Modified Full Text

I. Special Performance Requirements for Electronic Devices in Aerospace Scenarios In aerospace applications, electronic devices face two distinct extreme environmental dimensions: space orbits and celestial surfaces. The underlying physical failure mechanisms dictate stringent performance requirements.

Space Orbit Environment: In orbit, the high-vacuum environment renders convective heat dissipation ineffective, meaning system heat can only be dissipated via radiation. Meanwhile, devices are exposed to intense space radiation, which often necessitates radiation-resistant protective packaging for aerospace electronics.

Celestial Surface Exploration Environment: Devices must endure extremely wide temperature fluctuations (e.g., on the lunar surface, temperatures drop to -223 °C in polar regions and rise to 125 °C at the equator). Such extreme thermal cycling can cause fatigue cracking and delamination at packaging interfaces. At the electrical performance level, ultra-low temperatures induce a "carrier freeze-out" effect in semiconductor materials, leading to a sharp reduction in carrier concentration and severe degradation of device conductivity. Therefore, electronic devices in aerospace environments require corresponding thermal control systems.

Furthermore, aerospace electronics face extreme Size, Weight, and Power (SWaP) constraints.

II. Synergistic Complementarity and Substitutional Competition Between Silicon-Based and Wide Bandgap Semiconductors Silicon-based and wide bandgap semiconductors do not have an "either-or" relationship; rather, they synergistically complement each other in their primary domains and compete in specific scenarios.

1. Primary Domains Silicon-Based Semiconductors: Absolutely dominate the fields of microprocessors, high-computing-power logic chips, and high-density memory. Currently, there is a rigid demand for radiation-resistant protective packaging.

Silicon Carbide (SiC): Dominates high-power, high-voltage energy conversion scenarios. Under high-voltage bus architectures of 300V-400V and above (system power > 100kW), the wide bandgap and high thermal conductivity of SiC make it the preferred choice for deep space platforms and high-power satellites.

Gallium Nitride (GaN): Dominates high-frequency RF front-end and medium/low-voltage high-frequency switching scenarios. Based on the High Electron Mobility Transistor (HEMT) structure, GaN's switching characteristics in high-frequency communications and ultra-low temperatures are unmatched.

2. Analysis of Advantages and Disadvantages in Core Battlegrounds Power Devices (Power Management): In traditional satellite power systems, silicon-based IGBTs or MOSFETs are the mainstream. However, wide bandgap devices are engaging in substitutional competition.

The greatest advantage of SiC lies in reshaping the foundation of thermal design. After replacing silicon solutions, SiC chips can withstand junction temperatures above 200 °C, significantly increasing radiator surface temperatures. The heat dissipation power density can surge to over 2500W/m², greatly reducing satellite mass. Its disadvantages lie in radiation tolerance and ultra-low temperature performance. Under space high-voltage bus architectures, the single-event burnout (SEB) threshold of SiC power MOSFETs is relatively low, making radiation tolerance still a technical bottleneck. Meanwhile, in ultra-low temperature environments below -100 °C, the on-resistance of SiC MOSFETs abnormally increases significantly as the temperature drops, making them unsuitable for extremely cold environments.

In contrast, GaN HEMT devices exhibit strong substitution potential in the field of cryogenic power electronics. Studies show that GaN devices not only maintain stable and efficient switching performance at ultra-low temperatures of -223 °C (50K) without being affected by the carrier freeze-out effect, but also do not require consuming additional heat sources for preheating, making them an excellent choice for ultra-low temperature power converters.

RF Devices (Communication Electronics): In inter-satellite links and Earth-to-space communications, competition between silicon-based and GaN technologies is fierce. Silicon-based SiGe processes are mature and highly integrated, suitable for low-power transceiver link integration. However, in the power amplifier (PA) stage of the RF front-end, GaN holds an absolute advantage. Its extremely high power density and high-frequency characteristics allow phased array antennas to be made smaller and lighter. Meanwhile, its energy conversion efficiency during on-orbit operation far exceeds that of silicon-based devices, which is of decisive significance for satellite platforms constrained by solar panel area.

3. Analysis of Specific Application Scenarios GaN devices are preferred for ultra-cold/RF scenarios including lunar surface exploration and LEO inter-satellite links. In lunar rovers or hibernation-wakeup power distribution units (PDUs), GaN can achieve cold starts without the need for thermal enclosures, greatly reducing the burden on system size, power consumption, and weight.

III. Massive Capacity Demand for Space Computing Power — The TeraFab Project Based on currently available public information, we can basically confirm that aerospace electronics demand, represented by the mega TeraFab project, will become the largest growth driver for global semiconductor capacity in the long term.

1. Musk's Ambition: The Prototype of a Dyson Sphere and the Space Terawatt (TW) Computing Power Base The evolution of AI is facing a hard constraint at the energy level: the energy bottleneck. On Earth, rising power consumption from AI data centers is already straining terrestrial energy supplies.

To solve the energy bottleneck, Musk plans to deploy the AI computing power base in space orbits. In sun-synchronous orbits with minimal atmospheric attenuation, the power generation efficiency per unit area of space solar arrays is more than 5 times that of ground-based arrays.

