The research materials are Gmelina arborea, Tectona grandis (Teak), Terminalia superba (Afara), Ayin (Anogeissus leiocarpus), and Acacia (Robinia pseudoacacia). These species were chosen because of their availability in abundance in the South West of Nigeria. The materials were obtained at a furniture workshop in Iwo, Osun State (South Western region of Nigeria). The wood samples had been air dried for two months as at the time of collection.

Mechanical tests conducted on the selected wood species includes MOR, MOE, bending strength test, hardness, compressive strength and shear strength tests. The test samples of selected wood species were machined and trimmed to standard sizes specified by corresponding ASTM codes used for the determination of their mechanical properties. All the experiments were done at the Department of Metallurgical and Material Engineering Laboratory of Obafemi Awolowo University, Ile-Ife, Osun State, Nigeria.

Sample preparation and testing for bending strength were in accordance with the specification in [13] . All the samples were seasoned by air-drying because drying increases the physical properties, mechanical properties and the dimensional stability of wood [14] . Wood bending strength frame with digital dial gauge (Universal Testing Machine), operated at a load rate of 0.2 MPa/s, was used to carry out the tests using samples’ dimension of 10 mm × 10 mm × 300 mm. Samples were loaded to failure in three-point loading over a span of 280 mm. The loads at elastic limit (P) were recorded and used for computations of MOR in Nmm⁻² using Equations (1). To obtain stress—strain curves, a dial gauge with precision of 0.001 mm was used to determine sample’s deformation under load. The deformation was measured from the change in distance between the loading plates.

The slope of the linear part of the stress-strain curves was used to determine the MOE of the samples [15] , which is the load carrying capacity of members in bending and it is proportional to maximum moment borne by the wood species [11] . MOE and MOR are important properties of structural timbers and are commonly used in different standards as a basis for classifying timbers [16] . Forty test samples were tested per wood property per timber specie [16] .

M O R = 3 P L 2 b d 2 (1)

where, P is Maximum Load (N), L is Length of sample (mm), which is 280 mm, b is width of the sample (mm), which is 10 mm, d is thickness of the sample (mm), which is 10 mm.

The ruptured surfaces of the tested specimen were examined to ensure that failures were not due to internal hidden defects and tests results from specimens with failure modes due to hidden defects were rejected [16] .

The compressive strength test was conducted in accordance with [13] [17] . The dimension of the samples was measured using a vernier caliper. The samples were placed in the Universal Testing Machine (UTM) in such a way that the grain fibers were parallel to the applied compressive load. Deflection dial gauge was attached to the sample and adjusted to zero. The UTM was operated at a rate of 0.6 mm/min and the sample was gradually loaded in compression. The load was adjusted to its maximum capacity. The deflection reading from the dial gauge for each regular load increment was recorded. The deflection gauge was removed gradually when cracks were observed appearing in the sample. Then the sample was loaded to failure and the maximum crushing load, P_max, was recorded. The compressive strength of the wood sample was then calculated from the maximum crushing load and the contact area as shown in Equation (2).

σ c = P max b d (2)

where b is the width of the sample and d is the depth of the sample.

Shear strength of the timber samples was carried out following the specification in [17] . The size of the sample used for this test is 50 mm × 50 mm × 63 mm notched at one end (as shown in Figure 1) in order for the failure to occur on a 50 mm × 50 mm surface. The samples were held in position by gripping one end in the upper portion of the UTM and the other end gripped in the lower position in such a way that the edges of the samples were vertical and the end rest evenly on the support over the contact area. The UTM was switched on and the load was applied continuously throughout the test at a rate of motion of the movable cross head of 0.6 mm/min until the specimen shears. The load, F_max, at which the samples shear was noted and recorded. The machine was stopped and the specimen was removed. The experiment was repeated for all the samples of the species. The shear stress (τ) was calculated from Equation (3).

τ = F max b L (3)

where b is the width of the test piece, L is the length of the shearing plane and F is highest load.

Tensile test was also done in accordance with the method outlined in [13] [17] . The initial length (Li), initial cross-sectional area (Ai) and the initial diameter

(di) of the sample were measured and recorded. The specimen was then inserted into the grips of the testing machine which has strain-measuring device attached to it. Loads were applied in succession to the test sample and recorded with the corresponding elongation or extension. The readings were taken frequently as the yield point, due to the loads, were approached. The values of the elongations were measured with the aid of divider and a ruler. The test was run till fracture occurs. The final length (L_f) and the final diameter (d_f) of the sample were measured by joining the two broken halves of the sample. The Ultimate load (F) after the sample breaking was also recorded.

Brinell method of hardness test was used to test the hardness of all the species following procedure [17] . A steel ball of diameter 5 mm was inserted in the ball holder of Brinell hardness testing machine that was used to perform the experiment. With the aid of a Jack adjusting wheel, the ball was made to have contact with the test sample, whose surfaces were free from oil, grease, dust and the likes. The samples were loaded one after the other with a force of 30 KN for 15 seconds. The samples were then removed from the machine and their indentations were located, viewed through microscope and their diameters were measured using micro meter fixed on the microscope. The procedure was replicated for all the species’ test samples. The Brinell hardness of each of the samples of the species were estimated from Equation (4).

