CEBs buildings can be found on all continents. The quality of these buildings depends on the formulation of their materials. Mechanical resistance, swelling, and shrinkage are among the common issues encountered. These problems can

be aggravated by poor material study, characterization, or stabilization. Figure 1 presents some problems of earth construction in particular cracking.

In Cameroon, the MIPROMALO ( Figure 2 ) is responsible for promoting and disseminating the use of all local construction materials.

This research organization’s focus guided the selection of materials studied in this work to produce BTC, including simple earth, earth and cement, and earth and sawdust.

Samples of earth were collected at depths of −30 cm and −50 cm to avoid the organic part of the soil. For our study, we used earth from the city of Bangangté, specifically from the entrance of the city in an area of production of raw adobe bricks (see Figure 3 ) and with the following geodetic coordinates: X = 518382.046, Y = 1050043.553, Z = 1401.82.

Cameroon is the third most significant African country in terms of the importance of its forest resources, with an area of over 55% covered by forest [18] .

The forestry operation has a local processing capacity of 2.7 million m³ of logs. The initial processing of wood before export is divided into sawing, peeling and slicing activities. All of these activities result in the production of sawdust, which is estimated to be 400,000 m³. This production capacity is distributed across nearly 61 sawmills, 20% of which are located in the coastal region of Douala [18] .

Regarding the fate of the sawdust, a small quantity is utilised in sawmills situated in peri-urban areas, operated by local residents, to power domestic ovens. The majority of the sawdust was incinerated in the vicinity of the sawmills. For sawmills installed in rural areas where firewood is not scarce, the vast majority of sawdust production is burned or dumped in pits set up near the sawmills. In light of the ongoing increase in demand and the importation of pellets from countries such as South Africa onto the European market, it is possible to develop the pellet sector in the short and medium term. This can be achieved by exploring alternative avenues for pellet utilisation in the construction materials industry [18] .

Padouk

Figure 4 illustrates the type of Padauk wood found in the forest and its subsequent transformation into sawdust. The Padouk used in this study to produce wood shavings, scientifically known as Pterocarpus soyauxii, belongs to the Papilionaceae family (Leguminosae, Papilionoideae, Fabaceae) [19] . It is a highly durable wood that does not require preservation treatment, making it an environmentally friendly material of choice. It can be found in the equatorial forest of Nigeria to the Central African Republic, from the Democratic Republic of Congo to Angola. The sawdust used should be smaller than 1.6 mm in size and have a density of approximately 0.85 g/cm³.

The first attempts at stabilising roads with cement were made in the United States in 1915. The use of cement in building construction was independently developed in Germany from 1920 onwards. In the United States, from 1935 onwards, the use of ground-cement increased significantly, with applications in road construction and the development of airfield runways. Since that time, cement stabilization has been employed in countless applications worldwide, both in public works and construction [1] .

Cement is undoubtedly one of the most effective stabilisers for compressed earth blocks. The addition of cement before compression allows for the improvement of material characteristics, particularly in terms of water resistance. This is achieved through the irreversible bonds created between the largest particles. The cement will primarily affect the sand and gravel, acting in a manner similar to that of concrete or a sand-cement mortar. It is therefore inadvisable to use soils that are too clayey (>30%) [1] . In general, a minimum of 5% - 6% cement is required to achieve satisfactory results. The compressive strength is highly dependent on the dosage. An economically acceptable upper limit of 8% cement is often recommended [20] .

The cement used is grey Portland cement with characteristics prescribed by the NF EN 197-1 standard for hydraulic binders. It is equivalent to CEM II-42.5 type cement. Its compressive strength is 42.5 MPa and its density is 1.44 g/cm³.

This cement ( Figure 5 ) contains clinker (70 - 79%), pozzolan (21% - 25%) and gypsum (≤ 5%). Its fineness is between 3900 and 4000 cm2/g. The mineral ( Table 1 ) phases of CEM II 42.5 are: Alite, Belite, quartz, calcite, pyroxene and plagioclase. Pyroxene and plagioclase are residual crystalline phases of volcanic ash used as additives (Djobo et al. 2017; Jean Noël et al. 2020; Elie Kamseu et al. 2016; Leonelli et al. 2007). Table 1 presents the main mineral phases found in the cement used.

