Kinetics and Mechanical Performance of Bio-Based Polyurethane Wood Composites for Sustainable 3D-Printed Construction Materials

Abstract

Developing bio-based polyurethane (BPU) composites that incorporate bio-oil and wood dust as sources of hydroxyl groups (-OH) presents a compelling approach to advancing sustainable polymer systems. This study examines the impact of isocyanate-to-hydroxyl equivalent ratios and varying proportions of bio-oil and wood dust on the processability and mechanical properties of molded composite panels. Formulations were systematically optimized based on equivalent ratio calculations to enhance extrusion behavior and final structural performance. Extrusion trials demonstrated that an -NCO/-OH ratio of 1.5:1, with 50% wood dust serving as an -OH donor, resulted in the most stable material flow, characterized by minimized surface defects and an ideal viscosity for processing. Compression molding and mechanical testing revealed that a balanced formulation with 50% bio-oil and 50% wood dust, with an equivalent ratio of -OH groups, achieved the best combination of Young’s modulus, stress, and strain performance, even under wet conditions. SEM confirmed improved filler dispersion and interfacial adhesion in these optimized systems. Although full 3D-printing trials were not conducted, the observed extrusion stability and controlled curing behavior indicate strong potential for application in extrusion-based additive manufacturing. These results highlight that precise resin–filler balancing enables continuous extrusion, structural resilience, and reduced activation energy, reinforcing the viability of BPUs as scalable, sustainable materials for construction and additive manufacturing.

1. Introduction

Despite previous research investigating the synthesis or mechanical characteristics of bio-based polyurethanes, limited studies have quantitatively linked curing kinetics to mechanical properties within heterogeneous blend systems. This study combines in situ FTIR kinetics and mechanical testing to establish relationships between reaction, structure, and properties relevant to additive manufacturing and construction.

Incorporating sustainable polymers into additive manufacturing (AM), also known as 3D printing, offers a significant advance in sustainable materials engineering [1,2,3,4]. With growing environmental demands, bio-based polyurethanes (BPUs) have attracted attention for their renewable origins, including lignocellulosic biomass, vegetable oils, and bio-oils [5,6,7,8,9,10,11]. These resins provide mechanical strength and thermal stability while reducing carbon impact [12,13,14]. Techniques such as fused deposition modeling (FDM) enable the precise, low-waste fabrication of complex geometries, and their integration into additive manufacturing further allows for the creation of lightweight, structurally intricate, and application-specific components that advance circular economy strategies [5,15,16,17,18,19,20,21].

Advances in reactive extrusion and additive manufacturing now allow in situ polymerization during 3D printing, enabling high-performance thermoset structures to be fabricated directly from liquid precursors. This method, known as reactive additive manufacturing (RAM) [22,23,24], offers controlled crosslinking, adaptable curing kinetics, and real-time tuning of material properties. Such capabilities are accelerating the adoption of bio-based thermosets in extrusion-based platforms, driven by their chemical tunability and structural versatility [25,26,27,28].

The global thermosetting resin market is projected to grow substantially over the next decade, driven by increasing demand in the construction, automotive, and aerospace industries for materials with superior mechanical and thermal properties. This growth is also attributed to the increasing importance of advanced technologies utilizing bio-based resins, as well as a shift toward circular economy models. Growth is further reinforced by evolving environmental regulations and rising consumer interest in renewable alternatives. Biomass-derived feedstocks are increasingly explored as sustainable substitutes to reduce dependence on fossil carbon [29,30,31]. Reflecting this transition, the U.S. Department of Energy (DOE) and the U.S. Department of Agriculture (USDA) have prioritized the development of bioenergy and bioproducts [32].

Bio-based polyurethanes (BPUs), as thermosets, form densely crosslinked networks through irreversible curing, enabling control of stiffness, thermal resistance, and biodegradability [33,34,35,36,37]. Their reinforcement with OH-rich fillers and natural fibers enhances performance while valorizing agro-industrial residues [5,38,39]. These advances support global sustainability targets and the development of high-performance bio-composites. Market projections estimate BPU revenues of USD 63.90 million globally and USD 16.79 million in the U.S. by 2030, underscoring rising industrial adoption [40,41,42].

Extensive research has focused on renewable polyols from triglycerides, cellulose, lignin, and thermochemically converted biomass, such as pyrolysis oils, as sustainable alternatives to fossil-derived products [6,43]. These bio-based polyols reduce environmental impact while providing diverse hydroxyl groups that enhance reactivity, crosslinking, and network tunability [39,44,45,46,47,48,49,50]. Among them, pyrolysis-derived poplar-wood bio-oil is particularly promising, as its phenolic- and hydroxyl-rich composition enables efficient polyaddition with isocyanates to form robust polyurethane networks [51]. The hydroxyl distribution and density govern crosslink density, gel time, and mechanical properties, positioning bio-oils as versatile feedstocks for renewable polyurethane systems [52,53,54,55].

