Arsenic Removal from Drinking Water in Huanuara, Peru, Using Metalworking Residues: Characterization and Optimization

Abstract

Arsenic contamination in drinking water poses a critical environmental and public health threat, particularly in rural areas such as Huanuara, Peru, where concentrations exceed the 10 µg·L⁻¹ limit established by the World Health Organization (WHO). This study explores the potential use of iron-rich metalworking residues as an alternative adsorbent. Characterization using Scanning Electron Microscopy with Energy Dispersive Spectroscopy (SEM–EDS), X-ray Diffraction (XRD), and Brunauer–Emmett–Teller (BET) analysis revealed a specific surface area of 0.7469 m²·g⁻¹, with magnetite (Fe₃O₄) and metallic iron (Fe⁰) as the predominant phases. Arsenic concentrations were quantified by Inductively Coupled Plasma Mass Spectrometry (ICP–MS). A batch reactor system treated 16 L per cycle under conditions of 293–298 K and 95.46 kPa. Adsorption parameters were optimized using a Central Composite Rotatable Design (CCRD), with adsorbent mass (31.72–88.28 g) and contact time (4.17–9.83 h) as variables. Under optimal conditions (80 g, 9 h), 99.07% arsenic removal was achieved, reducing concentrations from 530.03 µg·L⁻¹ to ≤4.91 µg·L⁻¹. The quadratic regression model (R² = 0.90, p = 0.0006) was validated by ANOVA (p < 0.05; F = 22.02). These results demonstrate that metalworking residues offer a scalable and sustainable solution for arsenic remediation, supporting circular economy principles and decentralized water treatment.

1. Introduction

Arsenic contamination in drinking water represents a critical environmental and public health challenge, particularly in rural areas such as Huanuara, located in the Tacna region of southern Peru. Long-term exposure to arsenic through drinking water has been associated with severe health outcomes, including skin, lung, and bladder cancers, cardiovascular diseases, and developmental impairments in children [1,2]. The World Health Organization (WHO) classifies arsenic as a Group 1 carcinogen, recommending a maximum concentration of 10 µg·L⁻¹ in drinking water to mitigate its harmful effects [3]. However, hydrogeochemical studies indicate that arsenic concentrations in local water sources in Huanuara range from 60.55 µg·L⁻¹ to 759.68 µg·L⁻¹, significantly exceeding this threshold [3,4].

The primary source of arsenic contamination in the region is geogenic, resulting from the dissolution of arsenic-bearing minerals in hydrothermal systems. These natural processes are further intensified by redox-driven mobilization and prolonged water–rock interactions [1,5]. Additionally, anthropogenic activities, including mining, untreated wastewater discharge, and excessive groundwater extraction, contribute to elevated arsenic concentrations in regional water bodies [6,7]. Beyond human health risks, arsenic contamination affects soil quality, reduces agricultural productivity, and disrupts aquatic ecosystems due to bioaccumulation in sediments and organisms [8].

To address this issue, various arsenic removal methods have been developed, including coagulation–flocculation, ion exchange, membrane filtration, and advanced oxidation processes [9]. Among these, adsorption-based technologies are favored for their cost-effectiveness, simplicity, and high removal efficiency [10,11]. Iron-based materials, such as zero-valent iron (nZVI) and iron oxides, have demonstrated strong adsorption capacities due to their affinity for arsenic species [12,13]. However, large-scale applications of these methods involve high operational costs, complex infrastructure, and intensive maintenance.

Recently, processed metalworking residues, primarily composed of iron oxides, have emerged as a promising, sustainable, and low-cost alternative for arsenic removal. These industrial byproducts, derived from machining operations, present a readily available and economically viable solution compared to synthetic adsorbents [14,15]. Their heterogeneous surface morphology, high specific surface area (SSA), and redox-active properties enhance arsenic sequestration, making them suitable for water purification applications [14,15]. Studies have shown that these materials can achieve arsenic removal efficiencies of up to 92%, forming strong surface complexes with arsenic species [12,16]. This approach aligns with circular economic principles, repurposing industrial waste for water treatment.

Arsenic contamination in drinking water is a well-documented issue in the Tacna region, particularly in Huanuara, where communities face serious health risks due to elevated arsenic levels in household water sources [17,18,19]. While large-scale treatment plants have been proposed for Tacna, they remain financially and logistically inaccessible for rural populations, emphasizing the urgent need for affordable, locally available water treatment technologies [6].

This study aims to assess the effectiveness of using metalworking residues as adsorbents for arsenic removal from drinking water in Huanuara, Peru. The primary objectives are to characterize the physicochemical properties of these residues and optimize key adsorption parameters, including adsorbent mass and contact time, to maximize arsenic removal efficiency. The findings will contribute to the development of a low-cost, scalable, and sustainable water treatment solution, particularly for rural areas lacking access to advanced purification technologies. Additionally, by repurposing industrial byproducts, this research supports sustainable waste management practices in the context of water treatment [14].

2. Materials and Methods

2.1. Study Area and Sampling

The samples were collected from the Locumba River Basin, specifically in the district of Huanuara, Candarave Province, Tacna Region, Peru (Figure 1). This region, situated at altitudes between 3000 and 4500 m, experiences active volcanic and hydrothermal processes that significantly contribute to arsenic contamination in groundwater and surface water sources [4].

Huanuara has an arid to semi-arid climate [5,21], with temperatures ranging from 276 to 293 K [22]. The average atmospheric pressure is approximately 72.67 kPa, and annual precipitation is below 300 mm. These climatic conditions, combined with high evaporation rates, limit arsenic dilution in water sources. The prolonged dry season (April–November) further reduces groundwater recharge and increases arsenic mobility due to enhanced evaporation and decreased dilution [23].

