Green hydrogen produced via renewable-powered electrolysis is a key pathway for global decarbonization; however, its large-scale deployment remains limited by high costs and regional variability in renewable resources and infrastructure. To address the limitations of conventional deterministic techno-economic models, this study presents a machine learning-based framework that integrates spatial datasets, including renewable energy potential, electricity tariffs, and electrolyzer techno-economic parameters, across various regions of the world. A Random Forest model predicts the levelized cost of hydrogen with a test root mean square error of $0.28 per kg and an R² value of 0.94, significantly outperforming a simple linear regression model, which achieved an R² of 0.71. The main factors influencing cost are electricity price, availability of renewable energy, electrolyzer cost, and system efficiency. The model identifies the lowest-cost production areas, with hydrogen projected to cost between $1.8 and $2.5 per kg, in regions combining high solar and wind resources with stable infrastructure, such as North Africa, the Middle East, and Australia. This approach can inform policy, investment, and planning for cost-effective green hydrogen deployment.

Abstract

Introduction:
Rural electrification policy in Sub-Saharan Africa (SSA) increasingly prioritizes the productive use of electricity (PUE) to generate income and sustain decentralized energy systems. However, prevailing intervention models assume that financing, skills, and market access are the primary barriers to adoption. This study tests that assumption by examining whether infrastructure reliability functions not merely as one constraint among many, but as a gatekeeping mechanism that determines whether other factors can operate at all. Using evidence from rural Nigeria, where grid supply averages 2–4 hours daily and mini-grid sustainability is fragile, we test the hypothesis that below a critical reliability threshold, standard determinants of PUE adoption cease to predict behavior.

Methods:
In July 2025, we conducted a cross-sectional household survey (n=249) in two contrasting communities in southwestern Nigeria: one grid-connected with an unreliable supply (Ila-Orangun), and one where the community solar mini-grid had collapsed (Gbamu-Gbamu). Data were analyzed using descriptive statistics, logistic regression, and threshold modeling to examine (1) the magnitude of the underutilization of electricity or access–utilization gap, (2) the predictive power of income, education, and social capital under reliability constraints, (3) the hierarchy of adoption barriers, and (4) household adaptation strategies following infrastructure failure.

Results:
Analysis showed that only 32.4% of electrified households used electricity productively, resulting in a 67.6% access–utilization gap. Above 10–11 hours of daily supply, PUE adoption doubled to 57.7% (OR=7.6, p<0.001); below this threshold, adoption stagnated at 27–30%, and household characteristics (income, education, social capital) showed no significant predictive power. The barrier hierarchy inverted relative to Asian contexts: 76% of non-adopters cited unreliable supply as the primary barrier, versus 24% citing financing. Following mini-grid collapse in Gbamu-Gbamu, we see that adaptation strategies were stratified sharply by wealth: the poorest quintile experienced a 47.8% de-electrification rate, compared to 20% in the wealthiest quintile, revealing a 27.8 percentage-point equity gap.

Conclusions:
Findings support a conditional model of PUE adoption: standard determinants operate only above reliability thresholds. In severely constrained environments, infrastructure quality acts as a gatekeeper, rendering demand-side interventions (equipment subsidies, credit, training) ineffective unless reliability is first assured. Policy implications are as follows: (1) sequence interventions by prioritizing reliability improvements before deploying demand-side support; (2) redefine success metrics to reflect tier-specific reliability within frameworks like the Multi-Tier Framework; and (3) treat infrastructure sustainability as an equity imperative, given that system collapse disproportionately harms poor households. This study urges a reorientation of energy policy from access provision to quality-assured service delivery as a prerequisite for productive use and inclusive development.

