Journal of Colloid and Interface Science 302 (2006) 467–474www.elsevier.com/locate/jcis

Importance of bound water in hydration–dehydration behaviorof hydroxylated poly(N -isopropylacrylamide)

Tomohiro Maeda a, Kazuya Yamamoto a, Takao Aoyagi a,b,∗

a Department of Nanostructure and Advanced Materials, Graduate School of Science and Engineering, Kagoshima University,1-21-40, Korimoto, Kagoshima 890-0065, Japan

b PRESTO, Japan Science and Technology Agency, 4-1-8 Honcho Kawaguchi, Saitama, Japan

Received 21 April 2006; accepted 27 June 2006

Available online 30 June 2006

Abstract

In this study, a differential scanning calorimetric analysis was performed to investigate the role of water existing around the polymer chainson their thermoresponsive behaviors using the novel functional temperature-sensitive copolymer, poly(N -isopropylacrylamide-co-2-hydroxy-isopropylacrylamide) (poly(NIPAAm-co-HIPAAm)). The HIPAAm comonomers were incorporated into the polymeric chains as hydrophilicparameters, and then the hydration states of poly(NIPAAm-co-HIPAAm) with various HIPAAm compositions were examined. Bound water,which is affected by the polymeric chains to some extent, was produced by adding the copolymers to the water, and the temperature due to themelting of the bound water decreased as the HIPAAm content increased. On the basis of this result, we considered that the interaction betweenthe bound water and the polymeric chains is reinforced by the increasing HIPAAm composition. In addition, the cloud points of the copolymersshifted to a higher temperature, and the endothermic enthalpy for the phase transition decreased with increasing the HIPAAm content, suggestingthat the number of water molecules disassociated from the polymeric chains decreased. Based on these results, we postulate that the changes inthe interaction between the thermosensitive polymer and bound water exert a strong influence on its phase transition and/or separation, such asthe cloud point or dehydration behavior.© 2006 Published by Elsevier Inc.

Keywords: Thermoresponsive polymers; N -isopropylacrylamide; Cloud point; Differential scanning calorimetry; Bound water

1. Introduction

Stimuli-responsive polymers abruptly transform their physi-cochemical properties in response to external environmentalchanges such as temperature, light, electric field, pH, or concen-tration of the chemical species. A thermoresponsive polymer,which is a kind of stimuli-responsive polymer and responds totemperature, has been studied by many researchers. In partic-ular, poly(N -isopropylacrylamide) (PNIPAAm) has been usedin various fields, such as nanotechnology or biological sys-tems [1–5]. When the temperature is below the lower criticalsolution temperature (LCST, ∼31 ◦C), PNIPAAm exists in arandom coil and dissolves in water; however, above the LCST,its conformation is transformed into a globule and becomes in-

* Corresponding author. Fax: +81 99 285 7794.E-mail address: aoyagi@eng.kagoshima-u.ac.jp (T. Aoyagi).

0021-9797/$ – see front matter © 2006 Published by Elsevier Inc.doi:10.1016/j.jcis.2006.06.047

soluble [6]. This hydration–dehydration behavior of PNIPAAmin water is reversible. Schild reviewed some physicochemicalstudies concerning the PNIPAAm in order to understand itsphase transition mechanism [7]. Moreover, many researchershave investigated the unique solution properties of PNIPAAmand other thermoresponsive polymers [8–12].

It has been reported that introducing hydrophilic comono-mers into a thermosensitive polymer results in increasing theLCST, while hydrophobic comonomers decrease the LCST [13].Maeda et al. systematically investigated the hydration statesof PNIPAAm and its copolymers as well as other poly(N -alkylacrylamide)s above and below their LCSTs by Fouriertransform infrared (FTIR) spectroscopy [14–17]. In these pa-pers, they examined the influences of ions on the behaviorsof the IR spectra of these polymers, and showed that the IRspectra profiles of these polymer solutions containing saltsresemble those of polymer solutions in the absence of salts,

468 T. Maeda et al. / Journal of Colloid and Interface Science 302 (2006) 467–474

Fig. 1. Chemical structures of (a) CIPAAm, AIPAAm, HIPAAm, and (b) poly(NIPAAm-co-HIPAAm).

