IARPA Wants Drone Engines That Rot: New RFI Targets Bio-Derived Propulsion for Missions That Leave No Trace

The U.S. intelligence community is asking industry a question it has never asked out loud before: can a drone engine be grown instead of machined, and can it disappear after the mission ends? The Intelligence Advanced Research Projects Activity (IARPA) published RFI IARPA-RFI-26-01 on April 20, 2026, seeking information on bio-derived materials for transient unmanned aerial vehicle (UAV) propulsion systems, with responses due May 15, 2026. The request was first reported by Colton Jones at Defence Blog.

The request targets turbines, motors, and engine components that degrade through environmental triggers beyond ultraviolet light or water exposure, the two mechanisms defense researchers have relied on for the past decade. The framing is a direct admission that the previous generation of vanishing-drone programs stopped at the airframe. IARPA now wants the parts that actually generate thrust to disappear too. In nine years covering military drone R&D, this is the first time I have seen an intelligence agency treat biomanufacturing as a propulsion problem rather than a structural one.

The Numbers IARPA Wants Bio-Materials To Hit

The RFI sets specific survival thresholds respondents must address: operational temperatures above 500 degrees Celsius (932 degrees Fahrenheit) and mechanical stress above 100 megapascals. Those are real turbine and electric-motor numbers, not lab-bench targets. The agency also wants answers on manufacturing tolerance and scalable production at defensible unit cost.

The candidate material list is where the document gets interesting. IARPA names structural proteins like silk and collagen, polysaccharides like chitin and cellulose, mycelium-based composites, bio-acrylics, and bio-derived ceramics. Genetically modified structural proteins with tailored mechanical properties and enzymatically sensitive polymers with programmable degradation rates are called out as specific examples worth investigating.

The degradation triggers the agency is asking about are also telling. IARPA wants mechanisms that respond to enzymatic activity, microbial action, or oxidation. An engine that only degrades under direct sunlight is useless inside a forest canopy or a collapsed building. An engine that decays from exposure to soil microbes or ambient humidity works anywhere the mission ends.

Why ICARUS Stopped At The Airframe

The precedent for this work is DARPA’s Inbound, Controlled, Air-Releasable, Unrecoverable Systems program, better known as ICARUS. Launched in 2015 with $8 million over 26 months, ICARUS showed that structural drone components could be built from ultraviolet-triggered photopolymers that degrade on command. San Francisco-based Otherlab built the APSARA glider under that contract, a cardboard-and-mycelium delivery vehicle carrying a 1-kilogram (2.2-pound) payload and vanishing within hours of landing.

ICARUS worked because airframes are passive. A propulsion system is not. Turbine blades spin at tens of thousands of RPM inside enclosed combustion chambers, and UV light does not penetrate a sealed engine housing. That is the exact gap IARPA is now trying to close. The RFI acknowledges the difficulty directly, noting that large gaps remain between laboratory demonstrations and operational propulsion systems. Nobody has built a biological turbine yet, and IARPA knows it. The agency wants to find out who is closest.

The Intelligence Use Case Is Forensic Denial

The operational logic is straightforward. Intelligence missions increasingly happen in denied environments where recovery of a downed aircraft is impossible. A drone that finishes its mission but leaves a metal turbine in a field gives an adversary hardware to reverse-engineer and a serial number to leak. A drone whose engine decays into organic residue within days offers neither. Forensic denial becomes a design specification rather than an afterthought.

This matters now in a way it did not in 2015. Russia’s Shahed drones already use a wooden propeller and a basic four-cylinder Chinese engine. The plywood-and-foam Gerbera decoy that breached Polish airspace last September costs a fraction of a cruise missile. The low-cost expendable drone era is here, and its building blocks are already partially biological. IARPA is trying to militarize the degradation side of that trend before someone else does.

A counter-argument is worth naming. Bio-derived materials introduce supply chain fragility that metal alloys do not. Silk feedstock and mycelium cultures depend on biological pipelines that can fail in ways a titanium billet cannot. A program that trades forensic denial for brittle production runs may not survive contact with procurement reality. IARPA’s explicit request for scalability and cost data suggests the agency already sees this tension.

DroneXL’s Take

Biodegradable drone research has moved from hobbyist framing to intelligence-community priority in two years. DroneXL covered bamboo drones with open-source flight control software earlier this month out of China. IARPA-RFI-26-01 is the first time in this cycle that an intelligence agency has said the quiet part on the record: we want engines we can throw away on purpose. The Ukraine connection makes this feel overdue. I have written roughly a dozen pieces over the last eighteen months about wooden Russian drones and plywood decoys. The battlefield lesson is consistent: cheap and organic beats exquisite and recoverable in attritional drone warfare. IARPA read the same lessons.

Here is the call. IARPA will receive enough credible white papers by the May 15 deadline to justify a full Broad Agency Announcement by the end of 2026, with initial Phase 1 contracts awarded in the first half of 2027. The first award will go to a consortium pairing a synthetic biology firm with an established propulsion house, not to a pure biotech player. The technical risk is too high for a single-discipline team. IARPA’s track record on programs like Great Horned Owl and Little Horned Owl shows the agency prefers coupled teams with one partner who already knows how to build a flying thing that works.

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.