Water electrolysis is an electrochemical method for water splitting aimed at producing green hydrogen through the application of electricity. The basic reaction of water electrolysis is as follows in Equation 1. H 2 O l + E l e c t r i c i t y   237.2  kJ  mol − 1 + H e a t 48.6  kJ  mol − 1 →   H 2 g + 1 2 O 2 g (1)

For the above equation, the theoretical thermodynamic cell voltage required to electrolyze water into hydrogen and oxygen at room temperature is 1.23 V. Experimental results indicate that the optimal cell voltage for effective water splitting is 1.48 V. Additional voltage is necessary to surmount the kinetics and ohmic resistance associated with the electrolyte and cell components of the electrolyzer.

For an ideal 100 MWh PEM electrolyzer, the hydrogen generation rate is 100,000 (kW)/39.39 (kWh/kg) = 2538.715 kg/h. Assuming a conversion efficiency of 75% (for the Siemens Silyzer 300), the hydrogen production rate is 2538.715 kg/h x 0.75 = 1904.036 kg/h. Equation 1 clearly states that 1 mol of H₂O is required to make 1 mol of H₂ gas. The molar mass of H₂ is about 2 g/mol. The molar mass of H₂O is about 18 g/mol. To make 1 kg of hydrogen, we require 9 kg of H₂O. Therefore, the amount of water needed to make 1904.036 kg/h hydrogen is (1904 kg/h) x 9 = 1714 kg/h H₂O. The amount of water required to make 1904 kg/h hydrogen is 1714 kg/h. H₂O The density of H₂O is assumed to be 997 kg/m³. Thus, the total volume of H₂O required = 17,140/997 = 17.19 m³/h (412.5 m³/day). Literature validation (Bellotti et al., 2019; Nguyen et al., 2019) indicates that a 75% 100 MW electrolyzer yields a hydrogen production rate of around 1900 kg/h.

The data collected from literature for the capital (CAPEX) and operational (OPEX) costs estimation for water treatment, PEM electrolyzers and power generation options are presented in Table 3, Table 4, 5, respectively.

The range of CAPEX for PEM electrolyzer in the literature depends on various factors like, operating conditions, high catalyst and membrane cost, manufacturing techniques, high ohmic resistance, and high overpotential of oxygen evolution reaction (Capuano et al., 2021; Escobar-Yonoff et al., 2021; Ayers et al., 2024; Nizami and Purwanto, 2022).

Table 3 indicates that CAPEX and OPEX values in the literature range from 880 to 1326 $/kW and from 1% to 5% of CAPEX $/kW, respectively. The electrical and exergy efficiencies of PEM range from 60% to 80% and from 56.34% to 72.5% in the literature (Luo et al., 2023; Sin et al., 2024; Wang et al., 2022; T. International Renewable Energy Agency, 2020; Incer-Valverde et al., 2023), respectively.

The reported capacity for RO (Table 4) ranges from 54 to 2760 m³/day, with CAPEX between $10,416 and $218,900, and OPEX between $0.50 and $0.232 per m³. Exergy efficiency varies from 4% to 27.3% (without renewable) (Al Ghamdi and Mustafa, 2022) and from 37.67% to 54% (with renewable energy) (Kupcu et al., 2023; El Mansouri et al., 2020).

The reported capacity for UF (Table 4) varies from 60 to 2800 m³/day, with CAPEX between $46,730 and $138,020, and OPEX ranging from $0.66 to $0.04 per m³. The efficiency and exergy efficiency range from 80% to 90% (Gonzalez‐Ca et al., 2023) and 75.47% (Kupcu et al., 2023), respectively. In this study, the first energy source investigated is the national grid. And natural gas-generated electricity costs $37/MW, bringing the total energy cost for a 100 MW electrolyzer to $3700 per hour.

Subsequently, solar energy is examined as a renewable energy source, with Table 5 presenting the CAPEX and OPEX values for solar power (PV, ST, PTC) as reported in the literature.

