Understanding Semiconductors in Electronics: From Bandgaps to Burned MOSFETs

A semiconductor is a solid material, typically silicon or germanium, whose electrical conductivity falls between that of a conductor and an insulator, and can be precisely controlled by adding impurities (doping) or applying voltage. When you introduce semiconductors into a design, you change a circuit from a passive, linear network of resistors and capacitors into an active, non-linear system capable of switching, amplification, and digital logic. Hobbyists commonly confuse the raw semiconductor material (like a silicon wafer) with the finished component (like a diode or MOSFET), or they conflate solid-state switching with mechanical relay switching, ignoring the complex thermal and voltage-threshold behaviors unique to silicon.

If you are building power supplies, motor drivers, or microcontroller interfaces, you are relying on semiconductor physics every time you write a line of code or twist a wire. Let us break down how these materials actually behave on the workbench, complete with the math and failure modes that datasheets often bury.

The Physics of the Bandgap

To understand semiconductors in electronics, you have to understand the bandgap. In a pure silicon crystal at absolute zero, all electrons are locked in covalent bonds (the valence band). To conduct electricity, an electron must jump across an energy gap into the conduction band. Think of this gap like a highway toll booth: electrons must pay a specific energy toll to cross. For silicon at room temperature, that toll is exactly 1.12 eV (electron volts).

We manipulate this toll booth through doping:

• N-type doping: We inject phosphorus atoms, which have an extra electron. This creates a surplus of negative charge carriers sitting just below the conduction band, making it incredibly easy for current to flow.
• P-type doping: We inject boron atoms, which are missing an electron. This creates 'holes' (positive charge carriers) that act like bubbles in a liquid, allowing current to flow as electrons jump from hole to hole.

By sandwiching P-type and N-type silicon together, we create the PN junction—the foundational building block of every diode, bipolar junction transistor (BJT), and MOSFET you will ever solder. For a deeper dive into the atomic physics, the All About Circuits semiconductor textbook provides excellent lattice-level diagrams.

Where You Meet Semiconductors in Practice

You rarely interact with raw doped silicon; you interact with packaged components. Here is how the primary semiconductor devices divide up the work in a modern DC/AC circuit:

The critical takeaway for makers is the control mechanism. BJTs are current-controlled (you must supply continuous base current to keep them on), while MOSFETs and IGBTs are voltage-controlled (they draw virtually zero steady-state gate current, making them ideal for microcontroller GPIO pins).

Worked Numeric Example: MOSFET Power Dissipation

Let us look at a real-world calculation that separates hobbyists who melt their boards from those who build reliable hardware. Suppose you are using an IRLZ44N logic-level MOSFET to switch a 12V heater drawing 10A of continuous current.

According to the datasheet, the IRLZ44N has an on-state resistance ($R_{DS(on)}$) of 22 mΩ (0.022Ω) when the gate is driven at 5V. Let us calculate the power dissipated as heat inside the silicon die:

1. Calculate Power (P): $P = I^2 \times R_{DS(on)}$
2. $P = 10^2 \times 0.022 = 100 \times 0.022 = \mathbf{2.2W}$

2.2 watts does not sound like much, but a TO-220 package without a heatsink has a junction-to-ambient thermal resistance ($R_{\theta JA}$) of roughly 62 °C/W.

1. Calculate Temperature Rise: $\Delta T = P \times R_{\theta JA}$
2. $\Delta T = 2.2W \times 62 \text{ °C/W} = \mathbf{136.4 \text{ °C}}$

If your workshop ambient temperature is 25 °C, the silicon junction will hit 161.4 °C. The absolute maximum junction temperature for most silicon devices is 175 °C. You are operating just 13 degrees away from thermal destruction, and the plastic package will be too hot to touch. The fix: Add a small extruded aluminum heatsink with a thermal resistance of 15 °C/W, dropping the total thermal resistance and keeping the junction safely under 60 °C.

Real-World Scenario Walkthrough: The 3.3V GPIO Trap

Here is a classic bench failure involving ESP32 microcontrollers and power MOSFETs.

The Setup: You want to switch a 12V, 5A RGB LED strip using a 3.3V GPIO pin on an ESP32-WROOM-32. You dig through your parts bin and find an IRF520 N-channel MOSFET. You check the datasheet and see the Gate-Source Threshold Voltage ($V_{GS(th)}$) is listed as 2.0V to 4.0V. You reason that 3.3V falls right in the middle of that range, so the ESP32 will easily turn it on.

The Numbers: The IRF520 datasheet lists an $R_{DS(on)}$ of 0.27Ω. However, that specification is only guaranteed when $V_{GS} = 10V$. The $V_{GS(th)}$ of 2.0V-4.0V is defined as the voltage where the MOSFET conducts a microscopic 250µA—just enough to register on a multimeter, not enough to drive a load.

The Outcome: You wire it up and set the GPIO high. The LEDs glow dimly. Within five seconds, the IRF520 becomes scorching hot. The solder on your perfboard melts, and the component pops.

What Went Wrong: At 3.3V, the IRF520 is stuck in its linear (ohmic) region. It is acting like a partially closed valve. Instead of 0.27Ω, its effective resistance at 3.3V is closer to 3.0Ω. Pushing 5A through 3.0Ω results in $P = 5^2 \times 3.0 = \mathbf{75W}$ of heat dissipated in a tiny silicon die meant to handle maybe 2W without a massive heatsink. The silicon literally vaporized its internal bond wires.

The Fix: You must use a true logic-level MOSFET. A part like the IRLB8721 is specifically engineered to be fully enhanced (acting as a closed switch) at $V_{GS} = 2.8V$. At 3.3V, the IRLB8721 delivers an $R_{DS(on)}$ of about 8.7 mΩ, keeping your 5A load cool and your ESP32 safe.

Frequently Asked Questions