Home / Courses / GATE Physics / Practice / Dielectrics and capacitors

Dielectrics and capacitors

Electricity and magnetism · 40 questions and templates for GATE Physics, in order from warm-up to top rank. Check each answer, then read the full solution.

Loading questions… (they need JavaScript; nothing you type is sent anywhere)

Written questions with solutions

The same questions as a list, so you can print them or read without the practice box. Templates (fresh numbers) appear only in the practice box above.

  1. MCQStandard

    A dielectric sphere has polarisation P→=krrˆ (k a constant). The bound volume charge density inside it is
    1. −3k
    2. 3k
    3. −k
    4. −2k/r
    Answer and solution

    Answer: A

    ρb=−∇⋅P→=−1r2ddr(r2⋅kr)=−1r2(3kr2)=−3k, uniform. The surface gets σb=P(R)=kR; total bound charge −3k⋅43πR3+kR⋅4πR2=0, as it must be.

    Why people get this wrong: Dropping the minus sign in ρb=−∇⋅P→, or using the Cartesian d/dr without the r2.

  2. MSQStandard

    At the flat boundary between two linear dielectrics with no free charge on it, which quantities are continuous?
    1. The component of E→ along the boundary
    2. The component of D→ normal to the boundary
    3. The component of E→ normal to the boundary
    4. The component of D→ along the boundary
    Answer and solution

    Answer: A, B

    ∇×E→=0 (a thin rectangular loop) makes E∥ continuous. ∇⋅D→=ρf=0 (a thin pillbox) makes D⟂ continuous. Then E⟂=D⟂/ϵ jumps by ϵ1/ϵ2, and D∥=ϵE∥ jumps too.

    Why people get this wrong: Bound charge sits on the boundary, so E⟂ jumps even with no free charge.

  3. MCQExam level

    Field lines pass from a dielectric with ϵr1=2 into one with ϵr2=6 across a flat boundary carrying no free charge. In the first medium the field makes 30∘ with the normal. In the second it makes
    1. 10.9∘
    2. 30∘
    3. 45∘
    4. 60∘
    Answer and solution

    Answer: D

    E∥ continuous and ϵE⟂ continuous give tan⁡θ2tan⁡θ1=ϵ2ϵ1. So tan⁡θ2=3tan⁡30∘=3 and θ2=60∘. Lines bend away from the normal entering the higher-permittivity medium.

    Why people get this wrong: Using Snell's law with sines, or putting the ratio upside down (giving 10.9∘).

  4. MSQExam level

    A charged capacitor is isolated, and a dielectric slab (of the same thickness as the gap) is partly inserted between its plates, edge first. Neglect friction. Which statements are correct?
    1. The slab is pulled further in.
    2. The force on the slab is the same at every insertion depth.
    3. The stored energy decreases as the slab moves in.
    4. The voltage between the plates decreases as the slab moves in.
    Answer and solution

    Answer: A, C, D

    With Q fixed, U=Q2/(2C) and C(x) grows linearly with the inserted length x. The force F=−dU/dx=Q22C2dCdx is positive (inward) but falls as C grows, so it depends on x. U and V=Q/C both fall as the slab goes in.

    Why people get this wrong: With a battery connected the force is independent of x, since then F=12V2dC/dx with V fixed.

  5. NATExam level

    A parallel-plate capacitor has square plates 10 cm wide and a gap of 2.0 mm, and stays connected to a 200 V supply. A slab of dielectric constant 3.0 that just fits the gap is partly inserted. The force pulling it in is ______ μN (to one decimal place).
    Answer and solution

    Answer: 17.7 $\mu$N (±0.2)

    With inserted length x and plate width w: C=ϵ0wd[κx+(L−x)], so dCdx=ϵ0w(κ−1)d. At fixed V, F=12V2dCdx=ϵ0(κ−1)wV22d=(8.854×10−12)(2)(0.10)(4×104)4×10−3=1.77×10−5 N.

    Why people get this wrong: Using κ instead of κ−1: only the change in capacitance pulls.

  6. MCQExam level

    A capacitor connected to a battery has a dielectric slab pushed into it, raising its stored energy by ΔU. The work done by the battery in this process is
    1. ΔU
    2. zero
    3. ΔU/2
    4. 2ΔU
    Answer and solution

    Answer: D

    The battery moves extra charge ΔQ through voltage V: Wb=VΔQ=V2ΔC. The stored energy rises by 12V2ΔC=ΔU. So Wb=2ΔU; the other half is the work done on the slab as it is pulled in (or heat, if it slams in).

    Why people get this wrong: Assuming all the battery's work ends up stored.

