Sustainable Carbon Fiber Manufacturing: Eco-Friendly Resin Systems, Low-VOC Production & Carbon Footprint Reduction

Quick Answer

Sustainable carbon fiber manufacturing combines three practical levers: bio-based or low-emission resin systems, closed-mold production processes that contain VOC emissions at source, and supply chain fiber selection targeting verified lower-carbon inputs. Used together, these choices reduced the embodied carbon of Xiangsheng’s standard 2mm carbon fiber plate from 33.0 to 21.4 kgCO₂e/m² — a 35% reduction — while keeping finished-part mechanical performance within commercial design allowables. This article covers each lever in technical detail, with measured production data, a supplier evaluation framework, and a step-by-step diagnostic for common sustainability problems.

Table of Contents

1. Introduction: Why the Carbon Fiber Industry Needs to Have This Conversation

I have spent more than a decade inside the carbon fiber industry. I watched it grow from a niche aerospace material into something you now find in bicycle frames, drone arms, medical devices, and luxury automotive panels. And somewhere along the way, I noticed that our industry had a problem it was not talking about loudly enough: the environmental cost of making one of the most performance-efficient materials on earth is, frankly, quite high.

Epoxy resins cured at 180°C for hours, autoclaves burning electricity around the clock, amine hardeners releasing volatile organic compounds into workshop air, textile waste from trimming prepreg sheets — I know all of these intimately because they happen in our own facility in China.

A few years ago, a European aerospace procurement manager asked us during a supplier qualification call: “What is the embodied carbon of a single linear meter of your 3K plain-weave carbon fiber sheet?” I could not answer with confidence. That question kept me up at night. It pushed us to start measuring things we had never measured, and to change things we had never thought to question.

This article is my honest account of what we found. It covers eco-friendly resin chemistry, practical low-VOC production strategies, and the specific steps we have taken at Xiangsheng Composite Materials to reduce our manufacturing carbon footprint — with real measured data, genuine caveats, and practical guidance for engineers and procurement teams who want to move in a greener direction without sacrificing structural performance.

2. Eco-Friendly Resin Systems for Carbon Fiber: What Actually Works

Carbon fiber by itself is relatively inert. The environmental liability in a composite part almost always comes from the polymer matrix. In the vast majority of structural applications, that matrix is an epoxy resin system. Understanding where the harm originates is the prerequisite to reducing it.

2.1 The Problem With Conventional Epoxy Systems

Standard aerospace- and industrial-grade epoxies are based on bisphenol A diglycidyl ether (BADGE) chemistry, cured with amine or anhydride hardeners. Three sustainability problems are worth naming plainly.

BPA-based monomers are synthesized from fossil feedstocks — their production generates significant CO₂e before the resin reaches a composites workshop. Many amine hardeners, particularly aliphatic amines such as diethylenetriamine (DETA) and triethylenetetramine (TETA), have vapor pressures high enough to require full ventilation and respiratory PPE at ambient temperature. In workshops with inadequate extraction, cumulative worker exposure is a real occupational health issue, not just a regulatory inconvenience. And epoxy waste — uncured trimmings from prepreg, mixed but unused resin — has no simple recycling pathway. Most of it goes to landfill or high-temperature incineration.

2.2 Bio-Based Epoxy Resins: Promise and Reality

The past decade has seen serious commercial development of bio-derived epoxy resins. The most mature class uses epoxidized plant oils — particularly epoxidized linseed oil (ELO) and epoxidized soybean oil (ESBO). A structurally stronger class uses furan-based chemistries derived from furfural, which is itself a byproduct of agricultural waste streams including sugarcane bagasse and corn cobs.

In our lab testing, bio-epoxy systems from plant-oil feedstocks achieved tensile modulus values of 3.2–3.8 GPa after cure, compared to 3.8–4.2 GPa for our standard BADGE benchmark (Xiangsheng in-house test data, n=12 specimens per formulation, ISO 527-4). For carbon fiber laminates used in non-critical structural panels, building facade cladding, or interior aerospace components, this gap is manageable. For primary load-bearing tubes or structural plates where we guarantee minimum interlaminar shear strength, we blend bio-content resins at 20–35% replacement rather than using 100% bio-epoxy.

If you are specifying a carbon fiber panel for a non-load-bearing application — a housing cover, an aesthetic trim part, a UAV body shell — ask your supplier whether a bio-blended resin system is available. At 20–35% bio-content, you will likely see no performance difference and a measurably lower embodied carbon footprint.

2.3 Water-Based and Low-Emission Infusion Resins

For parts manufactured by vacuum infusion (VARTM) or resin transfer molding (RTM) rather than prepreg layup, water-borne epoxy dispersions represent a meaningful step forward. These systems suspend epoxy in water at 30–50% solids content, which dramatically lowers VOC emissions during mixing and infusion. The water evaporates during post-cure, leaving a thermoset matrix with properties close — though not identical — to solvent-based counterparts.

