Adhesive bonding of acetylated aspen flakes Part 3: adhesion with isocyanates

Adhesive bonding of acetylated aspen flakes Part 3. Adhesion with isocyanates J.A. Youngquist and R.M. Rowell (USDA Forest Service, Forest Products Laboratory, USA) Flakeboards made from control flakes bonded with an isocyanate resin outperformed phenolic-bonded control boards in thickness swelling tests in both liquid water and water vapour. Less irreversible swelling occurred in isocyanate-bonded control boards as compared to phenolic-bonded control boards. Flakeboards made from acetylated flakes, using either a phenolic or isocyanate resin, swelled in thickness much slower and to a lesser extent in liquid water and water vapour tests compared to control boards. Isocyanate-bonded control and acetylated aspen flakeboards had higher internal bond and modulus of rupture values and approximately equal modulus of elasticity properties when compared to a similar series of boards bonded with a phenolic resin, as reported in Part 1 of this series. Key words: acetylation; isocyanate adhesives; internal bond; dimensional stability; strength; flakeboard In Part I of this series, we investigated the surface changes and hydrophobicity that result from the acetylation of wood flakes I. We also investigated adhesive penetration, using a phenolic resin, and strength properties of the resulting control (untreated) and acetylated flakeboards. We found that the phenolic resin did not penetrate the acetylated wood structure because of the hydrophobic nature of the acetylated wood. In Part 2, we investigated the effects of the addition of emulsifier to the phenolic resin to improve wettability and adhesive penetration 2. We found that adding water-to-oil type emulsifiers to the water-soluble phenolic resin did improve resin penetration and strength properties. The purpose of this research was to (1) investigate the use of an isocyanate adhesive, (2) determine the rate and extent of thickness swelling in liquid water and water vapour using both isocyanate and phenolic resins in control and acetylated flakeboards, and (3) determine the strength properties of isocyanate-bonded control and acetylated flakeboards. *The Forest Products Laboratory is maintained in cooperation with the University of Wisconsin. This article was written and prepared by US Government employees on official time, and it is therefore in the public domain and not subject to copyright. Experimental details Flake preparation and acetylation Oven-dry ring-cut aspen flakes were acetylated using a simple dip procedure using acetic anhydride and reacted at 120°C as described earlier 3. Flakes with acetyi weight gains of 17% (based on the original ovendry weight) were produced. Flakeboard production Control or acetylated flakes were sprayed with an isocyanate adhesive to give a resin solids content of 3% (based on oven-dried treated flakes) or 5% phenolformaldehyde for comparison. Flakes were handformed into 597 × 686 mm randomly oriented mats. A fully automatic, programmable particleboard press system with data collection capability was used. The press was accurately controlled through an electrical servoactivated pump, actuated by pressure- and position-monitoring transducers. Position was reproducible to within 0.51 mm under a no-load condition and to within 0.127 mm under pressure. Line pressure was controlled to within 68.95 kPa of the targeted values at pressures below 3.45 MPa and to 0 1 4 3 - 7 4 9 6 / 9 0 / 0 4 0 2 7 3 - 0 4 © 1990 Butterworth-Heinemann Ltd INT.J.ADHESlON AND ADHESIVES VOL. 10 NO. 4 OCTOBER 1990 273 within 206.84 kPa at pressures above 3.45 MPa. The press was time-base programmed to operate to specific press openings or pressures, or alternated between these two modes of control. Control boards were pressed to a m a x i m u m pressure of 4.14 MPa for 8 min at 185°C. Acetylated boards were pressed to a m a x i m u m pressure of 4.48 MPa for 7.5 min at 185°C. All boards had an approximate density of 640 kg m -3 and were trimmed to a final size of approximately 56 by 66 cm. Equilibrium moisture content Equilibrium moisture content (EMC) of control and acetylated flakes and flakeboards was determined by placing weighed, oven-dried flakes in constant humidity rooms at 30%, 65%, and 90% relative humidity (RH) and 27°C. After 21 days, flakes and flakeboards were reweighed and EMC determined. Duplicate tests were run, and values were averaged. Water swelling tests Each flakeboard specimen (51 X 51 mm) was placed in a 10 X 10 cm container, 5 cm deep. The container was on a flatbed micrometer for the thickness measurements. Water (25°C) was added to the container, and the thickness was recorded as a function of time. Measurements were taken every 5 min for the first hour, every hour for the first 6 h, then once a day for 5 days. All water and humidity tests were done in duplicate. Water soaking tests Cyclic