Measuring LCSTs by novel temperature gradient methods: Evidence for intermolecular interactions in mixed polymer solutions

Article abstract of DOI:10.1021/ja029691k

Herein we describe studies of molecular interactions in thermoresponsive polymers as they go through phase transitions in aqueous solutions. By using our recently reported linear temperature gradient setup for studying the effects of temperature on chemical processes, we demonstrate the ability to probe lower critical solution temperature (LCST) behavior with excellent precision. This method also provides a simple and convenient way to assay the LCST of solutions containing more than one polymer and follow the clouding kinetics of polymer mixtures in real time. Copyright

Full text of DOI:10.1021/ja029691k

Published on Web 02/13/2003
Measuring LCSTs by Novel Temperature Gradient Methods: Evidence for
Intermolecular Interactions in Mixed Polymer Solutions
Hanbin Mao, Chunmei Li, Yanjie Zhang, David E. Bergbreiter,* and Paul S. Cremer*
Department of Chemistry, Texas A&M UniVersity, College Station, Texas 77845-3012
Received December 11, 2002 ; E-mail: cremer@mail.chem.tamu.edu
Polymers exhibiting a lower critical solution temperature (LCST)
are of interest in contexts as diverse as catalysis,1 green chemistry,2
drug delivery,3 sensor design,4 and permeability control in thin film
composites.5 Aqueous solutions of polymers such as poly(N-
isopropylacrylamide) (PNIPAM) are common examples of such
thermoresponsive materials in that they precipitate above their
LCSTs and redissolve below this temperature.6 Many proteins
display similar behavior.7,8
Although the precipitation mechanisms are often complex, it is
generally accepted that the desorption of water molecules from the
hydrophobic portions of the macromolecule causes the formation
of a more compact macromolecular structure and is directly
Figure 1. Schematic drawing of a temperature gradient device. Each CCD
image fits six capillary tubes simultaneously under a 2× objective.
associated with the phase transition.9 Beyond this, there exists a
measure of uncertainty about the mechanism of the LCST phase
behavior. For example, it has been suggested that the clouding of
a polymer solution proceeds in two cooperative steps with the first
stage involving the folding of individual macromolecules and the
second being aggregation of the folded polymers.10 The details of
the process whereby soluble polymers become precipitated ag-
gregates, however, are unknown.
Existing experimental methods have afforded many examples
of LCST behavior and have shown that LCST phenomena in poly-
(N-alkylacrylamide)s in particular are dependent on the identity of
the N-alkyl group and other solution components (e.g., salts).
However, while there are many studies of LCST phenomena, efforts
employing existing techniques can suffer from experimental dif-
ficulties. For example, if the reversibility of the process is being
studied, prolonged temperature ramping is required to minimize
hysteresis.11 Furthermore, the speed of data collection may be
limited by the detection method. Typical studies employ transmit-
Figure 2. Clouding curve for PNIPAM (10 mg/mL, Mw 340 kDa, LCST
30.5 °C) in DI water solution. Inset: CCD image of the PNIPAM clouding.
The temperatures at the two ends were at 26.4 and 34.5 °C.
tance, light scattering, and differential scanning calorimetry (DSC)
Rectangular borosilicate capillary tubes with a high aspect ratio
for measuring the LCST. These methods involve taking data
(100 µm × 1 mm × 2 cm) were placed parallel to the direction of
sequentially as a function of temperature. This Communication
the thermal gradient; therefore, determining the temperature along
describes a new high throughput approach for studying LCSTs that
the tube was a simple matter of measuring the lengthwise position.
only requires a few microliters of sample and provides excellent
As a demonstration, a tube was filled with an aqueous PNIPAM
temperature resolution capabilities.
solution. Since the cloudy portion of the solution scattered
Previous measurements of the LCSTs for PNIPAM and poly-
(N,N-diethylacrylamide) (PDEAM) show that these macromolecules
have very similar clouding points (31 °C ((1 °C) and 30 °C ((1
significantly more light than the clear portion, the clouding process
could be observed under an optical inverted microscope by
employing dark field microscopy (Figure 2). Interestingly, although
the solution has a typical polydispersity (1.89 for PNIPAM), the
temperature range of the phase transition only spanned about 0.2
°C. The LCST of PDEAM with a lower degree of polymerization
(Mw ) 230 kDa) and a comparable polydispersity (2.09) was
similarly precise, indicating that a narrow polydispersity is not
necessarily required to obtain such well-defined phase transitions.
LCSTs of PNIPAM and PDEAM measured with the device in
Figure 1 were 30.5 ( 0.16 °C and 29.3 ( 0.15 °C, respectively,
and remained constant with time. Using this device, up to six
samples of polymer solutions could be analyzed at once using two
°C)), respectively.12,13 We have measured these as well, using a
temperature gradient device14 developed in our laboratory, which
collects data as a function of position rather than time (Figure 1).
This device allows us to make multiple measurements of LCST
phenomena in a single experiment with microliter sample volumes.
Significantly, this device allows us to study mixed polymer systems
with sufficient resolution to reveal mechanistic insights into the
LCST behavior of mixed polymer systems.
In our setup, hot and cold brass tubes were placed in parallel on
a glass slide.14 A linear temperature gradient forms between them
in accordance with the Fourier heat diffusion equation.15
9
2850
J. AM. CHEM. SOC. 2003, 125, 2850-2851
10.1021/ja029691k CCC: $25.00 © 2003 American Chemical Society