  7. NATExam level

    A solid sphere of linear dielectric with ϵr=4 is placed in a uniform field of 300 V/m. The field inside the sphere is ______ V/m.
    Answer and solution

    Answer: 150 V/m

    Solving Laplace's equation inside and outside with the boundary conditions gives a uniform field inside: Ein=3ϵr+2E0=36(300)=150 V/m, along E0.

    Why people get this wrong: Using E0/ϵr=75 V/m, which is the slab result, not the sphere.

  8. MCQExam level

    The Clausius-Mossotti relation links the relative permittivity of a dense, non-polar medium to the polarisability α of its molecules (number density N). It reads
    1. ϵr−1=Nα/ϵ0
    2. ϵr−1ϵr+2=Nα3ϵ0
    3. ϵr+2ϵr−1=Nα3ϵ0
    4. ϵr=1+3Nα/ϵ0
    Answer and solution

    Answer: B

    Each molecule feels the local field E+P/(3ϵ0), not just E. With P=NαEloc and P=ϵ0(ϵr−1)E, eliminating Eloc gives ϵr−1ϵr+2=Nα3ϵ0. For dilute gases (ϵr≈1) it reduces to the first option.

    Why people get this wrong: The first option is only the dilute-gas limit; it ignores the field of neighbouring dipoles.

  9. MSQExam level

    Which of these statements about the electric displacement D→ are correct?
    1. Its flux through any closed surface equals the free charge enclosed.
    2. Its curl is always zero, so it is the gradient of a potential.
    3. In a region of linear dielectrics, its lines begin and end only on free charges.
    4. Its full vector is continuous across any boundary between dielectrics.
    Answer and solution

    Answer: A, C

    ∮D→⋅da→=Qf,enc, so D→ lines start and end only on free charge. But ∇×D→=∇×P→, which need not vanish (for example at the edges of a polarised bar), so D→ is not generally a gradient. At a boundary only D⟂ is continuous (with no free surface charge); D∥ jumps.

    Why people get this wrong: Treating D→ as "just like E→ but with free charge". Its curl is not zero in general, so you cannot use Coulomb's law with free charges to get D→.

  10. NATExam level

    A capacitor's plates carry free charge density ±6.0μC/m2 and the gap is filled by a dielectric of constant 3.0. The magnitude of the bound surface charge density on the face of the dielectric next to a plate is ______ μC/m2.
    Answer and solution

    Answer: 4 $\mu$C/m$^2$ (±0.05)

    D=σf, E=σf/(κϵ0), and P=D−ϵ0E=σf(1−1/κ)=6.0×23=4.0μC/m2. The bound charge on the face next to the positive plate is −4.0μC/m2, leaving a net 2.0=σf/κ to make the field.

    Why people get this wrong: Taking σb=σf/κ (that is the net charge seen by the field, not the bound charge).

  11. MCQExam level

    An atom's polarisability α is defined by p→=αE→. The quantity α/(4πε0) has the dimensions of
    1. area
    2. volume
    3. length
    4. nothing: it is a pure number
    Answer and solution

    Answer: B

    α=p/E has units C mV/m=C m2/V. 4πε0 has units C/(V m). Dividing: C m2/VC/(V m)=m3. This "polarisability volume" is of the order of the atom's size cubed (about 0.67×10−30 m³ for hydrogen).

    Why people get this wrong: Treating α itself as a volume; only after dividing by 4πε0 is it one.

  12. MSQExam level

    A short cylinder of electret has a permanent uniform polarisation P→ along its axis, and there is no free charge anywhere. Which statements are true?
    1. Its flat end faces carry bound charge ±P per unit area
    2. Its curved side carries bound charge P per unit area
    3. The bound volume charge density is zero
    4. At its centre the electric field points opposite to P→
    Answer and solution

    Answer: A, C, D

    Bound surface charge is σb=P→⋅nˆ: +P on the face P→ points out of, −P on the other, and zero on the curved side where nˆ⟂P→. Bound volume charge is ρb=−∇⋅P→=0 for uniform P→. The two charged faces act like a parallel-plate pair whose field runs from the + face to the − face, opposite to P→ (a "depolarising" field).

    Why people get this wrong: Expecting E→ to point along P→; inside a polarised body without free charge the field points back against it.

  13. NATHard

    The permittivity in the gap of a parallel-plate capacitor rises linearly from ϵ0 at one plate to 2ϵ0 at the other: ϵ(x)=ϵ0(1+x/d). The capacitance, in units of ϵ0A/d, is ______ (to three decimal places).
    Answer and solution

    Answer: 1.443 $\epsilon_0A/d$ (±0.005)

    The gap is a stack of thin layers in series: 1C=∫0ddxϵ(x)A=dϵ0A∫01du1+u=dln⁡2ϵ0A. So C=1ln⁡2ϵ0Ad=1.443ϵ0A/d.

