New Bio-Based Binding Systems for Plywood Panels

Abstract

This study investigated the use of non-formaldehyde binders in the production of plywood panels, focusing on mixtures containing 70% poly 4,4’-methylene diphenyl isocyanate (pMDI) and 30% soy flour (SF), along with blends of soy flour and agricultural residues (olive by-products—with and without extraction of their bioactive ingredients—and defatted hemp seeds). The basic properties of these biomaterials, such as moisture content, pH, and buffering capacity, were determined with laboratory analysis. Adhesive mixtures were characterized using attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTIR) and thermogravimetric analysis (TGA). The adhesive’s bonding ability was evaluated by manufacturing plywood panels on a laboratory scale, simulating industrial practices. The glue lines were visually inspected with a stereomicroscope. Micro-ATR-FTIR study of the cross-sections of plywood panels showed the full consumption of isocyanate groups indicating effective curing of the adhesive. Mixtures containing olive residues, particularly olive skin and stones, showed improved thermal stability in the TGA study. The mechanical properties of the plywood panels were assessed with three-point bending tests, while their shear strength and wood failure performance were tested according to the European standards used in the relevant industry (EN 314.1:2004 and EN 314.2:1993). In terms of flexural properties, the adhesive with non-extracted (NE) residual olive skin (ROS) showed the highest flexural strength of around 17 MPa and a flexural modulus of 650 MPa. The formulations containing extracted materials from hemp seeds (HSs) and residual olive skin (ROS) showed the best overall performance with wood failure values of 85% and 75% after the most severe cyclic test (EN314.1:2004-Pretreatment 5.1.3). Overall, the results showed that binders prepared with residual olive skin and defatted hemp seeds have promising performance and can be used in the manufacture of plywood panels.

1. Introduction

Although polymeric methylene diphenyl diisocyanate (pMDI) resins have been successfully used as non-formaldehyde binders in the manufacture of wood panel products such as particle board (PB) and medium density fiberboard (MDF), their use in the manufacture of plywood panels (PW) has been limited for a number of reasons. These include the difficulty of cleaning with water the equipment used to apply pMDI resins to veneers, the reaction of the NCO group in PMDI with the hydroxyl groups in wood, resulting in premature curing prior to hot pressing, the tendency of pMDI resin to stick to the caul plates during pressing making the plywood difficult to remove, and the incompatibility of pMDI with water-based adhesives commonly used in plywood production due to its high reactivity with water. A solution to overcome these drawbacks is to mix pMDI adhesives with starch to control their reactivity by forming urethane bonds between starch and pMDI, making them easier to apply to plywood [1]. The starch used industrially to make plywood is usually derived from wheat. However, in the context of a sustainable economy, it is desirable to use other lower value biomasses or wastes for this application.

In this study, residues of soybeans, hemp seeds, and olive fruit and leaves were investigated as pMDI resin fillers in the manufacture of plywood panels. All these materials are found in large quantities as wastes, and their use in this application gives them added value.

Soybean cultivation in Europe has grown steadily in recent years, reaching a record high of 11.5 million tonnes in 2023 [2], and is expected to increase by a further 5–10% in 2024 [3]. Soybeans are rich in protein and also contain fiber and oligosaccharides, making them a valuable food source for both humans and animals. However, soy has been studied for many applications beyond food, such as health products [4], dietary supplements, cosmetics [5,6,7], and chemical formulations. In the case of adhesives for plywood panels, He et al., 2012 [8], investigated a mixture of soy protein hydrolysate, different resins (phenol-formaldehyde, melamine-formaldehyde, phenol-melamine-formaldehyde, methylenediphenyl diisocyanate, and epoxy) and an amine polymer-epichlorohydrin adduct. It was found that the plywood panels were not successful in all cases, with the degree of hydrolysis of the soy playing a significant role. Lei et al., 2014 [9], improved the water resistance of a soy-based adhesive by mixing it with epoxy resin or melamine-formaldehyde resin or a mixture thereof, making it suitable for plywood production. Other researchers have investigated the performance of non-formaldehyde mixtures of soybean flour with cardanol-based epoxy resin or citric acid [10,11]. The results of these studies have shown the potential of such binders for plywood panel production. Alawode et al., 2022 [12], evaluated the potential of using soy flour (among other materials) as a partial replacement of pMDI resin for the production of both hardwood and softwood plywood panels. The results showed that replacing 30% of the pMDI resin with soy flour significantly improved its adhesion properties. Following this, it has been considered that the partial substitution of soy flour in pMDI should not exceed that of 30% because of the poor waterproof properties of soy flour itself [13,14].

