Research and Tests in North America and Europe (download brochure)
In 1955 Forest Products Laboratory in Madison, Wisconsin in cooperation with University of Wisconsin published report 1621 providing testing data for a so called Staybwood. This report was based on research going back to 1920-30-40th in Germany by Mr. Stamm and Hansen, who later were associated with the University of Wisconsin. For their test the thermo-treated wood was produced utilizing hot-melted metals at temperatures up to 300°C. The main result from this report is a prove that heat-treated wood becomes substantially more dimensionally stable, while losing weight and hardness.
Louisiana Forest Products Laboratory conducted test on the ability of thermo-treated wood to withstand termite attacks in 2002. The softwood, heat-treated with hot oil, showed no additional resistance to termite attacks compared to untreated softwood.
Forintek Canada Corp. was involved in tests for heat-treated wood produced by both Perdure and ThermoWood based processes. The tests were devoted to abrasion resistance, impact and static bending, dimensional stability and resistance to fungal decay.
Westwood Corp. and Keim Lumber Company provided a series of experiments on physical properties of Thermo-treated hardwood and softwood, compared with non-treated wood in 2008.
Test Results for hardwoods and softwoods, performed by Westwood Corp. and Keim Lumber Company (USA) - see Downloads
Experiment # 1. Shrinkage of wood in Thermo-treatment process.
The research method. Measurements of thickness and width of the boards before and after treatment. For each species 20-30 samples were used.
1. The shrinkage in average 2% in thickness and 3% in width.
2. The average volume decrease factor is 0.95 (5% loss). It reduced in proportion to the equilibrium moisture reduction of thermo-treated material.
Experiment # 2. Weight loss in Thermo-treatment process.
The research method. Measurements of weight of the boards before and after treatment. For each species 4-5 samples were used.
Results. The significant weight loss factor is caused by two reasons:
1. The equilibrium moisture decreases at least 2 times, compared with non-treated wood. This weight decrease factor is 4-5%.
2. The wood elementals emission while treated (15-18%). Because we know from Experiment # 1 the Volume decrease factor for each species, we can easily calculate the Density decrease factor (see results in the Table). The strength loss factor is proportional to the Density decrease factor, so this data also can be used as a Strength loss factor.
Experiment # 3. Water absorption of samples with the coated ends. Short time interval (3 hours).
The research method. Placing the samples of thermo-treated and untreated wood into the water and measuring its weight. For each species 4-5 samples were used.
1. We see that the water absorption of Pine is almost similar to non-treated wood. For Ash and Red Oak the water absorption decreased in a factor 2 compared with un-treated wood., for poplar – in a factor 1.3.
2. The thermo-treated coated pine (1 coating) shows a significant decrease of water absorption compared to the non-coated material (3 times less). To get more objective results we’ll use the parameter of Weight change to one inch of the sample (to decrease the influence of the factor of the sample's width variety).
3. Because the density of thermo-treated wood decreased by 15-18%, the real water resistance to the short water treatment caused by the thermo-treatment changes of wood at molecular level, increases 3 times for Ash and Oak, and twice for Poplar and by 15% for Pine.
Experiment # 4. Water absorption of samples with the cut (open) ends. Short time interval (3 hours).
The research method. Placing the samples of thermo-treated and untreated wood into the water and measuring its weight. For each species 1-2 samples were used.
1. We see a little increased water absorption for all the species with approximately the same proportion between treated and non-treated wood.
2. The thermo-treated coated pine (1 coating) with the cut ends shows the same water absorption than non-coated as not-treated material.
3. This factor shows that most of the water is going inside the wood from the ends. This means the critical significance of closing ends even in treated material!
Experiment # 5. Swelling of samples with the coated ends. Short time interval (3 hours).
The research method. Placing the samples of thermo-treated and untreated wood into the water and measuring its width. For each species 4-5 samples were used.
Results. The short time interval (like 3 hours) isnot enough for the swelling of the wood (even treated or non-treated).
Experiment # 6 Water absorption of samples with the coated ends. Long time interval (18 hours).
The research method. Placing the samples of thermo-treated and untreated wood into the water and measuring its weight. For each species 4-5 samples were used.
Results. The water absorption for treated wood is slowed up, compared with non-treated wood, and also with the Experiment # 3 for 3 hours. For non-treated wood we see the same dynamic of water absorption. So, the dynamic of water absorption depends on the time of water treatment.
