Modification of wood caused an increase in WPG with increasing concentrations of DMDHEU or GA due to incorporation of chemicals in the wood structures (Figure 1). After leaching in water, the WPGs of treated wood were reduced due to removal of non-reacted chemicals and water-soluble wood extractives. Veneers treated with DMDHEU obtained higher WPGs than did those treated with GA at the same treating concentration. This can be explained by the greater molar weight of DMDHEU (178 g mol^-1) compared with GA (100 g mol^-1). In addition, major parts of GA might evaporate during oven drying, while DMDHEU readily undergoes auto-condensation, which prevents evaporation. Modification with DMDHEU or GA apparently did not change the colour of wood strips (not shown). Treatments with the Fenton solution for 48 h apparently darkened the unmodified wood strips due to formation of chromophoric groups as a result of lignin oxidation (Xie et al. 2012). The modified wood exhibited less colour change; the colour of modified wood with 2.0 mol l^-1 chemicals was comparable with that of the unmodified veneer strips stored solely in buffer. The IR spectra of veneers treated with 2.0 mol l^-1 DMDHEU or GA wood did not change due to Fenton treatment (not shown). At comparable concentration, the lightness of GA-modified wood was greater than that of DMDHEU-modified wood (Figure 2), which shows that GA modification might be more efficient than DMDHEU modification in preventing lignin oxidation.

Figure 1

WPG of veneers treated with various concentrations of DMDHEU and GA before and after water leaching.

Figure 2

Appearances of veneer strips unmodified and modified with DMDHEU or GA after incubation in a solution containing Fenton reagent for 48 h.

(a) Unmodified, stored in buffer; (b) unmodified, Fenton treated; (c) 0.4 mol l^-1 DMDHEU, Fenton treated; (d) 1.2 mol l^-1 DMDHEU, Fenton treated; (e) 2.0 mol l^-1 DMDHEU, Fenton treated; (f) 0.4 mol l^-1 GA, Fenton treated; (g) 1.2 mol l^-1 GA, Fenton treated; (h) 2.0 mol l^-1 GA, Fenton treated. The number in the right corner shows the average value of lightness (n=3 veneers) and standard deviation (number in parenthesis) of 10 measurements.

The veneers exhibited minor ML in the initial 6 h of Fenton treatment (Figure 3a, b). In this stage, degradation of cell wall polymers is likely to occur, but the amounts of water-soluble compounds were obviously low. The ML of unmodified veneers after 48 h of exposure to Fenton reagent amounted to 18.0% and was somewhat lower than that described by Xie et al. (2012). This is attributed to the lower pH (3.0) and higher temperature (30°C) in the previous study. Under these conditions, the rate of the Fenton reaction is known to be higher (Pérez et al. 2002; Jung et al. 2009). Modification with DMDHEU and GA reduced the ML during the exposure to Fenton reagent; the increasing WPGs of the treated veneers further reduced the ML (Figure 3a,b). After 48 h of Fenton treatment, veneers modified with 2.0 mol l^-1 chemicals exhibited only an ML of 4.0% (DMDHEU) and 2.4% (GA). The MLs of GA-modified veneers were slightly less than those of DMDHEU-modified wood compared at equal treatment concentrations.

Figure 3

ML (a, b) and strength loss (c, d) of veneer strips treated with various concentrations of DMDHEU (a, c) and GA (b, d) during incubation with Fenton reagent.

As reported previously (Xie et al. 2005, 2007; Xiao et al. 2010), the process of chemical modification reduced wet tensile strength by up to 50%, especially at high chemical concentrations (see values at 0 h incubation time in Figure 3c,d). This tensile strength loss is attributed to hydrolysis of cell wall polysaccharides catalysed by the Lewis acid MgCl₂^·6 H₂ O, cross-linking of cell wall polymers, and embedding of the resin molecules in the cell wall matrix. Veneers unmodified and modified with 0.4 mol^-1 DMDHEU or GA, which exhibited a comparable strength prior to Fenton treatment, revealed a linear loss in tensile strength with the incubation time. At high chemical concentrations (1.2 and 2.0 mol l^-1), the tensile strength of the veneers was hardly affected by Fenton treatment.

