Physico-mechanical and durability characterization of earthen plaster stabilized with fermented rice husk for coating adobe walls

Abstract The present study aims to study the effect of fermented rice husk (RH) on earth plaster for coating and improving the durability of adobe walls. The mixtures of clayey earthen materials and 0%, 33%, 50%, and 67% of fermented RH of the total volume of the mixture are studied to assess their physicomechanical and durability properties. The RH is previously fermented separately in water for the periods of 2, 3, and 3 weeks. The pastes obtained after mixing the clayey earth material and fermented RH were used to mold 4 × 4 × 16 cm3 test specimens. The specimens were dried under ambient laboratory conditions and 35 HR) for 21 days before their characterization. The clay earthen material mainly contains silt (51%), clay (24%), sand (23%), and gravel (3%). It has a liquidity limit, plasticity index, and methylene blue value of 32%, 20%, and 2.66, respectively. The results show that apparent density, linear shrinkage, resistance to compression, and the thermal conductivity of the plaster specimens, respectively, decrease from 1.88 to 1.07 g/cm3, 4.52 to 0.83%, 3.88 to 0.82 MPa, and 0.87 to 0.05 W/m.K with increasing volumetric content of RH from 0% to 67%. Moreover, the resistance to abrasion increased. The resistance to capillary water absorption and water erosion was deficient for the content of RH above 50%. The mixture containing up to 33% of RH allows to improve engineering performances and reaches acceptable durability and would therefore be useful for coating adobe walls.


Introduction
In Burkina Faso, banco "adobe walls" remain the most-used material for the construction of habitats, according to the report of the General Census of Population and Housing (NISD National Institute of Statistics and Demography, 2022).The current development and the ecological awareness make sustainable and ecological constructions attract more attention.Thus, renewable and environmentally friendly materials are in high demand (Liu et al., 2017).In recent years, there has been a particular resurgence in earthen building materials.The main advantages of earthen materials are related to their significant reduction of environmental impact, local availability, low demand of energy throughout their life cycle, and recyclability (Losini et al., 2021).Clay earthen materials are abundant local resources, which are cost-effective and easy to work with, and with very low embodied energy (Labat et al., 2016).According to Aymerich et al. (2012), earth-based materials, such as adobes, can also provide a very high level of thermal comfort and acoustic satisfaction mainly due to their high moisture absorption/desorption and thermal and sound inertia.
However, adobes have shortcomings in terms of mechanical strength and durability, especially regarding water resistance.These issues are linked to the poor cohesion of the clay matrix and the appearance of large amounts of pores during the production of adobes, among others.It is therefore necessary to protect the adobe walls with quality coatings.This remains a major concern for low-income populations.
The commonly used cement or lime-based plasters are not suitable for coating these earthen walls and are relatively expensive.These coatings negatively induce the hygroscopic capacity of earthen walls and have negative consequences on the environment mainly greenhouse gas emissions (CO 2 ) during their life cycle.
Raw earthen plasters are also used for coating earthen walls but mostly have poor weathering resistance and require maintenance after each rainy season.A traditional practice of improvement consists in adding, to the earthen materials, plant or animal fibers such as rice husk, straw, kenaf fibers, fonio, cow dung, and horse manure, among others (Bamogo et al., 2020(Bamogo et al., , 2022;;Ouedraogo et al., 2017;Sunusi et al., 2020).Indeed, the fibers allow to reduce the crack shrinkage, which is particularly important for raw earth plasters, increase the resistance to compression, improve the thermal properties, and most often reduce the water sensitivity of the coatings.
Additionally, the rice production sector has been in the fourth position for cereal production in Burkina Faso, with an average production of 376,527 tons between 2019 and 2020 (Sawadogo et al., 2021).However, the use of rice husk in construction is still scientifically unknown to the population.Its application in construction will create added value for this sector.
Scientific research has primarily addressed the issues of earthen plasters by stabilization using agricultural by-products to improve their physical and mechanical properties and increase their resistance to weathering conditions.Bamogo et al. (Bamogo et al., 2020) conducted a study to evaluate the physical, mechanical, hydric, and thermal properties of earthen plasters amended by fermented cow dung at mass contents of 2 to 6%.The compressive and flexural strength of the plaster were increased by adding 6% cow dung to the total mass of soil.The water sensitivity and thermal conductivity were also improved.They justified this improvement by the formation of the amine silicate (a highly adhesive and insoluble molecule) that ensures matrix cohesion.The author also evaluated the effect of dolomitic lime on earth plasters (Bamogo et al., 2022).They showed that water erosion, capillary absorption and plaster abrasion were significantly reduced at an optimum content of 4% by mass.This was due to the formation of hydrated calcium silicates (CSH) and hydrated magnesium silicates (MSH), which reduced open-pore size.The interaction of these silicates with the isolated soil particles increased the cohesive strength in the mixture with a reduction of the capillary coefficient.The decrease in thermal conductivity of the plaster could be explained by the increase in air-filled closed pores.Ouedraogo et al. (Ouedraogo et al., 2017) studied the mechanical properties of adobes stabilized with kenaf fibers.They found that the compressive strength increases with the addition of 0.2% and 0.4% kenaf fibers 3 cm and 1.5 cm long, respectively.This is explained by the presence of kenaf fibers, which prevent crack propagation in the clay matrix due to the good adhesion of their rough surfaces.Other authors have reported the beneficial effect of cow dung as a stabilizer for adobes (Millogo et al., 2016;Ngowi, 1997;Vilane, 2010).
The present study specifically aims to identify the physical characteristics of the raw earthen materials (Nshimiyimana et al., 2020) and evaluate the physical and mechanical properties as well as the sensitivity to the water of the earthen plaster containing the fermented rice husk.

