Modifying wood with bio-polyesters: Analysis and performance

Abstract

Oligomers of lactic and glycolic acid (OLA, OGA) and poly(butylene succinate/adipate) (OBS, OBA) were impregnated and heated into wood for dimensional stabilisation. These bio-polyesters variously provide bulking and lumen filling treatments. Solvent extraction of treated wood showed that ca. 50% of the polymer remains in wood with in situ polymerisation evidenced by GPC. OGA readily penetrates the wood cell wall, which occurred only on heating OLA, whereas OBS/OBA was lumen filling with any cell wall penetration only evident after water soaking. Stability of OLA/OGA treated wood was improved, but did not withstand extended leaching, whereas OBS/OBA treated wood presented higher stability once leached. Bending resistance was influenced by OLA/OGA treatments, but original MOE values were recovered after leaching. Overall, the results show the bio-polyester treatment can confer specific properties of the modified wood to suit selected end use applications.

Keywords

Bio-polymers Dimensional stabilisation Poly(butylene adipate)Poly(butylene succinate)Poly(glycolic acid)Poly(lactic acid)Wood modification

Introduction

Natural resources are now to the fore with environmental conservation policies through replacement of fossil based, energy intensive or pollution causing materials. Unfortunately, most natural resources, particularly lignocellulosic materials, exhibit properties that do not reach the standard performances of petrochemical based materials. Wood is susceptible to many natural degradation processes, such as biological, thermal, aqueous, photochemical, chemical and mechanical degradation (Rowell 2005a, 2012). Composed of cellulose, hemicelluloses and lignin, wood is hydrophilic and hygroscopic, and constitutes a feedstock prone to mould, decay and insects. Historically pathogenic attack was first fought with creosote and then chromate-copper-arsenate (CCA) salts impregnated into the wood structure. Although conferring a very long service life to the wood, such products have been gradually regulated and banned due to environmental concern (Evans 2003; Schultz et al. 2007). To fight hygroscopicity, thermal treatments were developed that resulted in hemicellulose degradation and associated decrease in moisture sensitivity and improved wood stability. However, mechanical properties were affected due to relative increases in wood crystallinity contributing to decreased impact and bending resistance, which has hindered the thermo-treated wood market development (Boonstra et al. 2006; Beall 1971, 1972).

Modifying wood components by decreasing their sensitivity to water can lead to improved wood dimensional stability and promote use for outdoor applications. Chemical modification of cellulose and hemicelluloses without any degradation of lignin would result in a stable, high performance material. Chemical modification includes bulking and lumen filling treatments that can be active or passive depending on whether covalent bonding between chemicals and hydroxyl groups of wood is observed (Hill 2006). As active chemical treatments, acetylation typically consists of wood esterification with acetic anhydride, but other treatments have been made with other anhydrides, aldehydes, acid or alkyd chlorides, or methylation reactions (Rowell 2005b). The in situ polymerisation approach is mostly represented by wood furfurylation described as monomeric furfuryl alcohol polymerization within the wood structure (Lande et al. 2004). Many chemicals have been tested in a similar approach including formaldehyde-based or epoxy resins, DMDHEU, maleic anhydride with polyglycerols or unsaturated monomers including styrene (Ibach 2005; Hill 2006; Xie et al. 2005).

With the advantage of both lumen filling and bulking approaches, treatment of solid wood by impregnation and heating of lactic acid oligomers followed by in situ polymerisation with and without chemical catalysis has been reported (Noel et al. 2009a, 2009b). This entirely bio-based treatment, where lactic acid can be derived from corn fermentation, led to improved physical and mechanical properties of the treated wood including dimensional stability (70% ASE), increased hardness and decay resistance. The potential of other bio-polymers as wood stabilisation treatments is evaluated in this paper. The research on lactic acid oligomer wood treatment is extended to poly(glycolic acid), poly(butylene succinate) and poly(butylene adipate), with the treatment of wood using their oligomeric forms. As polar polyesters, all these bio-polymers are expected to show good affinity with wood. Furthermore their monomers and oligomers, as well as co-oligomers contain carboxylic acid groups capable of being grafted via esterification to hydroxyl groups of wood before or on polymerising within wood cell walls and lumens. The impact of molecular structure, steric volume and molecular weights of the different oligomerized monomers, as well as the influence of impregnation temperature and heating duration on treated wood properties have been assessed in terms of in situ polymerisation of bio-oligomers, grafting assessment, anti-swelling efficiency, bending resistance and Brinell hardness.

