All raw materials were sourced from Senegalese localities to reflect realistic construction practice. Dune sand was obtained from Bambilor (Dakar region), kaolin from Thiès (Thicky site), limestone (calcaire 0/3) from Rufisque, and laterite from Thiès. Typha australis stems were harvested from the Wakhinane Nimzat wetland area in Dakar. Water was supplied by the Société Sénégalaise des Eaux (SDE).

After harvest, Typha stems were sun-dried for at least seven days to ensure complete moisture evaporation, then mechanically chopped into short fibers of 5 - 20 mm length using a mechanical cutter; fibers were subsequently screened through a 20 mm sieve to remove oversized fragments and ensure uniform dispersion within the matrix. Laterite, kaolin, and limestone were dry-sieved through a 2 mm mesh to ensure a homogeneous texture prior to mixing.

Composite mixtures were designed according to a total dry mass of 2500 g per Proctor mould. The reference (control) mixture consisted of: laterite 1100 g, kaolin 500 g, limestone 500 g, and dune sand 400 g. This matrix composition was selected on the basis of preliminary trial mixes that balanced plasticity, workability, and sufficient cohesion for Proctor compaction. Typha fibers were introduced by substituting an equivalent mass of laterite, at dosages of 1%, 2%, 3%, 4%, 5%, and 7% of the total dry mass, corresponding to 25, 50, 75, 100, 125, and 175 g of fiber, respectively. The water content of all mixtures was fixed at 12% (300 mL), corresponding to the optimum moisture content (OMC) determined by the Standard Proctor compaction test (NF P 94-093) performed on the reference matrix, which yielded a maximum dry density of approximately 1.85 g/cm³. This single water content was applied to all dosage levels to isolate the effect of fiber content on strength, acknowledging that the OMC may shift slightly with increasing fiber addition.

The fabrication protocol followed four stages: 1) dry-mixing of laterite, limestone, Typha, and dune sand; 2) preparation of a kaolin slurry by mixing kaolin with the measured water; 3) combining the dry mix and the kaolin slurry; and 4) compaction using the Standard Proctor procedure (NF P 94-093) in cylindrical moulds (diameter: 100 mm, height: 110 mm). Three replicate specimens were prepared for each dosage level.

Compressive strength tests were performed 7 days after specimen fabrication using a concrete compression press. The uniaxial compressive strength (σc) was calculated as:

where F is the peak applied force (N) and A is the cross-sectional area of the specimen (mm²). For each dosage, the mean compressive strength was calculated from the three replicates. Results were compared against the minimum load-bearing threshold of 150 daN (≈1.91 MPa) specified by NF P14-301 for structural hollow blocks.

Thermal conductivity was determined by the heat-flow box method using prismatic specimens (270 mm × 270 mm × 50 mm) composed of 100% chopped Typha australis at varying bulk densities. For fiber-earth composites, thermal conductivity was assessed on cement-sand-Typha specimens (cement: 15.8%, water: 8.05%, sand: 76.14%) with Typha additions of 0.5% to 3% [8]. These reference values informed the thermal interpretation of the earth-Typha results.

3.1.1. Particle Size Distribution

Sieve analysis revealed distinct granular profiles for each material. The dune sand exhibited a stepped grading curve characteristic of medium-grained sands deficient in intermediate particle sizes, with 91.5% of particles passing the 0.1 mm sieve. The laterite showed well-graded behavior with particles distributed across the full sieve range (0.08 - 16 mm), including a coarse fraction (19.6% retained on the 16 mm sieve), indicating good granular interlocking potential. The kaolin was essentially composed of fine particles (<0.315 mm), with 99.5% passing the 0.08 mm sieve, reflecting its phyllosilicate mineralogy. The limestone (calcaire 0/3) presented a broadly distributed granulometry from 0 to 10 mm, with 61.4% passing the 0.08 mm sieve, confirming its role as a gap-filling aggregate.

3.1.2. Atterberg Limits and Plasticity

Atterberg limit tests were conducted on laterite, kaolin, and limestone. The laterite displayed a liquid limit (WL) of 44.20% and a plasticity index (IP) of 23.60, classifying it as a low-to-medium plasticity material (IP < 50), suitable for earth construction without excessive shrinkage risk. The kaolin exhibited markedly higher plasticity (WL = 90.35%, IP = 57.61), classifying it as a highly plastic and expansive clay. This property justifies its incorporation in small proportions to provide cohesion while being tempered by the other components. The limestone recorded WL = 49.94% and IP = 22.07, indicating low-plasticity behavior adequate for aggregate use; its moderate plasticity index reflects the fine fraction present in the 0/3 grading and does not preclude its role as a gap-filling aggregate in the composite matrix.

