A strategic decarbonization roadmap for a petrochemical refinery in Vietnam: integration of CCUS, green hydrogen, and the circular economy

Abstract

The decarbonization of energy-intensive industries, particularly in petrochemical refineries, is critical to achieving global net-zero targets. The objective of this study was to develop a structured methodological framework for refinery decarbonization planning under confined industrial data availability. This approach integrates Scope 1 greenhouse gas inventory, emission hotspot identification, technology assessment, and scenario-based transition analysis. A Tier 1 methodology based on the Intergovernmental Panel on Climate Change (IPCC) 2006 Guidelines was applied to quantify direct CO₂ emitted from major process units of a representative Petrochemical Refinery Y in Vietnam. The results identified the residue fluid catalytic cracking (RFCC) (25.95%), electricity and steam generation systems (24.20%), and hydrogen production (20.93%) as the major emission sources, contributing to a total Scope 1 emission of roughly four million tons of CO₂ per year. Based on the quantified emission profile and evaluation of applicable measures, a four-phase strategy toward carbon neutrality by 2050 was constructed to guide the staged deployment of energy efficiency improvements; electrification; renewable integration; carbon capture, utilization, and storage (CCUS); and green hydrogen (GH₂) systems. Scenario analysis indicates that cumulative emission reductions ranging from 80% (conservative) to 90% (optimistic) may be achievable under defined techno-economic assumptions regarding technology deployment and infrastructure development. This work contributes by potentially providing a scalable framework for refinery decarbonization in emerging economies, bridging the gap between emission assessment and transition strategies, while supporting national climate commitments.

1 Introduction

The refining industry plays a vital role in the energy conversion sector by providing products and raw feedstocks for several industrial activities, particularly transport fuels and chemicals. However, this industry is associated with a substantial environmental footprint, most notably high carbon dioxide (CO₂) emissions, which necessitate the measurement of carbon levels (Pinto et al., 2022; Kareem and Al-Shnawa, 2025). Across the petroleum life cycle, encompassing crude oil extraction, transportation, and fuel combustion, the refining stage accounts for more than 40% of the environmental impacts because it operates energy-intensive and high-temperature procedures that are difficult to decarbonize (Sunny et al., 2022). Hence, the decarbonization of refineries is crucial for mitigating carbon emissions and alleviating the pressing environmental impacts as a result of climate change. In response to global commitments to reach a net-zero target, the refining industry is undergoing a profound transition driven by not only stringent environmental regulations and corporate sustainability goals but also by growing market demand for cleaner energy and low-carbon products (Kareem and Al-Shnawa, 2025).

In essence, decarbonization consists of three primary actions: (1) replacing fossil fuel energy with low-carbon or renewable alternatives (e.g., wind, solar, and hydro), (2) enhancing energy efficiency across operations and processes, and (3) deploying carbon capture, utilization, and storage (CCUS) to remove and store CO₂ from industrial activities. The refining industry has long depended on fossil fuels, but there is now a growing adoption of more sustainable practices. As these processes keep advancing, adopting a broader perspective on the potential of circular economy (CE) strategies has become essential (Ghorbani et al., 2024). It is necessary to determine which CE activities that emphasize narrowing, slowing, and closing resource loops might successfully facilitate decarbonization while decreasing the need for primary sources, pollutant releases to nature, and detrimental effects on biodiversity (Haas et al., 2025; Morseletto, 2020). In the context of the refining sector, circular economy (CE) strategies could potentially mitigate life cycle GHGs and resource consumption by promoting renewable energy (RE) and recycled materials (Pinto et al., 2022).

In the context of Vietnam, the nation pledged to achieve net-zero emissions by 2050 at the 26th Conference of the Parties (COP26), demonstrating its commitment to the global climate goals (Tam et al., 2025). The government has introduced several national policies, for instance, the Power Development Plan VIII (PDP8), that prioritize the expansion of renewable energy sources and the exploration of modern approaches, such as CCUS and hydrogen (Son and Ky, 2026). The implementation of CE principles is also promoted through the Law on Environmental Protection (2020) and the National Scheme for Circular Economy Development, which accelerate resource efficiency and pave the way for the adoption of low-carbon measures. Moreover, with an estimated 11.8 Gt of geological CO₂ storage capacity, Vietnam has strong foundational resources required to deploy large-scale CCUS systems (Son and Ky, 2026). However, existing studies have predominantly examined national energy transition and individual decarbonization technologies as separate pathways. There remains a lack of systematic methodologies that integrate carbon accounting, emission hotspot identification, technology prioritization, and forward-looking deployment plans for refineries, particularly in emerging economies such as Vietnam, where accelerating demand, infrastructure deficits, and policy uncertainties interact. In addition, few studies have discussed the potential for integrating advanced CCUS, green hydrogen (GH₂), renewable alternatives, and CE principles into a unified pathway for refinery carbon abatement.

This study addresses these gaps by developing a methodological framework for decarbonization planning in refineries under limited data availability, comprising three sequential stages: (i) Scope 1 CO₂ emission quantification, (ii) assessment and prioritization of low-carbon technologies, and (iii) development of a long-term decarbonization roadmap. The emission calculations adhere to the standards set by the 2006 Intergovernmental Panel on Climate Change (IPCC) Guidelines (Gómez et al., 2006), with a specific emphasis on CO₂ as the GHG of interest, limited to Tier 1 of the IPCC. The resulting emission inventory is then used to qualitatively assess feasible green technologies, which are subsequently integrated into a four-phase implementation roadmap, outlining the deployment sequence, estimated capital expenditures (CAPEX), and strategic priorities to achieve net-zero targets by 2050. The methodology is demonstrated through petrochemical Refinery Y in Vietnam, where the assessment is confined to Scope 1 GHG emissions originating from process facilities, namely RFCC, electricity and steam generation, and hydrogen production systems. Beyond its application to Refinery Y, this approach provides a transparent and transferable basis for a refinery decarbonization roadmap in other refineries operating under similar conditions. This work also contributes to the literature on refinery energy transition in emerging countries.