The primary drinking water sources in Huanuara are two natural springs (Figure 1b). Manantial A, located at 3324 m (latitude −17.308803, longitude −70.323095), has an average flow rate of 1.23 L·s⁻¹ and supplies the lower section of the town via a pipeline distribution system. Manantial B, at 3310 m (−17.309003, −70.323052), provides 0.64 L·s⁻¹ and feeds a reservoir at 3275 m. Both springs exceed the WHO arsenic guideline, with concentrations ranging from 500 to 759.68 µg·L⁻¹ [22,24]. In 2024, monitoring by the Tacna Health Network confirmed arsenic levels up to 759.68 µg·L⁻¹, 75 times the WHO limit of 10 µg·L⁻¹. Even lower concentrations, such as 600.08 µg·L⁻¹, pose significant health risks [24]. These findings align with those from previous studies on the geological and climatic influences on water quality in the Locumba River Basin [4,22].

Water Sampling and Preservation

To evaluate the efficiency of arsenic removal using metalworking residues in a batch reactor, sampling campaigns were conducted. Water was collected from a representative site (−17.314436, −70.323634) in multiple 20 L polyethylene containers and georeferenced using a high-precision GPS (±5 m accuracy). A portion was preserved following ISO 5667-3:2018 [25] for the quantification of arsenic and iron, while the remaining samples were sent to the Physicochemical Laboratory at UNJBG Tacna for adsorption tests in the reactor. Post-treatment samples were preserved under the same standards for further analysis.

All samples were handled in accordance with ISO 5667-3:2018 to ensure integrity. Pre-washed high-density polyethylene (HDPE) containers were immediately sealed and stored at 277 K in insulated coolers. Sampling equipment was rinsed with distilled water to prevent cross-contamination and dried before reuse. Additionally, samples were transported to the laboratory within 48 h for analysis.

2.2. Adsorbent Material

The adsorbent material used in this study consisted of metalworking residues collected from several local metalworking shops in Tacna, Peru. A number of these shops granted permission to collect metallic residues, primarily iron shavings produced during machining processes such as cutting, grinding, filing, drilling, and welding. These shops specialize in shaping metals like mild steel, galvanized iron, wrought iron, aluminum, stainless steel (in limited applications), and various iron-based alloys. These materials are commonly used to fabricate structural components such as railings, facades, gates, doors, and metal furniture designed for corrosive environments. The residues collected represent a diverse mix of industrial byproducts, making them uniquely suited for arsenic adsorption applications [26].

Pretreatment Process of the Adsorbent

To optimize adsorption efficiency, the metalworking residues were subjected to a pretreatment protocol consisting of two main steps. First, magnetic separation was applied to selectively isolate ferromagnetic fractions, thereby enriching the material in reactive iron-bearing phases. Second, the recovered material was oven-dried and sieved through a No. 40 mesh (425.77 µm) for 10 min to ensure a uniform particle size distribution and improve the reproducibility of the adsorption experiments [27].

All adsorption tests were carried out under the natural pH conditions of the collected water samples, without any chemical adjustment, to evaluate the practical effectiveness of the adsorbent under environmentally relevant conditions [28].

2.3. Adsorbent Characterization

The physicochemical properties and adsorption capacity of the metalworking residues were analyzed using advanced characterization techniques. Scanning Electron Microscopy (SEM) was utilized to examine surface morphology and porosity, providing high-resolution imaging to assess surface texture and microstructure [29,30]. Energy Dispersive X-ray Spectroscopy (EDS) was employed to determine elemental composition and verify the presence of active sites crucial for arsenic adsorption [29,31].

To identify crystalline phases, X-ray Diffraction (XRD) was performed, providing structural information relevant to adsorption efficiency [31,32,33]. Additionally, Brunauer–Emmett–Teller (BET) analysis was conducted to quantify SSA and porosity, which are key parameters for adsorption optimization [34]. These integrated techniques ensured a comprehensive characterization of the material, which is essential for evaluating its suitability for arsenic removal [34,35].

2.4. Batch Reactor System for Arsenic Removal

A batch reactor system was developed to evaluate arsenic removal under controlled conditions, simulating real-world water treatment scenarios while ensuring operational reproducibility. The system processes 16 L of contaminated water per batch, equivalent to 80% of the reactor’s total capacity, optimizing mass transfer and minimizing vortex formation, which enhances adsorption efficiency [36,37]. The reactor operates at an ambient temperature of 286–298 K and an atmospheric pressure of 95.46 kPa, with adsorption experiments conducted in the Physicochemical Laboratory at UNJBG (Tacna). These controlled conditions ensured optimal adsorption performance while preserving experimental integrity. The treatment cycle, which ranges from 5 to 10 h, was determined using CCRD, an optimization methodology widely used in water treatment research.

The primary treatment unit consists of a 20 L HDPE batch reactor, selected for its high chemical resistance and mechanical durability. HDPE is NSF/ANSI 61-certified, ensuring no undesirable compounds leach into the treated water [38], and complies with CFR Title 21, making it suitable for potable water applications [39]. The reactor is specifically designed for arsenic adsorption, utilizing metalworking residues as a cost-effective and sustainable adsorbent.

