Investigating the Effectiveness of a Simple Water-Purifying Gadget Using Moringa oleifera Seeds as the Active Beads

Abstract

Clean water scarcity in developing countries like South Africa poses significant health risks. This study investigated the effectiveness of a simple water purification device using Moringa oleifera Lam. seeds as active beads, offering a novel, low-cost, and sustainable solution for water treatment in resource-limited settings. The device combined M. oleifera seed powder with activated charcoal and cotton wool, providing a locally adaptable and environmentally friendly solution. Water samples were collected from three sites along the Pienaars River during winter and summer, and M. oleifera seeds were ground into three particle sizes (710 µm, 1000 µm, and 2000 µm) for testing. Results showed that the device significantly reduced microbial loads, with the total coliforms decreasing by 60–85%, E. coli by 50–75%, Salmonella spp. by 40–70%, and Shigella spp. by 30–65% across sampling points. However, filtered samples still exceeded the WHO and SANS guidelines, with microbial counts remaining above 0 CFU/100 mL. Physicochemical properties, including pH (6.02–7.73), electrical conductivity (17.8–109.5 mS/m), and ion concentrations (e.g., nitrate: 0.21–39.55 mg/L; chloride: 8.57–73.55 mg/L), complied with the SANS 241:2015 and WHO drinking water standards. The finest particle size (710 µm) demonstrated the highest microbial reduction and increased magnesium concentrations by up to 30%. Seasonal variations influenced the performance, with summer samples showing a better microbial removal efficiency (70–85%) compared to winter (50–70%). This study highlights the potential of M. oleifera-based filtration as a low-cost, sustainable solution for reducing microbial contamination, though further refinement is needed to meet drinking water standards. This research introduces a novel approach to water purification by combining M. oleifera seed powder with activated charcoal and cotton wool, providing a locally adaptable and environmentally friendly solution. The findings contribute to the development of scalable, natural water treatment systems for resource-limited communities.

1. Introduction

Access to clean water is a fundamental necessity for human survival, yet it remains a significant challenge in many developing countries, including South Africa [1,2]. Water scarcity and pollution have led to a deterioration in both the quality and quantity of water resources, with climate change, droughts, and inadequate water management exacerbating the problem [2]. South Africa, ranked 26th out of 164 countries in terms of water availability per person, is projected to face a 17% deficit in its water supply by 2030 [3].

The World Health Organization (WHO) recommends a daily minimum of 2–3 L of clean drinking water for adults to maintain proper hydration and overall health [4]. However, approximately 159 million people globally consume untreated surface water from streams and lakes, putting them at risk of waterborne diseases [5]. In South Africa, the majority of households in urban areas have access to piped water; however, many rural areas still rely on unprotected water sources [6].

Conventional water treatment methods, such as coagulation–flocculation using synthetic materials, are often costly and may not be accessible to resource-limited populations [7]. Moreover, the use of synthetic coagulants has been associated with environmental and health concerns, including the release of toxic residues and the potential for aluminum accumulation in treated water [8]. These limitations have spurred interest in natural coagulants derived from tropical plants, such as Moringa oleifera Lam., which offer a low-cost, environmentally friendly alternative for water purification [8,9,10]. Historical evidence suggests that M. oleifera seeds have been used for centuries in traditional water treatment practices in regions such as Sudan and Ethiopia [11,12]. Recent studies have demonstrated its effectiveness in reducing turbidity, microbial loads, and heavy metal concentrations in water [12,13]. For instance, the investigation conducted by Madsen et al. [14] demonstrated that M. oleifera seeds are capable of achieving a reduction in turbidity by up to 99.5% and a decrease in bacterial counts ranging from 90% to 99% within the initial 1 to 2 h of treatment. In similar studies, research conducted by Narayasamy et al. [15] and Amagloh et al. [16] underscored the promising potential of M. oleifera as a viable and sustainable alternative to conventional synthetic coagulants in the context of developing nations.

This study introduces a novel approach by developing a simple, low-cost water purification device that combines M. oleifera seed powder with activated charcoal and cotton wool. Unlike previous studies that focused solely on M. oleifera as a coagulant, this research explores its integration into a multi-layered filtration system, offering a more comprehensive solution for microbial and physicochemical water purification. By evaluating the effectiveness of this device across different seasons and sampling sites, this study provides new insights into the adaptability and scalability of M. oleifera-based water treatment systems in diverse environmental conditions.

This study aims to investigate the effectiveness of a simple water purification device using M. oleifera seeds as active beads. By examining water samples from the Pienaars River in South Africa, we seek to evaluate the potential of this natural coagulant in reducing microbial contamination and improving water quality.

