Effect of the molecular weight on the lower critical solution temperature of poly(N,N,-diethylacrylamide) in aqueous solutions

Article abstract of DOI:10.1139/v01-180

The molecular weight dependence of the lower critical solution temperature of poly(N,N-diethylacrylamide) was studied with 11 samples of the polymer with a number-average molecular weight (M_n) ranging from 9.6 × 10³ to 1.3 × 10⁶ g mol^-1 and relatively narrow polydispersity indices from 1.19 and 2.60. These samples were obtained by fractional precipitation of the polymer. LCST values of polymers were measured by turbidimetry and microcalorimetry. An inverse dependence of LCST on the molar mass was found and the LCST of the samples remained more or less a constant above a critical molecular weight of ca. 2 × 10⁵ g mol^-1. The enthalpy and the entropy changes as well as the LCST of the polymer depend strongly on the molar mass of the polymer, especially in low molecular weight range.

Full text of DOI:10.1139/v01-180

1870
Effect of the molecular weight on the lower
critical solution temperature of poly(N,N-
diethylacrylamide) in aqueous solutions1
D.G. Lessard, M. Ousalem, and X.X. Zhu
Abstract: The molecular weight dependence of the lower critical solution temperature of poly(N,N-diethylacrylamide)
was studied with 11 samples of the polymer with a number-average molecular weight (Mn) ranging from 9.6 × 103 to
1.3 × 106 g mol–1 and relatively narrow polydispersity indices from 1.19 and 2.60. These samples were obtained by
fractional precipitation of the polymer. LCST values of polymers were measured by turbidimetry and microcalorimetry.
An inverse dependence of LCST on the molar mass was found and the LCST of the samples remained more or less a
constant above a critical molecular weight of ca. 2 × 105 g mol–1. The enthalpy and the entropy changes as well as the
LCST of the polymer depend strongly on the molar mass of the polymer, especially in low molecular weight range.
Key words: poly(N,N-diethylacrylamide), LCST, thermosensitive, phase diagram, effect of molecular weight.
Résumé : L’effet de la masse molaire sur la température critique de dissolution inférieure (LCST) du poly(N,N-dié-
thylacrylamide) a été étudié sur 11 fractions de masses molaires moyennes en nombre (Mn) variant entre 9.6 × 103 à
1.3 × 106 g mol–1 et ayant un indice de polydispersité situé entre 1.19 et 2.60. Ces échantillons ont été obtenus par
précipitation fractionée. La LCST des solutions aqueuses du polymère a été mesurée par turbidimétrie et par microcalo-
rimétrie. Il a été constaté que la LCST diminue avec la masse molaire de l’échantillon. Les valeurs de la LCST
demeurent cependant relativement constantes pour des masses molaires supérieures à 2 × 105 g mol–1. L’enthalpie et
l’entropie associées aux changements de phases diminuent de la même façon avec l’augmentation de la masse molaire.
Mots clés : poly(N,N-diéthylacrylamide), température critique de dissolution inférieure, thermosensible, diagramme de
phase, effet de la masse molaire.
Lessard et al. 1874
Introduction
solution (3, 4), the formation of hydrogen bonds between the
polymer and water (3, 5), the hydrophobic interactions be-
tween the polymer side chain groups (3, 6), and the disrup-
tion of specific hydrogen-bonded cyclic structure in alkyl
amide units and hydroxyl functions of water (7).
Only a few studies were conducted on PDEA even though
this polymer behaves similarly as PNIPAM without the ability
to form hydrogen bridges with a proton on the amide group
and the solvent. This polymer may be synthesized by anionic
polymerization (8, 9), group transfer polymerization (8, 10),
or free radical polymerization (7, 8, 10–14). It was observed
that for PDEA of low molar masses (Mn < 5000 g mol–1) the
transition temperature is 10°C higher for isotactic polymers
than for their syndiotactic homologues. Different initiators
were used for these tactic samples and the difference in the
end-group was attributed as the origin of the discrepancy in
their LCST values (8).
