The field work consisted of locating the various study sites and taking samples

for laboratory testing. Samples from the various quarries were taken from large, market-ready piles. Samples were taken in a representative manner from the top, middle and bottom of each pile, then packed in polystyrene bags, labeled and sent to the civil engineering laboratory of FOTSO Victor University Institute of Technology in Bandjoun for certain tests and to the civil engineering laboratory of the University Institute of Coast in Douala.

The water content was determined in accordance with standard NF P 94-050 (1995) . The particle size analysis was carried out by dry method according to NF P 94-056 (1996) . The sand equivalent was determined according to the requirements of standard NF P18-598 (1979) . The absolute density was determined by measuring the solid grains with a pycnometer and by successive weighing according to the recommendations of standard NF P 94-054 (1991) . The void index (e), porosity (η) and saturation (s_r) were determined by calculation using the following relationships:

e = d r − d d d d (1)

n = e 1 + e (2)

s r = w ⋅ d r e ⋅ γ ⋅ w (3)

where: d_r is real density, d_d is dry density, ω is Water content, γ is Volume weight.

In order to know with certainty whether the sand is mostly made up of fine or coarse grains, the modulus of fineness (MF) was determined according to the French standard NF P 18-540 (1997) and NF P 18-545 (2011) . Thus, the modulus of fineness is equal to 1/100^th of the sum of the cumulative refusals, expressed in percentage on the sieves of the following series: 0.16 - 0.315 - 0.63 - 1.25 - 2.5 - 5 mm:

Mf = % R 5 + % R 2.5 + % R 1.25 + % R 0.63 + % R 0.315 % R 0.160 100 (4)

Cumulative rejects in % of sieves (0.16 - 0.315 - 0.63 -1.25 - 2.5 - 5 mm).

The Hazen uniformity coefficient (C_u ) was obtained according to the French standard NF P 18-540 (1997) , by the following formula:

C u = D 60 / D 10 (5)

D₁₀ and D₆₀ are the abscissa values for the points on the grading curve that correspond to 10% and 60% ordinates. From this standard, the following reference is used to determine the uniformity of the curve:

- C_u < 3, then the grain size is uniform or tight;

- C_u > 3, then the grain size is varied or spread out.

The curvature coefficient (C_c ) allows the shape of the particle size curve to be described thanks to the effective diameter of the particles corresponding to 10%, 30% and 60% of passing. The formula for this coefficient is as follows:

C c = ( D 30 ) 2 / D 10 ∗ D 60 (6)

Therefore, for a well graded soil, a wide variety of diameters are present. When 1 < C_c < 3, the grading curve thus indicates the presence of a large variety of diameters. On the other hand, when the coefficient of curvature is too large or too small it indicates the absence of certain diameters between the effective diameters D₁₀ and D₆₀, and the grading is then said to be badly graded.

The values for natural water content, absolute density and sand equivalent of the different materials are given in Table 1.

From Table 1, it can be seen that the natural water contents of the different materials vary from 9.47% to 21.70%, with an average of 15.61%. However, the results of the absolute densities range from 2.42 to 2.55 with an average of 2.51. These absolute densities values are almost similar to those found by Tchouata et al. (2020) on the characterization of sands from Batié and Djione (West Cameroon), for use in civil engineering. The almost similar characteristics of these materials in terms of density could be due to the fact that they both originate from a granitic weathering complex. However, these values are slightly lower than those obtained by Tatiana et al. (2016) on the characterisation of the Santchou-West Cameroon sands, with a mean density value of 2.6. This variation in absolute density from one material to another could be due to a variation in the level of weathering of the granitic arena respectively for each sample point.

NB: LB2: Bamendou High School 2; LBT: Technical High School Bamendou.

Dreux (1985) qualifies concretes according to the degree of cleanliness of the sands used. The sand equivalent of the studied materials ranges from 30.54 to 85.77% with an average of 57.86%. This result is close to that obtained by Ngapgue et al. (2009) in Batié Cameroon (ES = 71.16), indicate that this is a slightly clayey sand with respect to the recommendations indicated by NF P18-598 (1979) standard. However, the value of 57.86% is far from that obtained by Tatiana et al. (2016) in Santchou Cameroon (ES = 87.76) during a study on the characterisation of alluvial sands from Santchou (West Cameroon) for use in civil engineering. In view of the above, this difference is sufficient proof for the difference in cleanliness between alluvial and eluvial sands from the West Cameroon highlands. The low value of sand equivalent for some materials is not surprising. These sands are associated with clays, which are derived from the alteration of feldspars (orthose or plagioclase) and micas in a dry and humid tropical climate. In addition, the method of extraction and washing of these construction materials also contributes to an increase in the fine particle content of these sands, rendering some samples with a sand equivalent of less than 70 unusable for the production of better quality concrete. Overall, Bamendou sand is not recommended for concrete production, as its cleanliness is too low. If used, this sand could contribute to the premature deterioration of structures.

The summary of void index, porosity and other identification parameters such as water content, absolute density and dry density are presented in Table 2.

The particle size composition of the natural and corrected materials is shown as a curve in Figure 2.

From the granulometric analysis of the studied sands, the results of the fineness moduli (FM) of the said materials are obtained. These values vary from 2.91 to 3.92 with an average of 3.31. These values being close to those obtained by Ngapgue et al. (2009) in Cameroon (MF = 3.32) and Tchouata et al., 2020, (MF = 3.10), are furthermore far from those obtained by Bouaziz et al. (2010) in Southern Algeria (MF = 2.8). These results indicate and confirm that these materials do not belong to the prescribed range to which a fineness modulus should belong, for obtaining a quality concrete (2.2 ≤ MF ≤ 2.8). In the Canadian General Specifications and Guidelines (CCDG) classification system, a soil is a sand when it contains less than 50% of the particles passing through the 80 μm sieve and therefore more than 50% of the particles retained on this sieve passing through the 5 mm sieve. A soil is fine sand when it contains more than 50% of particles passing the 315 μm sieve. Coarse sand is soil with less than 50% of particles passing through the 1.25 mm sieve. According to the Canadian General Specifications and Guidelines (CCDG) classification (Table 3), the sands of Bamendou are medium sands, except for the sands of Bamendou Technical High School, Troukwet and Nkounny 2 which are coarse sands.

With regard to the values of the uniformity and curvature coefficients (Table 4), the Bamendou sands are spread out and poorly graded, with the exception of sample S3. These sands are marked by the absence of certain diameters between D10 and D60.