Performance evaluation of a reverse osmosis membrane desalination equipment for application in rural communities in the Brazilian semi-arid region Open Access

Q_f

feed flow (L/h)

Q_p

permeate flow (L/h)

RR

recovery rate (%)

SPR

salt permeation rate (%)

SRR

salt removal rate (%)

TDS_c

total dissolved solids in the concentrate (mg/L)

TDS_f

total dissolved solids in the feed flow (mg/L)

TDS_p

total dissolved solids in the permeate (mg/L)

The increasing demand for water, driven by population growth, rising individual consumption, and economic development, coupled with a declining water supply due to climate change and pollution, has intensified water scarcity across much of the world (Richter et al. 2013; Djuma et al. 2016; Damania et al. 2017). Recent studies estimate that approximately 40% of the global population faces severe water scarcity, a figure projected to rise to 60% by 2025 (Schewe et al. 2014). Additionally, around 66% of the global population – equivalent to 4 billion people – currently experiences severe water scarcity for at least 1 month annually (Mekonnen Hoekstra 2016). These findings emphasize that traditional water sources, such as rain, snowmelt, and river runoff stored in lakes, rivers, and aquifers, are increasingly insufficient to meet human needs, particularly in regions afflicted by chronic water scarcity.

Countries affected by water scarcity must fundamentally reconsider their water resource planning and management. Current challenges, combined with the impacts of climate change, are expected to exacerbate water shortages, making unconventional water sources essential for supporting vulnerable populations (Trancoso et al. 2024). This necessitates innovative approaches to exploring viable yet unconventional water sources for diverse applications, including potable water supply, livelihoods, ecosystem conservation, climate change adaptation, and sustainable development (Qadir 2018). Although methodologies for water resource conservation and augmentation have been implemented in some water-scarce regions, their expansion is crucial, particularly in areas where water scarcity and quality degradation are worsening (van Vliet et al. 2017; Jones & van Vliet 2018).

In Brazil, the semi-arid region in the northeast exemplifies persistent water scarcity challenges. Despite Brazil's abundant freshwater resources, this region accounts for only 3% of the national supply yet sustains the country's second-largest population (Haguenauer et al. 2019). Furthermore, groundwater in this area is highly mineralized, with elevated concentrations of sodium, magnesium, and chloride, rendering it unsuitable for human consumption (Gurgel et al. 2024). These conditions exacerbate water scarcity, leaving many without access to clean water for basic needs and economic activities, thereby undermining social and economic sustainability.

To address these challenges, the Programa Água Doce (PAD), established in 2004 by the Federal Government, aims to provide sustainable access to quality water for human consumption through the utilization of groundwater desalination systems. This initiative incorporates technical, environmental, and social considerations in its implementation and management. By transforming brackish water into potable water, these systems offer a reliable and sustainable solution for local communities (MIDR 2020). The PAD predominantly operates in the Brazilian semi-arid region, which is consistently threatened by water scarcity due to limited rainfall and salinization of crystalline aquifers (Kreis et al. 2024). These aquifers serve as the primary source of brackish groundwater for communities equipped with desalination systems.

In the state of Paraíba, the PAD currently operates 101 desalination systems, providing water for domestic consumption, household activities, and livestock use. These systems benefit approximately 60,000 people directly and indirectly, in addition to over 400 public institutions, including associations, schools, healthcare facilities, and municipalities (Government of Paraíba 2023). The deployment of desalination technologies in rural communities of the northeastern semi-arid region has demonstrated success in improving water access (Haguenauer et al. 2019). Communities benefiting from these systems have shown greater resilience to water crises, reducing migration caused by water scarcity (Alves et al. 2015). Expanding the use of desalination technology and brackish water reserves could mitigate water shortages in the region. However, further experimental studies are required to optimize system performance and determine optimal operational parameters (Wang et al. 2024).