Polyurethane curing kinetics, driven by step-growth reactions between hydroxyl and isocyanate groups, dictate gelation, processability, and final performance [56,57,58,59]. Theoretical frameworks and predictive models [60] estimate gel points based on stoichiometry and functionality, linking isocyanate conversion with network connectivity [47]. While effective for well-defined systems, applying these models to bio-based formulations is more complex due to the heterogeneity of feedstock. Experimentally, kinetic models, such as nth-order [58,61,62], are used to extract rate constants, activation energies, and reaction orders [46,59,63,64,65]. Fourier-transform infrared (FTIR) spectroscopy enables in situ monitoring of isocyanate consumption and, when combined with multi-temperature kinetic analyses, provides detailed profiles of cure behavior. This integration of modeling and FTIR yields a comprehensive view of network development, highlighting the role of polyol chemistry and guiding formulation strategies to optimize the performance of sustainable polyurethane systems [58,61,66,67,68].

The mechanical performance of BPU composites depends on matrix microstructure and filler–matrix interactions, particularly with lignocellulosic fillers [69,70,71,72]. Wood flour enhances stiffness, strength, and stability by deflecting cracks and redistributing loads [73,74,75,76]. In extrusion-based 3D printing, optimizing filler loading and dispersion is crucial to balance printability with performance [77,78,79,80,81,82,83].

Processing lignocellulosic fillers is hindered by their hydrophilicity, heterogeneity, and agglomeration, which reduce dispersion and interfacial adhesion, potentially diminishing mechanical reliability [84,85]. Surface treatments, compatibilizers, and optimized loadings address these issues. Comprehensive evaluation requires ASTM-based mechanical testing alongside microstructural and chemical analyses to establish structure–property relationships and confirm formulation effectiveness.

While bio-based polyurethanes show promise for additive manufacturing, most studies treat curing kinetics and mechanical properties separately, often relying on simplified formulations. The chemical heterogeneity of pyrolytic bio-oils complicates predictions of gelation and network development, while the effects of fillers on mechanical reliability remain underexplored. This work addresses these gaps by integrating FTIR-based kinetic modeling with ASTM-standard mechanical testing of wood–flour composites. Linking cure kinetics with structural performance provides new insights into structure–property relationships and supports the development of sustainable polyurethane systems for extrusion-based construction.

2. Experimental Section

2.1. Materials

Polymeric methylene diphenyl diisocyanate (p-MDI) was purchased from Huntsman Corporation (The Woodlands, TX, USA). Poly(ethylene glycol), average M.W. 400 (PEG-400), was purchased from Thermo Scientific (Waltham, MA, USA).

The fast pyrolysis bio-oil (BO) used in this study was derived from poplar wood biomass through thermochemical decomposition at temperatures ranging from 400 to 600 °C in an oxygen-limited environment. The pyrolysis process was conducted in the Biosystems Engineering Laboratory at Auburn University, utilizing an electrostatic precipitator to enhance vapor condensation and liquid yield, in accordance with the methodology of Sakhakarmy et al. [51]. The resulting bio-oil, corresponding to the organic phase, was collected as a complex mixture of oxygenated hydrocarbons, including phenolics, carboxylic acids, aldehydes, and lignin-derived oligomers. To characterize the bio-oil’s functionality, Bruker 500 MHZ Nuclear Magnetic Resonance Spectroscopy (Billerica, MA, USA) (³¹P NMR) was used to quantify the concentration and distribution of hydroxyl (-OH) groups. Integration of the NMR spectra revealed distinct contributions from aliphatic OHs (3.49 mmol/g), phenolic OHs (4.27 mmol/g, including guaiacyl, catechol, and p-Hydroxyphenyl groups), and acidic OHs (0.79 mmol/g).

Wood dust (300 µm) was obtained from the bio-oil biorefinery process as a byproduct. Its accessible hydroxyl (-OH) content was measured using Surface Measurement Systems Ldt (Allentown, PA, USA) Dynamic Vapor Sorption (DVS) with deuterium exchange in collaboration with Surface Measurement Systems. The method involved drying the sample, exposing it to D₂O vapor at 95% relative humidity, and then re-drying to measure the mass change. The accessible hydroxyl content was calculated using the equation:

$$A={\displaystyle \frac{m_f-m_i}{m_i\times \left(M_D-M_H\right)}}\times 1000$$

(1)

where $A$ represents the accessible hydroxyl content (in mol/kg), $m_i$ and $m_f$ (in g) are the initial and final dry masses, respectively. $M_D$ and $M_H$ are the molar masses (in g/mol) of deuterium and hydrogen, respectively, and the factor 1000 converts mol/g into mol/Kg. The wood dust showed an accessible hydroxyl content of 4.06 mmol/g.

2.2. Synthesis of Bio-Based Polyurethanes (BPU)

Three polyurethane samples were prepared with varying ratios of PEG and bio-oil as -OH sources, while maintaining a constant NCO/OH ratio of 1:1 (

Table 1. Polyurethane samples for kinetic studies.

