From silicon atoms to the chip in your phone — with animated p-n junction formation, interactive diode V-I simulator, 3D doping visualization, rectifier waveform animations, and 20+ real-life examples.
Before diving in, here's the entire chapter on one page. This is your mental map — every topic is a link in this chain.
Semiconductors have intermediate conductivity — controllable by doping and temperature. Pure (intrinsic) Si/Ge has equal electrons and holes. Adding pentavalent/trivalent dopants creates n-type/p-type. Joining p and n creates a diode that conducts ONE way. Diodes rectify AC to DC — the foundation of every power supply and electronic circuit.
Before the 1950s, all electronics used vacuum tubes — glass bulbs with heated filaments, large, fragile, and power-hungry. Then came the transistor (1947, Bardeen, Brattain, Shockley) — a tiny solid-state device made of semiconductors. Today, a single chip smaller than your fingernail contains billions of transistors.
| Property | Vacuum Tube | Semiconductor Device |
|---|---|---|
| Size | Centimeters (bulb-sized) | Nanometers (chip-scale) |
| Power | High (needs heater) | Very low |
| Voltage | Hundreds of volts | A few volts |
| Lifetime | Limited (burnout) | Very long |
| Reliability | Fragile | Very high |
Even older CRTs (cathode-ray tubes) in old TVs and monitors are now replaced by LCD/OLED displays powered by solid-state transistors. Before the physics was fully understood, a natural galena (PbS) crystal with a metal point contact was already used as a radio detector!
Semiconductor devices replaced vacuum tubes — smaller, cheaper, more reliable, lower power. Today's world runs on silicon. All modern electronics depend on understanding p-n junctions, diodes, and transistors.
Metals: Very low resistivity (high conductivity)
ρ ~ 10⁻² – 10⁻⁸ Ω·m · σ ~ 10² – 10⁸ S/m
Semiconductors: Intermediate
ρ ~ 10⁻⁵ – 10⁶ Ω·m · σ ~ 10⁵ – 10⁶ S/m
Insulators: Very high resistivity (low conductivity)
ρ ~ 10¹¹ – 10¹⁹ Ω·m · σ ~ 10⁻¹¹ – 10⁻¹⁹ S/m
Note: Ranges are indicative — some materials may fall outside. Resistivity is not the only criterion.
Elemental: Si (silicon), Ge (germanium) — pure elements
Compound (inorganic): CdS, GaAs, CdSe, InP — combinations
Organic: Anthracene, doped phthalocyanines
Organic polymers: Polypyrrole, polyaniline, polythiophene
Most current devices use Si or Ge (elemental) and GaAs (compound). Organic semiconductors are an emerging field (OLED displays!).
Valence band: Energy levels of valence electrons (outer shell)
Conduction band: Higher energy levels where electrons can move freely
Band gap Eg: Energy difference between the top of valence band (E_v) and bottom of conduction band (E_c)
Metals: Eg ≈ 0 (bands overlap). Conduct even at 0 K.
Insulators: Eg > 3 eV. No electrons in conduction band even at room temperature.
Semiconductors: Eg < 3 eV. Some electrons are thermally excited to conduction band at room temperature.
Mistake: "Semiconductors have no conductivity."
Correction: They DO conduct — just not as much as metals. At room temperature, pure Si has ~10¹⁰ free electrons per cm³ — enough for measurable conductivity.
Mistake: "Semiconductors are always n-type or p-type."
Correction: Pure (intrinsic) semiconductors are neither! They become n- or p-type only after doping with impurities.
Three classes by band gap: metals (Eg ~ 0), insulators (Eg > 3 eV), semiconductors (Eg 0.2–3 eV). Silicon's 1.1 eV bandgap is ideal for room-temperature electronics. Conductivity is tunable via doping and temperature.
Si and Ge have a diamond-like crystal structure — each atom is bonded to 4 nearest neighbors by covalent bonds (sharing electrons).
Si and Ge are tetravalent (4 valence electrons). Each atom shares 1 electron with each of its 4 neighbors, forming 4 covalent bonds.
Lattice constants: Si = 5.43 Å, Ge = 5.66 Å, C (diamond) = 3.56 Å
At 0 K, all electrons are locked in bonds → no free electrons → semiconductor behaves as insulator.
At T > 0 K, thermal energy can break a covalent bond. The electron becomes free (conduction electron), leaving behind a hole (vacancy with effective positive charge +e).
