An assessment of a pressure impregnated,full-penetration,fire resisting chemical treatment for Laminated Veneer Lumber

Bushfire or wildfire is one of two significant disturbance events occurring in forest ecosystems, the other being insect attacks. Over the last decade, there has been an increased frequency of bushfires occurring in Australia (Dutta et al. 2016). The Commonwealth Scientific and Industrial Research Organisation (2018; CSIRO) in Australia reports that between 1901 and 2011, 260 bushfires were recorded in Australia with an associated 825 known civilian and firefighter fatalities. In 2008, extreme bushfire-weather conditions resulted in Australia's highest ever loss of life from a bushfire. In what has been termed Black Saturday, 173 people died and 414 were injured as a result of the fires. It is estimated that 400 individual fires were recorded on the 7th February 2008. Following the Black Saturday fires, Australia introduced specific requirements for bushfire-resisting materials to be used in the construction of new and renovated dwellings located in areas declared as ‘bushfire prone’.

The Australian Standard 3959:2009, in particular Appendix F (in accordance with AS3837), stipulates a number of naturally ‘bushfire-resisting timbers’ that are able to withstand fire ember attacks in accordance with several specific categories designated as Bushfire Attack Levels (BAL). In many cases, timber used as an external cladding or timber subfloor systems (LVL, I-Joist and Hardwood) will not exceed BAL 29 conditions. BAL 29 is characterised by an increasing level of ember attack and burning debris ignited by windborne embers together with an increasing heat flux between 19 and 29 kW/m². AS3837 compliant timber(s) must either have inherent fire resistant properties (such as the natural resistance to fire found in high-density hardwoods) or can be treated with the application/impregnation of fire retardant chemicals and tested to the required standard.

To satisfy the requirements of a bushfire-resisting timber to BAL 29, samples (100 × 100 mm) of timber must pass two conditions imposed by the use of the Cone Calorimeter device (Babrauskas et al. 1992), as pictured in Figure 1. The first is that the material samples must not achieve a Maximum Heat Release Rate (HRR) of greater than 100 kW/m², in addition, it must not exceed an average of 60 kW/m² HRR for 10 minutes following ignition, when the material samples are exposed to an irradiance level of 25 kW/m². If the timber product is used in an external application, such as timber weatherboard cladding, then an accelerated weathering test is also required prior to testing. The weathering testing is conducted in accordance with the American Society for Testing and Materials (ASTM) D2898, standard consisting of two parts, method A and method B. Method A consists purely of a ‘rain test’, whereas method B includes exposure to ultraviolet radiation (White, 2009).

OPEN IN VIEWER

Figure 1

Cone Calorimeter Testing Apparatus. Adopted from http://www.doctorfire.com/cone on the 23rd of August 2011.

Timber products can be treated in order to improve their fire performance characteristics. Treatments delay the ignition of the timber under fire conditions and reduce the HRR in order to slow down the spread of flame (Eloma et al. 1997; White and Dietenberger, 2010). As detailed in Figure 2, the Heat Release Curve for standard Douglas-Fir is significantly reduced for a fire retardant treated sample.

OPEN IN VIEWER

Figure 2

Heat Release Curves for Untreated and Fire Retardant Treated Douglas-Fir. Adopted from White & Dietenberger, 2010. Fire safety of wood construction.

Timber degradation as a result of exposure to fire forms a char layer. The char layer results in insulating the timber mass beneath, reducing/slowing further degradation. In the case of structural timber systems, as the fire diminishes, the cross-section of the timber also reduces the load carrying capacity of the member, therefore, the amount of charring of the cross-section of any timber member is a major contributing factor in fire resistance, specifically in relation to structural load capacities (White and Dietenberger, 2010). The application of chemical-based fire retardant additives, in varying formulations and compositions, can provide a reduced burn rate due to the improved charring rate of timber in a fire situation.