Sample Name	Equivalence Ratio -NCO: -OH	% -OH Equivalence Ratio PEG(400): BO
p-MDI1:1PEG(400)	1:1	100–0%
p-MDI1:0.5PEG(400):0.5BO		50–50%
p-MDI1:1BO		0–100%

). The equivalent ratios between PEG to bio-oil equivalence ratio (in %) were 100/0, 50/50, and 0/100. The experimental procedure involved mixing the appropriate PEG and bio-oil in a 50 mL beaker: stirring the mixture at 600 rpm using a hot plate magnetic stirrer for 2 min to ensure homogeneity. Subsequently, polymeric p-MDI (at an NCO/OH ratio of 1:1) was added, and the mixture was stirred for an additional 30 s at 600 rpm to initiate the polymerization reaction. The reaction mixtures were then cured isothermally at 25 °C, 40 °C, 60 °C, and 80 °C using a temperature-controlled hot plate to monitor the effect of temperature on the curing process.

2.3. Fourier-Transform Infrared Spectroscopy (FTIR-ATR)

Fourier-transform infrared (FTIR) spectroscopy measurements were performed using a Thermo Scientific (Waltham, MA, USA) Nicolet 6700 FT-IR spectrophotometer equipped with an attenuated total reflection (ATR) accessory and controlled by OMNIC 7.3 software. Every FTIR scan began with a background scan, followed by measurement of the sample, to achieve precise baseline correction. To provide a consistent environment of 33–35% relative humidity and 20 °C, an air dehumidifier was utilized, reducing the impact of surrounding moisture on the NCO absorption band (2270–2250 cm⁻¹). The IR spectra were collected in the wavenumber range from 400 cm⁻¹ to 4000 cm⁻¹ at a resolution of 4 cm⁻¹ and 32 scans per sample. Baseline spectra were collected prior to each sampling event. The primary absorption bands of interest were the isocyanate (-NCO) stretching band, typically found near 2270 cm^–−1, and the carbonyl (C=O) stretching band with a peak near 1673 cm⁻¹.

FTIR spectra were processed with OriginLab software, OriginPro 2025 (10.2), to facilitate the analysis of kinetic parameters. The spectra were adjusted using the Asymmetric Least Squares Smoothing (ALS) method, as provided by the software, which adjusts baseline variations without requiring anchor points. Parameters were optimized to ensure accurate correction. Subsequently, the spectra were normalized using min-max normalization to standardize peak intensities across samples. The bands attributed to -NCO and -CH₂, at approximately 2260 cm⁻¹ and 2860 cm⁻¹, respectively, were integrated using both the Integration Gadget and the Peak Analyzer tools in Origin to quantify reaction progress. The Region of Interest (ROI) was manually defined for each peak, and peak areas were calculated and recorded to facilitate subsequent analysis of isocyanate conversion.

2.4. Determination of Reaction Order and Rate Constant Using nth-Order Kinetics

To evaluate the conversion of isocyanate (-NCO) groups during the polyurethane curing reaction, the area ratio between the -NCO stretching band at 2260 cm⁻¹ and the -CH₂ stretching band at 2860 cm⁻¹, which was used as a reference band, was calculated using FTIR spectra at each time point. This NCO/CH₂ ratio reflects the relative consumption of isocyanate groups over time, assuming negligible side reactions. The isocyanate conversion, p, was calculated using the following equation:

$$Isocyanate conversion \left(p\right)=1-{\displaystyle \frac{{\left(\raisebox{1ex}{$A_{-NCO}$}\!\left/ \!\raisebox{-1ex}{$A_{{-CH}_2}$}\right.\right)}_{t=t}}{{\left(\raisebox{1ex}{$A_{-NCO}$}\!\left/ \!\raisebox{-1ex}{$A_{{-CH}_2}$}\right.\right)}_{t=0}}}$$

(2)

where ${\left(A_{-NCO}/A_{{-CH}_2}\right)}_{t=t}$ is the ratio at time t, and ${\left(A_{-NCO}/A_{{-CH}_2}\right)}_{t=0}$ represents the ratio at the initial reaction time. Since no FTIR data were collected at exactly zero time, this initial value was estimated by extrapolating a linear fit from the early data points in the NCO/CH₂ ratio curve. This approach provides a consistent baseline for calculating conversion throughout the reaction period.

The kinetics of polyurethane curing were effectively described by an nth-order reaction model, which characterizes the autocatalytic behavior commonly observed in polyurethane and other thermosetting polymers. The reaction rate in this nth-order model is given by:

$${\displaystyle \frac{d\alpha }{dt}}=A \exp \left(-{\displaystyle \frac{E}{RT}}\right){\left(1-\alpha \right)}^n$$

(3)

where α is the degree of conversion calculated from the experimental conversion data for kinetic analysis, t is the reaction time, A is the pre-exponential factor, E is the activation energy, R is the gas constant, T is the temperature, n is the reaction order, and $K$ is the rate constant at a given temperature. Experimental conversion data derived from FTIR analysis were used to compute α as a function of time. MATLAB R2022b Software was employed to implement non-linear curve fitting procedures, enabling the determination of the optimal values of n and the rate constant k at each curing temperature. The integrated form of the kinetic equation,

$${\int }_0^t\left(T\right)dt=K\times {\int }_0^{\alpha }{\displaystyle \frac{d\alpha }{{\left(1-\alpha \right)}^n}}$$

(4)

was solved numerically in cases where analytical solutions were not available. Conversion curves were generated for each formulation and temperature, facilitating the calibration of the model. Iterative fitting routines in MATLAB refined the parameters n and k to minimize the deviation between experimental and theoretical conversion profiles.

