0488-A2

ASSESSMENT OF THE DIMENSIONAL STABILITY OF CEMENT - BONDED PARTICLEBOARD FROM POST - HARVEST BANANA STEM RESIDUES AND SAWDUST

AJAYI, BABATUNDE ¹

ABSTRACT

Thin-size (6mm thickness), dense homogenous cement - bonded boards were produced from mixed shredded flakes of banana stem and sawdust of mixed hardwood species. The influences of weight proportion of the two types of ligno-cellulosic materials at five blended levels of 0:100, 25: 75, 50: 50, 75: 25 and 100: 0 of sawdust to flakes of banana stem; and cement mixing ratio at two levels of 2.0: 1.0 and 3.0: 1.0 on thickness swelling and water absorption properties of the experimental boards were assessed in the study. Every board was made at standard chemical additive concentration of 3.5% of the cement weight in board and density of 1150 kg/m ³ on oven dry weight nominal basis. Results showed that it was technically impossible to make boards from the pure banana fibres without the incorporation of the sawdust particles. All boards made at hat level failed to bond, as the cement binder did not set following 24 hours pressing cycle. Consequently, all such boards crumbled following demoulding processes. Panels obtained from the rest treatment combinations exhibited water absorption values which ranged from 3.69 to 22.22% and thickness swelling values of 0.27 to 6.50% following soak in water for 48 hours prior to testing. Increase in the wood fibre (sawdust particles) content manifested in the production of more dimensionally stable experimental boards. The result shows that for shredded banana fibres to be suitable in cement-board manufacture, pre-treatment process will be mandatory.

INTRODUCTION

The effect of human influence on the forest areas has reduced greatly the available number of valuable and economic tree species. The over-exploitation of Nigeria's hardwood species from the natural and plantation forests necessitates the need to focus on alternative sources of raw materials at least for use in panel products manufacturing which could substitute for use of sawnwood (lumber) and plywood for some specific end uses notably for production of different grades of furniture. Such potential panel products include particleboard and fibreboard. These alternative sources of raw materials specifically are logging residues; wastes emanating from industrial processing of wood in the sawmills, plywood mills and furniture mills; as well as agricultural residues. These raw materials have commercially been processed into pulp/paper; resin and cement-bonded particleboard; and fibreboard in different countries (Kollmann et.al. 1975, Sandermann 1970, Simatupang et. al. 1978). Profitable uses of these raw materials for value-added wood products manufacturing have substantially minimize periodic shortage of wood raw material and reduce exploitation pressure on the forest resources in different regions of the world. Among the agricultural residues, which have been used to-date, are bagasse, maize stalk and cotton stalk while laboratory scale research has focussed on use of many others (Sandermann 1970, Bison 1981, Elten 1981, Dada and Badejo 1981, Ajayi, 1982). Synthetic adhesive is a major binder component in the wood-based industries. It has provided waterproof and insect/fungus resistance bonds that have enhanced use of particleboards for furniture and structural uses in different countries. Scarcity of this group of adhesives in some countries and high cost in others as they are petrochemical industry by-products have promoted use of cement for particleboard manufacture most especially in the developing countries (Moslemi 1989, Badejo 1984, Oyagade 1995, Ajayi 2000). In Nigeria, inflation and foreign exchange problems have resulted in the frequent difficulties encountered in the production of resin-bonded particleboards. The resin binder has been noted to account for about 65% of the raw material input cost for particleboard production (Omoluabi, 1982). Therefore there is need to seek for alternative binder such as cement paste. The idea of using cement as a binder has been in existence as far back as 1930 when board was produced using magnesite cement as the binding agent (Maloney, 1977).

Types of panels produced using cement as a binder depends upon the geometry of the wood particles and the types of mineral binders used. The different board types available include the wood-wool excelsior boards, gypsum-bonded boards and cement-bonded particleboards. Wood-cement boards exhibit some inherent properties, which made them versatile as constructional material is for ceiling, roofing, flooring, partitioning, cladding, shuttering and wall panel members for low cost housing in many countries (Lee, 1991). Cement-bonded particleboard has many advantages over resin - bonded particleboard in that they are highly resistant to fire outbreak, exhibit low moisture uptake and swelling under prolonged soak in water or exposure to moisture, high resistance to insect, mould and fungi attack (Dinwoodie and Paxton, 1991; Simatupang, 1987,Sandermann, 1970).

