I've been looking at HEMT to use at low temperatures.

GaAs and GaN seem to be discussed often in the context of low temperature amplification.

My questions are twofold.

  1. To be suitable for low temperature operation, is it about having a small ionization energy value for electrons to break free?

  2. For low temperature operation, I naturally care about the transistors' power dissipation. I looked at the datasheet of ATF33143 (GaAs pHEMT) and it says its power dissipation is 600mW whereas the datasheet of CG2H40025 (GaN HEMT) does not seem to say anything about its power dissipation. It does mention "Saturated Power Output" as 6W, which I think is different. How does one go about finding the power dissipation? Is it just Idss*Vds? For ATF33143, Idss = 300mA, which was measured at Vds = 1.5V, which would give about 450mW, at least on the same order of magnitude as the power dissipation of 600mW given in the datasheet.


Do you mean HEMT devices made using GaAs and/or GaN to be used in an amplifier circuit which is then operated at low temperatures?

What I can find regarding GaAs and GaN in combination with low temperatures mostly refers to the fabrication of GaAs / GaN devices where the growing of those layers is done at a (relatively) low temperature. This appears beneficial as the crystal structure becomes better (less defects) compared to growing them at high temperatures.

If you mean something else, please include a link to where you found that information so we can look into it.

Regarding the power dissipation. You appear to treat the power dissipation as a property of the transistor. Like: "This transistor always has that power dissipation." It does not work like that.

To build an amplifier, a (HEMT) transistor is used in a circuit where it is biased at a certain voltage and/or current". All this determines the power consumption that the transistor will have. It is for example possible to take a transistor rated for use at 100 W and use it at 1 W. The 100 W is the maximum power the transistor can cope with (provided it is adequately cooled) but it can also work at a much lower power.

So the circuit and way you use it determine the power consumption. That's why you cannot find a direct answer for the power dissipation. There is no direct answer as it depends on how you use it.

| improve this answer | |
  • \$\begingroup\$ Thank you, I was asking about using GaAs pHEMT or GaN pHEMT transistors to amplify signals at cryogenic temperatures. In chapters 6 and 7 of the thesis here talks about it, (repository.tudelft.nl/islandora/object/…), (or you can go to this paper, which is basically chapter 6 of the thesis arxiv.org/abs/0708.0461)for example. \$\endgroup\$ – Blackwidow Sep 13 '18 at 19:14
  • \$\begingroup\$ I understand what you mean about power dissipation. But I guess I meant for a given target gain, you will have a different set of parameters, such as Vds and Id etc. I meant which kind of transisotors tends to give you the lowest power dissipation for the same gain. I guess there's no easy answer. \$\endgroup\$ – Blackwidow Sep 13 '18 at 19:16
  • \$\begingroup\$ I meant which kind of transisotors tends to give you the lowest power dissipation for the same gain Indeed, there is no easy answer as "it depends". Also gain by itself is already complex as a transistor by itself doesn't have "gain" it only has gain when used in a circuit where the gain also depends on input and output impedances to name only two parameters. The list of all parameters involved is quite long. \$\endgroup\$ – Bimpelrekkie Sep 13 '18 at 19:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.