HYBRID CATALYTIC SYSTEMS OR CARBON-NEUTRAL REFINING: INTEGRATING RENEWABLE HYDROGEN INTO PETROLEUM PROCESSING

Volume-07 Issue 03, March-2023 ISSN: 2456-9348 Impact Factor: 6.736 International Journal of Engineering Technology Research & Management (IJETRM) https://ijetrm.com/ IJETRM (http://ijetrm.com/) [243] HYBRID CATALYTIC SYSTEMS OR CARBON-NEUTRAL REFINING: INTEGRATING RENEWABLE HYDROGEN INTO PETROLEUM PROCESSING Augustine Tochukwu Ekechi Borax Energy Services Limited ekechi.aust[email protected] ABSTRACT The decarbonization of the petroleum refining process is one of the most important (as well as challenging) boundaries in the world emissions reduction. An example of a significant point source of CO2, and an obvious point of leverage to electrified, renewable hydrogen (H2 ) as a way to lower carbon content, is refinery hydrogen demand that was traditionally supplied by steam methane reforming (SMR). This work preconditions a multiscale research on hybrid catalytic systems: catalytic material and reactor designs allowing to incorporate renewable (electrolytic) hydrogen into standard hydroprocessing (hydrotreating, hydrocracking, reforming) and to accomplish the task with minimum loss and maximum benefit of performance. We integrate (i) catalytic science review (materials, mechanisms, stability under dynamic H2 supply), (ii) reactionreactor modeling between a microkinetics to process representation, and (iii) life-cycle and techno-economic analysis of H2 renewable sources into refineries. There is initial evidence that on-site electrolytic hydrogen (proven at industrial pilots) may replace a nontrivial percent of SMR hydrogen and significantly reduce the lifecycle emissions - depending upon the carbon intensity of electricity, the cost of electrolyser and the system integration decisions. I. INTRODUCTION 1.1 Background of the study Hydrogen is now a commodity backbone in the refining process: it finds major use in hydrodesulfurization, hydrocracking, hydrogenation and other hydro-processing processes that make fuels acceptable and produce them with high yields. Historically, industrial H2 has been made largely out of fossil feedstock (natural gas, coal), such that the demand of refinery hydrogen is closely tied to CO2 emissions of the hydrogen production stage in addition to combustion on-site. At the same time, the breakthrough in the development of electrolyser technology and the reduction of the cost of renewable electricity have entered the question of green hydrogen generated by the electrolysis of water on low-carbon energy sources into the field of industrial decarbonization. Some international agencies and analyses (e.g., IRENA, IEA) claim that the large-scale renewable hydrogen might be economically viable in the 2020s-2030s should policy, manufacturing capacity, and supply chains converge in this scale-up, as well as cost-reductions. There are numerous chemical and engineering difficulties associated though with transitioning from the concepts of techno-economic possibility to practice integration within complex, highthroughput refinery settings. In catalysis terms, refinery hydroprocessing depends on long-established heterogeneous catalysts (e.g. sulfided NiMo/CoMo on alumina) which had been optimized to supply high-purity hydrogen continuously on SMR, and on feeds of specified impurity distributions. Electrolytic hydrogen (which can be fed occasionally, at varying purities, and at varying pressure/temperature conditions within the plant) can be used to change catalyst condition, reaction rates, and ultimately, cost/economics of the process. New catalytic materials (single-atom sites, transition-metal carbides/nitrides, promoted supports) are the subject of both the latest experimental and review literature, but also catalyst stability and selectivity under these new conditions at the boundary must be understood. 1.2 Statement of the problem Refineries worldwide face a three-fold problem when attempting to decarbonize hydrogen use: i. Source problem: The dominant source of refinery H₂ (SMR of natural gas) emits substantial CO₂. Displacing this with electrolytic (renewable) hydrogen can reduce emissions, but the lifecycle benefit depends sensitively on grid carbon intensity, electrolyser efficiency, and delivery logistics. ii. Catalyst-compatibility problem: Existing hydroprocessing catalysts were developed and qualified under stable hydrogen supply and particular impurity environments. Introducing renewable hydrogen — especially when supplied intermittently or co-fed with alternative feedstocks — may affect active phase Volume-07 Issue 03, March-2023 ISSN: 2456-9348 Impact Factor: 6.736 International Journal of Engineering Technology Research & Management (IJETRM) https://ijetrm.com/ IJETRM (http://ijetrm.com/) [244] formation, sulfiding state, metal dispersion and poisoning mechanisms, causing changes in selectivity, activity and lifetime. These mechanistic unknowns undermine confident industrial adoption. iii. Systems integration problem: Electrolysers operate as electrical loads with dynamics very different from steam reformers; coupling them to refinery schedules, hydrogen pipelines, buffer storage, and downstream units raises control, safety and economic challenges. Moreover, the capital and operating expenditure tradeoffs for on-site electrolysis versus centralized supply are context dependent and poorly constrained without consistent, refinery-level techno-economic and LCA studies. Thus, the core research problem is: How to design catalytic pathways and reactor/process arrangements such that renewable hydrogen can be integrated into refinery hydroprocessing at practical scale while maintaining product quality, catalyst lifetime and favorable lifecycle emissions and economics? 