Developing Insights into Pretreatment Optimization: Effects of Eliminating Lime and Soda Ash in Groundwater RO Desalination

Abstract

In arid and water-stressed regions, groundwater desalination plants are critical for ensuring reliable potable water supplies, making improvements in their operational efficiency and cost effectiveness a priority for utilities. In many such facilities, lime and soda ash softening remain common pretreatment practices, which increase chemical consumption and sludge generation, prompting the need for alternative low-chemical strategies. This study evaluates the technical, operational, and economic implications of transitioning a full-scale brackish groundwater desalination plant, from lime–soda ash softening (old plan) to a low-chemical pretreatment strategy based on antiscalant dosing (new plan) upstream of reverse osmosis (RO). Key parameters, including pH, total hardness, calcium and magnesium hardness, silica, iron, alkalinity, and total dissolved solids (TDS), were measured and compared at multiple locations within the treatment plant under both the old and new plans. Removing lime and soda ash caused higher levels of hardness, alkalinity, and silica in the water before RO treatment, increasing the risk of scaling. Operationally, the feed pressure increased from 11.43 ± 0.16 bar (old plan) to a peak of 25.50 ± 0.10 bar in the new plan, accompanied by a decline in water production. Chemical cleaning effectively restored performance, reducing feed pressure to 13.13 ± 0.05 bar, confirming that fouling and scaling were the primary, reversible causes. Despite these challenges, the plant consistently produced water that complied with Saudi Standards for Unbottled Drinking Water (e.g., pH = 7.18 ± 0.09, TDS = 978.27 ± 9.26 mg/L). Economically, the new strategy reduced operating expenditure by approximately 54% (0.295 → 0.135 $/m³), largely due to substantial reductions in chemical and sludge handling costs, although these savings were partially offset by higher energy consumption and more frequent membrane maintenance. Overall, the findings emphasize the importance of systematic performance evaluation during operational transitions, providing guidance for utilities seeking to optimize pretreatment design while maintaining compliance, long-term membrane protection, and environmental sustainability.

Research Highlights

This study assessed the technical, operational, and economic impacts of replacing lime and soda ash softening with an antiscalant-based pretreatment in a brackish groundwater RO desalination plant.

Water quality remained within the Saudi Standards for Unbottled Drinking Water under both treatment plans, despite higher feedwater hardness and silica levels in the antiscalant-based system.

Operational monitoring (2023–2025) showed that feed pressures and permeate TDS increased under the new plan, indicating greater scaling propensity and the need for closer maintenance.

Economic assessment showed a 54% reduction in OPEX (0.295 → 0.135 $/m³), primarily from savings in chemical dosing and sludge handling.

The environmental footprint was reduced through lower sludge generation and simplified chemical handling, though balanced by the potential requirement for more frequent membrane cleaning.

1. Introduction

Groundwater serves as a main source of potable water globally, especially in arid and semi-arid regions where surface water resources are limited, seasonal, or unreliable [1,2,3]. In these areas, such as vast parts of the Middle East, North Africa, and Central Asia, groundwater provides a more stable, drought-resilient supply for domestic, agricultural, and industrial needs [4,5]. However, groundwater in these regions is often brackish and chemically complex, increasing the need for advanced treatments before use. For example, a comprehensive study in central Saudi Arabia reported that approximately 59% of sampled wells contain brackish water with total dissolved solids (TDS) between 2000 and 10,000 mg/L; major ions such as Ca²⁺, Mg²⁺, Na⁺, and Cl⁻ were predominant and frequently exceeded drinking water standards, while trace elements like iron and manganese were also detected at elevated concentrations [6]. Similarly, groundwater in the Hail region of northwest Saudi Arabia showed mean TDS values from 258 to 775 mg/L, with total hardness ranging from 72 to 310 mg/L and nitrate concentrations up to 65.6 mg/L, highlighting the scale of the treatment challenge [7]. Moreover, elevated manganese and iron levels are concerning because of their neurotoxic effects, as well as their interference with disinfection and distribution processes [8].

Many groundwater treatment plants in arid countries have responded to these challenges by adopting a multi-stage treatment sequence consisting of aeration, sedimentation, filtration, and reverse osmosis (RO). In this configuration, the initial stages function as pretreatment processes that remove particulates, iron, and hardness-forming species, thereby protecting the downstream RO membranes, which perform the bulk of dissolved salts removal. However, pretreatment quality has a direct influence on RO performance, and poor conditioning may lead to membrane scaling, fouling, increased chemical demand, and more frequent cleaning [9,10,11,12]. Recent studies reported that inadequate removal of calcium, magnesium, silica, iron, and manganese during pretreatment directly correlates with shortened membrane lifespan, increased energy consumption, and frequent chemical cleaning cycles [13,14]. In this context, balancing effective treatment with operational sustainability becomes a core objective for water utilities.

Historically, many treatment plants have relied on lime (Ca(OH)₂) and soda ash (Na₂CO₃) addition during the sedimentation stage to soften water, control pH, and reduce the load of scale-forming ions such as calcium and magnesium before reaching RO membranes. While effective in controlling scaling, this strategy also introduces economic, logistical, and environmental concerns, particularly in terms of chemical storage, dosing control, high operational cost, and the generation of large volumes of sludge [15,16]. In many cases, the sludge contains elevated levels of precipitated metals and carbonate complexes, which adds further disposal challenges in dry regions. Some studies investigated reducing chemical dosing in the sedimentation basin and reported lower operating costs and less sludge generation; however, this strategy was reported to increase membrane fouling under high recovery conditions [17,18]. Recent membrane performance assessments support this trade-off. In particular, higher recovery rates above 80% correlated with increased fouling rates in brackish water desalination systems [19,20].

