Effect of cement/wood ratio on the properties of cement-bonded particleboard using CCA-treated wood removed from service. (Composites and Manufactured Products).
D. Pascal Kamdem (*)
There is growing concern about the environmental risks and difficulty indisposing of preservative-treated wood products. Landfill disposal is becoming less acceptable and alternatives must be explored. A possible approach to reusing this waste material is to incorporate it in cement-bonded particleboard (CBP). This paper investigates the effect of cement/wood ratio on the properties of CBP using particles from wood treated with chromated copper arsenate (CCA) and removed from service. A total of 35 CBPs were laboratory-manufactured using Portland cement and wood particles from CCA-treated red pine. The following mechanical and physical properties were evaluated at various cement/wood ratios, ranging from 1.0 to 4.0, at 0.5 increments: modulus of elasticity (MOE) and rupture (MOR), from bending strength, internal bond (IB) strength, thickness swelling (TS), water uptake (WU), and water absorption (WA). The results indicate that bending strength increases with the cement/wood ratio from 1.0 to 3.0, and decreas es thereafter above a cement/wood ratio of 3.0. IB strength increases with cement/wood ratio and peaks at a cement/wood ratio of 3.5. The CBPs show low levels of TS after 24-hour water soaking. The TS and WA decrease as the cement/wood ratio increases from 1.0 to 3.0, and then level off at cement/wood ratio above 3.0. The laboratory-made CBP using particles from incorporating CCA-treated wood showed optimum bending strength, IB strength, and dimensional stability at a cement/wood ratio of 3.0, a finding that is in agreement with previously published work on CBP made with particles from untreated wood. This study suggests that CCA-treated wood removed from service can be used for manufacturing CBP at a cement/wood ratio of 3.0.
Currently, the total volume of treated wood products removed from service is estimated at 9 million [m.sup.3] in the United States (17). The quantity of treated wood removed from service each year will increase dramatically in the future. It is predicted that the total volume will reach up to 18 million [m.sup.3] in 2020 and approximately 90 percent of the wood removed from service will have been treated with chromated copper arsenate (CCA). There is growing concern about the environmental impacts and increasing difficulty in disposing of treated wood products. Landfill disposal is becoming less acceptable and alternatives must be explored (17,29).
A possible approach to recycling this waste material is to incorporate it in wood-cement composites (8,9,26). Wood-cement composites have been attracting renewed interest because of their potential applications in the building industry (3). They have specific advantages over conventional resin-bonded composites, including good dimensional stability, acoustic insulation, fire resistance, and decay resistance (21,22,24,26,31). Schmidt et al. (26) found that wood particles from CCA-treated specimens and chromic acid pretreated particles produced stronger wood-cement composites compared to untreated wood. Huang and Cooper (9) obtained a similar result. In general, water-soluble sugars and extractives inhibit the hydration of cement. There are several possible mechanisms by which chromic acid can react with wood and then increase the wood-cement compatibility. Treatments of wood with low pH solution (1.5 to 2) such as CCA and chromic acid result in the removal of some wood hemicellulose and extractives that may o therwise inhibit the hydration of cement. CCA treatment has also been reported to increase the hydrophobicity of the wood surface (17). The oxidation of wood hydroxyl groups by hexavalent chromium increases the hydrophobicity of wood (5) and then influences its compatibility with cement. Zamorani et al. (33) suggested that the addition of chromium to cement results in the formation of a chromium hydroxide gel that may participate in the hydration process by increasing the permeability of unhydrated cement clinker to moisture. Huang and Cooper (9) reported that the amounts of copper and arsenic leached from the wood-cement composites were considerably lower than those from CCA-treated wood particles. Also, with an increase in water/cement ratio (ranging from 0.4/1 to 0.6/1), the leaching losses of copper, arsenic, and chromium were reduced, and mechanical and physical properties of boards were enhanced. Studies of stabilization of chromium in Portland cement (2,32,33) show that trivalent chromium forms insolub le and stable chromium-III hydroxides.
The relatively high density of wood-cement composites, largely from the addition of the high-density cement, has been a significant drawback that adversely affects the economics of cement-bonded boards as a commercial building component. These high-density boards are difficult to handle, cut, nail, and transport. The addition of low-density wood appears to be an important factor that may influence the development of wood-cement composite panels by reducing the overall density of the panels.
