A Study on the Promoting Role of Renewable Hydrogen in the Transformation of Petroleum Refining Pathways

Abstract

The refining industry is shifting from decarbonization to hydrogenation for processing heavy fractions to reduce pollution and improve efficiency. However, the carbon footprint of hydrogen production presents significant environmental challenges. This study couples refinery linear programming models with life cycle assessment to evaluate, from a long-term perspective, the role of low-carbon hydrogen in promoting sustainable and profitable hydrogenation refining practices. Eight hydrogen-production pathways were examined, including those based on fossil fuels and renewable energy, providing hydrogen for three representative refineries adopting hydrogenation, decarbonization, and co-processing routes. Learning curves were used to predict future hydrogen cost trends. Currently, hydrogenation refineries using fossil fuels benefit from significant cost advantages in hydrogen production, demonstrating optimal economic performance. However, in the long term, with increasing carbon taxes, hydrogenation routes will be affected by the high carbon emissions associated with fossil-based hydrogen, losing economic advantages compared to decarbonization pathways. With increasing installed capacity and technological advancements, low-carbon hydrogen is anticipated to reach cost parity with fossil-based hydrogen before 2060. Coupling renewable hydrogen is expected to yield the most significant economic advantages for hydrogenation refineries in the long term. Renewable hydrogen drives the transition of refining processing routes from a decarbonization-oriented approach to a hydrogenation-oriented paradigm, resulting in cleaner refining processes and enhanced competitiveness under emission-reduction pressures.

1. Introduction

Petroleum refining stands as the backbone of global energy production, meeting the incessant demand for various fuels in our modern world. However, this vital industrial process comes with environmental costs [1]. Carbon emissions have emerged as a significant concern during the refining of crude oil into valuable petroleum products [2]. The refining industry ranks as the world’s third-largest stationary source of greenhouse gas emissions, placing it at the forefront of energy transition and pathways toward net-zero emissions [3].

The carbon footprint of petroleum refining spans from crude oil extraction to the final distribution of refined products [4]. Each stage entails high-energy operations, chemical transformations, and the use of materials with a high carbon footprint, resulting in substantial emissions of carbon dioxide and other greenhouse gases [5]. Between 2000 and 2021, cumulative greenhouse gas emissions from refineries totaled approximately 341 billion tons, with an average annual growth rate of 0.7% [6]. The magnitude of these emissions, along with the global scale of refining operations, highlights the pressing need to address and mitigate the environmental impact of this crucial industrial process.

To cope with the competition arising from market expansion and the increase in crude oil costs, most refineries have increased the proportion of heavy and low-quality crude oil processing. However, this exacerbates the contradiction between the carbon-to-hydrogen ratio of feedstocks and end products [7,8]. In distilled crude oil products, the proportion of straight-run fractions suitable for light fuel oil with a balanced hydrogen-to-carbon ratio is relatively low, while components requiring deep processing, such as vacuum gas oil (VGO) and vacuum residue (VR), are more prevalent. Analysis shows that heavy oil typically has a hydrogen-to-carbon ratio of around 1.37, whereas the primary products of the refining process—gasoline, jet fuel, and diesel—have carbon-to-hydrogen ratios of 1.85, 1.91~2.02, and 1.83, respectively [8]. Therefore, the most crucial task in the processing of heavy oil is to increase the hydrogen proportion. The imbalance of carbon and hydrogen elements leads to excess carbon emissions in the form of greenhouse gases [9]. This not only causes wastage of carbon resources but also results in increased energy consumption due to prolonged processing. In 2021, the global average carbon emission intensity of refineries was approximately 56.2 kg/bbl, with countries heavily reliant on deep processing technologies having a carbon emission intensity exceeding 50 kg/bbl [10]. Currently, widely acknowledged technological pathways for adjusting the hydrogen-to-carbon ratio include decarbonization pathways and hydrogenation pathways.

Delayed coking is a prominent heavy oil upgrading technology capable of directly processing low-quality heavy oils and effectively transforming them [11]. In addition to producing gasoline and diesel, it generates a significant quantity of gas oil, which is suitable for catalytic cracking or catalytic hydrocracking [12,13]. This process achieves comprehensive decarbonization and eliminates the majority of heavy metal impurities. Delayed coking offers strong feedstock adaptability, high conversion rates, lower investment requirements, and mature technology. Consequently, it is one of the primary methods used for processing high-asphaltene, high-metal-content, low-quality residuum. Beyond conventional residuum, delayed coking units can also process secondary residuum (such as slurry oil and ethylene tar), tank bottom oil, and refinery sludge [14,15]. However, as environmental regulations tighten and the demand for high-quality refinery products rises, delayed coking faces significant challenges. These challenges are primarily evident in several areas: firstly, delayed coking typically yields petroleum coke at rates exceeding 30%, resulting in lower overall petroleum resource efficiency due to the high coke yield [16]. Secondly, during periods of high oil prices, the economic viability of delayed coking units is compromised by the production of low-value coke products. Furthermore, the frequent operational tasks involved in delayed coking production, such as coke drum switching, hydraulic decoking, and the storage and transportation of petroleum coke, can lead to the release of oil and gas volatiles, thus impacting the environment. Additionally, other decarbonization processes like visbreaking and solvent deasphalting, though part of the decarbonization route, are not extensively discussed here due to their smaller processing capacities [17,18].

Central to the hydrogenation pathway are the pivotal units of VGO hydrocracking and Vacuum Residue Desulfurization (VRDS) [19,20]. Hydrocracking involves subjecting VGO and selecting secondary heavy oils to thermal breakdown under conditions characterized by elevated temperature, high hydrogen pressure, and the presence of catalysts. This process catalyzes the conversion of these feedstocks into a spectrum of products, including gases, gasoline, jet fuel, diesel, and hydrogenated tail oils. The VRDS process harnesses hydrogenation to treat atmospheric and vacuum residues, effectively purging impurities from the oils and enhancing the hydrogen-to-carbon ratio. Residue-processing techniques are typified by reactor configurations, ranging from fixed bed to moving bed, fluidized bed, and slurry bed hydroprocessing. Processes characterized by lower residue conversion rates serve as substrates for catalytic cracking and hydrocracking, while those exhibiting higher conversion rates yield lighter fuel oils directly [21,22]. In recent years, spurred by the reciprocal advancements in heavy crude oil upgrading and product purification, residue hydroprocessing technology has undergone rapid expansion. Projections suggest that over the next five years, the compound growth rate of residue hydroprocessing capacity will soar to 4.98%, while delayed coking is anticipated to exhibit a modest compound growth rate of merely 0.14% [23]. This trend underscores a pivotal evolution within refinery processing paradigms, driven by a dual pursuit of heightened operational efficacy and enhanced environmental sustainability. This will also bring about the carbon emissions issue associated with the hydrogen-supply process. Within the context of the ‘dual carbon goals’, providing low-carbon hydrogen to refineries is a mainstream research and development direction [24,25].

In a comprehensive assessment, low-carbon hydrogen emerges as a focal point within current academic discourse. The selection of hydrogen-production technologies is pivotal in determining both the cost and carbon emissions associated with hydrogen energy [26]. Presently, mature technologies such as fossil fuel-based hydrogen production and industrial by-product hydrogen processes demonstrate scalability and affordability, with hydrogen production costs typically in the range of 10~15 CNY/kg [27,28]. However, these methods exhibit high carbon intensity, notably coal-based hydrogen production, with a carbon intensity of 20 kgCO₂/kgH₂ [29]. In contrast, renewable energy electrolysis-based hydrogen production offers flexibility, high hydrogen purity, and negligible carbon emissions [30,31]. Yet, its current conversion rates remain modest, and hydrogen production costs are intricately linked to electricity sourcing and electrolyzer configurations, often exceeding 30 CNY/kgH₂. The integration of fossil fuel-based hydrogen production with carbon capture and storage (CCS) holds promise in significantly reducing hydrogen’s carbon intensity by over 90% [32]. However, commercialization efforts are still in nascent stages, necessitating higher initial investments.

Currently, the hydrogen strategies of mainstream energy companies and researchers are shifting from primarily relying on hydrogen based on fossil fuels to prioritizing the use of hydrogen based on renewable energy under carbon-neutral conditions [31]. Qin et al. integrated mature low- or no-carbon technologies, including green hydrogen, green electricity, biogas, and bio-oil co-processing technologies, into hydrocracking units [33]. The integrated co-processing technology reduces the input carbon emissions by 31%. Shi et al. coupled high-purity hydrogen produced from wind and solar energy with the coal-to-ethylene glycol process, enhancing ethylene glycol yields and significantly reducing product carbon footprints [34]. Nascimento da Silva et al. harnessed surplus wind energy to produce hydrogen for the hydrogenation units of petroleum refineries, replacing steam methane reforming for hydrogen production and enabling oxygen supply for oxygen combustion carbon dioxide capture in fluid catalytic cracking (FCC) units [35]. This process can reduce greenhouse gas emissions from refineries by up to 22.11%. Mastropasqua et al. developed a multi-energy system based on solid oxide fuel cells (SOFCs), aiming to achieve the simultaneous production of hydrogen, electricity, and process steam, along with carbon capture [36]. Additionally, research teams worldwide have investigated the coupling of wind energy, solar energy, biomass energy, nuclear energy, and traditional energy sources for the production of fuels and chemicals to achieve carbon reduction [27,30,35,37,38]. Shih and his teammates proposed the concept of ‘liquid sunshine’, which involves the reaction of hydrogen with carbon dioxide to produce methanol, successfully achieving industrialization [39]. Sinopec has launched the ‘Green Hydrogen Refining’ project, supplying renewable hydrogen to petroleum refining processes [40]. BP and offshore wind developer Ørsted are collaborating to complete a 50 MW electrolyzer project by 2024, aimed at providing large-scale green hydrogen for the Lingen refinery in Germany [41]. Shell has initiated several carbon-neutral green hydrogen projects globally, including the construction of Europe’s largest proton exchange membrane electrolyzer in July 2021. This facility is capable of producing 1300 tons of green hydrogen annually, intended for use in the Rhineland refinery’s hydrogenation unit in Germany [42]. As of the present, PetroChina is advancing four green hydrogen projects, among which is a 2 million kilowatt wind and solar power-to-hydrogen project in the Qinghai oilfield intended to supply hydrogen to refineries [43]. With ongoing technological advancements, the proliferation of similar projects is anticipated.

