1 Introduction

Global mobility is increasingly confronted with the societal need to significantly reduce air travel’s environmental impact. In alignment with the goal of decreasing emissions, the electrification of aircraft is currently intensely being researched. As a key component in the electric system, the constantly evolving field of power electronics contributes substantially to this development. High demands are imposed on power electronics. This paper reviews power electronics design challenges and opportunities in all- and hybrid-electric aircraft. Starting with a basic introduction to the electric propulsion architectures, their dependence on power electronics is highlighted in Sect. 2. Subsequently, key performance measures for power electronics in aircraft applications are identified and discussed in Sect. 3. Electric aircraft-specific requirements and challenges for power electronics are outlined in Sect. 4, and an exploration of the power electronics design space to address those challenges is conducted in Sect. 5. Cryogenic power electronics, functionally integrated drive systems, and multi-level inverters are the identified emerging technologies. Figure 1 serves as a visual representation of how the design space acts as a means to overcome the application-specific challenges along the path toward enhanced performance.

Fig. 1
figure 1

Optimization objectives, application-specific challenges, and degrees of design freedom

Full size image

2 Power electronics in electric aircraft propulsion

Toward electrification, in the first step, some systems previously powered by mechanical, hydraulic, or pneumatic actuators are already being replaced by electrical systems, resulting in the so-called more-electric aircraft (MEA). However, only replacing the large loads on board an aircraft, namely the propulsion system, will lead to the necessary reduction of the environmental impact. Several different propulsion architectures, ranging from hybrid- to all-electric aircraft, are under discussion for the design of such electric propulsion aircraft. Figure 2 gives examples. Power electronics are integral to the energy networks architecture [1, 2]. In the following subsections, to highlight this dependency, a selection of well-known [3, 4] general electric propulsion architectures is discussed, starting with turbo-electric architectures, followed by the various types of hybrid-electric architectures, and concluding with all-electric architectures. In addition to the different architectures, the relevance of the supply grid voltage and the centralization level is discussed.

2.1 Turbo-electric architectures

In the so-called partial turbo-electric architectures, a generator is driven by a gas turbine. From this generator, the electrical supply is fed to grid. This grid, usually in the form of a high-voltage direct current (HVDC) bus, supplies the electrical loads, specifically the electrical machines. Power electronics play an important role in this energy conversion process. Rectifiers (AC/DC converters, see Fig. 2) supply the electrical grid from the generator, and inverters (DC/AC converters) feed the electrical machine from this grid. Instead of the partial turbo-electric architecture’s turbofan, a turboshaft can be found in a full turbo-electric topology. This turboshaft, however, is not involved in the propulsion. In Fig. 2a, a simplified block diagram of a partial turbo-electric architecture is depicted.

Fig. 2
figure 2

Electrical propulsion architectures. Only propulsion-related elements are shown; additional components, for instance, to supply an AC bus or a low-voltage DC bus, may be present

Full size image

2.2 Hybrid-electric architectures

A number of different hybrid-electric architectures are under discussion [5]. In parallel hybrid-electric architectures, the propulsive element is a turbine-driven turbofan, which is assisted by an electric machine, either supplied from a battery or a fuel cell. In serial hybrid-electric architectures, no turbo-propulsion unit exists. A gas turbine burning fuel provides assistance to the electric propulsion unit, which is mainly supplied by battery- or fuel cell-based energy storage. The shaft drives no propeller. In Fig. 2b, a series–parallel combined hybrid-electric architecture is depicted. The gas turbine, which feeds the electric drive train, is also connected to propellers. In contrast to the parallel hybrid-electric architecture, however, not the same propulsive element is driven by the turbine and electric machine. Instead, different propulsive fans are present. In hybrid-electric architectures, power electronic rectifiers are required to transfer power from a turbofan or turboshaft to the high-voltage DC grid. To supply this grid from the energy storage, DC/DC converters are required. The electric motor is fed from the HVDC bus through an inverter.

2.3 All-electric architectures

All-electric aircraft are solely powered by electric energy storage systems. Battery- and fuel-cell-supplied systems, or combinations of both, are feasible. Depending on the type of energy storage, a distinction must be made in terms of the feasible solutions for power electronics. To supply the HVDC energy network from the energy sources and to, if present, interface between different voltage levels, DC/DC converters are required. Motor inverters are required for the speed-variable operation of electric machines. An example of an all-electric architecture that uses batteries as well as hydrogen as energy storage and an HVDC bus is shown in Fig. 2c.

2.4 HVDC voltage level

Generally, higher DC voltages are preferable for transmitting higher power, as they result in a lower current. This reduction in current leads to a decrease in conductor diameter and an overall reduction in the weight of the wiring harness and system. To decouple the supply grid voltage from the batteries, state-of-charge DC/DC converters are utilized. What is considered a high voltage in this context depends strongly on the overall system, requiring a holistic optimization that takes into account all other components of the system. In [6, 7], inverter topologies for an all-electric aircraft are evaluated for a DC-link voltage range from 1 to 4 kV, and it was estimated that values around 3 kV were advantageous in terms of efficiency and power density. The electrical architecture of the electrically powered short-range aircraft proposed in [1] also utilizes a 3 kV voltage level to connect the large loads.

