Braided CF/LMPAEK tube system reduces liquid hydrogen fuel line weight by 50-60%

Hydrogen (H₂) gas liquefies at a temperature of -253°C (20.28 K) under atmospheric pressure, just 20 degrees above absolute zero. This temperature is cold enough to make most structural materials brittle and H₂, among the smallest molecules that exist, are small enough to find any gap in a material and permeate straight through. For ground-based cryogenic infrastructure, those challenges are manageable: stainless steel and vacuum-jacketed lines are bulky and heavy, but when weight is not a constraint, they’re acceptable. 

Put those same requirements into a commercial aircraft and the equation changes entirely. A liquid hydrogen (LH₂) fuel cell-powered passenger aircraft must route LH₂ from the tank to the fuel cell through a fuel distribution system light enough to be viable, while surviving more than 10,000 thermal cycles over a 25-year service life, with each flight heating everything back to ambient before the next cryogenic soak begins. NASA (Hampton, Va., U.S.) research has shown that without adequate insulation, 50-70% of LH₂ can boil off in flight, a figure that makes H₂ aviation commercially unworkable if the fuel system is not designed correctly from the outset.

Conventional metallic cryogenic lines accommodate thermal contraction from this thermal cycling with bellows, O-rings, bolted flanges and mechanical seals; components that also multiply the number of potential leak points in the system. In a ground application, leaks are not desirable, but H₂ will rise and diffuse quickly in air, for example. In an occupied aircraft carrying a cryogenic, highly flammable fuel, every joint is a liability that both the designer and the regulator have to account for. The conventional metallic approach, borrowed from industrial cryo-technology, simply was not built with that constraint in mind.

Foundational TPC cryogenic design

Dresden-based herone GmbH (Germany) has spent the last several years re-engineering the design of cryogenic fluid lines from first principles, specifically for the aerospace operating environment, within the German government-funded LuFo projects WAKOS and ZEDI.

The company, founded in 2018 as a spin-off from TU Dresden’s Institute of Lightweight Engineering and Polymer Technology (ILK), has built its technology on a decade of research into thermoplastic composite (TPC) hollow profiles originating from co-founder Dr. Christian Garthaus and Dr. Daniel Barfuss’ doctoral work at ILK. That foundation produced herone’s patented continuous blow molding and injection forming processes, in the context of unitized thermoplastic driveshaft and gear demonstrators. The company focuses on carbon fiber-reinforced low-melt polyaryletherketone (CF/LMPAEK; LMPAEK is from Victrex, Clevelys, U.K.) and polyetheretherketone (PEEK) composite hollow profiles, creating a material system that offers 50-60% weight savings over stainless steel and a set of physical properties that make it unusually well suited to the demands of cryogenic H₂ applications. This material focus is the foundation of the company’s LH₂ fuel line system design.

That foundation has already produced flight hardware. Working with ArianeGroup GmbH (Bremen Germany) within the European Space Agency’s (ESA) Future Launchers Preparatory Programme (FLPP), herone recently completed the first full-scale CF/LMPAEK cryogenic line system component for the Ariane 6 launch vehicle with a near-net-shape, tape-preformed, out-of-autoclave (OOA)-consolidated assembly with integral thermoplastic fittings co-consolidated with the tubing in a single step, designed for the pressure loads and cryogenic conditions of launch vehicle service. This space application demands minimized mass above all else, and accepts a single-wall line design where brief mission durations and overboard venting manage any residual leakage risk. 

Aviation, by contrast, demands something considerably harder to achieve: a double-wall system with vacuum insulation, secondary containment and leak rates low enough to be safe in an occupied vehicle over thousands of flight cycles. The Ariane 6 component demonstrates the manufacturing process works at full scale; the aviation program is where the engineering requirements become genuinely uncharted.

Thermoplastic vs. thermoset matrix choice

To understand herone’s approach, it helps to think about what happens to a composite material during repeated cryogenic cycling. Epoxy-based thermoset composites behave, in a sense, like glass under those conditions: rigid and capable in normal service, but at temperatures approaching cryogenic (below -150°C), the brittleness latent in the material becomes structural. Under thermal cycling, matrix microcracking can initiate and propagate through the laminate. Each crack is a potential H₂ leakage pathway, not because the laminate has failed structurally, but because H₂’s molecular diameter is small enough to migrate through cracks that many structural assessments would dismiss as negligible.