The current studies on multi-energy coupling processes focus on conceptual system design. The refining process is characterized by its complexity, multitude of units, and diverse range of products [44]. For processes involving the coupling of multiple energy sources, global linear programming models are essential for conducting thorough techno-economic–environmental performance analysis and optimization [45]. Additionally, the swift advancement of low-carbon hydrogen production technologies in recent years has rendered many previous research findings outdated. However, similar studies do not include an analysis of future hydrogen production cost trends based on actual production data from China, nor do they provide comparisons of hydrogen production costs across different technologies within the same refinery hydrogen framework. Currently, refineries are ramping up the construction of hydrogenation unit equipment, leading to an increase in overall hydrogen consumption. Whether this technological transition can demonstrate sufficient economic competitiveness compared to the prevailing high-carbon hydrogen supply mode, which heavily relies on fossil fuels, remains understudied and warrants further investigation and validation.

This work extensively considers the coupling pathways between various hydrogen energy sources and oil refineries, providing a rigorous evaluation of the economic and product carbon footprints for a series of representative refinery configurations. In this context, we propose the following:

• (1). Establish an integrated linear programming (LP) model to assess the economic performance of current decarbonization and hydrogenation routes.

• (2). Develop a carbon footprint tracking model to track product carbon footprints and evaluate the environmental performance of decarbonization and hydrogenation routes.

• (3). Predict hydrogen prices at different time points within a unified multi-energy coupling framework, analyzing the process potential under scenarios of cost reduction due to technological advancements and cost increases due to carbon taxes.

• (4). Propose an evaluation index to demonstrate the driving role of low-carbon hydrogen in the transformation of refinery processing routes.

This comprehensive assessment not only provides a broad understanding of the sustainability value proposition of renewable hydrogen for refineries but also highlights the transformation of refinery element balance pathways under the impact of new energy sources.

2. Methods

In this study, an integrated economic and environmental analysis approach is employed to assess the potential added value of coupling external hydrogen with refineries. The hydrogen sources considered in this study encompass six types sourced from both conventional fossil fuels and renewable energy, which are either widely adopted in the market or poised for large-scale implementation.

The list of hydrogen sources includes coal gasification (CG), coal gasification with carbon capture (CGwCC), natural gas steam methane reforming (SMR), partial oxidation of heavy oil (POX), biomass gasification (BioG), wind-driven electrolysis (WtH), solar-driven electrolysis (StH), and wind–solar-driven electrolysis (WStH). These hydrogen sources are blended with the hydrogen produced by refineries in the hydrogen network to fill the hydrogen gap. Three representative refineries are designed and adopted in this study: deep hydrogenation, decarbonization (Dec), and hydrogen–decarbonization co-process (Cop).

2.1. Global LP Modeling

A multi-energy-coupled modern refinery was designed to maximize the gross margin of the oil refinery by establishing a linear programming model for the entire plant. The refinery process is illustrated in Figure 1. The model consists of three main parts: distillation, secondary processing, and blending. In the distillation section, the crude distillation unit (CDU) and vacuum distillation unit (VDU) were used for the initial separation of crude oil. The processed crude oil undergoes secondary processing in units such as gas plants, naphtha hydrotreater (HDT), catalytic reforming unit (CRU), benzene extraction unit (BEU), jet fuel HDT, diesel HDT, VGO HDT, VRDS, delayed coking (DCU), hydrocracking (HCU), fluid catalytic cracker (FCC), and gasoline HDT.

To ensure ample process flexibility and optimization potential, a network for deep processing of heavy fractions has been established, as depicted in Figure 1a, outlining the plant’s topology. This facilitates the arbitrary combination of hydrogenation, decarbonization, and the two processes in any proportion by establishing material flow relationships. A LP model was established with the aim of maximizing overall plant gross margin.

(1)$\mathrm{Max}GM=\sum m_p{C}_p-\sum m_r{C}_r-\sum m_u{C}_u$

where, m_p, m_r, and m_u represent the mass flow rates of products, raw materials, and utilities (water, electricity, steam, air, auxiliary materials, fuel, and catalysts) respectively, while C_p, C_r, and C_u represent the prices of products, raw materials, and utilities.

The consumption of utilities is further correlated with the feed mass flow rates of subprocesses to facilitate the calculation of utility material consumption:

(2)$m_{u,i}^j=a·m_i^{in}$

where $m_{u,i}^j$ represents the mass flow rate of the j-th utility material consumed by unit i, $m_i^{in}$ is the feed mass flow rate of unit i, and a is the utility consumption coefficient. The side stream yields and utility consumption coefficients for all units are predefined in the model, with specific values listed in Tables S1 and S2. In the network depicted in Figure 1a, each mentioned unit will be represented by a sub-model composed of multiple linear equations, expressing side-stream yield and utility consumption in terms of feed flow rates. These sub-models will be integrated into a global LP model through the plant-wide mass balance model. The linear programming model is solved using the Python Scipy library [46].

2.1.1. Refinery Configuration

Based on the treatment of heavy fractions, refineries can be categorized into decarbonization-oriented and hydrogenation-oriented types. Decarbonization pathways typically involve FCC and DCU, whereas hydrogenation pathways include HCU and VRDS. As the world’s largest refining company, China Petrochemical Corporation (Sinopec) has equipped its mainland Chinese refineries with hydrocracking (50%), residue hydrogenation (40%), catalytic cracking (100%), and delayed coking (86%). Initially, decarbonization routes dominated early design and construction phases due to their broad feedstock tolerance, technological maturity, early adoption, and lower investment costs. However, recent years have seen rapid development in residue hydrogenation technologies driven by advancements in heavy oil processing, hydrogenation techniques, and increasingly stringent environmental regulations [23].

In this study, the mainland China region was selected as the investigation area. Based on the operational data provided by China Petrochemical Corporation, three representative refinery configurations have been defined for detailed study in this paper: the hydrogenation route, the decarbonization route, and the hydrogenation–decarbonization co-processing route. As illustrated in Table 1, the hydrogenation route omits DCU and operates FCC units at lower design loads. The decarbonization route excludes VRDS units and HCU, while co-processing refineries incorporate all units, enabling the LP model to autonomously select processing routes for heavy fractions. In practice, hydrogenation and decarbonization processes in heavy fraction processing are not mutually exclusive; most enterprises utilize both processes and achieve maximized benefits through flexible process combinations. It is important to note that the concept of a refinery in this study does not refer to a specific refinery but rather reflects a broad representation of existing technological routes.

2.1.2. Sources of Hydrogen

Eight hydrogen supply routes were examined, and these hydrogen types are classified as gray hydrogen, blue hydrogen, and green hydrogen, as depicted in Figure 1b. External hydrogen and refinery-produced hydrogen converge in the hydrogen network and are supplied to various hydrogen-consuming units. In real refineries, hydrogen usage is divided into different pressure and purity networks based on its purity and the requirements of reaction units. Hydrogen carbon footprint data provided in the literature are utilized, with the statistical results shown in Table 2.

2.1.3. Blending Model

The sulfur content, nitrogen content, gum content, density, distillation range, olefin content, aromatic content, etc., of the product are linear indicators. For properties that are linearly blendable, the product properties are estimated based on the properties of each blending component using Formula (3).

(3)$q_p^k=\sum x_{ip}{q}_i^k$

where $x_{ip}$ represents the blending ratio of the i-th blending stream in product p. Particularly, the solidifying point ($t_{sp}$), vapor pressure ($p$), pour point ($t_{pp}$), smoke point ($t_{sp}$), viscosity ($\tau$), and flash point ($t_{fp}$) do not follow a linear blending rule. They are respectively transformed into indicators through Formulas (4) to (9), and then linear blending calculation is performed using Formula (10) [48,49].

(4)$\mathit{SPI}=\exp \left[0.03\times \left(1.8\times t_{sp}+32\right)\right]$

(5)$\mathit{VPI}=p^{1.25}$

(6)$\mathit{PPI}=\exp \left[0.03\times \left(1.8\times t_{pp}+32\right)\right]$

(7)$\mathit{SmPI}=1/(t_{sp}+0.0001)$

(8)$\mathit{VI}=\ln \left(\ln \left(\tau +0.8\right)\right)$

(9)$\mathit{FPI}=\exp [-6.1188+4345.2/\left(1.8\times t_{fp}+415.0\right)]$

(10)$I_p^k=\sum x_{ip}{I}_i^k$

where $I_p^k$ represents the k-th indicator value of product p, and $I_i^k$ represents the k-th indicator value of the i-th blending stream. Specifically, due to the sensitivity of most refinery catalysts to elements such as sulfur, nitrogen, nickel, and vanadium in the feedstock, which can easily reduce reaction activity and catalyst lifespan, reasonable constraints are imposed on the feedstock streams of the unit:

(11)$q_f^{k,l}\le q_f^k\le q_f^{k,u}$

where $q_f^k$ represents the property value k of feedstock f, and $q_s^{k,l}$ and $q_s^{k,u}$, respectively, denote the upper and lower bounds constraints on this property. The product property constraints for gasoline, jet fuel, and diesel are based on the Chinese National Standards GB 17930-2016 [50], GB 6537-2018 [51], and GB 19147-2016 [52].

