Abstract
The effect of annual ring number and width on the longitudinal (P) and transverse (S) ultrasonic wave velocities in the radial direction of black pine, Scots pine, Turkish red pine, and cedar softwoods was evaluated in this study. Annual rings were evaluated using high-resolution images captured with a Lumix camera. The 2.25 MHz P and 1 MHz S wave frequencies were propagated through the radial direction of small clear samples. An increase in ring number caused different changes in the P and S wave velocities. Only Scots pine and cedar presented continuous decreases in P and S wave velocities with the increase in annual rings. On the contrary, VR of Red pine slightly decreased and surpassed the initial value when the ring numbers increased from 5 to 10 and 15, respectively. Furthermore, the greatest decrease (4%) in the velocities was observed for VRL of Red pine. According to one-way ANOVA results, significant relations were only observed for VR vs. ring number of Black pine and cedar. R2 values ranged from 0.0001 (Red pine VR) to 0.18 (Cedar VR) for ring number and 0.0002 (Cedar VR) to 0.44 (Scots pine VR) for ring width. Furthermore, ANOVA results for linear regression analysis showed that VR of Scots pine and VR, VRL, VRT of Red pine can be statistically significantly predicted by the ring widths.
Download PDF
Full Article
Effects of Annual Ring Number and Width on Ultrasonic Waves in Some Softwood Species
Murat Aydın
The effect of annual ring number and width on the longitudinal (P) and transverse (S) ultrasonic wave velocities in the radial direction of black pine, Scots pine, Turkish red pine, and cedar softwoods was evaluated in this study. Annual rings were evaluated using high-resolution images captured with a Lumix camera. The 2.25 MHz P and 1 MHz S wave frequencies were propagated through the radial direction of small clear samples. An increase in ring number caused different changes in the P and S wave velocities. Only Scots pine and cedar presented continuous decreases in P and S wave velocities with the increase in annual rings. On the contrary, VR of Red pine slightly decreased and surpassed the initial value when the ring numbers increased from 5 to 10 and 15, respectively. Furthermore, the greatest decrease (4%) in the velocities was observed for VRL of Red pine. According to one-way ANOVA results, significant relations were only observed for VR vs. ring number of Black pine and cedar. R2 values ranged from 0.0001 (Red pine VR) to 0.18 (Cedar VR) for ring number and 0.0002 (Cedar VR) to 0.44 (Scots pine VR) for ring width. Furthermore, ANOVA results for linear regression analysis showed that VR of Scots pine and VR, VRL, VRT of Red pine can be statistically significantly predicted by the ring widths.
DOI: 10.15376/biores.17.1.1745-1763
Keywords: Annual rings; Ultrasound; Velocity; Softwood
Contact information: Department of Machine, Keçiborlu Vocational School, Isparta University of Applied Sciences, Isparta, Turkey; *Corresponding author: murataydin@isparta.edu.tr
INTRODUCTION
Ultrasonic testing is an evaluation method that is relatively less laboratory-dependent than other methods and that can be applied in various fields. Prominent properties of this type of testing include mobility and compatibility. However, interpretation of the ultrasonic signal is not easy, and an experienced user is needed for using the tools in manual mode. Also, the reproducibility of ultrasonic evaluation seems limited in manual mode. Therefore, operator-independent automated systems reduce the errors caused by misreading by the experienced or inexperienced user and improve the signal-to-noise ratio (Gros 1997). Interpretation of signals in wood science becomes more complicated due to the complex structure of wood material that presents different properties through the essential axis and planes. The anisotropic and inhomogeneous material structure of wood greatly attenuates ultrasound due to scattering from the inhomogeneity variations in the microstructure (Berndt and Johnson 1995). Therefore, an ultrasonic beam attenuates due to variation in composition, density, and porosity of the material (Gros 1997). Furthermore, these properties change due to environmental conditions and cause variations within the species that are much more complex than anisotropy (Brancheriau et al. 2012). The cellular microstructure of wood and the nanostructure of the cell walls control the elastic properties. Furthermore, the properties of wood are maximized in the longitudinal (L) direction (Ansell 2015). The same is true for ultrasonic wave velocities (UWV) in wood, and according to Dackermann et al. (2014), radial and tangential wave velocities are generally about a third of the longitudinal wave. Similar behavior is true for shear or transverse wave velocities, and transverse wave velocities of wood are remarkably slower than the longitudinal waves. These significant differences occur because of the longitudinally aligned anatomical elements such as fibers and tracheid, and radially aligned wood rays in growth rings. However, such structural elements are not observable through the tangential direction. When the polar orthotropic nature of wood is taken into consideration, the propagation of ultrasonic waves encounters these variations. Furthermore, micro-fibril angle and the length of the anatomical elements have influences on variations. For example, a small cell wall layer is responsible for higher acoustic wave velocities (Bergander and Salmén 2000). More comprehensive data for the propagation and distribution properties of longitudinal and transverse ultrasound waves in wood and wood-based materials are provided by Bucur (2006).