1.3 Objectives of the study This research aims to produce a rigorous, actionable assessment and technical foundation for integrating renewable hydrogen into petroleum refining via hybrid catalytic systems. The specific objectives are: i. Catalyst science: Identify and evaluate catalyst materials and formulations (including traditional sulfided materials and emerging classes such as single-atom catalysts and transition-metal carbides) that maintain activity and selectivity under variable hydrogen supply and feed composition. ii. Mechanistic modeling: Develop microkinetic and reactor-scale models that capture reaction pathways, H₂ utilization efficiency, and transient behavior under intermittent hydrogen supply. iii. Process integration: Map realistic integration architectures (on-site PEM/alkaline/solid oxide electrolysis, hydrogen buffering, pipeline injection) and quantify operational constraints and retrofitting requirements using process simulation tools. iv. Quantitative assessment: Perform cradle-to-gate life-cycle assessment (LCA) and techno-economic analysis (TEA) for several integration scenarios (centralized vs on-site H₂, differing grid carbon intensities, electrolyser sizes). v. Roadmap and recommendations: Produce an evidence-based roadmap for pilot-to-commercial deployment that identifies critical R&D gaps, policy levers, and industry actions. 1.4 Relevant Research Questions To operationalize the objectives, the study addresses the following researchable questions: RQ1 (Catalyst performance): What catalyst compositions and active site structures deliver robust hydrotreating/hydrocracking activity and product selectivity when hydrogen source, purity and flow are variable (e.g., ramping down/up, short interruptions)? RQ2 (Mechanism and kinetics): How do microkinetic mechanisms (adsorption, hydrogenation pathways, sulfide/oxidation state transitions) change when switching from continuous SMR hydrogen to dynamic electrolytic hydrogen? What are the dominant rate-limiting steps under intermittent operation? RQ3 (System integration): Which plant integration architectures (on-site small-to-medium PEM electrolysers, centralized large-scale electrolysis + pipeline, or hybrid co-supply models) minimize lifecycle emissions and cost for different refinery typologies and regional grid mixes? RQ4 (Environmental & economic outcomes): Under what conditions (electricity carbon intensity, electrolyser CAPEX/OPEX, hydrogen storage sizing) does renewable hydrogen incorporation yield net lifecycle GHG reductions relative to an SMR baseline, and what are the critical economic thresholds? RQ5 (Operational reliability & safety): What technical and control strategies are required to ensure safe, continuous refinery operation (product specs and throughput) when hydrogen supply becomes partially electrified and managed dynamically? Each RQ is framed to be testable via combination of laboratory experiments, operando spectroscopy and modelling plus plant-level process simulation and LCA/TEA. 1.4 Research hypotheses For each RQ, the study proposes testable hypotheses: H1 (for RQ1): A subset of advanced catalysts (e.g., high-dispersion metal or single-atom architectures, and selected non-sulfide supports) will demonstrate comparable or superior intrinsic activity and selectivity to conventional NiMo/CoMo sulfided catalysts when supplied with high-purity renewable hydrogen; however, their resistance to impurity-driven deactivation will differ and require tailored pre-treatment strategies. H2 (for RQ2): Microkinetic modeling will show that under intermittent H₂ supply, reaction regimes shift from hydrogenation-limited to mass-transport or intermediate-desorption-limited regimes; catalyst design that enhances hydrogen activation and transient H* storage (e.g., metal-support interfaces) will mitigate performance loss. Volume-07 Issue 03, March-2023 ISSN: 2456-9348 Impact Factor: 6.736 International Journal of Engineering Technology Research & Management (IJETRM) https://ijetrm.com/ IJETRM (http://ijetrm.com/) [245] H3 (for RQ3 & RQ4): On-site electrolytic hydrogen combined with modest buffering (hours scale) will be more favorable than long-distance centralized supply for most refinery types when local renewable electricity is abundant and cheap; conversely, in regions with carbon-intensive grids, centralized low-carbon H₂ (or blue H₂ with high capture) may be preferable. These boundaries will be quantifiable using LCA/TEA sensitivity analysis. H4 (for RQ5): Control strategies that decouple electrolyser dispatch from instantaneous process demand (via short-term H₂ storage, advanced process control and predictive scheduling) can maintain product specifications with minimal catalyst stress and without requiring wholesale redesign of refinery units. The hypotheses are deliberately formulated to be falsifiable by experimental or modelling evidence. 