J. Ocean Eng. Technol. Search

CLOSE


J. Ocean Eng. Technol. > Volume 40(3); 2026 > Article
Park and Kim: Mobility-Oriented Multiphysics, Insulation and Evidence Frameworks for Liquid Hydrogen Storage Tanks

Abstract

Liquid hydrogen (LH2) storage tanks are essential for long-range hydrogen transport and mobility, but their performance is controlled by coupled heat ingress, phase change, stratification, ullage pressure evolution, and motion-induced interface renewal. This review synthesizes the thermophysical fundamentals, tank architectures, insulation systems, quiescent storage behavior, operational transients, sloshing-coupled cryogenic dynamics, diagnostics, modeling hierarchies, and safety and approval requirements, with the fill ratio treated as a central design variable. The literature is organized through an evidence taxonomy spanning direct operation, controlled LH2 experiments, transitional evidence, and simulation-only studies. Three synthesis findings emerged. First, the fill ratio and motion jointly determine ullage stratification, wetted-wall distribution, and interfacial renewal so that intermediate-fill operation can trigger either boil-off amplification or short-duration pressure collapse. Second, mobility-relevant design cannot rely solely on quiescent boil-off or sloshing-load analysis. Reliable predictions require coupled treatment of thermodynamics, motion, controls, and validation. Third, the strongest evidence comes from benchmark self-pressurization datasets, trailer-route operation, and pilot-scale marine transport, whereas validation-grade motion-enabled datasets remain scarce for large maritime tanks and lightweight aviation concepts. Therefore, approval-oriented LH2 tank design should be evidence-graded, motion-aware, and explicitly linked to uncertainty-bounded verification and validation.

1. Introduction

1.1 Liquid Hydrogen Storage Tanks as a Multiphysics System

Liquid hydrogen (LH2) is increasingly regarded as a strategic storage medium for hydrogen systems that require a high gravimetric and volumetric energy density. Compared to compressed gaseous hydrogen, LH2 provides a much higher volumetric storage density and remains relevant to stationary storage, heavy-duty road transport, marine transport, aviation, and space applications (Kang et al., 2022; Naquash et al., 2024; Simanullang, 2025; Tiwari et al., 2024).
From an engineering perspective, however, an LH2 tank is not a passive cryogenic container. Its behavior is governed by a closed loop among external heat ingress, liquid–vapor phase change, thermal stratification, ullage pressure evolution, fill level, and operating response. Classical self-pressurization (SP) studies and later model reviews have shown that pressure rise cannot be explained by total heat leak alone because internal circulation, vapor compressibility, and interfacial transport redistribute the absorbed energy in a highly coupled manner (Hasan et al., 1991; Van Dresar et al., 1992; Barsi & Kassemi, 2008; Wang et al., 2022; Wang et al., 2024b).
That coupled viewpoint also changes how insulation should be interpreted. In practical LH2 storage, the thermal response depends on the insulation material, vacuum level, layer compression, vapor-cooled shielding, structural bridges, geometry, fill ratio, and operating history. Recent insulation reviews, coupled tank models, and integrated architecture studies support the view that insulation engineering and multiphysics tank behavior must be discussed together (Fesmire, 2015; Yin et al., 2024; Kang et al., 2024; Qu et al., 2025).

1.2 Application Landscape Across Stationary and Mobile Sectors

The governing questions of LH2 storage recur across sectors, but their relative importance shifts with application. Stationary systems emphasize long dormancy, predictable pressure rise, and low maintenance heat ingress; mobile systems also confront repeated filling and withdrawal, vibration, acceleration, and route-or voyage-dependent operating histories (Claussner et al., 2025; Schiaroli et al., 2025; Tiwari et al., 2024; Ebrahimi et al., 2024).
Among application-dependent factors, the motion environment is particularly important because it perturbs the free surface and renews the interface. Direct LH2 studies have already documented temperature, level, and pressure variations during truck and marine transport, while recent train, truck, and carrier-oriented studies have shown that sloshing can alter boil-off gas (BOG) generation, pressure response, and control demand (Takeda et al., 2015; Maekawa et al., 2018; Smith et al., 2022; Zhu et al., 2023; Ryu et al., 2024).
A critical caution follows from this mobility framing: designs can be misled if sloshing is treated only as a mechanical-load problem or if the boil-off rate (BOR) and SP are characterized only under quiescent conditions. Intermediate filling ratios under motion can trigger short-duration but high-consequence pressure spikes or collapses through interface renewal, mixing, and control interactions–responses that are often invisible to either single-phenomenon view (Jeon et al., 2024; Kartuzova et al., 2024; Wolf et al., 2025).
From a thermo-fluid standpoint, the filling ratio also governs the size and stability of the ullage region across land, air, and sea LH2 storage and fuel platforms. Classical LH2 stratification studies show that the vapor-side temperature field between the free surface and the tank roof is typically non-linear and can become multi-layered, particularly at low fill where the ullage volume is large (Bailey & Fearn, 1964; Tatom et al., 1964; Kumar et al., 2007; Van Dresar et al., 1992). This matters because many reduced-order BOR/SP models infer effective wall-to-liquid and wall-to-ullage convective heat-transfer coefficients from differences in the bulk-average temperature under quiescent conditions and then reuse them for mobility scenarios at the same filling ratio. Under motion, however, sloshing modifies the wetting patterns, boundary-layer renewal, and local evaporation/condensation, so quiescent-inferred coefficients can become systematically biased unless the closure is calibrated against motion-enabled evidence (Yu & Lim, 2024; Wang et al., 2024b; Nuwantha et al., 2024; Kartuzova et al., 2024).
Many sloshing analyses, specifically those aimed at structural load assessment, solve single-phase free-surface hydrodynamics or prescribe simplified wall heat-flux boundaries. Such approaches remain valuable, but they cannot be assumed to bound mobility-relevant pressure excursions and BOR unless they are coupled to thermodynamics, ullage dynamics, and phase-change modeling (Chung et al., 2022; Zhu et al., 2023; Lv et al., 2024b).

1.3 Scope, Taxonomy, and How This Review is Structured

The literature is valuable but fragmented. General LH2 storage reviews, insulation-focused reviews, no-vent thermal-behavior reviews, SP model reviews, and transfer-safety reviews have each clarified parts of the field, but they often leave the storage tank itself disassembled into separate topics (Al Ghafri et al., 2022; Morales-Ospino et al., 2023; Yin et al., 2024; Wang et al., 2024a; Wang et al., 2024b; Schiaroli et al., 2025; Claussner et al., 2025).
This paper adopts a tank-centered, phenomenon-based taxonomy. It moves from LH2 physical fundamentals to tank architecture, insulation, quiescent boil-off and SP, operational transients, sloshing-coupled dynamics, diagnostics, modeling, safety and standards, cross-sector lessons, and future research priorities. The intent is to connect the literature in the same order in which the tank problem is usually confronted in engineering practice.
Distinctively, mobility-driven operations and regulatory evidence demands are treated as first-order design constraints: the review is organized along a Mobility–Regulatory–Evidence axis, with the level of validation and documentation accompanying each modeling or materials claim, typically required for approval in principle (AiP) and eventual certification.
Table 1 summarizes representative LH2 cases across sectors, assigns each case one primary evidence class (E1–E4), and notes supplementary contextual or supporting evidence where relevant to anchor this taxonomy in actual deployment history. This is important because marine cargo, trailer distribution, and legacy space systems already provide direct operational LH2 evidence, whereas several aircraft and urban air mobility (UAM) examples remain pre-certification technology demonstrations (Wolf et al., 2025; HESC, 2025; Inomata et al., 2025; Airbus, 2025; Fabrum, 2025).

1.4 What “Comprehensive” Means Here and What is Out of Scope

In this review, comprehensive does not mean the entire liquid-hydrogen value chain. Rather, it refers to the coupled phenomena that govern the storage-tank performance: thermophysical properties, non-equilibrium behavior, ortho-para conversion, geometry and scale effects, insulation materials and configurations, boil-off and SP, filling-ratio effects, transfer transients, motion-coupled free-surface behavior, diagnostics, modeling hierarchy, and safety-related design constraints (Wang et al., 2024c; Claussner et al., 2025).
Hydrogen production, liquefaction-plant design, supply-chain economics, and downstream propulsion integration are discussed only when they directly constrain tank behavior. Within that boundary, the objectives are to consolidate dispersed literature into a single multiphysics framework, extract design implications, and identify the validation and scaling gaps most relevant to next-generation LH2 tank development (Yin et al., 2024; Wang et al., 2024b; Schiaroli et al., 2025). Fig. 1 summarizes the integrated scope of the review and the principal feedback loops discussed throughout the manuscript.
A consistent cross-sector message is that a single, quiescent BOR number is necessary but not sufficient. Once the tank becomes mobile, sloshing alters stratification, interfacial area, and convective heat transfer, while pressure-control actions feed back into the phase change. In regulated sectors, that coupling is no longer merely academic. It defines what must be demonstrated, measured, and validated in an approval-ready evidence package. This is why BOR–sloshing coupling is treated as a central, mobility-era design issue rather than a niche topic (Liu et al., 2018b; Wu and Ju, 2023; Kartuzova et al., 2024; Ahizi et al., 2024).
Table 1 shows that LH2 storage is already being engineered across space, road/rail mobility, aviation/UAM, and marine transport, but each sector foregrounds a different subset of the same coupled loop. Space programs forced the community to treat SP and stratification as measurable, model-validatable phenomena within tight heat-flux budgets, yielding reference datasets such as SHIIVER and the K-Site series (Johnson, 2024; Kartuzova et al., 2024). Road mobility has driven mission-aware BOR accounting and operational optimization for trailer fleets, where the route statistics and fill strategies influence losses and venting frequency (Wolf et al., 2025). Aviation and emerging UAM concepts prioritize the gravimetric efficiency and manufacturable lightweight cryotanks, motivating composite/resin tank demonstrations and system-level aircraft technology roadmaps (Airbus, 2025; Fabrum, 2025). Marine transport is now transitioning from pilot-voyage evidence and measured heat-ingress data (HESC, 2025; Inomata et al., 2025) to pre-commercial carrier concepts and evolving regulation, including large-scale carrier approvals in principle and updating the International Maritime Organization (IMO) interim recommendations for LH2 carriage in bulk (IMO, 2024; DNV, 2024; Bureau Veritas, 2024).

1.5 Evidence Taxonomy and Confidence Levels Used in This Review

The LH2 tank literature now spans direct operational data, controlled experiments, surrogate-fluid studies, design approvals, and simulation-only investigations. This review uses the following evidence taxonomy (from strongest to weakest) to indicate the strength and transferability of a given claim and avoid over-generalization.
  • (1) E1: direct LH2 operational evidence (in-service or voyage/route scale).

  • (2) E2: direct LH2 experimental evidence (ground facilities with controlled boundary conditions and diagnostics).

  • (3) E3: indirect or transitional evidence (other cryogens/surrogate fluids, component demonstrations, and class AiP concepts).

  • (4) E4: simulation-only or parametric evidence without validation against E1/E2 data.

Throughout this review, each case is assigned one primary evidence class (E1–E4) representing the strongest technical anchor for the main engineering claim. When deployment, demonstration, or approval context is also relevant, it is reported separately as supplementary contextual evidence and not merged into the primary E1–E4 label. In compact tables, that supplementary information is shown in brackets for readability rather than as an additional taxonomy level. Where possible, engineering guidance is distilled primarily from E1/E2 evidence, whereas E3/E4 studies are used mainly to identify trends, sensitivities, and priority gaps for future validation.

2. Fundamental Physics of Liquid Hydrogen Relevant to Storage Tanks

2.1 Thermophysical Properties and Sensitivities

The behavior of LH2 tanks is unusually sensitive to variations in fluid properties in the cryogenic range. Small changes in temperature and pressure near the normal boiling region produce substantial shifts in saturation pressure, density, compressibility, heat capacity, viscosity, and thermal conductivity, which is why reference equations of state and validated property databases are preferred over constant-property approximations (Leachman et al., 2009; Huber et al., 2022; McDonald et al., 2024).
These sensitivities matter because pressure rise and boil-off depend on the total heat input and on how vapor compressibility, buoyancy-driven circulation, and interfacial transport evolve inside the tank. For the same reason, apparent disagreements across the literature often trace back to differences in initial pressure, fill ratio, thermal boundary condition, or hydrogen composition rather than to contradictions in the underlying physics (Wang et al., 2024a; Wang et al., 2024b).

2.2 Two-Phase Equilibrium Versus Non-Equilibrium in Real Tanks

A useful first approximation is to assume that liquid and vapor remain near thermodynamic equilibrium, so that the bulk phases are represented by a common saturation state or by weakly non-equilibrium lumped states tied to pressure. That approximation is attractive for reduced-order models, but real LH2 tanks seldom remain in perfect equilibrium. Wall heat leakage creates boundary layers, differential warming of the ullage and liquid, and finite-rate interfacial transport, all of which drive the tank away from a fully uniform state.
Equilibrium models suppress spatial temperature differences and interfacial resistance, whereas non-equilibrium models can more faithfully represent mild surface evaporation, stratification, and transient pressure response. Therefore, the key issue is not whether one class of model is universally correct, but which level of approximation remains defensible under a given operating regime (Nuwantha et al., 2024; Wang et al., 2024b).
In the simplest near-equilibrium approximation, the interface may be represented by Tl,int ≈ Tv,int ≈ Tint and pint ≈ psat(Tint), where the subscript “int” denotes an effective interfacial state. This approximation is useful for hold-time screening, but its validity deteriorates once persistent stratification or strong transient phase redistribution develops.

2.3 Ortho–Para Conversion as an Additional Thermodynamic Source Term

Hydrogen exists as orthohydrogen and parahydrogen, and the equilibrium composition shifts strongly with temperature. If liquefaction and pre-storage conditioning do not sufficiently move the fluid toward the low-temperature para-rich equilibrium state, residual ortho-to-para conversion during storage releases heat and shortens dormancy (Leachman et al., 2009; Jiao et al., 2024).
Recent storage concepts also show that the role of spin conversion is path-dependent. In some integrated storage concepts, para-to-ortho conversion during warm-up can act as an endothermic sink along a specific thermal trajectory. Thus, ortho-para conversion is better interpreted as a thermodynamic source term embedded in the tank heat balance rather than as a detail confined to liquefaction (Peng et al., 2013; Matveev & Leachman, 2023b).

2.4 Cryogenic Heat-Transfer Regimes Relevant to LH2 Tanks

LH2 tanks operate in an unusual regime: the ambient-to-cryogen temperature difference is very large, but the allowable inward heat flux is very small. Heat reaches the fluid by solid conduction, radiation, and residual-gas conduction if the vacuum quality degrades; once inside the tank, it is redistributed by natural convection, interfacial evaporation and condensation, and by boiling processes adjacent to warm surfaces in local transients (Horie et al., 2015; Wang et al., 2024a).
Under long quiescent storage, the dominant response is usually mild evaporation with buoyancy-driven recirculation and stratification rather than vigorous boiling. Nevertheless, boiling correlations become much more important during chill-down or filling. The gravity level and orientation can also affect how the thermal layers develop, which is why the transfer of conclusions between ground tanks, mobile systems, and space applications must be considered carefully (Liu et al., 2018a; Claussner et al., 2025).

2.5 Interfacial Phenomena and Evaporation Drivers

The liquid-vapor interface in an LH2 tank is an active transport boundary. Net evaporation depends on the interface temperature, liquid subcooling, vapor superheat, hydrostatic pressure, and the rate at which the nearby fluid is renewed by convection. Hence, the same nominal heat leak can produce different evaporation responses depending on how the interface couples to the surrounding flow and temperature fields (Zuo et al., 2023; Nuwantha et al., 2024).
That importance is evident in classical and modern stratification studies. Bailey and Fearn (1964), Tatom et al. (1964), and Kumar et al. (2007) all show in different ways that evaporation is not merely an output calculated after the thermal field is solved; it is part of the thermal problem itself, and it provides a bridge between insulation and later system-level responses such as BOR, SP, and motion sensitivity.

3. Tank Concepts and Containment Architectures

3.1 Tank Concepts, Containment Architectures, and Structural-Material Considerations

3.1.1 Design drivers and constraints

The design of an LH2 tank is governed by competing requirements: gravimetric and volumetric efficiency, allowable boil-off, dormancy time, pressure-control capability, structural integrity at cryogenic temperature, inspectability, and compatibility with the intended duty cycle. In mobile applications, packaging, vibration, and repeated operation become particularly important. In stationary systems, long hold time and operational simplicity often dominate (Kang et al., 2022; Claussner et al., 2024).

3.1.2 Geometry families and their thermal-fluid implications

Geometry affects LH2 storage through the surface-area-to-volume ratio, wetted-wall distribution, ullage shape, natural-convection paths, and likely sloshing modes. Spherical tanks are often attractive thermally and structurally, whereas cylindrical or horizontal cylindrical tanks are easier to integrate into vehicles and transport systems. Comparative studies show that cylindrical and spherical LH2 tanks can exhibit materially different thermal performance, even at similar volumes, because geometry alters the heat ingress distribution and the internal response (Wang & Mérida, 2024).
The importance of geometry grows once the fill ratio is considered. The same tank can behave differently at different liquid levels because the proportions of wetted wall, dry wall, and free surface change together. Therefore, the geometry and fill level need to be treated jointly rather than independently (Matveev & Leachman, 2023a; Wang et al., 2024e).

3.1.3 Double-wall & vacuum-jacket architectures

Most high-performance LH2 tanks use a double-wall architecture in which an inner vessel is surrounded by an outer shell, with a vacuum-insulated annulus that may contain multilayer insulation (MLI), powders, or hybrid systems. The strength of this arrangement is that it separates containment from thermal protection. On the other hand, its weakness is that vacuum deterioration, penetrations, and local compression can sharply degrade performance (Claussner et al., 2024; Yin et al., 2024).

3.1.4 Internal structures and thermal bridges

In practical tanks, a substantial portion of heat ingress may arise not from the nominal insulation field but from supports, piping penetrations, instrumentation ports, valves, and internal hardware. These unavoidable details produce thermal bridges and often explain why the measured performance deviates from idealized wall-plus-insulation models. They also interact with flow if they function as baffles, supports, or internal appendages (Fesmire, 2015; Inomata et al., 2025).

3.1.5 Structural materials, support members, and joining constraints

Beyond geometry and vacuum architecture, LH2 containment is also constrained by the choice of structural material. Recent cryogenic materials reviews continue to identify austenitic stainless steels as the baseline family for LH2 service, while also noting renewed interest in higher-strength metallic and composite options for weight-sensitive applications. American Bureau of Shipping (ABS) material guidance similarly notes that chromium-nickel-manganese 200-series and chromium-nickel 300-series austenitic steels are preferred candidate families for low-temperature service, with low-carbon 304L and 316L remaining the traditional LH2 tank materials (Qiu et al., 2021; ABS, 2025a).
At the same time, the material landscape is no longer static. Nevertheless, the maturity of the available LH2-relevant evidence differs across candidate alloy families. High-strength 15% Ni steel already has direct LH2-relevant promise. Namegawa et al. (2025) reported sufficient 20 K toughness for base metal and weld joints, together with the possibility of reducing shell thickness relative to SUS304L in large LH2 tanks. By contrast, the high-manganese-related literature cited here is more heterogeneous, spanning Ni/Mn-replaced austenitic stainless steels and Fe-high-Mn austenitic steels, and remains less mature in terms of direct LH2 tank qualification. Therefore, the current LH2-oriented guidance remains more cautious for high-manganese-related systems, and recent LH2 design studies still treat their hydrogen compatibility and weld qualification as unresolved. Accordingly, both high-Ni and high-Mn-related families should be discussed as emerging candidates, but not at the same maturity level. High-Ni 15% Ni steel presently has the stronger LH2-specific evidence base, whereas high-Mn-related systems remain earlier-stage options for LH2 containment (Namegawa et al., 2025; Cho et al., 2024; Lee et al., 2024; Korean Register, 2024; Cha et al., 2025).
This trend should be interpreted carefully in the review. High-nickel concepts–especially 15% Ni steel–are attracting attention because they may enable thinner shell sections while retaining high cryogenic toughness, and because their LH2-specific 20 K evidence is already more direct than that of most other emerging alloys. High-manganese-related austenitic systems are being discussed primarily because of their potential cost and resource advantages, but their LH2 applicability remains less mature in terms of hydrogen compatibility, weld qualification, and code acceptance. Hence, these two families should not be presented as equally mature alternatives to the conventional stainless-steel baseline. High-Ni is the nearer-term structural candidate, whereas high-Mn-related systems remain more exploratory pending further LH2-specific validation (ABS, 2025a; Namegawa et al., 2025; Cho et al., 2024; Lee et al., 2024; Korean Register, 2024; Boddorff, 2025).
Support design is the point at which structural material selection and thermal performance meet directly. In most double-wall tanks, the inner and outer shells are linked by discrete support members that must carry gravity, pressure, and motion loads while minimizing conductive heat leak. Therefore, the review literature continues to cite G10, G10-CR, and G11 glass-epoxy laminate composite as attractive low-conductivity support candidates, whereas larger or mobility-oriented tanks often revert to stainless supports for manufacturability and robustness at the cost of increased heat leakage (Kang et al., 2022; Cha et al., 2025). The engineering difficulty is that replacing SUS-to-SUS support paths with G10/G11 or other composite supports does not eliminate the problem. It relocates it to dissimilar-material joining, differential thermal contraction, and long-term bond reliability. Cryogenic bonding studies on composite-metal joints show that adhesive selection is safety-critical, and National Aeronautics and Space Administration (NASA) bonding/sealing evaluations reported that stainless-steel penetrations through graphite/epoxy structures were difficult to bond and could compromise vacuum retention at cryogenic temperatures (Kang et al., 2007; Glass & Smith, 1997). For this reason, support members should be treated not as secondary hardware but as a decisive design feature in the LH2 tank architecture.
This is also where one of the most difficult practical design problems emerges. The inner and outer vessels are commonly metallic, while the most thermally attractive support materials may be G10, G10–CR, G11, or other glass-fiber-reinforced composites, which have much lower conductivity than stainless steel. Once these materials are introduced, however, the design challenge shifts from conductivity alone to interface reliability: differential thermal contraction, adhesive durability, mechanical preload retention, sealing, and inspectability all become important. Cryogenic joint studies and NASA bonding evaluations have long shown that dissimilar-material interfaces can become a limiting feature of a tank even when the candidate support material appears attractive in coupon-level testing (Glass & Smith, 1997; Kang et al., 2007; ABS, 2025a). Cryogenic epoxy testing for MLI standoff attachment already shows how sensitive bonded interfaces can be to differential contraction and substrate choice, which is highly relevant to metallic-shell/composite-support concepts (Alberts et al., 2017).
Table 2 organizes this selection problem in a compact engineering form. Its main message is that the thermally most attractive choice is not always the easiest choice to qualify. Low-conductivity composite supports reduce parasitic heat leak, but they also intensify concerns about differential contraction, bonded or bolted metallic-composite interfaces, inspectability, and long-term vacuum-service durability (ABS, 2025a; Boddorff, 2025; Cho et al., 2024).

