A thermo-responsive hydrogel for body temperature-induced spontaneous information decryption and self-encryption

Abstract

A thermo-responsive poly(N-isopropylacrylamide) (PNIPAM) hydrogel, exhibiting an interesting phenomenon of an opaque-transparent-opaque transition in the successive processes of heating and cooling, is reported. It is fabricated by means of both the porogenic effect of hydroxypropyl cellulose and the cononsolvency effect of PNIPAM in a mixed solvent of dimethyl sulfoxide and water. After being mildly triggered by body temperature, the hydrogel is used to spontaneously decrypt the quick response code within 4 min and then autonomously encrypts the code again within 10 min at room temperature. The mechanism for the transient transparency of hydrogels during the quenching process has been elucidated.

---

Smart hydrogels can rapidly respond to external stimuli such as light, temperature, pH and magnetic fields by autonomously changing their structure and properties, and have been widely used in smart actuators,^1,2 biosensors,³ drug delivery and controlled release,⁴ and information encryption.^5,6 Because smart hydrogels can display reversible transmittance or shape changes in response to external stimuli, they can realize dynamic information output and enhance the security of stored information, attracting increasing attention for information encryption.^7–12 In addition, information encryption based on smart hydrogels belongs to physical encryption, which has the advantage of crypticity and is widely used.¹³ Different kinds of smart hydrogel materials and devices for information encryption have been reported recently. However, smart hydrogels such as light- or solvent-responsive ones usually need a specific short-wavelength light or solvent to decode the information and encrypt after removal or to replace the specific stimuli, which are impractical to apply in daily life. Therefore, developing a smart hydrogel with easy access to external stimuli to realize spontaneous decryption/encryption is highly desired.

Thermo-responsive hydrogels can respond to an easily accessible physical and thermal stimulus as a decryption key.^7,14,15 Poly(N-isopropylacrylamide) (PNIPAM) hydrogels are the most popular thermo-responsive hydrogels, which exhibit outstanding and reversible volume and transparency changes at temperatures around the lower critical solution temperature (LCST), and have been used in information encryption.^16–20 Chen et al. prepared a PNIPAM hydrogel that experienced a transparent-opaque-transparent transition stimulated by solvent. A polar solvent, such as ethanol, was used as an ink to record information, and the encrypting behaviour was adjusted by altering the ethanol evaporation time and concentration, and then the information was decrypted by water or thermal stimulation.¹⁷ Bai et al. prepared a thermo-responsive ion conductive PNIPAM based hydrogel with a wide and adjustable dual temperature response in the range of 0–18 °C (Krafft point) and 37–80 °C (LCST). Combining hydrogels with different response temperatures allows them to display various patterns at different temperatures.¹⁹ The PNIPAM hydrogels show the functions of reversible decryption and encryption, by means of additional special solvents or the environmental temperature. However, developing a PNIPAM hydrogel with spontaneous decryption and self-encryption is still a challenge.

In this work, PNIPAM hydrogels, which are originally opaque at room temperature and show an opaque-transparent-opaque variation during the successive heating-quenching processes, are prepared by ultraviolet (UV) initiated polymerization, by means of both the cononsolvency effect of PNIPAM in a mixed solution of dimethyl sulfoxide (DMSO)–water²¹ and the pore-forming effect of hydroxypropyl cellulose (HPC)²² (Fig. 1). After the hydrogel synthesis, the co-solvent DMSO and porogen HPC are removed by extraction in water. The effects of N-isopropylacrylamide (NIPAM) concentration, HPC content and solvent composition on hydrogel microstructure and properties were investigated.

Fig. 1 Schematic illustration of the preparation and responsive process of a PNIPAM hydrogel. (a) The precursor solution containing NIPAM, HPC, N,N’-methylenebisacrylamide (MBA) and 2-hydroxy-2-methylpropiophenone (Darocur 1173) in a mixture of DMSO and water, (b) UV irradiated polymerization for 1 min and then washed with water to give an opaque hydrogel, then the hydrogel shrinks during heating (c), becomes transparent firstly (d) and then recovers to opaque (b) during cooling.