even though their LCSTs were changed by adding the salts.In other words, the amide I band or the amide II band didnot shift, which suggested that the added salts do not directlyinteract with the amide group, but change the structure of wa-ter existing around the polymeric chains. Furthermore, it wasdemonstrated that the conformational change of PNIPAAm inwater is governed by the dynamics of water molecules accord-ing to a computer simulation [18]. Based on these results, weconsidered that the thermosensitive behaviors of the polymerswould be closely related to the water molecules around thepolymer chains. To establish the hypothesis, we then tried toanalyze the hydration state of a thermosensitive polymer usingthe novel temperature-responsive polymer, poly(NIPAAm-co-2-hydroxyisopropylacrylamide) (poly(NIPAAm-co-HIPAAm))(Fig. 1b). In our previous research, we hypothesized that the fol-lowing three factors: (1) preserving the continuous and repeatedstructure of the isopropylamide group after copolymerization,(2) random monomer alignment of the polymer chains, and (3)uniformity of the comonomer content in each copolymer chain,are very important for constructing a functional polymer with asensitive temperature response. Therefore, we have newly de-signed three types of monomers, namely, an anionic monomer,2-carboxyisopropylacrylamide (CIPAAm) [19–21], a cationicmonomer, 2-aminoisopropylacrylamide (AIPAAm) [22], and anonionic monomer, HIPAAm (Fig. 1a) [23]. Since the struc-tures of these monomers are very similar to that of NIPAAm,the obtained copolymers by free radical copolymerization withNIPAAm could retain the continuous and repeated structureof the isopropylamide group after copolymerization, and themonomer reactivity ratios were very similar to the values (inpoly(NIPAAm-co-HIPAAm), r1 = 1.08 and r2 = 0.60, deter-mined by the Kelen–Tüdös method, where NIPAAm is (1) andHIPAAm is (2)) [23]. As expected, the obtained copolymersshowed very sensitive thermoresponses in aqueous media eventhough they contained many hydrophilic side chains.

In this study, we focused on water molecules around thepoly(NIPAAm-co-HIPAAm) chains and analyzed the hydra-tion behaviors of the copolymers by differential scanningcalorimetry (DSC). The surrounding water molecules wereclassified into three distinct modes, i.e., free water, boundwater and non-freezing water [24], and the correlation be-tween these three types of water and temperature-responsivebehaviors of the acrylamide-type thermosensitive polymers wasinvestigated. We employed poly(NIPAAm-co-HIPAAm) dueto the following reasons: (1) it is possible to introduce theHIPAAm comonomers into the NIPAAm-based copolymers ashydrophilic parameters without losing their sharp thermore-sponses, (2) because the chemical structures of NIPAAm andHIPAAm very closely resemble each other, the original hydra-tion state of PNIPAAm would not be totally destroyed, and (3)since poly(NIPAAm-co-HIPAAm) is nonionic, we could nottake into account the changes in the water structures by dis-sociation of an ionic moiety, therefore simplifying the system.Based on these important points, the HIPAAm could be verysuitable monomer for investigation of the hydration state of anacrylamide-type thermoresponsive polymer by introducing thehydrophilic comonomer into the NIPAAm-based copolymersas a hydrophilic parameter.

2. Experimental

2.1. Materials

NIPAAm was kindly supplied by Kohjin (Tokyo, Japan) andpurified by recrystallization from benzene–hexane. HIPAAmwas synthesized from D,L-2-amino-1-propanol and acryloylchloride which were purchased from Tokyo Kasei Kogyo(Tokyo, Japan) and used without further purification. Thismonomer synthesis was described in detail in our previous pa-per [23]. Poly(NIPAAm-co-HIPAAm) with various HIPAAm

T. Maeda et al. / Journal of Colloid and Interface Science 302 (2006) 467–474 469

compositions (NH-0, NH-10, NH-30, NH-50, NH-80, NH-100,where NH-X denotes the copolymers containing X mol% ofHIPAAm unit) were prepared by free radical copolymeriza-tion of NIPAAm with HIPAAm, and obtained in high yields.This preparation method was also previously mentioned [23].The chemical structures and the HIPAAm compositions ofthe copolymers were determined using 1H NMR spectroscopy(JEOL JNM-GSX400, 400 MHz spectrometer). The number-average molecular weights (Mn) and molecular weight dis-tributions (Mw/Mn) of the copolymers were determined bygel permeation chromatography (GPC, Jasco LC-2000 Plus,Tokyo, Japan). DMF containing LiBr (10 mM) was used as aneluent, and poly(ethylene glycol) was employed as the stan-dard for calibration. The number-average molecular weightsand molecular weight distributions of the obtained copolymerswere about 2.0–4.0 × 104 and 2.0–3.0, respectively.