Experiment # 7. Swelling of samples with the coated ends. Long time interval (18 hours).
The research method. Placing the samples of thermo-treated and untreated wood into the water and measuring its width. For each species 4-5 samples were used.
1. The swelling of thermo-treated wood after 18 hours under water is around 0.3-0.4% for Oak and Ash, around 0.7% for Poplar and 2% for Pine.
2. The swelling of thermo-treated material was reduced approximately at the same proportion as the water absorption been reduced with thermo-treated samples, compared with non-treated. (See results of Experiments # 3 and # 6).
3. The coating of wood significantly (10 times!) reduces the swelling of thermo-treated wood.
4. Compared with non-treated wood the swelling for treated wood reduced at the factor of 2-3 for the long time of water treatment. The following experiments with 5 days of water treatment shows the reducion of the water absorption factor for thermo-treated wood to 4-5. (see test data at the end of Report).
Experiment # 8. Water loss of samples with the coated ends. Time interval 7 hours.
The research method. Placing the samples of thermo-treated and untreated wood in open air and measuring its weight. For each species 4-5 samples were used.
Results. For getting more objective results we used the proportion of Vaporized and Absorbed water after 7 hours of drying in open air. The water loss of thermo-treated material approximately at the inverted proportion to the water absorption (the thermo-treated material lost twice more water absorbed before, compared with non-treated wood). (See also results of Experiments # 3 and # 6).
Experiment # 9. Swelling of samples with the coated ends after drying. Time interval 7 hours.
The research method. Placing the samples of thermo-treated and untreated wood open air and measuring its width. For each species 4-5 samples were used.
1. Non-treated wood continues to swell (instead of expected shrinkage) – in average 0,5%.
2. Treated wood stayed at the same size, achieved after 18 hours in water.
Experiment # 10. Data Analysis. Comparing additional (between 3 and 18 hours) water absorption and water loss after 7 hours of drying. Calculation of the percentage of vaporized water, compared with absorbed.
Results. Non-treated wood has lost approximately 50% of additional water compared with treated wood, wich lost 100% and more (the previously absorbed water).
1. Tests have confirmed a lot of well-known properties of thermo-treated wood, researched in Europe: the decreasing of swelling and water absorption in 1.5-2 times with direct water contact, the increasing of water loss for thermo-treated wood, the loss of weight (18-23%), volume (5%) and density (15-18%), also caused the corresponding loss of strength.
2. Tests show the significant improvement of all the properties of thermo-treated wood after applying the surface coating, especially to the ends of the boards.
3. The mentioned above are most critical for work with thermo-treated softwoods and soft (less density) hardwoods like poplar.
4. The first absorption of water by thermo-treated wood (restoration of the moistening) causes the swelling of wood by approximatelly 0.3-0.4% for Ash and Oak, 0.7% for Poplar and 2% for Pine. BUT! The drying and next absorption doesn’t shrink and swell the samples and additional absorbed water easily vaporizes in open air without changing the sizes of wood. The additional experiments show: the dimensions achieved after the first moistening of thermo-treated wood at the cooling stage at the treatment process and after the first moistening rest stable after the next cycles of moistening and drying of treated wood.
5. These factors are more significant for treated softwoods than for treated hardwoods. The results show the significant improvement of physical properties for thermo-treatment of hardwoods, than softwoods.
Changes in wood structure and chemical reactions (ThermoWood, Finland) - see Downloads
As a result of the heat treatment process the wood structure is re-formed, the following pictures show how the structure differs between normal untreated pine and heat treated pine.
Untreated pine Heat-treated pine
Heating wood permanently changes several of its chemical and physical properties. The change in properties is mainly caused by thermic degrading of hemicelluloses. Desired changes start to appear already at about 150 ºC, and the changes continue as the temperature is increased in stages. As a result, swelling and shrinkage due to moisture is decreased, biological durability is improved, colour darkens, several extractives flow from the wood, the wood becomes lighter, equilibrium moisture content decreases, pH decreases, and thermal insulation properties are improved. However, the wood’s rigidity and strength properties are also changed.
Chemical changes (ThermoWood, Finland)
VTT, the Helsinki University of Technology, and the University of Helsinki have published several scientific papers about chemical changes in heat-treated wood as part of their joint project entitled ‘Reaction Mechanisms of Modified Wood’ during 1998–2001. In addition, Risto Kotilainen from the University of Jyväskylä has written a dissertation called ‘Chemical Changes in Wood during Heating at 150–260 ºC’.