water soaking tests were run on boards (51 X 51 mm) as previously described 4. Each of six cycles consisted of water soaking for 5 days followed by oven-drying at 105°C for 2 days. Thickness swelling was calculated as a percentage of the original ovendried thickness. Humidity cycles Flakeboard specimens (51 X 51 mm) were placed in a humidity room at 30% RH and 27°C. Thickness was determined after 21 days. The specimens were then placed in a humidity room at 90% RH and 27°C for another 21 days, whereafter thickness was determined. This procedure was repeated for a total of four cycles of 30% to 90% RH. The specimens were then oven-dried and thickness was measured. Changes in thickness were calculated as a percentage of the ori~nal ovendried thickness. Strength properties of flakeboards Static bending tests were conducted on board specimens (76 × 330 mm) according to ASTM standard D 1037, using a 300 m m span. Moduli of rupture (MOR) and elasticity (MOE) were determined 5. Internal bond tests (ASTM D 1037) were carried out both with specimens (51 X 51 mm) cut from flakeboards for this purpose only and from the ends of specimens broken in static bending. Results and discussion Because of the limited number of specimens per individual test or treatment level, no statistical analysis of the data was possible. The results presented here should be considered as indicative of trends, and a larger, statistically valid experiment should be done to confirm these results. The flakeboards described in this paper were made using procedures that were identical to those used to make the phenolic-bonded boards described in Part 1 of this series. Density profiles taken on both series of boards were essentially identical, and the plots of temperature rise, pressure rise, and board thickness, which were made as each board was made. indicated essentially no difference between the two experiments. The same press programme was used in both series. Table 1 shows the EMC of control and acetylated flakes and flakeboards. At 30% RH, control flakes show a slightly higher EMC than the isocyanate-bonded board and about 1% higher EMC than the phenolicbonded board. Acetylated flakes and boards made from either adhesive show the same EMC at 30% RH. At 65% RH, control flakes show about a 3% higher EM¢ than either isocyanate- or phenolic-bonded boards. The acetylated flakes have a slightly higher EMC than isocyanate-bonded boards and a slightly lower EMC than phenolic-bonded boards. At 90% RH, control flakes have a 4% higher EMC than isocyanate-bonded boards and a 2% higher EMC than phenolic-bonded boards. Acetylated flakes alone show a higher EMC than isocyanate-bonded boards but show a lower EMC than phenolic-bonded boards. These results show that the phenolic resin is more hygroscopic than the acetylated furnish. The isocyanate resin is less hygroscopic than the acetylated furnish. Table 1. Equilibrium moisture content of control and acetylated aspen flakes and flakeboards at various relative humidities Equilibrium moisture content (%) 30% RH Control Acetylated 65% RH Flakes Phenolic boa rd * Isocyanate boa rd * Flakes Phenolic boa rd* Isocyanate boa rd t Flakes Phenolic boa rd * Isocyanate boa rd + 4.9 2.0 3.8 2.1 4.4 2.2 11.1 5.2 8.6 5.9 8.5 4.7 21.5 10.7 19.0 13.9 17.0 8.9 *5% phenolic adhesive ~3% isocyanateadhesive 274 90% RH INT.J.ADHESION AND ADHESIVES OCTOBER 1990 '° I 60 / ~ I- 40 L 3~ i s o c y a n a t e r e s i n O Control A ~ . . . . ,_= , 5~, phenolic r e s i n A Control = • .... • Acety,atod / T _ 80 phenolic resin A Control ~" I I 3~ isocyanate r e s i n o Control tylated 70 _ v ~ 60 e>" 10 0 15 30 45 60 (min) 1 2 3 4 (h) 5 6 1 2 3 4 (day] Time > o C 40 ~ 30 20 Fig. 1 Rate of thickness swelling in liquid water of aspen flakeboard made from control and acetylated flakes. (ML89 5855) 10 Fig. 1 shows the rate and extent of thickness swelling in liquid water of isocyanate- and phenolicbonded flakeboards. Phenolic-bonded control boards swelled much faster and to a greater extent than isocyanate-bonded boards. Boards made from acetylated flakes swelled at the same rate and to approximately the same extent regardless of which adhesive was used. Results of the cyclic water soaking/oven-drying thickness swelling tests are shown in Fig. 2. The data show a large amount of irreversible swelling (caused by the release of residual compressive stresses imparted to the board during pressing) in both control boards, more in the phenolic-bonded board than in the isocyanate-bonded board. The isocyanate-bonded control board showed slightly more reversible swelling (normal swelling of the wood substance) than did the phenolic-bonded board. Acetylation greatly reduced both irreversible and reversible swelling in both phenolic- and isocyanatebonded boards. More irreversible swelling occurred in the isocyanate-bonded acetylated board than in the phenolic-bonded acetylated board, but reversible swelling was about the same for both types of bonded acetylated boards. The very harsh conditions of this test possibly resulted in both adhesive and wood failure during the course of the test. In cyclic humidity tests (Fig. 3), results similar to those in liquid water tests were obtained. Irreversible Table 2. 