Polyvinyl alcohol solvent-free adhesives for biomass bonding via rapid water activation and heat treatment

Abstract

Bio-based adhesives have attracted significant attention from both academia and industry owing to their environmental friendliness and sustainability. However, conventional bio-based adhesives are predominantly solvent based, leading to major challenges, such as high viscosity and low solid content. In this context, this study presents the preparation of high-performance polyvinyl alcohol (PVA) solvent-free adhesives (PSFAs). These adhesives are prepared through a straightforward solution-casting method and exhibit substantial scalability, demonstrated herein by the production of a 1.1 m long sample. Following a rapid water activation (∼3 s), PSFAs show excellent adhesion capability, exhibiting excellent adhesive bonding performance with wood. Heat treatment initiates the formation of dual cross-linking networks between PVA chains, encompassing both chemical covalent bonds and physical hydrogen bonds, which leads to strong cohesion with the material. The wet strength of PSFAs reaches 2.25 MPa, exceeding that of the GB/T 9846-2015 standard (0.70 MPa) by more than two-fold. Moreover, PSFAs demonstrate exceptional durability and universality, enabling the preparation of diverse high-performance biomass composite materials. Compared to commercial adhesives, PSFAs are shown to be competitive across various dimensions, including adhesive preparation, wet bonding performance, cost-effectiveness, transportation, and storage time. In addition, the entire process of PSFAs only employs materials and chemicals recognized as safe in food by the US Food and Drug Administration (FDA). This study introduces a novel and promising approach for the development and application of environmentally friendly bio-based adhesives.

---

1. Introduction

Wood is a distinctive sustainable biomass material with many environmental benefits and notable attributes, such as renewability, carbon sequestration, and carbon storage; thus it is gathering widespread attention among people.^1–5 Wood-based panels, such as plywood, particleboard, and fiberboard, have been extensively used in cabinets, wardrobes, wood flooring, etc.⁶ According to the data sourced from the Food and Agriculture Organization of the United Nations, as of 2021, the global production of wood-based panels has reached 396.3 million m³, with a cumulative import and export value nearing 87.2 billion dollars.⁷ The processing of wood products requires the extensive use of adhesives and the performance of wood-based panels is determined by wood adhesives. However, existing adhesives predominantly comprise formaldehyde-based compounds, such as urea–formaldehyde (UF), phenol–formaldehyde, and melamine–formaldehyde, which are primarily sourced from non-renewable and unsustainable petrochemical resources. Furthermore, formaldehyde-based adhesives are known to release formaldehyde and volatile organic compounds, including benzene;^8–11 benzene and formaldehyde were classified by the International Agency for Research on Cancer as Group 1 carcinogens (carcinogenic to humans) in 1976 and 2004, respectively.^12–14 This poses substantial challenges to the sustainable development of the global wood product industry.

To mitigate the emission of formaldehyde and volatile organic compounds from wood products, researchers have developed several sustainable, formaldehyde-free alternatives for adhesives, including soy oil adhesives,¹⁵ lignin adhesives,¹⁶ soy protein adhesives,^17–20 starch-based adhesives,^21–23 and polyvinyl alcohol (PVA) adhesives.^24–26 PVA is among the most widely produced water-soluble polymers across the globe, and it can be manufactured on a large scale using non-petroleum raw materials such as bioethanol, and thus it is recognized for its general safety.²⁷ Nevertheless, macromolecules such as soybean protein, starch, and PVA form solutions with elevated viscosity, which reduces their processing adaptability.²⁸ Specifically, at a solid content of ≤12%, the viscosity of PVA solution is ≤3010 ± 20 mPa s at 25 °C (ESI Fig. 1†), sharply increasing to 13 900 ± 100 mPa s at 25 °C at a solid content of 16% (ESI Fig. 1†). Comparatively, the viscosity of UF resin at a solid content of 55% is only 45.70 ± 0.50 mPa s at 25 °C (ESI Fig. 2†). Consequently, the current sustainable formaldehyde-free adhesives employed for biomass bonding are not yet capable of substituting formaldehyde adhesives on a wide scale. To address the challenge of high viscosity, one approach involves the degradation of macromolecules via the increase in temperature or incorporation of fillers; however, this method introduces complexities and increases preparation costs.^29,30 Alternatively, addition of a diluent such as water can reduce viscosity; nonetheless, excessive water can inhibit effective bonding between adhesives and substrates. Moreover, solvent-based adhesives with a low solid content bring forward additional challenges, including increased transportation costs and a reduced product storage lifespan.

In response to these challenges, the present study proposed a novel strategy involving the preparation of solvent-free adhesives (SFAs) with a solid content of 100%, using the initially high-viscosity, low-solid-content water-based adhesives (such as PVA water-based adhesives). Notably, SFAs offer the advantage of significant reduction in transportation costs. For example, by converting PVA solutions with a solid content of 12% into SFAs, the transportation costs of the adhesive can be reduced by up to 88%. Furthermore, many wood adhesives typically require application in the solution state, and SFAs, without activation, remain in the dormant state. This characteristic prevents some common issues such as pre-curing and deterioration, associated with solvent-based adhesives, thereby maximizing the storage time. To qualify as ideal SFAs, materials must fulfill the following three conditions: (1) SFAs must be inherently safe and stable, devoid of any hazardous properties; (2) SFAs must exhibit controllable adhesion, allowing them to be activated through straightforward methods (such as water activation (WA)) to establish adhesion with substrates; (3) SFAs must produce strong cohesion after curing, as high cohesion is a key factor in preventing adhesive failure.³¹ Common examples of existing SFAs for biomass include isocyanate and thermoplastic polymers; however, these adhesives fall short of meeting the aforementioned three conditions. Although isocyanate adhesives boast exceptional performance, their high reactivity makes them toxic and hazardous.^32–34 On the other hand, thermoplastic polymers, such as low-density polyethylene, high-density polyethylene, and polyvinyl chloride (PVC), are safe and stable, yet they lack inherent activity, which mandates the need for intricate physical or chemical treatments, such as plasma treatment and addition of coupling agents, for activation and pretreatment.^35–38 Furthermore, the durability of thermoplastics as SFAs requires meticulous evaluation.