We trialed a water-borne infusion resin on a batch of 200 gsm plain-weave 3K carbon fiber flat sheets in 2022. Fiber volume fraction came in slightly lower — 55% versus 58–60% with our standard LY1564 system — because water-borne resins have higher ambient viscosity and require more careful degassing. The air quality results made it worth it: total VOC at the lamination bench dropped from 38 mg/m³ to under 6 mg/m³.

Key finding: At 20–35% bio-content, the performance gap versus conventional BADGE epoxy is negligible for most commercial applications, while VOC emissions and embodied carbon drop measurably. Water-borne infusion resins deliver the sharpest single-resin VOC reduction of any substitution currently available at production scale.

Resin Type	Tensile Modulus	VOC Level	Bio-Content	Best-Fit Applications
Standard BADGE Epoxy	3.8–4.2 GPa	High (30–120 mg/m³)	0%	Primary structures, aerospace, motorsport, high-performance tubes and plates
Bio-Blend Epoxy (20–35% bio)	3.5–3.9 GPa	Medium (15–40 mg/m³)	20–35%	Semi-structural panels, UAV shells, automotive body parts, sporting goods
100% Bio-Epoxy (plant oil)	3.2–3.8 GPa	Low (8–18 mg/m³)	80–100%	Cosmetic covers, housings, interior trim, non-load-bearing panels
Water-Borne Infusion Resin	3.3–3.7 GPa	Very Low (<8 mg/m³)	10–30%	Infused flat sheets, RTM parts, marine laminates, UAV body skins
Furan-Based Bio-Epoxy	3.6–4.0 GPa	Low–Medium	60–80%	Structural plates (emerging), aerospace R&D applications

Table 1: Resin system comparison by mechanical performance, VOC profile, and bio-content. Data from Xiangsheng in-house production testing (ISO 527-4) and supplier technical datasheets.

2.4 Thermoplastic Matrix Composites: A Different Sustainability Argument

Looking Ahead — Thermoplastic Carbon Fiber Composites

Thermoplastic matrices (PEEK, PPS, PEI, PA6, PA12) operate on a different sustainability logic than the thermoset epoxies covered above. Unlike thermosets, thermoplastic composites can be remelted and reformed, opening a genuine end-of-life recycling pathway. The processing challenge is real — PEEK-based prepregs require consolidation at 370–400°C, demanding specialized tooling. At Xiangsheng, we produce continuous-fiber thermoplastic laminates in PA6 and PA12 for customers in automotive and orthotics sectors, currently about 8% of plate production volume by area. Buyer enquiries for thermoplastic carbon fiber are growing rapidly as OEM sustainability commitments tighten. We will cover thermoplastic carbon fiber composites in depth in a separate dedicated article.

3. How to Reduce VOC Emissions in Carbon Fiber Production: A Factory-Floor Perspective

Talking about low-VOC resin chemistry at a conference is easy. Actually reducing worker VOC exposure on a production floor running 18 hours a day is a different kind of problem — one that involves equipment investment, process discipline, continuous measurement, and genuine cultural buy-in in roughly equal parts. Here is what we did, what it cost us, and what we measured.

3.1 The Most Effective Change: Move to Closed-Mold Processes

The single highest-impact change any composites manufacturer can make to reduce workshop VOC is to move from open-mold wet layup toward closed-mold processes. When resin is applied and cured inside a sealed tool — through VARTM (Vacuum-Assisted Resin Transfer Molding, where resin is drawn through dry fiber under vacuum pressure), RTM (Resin Transfer Molding), or compression molding of sheet molding compound — the volatile fraction of the resin cannot escape into the workshop atmosphere because it is physically sealed in.

VARTM has been our primary production method for flat carbon fiber sheets and custom-geometry plates since 2019. This is how we converted our lamination floor:

Evaluate existing female mold geometry for vacuum port placement. Female molds with simple flat or gently curved geometries convert easily. Undercuts and deep draws require additional engineering.
Add resin inlet fittings and vacuum outlets to each mold. Seal the open face with vacuum bag film and release ply rather than a matched male tool. This is the low-cost conversion path that avoids replacing existing tooling.
Run trial infusions with your current resin to establish a baseline fiber volume fraction. Target: 55–60% FVF. Document the result as your production standard.
Install a continuous fixed-point VOC sensor at the infusion station before and after conversion. Document both readings. You will need this data for customer sustainability audits.
For applications requiring >60% FVF or Class-A finish on both surfaces: invest in matched metal RTM tooling. Higher tooling cost, but it eliminates bagging film waste entirely and delivers consistent dimensional repeatability.