Experimental methods

Materials

Oven dried beech wood samples (Fagus sylvatica) were used for impregnation and treatment. All chemicals were sourced from Sigma-Aldrich (Switzerland): L(+)-lactic acid solution (≥85%), glycolic acid solution (70% in water, technical grade), 1,4-butanediol (99%), dimethyl succinate (98%), dimethyl adipate (98%) and titanium butoxide (97%).

Polyesters oligomers synthesis

As described by Noël et al. (2014), oligomeric polyesters were synthesized by their direct polymerisation under vacuum, using a four-necked flask (500 mL or 1 L) fitted with magnetic stirrer and reflux condenser linked to an inline cold trap and vacuum pump. Thermometers were used to observe the polymerization, condenser head and heater temperatures.

Lactic acid and glycolic acid oligomerisations

Lactic acid or glycolic acid (200–800 g as an aqueous solution) was poured into the flask. The solution was first heated at 70°C under reduced pressure (150 mbar) as an initial distillation step of 75 min. The initial oligomerisation step involved gradually increasing the temperature to 100°C over 100 min, yielding the oligoester of lactic acid OLA1 (or glycolic acid OGA1). Raising the temperature to 130°C over 160 min gave lactic acid oligoester OLA2 (or glycolic acid OGA2). At the end of the reaction, oligomers were poured in bottles, sealed and cooled to room temperature. OLA1, OLA2 are liquid products at ambient temperature. Glycolic acid oligomers OGA1 and OGA2 tend to solidify giving a wax-like solid product when cooled. Melting of OGA2 was rheologically determined at around 65°C (Noël et al. 2014).

Butylene succinate and butylene adipate oligomerisations

Oligoesters were synthesized by melt polymerisation of dimethyl succinate (OBS1 and OBS2) or dimethyl adipate (OBA2) and 1,4-butanediol, according to literature procedures (Nikolic and Djonlagic 2001). This was achieved by adding a 25% stoichiometric excess of 1,4-butanediol, either in absence (OBS1) or presence of titanium (IV) butoxide as esterification catalyst (OBS2 and OBA2). A mixture of dimethylester, 1,4-butanediol and catalyst (200–800 g) was poured into the flask under a nitrogen purge. The mixture was gradually heated to 180°C over 130 min under reduced pressure (150 mbar). At the end of the reaction, oligomers were poured into bottles, sealed and cooled. OBS1 is a clear liquid product at ambient temperature. OBS2 and OBA2 solidify as a white block when cooled. The melt temperatures of OBS2 and OBA2 were rheologically measured at ca. 75°C and 50°C respectively (Noël et al. 2014). Figure 1 shows the monomers chemical structures and their respective polymerization reactions.

Lactic acid, glycolic acid, dimethyl succinate and dimethyl adipate polymerisation reactions

Wood treatment

Wood samples were immersed in liquid oligomers, either at room temperature (OLA1, OLA2, OGA1) or at 90°C (OGA2, OBS2, OBA2). Containers were then placed in a vacuum oven under reduced pressure (580 mmHg) for 1 to 2 h, then atmospheric pressure over 1 to 2 h.

Impregnated samples were then wiped and set on aluminium foil in a ventilated oven at 120°C for 6 h. Anhydrous sample weight was measured before impregnation, after impregnation and after heat treatment. Weight uptake has been calculated according to equations (1) and (2) (1) (2) where w_i stands for the impregnated sample weight, w_t for the sample weight after complete treatment, and w₀ the oven dried sample weight before treatment.