Figure 6 represents the principle of the jar test. The sedimentation or jar test estimates the percentages of sand, clay, silt, and humus based on the thickness of each layer formed in the jar after settling. The soil texture can then be approximately determined using the soil texture triangle. Soil texture refers to the relative proportion of sand, silt, and clay particles in soil. Soil texture is often described by differentiating between heavy soils, which are rich in clay, and light soils, which are sandy or loamy. It is a major determinant of soil workability, water retention and drainage, as well as the level of organic matter and aggregation. The material’s resistant elements include gravel and sand. Clays provide cohesion and control plasticity, as well as influencing the earth’s reaction to humidity, which in turn affects shrinkage and swelling (Walker, 1995).

Some steps in determining the Atterberg limit are shown in Figure 7 . Atterberg limits are conventional geotechnical characteristics of soil that mark the thresholds between the transition of soil from liquid state to plastic state (liquid limit, Wl) and the transition of soil from plastic state to solid state (plastic limit, Wp). These limits are expressed as the soil’s water content at the considered transition state, expressed as a percentage of the mass of the raw material. The plasticity index is a measure of the quantity and type of soil present in a sample. It indicates the clay content of the sample [21] . These are used in specifications to control the compressibility properties and behaviour of soil mixtures. It is calculated using the equation (1) below:

$Ip=Wp-Wl$ (1)

where Ip is the plasticity index, $Wp$ is the plastic limit, and $Wl$ is the liquid limit.

The determination of the Atterberg limits was carried out according to the NF P 94-051 standard [18] [19] . Knowing these limitations allows for quick prediction of a soil’s construction possibilities.

The NBN EN 1097-5 standard describes a method for determining the total moisture content per unit of dry matter of an aggregate sample. Moisture can be found on the surface, inside the grains, or in pores accessible to water. The sample is weighed before and after drying at 110˚C. The percentage of water content is obtained by calculating the difference in mass between the wet and dry masses, expressed as a percentage of the dry mass [20] .

To determine the wet mass, the sample is spread out on a tray and dried in a ventilated oven at 110˚C ± 5˚C until the mass of the contents changes by less than 0.1% between two successive weighings spaced at least 1 hour apart, at which point it is considered to have reached a constant mass. For fine aggregates, it is possible to mix the sample with a spatula during the drying process to facilitate water evaporation. However, the spatula must always be left in the container to avoid losing particles.

The water content (W) is calculated using the formula

W = Mass of water/Mass of soil (2),

where Mass of water = Mass of soil − Mass of soil after drying.

The objective of the Proctor test is to ascertain the water content to which soil must be compacted in order to achieve maximum dry density. A 15 kg sample, sieved using 5 mm aperture sieves, was taken and divided into five piles of 3 kg. The materials used in the Proctor test were as follows: the Proctor mould, a cylindrical metal tube with an internal diameter of 10.15 cm and a height of 11.7 cm; the standard lady, a cylindrical sheep 5.1 cm in diameter guided by a rod inside a sheath over a certain centimetre in length. The modified Proctor test requires a sheep weight of 2.49 kg and a drop height of 30.5 cm. It also calls for trowels and spatulas for mixing, a scale with a sensitivity of at least one gram and a maximum capacity of at least 20 kg, and a precision scale with an accuracy of 0.01 g. Additionally, the test requires tars and an oven with a temperature range of 105˚C. The Proctor test begins with sample preparation on the mixing table. The sample is first watered with approximately the requisite quantity of water for the first compaction stage. It is then mixed to achieve uniform humidification of the mass. The compaction process is carried out in five stages, with each layer approximately 2.5 cm thick receiving 55 blows. Once compaction is complete, the upper part of the mould is removed to weigh the specimen and obtain the precise weight of the sample, with an accuracy of one gram. This process is repeated on five identical samples, each having received a different quantity of water.

Test specimens are made using a standardized prismatic mould ( Figure 8(a) ) with dimensions of 16 × 4 × 4 cm³ (NF P15-413 Standard). Compaction is carried out using a hydraulic press ( Figure 8(b) ) at a pressure of 10 MPa.

