Change in hydrogen bonding structures of a hydrogel with dehydration

Chemical Physics Letters 670 (2017) 84–88

Contents lists available at ScienceDirect

Chemical Physics Letters journal homepage: www.elsevier.com/locate/cplett

Research paper

Change in hydrogen bonding structures of a hydrogel with dehydration Ryo Naohara, Kentaro Narita, Tomoko Ikeda-Fukazawa ⇑ Department of Applied Chemistry, Meiji University, Kawasaki 214-8571, Japan

a r t i c l e

i n f o

Article history: Received 4 October 2016 Revised 30 December 2016 In final form 4 January 2017 Available online 5 January 2017 Keywords: Hydrogel Water X-ray diffraction

a b s t r a c t To investigate the mechanisms of structural changes in polymer network and water during dehydration, X-ray diffraction of poly-N,N-dimethylacrylamide (PDMAA) hydrogels was measured. The variation process in the individual structures of water and PDMAA were analyzed by decomposition of the diffraction patterns to separate the respective contributions. The results show that the short-range structures of PDMAA expand during dehydration, whereas the network structure as a whole shrinks. The average length of the hydrogen bonds between water molecules increases with the process. The present results provide a direct evidence of the structural changes of water and polymer in the hydrogel during dehydration. Ó 2017 Elsevier B.V. All rights reserved.

1. Introduction Hydrogels are unique materials consisting of large amounts of water within three-dimensional polymer networks. Because of their characteristic properties, hydrogels have been widely used as biomaterials, medicinal products, civil engineering products, and so on [1–3]. Furthermore, hydrogels are expected as functional materials such as drug delivery systems and scaffolding of cellular cultures [4,5]. The structures of the polymer network and the included water are important factors governing the chemical and physical properties of hydrogel materials. The water in hydrogels has been classified into three types: bound, intermediate, and free water [6]. Bound water forms hydrogen bonds with the hydrophilic groups in the polymer or interacts strongly with the polymer chains. Free water has a structure similar to that of bulk water and is ‘free’ from interactions with the surrounding polymer chains. Intermediate water exists between the bound and free water, and interacts weakly with the polymer. As a result of the differences in the structures of these three types of water, the properties of hydrogels are affected by the relative amounts of the three types of water. Because the relative amounts of the three types of water depend on water content [7], it is essential to understand the structural changes that the hydrogel undergoes during dehydration. The dehydration processes of hydrogels have been studied using various methods. Koshoubu et al. [8] analyzed mechanical properties of white gel in heat-treated eggs, and showed that the dehydration process of hydrogel can be classified into three stages ⇑ Corresponding author. E-mail address: [email protected] (T. Ikeda-Fukazawa). http://dx.doi.org/10.1016/j.cplett.2017.01.006 0009-2614/Ó 2017 Elsevier B.V. All rights reserved.

based on the differences in dehydration rates in the three types of water [9]. Sekine and Ikeda-Fukazawa [10] measured Raman spectra of poly-N,N-dimethylacrylamide (PDMAA) hydrogels and analyzed the structural changes in water and the polymer networks during dehydration. From these results, the following stages were proposed for the dehydration process of PDMAA hydrogels. In the beginning of the dehydration process (stage-I), free water mainly evaporates and the polymer network shrinks with the evaporation. Intermediate water begins to evaporate during the middle stage (stage-II). Due to a rapid reduction of water during stage-II, the polymer network collapses and undergoes a glass transition at the end of this stage. In stage-III, the remaining bound water evaporates slowly from the glassy dried gel. Using differential scanning calorimetry (DSC), Ikeda-Fukazawa et al. [11] showed that the boundaries between stages I–II and II–III for PDMAA hydrogel are 65 and 40 wt%, respectively. Although these studies indicated the macroscopic dehydration process of hydrogel, the microscopic process of structural changes of water and polymer network is less conclusive. In order to investigate the mechanism of microscopic structural changes of water and polymer network in a hydrogel during dehydration, we measured and analyzed X-ray diffraction (XRD) of a PDMAA hydrogel. XRD is a useful tool for the analyses of shortrange structure in liquid and amorphous materials, which have no long-range order structures [12–14]. Several studies of hydrogels using XRD have been previously reported. Ricciardi et al. [15] measured XRD of poly (vinyl alcohol) (PVA) hydrogels obtained using the freezing–thawing method. They showed that both the degree of crystallinity and the size of crystals in PVA hydrogels increase with an increasing number of the freezing– thawing cycles. Using XRD and small-angle neutron diffraction