2 Materials and methods

The research methodology was organized into two complementary parts consisting of seven sequential steps, as illustrated in Figure 1. Throughout Part 1, the main focus is on data acquisition and emission inventory development. Part 2 is mainly focused on translating the quantified emission profile into technology prioritization and proposing proper mitigations to address the emissions generated by Refinery Y processes. Each step builds upon the outputs of the previous stage, ensuring methodological consistency throughout the analysis.

FIGURE 1

Considering the criteria of flexibility and compatibility with the current case study, Tier 1 of the 2006 IPCC Guidelines was adequate to calculate the total CO₂ emissions of Refinery Y using Equations 1, 2 below (Gómez et al., 2006). Although Tier 2 and Tier 3 approaches generally provide higher accuracy in carbon accounting by using national emission factors (EFs) and direct facility measurements, they require detailed fuel properties, plant-specific EFs, or continuous emission monitoring data (CEMS), which were unavailable in this study. Consequently, the inventory was used primarily to compare emission sources and identify priority mitigation targets rather than to establish regulatory emission values. Nevertheless, the consistent application of the IPCC Tier 1 methodology across all process units enables a reliable comparison of emission sources and supports the identification of decarbonization priorities (Jamaludin et al., 2023).

For CO₂, it includes the carbon oxidation factor, assumed to be 1.

In addition, to calculate the total emissions by gas from the source category, the emissions as calculated in Equation 1 are summed over all fuels in Equation 2 below (Gómez et al., 2006):

2.1 Emission factors (EFs) and net calorific values (NCVs)

EFs and NCVs were identified for carbon footprint accounting, using the IPCC and UN Statistics Division (2023 Energy Statistics Yearbook) as references, as shown in Table 2 below.

Refinery gas and fuel gas from the off-gas stream are by-products of petroleum processing, especially catalytic cracking and reforming processes. Furthermore, they have a similar chemical composition, which is primarily composed of light hydrocarbons (methane, ethane, propane, and butane). Therefore, in this case, it is reasonable and scientifically sound to use the refinery gas EF value for fuel gas at Y, although there is no specific database on EF values of fuel gas referred to by the IPCC.

2.2 Sample calculations for the HGU

This study assumed that LPG consists only of saturated hydrocarbons, with a ratio of 50% C₄H₁₀:50% C₃H₈ that could yield the highest hydrogen recovery in the present analysis (Silva et al., 2015). Based on the selected LPG composition, the optimal operating conditions for hydrogen production were determined to be a feed molar ratio of H₂O/LPG/O₂ = 7.0/1.0/0.25 and a temperature of 973 K (Silva et al., 2015). Oxygen was introduced into the reaction system to facilitate the conversion of coke deposits into CO, thereby suppressing coke formation and mitigating catalyst deactivation.

2.2.1 CO₂ emissions from the reforming reaction

The steam reforming of LPG can be represented by Equations 3, 4:

The water-gas shift (WGS) reaction is given by Equation 5:

Assuming that the WGS reaction completely converted CO to CO₂ (Equation 5), along with Equation 3 and Equation 4, the overall reaction is given as Equation 6.

Referring to Silva et al. (2015), after the reforming reaction, the molar ratios of H₂ and CO obtained were 0.65 and 0.27, respectively. Thus, the molar ratio of CO was also the final molar ratio of CO₂, which is 0.27. The ratio of H₂/CO₂ by mass , as assumed above, is . This means that from 1 ton of H₂, we generated tons of CO₂. Then, with (Table 1), CO₂ emissions from the reforming reaction ( are 733,041.69 tons CO₂/year, given by Equation 7

2.2.2 CO₂ emissions from fuels

To obtain the amount of fuel combusted, including fuel gas (FG) and fuel oil (FO), we use Equations 8, 9:

The calculation of CO₂ emissions by FG and FO was done using Equation 1. Finally, Equation 2 was used to calculate the total CO₂ emissions from both the reforming reaction and fuels (the results are shown in Table 3).

In the final step, the decarbonization roadmap was developed using several assumptions. Investment costs were estimated from published refinery case studies, referred to in Part 1 above. The proposed implementation timeline was also established by considering readiness levels, infrastructure availability, and Vietnam’s net-zero emission target for 2050. Therefore, the roadmap should be interpreted as a strategic planning framework rather than a detailed project implementation plan.

3 Results and discussion

3.1 Overview of Refinery Y

3.1.1 Emission status

The primary source of CO₂ emissions from oil refineries typically stems from fuel combustion for plant operations and catalytic cracking units, with fuel combustion accounting for more than 60% of the total emissions (Sunny et al., 2022). Fuels are burned to provide energy and heat for equipment such as heaters, boilers, and combustion turbines. The emission levels depend on the type of crude oil processed, the capacity and energy efficiency of the plant, and the energy usage strategies implemented. GHG emissions from Y are estimated to average approximately four million tons annually.

3.1.2 Material sources

Crude oil from the Bach Ho oil field in Vung Tau is an excellent source of refinery feedstock due to its relatively low sulfur content (0.041%wt). However, as Vietnam began extracting Bach Ho crude oil before the establishment of domestic refineries, most of this crude was initially exported. With the construction and operational commencement of the Dung Quat Refinery in 2009, Bach Ho oil became its primary feedstock. Consequently, Middle Eastern crude oil was chosen to ensure the stable, long-term, and continuous operation of Y. Blending options involving Kuwait, Dubai, and Su Tu Den crude oils were studied, and it was determined that using 100% Kuwait crude oil yields the highest profit margins. Therefore, Kuwait crude oil was selected as the principal feedstock for the refinery (Universal, 2019).

3.1.3 Main technologies

Figure 2 below illustrates the overall process configuration of Refinery Y and highlights the major process units relevant to this study. Crude oil is first preheated through heat exchange, desalted in a two-stage desalting system, and heated to approximately 347 °C–359 °C before entering the atmospheric distillation unit (CDU), where it is separated into LPG, naphtha, kerosene, light gas oil (LGO), heavy gas oil (HGO), and atmospheric residue. Light-end streams from the CDU and other processing units are collected in the light-end recovery unit (LRU), where off-gas is treated, LPG is separated, and full-range naphtha is routed to the naphtha hydrodesulfurization (NHDS) unit. After hydrotreating to remove sulfur, nitrogen, oxygen, and other catalyst poisons, the naphtha stream is split into light naphtha for catalytic isomerization (PENEX) and heavy naphtha for the continuous catalyst regeneration (CCR) reforming unit, which produces high-octane reformate together with hydrogen for the refinery hydrogen network.