In this study, a simple solution-casting method was introduced for fabricating single-component polyvinyl alcohol (PVA) solvent-free adhesives (PSFAs), which was coupled with rapid WA (∼3 s) and hot-pressing to attain high-performance bonding (Fig. 1a). PVA particles were dissolved to create a homogeneous solution with a solid content of 12%, and subsequently a simple casting method was employed to fabricate PSFAs through hydrogen bonding self-assembly. PSFAs exhibit high hydrophilicity, which facilitates rapid WA and forms hydrogen bond cross-linking networks. This process exposes active hydroxyl groups, enabling the establishment of adhesion with the substrate. Following WA, PSFAs were assembled with wood veneers and hot pressed. The elevated temperature during hot pressing induces the cross-linking of PSFA, transitioning from single hydrogen bond cross-linking networks to dual cross-linking networks comprising both chemical covalent bonds and physical hydrogen bonds. This transformation contributes to the enhanced water resistance and heightened cohesion of PSFAs. PSFAs are environmentally friendly, easy to prepare, durable, and scalable materials with high performance. Furthermore, large-area PSFA boards with a length of 1.1 m and a width of 0.3 m (Fig. 1b) were easily prepared in this study, making PSFAs highly conducive to large-scale production. This research introduces innovative ideas and approaches for the development of sustainable, high-performance adhesives tailored for biomass materials such as wood and bamboo, significantly enhancing the industrial application prospects of green adhesives.

Fig. 1 Preparation and mechanism of polyvinyl alcohol (PVA) solvent-free adhesives (PSFAs) with rapid water activation (∼3 s) and hot pressing for high-performance bonding. (a) Schematic illustration of the preparation process, mechanism, and application of high-performance PSFAs. (b) Photograph of a large-area PSFA, measuring 1.1 m × 0.3 m; scale bar, 0.1 m.

2. Experimental

Materials

Poplar veneer was sourced from Shandong Xingang Group (Linyi, Shandong Province, China), while beech, birch, pine, and bamboo veneers were obtained from Natural Wood Modeling Material Store (Wuhan, Hubei Province, China). The thickness of wood veneers was 2 mm and that of bamboo veneers was 5 mm. PVA 1799 particles with a hydrolysis degree of 98–99% were procured from Macklin Chemical Co., Ltd (Shanghai, China). HCl and NaOH were purchased from Kunshan Jincheng Reagent Co., Ltd and Sigma-Aldrich, respectively. PVC was obtained from Nanya Plastics Co., Ltd (Nantong, Jiangsu Province, China) and UF resin from Qingjun Chemical Co., Ltd (Shijiazhuang, Hebei Province, China).

Preparation of PSFAs

PVA particles were dissolved in hot water at 95 °C and stirred for 6 h to prepare a homogeneous 12% PVA solution. The resulting PVA solutions were poured onto glass and cast using a doctor blade. After drying for 48 h, PSFAs with a thickness of ∼60 μm were obtained.

Preparation of heat-treated PSFAs

PSFAs were heated at 200 °C for 3 min to obtain heat-treated PSFAs.

Preparation of single-lap joints

The poplar veneers were cut into samples with dimensions of 100 mm × 100 mm. PSFAs were cut into samples with dimensions of 100 mm × 25 mm and rapidly subjected to WA for 3 s. Subsequently, they were adhered to two wood veneers lapped along the grain and then hot pressed at 1 MPa and 200 °C for 180 s to create wood single-lap joints. Next, the water-activated PSFAs were bonded to two bamboo veneers along the grain and hot-pressed (200 °C, 1 MPa, and 500 s) to produce bamboo single-lap joints. For PSFAs subjected to various treatments, PSFAs (without WA) were cold-pressed (without HT) at 30 °C and 1 MPa with two poplar veneers along the grain to create single-lap joints (control). The PSFAs activated with water were affixed to two pieces of poplar veneer along the grain and cold-pressed at 30 °C and 1 MPa to form single-lap joints (WA). PSFAs (without WA) were affixed to two pieces of poplar veneer along the grain and hot-pressed at 200 °C and 1 MPa for 180 s to produce single-lap joints (HT). The water-activated PSFAs were attached to two pieces of poplar veneer along the grain and hot-pressed at 200 °C and 1 MPa for 180 s to create single-lap joints (WA + HT). The single-lap joints prepared through various treatment methods and their corresponding names are presented in ESI Table 1.† The PVC was hot pressed at 200 °C and 1 MPa for 180 s with two pieces of poplar veneer along the grain to obtain single-lap joints. The UF was also hot-pressed (150 °C, 1 MPa, and 180 s) with two pieces of poplar veneer along the grain to obtain single-lap joints. Subsequently, all single-lap joints were cut into 25 mm width along the wood grain, with an overlapping area of 25 mm × 25 mm, for shear strength testing.

Preparation of biomass plywood

PSFAs with different molecular weights (PVA 0599 (M_w = 20 000–25 000), PVA 1099 (M_w = 40 000–50 000), PVA 1599 (M_w = 60 000–70 000), PVA 1799 (M_w = 70 000–80 000), and PVA 2099 (M_w = 85 000–90 000)) and thicknesses (40, 60, 80, and 100 μm) were cut into dimensions of 50 cm × 50 cm, swiftly activated by water, and then bonded to the wood veneer. The wood veneers were assembled and hot pressed at different temperatures (110, 140, 170, 200, and 230 °C) along the X–Y–X direction, following the wood grain, to produce a three-layer plywood. The hot-pressing time and pressure applied were 480 s and 1 MPa, respectively.

