EFFECTS OF HEAT TREATMENT ON THE ADHESION STRENGTH, PENDULUM HARDNESS, SURFACE ROUGHNESS,

The objective of this study was to investigate the surface properties of a UV-system applied on laminated parquet made with untreated and heat treated wood (ThermoWood). In this study, wood specimens prepared from Scots pine ( Pinus sylvestris ) wood were heat treated according to ThermoWood method at 190 o C for 2 hours and at 212 o C for 1 and 2 hours adhesion strength, pendulum hardness, surface roughness, colour and glossiness were determined. The UV-system was applied in two different types according to manufacturer recommendations. Results show that lightness and glossiness decreases and red colour tone increases with heat treatment. Pendulum hardness increased initially, decreasing afterwards with the intensity of the heat treatment. Tests showed that adhesion generally decreased with heat treatment. No significant differences were found for the surface roughness although a slight decrease was observed.


INTRODUCTION
Nowadays laminate flooring is preferred over solid wood due to its dimensional stability, lower cost and similar appearance.The flooring indoors is exposed to various conditions that can change color, brightness, surface roughness and hardness of laminate flooring surface.Testing the upper surface of laminate flooring is important to give us information for its utilization.One of the most important factors in laminated parquet production is the wood species selected for the outer layer.The most widely used wood species are American red oak, white oak, beech, pine, walnut, afrormosia, merbau, maple and iroko.Each species gives laminated parquet a characteristic color, gloss, surface roughness and hardness.
Heat treatment processes have been conquering a higher market share in the last years.One of the most successful processes is Thermowood®.The treatment is done with steam, with less than 3 to 5% oxygen without using pressure and with a minimum air speed of 10 m/s.The process begins with a rapid increase in temperature of the oven with heat and steam up to 100°C, followed by a gradual increase up to 130°C to near zero humidity.Then, heat treatment is carried at temperatures between 185°C and 230°C for 2 to 3 hours.This process is also one of the most studied processes in the last years (Korkut andAytin 2015, Moliński et al. 2016) along with other methods as for instance those based in heat treatment and vacuum like VAP HolzSysteme® (Batista et al. 2006a,b).
Color is the most studied surface property that changes with heat treatment.It is a well-known fact that wood becomes darker and at the same time lightness decreases as reported by many authors with different wood species (Bekhta and Niemz 2003, Chen et al. 2012, Dubey et al. 2012;Esteves et al. 2007, Mitsui et al. 2001, Sundqvist 2002).In the last few years gloss changes due to heat treatment have also been studied (Aksoy et al. 2011, Bekhta et al. 2014, Karamanoglu and Akyildiz 2013).
Wettability of wood surface decreases with heat treatments as reported by several authors (Hakkou et al. 2005, Pecina and Paprzycki 1988, Pétrissans et al. 2003, Dos Santos and Goncalves 2016).In accordance to Hakkou et al. (2005), that studied the wettability change with heat treatment for poplar, pine, spruce, and beech, this change could be due to a modification of the conformational arrangement of wood biopolymers resulting from the loss of residual water or, more probably, from the plasticization of lignin.This decrease in the surface wettability slows down the absorption of glues and varnishes and consequently reduces adhesion, which means that normal finishes are usually unfit to heat-treated wood (Sernek et al. 2007).Nevertheless there are varnishes and glues that can be adapted for this type of wood.
Hardness is a very important property in flooring, especially in the outer layer, and the results reported before prove that hardness changes depends on wood species, conditions of treatment and even with the direction of the tests (Shi et al. 2007).For instance Poncsak et al. (2006) studied heattreated birch and mentioned a slight hardness increase while Korkut et al. (2008b) concluded that janka-hardness decreased with heat-treatment for Scots pine wood.Boonstra et al. (2007) reported that Brinell hardness parallel to the grain increased significantly (48%) and hardness perpendicular to the grain increased slightly (5%) for spruce.Korkut and Guller (2008) studied the effects of heat treatment on the physical properties and surface roughness of red-bud maple and concluded that surface roughness decreased with increasing temperature treatment and treatment times.(Korkut et al. 2013) reported the effect of heat treatment on surface properties of wild cherry.These authors concluded that glossiness and surface roughness decreased with heat treatment compared to those of control specimens.Similar results were reported for heat treated red river gum tree (Eucalyptus camaldulensis) (Unsal and Ayrilmis 2005), black pine (Pinus nigra) (Gunduz et al. 2008) and turkish Hazel (Corylus colurna) (Korkut et al. 2008a).
Effects of heat treatment on the..: Gurleyen et al.