3.1.6 Scaling laws and the danger of “small-scale intuition”

Laboratory-scale studies are indispensable, but scale-up is not straightforward. As the tank size increases, the surface-area-to-volume ratio decreases, absolute heat ingress grows, localized bridges change their relative importance, and thermal time scales lengthen. Therefore, larger tanks can display more persistent stratification and different venting or hold behavior than small tanks, even under nominally similar normalized conditions (Matveev & Leachman, 2023a; Inomata et al., 2025).
The lesson from the review work is that the architecture, thermodynamics, and scale must be considered together. This is particularly important for maritime and terminal-scale storage, where the assumptions that hold at the laboratory scale may no longer be credible (Smith et al., 2022; Wang et al., 2024d).

3.2 Insulation Physics, Configurations, and Materials

3.2.1 Principal heat-ingress pathways and their relation to BOR

From a tank-engineering standpoint, boil-off does not arise from a single wall heat leak but from a network of thermal pathways: solid conduction through supports and penetrations, radiation across the annulus, residual-gas conduction under degraded vacuum, and localized leakage through practical hardware interfaces. This is why the effective performance of a cryogenic insulation system is better understood as a thermal-resistance network than as a uniform slab of material (Fesmire, 2015; Ratnakar et al., 2023; Yin et al., 2024).
The dominant path also changes with the vacuum strategy and scale. Small and medium vessels often target very high vacuum so that MLI can operate near its best regime, whereas larger tanks may accept a moderate vacuum and rely more on bulk-fill media because maintainability and annulus practicality become limiting (Sass et al., 2008; Kang et al., 2024; Mata et al., 2025). Fig. 2 presents a schematic diagram of this configuration-level interpretation of heat ingress.
In the simplest steady-state screening approximation, the ambient heat leak may be represented by an equivalent thermal-resistance network:
(1)
Q˙in≈(Thot-Tcold)Rtot
(2)
Rtot-1=Rcond-1+Rrad-1+Rgas-1+Rbridge-1+Rsupport-1
In Eqs. (1)–(2), Q̇in is the net heat ingress into the inner vessel (W); Thot and T are the effective warm- and cold-side boundary temperatures (K); Rtot is the equivalent thermal resistance between those boundaries (K/W); Rcond represents solid conduction through insulation/bulk media’ Rrad is the radiation across the annulus; Rgas is the residual-gas conduction under an imperfect vacuum, and Rbridge and Rsupport thermal-bridge paths through penetrations and supports. The inverse-sum form in Eq. (2) is a schematic representation of parallel heat-ingress paths and should be adapted to the actual tank architecture.
Although this resistance-network approach is highly useful for rapid screening, it should be interpreted as a first-order architecture model. As the analysis moves toward transient SP, filling, or sloshing, the governing problem shifts from an effective heat-leak estimation to a coupled redistribution of that heat within the wall, liquid, ullage, and interface.
Here, Rtot is interpreted as an equivalent series-parallel resistance assembled from the dominant ingress paths. This approximation is useful for rapid architectural screening and for relating coupon-level insulation data to tank-level heat ingress, even though more detailed models more explicitly resolve local wall-to-fluid heat partition (Crabb, 2022; Kim & Lee, 2025; Aasen et al., 2025).

3.2.2 Insulation performance metrics and test conditions

The insulation performance is reported in many ways: total heat leak, areal heat flux, apparent conductivity, effective conductivity, boil-off-equivalent loss, or dormancy gain. None of these metrics is fully portable unless the vacuum level, warm and cold boundary temperatures, layer density, geometry, and mechanical condition are reported simultaneously. Standardization work repeatedly emphasizes this point because cryogenic insulation performance is configuration-dependent rather than purely material-intrinsic (Fesmire, 2015; Ratnakar et al., 2023).
This distinction becomes even more important when coupon-level insulation data are translated into tank-level expectations. Apparent or effective conductivity measured under controlled laboratory conditions may be useful for ranking materials, but designers ultimately need architecture-level quantities such as the total heat leak, bridge sensitivity, and BOR-equivalent impact under the relevant fill policy and duty cycle. In other words, insulation testing and tank performance assessment should be linked explicitly rather than treated as separate evidence streams (Fesmire, 2015; Chen et al., 2024; Kang et al., 2024).

3.2.3 Multilayer insulation systems

MLI and related variable-density multilayer systems remain the benchmark passive insulation technologies for high-performance LH2 tanks under a sufficiently good vacuum. Their appeal lies in suppressing radiation with many reflective layers while keeping solid conduction very low through spacers. Recent reviews still identify MLI as the baseline option for compact, high-vacuum LH2 tanks where a minimal passive heat leak is the dominant design objective (Yatsenko et al., 2022; Yin et al., 2024; Simanullang, 2025).
At the same time, MLI is quite sensitive to installation conditions and environmental factors. Layer compression, blanket nonuniformity, degraded vacuum, and contact shortcuts can sharply increase conductive transfer. Experimental degraded-vacuum studies show that the nominal superiority of MLI holds only when blanket integrity and vacuum quality are maintained, making MLI a high-performance and high-discipline technology (Sun et al., 2009; Chen et al., 2024).

3.2.4 Bulk-fill and powder insulation

For larger tanks, bulk-fill systems such as perlite and hollow glass microspheres remain attractive because they can operate under moderate-vacuum conditions and may be easier to maintain at scale. In that regime, the design problem shifts from achieving the lowest conceivable heat flux to balancing heat leak, pump-down time, settling, annulus behavior, and operational practicality (Ratnakar et al., 2023; Mata et al., 2025).
The literature shows that bulk-fill selection cannot be reduced to nominal conductivity alone. Comparative work on perlite and glass bubbles demonstrates differences in evacuation behavior, vacuum sensitivity, and long-term handling that become important in real large-scale applications (Sass et al., 2008; Wang et al., 2024f; Mata et al., 2025).

3.2.5 Foam and composite cryogenic insulation

Foam-based insulation occupies a different niche from MLI. Closed-cell foams offer mechanical robustness, conformability, and utility where vacuum is unavailable or not reliably maintained, but they typically do not match the thermal performance of high-vacuum MLI in stringent dormant LH2 storage. Therefore, their value lies less in replacing MLI than in complementing it where mechanical resilience or atmospheric-stage operation matters (Yatsenko et al., 2022; Ratnakar et al., 2023).
Recent composite-insulation studies and aircraft-tank assessments reinforce this point. In system-level trade studies, foam can be useful in combination with multilayer or shield concepts, but it may lead to unacceptable venting penalties when used as a direct substitute for vacuum-jacketed high-performance insulation (Wang et al., 2020a; Johnson et al., 2023).
The mobile and UAM context helps explain why resin-based and thermoplastic composite concepts continue to attract attention despite the challenges of cryogenic service. In airborne systems, every kilogram saved at the tank level affects the payload, range, and integration flexibility, so lightweight composite shells and liners are being explored more aggressively than in stationary tanks. Recent work on polymeric and graphene-enhanced composite systems has focused on cryogenic cycling, impact tolerance, permeability, and recyclability, suggesting that the key question is no longer whether lightweight resin-based concepts are interesting, but whether they can retain structural and barrier performance after repeated ultra-low-temperature exposure (Gul et al., 2023; Johnson et al., 2023; Andrew et al., 2025; Olsson et al., 2024).

3.2.6 Hybrid systems and vapor-cooled shielding concepts

Hybrid insulation systems, especially those involving vapor-cooled shields (VCS), are one of the most active areas of current LH2 research. By intercepting heat at intermediate temperatures before it reaches the coldest boundary, VCS converts insulation from a purely passive barrier into a staged thermal-management architecture. Recent transient and CFD-based studies show that the shield location, operating time, mass flow, and integration with multilayer systems materially affect dormancy and vapor consumption (Jiang et al., 2021; Jiang et al., 2023; Li et al., 2024b; Zhu et al., 2024).
An important extension of this idea is its coupling with para-ortho conversion. Recent analyses suggest that when VCS is integrated with spin-conversion-related thermodynamic effects, dormancy can be extended further, and the required MLI layer count may be reduced under some conditions. These benefits are real, but they come with greater architectural and operational complexity (Shi et al., 2023; Lv et al., 2024a; Li et al., 2024b). Fig. 3 compares the resulting operating envelopes of the major insulation families.

3.2.7 Long-term degradation and maintainability

A major weakness of many insulation studies is that they emphasize the beginning-of-life calorimetric performance rather than the service-life performance. In real LH2 systems, vacuum degradation, blanket compaction, outgassing, handling damage, and localized compression near supports can all erode performance over time. Earlier degraded-vacuum MLI experiments already showed how sensitive heat leaks can be to rarefied-gas pressure (Sun et al., 2009; Fesmire, 2015).
Recent safety-oriented work adds another dimension by examining the insulation integrity under fire or accident conditions, reinforcing the point that inspection access, leak detection, retest logic, and replacement strategy belong within insulation design rather than outside it (Camplese et al., 2025). Fig. 4 presents the associated degradation-to-maintainability pathway.

3.2.8 Practical design guidelines distilled from literature

The literature supports several broad guidelines. High-vacuum MLI remains the baseline choice when minimal passive heat leak is the overriding objective and vacuum integrity can be maintained reliably. For larger tanks or moderate-vacuum service, bulk-fill systems such as perlite or glass microspheres become increasingly attractive. Foam is most useful when deployed as part of a composite strategy. Hybrid or VCS-based systems are justified when the dormancy targets are stringent enough to reward added complexity (Yin et al., 2024; Ratnakar et al., 2023; Kang et al., 2024; Mata et al., 2025).
The broader lesson is that insulation should not be selected by ranking materials from best to worst. Insulation should be selected by matching an insulation family to the tank scale, heat-ingress map, vacuum strategy, motion environment, and maintenance philosophy. This is exactly where insulation connects to the next Section on boil-off and SP.

4. Boil-Off, Self-Pressurization, and Operational Transients

4.1 Quiescent Tanks: Boil-Off and Self-Pressurization

4.1.1 BOR definitions and normalization choices

The term boil-off rate (BOR) is used widely in the LH2 literature, but not always consistently. Depending on the study, BOR may be reported as kg/day, percent/day, equivalent heat leak, or dormancy penalty. These metrics are all useful, but they are not directly interchangeable unless the inventory, pressure-control condition, and storage mode are reported clearly (Al Ghafri et al., 2022; Morales-Ospino et al., 2023).
The BOR and SP are also not identical. In a vented tank, evaporation appears directly as mass loss. In a no-vent tank, the first consequence of absorbed heat may be pressure build-up rather than immediate external loss. Therefore, the BOR, dormancy time, and pressure-rise rate should be interpreted together in quiescent LH2 storage (Wang et al., 2024a; Wang et al., 2024b).

4.1.2 Mechanistic sequence from heat ingress to pressure rise

SP begins when external heat ingress is partitioned among the wall, ullage, liquid bulk, and interface. The resulting process is not a single-step conversion of heat leak into vapor generation. Heat absorbed by the liquid can promote thermal expansion, stratification, and interfacial evaporation, while heat absorbed by the ullage changes vapor temperature and compressibility even before large net mass loss occurs (Hasan et al., 1991; Van Dresar et al., 1992; Barsi & Kassemi, 2008).
Later models sharpened this picture. Wang et al. (2022), Nuwantha et al. (2024), and Wang et al. (2024b) all showed that pressure rise depends strongly on the vapor temperature, interfacial mass transfer, liquid thermal expansion, and the internal partition of heat, meaning that identical total heat leak does not guarantee identical SP behavior. Fig. 5 presents the overall mechanistic sequence.

4.1.3 Thermal stratification and rollover-type concerns

Thermal stratification is one of the most influential internal processes in quiescent LH2 storage. As the tank absorbs heat, warmer regions form near the wall and in the upper vapor space while colder bulk liquid may persist below. Even modest vertical temperature gradients can alter the local density, buoyancy, saturation condition, and interfacial transfer because the properties of LH2 are strongly temperature-sensitive (Bailey & Fearn, 1964; Tatom et al., 1964; Kumar et al., 2007).
Surface evaporation does not merely result from stratification; it helps shape it. Kumar et al. (2007) showed that surface evaporation substantially modifies stratification development in tanks of different aspect ratios. In the LH2 context, it is more appropriate to speak of destratification-induced transients or rollover-type concerns than uncritically transfer liquefied natural gas (LNG) terminology, but the engineering message remains. Sudden mixing in a stratified tank can abruptly alter the pressure evolution (Wang et al., 2024a).

4.1.4 Filling ratio effects under quiescent conditions

Filling ratio is a primary thermo-fluid design variable rather than a simple inventory descriptor. It changes the ullage volume, free-surface area, wetted-wall fraction, liquid thermal mass, and the partition of heat ingress between the liquid and vapor. Consequently, it changes the magnitude of pressure rise and the mechanism through which that rise develops (Van Dresar et al., 1992; Yu & Lim, 2024).
Recent low-fill and holding-time studies reinforce the classical K-site results: lower fill often increases the ullage sensitivity and pressure responsiveness, while higher fill increases the liquid thermal inertia and alters the balance between the vapor-side and liquid-side warming. The practical implication is that there is no universal optimal fill ratio outside the context of a specific geometry and mission (Wang et al., 2024e; Yu & Lim, 2024; Qu et al., 2025). Fig. 6 summarizes the qualitative fill-ratio regimes used in the present review.
A practical perspective on why the fill ratio deserves special attention comes from LNG shipping heritage. For membrane LNG carriers, severe sloshing loads at intermediate fills led to operational filling restrictions (often expressed as barred bands in the vicinity of ~10%–70% of the tank height) and to sloshing risk-assessment frameworks used by classification societies and operators (Graczyk et al., 2007; Shell, 2022; Korean Register, 2022). In LH2 mobility applications–road trailers, aircraft/UAM tanks, and marine fuel systems–such avoidance is rarely feasible: normal consumption and refueling inevitably sweep the tank through intermediate fill ratios. Therefore, the design goal shifts from ‘avoid the critical fill window’ to ‘remain safe and predictable while passing through it,’ which motivates treating the fill ratio as a mission variable and linking it explicitly to motion-coupled BOR and pressure response (Sections 4.2–5).
Several first-order scaling arguments clarify why LH2 is particularly sensitive. According to IMO Resolution MSC.420(97), the density of saturated liquid hydrogen at the normal boiling point is approximately 70.8 kg/m3, whereas LNG typically lies in the range 430470 kg/m3 (GIIGNL, 2009; IMO, 2016). For comparable kinematic excitation, inertial contributions to the sloshing impact pressure scale roughly with ρ, suggesting that purely structural impact pressures for LH2 can be ~6–7 times lower than for LNG at the same velocity scale. By contrast, the normalized boil-off fraction increases as stored mass decreases. Under a steady heat leak Q̇, a common approximation is ṁBOG ≈ Q̇/hfg and BOR (%) ≈ (Q̇/(ρLVLhfg))× 86400 × 100, highlighting that for a fixed heat ingress per tank volume, BOR scales inversely with ρLhfg. This is can explain why LNG-like percentage BOR targets for LH2 generally require much lower heat ingress via vacuum jackets/MLI and careful mitigation of thermal bridges (Aasen et al., 2025). Here, Q̇ is the net heat ingress (W); hfg is the latent heat of vaporization (J/kg); ṁBOG is the boil-off-gas mass flow rate (kg/s); ρL is the liquid density (kg/m3), and VL is the liquid volume (m3); 86400 converts seconds to days.
Under motion, the fill ratio governs the hydrodynamic regime and the thermodynamic sign of the pressure response. Sloshing-driven mixing can enhance wall-to-fluid convection and renew the liquid–vapor interface, which may amplify evaporation and SP when heat ingress is high or when the vent capacity is limited, but the same mixing can also accelerate condensation and produce rapid ‘pressure collapse’ when a cold liquid surface is exposed to warm ullage. Experiments and validated models in cryogenic tanks–ranging from LN2 surrogate studies to LH2 K-site analyses–consistently report large pressure excursions and boil-off changes whose magnitude depends strongly on the fill level and excitation conditions (Jeon et al., 2024; Zhu et al., 2023; Kartuzova et al., 2024; Hauser, 2024; Gambioli et al., 2025). At low fill, the enlarged ullage region can sustain non-linear stratification. Therefore, the convection coefficients inferred from quiescent heat ingress should be treated cautiously when applied to motion-coupled BOR/SP predictions (Bailey & Fearn, 1964; Van Dresar et al., 1992; Yu & Lim, 2024).
For ships and offshore systems, the fill-ratio challenge is compounded by system-integration constraints. In large LH2 carrier concepts, the hull form and draft are explicitly influenced by the low specific gravity of liquefied hydrogen cargo, making ballast/draft management and mission profile part of the containment-system design envelope rather than an afterthought (JSEA, 2024). Moreover, unlike many land tanks that can be supported on dedicated insulated foundations, shipboard tanks must be supported and interfaced with hull structures and piping, increasing the prominence of penetrations and thermal bridges; voyage-scale heat-ingress estimation and uncertainty studies highlight the disproportionate contribution of supports and local cold spots to total heat leak (Inomata et al., 2025; Aasen et al., 2025).

Design Note 1

Mobility pitfalls of single-phenomenon studies (quiescent BOR-only or sloshing-only)
A near-equilibrium screening relation for ullage pressure evolution under fixed ullage volume can be expressed as:
(3)
dPudt≈RTuVu(m˙evap-m˙cond-m˙vent)
In Eq. (3), motion affects ṁevap and ṁcond through interface renewal and mixing, and the control system sets ṁvent (and PBU/recirculation if present).
  • (1) Track, at minimum: (i) heat-ingress budget and bridge inventory; (ii) fill ratio and ullage volume; (iii) motion input (acceleration spectra or time history); (iv) interface/mixing indicators; and (v) vent/PBU capacity and control logic.

  • (2) Beware: a sloshing-only structural program may size the impacts correctly but still under-size the pressure-control margin, while a quiescent BOR-only program may meet dormancy targets but miss transient pressure spikes/collapses under motion.

  • (3) Use this coupled view to define a mobility-ready design envelope and assemble the evidence package expected by IMO/class/AiP pathways (Section 8.4.1).

4.1.5 Pressure control and BOR mitigation

The most direct mitigation strategy is to reduce heat ingress through better insulation and reduced bridging, but once the residual heat leak remains non-negligible, pressure management must be treated explicitly. Options include venting, thermodynamic vent systems, active cooling, reliquefaction, zero-boil-off (ZBO) systems, and route-or mission-oriented operating strategies (Morales-Ospino et al., 2023; Wang et al., 2024b; Zheng et al., 2019).
Both classical and recent work show that active or hybrid mitigation can be effective when justified by the mission value. Thermodynamic venting, cryocooler-assisted boil-off reduction, and active pressure management each shift the problem from passive insulation alone toward integrated thermal management (Haberbusch et al., 2004; Plachta et al., 2014; Zuo et al., 2020; Ha et al., 2024; Wan et al., 2025; Blakseth et al., 2025).
Mitigation philosophy also matters. Some concepts attempt to cool a broad region of the tank wall, whereas others intercept heat only in the most dominant ingress zones or use local active devices to delay pressure rise. Recent large-scale analyses comparing broad-area and local-area cooling show that the most efficient choice depends on the tank size, support layout, dominant bridge paths, and allowable system complexity. This reinforces a central point of the present review. Boil-off reduction cannot be prescribed independently of the architecture, because the optimal mitigation strategy depends on where the residual heat leak actually enters the system (Haberbusch et al., 2004; Blakseth et al., 2025).

4.1.6 What data is missing

Despite decades of study, the field still depends heavily on a relatively small number of benchmark experiments. The K-site lineage remains foundational for model assessment, and modern validation work still returns to those classical tests. That continued reuse is useful, but it also reveals how few modern datasets provide simultaneous pressure, multi-point temperature, heat-flux partition, and geometry-specific realism (Hasan et al., 1991; Wang et al., 2024b; Kartuzova & Kassemi, 2025).
Therefore, the most important missing data are not just more pressure histories, but validation-grade datasets for horizontal tanks, low-fill tanks, and larger application-relevant systems with uncertainty budgets and clear boundary conditions. This gap becomes even more consequential once transfer, cyclic operation, and motion are considered (Wang et al., 2024a; Wan et al., 2025).