Introduction

The just transition agenda has become prominent in European Union energy policy but remains contested (Sabato, 2024). The research agenda on just transition and energy poverty alleviation is marked by a technical–economic bias and a neglect of socio-political questions (Bouzarovski, 2022). Moreover, there is a lack of empirical data on how just transition policies are implemented in fiscally constrained, southern European contexts. This study investigates Renewable Energy Communities (RECs) in Greece as promising eco-social and just transition policy tools (De Vidovich, 2024) and hybrid institutional spaces at the intersection of energy justice, social policy, and participatory energy governance frameworks. It employs a comparative analysis of a rural, top-down, municipality-initiated REC and an urban, bottom-up, citizen-initiated RECs which undertake actions to address energy poverty. This approach elucidates the relationships, interactions and interdependencies that determine how participatory energy policies and just transition agendas are mobilized and actualized within a local and structural context. This study thus highlights energy poverty not as a merely technical issue of resource distribution, but as a central just transition policy question entangled in broader questions over welfare distribution, recognition, and citizenship. The just transition is shown as a socio-technical issue that is spatially bound and embedded in pre-existing institutional contexts.

Methodology

This study asks whether and under what conditions RECs can deliver just transition policy goals such as energy poverty alleviation in resource-constrained environments. Utilizing a qualitative methodology, it reveals through semi-structured interviews and document analysis the taken-for-granted assumptions within the discourses and practices of the RECs. The interviews included 15 stakeholders from various backgrounds, such as REC members, civil society representatives, municipal government officeholders, and academic experts. Policy documents were analyzed from a variety of state, non-state, transnational and national institutions, enabling a broad coverage of discourses. The institutional context of Greece was also examined through careful analysis of the policies, laws, and discourses on the just transition and the role of RECs at the national level. This interdiscursive triangulation helps uncover the interlinkages between systemic, institutional, and subject-formation-level processes.

Results

This study highlights the tensions between inclusion and exclusion, market-logic and community governance, and philanthropy and solidarity in RECs seeking to achieve just transition goals. This exposes the pervasiveness of market logics within decentralized initiatives showing an institutional tension between market-based integration and community-based governance. Both initiatives are constrained by unfavorable grid access policies, lack of institutional support, and regulatory instability. Both initiatives respond to these constraints by implementing needs-based, residual redistribution for energy-poor households while excluding broader community members.

The governance formation and geography of the initiative influence their effectiveness in achieving just transition policy objectives. The top-down initiative is shown to more closely replicate the dominant policy paradigm of privatized energy, but local visibility gives rise to democratizing tendencies and claims-making based on citizenship, particularly from excluded island residents. Even though the top-down nature of the REC ensured adequate resources to cover all residents’ needs, initial mistrust led to low participation and uneven allocation of membership shares. As a result, the REC maintains 150 outstanding applications leading to tensions between included and excluded residents. The perceived injustice over access to the project and the distribution of benefits leads to lower stakeholder engagement threatening the project’s long-term social sustainability. The bottom-up initiative showcases discursive imaginaries oriented towards collective democratic control of energy and access to energy as a right but is tempered in practice by operational constraints including a lack of resources and consideration of beneficiaries’ anonymity, thus implementing technical–economic solutions for energy poverty. These operational constraints and regulatory instability have prompted fears over the cooperative’s financial viability leading to the project’s stagnation and inability to include more members.

Conclusion

The comparative approach allowed for the exploration of the effect of different governance structures (top-down bottom-up) and geography (urban-rural) on the success of just transition, community-based policies. This spatializes the just transition agenda, elucidating the translation of EU policy into national and local contexts. The results indicate the need for further empirical exploration of RECs and their potential to achieve just transition objectives as mediated by local contexts. It also highlights the need to not assume the success of community-based initiatives but rather to implement carefully designed policies that ensure adequate institutional support at the national and supranational level while accounting for local specificities. Energy transition policies aiming at strengthening decentralized production need to move beyond “one-size-fits-all” approaches and implement tailor-made solutions that empower local communities to achieve just transition objectives. Finally, this study underlines RECs as compensatory mechanisms for highly concentrated energy markets and inadequate redistribution policies in fiscally constrained southern-European policy environments.