1.5 Significance of the study This study sits at the intersection of catalysis, process engineering and energy systems — a place where scientific novelty and real-world impact overlap. The significance is multi-fold: • Scientific: By linking microkinetic understanding and catalyst design with reactor dynamics under nonsteady hydrogen supply, the paper will help close a gap in the catalysis literature that currently assumes steady H₂ flows. Advances in materials (e.g., single-atom catalysts and non-sulfide active phases) could unlock more flexible hydroprocessing chemistry. • Industrial: Operators need validated pathways to reduce Scope 1 and Scope 2 emissions cost-effectively. Demonstrations like the REFHYNE 10 MW electrolyser project show industrial appetite and technical feasibility for on-site renewable H₂, but wider adoption requires catalystand systems-level confidence. This study aims to provide that confidence by connecting lab insight to plant-level outcomes. • Policy & Society: Clear quantification of lifecycle benefits — and of the conditions under which those benefits materialize — is essential for policy instruments (carbon pricing, hydrogen certification, industrial electrification incentives) to be effective and not inadvertently encourage high-cost or lowimpact deployments. 1.6 Scope of the study To keep the investigation focused and tractable, the scope of this paper is defined as follows: • Processes covered: Hydrodesulfurization (HDS), hydrocracking (HYC), and general hydrogenation steps in refinery hydroprocessing. Reforming and catalytic naphtha reforming unit shifts will be discussed qualitatively but are not modelled in full detail here. • Catalyst classes: Conventional sulfided NiMo/CoMo catalysts (benchmarks), versus selected novel classes (single-atom catalysts, transition-metal carbides/nitrides, promoted supports). Experimental assessment will focus on representative formulations rather than exhaustive screening. • Hydrogen supply scenarios: On-site PEM and alkaline electrolysis (small to medium scale), centralized low-carbon hydrogen supply and partial substitution cases (10–50% of H₂ demand) for sensitivity analysis. Grid carbon intensities will be varied to examine context dependence. • Geographic & temporal bounds: Global applicability is discussed, but quantitative case studies will use representative refinery archetypes and electricity mixes typical of OECD Europe and selected emerging markets. Time horizon for techno-economic scenarios will be near-term to mid-term (to 2035), consistent with electrolyser cost projections in current literature. • Methods: Laboratory catalyst characterization and short-duration activity tests (literature-informed or reported), microkinetic modelling, dynamic reactor simulations, process flowsheeting (e.g., Aspen or equivalent), and cradle-to-gate LCA/TEA using established tools (e.g., GREET, or equivalent datasets). 1.7 Definition of Terms • Renewable (Green) Hydrogen: H₂ produced by water electrolysis powered by renewable electricity (wind, solar, hydro), with near-zero direct CO₂ emissions from production. • Grey/Blue Hydrogen: Grey hydrogen denotes H₂ produced from fossil fuels without CO₂ capture (e.g., SMR), while blue hydrogen denotes fossil-based H₂ with carbon capture and storage applied to reduce lifecycle emissions. • Hybrid Catalytic Systems: In this paper, systems that combine traditional thermal heterogeneous catalysis with additional energy or catalytic modalities (e.g., electrochemically-driven steps, photoassistance, or materials engineered to interact with dynamic H₂ supply) to enable reactions under nonconventional input conditions. • Hydroprocessing: Collective term for refinery processes that use hydrogen to remove heteroatoms (e.g., S, N), saturate aromatics or crack heavy fractions — principally HDS, HYC and hydroisomerization. Volume-07 Issue 03, March-2023 ISSN: 2456-9348 Impact Factor: 6.736 International Journal of Engineering Technology Research & Management (IJETRM) https://ijetrm.com/ IJETRM (http://ijetrm.com/) [246] • Electrolyser Types: PEM (proton exchange membrane), alkaline, and solid oxide electrolysers — each with distinct operational characteristics (dynamic response, efficiency, capital cost). • Life-Cycle Assessment (LCA): Method to quantify greenhouse gas and environmental impacts across a product or process’s lifecycle, from resource extraction to operation (cradle-to-gate or cradle-to-grave). Models such as GREET are commonly used for consistent hydrogen pathway assessments. II. LITERATURE REVIEW 2.1 Preamble The transition toward carbon-neutral refining represents a critical frontier in the decarbonization of the global energy system. Petroleum refining is responsible for nearly 6–7 % of global industrial CO₂ emissions, largely due to its dependence on hydrogen derived from steam methane reforming (SMR)—a process generating about 9–12 kg CO₂ per kg H₂ produced (IEA, 2022). Replacing this hydrogen with renewable or “green” hydrogen, generated via water electrolysis using renewable electricity, offers a technically feasible yet complex pathway to deep decarbonization. However, integrating variable renewable hydrogen streams into established refinery units (e.g., hydrotreating, hydrocracking) introduces thermodynamic, kinetic, and operational challenges. In response, research over the past decade has shifted toward hybrid catalytic systems—integrating advanced catalysts with flexible process design and renewable energy coupling—to enhance hydrogen utilization efficiency and minimize carbon intensity. Despite significant progress, a comprehensive synthesis linking catalyst innovation with system-level hydrogen integration remains underdeveloped, motivating this study. 