The system is structured within a robust metal frame and integrates key components to facilitate the treatment process. At the top, a Rushton-type impeller (a), mounted on a vertically aligned shaft and driven by a variable-speed motor, operates at 40 rpm, ensuring uniform mixing and improving mass transfer between the aqueous phase and the adsorbent [40,41]. The batch reactor (b) functions as the main adsorption unit, where arsenic-contaminated water interacts with the adsorbent material. Below the reactor, a sludge collection tank (c) facilitates the separation and storage of process residues, allowing for controlled disposal or potential reuse. The filtration unit (d) removes suspended solids, enhancing water clarity and ensuring compliance with water quality regulations [42]. Finally, the treated water storage tank (e) collects purified water for post-treatment physicochemical characterization and arsenic concentration analysis.

To ensure precise system design and functionality, a three-dimensional model of the reactor system was generated using SketchUp [43], providing a detailed structural and functional visualization (Figure 2). The reactor was specifically designed considering the hydrogeochemical conditions of Huanuara, Tacna Region, Peru, where metalworking residues were evaluated as a sustainable alternative for arsenic removal from drinking water.

This scalable and modular batch reactor system provides a cost-effective and adaptable solution for arsenic removal, particularly for decentralized water treatment applications in rural communities. By integrating adsorption, separation, and filtration into a single robust unit, the system complies with international environmental and public health standards [3,44] and represents a promising approach for improving water quality in arsenic-affected regions.

2.5. Experimental Design

This study employed a CCRD within a Response Surface Methodology (RSM) framework to optimize arsenic removal efficiency (%) in a batch adsorption system. The experimental design allowed for the evaluation of linear, quadratic, and interaction effects, facilitating the development of a predictive statistical model [45].

Two key independent variables were examined: mass of metalworking residue (x₁) and contact time (x₂). Residue mass levels were set at 40 g (−1), 60 g (0), and 80 g (+1), with axial points at 31.72 g (−1.41) and 88.28 g (+1.41). Contact times were evaluated at 5 h (−1), 7 h (0), and 9 h (+1), with axial points at 4.17 h (−1.41) and 9.83 h (+1.41). The agitation speed was maintained at 40 rpm to enhance mixing and mass transfer [46].

Arsenic removal efficiency (η, %) was calculated using

$$Arsenic removal efficiency\left(\%\right)={\displaystyle \frac{C_0-C_e}{C_0}}·100$$

(1)

where C₀ and C_e represent the initial and final arsenic concentrations (µg·L⁻¹), respectively. This equation quantifies the percentage of arsenic removed under different conditions.

Statistical analysis was performed using ANOVA (p < 0.05) to assess the significance of variables, and RSM was applied to model and optimize adsorption conditions. Statgraphics Centurion Version 19.6.05 was used for data processing and model validation, ensuring reliability and predictive accuracy [45].

2.6. Quantification of Arsenic, Iron, and Physicochemical Parameters

Arsenic and iron concentrations in water samples were quantified using Inductively Coupled Plasma Mass Spectrometry (ICP-MS), following US EPA Method 200.8 Rev. 5.4 (1994; validated and modified in 2016). This method ensures compliance with international standards for trace element determination in water and waste. The analysis included total arsenic and total iron to assess metal contamination in the samples.

Physicochemical parameters, including pH, electrical conductivity (EC, µS·cm⁻¹), turbidity (NTU), oxidation-reduction potential (ORP, mV), total dissolved solids (TDS, mg·L⁻¹), resistivity (Ω·m), salinity (g·kg⁻¹), and dissolved oxygen (DO, mg·L⁻¹), were measured in situ using an Environmental Express PCD380 waterproof three-channel multiparameter meter. Water quality characterization adhered to the WHO Guidelines for Drinking-Water Quality, ensuring compliance with international safety standards.

3. Results and Discussion

3.1. Characterization of Natural Water from Huanuara

The adsorption efficiency of metalworking residues was evaluated under different experimental conditions, with arsenic removal influenced by both adsorbent dosage and contact time.

Table 1. Residual arsenic and iron concentrations under different experimental conditions.

Initial concentrations: − Arsenic—530.03 ± 11.00 µg·L−1; − Iron—21.30 ± 15.00 µg·L−1; − WHO guideline limits—10 µg·L−1 for arsenic, 300 µg·L−1 for iron [47].
Assays	Metalworking Residues (Coded Value **) (g)	Contact Time (Coded Value) (h)	Residual Arsenic (µg·L−1)	Residual Iron * (µg·L−1)
1	−1 (40)	−1 (5)	25.21	13.80
2	+1 (80)	−1 (5)	16.10	105.00
3	−1 (40)	+1 (9)	18.83	19.10
4	+1 (80)	+1 (9)	4.91	88.50
5	−1.41 (31.72)	0 (7)	28.55	23.90
6	+1.41 (88.28)	0 (7)	19.13	132.70
7	0 (60)	−1.41 (4.17)	20.27	28.60
8	0 (60)	+1.41 (9.83)	5.46	15.10
9	0 (60)	0 (7)	27.78	83.00
10	0 (60)	0 (7)	29.97	77.00
11	0 (60)	0 (7)	25.83	75.10

* Uncertainty in residual arsenic: ±0.1 µg·L⁻¹. ** Coded values (−1, +1, −1.41, +1.41, 0) correspond to CCRD.

presents experimental results, highlighting the residual arsenic and iron concentrations under different conditions.

The application of metalworking residues effectively reduced arsenic levels, achieving a maximum removal efficiency in Assay 4 (80 g, 9 h) (Table 1), resulting in a final arsenic concentration of 4.91 µg·L⁻¹, which complies with WHO standards.

