Thermoplastic AFP: The Next Frontier in Composite Manufacturing

2025 35 min read thermoplastic AFP · aerospace · composites

Thermoplastic AFP is moving from research labs to production floors. Here's what it actually is, why the industry is shifting, and what it takes to run it — with a clear-eyed look at what's solved and what isn't yet.

Thermoplastic AFP manufacturing

Why Thermosets Built the Industry — and Why That's Changing

Thermoset composites built the aerospace industry. Carbon fibre reinforced epoxy is what the A320 is made of, what the 787 is made of, what makes modern aircraft structures 20–30% lighter than aluminium equivalents. For fifty years, the combination of thermoset resin and autoclave processing has been the gold standard for structural composite manufacturing.

Thermoset vs thermoplastic manufacturing

But a technology reaching maturity is also a technology approaching its ceiling.

Autoclave processing is expensive, capital-intensive, and a hard constraint on production rate. Thermoset parts cannot be welded, reformed, or recycled without significant energy input and material loss. The skills required to lay up thermoset prepreg are expensive and increasingly scarce. And as sustainability regulations tighten, the end-of-life fate of thermoset composites — largely landfill or incineration — is becoming a procurement issue, not just an environmental one.

Thermoplastic composites change the equation on all of these dimensions. And thermoplastic Automated Fiber Placement (AFP) is the manufacturing process that makes them viable for complex structural geometries at production rates aerospace programs actually need.

This is not a future technology. The physics works. The properties are demonstrated. The question now is execution — process robustness, supply chain depth, and accessible hardware that makes thermoplastic AFP available below the tier-1 contractor level.

This post covers what thermoplastic AFP is, why it matters, what the engineering challenges actually are, and where the technology is heading.

The Material Science: What Makes Thermoplastics Different

At the material level, the difference between thermosets and thermoplastics is a single thermodynamic property. Thermosets cure through an irreversible chemical cross-linking reaction. Once the polymer network forms, it doesn't unmake. Thermoplastics, by contrast, soften above their glass transition or melting temperature and re-solidify on cooling. That phase transition is repeatable — and it is the foundation for everything that makes thermoplastic composites interesting.

Thermoset vs. Thermoplastic — Phase Behaviour
Uncured
epoxy + CF
Heat + time
Cured
cross-linked network
🔒
Irreversible Reaction
Cannot be remelted, rewelded, or thermally reformed after cure.
End of life: landfill / incineration.

Solid
PEEK + CF
Heat above Tm
Melt
viscous
Cool
Solid
PEEK + CF
↺ Reversible phase change
♻️
Reversible Phase Change
Re-weldable, reformable, and fully recyclable.
End of life: pelletise + reuse.
Thermoset
High fibre-volume fraction
Mature, established process
Long cure cycles (autoclave)
No rework or repair by heat
End-of-life: landfill / incineration
Thermoplastic
In-situ consolidation possible (AFP)
Short consolidation cycles
Weldable joints — no fasteners
Reformable / repairable by heat
End-of-life: pelletise + reuse

This reversibility is not a minor footnote. It is what enables fastener-free assembly via welding, field repair without an autoclave, and end-of-life material recovery — three capabilities that thermosets categorically cannot provide.

Matrix Selection: The Thermal Processing Window

Not all thermoplastics are equal in AFP applications. The processing temperature — the temperature at which the matrix becomes workable — determines the hardware specification required, the energy input needed, and the thermal stresses induced in the laminate during processing.

Thermoplastic Matrix Comparison for AFP
Ordered by increasing process temperature — hover each row for details
Matrix
0°C 100°C 200°C 300°C 420°C
PA12
Tg ~50°C · Tm 178°C
AFP 220–260°C
Tg
~50°C
Tm
178°C
AFP
220–260°C
Automotive
Low cost, automotive & industrial applications
PA6
Tg ~75°C · Tm 220°C
AFP 250–290°C
Tg
~75°C
Tm
220°C
AFP
250–290°C
Automotive
High toughness, automotive structural
PPS
Tg ~90°C · Tm 285°C
AFP 320–360°C
Tg
~90°C
Tm
285°C
AFP
320–360°C
Industrial
Chemical resistance, flame retardant
PEI (Ultem)
Tg 217°C · Tm N/A (am)
AFP 340–380°C
Tg
217°C
Tm
Amorphous — no distinct melt
N/A (am)
AFP
340–380°C
Aerospace
Semi-structural, aerospace interior
PAEK
Tg ~145°C · Tm ~380°C
AFP 360–420°C
Tg
~145°C
Tm
~380°C
AFP
360–420°C
Primary Aero
High Tg, primary aerostructure
PEEK
Tg ~143°C · Tm 343°C
AFP 370–420°C
Tg
~143°C
Tm
343°C
AFP
370–420°C
Benchmark
The aerospace benchmark for structural AFP
PEKK
Tg ~156°C · Tm ~340°C
AFP 350–420°C
Tg
~156°C
Tm
~340°C
AFP
350–420°C
Process Window
Slower crystallisation — wider process window
Glass Transition Temp (Tg)
Melt Temp (Tm)
AFP Process Temp Range
Tg = glass transition temperature · Tm = melting temperature (semi-crystalline only).
Amorphous polymers (e.g. PEI) soften above Tg without a distinct melt point. All bars scaled to 420°C maximum.
CF/PEEK unidirectional tape

CF/PEEK unidirectional tape — no freezer, no expiry, no autoclave.

