The effects of nodes and resin on the mechanical properties of laminated
bamboo timber produced from
Gigantochloa scortechinii
Rogerson Anokye
a,b, Edi Suhaimi Bakar
a,c,⇑, Jegatheswaran Ratnasingam
a, Adrian Choo Cheng Yong
c,
Nova Noliza Bakar
daDepartment of Forest Production, Faculty of Forestry, Universiti Putra Malaysia, 43400 Serdang, Selangor Darul Ehsan, Malaysia bDepartment of Interior Architecture and Furniture Production, Kumasi Polytechnic, P.O. Box 854, Kumasi, Ghana
cInstitute of Tropical Forestry and Forest Products, UPM, 43400 Serdang, Selangor Darul Ehsan, Malaysia
dDepartment of Mathematics, Faculty of Mathematics and Natural Sciences, University of Andalas, 25163 Padang, West Sumatera, Indonesia
h i g h l i g h t s
Flexural performance of laminated bamboo timber with nodes improves with increase in node intervals. Weakness in node are attributed to low density and the irregular vascular bundle arrangements. Phenol formaldehyde provides higher performance than polyvinyl acetate in the LBT bonding.
Glue type and spread rate have utmost influence on the compression and shear bond strengths of the LBT.
a r t i c l e
i n f o
Article history: Received 1 July 2015
Received in revised form 12 November 2015 Accepted 13 December 2015
Keywords:
Laminated bamboo timber Mechanical properties Nodes
Phenol formaldehyde Polyvinyl acetate
a b s t r a c t
The objective of this work was to evaluate the mechanical properties of laminated bamboo timber (LBT) manufactured from bamboo (Gigantochloa scortechinii). Bamboo strips containing nodes were used to produce laminated samples. Each bamboo mat was arranged with 5 cm intervals ranging from 0 cm to 15 cm between the nodes in successive laminae. Phenol formaldehyde (PF) and polyvinyl acetate (PVAc) were used at two spread rates of 200 g/m2and 250 g/m2. The best mechanical properties were found in samples without nodes. Increasing intervals also resulted in increasing strengths. In all the mechanical properties studied, PF had higher strength with 200 g/m2spread rate except for shear where PVAc had similar values with PF. It appears that interval levels in the joints influenced the overall mechanical properties of the samples.
Ó2015 Elsevier Ltd. All rights reserved.
1. Introduction
Bamboo is native to Southeast Asia, South America, Africa, etc. where they grow in abundance. However, the economic develop-ment of bamboo is relatively slow. In Africa especially in Ghana, the most common non-timber forest product is wild bamboo. Bam-boo covers about 300,000 ha withBambusa vulgaris as the most common species (95% of the bamboo stands). Despite its abun-dance, bamboo has not been extensively utilized [1,2]. Eighteen exotic species has also been introduced from Hawaii and many of them are striving. Bamboo application in Africa has been limited to rural housing, handicrafts, temporal posts and props in the building industry, furniture and recently on charcoal production.
They are mostly used in the round culm form which does not require much complicated processing. As forest products continues to decrease with the ever increasing demand for wood and wood products in the country, there is a need to innovatively develop bamboo as a substitute to slow growing hardwoods for furniture manufacturing.
The Asian region has made some progress in the development of bamboo with China taking the lead of producing a new type of bamboo composite panel (over 90% bamboo) called laminated bamboo fibrillated-veneer lumber (LBL). LBL has been extensively promoted because of its high strength and stiffness[3,4]. In Malay-sia, many studies have also been conducted on the most common bamboo species which areGigantochloa scortechiniiandB. vulgaris.
The majority of the studies were concentrated on the properties of laminated bamboo as both boarding and structural members. Accordingly, most of the studies in Malaysia examined bending
http://dx.doi.org/10.1016/j.conbuildmat.2015.12.083 0950-0618/Ó2015 Elsevier Ltd. All rights reserved.
⇑ Corresponding author at: Department of Forest Production, Faculty of Forestry, Universiti Putra Malaysia, 43400 Serdang, Selangor Darul Ehsan, Malaysia.
E-mail address:edisuhaimi@upm.edu.my(E.S. Bakar).
Contents lists available atScienceDirect
Construction and Building Materials
and compression properties. The effects of layer structure, bamboo species, oil treatment, and glue type on the mechanical properties of laminated bamboo boarding have all been studied and found the bamboo strips to have yielded excellent mechanical properties
[5–9].