OLA2 treated samples for ASE, bending test and natural ageing have been impregnated at 90°C instead of ambient temperature (Noël et al. 2014).

For GPC analysis purpose, pure oligomers (OLA1, OLA2, OGA1, OGA2, OBS2 and OBA2) were poured in aluminium containers and set in a ventilated oven at 120°C for 6 h at the same time. They will be referred to as OLA1HT, OLA2HT, OGA1HT, OGA2HT, OBS2HT and OBA2HT for heat treatment.

Table 1 summarizes the different wood treatments, polymerisation and curing conditions.

Table 1

Summary of polymerisation reactions conditions and wood treatment/curing conditions

		Polymerisation conditions				State	Further polymer heating (GPC)		Wood impregnation		Wood treatment
		T_max/°C	Duration/min	Vacuum/mbar	Catalyst	State	T/°C	Duration/h	T/°C	Impregnation cycle	T/°C	Duration/h
OLA1	Lactic acid oligomers	100	100	150	…	Viscous liquid	…	…	22°C	Vaccum/P_atm, 2 to 4h	120	6
OLA1HT		100	100	150	…	Viscous liquid	120	6	…	…
OLA2		130	160	150	…	Viscous liquid	…	…	22°C	Vaccum/P_atm, 2 to 4h	120	6
OLA2HT		130	160	150	…	Viscous liquid	120	6	…	…
OGA1	Glycolic acid oligomers	100	100	150	…	Viscous liquid	…	…	22°C	Vaccum/P_atm, 2 to 4h	120	6
OGA1HT		100	100	150	…	Viscous liquid	120	6	…	…
OGA2		130	160	150	…	Crystallized solid	…	…	90°C	Vaccum/P_atm, 2 to 4h	120	6
OGA2HT		130	160	150	…	Crystallized solid	120	6	…	…
OBS1	Polybutylene succinate oligomers	180	130	150	…	Liquid	…	…	…
OBS2		180	130	150	Ti(OBu)₄	Solid block	…	…	90°C	Vaccum/P_atm, 2 to 4h	120	6
OBS2HT		180	130	150	Ti(OBu)₄	Solid block	120	6	…	…
OBA2	Polybutylene adipate oligomers	180	130	150	Ti(OBu)₄	Solid block	…	…	90°C	Vaccum/P_atm, 2 to 4h	120	6
OBA2HT	Polybutylene adipate oligomers	180	130	150	Ti(OBu)₄	Solid block	120	6	…	…

Extractions

Water leaching consisted of soaking treated samples in distilled water at 20°C for 7 days. After leaching, samples were oven dried at 103°C until constant weight. Weight loss due to water leaching and weight loss of the polymer were calculated as follows (3) where w_t stands for sample weight after complete treatment, and w_we for treated sample weight after water leaching.

Soxhlet extraction was carried out on small chips cut out of sample specimens in chloroform at 60°C for 48 h. After extraction, samples were oven dried at 103°C until constant weight. Weight loss due to solvent extraction has been calculated as (4) where w_t stands for sample weight after complete treatment, and w_se for treated sample weight after solvent extraction (Noël et al. 2014).

Size exclusion chromatography

Molecular weight profiles of oligomers, extracted oligomers and polymerized oligomers were evaluated by SEC in chloroform eluent. Molecular weight was estimated according to polystyrene standards (Noël et al. 2014).

Anti-swelling efficiency – moisture exclusion efficiency

All treated samples (15×15×15 mm, T×L×R) were placed in 95% relative air moisture (RH). Samples dimensions were measured before exposure at dry state (due to treatment process) and after weight stabilisation. Regular weight measurements were made in order to determine the equilibrium moisture content of samples with time.