The temperature dependence of the rate constant K was further characterized using the Arrhenius equation:

$$\mathrm{K}\left(\mathrm{T}\right)=\mathrm{A\; exp}\left(-{\displaystyle \frac{E}{RT}}\right)$$

(5)

Taking the natural logarithm of both sides linearizes this equation:

$$\ln \left(K\right)=\ln \left(A\right)-{\displaystyle \frac{E}{R}}·{\displaystyle \frac{1}{T}}$$

(6)

To extract the activation energy (E) and pre-exponential factor (A), ln(K) values were plotted against 1/T (K⁻¹) to construct Arrhenius plots for each composition. The slope and interception of the linear regression line provided E and ln(A), respectively. All T values were converted to reciprocal temperature (1/T), and corresponding K values to their natural logarithms.

2.5. Synthesis of Bio-Based Polyurethane Wood Composite Samples

Wood composite panels were fabricated using a 4 × 4-inch aluminum press mold with a fixed wood dust content of 8.5 g per panel. The amounts of bio-oil, PEG, and p-MDI were calculated based on equivalent ratio controls to ensure a stoichiometric balance between hydroxyl and isocyanate groups. The mixture was poured into the mold and subjected to a pressure of 5 tons-force for 1 h, followed by ambient curing for 24 h (

Table 2. Samples with different equivalent ratios between -NCO and -OH groups, and bio-oil and wood dust for panels.

Sample Name	Equivalent Ratio -NCO: -OH	% -OH Equivalent Ratio Bio-Oil: Wood Dust
Panel|p-MDI1.5:1(30BO-70WD)	1.5:1	30–70%
Panel|p-MDI1.5:1(50BO-50WD)		50–50%
Panel|p-MDI2:1(30BO-70WD)	2:1	30–70%

).

For extrusion, the resin and wood fiber (50 g) were blended using a grinder (Iceko-M200B, 200 W, Zhongshan, China) and processed through a 250 W single-screw extruder (Robot Digg, Shanghai, China) equipped with a 20 mm Ø barrel (L/D = 10) and a 9.45 mm Ø die. The formulation was prepared by calculating equivalent ratios based on the available hydroxyl (-OH) groups from the bio-oil and wood dust, as well as the isocyanate (-NCO) groups from p-MDI, ensuring stoichiometric control over the bio-based polyurethane system. The components were pre-mixed and manually fed into the extruder, which operated at a controlled screw speed of 17 rpm to promote uniform dispersion and minimize shear-induced variations. To prevent premature curing, the barrel temperature was maintained at approximately 25 °C by circulating ice water (380 L/h) through a tightly coiled 6 mm Ø copper tube wrapped around the barrel. The extrudate was collected after approximately 15–20 min of continuous operation, cooled to ambient temperature, and conditioned at room temperature for 48 h prior to mechanical and morphological testing.

2.6. Study of the Mechanical Properties of the Wood Composites

3-Point bending testing was performed on the panels after curing. Each panel was cut into rectangular specimens measuring 1 inch by 0.25 inches. Six samples were obtained from each panel: three were tested in dry conditions, while the remaining three were submerged in water at 20 ± 1 °C for 24 h and tested within 30–60 min of removal, following the water-soaked conditioning specified in ASTM D1037-12 [86] (Section 6 Moisture Content and Conditioning Requirements, 6.3.3. Water Soaked). This procedure is designed to assess the effects of short-term moisture. All mechanical testing was performed according to ASTM D1037-12 standards using a three-point bending setup on an Instron universal testing machine, Model 5982 Floor Model Series for Mechanical Testing (Norwood, MA, USA), equipped with a 100 N capacity load cell.

Compression testing experiments were conducted for the extruded samples by subjecting the cured composite samples to uniaxial loading in accordance with ASTM D695-23 [87], the standard for evaluating the mechanical properties of wood-based fiber and particle panel materials. Using a table saw, five specimens measuring 23.50 mm were cut from the extrudates. Tests were conducted on an Instron 5500R-1137 Universal Testing Machine (Norwood, MA, USA), equipped with a 50 kN load cell, and operated at a crosshead speed of 0.1 mm/min. Experimental data were collected and processed using the Bluehill Universal material testing software.

Flexural tests (ASTM D1037-12) and compressive tests (ASTM D695-23) were not meant to be directly compared in numerical terms, as they utilize distinct specimen shapes and stress conditions; rather, they served as complementary assessments to evaluate the mechanical performance of the formulations.

2.7. Scanning Electron Microscope (SEM)

Samples were tested for microstructural analysis using a Zeiss EVO-10 scanning electron microscope (SEM) (Oberkochen, Germany) operated at an accelerating voltage of 20 kV. Prior to imaging, the samples were sputter-coated with gold nanoparticles to ensure surface conductivity and enhance image resolution during SEM examination.

3. Results

3.1. Kinetic Assessment of Polyurethane Cure via ATR-FTIR and nth-Order Reaction Modeling

The polyurethane synthesis shown in Figure 1 follows a typical poly-addition reaction mechanism in which hydroxyl (-OH) groups from PEG-400 and bio-oil react with the isocyanate (-NCO) groups of p-MDI to form urethane linkages (-NHCOO-). This reaction is characteristic of polyurethane formation, where the isocyanate and hydroxyl functionalities undergo nucleophilic addition, resulting in the formation of a polymeric network.

