AN EXPERIMENTAL RESEARCH REGARDING THE OPTIMUM COMPOSITION FOR WOOD SLAG BOARDS

Wood slag boards (WSB) belong to the group of composites formed on the basis of mineral binders (cement, gypsum, blast-slag, etc.). They are multi-componential systems consisting of wood particles, slag-alkaline and mineralizator. The physico-mechanical properties of WSB completely answer the standard requirements valid for high-density wood-cement boards and therefore can be used as a competitive building material. The economic relevancy of their production is grounded in providing building industry with high-quality and at the same time cheap materials, as well as in utilizing scrap materials which would otherwise pollute the environment (blast slag and scrap wood). The optimizing of the mass parts of the components of WSB is a very important problem both from technological and economical point of view. The success in solving the optimizing problem depends to a great degree on the choice of method. This is way a planned experiment is applied, based on the simplex method for multicomponent system with preliminarily determined limits of calculation of the factors. With the help of specialized computer software, the optimum quantity of the components is determined through graphically presented optimum areas


Introduction
In recent years the production of different kinds of wood-mineral boards used in the building industry as thermal and acoustic materials (light boards with density of up to 600 kg/m 3 or highdensity boards, up to 1200 kg/m 3 , used for construction purposes) has rapidly developed. The basic advantage of these boards compared to the materials produced from wood is that they have higher degree of water-, fire-, and bioresistance, they are more easily processed with cutting instruments, and have smaller bulk mass.
WSB are considered to the type of wood-mineral boards whose production process presupposes not only high economic and ecological effectiveness, but also advanced technological characteristics. Carried out research out (Mihajlova, 1995) shows the following advantages of the production technology of WSB compared to that of wood cement boards: much shorter production [189] cycle due to the method of hot pressing; lower influence of the type of wood species and of the anisotropic properties of wood on the characteristics of the boards; not so significant dependence of the hardening of the slag-alkaline binder on the content of water-soluble and extractive substances in the composition of wood. The influence of the technological parameters on the properties of WSB has also been examined (Yossifov and collective, 1989;Mihajlova, 1995).
The aim of the present research work is to determine the optimum quantity of the mass parts of WSB components.
WSB consist of four basic components: slag-alkaline binder (X1), wood particles (X2), mineralizator (X3), and water (X4). Besides, in order to obtain a real product, variations of the components are allowed only within strictly determined boundaries (0≤ai≤xi≤bi) and the following condition has to be followed: Further on in the research the quantity of water is accepted to be a constant equal to 0.23, i.e. X4=0.23=const. In order to simplify the solution of the optimizing problem, it has to be accepted that the sum of the variable components gives a new unit of measurement: The restrictions on the varying of the components ai and bi, where i=1,2, and 3, have been determind on the basis of preliminary sifting experiments (Mihajlova, 1995). Together with that, the restrictions have been determined to answer the conditions Most often in this case the method Mc Lean -Anderson (Vouchkov, 1986) is applied. It presents the factor space as a polyhedron or a polygon whose form depends on the concrete value of the parameters ai and bi. A planned experiment is carried out with components changing in a simplex space. The two basic physico-mechanical properties of WSB, bending strength (σb) and thickness swelling after 24 h in water (αδ), form the optimum criterion (Yj).

Materials and Methods
Special laboratory experiments for optimizing the composition of WSB have been carried out following the matrix of planned experiment according to the method of Mc Lean -Anderson ( Table 1). The boards are obtained with set density of 1300 kg/m 3 , thickness 18 mm and moisture content 10±2 %. Mixed particles of different broad-leaved hardwood species have been used with a predominant fraction of 2/0,3 mm. A suitable, preliminarily determined (Mihajlova, 1995) composition of the binder has been used: dry component of granulated blast-slag grinded in advance and liquid component of technical caustic-alkali and soluble sodium water glass. Used sodium water glass serves as a mineralizator in order to neutralize the negative influence of simple  [190] sugar contained in wood (Pentosans and Hexans) on the adhesion of the slag-alkaline binder with the wood particles. The preparation of the wood-slag mixture for each board has been carried out in a laboratory mixing device dosing the mass of the components according to the matrix of the plan. The hot pressing followed a three-level diagram with a maximum of the specific pressure 2 MPa, temperature of the pressing plates 120°C and duration of the process 40 min. After pressing comes the conditioning of the boards that goes on for 5 days at temperature of 20 ± 2°C and 60 ± 5 % humidity.
The The plan of the experiment (Table 1) includes tests at the apexes and in the center of the tetragon, as well as in three additional points.