Study of synthesis parameters on the physical properties and morphology of smart PNIPAAm hydrogels

In the present work, three parameters were changed in the synthesis of PNIPAAm: the crosslinker agent, initiator concentrations, and magnetic agitation.

The monomer used was N-isopropylacrylamide (NIPAM), supplied by Sigma-Aldrich. The crosslinker was N-N’-methylen(bis)acrylamide (MBA), supplied by Sigma-Aldrich and the initiators were ammonium persulphate (APS) (Gold FM), and sodium metabisulphite (NMB) (Farmos).

The PNIPAAm synthesis was performed in an aqueous solution under a nitrogen atmosphere as prescribed by Queiroz [14]. NIPAM and MBA were weighed, diluted in approximately 5 mL of deionized water, and placed in a 50 mL glass balloon. The balloon was closed and placed in a nitrogen atmosphere for 5 min. The initiators APS and NMB were weighed and diluted in approximately 5 mL of deionized water. This solution was added to the glass balloon. The reaction was placed in water at 50°C. Seven hydrogel groups were prepared with the concentration shown in Table 1. The H1.A, H2.A, and H3A hydrogel groups were prepared with magnetic stirring during the polymerization reaction. The H1.0, H2.0, H3.0, and H4.0 hydrogel groups stayed in repose, without magnetic stirring, during the polymerization reaction. Table 1 shows that the concentrations of H1.0, H2.0, and H3.0 groups are the same as the H1.A, H2.A, and H3.A groups, and the difference among them was the synthesis process. The gelation started 30 s and 1 min after the introduction of the last reagent, varying according to the crosslinking ratio, being that it was inversely proportional to crosslink concentration. During this process, the solution changed from transparent to milky and viscous. After gelation, the samples stayed at room temperature for 24 h to complete the reaction. After that time, the samples were washed three times with approximately 15 mL of deionized water in a centrifuge at 3,500 rpm to eliminate any residues, and then they were placed in an oven at 60°C to remove residual water. The reagent concentrations in mmol and mol % are shown in Table 1.

The swelling ratio, also called the mass expansion ratio, is used to relate the mass of the water-swollen hydrogel (M_ge) to the mass of the dry hydrogel (M_g) [15]. It is given by the following equation 1 [16, 17]: (1) Q=[ (Mge−Mg)/Mg ]×100% Q = \left[ {\left( {{M_{ge}} - {M_g}} \right)/{M_g}} \right] \times 100\% Approximately 250 mg of each dry sample were placed in a 50 ml beaker, which was filled with Phosphate Buffered Saline (PBS) solution to 30 mL to let the gels expand at room temperature (25°C) and 37°C. After the time of 1, 5, 10, 15, 20, 30, 40, 50, 60, 80, 100, 120 min, the samples were removed from the beaker, dried with filter paper to remove the surface water, and weighed. To determine the dry mass, M_g, the samples were put in an oven at 60°C for 24 h before weighing. This experiment was carried out in triplicate.

Physical properties were characterized using ¹H-NMR spectroscopy to determine the number of hydrogen atoms in the sample molecules. The ¹H spectra were recorded in a Varian VNMRS spectrometer operating at a frequency of 500 MHz. The measurements were performed after the samples had been immersed for 25 min in deuterated water (D₂O) solvent, with a hydrogel/solvent ratio of 1:4 w/w. The temperature was about 30°C, and the spectra were collected on a preset mode to suppress the D₂O peak. The chemical dislocations (δ) were expressed as parts per million (ppm).

FTIR spectra were performed in a Varian 660 IR spectrometer in the 400–4,000 cm⁻¹ range using specular reflection [18]. The spectra were collected at room temperature with a nominal resolution of 4.00 cm⁻¹ and 32 scans, using the attenuated total reflectance (ATR) accessory. Approximately 3 mg of dry hydrogel powder was used in this analysis.

Assessment of the hydrogel composition was performed by energy dispersive spectroscopy (EDS) at various magnifications and accelerating voltages at randomly selected spots, using a Bruker detector controlled by Quantax CrystAlign Software attached to an FEI Quanta FEG 250 SEM.