The FTIR spectra presented in Figure 2 illustrate the evolution of the reaction, aligning with the expected polyaddition process and reinforcing the role of hydrogen bonding in stabilizing the developing polyurethane structure. The image shows the temporal evolution of polyurethane formation at 25 °C, highlighting the progressive consumption of isocyanate (-NCO) groups and the corresponding development of urethane linkages. The spectra exhibit a strong absorption band at approximately 2250 cm⁻¹ at the initial reaction stage (time = 1 min), which gradually decreases in absorbance over time, indicating the reaction of -NCO groups with hydroxyl (-OH) functional groups from PEG 400 and/or bio-oil. Simultaneously, the formation of urethane bonds is evidenced by the increasing intensity of the carbonyl (C=O) stretching peak around 1700 cm⁻¹, as well as the emergence of amide II bands (~1520 cm⁻¹) and C-N stretching bands (~1300 cm⁻¹). The broad absorption region between 3200 cm⁻¹ and 3600 cm⁻¹ further supports the development of hydrogen bonding interactions within the polymer network, which intensifies at later time intervals (time = 120–180 min).

The conversion versus time plots, Figure 3, analyzed at different temperatures, provide critical insights into the kinetics of polyurethane formation from PEG-400, bio-oil, and p-MDI. The experimental data and fitted curves follow a characteristic sigmoidal trend, where reaction rates increase with temperature due to enhanced molecular mobility and reactivity. The fitted curves obtained from non-linear regression, based on the Arrhenius equation and represented with scatter tendency, demonstrate a good correlation with the experimental values, allowing for the calculation of kinetic parameters such as the rate constant (k1) and reaction order (n). As expected, an increase in temperature leads to a significant enhancement in conversion rates, highlighting the temperature dependence of the reaction mechanism.

Notably, a distinct shift in the kinetic profile is observed when bio-oil is incorporated into the reaction system (upright and bottom plots). The systems containing bio-oil exhibit higher conversion rates, indicating that bio-based polyols enhance reaction efficiency. This behavior can be attributed to the presence of various reactive hydroxyl functionalities in bio-oil, which facilitate the formation of urethane bonds. The kinetic parameters extracted from the fitted curves suggest a higher effective reaction order in the bio-based systems, consistent with previous studies on the reactivity of lignin-derived polyols in polyurethane synthesis.

The reaction order (n) and rate constant (K), as illustrated in

Table 3. Samples with different equivalent ratios between -NCO and -OH groups, and bio-oil and wood dust for extrusion.

Sample Name	Equivalent Ratio -NCO: -OH	% -OH Equivalent Ratio Bio-Oil: Wood Dust
Ext|p-MDI1:1(30BO-70WD)	1:1	30–70%
Ext|p-MDI1:1(50BO-50WD)		50–50%
Ext|p-MDI1:1(60BO-40WD)		60–40%
Ext|p-MDI1.5:1(30BO-70WD)	1.5:1	30–70%
Ext|p-MDI1.5:1(50BO-50WD)		50–50%
Ext|p-MDI1.5:1(60BO-40WD)		60–40%

, show a significant variation across different formulations and temperatures. For the PEG-400 and p-MDI system, the reaction order gradually increases with temperature, indicating a progressive shift in the reaction mechanism as molecular mobility and secondary reactions become more prominent. However, when bio-oil is introduced into the system, the reaction order exhibits a more pronounced temperature dependence, suggesting a more complex reaction pathway, possibly due to the heterogeneous nature of bio-oil and its diverse hydroxyl functionalities or the mixture when taking the samples. Additionally, the rate constant (K) for bio-oil-containing systems is consistently higher than that of the PEG-based formulations, reinforcing the hypothesis that bio-oil enhances reaction kinetics. The system containing bio-oil exhibits the highest K values, suggesting that bio-based polyols facilitate urethane bond formation more efficiently than synthetic polyols. This behavior is consistent with prior studies highlighting the catalytic role of lignin-based polyols and their ability to modify polyurethane curing behavior, potentially enabling lower-temperature processing and improved sustainability in polymer manufacturing [88,89,90,91,92,93,94,95,96,97].

The Arrhenius analysis provides a deeper understanding of the curing kinetics of polyurethane formulations by quantifying the temperature dependence of the reaction rate constant (K). The activation energy and pre-exponential factor values presented in

Table 4. Calculated parameters for different BPU formulations.

