Dimensional Stability of Plywood Panels Made from Thermally Modified Poplar Veneers in the Conditions of Variable Air Humidity

ABSTRACT • Some properties of plywood panels made from untreated and thermally modified (90 min. at 200 oC) poplar veneers and their combinations have been analyzed. The change in equilibrium moisture content and the change in dimensions of the samples conditioned above water in closed containers were examined. The analysis using F-test (ANOVA) at the significance level of 95% confirmed that, regarding moisture content, all combinations of plywood panels differed significantly from the control panels. However, the analysis of thickness swelling showed that there was no significant difference among the plywood panels of similar construction. The highest values of anti-swelling efficiency were shown by plywood panels made from thermally modified veneers.


INTRODUCTION 1. UVOD
The purpose of thermal wood modification is to obtain the products of improved dimensional stability, decay resistance and durability. During thermal modification, the most thermally labile of wood polymeric components begin to degrade inside the cell wall (primarily hemicelluloses and in later stages also cellulose) resulting in the production of furan compounds, such as furfural and hydroxymethilfurfural (Rowell et al., 2009). At high temperatures, the links inside the lignin complex are degraded. This phase of increased lignin reactivity is characterized by the production of various condensation reactions between aldehyde and lignin, as well as by self condensation of lignin (Tjeerdsma et al., 2005).
The incurred chemical changes have a positive effect on the decrease in hydrofility of the treated material. According to Boonstra and Tjeerdsma (2006), the decrease in wood hygroscopicity during thermal treatments is the result of: -depolymerization of carbohydrates (especially hemicelluloses), which results in the decrease in the number of free hydroxyl groups; -increasing the share of crystal zone in the cellulose where hydroxyl groups are not readily available; further cross linking of the lignin, by which the availability of hydroxyl groups is additionally decreased.
The positive influence of thermal wood modification on the decrease in hydrofility, swelling and shrinkage of treated wood was confirmed by numerous researches (Kocaefe et al., 2008;Yongjian et al., 2010;Poncsak et al., 2010;Sinković et al., 2011;Zdravković and Lovrić, 2010). As opposed to these positive effects, thermal wood modification causes the decrease in most mechanical properties of wood. The two crucial factors affecting the final quality of treated wood are temperature and time (Kubojima et al., 2000;Poncsak et al., 2006;Shi et al., 2007;etc.).
Poplar (Populus nigra) is a species that, despite its relatively low density (ρ 0 =0.41g/cm 3 ), significantly changes its dimensions in the conditions of variable air humidity (α V =14.3%) (Šoškić and Popović, 2002). Also, poplar tends to form reaction wood (tension wood), which additionally complicates the drying process and influences the deformation of saw boards (Glavaški and Popadić, 1997). For this reason, poplar is most commonly used in the production of pulp and paper and in veneer and plywood production.
In the plywood panel production process, the crosslinking of adjacent veneer layers decreases plywood shrinkage, splitting and warping. In addition, thermal treatments cause the decrease in the treated material affinity for water. One of the measures for wood affinity for water is the contact angle. The measurements of contact angle in earlier papers showed that on the thermally treated veneers, (Zdravković and Lovrić, 2010), OSB panels (Unsal et al., 2010) and plywood panels (Candan et al., 2012), contact angle increases with increasing of temperature and treatment duration, which indicates decreasing treated material affinity for water. Improvement of physical properties of LVL made of thermally treated poplar veneers were proved by .
In this paper the possibility of production of plywood panels composed of thermally treated veneers was investigated in an attempt to obtain the material of improved dimensional stability.

