The goal of this LCA is to quantify the environmental impact of SSA production at two recycling facilities in Abu Dhabi, UAE, and to compare it with the footprint of NA to inform its adoption in construction and corporate environmental reporting.

The functional unit is 1 tonne of SSA produced; all results are normalised to this unit. A cradle-to-gate system boundary is adopted (Figure 1), spanning slag-feedstock acquisition and transport (local and imported), processing (crushing, screening, and magnetic separation), internal handling and stockpiling, and ancillary operations (dust suppression, equipment operation, and maintenance-related transport). The process description was provided by Mawaad Environmental Services, which operates both plants.

All foreground data represent calendar year 2024 and were obtained directly from Mawaad Environmental Services. Diesel and engine oil consumption were taken from procurement and fuel-dispensing records; water use was taken from utility/abstraction records; annual SSA output and the slag-feedstock split were taken from production logs; and transport distances were derived from fixed origin–destination routes (supplier sites, Khalifa Port, and Musaffah) using road-network and nautical distances. No foreground quantity was estimated where a metered or recorded value existed; the only modeled quantities are the transport activity values (tonne-kilometers), computed from recorded masses and fixed route distances as described in Section 2.4. Background emission factors are secondary data from the (Department for Environment, 2024-10-30), IPCC defaults, and published literature.

To evaluate the robustness of the results, a one-at-a-time sensitivity analysis was conducted for the parameters expected to have the greatest influence on the combined SSA intensity: road transport distance, HGV payload/load factor, marine-freight emission factor, and diesel combustion emission factor. Each parameter was varied by ±25% relative to the baseline while holding the other parameters constant. The sensitivity analysis was performed on the combined cradle-to-gate intensity and is reported in Section 3.3 together with a qualitative data-quality assessment.

2.3. System Boundary, Allocation, and Justified Exclusions

The two facilities produce two outputs: SSA (the product under study) and recovered ferrous metal, which is returned to steel mills. A cut-off (recycled-content) approach was applied: the recovered metal leaves the system as a recyclable co-product. Under this approach, no upstream burden was assigned to the recovered metal beyond the system boundary, and no mass or economic allocation was applied to the SSA footprint. GWP, expressed in t CO₂eq, is the reported impact category, consistent with prior work identifying it as the dominant category for aggregate production (Alzard et al., 2021).

Multiple assumptions were made to conduct this study. These assumptions include:

All electricity consumed in the plant is generated on-site by diesel generators; no grid electricity is used.

Total SSA production of 500,000 tonnes is achieved through two facilities: Al Fayah and KEZAD B.

50% of processed slag feedstock processed at the Al Fayah plant is imported from Qatar via sea route, and 50% is sourced locally from Musaffah. Transportation impacts are considered only for inbound loads; return trips are excluded. Maintenance materials and diesel are sourced from Musaffah. The transportation distances considered are listed in Table 1. The maintenance of transportation vehicles is excluded. Transport of recovered metal scrap is excluded from this LCA.

Intergovernmental Panel on Climate Change (IPCC) and GHG Conversion Factors 2024 are used to calculate the impact of both transportation modes.

Return (empty) trips were excluded because the DEFRA “average laden” tonne-kilometer factors already embed fleet-average empty-running; separately adding back-hauls would double-count. This avoids potential double-counting when average-laden factors are used and is examined in the sensitivity analysis.

Transport of recovered metal was excluded, consistent with the cut-off treatment of the recovered co-product, whose downstream handling belongs to the steelmaking system.

Vehicle manufacture and maintenance were excluded as capital goods/secondary processes that are typically negligible relative to operational fuel and freight in cradle-to-gate aggregate LCAs.

Downstream distribution, the use phase, and end-of-life were excluded by definition of the cradle-to-gate boundary because they depend on application-specific decisions outside the producer’s control.

Throughout the Results and Discussion section, total GHG emissions are reported in t CO₂eq, emission intensities are reported in kg CO₂eq/tonne SSA, and freight emission factors are reported in kg CO₂eq/tonne-km. Under the emission factors described in Section 2.4.3, Al Fayah reports 307.75 t CO₂eq from materials, 1,700.07 t CO₂eq from inbound logistics, and 0.0632 t CO₂eq from wastewater treatment, for a total of 2,007.88 t CO₂eq and an intensity of 8.03 kg CO₂eq/tonne SSA. KEZAD B reports 307.75 t CO₂eq from materials, 953.43 t CO₂eq from inbound logistics, and 0.0632 t CO₂eq from wastewater treatment, for a total of 1,262.33 t CO₂eq and an intensity of 5.05 kg CO₂eq/tonne SSA. Combined, the two plants total 3,270.22 t CO₂eq for 500,000 tonnes of SSA, yielding an average intensity of 6.54 kg CO₂eq/tonne SSA. Table 5 presents a summary of the results.

