The investigation consisted of collecting information from people living in rural areas of Burkina Faso on their knowledge of traditional water treatment with plant extracts. It covered four regions of the country with the lowest rates of access to drinking water namely East, Sahel, Mouhoun loop and Cascades regions. In these regions, 12 localities were taken into account, namely Djibo, Dori, Sebba, Fara, Pompoi, Tougan, Fada, Kantchari, Pama, Banfora, Niangologho, and Mangodara (Figure 1).

Once the survey was completed, the samples were collected across the country. The seeds of Accacia nilotica, Adansonia digitata, Balanites egyptiaca, Moringa Oleifera; Tamarindus indica, Aloe vera, and Opuntia ficus indica were obtained at the National Forest Seed Center (NFSC). Boscia senegalensis seeds and Capparis corymbosa leaves were collected in the urban park “bangr weeogo” of Ouagadougou (Figure 2).

The raw water samples used in this study were surface water collected in the Loumbila dam (12° 29 N, 01° 24 W). The collected samples were kept in the refrigerator at 4°C.

For the preparation of the solutions based on seeds of Accacia nilotica, Adansonia digitata, Balanites egyptiaca, Moringa Oleifera, Tamarindus indica, and Boscia senegalensis, the different seeds were ground into a fine powder. 100 g of the powder obtained in each case was diluted in 1 L of distilled water and placed under magnetic stirring for 1 h, to allow the release of the active molecules. The final concentration of each prepared solution was 100 g/L.

The solution based on Capparis corymbosa leaves were obtained by reducing them to a fine powder of which 100 g were macerated in 1 L of distilled water during 2 h.

For the preparation of the solutions containing Aloe vera, and Opuntia ficus indica, 50 g of their stems were cleaned, and crushed in 1 L of distilled water, to extract their sap. Thus, 02 stock solutions of 50 g/L each of their extracts were obtained.

After the preparation of different solutions based on plant extracts, their effectiveness was evaluated using the jar test. The jar test consists of carrying out small-scale potabilization tests in the laboratory using the coagulation-flocculation process. Thus, the previously prepared solutions were tested at different concentrations on the surface water samples using an electrically controlled six-station flocculator (brand name velp scientifica).

1 L of raw water was introduced into each flocculator beaker, followed by the addition of coagulant solutions at increasing concentrations. The agitation of the water after introduction of the coagulant was done in two phases: a fast agitation at 150 rpm for 5 min and a slow agitation at 45 rpm for 10 min. After at least 10 min of decantation, the supernatant of each beaker was taken and used to read the residual turbidity.

Coagulation/flocculation performance is evaluated by the turbidity removal rate, and our experiments were repeated in triplicate. Thus, the percent turbidity removal was noted (% T) and calculated by the following formula: %   T = T i − T f T i × 100 Where, Ti is initial turbidity of the water and Tf is the final turbidity after the jar test.

The physicochemical parameters of water samples were measured before and after treatment with plants extracts. The turbidity of the samples was measured by a laboratory turbidimeter (WTW Turb 550 IR) in accordance to the French standard NF ISO 7027 (2000). The pH of the water was measured with a pH meter (330i WTW), according to the NF 10523 standard (1994). The Alkalimetric Title (AT), the Complete Alkalimetric Title (CAT), and the concentration of calcium and magnesium ions (Hydrotimetric Title, HT) were determined by titrimetry. Each parameter was determined in triplicate. The removal efficiency of the analyzed parameters was determined by the formula below where, Ci represents the concentration of the parameter in the raw water and Cf represents the concentration of the same parameter in the treated water: %   C = C i − C f C i × 100

The water treated with our various coagulants and flocculants must be free of germs, and particular emphasis is placed on microbiological analyses. Thus, the main indicator germs of fecal contamination were retained for this study in accordance with the regulations in force. These are Escherichia Coli, total coliforms, and streptococci. These germs were all determined using the membrane filtration method and spread on specific culture media according to the French standard NF EN ISO 9308-1 (2000). Thus, for the research of total coliforms and Escherichia Coli, the Chromocult coliform Agar ES medium was used at an incubation temperature of 37°C and for streptococci, the Enterrococus agar medium was used at a temperature of 44°C.

Each microbiological parameter was determined in triplicate, and the percentage reduction of microorganisms (% M) was determined by the following formula: %   M = M i − M f M i × 100 Where, Mi is the number of colonies of the microorganism initially present in the water and Mf the number of colonies of the same microorganism after the jar test.

Graphs and the different concentrations calculations were performed using Microsoft Excel software, version 2016. Data were subjected to the analyses of variance (ANOVA) using XLSTAT software (2016) and significant differences between means were revealed via the Tukey test (p < 0.05). Principal component analysis was performed with R software, version 4.0.2 (2020).

Of 200 people surveyed (Figure 3), 26.5% had no knowledge of traditional water treatment processes. 23% knew only of physical processes such as filtration, heating and sand filtration. 50.5% were able to identify plants whose extracts could be used for water purification. Thus, we were able to identify nine plants, namely Accacia nilotica, Adansonia digitata, Aloe vera, Balanites egyptiaca, Boscia senegalensis, Capparis corymbosa, Moringa oleifera, Opuntia ficus indica, and Tamarindus indica (Table 1).