Raw materials
The clay earthen material was collected from Pabré located 20 km north of Ouagadougou in Burkina Faso (12°31'23.46"' North;1°34'22.8'" East).The material was taken from 50 cm below the top layer (vegetation layer) and has a clay texture and kaolinite clay mineralogy.It has a gray color and is used locally for the artisanal production of adobe bricks.This clay has already been studied for the production of stabilized compressed earth blocks (CEBs) (Nshimiyimana et al., 2020).The clay earthen material was crushed into particles <5 mm (Figure 1a).
The rice husk (RH) is a co-product from the processing of rice paddy and is used in its raw state.It was collected from Koupéla in Burkina Faso and has already been used in a calcined state for the stabilization of CEBs (Nshimiyimana et al., 2020).The RH was sieved through a 2 mm mesh sieve to extract the fine particles (Figure 1b).The two materials were dried at 105° for 24 h until constant masses to be able to control the consistency.Mixtures containing three (03) contents of RH were studied, namely 33%, 50%, and 67% of the total volume of the mixture, to assess the optimal content.

Production of specimens
The quantities of RH corresponding to the three studied contents (33%, 50%, and 67% by volume) were fermented in a quantity of water that would allow to reach good workability after mixing with the earth.The amount of water was determined using equation (1), referring to (Bamogo et al., 2020); where WL and WP are the liquid and plastic limits of the clay earthen material.The composition of the mixtures of earthen plaster is shown in Table 1.
At the end of the fermentation time (1, 2, and 3 weeks), the mixture of water and fermented RH was added to the earth and homogenized for 15 min to obtain a paste.The paste was used to produce the test specimen of plaster.The test specimens 40 × 40 × 160 mm 3 were produced by filling the molds in two layers, each layer is subjected to a series of 15 manual strokes.The specimens were removed from the mold after 24 h and are kept in ambient conditions of the laboratory (32 ± 5°C and 35 ± 5% RH) for at least 21 days until complete drying before characterization (Figure 2).The specimen is considered dry when the difference in mass is less than 0.5%, between two successive weighing in 24 h.