2. Materials and Methods

2.1. The Study Area and Sampling Points

This study was conducted along the Pienaars River, with three sampling points selected to represent varying environmental conditions and contamination levels (Figure 1). Moretele (25°07′50.1″ S 27°57′34.5″ E), located upstream in a rural area, serves as a baseline with minimal human impact, primarily influenced by agricultural runoff and livestock activities. Mamelodi (25°40′43.0″ S 28°21′27.0″ E), situated downstream in a peri-urban area, reflects high contamination from domestic wastewater, informal settlements, and small-scale industries. Tierpoort (25°52′16.0″ S 28°26′29.0″ E), further downstream in a semi-rural area, represents a transitional zone with mixed agricultural, residential, and light industrial influences. These sites were chosen to evaluate the filtration device’s effectiveness across diverse contamination scenarios, from low to high pollution levels. Water samples were collected during the winter (June to August) and summer (December to February) seasons, with four replicate samples collected per season at each sampling point.

2.2. Sampling Procedures

Standard water collection techniques were employed following Meybeck et al. [17] and DWAF [18] guidelines. Samples were collected in 1 L polyethylene sterile plastic bottles approximately 10–20 cm below the water surface in areas with gentle water movement. This depth was chosen to avoid surface debris and ensure representative sampling of the water column. Containers were sealed immediately to prevent contamination and stored in thermal containers with ice for transportation to the laboratory. Samples were refrigerated and analyzed within 16 h of collection.

2.3. Moringa oleifera Seed Preparation

Brown M. oleifera seeds were obtained from Lefakong Moringa farm in Boosplaas, North West Province, South Africa. The seeds were oven-dried at ±25 °C for 5 days, manually cracked to obtain kernels, and crushed into powder using a mortar and pestle (Figure 2). The powder was sieved to three different particle sizes: 710 µm, 1000 µm, and 2000 µm.

2.4. Prototype Water Filtering Device

A prototype water filtering device was constructed using a 500 mL capacity polypropylene sterile syringe with an internal and outer diameter of 6 mm and 9 mm depth, respectively. The device contained 20 g of M. oleifera seed powder (equivalent to a layer depth of approximately 15 mm), 10 g of activated charcoal (equivalent to a layer depth of approximately 10 mm), and 9 mm of cotton wool (Figure 3). The filter materials were arranged in the order of cotton wool (C), activated charcoal (AC), M. oleifera seed granules (MSG), and cotton wool (C) to optimize filtration efficiency. The top cotton layer removed large particles, preventing clogging. The activated charcoal was meant to adsorb organic pollutants and heavy metals, improving water taste and chemical quality. The M. oleifera granules acted as a natural coagulant and antimicrobial agent, targeting fine particles and microorganisms. The bottom cotton layer captured any remaining residues, ensuring clear and safe water. This layered design ensured a graded filtration process, addressing both physicochemical and microbial contaminants.

2.5. Water Quality Analysis

Physical parameters (pH and electrical conductivity) were measured using an HT-1202 Digital pH meter (Hanna Instruments, Cluj-Napoca, Romania) and a Bante901 Benchtop Conductivity Meter (Bante Instruments Co., Ltd., Shanghai, China). Metal elements were analyzed using an Agilent 725 Series ICP-OES instrument. Anions were determined using a Dionex 5000+ Ion Chromatography system (Thermo Fisher Scientific, Waltham, MA, USA). Bicarbonate concentration was measured through titration with a standardized HCl solution.

Table 1. Physicochemical and microbial parameters analyzed in water samples.

Parameter	Symbol	Method	Units
pH	pH	pH probe
Electrical Conductivity	E C	E C meter	mS/m
Nitrate	$\mathrm{N}{\mathrm{O}}_3^{-}$	IC	mg/L
Nitrite	$\mathrm{N}{\mathrm{O}}_2^{-}$	IC	mg/L
Chloride	${\mathrm{C}\mathrm{I}}^{-}$	IC	mg/L
Fluoride	${\mathrm{F}}^{-}$	IC	mg/L
Sulfate	${\mathrm{S}\mathrm{O}}_4^{2-}$	IC	mg/L
Phosphate	${\mathrm{P}\mathrm{O}}_4^{3-}$	IC	mg/L
Bicarbonate	${\mathrm{H}\mathrm{C}\mathrm{O}}_3^{-}$	Titration	mg/L
Sodium	Na	ICP-OES	mg/L
Potassium	K	ICP-OES	mg/L
Calcium	Ca	ICP-OES	mg/L
Magnesium	Mg	ICP-OES	mg/L
Boron	B	ICP-OES	mg/L
Total coliforms	TC	Viable plate count	CFU/mL
Escherichia coli	EC	Viable plate count	CFU/mL
Salmonella	Sal	Viable plate count	CFU/mL
Shigella	Shi	Viable plate count	CFU/mL

indicates the parameters that were measured.

