Effects of wood-preservative treatments on mechanical properties of E-glass/phenolic pultruded composite reinforcement for wood. (Composites and Manufactured Products).
Laminated timbers reinforced with fiber-reinforced polymer (FRP) material were treated with common preservative chemicals including oilborne (organo- and organometallic) and acid- and amine-based waterborne treatments, and the effects on the mechanical properties of FRP material were investigated. ASTM D 3039 (longitudinal and transverse tensile) and D 2344 (short beam interlaminar shear) tests were used for mechanical characterization of preservative-treated E-glass/phenolic pultruded material. Although the longitudinal elastic modulus was unaffected, some longitudinal strength losses were recorded for waterborne treated FRP coupons. A simple model was used to compute the average fiber strength within preservative-treated FRP coupons. These results were supported by scanning electron and light microscopy analyses of single glass fibers taken from failed FRP coupons. The work also includes a discussion of property losses that occur in the presence of threshold preservative retention levels, and how these loss es affect material capacity reduction factors (knock-down factors) used in design criteria. Recommendations for use of FRP composites for wood reinforcement in exterior environments are also provided for civil engineers and the FRP-wood reinforcement industry.
Modern production of laminated structural timbers for exterior use dates back to the development of resorcinol and phenol-resorcinol adhesives about 60 years ago (Selbo and Gronvold 1958, Kuenzel et al. 1953, Selbo 1957; Truax et al. 1953). Although biological deterioration of wood members has always occurred, it was not until the late 1970s that the American Institute of Timber Construction (AITC) developed recommendations requiring that all exterior use laminated members be treated with preservatives (AITC 1998). Wood, including laminated timbers and wood/FRP laminates, when used in exposed outdoor environments, is often treated with wood preservatives to prevent deterioration from decay fungi, insect attack, and other environmental agents (Kunezel et al. 1953).
More recent studies have demonstrated that FRP reinforcement in the order of 1.1 percent can increase the allowable bending strength of glulam beams by greater than 60 percent (Dagher et al. 1998, 1996). These wood hybrids, with proven mechanical properties, hold the promise of improving structures to support longer spans and heavier loads not previously possible with wood-only composites. However, to achieve long-lasting service life in exterior environments, preservative treatments must be applied to these composites. When a "post treatment" (preservative application following composite fabrication) method is used to produce reinforced wood composites, the composite is exposed to the preservative chemical as well as to the vacuum-pressure treatment process, both of which produce stresses in the final product (AWPA 1999a, AITC 1998).
Most of the wood preservation literature focuses only on wood and wood products (Blankenhorn et al. 1999, Kilmer et al. 1998, Manbeck et al. 1995, Baileys et al. 1994, Kimmel et al. 1994, Shaffer et al. 1991). Almost no information is available on the treatment of FRP materials, or these materials combined with wood. The production of preservative-treated composite reinforced laminated timbers represents the latest stage of investigation and development in the structural wood products industry and there is considerable interest by both the wood preserving and composite reinforced wood hybrid industry in the development of wood preservative compatible FRP systems.
The objectives of this study were as follows:
1) Characterize the effects of wood-preservative treatments on the mechanical properties of FRP material used in wood reinforcement.
2) Determine residual property retention after exposure to preservative chemicals and treatment, as a basis to develop material capacity reduction factors and threshold concentrations for common preservatives.
3) Develop preservative treatment recommendations and identify compatible preservatives with composite systems (matrix and fibers) for the preservative/pressure treated FRP-wood glued laminated industry.
MATERIALS AND METHODS:
Five wood preservative systems (acidic and basic waterborne and oilborne [organometallic]) (Table 1) were tested over a range of concentrations to determine the effect on tensile and apparent interlaminar shear properties of FRP reinforcing materials. Basic modeling was then performed to calculate the average fiber strength of the preservative-treated FRP material.
Only one type of FRP material was used in this experiment: E-glass/phenolic pultruded composite. This FRP material, identified as K-1, was developed by the Advanced Engineered Wood Composites Center at the University of Maine (Dagher et al. 1998) and manufactured by Strongwell Corporation, South Chatfield, Minnesota. The FRP material, 0.75 oz./[ft..sup.2] (0.023 g/[cm.sup.2]), consisted of reinforced unidirectional (0 degrees) E-glass continuous fiber rovings, faced with an E-glass randomly oriented chopped strand mat (CSM). The short-fiber CSM was initially bonded with melamine resin binder to form a mat suitable for the pultrusion process. In the pultrusion process, the continuous fiber rovings were oriented in the core and integrated with exterior CSM layers using a phenolic resin matrix. A variation of the pultrusion process was applied in which the continuous rovings were impregnated in phenolic resin, while the CSM mats were pulled dry into the die. As a result of this processing scheme, a resinstarved surface layer that improves bonding to wood was produced. The corresponding average volume contents for the fabricated FRP plate were: [V.sub.f] = 54 percent; [V.sub.m] = 21 percent; and [V.sub.v] = 25 percent where [V.sub.f] = E-glass fiber volume fraction; [V.sub.m] = matrix volume fraction; [V.sub.v] = void volume fraction. The resulting high void content leads to an open structure that favors movement and diffusion of moisture within the FRP composite material, and potentially could enhance the penetration of preservatives with the attendant capacity for increased chemical attack on the FRP material.