Despite the excellent performance of cement bonded boards, many wood species and agricultural residues may not bond well with cement to form suitable panel due the presence of some chemical substances in the wood which inhibit the proper setting of the cement binder ( Davis, 1966, Weatherwax and Tarkow, 1967, Simatupang et al, 1991, Fuwape, 1992; Ajayi, 2000;). These chemical substances include sugar, starch, hemicelluloses of the sapwood and extractive notably phenolic compounds (Biblis and Lo, 1968). While board has been made from different wood species commercially in different countries, agricultural residues have a similarly been used at least on laboratory scale basis (Sandermann 1970, Simatupang et al, 1978, Elten 1981). Post harvest banana stem residue is a potential raw material for particleboard production in some areas of the country, e.g. Ogun, Ondo, Ekiti, Edo and Cross-River and is available in large quantities. The objectives of this study therefore is to initiate laboratory -scale research investigation on use of flakes of banana stem residues in cement-board production and examine the feasibility of doing so if the residue were to be combined and mixed with sawmill sawdust residues at varying mixing ratios.

MATERIALS AND METHODS

The banana stems used in this study were collected from a plantation farm established at the Federal University of Technology, Akure - Nigeria. They were thereafter cut into billets and transported to the Forestry Research Institute of Nigeria wood workshop for further processing. The billets were later shredded into flakes on a tenoning machine, which carried scoring knives, which produced uniform flakes measuring 50 mm long and 25 mm wide. The flakes were thereafter dried to moisture content of 12% and bagged for further investigation. The flakes later received hot water pre-treatment in an aluminium bath at a temperature of 80 ^o C for a soaking period of 1 hour. This pre-treatment process was carried out in order to facilitate the removal of sugars and other chemical substances present in the raw material, which may possibly retard or completely inhibit the setting of the cement binder. After the end of the soaking period, the hot water was drained off while the materials (flakes) were air dried to a moisture content of 12% prior to use. They were then stored in polythene bags in readiness for board production. Sawdust of mixed species was collected from the sawmill operated by the Forestry Research Institute of Nigeria (FRIN) Ibadan. The sawdust particles were handled, treated and stored in a similar manner as the flakes of the banana residue.

The experiment was designed to include the following production variables:

i. Board density at 1150kg/m ³
ii. Blending proportion of sawdust to banana flakes at 0:100, 25:75, 50:50,75:25, 100:0 of the weight of these lingo-cellulosic raw materials in board.
iii. Mixing ratio of cement to wood at 2.0:1.0, 3.0:1.0 based on oven dry weight and volume of the board.
iv. Additive concentration at level of 3.5% of cement weight in each board.
v. Pressing pressure: 1.23N/mm ² .
vi. Board size: 350 mm (width) by 350 mm (length) by 6 mm (thickness)
The experimental design was 2 x 5 factorial experiment in complete randomised design, the combination of which gave 10 treatments.

According to each of the mixing ratio level used with board density level, the required quantity of the sawdust and banana flakes was weighed out and placed in a plastic bowl. The solution of calcium chloride and water was added uniformly and blended together. Wooden moulds of 350mm x 350mm were placed on caul with carried polythene sheets to prevent sticking of the formed boards on to the plates. The furnish was spread out on the plate, thereafter; a wooden press was used to press down the furnish within the mould. It was later covered with another polythene sheet, after which a top metal plate was placed on it and transferred to the press and cold pressed under a pressing pressure of 1.23N/mm ² to a thickness of 6mm for a period of 24 hours before removing them from the moulds for curing. After pressing, the demoulded boards were packed inside polythene bags for another 28 days to enhance further curing of the cement binder. Possible loss of water from the boards inside the bags was prevented through proper sealing of the bags.

The test specimens were cut on a circular saw. The edges were trimmed to avoid edge effect on the boards during testing. The board was further cut into various test specimens for evaluation in accordance with BS 5669: (1979). The parameters tested for are water absorption and thickness swelling. The data obtained form the experiment were analysed using descriptive statistical analysis, which gave summaries of the raw data; table presentation, which showed the variation in the variables at different levels of the factors considered and two ways analysis of variance for factorial experiments which estimated the importance of various sources of variation on the dependent variables.

RESULT AND DISCUSSION

The summary of the mean values of water absorption measured in each of the treatments combinations employed in this study is presented in Table 1. The mean values obtained for water absorption (WA) following 48 - hour immersion ranged from 3.69 to 22.22%. The lowest water absorption value of 3.69% was obtained from the mixing ratio level of cement to cellulosic materials and blending proportion ratio of 100:0 of sawdust to banana flakes which implies that increase in sawdust content of the board, resulted in decrease in water absorption.