XRD powder diffraction patterns (XRD) were recorded in a Bruker-AXS D8 Advance diffractometer, from 5° to 60° (2θ) with a step size of 0.02 at room temperature, using a Cu Kα (λ = 0.154 Å)′ radiation source. Before the measurement, the samples were macerated to avoid a preferential orientation. The X’Pert Data Viewer software was used to compute the areas of the amorphous halo (A_a) and the crystalline peaks (A_c) to estimate the degree of crystallinity (X_C) of the hydrogels using the following equation 2, proposed by Ruland [19]: (2) XC=(AC/Aa+AC)×100% {X_C} = \left( {{A_C}/{A_a} + {A_C}} \right) \times 100\%

Before SEM measurements, the samples were immersed for 24 h in deionized water, placed in liquid nitrogen for 1 min, freeze-dried for 3 days for complete water removal, and sputter-coated by platinum to enhance conductivity. The surface morphology was studied using an FEI Quanta FEG 250 SEM.

TGA measurements were performed using a TA Instruments Q500 thermogravimetric analyzer. Approximately 5 mg of each sample were placed in a platinum crucible and heated at a rate of 5°C/min from 25°C to 600°C in a nitrogen atmosphere. The first derivative of the TGA curve (DTA) was computed for better visualization of the thermal behavior of the samples.

Figure 2 shows the oscillatory shrinking-swelling property of the PNIPAAm hydrogels. The oscillation became less pronounced when the samples were heated to 37°C because of the LCST effect. The hydration/dehydration kinetics of H2.0 and H4.0 samples remained stable for 60 min and presented less variation of swelling ability than the samples H1.0 and H3.0 due to the higher concentration of MBA and APS, which increased the rigidity of the hydrogels. As the H1.0 and H1.A have lower MBA concentrations, they have higher swelling and oscillatory shrinking-swelling capacity, being the less rigid hydrogels, in order words, they have a high capacity to incorporate water inside their structures.

Fig. 2

As crosslink reagent concentration increased, more rigid was the polymer structure, and the less was the swelling capacity. This behavior is the consequence of the number of bridges formed on the polymeric structure. On the other hand, comparing H3.0 with H4.0 groups, which differ only in the APS concentration, one notices that the H4.0 sample, which has a higher APS concentration, is more rigid, showing that the redox initiator has an important rule to facilitate crosslinking. It was observed that the magnetic stirring during the synthesis process was important to determine the hydrogels’ behavior. The H1.A, H2.A, and H3.A groups showed a great oscillatory swelling capacity. The magnetic stirring could have contributed to forming polymeric chains of different sizes and crosslinking degrees.

Table 2 shows the NMR analysis was performed in H1.0, H2.0, H3.0, and H4.0 unheated samples because they were the ones with the largest swelling.

The chemical shifts are similar in all four un-heated hydrogel samples (Figure 3) and did not show the peaks at 5.80 and 6.27 ppm that appear in the Nipam spectrum. According to Wang et al. [21], the multiplet at about 1.2 ppm is due to a methyl group (CH₃), the multiplet at about 1.7 ppm is due to a methylene group (CH₂), the multiplet at about 2.1 ppm is associated with hydrogen close to the oxygen of an amide group, and the multiplet at about 4.0 ppm is due to one hydrogen atom close to the nitrogen of an amide group. It was previously mentioned that when a hydrated hydrogel is heated at temperatures close to the LCST, some water molecules are expelled from the polymer network to form free water (FW). The remaining water in the polymer is named confined water (CW). That is why the H4.0 sample, which has the highest crosslink concentration, exhibits the largest CW peak. The peaks labeled * in the spectra are due to rotational sidebands. The hydrogen of the amide group did not appear on spectra, probably because it was associated with interchangeable hydrogen atoms of the D₂O solvent [20, 21].

Fig. 3

Figure 4 shows the ATR-FTIR spectra of the hydrogels. All samples showed a band at 3,285 cm⁻¹ associated with stretching and a band at 1,535 cm⁻¹ associated with angular deformation of the N-H bond of secondary amide. The band at 3,080 cm⁻¹ is attributed to the unfolding of the N–H bond. The absence of a peak between in the 1,675–1,645 cm⁻¹ interval, which would be associated with the C=C bond, means there was no residual of NIPAM monomer in the hydrogels. The bands at 2,865, 2,930, and 2,974 cm⁻¹ are related to C–H stretching and the one at 1,623 cm⁻¹ is characteristic of the C=O stretching of carbonyl amide. Angular deformation of the C-H bond of methyl and isopropyl groups can be observed at 1,456 and 1,376 cm⁻¹, respectively.