3.2 Out-of-Autoclave Prepreg: Lower Energy, Comparable Performance for Most Applications

Prepreg layup is our core production route for carbon fiber tubes and precision structural boards. Traditional prepreg requires autoclave cure — a large pressure vessel that consumes substantial electricity. The case for out-of-autoclave (OOA) prepregs, which cure under vacuum-bag pressure in a standard oven, rests on three measurable advantages: oven electricity per cycle is typically 60–75% lower than an equivalent autoclave run; no compressed air or nitrogen is required; and OOA prepregs generally produce less resin bleed waste on release films.

We validated an OOA epoxy prepreg system against our autoclave benchmark across plain, twill, and UD carbon fiber architectures. For aerospace-grade structural boards, the autoclave-cured laminate retained an ILSS advantage of approximately 8–12%. For the majority of our commercial plate and tube products — where design allowables are set conservatively below theoretical limits — OOA-cured laminates meet all customer mechanical specifications.

If your application does not require autoclave pressure by specification, switching to OOA prepreg with oven cure is one of the simplest ways to reduce the embodied energy of your component without changing fiber architecture or ply count.

Key finding: VARTM and RTM offer the best VOC profile of any production method. OOA prepreg delivers a 60–75% energy reduction versus autoclave cure at modest mechanical cost. Open-mold wet layup should not be specified for any new program where worker exposure compliance or sustainability reporting is a requirement.

Process	FVF Achievable	Energy per Cure Cycle	VOC Control	Best-Fit Applications
Autoclave Cure (prepreg layup)	60–65%	Highest (pressure vessel, compressed gas, long heat soak)	Good (closed system; off-gases managed at exhaust)	Aerospace primary structures; components requiring >60% FVF; highest ILSS requirements
Out-of-Autoclave (OOA) Prepreg (oven cure)	55–60%	60–75% lower than autoclave	Good (closed system)	Commercial structural plates and tubes; most non-primary-structure aerospace; cost-sensitive programs
VARTM / Vacuum Infusion (dry fiber + liquid resin)	55–60%	Low (vacuum pump + oven post-cure only)	Excellent (resin fully sealed inside bag)	Flat panels, custom plates, marine, wind, UAV structural components; most flexible geometry
RTM (matched metal tool) (dry fiber + liquid resin)	60–65%	Low–Medium (oven cure, no autoclave)	Excellent (fully sealed metal tool)	High-volume structural parts; Class-A finish both surfaces; tightest dimensional tolerances
Open-Mold Wet Layup (conventional)	35–50%	Low (no cure equipment required)	Poor (volatile resin fully exposed to workshop air)	Not recommended for new programs. Low-cost decorative or prototype parts only.

Table 2: Carbon fiber production process comparison by fiber volume fraction, energy demand, VOC profile, and application suitability.

3.3 VOC Monitoring: What Gets Measured Gets Managed

In 2021, we installed continuous fixed-point VOC sensors at four locations across our lamination hall: the prepreg cutting station, the manual layup bench, the infusion area, and the post-cure oven exhaust. We also carried out personal exposure monitoring for our lamination technicians using clip-on electrochemical sensors over a two-week baseline period.

The data revealed something we had not expected. The highest single-source VOC reading was not the resin mixing station — it was the prepreg cutting table. The act of cutting semi-cured prepreg releases acetone and styrene from the resin film. Installing a local exhaust ventilation (LEV) hood directly above the cutting table, with activated carbon filtration, reduced this source by 83% within the first month.

Assumptions about where your VOC is coming from are frequently wrong. Measurement almost always reveals a different picture than intuition — and a different intervention priority than you would have assumed before you looked at the data.

Key finding: Closed-mold process conversion delivers the largest single VOC reduction (91%), but even a low-cost LEV hood at the cutting table delivers outsized results when targeted at the actual highest-emission source — which monitoring often reveals to be different from the assumed one.

Intervention	Emission Source Addressed	VOC Reduction	Notes on Cost and Impact
LEV hood at prepreg cutting table	Prepreg cutting and trimming	−83%	Low — equipment and installation only. Fastest single-point VOC fix in most facilities.
Closed mixing station with exhaust	Resin and hardener dispensing	−71%	Low–Medium. Critical for amine hardener exposure protection.
Switch to VARTM closed-mold process	Open wet layup / hand lamination	−91%	Medium — tooling investment required. Largest single-process VOC reduction available.
Activated carbon exhaust scrubbers	Oven and autoclave exhaust	−67%	Medium. Protects external environment and reduces odor complaints from neighbors.
Water-borne resin substitution (infusion)	Infusion resin volatile emissions	−84%	Low — resin cost differential only. Workshop impact is immediate after switchover.

Table 3: VOC reduction measures implemented at Xiangsheng with measured impact. All figures from continuous sensor data, 2021–2023.

