ANTI-FUNGAL AND ANTI-TERMITE ACTIVITY OF EXTRACTIVES COMPOUNDS FROM THERMALLY MODIFIED ASH WOODS

Thermal modification of wood is a promising alternative to chemical and biocidal modification processes, increasing the biological durability and dimensional stability of wood. However, the wood-decay resistance properties of heat-treated wood are still not well known. The main objective of this study was to determine the biological resistance of heat-treated ash wood, and assess the antifungal and anti-termite activity of extractive compounds from heat-treated ash woods, depending on the intensity of the modification process (2 hours at 170, 200, 215, 228 (°C) steam pressure). Untreated and heat-treated wood samples were extracted with water or acetone. The extracts were then used to determine inhibition effectiveness against white-rot (Trametes versicolor) and brown-rot (Rhodonia placenta) fungi. Whatman papers impregnated with extractives were used to evaluate the inhibition of termite feeding. Lastly, the extractives were analyzed by Gas Chromatography Mass Spectrometry (GC-MS) and compared for their level of anti-termite and antifungal activity. The results showed that the degree of antifungal activity of these extracts depended on the solvent used during the extraction process and varied depending on heat treatment intensity. The extracts were more effective against brown-rot than white-rot fungi. However, the anti-termite activity of heat-treated ash wood extracts was not really significant. A GC-MS analysis showed that the main share of the extractives in untreated wood was removed. In addition, new chemical elements were generated by the thermal degradation of wood polymers (lignin and hemicelluloses), including aliphatic acids, monosaccharides and other products resulting from their dehydration reaction. The most abundant element was syringaldehyde, from lignin derived compounds, which might explain the antifungal activities of thermally treated ash wood extracts.


