Statistical strategies for decision-making regarding the quality of particleboards with glass wool

Fassarella , M. V., Chaves, I. L. S., Paes , J. B., Lelis, R. C. C., Polvarini , G. S., Barros Junior , U. O., and Gonçalves, F. G. (2026). "Statistical strategies for decision-making regarding the quality of particleboards with glass wool," BioResources 21(2), 4110–4134.

Abstract

This study evaluated multivariate statistical strategies to select critical properties for the performance of particleboards bonded with urea-formaldehyde (UF) adhesive modified with glass wool residues. Panels were produced with six different proportions of glass wool incorporated into the UF adhesive (0%, 3.34%, 4.93%, 6.52%, 9.49%, 12.35%). These panels were characterized by physical, mechanical, fire-retardant, and acoustic properties. Three statistical tools were applied: hierarchical cluster analysis, principal component analysis (PCA), and Pearson correlation. PCA explained 80.3% of the total variance, revealing distinct patterns among treatments, especially at the lowest and highest filler contents. The correlation matrix showed the interdependence between the rheological properties of the adhesive and the final composite performance. Glass wool as a filling material, in the proportion of 3.34% of the adhesive, provided the best performance among the panels, as it promoted balance between mechanical, physical and acoustic properties. Up to the limit of 6.52% glass wool contributed to improving fire resistance without significant changes in mechanical strength but reduced dimensional stability due to changes in adhesive rheology. The combination of multivariate analyses provided a robust approach to identify key attributes and guide the formulation of panels with enhanced technical and functional performance.

Download PDF

Full Article

Statistical Strategies for Decision-making Regarding the Quality of Particleboards with Glass Wool

Michelângelo V. Fassarella ,^a Izabella Luzia S. Chaves ,^b Juarez B. Paes ,^a Roberto Carlos C. Lelis ,^c Geovanna S. Polvarini ,^a Udson O. Barros Junior ,^dand Fabricio G. Gonçalves ^a,*

DOI: 10.15376/biores.21.2.4110-4134

Keywords: Waste; Particleboard; Glass wool; Recycling; Alternative raw material

Contact information: a: Department of Forest and Wood Sciences, Federal University of Espírito Santo, Jerônimo Monteiro, Espírito Santo, Brazil; b: Department of Agrarian Sciences, Center for Exact and Technological Sciences, State University of Montes Claros, Janaúba, Minas Gerais, Brazil; c: Institute of Forests, Products Forests Department, Federal Rural University of Rio de Janeiro, Seropédica, Rio de Janeiro, Brazil; d: State University of the Tocantina Region of Maranhão, Maranhão Campus, Imperatriz, Brazil; *Corresponding author: fabricio.goncalves@ufes.br

Graphical Abstract

INTRODUCTION

In 2024, with a strengthened forestry scenario, Brazil reached the milestone of 10 million hectares of planted forests. This achievement reflects the availability of large-scale renewable raw material and the country’s commitment to advancing toward a more sustainable, low-carbon economy (IBÁ 2024).

To supply the flooring and panel industry, 4% of the planted forest area in 2023 was allocated to this segment. Among the highest-quality reconstituted wood panels, MDF has stood out in the market. Sales for MDF grew by 1.5%, totaling 7.1 million m³ (IBÁ 2024). In contrast, particleboards registered a 2.7% drop in domestic sales and continue to be the focus of innovations aimed at improving performance. These improvements involve modified resins or the incorporation of lignocellulosic additives (Mahieu et al. 2021; Papadopoulou et al. 2024; Mensah et al. 2025).

In the construction sector, these panels are part of industrialized building technologies that, while boosting productivity and profits, often impose performance challenges on the materials used (Haas et al. 2022). In this context, reconstituted panels gain relevance due to their functional versatility (Santos et al. 2021; Fehrmann et al. 2022, Cazella et al. 2024).

Adequate and sufficient resin application, as well as the control of pressing parameters, are key factors to ensuring the physical-mechanical integrity of the panels (ABNT, 2018). Among the core components of the reconstituted panel sector is urea-formaldehyde resin, which is widely used due to its low cost, fast curing, and high stiffness (Benhamou et al. 2022; Wibowo et al. 2022; Wang et al. 2023). However, its limitations in terms of moisture resistance, formaldehyde emissions, and durability have motivated further studies (Yildirim and Candan 2021; Kelleci et al. 2022; Moutousidis et al. 2023).

The incorporation of mineral materials such as glass wool and metallic nanoparticles emerges as a promising and effective strategy to alter technological properties of the adhesive, such as pH, viscosity, and gel time, in addition to improving the performance of reconstituted wood panels (Çavdar 2020; Liu et al. 2022; Khorramabadi et al. 2023, 2024; Gillela et al. 2024). Furthermore, the search for new adhesives (García et al. 2024; Gonçalves et al. 2008), resin enhancers (Sahoo et al. 2024), and methodologies that eliminate the need for synthetic adhesives (Guan et al. 2022), such as the use of alternatives like liquid glass (Lee and Thole 2018), has gained ground. Evaluating the application of glass wool residues in particleboards represents an innovative approach with significant technical and environmental potential.

Glass wool is produced by melting silicate materials in electric furnaces, followed by extrusion of the molten mass through rotary spinners under compressed air flow. The microfibers formed solidify upon rapid cooling, resulting in discontinuous, amorphous fibers (Tsukamoto et al. 2014). These fibers, with a density close to that of glass (≈2500 kg m⁻³) and diameters between 2 and 20 µm, are classified as solid and isotropic, with structural characteristics depending on the type of binder applied during the production stage (Meftah et al. 2019).

Chemically, glass wool consists primarily of silicon dioxide (SiO₂), sodium oxide (Na₂O), calcium oxide (CaO), aluminum oxide (Al₂O₃), ferric oxide (Fe_₂O₃), magnesium oxide (MgO), and boric oxide (B₂O₃), among other components (IARC 1988). This composition exhibits thermal stability, amorphous behavior, and a high surface area, which also contributes to its reactivity in alkaline systems (Yliniemi et al. 2020). Such attributes, combined with the importance of the material as a residue, make its incorporation into resin or cement matrices a viable alternative for modifying properties such as density, sound absorption, dynamic stiffness, and matrix-fiber adhesion (Adediran et al. 2021; Li et al. 2021; Machado et al. 2023). When incorporated into the adhesive as a filler material, it is relevant to assess the technological properties of the adhesive and its influence on the quality parameters of the panels produced.

The characterization of panels can be enhanced by methodologies that simultaneously evaluate the interactions among different material properties. The use of multivariate statistical techniques, such as principal component analysis (PCA), hierarchical cluster analysis, and Pearson correlation, enables the interpretation of complex data, the identification of patterns, and the establishment of significant relationships among properties (Fialho et al. 2022).

In this context, the present study aims to evaluate the effect of incorporating different levels of glass wool into urea-formaldehyde resin used in the production of particleboards. To this end, multivariate statistical tools were employed to understand correlations among performance variables, identify properties sensitive to the addition of the additive, and characterize behavioral patterns resulting from fiber incorporation.

The study was guided by two fundamental scientific questions: (i) under what conditions does the addition of glass wool positively influence the physical, mechanical, and thermal properties of the panels? (ii) what is the ability of different multivariate statistical approaches to identify patterns, correlations, and critical variables associated with panel performance?

These questions guide the critical analysis of the data and support the proposition of rational strategies to enhance the performance of lignocellulosic composites, focusing on the careful selection of target properties such as fire safety, acoustic insulation, dimensional stability, and mechanical resistance for industrial applications.

EXPERIMENTAL

Raw Material

The raw materials used in the production of medium-density particleboards consisted of Pinus spp. wood, glass wool residues, urea-formaldehyde (UF)-based adhesive, and catalyst. The UF adhesive and ammonium sulfate catalyst were purchased from a qualified supplier in the national market. The wood was obtained locally in the form of boards and subsequently crosscut into pieces of approximately 20 × 9 × 2 cm. Particle production was carried out using water-saturated wood, which was cut into flakes, air-dried, milled, and sieved, with particle sizes selected between 2.0 and 4.0 mm. The particles were then oven-dried until they reached 7% moisture and stored hermetically.

The glass wool, originating from urban waste, came from discarded household appliances in the southern region of Espírito Santo, Brazil. The mats were cut into segments of approximately 10 × 10 mm using a guillotine, and the material was manually cleaned to remove metallic particles and other possible impurities. The glass fibers present in the original mattress had a diameter ranging from 5 to 13 µm, with an estimated surface area between 8 and 20 g/m², and a density close to 2.5 g/cm³.