    Why people get this wrong: Using the average permittivity 1.5ϵ0. Series layers average 1/ϵ, not ϵ.

  14. MCQHard

    A sphere is uniformly polarised with polarisation P→ (a permanent electret, no external field). The field inside the sphere is
    1. −P→ϵ0
    2. zero
    3. −P→3ϵ0
    4. +P→3ϵ0
    Answer and solution

    Answer: C

    The bound charge is only on the surface, σb=Pcos⁡θ. A sphere with σ=σ0cos⁡θ has a uniform field −σ0/(3ϵ0) along the axis inside (and a dipole field outside). So E→in=−P→/(3ϵ0), opposite to P→: the depolarising field.

    Why people get this wrong: −P→/ϵ0 is the field inside a thin slab polarised across its thickness; the shape matters.

  15. NATHard

    A capacitor has capacitance 5.0μF with air in the gap. With a slab of constant 4.0 filling the gap, it is charged to 120 V and disconnected. The slab is then pulled out completely. The work done by the person pulling it is ______ J (to three decimal places).
    Answer and solution

    Answer: 0.432 J (±0.004)

    With the slab, Ci=20μF and Q=20×10−6×120=2.4 mC, which stays fixed. Ui=Q22Ci=0.144 J, Uf=Q22Cf=0.576 J. No battery and no heat (slowly), so the work done is Uf−Ui=0.432 J. The voltage rises to 480 V.

    Why people get this wrong: Using 12CV2 with V=120 V both times, as if the battery were still connected (that gives a negative answer).

  16. MSQHard

    A sphere of linear dielectric (ϵr>1) is placed in a uniform field E→0. Which statements are correct?
    1. The field inside is uniform.
    2. The field inside is weaker than E0.
    3. The polarisation is P→=3ϵ0ϵr−1ϵr+2E→0.
    4. The sphere acquires a net bound charge.
    Answer and solution

    Answer: A, B, C

    E→in=3ϵr+2E→0: uniform and smaller than E0 for ϵr>1. Then P→=ϵ0(ϵr−1)E→in=3ϵ0ϵr−1ϵr+2E→0. Bound charges always total zero.

    Why people get this wrong: As ϵr→∞ the result tends to the conducting sphere: Ein→0 and P→3ϵ0E0, matching σ=3ϵ0E0cos⁡θ.

  17. NATHard

    A point charge of +10 nC is embedded at the centre of a very large block of linear dielectric with ϵr=5. The total bound charge gathered on the surface of the small cavity round the charge is ______ nC.
    Answer and solution

    Answer: -8 nC (±0.05)

    Gauss for D→: D=q/(4πr2), so E=q/(4πϵ0ϵrr2): the charge looks like q/ϵr=2 nC. The difference comes from bound charge right next to it: qb=q/ϵr−q=−q(1−1/ϵr)=−8 nC. (Inside a uniform linear dielectric ρb=0 away from free charge, and the matching +8 nC sits on the far outer surface of the block.)

    Why people get this wrong: Answering −2 nC or +2 nC: 2 nC is the screened charge the field sees, not the bound charge.

  18. NATHard

    A capacitor of any shape is filled with a weakly conducting material of relative permittivity 4.0 and conductivity 1.0×10−10 S/m. Once charged and isolated, its charge leaks away with time constant ______ s (to three decimal places).
    Answer and solution

    Answer: 0.354 s (±0.004)

    For any shape: J=σE and Gauss's law for D→ (∮D→⋅da→=Q, with D=ϵE) give the leakage current from the positive electrode I=σ∮E→⋅da→=σQ/ϵ. So dQ/dt=−(σ/ϵ)Q and τ=RC=ϵ/σ. τ=4.0×8.854×10−121.0×10−10=0.354 s.

    Why people get this wrong: Trying to find R and C separately: the geometry cancels completely.

  19. MCQHard

    A capacitor with free charge ±σ on its plates has two dielectric layers: κ1=2 next to the positive plate and κ2=4 next to the negative plate. The bound surface charge at the interface between the two layers is
    1. +σ/4
    2. −σ/4
    3. zero, since the interface has no free charge
    4. −σ/2
    Answer and solution

    Answer: B

    D=σ throughout (pointing from + to −). P1=σ(1−1/2)=σ/2, P2=σ(1−1/4)=3σ/4. At the interface, layer 1's face (outward normal along D) carries +P1 and layer 2's face (outward normal against D) carries −P2. Total σ/2−3σ/4=−σ/4. Check with the field jump: ϵ0(E2−E1)=σ(1/4−1/2)=−σ/4.