DETERMINATION OF VOID CONTENT WITH IGNITION LOST TEST
Ignition testing was performed in accordance with ASTM D 2584 and D 2734 for Ignition Loss of Cured Reinforced Resins and Standard Test Methods for Void Content of Reinforced Plastics (ASTM 1994a, 1994b, 1991). The ignition mass loss is considered to be equivalent to the resin content of the sample (the small amount of volatiles, e.g., water, residual solvents, or sizing agents are ignored in this test). Confirmation of resin combustion was made using reflected light microscopy with a calibrated digital image analysis system. Thickness measurements were made on the surface CSM layer and the continuous fiber roving core, which were removed for measurement using a razor blade. An average of 15 measurements were taken on each sample. After ignition, the remaining fiber in the crucible was found to consist of the unidirectional continuous fiber roving core and the surface CSM layer. Both constituents were loose and had little to no cohesiveness, indicating that all the resin had been burned off.
MECHANICAL TEST METHODS
Two ASTM tests were used in this experiment:
1) ASTM D 3039 Standard Test Method for Tensile Properties of Polymer Matrix Composite Materials (ASTM 1995) modified for sample preparation and the inclusion of a post-treatment procedure (outlined later). The longitudinal tensile modulus ([E.sub.1]) and the longitudinal tensile strength ([F.sub.1t]) are two fiber-dominated composite properties. Conversely, the transverse tensile modulus ([E.sub.2]) is typically a matrix-dominated property. These composite properties have been used as performance indicators to characterize environmental effects on composite materials (Lopez-Anido and Wood 2001).
2) ASTM D 2344 Standard Test Method for Apparent Interlaminar Shear Strength of Parallel Fiber Composites by Short-Beam Method (ASTM 1984). The interlaminar shear strength (ILSS) of a fiber-reinforced composite is controlled by its matrix properties and the fiber-matrix interface properties. The apparent ILSS can be measured based on the short-beam test method according to ASTM D 2344. In this method, a composite specimen with a small span-to-depth (L/d) ratio is tested in three-point bending to induce the interlaminar shear mode of failure. The apparent shear strength is computed assuming a continuous parabolic shear stress distribution in the cross section, as predicted by elementary beam theory for homogeneous materials. However, it has been shown that the shear stress distribution is dominated by stress concentrations in the regions close to the loading nose and the supports (Muszynski et al. 2000). For this reason, the apparent shear strength from the short-beam test cannot be used for design data. In sp ite of these limitations, the short-beam test has become one of the most popular methods for determination of the interlaminar bond quality of composites due to the ease of specimen preparation and the simplicity of the experimental procedure. The ILSS of a composite laminate has been widely used as a performance indicator to assess the compatibility of fiber-matrix systems (fiber surface treatments), the effect of processing defects in the matrix (void content and micro cracks), and also environmental effects on composite materials (Lopez-Anido and Wood 2001).
Mechanical testing was conducted using an Instron 8801, 100 kN (22 kip) servo-hydraulic testing machine for the tensile tests, and an Instron 8001 9 kN (2 kip) electro-mechanical testing machine for the interlaminar shear tests. The test condition temperature ranged from 21[degrees] to 25[degrees]C (70[degrees] to 77[degrees]F) with 35 to 55 percent relative humidity (RH). A cross-head speed of 1.3 mm/min. (0.05 in./min.) was used for both tests. The tensile test specimens were 254 by 12.7 by 3.18 mm (10 by 0.5 by 0.125 in.) and the interlaminar shear sample size was 25.4 by 5.6 by 3.18 mm (1 by 0.22 by 0.125 in.). No tabbing was applied to the tensile test specimens, since hydraulic grips supplied sufficient frictional force without sample damage in preliminary tests.
For the samples treated with copper naphthanate preservative, a 5-minute high-strength epoxy adhesive was applied as a coating on both ends, covering the grip area to prevent surface CSM layer softening. This softening was observed in preliminary tests and was presumed to be due to a reaction of naphthenic acid with resin components of the FRP.
For the tensile test, 7 samples were used for each preservative treatment group including the untreated and carrier-treated (water, mineral spirits, diesel fuel) control and reference groups (128 samples total). For the interlaminar shear test, 144 samples were used with 9 samples in each treatment group including untreated, carrier control, and reference groups. The samples were cut using a fluid-cooled (Glasgrind[R] oil-free synthetic grinding fluid) diamond tip saw.