PEEK & PEKK

Dominate aerospace primary structure — combining mechanical properties (CAI, OHT, bearing strength) with the thermal stability needed for aerostructure service environments.

PPS & PEI

Widely used for secondary structure, brackets, clips, and interior panels where the processing temperature is more forgiving and the structural demands are lower.

PA-Family

Driving automotive adoption — where cycle time and cost pressure mean PEEK-grade properties are rarely justified, but fibre-reinforced thermoplastics still deliver significant structural benefit.

In-Situ Consolidation: Where the Engineering Complexity Lives

The defining capability of thermoplastic AFP — and the source of most of its engineering challenge — is in-situ consolidation (ISC). In conventional thermoset AFP, you lay up the laminate and then autoclave it separately. In thermoplastic AFP with ISC, you melt the matrix, press it, and consolidate it in a single pass as the head traverses the layup surface. No secondary oven. No autoclave. One step.

Laser in-situ consolidation at the nip point

Laser in-situ consolidation at the nip point — tape fed, heated above Tm in milliseconds, and compacted in a single pass.

In-Situ Consolidation — The Nip Point in Detail
Four interdependent process variables must be precisely controlled simultaneously
Incoming Thermoplastic Tape
PEEK/CF · PEKK/CF · PAEK/CF
HEAT SOURCE
400–450°C at nip point
Heats tape + substrate above Tm
in milliseconds
Laser / hot gas torch / ultrasonic. Rapid, localised heating of the nip zone only.
NIP
◄─── Direction of travel ───►
COMPACTION ROLLER
50–500 N force · controlled temp
Drives out voids, ensures contact
Transfers heat to substrate
Silicone or metallic roller. Force profile controls intimate contact at the ply interface.
Consolidated Laminate
Bonds to previous ply as it cools
Target: semi-crystalline morphology
The Four Interdependent Variables
1
Temperature
Must reach Tm at the nip point for bonding to occur.
↓ Too low → voids, incomplete consolidation
↑ Too high → matrix degradation, porosity
2
Pressure
Drives intimate contact and void collapse at the ply interface.
↓ Too low → voids, delamination risk
↑ Too high → fibre wash, geometry distortion
3
Speed
Controls dwell time under the heat source at the nip zone.
↓ Too slow → excess crystallinity (PEEK)
↑ Too fast → insufficient melt depth
4
Cooling Rate
Determines the final crystalline morphology of the matrix.
↓ Slow cool → high crystallinity → brittle
↑ Fast cool → amorphous → reduced Tg
✓ Target: semi-crystalline, consistent

The Coupling Problem

These four variables are coupled. Changing layup speed changes dwell time and therefore heat input, which changes the required laser power, which changes the temperature gradient, which changes the cooling rate, which changes crystallinity. On a complex geometry with varying surface angles, the coupling changes continuously along the tow path.

This is why the process software is not a secondary concern in thermoplastic AFP — it is central to the process. The thermal model that predicts nip-point temperature as a function of speed, geometry, ambient temperature, and prior ply state needs to be embedded in the trajectory generator, not bolted on after the fact.

Heat Source Comparison: Laser, Hot Gas, and Ultrasonic

The choice of heat source is one of the most consequential decisions in thermoplastic AFP system design. Three technologies dominate:

Heat Source Comparison for In-Situ Consolidation
Three primary heat sources used in thermoplastic AFP — advantages & limitations
🔆
Diode Laser
  • Precise energy delivery
  • Fast response time
  • Narrow heat-affected zone
  • Enables high layup speeds
  • Highly repeatable
  • High equipment cost
  • CF absorbs >90% — fibre overheating risk
  • Requires safety interlocks
  • Line-of-sight only
🔥
Hot Gas Torch
  • Lower equipment cost
  • Good for complex geometries
  • Mature, established technology
  • Low safety overhead
  • Wider heat-affected zone
  • Slower thermal response
  • Less energy-efficient
  • Convective → diffuse heating
  • Less suitable for high speed
〰️
Ultrasonic
  • Localised heating
  • No line-of-sight constraint
  • Suitable for semi-crystalline TP
  • Enables welding applications
  • Limited to semi-crystalline materials
  • Complex head mechanics
  • Not yet at production TRL for AFP
Precision & Control
Diode Laser
9.3
Hot Gas Torch
5.2
Ultrasonic
7.0
Accessibility & Cost
Diode Laser
3.5
Hot Gas Torch
8.5
Ultrasonic
5.0
Production Readiness (TRL)
Diode Laser
9.0
Hot Gas Torch
8.0
Ultrasonic
4.0
85°C
PEEK processing window at nip point (343°C → 430°C)
100–300 mm/s
Laser AFP layup speeds for PEEK/PEKK
3 Types
Dominant heat source technologies in thermoplastic AFP

Laser AFP is the dominant choice for PEEK and PEKK primary structure applications. The precision of energy delivery matters when the processing window is narrow — PEEK needs to be above ~343°C to consolidate and below ~430°C to avoid degradation, a window of roughly 85°C at the nip point. Laser diodes hit that window repeatably at layup speeds of 100–300 mm/s.

AFP-XS on a KUKA arm with laser heating

AFP-XS on a KUKA arm — laser heating at the nip point, in-process inspection running, thermoplastic tows consolidating in a single pass

Hot gas is widely used in lower-temperature applications (PPS, PEI) and in facilities where the capital cost of a laser system is not justified. DLR's research programmes have demonstrated excellent ISC results with hot gas at moderate layup speeds.