INTRODUCTION
Ash (Fraxinus spp.) is a significant hardwood resource largely used in the European wood industry. Given its very weak resistance to decay and insects, this timber is naturally in appropriate for class 3 use (EN 350 CEN 2016). Ash wood materials therefore need to be treated with biocide additives for outdoor use , Gérard et al. 2017. Industrial wood protection methods usually focus on ways of impregnating active formulations containing biocides. However, ash wood is only a moderately treatable timber for both its sap and heartwood (EN 350 CEN 2016, Gérard et al. 2017). In addition, due to their impact on human health, these chemicals are a subject of environmental pressures, which will restrict their use in the near future (Schultz et al. 2007). The Implementation of EU Environmental Legislation in recent years has resulted in the development of non-biocidal alternatives, such as thermal modification (Sandberg and Kutnar 2016), chemical impregnation (Gérardin 2016, Mantanis 2017, Guo et al. 2018, or a combination of the two (Salman et al. 2016, Sandberg et al. 2017. Using antagonistic microorganisms, combined or not with biochemicals, is also considered as a promising ecofriendly way of protecting wood: (i) biocontrol microorganisms consume the available nutrients and produce biochemical compounds that are toxic and/or repellent to fungal decay and, (ii) the biochemical compounds include cell wall-degrading enzymes, siderophores, chelating iron and other antibiotics. This last wood preservation method is not yet industrialized and requires additional studies, especially regarding the environmental impact of these formulations (Susi et al. 2011).
On the other hand, thermal treatment of wood, using slow pyrolysis in an inert atmosphere, has been widely studied and has now been industrially developed around the world, mainly to increase wood durability against fungi in line with the process parameters (Calonego et al. 2010, Candelier et al. 2016. According to the literature (Kamdem et al. 2002, Weiland andGuyonnet 2003), the improved resistance of heat-treated wood material to decay can be explained by the following four assumptions, all derived from chemical modifications of wood cell wall polymers due to their thermal degradation (Tjeerdsma and Militz 2005, Yildiz et al. 2006, Inari et al. 2007).
(i) The hydrophobic behavior of wood is increased, thus limiting water sorption within the material and reducing any fungal growth. This improved hydrophobicity reduces the capillary transfer of water into the thermally modified wood and thus limits fungal growth by removing the optimum conditions for its development (due to lack of water). In addition, heat-treated wood has a lower Fiber Saturation Point (FSP) than untreated wood, leading to better resistance to rot fungi Guyonnet 2003, Hakkou et al. 2006).
(ii) Wood polymers are modified and the enzymes involved in the fungal degradation of wood do not recognize it anymore as such. Firstly, the changes in lignin chemical composition prevent fungal enzymes from recognizing and attacking specific target molecules (Vallet et al. 2001, Lekounougou et al. 2009). Secondly, changes in the ligneous polymer network also seem to play a role in fungal inhibition. The improved resistance of modified wood to decay is mainly due to possible crosslinking between lignin and certain molecules derived from wood thermal degradation, such as furfural (Weiland and Guyonnet 2003). Such modifications lead to a substrate that is not recognized by the fungal enzyme system, resulting in an undecayed material (Hakkou et al. 2006). Moreover, cellulose may undergo an esterification reaction, due to the acetic acid generated during the thermal degradation of hemicelluloses Militz 2005, Gao et al. 2016).
(iii) Hemicelluloses are the main wood component degraded by thermal modification, inducing a significant loss in potential nutrients for fungal growth (Hakkou et al. 2006, Altgen et al. 2019. The chemical modification of polysaccharides is responsible for improving resistance to wood decay, but cannot be the only factor, as lignin is also an important source of nutrients for white-rot fungi (Lekounougou et al. 2009).
(iv) New extractive, fungicidal substances are generated by wood thermal degradation and can serve to prevent decay. In fact, some extracts from heat-treated beech, poplar and maritime pine are already known to have antifungal activities (Kamdem et al. 2000, Peters et al. 2009, Lovaglio et al. 2017. However, other studies showed that improvement in the resistance of beech wood to decay does not appear to be due to new extractible compounds formed during thermal treatment. Hakkou et al. (2006) reported that the new extractive compounds produced during heat treatment of beech wood, carried out between 200 °C and 280 °C, only had a slight effect on the heat-treated wood against Trametes versicolor. Kamdem et al. (2002) found similar results in a study focusing on heat-treated pine and spruce wood samples extracted with water and organic solvents.
The extractive content of heat-treated wood increased when treatment was carried out at low temperatures and decreased with treatment carried out at higher temperatures (> 220 °C). Most of the raw extractives disappeared and new compounds, such as anhydrosugars, mannosan, galactosan, levoglucosan and two C5 anhydrosugars, were generated. Syringaldehyde, sinapaldehyde and syringic acid appeared to be the products formed in the largest amounts, all of which came from lignin degradation (Esteves et al. 2008). This hypothesis regarding the formation of new extractives and their role in heat-treated wood durability therefore remains to be confirmed.
The resistance of thermally modified wood to termites has also been studied. Contrary to fungal resistance, the influence of heat treatment on wood resistance to termites appeared to be more variable depending on the processing conditions and wood species. In many cases, resistance to termites was random and, in some cases, even reduced (Doi et al. 1998, Momohara et al. 2003, Nunes et al. 2004, Sivrikaya et al. 2015, Salman et al. 2016. Some studies revealed that thermally modified wood had lower termite resistance than untreated wood (Sivrikaya et al. 2015, Salman et al. 2016, Salman et al. 2017. Generally, termite survival levels reveal an effect of thermal treatment on the biology of these insects. Termite survival rates remained virtually constant for treatments carried out at a low temperature (180 °C to 210 °C) and were lower for higher treatment temperatures. Treatments over 200 °C induced crucial modifications to the wood material, possibly reducing suitability for termites . As heat-treated wood is consumed when exposed to termites, the toxicity of ingested components may be the reason for the higher mortality rate observed (Surini et al. 2012) when treatments are carried out at high intensities (temperature -duration). In addition, when heat-treated and untreated control wood samples were exposed side-by-side (choice test), termites preferred to attack the untreated wood samples (Nunes et al. 2004).
In some cases, when the thermal modification process was carried out under optimum conditions, heat treatment enabled low natural durability timbers to be used by making them more resistant to decay, allowing them to be used in use classes 2 and 3 (use class 4 being excluded due to the occurrence of soft rots) (EN 335 CEN 2013), thus up-grading their economic value (Kamdem et al. 2002). These improved biological properties conferred to the wood by chemical modifications have a negative impact on the mechanical resistance of heat-treated wood (Dilik andHiziroglu 2012, De Oliveira-Araújo et al. 2016). Surface hardness seems to be slightly enhanced, while other characteristics, such as bending and compression strengths, stiffness and shear strength, are considerably weakened depending on the kind of thermal processes and treatment intensities used (Boonstra et al. 2008, Hannouz et al. 2015. The objective of this study was to investigate the antifungal and anti-termite activities of extracts from thermally modified ash wood depending on different treatment intensities. In order to ensure the reproducibility of the thermal treatment processes, as well as the quality of the treated wood products, the heat treatments were carried out under monitored conditions, using different heating temperatures (170, 200, 200, 215 and 228 (°C)) and a fixed residence time (2 h).