Conventionally, the aim is to maintain the DC voltage in as narrow a range as possible to achieve optimal power density [8]. However, due to various effects, which are discussed in Sect. 4, such as the increasing semiconductor failure rate with higher altitudes and, therefore, higher blocking voltage requirements, variable HVDC voltage levels are under discussion. Mission- and therefore flight altitude-dependent DC voltages with higher voltages during take-off with low cosmic radiation and lower voltages in the cruise phase with increased cosmic radiation and low air pressure would introduce new possibilities [9].

2.5 On-board network topology

The architectures shown in Fig. 2 generally depict centralized electric propulsion architectures. A decisive advantage of introducing electric propulsion systems on board an electric aircraft is the possibility of implementing a larger number of propulsion elements than previously possible in the conventional aircraft. However, the use of distributed drives also necessitates new distributed wiring harness structures. In such systems, the arrangement of energy storage units, the voltage level, and the control of the drives influence system efficiency and weight.

In a central onboard network topology, all energy sources and sinks are connected to a common HVDC onboard network. This enables a flexible design of the energy flows and, thus, a high level of redundancy in the energy supply. However, a great amount of cabling and redundancy of the onboard network itself is required. In contrast, in a decentralized onboard network topology, multiple energy storage units are positioned freely on board and near the largest loads, generally the drives. This way, the cabling effort is reduced. Redundancy must then be established by other means, for instance, through additional propulsive elements.

3 Key performance measures

As in all engineering fields, the performance of power electronic converters can be described by a set of performance measures or optimization goals. For power electronics in aviation, a specific weighting of these measures applies, and on route to improvement, the special requirements in this field of application have to be considered. As shown in the relative comparison in Fig. 3, the performance indicators efficiency, power density, and reliability are of paramount importance for aircraft applications. Typically, these performance measures are coupled, and certain trade-offs are required. The optimization goals will be discussed in the following subsections.

Fig. 3
figure 3

Application-dependent performance profiles for power electronics. Relative plotting with the importance of the measure increasing with the assigned number. Based on assessments by domain experts (c.f.  [10])

Full size image

3.1 Efficiency

The first challenge that power electronics face is achieving high efficiency. This is imperative due to the limited amount of energy stored on board the aircraft, which is restricted by its weight. This aspect is even more crucial than for ground-based electric vehicles, such as electric cars. Its improvement is under ongoing research and development. Relevant topics are covered in Sects. 5.1–5.4, including the application of wide band-gap (WBG) devices, advanced converter topologies, functional integration, and cryogenic power electronics.

Current inverter efficiencies are reported up to 98–99% [11,12,13]. However, since converters operate in a combined drive system, considering this value alone is not inherently reasonable. An increased switching frequency may negatively influence the inverter’s efficiency but can, for example, reduce losses in the electric motor. For measuring the system’s efficiency, though, no standards exist. Thus, given values may differ, including or excluding loss mechanisms, such as filter- or DC-link losses.

Efficiency impacts the total available energy on board and influences weight indirectly through cooling efforts. In contrast to conventional jet propulsion, which is mainly self-cooled, power electronics, as well as electric machines, need an additional cooling system. Since this also includes a re-cooling system, the total weight increases. High efficiency, in turn, results, due to the present thermal coupling, in a low on-state resistance of the power electronic device and low motor coil resistance.

Power electronic systems, consisting of multiple building blocks such as DC/DC converters and inverters, exhibit high overall efficiency. For DC/DC converters, it ranges from 94 to over 99% [14,15,16,17,18,19], and as mentioned, up to 99% for inverters. To reduce the energy consumption of the aircraft, other subsystems are more challenging, since they face higher losses. These include, but are not limited to, the propeller or fan drive system, HVDC bus, and fuel cells. In the case of a hybrid system, the combustion system strongly decreases overall efficiency.

3.2 Power density

While it is agreed that the energy storage outweighs the power converters’ weight and volume, still both low weight and volume of the power converter are desired. This is especially relevant in the face of an increase in propulsive electric power demand.

With regard to power density, it has to be distinguished between gravimetric (weight-sensitive) and volumetric (space-sensitive) power density. In aviation, gravimetric power density prevails over volumetrical power density and gravimetric power density for all converters is of great importance. Published power densities of converters and inverters spread widely from less than 10 up to 86 kVA/kg [11,12,13, 20]. These figures should, however, be treated with great caution. The wide range of power density values results from the absence of a generally agreed-upon standard for specifying the power densities of power electronic converter systems. While some references include the heat sink or heat exchanger, additional passive filtering components, and even the housing in this specification, others might only consider the power modules, DC-link, and associated busbar weight.

Increasing power density is often achieved by making the converter more compact, optimizing component usage, or improving the cooling system. However, this can lead to power electronic components operating at higher temperatures, necessitating some compromises in efficiency.

