Introduction

Stimuli-responsive materials are smart materials because they can sense and respond to changes in environmental conditions, such as pH, temperature, light, ionic strength, electric field, and magnetic field [1,2,3,4]. Changes in surrounding conditions trigger a change in the physical and chemical properties of these materials, such as size, shape, hydrophobicity/hydrophilicity, and degradation rate [2]. These responsive materials can, therefore, be used to fabricate responsive, active devices that can regulate and introduce new functionalities in a range of applications. Temperature is one of the most widely used external stimuli because it can be simply modulated by temperature-programmed equipment or ambient conditions. Temperature-responsive materials are generally those that undergo a volume transition or soluble-to-insoluble change in solvents at around the lower critical solution temperature (LCST). One of the most widely studied temperature-responsive polymers is based on poly(N-isopropylacrylamide) (PNIPAM) because it has an LCST of around 32 °C in water, which is near the human body temperature [5,6,7,8,9,10,11,12,13,14]. The non-toxicity, high water solubility, and biocompatibility of PNIPAM make it applicable in a range of biomedical applications. However, its mechanical properties are not particularly useful and, therefore, significant efforts have been directed toward improving its mechanical strength. A range of temperature-responsive copolymers have also been prepared that increase the strength of and impart new properties to hybrid materials. NIPAM has been copolymerized and graft copolymerized with several polymers, including cellulose [12], poly(N-vinylpyrrolidone) [15], poly(acrylic acid) [7], poly(ethylene oxide) [16], polysiloxanes [17], chitosan [11], and styrene–butadiene rubber [14]. NIPAM-based copolymers thus have the potential to be used for a wider range of applications.

Natural rubber (NR) is a well-known renewable elastomeric material obtained from the Hevea brasiliensis tree in the form of a milky white fluid called latex [18]. NR is a versatile material widely used in many applications. However, NR is poorly resistant to oxidation, ozone, weathering, and many chemicals and solvents because of unsaturation in cis-polyisoprene, a major polymer of NR [19, 20]. Several undesirable reactions, such as crosslinking or chain scission, occur on the unsaturated backbone of cis-polyisoprene. Although the presence of double bonds makes cis-polyisoprene susceptible to degradation, they are points at which new functionalities can be added. Several chemical reactions have been used to modify NR, producing hybrid structures with high strength and stability. Various monomers have been grafted with NR, including dimethyl aminoethyl methacrylate [21], glycidyl methacrylate [22], methyl methacrylate [23,24,25], styrene [26,27,28], 2-hydroxyethyl methacrylate [29, 30], vinyl alcohol [31], and maleic anhydride [32, 33]. Recently, our group reported the synthesis of temperature-responsive crosslinked natural rubber with NIPAM and N-vinylcaprolactam. However, no temperature-responsive materials have to date been reported via grafting reaction with NR.

In grafting monomers and polymers with NR, the main goal has been to improve the thermal, chemical, and mechanical properties of the NR, and the addition of responsive functions to NR via grafting reaction has never been reported. In this work, we propose a method for grafting NIPAM and NR as a means for preparing rubber-based, temperature-responsive materials. Grafting of hydrophilic monomers remains a challenge, because of the presence of surface proteins as rubber particle shells. Monomers are not well incorporated into the hydrophobic NR cores in latex and, therefore, often homopolymerize [34,35,36,37]. Therefore, in this work grafting was carried out using deproteinized NR latex and water-soluble potassium persulfate as an initiator. The LCST of the graft copolymers was measured and compared with that of pristine NR. Dye adsorption was investigated and the adsorption isotherms were derived. Based on these results, we show that temperature responsiveness can be introduced to NR, which will expand its range of applications.

Materials and methods

Materials

High-ammonia NR latex [60% dry rubber content (DRC)] was purchased from the Department of Agriculture, Thailand. All other chemicals were purchased from Sigma-Aldrich and used as received unless otherwise noted. NIPAM was purified by recrystallization in methanol. Organic solvents (Labscan, AR) were used as received. Deionized (DI) water was used throughout the experiments. PNIPAM was synthesized from NIPAM via free radical polymerization using benzoyl peroxide and used as a reference.