Thermosensitive polymers are known to undergo changes in
their physical properties when the temperature is varied. Many
N-substituted polyacrylamides in water exhibit reversible-phase
separation upon heating. The temperature at which precipita-
tion of the polymer occurs is called the lower critical solu-
tion temperature (LCST). Materials which show this
behavior in water are of a great interest for medical (1) and
industrial (2) applications. The most extensively studied
polymer in this family is probably poly(N-isopro-
pylacrylamide) (PNIPAM). Our research focuses on the ef-
fect of the molar mass on the LCST of aqueous solutions of
poly(N,N-diethylacrylmide) (PDEA), which behaves simi-
larly as PNIPAM.
Different researchers have explained the LCST phenome-
non of polymers in water. It was ascribed to the greater
entropy in the two-phase system than in a homogeneous
The aqueous solutions of PDEA can undergo precipitation
between 25 and 36°C as determined by means of turbi-
dimetry (8, 11, 15–17), differential scanning calorimetry (7,
8, 11, 17, 18), small angle neutron scattering (20), dynamic
and static light scattering (10), IR spectroscopy (7), and rhe-
ology (10, 14). Early research focused on the parameters
such as the addition of salts (11, 15, 17) and surfactant (11),
co-monomer content (13, 14, 17, 19), and concentration of
the polymer (8, 11, 16). The self-diffusion coefficients of
small molecules in PDEA aqueous solutions and gels were
also determined by the use of pulsed-gradient spin-echo
Received August 14, 2001. Published on the NRC Research
Press Web site at http://canjchem.nrc.ca on December 6, 2001.
D.G. Lessard, M. Ousalem, and X.X. Zhu.2 Département de
chimie, Université de Montréal, C.P. 6128, Succ. Centre-ville,
Montréal, QC H3C 3J7, Canada.
1This paper is dedicated to Professor G.R. Brown, who
passed away on June 19, 2001.
2Corresponding author (telephone: (514) 343-6733;
fax: (514) 343-7586; e-mail: julian.zhu@umontreal.ca).
Can. J. Chem. 79: 1870–1874 (2001)
DOI: 10.1139/cjc-79-12-1870
© 2001 NRC Canada

Lessard et al.
1871
NMR techniques (12). Most of these studies dealt with
polydisperse PDEA samples, and none of them addressed
clearly the question of the molecular weight effect on the
LCST of the polymers.
In the present work, using fractional precipitation, we pre-
pared 11 PDEA fractions with molar masses ranging from
9.6 × 103 to 1.3 × 106 g mol–1. Turbidimetry, also called
cloud point technique, and microcalorimetry were used to
study the effect of molecular weight on the LCST of these
fractions in aqueous solutions.
A number of parameters have been reported to influence
the LCST of thermosensitive polymers. It was shown that
the presence of salts in the solution generally decrease the
LCST of the solution (5, 20) and added surfactants improve
the solubility of the polymer, thereby raising the LCST (20–
22). Also, phase separations were studied as functions of the
amount of cosolvent (23), polymer concentration (24), and
pressure (25). Investigations of molecular weight influence
on the LCST, however, showed a large discrepancy in the re-
sults obtained. Some authors reported an inverse dependence
of the LCST on the molar mass (5, 24), while others claimed
the opposite (26). Several studies also reported that the mo-
lecular weight had no influence on the LCST (6, 27). It is
worthwhile to mention that the fractionation of this kind of
polymers is quite difficult due to the amphiphilicity of the
macromolecule.
allowed to cool down to the precipitation temperature. The
system was kept without stirring for at least 1 day before the
precipitated fraction was recovered. The remaining solution
underwent further precipitation in the same way as described
above. This process was repeated to obtain further fractions.
In total, 11 fractions were obtained.