Hydrogels are prepared with various NIPAM concentrations of 1.8, 2.7 and 3.6 M while with a constant HPC content of 1.5 wt% and a DMSO molar fraction in the mixed solvent of 0.55 (referred to as 1.8, 2.7 and 3.6 M hydrogel). All have interconnected pore structures throughout the cross-section, and the pore size obviously decreases with increasing NIPAM concentration (Fig. S1, ESI†). The 1.8 M hydrogel also has a loose and porous surface, whereas the 2.7 M and 3.6 M hydrogels have a dense skin layer. As the monomer concentration increases, the deswelling/swelling rates of the hydrogels decrease (Fig. S2c, ESI†), which is related to the smaller pores of the hydrogels. Compared with the 1.8 M and 2.7 M hydrogels, the 3.6 M hydrogel shows a significant increase and gradual decrease in transparency (discrepancy of 70%) during the rapid quenching process (transferred from water at 35 °C to 25 °C) followed by a sudden heating process (Fig. S2a and b, ESI†). The average monomer conversion (eqn (S1), ESI†) of all the PNIPAM hydrogels is in the range of 59–68%.

As a porogen, the feedstock amount of the HPC polymer affects the pore size and properties of the PNIPAM hydrogels. Hydrogels are prepared with a fixed monomer concentration of 3.6 M, a DMSO molar fraction of 0.55, and different amounts of HPC (0, 1.0, 1.5 and 2.0 wt%). All the hydrogels with HPC have a dense surface layer and an open pore structure, and the interconnectivity within the hydrogels is greater with an increasing HPC content (Fig. S3, ESI†). Compared with hydrogel without HPC (∼0.29 μm), the pore sizes of the other three hydrogels are 0.84, 0.86 and 0.9 μm, with HPC amounts of 1.0, 1.5 and 2.0 wt%, respectively. Correspondingly, the time for the hydrogels to reach the deswelling and swelling equilibria, as well as the time to recover their opaqueness during the rapid heating process is shorter (Fig. S4, ESI†). The peak of the transparency of the hydrogels decreases, as the HPC content increases.

The solvent composition affects the cononsolvency effect of PNIPAM in mixed solvents, and thus, has a significant impact on the pore size of the hydrogels. The PNIPAM hydrogels were prepared with a NIPAM concentration of 3.6 M, HPC content of 1.5 wt% and different DMSO molar fractions (referred to as X_DMSO hydrogel). With an increasing X_DMSO, the variation span of the relative diameter (RD) of hydrogels during the heating and quenching process is broadened; whereas the time to reach the deswelling and swelling equilibria becomes longer, as shown in Fig. 2a and c. Here, when the degree of shrinking or swelling of the hydrogel reaches 95% of the curve of diameter variation, the hydrogel is considered to approach the equilibrium state. The X_0.45, X_0.50, X_0.55 and X_0.60 hydrogels reach a deswelling equilibrium at 0.1, 0.8, 2 and 4 min; and a swelling equilibrium at 1.6, 6.1, 31 and 65 min, respectively. The shorter equilibrium time or the faster rate of deswelling/swelling originated from the higher interconnectivity of the pores inside the hydrogels, which results in lower diffusion resistances for the water molecules. All four hydrogels clearly show the dense surface layers but the cross sections show an open pore structure (Fig. S5, ESI†).

Fig. 2 Dynamic thermo-responsive properties and transparency changes of hydrogels prepared with different DMSO molar fractions with the constant NIPAM concentrations (3.6 M) and HPC content (1.5 wt%) during the successive heating (25 °C to 35 °C) and rapid quenching (35 °C to 25 °C) processes. (a) Optical images, (b) transmittance, and (c) relative diameter (RD). The scale bar is 1 cm.