Table 1: Average Water Absorption And Thickness Swelling Values Cement-bonded Boards Using Banana Flakes and Sawdust.

This observation is in agreement with the report of Fuwape, (1992), Oyagade, (1990) and Ajayi, (2000). The result showed that blending proportion and cement binder content have a positive influence on water absorption properties of cement-bonded board. As the cement content increased adequate and more cement binder is made available to thoroughly coat the banana flakes. Although water absorption (WA) decreased with increase in blending proportion of sawdust and the flakes, the decrease in water absorption values indicated that water could not easily penetrate the board when the proportion of sawdust in the board is high. The board with blending proportion of 0: 100 is found to be more porous and absorbed more water faster because of the increased quantity of banana flake present in it. The inter - flake void was variously filled with water thereby increasing the quantity of water constantly stocked in these voids, and eventually increasing the weight of the boards (Ajayi 2000).

The result showed that as the blending proportion of sawdust to banana flakes increased, water absorption decreased. The observed decrease in water absorption with increase in cement content is probably explained by the increase in cement gel mass encasing and penetrating wood particles (Oyagade, 1988; 1995). The occupation of the wood cell lumen by the cement gel could reduce the volume of the cell lumen available for water, and consequently, of the board as a whole.

The results of the analysis of variance for the test show that the materials to cement ratio and the interaction between the blending proportion and mixing ratio were strongly significant on water absorption (Table 2). Therefore, blending proportion is found to affect the result of WA obtained.

Table 2: ANOVA for Water Absorption (48-Hour)

The average thickness swelling after 48 - hour water immersion is presented in Table 1. The values obtained for thickness swelling (TS) ranged from 0.27 to 6.50%. The result presented compared adequately well with those reported in literatures: 1.8 to 3.1% by Geimer et al (1993), 0.98 to 3.62% by Badejo (1990) and Ajayi (2000). The values of thickness swelling decreased as the cement/mix ratio increase with increased blending proportion. Increase in proportion of sawdust to banana flakes improved the dimensional change of the boards. Boards produced at blending proportion of 50:50 and 75:25 appeared to be more stable and was enhanced by the combination of sawdust and banana flakes to produce a more dimensionally stable boards when compared with that obtained when ordinary banana flakes was used.

The result indicates that increase in blending proportion of sawdust to banana flake resulted in decreased thickness swelling. Thickness swelling (TS) values increase with increase in the flake content of board and decrease in the sawdust content. Similarly, thickness swelling (TS) values decreased at increasing level of cement to fibre ratio. Cement-bonded particleboards made at the blending proportion of 100: 0 of sawdust to banana flakes, mixing ratio of 3.0: 1.0 of cement to cellulosic materials produced the lowest thickness swelling (TS) value of 0.27%. The analysis of variance for the test shows that the blending proportions as well as interactions between different samples with mixing ratio are not significant at 5% level of probability (Table 3).

Table 3: ANOVA for Thickness Swelling (48 - Hours)

CONCLUSION

The results obtained from this study showed that board production is feasible from banana flakes when blended with sawdust. However, the use of the banana flakes without an inclusion of sawdust was found impossible. The study further revealed that blending proportion influences the dimensional stability. The values obtained for water absorption and thickness swelling following 48-hour water soak cycle ranged from 3.69 to 22.22% and 0. 27 - 6.50% respectively. These results indicate that increase in blending proportion and cement content resulted in the improvement of the dimensional stability of the boards. The result obtained from this study shows that board can be produced successfully from banana flakes when blended with sawdust at proportional level of sawdust/flakes of 50:50 and 75:25. Such boards could be used for internal construction, where there is no risk of prolonged wetting environment. This investigation is preliminary and further studies are therefore recommended.

REFERENCES

Ajayi, B. (1982). Investigation of the physical strength properties of maize stalk based particle - board. HND, Thesis, School of Forestry, For Res. Inst. of Nigeria.

Ajayi, B. (2000). Strength and dimensional stability of cement - bonded flake board produced from Gmelin arborea and Leucecina Leucocephala. A PhD Thesis, Department of Forestry and Wood technology pg 11 - 40.

Badejo, S. O. O. and Dada S. A. (1978): Effect of specific gravity and resin content on bending strength, water absorption and thickness swelling properties of baggase particleboard. Nigeria Journal of Forestry 8 (1&2):56-59.

Badejo, S. O.O. (1984). Effect of hot water and chemical pre- treatment of the physical and mechanical strength properties of wood- cement boards fabricated from three Nigerian plantation woods. Seminar series, VOL.1 NO. 142P FRIN, 1984.

