In this experimental study, rotary cut veneer sheets with dimensions of 0,2 ×50 × 50 cm thickness were obtained from freshly cut logs. While the alder and poplar veneers were manufactured from freshly cut logs, beech and Scots pine logs were steamed for 12 h before veneer production. The horizontal opening between knife and nosebar was 85 % of the veneer thickness, and the vertical opening was 0,5 mm in rotary cutting process. The veneers were then dried to 6-8 % moisture content with a veneer dryer. After drying, veneer sheets were treated with some fire retardant chemicals. For this aim, 5 % aqueous solutions of zinc borate, borax, monoammonium phosphate (MAP) and ammonium sulfate were used. The veneers were subjected to re-drying process at 110°C and then conditioned to 5-7 % moisture content again after they were immersed in the fire retardant solutions for 20 minutes. Twenty samples from each group were used to determine the level of retention. The retention level for each treatment solution was calculated with the following Equation 1, and they were presented in Table 2.

(1)

where; R is the retention level (kg/m³), G is the treatment solution absorbed by the sample in grams, C is the preservative or preservative solution in 100 g treatment solution, V is volume of sample in cm³.

Table 2:
Retention levels of fire retardant chemicals.

The colour measurements were carried out using a Minolta CM- 2600d spectrophotometer, equipped with an integrating sphere according to the CIE L*a*b* system (HAL 1996). Three samples from each group were used to determine the change of colour. Measurements were made over an 8 mm diameter spot with 10^o observer angle. The colour coordinates of each sample were measured before and after treatment of the veneer sheets at the same point. The ΔL*, Δa*, and Δb* colour coordinates, therefore total colour change (ΔE*) occurring in the specimens due to the treatment, were calculated on the tangential surface of replicates by comparing them to their initial values.

A fine stylus-type profilometer, Mitutoyo Surftest SJ-301 Surface Roughness Tester was used for the surface roughness test. Rz parameter (mean of the 10-point height of irregularities) was used to evaluate surface roughness of the samples according to DIN 4768 (1990). Ten samples at dimensions of 300×300×2 mm were used for each test group to evaluate surface roughness. Measurements were taken perpendicular to the grain.

The thermal conductivity of plywood was determined according to ASTM C-518 (2004). Two specimens with the dimensions of 300×300×2 mm were used for each group. The Lasercomp Fox-314 Heat Flow Meter shown in Figure 1 was used for the determination of thermal conductivity.

Figure 1:
Lasercomp Fox-314 heat flow meter.

The thermogravimetric analysis measurements were carried out using a TG/DTA 6300 (Seiko Instruments Inc. - SII NanoTechnology Inc., Japan). Before measurements, the samples of each group were powdered. About 5-10 mg of the wood powder was weighed in a platinum cup and scanned from room temperature (20°C) to 800°C at a heating rate of 20°C/min. The samples were measured in an atmosphere of nitrogen. The maximum thermal degradation rate was calculated as mass (%) at peak temperature divided by peak temperature.

The Rz (mean of the 10-point height of irregularities) and the ΔL*, Δa*, and Δb* colour coordinates, therefore total colour change (ΔE*) occurring in the specimens due to the treatment, were calculated and these values are shown in Table 3.

Table 3:
Surface properties changes of veneer sheets with treated different fire retardant chemicals.

According to Table 3, the fire retardant chemicals increased the surface roughness of the veneer sheets. The highest values of surface roughness were found for beech, alder and poplar veneers treated with borax and scots pine veneers treated with zinc borate. Ayrilmis et al. (2006) reported that fire retardant chemicals increased the surface roughness of veneer sheets. Dundar et al. (2008) also found that surface roughness of LVLs made from fire-retardant treated veneers was significantly increased as compared to untreated control specimens and this difference increased with increasing drying temperature. The second drying process performed after the impregnation may also have contributed to increase in surface roughness. Aydin (2004) indicated that the effects of fire retardant chemicals on surface roughness varied according to steaming pre-treatment and veneer drying temperature.

Positive and negative values show a tendency of wood surface to become reddish and greenish for Δa*, and yellowish and bluish for Δb*, respectively (HAL 1996). Treatment with zinc borate caused surfaces to become greenish and bluish in the all of veneer sheets. Borax, monoammonium phosphate and ammonium sulfate caused surfaces to become reddish and yellowish both for beech and scots pine veneers. In literature, Scots pine showed the yellowish and reddish hues due to impregnation (Simsek and Baysal 2012). In the alder veneers, borax and ammonium sulfate caused surfaces to become reddish and yellowish while monoammonium phosphate caused surfaces to become reddish and bluish. Borax caused surfaces to become greenish and bluish while monoammonium phosphate caused surfaces to become reddish and yellowish in the poplar veneers. Also, ammonium sulfate caused surfaces to become greenish and yellowish. Negative lightness stability (ΔL*) indicates a tendency of a wood surface to become darker (HAL 1996). Treatment with borax, monoammonium phosphate and ammonium sulfate caused surfaces to become darker in the all of veneer sheets except poplar. Simsek and Baysal (2012) found that borate treatments darkened both beech and Scots pine surfaces. In the poplar veneers, only monoammonium phosphate caused surfaces to become darker. The lowest darkening (-ΔL*) and total colour change (ΔE*) were observed with borax for beech and alder veneers. Aydin and Colakoglu (2005) stated that the reason for this could be the water-soluble extractive contents of beech and alder wood. In the poplar veneers, these values were obtained from monoammonium phosphate and zinc borate, respectively. They were obtained from ammonium sulfate and zinc borate in the Scots pine veneers, respectively.