Industrial hemp is another crop that is attracting increasing research interest due to its rapid growth and diverse applications in commercial products such as textiles, paper, medicine, food, animal feed, paints, biofuel, biodegradable plastics, and construction materials. Industrial hemp seeds, which are the fruits of Cannabis sativa, represent another valuable but underutilized biomass resource after the extraction of nutritional ingredients. Specifically, hemp seeds are covered by a thin, two-layered pericarp and contain an endosperm and two cotyledons. They contain approximately 25–30% oil, 25–30% protein, 30–40% fiber, and 6–7% moisture [15,16,17]. After extraction, the remaining biomass is considered waste. Although its unprecedented production growth (expected to grow at a CAGR of 22.44% over the next 10 years) will inevitably lead to large amounts of defatted residual material, there has been limited research into the use of defatted hemp seed residues. Kawalerczyk et al., 2020 [18], investigated the potential of incorporating hemp meal as a filler in melamine urea formaldehyde (MUF) adhesives for plywood production. They found that the incorporation of 20–25 pbw (parts by weight) hemp flour facilitated the production of plywood panels with mechanical properties comparable to the reference plywood bonded with MUF adhesive mixed with rye flour. More recently, in 2024, Barretto et al. [19] used extracted hemp proteins, both modified and unmodified, in combination with glutaraldehyde to synthesize wood adhesives. The adhesiveness of these resins was tested on cherry wood samples of specific dimensions. The results showed that hemp protein has significant potential as a renewable and environmentally friendly alternative to plant protein-based adhesives.

Regarding the olive tree, it is another crop whose cultivation produces waste that causes environmental problems and whose management entails significant economic costs. Olive growing and olive oil production processes generate significant solid residues, both from the processing of the fruit and from the waste olive leaves. An average mature olive tree can yield between 10 and 70 kg of olive oil, which is equivalent to roughly 2500 to 17,500 pieces of olive fruit, and a corresponding amount of residues [20]. It is therefore clear that a huge amount of olive leaf and fruit waste such as pomace, exhausted olive cake, pomace skins, and stones is produced every year. Of course, these residues are a source of valuable bioactive compounds such as simple phenols and phenolic acids, secoiridoids, flavonoids, lignans, and triterpenic acids, which are used in several applications including pharmaceuticals, dietary supplements, biofunctional foods, and cosmetics. Their extensive use is now the focus of commercial interest in the global market, but this also creates other waste in large quantities that usually remain unused.

Several scientists have investigated the use of olive residues—olive stones in particular—in various applications, such as fillers in thermoplastic resin composites [21,22,23], the recovery of various chemicals from them [24,25], as ingredients in skin care formulations, as a natural mulch for plants, and as a fuel [26,27]. However, there are only a few publications on wood adhesives, and none of them used olive residues after removal of the valuable bioactive compounds. Specifically, Benhamou et al., 2021 [28], compared cactus waste seeds and olive stones as fillers in phenol-formaldehyde (PF) to produce wood composites. The study showed that incorporation of up to 20 per cent cactus waste seeds and olive stones in PF resins resulted in higher viscosity, gel time, and solids content than the control PF, while the corresponding plywood panels had lower formaldehyde emissions than the control. Elsahli et al., 2016 [29], used olive stone/unsaturated polyester resin compositions at various ratios to produce particleboards that were evaluated for their mechanical and physical properties according to European standards. It was found that such panels have acceptable performance.

In this context, we investigated the possibility of replacing wheat flour, an edible resource, with biomass waste in the manufacture of pMDI-based adhesives for plywood panels, with the aim of producing a non-competitive and environmentally friendly product. Specifically, we examined the replacement of wheat flour with a combination of commercial defatted soybean flour and various biomass residues. The biomass residues investigated included olive leaves (OLs), exhausted olive pomace (EOP), residual olive skin (ROS), and olive stones (OSs), both with and without the removal of their bioactive compounds. Additionally, oil-free hemp seeds (HSs) were also studied.

Previous research has investigated the partial replacement of pMDI with up to 30% soybean meal [12,30]. Building on this, the present study investigates the partial replacement of soy flour (50%) with olive and hemp residues. In particular, the pMDI-based adhesives were prepared by mixing 70% of pMDI with 15% of defatted soybean flour and 15% of other biomass residues (by weight). Additionally, control formulations were prepared with only soybean flour and wheat/soybean flour mixtures. The adhesive mixtures were characterized by infrared spectroscopy to examine the chemical structure of the adhesives and by thermogravimetric analysis (TGA) to study their thermal stability. To test the properties of the novel pMDI-based adhesives, they were applied in the preparation of plywood panels according to conventional industrial practices. Finally, these plywood panels were further characterized by infrared spectroscopy, mechanical tests according to European standards, and TGA to verify, respectively, the curing of the adhesive, the mechanical performance of the plywood panels, and their thermal stability. Promising results were obtained, suggesting that the studied biomass residues can indeed be used to replace wheat flour in pMDI-based adhesives.