Changes in environmental factors may have essential effects on the anatomical properties of wood. One of the essential variations in wood structure is the annual rings (ARs) due to environmental factors in normal growth or engineered growth by sylvicultural practices. Tree-rings enclose the history of all experiences or developments that take place in the surrounding area within the tree stem. According to Carrer (2011), tree-rings are a valuable source of information due to the characteristics they have earned year by year. Many studies have evaluated tree-ring-related issues, and the fundamentals of the tree-ring research and evaluation are thoroughly presented by Schweingruber (1988) and Speer (2012). A hand lens with a scale is the simplest optical instrument for measuring the tree-rings. However, advanced tools, hardware, and software are the leading development for the evaluation of tree-ring and it’s effects on the properties. For example, Jackson et al. (2009) evaluated the ring-growth patterns of many wood specimens by terahertz time-domain reflectometry. Perlin et al. (2019) conducted an ultrasonic tomography investigation of timber slices through the tangential and radial directions regarding the growth ring direction. Mori et al. (2019) developed a non-destructive RW measurement technique and evaluated the effectiveness over the archaeologically excavated waterlogged wood. Adamopolous et al. (2009) determined the relationship between the ring-width (RW) and stem height for Pinus brutia using LINTAB and TSAP-Win. Trouillier et al. (2018) used optical scans and CooRecorder for RW measurements and CDendro for cross-dating of white spruce. Shishkova and Panayotov (2013) used CooRecorder and CDendro to measure and cross-date the tree-widths of Pinus nigra, respectively.
Knapic et al. (2007) investigated the relation between RW, the number of rings, density, and cambial age properties of oak wood. Kharrat et al. (2019) evaluated the RW, density, and modulus of elasticity (MOE) variations for black spruce and Jack pine woods using UWV measurement in terms of 5, 10, 15, and 20 annual ring numbers (ARNs) from the pith. They stated that the correlation between density and dynamic MOE is positive and statistically significant in rings. Dackermann et al. (2016) reported that AR acts as a barrier against waves and decreases the velocity of waves. Effects of inter and intra-tree (ring) variations were evaluated before. However, a comparison between the trees should be made over the same number of AR instead of similar size or diameter because results tend to be similar (Bendtsen 1978). From this point of view, figuring out the influence of AR on the ultrasonic wave by the same number instead of using the same sample size was aimed in this study. Consequently, this study tried to determine the effect of ARN and annual ring width (ARW) on the longitudinal and transverse UWV in the radial direction of four different softwood species in terms of the same number of rings.
EXPERIMENTAL
Materials and Methods
Black pine (Pinus nigra Arnold.), Scots pine (Pinus sylvestris L.), Turkish red pine (Pinus brutia Ten.), and cedar (Cedrus libani A. Rich.) woods were used in this study. Samples for each species were prepared with at least 5, 10, and 15 ARs. As seen in Figs. 1 and 2, samples were prepared according to the ARN instead of a certain dimension through the radial direction. Therefore, the dimensions of the samples in radial direction were different not only within the species but also between the species. The tangential and longitudinal dimensions were the minimum of 20 millimeters due to the diameter of the transverse transducers.