The interconnected pores are attributed to both the thermo-induced and solvent-induced microscale phase separation in the precursor solution of the hydrogels. On the one hand, the porogen HPC polymer shrinks into globules by thermo-induced phase separation at a curing temperature near 40 °C (higher than its LCST of 34 °C²³), which is completely removed as a sacrificed template after curing, forming interconnected pores in the hydrogels.²² This was verified by the strong absorption peak of the C–O–C groups in the HPC polymer at a wavenumber of 1086 cm⁻¹ which disappears from the FTIR spectra of the PNIPAM hydrogels (Fig. S6, ESI†). However, the absorption peaks of the isopropyl group at 1365 cm⁻¹ and 1385 cm⁻¹, as well as those of the amide group (N–H) and the carbonyl group (CO) groups at 1538 cm⁻¹ and 1639 cm⁻¹, respectively, indicates that the solvent composition has no effect on the chemical composition of the hydrogels. Moreover, all four hydrogels have high water contents at equilibrium (>80%), and a high average compression strength and strain at break around 285 kPa and 57%, respectively, (Fig. S7, ESI†). The LCST of all the hydrogels is calculated to be approximately 30 °C, namely the temperature corresponding to the maximum slope of the equilibrium relative diameter-temperature (ERD-T) curve (Fig. S8, ESI†). On the other hand, the solvent-induced phase separation is attributed to the cononsolvency effect of PNIPAM in the DMSO–water mixed solvent. The degree of phase separation decreases as the molar fraction of DMSO increases from 0.45 to 0.60.²⁴ Therefore, the pore size and interconnectivity of the cross-section decreases (Fig. S5, ESI†). The absence of a strong peak for the sulfoxide groups (S=O) in DMSO at a wavenumber of 668 cm⁻¹ in the Raman spectra shows there is no DMSO residual in the hydrogel (Fig. S9, ESI†).

At 25 °C (t = 0 min during heating to 35 °C), all the hydrogels are opaque with an extremely low transmittance as shown in Fig. 2a and b, when compared to the X_1.0 hydrogel (Fig. S10, ESI†). Their transparency slightly increases as the DMSO molar fraction and/or the observed wavelength (400–800 nm) increase (Fig. S11, ESI†). The transparency is reduced after the hydrogels are suddenly heated to 35 °C (0 < t < 10 min). This is due to the disappearance of the hydration layer surrounding the PNIPAM chains after heating,²⁵ resulting in a more pronounced interface between the polymer and water, and thus there is a high level of light scattering. Due to the power (n_w/n_h − 1) < 0 in eqn (S7) (ESI†), the transmitted intensity ratio I/I₀ is positively correlated with the pore size d in the hydrogel. The smaller the pore size, the lower the I/I₀, the stronger the light scattering, and the lower the hydrogel transparency. However, during the subsequent quenching process to 25 °C (10 < t < 90 min), the hydrogel transparency instantly and dramatically increases at first (see Movie S1, ESI†) to a high level, and then gradually decreases to the similar states as the original ones. Simultaneously, as the X_DMSO increases, the transparency increases and the time required to return to the opaque state also increases. The transmittance of the X_0.45, X_0.50, X_0.55 and X_0.60 hydrogels measured at a wavelength of 600 nm after rapid quenching for 0.5 min (the shortest time to get the data) is 0.3%, 26.8%, 69.9%, and 85.1%, respectively, and it gradually decreases to 0.1% (6 min), 0.2% (6 min), 12.0% (20 min) and 30.0% (40 min), respectively.