R. Naohara et al. / Chemical Physics Letters 670 (2017) 84–88

85

methods, Bosio et al. [16] analyzed the pore size of poly (2hydroxyethylmethacrylate) hydrogel. The results showed that the water in the pores or micro-channels of the hydrogel exists in the unfreezable state due to interaction with three-dimensional polymer networks. These studies suggest that XRD is an efficient technique for the investigation of hydrogel structures. 2. Experimental Poly-N,N-dimethylacrylamide (PDMAA) hydrogels were synthesized using the following procedure. N,N0 -dimethylacrylamide (DMAA), N,N-methylenebis(acrylamide) (BIS), potassium persulfate (KPS), and N,N,N0 ,N0 -tetramethylenediamine (TEMED) were used as the monomer, cross-linker, initiator, and catalyst, respectively. Distilled water was purged with N2 gas for 3 h prior to the sample preparation in order to remove the dissolved O2. An aqueous solution consisting of water (57 mL), BIS (0.276 g), and DMAA (5.94 g) was prepared. TEMED (48 lL) was then added to the solution. Finally, an aqueous solution of KPS (0.03 g) in water (3.0 mL) was added. The solution (8.0 mL) was dripped onto a silicon wafer with a diameter of 50 mm and a thickness of 1 mm. A radical polymerization was then preceded for 20 h at room temperature. The formed hydrogels, with a thickness of 3 mm, were soaked in pure water for 72 h in order to remove unreacted compounds. Fig. 1 shows the chemical structure of PDMAA. For hydrogel dehydration, the samples were stored in a desiccator at 298 ± 3 K and at a humidity of 50 ± 10% for 240 h. During the dehydration process, XRD of each sample was measured. The water content (W) of the dehydrated sample was determined by

W ¼ ½ðMgel  M fd Þ=M gel   100;

ð1Þ

where Mgel and Mfd are the mass of the sample and the fully dehydrated sample, respectively. The W values of the examined 29 samples were in range of 90.0–7.76 wt%. Diffraction patterns of the prepared hydrogels were measured using an X-ray diffractometer (Rigaku RINT-Ultima III). Using Cu Ka radiation of 1.5418 Å in wavelength, the diffraction patterns were measured in 2h range of 3–70 deg. The scan speed and step were 1.0 deg min1 and 0.01 deg, respectively. All measurements were carried out at room temperature. 3. Results and discussion Fig. 2 shows the XRD patterns of PDMAA hydrogels with W = 90.0, 53.0, and 7.76 wt%. The diffraction patterns have broad liquid-like features, and the band feature changes with water content. Three broad peaks are observed in the patterns for the hydrogels with water content of 53.0 and 7.76 wt% in the examined angle region. In the case of the hydrogel with W = 90.0 wt%, two well-defined peaks and the foot for the lowest-angle peak were observed. As shown in Fig. 2, the three broad peaks shift with water content. To analyze the dependence of the observed peak positions on water content, the XRD profiles were decomposed into three peaks

Fig. 1. Chemical structure of PDMAA.

Fig. 2. X-ray diffraction patterns of PDMAA hydrogels with water content W = (a) 90.0 wt%, (b) 53.0 wt%, and (c) 7.76 wt% at 298 K. The solid and dotted lines show the intensities observed experimentally and fitted curves determined using Gaussian functions, respectively.

using Gaussian functions. The dotted lines in Fig. 2 show the fitted curves. For instance, the angles of the three fitted peaks for the hydrogel with W = 7.76 wt% are 11.10, 20.59, and 36.69 deg. Using the Cu Ka wavelength of 1.5418 Å, these 2h values are converted into d-spacing values of 7.71, 4.31, and 2.45 Å respectively. Hereafter, these three peaks are referred to as p1, p2, and p3, respectively. Fig. 3 shows the dependence of the peak positions on the water content for p1, p2, and p3. The d-spacing of p1 decreases with dehydration, whereas those of p2 and p3 increase. These shifts can be attributed to the structural changes in the polymer networks and water within the hydrogel during dehydration. These results imply that short-range structures expand due to dehydration, while a long-range structures shrink. The p1 peak corresponds to a longrange structure between polymer chains, because the d-spacing of the peak is larger than 20 Å in swollen states. In addition to the three broad peaks observed at 11.10, 20.59, and 36.69 deg (i.e., p1, p2, and p3), the diffraction pattern of hydrogel with W = 7.76 wt% has three sharp peaks at 11.47, 30.72, and 54.34 deg. The arrows in Fig. 2(c) show the positions of the three sharp peaks. The peak at 30.72 deg. is assigned to the silicon wafer substrate. Because the thickness of the sample decreases as the dehydration progresses, the silicon peak was detected for the dried samples. The appearance of two peaks observed at 11.47 and

86

R. Naohara et al. / Chemical Physics Letters 670 (2017) 84–88

Fig. 3. Dependence of position of (a) p1, (b) p2 and (c) p3 peaks in d-spacing on water content.