Temperature	p-MDI1:1PEG(400) 1:1		p-MDI1:0.5PEG (400):0.5BO 1:0.5:0.5		p-MDI1:1BO 1:1
°C	n	K [s−1]	n	K [s−1]	n	K [s−1]
25	2.024	4.74 × 10−4	3.133	8.12 × 10−4	1.638	1.13 × 10−3
40	2.139	7.90 × 10−4	2.957	2.06 × 10−3	1.010	1.41 × 10−3
60	2.268	1.17 × 10−3	2.386	3.48 × 10−3	1.010	2.45 × 10−3
80	2.767	2.98 × 10−3	1.682	4.16 × 10−3	-	-
Average	2.30 ± 0.33	1.35 × 10−3 ± 1.12 × 10−3	2.54 ± 0.65	2.63 × 10−3 ± 1.49 × 10−3	1.22 ± 0.36	1.66 × 10−3 ± 6.96 × 10−4
R2	0.87	0.82	0.94	0.98	0.75	0.90

demonstrate the impact of varying polyol sources on the system’s reactivity. The PEG400 and p-MDI system exhibits the lowest activation energy (27.86 kJ/mol) and the smallest pre-exponential factor (34.1), suggesting that the curing process requires a higher thermal input to overcome the energy barriers associated with urethane bond formation. As bio-oil is gradually introduced into the formulation, first in combination with PEG400 and eventually as the sole hydroxyl source, a consistent decrease in activation energy is observed. This trend reflects the enhanced reactivity of bio-oil at room temperature, which accelerates the curing kinetics and reduces the energy required to initiate the reaction (

Table 5. Activation Energy and Pre-exponential factor for different Polyurethane compositions.

Composition	Equivalent Ratio	Activation Energy (KJ/mol)	Pre-Exponential Factor
p-MDI:PEG(400)	1:1	27.86	34.14
p-MDI:PEG(400):BO	1:0.5:0.5	25.52	29.64
p-MDI:BO	1:1	18.44	1.83

).

The most striking effect is observed in the bio-oil + p-MDI system, which exhibits the lowest activation energy and pre-exponential factor. These values indicate that as the proportion of bio-oil increases, the system becomes more reactive and kinetically efficient, with a more complex reaction mechanism, likely due to the presence of readily accessible hydroxyl groups and lower steric hindrance compared to conventional polyols. While the lower pre-exponential factor suggests a less temperature-dependent reaction pathway, the faster curing response at lower temperatures aligns with the potential of bio-oil to facilitate energy-efficient polyurethane synthesis. These results support the viability of bio-oil as a sustainable and reactive alternative for polyurethane formulations, enabling tunable kinetics based on its concentration. While bio-oil enhances the crosslinking density and final polymer structure, careful control over processing conditions is required to optimize curing behavior.

The reduced activation energy observed for bio-oil–based formulations may also be influenced by the presence of residual carboxylic acids inherent to lignocellulosic bio-oils. These acidic groups can promote proton-assisted catalytic pathways, accelerating isocyanate–hydroxyl addition and related side reactions, thereby lowering the apparent activation energy [94,98].

3.2. Processing, Mechanical Evaluation, and Microstructure of Bio-Based Polyurethane Wood Composite Formulations

Mechanical testing was conducted to evaluate the impact of formulation variables on structural performance. Specifically, compression tests were performed on bio-based polyurethane wood composite samples with varying wood dust content to assess the relationship between filler concentration, polymer crosslinking efficiency, and resulting mechanical properties.

Three-point bending experiments (

Table 6. BPU wood composite panels results: 3-point bending testing.

Condition	Sample Name	Strain [%]	Stress [MPa]	Modulus [MPa]
Dry	Panel|p-MDI1.5:1(30BO-70WD)	0.91 ± 0.3	6.83 ± 4.8	799.17 ± 312
	Panel|p-MDI1.5:1(50BO-50WD)	0.93 ± 0.1	13.10 ± 2.7	1542.41 ± 196
	Panel|p-MDI2:1(30BO-70WD)	1.35 ± 0.5	11.50 ± 4.3	1142.01 ± 303
Wet	Panel|p-MDI1.5:1(30BO-70WD)	1.36 ± 0.3	7.22 ± 4.8	599.42 ± 313
	Panel|p-MDI1.5:1(50BO-50WD)	1.87 ± 0.3	14.26 ± 4.3	1048.51 ± 225
	Panel|p-MDI2:1(30BO-70WD)	1.62 ± 0.5	8.42 ± 2.9	699.26 ± 223

) were conducted first, and compression-made samples (

Table 7. Extrusion results for different compositions of BPU wood composites and compression properties.

Sample Name	Compressed Screw	Non-Compressed Screw	Strain [%]	Stress [MPa]	Modulus [MPa]
Ext|p-MDI1:1(30BO-70WD)	Yes	No	6.3 ± 1.1	11.3 ± 1.7	311 ± 22
Ext|p-MDI1:1(50BO-50WD)	Yes	No	4.2 ± 0.7	6.7 ± 0.5	292 ± 38
Ext|p-MDI1:1(60BO-40WD)	Yes	No	3.5 ± 0.9	10.5 ± 4.1	431 ± 118
Ext|p-MDI1.5:1(30BO-70WD)	No	No	-	-	-
Ext|p-MDI1.5:1(50BO-50WD)	No	Yes	5.8 ± 0.9	32.2 ± 1.5	875 ± 126
Ext|p-MDI1.5:1(60BO-40WD)	No	No	-	-	-

) were prepared as a complementary assessment, allowing us to evaluate the same formulations under different loading modes rather than to directly compare absolute flexural and compressive values. This shift from weight-based to chemically balanced formulations enabled more precise control over the formation of polyurethane networks. By adjusting the equivalent ratios of p-MDI, bio-oil, and wood flour, the panel systems provided a deeper understanding of how resin chemistry, filler distribution, and moisture absorption collectively influence mechanical performance and structural behavior under flexural stress.