2.2. Carbon Emissions and Carbon Footprint Tracking Model

Carbon Footprint Lifecycle Boundary

To enable comprehensive tracking and assessment of carbon footprints across multi-energy coupling processes, this study introduces a ‘gate-to-gate’ product lifecycle carbon footprint accounting model, as depicted in Figure 2. This model encompasses pretreatment and primary processes, secondary processes, and blending stages [15].

In pretreatment and primary processes, the carbon footprint is calculated using the energy-allocation method. Initially, based on crude oil analysis data, processing volume, and the design and parameters of primary processing units obtained from Petro-Sim 7.3 software, the steam and fuel consumption and production in each side stream of the primary distillation unit are converted into corresponding carbon emissions [53]. This study employs a three-column distillation process, as illustrated in Figure 3. In the model, both the atmospheric tower and the vacuum tower are equipped with three stages of stripping, and the heat from the side products is partially recovered through heat exchange. By considering the carbon footprint generated from feedstock heat exchange energy (${CF}_{ex,i}$), heating energy (${CF}_{fur,i}$), heat released inside the distillation column (${CF}_{inre,i}$), and internal heat exchange energy of the unit (${CF}_{inex,i}$), a carbon footprint calculation model is established for each individual side stream.

(12)${CF}_i={CF}_{ex,i}+{CF}_{fur,i}-{CF}_{inre,i}-{CF}_{inex,i}$

(13)${CF}_{ex,i}=\frac{{\mathit{EF}}_Q·\sum Q_{ex}}{\sum m_i}$

(14)${CF}_{fur,i}=\frac{{\mathit{EF}}_Q·\sum Q_{fur}}{\sum m_i}$

(15)${CF}_{inre,i}=\frac{{\mathit{EF}}_Q·Q_{\mathit{inre}}}{m_i}$

(16)${CF}_{inex,i}=\frac{{\mathit{EF}}_Q·Q_{\mathit{inex}}}{m_i}$

where ${EF}_Q$, $Q_{ex}$, $Q_{fur}$, $Q_{inre}$, and $Q_{inex}$ represent the carbon footprint generated from the consumption of unit heat, energy obtained from heat exchange, energy obtained from heating, heat released inside the distillation column, and heat recovered from internal heat exchange of the unit, respectively.

Subsequently, the side streams from the distillation unit enter the secondary processing units along with purchased raw materials. The carbon footprint of the secondary processing side streams is calculated using the mass-allocation method: first, the carbon footprint of the feedstock and the carbon footprint generated by the consumption of common utilities in the unit are accounted for. Then, the carbon footprint is allocated according to the mass proportion of the side stream flows, with the carbon footprint of losses and waste being allocated to intermediate materials and products. The formula is as shown below (17):

(17)${CF}_{out}=A \times {CF}_{in}+B$

where A and B are transfer coefficients between the carbon footprint of the feedstock, CF_in, and the carbon footprint of the product, CF_out. A is assigned based on the unit properties, typically set to 1 under normal circumstances. B represents the carbon emissions from utilities and processes, calculated using Formula (18):

(18)$B=\sum _i{\mathit{CF}}_{u,i}+\sum _j{\mathit{CF}}_{f,j}$

where CF_u,i represents the carbon emission intensity of the i-th utility materials, and CF_f,j represents the carbon emission intensity of the j-th process, typically associated with the carbon emissions from catalyst coke burning. The utilities consumed by each unit and CF_u,i are listed in Table S2 and Table S3 respectively. When multiple streams converge, the carbon footprint is treated as a physical property and undergoes linear blending.

Furthermore, the integration of the carbon footprint tracking model with the global LP model achieves comprehensive carbon balance accounting for the entire plant, tracking the carbon footprint of intermediate materials, and tracing the carbon footprint of final products.

2.3. Price System

2.3.1. Price System Establishment

Crude oil procurement represents the largest cost in the refining business, as the properties of crude oil determine its processing difficulty and the yield of high-value products, thus influencing its price. Using the average price of West Texas Intermediate (WTI) crude oil from April 2020 to March 2023, provided by the U.S. Energy Information Administration (EIA) [54], as a benchmark, the price of crude oil is determined based on selected crude oil’s API, sulfur content, total acid number (TAN), and Formula (19) [48,55].

(19)$P_{\mathit{crude}}=m_1·\mathit{API}+m_2·\mathit{Sulfer}+m_3·\mathit{TAN}+m_4·P_{\mathit{WTI}}+b$

Based on historical WTI prices and product price data [56], linear regression was performed to calculate the corresponding product prices. The detailed fitting process and price data are provided in Figure S1.

2.3.2. Price Prediction

Benefiting from advancements in technology, including decreased manufacturing costs, enhanced power generation efficiency, the optimization of industrial chains, intensified market competition, and supportive government policies, the costs of wind power and photovoltaic power have witnessed a remarkable decline of 70% and 90%, respectively, over the past decade [57]. This downward trend is anticipated to persist in the foreseeable future. The reduction in power-generation costs will directly influence the price of hydrogen produced through water electrolysis [28].

At time t, the calculation of the average levelized cost of hydrogen (LCOH) is represented by Formula (20):

(20)$\mathit{LCOH}(t)=\frac{\mathit{CAPEX}·\mathit{CRF}+O\&M+F}{P_{H2}}$

where CAPEX represents the unit investment cost of the equipment; O&M denotes the operation and maintenance expenses; F stands for raw material costs, including water and electricity; and CRF is the capital recovery factor, calculated based on the discount rate i and the project lifespan T, as shown in Formula (21) [58]:

(21)$\mathit{CRF}=\frac{1}{{\sum }_{n=1}^T\frac{1}{{\left(1+i\right)}^n}}$

The reduction in future investment costs can be elucidated through learning curves. These curves embody the influence of ‘learning by doing’ on technological costs, illustrating how the unit investment cost of a technology diminishes by a consistent proportion (known as the learning rate) with each doubling of its cumulative installed capacity [58,59]. A higher learning rate signifies a more pronounced learning effect, leading to a swifter decline in costs as installed capacity expands. In the context of production processes incorporating multiple technologies, learning curves based on distinct components can be employed to delineate the phenomenon, with each technology characterized by its individual learning rate, collectively contributing to cost mitigation [57]. Formula (22) describes the learning curve for the decreasing cost Ct of hydrogen/power generation equipment:

(22)$C_t=C_0{\left(\frac{P_t}{P_0}\right)}^{-\alpha }$

(23)$lr=1-2^{-\alpha }$

where c₀, P_t, P₀, and α, respectively, represent the cost of the technology in the base year, the cumulative production/installation capacity at time t, the cumulative production/installation capacity of the technology in the base year, and the learning exponent. The learning exponent can be converted to the learning rate (lr) through Formula (23).

2.4. Case Studies

Based on the study presented in this paper regarding the catalytic role of low-carbon renewable hydrogen in driving the restructuring of refining process routes, on one hand, a refinery LP model integrating decarbonization, hydrogenation, and co-processing is developed. On the other hand, addressing the most criticized aspect of renewable hydrogen, its cost, a learning curve model is established to showcase the encouraging cost reductions driven by technological advancements such as wind and solar power generation. Similarly, the impact of external policy factors is also considered, particularly reflected in carbon taxes. Consequently, this study establishes 189 cases covering all the aforementioned considerations. As illustrated in Figure 4, this study incorporates three types of refineries (decarbonization-oriented, hydrogenation-oriented, and co-processing), seven hydrogen sources (CG, CGwCC, POX, SMR, WtH, StH, and WStH), hydrogen prices at three time points (current time C0, mid-term perspective C1, and long-term perspective C2), and carbon taxes at three time points (current Carbon Tax0, Carbon Tax1, and Carbon Tax2) as the cases under investigation.

The preset values for crude oil processing volume and raw material/product prices are listed in Table 3. Any costs not explicitly outlined in the text are reasonably estimated based on prevailing local market prices.

3. Discussion

To comprehensively assess the impact of renewable energy on the restructuring of petroleum refining routes, this study meticulously examines the economic and environmental evaluation outcomes of various hydrogen supply process routes at the present time point in Section 3.1 and Section 3.2, respectively. Building upon this groundwork, Section 3.3 meticulously scrutinizes the influence of technological cost reductions over time and the corresponding escalation in carbon emission costs on system performance. Lastly, Section 3.4 synthesizes the overarching analysis and proffers a calculation model for heavy oil hydrogenation rates, thus furnishing an exhaustive appraisal of the catalytic role of renewable low-carbon hydrogen in propelling the transformation of heavy oil processing routes.