Characterization

The morphology of the samples was characterized by scanning electron microscopy (SEM, Thermo Scientific Apreo S). X-ray diffraction (XRD) patterns were scanned from 5° to 60° using an X-ray diffractometer (SHIMADZU 6100, Japan) at a scan rate of 5° min⁻¹. Fourier transform infrared (FTIR) spectroscopy was conducted using an FTIR spectrophotometer (Thermo Scientific Nicolet iS5, USA), over the range of 4000–600 cm⁻¹. Raman spectra were obtained using a LabRAM HR Evolution instrument (HORIBA, France) with a 785 nm wavelength of the excitation laser. The laser with a power of 500 mW was employed, and all samples were measured in the range of 50 to 4000 cm⁻¹ with 100 scans and at a resolution of 4 cm⁻¹. Differential scanning calorimetry (DSC) analysis was conducted using a differential scanning calorimeter (Q2000, TA, USA) under a nitrogen atmosphere at a heating rate of 10 °C min⁻¹ and a temperature range from 50 to 250 °C. Ultraviolet–visible (UV–Vis) absorption spectra were recorded using a UV spectrophotometer (Cary100, Agilent, USA) over the range of 200–800 nm. X-ray photoelectron spectroscopy (XPS) was performed using a spectrometer (Escalab 250Xi, Thermo Scientific, USA) with a 148.6 eV monochromatic Al K source. A rotational viscometer (NDJ-8S, Shanghai Lichen Technology Co., Ltd, China) was utilized to measure the viscosity of PVA solutions with different solid contents (4%, 8%, 12%, and 16%).

Insolubility rate test

Acidic (pH 1) and alkaline (pH 13) solutions were prepared using HCl and NaOH, respectively. PSFAs, both before and after HT, were cut into sizes of 50 mm × 50 mm. They were placed in an oven at 80 °C and dried to a constant weight, and the initial mass (m₀) was recorded. Subsequently, the samples were soaked in acid (HCl, pH 1), water, and alkali (NaOH, pH 13) solutions for 1 day, respectively. Next, PSFAs were removed, washed with alcohol, dried to a constant weight, and the weight (m₁) was recorded. The insolubility rate (ω) was calculated as follows:

Tensile strength test

PSFAs were cut into rectangular test pieces with dimensions of 100 mm × 20 mm. Tensile strength tests were performed using a universal mechanical testing machine (Shimadzu, Japan, AGS-X) at a tensile speed of 5 mm min⁻¹, under conditions of 20 ± 2 °C temperature and 65 ± 5% relative humidity. Wet tensile strength was evaluated after soaking the samples in water at 63 °C for 3 h. Six parallel experiments were conducted and final results were obtained by calculating the average value.

Shear strength test

The shear strength test was conducted on a universal mechanical testing machine (Shimadzu, Japan, AGS-X) at a tensile rate of 5 mm min⁻¹. The wet shear strength of the samples was determined after immersing them in hot water at 63 °C for 3 h. Six parallel experiments were conducted and final results were obtained by calculating the average value. The test was carried out at a temperature of 20 ± 2 °C and a relative humidity of 65 ± 5%.

Cyclic test

The cyclic test was performed following the procedure presented in ESI Fig. 14.† The detailed test procedure is as follows: the sample (either the tensile sample or single-lap joint) was immersed in hot water at 63 °C for 3 h, which was followed by drying in an oven at 63 °C for 16 h. This cycle was repeated twice. All samples were designated as ‘Cycle X-wet/dry’ based on the completed steps, where ‘X’ represents the number of cycles, ‘wet’ signifies wet tensile/shear strength, and ‘dry’ indicates dry tensile/shear strength. The samples before the cycle test were labelled as ‘Cycle 0-Y’, where ‘Y’ stands for PSFA or single-lap joints. Specifically, the single-lap joints made of PVC were named ‘Cycle X-wet/dry-PVC’, and the samples before the cycle test were named ‘Cycle 0-Y-PVC’. Six parallel experiments were conducted and the average value of their results was calculated and used as the final result.

Wet shear strength test of plywood

The wet shear strength of plywood was tested in accordance with the GB/T 9846-2015 standard, with testing conditions set at a temperature of 20 ± 2 °C and a relative humidity of 65 ± 5%. The stability test for PSFAs involved the immersion of the plywood samples in hydrochloric acid (pH = 1), water, ethanol, and ethylene glycol for one day, which was followed by testing their shear strength. The shear sample and its dimensions are illustrated in Fig. 24.† The final results were obtained by calculating the average value of the results obtained by conducting six parallel experiments.

3. Results and discussion

A substantial quantity (up to 375 L) of PVA solutions was prepared using readily available raw materials and adhesive reaction equipment (Fig. 2a–c). Subsequently, a straightforward casting method was employed to prepare PSFAs (Fig. 2d). Notably, the color of PSFAs changed from transparent to yellow after 3 min of heat treatment (HT) (Fig. 2e and ESI Movie 1†). A thorough exploration of the changes in properties and cohesion mechanisms of PSFAs before and after HT was conducted.

Fig. 2 The preparation process as well as structural and compositional characterization of polyvinyl alcohol (PVA) solvent-free adhesives (PSFAs). (a) Photograph of pilot-scale PVA particles. (b) Photograph of a pilot-scale 250 L reactor. (c) Photograph of buckets filled with PVA solution (15 × 25 L). (d) Photograph of PSFAs prepared by a solution-casting method. (e) Photograph of PSFAs after HT. (f) Stress–strain curves of PSFAs before and after HT. (g) SEM image of the cross-sectional morphology of PSFAs before HT. (h) SEM image of the cross-sectional morphology of PSFAs after HT. (i) Insoluble rate of PSFAs before and after HT. (j) XPS C 1s spectra of PSFAs before and after HT. (k) Photographs of PSFAs before and after WA.

Prior to HT, the tensile strength of PSFA was 36.50 ± 2.10 MPa, whereas post-HT, the tensile strength increased to 105.70 ± 12.10 MPa, highlighting a two-fold increase (Fig. 2f). Subsequently, microscopic morphological characterization of the cross-sections of PSFAs before and after HT was performed. Before HT, the fracture surface was rough, indicating a ductile fracture in the cross-section of PSFA (Fig. 2g). Following HT, the fracture surface became smoother and acquired a greater density, which is indicative of a transition to brittle fracture in the cross-section of PSFA (Fig. 2h) and contributing to elevated mechanical properties in PSFA after HT.

Furthermore, the insoluble rate of PSFA before and after HT was analyzed. Prior to HT, after a 1-day soaking treatment, PSFAs were nearly completely dissolved in water and alkaline solutions, with substantial dissolution also observed in an acidic solution (ESI Fig. 3†). Following HT, PSFA exhibited near-complete insolubility in water, acid solution, and alkaline solutions (ESI Fig. 4†). The calculated insolubility rates of PSFA, before HT, were found to be 0%, 2.80 ± 0.05%, and 44.50 ± 0.52% in water, acid, and alkali solutions, respectively (Fig. 2i), increasing to 99.40 ± 0.49%, 99.10 ± 0.72%, and 99.50 ± 0.42%, respectively, post-HT, indicating virtually no dissolution (Fig. 2i). Thus, HT effectively enhances the water resistance of PSFA.

