The main objective of this study is to evaluate heat delamination characteristics of structural adhesives used for face bonding of cross-laminated timber, when subjecting both sides of CLT specimens to the small-scale flame test as required in the normative Annex A.2 of CSA O177 "Qualification Code for Manufacturers of Structural Glued-Laminated Timber" . While not technically in accordance with the CSA O177 Annex A.2 methodology, this flame exposure allows for a direct comparison to previous work performed on glulam specimens .
The long-term objective is to determine which currently-accepted test method allows for a better evaluation of heat delamination characteristics of adhesives used in structural engineered wood products, based on their actual end-use applications (e.g. bending, compression, combined stress, etc.) and their inherent manufacturing characteristics (e.g. cross-plies, etc.)
The objective of the current project is to develop a performance-based design process for wood-based design systems that would meet the objectives and functional statements set forth in the National Building Code of Canada.
More specifically, this report discusses the fire and seismic performance of buildings, as identified as a priority in a previous FPInnovations report.
The main objective of this study is to evaluate the heat release rate and fire growth contribution due to heat delamination characteristics of CLT manufactured with four types of adhesives used for face bonding, when exposed to a constant radiant heat flux. The evaluation is performed using the principles of ISO 5660-1 “Reaction-to-fire tests - Heat release, smoke production and mass loss rate – Part 1: Heat release rate (cone calorimeter method)”. The American version of this test method is ASTM E1354 « Standard Test Method for Heat and Visible Smoke Release Rates for Materials and Products Using an Oxygen Consumption Calorimeter ».
The long-term objective is to determine which currently accepted test methods allow for the most suitable evaluation of heat delamination characteristics of adhesives used in structural engineered wood products, based on their actual end-use applications (e.g. bending, compression, combined stress, cross-plies, etc.).
Transformative Technologies - Development of "Green" Wood Adhesives for Wood Composite Products
Chitosan is an amino polysaccharide obtained from the deacetylation of chitin, which is naturally occurring in the shells of a large number of marine crustaceans. Chitosan is soluble in weakly acidic aqueous solutions and possesses adhesive properties. Chitosan has received much attention for medical and industrial applications; however, only limited studies have been conducted on the application of chitosan as a wood adhesive, because its bonding properties on wood are poor. To improve the adhesive quality of chitosan resin, an innovative study on chitosan adhesives has been conducted to use selected fungal species to modify chitosan and improve its bonding properties, to synthesize non-formaldehyde resins with the fungus-modified chitosan, and to enhance urea-formaldehyde (UF) and phenol-formaldehyde (PF) resin performance with the fungus-modified chitosan.
The bonding properties of wood composites made with these chitosan-based green wood adhesives were significantly improved, in terms of lap-shear strength. Unmodified chitosan solution was not compatible with ammonium lignosulfonate, liquid PF resin, soybean resin, powder PF resin, or soybean flour, but was compatible with UF resin, polyvinyl acetate (PVA) resin, and phenol. With the addition of chitosan in UF and PVA resins, both the dry and wet shear strengths of plywood panels were improved, compared with those of panels bonded with the control UF and PVA resins, i.e. without chitosan. A number of chitosan and chitosan-reinforced UF resins were prepared as a binder for particleboard panel manufacturing. Six (6) types of particleboard panels with different levels of resin loadings and press conditions were manufactured. The resulting boards were tested to evaluate the bond quality of the chitosan and chitosan-reinforced UF resins. The test results showed that particleboard panels with good visual quality could be produced with all formulations of chitosan-UF adhesives, even with resin systems made with 1% of chitosan resin only. All chitosan resins used alone or added to UF resins yielded panels with better internal bond (IB) strength than those made with the UF control resin. The panels made with 1% chitosan resin plus 66% UF resin in a 1:1 ratio yielded panels with the highest IB strength and the best overall mechanical properties.
