutions of hy-
4
6
35
droxypropylcellulose and poly(N-vinylcaprolactam),
two polymers that possess a cloud point in water. The
enthalpy of transition is higher in D2O, compared to
H2O, for solutions of PNIPAM or PIPOZ, but the
enhancement in D2O is significantly larger in the case
of PIPOZ. Third, thermograms recorded with PNIPAM
solutions present a negative change in ∆Cp (∆Cp )
-
1
-1
Cp(T > Tm) - Cp(T < Tm) ) -63 J mol
Cp ∼ 0 for PIPOZ. A negative change of Cp was
reported also in studies of the phase transitions of
K ), while
∆
4
7
various pluronic-type block copolymers. Such a nega-
tive heat capacity change during the phase transition
may be taken as an indication of diminished interaction
between water molecules and polymer chains, before
and after the transition.
Ack n ow led gm en t. The work was supported by a
grant of the Natural Sciences and Engineering Research
Council of Canada (to F.M.W.) and by the Special
Coordination Funds for Science and Technology from
the Ministry of Education, Culture, Sports, Science and
∆V/V increase with increasing salt concentration (Table

2
562 Diab et al.
Macromolecules, Vol. 37, No. 7, 2004
Technology pf J apan (MEXT) (to K.K.). F.M.W. thanks
the Yamada Foundation for a Visiting Scientist award
for an extended stay in J apan.
(23) Chiu, T. T.; Thill, B. P.; Fairchok, W. J . In Water Soluble
Polymers; Advances in Chemistry Series 213; Glass, J . E.,
Ed.; American Chemical Society: Washington, DC, 1986; pp
4
25-433.
(
(
(
(
(
24) Lin, P.; Pearce, E. M.; Kwei, T. K. J . Polym. Sci., Part B:
Polym. Phys. 1988, 26, 603.
25) Chen, F. P.; Ames, A. E.; Taylor, L. D. Macromolecules 1990,
Refer en ces a n d Notes
(
(
(
(
(
(
(
(
(
1) Allen, C.; Maysinger, D.; Eisenberg, A. Colloids Surf. B:
Biointerfaces 1999, 16, 3.
2) Yokoyama, M.; Okano, T.; Sakurai, H.; Shibazaki, C.; Katao-
ka, K. Cancer Res. 1991, 51, 3229.
3) Kataoka, K.; Kwin, G. S.; Yokoyama, M.; Okano, T.; Sakurai,
Y. J . Controlled Release 1993, 24, 119.
4) Kataoka, K.; Harada, A.; Nagasaki, Y. Adv. Drug Deliv. Rev.
2
3, 4688.
26) Chen, C. H.; Wilson, J .; Chen, W.; Davis, R. M.; Riffle, J . S.
Polymer 1994, 35, 3587.
27) Christova, D.; Velichkova, R.; Loos, W.; Goethals, E. J .; Du
Prez, F. Polymer 2003, 44, 2255.
28) Heskins, M.; Guillet, J . E. Macromol. Sci. Chem. A2 1968,
1
441.
2
001, 47, 113.
5) Kwon, G. S.; Suwa, S.; Yokoyama, M.; Okano, T.; Sakurai,
Y.; Kataoka, K. J . Controlled Release 1994, 29, 17.
6) Yamamoto, Y.; Nagasaki, Y.; Kato, Y.; Sugiyama, Y.; Kataoka,
K. J . Controlled Release 2001, 77, 27.
7) Otsuka, H.; Nagasaki, Y.; Kataoka, K. Curr. Opinion Colloid
Interface Sci. 2001, 6, 3.
8) Yamamoto, Y.; Nagasaki, Y.; Kato, M.; Kataoka, K. Colloids
Surf. B: Biointerfaces 1999, 16, 113 and references therein.
9) Inoue, T.; Chen, G. H.; Hoffman, A. S.; Nakamae, K. J . Bioact.
Compat. Polym. 1998, 13, 50.
(29) Uyama, H.; Kobayashi, S. Chem. Lett. 1992, 1643.
(30) Lin, L.-N.; Brandts, J . F.; Brandts, M.; Plotnikov, V. Anal.
Biochem. 2002, 302, 144.
(31) Ravrindra, R.; Winter, R. Chem. Phys. Chem. 2003, 4, 359.