Deproteinization of natural rubber

NR latex was added to a mixture of 0.1 wt% urea and 1 wt% sodium dodecyl sulfate (SDS) solutions. The mixture was incubated for 60 min at 25 °C and then diluted with DI water to obtain 30% DRC latex. After centrifugation at 10,000 rpm for 45 min at 25 °C, the cream fraction was re-dispersed in 1 wt% SDS and was centrifuged twice to obtain deproteinized natural rubber (DPNR) latex. Finally, the DPNR was kept in 0.1 wt% SDS with a final DRC of 32%.

Grafting of NIPAM onto DPNR

The DPNR (10 g) was diluted with DI water to approximately 16 wt% DRC and placed in a 250-mL round-bottom flask. The mixture was then degassed under a nitrogen atmosphere for at least 45 min. SDS (1 phr) as an emulsifier was then added to the mixture and stirred for 15 min. Afterwards, potassium persulfate (KPS, 1 phr) as an initiator was added continuously to the solution while stirring for 30 min at 75 °C.

The N-isopropylacrylamide (NIPAM, 100 phr) was diluted with DI water in a 250-mL round-bottom flask. SDS (1 phr) was then added to the solution and stirred for 15 min. KPS (1 phr) was slowly added and stirred at 95 °C for 30 min.

The NIPAM-initiator solution and the NR-initiator solution were mixed. The reaction mixture was heated at 75 °C for 12 h. Next, the reaction was cooled to 30 °C and precipitated in acetone. The solid polymer product was purified by stirring in acetone twice to remove unreacted monomer and contaminants. The grafted rubber was dried under vacuum at 60 °C for 24 h.

Grafting efficiency is the fraction of grafted NIPAM on the backbone NR chains. The grafting efficiency was calculated using the following equation [38]:

$$ {\text{Grafting efficiency}}\left( \% \right) = \frac{{I_{4.0} }}{{I_{5.2} }} \times 100, $$
(1)

where I 4.0 is the integrated peak area of –CH–NH– from the NIPAM units and I 5.2 is the integrated peak area of –CH– in the isoprene units in the 1H-NMR spectrum.

Determination of LCST

The rubber samples were cut into small pieces (0.75 × 0.75 cm2) and compressed to displace air bubbles and reduce porosity by sandwiching them between two glass slides and heating at 60 °C overnight. The samples were then immersed in DI water for 2 h at temperatures of 28, 30, 32, 34, 36, and 38 °C to reach the saturation of water swelling. After gently removing the surface liquid with tissue paper, the weight of the materials was measured and the swelling percentage was calculated to determine the LCST, using the following equation [39]:

$$ \% {\text{Swelling}} = \frac{{W_{2} - W_{1} }}{{W_{1} }} \times 100, $$
(2)

where W 1 and W 2 are the weight of the material before and after immersion.

Adsorption studies

Five indigo carmine solutions (10, 20, 30, 40, and 50 ppm) were prepared using DI water. Their absorbance was measured using UV–visible spectrophotometry. The NIPAM-g-DPNR samples (0.2 g) were immersed in the indigo carmine solutions (10 mL) and were then placed on a shaker for 1 week [39]. The rubber samples were removed and the absorbance was measured. The concentration of indigo carmine remaining in the solution was then calculated from the absorbance, using the calibration curve.

The amount of indigo carmine adsorbed on the NIPAM-g-DPNR sample was calculated using the following equation:

$$ q_{\text{e}} = \frac{{(C_{\text{o}} - C_{\text{e}} )V}}{W}, $$
(3)

where q e (mg g−1) is the adsorption capacity of the rubber, C o (mg L−1) is the initial concentration of indigo carmine in the solution, C e (mg L−1) is the concentration of indigo carmine remaining in the solution, V (L) is the volume of the solution, and W (g) is the dry weight of the rubber sample. The data were fit using the Langmuir and Freundlich isotherm equations (Table 1).