Understanding the numerous changes that take place in the physical and chemical structure of wood during the heating process requires a good basic knowledge of its chemical composition, structure, and physical properties.
Reaction mechanisms of heat-treated wood (source: VTT).
The main components of wood (cellulose, hemicelluloses, and lignin) degrade in different ways under heat. Cellulose and lignin degrade more slowly and at higher temperatures than the hemicelluloses. The extractives in the wood degrade more easily, and these compounds evaporate from the wood during the heat treatment.
List of European standards (ThermoWood, Finland)
– EN 20 – 1 Wood preservatives. Determination of the protective effectiveness against Lyctus Brunneus (Stephens). Part 1: Application by surface treatment (laboratory method)
– EN 21 Wood preservatives. Determination of the toxic values against Anobium punctatum (De Geer) by larval transfer (Laboratory method)
– EN 46 Wood preservatives. Determination of the preventive action against recently hatched larvae of Hylotrupes bajulus (Linnaeus) (Laboratory method)
– EN 47 Wood preservatives. Determination of the toxic values against Hylotrupes bajulus (Linnaeus) larvae (Laboratory method)
– EN 84 Wood preservatives. Accelerated ageing of treated wood prior to biological testing. Leaching procedure
– EN 113 Wood preservatives. Test method for determining the protective effectiveness against wood destroying basidiomycetes. Determination of the toxic values
– EN 117 Wood preservatives. Determination of toxic values against Reticulitermes santonensis de Feytaud (Laboratory method)
– EN 252 Field test method for determining the relative protective effectiveness of a wood preservative in ground contact
– EN 302-2 Adhesives for load-bearing timber structures; test methods; part 2: determination of resistance to delamination (laboratory method)
– EN 335 – 1 Durability of wood and wood-based products - Definition of hazard classes of biological attack - Part 1: General
– EN 335 – 2 Durability of wood and wood-based products - Definition of hazard classes of biological attack - Part 2: Application to solid wood
– EN 350 – 1 Durability of wood and wood-based products. Natural durability of solid wood. Part 1: Guide to the principles of testing and classification of the natural durability of wood
– EN 350 – 2 Durability of wood and wood-based products. Natural durability of solid wood. Part 2: Guide to natural durability and treatability of selected wood species of importance in Europe
– EN 392 Glued laminated timber - Shear test glue lines
– EN 408 Timber structures. Structural timber and glued laminated timber. Determination of some physical and mechanical properties
– EN 460 Durability of wood and wood-based products - Natural durability of solid wood - Guide to the durability requirements for wood to be used in hazard classes
– ENV 807 Wood preservatives. Determination of the effectiveness against soft rotting micro-fungi and other soil inhabiting micro-organisms
– EN 927 – 1 Paints and varnishes. Coating materials and coating systems for exterior wood. Part 1: Classification and selection
– EN 927 – 3 Paints and varnishes. Coating materials and coating systems for exterior wood. Part 3: Natural weathering test
– EN 927 – 4 Paints and varnishes. Coating materials and coating systems for exterior wood. Part 4: Assessment of the water-vapour permeability
– EN 927 – 5 Paints and varnishes. Coating materials and coating systems for exterior wood. Part 5: Assessment of the liquid water permeability
– EN 12037 Wood preservatives - Field test method for determining the relative protective effectiveness of a wood preservative exposed out of ground contact - Horizontal lap-joint method
– ISO 5660 – 1 Fire tests; reaction to fire; part 1: rate of heat release from building products (cone calorimeter method)
– ISO 6341 Water quality -- Determination of the inhibition of the mobility of Daphnia magna Straus (Cladocera, Crustacea) -- Acute toxicity test
– ASTM D 3273 Test Method for Resistance to Growth of Mold on the Surface of Interior Coatings In an Environmental Chamber
Properties of heat treated wood (ThermoWood, Finland)
Density is determined by measuring the weight and dimensions of the sample. Thermowood has a lower density than untreated wood. This is mainly due to the changes of the sample mass during the treatment when wood loses its weight. As can be seen from the figure below, the density decreases as higher treatment temperatures are used. However, deviation is high and the coefficient of determination is low, due to natural variation in wood density.
Figure 2-4. The effect of treatment temperature on the density of pine treated for 3 hours at 160-240C. The average density in the temperature range T less than 160 C is 560 kg/m3. The test material was conditioned at RH 65% (source: VTT).