Efforts to enhance the efficiency of desalination systems have focused on determining optimal parameters for complex setups and general scenarios. Optimizing desalination efficiency remains critical for advancing this technology (Atia et al. 2023).

In this context, the present study aims to evaluate the performance of a membrane-based desalination system for brackish water treatment. The performance profile seeks to identify optimal operating conditions to facilitate system standardization and implementation in small rural communities of the semi-arid region. The system was tested under controlled conditions using well water sourced from a local reservoir to simulate real-life applications. The primary objective was to assess the system's viability and define specific operational parameters for implementation by local governments or private entities, establishing a framework for sustainable water resource management.

The entire analysis and research process was conducted at the Desalination Technologies Testing Center (CTTD), utilizing locally available measurement equipment and infrastructure. The CTTD is part of the Laboratory of Reference in Desalination (LABDES), affiliated with the Federal University of Campina Grande (Universidade Federal de Campina Grande, UFCG).

The pretreatment process began with the filtration of feedwater through a disc filter with a 130 μm mesh, effectively removing macroscopic particulates. Subsequently, the water was subjected to further treatment using a hollow fiber polyamide ultrafiltration membrane with a filtration surface area of 30 m². This stage was designed to eliminate microscopic particulate matter and microorganisms. The ultrafiltration process was operated in a dead-end configuration with scheduled backwash cycles every 40 min, allowing a complete operational cycle of 2 h per run.

Following the feedwater pretreatment process, the system was supplied with pretreated feedwater using a multi-stage high-pressure pump with a power capacity of 1 HP. This pump delivered the water to two polyamide reverse osmosis membranes, each with a surface area of 8.4 m², operating in series to carry out the desalination process.

Samples of concentrate water, feedwater, and permeate water were collected during the process for subsequent physicochemical analysis. The concentrate was stored in a separate tank to undergo further desalination in subsequent experimental runs.

The operating conditions for the membrane system were determined using the Winflows 4.03 software (Veolia 2023). These conditions were established by setting the expected permeate flow at 25% of the feed flow under initial conditions during the first experimental run (R1), as shown in Table 1. The parameters defined in this initial setup guided the selection of the pressure pump and served as the baseline for comparisons in subsequent experimental runs. Feed pressure, feed flow, and other operational parameters were consistently maintained across all runs to ensure uniformity.

The initial feedwater was sourced from a local well, characterized as brackish water with a total dissolved solids (TDS) concentration of 1,755.62 mg/L, calculated at the time of the experiment. For all subsequent runs, the feedwater consisted of the concentrate generated in the previous run, with TDS values determined through laboratory analysis.

The simulated water composition was determined through physicochemical analysis. Operating conditions were set to a feed flow rate of 2,000 L/h and an operating pressure of 7.1 kgf/m². Tests were conducted over a 120-min period, during which data were collected for subsequent sample analysis.

The concentrates recovered during the process executions were measured in terms of TDS for application at each stage of the run, as presented in Table 1.

All analyses were conducted in triplicate for each test batch, and the samples were analyzed for TDS in the permeate, concentrate, and feedwater. Using the TDS data, the parameters for the salt removal rate (SRR) and salt permeation rate (SPR) were calculated according to Equations (1) and (2).

To analyze the average flow rates produced by the equipment, repeated measurements were taken at fixed intervals of 15 min, resulting in a total of eight flow measurement results per run. These results were compiled to calculate an average flow rate, which was subsequently used to determine the system's average recovery rate (RR) as a percentage, using Equation (3).

At the conclusion of the process, permeate TDS values and permeate flow rates were regressed as functions of feedwater TDS to estimate process limits. The regression functions were used to determine the limit point for permeate quality, in compliance with standard MS 888/21, as well as the permeate flow rate at this defined point. These regressions were performed using Microsoft Excel, applying the least squared errors method to compare the experimental data. The selection between power and linear regression models was based on the highest determination coefficient (R²) value calculated for each model.