2. Materials and Methods

2.1. Materials

pMDI is a bulk material. Defatted soy flour (SF) with the trade name “Prolia” was provided by Cargill company (Wayzata, MN, USA). Hemp seeds (HSs) of the Felina 32 technical variety were provided by the Centre for Renewable Energy Sources and Saving (CRES) in Attiki, Greece. Regarding the olive residues, 3 categories of olive pomace waste (OPW) were collected and studied: (a) exhausted olive pomace (EOP); (b) residual olive skin (ROS); (c) olive stone (OS). The raw materials were obtained from a collaborating pomace oil factory located in Messinia region of Greece. For reference, commercial wheat flour (WF) was also included in the study.

2.2. Preparation of Materials

The exhausted olive pomace (EOP) by-product originated after chemically extracting the residual oil from two- and three-phase olive pomace waste. Subsequently, the raw material was dried in a laboratory oven at 140 °C, until most of its moisture was removed. Then, the moisture content of the EOP was less than 5%, which was sufficient for their pulverization with a standard laboratory blender. The raw materials of residual olive skin (ROS) and olive stone (OS) resulted from the exhausted olive pomace after processing it through sieves with a diameter of 1 mm. All dried materials were kept in sealed bags suitable for human use, at room temperature, prior to their dispatch to the laboratories for analysis and further separation of bioactive ingredients (Hydroxytyrosol, Tyrosol, Maslinic, and Oleanolic acids) by extraction using methanol as a solvent.

In parallel, waste olive leaves were collected, dried, and powdered, and then oleuropein was separated by ethanol/water (80/20) extraction. The remaining biomass was dried afterwards. Regarding hemp seeds, the oil was extracted by cold pressing using a screw press.

All the above materials before and after the extraction of oil and bioactive ingredients were used in adhesive mixtures with pMDI for plywood panel production (

Table 1. Materials used in the adhesive mixtures with pMDI for plywood panels.

Material	Characteristic
Soy flour Prolia FLR 200/90	Non-extracted (NE)
Exhausted Olive Pomace (EOP)	Non-extracted (NE)
Exhausted Olive Pomace (EOP)	Extracted (E)
Residual olive skin (ROS)	Non-extracted (NE)
Residual olive skin (ROS)	Extracted (E)
Olive stone (OS)	Non-extracted (NE)
Olive stone (OS)	Extracted (E)
Olive Leaves (OLs)	Non-extracted (NE)
Olive Leaves (OLs)	Extracted (E)
Hemp seeds (HSs)	Non-oil-extracted (NE)
Hemp seeds (HSs)	Extracted (Oil-free) (E)

), after being grounded and sieved to obtain particle size below 200 μm.

2.3. Preparation of Adhesive Mixtures

The adhesive mixtures were prepared by mechanical stirring of the materials at ambient temperature. The pMDI was 70% of the mixture and the flours 30%.

Details of the formulation of the mixtures are given in

Table 2. Composition of adhesive mixtures.

a/a	Materials	Weigh (g)
1	pMDI	70
	Soy flour	30
2	pMDI	70
	Soy flour	15
	Wheat flour	15
3	pMDI	70
	Soy flour	15
	EOP-non-extracted (NE)	15
4	pMDI	70
	Soy flour	15
	EOP-extracted (E)	15
5	pMDI	70
	Soy flour	15
	ROS-non-extracted (NE)	15
6	pMDI	70
	Soy flour	15
	ROS-extracted (E)	15
7	pMDI	70
	Soy flour	15
	OS-non-extracted (NE)	15
8	pMDI	70
	Soy flour	15
	OS-extracted (E)	15
9	pMDI	70
	Soy flour	15
	OL-non-extracted (NE)	15
10	pMDI	70
	Soy flour	15
	OL-extracted (E)	15
11	pMDI	70
	Soy flour	15
	HS-extracted (E) (oil-free HSs)	15
12	pMDI	Unsuitable for use *
	Soy flour
	HS-non-extracted (NE)

* Hemp seeds without oil removal could not be used, as their pulverization resulted in a paste-like mass that could not be sieved.

below.

The ingredients were agitated for 2–3 min and then applied on veneers.