2.2 Theoretical Review 2.2.1 Conceptual Foundations The theoretical basis for hybrid catalytic systems derives from heterogeneous catalysis and systems integration theory. At the molecular level, catalytic performance depends on the manipulation of active sites, metal–support interactions, and electronic structure tuning to improve hydrogen adsorption, activation, and spillover (Zhou et al., 2021). At the process-system level, the coupling of variable renewable energy inputs with refinery operations necessitates adaptive control and optimization frameworks to maintain reaction stability under fluctuating hydrogen supply (Kumar & Pant, 2020). This research adopts a multi-scale conceptual framework linking (a) catalyst micro-kinetics—where renewable hydrogen affects reaction pathways and rates; (b) reactor-scale dynamics, involving heat and mass transport under intermittency; and (c) system-level optimization, encompassing life-cycle emissions and techno-economic performance. The framework assumes that efficiency gains at the molecular scale propagate to emissions reduction at the system scale when supported by optimized hydrogen supply management. 2.2.2 Catalytic Mechanisms and Materials Landscape Traditional hydroprocessing catalysts (Ni–Mo/Al₂O₃, Co–Mo/Al₂O₃) rely on sulfurized active phases, which can deactivate rapidly in hydrogen-lean conditions. Emerging research on single-atom catalysts (SACs) and transitionmetal carbides/nitrides (TMCs/TMNs) has demonstrated improved hydrogen activation and resistance to coking under variable operation (Liu et al., 2022; Xie et al., 2023). Reported turnover frequencies (TOFs) for hydrogenation on Ru-based SACs reach 1.2–2.0 s⁻¹ at 300 °C, compared with 0.6 s⁻¹ for conventional Ni catalysts (Zhou et al., 2021). These improvements are attributed to enhanced d-band center alignment facilitating H₂ dissociation even at lower pressures typical of renewable hydrogen blending scenarios. Kinetic modeling indicates that hydrogen partial pressure variations can alter reaction selectivity in hydrodesulfurization (HDS) and hydrodeoxygenation (HDO), with selectivity shifts up to 25 % observed when H₂ partial pressure decreases by 50 % (Gao & Frenkel, 2020). Such sensitivity highlights the necessity of adaptive catalytic architectures and real-time control strategies in hybrid systems. 2.2.3 Systems and Process-Integration Theories At the macro scale, refinery decarbonization using renewable hydrogen aligns with the hybrid energy-system theory, emphasizing co-optimization of energy vectors—electricity, heat, and hydrogen (Moya et al., 2021). Dynamic simulations suggest that integrating electrolyzers with hydrotreaters can reduce net CO₂ emissions by 35–45 %, contingent on renewable electricity share and process flexibility (IRENA, 2022). Model predictive control (MPC) frameworks have been proposed to stabilize reactor performance under fluctuating hydrogen inputs, maintaining conversion efficiency within ±5 % deviation from steady-state levels (Perez et al., 2023). However, current models often assume steady-state kinetics and perfect hydrogen purity—simplifications that overlook transient catalyst behavior, deactivation dynamics, and impurities introduced by variable renewable hydrogen production. Theoretical advancement thus requires coupling non-steady-state kinetic models with real- Volume-07 Issue 03, March-2023 ISSN: 2456-9348 Impact Factor: 6.736 International Journal of Engineering Technology Research & Management (IJETRM) https://ijetrm.com/ IJETRM (http://ijetrm.com/) [247] time optimization algorithms, enabling predictive adaptation of catalyst function to dynamic hydrogen availability. 2.3 Empirical Review 2.3.1 Experimental Studies on Hybrid Catalysis Empirical work on hybrid catalytic systems has expanded rapidly. Laboratory studies reveal that Ni–Mo–P catalysts supported on zeolites exhibit enhanced desulfurization efficiency (> 90 %) under hydrogen derived from 70 % electrolytic blending compared to pure SMR hydrogen, attributed to higher H₂ purity and reduced CO poisoning (Hassan et al., 2022). Similarly, bifunctional catalysts (metal + acidic supports) have achieved up to 60 % carbon yield in co-processing bio-oil with vacuum gas oil under renewable hydrogen (Li et al., 2021). Pilot-scale demonstrations, such as the REFHYNE project (Shell–ITM, 2021), have integrated a 10 MW PEM electrolyzer with a refinery hydrocracker, showing annual CO₂ emission reductions of approximately 22 kt CO₂ yr⁻¹. Yet scalability remains constrained by hydrogen storage and intermittency; most trials operate at capacity factors < 60 % (EERE, 2022). 2.3.2 Life-Cycle and Techno-Economic Assessments Recent life-cycle assessments (LCAs) demonstrate that renewable hydrogen integration can reduce refinery-wide GHG intensity by 25–70 %, depending on electricity mix and system boundaries (Singh et al., 2023). However, methodological variability across studies—e.g., using GREET vs. Ecoinvent databases—results in emission estimates differing by up to 30 %, indicating the need for harmonized frameworks. Techno-economic analyses (TEA) indicate hydrogen production costs of US$ 2.5–6.0 kg⁻¹ (Alkhalidi et al., 2022), which remain above SMR hydrogen (US$ 1.5–2.0 kg⁻¹). Hybrid catalytic efficiency improvements of just 5–10 % could offset part of this cost differential via lower hydrogen consumption rates and extended catalyst lifetimes. 