Transient Chemical Activation of Covalent Bonds for Healing of Kinetically Stable and Multifunctional Organohydrogels

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE5 Apr 2022Transient Chemical Activation of Covalent Bonds for Healing of Kinetically Stable and Multifunctional Organohydrogels Liangying Jia, Lin Xu, Yaqing Liu, Jingcheng Hao and Xu Wang Liangying Jia National Engineering Research Center for Colloidal Materials and Key Laboratory of Colloid and Interface Chemistry of the Ministry of Education, School of Chemistry and Chemical Engineering, Shandong University, Jinan, Shandong 250100 Google Scholar More articles by this author , Lin Xu National Engineering Research Center for Colloidal Materials and Key Laboratory of Colloid and Interface Chemistry of the Ministry of Education, School of Chemistry and Chemical Engineering, Shandong University, Jinan, Shandong 250100 Google Scholar More articles by this author , Yaqing Liu National Engineering Research Center for Colloidal Materials and Key Laboratory of Colloid and Interface Chemistry of the Ministry of Education, School of Chemistry and Chemical Engineering, Shandong University, Jinan, Shandong 250100 Google Scholar More articles by this author , Jingcheng Hao National Engineering Research Center for Colloidal Materials and Key Laboratory of Colloid and Interface Chemistry of the Ministry of Education, School of Chemistry and Chemical Engineering, Shandong University, Jinan, Shandong 250100 Google Scholar More articles by this author and Xu Wang *Corresponding author: E-mail Address: [email protected] National Engineering Research Center for Colloidal Materials and Key Laboratory of Colloid and Interface Chemistry of the Ministry of Education, School of Chemistry and Chemical Engineering, Shandong University, Jinan, Shandong 250100 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.022.202101536 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail It remains a great challenge to balance the kinetic stability and intrinsic healing ability of polymer materials. Here, we present an efficient strategy of using a synthetic reaction cycle to regulate the intrinsic healing ability of thermodynamically stable and kinetically inert multifunctional organohydrogels. By combining a double decomposition reaction with spontaneous energy dissipation, we can construct the simplest synthetic reaction cycle that can induce a transient out-of-equilibrium state for achieving the healing of organohydrogels with kinetically locked acylhydrazone bonds. In addition to balancing kinetic stability and healing ability, the synthetic reaction cycle also enables the polymer materials to have high tolerance to organic solvents, high ionic strength, high and low temperatures, and other harsh conditions. Therefore, the kinetically stable and healable organohydrogels remain mechanically flexible and electrically conductive even down to −40 °C and are readily recyclable. The integration of chemical networks into healable polymers may provide novel, versatile materials for building next-generation electronic devices. Download figure Download PowerPoint Introduction As the most perfect material created by nature, the living organism, or a part of it, is a complex system integrating multiple functions into one whole entity. Taking the largest organ of the human body (skin) as an example, its functions include protection, body temperature regulation, sensing, breathing, and self-healing.1–3 The realization of most functions relies on the supply of chemical fuels to drive the formation of a transient out-of-equilibrium state at either macro or micro level.4–6 The effective consolidation of multiple functions in a synthetic material is also the goal that scientists have made great efforts to pursue. For synthetic gel materials widely used in the medical,7,8 electronics,9–11 and other fields, an ideal material would possess a collection of multiple functions or properties such as mechanical strength and flexibility, anti-freezing, electrical conductivity, high kinetic stability and inertness, and self-healing. Among these properties, the kinetic stability and self-healing ability are a pair of key elements that need to be seriously considered and balanced. The emergence of self-healing materials that contain microencapsulated12 and microvascularized2 healing agents reconciles the demands of kinetic stability and healing ability in both solid polymers and composites. However, the mechanical mismatch between the filling materials and bulk materials significantly restricts the application of these healing strategies in soft gels. In contrast, the self-healing strategies based on reversible