As can be seen in Fig. 1, predetermined cutting lines were marked on the laths by considering the start and finish points of AR. The laths were cut using a newly-sharped circular saw blade. Samples with improper or false rings and rings with compression wood were not tested. All the samples were prepared using only sapwood sections of the planks. Planks were prepared from the sections following the breast height of the logs. Furthermore, all the samples were cut from the same logs to prevent the probable variations dependent on the sampling. A total of 240 samples (20 for each ARN group) were tested.
Fig. 1. Sample preparation steps
Annual ring width (ARW) measurement was performed by image analysis and digital caliper. High-resolution images of the cross-section of the samples were taken using a Lumix GX1 camera and G macro 30 mm F/2.8 lens (Panasonic, Matsushita Electronics, Osaka, Japan). CooRecorder and CDendro (Cybis Elektronik and Data AB, Saltsjöbaden, Sweden) software were used to identify and point the AR borders for growth ring measurements, as seen in Fig. 2.
An Olympus Epoch 650 flaw detector (Olympus, Waltham, MA, USA) was used to measure the time of flight values in micro-second. Relatively high frequencies such as 1 Megahertz (MHz) can be used for the small (around 100 mm) specimens (Krause et al. 2015). Bucur and Kazemi-Najafi (2011) reported that 2.25 MHz transducer frequency provides greater energy and penetration into the material. Furthermore, Gonçalves et al. (2011, 2014), Hering et al. (2012), Vázquez et al. (2015), Niemz et al. (2017), Bachtiar et al. (2017), and Luis Gómez-Royuela et al. (2021) are some of the recent studies performed elastic characterization of wood using these central frequencies. Therefore, the effect of AR on UWV in the radial direction has been evaluated using longitudinal (2.25 MHz) and transverse (1 MHz) waves. Longitudinal wave velocity in the radial direction (VR), and transverse or shear wave velocities in the Radial direction and Longitudinal (VRL) or Tangential (VRT) polarizations were calculated using the transmitting time of ultrasound waves and dimensions of the samples through the radial direction.
Fig. 2. CooRecorder ring border pointing and RW for red pine (A), radial sections, and AR of the samples (B), and CDendro Plot details for cedar wood (C)
Defect-free samples were conditioned at 65% relative humidity and 20 ± 1 °C temperature. Densities were determined according to TS 2472 (2005) standard by stereo-metric method (dimension and weigh measurement using a digital caliper and precision scale, respectively).
One-way Analysis of Variance (ANOVA) was performed to interpret the effect of ARNs on velocities. Duncan’s multiple range (DMR) test was conducted to figure out the differences between the mean values of ARN groups. The confidence interval was 95%. The coefficients of determination (R2) by linear regression analysis were calculated to evaluate how differences in UWV can be expressed by ARW and ARN.
RESULTS AND DISCUSSION
The average values of ARW, density and UWV are presented in Table 1. The widths of the annual tree-ring strictly depend on the growth conditions. Temperature and precipitation are the responsible factors for the variations in tree-ring width (Dogan and Kose 2019). In general, widths are positively correlated with the amount of summer precipitation (Haneca et al. 2009), and it’s expected that location and environmental factors may make great variations in the ring properties. As can be seen in the table, variations for the ARW ranged from 7.7 to 17.6%. Reported variations for ARW were 20.4% (Mederski et al. 2013), 42.3% (Fabisiak and Fabisiak 2021), 15.3% (total chronology from 1958-2016) (Özel et al. 2021), and 60.5% (Büyüksari et al. 2017) for Scots pine, 40.8% (total chronology from 1970 to 2011) (Kantarci et al. 2013), 17.8 to 35.2% (Doğan and Köse 2015) for black pine, 39.2% (Adamopoulos et al. 2009) for red pine (Pinus brutia Ten.), and 60.9% (Öktem and Sözen 1992) for Cedrus libani. When compared to the reported data, it’s thought that relatively fewer variations for ARW were obtained due to sample preparations because samples were cut from the outer section of the matching planks, as mentioned in the experimental.