Experiences of Asian countries have shown undoubtedly that bamboo is a valuable and sustainable natural resource[10]. Its fast growing characteristic, capability to easily replenish itself after harvesting, advancement of processing technology, versatility of its use coupled with its low Eco-cost[3]has made bamboo furni-ture an imperative substitute to complement conventional wood furniture in the world market.
Laminated bamboo as the most popular bamboo product with good market acceptance requires the need for more study. The low density and the irregular vascular bundle arrangements of the nodes of bamboo affect most of the physical properties of lam-inated bamboo timber (LBT)[11,12]. These weak parts of the bam-boo culms have not been extensively studied. A thorough understanding of the carefully arranged node features in the LBT to assume the optimum strength is therefore required. Accordingly, the aim of this study was to verify the mechanical properties of LBT panels produced from G. scortechinii bamboo under different glue types, glue spreading rates and the node inter-val arrangements.
2. Materials and methods
2.1. Material preparation
G. scortechinii(buluh semantan) was used for this study. This was based on the results of our previous studies on the physical properties of the bamboo strips of the selected species[11]. The bamboo was harvested from the Forestry Research Insti-tute of Malaysia (FRIM).
Twelve (12) bamboo culms were selected and used for this investigation. Bam-boo strips were prepared from parts taken only up to 3 m height of the culm from the bottom. These were then reduced into groups containing only internodes and ones with nodes with an average length of 30 cm. The selected culms were split into 3 cm widths which provides an optimum yield of strip before preserved in a solu-tion of Borax for 15 min. After air-drying under a shed until 12% moisture content (MC), the splits were planed to final strip dimensions of 5 mm20 mm300 mm to remove the inner waxy and epidermal layers. The strips were then glued edge-to-edge with PVAc glue and pressed horizontally using clamps to obtain a wide mat. The glued members of 5 mm100 mm300 mm were subsequently plied together with the surface of the epidermal layer facing one direction to ensure an optimum adhesive performance[8,13]. The nodes were aligned alternatively in suc-cessive lamina at different interval groups of 5 cm, 10 cm and 15 cm to determine the optimal distance as shown inFig. 1.
Two different glue types – phenol formaldehyde (PF) of 49% solid content and viscosity of 6 cP at 25°C and polyvinyl acetate (PVAc) of 65% solid content and vis-cosity of 2 cP at 25°C and 65% and two glue spread rates (200 and 250 g/m2) were used for board lamination[14]. Pressing was carried out at 140°C and a pressure of
1.47 MPa for 30 min for the samples with PF and cold press at the same pressure for 4 h at room temperature for samples with PVAc. The experimental codes are PV and PF for glue types,LandHfor glue spread rates, andA,BandCfor the node intervals between the successive lamina (Table 1). Similar strips were prepared and lami-nated for specimens without nodes as control.
2.2. Mechanical properties of LBT
The basic mechanical properties of LBT for theG. scortechiniiwere evaluated in the static bending, compression and shear test. LBT without node and those con-taining nodes underwent static bending to examine the effect of the node intervals on the strength. The static bending was done in flatwise or perpendicular to the lamination glue lines. LBT without nodes underwent compression and shear tests to determine the effect of the glue types and their spread rates on the compressive and shear strengths. LBT specimens were reduced to sizes of 20 mm20 mm300 mm for static bending tests with 4 replicates totaling 48 specimens for the variables containing nodes and 16 specimens for those without nodes as control. The compressive and the shear tests were performed using 20 mm20 mm40 mm and 20 mm20 mm150 mm specimens, respec-tively. Ten replicates each were used for the various tests. The standard ASTM 3043-00[15]was referred to for the static bending test, with a consideration of lamination orientation. ASTM D143-09[16]and ASTM D7078-12[17]were also referred to for the compressive and shear tests respectively whiles, ASTM 5266-99[18]was referred to for the estimation of percentage of wood failure in adhesive bonding.
3. Results and discussion
3.1. Failure behavior of LBT under bending test
The bending test of 20 mm20 mm300 mm laminated
bamboo members with internode and those with nodes were
per-Fig. 1.Laminate bamboo timber at different node intervals. Table 1
Experimental design of the LBT at different node intervals.
Coding Glue type Glue spread rate (g/m2)
formed forG. scortechiniisampled from basal growth part for its wall thickness with horizontal lamination direction. The results indicated that most of the failures in the samples with node occurred at the node (Figs. 2–4). This may be due to the low den-sity, irregular vascular bundles arrangements, and consequently low strength of the strip in the node site[11,19,20].
