Questions: Boundary Conditions at Conducting and Dielectric Interfaces

Question 1

At a dielectric interface with no free surface charge, which statement correctly describes the boundary conditions for the electric field?

• A. Both the normal and tangential components of E are continuous across the interface
• B. The tangential component of E is continuous; the normal component of E can be discontinuous if ε changes
• C. The normal component of E is continuous; the tangential component can be discontinuous
• D. Both components are discontinuous — field vectors always change at a material boundary

Answer: (B)

At a dielectric interface with no free surface charge: (1) tangential E is always continuous — from Faraday's law applied to a thin loop at the boundary; (2) normal D is continuous (from Gauss's law with σ_free = 0), and since D = ε₀εE, if ε changes across the boundary, the normal component of E must be discontinuous to keep normal D continuous. The common misconception is that 'no surface charge' means all components are continuous — but it only guarantees continuity of normal D, not normal E.

Question 2

You solve for the electric field in two regions separated by a flat interface and find valid solutions in each region. Why aren't you done?

• A. You are done — if each solution satisfies Maxwell's equations in its region, the combined solution is automatically physical
• B. You must apply boundary conditions at the interface, which select the unique physical solution from infinitely many mathematically valid ones
• C. You must average the two solutions at the boundary to get the correct field there
• D. You must discard the solution in the lower-permittivity region

Answer: (B)

Maxwell's equations in each region are satisfied by many different field configurations — differential equations alone do not uniquely determine the solution. Boundary conditions are the matching conditions that stitch the two-region solutions together physically. Without them, you have freedom in the integration constants (or separation-of-variables coefficients) in each region that must be fixed by demanding the fields match correctly at the interface. Boundary conditions make the problem uniquely solvable; they are not an optional verification step.

Question 3

At a perfect conductor surface, the tangential component of E must be zero because any nonzero tangential E would drive an infinite current along the surface.

• T. True
• F. False

Answer: True

This is the physical reasoning behind the boundary condition. A perfect conductor has zero resistance, so by Ohm's law (J = σE), any nonzero tangential E would drive an infinite surface current — which is unphysical. Free charges in the conductor redistribute instantly to cancel any tangential E. The result is tangential E = 0 at a perfect conductor surface. This is why the tangential E condition at a conductor is more restrictive than at a general dielectric interface.

Question 4

Boundary conditions are separate postulates that should be added to Maxwell's equations — they contain physical information that Maxwell's equations alone do not capture.

• T. True
• F. False

Answer: False

Boundary conditions are *derived from* Maxwell's equations — they are what Maxwell's equations say in the limiting case where the integration region is a thin pillbox or loop at an interface. Applying Gauss's law to an infinitesimally thin pillbox gives the normal component conditions; applying Faraday's and Ampère's laws to a thin rectangular loop gives the tangential conditions. They introduce no independent physical postulates — they are Maxwell's equations applied at boundaries.

Question 5

Explain the 'pillbox derivation' and what physical quantity it determines at a boundary.

Model answer: The pillbox derivation applies Gauss's law to a thin cylindrical 'pillbox' straddling the interface, with flat faces on either side and a vanishingly thin curved side. As the height shrinks to zero, flux through the curved side vanishes, leaving only contributions from the two flat faces. The result is a relation between the *normal components* of the field on each side: (D₂ − D₁)·n̂ = σ_free for electric displacement, and (B₂ − B₁)·n̂ = 0 for the magnetic field (since there are no magnetic monopoles).

The companion derivation uses a thin rectangular loop shrunk to zero height for the tangential components. The vanishing short sides leave only the two long sides parallel to the interface. Faraday's law gives continuous tangential E; Ampère's law gives the discontinuity in tangential H equal to surface current K. Together, the pillbox and loop derivations cover all four boundary conditions and show they are geometrically natural consequences of Maxwell's integral laws — not separate postulates.