Current state-of-the-art materials in aircraft cabins are very stiff and lightweight sandwich panels used for instance in the ceilings, linings and hat racks. These sandwich panels contain a core made of aramid fibres and phenolic resin. Glass fibres are used in combination with the phenolic resin as prepreg-reinforcement for the top-layers of the sandwich (Fig. 3). These very light and stiff panels are manufactured in a heated press. No ecologically satisfactory end of life treatment is available for such sandwich panels because of the usage of phenolics and glass/aramid fibres. The separation of such heterogenous material combinations embedded in a crosslinked thermoset resin is very complex [2, 3].
The secondary structural parts require materials with higher mechanical properties. Because of weight and mechanical requirements, continuous carbon fibres are mainly used for such parts. To protect the electromagnetic interference and particularly the lightning strike, electrical conductive structural composites are desired to use in the critical zones of aircraft. The state-of-the-art carbon composites cannot satisfy the requirements. This is a big technological challenge for the polymer matrix composites [4,5,6,7,8,9,10].
A multitude of bio-fibres is available on the world market. In Europe, flax and hemp fibres are the most common bast-fibres used to reinforce composites. Ramie fibres that are grown in China are another suitable candidate. A drawback of natural fibre reinforced plastics (NFRP) is their lack of strength compared to glass and carbon fibre reinforced plastics (GFRP, CFRP). In theory, natural fibres can reach very high values for tensile strength of up to 1000 MPa but due to imperfections (kink bands) and incompatibilities to certain resin systems their potential cannot be fully used [2, 10]. The poor interfacial bonding properties between hydrophilic natural fibres and hydrophobic polymers lead to low mechanical properties of NFRP. Therefore, improving the interfacial strength and toughness is necessary in order to improve facilitate their full potential. On the other side, the low density of bio-fibres leads to good specific stiffness values comparable to GFRP with further advantages on acoustic and thermal damping due to their hollow structure, the lumen (Fig. 4) .
Based on previous successful research experiences on using chemicals and CNTs to improve the interfacial properties of NFRP, the application of nano-natural cellulose to modify the interfacial and interlaminar properties of the composites will be investigated . In a Chinese study, zirconia nanoparticles (ZrO2) were designed and grafted onto flax fibres by hydrogen bonds as seen in Fig. 5 . The results are an increased tensile strength of the grafted flax fibres while the tensile modulus is not affected. Another positive effect of ZrO2 grafted fibres is the reduced grow of fungal colonies on flax fibre reinforced epoxy composites.
Another topic of ECO-COMPASS is the evaluation of recycled carbon fibres (rCF). New carbon fibres are very expensive due to their energy intensive production process. It is, therefore, of high importance to reuse these valuable fibres in order to save energy, raw materials and cost. Pyrolysis of CFRP is a process that has made it into industrial use recently to regain carbon fibres from composite waste. These recycled carbon fibres (rCF) are available in milled and chopped form. The restricted length and the removal of the fibre sizing are their main drawbacks compared to virgin carbon fibres (vCF) [14, 15]. It is, therefore, at the moment not possible to give these “downcycled” fibres the same function as virgin carbon fibres.
In parallel to classic reinforcement types used in aviation, the combination of renewable bio-fibres and recycled carbon fibres in a hybrid non-woven will be evaluated as an alternative way to find applications for a rising amount of recycled carbon fibres (Fig. 6). Recycled carbon fibres have short and variable length which makes it very difficult to convert them into continuous yarns for traditional woven reinforcements. Non-woven processes are capable of combining different types of fibres of variable length in a single web structure for composite production. Nonwoven processes are also less expensive and more eco efficient compared to classic woven fabrics from bio-fibres due to their simpler production process . Renewable bio-fibres such as flax and PLA have been successfully combined to produce bio-degradable nonwoven composites [16,17,18]. The combination of recycled carbon fibres with bio-fibres could enable the designer to optimize multifunctional eco-efficient composites using their inherent advantages.
Bio-based polymeric resins
Today, thermosets are the most important polymer family used in the aviation industry , due to their versatility, high performance and the wide span of applications they comprise. Mainly phenolic and epoxy resins are used for the interior panels and secondary structures of aircraft [20, 21]. However, their petrochemical base and the difficulty of thermosets to be recycled, forces the industry to seek for feasible alternatives that can reduce the ecological footprint associated to their production .
In ECO-COMPASS, promising bio-based resins will be assessed for their use in aerospace applications. Thermosetting epoxies from rosin, itaconic and gallic acid (Fig. 7) have been identified as candidates from China, due to their aliphatic-cyclic structure which can outperform existing solutions in terms of mechanical properties and chemical resistance [22,23,24]. Other bio-based resins show very promising fire properties comparable to phenolic resins. As an example, Fig. 8 shows the results of flammability tests with glass and bio-fibre reinforced composites. The burn length is the distance of damaged area to specimen edge due to the combustion during the flammability test. Selected (at least partly) bio-based resins [epoxy (EP), furfuryl alcohol based resin (FUR) and linseed acrylate resin (ACR)] have been compared to phenolic resin (PF) as reference. The furfuryl alcohol based resin shows results in the range of the phenolic reference when it is reinforced with glass fibres. Furthermore, the need for flame retardants in case of using bio-sourced fibres is clearly visible as none of the specimens reinforced with flax fibres passed the test with the threshold of a maximal burn length of 152 mm .