FIGURE 2

Middle distillates and atmospheric residue are hydrotreated in the KHDS, GOHDS, and residue hydrodesulfurization (RHDS) units before further upgrading. Hydrotreated residue is processed in the residue fluid catalytic cracking (RFCC) unit to produce LPG, gasoline components, propylene, light cycle oil, clarified oil, and fuel gas. Coke deposited on the catalyst is continuously combusted during catalyst regeneration to provide process heat, making the RFCC one of the largest direct CO₂ emission sources in the refinery. Reformate from the CCR unit is subsequently upgraded in the aromatics complex to produce benzene, toluene, and para-xylene through the Sulfolane, Parex, Isomar, and Tatoray units. High-purity hydrogen required for hydrotreating and reforming is produced in the hydrogen manufacturing unit (HMU) via steam methane reforming followed by pressure swing adsorption (PSA) purification and is distributed through the hydrogen compression and distribution system (HCDS). Finally, sour water stripping, amine regeneration, Claus sulfur recovery, and tail gas treatment units (TGTU) remove sulfur compounds and recover sulfur, while waste heat boilers improve energy recovery and ensure compliance with environmental regulations. The CDU, HMU, CCR, hydrodesulfurization units, RFCC, and steam/electricity generation facilities represent the principal Scope 1 CO₂ emission sources and therefore constitute the focus of the carbon accounting and decarbonization assessment presented in this study.

3.2 Results of CO₂ emissions at Y

Applying the IPCC Tier 1 methodology outlined in Section 2, along with data in Tables 1, 2, the resulting Scope 1 CO₂ emissions of the six major refinery units are presented in Table 3, with the total amount of CO₂ emissions from Y in 2025 being approximately four million tons (Equation 2). It is obvious that the RFCC unit (coke burning) is the largest source of emissions, accounting for 25.9% (1,026,315 tons CO₂) of overall emissions at Y. This value comprises emissions from combustion of petroleum coke, FO (150,406.776 tons CO₂/year), and FG (206,141.76 tons CO₂/year), using Equations 1, 8, 9. In fact, this unit significantly consumes energy to process heavy hydrocarbons into lighter products, while generating large amounts of CO₂ from coke combustion (Hosseini et al., 2025). With the energy obtained from burning coke at 6,259.5 TJ, the emission from coke is 669,766.5 tons CO₂/year (Equation 1), which means that coke contributes 65.3% of RFCC emissions. This is followed by the electricity steam generation unit, which emitted 957,191 tons of CO₂, accounting for 24.20% of total emissions by using fossil fuels (404,628.624 tons of FO CO₂ and 552,562.56 tons of FG CO₂) to operate energy systems. Ranking in third place, as is presented in Section 2, the hydrogen generation unit (HGU) emitted 827,810 tons of CO₂, 20.93% of overall emissions. This is the process of converting LPG into higher-value products, but the reforming reaction also creates by-product CO₂.

Concerning the remaining units, the figures for the crude distillation unit (CDU), hydrotreating unit (HDT), and continuous catalyst regeneration unit (CCR) were 425,289 tons CO₂ (10.76%), 399,095 tons CO₂ (10.09%), and 318,959 tons CO₂ (8.07%), respectively. These units mainly comprise emissions from FO and FG. Although the emissions from these units were two times lower than those from the RFCC, HGU, and electricity steam generation units, they still contribute considerably to the plant’s total emissions. Therefore, in order to reduce total emissions, the plant must concentrate on optimizing its operations and enhancing technologies of such greatest sources of GHG emissions.

Collectively, RFCC, electricity and steam generation, and HGU account for more than 70% of the refinery’s total Scope 1 emissions. This concentration indicates that prioritizing these three units would provide the greatest opportunity for reducing refinery-wide emissions and therefore forms the analytical basis for the subsequent decarbonization pathway.

3.3 Discussion

In 2023, Refinery Y underwent a major maintenance shutdown, followed by completion of the RFCC unit. Y resumed operation at its full design capacity of 200,000 barrels of crude oil per day (approximately 10 million tons of crude oil per year). Assuming continuous full-capacity operation throughout the year (365 days), the estimated annual Scope 1 CO₂ emissions of 3,954,659 tons (Table 3) correspond to an emission intensity of approximately 54.17 kg CO₂/barrel of crude oil processed. This figure also shows that emissions from Y are higher than the commonly reported average (30–50 kg CO₂/barrel) (Jing et al., 2020; Ma et al., 2023). Because emission intensity is influenced by refinery configuration, process complexity, and system boundaries (Calzado Catalá et al., 2017), the comparison is intended as a general reference rather than a direct performance evaluation. In comparison with Dung Quat Refinery in Vietnam, with 1.6 million tons CO₂/year in an average emission, its capacity of approximately 130,000 barrels/day created nearly 33.72 kg CO₂/barrel of oil (PV, 2023). However, it is understandable that Y is a refinery and petrochemical complex with various petrochemical products from different units; therefore, the emissions rate from the production process and fuel combustion will be higher than that of the Dung Quat Refinery.

Figure 3 compares the distribution of Scope 1 CO₂ emissions among the major units of both refineries in Vietnam. The RFCC unit is the leading source, contributing 25.9% (1,026,315 tons CO₂/year) and 57% (910,421 t CO₂/year) of the total emissions at Refinery Y and Dung Quat, respectively. The differences in percentage may stem from the distribution of emissions. In addition, the electricity and steam generation system at Dung Quat experienced 423,729 tons of CO₂/year (26% of overall) (PV, 2023), compared to 24.2% at Y. Both plants rely on fossil fuels to operate their energy systems. At Y, the large emission sources are followed by HGU (LPG reforming), CDU, HDT, and CCR, as discussed above. The similarity in these structures between Y and Dung Quat indicates that the emission calculations for Y in this study are reasonable. The fact that RFCC and electricity and steam production are the two largest contributors to emissions shows that the emission model at Y is consistent with the general trend of the petrochemical refining industry. As a result, the emission calculation method applied to Y is well-founded and accurately reflects the plant’s operational reality.