A round robin study using the CSA O112.9-10 standard was conducted with three (3) participating test laboratories in Canada as recommended by the Standards Council of Canada (SCC). The study included only the block shear test under three test conditions (dry, vacuum-pressure, and 9-cycle boil-dry-freeze). Test specimens were prepared using black spruce as substrate and phenol-resorcinol formaldehyde adhesive as the bonding agent.
All participating test laboratories passed the shear test of the CSA O112.9-10 standard in which the requirements are based on quantile parameters (median shear strength, median and lower quartile wood failure). This finding showed that the block shear test method of this adhesive standard was reproducible among three well-known test laboratories in Canada, all of which have been qualified by the SCC for the ISO 17025 accreditation program.
The test laboratories, however, exhibited some differences in terms of average shear strength and wood failure in the three test conditions. Some possible reasons were suggested for these differences.
Chitosan is an amino polysaccharide deacetylated from chitin, which is naturally occurring in large amount in shells of marine crustaceans. Chitosan is soluble in weakly acidic aqueous solutions and possesses an adhesive property. Chitosan has received much attention for medical and industrial applications; however, only limited studies have been conducted on the application of chitosan as a wood adhesive because of its bonding properties on wood are poor. To improve the adhesive quality of chitosan resin, an innovative study on chitosan adhesives has been conducted to use selected fungal species to modify chitosan and improve its bonding property, to synthesize non-formaldehyde resin with the fungus-modified chitosan and to prepare UF and PF resins enhanced with the fungal modified chitosan. Bonding properties of wood composites made with these chitosan-based green wood adhesives in terms of lap-shear strength were significantly improved in this study. Unmodified chitosan solution was not compatible with ammonium lignosulfonate liquid, liquid PF resin, soybean resin, PF powder, or soybean flour, but was compatible with UF resin (liquid), PVA resin, or phenol. With addition of chitosan in UF and PVA resins, both dry and wet shear strengths of plywood panels were improved comparing with the use of the control UF and PVA resins without chitosan. A number of chitosan and chitosan-reinforced UF resins as binder for particleboard manufacturing have been prepared. Six (6) types of particleboards with different levels of resin loadings and press conditions were manufactured and evaluated for the bond quality of chitosan and chitosan-reinforced UF resins. The results showed that all formulations of chitosan-UF adhesives were able to produce particleboards with nice appearance, even those made of only with 1% of chitosan resin alone. All chitosan resins, alone or added to UF resins, had a better IB strength than UF control resin. The panels made of 1% of chitosan resin plus 66% of UF resin in a 1:1 ratio had the highest IB strength.
The main objective of this project was to introduce NCC into the phenolic resin system as a reinforcing agent for improving the resin’s bond quality and durability in the manufacture of oriented strand board (OSB). The approaches adopted in this project can be outlined as follows: Develop a procedure or a new process technology to uniformly incorporate NCC into the phenolic resin system including phenol-formaldehyde (PF) and lignin-based PF resin; Develop new formulations for the NCC-phenolic adhesive system in both liquid and powder forms; Characterize NCC-PF and NCC-lignin-PF resins with differential analytical techniques; Manufacture OSB panels with NCC-PF and NCC-lignin-PF resins; Quantify the performance improvement of OSB panel by evaluating the resulting panel for physical and mechanical properties.
Ajoutez cet article à votre liste de sélections pour demander le PDF - Add this item to your selection list to request the PDF
The relationship between proof load level of fingerjoined lumber and degree of cure of adhesive bonds was investigated. Tension tests were completed for two different degrees of cure for two different adhesives. The proof load level determined for the partially cured joints did not cause damage to the joints that survived the proof test.
Preliminary guidelines for determining appropriate proof load levels for testing fingerjoined lumber with partially cured joints were proposed. The proposed guidelines will need to be validated through mill trials to demonstrate their efficacy and reliability to the manufacturer and third party inspection agency.
Keywords: fingerjoined lumber; tension proof testing/loading; partially cured adhesive bonds.