The strain data suggest that wet samples exhibited higher strain values than their dry counterparts, with the highest strain recorded in the Panel|p-MDI1.5:1(50BO-50WD) (Wet) samples (1.87%). This can be attributed to the plasticizing effect of absorbed water, which reduces intermolecular forces within the polymer network, allowing for more significant deformation under applied stress [99,100]. Conversely, the lowest strain was observed in the Panel|p-MDI1.5:1(30BO-70WD) (Dry) samples (0.91%), indicating that a higher wood flour content enhances stiffness and limits elongation, but potentially makes the material more brittle. These findings highlight the role of wood dust as a reinforcing agent while also demonstrating that excessive filler content may introduce more interfacial defects, leading to stress concentration and reduced ductility.

The strength and modulus results further support these observations. The highest average strength was observed in the Panel p-MDI1.5:1 (50BO-50WD) (Wet) samples, indicating that the optimized bio-oil content in this formulation contributes to effective polymer-filler interactions, resulting in improved load distribution. The lowest strength was observed in the Panel|p-MDI1.5:1(30BO-70WD) (Dry) samples, indicating that a higher wood flour content may compromise stress transfer within the composite by creating voids or weak bonding sites. The modulus values follow a similar trend, with the Panel|p-MDI1.5:1(50BO-50WD) (Dry) samples displaying the highest modulus, suggesting a more tightly crosslinked polymer network and improved load-bearing capability. The lowest modulus was recorded in Panel p-MDI1.5:1 (30BO-70WD) (Wet) samples, indicating that a higher wood flour content and increased water absorption can significantly reduce the material’s overall stiffness.

Since both bio-oil and wood dust serve as sources of hydroxyl groups (-OH), their contribution to the polymer network is crucial in determining the extrusion behavior and final material properties. To optimize processability while maintaining mechanical integrity, the study focused on two different equivalent ratios: Ext|p-MDI1:1(BO-WD) and Ext|p-MDI1.5:1(BO-WD), evaluating how these ratios influence composite formation, adhesion, and flow characteristics.

To assess extrusion performance, the study analyzed the relationship between resin composition and processing stability, considering the torque and screw speed data as indicators of material flow and cohesion. The Ext|p-MDI1.5:1(50BO-50WD) sample exhibited the best extrusion quality, with a smooth, continuous extrudate, reduced surface roughness, and fewer visible cracks than the other compositions (Figure 4). This result suggests that increasing the isocyanate-to-hydroxyl ratio improves resin infiltration and adhesion, enhancing the homogeneity of the final composite. The increased polymeric MDI content also likely enhanced the wetting and adhesion of the resin to the wood particles, facilitating better cohesion and processability during extrusion. In contrast, Ext|p-MDI1:1 formulation displayed visible structural irregularities, particularly at higher wood flour contents (Ext|p-MDI1:1(30BO-70WD)), where increased viscosity led to processing challenges and surface cracking. Similarly, the Ext|p-MDI1:1(60BO-40WD) bio-oil-to-wood formulation exhibited irregular extrudate morphology, likely due to insufficient polymer coverage and reduced resin mobility (Figure 4). These findings confirm that balancing polymer crosslinking with filler content is crucial for achieving optimal processability and composite integrity, underscoring the importance of chemical compatibility and resin–filler interactions in bio-based polyurethane systems.

Mechanical compression testing for the formulations Ext|p-MDI1.5:1(30BO-70WD) and Ext|p-MDI1.5:1(60BO-40WD) could not be performed, as these compositions failed to produce extrudates suitable for testing. The poor processability of these samples, characterized by inconsistent flow and incomplete extrusion, underscores the limitations associated with high wood flour content or insufficient resin mobility, particularly when polymer crosslinking and filler dispersion are not optimally balanced.

Among the formulations successfully tested, Ext|p-MDI1.5:1(50BO-50WD) exhibited superior mechanical performance, with the highest stress (32.2 ± 1.5 MPa) and modulus (875 ± 126 MPa), along with good strain values. These results show qualitative agreement with the three-point bending data presented in Table 6, as both tests identify Panel|p-MDI1.5:1(50BO-50WD)/Ext|p-MDI1.5:1(50BO-50WD) as the composition with the most favorable combination of strength, stiffness, and strain capacity under their respective loading conditions, particularly with the corresponding Panel|p-MDI1.5:1(50BO-50WD) samples, which also showed high stress (14.26 ± 4.3 MPa) and strain (1.87 ± 0.3%). This consistency across both compression and bending tests confirms the favorable performance of the 50:50 bio-oil to wood dust formulation when processed under a 1.5:1 isocyanate-to-hydroxyl ratio.

Conversely, samples with either increased wood content (Ext|p-MDI1:1(30BO-70WD)) or increased bio-oil content (Ext|p-MDI1:1(60BO-40WD)) displayed lower mechanical performance, echoing the trends observed in Table 6. In these cases, high wood dust content likely introduced greater porosity and hindered stress transfer, while excessive bio-oil may have reduced structural stiffness due to lower crosslinking density. The alignment of extrusion-based mechanical testing with panel test results reinforces the role of formulation balance in optimizing both processing stability and mechanical integrity in bio-based polyurethane wood composites.