The blocks were manufactured at the Laboratory of Civil Engineering Technical Education of the University of Douala, following the Cameroonian standard NC 103. The mass of the dry specimens was kept constant during the experimental phase. We have formulations ( Table 2 ) with and without stabilizer, using cement and Padouk sawdust stabilizers at rates of 4%, 6%, and 8% relative to the total dry mass of the test specimens.

The compression resistance of samples is determined by dividing the maximum load at axial compression failure by the cross-sectional area of the sample perpendicular to the direction of the compressive force. Compression tests aim to determine the behavior and response of a material under a compressive load by measuring fundamental variables such as stress and strain.

Compression tests were conducted on 16 × 4 × 4 cm³ specimens. The compression strength is calculated using formula (3).

$R_c=\frac{F}{Lxl}$ (3).

With: R_c: Compressive strength of blocks in (MPa), F: Breaking force (N), L: Length of the specimen in cm, 1: Width of the specimen in cm.

The blocks used have dimensions of 4 × 4 × 4 cm³. The dry bulk density of the brick is determined using the formula below (expressed in kg/m³):

$\rho =\frac{M}{V}$ (5)

M: the mass in kilograms and V: represents the volume in cubic meters.

Dry the samples in an oven at a temperature of 105˚C ± 5˚C until a constant mass or dry mass (M_s) is obtained. After cooling, immerse them completely in water. Mass measurements were taken every 30 minutes after removing residual water with an absorbent cloth until the difference between two successive measurements was less than 0.2% or the saturated mass (M_sa) was reached [21] . The pore volume (V_p) is calculated as the difference between the final saturated mass and the initial dry mass, divided by the density of water (ρ_eau). The apparent porosity (P) was determined by dividing the pore volume by the total volume (V_t) of the specimen [21] .

$V_p=\frac{M_{sa-M_s}}{{\rho }_{eau}}$ (6)

$\rho =\frac{V_p}{V_t}$ (7)

The water absorption test determines the amount of water absorbed by capillarity. The test involves measuring the increase in mass of the specimen placed in a container with a constant water level. Prior to the absorption test, the specimens were dried in an oven at 105˚C for 24 hours. The mass of the completely dry specimen (M_s) was then recorded. The test tube was finally immersed in water for 24 hours and its soaked mass (M_t) was measured [22] . The absorption percentage is calculated using the following formula:

$A_{bs}=\frac{M_t-M_s}{M_s}\times 100$ (8)

We collected three soil samples from the study area, spaced an average of 30 metres apart, and passed them through a 5 mm sieve. The pH of a material has an impact on its properties [23] [24] . If the pH is below 6, soil pre-treatment is necessary. Table 3 shows the results of the jar test.

The texture triangle in Figure 9 helped us determine the type of soil under study. Based on our three samples, we conclude that the soil is sandy loam. Ideally, the soil should contain between 0% and 40% gravel, 25% and 80% sand, 10% and 25% silt, and between 8% and 30% clay [25] . Our analysis confirms that the soil is suitable for making earth blocks.

Atterberg limits are expressed in terms of liquidity limits (Wl), plasticity limits (Wp) and plasticity index (Ip). Their determination is necessary, because it represents a very important index for the classification of soils and also allows a better understanding of their behavior.

1) The Atterberg limits vary according to the cement content

Figure 10 illustrates the variation in the Atterberg limits as a function of cement content. The results demonstrate that the addition of cement to the base soil leads to a reduction in the plasticity index (from 15.9% to 11.28%), an increase in the liquidity limit (from 44.6% to 49.5%) and the plasticity limit (28.7% to 38.22%). This variation can be attributed to the hydration of the cement.

2) The Atterberg limits vary according to the sawdust content.

Table 4 presents the results of the Atterberg limits when we use sawdust as a stabilizer.

The test enables the optimal compaction characteristics of a material to be determined. It is important to note that this test determines the water content (W) that gives the maximum dry density of a material for a given compaction energy.