covalent bonds or supramolecular interactions are particularly suitable for improving the reliability of gels, not only because of their simple material fabrication procedures but also due to their effectiveness in dealing with multiple local healing events.4,13–15 Spontaneous intrinsic self-healing is widely preferred,15–18 but the weaknesses of such a strategy are the dynamic chemistry-related poor kinetic stability and inertness that leads to noticeable shape change, creep, and unwanted fusion of pristine materials (Scheme 1a)19,20 and a dramatically decreased healing ability as residence time increases after damage.21–23 Although external interventions such as light and heat can assist inactive solid films and coatings to generate a transient healing ability,13,24,25 and thus avoid the above-mentioned problems, such intervention methods not only accelerate the volatilization of the liquid in gels but also have low efficiency in interior healing of large bulk gels.24,26 Overall, it still lacks an effective way to achieve the healing of kinetically stable gels, as well as the development of reliable gels that integrate multiple desired functions. Scheme 1 | Schematic comparison of three different healing methods for gels. (a) Conventional healable gels based on permanently labile chemical bonds are spontaneously healable but usually lack kinetic stability and inertness. (b) Kinetically stable and healable hydrogels regulated by biochemical reactions have poor tolerance to harsh conditions. (c) Kinetically stable and healable gels regulated by a completely artificial reaction cycle in this work possess high tolerance to harsh conditions. Download figure Download PowerPoint Inspired by the hierarchically and temporally controlled wound healing process in biological systems, we recently approached to reconcile the contradiction between the kinetic stability and healing ability in hydrogels via kinetic control of bioactivity-regulated competing reactions and/or dissipative processes (Scheme 1b).19,26–29 Bioactive elements such enzymes and baker’s yeast play significant roles in mediating the transient out-of-equilibrium states on the damage site of polymer hydrogels for efficient structural healing and property recovery.19,26–28 Though bioactive ingredients are essential for wound healing in biological systems,5,30 their fragile nature may hinder the development of advanced, versatile synthetic materials. However, abiotic chemical fuels are not subject to this restriction. Recently, abiotic chemical fuels have been highly regarded in the field of dissipative self-assembly, which is a transient molecular behavior inspired by fuel-driven self-assembly in living organisms.31–34 For example, van Esch and colleagues34 activated certain molecular building blocks and achieved their transient self-assembly to form regenerative dynamic gels by employing artificial reaction cycles that are much cheaper and easier to operate than biological components. However, to the best of our knowledge, few examples of healing in multifunctional and kinetically stable gel materials regulated by artificial reaction cycles have been reported. In what follows, we describe a systematic method to regulate the intrinsic healing ability of kinetically stable and inert, anti-freezing, conductive, and recyclable organohydrogels by employing a double-decomposition-reaction-based artificial reaction cycle. The double decomposition reaction combined with spontaneous energy dissipation constructs the simplest artificial reaction cycle yet discovered for temporally programming the healing ability of kinetically stable and multifunctional organohydrogels. The robust integration of all these above-mentioned functions and properties in an organohydrogel system benefits from the use of such an artificial reaction cycle with high tolerance to harsh conditions (Scheme 1c). Experimental Methods Materials Diacetone acrylamide (DAAM), acrylamide (AM), 2,2-azoisobutyronitrile (AIBN), and 2-(dodecylthiocarbonothioylthio)-2-methylpropionic acid (CTA) were purchased from the Aladdin Reagent Company (Shanghai, China), and the AIBN was recrystallized from ether before use. Adipic dihydrazide (ADH) was purchased from the Energy Chemical Co. (Shanghai, China) and bromothymol blue (BTB) was purchased from the Kemiou Chemical Reagent Co., Ltd. (Tianjin, China). Finally, glycol and all other reagents were purchased from the Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China), and the above reagents were used as received unless otherwise specified. Polymer synthesis The polyacrylamide-co-poly(diacetone acrylamide) (PAM-co-PDAAM) polymer was synthesized according to our previous