What is transistor transconductance

Why is higher transconductance associated with a faster transistor switch?

Here is a quote from that site:

... higher "transconductance" - a measure of the performance of a transistor - than silicon transistors. The higher the transconductance, the faster the transistor can be switched on and off. This means that higher clock frequencies can be supported and that lower core voltages are required.

The argument in bold above is also shown in the Oxford dictionary under the example sentences: https://en.oxforddictionaries.com/definition/transconductance:

"The higher the transconductance, the faster the transistor can be switched on and off."

I know that the transconductance is the slope of the Iout Vin curve. The steeper the curve, the higher the transconductance of the transistor. However, this curve does not have a time axis. It seems that ∆Iout / ∆Vin via DC increments has nothing to do with time. Or is it? How can a faster change be related to a higher transconductance?

There is no time in the following representation (Ic Vbe are DC values ​​as they are capital letters, they are not instantaneous):

Brian Drummond

If the input voltage has a limited slew rate (volts / nanosecond), your time axis is given.

Jonk

Answering this question for MOSFETs is pretty straightforward as it is a relatively simple, symmetrical device. But it's a lot more complicated for the BJT. The BJT structure is inherently more complex, with internal nodes able to dominate performance. It will be interesting for me to see if a more comprehensive answer comes here that includes the BJT. I would probably learn something from it. But in general it is true that Gm is related to speed for both types of devices.

Andy aka

Very often the gate-source capacitance of a MOSFET is the dominant factor in how fast a MOSFET can be switched on or off. Rather high currents have to be injected into the gate capacitance in order to change the gate voltage quickly. So if the transconductance on MOSFET A is (about) twice as high as on MOSFET B, you only have to switch a certain load current appropriately, the gate voltage by half the amount compared to MOSFET B.

This normally leads to an increase in the switching speed for a given current injected into the gate.

Jonk

The OP read that this applies to the BJT (where it also applies). You went to the MOSFET where this issue is more directly related to C.Gs and calculates the RF response easily. Can you add anything to explain the BJT case? I think it can also apply as fT. = Gm2π (C.π + C.μ) for a BJT (although for BJT fT. Is also often only used as a number of the relative merit of the device and not as a function of its operating point cited.) And yes, I know the BJT is much more difficult to discuss on this point. Just curious if something could be added.

Andy aka

@jonk - I gave my answer before the OP edited their question to include the BJT's diagram and it's getting late. Also, the OP's first connection is all about carbon nanotubes, and the only transistor I can see mentioned was a MOSFET.

Jonk

OK. I'm fine. It's a complicated subject anyway. Looking forward to learning something is everything.

analogsystemsrf

Circuit time constants are C / g. Higher gm reduce the time constants.