Cryogenic Electronics: Part 1

by Dr. Randall K. Kirschman, consulting physicist, Silicon Valley, extelect@gmail.com

This is the first part of a two-part article about cryogenic electronics, the operation of electronic devices, circuits and systems at temperatures considerably below the “standard” electronics lower limit of −55° C. Alternative names are cryoelectronics, cold electronics and low-temperature electronics.

When cryogenic electronics is mentioned, superconductivity often comes to mind. Cryogenic electronics can be based either on superconductor devices or semiconductor devices, or on combinations of the two.[1] However, by far, semiconductor devices are the more widely used in practice for cryogenic electronics—and are the focus of this article.

Electronics are typically operated at room temperature because we humans live at room temperature, and room temperature electronics technology is well established. But room temperature is not necessarily the optimum temperature. This article’s aim is to present the reasons for lowering the operating temperature, why there is interest in cryogenic semiconductor electronics and to illustrate its range of applications with several examples.

Figure 2. Co-location can reduce microphonics (noise or variation in wiring capacitance resulting from vibration) , electromagnetic interference (EMI) , wiring capacitance, resistance and inductance , which reduce bandwidth. Also, preamplifier noise  may be reduced, and management of thermally generated voltages and impedance matching may be improved. Image: Randall K. Kirschman

Figure 2. Co-location can reduce microphonics (noise or variation in wiring capacitance resulting from vibration) , electromagnetic interference (EMI) , wiring capacitance, resistance and inductance , which reduce bandwidth. Also, preamplifier noise  may be reduced, and management of thermally generated voltages and impedance matching may be improved. Image: Randall K. Kirschman

Semiconductor-based electronic devices and circuits can operate over the entire cryogenic range, to the lowest temperatures: operation has been demonstrated down to ambient temperatures well below 1 K.

Beginnings
Cryogenic electronics is not a recent development: the operation of electronic devices, circuits and systems at cryogenic temperatures has been a practical and valuable technique for decades. It may be said to have begun—but neither superconductor nor semiconductor (and predating the Josephson Junction by a decade)—as early as 1951, when researchers evaluated a vacuum tube amplifier down to ≈14 K as a means to boost weak electrical signals from apparatus in a cryostat.[2]

During the 1960-70s, as they became available, semiconductor devices (diodes and transistors) of a variety of types were evaluated at cryogenic temperatures.[3] In the early years, conventional room-temperature electronic devices were used—and still are. But, as time passed, purpose-made devices and components have been designed and developed for cryogenic operation; some are available commercially.

Why?
Why is there interest in operating electronics at cryogenic temperatures? There are three primary motivations:

Figure 3. The SIDECARTM integrated circuit in a multichip package.  Image: Teledyne Imaging Sensors

Figure 3. The SIDECARTM integrated circuit in a multichip package. Image: Teledyne Imaging Sensors

• Improvement in performance, typically for some critical subsystem or performance “bottleneck”: e.g., improved signal-to-noise; increased efficiency and reduced power consumption; or increased processing speed;

• “Co-location”: locating part of the electronics of a system near a sensor, actuator or equipment operating at cryogenic temperatures (part of a distributed system);

• Desire or necessity of operating in an existing cold environment, artificial or natural: e.g., on a cold planet.
Often there is more than one motivation and the distinctions can be fuzzy. The following offers several practical examples to illustrate these motivations and the applications and capabilities of cryogenic semiconductor electronics.

Noise reduction for sensitive detection
Signal-processing for sensors—for example infrared detectors—is a frequent use for cryogenic electronics. To achieve the highest sensitivity, such detectors often must be operated at very low temperatures, sometimes within a fraction of a degree of absolute zero.

For certain transistors, lowering their temperature, thereby reducing thermal disturbances, decreases their noise and increases their gain (Figure 1); in other words, signal-to-noise ratio is enhanced. The initial stages for signal processing (e.g., preamps) are typically operated close to the sensor in the same—or nearby—cryogenic environment. This has been valuable to better realize the potential sensitivity of sensors that require cryogenic temperatures.

Figure 4. A three-stage microwave integrated circuit for the 4-12 GHz frequency range. Image: S. Weinreb, California Institute of Technology, Department of Astronomy

Figure 4. A three-stage microwave integrated circuit for the 4-12 GHz frequency range. Image: S. Weinreb, California Institute of Technology, Department of Astronomy

In addition to noise and gain benefits, locating the initial signal processing close to sensors has other benefits, which may be more important. Consider a sensor operated at cryogenic temperature: locating its preamplifier at room temperature (Figure 2, upper) would typically require a long connection from sensor to preamplifier, exposing the weak sensor signal to a number of hazards, which would typically degrade system performance. Susceptibility to these hazards can be greatly reduced by relocating the preamplifier close to the sensor, to directly reinforce the signal before sending it to room temperature (Figure 2, lower). Such initial signal processing at the cold sensor is termed close coupling or co-location; this arrangement can be considered a basic smart sensor in a distributed system.