3.1. Comparison of GM

By constructing a global linear programming model with the maximization of gross margins (GM) as the sole objective, the model optimizes product processing routes, yielding the scheme with the highest net profit. Based on this, 24 base scenarios were computed, considering 8 different hydrogen supply methods. The results are illustrated in Figure 5. As depicted in Figure 5, overall, the GM of processes exhibits a significant inverse correlation with hydrogen costs. Hydrogen, especially high-purity hydrogen, as a high-cost and high-consumption raw material, determines the GM and product processing routes in refining processes. The POX-Cop process, with the lowest hydrogen cost, achieves the highest GM at 816.53 CNY/t_oil (122.37 CNY/kg), followed by CG-Cop (784.92 CNY/t_oil (117.63 CNY/t_oil)) and SMR-Cop process (672.50 CNY/t_oil (100.78 CNY/kg)). Compared to these three high-carbon energy hydrogen production methods, profits from renewable energy hydrogen production methods, including biomass gasification, wind-driven electrolysis, solar-driven electrolysis, and wind–solar driven electrolysis, are significantly lower. The high cost of biomass raw materials, high electricity consumption in electrolysis hydrogen production processes, and the insufficient maturity of renewable energy hydrogen-related technologies contribute to the high cost of renewable hydrogen. Additionally, the application of carbon-capture technology also entails certain costs. For refineries employing gray hydrogen, the affordability of hydrogen enables the hydrogenation route to fully leverage its advantages of high conversion rates and increased product value, resulting in significantly higher GM compared to the decarbonization route. Conversely, for refineries utilizing green hydrogen, the higher cost of hydrogen favors the decarbonization process, which yields superior economic benefits. This underscores that currently, gray hydrogen sourced from fossil fuels is the most suitable hydrogen option for the hydrogenation route. Refineries adopting green hydrogen are not yet economically competitive in the short term. Therefore, expanding hydrogenation units in existing refineries appears to be a judicious choice in the short term. The LP model of co-processing refineries autonomously selects processing routes that optimize efficiency. When utilizing gray hydrogen, the GM aligns similarly to that of the hydrogenation route, while using either gray or green hydrogen allows for achieving higher GM through flexible adjustment of unit loads. The majority of refineries are equipped with both decarbonization and hydrogenation units, enabling them to withstand market fluctuations and achieve enhanced profitability.

The product compositions of all processes depicted in Figure 6 suggest a higher proportion of gasoline, jet fuel, and diesel in processes characterized by a high gross GM. A comprehensive analysis indicates that in the VGO hydrocracking process, the concurrent implementation of cracking and hydrogenation reactions effectively eliminates sulfur, nitrogen, oxygen, and metal impurities from the feedstock. This process typically achieves high conversion rates and yields a greater quantity of high-quality light oil products, particularly premium diesel and jet fuel. Furthermore, the liquid product yield of the VRDS unit surpasses 90%, providing feedstock for gasoline production in the CRU while simultaneously increasing diesel fuel output. The decarbonization route entails the extraction of carbon elements to yield light oil, concurrently generating petroleum coke products with an elevated carbon content. In the case of heavy and residue oil fractions exhibiting higher carbon-to-hydrogen ratios, the introduction of hydrogen elements can augment the production of high-value light oils, thus enhancing carbon utilization efficiency. This phenomenon often leads to hydrogenation routes gaining economic advantages over decarbonization routes in the majority of scenarios.

Hence, when both hydrogenation and decarbonization units coexist in the LP model, the model tends to opt for the hydrogenation route. Typically, as depicted in Figure 6b, all the VR from the CG-Cop process are directed to the VRDS unit. To meet the unit’s feedstock threshold, some VGO is blended into the feed while the remaining VGO is routed to the HCU. The availability of low-cost coal-derived hydrogen accentuates the advantages of the hydrogenation route, showcasing high conversion rates and elevated light oil yields.

However, the high cost of hydrogen from low-carbon energy sources makes the decarbonization route more economical than the hydrogenation route when adopting green hydrogen. When both hydrogenation and decarbonization units are configured, the loading of hydrogenation units is minimized to reduce raw material costs resulting from the high price of hydrogen. As shown in Figure 6a, the yield of light distillate products in the Bio-Cop, WtH-Cop, StH-Cop, CGwCC-Cop, and WStH-Cop processes significantly decreases compared to the hydrogenation route. Typically, as shown in Figure 6c, all VGO from the StH-Cop process is directed to the catalytic cracking unit, most of the VR is sent to the delayed coking unit, and a small portion is processed in the VRDS unit. The hydrocracking unit is completely shut down, and the VRDS unit operates at a reduced load, thereby achieving a reduction in hydrogen consumption.

Furthermore, the total hydrogen consumption and supply throughout the process were quantified and are presented in Figure 7a. In the decarbonization process, hydrogen consumption is primarily associated with the hydrotreating units, including those for jet fuel, diesel, gasoline, and naphtha, with the diesel HDT accounting for the largest share of hydrogen consumption. In the hydrogenation process, the majority of hydrogen consumption, exceeding 70%, is attributed to the HCU and VRDS. This is attributed to their increased capacity for processing heavy distillates and higher intensity of hydrogen consumption.

Through statistical analysis of the corresponding hydrogen sources, as illustrated in Figure 7b, it is observed that the cost of hydrogen not only affects the purchasing volume of externally sourced hydrogen but also influences the production of self-generated hydrogen. Self-generated hydrogen, besides originating from the CRU, also comprises a portion reclaimed through PSA units from the low-grade gas produced by the HCU and VRDS. The hydrogen network interconnects multiple units within the refinery, which directly impacts the processing route selection for the entire facility, serving as a significant factor influencing overall plant efficiency and carbon emissions.

3.2. Carbon Footprint Analysis

The carbon emissions of each unit were calculated using the carbon footprint tracking and carbon accounting model, with the results summarized in Figure 8. A comprehensive analysis indicates that the primary sources of carbon emissions in the process are hydrogen production and FCC units. The hydrogen production process from fossil fuels, exemplified by coal gasification, involves the preparation of synthesis gas, which is then subjected to the water–gas shift reaction to convert carbon monoxide and water into hydrogen and carbon dioxide. The high carbon-to-hydrogen ratio of the feedstock inevitably results in significant carbon emissions and the wastage of carbon during the production process. Another notable feature is that the carbon intensity from the decarbonization process is significantly lower than that of the hydrogenation process. In the decarbonization process, one of the main products from the core unit, the delayed coking unit (DCU), is petroleum coke, with a yield of approximately 30%. This process sequesters a substantial amount of carbon in the petroleum coke, thereby adjusting the hydrogen-to-carbon ratio. However, due to the low price and limited market demand for petroleum coke, its use as a fuel results in a higher carbon emission intensity.

The low-carbon characteristics of renewable energy significantly reduce overall plant carbon emissions. For the hydrogenation process, compared to coal gasification, coupling with renewable hydrogen reduces carbon emissions by 52% to 55%. The WtH-Cop process exhibits the lowest carbon intensity at 316.96 kgCO₂/t_oil. However, it is important to note that hydrotreated heavy oil produced from residue hydroprocessing can result in coke deposition on the catalyst surface during catalytic cracking, leading to CO₂ emissions when the coke is burned off in the regenerator section of the FCC unit. Overall, substituting fossil fuels with renewable energy in hydrogen production substantially reduces carbon emissions, significantly narrowing the carbon emission disadvantage of the hydrogenation route compared to the decarbonization route.

The carbon footprints of gasoline, jet fuel, and diesel were tracked using the carbon footprint calculation model, as shown in Figure 9. Light fuel oils, the main products of fuel-type refineries, exhibit higher carbon footprints. Among them, gasoline has the highest carbon footprint due to its longest processing chain and most complex blending components. The light fractions in crude oil, after undergoing CRU and efficient benzene extraction, can be incorporated into the gasoline blending system. Effective conversion of VGO into gasoline components is achieved through hydrocracking technology. High-residue fractions, such as VR, undergo a series of deep conversion and purification processes, including VRDS, FCC, and gasoline HDT units. These high-quality gasoline components, obtained through a series of advanced refining steps, are ultimately precisely blended into the gasoline blending pool. The long processing chain implies greater consumption of steam, fuel, and more direct carbon emissions, making gasoline the product with the highest carbon footprint. A portion of jet fuel is obtained through straight-run distillate hydrotreating, while another portion is derived from VGO hydrocracking, resulting in a production process that is both straightforward and environmentally friendly, with the lowest carbon footprint. The introduction of green hydrogen significantly reduces the carbon footprint of light fuel oil, with the most notable carbon reduction observed in the hydrogen-intensive gasoline production process.

The incorporation of green hydrogen significantly reduces the carbon footprint of light fuel oil, with the most notable impact observed in the hydrogen-intensive gasoline production process. Compared to the CG-Cop process, which has a carbon footprint of 748.03 kgCO₂/t, the WtH-Cop process achieves a 49.1% reduction, lowering the carbon footprint to 380.71 kgCO₂/t. Over the past decade, spurred by technological advancements and economies of scale, there has been a notable decrease in the cost of renewable power. Concurrently, the implementation of carbon taxes has commenced and is displaying a growing trend. Hence, in the subsequent sections, these factors will be meticulously examined.

3.3. Long-Term Analysis

3.3.1. Future Cost Trend of Hydrogen

In this section, future hydrogen prices are forecasted using the learning curve model provided in Section 2.3.2. As the case study is situated in China, data sourced from the reference are employed for the computation and fitting of the learning exponent [28]. For processes encompassing multiple subsystems with considerable potential for technological advancement, such as wind-driven electrolysis and solar-driven electrolysis, which entail both power generation equipment and electrolysis equipment, this study individually calculates the cost reduction of each subsystem. For systems with higher integration, the decrease in hydrogen production costs is directly computed. The parameters, installed capacity trends, and learning rates of the hydrogen-production technologies mentioned in this section are shown in Tables S4, S5, and S6, respectively. Currently, CG, SMR, and POX technologies have been widely commercialized. However, their potential for future cost reduction is limited, and their investment costs represent a relatively small portion of the overall hydrogen production costs. Therefore, it is assumed that the learning rate of these technologies is zero, implying that their equipment investment costs will remain unchanged in the future. This section particularly focuses on the most prominent renewable hydrogen production pathways, such as wind power electrolysis and photovoltaic power electrolysis technologies, as well as the alkaline electrolyzer (AE), which is characterized by its low cost and wide applicability.