2.6. Microbiological Analysis

Water samples were subjected to serial dilutions and plated on appropriate agar media. The E. coli and total coliforms were cultured on MacConkey agar, while Salmonella and Shigella were cultured on Xylose Lysine Deoxycholate (XLD) agar. Plates were incubated at 25 °C for 48 h, and colonies were counted using an automated colony counter (ZR-1101). We analyzed 4 treatments, replicated 4 times per site.

2.7. Data Analysis

Statistical analysis was performed using R statistical package version 3.6.3 [19]. One-way ANOVA was used to evaluate water quality variations among sampling points (p ≤ 0.05), with means separated using the least significant difference test. The results were visualized using GraphPad Prism software (version 5.0, GraphPad Software Inc., La Jolla, CA, USA).

3. Results

3.1. Physicochemical Parameters

Significant variations (p < 0.05) in the pH were observed across particle sizes and unfiltered water samples (

Table 2. Mean (±SE) values of physicochemical determinants of water quality based on samples collected from Moretele, Mamelodi, and Tierpoort sampling points along Pienaars River in summer, 2021. Large, medium, and fine represent particle sizes of Moringa oleifera seed powder.

Parameter		Moretele Site	Mamelodi Site	Tierpoort Site	SANS 241:2015 Limit	WHO Limit for Drinking Water	Conformity
pH	Unfiltered	6.98 ± 0.02 a	6.77 ± 0.04 a	6.76 ± 0.3 a	≥5 to ≤9.7	6.5–8.5	Conforms
	Large	6.16 ± 0.04 a	6.71 ± 0.01 b	6.37 ± 0.07 a
	Medium	6.64 ± 0.03 a	5.61 ± 0.01 b	6.62 ± 0.02 a
	Fine	6.02 ± 0.08 a	6.01 ± 0.01 a	6.23 ± 0.03 b
EC (mS/m)	Unfiltered	42.2 ± 0 b	34.35 ± 0.15 a	17.8 ± 0.2 c	≤170	-	Conforms
	Large	99.75 ± 0.25 b	80.38 ± 0.37 a	17.85 ± 0.25 c
	Medium	108.5 ± 0.5 b	92.5 ± 0.5 a	107.5 ± 0.5 b
	Fine	70.5 ± 0.5 b	65.55 ± 0.55 a	23.9 ± 0.1 c
$\mathrm{N}{\mathrm{O}}_3^{-}$ (mg/L)	Unfiltered	12.05 ± 0.05 a	4.26 ± 0.53 b	0.85 ± 0.01 c	<49	50	Conforms
	Large	0.35 ± 0.01 a	0.34 ± 0.02 a	0.63 ± 0.01 b
	Medium	11.45 ± 1.45 a	0.21 ± 0.01 b	0.57 ± 6.01 b
	Fine	10.5 ± 0.5 a	3.80 ± 0.8 b	0.52 ± 0.01 c
$\mathrm{N}{\mathrm{O}}_2^{-}$ (mg/L)	Unfiltered	0.07 ± 0.05 b	4.7 ± 0.24 a	0.05 ± 0.02 b	<3	<3	Non-Conforms
	Large	0.02 ± 0.01 a	0.02 ± 0.00 a	0.03 ± 0.0 b
	Medium	0.02 ± 0.01 b	0.02 ± 0.00 a	0.01 ± 0.00 a
	Fine	0.05 ± 0.01 b	0.03 ± 0.01 a	0.02 ± 0.10 a
Cl− (mg/L)	Unfiltered	36.85 ± 0.15 b	17.35 ± 0.05 a	8.57 ± 0.17 c	≤300	≤500	Conforms
	Large	65.3 ± 0.05 b	50.5 ± 0.5 a	8.91 ± 0.01 c
	Medium	72.94 ± 0.05 b	56.5 ± 1.5 a	12.5 ± 0.5 c