Regardless of the glue type and the glue spread rates, the nature of failures was rather based on the node intervals since almost all the specimens started failing from the node at the tension side of the load. Specimens bonded with 5 cm intervals broke from the bottom laminae and propagate through the nearest glue line until it emerges through the next node on the successive laminae (Fig. 2). This confirms that the bamboo material below the adhesive-wood interphase layer is weaker than the bond line
[21]. In some of them, the split continues through the glue line.
This is believed to have been caused by the closeness of the nodes. The 10 cm node intervals exhibited a similar trend of the 5 cm but most of them could not break through to the glue line and there-fore extends through a split in the laminae (Fig. 3). This was also found to be as same with that samples laminated 15 cm interval (Fig. 4). With these fracture behavioral results, the study suggests that nodes are very weak points and should not be aligned close together in successive lamina in LBT.
3.2. Effects of glue type on the static bending strength of LBT with node
The MOE of LBT fromG. scortechiniishown inFig. 5revealed that LBT specimens with nodes bonded with PF glue exhibited a higher performance by about 13.86% than PVAc. The MOR also showed a similar trend of increase over that of the PVAc. However,
Fig. 2.Failure behavior of LBT at 5 cm node intervals on static bending.
the result of the statistical analysis showed no significant difference in the glue type on both MOE and MOR (at 5% level of probability) (Table 2).
3.3. Effects of glue type, glue spread rate and node intervals on the static bending strength of LBT
The overall results of the effects of glue type, glue spread rate and the node intervals on the static bending (MOE and MOR) are
shown inTable 2. The table showed clearly that there was no sig-nificant difference in the MOE and MOR among the means of the glue types and the spread rates as well as the interactions between all the three factors. However, there was a significant difference among the means of the node intervals.
MOR recorded for samples with different node intervals showed almost the same strength with the node intervals of 5 cm and 10 cm. LBT with 10 cm interval joints yielded 1% and 7.6% higher MOR than 5 cm and 15 cm intervals respectively (Fig. 5). The best strength yield of LBT with nodes was found to have come from laminated specimens bonded with 200 g/m2 of PF aligned at an
interval of 10 cm. There was a significant difference among the means of the node intervals from samples without node to 15 cm (Table 2). In all, the control samples (without node) yielded signif-icantly higher strengths, indicating that samples without nodes have higher strength than samples with nodes at shorter intervals.
3.4. Effects of the glue type and the glue spread rate on the compression strength of LBT
The compressive strength test revealed that samples bonded with PF glue exhibited higher performance than those bonded with PVAc (Fig. 6). The average compressive strength of the specimens tested were 54.01 MPa with PF at 250 g/m2recording the highest Fig. 4.Failure behavior of LBT at 15 cm node intervals on static bending.
117.80
without node 5cm 10cm 15cm
14.17
without node 5cm 10cm 15cm
Fig. 5.Comparison of the effects of node interval on the bending strength of LBT.
Table 2
ANOVA on the effect of glue type, glue spread rate and the node interval on the LBT.
Source MOE MOR
Fvalues P values
Fvalues P values
Glue type 4.266ns .430 1.087ns .301
Spread rate .286ns .595 2.882ns .095
Node interval 13.38⁄⁄ .000 5.856⁄⁄ .001 Glue type⁄spread rate .654ns .421 3.049ns .084 Glue type⁄node interval 1.583ns .211 .401ns .671 Spread rate⁄node interval 3.004ns .055 1.856ns .163 Glue type⁄spread rate⁄node
interval
.447ns .641 .592ns .555
strength of 67.18 MPa. This is a bit higher than what was obtained by Anwar et al. when similar products from the same species were tested[22]. Yeh and Lin, also had an average compressive strength of 69.6 MPa with a different bamboo species[9]. Between the two glue types, samples bonded with PF showed 46.9% higher strength than samples bonded with PVAc. This large difference might be due to the plasticization of PF within the vascular bundles closer to the glue line from the hot pressing.
The failure phenomena of the specimens based on the various glue types and the glue spread rates of the LBT compressed parallel to grain showed damages from the top and propagating either along the glue line or through the bamboo material. It was evident that most of the specimens bonded with PVAc failed faster as their deformations could not reach the bottom. Similar crushing behav-ior was identified when strips of the same species was subjected to compression parallel to grain[8].
3.5. Effects of glue type and spread rate on the shear bond strength of LBT
According toFig. 7, the isolated effect of the type of glue and spread rate were analyzed. The values found for PF was statistically not different from values found for the PVAc. However, the values of glue line shear strength are lower when compared with the val-ues reported by Anwar et al.[23], who found an average value of 3.12 N/mm2for the untreated plybamboo from the same species.
The difference may have been as a result of the laminae orienta-tions used in our study.
