Mechanical Properties of Highly Oriented Recycled Carbon Fiber Tapes Using Automated Fiber Placement

Abstract

This study focuses on producing and processing highly aligned tapes from recycled carbon fibers (rCFs). The rCFs are processed with a modified carding machine, oriented through a specialized subsequent process and consolidated into a semi-finished product. These rCF-tapes are placed onto a two-dimensional tool using an adapted automated fiber placement (AFP) technology to demonstrate a novel approach of producing composites from highly oriented recycled materials. The semi-finished stacks are consolidated in a heating press and test coupons are tested according to corresponding standards. The rCF-tapes are evaluated using methods such as tensile and flexural testing and determination of fiber volume content. Mechanical values are assessed by processing various generations of rCF-tapes and comparing them to each other and to virgin fiber tapes (vCF-tapes) made of the same type of carbon fiber and matrix. Microscopic imaging is also performed to analyze the quality of the resulting composites. In this study, a tensile strength of up to 1100 MPa in the fiber direction and stiffness of up to 80 GPa at a fiber volume content (FVC) of approximately 40% were achieved. The results highlight the strong potential and benefits of using highly oriented rCF-tapes and demonstrate the suitability of fiber placement technologies for those recycled materials.

1. Introduction

The constant expansion of industry has greatly increased the demand for resources. As each new industrial product enters the market, interest in and consumption of its raw materials increases. However, resources are finite and can be rapidly depleted by industrial competition and economic growth. Societies, industries and governments are therefore called upon to promote resource-efficient development. In this context, reducing and avoiding waste, recycling and saving CO₂ are key elements of a sustainable future.

The use of carbon-fiber-reinforced plastics (CFRPs) and their associated material efficiency compared to metallic materials play an essential role in lightweight construction. However, the production of carbon fibers is based on fossil raw materials and is very energy intensive. Therefore, a circular economy approach should be applied to the carbon fiber industry, with fibers reused across multiple lifecycles. The recycling of carbon fibers allows for a sustainable material cycle, though this approach currently faces several limitations [1,2]. Recycled carbon fibers have significantly lower CO₂ emissions. Various studies report an energy consumption of approximately 30–49 MJ/kg [3,4,5], depending on the recycling method used (e.g., pyrolysis, solvolysis, steam thermolysis). In contrast, the production of virgin carbon fibers requires substantially more energy, ranging from 262 to 464 MJ/kg [6]. Using rCFs instead of virgin carbon fiber (vCF) has been shown to significantly reduce costs, with some studies indicating potential savings of up to 50% [7]. The production cost of rCFs is highly dependent on the recycling process used, but is reported to be below EUR 10 per kilogram for mechanically processed or pyrolyzed fibers [8,9]. In comparison, vCF typically costs more than EUR 10 per kilogram [8,9,10].

State-of-the-art materials made from recycled carbon fibers are mostly nonwoven with moderate mechanical properties [11,12]. Due to the isotropy and entangled arrangement of the fibers, it is difficult to achieve FVC above 35% without risking a significant reduction in mechanical properties due to fiber damage and shortening [12,13]. However, with increasing fiber orientation, higher fiber packing densities are possible [14,15]. These two characteristics—fiber orientation and high FVC—are particularly important when processing rCFs, as they are directly correlated with improved composite properties [16,17,18]. One way to enhance fiber orientation is the production of highly aligned thermoplastic rCF-tapes, which involves an additional drafting process after the card web formation [19,20]. Such card-sliver-based tape structures have been reported by multiple institutions [19,20,21]. Using automated fiber placement (AFP) technology, these tapes can be laid along exact load paths, making them suitable for use in demanding applications such as aerospace [22,23]. In addition, the discontinuous fiber structure of rCF-tape means that very high drapability- and flowability can be achieved, and it is possible to achieve tight lay-up radii (steering) that are not possible with current unidirectional materials as long as the matrix is in a molten state and the rCFs within the molten area are able to slide next to each other [20]. The aim of this study is to develop a production chain that improves the properties of rCF composites, enabling them to substitute for higher-grade materials or even virgin fibers. For this evaluation, TRL 6–7 machinery was used to demonstrate the high industrial applicability and transferability of recycled carbon fiber material. As a result, large quantities of material were required for production and processing. Due to the substantial material demand, the number of evaluations that could be conducted was limited, placing a strong emphasis on experimental trials. While previous studies have independently addressed aspects such as fiber alignment, mechanical performance, or tape formation from rCFs, the novelty of this work lies in combining all these aspects into a unified, industry-compatible process. By adapting AFP technology to work with highly oriented rCF tapes, this study presents a promising and scalable approach to elevate recycled materials into high-performance structural applications. Furthermore, a key distinguishing factor of this research is the successful processing of highly heterogeneous rCF feedstock, characterized by a wide range of fiber lengths and significant standard deviation.

