Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Properties of rotary peeled veneer and laminated veneer lumber (LVL) from New Zealand grown Eucalyptus globoidea

Properties of rotary peeled veneer and laminated veneer lumber (LVL) from New Zealand grown... Background: Eucalyptus species can be alternative plantation species to Pinus radiata D.Don (radiata pine) for New Zealand. One promising high value use for eucalypts is laminated veneer lumber (LVL) due to their fast growth and high stiffness. This study investigated the suitability of Eucalyptus globoidea Blakely for veneer and LVL production. Methods: Twenty-six logs were recovered from nine 30-year-old E. globoidea trees. Growth-strain was measured using the CIRAD method for each log before they were peeled into veneers. Veneer recovery, veneer splitting and wood properties were evaluated and correlated with growth-strain. Laminated veneer lumber (LVL) panels were made from eucalypt veneers only or mixed with radiata pine veneers to investigate the bonding performance of E. globoidea. Results: Veneers with no, or limited, defects can be obtained from E. globoidea. Veneer recovery (54.5%) correlated with growth-strain and was highly variable between logs ranging from 23.6% to 74.5%. Average splitting length in a veneer sheet was 3.01 m. There was a moderate positive association between splitting length and growth-strain (r =0. 73), but no significant association with wood stiffness (r =0.27). Bond quality of LVL panels prepared using E. globoidea veneer and a phenol formaldehyde adhesive did not satisfy AS/NZ 2098.2. Conclusion: Usable veneers for structural products could be obtained from E. globoidea at yields of up to 74.5%, but variation in the existing resource (which has not been genetically improved) was large. In particular, growth-strain reduced veneer recovery by splitting, largely independent of stiffness. The considerable variation in growth-strain and stiffness indicated a possibility for genetic improvement. Furthermore, a technical solution to improve bonding of E. globoidea veneers needs to be developed. Keywords: Growth-strain, Bonding, Splitting, Stiffness, LVL Background have been investigated previously for use in LVL. In gen- Eucalyptus species are hardwoods and make up 26% of eral, good veneer qualities (Acevedo et al. 2012), satisfac- the global forest plantation estate (FSC 2012). Plantation tory mechanical properties (de Carvalho et al. 2004; eucalypt species can grow fast, reaching up to 30 cm at Palma and Ballarin 2011) and no major gluing problems the base in 8 years (de Carvalho et al. 2004), and are were reported for eucalypt resources with air-dry dens- currently mostly grown for chip wood to supply the pulp ities less than 650 kg/m (Hague 2013, Ozarska 1999). & paper industry. However, eucalypt timber is generally A major obstacle to using eucalypts for veneers and of higher stiffness than that of most softwood species, LVL is the high level of growth-stresses present in the the main plantation resource for solid-wood processing. logs. These growth-stresses are generated by the newly High stiffness is beneficial for products used in struc- formed wood cells. The exact molecular mechanism by tural applications, such as in laminated veneer lumber which the cell walls generate such large stresses is (LVL) (Bal and Bektaş 2012). Plantation-grown eucalypts unknown (Alméras and Clair 2016; Okuyama et al. 1994; Toba et al. 2013; Yang et al. 2005). However, the newly formed cells tend to contract longitudinally and expand * Correspondence: clemens.altaner@canterbury.ac.nz New Zealand School of Forestry, College of Engineering, University of Canterbury, Private Bag 4800, Christchurch, New Zealand © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Guo and Altaner New Zealand Journal of Forestry Science (2018) 48:3 Page 2 of 10 transversely during cell wall maturation. As a conse- None of the currently commercially grown eucalypts quence, the centre of the stem is under axial compres- produce naturally ground-durable and coloured timber sion while the outside is under axial tension (Kubler even though the value of such a resource was identified 1987). These growth-stresses are released when cutting many years ago by early eucalypt enthusiasts (McWhan- into the stem i.e. during felling, sawing or veneer peel- nell 1960; Simmonds 1927). Interest in growing these ing. The release of growth-stresses can lead to severe eucalypt species to produce high-value speciality timbers end-splitting following a crosscut, board distortion continued in the forestry sector but smaller growers during sawing and breakage of veneers in the peeling favoured different species so no critical mass has been process (Archer 1987; Jacobs 1945; Yang and Waugh achieved to date. Furthermore, a successful plantation in- 2001). These defects are more prominent in smaller dustry needs to be supported by a breeding programme diameter logs, i.e. a plantation resource. Splitting of (Miller et al. 2000). Tree-breeding programmes require a veneers caused by growth-stress lowers veneer quality wide genetic basis and are costly, highlighting the need to and reduces yield. For example, only 20% usable veneers focus resources on a few species. were recovered from E. grandis W.Hill due to severe Three major research initiatives involving durable end-splitting (Margadant 1981). To date, no techno- eucalypts in New Zealand have been initiated in the last logical solution to reduce the effects of growth-stresses two decades. The Forest Research Institute (Scion) and has been implemented successfully. the New Zealand Forestry Association undertook a Unlike the global plantation estate, eucalypts and other series of trials on eucalypts with stringy bark. However, hardwood species account for only 2% of the New these were either discontinued due to a lack of funding Zealand plantation area, which is dominated by Pinus or have a narrow genetic base (van Ballekom and Millen radiata D.Don (radiata pine) (90%) (MPI 2016). Interest 2017). The New Zealand Dryland Forests Initiative in establishing commercial eucalypt plantations dates (NZDFI) has been working since 2008 to establish a eu- back to the late nineteenth century with the introduction calypt forest industry producing naturally durable timber and testing of many eucalypt species around that time based on a large scale-breeding programme of three spe- (Barr 1996; McWhannell 1960; Miller et al. 1992; Miller cies E. bosistoana F.Muell., E. quadrangulata H. Deane et al. 2000; Shelbourne et al. 2002; Simmonds 1927). & Maiden and E. globoidea (Millen 2009). This breeding Their work identified various Eucalyptus species that programme took a range of wood-quality traits into ac- suit New Zealand conditions. However, today E. nitens count (including low growth-stress). While primarily (H.Deane & Maiden) Maiden is the only Eucalyptus spe- chosen for the natural durability of their heartwood, cies that is currently grown commercially on a large these species also produce wood of high stiffness - up to scale. There are more than 10,000 ha E. nitens in South- 20 GPa (Bootle 2005). Demand for engineered timber land and Otago (in the southern South Island), but the products with exceptional stiffness has been generated species suffers from fungal and insect attack in the by the emergence of high-rise timber buildings (Van warmer North Island (McKenzie et al. 2003; Miller et al. de Kuilen et al. 2011). These species also have naturally 1992). Some small commercial plantings of E. fastigata durable heartwood so it may be possible to produce H. Deane & Maiden and a small amount of E. regnans preservative-free durable LVL (McKenzie 1993; Page and F.Muell. can also be found (Miller et al. 2000). The Singh 2014). Some information on the wood properties development of these three species is supported by of E. bosistoana, E. quadrangulata and E. globoidea is breeding programmes: E. nitens (Telfer et al. 2015); E. available from old-growth resources in Australia (Bootle fastigata (Kennedy et al. 2011); and E. regnans (Suon- 2005), but only young plantation-grown E. globoidea has tama et al. 2015). Eucalyptus nitens is currently grown been studied previously in New Zealand. Eucalyptus glo- for chip wood export for the pulp industry. Generally, it boidea has been reported to be well suited for plantation is possible to manufacture quality LVL from 15-year old forestry with good tree health, growth and adaptability E. nitens, which was reported to have an average MoE of combined with favourable timber properties of good 14.3 GPa and achieving F17 grade according to AS/NZS stiffness and natural durability (Barr 1996; Haslett 1990; 2269 (2012) (Gaunt et al. 2003). This compares Millner 2006). Additionally, it is easy to dry and has favourably to the majority of radiata pine LVL prod- relatively low growth-stress levels (Jones et al. 2010; ucts manufactured in New Zealand, the MoE of Poynton 1979). No information on peeling parameters, which range between 8 and 13 GPa, and which relies veneer drying or bonding has been reported for this on using the better part of the radiata pine resource. species, however. Apart from growth-stress, E. nitens has been reported Eucalyptus globoidea was selected for the present to suffer from collapse and internal checking during study to evaluate its suitability for veneer and LVL pro- drying (Lausberg et al. 1995; McKenzie et al. 2003; duction considering the fact that sufficiently large trees McKinley et al. 2002). could be sourced from a farm-forestry operation. To the Guo and Altaner New Zealand Journal of Forestry Science (2018) 48:3 Page 3 of 10 best of our knowledge, no sufficiently large E. bosistoana core diameter of 82 mm using a spindled lathe (Raute, or E. quadrangulata trees are available in New Zealand Finland) with three-stage chucks. The remaining three for processing research. Growth-strain of logs was logs fell off the lathe with a larger peeler core (133 to measured and then peeled into veneers. Green veneer 162 mm in diameter) due to severe end-splitting. Veneers recovery and peeling quality were evaluated and rela- with a thickness of 3.74 mm were produced from all 26 tionships between these attributes and both growth- logs. During the peeling process, the recovery and vol- strain and dynamic modulus of elasticity (MoE) were umes of different types of waste (core, round-up, spur, investigated. Physical properties including density, clipper defects) were recorded for each log. After clipping, shrinkage and moisture content of dried veneer were 296 veneer sheets were obtained. Recovery was defined as also monitored. E. globoidea veneers were used to manu- the ratio of the obtained veneer volume to log volume. facture pure eucalypts LVL and mixed LVL with radiata Veneer volume was calculated based on the number of pine veneers to investigate the bonding performances. sheets and the thickness and dimensions of sheets. For comparison, NPI provided the average recovery data of Methods more than 46,000 radiata