The molecular weight and its distribution of the polymer
samples were determined by size exclusion chromatography
(SEC) on a Waters system equipped with an online Waters
410 differential refractometer and a set of three Ultrastyragel
columns. Tetrahydrofuran was used as the mobile phase, and
polystyrene samples as the standards in the calibration of the
molar masses.
Determination of LCST
The cloud point of PDEA in water was set as the tempera-
ture at the inflexion point of the curve obtained by turbi-
dimetry. A CARY 1 BIO UV–vis spectrophotometer coupled
to a temperature controller was used in the absorbance mode
at a wavelength of 500 nm. A 1-cm sample cell containing
ca. 80 L of solution was used for the measurement against
deionized Milli-Q water as the reference. The cloud point of
sample with higher concentration was impossible to deter-
mine with enough precision due to the saturation of the sig-
nal on the spectrophotometer.
Microcalorimetry was also used to determine the LCST of
PDEA aqueous solutions. The endothermic signal at its max-
imum (corresponding to the LCST) was recorded for each
sample on a DSC VP-microcalorimeter from Microcal Inc.
A cell filled with deionized Milli-Q water was used as the
reference. Heating rates of 0.1 and 1.0°C min–1 were used,
with both methods, to see their effect on the LCST.
The LCST values of aqueous PDEA solutions with differ-
ent molecular weight were determined by means of turbi-
dimetry and microcalorimetry at a concentration of 1 wt% to
ensure an exact value of the cloud point and the LCST.
Experimental section
All chemicals were purchased from Sigma (St. Louis,
Missouri, U.S.A.) and Aldrich (Milwaukee, Wisconsin,
U.S.A.) and were used as received. Ammonium persulfate
was recrystallized before use.
Polymer synthesis
N,N-Diethylacrylamide (DEA) was prepared, as reported
previously (11), by reacting acryloyl chloride (97%) with an
excess of diethylamine (98%) in methylene chloride at 0°C.
The salt was filtered off and the solvent was evaporated.
Distillation of the liquid under 0.5 mmHg (1 mmHg =
133.322 Pa) vacuum at 40–50°C in the presence of hydro-
quinone yielded a clear liquid, which was kept in a freezer
until use.
Results and discussion
The fractionated samples of PDEA have a wide range of
molar masses, from 9.6 × 103 to 1.3 × 106 g mol–1. The
polydispersity of the fractions were lower than that of the
bulk sample (Mw/Mn = 3.59) as shown in Table 1.
PDEA was obtained by radical solution polymerization of
DEA using ammonium persulfate as the initiator and
N,N,N>,N>-tetramethylethylenediamine as the accelerator.
Purified water was degassed with nitrogen and was used as
the solvent. The reaction was carried out at room tempera-
ture for 4 h. The solvent was then evaporated and PDEA,
after dissolution into a small amount of acetone, was puri-
fied twice by precipitation in petroleum ether, resulting in a
white solid.
The LCST of a solute dissolved in a solvent can be ob-
tained from the phase diagram of the system. The phase dia-
grams of two polymer samples F2 and F9 are shown in
Fig. 1, where the cloud point is plotted as a function of con-
centrations ranging from 0.0001 to 10 wt%. The LCST is
given by the lower temperature at which the phase transition
occurs. It appears that the most accurate value of the LCST
can be obtained for concentrations ranging from 1 to 5 wt%
for low and high molecular weight samples, and no signifi-
cant differences on the LCST are observed within this inter-
val of polymer concentrations.
Fractionation and characterization
PDEA fractions were obtained by fractional precipitation
using acetone as the solvent and hexane as the nonsolvent.
To ensure good results, the concentration of the PDEA solu-
tion was kept low (less than 0.1 wt%) and the temperature
was maintained constant at 25°C. Hexane was added until
the solution turned cloudy. Then the temperature was raised
by 1 or 2°C to dissolve the precipitated polymer and then
As shown in Fig. 2a, the transmittance of the 1 wt%
PDEA solution decreases sharply at a certain temperature
owing to the turbidity of the solutions when precipitation oc-
curred. The exact values of the cloud point (LCST), DH, and
DS are listed in Table 1. The general trend observed for this
series of samples is that the LCST is lower for the higher
molecular weight polymers.
