Recovery of coagulants via acid treatment in potabilization sludges and their reuse in raw and urban wastewaters Open Access

Conventional water treatment plants have used aluminum-based coagulant solutions to remove colloidal substances and other suspended particles from raw water. During this process, a byproduct known as water treatment sludge (WTS) is generated that is typically discharged without prior treatment, causing serious environmental problems for surface waters and nearby ecosystems. Studies have been conducted to evaluate its potential reuse in various processes such as agriculture, construction material manufacturing, pollutant absorption, and its reuse as a coagulant. This study evaluated the recovery of aluminum via acid treatment in WTS from a drinking water treatment plant (DWTP) that supplies water for a population of 413,484 inhabitants, using different pH levels and mixing speeds. The efficiency of the recovered coagulants was evaluated for the removal of color and turbidity in raw and wastewater. The main results showed a maximum aluminum recovery at a pH of 1.5 with values up to 810.5 mg Al/L, with pH being the most influential factor in the process. The removal of color and turbidity of 95.84 and 97.06% were achieved in wastewater and 69.78 and 69.73% in raw water, respectively. The recovered coagulant could be used in DWTPs and in chemically enhanced primary treatment for sewage treatment.

Drinking water treatment plants (DWTPs) typically carry out coagulation, flocculation, and sedimentation processes to remove suspended solids, colloidal substances, and other particles present in the water. These processes generate an unavoidable product known as water treatment sludge (WTS), which accumulates mainly in sedimentation tanks. It is estimated that DWTPs produce 1–3% sludge regarding the water used in the potabilization process (Babatunde & Zhao 2007). These significant amounts of produced sludge generate additional costs for its proper management and disposal. For example, in the Netherlands, approximately $37–50 million is spent annually, and in Australia, $6.2 million is spent annually (Ackah et al. 2018; Kumar et al. 2020).

The characteristics of WTS will depend on the quality of the raw water, the efficiency of the operational structures involved in the treatment, and the final quality of the produced water. This sludge contains particles of sand, clay, colloids, microorganisms, and some natural organic materials reacted or adsorbed by iron (Fe) and/or aluminum (Al) hydroxide; the general physicochemical characteristics are mainly composed of SiO₂, followed by Al₂O₃ and Fe₂O₃, depending on the coagulant solution used (Yang et al. 2014; Zhao et al. 2021).

WTS is generally discharged directly into surface waters and adjacent lands after dehydration processes. However, this type of disposal, although less costly, is not an adequate solution due to the potential contamination of surface water. Such sludge discharges impact receiving water bodies, such as increased concentrations of metals (mainly Fe and Al), suspended solids, turbidity, changes in nutrient cycles, the development of anaerobic conditions, and changes in chemical composition, as well as the possibility of groundwater contamination (Ahmad et al. 2016; Wang et al. 2018).

Principles of Circular Economy that seek to promote regenerative design, eliminating waste and continual use of resources, are becoming increasingly important for water authorities (Nguyen et al. 2022). The health and environmental problems due to uncontrolled discharge of chemical sludge into soil and receiving waters have attracted the attention of researchers and experts to recovery/reuse of coagulants from the WTS as ecofriendly approach (Nayeri & Alireza Mousavi 2022).

In recent years, studies have reported alternative means for the management of WTS. These management methods are based on converting waste to resource principle; some examples of the potential reuse of WTS are: the use as special coagulating–flocculating agent for industrial effluent purification as well as for low-turbidity source water coagulation (Kang et al. 2022), the co-disposal mixed to clayey and sandy soils as barriers in bottom waterproofing layers, daily cover and final cover of landfills (Gonçalves et al. 2017; Mattoso et al. 2024), the production of construction materials and building materials (Shamaki et al. 2021; Tony 2022), its use in agriculture for fertilization purposes (Dassanayake et al. 2015), the recovery of coagulants present in the sludge for reuse in raw and wastewaters (Nair & Ahammed 2014; Ramadan et al. 2017; Ruziqna et al. 2020), and for the treatment of backwash water of DWTPs (Gavlak et al. 2024a).