As some treatments show a strong bulking effect in the cell wall whereas others only penetrate wood lumens, the ASE calculation was based on the corrected swelling calculation of treated samples (S_t) as defined below (Thybring 2013) (5) (6) where V_95%st stands for the treated sample volume after stabilisation at 95%RH, V_t for the treated sample volume before stabilization at 95%RH, V₀ the oven dried sample volume before treatment and S_nt the reference not treated sample swelling (7) where V_95%snt stands for the reference not treated sample volume after stabilisation at 95%RH and V_nt for the oven dried reference not treated sample volume.

As for calculation of the moisture exclusion efficiency, reduced EMC (EMC_R) has been used as basis. MEE is defined by equation (8) (8) where EMC_nt is the reference not treated samples EMC at 95%RH, and EMC_Rt stands for the reduced EMC of treated samples exposed at 95%RH. EMC_t and EMC_Rt are defined as follows (Hill 2008; Thybring 2013) (9) (10) where w_95%st stands for the treated sample weight after stabilisation at 95%RH, w_t for the treated sample weight before stabilisation at 95%RH, and w₀ the oven dried sample weight before treatment.

Swelling and anti-swelling efficiency of samples soaked into liquid water were calculated as well according to formulae (5)–(7) where ‘95%’ can be replaced by ‘lw’ standing for liquid water.

Mechanical performance

Samples of dimensions 140×12×7 mm (L×T×R) have been tested according to a four-point bending method using a Zwick universal testing machine (2·5 kN) and associated software. Samples were tested with linear force application in the elastic zone. Young's modulus was obtained from the stress–strain curve.

Brinell hardness of modified wood was evaluated according to EN 1534 standard procedures using a Zwick testing machine. A normal force load of 1 kN was applied within 15±3s for 25±3s, through a standard 10 mm diameter steel ball. After force releasing, the spherical indentation was measured in two places at 90° angle by means of an optical microscope. Brinell hardness is expressed as HB, calculated using equation (11) (11) where F is the normal load force in N, D is the steel ball diameter in mm, d the indentation print diameter in mm. HB is expressed in kg mm⁻².

Results and discussion

Treatment physical influence on wood

Shown in Table 2 are weight uptakes after oligomer impregnation and curing in the cell wall along with average swelling values of samples due to oligomer impregnation and curing. A significant difference was noticed between treatments according to their physical influence on the treated wood. All treatments penetrate easily into wood as confirmed by weight uptake values, however sample dimensional changes indicate the presence or not of the oligomer treatment within the cell wall.

Table 2

Physical influence of treatments on wood, weight uptake with impregnation (WU_i), weight uptake measured after impregnation and curing (WU_t), swelling observed with impregnation (S_i) and swelling measured after impregnation and curing (S_t)

Ref.	WU_i/%	WU_t/%	S_i/%	S_t/%
OLA1	80·6	56·0	0·9	24·5
OLA2	81·4	63·3	0·5	19·8
OGA1	149·8	91·8	9·1	26·5
OGA2	150·1	102·2	12·4	22·6
OBS2	64·5	64·4	0·6	1·7
OBA2	68·3	67·4	0·1	1·3

All oligomers were observed to penetrate easily in wood, even without an added over-pressure. However, only vacuum/atmospheric pressure cycles allowed weight uptakes from 64 to 150% (Table 2). The highest penetration was achieved by GA oligomers, which also penetrate the wood cell wall as demonstrated by impregnation swelling values of 9·1 and 12·4% for the OGA1 and OGA2 treatments respectively. GA oligomers may penetrate easily owing to a lower steric volume and viscosity. Heat treatment leads to a 38 and 32% decrease of weight uptake (OGA1 and OGA2 resp.) that could be due to residual GA monomer degradation or water loss from further in-situ polycondensation of GA oligomers (Noël et al. 2014). Temperature also promoted greater wood cell wall penetration by oligomers as indicated by swelling values after heat treatment.

LA oligomers only penetrate into wood lumens at ambient temperature as wood samples do not show significant swelling despite high weight uptakes (Table 2). A higher temperature promoted significant LA oligomer penetration into wood cell walls as demonstrated by swelling values (20–25%). Temperature was also associated with weight uptake decreases of 30 and 22% (OLA1 and OLA2 resp.) likely due to water loss on oligomer polycondensation and residual LA monomer degradation.