The term ‘intumescent reaction’ is synonymous with a subcategory of fire retardant protection mechanisms and can be defined as, the formation of a carbonaceous foam cellular layer in concert with the concurrent release of incombustible gases when exposed to fire (White and Dietenberger, 2010). The carbonaceous foam and incombustible gases prevent the accessibility of oxygen to the fire, reducing flame spread and protecting the timber (Elomaa et al., 1997; LeVan, 1984; Wladyka-Przybylak, 2000). The application of the intumescent chemicals falls into one of two categories, full-penetration pressure impregnated treatment or coating applications (Eloma et al., 1997; LeVan, 1984). The most appropriate application method depends on the application/usage of the wood product and stage of the building process. For example, materials that are used in a green field [new] building application can be treated off-site and delivered to site for installation, however, the application of treatments for in-situ structures may best be applied using fire retarding paint or primer solutions.

As an alternative to concrete and steel structural materials used in bushfire-prone areas throughout Australia, the present study sought to assess the suitability of a full-penetration, pressure impregnated, intumescent fire-retardant treated Laminated Veneer Lumber (LVL). The present study assessed the material properties and bushfire resistance qualities (to BAL 29) of treated LVL for use in building environments not exposed to weathering conditions, such as for a subfloor application.

Method

Materials

Test materials comprised of LVL manufactured by Pacific Woodtech Inc. in Washington State, in the United States of America. The sample material characteristics are presented in Table 1.

OPEN IN VIEWER

Table 1

Laminated Veneer Lumber sample properties.

Veneer	Thickness (normal)	2.5–3.2 mm
	Species	Douglas-fir (Pseudotsauga menziesii)
	Grade	CD (Metriguard graded) 15,300 MPa
	Joints	Face scarf and overlap
	Moisture content	8–15%
Dimensional tolerances
	Length	±10 mm
	Depth	≤ 200 ± 1 mm
		≥ 201 ± 2 mm
	Thickness	-0, +2 mm (at 12% moisture content) for 35 mm only
		-0, +4 mm (at 12% moisture content) for 42, 58 and 75 mm
Density		Average 600 kg m3
Adhesive		Phenol formaldehyde (Type ‘A’, AS 2754.1)
Formaldehyde emissions class		E0 (AS/NZS 4357)

The material samples were manufactured to include a Glue-line Permatek M30 termite treatment achieving a H2S hazard class, which is used in all areas in Australia south of the Tropic of Capricorn. The glue-line treatment offers termite protection as well as other decay protections. The samples were not subjected to accelerated weathering conditions as the product was tested on the understanding that it would be installed in a covered construction environment, typically a subfloor bearer and joist system/solution.

Results

The results of the Cone Calorimeter test indicate that the samples submitted for testing under AS3959:2009 Appendix F, AS/NZS3837, meet the necessary performance criteria, see Table 2.

OPEN IN VIEWER

Table 2

Results of the cone calorimeter testing on post-treated LVL samples.

Material	Thermex-FR sample one	Thermex-FR sample two	Thermex-FR sample three	Mean value
Specimen number	FH4572-25-1	FH4572-25-2	FH4572-25-3	–
Time to sustained flaming (sec)	289	662	577	509.3
Observations	i	i	i	–
Test duration (sec)	3516*	3085*	3449*	3350
Mass remaining Mr (gm)	153.0	170.9	162.6	162.2
Mass pyrolysed %	42.3%	34.8%	41.8%	39.6%
Specimen weight lossb (kg/m2)	11.93	8.52	11.87	10.77
Specimen weight loss rateb (g/m2, s)	3.7	3.5	4.1	3.8
Heat release rate
Peak kW m2	54.6	43.3	50.5	49.5
Over 60s from ignition kW/m2	28.9	26.5	23.1	26.2
Over 180s from ignition kW/m2	34.2	34.3	38.5	35.7
Over 300s from ignition kW/m2	28.7	30.7	39.3	32.9
Over 600s from ignition kW/m2	30.0	26.8	41.7	32.8
Total heat releasec MJ/m2	92.3	60.2	108.2	86.9
Average specific extinction areac	15.8	26.3	0.00	14.0
Effective heat of combustionc MJ/Kg	7.3	5.8	8.2	7.1

The results of the material properties testing of the post-treated LVL samples are presented in Table 3.

OPEN IN VIEWER

Table 3

Results of the various testing methods for material property values for LVL post-fire retardant treatment.