© 2001 NRC Canada

1872
Can. J. Chem. Vol. 79, 2001
Table 1. Molecular weight and LCST of the PDEA fractions.
LCST (°C)a
Samples
Mn (g mol–1)
Mw (g mol–1)
Mw /Mn
DH (J g–1)
DS (J g–1 K–1)
Turbidimetry
Microcalorimetry
F1
F2
F3
F4
F5
F6
F7
F8
F9
F10
F11
9 600
19 200
32 500
81 600
90 400
13 300
40 300
58 500
1.39
2.10
1.80
2.03
1.46
2.60
2.08
1.43
1.95
1.50
1.19
16.9
26.0
16.6
20.1
26.0
22.2
5.3
5.7
8.1
2.0
1.2
0.036
0.069
0.039
0.016
0.085
0.059
0.010
0.018
0.003
0.003
0.006
32.9
31.0
30.9
30.1
29.7
29.5
29.3
29.1
28.4
28.6
28.6
37.0
34.2
32.6
31.0
30.3
30.2
29.6
29.2
28.8
28.6
28.2
165 700
132 000
252 000
376 300
311 700
709 000
890 400
1 547 000
96 900
180 900
218 000
363 600
593 600
1 300 000
aHeating rate at 0.1°C min–1.
Fig 1. Phase diagram of the aqueous solution of PDEA of a low
molar mass fraction F2 (open circles) and high molar mass frac-
tion F9 (closed circles).
Fig. 2. LCST of 1 wt% aqueous solutions of selected PDEA
fractions (see Table 1) obtained at a heating rate of 0.1°C min–1
by: (a) turbidimetry; and (b) microcalorimetry.
40
38
36
34
32
30
28
100
80
60
40
20
0
(a)
F5
F4
F11
F3
(b)
0
5
10
15
F10
[PDEA] (wt%)
Microcalorimetric studies confirmed the general tendency
of the behavior of the PDEA fractions. The endothermic
phase transition occurs at lower temperatures for higher mo-
lecular weight samples as illustrated in Fig. 2b.
F9
F11
F7
The inverse dependence on molar mass of the LCST was
recently observed on PNIPAM (24). This phenomenon is at-
tributed to the difference in the free volumes caused by the
polymer chains and the solvent molecules. This difference is
much more significant for the longer polymer chains, since
they should precipitate at lower temperatures (28, 29). It was
demonstrated by Patterson (28) that the LCST is propor-
tional to the critical value of the Flory–Huggins interaction
parameter (cc) and that the LCST decreases with the ratio of
the molar volume of the polymer to that of the solvent (r):
27 28 29 30 31 32 33
Temperature (oC)
Mn. This may explain the broadening of the transmittance
curves (Fig. 2a) and of the endothermic peaks as Mn de-
creases (Fig. 2b). This implies a minor role of the poly-
dispersity on the width of transition for heavier fractions. A
critical value of Mn seems to lie at ca. 2 × 105 g mol–1, after
which the LCST values of the samples in water remained
nearly constant.
[1]
cc = 1/2(1 + r –1/2)²
In eq. [1], if the molar volume of the polymer increases with
the length of the chain, it is clear that the LCST of the poly-
mer must decrease with the molecular weight of the poly-
mer.
The shape of the curves presented in Fig. 3 suggests a
more pronounced effect for lower molecular weight frac-
tions. The slope is steep for values of Mn lower than 2 ×
105 g mol–1 and becomes more gradual for higher values of
The effect of the heating rate (1.0 and 0.1°C min–1) was
also investigated. At higher heating rates, the transition tem-
perature is expected to be overestimated because of the poor
heat transfer. When working with the 1.0 wt% PDEA frac-
tions, the cloud point technique was found to be more sensi-
tive to the heating rate as shown in Fig. 3a, where the LCST
© 2001 NRC Canada

Lessard et al.