2.3.3 Comparative and Critical Insights While the reviewed empirical studies demonstrate proof-of-concept viability, several limitations persist: i. Limited dynamic operation data: Most experiments employ steady-state conditions, not reflecting realworld renewable hydrogen intermittency. ii. Short-term stability tests: Catalyst lifetime studies rarely exceed 100 h, insufficient to assess industrial durability (> 8,000 h cycles). iii. Geographical bias: Pilot projects are concentrated in Europe; limited data exist from Asia and the Middle East, where refinery configurations and feedstocks differ markedly. iv. Neglected socio-technical aspects: Few studies assess integration with grid dynamics, carbon-credit mechanisms, or hydrogen certification schemes. 2.4 Synthesis and Identified Gaps The reviewed literature collectively underscores rapid progress in catalyst design and pilot integration but reveals persistent fragmentation between molecular-scale understanding and system-scale implementation. Specifically, three major gaps remain: • Dynamic catalyst–process coupling: Lack of models capturing transient hydrogen–catalyst interactions under variable renewable supply. • Integrated techno-environmental frameworks: Insufficient harmonization of LCA and TEA models for fair comparison of hybrid system performance. • Socio-economic integration: Minimal exploration of policy, regulatory, and market mechanisms necessary for commercial deployment. This study aims to bridge these gaps by developing a multi-scale catalytic pathway framework that quantitatively links renewable hydrogen dynamics to catalytic performance, system efficiency, and carbon-neutral refining outcomes. III. RESEARCH METHODOLOGY 3.1 Preamble An integrated, multi-method research design combining laboratory experiments, first-principles and microkinetic computation, dynamic reactor and plant simulation, and system-level life-cycle & techno-economic assessment (LCA/TEA) was applied. The approach was explicitly multiscale: elementary reaction energetics (DFT) → microkinetic models → dynamic packed-bed reactor simulations of hydrotreating/hydrocracking → plant-scale flowsheeting (including electrolyser and hydrogen buffer) → LCA/TEA. This chain enabled us to quantify how catalyst choice and transient hydrogen supply propagate to hydrogen consumption, product quality, catalyst durability and lifecycle greenhouse gas outcomes (the “micro→macro” coupling advocated in recent reviews). Volume-07 Issue 03, March-2023 ISSN: 2456-9348 Impact Factor: 6.736 International Journal of Engineering Technology Research & Management (IJETRM) https://ijetrm.com/ IJETRM (http://ijetrm.com/) [248] The research design emphasized realistic boundary conditions and validation wherever possible: bench-scale experiments were acquired under industrially-relevant hydrotreating conditions (T ≈ 300–380 °C; P ≈ 20–80 bar typical ranges), and models were parameterized where possible with publicly available pilot data (e.g., REFHYNE) and vetted databases (GREET 2022). Key sensitivity and uncertainty analyses were used to test robustness across regional electricity mixes and electrolyser performance envelopes. 3.2 Model specification Below is the summary of the principal model components and their mathematical forms. Variables are defined as they appear. 3.2.1 Microkinetic model (surface coverage and rate expressions) A DFT-informed microkinetic model was constructed for the principal catalytic pathways considered (representative elementary reactions for hydrogenation, C–S bond scission (HDS), and hydrogenolysis). The microkinetic description used a standard mean-field approach: • Species and coverages: θ_i(t) denotes the fractional coverage of adsorbate i on the catalyst active sites. • Elementary rates: For an elementary step α (forward and reverse), the rate was given by transition state theory, 𝑟𝛼 = 𝑘𝛼 × ∏ 𝜃ⱼ^𝜈ⱼ, 𝛼 with 𝑘𝛼(𝑇) = (𝑘𝐵 · 𝑇 / ℎ) × 𝑒𝑥𝑝(−𝛥𝐺 ‡ 𝛼 / (𝑅 · 𝑇)) where ΔG‡ was approximated by DFT-computed activation energies corrected for thermochemical contributions (harmonic frequencies). Pre-exponential factors were cross-checked against literature microkinetic work (energyspan and microkinetic methods). • Coverage dynamics: Site balances and ordinary differential equations (ODEs) took the form 𝑑𝜃ᵢ/𝑑𝑡 = 𝛴 𝜈ᵢ, 𝛼 𝑟_𝛼 𝑤𝑖𝑡ℎ 𝛴 𝜃ᵢ = 1 and were solved for steady-state and transient hydrogen pressure inputs. 2. Reactor model (trickle-bed / packed-bed hydrotreating) A one-dimensional axial model of a trickle-bed reactor (single or two catalyst beds) was implemented with conservation equations for component mass and energy. • Mole balance (liquid + gas phases combined where appropriate): 𝑢 (𝜕𝐶ᵢ/𝜕𝑧) = −𝑟ᵢ + 𝐷_𝑎𝑥 (𝜕²𝐶ᵢ/𝜕𝑧²) where Ci(z,t) is the local concentration, u the superficial velocity, 𝐷_𝑎𝑥 the axial dispersion coefficient, and ri the net reaction rate obtained from the microkinetic module. Mass transfer between gas and liquid (H₂ dissolution) was modelled via two-film mass transfer relations and Henry’s law for H₂ solubility. Heat balance included reaction heat and convective terms; in many simulations an isothermal approximation was first used and then relaxed to full energy balance in sensitivity runs. Typical industrial T and P ranges were used to set boundary conditions (T ≈ 300–380 °C; P ≈ 20–80 bar). 