This study examines whether bilateral green finance (BGF) can successfully reduce carbon dioxide emissions in West Africa by promoting the growth of renewable energy sources and enhancing energy efficiency. The study estimates linear panel models for lagged per-capita CO2 emissions using an unbalanced panel of 15 West African nations from 2000 to 2023. To account for contemporaneous correlation and heteroskedasticity, correlated panel-corrected standard errors and Driscoll–Kraay corrections are used. Lagged BGF is the primary explanatory variable, which is supplemented by foreign direct investment, energy intensity, trade openness, methane emissions, total renewable energy consumption, access to electricity in urban areas, renewable electricity output, and an SDG 7b infrastructure indicator that captures installed renewable capacity.

According to preliminary estimates, BGF is positively and significantly correlated with lagged CO2 emissions, suggesting that West Africa's current green finance allocation patterns are not yet translating into complete decarbonization. Reductions in per-capita CO2 emissions are also strongly correlated with higher percentages of renewable electricity output and lower energy intensity, underscoring the significance of allocating green finance to projects that improve efficiency and decarbonize the power sector. On the other hand, there is a positive correlation between emissions and both total renewable energy consumption and increased access to electricity, which is indicative of the region's rapid electrification, structural change, and economic growth.

By using quantile regressions to compare mean and distributional effects, the study demonstrates that the positive correlation between BGF and emissions is stronger at the median and upper quantiles of the emissions distribution. Building on these results, the study develops and evaluates a mediation hypothesis according to which energy efficiency and the deployment of renewable energy act as transmission channels through which BGF could, in theory, lower emissions, even if the reduced-form BGF–CO₂ relationship is currently positive. The findings offer policy-relevant insights for redesigning bilateral green finance instruments to better support West Africa's low-carbon transition by leveraging efficiency gains and renewable energy.

Low-carbon fuel standards (LCFS) such as Oregon’s Clean Fuels Program (CFP) rely on market-based credit prices to motivate reductions in the lifecycle carbon intensity (CI) of transportation fuels. In practice, those price signals operate alongside the physical build-out of low-carbon infrastructure, especially electric vehicle (EV) charging networks, which may amplify or dampen the extent to which regulated parties respond to credit-market incentives. Yet it remains unclear whether higher credit prices consistently produce proportionate CI reductions, or whether the credit price–CI relationship changes once prices cross key “turning points,” and whether those turning points depend on the level of EV charging infrastructure in place. This study tests for threshold behavior in the relationship between CFP credit prices, EV charging deployment, and ethanol CI. Using monthly data from January 2018 through December 2024 drawn primarily from Oregon CFP public reports and related market sources, the analysis focuses on ethanol CI (gCO₂e/MJ) as the outcome variable and the average CFP credit price as the main policy variable. Additional controls include B5 and E10 cost indices, measures of EV charging deployment, and lagged CI dynamics.

Empirically, the study applies threshold regression methods in the spirit of Hansen’s endogenous threshold model. The core specifications allow the marginal effect of credit prices on ethanol CI to differ across regimes defined by one or more estimated credit-price thresholds, selected by searching over candidate values and choosing thresholds that improve model fit (e.g., minimizing residual variance and improving AIC/BIC). Linear and quadratic models serve as benchmarks for comparison. The working hypothesis is that the credit price–CI relationship is piecewise: responses may be muted at low prices, strengthened in an intermediate range as incentives become binding, and weakened at very high prices as low-cost abatement options are exhausted. The study further tests whether EV charging infrastructure shifts these regimes by allowing charging variables and their interaction with credit prices to vary across threshold-defined regimes. Robustness checks examine alternative threshold variables and sensitivity to the lag structure and specification. By identifying where credit prices and EV infrastructure most strongly influence CI, the results aim to inform LCFS/CFP design choices, including benchmark setting, banking rules, and cost-containment mechanisms that shape where credit prices tend to fall relative to the most effective range for driving decarbonization.