2.4. Preparation of Plywood Panels

Plywood panels are typically constructed by gluing wood veneers together at right angles and then subjecting them to high temperatures and pressures. The adhesive used is usually a mixture of polymer resin (commonly PF resin), water and a filler, often wheat flour. This process produces strong boards that are widely used in construction and various other applications. When pMDI is used as the adhesive, the mixture is anhydrous to prevent any reaction between pMDI and water. The amount of adhesive used ranges from 0.1 to 0.25 kg/m². Hot pressing ensures good and sufficient contact between the boards and facilitates the curing of the adhesive for a strong bond. Special hydraulic presses with heated plates are used for this process, with temperatures ranging from 120 to 150 °C. The curing time depends on the thickness of the construction and the moisture, thermal conductivity and specific heat of the wood. In practice, the time required ranges from 2 to 15 min, depending on the thickness of the veneers, the type of adhesive used, and the degree of hardening required.

In this study, three-layer plywood panels were produced on a laboratory scale using typical production parameters [31,32] (

Table 3. Typical parameters of plywood panel production.

Wood of Veneers:	Softwood (Scots Pine)
Dimensions of veneers:	50 × 50 cm
Moisture of veneers:	3.5–6.5%
Assembling time:	10–20 min
Waiting time before hot pressing:	>1 h
Hot press temperature:	130–150 °C
Pressing time:	2–7 min
Specific pressure:	10–25 kg/cm2

). After pressing, the panels were stacked and left to allow the adhesive to fully cure. Finally, the boards were trimmed to their final dimensions, squared, and tested to determine their properties.

2.5. Characterization Methods

The biomasses of this study were characterized for their moisture content, pH, and buffer capacity.

The adhesive mixtures with pMDI were studied with Fourier transform infrared (FTIR) spectroscopy and thermogravimetric analysis (TGA).

All plywood panels were tested for their mechanical properties according to the European standards in force as well as with 3-point bend tests. Their thermal properties were determined via thermogravimetric analysis (TGA), while they were subjected to surface analysis with stereomicroscope. The various measurements were conducted as described below.

2.5.1. Moisture Content (MC) of Biomasses of Olive and Hemp Biomasses

The moisture content (MC) of the various biomasses was determined with the gravimetric method, also known as direct method, by comparing the weight of biomasses before and after drying them in an oven [2]. According to this method, firstly, a quantity of these samples is weighed using a precision balance. An oven dryer is then used to remove moisture from the chips. When the chips have reached a constant weight without further loss, the dry weights are measured again using the precision balance. The moisture content (MC) is then determined using the following equation:

$$\mathrm{M}\mathrm{C}\left(\%\right)=\frac{\left(\mathrm{w}-\mathrm{d}\right)}{w}\ast 100$$

(1)

where w = wet (initial) weight, and d = dry (final) weight.

2.5.2. pH and Buffer Capacity of Biomasses of Olive and Hemp Biomasses

Understanding the pH and buffering capacity of biomasses is essential for their effective use in gluing processes [33]. For pH measurements, 2 g (oven-dried basis) of ground biomass from each type were mixed with 30 mL of cold, freshly boiled distilled water (pH 7.3). The samples were stirred periodically at room temperature for 72 h, after which the pH was recorded. The measurement was repeated twice to obtain the average pH value [34]. The pH was measured by a CRISON pH meter (Crison Instruments, s.a., Alella, Spain) calibrated at three points (pH 4, pH 7, and pH 10) for high accuracy. The samples were then titrated to pH 3 with nominal 0.05 N HCl (for alkalinity determination). The alkalinity was expressed as ml of HCl consumed after correction for the volume of titrant required to adjust the distilled water to the target pH [34].

2.5.3. Fourier Transform Infrared (FTIR) Spectroscopy Study of Adhesive Mixtures and Plywood Panels

All the adhesive mixtures as mentioned in Table 2 were studied after being cured for 2 h in 120 °C. Spectra of pMDI alone and mixtures of pMDI/soy flour (70/30) and pMDI/soy flour/wheat flour (70/15/15) were studied for comparison reasons.

FTIR spectra of and the adhesive mixtures were recorded with a Cary 670 spectroscope (Agilent Technologies, Palo Alto, CA, USA) using a diamond attenuated total reflectance (ATR) accessory, (GladiATR, Pike Technologies, Madison, WI, USA). The spectra were collected in the mid IR area (4000–400 cm⁻¹), with 32 scans and a resolution of 4 cm⁻¹. The specimens were not manipulated for the measurements; they were placed on the ATR diamond and pressured by the instrument’s click-stop pressure clamp.

Furthermore, micro-ATR-FTIR measurements were performed on cross-sections of the manufactured plywood panels. The measurements were employed with a Cary 620 FTIR microscope (Agilent Technologies, Palo Alto, CA, USA)—coupled with the aforementioned Cary 670 spectroscope—equipped with a germanium micro-ATR accessory attached on the 15× objective lens of the microscope, using a 64 × 64 focal plane array (FPA) detector (Agilent Technologies, Palo Alto, CA, USA). Each FPA image consists of 4096 data points/individual spectra, each one of them collected from a 1.1 × 1.1 μm² independent observed area. The measurements were performed in the range of 3950–700 cm⁻¹, with a resolution of 4 cm⁻¹ and 64 co-added scans. For the preparation of the cross-sections, specimens from the panels were cut mechanically using a cutting machine to a thickness of 0.5 cm, and then polished with SiC papers up to 4000 mesh.

