Phase transition in hydrogels of thermoresponsive semi-interpenetrating and interpenetrating networks of poly(N,N-diethylacrylamide) and polyacrylamide

European Polymer Journal 85 (2016) 1–13

Contents lists available at ScienceDirect

European Polymer Journal journal homepage: www.elsevier.com/locate/europolj

Phase transition in hydrogels of thermoresponsive semi-interpenetrating and interpenetrating networks of poly(N,N-diethylacrylamide) and polyacrylamide Lenka Hanyková a,⇑, Jirˇí Speˇvácˇek b,⇑, Marek Radecki a, Alexander Zhigunov b, Hana Kourˇilová a, Zdenˇka Sedláková b a b

Faculty of Mathematics and Physics, Charles University, V Holešovicˇkách 2, 180 00 Prague 8, Czech Republic Institute of Macromolecular Chemistry, Academy of Sciences of the Czech Republic, Heyrovsky Sq. 2, 162 06 Prague 6, Czech Republic

a r t i c l e

i n f o

Article history: Received 19 August 2016 Received in revised form 22 September 2016 Accepted 6 October 2016 Available online 8 October 2016 Keywords: Temperature induced volume phase transition Semi-interpenetrating polymer network Interpenetrating polymer network Poly(N,N-diethylacrylamide) Polyacrylamide Hydrogel NMR

a b s t r a c t Temperature-induced phase transition in hydrogels of semi-interpenetrating (SIPNs) and interpenetrating (IPNs) polymer networks of thermoresponsive poly(N,N-diethylacrylamide) (PDEAAm) and hydrophilic polyacrylamide (PAAm) was studied by a combination of 1H NMR spectroscopy, small-angle neutron scattering (SANS) and DSC. The effect of various network composition and reverse sequence in the preparation procedure of the network on the volume phase transition was examined. Transition temperature linearly increased with the content of PAAm component in SIPNs and IPNs. SANS results revealed in investigated SIPN and IPN hydrogels two types of regions (aggregates), small and large. At temperature above the phase transition the small and large regions are related to compact multi-chain aggregates with smooth and fuzzy surface, respectively. From NMR results it follows that the individual polymer groups participated in the transition to various extent dependently on the type of hydrogel preparation. We found that parameters of the phase transition are tunable not only by composition of the SIPNs and IPNs but also by the preparation process. A certain portion of water bound in globular structures and a slow exchange regime between ‘‘bound” and ‘‘free” water were established by NMR methods for all investigated SIPNs and IPNs hydrogels. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction It is well known that some N-substituted acrylamide-based polymers exhibit in aqueous solutions a lower critical solution temperature (LCST); they are soluble at lower temperatures when adopting a coil conformation but above the LCST they become insoluble and their conformation changes to a globular one. This phenomenon is analogous to the volume phase transition phenomenon in crosslinked hydrogels [1,2]. At temperatures below volume phase transition temperature (VPTT) hydrogels absorb water to reach the swollen state and above VPTT they shrink and partly release their water. On the molecular level, both phase separation in solutions and volume phase transition (collapse) in crosslinked hydrogels are assumed to be a macroscopic manifestation of a coil–globule transition, as was shown for poly(N-isopropylacrylamide) (PNIPAm, VPTT
307 K) in water by light scattering [3]. ⇑ Corresponding authors. E-mail addresses: [email protected] (L. Hanyková), [email protected] (J. Speˇvácˇek). http://dx.doi.org/10.1016/j.eurpolymj.2016.10.010 0014-3057/Ó 2016 Elsevier Ltd. All rights reserved.