This sets the stage for a fundamental conflict between power density and efficiency. When efficiency is lower, resulting in increased power loss, there is a need for additional energy storage capacity, which in turn adds weight. Thus, a trade-off must be carefully managed.

It is worth noting that this balance is context-dependent and hinges on the specifics of the energy storage technology in use. With ongoing advancements in battery systems and hydrogen storage systems, this relationship may evolve over time [21].

3.3 Reliability

In engineering systems, such as aircraft, the various components within them experience a gradual deterioration process that eventually culminates in their failure. To mitigate the risk of unexpected failures and ensure safe and reliable operation, aircraft manufacturers establish maintenance intervals for the equipment incorporated into the aircraft’s design. This practice requires accurate lifetime estimation of these components before the aircraft is built.

The primary objective behind estimating component lifetimes is to guarantee dependable performance throughout the entire life cycle of the equipment. However, this goal must be achieved while keeping the size and cost of the components at a reasonable level. Excessive oversizing of components can lead to unnecessary expenses and increased weight.

Hence, the challenge lies in striking a delicate balance between component lifetime and cost-effectiveness during the design phase. By accurately predicting the lifetimes of the components used, engineers can ensure that maintenance intervals are appropriately planned, minimizing the risk of failure and maximizing the operational efficiency of the aircraft.

Over the past years, the industrial and automotive sectors have accumulated extensive experience with power electronics. This growing knowledge has highlighted the important roles of capacitors and semiconductors in power electronics. However, it has also revealed the vulnerability of these components to failures [22]. As a result, there have been extensive studies [23, 24] and the establishment of standards to guide the development of power electronics using lifetime-based approaches and accelerated aging techniques. The field of capacitor lifetime estimation [25] and lifetime testing of dielectric substrates for printed circuit boards [26], especially for fast-switching WBG semiconductors [27,28,29,30], is strongly being researched.

Numerous failure mechanisms impacting power systems are currently under investigation. Among these, one significant area of study involves the phenomenon of thermal cycling in semiconductor devices. The degradation mechanism is attributed to thermo-mechanical stress at the interface of materials with different coefficients of thermal expansion, exacerbated by vertical and lateral temperature gradients. This generates mechanical stress during temperature fluctuations, ultimately causing material and interconnection fatigue over time [31, 32]. While temperature cycling is the primary trigger for the above-mentioned degradation mechanisms, the average temperature and heating time also play significant roles in influencing the degradation process [33, 34].

Fig. 4
figure 4

Testbed for the accelerated aging of power semiconductors as used in [28, 30]

Full size image

The testbed illustrated in Fig. 4 serves as a crucial tool for gathering data essential to the lifetime modeling of semiconductors. To accomplish this, comprehensive knowledge of the aircraft’s mission profile is necessary. Loss calculations for the power electronic circuit employed and detailed information on the system’s thermal characteristics act as input variables for the modeling process. The junction’s temperature swings and average junction temperature can be estimated by utilizing these inputs. This approach is extensively discussed in [35].

In the testbed shown in Fig. 4, critical loads are emulated to subject the semiconductors to accelerated aging. This accelerated aging process facilitates the collection of valuable data that enables the fitting of lifetime models to them. Subsequently, based on these fitted data, the lifetime of the semiconductors can be estimated for the target application. A general observation is that higher temperature swings and elevated average junction temperatures reduce the lifetime of semiconductors.

In addition to thermal considerations, another area of study that impacts component lifetime is the phenomenon of cosmic radiation. This unique challenge, specific to aircraft and other high-altitude applications, is thoroughly examined in this paper’s Sect. 4.1.

Similar to the previously mentioned relationship between power density and efficiency in Sect. 3.2, the lifetime expectancy is also interconnected with power density: higher power density and, thus, lower efficiency lead to higher temperatures in the semiconductor, having a negative effect on the system’s lifetime. This again highlights the outstanding importance of the power electronics’ high efficiency.

4 Aircraft-specific challenges

As previously mentioned, power electronics design in the application area of electric aviation is subject to a number of distinct challenges. This section discusses challenges that arise from the environmental influences to which aircraft power electronics may be subjected. This especially includes the consideration of the influence of flight altitude and cosmic radiation. These specific challenges not only apply to aircraft power electronics but are particularly relevant to them.

4.1 Cosmic radiation

A major issue for the operation of power electronics is the rising concentration of cosmic radiation at high altitudes. Cosmic radiation also impacts power electronics on the ground, with the risk of failure increasing at higher altitudes. This is because high-energy particles have a lesser chance of colliding with atmospheric particles, making them more likely to enter semiconductor dies at high-energy levels. The effects of cosmic radiation were first described in [36], highlighting voltage-dependent device failures. Further investigations are presented in [37, 38], and the impact on aviation is discussed in [1, 39]. Different effects occur related to cosmic radiation. While some are of minor importance to power electronics, the so-called Single Event Effects (SEE) can cause destructive failures. Cosmic radiation, primarily composed of protons but also other particles, generates a particle shower in the atmosphere through spallation with atmospheric molecules. Some of these particles include highly accelerated neutrons. These cannot effectively be shielded, since they are not electrically charged. In the case of a collision with a silicon atomic core inside the semiconductor material, it creates an unstable isotope. This isotope quickly decays, generating a local plasma and a resulting high electric field strength gradient. Additional charge carriers can then be detached by impact ionization, and a chain reaction may lead to a short circuit of the device. This so-called Single Event Burnout (SEB) occurs suddenly and can hardly be predicted or prevented.

