An investigation into the use of acid treated sugar cane bagasse ash-based coagulants for water treatment Open Access

The provision of portable water requires the treatment of the raw water before being supplied to consumers. One of the processes involved in the treatment of raw water is coagulation (Katrivesis et al. 2019). Coagulation is a process sedimentation process of suspended solids in raw water through the destabilization of suspended particles to form microflocs (Pernitsky & Edzwald 2006). Coagulation has also been reported to remove or reduce colour in some waste streams (Aboulhassan et al. 2006). The most common coagulant in Africa is alum due to its relatively low cost and high effectiveness. The environmental and health effects of alum is of major concern due to their residual content in treated waters (Zhong et al. 2016). It has been shown that the residual Al from the use of alum has appositive association with Alzheimer's disease (Katrivesis et al. 2019). The use of alum is also associated with voluminous sludge which is not easily dewatered (Katrivesis et al. 2019). Synthetic coagulants are also associated with the increase in conductivity of treated water since they are added as salts (Nath et al. 2021).

Dictates of sustainable development and circular economy advocate for the use of waste materials as secondary resources. This allows the reduction of pressure to landfill waste and allows the waste to be kept in the economic cycle. Valorisation of waste is one of the benefits of recycling leading to a waste to wealth model. Many other coagulants require natural resources for their production, yet the use of sugarcane bagasse ash may prove to be inexpensive. It is in this vein that sugarcane bagasse ash was used as a possible precursor for the synthesis of a coagulant. The bulk of the bagasse ash is used as a fertiliser by the milling companies. It is well known that SCBA is capable of leaching metals into the environment. The alternative use of SCBA would assist in adding value to the waste for sugar milling companies. The use of secondary resource materials allows for recycling and also helps resource-starved countries to save the little bit of foreign currency they have. Some coal fly ash coagulants/flocculants have been reported in literature, but SBCA-based coagulants are rare (Li et al. 2009; Yan et al. 2012; Wang et al. 2020). Other natural coagulants that have been researched include chitosan, various plant seeds, bark, leaves and peels. These plants include pumpkin, moringa oleifera, banana, cassava to mention only a few (Abatneh et al. 2014; Al-Sameraiy 2017; Katrivesis et al. 2019; Owodunni & Ismail 2021).

SCBA is the incineration residue left after the burning of sugarcane bagasse in boilers of sugar mills for energy production (Chingono et al. 2018). SCBA contains unburnt carbon and hence the black colour of the ash due to low efficiencies of most sugar mill boilers (Rasul & Rudolph 2000). The main oxides associated with SCBA are SiO₂, Al₂O₃, Fe₂O₃ and CaO (Yadav et al. 2020). SCBA has successfully been used as an adsorbent for metals (Yadav et al. 2014), chemical oxygen demand removers (Chingono et al. 2018), supplementary cementitious material (Joshaghani et al. 2016) and production of zeolites through hydrothermal treatment (Shah et al. 2013).

Silica and alumina have been shown to be possibly responsible for the colloidal nature of coagulants (Yan et al. 2012). Acid treatment therefore removes the basic oxides of a coagulant and therefore increases the relative quantities of silica, alumina and iron oxide in the acid-treated coagulant. These oxides play an important role in coagulation.

The objective of this research was to study the (1) synthesise and characterise a coagulant by acid leaching of sugar cane bagasse ash; (2) evaluate the coagulant effectiveness in the reduction of turbidity of synthetic water and (3) investigate the practical applications of the coagulants in treating river water.

The SCBA was obtained from Illovo Sugar Company mill in Chikwawa Malawi. The main oxides in the SCBA were found to be (all % are m/m) MgO (2.594%), Al₂O₃ (5.117%), SiO₂ (56.241%), CaO (4.245%) and Fe₂O₃ (3.786%). Bentonite was obtained from G & W Base Minerals in South Africa. Analytical grades of H₂SO₄ and HCl were supplied by Rochelle Chemicals South Africa.

X-ray diffraction (XRD) and X-ray fluorescence (XRF) analysis were carried out at the Malawi Geological Survey Department. A Rigaku ZSX Primus II XRF spectrometer was used for the determination of oxide content, whilst an Ultima IV Rigaku XRD system using a reference internal ratio (RIR) method was used for the mineralogical determination. Turbidity measurements were carried out at the Malawi Bureau of standards using a turbidity meter (Orion AQ4500) under the ISO 7027-2016 method.

Here, 10 g of bentonite was added to 1,000 ml of distilled water, The mixture was stirred for 1 h at 250 revolutions per minute (rpm). The mixture was then left for 24 h to settle before the coagulation test could be done.

SCBA was mixed with 4 M HCl and 4 M H₂SO₄ in a solid to liquid ratio of 1:10. (Maila et al. 2020). The respective slurries were agitated at 200 rpm for 6 hours. Thereafter the slurries were filtered under gravity. The residues were then washed with distilled water to remove excess acid until the washings’ pH was around 7. The HCl coagulant was labelled as HSCBA, whilst the H₂SO₄ was labelled as SSCBA.