Despite a higher degree of polymerization and larger steric volume of monomers, OBS2 and OBA2 oligomers exhibit high weight uptakes as impregnation was carried out at 90°C which promoted decreased oligomer viscosity. There was no evidence for OBS or OBA oligomer penetration in the cell wall (Table 2). Oven heating does not decrease weight uptake nor lead to swelling. This suggests no further in situ polymerisation was achieved at this temperature.

In situ polymerisation

Compared to the starting oligomers, LA polymers extracted by solvent from treated samples show a significant shift in molecular weight distributions to higher molecular weight (Mw) fractions as shown in Fig. 2. This tendency to higher M_w was also observed with GA treated samples.

Comparison of lactic acid oligomers (OLA1), after further polymerisation (OLA1HT) and after in situ polymerisation and extraction (OLA1HT extracted)

As shown in Figs. 3 and 4, OBS and OBA oligomers do not show significant changes in polymerisation compared to LA and GA oligomers in wood during heating. A relatively small shift to higher molecular weight distributions was observed with OBS2, but temperature treatment (120°C) did not induce further polymerisation of OBA2 within wood. This interpretation of the extent of in situ polymerisation was based only on the oligomer and polymer fractions extractable from wood. No information is given for oligomeric or polymeric material retained, unextracted in the wood.

Comparison of OBS2 oligomers, after further polymerisation (OBS2HT) and after in situ polymerisation and extraction (OBS2HT extracted)

Comparison of OBA2 oligomers, after further polymerisation (OBA2HT) and after in situ polymerisation and extraction (OBA2HT extracted)

Chemical grafting

To assess the possible esterification reaction and grafting between carboxylic end groups of oligomers and wood hydroxyl groups, a relatively severe solvent extraction was carried out on treated samples. Moreover, water leaching of treated samples was conducted to also promote ester bond hydrolysis, whether this was wood-oligomer ester bond or ester bonds within oligomer chains.

Based on the high impregnation weight uptakes (Table 2), some 40% of LA oligomers were extracted by solvent (Table 3). LA oligomers were also sensitive to water extraction, with only small differences between OLA1 and OLA2 oligomer treatments.

Table 3

Polymer weight loss due to water leaching (WLP_wl) and to solvent extraction (WLP_se)

Ref.	WLP_wl/%	WLP_se/%
OLA1	45·4	38·7
OLA2	44·0	42·7
OGA1	36·4	13·9
OGA2	16·8	6·5
OBS2	22·1	51·2
OBA2	18·0	56·2

GA oligomers appeared to be retained more in wood, whether water or solvent was used. While higher molecular weight of impregnated GA oligomers may have influenced oligomers locking into the wood ultrastructure, in related work we have also observed poor solubility of GA oligomers (Noël et al. 2014).

With OBS2 and OBA2, results suggest these polymers were not readily hydrolysed and extractable in water. Furthermore, despite relatively low impregnation weight uptakes and further polymerisation, solvent extraction only removed half of these impregnated oligomers.

From Table 2 it can be assumed that all oligomers were impregnated in such excess that a significant extraction may have been expected. Nevertheless, despite severe extraction conditions, no more than 56% (OBA2) of impregnated oligomers were extractable. While further analysis will be required to determine any actual oligomer chemical grafting via covalent bonding, it is reasonable to assume in situ polymerisation has at least promoted entanglement of oligomeric and polymer chains around cellulose or hemicellulose fibrils.

Along with attempted hydrolysis, water leaching led to sample volume decreases of 3–8% for OLA and OGA treated samples, whereas OBS2 and OBA2 treated samples experienced a volume increase of 13·2 and 13·4% respectively. A volume decrease is a sign that some of the oligomers in the wood cell wall were extracted together with oligomers residing in wood lumens. A volume increase may suggest hydrolysed OBS2 and OBA2 oligomers were transported and deposited within the wood cell wall by water.