Table 1 Mathematical equations for the Langmuir and Freundlich isotherm models
Full size table

Characterization

CHN data were obtained using a CHN-2000 LECO analyzer. 1H (400 MHz) nuclear magnetic resonance (NMR) spectra were obtained using an AVANCE Bruker NMR Spectrometer with CDCl3 as a solvent. The Fourier transform infrared (FT-IR) spectra were obtained using a Perkin Elmer FT-IR (Spectrum GX model) spectrometer and NaCl salt windows. Thermogravimetric analysis (TGA) of the polymers was carried out using an SDTA851e Mettler Toledo analyzer. The glass transition temperatures of the polymers were obtained using a DSC822e Mettler Toledo differential scanning calorimeter (DSC). In both cases, the samples were heated at a rate of 10 °C min−1 under nitrogen atmosphere. UV–visible spectra were obtained using a UV-1700 PharmaSpecUV–visible spectrometer. The water contact angle was measured using an OCA35 Dataphysics contact angle meter. The size of the water droplet was 8 µL.

Results and discussion

Synthesis and characterization of graft copolymer

The graft copolymer of NIPAM and NR was prepared via free radical grafting in an aqueous system using KPS as the initiator. The NR latex was deproteinized prior to grafting, because previous studies have shown that deproteinization removes surface protein, improving grafting efficiency [40, 41]. Deproteinization was carried out using urea as the deproteinizing agent. The deproteinized NR (DPNR) was characterized using CHN analysis. It was found that the nitrogen content was 43% lower in the DPNR than the NR, whereas carbon and hydrogen content remained almost unchanged (Table 2). This indicates that deproteinization of the NR removed approximately half of protein from the NR latex, comparable with previous report [41].

Table 2 Elemental analysis of NR and DPNR
Full size table

Grafting of N-isopropylacrylamide (NIPAM) onto DPNR was carried out in an aqueous system using KPS as free radical initiator (Fig. 1). The formation of grafting copolymer is believed to proceed by two possible pathways: a reaction between the free radicals and double bonds of the isoprene units, and a hydrogen abstraction mechanism [39, 42,43,44,45]. The NIPAM-grafted DPNR (NIPAM-g-DPNR) was purified by stirring twice in acetone to remove impurities, unreacted NIPAM, and PNIPAM homopolymer. The NIPAM-g-DPNR was characterized using FT-IR and 1H-NMR and compared with the DPNR and PNIPAM. The FT-IR spectrum of the NIPAM-g-DPNR showed signature signals of both DPNR and PNIPAM (Fig. 2). Signals at 3000 and 1660 cm−1 corresponded to C–H and C=C stretching vibrations, respectively, while those at 3300, 2962, 1650, and 1540 cm−1 corresponded to N–H stretching, C–H stretching, C=O stretching (amide I), and C=O stretching (amide II) vibrations, respectively. The FT-IR results are consistent with those reported in previous studies [46]. Images were taken of the NIPAM-g-DPNR, DPNR, and NIPAM (Fig. 1b). DPNR appeared as a clear solid and NIPAM as a white powder. The NIPAM-g-DPNR was a yellowish solid and stiffer than the DPNR.

Fig. 1
figure 1

a Schematic representation for the synthesis of NIPAM-g-DPNR from DPNR and NIPAM and b their respective images

Full size image
Fig. 2
figure 2

FT-IR spectra of DPNR, NIPAM-g-DPNR, and PNIPAM

Full size image

The 1H-NMR spectrum of DPNR showed signals at 5.2, 2.1, and 1.7 ppm, corresponding to C–H, CH2, and CH3, respectively (Fig. 3), consistent with previous reports [2, 46, 47]. The 1H-NMR spectrum of the PNIPAM showed signals at 4.00, 1.82, 1.65, and 1.14 ppm, corresponding to CH–NH, C–H, CH2, and CH3, respectively. The 1H-NMR spectrum of NIPAM-g-DPNR showed signals corresponding to both DPNR and PNIPAM. The signal at 4.0 ppm corresponded to CH–NH in the N-isopropylacrylamide, while the signal at 5.2 ppm corresponded to the C–H of the isoprene units. These signals were used to calculate the grafting efficiency, which was found to be 18%.