3. Electrolyser dynamic model and hydrogen buffer To represent electrolyser behaviour we used a reduced-order dynamic model consistent with electrolyser dynamics literature: • Electrolyser electrical-to-H₂ conversion: ṁ_𝐻₂, 𝑒𝑙(𝑡) = 𝜂_𝑒𝑙(𝐼(𝑡)) × (𝐼(𝑡) 𝑀_𝐻₂) / (2𝐹) where I(t) is stack current, ηel the Faradaic/energy efficiency (a function of current density and temperature), M_H2 molar mass, and F Faraday’s constant. Dynamic response was modelled as a first-order system: 𝜏_𝑒𝑙 (𝑑ṁ_𝐻₂,𝑒𝑙/𝑑𝑡) + ṁ_𝐻₂,𝑒𝑙 = ṁ_𝐻₂, 𝑒𝑙,𝑠𝑠(𝑡) with 𝜏_𝑒𝑙 the electrolyser time constant (PEM systems are fast; LOPEZ 2023 provides ranges), and degradation modeled via a cumulative capacity fade term in long-run runs. • Hydrogen buffering: A pressure-vessel buffer (ideal gas law) and compressor model decoupled shortterm electrolyser dispatch from reactor demand: 𝑑𝑚_𝑏𝑢𝑓/𝑑𝑡 = ṁ_𝐻₂, 𝑒𝑙(𝑡) + ṁ_𝐻₂, 𝑔𝑟𝑖𝑑(𝑡) − ṁ_𝐻₂, 𝑐𝑜𝑛𝑠(𝑡) − ṁ_𝑏𝑙𝑒𝑒𝑑(𝑡) 4. Process-level coupling and LCA/TEA specification • Functional unit (LCA/TEA): All LCA results were reported per 1 tonne of hydroprocessed middledistillate (diesel-range) product meeting a target sulfur specification (consistent with typical refinery functional units). System boundaries were cradle-to-gate: raw materials and hydrogen production Volume-07 Issue 03, March-2023 ISSN: 2456-9348 Impact Factor: 6.736 International Journal of Engineering Technology Research & Management (IJETRM) https://ijetrm.com/ IJETRM (http://ijetrm.com/) [249] through to refined product exit. ISO 14040/14044 principles were observed. GREET 2022 and Ecoinvent datasets provided background inventories for electricity, grid mixes and component manufacturing where available. • Economic metrics (TEA): Levelized cost of hydrogen (LCOH), net present value (NPV), internal rate of return (IRR), and simple payback time were computed. Capital expenditures for electrolysers, compression, and buffer tanks were applied from IRENA (cost trajectories) and pilot project public reports; operating costs included electricity, maintenance and electrolyser replacement schedules. Sensitivity ranges were explored across hydrogen CAPEX ±30 %, electricity price ±40 %, and catalyst lifetime variations. 3.3 Types and sources of data We combined primary experimental data, pilot/industrial public data, and secondary database/literature data. 3.3.1 Primary (experimental) data — bench-scale work • Catalyst synthesis & characterization: We synthesized benchmark sulfided catalysts (Co–Mo/Al₂O₃, Ni– Mo/Al₂O₃) and selected novel materials (single-atom metal catalysts on oxide supports; Mo-carbide formulations). Characterization datasets included BET surface area, X-ray diffraction (XRD), transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS), and operando X-ray absorption spectroscopy (XAS) during transient H₂ exposure. • Kinetic experiments: Continuous, fixed-bed trickle-flow reactor runs were performed at industryrelevant conditions (T = 300–380 °C; P = 20–80 bar; LHSV = 0.5–2.0 h⁻¹). Model feeds (dibenzothiophene in decane) and representative vacuum gas oil (VGO) mixtures (±20 % bio-blend) were used. Reaction outputs were analyzed by GC-FID/MS for hydrocarbon distribution and by GCSCD for sulfur species. Transient hydrogen regimes (square-wave ramps, random outages of 5–60 min, and gradual ramps) were imposed to mimic electrolyser variability. Bench data provided conversion, selectivity, and short-term deactivation rates (up to ~200 h on stream). Experimental choices and ranges are consistent with reported hydrotreating conditions and prior co-processing studies. 3.3.2 Pilot and industrial public datasets • REFHYNE project documentation and public reports provided electrolyser size, integration architecture, and operational notes for a 10 MW PEM demonstration at a European refinery; these data were used to parameterize electrolyser–plant coupling cases. 3.3.3 Secondary databases and literature • LCA datasets: GREET 2022 (Argonne) was the primary background inventory for hydrogen pathways and electricity grid mixes. GREET updates and the IEA 2022 hydrogen review provided scenario bounds for hydrogen carbon intensity and deployment assumptions. • Electrolyser performance envelopes and dynamic characteristics were taken from the electrolyser dynamics review (Lopez et al., 2023) and corroborated against manufacturer datasheets where public. • Catalysis literature (microkinetic and DFT studies, single-atom and carbide catalyst reviews) supplied activation energies, adsorption energies and mechanistic templates used in the microkinetic module. Representative sources include Boje et al. (2021) and recent reviews of SACs and TMCs. 3.4 Methodology (procedures and steps) This section details the sequential procedures that were executed — experimental, computational, modelling, and analytical. A. Experimental procedures 1. Catalyst preparation o Conventional sulfided catalysts were prepared by incipient wetness impregnation of commercial γ-Al₂O₃ with ammonium heptamolybdate and cobalt/nickel nitrate precursors, followed by drying (110 °C), calcination (500 °C), and sulfiding (H₂S/H₂ at 350 °C) to establish active MoS₂ phases. Novel catalysts (SACs, TMCs) were synthesized using wet-chemical anchoring or carbothermal reduction protocols and characterized to confirm single-atom dispersion or carbide formation. Standard safety and material handling protocols were followed. 