Mixtures containing WTS and clayey and sandy soils were tested for application as barriers in bottom waterproofing layers, daily cover, and final cover of landfills.

The recovery of aluminum present in WTS via acid is one of the most implemented alternatives due to its high efficiency, low cost, and simple operation. This recovered aluminum can be reused as a coagulant, significantly reducing the consumption of pure coagulant and minimizing environmental impacts (Evuti & Lawal 2011; Ahmad et al. 2016). The recovery efficiency can be influenced by several factors, such as the raw water characteristics, the applied coagulant dose, pH, mixing speed, mixing time, and the amount of acid applied in acidification processes.

This research aimed to recover the coagulant present in the sludge from the Centenario, a full-scale DWTP through acid treatment and evaluate its coagulation capacity through the removal of color and turbidity in raw and domestic wastewaters. Unlike studies such as Nair & Ahammed (2014), who used low-density fresh mud collected directly in a clariflocculator, or that of Ramadan et al. (2017), who used sludge collected in sedimentation units, or that of Lebogang et al. (2023), who used sludge with a higher density and age collected in open sludge lagoons, in this study we used sludge of intermediate age, density and composition, results of the simultaneous washing of the sedimentation units of a DWTP on a real scale. The DWTP supplies 83% of the drinking water for the capital of the municipality of Pasto, whose population by 2024 was 413,484 inhabitants (Colombia-DNP 2025), with an operational flow that varies between 300 and 650 L/s. Since the city lacks an urban wastewater treatment plant, the coagulant recovery study is highly relevant due to its potential for reuse in the water treatment plant itself, as well as for the possible implementation of a chemically enhanced primary treatment for the future urban wastewater treatment plant.

The sludge used in this study was collected from the Centenario DWTP located in Pasto, Colombia, which has a conventional treatment system that uses poly aluminum chloride (PAC) as the coagulant solution. The main source of water that supplies the Centenario DWTP is the Pasto River, whose main sources of color and turbidity are of natural origin and come from the upper part of the river basin, where there are soils rich in carbon derived from the degradation of plant material from the surrounding forests (Moreno 2013). Historically, the DWTP used liquid aluminum sulfate type B in coagulation–flocculation processes; however, to obtain better results in color removal, as of 2012 the use of PAC was chosen, with an amount >19% of alumina and a basicity >72%. The doses of PAC normally applied in the DWTP range from 35 mg/L in low rainfall seasons, to 100 mg/L in high rainfall seasons, depending on the turbidity levels of the raw water, with an average annual value of 53.7 mg/L for the years 2012–2017.

The sampling was carried out at the inspection box where the drains used to evacuate the sedimentation tanks converge. The settlers are washed weekly, with a total volume of 2,680 m³.

The sludge was characterized in an aqueous phase, and the physicochemical parameters listed in Table 1 were evaluated. The HACH HQ40d Multiparameter device was used for pH measurements and complementarily for BOD₅; Imhoff Cone BRAND GMBH for settleable solids; HACH DR 6000 Spectrophotometer was used for chemical oxygen demand (COD) and for preliminary aluminum analysis; nylon filter discs of 0.45-μm pore size was used, crucible porcelain, diaphragm pump, analytical balance, universal oven, and muffle furnace kept at 1,000 °C were used for total dissolved solids (TDS) and total volatile solids (TVS) parameters. Due to the low sensitivity of the spectrophotometric method to quantify aluminum in sludge samples, the use of atomic absorption spectroscopy (AAS) was subsequently chosen in the quantification of aluminum recovered by the acid route.

The measured initial value for the aluminum present in the WTS was very low as a result of the low sensitivity of the spectrophotometric method for its quantification.

