All-cellulose colloidal adhesive

The study addresses the need to replace petrochemical-based adhesives—associated with environmental harm, health risks (e.g., formaldehyde, bisphenol A), and poor recyclability—with renewable, biodegradable, and recyclable alternatives. Although biomass-based adhesives exist, they often use biomass as fillers, require complex organic synthesis, incur high costs, and underperform, limiting industrial adoption. Cellulose, the most abundant natural polymer, offers renewability and functionality but its intrinsic hydrophilicity and hydrogen bonding with water have hindered development of water-resistant cellulose-based adhesives. The research question is whether cellulose alone can be transformed into a high-performance, water-resistant colloidal adhesive suitable for industrial wood bonding. The authors propose a two-step selective partial pyrolysis and acidification strategy to create an all-cellulose colloidal adhesive exhibiting strong bonding, excellent water resistance, and simple, low-cost processing.

The paper situates the work within efforts to develop sustainable, high-performance adhesives from biomass. Prior biomass-based adhesives often use lignin or cellulose as fillers or require complex syntheses, leading to inadequate performance and limited industrial uptake. Cellulose’s abundant hydroxyl groups and hydrogen-bonding yield hydrophilicity and poor water resistance in conventional cellulose-based adhesives. Recent research has explored nanocellulose colloids, rheology of cellulose-based hydrogels, and lignin-based adhesives, but an all-cellulose, high-strength, boiling-water-resistant adhesive with simple preparation remains unmet. The authors reference works on biomass pyrolysis, nanocellulose, wood adhesives (urea/formaldehyde, phenolic, melamine-formaldehyde, lignin), and computational tools (Cellulose-Builder, GROMACS, AMBER force fields) to contextualize their approach.

Materials: Microcrystalline cellulose (MCC), sulfuric acid (H2SO4), distilled water, beech and poplar veneers. Two-step preparation of all-cellulose colloidal adhesive:

• Step 1: Selective partial pyrolysis/depolymerization. Place 8 g MCC in a ceramic crucible; heat from 25 °C to 200 °C in a blast oven and hold for 3 h to generate intermediate active cellulose (MCC-SPP). Rapidly quench in ice water.
• Evidence of depolymerization and hydrophilicity reduction: FT-IR (β-(1→4) glycosidic bond peaks at 896 and 1165 cm⁻¹ decrease; O–H at 3420 cm⁻¹ decreases), UV (changes at 210 and 289 nm), ¹H NMR (primary OH at ~3.3 ppm and anomeric/glycosidic at ~4.5 ppm decrease), contact angle increases from 33.69° to 50.15°, GPC Mw decreases from 89,400 to 74,244, XRD crystallinity slightly decreases (MCC-SPP 42.28% vs 44.00% raw MCC), EDS shows reduced O content (520 eV), DSC indicates similar glass transition behavior, confirming preserved base structure with fewer hydrophilic groups.
• Step 2: Acidification and colloid formation. Disperse intermediate cellulose into 100 mL of 1% H2SO4; heat at 180 °C with stirring at 330 rpm for 2 h. Cool to room temperature; dry at 63 °C for 5 h to yield a black oily viscous intermediate. Add directly to 250 mL distilled water, disperse at 40 °C under stirring for 4 h to obtain the all-cellulose colloidal adhesive. Acid hydrolysis in dilute sulfuric acid yields broom-like filamentous cellulose colloidal particles with high aspect ratio and self-assembly propensity. Sulfate groups can esterify/replace OH (C6, C2, C3) generating reactive sites (-SO4/HSO4 protection) for curing. Characterization:
• FT-IR: monitor O–H/C–H (2915 cm⁻¹), sulfate-related bands (S=O 1170 cm⁻¹, C–O–S 880 cm⁻¹), ester C=O (1740 cm⁻¹), O–H (3420 cm⁻¹).
• XRD: observe amorphous broad peak (2θ ~15.12–16.80°), crystallinity trends in adhesive and cured films.
• ¹H NMR: chemical shifts indicating substitution and, post-curing, ether formation (e.g., 6.75 ppm), and protective HSO4-related signals (~1.22 ppm).
• DSC/TGA/DTG: identify curing transitions and thermal stages. Key DSC peaks at ~131.91 °C (solvent loss/transition to active molten state) and ~163.86 °C (secondary crosslinking/condensation), validating hot-press cure near 150 °C. TGA stages: 35.24–140.15 °C (water/solvent loss), 140.15–200.23 °C (crosslinking/cure), 200.23–249.43 °C (decomposition).
• DLS: particle size distribution and zeta potential; particle size mainly 250–550 nm; positive zeta potential 15–75 mV indicating colloidal stability.
• Rheology: flow curves show pronounced shear-thinning; thixotropy demonstrated; viscosity 11–15 mPa·s (representative 12.70 mPa·s) with high solid content versus conventional adhesives.
• Microscopy (SEM/TEM/optical): visualize broom-like filamentous particles and cured 3D networks; bonding interface morphology in plywood.
• EDS: elemental analysis before/after pyrolysis and after colloid preparation.
• Molecular dynamics (MD): Cellulose-Builder to construct nanocellulose; GROMACS simulations under periodic boundary conditions; SO4²⁻ and H+ added; AMBER14SB force field. Outputs include Gibbs free energy landscape, radius of gyration evolution, and hydrogen bond counts to support curing mechanism and structural stability. Plywood preparation and mechanical testing:
• Veneer cutting (10×10 cm), three-layer cross-lamination, single-sided gluing. Hot pressing at 150 °C, 1.3 MPa for 5 min.
• Adhesive variants: PCA (pyrolyzed then sulfated, i.e., the all-cellulose colloidal adhesive), UPCA (unpyrolyzed, directly sulfated), HPCA (pyrolyzed then hydrochloric acid), PCA stored 15 days.
• Tests: Dry shear strength; wet shear (63 °C water for 3 h); boiling–drying–boiling cycle: 100 °C boil 4 h → cold water cool → dry at 63 °C for 20 h → 100 °C boil 4 h. Use universal testing machine (1 mm/min), three specimens per group; report averages. Wood species: primarily beech (higher strength) for stringent boiling tests; poplar also tested (Supplementary).

• Performance: The all-cellulose colloidal adhesive (PCA) achieved dry shear strength of 1.97 MPa; wet strength after 63 °C water for 3 h of 1.96 MPa (no significant loss vs dry); boiling–drying–boiling (100 °C/4 h, dry 63 °C/20 h, 100 °C/4 h) strength of up to 0.81 MPa on beech; approximately 0.49 MPa on poplar due to substrate weakness.
• Failure mode: 100% wood failure (substrate failure) for dry and 63 °C wet tests; ~95% wood failure after boiling cycle, indicating strong interface bonding.
• Comparison: Boiling-water resistance superior to urea-formaldehyde and lignin adhesives, slightly lower than melamine-formaldehyde; dry and 63 °C wet strengths comparable to mainstream wood adhesives; outperforms UPCA (dry 1.28 MPa; wet 0.63 MPa) and HPCA (dry 1.09 MPa; wet 0.31 MPa). PCA retained good performance after 15 days storage (dry 1.66 MPa; wet 1.30 MPa).
• Colloidal properties: Particle size mostly 250–550 nm; positive zeta potential 15–75 mV, contributing to colloidal stability and favorable electrostatic interaction with negatively charged wood surfaces.
• Rheology: Ultra-low viscosity 11–15 mPa·s (e.g., 12.70 mPa·s) with high solid content; strong shear-thinning and thixotropy enable efficient spraying/roller application and deep penetration.
• Structural/chemical changes enabling water resistance: Selective partial pyrolysis reduces hydrophilic –OH (contact angle increase 33.69° → 50.15°); Mw reduction (89,400 → 74,244) increases reactivity; FT-IR shows decreased O–H, emergence of sulfate/ester bands (1170, 880, 1740 cm⁻¹); ¹H NMR shows substitution and, post-cure, ether formation (~6.75 ppm). XRD indicates amorphization upon curing (crystallinity drops from ~52.89% before curing to ~3.50% after curing). Pre-cure adhesive shows broad peak at 2θ ~15.12–16.80° and lower crystallinity (e.g., 58.22% → 52.89%).
• Thermal behavior: DSC peaks at 131.91 °C (solvent removal/activation) and 163.86 °C (secondary crosslinking/condensation) confirm optimal hot-press curing around 150 °C; TGA shows curing-related mass changes (35.24–140.15 °C and 140.15–200.23 °C) and decomposition onset 200.23–249.43 °C.
• Mechanism: Adhesion arises from mechanical interlocking (deep penetration), electrostatic interactions, hydrogen bonding, and covalent bonding (etherification) forming a dense 3D crosslinked network. MD simulations support increased hydrogen bonding and compact, stable cured structures (Gibbs energy landscape and radius of gyration analyses).

