United States

In response to the growing global interest in hydrogen as an energy carrier this study provides the first attempt to develop a baseline inventory of U.S. hydrogen emissions from production distribution and storage. The scope of this study was limited to pure hydrogen emissions and excludes emissions from low purity hydrogen streams and carriers. A detailed literature search was conducted utilizing various greenhouse gas emissions inventory protocol principles and guidelines to consolidate a list of activity data and emission factors. The best available activity data and emission factors were then selected via a Multi-Criteria-Based Decision Making Method named Technique for Order Preference by Similarity to Ideal Solution or modelled using best-engineering estimates. The study estimated total U.S. hydrogen emissions of 0.063 MMTA with emission bounds ranging from 0.02 to 0.11 MMTA. Given the total estimated H2 production capacity of 7.97 MMTA the study estimates a total U.S. hydrogen emission rate for production distribution and storage of 0.79% (0.26%–1.32%). To reduce the uncertainty in the estimated total hydrogen emissions future work should be conducted to measure facility-level hydrogen emission factors across multiple sectors. The inventory framework developed in this study can serve as a living document that can be updated and enhanced as more empirical data is obtained. This study also provides detailed insights regarding key emission or leakage sources and causes from each supply chain stage. The insights and conclusions from this study can provide direction for hydrogen production companies and safety professionals as they develop hydrogen emission mitigation measures and controls.

A new protocol is presented to directly characterise the toughness of microstructural regions present within the weld heat-affected zone (HAZ) the most vulnerable location governing the structural integrity of hydrogen transport pipelines. Heat treatments are tailored to obtain bulk specimens that replicate predominantly ferriticbainitic bainitic and martensitic microstructures present in the HAZ. These are applied to a range of pipeline steels to investigate the role of manufacturing era (vintage versus modern) chemical composition and grade. The heat treatments successfully reproduce the hardness levels and microstructures observed in the HAZ of existing natural gas pipelines. Subsequently fracture experiments are conducted in air and pure H2 at 100 bar revealing a reduced fracture resistance and higher hydrogen embrittlement susceptibility of the HAZ microstructures with initiation toughness values as low as 32 MPa√ m. The findings emphasise the need to adequately consider the influence of microstructure and hard brittle zones within the HAZ.

High-temperature (HT) proton exchange membrane (PEM) fuel cells (FC) offer key advantages for sustainable transportation especially in heavy-duty applications due to their improved thermal efficiency and water management. This study introduces an inverse design framework to develop flow field plates integrated with a gas diffusion layer (GDL) enabling scalable electrochemical performance from the unit cell to the plate level. A reduced-order homogenization-based multiphysics model is developed to evaluate designs with approximately 1000× faster computation. Flow channel orientation is optimized using a tensor field method and dehomogenized into manufacturable geometries. Optimized designs validated through high-fidelity 3D simulations show up to 12% higher average current density and 88% lower pressure drop compared to conventional parallel and mesh configurations. To address fabrication challenges solid-to-porous metal additive manufacturing is employed producing monolithic structures that integrate flow channels with a porous metal GDL. Both numerical and physical tests confirm high permeability and improved power output compared to carbon-based GDLs. These findings highlight the effectiveness of combining advanced computational modeling with metal 3D printing to enhance the performance and manufacturability of high-temperature PEMFC supporting their broader adoption in sustainable energy applications.