Exergy is the maximum useful work obtainable as a system moves to equilibrium with its environment, unlike energy, which is always conserved. Exergy analysis evaluates energy quality and identifies inefficiencies due to irreversibility’s, leading to entropy generation (Jorgensen and Fath, 2008; Rosen, 2004).

5 types of Exergies are majorly used for exergy analysis of a system (Chitsaz et al., 2019), they are:

• Physical Exergy: Exergy from temperature and pressure changes.

A PEM electrolyzer, used for hydrogen production, involves energy and exergy input to drive water electrolysis. The total energy demand includes electrical and thermal components. Exergy efficiency measures how effectively energy is converted to useful work, highlighting losses due to ohmic resistance, activation overpotentials, and heat dissipation (Ni et al., 2008). In our case we consider Kinetic and potential exergy to be negligible since we assume the electrolyzer is kept horizontally without any movement involved.

Exergy analysis helps pinpoint areas of inefficiency and guides improvements in system design and operation. It shows that while energy and exergy efficiencies are related, exergy analysis provides a more comprehensive view of energy quality and irreversibility, aiding the development of more efficient hydrogen production systems.

The environmental assessment for CO₂ emissions entails estimating greenhouse gas (GHG) emissions in CO₂ equivalents (CO₂e), accounting for all phases of a product or system’s life cycle. These stages often encompass manufacture, operation, and end-of-life phases. The scope of the environmental assessment is often organized around the three major stages of a product’s or system’s life cycle: (1) Direct emissions from owned or controlled sources (e.g., fuel burning transportation). (2) Indirect emissions result from the use of purchased power, steam, heating, and cooling. (3) Other indirect emissions, which encompass the full product value chain, such as supply chain emissions (for example, raw material extraction, transportation, and waste disposal).

The boundaries of the environmental assessment delineate the life cycle phases and the activities to be evaluated within each phase. These generally encompass manufacture, operation, and end-of-life phases. Manufacturing encompasses raw material extraction, material processing, assembly, and transportation. CO₂e emissions encompass energy utilized in manufacturing, alongside methane (CH₄) and nitrous oxide (N₂O) emissions from production processes, as well as emissions from energy-intensive machinery, manufacturing facilities, and transportation. During the operational phase, energy consumption associated with the typical use of the product or system, including electricity, fuel, or gas, is taken into account. The CO₂e emissions accounted for include those from fuel combustion, energy consumption resulting in indirect emissions (such as from power generation), and emissions from refrigerant leakage or other gases during utilization. The end-of-life phase encompasses decommissioning, recycling, and waste management. The CO₂e emissions accounted for include those from energy utilized in disposal or recycling, methane emissions from landfills, and emissions associated with trash transportation.

The CO₂ equivalent emissions from electricity generated by natural gas generally range from 0.4 to 0.6 kg of CO₂ equivalent per kilowatt-hour (kg CO₂e/kWh) (Hardisty et al., 2012). The CO₂ equivalent footprint of power generated from photovoltaic (PV) solar panels typically varies from 0.02 to 0.1 kg of CO₂ equivalent per kilowatt-hour (kg CO₂e/kWh) over the system’s lifespan (Peng et al., 2013). The carbon dioxide equivalent footprint of energy generated from concentrated solar power (CSP) systems and parabolic trough collectors (PTC) often varies from 0.03–0.1 and 0.03–0.09 kg of CO₂ equivalent per kilowatt-hour (kg CO₂e/kWh) over the system’s lifespan (Timilsina, 2021; Shahabuddin et al., 2021), respectively. The CO₂ equivalent footprint of RO and UF water treatment generally varies from 0.5 to 2.5 and 0.1–1 kg of CO₂ equivalent per cubic meter (kg CO₂e/m³) of processed water, respectively. The fluctuation within this range may be influenced by factors including the energy sources utilized for operation, system efficiency, and local conditions (Zhou et al., 2011).

Measuring sustainability performance and making decisions are crucial and difficult tasks. As a result, it is necessary to develop/implement a systematic approach that assures the ability to collect and abstract complex process operations, as well as provide straightforward analysis and communication using relevant sustainability indicators. Considering the aforementioned elements, this study proposes the conceptual framework as a powerful tool for any process scale, supporting the designer in getting a full performance evaluation. This technique has the ability and flexibility to be applied at various process sizes, such as equipment, process units, or the complete process, by allowing for direct comparison of viable alternative processes that target the same result but use different designs. Furthermore, designers and researchers can easily apply this approach to assess the sustainability performance following process changes.