Lignin or lignin residue by-product can be produced from black liquor of pulping process or from cellulosic ethanol process. This project studied these lignin by-products as a substitute for corncob/Superbond fillers in PF glue mixes for plywood production. Two types of lignins were evaluated in the lab tests and one of them was used in the mill trial production. PF glue mixes with up to 50% corncob/Superbond substituted with lignin were produced and evaluated for plywood production. The test glue mixes were comparable to the control in viscosity, stability and bond performance. It is technically viable that lignin or lignin residue by-product be used as a substitute for corncob/Superbond fillers in the plywood industry. Caution must be taken to ensure that all particles of lignin pass a 100-mesh screen and the glue mix passes a 40-mesh screen in order for the glue mix to be smoothly transferred from the storage tank to the glue mix applicator.
Full-scale fire resistance tests on wood-frame wall assemblies carried out by the wood industry in 2006 demonstrated that end-joined (finger-joined) studs may not perform on par with equivalent solid-sawn studs, as had long been accepted by Canadian and U.S. building codes. The test results were found to vary significantly with the type of adhesive used, and more importantly, walls constructed with finger-joints made with some adhesives did not exhibit expected fire-resistance ratings. Although no problems have been reported with fire performance in the field, the wood industry quickly moved to address this issue in order to ensure the continued safety of wood structures constructed with these products.
In February 2007, the American Forest & Paper Association (AF&PA) released a qualification procedure for adhesives to be used in fire-rated assemblies that is based on full-scale ASTM E119 fire resistance tests. In 2008 this test procedure was separated into two methods and published as two ASTM standards. These ASTM procedures have since been adopted by both the American Lumber Standard Committee (ALSC) and the Canadian Lumber Standards Accreditation Board (CLSAB).
Although the full-scale test is now accepted as a methodology to qualify adhesives for use in fingerjoined HRA (heat resistant adhesive) lumber, there are obvious advantages to developing a small-scale test procedure for use as an alternative to the full-scale procedure. This project worked to develop a new small-scale test methodology which proved useful for quantifying the elevated-temperature performance of adhesives used in fingerjoined lumber.
The work in promoting a small-scale test methodology to replace the full-scale fire resistance test.
Robatech’s glue application system for finger-jointer consists in a pumping unit and application heads with multiple nozzles intended for jetting glue directly into machined finger profiles. The novel system was tested onto a CRP 2000 finger-jointing machine located at the FPInnovation – Forintek laboratory with three types of glue onto 2 X 4 SPF blocks.
For the first run test, the system was found fairly complicated to adjust, as much in terms of its positioning in reference to the finger profiles as for the amount of glue applied. The trial with standard PVA glue resulted with a too important quantity of glue unevenly applied. There were also issues with clogging as the injectors are air tight, but not the nozzles. The test with the modified PVA glue was interrupted shortly after beginning when the mixture turned into a foamy substance that could barely be sucked by the pumping unit.
The delaminating test that was conducted for this run had 100% of the joints failing performance criteria. Following these results, the bending test scheduled was cancelled.
In its initial format, the glue application system assembled by Robatech is not suitable for industrial use. The application head required some modification to be less tedious to adjust and the manufacturer was compelled to provide an elegant solution for the issues regarding clogging and application quality of the initial nozzle configuration.
In the second trial, the system was much easier to set-up and the glue application was achieved in a more controlled fashion. There was no clogging issue as the type of glue used (Franklin Advantage 405) was more permissive in term of curing time. The delamination and bending tests for the second trial run had both positive outcomes.
Following the results of the tests performed at FPInnovations with Robatech’s gluing system; the latter has proven to be a potential substitution for existing applicators. More testing with different glues and some improvements are be needed to fully proof the system. Robatech has proven to provide adequate solutions to issues following the first testing session and came up with more ideas as a result of the second trial run.