According to wood species and fire retardant chemicals, the thermal conductivity mean values of veneer sheets are given in Figure 2.

Figure 2:
Thermal conductivity changes of the veneer sheets.

As can be seen from Figure 2, the highest values of thermal conductivity were obtained with ammonium sulfate for beech, poplar and scots pine veneers and monoammonium phosphate for alder veneers. When all test groups analyzed, untreated (control) groups were found to have the lowest thermal conductivity values. In the past studies, it was stated that thermal conductivity values of veneer sheets increased with impregnation, because the air in the cell lumen is replaced with a material having a better thermal conductivity (Kol et al. 2008). Another study suggests that fire retardants cause thermal conductivity to increase so that heat is absorbed by the chemical to prevent the wood surface from igniting (Rowell and Levan-Green 2005). Ozdemir et al. (2013) also found that thermal conductivity of high density fiberboard increased with using fire retardant chemicals and when the concentration of these chemicals increased, the thermal conductivity values increased. The differences among the thermal conductivity values in the study groups were very small. The reason for this is the low retention levels. Uysal et al. (2008) stated in their study that thermal conductivity values increased when retention levels were increased.

The temperature of the initial weight loss of pyrolysis (T_i), maximum degradation temperature (T_max), weight loss at maximum degradation temperature (W_max) and residual char at 800°C are given in Table 4. Figure 3 also show thermogravimetry curves according to wood species.

Figure 3:
Thermogravimetric curves of veneers treated fire retardant chemicals

Table 4:
The results of thermal degradation of the veneers treated the fire retardant chemicals.

The sharp decrease in thermogravimetric curves for the untreated control groups was seen between 180-380°C due to degradation of cellulose, hemicelluloses and lignin. Sinha et al. (2000) reported that the hemicelluloses decomposed easily to form gaseous products (CO, CO₂, condensable vapors, and etc.) between 200-280°C. Degradation of the hemicelluloses also causes the organic acid especially acetic acid formation. Acetic acid formation accelerates the decomposition of polysaccharides by means of acting as depolymerization catalyst (Brosse et al. 2010, Esteves and Pereira 2008). In temperature range between 250-300°C, the other main components of the wood products (lignin and cellulose) are degraded and gas, tar and char are produced during the pyrolysis (Pétrissans et al. 2014). In this study, untreated control groups gave similar results.

There are many theories that proposed to explain the mechanism of fire retardancy of chemicals. These theories are classified into five categories: Barrier theories, thermal theories, dilution by noncombustible gases theories, free radical trap theories and increased char/reduced volatiles theories. Generally, fire retardancy mechanism of the chemicals could operate by the combination of these theories (LeVan 1984). The most acceptable theory which is called “chemical theory” suggests that the retardant chemicals operate by altering the pyrolysis of wood, increasing of char amount and causing the formation of less volatile and combustible vapors (LeVan and Winandy 1990). Also, according to Kollman and Cote (1968) all known flame retardant chemicals increased charring of wood at relatively low temperatures, thus improving thermal insulation. In Table 4, it was seen that all of the fire retardant chemicals gave higher char amounts than untreated control panels. As stated by NIIR (2017), fire retardant chemicals lower the temperature at which pyrolysis occurs, directing the degradation pathway toward more char production and fewer volatiles. The pyrolysis reactions generally occur at low temperatures for the treated wood compared to the untreated wood (LeVan and Winandy 1990, Holmes 1977). According to the temperature of the initial weight loss of pyrolysis (T_i) and maximum degradation temperature (T_max) in Table 4, the pyrolysis of untreated veneer sheets generally began at higher temperatures than for treated veneer sheets.

The highest residual char amounts were obtained from ammonium sulfate for poplar and Scots pine; zinc borate for beech; borax for alder,. Boron materials as fire retardant chemicals were investigated by several researchers during the last decades (Tomak et al. 2012, Tondi et al. 2014, Salman et al. 2014, Yuksel et al. 2014, Istek et al. 2013, Roth et al. 2007, Jin et al. 2014, Qu et al. 2011). Parallel to the previous studies, our results showed that the impregnation of boron materials (Zinc borate and borax) enhanced the resistance against fire (Uner et al. 2016). Use of borates promotes the formation of CO and decrease CO₂ amount during the pyrolysis (Uysal and Ozciftci 2004). Gao et al. (2006) stated that boron material is protected with a physical mechanism; the melted boron material results in a film layer which inhibits the access of oxygen and temperature to material.

^♠Corresponding author: iaydin@ktu.edu.tr