Bio-inspired chitosan/polyvinyl alcohol/tannic acid hydrogels as plant grafting wood adhesives

Adhesives are widely used in various industries, such as construction, electronics, packaging, biomedical devices, and, particularly, wood industries.^1–5 Numerous industrial adhesives, such as formaldehyde-based amino resins, phenolic resins, isocyanate-containing adhesives, epoxy resins, and polyvinyl acetate composite adhesives, have been developed as wood adhesives.^5–7 For example, amino and phenolic resins exhibit remarkable mechanical properties, such as hardness and modulus of elasticity, when cured.⁵ Although the moduli of elasticity of isocyanate-containing adhesives, epoxy resins, and polyvinyl acetate composite adhesives are lower than those of amino or phenolic resins, they are still widely used in wood manufacturing and processing.⁵ However, most wood adhesives are prepared using petroleum-based resources and have considerable safety issues owing to the presence of toxic substances.^2,8–10 Therefore, the development of eco-friendly wood adhesive materials is highly desirable.

Polyphenols, which are abundant in the plant kingdom, have attracted considerable attention because of their adhesive properties on various substrates.^11–13 Polyphenols are rich in hydroxyl groups that are covalently linked to aromatic rings.^11–16 An example of a polyphenol is tannic acid (TA), a naturally occurring hydrolyzable condensed tannin.^14–17 TA is used as a component of adhesives owing to the presence of multiple galloyl groups, which facilitate binding to various organic and inorganic molecules via physical and chemical interactions.^18–21 The addition of TA to polymer solutions, including chitosan, gelatin, agarose, polyvinyl alcohol, polyethylene glycol, poly(N-isopropyl acrylamide), and Pluronic, promotes spontaneous solidification by intermolecular hydrogen bonding.^22–29 In addition, the addition of metal ions (i.e., Fe³⁺ ions) to the TA/polymer mixtures improved their mechanical properties and adhesive forces through coordination bonds.^20,30,31 Furthermore, TA/polymer composites exhibit oxidative cross-linking when oxidants, such as sodium periodate, are applied to the mixture.^31,32 Therefore, TA-based adhesives are promising candidates for application in the wood industry.

TA and polyvinyl alcohol (PVA) composites have been developed for various applications.^25,33,34 Owing to the hydrogen bonds between the hydroxyl groups of PVA and gallol groups of TA, the PVA/TA mixture exhibited excellent mechanical properties with rapid gelation.²⁵ In addition, self-assembled PVA/TA hydrogels with diverse and unique microstructures can be fabricated by controlling their concentration and temperature.²⁵ In addition, the PVA/TA hydrogels demonstrated nontoxic underwater adhesive properties between two stainless steel objects and maintained adhesive properties (∼100%) even after ten repetitions of attachment–detachment experiments.³³ Along with the PVA/TA composites, the addition of another component (polydopamine, borax, or 2-hydroxypropyl trimethyl ammonium chloride chitosan) to the PVA/TA networks improved the mechanical properties and adhesiveness of the materials.^35–37 In addition, the polydopamine-containing PVA/TA hydrogels exhibited excellent skin adhesive properties.³⁵ Therefore, the incorporation of various materials into PVA/TA hydrogels affects their mechanical and adhesive properties.

We hypothesized that the addition of gallic acid-conjugated chitosan (G-CS) to PVA/TA networks would improve both the mechanical and wood adhesive properties. G-CS has been previously synthesized as a tissue-adhesive material.^38,39 In this study, G-CS/polyvinyl alcohol/tannic acid (G-CPT) hydrogels were developed as plant-grafted wood adhesives. The G-CPT hydrogel adhesive exhibited improved elastic modulus (G′) values with rapid gelling and self-healing properties. In addition, the tensile strength of the G-CPT hydrogel films increased as a function of curing time. Furthermore, the G-CPT adhesives exhibited excellent wood adhesive properties. More importantly, the G-CPT hydrogels could be used as plant-grafted wood adhesives by covering the rootstock and scion of English ivy. Overall, the G-CPT hydrogel adhesives possess excellent mechanical and adhesive properties, with fixation abilities that can be used in various applications.