Adhesive preparation and technological characterization

For the adhesive formulations, glass wool was added as a filler to the commercial urea-formaldehyde (UF) adhesive of cascamite type (Redemite, Redelease, Brazil). The glass wool residues were incorporated at different concentrations into the adhesive using a homogenizing device (a homemade blender was used, with a speed of rotation between 3500 and 5000 revolutions per minute), resulting in six treatments (Table 1).

For each adhesive with different filler contents, it was first necessary to determine the solids content, i.e., the amount of resin solids contained in the commercial UF adhesive and in the respective filler concentrations studied.

The technological properties of the adhesives were evaluated in five replications, following previously established criteria (Table 2).

Table 1. Experimental Design According to the Content of Glass Wool Load in the Urea-Formaldehyde Adhesive and in the Particleboard Panels

Table 2. Analysis of the Technological Properties of Adhesives

The pH was measured in a controlled environment (≈ 25 °C) with a digital pH meter previously calibrated (standard buffer solutions – 4, 7 and 10), recording the value after the electrode stabilized in the sample.

The solids content was obtained by the ratio between dry mass (oven at 103 ± 2 °C for 12 h.) and initial mass of the sample (approximately 2 g of previously homogenized adhesive treatments).

The gel time of pure and filled urea-formaldehyde resin was evaluated considering the solids content and using ammonium sulfate (24%) as a catalyst, in a proportion of 2% based on the solids of the adhesive formed. The determination consisted of homogenizing the sample in a test tube under a water bath (90 °C), timing the interval (in seconds) until the transition from liquid to gel state, indicative of the beginning of curing.

Viscosity was determined according to D-2556-14 standard (ASTM, 2018), using a digital viscometer with a 3-inch stem and a rotation of 30 rpm, expressed in centipoise (cP).

Production of particleboard

The moisture level of the particles was initially determined using an infrared light-based moisture analyzer (LABORGLAS, MOC63u, São Paulo). Based on this value, the quantity of particles required to produce each panel was adjusted, with an additional 5% included to compensate for losses associated with particle moisture.

In all treatments, the same proportion of adhesive was used, corresponding to 12% of the oven-dry mass of the wood particles. The formulations were homogenized manually with the particles, following procedures described in the literature (Dhanapal et al. 2024; Fassarella et al. 2025), about 10 minutes at room temperature (≈28 ºC). All fractions (wood particles, adhesive and catalyst – ammonia sulfate, glass wool) were calculated based on the dry weight of the composite, for a target density of 700 kg/m³.

The particle mat was formed in a laboratory mold without a bottom, with dimensions of 42.5 × 42.5 cm and a removable cover, supported on an aluminum plate. The glued particles were uniformly distributed inside the mold, followed by pre-pressing on the forming cover, compacting and accommodating the particles on the aluminum sheet between two metal spacers with 1.25 cm edges.

The panels were produced using the compression molding method in a hydraulic press (SOLAB, SL12, Piracicaba) with heated plates. Pressing was performed at a temperature of 160 °C, a compression force of 72 tons (3.91 MPa), and a pressing time of 10 minutes.

After pressing and cooling at room temperature, the panels were trimmed to remove the edges and obtain the initial dimension of 40 × 40 cm. Three samples were produced for each treatment, totaling 18 panels, which were stored in a climate-controlled room (65 ± 5% relative humidity – RH and temperature of 25 ± 3 °C) until they reached an equilibrium moisture content of ≈ 12% (minimum 72 h), according to NBR 14810-2 (ABNT 2018).

The physical and mechanical tests proposed in this study and the number of replications for each test are presented in Table 3, along with the dimensions of the specimens according to the respective standards. Further details on sample preparation can be found in Fassarella (2025).

Table 3. Dimensions, Number of Test Specimens and Standards Associated with the Performed Test

Statistical analysis

Statistical analyses were conducted individually for each test, eliminating discrepant data based on the interquartile range (IQR) calculated from the 1^st and 3^rd quartiles, using Excel®. Values outside the minimum and maximum limits were considered outliers and excluded.

The experiments were conducted in a completely randomized design. All tests were initially subjected to regression analysis; since no statistical significance was observed, analysis of variance (ANOVA) followed by the Scott-Knott test (p < 0.05) was applied. Finally, hierarchical cluster analysis, principal component analysis (PCA), and Pearson correlation were performed to synthesize the relationships between variables, identify factors, and verify the strength of associations among parameters. Analyses were performed in R software (R Core Team 2025).

RESULTS AND DISCUSSION

The technological properties of the adhesives and the mechanical and physical properties of the panels are presented in Tables 4, 5, and 6. The data provides an overview of the general behavior of the formulated composites, highlighting the potential effect of mineral additive incorporation on the structure of the materials.

Table 4. Mean Values for the Technological Properties of the Adhesive

A gradual increase in pH was observed with the incorporation of glass wool to the urea-formaldehyde resin in the modified formulations. This behavior is associated with the mineral nature of the filler, whose composition based on silicates and metal oxides can give a slightly alkaline character to the adhesive. Studies indicate that increasing pH raises viscosity and gel time, which may compromise the physical and mechanical properties of the composites (Ghani et al. 2018; Kawalerczyk et al. 2023). To mitigate these effects, extending pressing time is recommended to ensure proper curing of resins with a more basic character (Ghani et al. 2018).

The solids content found was in the common range in urea-formaldehyde adhesives (59 to 66%) (Albuquerque et al. 2020). The treatment with 12.35% glass wool presented a higher value, evidencing the direct contribution of the mineral filler to the solid fraction of the formulation.

The increase in the solids content tended to favor the formation of a denser polymeric network, enhancing stiffness and adhesion. However, excessive elevations can intensify the viscosity of the system and compromise the processability during application (Pizzi, 2003). Similar results were reported by Moslemi et al. (2020), who observed the influence of the solid fraction on the gelation kinetics of modified UF adhesives, highlighting the role of concentration in the curing process.

The increase in viscosity with the use of additives is associated with the higher solid content and the modification of molecular interactions during adhesive formulation (Achchaq et al. 2009; Silva et al. 2024). The use of functional fillers, such as microfibrillated cellulose, nanoclays, and silica, demonstrates that moderate concentrations can optimize viscosity and improve the mechanical performance of composites (Lei et al. 2008; Veigel et al. 2011; Hosseini et al. 2020).

Controlled additions, generally below 2%, favor adhesive penetration and curing, while higher contents compromise adhesive flow, resin penetration into wood, and hinder industrial application (Roumeli et al. 2012; Salari et al. 2013; Kawalerczyk et al. 2025). These findings reinforce the importance of balancing formulation and reactivity (Hong and Park 2017), as also evidenced by the multivariate statistical analyses in this study, which identified the rheological parameters of the resin as determinant variables in the final performance of the panels.

The mechanical performance of particleboards is intrinsically linked to the adhesive formulation and its interaction with lignocellulosic particles (Arias et al. 2021; Baharuddin et al. 2023). The addition of fillers, such as glass wool, can directly influence structural cohesion and the efficiency of stress transfer within the composite matrix (Lima et al. 2020; Costa et al. 2024). In this study, when applied in moderate proportions, approximately 3.34%, glass wool provided better mechanical performance and dimensional stability to the panels (Table 5). High concentrations may compromise resin dispersion, leading to interfacial discontinuities and reduced mechanical strength (Baldin et al. 2016; Kawalerczyk et al. 2025). Therefore, careful adjustment of filler content is essential to optimize crosslinking performance and adhesion of bonded elements, preventing incompatibilities that affect the homogeneity and integrity of the material (Taquetti et al. 2023; Silva et al. 2024).

Table 5. Mean Values for the Mechanical Properties of the Particleboards Produced with Different Proportions of Glass Wool as Filler

The density of particleboards influences properties such as strength, dimensional stability, and adhesion efficiency (Nemli and Demirel 2007; Wong et al. 2020; Balea et al. 2022). In addition, its distribution, especially in the surface layers, affects fire performance (Harada et al. 2006; Najahi et al. 2023; Albert and Liew 2024). In the present study, the introduction of glass wool (a dense, friable, and inorganic material) modified the internal structure of the panel, reorganizing elements and filling voids without compromising compaction. However, higher filler contents may have reduced resin coverage on lignocellulosic particles, resulting in lower dimensional stability, as discussed by França et al. (2016) and Sozim et al. (2019).

Table 6. Mean Values Obtained for the Physical and Acoustic Properties of the Particleboards Produced with Different Proportions of Glass Wool in the Adhesive

The results indicated that increasing filler content can compromise the glue line due to the release of accumulated internal stresses (Bazzetto et al. 2019), while reducing resin content negatively influences moisture resistance (Iwakiri et al. 2012; Ayrilmis and Nemli 2017;). On the other hand, the presence of glass wool improved resistance to flammability, which can be attributed to its inorganic silica-rich composition, which acts as a thermal barrier, delaying flame spread (Evangelista et al. 2012; Lee and Thole 2018; Lemougna et al. 2020). This characteristic, combined with the surface hardness of the panels (Janka), reinforces the role of formulation in enhancing fire resistance (Harada et al. 2006; Najahi et al. 2023; Albert and Liew 2024), and in this specific case, higher levels (limited to 6.52%) contributed to increased fire resistance.