FIGURE 3

3.4 Feasible green solutions for Y

Following a similar pattern to global refineries, some optimal green solutions are proposed to reduce emissions at the primary units at Y, within the enabling framework and international standards of this review. In addition, the study provides Table 4 below, which shows an overview of the estimated contribution of each of the discussed mitigation solutions.

3.4.1 RFCC unit

Switching to light crude oil is impractical due to the technological configuration designed for processing heavy crude oil from the Middle East. Moreover, such oil is increasingly scarce and expensive, reducing economic efficiency and operational stability. In this context, viable measures include co-processing biogenic feedstocks and replacing fuel oil with natural gas, together reducing RFCC CO₂ emissions by 10–30% (Van Dyk et al., 2022). These solutions allow Y to change to low-carbon fuel sources while remaining consistent with the plant’s actual operating conditions and long-term energy strategy. Co-processing biodiesel from biomass pyrolysis with petroleum fractions (e.g., VGO) in RFCC units is regarded as a potential solution (Ourlin et al., 2026). When pre-treated through deoxygenation (HDO) or catalytic pyrolysis, biodiesel can be blended at a rate of 10–20%, improving production efficiency of gasoline and light olefins, while limiting coke formation (Bezergianni et al., 2018; Fogassy et al., 2011; Ourlin et al., 2026; Wang et al., 2016). It is estimated that this method could reduce 10–25% in Scope 1 CO₂ emissions (Lindfors et al., 2022). For example, Petrobras, a Brazilian state-owned oil company, conducted a pilot-scale study to co-process 5 wt% fast pyrolysis bio-oil (FPBO) with 95 wt% VGO, resulting in a decrease in coke yield (from 6.4 wt% (pure VGO) to 6.0 wt%), with gasoline yield slightly improved (Lammens and Gbordzoe, 2019). Switching to NG, which has a lower carbon content, helps reduce the carbon intensity during combustion. Most of the energy used in refineries comes from internally produced gas (refinery gas), mainly methane and ethane; however, this amount is often insufficient to meet all energy needs, and refineries often use additional heavy residual oil to compensate. Research by Elkamel et al. showed that converting all liquid fuels to NG during operation can reduce CO₂ emissions by up to 30% (Concawe, 2012; Elkamel et al., 2008). This measure can be viable in the medium term, especially given the expanding natural gas infrastructure in Vietnam, such as the Thi Vai LNG terminal (Thang et al., 2015) and the national liquefied NG network plan until 2030.

Enhancing energy efficiency is another priority because it directly reduces fuel consumption without altering refinery production. At Y, significant flue gas and process streams are available for heat recovery. Installing additional heat exchangers or improving heat integration could recover waste heat for feed preheating or steam generation, thereby reducing external fuel demand. This approach was proven effective at Motiva in Texas, where upgraded heat recovery systems cut energy use by 18% and CO₂ emissions by 17% (Wang, 2025). Advanced process control (APC) also helps fine-tune parameters such as feed flow, residence time, and catalyst regeneration rate, decreasing unnecessary fuel consumption and improving conversion efficiency. It represents one of the most cost-effective short-term decarbonization options for Y, as APC mainly involves control system upgrades rather than major equipment replacements.

Next, one of the most effective solutions to capture CO₂ from the RFCC unit is to deploy amine-based absorption technology in a post-combustion capture system that can be easily retrofitted (Hamdy et al., 2021; Hosseini et al., 2025; Li, 2023). For instance, refinery Albera in Canada, where this model was deployed, reduced 85% CO₂ emitted from its RFCC unit (Zhou et al., 2022). Once CO₂ is captured, it can be either stored or reused. Enhanced oil recovery (EOR) is also a widely applied option, both reducing emissions and boosting oil recovery (Zhou et al., 2022). Some refineries combining CCUS with EOR have achieved up to 90% CO₂ reduction while gaining additional economic value (Pilorgé et al., 2020). Beyond storage, CO₂ can also be converted into products (e.g., methanol, ethanol, or polycarbonates) that expand opportunities in petrochemicals and synthetic fuels. In the case of Y, post-combustion capture is considered the most suitable configuration because it requires fewer process modifications than pre-combustion alternatives, and the deployment of CCUS technology is considered to mitigate emissions by 50–90% under full-scale deployment.

Finally, electrifying RFCC auxiliary operations and integrating renewable energy can further reduce the carbon intensity of Refinery Y. Replacing steam-driven equipment with high-efficiency electric motors lowers fuel consumption, while electricity generated from renewable sources reduces indirect emissions (Sorknæs et al., 2022; Taqvi et al., 2021). This strategy is technically feasible because the Nghi Son Economic Zone has favorable renewable energy resources, with average solar irradiation of 4.2–4.8 kWh/m²/day (approximately 2,000–2,200 sunshine hours annually) and coastal wind speeds of 6.0–6.5 m/s, together with well-developed grid and port infrastructure that supports renewable energy deployment (IRENA, 2020). Consequently, rooftop solar photovoltaic systems and near-shore wind power could supply electricity for pumps, compressors, fans, and other auxiliary equipment. However, because the RFCC process requires sustained high-temperature heat, electrification cannot fully replace combustion-based energy in the near term and should therefore be regarded as a complementary medium-to-long-term decarbonization strategy.

3.4.2 Electricity steam generation unit (ESG)

The first priority is improving system efficiency to reduce fuel consumption while maintaining steam and electricity demand. Heat recovery steam generators (HRSGs) can recover thermal energy from turbine exhaust gases to generate additional steam without extra fuel combustion, while replacing aging steam turbines with high-efficiency units further improves power generation efficiency. According to the International Energy Agency (IEA), European refineries have adopted HRSG systems and reduced CO₂ emissions by 10–15%. Applying comparable improvements at Refinery Y is estimated to reduce emissions from the ESG unit by approximately 144,000–191,000 tons of CO₂ per year, depending on the extent of system modernization.