Wood samples
Untreated ash (Fraxinus excelsior L.) wood samples were obtained from a French wood company (Bois Durables de Bourgogne, 71120 Vendenesse-lès-Charolles, France). Twenty boards measuring 4000 x 110 x 25 (mm 3 ) (L x R x T) were selected. Special attention was paid to ensure these planks only had small variations in density (around 650 kg/m 3 ± 10 %) and a uniform width of annual rings. All the planks were sawn into two equal parts of 2 m in length. Half of each plank was used as reference material and the other half was thermally treated at different treatment intensities. The planks were then dried at 103°C in an industrial oven up to mass stabilization (m 0 ).

Heat treatment protocols
The thermal treatment processes were carried out at the same company. These operations were carried out in a 20 cubic meter industrial oven (Jartek ® ), in convection heat-transfer mode and under a steam pressure process (ThermoWood® method, Finnish Thermowood Association (2003)).
The temperatures of the wood core and that of the oven atmosphere were dynamically recorded and monitored throughout the treatment (Figure 1), in order to optimize the process and the final wood qualities.
The oven temperature was firstly increased rapidly (0,5 °C/min) to 100 °C, then slowly (0,2 °C/min) to 150 °C. This temperature level was maintained for 1 h. The temperature was finally increased (0,2 °C/min) to the desired level (from a low temperature (170 °C to 200 °C) to a high temperature (215 °C to 228 °C) and kept at that constant level for 2 hours. Steam was injected into the chamber to keep it oxygen-free and to influence chemical changes in the wood. The heating system was then stopped and the wood samples were cooled to room temperature in an oxygen-free atmosphere with a water spray.
Each heat treatment was carried out simultaneously on four ash wood planks.

Mass Loss (ML) due to wood thermal degradation
Each raw ash wood plank measuring 2000 x 110 x 25 (mm 3 ) (L x R x T) was dried at 103 °C up to mass stabilization and its anhydrous mass was measured (m 0 ). After a thermal modification process each treated ash wood plank was again dried at 103 °C and its anhydrous mass was re-measured (m 1 ). The Mass Loss (ML%) due to wood thermal degradation depending on treatment conditions was determined by the following Equation 1.

Decay resistance
Decay and termite resistance tests were carried out according to adaptations of XP CEN/TS 15083-1 (CEN 2006) and EN 117 (CEN 2012) standard criteria, respectively, in one of our previous studies . The protocols used for these durability tests are detailed in that previous work.

Extraction protocols
All the experimental procedures used to determine extractive contents were adapted, with minor modifications, from procedures found in the scientific literature (Rowell et al. 2005).
A mix of several pieces (One 20-cm long piece, per wood board) from untreated and heat-treated planks was crushed and sieved to obtain particle sizes from to 0,2 and 0,5 mm, for each temperature level. After a drying step at 103 °C up to mass stabilization (m 2 ), the sawdusts (10 g) were separately extracted in a Soxhlet with acetone (Sigma Aldrich, 32201-M) (6 h), or with hot water (2 x 6 h), and dried at 103 °C for 48 h to obtain the anhydrous mass (m 3 ).
For untreated and heat-treated ash wood under varying temperature conditions, extractive contents were determined by the following Equation 2.
Each Petri dish (9 cm diameter) was filled with 10 mL of malt-agar medium containing 1200 μL of the diluted extracts in water or acetone (Merck company, Darmstadt, Germany) (C= 2,5% m/m) and left to solidify. These dilutions were chosen according to past studies (Esteves et al. 2011, Pal et al. 2011. A 1-cm square section of a seven-day-old culture of Rhodonia placenta or Trametes versicolor was placed in the center of the Petri dish and stored in a climatic chamber regulated at 22 °C ± 2 °C and 70 % ± 5 % relative humidity (RH) for seven days' incubation. Three replicates were carried out per diluted wood extract sample and per fungus. Three media free of extractives and without solvent were used as a control. In addition, three media free of extractives and with solvent (water or acetone alone) were used to check that the water and acetone solvents did not have any impact on the activity of each fungus. Following the fungal exposure period, the mycelium growth diameter was measured in millimeters, for the control medium (M c ) and the extract-supplemented medium (M t ). The percentage of mycelium growth inhibition was determined by Equation 3.