According to the learning curve model calculations, Figure 10a–c illustrate the future capital expenditure (CAPEX) trends of PV power, wind power, and AE, respectively. Due to its high learning rate and rapid future capacity expansion, the CAPEX of PV power is anticipated to continue decreasing rapidly over the next 20 years. By 2040, it could decrease to 1661 CNY/kW, and by 2060, further reduction to 1362 CNY/kW is expected, representing a reduction of approximately 66% compared to 2020. Wind power exhibits a lower learning rate compared to PV power, resulting in a slower decrease in CAPEX. By 2040 and 2060, the CAPEX is projected to decrease to 5136 CNY/kW and 4674 CNY/kW, respectively. The CAPEX of electrolyzer systems is forecasted to decline rapidly before 2030. By 2040, the CAPEX of AE is expected to decrease to 1316 CNY/kW, representing a reduction of 78%. The International Energy Agency (IEA) predicts that, under the net-zero emission scenario, with the current learning rate, the CAPEX of electrolyzer systems is expected to drop below 2000 CNY/kW after 2050, thus supporting the validity of the predictions made in this study.

The future LCOH of various hydrogen production technologies is illustrated in Figure 10d. If coal prices remain at a relatively low level, green hydrogen LCOH will consistently exceed those of coal hydrogen production before 2055. Despite their current higher costs, the WtH, StH, and WStH routes exhibit the most significant cost-reduction trends with technological advancements, especially from 2020 to 2040. By 2040, the CGwCC, StH, and WStH routes demonstrate economic advantages over SMR, with LCOHs ranging from 13.41 to 19.93 CNY/kg (2.01 to 2.99 CNY/kg), where the WStH process maintains the lowest cost at 13.41 CNY/kg (2.01 CNY/kg) throughout. By 2056, WStH will become the lowest-cost low-carbon hydrogen production method, slightly below coal hydrogen, which currently holds the largest market share. By 2060, green hydrogen costs will further decrease to 10.49 to 17.39 CNY/kg (1.57 to 2.61 CNY/kg). The WStH process, through coupling with wind and PV power, partially mitigates the fluctuations in wind and solar energy, reducing system storage and hydrogen storage costs and thereby achieving lower hydrogen production costs, second only to POX hydrogen production by 2060. The POX hydrogen-production method maintains the lowest production costs throughout due to its reliance on refinery-derived feedstock, albeit at the expense of resource wastage. With the development of low-quality feedstock processing technologies, its market share is expected to further decrease.

Based on the trends in hydrogen cost variations, green hydrogen is likely to emerge as the most economically competitive refinery hydrogen supply method in the long term, particularly when considering carbon-taxation scenarios. The years 2020, 2040, and 2060 have been selected as the current, midterm, and long-term time points for further analysis.

3.3.2. Long-Term Economic and Environmental Performance Analysis

Based on case design and model calculations, the GM and carbon intensities for decarbonization routes, hydrogenation routes, and co-process routes were calculated, with results shown in Figure 11. When disregarding carbon emission costs, the hydrogen costs for traditional hydrogen production processes (CG, SMR, and POX) remain unchanged, resulting in no changes in the overall plant processing routes and, consequently, no changes in the overall plant benefits and carbon emissions. For decarbonization, hydrogenation, and co-process routes, the trend of benefit changes shows consistency. Benefiting from the decline in hydrogen production costs, refinery processes coupled with renewable energy (WtH, StH, and WStH routes) gradually narrow the economic performance gap with traditional hydrogen production methods, with some routes even achieving superiority. In the medium term, the StH and WStH routes have surpassed the SMR route, largely due to the rapid decline in solar power costs, which also gives the StH route an advantage over the WtH route. The BioG route is gradually losing its economic competitiveness, with still relatively high hydrogen production costs causing its GM to lag far behind other processes, reaching only about 74% of the WStH route. From a long-term perspective, the GM of green hydrogen processes exceeds that of the SMR process, and the WStH process also surpasses the CG process, becoming the second-best process in terms of economic performance, second only to POX. Analyzing from the processing route perspective, low hydrogen costs lead the co-process to adopt more hydrogenation units, leveraging the advantages of hydrogenation routes to produce more high-value products, with GM being 15–23% higher than that of decarbonization processes. Renewable energy starts to show competitiveness with traditional energy processes in medium-term prospects and becomes the dominant process in the long term. Overall, refineries exhibit high sensitivity to hydrogen prices due to their significant hydrogen consumption, particularly in hydrogenation processes. This sensitivity underscores the critical importance of reducing hydrogen costs for achieving economic competitiveness in adopting green hydrogen. If the current downward trend in green hydrogen costs persists, refineries adopting green hydrogen are projected to gradually narrow the gap with those using fossil fuels, potentially gaining economic advantages by around 2040.

It is important to note that carbon taxes are expected to become more stringent over time. A comprehensive analysis of the role of green hydrogen in promoting hydrogenation routes requires consideration of the impact of carbon taxes on the economic and environmental performance of the system.

3.3.3. Economic and Environmental Performance Analysis Considering Carbon Costs

Based on the projected carbon emission costs for China provided in the literature, corresponding carbon taxes were set at different time points in the LP model, and the entire process was optimized accordingly [28]. The results are shown in Figure 12. As depicted in the figure, the rate of increase in carbon taxes exhibits an approximately exponential growth trend over time, indicating an increasing demand for carbon reduction technologies in the energy industry.

The LP models incorporating carbon tax settings at three different time points were solved, and the benefits and carbon emissions results are illustrated in Figure 13. From the current to mid-term prospects, the significant cost-reduction effect of green hydrogen processes counteracts the additional costs incurred by increasing carbon taxes, leading to a marked increase in benefits, in stark contrast to other processes, with GM increasing by 23% to 60%. Specifically, the benefit of the WStH-Cop process reached 716.76 CNY/t_oil (107.42 CNY/t_oil), surpassing the CG process and approaching the POX process, which is 725.89 CNY/t_oil (108.79 CNY/t_oil). The imposition of high carbon taxes rapidly erodes the economic advantage of gray hydrogen-based processes, with a decrease in GM of around 15% across decarbonization, hydrogenation, and co-process routes. Although carbon capture technologies significantly alleviate the cost pressure from carbon emissions, GM still experiences a decline. Similarly, the BioG process witnesses a decrease in GM, trailing far behind other processes. The biomass-based hydrogen production process is relatively mature and lacks significant potential for technological advancements, resulting in consistently high production costs. Over the long term, this economic disadvantage indicates that it is not suitable for supplying hydrogen to large refineries. In contrast, the decrease in hydrogen supply costs for green hydrogen processes offsets the additional costs from carbon emissions, resulting in a GM increase of 23% to 60%. Meanwhile, the technical advantages of hydrogenation routes are demonstrated, with the co-process adopting processing schemes similar to hydrogenation routes. By 2060, carbon emission costs could reach as high as 1500 CNY/t_oil (224.79 CNY/t_oil), leading to varying degrees of GM decreases across all routes. Compared to 2040, traditional energy processes experience GM reductions of up to 63% to 70%, blue hydrogen processes see GM reductions of 55% to 56%, and green hydrogen processes witness GM reductions of 37% to 43%. Green hydrogen processes have already demonstrated significant advantages over other processes. Notably, the wind and solar energy coupled process (WStH-Cop) achieves a benefit of 440.23 CNY/t_oil (65.97 CNY/t_oil), far surpassing the CG-Cop process at 195.83 CNY/t_oil (29.35 CNY/t_oil).

The decarbonization route, characterized by its relatively fixed processes and its effective carbon sequestration, results in lower overall carbon emissions. Therefore, the high carbon emission costs do not impact its processing schemes and product distribution significantly. Conversely, the hydrogenation route, with its higher carbon emissions during hydrogen production, is more sensitive to the hydrogen carbon footprint. The imposition of high carbon emission costs prompts adjustments to processing routes in the LP model, aiming to minimize carbon footprint by shortening processing routes as much as possible and reducing energy consumption in high-carbon-emission units. The carbon emissions variation in co-process routes is more pronounced. In the mid-term prospect, the decline in green hydrogen prices enables it to cover the additional costs resulting from increased carbon taxes. Furthermore, despite the rapid increase in carbon taxes in the long-term prospect, the green hydrogen process can still withstand a certain level of carbon emissions intensity. This encourages an increase in the load of hydrogenation units to maximize benefits. This implies that traditional technological pathways relying on fossil fuel for hydrogen production will encounter carbon tax barriers during the transition from decarbonization to hydrogenation routes. Among all routes, only the green hydrogen route withstands the future high carbon taxes due to its superior economic and environmental performance.

3.4. The Role of Low-Carbon Renewable Energy in Driving Refinery Processing Route Transformation

To further and more comprehensively evaluate the role of the green hydrogen route in driving the transformation of refinery processing routes, the heavy fraction hydrogenation ratio (HFHR) is introduced to assess the impact of low-carbon energy on processing route changes. The HFHR represents the proportion of VGO and VR processed by HCU and VRDS units. A higher HFHR indicates a greater preference for hydrogenation routes. In the LP model, hydrogenation and decarbonization routes are not mutually exclusive; some scenarios simultaneously adopt both routes to achieve better benefits. This ratio provides a clear measure of the extent to which hydrogenation routes are employed in refinery processes, as shown in Formula (24).

