Analysis | Published:

# Technological, economic and environmental prospects of all-electric aircraft

## Abstract

Ever since the Wright brothers’ first powered flight in 1903, commercial aircraft have relied on liquid hydrocarbon fuels. However, the need for greenhouse gas emission reductions along with recent progress in battery technology for automobiles has generated strong interest in electric propulsion in aviation. This Analysis provides a first-order assessment of the energy, economic and environmental implications of all-electric aircraft. We show that batteries with significantly higher specific energy and lower cost, coupled with further reductions of costs and CO2 intensity of electricity, are necessary for exploiting the full range of economic and environmental benefits provided by all-electric aircraft. A global fleet of all-electric aircraft serving all flights up to a distance of 400–600 nautical miles (741–1,111 km) would demand an equivalent of 0.6–1.7% of worldwide electricity consumption in 2015. Although lifecycle CO2 emissions of all-electric aircraft depend on the power generation mix, all direct combustion emissions and thus direct air pollutants and direct non-CO2 warming impacts would be eliminated.

## Main

Owing to their high energy content per unit weight and volume, easy handling, global availability and manageable costs, liquid hydrocarbons have been a key enabler of commercial flight over the past century. In 2015, the global aircraft fleet consumed 276 million tonnes of jet fuel—7% of global oil products1.

However, reliance on oil products comes at an environmental cost. Aircraft CO2 emissions, owing to combustion of jet fuel, comprise 2.7% of energy-use-related CO2 emissions1,2. It is also estimated that the non-CO2 warming impacts of aircraft are of the same magnitude as aircraft CO2 emissions, thus approximately doubling aviation’s contribution to climate change3,4,5. The single largest non-CO2 contributor to warming may be the formation of contrails and contrail cirrus clouds3. In addition, aviation combustion emissions that affect air quality, such as NOx, are set to rise substantially6. This may increase the estimated 16,000 premature mortalities per year attributable to aviation emissions globally7. There is also growing evidence that noise from aircraft results in adverse health impacts and premature mortality among affected populations8.

Various options exist for reducing CO2 emissions from aircraft. For example, fuel burn per revenue passenger kilometre (RPK) of the US narrow-body aircraft fleet could be reduced by around 2% per year at no cost through 20509, whereas reductions obtainable for wide-body, long-distance aircraft would probably be smaller. However, these rates will be outpaced by the anticipated global aviation demand growth of around 4.5% per year10,11. In contrast to fuel-efficiency improvements, low-carbon fuels (for example, biofuels) could partially decouple CO2 emissions from aviation growth, although these options face cost and scale limitations and do not much help with non-CO2 impacts12,13, except for a potential thinning of contrails (the effect of which has an uncertain sign)14,15. Similarly, liquid hydrogen16 and liquified natural gas17 could greatly reduce direct CO2 emissions, but these fuels’ higher hydrogen content would result in enhanced contrail and cirrus cloud formation.

Until recently, energy carriers that do not entail in-flight combustion have not been considered. This work focuses on all-electric aircraft that have the potential to eliminate both direct CO2 emissions and direct non-CO2 impacts, although the net impact will depend on the power generation mix and associated emissions. However, exploiting these unparalleled benefits requires substantial technological advances, especially inbattery performance and cost.

## Technology trajectories towards all-electric aircraft

Two broad technology trajectories appear to lead to all-electric aircraft. The first trajectory builds upon the incremental electrification of jet engines. This class of hybrid-electric aircraft includes designs without batteries (that is, turbo-electric aircraft), in which the electric propulsion system serves to increase propulsive efficiency and provide for some degree of boundary-layer ingestion (in which ingestion and re-energizing of the aircraft boundary layer improves efficiency)18,19. The extent of fuel burn reductions is then the net effect of the increased propulsive efficiency and the additional weight of the electrical components. Hybrid-electric aircraft with batteries are also being considered, where the batteries may provide for additional power or regeneration at limited specific operating conditions. Although hybrid-electric aircraft with batteries would entail direct combustion emissions for the majority of flights, they could provide for reduced or eliminated emissions during particularly sensitive parts of a flight—such as flying through ice supersaturated parts of the atmosphere (to reduce contrails) or during takeoff and landing (to reduce near-airport emissions). With sufficient advancements in battery technology, however, the ultimate design is an all-electric aircraft, which would have no direct combustion emissions and thus have the potential to remove aviation-specific non-CO2 impacts and reduce CO2 emissions depending on the source of the electricity. In contrast, the second technology trajectory builds upon scaling up all-electric air taxis. Ref. 20 reports 55 such air vehicle designs, 80% of which are already all-electric. Progress in battery technology, especially specific energy, would then enable the scaling up of all-electric designs to larger vehicles, first to regional jets and then to narrow-body aircraft.

