Treatment of domestic greywater using fly ash as an adsorbent incorporated in a slow sand filtration system

Recent years have seen a sharp increase in population growth and industrialization globally. This coupled with poor management of the available water resource by overconsumption and pollution has led to water resource pollution and water scarcity (Nawwash & Ghunmi 2009). The practice of greywater recycling, if given the necessary attention, can help lessen the over-dependence on freshwater resources and decrease the pollution that results from the discharge of untreated greywater into freshwater sources (Oteng-Peprah et al. 2018). Greywater, also known as sullage, is wastewater from bathrooms, laundries, kitchens, etc. that does not contain fecal matter (WHO 2006; Siwila 2021). Implementation of greywater systems, at domestic, institutional, or even commercial scale, would mean the use of greywater at decentralized sites for landscape irrigation and toilet flushing among other non-potable uses. This lessens the amount of potable water distributed to these sites and consequently the measure of wastewater produced, conveyed, and treated at wastewater treatment plants. Explicitly, greywater reuse conserves resources – water, energy, and money (Giz 2014).

Application of greywater treatment systems is of particular importance in assisting developing countries in addressing Sustainable Development Goals (SDG) goal number 6: Guarantee accessibility and sustainable management of water and sanitation for all. More precisely, on objective 6.3, focussed on cutting in half the fraction of untreated wastewater and significantly growing recycling and safe re-use globally (UN 2015).

According to Oteng-Peprah et al. (2018), reclamation of greywater is a more reliable way of ensuring sustainable water supply for non-potable use than methods like rainwater harvesting which are greatly subjected to the hydrological conditions of an area. The composition of greywater is mostly dependent on lifestyle and type and choice of chemicals used for laundry, cleaning and bathing (Oteng-Peprah et al. 2018). Furthermore, the quantity of greywater generated also varies greatly according to household dynamics, and is influenced by factors like the number of occupants, their age distribution, their lifestyle characteristics, water-usage-patterns, the cost of water, and the prevailing climate.

For an average household, the daily total volume of grey water definitively produced is 356 litres, which translates to about 60% of the household-generated wastewater (WHO 2006). More than 85% of generated greywater is low-load greywater from showers and wash basins that has moderately great water quality boundaries and can be reused following simple treatment, without danger or harm. The observed high rate of greywater generation in homes, schools, and universities indicates a great potential for treatment and reclamation (Rodríguez et al. 2020).

Fly ash is an industrial by-product from thermal power plants, coal processing plants, and steel industries, which is majorly composed of fine particles. Its disposal causes major environmental problems, which has led to research on utilization of this waste product for various purposes (Saravanakumar et al. 2019). Fly ash is spherical in shape, and has high carbon content and a specific surface area between 2,000 and 6,000 cm²/g (Ganapathy et al. 2018). These characteristics increase the adsorption capacity of fly ash (Ganapathy et al. 2018), potentially making it a good candidate for greywater treatment.

Since adsorption can enhance contaminant removal performance by slow sand filtration (SSF) systems (Baruth 2005; Binnie & Kimber 2013), fly ash was incorporated in the designed SSF system to augment adsorption capacity. Moreover, fly ash can also serve as a filter media substitute or additional treatment step in SSF systems for enhanced removal of turbidity, colour, phenolic and other organic compounds, certain pesticides, and other micro-pollutants (Kawamura 2000; McAllister 2005; Ahmaruzzaman 2008). In addition, fly ash was incorporated because it has been indicated by others (Siabi 2003; Mihelcic et al. 2009) that adsorbent materials such as fly ash and activated carbon can enhance removal of inorganic and organic dissolved solids from water.

Reclaimed domestic greywater has various potable and non-potable uses, examples of which include flushing toilets, cleaning cars, cleaning driveways, cleaning paved yards, and watering lawns (Oteng-Peprah et al. 2018). The interest in the use of recycled greywater in domestic gardens and small-scale irrigation schemes owes to the fact that it generally contains lower concentration of organic matter and pathogens compared to mixed wastewater (Ochoa et al. 2015). However, there are some concerns with regard to safety of reclaimed greywater for irrigation purposes. The key issue being the potential for damaging effects of poor water quality on soil, plants, and humans (Khalaphallah 2012). Therefore, this study attempted to produce treated greywater that has minimal or negligible impact on soils, plants, and humans.

