3
\$\begingroup\$

I am taking a course on semiconductor physics and am learning about MOSFETs, MODFETs, and MESFETs.

I know the general structural differences between each one. A MOSFET is a metal, oxide, and semiconductor and you apply voltage to the gate to create an inversion layer in the semiconductor. A MODFET is similar to a MOSFET but its oxide is a doped AlGaAs combined with an undoped GaAs to create a strange dip in the band energy. This (somehow) creates a larger mobility for carriers in the 2-Dimensional Electron Gas layer.

A MESFET, I think, is the most dissimilar to the others because it is just a metal on a semiconductor. And you apply a voltage to create a large enough depletion region to completely deplete the layer so charge can't flow.

But what are the advantages and disadvantages to each one? Why would you want to use a MOSFET when the MODFET supposedly has a higher carrier mobility? Why would you use MODFET over a MESFET?

I am trying to construct a chart comparing and contrasting their characteristics, including (but not limited to): fabrication and structure and strain, parasitic effects, depletion region creation and shape, mobility of carriers, barrier height.

\$\endgroup\$
  • \$\begingroup\$ I took a shot at your first question. The second is more difficult. Remember there is more to choosing a device than performance: reliability, robustness to overstress, ESD, etc., and of course cost. Cost can have many un-obvious components: capital equipment required, finickiness of the process, yield, process time, ...; however I unfortunately have no idea how MODFETs stack up against MESFETs in any of these areas. \$\endgroup\$ – The Photon May 5 '14 at 0:55
  • \$\begingroup\$ Thanks for the accept, but really it's better to wait at least 24 hours before accepting any answer --- that gives people in every time zone around the world a chance to see your question and respond. Especially in this case there's a good chance someone will give a better answer if you wait a bit. \$\endgroup\$ – The Photon May 5 '14 at 2:36
5
\$\begingroup\$

I can only answer the first of your two main questions.

Why would you want to use a MOSFET when the MODFET supposedly has a higher carrier mobility?

Cost. First, if you want to make a GaAs/AlGaAs MODFET, that means you're working with GaAs wafers rather than silicon, and these are much more costly per device. Even if you make a MODFET with varying Ge content in SiGe, you are working with a more exotic and somewhat more costly material system than ordinary silicon.

Secondly, because the modified composition layer (AlGaAs or high-Ge SiGe) must be added by epitaxial growth, which is a slower and therefore more costly process than the ones needed to form an ordinary MOSFET on silicon.

Thirdly, because vast amounts of research have gone into reducing the cost of silicon CMOS processes over the past decades, allowing CMOS to be produced in high volumes, with high yields, and thus lower costs.

\$\endgroup\$

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.