## All-electric aircraft energy use

Aircraft energy use (E) per RPK during cruise flight can be described conveniently by the Breguet range equation21,22. Rearranged for energy intensity, equations (1) and (2) report energy use per RPK for jet engine aircraft (JEA) and all-electric aircraft (AEA), with PAX being the number of passengers transported, L/D the lift-to-drag ratio, ηtotal the total (tank-to-wake) efficiency of the jet engine or electric propulsion system, and W the weight of either fuel, the jet engine aircraft at the beginning (i) or the end (f) of the mission, or of the all-electric aircraft at any point during the mission.

$$E{{/{\rm{RPK}}_{{\rm{JEA}}}}} = 1{\mathrm{/}}\left( {\eta _{{{{\rm{total}}, {\rm{JEA}}}}}PAX\,(L{\mathrm{/}}D)} \right)W_{\mathrm{fuel}}{\mathrm{/}}{\mathrm{ln}}\left( {W_{\mathrm{i}}{\mathrm{/}}W_{\mathrm{f}}} \right)$$
(1)
$$E{\mathrm{/}}{{{\rm{RPK}}_{{\rm{AEA}}}}} = 1{\mathrm{/}}\left( {\eta _{{{{\rm{total}},\,{\rm{AEA}}}}}PAX\,(L{\mathrm{/}}D)} \right)W_{{\mathrm{AEA}}}$$
(2)

Assuming the same passenger count and lift-to-drag ratio between the jet engine and all-electric aircraft, equations 1 and 2 differ by only the propulsion system efficiencies and the weight factor. The latter is about 50–100% larger for all-electric aircraft as a consequence of the relatively low-specific energy batteries23,24. For narrow-body jet engine aircraft, Wi/Wf is typically 1.1–1.3; with Wfuel accounting for typically 10–30% of a narrow-body aircraft takeoff weight, the weight factor then roughly corresponds to the narrow-body aircraft takeoff weight. The resulting 50–100% higher energy intensity of all-electric aircraft is mitigated by the roughly two-fold tank-to-wake efficiency of electric propulsion systems compared to their jet engine counterparts22,25. We note that this calculation does not include the energy use associated with takeoff and climb, nor does it account for the upstream efficiency losses associated primarily with electricity generation. The latter strongly depend on the power generation technology and accounting practices for renewable energy.

A key enabler of electric flight and a critical determinant of energy intensity is the battery-pack specific energy. This variable enters the energy intensity of all-electric aircraft in equation (2) via the aircraft weight. If the on-board battery energy supply is kept constant, a higher specific energy leads to a lower all-electric aircraft weight and thus a lower aircraft energy use per RPK, which, in turn yields a longer range. In addition, a lighter aircraft would allow the downsizing of other components, such as landing gear, motor power, and so on, which yield additional energy intensity reductions and range gains.

Today’s best available Li-ion battery cells have a specific energy of around 250 Wh kg−1 (refs 26,27). Assuming a packing efficiency of 80%, which is at the lower end of projected future levels28 and below that of the recently developed Airbus E-Fan29, the pack-specific energy would result in roughly 200 Wh kg−1 and 1.7% of the jet fuel energy content. This battery would be capable of powering electric air taxis with 1–4 passengers over a distance of around 100 km20. However, short-range electric aircraft demand battery-pack specific energies of 750–2,000 Wh kg−1, which translates into 6–17% of the jet fuel energy content, depending on aircraft size and range22,23,24,30,31. Much of the required 4–10-fold increase in battery-pack specific energy could potentially be achieved with advanced Li–S technology, although Li-air chemistry may ultimately be required for the higher end of that range. Both of these battery technologies have low specific power, so an additional, high-power battery or another means of augmenting power may be required for takeoff and climb.

The historical long-term rate of increase in specific energy of the major battery chemistries has been around 3% per year, a doubling every 23 years32,33, although since 2000, specific energy has increased at a rate of 4% per year33. Although there is no ‘Moore’s Law’ equivalent for batteries—since significant advances require entirely new battery chemistries to be made practicable before incremental improvement can occur—this historical observation does suggest that the timescale for such progress to be made could be of the order of decades. On the basis of a continuation of the historical increase in specific energy, current levels of specific energy of 250 Wh kg−1 for advanced Li-ion battery cells, and a packing efficiency of 80%, a battery-pack specific energy of 800 Wh kg−1 could potentially be reached at around mid-century. This is consistent with the timescale of change in the aviation industry—for both the infrastructure and aircraft design lifecycles. For the purposes of this work we take the lower end of the above battery-pack specific energy range of 800 Wh kg−1 that is required for Airbus A320/Boeing 737-sized aircraft to be capable of up to 600 nautical miles (1,111 km) missions, depending on the specific layout and amount of batteries carried23.

In addition to battery pack specific energy, all-electric aircraft weight is determined by the power-to-weight ratio of the motors and the supporting infrastructure, consisting mainly of cables and power electronics. Whereas regional jets with about 50 seats are likely to require significantly improved mainstream technology, narrow-body aircraft with 100 seats and above may depend upon lightweight high-temperature superconducting electric motors due to the intrinsically high weight of conventional electric motors and the difficulty in providing cooling34.