In this process, the pH and mixing speed factors were evaluated at different levels until reaching 16 treatments (see Table 2). The samples were acidified by adding 98% sulfuric acid (H₂SO₄) to achieve four pH levels (1.5–2.0 to 2.5–3.0). The four pH levels and mixing rates were selected based on results reported by Ayoub & Abdelfattah (2016), Ramadan & El Sayed (2020) and Maraschin et al. (2020). Subsequently, mixing processes were carried out at four different speeds (125–150 to 175–200) rpm for 60 min. Finally, the samples were allowed to settle for 60 min, and the supernatant was withdrawn and used as the recovered coagulant in the subsequent experiments.

The volume used in the jar tests for coagulant recovery was 1 L, the mixing gradients correspond to each speed were as follows: 125 rpm (190 s⁻¹), 150 rpm (240 s⁻¹), 175 rpm (300 s⁻¹), and 200 rpm (390 s⁻¹).

For the quantification of recovered aluminum, AAS analysis was implemented using the SM 3500 Al-B method.

The recovered coagulant was evaluated in raw and wastewater matrices. The samples were obtained through grab sampling, and in situ parameters of pH, conductivity, dissolved oxygen, and temperature were evaluated. The raw water was collected in the Parshall flume at the Centenario DWTP before any treatment, and the wastewater was collected at the Juan XXII domestic wastewater discharge.

Fifteen millilitres of the recovered coagulants were added to the collected raw and wastewater samples, and jar tests were subsequently performed to evaluate their coagulation capacity according to the removal of color and turbidity. The volume used in the jar tests for reuse of the recovered coagulant was 1 L, the amount of recovered aluminum used in each test was 15 mL. An operational cycle of rapid mixing at 300 rpm (700 s⁻¹) for 10 s was used, followed by medium mixing at 150 rpm (240 s⁻¹) for 1 min, slow mixing at 30 rpm (25 s⁻¹) for 10 min, and finally followed by a sedimentation process of 30 min.

Table 3 presents the results obtained in the characterization of the raw water influent to the DWTP Centenario and the raw wastewater collected at the Juan XXII discharge point.

The results obtained for the evaluated parameters of the water treatment plant sludge are presented in Table 4.

The aluminum values (photometric method) were very low compared to those reported by Ayoub & Abdelfattah (2016), which recorded aluminum values in liquid sludge ranging from 1,124 to 4,096 mg Al/L in three different DWTPs. Additionally, Tony (2022) reported values between 7 and 142 mg Al/kg in solid-state sludge. The pH values registered in the characterization are very similar to those presented by other authors, who mentioned that this type of sludge has a neutral pH between 6.5 and 7.5. For example, Mazari et al. (2018) reported an average value of 7.47.

Volatile solids presented very low values compared to those reported by Mora-León (2022), who registered values exceeding 5,000 mg VSS/L for this parameter. In the case of suspended solids, the registered results are also very low compared to those reported in other studies where values exceeded 1,450 mg TSS/L (Ahmad et al. 2016; Rebosura et al. 2020).

In the case of COD, authors Dassanayake et al. (2015) and Tony et al. (2008) obtained lower results than those found in this study, with values ranging from 216 to 226 mg O₂/L. However, Barakwan et al. (2019) reported very high values up to 9,000 mg O₂/L. In the case of BOD, the results are much lower than those reported by authors such as Castaldi et al. (2014) and Rebosura et al. (2020), who reported results ranging from 45 to 104 mg O₂/L. The low values of TSS and BOD indicate that the source supplying the drinking water treatment plant has low contamination of organic origin; however, since in the essay of the COD both organic and inorganic components of a sample are subject to oxidation, it can be concluded that there is a predominance of inorganic components in the water of the Pasto River.

Table 5 presents the results of aluminum recovery using the acid method, utilizing combinations of the proposed factors: pH and mixing speed.

It can be observed that treatments with pH A = 1.5 were the most efficient in recovering aluminum, with values ranging from 566.0 to 810.5 mg Al/L, followed by treatments with pH B = 2.0 with values ranging from 294.5 to 386.5 mg Al/L. These results are higher than those reported by Ayoub & Abdelfattah (2016) and Ramadan & El Sayed (2020), who in their research found that acidification up to pH = 1.5 was the ideal point to recover the coagulant. In addition, Maraschin et al. (2020) obtained aluminum ion recoveries of up to 90% with acidification processes up to pH = 2.0. It was not possible to estimate the percentage of aluminum recovered, with respect to the coagulant applied in the PTAP, since the washing of the sedimentation units was carried out once a week and the doses used during the time of the research varied between 20 and 40 mg/L.