2.5.4. Visual Assessment of the Quality of Glue Lines in Plywood Panels

The glue lines of the panels were examined with a SteREO Discovery V.20, stereomicroscope (Zeiss, Oberkochen, Germany), equipped with a GRYPHAX Altair camera (Jenoptik, Jena, Germany).

2.5.5. Plywood Panels Performance Tests

The plywood panels were tested according to the European standards used by the relevant industry and refer to their mechanical strength and bonding quality. These standards are as follows:

• EN 314-1:2004 [35]. Plywood—Bonding quality—Part 1: Test methods.
• EN 314-2:1993 [36]. Plywood—Bonding quality—Part 2: Requirements.

The standard EN 314-1 outlines the methods for testing plywood panels, while EN 314-2 classifies the panels and establishes threshold values for their shear strength (measured with a Zwick/Roell machine) and wood failure performance. According to the EN314-1 standard, the panels are subjected to various pre-treatments before their testing for shear strength and wood failure evaluation. They are as follows:

• Pre-treatment 5.1.1 (EN314-1): Immersion in water of (20 ± 3) °C for 24 h.
• Pre-treatment 5.1.3 (EN314-1): Immersion in boiling water for 4 h, followed by drying in ventilated oven for 16 h to 20 h at (60 ± 3) °C and subsequent immersion in boiling water for 4 h followed by cooling in water at (20 ± 3) °C for at least 1 h.

In order for plywood panels to be suitable for various applications, they must meet certain combined shear strength and wood failure limits as shown in

Table 4. EN314-2:1993 standard bonding performance requirements.

Mean Shear Strength fv, N/mm2	Mean Apparent Cohesive Wood Failure w, %
0.2 < fv < 0.4	≥80
0.4 < fv < 0.6	≥60
0.6 < fv < 1.0	≥40
1.0 < fv	No requirement

below.

Although the above standards apply to industrial applications, they provide a reliable method of evaluation and laboratory boards as well.

In this study, the plywood panels were evaluated as boards for use in outdoor applications, and therefore, their properties were measured after pretreatments 5.1.1 and 5.1.3, which are the severest for plywood products. Shear force was applied to the central glue line of each test piece. After shearing, the test pieces were analyzed for the percentage of wood failure. The results were compared to those of panels prepared with the reference adhesive mixture. All evaluations were conducted according to the European standard EN 314-2: 1993, which correlates mean values of shear strength with mean values of wood failure (Table 4).

2.5.6. Three-Point Bending Tests of Plywood Panels

Three-point bending tests were performed using a Shimadzu EZ Flexural Tester Model EZ-LX (Kyoto, Japan), with a 2 kN load cell. Samples (26.5 mm wide and ca 9 mm thick) were tested flatwise on the support span according to ASTM D790-17 [37]. The support span-to-depth ratio was 16:1 (tolerance ±1), and a constant crosshead speed of 1.0 mm/min was applied.

2.5.7. Thermogravimetric Analysis (TGA) of Adhesive Mixtures and Plywood Panels

Thermogravimetric analysis was performed to the adhesive mixtures and the manufactured panels with a SETSYS TG-DTA 16/18 (Setaram, Lyon, France). In every case, the analysis was performed on samples of (3.0 ± 0.2) mg placed in alumina crucibles in dynamic conditions, from room temperature up to 800 °C for the adhesive mixtures and up to 500 °C for the plywood panels, with a heating rate of 20 °C/min, using nitrogen flow (50 mL/min).

The pMDI-based soy/flour mixture was measured (a) immediately after mixing its components and (b) after 2 h at room temperature to compare the progress of reactions between the components of the mixture. All the adhesive mixtures—as mentioned in Table 2—were studied after being cured for 2 h in 120 °C. For the study of the plywood panels, material was scraped from the surface of the middle veneer after their detachment, without further manipulation.

3. Results and Discussion

The aim of this study was to investigate the feasibility of replacing wheat flour with biomass residues in pMDI-based adhesives. To this end, a series of pMDI adhesives were prepared consisting of 70% pMDI and 30% biomass residues, including 15% defatted soybean flour and 15% other biomass residues (olive and hemp). The biomass materials used in this study were characterized for moisture content, pH, and buffer capacity. The adhesives were analyzed using FTIR and TGA and then applied to the production of plywood panels. These panels were produced by gluing wood veneers together at right angles and subjecting them to high temperature and pressure. The resulting plywood panels were characterized to evaluate the properties of the adhesive blends. The test results are presented below.