PRESSURE TREATMENT SCHEDULE
All pressure treatments were performed in a 3 by 0.5 m diameter (118 by 20 in. dia.) pressure treatment vessel. The pressure treatment schedule was the same for all preservative groups and included an initial vacuum of 635 mm Hg (25 in. Hg) for 10 minutes followed by a pressure of 1.034 MPa (150 psi.) for 15 minutes in a "full-cell" treatment (AWPA 1999b). The total contact time of the FRP samples with the solutions was approximately 60 minutes. Solvent-and waterborne preservative treatments (chromated copper arsenate [CCA-C], CDDC, copper naphthenate [CuNI) and their controls were treated at ambient temperature. Oilborne preservative treatments (creosote, pentachlorophenol [PCP]) and their controls were performed at 65 to 68[degrees]C (149 to 154[degrees]F).
Fixation period. -- A 3-day wet storage/fixation period of the samples was followed by air-drying and conditioning at 18[degrees]C (65[degrees]F) and 50 percent RH until the specimens reached equilibrium weight (15 days).
COMPUTATION OF LONGITUDINAL TENSILE STRENGTH ([F.sub.1T]) AND INTERLAMINAR SHEAR STRENGTH (ILSS)
The longitudinal tensile strength was calculated as:
[F.sub.1t] = [P.sub.max]/A 
where [P.sub.max] = ultimate tensile load prior to failure (lb.); A = average cross-sectional area ([in.sup.2]).
The interlaminar shear strength was calculated based on beam theory, as follows:
[S.sub.H] = 0.75 [P.sub.B]/b x d 
where [S.sub.H] = shear strength (psi); [P.sub.B] = ultimate applied transverse load prior to failure (lb.); b = width of specimen (in.); d = thickness of specimen (in.).
MODELING OF [[sigma].sub.fa] FOR FRP TREATED WITH WOOD PRESERVATIVES:
The average fiber strength ([[sigma].sub.fa]) was back-computed from the longitudinal strength of the FRP material according to the Rule of Mixtures (ROM), as suggested in Barbero (1998).
The ROM for computing the longitudinal elastic modulus is:
[E.sub.1] = [E.sub.f][V.sub.f]+[E.sub.m][V.sub.m] 
The property values and assumptions used in these computations are as follows: [E.sub.f] = 72.345 GPa for E-glass fiber; [E.sub.m] = 5.52 GPa for phenolic matrix. For the fiber roving core [V.sub.f] = 0.701 and [V.sub.m] = 0.124 were obtained from the ignition loss test (Table 2). Considering only the continuous roving core fibers (fibers oriented at 0 degrees) results in: [E.sub.1] = 72.345 (0.701) + 5.52 (0.124) = 51.4 GPa. This longitudinal elastic modulus is larger than the experimental value for the pultruded plate because the effect of the surface CSM layers is not considered.
Similarly, longitudinal tensile strength ([F.sub.1t]) values can be calculated based on the following ROM equation.
[F.sub.1t] = [[sigma].sub.fa][V.sub.f]+[[sigma].sub.m][V.sub.m] 
where [[sigma].sub.m] = [[sigma].sub.fa][E.sub.m]/[E.sub.f]
It is worth noting that for these pultruded FRP composite material results, (1-[V.sub.f])>[V.sub.m] due to the relatively high void content. Deriving [[sigma].sub.fa] from the [F.sub.1t] using the ROM equation, the following formula can be obtained.
[[sigma].sub.fa] = [F.sub.1t]/[[V.sub.f]+([E.sub.m]/[E.sub.f]) [V.sub.m]] 
Although the theoretical ultimate tensile strength value for unprocessed E-glass fibers is 3.45 GPa, the actual [[sigma].sub.fa] value is expected to be reduced by 50 percent or more during the production process (Barbero 1998).
In addition to average, standard deviation, and coefficient of variation values, SYSTAT Analysis of Variance (ANOVA) was also performed for longitudinal tensile strength ([E.sub.1t]), transverse tensile strength ([F.sub.2t]), ILSS, and longitudinal elastic modulus ([E.sub.1]) values. A pairwise probability test was also used for comparison of the treatments.
RESULTS AND DISCUSSION
The unidirectional continuous fiber roving core contained approximately 70.1 percent fiber volume fraction. This ratio is very close to [V.sub.fmax] = 78.5 percent, the theoretical maximum fiber volume fraction value, when fibers are arranged in a rectangular array (Barbero 1998).
Under the same treatment conditions, all specimens showed similar solution retentions. Small differences may be accounted for by the different densities of the preservative solutions used. As expected, the CDDC treatment resulted in the highest retention values because it is a dual treatment process. Due to differences in solution concentrations, the average active chemical retentions varied among the treatment groups. The highest active retention recorded was 259.36 kg/[m.sup.3] (16.21 pcf) for creosote when used as an undiluted solution (Table 3). For all treatments, the average solution uptake of 15.3 percent indicates that preservative treatments filled 85 percent of the void volume in FRP material, which highlights the importance of void volume content and its role in preservative uptake and retention.