pine logs (2.7 m long) collected Nine E. globoidea trees with straight form were ran- previously from the production line. The radiata pine logs domly chosen and felled from a 30-year-old stand in the used in this study were sourced from plantations and lower North Island (latitude 40° 11′ 12" S, longitude woodlots in Nelson and Marlborough. Veneer splitting 175° 20' 35" E, elevation 60 m) in May 2016. The stems was the major defect with few knots or other defects were manually debarked immediately after felling. From found so the aggregate defect rule according to AS/NZS these stems, 26 suitable logs for peeling of 2.7 m length 2269.0 (2012) was not applied. The green veneer sheets were recovered. The small end (SED) and large end were visually graded to four classes (face, core, composer, diameters (LED) were measured for each log in order to waste) according to their splitting severity by the technical calculate log volume. manager in NPI. Only face and core grades were consid- ered to be usable veneer. Growth-strain measurement From the dryer line with the Metriguard 2655 DFX For each log, the amount of growth-strain was deter- instrument (USA), ultrasonic propagation velocity and mined with the CIRAD method (Gerard et al. 1995). width data were made available for each veneer sheet. The growth-strain is variable on the surface of a stem The individual veneer sheets could be tracked back to (Gerard et al. 1995). Therefore, growth-strain was mea- the individual logs. With the clipping width known, the sured on four positions at ~1.35 m spaced by ~90 width data were used to calculate the tangential shrink- around the circumference of each 2.7 m log. The four age. The number of splits in each veneer was counted assessments for each log were averaged. Measuring from the Novascan grader (Grenzebach, Germany) points were chosen in straight-grained areas in close images. ImageJ software (National Institute of Health, proximity to the above described positions in order to USA) was used to analyse the splitting lengths. avoid knots. It was calculated from the measured change After drying, a strip approximately 200-300 mm wide in distance between the pins according to Eq. (1). was taken from a sheet near the start and the end of α = −φδ (1). each veneer mat. The dimension and weight of this strip where α is the strain in microstrain; δ is the measured were measured to obtain dried density. These test pieces displacement in μm and φ is a constant dependent on were then dried further in an oven at 103 °C until tree species. The published value for eucalypt of φ = 11.6 constant weight to obtain the moisture content. was used (Fournier et al. 1994). The dynamic modulus of elasticity (MoE) was calcu- lated from ultrasound acoustic propagation velocity and Rotary peeling and veneer evaluation wood density data according to Eq. (2). The trees were transported to Nelson Pine Industries E= V ρ (2). Ltd. (NPI), Richmond, NZ for processing. All logs were Where V is the ultrasonic velocity, E is the dynamic heated at 85 °C for 24 h in a water bath before peeling MoE and ρ is the density. Static MoE for the population 8 days after felling. 23 of the 26 logs were peeled to a of sheets was estimated by multiplying the dynamic Table 1 Green veneer recovery and amount of waste of E. globoidea compared to P. radiata data Green veneer Amount of waste recovery (%) Round-up (%) Spur (%) Core (%) Clipper defects (%) E. globoidea 54.5 4.6 2.5 12.0 20.0 P. radiata 69.8 11.2 2.7 6.0 8.3 Guo and Altaner New Zealand Journal of Forestry Science (2018) 48:3 Page 4 of 10 Table 2 Summary of veneer recovery and splitting Recovery (%) Useable veneer (%) Growth-strain (με) Splitting length (m) Split counts Mean 54.5 33.4 839.4 3.01 8.63 SD 14.2 23.7 181.7 2.57 5.11 Min 23.6 0.0 553.9 0.15 1.14 Max 74.5 74.5 1136.8 8.66 16.86 SD, Min and Max represent standard deviation, the minimum and maximum values respectively MoE with a factor of 0.868. The factor was empirically phenolic formaldehyde adhesive manufactured by Aica determined by laminating test panels of Eucalyptus globo- NZ Limited was used at a rate of 180 g/m .Panelswere diea veneers with known dynamic MoE and conducting hot-pressed at 160 °C with a pressure of 1.2 MPa. static 4-point bending tests in the edgewise direction. The quality of the glueline was assessed according to AS/ NZS 2098.2 (2012), which measures the percentage of area Bonding quality of eucalyptus veneer covered by wood after two veneers have been split apart. Ten laboratory-scale 10-ply LVL panels were manufac- According to AS/NZS 2269.0 (2012), bonding between the tured. Six panels were made of eucalypt veneer only, plies in LVL shall be a Type A bond. This specification re- choosing veneer sheets of defined MoE grades based on quires a phenolic adhesive complying with AS/NZS their dynamic MoE values. One panel each was made 2754.1 (2016) and also a bond quality of any single glueline from 12, 14, or 17.5 GPa sheets and three panels were not less than 2 and an average of all gluelines not less than made from 16 GPa sheets. Each LVL panel contained ven- 5 when tested according to AS/NZS 2098.2 (2012). Both a eer from one or two logs only. Another four panels were steam and a vacuum pressure method were used to assess made of five radiata pine and five eucalypt veneer plies. A glueline quality (AS/NZS 2098.2 2012). range of eucalypt grades were used in these panels. The first (14 GPa), third (17 GPa), sixth (14 GPa), eighth (12 GPa) and tenth layers (12 GPa) were eucalypt veneers Results and discussion and the rest were radiata pine. Two panels were made Rotary peeling and veneer recovery from each of G2 and G4 radiata-pine grades. A typical For the 26 logs of 2.7 m tested, the small end diameter averaged 34.4 cm with a standard deviation of 4.3 cm while the large end diameter averaged 38.9 cm with a standard deviation of 6.3 cm. The average diameter of the E. globoidea logs (36.3 cm) was comparable to the radiata pine logs (34.9 cm) used in the plant for LVL pro- duction. In a preliminary test, an additional E. globoidea log was peeled cold. This was unsuccessful and, therefore, preheating to soften the wood was deemed necessary. The average veneer recovery for the 26 E. globoidea logs (54.5%) was lower than for radiata pine logs (69.8%) Fig. 1 Face grade veneer with no splitting (top) and composer Fig. 2 Dependence of usable veneer conversions on growth-strain grade veneer with severe splitting (bottom) of the individual E. globoidea logs Guo and Altaner New Zealand Journal of Forestry Science (2018) 48:3 Page 5 of 10 Fig. 3 Association between splitting length and growth-strain (left) as well as number of splits (right) but the best E. globoidea log had a recovery of 74.5% Growth-strain and veneer splitting (Tables 1 and 2). This result was mainly due to a 6% High quality veneers can be obtained from E. globoidea higher loss in the peeler core caused by severe splits in although splitting significantly degraded the visual the supplied E. globoidea logs and an 11.7% greater clip- appearance of many veneers (55% of the veneers had per loss caused by end-splitting of the veneers compared splitting lengths longer than 2 m). Veneers with no and with the radiata pine logs. However, the amount of severe splitting are shown in Fig. 1. round-up waste was lower for the eucalypt logs than the The average recovery of useable veneer (face and core radiata pine logs, which indicated a better log form for grades) from E. globoidea was 33.4%. Severe splitting E. globoidea. Veneer recovery from individual logs was caused by growth-stress contributed to the low recovery. highly variable ranging from 23.6% to 74.5% (Table 2). A The average growth-strain was 839.4 με measured by the previous peeling study with E. nitens reported an overall CIRAD method. The log with the lowest growth-strain recovery of 59% (McKenzie et al. 2003). However, part- had approximately half the growth-strain compared with sheets were included in that study while only full sheets that with the highest. Usable veneer recovery was nega- were calculated in the present study. tively associated with growth-strain (Fig. 2). The average A spindled lathe was used to peel the logs in the useable veneer recovery for the logs in the bottom quar- current study but spindleless lathes are extensively used tile (growth-strain >965.7 με) was 5.8%, while that for in China. They are suitable for rotary peeling smaller the top quartile (growth-strain <701.8 με) was 57.2%. diameter logs from young and fast-grown hardwood The mean splitting length per veneer for individual plantations (McGavin 2016). Spindleless lathes achieve logs ranged from 0.15 m to 8.66 m. For a veneer 2.65 m higher yields because they can peel logs to smaller peeler in length and 1.26 m in width, the average total splitting core diameters compared to spindled lathes currently length was 3.01 m. This suggested splitting was limiting used for radiata pine. Moreover, spindleless lathes can veneer quality. control splits by pressing the splits together during peel- The splitting length was measured after the veneers ing. Therefore, using this type of lathe may generate were dried. Drying was likely to exaggerate the splitting higher recovery and quality of veneer sheets. lengths but was assumed to affect all veneers equally in this study. In addition, rough handling and peeling set- tings can also contribute to the splitting of veneers. It was assumed that all veneers were equally affected in these ways and the differences among them were mainly caused by growth-stress. A positive correlation (r = 0.73) was observed between splitting lengths in veneers and growth-strain of corre- sponding logs (Fig. 3). In this study, average splitting length was low when growth-strain was less than ~800 με (CIRAD). Longitudinal growth-strain was reported to be positively related to end-splitting of E. nitens and E. globulus logs (Valencia et al. 2011; Yang and Pongracic 2004). The number of splits in a veneer sheet is another measure to evaluate veneer spitting. A strong linear relation- Fig. 4 Splitting length for veneers obtained from the centre to the ship (r = 0.91) was obtained between splitting length and outside of five individual logs split numbers (Fig. 3). Considering that the measurement of Guo and Altaner New Zealand Journal of Forestry Science (2018) 48:3 Page 6 of 10 Table 3 Physical and mechanical properties of dried E. globoidea veneers Dried density (kg/m ) Moisture content (%) Shrinkage (%) Velocity (km/s) Dynamic MoE (GPa) Mean 688.13 7.31 9.85 4.657 15.14 SD 68.55 1.09 0.77 0.240 2.05 Min 557.41 5.51 8.46 4.322 11.04 Max 824.00 9.32 11.31 5.097 19.51 the number of splits is less time consuming than quantifying logs. Furthermore, the increasing curvature with de- splitting length, it might be a better option for future studies. creasing radius can facilitate splitting. However, the vari- This result indicated that higher veneer recovery and ation between the stems was much bigger than the quality would be possible