1. Introduction
The environmental impact of aviation has received attention because aviation accounts for 2.4% of global CO
2 emissions [
1] and these emissions are forecast to at least double by 2050 [
2]. Adding the effects of non-CO
2 emissions, such as NO
x, water vapour and sulphate and carbon particulates, potentially further doubles the contribution of aviation to climate change [
3]. The non-CO
2 emissions per unit of fuel burn lead to far more detrimental effects on the local air quality than on the ozone layer [
4], whilst the noise has a direct impact on human health.
To reduce the environmental impact of air travel, both the National Aeronautics and Space Administration (NASA) and the European Commission have put forward ambitious targets for the US and European aviation markets, respectively. The Advisory Council for Aviation Research and innovation in Europe (ACARE) has developed the Flightpath 2050 vision to achieve a 75% reduction in CO
2 emissions, a 90% reduction in NO
x and a 65% reduction in noise [
5]. The International Civil Aviation Organization (ICAO) is also promoting legislation targeted at regulating airline emissions internationally. The Carbon Offsetting and Reduction Scheme for International Aviation (CORSIA) requires airlines to offset CO
2 emissions of international flights that exceed 2020 levels [
6].
The projected incremental improvements in fuel efficiency of about 2% per year, however, are not sufficient to counteract the increase in passenger volume. Disruptive designs of propulsion systems, secondary systems and airframes will be required to reach the emissions objectives. The ultimate aspiration for the future of “green” aviation is a large scale
all-electric aircraft (AEA) [
7], as this configuration has the potential for zero emissions directly from flight. Hence, the UK has developed an aerospace technology strategy [
8] and has launched an industry–government partnership, the Jet Zero Council [
9], aiming to deliver zero-emissions flights through innovative technologies. The approaches toward achieving these emissions targets include improving the combustion efficiency of jet engines, and the potential adoption of hydrogen [
10] or other sustainable aviation (bio)fuels.
Aircraft electrification has been an increasing focus of academic and industry research over the last decade. All-electric designs have been demonstrated for small air vehicles. However, such prototypes have not been scaled up to more than ten passengers due to the specific energy (
E*) limitations of current battery technology. Note that in this article, all
E* values refer to useable or required pack-level values unless otherwise stated. A significant proportion of the energy expenditure would be used to transport the mass of the batteries; this mass would not decrease during a flight as would that of conventional fuel. Existing state-of-the-art Li-ion batteries (≈250 Wh/kg at the cell level) have
E* ≈ 170 Wh/kg when packed into a battery with a suitable casing [
11,
12]. This
E* is approximately fifty times lower than that of kerosene, even including the fuel tank weight (
E* ≈ 8.9 kWh/kg [
10]). Some promising battery technologies are projected to reach cell-level
E* ≈ 400–700 Wh/kg with the potential for scaling up to commercialization within five years [
13]. However, high
E* values potentially introduce a host of problems for aviation related to safety concerns and battery thermal runaway. Even current Li-ion batteries onboard conventional aircraft have caused fires due to overheating, notably on the Boeing 787 Dreamliner [
14]. Since electric propulsion is a key aspiration for future commercial airliner designs, several concepts have been developed in response to the engineering challenge of the electric propulsion system (
Table 1). In many of these studies, the range of the aircraft has been optimised against the batteries’ specific energy, which has been recognised as the most critical obstacle to large scale electric aviation. As a result, electrified aircraft concepts are typically differentiated by their approaches to reducing the power and energy demands on the battery packs.
A more immediate goal of commercial aviation is the
more-electric aircraft (MEA) which involves the electrification of only the subsystems. This configuration removes the pneumatic and hydraulic systems and has already been introduced into service to some extent in the Boeing 787 Dreamliner. AEA designs are based on the adoption of fully electrified subsystems. The initial approach for many AEA projects has involved researchers trying to recreate conventional aircraft configurations with electric propulsion and associated electronics that would be comparable with the existing gas generator turbine systems [
20]. Such approaches would lead to such electric aircraft designs being handicapped by the state of electronics technology. For this reason, scaling up existing small AEA design as a strategy towards an all-electric commercial airliner has not been fruitful. On the other hand, electrical propulsion allows the baseline configuration of an aircraft to change through disruptive integration strategies. Coupling different disciplines of study during the broader conceptual design phase could lead to a fundamentally unique vision for a future airliner.