1873
Fig. 4. Effect of the molecular weight on: (a) DH; and (b) DS
Fig. 3. Effect of the molecular weight on the LCST of the
PDEA fractions measured by: (a) spectrophotmetry (circles); and
(b) microcalorimetry (squares) for different heating rates. Open
symbols, heating at 1.0°C min–1; closed symbols, 0.1°C min–1.
for the PDEA fractions obtained by microcalorimetry at a heat-
ing rate of 0.1°C min–1.
30
(a)
40
(a)
38
20
10
36
34
32
30
0
0.10
28
40
(b)
(b)
38
0.05
36
34
32
30
28
0.00
0
500
3
Mn (1 × 10 g mol
1000
–
1500
1
)
0
500
1000
1500
–
1
3
M (1 × 10 g mol
n
)
(28). A low molar mass polymer at a given weight concen-
tration needs to interact with more water molecules to be
solvated, creating more order in the system. This increased
order decreases the overall entropy of the solution. The DS
of the transition is therefore greater.
It can also be demonstrated mathematically that lower
molecular weight polymers correspond to higher the DH val-
ues. It is known that cc decrease with the molecular weight
of the polymer (28) because the lower molar volume of the
polymer gives a lower value of the ratio (r) in eq. [1]. Since
cc is proportional to the energy required to break the contact
between the polymer and the solvent (28), the energy needed
to induce the phase separation (DH) will be greater for
shorter polymer chains.
obtained at 0.1°C min–1 were 3°C lower than those mea-
sured at 1.0°C min–1. Such a dependence of the LCST on the
heating rate was already reported for a polydisperse sample
(11). This was attributed to the poor heat transfer inherent to
the method where the recorded temperature is higher than
the effective temperature of the sample. Microcalorimetry
has a much better heat transfer and, therefore, this effect is
less pronounced as shown in Fig. 3b.
As seen in Fig. 4, microcalorimetry can be used also to
determine the enthalpy (DH) and the entropy (DS) at the
transition. These quantities seem to be strongly dependent
on the molecular weight. DH and DS follow the same trend
as the LCST for the samples, i.e., a hyperbolic decrease
from 21.0 J g–1 and 0.070 J g–1 K–1 for DH and DS to limit
values of 0.6 J g–1 and 0.003 J g–1 K–1, respectively. In each
case, the final plateau corresponds to a sample with a molar
mass of 2 × 105 g mol–1. The value obtained here for DH is
consistent with that reported by Idziak et al. (11) for a
PDEA sample (DH = 22.9 J g–1 for Mn = 2.0 × 104 g mol–1).
It is understandable that the entropy change is smaller for
the longer polymer chains. The DS values decrease as a
function of the average molar mass of the polymer and re-
main constant after a critical point at Mn = 2 × 105 g mol–1.
When a hydrophobic solute such as PDEA is dissolved in
water, the surrounding molecules tend to organize them-
selves in an ice-like structure (3). The order created de-
creases the entropy of the solution. The LCST phenomenon
for polymer in aqueous media is possible only if the dissolu-
tion is exothermic and the entropy of the mixture is negative
Conclusion
PDEA fractions were prepared by fractional precipitation
despite the amphiphilicity of the polymer. The phase dia-
grams showed that the changes of the LCSTs of the PDEA
aqueous solutions are not significant in the concentration
range from 1 to 5 wt%. Low heating rates provide more reli-
able results with turbidimetry, but this effect is negligible for
microcalorimetric measurements. Turbidimetric and micro-
calorimetric measurements have shown a sharp decrease in
their LCST, DH, and DS values as a function of the molar
mass of PDEA. The values become constant after a critical
molar mass at ca. Mn = 2 × 105 g mol–1. Polydispersity of
the polymer seems to plays a minor role for the heavier
PDEA fractions.
