This thesis work objective is the study of a novel forming process for the production of agromaterials: a high-pressure uniaxial thermocompression process for the molding of natural lignocellulosic materials, without pretreatment or binders. The first chapter is the state-of-the-art of « self-bonded » lignocellulosic materials as defined in the domain of wood-based panels. It comprises the study of the effect of the operating conditions even though reported pressures are much lower than the intended pressure in our study: the influence of the biomass type, the use of steam pretreatment and the fibers modification during processing. Data from closely related scientific domains are compared in order to discuss about the possible mechanisms of cohesion. A short technological part describes the process and its evolution over the prototyping steps in this work. Some limits are reported in particular about the major influence of the technological capability on the exploratory field and on the results. At first, cellulose is chosen as a model polymer. A method is developed in order to study the influence of the operating conditions on the compacted materials’ mechanical properties. Pressure has a limited effect above 100 MPa, the molding time has no significant effect (it can be reduced to 3 seconds) and a minimum of moisture content is necessary. Temperature is the most impacting factor and is correlated with higher mechanical properties, higher density and lower moisture uptake of the specimens as well as to a significant decrease of interparticular space on the surface. At 8% moisture and 200°C, steam accumulates in the core of the material which leads to the delamination of the samples. The moisture / temperature couple (0-8% and 175-250°C) is thus specifically studied; above 225°C the effect of delamination is decreased. The best mechanical properties are obtained at 2% and 250°C: 31/70 MPa of stress at break in tensile/bending and 2 and 8 GPa for the corresponding moduli. The structure / property relationship is discussed thanks to the data obtained. Water has a key role yet contradictory because of its plasticizing effect and higher thermal conductivity in one hand, and the accumulation of steam which hinders the cohesion at the core and leads to delamination. The addition of fatty compounds to cellulose increases the water resistance of the specimens: 5% of stearic acid and magnesium stearate increased the water drop penetration time and its surface contact angle. The grafting of octanoïc acid (and its anhydride equivalent) during the molding process is studied and confirmed by CPG analysis with a maximum DS value of 3.9 10-2 for the acid and 4.8 10-2 for the anhydride. Pretreatments (solvent exchange and high pressure homogenizer) are necessary in order to increase the cellulose / grafts contact and to obtain a significant grafting yield. The grafting yield is correlated with a decrease of the bending properties of the grafted materials. The process is then applied to a variety of raw plant materials (crop residues or byproducts of a first processing) with the aim of establishing a link between the physico-chemical properties of the plant materials and the compacted materials’ mechanical properties. Despite a mediocre statistical significance and lower properties of the materials compared to cellulose, the link is confirmed between high cellulose and lignin contents, low extractives, hemicelluloses, ashes and protein contents and high mechanical properties and water resistance. In response to the limits of this technique, an alternative process of transfer molding is proposed, together with preliminary tests that confirm numerous prospects.
展开▼