Hybrid-electric propulsion (
Figure 1) represents another strategy for combining different systems’ capabilities. As an intermediate step in realising the fully electric propulsion system,
hybrid-electric aircraft (HEA) integrate electric motors with gas turbines into the same designs, thereby combining combustion with electrical power to improve performance. In the turboelectric HEA, electric motors are driven by energy from turbofans which alleviates the need for battery energy storage. The engines can be downsized and optimised for cruising. Thy can operate closer to their design points during cruising, where they have maximum efficiency, than conventional engines.
Distributed propulsion (DP) (
Figure 2) represents an integration strategy that exploits the unique characteristics of electric propulsion.
Figure 2 shows a potential configuration where the traditional jet engines under the wings are replaced with multiple smaller electric engines towards the trailing edges of the wings. Electric motors can be scaled down without the performance penalties that limit the scaling of gas turbines. This allows engineers to distribute thrust along the wing and thus achieve superior aerodynamic performance. Moreover, the freedom that electric motors permit regarding their scaling and integration means they can also utilise boundary layer ingestion (BLI). Boundary layer ingestion decreases the engine’s inlet flow rate. This lower flow velocity reduces the engine’s power demands and thus reduces the energy storage requirements [
21]. Distributed propulsion with BLI is employed in the NASA N3-X concept [
19]. The Dragon concept, on the other hand, combines a turboelectric architecture with DP integrated with the wing section to reduce drag [
16].
Electric propulsion can also be coupled with aerodynamics through alternative wing designs. The BHL Ce-Liner concept utilizes a C-wing design with long C-shaped winglets to improve lift-to-drag ratio and thus decrease power and energy consumption [
18]. Other novel wing configurations that improve aerodynamic performance include the strut-braced wing (SBW) (
Figure 3a) and the box wing (BW) (
Figure 3b). NASA’s Subsonic Ultra Green Aircraft Research (SUGAR) Volt concept incorporates an SBW to improve drag and minimise weight [
17]. The wing–body design of the NASA N3-X is the most ambitious wing configuration which, when combined with DP, shows the greatest improvement in fuel efficiency [
19]. Each of these novel wing configurations leads to different challenges in terms of optimising aerodynamic and structural performance and passenger comfort. The improvements from propulsor and airframe/wing integration highlight the potential value in exploring energy storage and airframe integration through structural power composites (SPCs).
Structural power composites (SPCs) [
25] are multifunctional materials where one or more of the constituents of the material simultaneously perform(s) load carrying and energy-storing functions. For example, carbon fibres can be used to carry structural loads, store electrical energy and act as current collectors to conduct electrons. Similarly, structural electrolytes made from mixtures of epoxy and ionic liquid surrounding the carbon fibres can transfer both mechanical stresses and ionic charges. SPC technology has been developed over the last two decades, and its main advantages over existing systems are the potential for considerable systems-level mass and volume savings [
26], which are becoming increasingly important for many applications such as surface and air transport and consumer electronics. Structural power composites can be further categorized by the type of electrical energy storage device. The two types which receive the greatest attention are structural battery composites (SBC) [
27,
28] and structural supercapacitor composites (SSC) [
29]. SPC technology has been demonstrated in full scale automotive structures (
Figure 4a) incorporating SSCs (
Figure 4b) to power the rear lighting.
SPCs offer an additional or alternative solution to meeting the high electrical energy demands of AEA. Structural power reduces the parasitic weight penalty of traditional batteries and hence mitigates the high energy density requirements. Thus, SPCs potentially enable AEA which were previously considered unachievable or require alternative fuels such as liquid hydrogen. Although the technological, economic and environmental prospects of large AEA have been assessed in previous studies [
15,
31], such studies have not considered the adoption of SPCs in the aircraft. The study reported here investigated the potential impact that an SPC airframe would have on the design of future commercial aircraft. The electrochemical performance requirements of the SPC were evaluated for different aircraft configurations, including configurations previously regarded as non-optimal. The investigation was concluded by assessing the environmental impacts of the designs considered.
