GATE Electronics (EC): Electronic Devices (energy bands, carrier concentration, drift and diffusion, P-N junction, BJT, MOS capacitor, MOSFET, LED and solar cell), previous-year questions solved, part 3
Part 3 of 3 of the Electronic Devices: MOSFET, PN Junction hub: 8 more official GATE EC questions on MOS capacitor and MOSFET, each solved step by step and checked against the official answer key. Part 1 covers energy bands, carrier concentration and Fermi level, carrier transport: drift, diffusion, generation and recombination and p-N junction, Zener, BJT and optoelectronic devices.
Questions from official GATE EC papers (IIT Kharagpur 2022, IIT Kanpur 2023, IISc Bengaluru 2024, IIT Roorkee 2025, IIT Guwahati 2026); solutions by Skill-Lync. · Updated 17 Nov 2026 · 8 of 30 questions on this page
pMOS (W/L = 5) and nMOS (W/L = 1), gates tied to output, 4 V supply: is V_o equal to, above or below 2 V?
- (A)equal to
- (B)more than
- (C)less than
- (D)equal to
Answer: (C) Official key: C
- Each gate is tied to its own drain (the output node), so in both devices and each one is in saturation: is always true for .
- Steady state: the same current flows through both. With (the same cancels): and in the same units.
- , so .
- , which is less than 2 V: option (C).
- Check: of the pMOS and of the nMOS , both above 1 V, so both are on and saturated.
Trap: assuming a symmetric divider gives exactly 2 V, option (D). That is true only if . Here the nMOS is stronger (its is 300 against 200), so it pulls the node lower than the middle. Forgetting the mobilities (using only of 1 and 5) gives 2.38 V, option (B). We solved the current balance by bisection with full triode and saturation logic and got 1.899 V.
A diode-connected MOSFET, gate tied to drain, is the basic reference element of a current mirror and the load in many biasing cells. Its voltage follows the square-law equation, so SPICE model parameters for threshold voltage and mobility set the bias point directly.
nMOS and pMOS pass transistors (Vt = 1 V and -1 V) charging 1 uF capacitors: steady-state V1 and V2?
- (A)
- (B)
- (C)
- (D)
Answer: (C) Official key: C
- nMOS: the capacitor node is the source (the lower potential terminal). The channel stays on only while . Charging stops when , so . An nMOS passes a weak 1: it loses one threshold.
- pMOS: the 5 V terminal is the source (the higher potential terminal) and the capacitor node is the drain. and this never drops, because the source is held at 5 V. The channel stays on until , so . A pMOS passes a strong 1.
- So and : option (C).
Trap: assuming both devices pass the full 5 V (option A), or swapping the roles (option B). We simulated both capacitors charging with the square-law model: stopped at 4.000 V and reached 5.000 V.
This threshold drop is why a single nMOS pass transistor degrades a logic high, and why transmission gates in standard cells and multiplexers use an nMOS and pMOS in parallel. SRAM bit-cell designers also work around it, because an nMOS access transistor cannot pull a node above V_DD minus V_T.
Ideal p-type MOS capacitor: Q_IN = 2.2 uC/cm^2 at V_G = 2 V, C_OX = 1.7 uF/cm^2: Q_IN at V_G = 4 V?
Answer: Official key: 5.5 to 5.7
- In strong inversion the depletion charge has stopped growing, so any extra gate voltage goes into inversion charge across the oxide: .
- Use the first point to find the threshold voltage: , so . This is below 2 V, so is indeed strong inversion.
- At : .
- Same result without : rises by , so .
Trap: scaling in proportion to the gate voltage, . The charge is proportional to , not to . We computed both and checked that only lands in the official range 5.5 to 5.7.
Inversion charge C_ox times (V_GS minus V_T) is the starting point of every MOSFET current model, and a capacitance-voltage sweep of a MOS capacitor is how a fab extracts oxide capacitance and threshold voltage on test structures.
Long-channel nMOS, 100 um by 10 um, mu_n C_OX = 1 mA/V^2, Vt = 1 V, VGS = 2 - sin 2t, VDS = 1 V: peak I_D?
- (A)
- (B)
- (C)
- (D)
Answer: (C) Official key: C
- The current is largest when is largest. peaks at .