Chitosan (medium molecular weight, 200–800 cP), polyvinyl alcohol (PVA; M.W. 85–124 kDa), tannic acid (TA), gallic acid (GA), sodium chloride (NaCl), and N-hydroxysuccinimide were purchased from Sigma-Aldrich (Milwaukee, WI, USA). 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) was purchased from TCI-SU (Tokyo, Japan). All other chemicals were of analytical grade.

G-CS was synthesized by the formation of amide bonds between the carboxylic acid groups of gallic acid (GA) and amine groups of the chitosan (CS) backbone.^38,39 In brief, CS (1 g) was dissolved in a pH 2.0 HCl solution (100 ml), and the pH of the CS solution was adjusted to 5.0 using 5 N NaOH. GA (1.1 g) and EDC (1.2 g) were dissolved in a cosolvent of ethanol (25 ml) and DDW (distilled and deionized water) (1:1 v/v) and subsequently added to the CS solution slowly. The mixture was allowed to react for 12 h, and the pH of the reaction solution was maintained at 5.0. The product was purified by dialysis using a SpectraPor membrane with a molecular weight cutoff of 12–14 kDa against a pH 2.0 NaCl (10 mM) solution for 2 d, followed by dialysis in DDW for 4 h. The final products were lyophilized and stored in a desiccator until further use. The synthesis of G-CS was confirmed by using either ¹H NMR (Bruker Avance, 500 MHz) or UV–Vis spectroscopy (UV-1900i, Shimadzu) at the Core Facility for Supporting Analysis and Imaging of Biomedical Materials at Wonkwang University, which is supported by the National Research Facilities and Equipment Center. The degree of gallol substitution of G-CS was determined by comparing its absorbance at 265 nm caused by GA conjugation with the standard curves of GA concentrations. All measurements were performed in triplicate.

G-CS, polyvinyl alcohol (PVA), and tannic acid (TA) solutions were mixed to prepare G-CPT hydrogel adhesives. In brief, G-CS (100 mg) was dissolved in DDW (3.9 ml). Subsequently, solutions of PVA (10 wt. %) and TA (45 wt. %) were prepared in DDW. Following this, the solutions of G-CS (3.9 ml) and PVA (5 ml) were mixed and allowed to stabilize for 5 min. Subsequently, the TA solution (1 ml) was vigorously mixed with the G-CS/PVA mixture. The final G-CPT (G-CS/PVA/TA) hydrogels after centrifugation were then stored at 4 °C to promote the phase separation of the liquid and hydrogels. The final concentrations of G-CS, PVA, and TA were fixed at 1, 5, and 4.5 wt. %, respectively.

A rotational rheometer (Kinexus Lab+, NETZSCH, Germany) equipped with a 20 mm parallel plate geometry was used to monitor the viscoelastic properties of the G-CPT hydrogels. Frequency sweep measurements of the G-CPT hydrogels were performed to monitor the elastic modulus (G′) and viscous modulus (G″) as a function of frequency (0.1–10 Hz). G-CS, PVA, TA, G-CS/PVA, G-CS/TA, and PVA/TA hydrogels at the same concentrations as the G-CPT hydrogel components were used as controls. The self-healing properties of the G-CPT hydrogels were monitored using step-strain measurements. The G′ and G″ values were measured by changing the strain (0.5%, 1,000%, 0.5%, 1,000%, and 0.5%) for 120 s. All measurements were performed in triplicate.

The tensile strength was measured using a universal testing machine (UTM, TW-D1000, Taewan Tech., Seoul, Republic of Korea) equipped with a 500 N load cell. Both the PVA/TA and G-CPT hydrogels were placed in a 1 × 5 cm² rectangular mold and cured for 0, 12, or 24 h. After a predetermined time interval, both ends of the PVA/TA and G-CPT hydrogels were clamped to the UTM probes. The samples were pulled at a loading rate of 10 mm/min. The maximum force (N) at the point of the hydrogel fracture was monitored. All measurements were performed in triplicate.