The study was focussed on assessing if fly ash is a good material for greywater treatment for non-potable applications. The main objective of the study was to investigate the use of fly ash in the treatment of domestic greywater when incorporated in a SSF system. The specific objectives included: (i) to characterize and evaluate the fly ash for use as a greywater treatment media regarding its physical and chemical characteristics, (ii) to design and construct a small-scale SSF system incorporating fly ash as an adsorbent, and (iii) to compare treated greywater effluent with the Food and Agricultural Organisation (FAO) guidelines for irrigation water quality.

Class ‘F’ fly ash was collected from Mamba coal mine Plc in Zambia and was then analysed for physical and chemical properties. Elemental analysis was also done on the fly ash and the contained elements are presented in Table 1. The collected fly ash was further subjected to oven drying for 24 hours at a temperature of 110 °C and then sifted through a series of sieves before filter bed preparation using BS 1377-2:1990 (Soils for civil engineering purposes), which outlines the test procedure for particle size distribution. The minimum size of fly ash was restricted to 0.150 mm to avoid clogging the designed SSF system. A face mask was always worn when handling the fly ash in the laboratory as a precautionary measure against any respiratory issues. This is because fly ash contains a lot of fines. In addition, extra care was taken by using gloves and a spatula when handling the fly ash.

The river sand used in the designed system was obtained from Kafue river while the gravel was obtained from the Copperbelt University's new hostels construction site. The river sand and gravel were air dried for 24 hrs, then washed, and oven-dried for 24 hrs at 110 °C. After oven drying, sieve analysis was conducted on both materials according to BS 1377-1: 1990.

The low-load greywater samples used to test the designed system were collected from the student hostels at the Copperbelt University's main campus in Zambia. The collection of the sample was done 3 times per week, for 4 weeks. During each sampling, the collected greywater samples underwent plain pre-sedimentation for 2 hours before filtering the decanted raw greywater through each configuration of the designed SSF system.

Table 1 gives results for the chemical composition of the fly ash as tested at the Copperbelt University's school of mines laboratory using atomic absorption spectroscopy (AAS). The elements tested for were calcium (Ca), iron (Fe), silicon (Si), magnesium (Mg), aluminium (Al), sodium (Na), and potassium (K). Elements in fly ash, like most earth minerals, do not exist as individual elements but as compounds of oxygen. It is also worth noting that metal oxides (mostly groups 1 and 2) generally react with water to form ‘basic’ solutions (Zumdahl 2018). Based on the results obtained from the elemental analysis, Na₂O, MgO, and CaO fall into that category. From the listed compounds (Table 1), Na₂O forms NaOH with water, which is a strong base (Zumdahl 2018).

The analyses on the fly ash for some physical properties yielded the results shown in Table 2.

It was found that the fly ash used for the study had a low effective size (ES) and uniformity coefficient (UC). The size and uniformity of filter media are specified by the ES and UC (Steel & McGhee 1979; Siwila 2021). The ES (D₁₀) is the sieve size in millimeters that allows 10% of the grains of a granular media by weight to pass. The UC is the ratio between the sieve size that allows 60% of the grains by weight to pass and the effective size (Siwila 2021). According to (Siwila 2021), a UC close to 1 means a small range of grain sizes within the granular media. Consequently, few smaller grains will fit between the larger grains, and the granular media will be less tightly packed. This subsequently leads to a higher flow rate (CAWST 2010; Siwila 2021). On the other hand, a higher UC means a larger range of grain sizes within the granular media. As a result, the smaller grains will fill in the gaps between the bigger grains, resulting in more tightly packed granular media (CAWST 2010). This eventually leads to a lower flow rate CAWST (2010). This is why the UC in SSFs is usually greater than that of the rapid sand filters (RSFs) (CAWST 2010). Therefore, from literature, it was deduced that once incorporated in a filtration system, the low ES and UC of the fly ash would yield a low filtration rate (Siwila 2021), which was much desired in this case. Hence, low filtration rates are beneficial when high contact time between the water and an adsorbent material (e.g. fly ash) is required, and this enhances their contaminant removal efficiency (Siwila 2021).