if growth-stresses were radial effect, with the veneer splitting independent of reduced. Methods to control the growth-stress and min- radius for the worst and the best logs. imise splitting are difficult to perform, have not been It is worth noting that the log preheating and peeling successful in industrial applications and incur ongoing settings in this study were optimised for radiata pine. costs (Archer 1987; Malan 1995; Yang and Waugh Acevedo et al. (2012) reported that better quality veneers 2001). Growth-stresses are heritable and selecting trees were obtained from E. nitens by adjusting nose bar pres- with low growth-stress in a breeding programme can sure and peeling knife angle. potentially solve this problem for a future resource such as E. globoidea (Davies et al. 2017; Malan 1995). For Physical and mechanical properties of veneer existing eucalypt plantations grown for the lower-value After drying, the average dry density was 668 kg/m and the wood chips, such as E. nitens, segregation would be an average moisture content was 7.3% (Table 3). No excessively option. However, current methods of measuring growth- high or low moisture contents were found, which indicated stress are time consuming and cumbersome, making this homogeneous drying of the E. globoidea veneers. approach impractical (Yang and Waugh 2001). Rapid The average shrinkage of the E. globoidea veneers was and non-destructive segregation methods need to be 9.9% tangentially and varied between 8.5 and 11.3%. developed. For example, Yang et al. (2006) measured Most veneers were heartwood as the sapwood in E. growth-strain of 10-year-old E. globulus and found cor- globoidea is very narrow. For comparison, typical relations with cellulose crystallite width measured using tangential shrinkage values for radiata pine veneers are a SilviScan-2 instrument. 6.4% for sapwood and 4.4% for heartwood. Higher The relationship between splitting length and radial shrinkage will result in greater volume loss. It should be position for five logs is shown in Fig. 4. The splitting noted that, within species, heartwood typically displays length of the veneer sheets tended to increase towards lower shrinkage than sapwood. the centre of the stem (position 0). The decreasing The average dynamic MoE calculated for the E. globoi- circumference with decreasing radius results in shorter dea veneer sheets from Metriguard acoustic velocity and tangential distances between the radial end-splits of the interpolated lab density was 14.67 GPa ranging from 9.59 to 20.26 GPa (Fig. 5). The equivalent static MoE was estimated to be 12.73 GPa based on the empirical conversion equation. Common LVL products manufac- tured from radiata pine range from 8 to 11 GPa. Jones et al. (2010) investigated 25-year-old E. globoidea for high- quality solid wood production. Boards from the butt logs were reported to have an average density of 655 kg/m , dynamic MoE of 13 GPa and static MoE of 12 GPa. Ac- cording to Haslett (1990), the timber of E. globoidea (over 25 years old) in New Zealand has an MoE of 14.6 GPa at a moisture content of 12%. High stiffness wood tended to have higher growth- strain (Fig. 6). This is an unfavourable association as stiff wood with low growth-strain is desirable. However, the association between MoE and growth-strain was moder- ate (r = 0.65) implying the existence of stiff logs which Fig. 5 Cumulative distribution of dynamic and static MoE of are low in growth-strain. Several logs produced veneers veneer sheets with MoEs above 15 GPa and growth-strain levels below Guo and Altaner New Zealand Journal of Forestry Science (2018) 48:3 Page 7 of 10 Fig. 6 Association between MoE and growth-strain (left) as well as veneer splitting (right) 800 με. More importantly no association (r = 0.27) was was large with the average stiffness ranging from found between veneer splitting and stiffness, which 12.1 GPa to 18.0 GPa. Analysis of variance found signifi- demonstrated that peeling quality needs to be improved cant differences (P < 0.001) in MoE values of veneers through reducing growth-stress rather than through from different trees. MoE. The stiffest logs yielded veneers with an MoE of It has to be noted that the tested E. globoidea was up to 19.5 GPa, but these did not necessarily have a severe genetically unimproved material of unknown proven- splitting problem. Therefore, it seems possible to obtain a ance. Wood properties like growth-stress and MoE are stiff eucalypt resource that yields high-quality veneers. under genetic control (Davies et al. 2017). Murphy et al. A weak positive association (R = 0.26) between CIRAD (2005) reported a heritability of 0.3 to 0.5 for growth- longitudinal displacement and MoE has been found previ- strain in Eucalyptus dunnii Maiden and indicated tree ously in E. globulus (Yang et al. 2006). With the increase breeding can be an effective method to reduce growth- of longitudinal displacement, microfibril angle tended to stress. Considerable variation among trees was observed, decline while density increased. Similar associations be- indicating a potential for genetic improvement. The rela- tween growth-strain and wood properties have been re- tively high acoustic velocity of eucalypts in the corewood ported for wood from Populus deltoides Bartr (Fang et al. could allow peeling veneers to a smaller peeler core with 2008). However, no statistically significant associations spindleless lathes, improving yields and allowing the use were found between growth-stain and dynamic MoE or of a small diameter younger resource. density for E. nitens (Chauhan and Walker 2004). The average distribution of MoEs for the veneers Bonding quality obtained from the nine trees assessed is shown in Fig. 7. The bond test revealed poor bonding of the plies. None As for veneer splitting, the variance in MoE among trees of the panels made with E. globoidea alone passed the specifications for structural LVL (Table 4). Density seemed to exaggerate the bonding difficulty for the 100% E. globoidea LVL. Eucalyptus globoidea panels with densities higher than 800 kg/m had average bonding qualities lower than 3. Alternating E. globoidea and P. radiata veneers improved bond quality, and all samples Table 4 Bond tests of six LVL panels made from E. globoidea veneers (listed in order of increasing density) Grade Density Steam test Immersion test (GPa) (g/cm ) EE EE EE EE EE EE min max mean min max mean 12 640.51 1 9 5 1 5 2 16 696.92 3 8 6 4 9 6 16 702.52 1 9 6 3 8 7 16 806.83 0 5 1 0 3 2 Fig. 7 Box-and-whisker plots of veneer MoE from nine trees. Dotted 14 809.22 0 3 2 2 3 2 lines show minimum and maximum values; the thick black line 17.5 860.05 1 4 2 0 7 3 shows the median; the box represents the upper and lower EE represents bond quality values between E. globoidea veneers (0 no bond – quartiles and outliers are marked using open circles 10 excellent bond). The minimum, maximum and mean values are shown Guo and Altaner New Zealand Journal of Forestry Science (2018) 48:3 Page 8 of 10 Table 5 Bond tests of four LVL panels made from a mixture of E. globoidea and P. radiata veneers Sample Radiata pine grade Steam test Immersion test ER ER RR Mean ER ER RR Mean min max min max 1 G4 4 99 7 1 99 7 2G4 3 7 10 5 0 9 8 6 3 G2 2 99 7 3 99 6 4 G2 2 99 6 1 89 6 G4 and G2 are radiata pine veneer grades; 12, 14 and 17.5 GPa grade eucalypt veneers were used in each sample. ER represents the gluelines between eucalypts and radiata pine veneers and RR stands for the glueline between two radiata pine veneers. Mean is the average value of all 9 gluelines passed the steam test, but only one sample passed the between splitting length and stiffness. It should be immersion test. The glueline between radiata pine plies possible to find stiff logs, however, which would was excellent (Table 5). yield satisfactory veneers. The considerable variation Eucalypts have a higher density and extractive content in stiffness observed here indicated a potential for than radiata pine, and both these factors can make glu- genetic improvement. ing difficult. In Australia, difficulty in bonding veneers 6) The lack of association between stiffness and from dense eucalypts (air-dry density above 700 kg/cm ) splitting length suggested that quick acoustic are known and some special adhesive formulations have measurements for segregating eucalypt logs suitable been developed for these species (Carrick and Mathieu peeling are unlikely to be successful. 2005; Ozarska 1999). It is commonly stated that low and 7) Bond performance of E. globoidea LVL was poor and medium density (below 650 kg/cm ) eucalypts glue well did not meet the New Zealand standard. Alternating but young E. nitens was found to have bonding issues E. globoidea and P. radiata veneers improved bond however (de Carvalho et al. 2004, Farrell et al. 2011, quality, but bonding of E. globoidea veneers still Hague 2013). Plywood products made from various needs to be addressed. Eucalyptus species have been reported in China, Acknowledgements Malaysia, Uruguay and Brazil indicating that satisfactory We would like to thank Richard Barry from Nelson Pine Industries for bonding can be achieved for many Eucalyptus species overseeing the veneer peeling and LVL testing as well as for his contributions to the manuscript. Paul Millen's coordination in harvesting the (de Carvalho et al. 2004; Hague 2013; Turnbull 2007; Yu trees is also appreciated. et al. 2006). Therefore, it seems probable that a technical solution for gluing E. globoidea can be found. Funding This research project was financially supported by the Ministry of Primary Industries’ Sustainable Farming Fund (SFF407602). FG would also like to Conclusions thank the Chinese Scholarship Council (CSC) for supporting his study. Authors’ contributions 1) Veneer recovery ranged from 23.6% to 74.5% among FG carried out the field work, growth-strain measurement, veneer peeling, 26 E. globoidea logs. Larger peeler core and higher data analysis and prepared the manuscript. CMA designed and organised clipper losses occurred compared to radiata pine. the study contributed to data analysis and manuscript preparation. Both au- thors read and approved the final manuscript. 2) High-quality veneers could be obtained from E. globoidea. Low growth-strain logs produced more Competing interests usable veneers. Splitting degraded veneer quality. The authors declare that they have no competing interests. 