The glue spread rates (200 g/m2 and 250 g/m2) generally
were significantly different (atp60.05) in the bond shear. This
emphasized the work of Frihart and Hunt who found that several wood glues have poor ‘‘gap-filling” abilities[21]. That is, they bond tightly to wood but not to themselves and thereby requiring a min-imal glue line for maximum strength. Brady and Kamke also found better penetration of glue in a more porous and permeable wood causing more effective glue line[24]. It can therefore be inferred that the rate of glue spread can have a significant influence on the shear bond strength of the LBT. The results gathered shows a considerable improvement in the MOE and MOR results of many laminated bamboo and other wood products ranging from 65% to over 200% respectively[25].
Failure of the substrate (G. scortechinii) was observed in all spec-imens in glue line shear (Fig. 8). An average wood failure of 83.25% obtained indicated that spread rate used is, at least, enough to guarantee failure of the substrate. The spread rates did not show a significant difference in substrate failure. However, there was a significant difference among the two glue types in terms of their percentage of failure. Adamopoulos et al., found similar results with the two glues studied[26]. They attributed it to a higher pen-etration of the adhesives in the vessels by PVAc than with PF. The results proved that, the specimens met the minimum shear strength requirement of Malaysia Standard: MS 228[27].
4. Conclusions
1. The results revealed that most of the failures on the bending occurred at the node which was attributed to the low density and the irregular vascular bundle arrangements indicating the node as having low strength.
2. The flexural performance of the LBT containing nodes increases with increase in node intervals. Generally, 10 cm interval is ade-quate for overlapping the nodes in order to attain the maximum performance.
3. Phenol formaldehyde showed a higher performance than poly-vinyl acetate in the LBT bonding on both MOE and MOR. 4. The type of glue and the spread rate were also found to be of
utmost influence in the compression and shear bond strengths of the LBT.
However, a study on compression and shear at the nodes as well as the depth of penetration of the glue could be carried out in future to confirm this results of this study.
Acknowledgments
The authors would like to thank Universiti Putra Malaysia, Malaysia for providing facilities and financial support through
42.31a 45.18
Glue type / glue spread rate
Fig. 6.Effects of the glue type and the glue spread rate on the compressive strength parallel to grain of the LBT. Means denoted by the same letter are not significantly different at 5% probability level.
2.97a
Glue type / Glue spread rate
Fig. 7.Effects of the glue type and the glue spread rate on the glue line shear strength. Means denoted by the same letter are not significantly different at 5% probability level.
Percentage of bamboo failure (%)
Glue type / Spread rate
RUG Grant GP-IPB/2013/9413401, which made it possible to carry out this research. The authors also wish to express their gratitude to the staff of the Faculty of Forestry and FRIM for their support during this work. The results are part of the Ph.D. project of Roger-son Anokye.
References
[1]R. Anokye, G. Adu, The status of non-timber forest products (NTFPs) development in Ghana, J. Environ. Sci. Comput. Sci. Eng. Technol. (JECET) 3 (1) (2014) 144–155.
[2] M. Brink,Bambusa vulgaris Schrad. ex J.C. Wendl. [Internet]. Record from PROTA4U. in: D. Louppe, A.A. Oteng-Amoako, M. Brink (Eds.), Plant Resources of Tropical Africa (PROTA), Wageningen, Netherlands, 2008. <http://www. prota4u.org/search.asp> (accessed 18.03.15).
[3]X.F. Yu, H.L. Sun, W.J. Yu, Manufacturing technology and performance of structural laminated veneer lumber fromBambusa distegia, China Wood Ind. 25 (4) (2011) 1–3.
[4]F.D. Meng, Y.L. Yu, R.X. Zhu, Y.H. Zhang, W.J. Yu, Effect of resin impregnation rate on physical and mechanical properties of laminated bamboo fibrillated veneer lumber, China Wood Ind. 25 (2) (2011) 1–3.
[5]K. Rassiah, M.M.H. Megat, A. Ali, Mechanical properties of laminated bamboo strips from Gigantochloa scortechinii/polyester composite, Mater. Des. 57 (2014) (2014) 551–559.
[6]S. Talabgaew, V. Laemlaksakul, Experimental studies on the mechanical property of laminated bamboo in Thailand, World Acad. Sci. Eng. Technol. 34 (2007) 327–331.
[7] M. Mahdavi, P.L. Clouston, A.M. Asce, S.R. Arwade, Development of laminated bamboo lumber: review of processing, performance, and economical considerations, J. Mater. Civ. Eng. 23 (7) (2011) 1036–1042,http://dx.doi.org/ 10.1061/(ASCE)MT.1943-553.