The specific gravity of the fly ash was found to be similar to that of sand, i.e., 2.55–2.65, (Steel & McGhee 1979). On the other hand, the fly ash recorded a high pH of about 9.51, which signified that it had some basic compounds present in it as can be confirmed by Table 1 results. To test for pH, 20 g of fly ash was measured and put in a 100 mL beaker. Then 20 mL of distilled water was added to the sample and stirred for 20 minutes. The pH measurement was then done three (3) times and the average pH value was recorded as given in Table 1.

For the river sand and gravel, wet sieve analysis was carried out. It should be noted to the reader that dry sieving was done on the fly ash and corresponds to Table 2 results. In both cases the analyses were done in conformity with BS 1377-2. Table 3 gives a synopsis of the results acquired from the sieve analysis and particle size distribution curves for the filter media.

The ES and UC of the sand used in the study were within the range recommended for slow sand filters, i.e., 0.1–0.35 and 1.5–3, respectively (CAWST 2010; Siwila 2021). It is worth noting that the Low ES and high UC of the river sand also contributed to ensuring the desired slow rate of filtration. This is because the river sand had a high UC. According to literature (CAWST 2010; Siwila 2021) this resulted in a larger range of grain sizes within the granular media. Therefore, the smaller grains filled in the gaps between the larger grains, resulting in a more tightly packed granular media (CAWST 2010). This in turn led to the desired reduced filtration rate.

The small-scale SSF system used in this study was designed based on the biosand filter design, construction and installation manual presented by CAWST (2010). Thus, both configurations were scaled = down versions of the SSF model by CAWST (2010). The total filter material depths were kept between 20 and 40 cm, keeping in mind that affordable systems of this type are normally housed in low-cost buckets that are usually 20 to 40 cm high. The designed system consisted of two configurations. Configuration A, the control experiment, consisted of sand (192 mm depth) and gravel (45 mm depth) only (Figure 1), while configuration B, the main experiment, consisted of sand (137 mm depth), fly ash (30 mm depth), and gravel (45 mm depth) (Figure 1). The design parameters for each configuration are as given in Table 4.

The resulting filtration rates were 0.25 m/h for configuration A and 0.072 m/h for configuration B. The difference in filtration rates between the two configurations could be attributed to the presence of fly ash in configuration B which was absent in configuration A. The Class ‘F’ fly ash used was much finer than the river sand (Table 3) resulting in very low filtration rates.

Each system was charged with at least 4.0 litres of water per day, and a minimum of 500 ml samples were collected for testing from the treated effluent of each configuration (Table 5). The flow rate was measured using a 1 L jar and stopwatch at the fastest flow point in the filter, as this determines possible detachment of microbes and particles attached to filter media, and their subsequent flushing into the filtered water (NE-WTTAC 2014). It was ensured that clean river sand was used in the filters to ensure purity, as recommended by CAWST (2010). This was further complemented by sizing the fine sand according to recommendations by CAWST (2010) and Siwila (2021), with ES of 0.10 to 0.20 mm and UC of 1.5 to 2.5, giving a more tightly packed sand layer and subsequently a lower flow rate to yield better pollutant removals. This is expected to enhance pollutant removals by mechanical trapping and adsorption as well as other removal mechanisms, which occur within the sand body. The filtration rates for both configurations were maintained between the recommended 0.05 to 0.3 m/h (CAWST 2010; Siwila 2021) (see Table 4) to ensure adequate empty-bed contact time (EBCT).

During each sampling, both the raw and treated greywater samples were collected and tested for turbidity, colour, chloride, dissolved oxygen (DO), pH, total dissolved solids (TDS), total suspended solids (TSS), and electrical conductivity (EC). The TDS and EC were measured using the Vuro 651 TDS/conductivity meter. The TSS and colour were measured using a bench scale Ultraviolet-visible (UV-Vis) spectrophotometer. Turbidity was measured using the Hach 2100p turbidity meter while the pH parameter was measured using a Eutech Benchtop pH 700 meter. To test for DO and chloride concentrations, titration methods (APHA 2017) were used.

All the analyses for each tested water quality parameter were done in accordance with Standard Methods for water and wastewater analysis (APHA 2017) and all instruments were calibrated using fresh calibration solutions during each measurement.

The results (Figure 2 and Table 6) show that both system configurations were significantly effective in turbidity removal from the greywater. The average turbidity removal by configuration A was 79% while average turbidity removal by configuration B was 95%. The recorded average turbidity removal by configuration B was much higher than that of configuration A, probably due to presence of fly ash in configuration B which enhanced its adsorption capacity (Siabi 2003; Mihelcic et al. 2009). This indicated the need for adding an adsorption material like fly ash for substantial removal of turbidity from greywater by SSFs.

