Thermal modification of wood means controlled thermal treatment of timber at a temperature above 100 °C so that the treatment improves the dimensional stability of the wood and reduces its equilibrium moisture content and moisture life. Other beneficial effects of thermal modification comprise increased decay resistance and reduced pitch secretion. On the other hand, thermal modification results in darkened wood and impaired bending and splitting resistance. Owing to its properties, thermally modified wood is applicable especially to constructions and furniture exposed to contact with outdoor air.

Thermal wood modifying techniques have been described in WO 95/31680, which focuses on control of the temperature difference between the outer surface and the interior of the wood during the temperature increase so as to prevent cracks in the wood. In this reference, the temperature difference is in the range 10-30 °C and this range is maintained roughly constant during heating. According to this reference, a corresponding controlled temperature difference in the range 10-30 °C between the wood interior and its outer surface can also be maintained during cooling of the wood, during which, unlike the heating process, the higher temperature prevails within the wood.

WO 94/27102 discloses an improved method for producing thermally modified wood, in which the wood treatment temperature is above 150 °C, the treatment being performed with the aid of steam. A weight loss of at least 3% of the wood occurs in the course of the treatment.

WO 01/53812 further describes a procedure for recognizing the thermal modification in wood by measuring the number of free radicals in the wood and by comparing this number with that of free radicals in corresponding untreated wood.

The reference sets forth as a criterion for thermal modification that the number of free radicals in the wood should be at least 1.5-fold, preferably at least double compared to untreated wood.

The publications mentioned above, WO 95/31680, WO 94/27102 and WO 01/53812 are hereby incorporated by reference in the present patent description. In the following, thermal wood modification or thermally modified wood implies that the wood meets the criteria of thermal wood modification or of thermally modified wood defined in at least one of the three publications cited above.

Despite the reduced equilibrium moisture content of thermally modified wood in moist outdoor air, it has now been found that such wood, when brought into direct contact with water or moist soil, absorbs water in abundance. Although thermal modification as such enhances the decay resistance of wood, especially long-term waterlogging still has a detrimental effect on the resistance of the wood. In addition, soaked wood material gets a markedly increased weight.

The object of the present invention is thus to provide a solution in order to prevent waterlogging of thermally modified wood. The method of the invention comprises water-repellency treatment of thermally modified wood by means of a hydrophobic sizing agent reactive to cellulose. The treatment is preferably performed with a solution containing the hydrophobic sizing agent, followed by solvent evaporation by means of drying.

In accordance with the invention, it has been found that a size dissolved in a suittable organic solvent is readily absorbed into thermally modified wood, and this is due to the same fact as is waterlogging of wood, i. e. increased porosity of the wood structure. The preliminary tests conducted in accordance with the invention achieved a contact angle above 90 °, even above 100 °, between the surface of the treated wood and a water droplet deposited onto this, such an angle being an indication of high water repellence.

The entire wood material can be impregnated with the solution containing the sizing agent in order to achieve a constant size concentration. Optionally, the wood can be surface-treated by painting or by immerging it rapidly into an immersion basin, the surfaces of the piece of wood being treated throughout, while the size penetrates into the surface layers alone.

Sizing agents suitable for increasing the hydrophobicity of wood in accordance with the invention include sizing agents reactive with cellulose, the sizing being performed in the neutral or alkaline pH range. Such sizing agents comprise alkyl ketene dimers (AKD) and alkenyl succinic acid anhydrides (ASA) known in paper

and board sizing, among other applications. The paper and board industry uses these sizes in the form of water emulsions, whereas they are most expedient in impregnating thermally modified wood of the invention when dissolved into an organic solvent, such as acetone. In terms of absorbability, the size solution is preferably diluted, with the size concentration in the solution being e. g. 0.01-5% by weight, preferably 0.05-0. 2% by weight. However, the invention also allows for the use of a 100% AKD, ASA or similar size as such without any solvent, in the form of a concentrated solution, dispersion or emulsion, the medium of the dispersion or emulsion of the invention being e. g. water, and the proportion of sizing agent equalling the size concentrations in the solution mentioned above. AKD and ASA can further be jointly used when intermixed.