Heat Treatment Process
The heat treatment was performed according to ThermoWood® process at 190°C for 2 h, 212°C for 1 h and 2 h in a private commercial Novawood Factory in Gerede -Bolu, Turkey.Heat treated and untreated wood samples were conditioned to 12% equilibrium moisture content in a special room at 20 ± 2°C and 65 ± 5% relative humidity (ISO 554, 1976).

Laminated Parquet Flooring Material
In this study, laminated parquet flooring made with Scots pine (Pinus sylvestris L.), produced by KPS Company was selected for the tests.The production methods of laminated parquet flooring were shown in Figure 1.Five samples were prepared for each test and 6 measurements (5 x 6 = A total of 30 measurements) were made in each sample.The tangential surface of samples measuring 100 mm by 100 mm by 20 mm were used for all measurements.Wood colorant paint was used on the experimental samples.All laminated parquet flooring materials samples were conditioned to 12% MC in a conditioning room at 20°C (± 2) and with 65% (± 5) RH (ISO 554, 1976).The equilibrium MC of the samples was roughly 12 per cent after conditioning.
A43-0646 -UV sanding sealer is a type of varnish consisting of epoxy acrylic resin and ultraviolet ray curing sealers with solidity of (wt %) 95-97, and density of (20°C, g/cm 3 ) 1:15 to 1:20.N93-0910 nanolacke UV matt varnish is a type of varnish consisting of polyacrylic-based resin, nano-containing minerals, nanocomposites ultra violet curing (UV) varnish with solidity of (wt %) 95-100, and density of (20°C, g/cm 3 ) 1:09 to 1:15.Both of these varnishes are transparent and their application field are solid hardwood, chipboard and similar types of wooden materials.

Glossiness and Color Measurement
Glossiness measurements in coated wood with one and two layers were made in accordance to ISO 2813ISO (1994) ) in a Novo-Gloss Trio (Rhopoint Instruments Ltd., UK.).The measurements were made in perpendicular and parallel to the grain directions at an angle of 60 o (Figure 2a).The color change of laminated parquet produced from heat treated and untreated Scots pine, coated with one and two layers of finishing were analysed by a The Datacolor 110 (Wavelength resolution 10 nm, measurement geometry D/8 o ) with a D65 standard illuminant.Color parameters were measured using thirty replicates of each sample and an average value was reported.The CIELAB system characterized by three parameters, L*, a*, and b* was used.The L* axis represents the lightness, +a* is the red, minus a* for green, +b* for yellow, minus b* for blue, and L* varies from 100 (white) to zero (black) (Zhang et al. 2009).Total color difference (ΔE*) was calculated using Equation 1. (Figure 2b).A total of 480 data were measured for glossiness tests (Heat treatment 4 x layer 2 x measuring direction 2 x N 30= 480 measurement).The 960 data were measured for color tests (Heat treatment 4 x layer 2 x color difference 4 x N 30 = 960 measurement).

Surface Roughness
Surface roughness of all the samples was measured using a Mitutoyo Surftest SJ-301 device.Surface roughness of the samples was measured by using a stylus-type profilometer in Figure 3 (Mitutoyo 2015).In this device; R a , Ry, R z and R q test were measured according to ISO 4287 (1997).Four roughness parameters were determined: mean arithmetic deviation of profile (R a ), mean peak to-valley height (R z ), root mean square roughness (R q ), and maximum roughness (R y ).These parameters were commonly used in previous studies to evaluate surface characteristics of wood and wood composites such as veneer (Stumbo 1963).Roughness values were measured with a sensitivity of 0,5µm.The length of scanning line (Lt) was 15mm and the cut-off was λ=2,5mm.Specification of roughness parameters is described by Hiziroglu (1996), Hiziroglu and Graham (1998).

Adhesion Strength Test
In the study, the adhesion strength was determined in accordance to ASTM D-4541 (1995) in 1 ton (10 kN) ALSA electromechanical universal testing machine (Figure 4).404 plastic steel epoxy strong adhesive was used (Ayata 2014).The steel test cylinders with Ø 20 mm were attached to the sample surfaces at room temperature (~ 20ºC) via the help of a cast system (Demirci et al. 2013).Glued samples were then fixed by means of tools.All sample specimens were expected to dry for 24 hours.Glue residues were removed with a cutter.
The adhesion strength (X) was calculated in terms of MPa using the equation below (Budakci 2003).Effects of heat treatment on the..: Gurleyen et al.
A total of 480 data were measured for adhesion strength tests (Heat treatment 4 x Layer 2 x N 10 = 80 measurement).