The PAEK family of thermoplastics behave differently. “Think of them as the flexible polymer bottle rather than the glass bottle. They retain ductility at cryogenic extremes that thermosets lose,” explains Daniel Barfuss, co-founder and managing partner at herone. “When things get really cold, almost all materials become more fragile, and we need materials to be flexible enough to prevent tiny cracks which can lead to leaks. That’s why thermoplastics are valuable here.”

PEEK maintains approximately 3-4% elongation at break at -196°C (77 K) — the boiling point of liquid nitrogen (LN₂) and the standard temperature used in cryogenic materials characterization, representing a conservative proxy for the -253°C LH₂ service condition — compared to roughly 1.5% for glass fiber/epoxy systems. The standard cryogenic test temperature of 77 K is used in materials characterization because that cryogen is readily available in any laboratory, making it a practical and reproducible initial benchmark for assessing how materials behave, even when the actual service temperature, as in LH₂ applications, is colder still at -253°C.

The retained flexibility that PAEK polymers offer is the difference between a laminate that resists microcracking under thermal fatigue and one that does not. TPC in general also demonstrate significantly higher mode I interlaminar fracture toughness — the energy per unit area required to pry two bonded composite plies apart by opening them like a book — approximately five times greater than thermoset composites. As such, cracks not only initiate less readily, but require substantially more energy to propagate once they do.

Permeation is a separate but related problem. Even without cracking, H₂ diffuses through composite laminates under a concentration gradient. CF/LMPAEK laminates provide approximately 10 times lower H₂ permeability than epoxy systems at cryogenic temperatures, and at the -253°C of LH₂ service, permeation through the composite wall itself becomes negligible. The critical window is at ambient temperature during ground handling, refueling and warm-up phases, where a barrier layer is still required. 

Rather than applying a liner as a secondary postprocess step, herone integrates a metallic film permeation barrier directly between braided layers during preforming. The thermoplastic-functionalized barrier layer becomes part of the tube wall, co-processed into the structure, maintaining the homogeneity of the composite cross-section and avoiding the bonded interface that a separately applied liner creates. 

“When you use a high-quality thermoplastic and achieve a good molding surface, you get a resin-rich outer layer with no exposed fibers,” says Barfuss. “That surface seals. You don't need metal to do it, you just need resin. That's one of the things people don't expect thermoplastics to be able to do.”

Eliminating the joints

The materials herone has chosen resolve the key microcracking and permeation problems, but the deeper engineering question is structural: How do you build a cryogenic aircraft fuel line without the bellows, O-rings and bolted flanges that make conventional metallic assemblies so joint-heavy?

The answer lies in what PAEK TPC enable at the manufacturing level that thermoset composites do not. Because TPC can be reheated and reformed after initial consolidation, herone injection-forms or co-consolidates functional elements (flanges, fittings, ferrules, sealing surfaces) directly onto the composite tube body in a single integrated manufacturing sequence. Short fiber-reinforced PEEK is co-consolidated at 380°C with preheated PAEK preforms held at approximately 200°C, creating simultaneous cohesive molecular bonding at the polymer interface and geometric interlocking at the macro-scale. This produces what herone terms a “form-locking joint” — a connection that achieves 44% higher torque load capacity than cohesive bonding alone, without adhesives, fasteners or elastomeric seals. The flange is not attached to the tube; it is the tube, formed from the same material system and bonded at the molecular level.

“The co-consolidation is a technique that eliminates the need for postprocessing joining operations and additional joining specifications, such as knock-downs,” says Barfuss. “This process is inherently integrated into the fundamental consolidation specifications of the composite material itself. As a result, co-consolidation achieves shear design values that are three to four times greater compared to traditional metallic-composite adhesive bonding methods.”

For an aviation LH₂ application, herone is developing a double-wall configuration: a composite inner tube carrying the LH₂, separated from a composite outer containment tube by a vacuum-insulated annular gap maintained by 3D printed polymer spacers. The vacuum interspace provides thermal insulation which is critical to minimizing boil-off over flights lasting up to 5 hours and simultaneously acts as secondary containment if the inner line develops a leak. An interspace monitoring capability provides early detection before any failure cascades to the outer wall. 

By designing both walls from the same CF/LMPAEK material system and braiding each with independent laminate layups, herone independently tunes the coefficient of thermal expansion (CTE) of each tube. A near-zero axial CTE laminate on the inner tube suppresses axial contraction during cooldown. A matched CTE design between inner and outer walls removes the differential movement that conventional lines manage with bellows. Eliminating bellows reduces the joint count, system weight and the potential leak-point inventory simultaneously.

