Samples of the main trunks of four tropical timbers were obtained from various forests located in Ogwogo Nike in Enugu East of Enugu State, which is situated in the South Eastern part of Nigeria with the assistance of a well-trained government forestry official during the dry season. The selected timbers were: Nauclea diderichii (ubulu ani), Erythrophleum suavolens (inyi), Brachystegia eurycoma (achi), Prosopis africana (ugba). Recycled PET bottles and recycled PE (from sachet water packaging) were collected from homes and the environment in 3-3 areas of Onitsha, Nigeria. The collected timbers were pulverized using the sawing machine. They were milled and ground and passed through a sieve of 105 microns (mesh size) to remove impurities and then air-dried for 48 hrs. Clean consumer drinking water bottles and sachet bags collected were shredded by a shredding machine and a pair of scissors. The shredded plastics were air dried for 48 hours. The pulverized samples of the timbers and plastics are shown in Figure 1.

The composites were produced using a flat-press molding facility [12] . It comprises a digital temperature-controlled heater band of 500-watt capacity into which a cylindrical bronze mold (with lid) 80 × 50 mm outside dimension, and tapered internal dimension of 50 × 30 mm was fitted. A hand-operated hydraulic press coupled to a force load cell was applied to the blend and a digital force display was used to monitor the applied force/pressure during compounding. 20 g of each sample (constituting sawdust, rPE and rPET combined in fixed ratios to one another) was weighed. The selection (of 20 g) was based on the volume to weight ratio of the mold cavity. The respective weights of each constituent (sawdust, rPE and rPET) in each sample were therefore calculated based on 20 g total weight. Table 1 shows the experimental design.

After weighing, the constituents were mixed in a high-speed blender at 6000 Revolutions per minute (RPM) for 5 minutes. Subsequently, the properly mixed material was poured into a bronze mold, whose internal surface was coated with a thin film of graphite, and then fastened to the heater band. The temperature, force and time were set at 125^°C, 3000 N and 30 minutes respectively. Compounding was initiated by turning the start button on the heater unit. The heating rate was set at 25^°C per minute. After about 5 minutes of start-up, a drop in the force was observed indicative of rPE melting. When the force dropped significantly to as low as 500 N, after 10 minutes, a final force of up to 1500 N was applied and left unattended to thereafter to prevent the spilling of the rPE from the mold. The start button was then turned off after the alarm from the timer signaled the expiration of the set 30 minutes and the mold was allowed to cool slowly for 60 minutes, after which the composite was freely removed, facilitated by the tapered internal cavity and the graphite mold wash. Samples of the composites produced are shown in Figure 2.

The result of the densities of the wood-rPE/rPET composites produced is presented in Table 2 and shown in Figure 5 and Figure 6, respectively. The results showed that increase in the quantity of wood flour content affected the density of flat pressed WPCs in line with some recent studies [12] . From the results, the composites of Prosopis africana, W₄S₂ and W₄S₅ with SD-rPE/rPET mix ratios of 20:20:60 and 20:40:40 respectively, when compared to the other composites, had the two highest densities of 1.058 g/cm³ and 1.017 g/cm³ (respectively), while, W₃S₂ (0.801 g/cm³) and W₃S₃ (0.803 g/cm³) (composites of Brachystegia eurycoma) had the lowest densities. It was observed that the density of the SD-rPE/rPET composites decreased with the increase of SD percentage in the thermoplastic matrices from Sample Two (S₂) in W₁S₂ (0.970 g/cm³) and W₄S₂ (1.0581 g/cm³)

to Sample Three (S₃) in W₁S₃ (0.948 g/cm³) and W₄S₃ (0.952 g/cm³) respectively. This trend was also observed in the composites of W₂ and W₃ when going from Sample One (S₁) of W₂S₁ (0.936 g/cm³) and W₃S₁ (0.811 g/cm³) to Sample Two (S₂) of W₂S₂ (0.860 g/cm³) and W₃S₂ (0.801 g/cm³).

The decrease observed in the density of composites with an increase in the percentage of sawdust was in line with the literature which posited that the lower density of sawdust could be the cause of this reduction [4] . This means that the densities of the sawdust were actually enhanced by the plastic matrix which in turn is imparted with some level of biodegradability in line with the need to promote a sustainable environment [21] .