DSC and XRD were used to analyze the molecular structure of the materials. Before HT, an endothermic peak temperature of 107 °C was observed for PSFA via DSC, attributed to the evaporation of water (ESI Fig. 5†). Post-HT, the endothermic peak at 107 °C essentially disappeared, indicating dehydration of PVA through HT (ESI Fig. 5†). XRD test results further revealed that, after HT, PSFA exhibited higher crystallinity, resulting in a more compact structure post condensation (ESI Fig. 6†).

FTIR spectroscopy, UV–vis spectroscopy, Raman spectroscopy, and XPS were conducted to explore the changes in chemical groups. Before HT, the FTIR spectrum of PSFA shows absorption peaks at 3260 and 1140 cm⁻¹, corresponding to hydroxyl and ether bonds, respectively (ESI Fig. 7†). Post-HT, the intensity of the absorption peak for the hydroxyl group decreased, whereas that of the absorption peak for the ether bond increased notably, indicating that HT induces the condensation of hydroxyl groups in PVA molecules, leading to the formation of ether bonds. Furthermore, the UV–vis spectrum exhibited a new shoulder at ∼340 nm after HT, indicating the generation of a polyconjugated unsaturated structure (ESI Fig. 8†).³⁹ In the Raman spectrum, the bands at 2910, 1446, and 1520 cm⁻¹ were assigned to carbon–hydrogen bonds (–CH), methylene bonds (–CH₂–), and carbon–carbon double bonds (C=C), respectively (ESI Fig. 9†). Post-HT, the intensity of bands associated with –CH bonds (2910 cm⁻¹) and –CH₂– bonds (1446 cm⁻¹) decreased, and new bands related to C=C (1520 cm⁻¹) appeared. High-resolution XPS analysis showed that, after HT, the peak intensity attributed to C–OH at 286.08 eV was significantly weakened, whereas the peak intensity ascribed to C–C/C=C at 284.80 eV increased (Fig. 2j). Thus, following HT, in addition to the existing ether bond, the hydroxyl groups of PVA undergo dehydration condensation to generate C=C bonds. Consequently, HT induces cross-linking in PSFAs, leading to the formation of both C=C bonds and C–O–C bonds. This extensive transformation, affecting the macrostructure, microstructure, molecular structure, and functional groups, contributes to the heightened cohesion of PSFAs post-HT. Furthermore, PSFA exhibited enhanced adhesive properties following simple wetting with water. Without WA, PSFAs remained non-sticky and did not adhere to surfaces such as fingers, wood, or bamboo (Fig. 2k), whereas the reverse was true on WA. Therefore, PSFAs show promising potential for bonding biomass materials.

The ideal SFA should possess both excellent adhesion and cohesion properties. Adhesion of PSFAs is achieved through WA, while HT enhances the cohesive strength. PSFA adhesion was achieved on poplar veneer via a rapid WA strategy (∼3 s) (Fig. 3a and ESI Movie 2†). Subsequently, another poplar veneer was glued onto the water-activated PSFA, and single-lap joints were prepared through hot pressing (Fig. 3b and c). Hot pressing not only improves the mechanical interlocking effect between PSFAs and the wood surface, but also elevates the cohesive strength of PSFAs at high temperatures. More importantly, achieving high-performance bonding with PSFAs is contingent upon the synergistic effects of both WA and HT. To better understand the synergistic effect of WA and HT, PSFA bonding was explored through various methods (ESI Table 1†). Notably, the control group, which was subjected to neither WA nor HT, failed to achieve wood bonding (ESI Movie 3†). The dry shear strength results for the control, WA, HT, and WA + HT were 0, 1.35 ± 0.25, 0.75 ± 0.12, and 2.07 ± 0.32 MPa, respectively (ESI Fig. 10a and b†). The failure modes observed in the adhesive layer for control, WA, HT, and WA + HT were failure to bond, adhesion/substrate failure, adhesion failure, and substrate failure, respectively (ESI Fig. 11†). WA and WA + HT samples exhibited superior dry shear strength, whereas HT demonstrated lower dry shear strength, attributed to the low adhesion properties. Simultaneously, wet shear strength tests were performed, revealing values of 0, 0, 0.38 ± 0.15, and 1.87 ± 0.31 MPa for control, WA, HT, and WA + HT samples, respectively (Fig. 3d and e). The failure modes observed in the adhesive layer for control, WA, HT, and WA + HT samples were failure to bond, cohesion failure, adhesion failure, and substrate failure, respectively (Fig. 3f). Moreover, the WA sample exhibited cracking after soaking in cold water only for 3 days, while the WA + HT sample showed no signs of cracking even after 77 days of immersion (ESI Fig. 12†). Despite the good adhesive properties of the WA sample, it showed insufficient cohesive strength and water resistance, making the adhesive layer prone to breaking. In contrast, the dry shear strength and wet shear strength of the WA + HT sample exceeded the substrate strength, demonstrating ideal bonding performance. This superiority can be attributed to the synergistic effect of WA and HT processes.

Fig. 3 Preparation process of single-lap joints, the adhesive performance of polyvinyl alcohol solvent-free adhesives (PSFAs) and the bonding mechanism. (a) Photograph of a PSFA adhered poplar layer after water activation. (b) Photograph of hot-pressing equipment. (c) Photograph of a single-lap sample. (d) Stress–strain curves of PSFA bonding performance under different treatment conditions. (e) Bonding performance of PSFA under different treatments. (f) Photographs of wood failure in PSFA under different treatment conditions. (g) Wet lap sample, the top is a digital photograph, and the bottom is the corresponding SEM image of the sample (before breaking). (h) Wet single-lap joints, the top is a digital photograph, displaying the lotus silk-like mechanical interlocking structure, and the bottom is the corresponding SEM image of the sample (after breaking). (i) Schematic illustration of the adhesion and cohesion mechanisms of PSFAs. HT, heat treatment.

