How does the substrate material of a MMIC affect its noise figure at cryogenic temperatures?
Cryogenic MMIC Noise and Substrate Effects
Cryogenic MMIC amplifiers are the enabling technology for the most sensitive radio receivers in the world. The choice of semiconductor substrate and transistor technology (InP vs. GaAs vs. SiGe) directly determines the achievable noise performance at cryogenic temperatures.
| Parameter | Superheterodyne | Direct Conversion | Digital IF |
|---|---|---|---|
| Image Rejection | 60-90 dB (filter) | 30-50 dB (mismatch) | N/A (digital) |
| DC Offset | No issue | Major issue | No issue |
| LO Leakage | Low | High | Low |
| Integration | Difficult | Easy (single chip) | Moderate |
| Dynamic Range | 80-120 dB | 60-90 dB | 70-100 dB |
Frequently Asked Questions
Why is InP better than GaAs for cryogenic LNAs?
InP's lattice-matched InGaAs channel has higher electron mobility (15,000 vs. 8,000 cm^2/V-s at 77 K for GaAs) and higher electron velocity, resulting in lower channel resistance thermal noise. The InP substrate also has slightly better thermal conductivity at room temperature. At cryogenic temperatures, the InP HEMT noise advantage over GaAs grows because the mobility increase with cooling is larger for InGaAs. The approximately 2x noise advantage of InP over equivalent GaAs is well-documented across frequencies.
Can SiGe compete with InP for cryogenic LNAs?
At frequencies below approximately 2 GHz, SiGe HBTs can achieve comparable noise to InP HEMTs at cryogenic temperatures (2-5 K). The SiGe advantage is lower cost, mature fabrication, and excellent thermal conductivity of the Si substrate. Above approximately 5 GHz, InP HEMT dominates because its f_T and f_max are higher, providing more gain and lower noise. A hybrid approach (SiGe for DC-2 GHz, InP for 2-50 GHz) is used by some observatories.
How do I test a MMIC at cryogenic temperatures?
Mount the MMIC in a cryogenic test fixture inside a vacuum dewar cooled by a closed-cycle cryocooler. RF connections are made through coaxial feedthroughs or waveguide windows. Measure S-parameters using a VNA with cables routed through the dewar, and noise figure using the Y-factor method with a cold attenuator or cold-source technique. The measurement is complicated by: cable losses that change with temperature during cooldown, thermal gradients on the MMIC, and the need for very low-noise room-temperature post-amplification to measure the cryogenic DUT noise above the measurement system noise.