In recent years, there has been a notable shift toward “low-chemical” pretreatment strategies, emphasizing the use of advanced antiscalants as substitutes for bulk softening agents to reduce sludge generation and chemical consumption. These scaling inhibitors are designed to delay or inhibit the crystallization of sparingly soluble salts, such as calcium carbonate, calcium sulfate, barium sulfate, and silica, by several mechanisms, including ion sequestration, crystal lattice distortion, and particle dispersion [21]. Recent studies have found that novel antiscalants, including fluorescent-tagged phosphonate and polyacrylate formulations, can effectively inhibit the crystallization of sparingly soluble salts such as calcium carbonate, reducing crystal size and altering morphology, thereby mitigating RO fouling and improving operational stability [22,23]. Similarly, Shi et al. developed a novel nonionic starch-based antiscalant (St-g-GMA) that significantly reduced CaSO₄ scaling and flux decline in RO systems, with its performance attributed to chelation, dispersion, and lattice distortion mechanisms driven by grafted poly(GMA) chains [24]. These studies consistently demonstrate that even minor changes in operational design can significantly affect key performance indicators such as RO scaling, product water quality, membrane lifespan, and environmental compliance. However, there is limited research that investigates the combined effects of discontinuing lime/soda ash addition and changing the antiscalant on a system-wide scale, especially in full-scale groundwater desalination plants. This gap is especially critical in arid regions, where both sludge disposal and brine management represent operational and environmental challenges. A clear understanding of how operational changes influence both water treatment effectiveness and ecological impact is needed to guide future investments in treatment optimization.

The study provides a comparative assessment of the technical, operational, and economic impacts associated with eliminating lime and soda ash dosing and adopting an antiscalant-based approach in the pretreatment stage of a full-scale brackish groundwater RO plant. This transition represents a practical shift toward low-chemical operation, aiming to reduce sludge generation, simplify maintenance, and improve overall cost efficiency while maintaining stable RO performance. The evaluation covered key pretreatment, RO, and post-treatment stages, incorporating field observations and systematic sampling throughout the treatment line. Water samples collected from multiple locations were analyzed for pH, electrical conductivity, TDS, turbidity, and major ionic species, while operational data such as feed pressure, pressure drop, and water production rates were monitored in parallel. By comparing plant performance before and after the modification, the study highlights the direct influence of pretreatment quality on membrane scaling behavior, hydraulic stability, and economic sustainability. Overall, the results demonstrate that the new treatment plan maintained water quality comparable to the previous system while significantly reducing operational costs, particularly through lower chemical consumption and sludge handling. The study therefore contributes new applied evidence on the implementation and outcomes of low-chemical pretreatment strategies in groundwater desalination facilities, addressing both operational optimization and economic performance objectives.

2. Methodology

2.1. Study Area

This study was conducted at a groundwater desalination plant with an overall capacity of 60,000 m³/day, designed to produce potable water using integrated treatment processes. As the information is of a confidential nature, the name and location of the desalination plant cannot be disclosed. The treatment process flow diagram at the plant includes multiple pretreatment stages (including cooling and aeration, sedimentation, and rapid sand filtration), followed by an RO unit, which serves as the plant’s primary desalination and purification stage (Figure 1). The water source is moderately saline (brackish) groundwater with TDS ranging from 2000 to 2100 mg/L, extracted from deep aquifers at depths of 2000 to 2400 m below the ground surface. This depth suggests long groundwater residence times and potential for elevated concentrations of dissolved solids, hardness-forming ions, and trace elements.

2.2. Sample Collection

In this study, water samples were collected from multiple treatment stages within the plant and analyzed to evaluate the current water quality. In February 2024, the plant transitioned from the old treatment plan, which relied on lime and soda ash addition in the sedimentation basin, to the new treatment plan based on optimized antiscalant dosing. To ensure system stabilization and representative performance evaluation, no data were collected during the transition month (February 2024). Thus, data are divided into two operational periods: the old plan period (October 2023–January 2024) and the new plan period (March 2024–February 2025). These records were used to assess and compare the impact of the new operating strategy on the following key areas:

• Water quality following the sedimentation stage;
• The quality of water fed to the RO unit (particularly in terms of its compatibility with membrane design and operational specifications);
• Water quality after the RO unit;
• The final product water quality and its compliance with the Saudi Standards for Unbottled Drinking Water.

A minimum of five samples were collected and analyzed for water quality, and the average value was reported in this study. Samples were drawn from four critical locations within the treatment process flow diagram: (1) effluent from sedimentation units, (2) effluent from sand filters, (3) effluent from RO units, and (4) the final treated water produced by the plant (Figure 1). At each sampling location, approximately 2 L of water was collected in high-density polyethylene (HDPE) bottles that had been pre-rinsed with the sample water. Samples were immediately sealed and stored at 4 °C in insulated containers to minimize changes in physicochemical characteristics prior to laboratory analysis.

The primary focuses of this analysis were (1) the RO feedwater quality, as this unit is the most critical operationally and is directly affected by the performance of the adopted antiscalant, and (2) the quality of the final drinking water to assess any variations and ensure continued compliance with regulatory standards.

2.3. Water Quality Parameters and Analytical Tools

A range of physicochemical parameters was measured to assess water quality across the treatment process flow under both the previous and current operational conditions. The selected parameters were chosen based on their relevance to regulatory compliance, membrane performance, and operational efficiency. These included pH, TDS, electrical conductivity, total hardness, calcium and magnesium hardness, alkalinity, turbidity, silica, iron, chloride, sulfate, ammonia (NH₃), nitrite (NO₂⁻), and nitrate (NO₃⁻).

Each parameter was analyzed using standard, well-established methods. Specifically, pH and electrical conductivity were measured in situ using calibrated portable multi-parameter probes (Orion pH and conductivity multiparameter, Thermo-Fisher, Waltham, MA, USA). TDS was estimated from conductivity using a standard conversion factor. Total hardness and its calcium and magnesium fractions were measured via the EDTA titration method, while alkalinity was determined using titration with standardized 0.02 N sulfuric acid. Turbidity was measured using a nephelometric turbidity unit (NTU) meter (Hach 2100N, Loveland, CO, USA).

Silica concentrations were quantified using the molybdenum blue spectrophotometric method, and iron was analyzed using atomic absorption spectroscopy. Nitrogen species (NH₃, NO₂⁻, and NO₃⁻) were measured using colorimetric methods with UV-Vis spectrophotometry. Chloride and sulfate concentrations were determined via ion chromatography (IC 850 Professional, Metrohm, United Arab Emirates).