In the case of Wood 2, it could be observed from Figure 5 that all the densities of W₂S_1-3 (except in W₂S₂) were above the density of C₁ but the densities of W₂S_4-6 were well below the density of C₂. This showed that as the percentage of wood dust increased to 20%, the density also increased up to the optimum point (0.867 g/cm³). However, in Wood Two, 20% rPE (C₁) impacted more on the density of the composites than 40% rPE (C₂) in line with some work done elsewhere [12] . In the case of Wood Three, it was observed from Figure 5, that the densities of W₃S_1-3 decreased from 0.811 g/cm³ in W₃S₁ to 0.801 g/cm³ in W₂S₂ and finally to 0.802 g/cm³ in W₃S₃. All these values were below that of C₁. For W₃S_4-6, the density decreased from 0.867 g/cm³ to 0.838 g/cm³ in W₃S₄ to W₃S₅, but increased to 0.924 g/cm³ in W₃S₆. Apart from W₃S₆, the densities of W₃ samples were below C₁ and C₂. The relatively higher density observed for S_4-6 showed that formulation with 40% rPE impacted more on the densities of the composites than at 20% rPE. With respect to W₄S_1-3, it could be observed that the densities of the composites increased from 0.896 g/cm³ in W₄S₁ to 1.058 g/cm³ in W₄S₂ and decreased to 0.952 g/cm³ in W₄S₃. Similarly, in W₄S_4-6, the densities of the composites increased from 0.870 g/cm³ in W₄S₄ to 1.020 g/cm³ in W₄S₅ and decreased to 0.876 g/cm³ in W₄S₆. In both formulations, the density increased to the optimum point at S₂ and decreased as more wood dust was added. In Wood Four, the two formulations impacted positively on the density of the composites. Generally, as wood dust increased and rPET content decreased, the density of composites decreased except in W₁S₂. This may be due to poor binding arising from high rPET content in the composite according to some studies [22] . Furthermore, the decrease in the polymer content in the matrix slightly reduced the density values of the composite samples as a result of the viscoelastic behavior of wood fiber since the gaps in the lumens of lignocellulosic materials are crushed/jammed under high pressure and temperature [12] . For S_4-6, as wood dust increased, the density of composites also increased to an optimum point. This could be attributed to enhanced binding between wood dust and rPE, at low rPET content [23] .

From our discussion so far, it could be observed that the density of the SD-PET/PE composites did not show any consistent pattern of increase or decrease with the increased SD percentage in the thermos-plastic matrices This implies that the increase in SD content alone is not the only factor responsible for impacting on the densities of the composites. Considerations of the chemical compositions of the parent woods from which the SDs were derived and the anisotropic nature of trees could be contributory factors as has been noted in the literature [24] . However, the key factor which is important for the production industry is that all the SD-rPET-rPE composites formulated in this study met the required standards for high density particleboard (>800 kg/m³) by ANSI standards [25] .

Hardness is the ability of a material to maintain its physical features even in the face of applied force. The Shore D hardness values for the composites are presented Table 2 and shown in Figure 7 and Figure 8.

The results indicated that there was no particular order of increased or reduced hardness. The maximum hardness 56 Shore D was observed in W₄S₂, while minimum material hardness of 30 Shore D was observed in W₁S₃ as shown in Figure 8. Increase in material hardness is usually as a result of higher filler to matrix ratio [26] , while decrease in material hardness could be as a result of poor adhesion at the interfaces between SD particles and polymer matrixes [27] . In this investigation, the range of values showed a decrease in hardness when compared

to the values attributed to PET (85-95) and PE (70-80) in the shored D scale as expected because of the incorporation of sawdust from wood [28] .

The flexural strength of the composites of all four wood samples and their six different specimens with varying mixture ratio of rPET-rPE-SD are presented in Table 2 and in Figure 9 and Figure 10. W₁S₅ (14.995 N/mm²) had the highest flexural strength across all the formulations. Very low flexural strength was observed in majority of other wood samples with W₁S₃ having the lowest value of 0.636 N/mm².

The reason for low flexural strength of various composites could be as a result of poor interfacial interaction between the polymeric matrixes and wood particles, not allowing efficient stress transfer between the three phases of the material (Zou, Fan and Chen, 2016) [29] . All composite of rPET-rPE-SD except the 40:40:20 mixed ratio could not produce values close to their respective controls (C₁ had 4.643 and C₂ had 11.192). The results of the investigation showed that the composites of 40% rPE binder gave improved flexural strength than that with 20% rPE. This might be attributed to good dispersion of SD in the rPE/rPET blend (Durmaz, 2022) [12] . The inconsistencies with regard to trends, of the effect on the increase of wood flour with respect to its effect on the bending strength of WPCs have been reported [30] . The variation in flexural strength of the composites was as the result of anisotropic nature of wood such that