For further systematic exploration of the bonding mechanism of PSFAs, SEM was used to characterize the microscopic morphology of the dry and wet single-lap joints. Dry single-lap joints showed a dense mechanical interlocking structure formed by the interpenetration of the wood cell wall and PSFAs (ESI Fig. 13†). Furthermore, the hydroxyl groups of PVA can establish hydrogen bonds with the hydroxyl groups on the wood surface. In the case of wet single-lap joints, the interface between PSFAs and wood retained this mechanical interlocking structure (Fig. 3g). Remarkably, even after breaking, PSFAs recombined with wood, forming distinctive lotus silk-like structures at the wood interface (Fig. 3h). This phenomenon is attributed to the strong mechanical interlocking structure between the PSFA–wood interface, allowing the lapped sample to dissipate more energy during breaking, thereby achieving high-performance adhesion. The bonding mechanism between PSFAs and wood is depicted in Fig. 3i. Following WA, PSFA forms mechanical interlocking structures with the wood interface, ensuring robust adhesion. Concurrently, the cross-linking of the PVA chain undergoes transitions from single hydrogen bonds to a synergistic combination of chemical covalent bonds (carbon–carbon double bonds and ether bonds) and physical hydrogen bonds, which endows PSFAs with robust adhesive performance and water resistance.

Further tests were conducted on the cyclic tensile strength of heat-treated PSFAs. Notably, a reversibility in the high toughness and strength of heat-treated PSFAs was demonstrated during repeated cycles of soaking in hot water and subsequent drying (ESI Fig. 14 and 15a, b†), suggesting that PSFA may adapt well to both dry and wet conditions; this could potentially mitigate cracking caused by drying stress generated in wood and enhance the overall durability.

To validate this hypothesis, ‘soaking in hot water–drying’ cycle tests were conducted to assess the bonding performance of the samples and their applicability to different biomass sources (Fig. 4a and ESI Fig. 14†). Initially, cyclic shear strength tests were conducted using beech as a representative of hardwood. The adhesive layer of beech samples remained free of cracks throughout different stages of the cycle test (Fig. 4b). The shear strengths at various stages were measured to be 5.20 ± 0.40, 2.18 ± 0.26, 5.41 ± 0.20, 2.33 ± 0.52, and 4.78 ± 0.14 MPa, respectively (Fig. 4c and ESI Fig. 16†). Even after double cycle tests, substrate failure was shown to be the predominant failure mode (Fig. 4d). Similarly, the shear strengths of poplar single-lap joints at different stages were 2.07 ± 0.32, 1.87 ± 0.31, 1.91 ± 0.12, 1.81 ± 0.32, and 1.83 ± 0.16 MPa, respectively. The observed final failure mode after two cycle tests was also substrate failure (ESI Fig. 17a–d†). Pine lap joints were used as an example of a softwood, and no cracking occurred during different stages of the cycle tests, with shear strengths at the different stages being 2.09 ± 0.33, 1.8 ± 0.22, 1.93 ± 0.22, 1.52 ± 0.15, and 1.56 ± 0.14 MPa, respectively, with the final failure mode identified as substrate failure (Fig. 4e–g and ESI Fig. 18†). To elaborate on the number of cycles of PSFAs, multiple “soaking in hot water-drying” cycling tests were conducted using beech single-lap joints. The results showed that even after up to eight cycles of the ‘soaking in hot water-drying’ test, the shear strength of PSFAs remained at 4.56 ± 0.46 MPa, further demonstrating the excellent bonding performance of PSFAs on biomass materials (ESI Fig. 19†).

Fig. 4 Cycle tests and universality of polyvinyl alcohol solvent-free adhesives (PSFAs). (a) Schematic representation of cycle tests of single-lap joints. (b) Photographs of beech single-lap joints. (c) Bonding performance of beech single-lap joints. (d) Failure modes of beech single-lap joints. (e) Photographs of pine single-lap joints. (f) Bonding performance of pine single-lap joints. (g) Failure modes of pine single-lap joints. (h) Photographs of bamboo single-lap joints. (i) Bonding performance of bamboo single-lap joints. (j) Failure modes of pine single-lap joints.

In addition to wood, other biomasses were also assessed, including bamboo. The bamboo lap sample exhibited no cracking during the cycle tests, with shear strengths at the different stages measured to be 5.95 ± 0.33, 5.89 ± 0.90, 5.92 ± 0.74, 5.25 ± 1.04, and 5.72 ± 0.52 MPa, respectively. Notably, the final failure mode was still substrate failure (Fig. 4h–j and ESI Fig. 20†). For comparative analysis, PVC was selected as a representative thermoplastic polymer and subjected to the same cycle tests. Although the PVC adhesive layer remained intact during the initial stages, some cracking occurred in the final stage (ESI Fig. 21a†). The shear strengths at different stages were 1.37 ± 0.15, 0.42 ± 0.06, 0.70 ± 0.02, 0.46 ± 0.23, and 0.41 ± 0.03 MPa, indicating a noticeable downward trend (ESI Fig. 21b and c†). Furthermore, the failure mode shifted from substrate failure to adhesive failure (ESI Fig. 21d†). Given that PVC cannot be activated by water, it lacks good adhesion properties (ESI Fig. 22†). Although PVC can form bonds with wood through van der Waals forces and mechanical interlocking in the dry state (ESI Fig. 23a†), its deficiency in adhesive properties makes it prone to cracking in the wet state (ESI Fig. 23b†). In contrast, PSFA demonstrates exceptional durability and universality when applied to wood and biomass materials.

To assess the potential application of PSFAs in biomass composites, herein, poplar plywood (0.5 m × 0.5 m) was prepared through a process involving WA, assembly, and HT (Fig. 5a and ESI Fig. 24†).

Fig. 5 Performance comparison between polyvinyl alcohol solvent-free adhesives (PSFAs) and other biomass adhesives. (a) Illustration of the preparation process for plywood (0.5 m × 0.5 m). (b) Comparison of wet strength of PSFAs with those of various current green adhesives for biomass. (c) Comprehensive evaluation compared with existing commercial biomass adhesives. UF, urea–formaldehyde; MDI, methylene diphenyl diisocyanate; SB, soybean adhesive.