method,35 with some modification. Specifically, AM (7.28 g, 0.1024 mol), DAAM (1.92 g, 0.0113 mol), CTA (108.0 mg, 0.296 mmol), AIBN (9.8 mg, 0.0597 mmol), and dimethyl sulfoxide (50 mL) were mixed in a round-bottom flask, and the solution was bubbled with N2 gas for 30 min. The flask was then immersed in an oil bath at 70 °C and stirred for 24 h, and the obtained mixtures were precipitated in acetone (3×) and dried under vacuum at room temperature (RT) overnight to yield the PAM-co-PDAAM polymer. Fabrication of hydrogels and organohydrogels Aqueous solutions of PAM-co-PDAAM polymers, ADH, and BTB were uniformly mixed at pH 4 to reach final concentrations of 162, 15, and 0.1 mg/mL, respectively. The mixtures were then injected into a Teflon mold to fix their shape. The original hydrogels were formed in a few minutes at pH 4, and our pristine hydrogels with high kinetic stability were obtained by immersing the original hydrogels in 0.03 M sodium bicarbonate (NaHCO3) solution for 20 h. The organohydrogels with high kinetic stability were obtained by immersing the original hydrogels in various glycol-water mixtures containing NaHCO3 and sodium chloride (NaCl) for 20 h. The glycol volume fractions of the glycol–water mixtures were varied to be 0%, 30%, 60%, and 90%, and the resulting organohydrogels were defined as 0%, 30%, 60%, and 90% organohydrogels, respectively. Time-dependent weight variations of hydrogels and organohydrogels The hydrogels and organohydrogels were put into containers with different relative humidity (30%, 60%, and 100%), and the weights of the gels were recorded as the time increased. The weight variation of the gels was calculated by the equation ΔWt/W0 = (Wt − W0)/W0, where Wt is the weight of the gels at time t and W0 is the initial weight of the gels. Healing of hydrogels and organohydrogels Pristine hydrogel or organohydrogel with a size of 2.0 cm × 1.0 cm × 0.25 cm were cut into two completely separated pieces with a blade, and the healing of the damaged gels was achieved by applying 4 μL of chemical fuels (0.8 M citric acid-sodium citrate buffer, pH 3.4) to the fractured site (1.0 cm × 0.25 cm). The hydrogels, as well as the 0% and 30% organohydrogels, were healed under 100% relative humidity, while the healing of the 60% and 90% organohydrogels were conducted under ambient conditions. We used a blunt-edged probe from a TMS-Pro texture analyzer (Food Technology Corp., United States) to compress the middle of the pristine gels and the healed region of the healed gels to measure the mechanical strength of the samples. The test was conducted under a constant speed of 15 mm/min until complete fracture of the samples. After this, the fracture strengths of the pristine (Sp) and healed (Sh) gels were recorded and the healing efficiency was calculated as 100% × Sh/Sp. Determination of carbon dioxide generation kinetics in aqueous solutions A closed round-bottom flask containing 10 mL NaHCO3 aqueous solutions (0.8 M), a closed conical flask containing saturated NaHCO3 aqueous solutions, and an empty measuring cylinder were connected with two rubber tubes, and then 0.5 mL of different concentrations of chemical fuels (1, 1.5, and 2 M) were quickly injected into the round-bottomed flask, and the volume of saturated NaHCO3 aqueous solutions collected in the measuring cylinder was recorded in real time. Here, the volume of saturated NaHCO3 aqueous solutions collected in the measuring cylinder was approximated to the volume of carbon dioxide (CO2) generated by the reaction between NaHCO3 and chemical fuels. Measurement of BTB solubility BTB molecules (0.1 g) were dissolved in each of the 0%, 30%, 60%, and 90% solutions (100 g) at 20 °C (the solutions containing NaHCO3, NaCl, water, and glycol are designated as x% solutions, where x is the volume fraction of glycol in the glycol–water mixed solutions). The above solutions were then filtered to remove the undissolved BTB molecules to obtain the saturated BTB solutions. The saturated solutions were then diluted by 50 times before UV–vis spectroscopy analysis. We calculated the solubilities of BTB in the 0%, 30%, 60%, and 90% solutions according to the calibration curves of the absorbance of BTB in each of the solutions at λ = 616 nm. Organohydrogel touch strips We adapted a surface-capacitive system for our touch pad as a touch-sensing system. In our experiment, each end of the organohydrogel strip (7.5 cm × 1 cm × 0.25 cm) was connected to a copper electrode and a current meter, then connected to an alternating current (AC) power source (AFG 1062, Tektronix, United States). The AC voltage and operating frequency were ±0.6 V and 17 kHz, respectively, and we collected