MATERIJAL I METODE
The materials used in this study are poplar veneers with a nominal thickness of 3 mm. The veneers were selected by random sample method from the storage of plywood mill "Novi Drvni Kombinat" from Sremska Mitrovica. The materials were shipped to laboratory facility of the company "Tarket" at Bačka Palanka. Based on the previous laboratory research (Lovrić and Zdravković, 2009), it was found that thermal treatment at 200 ºC for 90 min gave the optimal ratio between the loss of volume shrinkage and the loss of mass of poplar veneer (loss of shrinkage was 7.62 % and loss of mass was 6.49 %). Most of the material was thermally treated by the above regime and some of the material was left untreated for the production of control samples.
Thermal modification was conducted in the presence of steam as protection agent, so the treatment can be assumed as steam-heat treatment. In the treatment regime, the conditioning phase was also included in addition to heating phase, thermal treatment phase and cooling phase. The conditioning phase of 2-hour duration started at the moment when the temperature dropped below 100 ºC.
After thermal modification, veneer sheets were cut into 80 cm by 80 cm and prepared for pressing. Melamine urea formaldehyde adhesive was applied by hand roller-spreader by spreading rate of 200 g/m 2 , veneers were arranged into the corresponding lay-ups and pressed according to the following regime: pressing temperature t = 85 ºC, total pressure P t = 15 MPa, pressing time Z 1 =10 min for three-layer plywood and Z 2 =13 min for five-layer plywood.
The following combinations of plywood were made: three-layer plywood composed of untreated veneers -3N (control group), three-layer plywood composed of outer layers of thermally treated veneers and inner layer of untreated veneer -TNT, three-layer plywood composed of treated veneers -3T, five-layer plywood of alternately composed treated and untreated veneers -TNTNT, five-layer plywood with outer layers of treated veneers and core layer of untreated veneers -T3NT, and five-layer plywood composed solely of treated veneers -5T.
Twenty samples of 5 cm by 5 cm were cut from each panel. The mass and dimensions of all samples were measured and dried to oven-dry condition, and measured again. Based on the measurements, the board moisture content (MC) (EN 322) and density (EN 323) were calculated. After measurements, the samples were arranged in closed containers above water and the changes in their MC and dimensions were monitored. The measurements of mass (for MC calculations) and thickness were performed every day during the first week, and after that every seven days.
After seven weeks (when the sample mass and thickness became stable), the samples were left to float on the water surface. The sample mass and thickness were measured again in the following week, and then ..........Zdravković, Lovrić, Stanković: Dimensional Stability of Plywood Panels Made ... they were immersed in water for another week. After this period, final measurements of mass and thickness were performed and the obtained data was used for calculations of swelling and MC for each plywood construction and for the construction of the corresponding curves representing the changes during the study period.

RESULTS AND DISCUSSION
3. REZULTATI I RASPRAVA 3.1 Initial plywood density, moisture content and thickness 3.1. Početna gustoća ploča, sadržaj vode i debljina It is obvious that, with the increase in the content of thermally treated veneers in the lay-up, plywood density increases and MC decreases. The decrease in MC is expected because during the thermal treatment, the number of free hydroxyl groups decreases and the possibility of moisture absorption from outdoor environment is reduced.
The increase in plywood density is probably caused by the following factors: -greater amount of plywood thickness loss during pressing due to diminished mechanical properties of thermally modified wood, caused by collapse of thermally modified cell walls, (Awoyemi and Jones, 2010); -higher plywood plasticity (in the process of thermal modification, hemicelluloses and celluloses are first decomposed, and the effect on lignin is much lower), which causes a lower "spring back" effect ( Table 2). Average thickness of the samples cut from the corresponding plywood panels are shown in Table 2. It can be seen that, under the same pressing regime, plywood thickness decreases if the content of thermally treated veneers in the lay-up increases. The exception is plywood TNTNT, which is somewhat thicker than plywood T3NT.