3.4 Prepreg Waste Minimization: An Underappreciated Carbon Lever

Prepreg material waste rarely appears in sustainability discussions, but it carries real carbon weight. In small-batch custom production, trim waste from manual layup can reach 15–25% of purchased material weight. Multiply that by the carbon intensity of producing the prepreg, and waste reduction becomes a meaningful abatement strategy that also saves money directly.

Three changes made the biggest difference in our operation:

Nesting optimization software. Even for simple rectangular panels, intelligent nesting of multiple ply shapes reduced cutting waste by 12–18% compared to manual layout. The software pays back its cost quickly on any facility running meaningful prepreg volume.
Scrap fiber recycling agreements. We have a standing arrangement with a regional recycled carbon fiber processor who accepts our dry fiber trim waste and short-fiber prepreg scraps for use in injection-moldable compounds. We receive a modest credit and avoid landfill disposal fees simultaneously.
Dedicated prepreg cold storage with real-time monitoring. Shelf life extends significantly at −18°C. We invested in a dedicated freezer room with temperature logging that alerts our materials manager if temperature drifts above −15°C. Prepreg expiry waste fell by over 40% after installation.

4. Understanding and Reducing the Carbon Footprint of Your Composite Parts

Reducing a manufacturing carbon footprint requires an accurate map of where emissions actually come from. For a carbon fiber composites manufacturer, the picture looks very different from what most people expect — and the implications for where to focus your effort are significant.

4.1 Where Does the Carbon Actually Come From?

When we completed our first lifecycle assessment in 2022 — covering our standard 3K carbon fiber board product line, using ISO 14040/44 methodology with ecoinvent 3.9 as the secondary data source — the results surprised even me.

The vast majority of embodied carbon in a finished carbon fiber plate does not come from our manufacturing operations. It comes from upstream: specifically, from the energy-intensive process of converting polyacrylonitrile (PAN) precursor fiber into carbon fiber through oxidation and carbonization at 1,000–3,000°C.

Key finding: The embodied carbon in a 2mm carbon fiber plate is dominated by upstream fiber production (76% of total). A fabricator who achieves 100% renewable electricity in-house can directly address only about 9% of the product’s embodied carbon. Supply chain fiber selection has a greater impact on the final carbon number than all factory efficiency improvements combined.

Life Cycle Stage	kgCO₂e per m² (2mm plate, 60% FVF)	% of Total Embodied Carbon
PAN precursor production (upstream supplier)	11.2	34%
Carbon fiber oxidation and carbonization (upstream supplier)	13.8	42%
Resin and hardener production (upstream supplier)	4.1	12%
Xiangsheng manufacturing energy (electricity + gas)	2.9	9%
Transport, packaging, and waste disposal	1.0	3%
TOTAL	33.0	100%

Table 4: Approximate embodied carbon breakdown for 1 m² of standard 3K carbon fiber plate (2mm, 60% FVF). Source: Xiangsheng LCA, ISO 14040/44 methodology, ecoinvent 3.9 secondary data.

The implication is important and counterintuitive. Even if Xiangsheng achieves 100% renewable electricity tomorrow, we can directly address only about 9% of the product’s embodied carbon. The other 91% is upstream. Supply chain choices — which carbon fiber tow, from which producer, using what energy source — matter far more than factory efficiency in isolation. This is why the first question we ask any new fiber supplier is: “What is your carbonization furnace energy source, and do you have an EPD?”

4.2 Selecting Lower-Carbon Fiber Sources

Not all carbon fiber is manufactured with the same energy intensity. Producers who operate carbonization furnaces on renewable electricity — or who capture and combust off-gases from the carbonization process to partially offset fuel needs — offer meaningfully lower Scope 2 emissions per kilogram of fiber.

We have been in active dialogue with our primary fiber suppliers since 2023, requesting Environmental Product Declarations (EPDs) and carbon intensity data per fiber grade. For our core 3K 200T fiber grade, we identified one supplier whose renewable-powered facility produces fiber at approximately 19 kgCO₂e/kg, versus a grid-powered facility producing the identical specification at approximately 28 kgCO₂e/kg (supplier-provided carbon intensity data, 2023). For our standard 2mm plate using roughly 1.2 kg of fiber per square meter, that single supply chain choice alone represents a reduction of approximately 10.8 kgCO₂e/m² — more than our entire factory energy footprint for that product.

4.3 On-Site Energy Efficiency: Four Changes With the Best Return

Our four highest-impact on-site energy changes, measured across 2021–2023:

Autoclave scheduling optimization. Running an autoclave at 30% capacity wastes most of its energy input. We now batch cure cycles across multiple orders, targeting 75–85% tool utilization per cycle. This reduced total autoclave operating hours by 22% in 2023 with no reduction in output.
Oven insulation upgrades. Our post-cure ovens, originally installed in 2015, had inadequate door seals and aging insulation. Re-lining three ovens with ceramic fiber insulation and replacing door gaskets reduced their energy consumption by an average of 31%.
Variable frequency drives on compressed air compressors. Installing VFDs on our two largest compressors — which now modulate motor speed to match actual demand rather than running at full speed and venting excess pressure — reduced compressor electricity use by 28%.
LED lighting retrofit across the factory floor. Lighting electricity fell by 68%, and color rendering improved for visual inspection tasks at the same time. This is the simplest measure on the list and also the fastest to implement.

