Simulation-Based Multi-Dimensional Evaluation of Ethanol as an Alternative Fuel for Marine Energy Systems

Abstract

The maritime sector accounts for approximately 3% of global greenhouse gas (GHG) emissions and faces binding decarbonization obligations under the International Maritime Organization’s (IMO) Net-Zero Framework and the FuelEU Maritime Regulation. Conventional marine fuels, including very low sulphur fuel oil (VLSFO) and liquefied natural gas (LNG), are insufficient to meet long-term regulatory intensity targets on a well-to-wake (WtW) lifecycle basis, creating an urgent need for credible fuel alternatives. This study investigates ethanol as a primary fuel for marine dual-fuel propulsion systems, assessed across four distinct production pathways, sugar beet, corn, sugarcane, and wheat straw, to determine its full decarbonization potential relative to VLSFO and LNG benchmarks. A simulation-based multi-dimensional evaluation framework is developed and applied, integrating dynamic operational simulation, energy analysis, environmental lifecycle modelling, and regulatory compliance assessment. The framework is calibrated against a high-resolution dataset from an active container ship, with scenario-specific engine data. While ethanol requires 39.1% more fuel mass than VLSFO due to its lower energy density, all four ethanol pathways deliver substantially superior WtW GHG reductions: from 50.2% (corn) to 76.9% (wheat straw), compared with 20.6% for LNG. All ethanol scenarios satisfy FuelEU compliance limits across the 2026–2045 horizon, with wheat straw ethanol achieving a GFI of 22.52 gCO₂e/MJ, compliant marginally with the 2040 IMO target. These findings demonstrate that bio-based ethanol, particularly from lignocellulosic feedstocks, is a technically viable and regulatorily superior alternative to LNG for maritime decarbonization, warranting accelerated research into production scale-up and bunkering infrastructure development.

1. Introduction

The global shipping industry, which accounts for approximately 3% of worldwide greenhouse gas (GHG) emissions, faces intensifying pressure to pursue a pathway toward decarbonization aligned with international climate commitments and the United Nations Sustainable Development Goals (SDGs) [1]. The International Maritime Organization (IMO) revised its GHG reduction strategy in 2023, setting net-zero emission targets for international shipping by or around 2050, with interim reduction checkpoints of at least 20% (striving for 30%) by 2030 and at least 70% (striving for 80%) by 2040, relative to 2008 baseline levels [2,3]. In April 2025, a landmark agreement was reached to adopt a legally binding framework comprising a global fuel standard and an emissions pricing mechanism, scheduled for entry into force in 2027, further cementing the industry’s obligations under the Paris Agreement [4]. Complementing these strategic commitments, mandatory technical and operational instruments, including the Energy Efficiency Design Index (EEDI) [5,6], the Energy Efficiency Existing Ship Index (EEXI) [7,8], and the Carbon Intensity Indicator (CII) [9,10], compel naval architects, ship operators, and fleet managers to pursue low-carbon propulsion solutions, novel technologies and cleaner fuel alternatives [11,12]. Among the fuels under active investigation for decarbonizing marine operations, liquefied natural gas (LNG), methanol, ammonia, hydrogen, and bioethanol have emerged as the principal candidates capable of delivering meaningful GHG reductions within the IMO framework [8].

The enforcement of the IMO 2020 global sulphur cap, combined with escalating EEXI and CII compliance requirements, has accelerated a broad industry transition away from conventional marine fuels such as heavy fuel oil (HFO) and very low sulphur fuel oil (VLSFO) toward cleaner substitutes [13,14]. LNG has gained considerable traction as a near-term transitional fuel given its comparatively lower carbon dioxide (CO₂) emissions profile and its relatively mature bunkering infrastructure; however, persistent concerns regarding methane slip and upstream extraction emissions continue to moderate its long-term decarbonization potential [14]. Beyond alternative fuels, the maritime sector is also exploring a range of renewable energy pathways to support long-term decarbonization objectives. Among these, ocean renewable energy technologies, particularly tidal current energy, have attracted increasing attention due to their high predictability, energy density, and potential contribution to low-carbon electricity generation for future integrated maritime energy systems [15,16].

Renewable liquid fuels, particularly bio-based alcohols such as methanol and ethanol, present compelling alternatives by offering reduced carbon footprints, compatibility with existing combustion infrastructure, and relative ease of handling [17,18]. Ethanol, in particular, has attracted growing interest owing to its renewable origin, elevated octane rating, and the breadth of viable production pathways. These pathways span first-generation (1G) processes utilizing food-based substrates such as sugarcane and corn [19], second-generation (2G) routes leveraging lignocellulosic agricultural residues [20], and third-generation (3G) technologies based on microalgal biomass [21]. While ethanol has been widely deployed in road transport fuel blending and is increasingly explored in dual-fuel (DF) marine engine applications [22,23], its adoption at scale within the maritime sector remains constrained relative to other alternative fuels.

Dual-fuel engines have emerged as one of the most promising technological pathways for near-term decarbonization in the shipping sector, owing to their inherent capability to switch between gaseous and liquid fuels while sustaining performance levels broadly comparable to conventional marine diesel engines [24,25]. The bulk of contemporary research in this space concentrates on LNG-diesel DF configurations [26], methanol-diesel blended systems [27], or ammonia-based propulsion [28]. Ethanol has been somewhat sidelined in these investigations, largely on account of its lower volumetric energy density, its mildly corrosive properties with respect to certain materials, and the limited scale of dedicated marine-grade ethanol production infrastructure [29,30]. Nevertheless, ethanol presents a compelling set of intrinsic attributes, including near-zero sulphur content, an inherently oxygenated molecular structure, and the possibility of achieving carbon-neutral or even net-negative lifecycle emissions via sustainable production routes, that position it as a credible candidate for short-to-medium-term decarbonization strategies [31]. Recent commercial developments have reinforced this potential: in late 2025, Everllence successfully demonstrated a small-bore four-stroke DF engine operating on ethanol, reporting that the ethanol fuel share surpassed that achieved with methanol during testing, underscoring growing industry confidence in ethanol’s technical viability [32].

The net environmental performance of ethanol as a marine fuel is fundamentally dependent upon the feedstock origin and the specific production pathway pursued [33]. First-generation ethanol derived from food crops such as corn or sugarcane carries a comparatively heavy lifecycle GHG burden due to agricultural inputs, processing energy demand, and direct as well as indirect land-use change [34]. Second-generation ethanol produced from lignocellulosic residues and waste biomass offers substantially improved carbon credentials, avoiding the land-use conflicts inherent to 1G production [20]. Third-generation ethanol derived from microalgae presents the most favourable theoretical lifecycle emissions profile, with studies suggesting the potential for net-negative CO₂ outcomes when carbon captured by algal cultivation is credited against overall emissions [21,35]. The environmental differentiation across these generational pathways is therefore critical when evaluating ethanol against competing marine fuels under full lifecycle accounting frameworks, an assessment requirement now formalized by the IMO’s 2025 agreement to apply well-to-wake emission criteria to its Global Fuel Standard [4].

Previous studies have examined various dimensions of ethanol’s applicability to marine propulsion, though a coherent and integrated assessment remains absent from the literature. Early technical appraisals established ethanol’s fundamental feasibility as a marine fuel while simultaneously identifying persistent limitations in bunkering logistics and material compatibility [36,37]. Economic comparisons for DF operation suggested that ethanol could be cost-competitive under specific market conditions [38], while lifecycle analyses of 2G wood-derived ethanol demonstrated clear human health and resource-use benefits despite moderate ecosystem-level trade-offs [39]. Subsequent investigations drew attention to the gap between the theoretical promise of 3G algae-based ethanol and the near-term commercialization barriers associated with large-scale microalgal cultivation and processing [40]. In parallel, simulation-based approaches have proven indispensable for evaluating DF engine behaviour across a range of alternative fuel conditions. Computational thermodynamic modelling, zero-dimensional cycle simulation, and computational fluid dynamics (CFD) have collectively enabled researchers to characterize fuel consumption, emissions profiles, and combustion stability in DF engines operating on LNG [24], methanol [27], and ammonia [28], without incurring the costs and operational complexities of full-scale sea trials. These simulation methodologies are increasingly recognized as the most resource-effective means of generating high-fidelity predictions of engine performance and emissions across varied fuel compositions and loading conditions [41].

Notwithstanding these individual contributions, the existing body of literature exhibits a notable absence of comprehensive, simulation-driven assessments that simultaneously address the technical performance, lifecycle environmental footprint, and regulatory compliance of ethanol in large-scale marine DF propulsion contexts. Prior studies have predominantly examined isolated parameters, combustion behaviour, or single-feedstock lifecycle emissions, without integrating these dimensions into a unified evaluation framework. Moreover, most assessments have been limited to small laboratory-scale or automotive engines, with very few addressing the specific thermodynamic and operational characteristics of large low-speed two-stroke marine engines representative of oceangoing vessels. This gap is particularly significant given that the IMO’s evolving regulatory architecture, anchored by the IMO NZF, demands precisely this kind of multi-dimensional evidence base to support credible fuel transition decisions. Furthermore, the sector-wide reliance on LNG as the default transitional fuel warrants critical re-examination in light of mounting evidence regarding upstream methane slip and the maturing availability of advanced biofuel pathways.