Overall, the behaviour of acoustic conversion efficiency was only marginally sensitive to variations in filler content, with improvements only when mechanical properties were simultaneously favoured. Acoustic damping effects were of low magnitude, reinforcing the need for integrated approaches between structural performance and vibroacoustic properties.

These data provide a fundamental basis for conducting multivariate statistical analyses, enabling the identification of patterns, correlations, and key variables in the characterization of the particleboards.

Cluster analysis

Hierarchical cluster analysis using Euclidean distance revealed the formation of four distinct clusters among the treatments (Fig. 1). Groups 1, 2, and 3 were composed, respectively, of treatments T2, T6, and T5, while Group 4, with greater multivariate stability, comprised treatments T4, T1, and T3.

Fig. 1. Dendrogram of the hierarchical cluster analysis of the treatments applied according to the content of glass wool added to the urea-formaldehyde adhesive (UF). Euclidean distance indicates dissimilarity; lower branches indicate greater similarity.

Treatment T2, with a lower filler content and greater Euclidean distance from the other groups, suggests that the adhesive formulation used in this case promoted a behavior that influenced the joint response of several parameters.

Similarly, treatments T6 (12.35%) and T5 (9.49%) formed separate branches, reflecting greater dissimilarity, which was likely due to the high concentration of additives. Elevated levels of inorganic fibers, such as glass wool, can impair the interaction among lignocellulosic particles, reduce internal bonding, and increase panel heterogeneity (Ülker and Burdurlu 2015; Sato et al. 2024).

Excessive filler levels in adhesives can increase viscosity, hinder homogeneous dispersion over wood particles, and compromise proper compaction. This not only reduces bonding uniformity but also weakens the structural integrity of the linkage, lowers workability, and may induce mechanical weaknesses and durability deficiencies in the composite (Dashti et al. 2012; Khanjanzadeh et al. 2019; Chen et al. 2022).

Although such effects interfere with the kinetics and efficiency of the curing process, thus justifying the isolation of treatment T6 in the hierarchical clustering, it is noteworthy that the panels still surpassed the minimum requirements of applicable standards, highlighting the structural resilience of the material.

The joint clustering of treatments T4, T1, and T3 indicates a zone of higher homogeneity, possibly related to the incorporation of moderate glass wool contents (between 0% and 6.52%), which may have favored a more uniform fiber dispersion within the lignocellulosic matrix of particleboards (Pintiaux et al. 2015).

According to Srivabut et al. (2018), the incorporation of mineral fillers such as nanoclay, talc, and calcium carbonate into lignocellulosic fiber composites can significantly enhance mechanical and physical properties, particularly when applied at optimal levels that improve dispersion within the polymeric matrix and interfacial interaction. These results corroborate the effects observed in treatment T2, in which the moderate glass wool incorporation into both the panel structure and the urea-formaldehyde adhesive likely contributed to improvements in the physical and mechanical properties of the material (Baharoǧlu et al. 2013; Cosereanu and Cerbu 2019).

This compatibility may have contributed to a more balanced performance in physical-mechanical properties. Similar outcomes were reported by Lu et al. (2016), who emphasized the role of chemical compatibility between fibers and wood particles in dimensional stability and internal bond strength of particleboards.

Principal components analysis (PCA)

The principal components analysis (PCA) revealed multivariate patterns among the physical-mechanical, chemical, and acoustic properties of the particleboards produced with varying levels of glass wool incorporated into urea-formaldehyde adhesive. The first two principal components accounted for 80.3% of the total variance, with Principal Component 1 (PC1) explaining 59% and Principal Component 2 (PC2) explaining 21.3% (Fig. 2).

Fig. 2. Principal components analysis (PCA) of the properties of particleboard panels. T: Treatment according to the content of glass wool added to the urea-formaldehyde adhesive. GT: Gel time. Visc: Viscosity; SC: Solid content; ρ(r-X): apparent density; CR: Compression ratio; WA: Water absorption; ST: Swelling in thickness; SB(MOE): Elastic modulus in static bending; SB(MOR): Modulus of rupture in static bending; IB: Internal bond; JH: Janka hardness; SP: Screw pulling (face); Flam: Flammability; Dl: Damping coefficient; ACE: Acoustic conversion efficiency; SII: Sound impact insulation.

In Fig. 2, treatments are represented by color-coded points according to increasing glass wool content. PC1 was strongly associated with variables such as thickness swelling (TS), water absorption (WA), viscosity (Visc), solid content (SC), pH, gel time (GT), damping coefficient (Dl), acoustic properties (ACE), and mechanical properties SB(MOE)) and SB(MOR)).

PC2 was related to apparent density (ρ r-X), compression ratio (CR), screw withdrawal resistance (SP), Janka hardness (JH), and internal bond (IB), thus reflecting the structural and strength-related characteristics of the panels.

The correlation coefficients of the variables with PC1 and PC2, which together explained most of the variance in the dataset, are presented in Table 7. These coefficients allow the identification of variables most sensitive to glass wool incorporation into urea-formaldehyde adhesive and support the selection of key attributes for determining particleboard performance.

Table 7. Correlation of the Evaluated Variables with the First Two Principal Components (PC1 and PC2) Obtained in the Principal Component Analysis (PCA)

The use of principal component analysis (PCA) in this study made it possible to identify complex interrelationships among the properties. The presented correlation matrix allows for understanding the degree of contribution of each property to the factorial directions that explain most of the variance observed in the data (Fialho et al. 2022). This multivariate approach not only highlights the properties most sensitive to the addition of glass wool but also identifies those with greater discriminating power among treatments, providing objective support for selecting key variables in the overall performance of particleboards.

PCA revealed a strong correlation of adhesive technological variables with the first principal component (PC1), such as viscosity (0.8241), pH (0.7801), and gel time (0.7039), indicating that the addition of glass wool significantly modified the behavior of the commercial resin.

The chemical changes in the adhesive system directly influenced the quality of the produced panels. Among the particleboard properties, high negative correlations were observed on PC1 for modulus of elasticity (–0.9441), modulus of rupture (–0.8694), screw withdrawal resistance (–0.908), flammability (–0.9331), and acoustic conversion efficiency (–0.8563), along with positive correlations with water absorption (0.9606) and thickness swelling (0.9733). These data indicate that modifying the adhesive formulation with glass wool incorporation plays a decisive role in the physical, mechanical, and thermal performance of particleboards, representing a potential strategy to optimize specific properties through rheological and chemical adjustments of the adhesive matrix.

This multivariate technique proved highly efficient by reducing an extensive set of variables into two principal components, which together represented 80.3% of the total variance of the experimental data—a percentage considered high and consistent with investigations involving lignocellulosic composites (Fehrmann et al. 2023).

Located in the lower-left quadrant, treatment T2 showed a strong association with the vectors modulus of rupture (SB(MOR)), acoustic conversion efficiency (ACE), and internal bond (IB), indicating satisfactory performance both in mechanical and acoustic terms.

This spatial configuration suggests that, despite the low glass wool content in the adhesive (3.34%), the treatment achieved a remarkable balance between structural stiffness and sound functionality. These results are consistent with the literature emphasizing the positive effects of moderate inorganic fiber incorporation in wood-based panels (Park et al. 2020; Yildirim and Candan 2021), indicating effective interaction between the lignocellulosic material and the modified adhesive.

T6 Treatment, located in isolation at the far right of PC1, showed high correlation with pH, viscosity, and solid content, evidencing the predominance of rheological factors. This behavior indicates that the high glass wool concentration (12.35%) significantly influenced adhesive characteristics, improving parameters such as gel time and colloidal stability. Although filler content reduced the uniformity of dispersion and adhesive penetration into wood particles, due to the lower amount of UF resin available in the bond (Hong and Park 2017), all panels met the minimum values required by the NBR 14-810-2 (2018) standard.

According to Calovi and Rossi (2023) and Iglesias et al. (2021), high levels of fibrous fillers promote increased adhesive viscosity and impair the wetting efficiency of lignocellulosic surfaces. Such limitations in anchorage capacity may justify the separation of T6 from variables associated with mechanical performance and dimensional stability.

Positioned in the upper-right quadrant, treatment T5 was close to the vectors ST, WA, and Dl, characterized by a profile strongly influenced by dimensional stability and specific acoustic attributes. These results suggest that the intermediate glass wool concentration (9.49%) favors a balance between acoustic performance and moisture absorption behavior, although increased ST may represent a limitation in terms of stability (Li et al. 2012; Sakamoto et al. 2024).