Fig. 3
figure 3

1H-NMR spectrum of a DPNR, b sample after washing PNIPAM/NR blend, c PNIPAM, and d NIPAM-g-DPNR

Full size image

To further confirm the formation of graft copolymer, a blend sample between DPNR latex and PNIPAM was prepared in water solution and dried as a thin film on a glass plate. The film was peeled off and purified via stirring twice in acetone and dried in an oven at 60 °C overnight. It was found that no PNIPAM appeared in the 1H-NMR spectrum as it was soluble in acetone (Fig. 3b). Therefore, the presence of both NIPAM and DPNR in the NIPAM-g-DPNR sample was attributed to the formation of a grafted structure that was inseparable by stirring in acetone. This also confirmed that stirring twice in acetone effectively removed the unbound PNIPAM homopolymer.

Effect of reaction conditions

The effect of reaction conditions and reagent concentrations on grafting efficiency was investigated. First, the effect of reaction temperature was investigated using a NIPAM concentration of 100 phr, KPS concentration of 1 phr, and reaction time of 12 h (Fig. 4a). It was found that when the reaction temperature was increased from 75 to 95 °C, which is the optimum working range for KPS [4], grafting increased from 12 to 21%. This dropped to 15% when the temperature increased further. Second, the grafting percentage as a function of NIPAM concentration was measured using a KPS concentration of 1 phr and reaction time of 12 h at 95 °C. It was found that, as the NIPAM concentration increased from 50 to 200 phr with respect to NR, grafting increased from 10 to 39% (Fig. 4b). Third, the effect of KPS concentration on grafting was studied using a NIPAM concentration of 100 phr and reaction time of 12 h at 95 °C (Fig. 4c). When the KPS concentration was increased from 0.5 to 2 phr, grafting increased from 10 to 25%. Lastly, the effect of reaction time on grafting was investigated using a NIPAM concentration of 100 phr, KPS concentration of 1 phr, and reaction temperature of 95 °C (Fig. 4d). When the reaction time was increased from 3 to 12 h, grafting increased from 12 to 45%.

Fig. 4
figure 4

Grafting percentage as a function of a temperature, b NIPAM concentration, c KPS concentration, and d reaction time

Full size image

Thermal properties

The thermal properties of the DPNR and NIPAM-g-DPNR were characterized using DSC and TGA. The DSC thermograms showed the glass transition temperatures of DPNR and NIPAM-g-DPNR to be −62.1 and −61.3 °C, respectively (Fig. 5). Low glass transition temperatures are due to the amorphous nature of NR. PNIPAM showed a broad endothermic transition between 60 and 150 °C, corresponding to the glass transition temperature, which is consistent with values reported in the literature [48]. A weak, broad endothermic transition between 70 and 150 °C was also observed from the NIPAM-g-DPNR sample, confirming the presence of small amounts of PNIPAM in the graft copolymer.

Fig. 5
figure 5

DSC thermograms of DPNR, PNIPAM, and NIPAM-g-DPNR (18% grafting efficiency)

Full size image

From TGA thermograms, the DPNR started to lose weight at 320 °C, consistent with previous reports (Fig. 6) [38, 39, 49]. PNIPAM started to lose weight at 100 °C due to breakage of the labile amide functional groups. The thermogram of the NIPAM-g-DPNR sample was similar to that of DPNR, but slightly less stable due to the presence of PNIPAM. This confirmed the presence of both DPNR and PNIPAM in the NIPAM-g-DPNR sample.

Fig. 6
figure 6

TGA thermograms of DPNR, PNIPAM, and NIPAM-g-DPNR (18% grafting efficiency)

Full size image

Temperature responsiveness

The temperature responsiveness of NR and NIPAM-g-DPNR in water were characterized at temperatures of 28, 30, 32, 34, 36, and 38 °C. At 28 °C, the NIPAM-g-DPNR swelled by 12% of its original weight, or about ten times more than the DPNR, due to the presence of hydrophilic NIPAM units. The NIPAM-g-DPNR can be considered a hydrogel-like material. The percentage swelling of the NR did not change with temperature, whereas that of the NIPAM-g-DPNR started to decrease at a water temperature of 30 °C and became constant at 34 °C (Fig. 7). The LCST of NIPAM-g-DPNR was, therefore, determined to fall in the range 30–34 °C, which is similar to the coil-to-globule transition temperature of PNIPAM. These results suggest that the presence of NR did not significantly affect the LCST of the PNIPAM. This is consistent with reports on other PNIPAM-based materials [9, 50,51,52,53].