Dimensional stability and moisture exclusion efficiency (MEE)

Both dimensional stabilities in wet conditions and in liquid water were considered in this study. As shown above, OLA2 treated samples were impregnated at 90°C instead of ambient temperature leading to the following properties: WU_i = 132·3%, WU_t = 119·0%, S_i = 21·4%, S_t = 23·4%.

As observed in Table 4 weight uptakes and bulking coefficients were different for each treatment, with significant differences between EMC_t and EMCR_t. As reviewed by Thybring (2013), it is commonly accepted that an EMC <25% is correlated to a MEE >40% which can be considered as a threshold under which wood decay appears to be negligible. With the exception of OLA2 treatment achieved with hot impregnation, no other treatment conferred an EMC lower than 25% for treated samples at 95% RH. The EMCR_t values at 95%RH seem to be dependent on the presence of polymers in the wood cell wall. Both OBS and OBA treated samples undergo a 20% swelling due to moisture conditioning, whereas OLA and OGA treated sample dimensions show minimal change where EMCR_t was in the same range. As observed by Hill and Jones (1999), Papadopoulos et al. (2004) and Hill and Mallon (1998) using carboxylic or cyclic anhydrides as wood modification agents, OLA and OGA treatments most likely swelled the wood cell wall creating empty volumes allowing water/moisture to penetrate. This hypothesis would explain the excellent ASE* values obtained together with high EMCR_t which was consistent with swelling on treatment (Table 2).

Table 4

Equilibrium moisture content (EMC), reduced EMC (EMC_Rt), swelling (S_t*), anti-swelling efficiency (ASE*) and moisture exclusion efficiency (MEE) of treated samples

	95% RH/23°C					Liquid water 23°C
Ref.	EMC_t/%	EMC_Rt/%	S_t*/%	ASE*/%	MEE/%	S_t*/%	ASE*/%
OLA1	35·5	56·8	0·5	97·2	−108·5	0·4	98·1
OLA2	8·4	18·5	1·6	91·3	32·2	0·9	95·7
OGA1	19·6	39·9	1·3	93·1	−46·5	−0·7	103·1
OGA2	18·3	38·7	1·2	93·6	−42·1	1·9	91·1
OBS2	21·7	37·2	19·2	−5·9	−36·7	17·2	18·5
OBA2	18·1	30·5	20·2	−11·2	−11·8	20·8	1·7
Ref.	27·2		18·2			21·1

According to ASE values obtained by water soaking, all treatments demonstrating a high bulking coefficient have a very high efficiency, whereas OBS and OBA treatments did not prevent dimensional changes in samples (Table 4).

Hot impregnation (OLA2) seems to have a positive influence on moisture exclusion efficiency, as it is the only treatment leading to an acceptable MEE result of 32·2%.

Samples submitted to an initial water soaking ASE cycle (ca. 700 h at 23°C) were then dried and subjected to a second ASE cycle at 95% RH/23°C, as consistent ASE values have to be calculated on treated samples leached by water. Table 5 shows the results of treatment efficiency values.

Table 5

Equilibrium moisture content (EMC), reduced EMC (EMC_Rt), swelling (S_t*), anti-swelling efficiency (ASE*) and moisture exclusion efficiency (MEE) of treated leached samples, along with polymer weight loss due to water leaching

	95%RH/23°C
Ref.	EMC_t/%	EMC_Rt/%	S_t*/%	ASE*/%	MEE/%	WLP_we/%
OLA1	20·4	22·7	12·9	23·6	19·1	80·5
OLA2	20·4	25·2	13·3	21·3	10·3	80·0
OGA1	27·9	32·3	18·8	−11·7	−15·3	85·0
OGA2	22·5	31·3	17·9	−6·3	−11·7	65·4
OBS2	17·7	27·6	9·6	42·9	1·8	23·4
OBA2	15·5	24·9	9·1	46·0	11·2	13·9
Ref.	28·1		16·9

Samples treated with OLA, OBS and OBA treatments reach an EMCR_t closer to 25%. However, high weight loss of polymer during water leaching very likely extracted part of the oligomers from the wood cell wall. Therefore significant swelling was noticed during ASE testing contributing to lower ASE* values. Water leaching of OGA treated samples lead to negative ASE* and MEE values. Unfortunately, hot impregnation did not prevent water leaching and decreased the MEE of the OLA2 treatment.