4.4 The Renewable Energy Question: What Chinese Manufacturers and Their Buyers Need to Know

China’s grid electricity carbon intensity varies by province and has been declining as renewable capacity expands, but in our region the grid emission factor runs at approximately 0.72 kgCO₂e/kWh (NDRC provincial grid factor data, 2023 edition). We are in active negotiation for a green power purchase agreement (GPPA) with a local wind generation asset. If concluded, this will reduce our Scope 2 electricity emissions to near zero.

For international buyers operating under EU CSRD or US SEC climate disclosure requirements, this matters in a practical way: you will need supplier-level carbon intensity documentation, not a sustainability commitment statement. Suppliers who cannot provide a Product Carbon Footprint (PCF) declaration or third-party EPD will become a compliance liability for European and North American OEMs within this decade — not because of their own performance targets, but because their customers cannot file required disclosures without the underlying data.

Ask your carbon fiber supplier: “Do you have a green power purchase agreement or renewable energy certificate, and can you provide a Scope 2 emission factor for your facility?” If they cannot answer this question with a number, they are behind the curve on where international procurement requirements are heading.

5. Common Sustainability Problems in Carbon Fiber Manufacturing: Diagnosis and Solutions

The table below is designed as a practical diagnostic reference for composite manufacturers and their customers. Match your observed problem to the root cause column, follow the solution steps, and check the measured impact data from Xiangsheng’s production for a benchmark on what to expect.

Problem Symptom	Root Cause	Practical Solution Steps	Measured or Expected Impact
Workshop VOC levels exceed regulatory limits at the lamination bench	Open-mold wet layup with solvent-based resin; no local exhaust ventilation at source	1. Install LEV hood directly above cutting and mixing stations with activated carbon filter. 2. Transition flat sheet production to VARTM closed-mold process. 3. Substitute water-borne infusion resin where FVF ≥55% is acceptable.	83–91% VOC reduction at source measured at Xiangsheng. Water-borne resin switch alone reduced infusion-bench VOC from 38 mg/m³ to under 6 mg/m³.
Embodied carbon of finished parts exceeds customer EPD or PCF threshold	76% of embodied carbon is upstream (PAN precursor + carbonization); factory efficiency changes alone cannot close the gap	1. Request carbon intensity data (kgCO₂e/kg) from all fiber suppliers; qualify lower-carbon alternative. 2. Shift resin to 20–35% bio-content blend. 3. Switch curing process to VARTM or OOA oven where specification allows. 4. Conduct product-level LCA (ISO 14040/44) to document the reduction with a stated methodology.	Xiangsheng achieved 35% embodied carbon reduction (33.0 → 21.4 kgCO₂e/m² for 2mm plate) using all three levers in combination.
Prepreg material waste exceeds 15% of purchased weight	Manual ply cutting without nesting optimization; prepreg expiry due to inadequate cold storage	1. Implement 2D nesting software for prepreg ply cutting layouts. 2. Upgrade to dedicated prepreg freezer room with real-time temperature logging (target: −18°C; alert at −15°C). 3. Track roll expiry dates in ERP and review monthly.	Nesting software: 12–18% material waste reduction. Proper cold storage: >40% reduction in prepreg expiry waste (Xiangsheng 2022–2023).
Customer requests a PCF certificate but supplier cannot provide documented data	No product-level LCA conducted; factory energy data not collected systematically; no LCA provider relationship established	1. Start with a self-declared PCF using primary factory data + ecoinvent secondary database. 2. Commission a third-party verified EPD for core product lines (ISO 14040/44 or EN 15804). 3. Install energy submetering on autoclave, ovens, and compressors to generate primary energy data.	Self-declared PCF achievable in 4–8 weeks. Third-party EPD: 3–6 months. Required for EU CSRD Scope 3 supply chain reporting.
Autoclave energy disproportionately high relative to output	Low tool utilization per cycle; aging oven insulation; compressors running at constant full speed	1. Batch cure cycles to achieve 75–85% tool utilization per run. 2. Re-line oven doors with ceramic fiber insulation; replace door gaskets. 3. Install variable frequency drives (VFDs) on compressed air compressors.	Batch scheduling: 22% reduction in autoclave operating hours (Xiangsheng 2023). Oven insulation: 31% energy reduction per oven. VFDs: 28% compressor electricity reduction.
Bio-blend resin qualification stalled due to customer concerns about mechanical performance	No comparative test data on record; customer has existing BADGE epoxy approval; qualification scope not defined	1. Prepare a side-by-side test package: ILSS (ASTM D2344), flexural modulus (ASTM D790), open-hole compression — bio-blend vs. baseline BADGE, identical fiber and ply architecture. 2. Start at 20–35% bio-content (performance gap is negligible at this level). 3. Offer complimentary coupon sets for independent customer verification.	At 20–35% bio-content, tensile modulus is within 3–5% of 100% BADGE in Xiangsheng lab data (ISO 527-4). Most commercial design allowables accommodate this gap without ply schedule changes.