The increase in feedwater TDS (TDS_f) during successive runs directly impacted the average production of the desalination system. As TDS_f increased, the osmotic pressure required to produce permeate also increased. Since permeate flow is directly influenced by osmotic pressure, as explained by Ismail et al. (2019), a decline in production was observed. This decline was measured to assess the system's capacity under varying conditions, aiming to evaluate its applicability in real-world scenarios.

Additionally, data on energy consumption per cubic meter of freshwater produced were collected to estimate energy expenditure in kilowatt-hours (kWh) and associated costs. The energy cost was calculated using the local electricity tariff of 0.67 BRL/kWh, equivalent to 0.12 USD/kWh based on the July 7, 2024 conversion rate (ANEEL 2023).

The observed data indicates a sharp decline in salt removal performance beginning at 3,000 mg/L of TDS. This increase results in a significant rise in permeate TDS, from 20.70 mg/L in R1 to 301.7 mg/L in R5. While these values remain within the permissible limits established by MS 888/21, the decline in performance highlights a pronounced diminishing trend in salt removal efficiency.

Considering the analysis of flow rate variations, a significant decrease in permeate production is observed with increasing feed TDS. This reduction is expected due to the rise in osmotic pressure of the feed, which is consistent with the literature on membrane desalination. The increase in osmotic pressure reduces permeate flow, as noted by Ismail et al. (2019). The data collected from the experimental runs are summarized in Table 3.

Regarding the expected operational costs of the equipment, water desalination costs increase with higher feed TDS (Table 4). Despite this rise, the cost of water production remains very low, suggesting that operational expenses should not pose a significant barrier to implementing desalination for populations in vulnerable situations.

According to the World Health Organization, basic human needs require a daily water volume of 50–100 L per person (WHO 2011). Using this metric, in the most expensive scenario analyzed, with a cost of USD 1.10/m³, the expense for providing water would range from USD 0.055 to USD 0.11 per person per day. This value is negligible for a governmental institution, making the equipment cost-effective for implementation in all cases studied.

In summary, the system offers an economical alternative for freshwater production with low energy costs, particularly when operating within the maximum recommended parameters, below 4,000 mg/L. Under these conditions, the reverse osmosis membrane maintains a salt rejection rate (SRR%) above 95%, reliably producing fresh water within a safe margin. However, at lower TDS levels, the system's cost-effectiveness diminishes, as the cost per cubic meter of fresh water increases by 86%, from USD 0.59 to USD 1.10 between the first and last measured runs. These findings indicate that the original operating conditions are optimal for the equipment, whereas higher salinity levels in the feedwater can significantly raise operational costs.

In conclusion, for the operation of this system in rural areas or small communities, this type of equipment can be used on a large scale, provided that the feedwater does not exceed the values determined in this article.

For brackish waters with TDS higher than 4,000 mg/L, this system is not recommended, requiring a higher-power pump to improve the SRR and the system recovery rate.

For use in well water with TDS up to 4,000 mg/L, the system may be suitable as it operates under optimal conditions. During periods of scarcity, the system can operate with concentrate recycling, even under non-ideal conditions, to mitigate water shortages in affected regions.

Overall, the evaluated system can be operated using well water with TDS levels up to 4,000 mg/L. It is not recommended for TDS levels higher than 7,500 mg/L due to operational inefficiencies, but it can still be managed, if necessary, under such conditions. It is not usable for desalination of well water with TDS levels exceeding 7,500 mg/L.

This study was conducted by the LABDES technical and academic team, supported by the UFCG. We acknowledge the MCTI, Brazilian Ministry for Science Technology and Innovation, who supported the creation of the CTTD.

O.N. involved in data curation, formal analysis, and investigation; M.R. involved in investigation and data curation; B.S. involved in writing – original draft, resources, and visualization; I.B. involved in writing – review and editing; N.A. involved in data curation; and K.F. involved in project administration, methodology, and conceptualization.

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

The authors declare there is no conflict of interest.