## Environmental impacts

All-electric aircraft would completely eliminate direct combustion emissions and thus remove associated direct CO2 and non-CO2 warming. The lifecycle CO2 intensity of all-electric aircraft is determined by the CO2 intensity of electricity used, losses associated with battery charging and electricity transmission/distribution, and the specific aircraft design and operation. Figure 1 depicts the warming intensity of a first-generation 180-seat, 150-passenger, all-electric aircraft over a mission of 400 nautical miles (741 km), which is projected to consume 180 Wh per RPK for a battery-pack specific energy of 800 Wh kg−1 (ref. 23). Using the 2015 average US grid CO2 intensity of 456 g of CO2 per kWh, this all-electric aircraft would generate 91 g of CO2 per RPK, if losses associated with electricity transmission/distribution and battery charging are included. This value is about 20% higher than the lifecycle CO2 intensity of its modern, jet engine counterparts (the ‘US’ dashed line in Fig. 1). However, if non-CO2 impacts are taken into account (by way of a factor of two3,4,5), the overall warming per RPK would be reduced by around 30%. The lifecycle CO2 intensity of all-electric aircraft would further decline with improved aircraft and battery technology and the potential transition of the grid towards renewable energy. Conversely, a longer range capability would result in a higher energy and thus CO2 intensity owing to the additional battery weight, as visible from equation (2). Note that CO2 emissions and non-CO2 impacts (such as cooling related to sulphur emissions from coal-fired power stations35) may still occur depending on the power generation mix.

If greenhouse gas emissions from battery production were taken into account, the warming intensity of all-electric aircraft shown in Fig. 1 would be slightly larger. Based on Li-ion battery studies, the increase in warming intensity would be 2–10 g of CO2 equivalent per RPK, depending upon the underlying assumptions36. However, employing end-of-economic-life high-performance batteries in stationary applications would significantly reduce these emission levels, as would the enhanced use of renewable electricity for battery production (see Methods).

In addition to removing direct non-CO2 impacts, all-electric aircraft would also eliminate direct air pollution. While indirect air pollution may occur depending on the power generation technologies employed, there is greater potential for emissions control of ground-based power generation compared to in-flight combustion.

Noise impacts of all-electric aircraft may be better or worse than conventional aircraft, depending on the design decisions made. Assuming a conventional tube and wing configuration, which does not take advantage of the design flexibility offered by electric propulsion, we estimated an overall improved noise performance of all-electric aircraft relying on a battery-pack specific energy of 800 Wh kg−1 compared to best-in-class current-generation short-haul aircraft. Considering both takeoff and landing operations, a 36% reduction in noise contour area is estimated, compared to the best-in-class aircraft (see Methods). This could allow extended airport operation hours, thus increasing aircraft utilization and airport capacity. During takeoff, aircraft noise is mainly determined by the thrust of the engines required. Owing to lower fan pressure ratios and the absence of combustion noise, we anticipate a more than 50% reduction in takeoff noise contour area. In contrast, during landing, the higher weight of all-electric aircraft means that the determinants of noise (principally lift, drag and landing speed) will result in a 15% larger noise contour area compared to those of best-in-class narrow-body aircraft. Higher battery-pack specific energy and future aircraft designs would provide the opportunity for reduced noise through novel aircraft design concepts and changes in operational procedures (such as highly distributed propulsion and steep approaches with propulsors in generating mode).

## All-electric aircraft economics

Compared to gas turbine engine aircraft, all-electric aircraft will have a different operating cost structure. Over its lifetime, an all-electric aircraft may require several generations of potentially expensive batteries, a factor that contributes to upfront investments (via the first set of batteries) and maintenance costs (via replacement batteries). In addition, its higher weight could increase the maintenance requirements of landing gear components. On the other hand, all-electric aircraft may also experience cost savings. For example, they would not require a fuel system or an additional gas turbine (auxiliary power unit) for generating electricity, engine starting, and so on. In addition, there may be potential for reductions in engine maintenance costs owing to the relative mechanical simplicity of electric motors, although this is uncertain for narrow-body aircraft due to the challenges of cooling high-temperature superconducting electric motors.