54.34 deg. suggests that a part of PDMAA or water is transformed into an ordered phase with a glass transition. The observed three broad peaks (i.e., p1, p2, and p3) includes contributions of both water and polymer matrix. To investigate the variation process in the distinct structures for the polymer networks and water, we attempted to decompose the diffraction patterns into the respective contributions from PDMAA and water. Fig. 4a) and (b) shows the XRD patterns of pure DMAA and water, respectively. As shown by the dotted and dashed lines in Fig. 4, the diffraction patterns of pure DMAA and water can be decomposed using Gaussian functions into three and two peaks, respectively. These results indicate that decomposition of the diffraction pattern of a hydrogel composed of water and PDMAA is expected to give rise to five peaks. For instance, Fig. 4(c) shows the result of the XRD pattern decomposition for the PDMAA hydrogel with W = 53.0 wt%. The three dotted curves show the contribution form polymer matrix and the two dashed curves are the peaks of water. From this result, p1 and p3 are assigned to PDMAA and water structures, respectively. The p2 peak includes contributions of both water and PDMAA. The fitting analyses were performed using the parameters determined from the decompositions for the pure DMAA and water as the initial input. Fig. 5 shows the decomposed diffraction patterns of PDMAA and water in PDMAA hydrogels with W = 90.0, 53.0, and 7.76 wt%. The patterns were obtained by summing the three fitted curves for PDMAA and the two curves for water, respectively. As shown in Fig. 5, the features of the diffraction patterns of water in the hydrogels are similar to the pattern of pure water (i.e., Fig. 4(a)),

Fig. 4. X-ray diffraction pattern of (a) pure DMAA, (b) pure water, and (c) PDMAA hydrogel with W = 53.0 wt%. The solid, dotted, and dashed lines show the observed intensities, and the curves fitted for the peaks assigned from DMAA and water, respectively.

although the peak positions shift slightly with dehydration. The relative intensity of the pattern determined for water is stronger than that of PDMAA at W = 90.0 wt%, and decreases with dehydration. Fig. 6 shows the dependence on the water content of the positions of the three peaks in the decomposed diffraction patterns of PDMAA in hydrogels. The d-spacing positions of two peaks, referred to as Dp2 and Dp3, are determined as 4.47, and 2.74 Å at W = 90.0 wt%. Although peak Dp1 is not observed in the swollen state, it appears at 32.58 Å in d-spacing at W = 79.7 wt%. The variation process in the peak positions can be classified into three stages: stage-I (W > 65 wt%), stage-II (40 wt% < W < 65 wt%), and stage-III (W < 40 wt%). In stage-I, The d-spacing position of Dp1 significantly decreases with dehydration, whereas those for Dp2 and Dp3 distributes around constants, which are close to the values of pure DMAA (i.e., the dotted line in Fig. 6). The Dp1 value decreases gradually with dehydration in stage-II and III, approaching the value of pure DMAA. Dp2 rapidly increases at the beginning of stage-II, and then gradually decreases. For Dp3, a significant increase is observed at the beginning of stage-III. Dp1 is assigned as the average distance between polymer chains, because the value approaches that of pure DMAA during dehydration. The value of Dp1 for pure DMAA (i.e., 5.25 Å) is close to the length of the polymer side chain (4.43 Å). In the swollen state at W > 65 wt%, the average interval between the polymer chains is around 30 Å. This interval decreases with dehydration,

R. Naohara et al. / Chemical Physics Letters 670 (2017) 84–88

Fig. 5. Decomposed diffraction patterns of PDMAA and water in PDMAA hydrogels with W = (a) 90.0 wt%, (b) 53.0 wt%, and (c) 7.76 wt% obtained using the results of fitting analyses. The dotted and dashed lines show the sum of the three curves fitted from the PDMAA peaks and of the two curves fitted from the water peaks, respectively.

and reaches a value close to that of DMAA in the dried state. Using the peak position of pure DMAA (4.32 Å), Dp2 can be assigned to the interatomic distance between the oxygen and the carbon atoms in the hydrophobic region (CH2, CH3). For Dp3, the peak position of 2.83 Å in pure DMAA is close to the interatomic distance between nitrogen and oxygen in the C@O group. The Dp2 and Dp3 values increase as the water content decreases because the hydrophobic regions are released from repulsive interactions with surrounding water molecules. The closed values of Dp2 and Dp2 with those of pure DMAA in the swollen state are probably due to the balance between the effects of attractive and repulsive interactions. The covalent bonds in the hydrophilic groups (i.e., C@O and CAN bonds) are elongated due to the hydrogenbonding interactions with water, and the bonds in the hydrophobic region (i.e., CAH bonds) are shorted as a result of the repulsion experienced from the surrounding water. Fig. 7 shows the dependence on water content of the positions of the two peaks associated with water in the hydrogels, referred to as Wp1 and Wp2. At water contents above 65 wt%, the positions of these peaks are close to the peak values of pure water (i.e., the dotted lines). At W < 65 wt%, the their values increase as the water content decreases. Wp1 and Wp2 are assigned to the distance between oxygen atoms in water molecules with and without formed hydrogen bonds, respectively. The observed increases in

87

Fig. 6. Dependence of position of (a) Dp1, (b) Dp2 and (c) Dp3 peak in d-spacing on water content. The dotted lines show the peak positions of pure DMAA.