The only effective countermeasure today is derating the voltage applied to the semiconductor according to its rated breakdown voltage. This derating depends on the altitude and, therefore, the concentration of cosmic radiation. It does not cancel out the possibility of an SEE but decreases its probability to an acceptable level. Experiments on this were conducted in [37], resulting in a failure in time (FIT) rate (failures per \(\mathrm {10^9 \, h}\)) for a reduced voltage applied to a power semiconductor and a given altitude. Figure 5 shows this relationship for different FIT rates of power semiconductors. Due to its altitude relation, SEE caused by cosmic radiation are a flight-specific issue. However, its effects play a major role in reliability and system lifetime, which is also addressed in Sect. 3.3.

Fig. 5
figure 5

Cosmic radiation and its voltage dependency, based on [1, 37]

Full size image

4.2 Temperature

The distinctive operational profile of aviation missions introduces a diverse range of thermal conditions to which power electronic systems are subjected. These conditions span from the elevated temperatures experienced during parking in high-temperature regions to the severe cold encountered during cruising at high altitudes. Also, other scenarios are to be examined. For example, fast transitions between contrasting thermal states—such as the rapid shift from cold, dark parking to the subsequent take-off phase— exacerbate stress on power electronic components. This thermal cycling contributes to an interplay of factors that can significantly impact the system’s operational lifespan, as discussed in Sect. 3.3.

Drawing an analogy to the operational characteristics of hybrid-electric automobiles, the utilization of power electronics in hybrid-electric aviation scenarios (see Sect. 2.2) entails exposure to elevated temperatures. For instance, [40] focuses on Silicon Carbide (SiC), offering comprehensive insights into the material’s behavior under extreme thermal conditions. The behavior of semiconductor components under elevated temperatures presents a dual challenge. Not only do these temperature conditions lead to increased power losses in the semiconductors, but they also contribute to a reduction in the overall operational lifespan of semiconductor devices.

The above-mentioned challenges underscore the need for thermal design considerations and robust temperature management systems. Addressing the multifaceted challenges of temperature dynamics in power electronics transcends traditional disciplinary boundaries. The ongoing research endeavors within the Cluster of Excellence-Sustainable and Energy-Efficient Aviation (SE\(^2\)A) framework emphasize a collaborative, cross-disciplinary approach, synergizing insights from thermodynamics and electrical engineering.

4.3 Pressure

Altitude, a determinant of low air pressure, contributes significantly to the domain of insulation coordination. Paschen’s law underlines challenges in this field. Paschen’s law can be defined as a mathematical relationship that provides the breakdown voltage, which signifies the minimum voltage required to initiate a discharge or electric arc between two electrodes placed within a gaseous environment. This breakdown voltage depends upon both the pressure of the gas and the distance between the electrodes.

However, it is important to note that Paschen’s law is applicable only to plane-parallel electrodes and specific gases. Investigations have been conducted to formulate empirical models on the phenomenon of creepage distances on printed circuit boards under the influences of temperature, relative humidity, and air pressure [41]. The current investigations focus on various geometries and materials, including the commonly used composite material Flame Retardant 4 (FR4), to address a broader range of scenarios beyond the scope of Paschen’s law. In short, to ensure that no electric arc occurs, the following rule applies: For a given voltage, the spacing between components needs to increase as pressure decreases, relative humidity increases, and contamination levels rise.

Challenges arise when addressing the trade-off between power density and required creepage and clearance distances. Augmentation of spacing reduces the power density, ultimately leading to a conflict of goals. The ambit of this challenge extends to semiconductor modules, where sufficient spacing must be ensured. Consequently, manufacturers encounter an imperative to expand their product spectrum, particularly to address the demands of aviation applications.

So far, no ideal solutions have manifested. However, some options are on hand to counter the challenges induced by Paschen’s law. The first one would be to hermetically seal the power electronics or connect it to the pressurized system on board. A defused variant would be to use a protective coating to ensure that no electric arcs can form due to the low air pressure.  [42] show how they employ the printed circuit board using double-sided mounting of devices. The printed circuit board substrate FR4 shows a higher breakdown voltage than air. Further research is needed, including in the area of materials science, to meet the competing demands for aerospace power electronics. This also includes, as outlined in [4], the development of new insulation materials with simultaneously high thermal conductivity.

4.4 Electromagnetic compatibility

Switched power electronic devices are the primary source of electromagnetic interference (EMI), generating significant conducted and radiated emissions due to their high voltage and current switching. The advent of fast-switching WBG devices has amplified electromagnetic compatibility (EMC) concerns in electronics and system design. In essence, EMI stems from steep voltage and current transitions inherent to power electronics, with steeper slopes leading to increased emissions.