The length of the alkyl chain contained in the alkyl ketene dimers may vary.

Commercial AKD sizes usually comprise a straight C16-C22 alkyl chain. However, AKD sizes comprising a branched or unsaturated carbon chain are also usable in the invention. AKD size preparations are available both in a solid and a liquid form.

Accordingly, the length of the alkenyl chain in alkenyl succinic acid anhydrides may vary. Commercial ASA sizes usually comprise a C16-C22 alkenyl chain.

In prior art, the article of C. A. S. Hill and S. Mallon in the Journal of Wood Chemistry and Technology, 18 (3), pages 299-311 (1998) explain chemical wood modification, in which one of the chemicals was 2-octene-1-yl succinic acid anhydride, which pertains to the ASA sizes, dissolved in pyridine. However, this was not a case of thermally modified wood, nor did it aim at increased hydrophobicity of the wood, but instead at increased dimensional stability and resistance of the wood and at an increased number of reactive functional groups in the wood by chemical means. The technique described in the article is thus associated with thermal wood modification rather than with wood hydrophobisation.

After the sizing and the evaporation of the solvent or emulsion or dispersion medium of the invention, the wood can be further processed by heating, e. g. by exposing it to a temperature in the range 100-140 °C for 15-60 minutes. The required heating depends on the sizing agent selected; preliminary tests indicate that such after-heating is almost insignificant in AKD sizing, whereas it improves the hydrophobicity obtained with ASA crucially.

The thermally modified wood of the invention treated as above is characterised in having been impregnated with a hydrophobic sizing agent reactive with cellulose, which makes the wood more water-repellent. The size has impregnated at least the surface layers of the piece of wood, so that the surfaces of the piece are protected by the size on all sides. The size may be evenly distributed throughout the wood, with the size concentration in the treated wood being approximately constant.

The invention further comprises the use of a hydrophobic size, such as AKD or ASA, in water-repellency treatment of thermally modified wood.

The invention is illuminated below by means of the following embodiment examples.

Example 1 AKD size containing a C16/C18 alkyl chain was dissolved in acetone to a solubility of 0. 1% by weight and thermal wood marketed under the name Thermowood was thoroughly impregnated with this solution. The wood was dried by evaporating the acetone solvent and the dried AKD treated wood was heated to 130 °C over periods <BR> <BR> of 15,30 and 60 min. , respectively. The water repellence of the wood was tested by a drop test, in which the contact angle formed by the water droplet placed on the wood and the wood itself was measured after contact periods of 1,5 and 10 s., respectively. The bar chart in enclosed figure 1 shows the results of five samples, a <BR> <BR> reference sample (ref. ), i. e. thermal wood not treated with AKD, AKD treated thermal wood, which had been dried but not subjected to a heating process, and AKD treated thermal wood, which had been dried and then heated to 130 °C over periods of 15,30 or 60 minutes. The results show the hydrophilicity of wood not treated with AKD, appearing as a low contact angle, which further decreases rapidly as the contact is prolonged. Thermal wood treated with AKD yielded contact angles of approx. 90-100°, which indicated that the surface was water repellent and devoid of a tendency of water absorption into the wood. It also indicates that a heating treatment following the AKD sizing has only a marginal bearing on the hydrophobicity of the wood and that heating of excessive duration may have a negative impact.

Example 2 The tests of example 1 were reproduced with the use of ASA containing a Cl8-C22- alkenyl chain in lieu of AKD. Using a process solution containing 0. 1% of ASA

dissolved in acetone, the contact angles of a water droplet were determined on dried and impregnated thermal wood as such or after heating treatments of 15,30 or 60 minutes at 130 °C, at the end of contact periods of 1,5 and 10 s. As in example 1, <BR> <BR> the reference (ref. ) comprised a commercial unsized thermal wood as such. The results appear in the bar chart of figure 2 in the drawing. It appears that ASA sizing alone, without heating, does not achieve any marked improvement in the hydrophobicity of thermal wood, whereas excellent hydrophobicity appearing as an approx. 100° contact angle is achieved when ASA sized and dried thermal wood is additionally subjected to a short heating process.