4.2 Filling, Withdrawal, and Operational Transients

4.2.1 Chill-down and pre-cooling as a dominant transient

Practical LH2 tanks rarely begin in a pre-conditioned cryogenic state. Before stable storage can begin, the tank wall, piping, and internal hardware must be chilled sufficiently so that the incoming LH2 is not consumed primarily by local boiling and flash evaporation. This makes chill-down and pre-cooling one of the dominant operational transients rather than a mere preparatory detail (Hartwig et al., 2023; Claussner et al., 2025).
Recent experiments show how important this stage is. Hartwig et al. (2023) reported rapid chilldown and fill tests on a flightweight aluminum tank and showed that the evolving boiling regime and wall thermal state strongly affect the pressure, fill level, and transfer time. Related fuelling studies also show that pre-cooling can materially reduce the transfer losses, especially in smaller or aviation-oriented tanks (Ghaffari-Tabrizi et al., 2022; Hartwig et al., 2023).

4.2.2 Filling dynamics and flashing

Once pre-cooling is underway, the filling process introduces its own complexity. Liquid injection changes the free-surface position, redistributes the thermal energy, and may cause flash evaporation or vapor condensation depending on the local state. No-vent fill remains the central framework for this problem. From Vaughan (1991) to more recent orbital and terrestrial studies, the literature consistently shows that the inlet condition, vent strategy, receiver-tank pressure, and initial temperature all strongly affect the fill efficiency and pressure development (Vaughan, 1991; Wang et al., 2015; Ma et al., 2017; Ma et al., 2023).
Modern transfer models connect these transients to practical decisions. Petitpas (2018), Gil-Esmendia et al. (2025), and Klopcic et al. (2025) reported that fill losses depend strongly on how transfer is driven and conditioned. Warm lines, pressure-driven transfer, and unfavorable receiver-tank conditions can sharply increase reverse gas or boil-off losses, making the filling dynamics a core storage problem rather than a peripheral station issue.

4.2.3 Withdrawal and pressure/temperature evolution

Withdrawal is not simply the reverse of filling. As liquid is removed, the ullage expands, the wetted-wall fraction changes, and the remaining fluid inherits thermal memory from prior operations. The pressure and temperature then evolve through a competition among decompression, flash evaporation, heat leak, and wall-to-fluid heat release (Tani et al., 2021; Appel et al., 2025).
Large-scale pressure-reduction experiments show that even apparently simple discharge events can exhibit nonintuitive behavior, such as early pressure recovery or delayed boiling response. This is important because it means that downstream use or unloading is inseparable from storage-state evolution (Tani et al., 2021; Lee et al., 2025).

4.2.4 Cyclic operation and mission profiles

In realistic service, LH2 tanks seldom see a single fill-and-hold cycle. They experience repeated sequences of filling, holding, withdrawal, venting, and refilling. Under such conditions, a single representative steady BOR becomes less meaningful because each new cycle begins from a thermodynamic state shaped by the previous one (Appel et al., 2025; Schiaroli et al., 2025).
This is particularly true in the mobile sectors. Trailer distribution, aircraft refueling, and marine unloading all introduce operation-dependent losses and constraints. Recent route-based trailer validation and aircraft refueling modeling both reinforce the idea that operational history is part of the tank state, not external to it (Wolf et al., 2025; ten Damme et al., 2025).
This is also why the fueling protocol deserves more attention than it often receives in simplified storage studies. Pre-cooling the receiving tank, managing its pre-transfer pressure state, and sequencing the transfer to minimize flash losses can materially reduce the turnaround time and cumulative hydrogen loss. Cryogenic fueling analyses aimed at reducing hydrogen boil-off during tank filling already suggest that operational preparation can be nearly as important as hardware choice in short-turnaround service (Ghaffari-Tabrizi et al., 2022). Fig. 7 presents the operational sequence and its inherited thermal memory.

4.2.5 Modeling needs for control-oriented predictions

These transients place special demands on modeling. High-fidelity CFD is essential for local boiling, flash, and interfacial field evolution, but it remains too expensive for repeated mission analysis or control studies. On the other hand, overly simple equilibrium models may miss wall thermal memory, non-equilibrium effects, or vent-logic behavior that dominate actual operation (Claussner et al., 2025).
Therefore, most useful operational models occupy a middle ground: physically meaningful yet fast enough for repeated scenario analysis. Recent node-based and reduced-order models for filling, transfer, and trailer operation show that this model class is likely to become increasingly important for digital twins, transfer planning, and system-level optimization (Ma et al., 2023; Appel et al., 2025; Gil-Esmendia et al., 2025; Klopcic et al., 2025).

5. Sloshing and Coupled Cryogenic Dynamics

5.1 Sloshing Regimes Across Filling Ratios

The sloshing response of an LH2 tank depends strongly on the fill ratio because the liquid inventory, ullage volume, free-surface extent, and natural frequencies all shift with the liquid level. Under weak excitation, the free surface may remain moderately wavy. Near resonance or under stronger forcing, the response can become strongly nonlinear and transition toward breaking, impact, spray, and entrainment (Ludwig et al., 2013; Jeon et al., 2022; Lv et al., 2024b).
The fill ratio is not just a geometric parameter in motion-coupled storage; it is a regime-defining variable because these hydrodynamic regimes change the interfacial area and renewal rate. That is why no single fill level can be labeled universally as the most severe for all LH2 tanks (Smith et al., 2022; Ryu et al., 2024).

5.2 Sloshing-Induced Mixing and Heat Transfer Enhancement

The primary thermal consequence of sloshing is intensified mixing and interfacial renewal. Oscillation, overturning, and breaking redistribute fluid with different thermal histories and repeatedly expose the interface to new liquid and vapor parcels, changing heat and mass transfer (Liu et al., 2018b; Jeon et al., 2024).
Direct LH2 modeling confirms this. Heavy-duty-truck and train studies show that large-amplitude sloshing can increase convective heat transfer, alter the liquid and vapor temperatures differently, and either reduce or increase the pressure depending on the duration and context (Zhu et al., 2023; Ryu et al., 2024; Im et al., 2025).
A key mobility modeling pitfall, which is relevant across road trailers, aircraft tanks, and ship tanks, is treating wall-to-fluid heat transfer and ullage stratification as quasi-static while solving only the hydrodynamics of sloshing. Under low fill, the ullage temperature field is non-linear and motion-sensitive. Therefore, representing it through a single bulk convection coefficient–or recycling a coefficient derived from quiescent heat-ingress tests at the same filling ratio–can mask the mechanisms that govern transient pressure spikes/collapse (Bailey & Fearn, 1964; Yu & Lim, 2024; Wang et al., 2024b; Nuwantha et al., 2024).

5.3 Coupled Thermo-Fluid Loop Under External Motion

The thermodynamic role of sloshing is best described as a closed loop. External motion deforms the free surface; the deformed surface renews the interface and changes wall wetting. This alters the evaporation–condensation balance and internal mixing, and the resulting redistribution of temperature and phase state changes tank pressure. The pressure change then feeds back into local saturation conditions and subsequent interface behavior (Kartuzova et al., 2024; Gambioli et al., 2025).
This loop explains why sloshing can be associated with short-term pressure collapse and with long-term BOG amplification. The dominant branch depends on the excitation duration, initial stratification, fill level, and control logic rather than solely on the sloshing intensity (Smith et al., 2022; Jeon et al., 2024; Im et al., 2025). The coupled pathways described above are synthesized in Fig. 8.

5.4 Sloshing-BOR Experimental Evidence

Direct evidence of LH2 sloshing remains rare but highly informative. Takeda et al. (2015) provided one of the earliest transport-oriented simulations of pressure, level, and temperature evolution in a moving 2000 L LH2 tank. Maekawa et al. (2018) then delivered direct marine-transport measurements using MgB2 level sensors synchronized with motion, temperature, and pressure.
Bridge experiments with other cryogens and the K-site hydrogen slosh dataset are also valuable. Ludwig et al. (2013), Himeno et al. (2018), Jeon et al. (2024), and Kartuzova et al. (2024) show that sloshing changes momentum exchange, condensation, destratification, and pressure-collapse dynamics, with different mechanisms dominating under moderate and violent conditions.
The scarcity of direct LH2 motion data also explains why caution is required when interpreting the sloshing literature. Some datasets are dominated by short pressure events, others by cumulative BOG behavior over longer journeys, and the instrumentation sets differ
significantly from one study to another. The practical consequence is that sloshing evidence must be read in terms of the time scale and observability: what looks like disagreement in outcome often reflects the differences in what the experiment was actually able to observe during the event (Maekawa et al., 2018; Ludwig et al., 2013; Jeon et al., 2024).

5.5 Numerical Approaches for Free-Surface Cryogenic Multiphase

Most recent sloshing studies rely on three-dimensional CFD using explicit interface-capturing, dynamic meshes, or moving frames, and phase-change closures. This modeling landscape has now progressed from generic sloshing calculations toward genuinely thermodynamic cryogenic sloshing simulations capable of resolving pressure drop, BOG response, and the role of baffles or structures (Liu & Li, 2018; Chung et al., 2022; Zhu et al., 2023; Lv et al., 2024b).
The central difficulty is the numerical resolution of wave motion, and the faithful coupling of interface kinematics with phase change and pressure evolution. Pressure-collapse prediction remains demanding because wave breaking, mixing, condensation, and transient transport coefficients all change rapidly during a single event (Kartuzova et al., 2024; Gambioli et al., 2025).

5.6 Cross-Sector Motion Environment and Design Implications

The same coupled physics appear across sectors, even though the forcing environments differ. Road applications emphasize braking, cornering, and trailer operation; rail and heavy-duty systems add vibration bands and packaging constraints; maritime systems add persistent roll and pitch spectra and voyage duration; aerospace introduces mass sensitivity and, in some cases, microgravity interface control (Takeda et al., 2015; Smith et al., 2022; Saltari et al., 2024; Wolf et al., 2025).
The main engineering implication is that mobility cannot be engineered by stitching together a quiescent BOR analysis and a separate sloshing-load study. The motion envelope, fill policy, internal structures, thermal architecture, and pressure-control logic should be co-designed whenever the application is motion-sensitive. Otherwise, a system may satisfy static BOR targets but fail under motion via pressure spikes or collapse, or it may pass structural sloshing checks but lack sufficient pressure-control margin and evidence for approval (Jeon et al., 2024; Kartuzova et al., 2024; Wolf et al., 2025). Fig. 9 shows the cross-sector motion frame used in this review. Therefore, the next step is to ask what can actually be measured well enough to validate that co-design.
From a mobility perspective, the key implication is that sloshing does not simply add a mechanical load. It can alter the effective interfacial area and heat/mass transfer pathways that govern BOR and SP. Accordingly, mobility-ready designs should report BOR/SP as quiescent values and as uncertainty-bounded responses under representative acceleration spectra, together with explicit pressurecontrol margins and evidence artifacts, as summarized in Design Note 2 and mapped to the modeling hierarchy (Table 3). These evidence artifacts should also align with the IMO/class/AiP evidence expectations summarized in Table 4.

6. Experimental Methods, Diagnostics, and Benchmark Datasets

6.1 What Can Be Measured and What Cannot

In LH2 storage experiments, pressure, the vented mass or BOG flow, fill level, and distributed temperature are the most accessible observables. They are already sufficient to infer dormancy, pressurization rate, and transfer efficiency. Nevertheless, the quantities that often matter most to model fidelity (local wall heat-flux partition, interfacial heat-transfer coefficients, ullage circulation, and local evaporation-condensation rates) remain much harder to measure directly and are often inferred instead (Hasan et al., 1991; Wang et al., 2024a; Kartuzova & Kassemi, 2025).
That limitation becomes especially important in sloshing or transfer tests, where the same pressure trace can arise from multiple internal mechanisms. Therefore, the central challenge is coherent multiphysics observability rather than simply higher sensor count (Maekawa et al., 2018; Jeon et al., 2024; Kartuzova et al., 2024).

6.2 Facilities: Quiescent Versus Motion-Enabled

The present experimental landscape can be grouped into quiescent SP facilities, chilldown and transfer rigs, motion-enabled facilities, and large-scale or architecture-rich system tests. The K-site lineage remains the best-known benchmark for quiescent pressurization, while the Hartwig chilldown/fill tests provide a modern transient-transfer benchmark. SHIIVER is particularly valuable as a bridge between insulation performance and tank-level thermodynamic behavior at a more realistic scale (Hartwig et al., 2023; Wang et al., 2020b; Johnson, 2024; Johnson et al., 2021a).
Motion-enabled direct LH2 datasets remain rare, which is why Maekawa et al. (2018) and newer K-site slosh-validation studies are so important. They connect the imposed motion, free-surface response, and thermodynamic behavior in ways that most purely static facilities cannot. This infrastructure gap also explains the continued use of bridge fluids and simulated motion environments in much of the literature (Maekawa et al., 2018; Kartuzova et al., 2024).

6.3 Instrumentation and Uncertainty

Pressure transducers, multi-point temperature arrays, level sensors, flow meters, and heat-flux sensors remain the core instrumentation families. Each gives only a partial view. The pressure is clean but highly integrated. The temperature arrays are indispensable for stratification but sensitive to placement, and level sensing becomes challenging under sloshing. Heat-flux sensors are particularly valuable because they help separate nominal insulation performance from actual architecture-level heat pathways (Hartwig et al., 2023; Johnson, 2024).
Recent NASA tank-testing work shows that even when absolute heat-flux uncertainty remains significant, spatial heat-flux trends can reveal architecture-level behavior that would be hard to infer from pressure alone. This reinforces the conclusion that pressure-only validation is insufficient for modern LH2 tank modeling (Johnson, 2024; Johnson et al., 2021b).
Instrumentation itself can also perturb the experiment. Cryogenic-compatible level sensors, external-heating sensor concepts, penetrations for wiring, and local heat-flux gauges may introduce small but non-negligible thermal disturbances or structural complexity. Therefore, the designer of an LH2 experiment faces a recurring trade-off. The more completely one instruments the tank, the greater the risk of altering the very heat-leak pattern one wishes to measure. This is another reason why carefully documented sensor layouts and uncertainty budgets are so important in reusable benchmark datasets (Maekawa et al., 2018; Johnson et al., 2021b).

6.4 Similarity and Scaling for Cryogenic Sloshing-BOR Problems

The similarity in LH2 experiments is more demanding than in purely hydrodynamic sloshing. Froude scaling is a useful starting point, but additional dimensionless groups become relevant once stratification, phase change, and pressure response are involved. Recent scaling analyses explicitly show that SP, relaxation, and lateral sloshing each involve different thermodynamic similarity requirements (Ahizi et al., 2024; Marques et al., 2025).
The implication is direct. A mechanically faithful replacement-fluid experiment may still fail to reproduce real LH2 pressure-collapse or BOG behavior. This is why scaling should be treated as part of the experimental design problem rather than as a post hoc justification for surrogate fluids (Marques et al., 2025).
A practical consequence is that replacement-fluid and reduced-scale studies remain valuable only when the target mechanism is stated clearly. If the objective is free-surface kinematics or impact loading, hydrodynamic similarity may be adequate. If the objective is pressure collapse, cumulative BOG generation, or interface-driven thermal response, then heat transfer and phase change similarity must also be explicitly addressed. Therefore, future benchmark reports should document the geometric scale and excitation amplitude as well as the thermodynamic rationale for the adopted scaling strategy (Ahizi et al., 2024; Marques et al., 2025).

6.5 Recommended Benchmark Cases

Currently, four benchmark families stand out. The first is the K-site/flightweight-tank SP lineage for quiescent pressure-rise validation. The second is the Hartwig chilldown/fill dataset for transient transfer. The third is the SHIIVER family, which brings realistic scale, fill-level evolution, and architecture effects together. The fourth is the emerging class of motion-enabled datasets, including marine transport measurements and validated slosh cases (Hartwig et al., 2023; Johnson et al., 2021a; Johnson, 2024; Maekawa et al., 2018; Kartuzova and Kassemi, 2025).
The immediate community need is not just more experiments but better-structured experiments: simultaneous measurements of pressure, temperature, level, heat flux, and motion with explicit uncertainty budgets and reusable metadata. Without that, the field will accumulate interesting case studies but too few transferable validation anchors (Wang et al., 2024a; Claussner et al., 2025). The benchmark landscape is organized in Fig. 10.

7. Modeling, Simulation, and Digital Engineering Frameworks

If Section 6 asks what can be measured, Section 7 asks what can be predicted credibly and efficiently. The modeling landscape for LH2 tanks is naturally hierarchical: lumped or nodal thermodynamic models for rapid screening, reduced-order models for operational analysis, CFD for local field resolution, and event-scale simulations for sloshing, chill-down, or pressure-collapse transients. Therefore, the engineering task is not to choose the most elaborate model in every case, but to select the model whose assumptions remain appropriate for the question being posed.
At the lowest analytical level, LH2 tanks are often represented by near-equilibrium control volumes linked by an equivalent thermal-resistance network and a small set of state variables. This representation remains valuable for first-pass sizing because it makes explicit how pressure rise depends on the total heat ingress, fill level, and available venting capacity, even before detailed field resolution is attempted.
(4)
dUtankdt=Q˙in+m˙inhin-m˙outhout-m˙vent hvent
In Eq. (4)Utank is the total internal energy of the tank contents (and, if included, the wall thermal mass); Q̇in is the net external heat ingress; ṁin, ṁout, and ṁvent are the inlet, outlet, and vent mass flow rates, respectively; and hin, hout, and hvent are the corresponding specific enthalpies.
In the closed no-vent limit, ṁvent = ṁin = ṁout = 0 and pressure rise emerges from the conversion of external heat ingress into internal energy and phase redistribution. This reduced statement of the problem is analytically transparent, but its limitations become evident once wall thermal inertia, liquid-vapor stratification, and localized heat partition must be represented explicitly.
For practical use, Table 3 summarizes the analytical hierarchy. The current trend is not to replace simple models with CFD indiscriminately, but to layer model classes according to the design question: thermal-resistance screening for first-pass insulation choices, lumped or reduced-order thermodynamics for dormancy and mission
studies, and CHT or Volume-of-Fluid (VOF)-based CFD when local heat-ingress pathways, stratification, chilldown, or sloshing control the answer (Wang et al., 2024b; Claussner et al., 2025; Majumdar et al., 2008; Kartuzova & Kassemi, 2025).
The current shift from conduction-only or prescribed-heat-flux analysis toward CHT does not invalidate the simpler approaches. Rather, it clarifies where their domain of credibility ends and where higher-fidelity treatment becomes necessary.

7.1 Lumped and Thermodynamic Network Models

Lumped and multi-node thermodynamic models remain indispensable because they capture overall state evolution at very low computational cost. They are well suited to dormancy estimation, fill-ratio studies, and rapid parametric screening when the design question concerns pressure rise or overall BOR rather than local fields (Al Ghafri et al., 2022; Wang et al., 2024b).
These models are not crude substitutes for CFD. When used transparently, they are physics-informed approximations whose value lies in speed and interpretability. Their main vulnerability is the need for closure assumptions regarding wall heat partition, interfacial transfer, and stratification thickness (Wang et al., 2024b; Nuwantha et al., 2024). In mobility contexts, the most fragile closures are often those that linearize the ullage temperature profiles or assume fill-independent convective coefficients inferred from quiescent data (Sections 4.1.4 and 5.2).

7.2 Reduced-Order Models for Mission and Operation Analysis

Reduced-order models extend lumped formulations into an operational context. They can represent filling, extraction, route-dependent operations, pressure-build-up logic, and repeated cycles fast enough for optimization and control studies. This is especially important for trailer distribution, aircraft refueling, and marine transfer, where repeated scenario evaluation matters more than one high-fidelity simulation (Adler & Martins, 2025; Appel et al., 2025; Wolf et al., 2025).
Because the storage tank is increasingly treated as an operating system rather than a static vessel, reduced-order models are likely to grow in importance. Their credibility, however, still depends on validation-grade benchmark data and careful treatment of uncertainty (Marques et al., 2024; Claussner et al., 2025).

7.3 CFD for Conjugate Heat Transfer and Insulation Coupling

Three-dimensional CFD becomes essential once the design question depends on local physics, such as conjugate wall-to-fluid heat transfer, multi-zone stratification, large-amplitude sloshing, wall wetting, internal structures, or event-scale pressure collapse. Its role is not simply to be more detailed than a lumped model, but to expose mechanisms that cannot be represented credibly with globally averaged coefficients alone.
A particularly important trend is the shift from prescribed wall-heat-flux or conduction-only analyses toward conjugate heat transfer (CHT), where solid-region conduction and fluid-region energy transport are solved as a coupled problem. This transition reflects the fact that supports, penetrations, local wetting, and fill-dependent contact areas alter the partitioning of heat between the liquid and the ullage.
In CHT-based tank simulations, the coupled energy equations in the solid (subscript s) and fluid (subscript f) regions can be expressed as
(5)
ρscp,s∂Ts∂t=∇·(ks∇Ts),ρfcp,f(∂ Tf∂ t+u·∇Tf)=∇·(kf∇Tf)+Spc
where ρ is the density (kg · m−3); cp is the specific heat (J · kg−1 · K−1); Tis the temperature (K); k is the thermal conductivity (W · m−1 · K−1); u is the fluid velocity vector (m · s−1), and Spc is a volumetric source term representing latent-heat exchange due to a phase change. The subscripts s and f denote solid and fluid domains, respectively.
In this CHT view, the inner wall is no longer a passive boundary with prescribed heat flux. Instead, it becomes an active component of the solution, and the computational emphasis shifts from external heat-leak imposition to coupled wall-fluid-interface energy redistribution.
This transition matters because the wall thickness, supports, penetrations, local wetting, and fill-level-dependent contact areas influence the partitioning of heat between the liquid and ullage. Recent LH2 modeling studies and reviews increasingly treat this wall-fluid coupling as an integral part of the problem rather than as a fixed boundary condition (Claussner et al., 2025; Pesich et al., 2024; Kartuzova and Kassemi, 2025).
Recent work has increasingly combined CFD with outer thermal models or reduced-order elements rather than treating it as a stand-alone silo. This trend is likely to continue because the most useful future tools will be hierarchical rather than monolithic (Qu et al., 2025; Kartuzova and Kassemi, 2025).