The AFP Head: Mechanical Design Requirements at 400°C

Running an AFP head at thermoplastic temperatures puts demands on the mechanical design that don't exist in thermoset systems. A thermoset AFP head operates at ambient temperature — it's a precision tape dispenser. A thermoplastic AFP head is operating at temperatures that exceed the service limit of most standard engineering plastics, many bearing materials, and some aluminium alloys.

Thermoplastic AFP Head — Key Design Requirements
Three integrated subsystems managing a >300°C thermal gradient across ~50 mm
Spool (ambient)
🧵
Material Feed System
  • Tape guide + steering mechanism
  • Receives tape at 20°C ambient from spool
  • Delivers tape to nip point at ~400°C
  • Thermal gradient management — insulation between feed path and heat zone
🔆
Heat Source Mount
  • Laser optics or hot gas nozzle
  • High-temp materials: ceramics, Inconel
  • Precisely positioned relative to nip point
  • Water-cooled housing to protect robot wrist
⚙️
Compaction Roller
  • Material: PTFE, silicon carbide, or high-temp metal alloy
  • Spring-loaded or servo-controlled actuation
  • Force feedback for closed-loop process control
  • Must maintain consistent contact on curved geometry
Thermal Management Requirements
Nip Point
380–450°C
PEEK / PEKK processing window
🔴
Head Body
< 150°C
Structural integrity of head assembly
🟠
Robot Wrist
< 60°C
Standard robot flange limit
↕️
Temperature Gradient
>300°C across ~50 mm
Requires insulation + active cooling of robot-side components

This thermal management challenge is why thermoplastic AFP heads are not trivially adapted from thermoset heads. The engineering of the thermal boundary between the robot interface (cool) and the processing zone (extremely hot) is a significant mechanical design problem in its own right.

AFP-XS thermoplastic head

Robot-Agnostic Design

Addcomposites' approach on the AFP-XS is to design the head as a self-contained thermal unit that mounts on any standard industrial robot flange without requiring robot modifications. The thermal isolation is managed within the head, not by the robot — which is what makes the system robot-agnostic and deployable on existing KUKA, Fanuc, ABB, and UR platforms.

Unique Capabilities: What Thermoplastic AFP Actually Unlocks

Beyond removing the autoclave from the process chain, thermoplastic composites enable manufacturing approaches that are structurally impossible with thermosets. These are not incremental improvements — they are qualitatively different capabilities.

1. Structural Welding: Fastener-Free Assembly

Thermoplastic Welding Methods
Three techniques for joining TP-CFRP parts without adhesive or mechanical fasteners
Resistance Welding
Resistive mesh at interface
1
Parts clamped with mesh at interface
2
Current applied → interface heats above Tm
3
Pressure applied, current removed
4
Cool under pressure → welded joint
30s–5m
Cycle Time
80–95%
Parent Strength
Suitable for: discrete joint locations, lap joints, reinforcement attachment
🌀
Induction Welding
Susceptor at interface
1
Induction coil passes overhead along seam
2
Eddy currents heat susceptor; matrix melts at interface
3
Pressure from roller consolidates joint
4
Continuous seam weld — like a zip
Continuous
Weld Mode
Full-scale
Demonstrated
Long seam welds, skin-stringer joints. Demonstrated on full-scale TP fuselage panels (Airbus / Clean Sky 2)
〰️
Ultrasonic Welding
Sonotrode at joint
1
Sonotrode pressed onto joint location
2
High-frequency vibration at 20–40 kHz
3
Frictional heat generated at interface
4
Rapid cool → spot weld complete
< 5s
Cycle Time
Spot
Weld Mode
Spot welds, small bracket attachment, interior fittings
Comparative Performance
Resistance
Induction
Ultrasonic
Bond Strength
90%
88%
75%
Throughput Speed
Med
High
High
Process Maturity
TRL 7
TRL 6–7
TRL 5
Seam Length
Discrete
Long seam
Spot only
Equipment Cost
Med
Med-High
Low-Med
Ultrasonic Welding

Ultrasonic Welding (USW) — Image credit: MDPI, from Advanced Thermoplastic Composite Welding for Aerospace Applications

Induction Welding

Induction Welding (IW) — Image credit: MDPI, from Advanced Thermoplastic Composite Welding for Aerospace Applications

Resistance Welding

Resistance Welding (RW) — Image credit: MDPI, from Advanced Thermoplastic Composite Welding for Aerospace Applications

The welding capability is where thermoplastic composites generate the most commercial interest at the assembly level. A conventional composite aircraft structure is held together with thousands of titanium fasteners. Each fastener requires a drilled hole — a stress concentration in the laminate — plus the fastener weight, sealant, and installation labour.

Thermoplastic welding removes the fastener. The joint strength is carried by the material, not by the fastener bearing on the hole. For secondary structure (clips, brackets, floor attachment), the cycle time reduction and weight saving are compelling. For primary structure skin-stringer interfaces, the structural efficiency gain is significant enough that multiple major OEMs are qualifying welded thermoplastic assembly processes for their next-generation aircraft programmes.