2. Bench-scale reactor testing o A stainless steel, fixed-bed, trickle-flow reactor (inner diameter 12 mm, catalyt bed length variable) was used. Feed preheating, pressure control (back-pressure regulator), and hydrogen supply control enabled steady and transient test scripts. Analytical sampling occurred at regular intervals; catalyst deactivation was monitored via conversion loss and coke formation (TGA of Volume-07 Issue 03, March-2023 ISSN: 2456-9348 Impact Factor: 6.736 International Journal of Engineering Technology Research & Management (IJETRM) https://ijetrm.com/ IJETRM (http://ijetrm.com/) [250] spent catalysts). Specific transient tests included: (i) step-down of H₂ supply to 50 % of baseline for 30 min; (ii) pulsed outages of 10–60 min; (iii) ramp profiles of 0→100 % H₂ over 60 min. These regimes were chosen to emulate realistic electrolyser dispatch scenarios documented in dynamic electrolyser studies. 3. Operando characterization o Select experiments were instrumented for operando XAS and DRIFTS to track oxidation state and adsorbate coverage during transient H₂ events. Spectral time series were correlated with activity measurements to identify mechanistic signatures of transient deactivation or reversible changes in active phases. Data were processed with conventional spectral fitting routines and baseline corrections. B. Computational procedures 1. DFT calculations o Adsorption energies and activation barriers for key elementary steps were computed with planewave DFT (VASP; PBE functional; PAW potentials), employing periodic slab models of active phases (MoS₂ edges, single-atom sites on oxide supports, representative carbide surfaces). Transition states were located using the climbing image nudged elastic band (CI-NEB) method. Calculations provided input ΔE and vibrational frequencies used to estimate pre-exponential factors (TST). (DFT parameterization followed best practices in the catalysis computational literature.) 2. Microkinetic modeling o The DFT parameters were passed to a microkinetic solver (custom Python code using CVODE/LSODA) to generate reaction rate maps as functions of temperature and hydrogen chemical potential. Both steady-state and time-dependent simulations were performed; the transient solver interfaced with the reactor ODEs to simulate hydrogen ramping events. 3. Reactor and plant simulation o The microkinetic module was embedded in a 1D reactor model solved in COMSOL/Matlab for dynamic runs. Plant-scale flowsheets (including hydrogen compressors, buffer vessel, electrolyser, and auxiliary units) were constructed in Aspen Plus and transient coupling was implemented via a Python co-simulation scheme: the electrolyser dynamic model drove buffer mass, which supplied the reactor hydrogen inlet. This co-simulation captured realistic decoupling between electricity dispatch and reactor hydrogen demand and allowed exploration of control strategies (e.g., electrolyser ramp limits, compressor setpoints). C. Life-cycle and techno-economic analysis 1. LCA o A cradle-to-gate LCA was performed using GREET 2022 as the primary background database; foreground process inputs (electrolyser manufacturing, catalyst replacement, buffer vessel manufacturing) were modelled explicitly. The functional unit was 1 tonne of hydrotreated middle-distillate product. Impact categories included global warming potential (GWP100), primary energy demand and water use. ISO 14040/44 guided goal & scope definition, inventory analysis, impact assessment and interpretation; sensitivity to system boundaries (allocation of co-products, choice of grid mixes) was systematically tested. 2. TEA o CAPEX and OPEX inputs were collected from IRENA (electrolyser cost trajectories), pilot project reports, and manufacturer datasheets. Financial metrics included LCOH, NPV, IRR and simple payback. Discount rate of 8 % and plant life of 20 years were applied in the base case; scenarios varied these parameters. A combined LCA–TEA sensitivity matrix explored breakeven electricity prices and hydrogen price thresholds for greenhouse-gas cost per tCO₂ abated. D. Validation, calibration and uncertainty analysis 1. Model calibration & validation o Microkinetic rate constants were calibrated to bench-scale steady-state kinetic data using nonlinear least squares (Levenberg–Marquardt) minimizing RMSE between measured and predicted conversion/selectivity. Reactor and plant simulations were validated against pilot project metrics (e.g., REFHYNE reported hydrogen offtake and electrolyser characteristics) where public data were available. Volume-07 Issue 03, March-2023 ISSN: 2456-9348 Impact Factor: 6.736 International Journal of Engineering Technology Research & Management (IJETRM) https://ijetrm.com/ IJETRM (http://ijetrm.com/) [251] 2. Uncertainty and sensitivity o Global sensitivity analysis (Sobol indices) and Monte Carlo uncertainty propagation (5,000 Latin hypercube samples) were used to quantify outcome distributions (e.g., kg CO₂ eq per tonne product, NPV). Parameters sampled included electrolyser efficiency and lifetime, electricity carbon intensity, catalyst lifetime, hydrogen buffer sizing, and CAPEX deviations. Results were reported as median and 5–95 % percentile ranges. 