In this study, PVA samples with various molecular weights were used to prepare PSFAs, these samples included PVA 0599 (M_w = 20 000–25 000), PVA 1099 (M_w = 40 000–50 000), PVA 1599 (M_w = 60 000–70 000), PVA 1799 (M_w = 70 000–80 000), and PVA 2099 (M_w = 85 000–95 000) and their wet strengths were tested (ESI Fig. 25 and 26†). The wet strengths of PVA 0599, PVA 1099, PVA 1599, PVA 1799, and PVA 2099 were found to be 0.56 ± 0.21, 1.49 ± 0.16, 1.58 ± 0.17, 1.63 ± 0.21, and 1.26 ± 0.06 MPa, respectively. The relatively low wet shear strength of PVA 0599 may be attributed to its low molecular weight, resulting in insufficient cohesive strength. Moreover, the wet strength of PVA 2099 decreases, which could be attributed to the excessively large molecular weight, resulting in reduced penetration (adhesion) on the wood surface.

Furthermore, the wet strength of PSFAs was also investigated at different hot-pressing temperatures and thicknesses (ESI Fig. 27 and 28†). At temperatures below 200 °C, the PSFA does not fully crosslink due to the lower curing temperature, resulting in poor adhesive performance. However, at higher temperatures (230 °C), the wet adhesive performance decreases, which can be attributed to the elevated temperature causing partial degradation of the cell wall and reducing the mechanical strength of the substrate.⁴⁰ Moreover, when the thickness is 60 μm, the wet strength of PSFAs is 1.63 ± 0.21 MPa. However, further increasing the thickness of PSFA does not significantly improve the bonding performance and would raise the cost. This result may be attributed to the fact that when the thickness is too small (<60 μm), it leads to insufficient adhesion at the wood bonding interface; in contrast, when the thickness is too large (>80 μm), it generates stress in the adhesive at the interface. Therefore, from a cost-effectiveness perspective, the 60 μm PSFA is a more ideal choice.

In this study, plywood products were also prepared using birch and beech, respectively. Furthermore, shear samples were prepared according to the GB/T 9846-2015 standard, and their wet shear strength was tested. The wet shear strengths for poplar plywood, birch plywood, and beech plywood were 1.63 ± 0.21, 1.62 ± 0.04, and 2.25 ± 0.12 MPa, respectively, which exceeded the class II plywood standards (0.7 MPa) by 132.86%, 131.43%, and 221.43%, respectively (ESI Fig. 29†). The failure mode observed was either substrate failure or cohesive/adhesive failure (ESI Fig. 30†), indicating that the bonding strength of PSFAs exceeds or is comparable to the strength of the substrates, signifying exceptional adhesive performance. In this study, the samples prepared with PSFA were also soaked in different solvents such as acid, water, ethanol, and ethylene glycol for one day (ESI Fig. 31†). Their bonding strengths were 1.13 ± 0.10, 1.79 ± 0.03, 1.49 ± 0.20, and 1.55 ± 0.04 MPa, respectively, demonstrating the excellent solvent stability of PSFAs. Compared to other reported green adhesives, such as soybean (SB) adhesives, lignin adhesives, starch adhesives, and other SFAs, PSFAs exhibit excellent wet shear performance (Fig. 5b and ESI Table 2†).

Compared with other multi-component biomass adhesives, PSFAs are single-component adhesives primarily composed of polyvinyl alcohol. No additional cross-linking agents are required during preparation or use, thereby maximizing atom economy. To better illustrate the green advancements of PSFAs in wood adhesives, we also compared PSFAs with urea-formaldehyde (UF) resin and methylene diphenyl diisocyanate (MDI) adhesives (ESI Table 3†). The starting material for polyvinyl alcohol can be bioethanol, obtained through the fermentation of biomass resources such as sugarcane and tubers, thereby eliminating reliance on fossil fuels.⁴¹ However, the primary starting materials for synthesizing UF resin are urea and formaldehyde. Additionally, the main starting materials for synthesizing MDI are aniline, phosgene, and formaldehyde. In terms of toxicity, the acute oral LD₅₀ of PVA is greater than 21.5 g kg⁻¹, indicating that it is generally nontoxic.⁴² The LD₅₀ (oral, mouse) and LC₅₀ (inhalation, mouse) of formaldehyde are 42 mg kg⁻¹ and 0.414 mg L⁻¹/4 h, respectively, indicating that excessive release of free formaldehyde from UF can cause significant harm to both human health and the environment.⁴³ In addition, the LC₅₀ (inhalation, rat) of MDI is 0.369 mg L⁻¹/4 h, which is also toxic.⁴⁴ It is worth noting that phosgene (LC₅₀ (inhalation, cat) = 0.190 mg L⁻¹/15 min) used in MDI synthesis is also a highly toxic substance.⁴⁵ PVA is generally recognized as safe (GRAS) by the US Food and Drug Administration (FDA) and is known for its good biodegradability and biocompatibility. However, UF and MDI are both non-biodegradable and non-biocompatible. Therefore, compared to existing wood adhesives, PSFAs show superior green advancements in terms of resources, starting materials, toxicity, biodegradability, and biocompatibility.

A comprehensive comparison was also conducted between PSFAs and existing commercial adhesives for biomass, including UF, MDI, and SB adhesives, across five dimensions: the manufacturing process of adhesives, wet shear strength, adhesive cost, transportation, and adhesive storage time (Fig. 5c). PSFAs can be obtained solely through the solution-casting method, requiring no additional processing, and boasting a straightforward manufacturing process. Moreover, PSFAs exhibit superior wet shear strength, while UF adhesive demonstrates lower wet shear strength (ESI Fig. 32 and 33†). The cost of PSFAs is only 0.09 USD m⁻² slightly higher than that of UF adhesive at 0.06 USD m⁻² (ESI Table 4†). In contrast, MDI and SB incur higher costs at 0.40 USD m⁻² and 0.20 USD m⁻², respectively (ESI Table 3†). Moreover, MDI and SB adhesives often necessitate additional fillers to adjust viscosity, contributing to further increased costs. In terms of transportation, the solid contents of UF and SB adhesives are within 50–60% and 15–40%, respectively, resulting in transportation costs that are 167–200% and 250–667% those of PSFAs consisting of 100% solid content (ESI Table 5†). Notably, PSFA, which contains only PVA, is non-toxic and safe during transport (ESI Table 6†). In contrast, UF and MDI adhesives are toxic and explosive, making their transportation unsafe (ESI Table 4†). Moreover, PSFAs also exhibit an extended storage time of over 1 year, while UF, MDI, and SB adhesives may undergo self-polymerization or decomposition, with the corresponding storage periods of 10–30 days, 120–180 days, and <30 days, respectively (ESI Table 5†). Impressively, even after 1 year, the shear strength of PSFAs remains comparable to its initial strength (ESI Fig. 34 and 35†). Therefore, PSFAs demonstrate a highly competitive commercial prospect.