Fig. 7. Dependence of position of (a) Wp1 and (b) Wp2 peak in d-spacing on water content. The dotted lines show the peak positions for pure water.

88

R. Naohara et al. / Chemical Physics Letters 670 (2017) 84–88

Fig. 8. Relationship between number of water molecules per unit structure of DMAA and hydrogel water content.

Wp1 and Wp2 are attributed to the decrease in the average strength of the hydrogen bonds as a result of the changes in the relative amount of the three types of water in the hydrogel during dehydration. As shown in Fig. 8, the number of water molecules per DMAA unit (i.e., C5H9NO) decreases as the water content decreases. Because the oxygen and the nitrogen atoms in DMAA can form two and one hydrogen bonds, respectively, three water molecules maximally form hydrogen bonds with a single DMAA unit. As shown in Fig. 8, four water molecules exist around each DMAA unit as bound water at W = 40 wt%, and three of them maximally form hydrogen bonds with a DMAA unit at any one time. Thus, the increase in the length of the hydrogen bonds which is observed in Fig. 7, results from the increase in the rate of hydrogen bonding between the water molecules and the polymer relative to the rate that between water molecules only. The result indicates that the hydrogen bonds between water and polymer are weaker than that between water molecules. 4. Conclusions The variation process in the structures of polymer network and water in PDMAA hydrogels during dehydration was analyzed using XRD. The significant structural changes were observed at water

contents around 65 and 40 wt%. These results show that the short-range structures expand during dehydration, whereas the long-range network structure shrinks. These structural changes are caused by changes in the relative amounts of the three types of water within the hydrogel. At high water content, the covalent bonds within the hydrophilic groups of the polymer are elongated as a result of their hydrogen-bonding interactions with water, and those of the hydrophobic group are shortened due to repulsive interactions with the surrounding water. Because the repulsive forces diminish with dehydration, the short-range structure of water and the polymer expand during this process. The present results provide a direct evidence of the structural changes of water and the polymer network within PDMAA hydrogels during dehydration. These outcomes also demonstrate that XRD is an efficient tool for the structural analysis of hydrogel materials. Neutron diffraction method is also an efficient technique to investigate the short range structure of hydrogel including hydrogen, because neutron is sensitive to hydrogen atoms in comparison with XRD. Acknowledgments This study was supported financially by JSPS KAKENHI (Grant numbers 25108004 and 26400522). References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16]

N. Peppas, R.E. Benner, Biomaterials 1 (1980) 158. E. Chiellini, A. Corti, S.D. Antone, R. Solaro, Prog. Polym. Sci. 28 (2003) 963. C.M. Cheng, P.R. Ledue, Mol. Biosyst. 2 (2006) 299. G. Otting, E. Liepinsh, K. Wuthrich, Science 254 (1991) 974. M. Nakasako, J. Mol. Biol. 289 (1999) 547. M.S. Jhon, J.D. Andrade, J. Biomed. Mater. Res. 7 (1973) 509. H.B. Lee, M.S. Jhon, J.D. Andrade, J. Colloid Interface Sci. 51 (1975) 225. N. Koshoubu, H. Kanaya, K. Hara, S. Taki, E. Takushi, K. Matsushige, Jpn. J. Appl. Phys. 32 (1993) 4038. K. Kudo, J. Ishida, G. Syuu, Y. Sekine, T. Ikeda-Fukazawa, J. Chem. Phys. 140 (2014) 044909. Y. Sekine, T. Ikeda-Fukazawa, J. Chem. Phys. 130 (2009) 034501. T. Ikeda-Fukazawa, N. Ikeda, M. Tabata, M. Hattori, M. Aizawa, S. Yunoki, Y. Sekine, J. Polym. Sci. Part B: Polym. Phys. 1251 (2013) 1017. A.H. Narten, H. Levy, J. Chem. Phys. 55 (1971) 2263. A.H. Narten, J. Chem. Phys. 70 (1979) 299. A.H. Narten, A. Habenschuss, J. Chem. Phys. 80 (1984) 3387. R. Ricciardi, F. Auriemma, C.D. Rosa, F. Laupretre, Macromolecules 37 (2004) 1921. L. Bosio, G.P. Johari, M. Oumezzine, J. Teixeira, Chem. Phys. Lett. 188 (1992) 113.