2.2. Differential scanning calorimetric measurement ofpoly(NIPAAm-co-HIPAAm) aqueous solution

A differential scanning calorimeter (DSC, DSC 6100, SeikoInstruments, Tokyo, Japan) was used to analyze the hydrationstate of poly(NIPAAm-co-HIPAAm) in water. Copolymers withvarious HIPAAm compositions were dissolved in ultrapure wa-ter and the sample solutions were prepared at concentrations of10, 20, and 30 w/v%. These polymer aqueous solutions werekept at 4 ◦C for 2 days, and about 30 µL of the copolymer solu-tions were then transferred to silver pans. To prevent water fromevaporating during the measurements, these pans were com-pletely sealed. Subsequently, the temperatures of the copolymeraqueous solutions were lowered to −60 ◦C, and these solutionswere frozen. We then carried out the DSC measurements bygradually raising the temperatures of the solutions. The mea-surements were performed between −60 and 120 ◦C at a scan-ning rate of 2.0 ◦C/min while heating. The reproducibility waschecked by running 3 experiments. �Hm (J/g), which is the ap-parent enthalpy of melting of water, was estimated by dividingthe total enthalpy of melting for the copolymer solution (J) bythe mass of water (g).

2.3. Analysis of dehydration behavior ofpoly(NIPAAm-co-HIPAAm) solution with 1H NMRspectroscopy

1H NMR measurements of the copolymers in deuteriumoxide (D2O) were carried out above and below the cloudpoints. The copolymers of various compositions were dissolvedin D2O (Wako Co., Tokyo, Japan, 99%) at a concentrationof 1.0 w/v%. 3-(Trimethylsilyl)propionic acid-d4 sodium salt(Aldrich Chem. Co.) was added to each copolymer solutionas the internal standard. Subsequently, we performed 1H NMRmeasurements of the copolymer solutions at 25 ◦C (below theircloud points) and at 10 ◦C higher than each cloud point of thecopolymers (above their cloud points).

3. Results and discussion

3.1. Thermoresponsive behavior ofpoly(NIPAAm-co-HIPAAm) in water

In our previous research, we demonstrated that poly(NIPA-Am-co-HIPAAm) aqueous solutions showed very sharp ther-mosensitive behaviors, and even the copolymers with a highcontent of hydroxyl groups (50 or 80 mol% for HIPAAmunit) were capable of exhibiting clear and discontinuous trans-mittance changes in aqueous media [23]. NH-100, i.e., theHIPAAm homopolymer, did not show a transmittance changebelow 100 ◦C at ambient pressure, but could cause a turbid-ity change by adding salts. On the basis of these results,poly(NIPAAm-co-HIPAAm) would have a strong intra- or in-termolecular association force. In general, introducing the hy-drophilic comonomers, such as acrylic acid, into temperature-responsive polymers or hydrogels, makes their thermosensitivebehaviors insensitive, and in some cases, cannot cause them[19,25–27]. However, since poly(NIPAAm-co-HIPAAm) is ca-pable of maintaining the sharp temperature-response behav-ior even in the copolymer containing a high content of hy-drophilic comonomers, this characteristic is very useful for thepurpose of analyzing the effects of introducing the hydrophiliccomonomers into the NIPAAm-based copolymers on the hy-dration states and the temperature-responsive behaviors of thecopolymers. Furthermore, the cloud points of poly(NIPAAm-co-HIPAAm) closely depended on the HIPAAm content, thatis, the cloud points increased with increasing the HIPAAm con-tent. These results are summarized in Table 1.