Strength of wood material in general has a strong correlation with density, and ThermoWood has slightly lower density after treatment. Therefore, it is obvious that ThermoWood in some cases has lower strength values. However, the weight-to-strength ratio can remain practically unchanged. The strength of wood is also highly dependent on the moisture content and its relative level below the grain saturation point. ThermoWood can benefit due to its lower equilibrium moisture content.
Two methods for testing bending strength have been used, one using defect-free material over a short span and the other utilizing pieces having natural defects over a longer span. The results (Figure 3-4) show that substantial strength loss in pine starts at temperatures over 220C.
Figure 3-4. The effect of heat treatment temperature on the bending strength of pine, average density 560 kg/m3 (source: VTT).
The results show that heat treatment does not significantly changes the modulus of elasticity of wood (Figure 4-4).
Figure 4-4. Effect of heat treatment temperature on the modulus of elasticity of pine, average density 560 kg/m3 (source: VTT).
The strength of heat-treated (230C, 5 hours) spruce was studied with larger test pieces according to EN 408. Prior to testing, the test pieces were conditioned at 45% and 65% relative humidity. The results are presented in table 1-4. With timber containing knots, the strength values for heat-treated wood are lower than those of untreated wood. This is due to, among other factors, the resins being extracted from the wood.
The reference values for untreated spruce at 12% moisture content are: bending strength 40-50 N/mm2 and modulus of elasticity 9,700-12,000 N/mm2.
In tests conducted on ungraded timber with defects and a span of 1,800 mm treated at 230C for 4 hours (Table 1-4), the bending strength was reduced by up to 40% compared with normal untreated wood. This was due to weakening of areas around the defects. However, with wood treated at lower temperatures of about 190C for 4 hours, the difference in bending strength was far less.
The majority of testing so far has been performed on small, defect-free pieces. More testing is needed on full-size test pieces and with varying numbers of knots and different knot types. In the absence of sufficient information, we recommend that ThermoWood NOT be employed in load-bearing structural usage for the time being.
Screw holding strength
Results from ‘heat treatment of timber’ study performed by the Institute of Environmental technology in 1999 showed that the major impact on screw holding strength was due more to the general variations in wood density than to the heat treatment itself. The study revealed that in lower-density material the results were better when smaller, pre-drilled holes were used.
Compression strength parallel to grain
According to tests conducted by VTT with timber treated at 195C for 3 hours, the compression strength parallel to the grain of the heat-treated timber was about 30% higher than that of normal untreated timber. The tests pieces in this study have been submerged in water before using.
Compression strength is mainly dependent on the actual density of the wood. Tests show that the heat treatment process does not have a negative effect on compression strength values. Actually, the results indicate that the compression strength values were better than with untreated wood even when a higher treatment temperature was used (Figure 5-4).
Figure 5-4. The compression strength (N/mm2) of spruce. Average density 420 kg/m3 (source: VTT).
Tests show that when the maximum compression load was achieved, the pieces broke into smaller sections but didn’t buckle like normal kiln-dried timber. This revealed clearly that heat-treated timber is not as elastic as normally kiln-dried timber.
Impact bending strength (dynamic bending).
From test results (CTBA), it can be understood that the impact strength value for ThermoWood is less than of normal kiln-dried timber. In testing spruce that had been treated for 3 hours at 220C, it was found that the impact strength was reduced by about 25%.
The tests were performed (VTT) by measuring both radial and tangential directions. It was found that with higher-temperature treatments (at 230C for 4 hours) the strength properties were reduced in radial tests from 1 to 25% and in tangential tests from 1 to 40%. However, lower-temperature treatments (at 190C) had very little effect on pine, although spruce showed a 1-20% decrease in both radial and tangential tests.
The splitting tests were performed at the Institute of Environmental Technology with spruce, pine and birch using an extensive range of treatment temperatures. From the test results, it can be concluded that the splitting strength is reduced by 30-40% and the decrease in strength is greater with treatment at higher temperatures.
Brinell hardness has been tested according to prEN 1534. The results show that the hardness increases as the treatment temperature increases (Figure 6-4). However, the relative changes are very small, therefore having no effect in practice. As with all wood species, the Brinell hardness is highly dependent on the density.
Figure 6-4. The effect of heat treatment on the Brinell hardness of pine. Treatment time of 3 hours (source: VTT).