A modified lap shear strength test using the UTM was performed to measure the adhesive properties of the G-CPT hydrogels on wood surfaces. In brief, wood sticks (P0000JOS, Uniart, Yangju, Republic of Korea) were cut into 1 × 5 cm² pieces and two wood sticks were overlapped by 1 × 1 cm². G-CPT hydrogels were subsequently applied between the wood sticks. The tensile strength was monitored by pulling the wood sticks at a loading rate of 1 mm/min. The detachment stresses of G-CS, PVA, TA, and PVA/TA were measured using the methods described above. In addition, a modified three-point bending test of the wood sticks attached using G-CPT hydrogels was performed. Similar to the detachment stress measurements, the wood sticks were cut into 1 × 5 cm² pieces, and the two wood sticks were overlapped by 1 × 1 cm². After applying the G-CPT hydrogels between the wood sticks, the sticks were placed onto two supporting pins. The distance between the pins was 100 mm. The upper pin imposed a controlled displacement (10 mm/min), and the corresponding forces (N) were measured. The maximum force (N) at the point of fracture was monitored. Commercial polyvinyl acetate-based craft glue (WCG70D12, AMOS) was used as the control. All measurements were performed in triplicate.

The grafting procedure was conducted using English ivy. The scion was prepared by cutting the end at an angle using a sterilized surgical blade to form a wedge shape, and the cut surface was immersed in water for 4 h. A rootstock with strong roots was prepared by removing dead leaves and small branches, followed by peeling off the outer layer of the stem to expose the cambium. The prepared scion was then carefully aligned with the cambium of the rootstock to ensure proper contact, and the graft site was secured using G-CPT hydrogel adhesive. Subsequently, the plants were divided into three groups based on days 7, 14, and 21, with three plants allocated to each group.

Micro-computed tomography (micro-CT) analysis of the English ivy grafting site was performed to monitor the internal structures between the rootstock and scion grafting as a function of time. Each sample was photographed from 2162 x-ray projections using a micro-computed tomography system (XT H 225, Nikon, Japan) to monitor the internal structures. The scan parameters were as follows: 140 kV of source voltage, 300 μA of source current, and 250 ms of exposure time. The data were visualized using the myVGL 3.0 program to view three-dimensional (3D) CT images of the internal structures and generate 3D videos.

To prepare G-CPT (gallic acid-conjugated chitosan/polyvinyl alcohol/tannic acid) hydrogel adhesives, gallic acid-conjugated chitosan (G-CS) was first synthesized using standard EDC chemistry [Fig. 1(a)]. The conjugation of gallol to the chitosan backbone was confirmed by ¹H NMR and UV–Vis spectroscopy. As shown in Fig. 1(b), the gallol protons of the G-CS backbone were observed at δ = 6.9 ppm (D₂O) in the ¹H NMR spectrum [Fig. 1(b)]. In addition, absorption peaks around 265 nm for G-CS were observed in the UV–Vis spectra [Fig. 1(c)]. The degree of gallol substitution was calculated to be 4.3% based on the absorption of G-CS at 265 nm relative to a standard curve derived from known concentrations of GA.

G-CPT hydrogel adhesives were fabricated by mixing G-CS, PVA, and TA solutions. As shown in Fig. 2(a), the G-CS solution (4 ml) was initially mixed with the PVA solution (5 ml). Subsequently, the TA solution (1 ml) was added to the mixture. When the TA solution was added to the G-CS/PVA mixtures, a water-immiscible liquid was observed around the TA solution. The G-CPT hydrogel adhesives were obtained by vigorous stirring and subsequent centrifugation. To confirm hydrogel formation, vial inversion tests were performed on the G-CPT hydrogels. As shown in Fig. 2(b), both PVA and TA solutions showed no gelation behavior, and the mixed solutions of G-CS and PVA formed hydrogels without any flow. This behavior can be attributed to the formation of 3D network structures facilitated by the hydrogen bonds between the gallol groups of G-CS and the hydroxyl groups of PVA [Fig. 2(c)]. Similarly, mixing G-CS and TA probably formed hydrogels because of the abundance of hydroxyl groups in the galloyl moieties, which may have facilitated hydrogen bonding [Fig. 2(c)]. As previously reported, a mixed solution of PVA and gallol-containing molecules exhibits a gelation behavior via hydrogen bonding.⁴⁰ In addition, the G-CPT hydrogels showed no flow after the G-CS/PVA and TA solutions were mixed.