To address this identified gap, the present study develops and applies a simulation-based multi-dimensional evaluation framework for assessing ethanol as an alternative fuel within a large marine DF propulsion system. A conceptual engine retrofit scenario is constructed by replacing a conventional two-stroke marine diesel engine with a gas-injection DF equivalent capable of operating on ethanol alongside pilot fuel. The simulation framework quantifies specific gas and pilot oil consumption, along with direct exhaust emissions of CO₂, methane (CH₄), and nitrous oxide (N₂O), and also incorporates upstream GHG contributions from the fuel supply chain to determine the total CO₂-equivalent (CO₂e) lifecycle footprint. The GHG intensity factor of IMO NZF and FuelEU are calculated to assess regulatory compliance, and comparative analyses against VLSFO and LNG are carried out to illuminate the operational versus upstream emissions trade-offs across different ethanol feedstock generations. The simulation methodology is grounded in an annual high-resolution operational dataset sourced from an active container ship, providing an empirically calibrated baseline that substantially exceeds the fidelity achievable through purely theoretical approaches. It is explicitly noted that ethanol–MGO blended operation is outside the scope of the present study; the investigation is confined to neat ethanol as the primary dual-fuel fuel supply, and blended co-injection strategies are reserved for future work.

2. Case Study Description

2.1. Vessel Under Investigation

The vessel selected as the case study for this investigation is a cellular container ship representative of the large oceangoing cargo carriers deployed on intercontinental trade routes. This class of vessel occupies a strategically important position within the global container fleet, as it is responsible for a disproportionately large share of total transport work and the emissions associated with it. Ultra-large container ships account for an estimated 25% of global container carrying capacity [42]. These vessels are predominantly assigned to high-frequency, long-haul trade corridors, including the Asia–Europe, Transpacific, and Transatlantic routes, where the combination of large payload and extended voyage distances produces an exceptionally high Twenty-foot Equivalent Unit (TEU)-mile demand per vessel. This concentration of cargo throughput on distance-intensive services renders ultra-large container ships a primary target for decarbonization interventions, since even incremental improvements in their GHG intensity translate into substantial absolute emission reductions at the fleet level. The principal particulars of the vessel under investigation are provided in

Table 1. Principal particulars of the container ship under investigation.

Parameter	Symbol	Value	Unit
Length Between Perpendiculars	Lpp	350	m
Beam	B	48.2	m
Summer Load Draft	T	15.5	m
Maximum Container Capacity	TEU	15,000	TEU
Deadweight (Summer)	DWT	145,527	t
Displacement	D	187,974	t

[43].

The vessel’s primary propulsion is provided by a single, direct-drive, two-stroke low-speed diesel engine, a MAN B&W 12K98ME-7 (Everllence, Augsburg, Germany), with a maximum continuous rating (MCR) of 71,770 kW. Under the baseline configuration, this engine operates on very low sulphur fuel oil (VLSFO) in conventional diesel mode. Onboard electrical demand, encompassing propulsion auxiliaries, navigation and communication systems, cargo cooling and ventilation, hotel loads, and deck machinery, is supplied by three four-stroke medium-speed auxiliary generator sets, each comprising a Doosan 9L32/40 (Doosan, Changwon, Republic of Korea) prime mover rated at 4500 kW MCR and fuelled by marine gas oil (MGO). The three-unit auxiliary configuration provides redundancy and operational flexibility, enabling a rotating dispatch strategy in which the number of sets in service is matched to the instantaneous electrical demand.

The operational dataset was obtained from the vessel’s onboard data logging system, which records engine power output, and navigational parameters at one-hour intervals. The dataset covers a complete annual round-trip cycle encompassing four operational modes: sea passage, port, canal transit, and anchor/waiting.

2.2. Retrofit Scenarios Under Investigation

To enable a systematic comparative assessment, six propulsion fuel scenarios are defined and evaluated within the multi-dimensional simulation framework. These comprise one VLSFO baseline, one LNG benchmark, and four ethanol variants, differentiated solely by production pathway. All retrofit scenarios share an identical engine replacement concept: the existing MAN B&W 12K98ME-7 and auxiliary engines are replaced by a dual-fuel (DF) equivalent capable of operating on the designated primary fuel with MGO as pilot ignition fuel. A summary of all six scenarios is provided in

Table 2. Summary of fuel scenarios investigated in this study.

Scenario	ID	Primary Fuel	Pilot Fuel	Pathway Generation	Benchmark/Alternative
VLSFO Baseline	S0	VLSFO	None	Fossil	Baseline
LNG	S1	LNG	MGO	Fossil	Benchmark
Ethanol—Sugar Beet	S2	Bioethanol	MGO	1G	Alternative
Ethanol—Corn	S3	Bioethanol	MGO	1G	Alternative
Ethanol—Sugarcane	S4	Bioethanol	MGO	1G	Alternative
Ethanol—Wheat Straw	S5	Bioethanol	MGO	2G	Alternative

1G = first-generation (food-crop feedstock); 2G = second-generation (lignocellulosic residue).

.

Since all four ethanol pathways combust the same molecular fuel in the engine cylinder, Scenarios S2–S5 are identical in terms of fuel mass consumption, thermal energy delivery, and direct TTW exhaust emissions. Their differentiation arises exclusively in the upstream WTT emission accounting and the Greenhouse Gas Fuel Intensity (GFI) and FuelEU regulatory assessments.

3. Computational Modelling and Methods

The present study develops and applies a simulation-based multi-dimensional evaluation framework for the comprehensive assessment of ethanol as an alternative fuel for large marine dual-fuel (DF) propulsion systems. The framework integrates four interlinked analytical dimensions, dynamic operational simulation, energy analysis, environmental modelling, and regulatory compliance assessment, enabling a holistic, evidence-based comparison between ethanol (across four distinct production pathways), liquefied natural gas (LNG), and conventional very low sulphur fuel oil (VLSFO) under identical operational conditions.

The analytical chain is anchored in an annual high-resolution dataset recorded aboard a container ship. This dataset provides mode-by-mode power demand, voyage duration, and operational context across the vessel’s full round-trip cycle. Rather than relying on steady-state or design-point assumptions, the framework propagates real operational variability through every analytical layer, ensuring that all performance indices, from fuel consumption to regulatory compliance indicators, reflect the stochastic nature of actual ship operations. Each analytical dimension is executed as a discrete computational module within the simulation environment, with outputs cascading sequentially to downstream modules. The overall structure of the framework is summarized in Figure 1.

The framework evaluates six scenarios in parallel: a VLSFO baseline (S0), an LNG benchmark (S1), and four ethanol production-pathway variants, sugar beet (S2), corn (S3), sugarcane (S4), and wheat straw (S5). For each scenario, scenario-specific Specific Gas Consumption (SGC) and Specific Pilot Oil Consumption (SPOC) data files are assigned to the main engine and auxiliary engines, ensuring that the thermodynamic SFC-load factor correlations correctly reflect the combustion behaviour of each fuel configuration. All computed quantities are resolved at the individual operating mode level and subsequently aggregated to annual totals for comparison.

3.1. Dynamic Simulation of the Operational Profile and Power Demand

The first module of the framework reconstructs the dynamic power profile of the vessel across its complete round-trip operational cycle. The ship’s operational data are structured as a sequence of discrete modes, each characterized by a required shaft power, P_req (kW), and a corresponding duration, Δt_i (h). Each mode corresponds to a distinct phase of the voyage, maneuvering, sea passage at varying speeds, and port waiting, and collectively spans the full annual operational profile of the vessel. Prior to simulation, the raw operational data underwent quality control processing to remove anomalous records (defined as power values exceeding 105% MCR or negative values, which accounted for fewer than 0.1% of records), followed by linear interpolation to fill any gaps in the one-hour time series.

For each operating mode i, the engine load factor (LF) is determined by relating the power demand to the maximum continuous rating (MCR) of the installed main engine. Where the power demand falls below a prescribed low-load threshold (set at 25% of total installed capacity), a single engine is assumed to be in service; otherwise, the full complement of installed engines is assumed to be operating. Accordingly, the power per operating engine and the corresponding load factor are evaluated as follows:

$$P_{eng,i}=P_{req,i}/N_{eng,i}$$

(1)

where N_eng,i is the number of engines in service during mode i, determined by:

$$\begin{gathered} N_{eng,i}=1,\hspace{1em}if{P}_{req,i}\le \alpha ·P_{MCR}·N_{total} \\ {N}_{eng,i}=N_{total},\hspace{1em}otherwise \end{gathered}$$

(2)

in which α is the low-load dispatch threshold (α = 0.25 in this study), P_MCR is the MCR of a single engine (kW), and N_total is the total number of installed engines. The engine load factor for mode i is then:

$$L{F}_i=P_{eng,i}/P_{MCR}$$

(3)

where LF_i ∈ [0, 1]. This load factor serves as the independent variable for all subsequent SFC-based performance correlations. The polynomial regression correlations relating SGC, and SPOC to the load factor are fitted to the engine-specific test-bed data using a second-degree least-squares polynomial of the form:

$$f\left(LF\right) = a_2 · {LF}^2 + a_1 · LF + a_0$$

(4)

where a₀, a₁, and a₂ are the regression coefficients determined from the scenario-specific engine data file for each fuel configuration. The goodness-of-fit is assessed using the coefficient of determination R². This polynomial mapping converts the time-varying operational load factor into a continuous approximation of engine performance across the entire power range, from idle to full MCR.

The instantaneous fuel mass flow rate (kg/h) for each engine in mode i is evaluated as:

$$m_{fuel,i}={\displaystyle \frac{SFC\left(L{F}_i\right)·P_{eng,i}·N_{eng,i}}{1000}}$$

(5)

where SFC(LF_i) is the specific fuel consumption evaluated at load factor LF_i from the polynomial correlation (g/kWh). The total fuel consumption per operating mode is obtained by integrating over the mode duration. Annual totals are accumulated by summing over all N_modes operating modes recorded across the dataset:

$$M_{fuel,annual}=\sum _i{m}_{fuel,i}·\Delta t_i$$

(6)

The same integration scheme is applied independently to the main engine primary fuel, the pilot fuel (MGO), and each auxiliary engine. The resulting annual fuel mass and mode-resolved profiles form the primary input to all downstream analytical modules.

