Clayey soils are widely distributed in various regions around the world, characterized by their fine composition [7] and high plasticity [8] . While these soils offer an abundance of cost-effective construction materials [9] their intrinsic mechanical properties can be limiting [6] . Indeed, unstabilized clay often exhibits low compressive strength [10] and bearing capacity, which can hinder their use in construction projects requiring strong and durable foundations.

The integration of plant fibers in soil stabilization has emerged as an innovative method addressed by several authors [11] - [16] to enhance the mechanical characteristics of clayey soils. Plant fibers such as jute fiber [17] [18] , palm fiber [19] [20] , sisal fiber [21] [22] , bamboo fiber [23] , flax fiber, and kapok fiber, due to their fibrous nature and strength, can act as natural reinforcements, strengthening the soil matrix and thereby improving its mechanical properties. This approach has the additional advantage of being cost-effective and environmentally friendly [24] [25] .

Several previous studies [26] [27] [28] have already explored the use of coconut fiber as a stabilizing agent for soils. In addition to its abundant availability - the world produces at least 30 million tons of coconuts [29] , coconut fiber possesses specific advantages such as its tensile strength [26] [27] [30] and natural durability. These studies have shown promising results, suggesting that the incorporation of coconut fiber can significantly enhance the mechanical properties of clayey soils. However, there are specific aspects that need further clarification and in-depth exploration for a broader and more effective application of this method.

Previous studies demonstrate a high potential for this approach, underscoring the importance of continuing research in this field.

These involve determining the Atterberg limits, water content, and dry density using the Proctor Test. The Atterberg limits are conventional geotechnical characteristics of soil that delineate the thresholds between:

This determination of Atterberg limits was carried out in accordance with the NF P 94-051 standard [31] and is illustrated in Figure 2 . This allows us to deduce the plasticity index $I_p$:

$I_p=L_l-L_p$ (1)

and the compactness index I_c:

$I_c=\frac{\left(L_l-W_e\right)}{I_p}$ (2)

where W_e is the water content of the clay, previously determined using the Proctor test.

The Proctor Test itself was conducted in accordance with the NF P94-093 standard [32] .

Adobe bricks of each composition in Table 1 were manually crafted, as shown in Figure 3 , using a metal mold with dimensions: 4 × 4 × 16 cm³. They were air-dried in natural convection for 12 days. The average masses of the different compounds were recorded every day during the drying process to monitor their progress. The results of these measurements are documented in Table 2 . It can be observed that the bricks were already dry between the 8th and 9th day as their masses remained constant from these days onward.

The mechanical properties, including flexural strength and compressive strength, were determined at the Eco Materials Laboratory of the International Institute for Water and Environmental Engineering (2ie) using a Controlab hydraulic press, as shown in Figure 4 , with a maximum applied load capacity of 300 kN. The breaking force is the maximum force that led to the sample's fracture or breakage. From this maximum force, the rupture pressure is deduced, referred to as either compressive strength or flexural strength, depending on the case. For this study, three-point bending was employed, as shown in Figure 4(a) . The two roller supports, each with a diameter of 10 mm, are spaced at a distance of L = 106.7 mm. The pressure force is applied through the third roller at the midpoint between them.

${\sigma }_f=\frac{3FL}{2b^3}$ (3)

where b (mm) is the edge of the square section of the prism, and F (N) is the breaking force.

As for compressive strength, it is determined using the relation:

${\sigma }_c=\frac{F}{S}$ (4)

where ${\sigma }_c$ is the compressive strength of the specimen in MPa, F is the maximum load sustained by the specimen in N, and S is the average value of the cross-sectional area in mm².

The force F is applied to the sample until it is crushed Figure 4(b) .

The press is connected to a computer, allowing real-time monitoring of the pressure force applied to the brick Figure 4(c) .