A low cement/wood ratio should yield a low-density panel, due to the lower density of wood particles (500 kg/[m.sup.3]) compared to that of cement. Wood particles are inexpensive compared to cement; adding high amounts of wood particles should reduce the cost of the board and the density as well. The economics of this cement-bonded particleboard (CBP) may become more favorable if cement/wood ratios can be reduced, with a minimum reduction of panel properties. Several studies have been reported on the effects of cement/wood ratio, particle geometry, particle treatment, wood species, and board density. (9-12,14,18,20,24). Information is limited regarding the mechanical and physical properties of CBP using CCA-treated wood removed from service at different cement/wood ratios. Huang and Cooper (9) examined the properties of CBP using particles from CCA-treated wood at cement/wood ratios of 2.0, 2.5, and 3.0. They stated that the strength and the dimensional stability of CBP increased with increasing cement/wood ratio.
The objective of this project was to study the influence of some processing parameters on the properties of CBP using CCA-treated wood removed from service. Specifically, this paper deals with the effect of cement/wood ratio on the mechanical properties and the dimensional stability of CBP containing particles from recycled CCA-treated wood.
MATERIALS AND METHODS
RAW MATERIAL COLLECTION AND CBP MANUFACTURE METHODS
CCA-treated red pine (Pinus resinosa Ait) poles retired after 15 years in service were collected in Wixom, Michigan. The poles were chipped with a Morbark Eager Beaver Chipper, and chips were reduced into particles with a laboratory hammer-mill. The average residual CCA retention in the particles was 8 kg/[m.sup.3] (19). The particles were sifted by size with a vibrating inclined screen, and only particles that passed through 10-mesh screens but were held by 16-mesh screens were selected for CBP production. Screened particles were air-dried to 6 [+ or -] 1 percent moisture content and used to manufacture CBP.
A commercial Portland cement (type I) was selected for this study. The cement/wood ratios on an ovendry weight basis were 1.0/1, 1.5/1, 2.0/1, 2.5/1, 3.0/1, 3.5/1, and 4.0/1. The cement/water ratio (weight basis) was 1/0.6. Wood particles, cement, and water were mixed in a blender for 5 minutes to produce a homogenous wood-cement-water mixture. The surfaces of caul plates were coated with mineral oil to reduce the adhesion of cement on the caul plates. The wood-cement mixture was spread consistently by hand on the caul plate on which a 305- by 305-mm forming frame was placed. The mat was pressed to a 10-mm-thick board using a hydraulic press under an initial pressure of 0.7 [+ or -] 0.1 MPa at room temperature. The pressure duration to reach the 10-mm thickness varied from 15 to 20 seconds, depending on the cement/wood ratios in the mat. The board was kept under 0.6 [+ or -] 0.1 MPa pressure for 24 hours in the hydraulic press. After 24 hours, the board was removed from the press and stored in a conditioning room maintained at 20[degrees]C and 100 percent relative humidity (RH) for 28 days to complete the curing of cement. Before testing, the CBP was conditioned at 20[degrees]C and 65 percent RH. Five CBPs were manufactured for each cement/wood ratio for a total of 35 boards. About 12-mm-wide edges of each board were trimmed off to remove the low-density and the poor-bonding area of the panel.
MECHANICAL AND PHYSICAL TESTING METHODS
For each cement/wood ratio, five specimens for three-point static bending tests, and five for internal bond (IB) strength, were cut from the CBPs according to ASTM D 1037-96a (1). Five specimens measuring 50.8 mm by 50.8 mm for testing thickness swelling (TS), water uptake (WU), and water absorption (WA) were prepared. All specimens were stored at a room conditioned at 20[degrees]C and 65 percent RH. Before testing, the weight and the size of each specimen were measured. Static bending and IB tests were conducted on a computercontrolled Model 4260 INSTRON machine. ForTS and WA tests, the specimens were submerged horizontally under 25 mm of water at 20 [+ or -] 2[degrees]C for 24 hours. After submersion, the specimens were suspended to drain for 10 minutes. The samples were blotted with tissue paper to remove the surface water. The weight and thickness was measured immediately (1).