3) Splitting length in veneers was correlated to growth- strain of the corresponding logs. Split number was Publisher’sNote Springer Nature remains neutral with regard to jurisdictional claims in strongly associated with splitting length and can be published maps and institutional affiliations. used to evaluate veneer splitting severity. Veneers from inner wood tended to a have longer splitting Received: 24 May 2017 Accepted: 3 January 2018 lengths; however differences among logs were more pronounced. References 4) Moisture contents of dried veneers indicated good Acevedo, A, Bustos, C, Lasserre, JP, Gacitua, W. (2012). Nose bar pressure effect in the lathe check morphology to Eucalyptus nitens veneers. Maderas. Ciencia y drying of E. globoidea. The average tangential tecnologia, 14(3), 289–301. shrinkage was 9.9% and the volume loss was higher Alméras, T, & Clair, B. (2016). Critical review on the mechanisms of maturation stress than for radiata pine. Larger shrinkage is a common generation in trees. Journal of the Royal Society Interface, 13(122), 20160550. Archer, RR (1987). Growth stresses and strains in trees. Berlin: Springer-Verlag Gmbh. feature of high-density timbers. AS/NZS 2098.2 (2012). Methods of test for veneer and plywood. Method 2: Bond 5) The average dynamic MoE was 14.67 GPa ranging quality of plywood (chisel test). Wellington: Standards Australia/ Standards from 9.59 to 20.26 GPa. No association was found New Zealand. Guo and Altaner New Zealand Journal of Forestry Science (2018) 48:3 Page 9 of 10 AS/NZS 2269.0 (2012). Plywood-structural. Part 0: Specifications. Wellington: McKinley, R, Shelbourne, C, Low, C, Penellum, B, Kimberley, M. (2002). Wood Standards Australia/ Standards New Zealand. properties of young Eucalyptus nitens, E. globulus,and E. maidenii in northland, AS/NZS 2754.1 (2016). Adhesives for timber and timber products. Part 1: Adhesives New Zealand. New Zealand Journal of Forestry Science, 32(3), 334–356. for manufacture of plywood and laminated veneer lumber (LVL). Wellington: McWhannell, F. B. (1960). Eucalypts for New Zealand farms, parks and gardens. Standards Australia/ Standards New Zealand. Hamilton, N.Z: Paul's Book Arcade. Bal, BC, & Bektaş, I. (2012). The effects of some factors on the impact bending Millen, P (2009). NZ Dryland forests initiative: A market focused durable eucalypt strength of laminated veneer lumber. BioResources, 7(4), 5855–5863. R&D project. In L Apiolaza, S Chauhan, J Walker (Eds.), Revisiting eucalypts 2009 : Workshop proceedings, (pp. 57–74). Christchurch: Wood Technology Barr, N. (1996). Growing eucalypt trees for milling on New Zealand farms. Wellington, New Zealand: New Zealand Farm Forestry Association. Research Centre, Univeristy of Canterbury. Bootle, KR (2005). Wood in Australia. Types, properties and uses, (2nd ed., ). Miller, JT, Cannon, P, Ecroyd, CE (1992). Introduced forest trees in New Zealand: Sydney: McGraw-Hill Book Company. Recognition, role, and seed source. 11. Eucalyptus nitens (Deane et maiden) Carrick, J, & Mathieu, K (2005). Durability of laminated veneer lumber made from maiden. (FRI Bulletin 124). Rotorua: New Zeland Forest Research Institute. Blackbutt. In Proceedings of the 10DBMC international conference on durability Miller, JT, Hay, AE, Ecroyd, CE (2000). Introduced forest trees in New Zealand: of building materials and components, France. Recognition, role, and seed source. 18. Ash eucalypts. (FRI Bulletin 124). Rotorua: Chauhan, SS, & Walker, J. (2004). Relationships between longitudinal growth New Zealand Forest Research Institute. strain and some wood properties in Eucalyptus nitens. Australian Forestry, Millner, J. (2006). The performance of eucalyptus species in hill country. Palmerston 67(4), 254–260. North, New Zealand: Massey University. Davies, NT, Apiolaza, LA, Sharma, M. (2017). Heritability of growth strain in Eucalyptus MPI (2016). National exotic forest description (NEFD). Wellington: Ministry for bosistoana: A Bayesian approach with left-censored data. New Zealand Journal Primary Industries. of Forestry Science, 47, 5. https://doi.org/10.1186/s40490-017-0086-2. Murphy, TN, Henson, M, Vanclay, JK. (2005). Growth stress in Eucalyptus dunnii. De Carvalho, AM, Lahr, FAR, Bortoletto Jr, G. (2004). Use of Brazilian eucalyptus to Australian Forestry, 68(2), 144–149. https://doi.org/10.1080/00049158.2005. produce LVL panels. Forest Products Journal, 54(11), 61. 10674958. Fang, C-H, Guibal, D, Clair, B, Gril, J, Liu, Y-M, Liu, S-Q. (2008). Relationships Okuyama, T, Yamamoto, H, Yoshida, M, Hattori, Y, Archer, R. (1994). Growth between growth stress and wood properties in poplar I-69 (Populus stresses in tension wood: Role of microfibrils and lignification. Annales des Deltoides Bartr. Cv. “Lux” ex I-69/55). Annals of Forest Science, 65(3), 307. Sciences Forestières, 51(3), 291–300. Farrell, R, Blum, S, Williams, D, Blackburn, D (2011). The potential to recover higher Ozarska, B. (1999). A review of the utilisation of hardwoods for LVL. Wood value veneer products from fibre managed plantation eucalypts and broaden Science and Technology, 33(4), 341–351. https://doi.org/10.1007/ market opportunities for this resource: Part A. (project no: PNB139-0809). s002260050120. Melbourne: Forest & Wood Products Australia. Page, D, & Singh, T. (2014). Durability of New Zealand grown timbers. New Fournier, M, Chanson, B, Thibaut, B, Guitard, D. (1994). Measurements of residual Zealand Journal of Forestry, 58(4), 26–30. growth strains at the stem surface. Observations on different species. Annales Palma, HAL, & Ballarin, AW. (2011). Physical and mechanical properties of LVL des Sciences Forestières, 51(3), 249–266. panels made from Eucalyptus grandis. Ciência Florestal, 21(3), 559–566. https:// FSC (2012). Strategic review on the future of forest plantations. Helsinki, Finland: doi.org/10.5902/198050983813. Forest Stewardship Council. Poynton, RJ (1979). Eucalyptus globoidea Blakely. In RJ Poynton (Ed.), Tree planting in Gaunt, D, Penellum, B, McKenzie, HM. (2003). Eucalyptus nitens laminated veneer southern Africa: The eucalypts, (pp. 316–324).South Africa:Department of Forestry. lumber structural properties. New Zealand Journal of Forestry Science, 33(1), Shelbourne, C, Bulloch, B, Low, C, McConnochie, R. (2002). Performance to 114–125. age 22 years of 49 eucalypts in the Wairarapa district, New Zealand, Gerard, J, Bailleres, H, Fournier, M, Thibaut, B. (1995). Wood quality in plantation reviewed with results from other trials. New Zealand Journal of Forestry eucalyptus - a study of variation in three reference properties. Bois et Forêts Science, 32(2), 256–278. des Tropiques, 245, 101–110. Simmonds, JH (1927). Trees from other lands for shelter and timber in New Hague, J. R. B. (2013). Utilisation of plantation eucalypts in engineered wood Zealand: Eucalypts. Auckland: Brett Printing and Publishing Company. products. Melbourne, Australia. Suontama, M, Low, CB, Stovold, GT, Miller, MA, Fleet, KR, Li, Y, Dungey, HS. (2015). Haslett, A. N. (1990). Properties and utilisation of exotic speciality timbers grown in Genetic parameters and genetic gains across three breeding cycles for New Zealand. Part VI: Eastern blue gums and stringy barks. Eucalyptus growth and form traits of Eucalyptus regnans in New Zealand. Tree Genetics & botryoides Sm. Eucalyptus saligna Sm. Eucalyptus globoidea Blakey. Eucalyptus Genomes, 11(6), 133. https://doi.org/10.1007/s11295-015-0957-8. muellerana Howitt. Eucalyptus pilularis Sm. (FRI Bulletin 119). Rotorua: New Telfer, EJ, Stovold, GT, Li, YJ, Silva, OB, Grattapaglia, DG, Dungey, HS. (2015). Zealand Forest Research Institute. Parentage reconstruction in Eucalyptus nitens using SNPs and microsatellite Jacobs, MR (1945). The growth stresses of woody stems. Canberra: Commonwealth markers: A comparative analysis of marker data power and robustness. PLoS Forestry Bureau. One, 10(7), e0130601. https://doi.org/10.1371/journal.pone.0130601. Jones, TG, McConnochie, RM, Shelbourne, T, Low, CB. (2010). Sawing and grade Toba, K, Yamamoto, H, Yoshida, M. (2013). Micromechanical detection of growth recovery of 25-year-old Eucalyptus fastigata, E. globoidea, E. muelleriana and E. stress in wood cell wall by wide-angle X-ray diffraction (WAX). Holzforschung, pilularis. New Zealand Journal of Forestry Science, 40,19–31. 67(3), 315–323. Kennedy, S, Dungey, H, Yanchuk, A, Low, C. (2011). Eucalyptus fastigata: Its current Turnbull, J (2007). Development of sustainable forestry plantations in China: A status in New Zealand and breeding objectives for the future. Silvae Genetica, review. (impact assessment series report no. 45). Canberra: Australian Centre for 60(6), 259–266. International Agricultural Research. Kubler, H. (1987). Growth stresses in trees and related wood properties. Forestry Valencia, J, Harwood, C, Washusen, R, Morrow, A, Wood, M, Volker, P. (2011). abstracts, 48(3), 131–189. Longitudinal growth strain as a log and wood quality predictor for Lausberg, M, Gilchrist, K, Skipwith, J. (1995). Wood properties of Eucalyptus nitens plantation-grown Eucalyptus nitens sawlogs. Wood Science and Technology, grown in New Zealand. New Zealand Journal of Forestry Science, 25(2), 147–163. 45(1), 15–34. https://doi.org/10.1007/s00226-010-0302-1. Malan, FS. (1995). Eucalyptus improvement for lumber production. Seminario Van Ballekom, S., & Millen, P. NZDFI: Achievements, constraints and opportunities. Internacional de Utilização da Madeira de Eucalipto, 15,1–19. In CM Altaner, TJ Murray, & J Morgenroth (Eds.), Durable eucalypts on Margadant, R. (1981). Why not peel locally grown eucalyptus. Wood Southern drylands: Protecting and enhancing value, Blenheim, New Zealand, 2017 (pp. Africa, 6(10), 8–22. 1-11): New Zealand School of Forestry McGavin, RL (2016). Analysis of small-log processing to achieve structural veneer Van de Kuilen, JWG, Ceccotti, A, Xia, Z, He, M. (2011). Very tall wooden buildings from juvenile hardwood plantations. Melbourne: The University of Melbourne. with cross laminated timber. Procedia Engineering, 14, 1621–1628. https://doi. org/10.1016/j.proeng.2011.07.204. McKenzie, H. (1993). Growing durable hardwoods - a research strategy. New Zealand Journal of Forestry, 38(3), 25–27. Yang, JL, Baillères, H, Evans, R, Downes, G. (2006). Evaluating growth strain of Eucalyptus McKenzie, H, Turner, J, Shelbourne, C. (2003). Processing young plantation-grown globulus Labill. From SilviScan measurements. Holzforschung, 60(5), 574–579. Eucalyptus nitens for solid-wood products. 1: Individual-tree variation in Yang, JL, Baillères, H, Okuyama, T, Muneri, A, Downes, G. (2005). Measurement quality and recovery of appearance-grade lumber and veneer. New Zealand methods for longitudinal surface strain in trees: A review. Australian Forestry, Journal of Forestry Science, 33(1), 62–78. 68(1), 34–43. https://doi.org/10.1080/00049158.2005.10676224. Guo and Altaner New Zealand Journal of Forestry Science (2018) 48:3 Page 10 of 10 Yang, J, & Pongracic, S (2004). The impact of growth stress on sawn distortion and log end splitting of 32-year old plantation blue gum. (project no: PN03.1312). Vicroria: Forest and Wood Products Research & Development Corporation. Yang, JL, & Waugh, G. (2001). Growth stress, its measurement and effects. Australian Forestry, 64(2), 127–135. Yu, Y, Ren, D, Zhou, Y, Yu, W. (2006). Preliminary study on gluability of E. urophylla veneer. China Forest Products Industry, 33(4), 20–23. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png New Zealand Journal of Forestry Science Springer Journals

Properties of rotary peeled veneer and laminated veneer lumber (LVL) from New Zealand grown Eucalyptus globoidea

Loading next page...