	Fine	64.5 ± 0.5 b	50.81 ± 0.19 a	9.3 ± 0.01 c
F− (mg/L)	Unfiltered	0.29 ± 0.05 b	0.22 ± 0.10 a	0.16 ± 0.01 c	≤1.5	1.5	Conforms
	Large	0.05 ± 0.03 a	0.05 ± 0.01 a	0.15 ± 0.03 b
	Medium	0.06 ± 0.01 a	0.06 ± 0.02 a	0.22 ± 0.01 b
	Fine	0.36 ± 0.10 b	0.15 ± 0.03 a	0.27 ± 0.02 c
${\mathrm{S}\mathrm{O}}_4^{2-}$ (mg/L)	Unfiltered	30.6 ± 0.1 b	27.15 ± 0.05 a	9.61 ± 0.02 c	≤250/≤500	-	Conforms
	Large	26.55 ± 0.35 a	25.75 ± 49.25 a	11.46 ± 0.46 b
	Medium	29.5 ± 0.5 a	25.5 ± 0.5 b	8.05 ± 0.5 c
	Fine	19.5 ± 0.5 b	13.7 ± 0.7 a	8.05 ± 0.05 c
${\mathrm{P}\mathrm{O}}_4^{3-}$ (mg/L)	Unfiltered	1.9 ± 0.03 b	0.75 ± 0.01 a	0.1 ± 0.0 c	-	-	Conforms
	Large	0.76 ± 0 ab	0.13 ± 2.12 a	0.1 ± 0.0 b
	Medium	0.95 ± 0.05 b	0.05 ± 0.5 a	0.1 ± 0 c
	Fine	0.14 ± 0.01 b	0.58 ± 0.16 a	0.1 ± 0 c
${\mathrm{H}\mathrm{C}\mathrm{O}}_3^{-}$ (mg/L)	Unfiltered	149.5 ± 0.5 b	140.5 ± 0.5 a	86 ± 0.6 c	-	-	Conforms
	Large	110.5 ± 0.5 a	140.2 ± 1.5 a	38.62 ± 38.62 a
	Medium	102.5 ± 0.5 b	98.5 ± 0.5 a	2.5 ± 0.5 a
	Fine	192.5 ± 0.5 b	128.5 ± 0.5 a	111.45 ± 0.45 c
Na (mg/L)	Unfiltered	31.65 ± 0.05 b	15.05 ± 0.05 a	5.92 ± 0.02 c	≤200	20	Conforms
	Large	20.51 ± 0.49 a	15.2 ± 0.02 a	3.9 ± 0.3 b
	Medium	13.1 ± 0.2 b	13.5 ± 0.5 b	2.5 ± 0.5 a
	Fine	9. 3 ± 1 b	9. 2 ± 2 b	2.2 ± 0 a
K (mg/L)	Unfiltered	7.28 ± 0.20 b	3.32 ± 0.02 a	1.89 ± 0.02 a	≤50	3000	Conforms
	Large	22.1 ± 4.89 c	3.31 ± 0.02 a	2.12 ± 0.12 a
	Medium	16.5 ± 0.50 b	13.95 ± 0.05 b	19.5 ± 0.5 c
	Fine	19.35 ± 0.05 c	8.1 ± 0.01 b	2 ± 0.10 a
Ca (mg/L)	Unfiltered	28.65 ± 0.05 b	27.7 ± 0.03 b	14.65 ± 0.05 a	≤150	-	Conforms
	Large	28.95 ± 3.43 b	27.85 ± 0.05 b	15.5 ± 0.5 a
	Medium	31.5 ± 0.5 c	28.01 ± 1.10 b	31 ± 1.13 c
	Fine	29.55 ± 0.64 b	30.08 ± 0.5 c	15.8 ± 0.14 a
Mg (mg/L)	Unfiltered	14.4 ± 0.01 b	17.45 ± 0.15 a	9.7 ± 0.11 c	≤70	≤50	Conforms
	Large	20.02 ± 1.17 a	17.45 ± 0.05 a	9.95 ± 0.03 c
	Medium	29.5 ± 0.5 b	21.5 ± 1.5 a	29.5 ± 0.5 b
	Fine	29.55 ± 0.35 b	18.75 ± 0.25 a	8 ± 0.01 c
B (mg/L)	Unfiltered	0.04 ± 0.03 b	0.02 ± 0.01 a	0.01± 0.00 c	<2400	2.4	Conforms
	Large	0.03 ± 0.01 b	0.01 ± 0.14 a	0.01 ± 0.00 a
	Medium	0.03 ± 0.02 b	0.01 ± 0.11 a	0.01 ± 0.00 a
	Fine	0.02 ± 0.03 b	0.01 ± 0.02 b	0.01 ± 0.00 c

Note: “Conforms” means that the parameter meets both SANS 241:2015 Limit and WHO Guidelines for drinking water. Unfiltered = unfiltered water (control), Means highlighted in bold means “non-conformance”, and Means in the same row indicated by different superscript letters are statistically significant (p ≤ 0.05).