Mechanical evaluation

To assess if the wood structure or cell cohesion has been degraded by the oligomer treatments, four-point bending testing was carried out. Compared to untreated reference samples, OGA treated samples were significantly softened, while the OLA treated samples MOE values were also significantly decreased (Table 5). However, both OBS and OBA treatments do not impact MOE substantially. For OLA and OGA, it can be assumed that the short heating procedure chosen for in situ polymerisation did not lead to complete polymerization of the oligomers in wood, but rather to an intermediate state where monomers, short oligomers and water as co-product are present in the wood ultra-structure leading to significant changes in plasticity. It might also be assumed that the acidic conditions could have partially degraded wood components, which, combined with oligomers presence, leads to wood component softening and mechanical slippage between wood fibres. MOE decrease was substantial for OGA1 and OGA2, and correlated with the high weight uptakes of both treatments (Table 2). The evaluation of MOE of the same samples after water leaching procedure (Table 6) indicates the first hypothesis. After water leaching and elimination of short to medium size oligomers, all samples recover their initial bending strength or exceed it. This observation indicates that on removal of monomers and short oligomers OLA and OGA treatments decrease wood elasticity. For OBS and OBA which have no oligomers or polymers impregnated into wood cell wall, any plasticization effect was not evident with OBS and OBA treatments having no significant influence on wood bending properties.

Table 6

Young modulus (MOE) of treated samples and treated/water leached samples obtained by four-point bending test and Brinell hardness (HB)

Ref.	MOE/MPa	MOE leached samples/MPa	HB/kg mm⁻²
OLA1	16·4	24·2	5·3
OLA2	18·2	22·9	5·9
OGA1	4·3	23·6	1·9
OGA2	10·9	29·7	4·8
OBS2	22·5	23·5	4·8
OBA2	25·4	21·7	4·2
Ref.	22·9	19·2	3·7

Brinell hardness testing confirmed the observation with OGA1 (Table 6) which was the only treatment leading to a significant hardness decrease when compared to untreated wood. All other treatments tended to increase surface hardness which indicates that in tangential compression, oligomers and polymers in cell walls or lumens participate to harden wood surface.

Conclusion

Wood chemical modification with OLA, OGA, OBS and OBA bio-polyesters can be divided into two kinds of treatment: bulking modification by OLA and OGA, and lumen filling treatments by OBS and OBA. Bulking treatments were proven to have further polymerized into wood structure and observed to be fixed with the wood (a max. of 15 to 45% of polymer extractable). Treatments also confer very high ASE* to modified samples (ca. 95% ASE* in high RH or in water) however, this was not retained after extended water leaching (<25% ASE*). High weight uptake and cell wall penetration along with a reduced polymerization state induces a large MOE decrease on bulking treatment.

Lumen filling treatments did not extensively polymerize within the wood structure and were extracted (55%) only on solvent extraction, indicative of some fixation within the wood structure. Being lumen filling only, these treatments did not penetrate the cell wall and cannot confer any significant ASE* to the modified wood. However, extensive water leaching appeared to transfer hydrolysed oligomer material into the cell wall as suggested by swelling and increased ASE* after water leaching. With an ASE* value of around 40% after leaching and stable mechanical properties, OBS and OBA treatments appear to be most promising for modification of the wood.

Overall, results show that an optimisation of oligomer treatment parameters can lead to significant improvement in wood properties. A lower oligomer weight uptake combined with higher treatment temperature could prevent excessive cell wall swelling due to treatment and improve MEE of OLA and OGA treatments. Higher treatment temperatures appear needed to induce further in situ polymerisation and promote the chemical reaction of all oligomer treatments within wood.