3.2. Energy Analysis of the Ethanol-Based Marine Energy System

3.2.1. Dual-Fuel Engine Selection and MCR Sizing

The energy analysis begins with the selection of a suitable DF engine architecture capable of accommodating ethanol as the primary gaseous fuel. Since no commercial ethanol-fuelled low-speed marine engine is currently available from MAN Energy Solutions, the MAN B&W ME-LGIM methanol engine platform was adopted as the reference architecture due to the physicochemical similarity between methanol and ethanol as low-flashpoint, oxygenated liquid fuels. Both fuels share analogous injection and safety system requirements, including low-flashpoint fuel supply systems, inert gas purging, and double-walled piping, making the ME-LGIM architecture the closest available commercial analogue for an ethanol DF engine at the present state of technology.

The required MCR of the selected DF engine is determined from the maximum power demand observed throughout the annual operational profile, augmented by a 10% design margin to ensure that the engine is never operated at its absolute rated output under normal conditions, in accordance with standard marine engineering practice.

This sizing criterion ensures adequate reserve power for adverse weather, shallow water resistance increases, and hull fouling effects encountered across the operational life of the vessel.

3.2.2. Specific Gas Consumption and Pilot Oil Consumption Evaluation

The Specific Gas Consumption (SGC, g/kWh) and Specific Pilot Oil Consumption (SPOC, g/kWh) of the selected MAN B&W ME-LGIM engine were evaluated using the Computer Engine Application System (CEAS) calculation tool provided by MAN Energy Solutions (Everllence). The CEAS tool generates load-dependent performance maps for the methanol variant of the ME-LGIM engine, providing SGC and SPOC values as continuous functions of the engine load factor across the full operating range from 25% to 100% MCR. These tabulated data constitute the scenario-specific engine data file assigned to the LNG and methanol reference cases.

Since no dedicated CEAS data are available for ethanol operation, the ethanol SGC was estimated from the methanol CEAS baseline through lower heating value (LHV) equivalence, under the assumption of comparable brake thermal efficiency (BTE) between the two fuels. This approach is physically justified by the structural similarity of the two alcohols and the well-established methodology for inter-fuel SFC conversion at equivalent energy output. The LHV-equivalence relation yields:

$$SG{C}_{EtOH}=SG{C}_{MeOH}·\left(LH{V}_{MeOH} / LH{V}_{EtOH}\right)$$

(7)

where SGC_EtOH and SGC_MeOH are the specific gas consumptions of ethanol and methanol, respectively (g/kWh), and LHV_MeOH = 19.9 MJ/kg and LHV_EtOH = 27.0 MJ/kg are the respective lower heating values.

The SPOC of the MGO pilot fuel is assumed to remain unchanged between the methanol and ethanol configurations, as the pilot injection system is fuel-agnostic and its quantity is governed by ignition requirements rather than the thermodynamic properties of the primary fuel.

Table 3. Second-degree polynomial regression coefficients for the SGC and SPOC-load factor correlations of each main engine and fuel scenario.

Scenario	Type	a2	a1	a0	R2
S0 (VLSFO)	SFC	0.0031	−0.3709	184.53	0.95
S1 (LNG)	SGC	0.0045	−0.5573	147.95	0.9681
S1 (LNG)	SPOC	0.0007	−0.14	9.4952	0.9897
S2–S5 (Ethanol)	SGC	0.0049	−0.4314	235.24	0.9754
S2–S5 (Ethanol)	SPOC	0.0024	−0.4669	31.709	0.9898

report second-degree polynomial regression coefficients for the SGC and SPOC-load factor correlations of each main engine and fuel scenario. The polynomial takes the form of Equation (4) where LF is the fractional load factor (0–1.00) and f(LF) is the specific fuel mass flow rate (g/kWh). R² values confirm goodness-of-fit across the operating range.

3.2.3. Mode-Level Energy Computation

The thermal energy content delivered by each fuel stream during operating mode i is evaluated by multiplying the mode-integrated fuel mass by the corresponding lower heating value:

$$E_{total,i}=E_{main,i}+E_{pilot,i}=m_{main,i}·LH{V}_{main}+m_{pilot,i}·LH{V}_{pilot}$$

(8)

where E_main,i and E_pilot,i are in MJ. Annual totals are obtained by summation over all operating modes. The energy split between the primary fuel and the MGO pilot provides a direct measure of the gaseous fuel substitution rate and is used as a comparative indicator of dual-fuel utilization efficiency across scenarios.

3.3. Environmental Modelling of Greenhouse Gas Emissions

The environmental module quantifies the full well-to-wake (WtW) greenhouse gas footprint of each scenario. The WtW system boundary adopted in this study encompasses all processes from primary energy source extraction through to exhaust gas discharge at the engine stack, including: (i) upstream fuel production, processing, and transport to the bunkering point (Well-to-Tank, WTT); and (ii) on-board combustion and direct exhaust emission (Tank-to-Wake, TTW). Vessel construction, dry-docking, and end-of-life decommissioning are excluded from the system boundary, consistent with the scope defined under Regulation (EU) 2023/1805 (FuelEU Maritime) [44] and the IMO Net-Zero Framework [45]. The GHG species considered are carbon dioxide (CO₂), methane (CH₄), and nitrous oxide (N₂O), with the latter two weighted by their respective 100-year global warming potentials (GWP₁₀₀) from the IPCC Sixth Assessment Report [46]: GWP(CH₄) = 27 and GWP(N₂O) = 273.

3.3.1. Tank-to-Wake (TTW) Exhaust Emissions

For each operating mode i, the direct CO₂ mass emitted by the primary fuel combustion is calculated from the mode-integrated fuel consumption and the fuel-specific CO₂ emission factor (EF_CO2, g CO₂/g fuel):

$$m_{CO2}=m_i·E{F}_{CO2}$$

(9)

The CO₂ emission factors applied in this study are: VLSFO: 3.151 g CO₂/g fuel; MGO: 3.206 g CO₂/g fuel; LNG: 2.750 g CO₂/g fuel; Ethanol (all pathways): 1.913 g CO₂/g fuel, consistent with the stoichiometric carbon content of each fuel [44,47].

Non-CO₂ GHG species (methane slip and nitrous oxide) are evaluated as power-based emission rates (g/kWh) applied to the total shaft power delivered in each mode. The TTW emission factors adopted for ethanol are: EF_CH4 = 0.001 g/kWh and EF_N2O = 0.003 g/kWh, reflecting the near-zero non-CO₂ GHG characteristics of alcohol combustion in a DF engine.

The total TTW GHG in CO₂-equivalent mass for mode i is:

$$m_{TTW,CO2e,i}=m_{CO2,main,i}+m_{CO2,pilot,i}+m_{CH4,i} · GW{P}_{CH4}+m_{N2O,i} · GW{P}_{N2O}$$

(10)

In accordance with the FuelEU Maritime accounting methodology [44] and RED II [48], the biogenic CO₂ emitted during ethanol combustion is treated differently from fossil CO₂: rather than assigning a positive TTW CO₂ emission factor, the biogenic carbon neutrality is captured through the WTT correction factor (Equation (11)), which applies a carbon correction factor (C_fCO2) to account for the CO₂ that was atmospherically sequestered during feedstock growth. The net effect is that the WTT intensity of biofuels reflects their full lifecycle carbon balance, including both agricultural emissions and the biogenic CO₂ credit. For the TTW calculations (Equations (9) and (10)), the stoichiometric CO₂ emission factor of 1.913 g CO₂/g fuel is applied uniformly to all ethanol scenarios, consistent with the carbon content of ethanol, irrespective of feedstock origin. Regarding LNG, this study considers only fossil-origin LNG, consistent with the current commercial bunker fuel supply chain. The WTT factor for LNG (18.5 gCO₂e/MJ from RED II [48]) reflects extraction, liquefaction, and transport emissions for fossil natural gas. Bio-LNG (biomethane liquefied) was not assessed in this study but is identified as a candidate for future comparative analysis.

3.3.2. Well-to-Tank (WTT) Upstream Emissions

The upstream GHG contribution associated with fuel production, processing, and transport to the vessel is evaluated using energy-based well-to-tank (WTT) emission intensity factors $\left({\gamma }_{WTT},{\mathrm{gCO}}_2\mathrm{e}/\mathrm{MJ}\right)$. In this work, the evaluation of biofuel lifecycle emissions follows the accounting approach adopted by FuelEU Maritime Regulation, where greenhouse gas intensity is assessed on a well-to-wake basis using upstream emission factors derived from Renewable Energy Directive II (RED II) [48]. For bio-based fuels such as ethanol, a carbon correction factor $\left(C_{fC{O}_2}\right)$, corresponding to the stoichiometric CO₂ emission factor of the fuel, is applied to account for the biogenic origin of combustion emissions. Accordingly, the FuelEU-adjusted WTT-equivalent emission intensity for biofuels is expressed as:

$${\gamma }_{WTT,FuelEU,i}={\gamma }_{REDII,i}-{\displaystyle \frac{C_{fCO\_2,i}}{LH{V}_i}}$$

(11)

where ${\gamma }_{WTT,FuelEU,i}$ is the FuelEU-adjusted WTT-equivalent GHG intensity of fuel $i \left({\mathrm{gCO}}_2\mathrm{e}/\mathrm{MJ}\right)$, ${\gamma }_{REDII,i}$ is the RED II default lifecycle emission factor, $C_{fC{O}_2,i}$ is the stoichiometric CO₂ emission factor of the fuel $\left({\mathrm{gCO}}_2/\mathrm{g} \mathrm{fuel}\right)$, and $LH{V}_i$ is the lower heating value of the fuel $\left(\mathrm{MJ}/\mathrm{g} \mathrm{fuel}\right)$. Then the WTT emission amount (m_WTT,i) is evaluated by multiplying the WTT-equivalent GHG intensity of fuel by the amount of fuel.