(24)$\mathit{HFHR}=\frac{m_{\mathit{HCR}}+m_{\mathit{VRDS}}}{m_{\mathit{VR}}+m_{\mathit{VGO}}}$

As shown in Figure 14a, the gray hydrogen processes benefit from low hydrogen costs, achieving high efficiency through hydrogenation units, with an HFHR of 1 until 2040. In 2040, under the influence of carbon taxes, the CG-Cop and SMR-Cop routes adjust their processing schemes by reducing the load on hydrogenation units to decrease hydrogen consumption and associated carbon emissions. By 2060, when carbon taxes are at their highest, the HFHR for the three gray hydrogen routes drops to between 0.44 and 0.48. All VR is processed through the DCU, while the majority of VGO (85–95%) is directed to the HCU, with a small portion processed by the FCC unit. This balance aims to mitigate carbon emissions from hydrogen and the high-intensity emissions from the FCC unit. As depicted in Figure 14b, carbon-capture technology enables the CCwCC route to maintain an HFHR of 1 until 2050. This indicates that carbon-capture technology can partially offset the high-intensity carbon emissions associated with coal-based hydrogen production while incurring some additional costs. Within the gate-to-gate LCA boundary, the higher costs and less pronounced carbon emission intensity advantages of the BioG route make it consistently favor the decarbonization route when faced with increasing carbon taxes, particularly due to its lower hydrogen consumption. As illustrated in Figure 14c, the three green hydrogen routes exhibit similar trends. In 2020, the high cost of hydrogen production led to the selection of the decarbonization processing route. However, after 2030, the rapid decline in green hydrogen costs results in the HFHR increasing to 1 and remaining so until 2060.

Comparative analysis through HFHR reveals that the presence of high carbon taxes imposes strict requirements on both the cost and environmental aspects of hydrogen. Traditional fossil fuel routes will incur higher costs under high carbon taxes. While carbon-capture and biomass-gasification technologies can partially alleviate carbon emission costs, their long-term effectiveness may not fully offset the exponential increase in carbon taxes. Moreover, besides reducing carbon emissions, phasing out decarbonization routes leads to decreased dust and noise pollution and improves energy efficiency.

To assess and visually demonstrate the holistic performance of the eight different processes in terms of economic viability, environmental protection, processing routes, and development potential, this study employs radar charts as a visualization tool for detailed comparative analysis. The best-performing values for each indicator are set to 1, and the values for other processes are shown as proportions relative to the maximum value. Specifically, the renewable hydrogen ratio represents the proportion of renewable hydrogen in the total hydrogen consumption of the system. Carbon reduction indicates the carbon emission reduction of each process compared to the CG process, which is set to 0, with the process achieving the best carbon reduction being set to 1.

Figure 15 provides a comprehensive comparison of the economic performance, carbon emissions, and potential of each route. Currently, the gray hydrogen process demonstrates significant economic advantages, albeit with the highest carbon emission intensity. In the mid-term and long-term outlooks, the green hydrogen process starts to narrow the economic performance gap and eventually achieves superiority. The complementary coupling of wind and solar energy for hydrogen production achieves optimal overall performance. By combining these two intermittent energy sources, the inherent low-carbon advantages of wind and solar power are preserved while fluctuations are partially mitigated, thereby reducing the capacity requirements for energy storage equipment. In the mid-term, its economic benefits are second only to the POX-Cop process, and in the long-term, it achieves the best economic performance. Traditional routes using coal gasification, natural gas SMR, and heavy oil POX as hydrogen sources will gradually lose their economic advantages over time. Faced with rising carbon taxes, these routes will be forced to reduce the load on hydrogenation units, leading to a decrease in the production of high-value products and resulting in increased pollution associated with decarbonization processes.

Refineries are typical profit-driven enterprises. Currently, major enterprises are advancing the transformation of refineries from decarbonization routes to hydrogenation routes, which seems to offer better short-term benefits and higher yields of light oil. However, the high carbon emissions resulting from high hydrogen consumption are an unavoidable issue. Through the analysis in this section, it is demonstrated that the cost-reduction potential and low-carbon attributes of renewable hydrogen will assist hydrogenation route refineries in lowering their carbon intensity, thereby enhancing their economic competitiveness in the face of future high carbon taxes.

4. Conclusions

In this study, a coupled planning model was developed to integrate various renewable energy sources with traditional refineries, rigorously assessing the economic and environmental performance of representative refinery configurations involving decarbonization, hydrogenation, and co-processing. Focusing on the period from 2020 to 2060, this research examines the role of renewable energy in transforming refinery processing routes amid technological advancements and increasing carbon taxes. The findings highlight the potential of renewable energy to significantly enhance the sustainability and efficiency of refinery operations, offering valuable insights for future energy policies and industrial strategies. The main conclusions are as follows:

• Currently, refineries using fossil fuel-based hydrogen exhibit the most favorable economic performance. While adopting the hydrogenation route enhances economic benefits, it also results in the highest carbon intensity and product carbon footprint. Refineries using green hydrogen achieve the lowest carbon emissions but face economic challenges due to high hydrogen costs, which can potentially be mitigated by choosing a decarbonization route.

• The cost of green hydrogen decreases rapidly with reductions in power generation and electrolyzer costs, coupled with an increased installed capacity. By 2040, the levelized cost of hydrogen (LCOH) for the solar thermal hydrogen (StH) route is projected to drop to 16.02 CNY/kg (2.40 CNY/kg). The wind–solar thermal hydrogen (WStH) process achieves even lower costs, reaching 10.49 CNY/kg (1.57 CNY/kg) by 2060, second only to hydrogen production via partial oxidation (POX).

• Considering increasing carbon taxes, the declining prices of low-carbon green hydrogen in the mid-term enable refineries to offset additional carbon tax costs. Despite anticipated rises in carbon taxes over the long term, processes based on green hydrogen can maintain economic viability and increase the hydrogenation rate of heavy fractions for greater profitability.

Under current technological trends, green hydrogen will be the primary driver of reform in refinery process routes. Its demonstrated low carbon emissions and cost-effectiveness will steer refinery processing routes away from the currently dominant decarbonization paths and toward hydrogenation routes, thereby reducing the noise and dust pollution associated with decarbonization processes. This shift will result in cleaner refinery processes and maintain better competitiveness under pressure for carbon reduction.

Author Contributions

Conceptualization, X.S. and B.C.; methodology, X.S.; software, B.C.; validation, G.W. and X.W.; formal analysis, X.S. and X.W.; investigation, X.S.; resources, G.W. and X.W.; data curation, X.W.; writing, X.S. and B.C. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement

Data are contained within the article.

Conflicts of Interest

Author Xiaofei Shi, Gang Wang, Xiaolin Wang and Bo Chen were employed by the company SINOPEC Dalian Research Institute of Petroleum and Petrochemicals Co., Ltd. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Nomenclature

AE	Alkaline electrolyzer.
API	American Petroleum Institute.
BEU	Benzene extraction unit.
BioG	Biomass gasification.
CAPEX	Capital expenditure.
CCS	Carbon capture and storage.
CDU	Crude oil distillation unit.
CG	Coal gasification.
CGwCC	Coal gasification with carbon capture.
Cop	Co-process.
CRU	Catalytic reforming unit.
DCU	Delay coke unit.
Dec	Decarbonization.
EIA	Energy Information Administration.
FCC	Fluid catalytic cracking.
GM	Gross margin.
HCU	Hydrocrak unit.
HDT	Hydrotreater.
HFHR	Heavy fraction hydrogenation ratio.
Hyd	Hydrogenation.
LCA	Life cycle assessment.
LCOH	Levelized cost of hydrogen.
LP	Linear programming.
POX	Partial oxidation of heavy oil.
SMR	Steam methane reforming.
SOFC	Solid oxide fuel cell.
StH	Solar-driven electrolysis.
TAN	Total acid number.
VDU	Vacuum distillation unit.
VGO	Vacuum gas oil.
VR	Vacuum residue.
VRDS	Vacuum residue desulfurization.
WStH	Wind–solar-driven electrolysis.
WtH	Wind-driven electrolysis.
WTI	West Texas intermediate.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Tables

Figure 1. Refinery configurations and structure of LP models: (a) Flowsheet; (b) Hydrogen plant. The hydrogen resources are categorized into three types: Gray hydrogen, blue hydrogen, and green hydrogen.

Figure 2. Carbon footprint life cycle boundary.

Figure 3. Distillation unit process flowsheet.

Figure 4. Case design.

Figure 5. GM of the considered technological routes in the current scenario.

Figure 6. (a) Product composition; (b) CG-Cop process heavy distillate mass flow; (c) StH-Cop process heavy distillate mass flow.

Figure 6. (a) Product composition; (b) CG-Cop process heavy distillate mass flow; (c) StH-Cop process heavy distillate mass flow.

Figure 7. Hydrogen consumption: (a) Unit hydrogen consumption; (b) Hydrogen sources.

Figure 7. Hydrogen consumption: (a) Unit hydrogen consumption; (b) Hydrogen sources.

Figure 8. The carbon emission intensity of the units.

Figure 9. Primary product carbon footprint.

Figure 10. CAPEX and LCOH trends. (a) PV power CAPEX. (b) Wind power CAPEX. (c) AE CAPEX. (d) LCOH trends.

Figure 11. Long-term economic performance: (a) Decarbonization routes; (b) Hydrogenation routes; (c) Co-process routes.