The viscoelastic behavior of G-CPT hydrogels was measured using a rotational rheometer. Frequency sweep measurements of G-CS, PVA, TA, G-CS/PVA, G-CS/TA, PVA/TA, and G-CS/PVA/TA hydrogels were performed to monitor the elastic (G′) and viscous (G″) values. Along with the vial inverting test, G-CS [Fig. 3(a)], PVA [Fig. 3(b)], and TA [Fig. 3(c)] could not form the hydrogels in rheological analysis. When the G-CS was mixed with PVA, the mixture solidified to form hydrogels [Fig. 3(d)]. As mentioned previously, the addition of polyphenolic compounds to PVA enhances the mechanical properties of PVA-based hydrogels by forming hydrogen bonds among hydroxyl groups.^40,41 Similarly, the mixture of G-CS and TA formed a hydrogel [Fig. 3(e)]. In addition, the mixture solutions of PVA and TA showed an increase in the elastic modulus (G′) value in all frequency ranges, as shown in Fig. 3(f). As previously reported, mixed solutions of PVA and TA exhibit immediate gelation with self-healing properties via hydrogen bonding between the hydroxyl groups of PVA and TA.^20,42 G-CPT hydrogels showed further enhancement in the G′ values [Fig. 3(g)]. Figure 3(h) shows the average G′ values for G-CS, PVA, TA, G-CS/PVA, G-CS/TA, PVA/TA, and G-CS/PVA/TA hydrogels. Therefore, the incorporation of G-CS into PVA/TA enhances the mechanical properties of the hydrogels.

Self-healing properties of hydrogels are required for their application as wood adhesives. Figure 4(a) shows a brownish G-CPT hydrogel. After cutting the G-CPT hydrogel [Fig. 4(b)], both parts of the hydrogel were attached to each other [Fig. 4(c)]. Immediately after attachment, the hydrogel was pulled using forceps. As shown in Fig. 4(d), new networks were formed between the cutting edges of the hydrogel. Step-strain measurements were performed to confirm the self-healing behavior of the hydrogels [Fig. 4(e)]. When 0.5% strain was applied to the hydrogel, the G′ values of the hydrogels were 92.3 ± 2.6 kPa. Upon applying 1,000% strain, the G′ values decreased to 0.1 ± 0.09 kPa. Subsequently, when the strain was reduced to 0.5%, the G′ values fully recovered (∼100%). In addition, we monitored the elongation of the G-CPT hydrogels. As shown in Fig. 4(f), the G-CPT hydrogels were extensible when pulled using forceps. In addition, the elongation of the G-CPT hydrogels was marginally higher than that of the PVA/TA hydrogels, which may be due to the physical entanglement of the hydrogels from G-CS [Fig. 4(g)].

The tensile strength of the G-CPT hydrogels prepared was monitored using a UTM as a function of time. After preparing the G-CPT hydrogels with dimensions of 1 × 5 cm², the tensile strength was measured by pulling the UTM probe [Fig. 5(a)]. The tensile strength of the G-CPT hydrogels was ∼0.2 N immediately after preparing the adhesives. However, the tensile strengths of the G-CPT hydrogels were increased to ∼0.6 and ∼8.1 N following 12 and 24 h curing, respectively [Fig. 5(b)]. The PVA/TA hydrogels without G-CS exhibited remarkable tensile strengths (∼0.6, ∼0.9, and ∼8.1 N after 0, 12, and 24 h of curing, respectively). Before 12 h, the tensile strengths of the PVA/TA hydrogels were marginally higher than those of the G-CPT hydrogels [Fig. 5(b) and the inset of Fig. 5(b)]. However, the tensile strengths of the PVA/TA and G-CPT hydrogels after 24 h were found to be similar.