RESULTS AND DISCUSSION
Tables 1 and 2 summarize the mean value and standard deviation of modulus of elasticity (MOE), modulus of rupture (MOR), IB, density, TS, WU, and WA for CBPs at cement/wood ratios ranging from 1.0 to 4.0, at 0.5 increments. The board density increases with the cement/wood ratio (Tables 1 and 2). The average values of density range from 815 to 1303 kg/[m.sup.3] for various cement/ wood ratios. This was thought to be mainly due to the content and the relatively high density of Portland cement (2860 kg/[m.sup.3]) compared to that of particles from CCA-treated red pine wood (700 kg/[m.sup.3]). The mean thickness of specimens ranged from 10.2 to 10.9 mm and was greater than the 10-mm target board thickness. This deviation from the initial target density was attributed to the springback of the wood particles.
Figures 1a and lb show that MOE and MOR increase with the cement/ wood ratio and then exhibit decreases at cement/wood ratio above 3.0. It can be seen in Table 1 that the mean MOEs of the CBPs vary from 1.67 to 7.95 GPa, which are comparable with those of resin-bonded particleboard (0.55 to 2.75 GPa) and oriented strandboard (4.83 to 8.27 GPa) listed in the literature (30). The mean values of MOR of the CBPs vary from 5.09 to 9.52 MPa, which are smaller than those of resin-bonded particleboard (5.0 to 23.5 MPa) and oriented strandboard (20.7 to 27.6 MPa) (30). The optimum values of MOE and MOR, 7.95 GPa and 9.52 MPa, respectively, are obtained with boards made at a cement/wood ratio of 3.0 (Table 1 and Figs. la and 1b). The same cement/ wood ratio is used in manufacturing commercial cement-bonded excelsior board (27). Ma et al. (15) studied the properties of bamboo-cement composites at cement/bamboo ratios ranging from 1.4 to 3.0. They found that the optimum values of MOE and MOR were obtained at a cement/wo od ratio of 2.6. Moslemi and Pfister (18) investigated the bending strength of wood-cement composite panels made of lodgepole pine at cement/wood ratios ranging from 1.5 to 3.0. They reported that MOE increases proportionally with increasing cement/wood ratio and exhibits a maximum at a ratio of 3.0. They also noticed that MOR increases as cement/wood ratio increases from 1.5 to 2.0 and then decreases as ratio increases from 2.0 to 3.0. Prestemon (24) indicated that wood-cement composite board made from wood slivers, sawdust, and cement has a higher strength when the cement/wood ratio increases from 0.75 to 1.5. Lee (13) reported that at a lower cement/wood ratio, wood excelsior in cement-bonded excelsior board does not receive adequate cement coating, resulting in poor bonding. At a higher cement/wood ratio, the compaction ratio (mat-to-board thickness ratio) is reduced, leading to lower bending strength. Moslemi and Pfister (18) explained that the increase of MOE with the increase of cement/wood ratio is be cause cement is inherently a more rigid material than wood. The presence of aggregate in concrete induces stress concentrations at the aggregate-cement interface. As wood particle volume increases, the regions of stress concentration around adjacent wood particles become more diffuse, resulting in an increased resistance to the stress applied. However, the amount of cement must be sufficient for a complete matrix formation.
Figure 1c shows IB strength of the CBPs at various cement/wood ratios. LB increases with increasing cement/wood ratio. IB values vary from 0.37 to 1.55 MPa, with the highest value at a cement/ wood ratio of 3.5 (Table 1). The TB of the CBPs in this study is comparable with that of resin-bonded particleboard (0.10 to 1.00 MPa) (30). With increasing cement/wood ratio, the wood particles are well coated with cement, which acts as an adhesive to bind the aggregates together. There are variations in TB values at a given cement/wood ratio as noted in the large standard deviation listed in Table 1. This may be due to the variations in density and wood particle distribution in the panel due to a poor mixing process.