7.4 Phase-Change Modeling: Evaporation and Condensation Closures

The most difficult issue in an LH2 tank simulation is not merely free-surface tracking, but a representation of interfacial evaporation and condensation across quiescent storage, venting, sloshing, and transfer. Different closures can match the same pressure history while implying different internal mechanisms, which makes interfacial modeling the central structural uncertainty in many simulations (Kartuzova et al., 2024; Gambioli et al., 2025).
A closure suitable for long-duration SP may perform poorly in chilldown or violent sloshing, and vice versa. This is why model selection must remain tied to regime, validation evidence, and intended use rather than to an abstract hierarchy of complexity (Wang et al., 2024b; Nuwantha et al., 2024).

7.5 Verification, Validation, and Uncertainty Quantification

For LH2 storage models, validation remains more limiting than formulation. Recent K-site work is a good example. Modern VOF-based CFD has been validated against classical SP data and then extended to new fill levels, showing how benchmark reuse can support disciplined model extension rather than one-off curve fitting (Kartuzova & Kassemi, 2025).
Nevertheless, validation should not be reduced to pressure matching. Structural uncertainty in heat-leak partition, interfacial closure, and sensor interpretation often dominates numerical error. Recent trailer-model validation and data-assimilation work make this especially clear by showing that the measurement uncertainty can be a major contributor to apparent model mismatch (Wolf et al., 2025; Marques et al., 2024).

7.6 AI-assisted Surrogates and Data-Driven Closures

AI-assisted surrogates and data-assimilation methods are now sufficiently plausible to be useful, provided they remain anchored to physics. Marques et al. (2024) demonstrated real-time assimilation of a 0D cryogenic storage model using streaming data, explicitly targeting heat-leak and sloshing-enhanced losses. This is promising because it moves the field toward adaptive calibration rather than static parameter fitting.
The most credible near-term role for AI is not replacing thermo-fluid models but augmenting them: surrogate closures, emulator-assisted optimization, uncertainty reduction, and digital-twin support for route planning or long-duration storage. In LH2 storage, data remain too sparse and too regime-dependent for purely data-driven modeling to replace mechanistic reasoning (Marques et al., 2024; Claussner et al., 2025).
From a practical engineering viewpoint, this is the point at which digital transformation (DX) and AI-enabled augmentation (AX) become operationally relevant for LH2 tanks. In this context, DX refers to the systematic digitization of the tank state, route history, maintenance records, and operational decisions, while AX denotes the use of AI-assisted tools to support interpretation, prediction, and control within that digital framework. Fig. 11 summarizes the corresponding model ladder and validation logic.

8. Safety, Standards, and Engineering Constraints

8.1 Hazard Pathways Specific to LH2 storage

The primary hazard pathways in LH2 storage arise from the combination of cryogenic temperature, flammability, and pressure accumulation. Heat ingress can produce SP and vent demand; leaks can create both cryogenic exposure and flammable-gas risk; and transfer upsets can challenge pressure control and ventilation. These hazards are tightly linked rather than independent, which is why recent transfer-safety reviews treat storage, transfer, and vent handling as a single hazard chain (Schiaroli et al., 2025). The presence of such coupled hazard pathways also suggests that the safety behavior of LH2 storage systems is closely linked to the underlying thermodynamic processes. Variations in heat ingress, insulation performance, or pressure management can affect the rate of boil-off generation and, consequently, the pressure rise within the tank. In situations where the pressure increase exceeds the intended operating range, venting or pressure-relief activation may occur, which in turn may lead to hydrogen release (Molkov, 2012). In this respect, many safety issues in LH2 storage systems originate from the same heat-ingress and boil-off mechanisms discussed in the preceding sections.
From an accident perspective, several types of initiating events may result in hydrogen release from LH2 storage systems. Typical examples include loss of containment from tanks, piping, or valves, abnormal pressure development associated with excessive boil-off, and disturbances during storage or transfer operations (ISO, 2026). Although the initiating causes differ, these events often evolve through similar thermodynamic pathways involving vapor generation and pressure rise within the storage system. When liquid hydrogen is released, rapid flashing and evaporation can initially produce a cold hydrogen cloud before the gas warms and becomes buoyant. Experimental studies have shown that this early stage may involve complex dispersion behavior, which can affect the ignition conditions and the spatial extent of flammable mixtures (Houf & Schefer, 2007). As mixing with the surrounding air progresses, buoyancy generally becomes dominant, and the hydrogen plume rises rapidly.
Overall, these aspects indicate that LH2 storage safety depends on the containment integrity and on how effectively heat ingress, pressure evolution, and hydrogen release are managed at the system level. These considerations form an important basis for the engineering constraints and system-level aspects discussed in the following sections.

8.2 Structural Integrity Under Cryogenic and Motion Loads

Structural integrity is governed by more than static pressure design. Low temperature introduces contraction, seal challenges, and possible embrittlement; motion or vibration adds dynamic loading, particularly in mobile and marine applications. Kim and Chun (2023) clearly showed this in the marine context, using LNG-derived material requirements as a baseline for LH2 cargo and fuel containment. The practical lesson is that thermal efficiency and structural robustness cannot be decoupled cleanly in real tanks.
This structural problem is now broadening beyond the traditional stainless-steel baseline. As discussed earlier, 304L and 316L remain the most established inner-shell materials. Among the emerging metallic alternatives, current research and guidance place high-strength 15% Ni steel closer to near-term LH2 application because direct 20 K toughness and weld-joint data are already available, whereas high-manganese-related candidates are treated more cautiously because their LH2-specific hydrogen compatibility and weld qualification are still being resolved (ABS, 2025a; Namegawa et al., 2025; Cho et al., 2024; Lee et al., 2024; Korean Register, 2024). The same issue extends to supports and attachments. A low-conductivity G11 support may improve thermal efficiency, but the associated metal-composite or metal–glass-fiber-reinforced polymer (metal-GFRP) interface can become a structural hot spot under thermal contraction and motion loading. Thus, structural integrity for LH2 tanks is inseparable from joining technology, support architecture, and cryogenic material compatibility.
A second structural issue is the connection between the tank walls and their supporting members. Even if stainless steel remains the baseline material for the primary shells, the most thermally efficient support concepts may use G10/G11-type composites or other low-conductivity inserts. In that case, the engineering bottleneck often shifts from the bulk material strength to the reliability of the metal-composite interface. Adhesive performance, local stress concentration, thermal contraction mismatch, and long-term joint integrity under repeated cryogenic cycles become just as important as the support material itself. This is why cryogenic joint qualification should be considered part of a structural integrity assessment rather than as a secondary manufacturing issue (Glass & Smith, 1997; Kang et al., 2007; ABS, 2025a).

8.3 System Integration: Valves, Sensors, and Pressure-Management Hardware

Pressure-relief valves (PRVs), pressure-build-up units (PBUs), vent stacks, transfer lines, sensors, and control logic all shape how residual heat ingress appears in service. Two tanks with similar insulation and geometry can behave differently if their pressure-management philosophies differ. Trailer and transfer studies repeatedly identified auxiliary hardware–especially the PBU–as a dominant factor in operational response and model fitting (Wolf et al., 2025; Schiaroli et al., 2025).
Thus, venting and instrumentation should not be seen as appendages added after the tank is designed. They are part of the storage response itself (Ha et al., 2024; Lee et al., 2025). These components operate as part of an integrated monitoring and pressure management architecture rather than as independent devices. The sensors provide continuous information on pressure and temperature conditions within the storage system, while the valves and related hardware regulate venting and pressure control functions (Davies et al., 2024). Therefore, safe operation depends on the coordinated interaction between sensing, control, and actuation functions within the overall system.
Because these functions are closely coupled, certain hazards may arise from failures of individual components and from interactions within the control structure itself. For example, delayed sensor response, inappropriate control actions, or unintended valve operations during abnormal pressure evolution may lead to undesired system behavior. Such situations show that potential hazards can arise from the way monitoring and control functions interact within the system. Traditional component-oriented approaches, such as Failure Modes, Effects, and Criticality Analysis (FMECA), are used widely to analyze hardware failures in engineering systems. On the other hand, these methods may have limitations in capturing hazards arising from complex interactions between sensing, control, and actuation functions. For systems with significant control interactions, complementary system-oriented approaches such as Systems Theoretic Process Analysis (STPA) have been proposed to support the identification of interaction-related hazards (Leveson, 2011).
Considering these aspects can be useful when designing the monitoring and pressure management architecture of LH2 storage systems, particularly during operations such as pressure regulation, bunkering, and hydrogen transfer, where coordinated system response is required to maintain safe operating conditions.

8.4 Codes, Standards, and Classification Perspectives

The standards landscape for LH2 storage is developing rapidly but remains application-specific. The International Organization for Standardization (ISO) 13985:2006 specifies construction and testing requirements for refillable liquid-hydrogen fuel tanks attached to land vehicles, while ISO 11326:2024 specifies inspection and test procedures for liquid-hydrogen cargo tanks on hydrogen ships (ISO, 2006; 2024). For vacuum-insulated aboveground storage in the U.S. context, the Compressed Gas Association (CGA) H-3 provides minimum design and performance requirements for cryogenic LH2 tanks (CGA H-3, 2024).
In maritime carriage, IMO Resolution Maritime Safety Committee (MSC) Resolution MSC.565(108) provides revised interim recommendations for the carriage of liquefied hydrogen in bulk. These recommendations support project approvals under the International Code for the Construction and Equipment of Ships Carrying Liquefied Gases in Bulk (IGC Code) framework via the tripartite agreement mechanism while prescriptive provisions continue to mature (IMO, 2024; Maritime Department, Hong Kong SAR Government, 2024). The IMO Sub-Committee on Carriage of Cargoes and Containers (CCC) 10 (September 2024) explicitly noted that the revised recommendations did not cover membrane-type cargo containment systems and anticipated the development of a dedicated Part D for vacuum membrane concepts (Lloyd’s Register, 2024; IMO, 2024). CCC 11 (September 2025) subsequently completed draft revisions to MSC.565(108), adding Part D for membrane-type systems that maintain insulation spaces under vacuum, and finalizing the interim guidelines for the use of hydrogen as fuel (Lloyd’s Register, 2025; DNV, 2025c).
Beyond safety compliance, mobility-era boil-off is increasingly coupled to decarbonization economics. The 2023 IMO Green House Gas (GHG) Strategy and the IMO Net-Zero Framework (approved at MEPC 83) introduce a lifecycle fuel-intensity standard and a global economic measure in which ships exceeding thresholds must acquire remedial units, while ships using zero or near-zero GHG fuels may be eligible for financial rewards (IMO, 2023; 2025). In this context, any filling-ratio-dependent BOR amplification under motion—which manifests as additional boil-off handling energy, vent losses, or extra propulsion energy to compensate for lost stored energy—directly penalizes voyage efficiency and can translate into operating-cost burdens for ship operators and cargo owners.
Complementing IMO guidance, classification societies have accelerated technology screening through dedicated guidance notes and multiple approvals-in-principle (AiPs). For example, Det Norske Veritas (DNV) has issued AiPs for several 80,000 m3-class LH2 carrier concepts featuring large, vacuum-insulated cargo tanks and hydrogen-compatible propulsion options (DNV, 2024; DNV, 2025a). Related carrier-scale AiPs have also been granted for alternative cargo-containment developments led by major shipbuilding and engineering partners, including an 80,000 m3 LH2 carrier design and cargo containment system jointly developed by Hanwha Ocean and CB&I (CB&I, 2024). In parallel, a Bureau Veritas-reviewed joint development project with Gaztransport & Technigaz (GTT) and TotalEnergies obtained AiPs for a cryogenic membrane containment system and a preliminary 150,000 m3 large-scale LH2 carrier concept, highlighting a second architectural branch in which a vacuum is maintained in the membrane-tank insulation spaces (GTT, 2024; Bureau Veritas, 2024). These initiatives do not replace statutory approval, but they are shaping the near-term engineering definition of ‘practical’ LH2 tanks by bringing the BOR performance, vacuum integrity, and motion/operability considerations into early-stage hazard and feasibility reviews. Table 4 lists selected milestones and their typical evidence expectations, and Section 8.4.1 distills these into an engineering checklist suited to mobility-era tanks.

8.4.1 Evidence packages for IMO/Class/AiP decisions in mobility-era LH2 tanks

For emerging LH2 concepts, regulatory texts and class approvals in principle effectively translate into an evidence package: a structured set of analyses, tests, and safety arguments that link the mission envelope to credible performance and hazard margins. That package must explicitly address motion-coupled thermo-fluid dynamics (sloshing–BOR–SP coupling, i.e., sloshing-boil-off-SP coupling) in mobility-dominant applications, rather than relying on quiescent BOR alone. Section 9.1.5 summarizes the representative mini-cases that exemplify typical evidence patterns and remaining gaps.
  • (1) Mission and mobility envelope: route–voyage duration, motion spectra or acceleration environment, vibration, fill policy, allowable pressure window, and permissible venting philosophy.

  • (2) Heat-ingress budget and thermal-resistance screening: insulation-family selection, vacuum-level envelope, radiation/contact sensitivities, and an explicit thermal-bridge inventory for supports and penetrations.

  • (3) Quiescent BOR/SP credibility: near-equilibrium assumptions stated, lumped thermodynamic model anchored to E1/E2 evidence, uncertainty bounds reported.

  • (4) Operational transients: chill-down, filling/withdrawal, pressure-build-up unit (PBU) operation, and control logic verified against representative sequences.

  • (5) Motion-coupled phenomena: sloshing regime map versus filling ratio; coupling pathway (motion → mixing/interface renewal → evaporation → pressure); and at least one validation-grade dataset or benchmark targeted for verification and validation (V&V).

  • (6) Structural integrity and joining: shell/support/joint design validated under cryogenic contraction, thermal cycling, and fatigue; leak tightness and compatibility evidence for dissimilar materials (e.g., SUS-to-G10/G11 interfaces) documented.

  • (7) Safety case and consequence analysis: hazard identification (HAZID)/environmental hazard identification (ENVID) framing, failure modes and effects analysis (FMEA)/fault tree analysis (FTA) as appropriate, vent/dispersion philosophy, ignition controls, and emergency response for vacuum-space leakage scenarios.

  • (8) Instrumentation and monitoring: vacuum integrity monitoring, pressure/temperature mapping, level sensing, acceleration logging, and data management sufficient for operational diagnostics and DX readiness.

  • (9) Verification, validation, and uncertainty quantification: model hierarchy chosen for decisions, numerical verification, validation plan, and uncertainty budgets for key performance metrics (BOR, pressure rise, and vented mass).

  • (10) Maintainability and inspection: vacuum re-pump strategy, degradation detection, access and replacement concepts for supports/penetrations, and inspection intervals consistent with the approval basis.

Table 5 presents a mobility-ready design envelope that links fill ratio, motion severity, and pressure-control margin to the minimum analysis fidelity and evidence artifacts (Design Note 2).
In approval-package terms (Table 5), the ‘Low fill + low motion’ row is linked to the mission envelope, allowable pressure window, quiescent boil-off/self-pressurization credibility, and baseline instrumentation items. The ‘Intermediate fill + low motion’ row is linked to repeatable boundary-condition definition, stratification-sensitive BOR/SP assessment, and uncertainty-aware model anchoring. The ‘Intermediate fill + mobility motion’ row is linked directly to motion/sloshing operability assessment, acceleration logging, vent/PBU capacity margin, and validation of motion-coupled thermo-fluid behavior. The ‘High fill + mobility motion’ row is linked to thermal-bridge inventory, structural/inspection considerations, and vacuum-monitoring requirements. The ‘Filling/withdrawal transients under motion’ row is linked to operational-transient verification, control-logic confirmation, flow/valve-state logging, and mass/energy-loss accounting.

Design Note 3

What goes wrong if you optimize only BOR or only sloshing?
Mobility transforms LH2 tank performance from a single dormancy metric into a coupled response. A quiescent BOR-only design may satisfy steady heat-leak targets yet still experience motion-triggered pressure spikes/collapses because sloshing renews the interface and disrupts stratification; the result can be vent/PBU saturation, unexpected PRV lift, and unplanned loss. Conversely, a sloshing-only structural campaign may size impact loads correctly but miss thermodynamic pressure excursions and the distribution of operational losses, leading to an under-designed pressure-control margin and incomplete approval evidence. Therefore, mobility-ready design requires coupled metrics and uncertainty-bounded responses rather than stitched single-phenomenon assessments (Jeon et al., 2024; Wolf et al., 2025; Kartuzova et al., 2024).
Minimum coupled checklist (screening):
  • (1) Fill-ratio sweep across the mission (including intermediate regimes) under representative motion spectra.

  • (2) Coupled pressure-response metrics: ΔPpeak/Δt, dP/dt, and vent/PBU flow versus capacity (margin).

  • (3) Thermal state observables: multi-point liquid/ullage temperatures, stratification indicators, and heat-leak partition (bridges).

  • (4) Free-surface/mixing proxies: slosh intensity measures, interface renewal indicators, and mixing events.

Control and operability logs: vent/PBU/valve states, set-points, and uncertainty-bounded mass/energy loss accounting.
One practical consequence of this evolving standards landscape is that material innovation and code maturity do not necessarily progress at the same speed. Even when a candidate such as 15% Ni steel shows encouraging LH2-specific structural evidence, it remains less code-mature than the conventional 304L/316L baseline. The maturity gap is currently larger for high-manganese-related systems, whose LH2-specific qualification routes, welding practices, inspection criteria, and service evidence are still less established. For designers, this means that code readiness is itself a design variable: a technically attractive material may still carry project risk if its qualification pathway is immature (ABS, 2025a; Korean Register, 2024; Namegawa et al., 2025).
From a review perspective, standards should also be read as indicators of technology maturity. Topics that are already codified usually correspond to design assumptions and test procedures that have stabilized, whereas topics still governed by interim recommendations or project-specific approval often signal unresolved issues in validation, material qualification, or operating practice. This is particularly relevant for mobile and marine LH2 systems, where code evolution is occurring alongside hardware deployment (ISO, 2024; IMO, 2024; Li et al., 2024a; Schiaroli et al., 2025).

8.5 Design Trade-Offs

At the system level, LH2 tank design is a constrained trade-off among BOR, pressure control, mass, volume, inspectability, safety, and cost. Lower heat leak often demands more sophisticated insulation and vacuum management, safer and more maintainable systems often introduce hardware that increases heat ingress, and aggressive boil-off suppression can increase control complexity or power demand. Therefore, the best design is not the one with the lowest theoretical heat leak, but the one that meets mission requirements under realistic operating and compliance constraints (Kim & Chun, 2023; ISO, 2024; Schiaroli et al., 2025).
The trade-off perspective becomes even more important when emerging materials or composite support concepts are introduced. A change that lowers the structural mass or conductive heat leak may at the same time complicate qualification, joint design, repairability, or inspection access. For that reason, a design comparison should include mass and BOR, as well as qualification pathway, joining complexity, maintainability, and code readiness (ABS, 2025a; Kim & Chun, 2023).

9. Cross-Sector Lessons, Research Gaps, and Future Directions

9.1 Cross-Sector Lessons and Maritime Implications

9.1.1 Aerospace heritage and what transfers well

Aerospace heritage contributes three durable lessons: benchmark culture, extreme mass sensitivity, and the need to validate thermal management under demanding mission constraints. NASA SP, ZBO, and sloshing work remain central to the field because they prioritize predictive credibility rather than approximate storage heuristics (Kassemi et al., 2018; Johnson et al., 2021a; Tiwari et al., 2024).
That heritage is also embodied in actual hardware lineages. The Centaur upper stage established long-lived stainless-steel LH2 cryotank practice in flight systems, while NASA cryogenic test programs later translated that heritage into benchmark datasets for SP and thermal management validation. In the current aviation context, Airbus has moved from generic concept studies toward a liquid-hydrogen BreadBoard and announced integrated ground testing of the hydrogen distribution system and propulsive bench for 2027. These examples matter because they show a real progression from launch-vehicle heritage, to benchmark science, to transport-oriented cryotank development (Dawson & Bowles, 2004; Airbus, 2025).
The exact gravity and mission environment are less directly transferable, but the methodological rigor of aerospace benchmarking transfers very well. For LH2 tanks outside spaceflight, that rigor may be as valuable as any single hardware concept.
Recent NASA aviation-oriented cryogenic hydrogen studies sharpen this lesson further by stating the design requirements for future aircraft tanks more explicitly: durability over thousands of pressure and thermal cycles, crashworthiness, maintainability and inspectability, rapid turn-around refueling, and lightweight tanks with high thermal-performance insulation are all treated as core requirements rather than secondary aspirations (Lvovich et al., 2025). This means that aerospace and UAM-oriented LH2 storage is pushing the field toward material systems that are cryogenically robust, yet lightweight, certifiable, and operationally repeatable.