3.2. The correlation between the temperature-responsivebehavior of poly(NIPAAm-co-HIPAAm) and water moleculesaround the copolymer chains

In some polymer aqueous solutions, it has been reportedthat the condition of the water molecules around the poly-meric chains is different from normal one [28,29]. Uedaira andother researchers have described that three kinds of waters, i.e.,free water, bound water, and non-freezing water, would par-ticipate in the hydration of the water-soluble polymers [24,28–30]. Therefore, we also divided water molecules existingin the copolymer solutions into three distinct states (free wa-ter, bound water, and non-freezing water) in imitation of them.Free water is hardly affected by the polymer chains and its be-havior is identical with ordinary water molecules. This water

Table 1The cloud points for 1.0 w/v% aqueous solutions of poly(NIPAAm-co-HIPAAm)

Code Cloud point (◦C)

NH-0 31.6NH-10 36.7NH-30 41.8NH-50 55.0NH-80 80.0NH-100 –

470 T. Maeda et al. / Journal of Colloid and Interface Science 302 (2006) 467–474

Fig. 2. DSC curves for NH-10 aqueous solutions of poly(NIPAAm-co-HIPAAm) with various polymer concentrations. �Hm denotes the apparent en-thalpy of melting of water.

freezes at 0 ◦C under the atmospheric pressure. Bound wa-ter is somewhat influenced by the polymeric chains and isknown to freeze at a lower temperature than the usual freezingpoint [24,31]. Non-freezing water is strongly associated withthe polymer chains and cannot freeze at even −196 ◦C [24].Therefore, the stronger the water molecules interact with thepolymeric chains, the more difficult they tend to crystallize. Inother words, the stronger interaction between water moleculesand the polymer chains results in lowering the melting point (orfreezing point). We examined the relationship between thesethree distinct waters and the thermo-responsive behavior of thenonionic poly(NIPAAm-co-HIPAAm) by differential scanningcalorimetry. DSC measurements were carried out between −60and 120 ◦C during a heating process, and, as described below,two endothermic peaks were observed. The one is an endother-mic peak arising from the melting of water, and the other is dueto dehydration of the polymer chains that occurred during thephase transition and/or separation. The former consists of twocomponents: the one is the endothermic peak at ca. 0 ◦C, whichoriginates in the melting of free water, and the other is the shoul-der peak observed at the lower temperature (∼−10 ◦C), beingdue to melting of the bound water.

Fig. 2 shows DSC thermograms due to the melting of waterin the copolymer aqueous solutions and only water. In only wa-ter (Fig. 2 (a)), a very sharp endothermic peak at ca. 0 ◦C wasobserved, however, in the poly(NIPAAm-co-HIPAAm) aqueoussolutions (Fig. 2 (b–d)), the broader endothermic peaks havingthe shoulder peaks at lower temperature were found. Accord-ingly, it was revealed that the bound water was produced inthe polymer aqueous milieu by adding the copolymers to wa-ter. Moreover, the apparent enthalpy of melting of water, �Hm,decreased by adding the copolymers to water (Fig. 2 (a, b)). We

Fig. 3. DSC curves for 30 w/v% aqueous solutions of poly(NIPAAm-co-HIPAAm) with various compositions. Tm denotes the temperature due to themelting of the bound water.

considered that this result is due to not only the appearance ofthe bound water whose �Hm might be smaller than that of freewater, but also the generation of non-freezing water. It has beenreported that the �Hm of bound water is different from thatof free water [28], and because bound water melts at a lowertemperature than the melting point of free water, the �Hm ofthe bound water seems to be lower than that of free water. Be-sides, since non-freezing water cannot freeze, its �Hm couldbe equal to zero. Therefore, it is assumed that the �Hm of freewater (�Hm,f) is the largest among the three types of waters,and the �Hm of bound water (�Hm,b) is greater than that ofthe non-freezing water (�Hm,n) (that is, �Hm,f > �Hm,b >

�Hm,n = 0). Moreover, �Hm of the copolymer aqueous so-lutions decreased with increasing the polymer concentrationbecause the fractions of bound water and non-freezing waterrelative to the free water could increase by adding the polymersto water (Fig. 2 (b–d)). This is also suggested by the fact thatthe endothermic peaks due to free water decreased, whereas theshoulder peaks resulting from the bound water increased withincreasing the polymer concentration. Also, �Hm measured inonly water (�Hm = 333 J/g, Fig. 2 (a) was consistent with thevalue in the literature (6.008 kJ/mol, which corresponds to ca.333.5 J/g).