4. Conclusion

In this study, polyvinyl alcohol solvent-free adhesives (PSFAs) derived from PVA solutions through hydrogen bonding self-assembly exhibited exceptional bonding performance. The achievement of high-performance bonding with PSFAs involved a rapid WA step (∼3 s) followed by hot pressing. WA played an important role in opening the hydrogen bond cross-linking network of PVA chains, facilitating adhesion of PSFAs. Hot pressing further enhanced the mechanical interlocking effect between PSFAs and the substrate, including high cohesion within PSFAs through synergistic crosslinking networks involving ether bonds, carbon–carbon double bonds, and hydrogen bonds. Moreover, PSFAs demonstrated exceptional durability, maintaining a dry strength equivalent to the initial strength even after two consecutive dry/wet cycles. The universality of PSFAs was evident in their successful bonding with various biomass materials such as poplar, beech, pine, and bamboo. With PSFAs, diverse biomass composite materials such as poplar plywood, beech plywood, and birch plywood were successfully prepared. In a comprehensive comparison with other commercial adhesives, such as UF, SB, and MDI, PSFA exhibited superiority in adhesive preparation, wet bonding performance, cost-effectiveness, transportation, and storage time. Furthermore, compared with other multi-component green adhesives and commercial adhesives, PSFAs exhibit better advantages in green chemistry, including high atom economy, renewable raw materials, and recognition as safe for food use by the U.S. Food and Drug Administration (FDA). PSFAs exhibit the potential to overcome the limitations associated with traditional solvent-based adhesives characterized by high viscosity and low solid content. This study opens up new avenues for the development and application of high-performance green adhesives for biomass.

Author contributions

Liangxian Liu: Writing – original draft, writing – review & editing, data curation, investigation, methodology, validation, and formal analysis; Ming wei: investigation and resources; Haiyu Li: investigation and validation; Yutong Chen: investigation and validation; Yuyan Jiang: investigation and validation; Tian Ju: investigation; Zetan Lu: investigation; Guoqing Mu: investigation; Lijian Cai: investigation; Dexiu Min: investigation; Yanjun Xie: investigation; Jian Li: supervision and conceptualization; Shaoliang Xiao: supervision, funding acquisition, writing – review & editing, conceptualization, and project administration. All authors commented on the final manuscript.

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

S. X., L. L., J. L., and M. W. are inventors on a patent application related to this work filed by Northeast Forestry University (Application number: 2024103509093, filling date: 26 March 2024) titled “A Method for Preparing a Solvent-Free Polyvinyl Alcohol Adhesive and Its Bonding Method”. All other authors declare that there is no conflict of interests regarding the publication of this article.

Acknowledgements

We acknowledge funding support from the financial supports provided by the National Natural Science Foundation of China (32401500), the National Key Research and Development Program of China during the 14th Five-Year Plan Period (grant number: 2023YFD2201404), the Major Science and Technology Innovation Projects in Shandong Province (grant number: 2023CXGC010312), the Natural Science Foundation of Heilongjiang Province (grant number: YQ2023C024), and the Fundamental Research Funds for the Central Universities (grant number: 2572023CT07).