2. Materials and Methods

2.1. Materials

The study focuses on tapes made from recycled carbon staple fibers combined with polyamide 6 (PA6) fibers (EMS Grilon P300) as thermoplastic matrix. The rCF originates from cut-off waste of T700S virgin carbon fiber (vCF) fabrics used in the aircraft industry. The characteristics of both the carbon and polymer fibers are shown in

Table 1. Properties of recycled carbon fibers and polymer feedstock.

Fiber Properties	rCF (T700S)	Polymer Fibers (PA6)
Mean Staple length [mm]	84.6 ± 88.7	60
Linear Density [dtex]	0.68 ± 0.06	3.3
Density [g/cm3]	1.8	1.14
Tensile Strength [GPa]	4.01 ± 0.87	0.1
Tensile Modulus [GPa]	212.45 ± 5.38	0.75
Initial Modulus [GPa]	254.24 ± 7.33
Tenacity [cN/tex]	226.71 ± 49.1	46

. Due to the high variation in fiber length of the rCFs and the presence of very long fibers in the feedstock resulting from cutting errors, the material exhibits a high standard deviation in fiber length. After processing via roller carding, the fiber length distribution becomes narrower, and the average fiber length is reduced.

2.2. rCF-Tape Production

To produce highly oriented tape intermediates, the rCFs are first opened and mixed with PA6 matrix fibers. After pre-opening and blending, the rCF/PA6 mixture passes through a modified roller carding machine, where the fibers are further separated, oriented and combined into a fiber sliver. The process is shown in Figure 1.

The resulting rCF/PA6 card sliver serves as feedstock material for the subsequent tape-forming process illustrated in Figure 2. The sliver is stretched to increase the alignment of the fibers and to achieve the final linear weight. It is then formed into the chosen geometry and fixed into an endless tape structure by first melting the PA6 fibers and subsequently cooling them down again. This fixation is carefully controlled to retain the desired cross-section while keeping the tape flexible enough to be wound and unwound without causing fiber damage.

The rCF-tape can currently be produced with a maximum width of approx. 30 mm. For this reason, the standard width of 1 inch (25.4 mm) is used when laying using AFP. Two generations of rCF-tape are produced, differing in their manufacturing process. The first-generation tapes are approximately 30 mm wide (Figure 3a) and require cutting to 1 inch on the sides (Figure 3b). The second-generation tapes are produced with a new tape-forming tool that allows direct production of a 1-inch-wide tape (Figure 3c). Therefore, these tapes do not require slitting, which reduces the process by one step and improves the environmental footprint. The second-generation rCF-tapes also have a higher basis weight, which improves stiffness, stability and processability.

To provide a benchmark for mechanical properties, samples are taken from sheets of vCF-tapes. The carbon fibers in these vCF-tapes are identical to those in the rCF-tapes and are type T700S from Toray Industries, Inc., Tokyo, Japan.

2.3. Production of Test Panels

To produce semi-finished carbon fiber products and process them into recycled carbon-fiber-reinforced plastic (rCFRP) parts, the rCF-tapes are laid up using the AFP process. This is carried out using the Coriolis Composites Csolo machine (Figure 4a). The rCF tapes are laid parallel to each other in a 0° direction with a varying number of plies depending on the desired thickness. To fill any gaps between the parallel tapes, the tapes in the subsequent ply are placed on the centerline of the tapes in the previous ply. This both fills gaps and ensures that the plies are bonded together in the scrim (Figure 4b). The parameters of the Csolo machine are needed to be adapted o process these rCF-tapes. Due to the variability in thickness and width inherent to recycled materials, the tape dimensions are less consistent compared to those made from virgin materials. To compensate for this, the lay-up speed was reduced to 250 mm/s and the laser power was increased to 400 W. In addition, a compaction force of 500 N was set for the AFP layup roller in order to generate sufficient local surface pressure and to adequately fix the tape. This adjustment was necessary because the binder in the tape did not appear to be sufficiently homogeneous, causing local tape detachment from the substrate. In combination with the laser power of 400 W, which corresponds to a generated temperature of approx. 220 °C at the nip point, and the already mentioned reduced layup speed of 250 mm/s, the layup flow and quality were significantly improved.

The stacks are pressed into panels (Figure 4c) using a variothermal hot-pressing technique with the LZT-OK-130-L press from Langzauner GmbH. The heating temperature is set at 245 °C and the pressing pressure is maintained at 20 bar. Test specimens for tensile and flexural tests are taken from these panels and their properties are evaluated according to standard test methods.