Characterization of raw materials
The particle size distribution of the clay earthen material was determined by referring to two standards, NF P 94-056 (for the coarse fraction > 80 ) and NF P 94-057 (for the fine fraction ≤ 80 ) (NF EN ISO 17892-4).The Atterberg limits (liquidity and plasticity) were determined by referring to standard NF P 94-051 (NF P 94-051, 1993).The specific density of the particles was determined by referring to (NF EN 1097-6).The apparent density of the clay and the rice husk were determined by mass over the total volume occupied by the particles.
To evaluate the water absorption of rice husk, A b (%), a quantity of 20 g of rice husk was ovendried at 105°C for 24 h to obtain a dry mass (M s ) before being immersed in water and measuring the mass of soaked samples (M h ).Equation 2was used to calculate the water absorption, referring to (Bamogo et al., 2022).The evolution of the pH of water containing fermented RH was monitored using a pHmeter for 7 weeks of the fermentation.

Characterization of earthen plaster
2.2.2.1.Apparent density and total porosity.The apparent density, ρ a (g/cm 3 ), was determined according to the German standard (DIN German Institute for Standardization, 2013) using equation 3, referring to (Bamogo et al., 2020), where m (g) is the dry mass of the specimens and v (cm 3 ) is its volume.The total porosity, P t % ð Þ, was evaluated from equation 4, referring to (Bamogo et al., 2022), using the apparent density ρ s ð Þof the specimen, and the equivalent specific density ρ s:eq � � of the materials.

Linear shrinkage.
The linear shrinkage, α (%), of earthen plaster was determined in equation 5; from the difference between the initial lengths (lo) and final length (l) of the mold measured before and after drying on three prismatic test specimens (40 × 40 * 160 mm 3 ), referring to (Bamogo et al., 2020;Montana et al., 2013).

Abrasion resistance.
The abrasion resistance test was carried out on test specimens by subjecting them to mechanical abrasion using a 3 kg wire brush (DIN German Institute for Standardization, 2013).The test consists of passing the brush over the length of the specimen for 60 cycles.The abrasion coefficient, C a g=cm ð Þ; expresses the ratio between the quantity of material lost from the surface and the brushed surface in equation ( 7).m 0 -m 1 (g) is the lost mass (g) and s (cm 2 ) is the brushed surface.

Capillary absorption.
Water absorption by capillarity is measured on 40 × 40 * 160 mm 3 specimens referring to the standard NF XP 13-901 (Antunes et al., 2019).This test consists of measuring the increase in mass caused by the capillary rise on a prismatic specimen.The mass of the sample was measured before and after the immersion time, t (min), of 10 min.The mass of absorbed water, m 1 -m 0 (g), per unit of surface, S (cm 2 ), exposed to water rise allowed to determine the coefficient of capillary water absorption, Cb (g/cm 2 .min 0.5 ), by equation 8, referring to (Bamogo et al., 2022).

Resistance to water erosion.
The test used to assess the resistance of earth plaster to rain erosion consisted in projecting water at a constant flow rate of 1.5 L/min for 10 min on specimens placed on an inclined plane at an angle of 30° with respect to the horizontal (Figure 4), referring to Bamogo et al. (Bamogo et al., 2020).The mass loss, C (%), was estimated from equation 9. M 0 is the dry mass before the test, and M 1 is the mass after the test.