The default lifecycle emission factors specified in RED II [48] were applied in this study to distinguish between different ethanol production pathways and benchmark fuels. The corresponding WTT and TTW emission factors are presented in

Table 4. Fuel TTW emission factors, and WTT intensity factors for all scenarios.

Fuel	EFCO2 (g CO2/g Fuel)	EFCH4 (g/kWh)	EFN2O (g/kWh)	RED II Factor (gCO2e/MJ)	WTT Factor (gCO2e/MJ)	Source
VLSFO	3.151	0.010	0.030	13.2	—	[44,47]
MGO	3.206	0.010	0.030	14.4	—	[44,47]
LNG	2.750	0.200	0.020	18.5	—	[44,47]
Ethanol—Sugar Beet	1.913	0.001	0.003	25.1	−45.75	[48]
Ethanol—Corn	1.913	0.001	0.003	42.5	−28.35	[48]
Ethanol—Sugarcane	1.913	0.001	0.003	28.1	−42.75	[48]
Ethanol—Wheat Straw	1.913	0.001	0.003	13.7	−57.15	[48]

.

The system boundary adopted for WTT accounting in this study encompasses all processes from primary feedstock cultivation or extraction through to fuel delivery at the vessel’s bunkering point (“field-to-tank”), including agricultural production, feedstock transport, conversion/refining, and distribution logistics. Land-use change (LUC) emissions are included in the RED II default values for first-generation feedstocks (corn and sugarcane) but not for the lignocellulosic residue pathway (wheat straw), where no additional land cultivation is required. The WTT emission factors are drawn from the Renewable Energy Directive II (RED II) Annex V default lifecycle values [48], which reflect the EU regulatory reference dataset and are widely applied in maritime lifecycle assessments [44,47]. It is acknowledged that actual WTT intensities can vary from these defaults depending on regional electricity grid carbon intensity, agricultural practices, and logistics distances; the sensitivity of the results to ±10% variation in WTT factors is assessed in Section 4.5.

3.3.3. Well-to-Wake Totals

The total WtW GHG emission for each operating mode is:

$$m_{WtW,CO2e,i}=m_{TTW,CO2e,i}+m_{WTT,main,i}+m_{WTT,pilot,i}$$

(12)

Annual WtW totals are accumulated by summation over all operating modes. These aggregated values form the basis for both the regulatory compliance assessment and the comparative scenario ranking.

3.4. Regulatory Compliance Assessment

The regulatory module assesses the compliance of each investigated fuel scenario with the two principal decarbonization frameworks currently applicable to international shipping: the regional FuelEU Maritime regulation (EU 2023/1805) [44] and the global IMO Net-Zero Framework (NZF) based on the Greenhouse Gas Fuel Intensity (GFI) methodology [45,49]. Both frameworks operate on an energy-intensity basis (gCO₂e/MJ of fuel energy consumed), enabling direct benchmarking of ethanol against reference fuels.

3.4.1. Greenhouse Gas Fuel Intensity (GFI)

The Greenhouse Gas Fuel Intensity (GFI) is the primary performance metric used by the IMO NZF and forms the cornerstone of the regulatory module. The attained GFI on a well-to-wake basis is defined as the total WtW GHG emission normalized by the total fuel energy consumed by the vessel during the assessment period, as shown in Equation (13) [45].

$$GF{I}_{attained}={\displaystyle \frac{{\sum }_i{m_{WtW}}_{,CO2e,i} }{{\sum }_i{E}_{total,i}}}$$

(13)

The TTW and WTT components of the GFI are evaluated analogously to permit separation of the operational and upstream contributions.

The IMO 2008 reference GFI baseline is 93.3 gCO₂e/MJ, derived from the energy-weighted fleet average emission intensity of the 2008 international shipping fleet as specified in the IMO Net-Zero Framework documentation [45]. This reference value is used to assess the relative decarbonization performance of each scenario.

The forthcoming IMO NZF employs a two-tiered compliance structure to distinguish between a minimum mandatory threshold and an aspirational performance level. The Tier 2 (Base) target defines the legally binding annual GFI reduction floor below which all vessels must operate; non-compliance triggers the GFI levy mechanism. The Tier 1 (Direct) target is set more stringently and represents the ambition level aligned with the 1.5 °C climate pathway; vessels achieving Tier 2 compliance may qualify for reduced levy rates or surplus credits under the IMO pricing mechanism. Emissions that exceed the direct compliance target but remain below the base target are classified as Tier 1. These emissions incur a charge of US$100 per tonne of CO₂-equivalent (CO₂eq). Emissions that surpass the base target fall into Tier 2, where a higher charge of US$380 per tonne of CO₂eq applies.

The annual GFI compliance should be assessed against the two-tiered targets [45] per year t as shown in Equation (14).

$$\begin{array}{c}Base Target: GF{I}_{base,t}=GF{I}_{ref}· \left(1-r_{base,t}\right)\\ Direct Targe: GF{I}_{direct,t}=GF{I}_{ref}· \left(1-r_{direct,t}\right)\end{array}$$

(14)

where GFI_ref = 93.3 gCO₂e/MJ is the 2008 baseline and r_base,t and r_direct,t are the percentage reduction mandates for year t defined by the IMO NZF (the reduction requirements begin in 2028 with initial values of 4% for the Base Target and 17% for the Direct Target, followed by progressively increasing reduction levels to 65% for Base target in year 2040) [45]. The annual GFI deviation from the Base and Direct targets is evaluated; a negative ΔGFI indicates that the vessel’s GFI is below the target (compliance), while a positive value signals non-compliance and triggers the applicable cost mechanism.

3.4.2. FuelEU Maritime

At the regional level, the FuelEU Maritime Regulation (EU 2023/1805), applicable to ships of 5000 GT and above, mandates progressive reductions in the annual average WtW GHG intensity of energy used on board. The regulation adopts a reference GHG intensity of 91.16 gCO₂e/MJ and applies the following reduction trajectory relative to the reference value: −2% from 2025; −6% from 2030; −14.5% from 2035; −31% from 2040; −62% from 2045; −80% from 2050. For a given reporting year, the FuelEU annual GHG intensity target is evaluated as shown in Equation (15) [44,47].

$$GHG{I}_{target}=GHG{I}_{ref}·\left(1-R_t\right)$$

(15)

where R_t is the applicable percentage reduction for year t and GHG_ref = 91.16 gCO₂e/MJ. The attained GHG intensity factor is evaluated as shown in Equation (16) [44,47].

$$GHG{I}_{attained}={\displaystyle \frac{{\sum }_i{m_{WtW}}_{,CO2e,i} }{{\sum }_i{E}_{total,i}·RW{D}_i+{\sum }_k{E}_k}}$$

(16)

The parameter RWD_i refers to the reward multiplier assigned to renewable fuels of non-biological origin, with a value of 2. This multiplier remains applicable until the end of 2033; after this period, the reward factor reverts to 1 [44]. The variable E_k denotes the amount of electrical energy, measured in MJ, delivered to the vessel through the onshore power supply (OPS) connection point k. Beginning in 2030, vessels lacking zero-emission technologies for port operations will be required to include this electricity consumption in the total energy denominator [44]. Examples of zero-emission technologies include fuel cells, onboard energy storage systems, and renewable energy generation technologies such as wind and solar power. According to the FuelEU Maritime regulations, electricity supplied through OPS is considered to produce zero emissions [44].

The FuelEU compliance balance (ΔGHGI) is defined as:

$$\Delta GHGI=\left(GHG{I}_{target}-GHG{I}_{attained}\right)·E_{total,annual}$$

(17)

A positive ΔGHGI corresponds to a surplus (the vessel emits less than the target), which may be banked or sold to other ships under the FuelEU pooling mechanism. A negative ΔGHGI corresponds to a deficit, triggering a FuelEU compliance penalty. In this study, the FuelEU assessment is applied to all operational years within the study horizon to quantify the cumulative compliance trajectory of each ethanol pathway relative to the benchmarks.

4. Results and Discussion

4.1. Dynamic Operational Simulation Results

The dynamic operational simulation module processes the full round-trip power profile of the vessel, encompassing the main propulsion engine and the three auxiliary generator sets, to reconstruct the time-varying power demand, load factor, and energy consumption across every discrete operating mode of the voyage. The complete power profile of all four engines over the full round trip is presented in Figure 2.

The complete round trip spans around 2000 h; the trip is divided into four distinct operating modes: sea passage, port, canal transit, and anchor/waiting. Sea passage dominates the profile, accounting for 1350 h (67.6%) of the total trip duration, confirming that propulsion power demand governs the bulk of the vessel’s energy and emission budget. Port operations constitute 459 h (23.0%), during which the main engine is shut down, and electrical load is served exclusively by the auxiliary generators. Canal transit represents 137 h (6.9%) and is characterized by low main engine power operation, reflecting the speed and maneuvering restrictions imposed by restricted waterway navigation. Anchor and waiting periods contribute a relatively minor 53 h (2.6%) to the total trip duration.