The second measure is reducing the carbon intensity of energy production through fuel switching and renewable energy integration. Because the ESG unit supplies electricity and steam to the entire refinery, it represents the most suitable location for incorporating renewable electricity and low-carbon fuels. Electricity generated from onsite solar or wind installations could partially replace grid electricity, while biomass, biogas, or green hydrogen could gradually substitute fossil fuels in boilers and gas turbines (Cherwoo et al., 2023; Kaur et al., 2025; Nnabuife et al., 2025). Previous studies indicate that supplying approximately 10% of electricity demand from renewable sources can reduce CO₂ emissions by 8–12% (Kongkuah et al., 2025), whereas partial replacement of fossil fuels with biomass or green hydrogen can substantially lower the carbon intensity of steam generation. When implemented together, these measures could reduce emissions from the ESG unit by approximately 15–25%. Finally, CCUS should be considered a long-term decarbonization option for this unit. Because combustion-related CO₂ emissions originate mainly from boiler and turbine flue gases, post-combustion carbon capture can be integrated into the existing exhaust system using infrastructure similar to that proposed for the RFCC unit.

3.4.3 HGU unit

At Y, the hydrogen generation unit is of high importance in providing hydro for deep processes such as hydrocracking (HCU) and hydrodesulfurization (HDS), enhancing the quality of fuel outputs. Otherwise, the current gray hydrogen is produced at the HGU unit, which relies on steam methane reforming (SMR) technology, directly leading to 827,810 tons of CO₂/year. Therefore, the three main solutions for this unit are (1) improvement in technologies and efficiency, (2) GH₂, and (3) CCUS. First, it is practical to optimize catalytic yield and operating conditions to increase hydrocarbon conversion rates while reducing fuel consumption. HRSG systems can also be applied as in the above unit, yielding the same result. Next, switching to production of GH₂ by water electrolysis using renewable energy (solar or wind) has the potential to eliminate emissions (Nguyen, 2024). It is assumed that if at least 30% of gray hydrogen production were replaced with GH₂, CO₂ emissions would be reduced by 25–35% (Franco, 2025). Despite the expensive production cost of GH₂ (2–3 times higher than gray hydrogen), the downward trend in the cost of electrolysis technology and renewables development would reduce this cost in the future. Finally, CCUS is one of the most effective measures that could be applied.

3.4.4 Entire refinery

Regarding approachable solutions for the whole plant, all the solutions identified for the major emitting units are inherently scalable and expandable. In addition, based on Concawe’s Refinery 2050 study, there will be strong shifts in fuel demand by 2050 (Concawe, 2020). That report evaluates several decarbonization pathways for EU refineries, including electrification, electrolytic hydrogen, CCUS, and alternative feedstocks (e.g., biomass, lipids, and CO₂-derived syngas) for Fischer–Tropsch (FT) synthesis of e-fuels (Concawe, 2020; 2024). As a result, electrolytic hydrogen can cut refinery-site emissions by up to 75% but does not significantly reduce combustion emissions of the final fuels. In contrast, co-processing or fully switching to alternative feedstocks reduces product carbon intensity (up to >80% with CCS on biogenic CO₂) but may increase onsite emissions unless CO₂ is captured. These findings reveal a trade-off: hydrogen solutions offer large onsite reductions with limited changes, whereas renewable feedstocks deliver deeper lifecycle decarbonization but require major technological transformation. Although based on the EU context, constraints such as biomass and lipid availability also apply to other regions, suggesting that future refinery configurations should emphasize clean hydrogen and CO₂ recycling, forming the basis of a “circular refinery” (Concawe, 2020).

To ensure the reduction contribution of each mitigation measure was quantitatively traceable rather than assumed at the whole-refinery level, the individual reduction percentages in Table 4 were derived by weighting each measure’s within-unit efficiency (supported by cited literature ranges) against that unit’s share of total baseline Scope 1 emissions, as established in Table 3. For measures with technology-dependent efficiency (e.g., green hydrogen substitution), the within-unit reduction was further derived from assumed electrolyzer capacity and hydrogen output relative to Refinery Y’s current hydrogen production, using the relation: substitution fraction = (H₂ produced by electrolysis)/(current HGU hydrogen output). Because component-level projections carry compounding uncertainty, Table 4 is presented as two bounding scenarios, conservative and optimistic, corresponding to the lower and upper efficiency ranges reported in the literature for each technology. A residual reduction requirement, attributed to direct air capture (DAC) and carbon offsets consistent with the Phase 4 measures described in Table 5, is included explicitly to close the gap between technology-specific abatement and the refinery’s net-zero target rather than leaving this contribution implicit. This revised structure yields a cumulative reduction range of 80–90% of total refinery Scope 1 emissions, consistent with the range reported in Table 5 and the abstract, and provides a transparent, reproducible basis for the claimed reduction pathway.

3.5 Proposed strategic plan to decarbonize Y toward net zero by 2050

In alignment with Vietnam’s commitment to reducing 86 million tons in industrial processes by 2030 and achieving net zero by 2050, a four-stage decarbonization plan is clearly proposed as shown in Table 5, according to the assessment of each of the above feasible green solutions for Y.

In general, each stage contributes incrementally to Y’s decarbonization pathway. Starting with Phase 1, rapid economic and environmental benefits are established as a solid foundation to deploy medium-to-long-term technologies at the refinery. Subsequently, Phase 2 represents a critical transition from operational optimization to structural energy, followed by some initial preparation steps for CCUS and GH₂ production. After that, Phase 3 serves as a technically operational bridge that introduces pilot-scale CCUS on the RFCC unit alongside bio-feedstock co-processing and pilot GH₂, together adding approximately 13–23% of additional cumulative CO₂ reduction, while establishing the technical and operational basis for full-scale CCUS expansion to HGU in Phase 4. Finally, the plan ends with Phase 4, which represents the decisive step toward net zero. It allows Y to meet emerging low-carbon fuel standards, thereby accessing the international market and engaging in trading mechanisms.

Table 6 summarizes the implementation feasibility and key economic characteristics of the major decarbonization technologies considered for Refinery Y. The feasibility levels (high, moderate, and low) were assigned through a qualitative assessment based on a combination of evidence from the literature, technology maturity, and site-specific technical constraints at Refinery Y, along with the current infrastructure and policy conditions in Vietnam. In this study, the feasibility ratings should be interpreted as comparative indicators to support technology prioritization rather than quantitative scores.

