The specimens used in the study were prepared according to Australian Standards for mechanically testing small clear specimens of timber (Mack 1979) and were of the following sizes:

The samples were taken from the sapwood of 5 years old back-sawn Jabon wood (Anthochepalus cadamba (Roxb) Miq) from West Java Province, Indonesia. As the Jabon wood was only 5 years old the heartwood part was very small. The test samples were prepared from “clear” and “straight grained” wood which did not contain unfavourable features such as knots, cross grain, checks, and splits (Green et al. 1999). A total of 768 specimens were prepared and divided into 4 groups based on the type of treatment applied, as follows:

Impregnation material included Merbau extractives (ME) which were prepared using the following method.

Merbau wood powder, with a moisture content of 18%, was prepared using a grinding machine. Extraction was done by maceration, in which 500 mL of the powder passed through a 40-mesh screen, was dispersed in hot water of 80 °C, and stirred every 3 h at room temperature for two 24-h periods. The mixture was filtered with a glass filter, resulting in the separation of the first filtrate (1) and the residue. This maceration procedure was repeated for the residue, resulting in a second filtrate (2). The first (1) and the second (2) extracts were combined to make a concentrate (with the use of a rotary evaporator), which was freeze-dried to powder. The properties the ME have been thoroughly investigated by Malik et al. (2016). Hot water was used as the solvent in this study because according to Kislik (2012), water is the best alternative for organic solvent. It is environmentally friendly, relatively cheaper, safe, nontoxic, imflammable, and recyclable. Merbau extractive (ME), being a water soluble material, is easy to extract by water. The polymerised Merbau extractives (PME22 and PME33) were made from the ME which has been polymerised in base condition with formaldehyde and resorcinol addition as the copolymer. The resin is classified as a resole because the mole ratio of R/F should be <1 or F/R ≥1, and polymerisation can be done at room temperature. Impregnation treatment was done in a vacuum-pressure vessel where the wood samples were placed into open-top container of a vacuum-pressure unit where connected by a hose to the container of liquid of polymerized Merbau extract (PME) that has been prepared. The open-top container containing the samples was vacuumed using the compressor at the power of 0,1 kg/cm² for 30 minutes. The vacuum was then released and the container was allowed to be filled by the liquid of impregnant, so the liquid level was 20 cm higher than the submerged wood samples. Pressure was then applied and maintained at 15 kg/cm² for 1 hour. This procedure was applied for the samples of the three impregnating solutions (ME, PME22 and PME33). After that, all specimens were drained and then put into a conditioning room until a constant weight was reached (Malik et al. 2015, Malik et al. 2016, Malik et al. 2018). The moisture content of all tested samples was 12% to avoid - according to Rowell (1996) - a misleading or even invalid values if the samples were tested at different moisture levels.

Thirty two replicate samples measuring 30 x 30 x 90 mm (ratio 1 : 1 : 3) were prepared from each treatment. According to the “Australian Standards for Mechanically Testing SmallClear Specimens of Timber” (Mack 1979), the dimensions of hardness specimens are 50 x 50 x 150 mm, but the test may also be made on a specimen of any other size, as long as the thickness is at least 25 mm and the ratio of 1:1:3 is maintained. In this study specimen dimensions were 30 x 30 x 90 mm as it was impossible to obtain 50 mm thickness. The hardness was tested using an Instron® Universal Testing Machine according to the above standard. The hardness test was done on all faces of the specimens to obtain side-hardness (4 sides) and 2 cross-sections. The reason for testing all faces of each specimen was to ensure a more representative hardness value (Green et al. 2006). Before and after testing, the weight and dimensions of each specimen were measured for determination of wood density and moisture content.

The shear strength test was carried out according to the procedure of Mack (1979). The loading was applied continuously throughout the test at a rate of motion of the movable crosshead of 0,6 mm/min until a break or separation occurred on the surface of the test samples. The test was done on 256 specimens with 32 repetitions for each treatment on radial and tangential sections.

A number of 128 specimens were prepared for bending strength tests (MOE and MOR) using an Instron® Universal Testing Machine according to The Australian StandardsProcedure for Mechanically Testing Small Clear Specimens of Timber (Mack 1979) for specimen size adjustment. The loading rate was 1,0 mm per-minute and the specimens were loaded on the radial face.