2

L. Hanyková et al. / European Polymer Journal 85 (2016) 1–13

Due to attractive stimuli-responsive properties of smart hydrogels, several potential applications in drug delivery systems [4], bioseparation [5], artificial muscles [6], microfluidics [7], tissue engineering [8], micro/nanoactuators [9], etc., were suggested. However, in connection with application of PNIPAm some concerns appeared about potential release of toxic low-molecular-weight amines due to hydrolysis and other thermoresponsive polymers have been proposed as alternatives for bioapplications [10]. Poly(N,N-diethylacrylamide) (PDEAAm), with VPTT values
304–307 K [11–13] and the capacity to store and release active agents, is attractive candidate for biomedical applications, particularly drug delivery. The cytotoxicity of PDEAAm hydrogels is less pronounced than for PNIPAm ones, though it is rather low in both cases [14]. PDEAAm is often used for comparison with PNIPAm and it has been found that the swelling ratio of PDEAAm hydrogels below VPTT is much lower than that of PNIPAm hydrogels which is attributed to the different chemical structure of monomeric side chains [15,16]. Besides investigation of the neat PDEAAm hydrogels, copolymerization of DEAAm with a hydrophilic (hydrophobic) monomer or introducing another polymer into PDEAAm hydrogel to form interpenetrating polymer network (IPN) were also used to control phase transition behavior. The effect of copolymerization of DEAAm with various hydrophilic comonomers on the swelling properties and parameters of the phase transition were studied using nuclear magnetic resonance (NMR), Fourier transform infrared spectroscopy (FTIR) and modulated differential scanning calorimetry (DSC) [17]. The crosslinking density and the nature of the incorporated hydrophilic component were found to impact hydrogel structure, mechanical properties and swelling kinetics. Temperature-dependent FTIR was employed to study the phase transition process in random copolymers and microgels based on NIPAm and DEAAm and the transition mechanism by the difference of various hydrogen bonds was explained [18]. The thermo- and pH-sensitive behavior of semi-interpenetrating (SIPN) hydrogels based on poly-((2-dimethylamino)ethyl methacrylate) and PDEAAm were investigated for various compositions [19]. In recent years, our research group has been paid attention to investigation of temperature-induced phase transition in hydrogels of IPNs consisted of thermoresponsive PNIPAm, poly(N-isopropylmethacrylamide) (PNIPMAm) and poly(N-vinylcaprolactam) (PVCL), and hydrophilic polyacrylamide (PAAm) [20–23]. The behavior of the hydrogels (one or two transitions) was found to depend on the ratio of the two IPN components. Separate transitions for the two components were revealed in hydrogels of IPNs containing around 50 mol% of PNIPAm monomer units. A certain portion of spatially restricted bound water was established from NMR relaxation measurements at temperature above the phase transition. In the present work we combined NMR spectroscopy with swelling measurements, DSC and small angle neutron scattering (SANS) to characterize the volume phase transition in SIPNs and IPNs consisting of temperature-sensitive PDEAAm and hydrophilic PAAm. Besides the effect of network composition on the volume phase transition we investigated SIPNs and IPNs with a reversed sequence of the two components in the preparation procedure, i.e., SIPNs of crosslinked PDEAAm and linear PAAm and inversely SIPNs of crosslinked PAAm and linear PDEAAm, and both types of IPNs PDEAAm/PAAm and PAAm/ PDEAAm, as there are virtually no studies of this aspect in the literature. The behavior of prepared networks was compared with aqueous solution of linear PDEAAm and neat hydrogel of crosslinked PDEAAm. 2. Experimental 2.1. Synthesis of homopolymers and SIPNs and IPNs PDEAAm and PAAm homopolymers were prepared by free radical polymerization of DEAAm (Sigma-Aldrich) and AAm (Fluka) monomers in ethanol at 343 K for 24 h using 2,20 -azobis(2-methylpropionitrile) (ABIN, Fluka) as initiator. For the preparation of PDEAAm/PAAm (and PAAm/PDEAAm) SIPNs of various composition DEAAm (or AAm) monomer and crosslinker, N,N0 -methylene bisacrylamide (MBAAm, Fluka), were dissolved in aqueous solution of PAAm (or PDEAAm). Hydrogels were prepared using the redox initiating system ammonium persulfate-N,N,N0 ,N0 -tetramethylenediamine (TEMED). In the synthesis of IPNs PDEAAm/PAAm the PDEAAm network was prepared first in water at 277 K using the redox initiating system [(NH4)2S2O8-TEMED] and crosslinking agent MBAAm. Subsequently, the first network was swollen in aqueous solution of AAm monomer, redox initiating system and MBAAm, polymerization was then carried out at room temperature. The IPNs PAAm/PDEAAm were prepared in a similar way but the PAAm network was prepared as the first one. 2.2. Methods 1 H NMR measurements were performed with a Bruker Avance 500 liquid-state spectrometer operating at 500.1 MHz. Typical conditions were as follows: p/2 pulse width of 12.5 ls, relaxation delay of 20 s, spectral width of 5 kHz, acquisition time of 1.64 s, and 16 scans. The integrated intensities were determined by spectrometer integration software with an accuracy of ±1%. Deconvolution was used to determine integrated intensities of two HDO signals. The 1H spin-spin relaxation times T2 of HDO were measured using the same instrument and the CPMG [24] pulse sequence 90°x  (td  180°y  td)nacquisition with td = 0.5 ms, relaxation delay 100 s and 8 scans. The total time for T2 relaxation was an array of
35 values. The relative error for T2 values did not exceed ±5%. All obtained T2 relaxation curves had the monoexponential character. Constant temperature within ±0.2 K was maintained in all measurements using a BVT 3000 temperature unit. The samples were always kept at the experimental temperature for 15 min before the measurement. DSC measurements were carried out using a Pyris 1 differential scanning calorimeter (PerkinElmer). The measurements were done with a heating rate of 5 K/min in the temperature range from 295 to 335 K.

3

L. Hanyková et al. / European Polymer Journal 85 (2016) 1–13

Small angle neutron scattering (SANS) experiments were performed with YuMO-TOF (time-of-flight) spectrometer on the high-flux pulsed reactor IBR-2 at the Frank Laboratory of Neutron Physics, Joint Institute for Nuclear Research (JINR), Dubna, Russia. Two detectors were used to cover larger q-range. The distance from sample to detectors were 5.28 m and 12.965 m for the first and second detector, respectively. The obtained spectra were normalized and absolute values of the neutron scattering intensity were obtained using a vanadium scatterer by the standard procedure with SAS software [25]. The beam was collimated to a diameter of 10 mm on the sample. The measurements were done at two temperatures, namely at 288 K (temperature below the VPTT) and subsequently at 328 K (temperature above the VPTT). Temperature control and stabilization were made by a special thermal box connected to a Lauda computer-controlled thermostat. Exposure time for each sample at both temperatures was 10 min. The fitting procedure was done using SASFit software [26]. 3. Results and discussion 3.1. Characterization of SIPNs and IPNs hydrogel samples The real ratios of the two components in samples of SIPNs and IPNs were determined from integrated intensities of 1H NMR signals C and D corresponding to the main chain CH protons of DEAAm and AAm units, respectively (cf. Fig. 3 and Scheme 1). The molar ratio of respective monomer units of investigated networks determined in this way are shown in Table 1. The samples D sol and D in Table 1 denotes 5 wt.% D2O solution of linear PDEAAm and the hydrogel of PDEAAm network. The equilibrium swelling ratio was determined at 300 K (below VPTT) after 3 days as the ratio (mT  md)/md, where mT and md denote the weight at temperature 300 K and the weight of dried sample, respectively. The values are given in Table 2. It should be emphasized that with the exception of IPNs A/D 20/80 and 30/70, all hydrogels of SIPNs and IPNs show lower swelling ratios in comparison with hydrogel of the neat PDEAAm. Relatively high swelling ratio observed for IPNs A/D is obviously a consequence of preparation procedure used for these samples when hydrogel of neat hydrophilic PAAm with high swelling ratio is prepared as the first component and subsequently immersed into solution od DEAAm monomer.

Scheme 1. Structural formulae of PDEAAm and PAAm monomer units.