Proper and sufficient indicators need to be selected (directly can be adapted or modified from the literature) or developed. This phase is very critical since the selected indicators require measuring sustainability performance and making decisions. For sustainability assessment, the classified indicators are used for the alternative proposed processes’ designs to determine process potential which is defined as “a metric’s maximum level of improvement, over a technology’s current state of the art, that can be attained by any conceivable design employing the same technology”. The ratings of sustainability indicators lack relevance without the provision or selection of reference states. At present, the majority of sustainability indicator outcomes are articulated in terms of total quantities of materials or energy utilized by the process or per unit of product. This technique for existing processes is implemented by comparing current outcomes with prior trials or by evaluating other procedures yielding the same product. Nonetheless, this technique will become more challenging when the analysis is conducted across varying process design scales and boundaries. A system for identifying and selecting a sustainability scale for each indicator, framed by two scenarios reflecting optimal (100% sustainability) and minimal (0% sustainability) targets, was presented to address this issue. This sustainability scale facilitates the conversion of any indicator score into a dimensionless format by employing the worst and best situations, as delineated in Equation 19. This equation facilitates the visualization and comparison of sustainability assessment outcomes for each indicator across the four domains: Energy, Efficiency, Environment, and Economic.

Supplementary Tables 1–4 in the Supplementary Material defines the sustainability indicators—efficiency, energy, environmental, and economic—used in this study. I n d i c a t o r   m e a s u r e m e n t   i , % = A c t u a l   S c o r e   − W o r s t   S c o r e B e s t   S c o r e − W o r s t   S c o r e   x   100 % (19)

Table 6; Figure 7 indicate that PEM, ST, and PTC technologies necessitate significant capital investment, with worst-case and best-case scenarios ranging from 132.63 to 88.03 million dollars, 732.521 to 646.76 million dollars, and 655.7 to 646.1 million dollars, respectively. In the comparison of the eight cases (Figure 7), Case 1 and Case 5 exhibit the lowest capital cost investment, both averaging approximately 110 million dollars in CAPEX. This is attributed to their reliance on grid electricity, eliminating the need for direct capital investment. Cases 3, 4, 7, and 8 demonstrate the highest capital investment due to incorporating technologies requiring significant capital, specifically PEM, ST, and PTC, with an average CAPEX value of approximately 750 million dollars. CAPEX values for concentrated solar power cases are relatively higher than those for photovoltaic cases (Approximately four–4.4 times).

Table 6; Figure 8 indicate that PEM and PV technologies exhibit the highest operational costs, with worst-case and best-case scenarios ranging from 6.63 to 0.88 million dollars and 1.3 to 0.625 million dollars, respectively. Operating expenses (Figure 8) for photovoltaic cases are significantly higher than those for concentrated solar power. CAPEX and OPEX remain unaffected by the type of water treatment employed, whether RO or ultrafiltration (Figures 7, 8).

Table 6; Figure 9 present the calculation of CO₂ equivalent emissions per kg of H₂ produced via PEM, derived by aggregating the contributions from the energy source and the water treatment method. This calculation includes factors such as direct emissions from combustion and indirect emissions associated with the extraction, processing, and transportation of natural gas, as well as the manufacturing, installation, operation, and decommissioning of solar panels, operational efficiency, and local conditions. Substituting grid-based energy with PV, ST, and PTC technologies significantly reduces the kg CO₂e/kg H₂ (appromitaly averaged %88).

This study assesses the sustainability of the Siemens Silyzer 300 PEM (100 MWh) with an emphasis on renewable energy sources. The first step is to calculate the exergy efficiency of the aforementioned electrolyzer to use in the energy efficiency calculation regarding its impact on overall sustainability. Then, the overall sustainability assessment results are represented in the following paragraphs as explained in the previous section (3.5. Sustainability Assessment).