In the safety-critical context of aviation, good EMC is crucial. To prevent interference between different systems, EMI must be shielded properly or, preferably, avoided as good as possible. Since traditional countermeasures consist, for instance, of metal shielding, which lowers gravimetric power density, or inductive filtering, which adds weight and lowers efficiency, EMC should be already addressed in the first design stage. Designed in from the very first stage, EMI occurrence can be reduced, lowering the effort for countermeasures.

An example of an early stage design-in is given in [20]. By optimizing the power commutation loop between a low inductive DC-link and the semiconductors, overvoltages and ripple frequencies are significantly reduced, mitigating radiated as well as conducted emissions. Another approach is functional integration as described in Sect. 5.3. This covers two issues: the minimization or omitting of motor cables, which act as an antenna for the applied switched voltage, and the possibility of a combined shielding for multiple systems and, therefore, weight savings.

4.5 Mechanical loads due to vibration and shock

The use of power electronics in aircraft also places new demands in terms of resilience to mechanical shocks and vibrations. Qualification tests for these loads are already standardized in the automotive environment. In principle, it is necessary to be able to simulate real loads with the help of field data. Methods similar to those described in Sect. 3.3 can be used. The analyses can range from deep inside the component [43, 44], board level [45] to the entire assembly [46]. The parameters for such a vibration test are the test duration, test frequency, stress orientation, vibration energy, and the stress application method (swept sine or random) [45].

The mechanical loads are already taken into account during the development phase. For example, finite element analyses are carried out to evaluate the effects of vibration and shock at an early stage [44]. Real test bench experiments, as described in [46], validate the assumptions used in simulations and demonstrate the robustness of the assembly. The longstanding use of power electronic components in the automotive sector enables manufacturers to respond to the high requirements in the aviation environment. This is supported by the fact that many components are encapsulated, also for thermal reasons, which greatly reduces the likelihood of premature failure in the event of vibration and shock.

4.6 Standards for electric aircraft development

Standardized regulations and requirements have been developed for conventional aircraft to ensure functionality and reliability. The aim is to minimize failure conditions of the aircraft and its components when exposed to the associated environmental conditions. The increased integration of power electronics, and therefore the introduction of new technologies, into the aircraft propulsion system makes amendments to existing standards necessary.

A large number of organizations are involved in the development and refinement of these standards. These include, for example, the International Civil Aviation Organization (ICAO), the Federal Aviation Administration (FAA), and the European Union Aviation Safety Agency (EASA). The latter organization is, in addition to releasing standards, especially active in publishing supporting documents on emerging technologies. These include so-called Certification Specifications (CS), for example for “Normal, Utility, Aerobatic and Commuter Aeroplanes” (CS-23) [47] and “Large Aeroplanes” (CS-25) [48], as well as Special Conditions (SC) for new designs not covered by CS. An example would be the SC E-19 on “Electric/Hybrid Propulsion System” [49].

Additional standardization bodies exist. “Environmental Conditions and Test Procedures for Airborne Equipment” are defined in the Radio Technical Commission for Aeronautics’ (RTCA) standard DO-160 [50], which is released in coordination with the European Organization for Civil Aviation Equipment’s (EUROCAE) ED-14. ASTM International offers various standards, such as the “Standard Guide for Aircraft Electrical Load and Power Source Capacity Analysis” (ASTM F2490-20) [51], the “Standard Specification for Electrical Systems for Aircraft with Electric or Hybrid-Electric Propulsion” (ASTM F3316/F3316M-19) [52], or the “Standard Specification for Design of Electric Engines for General Aviation Aircraft” (ASTM F3338-21) [53]. General standards for civil aircraft published by the organization SAE International, for example, are the “Guidelines for Development of Civil Aircraft and Systems” (SAE ARP4754) [54] or the “Guidelines for Conducting the Safety Assessment Process on Civil Aircraft, Systems, and Equipment” (SAE ARP4761) [55]. These are supplemented by additional standards for electrified propulsion aircraft. These include, for example, the standards SAE AIR6127, SAE AIR8678, SAE AIR7506, and SAE AIR6540, which deal with “Managing Higher Voltages in Aerospace Electrical Systems” [56], “Architecture Examples for Electrified Propulsion Aircraft” [57], “Impact of High Voltage on Wiring” [58], and “Fundamentals in Wire Selection and Sizing for Aerospace Applications” [59], respectively. With SAE ARP5584, a “Document for Electric Power Management” [60] is provided. In this context, MIL-STD 704F (“Aircraft Electrical Power Characteristics” [61]), which standardizes power quality and characteristics, is also frequently referred to.

5 Power electronics design space

To optimize power electronics accordingly to the performance measures highlighted in Sect. 3, a large number of degrees of freedom are available in power electronics design. A multi-dimensional design space results from taking into account the application-specific challenges discussed in Sect. 4 (see Fig. 1).