Taking into account only the differences in the largest-expenditure items between an all-electric aircraft and a jet engine aircraft in terms of capital costs (energy storage and propulsion system) and maintenance costs (landing gear and battery replacement), Fig. 2 depicts the potential range of breakeven electricity prices for a first-generation Airbus A320/Boeing 737-sized all-electric aircraft with a range of 400 nautical miles (741 km). Two sets of lines are shown with different levels of specific energy. Each set represents battery costs of US$100 kWh−1 and US$ 200 kWh−1, which reflect the target and 2017 level of Li-ion batteries37. At the 2015 US jet fuel price of US$1.8 per gallon, the breakeven electricity prices of only the all-electric aircraft with a battery-pack specific energy of 1,200 Wh kg−1 and battery costs of US$ 100 kWh−1 would fall within the 2015 US electricity price range of 6.9–12.7 cents per kWh, depending on the end-use sector38. In contrast, a first-generation all-electric aircraft with a battery-pack specific energy of 800 Wh kg−1 and a range of 400 nautical miles (741 km) would be economically viable only with battery costs of around US$100 kWh−1 or less and policies that result in significant reductions in electricity prices or increases in jet fuel prices. A carbon tax of US$ 100 per tonne of CO2, which translates into US$0.97 per gallon of jet fuel, would increase the break-even electricity price of the first-generation all-electric aircraft with a battery-pack specific energy of 800 Wh kg−1 to levels observed within the USA, if electricity is produced from renewable sources. This suggests that policies that support both low-carbon electricity and the introduction of a carbon tax may be central prerequisites for introducing all-electric aircraft if today’s market conditions prevail until all-electric aviation becomes technically feasible. However, as battery-pack specific energy increases and costs of renewable power decline, the cost-effectiveness of all-electric aircraft will improve and the need for supportive policies will diminish. The conditions required for cost parity with jet engine aircraft are also more relaxed for shorter missions but more stringent for longer missions, primarily owing to the extra battery weight and its impact on energy use. ## All-electric aircraft adoption potential Since advanced batteries with 5–10 times the pack specific energy of today’s Li-ion batteries would still contain only 8–17% of the energy content per unit weight of jet fuel (although this does not credit electrochemical storage with the higher energy conversion efficiency compared to gas turbines), all-electric aircraft would be constrained to short-range missions, at least initially. The limitation to short-distance operations of all-electric aircraft can be seen in Fig. 3, which depicts the global air transportation network in 2015 by distance band. The range of 600 nautical miles (1,111 km) (yellow trajectories) could be covered with all-electric aircraft relying on a battery-pack specific energy of 800 Wh kg−1 (ref. 23). Although a higher battery pack specific energy could lead to a more integrated flight network, there are technological limits. Operating beyond distances of 1,200 nautical miles (2,222 km) in a single-stage flight would require a battery- pack specific energy of at least 1,600 Wh kg−1 (ref. 23), which may remain a significant technological challenge for decades to come. From today’s perspective, the only way to further expand the all-electric aircraft network by operating over flight distances longer than 1,200 nautical miles would be via multistage flights with at least one intermediate stop. (This, of course, is contingent on achieving a battery-pack specific energy of 800 Wh kg−1). However, this strategy would probably lead to reduced travel demand owing to the associated increase in travel time. In addition, multistage flights may be limited by airport capacity and noise regulations. Thus, all-electric aircraft operations would probably remain limited to intra-continental traffic, in the absence of notable breakthroughs in battery technology or changes in consumer behaviour. Yet a short-range all-electric aircraft market can generate large-scale impacts. As shown in Fig. 4, an all-electric aircraft fleet with a useful range of 600 nautical miles (1,111 km) could substitute up to 15% of global RPK and up to half of global departures. In addition, it could substitute almost 15% of commercial aircraft fuel use and eliminate around 40% of global landing-and-takeoff-related NOx emissions. ## Impact on electricity generation Using the aircraft performance characteristics specified by ref. 23, we simulate the electricity demand of a hypothetical, all-electric aircraft fleet operating within the global 2015 flight network. This analysis, using the AIM2015 integrated model39, suggests that the energy demand by all-electric narrow-body aircraft operating at flight distances up to 400–600 nautical miles (741–1,111 km) would correspond to 112–344 TWh or 0.6–1.7% of 2015 global electricity consumption (see Methods). This percentage range reflects the global average of variable country-level data, culminating in slightly higher percentages within the industrialized world of 0.6–2.2% of total US electricity consumption and 1.3–3.7% for the UK. Assuming that the aircraft batteries for each first morning flight would be charged overnight, around 85% of recharging would occur over the course of a day. This would lead to extra power generation capacity requirements of 1.2–3.6 GW in the UK, 6.6–27 GW in the US and 31–118 GW globally for aircraft operating ranges of 400–600 nautical miles, assuming a 35% capacity factor as typical for renewable power systems. If world population and income levels follow the IPCC SSP2 ‘Middle-of-the-Road’ scenario, the resulting increase in air travel demand would imply that electricity requirements triple by 2050. ## Discussion All-electric aircraft could greatly reduce the environmental impact of aviation. Most importantly, they could eliminate direct CO2 and non-CO2 warming, in addition to removing all air pollutants. Moreover, all-electric aircraft have the potential to mitigate noise, especially during takeoff. The extent to which these benefits can be exploited from the global aircraft fleet will depend critically upon battery-pack specific energy. All-electric aircraft with battery packs of 800 Wh kg−1, enabling a range up to 600 nautical miles (1,111 km), could replace half of all aircraft departures, mitigate airport area NOx emissions by 40%, and reduce fuel use and direct CO2 emissions by 15%. Assuming strong progress in battery technology, aircraft with the two-fold endurance leading to a range of 1,200 nautical miles (2,222 km) could replace more than 80% of all aircraft departures, mitigate airport area NOx emissions by more than 60%, and reduce fuel use and direct CO2 emissions by around 40%. Although a realization of these prospects may fall well into the second half of this century, they seem too large to ignore. This analysis has shown that future, first-generation all-electric narrow-body aircraft may not be economically competitive to jet engine aircraft under today’s market conditions. To reach cost effectiveness with conventional aircraft, jet fuel prices would need to be in excess of US$ 100 per barrel. Conversely, if jet fuel prices remain at their 2015 level, end-use electricity prices would need to be below 4–6 cents per kWh, depending on battery costs, to ensure the economic competitiveness of all-electric aircraft. In addition, today’s CO2 intensity of electricity would lead typically to higher lifecycle CO2 emission levels compared to jet engine aircraft over the same mission, although the total warming impact may be reduced in most parts of the world.

