NASA NTRS2012
A large category of scientific investigation takes advantage of the interactions of signals in the frequency range from 300 to 1,000 GHz and higher. This includes astronomy and atmospheric science, where spectral observations in this frequency range give information about molecular abundances, pressures, and temperatures of small-sized molecules such as water. Additionally, there is a minimum in the atmospheric absorption at around 670 GHz that makes this frequency useful for terrestrial imaging, radar, and possibly communications purposes. This is because 670 GHz is a good compromise for imaging and radar applications between spatial resolution (for a given antenna size) that favors higher frequencies, and atmospheric losses that favor lower frequencies. A similar trade-off applies to communications link budgets: higher frequencies allow smaller antennas, but incur a higher loss. All of these applications usually require converting the RF (radio frequency) signal at 670 GHz to a lower IF (intermediate frequency) for processing. Further, transmitting for communication and radar generally requires up-conversion from IF to the RF. The current state-of-the-art device for performing the frequency conversion is based on Schottky diode mixers for both up and down conversion in this frequency range for room-temperature operation. Devices that can operate at room temperature are generally required for terrestrial, military, and planetary applications that cannot tolerate the mass, bulk, and power consumption of cryogenic cooling. The technology has recently advanced to the point that amplifiers in the region up to nearly 1,000 GHz are feasible. Almost all of these have been based on indium phosphide pseudomorphic high-electron mobility transistors (pHEMTs), in the form of monolithic microwave integrated circuits (MMICs). Since the processing of HEMT amplifiers is quite differ en t from that of Schottky diodes, use of Schottky mixers requires separate MMICs for the mixers and amplifiers. Fabrication of all the down-/up-conversion circuitry on single MMICs, using a ll-HEMT circuits, would constitute a major advance in circuit simplicity.