Pendulum Hardness
Conditioned samples were subsequently subjected to the König pendulum hardness test to detect the hardness of the varnish coating according to ASTM D 4366-95 (1984).Test panels were placed on the panel table and a pendulum was gently placed on the panel surface.The pendulum was then deflected through 6° and released while simultaneously starting the oscillation counter.The number of oscillations for the amplitude to decrease from 6° to 3° was determined to be the König hardness.Thirty replications were conducted on separate specimens for each treatment group (Figure 5) (Cakicier et al. 2011a andCakicier at al. 2011b).Devices in the sample surface (63 ± 3,3 in HRC hardness and 5 ± 0,0005 mm in diameter) determine the hardness layer according to the oscillating pendulum swing with two balls (Ayata 2014).A total of 240 data were measured for pendulum hardness tests (Heat treatment 4 x Layer 2 x N 30 = 240 measurement).

Statistical Analysis
In the experiments, statistical analysis was performed for a total of 2720 data (Glossiness 480 + color 960 + surface roughness 960 + adhesion strength 80 + pendulum hardness 240 = 2720 measurement).Statistical evaluations were analyzed by IBM SPSS 17 Software Package program.In relation to color parameters, lightness decreased with the increase of the heat treatment despite the number of layers of the coating.For the less intense treatment (190ºC, 2 hours), lightness decreased approximately 38% in relation to initial lightness and for the most intense treatment (212ºC, 2 hours) the decrease was approximately 48%.Simultaneously, red color tone increased while yellow color tone did not show any consistent change.Similar results were reported before by several authors with uncoated heat treated wood (Bekhta and Niemz, 2003;Dubey et al. 2012;Esteves et al. 2007;Mitsui et al. 2001).Glossiness, measured parallel and perpendicular to the grain decreased with the intensity of the heat treatment which is in accordance to the reported by other authors (Aksoy et al. 2011;Karamanoglu and Akyildiz 2013).There seems to be no influence on glossiness due to the coating.

Pendulum Hardness, Adhesion and Surface Roughness
Pendulum hardness increased initially, decreasing afterwards with the intensity of the heat treatment.Coating increased the hardness of the surface since pendulum hardness is higher for the samples with two coats.These results are in accordance to results presented before that stated that Maderas.Ciencia y tecnología 19(2): 213 -224, 2017 Effects of heat treatment on the..: Gurleyen et al. hardness changes depends on wood species, conditions of treatment and even with the direction of the tests (Shi et al. 2007).Tests showed that adhesion strength generally decreased with heat treatment, nevertheless the decrease was much smaller when two layers coating is used.
No significant differences were found for the surface roughness.Although there is no difference between R a parameter of treated and untreated wood, this parameter is higher for wood with two layers coating.Maximum roughness (R y ) increases slightly with heat treatment and with the number of layers in the coating.This parameter represents the distance between peak and valley points of the profile and can be used as an indicator of the maximum defect height within the assessed profile (Mummery 1993).Similar variation was found for R z and somewhat for R q parameters.Roughness measurements by the stylus method with uncoated untreated and heat treated Turkish river red gum showed that surface roughness values decreased with increasing treatment temperature and treatment times (Unsal and Ayrilmis, 2005).Similar results were reported by (Gunduz et al. 2008) with Camiyani Black Pine (Pinus nigra), with Turkish Hazel (Corylus colurna) (Korkut et al. 2008a) and by (Korkut and Guller 2008) with red-bud maple (Acer trautvetteri Medw.).

CONCLUSIONS
Overall, lightness and glossiness decreased and red colour tone increased with heat treatment and no significant differences were found between one or two layers coating.In relation to pendulum hardness there was an increase initially, decreasing afterwards with the intensity of the heat treatment.This decrease was smaller if two layers coatings are used.Although adhesion strength generally decreased with heat treatment the use of a double layer coating improved the adhesion.No significant differences were found for the surface roughness although a slight decrease was observed.

Figure 1 .
Figure 1.Two different types of UV varnish application process.

Figure 4 .
Figure 4. Adhesion test machine and test sample.

Table 1 ,
table 2, table 3 and table 4 presents colour parameters, glossiness, pendulum hardness, adhesion strength and surface roughness parameters for untreated and heat treated wood coated with one or two layers.

Table 1 .
Statistical data for color difference values in untreated and heat treated wood coated with one and two layers Scots pine.Homogeneity Group, Std.Dev.: Standard Deviation, Mean: Average, N: Number of Measurements

Table 2 .
Statistical data for glossiness values in untreated and heat treated wood coated with one and two layers Scots pine.Homogeneity Group, Std.Dev.: Standard Deviation, Mean: Average, N: Number of Measurements

Table 3 .
Statistical data for surface hardness and adhesion values in untreated and heat treated wood coated with one and two layers Scots pine.

Table 4 .
Statistical data for surface roughness values in untreated and heat treated wood coated with one and two layers Scots pine.Homogeneity Group, Std.Dev.: Standard Deviation, Mean: Average, N: Number of Measurements