Arsenic removal efficiency was directly influenced by the dosage of metalworking residues and contact time. Assay 8 (60 g, 9.83 h) resulted in a residual arsenic concentration of 5.46 µg·L⁻¹, demonstrating the beneficial effect of prolonged exposure. Conversely, lower dosages, such as in Assay 5 (31.72 g, 7 h), left an arsenic concentration of 28.55 µg·L⁻¹, indicating that insufficient adsorbent levels hindered compliance with drinking water standards.

These findings align with previous research on iron-based adsorbents, such as ZVI, iron-coated sand, and iron oxide-loaded biochar, which have shown that increasing adsorbent dosage enhances arsenic retention due to the greater availability of active sites [6,48]. Studies on industrial iron residues, particularly those containing iron and aluminum oxides, have reported arsenic removal efficiencies exceeding 90% under optimized conditions [2,49].

3.1.1. Effect of pH and Physicochemical Parameters on Water Stability

The initial pH of the water was recorded at 8.10, while the treated water maintained a stable pH range of 8.18–8.37, remaining within the WHO’s recommended limits (6.5–8.5) (

Table 2. Physicochemical parameters of water samples.

Assays	pH	ORP (mV)	TDS (mg·L−1)	Resistivity (Ω·m)	EC (mS·m−1)	Salinity (g·kg−1)	Temperature (K)	DO (mg·L−1)
Initial Water Sample	8.10	67.5	633	7.88	12.70	0.70	291.15	8.00
1	8.19	68.3	660	7.66	13.14	0.55	286.35	7.43
2	8.26	74.6	676	7.40	13.48	0.57	286.55	8.86
3	8.37	85.4	663	7.54	13.20	0.59	288.85	8.07
4	8.24	78.7	665	7.49	13.39	0.56	288.15	7.96
5	8.35	83.3	666	7.50	13.18	0.60	290.05	8.48
6	8.30	81.6	673	7.46	13.42	0.57	287.25	7.94
7	8.18	70.2	689	7.26	13.78	0.61	288.45	8.99
8	8.26	86.1	677	7.37	13.49	0.56	288.35	7.87
9	8.30	84.9	669	7.38	13.54	0.58	287.35	7.91
10	8.32	86.5	662	7.54	13.22	0.66	287.55	7.85
11	8.34	87.3	659	7.45	13.42	0.62	288.75	7.92

). This minimal variation suggests that the application of metalworking residues for arsenic removal did not significantly alter the water’s pH. The adsorption of arsenic onto iron-based materials is highly pH-dependent, with previous studies on iron oxides and ferrihydrite confirming that removal efficiency is highest at pH 6–8. Extreme values reduce electrostatic attraction between arsenate species and the adsorbent surface, thereby decreasing adsorption efficiency [50,51].

Research on steel slag and iron-modified adsorbents suggests that alkaline conditions promote the formation of calcium arsenate (Ca₃(AsO₄)₂), which further enhances arsenic removal [52]. However, maintaining pH below 8.5 is critical to prevent arsenic desorption and efficiency losses [10]. The observed pH stability in this study indicates that metalworking residues create favorable conditions for arsenic adsorption without requiring additional pH adjustments, thereby enhancing their practicality in rural water treatment applications [53].

Beyond pH stability, the ORP increased from 67.5 mV in the initial water sample to 68.3–87.3 mV in the treated samples, with the highest ORP observed in Assay 11 (87.3 mV). This shift indicates a transition toward more oxidizing conditions, favoring arsenite (As (III)) oxidation to arsenate (As (V)), a form more readily adsorbed onto iron-based materials [54].

TDS increased slightly from 633 mg·L⁻¹ in the initial sample to a maximum of 689 mg·L⁻¹ in Assay 7. The lowest TDS among treated samples (659 mg·L⁻¹ in Assay 11) suggests minor variations due to mineral dissolution from metalworking residues. EC followed a similar trend, rising from 12.70 mS·m⁻¹ to 13.14–13.78 mS·m⁻¹, with the highest value in Assay 7 (13.78 mS·m⁻¹). These changes remained within the WHO’s recommended range for drinking water (500–1000 mg·L⁻¹).

Resistivity decreased from 7.88 Ω·m in the initial sample to 7.26–7.66 Ω·m in the treated samples, confirming the inverse relationship with EC. The lowest resistivity was recorded in Assay 7 (7.26 Ω·m), correlating with the highest EC. This behavior aligns with that seen in previous studies on iron oxide interactions with ionic contaminants [55]. Salinity levels ranged from 0.55 to 0.70 g·kg⁻¹, with the highest value in the initial water sample. The treated water exhibited slight variations (0.55–0.66 g·kg⁻¹), indicating that the process did not introduce excessive salt content [1].

Water temperature remained stable (286.35–291.15 K), with the lowest recorded in Assay 1 (286.35 K) and the highest in the untreated sample (291.15 K), suggesting minimal thermal impact from treatment. DO levels varied from 7.43 to 8.99 mg·L⁻¹, with the lowest in Assay 1 and the highest in Assay 7, preventing anaerobic conditions that could lead to undesirable arsenic transformations [56].

The stability of pH and key physicochemical parameters confirms that metalworking residues provide a reliable and sustainable approach for arsenic remediation. The minimal pH impact eliminates the need for additional chemical adjustments, improving operational feasibility.

3.1.2. Iron Leaching and Secondary Contamination

Residual iron concentrations in the treated water ranged from 13.80 µg·L⁻¹ to 132.70 µg·L⁻¹, remaining well below the WHO’s threshold of 300 µg·L⁻¹. The highest iron concentration, observed in Assay 6 (88.28 g, 7 h) at 132.70 µg·L⁻¹, suggests that excessive residue dosages may contribute to iron leaching.