The stacked bar chart presented in Figure 3 illustrates the dominance of transport over materials at both plants, with the effect being more pronounced at Al Fayah due to its longer road legs. The transport-by-mode presented in Figure 4 shows that road movements account for 88% of Al Fayah’s transport GWP and almost 59% at KEZAD B, reflecting each plant’s last-mile distance.

The calculations adhere to the ISO 14040/14044 structure for inventory compilation, ensuring that data, factors, and boundaries are transparent for subsequent impact assessment and interpretation. The Scope 3 Category 4 mapping and methods adhere to the GHG Protocol technical guidance.(Protocol, 2013). For market context, DEFRA (2024) lists NA at 7.75 kg CO₂eq/tonne. Benchmarking against this value, the modelled SSA intensity of 6.54 kg CO₂eq/tonne is roughly 16% lower on a cradle-to-gate basis under the same factor set.

Road freight is the primary driver of GWP. At Al Fayah, road transportation accounts for approximately 88% of transport emissions. At KEZAD B, the road accounts for about 59%. Marine shipments help reduce intensity, but the last-mile road legs still control the totals, especially where distances are longer. Scope 1 diesel for on-site generation is the largest materials-related source. Engine oil and water treatment are minor contributors in comparison. These findings are consistent with DEFRA factors and the GHG Protocol approach to upstream logistics.

The transportation contribution to the overall environmental impact can be reduced by shifting more inbound mass to sea where feasible, shortening last-mile distances through routing, and prioritizing higher-capacity vehicles that match the articulated HGV profile used in the model. The recycling plants‘ operating company can also work with carriers to validate actual vehicle classes and load factors and update tonne-km factors accordingly. These steps ensure that the methods align with the GHG Protocol’s Scope 3 calculation guidance and DEFRA documentation, while also improving representativeness. Scope 1 intensity could be reduced by improving generator efficiency or, where practical, integrating on-site renewable energy. Recording generator operating hours and average load would also allow the development of a kWh-based indicator for future audits.

The results are consistent with previous LCA studies showing that the environmental benefit of recycled or secondary aggregates depends strongly on system boundary, transport distance, and the avoided use of virgin quarry materials. A study by (Sousa et al., 2026) reported an emission intensity of 6.64 kg CO₂eq/tonne RA for recycled aggregate production, which is nearly identical to the SSA value reported in this study (Programme, 2022). This high similarity in the reported values is probably due to the additional processing (crushing, sorting) both aggregates require, potentially high diesel/electricity use, and the substantial transport. (Alzard et al., 2021) reported that replacing NA with RA in UAE concrete mixtures can improve environmental performance, although the magnitude of the benefit depends on the full concrete mix design and associated binder contributions.

Similarly, (Loureiro et al., 2022) highlighted the potential of steel slag and recycled concrete aggregates to reduce reliance on quarry materials in asphalt applications, while emphasizing that environmental performance is affected by processing and transport requirements. In the present study, the average SSA intensity of 6.54 kg CO₂eq/tonne SSA is below the DEFRA 2024 natural aggregate benchmark of 7.75 kg CO₂eq/tonne under the same factor set. However, the plant-level results show that this advantage is not uniform: KEZAD B performs better due to the shorter last-mile road distance, whereas Al Fayah has a higher intensity due to longer road transport. Therefore, the findings are broadly consistent with earlier recycled-aggregate studies but also show that regional logistics and facility location are decisive factors in determining whether SSA provides a lower-carbon aggregate option.

It is worth mentioning that steel slag is commonly managed as inert or construction-type waste when landfilled. DEFRA’s 2024 waste factors list inert mineral streams, such as aggregates, bricks, and concrete, at approximately 1.234 kg CO₂eq/tonne for landfill, reflecting site operations and negligible methane formation. Using this factor for steel slag, diverting 515,000 tonnes of steel waste from landfill in 2024 avoids approximately 635,510 kg CO₂eq, equal to 635.51 t CO₂eq. This avoided burden equals about 20% of the modeled SSA cradle-to-gate footprint of 3,270.22 t CO₂eq under the factor set used in this study. DEFRA cautions that waste factors are provided for accounting purposes and may not represent full comparative life-cycle outcomes; the estimate is therefore presented as a contextual indicator, not included in the main LCA result.