This study investigates the socio-economic determinants of household inverter adoption in rural India, positioning it as a critical proxy for understanding future distributed energy storage transitions. As India accelerates its renewable energy deployment, integrating storage technologies at the household level will be crucial for maintaining grid stability and ensuring energy security. However, there is little empirical evidence on what drives the early adoption of storage-like technologies in developing economies. Leveraging nationally representative panel data from the Access to Clean Cooking Energy and Electricity Survey (2015–2018) across six states, we employ random-effects and rare-events logistic regression models to address the inherent econometric challenges of analysing low-probability adoption events.
Our analysis reveals that, despite the widespread unreliability of the grid, only 4-5% of households own inverters, with these devices primarily concentrated among affluent, literate, and grid-connected individuals. This pattern indicates a significant socio-economic stratification in the early stages of storage adoption. Key findings challenge conventional expectations: Grid connection is a strong positive predictor, confirming its role as a reliability-coping mechanism. In contrast, renewable energy sources show no significant complementarity, indicating a market disconnect between standalone solar adoption and storage integration. A striking and consistent result is the significantly lower adoption rates among male-headed households, suggesting important gendered differences in reliability valuation, financial prioritisation, or intra-household decision-making dynamics.
The persistence of these patterns across model specifications underscores that inverters are not merely backup devices but represent a household’s first capital-intensive investment in energy security. We argue that current inverter adopters constitute a clearly identifiable “storage-ready” segment, providing a behavioural and economic baseline for forecasting the adoption of more advanced storage technologies, such as lithium-ion battery systems. The study concludes with urgent policy implications: without targeted financing mechanisms, enhanced grid reliability initiatives, and gender-sensitive energy planning, the transition to distributed storage risks reinforcing existing energy inequities. These insights are vital for designing inclusive energy policies that can bridge the reliability gap, prevent a “storage divide,” and accelerate India’s sustainable energy transition.

The global aviation sector faces mounting pressure to decarbonize, with sustainable aviation fuels (SAFs) representing a critical transitional solution. Brazil, endowed with abundant biomass resources and a mature biofuel industry, holds significant potential for SAF production. However, the economic viability, energy efficiency, and environmental impact of SAF pathways remain underexplored through integrative energy economics frameworks. This study applies principles of thermo-economics and exergy economics to evaluate the technical, economic, and policy dimensions of SAF production in Brazil. By examining the exergetic efficiency and life-cycle costs of key production routes, such as HEFA, ATJ, Fischer–Tropsch, and SIP, we aim to identify optimal pathways for scaling SAF within Brazil’s energy and climate policy landscape.

The methodology integrates a systematic review of Brazilian SAF initiatives, regulatory frameworks, and production technologies with thermo-economic and exergy analyses. Data were drawn from government publications, technical reports, academic literature, and industry roadmaps (2010–2025). Exergy analysis was applied to quantify the useful energy potential and irreversibilities of each SAF conversion pathway. Thermo-economic models were developed to assess the cost of exergy destruction and the trade-offs between capital investment, operating costs, and thermodynamic performance. Policy instruments such as RenovaBio, the Fuel of the Future Program, and tax incentives were evaluated for their impact on the exergy–economic competitiveness of SAFs relative to conventional aviation kerosene.

The analysis reveals that while Brazil possesses sufficient feedstock potential (particularly from sugarcane residues, wood waste, and used cooking oil), the exergy efficiency of SAF production varies significantly across technologies. Fischer–Tropsch and ATJ routes showed higher exergetic performance for lignocellulosic feedstocks, whereas HEFA exhibited lower exergy losses but faced feedstock limitations and higher hydrogen demand. Thermo-economic assessment indicated that SAF production costs remain 20–80% above fossil kerosene, largely due to exergy inefficiencies in conversion and upgrading stages. The study also highlights regulatory gaps, including the absence of a SAF-specific mandate, complex environmental licensing, and post-blending taxation issues, which collectively hinder investment and scale-up. The integration of exergy-based metrics into policy design could prioritize high-efficiency pathways and align incentives with decarbonization goals.