References

1. Y. Zhu, C. Romain and C. K. Williams, Nature, 2016, 540, 354–362.
2. A. K. Mohanty, F. Wu, R. Mincheva, M. Hakkarainen, J.-M. Raquez, D. F. Mielewski, R. Narayan, A. N. Netravali and M. Misra, Nat. Rev. Methods Primers, 2022, 2, 46.
3. Q. Xia, C. Chen, Y. Yao, S. He, X. Wang, J. Li, J. Gao, W. Gan, B. Jiang, M. Cui and L. Hu, Adv. Mater., 2021, 33, e2001588.
4. W. Gan, C. Chen, H. T. Kim, Z. Lin, J. Dai, Z. Dong, Z. Zhou, W. Ping, S. He, S. Xiao, M. Yu and L. Hu, Nat. Commun., 2019, 10, 5084.
5. C. H. Dreimol, H. Guo, M. Ritter, T. Keplinger, Y. Ding, R. Gunther, E. Poloni, I. Burgert and G. Panzarasa, Nat. Commun., 2022, 13, 3680.
6. Z. Hao, X. Xi, D. Hou, H. Lei, C. Li, G. Xu and G. Du, Int. J. Biol. Macromol., 2023, 253, 127135.
7. . Forestry Production and Trade, https://www.fao.org/faostat/en/#data/FO, (accessed January 10, 2024).
8. Y. Ma, Z. Kou, Y. Hu, J. Zhou, Y. Bei, L. Hu, Q. Huang, P. Jia and Y. Zhou, Int. J. Adhes. Adhes., 2023, 126, 103444.
9. X. Shi, S. Gao, C. Jin, D. Zhang, C. Lai, C. Wang, F. Chu, A. J. Ragauskas and M. Li, Green Chem., 2023, 25, 5907–5915.
10. A. Arias, E. Entrena-Barbero, G. Feijoo and M. T. Moreira, J. Environ. Chem. Eng., 2022, 10, 107053.
11. S. Ge, N. L. Ma, S. Jiang, Y. S. Ok, S. S. Lam, C. Li, S. Q. Shi, X. Nie, Y. Qiu, D. Li, Q. Wu, D. C. W. Tsang, W. Peng and C. Sonne, ACS Appl. Mater. Interfaces, 2020, 12, 30824–30832.
12. V. J. Cogliano, R. Baan, K. Straif, Y. Grosse, B. Lauby-Secretan, F. El Ghissassi, V. Bouvard, L. Benbrahim-Tallaa, N. Guha, C. Freeman, L. Galichet and C. P. Wild, J. Natl. Cancer Inst., 2011, 103, 1827–1839.
13. D. Loomis, K. Z. Guyton, Y. Grosse, F. El Ghissassi, V. Bouvard, L. Benbrahim-Tallaa, N. Guha, N. Vilahur, H. Mattock and K. Straif, Lancet Oncol., 2017, 18, 1574–1575.
14. C. Protano, G. Buomprisco, V. Cammalleri, R. N. Pocino, D. Marotta, S. Simonazzi, F. Cardoni, M. Petyx, S. Iavicoli and M. Vitali, Cancers, 2022, 14, 165.
15. C. R. Westerman, B. C. McGill and J. J. Wilker, Nature, 2023, 621, 306–311.
16. G. Yang, Z. Gong, X. Luo, L. Chen and L. Shuai, Nature, 2023, 621, 511–515.
17. Y. Xu, Y. Han, M. Chen, J. Luo, S. Q. Shi, J. Li and Q. Gao, Composites, Part B, 2021, 211, 108677.
18. Y. Wei, S. Jiang, C. Li, J. Li, X. Li, J. Li and Z. Fang, Ind. Crops Prod., 2022, 189, 115850.
19. S. Jin, K. Li, Q. Gao, W. Zhang, H. Chen, J. Li and S. Q. Shi, J. Cleaner Prod., 2020, 254, 120143.
20. Y. Zeng, W. Yang, P. Xu, X. Cai, W. Dong, M. Chen, M. Du, T. Liu, P. J. Lemstra and P. Ma, Chem. Eng. J., 2022, 428, 132390.
21. L. Chen, Z. Xiong, Q. Li, Z. U. Din and H. Xiong, Int. J. Biol. Macromol., 2019, 140, 1026–1036.
22. W. Shao, P. Wang, J. Liu, H. Xu, X. Cai, Q. Wu, N. Xia and F. Kong, Ind. Crops Prod., 2022, 181, 114809.
23. Y. Sun, J. Gu, H. Tan, Y. Zhang and P. Huo, Int. J. Biol. Macromol., 2018, 112, 1257–1263.
24. J. Lamaming, N. B. Heng, A. A. Owodunni, S. Z. Lamaming, N. K. A. Khadir, R. Hashim, O. Sulaiman, M. H. M. Kassim, M. H. Hussin, Y. Bustami, M. H. M. Amini and S. Hiziroglu, Composites, Part B, 2020, 183, 107731.
25. D. Lee, H. Hwang, J. S. Kim, J. Park, D. Youn, D. Kim, J. Hahn, M. Seo and H. Lee, ACS Appl. Mater. Interfaces, 2020, 12, 20933–20941.
26. J. T. Aladejana, Z. Wu, D. Li, K. Guelifack, W. Wei, X. A. Wang and Y. Xie, ACS Sustainable Chem. Eng., 2019, 7, 18524–18533.
27. M. Rostagno, S. Shen, I. Ghiviriga and S. A. Miller, Polym. Chem., 2017, 8, 5049–5059.
28. Z. u. Din, L. Chen, H. Xiong, Z. Wang, I. Ullah, W. Lei, D. Shi, M. Alam, H. Ullah and S. A. Khan, Starch/Staerke, 2020, 72, 1900276.
29. J. Luo, Y. Zhou, Q. Gao, J. Li and N. Yan, ACS Sustainable Chem. Eng., 2020, 8, 10767–10773.
30. G. Zeng, F. Zhu, J. T. Aladejana, Y. Zhou, K. Li, J. Luo, X. Li, Y. Dong, K. Wang and J. Li, J. Mater. Chem. A, 2023, 11, 11310–11325.
31. H. Fan and J. P. Gong, Adv. Mater., 2021, 33, e2102983.
32. J. Sternberg and S. Pilla, Green Chem., 2020, 22, 6922–6935.
33. E. Delebecq, J. P. Pascault, B. Boutevin and F. Ganachaud, Chem. Rev., 2013, 113, 80–118.
34. M. Yamada, H. Kato and K. Taki, J. Adhes. Soc. Jpn., 2012, 48, 193–203.
35. K. N. Pandiyaraj, V. Selvarajan, R. R. Deshmukh and C. Gao, Vacuum, 2008, 83, 332–339.
36. L. M. Matuana, J. J. Balatinecz and C. B. Park, Polym. Eng. Sci., 1998, 38, 765–773.
37. X. Zhou, Y. Cao, K. Yang, P. Yu, W. Chen, S. Wang and M. Chen, J. Cleaner Prod., 2020, 269, 122196.
38. M. Tahara, N. K. Cuong and Y. Nakashima, Surf. Coat. Technol., 2003, 174–175, 826–830.
39. N. i. V. Alekseeva, A. M. Evtushenko, I. P. Chikhacheva, V. P. Zubov and I. V. Kubrakova, Mendeleev Commun., 2005, 15, 170–172.
40. K. Srinivas and K. K. Pandey, J. Wood Chem. Technol., 2012, 32, 304–316.
41. M. Rostagno, S. Shen, I. Ghiviriga and S. A. Miller, Polym. Chem., 2017, 8, 5049–5059.
42. B. Nair, Int. J. Clin. Toxicol., 1998, 17, 67–92.
43. . National Center for Biotechnology Information. PubChem Compound Summary for CID 712, Formaldehyde, https://pubchem.ncbi.nlm.nih.gov/compound/Formaldehyde, (accessed July 1, 2024).
44. . National Center for Biotechnology Information. PubChem Compound Summary for CID 7570, 4,4′-diphenylmethane diisocyanate, https://pubchem.ncbi.nlm.nih.gov/compound/4_4_-Diphenylmethane-diisocyanate, (accessed July 1, 2024).
45. . National Center for Biotechnology Information. PubChem Compound Summary for CID 6371, Phosgene, https://pubchem.ncbi.nlm.nih.gov/compound/Phosgene, (accessed July 1, 2024).

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4gc03386h

---