    Why people get this wrong: No free charge does not mean no charge: the bound charges of the two layers do not cancel when the constants differ.

  20. MCQHard

    A coaxial cable has a fixed outer conductor radius b, and its insulation breaks down when the field anywhere reaches Emax⁡. To make the largest possible voltage safe, the inner conductor radius a should be
    1. b/2
    2. b/e
    3. b/e2
    4. as small as possible
    Answer and solution

    Answer: B

    Between the conductors E=Vsln⁡(b/a), largest at the inner surface: E(a)=Valn⁡(b/a). Setting this equal to Emax⁡: V=Emax⁡aln⁡(b/a). Then dVda=Emax⁡[ln⁡ba−1]=0 gives ln⁡(b/a)=1, a=b/e, and Vmax⁡=Emax⁡b/e.

    Why people get this wrong: A thinner inner wire is not safer: the field at its surface grows like 1/a, so very thin wires break down at low voltage.

  21. MSQHard

    A linear dielectric whose permittivity ϵ(r→) varies smoothly from point to point contains no free charge, and a static field is set up in it. Which statements are correct?
    1. ∇⋅E→=0 everywhere in it.
    2. ∇⋅D→=0 everywhere in it.
    3. The bound volume charge density is zero everywhere.
    4. ∇×E→=0 in it.
    Answer and solution

    Answer: B, D

    B: Gauss for D→ counts only free charge. D: any static field is curl-free. But E→=D→/ϵ, so ∇⋅E→=D→⋅∇(1/ϵ)≠0 wherever D→ has a component along ∇ϵ (A false). Since ϵ0∇⋅E→=ρb here, there is bound volume charge (C false).

    Why people get this wrong: Carrying over "no free charge means no bound volume charge", which is true only for a uniform dielectric.

  22. NATHard

    A coaxial cable has inner radius 1.0 mm and outer radius 4.0 mm. From 1.0 to 2.0 mm the space is filled with a dielectric of constant 4, and from 2.0 to 4.0 mm with one of constant 2. Its capacitance per unit length is ______ pF/m (to one decimal place).
    Answer and solution

    Answer: 107 pF/m (±0.6)

    With charge λ per length, D=λ/(2πs) in both layers, and E=D/(κϵ0). Voltage: V=λ2πϵ0[ln⁡24+ln⁡22]=λ2πϵ0(0.520). So C/L=2πϵ00.520=5.563×10−110.520=1.070×10−10 F/m =107.0 pF/m. (The layers act as two cylindrical capacitors in series.)

    Why people get this wrong: Averaging the constants to 3 and using ln⁡4 (gives 120 pF/m); layers in series add their ln⁡(…)/κ terms.

  23. NATHard

    Model an atom as a point nucleus +e inside a uniform spherical electron cloud of radius 0.50×10−10 m and charge −e. In a field of 1.0×106 V/m, the nucleus shifts from the centre of the cloud by about ______ fm (three significant figures; 1 fm =10−15 m).
    Answer and solution

    Answer: 0.0868 fm (±0.0009)

    Inside a uniform ball of charge, the field at distance d from the centre is ed4πε0a3 (Gauss's law). Equilibrium with the applied field: ed4πε0a3=E, so d=4πε0a3Ee=1.113×10−10×1.25×10−31×1061.602×10−19=8.68×10−17 m =0.0868 fm, about a millionth of the atom's size. The model gives α=4πε0a3.

    Why people get this wrong: Using the field of a point charge, e/(4πε0d2); inside the cloud only the charge within radius d pulls back.

  24. MCQHard

    Water has a static relative permittivity of about 80, yet its refractive index for visible light is 1.33 (so εr=n2≈1.8 at those frequencies). The best explanation is
    1. light is too short in wavelength to enter water molecules
    2. the refractive index and permittivity are unrelated quantities
    3. water has no electronic polarisation
    4. the permanent molecular dipoles cannot turn fast enough to follow a field oscillating at optical frequencies, so only electronic polarisation remains
    Answer and solution

    Answer: D

    The static value includes orientational polarisation: water's permanent dipoles line up with a slow field. Rotating a molecule takes about 10−11 s, far longer than one optical cycle (∼10−15 s), so at optical frequencies that contribution vanishes. Only the fast electronic response is left, giving εr=n2≈1.8. This is dielectric dispersion.

    Why people get this wrong: Expecting n=εr to use the static value; εr depends on frequency.