Braiding to functional assembly

The company’s manufacturing sequence begins with automated tape braiding. Victrex tape AE250 supplied in LMPAEK-matrix form with PAEK-compatible sizing that delivers 20% higher fiber-matrix adhesion than unsized fibers, are braided over a mandrel by a robotic system that controls feed rate, braid angle and layer sequence. Braid angles from ±15° to 70° are selectable as well as pure 0° layer integration, enabling the laminate architecture to be tuned for each application — including specific multi-axial angles for CTE management and higher helical angles for hoop-stress capacity under internal pressure. For curved sections, the mandrel geometry routes the tube through bends with radii more than twice the diameter without fiber wrinkling, a direct advantage of the TPC tape architecture over dry fiber braiding that must be consolidated separately.

Following the completion of the braided preform, including the metallic barrier film positioned between designated laminate layers at this stage, the assembly transfers to a heated press. An internal inflatable bladder inserted through the tube bore applies radial consolidation pressure from inside the preform against the tool face while the  press assembly heats to processing temperature: 305-340°C for plies of carbon fiber-reinforced prepreg made with LMPAEK polymer 385°C for PEEK. This OOA consolidation produces void content below 2% in approximately 15 minutes, compared to 240 minutes for autoclave-cured thermoset prepregs. 

The company’s Dresden facility, representing more than €4 million in production investment, targets 20,000 parts annually, and that throughput only makes sense at 15-minute consolidation cycles. The metallic barrier film is positioned between braided layers before consolidation and thermally fused into the finished wall during the same press cycle, requiring no separate process step.

For field assembly, herone has also developed a PEEK-based electrofusion socket system: a resistance-heating element embedded in a thermoplastic sleeve heats the joint to fusion temperature when energized, welding two line sections together on-site without additional tooling or external heat sources. 

“We’ve developed a joining approach that works like plumbing,” says Barfuss. “You bring the socket to the site, clip on a simple electrical connection, and the heat does the rest. No tooling, no external press; the joint fuses in place.”

This concept brings the simplicity of established plumbing-industry joining techniques to enable reliable fusing of pipe sections to an aerospace-grade composite cryogenic line.

Making a case for CF/LMPAEK in aerospace programs

Micrograph analysis of CF/LMPAEK laminate specimens subjected to extended cryogenic thermal cycling shows no measurable microcracking in either flat coupons or tubular geometry. This represents the critical distinction from thermoset composite test results under equivalent conditions. Permeation measurements on cycled and uncycled specimens confirm that barrier-integrated laminates meet aviation LH₂ service requirements. The material system carries qualification data from broader aerospace programs under PAEK-class material approvals (see sidebar), and herone holds AS/EN9100 manufacturing certification. 

Compared to aerospace-grade stainless steel, the CF/LMPAEK tape-braided tube assembly is projected to reduce line system weight by 50-60%. Additional savings from integral CF/LMPAEK flanges displace separate metallic flange hardware, as flanges account for roughly one-third of a metallic line assembly’s total mass.

The technology currently sits at technology readiness level (TRL) 3 for the aviation double-wall configuration, with TRL 6 targeted within the coming year. No established certification standards yet exist specifically for LH₂ piping in passenger aircraft; EASA CS-25 specifications are being adapted, and the FAA’s December 2024 Hydrogen-Fueled Aircraft Roadmap sets development targets through 2028 and 2032. Nonetheless, the failure behavior of herone’s CF/LMPAEK-based tube assemblies aligns well with the regulatory intent of those frameworks. Unlike metallic lines that can fail suddenly under overpressure, the TPC tubes fail first in the polymer matrix, producing slow localized leakage detectable through the interspace monitoring system before any structural event occurs. That failure-mode predictability is an engineering argument for CF/LMPAEK as much as a safety one.

The design’s thermoplastic matrix also closes the sustainability case. Because LMPAEK can be remelted, production off-cuts and end-of-life components can be reprocessed into chopped TPC feedstock, avoiding the landfill destination typical for thermoset composite scrap. For an industry beginning to treat circular economy obligations as genuine design constraints rather than compliance exercises, that reprocessability matters.

“We’re not just replacing metal with composites,” Barfuss notes. “We’re creating a system that aviation’s H₂ infrastructure can actually qualify, maintain and eventually recycle, and doing it in a way that can be manufactured at the rates the industry will eventually need.”