Determining the difference in contaminant levels before and after filtration is an essential step in evaluating the performance of any water treatment system, including activated sand-based filters. The effectiveness of such filters is typically assessed by measuring the reduction in contaminants such as turbidity, microbial pathogens, and chemical pollutants. These analyses help to quantify the filter’s removal efficiency and establish whether it meets the required water quality standards. Below, we explore the factors influencing the removal of contaminants and the methodologies used to assess these differences.

Turbidity, which is the cloudiness or haziness of a liquid caused by suspended particles such as silt, clay, and organic matter, is one of the most common indicators of water quality. High turbidity levels can reduce the effectiveness of disinfection methods (like chlorine or UV light) and impede the performance of filtration systems by clogging filter media [5],[6]. Therefore, measuring the reduction in turbidity before and after filtration is critical for evaluating the efficacy of activated sand-based filters.

Activated sand filters are particularly effective at removing suspended solids, and studies have shown that these filters can achieve significant reductions in turbidity. For instance, research by Zhang, [7] demonstrated that activated sand filters can reduce turbidity levels by up to 90%, depending on factors such as the filter design, grain size, flow rate, and the characteristics of the water being filtered. A similar study by [8] confirmed that turbidity reduction is highly influenced by the media’s porosity and surface area, which allows for better particle trapping and sedimentation.

The efficiency of turbidity removal also depends on the nature of the suspended particles. Coarser particles may be removed more effectively by filtration, while finer particles or colloids, which tend to stay suspended in water for longer periods, may be more challenging to remove. Therefore, studies typically consider not just the overall turbidity reduction but also the types of particles removed during filtration [8]. In some cases, pre-treatment steps like coagulation or flocculation may be used in combination with activated sand filtration to enhance turbidity removal.

While activated sand filters are known for their ability to remove suspended solids and microorganisms, their ability to remove dissolved chemical contaminants, such as heavy metals, pesticides, and organic chemicals, is less straightforward and depends on the specific contaminants involved. Activated sand filtration works primarily through physical adsorption and surface interactions, and the degree to which it can adsorb chemical pollutants is influenced by the surface area and porosity of the sand, as well as the chemical characteristics of the contaminants.

Studies have shown that activated sand can be effective in removing certain chemical pollutants, particularly heavy metals like lead, arsenic, and mercury. For example, a study by Liu et al. [6] found that activated sand filters were able to significantly reduce concentrations of lead and arsenic in contaminated water. The removal process is primarily based on adsorption, where the chemical contaminants are attracted to and held on the surface of the sand particles. The efficiency of heavy metal removal depends on several factors, including the pH of the water, the ionic strength, and the presence of other competing ions.

For pesticides and herbicides, the adsorption capacity of activated sand may be less effective, especially for hydrophobic compounds. The objective of this research is to design and evaluate the physicochemical performance of an activated-sand-based Point of Use water treatment system for household applications in rural and informal settlements in Kenya.

In this research, a 2^k factorial design with five control factors (sand grain size, flow rate, residence time and concentration, and loading rate) was employed to optimize the activated sand filter performance in the removal of physicochemical contaminants. The experimental setup consisted of two cartridge configurations. The factors and levels are shown in Table 1.

The activated carbon was sourced from Panthera-activated Carbon^TM, Silver activated sand was sourced from Nikken Corporation (Japan), with commercial name Clinca205 and the zeolite was sourced from Rift African gems Kenya. Leaching Test of Silver in Activated Sand: A leaching test will be performed on the sample of activated sand according to Notification No. 45(2000), “Test for materials of mechanical equipment and materials” based on Ministerial Ordinance No 15(2000) Japan, “Technical standards for water Utilities”.

Sample collection was done according to WHO guideline procedures [2] in order to avoid contamination and to ensure accurate results.

Sterilization of sample bottles: Plastic bottles of at least 200 mL capacity with plastic screw caps were cleaned thoroughly and then rinsed with distilled water and sterilized by autoclaving at 121 °C for 15 min. Other materials used in the bacteriological analysis, such as dilution bottles, small cylinders, beakers, tubes, buffer solutions, and pipettes, were also sterilized. Water was sampled from the two selected rivers and the baseline water characteristics of the rivers were performed.

Sampling procedures: Water samples for physicochemical analysis were collected in sterile bottles, which were kept unopened until the time of filling, usually less than 24 h [2]. The cork was removed, and the bottle was held by the other hand around the base of the bottle. The researcher ensured that the bottle was not rinsed with the sample during collection and that the bottle was not completely filled to allow for shaking prior to analysis. Surface waters such as rivers were sampled away from the banks as much as possible.

