Interrelationship between Static a nd Dynamic Strength Properties of Wood and its Structural Integrity

Various biotic and abiotic agents affect the performance of wood products. Chemicals, thermal energy, radiation, as well as different organisms have the potential to alter the optical, haptic and functional performance of wood. These effects come along with a change of structural integrity of wood, which in turn affects its strength properties. Therefore, a test was developed to quantify the structural integrity of wood in terms of its Resistance to Impact Milling (RIM). In a High-Energy Multiple Impact (HEMI) – test, steel balls were used in a heavy vibratory mill for crushing wood samples. Thousands of single events were captured by analyzing the fragments. Based on the degree of integrity and the percentage of fi ne fragments (< 1mm), an indicator has been defi ned to detect structural changes on cell wall level with high sensitivity. The aim of this study was to investigate the variation of structural integrity within and between ten different wood species in comparison with some strength properties according to standardized test protocols and in dependence of wood density. HEMI-tests, bending tests, and impact bending tests were performed with matched specimens. Wood density turned out to have only a subsidiary effect on structural integrity, but is dominanting standard strength properties. Thus, RIM was found to be only slightly correlated with the impact bending strength (IBS) and bending strength (MOR). On the other hand, the method shows clear insensitivity to natural variation in anatomy of wood.

Ključne riječi: čvrstoća na savijanje, test višestrukih udaraca visoke energije (HEMI), žilavost materijala, modul elastičnosti, otpornost na udarce (RIM) Various biotic and abiotic agents are affecting the performance of wood and wood based products.Chemicals, thermal energy, radiation, as well as different organisms have the potential to alter the optical, haptic and fi nally functional performance of wood (e.g.Berger et al., 2006;Sandak et al., 2015;Willems et al., 2015).Many of these effects come along with a change of the structural integrity of wood, which in turn affects its strength properties.Therefore, a test was developed to determine the structural integrity of wood in terms of its resistance to impact milling.A High-Energy Multiple Impact (HEMI) -test has been designed using steel balls in a heavy vibratory mill for crushing wood samples.The development work started after realizing that bending and impact bending strength tests were not suitable for quality control of -in this case -thermally modifi ed wood (Brischke et al., 2006a).They suffered from insuffi cient reliability and reproducibility and required a high number of carefully selected and precisely manufactured replicate samples.Consequently, time and costs became unacceptably high for industrial quality control (Rapp et al., 2006).
The aim was to overcome the drawbacks of standard strength testing, but keeping the advantage of examining a highly sensitive wood property such as its dynamic strength, which is strongly affected by structural changes of the material.Instead of using multiple replicates, the number of events affecting the wood should be multiplied.After a fi rst attempt to use a shotgun to apply several hundred pellets at a time on wooden boards -which might cause signifi cant security problems in the lab -a heavy vibratory ball mill was used for crushing wood samples (Brischke et al., 2006a).The Resistance to Impact Milling (RIM), which can vary between 0 and 100 %, is used here as a measure of wood structural integrity.
Heat treatment of wood goes along with a drastic strength loss (e.g.Esteves and Pereira, 2008).In particular, the dynamic strength properties are affected, and so is RIM.Excellent correlation was obtained between RIM and the severity of thermal modifi cation expressed as decrease in mass (dm) or in terms of color changes (Brischke et al., 2006a(Brischke et al., , 2012;;Rapp et al., 2006).As shown for 14 different wood species by Welzbacher et al. (2012), the treatment intensity can be estimated from RIM with fairly high precision.
Different kinds of chemical modifi cation were also examined.Their very different effects on the structural integrity of wood were found to be detectable.While furfurylation and treatments with DM-DHEU (dimethylol dihydroxyethyleneurea ) and melamine resin led to a signifi cant decrease in RIM, hydrophobation with oils and waxes increased the structural integrity (Brischke et al., 2012).Impregnation with oil and wax obviously did not weaken the wood structure, but increased the RIM, which might be explained by hydraulic effects (e.g.Ulvcrona et al., 2006).Furthermore, a remarkably reduced amount of fi ne fragments indicated 'adhesion effects', which might also have a positive effect on the structural integrity.
Hydraulic effects have also been observed on water saturated samples by Brischke et al. (2014), when testing different wood species used in the marine environment.While the structural integrity decreased with increasing moisture content in the hygroscopic range, it increased again up to full water saturation due to hampered short term compression of the wooden cells when the lumens were fi lled with water.
The HEMI test was furthermore used to detect incipient decay (Brischke et al., 2006b).Brown and white rot had clearly different effects on the structural integrity.Partly, differences were found even on fungal species level (Brischke et al., 2008).Fungal decay was found to be detectable before signifi cant mass loss was determined.Furthermore, Huckfeldt et al. (2010) showed that drilling cores taken from full size structures can also be used for HEMI tests for early detection of fungal degradation.Samples from archaeological objects, such as the Vasa shipwreck in Stockholm, Sweden, were investigated with the HEMI method (Rapp et al., 2008).
Gamma radiation, which is a common method for sterilization of wood samples, e.g. for laboratory resistance tests, did negatively affect the structural integrity of wood (Despot et al., 2007).Degradation of cellulose through radiation led to signifi cantly reduced RIM.On the other hand, the HEMI method was found to be almost unaffected by wood density, weathering

