Review of the International Conference on Composite Material (ICCM), Belfast 2023

At Belfast 30 July-4 August, there were more than 1,100 participants, with 500 oral (in up to 14 parallel sessions) and 300 poster presentations. The industry was well represented in the plenary and keynote lectures but much less in regular oral presentations and posters, which were often presented by PhD students or young postdocs, hence illustrating the “academic orientation” of the conference.

However, the industry greatly benefits from attending because of the opportunity to discover the most recent inventions, early-stage developments and to meet/recruit the best potential employees. Similarly, academics benefit from more substantial input from the industry to ensure they work on the most pressing problems that the industry is facing.

The focus of the conference was clearly on sustainability, sometimes more in a ‘greenwashing’ style of presentations (too easy claims like “We all work towards a more sustainable composites industry”), but also with exciting innovations (from academia) and serious commitments (from industry). This is a breakthrough compared to the previous ICCM conferences four and six years ago.

In the following text, all plenary and most keynote lectures will be covered, infused with impressions from the regular oral presentations.

Composite innovations towards sustainability

In his opening Scala lecture, Mike Hinton (Catapult, UK) stated that we live in interesting times, at the end of one long S-curve of innovations (from ‘fermentation’ to ‘maturity’) within the field of composites. The ICCM community is essential in supporting growth along these curves to mature existing technologies further, even if this does not create a disruptive technology. However, we should not forget that we also help to develop the next S-curve. We are living in disruptive times, with the advent of artificial intelligence, hydrogen technology and the enormously increased awareness of climate change. We should therefore work on technologies that may start a new S-curve.

At the end of his talk, Hinton made an interesting remark on the life cycle impact in terms of CO₂ emission of carbon fibre-reinforced plastic (CFRP) versus aluminium airplanes. Because of the 10:1 buy-to-fly – ratio of aluminium, and despite the higher ecological impact of CFRP (35 compared to 8 kg CO₂/kg raw material [1]), the life cycle impact of an aluminium aircraft is many times higher than for a similar CFRP airplane.

Chantal Fualdes (Airbus) claimed that the Airbus experience proved that a weight reduction of up to 30% for carbon fibre composite wings and up to 20% for CFRP fuselages is possible compared to equivalent aluminium structures. However, four challenges still remain to be addressed in the future. When designing for weight reduction, a ‘pyramid of simulation’ should supplement the time-consuming and expensive experimental testing and certification pyramid as quickly as possible. Maintenance costs should be further reduced, a major challenge being the composite fuselage dent assessment, for which a large database (> 2,500 damage reports) on impacts in metallic, short-range airplanes forms a precious source of experience. Increasing the production rate at an affordable cost is the third challenge, but the most challenging will be realising an end-to-end circularity by using innovative materials and structural concepts. Chantal Fualdes concluded that “collaboration with labs and universities is a strong asset, which will strongly support reliable modelling and simulation.”

Kenichi Yoshioka (Toray Industries) focused on the current innovations Toray is pursuing in developing more sustainable carbon fibres. As an introduction, he made the interesting statement that, although carbon fibres are rather energy intensive when being produced (from aniline synthesis over PAN polymerisation and fibre drawing to carbonisation), the reduction in energy consumption (and CO₂-emission) during the use of CFRP-intensive structures is impressive and outperforming the initially higher energy consumption during manufacturing. As an example, Yoshioka provided data on the lifetime CO₂-emission reduction during 10 years of use, for airplanes (medium-size, domestic flight type) in which 1 tonne of carbon fibres where used (replacing aluminium). The CO₂-emission reduction is estimated to be 1,400 tonnes, whereas only 20 tonnes is needed for carbon fibre production. Additionally, CO₂-emission data should be represented not only in terms of kg CO₂ per kg CF, but also per volume or per strength or stiffness when comparing carbon fibres with other structural materials.

Still, a further reduction of energy consumption and CO₂-emission during carbon fibre manufacturing should be pursued. Yoshioka commented on two interesting routes: first, biobased precursors are an option, both by using bio-sourced aniline or exploring alternative molecules like lignin. Second, important CO₂-emission reductions can be realised by using recycled carbon fibres: state-of-the-art pyrolysis can reduce the CO₂-emission for 1 kg of carbon fibre to less than 1/5, and Toray is developing a new pyrolysis method which further reduces it to less than one-tenth. But the continuous effort to make the (traditional) production process more energy efficient has in itself, over the past 25 years, reduced the CO₂-emission from 30 kg to 20 kg CO₂/kg CF, according to JCMA-data. Toray is performing fundamental studies to improve energy management in their operations further.

Tia Benson Tolle (Boeing) summarised the intentions of Boeing on how the company will concretise the circularity and sustainability goals as they have been laid down in the seventeen sustainability development goals of the United Nations Organisation. Because, according to Benson Tolle, aircraft is responsible for 2.5% of global CO₂-emissions and 12% of the emissions related to transport (!), Boeing has clear plans to reduce these numbers by taking different short-term measures but also identifying long-term goals: increase the use of SAF (sustainable aircraft fuel), increase efforts for an effective dismantling and recycling of aircraft via active participation in AFRA (Aircraft Fleet Recycling Association) and working towards a ‘commitment to fly net zero by 2050’, together with the other 40 members of ATAG, the Air Transport Action Group.