For the base case, the hydrogen production rate is 1904 kg/h or 944.5 kmol/h, the electrical efficiency of the electrolyzer is 75%, and the water temperature is 75^°C.

The rate of energy content of H₂ stream: E x , H 2 = m ˙ e x p h , H 2 + e x c h , H 2 (20) e x p h , H 2 = h − h 0 − T 0 s − s 0 = 4650.4 − 3931.7 k J k g − 298.15   K 55.657 − 53.429 k J k g · K = 54.4218 k J k g = = 109.7143488 k J k m o l e x c h , H 2 = 1 * 236090 J m o l + 8.314 J m o l * K * 298.15 K * ⁡ ln ⁡ 1 = 236090 J m o l E x , H 2 = m ˙ e x p h , 2 + e x c h , 2 = 944.4625784   kmol h * 109.7143488 + 236090 k J k m o l = 223081791.3 k J h

The rate of electric energy input: E e l e c t r i c = 100 M W = 100000 k J s * 3600 s h = 360000000 k J h

The rate of thermal energy input for the PEM electrolyzer: E h e a t , P E M ≈ 0

The rate of thermal energy input to the heat exchanger for heating up the H₂O from 25^°C to 75^°C; E h e a t , H 2 O = 944462.5784   mol h * 75.52662 J m o l * K * 348.15 − 298.15 K = 3566603.313 kJ h

The exergy efficiency: η e x = E H 2 × N ˙ H 2 , o u t E e l e c t r i c + E h e a t , P E M + E h e a t , H 2 O = 223081791.3 k J h 360000000 k J h + 0 + 3566603.313 kJ h = 0.6136

Figure 10 illustrates the sustainability assessment for the integration of PEM, RO/UF, and a renewable energy source, considering material efficiency, energy efficiency, environmental impact, and economics. The efficiency of a process or unit operation is indicated by the quantity of materials and services needed to produce the desired product or accomplish a specific task. Mass transfer operations significantly affect energy demand, equipment size, costs, raw materials, and emissions. Efficiency indicators offer sustainability assessments that are valuable for identifying opportunities in process design during the conceptual stages, significantly impacting all aspects of process sustainability. The integrated process exhibits an average material efficiency of 94.866%, which can be deemed successful concerning sustainability objectives.

The primary significance of this study is to suggest an integrated method that mitigates adverse environmental impacts, including global warming. The attainment of these environmental impact minimization goals should start from the process input, such as considering the nature of the raw material and energy sources. Figure 10 illustrates that the suggested process integration achieves an average environmental efficiency of 98.34% through the utilization of renewable energy sources. Utilizing grid-based energy in the process diminishes the average environmental efficiency to 62.59% due to its significant global warming potential.

The demand for energy significantly impacts the sustainability performance of a chemical process or equipment, as shown in overall product costs, energy goods and services, and heat emissions. Like other indicators of process sustainability, energy indicators must be scientifically valid, easily calculable, and consistent, as the computation of thermodynamic characteristics relies on many data and reference states that should be accessible, particularly for novel chemical processes. The average energy sustainability is 83.06% due to an exergy efficiency of roughly 61%.

Enhancing PEM efficiency can augment the average energy sustainability. Improving the efficiency of PEM electrolyzers is essential to reducing the cost of green hydrogen and expanding its role in the energy transition. Key strategies and technologies such as advanced catalyst materials, improved membrane materials, enhanced electrode structures, optimized operating conditions, innovative system designs and integrating artificial intelligence and machine learning are implemented to enhance their efficiency. Catalysts are crucial for improving reaction kinetics in PEM electrolyzers. Current advancements focus on iridium-based catalysts, non-noble metal catalysts and core-shell structures (Carmo et al., 2013). Membrane properties significantly influence the efficiency and durability of PEM electrolyzers. Advances in perfluorosulfonic acid (PFSA) membranes have improved proton conductivity while maintaining mechanical and chemical stability. Hybrid structures combining PFSA with additives such as inorganic oxides enhance conductivity, reduce gas crossover, and improve thermal stability. Efficient electrode design improves the utilization of catalyst materials and facilitates better mass transport. Catalyst-coated membranes (CCMs) and porous transport layers (PTLs) are primary focal points (Haider et al., 2021). Improvements in system-level design, including modular systems and system heat recovery, enhance the overall efficiency and scalability of PEM electrolyzers (Lange et al., 2024). AI tools are increasingly being used to optimize PEM electrolyzer performance: Predictive maintenance analyzes operational data to predict and prevent failures and machine learning models identify optimal operating conditions to minimize energy consumption (Lange et al., 2024).