This establishes a clear connection between extrusion processability and final mechanical performance, demonstrating that optimized resin-to-filler ratios are essential for achieving structural integrity and efficient manufacturing. The SEM images highlighted how higher wood dust content led to increased porosity and reduced adhesion, directly impacting the extrusion behavior observed in the different formulations. The Ext|p-MDI1.5:1(50BO-50WD) provided the most stable extrusion process and minimized defects that could otherwise compromise the composite’s modulus and specific modulus.

The modulus trends observed in the tests correlate closely with the internal structure revealed by SEM imaging, linking mechanical reinforcement to the quality of filler dispersion and interfacial adhesion (Figure 5).

The Scanning Electron Microscope (SEM) analysis provides direct microstructural evidence supporting the mechanical results. The 40 wt% wood-dust composite exhibited a compact and well-bonded microstructure, characterized by uniform particle distribution, reduced void content, and strong interfacial adhesion between the lignocellulosic filler and the polyurethane matrix. These morphological features are consistent with the enhanced compressive and flexural strengths observed for this formulation, as efficient stress transfer occurs through the continuous polymer network and well-anchored wood particles. In contrast, the 50 wt% sample displayed noticeable porosity and interparticle cavities, suggesting incomplete resin wetting and limited matrix penetration. Such interfacial discontinuities correlate with the slight decrease in mechanical performance at higher filler loadings, confirming that excessive wood flour content impedes polymer infiltration and weakens stress propagation across the composite. Overall, the SEM observations corroborate the mechanical behavior trends, establishing a clear structure–property relationship governed by filler dispersion and interfacial bonding quality.

4. Conclusions

This research highlights a distinctive combination of kinetic modeling and mechanical characterization to uncover formulation-specific curing pathways and correlations between structure and properties in bio-oil-based polyurethane composites. By integrating in situ FTIR kinetic analysis with thermally driven n-th-order modeling, the study establishes direct quantitative relationships among reaction rate, activation energy, and the progression of mechanical integrity at various NCO:OH ratios. This combined methodology provides a comprehensive understanding of how chemical conversion affects the development of stiffness, strength, and interfacial adhesion in bio-polyurethane materials. In addition to providing essential insights into the curing behavior of bio-oil/wood–flour mixtures, the findings also present a predictive framework for customizing resin formulations and processing conditions to achieve the desired performance in extrusion-based additive manufacturing and other sustainable construction endeavors.

This study presents a comprehensive examination into the synthesis, curing behavior, mechanical performance, and processability of bio-based polyurethane (BPU) composites formulated from poplar-wood-derived bio-oil, wood dust, and polymeric methylene diphenyl diisocyanate (p-MDI). Utilizing FTIR-ATR spectral analysis in conjunction with n-th order kinetic modeling, this research quantitatively explained the curing characteristics of bio-based polyurethane (BPU) composites produced from poplar-derived bio-oil and wood dust. The findings indicate that adding bio-oil significantly lowers the activation energy compared to traditional PEG-400 systems, facilitating faster curing kinetics at reduced temperatures. This increased reactivity results from the high quantity of phenolic and aliphatic hydroxyl groups present in bio-oil, confirming its function as a multifunctional polyol that enhances the formation of dense and thermally stable polyurethane networks.

By integrating kinetic modeling with thorough mechanical characterization, this study fills a critical gap concerning the relationship between reaction kinetics and mechanical properties in hybrid bio-polymer systems. The optimized formulation, characterized by an NCO:OH ratio of 1.5:1 and a 50:50 ratio of bio-oil to wood dust, achieved the optimal combination of stiffness, strength, and moisture resistance. SEM analysis confirmed that these enhanced properties resulted from better filler dispersion, strong interfacial bonding, and reduced porosity, which collectively facilitate effective stress transfer throughout the composite matrix. Conversely, excess bio-oil or wood dust impaired matrix continuity and weakened mechanical integrity, underscoring the significance of accurate resin-to-filler balancing in the design of sustainable composites.

Extrusion experiments further validated the processability of the optimized BPU system, resulting in a continuous, homogeneous, and defect-free material flow. Although layer-by-layer additive manufacturing was not executed, the successful extrusion of the reactive blend under controlled circumstances strongly indicates its scalability and potential for adaptation in large-format 3D printing applications. This discovery positions bio-oil-based polyurethane composites as a valuable feedstock for sustainable construction, especially in the extrusion-based production of structural elements and modular housing.

Although 3D printing trials were not conducted, the extrusion and curing behavior indicate potential applicability in extrusion-based additive manufacturing.

Overall, this research lays a foundational framework for linking curing kinetics, composition, and mechanical properties in bio-derived polyurethane systems. The combination of renewable resources, predictable curing characteristics, and strong mechanical performance paves the way for circular, high-performance building materials. These findings support the viability of bio-based polyurethanes as scalable, circular alternatives to petroleum-derived materials in the next generation of eco-efficient building technologies. By demonstrating the successful extrusion of bio-oil and wood-dust composites, this study advances the development of sustainable polymer systems capable of reducing dependence on petrochemical resins and accelerating the transition toward carbon-conscious, resource-efficient construction materials.