Introducing SPCs into aviation is especially challenging due to certification issues, including fire resistance, damage tolerance and cyclability [
25]; however, the development of predictive modelling [
32] may help to address these issues. SPCs are still emerging technology, but the prospects they can offer will have profound implications for the aviation industry. The mass savings resulting from the integration of SPCs, as confirmed by various studies concerning their integration in electric vehicles [
33], are a very attractive feature for aerospace applications. Electric aircraft range increases of 11% to 66% have been predicted if a battery could be substituted with SPCs in airframes [
34]. The structural electrolyte in the SPC may mitigate issues associated with separator shrinkage and short-circuiting [
35] at low temperatures, and therefore may offer superior fire-resistance to that of batteries [
34]. Distributed energy storage using SPCs may offer different thermal management options and may self-passivate in the event of a fault or penetration. Moreover, SPCs can offer localization of power sources, which has the potential benefits of increased safety and further reductions in wiring and cooling system masses.
Studies on early adoption routes for SPCs in conventional aircraft have considered powering auxiliary systems such as those in the aircraft cabin. Replacing the floor panels of the cabin with SPCs that would power the in-flight entertainment system has been modelled, demonstrating that mass saving of over 260 kg per aircraft (approximately 2% of the maximum payload mass on the A220-100) could be achieved if the SPC could meet specific energy, specific power and in-plane elastic modulus targets of approximately 144 Wh/kg, 300 W/kg and 28 GPa, respectively [
36]. A feasibility analysis of SPC integration for small electric aircraft focused on replacing the structures within the Airbus E-fan 1.0 and Bristol Eco-Flyer, both two-seater AEA, with SPCs [
37]. A specific energy of 52 Wh/kg and specific power of 103 W/kg would be required for the SPC to fully power the aircraft. A higher specific energy of 122 Wh/kg could increase endurance by 31%. A more detailed structural analysis considering the adoption of structural batteries in a four-seater general aviation aircraft with a serial hybrid-electric propulsion (Hybris) presented a procedure to size an airframe based on a weight-optimal approach [
38]. In contrast, the study reported here considered such analyses on large commercial aircraft together with slender wing and electric propulsion system configurations.
SPCs have the potential to replace batteries and improve the system-level performances of aircraft. However, perhaps more critically, SPCs are such a profoundly different way to power aircraft that they warrant rethinking the traditional approach to conceptual aircraft design. Instead of bracketing the design of different systems within the aircraft, a more integrated approach will be essential to enable novel, disruptive aircraft configurations using SPCs and slender wings to reduce drag. For internal combustion engine aircraft, a drawback of slender wings is that there is a significant reduction in the fuel tanks’ total volume inside the wings. For SPC airframes, there may be no (or less) need for fuel tank (or battery pack) volume; hence these slender wing configurations both facilitate efficient electric aircraft designs and are synergistic with an SPC airframe.
Since SPC adoption presents significant challenges, the first step is to assess the feasibility of using SPCs for this application. The aim of this study was to adopt a system-level approach and exploit the synergy between the airframe and the power system of future large aircraft using SPCs. Due to the current low level of maturity of SPCs, the performance of existing SPCs was not used as a constraint, but rather the performance requirements were prescribed from the analysis. Ultimately, the main aim was to guide research efforts to the performance levels that need to be reached if SPCs are to be integrated into large commercial aircraft. A secondary aim was to provide insights into potential adoption strategies for these materials that are unconstrained by the current performances of existing materials.
This multifunctional design study focused specifically on the electrification of large civil aircraft. Since both technologies, SPCs and electric propulsion for large aircraft, are projected to mature over the next thirty years, it was fitting to analyse them in the same context. However, we do not present comparisons against either hydrogen or alternative fuels because there are still many unknowns regarding issues such as cryogenic storage and carbon capture. Furthermore, SPC technology could potentially be used, not only as an alternative, but together with other low carbon technologies, such as hydrogen or sustainable aviation fuels, to improve fuel efficiency. Conceptual designs of future aircraft configurations are outlined herein to identify the role structural power could have in developing those concepts further. By outlining the potential and the limitations of structural power for future aviation, definitive goals can be set for the further development of the technology.