The main engine power profile is characterized by an extended low-to-moderate load regime during sea passage, with abrupt power reductions to near-zero or zero during port calls and short-duration peaks associated with river transit maneuvers. The maximum recorded main engine power is 37,260 kW, corresponding to 51.9% of MCR, a notably low peak utilization that reveals the vessel is operating well below its design capacity throughout the entire voyage. This finding is consistent with slow-steaming practices adopted widely in the container ship sector as a fuel and emission reduction measure. The mean power during operating hours is 14,757 kW, equivalent to an average operational load factor of 20.56% MCR, with a standard deviation of 10.75 percentage points reflecting the variability introduced by different sea conditions, vessel draught changes, and speed orders across the voyage legs.

The load factor distribution of the main engine reveals a strong concentration of operating hours in the low-to-moderate load range. 30.8% of all operating hours fall within the 10–20% MCR band and 28.9% in the 20–30% MCR band, making these two bins collectively responsible for 59.7% of total operating time. A further 24.4% of hours are spent in the 30–40% MCR range, bringing the cumulative fraction below 40% MCR to 99.9% of all operating hours. The engine never exceeds 55% MCR across the entire voyage, and operation above 50% MCR accounts for a negligible 0.1% of operating time.

From an engine design perspective, the dominance of low-load operation confirms the appropriateness of the engine selection criterion applied in Section 3.2.1, where the retrofit MCR was sized at 110% of the maximum observed power demand (37,260 × 1.10 = 41,200 kW) rather than the original engine’s 71,770 kW nameplate rating. Operating the original engine at an average of 20.6% MCR implies that the propulsion plant is significantly oversized relative to the actual voyage power demand, a common characteristic of older ship designs that were specified for higher service speeds subsequently abandoned under slow-steaming regimes. The total mechanical energy delivered by the main engine over the round trip is 80.6 TJ, representing the primary driver of fuel consumption and emissions across all scenarios.

The vessel’s auxiliary power system consists of three diesel generator sets (AE1, AE2, and AE3), each rated at 4500 kW maximum continuous rating (MCR). These generators supply electrical power for ship operations, including propulsion support systems, cargo handling, hotel loads, and deck equipment. The operational simulation demonstrates a balanced dispatch strategy, where the number of active generators varies according to electrical demand during different voyage phases. AE1 recorded the highest operating duration at 1028 h (51.4% of the voyage), followed closely by AE3 with 1004 h (50.3%), while AE2 operated for 879 h (44.0%). Despite similar utilization rates, AE2 functioned mainly as the dedicated port-service generator, operating during 96% of port hours compared with only 14.8% and 21.3% for AE1 and AE3, respectively.

The auxiliary engines consistently operated at relatively low load factors. Average load factors during operation were 29.3% MCR for AE1, 27.2% for AE3, and only 20.1% for AE2. Even under peak conditions, no generator exceeded 48.5% MCR, confirming that the auxiliary plant is substantially oversized relative to actual electrical demand. Combined auxiliary power demand averaged 1679 kW, with a maximum recorded demand of 3139 kW, representing only 23.3% of the total installed auxiliary capacity of 13,500 kW. While this oversizing improves operational reliability, it also results in prolonged low-load operation, which negatively affects fuel efficiency.

Analysis of auxiliary engine dispatch patterns shows that single-engine operation dominates the voyage, accounting for 61.4% of total trip duration. Two-engine operation occurs during 30.6% of the voyage, mainly during periods of increased electrical demand such as cargo handling and ballast operations, while simultaneous operation of all three generators is limited to only 7.7% of voyage time during peak demand events. This operational pattern significantly influences fuel consumption because higher load operation on a single engine generally provides better specific fuel consumption than splitting the same load between multiple engines.

Role differentiation between sailing and port operations is also evident. AE1 and AE3 primarily support sea-passage operations at moderately higher load factors, whereas AE2 mainly operates during low-load port conditions. Overall, the auxiliary engines delivered 12.16 TJ of electrical energy during the round trip, equivalent to approximately 13.1% of the main engine’s total energy output. Therefore, the simulation results characterize the vessel as operating under a predominantly low-load duty cycle, providing important input data for fuel consumption, emissions, and regulatory compliance analyses.

4.2. Energy Analysis Results

Before presenting the energy analysis results, it is important to highlight the key modelling assumption underpinning this study: the lower heating value (LHV)-equivalence approach used to derive the ethanol specific gas consumption (SGC) from the methanol CEAS baseline. This assumption directly influences the predicted fuel consumption and, consequently, affects all subsequent energy, environmental, and regulatory performance results. To evaluate the robustness of the findings, a sensitivity analysis is conducted in Section 4.5, where the two most influential and uncertain parameters, the ethanol SGC and the well-to-tank (WTT) emission factor, are varied by ±10% relative to their baseline values.

The energy analysis quantifies the annual fuel consumption and corresponding energy demand for each fuel scenario, with results disaggregated into primary fuel and pilot fuel contributions. It should be noted that the four ethanol pathways considered (S2–S5) differ only in their upstream production routes and associated lifecycle emission factors. As the combustion characteristics, engine performance assumptions, and fuel consumption parameters remain identical across these pathways, they produce the same fuel consumption and energy demand results. Consequently, differences among the ethanol scenarios emerge only in the environmental and regulatory assessments presented in the subsequent sections. The energy performance results for the three fuel configurations, VLSFO (S0), LNG (S1), and ethanol (S2–S5), are presented in Figure 3.

Figure 3 presents the annual primary and pilot fuel mass consumption for each scenario. The VLSFO baseline (S0) consumes 18,734 t/year of primary fuel with no pilot fuel requirement. LNG (S1) requires 14,115 t/year of primary LNG supplemented by 771 t/year of MGO pilot fuel, for a combined total of 14,886 t/year, a 20.5% reduction in total fuel mass relative to VLSFO. This reduction is a direct consequence of LNG’s substantially higher lower heating value (LHV = 50.0 MJ/kg compared with 41.2 MJ/kg for VLSFO), which lowers the mass of fuel required to deliver an equivalent energy output for any given load.

The ethanol scenarios (S2–S5) display a fundamentally different consumption profile. The annual primary ethanol consumption reaches 24,159 t/year, with an additional 1901 t/year of MGO pilot fuel, yielding a combined total of 26,060 t/year. This represents a 39.1% increase in total fuel mass over the VLSFO baseline and a 75.1% increase over LNG. This is a thermodynamically unavoidable consequence of ethanol’s lower energy density (LHV = 27.0 MJ/kg), approximately 34% lower than VLSFO and 46% lower than LNG. To deliver the same shaft power output across the identical operational profile, the engine combusts a substantially larger mass of ethanol per unit time, as formalized by the LHV equivalence relation applied in the SFC estimation (Section 3.2.2). This gravimetric penalty carries direct practical implications for bunker tank sizing, bunkering frequency, and cargo deadweight utilization, all key considerations for large-scale ethanol adoption in commercial shipping.

The MGO pilot fuel fraction also differs meaningfully between the two DF configurations. In the LNG scenario, the pilot fuel constitutes 5.2% of total fuel mass, whereas in the ethanol scenarios, it accounts for 7.3%. This elevated pilot ratio for ethanol reflects the engine SFC characteristics of the ME-LGIM reference platform when adapted to ethanol operation: because ethanol’s LHV is lower than that of methanol (the basis for the CEAS-derived SPOC values), the absolute pilot oil mass per unit energy delivered is comparatively higher. This increased MGO pilot dependency partially offsets the GHG advantages of ethanol as a low-carbon primary fuel, a trade-off captured in detail in the environmental analysis.

The pilot fuel ratio is not constant across the load range but varies as a function of engine load factor. At low loads (LF < 25% MCR), the pilot fuel requirement per unit of primary fuel is proportionally higher because the combustion chamber temperature and pressure are insufficient to reliably auto-ignite low-flashpoint liquid fuels without a larger ignition energy contribution from the pilot spray. At higher loads (LF > 40% MCR), the elevated in-cylinder conditions reduce the minimum pilot quantity needed for stable ignition, allowing the SPOC to decrease as a fraction of total fuel consumption. In the present study, the SPOC is evaluated from the CEAS load-dependent performance map and therefore already captures this load-dependent variation through the polynomial correlation of Equation (4).

Figure 4 presents the annual thermal energy consumption in TJ/year, eliminating the mass-based distortion introduced by differing fuel densities. The VLSFO baseline (S0) records the highest total energy input of 776.5 TJ/year, entirely from primary fuel. LNG (S1) delivers the lowest total energy consumption of 738.7 TJ/year (705.7 TJ primary LNG + 32.9 TJ MGO pilot), representing a 4.9% reduction relative to VLSFO. This modest but meaningful efficiency advantage reflects the improved brake thermal efficiency associated with the lean-premixed combustion characteristics of natural gas in a low-speed two-stroke DF engine, reducing heat rejection and exhaust enthalpy losses relative to conventional diesel-cycle operation.

The ethanol scenarios (S2–S5) deliver a total annual energy input of 733.5 TJ/year (652.3 TJ primary ethanol + 81.2 TJ MGO pilot). Notably, this total is 5.5% lower than the VLSFO baseline and 0.7% lower than the LNG scenario, a result that initially appears counterintuitive given the large fuel mass penalty identified in Figure 3. The explanation lies in the LHV equivalence methodology applied in the SFC estimation: the polynomial SFC correlations for ethanol are derived by scaling the methanol CEAS data by LHV ratio (LHV_MeOH/LHV_EtOH = 19.9/27.0 = 0.737), which reduces the energy delivered per unit SFC relative to methanol. The net result is that despite consuming 39.1% more fuel by mass, the ethanol scenarios deliver slightly less total thermal energy than VLSFO, indicating that the effective specific energy conversion in ethanol DF operates at a marginally lower thermodynamic efficiency than VLSFO baseline under equivalent power demand conditions.