Since timescales of mutually reinforcing technologies are measured in decades (that is, new aircraft design, battery development, electricity grid decarbonization and sufficiently strong decline in electricity prices from renewable power to increase cost effectiveness), research and development of critical all-electric aircraft components would need to start immediately in order to exploit the opportunities provided by an all-electric aircraft system in the decades to come. A potential path of manageable risk would be the development first of turbo-electric and then hybrid-electric technology, with the possible exception of all-electric regional aircraft, which can rely on less stringent requirements for battery-pack specific energy and power and may not require high-temperature superconducting technology. Although these transition technologies will not result in significant reductions of greenhouse gas emissions, they are critical enablers of and technology milestones towards an all-electric aircraft system.

## Methods

### Distribution of passenger kilometres and fuel burn by distance

Departures and fuel burn by distance is derived from flight schedules and passenger numbers from the Sabre Market Intelligence Database40, assuming great circle routing. To estimate fuel burn and landing- and- takeoff NOx emissions, we use the aircraft performance model from the Aviation Integrated Model AIM201541, the updated version of AIM42.

### Electric aircraft noise assessment

The impact of aircraft noise on communities near airports depends not only on noise levels from the aircraft but also on its operational characteristics. Quantification of this impact is usually mapped using noise contours, which, in turn, depend upon the noise–power–distance (NPD) curves of the aircraft. For existing aircraft, NPD curves are publicly available43 but for novel aircraft, they need to be estimated.

In the present study, the all-electric aircraft NPD curves have been derived from those of a baseline A320-232 aircraft using a method that accounts for both operational and technological variations of the aircraft from the baseline case44,45,46,47. The all-electric aircraft airframe and propulsor fans are assumed to behave acoustically in a similar manner to their conventional equivalents. Propulsor weight is estimated based on the method of ref. 48. Together with nacelle drag and an estimation of battery and cabling weight, the NPD curves for a number of distributed propulsion configurations and missions can be calculated49. In these calculations, airframe, fan and jet mixing noise are considered but motor noise has been ignored. Based on predictions by ref. 50, motor noise can be presumed to be negligible compared to fan and jet mixing noise contributions. From the NPDs, aircraft noise contours have been calculated using a method known as RANE (rapid airport noise estimation) that has been benchmarked against the US Federal Aviation Administration’s Integrated Noise Model (INM 7.0c)51. Typical results are illustrated in the Supplementary Information.

### Aircraft warming impact of battery production

The warming intensity in Fig. 1 excludes greenhouse gas emissions associated with battery production. According to ref. 36, the literature-based values range from 39–196 kg of CO2 equivalent per kWh, depending on the methodological approach, the method for imputing missing data, the carbon intensity of electricity and other factors. Given a battery capacity of 64,000 kWh23, the amount of greenhouse gasemissions due to battery production would result in 2,500–12,500 tonnes of CO2 equivalent. Assuming an average of 150 passengers per aircraft, a block speed of 800 km per hour, an average utilization of 10 h per day, and a battery lifetime of 3 years, battery-production-related greenhouse gasemissions would result in 2–10 g of CO2 equivalent per RPK or 2–10% of the warming intensity of an all-electric aircraft, provided the carbon intensity of electricity corresponds to the world average of around 500 g of CO2 per kWh. Note that this range represents an upper limit, because end-of-life high-performance batteries will probably experience a second life in stationary applications. In addition, a lower carbon intensity of electricity will result in further reductions52.

### Cost-effectiveness of all-electric aircraft

The key difference between the A320neo reference aircraft and the derived all-electric aircraft is the energy storage and propulsion system. Our all-electric aircraft capital cost estimate (referring only to recurring costs) is based upon the reference aircraft average retail price of US$46 million, which includes the price of two gas turbine engines at US$ 5.5 million, after a whole-aircraft discount of 57%53. Not taking into account the credit for the obsolete fuel system and auxiliary power unit, we add the cost of batteries at US$100 kWh−1 and US$ 200 kWh−1. These numbers reflect the projected future and current costs of Li-ion batteries. Given the projected battery capacity of 28 MWh (21 MWh) for first-generation all-electric aircraft with a battery specific energy of 800 Wh kg−1 (1,200 Wh kg−1), the total cost of batteries results in US$2.8 million (US$ 2.1 million) and US$5.6 million (US$ 4.2 million), respectively. The replacement costs of the batteries after their useful life of 5,000 cycles is then accounted for in the maintenance costs.

Our estimate of the cost range of the electric propulsion system is based upon two limiting cases. The lower-end estimate assumes electric propulsor costs without high-temperature superconducting motors. It is based upon electric propulsion system costs of US$8 kW−1, which corresponds to the 2022 US Department of Energy target for electric motors plus inverters for automobile applications54. Based upon a maximum power requirement of 12.5 MW for each of the four propulsion units during take-off for the aircraft with a battery specific energy of 800 Wh kg−1, the cost of one electric motor plus inverter amounts to US$ 100,000. These costs exclude the fan, which costs about 15% of the cost of a gas turbine engine55 or US$410,000. Hence, the cost of one propulsion system totals US$ 510,000, which translates into around US$2 million for the four units. For the aircraft with a battery specific energy of 1,200 Wh kg−1, the lower maximum power requirement leads to propulsion system costs that are only around half as high. The higher-end cost case accounts for a high-temperature super-conducting electric propulsion system. Perhaps conservatively, it corresponds to the cost of two jet engines, or US$ 5.5 million. Subtracting the costs of four fans would lead to motor plus power electronics costs of US$3.9 million. In light of the maximum aircraft power requirement of 50 MW, these costs would then translate into US$ 78 kW−1. The latter are within the range of the high-temperature super-conducting motor costs cited by Hoelzen et al.56. However, with progress in high-temperature super-conducting wire technology, especially, and increases in production scale, high-temperature super-conducting motor costs are expected to decline drastically57,58.