Anti-termite activity test
The anti-termite activities of each heat-treated ash wood extract were tested by screening tests. Anhydrous cellulose paper measuring 2,5 cm in diameter was weighed (m 4 ) and then impregnated with 70 µL of the diluted extracts in water or acetone (C= 2,5% m/m), air dried (20 °C ± 2 °C and 65 % ± 5 %) for 2 hours (m 4 ), then placed in the center of a Petri dish (5,5 cm diameter). Fifteen grams of wet sand (4 vol. sand/1 vol. water) was placed evenly around the paper and 20 termite workers (Reticulitermes flavipes, ex. santonensis) were added to each test set-up. Three replicates were carried out for each diluted wood extract. Three papers impregnated with water or acetone alone were tested to estimate the impact of the solvent on termite activity. Tests with water only were considered as a virulence control. For each test set-up, the paper samples were placed on a plastic grid. Lastly, three diet control set-ups containing only 15 g of wet sand and 20 termites were used to check termite survival without any feeding possibilities/without any trophic sources. All test set-ups were kept at 27 °C and > 75 % RH ( Figure 2). Every two days, each test set-up was observed to check sand humidity, add water if needed and keep track of termite behavior and activity. When all the termites contained in the diet control set-ups had died, the test was stopped. The termite survival rate was then determined, the anhydrous mass of the cellulose papers was measured (m 5 ) and the Weight Losses (WL term. %) due to termite degradation were calculated by the following Equation 4.
( ) Figure 2: Termite screening test set-ups at the start of termite exposure.
Before being injected into the GC-MS analyzer, the extractive substance samples were derivatized in order to improve the detection of all chemical compounds. To perform this silylation derivatization process, 2 mg of extract was solubilized in a glass tube with 50 μL of BSTFA + 1 % TMCS solution (Bistrimethylsilyltrifluoroacetamide + Trimethylchlorosilane) (Acros Organics). The glass tube was sealed and dried in an oven at 70 °C for 120 minutes, then opened to evaporate the BSTFA. The derivatized wood extract was dissolved in 1 mL of C 4 H 8 O 2 solution (ethyl acetate (Acros Organics)). One μL of this solution was injected into the Gas Chromatograph at a 250 °C inlet temperature in splitless mode. Helium was used as the carrier gas. The temperature program was: 80 °C (2 min), 10 °C/min to 190 °C 15 °C/min −1 to 280 °C -maintained for 10 minutes, 10 °C/min to 300 °C 15 °C/min −1 -maintained for 14 minutes. A helium flow of 1 mL/min was used as the mobile phase. After this separation step, compounds were transferred to the Mass Spectrometer by a transfer line heated at 250 °C and ionization was achieved by the Electron Impact method (70 eV ionization energy).
The recognition of each component was achieved by comparing its mass spectrum with the NIST Library 2005 using NIST MS Search 2.0 (2011) software. The identification was deemed to be relevant when the comparison coefficient was higher than 900 (The comparison coefficient corresponds to the match factor between the mass spectrum obtained by a GC-MS analysis and the mass spectrum from the NIST MS Search 2.0 (2011). The match factors ranged from 0 to 1000, with 0 meaning no match and 1000 meaning a total match).

Determination of pH Value
pH measurements were conducted according to Wang et al. (2012). Untreated and heat-treated samples were ground and passed through 40-60 mesh screens. Three grams of oven-dried sawdust samples was soaked in 30 mL of distilled water and then stirred for 5 min, allowed to stand for 15 min, stirred for another 5 min, and then left to stand for another 20 min. After this procedure, the pH values of the liquid were determined using a pH meter (PH/mVmeter Knick 911 ATEX; Knick Elektronische™, Berlin, Germany).

Statistical analysis
The impact of heat treatment intensity compared to the untreated ash wood on (i) the termite and fungal resistance of the wood samples, (ii) the content and (iii) the effects of water and acetone extractives on their antifungal and anti-termite efficiency were evaluated using an ANOVA (one-way analysis of variance) and Duncan's comparison test. These statistical analyses were carried out by the JMP 10.0.2 program (SAS 2012) by applying the Fisher test. The results were then ranked into several categories, from "a" to "e" for the water-extractives and from "A" to "E" for the acetone-extractives. The impact of a parameter on a system not connected by the same letter was considered as non-significant at the 5 % level.