Table 1 Compositions of SIPN and IPN hydrogels. Sample description

Sample designation

Composition Monomer ratio in the synthesis

Molar ratio determined from 1H NMR spectra

PDEAAm solution

D sol

100/0

100/0

PDEAAm hydrogel

D

100/0

100/0

SIPN of crosslinked PDEAAm and linear PAAm

SIPN D/A 77/23 SIPN D/A 64/36 SIPN D/A 60/40

80/20 70/30 60/40

77/23 64/36 60/40

SIPN of crosslinked PAAm and linear PDEAAm

SIPN SIPN SIPN SIPN

20/80 30/70 60/40 70/30

30/70 32/68 43/57 56/44

IPN of PDEAAm (first component) and PAAm

IPN D/A 89/11 IPN D/A 85/15 IPN D/A 64/36

70/30 60/40 50/50

89/11 85/15 64/36

IPN of PAAm (first component) and PDEAAm

IPN A/D 15/85 IPN A/D 20/80 IPN A/D 30/70

20/80 30/70 40/60

15/85 20/80 30/70

A/D A/D A/D A/D

30/70 32/68 43/57 56/44

4

L. Hanyková et al. / European Polymer Journal 85 (2016) 1–13

Table 2 Equilibrium swelling ratio at 300 K and parameters of the phase transition obtained from DSC measurements (see Fig. 1), where Ton is the onset temperature, Tpeak is the peak temperature and DH is the enthalpy of the transition. Samples

Swelling ratio

Ton (K)

Tpeak (K)

DH (J/g of dry polymer)

DH (kJ/mol of PDEAAm units in dry polymer)

D sol

–

306.5

314.3

29.6

3.76

D 100/0

24.7

302.9

312.4

41.2

5.24

SIPN D/A 77/23 SIPN D/A 64/36 SIPN D/A 60/40

16.3 19.1 21.7

304.6 307.2 308.8

311.7 315.0 315.2

26.9 21.8 18.6

4.44 4.33 3.94

SIPN SIPN SIPN SIPN

10.4 16.3 14.6 14.1

305.0 306.2 307.3 309.1

310.7 313.4 316.2 317.7

26.1 24.2 20.0 15.1

4.75 4.52 4.47 4.36

IPN D/A 89/11 IPN D/A 85/15 IPN D/A 64/36

13.7 14.7 14.2

302.3 302.8 305.0

310.4 310.9 315.4

27.7 26.6 19.8

3.95 3.98 3.93

IPN A/D 15/85 IPN A/D 20/80 IPN A/D 30/70

17.7 35.6 33.3

303.6 306.4 309.7

313.3 313.5 316.8

22.9 16.7 5.9

3.43 2.67 1.07

A/D A/D A/D A/D

30/70 32/68 43/57 56/44

3.2. DSC Examples of DSC thermograms as obtained for PDEAAm hydrogel and the set of samples SIPN D/A are shown in Fig. 1. One rather broad endothermic transition was observed in all SIPNs and IPNs as well as in PDEAAm hydrogel and solution upon heating. Previously, broad DSC peaks were detected for PDEAAm solutions [27] and we suppose that a significant peak broadness (10 K) is related to chemical structure of the DEAAm side chains as it was observed for all samples, even for solution of linear PDEAAm. The onset and peak temperatures Ton and Tpeak, and enthalpy DH of the transition were extracted from the thermograms and are summarized in Table 2. Transition temperature of PDEAAm solution (Ton = 306.5 K) is in accord with the interval of LCST values 304–307 K reported for PDEAAm solutions from turbidimetric and microcalorimetric measurements [11–13]. In comparison with PDEAAm solution, hydrogel of neat PDEAAm shows a shift of the transition temperature to lower value. This effect was observed previously for linear and crosslinked PDEAAm by NMR [28] and it was attributed to the lower mobility of PDEAAm network segments due to crosslinks. The increasing content of hydrophilic PAAm component in SIPNs or IPNs shifts the transition toward higher values. The dependences of critical temperatures on composition of hydrogels will be discussed in detail together with NMR results. The enthalpy values corresponding to the phase transition were recalculated to the weight of dry sample (J/g of dry polymer) and to PDEAAm monomer units (kJ/mol of PDEAAm units in dry polymer). The enthalpy values obtained for SIPNs and IPNs samples are somewhat lower than the values of PDEAAm hydrogel and they slightly decrease with increasing content of AAm units in the hydrogel. Significant lowering of enthalpy values was detected for IPNs A/D 20/80 and 30/70, i.e., for samples which showed significantly higher swelling ratio.

SIPN D/A 60/40 0.75

cP (J/gK)

SIPN D/A 64/36 0.50

SIPN D/A 77/23 0.25

D 0.00 280

290

300

310

320

330

340

temperature (K) Fig. 1. DSC thermograms of PDEAAm hydrogel (D) and hydrogels SIPNs D/A with various composition.

5

L. Hanyková et al. / European Polymer Journal 85 (2016) 1–13

3.3. SANS Fig. 2 shows SANS curves for hydrogels of PDEAAm (D), SIPN 64/36, SIPN A/D 56/44 and 32/68, and IPN D/A 89/11 at 283 K and 328 K. Experimental points are denoted as square or circle symbols, depending on temperature, while full line shows the best fit achieved. A markedly different shape of scattering curves at temperature above the transition (328 K) and below the transition (283 K) is clearly visible in all cases. Scattering curve of the PDEAAm hydrogel (D) at 283 K can be described by the function of generalized Gaussian coil [26]. The Flory excluded volume parameter m is at this temperature equal to 0.67, which corresponds to the swollen chains. The situation is changing at temperature 328 K, i.e., above the phase transition. In this case scattering profile for the sample D can be fitted by sphere function with high polydispersity of the radius R [29]. We obtained R = 14.6 nm and polydispersity PD = 0.52 (the ratio of the standard deviation from the mean value and the mean value). The behavior of SIPN A/D, SIPN D/A and IPN D/A hydrogels is more complex and especially at temperature above the phase transition and after background subtraction it can be described considering two types of regions with increased chains density, representing the smaller and bigger regions. To get some information about sizes of these regions we assumed their fractal-like shape and we fitted experimental curves by mass fractal function [30]. During fit with only one function we observed deviation of theoretical function from experimental points in q-range 0.3–0.6 nm1. Introduction of additional function which describes objects with the size about 14 nm helped us to obtain a very good fit. A necessity to use combination of two functions to fit the experimental scattering curve is best demonstrated on the scattering profile of the sample SIPN A/D 32/68 at 328 K (in Fig. 2 star symbols and green line for experimental points and best fit, respectively), while for other samples it is not so pronounced. Therefore for all SIPN and IPN hydrogels the scattering curves were fitted using combination of two functions which show formation of regions (aggregates) of two sizes. Due to specific manipulation with the sample and sample holder only relative scale of intensity is presented, but some conclusions can be made on the basis of the ratio of intensities I at q = 0 (I0). During fitting procedure we always obtained

0.1

1

0. 1

103

1

SIPN D/A 64/36 283 K 328 K Best fit

D 283 K 328 K Best fit

102

101

Intensity, arb. u.