The MGO pilot energy fraction shows a more pronounced divergence in energy terms than in mass terms. In the LNG scenario, pilot fuel contributes 32.9 TJ, or 4.5% of total energy. In the ethanol scenarios, the pilot contributes 81.2 TJ, or 11.1% of total energy. This more than doubling of the pilot energy share reflects both the higher SPOC of the ethanol configuration and the proportional amplification caused by ethanol’s lower LHV. This elevated pilot energy fraction is consequential from an environmental standpoint: each megajoule of MGO pilot carries a WtT upstream emission factor of 17.7 gCO₂e/MJ and a TTW CO₂ factor of 3.206 g CO₂/g fuel, meaning that the pilot fuel becomes a non-negligible contributor to the overall GHG footprint of the ethanol system.

Table 5. Summary of annual fuel and energy consumption results by scenario.

Scenario	Primary Fuel (t/yr)	Pilot MGO (t/yr)	Total Fuel (t/yr)	Total Energy (TJ/yr)	Pilot Energy Fraction (%)
S0—VLSFO	18,734	0	18,734	776.5	0.0%
S1—LNG	14,115	771	14,886	738.7	4.5%
S2–S5—Ethanol	24,159	1901	26,060	733.5	11.1%

consolidates the key energy results for all scenarios.

4.3. Environmental Analysis Results

The environmental module evaluates the full WtW GHG footprint of each scenario by combining the TTW direct combustion emissions with the WTT upstream emissions from fuel production and supply. Unlike the energy analysis, the environmental assessment reveals significant differentiation among the four ethanol feedstock scenarios. The results are presented in Figure 5.

The TTW emissions represent the total in-cycle GHG output from combustion of primary and pilot fuels, accounting for CO₂, CH₄ slip, and N₂O weighted by 100-year GWPs. As shown in Figure 5, the VLSFO baseline (S0) records the highest TTW total of 60,083 t CO₂e/year, arising entirely from primary fuel combustion with no pilot contribution. The LNG scenario (S1) achieves a TTW total of 42,421 t CO₂e/year, a 29.4% reduction relative to the baseline. This reduction stems primarily from methane’s lower carbon-to-hydrogen ratio relative to the long-chain hydrocarbons in VLSFO, though the methane slip term imposes a GHG penalty on the LNG configuration that partially erodes its combustion carbon advantage.

All four ethanol scenarios (S2–S5) yield identical TTW emissions of 52,400 t CO₂e/year, comprising 46,305 t CO₂e/year from primary ethanol combustion and 6096 t CO₂e/year from the MGO pilot fuel. This represents a 12.8% reduction relative to VLSFO, but still modest compared with the LNG TTW advantage of 29.4%. Two competing mechanisms determine this outcome. On one hand, ethanol’s substantially lower carbon-to-energy ratio (EF_CO2 = 1.913 g CO₂/g fuel, versus 3.151 for VLSFO) confers a clear TTW benefit per unit mass burned. On the other hand, the 29.1% higher primary fuel mass consumed (24,159 vs. 18,734 t/yr) partially erodes this per-unit advantage when expressed as an annual total. Furthermore, the elevated MGO pilot burden of 6096 t CO₂e/year, absent in the VLSFO scenario, contributes an additional TTW penalty with no offset. The net result is that ethanol’s operational GHG performance remains substantially inferior to LNG at the TTW level, a finding that fundamentally motivates the lifecycle WtW perspective.

The WTT upstream contributions reveal the most decisive differentiator across all scenarios investigated. As shown in Figure 5, VLSFO (S0) and LNG (S1) carry positive WTT burdens of 10,408 t CO₂e/year and 13,531 t CO₂e/year, respectively, reflecting the GHG intensity of fossil fuel extraction, refining, and liquefaction. The LNG upstream burden marginally exceeds that of VLSFO in absolute terms, primarily because the total energy consumed by the LNG system (738.7 TJ/year) is slightly lower than for VLSFO (776.5 TJ/year), but the WTT intensity factor applied to LNG’s supply chain incorporates contributions from natural gas field-to-liquefaction energy losses that offset part of the lower energy throughput.

The four ethanol scenarios exhibit strongly negative WTT contributions, indicating net upstream carbon credits arising from bio-based production pathways that displace fossil carbon accounting. The magnitude varies substantially by feedstock:

Corn ethanol (S3) carries the least favourable upstream profile among the ethanol variants, with a WTT contribution of −17,324 t CO₂e/year. Despite being negative, the corn RED II default lifecycle emission factor of 42.5 gCO₂e/MJ, the highest among the four pathways, reflects the significant agricultural energy inputs, fertilizer-derived N₂O, and land-use implications of corn-based fermentation, which limit the extent of the upstream credit.

Sugarcane (S4) and sugar beet (S2) deliver intermediate upstream credits of −26,717 t CO₂e/year and −28,674 t CO₂e/year, respectively. Sugar beet’s slightly larger credit compared with sugarcane reflects its marginally lower RED II factor, which stems from the more favourable energy balance of beet-to-ethanol processing chains in Northern European production contexts.

Wheat straw ethanol (S5) achieves the most favourable upstream profile of all scenarios, with a WTT credit of −36,110 t CO₂e/year driven by its RED II factor of 13.7 gCO₂e/MJ and WTT factor of −57.15 gCO₂e/MJ. As a second-generation lignocellulosic residue, wheat straw utilizes a material that would otherwise decompose or be field-burned, incurring minimal additional land-use pressure and substantially lower net process emissions relative to first-generation crops.

The WtW total, which integrates TTW direct and WTT upstream contributions, provides the most complete policy-relevant measure of lifecycle GHG performance and is illustrated by the relative reduction percentage in Figure 6. The VLSFO baseline (S0) is the highest WtW emitter at 70,491 t CO₂e/year. LNG (S1) achieves 55,951 t CO₂e/year, a 20.6% reduction over VLSFO. Notably, this is smaller than LNG’s 29.4% TTW advantage, because the LNG upstream burden (13,531 t CO₂e/year) marginally exceeds that of VLSFO (10,408 t CO₂e/year), penalizing the WtW balance relative to the TTW comparison.

All four ethanol pathways substantially outperform both benchmarks at the WtW level. Corn ethanol (S3) achieves the most modest reduction of 50.2% (35,077 t CO₂e/year), yet this is still more than double LNG’s lifecycle saving relative to VLSFO. Sugarcane (S4) and sugar beet (S2) deliver reductions of 63.6% and 66.3%, respectively. Wheat straw ethanol (S5) records the lowest WtW total of all scenarios at 16,290 t CO₂e/year, representing a 76.9% reduction relative to the VLSFO baseline, the most substantial decarbonization outcome in this study, and more than three times the proportional reduction achieved by LNG.

From a practical deployment perspective, a key engineering constraint associated with ethanol adoption in marine fuel systems is material compatibility. Ethanol’s hydroxyl group (–OH) renders it mildly corrosive toward certain elastomers, aluminum alloys, and zinc-coated components commonly found in conventional fuel storage, piping, and injection systems designed for VLSFO or MGO. Specifically, nitrile rubber (NBR) seals widely used in legacy fuel systems, exhibit swelling and degradation in contact with ethanol blends above approximately 20% vol [30,36]. For the ethanol DF retrofit scenario considered in this study, the low-flashpoint fuel supply system of the ME-LGIM reference platform is already specified with fluorinated elastomers (FKM) and stainless-steel pipework, which are fully compatible with ethanol. However, the bunkering infrastructure (shore-side tanks, transfer hoses, and manifolds) would require targeted material verification and, in many cases, replacement of legacy components prior to ethanol service entry. This compatibility requirement represents an additional upfront capital cost recommended for quantitative assessment in future techno-economic studies.

4.4. Regulatory Compliance Results

The regulatory compliance module evaluates the performance of each scenario against the two principal decarbonization frameworks currently applicable to international shipping: the global IMO Net-Zero Framework (NZF) based on the Greenhouse Gas Fuel Intensity (GFI) methodology, and the regional FuelEU Maritime Regulation (EU 2023/1805), which mandates progressive reductions in the annual average WtW GHG intensity of ship energy use. Both frameworks operate on an energy-intensity basis (gCO₂e/MJ) and adopt a common reference GHG intensity derived from the 2008 fleet average. The GFI results are presented in Figure 7.

Figure 7 presents the well-to-wake GFI for each scenario plotted against the evolving IMO NZF Tier 1 (Direct target) and Tier 2 (Base target) annual reduction trajectories relative to the reference value of 93.3 gCO₂e/MJ. The VLSFO baseline (S0) records a GFI of 90.78 gCO₂e/MJ in the pre-2030 period, composed of a TTW component of 77.38 gCO₂e/MJ and a WTT upstream contribution of 13.4 gCO₂e/MJ. This places the VLSFO scenario marginally below the IMO reference value of 93.3 gCO₂e/MJ and above the Tier 1 target applicable from 2028 (77.44 gCO₂e/MJ), confirming that VLSFO operation is already in a non-compliant zone under the IMO NZF from the framework’s entry into force. After 2030, accounting for changes in port electricity emissions contributions, the VLSFO GFI adjusts marginally to 90.55 gCO₂e/MJ, remaining structurally non-compliant throughout the entire 2028–2050 horizon as the Tier 1 target progressively tightens from 77.44 (2028) to 73.71 (2030) to 53.18 (2035) to 18.70 (2040) and ultimately to 0 gCO₂e/MJ (2050). The divergence between the VLSFO GFI and the compliance corridor, therefore, widens monotonically over time, rendering continued VLSFO operation increasingly untenable from a regulatory standpoint beyond 2028.