The mechanism for the special opaque-transparent-opaque transition during the heating/cooling process were determined in combination with the equilibrium states at high and low temperatures, and the dynamic microstructure change of the hydrogels during cooling. In detail, during sudden cooling, the hydration layer surrounding the PNIPAM polymeric chain is re-generated,²⁶ resulting in the interface elimination between the polymer and water; simultaneously, due to the wide spectrum of size variation, the pore size of the hydrogels at the beginning of the cooling process or at the end of the preceding heating process (t = 10 min) decreased to a low level, assuming there was one order of magnitude of light wavelength. Therefore, the scattering effect of the light is significantly reduced due to both the interface elimination and the small pore size, and the hydrogels become transparent. As the cooling continues, the internal pore size of the hydrogels gradually increases due to the hydrophilic swelling of the polymer network at a temperature lower than the LCST. In this case, the I/I₀ value decreases with the increase of the pore size d according to eqn (S7) (ESI†), and thus the degree of light scattering increases, and the hydrogels gradually become opaque. To discover the microstructures of the hydrogels during the cooling process, the cross-sectional SEM images of the hydrogels were obtained after rapid quenching for 0.5 min, 6 min, 20 min and 40 min (Fig. 3). At 10.5 min (cooling for 0.5 min), the cross-sections of all four hydrogels have a “sandwich” structure (Fig. 3a1–d1), and the thickness of the middle layer increases but the pore size decreases with the increasing amounts of X_DMSO (Fig. 3a2–d2). This is because the upper and lower surfaces of the hydrogel sheet has first contact with the water at 25 °C, and begin to swell earlier than the middle part, and thus the swelling degree of the middle parts is smaller than that of the outer parts at the same time. Although the pore size of the hydrogels at 10.5 min is much smaller than that at original state (0 min, Fig. S5a2–d2, ESI†) after experiencing the heating, the pore sizes of the X_0.45 and X_0.50 hydrogels was large enough to scatter light and thus the hydrogels remained opaque (Fig. 2a) during cooling. As the cooling further continues, the middle part of the hydrogels gradually swells, and the sandwich structure disappears (Fig. 3a3–d3) and the corresponding pore sizes recovers to their original states (Fig. 3a4–d4 and Fig. S5a2–d2, ESI†). The hydrogels become opaque, macroscopically at the 16 min (X_0.45), 16 min (X_0.50), 30 min (X_0.55) and 50 min (X_0.60), see Fig. 2a.

Fig. 3 Cross-sectional SEM images of hydrogels prepared with different DMSO molar fractions after rapid quenching for 0.5 min (a1–d1, a2–d2) and for a certain period of time (a3–d3, a4–d4). The scale bars are 200 μm (a1–d1, a3–d3) and 10 μm (a2–d2, a4–d4). (a) X_0.45, (b) X_0.50, (c) X_0.55, (d) X_0.60.

After comprehensively considering both the larger transparency discrepancy and the faster response rate during cooling, the X_0.55 hydrogel was selected for testing the repeatability, and the results are shown in Fig. 4a. During the consecutive five heating/cooling circles between 25 °C and 35 °C, the RD values and the transmittance of the hydrogel were easily reversible. During the 5th cycles, the average transmittance of the hydrogel after cooling for 0.5 min at 600 nm was about 70%. The RD-t curves in the five cycling experiments were consistent, and the hydrogels reached deswelling and swelling equilibria at approximately the same time and to the same degree. The RD value of this hydrogel was shrinking to 0.74 at 2 min during heating, and swelling to 0.99 at 30 min during cooling.

Fig. 4 Repeatability of hydrogel and its applications in information encryption. (a) RD and transmittance change of the X_0.55 hydrogel during 5 heating/cooling cycles. (b) Decryption process after heating by body temperature and self-encryption process of hydrogels at room temperature.

Since the as-prepared X_0.55 hydrogel are provided with an opaque-transparent-opaque transition during consecutive heating and cooling process, it is used to demonstrate the application in information decryption at body temperature and self-encryption at room temperature (Fig. 4b). The information encryption device with a hidden quick response (QR) code is covered by a piece of X_0.55 hydrogel. At room temperature at about 25 °C, the QR code in the device is invisible due to the opaque property of the X_0.55 hydrogels (t = 0 min in Fig. 2a). To decrypt the code, two fingers (body surface temperature ∼35 °C) were used to heat the device for 2 min, so that the hydrogel shrinks and becomes more opaque, and then the whole device was left in a 25 °C air environment. The cooling process is recorded and shown in Fig. 4b. At t = 0 min, the QR code is also invisible due to the decrease of transmittance of the hydrogels after heating (t = 10 min in Fig. 2a). As the time increases, the transparency of the hydrogel gradually increases, and the QR code gradually appears. At 4 min, the QR code can be scanned with a smartphone and to obtain the hidden information. With a further increase of time, the hydrogel transparency begins to decrease, and the QR code is no longer scannable after 10 min, and completely hidden again at 90 min. Such a spontaneous decryption/self-encryption process is quite efficient and feasible with mild energy input.