2. Re-Forming: From Flat to Shape Without Separate Layup

Thermoplastic Re-forming Process
Lay up flat via AFP — then form to complex geometry in a separate thermoforming step
🗂️
Flat AFP Laminate
Thermoplastic tape laid up on flat or low-curvature mandrel via AFP head
Simplified trajectories · Standard consolidation parameters · In-situ or post-consolidation
Heat
above Tg
🔧
Press Tool
Laminate heated uniformly above Tg using infrared, oven, or direct-contact heating
Matched metal press · diaphragm forming · or bladder press tooling
Pressure
+ cool
Final Geometry
Part cooled under pressure in tool — holds formed shape with precise dimensional accuracy
No cure cycle · No autoclave · Net-shape or near-net-shape output
Why Re-forming Changes the Economics
Lay up flat — far simpler AFP trajectory programming
Standard thermoforming tooling — significantly less expensive than complex layup mandrels
Fast geometry iteration without re-programming the AFP cell
Thermoset impossible — only thermoplastic enables this route
Applicable Geometries
📐 L-angle
🔲 C-channel
〽️ Z-section
🎩 Hat stiffener
🌐 Simple doubly-curved panel

The ability to lay up flat and re-form is particularly valuable in prototyping and low-volume production. Instead of programming complex layup trajectories over a curved mandrel, you lay up on a flat tool (simple, fast, accurate), then thermoform the blank to the target geometry. The process is well-understood from the thermoplastic sheet forming industry and does not require specialised AFP-specific expertise.

3. In-Situ Repair

Thermoset vs. Thermoplastic Repair Comparison
Step-by-step repair workflows — time, access requirements, and field viability
🔒
Thermoset Repair
1
Damage Detected
NDT inspection, visual assessment
2
Grind Out Damage
Controlled scarfing to expose clean substrate
3
Apply Repair Plies
Prepreg or wet layup matched to parent laminate
4
Vacuum Bag + Heat Blanket
Out-of-autoclave on-part curing setup
5
Cure Cycle
60–180 min at 120–180°C under vacuum pressure
6
NDT Verification
Ultrasonic / thermography inspection
⏱️Total time: 4–24 hours
🏭Autoclave-quality repair requires autoclave access
⚠️Field repair achievable — quality risk without autoclave
♻️
Thermoplastic Repair
1
Damage Detected
NDT inspection, visual assessment
2
Clean Damage Zone
No scarfing required — surface preparation only
3
Apply Thermoplastic Patch
Same material family — fully compatible interface
4
Local Heating
Heat gun / laser / resistive blanket — portable equipment
5
Apply Pressure, Cool
Manual or mechanical pressure during cool-down
6
NDT Verification
Ultrasonic / thermography inspection
⏱️Total time: 30 minutes – 4 hours
No autoclave required. Repair quality approaches parent material
🚁Fully field-deployable with portable equipment
Comparative Repair Performance
Thermoset
Thermoplastic
Repair Time
4–24 hrs
0.5–4 hrs
Equipment Need
Heavy (autoclave)
Portable only
Bond Strength
~75% (field OOA)
~90% parent
Field Viability
Limited / risk
Fully deployable

For defence and space applications where field repair without depot-level equipment is a genuine operational requirement, thermoplastic's repairability is not a nice-to-have — it is a programme requirement. Several advanced UAV and space launch vehicle programmes are specifying thermoplastic composites explicitly for this reason.

4. End-of-Life Recyclability

Composite End-of-Life Pathways
Material recovery options for thermoset vs. thermoplastic composites — current state vs. enabled future
🔒
Thermoset
Current State
📦 Decommissioned Part
🗑️
Landfill
Most common current route. Increasingly regulated in EU markets — CFRP waste bans emerging.
⚠ Regulated / Restricted
🔥
Incineration
Energy recovery in some EU markets. Carbon fibre destroyed — no material retention.
◑ Energy only
⚗️
Pyrolysis → rCF
Burns off matrix to recover carbon fibre. High cost, limited infrastructure, not yet widely scaled.
◑ Partial recovery
♻️
Thermoplastic
Enabled Pathway
📦 Decommissioned Part
🔄
Shred + Melt → Pellets
Downcycled to injection moulding feedstock. Material retained in supply chain, shorter fibre lengths.
✓ Material retained
🔁
Re-consolidation → New Part
Reform to new geometry via thermoforming. Near-equivalent properties if matrix not thermally degraded.
★ High value recovery
🧪
Solvent Process → rCF
Emerging process dissolving matrix, recovering long continuous fibres. Higher value than pyrolysis route.
✓ Higher value than pyrolysis
Material Value Retained by Pathway
🗑️ Landfill (TS)
~0%
🔥 Incineration (TS)
~5%
⚗️ Pyrolysis rCF (TS)
~30%
🔄 Shred + Pellets (TP)
~45%
🧪 Solvent rCF (TP)
~65%
🔁 Re-consolidation (TP)
~85%
Chopped CFRTP recyclate

Chopped CFRTP recyclate — thermoplastic composite material recovered back into the supply chain as injection moulding feedstock.

The EU's End-of-Life Vehicle Directive (ELV) and evolving aircraft end-of-life regulations are forcing OEMs to quantify and report the recyclability of structural materials. For programmes where end-of-life recyclability is a contract requirement or a public-facing sustainability commitment, thermoplastic composites offer a documented pathway that thermosets simply do not.

The Challenges That Remain Real

The thermoplastic AFP story would be incomplete without an honest account of what is still being solved. Anyone telling you thermoplastic AFP is a mature, plug-and-play process is either selling something or working at a very different level of production than most manufacturers.