The LNG scenario (S1) achieves a pre-2030 GFI of 75.74 gCO₂e/MJ (TTW: 57.43; WTT: 18.32 gCO₂e/MJ) and an adjusted post-2030 value of 75.63 gCO₂e/MJ. This positions LNG below the Tier 1 target from 2028 through 2030, confirming Tier 1 compliance throughout the early NZF period. However, as the Tier 1 trajectory steepens beyond 2030, LNG crosses into non-compliance progressively: the Tier 1 target falls to 69.60 gCO₂e/MJ in 2031, 65.50 in 2032, 61.39 in 2033, and 57.29 in 2034, values that converge on and then fall below the LNG GFI of 75.78 gCO₂e/MJ. LNG therefore faces Tier 1 non-compliance from approximately 2031 onwards and remains well above the Tier 2 targets throughout the entire assessment horizon. This finding is highly significant for fleet investment planning: vessels retrofitted for LNG operation today will face regulatory non-compliance within the first decade of operation under the IMO NZF, underscoring the transitional rather than long-term nature of LNG as a decarbonization strategy.

The four ethanol scenarios (S2–S5) present a markedly different compliance picture, as illustrated by the wide spread of GFI values visible in Figure 7. The pre-2030 GFI values are: sugar beet 32.35 gCO₂e/MJ, corn 47.82 gCO₂e/MJ, sugarcane 35.02 gCO₂e/MJ, and wheat straw 22.21 gCO₂e/MJ. All four values lie substantially below the Tier 1 trajectory of 2028. More significantly, two of the four pathways, sugar beet, and sugarcane also fall below the Tier 1 target from 2028 until 2038, placing them in the over-compliant zone that enables the accrual of GFI compliance surplus credits. Corn ethanol (S3), with the highest GFI among the four, falls within the Tier 1–Tier 2 corridor beyond 2036.

After 2030, accounting for the port electricity emission factor contribution (67 gCO₂eq/MJ applied to the 6.93 TJ port electricity demand), all ethanol GFI values undergo a marginal upward adjustment but remain structurally compliant: sugar beet 32.58, corn 47.93, sugarcane 35.23, wheat straw 22.52 gCO₂e/MJ. Wheat straw ethanol (S5) remains the strongest performer throughout the timeline, achieving a GFI 70.9% below the 2028 Tier 1 target of 77.44 gCO₂e/MJ and already compliant with the 2039 Tier 1 target of 25.60 gCO₂e/MJ. The WTT component of the wheat straw GFI is −49.23 gCO₂e/MJ, indicating that the upstream carbon credit from lignocellulosic production more than offsets the positive TTW component (71.44 gCO₂e/MJ), yielding a net WtW GFI substantially below even the long-term targets. This result demonstrates that, under a WtW accounting framework, second-generation lignocellulosic ethanol already satisfies the most stringent mid-century IMO compliance thresholds based on present-day production technology.

Figure 8 presents the annual GHG Energy Intensity (GHGI) of each scenario plotted against the FuelEU Maritime compliance trajectory, which mandates progressive intensity reductions relative to the 2020 reference value of 91.16 gCO₂e/MJ. The applicable annual FuelEU targets are: 89.34 gCO₂e/MJ from 2025 (−2%); 85.69 from 2030 (−6%); 77.94 from 2035 (−14.5%); 62.90 from 2040 (−31%); 34.64 from 2045 (−62%); and 18.23 from 2050 (−80%).

The VLSFO baseline (S0) records a GHGI of 90.78 gCO₂e/MJ in the pre-2030 period, exceeding the FuelEU target from its first compliance year in 2026 (89.34 gCO₂e/MJ). This non-compliance becomes progressively more severe as the target tightens: the VLSFO GHGI exceeds the 2030 target (85.69) by 4.86 gCO₂e/MJ, the 2035 target (77.94) by 12.61 gCO₂e/MJ, and the 2040 target (62.90) by 27.65 gCO₂e/MJ. This confirms that conventional fossil-fuel operations are entirely incompatible with FuelEU Maritime compliance across the full study horizon.

The LNG scenario (S1) achieves a GHGI of 75.74 gCO₂e/MJ (pre-2030) and 75.63 gCO₂e/MJ (post-2030), placing it below the FuelEU compliance threshold from 2026 through 2039. LNG satisfies the −2% target (89.34 gCO₂e/MJ, applicable 2025–2029) and the −6% target (85.69, applicable 2030–2034) with ample margin. However, as the FuelEU target drops to 62.9 gCO₂e/MJ from 2040, the LNG post-2030 GHGI of 75.63 gCO₂e/MJ does not satisfy the 2040–2044 target, with the deficit growing from 12.73 gCO₂e/MJ in 2040 to 57.40 gCO₂e/MJ in 2050. This result reinforces the conclusion from the IMO GFI analysis: LNG provides meaningful regulatory relief only in the short-to-medium term (2025–2039 under FuelEU) and cannot support long-term compliance trajectories without supplementary measures such as carbon capture or blending with bio-LNG.

The four ethanol scenarios (S2–S5) demonstrate a categorically different compliance profile. All four pathways record GHGI values substantially below the FuelEU compliance limit at every point in the regulatory timeline from 2026 to 2045, confirming full FuelEU Maritime compliance throughout the entire assessment horizon without exception.

As shown in Figure 8, sugar beet (S2) achieves a pre-2030 GHGI of 32.35 gCO₂e/MJ, falling 63.8% below the 2026 FuelEU target of 89.34 gCO₂e/MJ. Post-2030, its GHGI adjusts marginally to 32.58 gCO₂e/MJ, remaining below even the most stringent 2045 FuelEU threshold of 34.64 gCO₂e/MJ by 2.06 gCO₂e/MJ. Similarly, sugarcane (S4) records a GHGI of 35.02 gCO₂e/MJ (pre-2030) and 34.23 gCO₂e/MJ (post-2030), achieving compliance margins against the 2045 target, respectively.

Corn ethanol (S3), the least favourable ethanol variant, records a GHGI of 47.82 gCO₂e/MJ (pre-2030) and 47.93 gCO₂e/MJ (post-2030). This remains below the 2025–2044 FuelEU targets with a comfortable margin. However, the 2045 FuelEU target (34.64 gCO₂e/MJ) is tighter than the corn ethanol post-2030 GHGI, indicating that corn-based ethanol would enter non-compliance from 2045 onwards under FuelEU without improvement in upstream emission performance. This represents a critical planning horizon: shipowners adopting corn ethanol as their primary fuel need to account for the potential need to transition to lower-carbon feedstocks or blend with 2G/3G ethanol before 2045 to maintain FuelEU compliance.

Wheat straw ethanol (S5) achieves the strongest FuelEU performance, with a pre-2030 GHGI of 22.21 gCO₂e/MJ and a post-2030 value of 22.52 gCO₂e/MJ. This GHGI is more than the most stringent FuelEU 2050 target (18.23 gCO₂e/MJ) by only 4.29 gCO₂e/MJ. Therefore, the wheat straw scenario represents the only configuration in this study that satisfies both the IMO NZF and FuelEU near 2050 targets without requiring additional technological interventions such as carbon capture or blending with electrofuels. Achieving this performance level in practice, however, depends on maintaining the upstream emission intensity assumed for lignocellulosic ethanol production, which in turn requires rigorous supply chain management to prevent scope 1 and scope 2 emissions from escalating with production scale.

Taken together, the GFI and GHGI analyses converge on three overarching regulatory conclusions. First, VLSFO is incompatible with both frameworks from their respective compliance start dates and becomes increasingly penalized as the intensity targets tighten. Second, LNG provides short-to-medium term compliance relief under both IMO NZF (Tier 1 compliance to ~2030) and FuelEU (compliance to ~2034–2039) but faces certain non-compliance beyond these horizons, rendering it a transitional rather than long-term solution. Third, all ethanol pathways except corn deliver structural long-term compliance under both frameworks across the full 2026–2050 horizon, with wheat straw ethanol demonstrating sufficient compliance margin to satisfy even the most ambitious 2050 targets. Corn ethanol satisfies all targets through 2044 but faces a compliance gap from 2045 under FuelEU, identifying the specific policy-relevant threshold at which feedstock upgrading becomes mandatory. These findings provide a quantitative evidence base for fuel transition decision-making that goes beyond the binary “compliant/non-compliant” framing typically applied in regulatory assessments.

4.5. Sensitivity Analysis Results

To assess the robustness of the simulation results against key modelling uncertainties, a parametric sensitivity analysis was conducted by independently varying the two most uncertain input parameters, the ethanol Specific Gas Consumption (SGC), and the WTT upstream emission factor, by ±10% relative to their baseline values. The influence of each perturbation on the three principal KPIs (annual fuel consumption, annual WtW GHG emissions, and GFI) was evaluated for the wheat straw ethanol scenario (S5, the strongest performer) and the LNG benchmark (S1), as these represent the bounding cases of the fuel performance spectrum. The results are presented in

Table 6. Sensitivity analysis results: effect of ±10% variation in key uncertain parameters on principal KPIs.

Parameter Varied	Variation	S5 Fuel Con. (t/yr)	S5 WtW GHG (t CO2e/yr)	S5 GFI (gCO2e/MJ)	S1 WtW GHG (t CO2e/yr)	S1 GFI (gCO2e/MJ)
Baseline	—	26,060	16,290	22.21	55,951	75.74
SGC (ethanol)	+10%	28,666	17,911	22.20	55,951	75.74
SGC (ethanol)	−10%	23,454	14,670	22.22	55,951	75.74
WTT factor	+10%	26,060	20,132	27.45	57,304	77.58
WTT factor	−10%	26,060	12,442	16.96	54,598	73.91

.