In summary, a PNIPAM hydrogel with opaque-transparent-opaque transition during successive heating and cooling processes is developed. Both the porogen HPC polymer to construct the micrometer pores based on thermo-induced phase separation and co-solvent DMSO to generate the submicron pores based on solvent-induced weak phase separation are indispensable. The optimized hydrogel reaches the equilibria of deswelling and swelling in 2 min and 30 min, respectively, with satisfactory repeatability. The mechanism for the transient transparent state during the cooling process is attributed to the reduced light scattering, which in turn originates from the regenerated hydration layer surrounding the PNIPAM polymeric chain as well as the small pore size reached after heating. This study provides guidance for the development of novel smart materials in the field of information encryption.

The authors gratefully acknowledge support from the Sichuan Province Science and Technology Support Program (2019JDJQ0026).

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. C. Yao, et al. , Adv. Funct. Mater., 2015, 25, 2980–2991.
2. A. K. Mishra, et al. , Sci. Robot, 2020, 5, eaaz3918.
3. Z.-H. Jia, et al. , Macromol. Rapid Commun., 2021, 42, 2100200.
4. M. Santhamoorthy, et al. , Polymers, 2022, 14, 4128.
5. H. Shang, et al. , Adv. Optical Mater., 2022, 10, 2200608.
6. J. Hai, et al. , Angew. Chem., Int. Ed., 2018, 57, 6786–6790.
7. Y. Sun, et al. , Adv. Mater., 2022, 34, 2201262.
8. F. Shan, et al. , ACS Appl. Mater. Interfaces, 2023, 15, 7405–7413.
9. J. Deng, et al. , ACS Appl. Mater. Interfaces, 2021, 13, 39967–39975.
10. C. Yang, et al. , ACS Appl. Mater. Interfaces, 2021, 13, 36538–36547.
11. H. Wen, et al. , J. Mater. Chem. C, 2021, 9, 2455–2463.
12. H. Lu, et al. , Adv. Funct. Mater., 2022, 32, 2206912.
13. M. Li, et al. , Small, 2022, 18, 2205359.
14. X. Yang, et al. , Adv. Funct. Mater., 2022, 32, 2205177.
15. F. Yin, et al. , Macromol. Rapid Commun., 2022, 43, 2200401.
16. L. Tang, et al. , Prog. Mater. Sci., 2021, 115, 100702.
17. Z. Chen, et al. , Adv. Funct. Mater., 2022, 32, 2201009.
18. Y. Yang, et al. , Biomacromolecules, 2023, 24, 1522–1531.
19. L. Bai, et al. , Chem. Eng. J., 2023, 456, 141082.
20. X. Zhang, et al. , Chem. Eng. J., 2023, 468, 143734.
21. H. Yamauchi and Y. Maeda, J. Phys. Chem. B, 2007, 111, 12964–12968.
22. X. S. Wu, et al. , J. Polym. Sci., Polym. Chem., 1992, 30, 2121–2129.
23. Y. Feng, et al. , ACS Appl. Mater. Interfaces, 2023, 15, 5836–5844.
24. Y. Alsaid, et al. , Adv. Mater., 2021, 33, 2008235.
25. Y. Ono and T. Shikata, J. Am. Chem. Soc., 2006, 128, 10030–10031.
26. F. M. Winnik, Macromolecules, 1990, 23, 233–242.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4cc01349b

---