To estimate the risk of scaling due to calcium carbonate (CaCO₃) in the feedwater of the RO unit, the saturation index (SI) was determined using Equation (1) through Visual MINTEQ 3.1 software. This precipitate was selected as a model reaction due to the presence of high calcium and carbonate levels in the raw water. The SI calculation was performed under equilibrium conditions in a closed system with respect to CO₂, using the measured RO feedwater composition (Table 2) at 25 °C. A negative SI value is indicative of undersaturation with respect to CaCO₃ (no scaling risk), while a positive SI means that scaling of CaCO₃ is thermodynamically possible.

$$\mathrm{S}\mathrm{I}=\mathrm{l}\mathrm{o}\mathrm{g}({\displaystyle \frac{\mathrm{I}\mathrm{A}\mathrm{P}}{\mathrm{K}\mathrm{s}\mathrm{p}}})$$

(1)

where IAP is the ionic activity product of possible solids, and Ksp is the solubility product.

2.4. RO Membrane Performance

The RO unit consists of spiral-wound thin-film composite (TFC) polyamide membranes (TM720-400, Toray, Japan) housed in standard 8-inch pressure vessels configured for brackish water desalination. The performance of the RO unit was evaluated using operational data collected between October 2023 and February 2025. Key indicators included feed pressure, pressure drops across the three RO stages (ΔP1, ΔP2, ΔP3), product and reject flow rates, and permeate TDS. These parameters are widely recognized as reliable indicators of membrane condition, hydraulic stability, and desalination efficiency [25,26]. Changes in feed pressure and pressure drops across the RO stages were analyzed to identify signs of membrane fouling or scaling, while variations in product and reject flow rates were used to evaluate hydraulic efficiency and recovery. Permeate TDS was monitored as a direct measure of salt rejection and overall water quality. Together, these performance indicators enabled a consistent comparison of RO operation before and after the treatment plan shift and provided the basis for linking operational performance to the subsequent economic assessment.

2.5. Economic Assessment Calculations

An economic assessment was conducted to compare the operational expenditure (OPEX) of the treatment plant under the old and new treatment plans. Capital expenditure (CAPEX) was not considered, as the plant infrastructure remained unchanged during the shift in treatment strategy and because CAPEX is typically excluded in standard economic evaluations of brackish water RO (BWRO) facilities [27,28]. Instead, the analysis focused on recurring operational costs, including chemicals, energy, membrane replacement, sludge handling, and labor and maintenance, which directly reflect the impact of the treatment plan on cost efficiency.

Chemical costs were estimated using a combination of stoichiometric calculations and plant dosing records. For lime and soda ash, stoichiometric coefficients from the relevant softening reactions were used to calculate chemical consumption and costs (Equations (2)–(5)) [29]. In contrast, anticipants and chlorine dosing were taken directly from plant operation records and converted to costs using local market prices.

Ca(HCO₃)₂ + Ca(OH)₂ → 2 CaCO₃ ↓ + H₂O

(2)

Mg(HCO₃)₂ + 2 Ca(OH)₂ → 2 CaCO₃ ↓ + Mg(OH)₂ ↓ + 2 H₂O

(3)

CaSO₄ + Na₂CO₃ → CaCO₃ ↓ + Na₂SO₄

(4)

CaCl₂ + Na₂CO₃ → CaCO₃ ↓ + 2 NaCl

(5)

Sludge handling costs were then estimated from the mass of precipitated solids, expressed as wet cake with 80% water content [30]. Energy costs were estimated from measured feed pressures, using a pump efficiency of 80% to approximate specific energy consumption, Equation (6) [31,32,33].

$$\mathrm{E} \left({\displaystyle \frac{\mathrm{k}\mathrm{W}\mathrm{h}}{{\mathrm{m}}^{3 }}}\right)={\displaystyle \frac{\mathrm{P}}{\eta · 36}}$$

(6)

where P is the average feed pressure (bar), η is pump efficiency, and the constant converts bar to kWh/m³.

Membrane replacement costs were calculated using plant records on the number of installed elements, their replacement frequency, and actual unit prices, ensuring that this cost component reflects site-specific operational practices rather than generalized estimates. Finally, labor and routine maintenance costs were treated as a fixed value for both scenarios [34]. All cost components were normalized by the total volume of permeate produced to allow direct comparison between the two operational strategies, and the overall OPEX was then calculated as the sum of individual contributions using Equation (7):

$$\mathrm{O}\mathrm{P}\mathrm{E}\mathrm{X}\left({\displaystyle \frac{\mathrm{\$}}{{\mathrm{m}}^3}}\right)={\mathrm{C}\mathrm{o}\mathrm{s}\mathrm{t}}_{\mathrm{c}\mathrm{h}\mathrm{e}\mathrm{m}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{l}}+ {\mathrm{C}\mathrm{o}\mathrm{s}\mathrm{t}}_{\mathrm{e}\mathrm{n}\mathrm{e}\mathrm{r}\mathrm{g}\mathrm{y}}+ {\mathrm{C}\mathrm{o}\mathrm{s}\mathrm{t}}_{\mathrm{m}\mathrm{e}\mathrm{m}\mathrm{b}\mathrm{r}\mathrm{a}\mathrm{n}\mathrm{e}\mathrm{s}}+{\mathrm{C}\mathrm{o}\mathrm{s}\mathrm{t}}_{\mathrm{s}\mathrm{l}\mathrm{u}\mathrm{d}\mathrm{g}\mathrm{e}}+ {\mathrm{C}\mathrm{o}\mathrm{s}\mathrm{t}}_{\mathrm{l}\mathrm{a}\mathrm{b}\mathrm{o}\mathrm{r}/\mathrm{m}\mathrm{a}\mathrm{i}\mathrm{n}\mathrm{t}\mathrm{e}\mathrm{n}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{e}}$$

(7)

3. Results and Discussion

3.1. Water Quality Analysis After the Sedimentation Stage

A comparative analysis of water quality parameters measured at the sedimentation basin effluent under the old and new treatment plans showed substantial changes in key chemical characteristics (

Table 1. Water quality at the effluent of the sedimentation basin.