). During summer, the pH at the Tierpoort site for fine particles was higher compared to the Mamelodi and Moretele sites. Conversely, in winter, the Moretele site exhibited a higher pH. Overall, summer pH values were lower than winter pH values.

Electrical conductivity (EC) values showed significant variations across sites and particle sizes in both seasons. In winter, the Moretele site exhibited higher EC and chloride levels in the unfiltered water and in large and fine particle sizes. Sodium values were also elevated in winter in both the unfiltered water and particle sizes compared to the Mamelodi and Tierpoort sites (

Table 3. Mean (±SE) values of physicochemical determinants of water quality based on samples collected from Moretele, Mamelodi, and Tierpoort sampling points of Pienaars River in winter, 2021. Large, medium, and fine represent particle sizes of Moringa oleifera seed powder.

Parameter		Moretele Site	Mamelodi Site	Tierpoort Site	SANS 241:2015 Limit	WHO Limit for Drinking Water	Conformity
pH	Unfiltered	7.63 ± 0.01 b	7.51 ± 0.01 a	7.73 ± 0.01 c	≥5 to ≤9.7	≤6.5–8.5	Conforms
	Large	6.14 ± 0.12 ab	5.57 ± 0.32 b	7.73 ± 0.02 c
	Medium	7.19 ± 0.06 a	7.03 ± 0.02 a	7.5 ± 0.24 a
	Fine	6.06 ± 0.04 a	5.35 ± 0.02 b	6.01± 0.14 a
EC (mS/m)	Unfiltered	74.78 ± 1.93 a	70.53 ± 0.47 a	21.5 ± 1.5 b	≤170	-	N/A
	Large	100.3 ± 0.7 b	80.75 ± 0.75 a	21.9 ± 1.9 c
	Medium	71.8 ± 1.7 a	66.45 ± 1.25 a	22.15 ± 1.65 b
	Fine	74.1 ± 0.9 b	109.5 ± 0.5 a	94.3 ± 5.5 ab
$\mathrm{N}{\mathrm{O}}_3^{-}$ (mg/L)	Unfiltered	27.5 ± 1.5 b	2.16 ± 0.01 a	0.7 ± 0.05 a	<49	<50	Conforms
	Large	39.55 ± 1.15 a	0.35 ± 0.01 b	0.64 ± 0.11 b
	Medium	36.25 ± 1.05 a	0.30 ± 0.65 b	0.54 ± 0.00 b
	Fine	0.88 ± 0.03 a	33.05± 1.05 b	0.21± 0.0 a
$\mathrm{N}{\mathrm{O}}_2^{-}$ (mg/L)	Unfiltered	2.9 ± 0.01 b	0.7 ± 0.01 a	0.02 ± 0.01 c	<3	<3	Conforms
	Large	0.02 ± 0.01 a	0.02 ± 0.01 a	0.02 ± 0.01 a
	Medium	0.02 ± 0.02 a	0.02 ± 0.02 a	0.02 ± 0.02 a
	Fine	0.02 ± 0.02 a	0.02 ± 0.02 a	0.02 ± 0.02 a
${\mathrm{C}\mathrm{I}}^{-}$ (mg/L)	Unfiltered	71.2 ± 1.7 b	52.75 ± 0.35 a	8.89 ± 0.39 c	≤300	<500	Conforms
	Large	65.35 ± 1.15 b	52 ± 1.1 a	8.91 ± 0.41 c
	Medium	64.55 ± 1.35 b	50.65 ± 1.25 a	9.66 ± 0.34 c
	Fine	11.35 ± 0.45 c	73.55 ± 1.05 b	58.4 ± 0.5 a
F (mg/L)	Unfiltered	0.31 ± 0.02 a	0.26 ± 0.01 a	0.15 ± 0.03 b	≤1.5	≤1.5	Conforms
	Large	0.02 ± 0.01 a	0.03 ± 0.02 a	0.16 ± 0.04 b
	Medium	0.36 ± 0.01 b	0.16 ± 0.01 a	0.28 ± 0.01 c
	Fine	0.04 ± 0.00 a	0.04 ± 0.00 a	0.03 ± 0.01 a
${\mathrm{S}\mathrm{O}}_4^{2-}$ (mg/L)	Unfiltered	27.5 ± 1.5 b	2.16 ± 0.01 a	0.7 ± 0.05 a	≤250	≤500	Conforms
	Large	296.5 ± 2.5 b	196.5 ± 1.5 a	11.95 ± 1.85 c
	Medium	60.8 ± 0.2 b	64.55 ± 0.45 a	13.35 ± 0.45 c