 
/lp/springer-journals/properties-of-rotary-peeled-veneer-and-laminated-veneer-lumber-lvl-F44bQtfxAr
Publisher
Springer Journals
Copyright
2018 The Author(s).
eISSN
1179-5395
DOI
10.1186/s40490-018-0109-7
Publisher site
See Article on Publisher Site

Abstract

Background: Eucalyptus species can be alternative plantation species to Pinus radiata D.Don (radiata pine) for New Zealand. One promising high value use for eucalypts is laminated veneer lumber (LVL) due to their fast growth and high stiffness. This study investigated the suitability of Eucalyptus globoidea Blakely for veneer and LVL production. Methods: Twenty-six logs were recovered from nine 30-year-old E. globoidea trees. Growth-strain was measured using the CIRAD method for each log before they were peeled into veneers. Veneer recovery, veneer splitting and wood properties were evaluated and correlated with growth-strain. Laminated veneer lumber (LVL) panels were made from eucalypt veneers only or mixed with radiata pine veneers to investigate the bonding performance of E. globoidea. Results: Veneers with no, or limited, defects can be obtained from E. globoidea. Veneer recovery (54.5%) correlated with growth-strain and was highly variable between logs ranging from 23.6% to 74.5%. Average splitting length in a veneer sheet was 3.01 m. There was a moderate positive association between splitting length and growth-strain (r =0. 73), but no significant association with wood stiffness (r =0.27). Bond quality of LVL panels prepared using E. globoidea veneer and a phenol formaldehyde adhesive did not satisfy AS/NZ 2098.2. Conclusion: Usable veneers for structural products could be obtained from E. globoidea at yields of up to 74.5%, but variation in the existing resource (which has not been genetically improved) was large. In particular, growth-strain reduced veneer recovery by splitting, largely independent of stiffness. The considerable variation in growth-strain and stiffness indicated a possibility for genetic improvement. Furthermore, a technical solution to improve bonding of E. globoidea veneers needs to be developed. Keywords: Growth-strain, Bonding, Splitting, Stiffness, LVL Background have been investigated previously for use in LVL. In gen- Eucalyptus species are hardwoods and make up 26% of eral, good veneer qualities (Acevedo et al. 2012), satisfac- the global forest plantation estate (FSC 2012). Plantation tory mechanical properties (de Carvalho et al. 2004; eucalypt species can grow fast, reaching up to 30 cm at Palma and Ballarin 2011) and no major gluing problems the base in 8 years (de Carvalho et al. 2004), and are were reported for eucalypt resources with air-dry dens- currently mostly grown for chip wood to supply the pulp ities less than 650 kg/m (Hague 2013, Ozarska 1999). & paper industry. However, eucalypt timber is generally A major obstacle to using eucalypts for veneers and of higher stiffness than that of most softwood species, LVL is the high level of growth-stresses present in the the main plantation resource for solid-wood processing. logs. These growth-stresses are generated by the newly High stiffness is beneficial for products used in struc- formed wood cells. The exact molecular mechanism by tural applications, such as in laminated veneer lumber which the cell walls generate such large stresses is (LVL) (Bal and Bektaş 2012). Plantation-grown eucalypts unknown (Alméras and Clair 2016; Okuyama et al. 1994; Toba et al. 2013; Yang et al. 2005). However, the newly formed cells tend to contract longitudinally and expand * Correspondence: clemens.altaner@canterbury.ac.nz New Zealand School of Forestry, College of Engineering, University of Canterbury, Private Bag 4800, Christchurch, New Zealand © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Guo and Altaner New Zealand Journal of Forestry Science (2018) 48:3 Page 2 of 10 transversely during cell wall maturation. As a conse- None of the currently commercially grown eucalypts quence, the centre of the stem is under axial compres- produce naturally ground-durable and coloured timber sion while the outside is under axial tension (Kubler even though the value of such a resource was identified 1987). These growth-stresses are released when cutting many years ago by early eucalypt enthusiasts (McWhan- into the stem i.e. during felling, sawing or veneer peel- nell 1960; Simmonds 1927). Interest in growing these ing. The release of growth-stresses can lead to severe eucalypt species to produce high-value speciality timbers end-splitting following a crosscut, board distortion continued in the forestry sector but smaller growers during sawing and breakage of veneers in the peeling favoured different species so no critical mass has been process (Archer 1987; Jacobs 1945; Yang and Waugh achieved to date. Furthermore, a successful plantation in- 2001). These defects are more prominent in smaller dustry needs to be supported by a breeding programme diameter logs, i.e. a plantation resource. Splitting of (Miller et al. 2000). Tree-breeding programmes require a veneers caused by growth-stress lowers veneer quality wide genetic basis and are costly, highlighting the need to and reduces yield. For example, only 20% usable veneers focus resources on a few species. were recovered from E. grandis W.Hill due to severe Three major research initiatives involving durable end-splitting (Margadant 1981). To date, no techno- eucalypts in New Zealand have been initiated in the last logical solution to reduce the effects of growth-stresses two decades. The Forest Research Institute (Scion) and has been implemented successfully. the New Zealand Forestry Association undertook a Unlike the global plantation estate, eucalypts and other series of trials on eucalypts with stringy bark. However, hardwood species account for only 2% of the New these were either discontinued due to a lack of funding Zealand plantation area, which is dominated by Pinus or have a narrow genetic base (van Ballekom and Millen radiata D.Don (radiata pine) (90%) (MPI 2016). Interest 2017). The New Zealand Dryland Forests Initiative in establishing commercial eucalypt plantations dates (NZDFI) has been working since 2008 to establish a eu- back to the late nineteenth century with the introduction calypt forest industry producing naturally durable timber and testing of many eucalypt species around that time based on a large scale-breeding programme of three spe- (Barr 1996; McWhannell 1960; Miller et al. 1992; Miller cies E. bosistoana F.Muell., E. quadrangulata H. Deane et al. 2000; Shelbourne et al. 2002; Simmonds 1927). & Maiden and E. globoidea (Millen 2009). This breeding Their work identified various Eucalyptus species that programme took a range of wood-quality traits into ac- suit New Zealand conditions. However, today E. nitens count (including low growth-stress). While primarily (H.Deane & Maiden) Maiden is the only Eucalyptus spe- chosen for the natural durability of their heartwood, cies that is currently grown commercially on a large these species also produce wood of high stiffness - up to scale. There are more than 10,000 ha E. nitens in South- 20 GPa (Bootle 2005). Demand for engineered timber land and Otago (in the southern South Island), but the products with exceptional stiffness has been generated species suffers from fungal and insect attack in the by the emergence of high-rise timber buildings (Van warmer North Island (McKenzie et al. 2003; Miller et al. de Kuilen et al. 2011). These species also have naturally 1992). Some small commercial plantings of E. fastigata durable heartwood so it may be possible to produce H. Deane & Maiden and a small amount of E. regnans preservative-free durable LVL (McKenzie 1993; Page and F.Muell. can also be found (Miller et al. 2000). The Singh 2014). Some information on the wood properties development of these three species is supported by of E. bosistoana, E. quadrangulata and E. globoidea is breeding programmes: E. nitens (Telfer et al. 2015); E. available from old-growth resources in Australia (Bootle fastigata (Kennedy et al. 2011); and E. regnans (Suon- 2005), but only young plantation-grown E. globoidea has tama et al. 2015). Eucalyptus nitens is currently grown been studied previously in New Zealand. Eucalyptus glo- for chip wood export for the pulp industry. Generally, it boidea has been reported to be well suited for plantation is possible to manufacture quality LVL from 15-year old forestry with good tree health, growth and adaptability E. nitens, which was reported to have an average MoE of combined with favourable timber properties of good 14.3 GPa and achieving F17 grade according to AS/NZS stiffness and natural durability (Barr 1996; Haslett 1990; 2269 (2012) (Gaunt et al. 2003). This compares Millner 2006). Additionally, it is easy to dry and has favourably to the majority of radiata pine LVL prod- relatively low growth-stress levels (Jones et al. 2010; ucts manufactured in New Zealand, the MoE of Poynton 1979). No information on peeling parameters, which range between 8 and 13 GPa, and which relies veneer drying or bonding has been reported for this on using the better part of the radiata pine resource. species, however. Apart from growth-stress, E. nitens has been reported Eucalyptus globoidea was selected for the present to suffer from collapse and internal checking during study to evaluate its suitability for veneer and LVL pro- drying (Lausberg et al. 1995; McKenzie et al. 2003; duction considering the fact that sufficiently large trees McKinley et al. 2002). could be sourced from a farm-forestry operation. To the Guo and Altaner New Zealand Journal of Forestry Science (2018) 48:3 Page 3 of 10 best of our knowledge, no sufficiently large E. bosistoana core diameter of 82 mm using a spindled lathe (Raute, or E. quadrangulata trees are available in New Zealand Finland) with three-stage chucks. The remaining three for processing research. Growth-strain of logs was logs fell off the lathe with a larger peeler core (133 to measured and then peeled into veneers. Green veneer 162 mm in diameter) due to severe end-splitting. Veneers recovery and peeling quality were evaluated and rela- with a thickness of 3.74 mm were produced from all 26 tionships between these attributes and both growth- logs. During the peeling process, the recovery and vol- strain and dynamic modulus of elasticity (MoE) were umes of different types of waste (core, round-up, spur, investigated. Physical properties including density, clipper defects) were recorded for each log. After clipping, shrinkage and moisture content of dried veneer were 296 veneer sheets were obtained. Recovery was defined as also monitored. E. globoidea veneers were used to manu- the ratio of the obtained veneer volume to log volume. facture pure eucalypts LVL and mixed LVL with radiata Veneer volume was calculated based on the number of pine veneers to investigate the bonding performances. sheets and the thickness and dimensions of sheets. For comparison, NPI provided the average recovery