Parameter	Old	New
pH	9.94 ± 0.25	8.08 ± 0.59
TOT hardness (mg/L as CaCO3)	371.86 ± 9.75	821.56 ± 23.49
Ca hardness (mg/L as CaCO3)	202.46 ± 9.07	563.18 ± 19.09
Mg hardness (mg/L as CaCO3)	168.73 ± 7.25	213.56 ± 9.51
Silica (mg/L)	17.06 ± 2.30	37.23 ± 4.85
Iron (mg/L)	0.08 ± 0.00	0.18 ± 0.00
Alkalinity (mg/L as CaCO3)	43.86 ± 4.47	70.06 ± 14.31

). The most pronounced shift was observed in pH, which decreased significantly from 9.94 ± 0.25 in the old treatment plan to 8.08 ± 0.59 in the new system (p < 0.05). This reduction is directly attributed to the elimination of lime and soda ash dosing, which previously served to elevate pH and precipitate hardness like calcium carbonate and magnesium hydroxide [35]. Correspondingly, the total hardness increased more than double from 371.86 ± 9.75 to 821.56 ± 23.49 mg/L as CaCO₃, indicating that a greater fraction of calcium and magnesium remained in solution due to the absence of chemical softening.

In particular, calcium hardness increased from 202.46 ± 9.07 to 563.18 ± 19.09 mg/L as CaCO₃, and magnesium hardness increased from 168.73 ± 7.25 to 213.56 ± 9.51 mg/L as CaCO₃, confirming the diminished removal efficiency of these divalent cations in the new treatment system. In parallel, alkalinity also increased from 43.86 ± 4.47 to 70.06 ± 14.31 mg/L as CaCO₃, possibly due to the reduced consumption of carbonate species that would otherwise precipitate during lime/soda ash treatment. Moreover, silica concentrations increased more than double (17.06 ± 2.30 to 37.23 ± 4.85 mg/L), likely due to the lower pH, reducing the tendency of silica to polymerize or adsorb onto precipitated solids [36,37]. Additionally, iron concentrations increased from 0.08 ± 0.00 to 0.18 ± 0.00 mg/L, which may reflect reduced pH-dependent iron precipitation or destabilization of iron particulates under the new operating conditions [38]. Collectively, these results suggest that the new treatment approach, while operationally simpler and possibly more cost-effective, results in a more chemically complex water with elevated scaling risks, particularly for the downstream RO unit.

3.2. Water Quality Analysis After the Filtration Stage

Following the rapid sand filtration stage, where sulfuric acid was injected prior to filtration in both treatment schemes, some significant differences in water quality remained clear between the old and new operational plans. In the old system, acid dosing effectively reduced the pH to 6.34 ± 0.26, enhancing the removal of residual precipitates formed during upstream lime and soda ash treatment (

Table 2. Water quality at the effluent of the filtration basin.

Parameter	Old	New
pH	6.34 ± 0.26	6.99 ± 0.25
TDS (mg/L)	1919.16 ± 21.99	1810.31 ± 31.07
TOT hardness (mg/L as CaCO3)	363.97 ± 29.19	818.93 ± 11.55
Ca hardness (mg/L as CaCO3)	186.69 ± 14.66	562.17 ± 8.60
Mg hardness (mg/L as CaCO3)	162.16 ± 5.79	256.83 ± 12.99
Silica (mg/L)	15.94 ± 2.56	36.46 ± 5.31
Iron (mg/L)	0.02 ± 0.00	0.04 ± 0.00
Alkalinity (mg/L as CaCO3)	37.72 ± 4.19	68.09 ± 4.87

). The pH under the new treatment plan was slightly higher (6.99 ± 0.25), suggesting an increased buffering capacity and limited acid neutralization, likely due to the presence of elevated carbonate alkalinity in water. Moreover, despite similar TDS levels (1919.16 ± 21.99 mg/L in the old vs. 1810.31 ± 31.07 mg/L in the new system), hardness-related parameters remained significantly elevated under the new plan. Total hardness, for instance, increased from 363.97 ± 29.19 to 818.93 ± 11.55 mg/L, with both calcium and magnesium hardness showing similar trends compared to post-sedimentation analysis. These higher hardness levels observed in the pretreatment (which serves as the RO feedwater) under the new plan are a direct consequence of eliminating lime and soda ash softening. This indicates that the filtration process alone could not compensate for the lack of chemical precipitation in the upstream sedimentation unit in the new treatment plan.

In addition, the silica concentration remained high under the new operational plan (36.46 ± 5.31 mg/L vs. 15.94 ± 2.56 mg/L), likely due to reduced co-precipitation or adsorption with metal hydroxides at higher pH levels [36,37,39]. Although these concentrations remain within an acceptable operational range, it is critical to continuously monitor the silica level in the RO feedwater, given that silica scaling is challenging to clean and often irreversible using standard chemical cleaning procedures [40]. Continuous accumulation of silica, especially in the presence of elevated magnesium and high pH, can accelerate RO membrane fouling and deterioration, which in turn reduces the efficiency of long-term system desalination performance [41,42].

Moreover, the concentration of iron also increased from 0.02 ± 0.00 to 0.04 ± 0.00 mg/L, reflecting less effective removal or possibly partial solubilization under near-neutral pH. Similarly, alkalinity increased from 37.72 ± 4.19 mg/L to 68.09 ± 4.87 mg/L, suggesting less precipitation of carbonate species in the absence of lime/soda ash treatment. These water quality changes suggest that the new system, while avoiding chemical dosing costs, may lead to higher fouling or scaling risk in downstream membranes. As reported in previous studies, elevated hardness and silica are known contributors to membrane scaling and irreversible performance degradation in RO systems [43,44]. Therefore, while the new plan offers simplified operations, it introduces trade-offs that must be carefully managed to protect RO performance and extend membrane lifespan.

A comparative analysis of post-filtration water quality parameters showed a notable increase in the potential for membrane scaling under the new operational strategy. The marked increases in key chemical indicators in the RO feedwater (i.e., the total and calcium hardness, alkalinity, silica, and pH shown in Table 2), all of which are critical determinants of scaling risk in the RO unit of the new treatment plan. To quantify this risk, the SI was estimated for both treatment scenarios using the measured water quality parameters (Table 2). The old system showed a significantly negative SI (~−1.5), indicating undersaturation with respect to calcium carbonate and a minimal risk of scaling. In contrast, the new system yielded an SI of approximately +0.2, indicating slight supersaturation and a mild risk of calcium carbonate scaling. This increase is primarily due to the elimination of lime and soda ash dosing, which had previously removed a substantial portion of calcium and magnesium through precipitation.