Fig. 7
figure 7

Water swelling percentage of DPNR and NIPAM-g-DPNR (18% grafting efficiency) in aqueous solutions at different temperatures

Full size image

The LCSTs of NR and NIPAM-g-DPNR were also determined from the temperature-controlled contact angle. At 28 °C, the contact angles of NR and NIPAM-g-DPNR were 98° and 63°, respectively, indicating a greater hydrophobicity of NR than NIPAM-g-DPNR (Fig. 8). The contact angles of the NR sample were almost constant across the measured temperature range. In contrast, the contact angles of the NIPAM-g-DPNR sample started to increase at the temperature of 32 °C reaching 97° at 38 °C, which is close to those of the NR sample. The increase in contact angle was due to an increase in hydrophobicity of the NIPAM units resulting from breakage of the hydrogen bonds between the water molecules and NIPAM. Based on this abrupt change in contact angles, the LCSTs were determined to be approximately 32–36 °C, similar to those determined by the water swelling method.

Fig. 8
figure 8

Contact angles of DPNR and NIPAM-g-DPNR (18% grafting efficiency) at different temperatures

Full size image

The temperature responsiveness of the DPNR and NIPAM-g-DPNR specimens was monitored on a hot plate, whose temperature was slowly increased from 25 to 40 °C over a period of 7 min. The NIPAM-g-DPNR sample changed its shape close to its LCST, whereas the DPNR sample retained its original shape throughout (Fig. 9). This indicates that NIPAM-g-DPNR remains temperature responsive in the solid phase.

Fig. 9
figure 9

Images of NIPAM-g-DPNR (18% grafting efficiency) and DPNR specimen on a hot plate

Full size image

Dye adsorption studies

Next, the absorption of indigo carmine dye by NIPAM-g-DPNR was investigated. Indigo carmine was chosen to represent a water-soluble anionic model drug. The rubber samples were incubated in indigo carmine solution for 1 week at room temperature. The Langmuir and Freundlich isotherms were used to investigate the interaction between the indigo carmine and the NIPAM-g-DPNR. The Langmuir adsorption isotherm is based on the assumption that a monolayer coverage is formed on the surface of the adsorbent whereas the Freundlich adsorption isotherm describes the characteristics of multilayer adsorption [54,55,56]. On the basis of the R-square (R 2) of these models (Fig. 10), the Langmuir model showed the highest value of R 2 as 0.754, being the best fit to the experimental data between the two models. Our analysis suggests that the adsorption of indigo carmine onto the surface of the NIPAM-g-DPNR was of the monolayer form. NIPAM-g-DPNR is thus a potentially useful temperature-responsive material for biomedical applications.

Fig. 10
figure 10

Plots of a Langmuir and b Freundlich isotherms from the adsorption of indigo carmine on NIPAM-g-DPNR (18% grafting efficiency)

Full size image

Conclusions

In this work, graft copolymers between NIPAM and DPNR were successfully prepared via reaction in emulsion using KPS as initiator. The graft materials were rubber-like with thermal properties close to pristine NR. The temperature responsiveness of the graft copolymers in water and solid state showed an LCST of 32–34 °C close to that of PNIPAM. The dye adsorption studies revealed the Langmuir isotherm, indicating monolayer coverage. Therefore, these temperature-responsive, hydrogel-like rubbers have many potential biomedical and sensing applications. This includes smart bandages and facial masks that release medicine or therapeutic ingredients upon contact with human skin. Other applications include smart tires for automobiles that change adhesion in response to temperature. The strategy presented in this work allows multi-stimuli-responsive rubbers to be prepared through reactions between multiple responsive materials. This greatly extends the properties of the rubbers.