the measured currents of I1 and I2 on two current meters of A1 and A2 through two independent channels in a data acquisition card (Model 7700 Multiplexer Modules, KEITHLEY, United States). A software package (Kickstart, KEITHLEY) was employed to read the current signals by time. Because identical voltage was applied to all sites of the touch pad in the surface-capacitive system, a uniform electrostatic field across the touch pad was created. The touch pad system works as follows. When a finger, as a conductor, touches the touch pad, the touch point becomes grounded, and a potential difference is induced between the touch point and the electrode. The potential difference allows the current to flow from the electrode through the finger, and the magnitude of the current varies inversely with the change in the distance between the touch point and the electrode. Characterizations The 1H NMR spectrum was recorded on a Bruker Avance 400 spectrometer (Bruker, Switzerland) at 400 MHz. The number-average molecular weight (Mn) and dispersity (Đ = Mw/Mn) of the polymer were determined by using a size-exclusion chromatography system equipped with a Waters 1515 isocratic high-performance liquid chromatography pump, an ULTRAHYDROGEL 120 PKGD, and a Waters 2414 refractive index detector (Waters, United States). The eluent was H2O with 0.1 M NaNO3 at a flow rate of 1.0 mL/min. Polyethylene glycol was used as the standard. The scanning electron microscopy (SEM) image was observed using a Gemini 300 (Zeiss, Germany) field-emission SEM, where the sample was lyophilized before analysis. Rheological experiments were conducted on a HAAKE RheoStress 6000 rheometer (HAAKE, Germany) with a cone-plate sensor system (C35/1° Ti L07116, diameter: 35 mm, core angle: 1°) at 25 °C with a frequency range from 0.1 to 100 Hz at 10 Pa. All dynamic rheological measurements were performed on the linear viscoelastic regimes. The freezing points of the samples were measured by differential scanning calorimetry (DSC) on a NETZSCH DSC3500 instrument (NETZSCH, Germany), and the cooling–heating process was conducted at a rate of 10 °C/min. Additionally, UV–vis spectra of the samples in 96-well plates were measured using a microplate reader (Spark®, TECAN, Switzerland). Impedance tests were carried out by sandwiching the samples with two stainless steel electrodes on an electrochemical workstation (CHI760E, Shanghai Chenhua Instrument Co., Ltd., China), and the resistivity of the samples was calculated using the equation ρ = RS/L, where S is the contact area between the stainless steel electrode and the sample, R is the resistance of the sample, and L is the thickness of the sample. Finally, the electrical signals of the sample during the damage-healing process were recorded using the electrochemical workstation. Statistical analysis All data are presented as mean ± standard deviation (n = 3) unless otherwise stated. Results and Discussion We demonstrated our concept by employing a kinetically stable acylhydrazone-based gel system, which was built by the cross-linking of PAM-co-PDAAM (the characterizations can be found in Supporting Information Figures S1 and S2 and Table S1) with ADH, followed by NaHCO3 addition. The presence of NaHCO3 in the gels provides a basic environment that is critical for deactivating the acylhydrazone bonds to produce thermodynamically stable and kinetically inert gels to avoid unwanted fusions. When healing or fusion of the gels is needed, the acylhydrazone bonds can be temporarily activated by local acid (chemical fuel) treatments (Scheme 2). The acid treatments facilitate the structural healing of the gels and trigger the occurrence of the double decomposition reaction to produce CO2 that can slowly release to the atmosphere to recover a basic pH with an overall gain in entropy. Most importantly, both the structure and kinetic stability of the gels can be completely restored by undergoing this purely chemical process. Notably, the chemical fuel can be applied locally to the fracture surfaces of the damaged gels before the pieces are brought back into close contact for healing ( Supporting Information Figure S3a). This fuel application method leads to a transient activated state of the damage surfaces that survive for a few minutes to several hours, depending on the composition of the chemical fuel, which means that the entire fracture surface becomes uniformly activated. Thus, the healing is independent of the size of the materials. In contrast, when healing is triggered by heat or light, the damaged pieces must first be brought close together ( Supporting Information Figure S3b) due to the poor activation persistence after removal of the