(32) Heerklotz, H.; Seelig, J . Biophys. J . 2002, 82, 1445.
(33) Kujawa, P.; Winnik, F. M. Macromolecules 2001, 34, 4130.
(34) Kujawa, P.; Goh, C. C. E.; Calvet, D.; Winnik, F. M.
Macromolecules 2001, 34, 6387.
(
35) Laukkanen, A.; Valtola, L.; Winnik, F. M.; Tenhu, H.
Macromolecules, in press.
(
(
(
(
(
(
(
10) Francis, M. F.; Piredda, M.; Winnik. F. M. J . Controlled
Release 2003, 93, 59.
11) Francis, M. F.; Lavoie, L.; Winnik, F. M.; Leroux, J . C. Eur.
J . Pharmacol. Biopharm. 2003, 56, 337.
12) Lee, S. C.; Chang, Y.; Yoon, J .-S.; Kim, C.; Kwon, I. C.; Kim,
Y.-H.; J eong, S. Y. Macromolecules 1999, 32, 1847.
13) Lee, S. C.; Kim, C.; Kwon, I. C.; Chung, H.; J eong, S. Y. J .
Controlled Release 2003, 89, 437.
14) Kobayashi, S.; Iijima, S.; Igarashi, T.; Saegusa, T. Macro-
molecules 1987, 20, 1729.
15) J ordan, J .; Graf, K.; Riegler, H.; Unger, K. K. Chem. Com-
mun. 1996, 9, 1024.
16) J ordan, R.; Martin, K.; R a¨ der, H.-J .; Unger, K. K. Macro-
molecules 2001, 34, 8858.
(36) Nemethy, G.; Sheraga, H. A. J . Chem. Phys. 1964, 41, 680.
(37) Ben-Naim. A Hydrophobic Interactions; Plenum Press: New
York, 1980.
(38) Murray, J . M.Sc. Dissertation, University of Montreal, 2004.
(39) Seeliger, W.; Aufderhaar, E.; Diepers, W.; Feinauer, R.;
Nehring, R.; Their, W.; Hellmann, H. Angew. Chem., Int. Ed.
1
996, 5, 875.
(
(
40) Zana, R. J . J . Polym. Sci., Polym. Phys. Ed. 1980, 18, 121.
41) Israelachvili, J . N. Intermolecular and Surface Forces; Aca-
demic Press: London, 1985.
(
(
42) Schild, H. G. Prog. Polym. Sci. 1992, 17, 163.
43) Baltes, T.; Garret-Flaudy, F.; Freitag, R. J . Polym. Sci., Part
A: Polym. Chem. 1999, 37, 2977.
(
(
17) Kobayashi, S. Prog. Polym. Sci. 1990, 15, 751.
18) Rehfeldt, F.; Tanaka, M.; Pagnoni, L.; J ordan, R. Langmuir
(44) Freitag, R.; Garret-Flaudy, F. Macromolecules 2002, 18, 3434.
(45) Tiktopoulo, E.; Uversky, V. N.; Lushchik, V. B.; Klenin, S.
I.; Bychkova, V. E.; Ptisyn, O. B. Macromolecules 1995, 32,
1471.
2
002, 18, 4908.
(
(
(
(
19) Goddard, P.; Hutchinson, L. E.; Brown, J .; Brookman, L. J .
J . Controlled Release, 1989, 10, 5.
20) Wang, C. H.; Hsiue, G. H. J . Polym. Sci., Polym. Chem. 2002,
(46) Winnik, F. M. J . Phys. Chem. 1989, 93, 7452.
4
0, 1112.
(
47) Beezer, A. E.; Loh, W.; Mitchell, J . C.; Royall, P. G.; Smith,
D. O.; Tute, M. S.; Armstrong, J . K.; Chowdhry, B. Z.;
Leharne, S. A.; Eagland, D.; Crowtherg, N. J . Langmuir 1994,
21) Woodle, M. C.; Engbers, C. M.; Zalipsky, S. Bioconjugate
Chem. 1994, 5, 493.
22) Poly(ethylene glycol) Chemistry: Biotechnical and Biomedical
Applications; Harris, J . M., Ed.; Plenum Press: New York,
1
0, 4001.
1
992.
MA0358733