Footnotes

Acknowledgements

The authors would like to thank the COST Action FP1006 for financing this research project (funded project no. C11·0129), the COST Office for STSM funding (COSTS-STSM-RA – New Zealand-06400) and contributing to ECMW7 travel. We would also like to acknowledge Scion (& Dr. Elizabeth Dunningham) for availability of personnel and equipment resourcing and Armin Thumm for GPC measurements.

The paper is based on the presentation given at the Seventh European Conference on Wood Modification, 2014.

References

Beall

. 1971. Differential calometric analysis of wood and wood components. Wood Science and Technology, 5, pp. 159–75.

Beall

. 1972. Introduction to thermal analysis in the combustion of wood. Wood Science, 5, pp. 102–108.

Boonstra

et al. 2006. Microstructural and physical aspects of heat treated wood. Part 1. Softwoods. Maderas: Ciencia y Tecnologia, 8, pp. 193–208.

Boonstra

et al. 2006. Microstructural and physical aspects of heat treated wood. Part 2. Hardwoods. Maderas: Ciencia y Tecnologia, 8, pp. 209–17.

Evans

. 2003. Emerging technologies in wood protection. Forest Products Journal, 53 (1), pp. 14–22.

Ibach

and Ellis

. 2005. Lumen modifications. In: Handbook of wood chemistry and wood composites. Boca Raton, FL: CRC press. p. 15.

Hill

CAS

and Mallon

. 1998. The chemical modification of Scots pine with succinic anhydride or octenyl succinic anhydride. I. Dimensional stabilisation. Holzforschung, 52, pp. 427–33.

Hill

CAS

and Jones

. 1999. Dimensional changes in Corsican pine sapwood due to chemical modification with linear chain anhydrides. Holzforschung, 53, pp. 267–71.

Lande

, Westin

and Schneider

. 2004. Eco-efficient wood protection: furfurylated wood as alternative to traditional wood preservation. Management of Environmental Quality and International Journal, 15 (5), pp. 529–40.

10.

Noël

et al. 2009a. Lactic acid/wood based composite material. Part 1: Synthesis and characterization. Bioresource Technology, 100 (20), pp. 4711–16.

11.

Noël

et al. 2009b. Lactic acid/wood based composite material. Part 2: Physical and mechanical performance. Bioresource Technology, 100 (20), pp. 4717–22.

12.

Noël

, Grigsby

and Volkmer

. 2014. Evaluating the extent of bio-polyester polymerization in wood by thermogravimetric analysis. in J. Wood Chem. Technol, DOI: 10.1080/02773813.2014.962154.

13.

Papadopoulos

, Hill

CAS

and Gkaraveli

. 2004. Analysis of the swelling behaviour of chemically modified softwood: a novel approach. Holz als Roh- und Werkstoff, 62, pp. 107–12.

14.

Rowell

. 2005a. Moisture properties. In: Handbook of wood chemistry and wood composites. Boca Raton, FL: Taylor and Francis. p. 4.

15.

Rowell

., 2005b. Chemical modification of wood. In: Handbook of wood chemistry and wood composites. Boca Raton, FL: Taylor and Francis. p. 14.

16.

Rowell

, Pettersen

and Tshabalala

. 2012. Cell wall chemistry, in: Handbook of wood chemistry and wood composites. 2nd ed. Boca Raton, FL: Taylor and Francis.

17.

Schultz

, Nicholas

and Preston

. 2007. A brief review of the past, present and future of wood preservation. Pest Management Science, 63 (8), pp. 784–88.

18.

Thybring

. 2013. The decay resistance of modified wood influenced by moisture exclusion and swelling reduction. International Biodeterioration and Biodegradation, 82, pp. 87–95.

19.

Xie

et al. 2005. Weathering of wood modified with the N-methylol compound 1, 3-dimethylol-4,5-dihydroxyethyleneurea. Polym. Degrad. Stab. 89, pp. 189–99.