Equilibrium moisture content.
Heat treatment of wood reduces the equilibrium moisture content. Comparisons have been made of heat-treated wood with normal untreated wood at various relative humilities.
Heat treatment clearly reduces the equilibrium moisture content of wood, and at high temperatures (220C) the equilibrium moisture content is about half that of untreated wood. The difference in wood moisture values is higher when the relative humidity is higher. The figure below shows the effects on material treated at 220-225C for 1-3 hours and varying humidities.
Figure 7-4. The effect of relative humidity on moisture content of heat-treated spruce (source: VTT).
Swelling and shrinkage due to moisture.
Heat treatment significantly reduces the tangential and radial swelling (Figures 8-4 and 9-4).
The effect of heat treatment in terms of reduced swelling and shrinkage of wood was clearly shown in relation to cupping of the final product. According to VTT tests, heat-treated wood both with and without a coating maintained its form but CCA-treated and untreated wood were affected by cupping.
Unlike timber in general, heat-treated wood does not feature drying stress. This is a clear advantage, seen when, for example, splitting the material and manufacturing carpentry products. In addition, the wood’s swelling and shrinkage are very low.
The water permeability of heat-treated wood has been tested by CTBA, examining end grain penetration. This feature is important in, for example, windows. Samples were dipped in demineralised water and then kept in a room with a relative humidity of 65% and temperature of 20C. The samples were periodically weighed over a period of 9 days. The conclusion was that during a short period the water permeability of heat-treated spruce was 20-30% lower than that of normal kiln dried spruce.
VTT has tested the steam permeability of heat-treated wood according to EN 927-4. The results are shown in the figure below (Figure 10-4).
Figure 10-4. The effect of heat treatment on steam permeability (source: VTT).
Water permeability was tested by VTT according to EN 927-5 too. Permeability was determined after the pieces soaked in water for 72 hours with their end surfaces sealed. Untreated spruce gained a moisture content of 22%, while the moisture contents of wood, treated at 195C and at 210C were about 12% and 10% respectively.
Tests have shown that the thermal conductivity of heat treated wood is reduced by 20-25% when compared with normal untreated softwoods. Therefore the heat treated material is ideal for outdoor applications like outer doors, cladding, windows, siding, and saunas.
SBI test (EN 13823)
The fire resistance of construction products according to the new Euroclasses was assessed with SBI (Single Burning Item) test. In this test, a specimen consisting of two vertical wings forming a right-angled corner is exposed to flames from a gas burner. The height of the specimen wings is 1,5 m, and their widths are 0,5 and 1,0 m. The gas burner placed at the bottom of the corner stands for single burning item producing a heat attack with a maximum of about 40 kW/m2 on the product tested.
The effect of heat treatment on RHR (Rate of Heat Release) is shown in figure 11-4. The RHR level of heat-treated pine was about 10 kW greater than that of untreated pine. The earlier increase of RHR towards the end of the test for the specimen without heat treatment was due to its smaller thickness. In THR, an increase of about 15% due to heat treatment was observed. Smoke production was roughly doubled. In addition, the ignition time (based on 5 kW increase in RHR) was shortened by 30%. In conclusion, heat treatment seems to degrade the fire resistance of wood. This is probably related to release of volatile compounds during heat treatment. Although the temperature during the treatment is not near the ignition temperature of wood, the constituents of wood can still gradually disintegrate. Consequently, the material properties change, leading to slightly degraded fire resistance.
The number of tests made on ThermoWood has been too low to establish exact values. However it can be stated that ThermoWood does not differ significantly from normal wood when it comes to fire safety. ThermoWood is in fire class D.
Figure 11-4. Rate of heat release of pine specimens with (2/1) and without (3/1) heat treatment. The specimen thickness was 21 and 25 mm for untreated and heat-treated pine boards, respectively.
ISO 5660 test
VTT tested the fire resistance properties of ThermoWood according to ISO 5660. Heat treatment decreased the ignition time for both pine and spruce samples (Tables 4-4 and 5-4) to half that of untreated wood. With pine samples, the rate of heat release (RHR) decreased 32%. The heat-treated spruce samples showed no difference. The production of smoke was small with heat-treated pine and spruce samples in comparison to untreated samples.
Test according to NF B 52501
Tests were carried out by CTBA according to the NF B 52501. All samples studied can be classified in Class M3. The tests indicate that the fire resistance of heat-treated wood has to be considered to be the same as that of untreated wood of corresponding species.
