Field study of an arsenic removal plant for drinking water using activated carbon and iron in a rural community in the province of Pisco, Peru

AC-Fe

adsorbent material based on activated carbon and iron

INS

Instituto Nacional de Salud

It is estimated that globally 200 million people are exposed to arsenic from drinking water (Shakoor et al. 2015), and it has been estimated that approximately 250,000 people in Peru have been chronically exposed to arsenic in drinking water with concentrations greater than 50 μg/L of arsenic (Bundschuh et al. 2012). The government has established some regulations to improve the water access, but the main problem is the inequality of access and quality of water between the urban and rural areas. For instance, in rural areas only 2.7% of the population consumes water with an adequate level of free residual chlorine, while in urban areas 50.6% of the population does so (MINVIV 2023). Moreover, there are no specific public policies to address the problem of arsenic in drinking water.

The health damages caused by chronic ingestion of such toxic agents have been widely described. Among the main diseases described are the risk of skin, bladder, and liver cancer; respiratory and cardiovascular diseases; and neurological damage (George et al. 2014).

Until the year 2015, more than 8,000 environmental liabilities were registered in the Peruvian territory, which are against potential polluting sources of arsenic and other toxic metals (MINSA 2016).

In 1993, the World Health Organization (WHO) reduced the recommended value of arsenic in drinking water from 50 to 10 μg/L (WHO 2003), increasing the challenge of reducing arsenic content in drinking water. Some technological options are available in the market for the removal of arsenic from drinking water, including coagulation/filtration, ion exchange, reverse osmosis, and electrocoagulation (Glade et al. 2021); however, these technologies may not be economically accessible to rural communities.

Bilal et al. (2021) have recently described the advantages of adsorption techniques over other techniques for the removal of heavy metals from water, among which are the wide pH range in which it can operate, the low costs that are usually associated, and the simplicity of the application procedure.

In this field study, we have developed and tested a pilot plant for the removal of arsenic from a groundwater source. We have used a filter media developed by the Instituto Nacional de Salud (INS), which has been developed based on activated carbon impregnated with iron.

The presence of ions such as chlorides and sulfates were studied with the adsorption capacity of the AC-Fe, and the adsorption capacity for the arsenic species As(III) and As(V) was also evaluated.

The operating design has been studied under real conditions and some changes have been made during the operation process to maximize the arsenic-safe water production capacity.

A filter medium has been developed in the laboratories of INS, which has as its main component activated carbon from coconut shells and impregnated with iron (AC-Fe), which has been protected by patent No. 0001198-2014/DIN-INDECOPI (INS 2016); a size of 12 × 40 mesh has been used to reduce the resistance to the passage of water.

Arsenic speciation in water was carried out using a methodology previously described (Chávez 2009), briefly, using ion exchange resins (DowexTM x 50) previously conditioned with acetic acid, and finally packed in columns. Water samples were passed through these columns and the elutions were analyzed by hydride generation atomic absorption spectroscopy (HGAAS). A Perkin Elmer Analyst 400 model, coupled with a flow injection system of the same brand, model FIAS 100, was used. The result of As(V) was subtracted from the result of total arsenic to determine the concentration of As(III).

Adsorption tests were carried out for the two inorganic arsenic species, for this purpose. A mass of 100 mg of the AC-Fe was agitated with 50 mL of As(V) and As(III) solutions at different concentrations for 24 h. To ensure that the system reaches equilibrium, the adsorption isotherms of both species were obtained, and the maximum adsorption capacity (q_max) was estimated. The test was carried out in triplicate, the average of the q_max obtained is reported.

According to the results of the characterization of the water source, the following results in mg/L were found: phosphate < 0.01, iron < 0.01, sulphate 57.8, and chloride 24.0. Therefore, we carried out the tests of possible interferences caused by sulfate and chloride ions on the adsorption capacity of the AC-Fe, solutions of As(V) and As(III) of initial concentration 1 mg L⁻¹, in the presence of sulfate and chloride ions in concentrations of 30, 60, and 90 mg L⁻¹, low and high concentrations as found in the water source samples (results not shown in the present article). The dose of AC-Fe was 2.0 g L⁻¹, and agitation was maintained for 24 h until equilibrium was reached. Then the solution was filtered with a 0.45 μm porosity membrane filter, and finally the concentrations of As(V) and As(III) in their respective solutions were analyzed. Each interference assay was performed in duplicate, the average is reported.

Initially, the columns worked sequentially, first the water passed through the column filled with a zeolite-based filter medium, and then through the second column, containing the AC-Fe. However, later this design was modified so that the two columns contain AC-Fe, and the water passes in parallel; a second modification was made to the flow rate, which was reduced from 3.0 to 1.8 m³ h⁻¹. The samples were obtained once a week and the analyses were carried out in duplicate. The average obtained was used.