Currently, SpaceX's reusable spacecraft have significantly reduced space launch costs. In the long run, deploying distributed terawatt (TW) AI computing nodes in space will be more cost-effective than expanding power grids and data centers on Earth. In the future, it is entirely possible for countries to collaborate to build electromagnetic mass drivers on the Moon, or even construct ultra-large-scale energy networks similar to a Dyson sphere, pushing computing power to the petawatt level.

2. Massive Capacity Demand Supporting this ambition requires ultra-large-scale chip capacity. The TeraFab project aims to achieve a total computing power chip capacity of 1 TW. According to SpaceX's plans, the annual target for deploying space computing power to orbit is as high as 100 GW. Calculated based on each ton of satellite carrying 100 kilowatts (100 kW) of computing power, deploying this computing power in space would require launching roughly 1 million tons of hardware to orbit each year. This requires chip manufacturing for aerospace electronics applications to break through the existing foundry capacity ceiling and custom design model, achieving an ultra-large-scale, highly standardized production model that far exceeds current global semiconductor manufacturing capacity.
3. Capacity Driving Effect on Power Semiconductors: Moving Towards "Ultra-Large-Scale Mass Production" Although space computing power projects and TeraFab primarily focus on advanced computing chips (logic and memory), the massive deployment of computing power in space will trigger a matching massive demand for space energy conversion/management and interstellar data transmission. In space, the DC power generated by terawatt-level solar arrays must be continuously, efficiently, and with low loss converted into low-voltage, high-current power to supply AI chips. Simultaneously, massive AI inference results must be transmitted back to Earth in real time through tens of thousands of inter-satellite laser links (Laser Mesh Network).

This system-level demand will have an unprecedented capacity-driving effect on photovoltaic panels and high-voltage/high-frequency wide bandgap semiconductors. The massive DC-DC power conversion networks and RF/laser communication matrices will depend heavily on SiC (for high-voltage bus power management) and GaN devices (for high-frequency communications and low-voltage switching). This means that special wide bandgap semiconductors, which were originally only "customized in small quantities for national key projects at high prices" in aerospace systems, will be forced to cater to the bill of materials for tens of thousands of AI computing satellites annually in the future. Demand of this magnitude will directly drive the power semiconductor supply chain towards "ultra-large-scale mass production."

IV. Current Progress and Technical Challenges of Semiconductors in Aerospace Applications

1. Current Progress: Commercial Chips Have Been Verified "Anti-Freezing" New Materials Shine: In the past, chips in spacecraft were extremely delicate and had to operate in "thermal enclosures" that consume substantial power. However, new wide bandgap semiconductors such as Gallium Nitride (GaN) and Silicon-Germanium (SiGe) have now been proven to work stably in the extreme cold of over minus 200 degrees (such as the lunar polar night environment).

Commercial Off-The-Shelf (COTS) Space Verification: Aerospace chips are breaking the curse of "only small-batch, high-cost custom manufacturing." Mature commercial semiconductor devices and lithium batteries on the market have been proven to withstand the extreme "freeze-thaw" cycles in space and successfully wake up. This means that some equipment can shed bulky thermal protection devices, reducing spacecraft size and cost substantially.

2. Technical Challenges: "Hardcore" Challenges Brought by Extreme Environments The "Physical Limits" of Extreme Temperature Swings: Celestial bodies such as the Moon experience extreme thermal swings. Traditional silicon-based chips will directly stop functioning due to carrier freeze-out in extreme cold, while silicon carbide devices (mainly SiC MOSFETs) will also experience an abnormal surge in on-resistance at ultra-low temperatures. Additionally, alternating cold and heat will cause cracking of chip packaging materials due to mismatched thermal expansion and contraction, as well as solder joint fatigue. This is currently a major bottleneck hindering long-life aerospace missions.

The "Fatal Blow" of Space Radiation: Cosmic high-energy particles have extremely strong penetrating power. In high-power, high-voltage aerospace applications, high-energy particles easily cause catastrophic "single-event burnout" in power chips such as Silicon Carbide (SiC). Key technical roadblocks include balancing high-voltage operation and radiation tolerance for power devices, and reconciling protective packaging requirements with SWaP constraints for core computing chips.

Lack of Industry Certification Standards: Globally, there are no unified certification standards for ultra-low-temperature (below -55 °C) aerospace chips, and foundries lack corresponding test models. This leads to long R&D cycles and extremely high trial-and-error costs for new aerospace chips.

V. List of Industry-Academia-Research Stakeholders in the Aerospace Electronics Fields of China and the US (Partial) Note: The aerospace sector in all countries is closely tied to the defense industry. Therefore, the tables below only list selected publicly disclosed organizations.

China

United States

References

1. SpaceX Prospectus
2. SpaceX and FCC International Public Communication Concentration "Orbital Data Center" Project Application Materials
3. NASA - Cold Electronics for Lunar Missions Technical White Paper
4. TeraFab Project Launch Speech
5. Beijing Satellite Manufacturing Plant, CAST (China Academy of Space Technology), Wan Chengan, Discussion on the Aerospace Application of Wide Bandgap Power Devices, academic report