LONGITUDINAL TENSILE, TRANSVERSE TENSILE, AND INTERLAMINAR SHEAR STRENGTH
The experimental results and related statistical analysis show that the longitudinal elastic modulus ([E.sub.1]) values of preservative-treated FRP coupons in the tensile test were not affected by the preservative chemical treatments (Fig. 1). These observations are in general agreement with findings from durability studies where E-glass epoxy FRP reinforcements for concrete were exposed to water and salt-water (Kshirsagar et al. 2000). However, the longitudinal tensile strength ([F.sub.lt]) data show statistically significant reductions in strength values of FRP treated with CDDC (27% reduction) and the high concentration of CCA (25% reduction) when compared to other treatments and the untreated controls (Tables 4 and 5). This can be explained by the extreme pH values of these preservative solutions (pH 1.5 to 2 for CCA and pH 11 to 12 for CDDC), and the likelihood of chemical attack on E-glass fibers in FRP composites by alkali and acid chemicals (Ranney and Parker 1995, Ramachandran et al. 1980). The chemi cals in CCA or CDDC solutions (Cr[O.sub.3], CuO, [As.sub.2][O.sub.5], monoethanolamine, alkylamine, and sodium dimethyldithiocarbamate) may oxidize or erode the glass fibers, causing reduction in tensile properties of the treated material. Light and scanning electron microscopy (SEM) analysis revealed spiral cracks and longitudinal fissures on some individual glass fibers taken from CCA- and CDDC-treated FRP coupons while untreated fibers had undamaged surfaces (Fig. 2). Treatment residues were present on the surface of fibers in some instances, but these did not appear to affect the integrity of the fibers. If spiral cracks occur on the surface of the glass with CCA or CDDC treatment, the resulting flaw will reduce fiber tensile strength as predicted by the well-known Griffith's formula. The surface failures would then promote weakening of the FRP. The corrosion of glass fibers in aqueous environments (acidic or basic) has been extensively documented in the literature (Kshirsagar et al. 2000; Hojo et al. 199 8, Nagae and Otsuka 1996, Fujii et al. 1993, Ehrenstein et al. 1990, Rodriguez 1987). Chemical attack mechanisms on glass surfaces have been identified as leaching and/or etching processes. Leaching can be described as selective removal of soluble contents via ion exchange and it is more common in acidic media. Etching, on the other hand, is known as a first order reaction that involves hydration followed by total dissolution of the glass. Etching is more common in alkaline media (Rodriguez 1987). Water, particularly at an elevated temperature, can weaken glass fibers (Lopez-Anido and Wood 2001, Ehrenstein et al. 1990) by etching them and by leaching out some of the glass constituents. This results in fissures or crevices, as observed in the SEM, which may lead to stress concentrations around the flaws reducing the average fiber strength. The reactions that occur between glass and solutions also depend on the chemical composition of the glass (Rodriguez 1987). Different types of glass fiber (S-glass or C-glas s), or other fibers produced from different materials (carbon, aramid, boron etc.), would be expected to be more or less resistant to degradation by the preservative solution components. However, the use of more resistant glass products may be cost-prohibitive, and E-glass is currently the predominant material used in FRP pultruded materials by industry.
The reduction in average fiber strength ([[sigma].sub.fa]), following exposure to preservative treatments is correlated (Eq. ) with a reduction in the longitudinal tensile strength of the composite ([F.sub.1t]). Equation  was applied to back-compute approximate ([[sigma].sub.fa]) values for treated and control groups by neglecting the effect of the surface CSM layers.
CCA did not cause any reduction in the ILSS of the FRP, indicating that no drastic effects on the phenolic matrix or fiber-matrix interface occurred (Figs. 3 and 4). However, the CDDC and creosote ILSS data showed a statistical reduction in ILSS, suggesting that the alkaline nature of CDDC and the aromatic nature of creosote may affect the matrix and/or fiber-matrix interface. It is possible that the fiber-matrix interface was weakened by removing or neutralizing the chemical coupling agent used to coat the E-glass fiber surface that promotes adhesion between the fibers and the matrix (Nagae and Otsuka 1996). Creosote accumulation on glass fiber surfaces (Fig. 5) suggests that creosote may penetrate the fiber/matrix interface and promote ILSS reduction. Since aromatic phenolic matrixes have previously been reported to be chemically inert to hydrocarbons in 25[degrees] and 93[degrees]C (77[degrees] and 199[degrees]F) environments (Bauccio 1994) and attack of the fiber was not observable by SEM, the likely expl anation is degradation of the fiber-matrix interface. Exposure of samples to the higher concentrations of CCA or CDDC did not cause any significant reduction of tensile strength at the 95 percent (p = 0.05) significance level; however, at the 90 percent (p = 0.10) level, a 19 percent reduction in the strength of the CCA-treated samples was observed.