The influence of the cement-wood ratio on the TS of the CBPs is illustrated in Figure id. The TS of the specimens after 24-hour water soaking decreases as cement/wood ratio increases from 1.0 to 2.5 and levels off at a ratio above 2.5. The average TS values vary from 0.51 to 0.98 percent at different cement/wood ratios (Table 2). They are remarkably smaller than those of the normal TS (5 to 20%) of resin-bonded particleboards after 24-hour water soaking. Lee (12) investigated the physical and mechanical properties of cement-bonded southern pine excelsior board with a cement/wood ratio of 2/1. The results indicated that a 48-hour water soaking produced 0.84 percent TS. TS is closely related to the TB strength and cement content of the boards. Great IB strength of the boards should lead to small TS. High adhesive content is known to have a significant effect in reducing TS and maintaining panel strength. Hydrated cement swells when it is exposed to high RH conditions (4,6,7,23,25,28). The swelling of hydrated cement may be attributed to the fact that prolonged moisture adsorption induces tensile stresses in inter-solid bond, disrupting some of the weak links and causing further adsorption (4). The swelling of cement block when going from dry to wet conditions is considerably smaller than solid wood. Fan et al. (4) indicated that the TS of commercial CBP was about 7.8 times that of laboratory-made cement paste. During water soaking, the swelling of wood and the compression recovery of wood particles in CBPs resulting from press operation are restrained to some extent by the cement coats on the particles.
Figures 1e and 1f indicated that WU and WA of the specimens after 24-hour water soaking decrease as cement/wood ratio increases from 1.0 to 3.0, and level off at the ratios above 3.0. WU and WA values vary from 2.14 to 7.17 g and from 6.9 to 27.8 percent, respectively, at different cement/wood ratios (Table 2). An increase in weight of CBP specimens during water soaking was attributed to increases of bound water (adsorbed by wood cell walls and cement) and free water that penetrates into the specimens and fills the internal voids within the specimens. With increasing cement/wood ratio, the density increases (Table 2) and the internal voids decrease, resulting in a smaller amount of free water in the boards.
A total of 35 CBPs were manufactured in the laboratory using Portland cement and CCA-treated red pine removed from service. MOE, MOR, IB strength, TS, WU, and WA of the CBPs at various cement/wood ratios were tested. MOE and MOR increase as cement/wood ratio increase from 1.0 to 3.0, and then decrease as the ratio increases from 3.0 to 4.0. A ratio of 3.0 was found to offer optimum bending strength. LB strength increases with increasing cement/wood ratio with the highest value at a ratio of 3.5. The CBPs show low levels of TS when subjected to 24-hour water soaking. TS, WU, and WA decrease as cement/wood ratio increases from 1.0 to 3.0 and then levels off at ratios above 3.0. The laboratory-made CBP, using particles from retired CCA-treated wood, showed optimum bending strength, LB strength, and dimensional stability at a cement/wood ratio of 3.0. This finding is in agreement with previously published work on CBP made with untreated wood. Further work regarding the effects of wood particle geometry and board density on the physical and mechanical properties of CBPs and the mobility of copper, chromium, and arsenate in CBPs is underway.
[Figure 1 omitted]
