Holzforschung, Vol. 59, pp. 389–396, 2005 • Copyright  by Walter de Gruyter • Berlin • New York. DOI 10.1515/HF.2005.064 Hydrophobisation and densification of wood by different chemical treatments Sylvie Bach 1 , Mohamed Naceur Belgacem 1 and Alessandro Gandini 1,2, * 1 LGP2, Ecole Francaise de Papeterie et des Industries ¸ Graphiques (INPG), St Martin d’He ` res, France 2 Instituto de Quı´mica de Sa ˜ o Carlos, Universidade de Sa ˜ o Paulo, Sa ˜o Carlos, SP, Brazil *Corresponding author. Instituto de Quı ´mica de Sa ˜ o Carlos, Universidade de Sa ˜o Paulo, Av. Trabalhador Sa ˜ o Carlense, 400, 13566-590 Sa ˜o Carlos, SP, Brazil Fax: q55-16-3373-9952 E-mail: gandini@iqsc.usp.br Abstract Samples of Pinus sylvestris were treated with different reagents bearing isocyanate, carboxylic anhydride or oxi- rane functions to induce reaction with the OH groups of the various components of wood. Conditions were opti- mised to achieve maximum grafting. When the reagents also carried a polymerisable function, a monomer (sty- rene or methyl methacrylate) was incorporated into the wood morphology after the derivatisation reaction. Sub- sequent radical polymerisation produced chemical incor- poration of some of the resulting macromolecules. The properties of the doubly modified samples were assessed using a wide variety of tests. Both the hydro- phobic character and the density increased considerably, suggesting improved lifetime and extended domains of application. Keywords: acid-base properties; carboxylic anhydrides; densification; hydrophobisation; inverse gas chromatog- raphy (IGC); isocyanates; oxiranes; wood chemical modification. Introduction The modification of wood with the aim of preserving it from various sources of physical, chemical and biochem- ical deterioration has attracted the interest of scientists and technologists for a long time (Banks 1990) and has recently been thoroughly discussed (Goodell et al. 2003). The chemical modification of wood is achieved by the conversion of the hydroxyl groups of cellulose, hemicel- luloses and lignins, mostly into ester or urethane func- tions. Recently, a new strategy for wood modification was proposed, consisting of the use of reagents bearing two reactive functions, one to be used for condensation with the OH groups of the wood components and the other for further modification. The reaction of wood with glycidyl methacrylate (OHqoxirane), followed by the copolymerisation of the appended methacrylic moieties with methyl methacrylate is a typical example illustrating this approach (Rozman et al. 1997). Other suitable re- agents are crotonic and methacrylic anhydride, as stud- ied by Hill and Cetin (2000). In our view, the very few pioneering investigations published within this context lacked thoroughness, particularly in the characterisation of the modified samples, and we felt therefore that a more comprehensive study was required to test the gen- eral validity of the working hypothesis. The present study was therefore based on the assumptions of previous studies and called upon the use of a variety of reagents combining one of three alternative reactive functions (isocyanate, carboxylic anhydride or oxirane) with styrenic or acrylic moieties. After appending each of these molecules to a wood substrate, a mono- mer (styrene or methyl methacrylate) was infiltrated into the medium and its free radical polymerisation induced by typical thermal initiators. The main issues here were that: (1) the actual occurrence of both reactions, first with the wood components and then by copolymerisation with the appended double bonds, had to be proved; and (2) the modified wood should display improved properties. We recently conducted a first study in this context, albeit using a slightly different approach, in which cellu- lose fibres and wood were treated with aromatic reagents bearing two isocyanate or two carboxylic anhydride func- tions at the opposite ends of the molecule (Gandini et al. 2001). Heterogeneous conditions coupled with the choice of stiff molecular structures led to couplings with the substrate that involved only one of the two reactive moieties, thus leaving the second available for further exploitation. Linear polyesters and polyurethanes, syn- thesised in the presence of these modified substrates, were shown to produce covalent bridges between the macromolecule and the substrate surface, since the NCO or anhydride functions, respectively, had copolymerised with the growing chains. This initial investigation allowed us to establish the optimal conditions (temperature, sol- vent, catalyst) to be applied in the present context deal- ing with wood grafting by coupling agents bearing two identical functions. Materials and methods Materials All pine wood samples (Pinus sylvestris L.) were kindly provided by CTBA, France. 3-Isopropenyl-a,a9-dimethylbenzyl isocyanate (IPI), 2-isocyanatoethyl methacrylate (IEM), methacrylic anhy- dride (MA), glycidyl methacrylate (GM), styrene (S) and methyl methacrylate (M) were commercial products supplied by Aldrich (see Figure 1). The double reagents were used as received, whereas the two latter monomers were purified by fractional dis- tillation after removing the inhibitor. All other reagents, catalysts Brought to you by | University of Georgia Libraries Authenticated Download Date | 5/25/15 12:56 AM