Influence of carding parameters on the mechanical properties of carded recycled carbon fibers with a Weibull statistical approach
Topic(s) : Material science
Co-authors :
Jean IVARS (FRANCE), Ahmad Rashed LABANIEH , Kim-Phuc TRAN , Damien SOULAT (FRANCE)Abstract :
By 2050, the aerospace industry alone is expected to produce approximately 500,000 tons of carbon fiber reinforced plastic (CFRP) waste [1]. Thermal and chemical recycling techniques are used to recover and preserve the carbon fibers' mechanical properties. One of the main challenges is then to effectively realign and repurpose these fibers for high-performance applications.
Addressing this challenge, several projects at laboratory scale have emerged in recent years, each employing unique processing techniques. HiPerDiF aligns fibers using a process involving fiber-water jets and slotted plates [2]. In a different approach, MANIFICA produces highly aligned recycled carbon fibers (rCF) conveyed through a vibrating lattice [3], and, more recently, friction yarns made from rCF carded slivers have been developed using a DREF machine [4]. These innovative methods are some ways for the manufacturing of rovings based on rCF, for which orientation and properties need to be controlled. To achieve this aim, the influence of the parameters of each stage of the manufacturing process, starting with the carding stage, on these properties must be precisely analyzed.
In this study, three parameters are examined for their impact on the elementary properties of rCF extracted from webs. These parameters include the ratio of thermoplastic fibers mixed with rCF in the web, the speed ratio between the cylinders, and the number of carding cycles. For tensile characteristics individual fibers are mounted on cardboard frames with a 25mm gauge length with a minimum sampling of 50 fibers for each web. As described in literature, carbon fiber exhibits brittle fracture behavior [5], for this type of mechanical behavior, a common statistical method is the Weibull statistical model, which evaluates the probability of failure on all tested samples.
For all parameters considered in the experimental design, the analysis of all data is based on the Weibull statistical approach to stress at break, as well as on the values of tensile modulus and elongation at break. The first step involves determining which parameters of the Weibull distribution best match the experimental data, using methods such as the Maximum Likelihood Estimator. Next, a statistical analysis, involving the Sum Square Error (SSE) and the Kolmogorov-Smirnov test (K-S Test) will be conducted to ensure theoretical distribution matches the experimental data.
Finally, by comparing the parameters of the Weibull distribution and the conventional statistical parameters (mean and standard deviation), the study aims to assess the impact of the three carding parameters on the mechanical properties of rCF.
Addressing this challenge, several projects at laboratory scale have emerged in recent years, each employing unique processing techniques. HiPerDiF aligns fibers using a process involving fiber-water jets and slotted plates [2]. In a different approach, MANIFICA produces highly aligned recycled carbon fibers (rCF) conveyed through a vibrating lattice [3], and, more recently, friction yarns made from rCF carded slivers have been developed using a DREF machine [4]. These innovative methods are some ways for the manufacturing of rovings based on rCF, for which orientation and properties need to be controlled. To achieve this aim, the influence of the parameters of each stage of the manufacturing process, starting with the carding stage, on these properties must be precisely analyzed.
In this study, three parameters are examined for their impact on the elementary properties of rCF extracted from webs. These parameters include the ratio of thermoplastic fibers mixed with rCF in the web, the speed ratio between the cylinders, and the number of carding cycles. For tensile characteristics individual fibers are mounted on cardboard frames with a 25mm gauge length with a minimum sampling of 50 fibers for each web. As described in literature, carbon fiber exhibits brittle fracture behavior [5], for this type of mechanical behavior, a common statistical method is the Weibull statistical model, which evaluates the probability of failure on all tested samples.
For all parameters considered in the experimental design, the analysis of all data is based on the Weibull statistical approach to stress at break, as well as on the values of tensile modulus and elongation at break. The first step involves determining which parameters of the Weibull distribution best match the experimental data, using methods such as the Maximum Likelihood Estimator. Next, a statistical analysis, involving the Sum Square Error (SSE) and the Kolmogorov-Smirnov test (K-S Test) will be conducted to ensure theoretical distribution matches the experimental data.
Finally, by comparing the parameters of the Weibull distribution and the conventional statistical parameters (mean and standard deviation), the study aims to assess the impact of the three carding parameters on the mechanical properties of rCF.