	Fine	170.50 ± 5.5 b	133 ± 0.5 a	130 ± 0.5 ab
${\mathrm{P}\mathrm{O}}_4^{3-}$ (mg/L)	Unfiltered	8.7 ± 0.34 b	3.88 ± 0.07 a	0.1 ± 0.00 c	-		Conforms
	Large	23.4 ± 0.4 b	10.65 ± 0.55 a	0.1 ± 0.00 c
	Medium	7.21 ± 0.22 b	4.75 ± 0.2 a	0.1 ± 0.00 c
	Fine	41.75 ± 0.05 b	24.75 ± 0.15 a	0.10 ± 1.25 c
${\mathrm{H}\mathrm{C}\mathrm{O}}_3^{-}$ (mg/L)	Unfiltered	171.5 ± 11.5 a	204 ± 4.03 a	92.25 ± 5.35 b			Conforms
	Large	90.36 ± 1.20 b	77.25± 1.65 a	110.5 ± 1.5 c
	Medium	193.5 ± 1.5 b	218.5 ± 1.5 a	113.5 ± 1.5 c
	Fine	104 ± 1.10 a	99.9 ± 0.1 a	32.15 ± 0.55 b
Na (mg/L)	Unfiltered	83.95 ± 1.35 b	62.19 ± 2.01 a	9.45 ± 0.53 c	≤200	≤200	Conforms
	Large	77.85 ± 0.65 b	61.2 ± 1.1 a	21.8 ± 1.3 c
	Medium	94.5 ± 5.2 a	94.55 ± 4.65 a	22.6 ± 0.5 b
	Fine	10.04 ± 0.05 c	84.4 ± 0.9 b	65.2 ± 0.7 a
K (mg/L)	Unfiltered	13.74 ± 1.34 a	9.28 ± 0.25 a	0.95 ± 0.03 b	≤50	3000	Conforms
	Large	15.95 ± 1.65 b	39.1 ± 0.2 a	2.26 ± 0.25 c
	Medium	9.55 ± 0.05 b	8.27 ± 0.01 a	2.13 ± 0.02 c
	Fine	14.52 ± 0.48 b	20.3 ± 0.5 a	3.25 ± 0.45 c
Ca (mg/L)	Unfiltered	39.55 ± 6.65 b	30.48 ± 1.42 a	20.3 ± 0.8 a	≤150	-	Conforms
	Large	41.5 ± 1.4 b	34.15 ± 1.35 a	25.05 ±0.75 a
	Medium	40.55 ± 0.65 b	46.8 ± 0.5 b	35.35 ± 0.45 a
	Fine	13.75 ± 0.85 c	32.3 ± 0.6 b	21.8 ± 0.5 a
Mg (mg/L)	Unfiltered	21.8 ± 1.1 ab	22.05 ± 1.45 a	13.8 ± 1 c	≤70	0.08	Conforms
	Large	21.25 ± 0.55 a	28.85 ± 1.55 b	17.15 ± 0.65 c
	Medium	26 ± 0.8 a	22.55 ± 0.35 a	18.11 ± 0.31 b
	Fine	21.2 ± 0.6 a	30.7 ± 0.7 b	23.35 ± 0.45 a
B (mg/L)	Unfiltered	0.07 ± 0.01 ab	0.01 ± 0.00 b	0.03 ± 0.01 a	<2400	2.4	Conforms
	Large	0.05 ± 0.01 b	0.01 ± 0.00 c	0.02 ± 0.01 a
	Medium	0.06 ± 0.01 b	0.01 ± 0.00 c	0.03 ± 0.01 a
	Fine	0.05 ± 0.02 b	0.01 ± 0.0 a	0.01 ± 0.01 a

Note: “Conforms” means that the parameter meets both SANS 241:2015 Limit and WHO Guidelines for drinking water. Unfiltered = unfiltered water (control), and Means in the same row indicated by different superscript letters are statistically significant (p ≤ 0.05).

). The mean Mg concentration ranged from 13.81 mg/L in summer to 30.7 mg/L in winter. The filtration with the M. oleifera seed particle size of 710 µm resulted in higher concentrations of Mg, suggesting that the addition of the M. oleifera seed powder increases the concentration of metals in treated water.

The results revealed several significant relationships among water quality parameters. The pH exhibited weak to moderate positive correlations with nitrate, chlorine, bicarbonate, and calcium, while showing negative correlations with sulfate, potassium, and magnesium. The EC displayed strong positive correlations with pH and iron. In summer, the EC exhibited strong positive correlations with sodium, chloride, and fluoride (Table 3).