Failure modes of tension coupons were determined visually and recorded for all specimens after the specimens failed according to ASTM D 3039. In general, the dominant feature of specimen failure was categorized as massive debonding and rupture of the glass fibers at the center of specimens, with fractured glass fibers protruding from the narrow (thickness) edge of the specimens (Table 6). Explosive failures (ASTM 1995, 1987) were observed in almost all cases in the untreated samples, the oilborne preservative treated samples, and the oilborne controls. The CCA- and CDDC-treated FRP coupons failed with longitudinal splitting and less explosive type of fractures. These later coupons had very brittle properties associated with the presence of flaws in the glass fiber reinforcement.
Following are some recommendations for civil engineers and the wood reinforcement industry.
The retention levels of wood preservatives utilized in this work simulated different exposure levels of FRP material to preservatives that would be used in industry for a range of treated wood applications. Our results indicate that a reduction in tensile strength and ILSS occurred in waterborne preservative treated FRP composites when either ground contact or marine retention levels (19% to 28% reduction, respectively) were used. These reductions should be taken into account in design calculations. The ground contact retention of CCA caused a non-significant 19 percent tensile reduction (p value = 0.081) (Table 7). Further study is needed, but these data indicate that CCA retentions, at least those at the marine exposure level, will cause significant strength loss of the E-glass/phenolic pultruded FRP. For CDDC treatments, however, increased retention did not cause a change in tensile strength. No statistically significant reduction in longitudinal tensile strength was recorded in the oilborne treated sample s, regardless of preservativde type (PCP in diesel fuel, Cu-N in mineral spirits, or creosote) or retention level. Therefore, oilborne treatments used in this experiment could be considered compatible with E-glass phenolic pultruded material. Creosote treatment, however, resulted in a statistically significant (10%) reduction in ILSS; other oilborne and CCA treatments had no effect on this mechanical property. These results indicate that the differing chemical composition of preservatives can affect different mechanical properties in FRPs, which should be taken into account in the design criteria for preservative-treated FRP-wood hybrids. Structural engineers should be aware of the effects of wood preservatives to allow the selection of compatible fiber and matrix systems. Compatibility tests with potential wood preservatives and different fiber and matrix systems for wood reinforcement should be performed in advance. Use of "rating values" to estimate the effects of preservative treatment on FRP strength pro perties can be developed; however, these would need to be corrected based on the environment where the material would ultimately be expose
Based on this work, the following issues have been identified that require further study:
1) Material capacity reduction factors for design specifications must be developed using broader retention levels for aboveground, ground-contact, and marine exposure, taking into consideration the short- and long-term strength loss effects of preservative treatments.
2) Threshold concentration limits, for effective treatment with minimal or no strength loss, must be incorporated into written recommendations for structural engineers and glulam beam manufacturers.
3) Guidelines should be prepared for pre- and post-preservative treatment schedules for the FRP-reinforced glulam beam industry.
[FIGURE 1 OMITTED]
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
TABLE 1 Type and percent concentrations (w/w) of preservatives used. (a) Preservative CCA-C (b) CDDC (b) Cu-N (c) PCP (d) Creostoe (e) (%) Low conc. 2.5 2.5 0.5 5 100 High conc. 10 5 8 10 100 (a)A range of preservative concentrations were used to target the aboveground, ground contact, and marine application retention levels for wood recommended by AWPA (AWPA-C14 and C28-99). CCA-C = chromated copper arsenate type C: arsenic acid 17.0 percent, chromic acid 23.75 percent, and copper oxide 9.25 percent; CDDC is a dual treatment process (Kodiak[R] ISK Biosciences, Memphis, TN.) The process consists of a monoethanolamine treatment followed by sodium dimethyldithiocarbamate treatment; Cu-N = copper naphthenate: naphthenic acid, copper salt 60 to 80 percent, mineral spirits 15 to 25 percent; PCP = pentachlorophenol: pentachlorophenol 90 to 94 percent, 2,3,4,6,-tetrachlorophenol 0 to 1.5 percent, hydroxypolychlorodibenzo ehters 4 to 7 percent; creosote = complex mixture of hydrocarbons 