Characteristics of the raw materials
Table 2 shows the density and total porosity of clay earthen material and RH.The apparent density of the earth is 1340 kg/m 3 ; within the range recommended for the use of plaster or adobe soil (1200 to 2100 Kg/m 3 ) (XP P13-901, 2001).The apparent density of RH of 110 kg/m 3 is lower than that of earth; which can be explained by its higher porosity and composition of less dense plant fibers.
Figure 6 presents the granulometric curve of the earth.More than 75% of grains pass through an 80 µm sieve.It is mainly made of silt (51%), clay (24%), sand (23%), and a small amount of gravel (3%).The clay content (20 and 40%) is within the range recommended for the production of plaster (XP P13-901, 2001).The granulometric curve of the soil falls within the zone recommended for the use of plasters or adobes.The earth has a liquid limit of 32%, plastic limit of 12%, plasticity index of 20%, and methylene blue value of 2.66 g/100 g.The earthen material is located in the boundaries of the earth recommended for plasters and adobes (Figure 7).The methylene blue value of 2.66 g/100 g confirms the silty-clayey characteristics of the soil.The earthen material is classified in the category of materials  of medium plasticity.According to NF P 11-300, the soil is type A 2 (clayey silt of low plasticity and sensitive to water).
Figure 8 shows that the rich husk absorbs a large amount of water immediately after contact with water with an absorption capacity of 250% after 24 h.This water absorption capacity would be linked to the hydrophilic nature of the rice husk due to its content of cellulose and hemicellulose.This absorption capacity is comparable to that reported by Ouédraogo et al. (Bamogo et al., 2022).Damene et al. (Damene1 et al., 2020) found a higher absorption of 502% for barley straw.The high-water absorption could cause a problem of aging of the plaster with the clayey material.
The evolution of the pH of the fermented rice husk in water over time is presented in Figure 9.After 1 week of fermentation, the pH value changes from 7.51 to 5.81, for the solution changing from the basic state to the acidic state.This change in pH would be due to the extraction of polysaccharide substances (cellulose, hemicelluloses, starches, and pectin) from the rice husk.However, after this decrease of the pH at the end of the 1 st week of fermentation, an increase in pH is noticed over time.Indeed, the pH value increased from 5.81 at 1 week of fermentation to 6.32, 6.74 and 7.31, respectively, at 2, 3, and 4 weeks of fermentation.The solution therefore became basic again after 4 weeks.The pH varies little from beyond the 7 th week of fermentation.This would be due to the use of the previously extracted substances (cellulose, hemicelluloses, starches, and pectin) by the water molecules with the probable end of the extraction process being around 4 weeks after which the pH does not evolve almost anymore.According to (Vissac et   2017), the fermentation time of the rice husk is between 3 and 4 weeks so the return to the basic medium would probably mark the end of the macerating process.

Apparent density
Figure 10 shows that the density of the specimens with 33% fermented RH (M-33%RH) hardly changes compared to the reference specimens M-0% RH for all fermentation time.From 50% RH, there is a decrease in the density of the plasters from 1.88 g/cm 3 for M-0% RH to 1.57 g/cm 3 and 1.07 g/cm 3 respectively for M-50% RH and M-67% RH at 1 week of fermentation.This decrease would be due to the large volume of RH in a less dense bio-material (Lekshmi et al., 2020;Millogo et al., 2014).Šál et al. (Šál & Nováková, 2019) reported similar results for biochar-stabilized adobes.
By considering the fermentation time, it can be seen that time has a slight influence on the density of the coatings.There is a very slight increase in density with time, especially for M50% RH and M-67% RH.DIN 18,947 (DIN German Institute for Standardization, 2013) recommends the use of earth plaster whose density is between 1.60 and 1.88 g/cm 3 .Therefore, only the M-33% plaster could be suitable.

Total apparent porosity
Figure 11 presents the total apparent porosity of the specimens which reduces when 33% of RH is introduced into the soil.It decreased from 28% for M-0% RH to 18% for M-33% RH after 2 weeks of  fermentation.An increase in the total apparent porosity is, however, noticed with the increase in the content of RH to M-67% RH.This increase could be explained by a disturbance of the interactions in the mixture due to the addition of a large quantity of RH.
The results also show that the fermentation time has very little influence on the total porosity of the plaster, with a slight decrease in porosity.

Linear shrinkage
Figure 12 presents the evolutions of the linear shrinkage of the specimens after 21 days of maturation at ambient laboratory temperature (32 ± 5°C; 35 �5% HR).The shrinkage of the specimens decreased with the increase in the content of RH.After 2 weeks of fermentation, the shrinkage evolved from 4.52% for the reference M-0% RH to 3.96%, 2.71% and 1.04%, respectively, for M-33% RH, M-50% RH and M-67% RH.This decrease in shrinkage is due to the reinforcement role of the RH fibers in the mixture, due to their rough surfaces.Similar results have been reported in the literature on earth plasters reinforced with plant fibers (Faria et al., 2016;Lekshmi et al., 2020).
Considering the fermentation time, a decrease in shrinkage is noticed after 4 weeks of fermentation.This could be due to a possible formation of amine silicate, an adhesive molecule resulting from the reaction between the substances extracted from the RH during the fermentation (Bamogo et al., 2020).There is, however, an increase in linear shrinkage, for M-50% RH and M-67% RH, between 2 and 3 weeks of fermentation.This phenomenon could be justified by a large decomposition of the fibers after 3 weeks compared to 2 weeks (Bamogo, 2020).According to (DIN German Institute for Standardization, 2013), the shrinkage of earthen plasters stabilized with vegetable fibers must be less than 3%, which is reached with 50% RH.