Fig. 3 shows DSC thermograms and the �Hm of the copoly-mer aqueous solutions of various HIPAAm compositions. Itshould be noted that the �Hm values were almost same al-though the HIPAAm contents were different. As describedabove, it has been reported that the �Hm,b is different from�Hm,f [28], and estimating the value of �Hm,b is exceed-ingly difficult. In this study, hence, we could not calculate theamount of each water existing in the copolymer solutions. How-

T. Maeda et al. / Journal of Colloid and Interface Science 302 (2006) 467–474 471

Fig. 4. Relationship between Tm and the HIPAAm content for poly(NIPA-Am-co-HIPAAm) aqueous solutions. Polymer concentration: 10 w/v% (!),20 w/v% (P), 30 w/v% (1).

ever, on the basis of these results that the �Hm values werealmost constant, it seems that the quantitative balance of freewater, bound water, and non-freezing water was unchangedin spite of the variation in the HIPAAm composition. Sincethe HIPAAm monomer is hydrophilic, we expected that the�Hm might decrease with increasing the HIPAAm composi-tion due to the change in the bound water and non-freezingwater. Consequently, these data were unexpected results. Inter-estingly however, as seen in Fig. 3, the shoulder peaks assignedto the bound water became broader and shifted to a lower tem-perature as the HIPAAm content increased. We then definedthe Tm, which is the temperature resulting from the melting ofthe bound water as shown in Fig. 3, and plotted it in Fig. 4as a function of the HIPAAm composition. It is realized thatthe Tm closely depends on the HIPAAm content and decreaseswith an increase in the HIPAAm composition. Moreover, it wasfound that the Tm is independent of the polymer concentra-tion, suggesting that the shift in Tm is not due to the freezingpoint depression. As described above, water molecules, whichstronger associate with the polymeric chains, become difficultto crystallize, that is, leading to a lower melting point. There-fore, we consider that the interaction between the bound waterand the copolymer chains is reinforced as the HIPAAm compo-sition increases.

3.3. Dehydration behavior induced in the phase transitionand/or separation for poly(NIPAAm-co-HIPAAm) aqueoussolution

In our previous research, we performed DSC measurementsfor the purpose of investigating the phase transition and/orseparation behavior of poly(NIPAAm-co-HIPAAm) in aque-ous media [23]. From these measurements, it became clearthat although the copolymers showed very sensitive transmit-tance changes in water, the endothermic peaks due to dehy-dration of the polymeric chains that occurred during the phasetransition and/or separation gradually became small with in-creasing the HIPAAm content, and finally could not be ob-served above NH-50. In this study, the polymer concentra-tion dependency for the phase transition and/or separation be-

Fig. 5. �Ht for aqueous solutions of poly(NIPAAm-co-HIPAAm) as a functionof polymer concentration: NH-0 ("), NH-10 (Q), NH-30 (2). �Ht denotes theendothermic enthalpy per mol of monomer unit, which is due to the dehydrationinduced during the phase transition and/or separation.

havior of poly(NIPAAm-co-HIPAAm) was examined. Fig. 5shows the endothermic enthalpy, �Ht, resulting from dehy-dration induced during the phase transition and/or separation,as a function of the polymer concentration. In NH-0–NH-30,no noticeable changes in �Ht were observed even though thepolymer concentration changed, which implied that the dehy-dration profiles of the copolymers were independent of thepolymer concentration. However, �Ht decreased with increas-ing the HIPAAm content, and NH-50 and NH-80 did not showany endothermic peak for all the polymer concentrations. Wehave already demonstrated that poly(NIPAAm-co-HIPAAm)with a high HIPAAm content, such as NH-50, showed a liquid–liquid phase separation accompanied by coacervation [23]. Ithas been reported that the �Ht of temperature-responsive poly-mers causing coacervation in the phase separation are quitesmall [32,33]. Thus, the �Ht of NH-50 and NH-80 were toosmall to be estimated in these measurements. These results in-dicate that dehydration of the polymeric chains becomes insuf-ficient with an increase in the HIPAAm composition. In otherwords, it is considered that the number of water molecules,which are dissociated from the copolymer chains in the phasetransition and/or separation, decreases as the HIPAAm contentincreases.