100 percent. (b)In distilled water carrier. (c)In mineral spirits carrier. (d)In diesel fuel carrier. (e)As original solution. TABLE 2 Fiber architecture of E-glass/phenolic FRP based on the Ignition Loss Test (ASTM D 2584). Ignition Fiber weight Sample ID Thickness loss fraction (mm) [in] (%) Plate (roving core + surface CSM) 3.34 14.19 85.81 (0.81 #) 0.48 * 0.48 * [0.131] 3.40 ** 0.56 ** Roving core layer only 2.19 6.98 93.02 (1.88 #) 0.38 * 0.38 * [0.086] 5.39 ** 0.40 ** Surface CSM layer only 0.59 -- -- (1.69 #) [0.023] Matrix weight Fiber volume Sample ID fraction fraction (%) Plate (roving core + surface CSM) 14.19 54.2 0.48 * 0.71 * 3.40 ** 1.31 ** Roving core layer only 6.98 70.1 0.38 * 0.34 * 5.39 ** 0.48 ** Surface CSM layer only -- 24.0 Matrix volume Void volume Sample ID fraction fraction (%) Plate (roving core + surface CSM) 21.0 24.7 0.61 * 0.46 * 2.89 ** 1.87 ** Roving core layer only 12.4 17.6 0.66 * 0.32 * 5.31 ** 1.82 ** Surface CSM layer only 37.7 38.3 (a)* = standard deviation of 6 specimens; ** = coefficient of variation (COV %) of 6 specimens; # = coefficient of variation (COV %) of 15 measurements. TABLE 3 Average solution and active ingredient retention values of pressure treated FRP specimens. Avg. Avg. Avg. Avg. solution solution ingredient weight Treatment uptake retention retention change (g/plate) (kg/ [m.sup.3] [pcf]) (%) Untreated -- -- 0  0 Distilled water 22.55 228.0 [14.25] 0  14 CCA-C 10% 24.60 229.6 [15.60] 24.96 [1.56] 15 CCA-C 2.5% 23.60 237.6 [14.85] 5.92 [0.37] 15 CDDC 2.5% 28.30 228.8 [17.68] 7.04 [0.44] 18 CDDC 5% 26.70 268.8 [16.80] 13.44 [0.84] 17 Cu-N 8% 21.30 212.8 [13.30] 16.96 [1.06] 13 Cu-N 2.5% 19.90 199.0 [12.44] 4.96 [0.31] 12 Cu-N 0.5% 21.40 214.0 [13.38] 1.12 [0.07] 13 Creosote 100% 26.15 259.3 [16.21] 259.36 [16.21] 17 PCP 5% 27.10 270.8 [16.93] 13.6 [0.85] 17 PCP 10% 25.00 251.6 [15.73] 25.12 [1.57] 16 Mineral spirits 19.30 194.2 [12.14] 0  12 Diesel fuel 21.20 213.4 [13.34] 0  13 Initial solution Treatment pH Untreated -- Distilled water -- CCA-C 10% 2 CCA-C 2.5% 2.5 CDDC 2.5% 11 CDDC 5% 11 Cu-N 8% N/A Cu-N 2.5% N/A Cu-N 0.5% N/A Creosote 100% N/A PCP 5% N/A PCP 10% N/A Mineral spirits N/A Diesel fuel N/A TABLE 4 Summary of statistically significant strength reduction in longitudinal, transverse tensile and interlaminar shear tests of preservative-treated E-glass/phenolic pultruded FRP. (a) Waterborne preservatives Oilborne preservati ves CCA CCA CDDC CDDC Cu-N Strength reduction 2.5% 10% 2.5% 5% 0.5% Long. tensile test B A A A C (ASTM D 3039) Trans. tensile test -- C -- C -- (ASTM D 3039) Interlaminar shear C C A A C (ASTM D 2344) Oilborne preservatives Cu-N Cu-N PCP PCP Strength reduction 2.5% 8% 5% 10% Creosote Long. tensile test C C C C C (ASTM D 3039) Trans. tensile test C -- C -- C (ASTM D 3039) Interlaminar shear C C C C A (ASTM D 2344) (a)A = statistically significant reduction at 95 percent confidence level (p-value is between 0.000 to 0.050); B = statistically significant significant reduction at 90 percent confidence level (p-value is between 0.000 to 0.100); C = no statistically significant reduction. TABLE 5 Average longitudinal tensile, interlaminar shear strength (ILSS), longitudinal tensile modulus values of FRP composite plate and [[sigma].sub.fa] "average fiber strength" of E-glass fibers. Composite long. E-glass fiber Composite Treatments tensile strength strength [[sigma].sub.fa] ILSS (MPa) Untreated 705.4 1264 27.06 Water 695.4 1246 26.38 CCA (10%) 530.9 951 26.36 CCA (2.5%) 571.5 1024 26.24 CuN (8%) 658.5 1180 26.76 CuN (2.5%) 632.9 1134 25.91 GuN (0.5%) 705.1 1263 26.64 Mineral spirits 651.4 1167 26.33 Diesel fuel 728.9 1306 25.55 Creosote 722.8 1295 24.71 PCP (10%) 672.0 1204 26.75 PCP (5%) 605.9 1086 26.73 CDDC (5%) 509.8 913 23.40 CDDC (2.5%) 506.7 908 22.98 Composite long. Treatments elastic modulus (GPa) Untreated 41.53 Water 42.54 CCA (10%) 41.05 CCA (2.5%) 41.31 CuN (8%) 41.51 CuN (2.5%) 41.53 GuN (0.5%) 41.37 Mineral spirits 41.60 Diesel fuel 41.01 Creosote 42.08 PCP (10%) 43.01 PCP (5%) 42.71 CDDC (5%) 40.72 CDDC (2.5%) 40.52 TABLE 6 Failure modes of the transversal tensile coupons according to ASTM D 3039. (a) ASTM Overall failure mode percentage (%) XGM 85 SGM 11 GAT 4 (a)XGM = explosive, gage, middle; SGM = longitudinal splitting, gage, middle; GAT = grip/tab, at grip/tab, middle. TABLE 7 Percent reductions in longitudinal tensile strength and ILSS tests. % reduction p-value Treatments for long, tensile strength CCA (2.5%) 19 0.081 CCA (10%) 25 0.004 CDDC (2.5%) 28 0.001 CDDC (5%) 28 0.000 Treatments for ILSS Creosote 9 0.000 CDDC (2.5%) 14 0.000 CDDC (5%) 15 0.000
American Institute of Timber Construction (AITC). 1998. Standard for preservative treatment of structural glued laminated timber. AITC 109-98. AITC, Englewood, CO. pp. 1-11.