Figure 12. Prediction of carbon taxes.

Figure 13. Long-term economic and environmental performance: (a) GM of decarbonization routes; (b) GM of hydrogenation routes; (c) GM of co-process routes; (d) Carbon intensity of decarbonization routes; (e) Carbon intensity of hydrogenation routes; (f) Carbon intensity of co-process routes.

Figure 14. HFHR of considered routes: (a) gray hydrogen routes; (b) blue hydrogen routes; (c) green hydrogen routes.

Figure 15. Comprehensive comparison of case processes.

Table 1

Design parameters of multi-energy coupling process.

Design Capacities	Hydrogenation Route, Mt/a	Decarbonization Route, Mt/a	Co-Process Route, Mt/a
CDU	12.0	12.0	12.0
VDU	8.7	8.7	8.7
Gas plant	1.7	1.7	1.7
Naphtha HDT	2.0	2.0	2.0
CRU	2.2	2.2	2.2
BEU	1.8	1.8	1.8
Diesel HDT	1.6	1.6	1.6
VGO HDT	1.3	1.3	1.3
VRDS	3.0	0	3.0
DCU	0	3.6	3.6
HCU	1.5	0	1.5
FCC	5.0	6.0	6.0
Gasoline HDT	3.0	3.0	3.0

Table 2

Hydrogen parameters of different energy sources.

Hydrogen Production Routes	Carbon Footprint, kgCO2/kgH2	References	Adopted in This Study, kgCO2/kgH2
CG	14.72~26.09	[27,47]	20.41
SMR	10.09~17.21	[27,28]	13.65
BioG	0.31~8.63	[27,28]	4.47
CGwCC	4.92~10.90	[28,47]	6.81
WtH	0.52~1.14	[27,28,29]	0.83
StH	1.32~2.50	[27,28,29]	1.90
WStH	1.10~1.31	[29]	1.30
POX	14.41	Estimates based on factory data	14.41

Table 3

Raw materials and product prices.

Raw Materials and Products	Value	Units
Marlim oil	6	Mt/a
Saudi Arabia heavy oil	6
Marlim oil	3037	CNY/t
Saudi Arabia heavy oil	3066
CG H2	11,000
SMR H2	18,000
BioG H2	30,000
CGwCC H2	22,000
WtH H2	32,000
StH H2	42,000
WStH H2	38,000
POX H2	9000
Gasoline	4775
Jet fuel	4767
Diesel	4281
Aromatics	6075
LPG	2787
Petroleum Coke	1383

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/pr12071317/s1, Figure S1: Product price fitting; Table S1: Side streams yields; Table S2: Utilities consumption; Table S3: Material carbon footprint; Table S4: Parameters of hydrogen production technology routes; Table S5: Installed capacity trend; Table S6: Learning rates of different technologies.

References

1. Zhao, S.; Song, Q.; Liu, L.; Li, J.; Zhao, D. Uncovering the lifecycle carbon emissions and its reduction pathways: A case study of petroleum refining enterprise. Energy Convers. Manag.; 2024; 301, 118048. [DOI: https://dx.doi.org/10.1016/j.enconman.2023.118048]

2. IEA. CO₂ Emissions in 2022; IEA: Paris, France, 2023.

3. Jing, L.; El-Houjeiri, H.M.; Monfort, J.-C.; Brandt, A.R.; Masnadi, M.S.; Gordon, D.; Bergerson, J.A. Carbon intensity of global crude oil refining and mitigation potential. Nat. Clim. Chang.; 2020; 10, pp. 526-532. [DOI: https://dx.doi.org/10.1038/s41558-020-0775-3]

4. Abdul-Manan, A.F.N.; Arfaj, A.; Babiker, H. Oil refining in a CO₂ constrained world: Effects of carbon pricing on refineries globally. Energy; 2017; 121, pp. 264-275. [DOI: https://dx.doi.org/10.1016/j.energy.2017.01.046]

5. Eslamidoost, Z.; Arabzadeh, M.; Oskoie, V.; Dehghani, S. Carbon footprint calculation in one of the largest Gas Refinery Companies in the Middle East. Environ. Sci. Pollut. Res.; 2022; 29, pp. 81609-81623. [DOI: https://dx.doi.org/10.1007/s11356-022-21482-6]

6. Ma, S.; Lei, T.; Meng, J.; Liang, X.; Guan, D. Global oil refining’s contribution to greenhouse gas emissions from 2000 to 2021. Innovation; 2023; 4, 100361. [DOI: https://dx.doi.org/10.1016/j.xinn.2022.100361]

7. Burnes, D.; Camou, A. Impact of fuel composition on gas turbine engine performance. J. Eng. Gas Turbines Power; 2019; 141, 101006. [DOI: https://dx.doi.org/10.1115/1.4044238]

8. Hussain, N. Composition of Crude Oil. Available online: https://thepetrosolutions.com/composition-of-crude-oil/ (accessed on 27 May 2024).

9. Ancheyta, J.; Sánchez, S.; Rodríguez, M.A. Kinetic modeling of hydrocracking of heavy oil fractions: A review. Catal. Today; 2005; 109, pp. 76-92. [DOI: https://dx.doi.org/10.1016/j.cattod.2005.08.015]

10. IEA. Global Energy Review: CO₂ Emissions in 2021; IEA: Paris, France, 2022.

11. Sawarkar, A.N.; Pandit, A.B.; Samant, S.D.; Joshi, J.B. Petroleum Residue Upgrading Via Delayed Coking: A Review. Can. J. Chem. Eng.; 2007; 85, pp. 1-24. [DOI: https://dx.doi.org/10.1002/cjce.5450850101]

12. Chen, Y.-M. Recent advances in FCC technology. Powder Technol.; 2006; 163, pp. 2-8. [DOI: https://dx.doi.org/10.1016/j.powtec.2006.01.001]

13. Weitkamp, J. Catalytic Hydrocracking—Mechanisms and Versatility of the Process. ChemCatChem; 2012; 4, pp. 292-306.

14. Xing, B.; Ye, L.; Liu, J.; Qin, X.; Yu, W.; Xie, J.; Hou, L.; Wang, H.; Ji, Y.; Lu, D. et al. Reaction network of sulfur compounds in delayed coking process. Chem. Eng. J.; 2021; 422, 129903. [DOI: https://dx.doi.org/10.1016/j.cej.2021.129903]

15. Muñoz, J.A.D.; Aguilar, R.; Castañeda, L.C.; Ancheyta, J. Comparison of Correlations for Estimating Product Yields from Delayed Coking. Energy Fuels; 2013; 27, pp. 7179-7190. [DOI: https://dx.doi.org/10.1021/ef4014423]

16. Dongdong, C.; Xingchun, L.; Ming, X.; Wenjia, X. Emission characteristics and environmental implication of volatile organic compounds from delayed coking unit. Pet. Refin. Eng.; 2022; 52, 58.

17. Speight, J. Visbreaking: A technology of the past and the future. Sci. Iran.; 2012; 19, pp. 569-573. [DOI: https://dx.doi.org/10.1016/j.scient.2011.12.014]

18. Lee, J.M.; Shin, S.; Ahn, S.; Chun, J.H.; Lee, K.B.; Mun, S.; Jeon, S.G.; Na, J.G.; Nho, N.S. Separation of solvent and deasphalted oil for solvent deasphalting process. Fuel Process. Technol.; 2014; 119, pp. 204-210. [DOI: https://dx.doi.org/10.1016/j.fuproc.2013.11.014]

19. Sadighi, S.; Ahmad, A. An optimisation approach for increasing the profit of a commercial VGO hydrocracking process. Can. J. Chem. Eng.; 2013; 91, pp. 1077-1091. [DOI: https://dx.doi.org/10.1002/cjce.21720]

20. Threlkel, R.; Dillon, C.; Singh, U.G.; Ziebarth, M. Increase flexibility to upgrade residuum using recent advances in RDS/VRDS-RFCC process and catalyst technology. J. Jpn. Pet. Inst.; 2010; 53, pp. 65-74. [DOI: https://dx.doi.org/10.1627/jpi.53.65]

21. Okunev, A.G.; Parkhomchuk, E.V.e.; Lysikov, A.I.; Parunin, P.D.; Semeykina, V.; Parmon, V.N. Catalytic hydroprocessing of heavy oil feedstocks. Russ. Chem. Rev.; 2015; 84, 981. [DOI: https://dx.doi.org/10.1070/RCR4486]

22. Matsushita, K.; Goto, Y.; Takamura, T.; Iwama, M. Development of Residue Upgrading Technology Integrated Solvent De-asphalting and Residue Hydrotreating System: De-asphalted Oil Properties and Reactivity Estimation. J. Jpn. Pet. Inst.; 2020; 63, pp. 308-314. [DOI: https://dx.doi.org/10.1627/jpi.63.308]

23. Semih Eser. Processing and Conversion of Vacuum Distillation Residue. Available online: https://www.e-education.psu.edu/fsc432/content/processing-and-conversion-vacuum-distillation-residue (accessed on 27 May 2024).