Recycling steel slag prevents material from entering landfills, recovers metal, and returns the mineral fraction to construction, thereby increasing resource efficiency and supporting a circular economy. The World Steel Association notes that co-product use prevents landfills and reduces CO₂ emissions.(Association, 2020). Substituting slag for natural aggregates also eases pressure on sand and gravel extraction, which the United Nations Environment Program links to erosion, biodiversity loss, and other impacts.(Programme, 2022). Studies on using by-products as aggregates report environmental gains from reduced virgin extraction and avoided disposal (Loureiro et al., 2022).

3.3. Sensitivity and Uncertainty Analysis

Because transportation accounted for approximately 81% of the combined cradle-to-gate footprint, a one-at-a-time (OAT) sensitivity analysis was conducted for the parameters with the greatest influence and uncertainty. The tested parameters were road transport distance, HGV payload/load factor, marine-freight emission factor, and diesel combustion emission factor. In each case, one parameter was varied while all other parameters were held constant at their baseline values. Continuous parameters were varied by ±25%. The resulting change in the combined baseline intensity of 6.54 kg CO₂eq/tonne SSA is reported in Table 6.

The results show that the combined SSA intensity is most sensitive to road transport distance and HGV payload/load assumptions, which is consistent with the dominance of last-mile road haulage in the inventory. In contrast, the marine freight and diesel combustion factors produce smaller changes in the combined result. The sensitivity analysis, therefore, confirms that the main uncertainty lies in freight assumptions, particularly road transport distance and loading conditions. Although the SSA intensity remains below the natural aggregate benchmark in several sensitivity cases, the margin narrows and may reverse under unfavorable road-freight assumptions. These findings reinforce the importance of obtaining carrier-specific data on actual payloads, routing, and return-trip conditions in future updates.

The qualitative data-quality assessment indicates good temporal representativeness because all foreground data refer to the 2024 reporting year, and good technological representativeness because the data were obtained from the two facilities under study. The main limitation is the geographical representativeness of the secondary transport emission factors, which are based on the UK Government 2024 dataset rather than UAE-specific freight factors. Accordingly, the results should be interpreted as representative of the stated facilities, year, boundary, and emission-factor set, rather than as a generic emission factor for all SSA production conditions.

A practical way forward centres on stronger primary data and standardized accounting. Future work should meter generator output and diesel consumption to express on-site energy as kg CO₂eq per kWh, which will improve traceability within the ISO 14040 framework. (International Organization for Standardization, 2020). Suppliers should contribute towards shifting the inbound logistics from generic distance-based factors to carrier data by actively engaging in capturing the actual vehicle classes, payloads, and routing from hauliers. Until robust UAE-specific factors are available, transport emissions can remain anchored to the UK Government 2024 conversion factors while documenting assumptions and data quality.(Department for Environment, 2024-10-30). Additionally, for logistics, adopting the Smart Freight Centre’s Global Logistics Emission Council Framework enables alignment with widely used freight reporting practices and eases integration of carrier data; accounting under the GHG Protocol remains the reference for boundary and category mapping.(Centre, 2025). These steps maintain consistency with ISO 14040/14044 while incrementally replacing secondary factors with primary measurements and verified activity data.

The results can support multiple decisions and disclosures. Projects and tenders can reference the quantified cradle-to-gate intensities when specifying aggregates, while scenario analysis explores route optimization, higher-capacity vehicles, and modal choices to reduce tonne-kilometres. Additionally, this study can be extended to a third-party verified Environmental Product Declaration (EPD) to communicate performance in the construction market. This pathway formalizes data quality control and comparability, positioning the product for green procurement. Periodic updates should refresh factor sets and recalculate intensities as new carrier data, operational metering, or local factors become available. Over time, uncertainty and sensitivity analyses can be expanded to quantify the influence of fleet mix, load factors, and seasonal operations, while keeping interpretation within the ISO four-phase structure. These actions convert this LCA from a one-off assessment into a living dataset that supports procurement, supplier engagement, and transparent reporting across reporting cycles.

The authors thank Mawaad Environmental Services for providing primary operational data and technical clarifications that enabled this study. The authors also acknowledge that work has been supported by Riga Technical University, the 1.1.1.9 Research application No 1.1.1.9/LZP/1/24/086 of the Activity "Post-doctoral Research".

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sector/ individuals

Not applicable.

The author(s) declare(s) that there is no competing interest.