light and heat sources. Therefore, although light and heat can effectively trigger the healing in coatings or films,13,25 they may not be good candidates to trigger the healing of large bulk hydrogels due to the limited penetration depth in materials. Scheme 2 | Schematic illustration of the healing of kinetically stable gels via transient chemical activation of acylhydrazone bonds, which is regulated by a double-decomposition-reaction-based synthetic reaction cycle. Download figure Download PowerPoint The PAM-co-PDAAM polymer aqueous solution in our experiment was mixed with the cross-linker (ADH) and deactivator (NaHCO3) in two different ways. First, simultaneously mixing the three components led to the formation of a liquid with a low viscosity (Figure 1a) because the cross-linking reaction was inhibited under high pH in the presence of NaHCO3. The pH of the mixture solution in this case was ∼9, which we determined by colorimetry using BTB (its chemical structure can be found in Supporting Information Figure S4a) as a pH indicator (Figure 1b). Second, the cross-linking of the polymer by ADH was first achieved at pH 4 to obtain a yellow hydrogel (the color arose from the pre-stored BTB under acidic conditions), and a post-immersion process was then used to allow the deactivator to diffuse into the bulk materials to produce polymer hydrogels with kinetically locked acylhydrazone bonds (Figure 1a). Figure 1 | Fabrication and healing of the hydrogels. (a) Photographs showing the fabrication of kinetically inert hydrogels and a SEM image showing the microstructure of the hydrogels. (b) Photographs of the BTB-containing hydrogels at different pH levels. (c) Rheological characterizations of the original and pristine hydrogels. (d) Photographs showing the poor kinetic inertness of the original hydrogels (the top row), and the high kinetic inertness of the pristine hydrogels before (the middle row) and after (the bottom row) healing. (e) Time-dependent modulus changes of the aqueous systems containing PAM-co-PDAAM and ADH at different pH levels. (f) Healing processes of the kinetically inert hydrogels by treating with different concentrations of acidic buffers (pH 3.4). (g) Representative compressive stress versus strain curves of the hydrogels before damage and after damage and healing for 12 h by treating with different concentrations of acidic buffers. (h) Healing efficiency for the hydrogels treated with different concentrations of acidic buffers. Download figure Download PowerPoint For ease of understanding, the hydrogels before and after NaHCO3 addition are designated as original and pristine hydrogels, respectively (details of the samples can be found in Supporting Information Table S2). The SEM image in Figure 1a reveals that the pristine hydrogels had a porous network structure with a pore size of ∼150 μm. Notably, BTB played an important role in the timely monitoring of the pH change of the hydrogels during the post-immersion process. The addition of BTB did not change the mechanical, pH activation, or healing properties of the gels because the BTB amount was sufficiently low in every case. However, some of the pre-stored BTB molecules in the samples escaped when they were immersed in NaHCO3 aqueous solution. The BTB retention in the pristine hydrogels was calculated to be ∼70%, based on a pre-established calibration curve ( Supporting Information Figure S4b). Figure 1c and Supporting Information Figure S5 show that both the storage moduli (G′) and the mechanical strength of the hydrogels slightly decreased after immersion in NaHCO3 solutions because the weight and volume of the hydrogels increased ( Supporting Information Figure S6), accompanied by a decrease in hydrogel solid content. Supporting Information Figure S7 shows that the complex viscosity of the original and pristine hydrogels decreased as frequency increased, revealing a shear-thinning behavior stemming from polymer network structure shearing force damage. The original hydrogels without NaHCO3 showed poor kinetic stability, and two original hydrogels fused into one whole piece after contacting each other for 5 h (Figure 1d, the top row) due to the highly active acylhydrazone exchange under low pH conditions. Though the fusion of bulk polymers with poor kinetic inertness has been academically studied in a disulfide-based system,36 high kinetic stability and inertness have always been a pursuit for polymer materials for many potential applications.27,37 Here, our pristine hydrogels containing NaHCO3 are typical bulk materials that have high kinetic stability as evidenced by the fact that two pristine hydrogels could be easily separated after contacting each other