The physico-chemical analysis was employed to determine whether there is a chemical change on Jabon wood after impregnation treatment that affect the mechanical characteristics. Two methods were used: Fourier Transform Infrared Spectroscopy (FT-IR) that aimed to analysis functional groups and finger print alteration and X-Ray Diffraction (XRD) Analysis to determine degree of crystallinity of the samples from untreated and treated wood. The amount of 15 mg of treated and untreated samples of 120 mesh size were directly used in the FTIR spectroscopy measurement. They were then embedded in kalium bromide (KBr) pellets and analysed using a MB3000 (ABB, Canada) spectrometer. They were scanned in the absorption mode in the range of 4000 to 500 cm⁻¹ with the resolution of 2 cm⁻¹. The crystallinity of the samples from untreated and treated wood was measured using X-Ray Diffraction (XRD) (MAXima_X Shimadzu® XRD-7000). The X-ray beam was powered with a 40kV, 30mA source and scans made in the range from 0-40 degrees at a scan speed of two degrees per minute. The degree of crystallinity was calculated as the ratio of the intensity differences in the peak positions.

The changes in mechanical properties of impregnated Jabon wood samples were qualitatively analysed by comparing through the changes of physico-chemical characteristics of impregnated and non-impregnated samples. All data of mechanical properties changes were then analysed by one-way analysis of variance to determine if there were any significant differences between the samples. Then, a post-hoc multiple comparison by Tukey was done to determine whether there was a significant difference between the groups.

The changes of physical and mechanical properties of wood can occur by chemical modification due to changes in the chemistry of the wood cell wall polymers. These properties can vary from simple colour change in the wood to major changes in brittleness, hardness, strength, stiffness, density, and moisture sorption (Rowell 1996). The changes may include decreasing, increasing, or even no effect on the properties, depending on the type of chemical used in the treatment. The results of tests of mechanical properties of impregnated Jabon wood i.e hardness, shear and bending strength are presented in Table 1. The analysis showed that the density values of the samples were significantly different for different treatments (Sig = 0,000) (Table 2). As reported by Malik et al. (2015), the density of Jabon wood after impregnating treatment using polymerise merbau extractives - PME22 and PME33 - increased by 26 and 30 %, consecutively. Generally, almost all mechanical properties of wood are known to be closely associated with wood density (Brown et al. 1952, Panshin et al. 1964, Sharp 2003). Table 1 shows that there are increments in mechanical properties after impregnation treatment with PME22 and PME33, but not in the samples impregnated with merbau extractives (ME).

Table 1:
Mechanical characteristics of treated and untreated Jabon wood.

Table 2:
Analysis of variance of and density of treated and untreated Jabon wood samples.

The increment of mechanical properties of treated Jabon wood after impregnation treatment have been confirmed by FTIR spectra and crystallinity (XRD) analysis (Malik et al. 2018). A coupling reaction between wood and the impregnant can be confirmed by FTIR spectra. The reaction of treated Jabon wood by impregnation using polymerised Merbau extractives (PME22 and PME33) was shown by the presence of the peaks at 2922,16 (PME22) and 2895,15 (PME33), 2002,11 (PME22) and 1928,82 (PME33), 1327,03 (PME22) and 1330,88 (PME33), 1033,85 and 1041,56 for C-H and CH₃ stretching in polymerized Merbau extractives (PME), aromatic combination bands, C-H deformation in polymers (aromatic aldehyde) and C-O ether bond stretching vibrations in PME reaction, respectively, where they do not exist in untreated (UT) Jabon wood samples (Malik et al. 2018). This permanent chemical presence has not only resulted in the colour change but also in increased mechanical properties. Using FTIR spectra analysis to confirm the coupling reaction between solid wood and impregnating material and its effect on the wood mechanical characteristics was also carried out by other researchers (Hamdan and Islam 2012, Dong et al. 2016). Hamdan and Islam (2012) showed an optimistic improvement in physical and mechanical properties that were confirmed by FTIR spectra. The results revealed the reaction of wood based hydroxyl groups with diazonium salt woods which were modified by benzene diazonium salt formulation. Using rosin as impregnant, from FTIR analysis Dong et al. (2016) stated that there may be hydrogen bonds formed by the reason of the affinity between hydroxyl groups in wood and carboxyl groups in rosin, which could improve the interaction between the resin and wood. In addition, results of X-ray diffractometer (XRD) analysis showed that the crystallinity of both PME22 and PME33 was very low by 8,76 and 10,03%, respectively. However, this resulted in increased crystallinity of the impregnated Jabon wood from 46,55% (UT) to 48,25% (impregnated by PME22) and 50,84 % (impregnated by PME22). In contrast, a decrement of the crystallinity occurred in the samples impregnated with ME by 1,02%. According to Santoso (2014), these would have an implication on hardness, shear and bending strengths. The changes of crystallinity values are shown in Figure 1.