The G-CPT hydrogel exhibited remarkable adhesive properties for wood. To evaluate the wood adhesive forces of the G-CPT hydrogel adhesives, a modified lap shear test was performed. As shown in Fig. 6(a), the wood sticks were cut into 1 × 5 cm² pieces, and the two wood sticks were overlapped by 1 × 1 cm². The G-CPT hydrogel adhesives were subsequently added to the wood sticks. The measured detachment stress of the G-CPT hydrogel adhesives (142.2 ± 7.2 MPa) was substantially higher than that of G-CS (5.3 ± 1.4 MPa), PVA (2.2 ± 0.2 kPa), TA (0.4 ± 0.01 kPa), and PVA/TA hydrogels (106.5 ± 2.5 MPa) [Fig. 6(b) and the inset of Fig. 6(b)]. In addition, a modified three-point bending test of the wood sticks attached to the G-CPT hydrogels was performed to confirm their adhesive properties [Fig. 6(c)]. As shown in Fig. 6(d), the compressive strength of the G-CPT hydrogel groups (13.8 ± 1.0 N) was similar to that of commercial wood glues (15.2 ± 1.6 N). In addition, a miniature wooden house was prepared using G-CPT hydrogel adhesive [Fig. 6(e)]. Without any adhesive, the wooden structure could not maintain its form under a 3.48 kg burette support stand [Fig. 6(f)]. However, after applying the G-CPT hydrogel adhesive, the miniature house could withstand at least three burette support stands, allowing it to be set for 24 h [Fig. 6(g)]. Thus, G-CPT hydrogels are expected to have different applications as wood adhesives.

G-CPT hydrogel adhesives can be used as wood adhesives for plant grafting applications. As shown in Fig. 7(a), the English ivy rootstock was prepared using a scion prepared from another English ivy. Scion grafting onto the rootstock of the English ivy was performed. G-CPT hydrogel adhesives, composed of G-CS, PVA, and TA, were covered with graft unions without wraps or waxes. Figure 7(b) shows the photographic images of the plant grafting procedures using G-CPT hydrogel adhesives. The color of the G-CPT hydrogel adhesives changed slightly from light brown to dark brown, which may be due to the oxidization of the gallol moieties in the networks [Fig. 7(c)]. As previously reported, the color changes of gallol-containing molecules as a function of time indicate oxidative reactions of the gallol moieties.⁴³ Plant grafting with the G-CPT hydrogel adhesives was confirmed using micro-computed tomography (micro-CT). The xylem and phloem inside the plants were confirmed at 7 d [Fig. 7(d)], 14 d [Fig. 7(e)], and 21 d [Figs. 7(f)–7(h)] after grafting. Figure 7(f) shows a 3D graphic representing the x- and y-axis positions acquired by micro-CT imaging 21 d after grafting. After 7 and 14 d, there was a lack of integration between the rootstock and the scion. As shown in Figs. 7(g) and 7(h), the plants were fully interconnected between the rootstock and scion, indicating that the grafted plants could successfully survive using G-CPT hydrogel adhesives. Thus, G-CPT hydrogel adhesives can be used for various eco-friendly purposes in plant and wood industries.

In summary, gallol-containing chitosan (G-CS), polyvinyl alcohol (PVA), and tannic acid (TA) composite (G-CPT) hydrogel adhesives were developed for use as wood adhesives. The G-CS, PVA, and TA solutions solidified to form hydrogel adhesives immediately after mixing. G-CPT hydrogels exhibited enhanced mechanical, self-healing, and extensible properties. In addition, the G-CPT hydrogel adhesive exhibited excellent adhesion to wood sticks (∼142 MPa). The tensile strengths of the G-CPT hydrogel films were substantially increased to 8.1 ± 1.6 N following curing for 24 h. Therefore, G-CPT can be used for plant grafting applications as a wood adhesive owing to its remarkable mechanical strength, self-healing properties, and superior adhesiveness. G-CPT could support the interconnection of the phloem and xylem between the rootstock and scion of English ivy. Thus, G-CPT is expected to have various applications as a wood adhesive and a support material.

This study was supported by a grant from the National Institute of Ecology (NIE) funded by the Ministry of Environment of the Republic of Korea (Grant No. NIE-B-2024-18). This study was also supported by a National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT; RS-2024-00438808, J.H.R.) and a Korean Fund for Regenerative Medicine grant funded by the Korean government (Ministry of Science and ICT and Ministry of Health and Welfare; 22A0103L1, J.H.R.).

The authors have no conflicts to disclose.

Haejin Bae: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Validation (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Hyun Ho Shin: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Investigation (equal); Methodology (equal); Validation (equal); Visualization (equal). Ji Hyun Ryu: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Project administration (equal); Resources (equal); Supervision (equal); Writing – original draft (equal); Writing – review & editing (equal).

The data that support the findings of this study are available from the corresponding author upon reasonable request.