To further investigate this consideration, we performed 1HNMR measurements of the copolymers in deuterium oxideabove and below their cloud points. By using 1H NMR spec-troscopy, we can obtain information concerning the hydra-tion states of the copolymers. Below their cloud points, sincethe copolymers dissolve in the solvent (D2O), the peaks as-signed to each proton of the copolymer are detected. How-ever above their cloud points, the copolymer chains would bemore or less dehydrated, therefore, the peak intensities of thecopolymers might decrease or the peaks might be undetectable.Fig. 6 shows the 1H NMR spectra of NH-0 and NH-50 inD2O above and below each of their cloud points. Below theircloud points, the peaks assignable to each proton of the copoly-mers were clearly detectable. When the polymer solutions wereheated to 10 ◦C higher than each cloud point of the copoly-mers, the peak intensities were drastically reduced in NH-0

472 T. Maeda et al. / Journal of Colloid and Interface Science 302 (2006) 467–474

Fig. 6. 1H NMR spectra of poly(NIPAAm-co-HIPAAm) in D2O: NH-0 (a) at room temperature (r.t.) and (b) above the cloud point; NH-50 (c) at r.t. and (d) abovethe cloud point. Polymer concentration: 1.0 w/v%.

(Fig. 6 (c)), however, in NH-50, although a certain degree ofdecrease in the peak intensity was observed, it did not dramat-ically decrease (Fig. 6 (d)). Because PNIPAAm gives rise to anabrupt hydration–dehydration change accompanied by a coil-to-globule transition at its LCST, the polymeric chains weredehydrated above the LCST and a large decrease in the peakintensity should occur. On the other hand, in view of the re-sult that the peaks of the copolymer remained up to a point,we postulated that NH-50 does not cause a drastic dehydra-tion at its cloud point. In fact, NH-50 shows coacervation andits dehydration could partially occur. Accordingly, we demon-strated that there is a significant difference in the 1H NMRspectrum between the coil–globule transition and coacerva-tion. Also, these results were in good agreement with those ofthe DSC measurements which suggested that NH-0 showed alarge enthalpy of transition (�Ht), whereas the NH-50 copoly-mer did not exhibit an endothermic peak at its cloud pointfor every polymer concentration. Fig. 7 shows the 1H NMRspectra of poly(NIPAAm-co-HIPAAm) with various HIPAAmcompositions in D2O above their cloud points (10 ◦C higherthan each cloud point of the copolymers). It was revealed thatthe remaining peaks gradually increased with an increase in theHIPAAm content. Consequently, we suggest that dehydrationof the copolymer chains in the phase transition and/or separa-tion would become more difficult to induce with increasing theHIPAAm content, that is, the amount of water molecules dis-associated from the polymeric chains decreases. These resultswere also consistent with the results obtained by the DSC mea-surements.

3.4. The relationship between the cloud point, Tm, �Ht, andthe HIPAAm composition for poly(NIPAAm-co-HIPAAm)aqueous solutions

Based on the results obtained in this study, we derived thecorrelation between the Tm and the thermoresponsive behaviorof poly(NIPAAm-co-HIPAAm) in water. Fig. 8 shows the rela-tionship between Tm, �Ht, and the HIPAAm composition, andFig. 9 shows the relationship between Tm, the cloud points (de-termined by UV–vis measurements), and the HIPAAm content.It was revealed that the Tm and �Ht decreased, whereas thecloud points of the copolymers increased as the HIPAAm com-position increased. Based on these results, we postulated thefollowing: the hydrophilicity of the copolymers increases withincreasing the HIPAAm content, thus reinforcing the interac-tion between the bound water and copolymer chains (that is,Tm decreases). Moreover, the stronger interaction between thebound water and the copolymers results in becoming more dif-ficult for the polymeric chains to cause dehydration in the phasetransition and/or separation. Consequently, the number of wa-ter molecules, which are dissociated from the polymeric chains,decreases (that is, �Ht decreases), and the cloud points of thecopolymer solutions should shift to a higher temperature. It hasbeen demonstrated that PNIPAAm includes water molecules tosome extent even in the globule state, and is not completely de-hydrated [14,34]. Thus, water molecules, such as non-freezingwater, which strongly associate with the polymeric chains, maynot be dissociated and partly remain even during the phasetransition and/or separation. In addition, because free water ishardly influenced by the polymer chains, it could not affect the

T. Maeda et al. / Journal of Colloid and Interface Science 302 (2006) 467–474 473

Fig. 7. 1H NMR spectra of poly(NIPAAm-co-HIPAAm) with a variety of compositions in D2O: (a) NH-0, (b) NH-10, (c) NH-30, (d) NH-50, (e) NH-80 above thecloud points. Polymer concentration: 1.0 w/v%.