- Overdrive . This is more than , so the device is in the triode region, not saturation.
- Triode: with .
- : option (C).
Trap: using the saturation formula , option (B), without checking against the overdrive. Forgetting the gives 40 mA, option (A). Taking (so ) gives 5 mA, option (D). We scanned a full period of with region logic: the maximum was 15 mA, in the triode region.
Checking the region before choosing the current equation is routine in device modelling. A transistor used as a switch or resistor with a small drain voltage sits in the triode region, which is why on-resistance, set by the overdrive voltage, is the figure quoted on power-MOSFET datasheets.
MOS capacitor: which plot shows depletion width W_dep against gate voltage V_g (V_fb, V_t marked)?
- (A)
- (B)
- (C)
- (D)
Answer: (B) Official key: B
- Below the semiconductor surface is in accumulation: majority carriers pile up and there is no depletion region, so .
- Between and the surface is depleted. A larger gate voltage uncovers more ionised acceptors, so grows. It follows , so rises steadily from zero and the curve flattens as it climbs, like a square root.
- At strong inversion starts. The inversion layer now supplies the extra charge, so stops growing and stays at its maximum .
- That is plot (B): zero up to , a rise that flattens, a constant level from onward.
Trap: plot (A) levels off well before , with a vertical start. (C) and (D) fall again, as if depletion shrank with more positive gate voltage. We computed for a p-type substrate with a 10 nm oxide: it was zero at , increasing and flattening, and reached exactly at . At 60% of the way from to it was at 73% of , so it had not yet saturated. We then read the sampled curve of each plot: only (B) passes all four checks.
The maximum depletion width sets the ultimate depth of the channel depletion layer, which is why substrate doping is chosen to trade threshold voltage against short-channel control. Capacitance-voltage measurements on test structures read this width back as the dip and flat floor of the C-V curve.
MOS capacitor on p-type silicon, 100 nm oxide, VFB = 0 V: find the silicon work function
Answer: Official key: 4.10 to 4.50
- Oxide capacitance per area: .
- Flat-band voltage: , with .
- Oxide-charge term: .
- gives , so .
- .
Trap: the sign of the charge term. A positive fixed oxide charge shifts negative, so must be positive to bring it back to 0 V. Adding the term instead gives 4.88 eV, outside the official range. The code recomputed , and and confirmed the wrong-sign value falls outside the key.
Flat-band voltage is measured on capacitor test structures at fab to monitor oxide charge and gate work function, and it feeds the threshold voltage in the SPICE model. High-k metal-gate processes tune the metal work function precisely for this reason.
NMOS in linear region gives 5 uA at VDS = 0.1 V: transconductance after VDS goes to 1.5 V
Answer: Official key: 52.40 to 52.60
- In the linear region with .
- At : , so .
- Now . The device is no longer in the linear region. It is in saturation.
- In saturation and .
- .
Trap: using the triode formula because the question starts in the linear region. The new exceeds , so the device has moved to saturation. The code compared with and confirmed the saturation value lies inside the key and the triode value does not.
Compact models such as BSIM switch between the triode and saturation expressions at and keep the current and its derivatives continuous across it. A wrong region in a hand model gives the wrong , which is why transistor sizing for a given gain starts by checking against .
n-channel MOSFET with gate tied to drain and a 1 kOhm load, VTH = 1 V: find the gate voltage VG
Assume , , and and neglect channel length modulation effects.
The gate voltage () of the n-channel MOSFET (in Volt) is __.
(rounded off to two decimal places)
Answer: Official key: 2.50 to 2.60
- The gate is tied to the drain, so . Then always holds, and the device is in saturation.
- Saturation square law: .
- The same current flows through the 1 kΩ resistor: .
- Equate, with in mA and in volts: , so .
- , and , so the device is indeed on. (The other root, , is below threshold.) .
Trap: using and forgetting the factor of the square law. That gives , outside the key range. We solved the node equation both by bisection and by the quadratic formula and got the same .
A diode-connected MOSFET is the basic building block of a current mirror and of bias generators in analog and mixed-signal chips, and its for a given current is how threshold voltage and are extracted from measurements in SPICE model fitting.