T1, T3, and T4 Treatments exhibited closer distribution to each other, indicating greater uniformity from the multivariate perspective. Treatments T3 and T4, with moderate glass wool levels (4.93% and 6.52%, respectively), behaved similarly to the commercial panel (T1), suggesting that controlled fiber incorporation does not compromise structural integrity nor cause significant variations in physical-chemical and mechanical properties (Benthien et al. 2019; Engehausen et al. 2024).

In the factorial plane, the vectors corresponding to gel time (GT) and water absorption (WA) showed similar orientation, suggesting a positive correlation between these variables, that is, higher GT values may be associated with increased water absorption. However, the GT vector had a lower magnitude, indicating a relatively smaller contribution to overall data variability compared with panel water absorption.

Similarly, the proximity of the vectors for pH, viscosity, and solid content reveals a trend of co-evolution of these variables depending on adhesive formulation (Gultom et al. 2013). Interestingly, the proximity of distinct properties such as flammability (Flam), modulus of elasticity (SB(MOE)) and acoustic impedance (SII) suggests potential correlations among them.

Conversely, vectors oriented in opposite directions indicate relevant inverse relationships. The damping coefficient (Dl) vector was oriented opposite to the acoustic conversion efficiency (ACE) and modulus of rupture (SB(MOR)) vectors, suggesting that higher levels of viscoelastic energy dissipation may compromise both acoustic performance and rupture strength (Krushynska et al. 2021; Kumar et al. 2025).

Likewise, the opposition between the modulus of elasticity (MOE) and thickness swelling index (ST) vectors highlights a possible antagonism between structural stiffness and dimensional stability, possibly due to limitations in adhesive formulation regarding its interaction with the particulate matrix (Gultom et al. 2013; Iswanto et al. 2023).

The results reinforce the hypothesis that incorporating glass wool into urea-formaldehyde adhesive simultaneously and interdependently influences multiple particleboard properties. When properly adjusted, parameters such as viscosity, pH, and gel time can contribute to panel performance optimization, promoting better adhesion between adhesive and particles and resulting in joint advances in mechanical, acoustic, and dimensional stability properties (Gultom et al. 2008; Gonçalves et al. 2017; Iswanto et al. 2023).

In this context, PCA has been shown to be a robust statistical tool, capable of identifying latent multivariate patterns and providing an integrated interpretation of treatment effects, effectively complementing traditional analyses.

Pearson correlation

The Pearson correlation was applied to quantify the intensity of linear relationships among the physical-mechanical, rheological, and acoustic properties of the modified particleboards (Fig. 3).

Fig. 3. Observed values for Pearson correlation. **: significant correlations at 1%; *: significant correlations at 5%; ns: non-significant correlations (p > 0,05); r: Pearson correlation coefficient; * pH: Hydrogen potential; SC: Solids content; Visc: Viscosity; GT: Gel time; ρ (r-X): Apparent density; CR: Compression ratio; WA: Water absorption; ST: Swelling in thickness; SB(MOE): Modulus of elastic modulus in static bending; SB(MOR): Modulus of rupture in static bending; IB: Internal bound; JH: Janka hardness; SP: Screw pulling (face); Flam: Flammability; SII: Sound impact insulation; Dl: Damping coefficient; ACE: Acoustic conversion efficiency.

For interpretation of the coefficients, the classification proposed by Navarro and Foxcroft (2025) and Weisburd et al. (2020) was adopted, considering strong correlations within the range 0.60 < r ≤ 0.90 and very strong within 0.90 > r < 1.

The resulting matrix revealed a high degree of interdependence among the analyzed variables, with than 49.26% of the correlations classified as strong or very strong, and about 29.41% as moderate. This indicates statistical consistency of the data and coherence in the response patterns of the panels.

Among the main results, a very strong correlation was observed between pH and solid content (r = 0.95**), as well as a strong correlation between viscosity and solid content (r = 0.80). These findings suggest that urea-formaldehyde adhesive formulations with higher alkalinity and solid content tend to exhibit higher viscosity. This behavior is consistent with the literature, which highlights the sensitivity of viscosity to solid concentration in colloidal systems with particulate additives (Akpabio 2012; Bacigalupe et al. 2020).

Conversely, viscosity showed a strong negative correlation with the modulus of elasticity (SB(MOE)) (r = –0.79), indicating that more viscous formulations may hinder adequate adhesive penetration into the lignocellulosic particles, reducing internal cohesion and structural stiffness of the panels, as also reported by Cesprini et al. (2022).

A very strong correlation was also observed between pH and Janka hardness (JH) (r = 0.92), indicating that more alkaline conditions favor more efficient resin curing and greater resistance to penetration. This effect is particularly relevant in formulations with higher glass wool content, in which alkalinity contributes to more complete polycondensation reactions and formation of a stronger adhesive matrix (Wang et al. 2016).

Regarding dimensional stability, a very strong negative correlation was found between flammability (Flam) and thickness swelling (ST) (r = –0.92**), suggesting that lower moisture absorption is associated with higher fire resistance. This behavior may be linked to greater porosity in the panels, which facilitates water penetration and contributes to swelling. Nevertheless, despite this less compact structure, the presence of silica derived from glass wool acts as a thermal barrier, delaying flame propagation and explaining the observed reduction in flammability (Costa et al. 2020) as well as enhancing fire resistance (Çavdar 2020; Najahi et al. 2023). This trend aligns with several studies emphasizing the role of material composition and structure in determining both fire resistance and dimensional stability (Maminski et al. 2011; Kweon et al. 2012; Yoo and Kim 2014; Jeon et al. 2017).

Acoustic conversion efficiency (ACE) showed a very strong negative correlation with the damping coefficient (Dl) (r = –0.94**) and a strong positive correlation with acoustic impedance (SII) (r = 0.71), indicating that stiffer structures with lower viscoelastic dissipation provide better sound transmission (He et al. 2018).

The modulus of rupture (SB(MOR)) correlated strongly with the acoustic conversion efficiency (ACE) (r = 0.95) and negatively with the damping coefficient (Dl) (r = -0,73ns), suggesting that both acoustic and mechanical performance are influenced by viscoelastic mechanisms, especially those related to internal cohesion, stiffness, and uniformity of the lignocellulosic matrix (Taghiyari et al. 2017; Bertolini et al. 2019).

Properties such as screw withdrawal (SP) and internal bond (IB) exhibited weak correlations with rheological variables, including pH, viscosity, and solid content, indicating that their variation may be more closely related to structural aspects such as particle orientation, localized density, and mixture uniformity (Arabi et al. 2012; Engehausen et al. 2024). However, IB showed meaningful correlations with SB(MOR) (r = 0.62ns) and ACE (r = 0.87), reinforcing its importance in the functional response of the panels.

Overall, these results demonstrate that the rheological properties of the adhesive play a crucial role in defining the structural and functional characteristics of the panels. Adjustments in formulation parameters—particularly pH, viscosity, and solid content—can directly affect mechanical stiffness, acoustic performance, and dimensional stability (Ramesh et al. 2022; Jalowy et al. 2025). Within this context, the Pearson correlation matrix complements principal component analysis (PCA) by enabling the identification of directed associations between specific variables, thereby supporting the development of consistent models for the integrated performance of particleboards.

In an integrated framework, the combination of Pearson correlation, PCA, and hierarchical clustering made it possible to identify critical properties and similarity patterns across treatments. This joint approach enhances interpretive robustness and contributes to the optimized formulation of panels with functional additives. Such an approach supports targeted variable selection, increasing efficiency in the development of sustainable materials with improved mechanical, physical, and functional performance.

CONCLUSIONS

The application of multivariate statistical techniques allowed the effects of glass wool addition to urea-formaldehyde adhesive in wood particleboards to be understood. The study demonstrated that incorporating glass wool residue into urea-formaldehyde (UF) resin significantly alters adhesive rheological properties (viscosity, pH, and solid content).
PCA revealed that 80.3% of the total variance was explained by the first two principal components, establishing strong correlations between adhesive rheological properties and the final performance of the panels. Hierarchical clustering analysis identified four distinct groups, and the Pearson correlation matrix revealed expressive associations between adhesive rheological properties and the physical-mechanical and acoustic performance of the panels, including very strong correlations between pH and Janka hardness (r = 0.92) and between ACE and the damping coefficient (r = –0.94).
The strong correlations detected for MOE, MOR, WA, TS, Flam, and ACE with multiple parameters reinforce their role as critical indicators for quality control, dimensional stability, mechanical resistance, and fire performance. Thus, these variables make the greatest contribution to treatment differentiation.
The adhesive with 3.34% glass wool content enhanced the mechanical, physical, and acoustic performance of the panels, while higher addition levels (>9%) improved fire resistance and reduced sound impedance without compromising compliance with normative requirements for other parameters.
The results indicate that modification of the adhesive with glass wool contributes to the optimization of the physical, mechanical, fire-retardant, and acoustic properties of particleboards. Therefore, this study validates glass wool as a functional additive of technological interest, offering scientific evidence to support the design of products with improved performance efficiency, safety, and environmental relevance.