CCUS implementation in Vietnam remains constrained due to the absence of a commercial CO₂ transport and storage network (Liu et al., 2025; Nguyen, 2025). Previous geological assessments have indicated significant offshore storage potential in sedimentary basins, such as the Cuu Long, Nam Con Son, Malay-Tho Chu, and Song Hong, with an estimated theoretical storage capacity of several gigatons of CO₂ (Bokka and Lau, 2023). Restrictions remain on detailed site characterization, monitoring protocols, and liability regulations. In general, commercial CCUS deployment at Refinery Y is considered feasible only in the long term and would require coordinated national investment in CO₂ storage infrastructure and regulatory development. Moreover, electrification and green hydrogen production will substantially increase refinery electricity demand (Quoc Thang, 2023). According to Vietnam’s Power Development Plan VIII (PDP VIII), the national energy strategy targets rapid expansion of renewable electricity, particularly offshore wind and solar photovoltaic generation (Vneconomy, 2025). Nevertheless, the availability of stable renewable electricity remains uncertain in the short term because of transmission constraints, renewable curtailment, and regional imbalance between electricity generation and industrial demand. Hence, phased implementation combined with long-term power purchase agreements (DPPAs) is considered a more practical approach for Refinery Y. Water availability is another critical consideration for green hydrogen production. Proton exchange membrane and alkaline electrolyzers typically require approximately 9–15 L of high-purity water/kg of hydrogen produced, excluding additional cooling water requirements. Refinery Y already operates industrial water treatment facilities; however, commercial-scale hydrogen production would require expanded water purification capacity and improved water recycling systems to minimize freshwater consumption and ensure long-term operational sustainability.

4 Conclusion

In conclusion, the proposed framework in this study aimed to provide a systematic approach for identifying major emission sources and evaluating potential mitigation strategies. The application of the 2006 IPCC Tier 1 provided a transparent approach for estimating fuel combustion emissions using standard EFs and NCVs, which is appropriate under the current data availability. The results indicate that the refinery emits approximately 4 Mt CO₂ under full-capacity operation in 2025, corresponding to an average carbon intensity of 54.17 kg CO₂/barrel of oil processed, which is above the lower range reported for leading low-carbon refineries. It concentrates on three units with priority during upcoming transition phases, namely RFCC (25.95%), ESG (24.20%), and HGU (20.93%), due to their large contribution to total emissions. The findings demonstrate that prioritizing high-emission process units enables a more targeted allocation of decarbonization technologies and investment, providing a practical basis for long-term transition planning in the refining sector.

On the other hand, the estimated investment requirements increase substantially across implementation phases, with Phase 4 requiring between $740M and $960M. These figures reflect the high capital intensity of large-scale CCUS and green hydrogen deployment. Rather than predicting specific emission outcomes, the four-phase proposed roadmap is intended to provide a structured and technically informed pathway that can support long-term emission reduction planning while remaining subject to future technical, economic, and policy developments. In the future, research should incorporate Tier 2 or Tier 3 emission inventories using refinery-specific operating data, together with uncertainty analysis, techno-economic optimization of mitigation pathways, assessment of domestic CO₂ storage capacity, and evaluation of green hydrogen supply chains, to improve the robustness of long-term decarbonization planning. Overall, although developed for Refinery Y, the methodology can be adapted to other refineries in Vietnam as well as in similar developing countries where site-specific data are restricted and staged decarbonization planning is required.

References

• 1

  AsmelashE.GoriniR. (2021). International Oil Companies and the Energy Transition. Abu Dhabi: IRENA. Available online at: https://www.irena.org/Publications/2021/Feb/Oil-companies-and-the-energy-transition (Accessed July 9, 2026).

  • Google Scholar

• 2

  BezergianniS.DimitriadisA.KikhtyaninO.KubičkaD. (2018). Refinery co-processing of renewable feeds. Prog. Energy Combust. Sci.68, 29–64. 10.1016/j.pecs.2018.04.002

  • CrossRef
  • Google Scholar

• 3

  BokkaH. K.LauH. C. (2023). Decarbonizing Vietnam’s power and industry sectors by carbon capture and storage. Energy262, 125361. 10.1016/j.energy.2022.125361

  • CrossRef
  • Google Scholar

• 4

  ByrumZ.PilorgéH.WilcoxJ. (2021). Technological Pathways for Decarbonizing Petroleum Refining. North America: World Resources Institute. 10.46830/wriwp.21.00004

  • CrossRef
  • Google Scholar

• 5

  Calzado CataláF.De VuystK.De MeerleerW.GardzinskiW.HueskenS.Iglesias LopezA.et al (2017). Estimating the marginal CO2 intensities of EU refinery products. Brussels: CONCAWE Reports.

  • Google Scholar

• 6

  CherwooL.GuptaI.FloraG.VermaR.KapilM.AryaS. K.et al (2023). Biofuels an alternative to traditional fossil fuels: a comprehensive review. Sustain. Energy Technol. Assessments60, 103503. 10.1016/j.seta.2023.103503

  • CrossRef
  • Google Scholar

• 7

  Concawe (2012). EU Refinery Energy Systems and Efficiency. Brussels: CONCAWE Reports. Available online at: https://www.concawe.eu (Accessed July 8, 2026).

  • Google Scholar

• 8

  Concawe (2020). Exploring Possible Pathways For The EU Refining System To Contribute To A Low-CO2 Economy In The 2030-2050 Time Frame. Brussels: CONCAWE Reports. Available online at: https://www.concawe.eu (Accessed July 8, 2026).

  • Google Scholar

• 9

  Concawe (2024). E-Fuels: a techno-economic assessment of European domestic production and imports towards 2050. Brussels: CONCAWE Reports. Available online at: https://www.concawe.eu (Accessed July 8, 2026).