In addition to the elevated scaling tendency indicated by the SI results, operational monitoring also revealed changes in the particulate and colloidal characteristics of the RO feedwater. The feedwater characterization showed that the Silt Density Index (SDI) increased from 1.5 to 2.5 under the old treatment plan to between 3 and 4 after adopting the new antiscalant-based pretreatment. This moderate rise, along with more frequent cartridge filter replacements, suggests greater colloidal loading and reduced filtration efficiency under the modified water chemistry. Optimizing the pretreatment configuration, particularly the sand and cartridge filters, would help maintain lower SDI values and minimize colloidal fouling risks, thereby improving the long-term stability of the RO system.

These findings align with previous research that highlights the critical role of pretreatment in minimizing fouling and scaling risks [15]. While the operational changes may offer cost benefits and simplify the process, the increased SI and SDI indicators suggest a need for careful evaluation of long-term RO performance and membrane protection strategies in the plant.

3.3. Performance of the RO Stage

3.3.1. Water Quality

Water quality analysis at the effluent of the RO unit was performed and showed that, despite significant differences in pretreatment chemistry and RO feedwater quality, the RO unit continued to produce high-quality water (

Table 3. Water quality at the effluent of the RO unit.

Parameter	Old	New
pH	7.07 ± 0.35	7.21 ± 0.11
TDS (mg/L)	79.83 ± 4.48	162.25 ± 17.36
TOT hardness (mg/L as CaCO3)	37.56 ± 2.07	80.43 ± 5.79
Ca hardness (mg/L as CaCO3)	21.93 ± 1.64	45.93 ± 1.68
Mg hardness (mg/L as CaCO3)	17.67± 2.07	29.07 ± 4.99
Alkalinity (mg/L as CaCO3)	26.38 ± 4.07	32.42 ± 2.59
Chlorine (mg/L)	0.29 ± 0.02	0.31 ± 0.03

). In particular, the pH of the water produced under both treatment plans was neutral and nearly identical (7.07 ± 0.35 for the old plan vs. 7.21 ± 0.11 for the new plan), which, based on statistical analysis, were not significantly different (p = 0.28).

Moreover, the RO membranes achieved substantial removals of TDS, hardness, and alkalinity in both cases. Under the new plan, the TDS level significantly dropped from 1810.31 ± 31.07 to 162.25 ± 17.36 mg/L, the total hardness was reduced from 818.93 ± 11.55 mg/L in the feed to 80.43 ± 5.79 mg/L in the product, while calcium hardness and alkalinity declined from 562.17 ± 8.60 mg/L to 45.93 ± 1.68 mg/L, and from 68.09 ± 4.87 to 32.42 ± 2.59 mg/L as CaCO₃, respectively. Although these values are still acceptable and within the Saudi Standards for Unbottled Drinking Water (SSUDW), they remain slightly greater than those obtained under the old plan (feed 1919.16 ± 21.99 → product 79.83 ± 4.48 mg/L for TDS, feed 363.97 ± 29.19 → product 37.56 ± 2.07 mg/L for total hardness; feed 186.69 ± 14.66 → product 21.93 ± 1.64 mg/L for calcium hardness; feed 37.72 ± 4.19 → product 26.38 ± 4.07 mg/L as CaCO₃ for alkalinity), indicating that the membranes maintained high rejection in the new treatment plan despite the increased scaling risks.

The RO membranes reliably remove hardness and alkalinity even at high feedwater concentrations, owing to the combined effect of antiscalant dosing and membrane separation. This shows that the membranes are robust and that the antiscalant effectively prevents scale formation during normal operation. Nevertheless, the slightly higher TDS and calcium hardness in the product water under the new plan suggest that prolonged operation without softening may increase the risk of localized scaling and salt passage [45,46]. This emphasizes the earlier observation that, although the new operation strategy simplifies pretreatment and potentially reduces chemical costs, it may require more frequent membrane maintenance and cleaning or optimized antiscalant dosing to sustain RO efficiency.

3.3.2. Feed Pressure and Hydraulic Behavior

Because the water quality after the RO suggested higher risks of scaling and fouling under the new treatment plan, we systematically evaluated the RO unit’s performance. Key indicators such as feed pressure, pressure drops across the three RO stages (ΔP1, ΔP2, ΔP3), water production rates, and permeate TDS were monitored from October 2023 to February 2025. These measurements provided insight into membrane condition, hydraulic stability, and desalination efficiency before and after the operational shift. The operational records showed clear differences between the two pretreatment strategies. Under the old treatment plan (October 2023–January 2024), when lime and soda ash were applied, the feed pressure remained relatively stable, ranging from 12.46 ± 0.15 to 12.77 ± 0.17 bar (Figure 2A). This stability indicates low fouling and scaling tendencies during this period, consistent with the chemical softening effect of lime and soda ash, which reduces hardness and alkalinity before membrane contact [45].

After the shift to the new treatment plan in February 2024, which eliminated lime and soda ash and relied solely on antiscalant addition, the system experienced gradual pressure increases. In particular, feed pressure increased by 127% (from 11.43 ± 0.16 bar in March 2024 to 25.50 ± 0.10 bar by September 2024), suggesting progressive scaling and/or fouling (Figure 2A). The increase was especially sharp between May and September, aligning with reduced water production rate (128.2 ± 1.31–114.67 ± 1.52 m³/h) (Figure 2C). Such behavior is characteristic of inorganic scaling, particularly calcium carbonate and silica, which can elevate feed pressure by reducing membrane permeability [18].

Following a chemical cleaning of the membranes in October 2024, a marked improvement in hydraulic performance was observed, as evidenced by the sharp decline in feed pressure from 26.36 ± 0.06 bar immediately before cleaning to 13.13 ± 0.05 bar by December. This confirms that fouling/scaling rather than irreversible membrane damage was the dominant cause of deterioration [46,47,48].