Void Content: Closing the Gap

Void Content Benchmarks
Typical void % by process — measured against the <2% aerospace qualification threshold
0%
1%
2%
3%
4%
5%
<2% threshold
Thermoset Autoclave
Prepreg · benchmark
< 1%
✓ Qual met
TP In-Situ (early research)
Pre-2018
3–5%
— not tested
TP In-Situ (2020 state)
~2020
1–3%
— not tested
TP In-Situ (2024 optimised)
Best-in-class ISC · 2024
0.5–1.5%
✓ Approaching
TP Press Consolidation
OOA · post-form press
< 1%
✓ Qual met
TP In-Situ Consolidation — Void Reduction Progress
Pre-2018
3–5%
~2020
1–3%
2024
0.5–1.5%
Gap to autoclave: ~0.5–1% remaining for best-in-class ISC systems. Status: engineerable, process-dependent — not a fundamental material barrier. Press consolidation already meets the <2% qualification threshold without an autoclave.

Void content is the primary mechanical property differentiator between ISC-consolidated laminates and autoclave-consolidated laminates. Voids act as stress concentrations under compression loading. The gap has closed significantly over the past decade of research, and several programmes have demonstrated autoclave-equivalent properties with optimised ISC processes. But it requires careful process control — it does not happen automatically.

Crystallinity Management in PEEK and PEKK

Semi-crystalline thermoplastics develop their mechanical properties through the formation of crystalline regions in the polymer matrix during cooling. The degree of crystallinity — and crucially, its uniformity — depends on the cooling rate during processing.

Crystallinity vs. Cooling Rate — PEEK
Four cooling regimes and their effect on crystallinity, Tg, and mechanical behaviour
Very Slow
< 5°C/min
Crystallinity
35–40%
Tg elevated
High stiffness, more brittle — reduced impact resistance
⚠ Over-crystallised
Moderate
5–50°C/min
Crystallinity
25–35%
Tg target
Balanced stiffness and toughness — target morphology for aerostructure
★ Target regime
Fast
> 100°C/min
Crystallinity
10–20%
Tg lower
Increased toughness, lower stiffness — may be acceptable for some applications
◑ Application-specific
Quench
Very rapid
Crystallinity
< 5%
Tg much lower
Quasi-amorphous — inadequate service temperature resistance for aerostructure
✗ Not suitable
Crystallinity Range Across Regimes
Quench
<5%
Fast
10–20%
Moderate ★
25–35%
Very Slow
35–40%
0% 10% 20% 30% 40%
The AFP Challenge
Cooling rate varies continuously during AFP deposition — the process control software must manage this variability actively to maintain the target semi-crystalline morphology.
📐 Geometry changes alter thermal contact
🤖 Robot speed changes dwell time
📦 Prior ply thermal mass varies
⚙️ Closed-loop control essential

PEKK's slower crystallisation kinetics give it a wider process window than PEEK, which is one reason several aerospace programmes have preferred PEKK despite its higher cost — it is more forgiving of process variation.

Material Cost Premium

Thermoplastic Tape vs. Thermoset Prepreg — Cost Comparison
Raw material cost per kg — thermoset baseline vs. thermoplastic grades · scale: €0–200/kg
€0 €40 €80 €120 €160 €200
Prepreg ceiling €80
Carbon-epoxy Prepreg
IMA / T700 · aerospace grade
€30–80 /kg
Thermoplastic tapes
PPS-CF UD Tape
Secondary structure grades
€60–100 /kg
PEI (Ultem)-CF UD Tape
Interior / semi-structural
€80–130 /kg
PAEK-CF UD Tape
Intermediate performance
€100–180 /kg
PEEK-CF UD Tape
Aerospace primary structure
€90–160 /kg
PEKK-CF UD Tape
Wide process window premium
€110–200 /kg
📈
TP tape premium over thermoset prepreg — 1.5× to 4× depending on grade and volume. Premium is offset by downstream savings across the production system.
1.5×–4×
material premium
Cost Offset Factors — Downstream Savings
🛍️
No autoclave consumables — vacuum bags, bleeder, release film eliminated
No autoclave energy cost per cycle
✂️
No cure-related scrap — no exotherm, no void-induced rejection
🔩
Weldable assembly — fastener count and cost reduction
♻️
Recyclability — material value recovery at end of life
📦
No cold storage or shelf-life handling costs
1.5×–4×
TP tape cost premium over thermoset prepreg (grade and volume dependent)
€90–200/kg
Typical PEEK/PEKK UD tape cost range (aerospace grade)
Favourable
Whole-process economics when autoclave removal, fastener reduction & EOL recovery are included

The material cost premium is real and it does not fully disappear even at volume. But it is increasingly being evaluated in the context of whole-process economics — when you remove autoclave processing, reduce fastener count, and account for end-of-life material recovery, the lifetime cost per kilogram of finished structure often compares favourably.

Aerospace Applications: What Is Qualifying Now

The programmes in qualification today will set the standard process envelopes and material databases that the rest of the industry will inherit. Understanding where thermoplastic AFP is qualifying — and for what structural applications — tells you where the technology is genuinely headed.