9.1.2 Land-based, vehicular, and aviation/UAM experience

Land-based and vehicular LH2 systems teach a different lesson: operation matters as much as static design. Trailer distribution, refueling, and route-dependent mobile service expose the importance of pressure-build-up units, partial-fill operation, repeated transients, and practical transfer losses. Recent route-based trailer validation makes this point especially clearly, because it links model performance to real distribution data rather than idealized cases (Wolf et al., 2025).
Vehicle and trailer work also shows that internal structures can materially change outcomes. Simulated road transport studies report a strong sensitivity of BOG generation to slosh-damping structures and motion history, which reinforces the view that mobile LH2 storage is a route- and structure-dependent thermodynamic problem (Im et al., 2025).
This same logic is now extending into lightweight aviation and UAM contexts, where the tank mass directly constrains the payload and range. Recent hydrogen-tank materials work has examined recyclable liquid thermoplastic and graphene-modified carbon-fiber composite systems, and polymer-composite characterization studies identify Elium 591 and related resin systems as filament-windable candidates for hydrogen tanks with potential end-of-life recyclability (Gul et al., 2023; Andrew et al., 2025). At the industrial-demonstration level, Australasian teams led by AMSL Aero, Stralis Aircraft, and Fabrum have already moved to registered Dewar vessels and filled lightweight composite LH2 aviation tanks for pre-flight programs, suggesting that lightweight mobile LH2 storage is no longer only a conceptual topic for aircraft and eVTOL platforms (Fabrum, 2025; IP Group, 2025).
The relevance of this trend is no longer purely conceptual. Alongside peer-reviewed work on lightweight cryogenic composites, recent Australasian demonstration activity has begun to move liquid-hydrogen aviation tanks from design studies toward registered hardware and refueling demonstrations. In particular, AMSL Aero has reported receiving a registered liquid-hydrogen tank for zero-emission test-flight preparation, while Fabrum, AMSL Aero, and Stralis have shown airport-based filling of lightweight composite LH2 aviation tanks. These milestones do not yet replace the need for long-duration durability data or certification evidence, but they do show that lightweight LH2 tanks for regional aircraft or UAM-like mobility are becoming an engineering deployment question rather than only a speculative research topic (Fabrum, 2025; IP Group, 2025).
A useful way to read these aviation and UAM developments is to separate design evidence from deployment evidence. Peer-reviewed composite-tank studies and materials papers have addressed cryogenic cycling, permeability, and lightweight architectures. Recent Australasian demonstrations by Fabrum, AMSL Aero, and Stralis, however, have shown that lightweight composite LH2 tanks can now be filled in an airport-based liquid-hydrogen test facility prior to flight testing. That does not amount to certification-grade durability evidence, but it is a meaningful transition from laboratory materials development to a system-level demonstration in a mobility context (Andrew et al., 2025; Fabrum, 2025).

9.1.3 Marine-specific gaps

Marine LH2 storage differs mainly by scale, voyage duration, and exposure to persistent motion. Carrier-oriented modeling indicates that for LNG-like tank volumes and insulation thicknesses, an LH2 carrier stores much less fuel energy while being more sensitive to boil-off and sloshing. Voyage-scale measurements now begin to confirm these system-level concerns (Smith et al., 2022; Inomata et al., 2025).
This sector is now supported by a clearer case history than is sometimes recognized. The HESC pilot project completed the first Australia-Japan maritime transport chain for liquefied hydrogen in 2022 using the 1250 m3 SUISO FRONTIER, while Kawasaki later reported pilot-test BOR values of approximately 0.3%/day for the ship tank and approximately 0.06%/day for the Hy touch Kobe onshore storage tank. Inomata et al. (2025) then added an academic voyage-scale comparison of measured and calculated heat ingress for the same carrier class, which is exactly the type of evidence required to move marine LH2 design beyond extrapolation from small tanks (HESC, 2025; Kawasaki Heavy Industries, 2023; Inomata et al., 2025).
The most important marine-specific gaps are long-duration voyage data, cargo-scale motion-coupled BOR validation, and still-evolving regulatory frameworks. These gaps matter because they determine if the conclusions drawn from small mobile tanks or aerospace tanks remain credible after scale-up (ISO, 2024; IMO, 2024; Inomata et al., 2025).
Beyond pilot voyages, the marine sector is also generating a dense pre-commercial evidence trail through approvals in principle (AiPs), joint development projects, and yard-scale demonstrations. Within a two-year window, DNV has issued AiPs for multiple 80,000 m3-class LH2 carrier concepts from Korean shipbuilders and partners, while Bureau Veritas reviewed a 150,000 m3-class membrane-tank concept involving GTT and TotalEnergies. Together, these activities suggest that the bottleneck is shifting from “can a cryogenic tank be built?” toward “can vacuum integrity, BOR, and sloshing-aware operability be validated at the carrier scale under class and IMO scrutiny?” (DNV, 2024; DNV, 2025a; DNV, 2025b; Bureau Veritas, 2024; GTT, 2024).

9.1.4 A unified design workflow for marine LH2 tanks

For marine applications, the most useful design workflow begins with the mission envelope–voyage duration, motion spectrum, allowable pressure window, fill and unloading policy–and then proceeds iteratively through geometry, insulation, BOR/SP prediction, unloading architecture, and safety/control design. Recent maritime modeling supports this loop rather than a linear design sequence (Wang et al., 2024d; Wang et al., 2024e; Lee et al., 2025).
This workflow is already beginning to acquire a commercialization trajectory. In early 2026, Kawasaki announced the construction of a 40000 m3-class liquefied-hydrogen carrier as the next step beyond the SUISO FRONTIER pilot scale, making explicit the transition of the field from proof-of-concept transport to commercial-scale design. That transition underscores the importance of thermal-architecture credibility, unloading logic, support qualification, and voyage-scale validation rather than reducing it (Kawasaki Heavy Industries, 2026).
The same workflow benefits from cross-sector borrowing. Aerospace contributes benchmark discipline; land transport contributes operational realism, and the maritime application provides the decisive test of scale and persistence. In that sense, cross-sector learning is useful not because all applications are identical, but because each reveals a different part of the same coupled design loop (Tiwari et al., 2024; Wolf et al., 2025; Inomata et al., 2025).
In practical terms, this workflow is also where digital engineering becomes operationally consequential. Once the mission envelope, fill policy, allowable pressure window, and unloading logic are represented explicitly, data from operations and validation can be fed back into design updates, maintenance planning, and control refinement in a disciplined manner. Fig. 12 presents the resulting closed-design loop.

9.1.5 Mini-case portfolio across sectors: evidence, gaps, and approval relevance

This section summarizes four representative mini-cases that span direct operational evidence, validation-grade experiments, and approval-oriented concept maturation to reduce the risk of a “broad but shallow” impression. Each mini-case is labeled by evidence class (E1–E4; Section 1.5) and highlights what is demonstrated, what remains uncertain, and which design or approval decisions are most affected.

Mini-case M1

Marine/shipbuilding scale-up anchored by pilot operations (E1 [supplementary context: E3 scale-up/AiP trajectory])
Context: The HESC pilot chain and the 1,250 m3-class liquefied-hydrogen carrier provide rare voyage-scale data linking heat ingress, pressure management, and operability under marine motion.
Evidence: Voyage measurements and measured-versus-calculated heating ingress comparison (E1) combined with an active scale-up trajectory supported by class/AiP and commercial planning (E3).
What it demonstrates: (i) tank-level heat ingress can be bounded and back-calculated at the voyage scale; (ii) vacuum/insulation integrity and thermal-bridge management remain decisive even when the nominal insulation performance is high; (iii) operability margins depend on how motion-driven mixing interacts with pressure control.
What is still missing: public, validation-grade datasets that co-measure acceleration spectra, internal temperature fields, interface dynamics, and vent/control actions during real voyages; and benchmark cases for membrane-type/vacuum systems that align with IMO CCC 11 Part D discussions.
Approval bottleneck: showing the motion-coupled pressure response and vacuum/thermal-bridge integrity across a voyage envelope in a form acceptable to IMO/Class and transferable to next-scale carrier concepts.
Why it matters for mobility-era design: the “effective BOR” becomes route- and motion-dependent; approval-ready designs require evidence packages that connect quiescent BOR targets to motion-coupled operability margins (see Table 4 and Section 8.4.1). (HESC, 2025; Inomata et al., 2025; Kawasaki Heavy Industries, 2026; Lloyd’s Register, 2025).
Minimum evidence artifacts (aligning with Table 4): motion input (acceleration/sea-state proxy); multi-point T fields + level; pressure/vent mass-flow with control states; heat-ingress/BOR mass–energy balance with uncertainty.
Evidence artifacts (typical): voyage motion/acceleration spectra; pressure–temperature–level time histories; heat-ingress budget with uncertainty; vent/pressure-control logic; vacuum/support inspection logs.
Design envelope mapping: Design Note 2, rows 1 and 3.

Mini-case R1

Road-mobile LH2 distribution–route dependence and control dominance (E1)
Context: Mobile distribution trailers operate under repeated fill/withdrawal cycles with route-specific transients and pressure-build-up unit (PBU) interactions.
Evidence: Fleet-route validation against 13 full distribution routes provides direct operational evidence (E1).
What it demonstrates: (i) mission dependence is first-order–route history and control logic can dominate the BOR outcomes; (ii) PBU sizing/control and transfer scheduling are as important as a static heat leak; (iii) instrumentation quality limits model credibility, motivating DX-ready sensing for fleet operation.
What is still missing: open datasets that include synchronized motion input, internal thermal stratification, and interface dynamics under road excitation; and systematic validation of sloshing–BOR coupling mitigation devices (e.g., multi-orifice baffles) under LH2 conditions.
Approval bottleneck: auditable proof that route- and control-dependent pressure margins (vent/PBU/PRV) remain acceptable across the fleet distribution envelope, with uncertainty-bounded loss accounting.
Why it matters for mobility-era design: road data provide the clearest current example of “BOR as an operational distribution,” not a single number, motivating route-aware models and evidence packages for approval and procurement. (Wolf et al., 2025; Schiaroli et al., 2025; Im et al., 2025).
Minimum evidence artifacts (aligning with Table 4): GPS/route + acceleration/vibration; pressure + PBU/valve states; multi-point temperature/level; venting/mass-loss accounting and ambient conditions.
Evidence artifacts (typical): route-resolved fill/withdrawal logs; PBU/vent state histories; pressure–temperature–level time series; acceleration/braking statistics; loss accounting (mass/energy balance).
Design envelope mapping: Design Note 2 rows 3 and 5.

Mini-case A1

Aviation/UAM–lightweight architecture and certification evidence gap (E3 [limited supporting E2 facility-level evidence])
Context: Aviation and UAM demand aggressive mass minimization, rapid turnaround, and high-cycle thermal durability, making shell/support/joining choices and insulation integration central design drivers.
Evidence: Technology programs and ground demonstrations provide concept maturation evidence (E3), while validation-grade, public LH2 durability datasets remain limited (scarce E2).
What it demonstrates: (i) lightweight containment concepts (including composites and hybrid architectures) are being tested; (ii) refueling demonstrations highlight practical constraints on chill-down, venting policy, and operability; (iii) materials selection is tightly linked to certification evidence planning.
What is still missing: long-life cryogenic fatigue/thermal-cycle qualification for lightweight shells and low-k supports; systematic evidence on bonding/joining reliability (e.g., stainless-to-G10/G11 interfaces) under realistic cycling; and public benchmarks for rapid no-vent filling with motion disturbances.
Approval bottleneck: certification-grade durability and leak/vacuum-integrity evidence for lightweight shells, supports, and joints (including dissimilar-material interfaces) under high-cycle thermal and operational transients.
Why it matters for mobility-era design: the bottleneck is increasingly “evidence for certification” rather than the feasibility of a single component; approval-ready designs require integrated evidence packages spanning materials, supports, insulation, and control. (Tiwari et al., 2024; Airbus, 2025; Boddorff, 2025; Fabrum, 2025).
Minimum evidence artifacts (aligning with Table 4): thermal-cycle and cryo-fatigue durability; vacuum hold/leak-rate metrics; joining/bonding reliability data (e.g., SUS–G10/G11); pressure-control performance under representative maneuvers.
Evidence artifacts (typical): cyclic thermal/vacuum durability tests; joint/support qualification data (metal–composite interfaces); leak and vacuum-integrity metrics; ground demonstration logs; uncertainty-bounded performance margins.
Design envelope mapping: Design Note 2 rows 4 and 5.

Mini-case S1

Space heritage–benchmark culture and validation ladder (E2 [supplementary context: E1 heritage])
Context: Space-propellant cryotanks established benchmark culture for SP, vent/pressure control, and interface-driven transients, with datasets that remain foundational for validation.
Evidence: Validation-grade experiments and benchmark programs (E2), including K-site SP and SHIIVER-type architecture-rich testing, support disciplined V&V and model hierarchy development.
What it demonstrates: (i) pressure evolution depends on internal stratification and interfacial transport, not on the total heat leak alone; (ii) benchmarking enables controlled extension across fill ratios and heat-flux conditions; (iii) heat-flux instrumentation helps separate insulation performance from architecture-level bridges.
What is still missing for mobility transfer: direct mapping from space-centric scenarios to terrestrial mobility envelopes; specifically, combined motion–thermo datasets at 1 g with realistic acceleration spectra and modern instrumentation.
Approval bottleneck (for mobility transfer): closing the validation ladder with motion-coupled E2/E1 datasets at 1 g, so that approvals do not depend solely on conservative margins or surrogate extrapolation.
Why it matters for mobility-era design: space heritage provides the validation ladder and methodology; mobility applications must add motion-coupled evidence to complete the ladder. (Hasan et al., 1991; Van Dresar et al., 1992; Johnson, 2024; Kartuzova & Kassemi, 2025).
Minimum evidence artifacts (aligning with Table 4): controlled boundary heat flux; pressure history + vent states; stratification/interface measurements (liquid & ullage); benchmark uncertainty budget and reproducibility metrics.
Evidence artifacts (typical): benchmark boundary-condition definitions; calibrated heat-flux/heat-leak partition; multi-point temperature fields; pressure/vent histories; documented uncertainty budgets enabling V&V and UQ.
Design envelope mapping: Design Note 2 rows 2 and 4.
Overall, these mini-cases show that the field is not short of concepts, but short of reusable, validation-grade datasets that couple motion, stratification, phase change, and control actions. In a mobility-era regulatory environment, the central engineering output is an evidence package that links the modeling hierarchy (Table 3) to approval expectations (Table 4) through benchmarkable measurements and uncertainty-bounded predictions (Fig. 12).

9.2 Research Gaps and Future Directions

9.2.1 Insulation durability under realistic service conditions

A recurring weakness of the current literature is that insulation is often assessed at the beginning of life, while service degradation remains under-characterized. Vacuum deterioration, MLI compression, localized thermal bridges, accident tolerance, and inspection accessibility warrant more systematic study under realistic duty cycles (Yin et al., 2024; Camplese et al., 2025).

9.2.2 Reliable coupled datasets: motion + BOR + SP

The field urgently needs more datasets in which motion, pressure, temperature redistribution, fill level, and heat leak are measured simultaneously. Direct motion-enabled LH2 datasets remain rare, which limits the validation of the CFD and reduced-order models in the very regimes that matter for transport applications (Maekawa et al., 2018; Kartuzova et al., 2024; Marques et al., 2025). In particular, mobility-relevant benchmarks should resolve the ullage temperature profiles (not only bulk averages) across low-to-intermediate filling ratios under measured motion inputs, so that wall-to-fluid convection and interfacial-transfer closures can be calibrated rather than assumed (Bailey & Fearn, 1964; Yu & Lim, 2024; Kartuzova et al., 2024).

9.2.3 Standardized benchmark problems and open data

Benchmark reuse has already proven its value in the K-site and SHIIVER lineages. The next step is a broader benchmark portfolio spanning quiescent SP, chilldown and fill, horizontal tanks, and motion-coupled events, with standardized reporting of heat-leak partition, initial state, and uncertainty (Johnson et al., 2021a; Johnson, 2024; Hartwig et al., 2023; Kartuzova & Kassemi, 2025).

9.2.4 Next-generation mitigation: active control and recondensation

The trend toward selectively active mitigation is likely to continue. Recent studies on thermodynamic vent systems (TVS), cryocooler-assisted boil-off reduction, active pressure management, and cargo-transfer architectures suggest that future LH2 systems will combine better passive insulation with smarter active interception or control whenever mission value justifies it (Plachta et al., 2014; Ha et al., 2024; Lee et al., 2025; Blakseth et al., 2025).

9.2.5 Digital twins, AI-enabled modeling, and DX/AX readiness

Digital twins are now plausible enough to be discussed seriously, but are not mature enough to be used casually. Real-time assimilation has already been demonstrated in reduced-order cryogenic models, but practical deployment still depends on the streaming measurements, well-benchmarked reduced-order models, and quantified uncertainty (Marques et al., 2024; Wolf et al., 2025). Recent hydrogen-sector reviews frame digital twins as a route to real-time monitoring, predictive maintenance, and optimized operation, while cryogenic-hydrogen safety reviews go further by proposing integrated CFD-machine learning (ML)-sensor workflows for leak detection and risk-aware operation (Naanani et al., 2025; Vogiatzaki et al., 2025).
For this review, it is useful to interpret DX in practical tank-engineering terms. Digital transformation means converting LH2 tanks from periodically assessed hardware into continuously interpretable systems. In that setting, AI or AX should not be presented as a substitute for thermo-fluid physics, but as a physics-informed augmentation layer. Recent work on AI-aided LH2 tank modeling for sustainable aviation shows that surrogate models can evaluate the gravimetric efficiency across many design variables within milliseconds, which is quite relevant for lightweight mobile applications (Ford et al., 2025). Therefore, the near-term opportunity is hybrid rather than purely data-driven: AI-assisted design-space screening, adaptive reduced-order model calibration, predictive maintenance of vacuum/support systems, route-aware operating strategy, and anomaly detection for transfer and vent control. The field is not yet ready for black-box automation, but it is ready for physics-informed DX and AI-enabled decision support.
Therefore, AX should be understood not as an abstract management slogan but as a concrete engineering workflow. Possible applications include AI-assisted screening of insulation and mission combinations, online estimation of effective heat leak from pressure and temperature streams, anomaly detection for support or vacuum degradation, and route-aware operating recommendations for mobile tanks. The most credible implementations will remain hybrid. Physics-based storage models augmented by data assimilation, surrogate acceleration, and targeted machine-learning components rather than replaced by black-box prediction alone (Marques et al., 2024; Ford et al., 2025; Wolf et al., 2025).
A staged DX roadmap is more realistic than a single leap toward full digital twins. The first stage is better sensing and structured metadata from experiments and operations. The second is validated reduced-order monitoring. The third is data assimilation and anomaly detection. Only after those layers are in place do AI-enabled recommendations and semi-autonomous optimization become credible. Framed in this way, DX and AI are not parallel to tank engineering. They are extensions of the same validation discipline discussed throughout this review (Marques et al., 2024; Naanani et al., 2025; Vogiatzaki et al., 2025).

9.2.6 Scale-up to large tanks: what breaks first?

Scale-up remains the decisive challenge for marine and terminal-scale LH2 storage. What usually breaks first is not a single material or component, but the assumption that small-scale intuition remains valid. On a larger scale, heat-ingress pathways, operational timescales, maintainability, and fill-level sensitivity all change in ways that require their own validation logic (Smith et al., 2022; Inomata et al., 2025; Wang et al., 2024d).

9.2.7 Priority list (Top 10) with expected impact

The literature reviewed in this paper suggests a practical top-priority set for next-generation LH2 tanks: validation-grade quiescent datasets; direct motion-enabled LH2 benchmarks; durable and diagnosable insulation; standardized reporting protocols; benchmark-driven validation across scales; reduced-order mission models; digital-twin-ready data assimilation; large-scale maritime heat-ingress and voyage datasets; integrated transfer architectures that use BOG intelligently; and continued maturation and harmonization of standards (Wang et al., 2024a; Claussner et al., 2025; Johnson, 2024; Li et al., 2024a; Schiaroli et al., 2025).
The key point is that the field is now constrained less by a shortage of concepts than by a shortage of validated integration. Fig. 13 presents the resulting research roadmap.

10. Conclusions

10.1 Major Findings

This review has argued that an LH2 storage tank should be treated as a closed multiphysics system rather than a passive cryogenic vessel. Insulation, heat ingress, interfacial phase change, stratification, pressure evolution, fill policy, and motion environment interact continuously, and the literature is already rich in each subtopic despite remaining fragmented when read tank-by-tank (Kang et al., 2022; Yin et al., 2024; Wang et al., 2024a; Claussner et al., 2025).
Five design-level findings emerge repeatedly. First, insulation performance must be evaluated at the configuration level, not solely at the material level. Second, the BOR and SP are dynamic responses to internal thermal redistribution rather than scalar heat-leak penalties. Third, the filling ratio is a regime-defining variable. Fourth, motion-sensitive applications require sloshing to be treated as a thermodynamic coupling problem, not solely a structural-load problem. Fifth, the decisive bottleneck is now validation-grade data rather than a lack of modeling concepts (Wang et al., 2024b; Smith et al., 2022; Johnson, 2024; Kartuzova & Kassemi, 2025).

10.2 Engineering Implications

For designers, the practical implication is straightforward: mission envelope, geometry, insulation architecture, BOR/SP prediction, and pressure-control philosophy should be developed iteratively rather than sequentially. This is especially true in cross-sector and maritime applications where motion, scale, and operating duration interact strongly (Inomata et al., 2025; Lee et al., 2025; Wolf et al., 2025). In practice, regulated projects also require an evidence package that ties those design choices to credible margins–combining benchmark tests, validated models, and a transparent uncertainty budget. The checklist in Section 8.4.1 summarizes the minimum requirements for translating concepts into approval-ready tank systems. Early alignment with evolving IMO provisions and class-society expectations is prudent because containment concepts, venting/pressure-control philosophy, and vacuum integrity requirements are being actively refined (IMO, 2024; DNV, 2025c; Lloyd’s Register, 2025).
A specific warning follows for mobility-oriented tanks; it is risky to optimize in a single discipline. A sloshing-only program may treat the tank correctly as a load-bearing structure but miss thermodynamic pressure excursions and control saturation. A quiescent BOR-only program may satisfy dormancy metrics but miss motion-triggered pressure spikes/collapses. For this reason, approval-oriented evidence packages should be organized around coupled metrics and uncertainty-bounded responses (Section 8.4.1 and Design Note 2), not isolated component results (Jeon et al., 2024; Wolf et al., 2025).