100

10-1 103

IPN D/A 89/11 283 K 328 K Best fit

SIPN A/D 56/44 283 K 328 K Best fit

102

SIPN A/D 32/68 328 K Best fit

101

100

10-1 0.1

1

0.1

1 -1

q, nm

Fig. 2. SANS curves of hydrogels in D2O of PDEAAm (D), SIPN 64/36, SIPN A/D 56/44 and 32/68, and IPN D/A 89/11 measured at 283 K (squares) and 328 K (circles) (for the sample SIPN A/D 32/68 only the curve at 328 K is shown; stars and green fit). Solid lines show best fits. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

6

L. Hanyková et al. / European Polymer Journal 85 (2016) 1–13

Table 3 Radius of gyration Rg for two types of regions (aggregates) existing in SIPN A/D, SIPN D/A and IPN D/A hydrogels at 283 K and 328 K. Sample

283 K

328 K

Rg (big) (nm)

Rg (small) (nm)

Rg (big) (nm)

Rg (small) (nm)

30/70 32/68 43/57 56/44

8.4 18.7 13.9 13.0

2.6 2.0 2.3 2.5

42.4 39.1 21.0 19.1

12.0 12.1 13.0 13.7

SIPN D/A 77/23 SIPN D/A 64/36 SIPN D/A 60/40

12.7 9.9 9.7

1.7 2.0 1.8

26.4 23.9 25.4

16.2 12.5 13.9

IPN D/A 89/11 IPN D/A 85/15

12.8 15.7

1.6 0.9

23.2 26.0

13.3 13.5

SIPN SIPN SIPN SIPN

A/D A/D A/D A/D

two values of I0 for each scattering curve, with exception of the sample D. At temperature 283 K the highest ratio of I0 of larger regions to I0 of smaller regions was obtained for the IPN D/A samples. Three times lower values of this ratio (values of I0 are more close the one with the other) were obtained for the SIPN D/A samples where the lowest value was obtained for the sample with the highest content of PAAm (SIPN D/A 60/40). This ratio does not change much with the content of PAAm component. When we prepare PAAm network inside PDEAAm solution (SIPN A/D samples) and if we assume formation of smaller and larger regions with virtually the same electron densities then we obtain lowest ratios I0big/I0small, i.e., a larger amount of smaller regions. Values of radius of gyration for both types of regions are shown in Table 3. When we used for fitting of scattering curves at 283 K generalized Gaussian coil functions, Flory exponent m was 0.50– 0.64, if we chose mass fractal function, fractal dimension was 1.73. This means that at 283 K both regions are represented by swollen/relaxed chains. At temperature 328 K, which is above the phase transition, compact aggregates are formed. Here again we used two mass fractal functions for fitting and we obtained fractal dimension 2.97 for the larger aggregates which shows compact structures with fuzzy surface. For the smaller aggregates we obtained fractal dimension 3.7 which shows compact structures with smooth surface. Both the model where large and small regions (aggregates) are separated and model where smaller regions are inside larger regions can be in accord with obtained results. Table 3 also shows that for all samples and both types of regions the Rg values obtained at 328 K are larger in comparison with values obtained at 283 K. At the same time while the Rg values of large regions are approx. twice larger at 328 K, Rg values of small regions are 5–10 times larger at 328 K in comparison with 283 K. Larger Rg values at temperatures above the phase transition we previously reported for IPN hydrogels of PNIPMAm/PNIPAm, PVCL/PNIPAm and PNIPAm/PAAm, as well as for PVCL aqueous solutions [21,22,31] and explained that above the transition they are related to multi-chain globular superstructures (aggregates). We assume that in general tendency to aggregation is a consequence of predominant hydrophobic interactions at temperatures above the transition. From Table 3 it also follows that while the sizes of smaller regions (aggregates) are at both temperatures very similar for all samples, there is a pronounced difference in behavior of Rg values of larger aggregates at 328 K between SIPN A/D samples on the one hand and SIPN D/A and IPN D/A samples on the other hand. For the collapsed SIPN A/D hydrogels the size of larger aggregates decreases with decreasing content of PDEAAm component, while such dependence was not found for collapsed SIPN D/A and IPN D/A hydrogels. Interestingly enough, from Table 3 it also follows that Rg values of both large and small regions are very similar for SIPN D/A and IPN D/A samples. 3.4. NMR Fig. 3 shows high-resolution 1H NMR spectra of SIPN A/D 56/44 measured under the same instrumental conditions at three temperatures. The assignment of resonances to various proton types (cf. Scheme 1) is shown directly in the spectrum measured at 300 K, i.e., below the VPTT of PDEAAm. The strong peak A corresponds to HDO. The most important effect observed in the NMR spectra measured at higher temperatures is a marked decrease in the integrated intensity of all polymer signals. Evidently, the mobility of most polymer units is significantly reduced to such an extent that the corresponding signals are too broad to be detected in high resolution NMR spectra [32]. To quantitatively characterize the phase transition, we have used the values of the collapsed p-fraction (degree of collapsing) obtained as

pðTÞ ¼ 1 

I I0

ð1Þ

where I is the integrated intensity of the given polymer signal in the spectrum of partly collapsed hydrogel and I0 is the integrated intensity of this signal if no collapse transition occurs. For I0, we took values based on integrated intensities as obtained at 300 K and taking into account the fact that the integrated intensities should decrease with absolute temperature as 1/T, i.e., I0(T) = I300 0 (300/T) [20,21,32]. Fig. 4 shows the temperature dependences of the p-fraction as obtained for all polymer signals in hydrogels of neat PDEAAm D 100/0 (Fig. 4a), semi-interpenetrating networks SIPN A/D 56/44 (Fig. 4b) and D/A 64/36 (Fig. 4d) and interpenetrating network D/A 37/64 (Fig. 4c). From Fig. 4a it follows that p fraction of all groups in the

L. Hanyková et al. / European Polymer Journal 85 (2016) 1–13

7

300 K A

F

B E

C D 310 K

320 K

5

4

3

2

1

chemical shift (ppm) Fig. 3. 1H NMR spectra (500.1 MHz) of SIPN A/D 56/44 in D2O recorded at 300 K, 310 K, and 320 K at the same instrumental conditions. Peak assignments are explained in the text and in Scheme 1.