For intrinsic semiconductor: n_e = n_h = n_i (intrinsic carrier concentration)
Both electrons and holes contribute to conduction. Total current: I = I_e + I_h
n_e = n_h = n_i (carrier concentration)
Both electrons (negative) and holes (positive) contribute to current:
I = I_e + I_h
Direction of current:
Generation ↔ Recombination equilibrium: at steady state, generation rate = recombination rate.
Because when a bond breaks, one electron leaves its valence site and one hole is left behind. The process is inherently paired — the same thermal energy creates both. So n_e = n_h for pure semiconductors. This balance holds until we add impurities (doping).
Intrinsic semiconductor: pure Si/Ge. Each atom shares 4 valence electrons with neighbors (covalent bonds). Thermal energy creates electron-hole pairs. n_e = n_h = n_i. Both carriers contribute to current. At 0 K, behaves as insulator.
Pure silicon has too few free carriers for practical devices. Solution: add a tiny amount of impurity — a few parts per million — to dramatically increase conductivity. This is doping.
Doping: Deliberate addition of small amounts of impurity atoms to a pure semiconductor.
Dopant: The impurity added.
Requirements for doping to work:
Dopant: Pentavalent (5 valence electrons) — P, As, Sb
4 of the 5 valence electrons form covalent bonds with neighboring Si atoms. The 5th electron is very weakly bound (ionization energy only ~0.01–0.05 eV) and easily becomes a free conduction electron.
Called "donor" impurity because it "donates" a free electron.
Result: n_e >> n_h (electrons are majority carriers, holes minority).
Dopant: Trivalent (3 valence electrons) — B, Al, In
3 of the 4 covalent bonds form normally. The 4th bond has a missing electron = a hole. The dopant is called an acceptor — it can accept an electron from a neighboring bond.
Result: n_h >> n_e (holes are majority carriers, electrons minority).
For any extrinsic semiconductor (n or p):
n_e × n_h = n_i² (always, in thermal equilibrium)
Where n_i = intrinsic carrier concentration.
So if we dope with donors (increasing n_e), the hole count n_h must decrease to maintain the product.
Crystal remains electrically neutral: donor ions (positive) + extra electrons (negative) balance out.
Si has 5×10²⁸ atoms/m³. Doped with 1 ppm As (pentavalent):
N_D = 5×10²⁸ × 10⁻⁶ = 5×10²² donors/m³
Since N_D >> n_i (~10¹⁶), we get: n_e ≈ N_D = 5×10²² m⁻³
Using n_e n_h = n_i² = (1.5×10¹⁶)² = 2.25×10³²:
n_h = 2.25×10³² / 5×10²² ≈ 4.5×10⁹ m⁻³
So n_e / n_h ≈ 10¹³ — electrons are 13 orders of magnitude more abundant than holes!
Mistake: "n-type is negatively charged."
Correction: n-type is electrically NEUTRAL. The donor atom becomes positive when it donates an electron, but that positive charge balances the added electron. Overall charge = 0.
Mistake: "n-type has more electrons than holes only because of doping."
Correction: True, but there's a subtlety — the number of intrinsic (thermally generated) holes DECREASES when you add donors. The product n_e × n_h is fixed at n_i².
Doping (ppm level) dramatically increases conductivity. n-type (pentavalent dopant): electrons majority. p-type (trivalent dopant): holes majority. Mass action law: n_e × n_h = n_i². Crystal remains neutral.
A p-n junction is formed by joining p-type and n-type semiconductor materials. This is the basic building block of diodes, transistors, LEDs, solar cells — nearly all semiconductor devices.
Take a thin p-type Si wafer. Dope one side with pentavalent (n-type) impurity. The junction forms at the boundary.
Note: You CANNOT just glue p-type and n-type slabs — the atomic-scale contact requires proper crystal continuity (Example 14.3 in NCERT).
Depletion region: A thin layer (~10⁻⁷ m) at the junction where free charges have diffused away, leaving behind immobile ions.
p-side edge: negative ions (from acceptor atoms that gained electrons)
n-side edge: positive ions (from donor atoms that lost electrons)
Electric field: Points from n (positive) to p (negative)
Barrier potential V₀: The potential difference across the depletion region. Typical values:
Diffusion current: Due to concentration gradient — holes p→n, electrons n→p. Direction: p→n internally.
Drift current: Due to the E-field — minority carriers swept across junction. Direction: n→p internally (opposite to diffusion).