TABLE 1 Mechanical properties of CBPs at various cement/wood ratios. (a) Cement/wood ratio MOE MOR Density (b) IB (Gpa) (MPa) (kg/[m.sup.3]) (MPa) 1.0/1 1.67 (0.18) 5.09 (1.08) 907 (52) 0.37 (0.10) 1.5/1 3.10 (0.32) 7.98 (0.97) 935 (46) 0.62 (0.08) 2.0/1 4.50 (0.40) 8.45 (0.39) 1044 (39) 1.13 (0.33) 2.5/1 5.44 (0.80) 8.44 (1.29) 1125 (34) 0.95 (0.22) 3.0/1 7.95 (0.70) 9.52 (0.65) 1173 (26) 1.10 (0.19) 3.5/1 7.28 (0.31) 9.13 (0.29) 1208 (44) 1.55 (0.24) 4.0/1 5.81 (0.84) 8.92 (0.71) 1276 (31) 1.33 (0.37) Cement/wood ratio Density (c) (kg/[m.sup.3]) 1.0/1 815 (52) 1.5/1 888 (20) 2.0/1 1089 (54) 2.5/1 1147 (28) 3.0/1 1220 (25) 3.5/1 1286 (41) 4.0/1 1303 (33) (a)Value listed in parentheses is the standard deviation based on five specimens. (b)The density of the specimens for bending test. (c)The density of the specimens for IB test. TBALE 2 Thickness, TS, WU, Aand WA of CBPs at various cement/wood ratios. (a) Cement/wood ratio Thickness (b) TS WU WA (mm) (%) (g) (%) 1.0/1 10.8 (0.3) 0.98 (0.57) 7.17 (0.50) 27.8 (3.7) 1.5/1 10.9 (0.1) 0.72 (0.52) 4.30 (0.23) 16.8 (1.4) 2.0/1 10.5 (0.3) 0.60 (0.39) 3.93 (0.52) 14.0 (1.8) 2.5/1 10.4 (0.2) 0.51 (0.27) 3.50 (0.30) 12.0 (1.2) 3.0/1 10.2 (0.1) 0.53 (0.24) 2.14 (0.30) 6.9 (0.8) 3.5/1 10.2 (0.1) 0.55 (0.30) 3.14 (0.57) 10.0 (2.0) 4.0/1 10.2 (0.1) 0.53 (0.51) 2.70 (0.27) 8.3 (0.6) Cement/wood ratio Density (kg/[m.sup.3]) 1.0/1 946 (60) 1.5/1 928 (43) 2.0/1 1058 (59) 2.5/1 1108 (43) 3.0/1 1193 (52) 3.5/1 1213 (38) 4.0/1 1255 (37) (a)Value listed in parentheses is the standard deviation based on five specimens. (b)The thickness was measured at 20[degrees]C and 65 percent RH before 24-hour water soaking.
(*.)Forest Products Society Member.
[C]Forest Products Society 2002.
Forest Prod. J. 52(3):77-81.
(1.) American Society for Testing and Materials. 1999. Standard methods for evaluating properties of wood-based fiber and particle panel materials. ASTM D 1037-96a. Philadelphia, PA. pp. 139-168.
(2.) Cocke, D.L. 1990. The binding chemistry and leaching mechanisms of hazardous substances in cementitious solidification/stabilization systems. J. of Hazardous Material 24:231-253.
(3.) Dinwoodie, J.M. and B.M. Paxton. 1983. Wood cement particleboard: A technical assessment. Building Res. Establishment, Watford, England. 4/83:1-4.
(4.) Fan, M., J.M. Dinwoodie, P.W. Bonfield, and M.C. Breese. 1999. Dimensional instability of cement-bonded particleboard: Behavior of cement paste and its contribution to the composite. Wood and Fiber Sci. 31(3):306-318.
(5.) Feist, W.C. and W.D. Ellis. 1978. Fixation of hexavalent chromium on wood surface. Wood Sci. 1l(2):76-81.
(6.) Feldman, R.F., and P. J. Sereda. 1963. Use of compacts to study the sorption characteristics of powdered plaster of paris. J. Appl. Chem. 13:87-93.
(7.) _____ and _____. 1964. Sorption of water on compacts of bottle-hydrated cement. Part I. The sorption and length change isotherms. 3. Appl. Chem. 14:87-93.
(8.) Hsu, W.E. 1994. Cement bonded particle-board from recycled CCA-treated and virgin wood, In: Proc. Inorganic Bonded Wood and Fiber Composite Materials. A.A. Moslemi, ed. Univ. of Idaho, Moscow, ID. 4:3-5.
(9.) Huang, C., and P.A. Cooper. 2000. Cement-bonded particleboards using CCA-treated wood removed from service. Forest Prod. 3. 50(6):49-56.
(10.) Huffaker, E.M. 1962. Use of planer mill residues in wood-fiber concrete. Forest Prod. 3. 12(7):298-301.
(11.) Kayahara,M., K. Tajika., and H. Nakagawa. 1979. Increase of strength of wood-cement composites. Mokuzai Gakkaishi 25(8): 552-557.
(12.) Lee, A.W.C. 1984. Physical and mechanical properties of cement bonded southern pine excelsior board. Forest Prod. J. 34(4):30-34.
(13.) _____. 1985. Effect of cement/wood ratio on bending properties of cement bonded southern pine excelsior board. Wood and Fiber Sci. 17(3):361-364.