3.1. Properties of Studied Biomaterials

The properties of the biomasses are shown in

Table 5. Properties of biomasses olive and hemp biomasses.

Material	Solids, %	pH at 21 °C, [ ]	Buffer Capacity, mL
EOP-non-extracted (NE)	92.14	5.95	63
EOP-extracted (E)	91.76	7.10	18
ROS-non-extracted (NE)	92.28	5.50	44
ROS-extracted (E)	90.95	7.01	12
OS-non-extracted (NE)	88.9	7.27	12
OS-extracted (E)	89.87	7.60	11
OL-non-extracted (NE)	94.58	5.59	46
OL-extracted (E)	89.61	5.95	18
HS-extracted (E)	96.00	5.85	10

.

The above results show that non-extracted (NE) materials are more acidic than extracted biomass. This also affects their buffer capacity as non-extracted materials present a higher buffer capacity than the extracted ones.

3.2. ATR-FTIR Analysis of the Adhesive Mixtures

Figure 1a shows the ATR-FTIR spectrum collected from the neat and uncured pMDI together with the spectra of cured pMDI adhesive mixtures with soy, and with soy and wheat. Neat and uncured pMDI presents all the characteristic bands as supported in numerous references [38,39,40]. The difficulty to observe FTIR bands in the 2300–2200 cm⁻¹ region is noted, due to the diamond ATR accessory that was employed for the collection of the spectra. For this reason, the isocyanate band at 2278 cm⁻¹—according to the bibliography—is biased and located at 2242 cm⁻¹.

Both the cured adhesive mixtures—with plain soy flour and soy/wheat flours—present the same spectral characteristics (Figure 1a). The isocyanate band decreases and shifts to 2270 cm⁻¹, but is not fully consumed; this is indicative that the pMDI polymerization with the additives is not full but isocyanates remain to react with the wood of the veneers. A new band of C=O at 1663 cm⁻¹ is observed, indicative of NH-CO-NH linkages [38]. The bands at 1610 and 1577 cm⁻¹ of neat and uncured pMDI disappear and a new band at 1595 cm⁻¹ rises, indicating the formation of secondary urea linkages. Finally, the shoulder at 1530 cm⁻¹ is attributed to CHN groups of secondary urethane formation.

Figure 1b shows the ATR-FTIR spectra of the cured adhesive mixtures, as they are described in Table 2. In Figure 1a, the glue mixtures with soy flour and with both soy and wheat flours are again presented for comparative reasons. In every case, all the spectra of the cured glue mixtures present similar characteristics in comparison with the cured mixtures of soy flour, and of soy and wheat flour, as already discussed; the band of the isocyanate group at 2270 cm⁻¹ is decreased in comparison with the neat and uncured pMDI, but the group is not fully consumed, as isocyanates remain to react with the wood of the veneers.

3.3. TGA Analysis of the Adhesive Mixtures

Figure 2 shows the TGA curves of mass loss and its derivative for the adhesive mixture of pMDI with 30% soy flour. The adhesive mixture was studied immediately after combining its ingredients (uncured), two hours after its preparation (prepolymer) and after its curing (cured). The uncured adhesive mixture presents a 7% mass loss up to 160 °C, which is attributed to humidity. Several local maxima of the mass loss rate can be observed, which correspond to its unreacted ingredients, such as the degradation of pMDI at 245 °C, the degradation of MDI monomers at 430 °C, and, finally, the degradation of the remaining isomers of MDI at 500 °C [41]. Moreover, and due to the soy flour, two additional local mass loss rate maxima are reported, at ~350 °C, due to starch and protein decomposition [30,42]. Regarding the cured adhesive mixture, its decomposition presents a mass loss rate maximum at 344 °C, in accordance with the literature [43]. Finally, the prepolymer presents a mass loss of 0.5% due to humidity up to 160 °C, which indicates that the moisture content of soy flour is immediately consumed as pMDI starts to cure. This is supported from the major step of degradation that the three analyzed samples present; their corresponding maxima of mass loss rate are located at 250, 322, and 343 °C for the uncured, the prepolymer, and the cured mixture, respectively.