Biological resistance tests on wood block samples
Decay and termite resistance tests were previously undertaken in . In order to put into perspective, the biological durability of untreated and heat-treated solid wood with the following antifungal and anti-termite activities of the respective wood extractive fractions, this section focuses solely on the main results obtained by .

Decay resistance
Similar results were found for both of the rots tested. However, Rhodonia placenta was the most aggressive rot on beech control samples (WL 49,9 %), whereas Trametes versicolor was the most degrading rot on ash control samples (WL 48 %). Thermal modification increased the durability of all wood materials, which was in agreement with previous studies (Kamdem et al. 2002, Esteves andPereira 2009). These results were expected as Rousset et al. (2004) and Metsä-Kortelainen et al. (2005) also found that the thermal treatment of wood at high temperatures increases the resistance of wood to decay.
According to Table 1, the thermal treatment carried out at temperatures over 200 °C conferred the "very durable" durability class 1 to the heat-treated wood materials, according to the classification method of XP CEN/TS 15083-1 (CEN 2006).  (2017)).

WL% -Median values ML% -Average values
Values within each column and factor followed by the same letter are not significantly different. * ''0'' for no attack ''1'' for attempted attack, ''2'' for slight attack, ''3'' for average attack, ''4'' for a strong attack. ** According to a one-way analysis of variance, systems not connected by the same letter are largely different at the 5 % level.

Termite resistance
Termite survival revealed an effect on termite biology. Although the survival rate was similar for treatments at 170 ºC and 200 °C, it was lower above those temperatures (Table 1). Heat treatments at temperatures over 200 °C caused critical changes to the wood and consequently its durability could be improved. As the wood was nonetheless degraded, the toxicity of the components consumed may explain the higher termite mortality rate (Surini et al. 2012) depending on the increase in heat treatment intensity. However, it was not particularly effective over a short time scale, as 51,33 % of the termites were still alive after the test, with no significant differences between treated and untreated wood. The current results are in agreement with those of Nunes et al. (2004), who studied the resistance to termites of the species Reticulitermes grassei with wood treated by the German method (OHT) and concluded that, in spite of the slightly higher mortality of termites in treated samples and a smaller weight loss, the differences were not significant.

Extractive content
For thermal treatment carried out at a low temperature (< 215 °C), the proportions of heattreated wood extractives increased in comparison with those of untreated ash wood, whatever the solvent used during the extraction process (Figure 3). This result tallied with past studies, indicating an increase in extractive rate resulting from the generation of different degradation products (Poncsak et al. 2009, Esteves et al. 2008. A past study by Esteves et al. (2008), carried out on eucalypt wood, showed that thermal degradation of lignin and hemicelluloses led to an increase in extractive content. In fact, the formation of new extractive compounds arising from polysaccharide degradation at around 160 °C may be one of the reasons for that phenomenon. At higher temperatures (>215 °C), these new compounds are also generated but, under the heat effect, they are then converted into volatile products that leave the structure of the wood, thus causing a loss of matter, leading to a decrease in extractive content (Wang et al. 2016).
The relative percentage contents of extractives ( Figure 3) increased at low treatment temperatures, reaching maximum values of 4,59 % and 2,19 % for thermal treatment carried out at 170°C and for water and acetone extractions, respectively. Beyond that temperature the relative extractive percentage decreased.

Antifungal activity test
Fungal activity was inhibited by the heat-treated and untreated ash wood extracts. Table 2 and Figure 4 shows how fungal development was hindered depending on the extractive compounds. However, for both of the fungus species tested, the effectiveness of the extracts varied depending on the heat treatment temperature and the solvent used during the extraction process. It clearly appeared that untreated and heat-treated ash wood extracts were more efficient against brown-rot (Rhodonia placenta) than against white-rot (Trametes versicolor) growth, whatever the extraction solvent used. In both cases, the efficiency of the extracts seemed to be correlated to the heat treatment temperature.   2 indicates that heat-treated ash wood extracts were more efficient against Trametes versicolor and Rhodonia placenta after 3,5 days than after 7 days, except for the modified ash wood extracts extracted with acetone against Rhodonia placenta. It turned out that, 7 days after fungal exposure, acetone-extracts from ash wood modified at a temperature of 228 °C delayed the growth of Trametes versicolor and Rhodonia placenta most. Values within each column and factor followed by the same letter are not significantly different (letters from A to E were used to compare acetone-extractives and letters form a to e were used to compare water-extractives).
The different degrees of fungal development for untreated and thermally modified wood reflecting the contrasts in extractive chemical composition are displayed in Figure 6 and Figure 7. Although some effects of water and acetone extracts from heat-treated ash could be seen on fungal growth depending on the treatment intensity, this phenomenon was not significant. It appeared that the fungi adapted themselves to the new medium and regained normal growth after a few days of incubation. In other words, heat-treated wood extracts seem to have shortterm fungistatic action. Table 3 and Figure 5 show that the heat-treated ash extract yields and their anti-termite activities were largely linked to the polar or nonpolar nature of the solvent used during the extraction process. In fact, anti-termite compounds can have many varied chemical characteristics and different polarity levels, which may or may not be soluble in a specific solvent (Kadir et al. 2015).