Nine structural adhesive formulations were selected to evaluate the effect of different curing methods on pH and alkalinity and/or acidity of adhesives. These included four phenol-formaldehyde (PF) resins with high pH, one phenol-resorcinol-formaldehyde (PRF) resin with intermediate pH, two melamine-urea-formaldehyde resins (MUF) with low pH and two melamine-formaldehyde (MF) resins with low pH. Four curing methods were used to prepare cured resin samples for the study: 1) curing at 102-105oC for one hour based on the CSA O112.6-1977 Standard; 2) curing for four hours at 66oC, followed by one hour at 150oC based on the ASTM D 1583-01 Standard; 3) curing at room temperature overnight based on the ASTM D 1583-01 Standard; and 4) cured adhesive collected from glue line squeezed-out from block shear assembly.
The effect of the curing method on pH of the cured adhesive strongly depended on the individual adhesives. For the PF, the alkalinity observed was different for each resin tested in the liquid form, while in the cured form, the difference in the alkalinity depended on the curing method. The MUF and MF were the most affected by the curing method, particularly the MUF, which showed much higher cured film pH values when tested by method 2 compared to the other three methods, while both the cured MF and MUF exhibited quite variable acidity values when measured with the different methods. The PRF showed reasonably uniform cured film pH but varying acidity values when measured with the different methods.
A reasonable relationship was observed between pH and alkalinity and between pH and acidity when the adhesives were considered as a group (i.e., adhesives of high pH as one group and adhesives of low pH as another group). Such a relationship was weaker when the adhesives were considered individually.
The use of a liquid sample in the determination of alkalinity/acidity of adhesives by titration was more convenient than using a cured film sample.
Full title: Impact of extreme pH of structural adhesives on bond durability as related to development and modification of CSA O112 wood adhesive standards. Part I. Investigation of different test methods for measuring pH, alkalinity and/or acidity of cured adhesive films and cured adhesives
Nine structural adhesives with varying pH were selected to examine the effect of pH on wood-adhesive bond quality. These included four high pH phenol-formaldehyde (PF), one intermediate pH phenol-resorcinol-formaldehyde (PRF), two low pH melamine-urea-formaldehyde (MUF), and two low pH melamine-formaldehyde (MF) adhesives. Block shear specimens were prepared with these adhesives using Douglas fir and black spruce. The adhesive performance was evaluated by measuring the shear properties (strength and wood failure) of the specimens tested at the dry and vacuum-pressure / re-dry (VPD) conditions.
Adhesive pH, test condition, and wood species showed significant effects on the shear properties. Douglas fir yielded about 40% higher shear strength at the dry condition compared to the VPD condition. Black spruce showed smaller difference in shear strength between the dry and the VPD conditions, the difference being only about 6%.
The different adhesives performed differently at the dry and VPD conditions. The high pH adhesives showed similar wood failures at both test conditions. On the other hand, the low pH adhesives showed high wood failure at the dry condition, but dropped significantly at the VPD condition for both species. This indicates that the low pH adhesives were less durable than the high pH adhesives.
Some correlation was observed between shear properties (strength and wood failure) and cured adhesive pH in the VPD condition, but not in the dry condition. Such a correlation was stronger in Douglas-fir than in black spruce.
Full title: Impact of extreme pH of structural adhesives on bond durability as related to development and modification of CSA O112 wood adhesive standards. Part II. Evaluation of block shear properties of selected wood adhesives by short term exposure test
This is a continuation of the short-term testing performed in Phase II of this study to determine the effects of adhesive pH on wood-adhesive bond durability. In this phase, the Douglas-fir block shear specimens prepared in Phase II using the nine structural adhesives, viz. four high pH phenol-formaldehyde (PF), one intermediate pH phenol-resorcinol-formaldehyde (PRF), two low pH melamine-urea-formaldehyde (MUF), and two low pH melamine-formaldehyde (MF), were tested periodically for up to 12 months under long-term vacuum-pressure / re-dry (VPD) condition. The VPD consisted of vacuum-pressure treatment followed by 0, 4, 8, and 12-month exposure durations at 50°C. The specimens were dried, in each exposure period, to their original moisture content prior to testing for shear strength and wood failure.