The work demonstrates that cellulose alone, after selective partial pyrolysis and dilute sulfuric-acid treatment, can form a stable colloidal adhesive that cures into a dense three-dimensional network, overcoming the traditional water-resistance limitations of cellulose-based systems. Reducing the density of hydrophilic hydroxyl groups by partial pyrolysis enhances hydrophobicity and water resistance while preserving reactive sites. Sulfation/protection of hydroxyls creates active groups that, under mild hot-pressing (150 °C, 1.3 MPa, 5 min), participate in condensation and etherification reactions, corroborated by FT-IR, ¹H NMR, DSC/TGA, and XRD. The colloid’s nano–micro particle sizes and high positive zeta potential facilitate deep penetration, mechanical interlocking, and strong electrostatic interactions with wood, leading to high dry and wet strengths and retention after severe boiling cycles with predominantly wood failure. Comparative results place the adhesive’s performance at or above several commercial and biomass adhesives, with process advantages of ultra-low viscosity, high solids, and simple, low-cost preparation. MD simulations provide molecular-level evidence for network formation and stability through increased hydrogen bonding and compact conformations, supporting the proposed curing and adhesion mechanisms.

A simple two-step process—selective partial pyrolysis of microcrystalline cellulose followed by dilute sulfuric-acid treatment—yields an all-cellulose colloidal adhesive with excellent adhesion and boiling-water resistance. The adhesive achieves 1.97 MPa dry shear strength, 1.96 MPa after 63 °C water exposure, and 0.81 MPa after boiling–drying–boiling cycles, with failure predominantly in the wood substrate. Colloidal stability (positive zeta potential), nano–micro particle sizes, ultra-low viscosity (~12.70 mPa·s), and mild curing (150 °C, 1.3 MPa, 5 min) enable effective penetration, strong interfacial bonding, and industrially favorable application (e.g., spraying). Structural analyses confirm reduced hydrophilicity, activation of reactive groups, and curing into a stable, largely amorphous crosslinked network. The approach leverages renewable, biodegradable cellulose as the sole raw material, offering a low-cost, potentially scalable alternative to petrochemical adhesives. Future work could explore performance across more substrates, long-term durability and aging, optimization of sulfate content and curing protocols, and fully green processing/neutralization strategies.

• Substrate scope: The study focuses on wood (primarily beech; poplar also tested). Generalizability to non-wood substrates or wood composites beyond plywood remains untested.
• Comparative performance: While boiling-water resistance surpasses urea-formaldehyde and lignin adhesives, it is slightly lower than melamine-formaldehyde adhesives.
• Aging/storage: Adhesive strength decreases after extended storage in acidic conditions (e.g., dry strength from 1.97 MPa to 1.66 MPa after 15 days), indicating potential sensitivity to storage environment.
• Thermal limits: TGA indicates onset of significant decomposition above ~200–250 °C, limiting high-temperature applications.
• Process chemistry: The system relies on sulfuric acid; potential issues related to corrosion, acid handling, and post-processing neutralization are not detailed.