American Society for Testing and Materials (ASTM). 1984. Standard test method for apparent interlaminar shear strength of parallel fiber composites by short beam method. ASTM D 2344-84. Annual Book of ASTM Standards. ASTM, West Conshohocken, PA. pp. 15-17.
_____. 1987. Standard terminology of high-modulus reinforcing fibers and their composites. ASTM D 3878-87. Annual Book of ASTM Standards. ASTM, West Conshohocken, PA. pp. 177-178.
_____. 1991. Standard test methods for density and specific gravity (relative density) of plastics by displacement. ASTM D 792-91. Annual Book of ASTM Standards. ASTM, West Conshohocken, PA. pp. 153-156.
_____. 1994a. Standard test methods for void content of reinforced plastics. ASTM D 2734-94. Annual Book of ASTM Standards. ASTM, West Conshohocken, PA. pp. 103-105.
_____. 1994b. Standard test method for ignition loss of cured reinforced resins. ASTM D 2584-94. Annual Book of ASTM Standards. ASTM, West Conshohocken, PA. pp. 81-82.
_____. 1995. Standard test method for tensile properties of polymer matrix composite materials. ASTM D 3039-95. Annual Book of ASTM Standards. ASTM, West Conshohocken, PA. pp. 111-121.
American Wood Preservers' Association (AWPA). 1999a. Standard for preservative treatment by pressure processes of structural glued laminated members and laminations before gluing. AWPA C28-99. AWPA, Granbury, TX. pp. 106-111.
_____. 1999b. Preservative pressure treatment of composite wood products. AWPA Use category system. AWPA, Granbury, TX. pp. 129-182.
Baileys, R.T., D.A. Webb, P.R. Blankenhom, J. J. Janowiak, P. Labosky, Jr., K.R. Kessler, W.R. Kilmer, H.B. Manbeck, and K.R. Schaffer. 1994. Hardwood glulam and creosote treatment - An evaluation of glue bond performance and structural strength characteristics, In: Proc. of AWPA. AWPA, Granbury, TX. pp. 1-8.
Bauccio, M. (ed.) 1994. Engineering Materials Reference Book. 2nd ed. ASM International, The Materials Information Soc., Materials Park, OH. pp. 353-355.
Blankenhorn, P.R., P. Labosky, Jr., J.J. Janowiak, H.B. Manbeck, D.A. Webb, and R.T. Baileys. 1999. The development of preservative treatment recommendations for red oak and red maple glued-laminated timber bridge members. Forest Prod. J. 49(5):87-93.
Barbero, E.J., 1998. Introduction to Composite Materials Design. Taylor & Francis, Philadelphia, PA. pp.78-99.
Dagher, H.J., T.E. Kimball, and S.M. Shaler. 1996. Effect of FRP-reinforcement on low grade eastern hemlock glulam. In: Proc. Of National Conf. on Wood Transportation Structures. M.A. Ritter, S.R. Duwadi, and P.D.H. Lee, eds, General. Tech. Rept. FPL-GTR-94. USDA Forest Serv., Forest Prod. Lab., Madison, WI.
_____. J. Poulin, B. Abdel-Magid, S.M. Shaler, W. Tjoelker, and B. Yeh. 1998. FRP reinforcement of Douglas-fir and western hemlock glulam beams, In: Proc. Inter. Composites Expo (ICE). pp. 1-3. Composites Institute, Harrison, NY.
Ehrenstein, G.W., A. Schmiemann, A. Bledzki, and R. Spaude. 1990. Handbook of Ceramics and Composites: Corrosion Phenomena in Glass-Fiber Reinforced Thermosetting Resins. N.P. Cheremisinoff, ed. Marcel Dekker Inc., New York, pp. 231-268.
Fujii, Y., A. Murakami, K. Katou, T. Yoshiki, Z. Maekawa, and H. Hamada. 1993. Durability of GFRP in corrosive environments. In: Proc. of the Third Inter. Offshore and Polar Engineering Conf. The Inter. Soc. of Offshore and Polar Engineers, Golden, CO. pp. 317-320.