10.3 Outlook

The next generation of LH2 storage research will likely be shaped less by isolated gains in any one component and more by integration: co-designed experiments and models, durable insulation tied to architecture-level heat-ingress control, support and joining solutions that reconcile thermal efficiency with structural reliability, reduced-order tools anchored to benchmark data, and digital workflows that bring AI-assisted design and operation into genuinely physics-informed practice. This is especially important for mobile sectors such as trailers, aircraft, UAM platforms, and marine transport, where lightweight materials, route-dependent operation, and safety certification all interact. If that integration is achieved, LH2 tanks can evolve from specialized cryogenic vessels into robust, designable infrastructure elements for stationary, mobile, and maritime hydrogen systems.
That integration will need to occur simultaneously across materials, architecture, and digital capability. The storage community will require better LH2-specific structural-material qualification, particularly for high-manganese-related systems and for weld/support interfaces, while also clarifying how far high-strength 15% Ni steel can move toward practical approval as a lighter large-tank option. At the same time, the field will benefit from DX-oriented infrastructure in which benchmark experiments, route data, and reduced-order models support AI-assisted design and operation without detaching the engineering workflow from validated thermo-fluid physics (ABS, 2025a; Johnson et al., 2023; Marques et al., 2024; Ford et al., 2025).
This perspective also explains why mobile LH2 systems deserve continued emphasis. Trailer fleets, aircraft, and UAM platforms, and marine transport are likely to be the applications that force the community to integrate lightweight materials, support-member and joining design, sloshing-aware operation, and digital decision support most quickly. In that sense, mobility is not a niche subtopic of LH2 storage; it is a primary driver that makes motion-coupled BOR evidence and approval-ready evidence packages unavoidable. Evidence classes (E1–E4) provide a pragmatic way to communicate confidence when transferring results across sectors and scales (Tiwari et al., 2024; Wolf et al., 2025; Inomata et al., 2025).

Conflict of Interest

Jong-Chun Park serves as an editorial board member of the Journal of Ocean Engineering and Technology, but played no role in deciding the publication of this article. No potential conflicts of interest relevant to this study are reported.

Funding

This research was supported by the Industrial Strategic Technology Development Program–Development of a high-accuracy prediction method of phase change, BOR, and pressure change in the cargo hold of LNG/LH2 ships in consideration of the operating environment (20026368) funded by the Ministry of Trade, Industry & Energy (MOTIE, Republic of Korea).

Acknowledgments / AI disclosure

The authors acknowledge the use of Google Gemini 3 in drafting selected schematic illustrations. All generated graphics were critically reviewed and edited by the authors, and the authors take full responsibility for their scientific accuracy and final presentation.

Fig. 1
Integrated multiphysics framework for LH2 storage tanks
ksoe-2026-021f1.jpg
Fig. 2
Heat-ingress pathways in a double-wall LH2 storage tank and their coupling to internal circulation and interfacial phase changes
ksoe-2026-021f2.jpg
Fig. 3
Operating envelopes of the major insulation families for LH2 storage tanks
ksoe-2026-021f3.jpg
Fig. 4
Degradation and maintainability pathway for LH2 tank insulation systems
ksoe-2026-021f4.jpg
Fig. 5
Mechanism loop for boil-off and self-pressurization in a quiescent LH2 tank
ksoe-2026-021f5.jpg
Fig. 6
Qualitative map of the filling-ratio effects under quiescent LH2 storage
ksoe-2026-021f6.jpg
Fig. 7
Operational-transient sequence for LH2 storage tanks
ksoe-2026-021f7.jpg
Fig. 8
Sloshing-induced thermo-fluid coupling pathways in LH2 tanks
ksoe-2026-021f8.jpg
Fig. 9
Cross-sector motion envelope for mobile LH2 storage problems
ksoe-2026-021f9.jpg
Fig. 10
Benchmark and diagnostics landscape for LH2 storage experiments
ksoe-2026-021f10.jpg
Fig. 11
Modeling hierarchy and validation ladder for LH2 storage tanks
ksoe-2026-021f11.jpg
Fig. 12
Evidence-backed design workflow linking mission definition, architecture/insulation selection, modeling hierarchy, tests and benchmarks, and IMO/class AiP requirements, closing the loop through DX/AX-oriented operation and digital twins
ksoe-2026-021f12.jpg
Fig. 13
Research roadmap for next-generation LH2 storage tanks
ksoe-2026-021f13.jpg
Table 1
Representative LH2 deployment and demonstration cases used to anchor this review, tagged by evidence class (E1–E4), with supplementary context or supporting evidence indication in brackets where relevant, and the design decisions they most strongly affect.
Sector Representative case or program Evidence class (E1–E4) Design decisions impacted Key lesson (review anchor)
Space S1 — Centaur LH2 cryotank heritage; NASA K-site and SHIIVER benchmark programs (Johnson, 2024; Kartuzova & Kassemi, 2025) (see Sec. 9.1.5). E2 [supplementary context: E1 heritage] Validation ladder for BOR/SP models; insulation and thermal-bridge accounting; pressurization/venting philosophy; scale extrapolation rules. Space heritage anchors benchmark culture and validation-first modeling, providing transferable confidence bounds for mobility-era tanks.
Road/mobile R1 — Operational LH2 trailer distribution validated against 13 complete routes (Wolf et al., 2025) (see Sec. 9.1.5). E1 Route-aware BOR prediction; pressure build-up unit (PBU) sizing and control; fill policy and transfer scheduling; instrumentation/DX for fleet operation. Mobility introduces mission dependence; operational data highlight sensitivity to pressure build-up and transients.
Marine M1 — Hydrogen Energy Supply Chain (HESC)/SUISO FRONTIER pilot chain; Hy touch Kobe; voyage heating ingress validation for a 1,250 m3 carrier tank, and scale-up trajectory toward commercial carriers (HESC, 2025; Inomata et al., 2025; Kawasaki Heavy Industries, 2026) (see Sec. 9.1.5). E1 [supplementary context: E3 scale-up/AiP trajectory] Tank type and insulation concept; vacuum integrity and maintainability; motion-coupled operability margins; evidence packages for IMO/class approval. Marine scale-up makes vacuum management and motion-coupled BOR central, linking design directly to regulatory evidence.
Aircraft/UAM A1 — Airbus LH2 aircraft demonstrators; AMSL/Stralis–Fabrum composite-tank refueling demonstration (Airbus, 2025; Fabrum, 2025) (see Sec. 9.1.5). E3 [limited supporting E2 facility-level evidence] Lightweight shell/support/joining; rapid refueling/no-vent fills; fatigue and thermal-cycle durability; certification evidence planning. Weight drives materials innovation; direct LH2 durability and validation evidence remain decisive gaps.

Note. Across these sectors, the evidence base remains uneven. Some conclusions can be anchored directly in E1/E2 evidence (operational data and controlled LH2 experiments), whereas others, especially for lightweight aircraft/UAM tanks and large-scale carrier concepts, are still extrapolations from E3/E4 evidence (surrogate studies, AiPs, and simulation-only work).

Table 2
Shell, support, and joining options frequently discussed for LH2 tanks and cryogenic hydrogen systems
Configuration family Primary thermal/structural attraction Key joining or qualification issue Likely application domain
304L/316L shells with metallic supports Mature cryogenic weldability, durability, and inspectability Support conduction can dominate local heat leak; thermal bridges remain significant Baseline stationary, road, and marine tanks
304L/316L shells with G10/G11 or GFRP supports Much lower support conductivity than metallic supports Metal-to-composite joining, CTE mismatch, bond reliability, and inspection route become critical Small and medium vacuum-insulated tanks; selected high-performance concepts
High-Ni steel shell concepts (especially 15% Ni steel) Direct 20 K LH2-relevant toughness evidence; possible shell-thickness and weight reduction relative to 304L Weld qualification, fracture assessment, cost, and project-specific approval route remain application-dependent Near-term candidate for large LH2 tanks and transport concepts
High-Mn related austenitic concepts (including Fe-high-Mn and Ni/Mn-replaced systems) Cost/resource advantage with promising cryogenic-service potential LH2-specific hydrogen compatibility, weld qualification, and code acceptance remain limited Exploratory or longer-term LH2 containment research
Composite or resin-rich aircraft tank concepts with metallic interfaces (Olsson et al., 2024; Boddorff, 2025). High specific stiffness and strong lightweight potential for aviation/UAM Permeability, impact tolerance, boss-interface durability, and certification evidence remain major issues Aircraft, UAM, and advanced mobility demonstrators
Table 3
Analytical hierarchy commonly used in LH2 tank analysis and design
Model tier Core assumption set Best use in practice Main limitation
Thermal-resistance/near-equilibrium screening Few temperatures, equivalent resistances, and prescribed heat leak Early insulation screening and hold-time bounding Cannot resolve stratification, interface physics, or motion effects
Lumped/multi-node thermodynamics Small number of control volumes with simplified interface transfer Quiescent BOR/SP prediction and fill-level trends Requires closure assumptions for heat partition and interfacial transport
Reduced-order mission models Lumped physics plus controls, route, or cycle logic Trailer routes, repeated filling/withdrawal, and digital-twin-ready planning Still weak for local boiling, violent sloshing, or wall hot-spots
CHT CFD Solid conduction and fluid convection solved together Thermal bridges, wall-fluid coupling, shield layout, and local stratification Expensive for broad mission screening; still depends on phase-change closure
VOF/phase-change multiphase CFD Resolved free surface with transient evaporation/condensation and motion Sloshing, pressure collapse, chilldown, splash, and event-scale transients High computational cost and strong sensitivity to interface modeling choices
Table 4
Selected regulatory and approval milestones (IMO/class/AiP) relevant to maritime LH2 carriage, including typical evidence-package expectations and the design decisions they most strongly constrain
Date Entity Milestone (abridged) Evidence package expectation (typical) Design decisions impacted
Jan 2024 Bureau Veritas /GTT /TotalEnergies JDP AiPs for a membrane-type LH2 containment system and a preliminary 150,000 m3 carrier concept (GTT, 2024; Bureau Veritas, 2024). Concept safety case; vacuum-space hazard controls; prototype insulation/vacuum tests; inspection/maintainability concept; BOR/SP predictions with sensitivities. Choice of membrane + vacuum insulation; vacuum monitoring and maintenance strategy; BOR margin policy; failure-mode coverage for inter-barrier spaces.
May 2024 IMO (MSC 108) Adopted revised interim recommendations for carriage of liquefied hydrogen in bulk (Resolution MSC.565(108)) (IMO, 2024). Documented containment philosophy; insulation/vacuum-space provisions; PRV/venting philosophy; hazard analysis; compatibility and testing requirements under an interim approval basis. Defines interim approval baseline for LH2 bulk carriage; constrains insulation-space design, vent/relief logic, and hazard controls used in class approval packages.
Sep 2024 IMO CCC 10 Identified that MSC.565(108) did not yet cover membrane-type cargo tanks and anticipated a dedicated Part D for vacuum membrane systems (Lloyd’s Register, 2024; IMO, 2024). Regulatory gap analysis for membrane-type CCS; definition of additional test/analysis expectations for vacuum insulation spaces and leakage response. Triggers extra evidence beyond independent tanks for membrane concepts; elevates vacuum integrity, leak control, and emergency response as approval-critical design decisions.
Sep 2024 DNV / HD KSOE consortium AiP for an 80,000 m3 electric-propulsion LH2 carrier concept with large vacuum-insulated tanks (DNV, 2024). Preliminary design review; global/local structural checks; insulation/vacuum concept definition; BOR/SP prediction; motion/sloshing operability assessment; HAZID/ENVID framing. Carrier-scale architecture selection and vacuum strategy; motion-aware operability margins; alignment of modelling+testing plan with class evidence expectations.
Sep 2024 DNV / CB&I / Hanwha Ocean AiP for a cargo containment system and an 80,000 m3 LH2 carrier concept using vacuum-insulated double-wall spheres (CB&I, 2024). Architecture-level thermal budget; vacuum retention and leak detection plan; motion/sloshing load assessment; integrated safety case and operability strategy. Reinforces vacuum-insulated spherical containment as a near-term carrier candidate; highlights maintainability, inspection access, and motion-coupled BOR as design constraints.
Jan 2025 DNV / HD KSOE AiP for vacuum-insulated large-scale LH2 tank technology and a shipyard-scale vacuum creation method (DNV, 2025b). Qualification plan for vacuum creation/retention at scale; leak detection and monitoring; re-pump/maintenance strategy; representative thermal testing and degradation tracking. Addresses a principal scale-up barrier: creating and maintaining a vacuum at large tank sizes; links BOR targets to maintainability and lifetime vacuum performance.
Sep 2025 IMO CCC 11 Finalized draft revisions to MSC.565(108) for submission to MSC 111, including a new Part D for membrane-type cargo containment systems maintaining insulation spaces under vacuum (Lloyd’s Register, 2025; DNV, 2025c; ABS, 2025b; Korean Register, 2025). Explicit expectations for membrane-vacuum systems: vacuum control and monitoring; leakage management; inspection/maintenance; emergency response; and evidence of operability under motion and transients. Broadens recognized containment options and formalizes evidence needs for membrane-vacuum concepts; strengthens the need for motion-coupled BOR/SP evidence in approval packages.
Sep 2025 DNV / SHI AiP for an 80,000 m3 LH2 carrier design concept (DNV, 2025a). Independent concept review; tank operability assessment under representative motion spectra; validated BOR/SP prediction plan; hazard identification and contingency planning. Adds an independent validation stream for large carrier concepts; reinforces that mobility-era operability (sloshing–BOR coupling) is a class-level approval concern.
Table 5
Mobility-ready design envelope for LH2 tanks: fill-ratio/motion regimes to coupled BOR/SP risk modes, required analysis fidelity, and evidence artifacts.
Regime (fill ratio & motion) Coupled risk mode Key metrics / constraints Minimum analysis tier Evidence artifacts (typical) Case anchor(s) & exemplar sources
Low fill (heel) + low motion High ullage sensitivity; rapid pressure response dP/dt; ullage T stratification; venting philosophy Tier 1–2 (lumped/2-node) + uncertainty bounds Multi-point Tu/Tl, P(t), level, heat-leak budget M1 (Sec. 9.1.5) — Direct anchor: Inomata et al. (2025) [E1]; context: Wang et al. (2024e) [E4].
Intermediate fill + low motion Stratification–interface dynamics dominate BOR excursion; rollover/mixing triggers Tier 2–3 (multi-node + calibrated heat-leak partition) P–T–level logs; repeatable boundary conditions; UQ S1 (Sec. 9.1.5) — Direct anchor: Kartuzova and Kassemi (2025) [E2]; context: Wang et al. (2024b) [E4].
Intermediate fill + mobility motion (roll/pitch/braking) Sloshing–evaporation coupling; pressure spikes/collapse ΔPpeak; vent/PBU capacity margin; interface renewal proxy Tier 3–5 (ROM/CHT + targeted free-surface modeling) Acceleration spectra; PBU/vent states; high-rate P(t); T field R1 (Sec. 9.1.5); M1 (Sec. 9.1.5) — Direct anchor: Wolf et al. (2025) [E1]; proxy trend: Jeon et al. (2024) [E3].
High fill + mobility motion Wall wetting changes; bridge paths dominate heat ingress Bridge heat-leak partition; local cold spots; fatigue hotspots Tier 3–4 (CHT for bridges/supports) + structural checks Support/penetration inventory; inspection/vacuum monitoring S1 (Sec. 9.1.5); A1 (Sec. 9.1.5) — Direct anchor: Johnson (2024) [E2]; context: Boddorff (2025) [E4].
Filling/withdrawal transients under motion Chilldown/flash + motion; control interactions Peak vent flow; allowable pressure window; mission loss budget Tier 2–4 (transient thermodynamics + control-oriented ROM) Flow/valve states; mass balance; transient P–T–level; repeatability R1 (Sec. 9.1.5); A1 (Sec. 9.1.5) — Direct anchors: Hartwig et al. (2023) [E2] and Wolf et al. (2025) [E1].

Note. Design Note 2 is presented in Table 5 and links the operating regime (fill ratio and motion) to risk modes, required analysis fidelity, and typical evidence artifacts aligned with the approval-package items in Section 8.4.1.

References

Aasen, A., Blakseth, S. S., Massing, A., Nekså, P., & Gjennestad, M. A. (2025). Thermal performance estimation for cryogenic storage tanks: Application to liquid hydrogen. International Journal of Hydrogen Energy, 144, 42-54. https://doi.org/10.1016/j.ijhydene.2025.05.134
crossref
ABS. (2025a). Materials for Liquid Hydrogen, Liquid Ammonia and Liquid Carbon Dioxide Containment Systems.

ABS. (2025b) September 16. News Brief: CCC 11 Brief. https://ww2.eagle.org/content/dam/eagle/regulatory-news/2025/CCC%2011%20Brief.pdf

Adler, E. J., & Martins, J. R. R. A. (2025). Liquid hydrogen tank boil-off model for design and optimization. Journal of Thermophysics and Heat Transfer, 39(3), 505-519. https://doi.org/10.2514/1.T7005
crossref
Ahizi, S., Marques, P. A., Monteiro, F. M. F., Gonzalez, L. M., Gambioli, F., Scheufler, H., Abarca, R., & Mendez, M. A. (2024). Scaling laws for experimental characterization of sloshing effect in liquid hydrogen tanks. In the 34th International Ocean and Polar Engineering Conference, ISOPE-I-24-387.

Airbus. (2025) March 25. Airbus showcases hydrogen aircraft technologies during its 2025 Airbus Summit. Airbus Newsroom.

Al Ghafri, S. Z. S., Swanger, A., Jusko, V., Siahvashi, A., Perez, F., Johns, M. L., & May, E. F. (2022). Modelling of liquid hydrogen boil-off. Energies, 15(3), 1149. https://doi.org/10.3390/en15031149
crossref
Alberts, S. J., Doehne, C. J., & Johnson, W. L. (2017). Testing tensile and shear epoxy strength at cryogenic temperatures. Journal of Physics: Conference Series, 922, 012006.

Andrew, J. J., Ubaid, J., Cantwell, W. J., Khan, K. A., & Umer, R. (2025). The effect of cryogenic thermal cycling on impact performance of graphene-enhanced recyclable liquid thermoplastic/carbon fiber composites for hydrogen storage applications. Composites Part C: Open Access, 17, 100601. https://doi.org/10.1016/j.jcomc.2025.100601
crossref
Appel, K. R., Matveev, K. I., & Leachman, J. W. (2025). Modeling the effects of liquid hydrogen tank operations on boil-off losses. Cryogenics, 150, 104161. https://doi.org/10.1016/j.cryogenics.2025.104161
crossref
Bailey, T. E., & Fearn, R. F. (1964). Analytical and experimental determination of liquid-hydrogen temperature stratification. Advances in Cryogenic Engineering, 9, 254-264. https://doi.org/10.1007/978-1-4757-0525-6_30
crossref
Barsi, S., & Kassemi, M. (2008). Numerical and experimental comparisons of the self-pressurization behavior of an LH2 tank in normal gravity. Cryogenics, 48(3–4), 122-129. https://doi.org/10.1016/j.cryogenics.2008.01.003
crossref
Blakseth, S. S., Aasen, A., Massing, A., & Nekså, P. (2025). Local area cooling versus broad area cooling for boil-off reduction in large-scale liquid hydrogen storage tanks. Cryogenics, 148, 104150. https://doi.org/10.1016/j.cryogenics.2025.104065
crossref
Boddorff, A. K. (2025). Choosing the right stuff: Material selection for liquid hydrogen aircraft cryotanks (NASA/TM-20250007133). National Aeronautics and Space Administration.