B H N = Loadapplied ( N ) Specialsurfaceareaofindentation = 2 P π D ( D − ( D 2 − d 2 ) ) (4)

where D is the diameter of the ball, d is diameter of indentation and P is load.

The results of bending strength tests on all the species presented in Figures 2(a)-(e) is the graphical relationship between the flexural stress and flexural strain. The graphs indicated the results of resistance of the species to bending using three-point bending test. Table 1 shows that Anogeissus Leiocarpus (Ayin) has the highest modulus of rupture (averagely, 361.75 N/mm²) and has the highest bending strength (90.44 N/mm² averagely) respectively. This is followed by Acacia (Robinia Pseudoacacia), Gmelina Arborea, and Tectona Grandis (Teak) respectively. Terminalia Superba (Afara) has the least MOR and has the least resistance to bending or flexural load.

However, in terms of the stiffness of the species which is measured by MOE, Gmelina arborea has the highest value (on the average) of MOE (3071.11 N/mm²), followed by Anogeissus leiocarpus (Ayin, 2815.55 N/mm²), Acacia (Robinia pseudoacacia, 2713.72 N/mm²), Terminalia superba (Afara) and Tectona grandis (Teak) respectively. This means that, in this set of species, Anogeissus leiocarpus (Ayin) is most reliable in bending than others while Terminalia superba (Afara) has the least reliability being least reliable does not indicate that it is not useful but it is just that the use has its own limit.

Properties of wood which include; compressive strength, tensile strength, MOE, MOR, moisture content, density, seasoning and shrinkage characteristics, anatomy, chemical characteristics and technology of utilization are among the major factors that determines the quality, suitability and material utilization of wood [7] . The compressive strength properties of Anogeissus leiocarpus (Ayin), Robinia pseudoacacia (Acacia), Gmelina arborea, Tectona grandis (Teak) and Terminalia superba (Afara) were graphically represented in Figures 3(a)-(e) The maximum compressive strengths of each of the species were read from the graph and presented in Figure 4. This represents the maximum compressive stress that each of the species can withstand without breaking and beyond which they will fail to sustain the load causing the compressive stress. The Anogeissus leiocarpus (Ayin) has the maximum resistance to compressive stress followed by Acacia (Robinia pseudoacacia) while Terminalia superba (Afara) has the least compressive strength.

Figures 5(a)-(e) shows the graphs of the responses of each of the species to tensile loads applied to the woods during the laboratory experiments. The maximum resistance that each of the species can offer against stretching due to the applied tensile loads were read from the graphs and were presented in Figure 6. The results show that Acacia (Robinia pseudoacacia) has the highest resistance to tensile stress followed by Anogeissus leiocarpus (Ayin) and Terminalia superba (Afara) has the least tensile stress resistance. Gmelina arborea and Tectona grandis (Teak) have very close values of tensile strength, although that of Gmelina arborea is still insignificantly higher than that of Tectona grandis (Teak).

Figure 7 shows the results of the shear strength of the tested wood species. It indicates that Tectona grandis (Teak) has the highest resistant to shear. Next in the order is Acacia (Robinia pseudoacacia) and Anogeissus leiocarpus (Ayin). The least shear resistant of all the studied species is Terminalia superba (Afara).

Density has a strong influence on the physical characteristics, seasoning and shrinkage rate, defects and mechanical characteristics of wood. The mean densities of the selected species were calculated from the experimental results, recorded and arranged in descending order in Figure 8. The results show that Acacia (Robinia pseudoacacia) is the densest of all the species, followed by Tectona grandis (Teak), Anogeissus leiocarpus (Ayin) and Gmelina arborea respectively. The least dense in the rank is Terminalia superba (Afara).

Table 8 of [18] has timber species’ strength classes based on the values of their mechanical properties; tension parallel to grain, compression parallel to grain, bending parallel to grain, shear parallel to grain, density and MOE. The results of the experiments conducted on the test species were compared with the values in this code to ascertain the classes that these test species may fall into. The highest strength class in the code is class D70 while the lowest is C14. Reference [18] states that the strength classes C14 to C40 are softwood and D30 to D70 are hardwood.

All the tested species cannot be categorized into any of the classes of strength in the code because all the species have very lower values of MOE to the classes in the code, four (4) of the species; Anogeissus leiocarpus (Ayin), Acacia (Robinia pseudoacacia), Gmelina arborea, and Tectona grandis (Teak), have values of their tension parallel to grain, compression parallel to grain and bending parallel to grain far higher than the values of the same parameters for Class D70 timber in the code, hence cannot be classified as any of the classes in the code. Terminalia superba (Afara) values of the parameters does not only lower the value of class D70 but each parameter; tension parallel to grain, compression parallel to grain and bending parallel, falls into three different classes; Class D60, Class D30 and Class D70 respectively.

The result is the same for all the species in terms of shear parallel to grain, density and MOE. There is no clear-cut class of strength for all the species. Although, the experimental values are not higher than the code provisions but for a species, the three parameters can fall into different classes. For example, in terms of shear, Anogeissus leiocarpus (Ayin) falls into class D40 and class D50.