Improved performance is usually achieved by introducing new building blocks for converter design, by different converter topologies, or system approaches. In Sect. 5.1, the application of WBG semiconductor devices is discussed, as they pose the most influential building block in power converter design. As a topological option, multi-level inverters are discussed, and functionally integrated drive systems, as well as cryogenic power electronics, are presented as emerging technologies to optimize the overall system.

5.1 Application of wide band-gap semiconductor devices

The design of power electronic converter systems is centered around its switching elements, the power semiconductors. In conventional converters, silicon-based insulated-gate bipolar transistors (IGBT) in combination with anti-parallel freewheeling diodes are used as switching elements. The use of these components and the associated disadvantages, such as their bipolar forward characteristic, including a fixed on-state voltage drop, and high switching energies, result in IGBT-based converters often being operated at low switching frequencies in the 2–15 kHz range [62]. While there may be technological progress in the development of these conventional components, WBG semiconductors, such as switches based on SiC and GaN (Gallium Nitride), will play a key role in future power electronic systems [63, 64]. This is due to their excellent properties, for instance in terms of switching performance. These materials shine when efficiency and power density matter most. WGB devices boost both, especially at light loads, and enable efficient use of higher switching frequencies ranging from 50 to 200 kHz [62], reducing passive component size.

However, the use of WBG devices in variable-speed drive applications also poses a number of challenges. Especially the fast switching transitions at the inverter output can have a negative impact on the driven electrical machine. Either additional stress and losses in the machine are imposed, or the effect is counteracted by dedicated filtering at the inverter output, which in turn negatively affects the power density of the system. Furthermore, compliance with EMC requirements (see Sect. 4.4) has to be ensured. While Si power devices still are attractive in terms of cost, the recent adoption and commercialization of SiC-MOSFET lead to the assumption that the trend over the coming decade will be for development to focus on SiC [65]. For WBG, in particular, there currently is a strong focus on researching reliability and lifetime. By detecting shortcomings, semiconductors become more robust, more durable, and thus cheaper in the lifecycle. A detailed explanation is given in Sect. 3.3.

As described, the application of new power semiconductor materials poses certain challenges. “Drop-in replacements” into conventional power electronics converters are possible but do not fully exploit the technological potential. Developments at the component level (semiconductor material, but also packaging and passive components) significantly influence the possible converter design options. The full potential of WBG devices can be fully harnessed with the advance of integrated concepts (see Sect. 5.3) and specific inverter topologies optimized for WBG switches, such as multi-level inverters (see Sect. 5.2). Highly adapted designs are required [8, 20, 66].

5.2 Multi-level topologies

As discussed in Sect. 5.1, while there are many advantages in applying WBG devices, some limitations apply. In addition to considering steep voltage slopes [67], the application of such WBG devices is also limited by the DC-link voltage level. As discussed in Sect. 2.4, generally high DC voltages are preferred for transmitting high power. Both the rated and breakdown voltage of the power semiconductor result from the required DC-link voltage, which in turn depends on the topology the semiconductors are employed in [6]. Environmental influences play a role in evaluating the suitability of switches for a given application. Due to these influences (see Sect. 4), an additional voltage safety margin has to be considered in aerospace applications. This is especially relevant for WBG devices, where the component development in terms of breakdown voltages is currently still not as advanced as for their silicon counterparts.

Fig. 6
figure 6

Standard 2-level- (left) and 3-level ANPC inverter topology (right)

Full size image

Multi-level inverters, which do not stress the individual switches with the full DC-link voltage, are a viable option to increase HVDC voltage without having to increase the voltage requirements for the individual switches.

Since multi-level topologies contribute to several previously identified design goals, they are strongly being researched for electrified aircraft. A number of different topologies are researched and megawatt-scale multi-level designs by different research teams exist [11, 68, 69]. Especially inverters based on the active neutral-point clamped (ANPC) topology (Fig. 6) are considered a promising solution.

The ANPC topology offers superior efficiency over a wide operating range. Especially for aircraft applications, where the amount of energy that can be stored on board is strongly limited, high efficiency over the full mission profile is crucial to minimize power losses and extend aircraft operating range and flight times.

While the increased number of switch positions in such inverters (see Fig. 6) requires more power devices and increases circuit complexity, losses are distributed over many semiconductors, which provides extended degrees of freedom in converter operation and thermal design. A comprehensive overview of the degrees of freedom and associated design aspects for 3-level ANPC inverters is provided in [70].

A more complex topology inevitably raises questions as to its influence on reliability. A meaningful reliability analysis requires a probabilistic reliability assessment taking into account a multitude of degradation mechanisms. Intensive research efforts are currently being undertaken in this area. Recent studies, however, have shown, for example, that the lifetime of semiconductor switches (under consideration of the influence of bond wire fatigue) can be assessed positively for the 3-level ANPC compared to the standard 2-level topology [35]. For some topologies, the flexibility offered by the increased number of switch positions and resulting multiple redundant commutation paths results in improved post-fault performance. “Limp home” operational mode can be implemented, and the inverter can continue to operate at reduced power [71, 72]. This is essential for safety-critical application areas such as aviation.

