R&D tax incentives for energy and cleantech

Energy and cleantech development sits comfortably inside the definition of R&D in all seven schemes: a new cell chemistry, an inverter control strategy under fault conditions, a process route for a fuel. The science is rarely in doubt. What sets the sector apart is how the work is paid for and what it is done with. Projects commonly carry government grants and co-funding, and each country treats that money differently, from removing the funded costs to clawing back the benefit to ignoring the grant. The work is also capital intensive, so test rigs and pilot plants raise equipment questions a software claim never meets.

This guide follows the shape of the manufacturers guide, adds pilot and demonstration plants, and ends with each country's rule on grants.

What this guide covers

• Where technical uncertainty sits in materials, storage, power electronics and process routes.
• What to exclude, including installation of proven systems, compliance testing and exploration.
• Where the qualifying work ends in a pilot or demonstration plant.
• Evidence from test rigs and funder reports, and the equipment and energy costs often missed.
• Each country's rule on grants and co-funding.

Where to look

Materials and electrochemistry

Electrode materials, catalysts, membranes and electrolytes where a performance, degradation or cost target had not been reached by known compositions.

Storage systems

Battery management, thermal management and pack design where the combined behaviour of cells, cooling and control under real duty cycles could not be predicted and had to be tested. Assembling a pack from a supplier's reference design is not.

Power electronics and grid integration

Converter topologies and control strategies that had to hold stability or ride through faults in a configuration the standards and the vendors did not cover, established by simulation and hardware testing.

Fuels and process routes

Electrolysis, synthesis and conversion processes where yield, purity or efficiency at scale was uncertain, and scale-up from bench to pilot produced behaviour that had to be investigated. Generation components and building energy systems follow the same pattern: a new geometry, material or combination proven by test rather than by established design data.

Monitoring, forecasting and optimisation software follows each country's software rules as well; the industry examples module shows where the software boundary sits.

What to exclude

• Installation, commissioning and operation of proven systems to the manufacturer's specification.
• Site selection, feasibility, planning and permitting.
• Compliance testing and certification.
• Prospecting, exploring or drilling for minerals, petroleum or natural gas, which several schemes exclude expressly.
• Demonstration of commercial viability once the technical questions are answered.

Pilot and demonstration plants

A pilot plant can be the experiment, the place where the experiment happens, or a commercial asset, and the schemes treat each differently. Where the uncertainty can only be resolved at scale, building and running the pilot is part of the work: Australia treats building a prototype or pilot plant as eligible where it supports the experiment, or as part of the experiment where the prototype is its subject, and the United Kingdom treats prototypes as R&D where they resolve uncertainty, but not a first of class built for sale beyond the point the uncertainty is resolved. Where the plant also produces for sale or demonstrates viability, several schemes stop: Ireland excludes pilot plants that are not wholly and exclusively undertaken for R&D purposes, and New Zealand excludes preproduction activities such as demonstration of commercial viability, tooling up and trial runs. Where output is sold, Australia applies a feedstock adjustment whose feedstock inputs include energy, New Zealand allows materials that become part of products sold only to the extent the cost exceeds the amount derived, and Canada applies recapture rules. Record the date on which the technical questions were answered; the claim ends there even if the plant runs on.

Evidence in energy and cleantech

• A dated note at the start of each technical unknown stating the target, why existing data did not predict the result, and who judged that.
• Test rig logs, cycle test data and performance curves against the predicted values, including the failures.
• Failure analyses and the design changes that followed them.
• Pilot plant run logs with parameters, output and the question each run addressed.
• Grant milestone reports, which already describe the work but are written for a funder, so add the uncertainty framing a reviewer needs.
• Funding agreements and customer contracts, read for who bears the risk and owns the results.
• Time by engineer by project, captured monthly at the latest.

Costs often missed

Cells, materials and components destroyed in testing are eligible costs in most schemes. Test equipment is capital and the treatment varies widely; energy consumed in testing is a consumable in some schemes and an overhead in others.

Getting started

List the year's projects with their funding sources and the equipment used, then run a project identification workshop on the technical content. For each project that survives the uncertainty test, model the grant rule before estimating the benefit, because the same grant can leave the claim untouched in one country and remove most of it in another. The finance module on grants works through the mechanics.

Illustrative example

Illustrative example. Kestrel Grid Systems Ltd, a hypothetical company, spent 500,000 on a battery thermal management project to keep cell temperatures within a stated band under a fast charging duty cycle that the supplier's reference design could not sustain. The engineers did not know whether any coolant path and control strategy could hold the band within the pack's weight limit. They built three rig configurations, cycled 200 cells to failure, rejected two configurations on temperature spread and fixed the third, at which point the work ended. The spend was 300,000 of engineering time, 120,000 of cells and materials destroyed in testing and 80,000 of test chamber costs. A government agency grant of 200,000 covered 40 percent. A further 50,000 spent installing a proven inverter at the demonstration site is excluded in every scheme. The grant now decides the claim. In a country that removes expenditure met by grant assistance, the candidate base falls to 300,000. In a country that applies a clawback, the base stays at 500,000 and an adjustment later removes the premium benefit on the 200,000 recouped. In the United Kingdom, the merged scheme contains no restriction for subsidised or grant-funded expenditure for accounting periods beginning on or after 1 April 2024, so the base stays at 500,000. The benefit at that country's rate applies to the resulting base.

Country differences

One rule per country, in the scheme's own wording, on grants and government co-funding. The country pages hold the detail.

Related pages

• Guide: R&D tax incentives for manufacturers
• Technical Teams Program, Module 5: Industry examples
• In-House Finance Team Program, Module 5: Grants and subsidised expenditure
• Startups and Scale-ups, Module 3: Funding interactions
• Guide: qualifying versus non-qualifying activities
• Resource hub: eligibility

Educational material, not advice. Eligibility and the treatment of grants depend on the scheme, the funding agreement and the facts.