© 2001 NRC Canada

1874
Can. J. Chem. Vol. 79, 2001
12. L. Masaro, M. Ousalem, W.E. Baille, D. Lessard, and X.X.
Zhu. Macromolecules, 32, 4375 (1999).
Acknowledgements
13. L. Bromberg and G. Levin. J. Polym. Sci. Part A: Polym.
Chem. 34, 2595 (1996).
14. J. Hrouz and M. Ilavsky. Polym. Bull. 22, 271 (1989).
15. R. Freitag, T. Baltes, M. Eggert, K. Schügerl, and U. Bahr.
Bioseparation, 4, 353 (1994).
The financial support from Natural Sciences and Engi-
neering Research Council of Canada (NSERC) and Fonds
FCAR of Quebec is gratefully acknowledged.
16. M. Eggert and R. Freitag. J. Polym. Sci. Part A: Polym. Chem.
32, 803 (1994).
17. A. Percot, M. Lafleur, and X.X. Zhu. Polymer, 41, 7231 (2000).
18. L.D. Taylor and L.D. Cerankosky. J. Polym. Sci. 13, 2551
(1975).
References
1. R. Yoshida, K. Sakai, T. Okano, Y. Sakurai, Y.H. Bae, and
S.W. Kim. J. Biomater. Sci. Polym. Ed. 3, 155 (1991).
2. K.S. Kazanskii and S.A. Dubrovskii. Adv. Polym. Sci. 104, 97
(1992).
19. J. Plestil, Y.M. Ostanevich, S. Borbely, J. Stejskal, and M.
Ilavsky. Polym. Bull. 17, 465 (1987).
3. H.G. Schild. Prog. Polym. Sci. 17, 163 (1992).
4. M. Heskins and J.E. Guillet. J. Macromol. Sci.-Chem. Part A,
8, 1441 (1968).
5. H.G. Schild and D.A. Tirrell. J. Phys. Chem. 94, 4352 (1990).
6. S. Fujishige, K. Kubota, and I. Ando. J. Phys. Chem. 93, 3311
(1989).
7. N.A. Platé, T.L. Lebedeva, and L. Valuev. Polym. J. (Tokyo),
31, 21 (1999).
8. R. Freitag, T. Baltes, and M. Eggert. J. Polym. Sci. Part A:
Polym. Chem. 32, 3019 (1994).
9. M. Kobayashi, T. Ishizone, and S. Nakahama. Macromolecules,
33, 4411 (2000).
10. M. Itakura, K. Inomata, and T. Nose. Polymer, 41, 8681 (2000).
11. I. Idziak, D. Avoce, D. Lessard, D. Gravel, and X.X. Zhu.
Macromolecules, 32, 1260 (1999).
20. J. Eliassaf. J. Appl. Polym. Sci. 22, 873 (1978).
21. L.T. Lee and B. Cabane. Macromolecules, 30, 6559 (1997).
22. G. Staikos. Macromol. Rapid Commun. 16, 913 (1995).
23. H.G. Schild, M. Muthukumar, and D.A.Tirrell. Macromolecules,
24, 948 (1991).
24. Z. Tong, F. Zeng, and X. Zheng. Macromolecules, 32, 4488
(1999).
25. K. Otake, R. Karaki, T. Ebina, C. Yokoyama, and S. Takahashi.
Macromolecules, 26, 2194 (1993).
26. X. Zheng, Z. Tong, X. Xie, and F. Zeng. Polym. J. (Tokyo),
30, 284 (1998).
27. E.I. Tiktopoulo, V.N. Uversky, V.B. Lushchik, S.I. Klenin, V.E.
Bychkova, and O.B. Ptitsyn. Macromolecules, 28, 7519 (1995).
28. D. Patterson. Macromolecules, 2, 672 (1969).
29. F. Wang, S. Saeki, and T. Yamaguchi. Polymer, 40, 2779 (1999).
© 2001 NRC Canada