REFERENCES CITED

ABNT NBR 14810-2 (2018). “Painéis de partículas de média densidade. Parte 2: Requisitos e métodos de ensaio,” Associação Brasileira de Normas Técnicas, Rio de Janeiro.

Achchaq, F., Djellab, K., and Beji, H. (2009). “Hydric, morphological and thermo-physical characterization of glass wools: From macroscopic to microscopic approach,” Construction and Building Materials 23(10), 3214-3219. https://doi.org.10.1016/j.conbuildmat.2009.06.018

Adediran, A., Lemougna, P. N., Yliniemi, J., Tanskanen, P., Kinnunen, P., Roning, J., and Illikainen, M. (2021). “Recycling glass wool as a fluxing agent in the production of clay- and waste-based ceramics,” Journal of Cleaner Production 289, article 125673. https://doi.org.10.1016/j.jclepro.2020.125673

Akpabio, U. D. (2012). “Effect of pH on the properties of urea formaldehyde adhesives,” International Journal of Modern Chemistry 2(1), 15-19.

Albert, C. M., and Liew, K. C. (2024). “Recent development and challenges in enhancing fire performance on wood and wood-based composites: A 10-year review from 2012 to 2021,” Journal of Bioresources and Bioproducts 9(1), 27-42. https://doi.org.10.1016/j.jobab.2023.10.004

Albuquerque, C. E. C., Iwakiri, S., Keinert Junior, S., and Trianoski, R. (2020). “Adesão e Adesivos,” in: Painéis de Madeira Reconstituída, FUPEF, Curitiba, PR, Brazil.

ANSI A208.1 (1999). “Particleboard,” American National Standard Institute, Gaithersburg, MD, USA.

ASTM D2526-14 (2018). “Standard test method for apparent viscosity of adhesives having shear-rate-dependent flow properties using rotational viscometry,” American Society for Testing and Materials, West Conshohocken, PA, USA.

Arabi, M., Faezipour, M., Haftkhani, A. R., and Maleki, S. (2012). “The effect of particle size on the prediction accuracy of screw withdrawal resistance (SWR) models,” Journal of the Indian Academy of Wood Science 9, 53-56. https://doi.org.10.1007/s13196-012-0063-6

Arias, A., González-Rodríguez, S., Barros, M. V., Salvador, R., Francisco, A. C., Piekarski, C. M., and Moreira, M. T. (2021). “Recent developments in bio-based adhesives from renewable natural resources,” Journal of Cleaner Production 314, article 127892. https://doi.org.10.1016/j.jclepro.2021.127892

Ayrilmis, N., and Nemli, G. (2017). “Effect of adhesive type on the quality properties of particleboard,” International Scientific Journals 11(7), 364-365.

Bacigalupe, A., Molinari, F., Eisenberg, P., and Escobar, M. M. (2020). “Adhesive properties of urea-formaldehyde resins blended with soy protein concentrate,” Advanced Composites and Hybrid Materials 3, 213-221. https://doi.org.10.1007/s42114-020-00151-7

Baharoǧlu, M., Nemli, G., Sarı, B., Birtürk, T., and Bardak, S. (2013). “Effects of anatomical and chemical properties of wood on the quality of particleboard,” Composites Part B: Engineering 52, 282-285. https://doi.org.10.1016/j.compositesb.2013.04.009

Baharuddin, M. N. M., Zain, N. M., Harun, W. S. H., Roslin, E. N., Ghazali, F. A., Md Som, S. N. (2023). “Development and performance of particleboard from various types of organic waste and adhesives: A review,” International Journal of Adhesion and Adhesives 124. https://doi.org.10.1016/j.ijadhadh.2023.103378

Baldin, T., Silveira, A. G., Vidrano, B. R. A., Cancian, L. C., Spatt, L. L., and Haselein, C. R. (2016). “Qualidade de painéis aglomerados produzidos com diferentes proporções de madeira e capim-annoni,” Revista Brasileira de Ciências Agrárias 11(3), 230-237. https://doi.org.10.5039/agraria.v11i3a5376

Balea, G., Lunguleasa, A., Zeleniuc, O., and Coşereanu, C. (2022). “Three adhesive recipes based on magnesium lignosulfonate, used to manufacture particleboards with low formaldehyde emissions and good mechanical properties,” Forests 13(5), article 737. https://doi.org.10.3390/f13050737

Bazzetto, J. T. L., Bortoletto Junior, G., and Brito, F. M. S. (2019). “Effect of particle size on bamboo particle board properties,” Floresta e Ambiente 26(2), article e20170125. https://doi.org.10.1590/2179-8087.012517

Benhamou, A. A., Boussetta, A., Kassab, Z., Nadifiyine, M., Sehaqui, H., El Achaby, M., and Moubarik, A. (2022). “Application of UF adhesives containing unmodified and phosphate-modified cellulose microfibers in the manufacturing of particleboard composites,” Industrial Crops and Products 176, article 114318. https://doi.org.10.1016/j.indcrop.2021.114318

Benthien, J. T., Lüdtke, J., and Ohlmeyer, M. (2019). “Effect of increasing core layer particle thickness on lightweight particleboard properties,” Eur. J. Wood and Wood Products 77(6), 1029-1043. https://doi.org.10.1007/s00107-019-01452-5

Bertolini, M. S., Morais, C. A. G., Christoforo, A. L., Bertoli, S. R., Santos, W. N., and Rocco Lahr, F. A. (2019). “Acoustic absorption and thermal insulation of wood panels: Influence of porosity,” BioResources 14(2), 3746-3757. https://doi.org.10.15376/biores.14.2.3746-3757

Calovi, M., and Rossi, S. (2023). “Impact of high concentrations of cellulose fibers on the morphology, durability and protective properties of wood paint,” Coatings 13(4), article 721. https://doi.org.10.3390/coatings13040721

Çavdar, A. D. (2020). “Effect of zeolite as filler in medium density fiberboards bonded with urea formaldehyde and melamine formaldehyde resins,” Journal of Building Engineering 27, article 101000. https://doi.org.10.1016/j.jobe.2019.101000

Cazella, P. H. S., Souza, M. V., Rodrigues, F. R., Silva, S. A. M., Bispo, R. A., Araujo, V. A., and Christoforo, A. L. (2024). “Polyethylene terephthalate (PET) as a recycled raw material for particleboards produced from pinus wood and biopolymer resin,” Cleaner Production Journal 447, article 141460. https://doi.org.10.1016/j.jclepro.2024.141460

Cesprini, E., Causin, V., De Iseppi, A., Zanetti, M., Marangon, M., Barbu, M. C., and Tondi, G. (2022). “Renewable tannin-based adhesive from Quebracho extract and furfural for particleboards,” Forests 13(11), article 1781. https://doi.org.10.3390/f13111781

Chen, X., Sun, C., Wang, Q., Tan, H., and Zhang, Y. (2022). “Preparation of glycidyl methacrylate grafted starch adhesive to apply in high-performance and environment-friendly plywood,” International Journal of Biological Macromolecules 194, 954-961. https://doi.org.10.1016/j.ijbiomac.2021.11.152

Cosereanu, C., and Cerbu, C. (2019). “Morphology, physical, and mechanical properties of particleboard made from rape straw and wood particles glued with urea-formalde-hyde resin,” BioResources 14(2), 2903-2918. https://doi.org.10.15376/biores.14.2.2903-2918

Costa, A. A., Mascarenhas, A. R. P., Santos, C. M. M., Faria, C. E. T., Duarte, P. J., and Cruz, T. M. (2020). “Technological characterization of engineered panels produced with Paricá wood,” Research, Society and Development 9(8), article e786986089. https://doi.org.10.33448/rsd-v9i8.6089

Costa, D., Serra, J., Quinteiro, P., and Dias, A. C. (2024). “Life cycle assessment of wood-based panels: A review,” Journal of Cleaner Production 444, article 140955. https://doi.org.10.1016/j.jclepro.2024.140955

Dashti, H., Salehpour, S., Taghiyari, H. R., Far, F. A., and Heshmati, S. (2012). “The effect of nanoclay on the mass transfer properties of plywood,” Digest Journal of Nanomaterials and Biostructures 7(3), 853-860.