Similar trends have been observed in steel slag-based technologies, where iron oxides facilitate arsenic adsorption, but excessive iron dissolution can lead to secondary contamination [57]. Iron leaching in iron-coated sand systems varies with water chemistry but typically remains within safe drinking water limits [58]. Despite potential concerns, the findings confirm that iron release in this study remained controlled, ensuring compliance with international drinking water quality standards.

3.1.3. Interrelationship Between Arsenic Removal, pH, and Iron Concentration

The interplay between arsenic removal, pH, and iron concentration is critical in adsorption-based water treatment. The slight increase in pH observed in this study may have contributed to maintaining optimal arsenic adsorption conditions, as previous research suggests that slightly alkaline environments enhance arsenic removal efficiency [59].

Furthermore, dissolved iron may play a secondary role in arsenic removal through co-precipitation and complexation reactions, forming iron–arsenic complexes that enhance overall adsorption efficiency [60]. However, the results suggest that arsenic removal in this study was primarily driven by direct adsorption onto metal oxides, as assays with lower iron leaching still achieved high removal efficiencies.

Studies on similar technologies emphasize the importance of balancing iron release and arsenic adsorption to ensure both efficiency and compliance with water quality regulations [61]. These findings confirm that metalworking residues present a viable and sustainable solution for arsenic remediation, provided that operational parameters are optimized to prevent secondary contamination.

3.2. Characterization of Processed Metalworking Residues

The SEM analysis was conducted using a TESCAN VEGA3 LMU (Brno, Czech Republic), revealing a heterogeneous morphology composed of angular and spherical particles (Figure 3a). This structural diversity enhances surface area and adsorption efficiency by increasing the availability of active sites for arsenic binding [11,27]. The angular particles feature sharp edges, promoting physical adsorption, while the spherical particles contribute to porosity, facilitating higher arsenic retention through electrostatic attraction and redox interactions.

SEM images at 150× magnification (Figure 3a) highlight the rough and irregular surfaces of the particles, confirming a high density of active adsorption sites, which significantly improves adsorption efficiency [49]. The specific surface area of the samples was determined using the BET method with a Nova 800 BET analyzer (Anton Paar, Austria). The BET-specific surface area (SSA) of the metalworking residues was measured at 0.75 m²·g⁻¹. These results, also detailed in Table 2, confirm that this SSA is typical for bulk mixtures of ZVI and magnetite, influenced by the substantial particle size, which limits access to active adsorption sites. The inverse correlation between particle size and SSA is supported by the literature [62,63]. For instance, 100-mesh ZVI typically has an SSA of about 0.25 m²·g⁻¹ [64], while magnetite (Fe₃O₄) produced from hematite reduction ranges from 4 to 10 m²·g⁻¹ and can reach up to 100 m²·g⁻¹ for particles approximately 50 nm in size [62]. More reactive materials like nanoscale ZVI display SSAs between 25 and 54 m²·g⁻¹ due to their smaller particle sizes and higher surface energies [65]. Ferrihydrite, being highly porous, can exceed 300 m²·g⁻¹ due to its amorphous and hydrated structure [62].

The average pore diameters were determined using the Barrett–Joyner–Halenda (BJH) method [66], recorded at 5.97 nm during adsorption and 6.30 nm during desorption. These results, detailed in

Table 3. Pore size and surface area analysis.

Parameter	Unit	Result
Single-point surface area	m2·g−1	0.7056
Specific surface area (BET)	m2·g−1	0.7469
t-Plot micropore area	m2·g−1	0.1960
t-Plot external surface area	m2·g−1	0.9276
t-Plot micropore volume	m2·g−1	0.00006
BJH adsorption average pore diameter	nm	5.9732
BJH desorption average pore diameter	nm	6.2978

, confirm the mesoporous structure of the material, which facilitates rapid contaminant diffusion and enhances adsorption kinetics. The structural attributes of mesoporous materials are critical, improving interaction dynamics between the contaminant and the adsorbent surface and thus boosting the efficacy of water treatment processes. Such properties are essential for the design and application of adsorbents in environmental water treatment, where rapid and efficient contaminant removal is crucial [67].

Elemental Mapping (Figure 3b) and the EDS spectrum (Figure 3c) confirm that the material primarily consists of iron (Fe) and oxygen (O), with trace amounts of carbon (C), aluminum (Al), silicon (Si), calcium (Ca), sulfur (S), chlorine (Cl), and potassium (K), elements that may influence arsenic (As) adsorption. Aluminum functions as a coagulant, facilitating As(V) removal via precipitation and adsorption onto metal oxides [28], whereas silicon enhances structural integrity but competes with arsenic for adsorption sites, potentially diminishing efficiency [68]. Calcium contributes significantly to arsenic sequestration by forming stable As(III)–Ca complexes, achieving up to 85% removal at pH 10.75, and enabling As(V) coprecipitation as Ca₃(AsO₄)₂, thereby limiting arsenic mobility in aqueous systems, with efficiency directly influenced by calcium concentration [52,69]. Sulfur enhances arsenic immobilization through the formation of metal sulfides, generating As₂S₃, which reduces arsenic solubility under reducing conditions [70]. The presence of chlorine (Cl⁻) in aqueous solutions affects arsenic (As) adsorption by forming Fe–As–Cl complexes, which enhance arsenic solubility and mobility, particularly under acidic conditions, potentially reducing the efficiency of iron-based adsorbents in water treatment applications [28,71]. Similarly, potassium (K⁺) interacts with silicates and iron compounds, modifying the surface charge of the adsorbent and influencing the distribution of active sites. The impact of potassium on arsenic removal is concentration-dependent and influenced by competing ions in the solution, which can significantly alter adsorption efficiency [52,65]. The interplay of these trace elements dictates arsenic speciation, retention, and mobilization, emphasizing the need to optimize pH, ionic strength, and redox potential to enhance arsenic adsorption efficiency in practical water treatment applications.