Promjene dimenzija ploča i sadržaja vode
To determine how the study plywood panels react to the conditions of higher air humidity, the plywood samples were arranged in closed containers above water. During the first week, MC and thickness of the samples increased rapidly, especially during the first three days (Figure 1). The control sample -3N showed significantly greater changes in MC and thickness compared to other plywood construction. Similar results were shown in the research by Nazerian et al.  3N (control group) -three-layer plywood composed of untreated veneers / troslojne ploče od nemodificiranih furnira; TNT -three-layer plywood composed of outer layers of thermally treated veneers and inner layer of untreated veneer / troslojne ploče s vanjskim toplinski modificiranim furnirima i unutarnjim nemodificiranim furnirom; 3T -three-layer plywood composed of treated veneers / troslojne ploče od toplinski modificiranih furnira; TNTNT -five-layer plywood of alternately composed treated and untreated veneers / peteroslojne ploče izrađene od kombinacije modificiranih i nemodificiranih furnira; T3NT -five-layer plywood with outer layers of treated veneers and core layer of untreated veneers / peteroslojne ploče izrađene s vanjskim modificiranim furnirima i unutarnjim nemodificiranim furnirima; 5T -five-layer plywood composed solely of treated veneers / peteroslojne ploče izrađene samo od modificiranih furnira With the progression of the experiment, all the observed values were slightly increased until the end of the sixth week from the beginning, and they became stable in the seventh week ( Figure 2). Plywood MC and total thickness swelling reached the maximum after floating and immersing of the samples for eight and nine weeks ( Table 3).
The analysis of graphs in Figure 1 shows that the separation of the presented curves into three groups starts after the first week. The lowest values of thickness swelling were attained by the samples cut from plywood made exclusively of treated veneers (boards 3T and 5T), the middle group of curves consisted of plywood made of a combination of treated and untreat- Maximum changes in dimensions were achieved only after the samples were immersed in water (Table 3) without an essential effect on the observed trend. Figure 3 shows the effect of plywood composition on dimension   The values of maximum MC (calculated after immersion - Table 3) were not so strictly grouped. The smallest value of MC was observed for five-layer plywood exclusively composed of thermally treated veneers (5T -MC=101.74 %). The three-layer plywood composed of only thermally treated veneers (3T) showed the maximum MC values, similar to plywood TNTNT, while plywood T3NT was the closest to the control plywood 3N.
Statistical analysis of the obtained values was performed to determine whether the calculated values of average changes in MC and thickness swelling were due to coincidence, or due to different behavior of the study boards. The values calculated after seven weeks from the beginning of the experiment were taken as the reference point. The main statistical indicators of MC and thickness swelling are shown in Table 4 and 5.    Based on the analysis of data on MC reached after seven weeks (Table 6), it can be concluded that there is a significant difference among all boards, i.e. that in the conditions of variable air humidity each board behaves differently. However, the conclusions are different if the data on thickness swelling reached during the study period (Table 7) are considered.
All types of boards were significantly different compared to the control board 3N, while there was no difference among the boards of similar construction. So, there was no significant difference between the boards 3T and 5T (boards composed of treated veneers only), nor in the comparison of boards composed of different combinations of thermally treated and untreated veneers -TNT, TNTNT and T3NT. It is interesting that there was no significant difference in the comparison of 5T and TNTNT boards, which means that the middle veneer sheet, in spite of the lowest exposure to environmental moisture, contributed to the decrease in thickness swelling of the observed board.
As in practice, the change in thickness is more important than the change in moisture content, it can be concluded that the observed categories of boards present the boards of similar properties.

ZAKLJUČAK
Based on the examination of plywood panels produced by combining the thermally treated and untreated poplar veneers, it was concluded that the thermal modification process had a positive effect on the decrease in moisture absorption of the panels. It was also concluded that, with the increased amount of thermally treated veneers in the plywood panels, the values of the observed properties decreased.
The calculated values of ASE (anti-swelling efficiency) showed that the three-layer board exclusively composed of thermally treated veneers (3T), and the five-layer board (5T) achieved the highest values, which was expected.
The analysis using F-test at the significance level of 95 % confirmed that all boards had significantly different moisture contents. However, the analysis of thickness swelling showed that there were no significant differences among the boards of similar construction.
The results presented in this paper proved that the use of thermally treated veneers in plywood production contributed to the improvement of their properties and their resistance to higher air humidity and moisture.