Strength tests 2.3. Ispitivanje čvrstoće
Impact bending tests were performed on 300 x 20 x 20 mm³ specimens using a Otto-Wolpert pendulum impact machine according to DIN 52 189-1 (DIN 1981).The static bending tests were run to determine the bending strength (modulus of rupture MOR) and the modulus of elasticity (MOE) of wood according to DIN 52 186 (1978).Three-point bending tests were performed on 360 x 20 x 20 mm 3 specimens with concentric force parallel to the grain on a universal testing machine Zwick/Roell Z100.Before the mechanical strength tests, all samples were conditioned in a standard climate at 20 °C and 65 % relative humidity until constant mass was achieved.

High-energy multiple impact (HEMI) -tests 2.4. Testovi višestrukih udaraca visoke energije (HEMI testovi)
The development and optimization of the High-Energy Multiple Impact (HEMI) -test have been described by Rapp et al. (2005) and Brischke et al. (2006a, b).In the present study, the following procedure was applied: 20 oven-dried specimens of 10 (ax.) x 5 x 20 mm³ were placed in the bowl (140 mm in diameter) of a heavy-impact ball mill (Herzog HSM 100-H; Herzog Maschinenfabrik, Osnabrück, Germany), together with one steel ball of 35 mm diameter for crushing the specimens.Three balls of 12 mm diameter and three of 6 mm diameter were added to ensure impact with smaller wood fragments.The bowl was shaken for 60 s at a rotary frequency of 23.3 s -1 and a stroke of 12 mm.The fragments of the 20 specimens were fractionated on a slit sieve according to ISO 5223 (1996) with a slit width of 1 mm using an orbital shaker at an amplitude of 25 mm and a rotary frequency of 350 min -1 for 2 min.The following values were calculated: (2) Where the degree of integrity I is the ratio of the mass of 20 biggest fragments m 20 to the mass of all fractions m all after crushing.effects (e.g.cracking) and blue stain (Brischke et al. 2009), which requires further testing.
Previous tests with different wood-based materials indicated that RIM and mechanical properties, such as impact bending strength and bending strength, are only poorly correlated, although both are negatively affected e.g. by heat, fungal decay or radiation, but are obviously affected by different structural features (Welzbacher et al., 2011;Brischke et al., 2014).
Therefore, in this study, the structural integrity of different wood species was determined on samples, which had previously been submitted to standard impact bending tests.Furthermore, matched specimens were subjected to three-point-bending tests to determine the modulus of elasticity (MOE) and the modulus of rupture (MOR).

Drvo
Specimens were prepared from European-grown wood species, i.e. four softwoods and six hardwoods as shown in Table 1.Specimens for impact bending and bending tests were cut from the same planks; HEMI test specimens were cut from impact bending test specimens after the tests to match them exactly in axial direction.