Most chemical quality parameters, including pH, EC, $\mathrm{N}{\mathrm{O}}_3^{-},$ Cl⁻, F⁻, ${\mathrm{S}\mathrm{O}}_4^{2-},$ Na, K, Ca, Mg, and B, observed at the three sampling points were compliant with the set standards for drinking water in both seasons. However, notably high concentrations of nitrite were obtained from the Mamelodi site before filtration.

3.2. Microbiological Parameters

The analysis of unfiltered water samples revealed elevated counts of total coliforms, Salmonella spp., Shigella spp., and E. coli in both seasons (Figure 1 and Figure 2). Mamelodi had the highest amounts of Salmonella spp., E. coli, and total coliforms in summer (3.9 × 10⁴, 7, and 7.0 × 10⁴ CFU/100 mL, respectively) and winter (2.3 × 10⁴, 11, and 7.0 × 10⁴ CFU/100 mL, respectively).

After filtration with the M. oleifera seed-based gadget, the microbiological load and values of the physicochemical parameters decreased significantly (p < 0.05) in all test samples, except for Mamelodi’s Shigella and E. coli counts. Among the three sites, the finest particle size (710 µm) resulted in the lowest microbial counts in both seasons (Figure 4 and Figure 5).

The Moretele sampling point, which is downstream, had the highest colony forming units (CFUs) for E. coli in summer. Winter results were characterized by high levels of E. coli in the Mamelodi samples (10 CFU/100 mL in the control) and no CFUs in the Moretele and Tierpoort samples. The winter results also exhibited higher levels of total coliforms, Salmonella spp., and Shigella spp. in the Mamelodi samples compared to the other two sites.

Although the overall CFUs for Salmonella, Shigella, and E. coli were reduced after the treatment with the M. oleifera seed powder, the counts remained higher than the WHO limit of 0 CFU/100 mL for drinking water quality, indicating the presence of pathogens and the unsuitability of the water for direct consumption.

4. Discussion

The effectiveness of M. oleifera seed powder as a water purification agent was investigated across three sampling sites along the Pienaars River during both the summer and winter seasons. The results reveal complex interactions between the M. oleifera seed powder and various water quality parameters, highlighting both the potential and limitations of this natural purification method.

4.1. Physicochemical Parameters

The application of M. oleifera seed powder generally resulted in a decrease in pH, which was particularly evident in the winter samples. This effect was most pronounced with the large and fine particle sizes at the Mamelodi site, where pH values dropped below the WHO-recommended range (6.5–8.5). This pH reduction could be attributed to the release of organic acids from the seed powder during the filtration process [20]. While slight pH adjustments may be beneficial for certain water sources, care must be taken to ensure the final pH remains within acceptable limits for drinking water.

Interestingly, filtration with M. oleifera seed powder (710 μm particle size) led to increased concentrations of certain minerals, notably magnesium and, in some cases, calcium. This observation aligns with previous studies on the mineral composition of M. oleifera seeds [21]. For instance, in the winter samples, magnesium levels increased across all sites after the filtration with medium-sized particles. While this mineral enrichment could have potential nutritional benefits, it underscores the need for the careful optimization of the filtration process to maintain compliance with drinking water standards.

The effectiveness of the M. oleifera filtration varied between the summer and winter seasons. For instance, the impact on pH was more pronounced in winter samples. Additionally, the higher baseline levels of certain ions (e.g., sodium and chloride) in the winter samples at some sites influenced the overall effectiveness of the filtration process. These seasonal variations underscore the importance of adapting water treatment strategies to local and temporal conditions.

This study investigated three particle sizes of M. oleifera seed powder (large, medium, and fine), each showing different effects on water quality parameters. In general, finer particles tended to have a more pronounced effect on parameters, such as the EC and ion concentrations, possibly due to their larger surface area facilitating more interactions with water constituents [22]. However, the optimal particle size varied depending on the specific parameter and water source, suggesting that a one-size-fits-all approach may not be suitable for all applications.

4.2. Microbiological Parameters

The elevated microbial counts, particularly at the Mamelodi site, underscore the pressing issue of microbial contamination in the Pienaars River. Seasonal variations were evident, with summer samples generally showing higher levels of contamination. This aligns with the observations of Du Plessis et al. [23] and De Klerk et al. [24], who noted that seasonal fluctuations in microbial loads can be influenced by changes in temperature, flow rates, and human activities.