  • Google Scholar

• 10

  ElkamelA.Ba-ShammakhM.DouglasP.CroisetE. (2008). An optimization approach for integrating planning and CO2 emission reduction in the petroleum refining industry. Industrial Eng. Chem. Res.47 (3), 760–776. 10.1021/ie070426n

  • CrossRef
  • Google Scholar

• 11

  FogassyG.ThegaridN.SchuurmanY.MirodatosC. (2011). From biomass to bio-gasoline by FCC co-processing: effect of feed composition and catalyst structure on product quality. Energy Environ. Sci.4 (12), 5068–5076. 10.1039/c1ee02012a

  • CrossRef
  • Google Scholar

• 12

  FrancoA. (2025). Green hydrogen and the energy transition: hopes, challenges, and realistic opportunities. Hydrogen6 (2), 28. 10.3390/hydrogen6020028

  • CrossRef
  • Google Scholar

• 13

  GhorbaniY.ZhangS. E.NwailaG. T.BourdeauJ. E.RoseD. H. (2024). Embracing a diverse approach to a globally inclusive green energy transition: moving beyond decarbonization and recognizing realistic carbon reduction strategies. J. Clean. Prod.434, 140414. 10.1016/j.jclepro.2023.140414

  • CrossRef
  • Google Scholar

• 14

  GómezD.WattersonJ.AmericanoB.HaC.MarlandG.MatsikaE.et al (2006). Chapter 2: stationary combustion. 2006 IPCC Guidelines for National Greenhouse Gas Inventories, 2.1–2.47.

  • Google Scholar

• 15

  Greenhouse Gas Protocol (2011). Sector Toolsets for Iron and Steel-Guidance Document. Washington, DC: WRI and WBCSD, 1–12.

  • Google Scholar

• 16

  HaasW.BaumgartA.EisenmengerN.VirágD.KaltG.SommerM.et al (2025). How decarbonization and the circular economy interact: benefits and trade-offs in the case of the buildings, transport, and electricity sectors in Austria. J. Industrial Ecol.29 (2), 531–545. 10.1111/jiec.13619

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 17

  HamdyB.GoelC.RuddA.J.Barron, A.R.AndreoliE. (2021). The application of amine-based materials for carbon capture and utilization: an overarching view. Mater. Adv.2 (18), 5843–5880. 10.1039/D1MA00360G

  • CrossRef
  • Google Scholar

• 18

  HosseiniS. M.MoghadamP.Afshar EbrahimiA. (2025). Boosting CO₂ capture efficiency of the exhausted RFCC flue gas by using intercooler exchangers: leveraging ANN in MDEA-Based approach. J. CO₂ Util.95, 103091. 10.1016/j.jcou.2025.103091

  • CrossRef
  • Google Scholar

• 19

  HoweC.O’DellK.NehaR.ChristianT. (2024). Pathways to Commercial Liftoff: Clean Hydrogen. U.S Department of Energy. Available online at: https://docs.cpuc.ca.gov/PublishedDocs/SupDoc/A2209006/8462/577291710.pdf (Accessed July 9th, 2026).

  • Google Scholar

• 20

  IEA (2021). Net Zero by 2050. International Energy Agency. Available online at: https://www.iea.org/reports/net-zero-by-2050 (Accessed July 3, 2026).

  • Google Scholar

• 21

  IRENA (2020). Renewable Energy Statistics 2020. Available online at: https://www.irena.org/publications/2020/Jul/Renewable-energy-statistics-2020 (Accessed July 9, 2026).

  • Google Scholar

• 22

  JamaludinN. F.ShuhaimiN. A.MohamedO. Y.HadipornamaM. F. (2023). Quantification of carbon footprint in petroleum refinery process. Chem. Eng. Trans.106, 97–102. 10.3303/CET23106017

  • CrossRef
  • Google Scholar

• 23

  JaradD. (2024). Global Status of CCS 2024. Global CCS Institute. Available online at: https://www.globalccsinstitute.com/publications/global-status-of-ccs-2024/ (Accessed July 9, 2026).

  • Google Scholar

• 24

  JingL.El-HoujeiriH. M.MonfortJ.-C.BrandtA. R.MasnadiM. S.GordonD.et al (2020). Carbon intensity of global crude oil refining and mitigation potential. Nat. Clim. Change10 (6), 526–532. 10.1038/s41558-020-0775-3

  • CrossRef
  • Google Scholar

• 25

  KareemM. A.Al-ShnawaA. (2025). Decarbonizing oil refineries: the transition to green electricity in high-temperature processes. ISAR J. Sci. Technol.3 (11), 11–24. 10.5281/zenodo.18066131

  • CrossRef
  • Google Scholar

• 26

  KaurS.KumarR.SinghK.SinghS. (2025). Systematic review of hydrogen, biomass, biogas, and solar photovoltaics in hybrid renewable energy systems: advancements, challenges, and future directions. Int. J. Hydrogen Energy137, 160–189. 10.1016/j.ijhydene.2025.04.525

  • CrossRef
  • Google Scholar

• 27

  KongkuahM.AlessaN.KongkuahM.AlessaN. (2025). Renewable energy and carbon intensity: Global evidence from 184 countries (2000–2020). Energies18 (13), 1–29. 10.3390/en18133236

  • CrossRef
  • Google Scholar

• 28

  LammensT.GbordzoeE. (2019). Co-Processing fast pyrolysis bio-oil. TechnipFMC.

  • Google Scholar

• 29

  LiH. (2023). Advancing “Carbon Peak” and “Carbon Neutrality” in China: a comprehensive review of current global research on carbon capture, utilization, and storage technology and its implications. ACS Omega8 (45), 42086–42101. 10.1021/acsomega.3c06422

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 30

  LindforsC.ElliottD.PrinsW.OasmaaA.LehtonenJ. (2022). Co-processing of biocrudes in oil refineries. Energy Fuel37, 799–804. 10.1021/acs.energyfuels.2c04238

  • CrossRef
  • Google Scholar

• 31

  LiuN.LiuD.ZhanY.ChenQ.QiaoY.TanR. (2025). From innovation to integration: the evolving role of carbon capture and storage in global decarbonization. Science1 (1), e70008. 10.1002/csc3.70008

  • CrossRef
  • Google Scholar

• 32

  MaS.LeiT.MengJ.LiangX.GuanD. (2023). Global oil refining’s contribution to greenhouse gas emissions from 2000 to 2021. Innovation4 (1), 100361. 10.1016/j.xinn.2022.100361

  • CrossRef
  • Google Scholar

• 33

  MorselettoP. (2020). Targets for a circular economy. Resour. Conservation Recycl.153, 104553. 10.1016/j.resconrec.2019.104553