Figure 2 illustrates a summary of the observed turbidity levels for the raw and treated greywater over the study period and the line representing the Zambia Environmental Management Agency (ZEMA) effluent discharge limit (15 NTU). It can be clearly seen that the filtered water quality using configuration B consistently met the ZEMA wastewater discharge limit for turbidity but this was not so with configuration A. Hence the addition of fly ash to the designed SSF system for enhanced contaminant removal was important to improve the quality of the treated greywater.

Substantial TSS removal from the domestic greywater was recorded by both system configurations (Figure 3 and Table 6). Just like for turbidity, configuration B was much better than configuration A for TSS removal, but particle removal was significant by both. Average TSS removal was 98% for configuration B, while average turbidity removal for configuration A was 90%. The higher TSS removal by configuration B can also be attributed to the presence of the fly ash, which substantially increased the filter's adsorption capacity (Kawamura 2000; McAllister 2005; Ahmaruzzaman 2008). In addition, the fly ash mostly contained fine grains which subsequently caused the suspended particles to be substantially trapped among the fly ash media grains. However, both system configurations consistently met the ZEMA TSS discharge limit and FAO guideline (>100 mg/L) rendering the water aesthetically appealing for the desired purpose of lawn grass watering. Since both configurations can produce clear water, with significant particle removals (Figures 2 and 3, Table 6) they can be handy for affordable small-scale domestic greywater treatment for non-potable uses like irrigation.

The colour levels in the raw greywater were variable, the highest being 2,222 PtCo (Figure 4). The results (Figure 4 and Table 6) show that both system configurations substantially removed colour from the domestic greywater used. The average colour removal by configuration A was 93% while the average colour removal by configuration B was 98%. Therefore, the recorded average colour removal by configuration B was significantly much higher than that of configuration A. It should be noted that the ZEMA effluent discharge limit for colour is 20 PtCo with all configuration B treated effluents meeting this standard throughout the study (Figure 4).

The better performance in terms of colour removals by configuration B can be attributed to the presence of fly ash which most probably enhanced its adsorption capacity. According to Ganapathy et al. (2018), Fly ash is spherical in shape, has high carbon content and specific surface area between 2,000 and 6,000 cm²/g. These characteristics increase the adsorption capacity of fly ash (Ganapathy et al. 2018). Therefore, the importance of incorporating fly ash as an adsorbent material is hereby indicated again. Moreover, fly ash can also serve as a filter media substitute or additional treatment step in SSF systems for enhanced removal of turbidity (Figure 2), colour (Figure 4 and Table 6), certain organic compounds, and other pollutants (Kawamura 2000; McAllister 2005; Ahmaruzzaman 2008).

Chloride, which is a common constituent in water, is highly soluble, and once in solution tends to accumulate (Department of Water Affairs & Forestry 1996). It is one of the parameters outlined in the FAO guideline for irrigation water; it brings about specific ion toxicity, which can affect sensitive crops (Zaman et al. 2018). Upon comparison of the treated effluent values with FAO guidelines for irrigation water (Misstear et al. 2017), it was found that treated greywater effluent consistently contained chloride levels within recommended levels of <350 mg/L (Figure 5) for both configurations. Thus, the treated greywater from both configurations contained safe chloride levels for watering of plants (Misstear et al. 2017).

The lowest observed DO concentration was recorded on day 6 for raw domestic greywater obtained from the hand basin (Figure 6). Surprisingly, the treated effluent showed increased levels of DO by both configurations after filtration, with configuration B showing higher increase on each sampling day. Configuration A showed an average increase in DO concentration of 27% whereas that for DO increase configuration B was 33%. The increase in DO in the effluent of both configurations could be somewhat attributed to microbial growth in the system (Yildiz 2012) and the mode of greywater feeding, which might have induced some form of aeration as the water was flowing onto the filter bed surfaces (Figure 1). Both configurations consistently met the ZEMA environmental discharge limit (Figure 6) for ensuring a minimum DO level of 5 mg/L in the treated effluent (Siwila 2021).