Anti-termite activity test
The test was stopped after 21 days, when the termites in the diet control set-ups had died. The results in Table 2 show that the anti-termite activity of the heat-treated ash wood extracts was not significant. Values within each column and factor followed by the same letter are not significantly different (letters from A to C were used to compare acetone-extractives and letters form a to e were used to compare water-extractives).
For untreated and treated ash wood under different treatment intensities, the mass loss due to termite deg-radation indicated that the relative extracts obtained with water were statistically more attractive for termites and the termite survival rate was quite similar to that of the control sample. This effect was related to the final temperature level of the thermal treatment. On the other hand, the same extractives obtained with acetone seemed to have no significant effect on termites compared to the water control samples ( Figure 5).
The termite survival rate revealed no significant effect on the biology of these insects. The survival rate was not significantly dissimilar for the different treatments compared to the control samples. The effects of untreated wood extracts on the termite survival rate were very similar to those for heat-treated wood, confirming the innocuous property of these treated wood extracts in terms of termite toxicity.

Water extractives
The typical chromatograms from GC-MS analyses of the chemical composition of untreated and thermally modified ash wood extracts by the water extraction process are presented in Figure 6. For untreated ash wood, the GC-MS analysis highlighted that almost all of the original water extractives resulted from polysaccharides or their silylated fragments. Only the weak peak observed at rt = 11,88 min corresponded to the antioxidant Tyrosol (4-hydroxyphenylethanol).
For the ash wood treated at 170 °C, it was found that the respective water extracts mainly comprised low molecular weight organic 2-hydroxy-acids (propanoic acid (rt = 4,84 min and 6,09 min), acetic acid (rt = 5,10 min), butanoic acid (rt = 5,80 min)) ( Table 5) and derivated phenylpropanoid compounds (rt = 11,00 min and 11,35 min). The presence of these acid compounds, decreasing the pH of the wood, may have played a role in fungal inhibition. Indeed, according to the results presented in Table 4, the pH values of ash wood decreased depending on the temperature used during the heat treatment process. The pH values varied from 5,16 for untreated wood to 3,92 for wood thermally modified at 228 °C. Similar trends were observed on softwood and hardwood species by Niemz et al. (2010), showing that the pH decreases with increasing heat treatment intensity. Table 4: pH values of untreated and heat-treated ash wood.
Values followed by the same letter are not significantly different.
Indeed, Yalcin and Sahin (2015) reported that heat treatment conferred on narrow-leaved ash wood a pH that was less conducive to fungal growth, with the untreated wood pH of 5,7 falling to 3,9 in the modified wood, while the pH for optimum fungal growth is around 5-6 (Bozkurt et al. 1993). In addition, such a decrease in wood pH could disturb the enzymatic digestion of termites during their wood destruction (Lima et al. 2014). In fact, pH widely affects the activity of fungal enzymes. The pH dependence is usually due to the side groups of the amino acids. A diminution in pH changes the protonation pattern resulting in protein denaturation (in shape, charge and location within the substrate) (Purich 2010).
For ash woods treated at higher temperatures, from 200 °C to 228 °C, only traces of the previously mentioned low molecular weight organic acids were observed. In addition, a large peak located in the polysaccharide area, from rt = 10 min and rt = 13 min, was observed but remained unidentified. These last peaks may have been due to the condensation of several chemical compounds produced by wood thermal degradation and might also partially explain the greater fungal resistance of wood thermally treated at high temperatures (Hakkou et al. 2006).