Thermoplastic AFP Qualification Landscape (2025)
Readiness by application tier — from qualified production to advanced demonstrators
Qualified / In Production
Qualification in Progress
Advanced Demonstrators
Specification-Driven
🔵
Secondary Structure
Non-structural panels, low primary load
Brackets, clips, floor attachment fittings, drain masts, fairings
✓ Qualified
Multiple OEM primes
TRL
8–9
Airbus A350
Fokker / GKN
In production at several OEMs
Semi-Structural
Loaded but not primary flight-critical
Pressure bulkheads, movables, spoilers, access panels
◑ In Progress
Programme-driven
TRL
6–7
Clean Sky 2
Wing of Tomorrow
Airbus / NLR
🔴
Primary Structure
Safety-critical, full certification required
Fuselage skin panels, welded stringer-skin assemblies, empennage skins
▲ Demonstrators
Full-scale advanced
TRL
5–6
NLR
DLR
Airbus
STELIA
Full-scale demo panels
🚀
Space Launch Vehicles
Non-reusable or single-use primary structures
Payload fairings, inter-stage structures, satellite panels
✓ Qualified
Thermoplastics dominate
TRL
9
SpaceX
Ariane Group
New Space OEMs
🛡️
Defence / UAV
Specification-driven, field repairability a key driver
UAV primary structure, missile fins, drone frames
◈ Spec-Driven
Field repair requirement
TRL
7
Various defence primes
NATO requirements
Loitering munitions
Qualification Maturity by Tier
🚀 Space Launch
TRL 9 — Production
🔵 Secondary
TRL 8–9 — Qualified
🛡️ Defence / UAV
TRL 7 — Spec-driven
⚪ Semi-Structural
TRL 6–7 — In Progress
🔴 Primary Structure
TRL 5–6 — Demonstrators

The Airbus Wing of Tomorrow programme — the most significant composite primary structure development programme currently running — has demonstrated thermoplastic AFP stringer fabrication with induction welding for skin-stringer assembly. The weight and recurring cost targets for that programme cannot be met with thermoset processing. Thermoplastic AFP is not an option being evaluated; it is the enabling technology.

Fokker (now part of GKN Aerospace) has been producing thermoplastic control surfaces in series production for over a decade. The Gulfstream G500/G600 tail structures use thermoplastic composites. These are not research programmes — they are production programmes running today.

Thermoplastic fuselage demonstration

Image credit: Composites United, from New Architecture for Automated Production of the World's Largest Thermoplastic Aircraft Fuselage Demonstrated on a 1:1 Scale

The Process Chain: What It Takes to Run Thermoplastic AFP

If you are evaluating thermoplastic AFP for your facility, the full process chain looks like this:

Thermoplastic AFP — Full Process Chain
Six sequential stages from incoming tape to qualified part — no autoclave at any step
1
📦
Material Preparation
TP UD tape received from supplier (PEEK, PEKK, PPS, PEI)
Incoming QC: thickness, width, fibre volume fraction
Slit to required tow width if supplied wider
Store on climate-controlled creel — no shelf-life issue
Moisture conditioning critical for PPS: absorbed moisture converts to steam at process temperatures
2
🔧
Tool Preparation
Tool material: Invar, carbon/carbon, or high-temp tooling composite
Surface: high-temp release agent or PTFE film
Thermal mass management — tool must survive repeated heating cycles
Dimensional stability critical: TP has higher thermal residual stress than thermoset
3
🤖
AFP System — Robot + Head + Software
Robot: standard 6-axis industrial (KUKA, Fanuc, ABB, UR)
Head: TP-capable — laser + compaction roller, up to 450°C
CAM software: generates tow paths, embeds thermal model
Process parameters: speed, power, roller force, tape tension
4
Layup + In-Situ Consolidation
Head traverses tool surface, laying and consolidating each tow
Real-time temperature monitoring via pyrometer at nip point
In-process inspection: camera-based coverage and gap detection
No intermediate debulk cycles required — key advantage over thermoset prepreg
5
⚙️
Post-Processing (if any)
ISC with good process control: no post-process typically required
Optional press consolidation: further reduces voids if needed
Welding: resistance, induction or ultrasonic for sub-assembly joining
Machining: trim, drill — thermoplastics are easier to machine than thermoset
6
🔬
NDT and Qualification
Ultrasonic C-scan: void content, delamination detection
CT scan: high-resolution void mapping for qualification coupons
Mechanical coupons: ILSS, CAI, OHT, bearing strength
Crystallinity verification (DSC) for PEEK / PEKK applications
Key Process Advantages vs. Thermoset
No autoclave at any stage
No debulk cycles
No cold storage / shelf life
In-situ consolidation
Weldable assembly
Thermoplastic machinability
Field-repairable

The Most Underestimated Step

The step that is most often underestimated by teams new to thermoplastic AFP is step [3] — specifically the coupling between the CAM software and the thermal model. Thermoset AFP CAM software generates a valid programme if the geometry is correct and the steering radius is within limits. Thermoplastic AFP CAM software needs to additionally ensure that the thermal input at every point of every tow path is within the process window for that material, at that layup speed, on that surface angle, in that ply sequence.

That requirement is what makes software-first AFP development — designing the process parameters into the software from the beginning, not patching them in — the right architecture for thermoplastic AFP.

The Hardware Accessibility Shift

Historically, thermoplastic AFP required custom-built machines from a small number of specialist manufacturers. The machines were expensive, required specialist integration, and were effectively only accessible to tier-1 aerospace contractors and national research institutes.

That landscape has changed significantly in the past five years.