Figure 1:
Diffractograph of untreated (UT) and treated Jabon wood samples using ME, PME22 andPME33. Remarks: UT, ME, PME22 and PME33 refers to Table 1.

Figure 2:
Overlaying diffractograph of the increment of crystallinity treated Jabon wood samples using PME22 and PME33.

Table 3:
Crystallinity and molecular weight of treated and untreated Jabon wood samples.

PME22 and PME33 can be classified in low molecular weight (MW) polymerized material (PME22 = 3164 and PME33 = 3615). Due to the low MW, these materials can penetrate the cell wall of the treated wood. Thus, crystallinity of the treated wood with PME22 and PME33 increased, or in other words, the increments of crystallinity of the treated wood samples is an evidence that both PMEs penetrated the cell wall. Crosslinking reaction occurred between functional groups of PMEs and the cell wall material. This finding is in line with Wu et al. (2011), who stated that a quasi-crystalline was formed as the result of the crosslinking reaction. The crosslinking material not only entered the cell cavity and cell gap, but also filled in the cell walls and the micro-fibrils (Wang et al. 2003, Liu et al. 2009, Ma etal. 2013). This led to the polymerization between the functional groups of wood and the chemicals which resulted in increase of physical and mechanical properties of the impregnated (Chen et al. 2013). These results have been confirmed by this current study as presented in the following sections.

Hardness is an important property of timber for various applications such as for flooring, furniture and some structural uses. It provides an indication of how well the wood performs in relation to wear and denting (Thomas et al. 2009.). Impregnation treatment can improve the hardness, as revealed by many researchers. Hamdan and Islam (2012) reported that the hardness of five types of selected tropical light hardwoods had higher hardness after impregnation using benzene diazonium salt, compared to their corresponding non- impregnated control wood. The increase of wood hardness is due to the increase of interfacial adhesion between the polymer and wood that improves water exclusion, decreases the rate of swelling, and increases hardness of WPC (Ellis and O'Dell 1999). Hardness can be also generally enhanced through impregnation and in-situ polymerization (Zhang et al. 2006, Keskin et al. 2004). The wood hardness increases with the increasing density (Forest Product Laboratory 1940). According to Rowell (1999), the increase in hardness is more than proportional to the increase in specific gravity. In present study, as shown in Table 1 and Figure 3 both end- and side- hardness increased following the increment in the density from 340 kg/cm³ (UT) to 430 kg/cm³ (PME22) and 440 kg/cm³ (PME33). Statistically, the differences in hardness increments after impregnating treatment using PME22 and PME33 were significant as determined by one-way Anova at α = 0,05 (F(3119) = 26,386 Sig = 0,000). Post-hoc analysis by Tukey showed that the hardness resulted from impregnating 300 treatment using PME22 (He = 3934,43 ± 667,44 N; Hs = 2777,24 ± 418,16 N, Sig = 0,000) 301 and PME33 (He = 4350,52 ± 655,87 N, Hs = 2976,12 ± 727,85 N, Sig = 0,000) were significantly different from non-impregnated samples/UT (He = 3277,58 ± 543,68 N, Hs = 2127,45 ± 626,45 N) and from samples impregnated with ME (He = 3246,0 2345,36 ± 480,33 N). Even, the hardness of impregnated samples using PME22 and PME33 were significantly different from each other. There was no significant difference between the hardness of UT and ME, as well as their density.

Figure 3:
Hardness increment of Jabon wood after impregnating treatment using PME22 and PME33: (a) End-hardness; (b) Side-hardness.