Test to British Standard, surface spread of flame, BS 476 Part 7.
A very limited number of pine and spruce pieces treated at 210C were tested in the United Kingdom in accordance with Class 1 surface spread of flame standard BS 476, Part 7. The results showed that both heat-treated wood species attained a class 4 rating. The standard rating for normally treated wood is class 3. The heat-treated wood exceeded the limit for class 3 within the first minute.
Due to the very small number of tests pieces used, it is concluded that the results cannot be relied upon and more extensive testing is needed using material treated at varying temperatures and moisture contents. The BS tests and results only focused on flame spread speed, and this element is only one part of the testing procedure set forth in new EN standards. The heat-treated wood had a clearly shorter ignition time but was better than the normally dried softwoods in items of heat and smoke release.
VTT carried out three tests to determine the biological durability of heat-treated timber. The tests were carried out in accordance with the EN 113 standard, with a 16-week decay time. In addition, a modification of the EN 113 test was used; the test time was accelerated by using smaller test pieces and a shorter decay time (6 weeks). The third test was made in soil contact according to ENV 807, the test times being 8, 16, 24 and 32 weeks. The test fungi were Coniophora puteana and Poria placenta since these are regarded as the most common and problematic fungi.
The results revealed a remarkable ability of the heat-treated wood to resist decay by brown rot. Against the two fungi, the heat-treated wood showed varying results. The heat-treated wood required a higher treatment temperature in order to gain maximum resistance against Poria placenta compared to resistance against Coniophora puteana (Figure 12-4).
Figure 12-4. The effect of heat treatment on decay by brown rot in a modified EN 113 test. Heat-treated pine, treatment time of 4 hours (source VTT).
The biological resistance test in accordance with EN 113 revealed very good durability depending on the treatment temperature and time. In order to treat the wood to meet the class 1 (very durable) requirements, temperatures of over 220°C for 3 hours are required, and to gain class 2 (durable) status, the desired result is achieved at about 210°C (Figure 13-4).
Figure 13-4. The effect of temperature on the weight loss ratio. Pine, treatment time 3 hours. Standard EN 350-1. Natural durability (source: VTT).
Based on the results of the field test (EN 252), it is recommended that ThermoWood not be used in deep ground applications where structural performance is required. It is assumed that the indicated loss of strength is due to a moisture and not caused by any micro-organism. Establishing the reason behind this phenomenon will require further study. However practical experience has found that usage of Thermo-D material in ground contact where structural performance is not critical and periodic drying of the surfaces is allowed does not cause any significant deterioration to the material. This is especially apparent when the ground has good drainage and is made up of sand or shingle.
Resistance to insects.
Test were carried out by the CBTA in France. Longhom beetles are found in sapwood of softwoods. The common furniture beetle (Anobium punctatum) attacks hardwoods in particular. Lyctus Bruneus is found in some hardwood species. The tests showed that ThermoWood was resistant to all three of the above insects.
Tests made at the University of Kuopio also prove that ThermoWood has good resistance against longhom beetles. The test report concludes that beetles recognize pine from its terpene emissions to be a suitable place for egg laying. Because terpene emissions from ThermoWood are drastically reduced in comparison to normal wood, it is expected that beetles will choose normal wood over ThermoWood, whenever possible. According to the report, the same phenomena can apply also to termites. However, more testing is needed in this area.
Concerning termites, the problem is currently more apparent in southern hemisphere locations, but termites have already spread through France and cases have also been reported in countries further north in Europe. Termites attack buildings from the earth below, avoiding direct sunlight whenever possible. Termites will attack both wood and concrete-based materials in their quest to nutrition. Various measures have been developed to control the problem; these include polythene membranes being installed in the foundations. Also various bituminous paint products are available to seal possible routes up the building. So far the test results indicate that ThermoWood is unable to resist attack from termites. However, local tests are recommended since termite types vary from one region to another.
Durability Class of Thermo-treated wood (Termoholtz, Austria)
Wood endurance increase by means of retification (Institution Technical Wood and Furniture Centre, France) - see Downloads
Retification allows materials to acquire new qualitative characteristics. This process involves reduced pyrolysis of wood that makes the wood more stable and more mycosis-resistant, but slightly impairs its mechanical properties.