(15.) Ma, L., Y. Kuroki, W. Nagadom, Q.R. Pulido, S. Kawai, and H. Sasaki. 1998. Manufacture of bamboo-cement composites. Part IV. Effects of sodium silicate on cement curing by steam injection pressing. Mokuzai Gakkaishi 44(4):23t-312.
(16.) Maldas, D.C. and D.P. Kamdem. 1998. Surface tension and wettability of CCA-treated red maple. Wood and Fiber Sci. 30(4):368-373.
(17.) McQueen, J. and J. Stevens. 1998. Disposal of CCA-treated wood. Forest Prod. J. 48(11/12):86-90.
(18.) Moslemi, A.A. and S.C. Pfister. 1987. The influence of cement/wood ratio and cement type on bending strength and dimensional stability of wood-cement composite panels. Wood and Fiber Sci. 19(2):165-175.
(19.) Munson, J.M. and D.P. Kamdem. 1998, Reconstituted particleboards from CCA-treated red pine utility poles. Forest Prod. J. 48(3):55-62.
(20.) Namioka, Y., T. Takahashi, T. Anazawa, and M., Kitazawa. 1976. Studies on the manufacturing of wood-based cement boards. J. of Hokkaido Forest Prod. Inst. 65:87-141.
(21.) Parameswaran, V.N. and F.N. Broker. 1979. Micromorphological investigation on woodcement composites after long-term use. Holzforschung 33:97-102.
(22.) _____,_____, and M.H. Simatupang. 1977. Micromorphological studies on mineral-bonded wood composites. Interaction between binders and wood. Holzforschung 31:173-178.
(23.) Powers, T.C. and T.L. Brownyard. 1948. Studies of the physical properties of hardened Portland cement past. The PCA Res. Dev. Labs. 992 pp.
(24.) Prestemon, D.R. 1976. Preliminary evaluation of a wood-cement composite. Forest Prod. J. 26(2):43-45.
(25.) Resch, H.A. and D.H. Turk. 1967. The reversible and irreversible drying shrinkage of hydrated Portland cement and tricalcium silicate pastes. J. PCA Res. Dev. Labs. May 8-21.
(26.) Schmidt, R., R. Marsh, J.J. Balatinecz, and P.A. Cooper. 1994. Increased wood-cement compatibility of chromate-treated wood. Forest Prod. J. 44(7/8):44-46.
(27.) Semple, K. and P.D. Evans. 2000. Adverse effects of heartwood on the mechanical properties of wood-wool cement boards manufactured from radiata pine wood. Wood and Fiber Sci. 32(1):37-43.
(28.) Sereda, P.J. and R.F. Feldman. 1963. Compacts of powered materials as porous bodies for use in sorption studies. 3. Appl. Chem. 13:150-158.
(29.) Solo-Gabriele, H., V. Calitu, and M. Kormienko. 1999. Disposal of CCA-treated wood: An evaluation of existing and alternative management options. Rept. 99-6. Florida Center for Solid and Hazardous Waste Management, Gainesville, FL.
(30.) USDA Forest Service, Forest Products Laboratory. 1999. Wood Handbook: Wood as an Engineering Material. Forest Prod. Soc., Madison, WI.
(31.) Weatherwax, R.C. and H. Tarkow. 1964. Effect of wood setting of Portland cement. Forest Prod. 3. 14(12):567-570.
(32.) Wilk, C.M. 1997. Stabilization of heavy metals with Portland cement: Research synopsis. Waste Management Information, Portland Cement Assoc., Skokie, IL.
(33.) Zamorani, E., I.A. Sheikh, and G. Serrini. 1988. Physical properties measurements and leaching behavior of chromium compounds solidified in a cement matrix. Nuclear and Chemical Waste Management. 8:239-245.
The authors are, respectively, Research Associate and Associate Professor, Dept. of Forestry, Michigan State Univ., East Lansing, MI 48824. The authors wish to appreciate Dr. Sia Ravan, Dept. of Civil Engineering, Michigan State Univ. for his technical assistance. Financial support from the USDA Special Project No.99-34158-7838 is gratefully acknowledged. This paper was received for publication in January 2001. Reprint No. 9257.