Pressure drops (ΔP) provide further evidence of fouling location and type. Under the old treatment strategy (October 2023–January 2024), ΔP across the three stages remained low and relatively stable (ΔP1 ≈ 1.5 ± 0.04 bar, ΔP2 ≈ 1.05 ± 0.34 bar, ΔP3 ≈ 2.1 ± 0.69 bar) (Figure 2B). The slightly higher ΔP in the third stage likely reflects concentration polarization and localized scaling in high-recovery membranes, as expected in brackish RO systems [49,50]. After the treatment shift in February 2024, the pressure profiles changed. ΔP1 increased progressively, reaching 2.67 ± 0.06 bar by June 2024, while ΔP2 and ΔP3 remained relatively stable. Interestingly, ΔP1 decreased temporarily between July and September 2024 despite rising feed pressure, suggesting that the dominant resistance during that period was more related to surface scaling rather than particulate or biofouling blockage within the pressure vessels [51].

The chemical cleaning in October 2024 restored ΔP across all stages to near-initial levels (ΔP1 ≈ 2.1 ± 0.05 bar, ΔP2 ≈ 0.87 ± 0.06 bar, ΔP3 ≈ 1.06 ± 0.02 bar), confirming that fouling was chemically removable. Nevertheless, ΔP1 and ΔP2 began rising again by January–February 2025, consistent with the gradual re-accumulation of deposits. These findings emphasize the importance of antiscalant optimization, as under-dosing or inappropriate formulations can fail to provide long-term protection [52,53].

Moreover, the average feed pressure increased by approximately 2.3 bar/month between March and September 2024, indicating a gradual decline in membrane permeability consistent with inorganic scaling. Considering the measured hardness and silica levels in the RO feedwater, this rate aligns with values reported for calcium carbonate and silica scaling in full-scale brackish RO plants operating at high recovery ratios [54]. The subsequent restoration of feed pressure to 13.1 bar after the chemical cleaning in October 2024 demonstrates that the fouling was largely reversible and inorganic in nature, as such deposits are typically removed by acid-based cleaning. The combined pattern of feed pressure, pressure-drop profiles, and post-cleaning recovery provides quantitative evidence of scaling-induced hydraulic resistance under the modified pretreatment plan, confirming that the pressure behavior observed in this study is consistent with previously reported carbonate- and silica-related scaling mechanisms.

Our results in permeate flow rates indicate that overall water production remained largely consistent after the transition to the antiscalant-based pretreatment (Figure 2C). During operation with lime–soda ash, the average permeate flow was approximately 149.9 ± 0.25 m³/h, whereas under the new treatment it averaged ~124.5 ± 14.54 m³/h. Month-to-month variations under the new plan reflect the same hydraulic trends discussed earlier: a temporary decline in production corresponded with the period of elevated feed pressure and increased ΔP (spring–summer 2024; Figure 2A,B), followed by recovery after the October cleaning when feed pressure normalized and ΔP stabilized.

We also calculated the water recovery achieved by the RO system using the recorded permeate and reject flow rates. On average, recovery was 85.9% under the lime and soda ash pretreatment and 83.0% under the antiscalant-based treatment, excluding the transitional month of March 2024. Maintaining comparable recovery levels is important for ensuring steady water output and minimizing specific energy consumption, as reductions in recovery directly increase feed volume and pumping requirements. The new pretreatment approach maintained water recovery within the same range as reported for full-scale BWRO systems using microfiltration and antiscalant dosing [55,56]. This consistency confirms that the simplified low-chemical pretreatment sustained desalination efficiency and effective scaling control, highlighting its suitability for long-term groundwater desalination operations.

3.4. Final Produced Water Quality and Regulatory Compliance

The final treated water collected from the storage tank under the new operational system was analyzed to verify compliance with SSUDW [57]. The pH of the final water product was 7.18 ± 0.09, within the SSUDW range of 6.5 to 8.5, which decreases the risks of corrosivity or scaling in the distribution system (

Table 4. Comparison of the final desalinated water quality with the Saudi Standards for Unbottled Drinking Water (SSUDW).

Parameter	Concentration	SSUDW [57]
pH	7.18 ± 0.09	6.5–8.5
TDS (mg/L)	978.27 ± 9.26	1000
Conductivity (μS/cm)	1387.33 ± 8.57	1500
Free chlorine residual (mg/L)	0.15 ± 0.00	0.5
Turbidity (NTU)	1.05 ± 0.08	5
Total hardness (mg/L as CaCO3)	447.66 ± 6.07	320
Ca hardness (mg/L as CaCO3)	280.33 ± 3.56	>30
Mg hardness (mg/L as CaCO3)	168.16 ± 10.38	Not regulated
Alkalinity (mg/L as CaCO3)	86.86 ± 2.60	Not regulated
Chloride (mg/L)	220.6 ± 12.50	250
Sulfate (mg/L)	210.67 ± 8.11	250
Ammonia (mg/L)	0.04 ± 0.00	0.5
Nitrite (mg/L as N)	0.007 ± 0.00	3
Nitrate (mg/L as N)	5.72 ± 0.08	50
Iron (mg/L)	0.03 ± 0.00	0.3

). Similarly, the measured TDS was 978.27 ± 9.26 mg/L, just below the 1000 mg/L threshold set by SSUDW, indicating that the plant under the modified treatment plan consistently produced potable water with acceptable salinity levels [58]. Electrical conductivity (1387.33 ± 8.57 µS cm⁻¹) was also well below the standard limit of 1500 µS/cm, further confirming effective salt rejection. Turbidity (1.05 ± 0.08 NTU) was lower than the 5 NTU guideline, reflecting efficient removal of particulates. Free chlorine residual was maintained at 0.15 ± 0.00 mg/L, within the recommended concentration of 0.5 mg/L, ensuring microbiological safety without exceeding taste and odor thresholds.