The XRD analysis was performed using an Aeris diffractometer (Malvern Panalytical, Almelo, The Netherlands) with Cu-Kα radiation (λ = 1.5406 × 10⁻¹⁰ m). The diffraction patterns were obtained over a 2θ range of 10.006° to 79.988°, with a step size of 0.011°. EDS analysis confirmed that the material consists primarily of iron (Fe) and oxygen (O). Additionally, the XRD patterns confirm the presence of magnetite and metallic iron (Fe⁰), consistent with the reference patterns from the Inorganic Crystal Structure Database (ICSD 26410 for magnetite and ICSD 14754 for metallic iron). These results highlight the dominant presence of iron oxides in the residues, which play a crucial role in arsenic adsorption. The XRD spectrum further confirms the presence of these crystalline phases, with diffraction peaks matching the standard reference patterns (Figure 4).

The magnetite in the residues is particularly instrumental in arsenic adsorption. Its structure provides active surface sites, facilitating the adsorption of both As(V) and As(III) species. Previous studies have demonstrated that magnetite exhibits a higher adsorption capacity for As(V) due to its stable ionic form, making it more readily adsorbed than As(III) [72,73]. Additionally, the presence of metallic iron contributes to the arsenic removal process by reducing As(III) to As(V), thereby enhancing its subsequent adsorption onto the magnetite surface [74]. This redox transformation mechanism significantly improves the overall efficiency of arsenic removal, as As(III) is less adsorbable than As(V).

The metalworking residues exhibit a surface area of 0.75 m²·g⁻¹, which is relatively low compared to other iron oxide-based adsorbents. Despite this limitation, the material demonstrates notable efficiency in arsenic removal. The synergistic interaction between magnetite and metallic iron compensates for the lower surface area, thereby enabling the effective removal of both As(V) and As(III). While a reduced surface area may constrain the overall adsorption capacity, the high reactivity of magnetite and the redox properties of metallic iron enhance the efficacy of metalworking residues as an arsenic adsorbent [74,75].

A comparative analysis between metalworking residues and other iron oxide-based adsorbents and industrial byproducts provides further insights into their effectiveness in arsenic removal. Magnetite has demonstrated superior performance compared to other iron oxides, such as goethite and ferrihydrite, due to its higher adsorption capacity and favorable magnetic properties, which facilitate easier post-removal separation [76].

Additionally, industrial byproducts such as mill scale and slag, which also contain iron oxides, have been explored for arsenic removal. Although these materials exhibit some degree of arsenic removal capability, they generally present limitations in adsorption capacity and regeneration potential when compared to magnetite. Metalworking residues, owing to their magnetite and metallic iron content, offer a distinct advantage. The combined adsorption mechanism and redox conversion of As(III) to As(V) further enhance their arsenic removal performance [75].

In summary, metalworking residues, composed of both magnetite and metallic iron, exhibit high efficiency in arsenic removal from contaminated water. Magnetite facilitates As(V) adsorption, while metallic iron enhances the overall removal process by reducing As(III) to As(V). This dual mechanism of adsorption and redox transformation establishes metalworking residues as a highly effective material for arsenic remediation.

3.3. Optimization and Performance Analysis of Arsenic Removal Using Metalworking Residues

3.3.1. Statistical Validation and Model Significance

The efficiency of arsenic removal using processed metalworking residues was evaluated through a CCRD, as shown in

Table 4. Matrix and arsenic removal efficiencies using metalworking residues at natural pH, 287–293 K, and 40 rpm.

Independent Variables Coded (Actual)			Arsenic Removal Efficiency (%)
Assays	Mass of Metalworking Residues (g)	Contact Time (h)	Experimental	Predicted
1	−1 (40)	−1 (5)	95.24	94.80
2	+1 (80)	−1 (5)	96.96	96.52
3	−1 (40)	+1 (9)	96.45	96.62
4	+1 (80)	+1 (9)	99.07	98.33
5	−1.41 (31.72)	0 (7)	94.61	93.98
6	+1.41 (88.28)	0 (7)	96.39	96.41
7	0 (60)	−1.41 (4.17)	96.18	96.66
8	0 (60)	+1.41 (9.83)	98.97	99.22
9	0 (60)	0 (7)	94.76	95.19
10	0 (60)	0 (7)	94.35	95.19
11	0 (60)	0 (7)	95.13	95.19
The uncertainty associated with the values in the table can be shown as follows: Arsenic removal efficiency (Experimental)= ±2.07%

. The experimental data were analyzed using a quadratic regression model, with its significance assessed through Analysis of Variance (ANOVA), as presented in

Table 5. Analysis of Variance (ANOVA) for the model representing arsenic removal efficiency.

Source of Variation	Sum of Squares	Degree of Freedom	Mean Square	F-Test	p-Value
Regression	24.12	3.00	8.04	22.02	0.0006
Residual	2.56	7.00	0.37
Lack of fit	2.25
Pure error	0.30
Total	26,68
R2 = 0.90; F3:7:0.05 = 4.35

. The final quadratic model obtained is given by

$$\mathrm{A}\mathrm{s} \mathrm{r}\mathrm{e}\mathrm{m}\mathrm{o}\mathrm{v}\mathrm{a}\mathrm{l} \mathrm{e}\mathrm{f}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{c}\mathrm{y}\left(\%\right)=106241+0.043x_1-4.335x_2+0.343x_2^2$$

(2)

where x₁ represents the mass of metalworking residues (g) and x₂ represents the contact time (h).