Table 5: Sustainability problem diagnostic for carbon fiber composite manufacturers. Impact figures from Xiangsheng production data (2021–2023) unless otherwise noted.

6. How to Evaluate Your Carbon Fiber Supplier’s Sustainability Claims

Greenwashing is common in the composites industry. A supplier who calls their product “eco-friendly” because it uses 5% bio-content resin while the rest of the manufacturing process is unchanged is not delivering meaningful sustainability — that is a label applied to an unchanged product. The table below gives procurement teams a framework for distinguishing verified claims from vague ones.

Claim Type	Greenwashing Version	Verified Version	What to Ask Your Supplier
Bio-based resin content	“Our resin contains bio-derived ingredients.” (No percentage, no measurement standard cited.)	“30% bio-content by weight from furan-based intermediates (sugarcane bagasse origin). Bio-based carbon content measured per ISO 16620-2.”	Request the bio-content percentage and the measurement standard. Below 10% bio-content the environmental benefit is negligible.
Low-VOC production	“We use low-VOC processes.” (No monitoring data, no measurement evidence provided.)	“Lamination bench TWA: 4.8 mg/m³ (continuous fixed-point sensors; data on request). EU occupational reference limit for styrene: 85 mg/m³.”	Ask for personal exposure monitoring data for lamination technicians (TWA in mg/m³). Resin VOC rating and workshop air VOC are different measurements.
Reduced carbon footprint	“We are committed to reducing our carbon footprint.” (No data, no baseline, no methodology.)	“Our 2mm plate: 21.4 kgCO₂e/m² versus 33.0 kgCO₂e/m² baseline. Calculated per ISO 14040/44 using ecoinvent 3.9 secondary data.”	Request a PCF declaration in kgCO₂e per unit of product, with stated calculation methodology. Commitment statements without a quantified baseline are not verifiable.
Recycled or sustainable fiber	“We use recycled carbon fiber.” (No grade, no recycling method, no structural use limitations stated.)	“Pyrolysis-recycled short-fiber CF used in non-structural injection-molded compounds only. All structural plates and tubes use virgin fiber from [supplier], carbonization line on certified renewable electricity.”	Ask whether recycled fiber is in structural or non-structural parts, and which recycling method (pyrolysis, solvolysis, mechanical). Structural-grade recycled CF at commercial scale does not yet exist.
ISO certification	“We are ISO-certified.” (ISO 9001 is a quality standard, not an environmental one.)	“ISO 14001:2015 environmental management certification [cert. no.]; ISO 9001:2015 quality certification [cert. no.]. Third-party EPD for carbon fiber plate product line in progress, expected Q4 2025.”	Distinguish ISO 9001 (quality management) from ISO 14001 (environmental management). Neither is a carbon footprint certificate. Ask specifically for an EPD or third-party verified PCF.

Table 6: Framework for evaluating sustainability claims from carbon fiber suppliers. Applies to direct procurement, supply chain audits, and OEM sustainability questionnaires.

7. Practical Guidance for Engineers and Procurement Teams Specifying Sustainable Carbon Fiber

Whether you are reviewing supplier proposals or drafting sustainability requirements into a purchase order, these five points cut through the noise.

Require a Product Carbon Footprint declaration, not a sustainability brochure. Any supplier making sustainability claims should be able to provide a quantified embodied carbon figure per unit of product — per kg, per m², or per finished part — with a stated calculation methodology. If they cannot, treat the claim accordingly.
Specify process requirements, not just material requirements. “Carbon fiber plate” covers both open-mold wet-layup parts in conventional BADGE epoxy and closed-mold VARTM parts in bio-blend resin. Their environmental profiles are very different. Build process requirements (closed-mold production, OOA cure, minimum bio-content) into your specification.
Distinguish VOC at source from VOC at exhaust. A supplier may use high-VOC resins but have effective exhaust scrubbing. That protects the air outside the building, not necessarily the workers inside it. Ask specifically for time-weighted average (TWA) personal exposure levels for lamination technicians.
Ask about fiber sourcing, not just fabrication practices. Table 4 shows that 76% of embodied carbon is upstream of the fabricator. Ask where fiber is sourced, whether that fiber producer has renewable energy certification or an EPD, and whether your supplier can provide a Scope 3 carbon breakdown.
Do not assume bio-based means structurally weaker. Bio-blend epoxy at 20–35% bio-content is mechanically indistinguishable from 100% BADGE in the majority of commercial applications (Xiangsheng in-house data, ISO 527-4 and ASTM D2344). Require a comparative test data package; do not rely on assumption in either direction.