data of Methods more than 46,000 radiata pine logs (2.7 m long) collected Nine E. globoidea trees with straight form were ran- previously from the production line. The radiata pine logs domly chosen and felled from a 30-year-old stand in the used in this study were sourced from plantations and lower North Island (latitude 40° 11′ 12" S, longitude woodlots in Nelson and Marlborough. Veneer splitting 175° 20' 35" E, elevation 60 m) in May 2016. The stems was the major defect with few knots or other defects were manually debarked immediately after felling. From found so the aggregate defect rule according to AS/NZS these stems, 26 suitable logs for peeling of 2.7 m length 2269.0 (2012) was not applied. The green veneer sheets were recovered. The small end (SED) and large end were visually graded to four classes (face, core, composer, diameters (LED) were measured for each log in order to waste) according to their splitting severity by the technical calculate log volume. manager in NPI. Only face and core grades were consid- ered to be usable veneer. Growth-strain measurement From the dryer line with the Metriguard 2655 DFX For each log, the amount of growth-strain was deter- instrument (USA), ultrasonic propagation velocity and mined with the CIRAD method (Gerard et al. 1995). width data were made available for each veneer sheet. The growth-strain is variable on the surface of a stem The individual veneer sheets could be tracked back to (Gerard et al. 1995). Therefore, growth-strain was mea- the individual logs. With the clipping width known, the sured on four positions at ~1.35 m spaced by ~90 width data were used to calculate the tangential shrink- around the circumference of each 2.7 m log. The four age. The number of splits in each veneer was counted assessments for each log were averaged. Measuring from the Novascan grader (Grenzebach, Germany) points were chosen in straight-grained areas in close images. ImageJ software (National Institute of Health, proximity to the above described positions in order to USA) was used to analyse the splitting lengths. avoid knots. It was calculated from the measured change After drying, a strip approximately 200-300 mm wide in distance between the pins according to Eq. (1). was taken from a sheet near the start and the end of α = −φδ (1). each veneer mat. The dimension and weight of this strip where α is the strain in microstrain; δ is the measured were measured to obtain dried density. These test pieces displacement in μm and φ is a constant dependent on were then dried further in an oven at 103 °C until tree species. The published value for eucalypt of φ = 11.6 constant weight to obtain the moisture content. was used (Fournier et al. 1994). The dynamic modulus of elasticity (MoE) was calcu- lated from ultrasound acoustic propagation velocity and Rotary peeling and veneer evaluation wood density data according to Eq. (2). The trees were transported to Nelson Pine Industries E= V ρ (2). Ltd. (NPI), Richmond, NZ for processing. All logs were Where V is the ultrasonic velocity, E is the dynamic heated at 85 °C for 24 h in a water bath before peeling MoE and ρ is the density. Static MoE for the population 8 days after felling. 23 of the 26 logs were peeled to a of sheets was estimated by multiplying the dynamic Table 1 Green veneer recovery and amount of waste of E. globoidea compared to P. radiata data Green veneer Amount of waste recovery (%) Round-up (%) Spur (%) Core (%) Clipper defects (%) E. globoidea 54.5 4.6 2.5 12.0 20.0 P. radiata 69.8 11.2 2.7 6.0 8.3 Guo and Altaner New Zealand Journal of Forestry Science (2018) 48:3 Page 4 of 10 Table 2 Summary of veneer recovery and splitting Recovery (%) Useable veneer (%) Growth-strain (με) Splitting length (m) Split counts Mean 54.5 33.4 839.4 3.01 8.63 SD 14.2 23.7 181.7 2.57 5.11 Min 23.6 0.0 553.9 0.15 1.14 Max 74.5 74.5 1136.8 8.66 16.86 SD, Min and Max represent standard deviation, the minimum and maximum values respectively MoE with a factor of 0.868. The factor was empirically phenolic formaldehyde adhesive manufactured by Aica determined by laminating test panels of Eucalyptus globo- NZ Limited was used at a rate of 180 g/m .Panelswere diea veneers with known dynamic MoE and conducting hot-pressed at 160 °C with a pressure of 1.2 MPa. static 4-point bending tests in the edgewise direction. The quality of the glueline was assessed according to AS/ NZS 2098.2 (2012), which measures the percentage of area Bonding quality of eucalyptus veneer covered by wood after two veneers have been split apart. Ten laboratory-scale 10-ply LVL panels were manufac- According to AS/NZS 2269.0 (2012), bonding between the tured. Six panels were made of eucalypt veneer only, plies in LVL shall be a Type A bond. This specification re- choosing veneer sheets of defined MoE grades based on quires a phenolic adhesive complying with AS/NZS their dynamic MoE values. One panel each was made 2754.1 (2016) and also a bond quality of any single glueline from 12, 14, or 17.5 GPa sheets and three panels were not less than 2 and an average of all gluelines not less than made from 16 GPa sheets. Each LVL panel contained ven- 5 when tested according to AS/NZS 2098.2 (2012). Both a eer from one or two logs only. Another four panels were steam and a vacuum pressure method were used to assess made of five radiata pine and five eucalypt veneer plies. A glueline quality (AS/NZS 2098.2 2012). range of eucalypt grades were used in these panels. The first (14 GPa), third (17 GPa), sixth (14 GPa), eighth (12 GPa) and tenth layers (12 GPa) were eucalypt veneers Results and discussion and the rest were radiata pine. Two panels were made Rotary peeling and veneer recovery from each of G2 and G4 radiata-pine grades. A typical For the 26 logs of 2.7 m tested, the small end diameter averaged 34.4 cm with a standard deviation of 4.3 cm while the large end diameter averaged 38.9 cm with a standard deviation of 6.3 cm. The average diameter of the E. globoidea logs (36.3 cm) was comparable to the radiata pine logs (34.9 cm) used in the plant for LVL pro- duction. In a preliminary test, an additional E. globoidea log was peeled cold. This was unsuccessful and, therefore, preheating to soften the wood was deemed necessary. The average veneer recovery for the 26 E. globoidea logs (54.5%) was lower than for radiata pine logs (69.8%) Fig. 1 Face grade veneer with no splitting (top) and composer Fig. 2 Dependence of usable veneer conversions on growth-strain grade veneer with severe splitting (bottom) of the individual E. globoidea logs Guo and Altaner New Zealand Journal of Forestry Science (2018) 48:3 Page 5 of 10 Fig. 3 Association between splitting length and growth-strain (left) as well as number of splits (right) but the best E. globoidea log had a recovery of 74.5% Growth-strain and veneer splitting (Tables 1 and 2). This result was mainly due to a 6% High quality veneers can be obtained from E. globoidea higher loss in the peeler core caused by severe splits in although splitting significantly degraded the visual the supplied E. globoidea logs and an 11.7% greater clip- appearance of many veneers (55% of the veneers had per loss caused by end-splitting of the veneers compared splitting lengths longer than 2 m). Veneers with no and with the radiata pine logs. However, the amount of severe splitting are shown in Fig. 1. round-up waste was lower for the eucalypt logs than the The average recovery of useable veneer (face and core radiata pine logs, which indicated a better log form for grades) from E. globoidea was 33.4%. Severe splitting E. globoidea. Veneer recovery from individual logs was caused by growth-stress contributed to the low recovery. highly variable ranging from 23.6% to 74.5% (Table 2). A The average growth-strain was 839.4 με measured by the previous peeling study with E. nitens reported an overall CIRAD method. The log with the lowest growth-strain recovery of 59% (McKenzie et al. 2003). However, part- had approximately half the growth-strain compared with sheets were included in that study while only full sheets that with the highest. Usable veneer recovery was nega- were calculated in the present study. tively associated with growth-strain (Fig. 2). The average A spindled lathe was used to peel the logs in the useable veneer recovery for the logs in the bottom quar- current study but spindleless lathes are extensively used tile (growth-strain >965.7 με) was 5.8%, while that for in China. They are suitable for rotary peeling smaller the top quartile (growth-strain <701.8 με) was 57.2%. diameter logs from young and fast-grown hardwood The mean splitting length per veneer for individual plantations (McGavin 2016). Spindleless lathes achieve logs ranged from 0.15 m to 8.66 m. For a veneer 2.65 m higher yields because they can peel logs to smaller peeler in length and 1.26 m in width, the average total splitting core diameters compared to spindled lathes currently length was 3.01 m. This suggested splitting was limiting used for radiata pine. Moreover, spindleless lathes can veneer quality. control splits by pressing the splits together during peel- The splitting length was measured after the veneers ing. Therefore, using this type of lathe may generate were dried. Drying was likely to exaggerate the splitting higher recovery and quality of veneer sheets. lengths but was assumed to affect all veneers equally in this study. In addition, rough handling and peeling set- tings can also contribute to the splitting of veneers. It was assumed that all veneers were equally affected in these ways and the differences among them were mainly caused by growth-stress. A positive correlation (r = 0.73) was observed between splitting lengths in veneers and growth-strain of corre- sponding logs (Fig. 3). In this study, average splitting length was low when growth-strain was less than ~800 με (CIRAD). Longitudinal growth-strain was reported to be positively related to end-splitting of E. nitens and E. globulus logs (Valencia et al. 2011; Yang and Pongracic 2004). The number of splits in a veneer sheet is another measure to evaluate veneer spitting. A strong linear relation- Fig. 4 Splitting length for veneers obtained from the centre to the ship (r = 0.91) was obtained between splitting length and outside of five individual logs split numbers (Fig. 3). Considering that the measurement of Guo and Altaner New Zealand Journal of Forestry Science (2018) 48:3 Page 6 of 10 Table 3 Physical and mechanical properties of dried E. globoidea veneers Dried density (kg/m ) Moisture content (%) Shrinkage (%) Velocity (km/s) Dynamic MoE (GPa) Mean 688.13 7.31 9.85 4.657 15.14 SD 68.55 1.09 0.77 0.240 2.05 Min 557.41 5.51 8.46 4.322 11.04 Max 824.00 9.32 11.31 5.097 19.51 the number of splits is less time consuming than quantifying logs. Furthermore, the increasing curvature with de- splitting length, it might be a better option for future studies. creasing radius can facilitate splitting. However, the vari- This result indicated that higher veneer recovery and