The nitrogen content of the unmodified veneers was below the detection limit (0.05%). Modification with DMDHEU increased the N content as a function of the concentration of the modification chemical (Figure 4a). The N content in veneers treated with 0.4 mol l^-1 DMDHEU was reduced considerably during Fenton treatment, which was accompanied with increased detection of gaseous nitrogen compounds (Figure 4b). This is a sign of degradation of the insoluble DMDHEU-cell wall complex, which leads to mineralised gaseous products. The N content in the treating solution was not determined, but it is assumed that the modified veneers release water-soluble degradation products prior to mineralisation.

Figure 4

Nitrogen content in DMDHEU-modified veneers (a) and nitrogen gas in serum bottles produced during incubation with Fenton reagent (b).

A similar degradation process of DMDHEU-modified veneers due to weathering has been shown previously, where the ML of DMDHEU occurred faster than the ML of veneer (Xie et al. 2005). N loss is consistent with ML and tensile strength reduction. Veneers modified with 2.0 mol l^-1 DMDHEU exhibited a slight N reduction within the first 2 h of exposure, which might be attributable to leaching of unfixed DMDHEU rather than to degradation by Fenton reagent.

Despite higher N contents, veneers with higher WPGs of DMDHEU produced minor amounts of nitrogen gas after exposure to Fenton reagent for 6 h, after which gas production did not increase. It is therefore assumed that the enhanced initial nitrogen gas production is due to mineralisation of leached, unfixed DMDHEU rather than the mineralisation of the DMDHEU-containing cell wall complex. In total, the amount of DMDHEU released from the veneers is relatively small compared with the WPGs of wood.

The concentration of H₂ O₂ in the solution containing untreated veneers decreased linearly from 50 to 30 mmol l^-1 in the course of the incubation (Figure 5). Solutions containing veneers modified with 0.4 mol l^-1 DMDHEU or GA revealed a slightly lower decomposition rate of H₂ O₂ than did those with untreated veneers. The decomposition rate further reduced in the presence of veneers of increasing WPG. At the highest chemical concentration (2.0 mol l^-1), H₂ O₂ concentration decreased by only 6.5% (DMDHEU) and 8.5% (GA). This shows the inhibiting effect of the treatments with respect to the reducing decay of H₂ O₂, which results in the formation of hydroxyl radicals.

Figure 5

Hydrogen peroxide concentration (a, b), concentration of CO₂ (c, d), and O₂ (e, f) in the serum bottles containing veneers treated with various concentrations of DMDHEU (a, c, e) and GA (b, d, f) during incubation with Fenton reagent.

Under the conditions of this study, decomposition of H₂ O₂ proceeds mainly via the Fenton reaction. To enhance the turnover of the reaction, ferric ions must be re-reduced to ferrous ions. Phenolic constituents of wood might function as redox partners, such as lignin (with its free phenolic groups), its degradation products, or phenolic extractives (Xie et al. 2010). Chemical modification might block phenolic hydroxyl groups and thus inhibit the reduction of ferric ions; this would result in a reduced decomposition rate of H₂ O₂ via the Fenton reaction. However, other modes of action are possible (see below).

Production of CO₂ due to the Fenton reaction was highest in the solution containing untreated veneers; the amount increased exponentially with the incubation time (Figure 5a, b). The increasing rate of CO₂ can be explained by the increasing volume of pores in the cell wall, which leads to exposure of more cell wall material and/or by the enhanced reduction of ferric ions by fragments of cell wall polymers, particularly lignin, which drive the Fenton reaction. Fenton reagent in acetate buffer at pH 3.0 (without wood or chelators) produces CO₂ by oxidative degradation of acetate (Xie et al. 2012). In the present study, however, which applies acetate buffer at pH 4.0, CO₂ was not produced from buffer constituents during the incubation (not shown). This indicates that the Fenton reaction hardly proceeds in buffer alone because reduction of ferric ions does not occur. If it is assumed that this reduction initially proceeds in the wood with the participation of wood constituents, particularly lignin, the CO₂ produced should derive mainly from mineralisation of wood and not of buffer. In a later phase, however, these degradation products might also promote the formation of hydroxyl radials outside the veneers that may also mineralise acetate buffer.

The total mass of unmodified veneer strips was initially 75 mg, and 18% was lost after 48 h of Fenton treatment (Figure 3a). As the carbon content of Scots pine sapwood amounts to 47.4% (Xie et al. 2005), the Fenton system should produce 10.6 mmol l^-1 CO₂, if the ML would be completely due to mineralisation. However, approximately 8.0 mmol l^-1 was actually produced (Figure 5c), indicating that at least 2% of decomposed cell wall fragments remained in the solution. Modification of the veneers with DMDHEU and GA caused a decrease in the production rate and total amount of CO₂ during the treatment with Fenton reagent (Figure 5c, d).