Compressive strength
Figure 13 presents the evolution of the compressive strength of the plaster containing RH.It can be seen that the compressive strength of the plaster does not vary for 33% of RH compared with the reference plaster without RH.However, the strength decreased from 50% of RH compared to the reference plaster.This could be due to the high content of RH and the high porosity in these plasters.Thus, at 2 weeks of fermentation, the compressive strength reached the value of 3.88 MPa, 3.85 MPa, 2.15 MPa, 0.82 MPa, respectively, for M-0% RH, M33% RH, M −50% RH, M-67% RH.
The same results have been reported in the literature with the use of fibers for the stabilization of rammed earth plasters and adobes (Lekshmi et al., 2020;Šál & Nováková, 2019).
There is also a slight increase in the compressive strength with the fermentation time which would be due to the production of more amine silicate which ensures the adhesion between the fibers and the isolated clay particles (Bamogo et al., 2020).
According to DIN 18,947 (DIN German Institute for Standardization, 2013), the raw earth plasters must have a minimum compressive strength of 1 MPa (class SI).The strength of M-33% RH and M-50% RH are greater than 1.5 MPa (class SII) and would be better suited as raw earth plaster.

Abrasion resistance
Figure 14 shows that with 33% RH, the plaster has almost the same abrasion resistance as the reference coating without RH.However, a decrease in this resistance is noted, as the coefficient of abrasion increased with the increase in RH.This is characterized by an increase in the loss of mass of the specimens during the abrasion test.The abrasion coefficient thus increased after 2 weeks of fermentation from 0.26 g/cm 3 for M-33% RH to 0.67 g/cm 3 and 2.69 g/cm 3 respectively for M-50% RH and M −67% RH.This increase would be linked to the addition of RH fiber which are less resistant than clay particles.This decreasing trend agrees with the results reported by Šál et al. (Šál & Nováková, 2019) on bricks stabilized by fermented biochar.
The fermentation time has very little influence on abrasion resistance with M-67% RH, on the one hand.On the other hand, there is an increase in mass loss with M-33% RH and M-50% RH, especially at 4 weeks compared to 3 weeks of fermentation.This is probably due to a high disaggregation of the RH fibers, reducing the cohesion in the plaster (Bamogo, 2020).
The DIN standard classifies plasters into two classes with regards to abrasion resistance.Up to 50% BR, the plasters are in class SII with Ca >1.5 g/cm 3 .Therefore, the M-33% RH and M50% RH will be better suited for abrasion resistance.

Thermal conductivity
Figure 15 shows that the thermal conductivity of the plaster decreases with the increase in the content of RH.This would be linked to the cellulose contained in the RH, which is a very insulating molecule (Ouedraogo et al., 2019).After 2 weeks of fermentation, the thermal conductivity value evolved from 0.87 W/m.K for the reference plaster without RH to 0.69 W/m.K, 0.42 W/m.K, and 0.13 W/m.K, respectively, for M-33% RH, M-50% RH, and M-67% RH.These results are within the same range as those of the plaster stabilized by cow dung, where the thermal conductivity decreased from 0.85 W/m.K without cow dung to 0.75 W/m.K with 6% addition by weight of cow dung (Bamogo et al., 2020).Ouedraogo et al. (Ouedraogo et al., 2019) also reported that the thermal conductivity of adobes evolved from 1.1 W/m.K to 0.35 W/m.K after addition of 1% by weight of fonio.
The thermal conductivity slightly increases for M-33% RH and M-50% RH with the increase in fermentation time.This could be justified by a degradation of the cellulosic wall of the RH fibers.On the contrary, the trend is rather reversed for M-67% RH with a decrease in conductivity over time.