Hojo, H., K. Tsuda, M. Kubouchi, and D-S. Kim. 1998. Corrosion of plastics and composites in chemical environments. Metals and Materials 4(6):1191-1197.
Kilmer, R.W., P.R. Blankenhorn, P. Labosky, and J.J. Janowiak. 1998. Laminating creosote-treated hardwoods. Wood and Fiber Sci. 30(2):175-184.
Kimmel, J.D., J.J. Janoviak, R.T. Baileys, and P. Merrick. 1994. Characteristics of creosote treated LVL materials. Forest Prod. J. 44(5):49-53.
Kshirsagar, S., R. Lopez-Anido, and R.K. Gupta. 2000. Environmental aging of fiber-reinforced polymer-wrapped concrete cylinders. ACt Materials J., American Concrete Inst. 97(6):703-712.
Kuenzel, J.G., N.V. Poletika, and H.B. McKean. 1953. The gluing of preservative-treated wood for severe service conditions. Forest Prod. J. 3(12):35-40.
Lopez-Anido, R. and K. Wood. 2001. Environmental exposure characterization of fiber reinforced polymer materials used in bridge deck systems. Res. Rept. No. AEWC 01-26. Advanced Engineered Wood Composites Center, Advanced Structures and Composites Lab., Univ. of Maine, Orono, ME.
Manbeck, H.B., K.R. Shaffer, J.J. Janowiak, P.R. Blankenhorn, P. Labosky, Jr., R.T. Baileys, and D.A. Webb. 1995. Creosote treatment effect on hardwood glulam beam properties. Wood and Fiber Sci. 27(3):239-249.
Muszynski, L., R. Lopez-Anido, and S.M. Shaler. 2000. Image correlation analysis applied to measurement of shear strains in laminated composites. In: Proc. SEM IX. Inter. Congress and Exposition on Experimental Mechanics. Soc. for Experimental Mechanics, Bethel, CT. pp. 163-166.
Nagae, Y. and Y. Otsuka. 1996. Effect of sizing agent on corrosion of glass fiber reinforced plastics (GFRP) in water. J. of Material Sci. Letters. 15(1):83-85.
Ramachandran, B.E., B.C. Pai, and N. Balasubramanian. 1980. Studies on the acid resistance of E-glass. J. of the American Ceramic Soc. 1(63):1-3.
Ranney, T.A. and L.V. Parker. 1995. Susceptibility of ABS. FEP, FRE, FRP, PTFE and PVC well casing to degradation by chemicals. Special Rept. 95-1. U.S. Army Corp. of Engineers, Cold Regions Research and Engineering Laboratory (CRREL), Hanover, NH. 11 pp.
Rodriguez, E.L. 1987. Corrosion of glass fibers. J. of Materials Sci. Letters 6(6):718-720.
Selbo, M.L. 1957. Laminating of preservative treated wood. In: AWPA Proc. 53:48-55.
_____ and O. Gronvold. 1958. Laminating preservative treated scotch pine. Forest Prod. J. 8(9):25-26.
Shaffer, K.R., H.B. Manbeck, P.R. Blankenhorn, J.J. Janowiak, P.L. Labosky. 1991. Structural performance of treated and untreated northern red oak glued laminated members. ASAE Pap. No.91-4658. Am, Soc. of Agricultural Engineers, St. Joseph, MI. pp. 1-23.
Truax, T.R., J.O. Blew, and M.L. Selbo. 1953. Production of preservative-treated laminated timbers. In: AWPA Proc. 49:113-123. AWPA, Granbury, TX.
The authors are, respectively, Graduate Research Assistant and Professor, Wood Sci. and Technology and Advanced Engineered Wood Composites Center, Univ. of Maine (UM), Orono, ME, 04473; and Assistant Professor, Dept. of Civil and Environmental Engineering and Advanced Engineered Wood Composites Center, UM. This is paper 2551 of the Maine Agriculture and Forest Expt. Sta. We thank the National Science Foundation EPSCORe program and the UM Wood Utilization Research program for support of this research. This paper was received for publication in October 2001. Reprint No. 9376.
CIHAT TASCIOGLU *
BARRY GOODELL *
ROBERTO LEPEZ-ANIDO *
* Forest Products Society Member.
[C] Forest Products Society 2002.
Forest Prod. J. 52(11/12):53-61.
|Printer friendly Cite/link Email Feedback|
|Author:||Tascioglu, Cihat; Goodell, Barry; Lopez-Anido, Roberto|
|Publication:||Forest Products Journal|
|Date:||Nov 1, 2002|
|Previous Article:||A new fungicide combination for prevention of kiln brown stain of white pine. (Solid Wood Products).|
|Next Article:||Effects of zinc borate on the properties of medium density fiberboard. (Composites and Manufactured Products).|