Anion concentrations also fell within acceptable ranges. Chloride and sulfate were measured at 220.6 ± 12.50 and 210.67 ± 8.11 mg/L, respectively, both well below the maximum contaminant level of 250 mg/L. Nitrogen species were effectively controlled, with nitrate (5.72 ± 0.08 mg/L) and nitrite (0.007 ± 0.00 mg/L) remaining far below the regulatory thresholds of 50 mg/L and 3 mg/L, respectively [59,60,61]. Ammonia was detected at trace levels (0.04 ± 0.00 mg/L), far below the 0.5 mg/L reference limit. Iron was similarly low at 0.03 ± 0.00 mg/L, an order of magnitude lower than the 0.3 mg/L standard limit, confirming the effectiveness of pretreatment and membrane removal of trace metals. Moreover, the total hardness in the finished water was high and exceeded the SSUDW of 320 mg. In particular, the level of total hardness was 447.66 ± 6.07 mg/L as CaCO₃, with calcium hardness at 280.33 ± 3.56 mg/L and magnesium hardness at 168.16 ± 10.38 mg/L. While these values classify the water as “very hard,” they pose no health risks and are within the range commonly reported for brackish groundwater desalination plants [62,63]. Alkalinity was 86.86 ± 2.60 mg/L as CaCO₃, a level that provides moderate buffering capacity while reducing the risk of post-treatment scaling.

In summary, the final produced water complied fully with SSUDW. Although the water remained relatively hard, this parameter is primarily aesthetic and does not compromise potability. The results confirm that, despite operational differences between pretreatment strategies, the plant consistently produced water of suitable quality for municipal distribution.

3.5. Economic Assessment

A comparative economic analysis was conducted to evaluate the operational expenditure (OPEX) of the groundwater treatment plant under the two treatment strategies: the old plan (lime–soda ash softening with RO) and the new plan (direct RO with antiscalant dosing). The OPEX was broken down into five major components (chemicals, energy, RO membrane replacement, sludge handling, and labor/maintenance) to identify the main cost drivers and to quantify the impact of the operational change (Figure 3). Under the old plan, the total OPEX was 0.295 $/m³, whereas the new plan reduced the total OPEX to 0.135 $/m³. This represents a 54.2% decrease in unit treatment cost, highlighting the significant economic benefit of transitioning to the new treatment strategy.

The most pronounced difference was observed in chemical costs and sludge handling. In the old plan, chemical dosing (lime and soda ash) accounted for 0.128 $/m³, while sludge management arising from softening reactions added a further 0.083 $/m³ (Figure 3). Together, these two categories comprised nearly 71% of the total OPEX. In contrast, the new plan effectively eliminated lime dosing and considerably reduced soda ash consumption, replacing them with a low-dose antiscalant. This shift lowered the chemical cost to 0.032 $/m³, while sludge handling costs dropped by more than 95% (to 0.004 $/m³). These reductions illustrate the dual economic and operational advantage of moving away from conventional chemical softening toward an antiscalant-based pretreatment scheme.

Energy consumption exhibited an opposite trend. The higher average RO feed pressure required under the new plan (17.7 bar vs. 12.7 bar in the old plan) increased the unit energy cost from 0.030 to 0.043 $/m³ (Figure 3). Although this represents a 39.7% rise in energy-related expenditure, the absolute increase was small compared with the savings in chemical and sludge management costs. Similarly, the cost of RO membranes increased from 0.001 to 0.004 $/m³ due to higher cleaning and replacement needs associated with the new operating conditions; however, these increases were more than offset by the major reductions in other categories. Moreover, the labor and routine maintenance were treated as a fixed contribution of 0.053 $/m³ in both scenarios for simplicity, consistent with the assumption that staffing levels and day-to-day maintenance requirements remain unchanged between the two treatment plans.

When compared with published benchmarks for brackish groundwater RO plants, the new treatment plan shows an operating cost of about 0.135 $/m³, which sits at the lower end of reported ranges (0.14–0.63 $/m³) [64,65]. This lower value is reasonable in light of the plant’s relatively low feedwater salinity, the elimination of sludge disposal costs, and the advantage of low energy tariffs on desalination plants locally (0.070 $/kWh) [66,67]. The OPEX of the old treatment plan, on the other hand, falls closer to the middle of reported values (0.295 $/m³), reflecting the added expense of chemical softening and sludge handling [68,69].

To assess the robustness of the reported 54.2% reduction in OPEX, a sensitivity analysis was performed by varying key cost parameters, including chemical prices, energy tariffs, and membrane replacement costs, by ±20% to represent typical market fluctuations. The total OPEX of the old treatment plan ranged between 0.270 and 0.321 $/m³, while that of the new plan ranged between 0.124 and 0.146 $/m³. These sensitivity ranges were derived directly from the baseline cost components presented in Figure 3. Chemical cost exhibited the highest sensitivity, followed by sludge handling cost, whereas membrane replacement cost had a comparatively minor effect. Even under the most unfavorable scenario, combining a 20% decrease in chemical cost for the old plan and a 20% increase in labor and maintenance cost for the new plan, the OPEX reduction remained at approximately 45.9% relative to the old lime and soda ash plan. These findings confirm that the economic advantage of the new antiscalant-based strategy remains substantial and robust against realistic market fluctuations.

Moreover, to illustrate the sensitivity of OPEX to maintenance activities, a simple scenario-based estimate was developed using the observed operational data. Increasing the membrane cleaning frequency from one to two or three cycles per year would raise the overall OPEX of the new treatment plan by only 3–7%, without altering the overall 54% cost reduction relative to the lime–soda-ash plan. This limited variation indicates that moderate increases in maintenance or cleaning frequency have minimal impact on the strong economic advantage of the low-chemical pretreatment strategy.

Overall, the economic assessment confirms that the new treatment plan delivers a more cost-efficient and sustainable operation. It is important to note that the shift toward simpler, low-chemical pretreatment systems, particularly those combining ultrafiltration and antiscalant dosing, has been widely reported to reduce chemical use, sludge generation, and overall operating costs in desalination plants [11,65,69]. The results of this study align with that direction, showing that replacing lime and soda ash softening with an antiscalant-based approach achieved comparable water quality at substantially lower OPEX. While the specific cost impact of membrane-based pretreatment options depends on local water chemistry and operational conditions [67], our findings collectively support the broader movement toward sustainable, low-chemical pretreatment optimization in groundwater RO desalination, where the greatest opportunities for cost savings often lie not in marginal energy reductions but in strategic decisions about pretreatment chemistry and sludge management.