2.4. Mechanical Evaluation

Due to the high orientation of the staple fibers in the rCF-tapes, this material exhibits characteristics that lie between unidirectional (UD) materials and isotropic nonwoven materials. Currently, there is no clearly defined standard testing method specifically tailored for this hybrid material type. Therefore, to comprehensively evaluate the properties of the rCF-tapes, the produced composites are tested according to the standard for both unidirectional and nonwoven materials. Tensile tests are conducted in a 0° direction according to the requirements of both Part 4 (for isotropic and orthotropic fiber-reinforced plastic composites) and Part 5 (for unidirectional fiber-reinforced plastic composites) of DIN EN ISO 527. The specimens have different dimensions depending on the applicable standard: according to DIN EN ISO 527-4, the dimensions are 250 mm × 25 mm × 2 mm, while according to DIN EN ISO 527-5, the dimensions are 250 mm × 15 mm × 1 mm.

The four-point bending tests are performed in accordance with DIN EN ISO 14125. Again, the rCF material is evaluated under two classifications: Class II (plastics reinforced with mats, continuous mats and fabrics as well as mixed formats) and Class IV (unidirectional composites with carbon fiber systems). The specimens here are 15 mm wide and 2 mm thick, while their length varies depending on the class: 40 mm for Class II and 100 mm for Class IV.

3. Results

3.1. Fiber Volume Content

The FVC of the CFRP is determined chemically at the Fraunhofer IGCV laboratory in accordance with DIN EN 2564. This value is essential for comparison, as mechanical properties vary depending on the fiber volume content. The results indicate that vCFRP has a FVC of 52.75%, while rCFRPs made from the first- (rCFRP_Gen1) and second-generation tape (rCFRP_Gen2) have fiber volume contents of 40.72% (±0.16%) and 40.45% (±1.51%), respectively.

3.2. Tensile Tests of rCFRP_Gen1 Compared to rCFRP_Gen2

As shown in Figure 5, there are no significant differences between the samples of the same generation, despite using different test standards. However, the tensile properties of rCFRP_Gen1 are generally higher than those of rCFRP_Gen2.

The optical measurement with image editing and processing program ImageJ (Version 1.54d) shows that, on average, 0.58% of the sample area of rCFRP_Gen1 and 1.13% of the sample area of rCFRP_Gen2 has voids (see Figure 6). This means that the sample of rCFRP_Gen2 exhibits nearly twice as many potential weak points as the sample rCFRP_Gen1. The difference cannot be conclusively attributed to a specific influencing factor, but it may be due to quality fluctuations in the production of the rCF-tape or the manufacturing of the test panels. One likely influencing factor is that second-generation rCF tapes appear to have poorer consolidation. An increase in pressure from 20 bar to 30 bar could have optimized the consolidation quality and reduced porosity. Also, a higher temperature would increase the flowability of the molten PA6 matrix fibers. Due to limited material availability, those optimizations could not be performed in subsequent trials.

3.3. Tensile Tests of vCFRP Compared to rCFRP_Gen1

The tensile strength and modulus of vCFRP are adjusted (normalized) from 52% to 40% fiber volume content using the Rule of Mixture (ROM) equation to allow for better comparison with rCFRP [2]. This normalization is more accurate for vCFRP than for rCFRP, as the carbon fibers in the rCF-tape are not completely unidirectional and have finite length. Therefore, strict linearity cannot be assumed. A comparison of the tensile values (DIN EN ISO 527-5) shows that rCFRP achieves more than 68% of the strength of vCFRP, despite the difference in FVC. However, when the strength is normalized to the same FVC, rCFRP achieves approximately 88% of the tensile strength and 89% of the tensile modulus of vCFRP (see Figure 7). The rCFRP_Gen1 material achieved the highest tensile strength of 1141.33 MPa (±74.09 MPa) with a tensile stiffness of 77.04 GPa (±7.99 GPa) when tested according to DIN EN ISO 527-4 using 2 mm samples. When tested under DIN EN ISO 527-5 with a 1 mm sample thickness, it attained a tensile strength of 1037.15 MPa (±127.98 MPa) and a tensile stiffness of 79.65 GPa (±5.93 GPa). Thus, testing with ISO 527-4 and 527-5 leads to similar results for both tensile strength and tensile stiffness.

3.4. Flexural Tests of rCFRP Compared to vCFRP in 0° and 90°

As shown in Figure 8, the flexural strength of rCFRP_Gen1 (class II) reaches 81% of the stress observed for vCFRP (class IV) in a 0° direction despite the differences in FVC. In contrast, rCFRP_Gen2 (class IV) achieves only 70% of the stress of vCFRP (class IV) in a 0° direction.