neat PDEAAm show virtually identical dependences on temperature and maximum value pmax is equal 1 which means that all PDEAAm units are involved in phase separated globular structures. Fig. 4b–d demonstrate diverse temperature dependences of p-fraction for particular polymer signals as detected for SIPNs and IPNs. In the case of SIPN A/D 56/44 in Fig. 4b, pmax values approach 0.67 and 0.57 for the PDEAAm side diethyl group (signals B and F) and PDEAAm main chain group CH, respectively. Mixed PDEAAm and PAAm signal E of the main chain groups CH2 gives pmax values only 0.23 and the both main chain groups CH and CH2 show rather more gradual increase of p-fraction with temperature. Signal D of PAAm CH group practically does not change with temperature which implies that virtually all AAm units show high mobility even at elevated temperatures. Similar dependences as shown in Fig. 4b were obtained also for other SIPN A/D samples and for all IPN hydrogels as it is demonstrated for IPN D/A 37/64 in Fig. 4c. Evidently, the hydrophilic AAm units do not participate in collapse transition and moreover, they prevent DEAAm units to pack into globular structures and this results in smaller values of the p-fraction of PDEAAm proton groups. In this respect similar behavior was previously found for IPNs of PNIPAm and PAAm [22]. Fig. 4d depicts temperature dependences of p-fraction typical for semi-interpenetrating networks SIPNs D/A. PDEAAm signals show pmax values lower than 1 as a consequence of introduction of AAm component into hydrogel. In contrast to SIPNs A/D, here PAAm signals decrease in their intensities and corresponding temperature dependences of p-fraction show rather gradual two-step character with value of pmax
0.39

8

L. Hanyková et al. / European Polymer Journal 85 (2016) 1–13

0.7

b SIPN A/D 56/44

0.6

0.8

D 100/0 B C D F

0.6 0.4

B C D E F

0.5

p-fraction

p-fraction

0.8

a

1.0

0.4 0.3 0.2

0.2

0.1 0.0

0.0 295

300

305

310

315

320

325

300

310

p-fraction

0.6

1.0

c

0.9

340

SIPN D/A 64/36 B C D E F

0.7

B C D E F

0.4

330

d

0.8

IPN D/A 37/64

0.6

p-fraction

0.8

320

temperature (K)

temperature (K)

0.2

0.5 0.4 0.3 0.2 0.1

0.0

0.0

290

300

310

320

330

340

290

300

310

320

330

340

350

temperature (K)

temperature (K)

Fig. 4. Temperature dependences of the p-fraction as determined from various PDEAAm and PAAm NMR signals for D 100/0 (a), SIPN A/D 56/44 (b), IPN D/A 37/64 (c) and SIPN D/A 64/36 (d).

in the case of CH protons (signal D) and pmax
0.62 in the case of mixed CH2 band. Evidently, shrinking of crosslinked PDEAAm chains in SIPNs D/A leads to the partial collapse of linear PAAm chains, but this process is slower than the initial process when PDEAAm chains are shrinked. This can indicate that in collapsed SIPN D/A hydrogels PAAm chains form small regions as detected by SANS which are inside large regions (aggregates) preferably formed by chains of crosslinked PDEAAm. The effect of polymer composition is shown in Fig. 5 where the temperature dependences of the p-fraction in PAAm/ PDEAAm SIPN and IPN hydrogels, determined from the integrated intensities of the methyl signal of PDEAAm (signal F), 0.8 1.0

a

0.7

0.4

0.5

SIPN A/D 56/44 43/57 32/68 30/70

p-fraction

p-fraction

0.8 0.6

b

0.6

0.4 0.3 IPN A/D 30/70 20/80 15/85

0.2 0.2

0.1 0.0

0.0 295

300

305

310

315

320

temperature (K)

325

330

335

300 305 310 315 320 325 330 335 340 345

temperature (K)

Fig. 5. Temperature dependences of the p-fraction in hydrogels SIPN A/D (a) and IPN A/D (b) with various composition.

L. Hanyková et al. / European Polymer Journal 85 (2016) 1–13

9

are depicted. The results are reasonably in accord with DSC results and transition temperatures determined by NMR, TNMR (temperature in the middle of the transition corresponding to pmax/2) agree rather well with Tpeak values determined by DSC (cf. Table 2). Increasing content of PAAm component shifts VPTT to higher temperatures and maximum values pmax are lower. Temperature dependences of the p-fractions, determined from PDEAAm methyl signal were used to obtain maximum values pmax and critical temperatures TNMR for all measured samples. Dependences of pmax and TNMR on the composition of SIPN and IPN hydrogels are depicted in Fig. 6. The maximum value pmax of the neat PDEAAm hydrogel has reached almost 1 but it significantly decreases with increasing AAm content in SIPNs and IPNs as it is seen in Fig. 6a. At the same time, SIPN hydrogels show higher values of pmax in comparison with IPN hydrogels. This could indicate that in interpenetrating networks, where both components are crosslinked, more DEAAm units remain sufficiently mobile above the transition as crosslinked structure hampers them to pack tightly into collapsed phase. As it follows from Fig. 6b, critical temperature TNMR is proportional almost linearly to the content of DEAAm units in SIPNs and IPNs; linear dependence of critical temperature on the composition was already observed in systems of linear random copolymers [33–35] and IPN hydrogels based on chemically modified poly(vinyl alcohol) and PNIPAAm [36]. Surprisingly, all TNMR values except those corresponding to IPN A/D fit well to one linear dependence regardless of the preparation procedure. Much steeper dependence TNMR vs. composition as detected for hydrogels of IPNs A/D is obviously attributed to the substantially higher swelling ratios of these samples in comparison with other hydrogels (Table 2). It was shown that the VPTT of polymer hydrogels having the same swelling ratio is equal, indicating that the VPTT is determined by the overall hydrophilicity of the polymer [33]. Presence of PAAm network prepared as the first component of IPNs A/D thus leads to higher hydrophilicity of hydrogel and tends to increase the value of the VPTT and to reduce enthalpies as detected by DSC. Fig. 7 shows a part of 1H NMR spectrum with signals of water (HDO) for SIPN D/A 64/36 at room temperature and at two elevated temperatures. From this figure it follows that a new signal of HDO with a smaller chemical shift of 0.03–0.09 ppm in comparison with the main HDO peak was detected at higher temperatures. This new HDO signal which appears in spectra of all the investigated SIPN and IPN hydrogels but only in the transition region and at temperatures above the phase transition evidently corresponds to water molecules bound in globular structures of the collapsed hydrogel. Two separate NMR signals of water were previously detected also in highly concentrated poly(vinyl methyl ether) (PVME)/D2O solutions [37,38], collapsed PVME and PNIPAm hydrogels [39–42], homo- and copolymer microgels of NIPAm and DEAAm in water/methanol mixtures [43] and IPN hydrogels of PNIPMAm/PNIPAm, PVCL/PNIPAm and PNIPAm/PAAm [21–23], and

