How much circuits knowledge is required for device physics?

[spongehateworl]

Hey everyone, I'm graduating with my bachelor degree in physics soon and intend to study physical electronics and device/semiconductor physics in graduate school. I've taken the standard undergrad physics curriculum (quantum mech, stat mech, electromagnetism, circuits, experimental physics) with an emphasis in solid state physics and electives in electrical engineering and materials science.

My knowledge in circuitry only consists of mostly basic semiconductor circuits with little design experience. Would it be beneficial for me to take more circuitry classes before graduating to prepare me for this field in graduate school?

 

[rbj]

if you're going to design transistors and integrated circuit components, you should know how transistors are hooked up to become basic logic gates in digital electronics. like NAND gates, NOR gates, NOT gates, a simple RS flip-flop, a D flip-flop (a.k.a. a "latch" or a single bit of memory). maybe understand more logic techniques like implementing a Boolean expression with gates and deriving a Boolean expression from a truth table.

and, for analog electronics, you should know how a basic amplifier works, how a differential amplifier works, how transistors are used for active loads (instead of resistors) of other transistors, and how transistors are used for constant current sources for differential amplifiers.

maybe look into rectifiers and power supplies (like how a Zener diode is used as a voltage reference).

those would be the most fundamental of circuits where your devices are one layer lower in terms of primitivity.

 

[Enthalpy]

Knowing circuitry would be very useful. Though, this doesn't meana course. Reading papers like Elektor would be at least as good (well, I mean, far better). Books for amateurs exist as well, also better than a course.

 

[Kholdstare]

Basic circuit theory is enough. Dont need to know graph theory, state theory, transfer function, pole-zero etc stuff.

 

[Kholdstare]

In fact only KCL and KVL is enough. Rest is just material physics.

 

[Studiot]

I'm really suprised no one mentioned two port network theory.

If you are working with device physics in the manufacture of components how would you do without it?

A good start would be to look at manufacturers component sheets and make sure you can understand all the parameters presented and their method of measurement and the test circuit used.

For instance a few examples

Rise Time
Fall (decay) time
h_fe
T_jc
R_th
Tristate output leakage current
f_t

 

[Kholdstare]

@Studiot - Touche. Although you're right a detailed knowledge of two port network is not really needed. Isn't it? I mean the device specifications (hybrid parameters/model params) will always be there from the project spec. He just need to adjust the device so that the two port parameters come out correctly.