Both treated effluents (i.e., from configuration A and configuration B) were well within the FAO recommended irrigation water quality with regards to EC and TDS levels. Thus, all recorded EC values were way below 3,000 μS/cm (Figure 7) and all recorded TDS values were way below 450 mg/L (Figure 8) as recommended by FAO for irrigation purposes (Misstear et al. 2017). Therefore the treated greywater consistently met the EC and TDS limits for irrigation purposes (Figures 7 and 8) making the water fit for the intended use of lawn watering.

It is however worthy noting that the effluent from configuration B, which contained fly ash, in some cases, recorded slightly higher EC values than the raw greywater values (Figure 7) and correspondingly slightly higher values of TDS (Figure 8). Surprisingly and as an exception, it was noted that for raw greywater samples with EC and TDS values above FAO recommended values (Figures 7 and 8), configuration B consistently recorded significantly reduced EC and TDS levels in the treated effluent (Figures 7 and 8). Use of synthetic raw greywater containing values of EC and TDS higher than FAO recommended values is therefore recommended for testing the designed system to ascertain this observation. The June 10 sample was a highly turbid laundry sample obtained from observed laundry activities on that day. From observation, it had a very high proportion of suspended solids and colloids as well as very dark colour even after pre-sedimentation. This might have been the reason for the high EC and TDS levels on the said day.

It was observed that raw greywater had variable pH, in most cases near neutral (7.0) (Figure 9). The effluent from configuration A generally had pH values close to those of raw greywater, while the effluent from configuration B showed a slight increase in pH (Figure 9). The slight increase in pH in the effluent of configuration B coud be attributed to the pH of the fly ash that averaged around 9.51 (Table 2). Hence due to high pH value of the fly ash, the pH of configuration B's effluent was generally above that recommended by the FAO guidelines for irrigation (6.0 to 8.4) (Misstear et al. 2017). It should be however noted that similar studies reported a reduction in the pH of the wastewater effluent treated with fly ash (Chandrakala et al. 2017; Ganapathy et al. 2018). However, Ganapathy et al. (2018), found that the pH increased with an increase in the fly ash layer thickness in the filtration system. But the pH of the treated effluent was still lower than that of the raw water. This disparity could be attributed to the difference in the source of the fly ash used where in most cases it had low pH. Use of fly ash or an alternative adsorbent material that may have minimal impact on the pH of the treated greywater is therefore proposed.

From the study, it can be concluded that the incorporation of fly ash in the designed SSF system significantly reduced turbidity, colour, suspended solids, and chloride, as well as appreciably increased the DO concentration. It may therefore be applicable for affordable small-scale domestic greywater treatment for non-potable uses such as irrigation.

Future studies should also consider carrying out toxicity characteristic leaching procedure tests to determine if there are hazardous elements present in the fly ash sample. Besides, elemental analysis of fly ash for comprehensive assessment of heavy metal presence and analyses of the filtrate for the presence of metal ions accordingly is highly recommended. Furthermore, investigating the possible use of fly ash in form of granules or pellets is also recommended. Thus, formation of fly ash pellets/granules is proposed for further research.

In addition, research should also be done on the storage and how long the treated water can be kept before unwanted organisms and foul odours start to develop in the water. Furthermore, responsible government bodies should come up with local standards for irrigation water quality, particularly with regards to recycled greywater to help foster a mindset of sustainability in respective nations and regions. The local recycled wastewater standards should also be accompanied by safety guidelines for greywater handling. Scaled up domestic greywater treatment systems based on the proposed system, should be tested and implemented on a large scale at institutional or community level for non-potable use. Moreover, a specially designed and engineered grey water collection system to separately collect low-load grey water from showers, sinks, and wash basins before these streams are mixed with blackwater (water from the toilets) and high load grey water (e.g. from the kitchen) should be designed in relation to the existing infrastructure on each site. Also, obtaining fly ash from sources that will cause zero increase in the parameters of the treated effluent is recommended to ensure full compliance with irrigation water quality and any other desired non-potable use. Practical issues of incorporating the use of fly ash into a full-scale application should be explored further. These may include the need to carry out the TCLP test on the fly ash and use of greywater from several varying greywater sources as well as heating the fly ash to kill any bacteria that may be present. Additionally, further research is proposed to establish suitable replacement/disposal, monitoring, and any health side effects.

All relevant data are included in the paper or its Supplementary Information.