Terminology » GaN-on-Si vs GaAs: Key Differences for RF Modules

GaN-on-Si vs GaAs: Key Differences for RF Modules

gaas gan-on-si rf front end semiconductor

Introduction : Wireless communication technologies are advancing rapidly from wi-fi 7, 5G advanced and 6G. Hardware powering these technologies require stringent RF front end modules (FEMs). These modules are placed between antenna and radio transceiver and takes care of transmission and reception of wireless signals. For decades, Gallium Arsenide (GaAs) was the material choice for RF front end modules (FEMs). As frequence and power demands increases, GaN on Si has been introduced to meet these challenging requirements. we will explore what these two semiconductor technologies are, why the industry is shifting and the key differences between them.

A typical FEM consists of following three main components.

  • Power Amplifier : Boosts the signal strength for transmission.
  • Low Noise Amplifier : Amplifies weak incoming signals from antenna without adding significant noise.
  • RF Switches : Toggles antenna connection between transmitting and receiving paths.

What is GaAs (Gallium Arsenide) ?

Gallium Arsenide is a mature and legacy compound semiconductor. Following are its advantages and disadvantages.

  • Benefits : GaAs heterojunction bipolar transistors (HBTs) and pseudomorphic high electron mobility transistors (pHEMTs) offer excellent linearity and power efficiency for traditional sub-6 GHz frequency bands.
  • Limitations : As modern wireless standards push into higher frequency ranges such as the FR3 band (i.e. 6 to 15 GHz) and millimeter wave (mmWave) bands; the transmit power requirements skyrocket. Under these extreme high frequency, high power conditions, traditional GaAs technology struggles with power efficiency and thermal management.

What is GaN on Si ?

Gallium Nitride (GaN) is a wide-bandgap semiconductor known for its ability to handle incredibly high power and frequencies. Historically, GaN was grown on Silicon Carbide (SiC) substrates, which offered excellent performance but was too expensive for consumer electronics like smartphones.

Modern innovations have allowed GaN-on-Si to be manufactured using 8-inch, CMOS compatible silicon foundries. Furthermore, the development of Enhancement mode (E-mode) GaN has allowed engineers to build “normally off” transistors. This eliminates the need for complex, space consuming negative voltage biasing circuits that were previously required for older GaN iterations.

Difference between GaAs and GaN on Si

FeatureGaAsGaN on Si
Break down voltageLowerMuch Higher
High Frequency EfficiencyDecreases significantly above 6 GHzExcellent up to 40 GHz
Wafer ScalabilityTypically 6-inch specialized foundries8-inch standard CMOS-compatible foundries
Integration LevelMulti chip modules usually requiredHigh (Single die PA, LNA, and Switch possible)
Best Use CaseLegacy sub 6 GHz mobile & wi-fiWi-Fi 7, 6G FR3 and mmwave

Conclusion

While GaAs has served the wireless industry faithfully for decades and will continue to be used in cost sensitive and lower frequency applications. GaN on Si represents the future of RF Front End Modules. By combining the raw power and high frequency capabilities of Gallium Nitride with the manufacturing scale and integration potential of Silicon, GaN on Si delivers the exact mix of performance, miniaturization and cost effectiveness required to make next generation networks like Wi-Fi 7 and 6G a reality.