Bipolar Junction Transistor (BJT): NPN and PNP Structure

Every analog design—from a simple LED driver to a GHz LNA—ultimately reduces to controlling current with current or voltage. The bipolar junction transistor (BJT) was the device that made this practical in silicon, and understanding its physical structure (not just its "amplifier symbol" behavior) is what lets you predict Early effect, saturation voltage, breakdown, and thermal runaway instead of just looking them up.

Physical Structure

A BJT is two back-to-back p-n junctions sharing a thin common region, formed by three doped layers in sequence:

NPN:            PNP:
  E   B   C       E   B   C
  N | P | N       P | N | P

• Emitter (E): heavily doped (denoted N⁺ or P⁺). Its job is to inject carriers efficiently into the base.
• Base (B): very thin (typically 0.1–1 µm) and lightly doped. Thinness is what lets injected carriers cross it by diffusion before recombining.
• Collector (C): moderately doped, physically the largest region, and lightly enough doped that it can sustain a wide depletion region for high breakdown voltage.

The doping is deliberately asymmetric — emitter >> base, collector < emitter. This is not incidental; it's the entire reason a BJT amplifies. The base-emitter junction is optimized for efficient carrier injection (asymmetric doping maximizes the fraction of current carried by minority carriers into the base rather than back into the emitter), while the base-collector junction is optimized to support a large reverse voltage without breaking down.

NPN vs. PNP: Same Physics, Mirrored Polarity

NPN and PNP are mirror images: swap every polarity (voltage, current direction, doping type) and the equations are identical. NPN devices dominate in practice — especially at high frequency — because electron mobility in silicon is 2–3× higher than hole mobility, giving NPN devices higher fT for the same geometry. PNP transistors remain essential wherever a design needs to pull current from a positive supply or complement an NPN (push-pull output stages, current mirrors, level shifting).

Why the Structure Produces Current Gain

Consider an NPN in the active region: base-emitter junction forward-biased, base-collector junction reverse-biased.

1. Forward bias on the B-E junction injects electrons from the N⁺ emitter into the thin P base (and, much less significantly, holes from base into emitter — minimized by the emitter's much higher doping).
2. Because the base is thin and lightly doped, the vast majority of injected electrons diffuse straight across it rather than recombining with base majority carriers (holes).
3. On reaching the base-collector depletion region, the strong reverse-bias field sweeps these electrons into the collector — this becomes collector current IC.
4. The small fraction of electrons that do recombine in the base must be resupplied by the base terminal — this is the base current IB, and it's small precisely because recombination is minimized by design (thin base, high emitter doping, long minority-carrier lifetime).

The ratio of these is the current gain:

β (hFE) = IC / IB        (typically 50–300 for small-signal BJTs)
IE = IC + IB              (KCL at the three terminals)
IC ≈ α·IE,   α = β/(β+1) ,  α typically 0.98–0.997

Worked check: if β = 150 and IB = 20 µA, then IC = β·IB = 150 × 20 µA = 3.0 mA. Emitter current IE = IC + IB = 3.0 mA + 0.02 mA = 3.02 mA. Verify with α: α = 150/151 = 0.9934; IC = α·IE = 0.9934 × 3.02 mA = 3.00 mA. ✓ Matches, confirming the KCL relationship holds self-consistently.

This structural picture — thin lightly-doped base sandwiched between an efficient emitter and a breakdown-tolerant collector — is the physical origin of every first-order BJT parameter: β depends on base width and doping ratio, breakdown voltage (BVCEO) depends on collector doping and depletion width, and fT depends on how fast carriers transit the base.

Structural Asymmetry Has a Practical Consequence

Because the emitter and collector are doped and sized differently, you cannot simply swap them and expect the same performance. Running a BJT "backwards" — using the collector terminal as emitter and vice versa — still conducts (both junctions still exist), but with drastically reduced current gain, often β < 5, because the collector is not optimized for efficient carrier injection and the base geometry seen from that side is different. This is why datasheets carefully distinguish pin assignments, and why accidentally reversing E and C on a board is a common, hard-to-spot bring-up bug: the circuit may partially work but with far lower gain than expected.

Design Implications of the Physical Picture

• Base width sets both β and fT tradeoffs: a thinner base raises β and speeds up transit time (higher fT) but lowers the base-collector breakdown voltage and increases the risk of punch-through at high VCB.
• Emitter doping sets injection efficiency: heavier emitter doping raises β but can reduce breakdown voltage (BVEBO) and increase base-emitter junction capacitance.
• Collector doping/thickness sets voltage rating: power BJTs use a lightly doped, wider collector (sometimes with an added epitaxial layer) specifically to support high BVCEO, at the cost of higher ohmic resistance and larger VCE(sat).
• Thermal behavior follows structure: VBE has a negative temperature coefficient (~ −2 mV/°C), and β increases with temperature — this combination is why paralleled BJTs need emitter degeneration resistors to avoid thermal runaway (covered in detail under device matching).
• NPN vs. PNP selection: default to NPN for speed-critical or high-current low-side switching; use PNP for high-side switching from a positive rail or for complementary output stages where symmetric push-pull drive is needed.

Key Takeaways

• A BJT is two p-n junctions sharing a thin, lightly-doped base; asymmetric doping (emitter heaviest, base thinnest, collector for breakdown) is what enables current gain rather than just two back-to-back diodes.
• NPN and PNP are physical mirror images — same equations, all polarities reversed — with NPN generally faster due to higher electron mobility in silicon.
• Current gain arises because injected minority carriers mostly diffuse across the thin base to the collector, with only a small recombination-driven base current needed to sustain the process: β = IC/IB, IC ≈ α·IE.
• Emitter and collector are not interchangeable despite the symbol's apparent symmetry — swapping them collapses β due to the collector's lower doping and different geometry.
• Base width, emitter doping, and collector doping/thickness are the physical knobs behind the β/fT/breakdown-voltage tradeoffs seen on every datasheet.