Bureau Veritas. (2024) February 5. GTT, TotalEnergies, LMG Marin and Bureau Veritas reach a new milestone in liquefied hydrogen transport, Bureau Veritas Marine & Offshore Newsroom; https://marine-offshore.bureauveritas.com/newsroom/gtt-totalenergies-lmg-marin-and-bureau-veritas-reach-new-milestone-liquefied-hydrogen

Camplese, D., Scarponi, G. E., Eberwein, R., Hajhariri, A., Otremba, F., & Cozzani, V. (2025). Comparative performance assessment of multilayer insulation (MLI) systems for liquid hydrogen vessels in fire scenarios. International Journal of Hydrogen Energy, 135, 537-552. https://doi.org/10.1016/j.ijhydene.2025.04.534
crossref
CB&I. (2024) September 18. CB&I and Hanwha Ocean receive approval for cargo containment system and ship design for liquid hydrogen carriers from DNV. PRNewswire press release. https://www.prnewswire.com/news-releases/cbi-and-hanwha-ocean-receive-approval-for-cargo-containment-system-and-ship-design-for-liquid-hydrogen-carriers-from-dnv-302250931.html

CGA H-3. (2024). Standard for cryogenic hydrogen storage (ANSI/CGA H-3-2024. 4th ed Compressed Gas Association (CGA); https://webstore.ansi.org/standards/cga/ansicga2024-2543900

Cha, S.-J., Tak, H.-J., Hwang, B.-K., Lee, J.-P., Kim, J.-H., & Lee, J.-M. (2025). Structural assessment of independent type-C liquid hydrogen fuel tank. Journal of Marine Science and Engineering, 13(4), 730. https://doi.org/10.3390/jmse13040730
crossref
Chen, L., Yao, X., Hu, C., & Han, Y. (2024). Strategies for improving the thermal insulation performance of liquid hydrogen tanks: A review of multilayer insulation structure. International Journal of Hydrogen Energy, 94, 1419-1434. https://doi.org/10.1016/j.ijhydene.2024.10.145
crossref
Cho, H. J., Kim, D. H., Lee, S., & Kim, S. J. (2024). Cryogenic impact toughness characteristics of Ni/Mn replaced austenitic stainless steels subjected to high-pressure gaseous hydrogen. International Journal of Hydrogen Energy, 92, 877-894. https://doi.org/10.1016/j.ijhydene.2024.10.322
crossref
Chung, S.-M., Jeon, G.-M., & Park, J.-C. (2022). Numerical approach to analyze fluid flow in a type C tank for liquefied hydrogen carrier (part 1: Sloshing flow). International Journal of Hydrogen Energy, 47, 5609-5626. https://doi.org/10.1016/j.ijhydene.2021.11.156
crossref
Claussner, L. M., Scarponi, G. E., & Ustolin, F. (2025). Thermo-fluid dynamics modelling of liquid hydrogen storage and transfer processes. Hydrogen, 6(4), 122. https://doi.org/10.3390/hydrogen6040122
crossref
Claussner, L. M., Ustolin, F., & Scarponi, G. E. (2024). Design and operation of liquid hydrogen storage tanks. Chemical Engineering Transactions, 111, 31-36. https://doi.org/10.3303/CET24111006
crossref
Crabb, C. (2022). Thermal resistance network for the design and testing of a liquid hydrogen vapor cooled shielded fuel tank. M.S. thesis. Washington State University; https://doi.org/10.7273/000004577
crossref
Davies, E., Ehrmann, A., & Schwenzfeier-Hellkamp, E. (2024). Safety of hydrogen storage technologies. Processes, 12(10), 2182. https://doi.org/10.3390/pr12102182
crossref
Dawson, V. P., & Bowles, M. D. (2004). Taming Liquid Hydrogen: The Centaur Upper Stage Rocket, 1958–2002 (NASA/SP-2004-4230). NASA.

DNV. (2024) September 16. DNV grants AiP for HD KSOE’s electric propulsion LH2 carrier design. Safety4sea; https://safety4sea.com/dnv-grants-aip-for-hd-ksoes-electric-propulsion-lh%E2%82%82-carrier-design/

DNV. (2025a) September 10. DNV awards AiP to SHI for 80K LH2 carrier design. DNV; https://www.dnv.com/news/2025/dnv-awards-aip-to-shi-for-80k-lh-carrier-design/

DNV. (2025b) January 20. DNV awards approval in principle (AiP) to KSOE for vacuum-insulated large-scale liquefied hydrogen tank technology. DNV News. https://www.dnv.com/news/2025/dnv-awards-approval-in-principle-to-ksoe-for-vacuum-insulated-large-scale-liquefied-hydrogen-tank-technology/

DNV. (2025c) September 17. IMO CCC 11: Interim guidelines for hydrogen as fuel completed. DNV News. https://www.dnv.com/news/2025/imo-ccc-11-interim-guidelines-for-hydrogen-as-fuel-completed/

Ebrahimi, A., Rolt, A., Jafari, S., & Anton, J. H. (2024). A review on liquid hydrogen fuel systems in aircraft applications for gas turbine engines. International Journal of Hydrogen Energy, 91, 88-110. https://doi.org/10.1016/j.ijhydene.2024.10.121
crossref
Fabrum. (2025) October 28. World-first liquid-hydrogen refuelling delivers major step toward Australasia’s first hydrogen-electric flight. Fabrum news release. https://fabrum.nz/world-first-liquid-hydrogen-refuelling-delivers-major-step-toward-australasias-first-hydrogen-electric-flight/

Fesmire, J. E. (2015). Standardization in cryogenic insulation systems testing and performance data. Physics Procedia, 67, 1089-1097. https://doi.org/10.1016/j.phpro.2015.06.205
crossref
Ford, O., Alipour Bonab, S., & Yazdani-Asrami, M. (2025). Artificial intelligence-aided liquid hydrogen tank modelling for sustainable aviation. Energy Reports, 14, 4609-4623. https://doi.org/10.1016/j.egyr.2025.11.054
crossref
Gambioli, F., Malan, A., & Mastroddi, F. (2025). On the relationship between pressure collapse rate and Nusselt number during sloshing in cryogenic liquid hydrogen tanks. International Journal of Hydrogen Energy, 183, 151748. https://doi.org/10.1016/j.ijhydene.2025.151748
crossref
Ghaffari-Tabrizi, F., Haemisch, J., & Lindner, D. (2022). Reducing hydrogen boil-off losses during fuelling by pre-cooling cryogenic tank. Hydrogen, 3(2), 255-269. https://doi.org/10.3390/hydrogen3020015
crossref
GIIGNL. (2009). LNG information paper No. 1 Basic properties of LNG. International Group of Liquefied Natural Gas Importers (GIIGNL); https://cdn.prod.website-files.com/67caa5e229e50c8dab63ce81/685d52efcd0295d5a9965f6a_GIIGNL%20Information%20Paper%20N%C2%B01%20%E2%80%93%20Basic%20Properties%20of%20LNG.pdf

Gil-Esmendia, A., Flores, R. J., & Brouwer, J. (2025). Modeling and improving liquid hydrogen transfer processes. Applied Energy, 390, 125779. https://doi.org/10.1016/j.apenergy.2025.125779
crossref
Glass, D. E. (1997) August. Bonding and sealing evaluations for cryogenic tanks (NASA Contractor Report NASA/CR-201734). https://ntrs.nasa.gov/api/citations/19970029012/downloads/19970029012.pdf

Graczyk, M., Moan, T., & Wu, M. K. (2007). Extreme sloshing and whipping-induced pressures and structural response in membrane LNG tanks. Ships and Offshore Structures, 2(3), 201-216. https://doi.org/10.1080/17445300701423049
crossref
GTT. (2024) January 31. GTT, TotalEnergies, LMG Marin and Bureau Veritas reach a new milestone in liquefied hydrogen transport. GTT press release. https://www.gtt.fr/sites/default/files/CP%20-%20AIPs_LH2%20Carrier%20Membrane_EN_def3.pdf

Gul, W., Qarssis, Y., Tounsi, A., & Bousahla, A. A. (2023). Characterization of polymeric composites for hydrogen tank. Polymers, 15(18), 3716. https://doi.org/10.3390/polym15183716
crossref pmid pmc
Ha, D. W., Noh, H. W., Koo, T. H., Ko, R. K., & Seo, Y. M. (2024). Experimental study on the liquid hydrogen zero boil-off in a small-scale liquefaction system with an automatic control technology. International Journal of Hydrogen Energy, 83, 933-945. https://doi.org/10.1016/j.ijhydene.2024.07.167
crossref
Haberbusch, M. S., Stochl, R. J., & Culler, A. J. (2004). Thermally optimized zero boil-off densified cryogen storage system for space. Cryogenics, 44, 485-491. https://doi.org/10.1016/j.cryogenics.2004.02.016
crossref
Hartwig, J., Moore, R., & Mercado, M. (2023). Test data analysis of the liquid hydrogen chilldown and fill transfer experiments on a flightweight aluminum tank. International Journal of Heat and Mass Transfer, 202, 123766. https://doi.org/10.1016/j.ijheatmasstransfer.2022.123766
crossref
Hasan, M. M., Lin, C. S., & Van Dresar, N. T. (1991). Self-pressurization of a flightweight liquid hydrogen storage tank subjected to low heat flux (NASA Technical Memorandum 103804).

Hauser, D., Kassemi, M., Kartuzova, O., Baker, M., & Mishic, J. (2024) February 1-2. Mastering the cryogenic frontier: Predicting sloshing in cryogenic propellant tanks. [Conference presentation]. 46th Annual AAS Guidance, Navigation and Control Conference, AAS 24-094 https://ntrs.nasa.gov/api/citations/20230018658/downloads/Masteringthecryogenicfrontier_2_22_24.pdf

HESC. (2025). The world-first Hydrogen Energy Supply Chain (HESC) Project. Hydrogen Energy Supply Chain project website.

Himeno, T., Ohashi, A., Anii, K., Haba, D., Sakuma, Y., Watanabe, T., Inoue, C., Umemura, Y., Negishi, H., & Nonaka, S. (2018). Investigation on phase change and pressure drop enhanced by violent sloshing of cryogenic fluid. AIAA Propulsion and Energy Forum. AIAA 2018-4755; https://doi.org/10.2514/6.2018-4755
crossref
Horie, Y., Shirai, Y., Shiotsu, M., Matsuzawa, T., Yoneda, K., Shigeta, H., Tatsumoto, H., Hata, K., Naruo, Y., Kobayashi, H., & Inatani, Y. (2015). Film boiling heat transfer properties of liquid hydrogen in natural convection. Physics Procedia, 67, 643-648. https://doi.org/10.1016/j.phpro.2015.06.109
crossref
Houf, W., & Schefer, R. (2007). Predicting radiative heat fluxes and flammability envelopes from unintended releases of hydrogen. International Journal of Hydrogen Energy, 32(1), 136-151. https://doi.org/10.1016/j.ijhydene.2006.04.009
crossref
Huber, M. L., Lemmon, E. W., Bell, I. H., & McLinden, M. O. (2022). The NIST REFPROP Database for highly accurate properties of industrially important fluids. Industrial & Engineering Chemistry Research, 61(42), 15449-15472. https://doi.org/10.1021/acs.iecr.2c01427
crossref pmid pmc
Im, J., Kwon, T., Lee, S., Kim, D., Cho, H., Won, W., & Gu, B. (2025). Unveiling the impact of internal structure on boil-off gas generation in moving liquid hydrogen (LH2) transport trailer. Renewable and Sustainable Energy Reviews, 216, 115634. https://doi.org/10.1016/j.rser.2025.115634
crossref
IMO. (2024). Sub-Committee on Carriage of Cargoes and Containers, 10th session (CCC 10): Meeting summary. International Maritime Organization; https://www.imo.org/en/mediacentre/meetingsummaries/pages/ccc-10th-session.aspx

IMO. (2025) April 11. IMO approves net-zero regulations for global shipping. International Maritime Organization; https://www.imo.org/en/mediacentre/pressbriefings/pages/imo-approves-netzero-regulations.aspx

Inomata, A., Gokan, K., Yoshiyama, T., Hosono, T., Iwasaki, H., Maekawa, K., & Takeda, M. (2025). Comparison of measured and calculated heat ingress into cargo tanks of liquefied hydrogen carriers during international voyages. Journal of Marine Science and Technology, 30, 816-825. https://doi.org/10.1007/s00773-025-01085-9
crossref
IP Group. (2025) June 2. AMSL Aero - First Australian registered liquid hydrogen tank clears runway for zero-emission test flights. Portfolio news release.

International Maritime Organization (IMO). (2023). 2023 IMO Strategy on Reduction of GHG Emissions from Ships (MEPC.377[80]). https://wwwcdn.imo.org/localresources/en/KnowledgeCentre/IndexofIMOResolutions/MEPCDocuments/MEPC.377%2880%29.pdf

International Maritime Organization (IMO). (2016). Interim Recommendations for the Carriage of Liquefied Hydrogen in Bulk (MSC.420[97])..

International Maritime Organization (IMO). (2024). Revised Interim Recommendations for Carriage of Liquefied Hydrogen in Bulk (MSC.565[108])..

International Organization for Standardization (ISO). (2024). Ships and marine technology - Test procedures for liquid hydrogen storage tank of hydrogen ships (ISO 11326:2044).

International Organization for Standardization (ISO). (2006). Liquid hydrogen - Land vehicle fuel tanks. (ISO 13985:2026).

International Organization for Standardization (ISO). (2026). Hydrogen technologies– Basic considerations for the safety of hydrogen systems. (ISO/TS 15916). https://www.iso.org/standard/89423.html

JSEA. (2024). Japan Built Ships Latest: Technology Development Trends (by ship type). Japan Ship Exporters’ Association report. https://www.jsea.or.jp/wp_2022/wp-content/themes/nihon_senpaku/img/publications/sameij_2024/by-ship-type/Technology%20Development%20Trends.pdf

Jeon, G.-M., Jeong, S.-M., & Park, J.-C. (2024). Experimental and numerical investigation of the influences of sloshing motion on the change in boil-off gas and boil-off rate in a cryogenic liquid tank. Ocean Engineering, 298, 117173. https://doi.org/10.1016/j.oceaneng.2024.117173
crossref
Jeon, G.-M., Park, J.-C., Kim, J.-W., Lee, Y.-B., Kim, D.-S., Kang, D.-E., Lee, S.-B., Lee, S.-W., & Ryu, M.-C. (2022). Experimental and numerical investigation of change in boil-off gas and thermodynamic characteristics according to filling ratio in a C-type cryogenic liquid fuel tank. Energy, 255, 124530. https://doi.org/10.1016/j.energy.2022.124530
crossref
Jiang, W., Sun, P., Li, P., Zuo, Z., & Huang, Y. (2021). Transient thermal behavior of multi-layer insulation coupled with vapor cooled shield used for liquid hydrogen storage tank. Energy, 231, 120859. https://doi.org/10.1016/j.energy.2021.120859
crossref
Jiang, W., Yang, Y., Hu, C., Li, P., Sun, P., & Huang, Y. (2023). Experimental study on composite insulation with foam, multilayer and vapor cooled shield for cryogen storage under different vacuum conditions. Cryogenics, 129, 102604. https://doi.org/10.1016/j.cryogenics.2022.103604
crossref
Jiao, F., Al Ghafri, S., O’Neill, K. T., Stanwix, P. S., Sellner, G. M., Fridjonsson, E. O., Robinson, N., May, E. F., & Johns, M. L. (2024). Review of hydrogen ortho-para conversion: experimental data and reaction kinetics. Reaction Chemistry & Engineering, 9(11), 2846-2862. https://doi.org/10.1039/D4RE00259H
crossref
Johnson, W. L. (2024). Performance testing of liquid hydrogen tanks. Cold Facts Magazine. 40(1), 27-32. https://ntrs.nasa.gov/api/citations/20240000753/downloads/ColdFacts_H2tank_v3.pdf

Johnson, W. L., Balasubramaniam, R., Hibbs, R., Zimmerli, G. A., Asipauskas, M., Bittinger, S., Dardano, C., & Koci, F. D. (2021a). Demonstration of Multilayer Insulation, Vapor Cooling of Structure, and Mass Gauging for Large Scale Upper Stages: Structural Heat Intercept, Insulation, and Vibration Evaluation Rig (SHIIVER) Final Report (NASA TP-20205008233).

Johnson, W. L., Baltman, E., & Koci, F. D. (2023). Assessment of insulation systems for aircraft liquid hydrogen tanks. IOP Conference Series: Materials Science and Engineering, 1301, 012062. https://doi.org/10.1088/1757-899X/1301/1/012062
crossref
Johnson, W. L., Balasubramaniam, R., & Hibbs, R. (2021b). Results of use of heat flux sensors on liquid hydrogen tanks. Proceeidings of Advances in Cryogenic Engineering: Proceedings of the Cryogenic Engineering Conference. 37-44. https://www.proceedings.com/content/065/065204webtoc
crossref pdf
Kang, D.-H., An, J.-H., & Lee, C.-J. (2024). Numerical modeling and optimization of thermal insulation for liquid hydrogen storage tanks. Energy, 291, 130143. https://doi.org/10.1016/j.energy.2023.130143
crossref
Kang, D., Yun, S., & Kim, B.-k. (2022). Review of the liquid hydrogen storage tank and insulation system for the high-power Locomotive. Energies, 15(12), 4357. https://doi.org/10.3390/en15124357
crossref
Kang, S. G., Lee, D. G., & Kim, B. C. (2007). Evaluation of cryogenic performance of adhesives using composite-aluminum double-lap joints. Composite Structures, 78(3), 440-446. https://doi.org/10.1016/j.compstruct.2005.11.005
crossref
Kartuzova, O. V., & Kassemi, M. (2025). Validation of a two-phase CFD model for predicting tank self-pressurization in the ground-based K-site experiment. AIAA SciTech 2025 Forum. AIAA 2025-0353; https://doi.org/10.2514/6.2025-0353
crossref
Kartuzova, O. V., Kassemi, M., & Hauser, D. M. (2024). Validation of a two-phase CFD model for predicting propellant tank pressurization and pressure collapse in the cround-based K-site hydrogen slosh experiment. In AIAA SciTech 2024 Forum. AIAA ; 2024-0547https://doi.org/10.2514/6.2024-0547
crossref
Kassemi, M., Kartuzova, O., & Hylton, S. (2018). Validation of two-phase CFD models for propellant tank self-pressurization: Crossing fluid types, scales, and cravity levels. Cryogenics, 89, 1-15. https://doi.org/10.1016/j.cryogenics.2017.10.019
crossref
Kawasaki Heavy Industries. (2023) December 11. Kawasaki proves excellent thermal-insulation performance for liquefied hydrogen storage tanks. Kawasaki Heavy Industries news release.

Kawasaki Heavy Industries. (2026) January 6. Constructing a commercial-scale liquefied hydrogen supply chain: Contract signed to build world’s largest 40000 m3 liquefied hydrogen carrier. Kawasaki Heavy Industries news release.

Kim, M.-S., & Chun, K. W. (2023). A Comprehensive review on material compatibility and safety standards for liquid hydrogen cargo and fuel containment systems in marine applications. Journal of Marine Science and Engineering, 11(10), 1927. https://doi.org/10.3390/jmse11101927
crossref
Kim, M.-S., & Lee, J. H. (2025). Prediction of Boil-Off Gas in Cryogenic Tanks with a Coupled Thermal Resistance and Thermodynamic Model. Processes, 13(11), 3584. https://doi.org/10.3390/pr13113584
crossref
Klopcic, N., Schatz, M., Sartory, M., Pertl, P., Winkler, F., Drexler, A., Hermann, R., & Trattner, A. (2025). Thermodynamic modeling and simulation of LH2 refueling with conditioning vessel. International Journal of Hydrogen Energy, 134, 164-180. https://doi.org/10.1016/j.ijhydene.2025.04.489
crossref
Korean Register. (2022). Guidance for strength assessment of membrane type LNG cargo containment system under sloshing loads. https://www.krs.co.kr/KRRules/KRRules2025/data/data_other/english/Guidance%20for%20Strength%20Assessment%20of%20Membrane-Type%20LNG%20Tanks%20under%20Sloshing%20Loads_2022.pdf

Korean Register. (2024). Guidelines for selection of metallic materials of containment system for alternative fuels for ships. https://www.krs.co.kr/TECHNICAL_FILE/Guidelines%20for%20Selection%20of%20Metallic%20Materials%20of%20Containment%20System.pdf

Kumar, S. P., Prasad, B. V. S. S. S., Venkatarathnam, G., Ramamurthi, K., & Murthy, S. S. (2007). Influence of surface evaporation on stratification in liquid hydrogen tanks of different aspect ratios. International Journal of Hydrogen Energy, 32(12), 1954-1960. https://doi.org/10.1016/j.ijhydene.2006.08.052
crossref
Leachman, J. W., Jacobsen, R. T., Penoncello, S. G., & Lemmon, E. W. (2009). Fundamental equations of state for parahydrogen, normal hydrogen, and orthohydrogen. Journal of Physical and Chemical Reference Data, 38(3), 721-748. https://doi.org/10.1063/1.3160306
crossref
Lee, E., Kim, J., Jung, W., Lee, J., & Chang, D. (2025). Design and modeling of liquid hydrogen cargo transfer system with make-up boil-off gas compression. International Journal of Hydrogen Energy, 179, 151675. https://doi.org/10.1016/j.ijhydene.2025.151675
crossref
Lee, W. J., Lee, J.-Y., Oh, S.-K., Yang, H., Park, J., Baek, U. B., & Lee, Y.-K. (2024). Temperature dependency of hydrogen embrittlement resistance of austenitic Fe-24Mn-3Cr-0.5Cu-0.47C steel. Materials Science and Engineering: A, 889, 145838. https://doi.org/10.1016/j.msea.2023.145838
crossref
Leveson, N. G. (2011). Engineering a safer world: Systems thinking applied to safety, MIT Press.