Parametric optimisation was carried out by varying one parameter at a time whilst keeping others constant. Here, 500 ml of turbid waste was poured into 1,000 ml beaker. The coagulants were added to the turbid water followed by rapid stirring at 250 rpm for 3 minutes followed by gentle stirring for 5 minutes at 50 rpm. Four test jars were used per each test. Three jars were dosed at the same solid loading whilst the fourth was the control without coagulant dosing. The turbidity of the supernatant was measured using this method. The parameters investigated were the effect of coagulant dosing, time and initial pH on the coagulation process.

Each result reported was an average of three tests. The average was taken if the individual results were within 10% of each other. Statistical significance was measured at a 95% confidence interval (CI) using a p-value of 0.05. The error bars in graphs were also at 95% CI.

A comparison in the oxide composition of the three coagulants is shown in Table 1.

Acid activation of the SCBA resulted in the reduction of most elemental oxides die to dissolution, however the dissolution was higher in SSCBA than HSCBA (Table 1). This is because H₂SO₄ is a stronger acid than HCl. XRF is a semiquantitative method which would explain the increase in SiO₂ relative content.

There was a general increase in turbidity removal with an increase coagulant dosage. Maximum removal was achieved at 8, 6 and 8% for the SCBA, HSCBA and SSCBA respectively. The increase in turbidity removal to the highest point for each coagulant was significant as the p-value calculation was less than 0.05. The respective highest removal points were followed by a drop in turbidity removal. The removal mechanism may be through interaction between colloids and coagulant which results in the increase in mass of colloid leading to sedimentation (Yimer & Dame 2021). 6HSCBA was the only coagulant to achieve a residual turbidity of less than 5 NTU, the maximum recommended residual NTU by the World Health Organization (WHO 2008). Al- and Fe-bearing species have been shown to be involved in the coagulation process, HSCBA had more Al and Fe compared with SSCBA (Table 1). This may explain why there was more turbidity removal with HSCBA than with SSCBA.

Highly turbid water from Lichenza River obtained and subjected to coagulation using 6HSCBA and 8SSCBA. The turbidity of the water was 1,644 NTU. The outcome for the coagulation test using SSCBA is not shown as turbidity reduction was below 10%.

Table 2 shows that the higher the overflow rates (settling rates) the lower the removal efficiency. Therefore, the design for settling using 6HSCBA should have an overflow rate between 0.01 and 0.02 cm/s to achieve sufficient turbidity removal.

The monitoring of residual effect a coagulant has on treated water is of paramount importance. Synthetic coagulants like alum leave traces of metals in treated water which results in so many health effects (Katrivesis et al. 2019). A sustainable coagulant therefore has to be a natural coagulant which does not leave a significant residual content in the treated water. The river water quality characteristics were monitored before and after coagulation. It could be seen that the most significant change was in turbidity achieving a 99.9% removal (Table 3). The pH change was insignificant and was within the international statutes for wastewater discharge (EPA or WHO statute). There was a significant drop in the content of Fe and Zn (p < 0.05). This may be attributed to the fact that SCBA may have acted as an adsorbent also for the heavy metals, however the reduction in Cu content was insignificant and was within the error limits of the measurement.

The study showed that HCl-treated SCBA (HSCBA) was a better coagulant than H₂SO₄-treated SCBA (SSCBA) as HSCBA could reduce turbidity of a 110 NTU synthetic wastewater by 97.7% and SSCBA by 90.9%. Turbidity removal increased with solid loading for both coagulants up to a 6% m/v S/L and 8% m/v S/L for HSCBA and SSCBA, respectively. After the respective peaks, there was a reduction in turbidity removal for both coagulants. HSCBA was a better coagulant as it had more Fe and Al than SSCBA. HSCBA had a higher amorphous content which allowed for a greater surface area for particle interaction compared with SSCBA. Both coagulants could function effectively in the 6–10 pH range. Only HSCBA could reduce the turbidity of a 1,644 NTU river sample by 99.9% whilst the SSCBA was not effective. A coagulant that does not leave significant residual content in treated water is preferred and is a healthy alternative to synthetic inorganic coagulants. The residual content of the treated water was insignificant compared with the untreated water, further enhancing the possibility to use HSCBA as a natural coagulant.

The author would like to acknowledge the Royal Academy of Engineering who supported this research through the Higher Education Programme of Sub-Sahara Africa (HEPSSA1921/3/31). The author thanks also the students who did the laboratory work and these are Alfred Levison, Isaac Maso, Hassan Allison and Bridget Kesakudza.

Funding for this research was through a grant from the Royal Academy of Engineering who supported this research through the Higher Education Programme of Sub-Sahara Africa (HEPSSA1921/3/31). The funder has no role in the design, data collection and analysis, decision to publish and preparation of manuscript.

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

The authors declare there is no conflict.