To accelerate SAF deployment, Brazil must adopt an integrated energy policy approach that combines thermo-economic optimization with supportive regulation. The principal recommendations include (1) introducing exergy-based subsidies to reward high-efficiency SAF pathways; (2) simplifying environmental licensing through multi-level governance coordination; (3) establishing an SAF blending mandate linked to carbon intensity reduction; and (4) fostering public–private partnerships for R&D in exergy-efficient conversion technologies. This study demonstrates that thermo-economic and exergy analyses provide a robust framework for guiding SAF policy, enabling Brazil to leverage its bioenergy potential while advancing toward its climate commitments under the Paris Agreement.

The Inflation Reduction Act (IRA) introduced the Energy Community Tax Credit Bonus, providing an additional 10% investment tax credit for renewable energy projects located in designated “energy communities” with a legacy of coal mining and coal-fired power generation. While this policy aims to accelerate clean energy investment and economic revitalization in historically fossil fuel-dependent regions, less is known about how these communities are responding to new renewable development on the ground. This study examines local responses to large-scale solar projects in Southern Indiana, a state characterized by strong local siting authority and county-level control over energy project approval. We conducted 22 semi-structured telephone interviews with county commissioners and local officials across 32 counties between January and March 2025, including 13 interviews in counties designated as energy communities due to coal mine or coal plant retirements, and 9 interviews in adjacent counties without energy community status. Our findings indicate that coal’s legacy continues to shape local identity, political culture, and economic narratives, with 86% of respondents citing its historic importance through tax revenues, employment, industrial attraction, and community stability. At the same time, economic considerations were central to solar energy acceptance, as 73% of respondents emphasized tax revenue, landowner lease income, and local economic activity as key benefits of solar development. However, despite these perceived advantages and the added incentives from the IRA, officials consistently identified constituent pressure and local political dynamics as the primary factors influencing project approvals and denials, often superseding technical siting criteria or economic rationale. These results suggest that federal place-based incentives alone are insufficient to ensure local acceptance of renewable energy projects in post-coal regions, as historical energy identities, land use values, and political contexts mediate how such policies are interpreted and implemented at the local level. The findings also point to the potential role of clearer state-level siting guidance, and early, sustained community engagement in improving the prospects for large-scale solar deployment in these areas. By situating county-level decision-making within broader federal decarbonization policy and regional energy transition dynamics, this study contributes to emerging scholarship on post-coal transitions, place-based policy, and the governance challenges surrounding utility-scale renewable energy development.