after a limit, load cannot be proportional to deformation [31] . Generally, as wood dust increased for S_1-3, with relatively higher PET content, flexural strength decreased as reported in some studies (Chaharmahali et al. 2008 [8] , Ayrilmis and Jarusombuti, 2011 [32] ; Durmaz, 2002) [13] . The decrease in bending strength of up to 58% was prominent after 50% of wood flour content, limiting the usability of WPC where high mechanical properties are vital (Durmaz, 2022) 13. Composite of W₂ (W₂S₁) had the best flexural strength at 20/70 rPE/rPET blend (10% wood dust). But at 40/40 rPE/rPET combination, Wood₁ (W₁S₅) gave the best flexural strength (14.995 N/mm²). It is worthwhile to note also that W₁S₅ that produced the highest of flexural strength (14.995 N/mm²), also produced the highest value in terms of percentage elongation of 26.65%. This is an indication that with the mixture of 20% SD of wood sample one (W₁) and 40% each of rPET and rPE, the composite will attain a very good flexural strength which is a major mechanical property of composite for resistance to bending. This property could be further improved by the addition of fillers like glass and carbon fibers [33] .

The ultimate tensile capabilities of all six different specimens for all four wood samples under study are indicated in Table 2 and in Figure 11 and Figure 12

respectively. It was observed that the UTS increased with an increase in both rPE and SD content as has been reported (Manaila, Stelescu, Craciun, and Ighigeanu, 2016) [34] . The UTS could be correlated with the density of composite materials as has been reported [35] With respect to Wood One Composite Samples (W₁S_1-3), it was observed that the UTS decreased from 0.99 N/mm² inW₁S₁ to 0.29 N/mm² in W₁S₂ and rose to 2.01 N/mm² in W₁S₃.

For W₁S_4-6, the UTS decreased from 4.32 N/mm² in W₁S₄ to 3.31 N/mm² in W₁S₅ and then to 3.32 N/mm² in W₁S₆. In the case of Wood 2 Composite Samples (W₂S_1-4), the UTS decreased from W₂S₁ to W₂S₃. For W₂S_4-6, the UTS increased from 1.70 N/mm² to 2.79 N/mm². The UTS of Wood 3 Composite Samples (W₃S_1-3) increased from 0.58 N/mm² in W₃S₁ to the optimum point of 1.67 N/mm² in W₃S₂, while for W₃S_4-6, it decreased from 1.950 N/mm² in W₃S₄ to 1.435 N/mm² in W₃S₆. For Wood Four Composite Samples (W₄S_1-3), a similar trend was obtained as in Wood Three (W₃S_1-3).

The highest UTS for W_1-4S_4-6, was obtained at W₁S₄ (14.32 N/mm²) while the least was 0.29 N/mm² in W₁S₂ (Figure 11). The highest value at W₁S₄ (14.32 N/mm²) was 220% greater than C₁ and 122% greater than C₂. Hence, W₁S₄ had the highest ability to withstand tearing due to tension than other wood species. The high value of W₁S₄ (14.32 N/mm²) could be attributed to the enhanced compatibility of 40/50% rPE/rPET and wood dust filler at 10% loading compared to 30% loading.

The results of the percentage elongation of the composites studied are presented in Table 2 and in Figure 13 and Figure 14.

The percentage elongation of these wood composites was generally above that of C₂. The highest percentage elongation for S_1-3 was in W₄S₁ (17.08%), which was 16.08% above C₁.

Similarly, the highest percentage elongation for S_4-6 formulation was at W₁S₅ (26.84%), which was 174.71% above C₂. This implied that W₁S₅ composite gave the greatest ductility and should have the greatest toughness. This showed that formulation involving 40% PE was superior in ductility than that with 20% PE as a binder. This is in line with the result of the flexural strength experiment discussed earlier in section 3.3. Statistical analysis of data confirmed the prime position of W₁S₅ in the scheme of things as well. The analysis of variance (ANOVA) revealed that there was a statistically significant variation in the wood composites of Wood 1 based on the Physico-mechanical values as the p-value (0.020) obtained was less than the critical level of α = 0.05. Hence, to further determine which wood composites significantly varied across the group based on their physico-mechanical values, the multiple comparison test was performed and results obtained showed that W₁S₅ composites are significantly different from W₁S₁, W₁S₂, and W₁S₃ composites with p-value (0.018, 0.009 and 0.001) respectively. Thus, W₁S₅ had a significantly higher proportion of the desired physico-mechanical values compared to other wood composites.

The results of thickness swelling (TS) are presented in Table 2 and in Figures 15-18.

Overall, it could be deduced that the TS did not follow any particular pattern of increase or decrease based on the ratio of SD-rPET-rPE mixture. The differences observed in TS of the composites can be attributed to variations in blending efficiency, wood percentages, density and pressing condition (Halligan, 1970) [36] . But it was observed that the SD-rPET-rPE composites having lower percentages of rPET-rPE were susceptible to the thickness swelling than those of panels having higher rPET-rPE content. This might be due to the increasing SD content in the WPC formulation.