Although estimating the cost of all-electric aircraft propulsors in decades is highly uncertain, these numbers may be indicative of the order-of-magnitude cost. The results imply (see Fig. 2) that the uncertainty in the electric propulsion system costs is unimportant relative to the uncertainty in battery cost or overall aircraft performance, even if propulsion system costs are a factor of two or more greater than our higher case.

In addition to capital costs, the cost-effectiveness analysis takes into account maintenance costs and energy costs. Expenditures for crew and airport/airspace were assumed to be identical for the two competing aircraft types. Maintenance costs of the A320neo were computed with data from Aircraft Commerce on the basis of the A320–20059 and resulted in US$960 per flight hour. This number compares well with US Form 41 data60. In contrast, the maintenance costs of the all-electric aircraft range from US$ 1,170 per flight hour for batteries with a specific energy of 1,200 Wh kg−1 and costs of US$100 kWh−1 to US$ 1,500 per flight hour for batteries with a specific energy of 800 Wh kg−1 and costs of US$200 kWh−1. The higher maintenance costs of all-electric aircraft can be attributed mainly to battery maintenance. Using a discount rate of 7%, maintenance costs for the aircraft with 800 Wh kg−1 batteries result in US$ 265 and US$530 per flight hour for the US$ 100 and US$200 kWh−1 battery costs, respectively. Owing to the required smaller battery capacity, maintenance costs for the aircraft with 1,200 Wh kg−1 batteries result in US$ 205 and US\$ 410 per flight hour.

### Impact on electricity generation

The hypothetical year-2015 and year-2050 electricity demand projections are obtained using the global aviation systems model AIM39. For 2015, we take the baseline global network as represented in AIM, which is obtained from a global scheduled passenger and flight database for 201540. For each flight segment up to an assumed range of 400–600 nautical miles, we calculate the electricity demand under the assumption that all passengers are carried on all-electric narrow-body aircraft of the type and size specified in ref. 23. We use a performance model fitted to the electricity demand of an all-electric aircraft with a battery specific energy of 800 Wh kg−1, a 400- or 600- nautical-mile design range, and different passenger load factors, assuming passenger load factors similar to those historically flown on each segment. This procedure provides an estimate of the electricity demand per airport.

We use the central SSP2 reference case from ref. 39 to project demand by flight segment in 2050. The mid-range trends for future socioeconomic characteristics underlying this projection results in 2017–2037 demand growth rates consistent to those from the most recent Airbus and Boeing forecasts10,11. Total RPK in 2050 is around 3.7 times the value in 2015. The same procedure as for 2015 is used to estimate electricity demand; the increase in electricity demand is lower compared to total RPK because of a shift towards longer-haul flights, which cannot be served by all-electric aircraft.

## Data availability

The data that support the plots within this paper and other findings of this study are available from the corresponding author upon reasonable request.

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

## References

1. 1.

International Energy Agency World Energy Statistics 2017 (IEA, 2017)

2. 2.

International Energy Agency CO 2 Emissions from Fuel Combustion 2017 (IEA, 2017)

3. 3.

Lee, D. S. et al. Aviation and global climate change in the 21st century. Atmos. Environ. 43, 3520–3537 (2009).

4. 4.

Dorbian, C. S., Wolfe, P. J. & Waitz, I. A. Estimating the climate and air quality benefits of aviation fuel and emissions reductions. Atmos. Environ. 45, 2750–2759 (2011).

5. 5.

Brasseur, G. P. et al. Impact of aviation on climate: FAA’s Aviation Climate Change Research Initiative (ACCRI) phase II. Bull. Am. Meteorol. Soc. 97, 561–583 (2016).

6. 6.

European Aviation Safety Agency European Aviation Environmental Report 2016 (EASA, 2016).

7. 7.

Yim, S. H. L. et al. Global, regional and local health impacts of civil aviation emissions. Environ. Res. Lett. 10, 034001 (2015).

8. 8.

Wolfe, P. J., Kramer, J. L. & Barrett, S. R. H. Current and future noise impacts of the UK hub airport. J. Air Transp. Manag. 58, 91–99 (2017).

9. 9.

Schäfer, A. W., Evans, A. D., Reynolds, T. G. & Dray, L. Costs of mitigating CO2 emissions from passenger aircraft. Nature Clim. Change 6, 412–417 (2016).

10. 10.

Airbus Global Market Forecast 2017–2037 (Airbus Commercial Aircraft, Toulouse, 2017).

11. 11.

Boeing Current Market Outlook 2017–2036 (Boeing Commercial Airplanes, Seattle, 2017).

12. 12.

Bann, S. J. et al. The costs of production of alternative jet fuel: a harmonized stochastic assessment. Bioresour. Technol. 227, 179–187 (2017).

13. 13.

Stratton, R. W., Wolfe, P. J. & Hileman, J. I. Impact of aviation non-CO2 combustion effects on the environmental feasibility of alternative jet fuels. Environ. Sci. Technol. 45, 10736–10743 (2011).