The results found in the arsenic speciation in the water source are provided in Table 1. It was found that more than 80% of the arsenic is present as As(V). These results are similar to those found in other studies conducted in the Andean region of South America, where it has been reported that more than 90% of arsenic in the water of Pozuelos, Bolivia, was present as As(V) (Hudson-Edwards & Archer 2012). Different rivers in the department of Tacna, Peru, have been studied, with the finding that As(V) represented between 83 and 100% in the samples studied (Chávez 2009). In another study conducted on the lakes and streams in Ontario, Canada, it was found that in general the predominant species is As(V) in a proportion often greater than 99% (Chen et al. 2019).

In this regard, q_max values have been reported for other adsorbent media, which used iron salts in their composition with q_max values for As(V) in the range of 0.04–0.50 mg g⁻¹ and for As(III) in the range of 0.04–1.25 mg g⁻¹ (Thirunavukkarasu et al. 2003; Panthi & Wareham 2011; Aremu et al. 2019). Materials with higher adsorption capacities have been developed; Rodriguez-Romero et al. (2020) have obtained a material from the pyrolysis of prickly pear cactus (Opuntia ficus) achieving a q_max of 8.17 mg g⁻¹ for As(V). An activated carbon with impregnated cerium oxide has achieved a q_max of 16.5 and 21 mg g⁻¹ for As(V) and As(III), respectively (Hoang et al. 2022).

The study of interferences with sulfate and chloride ions showed that there is no reduction in the adsorption capacity of any of the two arsenic species; similar results have been described in materials developed based on activated carbon, where nitrate, sulfate, carbonate, and phosphate ions were evaluated for the adsorption of As(V), in which it is concluded that none of these ions affects the adsorption capacity (Velazquez-Jimenez et al. 2018). In another study where the samples contained high levels of chlorides and sulfates it was concluded that they did not affect the effectiveness of arsenic adsorption by a material prepared from sand impregnated with iron (Van den Bergh et al. 2010).

Our results show a slight increase in arsenic adsorption capacity with increasing sulfate concentration. This phenomenon has also been reported by Sun et al. (2006), who proposes a possible formation of an Fe/As/S complex as a possible removal mechanism under reducing conditions; this indicates that the presence of sulfates in the water sources to be treated favors the removal of arsenic. Another aspect to consider is that studies could be guided on the development of adsorbent materials that include the application of sulfur.

In a study performed in a rural community in Bolivia, using materials impregnated with iron oxides, a breakthrough between 40 and 50 bed volumes was obtained (Van den Bergh et al. 2010). In a pilot study whose water contained 16.44 μg L⁻¹ of arsenic, quartz sand with ferrous sulfate was used, obtaining a removal efficiency of up to 97% (Gottinger et al. 2010). Likewise, in another study, in which iron precipitation and subsequent adsorption was used as the most efficient technique, a removal efficiency between 68 and 90% was reported at a flow rate of 50 L h⁻¹. In this work, the breakthrough volume was not reported (Awuah et al. 2009). Many studies have been conducted on activated carbon-based adsorbents; however, there is a lack of full-scale studies due to the presence of coexisting ions, which can vary considerably the results obtained (Amen et al. 2020).

The arsenic content in the water source of the present work contained relatively low concentrations of arsenic. The application to water sources with high arsenic content could drastically decrease the efficiency of safe water production. However, this work can serve as a baseline to project the dimensions and capacity of the columns to be used based on the arsenic concentration of the water source.

Among other sustainable technologies tested in the field for arsenic removal, there is phytoremediation with the use of plants such as vetiver (Goykovic-Cortés et al. 2021). Granular adsorbers and hybrid membranes using milk protein nanofibril-carbon hybrid materials were applied in point-of-use devices, in Peru (Bolisetty et al. 2021). Gonzalez et al. (2019) evaluated the effect of different fluxes (16, 23, and 30 L/m² h) on the rejection of As(V) during nanofiltration in a pilot plant powered by solar panels. The breaking point was not reported, and they managed to produce permeate concentration of 5 μg/L. The flows achieved in the present work are much higher, 25,412 L/m² h. (Gonzalez et al. 2019).

The installation of treatment systems for the removal of arsenic on a large scale in rural areas must consider sustainability. For this, it is important to have the participation of the population from the beginning of the project, as well as that of the authorities.

The present study has demonstrated that the arsenic removal capacity of the AC-Fe media, developed by INS applied in the full-scale water treatment proposed in this work, was able to produce arsenic-safe water with levels below the WHO recommended level (less than 0.010 mg L⁻¹) under the current conditions and parameters of the natural water source of the Pisco River.

Interference tests indicated that the presence of chloride and sulfate ions did not produce a negative effect on the adsorption capacity, indicating a surface selectivity of the AC-Fe media material toward arsenic.

In the treatment plant under the field study, in the community of Montesierpe – Humay, Pisco, Peru, it was found that reducing the flow of each adsorption column from 3.0 to 1.8 m³ h⁻¹ increased the water production capacity by 79% until reaching the break point.

The zeolite filtration column did not contribute to arsenic removal, nor did it reduce the amount of dissolved solids, so it was decided to replace it and increase a second column of AC-Fe to double the production of arsenic-safe water for the population.

Studies should be carried out on the regeneration of spent adsorbent for reuse in treatment plants for removing arsenic from water to reduce operating costs and reduce solid waste generated.

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

The authors declare there is no conflict.