1.0

a

pmax

0.9 0.8 0.7

D 100/0 SIPN D/A SIPN A/D IPN D/A IPN A/D

0.6 0.5 40

50

60

70

80

90

100

content of PDEAAm units, mol%

320

D 100/0 SIPN D/A SIPN A/D IPN D/A IPN A/D

b

TNMR, K

318 316 314 312 310 308 40

50

60

70

80

90

100

content of PDEAAm units, mol% Fig. 6. Dependences of maximum value pmax (a) and the critical temperatures TNMR (b) on composition of SIPN and IPN hydrogels.

10

L. Hanyková et al. / European Polymer Journal 85 (2016) 1–13

a 300 K

b 311 K

c 330 K

5.5

5.0

4.5

4.0

chemical shift (ppm) Fig. 7. HDO signals in 1H NMR spectra of SIPN D/A 64/36 hydrogel in D2O measured at 300 K (a), 311 K (b) and 330 K (c).

similarly assigned to ‘‘free” and ‘‘bound” water. The existence of two separate HDO signals in studied hydrogels gives evidence of a slow exchange between ‘‘bound” and ‘‘free” water molecules. From the condition [44] 1/s  Dt, where s is the residence time and Dt is the difference of the respective chemical shifts in Hz, it follows that for the residence time of ‘‘bound” HDO molecules it holds s  20 ms. Substantial differences between ‘‘bound” and ‘‘free” water were found from measurements of spin–spin relaxation time T2 of HDO molecules. So in the collapsed hydrogel of SIPN D/A 64/36, T2 value 0.024 s detected at 330 K for ‘‘bound” (confined) water was two orders of magnitude smaller in comparison with T2 value 2.32 s of ‘‘free” water molecules. The main source of these differences is evidently the fact that the motion of ‘‘bound” water is spatially restricted and anisotropic [37]. It follows from the comparison with temperature dependences of p-fraction (cf. Fig. 4) that separate signal of the ‘‘bound” water appears already at temperatures where polymer segments just begin to form collapsed structures and p-fraction starts to increase above zero value. This behavior is probably connected with a self-organization process which includes water molecules and which begins in the early stage of the phase transition in hydrogels. Further it is clearly visible in Fig. 7 that during heating process, the intensity of the ‘‘bound” water signal is decreasing; fraction of ‘‘bound” water at 311 K and 330 K is 0.42 and 0.19, respectively. We attribute this effect to releasing of ‘‘bound” water from collapsed structures and besides increasing temperature this process is affected significantly by time as we recently found for IPN hydrogels PNIPAm/PAAm and PVCL/PNIPAm [22,23]. Interestingly enough, it follows from Figs. 7 and 8 that the difference of chemical shifts between signals of the ‘‘free” and ‘‘bound” water increases with increasing temperature and this dependence has asymptotic behavior.

L. Hanyková et al. / European Polymer Journal 85 (2016) 1–13

11

chemical shift difference (ppm)

0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 310

320

330

temperature (K) Fig. 8. Temperature dependence of chemical shift difference between signals of the ‘‘free” and ‘‘bound” water measured for SIPN D/A 64/36 hydrogel.

Such observation was for the first time reported for methanol signals in microgels of NIPAm and DEAAm [43] and indicates changes in the polymer structure of the collapsed hydrogel which surrounds the restricted solvent (water) molecules. When hydrogel network begins to shrink, a part of water is entrapped into collapsed structures and its nearest environment is then different from bulk water. During subsequent heating, continuous shrinking is accompanied with releasing of the ‘‘bound” water and this leads to increasing difference in the environment surrounding restricted water molecules which remain in the collapsed globular structures. Consequently, chemical shift difference between signals of the ‘‘free” water and ‘‘bound” water is increasing. It should be mentioned that we did not detect separate peaks of ‘‘free” and ‘‘bound” water in samples of the neat PDEAAm solution and hydrogel. The existence of two separate HDO signals means that there is a slow exchange between bound and free sites and it was shown for concentrated PVME solutions that the rate of the exchange process (slow or fast) is directly correlated to the size of phase-separated mesoglobules [38]. However, SANS results (cf. Table 3 and text in Section 3.3) do not show any significant difference in Rg values of globular structures and/or aggregates for collapsed hydrogel of the neat PDEAAm on the one hand and collapsed hydrogels of SIPNs and IPNs on the other hand. This indicates that there is another source of the different shape of the water (HDO) signal for these two types of collapsed hydrogels, such as a potentially different porous structure.

4. Conclusions In this work we combined methods of NMR spectroscopy, SANS and DSC to investigate temperature-induced phase transition in hydrogels of PDEAAm and PAAm with interpenetrating and semiinterpenetrating architecture, various composition and sequence in the preparation procedure. In summary, parameters of the phase transition were found to be tunable not only by composition of the SIPNs and IPNs but also by the preparation process and this knowledge can be useful in the design of new responsive materials. The increasing content of hydrophilic PAAm component in SIPNs and IPNs shifts the transition linearly toward higher temperatures but at the same time the value of VPTT is significantly influenced by swelling ratio of hydrogels. Exceptional behavior was detected for IPNs A/D where hydrophilic PAAm network was prepared as the first component. These networks showed higher swelling ratio and consequently higher transition temperatures and lower enthalpy values in comparison with other hydrogels. SANS results revealed in investigated SIPN and IPN hydrogels two types of regions, small and large. At temperature above the phase transition the small and large regions are related to compact multi-chain aggregates with smooth and fuzzy surface, respectively. While for the collapsed SIPN A/D hydrogels the size of larger aggregates decreases with decreasing content of PDEAAm component, for collapsed SIPN D/A and IPN D/A hydrogels their size does not depend on sample composition. Moreover, the size of both large and small regions is very similar for SIPN D/A and IPN D/A samples. Fraction pmax of polymer units with significantly reduced mobility determined by NMR is lower for hydrogels of IPNs in comparison with SIPNs. From detailed analysis of temperature dependences of the p-fraction it further follows that individual polymer groups participate in the collapse to various extent dependently on the type of hydrogel preparation. AAm units do not participate in collapse transition in SIPN A/D and all IPN hydrogels and they prevent DEAAm units to pack into globular structures; this leads to smaller values of the p-fraction of PDEAAm protons. On the other hand in SIPNs D/A hydrogels, linear PAAm chains partially participate in collapse as a consequence of shrinking of crosslinked PDEAAm chains. This indicates that in collapsed SIPN D/A hydrogels PAAm chains form small regions as detected by SANS which are inside large regions preferably formed by chains of crosslinked PDEAAm.