[8]H. Hamdan, A. Zaidon, M.M. Tamizi, Mechanical properties and failure behaviour ofGigantochloa scortechinii, J. Trop. For. Sci. 21 (4) (2009) 336–344. [9] M.-C. Yeh, Y.-L. Lin, Finger joint performance of structural laminated bamboo member, J. Wood Sci. 58 (2) (2011) 120–127, http://dx.doi.org/10.1007/ s10086-011-1233-7.
[10] B.M.E. Dannenmann, C. Choocharoen, W. Spreer, M. Nagle, H. Leis, A. Neef, J. Mueller, Tropentag 2007 University of Kassel-Witzenhausen and University of Göttingen, October 9–11, 2007.
[11]R. Anokye, R.M. Kalong, E.S. Bakar, J. Ratnasingam, M. Jawaid, K. Awang, Variations in moisture content affect the shrinkage ofGigantochloa scortechinii
andBambusa vulgarisat different heights of the bamboo culm, BioResources 9 (4) (2014) 7484–7493.
[12]A.S. Nordahlia, U.M.K. Anwar, H. Hamdan, M.A.L.S. Mahanim, Anatomical, physical and strength properties of Schizostachyum brachycladum (buluh lemang), J. Bamboo Rattan 10 (4) (2011) 111–122.
[13] D. Liu, J. Song, D.P. Anderson, P.R. Chang, Y. Hua, Bamboo fiber and its reinforced composites: structure and properties, Cellulose 19 (5) (2012) 1449– 1480,http://dx.doi.org/10.1007/s10570-012-9741-1.
[14] S. Aicher, H. Reinhardt, H. Garrecht, Material and Joints in Timber Structures: Recent Developments of Technology, Springer Science & Business Media (2013),http://dx.doi.org/10.1007/978-94-007-7811-5_32.
[15] ASTM D3043-2011, Standard Test Method for Structural Panels in Flexure, Annual Book of ASTM Standards, vol. 04.10, Philadelphia, PA.
[16] ASTM. D143-09, Standard Methods of Testing Small Clear Specimens of Timber, D143, West Conshohocken, PA.
[17] ASTM D7078-12, Standard Test Method for Shear Properties of Composite Materials, West Conshohocken, PA.
[18] ASTM D5266-99, Standard Practice for Estimating the Percentage of Wood Failure in Adhesive Bonded Joints, Annual book of ASTM standards, vol. 15.06, West Conshohocken, PA, 2005.
[19] Shamsun Nahar, M. Hasan, Effect of chemical composition, anatomy and cell wall structure on tensile properties of bamboo fiber, Eng. J. 17 (1) (2013) 61– 68,http://dx.doi.org/10.4186/ej.2013.17.1.61.
[20] S. Amada, Y. Ichikawa, T. Munekata, Y. Nagase, H. Shimizu, Fiber texture and mechanical graded structure of bamboo, Compos. B Eng. 28 (1–2) (1997) 13– 20,http://dx.doi.org/10.1016/S1359-8368(96)00020-0.
[21] C.R. Frihart, C.G. Hunt, Adhesives with Wood Materials- Bond Formation and Performance, Wood Handbook, Chapter 10: <http://www.fpl.fs.fed. us/documnts/fplgtr/fplgtr190/chapter_10.pdf> (retrieved 20.03.15). [22] D. Biswas, B.S. Kanti, M. Mozaffar Hossain, Physical and mechanical properties
of urea formaldehyde-bonded particleboard made from bamboo waste, Int. J. Adhes. Adhes. 31 (2) (2011) 84–87, http://dx.doi.org/10.1016/j.ijadhadh. 2010.11.006.
[23]U.M.K. Anwar, M.T. Paridah, H. Hamdan, M. Abd Latif, A. Ziadon, Adhesion and bonding fromGigantochloa scortechinii, Am. J. Appl. Sci. (2005) 53–58. [24]D.A. Brady, F.A. Kamke, Effects of hot-pressing parameters on resin
penetration, For. Prod. J. 38 (11/12) (1988).
[25] J.F. Correal, J.S. Echeverry, F. Ramirez, L.E. Yamín, Constr. Build. Mater. 73 (2014) (2014) 105–112,http://dx.doi.org/10.1016/j.conbuildmat.2014.09.056. [26]S. Adamopoulos, A. Bastani, P. Gascón-Garrido, H. Militz, C. Mai, Adhesive bonding of beech wood modified with a phenol formaldehyde compound, Eur. J. Wood Wood Prod. 70 (6) (2012) 897–901.