Thermoplastic AFP Hardware — Accessibility Evolution
From bespoke Tier 1 capital equipment to accessible modular systems — 2015–2025
🏭
Before
~2015–2020
Machine Type
Bespoke gantry or large industrial robot cell
Capital Cost
€1.5M – €5M+
Footprint
200–500 m²
Integration
6–18 months, specialist integrator required
Temp Cap
< 350°C — limited to lower-Tg materials
Access
Tier 1 primes and national labs only
🚀
Now
~2022–2025
Machine Type
Robot-mounted AFP head on standard industrial robot
Capital Cost
€50K – €300K (head + integration)
Footprint
30–80 m² (standard production bay)
Integration
4–8 weeks
Temp Cap
Up to 450°C — PEEK, PEKK, PAEK compatible
Access
Tier 2/3 suppliers, research institutes, SMEs, startups
Quantified Reduction — Before vs. Now
Before
Now
Entry Cost
€1.5M–€5M+
€50K–€300K
Footprint
200–500 m²
30–80 m²
Integration Time
6–18 months
4–8 weeks
Temp Capability
< 350°C
Up to 450°C
The accessibility barrier that kept thermoplastic AFP out of mid-tier manufacturing has been substantially removed.
What remains is process knowledge — and that is transferable.
€50K–€300K
Current robot-mounted AFP head (vs €1.5M–€5M+ previously)
4–8 weeks
Integration time (vs 6–18 months previously)
450°C
Maximum processing temperature — PEEK, PEKK, PAEK compatible

This shift matters because the qualification burden in aerospace is front-loaded at the prime level. By the time thermoplastic AFP processes are qualified for primary structure programmes at Airbus and Boeing, the material databases, process windows, and design allowables will be published and available to the supply chain. Tier-2 and tier-3 suppliers who have built their thermoplastic AFP process capability now will be positioned to execute those programmes when qualification transfers down the chain.

Traditional AFP gantry vs AFP-XS by Addcomposites

Traditional AFP gantry vs. AFP-XS by Addcomposites — the accessibility shift that brought thermoplastic AFP within reach of Tier 2/3 manufacturers and research institutes.

Where the Technology Is Heading: 2025–2035

The adoption curve for thermoplastic AFP is following a pattern composites manufacturing has seen before — aerospace primary qualification happens first, then the knowledge base and supply chain mature enough to enable industrial and defence adoption, then cost normalisation opens automotive and high-volume applications.

Thermoplastic AFP — Adoption Trajectory
Indicative production volume ramp by sector — staggered wave adoption 2020–2035
↑ We are here (~2026)
Space / Advanced Demonstrators
Aero Primary Structure
Aero Secondary + Semi-Structural
Defence / UAM / eVTOL
Automotive / High-rate EV

Technology Milestones Most Likely to Accelerate the Curve

Materials

Tape quality consistency from tier-2 tape suppliers is improving. PEKK pricing is coming down as volume increases. PA-CF tape quality for automotive AFP is reaching the consistency needed for structural applications.

Process Automation

Closed-loop process control — where the AFP system adjusts laser power and speed in real-time based on in-situ temperature feedback — is transitioning from research to production. This removes the process engineer from the parameter-adjustment loop.

Software

Thermal modelling fidelity is improving. Process simulation tools that predict consolidation quality before the first ply is laid are becoming commercially available. This compresses process development time from months to weeks.

Assembly Qualification

Resistance and induction welding of thermoplastic composite structures is in active qualification at multiple primes. Once skin-stringer welded assembly qualifies for primary structure, the recurring cost and weight arguments for thermoplastic AFP become undeniable.

The Bottom Line

Thermoplastic AFP is not a future technology waiting to be invented. It is a current technology in the middle of its industrialisation phase.

The physics is understood. The material properties are demonstrated. The manufacturing economics — no autoclave, weldable assembly, end-of-life recyclability, field-repairable structures — are compelling at multiple levels of the aerospace and industrial supply chain.

Thermoplastic AFP — Decision Summary (2025 State of the Art)
Head-to-head comparison across 11 dimensions — thermoset AFP vs. thermoplastic AFP
Dimension
🔵 Thermoset AFP
🔴 Thermoplastic AFP
Autoclave required
Yes — most programs
No
Post-cure step
Yes — always
No
Weldable assembly
No
Yes — induction, resistance
In-situ repair
Limited
Yes
Recyclability
No (practical)
♻️ Yes — melt + reform
Shelf life (material)
6–18 months (frozen)
Unlimited
Process window
Wide
Narrow (high-temp mats)
Void content (current ISC)
< 1%
0.5–2% (process dep.)
Material cost
€30–80 /kg
€80–200 /kg
Qualification maturity
Fully mature
Qualifying now (primary)
Accessible hardware
Yes
Yes — 2022 onwards
Process Simplicity
Thermoset AFP
4.2
Thermoplastic AFP
7.8
Lifecycle Sustainability
Thermoset AFP
2.5
Thermoplastic AFP
8.5
Qualification Readiness
Thermoset AFP
9.5
Thermoplastic AFP
5.5
Total Cost of Ownership
Thermoset AFP
5.0
Thermoplastic AFP
7.2

The questions now are process robustness at production rate, supply chain depth for thermoplastic tapes at volume, and software sophistication for managing the thermal process in complex geometry. None of these are fundamental barriers — they are engineering problems with known solution paths.

The manufacturers who build their thermoplastic AFP process capability now — before the qualification documents are published and the supply chain demand surge arrives — are the ones who will be positioned to capture the programmes when they come.

What Addcomposites Offers for Thermoplastic AFP

AFP-XS and AFP-X by Addcomposites

AFP-XS and AFP-X by Addcomposites — thermoplastic AFP from first demonstrator to full production rate, on any standard industrial robot.

The AFP-XS and AFP-X heads support thermoplastic processing up to 450°C, compatible with PEEK, PEKK, PAEK, PPS, and PEI tape on any standard industrial robot. AddPath, the open CAM software, includes thermal process modelling for thermoplastic trajectory generation.