1) Base soil and earth/Cement

Figure 11 illustrates the evolution of density as a function of water content. Figure 11 Proctor demonstrates a curve that allows for the determination of the optimal water content corresponding to the maximum dry density of the earth. Furthermore, the optimal water content of cement-stabilised earth is presented. The results demonstrate that the introduction of cement into the base soil results in a gradual decline in the maximum dry density, with a maximum compaction density of 19.7 KN/m³ reducing to 18.35 KN/m³. As the sawdust content increases from 0% to 8%, however, a slight increase in water content from 12% to 13.5% is observed when dry density drops slightly.

2) Base soil and earth/Sawdust

Figure 12 shows the variation in dry density as a function of water content. This is done with sawdust as a stabilizer. The results demonstrate that the introduction of sawdust into the base soil results in a gradual decline in the maximum dry density, reaching a maximum compaction density of 19.7 KN/m³, which then reduces to 17.8 KN/m³. As the sawdust content increases from 0% to 8%, however, a slight decrease in dry density is accompanied by an increase in water content from 12% to 13.5%. This reduction is due to the density of padauk, the original species of our sawdust, which is lower than that of the base soil.

Solid density is an important property of construction materials used to control quality from raw powders to final products. Geometric and skeletal densities allow researchers and producers to calculate the open porosity of a solid and hardened sample, which provides critical properties such as strength, temperature tolerance, and transformability. Figure 13 shows the evolution of density for different samples.

The addition of stabilizers to CEBs results in a decrease in density. At 0% sawdust, the density is 1.58 g/cm³, while at 8% sawdust, it is 1.41 g/cm³. As the percentage of sawdust increases, the density of BTC decreases further. As the percentage of sawdust increases, the density of CEBs decreases further. These values are lower than those of terracotta (1.8 - 2.0 g/cm³) and rammed earth (1.7 g/cm³) [26] . The reason for this phenomenon is the gradual decrease in the volume of the earth in the prepared BTC. Its density is estimated to be 1.64 g/cm³, which is higher than that of cement (1.44 g/cm³) and sawdust (0.85 g/cm³).

CEBs with stabilizers, such as cement or sawdust, have lower porosity compared to those without stabilizers (25% porosity).

The reduction in porosity can be attributed to the smaller size of cement particles, which can infiltrate the pores more easily. For instance, CEBs with 8% cement have a porosity of 18%.

The porosity of CEBs stabilized with sawdust is reduced due to the smaller diameter of the sawdust particles passing through a 1.6 mm sieve, which facilitates infiltration between the voids. Compression also contributes to the reduction of voids. When using sawdust as a stabilizer, the porosity decreases from 25% to a minimum of 20% with 8% sawdust.

Figure 15 shows the curve representing the variation in absorption of the different formulations.

The absorption capacity generally increases by 17% to 20% when sawdust residue is added. This is due to the absorption capacity of agricultural residues. [27] demonstrated that the absorption capacity increases with the addition of fibers, depending on their length.

However, when cement is added, the absorption rate decreases. The total absorption value ranges from 17% for blocks without cement stabilizer to 12% for blocks containing 8% cement. This can be attributed to cement’s ability to fill voids, reduce porosity between soil particles, and decrease infiltration. According to [28] , the addition of cement also reduces the absorption capacity of earth blocks.

The resistance of CEBs in compression is shown in Figure 16 .

Figure 17 shows the variation in stress as a function of displacement during the compression test. The compressive strength increases as stabilizers are added. Specifically, the strength increases from 5.1 MPa with 0% stabilizer to 12.8 MPa with 8% cement, and from 5.2 MPa with 0% sawdust to 11.5 MPa with 8% sawdust.

The increase in strength for cement is due to infiltration and void reduction caused by the stabilizer and the compressive force. The results confirm previous findings by [29] [30] that adding cement increases compressive strength.

Additionally, introducing fibers to sawdust improves its compressive strength. Other studies have shown that stabilizing with plant fibers also increases compressive strength of CEBs, as reported by [31] [32] for low sawdust rates.

The curve of the evolution of the stresses as a function of the displacement also demonstrates that the compressive resistance increases with the stabiliser rate.