Indications of the extent of degradation of the wood layer, adjacent to the glue line due to pH during the long-term exposure, were also examined by the 1 % sodium hydroxide solubility test. The results indicated that the wood-layer samples closest to the glue line, which contained included-glue, showed higher solubility compared to those farther from the glue line. This suggests that wood degradation and/or potential glue decomposition occurred and is considered to be induced by the adhesive alkalinity or acidity under the long-term exposure conditions.
The PF showed the best durability performance followed by PRF and MF/MUF. The MF/MUF degraded completely after the 12-month exposure period.
For the PF, there are indications that some degree of degradation occurred in the wood layer adjacent to the bond line during the 12-month exposure period, which could be attributed to the high pH of the adhesives. This observation was not apparent for the PRF, and is considered inconclusive for the MF/MUF since they degraded during the exposure period.
Full title: Impact of extreme pH of structural adhesives on bond durability as related to development and modification of CSA O112 wood adhesive CSA standards. Part III. Evaluation of block shear properties of selected wood adhesives by long term exposure test
A continually growing number of jurisdictions in Canada and the United States are placing restrictions on the use of engineered wood products manufactured with structural adhesives. Since vertical-use-only fingerjoined wood studs manufactured with a polyvinyl-acetate (PVA) adhesive are one of the products of concern, Forintek’s L. Richardson was appointed to chair a special task group of the American Wood Council (AWC) Subcommittee on Fire Performance of Wood to determine what if any fire testing should be carried out and then to oversee those fire tests. In 2005, three Canadian manufacturers of fingerjoined studs requested that Forintek develop a strategy for assessing the substitutability/inter-changeability of SPS-3 vertical-use-only fingerjoined studs manufactured with polyvinyl-acetate (PVA) adhesives for solid-sawn wood studs in the construction of wood-frame loadbearing walls. Therefore, Forintek concluded that it should focus its efforts in 2005-2006 on assessing the fire performance of vertical-use-only fingerjoined wood studs manufactured with PVA adhesives. This change in direction for the project was supported by the project’s technical liaisons and mid-year revisions were made to reflect the change.
In March 2006, a standard ASTM E 119 fire test was conducted on a wall assembly constructed with nominal 2x4 No. 2 (and better) SPF SPS-3 fingerjoined studs manufactured with a PVA adhesive, one layer of 15.9-mm Type X gypsum board on each face and 90-mm thick batts of mineral-fibre insulation filling the stud cavities. The load applied to the assembly during the test was 100% of the maximum load permitted according to the National Design Specification® for Wood Construction. An identical wall constructed with solid-sawn Douglas fir studs had exhibited 71 minutes of fire resistance during tests carried out for AWC several years earlier at the same facility.
This wall assembly constructed with fingerjoined studs was unable to support the applied structural load after 49 minutes of exposure during the fire test. Structural failure occurred rapidly and was total. An examination of the wall following the test determined that the nearest fingerjoints in each stud to a horizontal joint in the gypsum board on the unexposed face about 1.8-m above the bottom of the wall separated and acted as a hinge thereby permitting the studs to fold over. There was no evidence of wood failure in the fingerjoints after they separated and the studs “jack-knifed”. Tensile strength tests on one fingerjoint that did not separate during the fire test indicated that the average peak stress at failure for specimens obtained from that fingerjoint were about 70 percent of the average peak stress at failure for fingerjoined specimens obtained from a stud that was not fire-tested. Also, in the case of the specimens obtained from one fingerjoint that did not separate during the fire test, the failure occurred mostly along the bondline surfaces of the joint with little wood failure. In the case of the specimens from a fingerjoined stud that had not been fire tested, there was a great deal of wood failure (nearly 100 percent).