Fig. 8. Tm and �Ht for aqueous solutions of poly(NIPAAm-co-HIPAAm) asa function of the HIPAAm content. Polymer concentration: 10 w/v%; Tm (!),�Ht (").

phase transition and/or separation behaviors of the temperature-responsive polymers. Hence, it is presumed that the behavior ofbound water plays an important role in the thermoresponsiveprofiles of the polymers. So far, it has been reported that theLCSTs of thermosensitive copolymers depend on their �Ht,i.e., the LCST increases with decreasing the �Ht [13,15,16]. Inthis study, however, it was first demonstrated that Tm, which isthe temperature resulting from the melting of bound water, af-fected the �Ht and the cloud points of the thermoresponsivecopolymers, therefore, the temperature-responsive behaviors ofthe polymers significantly depend on the bound water.

4. Conclusion

In this study, we analyzed the hydration state of a ther-moresponsive polymer with a novel nonionic temperature-

Fig. 9. Tm and the cloud points for aqueous solutions of poly(NIPAAm-co-HIPAAm) as a function of the HIPAAm content. Polymer concentration:10 w/v% for Tm and 1.0 w/v% for the cloud points; Tm (!), cloud points (2);Tm and the cloud points were determined from the DSC and UV–vis measure-ments, respectively.

sensitive polymer, poly(NIPAAm-co-HIPAAm). Particularly,water molecules existing around the copolymer chains weredetermined and divided into three distinct states, namely, freewater, bound water, and non-freezing water. By adding thecopolymers to water, the shoulder peaks resulting from themelting of the bound water appeared in the DSC thermograms.Also, the Tm, which is the temperature due to the melting ofthe bound water, shifted to lower temperatures with increas-ing the HIPAAm composition. Based on these results, it wasbelieved that the bound water more strongly interacts withthe polymeric chains with an increase in the HIPAAm com-position. Furthermore, the endothermic enthalpy for the phasetransition and/or separation, �Ht, decreased with increasing theHIPAAm composition. These results indicate that the amount

474 T. Maeda et al. / Journal of Colloid and Interface Science 302 (2006) 467–474

of water molecules dissociated from the polymeric chains de-creases with increasing the HIPAAm content. In 1H NMRspectra, the intensities of the peaks arising from each protonof the copolymers with a low HIPAAm content, such as NH-0 and NH-10, were dramatically reduced above their cloudpoints. These results suggest that the polymeric chains broughtabout an abrupt dehydration during the phase transition andthe copolymers were insoluble. On the other hand, the peakintensity was not drastically reduced in the copolymer contain-ing a high content of HIPAAm above each cloud point of thecopolymer solutions, and the remaining peaks progressively in-creased as the HIPAAm composition increased. Consequently,we suggest that the dehydration of the copolymer chains thatoccurred in the phase transition and/or separation becomesincomplete with increasing the HIPAAm content, and theseresults were correlated with the ones obtained from the DSCmeasurements.

We determined the correlation between the Tm, the cloudpoints of the copolymers, �Ht and the HIPAAm content forpoly(NIPAAm-co-HIPAAm) aqueous solutions. With an in-crease in the HIPAAm composition, the Tm and �Ht decreased,and the cloud points shifted to a higher temperature. An in-crease in the hydrophilic comonomer (HIPAAm) content raisesthe hydrophilicity of the copolymer chains, thus, the interac-tion between the bound water and the polymeric chains couldbecome stronger. This occurrence accounts for the fact that thedehydration of the polymeric chains induced during the phasetransition and/or separation becomes more difficult to occur.Therefore, it was demonstrated that the amount of dissociatedwater molecules is reduced and the cloud points of the copoly-mer solutions increase. That is, the relationship between the Tm,the cloud point, �Ht and the HIPAAm composition is deter-mined for poly(NIPAAm-co-HIPAAm), and it becomes clearthat the thermosensitive behavior of a temperature-responsivepolymer is dependent on the bound water.

Acknowledgment

This research was partially supported by the Ministry of Ed-ucation, Science, Sports and Culture, through a Grant-in-Aidfor Scientific Research (B) 15300167, 2003.

References

[1] T. Okano, N. Yamada, M. Okuhara, H. Sakai, Y. Sakurai, Biomaterials 16(1995) 297.