Because Spirit Aerosystems consumes more carbon fibres than any of the aerospace OEMs, the opinion voiced by Sean Black should not be undervalued. Black sees three major trends: completely new aircraft structures are underway, sustainability is getting more important, and automation will render composites ultra-competitive. In his opinion, electric aircraft will require much higher weight reduction, leading to different design solutions (blended wing body) and the use of new and optimised composite materials.

Modelling and simulations are key to further structural and manufacturing optimisation

Pedro Camanho (University of Porto) identified four drivers for innovation in composites: digitalisation of design and manufacturing processes, uncertainty quantification, and highly optimised and non-conventional composite materials. The former two drivers could be materialised by simulation across the scales, similar to multiscale modelling, as was advocated by many presenters during the conference. He introduced a meso-mechanical model that accurately represents directional, pre-localisation non-linearities and does not require pre-knowledge of fracture planes, which is realised by a novel self-consistent clustering analysis of damage. To address the challenge of uncertainty quantification, Camanho used a combination of higher-order theories, global-local approaches, and perturbation methods, leading to the calculation of probabilistic distributions of quantified properties of interest of complex composite structures.

Suresh Advani (University of Delaware) emphasised in his plenary lecture the importance of breaking the wall between manufacturing science and practice. This will help the industry move away from trial and error to set up manufacturing processes and enable a more deliberate and efficient methodology to optimize them. Modelling the manufacturing processes is vital in this regard, as it improves reliability, reduces development time, mitigates cost, and leads to better performance. He discussed modelling strategies for three main process types: fibre suspensions, squeeze flow and flow through porous media. Here are just three examples:

• modelling reorientation of fibres inside fibre suspensions can help improve the alignment of short fibres, which is critical for re-using recycled carbon fibres;
• modelling squeeze and percolation flow in the automated fibre placement process can help to ensure gaps are filled and overlaps are minimised;
• permeability and its variability are key to predicting mould filling and optimising any liquid moulding process.

In his keynote lecture, Tong-earn Tay (National University of Singapore) claimed that modelling should not only be multiscale (bridging all scales from micro to macro) but also ‘multi-fidelity’, meaning that a range of different mechanical performances of a composite structure should be reliably predicted. He elaborated on the difference between discrete and ‘smeared’ modelling approaches, the former being precise but time-consuming, the latter faster but encountering convergence problems. Tong-earn Tay proposed integrating both approaches, resulting in high accuracy and faster modelling, up to 20% faster for OHT (open hole tension) and up to 50% faster for impact modelling.

Reza Vaziri (University of British Columbia) discussed progressive damage models and their importance in advancing composites science. These simulations require high-quality experimental data to set up the constitutive models, which is often challenging to obtain. The quality of the predictions is governed more by the quality of these constitutive models than by the details of the model. Setting up these constitutive models often involves calibration based on time-consuming tests with complex data reduction. However, there are potential solutions to circumvent these challenges. Theory-guided machine learning and physics-informed neural networks can help embed more physical behaviour, thereby improving the reliability of inverse identification. Combining machine learning with finite element analysis and adding robust statistical sampling is a crucial enabler for the simulation of large-scale composite components, considering the inherent uncertainty of composites.

Some recent innovations in composite applications

In the framework of NESTS, the New European Space Transporation Solutions programme of ESA, launched in 2021, Ariane Group has been awarded several design contracts for space transport solutions for the post-Ariane6 period 2030-2050. However, already for the Ariane 6, composites will be crucial to achieve significant weight reduction, as explained by Markus Quadt of the Ariane Group, in Bremen. For instance, for the ICARUS module, which is the future upper stage of Ariane 6, an innovative carbon fibre composite architecture will enable an ultra-lightweight design and integrate cryogenic functions in the structure itself. Moreover, a reduction in recurring costs and in greenhouse gas emissions will be achieved compared to a more traditional aluminium architecture. To reduce the manufacturing costs, the manufacturing technology will be matured by building the Phoebus demonstrator (using amongst others automated fibre placement), which should be ready by 2026. The lightweight ICARUS should be ready by 2029, allowing for an increase of 33% of the payload (from 1,500 to 2,000 kg).

Mark Braniff (Spirit Aerosystems, Belfast) presented a nice didactical story on how the company did build upon the centuries old textile machinery and weaving industry in Belfast, when they decided to develop 3D textiles to create T-shaped preforms for structural wing parts (in collaboration with Ulster University). Unfortunately, the idea never could be upgraded to industrial level, but it initiated fundamental research in Resin Transfer Infusion (RTI), which was needed to infuse these rather complex 3D textile structures. In this patented RTI process, the resin is injected into the dry fabrics structure while it is in an autoclave. The process allows for the manufacture of large, one-piece complex structures, reducing the need for many different parts and mechanical fasteners. It is now heavily used by Spirit Aerosystems for manufacturing the composite wing for the Airbus A220. Many challenges, like keeping the porosity below 0.25% and reducing the fibre waviness, have been successfully addressed to achieve this important milestone. In the future, automated fibre placement will be introduced as a supplementary technology.