Although Equation (2) provides a strong fit to the experimental data, its validity is limited to the range of the studied parameters. The quadratic model effectively captures the relationship between arsenic removal efficiency, the mass of metalworking residues, and contact time, but it does not fully account for adsorption kinetics or equilibrium dynamics beyond these limits. Specifically, at low contact times, the model predicts removal efficiencies exceeding 100%, which are physically unrealistic and indicate an overestimation of removal performance outside the experimental domain. This limitation suggests that, while statistically robust, the model should be interpreted with caution when extrapolating beyond the tested conditions. Future studies should refine the model by incorporating kinetic constraints or alternative adsorption models to improve predictive accuracy and provide realistic estimates across various operational scenarios.

The ANOVA results confirm the statistical significance of the model (p < 0.05), with both adsorbent mass (p = 0.0249) and contact time (p = 0.0223) being significant factors. The quadratic term for adsorbent mass (0.343x₁², p = 0.0128) [45] indicates a non-linear relationship, suggesting that arsenic removal efficiency reaches a saturation point at higher adsorbent masses. These findings align with previous studies on the optimization of iron-based adsorbents for arsenic removal, where saturation effects limit further efficiency improvements beyond a critical adsorbent mass [77].

Additionally, the lack-of-fit test (p = 0.27) is not significant, indicating that the model adequately describes the experimental data. The model explains 90.42% of the variability in arsenic removal efficiency (R² = 90.42), confirming its predictive reliability. These results are consistent with prior adsorption models using iron-based materials, where similarly high R² values have been reported [56].

The Durbin–Watson test (DW = 1.05, p = 0.0655) confirmed the absence of autocorrelation, supporting the robustness of the model’s assumptions. Similar findings have been reported in studies applying response surface methodology (RSM) for optimizing heavy metal removal processes, validating the approach used in this study.

3.3.2. Effects of Adsorbent Mass and Contact Time on Arsenic Removal

Effect of adsorbent mass: The results indicate that increasing the adsorbent mass improves arsenic removal efficiency up to approximately 85 g, beyond which additional increases do not yield significant improvements. The quadratic term (0.343x₁², p < 0.05) confirms that higher adsorbent masses lead to adsorption site saturation, reducing efficiency gains [45]. This behavior has also been observed in studies with iron oxide nanoparticles, where excessive adsorbent loading led to particle aggregation, thereby limiting adsorption efficiency [27,71].

At 40 g, the efficiency ranges between 94.00% and 95.24%, depending on contact time. The highest efficiency (98.97%) occurs at 60 g (Assay 9), while an increase to 88.28 g (Assay 6) results in a slight decline (96.39%). Similar trends have been observed in experiments using ZVI and iron-based composites, where arsenic adsorption efficiency decreased due to diffusion limitations at high adsorbent dosages [16].

Effect of contact time: The contribution of contact time in Equation (2) (% removal) (0.343x₂ − 4.353x) demonstrates a non-linear behavior. These data are valid only within the tested range (5–9.83 h), as other studies have shown that prolonged contact times may lead to desorption or competitive adsorption effects. This phenomenon has been reported in studies involving iron-based adsorbents, where extended exposure resulted in arsenic re-equilibration and partial desorption [28].

The highest efficiencies (≥98%) were observed at contact times of 5 to 9 h, beyond which efficiency stabilizes. This behavior is attributed to the adsorption equilibrium, where excessive contact times can lead to surface re-equilibration and the potential release of weakly bound arsenic [78]. Studies on iron-based materials have shown similar equilibrium times, reinforcing the importance of this parameter for optimizing adsorption processes [56,70].

3.3.3. Process Optimization and Practical Implications

The response surface plots (Figure 5) illustrate the relationship between adsorbent mass, contact time, and arsenic removal efficiency.

Efficiency gains diminish beyond 85 g, supporting the adsorption site saturation hypothesis (Figure 5a) [6]. Optimal conditions for achieving an efficiency of ≥97% are observed at an adsorbent mass of 60–85 g and contact times of 5–9 h (Figure 5b). The absence of significant interaction effects (x₁x₂, p > 0.05) indicates that adsorbent mass and contact time influence efficiency independently, simplifying process optimization [14]. This finding aligns with those of studies that employed response surface methodology for heavy metal removal, demonstrating that optimizing independent factors enhances adsorption efficiency without complex interactions [56].

From an operational perspective, the optimal conditions for arsenic removal using metalworking residues are an adsorbent mass of 50–85 g and a contact time of 5–9 h.

Using more than 85 g of adsorbent or extending the contact time beyond 9 h does not significantly improve efficiency, making these conditions the most cost-effective and sustainable [79]. These results are consistent with those from studies using iron oxide nanoparticles, where similar mass and time constraints were identified as optimal for arsenic removal [71,77].