Finally, modifying operational parameters to optimize efficiency while ensuring durability involves dynamic operation and elevated operating temperatures. Especially, optimizing its operational temperature and integrating waste heat recovery are crucial for improving its efficiency and overall system performance. The operational temperature of PEM electrolyzers has an effect on reaction kinetics and membrane conductivity. Higher temperatures improve reaction kinetics by reducing activation energy barriers for the hydrogen evolution reaction (HER) and oxygen evolution reaction (OER). This leads to increased electrochemical performance and lower overpotentials. Proton conductivity of the PEM increases with temperature, reducing ohmic losses. However, excessive temperatures can degrade the membrane and other components, such as the catalyst layer and gas diffusion layers. PEM electrolyzers typically operate within a range of 50°C–80°C. Balancing temperature within this range ensures enhanced performance without compromising the durability of components. However, Future research focuses on developing high-temperature membranes that operate efficiently above 80°C, enhancing the potential for waste heat utilization. Advanced thermal management systems using sensors and controllers are deployed to maintain optimal operating temperatures and minimize fluctuations during variable power inputs, especially when integrated with renewable energy sources (Lange et al., 2024; Mao et al., 2024). PEM electrolyzers generate waste heat during operation, which can be recovered and utilized to improve overall energy efficiency. Electrochemical reactions in the cell produce heat as a byproduct and heat losses occur through electrical resistance in the membrane, electrodes, and external circuitry. Waste heat can be used to preheat the feedwater for electrolysis, reducing the energy required to reach operational temperatures. Heat recovery techniques such as heat exchangers and thermal storage systems can be used. Heat exchangers transfer waste heat from the electrolyzer stack or coolant systems to a secondary process. Captured heat can be stored in phase-change materials or insulated tanks for later use, enabling more flexible operation (Carmo et al., 2013; Chen Y. et al., 2023).

Nevertheless, the most challenging aspect of the proposed integrated process is the economic factor due to having high production costs (the current estimate for the cost of producing hydrogen via electrolysis is between $5 and $6 per kilogram). To achieve cost parity with steam methane reforming (SMR) at all capacities involves lowering the cost of hydrogen production via electrolysis to between $2.08 and $2.27 per kg. We need to further study profitability analysis for green hydrogen production from wastewater at varying scale that considers the generated income from oxygen production as well as other costs such infrastructure, conditioning and storage. Such detailed study can quantify the actual production cost of green hydrogen for the integrated system. Currently, the average economic sustainability of the process is approximately 57%. Reducing the production cost to $2.08 and $2.27 per kg can enhances sustainability to approximately 89.65%.

a) CSP scenarios show the highest CAPEX, approximately 4 times higher than PV cases.

a) The integrated process achieved high average material efficiency (94.87%) and high average environmental efficiency (98.34%) when using renewable energy, as opposed to lower environmental efficiency when using grid electricity(62.59%).

a) Hydrogen production via electrolysis currently costs $5 to $6 per kg, which is higher than SMR.

These findings emphasize the need for further research to improve efficiency and reduce the costs of the integrated system. With advancements in technology, green hydrogen can be a cornerstone in the transition toward sustainable energy systems, especially in regions like GCC that are rich in solar energy but limited in fresh water. Also, the conclusions highlight that although the integration of PEM electrolyzers, solar energy, and wastewater systems presents significant opportunities for clean energy and resource recovery, the scalability of each technology constrains the validity of the findings. The results and conclusions are valid for specific ranges of scale and may not be generalized to very small or very large systems without further optimization studies.