Capillary absorption
Figure 16 shows the capillary absorption capacity of the plaster after 10 min.The M-33% RH plaster has almost the same capillarity as the reference plaster without RH.However, there is an increase in capillarity for M-50% RH, while the M-67% RH plaster disintegrated in water before the end of the test.This shows that 67% of RH is too much to use in plasters.This evolution could be explained by the high-water absorption of the cellulose-rich RH which contributes to increasing the absorption of the plaster with the increase in the RH content.These results corroborate the report of Šál et al. (Šál & Nováková, 2019) on bricks stabilized by fermented biochar.The disintegration of the specimens of the M-50% RH plaster indicates that the 50% RH content would also be too much.

Resistance to water erosion
Figure 17 presents the evolution of the water erosion resistance of the specimens after 10 min of testing.The loss of mass of the specimens increased with the increase in the RH content.The M-67% RH plaster broke before the end of the test, confirming that this RH content is too high.The increase in mass loss can be related to the decrease in the abrasion resistance of plaster and the high-water absorption capacity of RH.The study by Bamogo et al. (Bamogo et al., 2020) showed that the water erosion of the plaster was stabilized by the addition of 6% cow dung  decreased.Šál et al. (Šál & Nováková, 2019) obtained the same decreasing trend on bricks stabilised with fermented biochar.The different evolution in the present study could be linked to the biochemical nature of the fibers, the grain size of the soil, and the content of fibers.
There was no variation between 2 and 3 weeks of fermentation.There is a reduction in the loss of mass for M-33% RH at 4 weeks, which can be explained by the formation of more silicate amine which is relatively insensitive to water.The M-50% RH plaster disintegrates as in the case of the absorption test confirming that the content of 50% would be too high.

Conclusion
This study focused on the design of an earthen plaster stabilized with fermented rice husk and intended for coating adobe walls.The study consisted in characterizing the earth and the rice husk and designing the plaster using a previously fermented rice husk at different fermentation times and at different rice husk contents.The study characterized the physico-mechanical, thermal and hydric properties of designed plaster.The following key conclusions were drawn: • The rice husk allowed to reach similar compressive strength for M-33% RH and the reference, beyond which it decreased from 3.88 to 0.82 MPa.However, it also reduced the shrinkage from 4.52 to 0.83% and the risks of cracks due to the drying shrinkage.
• The density is also constant for M-33% RH, beyond which it decreases from 1.88 to 1.07 g/cm 3 .This could explain the negative impact on the mechanical properties with 50% and 67% of RH.
• The thermal conductivity of adobes decreased sharply from 0.87 to 0.05 W/m.K with the addition of rice husk due to the insulating nature of the cellulose contained in the rice husk.
• The capillary absorption, water erosion, and abrasion resistance were rather negatively affected with the increase in rice husk.This evolution can be justified by the high-water absorption of the rice husk, the large quantity of rice husk which are less resistant than the soil.
• Regarding the current results, the optimal design of plaster is M-33% RH for 3 weeks of fermentation.
To fill the gaps observed through the current study, the following recommendation should be addressed in future work: • Perform chemical, mineralogical, and hygroscopic characterizations of raw materials and plaster for a better understanding of the mechanisms of stabilization by fermentation.
• Redesign the content of rice husk in the plaster and fermentation method of rice husk to improve the cohesion of the mixture and use a more clayey earth material; • Study the compatibility of the plaster and adobe wall and use other agricultural byproducts for the improvement of the plaster.
Figure 1.a) Visual appearance of the a) clay earthen material under 5 mm b) the rice husk under 2 mm.

Figure
Figure 2. Schematic procedure for the production and characterization of specimens.
2.2.2.3.Compressive strength.The compressive strength, Rc (MPa) was tested referring to the German standard DIN 18,947 (DIN German Institute for Standardization, 2013).It is the ratio of the force, F c (N), of the failure applied on the surface, S (mm 2 ) (equation 6).It was evaluated on the six half-prisms obtained during the 3-point bending test using a hydraulic press (Figure3).

Figure
Figure 3. Experimental setup for compression test.

Figure
Figure 4. Experimental setup for water erosion test.

Figure
Figure 5. Experimental setup for thermal conductivity test.

Figure 6 .
Figure 6.Particle size curve of earthen material.

Figure
Figure 10.Variation in the apparent density of the plaster with the content of RH.