For the pure clay, we obtained an optimal water content (W_e) de 22.3% at a dry density γ_b = 1.42 g/cm³. For different proportions of coconut fiber (0.25%; 0.50% and 0.75%), we obtained the pairs: (W_e; γ_b) as follows (W_e = 28.2%; γ_b = 1.47 g/cm³); (W_e = 24.9%; γ_b = 1.48 g/cm³); (W_e = 24.2%; γ_b = 1.53 g/cm³). It is observed that the water content decreases with the fiber content, while the dry density increases. The decrease in W_e indicates that after mixing, the fibers absorb a portion of the added water, explaining the regression of water content with increasing fiber content in the mixtures. The increase in dry density reflects high compaction of the clay-fiber composite, indicating strong internal cohesion after compaction.

Liquid limit L_l = 63.6, plastic limit L_p = 27.9, plasticity index I_p = 35.7, consistency index I_c = 1.456. Based on this, the clay is classified as follows: 25 < I_p = 35.7 < 40, making it a marly clay of class A3. According to the Road Earthwork Guide (GTR) [33] , if this clay is to be used in an earthwork project, lime treatment is required for soils of types A3 and A4, which means soils with a significant liquid limit and a plasticity index greater than 20. Additionally, I_c= 1.456 >1.3, according to the hydric classification, placing it in class A3ts.

With a plasticity of I_p = 35.7 > I_p = 14% recommended by [34] , it is suitable for making compressed earth blocks (CEBs).

The evolution of the pressure force applied to the samples of different compositions is shown in Figure 5 for the pure clay (BA) and in Figure 6 for the clay-fiber compositions. The terms E₁, E₂ and E₃ respectively denote sample number 1, number 2, and number 3. For BA, the breaking forces are 136.21 N and 200.46 N for E₁ and E₂ respectively. In contrast, samples composed of clays stabilized with fiber proportions of 0.25%, 0.50%, 0.75%, and 1% exhibit greater breaking forces. The stabilized bricks with 0.25% fibers have a breaking force ranging from 240 N to 250 N, and those stabilized with 0.5% fibers between 440 N and 460 N. The Clay-Fiber Bricks of 0.75% have a breaking force of 500 N; Stabilizations with 1% fibers have a range of 960N to 1100N. In some cases, the test did not lead to a clear rupture. These cases are considered aberrations and are not taken into account. Thus, using Equation 3, we deduce flexural strengths of 0.63 MPa for pure clay (BA); 0.89 MPa for clay + 0.25% fiber (BAF1/4); 1.68 MPa for clay + 0.5% fiber (BAF1/2); 1.87 MPa for clay + 0.75% fiber (BAF3/4); 3.91 MPa for clay + 1% fiber (BAF1), which is slightly better than that found by [35] who used lime-fiber mixture as stabilizers. Flexural strength increases with fiber content, indicating that coconut fibers enhance the flexibility of the bricks by reinforcing the bonds between the solid clay particles.

The behaviors of the bricks under the effect of compressive load are illustrated in Figure 7 for BA and Figure 8 for the different clay-fiber compositions at 0.25%, 0.5%, 0.75%, and 1%. By applying Equation (4), we obtain the values of the compressive stress σ_c (MPa). Thus, we have: compression strength for BA = 5.90 MPa, BAF1/4 = 6.395 MPa, BAF1/2 = 6.292 MPa, BAF3/4 = 6.065 MPa, and BAF1 = 5.423 MPa. It is noted that the compressive strength σ_c increases with the increase in fiber content up to 0.5% of it. Beyond that, at 0.75% and 1% fiber content, the compressive stress σ_c decreases. This means that when the fiber content is high, it makes the compound more ductile, leading to a decrease in σ_c. Therefore, it can be deduced that there is an optimal amount of fiber for good compressive strength. On the other hand, the presence of fiber results in a reduction in crack propagation under tension after initial deformation, a decrease in the number of cracks caused by shrinkage, and a decrease in the hydraulic conductivity of compacted clayey soils [36] [37] .