At equilibrium: diffusion = drift → net current = 0.
p-n junction formation: diffusion creates depletion region + electric field. Barrier potential V₀ (0.7V Si, 0.3V Ge). Equilibrium: diffusion current = drift current, net current = 0. This is the foundational structure of all semiconductor devices!
A diode is a p-n junction with metallic contacts. Its key property: conducts current easily in ONE direction (forward), blocks it in the other (reverse).
Triangle pointing from p to n:
p →|— n
Arrow indicates direction of conventional current in forward bias.
Setup: p-side connected to + terminal of battery, n-side to − terminal.
Effect: Applied voltage opposes barrier potential → effective barrier becomes (V₀ − V) → depletion region narrows → electrons/holes easily cross junction.
Result: Forward current flows, typically milliamperes (mA).
Threshold voltage: Below this voltage, current is small. Above it, current rises steeply. Threshold ≈ 0.7 V (Si) or 0.2–0.3 V (Ge).
Setup: n-side connected to + terminal, p-side to − terminal.
Effect: Applied voltage adds to barrier → effective barrier becomes (V₀ + V) → depletion region widens → diffusion current suppressed.
Result: Only tiny drift current (μA) flows due to minority carriers. This is called the reverse saturation current — nearly independent of applied voltage.
Breakdown: At very high reverse bias (V_br), current increases sharply. If not limited, the diode may be destroyed.
I = I₀ (e^(eV/kT) − 1)
Where I₀ = reverse saturation current, k = Boltzmann constant, T = temperature.
For forward bias (V >> kT/e): I ≈ I₀ e^(eV/kT) — exponential rise.
For reverse bias (V << 0): I ≈ −I₀ — nearly constant small negative current.
Trap: "Diode conducts in both directions."
Correction: NO — a regular diode conducts easily only in forward bias. In reverse bias, only a tiny leakage current (μA) flows. The diode is fundamentally a one-way valve.
Trap: "Threshold voltage is the same for all diodes."
Correction: It depends on material: Si ≈ 0.7 V, Ge ≈ 0.2–0.3 V. Also depends on temperature and current level (slightly).
Diode = p-n junction with one-way conduction. Forward bias: barrier reduced (V₀−V), current flows (mA). Reverse bias: barrier increased (V₀+V), only tiny drift current (μA). Threshold: 0.7 V Si, 0.2 V Ge. Shockley equation: I = I₀(e^(eV/kT) − 1).
A rectifier converts AC (which reverses direction) to DC (which flows one way). It uses the one-way property of diodes.
Circuit: AC source → transformer (optional) → diode → load resistor R_L
Operation:
Result: Pulsating DC output — only during positive half-cycles
Output frequency = input frequency (50 Hz in India)
Efficiency: 40.6% (only half the AC wave is used)
Circuit: Centre-tapped transformer + 2 diodes (D₁, D₂) + load R_L
Operation:
Result: Both half-cycles used → smoother DC output
Output frequency = 2× input frequency (100 Hz for 50 Hz input)
Efficiency: 81.2% (double the half-wave)
Alternative: Bridge rectifier uses 4 diodes but no centre-tap.
A capacitor connected in parallel with R_L smooths the pulsating DC.
During peaks: Capacitor charges to peak voltage.
Between peaks: Capacitor discharges slowly through R_L, filling in the "gaps."
Time constant τ = R_L·C — larger = smoother output.
Output is closer to peak voltage — nearly steady DC!
Most electronic circuits (ICs, microprocessors, LEDs, batteries) require DC. AC is great for transmission (easy to step up/down via transformers), but devices need steady DC to operate. The rectifier is the bridge between AC grid and DC electronics.
Rectifier: AC → DC using diode's one-way property. Half-wave: 1 diode, uses positive half-cycle, 40.6% efficiency. Full-wave: 2 diodes + centre-tap (or 4 in bridge), uses both halves, 81.2% efficiency. Capacitor filter smooths output to nearly steady DC. Universal in all power supplies!
Q1. C, Si, and Ge have the same lattice structure. Why is C (diamond) an insulator while Si and Ge are intrinsic semiconductors?
All three have 4 valence electrons. The energy to free an electron depends on which orbit those electrons are in.
C (diamond): Eg = 5.4 eV (electrons in n=2 orbit — tightly bound)
Si: Eg = 1.1 eV (n=3 orbit)
Ge: Eg = 0.7 eV (n=4 orbit)
The valence electrons are in higher n orbits for Ge and Si — easier to free.