50 i e- 20 0 OD Wet OD Wet OD Wet OD Wet OD Wet OD Wet OD (i) (z) (3) Oven-dry/water (4) (s) (6) s o a k i n g cycles Fig. 2 Change in thickness of aspen flakeboards made from control and acetylated flakes in repeated water swelling test. {ML89 5856) swelling was greater in phenolic-bonded control boards than in isocyanate-bonded control boards. Under the conditions of that test, however, more reversible swelling was observed in phenolic-bonded control boards than in isocyanate-bonded control boards. Boards made from acetylated flakes bonded with phenolic or isocyanate resins showed similar results. Swelling after the four-cycle test was much less in acetylated boards as compared to control boards. Strength properties (internal bond (IB), MOR, MOE), for the control and acetylated boards bonded with the isocyanate adhesive are reported in Table 2. A reference comparison is also made to the American National Standards Institute (ANSI) specification for mat-formed wood particleboard category for the Type 2 Waferboard. The m values for both control and acetylated boards were approximately equal, and both substantially exceeded the ANSI published standard. The MOR and MOE values for the acetylated boards were 77% and 85% of the control values, respectively. Strength properties of aspen isocyanate-bonded flakeboards made from control and acetylated flakes Specimen Control Acetylated* ANSI standard A 208.1 + (Type 2-waferboard) Internal bond strength (kPa) Modulus of rupture (M Pa) Modulus of elasticity (M Pa) Mean Range Mean Range Mean Range 641 627 448-717 414-896 34.4 26.6 25.0-47.0 9.3-42.6 4475 3785 3874-5701 3194-4916 20.7 -- 3447 349 *Acetylated to a 16.2 weight percent gain of bonded chemical +American National Standards Institute. Mat-formed wood particleboard; ANSI A 208,1 ; New York, 1979 INT.J.ADHESION AND ADHESIVES OCTOBER 1990 275 36 34 3~ isocyanate resin O Control • Acetylated in Part 1 of this series), isocyanate-bonded boards had higher Ia and MOR values and approximately equal MOE properties. 5~ phenolic resin A Control • Acetylated 32 Conclusions 30 No processing or assembly differences were found between phenolic-resin-bonded boards made in Part 1 of this series and the isocyanate-bonded boards made for this phase of the research. In both liquid water and water vapour tests, isocyanate-bonded control boards outperformed phenolic-bonded control boards. Less irreversible swelling occurred in isocyanate-bonded control boards as compared to phenolic-bonded control boards. Flakeboards made from acetylated flakes swelled in thickness much more slowly and to a much lesser extent than control boards, regardless of the resin used to bond the boards, in either liquid water or water vapour tests. There was a slight improvement in bonding properties of isocyanate-bonded acetylated boards compared to phenolic-bonded acetylated boards. 28 26 24 ,¢ .u_ >. t. x~ i 22 20 18 = o o > 16 ._c • t. 14 12 10 References 0 OD 30 90 (1) 30 90 (2) 30 90 (3) 30 90 (q) 30 OD H u m i d i t y cycle (~) Fig. 3 Change in thickness of aspen fibreboards made from control and acetylated flakes at 27°C and 30% and 90% relative humidity. (ML89 5857) These percentages did not change, even when the lowest and the highest values were excluded from the averages. The MOR and MOE levels were about equal to or higher than the ANSI standard. When comparing strength properties reported here to those for flakeboards bonded with phenolic adhesives (reported 276 INT.J.ADHESION A N D ADHESIVES OCTOBER 1 9 9 0 I Rowell, R.M., Youngquist. J.A. and Sachs, I.B. 'Adhesive bonding of acetylated aspen flakes. Part 1. Surface changes, hydrophobicity. adhesive penetration, and strength' Int J Adhesion Adhesives 7 4 (1987) pp 183-188 2 Youngquist, J.A., Sachs, I.B. and Rowell, R.M. 'Adhesive bonding of acetylated aspen flakes. Part 2. Effects of emulsifiers on phenolic resin bonding' Int J Adhesion Adhesives 8 4 (1988) pp 197-200 3 Rowell, R.M., Tillman, A.-M. and Simonson, R. 'A simplified procedure for the acetylation of hardwood and softwood flakeboard production' J Wood Chem Technol 6 3 (1986)" pp 427-448 Rowell, R.M. and Ellis, W.D. 'Determination of dimensional stabilization of wood using the water-soak method' Wood Fiber 10 2 (1978) pp 104-111 ASTM 'Standard methods of evaluating the properties of wood-base fiber and particle panel materials' ASTM D 1037-38 (American Society for Testing and Materials, Philadelphia, PA, USA, 1982) Authors The authors are with the USDA Forest Service, Forest Products Laboratory, One Gifford Pinchot Drive, Madison, WI 53705-2398, USA.