Hydrogen: Where it Makes Sense — and Where It Doesn’t

The tale of two hydrogens: expensive buses, giant caverns, and policy loopholes

In the same news cycle, hydrogen looks simultaneously overhyped and indispensable. On one end, Aberdeen is retiring its 25 hydrogen double‑decker buses after a costly, high‑profile stumble. In Germany’s Düren district, a local hydrogen vision for buses and energy is drawing warnings of stranded assets as costs and utilization fall short. On the other, Utah’s ACES Delta hub is approaching full operation with 220 MW of electrolyzers and salt‑cavern storage reportedly capable of holding two to three times the energy of all U.S. grid‑connected batteries combined—pointing to a real, system‑level role for hydrogen as long‑duration storage. Meanwhile, Europe’s renewable fuel mandates risk being satisfied “on paper” through accounting pathways that don’t actually put hydrogen into vehicles.

Taken together, these stories offer more than contradiction: they offer a decision framework. Hydrogen can deliver net system benefit in specific niches. It also creates fiscal and emissions liabilities when deployed where direct electrification is cheaper, simpler, and cleaner. Policymakers, investors, and planners need a practical way to tell the difference—and to hard‑wire safeguards that avoid the next Aberdeen.

What the bus failures are telling us

Hydrogen buses have struggled for three structural reasons: end‑use inefficiency, low utilization of bespoke infrastructure, and fuel‑supply risk.

• End‑use efficiency gap: A typical hydrogen fuel‑cell bus consumes around 8–10 kg H2 per 100 km. At delivered prices of €8–12/kg (common in Europe in recent pilots), that’s €0.64–1.20 per km for fuel alone. A battery‑electric bus averages roughly 1.2–1.4 kWh/km. At €0.15/kWh wholesale‑plus‑distribution cost, that’s €0.18–0.21 per km—three to five times cheaper to “fuel.” Maintenance tends to be lower for BEBs as well. Capital costs for hydrogen buses and refueling stations remain materially higher, with vehicle premiums of 30–60% versus BEBs and multimillion‑euro station costs that are hard to amortize without very high throughput.

• Utilization and the tyranny of fixed costs: Refueling stations and electrolyzers are capital‑intensive. If a depot dispenses a few hundred kilograms per day instead of the thousands assumed in pro formas, the cost per kilogram spikes. Aberdeen’s 25‑bus fleet never delivered the steady, high‑load demand profile the station needed; downtime and spares compounded the problem. Düren’s risk profile looks similar on paper: high capex, uncertain offtake, and limited ability to repurpose assets if economics sour.

• Supply‑chain fragility: Green hydrogen requires cheap, clean electricity for a large share of hours. Where low‑cost power is scarce or grid carbon intensity is high, delivered H2 is expensive and may not be very green. Trucked hydrogen adds logistics cost and vulnerability.

The lesson isn’t “hydrogen doesn’t work”—it’s that public transit presents all the worst boundary conditions for it. Buses are a solved problem for batteries: predictable routes, depot charging, regenerative braking, and falling pack prices. Hydrogen stacks the deck against itself in this segment.

The system case for (and against) hydrogen storage

Hydrogen’s appeal for grids lies in its unique storage scale and duration, not round‑trip efficiency. Electrolysis to storage to power has a round‑trip efficiency of roughly 25–40% (65–75% electrolysis efficiency, 90–95% storage/compression, 50–60% turbine or fuel‑cell conversion). By contrast, lithium‑ion batteries deliver 85–92% round‑trip.

So why build hydrogen storage at all? Because for multi‑day to seasonal balancing, cost per unit of stored energy becomes decisive. Salt caverns can store energy at single‑digit dollars per kWh of capacity, orders of magnitude cheaper than adding the equivalent battery energy capacity. ACES Delta leverages exactly this: large‑scale electrolysis linked to massive, low‑cost subsurface storage. Public figures indicate the project targets around 100+ metric tons of hydrogen production per day at full operation, with caverns sized in the hundreds of GWh of chemical energy—enough to firm long lulls in wind and sun.

When does that make sense?

• Where there is suitable geology (salt caverns) and a firming need that’s truly multi‑day to seasonal.
• Where very low‑cost, low‑carbon electricity is available a large share of the time (e.g., high renewable penetration with significant curtailment or dedicated co‑located renewables).
• Where there is a guaranteed offtaker (e.g., hydrogen‑capable turbines) that can run at high value during scarcity events.

Where it doesn’t: using hydrogen for daily arbitrage or as a first‑line storage option on a grid that can meet most variability with demand response, transmission, pumped hydro, thermal storage, and batteries. In those systems, hydrogen’s conversion losses impose unnecessary costs.

Policy design: avoid decarbonization on paper

Europe’s forthcoming renewable fuel mandates for transport (including RFNBOs) risk being met through accounting that never puts hydrogen‑derived fuels into vehicles. For instance, refineries can substitute electrolytic hydrogen for grey hydrogen in fuel production and claim compliance, even if road transport itself doesn’t decarbonize. Book‑and‑claim or mass‑balance approaches can create double‑counting risks if the same “green molecule” is used to satisfy multiple objectives.

There is climate value in cleaning up industrial hydrogen. But if the policy intent is to decarbonize transport, credits must reflect real delivered fuel use in vehicles, verifiable emissions reductions, and additional renewable generation. Otherwise, mandates backstop stranded refueling assets and keep legacy fossil infrastructure running longer.