The tests were carried out on three major healthful wood types (poplar, fir, and spruce) used in constructions, but hard to apply common products for wood conservation. Test samples of wood retified at different temperatures were exposed to basiodimicite fungi. The test results proved to be stunning. This interesting property can successfully be used for increasing the endurance of materials, originally not resistant to biological effects. The treatment process optimization has improved both the procedure itself and the wood endurance.
Wood retification is heat treatment at 200º - 260º? in the oxygen-unsaturated atmosphere. The treatment combines a series of physical, mechanical, and chemical wood transformations that impart unique endurance to the treated material.
Wood represents a complex material in terms of its elements heterogeneity, their anisotropy, and hydroscopic and biological properties. Under special heat treatment, or retification, that represents a reduced low-temperature pyrolysis (200º - 260º ?), the water absorption capacity of wood noticeably decreases, the volume stability improves, and the resistance to microorganism damage enhances.
The aim of the present investigation is to present the results obtained during the treatment of three major wood types, the two soft (poplar and fir), and the one hard (spruce) ones.
2. Description of the treatment process
Retification is a heat treatment at the temperature of 200º - 260º ? and short time deficiency of air.
The term «retified wood» implies the material obtained by retification of natural wood, i.e. as a result of chemical transformation (creation of new bonds) on molecular level wood components crystallize. This thermal process occurs under specific conditions of pressure, temperature, and at an accurately set level of temperature. The physicochemical composition of natural and treated wood is shown in Table 1.
During the high-temperature treatment, a portion of water contained in wood is extracted. Under these conditions and in an inert atmosphere carbon monoxides and dioxides are released, thus, resulting in the alteration of wood constituents. This is a complicated and multilateral process of heat wood treatment that leads to numerous reactions that occur at different stages of the treatment. However, real-life control of temperature, duration, gas pressure and cooling atmosphere facilitates the thermal condensation reaction of certain constituents of internal wood structure without any losses of main ingredients (cellulose and lignin).
Wood becomes much more moisture-proof during the first minutes of retification; the material releases 4% of moisture into the external atmosphere. The weight decline is accounted for by the fact that water filling up the cracks of pentosan derivatives (semichemical pulp) provides the stability of the dimensions.
The stability of the dimensions appears due to furfural polymers derived from the destruction of sugars that are less hydroscopic than hemocellulose.
3. Improvement of biological damage resistance.
In order to control the improvement of material endurance, three major samples of retified material (Bravery test, 1979) were tested.
The test checked the resistance to the following mycosis:
The poplar samples were also tested for the resistance to Chaetomium globosum Kunze.
12 retified wood samples (30x10x5 mm) were placed into containers with fungi crops for 6 months. At the same time, samples of untreated wood were exposed to the same conditions.
The results of the tests are represented in Tables 2 – 4 (see Downloads).
In all cases the loss of weight by the retified wood samples is considerably less (<1%).
The samples have demonstrated the resistance to biological damage in accordance with standards.
In all cases, the moisture of retified wood was much less, than that of natural wood; and this fact is especially important for the poplar samples.
Putrefactive processes did not affect the poplar samples.
Retification imparts the treated wood with exceptional resistance to biological damage. The low moisture level of all retified wood samples has influenced the improvement of this characteristic, since the absorption level of wood significantly diminishes. Notwithstanding the fact that the hemicellulose content decreases, the potential for destructive processes development is much less than in natural wood.
This increase in wood endurance becomes possible due to chemical and physical modifications of wood during the treatment. This fact suggests that the «new» material possesses interesting properties, and offers ample prospects for the application of soft wood types under environment conditions.
Researches of wood retification (France) - see Downloads
We present here the results concerning common species in Europe and compare them with imported species reputed for their stability (western red cedar, teak). Measurements were carried out according to French standards , [7l, ,  and . Fungi used to determine wood durability were Poria placenla for resinous trees and Serpula lacrymans for poplar.
Reduction of volumetric shrinkage
Western red cedar and teak are classified from very durable to moderately durable (respectively 2-3 and 1-3). Retified wood dimensional stability and resistance to decay are very competitive with those of these species. We even succeed in upgrading fir and poplar from non durable (5) to very durable (1).
Nevertheless, retification may lightly decrease wood mechanical strength because of the modification of its chemical components. On the other hand, a carefully driven retification schedule requires a very good quality pile stacking within kiln: imposing such rules to wood industrialists should lead to an overall improvement of dried wood quality and, as a consequence, to retified wood quality.
Evolution of mechanical properties