Acidification can also cause leaching of other metals present in WTS, such as iron, chromium, and manganese; however, these pose minimal risk since their concentrations in the sludge are very low compared to aluminum (Fouad et al. 2016). Other combined methodologies have also been studied for the recovery of ferric coagulants such as alkalization, acidification, and ultrafiltration, with positive results. However, these methodologies can generate high costs (Keeley et al. 2016).

The Shapiro–Wilk test reported a P-value of 0.01995, leading to the conclusion that the data did not follow a normal distribution. The Fligner–Killeen test reported a P-value of 0.03332, indicating that the data did not present homoscedasticity.

The results of the statistical analysis show that pH directly affects the recovery of aluminum, making it necessary to standardize efficient acid addition processes on a real scale for potential reuse applications in DWTP and subsequently conduct cost analyses to assess the economic viability of this alternative.

The Shapiro–Wilk test reported a P-value of 0.01995, leading to the conclusion that the data did not follow a normal distribution. The Fligner–Killeen test reported a P-value of 0.4259, indicating that the data presented homoscedasticity.

As no statistically significant differences were found in the recovery of aluminum as a function of mixing speeds, in an eventual scale-up of the process, the equivalent of the lowest speed of 125 rpm (190 s⁻¹) can be implemented for the purpose of saving energy.

The Shapiro–Wilk test reported a P-value of 8.426e⁻⁰⁶, leading to the conclusion that the data did not follow a normal distribution. The Fligner–Killeen test reported a P-value of 1.928e⁻⁰⁹, indicating that the data did not present homoscedasticity. The non-parametric Kruskal–Wallis analysis demonstrated a statistically significant difference between the different treatments used to remove color in wastewater with a P-value of 1.312e⁻⁰⁶.

The Shapiro–Wilk test reported a P-value of 1.379e⁻⁰⁵, leading to the conclusion that the data did not follow a normal distribution. The Fligner–Killeen test reported a P-value of 7.214e⁻⁰⁶, indicating that the data did not present homoscedasticity. The non-parametric Kruskal–Wallis analysis demonstrated a statistically significant difference between the different treatments used to remove turbidity in wastewater with a P-value of 1.757e⁻⁰⁷.

The obtained results are similar to those obtained in other studies, such as the conducted by Mora-León (2022) and Xu et al. (2009), who recovered coagulants from aluminum sludge using sulfuric acid and achieved maximum turbidity removal of 96% in domestic wastewater. On the other hand, Nair & Ahammed (2014) recovered coagulant by the acid method using sulfuric acid in potabilization sludge, where PAC was used as a coagulant solution, and was also used to treat the effluent from a upflow anaerobic sludge blanket (UASB) reactor that received domestic wastewater, achieving turbidity removal of up to 80%. Meanwhile, Lebogang et al. (2023) obtained removal values of 91–99% in synthetic waters of medium and high turbidity. However, it is recommended to verify the pH of the water after the coagulation process and adjusting it using calcium hydroxide if necessary.

In the case of wastewater, the current legal regulations that correspond to Resolution 0631 of the Ministry of Environment and Sustainable Development (Colombia 2015), do not contemplate either turbidity or color as water quality parameters for the discharge of effluents into water bodies; however, associated with the reduction of turbidity will most likely be the removal of organic matter.

As the implementation of a plant for the treatment of municipal wastewater is an obligation that must be fulfilled in a very short term, the recovery of the coagulant and its reuse for advanced primary treatment could be a priority for the municipal sanitation company.