The incorporation of additives within the resins will be analysed to enhance the performance of neat resins and subsequent composite manufacturing. With this purpose, application of carbon based additives such as graphene, graphene oxide  or CNT , as well as silicon carbide nanoparticles (nanotubes or nanowhiskers) could be highly efficient in terms of thermal conductivity enhancement [27,28,29], along with electric conductivity  and enhancement of mechanical properties [31, 32]. Furthermore, specific coupling agents will be employed to enhance the compatibility with additives and fibres. On the other hand, fire resistance properties will be enhanced by means of nanotechnology and innovative halogen-free phosphorous based reactive fire retardant additives, covalently linked to the neat resin .
The final composite material properties are related to the control of the processing parameters during the manufacturing. The choices of the appropriate pressure, temperature and curing time are matter of importance to obtain low porosity contents and an important extend of cure.
As different bio based materials will be considered in the ECO-COMPASS project, a physico chemical characterization campaign will be done to identify the polymerization kinetics and the rheological behaviour of the materials to be transformed. These tests, which will also be carried out on formulations with fillers for multifunctional properties enhancement, will allow the identification of the most suitable process windows to guarantee the best compromise between high material properties, and limited curing cycle time. As bio based materials are usually very sensitive to the environment and show bigger deviation standards on properties than “non bio-based” materials, such definition of the curing cycles parameters are critically important to maximize the process reliability and to increase the repeatability of the final properties of the materials manufactured.
Then, composite laminates and sandwich composites will be manufactured by autoclave and hot press. Thanks to resin and fiber formulations with fillers, an enhancement of the material performances is targeted, such as structural and mechanical properties, structural damping, fire/smoke/toxicity and hygrothermal ageing properties. Samples lay ups and dimensions will be chosen according to aeronautical specifications and standards, and material health will be preliminary controlled by ultrasonic testing before characterization.
Material protection will also be evaluated to improve the durability of the material manufactured regarding environmental attacks and fire smoke and toxicity properties, but will require a study of the compatibility of the material with these solutions. For the different applications (secondary structures and interior parts), a special attention will be paid to the total weight increase induced by the material protections, which will be parameter for the choice of the coating. Of course, synergies between the different requirements will be looked for between the different solutions evaluated.
Modelling and simulation
The objective of ECO-COMPASS research is to design, improve and optimize the eco-composites to be used in an efficient way in (semi-)structural parts. The mechanical–numerical proposal for such purpose consists of an investigation addressed to obtain good mechanical properties, durable, and resistant eco-composites based on rational analysis that provides by the adaptation of generalized mixing theory and/or multiple scale homogenization theory [34,35,36], derived from the formulations for the classical composite material, and all these mechanical formulations within a framework provided by the genetic algorithms optimization [37, 38]. So, the proposed procedure promises a detailed behaviour study of the whole composite, starting from each one’s simple component behaviour. It seeks to obtain sustainable materials which are also mechanically and thermally efficient.
Multiscale procedures are based in analysing a material model (micro-model), assuming a periodic distribution of the material within the structure. This analysis provides the material response, which can be used in a structural model (macro-model) to obtain the global performance of the structure. There are several approaches in which a multiscale procedure can be defined. The generalized theory of mixtures or serial/parallel mixing theory [34, 39] proposes a phenomenological homogenization in which the composite performance is obtained from the constitutive models of its components and some closing equations that define how these components interact among them. This formulation is capable of accounting for complex failure procedures such as delaminations, with an affordable computational cost [40, 41].
The other multiscale approach that will be used consists in obtaining the composite performance from the analysis of a numerical model of a representative volume element (RVE). The boundary conditions to be applied at the RVE come from the macro-model and the response obtained from the RVE is transferred to the structural model. An example on how a multiscale analysis works is shown in Fig. 9, in which are plotted the stresses obtained in the macro and micro scales for a clamped beam made with a pultrusion composite. As it is shown by Otero et al. in  the main drawback of this approach is its computational cost. For this reason, ECO-COMPASS project will look into two different strategies to reduce it. One consists of defining a comparison parameter capable of predicting if a given material point might have reached its threshold stress–strain state. His approach has been formulated in  and has shown an excellent performance. The second approach consists of analysing the failure of RVE under different stress–strain states to create a material database defining the failure threshold and its evolution. This approach requires an initial computational effort to define the database but, afterwards, the structural simulation can be conducted quite easily.
All formulations developed to analyse composite structures require, also of accurate material models to characterize the constituent materials. The working group involved in ECO-COMPASS will use advanced constitutive models for the materials [44, 45], together with specific formulations to account for large anisotropy behaviour , or plastic mechanical damage and moisture content . The optimal material design [37, 38] in terms of the structure and its uses depending on the climatic conditions of the place also will be considered through thermal treatment of conduction and diffusion.
The use of virtual design and optimization models that will be developed in ECO-COMPASS will reduce the development time and cost of multifunctional eco-composites by reducing the number of manufacturing trials and experiments. Investigation is carried out on the understanding of material parameters and processing factors affecting the mechanical, thermal and electrical properties of bio-polymer nanocomposites as well as the electromagnetic shielding properties and lightning strike behaviour of eco-composites using numerical [48,49,50,51] and/or analytical models. The material parameters considered are the nanofiller dimensions, the nanofiller configuration, the properties of the nanofiller/matrix interphase and the properties of the bio-polymer and fibres. The processing factors considered are the nanofiller volume fraction and the formation of agglomerates of nanofillers. The investigation is conducted by means of representative unit cells (RUCs) of nanofiller agglomerates developed using the DIGIMAT software. The RUCs are solved numerically using the finite element method and analytically using the Mori–Tanaka method. At the same time, homogenization of the RUCs is applied through the use of periodic boundary conditions. The models will receive input from microscopy images (SEM, AFM, etc.) and will be validated against mechanical, thermal, electrical, EMI and lightning strike tests.