14. 14.

Moore, R. H. et al. Biofuel blending reduces particle emissions from aircraft engines at cruise conditions. Nature 543, 411–415 (2017).

15. 15.

Caiazzo, F., Agarwal, A., Speth, R. L. & Barrett, S. R. H. Impact of biofuels on contrail warming. Environ. Res. Lett. 12, 114013 (2017).

16. 16.

Brewer, G. D. Hydrogen Aircraft Technology (CRC Press, 1990).

17. 17.

Withers, M. R. et al. Economic and environmental assessment of liquified natural gas as a supplemental aircraft fuel. Prog. Aerosp. Sci. 66, 17–36 (2014).

18. 18.

Drela, M. Power balance in aerodynamic flows. AIAA J. 47, 1761–1771 (2009).

19. 19.

Hall, D. K. et al. Boundary layer ingestion propulsion benefit for transport aircraft. J. Propul. Power 33, 1118–1129 (2017).

20. 20.

Datta, A. Commercial Intra-City On-Demand Electric-VTOL Status of Technology (University Maryland, College Park, 2018).

21. 21.

Lee, J. J., Lukachko, S. P., Waitz, I. A. & Schäfer, A. Historical and future trends in aircraft performance, cost, and emissions. Annu. Rev. Ener. Environ. 26, 167–200 (2001).

22. 22.

Hepperle, M. Electric Flight Potential and Limitations (DLR Electronic Library, German Aerospace Centre, Cologne, 2012); https://elib.dlr.de/78726/

23. 23.

Gnadt, A. R. Technical and Environmental Assessment of All-Electric 180-Passenger Commercial Aircraft. SM thesis, Massachusetts Inst. Technol. (2018).

24. 24.

Hornung, M., Isikveren, A. T., Cole, M. & Sizmann, A. Ce-Liner—case study for emobility in air transportation. In 13th AIAA Aviation Technology, Integration, and Operations Conference (AIAA, 2013); https://doi.org/10.2514/6.2013-4302

25. 25.

Delhaye, J. L. & Rostek, P. Electrical Technologies for the Aviation of the Future (Europe-Japan Symposium, Tokyo, 2015).

26. 26.

Panasonic Lithium Ion NCR18650B (Panasonic, Newark, 2012).

27. 27.

Muenzel, V. et al. A comparative testing study of commercial 18650-format lithium-ion battery cells. J. Electrochem. Soc. 162, A1592–A1600 (2015).

28. 28.

Dever, T. P. et al. Assessment of Technologies for Noncryogenic Hybrid Electric Propulsion (NASA Glenn Research Center, Cleveland, 2015).

29. 29.

Airbus E-FAN, The New Way to Fly (Airbus, Munich, 2015).

30. 30.

Bradley, M. K. & Droney, C. K. Subsonic Ultra Green Aircraft Research: Phase I Final Report NASA/CR-2011-216847 (Boeing Research & Technology, Huntington Beach, 2011).

31. 31.

Bradley, M. K. & Droney, C. K. Subsonic Ultra Green Aircraft Research Phase II: N+4 Advanced Concept Development NASA/CR-2012-217556 (Boeing Research & Technology, Huntington Beach, 2012).

32. 32.

Koh, H. & Magee, C. L. A functional approach for studying technological progress: extension to energy technology. Technol. Forecast. Soc. 75, 735–758 (2008).

33. 33.

Crabtree, G., Kócs, E. & Trahey, L. The energy-storage frontier: lithium-ion batteries and beyond. MRS Bull. 40, 1067–1076 (2015).

34. 34.

Sinsay, J. D. et al. Air Vehicle Design and Technology Considerations for an Electric VTOL Metro-regional Public Transportation System (12th AIAA Aviation Technology, Integration, and Operations Conference, 2012); https://doi.org/10.2514/6.2012-5404.

35. 35.

Unger, N. et al. Attribution of climate forcing to economic sectors. Proc. Natl Acad. Sci. USA 107, 3382–3387 (2010).

36. 36.

Kim, H. C. et al. Cradle-to-gate emissions from a commercial electric vehicle Li-ion battery: a comparative analysis. Environ. Sci. Technol. 50, 7715–7722 (2016).

37. 37.

Chediak, M. The Latest Bull Case for Electric Cars: the Cheapest Batteries Ever (Bloomberg, 2017); https://www.bloomberg.com/news/articles/2017-12-05/latest-bull-case-for-electric-cars-the-cheapest-batteries-ever.

38. 38.

Energy Information Administration United States Electricity Profile 2015. Table 1 (EIA, 2017); https://www.eia.gov/electricity/state/unitedstates/.

39. 39.

Dray, L. M. AIM2015 Documentation (Air Transportation Systems Laboratory, UCL, 2018); http://www.atslab.org/wp-content/uploads/2018/08/AIM-2015-Documentation-v9-30082018.pdf.

40. 40.

Sabre Market Intelligence Schedule and Passenger Database (Sabre, 2017); https://www.sabreairlinesolutions.com/home/software_solutions/product/market_competitive_intelligence/.

41. 41.

Dray, L. M., Schäfer, A. W. & Al Zayat, K. The global potential for CO2 emissions reduction from jet engine passenger aircraft Transp. Res. Rec. 1-12, 18-04002 (2018).