Sampling location: Two sampling points were selected in reference to the objective of the research to provide a point-of-use device for rural and urban informal settlements.

The sampling points are:

1. Kesses River as shown in Figure 3, is a representative of a rural settlement.

   Location: Kesses Sub County, Uasin Gishu County, Kenya;

   Main feature: Kesses Dam, a significant water reservoir;

   Water source: Primarily fed by the Tarakwa and Nderugut rivers;

   Terrain: High altitude plateau within the Rift Valley;

   Usage: Water supply for domestic use, irrigation, and recreational activities;

2. Sosiani River, as shown in Figure 4, represents the urban settlements.

   Source: Kaptagat Forest

   Major Town: Eldoret

   Drainage Basin: Nzoia River basin, which flows into Lake Victoria

   Catchment Area: Primarily agricultural land use

Figure 3.

Kesses river site coordinates and sampling points

Figure 4.

Sosiani river sampling site coordinates

Figure 5 shows the basic design of the point-of-use water treatment prototype developed by the researchers, whereas Figure 6 represents the filter cartridge configurations, which allow for the replacement of spent cartridge media.

Figure 5.

Basic design of activated sand based device

Figure 6.

Constructed filter cartridge configurations

Design and Construction of prototype: The fixed bed was designed and constructed to the specifications below:

• Total Bed Height = 120 cm

• Total bed Width = 80 cm

The container volume is calculated according to Eq.(1):

The flow rate was measured as 29.07 cm³/s. The Empty Bed Contact Time (EBCT) is calculated according to Eq.(2).

The filter cartridges were packed as shown in Table 2.

X-ray Fluorescence (XRF) analysis is an elemental analysis technique used to determine the composition of materials based on their unique fluorescence properties when exposed to X-rays. In the context of assessing filter media for water treatment, XRF can provide crucial information regarding the elemental makeup and potential contaminants present in the media.

The results obtained from XRF analysis of the filter materials are shown in 3 reveals that all the filter materials, contain Silica (SiO₂), Alumina (AI₂O₃), Ferric Oxide (Fe₂O₃), Calcium Oxide, Phosphorus pentoxide(P₂O₅), Chloride, Potassium Oxide and other oxides. These results compare to those of [11] and [12].

Activated sand and river sand show presence of MgO, which has remarkable adsorption capacity. Their unique structural properties and large surface area enable exceptional pollutant removal, including heavy metals, organic compounds, and dyes, from wastewater [13]. Additionally, synthesis and functionalization techniques show promise in enhancing adsorption performance and selectivity, offering tailored solutions to specific treatment challenges. Tailoring the surface chemistry and morphology of MgO nanoparticles allows for improved pollutant removal efficiency and paves the way for tailored solutions to specific wastewater treatment challenges [13].

Fourier Transform Infrared Spectroscopy (FTIR) is a valuable analytical technique used to examine the chemical composition and functional groups present in filter media for water treatment. By analyzing materials like activated carbon, sand, and zeolite, FTIR provides insights into their properties, adsorption characteristics, and interactions with contaminants.

To prepare samples for analysis, filter media are typically dried and ground into a fine powder, enhancing the penetration of infrared light for accurate results. The resulting spectra display peaks that correlate to specific molecular vibrations, allowing researchers to identify functional groups such as hydroxyl, carbonyl, and amine. This identification can indicate the presence of pollutants or the condition of the filter media. The recorded spectra for activated charcoal is shown in Figure 7.

Figure 7.

FTIR spectra of activated charcoal

The infrared transmission peaks (Figure 7) recorded at wavelengths 3774.97 cm^-1 and 3693.01 cm⁻¹ exhibited a wide and broad transmission band. Here, the presence of a broad transmission band was attributed to the O-H stretching mode of non-bonded hydrogen hydroxy groups attached to the structure or surface-adsorbed moisture, this is characteristic of phenols [14]. The absorption peaks at 3515.60 cm^-1 is characteristic of Dimeric OH stretch, whereas 3441.35 cm^-1 allude to a non-polymeric OH stretch. The bands at 1057.76 cm^-1 and 1106.94 cm^-1 be assigned to phosphate ions [15] which aid in precipitate formation [16], whereas the peaks at 725.10 cm^-1, 766.57 cm^-1 and 865.88 cm^-1 can be assigned to the Aryl group [14]. There is also the presence of amino and organo-halogen groups. With the major bonds being carbon or oxygen based, we can confirm that carbon and oxygen are the main elements in activated charcoal as attributed to in the elemental analysis in 3, confirming its organic nature.