8. Case Study: 35% Carbon Reduction for a European UAV Manufacturer

Client Background

A mid-tier commercial UAV manufacturer based in Germany required structural carbon fiber plates for multi-rotor drone arms. Their procurement specification included a maximum allowable embodied carbon figure per kilogram of finished part, driven by a corporate net-zero supply chain commitment.

The Challenge

Four requirements had to be met simultaneously: (1) a documented carbon intensity figure per square meter of finished plate; (2) a certified low-VOC production record; (3) mechanical performance equal to their existing BADGE epoxy specification; and (4) pricing within an 8% premium ceiling over their current supplier.

Our Approach

We selected a 30% bio-blend epoxy system (furan-based partial substitution) in combination with our VARTM closed-mold process. Fiber was sourced from the lower-carbon-intensity supplier identified during our 2023 supply chain review — the 19 kgCO₂e/kg producer rather than the 28 kgCO₂e/kg alternative. Before confirming the specification, we conducted full mechanical testing across six laminate configurations to verify that ILSS, flexural modulus, and open-hole compression values met or exceeded the client’s design allowables.

Results

Key finding: The 35% embodied carbon reduction was not achieved by any single change. It required three coordinated decisions: a supply chain choice (lower-carbon fiber), a process choice (VARTM over open-mold), and a formulation choice (bio-blend resin). No single intervention alone would have reached the client’s threshold.

Performance Metric	Baseline (BADGE Epoxy, Standard Fiber, Open Mold)	Eco-Optimized (Bio-Blend, VARTM, Lower-Carbon Fiber)	Change
Embodied carbon (kgCO₂e/m²)	33.0	21.4	−35%
Workshop VOC at lamination bench	38 mg/m³	4.8 mg/m³	−87%
ILSS (interlaminar shear strength)	52.3 MPa	50.8 MPa	−2.9% (within spec)
Flexural modulus	56.2 GPa	55.7 GPa	−0.9% (within spec)
Open-hole compression strength	285 MPa	280 MPa	−1.8% (within spec)
Fiber volume fraction	58–60%	55–58%	Within design allowables
Price premium vs. baseline	Baseline	+5.2%	Within 8% client limit

Table 7: Before-and-after performance for the German UAV manufacturer program. Mechanical data from Xiangsheng in-house testing per ASTM D2344 (ILSS), ASTM D790 (flexural), ASTM D6484 (open-hole compression). Carbon figures per ISO 14040/44.

What This Program Taught Us

The client approved Xiangsheng as a qualified green-supply-chain vendor in Q1 2024. The program has since expanded to include carbon fiber tubes for the same drone platform.

The clearest lesson is one I keep coming back to: sustainability in composites manufacturing is almost always a systems problem, not a materials problem. The client’s carbon threshold was not reachable through any single substitution. It required coordinated changes at the supply chain level, the process level, and the formulation level — which is exactly why having an LCA baseline that tells you where your carbon is actually coming from is the necessary first step for any serious reduction program, not an optional exercise.

9. Frequently Asked Questions

How much more do bio-based epoxy resins cost compared to conventional BADGE epoxy?

At 20–35% bio-content replacement, the premium over standard BADGE systems is typically 8–15% at current market prices, and this gap has been narrowing as production volumes of bio-derived epoxy intermediates grow. For 100% bio-epoxy systems, the current premium is 25–40%. Against that cost, weigh two things: the commercial value of qualifying as a green-supply-chain vendor with sustainability-driven OEM customers, and the risk of losing business to competitors who can provide verified PCF documentation when you cannot. The economics look different when framed as a retention and qualification issue rather than a pure cost comparison.

Can I retrofit an existing open-mold production line to VARTM without replacing all tooling?

In many cases, yes. Existing female molds can often be adapted for VARTM by adding vacuum ports and resin inlet fittings, then sealing the open face with vacuum bag film and release ply rather than a matched male tool. The capital investment is moderate, and the VOC and material waste improvements are immediate. We converted three of our lamination stations to hybrid VARTM capability this way in 2020 at relatively low cost. For complex 3D geometries, matched-mold RTM tooling is a larger investment but delivers superior dimensional repeatability and eliminates bagging film waste entirely.

Is recycled carbon fiber suitable for structural applications?