Veneers modified with 2.0 mol l^-1 DMDHEU and GA revealed MLs of 4.0% and 2.4%, respectively, after 48 h of exposure to Fenton reagent (Figure 3a, b). If these MLs would derive exclusively from leached DMDHEU and GA, which were completely mineralised, they would produce 1.7 and 1.8 mmol l^-1 CO₂, respectively. As the actual amount of CO₂ produced was 3.0 mmol l^-1, some mineralisation of the modified veneers must have occurred.

The Fenton reaction also produced molecular oxygen, which derives from decayed H₂ O₂ (Figure 5e,f). In the presence of unmodified veneers, 20 mmol l^-1 H₂ O₂ was decomposed in the solution within 48 h (Figure 5a, b). Complete conversion of H₂ O₂ would produce 10 mmol l^-1 O₂. Actually, only 2.3 mmol l^-1 O₂ was obtained; the rest was converted to CO₂ via oxidative decay of wood and acetate/acetic acid. Modification of veneers with 0.4 mmol l^-1 DMDHEU did not reduce the production of O₂ during incubation with Fenton reagent as compared with the unmodified veneers. Increasing WPGs of DMDHEU, however, decreased O₂ production (Figure 5e). This is consistent with less decomposition of H₂ O₂ and lower production of CO₂.

The Fenton system containing GA-modified veneers exhibited a similar trend in production of O₂ to that containing DMDHEU-modified veneers, but the former produced slightly less amounts of O₂ and CO₂ compared at the same treatment concentrations. Accordingly, modified wood is less susceptible to decay by the Fenton reaction.

Hill et al. (2005) demonstrated that acetylated wood suffers less ML due to brown rot decay as a function of increasing WPG through acetylation and attributed this to micro-pore blocking, which prevents the penetration of the decay agent into the cell wall (Hill et al. 2005) and/or to reduced cell wall MC (Hill 2009). A direct effect of chemical modification on the resistance of wood to isolated decay agent has not yet been shown. Reduced mineralisation of the veneers reveals that modified veneers with increasing WPGs became increasingly resistant to oxidative degradation by hydroxyl radicals produced by Fenton reagent. Less tensile strength loss indirectly indicates less degradation of polysaccharides, while lower degrees of discoloration indicate lower oxidative decay of lignin. There are two interpretations for these findings: (a) Fenton reagent is unable to penetrate into the cell wall matrix or (b) chemical modification prevents the reduction of the ferric form of iron to the ferrous form. This reduction might be caused by wood constituents such as phenolic compounds of lignin or extractives or by phenolic compounds produced by the fungus (Goodell et al. 1997; Kerem et al. 1999; Xu and Goodell 2001; Qian et al. 2002). In previous studies, however, the addition of metal-binding, phenolic compound with MW_low DHBA to Fenton reagent did not enhance strength loss and mineralisation of unmodified micro-veneers (Xie et al. 2010, 2012) and was therefore not considered in this study.

It is assumed that ferrous ions are chelated by two acetate molecules, and ferric ions, by three acetate molecules; these complexes are much smaller than extracellular enzymes responsible for cell wall degradation. Increased cell wall bulking or cross-linking due to the modifications might reduce the diameters of cell wall micro-pores and the availability of water as a transport medium, thereby inhibiting penetration by iron complexes (mechanism a; Papadopoulos and Hill 2002; Xiao et al. 2012). Reduced penetration of iron complexes might also disable reduction of ferric irons by free phenolic groups of lignin. Despite lower WPGs, GA-modified veneers were somewhat more resistant to degradation by Fenton reagent than DMDHEU-modified ones were. This might be attributed to a stronger degree of cross-linking by GA and smaller pore diameters resulting thereof.

Both DMDHEU and GA might react with phenolic hydroxyl groups in lignin and thus prevent reduction of ferric ions via free phenolic groups (mechanism b). At this juncture, it cannot be elucidated which mechanism is predominant. DMDHEU mainly undergoes auto-condensation and GA also undergoes electrophilic aromatic substitution of lignin or condensation with hydroxyl groups of polysaccharides. Still, for both chemicals, reaction with phenolic hydroxyl groups in lignin is possible.