The higher levels of E. coli at the Moretele site during summer can be attributed to increased rainfall and runoff, which may carry fecal contaminants from surrounding areas into the river [25]. This seasonal pattern was particularly pronounced for the total coliforms and E. coli, while Salmonella and Shigella species showed more consistent levels across seasons, suggesting different sources or survival mechanisms for these pathogens.

The reduction in microbial loads after the filtration with M. oleifera seed powder is encouraging and consistent with previous research [12]. Across all microbial types and sampling sites, the filtration demonstrated a clear positive impact on water quality.

This study revealed that the finest particle size (710 μm) consistently showed the greatest efficacy in reducing microbial counts. This suggests that optimizing the particle size could be a key factor in improving the filtration process, possibly due to the increased surface area for microbial interaction. The effectiveness of the M. oleifera filtration varied between summer and winter, with a generally better performance observed in summer samples. This could be due to differences in initial microbial loads or changes in the water chemistry affecting the antimicrobial properties of M. oleifera compounds. The efficacy of the filtration varied across sampling sites, indicating that local water characteristics influence the performance of the M. oleifera seed powder. This highlights the need for site-specific optimization when implementing this technology. The filtration showed varying degrees of effectiveness against different microbial types. It was particularly effective against E. coli and total coliforms, while showing more variable results for Salmonella and Shigella species.

The elevated microbial counts, particularly at the Mamelodi site, highlight the significant challenge of the microbial contamination in the Pienaars River. The higher levels of E. coli at the Moretele site during summer can be attributed to increased rainfall and runoff, which may carry fecal contaminants from surrounding areas into the river [25].

4.3. Implications and Future Directions

This study demonstrates the potential of M. oleifera seed-based filtration as a sustainable and locally available water treatment option in developing countries. However, the results also highlight the need for further research and development to optimize the filtration process.

Future studies should focus on investigating combinations of M. oleifera filtration with other treatment methods to achieve potable water standards. Additionally, research should explore the impact of different M. oleifera seed preparation methods on the filtration efficacy. Conducting long-term studies is essential to assess the sustainability and consistency of M. oleifera-based water treatment systems. Furthermore, evaluating the potential environmental and health impacts of the increased mineral content in treated water is necessary for ensuring the safety and viability of this treatment approach.

4.4. Limitations of the Study

In the present study, although the filtration device significantly reduced microbial loads, the colony forming units (CFUs) for pathogens such as E. coli, Salmonella spp., and Shigella spp. remained above the WHO and SANS guidelines for drinking water. This indicates that the current prototype is not sufficient as a standalone solution for producing potable water and may require additional treatment steps. Additionally, this study tested three particle sizes of M. oleifera seed powder (710 µm, 1000 µm, and 2000 µm). While finer particles showed a better microbial reduction, they also led to increased mineral concentrations in some cases. Further research is needed to identify the optimal particle size that balances microbial removal with a minimal alteration of water chemistry. Furthermore, the effectiveness of the filtration device varied between summer and winter, likely due to differences in the initial water quality and microbial loads. This highlights the need for seasonal adjustments in the filtration process, which may limit its applicability in regions with highly variable water conditions. Lastly, this study was conducted at only three sampling points along the Pienaars River. While these sites provided valuable data, the results may not be fully representative of other water sources with different contamination profiles. Expanding the study to include more diverse water sources would enhance the generalizability of the findings.

5. Conclusions

This study evaluated the effectiveness of a M. oleifera-based water purification device combined with activated charcoal and cotton wool for treating water from the Pienaars River in South Africa. The device significantly reduced microbial loads, including the total coliforms, E. coli, Salmonella spp., and Shigella spp., but filtered samples still exceeded the WHO and SANS drinking water guidelines, indicating the need for additional treatment steps. Finer M. oleifera seed particles (710 µm) were more effective in microbial reduction but increased mineral concentrations, highlighting the importance of optimizing the particle size. Seasonal variations influenced the device’s performance, with better results in summer due to differences in water quality. Physicochemical properties, such as pH and ion concentrations, generally met drinking water standards, though slight pH reductions and mineral increases were observed. The site-specific performance varied, with the Mamelodi site showing the highest contamination and least consistent microbial reduction. This study introduces a novel, low-cost, and sustainable filtration approach, but further refinement is needed to address limitations, such as incomplete microbial removal, seasonal variability, and scalability. Future research should focus on optimizing the system, integrating complementary treatments, and conducting long-term field trials to ensure practicality and safety. This work provides a foundation for developing scalable, natural water treatment solutions to combat waterborne diseases in resource-limited settings.