Fig. 4

Table 3 shows the hydrogels’ chemical composition using EDS analysis. All samples contained carbon, oxygen, and platinum, which was the element used for coating. This result suggests that the synthesis did not incorporate impurities such as sulfur from the redox initiator. Nitrogen was detected only in samples with a relatively low concentration of MBA (hydrogels groups H1.A, H1.0, H2.A, and H2.0), probably because the other samples had a large number of pores, so nitrogen detection was impaired. Figure 5 shows the EDS spectrum and the corresponding SEM image of the hydrogel H1.A.

Fig. 5

All samples showed an amorphous XRD halo, characteristic of PNIPAAm (Figure 6). Crystalline peaks were observed in hydrogels H2.0, H3.0, and H4.0 groups. This suggests that these hydrogels had crystalline regions not described in the literature, which could be an indication of a trend in the polymer chain to be organized. The degrees of crystallinity, determined by the Ruland method [19], were 7.3%, 5.5%, and 5.6% for hydrogels H2.0, H3.0, and H4.0, respectively.

Fig. 6

Figure 7 shows different hydrogel morphologies. Samples from H2.0, H3.0, and H4.0 groups display some degree of organization. These results can be associated with the small movements of their molecules during synthesis. The smaller the movement, the lower will be the entropy of the system. This behavior allows for organized polymerization [22]. The lamellar structure was observed in samples with many pores. The H4 hydrogel showed a structure with a greater formation of layers. The distance between the layers is approximately 10 μm.

Fig. 7

The sample of H1.0 does not show organized regions (Figure 8), with many cavities in some regions. This is consistent with XRD results since its diffractogram does not have crystalline peaks.

Fig. 8

All synthesized samples with magnetic stirring during the polymerization process (H1.A, H2.A, and H3.A) showed no ordering and different morphologies with cavities of different sizes (in Figure 9), so these results corroborate the oscillatory swelling capacity. The sample of H1.A had two different regions, one with large cavities and another similar to a spiderweb. Sample H2.A displayed diagonal layers and cavities of several sizes. Both H1.A and H2.A showed a nonporous surface. The sample of H3.A showed a globular surface with a convoluted morphology that gives it a spongy appearance. Of the three, H3.A is the one with the largest swelling capacity because the bigger the cavities the more solvent would be warehoused. Magnetic agitation has been an indication of the influence of entropy on the system: when the hydrogel is not at rest during the synthesis process, entropy increases, and so does the number of events that polymer chains have to connect to establish a diversified number of polymeric bridges and create hydrogels with different behaviors.

Fig. 9

Thermogravimetric analysis of the hydrogels (Figure 10) showed that the material had degraded in different events. All hydrogels had a mass decrease at a temperature below 150°C, which is due to the loss of the water absorbed by the sample. The samples of H2.A and H3A had only two degradation events, meaning that these hydrogels have chains with similar size distribution. The samples of H1.A, H1.0, H2.0, H3.0, and H4.0 had three degradation events. The degradation events are shown in Table 4, with values of T_max and T_peak. The second degradation event occurred below 300°C for the samples of H1.A, H1.0, H3.0, and H4.0; for the samples of H2.A and H3.A, they happened around 380°C. The third event occurred below 415°C for the samples of H1.A, H1.0, H2.0, H3.0, and H4.0. According to Lucas et al. [23], this degradation event can be an indication that these hydrogels have smaller chains with a wider size distribution. The temperature at which this degradation event occurs depends on the chain length: the higher the value of T_peak, the longer the chains and the more rigid the structure. The sample of H2.0 had the largest value of T_peak, 412°C and the smallest swelling capacity, showing that it has the longest and most rigid chains. The sample of H4.0 had the smallest value of T_peak, 376°C, suggesting that it had the shortest chains with a highly cross-linked network according to its small swelling capacity. Samples synthesized with stirring (H2.A and H3.A) showed greater thermal stability than those synthesized without stirring. With respect to swelling capacity, the sample of H2.A is more rigid than the sample of H3.A, and this is confirmed by thermogravimetric analysis, since for the second event, T_peak of the sample of H2.A is 8°C higher than for the samples of H3.A. Moreover, these hydrogels were the most stable with values of T_peak for the first event of 104°C and 111°C, respectively.

Fig. 10