Shear strength is defined as an ability to resist internal slipping of one part upon another horizontally along the wood fibre (Green et al. 1999, Shmulsky et al. 2011). As well as hardness, shear also increased with increased density due to the impregnating treatment (Chow 1966, Rowell 1999). Table 1 and Figure 4 shows that both radial (R)- and tangential (T)- shear strength increased after impregnating treatment using polymerised Merbau extractives (PME22 and PME33) from 6213,49 ± 381,48 kPa (R) and 6635,18 ± 704,12 kPa (T) at UT to 8815,20 ± 1367,05 kPa (R) and 9925,31 ± 1668,11 kPa (T) at PME22 and to 10812,81 ± 1234,66 kPa (R) and 11883,70 ± 1289,57 kPa (T) at PME33 Jabon wood samples. Meanwhile, there was no significant improvement of the shear strength of Jabon wood samples after impregnating treatment with ME; the shear slightly decreased on radial section. Statistically, the shear strength changed after impregnating treatment using PME22 and PME33. The shear values were significantly different as determined by one-way Anova at α = 0,05 (F(7248) = 143,460 Sig = 0,000). Further, post-hoc analysis by Tukey method showed that the shear values resulted from impregnating treatment using PME22 and PME33 were significant different from UT and ME samples both on radial and tangential sides. This analysis also revealed that there was a significant difference in the shear strength between radial and tangential sides of UT samples. However, there was no significant difference between radial- and tangential- shear strength of ME samples.

Figure 4:
Shear increment of Jabon wood after impregnating treatment using PME22 and PME33. UTR = Non-impregnated samples (controls) on Radial section; UTT = Non-impregnated samples (controls) on Tangential section: MER = Impregnated samples using Merbau extractives on Radial section; MET = Impregnated samples using Merbau extractives on Tangential section; PME22R = Impregnated samples using PME22 on Radial section; PME22T = Impregnated samples using PME22 on Tangential section; PME33R = Impregnated samples using PME33 on Radial section;.PME33T = Impregnated samples using PME33 on Tangential section.

Impregnation treatment generally improves flexure characteristics of impregnated wood. In regards to using polymeric material as the impregnant, Zhang et al. (2006) concluded that modulus of elasticity was enhanced through impregnation and in-situ polymerization. Moreover, the increase of modulus of rupture (MOR) and modulus of elasticity (MOE) was related with weight percent gain of the impregnated wood (Deka and Saikia 2000). In line with these earlier studies, it should be pointed out that the increments of the flexure characteristics have been also observed in impregnated Jabon wood, as shown in Table 1 and Figure 5. From the results presented in Table 1, it can be seen that MOE increased from 4888,62 ± 734,93 MPa (UT) to 6038,47 ± 658,06 MPa in PME22 samples (23,52%) and to 6849,99 ± 597,26 MPa in PME33 samples (40,12%). MOE value decreased by -1,04% on impregnated samples using ME. MOR values increase by 3,27; 28,50 and 41,19 % after impregnated with ME, PME22 and PME33, respectively. Statistical analysis indicate that the increments of MOE and MOR were significantly different (F(3124)= 6038,47 MPa, Sig = 0,000 for MOE; F(3124)= 54,98 MPa, Sig = 0,000 for MOR). Furthermore, post-hoc analysis by Tukey showed that MOE and MOR values of the impregnated samples using PME22 and PME33 samples were significant different compared to UT and ME samples. The MOE values of PME22 and PME33 samples were also significant different. However, the MOR values of PME22 and PME33 samples were not significantly different as well as MOR of UT and ME samples and be amorphous area or by para-crystallin of microfibril. In addition by para-crystallin can be related to crystallization process of cellulose in In addition, Yunianti (2012) reported that there is a strong linear correlation (r=0,83) between MOE with- in certain limit - crystallinity. In the present study, as demonstrated in Figure 2 and Table 3, it could be assumed that the increment of the bending strength of impregnated Jabon wood is related to the crystallinity increment. Chen et al. (2013) stated that there is no doubt that the mechanical properties are closely related with the strength of the wood cell wall.

Figure 5:
The increment of bending strength of Jabon wood after impregnating treatment using PME22 and PME33: (a) MOE and (b) MOR.

^♠Corresponding author: jmalik.unimelb2012@gmail.com