Test	Test method (ASTM D)	Unit	Sample to control (Δ %)
Moisture content	4442-92-5	%	+ 4%
Specific gravity	–	10^6 psi	+ 2%
Edge bending MOE	5456-10 6.5.1	psi	−2%
Edge bending MOR	5456-10 6.5.1	psi	−12%
Tension parallel	5456-10 6.5.2	psi	−8%
Compression parallel	5456-10 6.5.3	psi	−9%
Edged compression perpendicular	5456-10 6.5.4	psi	−14%
Flat compression perpendicular	5456-10 6.5.4	psi	−13%
Edges shear	5456-10 6.5.5.1 (block)	psi	No change
Flat shear	AS/NZS 4063.1.2010 2.7	psi	+1%
Edge nail bearing (L-direction)	5456-10 A2.5	psi	+31%
Edge nail bearing (X-direction)	5456-10 A2.5	psi	+8%
Edge nail bearing (average)	–	psi	+20%
Edge bolt bearing (L-direction)	5456-10 A2.5	psi	+13%
Edge bolt bearing (X-direction)	5456-10 A2.5	psi	–12%
Edge bolt bearing (average)		psi	+3%

The results indicate the treatment process had little effect on the Modulus of Elasticity (MOE), or stiffness, however varying effects on the bending and compression values. Results indicate a conservative 20% decrease in the overall strength of the product, yet the stiffness rating remained intact. While many of the fastening/nailing properties evaluated indicate improved capabilities following treatment. Testing in accordance with ASTM E84-10b, Test Method for Surface Burning Characteristics of Materials, achieved the following results found in Table 4.

OPEN IN VIEWER

Table 4

Flame spread and smoke indexation classification.

Flame spread	Flame spread	Flame spread classification
	11	10
Smoke indexation	Smoke developed	Smoke development classification
	124	125

During the test, the surfaces of the samples ignited at approximately 93 seconds, the flame began to progress along the sample until it reached the end of the flame spread at 580 seconds. For reference purposes, the flame spread and smoke developed index falls within ‘Classification A’ of the International Building Code (2003) chapter 8 Interior Finishes s803 wall and ceiling finishes, NFPA 5000, chapter 10 Interior Finishes, s10.3 interior wall and ceiling finish testing and classification. Classification A is the best class in the scale.

Discussion

The present study sought to assess the suitability of LVL treated with a full-penetration, pressure impregnated, fire retardant as an alternative to concrete and steel for use in bushfire prone areas to BAL 29 in Australia. The present study provides evidence LVL treated with an intumescent fire retardant chemical complies with AS3959:2009, Appendix F ‘Fire Resistant Timber’. Therefore, its application for use in the construction of dwellings and other buildings in bushfire-prone areas up to and including BAL -29 categorisation is supported. Further, the LVL material properties remained fairly constant post-treatment. The strength degradation of the LVL was to be expected following pressure treatment. However, the slight increases in the fastener properties values was a welcome side effect to the treatment process. In particular, for the use of nail plates, which are commonly used in manufacturing in Australia.

The LVL samples achieved modest flame spread and a high smoke indexation. The purpose of this test is to determine the relative burning behaviour of the material by observing the flame spread along the specimen. Flame spread and smoke developed index scores are used to regulate the flame spread properties of all combustible building materials in Australia (Gardner and Thomson, 1987). While the flame spread indexation is suitable, a second comparative test should validate the reliability of the smoke indexation scores as the results are quite high compared to other products within the building materials classification. A more thorough investigation may uncover a discrepancy in the methods used (in North American/Canadian versus Australian Laboratories) for determining smoke indexation scores and a greater sample population should validate or contest this project's results.

A limitation of the current study included the lack of testing for a chemical interaction between the fire retardant and the active ingredient in the termite and decay protection treatment of the LVL. In addition, the author is uncertain of the long-term effect of the pressure-treating process on the glue bonding of the LVL. Future research may look at the chemical agents and examine any interaction effects between the agents. A potential aim could be to determine if fire retardant chemicals act as an effective termite treatment, in the absence of using active termiticide ingredients within the glue-line.

The totality of the results from this research culminates to reveal that LVL bearer and joist subfloor systems treated with Thermex-FR chemicals are suitable for use in bushfire-prone areas across Australia up to BAL-29.

Recommendations for industry;

For non-exposed products, such as the LVL tested within this paper, a duty of care exists for distributors and suppliers. The treated subfloor system should be sold as a complete package. In other words, the product should be sold as an entire ‘treated system’; avoiding the potential for non-treated product contamination.