Badejo, S. O. (1990): Sawmill Wood residues in Nigeria and their utilization. Invited paper, Proceeding of the National workshop on forestry management strategies of self-sufficiency in wood production. Forestry Research Institute of Nigeria, Ibadan, 12-15 July 1990.

Biblis, E.J. and Lo, C. (1988). Sugars and other wood extractives effect on setting of southern pine- cement mixture. Forest product Journal 6pq 28-34.

British Standard Institution: B.S. 5669:1979. Specification for wood chipboard and methods of test for particleboard BSI, London.

Biston -Werke, Co. (1981). Wood- cement board plants. Technical literature, spring 1, West Germany

Davis, T.C. (1966): Effect of blue stain on setting of excelsior-cement mixture. For. Prod. J. 16(6):49-50

Dinwoodie, J. M. and Paxton B. H. (1991): The long -term performance of cement-bonded wood particleboard. In proc. Inorganic bonded wood and fibre composite materials conference. For. Prod. Res. Soc. Madison W/S. pp45-54. In Moslemi, A. A. (ed.).

Fuwape, J. A. (1992) Sorption properties of Wood - Cement particles - board as influenced by cement/wood ratio. J. Ind. Acad. Wood Sci. 23(1): 1 -9.

Geimer, R. L.; Souza, M. R.; Moslemi, A. A. and Sumatupang N. H. (1993). Carbondioxide application for rapid production of cement - bonded particle board. Pp 31 - 41 in Moslemi, A. A. (Ed) Inorganic Bonded Wood and Fibre Composite Materials. Vol. 3 Forest Prod. Res. Soc. Madison, Wis Pp 32 - 34. In Moslemi A. A. (Ed).

Kollmann, F.F. P., Kuenzi, E. W. and Stamm, A. J. (1975): Principles of Wood Science and Technology, Vol.II. Wood-based materials. Springer Verlag Inc. Press, New York.

Lee, A. W. C. (1991). The latest development in the cement - bonded wood excelsior (wood wool) board industry. In. proc. Second international Inorganic bonded wood and Fibre composite Materials Forest Prod. Res. Sec. Madison. WIS. Pp. 103 - 107.

Maloney, T.M. (1977): Modern particleboard and dry - process fibreboard manufacturing. Miller freeman, San Francisco.

Moslemi, A.A. (1989): Wood - Cement panel products: Coming of age. In: Proc. Fibre and particleboards bonded with inorganic binders, forest prod. Res. Soc. Madison, Wis. Pp 12 - 18.

Omoluabi.A.B.(1982). Demand prospects for particleboard in Nigeria. M.sc. thesis, Department of forest, Res. Manag. University of Ibadan, Nigeria.

Oyagade, A. O. (1988). Thickness Swelling and water absorption of cement - bonded particle - board as influenced by three process variables. Journal of Forestry 18 (1 and 2): 20 -27.

Oyagade, A.O. (1990). Effect of cement/wood ratio on the relationship between cement-boarded particleboard density and bending properties. Journal of Tropical for. Sc. 7pp.

Oyagade, A. O. (1995). Influence of Portland Cement alkalinity on the dimensional stability of cement - wood particle - board. Nigerian Journal For. 24 (1 and 2) pp 21 - 25.

Sandermann, W. (1970). Technical processes for the manufacture of wood-wool cement boards and their adaptation for the utilization of agricultural wastes. UNIDO. ID/WG 82/4. 1970.

Simatupang, M. H.; Seddiq, N.; Habighorst, C. and Geimer, R. H. (1991). Technology for rapid production of Mineral - bonded wood composite boards. Proc. Second Int. Inorganic bonded wood and Fibre composite materials. For. Prod. Res. Soc. Madison. WIS. Pp 18 - 27. In Moslemi A. A. (Ed).

Simatupang, M. H., Schwarz, H. G. and Broker, F. W. (1976): Small scale plants for the manufacture of mineral-bonded wood composites. 8 ^th World Forestry Congress, FID-11/21-3, Jarkata, Indonesia, 16-28 Oct. 1978, Special mpaper.

Simatupang, M. H. (1987): Manufacturing process and durability of cement-bonded wood composites. Proc. 4 ^th Int. conf. on Durability of building materials and building components vol.1. Pergamon, Oxford, pp128-135.

Weatherwax R. C. and Tarkow, H. (1967): Effect of wood setting of Portland Cement: Decayed wood as an inhibitor. For. Prod. J. 17(7):30-32.

¹ Department of Forestry and Wood Technology,
Federal University of University, P.M.B 704, Akure