3.1.3. Physical Properties

Table 1 summarizes the key physical properties of all constituent materials. The specific gravity values of sand (2.661 g/cm³), laterite (2.50 g/cm³), kaolin (2.485 g/cm³), and limestone (2.59 g/cm³) are typical of silicate-based minerals. The Typha fiber bulk density was extremely low (0.044 g/cm³), consistent with its highly porous lignocellulosic structure, which is the primary driver of its insulating capacity. The sand equivalent of 96.44% > 80% confirms the cleanliness and absence of clay fines in the dune sand.

Table 1. Physical properties of constituent materials.

*Absolute density measured by helium pycnometry [7].

Figure 1 presents the mean uniaxial compressive strength (UCS) values obtained for all dosage levels. All formulations comfortably exceeded the 1.91 MPa threshold corresponding to the normative 150 daN load-bearing requirement (NF P14-301), demonstrating the structural viability of the earth-Typha composite for construction applications.

The strength evolution followed a non-monotonic pattern with a clear optimum at 4% Typha content. This trend can be interpreted through the fiber matrix interaction mechanism: at low dosages (1% - 4%), the finely chopped Typha fibers penetrate the micropores of the earth matrix, increasing compaction density and improving particle interlocking, which progressively enhances load transfer capacity [9]. The maximum UCS of 2.86 MPa at 4% represents a 33% improvement over the control mixture (2.15 MPa), a substantial reinforcement effect attributable to the bridging action of fibers across potential crack planes.

Beyond 4%, the excess fiber volume disrupts matrix continuity: the low-density Typha material (0.044 g/cm³) introduces a high pore volume that counteracts compaction, reduces inter-particle contact, and creates preferential failure planes. This behavior is consistent with observations reported for other plant-fiber-earth composites, including hemp lime [10] and sisal-clay systems, where an optimal fiber content exists, beyond which strength decreases while porosity and hence thermal resistance increase [3].

Figure 1.Mean uniaxial compressive strength of earth-Typha composites as a function of fiber content.

The mean thermal conductivity of 100% Typha australis specimens was 0.045 W/m·K at a mean bulk density of approximately 69.7 kg/m³, confirming its classification as a thermal insulator (λ < 0.065 W/m·K). This value is comparable to that of hemp-lime composites (0.06 - 0.12 W/m·K) [11][12] and loose mineral wool (0.035 - 0.045 W/m·K), positioning Typha as a competitive alternative insulating material [8].

In the earth Typha composites, the thermal behavior exhibits an inverse relationship with mechanical strength: while compressive strength peaks at 4% and declines thereafter, thermal resistance continues to improve as Typha content increases from 4% to 7%. This trade-off reflects the dual nature of Typha: its fibrous, porous structure simultaneously introduces crack-bridging reinforcement (beneficial at low dosages) and thermal barrier properties (dominant at higher dosages). Consequently, the optimal dosage depends on the intended application: 4% Typha is recommended where structural load-bearing capacity is the primary criterion, while dosages of 5% - 7% are more appropriate for non-load-bearing thermal insulation applications such as partition walls, ceiling panels, or external cladding.

It is important to note that in this study, the thermal conductivity of the earth-Typha composites was not measured directly on the Proctor-compacted specimens, but was inferred from the literature data on Typha-cement-sand composites and the known trend of conductivity decrease with increasing Typha content [8]. Direct measurement of λ on earth-Typha composites using a heat-flow meter or hot-disk apparatus is recommended in future work to quantify thermal performance precisely across the full dosage range.

The results of this study have significant practical implications for the construction sector in Senegal and West Africa more broadly. Laterite, kaolin, limestone, and dune sand are all locally and abundantly available raw materials, while Typha australis constitutes an invasive biomass whose management represents a socio-environmental challenge. The valorization of Typha into composite building elements therefore contributes simultaneously to: 1) reducing construction costs by replacing imported materials; 2) improving the energy performance of buildings through enhanced thermal mass and insulation; 3) reducing the carbon footprint of construction by substituting cement-based products with earth composites that sequester biogenic carbon [5]; and 4) creating local economic value chains around Typha harvesting and processing.

The compressive strength values obtained (2.15 - 2.86 MPa) are consistent with, and in several cases superior to, values reported for stabilized compressed earth blocks (CEB) in the literature, which typically range from 1.0 to 3.0 MPa [13]. The compliance of all formulations with the NF P14-301 threshold is promising and suggests the preliminary suitability of earth-Typha composites as structural masonry elements in low-rise buildings under the reported laboratory conditions. However, block-scale validation, long-term durability testing (moisture resistance, freeze-thaw cycling, abrasion), and in-situ performance monitoring are required before definitive structural recommendations can be made. Beyond mechanical compliance, comparative life-cycle assessments of façade systems and thermal insulation materials across different climatic conditions further support the environmental relevance of adopting such low-impact, locally sourced solutions [14].