In this study, the processed metalworking residues achieved 99.07% efficiency, surpassing conventional iron-based materials, which typically range from 90% to 98% [6,80]. This superior performance is attributed to the synergistic interaction between magnetite and metallic iron, enabling electrostatic attraction and As(V) reduction to As(III) [14,77]. These results align with those from previous studies on iron-based adsorbents, such as ZVI, iron-coated sand, and iron oxide-loaded biochar, which demonstrate that increasing adsorbent dosage enhances arsenic retention by providing more active sites [81]. Industrial iron residues, particularly those containing iron and aluminum oxides, have shown arsenic removal efficiencies exceeding 90% under optimized conditions [82]. Compared to these materials, the processed residues exhibited higher efficiency due to their greater surface area, improved porosity, and the combined presence of magnetite and Fe, which enhance arsenic binding and redox transformation [14]. While zero-valent iron nanoparticles (nZVI), iron oxides (e.g., Fe₂O₃, FeOOH), and iron-coated materials have been widely used for arsenic remediation, their adsorption capacities are often limited by lower porosity and surface reactivity [83].

3.4. Environmental and Practical Implications of Arsenic Removal Using Metalworking Residues

The utilization of metalworking residues for arsenic removal offers a cost-effective and sustainable alternative, particularly in resource-limited settings where conventional methods such as reverse osmosis and ion exchange are often unfeasible due to their high operational costs and infrastructure demands [8]. These findings underscore the potential of metalworking residues as an efficient, decentralized water treatment solution, making them especially suitable for rural and peri-urban communities [54].

Furthermore, this approach aligns with circular economy principles by repurposing metal-based waste materials for water purification, thereby enhancing both environmental sustainability and economic feasibility [8]. Studies indicate that repurposing industrial by-products for water treatment not only reduces the environmental footprint associated with arsenic remediation, but also minimizes waste disposal challenges [84]. Additionally, hybrid approaches incorporating adsorption and photocatalytic oxidation methods have demonstrated promising results, further improving arsenic removal efficiency while maintaining cost-effectiveness [8].

In regions with limited access to safe drinking water, integrating low-cost arsenic removal techniques such as the use of iron-based waste materials could significantly enhance public health outcomes while ensuring long-term sustainability [8]. The life cycle analysis of these methods suggests a lower environmental impact compared to conventional high-energy treatment technologies, reinforcing their viability as a scalable solution for arsenic-contaminated water sources [85].

4. Conclusions

This study has demonstrated that processed metalworking residues serve as a viable and sustainable adsorbent for arsenic removal in drinking water treatment. Material characterization, conducted via XRD, SEM-EDS, and BET analysis, identified magnetite and metallic iron as the predominant phases, which facilitated arsenic removal through adsorption and redox transformation mechanisms.

The adsorption process was optimized using CCRD within a RSM framework, determining that an adsorbent mass of 80 g and a contact time of 9 h yielded the highest efficiency, achieving 99.07% arsenic removal and reducing concentrations from 530.03 µg·L⁻¹ to ≤4.91 µg·L⁻¹, complying with the WHO’s standards. The quadratic regression model (R² = 0.90, p = 0.0006) exhibited high predictive accuracy, validated through ANOVA (p < 0.05) and an F-test value of 22.02. Additionally, response surface analysis confirmed that increasing the adsorbent mass or contact time beyond these optimal conditions did not significantly enhance removal efficiency, establishing an operational threshold.

To enhance the understanding of adsorption mechanisms and optimize process efficiency, future research should focus on the kinetics, equilibrium (through isotherm analysis), and thermodynamics of the adsorbent derived from metalworking residues. The absence of these studies limits the precise characterization of the process and its scalability. Additionally, long-term stability assessments under real environmental conditions are crucial, along with evaluating the adsorbent’s performance in natural aqueous matrices containing competing ionic species. Furthermore, developing optimized regeneration strategies will be essential to ensuring process sustainability and operational feasibility on an industrial scale.

The arsenic removal system, based on a 20 L batch reactor, operates passively for 8 to 9 h with minimal intervention. Constructed from HDPE, this reactor treats 16 L of natural water, optimizing the adsorbent dosage between 60 and 80 g and adjusting the residence time to maximize removal efficiency. Following adsorption, the treated effluent undergoes filtration using granular media such as sand, activated carbon, or cartridge filters, removing suspended solids and ensuring compliance with water quality standards. To maintain microbiological safety, a post-treatment disinfection step such as chlorination or other validated methods is recommended. The system’s low-maintenance operation and chemical-free process make it easily integrable into the daily routines of communities with limited infrastructure. Its operational flexibility allows treatment cycles to be conducted overnight or throughout the day, ensuring a continuous supply of safe water without disrupting users’ daily activities.

Although the 20 L reactor has demonstrated its effectiveness, its 9 h treatment cycle for processing 16 L of water may limit its applicability in high-demand scenarios. Therefore, future research should focus on optimizing system performance by reducing treatment times and refining adsorbent dosage without compromising arsenic removal efficiency. Additionally, the development of more efficient and scalable systems would facilitate deployment in communities with limited access to advanced infrastructure. Transitioning to continuous-flow systems or improving batch reactor designs would increase treatment capacity, making large-scale implementation feasible at both community and industrial levels.

The environmental sustainability of this process depends not only on its operational efficiency, but also on the responsible management of spent adsorbent material. Proper disposal is critical to prevent the leaching of toxic species and secondary environmental contamination. Therefore, stabilization through encapsulation with cement or other immobilizing agents, as well as controlled disposal, is necessary to mitigate environmental risks.

Finally, assessing the socioeconomic and environmental impacts of utilizing processed metalworking waste as an adsorbent is essential to ensuring the long-term viability of this water purification strategy. Integrating this approach into circular economy frameworks will enable the valorization of industrial waste, reducing reliance on high-cost conventional technologies while promoting cost-effective, efficient, and environmentally responsible solutions for potable water supply in underserved communities and regions with limited infrastructure.