  • CrossRef
  • Google Scholar

• 34

  NguyenH. T. (2025). Integrated carbon capture and storage strategies in hard-to-abate industries for achieving vietnam’s net zero by 2050. Greenh. Gases Sci. Technol.15 (3), 333–345. 10.1002/ghg.2341

  • CrossRef
  • Google Scholar

• 35

  NnabuifeS. G.HamzatA. K.WhidborneJ.KuangB.JenkinsK. W. (2025). Integration of renewable energy sources in tandem with electrolysis: a technology review for green hydrogen production. Int. J. Hydrogen Energy107, 218–240. 10.1016/j.ijhydene.2024.06.342

  • CrossRef
  • Google Scholar

• 36

  OurlinT.MartiF. T.VenderboschR.ÇelebiS.KarahanS.MathieuY.et al (2026). Integration of pyrolysis bio-oils of varying upgrading levels in FCC co-processing with VGO: benefits and limitations. Fuel417, 138622. 10.1016/j.fuel.2026.138622

  • CrossRef
  • Google Scholar

• 37

  PilorgéH.McQueenN.MaynardD.PsarrasP.HeJ.RufaelT.et al (2020). Cost analysis of carbon capture and sequestration of process emissions from the U.S. industrial sector. Environ. Sci. Technol.54 (12), 7524–7532. 10.1021/acs.est.9b07930

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 38

  PintoL. F. R.TucciH. N. P.MummoloG.NetoG. C. de O.FacchiniF. (2022). Circular economy approach on energy cogeneration in petroleum refining. Energies15 (5), 1713. 10.3390/en15051713

  • CrossRef
  • Google Scholar

• 39

  PV (2023). Dung Quat Refinery Aims for Green Energy. VIETNAM.VN. Available online at: https://www.vietnam.vn/en/nmld-dung-quat-huong-den-nang-luong-xanh (Accessed July 8, 2026).

  • Google Scholar

• 40

  QuocT. (2023). Green Hydrogen Production In Vietnam: Potential, Demand, And Obstacles. EVNPECC3. Available online at: https://www.pecc3.com.vn (Accessed July 9, 2026).

  • Google Scholar

• 41

  SilvaP. P.FerreiraR. A.NunesJ. F.SousaJ. A.RomanieloL. L.NoronhaF. B.et al (2015). Production of hydrogen from the steam and oxidative reforming of LPG: thermodynamic and experimental study. Braz. J. Chem. Eng.32, 647–662. 10.1590/0104-6632.20150323s00003441

  • CrossRef
  • Google Scholar

• 42

  SonN. T.KyN. T. (2026). Integrated carbon-neutral LNG and hydrogen pathways for Vietnam’s net-zero transition: a techno-economic and system-level assessment. Int. J. Adv. Signal Image Sci.12, 1418–1425. 10.29284/2y7z5848

  • CrossRef
  • Google Scholar

• 43

  SorknæsP.JohannsenR. M.KorbergA. D.NielsenT. B.PetersenU. R.MathiesenB. V. (2022). Electrification of the industrial sector in 100% renewable energy scenarios. Energy254, 124339. 10.1016/j.energy.2022.124339

  • CrossRef
  • Google Scholar

• 44

  SunnyN.BernardiA.DanaciD.BuiM.Gonzalez-GarayA.ChachuatB. (2022). A pathway towards net-zero emissions in oil refineries. Front. Chem. Eng.4, 804163. 10.3389/fceng.2022.804163

  • CrossRef
  • Google Scholar

• 45

  TamN. V.QuynhT. T. H.ToanN. Q. (2025). How Vietnam can achieve net-zero carbon emissions in construction and built environment by 2050: an integrated AHP and DEMATEL approach. Build. Environ.274, 112752. 10.1016/j.buildenv.2025.112752

  • CrossRef
  • Google Scholar

• 46

  TaqviS.AlmansooriA.ElkamelA. (2021). Optimal renewable energy integration into the process industry using multi-energy hub approach with economic and environmental considerations: refinery-wide case study. Comput. Chem. Eng.151, 107345. 10.1016/j.compchemeng.2021.107345

  • CrossRef
  • Google Scholar

• 47

  ThangN. V.HoaH. T.HaN. T. (2015). Method of weighted criteria evaluation to select LNG suppliers. Petrovietnam J.2, 53–59. Available online at: https://pvj.com.vn/index.php/TCDK/article/view/429.

  • Google Scholar

• 48

  UniversalT. (2019). Overview about NSRP. Nghi Son Refinery and Petrochemical LLC. Available online at: https://nsrp.vn/about-overview/ (Accessed July 8, 2026).

  • Google Scholar

• 49

  Van DykS.SuJ.EbadianM.SaddlerJ. (2022). Production of lower carbon-intensity fuels by co-processing biogenic feedstocks: potential and challenges for refineries. Fuel324, 124636. 10.1016/j.fuel.2022.124636

  • CrossRef
  • Google Scholar

• 50

  Vneconomy (2025). Vietnam approves revised power development plan 8, signaling massive renewable energy expansion. Sustainable Energy Transition Programme (SETP). Available online at: https://setp.vn/eu_news/vietnam-approves-revised-power-development-plan-8-signaling-massive-renewable-energy-expansion/.

  • Google Scholar

• 51

  WangY. (2025). Waste heat recovery and energy reutilization technologies in industrial processes: research on improving energy efficiency and reducing emissions. E3S Web of Conferences60605003. 10.1051/e3sconf/202560605003

  • CrossRef
  • Google Scholar

• 52

  WangC.LiM.FangY. (2016). Coprocessing of catalytic-pyrolysis-derived bio-oil with VGO in a pilot-scale FCC riser. Industrial Eng. Chem. Res.55 (12), 3525–3534. 10.1021/acs.iecr.5b03008

  • CrossRef
  • Google Scholar

• 53

  ZhouJ.Van Den AssemD.ChalaturnykR.MeikleG.GrayM.DarS.et al (2022). Carbon Capture, Utilization, and Storage (CCUS): Technology Innovation to Accelerate Broad Deployment in Alberta. Alberta Innovates. Available online at: https://albertainnovates.ca/wp-content/uploads/2022/06/AI-CCUS-WHITE-PAPER-2022-WEB.pdf.

  • Google Scholar