The Shapiro–Wilk test reported a P-value of 0.002086, leading to the conclusion that the data did not follow a normal distribution. The Fligner–Killeen test reported a P-value of 2.957e⁻⁰⁷, indicating that the data did not present homoscedasticity. The non-parametric Kruskal–Wallis analysis demonstrated a statistically significant variation between the different treatments used to remove color in raw water with a P-value of 3.638e⁻⁰⁷.

The obtained values are similar to those reported by Pedretti & Medeiros (2022) who recovered coagulants from aluminum sludge using sulfuric acid and achieved color removal near to 75%. The obtained color removal results are low compared to those obtained by Vilela (2020), who achieved color removals of up to 99% in raw water using coagulants recovered via the acid–alkaline method employing sodium hydroxide (NaOH – 20%), sulfuric acid (H₂SO₄ – 20%), and hydrochloric acid (HCl – 20%). Gavlak et al. (2024b) achieved color removals of up to 98% in water using coagulants recovered from aluminum sludge acidified with sulfuric acid.

The Shapiro–Wilk test reported a P-value of 0.002258, leading to the conclusion that the data did not follow a normal distribution. The Fligner–Killeen test reported a P-value of 5.055e⁻⁰⁵. The non-parametric Kruskal–Wallis analysis demonstrated a statistically significant variation between the different treatments used to remove color in raw water with a P-value of 3.587e⁻⁰⁵.

The coagulants recovered at pH 1.5 had the highest removal efficiencies, with values ranging from 54.17 to 69.73%. These values are low compared to those reported in other studies where coagulants were recovered by acidification and used in raw water, Hamzah et al. (2022) showed turbidity removal results of up to 99.47%, very similar to those reported by Ruziqna et al. (2020), which demonstrated removal efficiencies of up to 93.28%. However, the results obtained are similar to those found in another study where coagulants recovered through acidification processes combined with filtration processes managed to remove between 60 and 70% of turbidity in raw water (Keeley et al. 2014). Additionally, Dahasahastra et al. (2022) presented similar results with coagulants recovered from aluminum sludge acidified with nitric acid. These coagulants achieved a turbidity removal of 74%.

According to Resolution 2115 of the Ministry of Environment, Housing and Territorial Development and the Ministry of Social Protection (Colombia 2007), which regulates the maximum permissible values of the water quality parameters for treated water in Colombia, the treatments achieved final turbidity values very close to those referred to the standard (2 NTU). With values between 2.47 and 3.79 NTU, demonstrating that the implementation of this type of coagulants recovered by acidification may be a viable option in DWTP; however, for the color parameter, none of the treatments obtained final values close to the maximum permissible value referred to in the standard of 15 PCU.

It is advisable to carry out a cost–benefit analysis to assess the economic feasibility of large-scale use of recovered coagulants for the specific conditions in the municipality of Pasto. Esmeraldas (Ecuador) Torres-Mendoza et al. (2023) reported a 20% recovery of sulfate, but the high costs associated with pumping the sludge, the acquisition of land and the infrastructure necessary for the recovery and use of the coagulant showed that it would be more economical to continue acquiring pure coagulant; however, in the case of recovery and reuse for wastewater treatment in Pasto it would not imply investments in pumping equipment and its respective energy consumption because the present wastewater treatment plant is located in the upper part of the municipality. The future wastewater treatment plant would be located in the lowest part of the municipality.

In this study, the aluminum recovery from the WTS using acidification and the effects of different pH levels and mixing speeds were investigated.

The pH had the greatest influence, with treatments at pH 1.5 reporting the highest recovery values, reaching up to 810.5 mg Al/L. The mixing speed variable and its different levels did not affect the recovery process.

The highest removal efficiencies were registered in wastewater with maximum values of 95.84% for color and 97.06% for turbidity at doses of 15 mL. In raw water, maximum removal values of 69.78% for color and 69.73% for turbidity were obtained.

The treatments proposed in this study show promising results for the reuse of the recovered coagulant in the operation of DWTPs or especially as an advanced primary treatment in wastewater treatment plants. This alternative could reduce the pure coagulant acquisition costs.

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

The authors declare there is no conflict.