for 5 h (Figure 1d, the middle row). We further investigated the kinetics of acylhydrazone-based hydrogel formation under different pH levels to obtain quantitative data to better understand the pH-dependent kinetic stability and inertness of our hydrogels. To accomplish this, PAM-co-PDAAM and ADH were mixed in situ on a rheometer, and the time-dependent modulus changes were recorded. This method has been previously adopted by Anseth and colleagues38 to study the modulus evolution of bis-aliphatic hydrazine hydrogels. Figure 1e shows that the G′ of the sample increased rapidly and reached 10 kPa within 10 min at pH 4. In addition, the crossover time of G′ and G″ was observed to be only 2 min for this sample, suggesting the rapid formation of acylhydrazone bonds under low pH. The crossover time of G′ and G″ grew as pH increased under acidic conditions (6 and 30 min for pH 5 and 6, respectively). In contrast, the moduli of the samples remained at low levels during the whole measurement process when pH ≥ 7, implying that the cross-linking reaction was significantly inhibited in neutral and basic environments. The above results demonstrate that the formation of acylhydrazone bonds was catalyzed by acids and that the activity for acylhydrazone bonds decreased with increasing pH under acidic conditions. At pH ≥ 7, the acylhydrazone bonds became completely inactive, and these conclusions are further supported by the time-dependent complex viscosity variation ( Supporting Information Figure S8). Clearly, the difference in the activity of acylhydrazone bonds between low and high pH levels enables the activation of kinetically stable materials by relying on acid treatments for self-healing purposes. The pristine hydrogels containing NaHCO3 had high kinetic stability and inertness, but they could not heal themselves spontaneously. To obtain the transient healing ability, acidic buffers with different concentrations were applied locally to the fracture surfaces of the damaged NaHCO3-containing hydrogels. Figure 1f shows that stretching the healed hydrogels treated with 0.8 and 1 M chemical fuels led to ruptures at regions different than the original damage sites, suggesting that the hydrogels completely recovered their mechanical strength when treated with high concentrations of chemical fuels. In contrast, the healed hydrogels treated with 0.5 M acidic buffers broke at the original-damage sites when a tension force was applied, indicating that the hydrogels failed to heal completely in this case. Notably, the healing of the hydrogels was conducted under 100% relative humidity to avoid the water loss during the healing process ( Supporting Information Figure S9). We further studied healing efficiency with compression experiments. Healing efficiency reached over 90% when the damaged hydrogels were treated with 0.8 and 1 M chemical fuels but was only 67% when 0.5 M chemical fuels were used (Figures 1g and 1h) because the addition of adequate acids locally was necessary to decrease the pH of the fracture surface and activate the local acylhydrazone bonds for structural healing. Notably, the acid treatments only temporarily decreased the local pH of the damage site for structural healing. In addition, the NaHCO3 molecules in the hydrogels slowly migrated to the damage site and neutralized the acids to restore the kinetic stability of the hydrogels. The restoration of the pH also arose from the release of CO2 produced by the double decomposition reaction between NaHCO3 and the acids (Scheme 2). In our system, no visible bubbles were observed during the healing process because the CO2 liberation was slow ( Supporting Information Figure S10). Furthermore, the CO2 liberation did not change the mechanical strength of the hydrogels. Although both 0.8 and 1 M acidic buffers resulted in efficient healing, the dissipation for 0.8 M only took 5 h but took 12 h for 1 M. Therefore, 0.8 M acidic buffers were chosen as the optimal concentration. The healed regions of two healed hydrogels were put in contact with each other for 5 h to confirm their kinetic stability, and the results show that the two healed hydrogels could be simply separated (Figure 1d, the bottom row), suggesting superior kinetic stability of the healed samples. These results support the feasibility of using a competing chemical process to balance the kinetic stability and healing ability of polymer hydrogels. Hydrogels have inferior freezing tolerance, and they become rigid and fragile when stored at low temperatures.10,39 Our kinetically stable hydrogel system also exhibits freezing intolerance ( Supporting Information Figure S11). To overcome this shortcoming, glycol was introduced