The spectral analysis of activated sand is shown in Figure 8. The infrared transmission peaks recorded at the wavelengths 3603.34 cm^-1, 3694.94 cm^-1, 3730.62 cm^-1and 3820.29 cm⁻¹ exhibited strong narrow transmission bands. Here, the presence of a narrow transmission band was attributed to the O-H stretching mode of non-bonded hydrogen hydroxy groups, this is characteristic of phenols [14].

Figure 8.

FTIR spectra of activated sand

The absorption peaks at 3514.63 cm^-1, 3440.39 cm^-1, 3319.86 cm^-1 and 3200.29 cm^-1, are attributed to H-bonded with an OH stretch hydroxy group. The strong peaks at 2876.31 cm^-1 and 2823.28 cm^-1 are characteristic of Carboxylic acid [17] groups. The peaks at 1526.36 cm^‑1 and 1570.74 cm^-1 are indicated for aromatic ring stretch of C=C-C bond typical of methylene. The band peaks of 1107.90 cm^-1 and 1180.22 cm^-1 are assigned to phosphate ions [15] which aid in precipitate formation [16], whereas the peaks at 984.48 cm^-1 and 1057.76 cm^-1 is indicative of silicate ions [18]. Transmission at 1249.65 cm^-1 is indicative of aromatic phosphates group. There is also the presence of alkenes, amino, aryl and organo-halogen groups.

The existence of phosphate and silicate ions in the activated sand, we can confirm that silica, aluminum oxide, magnesium oxide and Potassium dioxide are the main elements in activated sand as attributed to in the elemental analysis in Table 3, thus confirming its inorganic nature.

The sand particle analysis is listed in Table 4. The uniformity factor obtained is 1.83 which is less than the recommended factor for slow sand filtration which is 2 [19]. The sand particles were categorized into two regimes: fine sand particles with mesh size of less than 500 µm and Coarse sand between 500 and 1500 µm.

Figure 9

Sand particle analysis

Turbidity reduction in the cartridge filter

The experimental data on turbidity reduction (Figure 10), shows an average turbidity reduction of 85.3 % and 86.4 % on cartridges 1 and 2 on Kesses river water point 1, 91.2 % and 85.9 % on Kesses river water point 2, 80 % and 75.1 % on Sosiani River water and 88 % and 83.2 % on Boundary River water. Overall Cartridge 1 performed better than cartridge 2 in all the sampled waters, this is attributed to the higher volume of silica in the cartridge. These results are supported by those of Fitriani et al. [20] who managed removal efficiencies of up to 94.34 % and Das et al. who reported average removal efficiency of 79.96 % [21].

Figure 10.

Turbidity removal efficiency

Turbidity was caused by suspended and insoluble species which were removed by both filters efficiently to less than 5 NTU. However, it is noteworthy that any turbidity above 20 NTU, the sample was first pretreated in a bio-charcoal sand system. The main mechanism of turbidity removal by the filter cartridges was surface straining by sand particles as well as other processes such as interception, sedimentation and adsorption [22],[23].

Removal of Total Iron in the cartridge filter:

The average iron removal rate was 77 % and 70.5 % in cartridges 1 and 2 respectively (see Figure 11) while running on Kesses water, and 79.2 % and 69.8 % on Sosiani river water. These results are supported by those of Eniola and Sizirici [24], who reported 60.18 and 78.94 % reduction in iron. Other studies have reported 95.28 % reduction in iron [25], 97.9-99.9 % [26], Fitriani et al. who reported removal rates of 45.4-50 % [20] and Endang reported 96.39 % effectiveness.

Figure 11.

Total iron removal efficiency

This research agrees with the observations of Hidayatullah, that a filter with a silica sand height of 75 cm can achieve an iron parameter removal of up to 72.96 % [27]. The silica sand in the Slow Sand Filter functions as an adsorbent and oxidizing agent for water pollutants. Silica sand has a (-) charge, so that it can attract positively charged particles such as cations from the trapped iron molecules [27].