Key guardrails include:

• Additionality: new renewable generation dedicated (physically or contractually) to H2 production.
• Temporal matching: hourly correlation between renewable production and electrolysis to avoid piggybacking on a fossil‑heavy grid mix.
• Geographic correlation: same bidding zone or grid region to reflect real system impacts.
• No double‑counting: one green claim per physical use, with clear registry rules.

A practical decision framework: five gates before hydrogen

Before committing public money or long‑lived assets, run every hydrogen proposal through these gates. If it fails any one, pause or pivot.

1. End‑use efficiency gate

• Question: Is there a commercially viable direct‑electrification alternative?
• Rule of thumb: If electric options deliver equal service at <50% of the energy cost per km/tonne‑km/MWh and fit operationally, choose electricity.
• Likely pass: iron/steel DRI, ammonia and other hydrogen‑based chemicals, some high‑temperature process heat (>600°C), long‑haul maritime fuels (ammonia/methanol), synthetic aviation fuels.
• Likely fail: city buses, passenger cars, residential heating via hydrogen blending, most local trucking.

2. Utilization and scale gate

• Question: Will the asset run hard enough to amortize capex?
• Rule of thumb: Electrolyzers need >50% capacity factor with power <€20–30/MWh to approach €2–3/kg LCOH; refueling stations should be sized for >70% of expected peak throughput and anchored by firm contracts.
• Red flag: Pilots with thin fleets (dozens of vehicles) supporting bespoke stations.

3. System value gate

• Question: Does the project solve a problem other tools can’t solve more cheaply?
• Rule of thumb: Use hydrogen storage when duration needs exceed ~24–100 hours and there’s suitable geologic storage; for shorter durations, prioritize batteries, pumped hydro, thermal storage, and demand flexibility.
• Bonus: Co‑benefits such as oxygen/heat valorization and grid‑services revenue streams.

4. Supply‑chain and siting gate

• Question: Is there low‑cost, low‑carbon power and logistics to deliver H2 at target cost?
• Rule of thumb: Co‑locate electrolysis with dedicated renewables and, where possible, storage caverns; avoid trucking hydrogen long distances; design modular plants that can be redeployed.

5. Accounting integrity gate

• Question: Will claimed emissions reductions stand up to audit?
• Rule of thumb: Enforce additionality, hourly matching, and single‑use crediting; publish metered data; tie subsidies to verified kg and carbon intensity.

Safeguards to prevent stranded assets

• Stage‑gated capital and off‑ramps: Release public funds in tranches tied to utilization, delivered cost, and emissions‑intensity milestones. Pre‑agree off‑ramps if thresholds aren’t met.
• Open‑access and repurposability: Where possible, make H2 production and storage hubs multi‑user and design pipelines, compressors, and power interconnections that can be repurposed (e.g., for industrial H2 if mobility demand disappoints).
• Anchor‑tenant contracts: Secure long‑term offtake (5–10 years) with creditworthy industrial users or grid operators before building large assets.
• Competitive alternatives test: Require a formal options analysis that quantifies total system cost versus electrification. If BEVs or heat pumps win on a 10‑ to 15‑year net present cost and emissions basis, they should prevail.
• Transparency and verification: Mandate third‑party MRV for carbon intensity at the project level and publish performance data. Tie incentives to performance, not nameplate capacity.
• Local air‑quality and equity lens: If hydrogen is used for backup or peaking power, assess local NOx and noise impacts and compare with cleaner alternatives.

Where to point hydrogen capital now

• Industrial feedstocks: Replace grey hydrogen in ammonia, methanol, and refining, prioritizing plants with access to low‑cost renewables and, where relevant, salt‑cavern storage to smooth operations.
• Green steel: Direct‑reduced iron with hydrogen at sites co‑located with high‑capacity‑factor renewables.
• Long‑duration storage hubs: Geologies with salt caverns and high renewable build‑out (e.g., U.S. Interior, parts of the Middle East, North Sea basin) to backstop seasonal variability and enable higher VRE shares.
• Maritime fuels: Ammonia or methanol e‑fuels for deep‑sea shipping on defined corridors with bunkering infrastructure and credible fuel‑quality/safety frameworks.
• Aviation e‑fuels: Power‑to‑liquids for synthetic kerosene at locations with exceptional renewable resources and CO2 supply, recognizing higher cost but scarce alternatives for long‑haul.

Cautious or avoid for now:

• Urban buses and light‑duty vehicles: Batteries dominate on cost and performance.
• Residential and commercial heating: Heat pumps and district energy are more efficient; hydrogen blending delivers little decarbonization per euro.
• Daily grid cycling: Batteries, demand flexibility, and thermal storage outcompete hydrogen on round‑trip efficiency and cost.

The bottom line: hydrogen as a scalpel, not a sledgehammer

Aberdeen’s buses and Düren’s ambitions warn against using hydrogen where electrification already wins. ACES Delta illustrates hydrogen’s unique leverage when the goal is seasonal energy shifting at immense scale. Europe’s RFNBO pathway debate shows how policy can either sharpen or blunt the tool through accounting.

A good rule for the 2020s: treat hydrogen as a scarce, strategic vector. Spend it where its physics align with the problem—molecules for molecules, long duration for long gaps, heat for high temperatures—and insist on utilization, additionality, and verifiable emissions cuts. That’s how to avoid stranded assets and build the parts of the hydrogen economy that the clean‑energy system actually needs.