ation between the stems was much bigger than the quality would be possible if growth-stresses were radial effect, with the veneer splitting independent of reduced. Methods to control the growth-stress and min- radius for the worst and the best logs. imise splitting are difficult to perform, have not been It is worth noting that the log preheating and peeling successful in industrial applications and incur ongoing settings in this study were optimised for radiata pine. costs (Archer 1987; Malan 1995; Yang and Waugh Acevedo et al. (2012) reported that better quality veneers 2001). Growth-stresses are heritable and selecting trees were obtained from E. nitens by adjusting nose bar pres- with low growth-stress in a breeding programme can sure and peeling knife angle. potentially solve this problem for a future resource such as E. globoidea (Davies et al. 2017; Malan 1995). For Physical and mechanical properties of veneer existing eucalypt plantations grown for the lower-value After drying, the average dry density was 668 kg/m and the wood chips, such as E. nitens, segregation would be an average moisture content was 7.3% (Table 3). No excessively option. However, current methods of measuring growth- high or low moisture contents were found, which indicated stress are time consuming and cumbersome, making this homogeneous drying of the E. globoidea veneers. approach impractical (Yang and Waugh 2001). Rapid The average shrinkage of the E. globoidea veneers was and non-destructive segregation methods need to be 9.9% tangentially and varied between 8.5 and 11.3%. developed. For example, Yang et al. (2006) measured Most veneers were heartwood as the sapwood in E. growth-strain of 10-year-old E. globulus and found cor- globoidea is very narrow. For comparison, typical relations with cellulose crystallite width measured using tangential shrinkage values for radiata pine veneers are a SilviScan-2 instrument. 6.4% for sapwood and 4.4% for heartwood. Higher The relationship between splitting length and radial shrinkage will result in greater volume loss. It should be position for five logs is shown in Fig. 4. The splitting noted that, within species, heartwood typically displays length of the veneer sheets tended to increase towards lower shrinkage than sapwood. the centre of the stem (position 0). The decreasing The average dynamic MoE calculated for the E. globoi- circumference with decreasing radius results in shorter dea veneer sheets from Metriguard acoustic velocity and tangential distances between the radial end-splits of the interpolated lab density was 14.67 GPa ranging from 9.59 to 20.26 GPa (Fig. 5). The equivalent static MoE was estimated to be 12.73 GPa based on the empirical conversion equation. Common LVL products manufac- tured from radiata pine range from 8 to 11 GPa. Jones et al. (2010) investigated 25-year-old E. globoidea for high- quality solid wood production. Boards from the butt logs were reported to have an average density of 655 kg/m , dynamic MoE of 13 GPa and static MoE of 12 GPa. Ac- cording to Haslett (1990), the timber of E. globoidea (over 25 years old) in New Zealand has an MoE of 14.6 GPa at a moisture content of 12%. High stiffness wood tended to have higher growth- strain (Fig. 6). This is an unfavourable association as stiff wood with low growth-strain is desirable. However, the association between MoE and growth-strain was moder- ate (r = 0.65) implying the existence of stiff logs which Fig. 5 Cumulative distribution of dynamic and static MoE of are low in growth-strain. Several logs produced veneers veneer sheets with MoEs above 15 GPa and growth-strain levels below Guo and Altaner New Zealand Journal of Forestry Science (2018) 48:3 Page 7 of 10 Fig. 6 Association between MoE and growth-strain (left) as well as veneer splitting (right) 800 με. More importantly no association (r = 0.27) was was large with the average stiffness ranging from found between veneer splitting and stiffness, which 12.1 GPa to 18.0 GPa. Analysis of variance found signifi- demonstrated that peeling quality needs to be improved cant differences (P < 0.001) in MoE values of veneers through reducing growth-stress rather than through from different trees. MoE. The stiffest logs yielded veneers with an MoE of It has to be noted that the tested E. globoidea was up to 19.5 GPa, but these did not necessarily have a severe genetically unimproved material of unknown proven- splitting problem. Therefore, it seems possible to obtain a ance. Wood properties like growth-stress and MoE are stiff eucalypt resource that yields high-quality veneers. under genetic control (Davies et al. 2017). Murphy et al. A weak positive association (R = 0.26) between CIRAD (2005) reported a heritability of 0.3 to 0.5 for growth- longitudinal displacement and MoE has been found previ- strain in Eucalyptus dunnii Maiden and indicated tree ously in E. globulus (Yang et al. 2006). With the increase breeding can be an effective method to reduce growth- of longitudinal displacement, microfibril angle tended to stress. Considerable variation among trees was observed, decline while density increased. Similar associations be- indicating a potential for genetic improvement. The rela- tween growth-strain and wood properties have been re- tively high acoustic velocity of eucalypts in the corewood ported for wood from Populus deltoides Bartr (Fang et al. could allow peeling veneers to a smaller peeler core with 2008). However, no statistically significant associations spindleless lathes, improving yields and allowing the use were found between growth-stain and dynamic MoE or of a small diameter younger resource. density for E. nitens (Chauhan and Walker 2004). The average distribution of MoEs for the veneers Bonding quality obtained from the nine trees assessed is shown in Fig. 7. The bond test revealed poor bonding of the plies. None As for veneer splitting, the variance in MoE among trees of the panels made with E. globoidea alone passed the specifications for structural LVL (Table 4). Density seemed to exaggerate the bonding difficulty for the 100% E. globoidea LVL. Eucalyptus globoidea panels with densities higher than 800 kg/m had average bonding qualities lower than 3. Alternating E. globoidea and P. radiata veneers improved bond quality, and all samples Table 4 Bond tests of six LVL panels made from E. globoidea veneers (listed in order of increasing density) Grade Density Steam test Immersion test (GPa) (g/cm ) EE EE EE EE EE EE min max mean min max mean 12 640.51 1 9 5 1 5 2 16 696.92 3 8 6 4 9 6 16 702.52 1 9 6 3 8 7 16 806.83 0 5 1 0 3 2 Fig. 7 Box-and-whisker plots of veneer MoE from nine trees. Dotted 14 809.22 0 3 2 2 3 2 lines show minimum and maximum values; the thick black line 17.5 860.05 1 4 2 0 7 3 shows the median; the box represents the upper and lower EE represents bond quality values between E. globoidea veneers (0 no bond – quartiles and outliers are marked using open circles 10 excellent bond). The minimum, maximum and mean values are shown Guo and Altaner New Zealand Journal of Forestry Science (2018) 48:3 Page 8 of 10 Table 5 Bond tests of four LVL panels made from a mixture of E. globoidea and P. radiata veneers Sample Radiata pine grade Steam test Immersion test ER ER RR Mean ER ER RR Mean min max min max 1 G4 4 99 7 1 99 7 2G4 3 7 10 5 0 9 8 6 3 G2 2 99 7 3 99 6 4 G2 2 99 6 1 89 6 G4 and G2 are radiata pine veneer grades; 12, 14 and 17.5 GPa grade eucalypt veneers were used in each sample. ER represents the gluelines between eucalypts and radiata pine veneers and RR stands for the glueline between two radiata pine veneers. Mean is the average value of all 9 gluelines passed the steam test, but only one sample passed the between splitting length and stiffness. It should be immersion test. The glueline between radiata pine plies possible to find stiff logs, however, which would was excellent (Table 5). yield satisfactory veneers. The considerable variation Eucalypts have a higher density and extractive content in stiffness observed here indicated a potential for than radiata pine, and both these factors can make glu- genetic improvement. ing difficult. In Australia, difficulty in bonding veneers 6) The lack of association between stiffness and from dense eucalypts (air-dry density above 700 kg/cm ) splitting length suggested that quick acoustic are known and some special adhesive formulations have measurements for segregating eucalypt logs suitable been developed for these species (Carrick and Mathieu peeling are unlikely to be successful. 2005; Ozarska 1999). It is commonly stated that low and 7) Bond performance of E. globoidea LVL was poor and medium density (below 650 kg/cm ) eucalypts glue well did not meet the New Zealand standard. Alternating but young E. nitens was found to have bonding issues E. globoidea and P. radiata veneers improved bond however (de Carvalho et al. 2004, Farrell et al. 2011, quality, but bonding of E. globoidea veneers still Hague 2013). Plywood products made from various needs to be addressed. Eucalyptus species have been reported in China, Acknowledgements Malaysia, Uruguay and Brazil indicating that satisfactory We would like to thank Richard Barry from Nelson Pine Industries for bonding can be achieved for many Eucalyptus species overseeing the veneer peeling and LVL testing as well as for his contributions to the manuscript. Paul Millen's coordination in harvesting the (de Carvalho et al. 2004; Hague 2013; Turnbull 2007; Yu trees is also appreciated. et al. 2006). Therefore, it seems probable that a technical solution for gluing E. globoidea can be found. Funding This research project was financially supported by the Ministry of Primary Industries’ Sustainable Farming Fund (SFF407602). FG would also like to Conclusions thank the Chinese Scholarship Council (CSC) for supporting his study. Authors’ contributions 1) Veneer recovery ranged from 23.6% to 74.5% among FG carried out the field work, growth-strain measurement, veneer peeling, 26 E. globoidea logs. Larger peeler core and higher data analysis and prepared the manuscript. CMA designed and organised clipper losses occurred compared to radiata pine. the study contributed to data analysis and manuscript preparation. Both au- thors read and approved the final manuscript. 2) High-quality veneers could be obtained from E. globoidea. Low growth-strain logs produced more Competing interests usable veneers. Splitting degraded veneer quality. The authors declare that they have no competing interests. 