The efficiency of iron removal was caused by the filter media that can work with ion exchange processes to reduce iron in water. Ion exchange process occurs because of two things, namely there are natural particles such as shells and there are synthetic particles, i.e., organic resins. In research, the ion exchange process occurs because there are natural particles. According to [20], the absorption mechanism of the Fe²⁺ metal ions can occur if it interacts with CaCO₃. This was supported because the ionic radii of Fe²⁺ and Ca²⁺ atoms are similar so that they were able to replace the position of the ions. The cycles that occur in the process are as follows:

CaCO3→CO32−+Ca2+(3) CaCO3(s)+Fe2++CO32−→FeCO3+Ca2+(4) Fe2++CO32−→FeCO3(s)(5)

The insoluble Fe metal will be retained in the pores of the filter media when water passes through the media, so that it can reduce the Fe content in the water as it exits the outlet [27]. In addition, iron removal was influenced by the transport mechanism so that an adsorption process occurs in the filtration unit. The adsorption process occurs because there is a dif­fusion process so that the adsorbate is adsorbed. In addition, there is an attachment mechanism caused by the van der Waals force [28]. Silica sand have a negative (-) charge, so they can attract positively (+) charged particles such as cations from iron molecules [20].

Variation of Calcium in the cartridge filter

The graphs in Figure 12 show the variation of calcium between the influent waters and effluent. There was an average increase in Calcium of 7.7 % and 6.9 % in cartridges 1 and 2 while running on Kesses water and 57.4 % and 43.4 % respectively on Sosiani river water. These figures are supported by the study by Sabogal et al. who reported increased levels of Calcium caused by leaching of the elements in the filtration media into the effluent water [29].

Figure 12.

Calcium variation

Variation of Alkalinity in the cartridge filter

The graphs in Figure 13 show that there was an average increase in alkalinity of 7.7 % and 6.9 % in cartridge 1 and 2 respectively running on Kesses water, and 26.1 % and 35.9 % in Sosiani river water. These results are corroborated by those of Ahmed et al. who showed increase in alkalinity for all the filtrates [30].

Figure 13.

Variation of alkalinity

Removal of Chlorides in the cartridge filter

The graphs in Figure 14 show the removal efficiency of Chloride by the Cartridge filters in different water sources. The average removal efficiency is 7.8 % and 5.2 % in cartridges 1 and 2 while running on Kesses water and 38.9 % and 34.3 % when running on Sosiani river water. Overall Cartridge 1 has better removal efficiency. These results are supported by a study by Ahmed, et al who reported a removal efficiency of 13 % [30].

Figure 14.

Removal of chlorides

Removal of Fluorides in the cartridge filter

Fluoride removal efficiency was 6.5 % and 5.4 % for cartridge 1 and 2 respectively, while running Kesses river water and 7.7 % and 2.8 % while running Sosiani river water as shown in Figure 15. Overall Cartridge 1 was more efficient at Fluoride removal. These figures are comparatively lower with those of Madhusha et al who in their study reported efficiencies of up to 45 % [31].

Figure 15.

Removal of fluorides

The presence of carboxyl and hydroxyl groups on the surface of the activated carbon influences the material's adsorption characteristics. The location and number of functional groups on the pore structure of activated carbon play a significant role in the adsorption process [32]. Additionally, activated carbon can adsorb anions such as fluoride ions depending on the surface chemistry of activated carbon along with its well-developed porosity, in particular, the presence of OH and phosphate groups leads to such favorable interactions with fluoride.

Variation of Sulphates in the cartridge filter

There was an average increase of 59.5 % and 77.7 % in sulphates for cartridge 1 and 2 respectively running on Kesses water and 19.5 % and 21.3 % on Sosiani river as shown in Figure 16. The increase in sulphate is due to formation of sodium, potassium and magnesium sulphates as the water is filtered through the filter media. In a Study be Fernando et al the average sulfate before modernization was 3.02 mg/L and after the changes it was 17.85 mg/L, almost 500 % increase, due to aluminum sulfate, which leaves a residual sulfate in the water [33]. A similar study by Ohmaid, also showed increased levels of sulphates in product water after filtration [34] This confirms that there are some impurities like sulphates, that cannot be effectively removed by the developed system [35].

Figure 16.

Sulphate removal

Removal of Phosphates in the cartridge filter: The average Phosphate removal was 39.2 % and 34 % in cartridge 1 and 2 on Kesses river water and 12.2 % and 9.4 % on Sosiani river water as shown in Figure 17. The performance of Cartridge 1 was better than that of cartridge 2. These results are supported by the research by De Rozari who obtained removal efficiencies of between 35.9 % and 44.2 % [36].

Figure 17.

Phosphate removal

The presence of Magnesium in the form of Magnesium oxide in the activated sand significantly promotes the adsorption capacity of phosphates due to the strong di-valent cation bridging between Mg and P [37]. The formation of magnesium and calcium phosphates on activated carbon surfaces is considered the primary mechanism of Phosphate removal from water [38].

There are no conflicts to declare.