The rCFRP_Gen1 (Class II) specimens achieve 60% of the flexural modulus of vCFRP (Class IV). Compared to the same test class (Class IV), rCFRP_Gen2 reach 61% of the flexural modulus of vCFRP. In numbers the rCFRP_Gen1 material reaches 742.99 MPa (±58.18 MPa) as flexural strength in 0° and 66.49 GPa (±2.06 GPa) as flexural stiffness in 0°. rCFRP_Gen2 material shows slightly lower strength and slightly higher stiffness with 640.01 MPa (±26.72 MPa) and 68.08 GPa (±1.88 GPa). In 90° rCFRP_Gen1 exhibits significantly higher flexural strength, with 126.39 MPa (±15.06 MPa), than Gen2 material (85.87 MPa (±4.37 MPa). In general, the rCFRPs have higher flexural properties at 90° than the vCFRPs. This is due to the more homogeneous and anisotropic fiber orientation in 0° and therefore higher number of fibers in 90°.

4. Discussion

In summary, it can be demonstrated that second-generation rCF-tapes perform better during processing via AFP compared to first-generation rCF-tapes. This improvement is likely due to their higher areal weight, greater homogeneity, and the resulting more uniform thickness and stiffness of the tape. Furthermore, the second-generation rCF-tapes do not have to be slitted, which is an additional benefit of the material, as no material is wasted during tape production, thereby reducing the carbon footprint. However, the mechanical properties of rCFRP_Gen2 are lower than those of rCFRP_Gen1. This may be due to the fact that rCFRP_Gen2 tapes were not slit in order to minimize material loss. Another potential explanation is that rCFRP_Gen1 underwent a more elaborate manual preparation process to handle the inhomogeneous rCF mixture, whereas a more cost-effective approach was chosen for rCFRP_Gen2. In the latter case, the fibers were only roughly processed, and short fibers were not sorted out. Since mechanical performance strongly depends on fiber orientation and overlap length, these factors could significantly influence the results. Furthermore, the lower mechanical properties of rCFRP_Gen2 could be explained by the number of pores observed in its panels. The microscopic images not only show a higher number of these pores, but also larger pores than in rCFRP_Gen1. This could indicate poorer consolidation of the second-generation tapes. Furthermore, the gaps between the layers in the stack play a role in the mechanical properties of CFRPs. Due to the higher basis weight of the second-generation tapes, one layer less was used in the stack. It can be assumed that a lower number of layers with the same panel thickness leads to poorer mechanical performance. However, further investigations are required to confirm these suspected causes, which were beyond the scope of the project. Despite the processing challenges, the rCF_Gen1 tapes exhibit very good mechanical properties.

Compared to vCFRP, rCFRP_Gen1 achieves 68% of the tensile strength and 68% of the tensile modulus without normalization of FVC, and 88% of the strength and 89% of the tensile modulus when normalized to the same FVC. In terms of flexural testing, the samples of rCFRP_Gen1 achieve 81% of the stress and 60% of the flexural modulus of vCFRP in a 0° direction. In 90° direction, the rCF-tapes outperform virgin tape material because of their higher degree of isotropy. These properties are more than twice of those of rCF nonwoven materials produced by comparable state-of-the-art technologies and show the clear potential of recycled fiber in tape form [11,12,13,14,15]. It should be noted that the rCF-tapes are not yet of consistent quality. Further optimization is required with regard to fiber orientation and distribution. In addition, rCF-tapes need to be produced with a higher FVC to allow a more accurate comparison with vCF-tapes. Whether a higher FVC leads to better mechanical properties remains to be clarified by further investigations. Although tapes made from recycled carbon fibers are still at an early stage of development, the results of this work demonstrate their great potential. They already exhibit attractive mechanical properties that are advantageous for the reuse of CFRPs. In addition, they can play an important role in closing a so far uncovered gap in the recycling loop. In contrast to other rCF semi-finished products, such as rCF nonwoven materials, they can be used to produce highly stressed structural components. Due to the better drapability of rCF-tapes compared to vCF intermediates, it is also possible to place more curved tracks in the AFP process without creating wrinkles and to realize more complex shapes. The effect is enabled by the short flowpaths of the thermoplastic matrix and inter-fiber sliding of the carbon staple fibers as long as the matrix material is in a molten state.

In summary, it can be stated that CFRPs made from rCF-tapes are not only environmentally friendly, but also have clear advantages over CFRPs made from vCF. Traditional applications such as aerospace and the automotive industry can also benefit from this. Sustainable lightweight construction can therefore make a valuable contribution to the energy transition.