Philip Crothers (Boeing) talked about the need for high-rate manufacturing. While this is typically not seen as an issue for the aerospace sector, the current outlook for future aircraft would require 23 million parts per month. Composites can achieve these rates, but many practical challenges remain to overcome. He identified three main challenges that the composites community should try to address: rate, scale, and sustainability. Another vital component is the repeatability of the process and part quality. Process automation is key to achieving this, which would also help address the rate challenge.

Luke Djukic (Omnitanker, Australia) designs aerospace and transport tanks, including for chemical transport. The company has a unique, patented method for bonding thermosets to thermoplastics. He discussed the importance of avoiding welds to help minimise or prevent environmental stress cracking. Regulations are essential to ensure the safe operation of composite tanks, especially when they contain hydrogen or chemicals. The United Nations regulations play a vital role here, but unfortunately, there are no regulations yet for liquid hydrogen composite tanks at -253°C. Given their importance in enabling zero-emission airplanes and the amount of work presented at ICCM on such tanks, such regulations are urgently needed.

Some breakthrough research achievements in universities.

Conchúr Ó Brádaigh (University of Edinburgh) talked about the latest developments in the field of in-situ polymerisation of high fibre volume fraction composites. These resins are processed like thermosets but become thermoplastic polymers upon curing. There are exciting developments in this field that can help make composites more sustainable. Examples were shown where placing activators inside the sizing can help improve the in-situ polymers’ performance. One remarkable result was that a composite corner was reprocessed and showed no drop in flexural strength, highlighting the benefits of thermoplastic over thermoset composites.

Paul Weaver (University of Limerick) discussed the concept of ‘varicomposites’, which are composites that can vary spatially and temporally. Given the complexity of such composites, they must be manufactured with minimal defects. One example of varicomposites is tow-steering or variable angle tow composites. In such composites, the local tow orientation is modified to improve buckling resistance or the performance around holes. Examples were shown of a non-spherical pressure vessel made of these composites, which enabled it to be bend free. This would enable vessels that can be packed into small spaces more efficiently. Another example is morphing composites, which can exploit buckling to reduce the stiffness locally and temporarily. Examples were shown of the use of morphing composites to fold blades, so they can be transported more efficiently.

Philip Withers (University of Manchester) explained the difference between synchrotron and lab-scale computed tomography (CT). Synchrotron CT tends to have a parallel beam and target high resolutions, whereas lab-scale CT has a conical beam and typically lower resolutions and slower acquisition. Both techniques are crucial to understanding composite microstructures and failure behaviour. Many examples were shown to highlight the capabilities of CT and to inspire people on how they can use CT in their research. A few notable highlights:
• cracks that are difficult to detect with conventional techniques due to the lack of crack opening can be visualised through by exploiting phase contrast in synchrotron CT or staining agents;
• 3D imaging can also help to monitor the infusion process, helping to establish how voids develop;
• CT images can be used to build 3D models of the microstructure.

In his keynote lecture, Connor Mc Carthy (Uni Limerick) reported on a European project aiming at developing technologies for “designing for disassembly”. Joining of composites to metallic structures can be either chemical (bonding), thermal (welding) or mechanical (riveting). The FALCOM joining concept is a combination of chemical and mechanical joining: two interlocking surfaces are created, in the metallic part by laser cutting and in the composite part by laser cutting or compression moulding, and the parts are then joined by adhesive bonding. The strength and energy to failure, both measured using a lap shear test, increased by respectively 65% and 800%. When using a reversible adhesive, containing iron oxide-based nanoparticles which swell during heating, the joint becomes reversible, and hence could be used in automotive applications.

A final remark

In total, almost 500 oral papers have been presented in 14 parallel sessions. It is simply impossible to give a global overview, but one trend was remarkable: several large, mainly European collaborative research projects have been organising a series of sessions presenting the ongoing research in these projects. This was not always successful because some of these projects were still in a start-up or early stage, and hence little results and merely only planned research could be presented. A strong exception was the series of sessions on compressive strength of composites, jointly organised by Imperial College London and the University of Bristol. The lectures not only covered research outcomes of their large joint project NextComp but also invited researchers from other universities to present their work on the compression strength of composites, an often-under-estimated challenge for composite applications. Combining presentations from a large research project with others from outside the team seems to be a concept to be followed up on in future ICCMs.

This article has been written by Ignaas Verpoest, Emeritus Professor at the Composite Materials Group of the Department of Materials Engineering, KU Leuven, and Yentl Swolfs, Research professor and the coordinator of the Composite Materials Group at KU Leuven.

Reference:
[1] There is much debate on the exact values for CFRP. JCMA for example claims much lower values.