Dhanapal, S., Singh, M., and Nagammanavar, U. (2024). “Influence of rice straw and wood fiber combination on physical and mechanical properties of rice straw pulverized composite board,” Maderas. Ciencia y Tecnología 26, article e3724. https://doi.org.10.22320/s0718221x/2024.37

Engehausen, N., Benthien, J. T., and Lüdtke, J. (2024). “Influence of particle size on the mechanical properties of single-layer particleboards,” Fibers 12(4), article 32. https://doi.org.10.3390/fib12040032

Evangelista, N., Tenório, J. A. S., and Oliveira, J. R. (2012). “Pozolanicidade dos resíduos industriais, lã de vidro e lã cerâmica,” Revista Escola de Minas 65(1), 79-85. https://doi.org.10.1590/S0370-44672012000100011

Fassarella, M. V. (2025). Efeitos da Adição de lã de Vidro à Resina Ureia-Formaldeído nas Propriedades de Painel Aglomerado Homogêneo, Master’s Thesis, Universidade Federal do Espírito Santo, Jerônimo Monteiro, ES, Brazil.

Fassarella, M. V., Chaves, I. L. S., Segundinho, P. G. A., Paes, J. B., Lelis, R. C. C., Oliveira, M. P., Silva, E. S. G., and Gonçalves, F. G. (2025). “The potential for glass wool waste as a filler in UF adhesive to promote particleboard strength,” Recycling 10(6), article 220. https://doi.org/10.3390/recycling10060220

Fehrmann, J., Belleville, B., Ozarska, B., Gutowski, W. S., and Wilson, D. (2023). “Influence of particle granulometry and panel composition on the physico-mechanical properties of ultra-low-density hemp hurd particleboard,” Polymer Composites 44(11), 7363-7383. https://doi.org.10.1002/pc.27631

Fehrmann, J., Belleville, B., and Ozarska, B. (2022). “Effects of particle dimension and constituent proportions on internal bond strength of ultra-low-density hemp hurd particleboard,” Forests 13(11), article 1967. https://doi.org.10.3390/f13111967

Fialho, L. F., Carneiro, A.C. O., Figueiró, C. G., Peres, L. C., Carneiro, A. P. S., and Surdi, P. G. (2022). “Application of univariate and multivariate statistical analyzes in clonal selection of Eucalyptus spp. for charcoal production,” Ciência Florestal 32(3), 1659-1683. https://doi.org.10.5902/1980509840443

França, M. C., Cunha, A. B., Trianoski, R., Schimalski, M. B., and Rios, P. D. (2016). “Produção de painéis aglomerados homogêneos a partir de fibras oversize residuais de uma indústria de MDF,” Scientia Forestalis 44(111), 665-674. https://doi.org.10.18671/scifor.v44n111.12

García, F. D., Aigner, S. N., Cedres, J. P., Luna, A., Escobar, M. M., Mansilla, M. A., and Bacigalupe, A. (2024). “Novel adhesive based on black soldier fly larvae flour for particleboard production,” Construction and Building Materials 411, article 134758. https://doi.org.10.1016/j.conbuildmat.2023.134758

Ghani, A., Ashaari, Z., Bawon, P., and Lee, S. H. (2018). “Reducing formaldehyde emission of urea formaldehyde-bonded particleboard by addition of amines as formaldehyde scavenger,” Building and Environment 142, 188-194. https://doi.org.10.1016/j.buildenv.2018.06.020

Gillela, S., Yadav, S. M., Sihag, K., Kelkar, B. U., and Dangtungee, R. (2024). “Effects of addition of nanoclay in phenol-formaldehyde resins on the properties of Lantana camara fibre composites,” Journal of Tropical Forest Science 36(1), 40-50. https://doi.org.10.26525/jtfs2024.36.1.40

Gonçalves, F. G., Lelis, R. C. C., Oliveira, J. T. S., Garcia, R. A., and Brito, E. O. (2017). “Chapas aglomeradas confeccionadas com ureia-formaldeído sob adição de tanino em pó,” Ciência Florestal, 27(4), 1349-1363. https://doi.org.10.5902/1980509830216

Gonçalves, F. G., Lelis, R. C. C., and Silva Oliveira, J. T. (2008). “Influência da composição da resina tanino-uréia-formaldeído nas propriedades físicas e mecânicas de chapas aglomeradas,” Revista Arvore 32(4), 715-722. https://doi.org.10.1590/S0100-67622008000400013

Guan, M., Fu, R., Yong, C., Li, Y., and Xu, X. (2022). “Properties of binderless bamboo particleboards derived from biologically fermented bamboo green residues,” Waste Management 151, 195-204. https://doi.org.10.1016/j.wasman.2022.07.040

Gultom, L. A., Dirhamsyah and Setyawati, D. (2013). “Mechanical and physical properties of particleboard rice straw,” Hutan Jurnal Lestari 1(3), 458-465.

Haas, A., Lourenço, W. M., Santos, J. C. P., and Santos J. L. P. (2022). “Isolamento ao ruído de impacto de laje pré-fabricada nervurada com vigotas protendidas e lajotas cerâmicas,” Ambiente Construído 22(1), 105-123. https://doi.org.10.1590/s1678-86212022000100581

Harada, T., Uesugi, S., and Masuda, H. (2006). ‘Fire resistance of thick wood-based boards,” J. Wood Science 52(6), 544-551. https://doi.org.10.1007/s10086-006-0805-4

He, M., Perrot, C., Guilleminot, J., Leroy, P., and Jacqus, G. (2018). “Multiscale prediction of acoustic properties for glass wools: Computational study and experimental validation,” Journal of the Acoustical Society of America 143(6), article 3283. https://doi.org.10.1121/1.5040479

Hong, M. K., and Park, B. D. (2017). “Effect of urea-formaldehyde resin adhesive viscosity on plywood adhesion,” Journal of the Korean Wood Science and Technology 45(2), 223-231. https://doi.org.10.5658/WOOD.2017.45.2.223

Hosseini, S. B., Asadollahzadeh, M., Kazemi Najfai, S., and Taherzadeh, M. J. (2020). “Partial replacement of urea-formaldehyde adhesive with fungal biomass and soy flour in plywood fabrication,” Journal of Adhesion Science and Technology 34(13), 1371-1384. https://doi.org.10.1080/01694243.2019.1707948

IARC. (1988). “Man-made mineral fibres,” in: Evaluation of Carcinogenic Risks to Humans: Man-Made Mineral Fibres and Radon,” v. 43. Cancer by the Secretariat of the World Health Organization, Lyon, France. https://www.ncbi.nlm.nih.gov/books/NBK316363/

Indústria Brasileira de Árvores (IBÁ) (2024). “Relatório Anual,” Brasília. https://iba.org/wp-content/uploads/2025/05/relatorio2024.pdf

Iglesias, M. C., McMichael, P. S., Asafu-Adjaye, O., Via, B. K., and Peresin, M. S. (2021). “Interfacial interactions between urea formaldehyde and cellulose nanofibrils (CNFs) of varying chemical composition and their impact on particle board (PB) manufacture,” Cellulose 28, 7969-7979. https://doi.org.10.1007/s10570-021-04007-1

ISO 11925-2 (2020). “Reaction to fire tests – Ignitability of products subjected to direct impingement of flame – Part 2: Single-flame source test,” International Organization for Standardization, Geneva, Switzerland.

Iswanto, A. H., Sutiawan, J., Darwis, A., Lubis, M. A. R., Pędzik, M., Rogoziński, T., and Fatriasari, W. (2023). “Influence of isocyanate content and hot-pressing temperatures on the physical-mechanical properties of particleboard bonded with a hybrid urea-formaldehyde/isocyanate adhesive,” Forests 14(2), article 320. https://doi.org.10.3390/f14020320

Iwakiri, S., Matos, J. L. M., Ferreira, E. S., Prata, J. G., and Trianoski, R. (2012). “Produção de painéis compensados estruturais com diferentes composições de lâminas de Eucalyptus saligna e Pinus caribaea,” Revista Árvore 36(3), 569-576. https://doi.org.10.1590/S0100-67622012000300019

Machado, J. P., Silva, T. C., Borgert, C. H., Rosso Neto, L., Gesuino, D. B., Oliveira, J. R., Frizon, T. E. A., Grillo, F. F., and Junca, E. (2023). “Mechanical behavior of cementitious composites reinforced with the fiber of sugarcane bagasse and glass wool waste,” International Journal of Environmental Science and Technology 20, 3765-3774. https://doi.org.10.1007/s13762-022-04224-6

Jalowy, L., Nemec, D., and Ilhan, O. (2025). “Comparison of dispersing processes of bio-based and synthetic materials: A review,” ChemEngineering 9(2), 36. https://doi.org.10.3390/chemengineering9020036