Figure 3 shows the TGA results of the various studied adhesive mixtures of pMDI with 15% soy flour, and 15% flour of the agricultural wastes, according to Table 2, while the mixture of pMDI with 30% soy flour—as aforementioned—is also presented for comparative reasons. All the specimens were studied after being cured. The adhesive mixtures present a major degradation step, with a local mass loss rate maximum at 330–360 °C, which is in total agreement with cured pMDI mixtures according to the literature [43]. As shown by the dTG curves (Figure 3b), the glue mixtures with residual olive skin (ROS), olive stone (OS), and olive leaves (OLs)—all extracted—present a better thermal stability of all the other mixtures, while the glue mixtures with wheat flour (WF), extracted (EOP-E) and non-extracted exhausted olive pomace (EOP-NE), residual olive skin non-extracted (ROS-NE), and extracted hemp seeds (HS-E) have a lower thermal stability.

3.4. Plywood Test Results

All of the adhesive formulations in Table 2 were used in the production of three-layer plywood panels using the process described in Section 2.4. The results of their evaluations are given below.

3.4.1. Plywood Visual Observations

The quality of the glue lines was also assessed by visual observation (Figure 4).

Figure 4 above shows the cross-sections of the panels and allows a qualitative assessment of the glue lines. This study has shown the successful and uniform coating of the wood surfaces by the adhesive mixture, which is a critical parameter for their good bonding.

3.4.2. Micro-ATR-FTIR Analysis of Plywood Panels

Figure 5 presents the FPA micro-FTIR analysis of the cross-section of the plywood panel manufactured with the heartwood non-extracted adhesive mixture. The analysis was performed in micro-ATR mode, using a germanium crystal which does not affect the 2400–2000 cm⁻¹ area as a diamond one. The FPA chemical image of the area of interest is indicated in Figure 5a and presented in Figure 5b, showing the intensity of the band at 1710 cm⁻¹ in a color scale from blue (absence of band) to red (most intense). The band at ~1710 cm⁻¹ is attributed to urethane linkages (R-NH-COOR) that are formed, due to the reaction of the isocyanate groups with hydroxyls [38]. The glue mixture is distributed in the bonding line, within a 20 μm thickness. The micro-ATR spectrum of the glue mixture (Figure 5c) shows that the isocyanate group is fully consumed, as no band is noted at ~2240 cm⁻¹, except from the bands at 2400–2300 cm⁻¹ due to atmospheric CO₂.

3.4.3. Mechanical Properties of Plywood Panels

Plywood Panels Evaluation According to the European Standards

The shear strength and wood failure performance of panels was evaluated according to the European standards EN314.1 and EN 314.2 as described in Section 2.5.4. The results are shown in the following

Table 6. Properties of plywood panels made with pMDI/SF/various OPW and HS.

Formulation:	1 (Control 1)	2 (Control 2)	3	4	5	6	7	8	9	10	11
Resin:		pMDI/SF
Biomaterial:	-	WF	EOP	EOP	ROS	ROS	OS	OS	OL	OL	HS
Type	-		NE	E	NE	E	NE	E	NE	E	E
EN314.1
Pre-treatment 5.1.1		Immersion in water of 20 °C for 24 h
Shear strength, N/mm2	1.65	1.80	1.66	2.03	1.90	1.75	1.78	1.76	1.72	1.75	1.70
Wood failure, %	75	70	40	55	70	85	45	0	40	25	85
Pre-treatment 5.1.3		4 h in boiling water-16 h drying at 60 °C-4 h in boiling water-1 h in cool water
Shear strength, N/mm2	1.56	1.32	1.54	1.60	1.65	1.64	1.69	1.83	1.67	1.36	1.56
Wood failure, %	75	40	20	35	55	75	40	50	25	20	85

.

The above results show that all adhesive formulations have a shear strength greater than 1 N/mm², and therefore, according to the specifications of the EN314-2 standard for bonding performance requirements (Table 4), there is no requirement for a specific wood failure performance. In practice, however, the industry requires a wood failure of more than 50%. In this light, the mixtures with ROS and especially the extracted material (ROS-E) together with hemp seeds (HS-E) have the best overall performance after both pre-treatments (5.1.1. and 5.1.3.)—even from the reference formulations (formulations 1 and 2 of Table 6).

Plywood Flexural Test Results

The flexural properties of the panels were studied by subjecting them to three-point bending tests. This measurement shows the resistance of the panels to bending as well as the maximum strength that the panels can withstand. Figure 6 shows the flexural stress–strain curves of the examined panels, and the corresponding flexural data are presented in

Table 7. Flexural data of the prepared plywood panels.