Furthermore, compared to the standard 2-level inverter topology, multi-level inverters incorporate intrinsically lower EMI noise emissions at the inverter output, since its waveform approximates the desired sinusoidal shape better [73]. This is beneficial in applications with stringent noise emission regulations but also helps to reduce the size of passive filter components and shift losses away from the already heavily stressed electrical machines. [74] provides a good overview of two- to three-level midpoint clamped inverter topologies.

5.3 Functionally integrated drive systems

Recent technological developments on the device and on the topological level, as shown in Sects. 5.1 and 5.2, already hold great potential for aircraft applications. Inverters for electric propulsors are already demonstrated for power ratings up to several 100 kVA and gravimetric power densities of up to 40 kVA/kg [12, 75]. Short-term developments in SiC-MOSFET technology and system integration will further improve power rating, efficiency, and weight [11, 20, 76].

With functional integration, some beneficial points can be addressed. Conventional drive systems usually consist of distinguished functional parts, such as inverter, motor, and gearbox. The overall system can be optimized by integrating the individual functional units into a single, cohesive unit rather than using separate parts. A separated enclosure is not necessarily required for the inverter. Smartly distributing the power electronic components with respect to the motor phases allows lengthy motor cables to be avoided, saving space and weight and contributing to improved EMI behavior. Motor cables often are a significant source of EMI emissions, since they are exposed to switched voltages and nonlinear currents.

This emphasizes the holistic approach in which efficiency should be viewed as system efficiency instead of inverter, motor, and mechanics efficiency separately. By increasing the inverter switching frequency, losses can be decreased in the motor and increased in the inverter and vice versa. However, this has to be evaluated in each particular case, since the dependencies are complex and differ for different motor- and inverter topologies and power ratings. Similarly, by coupling these elements systematically, a shared cooling system can be implemented, allowing for further weight savings.

A recent aviation drive system design is given in Fig. 7. Here, a nacelle is equipped with a newly conceptualized propeller, an electric motor, and an integrated inverter [20]. The drive works gearless with the propeller attached to the outrunner rotor of the motor. As a scaled system, it is designed for a power rating of 200 kW at a rotational speed of 6000 rpm.

Fig. 7
figure 7

Integrated aviation propulsion system: Nacelle with an inverter, electric machine, and propeller (taken from [20])

Full size image

Its inverter, working at a DC-link voltage of VDC = 1000 V, is based on 1700 V rated SiC power modules, having a voltage derating of 42% to obtain an FIT rate of 1 at an altitude of 6000 m. The complete system has a redundant electrical architecture, consisting of two independent three-phase systems, where both systems can be driven independently. Its six phase’s power modules are distributed around the motor, directly connected to the motor phases, being controlled by one central controller.

A planar inverter prototype for laboratory experiments, designed as an electrical twin in a simplified environment, was built. It achieves an efficiency of 99.3% at rated power and a power density of 41.4 kVA/kg. These values are valid for the inverter alone. No auxiliary hardware is taken into account, and the power density does not include the cooling system nor the housing, as both are to be shared with the electrical machine in the final system design. A picture of the prototype is given in Fig. 8.

Fig. 8
figure 8

Planar inverter prototype (taken from [20])

Full size image

5.4 Cryogenic power electronics

The use of disruptive technologies such as cryogenic cooling [77] is expected to play a crucial role in achieving the ambitious goal of aircraft electrification. In cryogenic power electronics utilizing fuel cells, liquid hydrogen is a fuel source, which can also serve a dual purpose as a coolant for the components to achieve cryogenic temperatures [78]. This innovative approach not only ensures the improved performance of semiconductors [79, 80] but also utilizes the waste heat generated by the components to pre-heat the incoming hydrogen before it enters the fuel cell. The pre-heating of hydrogen is essential to maintain its temperature above the freezing point of water, thereby preventing ice formation and potential blockages in the system. Additionally, this approach is the enabler to use superconductors instead of heavy copper wires.

The behavior of different semiconductor materials Si (Silicon), SiC, and GaN differ under cryogenic temperatures. GaN, being a WBG semiconductor, exhibits a notable characteristic wherein the normalized on-resistance Ron decreases as the temperature decreases. This favorable feature ensures that GaN-based devices experience improved electrical conductivity and reduced losses at cryogenic temperatures. Consequently, the overall performance of GaN-based power electronics is enhanced, making them a compelling choice for cryogenic applications[77, 79, 80].

On the other hand, Si shows a slightly different behavior compared to GaN. While its Ron also decreases with decreasing temperatures, there is a noticeable increase in Ron visible at extremely low temperatures. This increase, though slight, poses certain challenges for Si-based devices under deep cold cryogenic conditions. However, for most practical applications, the decrease in Ron at higher cryogenic temperatures still leads to improved performance compared to normal temperature operation [77, 79, 80].