Jeon, C. K., Lee, J. -S., Chung, H., Kim, J. -H., and Park, J. -P. (2017). “A study on insulation characteristics of glass wool and mineral wool coated with a polysiloxane agent,” Advances in Materials Science and Engineering 2017, article 3938965. https://doi.org.10.1155/2017/3938965

Kawalerczyk, J., Antov, P., Dziurka, D., Mirski, R., and Lee, S. H. (2023). “The effect of pressing parameters and hardener content on the properties of plywood bonded with propylamine-UF adhesive,” Wood Material Science and Engineering 19(3), 710-717. https://doi.org.10.1080/17480272.2023.2286633

Kawalerczyk, J., Barczewski, M., Woźniak, M., Kuliński, M., Smogór, H., Dukarska, D., and Mirski, R. (2025). “Propolis extract as a bio-based modifier of urea-formaldehyde adhesive in particleboard production,” European Journal of Wood and Wood Products 83, article 84. https://doi.org.10.1007/s00107-025-02242-y

Kelleci, O., Koksal, S. E., Aydemir, D., and Sancar, S. (2022). “Eco-friendly particleboards with low formaldehyde emission and enhanced mechanical properties produced with foamed urea-formaldehyde resins,” Journal of Cleaner Production 379, article 134785. https://doi.org.10.1016/j.jclepro.2022.134785

Khanjanzadeh, H., Behrooz, R., Bahramifar, N., Pinkl, S., and Gindl-Altmutter, W. (2019). “Application of surface chemical functionalized cellulose nanocrystals to improve the performance of UF adhesives used in wood based composites – MDF type,” Carbohydrate Polymers 206, 11-20. https://doi.org.10.1016/j.carbpol.2018.10.115

Khorramabadi, L. A., Behrooz, R., and Najafi, S. K. (2023). “Reduction of formaldehyde emission from medium density fiberboard using nanoclay modified with 3-aminopropyltriethoxysilane and L-Lysine as additives to urea-formaldehyde adhesive,” International Journal of Adhesion and Adhesives 125, article 103426. https://doi.org.10.1016/j.ijadhadh.2023.103426

Khorramabadi, L. A., Behrooz, R., and Najafi, S. K. (2024). “Effect of blending L-lysine-modified montmorillonite into urea-formaldehyde resin on formaldehyde emission and physicomechanical properties of medium density fiberboard,” European J. Wood and Wood Products 82(5), 1393-1405. https://doi.org.10.1007/s00107-024-02090-2

Krushynska, A. O., Gliozzi, A. S., Fina, A., Krushinsky, D., Battegazzore, D., Badillo-Ávila, M. A., Acuautla, M., Stassi, S., Noè, C., Pugno, N. M., and Bosia, F. (2021). “Dissipative dynamics of polymer phononic materials,” Advanced Functional Materials 31(30), article 2103424. https://doi.org.10.1002/adfm.202103424

Kumar, A., Vijaya Kumar, K. R., Suresh, G., Vezhavendhan, R., Chandramohan, P., Saranya, R., and Rathinasabapathi, G. (2025). “A comprehensive study on the influence of polyurethane blending in melamine-urea-formaldehyde-based plywood composites: an interpenetrating polymer network (IPN) approach,” Journal of the Indian Academy of Wood Science. https://doi.org.10.1007/s13196-025-00383-4

Kweon, O. -S., Yoo, Y. -H., Kim, H. -Y., and Min, S. -H. (2012). “An experimental study on fire safety performance of glass wool sandwich panel,” 26(5), 21-27. https://doi.org.10.7731/KIFSE.2012.26.5.021

Lee, S. J., and Thole, V. (2018). “Investigation of modified water glass as adhesive for wood and particleboard: mechanical, thermal and flame retardant properties,” European Journal of Wood and Wood Products 76(5), 1427-1434. https://doi.org.10.1007/s00107-018-1324-x

Lei, H., Du, G., Pizzi, A., and Celzard, A. (2008). “Influence of nanoclay on urea-formaldehyde resins for wood adhesives and its model,” Journal of Applied Polymer Science 116(5), 2658-2667. https://doi.org.10.1002/app.28359

Lemougna, P. N., Adediran, A., Yliniemi, J., Ismailov, A., Levanen, E., Tanskanen, P., Kinnunen, P., Roning, J., and Illikainen, M. (2020). “Thermal stability of one-part metakaolin geopolymer composites containing high volume of spodumene tailings and glass wool,” Cement and Concrete Composites 114, article 103792. https://doi.org.10.1016/j.cemconcomp.2020.103792

Li, C., Chen, Z., Zhou, J., Li, B., Wu, W., Chen, Z., Xu, T., and Qiu, J. (2012). “Material parameter determination in glass wool mat for sound absorption,” Mechanical Engineering, Materials and Energy 1448-149, pp. 1271-1275. https://doi.org.10.4028/www.scientific.net/AMM.148-149.1271

Li, Z., Beltran, I. A. F., Chen, Y., Šavija, B., and Ye, g. (2021). “Early-age properties of alkali-activated slag and glass wool paste,” Construction and Building Materials 291, article 123326. https://doi.org.10.1016/j.conbuildmat.2021.123326

Lima, A. J. M., Iwakiri, S., Satyanarayana, K. G., and Lomelí-Ramírez, M. G. (2020). “Studies on the durability of wood-cement particleboards produced with residues of Pinus spp., silica fume, and rice husk ash,” BioResources 15(2), 3064-3086. https://doi.org.10.15376/biores.15.2.3064-3086

Liu, J., Li, Y., Mo, H., Xie, E., Fang, J., and Gan, W. (2022). “Current utilization of waste biomass as filler for wood adhesives: A review,” Journal of Industrial and Engineering Chemistry 115, 48-61. https://doi.org.10.1016/j.jiec.2022.08.016

Lu, G. -Z., Zheng, X. -S., Ding, X. J., and Zhao, W. -X. (2016). “The performance study of modified glass wool board external thermal insulation system and the application in the energy-efficient construction in existing building,” in: 2^nd Annual International Conference on Advanced Material Engineering, Atlantis Press, pp. 825-839.

Mahieu, A., Vivet, A., Poilane, C., and Leblanc, N. (2021). “Performance of particleboards based on annual plant byproducts bound with bio-adhesives,” International Journal of Adhesion and Adhesives 107, article 102847. https://doi.org.10.1016/j.ijadhadh.2021.102847

Maminski, M. Ł., Król, M., Jaskółowski, W., and Borysiuk, P. (2011). “Wood-mineral wool hybrid particleboards,” European Journal of Wood and Wood Products 69(2), 337-339. https://doi.org.10.1007/s00107-010-0 470-6

Meftah, R., Berger, S., Jacqus, G., Laluet, J. Y., and Cnudde, V. (2019). “Multiscale characterization of glass wools using X-ray micro-CT,” Materials Characterization 156, article 109852. https://doi.org.10.1016/j.matchar.2019.109852

Mensah, P., Chrysafi, I., Karidi, K., Mitani, A., and Bikiaris, D. N. (2025). “Eco-friendly particleboards produced with banana tree (Musa paradisiaca) pseudostem fibers bonded with cassava starch and urea-formaldehyde adhesives,” Journal of Renewable Materials 13(7), 1475-1489. https://doi.org.10.32604/jrm.2025.02025-0047

Moslemi, A., Koohi, M. Z., Behzad, T., and Pizzi, A. (2020). “Addition of cellulose nanofibers extracted from rice straw to urea formaldehyde resin; effect on the adhesive characteristics and medium density fiberboard properties,” International Journal of Adhesion and Adhesives 99, article 102582. https://doi.org/10.1016/j.ijadhadh.2020.102582

Moutousidis, D., Karidi, K., Athanassiadou, E., Stylianou, E., Giannakis, N., and Koutinas, A. (2023). “Reinforcement of urea formaldehyde resins with pectins derived from orange peel residues for the production of wood-based panels,” Sustainable Chemistry for the Environment 4, article 100037. https://doi.org.10.1016/j.scenv.2023.100037

Najahi, A., Aguado, R. J., Tarrés, Q., Boufi, S., and Delgado-Aguilar, M. (2023). “Harvesting value from agricultural waste: Dimensionally stable fiberboards and particleboards with enhanced mechanical performance and fire retardancy through the use of lignocellulosic nanofibers,” Industrial Crops and Products 204, article 117336. https://doi.org.10.1016/j.indcrop.2023.117336

Navarro, D. J., and Foxcroft, D. R. (2025). “Correlation and linear regression,” in: Learning Statistics with Jamovi: A Tutorial for Beginners in Statistical Analysis, Open Book Publishers, Cambridge, UK, pp. 225-294. https://doi.org.10.11647/OBP.0333.12