The results confirm that the WTT emission factor is the most influential uncertain parameter, driving a ±23.6% variation in WtW GHG and GFI for the ethanol scenario. However, the SGC uncertainty, which propagates a ±10% change directly into fuel consumption, represents the second most important source of variability, as it drives a ±9.9% variation in WtW GHG for the ethanol scenario and a near-zero effect on GFI because of the compensation of energy consumption in the denominator. Compared to the LNG benchmark case, the WTT emission factor drives a variation of ±2.4% only in WtW GHG. Therefore, under the least favourable condition (WTT +10%), the wheat straw ethanol GFI reaches 27.45 gCO₂e/MJ, still 64.6% below the IMO NZF 2028 Tier 1 threshold of 77.44 gCO₂e/MJ and fully compliant until 2038 under IMO NZF while it compliant until 2049 under FuelEU GHGI. These results confirm that the principal conclusions of this study, particularly the lifecycle superiority of ethanol over LNG, are robust to the parameter uncertainties inherent in the WTT factor and LHV-based SFC estimation methodology.

4.6. Ethanol Deployment Considerations

While a full techno-economic assessment of ethanol bunkering infrastructure is beyond the scope of this study, a qualitative comparative overview of the key infrastructure and cost dimensions of ethanol relative to methanol and LNG, the two most commercially mature alternative marine fuels, is provided in

Table 7. Qualitative comparison of ethanol, methanol, and LNG marine fuel infrastructure and economic dimensions.

Dimension	LNG	Methanol	Ethanol (2G)
Bunkering infrastructure maturity	High	Low–Medium	Very Low
Storage temperature requirement	Cryogenic (−162 °C)	Ambient	Ambient
Energy density (MJ/kg)	50.0	19.9	27.0
Relative fuel tank volume (vs. VLSFO)	×1.8	×3.7	×2.7
Material compatibility concern	High (cryogenic materials)	Moderate (elastomers)	Moderate (elastomers)
Estimated fuel cost (2024, $/tonne)	550–700	400–600	600–900 (1G)l; 900–1500 (2G)
WtW GHG vs. VLSFO (this study)	−20.6%	−	−50.2% to −76.9%
IMO NZF 2040 compliance	✗	✓ (blue)	✓ (wheat straw pathway)
FuelEU 2050 compliance	✗	Marginal	✓ (all pathways except corn from 2045)
Production scalabilitym	High	High	Low–Medium (2G)

Sources: infrastructure maturity and cost estimates from [29,30,36,37]; volume ratios calculated from LHV and density data; regulatory compliance from this study.

, drawing on published lifecycle cost and infrastructure assessments.

It is explicitly noted that blended-fuel operation, wherein ethanol is co-injected with MGO or other distillate fuels at variable volumetric ratios, falls outside the scope of the present study. The current investigation is confined to neat ethanol as the primary dual-fuel gas supply in a dedicated low-flashpoint fuel system, consistent with the MAN B&W ME-LGIM reference architecture. This scope boundary was adopted deliberately to provide a clean baseline assessment of ethanol’s maximum decarbonization potential before the complicating effects of blend ratio variability, partial substitution rates, and combustion interaction phenomena are introduced. Existing literature on alcohol–diesel blending in marine and stationary diesel engines [22] indicates that ethanol–MGO blends below 20% vol are broadly compatible with standard medium-speed injection systems with modest material modifications and deliver proportional but sub-linear reductions in CO₂ emissions relative to the neat diesel baseline. For large two-stroke low-speed engines of the type considered in this work, however, dedicated blended-operation data remain scarce, and the combustion stability, ignition delay, and emission profiles of ethanol–MGO co-injection under slow-steaming low-load conditions (which dominate this vessel’s duty cycle, as shown in Section 4.1) are yet to be characterized experimentally. A dedicated simulation and experimental study of blended ethanol–MGO operation across the representative load factor range of 10–40% MCR is therefore recommended as a near-term research priority, given its practical relevance as a lower-barrier entry pathway to ethanol adoption that avoids the full capital commitment of a dedicated low-flashpoint fuel supply system retrofit.

5. Conclusions

5.1. Contribution of This Work

This study presents one of the few full-scale, simulation-driven assessments of ethanol as a direct primary fuel for large marine dual-fuel propulsion systems, addressing a significant gap in the existing maritime decarbonization literature where ethanol has been largely overlooked in favour of LNG, methanol, and ammonia. The principal novelty lies in the integration of analytical dimensions, energy, environmental, and regulatory within a single coherent framework calibrated against real operational data from an active container ship. Unlike previous studies that have examined isolated performance parameters, this work delivers a comparative multi-pathway evaluation that simultaneously captures the combustion-level behaviour and the full lifecycle GHG consequences of four distinct ethanol production routes. A further original contribution is the explicit identification of a regulatory paradox: while ethanol demonstrates a substantially superior WtW lifecycle carbon profile relative to LNG across all feedstock pathways, its higher fuel mass consumption under current IMO operational accounting frameworks creates compliance challenges under energy-intensity indicators that do not yet fully reward upstream carbon credits. The principal findings of this study can be summarized as follows:

• The fuel mass penalty of ethanol (+39.1% vs. VLSFO) is mass-based rather than energy-based: total thermal energy input for ethanol scenarios is 5.5% lower than VLSFO, because the LHV-equivalence SFC estimation framework correctly accounts for the lower energy density. This distinction is critical for tank-sizing and bunkering frequency assessments.
• All four ethanol production pathways deliver superior well-to-wake GHG reductions compared with LNG. Even the least favourable pathway (corn) delivers a lifecycle GHG reduction more than double that of LNG relative to VLSFO, and the best-performing pathway (wheat straw) achieves a reduction more than three times larger.
• The lifecycle inversion effect is the defining mechanism. At the operational (TTW) level, ethanol’s direct combustion advantage over VLSFO is modest and falls well short of LNG’s in-cycle benefit, due to elevated fuel mass and MGO pilot consumption. However, the large negative upstream (WTT) carbon credits from bio-based production pathways fully reverse this disadvantage at the WtW level, a reversal that fundamentally challenges the prevailing industry assumption that LNG is the superior transitional fuel.
• Wheat straw ethanol satisfies 2040 IMO NZF targets with current technology. With a WtW GFI of 22.52 gCO₂e/MJ, wheat straw ethanol is the only scenario in this study that meets the IMO NZF Tier 1 target applicable in 2040 (18.70 gCO₂e/MJ) with a limited compliance margin, confirming that second-generation lignocellulosic ethanol already satisfies mid-century regulatory intensity thresholds under a well-to-wake accounting framework.
• LNG compliance is time-limited under both regulatory frameworks. LNG achieves IMO NZF Tier 1 compliance only through approximately 2028–2029 and FuelEU compliance through 2026–2039, beyond which its GFI of 75.63 gCO₂e/MJ exceeds the tightening trajectory targets. LNG cannot satisfy the 2030 IMO NZF target or the FuelEU 2040 thresholds, confirming its role as a transitional rather than a long-term decarbonization solution.
• Corn ethanol faces a post-2044 FuelEU compliance gap. Despite satisfying all FuelEU targets through 2044, corn ethanol’s GHGI of 47.93 gCO₂e/MJ exceeds the 2045 FuelEU target of 34.64 gCO₂e/MJ, identifying the specific regulatory threshold at which feedstock upgrading from first-generation to second-generation production becomes mandatory to maintain compliance.

The findings collectively demonstrate that bio-based ethanol is a technically feasible, environmentally superior, and regulatorily compliant alternative marine fuel for large dual-fuel propulsion systems, with lifecycle GHG performance that substantially surpasses LNG across all four production pathways investigated. The conventional maritime industry narrative positioning LNG as the optimal transitional pathway to 2030 and beyond is directly challenged by this evidence. The principal barriers to ethanol adoption are not environmental but structural: elevated fuel mass requirements, immature bunkering infrastructure, and production cost uncertainties that currently render advanced bioethanol economically uncompetitive relative to fossil fuels. Critically, the existing IMO operational accounting framework, which evaluates GHG intensity without fully crediting upstream biogenic carbon sequestration, creates a structural policy misalignment that must be corrected through the transition to full lifecycle accounting if renewable alcohols are to compete on equitable terms. This study provides a quantitative evidence base for that policy dialogue.

5.2. Recommendations for Future Work

The following directions are identified as priorities for future research emerging from the limitations and findings of this study:

• Full-scale experimental validation on ethanol-fuelled two-stroke engines. The engine SFC data in this study are derived from the MAN B&W ME-LGIM methanol platform via LHV equivalence. Dedicated test-bed measurements on ethanol-adapted two-stroke DF engines, particularly covering methane slip rates, ignition delay, and combustion stability at low loads, are essential to validate and refine the polynomial SFC correlations applied here.
• Future work should quantify the capital and operating cost implications of establishing dedicated ethanol bunkering networks at key maritime hubs, including cold-chain requirements, material compatibility retrofits, and storage safety systems, to provide a complete total cost of ownership assessment.
• Blended fuel operation, ethanol-MGO co-injection at variable ratios, represents a near-term, lower-barrier deployment pathway that avoids the full capital commitment of a dedicated low-flashpoint fuel supply system. Future work should develop a dedicated simulation and experimental study covering combustion stability, ignition delay, and emission profiles of ethanol–MGO blends across the 10–40% MCR load range representative of this vessel’s duty cycle, where blending behaviour under slow-steaming conditions remains largely uncharacterized for large two-stroke low-speed engines.
• The economic feasibility of ethanol adoption is sensitive to the precise carbon charge rates that will be adopted under the IMO NZF levy mechanism, which remain subject to ongoing negotiation. Future work should perform scenario analysis across a range of levy structures to identify the carbon price thresholds at which each ethanol pathway becomes economically competitive with LNG and VLSFO.