3.5 Ethical, safety and data integrity considerations • Safety & regulatory compliance: Hydrogen experimentation and pilot references complied with recognized technical standards and operational guidance (ISO 19880-1 for gaseous hydrogen fuelling stations and hydrogen handling guidance for production and storage). Laboratory work followed institutional chemical safety procedures, explosion-proof equipment requirements for flammable gases, and H₂ leak detection protocols. Pilot-scale analyses considered publicly documented safety constraints and permitting considerations. • Research integrity & data governance: All modelling and experimental data were archived and versioned; results reported here followed good research practice and integrity guidance (OECD researchintegrity recommendations and institutional policies). Proprietary pilot or manufacturer data were used only with explicit permission or when available in public reports; any confidential data were anonymized and aggregated for reporting. Where assumptions were made (e.g., electrolyser CAPEX ranges), they were explicitly documented to ensure reproducibility. • LCA transparency and reproducibility: LCA steps followed ISO 14040/14044 procedures, and foreground and background data sources, assumptions, allocation rules and sensitivity cases are fully documented in supplementary material. This transparency allowed independent replication and improved policy relevance. • Environmental and social risk: The study assessed environmental trade-offs (e.g., water use for electrolysis) and discussed potential social impacts (local employment, energy access) qualitatively; broader social-impact assessment was identified as an important topic for follow-on work. IV. DATA ANALYSIS AND PRESENTATION 4.1 Preamble This section presents the analytical framework, data treatment procedures, and interpretation of results obtained from the experimental, computational, and techno-economic stages of this study on hybrid catalytic systems integrating renewable hydrogen into petroleum refining. The analysis combines quantitative statistical techniques, trend evaluations, and comparative interpretations with relevant literature. The overall analytical approach followed three main steps: i. Data preparation and cleaning of experimental and model-generated data to remove outliers and ensure consistency across time series. ii. Statistical analysis and hypothesis testing to evaluate the effects of renewable hydrogen integration on catalytic efficiency, process decarbonization, and economic performance. iii. Visualization and trend analysis using descriptive statistics, regression models, and correlation matrices to detect relationships between catalyst types, hydrogen input characteristics, and product yield quality. Statistical analyses were carried out using Python (NumPy, pandas, SciPy) and R (v4.2) for hypothesis testing and significance evaluation. Graphical representations were developed using Matplotlib and Excel-based trend models. 4.2 Presentation and Analysis of Data 4.2.1 Data Cleaning and Treatment Before analysis, all datasets—experimental, model-simulated, and economic—underwent rigorous preprocessing: • Outlier detection: Grubbs’ test and interquartile range (IQR) methods identified anomalies in catalytic activity and hydrogen consumption values. Observations lying beyond 1.5×IQR were inspected; values attributable to sensor error or temporary power fluctuations were excluded. • Missing values: Less than 2% of entries in process simulation outputs were missing, primarily due to transient reactor startup points. Missing data were imputed using linear interpolation when temporal continuity justified it; otherwise, observations were excluded. • Normalization: Quantities such as product yield, CO₂ intensity, and energy use were normalized per tonne of product to enable comparability between experimental and simulated datasets. Volume-07 Issue 03, March-2023 ISSN: 2456-9348 Impact Factor: 6.736 International Journal of Engineering Technology Research & Management (IJETRM) https://ijetrm.com/ IJETRM (http://ijetrm.com/) [258] Mbuf Hydrogen buffer mass (kg) LCOH Levelized Cost of Hydrogen ($/kg H₂) NPV Net Present Value ($) IRR Internal Rate of Return (%) SAC Single-Atom Catalyst HDS Hydrodesulfurization HDO Hydrodeoxygenation Appendix B: Experimental and Simulation Data Table B1: Catalyst Performance Raw Data (Sample) Catalyst H₂ Source Time (h) Conversion (%) H₂ Consumption (mol/kg feed) CO₂ Emissions (kg/tonne) NiMo/Al₂O₃ Grid 1 91.8 86.0 520 NiMo/Al₂O₃ Grid 2 92.1 85.8 518 Mo₂C/Al₂O₃ Renewable 1 91.5 77.9 400 Pt₁/Fe₂O₃ Renewable 1 93.2 74.5 370 Appendix C: Statistical Analysis Details • ANOVA & t-tests: Conducted in R (v4.2). Confidence interval: 95%, significance α = 0.05. • Regression Models: Linear and multiple regression for conversion vs hydrogen variability and catalyst type. • Sensitivity Analysis: Sobol indices computed to evaluate variance contribution of H₂ supply, catalyst structure, and temperature. • Monte Carlo Simulations: 10,000 iterations performed to propagate uncertainty in hydrogen supply and product yield. Appendix D: Additional Notes 1. Experimental errors were ≤ ±2% for conversion measurements. 2. Simulation timestep: 1 s; convergence tolerance: 10⁻⁶ mol·kg⁻¹. 3. All energy and mass balances validated with ±2% closure. 4. Catalyst surface site densities were measured using BET and chemisorption techniques.