Ayrilmis et al., 2011 [32] reported that for thickness swelling and water absorption of WPCs, wood fibres were mainly responsible. The TS at room temperature (25˚C) ranged between 0.07 to 0.30 for composites with up to 30% of SD content. Averagely, the lower thickness swelling was found in composite mixtures with the following percentage ratio rPE 20% rPET 60% SD 20%. Wood has a critical surface energy in the range of 40 - 60 ml/m² (Gupta et al., 2007) [37] which is higher than that of PET and PE. The large difference in surface energy between PET, PE and wood could make the various composite mixtures become hydrophobic. From the statistical analysis, it was observed that there was

no significant difference for all samples, that is the composites of Wood One through Wood Four (W_1-4S_1-6) at three different temperatures (25˚C, 50˚C and 75^°C) as their p-values were greater than the critical level of α = 0.05 respectively.

The results of water absorption capacity of the composites are presented in Table 3 and in Figures 19-21. With respect to the Composites of Wood One (W₁S_1-6), it was observed that at 25^°C, WAC of W₁S_1-3 increased from 31.40% in W₁S₁ to a maximum of 47.04% in W₁S₂ and then decreased to a minimum of

26.14% in W₁S₃. A similar trend was observed in W₁S_4-6, with a maximum WAC of 23.65% inW₁S₅. The highest water absorption capacity of the composites at 47.04% for W₁S₂ was 66.28% greater than C₁, while that of W₁S_4-6 which was highest at W₁S₅ was 16.81% greater than C₂. This shows that the composite containing 20% PE had more WAC and hence may be more porous and more prone to performance problem than the one containing 40% PE. A comparison of

water absorption capacity for the composites of the four wood samples at the two temperatures 25^°C and 75^°C studied, showed that at 25˚C for S_1-3, the wood dust increased in the composites as the WAC increased with increase in SD in the composites for W₁ and W₃. But for W₂ and W₄, it decreased. Similarly, for S_4-6, as the wood dust increased, the WAC of the wood composites also increased for W₁ and W₃ but not for W₂ and W₄.

This showed that the two composite formulations for the wood species gave the same trend in water absorption. At 75^°C, it was observed that for S_1-3, as the wood dust increased in the composite, the WAC for composites of W₁ and W₄ decreased while those of W₂ and W₃ increased. The same trend was observed at S_4-6 compositions.

The increase in WAC of composites as wood dust increased has been evaluated. It was observed that as wood dust increased with a decrease in the volume fraction of PET, WAC also increased (Atoyebi, Adediran and Adisa, 2018) [38] . WAC of composites of wood tends to increase with wood dust loading because of the structural similarity between water and wood species. The ability of the composites to absorb water is attributed to the hydrophilic nature of SD particles. Wood which is the main content of SD is a hydrophilic porous composite which consists of cellulose, lignin and hemicellulose, polymers that are rich in functional groups such as hydroxyls, which readily interact with water molecules by hydrogen bonding (Clemons, 2002) [39] and due to this reason, the WPCs have the potential ability to uptake water under humid condition (Adhikary et al., 2008) [40] . On the other hand, the higher water resistance of composites can be attributed to the hydrophobic characteristics of both PE and PET, though they are semi-crystalline in nature. Composite materials are known to degrade when subjected to humid conditions, and therefore the humidity confounds the difficulty of determining the high strain rate behavior of composites. Nevertheless, it was noted that the trend for W₁ and W₂ in WAC with temperature was reversed in the two temperatures (25˚C and 75˚C) studied. However, in W₃ and W₄, such reversals did not occur. The reversal of trend in WAC with temperature may be attributed to a change in the structure of the composites at higher temperatures for W₁ and W₂ but not for W₃ and W₄. Overall, it could be deduced that the WAC did not follow any particular pattern of increase or decrease based on the ratio of SD-rPET-rPE mixture. The highest water absorption content of 47.04% was noticed in W₂S₂ at 25^°C and 42.58% was noticed in W₄S₄ at 75^°C, while the lowest water absorption content of 1.06% was observed in W₃S₃ at 25^°C and 1.73% was observed at 75^°C. Several researchers have found that water absorption by composites brings about a breakdown of matrix dominated quasi-static properties (Woldesenbet, Gupta and Vinson, 2002) [41] . Composites with high water absorption tend to decrease in performance with time (Nayak, Mahato, and Ray, 2016) [42] . In light of this argument, the results as shown in Figure 20 and Figure 21 a show indicate that sample W₁S₅ has the lowest water absorptive quality since it has the lowest average value for both 25% and 75% combined and thus could be the best among the rest in performance. Furthermore, this composite could be the most environmentally friendly since studies have indicated that it could have the least carbon footprint in the environment (Duruaku, Ajiwe, Okoye and Arinze, 2016; Duruaku et al., 2023) [43] [44] .