[2] J. Li, X. Hong, Y. Liu, D. Li, Y. Wang, J. Li, Y. Bai, T. Li, Adv. Mater. 17(2005) 163.

[3] K. Kataoka, H. Miyazaki, M. Bunya, T. Okano, Y. Sakurai, J. Am. Chem.Soc. 120 (1998) 12694.

[4] P.S. Stayton, T. Shimoboji, C. Long, A. Chilkoti, G. Chen, J.M. Harris,A.S. Hoffman, Nature 378 (1995) 472.

[5] H. Tsutsui, M. Mikami, R. Akashi, Adv. Mater. 16 (2004) 1925.[6] S. Fujishige, K. Kubota, I. Ando, J. Phys. Chem. 93 (1989) 3311.[7] H.G. Schild, Prog. Polym. Sci. 17 (1992) 163.[8] C. Wu, S. Zhou, Macromolecules 28 (1995) 5388.[9] L.-T. Lee, B. Cabane, Macromolecules 30 (1997) 6559.

[10] S. Kunugi, Y. Yamazaki, K. Takano, N. Tanaka, M. Akashi, Langmuir 15(1999) 4056.

[11] D. Nakayama, Y. Nishio, M. Watanabe, Langmuir 19 (2003) 8542.[12] I.C. Barker, J.M.G. Cowie, T.N. Huckerby, D.A. Shaw, I. Soutar, L. Swan-

son, Macromolecules 36 (2003) 7765.[13] H. Feil, Y.H. Bae, J. Feijen, S.W. Kim, Macromolecules 26 (1993) 2496.[14] Y. Maeda, T. Higuchi, I. Ikeda, Langmuir 16 (2000) 7503.[15] Y. Maeda, T. Nakamura, I. Ikeda, Macromolecules 34 (2001) 1391.[16] Y. Maeda, T. Higuchi, I. Ikeda, Langmuir 17 (2001) 7535.[17] Y. Maeda, T. Nakamura, I. Ikeda, Macromolecules 35 (2002) 10172.[18] P.R. ten Wolde, D. Chandler, Proc. Natl. Acad. Sci. USA 99 (2002) 6539.[19] T. Aoyagi, M. Ebara, K. Sakai, Y. Sakurai, T. Okano, J. Biomater. Sci.

Polym. Ed. 11 (2000) 101.[20] M. Ebara, T. Aoyagi, K. Sakai, T. Okano, Macromolecules 33 (2000)

8312.[21] M. Ebara, T. Aoyagi, K. Sakai, T. Okano, J. Polym. Sci. Part A Polym.

Chem. 39 (2001) 335.[22] T. Yoshida, T. Aoyagi, E. Kokufuta, T. Okano, J. Polym. Sci. Part A

Polym. Chem. 41 (2003) 779.[23] T. Maeda, T. Kanda, Y. Yonekura, K. Yamamoto, T. Aoyagi, Biomacro-

molecules 7 (2006) 545.[24] H. Uedaira, Membrane 8 (1983) 351.[25] S. Beltran, J.P. Baker, H.H. Hooper, H.W. Blanch, J.M. Prausnitz, Macro-

molecules 24 (1991) 549.[26] A. Gutowska, Y.H. Bae, J. Feijan, S.W. Kim, J. Controlled Rel. 22 (1992)

95.[27] X. Yin, H.D.H. Stöver, Macromolecules 36 (2003) 9817.[28] H. Hatakeyama, T. Hatakeyama, Thermochim. Acta 308 (1998) 3.[29] K. Nakamura, Y. Minagawa, T. Hatakeyama, H. Hatakeyama, Ther-

mochim. Acta 416 (2004) 135.[30] C. Alvarez-Lorenzo, J.L. Gómez-Amoza, R. Martínez-Pacheco, C. Souto,

A. Concheiro, Eur. J. Pharm. Biopharm. 50 (2000) 307.[31] A.K. Lele, M.M. Hirve, M.V. Badiger, R.A. Mashelkar, Macromole-

cules 30 (1997) 157.[32] H. Miyazaki, K. Kataoka, Polymer 37 (1996) 681.[33] S. Sugihara, S. Kanaoka, S. Aoshima, Macromolecules 37 (2004) 1711.[34] X. Wang, X. Qiu, C. Wu, Macromolecules 31 (1998) 2972.