Nemli, G., and Demirel, S. (2007). “Relationship between the density profile and the technological properties of the particleboard composite,” Journal of Composite Materials 41(15), 1793-1802. https://doi.org.10.1177/0021998307069892

Papadopoulou, E., Chrysafi, I., Karidi, K., Mitani, A., and Bikiaris, D.N. (2024). “Particleboards with recycled material from hemp-based panels,” Materials 17(1), article 139. https://doi.org.10.3390/ma17010139

Park, S. -H., Lee, M., Seo, P. -N., Kang, E., and Kang, C. -W. (2020). “Acoustical properties of wood fiberboards prepared with different densities and resin contents,” BioResources 15(3), 5291-5304. https://doi.org.10.15376/biores.15.3.5291-5304

Pereira, H. M. L., Lelis, R. C. C., Gomes, F. J. B., Sousa, N. D., Lopez, Y. M., Gonçalves, F. G., Chaves, I. L. S., and Brito, A. S. (2024). “Effect of tannin and calcium lignosulfonate-based adhesives on particleboards production,” International Journal of Adhesion and Adhesives 132, article 103722. https://doi.org/10.1016/j.ijadhadh.2024.103722

Pintiaux, T., Viet, D., Vandenbossche, V., Rigal, L., and Rouilly, A. (2015). “Binderless materials obtained by thermo-compressive processing of lignocellulosic fibers: A comprehensive review,” BioResources 10(1), 1915-1963. https://doi.org.10.15376/biores.10.1.1915-1963

Pizzi, A. (2003). “Urea-formaldehyde adhesives,” in: Handbook of Adhesive Technology, CRC Press, New York, USA.

Ramesh, M., Rajeshkumar, L., Sasikala, G., Balaji, D., Saravanakumar, A., Bhuvaneswari, V., and Bhoopathi, R. (2022). “A critical review on wood-based polymer composites: Processing, properties and prospects,” Polymers 14(3), article 589. https://doi.org.10.3390/polym14030589

Roumeli, E., Papadopoulou, E., Pavlidou, E., Vourlias, G., Bikiaris, D., Paraskevopoulos, K.M., and Chrissafis, K. (2012). “Synthesis, characterization and thermal analysis of urea-formaldehyde/nanoSiO₂ resins,” Thermochimica Acta 527, 33-39. https://doi.org.10.1016/j.tca.2011.10.007

Sahoo, S. C., Sil, A., Solanki, R., and Dutta, A. (2024). “Exploring the impact of resin booster (Polymate-888) on urea formaldehyde resin to enhance plywood,” International Journal of Wood Science and Technology 2(1), 1-9.

Sakamoto, S., Sato, K., and Muroi, G. (2024). “Improvement of sound-absorbing wool material by laminating permeable nonwoven fabric sheet and nonpermeable mem-brane,” Technol. 12(10), article 195. https://doi.org.10.3390/technologies12100195

Salari, A., Tabarsa, T., Khazaeian, A., and Saraeian, A. (2013). “Improving some of applied properties of oriented strand board (OSB) made from underutilized low quality paulownia (Paulownia fortunie) wood employing,” Industrial Crops and Products 42, 1-9. https://doi.org.10.1016/j.indcrop.2012.05.010

Santos, M. M., Diez, M. A., Suárez, M., and Centeno, T. A. (2021). “Innovative particleboard material from the organic fraction of municipal solid waste,” Journal of Building Engineering 44, article 103375. https://doi.org.10.1016/j.jobe.2021.103375

Sato, N., Takahashi, A., Kiyohara, S., Terayama, K., Tamura, R., and Oba, F. (2024). “Target material property-dependent cluster analysis of inorganic compounds,” Advanced Intelligent Systems 6(12), article 2400253. https://doi.org.10.1002/aisy.202400253

Silva, L. C. L., Lima, F. O., Araujo, V. A., Santos, H. F., Lahr, F. A. R., Christoforo, A. L., Favarim, H. R., and Campos, C. I. (2024). “Influence of pressing temperatures on physical-mechanical properties of wood particleboards made with urea-formaldehyde adhesive containing Al₂O₃ and CuO nanoparticles,” Polymers 16(12), article 1652. https://doi.org.10.3390/polym16121652

Sozim, P. C. L., Napoli, L. M., Ferro, F. S., Mustefaga, E. C., and Hillig, E. (2019). “Propriedades de painéis aglomerados produzidos com madeiras de Ligustrum lucidum e Pinus taeda,” Pesquisa Florestal Brasileira 39(1), article 201801696. https://doi.org.10.4336/2019.pfb.39e201801696

Srivabut, C., Ratanawilai, T., and Hiziroglu, S. (2018). “Effect of nanoclay, talcum, and calcium carbonate as filler on properties of composites manufactured from recycled polypropylene and rubberwood fiber,” Construction and Building Materials 162, 450-458. https://doi.org.10.1016/j.conbuildmat.2017.12.048

Taghiyari, H. R., Taheri, A., and Omrani, P. (2017). “Correlation between acoustic and physical – mechanical properties of insulating composite boards made from sunflower stalk and wood chips,” European Journal of Wood and Wood Products 75(3), 409-418. https://doi.org.10.1007/s00107-016-1101-7

Taquetti, V. B., Silva, V. V., Chaves, I. L. S., Oliveira, R. G. E., Maffioletti, F. D., Ferreira, G., Mendonça, J. P. C., Silva, E. S. G., and Gonçalves, F. G. (2023). “Performance of eucalypt particleboard with the addition of farm waste,” Heliyon 9(12), article 22760. https://doi.org.10.1016/j.heliyon.2023.e22760

Tsukamoto, M., Murakami, T., Yoshimura, Y., Kuroki, Y., Okamoto, T., and Takata, M. (2014). “Evaluation of the tensile strength of polypropylene-based composites containing glass wool,” Materials Letters 132, 267-269. https://doi.org.10.1016/j.matlet.2014.06.039

Ülker, O., and Burdurlu, E. (2015). “Effects of some mineral wools and adhesives on burning characteristics of particleboard,” BioResources 10, 3775-3789. https://doi.org.10.15376/biores.10.2.3775-3789

Veigel, S., Müller, U., Keckes, J., Obersriebnig, M., and Gindl-Altmutter, W. (2011). “Cellulose nanofibrils as filler for adhesives: Effect on specific fracture energy of solid wood-adhesive bonds,” Cellulose 18(5), 1227-1237. https://doi.org.10.1007/s10570-011-9576-1

Wang, H., Wang, H., Liao, J., Zhou, X., and Du, G. (2023). “Technological properties of a branched polyethyleneimine derivative as a cross-linker for low molar ratio urea-formaldehyde resins,” Polymer Testing 118, article 107914. https://doi.org.10.1016/j.polymertesting.2022.107914

Wang, X. -M., Casilla, R., Zhang, Y., Cooper, P., Huang, Z., and Wang, X. (2016). “Effect of extreme pH on bond durability of selected structural wood adhesive,” Wood and Fiber Science 48(4), 245-259.

Weisburd, D., Britt, C., Wilson, D.B., and Wooditch, A. (2020). “Measuring association for scaled data: Pearson’s correlation coefficient,” in: Basic Stati. Criminology and Crim. Just., Springer: Cham, 479-530. https://doi.org.10.1007/978-3-030-47967-1_14

Wibowo, E. S., Park, B. D., and Causin, V. (2022). “Recent advances in urea-formalde-hyde resins: Converting crystalline thermosetting polymers back to amorphous ones,” Polymer Reviews 62(4), 722-756. https://doi.org.10.1080/15583724.2021.2014520

Wong, M. C., Hendrikse, S. I. S., Sherrell, P. C., and Ellis, A. V. (2020). “Grapevine waste in sustainable hybrid particleboard production,” Waste Management 118, 501-509. https://doi.org.10.1016/j.wasman.2020.09.007

Yildirim, M., and Candan, Z. (2021). “Performance properties of particleboard panels modified with nanocellulose/boric acid,” BioResources 16(1), 1875-1890. https://doi.org.10.15376/biores.16.1.1875-1890

Yliniemi, J., Walkley, B., Provis, J., Kinnunen, P., and Illikainen, M. (2020). “Nanostructural evolution of alkali-activated mineral wools,” Cement and Concrete Composites 106, article 103472. https://doi.org.10.1016/j.cemconcomp.2019.103472

Yoo, Y. -H., and Kim, Y. -N. (2014). “A fire risk assessment of sandwich panel by comparison fire test,” Korean Society of Hazard Mitigation 14(3), 21-26. https://doi.org.10.9798/KOSHAM.2014.14.3.21

Article submitted: October 8, 2025; Peer review completed: February 13, 2026; Revised version received and accepted: March 11, 2026; Published: March 24, 2026.

DOI: 10.15376/biores.21.2.4110-4134