Sample	Flexural Stress at Break (MPa)	Flexural Strain (%)	Flexural Modulus (MPa)	Flexural Strength (MPa) *
pMDI/SF	10.20	3.00	604.6	10.60
pMDI/SF/WF	12.60	3.16	580.2	12.97
pMDI/SF/EOP-NE	10.08	2.31	592.1	10.17
pMDI/SF/EOP-E	14.10	4.33	557.0	14.20
pMDI/SF/ROS-NE	16.52	3.81	655.2	17.25
pMDI/SF/E	12.10	2.70	647.3	12.41
pMDI/SF/OS-NE	7.45	1.54	563.9	7.60
pMDI/SF/OS-E	10.10	2.49	583.1	10.55
pMDI/SF/OL-NE	8.95	2.83	494.2	9.05
pMDI/SF/OL-E	7.57	2.95	467.8	9.01
pMDI/SF/HS-E	9.20	3.11	536.8	9.63

Flexural strength (MPa) *: It is the maximum flexural stress sustained by the samples during the three-point bending tests.

.

With some exceptions, most plywood panels with olive and hemp residue binding systems exhibited similar mechanical behavior compared to the reference panel, indicating that the novel biobased adhesives under study could potentially replace the conventional ones in terms of mechanical behavior. The panels with pMDI/SF/ROS-NE resin exhibited the best mechanical performance among all other panels, meaning it exhibited the highest endurance to bending deformation. Specifically, it could withstand maximum flexural stress (flexural strength) and flexural modulus close to 17 MPa and 650 MPa, respectively. Chen et al., 2024 [44], investigated starch-based gels as green alternatives adhesives for plywood applications providing flexural data comparable to the present bio-based binders.

3.4.4. TGA of the Plywood Panels

The TGA performed on material removed from the middle veneer of each plywood panel manufactured with the adhesive mixtures under study is presented in Figure 7. In every case, the thermal degradation of the adhesive mixtures follows that of the wood, as it was impossible to extract just the adhesive mixture that was cured between the veneers; following this, the mass loss and dTG curves of the plain wood veneer are shown for comparative reasons [45]. A slight mass loss due to humidity that does not exceed 3.5% is reported up to 160 °C. All the samples present a local maximum in the mass loss rate at about 375–390 °C due to wood degradation predomination and, in particular, the cellulose decomposition. Plain wood is considered as the most thermally stable one from all the samples under study, as its thermal decomposition maximum is located at 390 °C. The samples containing adhesive mixtures with olive tree residues, and especially that with residual olive skin (ROS) and olive stone (OS), present a somewhat better thermal stability in comparison with the other glue mixtures, with local degradation maxima above 380 °C. It should be mentioned that although all the adhesive mixtures—as collected from the wooden panels after their curing—present similar thermal characteristics, the one composed of extracted hemp seeds (HS-E) differs, as it presents two more local degradation maxima, at 310 and 439 °C. The first one is also observed in other adhesive mixtures, mainly the one with olive leaves (OL-NE), and can be related with the depolymerization of hemicellulose, cellulose, lignin, and pectins, which are associated with the agriculture residues under study [46]. The latter could be associated with lignin pyrolysis or the decomposition of char residues [47].

Summarizing all the above results, it can be seen that the removal of bio-components from biomasses increases their pH, resulting in less acidic materials and a reduction in their buffer capacity (Table 5). The FTIR analysis of the adhesive mixtures with the agricultural residues before and after their curing shows that they exhibit the exact characteristics with those of the glue mixture with soy flour. Furthermore, it is noted that in the cured adhesive mixtures, the peak corresponding to the isocyanate group is not totally consumed, and isocyanate groups still remain to react during the adhesion stage of the manufacturing of the plywood panels. This is supported by the FPA micro-FTIR analysis of the plywood panels, where the isocyanate groups were totally consumed. Visual observation of the glue lines showed good adhesion in all cases, and mechanical tests revealed that all the panels made from all studied biomaterials exhibit excellent mechanical properties and high wood failure. Notably, the panels with ROS and HSs demonstrated mechanical properties and wood failure performance that are comparable to, or even superior to the reference compositions (Table 6 and Table 7). The TGA of the adhesive mixtures showed that the ones containing ROS, OS, and OL had better thermal stability compared to all the other blends. In the case of the material collected from the bonding lines of the plywood panels, the TGA indicated that their thermal behavior closely follows that of plain wood.

4. Conclusions

This study demonstrated that agricultural biomass residues, such as soybean meal, and olive and hemp cultivation wastes can serve as effective substitutes for the traditionally used wheat flour in pMDI-based formulations for plywood panels. Of the various residues tested, the bio-ingredient-free olive fruit skin residues and oil-free (defatted) hemp seeds showed the best overall combined performance in terms of shear strength and wood failure, which are the most important properties for the relevant industrial sector. Similarly, the adhesive system with non-extracted olive skin (ROS-NE) showed the highest resistance in the three-point flexure tests, with a flexural strength and modulus of around 17 MPa and 650 MPa, respectively, while all ROS-based formulations were among the most thermally stable formulations. The use of such waste materials, which have no other use, in this application was investigated for the first time in this work, and the results of the study pave the way for their use in this value-added application.