C O MMU N I C A T I O N S
chains that later segregate into pure PNIPAM and PDEAM. In other
words, early aggregate formation most likely involves the mixing
of PNIPAM and PDEAM. Such mixed aggregates would have
higher entropy than the corresponding structures containing only
one type of polymer. The difference in entropy between the pure
and mixed systems should lead to a lower transition state free
energy, ∆Gq, to precipitation for the mixed complexes. Interestingly,
this initial precipitate then begins to phase segregate into pure
PNIPAM and PDEAM particles presumably on enthalpic grounds.
To verify these assumptions, DSC studies of PNIPAM and PDEAM
were performed as a function of mole fraction of PNIPAM. The
results showed that the onset to the phase transition came at a lower
temperature for mixed systems by amounts that exactly matched
the data in Figure 4. Furthermore, the overall enthalpy for the entire
precipitation process rose continuously as the amount of PNIPAM
was increased. This latter result is consistent with the notion that
the two polymers must ultimately segregate upon undergoing the
phase transition and that PNIPAM has the higher phase transition
enthalpy. The DSC results are provided in the Supporting Informa-
tion.
Figure 3. Kinetic studies of PNIPAM and PDEAM mixtures. PNIPAM
and PDEAM were mixed as 1:1.5 molar ratio in DI water.
The methods and results described above provide a new and rapid
protocol for studying the precipitation/solubilization behavior of
thermoresponsive polymers with minute quantities of materials and
excellent signal-to-noise ratios. Indeed, the high throughput potential
for this method should make it possible in the near future to collect
large amounts of data with great temperature precision quite easily.
Figure 4. Plots of the initial LCST at different mole fractions of PNIPAM
for PNIPAM and PDEAM mixtures. The solid line represents a fit to the
data.
Acknowledgment. We thank the National Science Foundation
(DMR-9977911) (D.E.B.), the Army Research Office (DAAD19-
01-1-0346) (P.S.C.), and the Office of Naval Research (YIP Award
NOOO14-00-1-0664) (P.S.C.). P.S.C. also acknowledges the fol-
lowing: Beckman Young Investigator Award, an Alfred P. Sloan
Fellowship, and a Nontenured Faculty Award from 3M Corporation.
H.M. acknowledges support from a Proctor & Gamble Fellowship.
different polymer solutions with known LCSTs as internal controls
to aid in temperature readout and prevent systematic errors between
experiments.
Since the clouding curves of macromolecules in solution can be
obtained with good temperature resolution over relatively short
times, the time-dependent evolution of the clouding process was
studied simply by recording a series of time-lapse images of a given
sample with a standard CCD camera. When this was done for a
mixed solution of PNIPAM and PDEAM, a series of time-
dependent phenomena were revealed (Figure 3). The first image