For C, thermal energy at room temperature (0.026 eV) is too small to jump the 5.4 eV gap → effectively insulator. For Si/Ge, gap is small enough → some electrons jump → intrinsic semiconductor.
Q2. Pure Si crystal has 5×10²⁸ atoms/m³. Doped by 1 ppm of pentavalent As. Find n_e and n_h. Given n_i = 1.5×10¹⁶ m⁻³.
N_D = 5×10²⁸ × 10⁻⁶ = 5×10²² m⁻³
Since N_D >> n_i, we have n_e ≈ N_D = 5×10²² m⁻³
n_e × n_h = n_i² = (1.5×10¹⁶)² = 2.25×10³²
n_h = 2.25×10³² / 5×10²² = 4.5×10⁹ m⁻³
n_e / n_h = 10¹³ — electrons dominate by 13 orders of magnitude!
Q3. Can we take a p-type semiconductor slab and physically join it to an n-type slab to get a p-n junction?
NO!
Even the flattest slab has surface roughness much larger than atomic spacing (~2–3 Å). Continuous contact at the atomic level is impossible.
The gap would act as a discontinuity — charge carriers would have to tunnel across a vacuum/air gap, which doesn't form a proper junction. The device would have poor electrical properties and not behave as a diode.
By diffusing dopant atoms into one side of a single crystal (e.g., heating phosphorus onto a p-type wafer). Or by epitaxial growth of n-type on p-type. The lattice must be continuous!
Q4. From the given Si diode V-I curve: at I = 20 mA, V = 0.8 V; at I = 10 mA, V = 0.7 V. Find forward resistance. Also find reverse resistance at V = −10 V, I = −1 μA.
r_f = ΔV / ΔI = (0.8 − 0.7) / (20 − 10) mA
= 0.1 V / 10 mA = 0.1 / 10×10⁻³ = 10 Ω
r_r = V / I = −10 V / (−1 μA)
= 10 / 10⁻⁶ = 10⁷ Ω = 10 MΩ
Forward: 10 Ω. Reverse: 10 MΩ. That's a ratio of 1,000,000:1 — that's how the diode acts as a one-way valve.
Q5. In a half-wave rectifier, input frequency is 50 Hz. What is the output frequency? What if full-wave?
One pulse per input cycle → output frequency = input frequency = 50 Hz
Two pulses per input cycle → output frequency = 2 × 50 = 100 Hz
Higher output frequency means less ripple → easier to filter → smoother DC. Full-wave is preferred in most applications.
Q6. A silicon diode is forward biased at 0.75 V. If I₀ (saturation current) = 10⁻⁸ A and kT/e = 0.026 V at 300 K, find the current.
I = I₀ (e^(eV/kT) − 1)
eV/kT = 0.75 / 0.026 ≈ 28.85
e^28.85 ≈ 3.4 × 10¹²
I ≈ I₀ × e^(eV/kT) = 10⁻⁸ × 3.4×10¹² = 3.4×10⁴ A
Whoa — that's huge! But real diodes have series resistance that limits current. Still, it shows exponential sensitivity: 0.75 V gives vastly more current than 0.70 V.
This is why the diode's forward voltage is nearly "constant" at ~0.7 V — small voltage changes cause huge current changes. Real diodes have parasitic resistances that limit actual current.
Q7. Distinguish between intrinsic and extrinsic semiconductors.
Pure semiconductor (Si or Ge). No impurities. n_e = n_h = n_i. Conductivity is low (only from thermal generation). Depends strongly on temperature. Behaves as insulator at 0 K.
Doped with small amount of impurity (pentavalent or trivalent). Conductivity greatly increased. Majority carriers (electrons for n-type, holes for p-type) dominate. Conductivity depends on doping level. Much more useful for devices.
| n_e, n_h | Electron, hole concentration (m⁻³) |
| n_i | Intrinsic carrier concentration |
| N_D, N_A | Donor, acceptor concentration |
| V₀ | Barrier potential |
| V_th | Threshold (cut-in) voltage |
| I₀ | Reverse saturation current |
| kT/e | Thermal voltage (≈0.026 V at 300 K) |
| E_g | Energy band gap |
| r_d | Dynamic resistance |
| ρ, σ | Resistivity, conductivity |
(A) Both true, R explains A. (B) Both true, R does NOT explain A. (C) A true, R false. (D) A false, R true.