For example, this concept was used for the infrared detectors in the Earth-orbiting IRAS (Infrared Astronomy Satellite, 1983), which used two-transistor preamplifiers co-located and operating at ≈60 K.[6] This was the first use of cryogenic electronics in space.. Image: Randall K. Kirschman”]Figure 5. Four out of many examples of microwave/mm-wave receiver noise reduction by cooling from RT (●) to deep cryogenic temperatures as indicated in K (●)[Data from 8-11]. Image: Randall K. Kirschman

Spacecraft signal processing
Cryogenic signal-processing electronics for spacecraft has since become much more sophisticated, with complex circuitry performing many functions, and is an essential ingredient in many observational spacecrafts. As a prime example, Teledyne’s purpose-designed SIDECAR™ (System Image, Digitizing, Enhancing, Controlling and Retrieving) system is capable of operating down to 30 K (Figure 3).[7] It comprises microprocessing, digital/analog, memory and interfacing circuits. It will be onboard several spacecraft due for launch during the 2020s; e.g, operating at approximately 40 K in the James Webb Space Telescope (JSWT), and operating below 100 K in the NEO Surveyor. Also due for launch during the 2020s is the Roman Space Telescope, which will incorporate NASA’s ACADIA electronics, operating at 140 K to support a detector array operating at 95 K.

Figure 6. Part of the VLA (Very Large Array): twenty-seven radio-astronomy antennas located at the National Radio Astronomy Observatory site in New Mexico. Each antenna in the array is 25 m (82 ft) diameter. Image: National Radio Astronomy Observatory/Associated Universities, Inc./National Science Foundation

Figure 6. Part of the VLA (Very Large Array): twenty-seven radio-astronomy antennas located at the National Radio Astronomy Observatory site in New Mexico. Each antenna in the array is 25 m (82 ft) diameter. Image: National Radio Astronomy Observatory/Associated Universities, Inc./National Science Foundation

Other co-location examples of cryogenic signal-processing electronics include gravitational wave antennas, scanning tunneling microscopes (STMs) and superconductive detectors.

Microwave receivers
A further example of noise reduction and gain improvement is provided by microwave/mm-wave preamplifiers (Figure 4). In this application there is no sensor: rather the signal is from an antenna. Cryogenic operation can reduce noise levels by an order of magnitude (Figure 5). Accordingly, hundreds of cryogenically cooled microwave/mm-wave receivers are used in radio-telescope arrays for radio astronomy (Figure 6) and deep-space communication. As Figure 5 indicates, typical operating temperatures are ≈10 to 30 K. A more familiar application, in the low microwave frequencies (approximately 1-2 GHz) is cell-phone station preamplifiers.

More efficient power conversion
In a very different realm, in terms of much lower frequency and much higher power level, cooling enables higher efficiency in power-conversion circuits, via reduction of parasitics. As one example, electrical resistance is reduced—thus power loss is reduced—when transistors are switched on (Figure 7 left). There can be three additional contributors to increased efficiency: (1) higher resistance (lower leakage current) when switched off, (2) lower dielectric losses and (3) reduced switching transients. All these may be reduced by cryogenic operation; as a result, waste heat from a power converter may be reduced by a factor of two to three (Figure 7 page 16). An additional advantage of cryogenic operation is increased switching frequency, enabling reduction of size and mass of components. and a SiGe bipolar power transistor[Data from 14]; right: power-loss reduction in a 100-W-output cryogenic power converter[Data from 15]. Image: Randall K. Kirschman”]Figure 7. Left: three of many examples of transistor on-resistance vs temperature: Si and GaN field-effect power transistors[Data from 12,13] and a SiGe bipolar power transistor[Data from 14]; right: power-loss reduction in a 100-W-output cryogenic power converter[Data from 15]. Image: Randall K. Kirschman

Even when waste heat is reduced, its removal is vital. Cooling to cryogenic temperatures facilitates heat removal because thermal conductivities of good thermal conductors (e.g., copper, aluminum, sapphire and semiconductors) are increased—typically by more than an order of magnitude (Figure 8). The electrical resistance reduction for copper can also reduce power loss in wiring.

The advantages of cryogenics for power electronics are being considered for land, sea and air transportation: hybrid and fully electric automobiles as well as trains, ships and aircraft. An associated concept is to use cryogenic liquid hydrogen (20 K) as a fuel, either for combustion engines or for fuel cells powering electric motors. A natural extension would be cryogenic operation of the associated power-conversion electronics for higher efficiency, to reduce waste heat and facilitate its removal. A range of possible applications of cryogenic power electronics—for power generation, management, distribution, and use—is described in Cold Facts, v. 29, n. 2, pp. 45, Spring 2013. [16]Figure-8

To be concluded in the next issue of Cold Facts.