12

L. Hanyková et al. / European Polymer Journal 85 (2016) 1–13

A certain portion of water (HDO) bound (confined) in globular structures was revealed from measurements of 1H NMR spectra for all SIPNs and IPNs hydrogels. A slow exchange regime between ‘‘bound” and ‘‘free” water was established and for the residence time s of bound HDO it holds s >> 20 ms. The ‘‘bound” HDO appears already at temperatures when polymer segments just begin to form collapsed structures. During subsequent heating, water molecules bound in the collapsed structures are releasing and this leads to more different environment surrounding restricted water molecules which remain in the collapsed structures. Acknowledgment Support by the Czech Science Foundation – Czech Republic (Project 13-23392S) is gratefully acknowledged. SANS measurements were performed in the framework of collaboration of the Institute of Macromolecular Chemistry AS CR with Frank Laboratory of Neutron Physics, Joint Institute for Nuclear Research (Dubna, Russia) (Theme 04-4-1121-2015/2017). References [1] T. Tanaka, Collapse of gels and the critical endpoint, Phys. Rev. Lett. 40 (1978) 820–823. [2] K. Dušek (Ed.), Responsive Gels: Volume Transitions I, IIAdv. Polym. Sci. 109 and 110 (1993). [3] S. Fujishige, K. Kubota, I. Ando, Phase transition of aqueous solutions of poly(N-isopropylacrylamide) and poly(N-isopropylmethacrylamide), J. Phys. Chem. 93 (1989) 3311–3313. [4] L. Liu, W. Wang, X.J. Ju, R. Xie, L.Y. Chu, Smart thermo-triggered squirting capsules for nanoparticle delivery, Soft Matter 6 (2010) 3759–3763. [5] K. Nagase, J. Kobayashi, T. Okano, Temperature-responsive intelligent interfaces for biomolecular separation and cell sheet engineering, J. R. Soc. Interface 6 (2009) S293–S309. [6] Y. Takashima, S. Hatanaka, M. Otsubo, M. Nakahata, T. Kakuta, A. Hashidzume, H. Yamaguchi, A. Harada, Expansion-contraction of photoresponsive artificial muscle regulated by host-guest interactions, Nat. Commun. 3 (2012) 1270. [7] D.J. Beebe, J.S. Moore, J.M. Bauer, Q. Yu, R.H. Liu, C. Devadoss, B.H. Jo, Functional hydrogel structures for autonomous flow control inside microfluidic channels, Nature 404 (2000) 588–590. [8] D. Seliktar, Designing cell-compatible hydrogels for biomedical applications, Science 336 (2012) 1124–1128. [9] T.A. Asoh, M. Matsusaki, T. Kaneko, M. Akashi, Fabrication of temperature-responsive bending hydrogels with a nanostructured gradient, Adv. Mater. 20 (2008) 2080–2083. [10] L. Kris Kostanski, R. Huang, R. Ghosh, C.D.M. Filipe, Biocompatible poly(N-vinyllactam)-based materials with environmentally-responsive permeability, J. Biomater. Sci. Polym. Edn. 19 (2008) 275–290. [11] T. Baltes, F. Garret-Flaudy, R. Freitag, Investigation of the LCST of polyacrylamides as a function of molecular parameters and the solvent composition, J. Polym. Sci. Part A: Polym. Chem. 37 (1999) 2977–2989. [12] I. Idziak, D. Avoce, D. Lessard, D. Gravel, X.X. Zhu, Thermosensitivity of aqueous solutions of poly(N,N-diethylacrylamide), Macromolecules 32 (1999) 1260–1263. [13] H.Y. Liu, X.X. Zhu, Lower critical solution temperatures of N-substituted acrylamide copolymers in aqueous solutions, Polymer 40 (1999) 6985–6990. [14] M. Panayiotou, C. Pöhner, C. Vandevyver, C. Wandrey, F. Hilbrig, R. Freitag, Synthesis and characterisation of thermo-responsive poly(N,N0 diethylacrylamide) microgels, React. Funct. Polym. 67 (2007) 807–819. [15] M. Panayiotou, R. Freitag, Synthesis and characterisation of stimuli-responsive poly(N,N0 -diethylacrylamide) hydrogels, Polymer 46 (2005) 615–621. [16] Y. Maeda, M. Yamabe, A unique phase behavior of random copolymer of N-isopropylacrylamide and N,N-diethylacrylamide in water, Polymer 50 (2009) 519–523. [17] J.I. Ngadaonye, L.M. Geever, M.O. Cloonan, C.L. Higginbotham, Photopolymerised thermo-responsive poly(N,N-diethylacrylamide)-based copolymer hydrogels for potential drug delivery applications, J. Polym. Res. 19 (2012) 9822. [18] M. Keerl, V. Smirnovas, R. Winter, W. Richtering, Interplay between hydrogen bonding and macromolecular architecture leading to unusual phase behavior in thermosensitive microgels, Angew. Chem. Int. Ed. 47 (2008) 338–341. [19] N. Zhang, M. Liu, Y. Shen, J. Chen, L. Dai, C. Gao, Preparation, properties, and drug release of thermo- and pH-sensitive poly((2-dimethylamino)ethyl methacrylate)/poly(N,N-diethylacrylamide) semi-IPN hydrogels, J. Mater. Sci. 46 (2011) 1523–1534. [20] J. Štˇastná, L. Hanyková, Z. Sedláková, H. Valentová, J. Speˇvácˇek, Temperature-induced phase transition in hydrogels of interpenetrating networks poly(N-isopropylmethacrylamide)/poly(N-isopropylacrylamide), Colloid. Polym. Sci. 291 (2013) 2409–2417. [21] L. Hanyková, J. Speˇvácˇek, M. Radecki, A. Zhigunov, J. Štˇastná, H. Valentová, Z. Sedláková, Structures and interactions in collapsed hydrogels of thermoresponsive interpenetrating polymer networks, Colloid Polym. Sci. 293 (2015) 709–720. [22] M. Radecki, J. Speˇvácˇek, A. Zhigunov, Z. Sedláková, L. Hanyková, Temperature-induced phase transition in hydrogels of interpenetrating networks of poly(N-isopropylacrylamide) and polyacrylamide, Eur. Polym. J. 68 (2015) 68–79. [23] J. Speˇvácˇek, L. Hanyková, Phase transition in hydrogels of thermoresponsive interpenetrating polymer networks, Macromol. Symp. 359 (2016) 111– 117. [24] T.C. Farrar, E.D. Becker, Pulse and Fourier transform NMR, Academic Press, 1971. p 27–29. [25] A.G. Soloviev, T.M. Solovieva, A.V. Stadnik, A.K. Islamov, A.I. Kuklin, Programma dlja pervicnoj obrabotki spektrov malouglovovo rassejanija, Communication of JINR, 2003, P10-2003-86. [26] J. Kohlbrecher, I. Bressler, Software package SASFit, . [27] L.H. Gan, C. Wensheng, K.C. Tam, Studies of phase transition of aqueous solution of poly(N,N-diethylacrylamide-co-acrylic acid) by differential scanning calorimetry and spectrophotometry, Eur. Polym. J. 37 (2001) 1773–1778. [28] J. Speˇvácˇek, D. Geschke, M. Ilavsky´, 1H NMR study of temperature collapse of linear and crosslinked poly(N,N-diethylacrylamide) in D2O, Polymer 42 (2001) 463–468. [29] A. Guinier, G. Fournet, Small-Angle Scattering of X-rays, John Wiley and Sons, New York, 1955. [30] C.M. Sorensen, G.M. Wang, Size distribution effect on the power law regime of the structure factor of fractal aggregates, Phys. Rev. E 60 (1999) 7143– 7148. [31] J. Speˇvácˇek, J. Dybal, L. Starovoytova, A. Zhigunov, Z. Sedláková, Temperature-induced phase separation and hydration in poly(N-vinylcaprolactam) aqueous solutions: a study by NMR and IR spectroscopy, SAXS, and quantum-chemical calculations, Soft Matter 8 (2012) 6110–6119. [32] J. Speˇvácˇek, NMR investigations of phase transition in aqueous polymer solutions and gels, Curr. Opin. Colloid Interface Sci. 14 (2009) 184–191. [33] H. Feil, Y.H. Bae, J. Feijen, S.W. Kim, Effect of comonomer hydrophilicity and ionization on the lower critical solution temperature of N-isopropylacrylamide copolymers, Macromolecules 26 (1993) 2496–2500. [34] E. Djokpé, W. Vogt, N-isopropylacrylamide and N-isopropylmethacrylamide: cloud points of mixtures and copolymers, Macromol. Chem. Phys. 202 (2001) 750–757. [35] L. Starovoytova, J. Speˇvácˇek, M. Ilavsky´, 1H NMR study of temperature-induced phase transitions in D2O solutions of poly(N-isopropylmethacrylamide)/ poly(N-isopropylacrylamide) mixtures and random copolymers, Polymer 46 (2005) 677–683.

