In 1994 approximately 400 000 m³ of Pinus radiata D. Don was treated with boron based preservatives. The current preservative process involves either the immersion or spraying of green timber with concentrated boric acid solutions followed by stacking under cover for considerable lengths of time. The New Zealand Forest Research Institute has proposed an alternative to this aqueous phase process which is commonly called the vapour boron timber-treatment process.

The vapour boron timber-treatment process involves drying the timber to a low moisture content (preferably in a high-temperature kiln) and then placing the timber which is to be treated in the treatment vessel. The treatment vessel is then heated and evacuated and the preservative (usually trimethyl borate / methanol azeotrope) is injected into the vessel as a liquid. As the preservative evaporates in the reduced pressure and the elevated temperature of the vessel, the pressure in the vessel rises rapidly causing a pressure wave of gas to flow into the timber. As this pressure wave travels through the timber the trimethyl borate reacts with any water vapour still present and the adsorbed water in timber to form boric acid (which acts as the preservative) and methanol. Subsequently, the trimethyl borate left in the treatment vessel diffuses through the timber and reacts with the adsorbed water. A second vacuum removes the methanol from the timber.

The modelling of this process involves both the study of the chemical kinetics of the reaction between trimethyl borate and water and the study of the flow of gases through and/or into Pinus radiata D. Don.

The trimethyl borate reacts with both adsorbed water and water vapour. Both reactions proceed by the sequential replacement of the three methyl groups with hydroxyl groups until the boric acid molecule is produced. The reaction rate constant for the reaction between trimethyl borate and water vapour has been experimentally determined by drawing water vapour and trimethyl borate through a transparent glass tube, noting the position of the build-up of boric acid on the walls of the tube. The reaction rate constant was determined to be 6.8 x 10⁻³ ± 0. 7 x 10⁻³ Pa⁻¹ s⁻¹.

The reaction rate between the adsorbed water and trimethyl borate is related to the moisture content of the wood. At low moisture contents the water that is adsorbed onto the wood surface is more tightly bound than water which is adsorbed at higher moisture contents. As the activation energy for any reaction between adsorbed water and trimethyl borate will be a function of the adsorption energy, the reaction rate also is a function of the moisture content.

The predominant flowpath for the flow of gases and liquids under a constant pressure gradient in Pinus radiata D. Don is via the axial and radial resin canal network. There is some evidence that some of the flowpaths may be less than the mean free path-length of the gas as molecular flow occurred in a few samples. However, the flow can be modelled accurately by Darcy's Law. The radial permeability of Pinus radiata D. Don increases with the severity of drying; air-dried wood has a lower permeability than wood dried in a kiln with a wet-bulb temperature of 60°C and a dry-bulb temperature of 70°C. High-temperature dried Pinus radiata D. Don has a still higher permeability. As both heart- and sap-wood displayed the same trend, the increase in permeability is presumed to be caused by modification of the resinous deposits and blockages in the resin canals.

The unsteady-state flow of gases (where the pressure gradients change with time) can be modelled using only one resistance, however where molecular flow was present this assumption was not exact.

The proposed model of the vapour boron timber-treatment process involves flow of gases by permeation and diffusion. The diffusional flow has been approximated to a binary system with methanol, air, and water vapour making up one pseudo-species, and trimethyl borate the other. The model has been solved numerically using the method of lines and an implementation of the Adams-Moulton method.

The equations which make up the model form a family of four, coupled, partial differential equations. These equations appear to be parabolic (second order); however, as some of the species have a zero concentration (due to reaction) the equations become first-order in nature. This causes severe numerical difficulties when a central difference approximation to the partial differential equation is used. These numerical problems have been solved by a using mixed central and backwards approximation to the partial differential equations.

Simulations of laboratory-scale experimental treatments have shown that the model accurately predicts the boric acid deposition profile. These simulations show that moisture content is by far the dominant factor in determining the shape of the boric acid retention profile. At low moisture contents, the profile is quite flat; at high moisture contents the majority of the boric acid is deposited in the outer 10 mm of the sample.