Dr. Randall Kirschman is based in Silicon Valley and provides consulting related to R&D for electronic devices and circuits, as well as assistance in obtaining funding, particularly for extreme temperatures. He also presents professional development courses covering the principles and practice of low-temperature and high-temperature electronics. He received his Ph.D. from Caltech in physics and electrical engineering. E-mail: ExtElect@gmail.com, Web site: http://www.ExtremeTemperatureElectronics.com

References
[1] R. K. Kirschman, “Low-temperature electronics,” IEEE Circuits and Devices Magazine, v. 6, n. 2, pp. 12-24, March 1990; doi: 10.1109/101.46054

[2] A. N. Gerritsen and F. van den Burg, “The possibility of using an amplifier at low temperatures,” Physica, v. 17, n. 10, pp. 930-932, Oct. 1951; doi: 10.1016/0031-8914(51)90047-X (reprinted in [3]).

[3] “Low-Temperature Electronics,” ed. R. K. Kirschman, IEEE Press, 1986, ISBN: 0-87942-206-8, Part VII.

[4] A. T. Lee, B. Cabrera and B. A. Young, “Cryogenic preamplifiers for low-temperature particle detectors,” Low Temperature Detectors for Neutrinos and Dark Matter III, Editions Frontières, 1990, pp. 313-320.

[5] Y. Christoforou and O. Rossetto, “GaAs preamplifier and LED driver for use in cryogenic and highly irradiated environments,” Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, v. 425, n. 1-2, pp. 347-356, 1 April 1999; doi.org/10.1016/S0168-9002(98)01398-9

[6] F. J. Low “Application of JFets to low background focal planes in space,” Proc. SPIE 0280, Infrared Astronomy: Scientific/Military Thrusts and Instrumentation, Washington, D.C., 27 July 1981; doi.org/10.1117/12.931946

[7] http://www.teledyne-si.com/products-and-services/imaging-sensors/sidecar-asic

[8] S. Weinreb, J. C. Bardin and H. Mani, “Design of cryogenic SiGe low-noise amplifiers” IEEE Trans. Microwave Theory and Techniques, v. 55, n. 11, pp. 2306-2312, Nov. 2007; doi: 10.1109/TMTT.2007.907729

[9] A. H. Coskun and J. C. Bardin, “Cryogenic small-signal and noise performance of 32nm SOI CMOS,” 2014 IEEE MTT-S International Microwave Symposium (IMS2014), Tampa, Florida, 1-6 June 2014; doi: 10.1109/MWSYM.2014.6848614

[10] F. Thome, A. Leuther, J. D. Gallego, F. Schäfer, M. Schlechtweg and O. Ambacher, “70–116-GHz LNAs in 35-nm and 50-nm gate-length metamorphic HEMT technologies for cryogenic and room-temperature operation,” 2018 IEEE/MTT-S International Microwave Symposium (IMS), Philadelphia, Pennsylvania, 10-15 June 2018, pp. 1495-1498; doi: 10.1109/MWSYM.2018.8439685

[11] T. J. Reck, W. Deal and G. Chattopadhyay, “Cryogenic performance of HEMT amplifiers at 340GHz and 670GHz,” 2014 IEEE MTT-S International Microwave Symposium (IMS2014), Tampa, Florida, 1-6 June 2014; doi: 10.1109/MWSYM.2014.6848250

[12] A. J. Forsyth, C. Jia, D. Wu, C. H. Tan, S. Dimler, Y. Yang and W. Bailey, “Cryogenic converter for superconducting coil control,” IET Power Electronics, v. 5, n.6, pp. 739-746, July 2012, doi. 10.1049/iet-pel.2011.0287

[13] J. Colmenares, T. Foulkes, C. Barth, T. Modeert, R. C. N. Pilawa-Podgurski, “Experimental characterization of enhancement mode gallium-nitride power field-effect transistors at cryogenic temperatures,” 2016 IEEE 4th Workshop on Wide Bandgap Power Devices and Applications (WiPDA), Fayetteville, Arkansas, 7-9 Nov 2016, pp. 129-134; doi: 10.1109/WiPDA.2016.7799923

[14] R. K. Kirschman, Low-Temperature Electronics Course Notes, 2019.

[15] R. R. Ward, W. J. Dawson, L. Zhu, R. K. Kirschman, G. Niu, R. M. Nelms, O. Mueller, M. J. Hennessy and E. K. Mueller, “Novel SiGe semiconductor devices for cryogenic power electronics,” AIP Conference Proceedings, v. 824, n. 1, pp 351-358, 2006; doi: 10.1063/1.2192371

[16] R. K. Kirschman, “Does cryogenic electronics have a role in electric power?” Cold Facts, v. 29, n. 2, pp. 45, Spring 2013. ■