L. Hanyková et al. / European Polymer Journal 85 (2016) 1–13

13

[36] A.C. Wenceslau, F.G. dos Santos, É.R.F. Ramos, C.V. Nakamura, A.F. Rubira, E.C. Muniz, Thermo- and pH-sensitive IPN hydrogels based on PNIPAAm and PVA-Ma networks with LCST tailored close to human body temperature, Mater Sci Eng, C 32 (2012) 1259–1265. [37] J. Speˇvácˇek, L. Hanyková, 1H NMR study on the hydration during temperature-induced phase separation in concentrated poly(vinyl methyl ether)/D2O solutions, Macromolecules 38 (2005) 9187–9191. [38] J. Speˇvácˇek, L. Hanyková, J. Labuta, Behavior of water during temperature-induced phase separation in poly(vinyl methyl ether) aqueous solutions. NMR and optical microscopy study, Macromolecules 44 (2011) 2149–2153. [39] L. Hanyková, J. Speˇvácˇek, M. Ilavsky´, 1H NMR study of thermotropic phase transition of linear and crosslinked poly(vinyl methyl ether) in D2O, Polymer 42 (2001) 8607–8612. [40] E. Díez-Peña, I. Quijada-Garrido, J.M. Barrales-Rienda, M. Wilhelm, H.W. Spiess, NMR studies of the structure and dynamics of polymers gels based on N-isopropylacrylamide (N-iPAAm) and methacrylic acid (MAA), Macromol. Chem. Phys. 203 (2002) 491–502. [41] N. Wang, G. Ru, L. Wang, J. Feng, 1H MAS NMR studies of the phase separation of poly(N-isopropylacrylamide) gel in binary solvents, Langmuir 25 (2009) 5898–5902. [42] T.M. Alam, K.K. Childress, K. Pastoor, C.V. Rice, Characterization of free, restricted, and entrapped water environments in poly(N-isopropyl acrylamide) hydrogels via 1H HRMAS PFG NMR spectroscopy, J. Polym. Sci. Part B: Polym. Phys. 52 (2014) 1521–1527. [43] C.H. Hofmann, F.A. Plamper, C. Scherzinger, S. Hietala, W. Richtering, Cononsolvency revisited: solvent entrapment by N-isopropylacrylamide and N,N-diethylacrylamide microgels in different water/methanol mixtures, Macromolecules 46 (2013) 523–532. [44] P.A. Mirau, A Practical Guide to Understanding the NMR of Polymers, Wiley, 2004. pp. 24–27.