AddPath trajectory generation

AddPath: Software-First Thermoplastic AFP

AddPath in action — trajectory generation and thermal process modelling running before the first tow is laid.

If you're evaluating thermoplastic AFP for your facility — whether that's a first demonstrator, a process development programme, or a production readiness exercise — we're happy to discuss your material, geometry, and throughput requirements.

Thermoplastic AFP manufacturing overview

Learn More

Get in touch to discuss your thermoplastic AFP application →

Contact Us for a Consultation

References

  1. Airbus / Clean Sky 2 — Thermoplastic Wing of Tomorrow Programme. Summary of the Clean Sky 2 Wing of Tomorrow initiative, covering thermoplastic composite stringer fabrication, induction welding for skin-stringer assembly, and large-scale demonstrator results at Airbus Broughton and partner facilities. https://www.cleansky.eu/wing-of-tomorrow
  2. DLR (German Aerospace Center) — Thermoplastic Composite Fuselage Research. DLR's Centre for Lightweight Production Technology (ZLP) thermoplastic AFP research, covering in-situ consolidation parameter optimisation, hot-gas torch AFP, and crystallinity management in PEEK and PEKK laminates. https://www.dlr.de/zlp
  3. NLR (Netherlands Aerospace Centre) — Thermoplastic Composite Manufacturing. NLR's published research on resistance welding of thermoplastic composites for aerospace primary structure, including ILSS, fatigue, and impact data for welded CF/PEEK joints. https://www.nlr.org/research/thermoplastic-composites
  4. CompositesWorld — "Thermoplastic Composites: Aerospace Applications and Market Trends" (2024). Market analysis and technology overview of thermoplastic AFP adoption across aerospace primary and secondary structure, covering major OEM programmes and supply chain development. https://www.compositesworld.com/articles/thermoplastic-composites-aerospace
  5. Victrex — PEEK and PAEK for AFP Applications: Processing Guide. Technical data and processing guidance for Victrex PEEK and PAEK UD tape in AFP applications, including recommended process windows, crystallinity development curves, and mechanical property benchmarks. https://www.victrex.com/composites/afp
  6. Solvay (now Syensqo) — APC-2 PEEK Unidirectional Tape Technical Data Sheet. Material specification for CF/PEEK APC-2 tape including Tg, Tm, melt viscosity, and AFP processing parameters. Industry reference data for PEEK in-situ consolidation process window. https://www.syensqo.com/composites
  7. Dávid Šmíd et al. — "In-Situ Consolidation of Thermoplastic Composites by AFP: Process Parameters and Mechanical Properties" (Composites Part A, 2022). Peer-reviewed study quantifying the relationship between AFP process parameters (speed, temperature, roller force) and void content in ISC-consolidated CF/PEEK laminates. https://doi.org/10.1016/j.compositesa.2022.107015
  8. Grouve, W.J.B. et al. — "Thermoplastic Composite Welding: A Review" (Composites Part B, 2023). Comprehensive review of resistance, induction, and ultrasonic welding of thermoplastic composites — joint strength data, process parameters, and qualification status for aerospace primary structure applications. https://doi.org/10.1016/j.compositesb.2023.110715
  9. MarketsandMarkets — Thermoplastic Composites Market Report 2024–2031. Market sizing (USD 14.3B by 2031), CAGR projections, and adoption segment analysis for thermoplastic composites with breakdown by resin type, fibre type, and end-use application. https://www.marketsandmarkets.com/Market-Reports/thermoplastic-composites-market-194.html
  10. JEC Composites — "Thermoplastic AFP: From Lab to Production" (JEC World 2025 Conference Proceedings). Conference proceedings from JEC World 2025 covering process maturity assessments, production-rate results from leading AFP thermoplastic programmes, and material supply chain development status. https://www.jeccomposites.com/jec-world
  11. Fokker Aerostructures / GKN Aerospace — Thermoplastic Composite Control Surfaces. Overview of Fokker's long-running thermoplastic composite production programmes for Gulfstream G500/G600 tail structures and Airbus A350 thermoplastic brackets — evidence of series production maturity. https://www.gknaerospace.com/thermoplastic-composites
  12. Addcomposites — AFP-XS Thermoplastic Processing Technical Documentation. Addcomposites technical documentation covering AFP-XS thermoplastic head specifications, supported material systems (PEEK, PEKK, PPS, PEI up to 450°C), robot compatibility, and AddPath thermal process modelling for thermoplastic trajectory generation. https://www.addcomposites.com/afp-xs
Pravin Luthada

Pravin Luthada

CEO & Co-founder, Addcomposites

About Author

As the author of the Addcomposites blog, Pravin Luthada's insights are forged from a distinguished career in advanced materials, beginning as a space scientist at the Indian Space Research Organisation (ISRO). During his tenure, he gained hands-on expertise in manufacturing composite components for satellites and launch vehicles, where he witnessed firsthand the prohibitive costs of traditional Automated Fiber Placement (AFP) systems. This experience became the driving force behind his entrepreneurial venture, Addcomposites Oy, which he co-founded and now leads as CEO. The company is dedicated to democratizing advanced manufacturing by developing patented, plug-and-play AFP toolheads that make automation accessible and affordable. This unique journey from designing space-grade hardware to leading a disruptive technology company provides Pravin with a comprehensive, real-world perspective that informs his writing on the future of the composites industry.