Hydrogen’s increasing potential as an alternative fuel for heavy-duty transport has led to the conversion of conventional diesel compression-ignition engines to spark-ignition hydrogen operation. Hydrogen’s broad flammability range enables leaner operation achieving both higher engine efficiency and near-zero emissions. In particular direct injection hydrogen combustion improves volumetric efficiency and reduces problems including pre-ignition and knock related to hydrogen port-fuel injection. In the present work we performed an optical investigation of direct injection (DI) hydrogen combustion under leaner mixture conditions. The study was conducted using a heavy-duty optical diesel engine modified for spark-ignition operation. Bottom-view natural flame luminosity and OH-PLIF imaging were conducted along with in-cylinder pressure measurements. Experiments were conducted at three air-excess ratios (3 3.4 and 3.8) with spark timings (ST) varied from − 15 ◦CA aTDC to − 30 ◦CA aTDC. Hydrogen injection ended at − 30 ◦CA aTDC with the start of injection adjusted accordingly to achieve the desired lambda conditions. The maximum IMEPg corresponded to the lowest COV of the IMEPg indicating optimal spark timing for lean DI hydrogen combustion. The optimized spark timing for λ = 3 λ = 3.4 and λ = 3.8 were occurred at − 25 ◦CA aTDC − 25 ◦CA aTDC and − 30 ◦CA aTDC respectively. The corresponding COV of IMEPg values were below 5 % indicating stable combustion. The flame kernel first initiates at the spark plug and then propagates toward the piston’s outer boundary however the flame propagation does not remain as a continuous front unlike port-fuel injected hydrogen combustion. The effect of fuel stratification is evident in combustion luminosity and OH-PLIF images showing pockets of varying intensity within the combustion chamber. Natural flame luminosity images reveal incomplete flame coverage and asymmetric combustion emphasizing the need for metal engine experiments to further quantify the unburned hydrogen and associated combustion losses.

The development of efficient and sustainable hydrogen storage materials is a key challenge for realizing hydrogen as a clean and flexible energy carrier. Among various options metal hydrides offer high volumetric storage density and operational safety yet their application is limited by thermodynamic kinetic and compositional constraints. In this work we investigate the potential of machine learning (ML) to predict key thermodynamic properties—equilibrium plateau pressure enthalpy and entropy of hydride formation—based solely on alloy composition using Magpie-generated descriptors. We significantly expand an existing experimental dataset from ~400 to 806 entries and assess the impact of dataset size and data augmentation using the PADRE algorithm on model performance. Models including Support Vector Machines and Gradient Boosted Random Forests were trained and optimized via grid search and cross-validation. Results show a marked improvement in predictive accuracy with increased dataset size while data augmentation benefits are limited to smaller datasets and do not improve accuracy in underrepresented pressure regimes. Furthermore clustering and cross-validation analyses highlight the limited generalizability of models across different material classes though high accuracy is achieved when training and testing within a single hydride family (e.g. AB2). The study demonstrates the viability and limitations of ML for accelerating hydride discovery emphasizing the importance of dataset diversity and representation for robust property prediction.

The cement industry a foundation of infrastructure development is responsible for nearly 7 % of global CO2 emissions highlighting an urgent need for scalable decarbonization strategies. This study investigates the technoeconomic feasibility of integrating on-site solar-powered green hydrogen production into cement manufacturing processes. A mixed-integer linear programming (MILP) model optimizes the design and operation of solar photovoltaics (PV) proton exchange membrane (PEM) electrolyzer and hydrogen storage for a representative cement plant in Texas. Five hydrogen substitution scenarios (10–30 % of thermal demand) were evaluated based on net present cost (NPC) levelized cost of hydrogen (LCOH) cost of CO2 avoided and greenhouse gas (GHG) emissions reduction. Hydrogen integration up to 30 % is technically viable but economically constrained with LCOH rising non-linearly from $58.7 to $95.3 GJ− 1 due to escalating component costs. Environmentally a 30 % hydrogen share could reduce total U.S. cement sector emissions by 22 %. While significant this confirms at present the solar-driven hydrogen serves as a partial solution rather than a standalone pathway to deep decarbonization suggesting it must complement other strategies like carbon capture electrification and other complementary technologies. The economic viability of this approach is entirely contingent on financial incentives as the investment tax credits of 80 % or higher are essential to enable cost parity with fossil fuels. This work provides a comprehensive techno-economic and environmental framework concluding immense economic barriers and that aggressive policy support is indispensable for enabling the transition to low-carbon cement manufacturing.