  25. NATHard

    A spherical capacitor has radii 5.0 cm and 10 cm. The lower half of the space between them (a hemispherical shell) is filled with a dielectric of constant 3.0; the upper half is air. Its capacitance is ______ pF (one decimal place).
    Answer and solution

    Answer: 22.3 pF (±0.2)

    The field stays radial: tangential E→ is continuous across the flat boundary between the halves, and a radial field Ar2 in both halves fits that (with different free charge densities on each half). So the two halves are capacitors in parallel, each with half the solid angle: C=C02+κC02=1+κ2C0. With C0=4πε0abb−a=11.1 pF, C=2×11.1=22.3 pF.

    Why people get this wrong: Treating the halves as being in series; they share the same potential difference, so they are in parallel.

  26. NATHard

    In a dense non-polar medium, Nα3ε0=0.25. Using the Clausius-Mossotti relation, its relative permittivity is ______ (two decimal places).
    Answer and solution

    Answer: 2 (±0.01)

    Clausius-Mossotti: εr−1εr+2=Nα3ε0=x, so εr=1+2x1−x=1.50.75=2.00. The local field E→+P→/3ε0 felt by each molecule is larger than the average field, which raises εr above the dilute-gas estimate 1+Nα/ε0=1.75.

    Why people get this wrong: Using the dilute formula εr=1+3x (1.75), which ignores the local-field correction.

  27. MCQHard

    For free charges placed in a linear dielectric, the work needed to assemble them is given by
    1. 12∫D→⋅E→dτ
    2. ε02∫E2dτ
    3. 12∫ρbVdτ
    4. ∫P→⋅E→dτ
    Answer and solution

    Answer: A

    Bringing in free charge δρf costs ∫Vδρfdτ=∫V∇⋅(δD→)dτ=∫E→⋅δD→dτ. For a linear medium E→⋅δD→=12δ(D→⋅E→), so W=12∫D→⋅E→dτ. This includes the energy stored in stretching the molecular "springs"; ε02∫E2 counts only the field and would be the work to assemble all charges, bound ones included.

    Why people get this wrong: Using the vacuum formula; it leaves out the energy that goes into polarising the material.

  28. MSQHard

    A gas of polar molecules (permanent dipole moment p0, polarisability α, number density N) is studied at different temperatures T. Which statements are true?
    1. The orientational part of its susceptibility falls roughly as 1/T
    2. The electronic (induced) part hardly depends on temperature
    3. Its susceptibility is about Nε0(α+p023kBT)
    4. Non-polar molecules cannot be polarised by a field
    Answer and solution

    Answer: A, B, C

    Thermal agitation fights the lining up of permanent dipoles; for weak fields the average dipole along E→ is p02E3kBT (Langevin-Debye), so that part falls as 1/T. The induced moment αE comes from distorting the electron clouds and barely depends on T. Adding the two gives the Debye formula. Every atom or molecule, polar or not, has an induced moment.

    Why people get this wrong: Thinking non-polar means non-polarisable; non-polar only means no permanent moment.

  29. NATTop rank

    Two vertical parallel plates 2.0 mm apart dip into an oil of dielectric constant 3.0 and density 900 kg/m3, and are held at 5.0 kV. Taking g=9.8 m/s2, the oil between the plates rises above the outside level by ______ mm (to two decimal places).
    Answer and solution

    Answer: 6.27 mm (±0.07)

    The upward electric force per unit width is F/w=ϵ0(κ−1)V22d (the slab force at fixed V). It holds up a column of mass ρghd per unit width. So h=ϵ0(κ−1)V22ρgd2=(8.854×10−12)(2)(2.5×107)2(900)(9.8)(4×10−6)=6.27×10−3 m.

    Why people get this wrong: Using κ in place of κ−1, or forgetting that the column's weight per width involves d once and the force divides by d once more.

  30. MSQTop rank

    A parallel-plate capacitor (gap d) is kept at voltage V by a battery, and a dielectric of constant κ fills the gap. Compared with the empty capacitor at the same V, which statements are correct?
    1. The free charge on the plates is κ times larger.
    2. The field in the gap is still V/d.
    3. The free surface charge density is κϵ0V/d.
    4. The bound surface charge density on the dielectric is κϵ0V/d.
    Answer and solution

    Answer: A, B, C

    With V fixed, E=V/d is unchanged. Then D=κϵ0V/d=σf, κ times the empty value. P=D−ϵ0E=(κ−1)ϵ0V/d, which is the bound density magnitude, not κϵ0V/d.

    Why people get this wrong: The bound charge is always smaller than the free charge by exactly ϵ0E; that leftover is what makes the field.

← All topics · Timed mock test →