3) Splitting length in veneers was correlated to growth- strain of the corresponding logs. Split number was Publisher’sNote Springer Nature remains neutral with regard to jurisdictional claims in strongly associated with splitting length and can be published maps and institutional affiliations. used to evaluate veneer splitting severity. Veneers from inner wood tended to a have longer splitting Received: 24 May 2017 Accepted: 3 January 2018 lengths; however differences among logs were more pronounced. References 4) Moisture contents of dried veneers indicated good Acevedo, A, Bustos, C, Lasserre, JP, Gacitua, W. (2012). Nose bar pressure effect in the lathe check morphology to Eucalyptus nitens veneers. Maderas. Ciencia y drying of E. globoidea. The average tangential tecnologia, 14(3), 289–301. shrinkage was 9.9% and the volume loss was higher Alméras, T, & Clair, B. (2016). Critical review on the mechanisms of maturation stress than for radiata pine. Larger shrinkage is a common generation in trees. Journal of the Royal Society Interface, 13(122), 20160550. Archer, RR (1987). Growth stresses and strains in trees. Berlin: Springer-Verlag Gmbh. feature of high-density timbers. AS/NZS 2098.2 (2012). Methods of test for veneer and plywood. Method 2: Bond 5) The average dynamic MoE was 14.67 GPa ranging quality of plywood (chisel test). Wellington: Standards Australia/ Standards from 9.59 to 20.26 GPa. No association was found New Zealand. Guo and Altaner New Zealand Journal of Forestry Science (2018) 48:3 Page 9 of 10 AS/NZS 2269.0 (2012). Plywood-structural. Part 0: Specifications. Wellington: McKinley, R, Shelbourne, C, Low, C, Penellum, B, Kimberley, M. (2002). Wood Standards Australia/ Standards New Zealand. properties of young Eucalyptus nitens, E. globulus,and E. maidenii in northland, AS/NZS 2754.1 (2016). Adhesives for timber and timber products. Part 1: Adhesives New Zealand. New Zealand Journal of Forestry Science, 32(3), 334–356. for manufacture of plywood and laminated veneer lumber (LVL). Wellington: McWhannell, F. B. (1960). Eucalypts for New Zealand farms, parks and gardens. Standards Australia/ Standards New Zealand. Hamilton, N.Z: Paul's Book Arcade. Bal, BC, & Bektaş, I. (2012). The effects of some factors on the impact bending Millen, P (2009). NZ Dryland forests initiative: A market focused durable eucalypt strength of laminated veneer lumber. BioResources, 7(4), 5855–5863. R&D project. In L Apiolaza, S Chauhan, J Walker (Eds.), Revisiting eucalypts 2009 : Workshop proceedings, (pp. 57–74). Christchurch: Wood Technology Barr, N. (1996). Growing eucalypt trees for milling on New Zealand farms. Wellington, New Zealand: New Zealand Farm Forestry Association. Research Centre, Univeristy of Canterbury. Bootle, KR (2005). Wood in Australia. Types, properties and uses, (2nd ed., ). Miller, JT, Cannon, P, Ecroyd, CE (1992). Introduced forest trees in New Zealand: Sydney: McGraw-Hill Book Company. Recognition, role, and seed source. 11. Eucalyptus nitens (Deane et maiden) Carrick, J, & Mathieu, K (2005). Durability of laminated veneer lumber made from maiden. (FRI Bulletin 124). Rotorua: New Zeland Forest Research Institute. Blackbutt. In Proceedings of the 10DBMC international conference on durability Miller, JT, Hay, AE, Ecroyd, CE (2000). Introduced forest trees in New Zealand: of building materials and components, France. Recognition, role, and seed source. 18. Ash eucalypts. (FRI Bulletin 124). Rotorua: Chauhan, SS, & Walker, J. (2004). Relationships between longitudinal growth New Zealand Forest Research Institute. strain and some wood properties in Eucalyptus nitens. Australian Forestry, Millner, J. (2006). The performance of eucalyptus species in hill country. Palmerston 67(4), 254–260. North, New Zealand: Massey University. Davies, NT, Apiolaza, LA, Sharma, M. (2017). Heritability of growth strain in Eucalyptus MPI (2016). National exotic forest description (NEFD). Wellington: Ministry for bosistoana: A Bayesian approach with left-censored data. New Zealand Journal Primary Industries. of Forestry Science, 47, 5. https://doi.org/10.1186/s40490-017-0086-2. Murphy, TN, Henson, M, Vanclay, JK. (2005). Growth stress in Eucalyptus dunnii. De Carvalho, AM, Lahr, FAR, Bortoletto Jr, G. (2004). Use of Brazilian eucalyptus to Australian Forestry, 68(2), 144–149. https://doi.org/10.1080/00049158.2005. produce LVL panels. Forest Products Journal, 54(11), 61. 10674958. Fang, C-H, Guibal, D, Clair, B, Gril, J, Liu, Y-M, Liu, S-Q. (2008). Relationships Okuyama, T, Yamamoto, H, Yoshida, M, Hattori, Y, Archer, R. (1994). Growth between growth stress and wood properties in poplar I-69 (Populus stresses in tension wood: Role of microfibrils and lignification. Annales des Deltoides Bartr. Cv. “Lux” ex I-69/55). Annals of Forest Science, 65(3), 307. Sciences Forestières, 51(3), 291–300. Farrell, R, Blum, S, Williams, D, Blackburn, D (2011). The potential to recover higher Ozarska, B. (1999). A review of the utilisation of hardwoods for LVL. Wood value veneer products from fibre managed plantation eucalypts and broaden Science and Technology, 33(4), 341–351. https://doi.org/10.1007/ market opportunities for this resource: Part A. (project no: PNB139-0809). s002260050120. Melbourne: Forest & Wood Products Australia. Page, D, & Singh, T. (2014). Durability of New Zealand grown timbers. New Fournier, M, Chanson, B, Thibaut, B, Guitard, D. (1994). Measurements of residual Zealand Journal of Forestry, 58(4), 26–30. growth strains at the stem surface. Observations on different species. Annales Palma, HAL, & Ballarin, AW. (2011). Physical and mechanical properties of LVL des Sciences Forestières, 51(3), 249–266. panels made from Eucalyptus grandis. Ciência Florestal, 21(3), 559–566. https:// FSC (2012). Strategic review on the future of forest plantations. Helsinki, Finland: doi.org/10.5902/198050983813. Forest Stewardship Council. Poynton, RJ (1979). Eucalyptus globoidea Blakely. In RJ Poynton (Ed.), Tree planting in Gaunt, D, Penellum, B, McKenzie, HM. (2003). Eucalyptus nitens laminated veneer southern Africa: The eucalypts, (pp. 316–324).South Africa:Department of Forestry. lumber structural properties. New Zealand Journal of Forestry Science, 33(1), Shelbourne, C, Bulloch, B, Low, C, McConnochie, R. (2002). Performance to 114–125. age 22 years of 49 eucalypts in the Wairarapa district, New Zealand, Gerard, J, Bailleres, H, Fournier, M, Thibaut, B. (1995). Wood quality in plantation reviewed with results from other trials. New Zealand Journal of Forestry eucalyptus - a study of variation in three reference properties. Bois et Forêts Science, 32(2), 256–278. des Tropiques, 245, 101–110. Simmonds, JH (1927). Trees from other lands for shelter and timber in New Hague, J. R. B. (2013). Utilisation of plantation eucalypts in engineered wood Zealand: Eucalypts. Auckland: Brett Printing and Publishing Company. products. Melbourne, Australia. Suontama, M, Low, CB, Stovold, GT, Miller, MA, Fleet, KR, Li, Y, Dungey, HS. (2015). Haslett, A. N. (1990). Properties and utilisation of exotic speciality timbers grown in Genetic parameters and genetic gains across three breeding cycles for New Zealand. Part VI: Eastern blue gums and stringy barks. Eucalyptus growth and form traits of Eucalyptus regnans in New Zealand. Tree Genetics & botryoides Sm. Eucalyptus saligna Sm. Eucalyptus globoidea Blakey. Eucalyptus Genomes, 11(6), 133. https://doi.org/10.1007/s11295-015-0957-8. muellerana Howitt. Eucalyptus pilularis Sm. (FRI Bulletin 119). Rotorua: New Telfer, EJ, Stovold, GT, Li, YJ, Silva, OB, Grattapaglia, DG, Dungey, HS. (2015). Zealand Forest Research Institute. Parentage reconstruction in Eucalyptus nitens using SNPs and microsatellite Jacobs, MR (1945). The growth stresses of woody stems. Canberra: Commonwealth markers: A comparative analysis of marker data power and robustness. PLoS Forestry Bureau. One, 10(7), e0130601. https://doi.org/10.1371/journal.pone.0130601. Jones, TG, McConnochie, RM, Shelbourne, T, Low, CB. (2010). Sawing and grade Toba, K, Yamamoto, H, Yoshida, M. (2013). Micromechanical detection of growth recovery of 25-year-old Eucalyptus fastigata, E. globoidea, E. muelleriana and E. stress in wood cell wall by wide-angle X-ray diffraction (WAX). Holzforschung, pilularis. New Zealand Journal of Forestry Science, 40,19–31. 67(3), 315–323. Kennedy, S, Dungey, H, Yanchuk, A, Low, C. (2011). Eucalyptus fastigata: Its current Turnbull, J (2007). Development of sustainable forestry plantations in China: A status in New Zealand and breeding objectives for the future. Silvae Genetica, review. (impact assessment series report no. 45). Canberra: Australian Centre for 60(6), 259–266. International Agricultural Research. Kubler, H. (1987). Growth stresses in trees and related wood properties. Forestry Valencia, J, Harwood, C, Washusen, R, Morrow, A, Wood, M, Volker, P. (2011). abstracts, 48(3), 131–189. Longitudinal growth strain as a log and wood quality predictor for Lausberg, M, Gilchrist, K, Skipwith, J. (1995). Wood properties of Eucalyptus nitens plantation-grown Eucalyptus nitens sawlogs. Wood Science and Technology, grown in New Zealand. New Zealand Journal of Forestry Science, 25(2), 147–163. 45(1), 15–34. https://doi.org/10.1007/s00226-010-0302-1. Malan, FS. (1995). Eucalyptus improvement for lumber production. Seminario Van Ballekom, S., & Millen, P. NZDFI: Achievements, constraints and opportunities. Internacional de Utilização da Madeira de Eucalipto, 15,1–19. In CM Altaner, TJ Murray, & J Morgenroth (Eds.), Durable eucalypts on Margadant, R. (1981). Why not peel locally grown eucalyptus. Wood Southern drylands: Protecting and enhancing value, Blenheim, New Zealand, 2017 (pp. Africa, 6(10), 8–22. 1-11): New Zealand School of Forestry McGavin, RL (2016). Analysis of small-log processing to achieve structural veneer Van de Kuilen, JWG, Ceccotti, A, Xia, Z, He, M. (2011). Very tall wooden buildings from juvenile hardwood plantations. Melbourne: The University of Melbourne. with cross laminated timber. Procedia Engineering, 14, 1621–1628. https://doi. org/10.1016/j.proeng.2011.07.204. McKenzie, H. (1993). Growing durable hardwoods - a research strategy. New Zealand Journal of Forestry, 38(3), 25–27. Yang, JL, Baillères, H, Evans, R, Downes, G. (2006). Evaluating growth strain of Eucalyptus McKenzie, H, Turner, J, Shelbourne, C. (2003). Processing young plantation-grown globulus Labill. From SilviScan measurements. Holzforschung, 60(5), 574–579. Eucalyptus nitens for solid-wood products. 1: Individual-tree variation in Yang, JL, Baillères, H, Okuyama, T, Muneri, A, Downes, G. (2005). Measurement quality and recovery of appearance-grade lumber and veneer. New Zealand methods for longitudinal surface strain in trees: A review. Australian Forestry, Journal of Forestry Science, 33(1), 62–78. 68(1), 34–43. https://doi.org/10.1080/00049158.2005.10676224. Guo and Altaner New Zealand Journal of Forestry Science (2018) 48:3 Page 10 of 10 Yang, J, & Pongracic, S (2004). The impact of growth stress on sawn distortion and log end splitting of 32-year old plantation blue gum. (project no: PN03.1312). Vicroria: Forest and Wood Products Research & Development Corporation. Yang, JL, & Waugh, G. (2001). Growth stress, its measurement and effects. Australian Forestry, 64(2), 127–135. Yu, Y, Ren, D, Zhou, Y, Yu, W. (2006). Preliminary study on gluability of E. urophylla veneer. China Forest Products Industry, 33(4), 20–23.

Journal

New Zealand Journal of Forestry ScienceSpringer Journals

Published: Dec 1, 2018

References