24. Jia, X.; Zhang, Y.; Tan, R.R.; Li, Z.; Wang, S.; Wang, F.; Fang, K. Multi-objective energy planning for China’s dual carbon goals. Sustain. Prod. Consum.; 2022; 34, pp. 552-564. [DOI: https://dx.doi.org/10.1016/j.spc.2022.10.009]

25. Hao, J.; Gao, F.; Fang, X.; Nong, X.; Zhang, Y.; Hong, F. Multi-factor decomposition and multi-scenario prediction decoupling analysis of China’s carbon emission under dual carbon goal. Sci. Total Environ.; 2022; 841, 156788. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2022.156788]

26. Pareek, A.; Dom, R.; Gupta, J.; Chandran, J.; Adepu, V.; Borse, P.H. Insights into renewable hydrogen energy: Recent advances and prospects. Mater. Sci. Energy Technol.; 2020; 3, pp. 319-327. [DOI: https://dx.doi.org/10.1016/j.mset.2019.12.002]

27. Parkinson, B.; Balcombe, P.; Speirs, J.F.; Hawkes, A.D.; Hellgardt, K. Levelized cost of CO₂ mitigation from hydrogen production routes. Energy Environ. Sci.; 2019; 12, pp. 19-40. [DOI: https://dx.doi.org/10.1039/C8EE02079E]

28. Wang, Y.Z.; Ou, X.M.; Zhou, S. Future cost trend of hydrogen production in China based on learning curve. Clim. Chang. Res.; 2022; 18, pp. 283-293.

29. Shi, X.; Qian, Y.; Yang, S. Fluctuation analysis of a complementary wind–solar energy system and integration for large scale hydrogen production. ACS Sustain. Chem. Eng.; 2020; 8, pp. 7097-7110. [DOI: https://dx.doi.org/10.1021/acssuschemeng.0c01054]

30. Muñoz Díaz, M.T.; Chávez Oróstica, H.; Guajardo, J. Economic Analysis: Green Hydrogen Production Systems. Processes; 2023; 11, 1390. [DOI: https://dx.doi.org/10.3390/pr11051390]

31. Liu, W.; Wan, Y.; Xiong, Y.; Gao, P. Green hydrogen standard in China: Standard and evaluation of low-carbon hydrogen, clean hydrogen, and renewable hydrogen. Int. J. Hydrogen Energy; 2022; 47, pp. 24584-24591. [DOI: https://dx.doi.org/10.1016/j.ijhydene.2021.10.193]

32. Wu, X.D.; Yang, Q.; Chen, G.Q.; Hayat, T.; Alsaedi, A. Progress and prospect of CCS in China: Using learning curve to assess the cost-viability of a 2 × 600MW retrofitted oxyfuel power plant as a case study. Renew. Sustain. Energy Rev.; 2016; 60, pp. 1274-1285. [DOI: https://dx.doi.org/10.1016/j.rser.2016.03.015]

33. Qin, K.; Ye, S.; Wu, L. Process design and analysis of a net-zero carbon emissions hydrocracking unit integrating co-processing technique with green hydrogen and electricity. Energy; 2024; 295, 131080. [DOI: https://dx.doi.org/10.1016/j.energy.2024.131080]

34. Shi, X.; Wang, Q.; Yang, S.; Qian, Y. Low-carbon path of geographically matched hybrid energy utilization: A coal-to-ethylene glycol process with hydrogen from the coupled wind/solar power. ACS Sustain. Chem. Eng.; 2021; 9, pp. 4583-4599. [DOI: https://dx.doi.org/10.1021/acssuschemeng.0c09263]

35. Nascimento da Silva, G.; Rochedo, P.R.R.; Szklo, A. Renewable hydrogen production to deal with wind power surpluses and mitigate carbon dioxide emissions from oil refineries. Appl. Energy; 2022; 311, 118631. [DOI: https://dx.doi.org/10.1016/j.apenergy.2022.118631]

36. Mastropasqua, L.; Pegorin, A.; Campanari, S. Low fuel utilisation solid oxide fuel cell system for CO2-free hydrogen production in oil refineries. J. Power Sources; 2020; 448, 227461. [DOI: https://dx.doi.org/10.1016/j.jpowsour.2019.227461]

37. Shi, T.; Gu, L.; Xu, Z.; Sheng, J. Research on Multi-Objective Energy Management of Renewable Energy Power Plant with Electrolytic Hydrogen Production. Processes; 2024; 12, 541. [DOI: https://dx.doi.org/10.3390/pr12030541]

38. Zhang, N.; Li, Z.; Zheng, X.; Liu, P. How Would Structural Change in Electricity and Hydrogen End Use Impact Low-Carbon Transition of an Energy System? A Case Study of China. Processes; 2024; 12, 437. [DOI: https://dx.doi.org/10.3390/pr12030437]

39. Shih, C.F.; Zhang, T.; Li, J.; Bai, C. Powering the future with liquid sunshine. Joule; 2018; 2, pp. 1925-1949. [DOI: https://dx.doi.org/10.1016/j.joule.2018.08.016]

40. Sinopec. Sinopec Xinjiang Kuqa Green Hydrogen Pilot Project Enters Operation, Leading China’s Green Hydrogen Development. Available online: http://www.sinopecgroup.com/group/en/Sinopecnews/20230704/news_20230704_299217593563.shtml (accessed on 27 May 2024).

41. BP. BP and Ørsted to Create Renewable Hydrogen Partnership in Germany. Available online: https://www.bp.com/en/global/corporate/news-and-insights/press-releases/bp-and-orsted-to-create-renewable-hydrogen-partnership-in-germany.html (accessed on 15 June 2024).

42. Shell. Shell Hydrogen Projects. Available online: https://www.shell.com/what-we-do/hydrogen.html (accessed on 15 June 2024).

43. FuelCellsWorks. PetroChina’s First Large-Scale Hydrogen Project Starts. Available online: https://fuelcellsworks.com/news/petrochinas-first-large-scale-hydrogen-project-starts/ (accessed on 15 June 2024).

44. Khor, C.; Elkamel, A. Superstructure optimization for oil refinery design. Pet. Sci. Technol.; 2010; 28, pp. 1457-1465. [DOI: https://dx.doi.org/10.1080/10916460903096806]

45. Khor, C.; Yeoh, X.Q.; Shah, N. Optimal design of petroleum refinery topology using a discrete optimization approach with logical constraints. J. Appl. Sci.; 2011; 11, pp. 3571-3578. [DOI: https://dx.doi.org/10.3923/jas.2011.3571.3578]

46. SciPy: Open Source Scientific Tools for Python. Available online: https://docs.scipy.org/doc/scipy/reference/generated/scipy.optimize.linprog.html (accessed on 27 May 2024).

47. Li, J.; Wei, Y.-M.; Liu, L.; Li, X.; Yan, R. The carbon footprint and cost of coal-based hydrogen production with and without carbon capture and storage technology in China. J. Clean. Prod.; 2022; 362, 132514. [DOI: https://dx.doi.org/10.1016/j.jclepro.2022.132514]

48. Carlson, N.A.; Singh, A.; Talmadge, M.S.; Jiang, Y.; Zaimes, G.G.; Li, S.; Hawkins, T.R.; Sittler, L.; Brooker, A.; Gaspar, D.J. et al. Economic analysis of the benefits to petroleum refiners for low carbon boosted spark ignition biofuels. Fuel; 2023; 334, 126183. [DOI: https://dx.doi.org/10.1016/j.fuel.2022.126183]

49. Jiang, Y.; Zaimes, G.G.; Li, S.; Hawkins, T.R.; Singh, A.; Carlson, N.; Talmadge, M.; Gaspar, D.J.; Ramirez-Corredores, M.M.; Beck, A.W. et al. Economic and environmental analysis to evaluate the potential value of co-optima diesel bioblendstocks to petroleum refiners. Fuel; 2023; 333, 126233. [DOI: https://dx.doi.org/10.1016/j.fuel.2022.126233]

50. GB 17930-2016 Standardization Administration of China; Gasoline for Motor Vehicles: Beijing, China, 2017.

51. GB 6537-2018 Standardization Administration of China; No.3 Jet Fuel: Beijing, China, 2018.

52. GB 19147-2016 Standardization Administration of China; Automobile Diesel Fuels: Beijing, China, 2016.

53. Tian Tao, R.X.; Wang, L.; Tian, Z.; Wang, B. Study on the Carbon Footprint Evaluation of Total Product for the Atmospheric and Vacuum Distillation Unit and Its Application. Acta Pet. Sin. Pet. Process. Sect.; 2023; 39, pp. 279-288.

54. EIA. Petroleum & Others Liquids Spot Price. Available online: https://www.eia.gov/dnav/pet/PET_PRI_SPT_S1_D.htm (accessed on 28 May 2024).

55. Khani, L.; Mohammadpourfard, M. Investigation of a New Methanol, Hydrogen, and Electricity Production System Based on Carbon Capture and Utilization. Energy Systems Transition: Digitalization, Decarbonization, Decentralization and Democratization; Vahidinasab, V.; Mohammadi-Ivatloo, B. Springer International Publishing: Cham, Switzerland, 2023; pp. 87-129.

56. EIA. Refiner Petroleum Product Prices by Sales Type. Available online: https://www.eia.gov/dnav/pet/pet_pri_refoth_dcu_nus_m.htm (accessed on 27 May 2024).

57. IRENA. Renewable Power Generation Costs in 2022; IRENA: Abu Dhabi, United Arab Emirates, 2023.

58. Detz, R.J.; Reek, J.N.H.; van der Zwaan, B.C.C. The future of solar fuels: When could they become competitive?. Energy Environ. Sci.; 2018; 11, pp. 1653-1669. [DOI: https://dx.doi.org/10.1039/C8EE00111A]

59. Rivera-Tinoco, R.; Schoots, K.; van der Zwaan, B. Learning curves for solid oxide fuel cells. Energy Convers. Manag.; 2012; 57, pp. 86-96. [DOI: https://dx.doi.org/10.1016/j.enconman.2011.11.018]