4. Environmental Perspective

The shift in treatment strategy, particularly the discontinuation of lime and soda ash addition, has implications not only for water quality and operational efficiency but also for the environmental footprint of the treatment plant. Lime and soda ash are commonly used in water softening processes to remove hardness-causing ions through chemical precipitation. This operational change significantly reduces the volume and mass of generated sludge wastes (e.g., calcium carbonate and magnesium hydroxide), requiring less handling, dewatering, and disposal processes, thereby lowering the environmental burden associated with sludge management and transport [70]. Furthermore, eliminating these chemicals lowers the embedded energy and carbon emissions associated with their production, transportation, and handling, contributing to a more sustainable plant operation in line with green desalination strategies [71,72].

However, the new treatment approach may introduce other environmental trade-offs, particularly in relation to the characteristics of the reject stream from the RO unit. Higher concentrations of hardness ions and silica that are not removed in the pretreatment stages are ultimately rejected by the RO membranes and discharged as concentrate brine, often to nearby lagoons or surface discharge areas. This reject stream, now potentially more saturated with scale-forming ions and trace treatment chemicals, may pose a greater ecological risk to the receiving environment, especially in arid regions with limited natural dilution capacity [73,74]. If not properly managed, such brine disposal can lead to localized salinization of soil and groundwater, harming vegetation and aquatic ecosystems. Moreover, the observed increase in iron and silica concentrations measured in this study may raise concerns about potential long-term impacts on lagoon sediment chemistry and microbial processes. As desalination and brackish water treatment expand globally, the importance of integrated brine management and environmentally friendly pretreatment design becomes more critical. Therefore, future studies should incorporate detailed chemical modeling and ecological toxicity testing to evaluate the downstream effects of concentrate quality. They should also explore opportunities for brine minimization or reuse, particularly in inland groundwater desalination plants, to support the development of more sustainable reject disposal strategies.

5. Recommendations and Future Work

The outcomes of this study are consistent with broader operational experiences from Gulf-region desalination facilities, where transitioning toward “low-chemical” pretreatment strategies often involves balancing environmental gains with operational robustness [75,76]. By providing a rare, long-term operational dataset, this work illustrates both the opportunities and constraints associated with antiscalant-based approaches in groundwater RO systems. To support utilities in advancing more sustainable operations, the following recommendations are proposed:

• Investigate Seasonal and Hydrogeochemical Variability

Future research should explore how seasonal and hydrogeochemical fluctuations in groundwater sources influence scaling behaviour under modified pretreatment schemes. Long-duration pilot or plant-scale studies can help determine how changes in temperature, recharge cycles, and geochemical composition affect fouling and scaling risks. Integrating water quality monitoring with operational performance and economic analysis will enable utilities to make evidence-based decisions when evaluating alternative pretreatment configurations.

2.

Expand Temporal and Spatial Scope of Data Collection

Extending the study period across multiple years and incorporating datasets from different geographic and hydrogeologic settings would provide a more robust understanding of system resilience. Examining multi-seasonal variations in feedwater characteristics, such as temperature, recharge dynamics, and aquifer chemistry, can yield critical insights into the adaptability and stability of low-chemical pretreatment strategies under diverse operational conditions.

3.

Incorporate Comprehensive Economic Assessment

While the current study focused primarily on OPEX, future analyses should integrate CAPEX and membrane life-cycle data. Evaluating impacts on membrane aging, replacement intervals, and overall life-cycle costs will offer a more holistic cost–benefit perspective, enabling more informed decision-making for sustainable desalination operations.

4.

Optimize Pretreatment Configurations

The modified pretreatment approach, which includes aeration, sedimentation, and dual-media filtration, proved effective in supporting low-chemical operation by reducing lime and soda ash usage, minimizing sludge production, and lowering operational costs. Building on this success, future optimization efforts should explore membrane-based pretreatment technologies, such as microfiltration or ultrafiltration, which have demonstrated stable performance in full-scale brackish groundwater RO systems [77]. These approaches could further reduce chemical dependency and enhance process sustainability.

5.

Standardize Post-Treatment Blending and Remineralization

Consistent final water quality can be achieved by optimizing and standardizing post-treatment blending and remineralization practices. Establishing clear operational guidelines and implementing rigorous monitoring protocols will help maintain target TDS levels, improve product water stability, and minimize variations resulting from uncontrolled blending or salt addition.

6. Conclusions

This study evaluated the operational, technical, and economic impacts of transitioning a full-scale brackish groundwater desalination plant from lime–soda ash softening (old plan) to an antiscalant-based (new plan) pretreatment strategy upstream of reverse osmosis (RO). Eliminating lime and soda ash simplified operations, reduced chemical consumption, and lowered sludge generation while maintaining compliance with the Saudi Standards for Unbottled Drinking Water (SSUDW).

The transition introduced measurable trade-offs between reduced chemical use and increased scaling risks. The total hardness in the RO feedwater increased from 363.97 ± 29.19 mg/L (old plan) to 818.93 ± 11.55 mg/L (new plan) as CaCO₃, in response to the elimination of lime–soda ash softening; however, the RO permeate continued to show acceptable total hardness levels (<85 mg/L as CaCO₃) during normal operation. Feed pressure increased from 11.43 ± 0.16 to 25.50 ± 0.10 bar before cleaning and recovered to 13.13 ± 0.05 bar afterward, demonstrating that fouling and scaling effects were reversible. Water recovery remained stable between 83 and 85% over 16 months of operation, indicating that hydraulic efficiency was not compromised.

The economic assessment indicated that the new operational strategy reduced overall operating expenditure by approximately 54%, primarily due to savings in chemical use and sludge handling. These benefits outweighed moderate increases in energy demand and membrane maintenance costs, positioning the new system as a more cost-effective and environmentally sustainable option. From an environmental standpoint, the reduced chemical and sludge footprint represented a significant advantage, although the concentrate stream contained elevated levels of hardness, silica, and antiscalant residues that require further investigation to ensure responsible brine management.

Overall, the results emphasize that operational transitions in desalination plants must be evaluated holistically, balancing chemical reduction and cost savings against the RO membrane lifespan and environmental impacts.