The European Union’s electricity markets face escalating volatility due to the rapid expansion of variable renewable energy sources (RES), deeper cross-border integration, and technology-driven demand flexibility. These dynamics increase extreme price events, elevate balancing costs, and challenge system stability. To address this, we propose the Market Security Index (MSI), an author-developed, composite metric to quantify market security (higher MSI = greater resilience) across EU bidding zones. MSI tracks resilience, benchmarks heterogeneous systems, and guides investments in flexibility, aligning with IOCEN 2026’s focus on sustainable energy systems, smart grids, and AI applications. Concept and Construction: The MSI integrates three pillars capturing short-term risk, operational stress, and system buffers: (i) Price risk—frequency and intensity of extreme price episodes and volatility in day-ahead (DA) and intraday (ID) markets; (ii) Balancing stress—costs of reserve activation and imbalance settlement relative to system load; (iii) Buffer capacity—availability of interconnectors (net transfer capacity-to-load ratio), storage (batteries, pumped hydro), and demand-side response (DSR) activation during peak stress. Indicators are normalized, directionally aligned, and aggregated using equal weights, with robustness validated via principal component analysis and constrained optimization. Data and Empirical Strategy: We construct a multi-year, monthly panel dataset from ENTSO-E, ACER/CEER, and national transmission system operators, covering EU bidding zones. Control variables include RES shares, network constraints, interconnector outages, and weather (heating/cooling degree days, wind/solar output). Fixed-effects panel models estimate associations between MSI and market outcomes (balancing costs, extreme price incidence), while quantile regressions test stability across distributions. Case studies, such as the 2025 Iberian blackout, validate MSI’s responsiveness to shocks, serving as plausibility checks. Findings: Higher MSI consistently correlates with lower balancing costs and fewer extreme price events. Key insights include the following: (i) Zones with higher interconnector capacity (NTC-to-load) mitigate price shock transmission to balancing costs, supporting cross-border buffering (H3); (ii) DSR activation narrows DA–ID price spreads and reduces imbalances, especially in high-RES systems (H4); and (iii) Storage attenuates positive price spikes, with stronger effects in RES-heavy zones (H5). These findings are robust to alternative normalizations and weather/network controls, with shock-period MSI behavior aligning with operational narratives. Policy Implications: MSI serves as a policy dashboard to benchmark zones, identify vulnerabilities (e.g., low buffers in RES-dominated systems), and prioritize cost-effective interventions like interconnector upgrades, storage deployment, or DSR expansion. It complements regulatory assessments of CfD/PPA schemes by highlighting the need for physical flexibility to stabilize short-term operations, supporting the EU’s Fit for 55 decarbonization goals. AI-Enabled Extensions: MSI incorporates random forest models to forecast demand and RES generation errors, enhancing MSI-F, a forward-looking variant for probabilistic imbalance risk and tail price event prediction. This aligns with IOCEN 2026’s AI focus, enabling scenario analysis for RES penetration and DSR adoption. Limitations and Future Work: The MSI is descriptive, not causal, requiring future quasi-experimental validation. Planned extensions include integrating electric vehicle flexibility, hydrogen storage, and network-aware spatial models. Contribution: The MSI offers a transparent, modular tool bridging market analytics and system operations. It enables regulators and operators to track resilience, compare zones, and allocate flexibility investments for maximum stability, advancing the EU’s sustainable energy transition.

1. Introduction

Magnesium hydride (MgH₂) is regarded as a solid-state hydrogen storage material with high potential. Its avantages include low cost, excellent reversibility of release/uptake of hydrogen and abundance on earth compared to other solid-state hydrogen storage materials. Unfortunately, MgH₂ released hydrogen at very high temperatures (> 400 °C for the pure MgH₂), sluggishly absorbed and desorbed hydrogen, and stable thermodynamic properties restricted commercialization of MgH₂ for hydrogen storage applications. To overcome this problem, many approaches have been studied, such as using high-energy ball milling to synthesize nanoparticles of MgH₂ using catalysts or additives that lower the energy barrier for Mg-H bond breaking and shorten hydrogen diffusion pathways, and mixing with other metals or hydrides (destabilized concept) that will alter the thermodynamic properties of MgH₂.

2. Methods

This study examines the impact of zirconium fluoride (ZrF₄) addition on the MgH₂ synthesized using ball milling to improve hydrogen storage properties. The prepared 10 wt.% ZrF₄-doped MgH₂ sample was characterized by X-ray diffraction, pressure–composition–temperature (PCT) curves and differential scanning calorimetry (DSC).

3. Results

From the PCT characterization, the 10 wt.% ZrF₄-doped MgH₂ sample demonstrated improved hydrogen storage properties relative to as-milled MgH₂, with the onset decomposition temperature reduced from 340 °C to 220 °C. The absorption/desorption kinetics measurements proved that the sample of 10 wt.% ZrF₄-doped MgH₂ can release and uptake hydrogen much faster than the pure as-milled MgH₂. From the DSC measurement at different heating rates, the apparent activation energy for hydrogen release from 10 wt.% ZrF₄-doped MgH₂ is reduced significantly compared to the undoped MgH₂.

4. Conclusions

The improved hydrogen storage properties of MgH₂ in the presence of ZrF₄ are thought to be due to the function of in situ produced active species that form during the dehydrogenation process. This active species plays a critical role in catalyzing the hydrogen release and uptake in the ZrF₄-doped MgH₂ composite system.