Determination of oven-dry density 2.2. Određivanje gustoće apsolutno suhog drva
The oven-dry density was determined on HEMI test specimens since they were axially matched with the impact bending test specimens.The specimens were oven dried at 103 °C till constant mass, weighed to the nearest 0.01 g and the dimensions were measured to the nearest 0.01 mm.The oven dry density was calculated according to the following equation: Where ρ 0 is the oven-dry density, m 0 is the ovendry mass, and V 0 is the oven-dry volume of specimens.
Where the fi ne fraction F is the ratio of the mass of fragments < 1 mm to the mass of all fractions m all multiplied by 100.

(4)
Where RIM is the resistance to impact milling as a measure for the structural integrity of the material.

REZULTATI I RASPRAVA
The results of various mechanical tests are summarized in Table 2.A wide range of density and different strength related properties was covered with Poplar showing lowest density and mechanical strength and Robinia with the best property values.However, Beech and Hornbeam showed similar or even higher RIM compared to Robinia.Compared to other mechanical properties, RIM differed significantly less between wood species.In contrast, I and F showed much higher variations than the resulting RIM, which points to the anatomical peculiarities of different wood species.For instance, high F values were found for softwoods, which also showed generally lower IBS, and English oak showed the highest F value of all.The softwoods are characterized by weak earlywood portions that alternate with latewood areas, whereby the transition from one tissue to the other can be more or less abrupt (e.g.Schweingruber, 2012).English oak, however, suffers from very thick parenchyma rays and large early wood vessels, which in combination might explain both, low RIM and IBS.

Wood species
As expected, an excellent correlation was established between IBS and MOR (R² = 0.9034, Figure 1.Rather unexpected, the correlation between MOE and MOR was poor (R² = 0.491, Figure 2  The reason for MOR and IBS being so well correlated indicates a density dependency, which is illustrated in Figure 3.If only the mean values for different wood species are considered, IBS and MOR are well correlated with density, while RIM and MOE are not (Figure 3).However, if single values within one species are considered, as exemplarily shown for Norway spruce and Siberian larch in Figure 4, no correlation is achieved, neither between RIM nor IBS and density.This coincides with previous studies, where RIM appeared to be unaffected by density within one wood   species (Brischke et al., 2009), but also across different wood species (Brischke et al., 2014).Consequently, other anatomical characteristics need to be responsible for the signifi cant differences in structural integrity between wood species.Nevertheless, as shown in Figure 5, MOR (R² = 0.5561) and IBS (R² = 0.67) seem to be correlated with RIM, but MOE (R² = 0.1051) is obviously not.An overview of cross-correlation of the investigated wood properties is given in Table 3. Density is the dominating parameter for the majority of strength properties of wood, but numerous further factors have the potential to affect different strength and elasto-mechanical properties of wood.Fiber length, composition, size and amount of rays, lignin content, the micro fi bril angle, and others lead to intra-and inter-species specifi c variation of wood properties (Niemz and Sonderegger, 2003) and are not necessarily the same for different wood properties such as the investigated parameters in this study.Dynamic strength properties, such as IBS and RIM, are affected by anatomical abnormalities, which can easily overrule the infl uence of wood density (Ghelmeziu, 1937; von Pechmann, 1953; Niemz and Sonderegger, 2003).

ZAKLJUČAK
Wood density seems to have only a subsidiary effect on the structural integrity of wood as determined in high-energy multiple impact (HEMI) tests.Furthermore, the RIM seems only slightly correlated with standard strength properties of wood such as IBS and MOR.More likely, anatomical features within one wood species, and in particular between wood species, have stronger effects on the structural integrity of wood and thus on its brittleness.Consequently, future studies need to include microscopic studies to further elaborate the effect of anatomical characteristics such as micro fi bril angles and cell and tissue volume ratios.
The limited transferability of RIM to established strength properties of wood comes along with its insensitivity to natural variation in anatomy of wood.This retrieves the advantage of a high discriminatory power for detecting structural changes, e.g.caused by fungal decay or cell wall modifi cation.