42. 42.

Reynolds, T. G. et al. Modelling environmental and economic impacts of aviation: introducing the aviation integrated modelling project. In Proc. 7th AIAA/ATIO Conf. (AIAA, 2007); https://doi.org/10.2514/6.2007-7751.

43. 43.

Eurocontrol Experimental Centre Aircraft Noise and Performance (ANP) Database v2.1 (EEC, 2016); http://www.aircraftnoisemodel.org.

44. 44.

Synodinos, P. A., Self, R. H., Flindell, I. H. & Torija, A. J. Estimating variation in community noise due to variation in aircraft operations. In Proc. 10th Eur. Congress Exposition Noise Control Eng. (European Acoustics Association, 2015).

45. 45.

Synodinos, P. A., Self, R. H. & Torija, A. J. A new method for estimating community noise changes due to aircraft technology variations. In Proc. 23rd Int. Congress Sound Vibration (International Institute of Acoustics and Vibration, 2016).

46. 46.

Synodinos, P. A., Self, R. H. & Torija, A. J. A framework for predicting noise-power-distance curves for novel aircraft designs. J. Aircraft 55, 781–791 (2018).

47. 47.

Synodinos, P. A., Self, R. H. & Torija, A. J. Noise assessment of aircraft with distributed electric propulsion using a new noise estimation framework. In Proc. 24th Int. Congress Sound Vibration (International Institute of Acoustics and Vibration, 2017).

48. 48.

Liu, C. X., Teng, J. & Ihiabe, D. Method to explore the design space of a turbo-electric distributed propulsion system. J. Aerospace Eng. 29, 04016027 (2016).

49. 49.

Synodinos, P. A. A New Framework For Estimating The Noise Impact Of Novel Aircraft. PhD thesis, Univ. Southampton (2017).

50. 50.

Huff, D. L., Henderson, B. S. & Envia, E. A First Look at Electric Motor Noise for Future Propulsion Systems. Oral/visual presentation GRC-E-DAA-TN31506 (NASA Glenn Research Center, Cleveland, 2016).

51. 51.

Torija, A. J., Self, R. H. & Flindell, I. H. A model for the rapid assessment of the impact of aviation noise near airports. J. Acoust. Soc. Am. 141, 981–995 (2017).

52. 52.

Ager-Wick Ellingsen, L., Singh, B. & Hammer Strømman, A. The size and range effect: lifecycle greenhouse gas emissions of electric vehicles. Environ. Res. Lett. 11, 1–8 (2016).

53. 53.

FlightGlobal Ascend Fleets (FlightGlobal, 2017); https://www.flightglobal.com/products/.

54. 54.

Howell, D., Cunningham, B., Duong, T. & Faguy, P. Overview of the DOE VTO Advanced Battery R&D Program (US Department of Energy, Vehicle Technologies Office, Washington DC, 2016).

55. 55.

Fancher, D. Defining Technologies for the next Millennium (GE Aircraft Engines, Evendale, 1999).

56. 56.

Hoelzen, J. et al. Conceptual design of operation strategies for hybrid electric aircraft. Energies 11, 217 (2018).

57. 57.

Schiferl, R., Flory, A., Livoti, W. C. & Umans, S. D. High-temperature superconducting synchronous motors: economic issues for industrial applications. IEEE T. Ind. Appl. 44, 1376–1384 (2008).

58. 58.

Haran, K. S. et al. High power density superconducting rotating machines—development status and technology roadmap. Supercond. Sci. Tech. 30, 123002 (2017).

59. 59.

Aircraft Commerce Maintenance and Engineering (Aircraft Commerce, 2017); http://www.aircraft-commerce.com/default.asp.

60. 60.

US Bureau of Transportation Statistics Air Carrier Financial: Schedule P-5.2 (2017); https://www.transtats.bts.gov/DL_SelectFields.asp.

61. 61.

Piano-X (Lissys, 2010); http://www.lissys.demon.co.uk/PianoX.html.

62. 62.

Schäfer, A. W., Heywood, J. B., Jacoby, H. D. & Waitz, I. A. Transportation in a Climate-Constrained World (MIT Press, 2009).

## Acknowledgements

Research underlying this work was made possible by the UK Engineering and Physical Sciences Research Council (EP/P511262/1) and the National Science Foundation Graduate Research Fellowship (grant number 1122374). We thank M. Schofield, J. Sabnis and R. Gardner for discussions and K. Al Zayat for early contributions to this work.

## Author information

A.W.S. led the overall study, the analysis of the results and the preparation of the manuscript. S.R.H.B. led the all-electric aircraft performance study and contributed to the analysis of the results and to the preparation of the manuscript. R.S. led the all-electric aircraft noise study and contributed to the preparation of the manuscript. A.R.G. carried out the all-electric aircraft performance simulations and contributed to the preparation of the manuscript. L.M.D. carried out the analysis of the results. K.D. and A.O’S. contributed to the analysis of the results. A.P.S. and A.J.T. contributed to the all-electric aircraft noise study.

### Competing interests

The authors declare no competing interests.

Correspondence to Andreas W. Schäfer.

## Supplementary information

1. ### Supplementary Information

Supplementary Figure 1.

## Rights and permissions

### DOI

https://doi.org/10.1038/s41560-018-0294-x