For SiC-MOSFET, the behavior is quite distinctive. SiC exhibits a significant increase in normalized Ron at deep cold cryogenic temperatures. This characteristic can potentially limit the performance of SiC-based power electronics when operating under such conditions. Careful consideration and additional optimization may be required when designing SiC-based devices for cryogenic applications [77, 79, 80].

Regarding total switching losses, all types of semiconductors benefit from cryogenic cooling. As the temperature decreases, the total switching losses in these materials reduce. This improvement in switching losses at cryogenic temperatures contributes to enhanced efficiency and power handling capabilities of the power electronics, making them even more suitable for high-performance aircraft electrification systems [77, 79, 80].

However, one critical challenge across all semiconductor types is the reduction in their breakdown voltage with decreasing temperature. This poses challenges, especially for high-voltage approaches and flights in high altitudes, where cosmic radiation can influence the system’s performance [77, 79, 80].

With the reduced losses achieved through cryogenic cooling, semiconductor devices offer new possibilities in aircraft electrification, enabling improved efficiency, higher power density, or an optimal combination of both. First, the lower losses enable the attainment of higher efficiency in power electronic systems. With decreased energy dissipation as heat, a larger proportion of the generated energy is utilized for propulsion. This enhanced efficiency translates into significant benefits for electric aircraft range requirements. By consuming less energy for the same power output, electric aircraft can achieve longer flight ranges, reducing the need for frequent recharging or refueling and thereby increasing their suitability for various missions and operational scenarios.

Second, the possibility of using semiconductor devices in “overdrive” operation at cryogenic temperatures opens doors to higher power densities, meaning that more power can be delivered in a compact form factor. The ability to achieve higher power densities is especially crucial for aerospace applications, where space and weight constraints are of importance. In combination with lower losses this results in an opportunity to strike an optimal balance between efficiency and power density. Engineers can fine-tune the design and operation of cryogenically cooled semiconductor devices to achieve a sweet spot where efficiency and power density are maximized simultaneously. This approach ensures that power electronic systems are efficient and capable of delivering the required power output for demanding aerospace applications. Striking this balance is vital for designing advanced aircraft propulsion systems that are both high-performing and energy-efficient. However, new challenges emerge, such as the need for additional components. These include but are not limited to, air compressors for the fuel cell, thick thermal insulation, and hydrogen tanks, affecting overall system power density [79, 81]. Bearing this in mind, such a disruptive approach only makes sense for new AEA developments. Current studies are investigating cryogenically cooled, fully electric, concepts for short [78] and long-haul aircraft [77].

Four major challenges for power electronics arise with the usage of hydrogen. First, hydrogen’s lower volume-related heat capacity compared to water, which is commonly used for cooling, poses a significant challenge. This lower heat capacity implies that hydrogen is less efficient in absorbing and dissipating heat. This limitation, coupled with hydrogen’s inherent volatility, complicates the cooler design.

Moreover, the phenomenon of hydrogen embrittlement adds another layer of complexity to the utilization of hydrogen as a coolant. Hydrogen embrittlement occurs when hydrogen atoms penetrate into the crystalline structure of metals, leading to potential structural integrity issues.

Furthermore, the hydrogen flow rate in cooling systems presents a challenge that demands careful consideration. The flow rate of hydrogen is directly proportional to the power demand of the system. This dynamic relationship necessitates detailed modeling to predict and analyze the temporal course of temperature changes and the spatial temperature distribution within the power electronics components.

Finally, the lifetime of power electronics requires further research. Comprehensive research on component longevity is conducted, encompassing both active and passive elements, within non-cryogenic conditions [25, 26, 28,29,30, 35, 41]. The extension of lifetime investigation to cryogenic temperatures is part of current research efforts.

6 Conclusion

The era of traditional fuels is rapidly coming to an end, paving the way for a new era centered on environmentally conscious electric aviation. This transformation is gaining momentum as various approaches are intensively explored. Power electronics is crucial for this transformation process. The large, multi-dimensional design space in this area requires a holistic approach to identify the optimal solution. There exists no generic solution that fits every aircraft type or flight mission. However, in the transition to fully electrifying aircraft, some distinct approaches are heavily researched and, therefore, discussed in this paper. Cryogenic systems are a promising approach for future long-range aircraft to make energy distribution and propulsion systems efficient and lightweight. Regardless of the cooling concept, particular design options exist to enhance the power electronics and overall propulsion system. These options include converter topologies, for instance, multi-level propulsion inverters, which contribute to improved output power quality, resulting in low filter effort. They also enable the usage of WBG switches in their current state of technology. Using WBG devices, efficiency as well as power density are improved, resulting in prolonged flight times and distances. Moreover, integrated drive systems can maximize power density through weight savings. Further advantages are, for instance, minimized EMI sources and optimized overall system efficiency. For the development of future hybrid- and full-electric aircraft applications, there are multiple points to address. Improved lifetime estimation of power electronics is under ongoing research. Subsystem refinement in the context of topology optimization to fully exploit the advantages of multi-level topologies is another field of research. Finally, the influences of harsh environmental impacts, such as cosmic radiation and low air pressure, are heavily researched due to their importance on power electronic devices.