Preliminary tests in the laboratory showed a very insignificant turbidity reduction, or even almost zero for the solutions based on Acacia nilotica, Adansonia digitata, Balanites aegyptiaca, Capparis corymbosa, and Tamarindus indica (Figure 4). These solutions were all applied at increasing concentrations (0.1 g/L to 10 g/L) to waters ranging in turbidity from 241 to 386 NTU to observe kinetics. However, no significant reduction in turbidity was observed after 2 h of settling for any of the concentrations applied.

For the Aloe vera and cactus (Opuntia ficus indica) treatments, the turbidity abatement was low and not very satisfactory after 1 h of decantation (Figure 5). Application of the cactus-based coagulant solution to 184.8 NTU water reduced it to 125.0 NTU. The Aloe vera solution reduced to 167.0 NTU, a water whose initial turbidity was 210.6 NTU. Thus the turbidity abatement rate was 32.35% for the cactus-based treatment, and 20.7% for the Aloe vera-based treatment.

Solutions based on Boscia senegalensis seeds and Moringa oleifera seeds, on the other hand, showed a significant reduction in turbidity after only 1 h of decantation (Figure 6). The turbidity of the water treated by the first coagulant solution (Boscia) was reduced from 377.1 NTU to 57.2 NTU; i.e., an abatement of 84.83% obtained for a concentration of 2.5 g/L. The second coagulant solution (Moringa) reduced the turbidity of water from 276.6 NTU to 47.6 NTU; i.e., an abatement of 82.97%, obtained for a concentration of 1 g/L.

To compare the efficacy of Moringa and Boscia biocoagulants with that of aluminum sulfate, water samples of 352.80 NTU were used for coagulation-flocculation tests with increasing doses of aluminum sulfate, Moringa oleifera seed extract, and Boscia senegalesis seed extract. The variations in residual turbidity are recorded in Figures 9A, B. Figure 9A allows us to observe a decrease in turbidity as a function of the concentration of aluminum sulfate used. As its concentration increases, the turbidity of the water decreases. However, this dependence is not linear. The maximum reduction in turbidity is 92.8% and was obtained with a dose of 0.05 g/L of aluminum sulfate solution.

Determination of the optimal concentration for Moringa and Boscia seeds show that at low doses, the turbidity of the water sample decreased with increasing biocoagulant concentration (Figure 9B). However, above 0.9 g/L of Moringa solution used, and 1 g/L of Boscia solution, a slight increase in turbidity was observed. Thus, the optimal concentration of Moringa coagulant needed to treat the raw water sample was 0.9 g/L, while that of Boscia senegalensis was 1 g/L. At these concentrations, we obtained a turbidity reduction of 82.06% for Moringa and 80.78% for Boscia, respectively.

These results thus show that aluminum sulfate, Moringa oleifera and Boscia senegalensis seeds are all effective coagulants in reducing raw water turbidity, although to different degrees. Indeed, for the same 352.80 NTU water, higher concentrations of biocoagulant (0.9 g/L for Moringa and 1 g/L for Boscia) than aluminum sulfate (0.05 g/L) were required to obtain the lowest turbidity.

Figures 10A, B show that the Aluminum Sulfate/cactus, Moringa/Aloe and Boscia/cactus combinations provide considerable turbidity reduction after 15 min of decantation.

The comparison of these three (3) combinations reveals a clear improvement in the coagulation capacity of aluminum sulfate, Moringa seeds, and Boscia seeds. The turbidity obtained with these three (3) combinations after 15 min of decantation is in accordance with the WHO standard (≤5 NTU). In addition to the improvement of the water clarification speed, the addition of cactus mucilage (0.6 mL) and Aloe (0.7 mL) to the coagulation-flocculation process resulted in the reduction of the optimal concentration of aluminum sulfate. Thus, the optimal concentration of aluminum sulfate used to obtain a turbidity in compliance with the standard went from 0.05 g/L to 0.04 g/L, while the concentration of Moringa and Bocia remained the same (respectively 0.9 and 1 g/L). Even with the addition of mucilage, the coagulant requirement for Moringa and Boscia remains higher than that for aluminum sulfate to achieve turbidity that meets the standard.

In addition to turbidity, measurements of pH, conductivity, HT, CAT, calcium ions (Ca²⁺), sodium ions (Na⁺) and potassium ions (K⁺) showed a slight increase (Figure 11), after treatment of the water with Moringa oleifera seeds and Boscia senegalensis seeds (Table 2).

A total absence of Enterococci in the raw water and treated water was observed. Moreover, the total coliforms that were present in the raw water (144 CFU/100 mL) have clearly decreased in the water treated with Moringa Oleifera seeds (about 90.27%) and Boscia Senegalensis seeds (about 85.41%)

Escherichia Coli was present in the raw water (21 CFU/100 mL). With the Moringa oleifera and the Boscia senegalensis seeds, the treatment reduced E. coli to 1 CFU/100 mL (about an abatement of 95.23%) and 3 CFU/100 mL (about abatement of 85%), respectively (Figure 12).

However, it should be noted that even if these treatments were allowed to obtain good yields in terms of elimination of microorganisms, these last ones nevertheless experienced a slow proliferation as the treated water remained stored at room temperature (Figures 13A, B).