During a series of telephone conferences to discuss the results of this fire test, Robert Glowinski, Vice-president of the American Forest & Paper Association (AF&PA) and a lawyer, and prior to becoming Vice-president of AF&PA, the AWC staff member responsible for fire issues, stated that because building codes in the United States were specifically revised at the request of AF&PA to permit inter-changeability/substitutability of vertical-use-only fingerjoined studs for solid sawn studs, to do nothing as a consequence of the results of this test is not an option. AF&PA has legal obligations, as do each of the manufacturers of these products. AF&PA and all manufacturers of these products have a legal obligation to move continuously and directly to address the issues raised by the results of the fire test and/or to make the necessary notification.
A Task Group consisting of representatives from the Southern Forest Products Association, Western Wood Products Association, West Coast Lumber Inspection Bureau, Canada’s National Lumber Grades Authority (NLGA), CWC, Forintek and Weyerhaeuser was appointed to work with AWC staff to identify and contact stakeholders in the United States and Canada and provide them with background information about the test. The Task Group was also directed to develop a brief communication document to inform all United States and Canadian stakeholders (manufacturers) and lumber grading agencies of the results of the fire test, its implication for these products within the context of current building codes, and to invite them to a meeting to quickly develop one or more study plans on how to proceed. At the stakeholders’ meeting, both technical and legal issues will be considered. Once the stakeholders have identified the components of the study plan(s), they will prioritize each component of the plan and identify how each is to be funded. Forintek will work with the AWC Task Group, NLGA, CWC and Canadian manufacturers of vertical-use-only fingerjoined studs in development and implementation of those study plans.
Waferboard/oriented strandboard (OSB) has been traditionally manufactured almost exclusively with trembling aspen in Canada. With the declining availability of aspen wood resource, OSB mills have begun to use alternative species in their production, usually other hardwood species. Meanwhile the mills have also begun experiencing some constraints in the use of phenol-formaldehyde (PF) resins for bonding these species, such as poor resin distribution and low retention on the surfaces of strands, particularly for a powdered form. The change of binder systems for bonding dense hardwoods can be extremely costly to OSB producers. The objective of this study is to determine the optimum adhesive and process requirements for manufacturing OSB from high-density hardwood furnish.
Study in 1997-1999 has shown that a dense wood is more difficult to bind than a less dense wood with a powder phenolic resin due to the poor resin distribution and retention. A liquid resin appears to produce stronger panels compared to a powder resin. The powder resin blending and bonding efficiency could be apparently improved by different resin application methods: (1) enhancing the wax distribution; (2) separately applying wax and resin to individual species strands and then mixing them up; (3) spraying a small amount of liquid additives after resin application; and (4) finally using small particle-size powders. A series of strandboards were constructed with aspen, birch, southern yellow pine, and sweetgum, using powder PF resins in the face and powder PF, liquid PF or diphenylmethane diisocyanate (MDI) in the core. An overall comparison showed that the aspen panels performed best followed by the southern yellow pine panel while the birch and sweetgum panels performed similarly with regard to both physical and mechanical properties.
Work in 1999-2000 focused on some fundamental studies on wood chemical and physical properties. Seven wood species were characterised for pH, base and buffering capacities, bound and soluble acids, and water and ethanol-toluene solubility. The wood species included aspen, white birch, yellow birch, red maple, sugar maple, southern yellow pine, and sweetgum. The study was also extended to the mixed wood species, included white/yellow birch, aspen/birch, southern yellow pine/sweetgum, aspen/red maple, aspen/sugar maple, and aspen/red maple/birch. This work indicated that there were significant differences in the chemical characteristics between the species investigated. These wood species were also characterised for surface roughness using a surface roughness tester. It was found that aspen strands showed significantly rougher surfaces than did southern yellow pine, sweetgum, and sugar maple. Strand surface characteristics seem to be related to the wood anatomical structure. A species (like aspen) having low density appears to yield a rougher surface than does one having high density (like sweetgum).
In the coming year, the efforts will focus on characterisation of wood/resin interaction, modification of phenolic resin, and optimization of panel manufacturing process parameters in order to more efficiently utilise various high-density hardwood furnishes for OSB production based on the information obtained in the previous studies of this project. The detailed information on project plan and milestone is illustrated in Appendixes 1 and 2.