Variations in the ARW are to be expected due to climate changes. As noted by Köse et al. (2012), year-by-year alterations in environmental factors cause discrepancies in tree-ring widths, and in the literature, great varieties of ARW were reported for tested species. For example, variations of approximately 0.25 to 1.8 mm (Kemalpaşa Mountain) and from 1 to 6 mm (Karabelen Mountain) were observed for Pinus brutia Ten through 1942 to 1998 period (Tolunay 2003). Therefore, great variations in ARW are possible not only due to the growth conditions and region but also by the periods. Therefore, ARWs can be in accordance with the literature or not due to lots of affecting factors. However, as can be seen in Table 1, ARW averages of Scotch pine wood ranged from 1.26 to 1.37 mm, which are in accordance with 1.34 mm reported by Büyüksarı et al. (2017) and in the range of 0.79 to 2.6 mm and 0.37 to 3.91 mm reported by Dündar (2005) and Sensuła et al. (2017), respectively. Similar harmonies are valid for black pine such as 0.11 to 4.82 mm (from 1744 to 2011 – Hodul Mountain) (Kantarci et al. 2013), and cedar woods such as 2.45 to 4.11 mm (Akkemik 2003).
According to results shown in Table 1, all longitudinal and transverse wave velocities were decreased when ARNs were increased from five to ten. However, VR of red pine, VRL of black pine, and VRT of red pine were increased when ARN was increased from 10 to 15. However, almost all increased values were below the average of the initial values except VR of red pine. The VR of red pine slightly surpassed (0.1%) the initial value, but it was just 2 m/s and can be regarded as negligible. Also, VRT of red pine and cedar for 10 and 15 rings were almost equal. Therefore, constant but various decreases in all velocities by the increase in ARN were observed only for Scots pine and cedar.
The maximum decrease (-4.0%) in the velocity was seen in the transverse ultrasonic wave of red pine. The changes in longitudinal wave velocities with the increase in ARN ranged from -3.7% (black pine, 15 rings) to +0.1% (red pine, 15 rings). Decreases in VRL ranged from -4% (red pine, 15 rings) to -0.8 (red pine, 10 rings). Decreases in VRT ranged from -3.7% (black pine, 15 rings) to -0.5 (Scots pine, 10 rings). The tendencies of VRL and VRT for red pine and Scots pine were nearly the same. Moreover, velocities decreased more specifically following the ten AR, except VRL for black pine.
Table 1. Average ARW, Density, UWV Values of the Species
The reported VR, VRL, and VRT values for the evaluated species are presented in Table 2. Some of the velocities were in harmony, while others were not. However, as reported by Yılmaz Aydın and Küçükköse (2020), ultrasound propagation is a reliable non-destructive test method for the characterization of wood stiffness.
Table 2. UWV in the Radial Direction
The ANOVA results for ARN are presented in Table 3, and changes in the means of UWV due to an increase in ARN were statistically significant only for VR of black pine and cedar woods. According to Duncan’s multiple range test classifications (Table 4), there were no statistically significant differences between the means of UWV and ARN except VR for cedar and black pine species. Furthermore, the mean values of VR for 10 and 15 rings were statistically the same.
Generally, a high R2 value indicates better prediction of the variable. The values closer to 1 reveal that the association is greater, but do not express the statistical significance. However, according to the results of regression analysis seen through Figs. 3 to 6, a maximum of 44% of the data could be predicted by the ARW.
Fig. 3. The relationship between UWV and ARN or ARW for Scots pine
Table 3. ANOVA Results for ARN
Table 4. Duncan Homogeneity Groups of Ultrasonic Velocities According to ARN
Fig. 4. The relationship between UWV and ARN or ARW for Red pine
Fig. 5. The relationship between UWV and ARN or ARW for Black pine