was taken immediately after the mixed polymer solution was placed
over the temperature gradient. As can be seen, the original clouding
curve of the mixture appeared to be smooth. After 6 min, however,
a higher temperature kink appeared in the curve. This further
evolved into a pronounced dip after 9 min. At this point, cycling
between the latter two stages occurred; however, the initial sequence
of clouding behavior as well as the migration of the LCST toward
higher temperature were reproducible over many samples and within
a given sample. Reversibility within the same sample was observed
when the TLow and the THigh sides of the tube were switched.
To investigate the significance of the 0 min conditions, experi-
ments for mixtures of PNIPAM and PDEAM were carried out as
a function of mole fraction of PNIPAM. A plot of the nascent LCST
temperature is shown in Figure 4. The results bear a striking
resemblance to the curve obtained for the Gibbs free energy of
mixing of two substances: G ) ø1G1 + ø2G2 + nRT(ø1 ln ø1 + ø2
ln ø2). In this case, G1 and G2 are the Gibbs free energies of pure
substances 1 and 2, respectively. ø1 and ø2 are the mole fractions
of substances 1 and 2, respectively. This correlation implies that
the aggregation process involves the rapid (i.e., kinetically con-
trolled) formation of aggregates of a small number of polymer
Supporting Information Available: Experimental procedures and
characterization data (PDF). This material is available free of charge
via the Internet at http://pubs.acs.org.
References
(1) Mariagnanam, V. M.; Zhang, L.; Bergbreiter, D. E. AdV. Mater. 1995, 7,
69-71.
(2) Huang, X.; Witte, K. L.; Bergbreiter, D. E.; Wong, C.-H. AdV. Synth.
Catal. 2001, 343, 675-681.
(3) Hoffman, A. S. J. Controlled Release 1987, 6, 297-305.
(4) Ito, T.; Hioki, T.; Yamaguchi, T.; Shinbo, T.; Nakao, S.; Kimura, S. J.
Am. Chem. Soc. 2002, 124, 7840-7846.
(5) Rao, G. V. R.; Balamurugan, S.; Meyer, D. E.; Chilkoti, A.; Lopez, G. P.
Langmuir 2002, 18, 1819-1824.
(6) Schild, H. G. Prog. Polym. Sci. 1992, 17, 163-249.
(7) Urry, D. W. J. Phys. Chem. B 1997, 101, 11007-11028.
(8) Urry, D. W.; Starcher, B.; Partridg, S. M. Nature 1969, 222, 795.
(9) Badiger, M. V.; Lele, A. K.; Bhalerao, V. S.; Varghese, S.; Mashelkar,
R. A. J. Chem. Phys. 1998, 109, 1175-1184.
(10) Winnik, F. M. Polymer 1990, 31, 2125.
(11) Boutris, C.; Chatzi, E. G.; Kiparissides, C. Polymer 1997, 38, 2567-
2570.
(12) Taylor, L. D.; Cerankowski, L. D. J. Polym. Sci.: Polym. Chem. Ed. 1975,
13, 2551-2570.
(13) Freitag, R.; Baltes, T.; Eggert, M. J. Polym. Sci.: Part A: Polym. Chem.
1994, 32, 3019-3030.
(14) (a) Mao, H.; Yang, T.; Cremer, P. S. J. Am. Chem. Soc. 2002, 124, 4432-
4435. (b) Mao, H.; Holden, M. A.; You, M.; Cremer, P. S. Anal. Chem.
2002, 74, 5071-5075.
(15) Incropera, F. P.; DeWitt, D. P. Fundamentals of Heat and Mass Transfer,
3rd ed.; John Wiley & Sons: New York, 1990.
JA029691K
9
J. AM. CHEM. SOC. VOL. 125, NO. 10, 2003 2851