Li, K., Guo, X., Shen, T., Gao, Y., Han, Y., & Zheng, J. (2024a). Review of standards for liquid hydrogen storage vessels. Proceedingds of ASME Pressure Vessels and Piping Conference, PVP2024-122708-V001T01A030. https://doi.org/10.1115/PVP2024-122708
crossref
Li, K., Wen, J., & Wang, S. (2024b). Insulation and cost optimization of vapor-cooled shield coupled with para-to-ortho hydrogen conversion based on self-pressurization model of liquid hydrogen tank and NSGA-II. Journal of Energy Storage, 99(Part A), 113229. https://doi.org/10.1016/j.est.2024.113229
crossref
Lloyd’s Register. (2024). CCC 10 summary report (Carriage of Cargoes and Containers Sub-Committee, 16–20 September 2024). Lloyd’s Register Regulatory Affairs. https://www.lr.org/en/knowledge/regulatory-updates/imo-meetings-and-future-legislation/ccc-10-summary-report/

Lloyd’s Register. (2025). CCC 11 summary report (Carriage of Cargoes and Containers Sub-Committee, 8–12 September 2025). Lloyd’s Register Regulatory Affairs. https://www.lr.org/en/knowledge/regulatory-updates/imo-meetings-and-future-legislation/ccc-11-summary-report/

Liu, Z., & Li, C. (2018). Influence of slosh baffles on thermodynamic performance in liquid hydrogen tank. Journal of Hazardous Materials, 346, 253-262. https://doi.org/10.1016/j.jhazmat.2017.12.047
crossref pmid
Liu, Z., Li, Y., & Zhou, G. (2018a). Study on thermal stratification in liquid hydrogen tank under different gravity levels. International Journal of Hydrogen Energy, 43(19), 9369-9378. https://doi.org/10.1016/j.ijhydene.2018.04.001
crossref
Liu, Z., Feng, Y., Lei, G., & Li, Y. (2018b). Fluid thermal stratification in a non-isothermal liquid hydrogen tank under sloshing excitation. International Journal of Hydrogen Energy, 43(50), 22622-22635. https://doi.org/10.1016/j.ijhydene.2018.10.113
crossref
Ludwig, C., Dreyer, M. E., & Hopfinger, E. J. (2013). Pressure variations in a cryogenic liquid storage tank subjected to periodic excitations. International Journal of Heat and Mass Transfer, 66, 223-234. https://doi.org/10.1016/j.ijheatmasstransfer.2013.06.072
crossref
Lv, H., Zhang, Z., Chen, L., Zhang, Z., Chen, S., & Hou, Y. (2024a). Thermodynamic analysis of vapor-cooled shield with para-ortho hydrogen conversion for liquid hydrogen tanks. International Journal of Hydrogen Energy, 50(Part D), 1448-1462. https://doi.org/10.1016/j.ijhydene.2023.09.290
crossref
Lv, H., Chen, L., Zhang, Z., Chen, S., & Hou, Y. (2024b). Numerical study on thermodynamic characteristics of large-scale liquid hydrogen tank with baffles under sloshing conditions. International Journal of Hydrogen Energy, 57, 562-574. https://doi.org/10.1016/j.ijhydene.2023.12.182
crossref
Lvovich, V. F., Johnson, W. L., & Vivod, S. L. (2025). Cryogenic hydrogen studies and testing for advanced aircraft at NASA[Conference presentation]. 4th Annual VFS H2-Aero Symposium. https://ntrs.nasa.gov/api/citations/20250002550/downloads/AeroTech%20Cryogenic%20H2%20testing%20Mar%202025.pdf

Ma, Y., Li, Y., Zhu, K., Wang, Y., Wang, L., & Tan, H. (2017). Investigation on no-vent filling process of liquid hydrogen tank under microgravity condition. International Journal of Hydrogen Energy, 42(12), 8264-8277. https://doi.org/10.1016/j.ijhydene.2017.02.198
crossref
Ma, Y., Zhang, Y., Luo, X., Li, Y., & Xie, F. (2023). Analysis and modeling of no-vent filling process for liquid-hydrogen tank in orbital conditions. Processes, 11(5), 1315. https://doi.org/10.3390/pr11051315
crossref
Maekawa, K., Takeda, M., Miyake, Y., & Kumakura, H. (2018). Sloshing measurements inside a liquid hydrogen tank with external-heating-type MgB2 level sensors during marine transportation by the training ship Fukae-Maru. Sensors, 18(11), 3694. https://doi.org/10.3390/s18113694
crossref pmid pmc
Majumdar, A. K., Steadman, T. E., Maroney, J. L., Sass, J. P., & Fesmire, J. E. (2008). Numerical modeling of propellant boil-off in a cryogenic storage tank. AIP Conference Proceedings, 985, 1507-1514. https://doi.org/10.1063/1.2908513
crossref
Maritime Department, Hong Kong SAR Government. (2024). Revised Interim Recommendations for Carriage of Liquefied Hydrogen in Bulk. Hong Kong Merchant Shipping Information Note, 47/2024. https://www.mardep.gov.hk/filemanager/en/share/msnote/pdf/msin2447.pdf

Marques, P. A., Ahizi, S., & Mendez, M. A. (2024). Real-time data assimilation for the thermodynamic modeling of cryogenic storage tanks. Energy, 302, 131739. https://doi.org/10.1016/j.energy.2024.131739
crossref
Marques, P. A., Ahizi, S., Monteiro, F., Scheid, B., & Mendez, M. A. (2025). On the scaling of heat and mass transfer in cryogenic propellant tanks: a model-based and experimental analysis under static conditions and lateral sloshing. Applied Thermal Engineering, 27(9), 127751. https://doi.org/10.1016/j.applthermaleng.2025.127751
crossref
Mata, C., Yoo, S.-H., & Kim, K.-Y. (2025). Study on the evacuation of gas in bulk-fill insulation materials used in large-scale LH2 storage tanks. International Journal of Hydrogen Energy, 97(7), 1498-1506. https://doi.org/10.1016/j.ijhydene.2024.11.434
crossref
Matveev, K. I., & Leachman, J. W. (2023a). The effect of liquid hydrogen tank size on self-pressurization and constant-pressure venting. Hydrogen, 4(3), 444-455. https://doi.org/10.3390/hydrogen4030030
crossref
Matveev, K. I., & Leachman, J. W. (2023b). Modeling of liquid hydrogen tank cooled with para-orthohydrogen conversion. Hydrogen, 4(1), 146-153. https://doi.org/10.3390/hydrogen4010010
crossref
McDonald, J. P., Smith, C. D., & Boylston, A. D. (2024). Parahydrogen thermophysical properties V05 final report (Report No SNP-DOC-0046). NASA Technical Reports Server (NTRS). https://ntrs.nasa.gov/citations/20240008350

Molkov, V. (2012). Fundamentals of hydrogen safety engineering II. bookboon.com.

Morales-Ospino, R., Celzard, A., & Fierro, V. (2023). Strategies to recover and minimize boil-off losses during liquid hydrogen storage. Renewable and Sustainable Energy Reviews, 182, 113360. https://doi.org/10.1016/j.rser.2023.113360
crossref
Naanani, H., Nachtane, M., & Faik, A. (2025). Advancing hydrogen safety and reliability through digital twins: Applications, models, and future prospects. International Journal of Hydrogen Energy, 115, 344-360. https://doi.org/10.1016/j.ijhydene.2025.02.440
crossref
Namegawa, T., Hoshino, M., & Okawa, T. (2025). Suitability Evaluation of High-strength 15% Ni Steel for Liquid Hydrogen Tank. Nippon Steel Technical Report. 132: 39-43. https://www.nipponsteel.com/common/secure/en/tech/report/pdf/132-07.pdf

Naquash, A., Agarwal, N., & Lee, M. (2024). A review on liquid hydrogen storage: Current status, challenges and future directions. Sustainability, 16(18), 8270. https://doi.org/10.3390/su16188270
crossref
Nuwantha, H. D. N., Jung, D., & Huh, C. (2024). Improvement in predicting the thermal behavior of liquid hydrogen storage through novel thermal modeling. International Journal of Hydrogen Energy, 72, 1143-1158. https://doi.org/10.1016/j.ijhydene.2024.05.382
crossref
Olsson, R., Cameron, C., Moreau, F., Marklund, E., Merzkirch, M., & Pettersson, J. (2024). Design, manufacture, and cryogenic testing of a linerless composite tank for liquid hydrogen. Applied Composite Materials, 31, 1131-1154. https://doi.org/10.1007/s10443-024-10219-y
crossref
Peng, J. K., & Ahluwalia, R. K. (2013). Enhanced dormancy due to para-to-ortho hydrogen conversion in insulated cryogenic pressure vessels for automotive applications. International Journal of Hydrogen Energy, 38(31), 13664-13672. https://doi.org/10.1016/j.ijhydene.2013.08.039
crossref
Pesich, J. M., Hauser, D. M., Yamashita, K. G., Baker, M. C., & Sakowski, B. A. (2024). Liquid hydrogen tank chill and no-vent fill prediction using computational fluid dynamics. IOP Conference Series: Materials Science and Engineering, 1301, 012065. https://doi.org/10.1088/1757-899X/1301/1/012065
crossref
Petitpas, G. (2018). Simulation of boil-off losses during transfer at a LH2 based hydrogen refueling station. International Journal of Hydrogen Energy, 43(46), 21451-21463. https://doi.org/10.1016/j.ijhydene.2018.09.132
crossref
Plachta, D. W., Johnson, W. L., & Feller, J. R. (2014). Cryogenic boil-off reduction system testing. AIAA 2014-3579.
crossref
Qiu, Y., Yang, H., Tong, L., & Wang, L. (2021). Research progress of cryogenic materials for storage and transportation of liquid hydrogen. Metals, 11(7), 1101. https://doi.org/10.3390/met11071101
crossref
Qu, Z., Cai, Z., Ma, Z., Zhang, X., Shu, Z., Liang, W., & Cai, L. (2025). Co-simulation of thermal behavior in liquid hydrogen tanks under vacuum degradation. Energy, 329, 136841. https://doi.org/10.1016/j.energy.2025.136841
crossref
Ratnakar, R. R., Sun, Z., & Blakotaiah, V. (2023). Effective thermal conductivity of insulation materials for cryogenic LH2 storage tanks: A review. International Journal of Hydrogen Energy, 48(21), 7770-7793. https://doi.org/10.1016/j.ijhydene.2022.11.130
crossref
Ryu, M.-R., Yun, S., Kim, B.-K., Kang, D., Kim, G., & Lee, H. (2024). Numerical analysis of sloshing effects in cryogenic liquefied-hydrogen storage tanks for trains under various vibration conditions. Cryogenics, 143, 103961. https://doi.org/10.1016/j.cryogenics.2024.103961
crossref
Saltari, F., Pizzoli, M., Migliorino, M. T., Binni, A., Coppetelli, G., Mastroddi, F., Pagliaroli, T., Duchetto, F. D., Gambioli, F., Abarca, R., & Scheufler, H. (2024). Experimental investigations on the sloshing-induced pressure drop in tanks for hydrogen-powered aircraft. AIAA SciTech 2024 Forum, AIAA 2024-2158.
crossref
Sass, J. P., Fesmire, J. E., Nagy, Z. F., Sojourner, S. J., Morris, D. L., & Augustynowicz, S. D. (2008). Thermal performance comparison of glass microsphere and perlite insulation systems for liquid hydrogen storage tanks. AIP Conference Proceedings, 985, 1375-1382. https://doi.org/10.1063/1.2908497
crossref
Schiaroli, A., Claussner, L., Campari, A., Cirrone, D., Linseisen, B., Friedrich, A., de Ritter, E. T., Kuznetsov, M., & Ustolin, F. (2025). A comprehensive review on liquid hydrogen transfer operations and safety considerations for mobile applications. International Journal of Hydrogen Energy, 107, 164-182. https://doi.org/10.1016/j.ijhydene.2024.12.266
crossref
Shi, C., Zhu, S., Wan, C., Bao, S., Zhi, X., Qiu, L., & Wang, K. (2023). Performance analysis of vapor-cooled shield insulation integrated with para-ortho hydrogen conversion for liquid hydrogen tanks. International Journal of Hydrogen Energy, 48(8), 3078-3090. https://doi.org/10.1016/j.ijhydene.2022.10.154
crossref
Simanullang, M. (2025). Liquid hydrogen storage and insulation materials for liquid hydrogen storage tanks: Trends and challenges. International Journal of Hydrogen Energy, 140, 881-888. https://doi.org/10.1016/j.ijhydene.2025.03.453
crossref
Smith, J. R., Gkantonas, S., & Mastorakos, E. (2022). Modelling of boil-off and sloshing relevant to future liquid hydrogen carriers. Energies, 15(6), 2046. https://doi.org/10.3390/en15062046
crossref
Sun, P. J., Wu, J. Y., Zhang, P., Xu, L., & Jiang, L. M. (2009). Experimental study of the influences of degraded vacuum on multilayer insulation blankets. Cryogenics, 49(12), 719-726. https://doi.org/10.1016/j.cryogenics.2009.09.003
crossref
Takeda, M., Nara, H., Maekawa, K., Fujikawa, S., Matsuno, Y., Kuroda, T., & Kumakura, H. (2015). Simulation of liquid lLevel, temperature and pressure inside a 2000 liter liquid hydrogen tank during truck transportation. Physics Procedia, 67, 208-214. https://doi.org/10.1016/j.phpro.2015.06.036
crossref
Tani, K., Himeno, T., Sakuma, Y., Watanabe, T., Kobayashi, H., Toge, T., Unno, S., Kamiya, S., Muragishi, O., & Kanbe, K. (2021). Pressure recovery during pressure reduction experiment with large-scale liquid hydrogen tank. International Journal of Hydrogen Energy, 46(57), 29583-29596. https://doi.org/10.1016/j.ijhydene.2020.12.184
crossref
Tatom, J. W., Brown, W. H., Knight, L. H., & Coxe, E. F. (1964). Analysis of thermal stratification of liquid hydrogen in rocket propellant tanks. Advances in Cryogenic Engineering, 9, 265-272. https://doi.org/10.1007/978-1-4757-0525-6_31
crossref
ten Damme, L., van Put, M., & Gangoli Rao, A. (2025). Simulation of the refuelling process for an LH2-powered commercial aircraft Part 1 - Modelling and validation. International Journal of Hydrogen Energy, 195, 152168. https://doi.org/10.1016/j.ijhydene.2025.152168
crossref
Tiwari, S., Pekris, M. J., & Doherty, J. J. (2024). A review of liquid hydrogen aircraft and propulsion technologies. International Journal of Hydrogen Energy, 57, 1174-1196. https://doi.org/10.1016/j.ijhydene.2023.12.263
crossref
Van Dresar, N. T., Lin, C. S., & Hasan, M. M. (1992). Self-Pressurization of a Flightweight Liquid Hydrogen Tank: Effects of Fill Level at Low Wall Heat Flux. AIAA 92-0818.

Vaughan, D. A. (1991). Analytical modeling of no-vent fill process. Journal of Spacecraft and Rockets, 28(5), 574-579. https://doi.org/10.2514/3.26283
crossref
Vogiatzaki, K., Simonetto, M., & Cozzani, V. (2025). Digital twins for cryogenic hydrogen safety: Integrating computational fluid dynamics and machine learning. Hydrogen, 6(4), 110. https://doi.org/10.3390/hydrogen6040110
crossref
Wan, C., Shi, C., Zhu, S., Fang, S., Shi, G., Li, D., Shao, S., Wang, K., & Qiu, L. (2025). Experimental study on self-pressurization and active pressure management in a horizontal liquid hydrogen tank. International Journal of Hydrogen Energy, 152, 149399. https://doi.org/10.1016/j.ijhydene.2025.05.029
crossref
Wang, H. R., Wang, B., Pan, Q. W., Wu, Y. Z., Jiang, L., Wang, Z. H., & Gan, Z. H. (2022). Modeling and thermodynamic analysis of thermal performance in self-pressurized liquid hydrogen tanks. International Journal of Hydrogen Energy, 47(71), 30530-30545. https://doi.org/10.1016/j.ijhydene.2022.07.027
crossref
Wang, H., Wang, B., Sun, J., Pan, Q., Luo, G., Tao, X., He, Y., Pfotenhauer, J., Jin, T., & Gan, Z. (2024a). Experimental and computational fluid dynamic investigation on thermal behaviors of liquid hydrogen during the no-vented storage process: a literature review. International Journal of Hydrogen Energy, 57, 822-843. https://doi.org/10.1016/j.ijhydene.2024.01.062
crossref
Wang, H., Wang, B., Xu, T., Shen, X., He, Y., Zhou, W., Pfotenhauer, J., Jin, T., & Gan, Z. (2024b). Thermal models for self-pressurization prediction of liquid hydrogen tanks: formulation, validation, assessment, and prospects. Fuel, 365, 131247. https://doi.org/10.1016/j.fuel.2024.131247
crossref
Wang, H., Gao, Y., Wang, B., Pan, Q., & Gan, Z. (2024c). Perspective for the Safe and High-Efficiency Storage of Liquid Hydrogen: Thermal Behaviors and Insulation. Hydrogen, 5(3), 559-573. https://doi.org/10.3390/hydrogen5030031
crossref
Wang, J., Webley, P. A., & Hughes, T. J. (2024d). Thermodynamic modelling of pressurised storage and transportation of liquid hydrogen for maritime export. International Journal of Hydrogen Energy, 62, 1273-1285. https://doi.org/10.1016/j.ijhydene.2024.02.285
crossref
Wang, J., Webley, P. A., & Hughes, T. J. (2024e). Thermodynamic modelling of low fill levels in cryogenic storage tanks for application to liquid hydrogen maritime transport. Applied Thermal Engineering, 256, 124054. https://doi.org/10.1016/j.applthermaleng.2024.124054
crossref
Wang, Y., Chen, S., Dang, Z., Jiao, J., & Pang, C. (2024f). Research on effective thermal conductivity of hollow glass microspheres under vacuum at low temperature. Applied Thermal Engineering, 252, 123675. https://doi.org/10.1016/j.applthermaleng.2024.123675
crossref
Wang, L., Li, Y., Zhang, F., & Ma, Y. (2015). Performance analysis of no-vent fill process for liquid hydrogen tank in terrestrial and on-orbit environments. Cryogenics, 72, 161-171. https://doi.org/10.1016/j.cryogenics.2015.10.001
crossref
Wang, P., Ji, L., Yuan, J., An, Z., Yan, K., & Zhang, J. (2020a). Modeling and optimization of composite thermal insulation system with HGMs and VDMLI for liquid hydrogen on orbit storage. International Journal of Hydrogen Energy, 45(11), 7088-7097. https://doi.org/10.1016/j.ijhydene.2019.12.110
crossref
Wang, X-YJ., Johnson, W. L., & Hibbs, R. G. (2020b). SHIIVER Assembly Thermal Model Prediction for Transient Test Conditions. AIAA Propulsion and Energy Forum. AIAA 2020-3948; https://doi.org/10.2514/6.2020-3948
crossref
Wang, Z., & Mérida, W. (2024). Thermal performance of cylindrical and spherical liquid hydrogen tanks. International Journal of Hydrogen Energy, 53, 667-683. https://doi.org/10.1016/j.ijhydene.2023.11.287
crossref
Wolf, C., Primke, F., Alekseev, A., Rehfeldt, S., & Klein, H. (2025). Validation of a thermodynamic model for the liquid hydrogen distribution utilizing novel data from an operational trailer. International Journal of Hydrogen Energy, 193, 152397. https://doi.org/10.1016/j.ijhydene.2025.152397
crossref
Wu, S., & Ju, Y. (2023). Numerical study on pressure drops and thermal response of cryogenic fuel storage tanks under sinusoidal sloshing excitation. International Journal of Hydrogen Energy, 48(93), 36523-36540. https://doi.org/10.1016/j.ijhydene.2023.05.270
crossref
Yatsenko, E. A., Goltsman, B. M., Novikov, Y. V., Izvarin, A. I., & Rusakevich, I. V. (2022). Review on modern ways of insulation of reservoirs for liquid hydrogen storage. International Journal of Hydrogen Energy, 47(97), 41046-41054. https://doi.org/10.1016/j.ijhydene.2022.09.211
crossref
Yin, L., Yang, H., & Ju, Y. (2024). Review on the key technologies and future development of insulation structure for liquid hydrogen storage tanks. International Journal of Hydrogen Energy, 57, 1302-1315. https://doi.org/10.1016/j.ijhydene.2024.01.093
crossref
Yu, T., & Lim, Y. (2024). Changes in holding time depending liquid filling ratio, heat transfer parameters, and shape of the tank. International Journal of Hydrogen Energy, 82, 836-852. https://doi.org/10.1016/j.ijhydene.2024.07.333
crossref
Zheng, J., Chen, L., Xu, X., Guo, L., Zhou, Y., & Wang, J. (2019). A novel insulation system based on active cooling without power input for liquid hydrogen storage. Energy, 182, 1-10. https://doi.org/10.1016/j.energy.2019.06.050
crossref
Zhu, X. Y., Lee, J. H., Kim, K. H., Lim, C. H., & Lee, S. H. (2024). Coupled CFD modeling and thermal analysis of multi-layered insulation structures in liquid hydrogen storage tanks for various vapor-cooled shields. Case Studies in Thermal Engineering 63, 105317, https://doi.org/10.1016/j.csite.2024.105317
crossref
Zhu, Y., Bu, Y., Gao, W., Xie, F., Guo, W., & Li, Y. (2023). Numerical study on thermodynamic coupling characteristics of fluid sloshing in a liquid hydrogen tank for heavy-duty trucks. Energies, 16(4), 1851. https://doi.org/10.3390/en16041851
crossref
Zuo, Z., Jiang, W. B., Qin, X., & Huang, Y. (2020). Numerical investigation on full thermodynamic venting process of liquid hydrogen in an on-orbit storage tank. International Journal of Hydrogen Energy, 45(51), 27792-27805. https://doi.org/10.1016/j.ijhydene.2020.07.099
crossref
Zuo, Z., Zhu, W., Huang, Y., Wang, L., & Tong, L. (2023). A review of cryogenic quasi-steady liquid-vapor phase change: theories, models, and state-of-the-art applications. International Journal of Heat and Mass Transfer, 205, 123916. https://doi.org/10.1016/j.ijheatmasstransfer.2023.123916
crossref
TOOLS
Share :
Facebook Twitter Linked In Google+ Line it
METRICS Graph View
  • 3 Crossref
  •   Scopus
  • 3,096 View
  • 56 Download


ABOUT
BROWSE ARTICLES
ARTICLE CATEGORY

Browse all articles >

PUBLICATION ETHICS
FOR CONTRIBUTORS
Editorial Office
President Office BD Rm. 1302, 13 Jungang-daero 180beon-gil, Dong-gu, Busan 48821, Republic of Korea
Tel: +82-51-759-0656    Fax: +82-51-759-0656    E-mail: ksoehj@ksoe.or.kr                

Copyright © 2026 by The Korean Society of Ocean Engineers.

Developed in M2PI

Close layer
prev next