Relationship between Anatomical Properties and some Physical and Mechanical properties for Five Wood Species Growing in North Darfur- Sudan

Abstract

This study was conducted to assess the Relationships between anatomical properties and some physical and mechanical properties of five wood species growing in North Darfur, Sudan. which consist of the following species (Acacia seyal (Talh), Acacia nilotica (Sunt), Balanites aegyptiaca (Heglig), Eucalyptus camaldulensis (Kafour), and Faidherbia Albida Haraz). The main objective of this study was to determine some anatomical, physical, mechanical, and technological properties of the studied wood species and establish relationships between anatomical properties, as independent variables, and the other properties, as dependent variables, in the form of mathematical models with which we can predict physical and mechanical properties from anatomical properties and find the most important anatomical properties affecting the other properties. All anatomical properties were determined from macerated fibers. Wood chips from the five wood species were macerated by boiling them in concentrated nitric acid for ten minutes. The liberated fibers were stained in safranin for five minutes, washed by alcohol and water and mounted on glass slides ready for examination and measurements. The fiber length was determined using stereological techniques on macerated fiber. The other anatomical characteristics measured included fiber length (FLmm), fiber diameter (FDµm), fiber lumen (FLDµm), double cell wall thickness (DCWTµm), rankles ratio (RR), coefficient of cell rigidity (CR), and fiber flexibility (FF). The physical properties tested comprised density, moisture content and shrinkage. Static bending was also carried out for determining modulus of rupture (MOR MPa) and Modulus of Elasticity (MOE MPa).This, in addition to compression parallel to the grain (COM MPa). As well as Modulus of Elasticity from compression (MOC MPa) . All physical and mechanical properties were carried out according to standard procedures. The results of the analysis of variance revealed significant differences in most of the anatomical properties between species. Fiber length ranged between (2.28mm) for kafour and (1.45mm) for haraz. Fiber diameter, on the other hand ranged between (16.7470µm) for sunt to (14.113µm) for hraz. Fiber lumen diameter was highest for haraz (7.832µm) and lowest for talh (6.122µm). Double call wall thickness, on the other hand, ranged between (8.95 µm) for sunt to 6.22 µm for haraz. Rankles ratio, however, was highest for talh (1.834) and lowest for haraz (0.842). For coefficient of cell rigidity talh was highest (0.288), while haraz was the lowest (0.223). Fiber Flexibility ranged between (0.55) for haraz to (0.433) for Kafour. Modulus of Rupture ranged between (139.701) for sunt to (68.736) for haraz. Compression parallel to the grain ranged between (68.409) for sunt to (35.864) for haraz. While modulus of elasticity ranged from (12873.6) sunt to (6507) for haraz. Density ranged from (0.95g/cm3) for sunt to (0.50g/cm3) for haraz. Shrinkage, however, showed no significance variation between the species. The correlation analysis revealed significant correlations between MOR, MOE, COM, MOC and density on one hand and some anatomical characteristics, namely FD, DCWT, rankles ratio, coefficient of fiber rigidity, volume fraction of cell lumen(PPCL) and fraction of cell wall(PPCW). When these correlated variables were used in simple regression with individual physical and mechanical properties as dependent variables and correlated anatomical properties as the independent variables, they revealed that PPCW , PPCL and DCWT was the most important anatomical properties affecting the dependent variables (MOR, MOE, COM,MOC and DEN), followed by FD, RR,CR and FF. The multiple regression used for modelling these relationships showed that the most important independent factors affecting the dependent variables (MOR, MOE, COM, MOC, and DEN) were the same namely: PPCW, PPCL, DCWT, FD, RR, CR. R-square for MOR model was 0.6219 at P<0.0001significant level; for MOE model R-square was 0.6057 at P< 0.0001; while for COM model R-square was 0.6774 and P< 0.0001. Hence R-square for model of MOC was 0.589 at P<0.0001significant level. However the principal factors influencing density were DCWT and CR with R-square = 0.6539 and P<0.0001.