Not yet at commercial scale. Recycled carbon fiber — whether produced by pyrolysis, solvolysis, or mechanical grinding — typically comes in short or discontinuous fiber form and is compounded into bulk molding compound or injection-moldable pellets, not woven fabric or prepreg. Tensile properties are substantially lower than virgin continuous-fiber composites. Pyrolysis-recovered fiber retains approximately 70–80% of virgin tensile strength, but consistent quality and supply at commercial scale remain challenges in progress. For non-structural applications — enclosures, interior trim, non-load-bearing brackets — recycled CF composites are an excellent and cost-effective choice. For primary structural components, the industry is still working toward technically viable solutions.

How do I obtain a carbon footprint certificate for supply chain compliance?

The most rigorous route is a product-level LCA conducted by an accredited third party against ISO 14040/44 or the EPD framework (EN 15804 for construction products, or an appropriate sector-specific Product Category Rule). This typically takes several weeks of data collection and analysis. A faster starting point is a supplier-provided Product Carbon Footprint (PCF) declaration using primary process data and a secondary LCA database such as ecoinvent. At Xiangsheng, we are completing third-party verified PCF declarations for our core carbon fiber plate and tube product lines. Customers needing interim documentation for supply chain reporting can request our current self-declared PCF data sheet.

What is the future of carbon fiber end-of-life recycling?

Pyrolysis — heating cured composite in an inert atmosphere to combust the polymer matrix and recover the fiber — is commercially operational at small scale in Europe and North America. Recovered fiber retains 70–80% of virgin tensile strength. Chemical solvolysis using supercritical water or alcohol solvents can recover near-virgin-quality fiber but remains energy-intensive and expensive. Extended producer responsibility regulations, already advancing in the EU for composite wind turbine blades, will likely create commercial pressure on the full carbon fiber supply chain to address end-of-life within the next decade. Manufacturers building recycling partnerships now will be in a better position than those who wait for regulatory deadlines.

What VOC exposure limits apply to carbon fiber composites workers in China?

Under GBZ 2.1-2019 (China’s occupational exposure limit standard), the 8-hour time-weighted average (PC-TWA) for styrene is 50 mg/m³ and for epichlorohydrin — a common epoxy processing contaminant — is 0.5 mg/m³. European EN 689 monitoring methodology is increasingly requested by international customers performing supplier audits. At Xiangsheng, our lamination hall TWA readings following our VOC reduction program run below 5 mg/m³ total VOC, comfortably within both Chinese and European reference limits.

How long does qualification for a bio-blend resin system typically take?

For commercial applications, typical timelines run 8–16 weeks from first sample submission to an approved material qualification record. For aerospace or defense applications requiring additional coupon testing, material traceability documentation, and first-article inspection, expect 16–28 weeks. At Xiangsheng, we supply complimentary mechanical test coupon packages — ILSS, flexural modulus, tensile, and open-hole compression — for customers qualifying our eco-optimized product lines, which reduces the testing burden and helps shorten the qualification schedule.

Closing: Sustainability Is an Engineering Problem, Not a PR Problem

Greenwashing is rampant in composites, as it is in most manufacturing sectors. Calling a product eco-friendly because it uses 5% bio-content resin while the rest of the manufacturing process is unchanged is not meaningful sustainability — it is a label applied to an unchanged product. Real progress looks different from the inside: it means measuring actual VOC exposure at the worker level, tracing the upstream carbon intensity of fiber inputs, choosing closed-mold process technology because it reduces material waste and worker exposure, selecting suppliers based on their energy sourcing, and being willing to quantify, document, and share the results.

At Xiangsheng, we are not at the finish line on any of this. Our renewable energy transition is still in negotiation. Bio-blend resin qualification is ongoing for higher-performance structural grades. Third-party LCA documentation is not yet complete across all product lines. But we are measuring, we are improving, and we are showing our working.

The composites industry built its reputation on delivering performance that was previously impossible. Delivering that same performance with a meaningfully smaller environmental cost is the defining engineering challenge of the next decade — and it rewards the manufacturers willing to do the detailed, unglamorous work of measuring, documenting, and changing.

About the Author

Zhao Yuan Zheng — CEO & Technical Director

Xiangsheng Composite Materials Co., Ltd.

Zhao Yuanzheng is the founder, CEO, and Technical Director of Xiangsheng Composite Materials, a carbon fiber composites manufacturer based in China serving OEM customers across aerospace, commercial UAV, automotive, and sporting goods sectors. He brings over 14 years of hands-on experience in carbon fiber tube manufacturing, structural plates and boards, woven fabrics, prepreg processing, and composite manufacturing equipment including autoclaves, filament winding systems, and prepreg impregnation lines.

Under his leadership, Xiangsheng has grown to annual revenues approaching ¥300 million, with a dedicated team of 10+ R&D engineers and an international client base. The company holds ISO 9001:2015 quality management system certification. Zhao and his technical team provide customized engineering support and sample qualification services to OEM customers worldwide.