Acetone extractives
The typical chromatograms from GC-MS analyses for the chemical composition of untreated and thermally modified ash wood extracts by the acetone extraction process are presented in Figure 7. The GC-MS analysis revealed that almost all of the untreated acetone-extractives disappeared after heat treatment, but on the other hand the thermal degradation of hemicelluloses and lignin generated new compounds. Monosaccharides and some products derived from their dehydration reactions, along with syringaldehyde (13,18 min, Figure 7), were the most abundant products generated by the heat treatment. Syringaldehyde appeared to be the product formed in the largest amounts, arising from lignin degradation, and was generally the major detected component according to the TIC chromatograms, Table 5. *Rt = Retention time (in minutes) shows the time taken for the analytes to pass through the column and reach the mass spectrometer detector. **Intensity (as a %) represents a reflection of the amount of a specific analyte that was present, compared to the predominant one (e.g. Intensity = 100%).
Similar results for syringaldehyde extraction, by water and ethanol, from heat-treated wood were found by Ibrahim et al. (2012). The syringaldehyde content increased depending on the thermal treatment intensity, more particularly for wood treated at temperatures over 200 °C (Figure 7), and might explain the fungal inhibition activities of the heat-treated wood extractives. Indeed, previous studies highlighted the antifungal and anti-microbial activities of the phenolic compound syringaldehyde. Murugesan et al. (2009) also showed that syringaldehyde inhibited the growth of bacteria and of the fungus Ganoderma lucidum in syringaldehyde-treated malachite green samples. De Souza et al. (2005) also reported that syringaldehyde expounded anti-fungal activity against Leucoagaricus gongylophorus. According to Ibrahim et al. (2012), syringaldehyde (also called 3,5-dimethoxy-4-hydroxybenzaldehyde), has the same kind of structure as vanillin and is a unique natural compound that has bioactive properties (antioxidant, antifungal, anti-microbial and anti-tumorigene-sis), which belongs to the phenolic aldehyde family. In a past study, syringaldehyde displayed a successful role in inhibiting fungal growth rates (Kelly et al. 2008). Its fungicidal effects were mainly due to aldehyde moiety and the hydroxyl substituent present in syringaldehyde (De Souza et al. 2005). Total hydroxyl groups from lignin also had a large impact on cellulase adsorption and enzymatic hydrolysis occurring during fungal degradation (Yu et al. 2014), and more particularly acetosyringone, vanillin and syringaldehyde (Qin et al. 2016). Ximenes et al. (2011) confirmed that syringaldehyde and vanillin inhibit cellulose hydrolysis. Although syringaldehyde displayed insecticidal properties against Acanthoscelides obectus beetles in a past study (Regnault-Roger et al. 2003), the concentrations tested in this work did not make it possible to confirm its insecticidal behavior on termite activity.

CONCLUSIONS
Heat treatment improved wood durability, clearly increasing resistance to brown and white rots, but it had only a slight effect in improving resistance to termites. The process carried out at 228 °C was found to give ash wood the best biological resistance properties. However, it can be expected that such treatment will greatly degrade the mechanical properties of the wood.
In addition, the results presented in this work confirmed that thermal treatment generates new extractive substances within the modified wood material, depending on the treatment temperature level, which could act as fungistatics and prevent fungus from growing, improving the resistance of the modified wood to decay. The antifungal activity of heat-treated ash wood extracts varied depending on heat treatment intensity and the solvent used during the extraction process, but these effects were not efficient over the long term and were not significant. It also clearly appeared that extracts from heat-treated ash wood were more efficient against brownrot growth than against white-rot growth. The analysis of these extractive compounds showed that the most abundant was syringaldehyde. In addition, the syringaldehyde content increased in line with the thermal treatment intensity, whatever the extraction process used. Although syringaldehyde has shown insecticidal properties in past studies, it could not be shown in our study that it has an insecticidal role against termites. The generation of low molecular weight organic acids might decrease the pH of wood, also impacting fungal inhibition and the enzymatic digestive system of termites. Lastly, the condensation of several chemical compounds from wood thermal degradation might also partially explain the better decay resistance of wood treated at high temperature. However, taken separately, the extractive compounds might act differently from the same extracts present within the wood. The increased durability of heat-treated ash wood is due to thermal degradation of wood cell wall polymers involving a combination of modifications: hydrophobic behavior, lignin modification, hemicellulose degradation and also extractive generation. To conclude, the results obtained in this study showed that the extractive compounds of heat-treated wood have a slight impact on the better durability of thermally modified wood, but are certainly not the only reason.