Inside a dielectric material, the electric field is modified by material polarization. The displacement field D = ε₀κE is continuous across boundaries (without free surface charges), while E is discontinuous. The bound charge density relates to polarization by ρ_bound = -∇·P.
Work through boundary conditions at dielectric interfaces. Apply Gauss's law in integral and differential forms for both D and E.
When you place a dielectric material in an electric field, the field doesn't simply pass through undisturbed. The material responds: its molecules, which may be polar or may become induced dipoles, align partially with the external field. This alignment is called polarization, denoted P, and it represents the average dipole moment per unit volume. The consequence is a weakening of the total electric field inside the material — this is why you learned that the capacitance of a parallel-plate capacitor increases by the factor κ (the dielectric constant) when you insert a dielectric. The internal field is reduced to E_inside = E_vacuum/κ.
The physical origin of this field reduction is the appearance of bound charges at the surfaces and within the bulk of the dielectric. When the molecular dipoles align, the positive ends of one dipole are adjacent to the negative ends of the next, and the interior charges cancel — but at the surfaces, charges are left exposed. These surface bound charges produce a field opposing the external field, reducing the net field inside. The volume bound charge density satisfies ρ_bound = −∇·P: wherever the polarization is non-uniform, bound charges pile up in the bulk.
To handle dielectrics cleanly, physicists introduce the displacement field D = ε₀E + P = ε₀κE (in a linear isotropic medium). The beauty of D is that its divergence depends only on free charges: ∇·D = ρ_free. Gauss's law in terms of D, ∮D·dA = Q_free_enclosed, has the same form as the vacuum version but with D replacing ε₀E. This is powerful: when designing capacitors or waveguides, you often know the free charges but not the bound charges, so working with D sidesteps the complication.
The distinction between D and E becomes critical at boundaries between materials. At an interface with no free surface charge, the normal component of D is continuous: D₁ₙ = D₂ₙ. But the normal component of E is discontinuous by a factor of κ₁/κ₂. Meanwhile, the tangential component of E is always continuous (from ∇×E = 0 in electrostatics), but the tangential component of D is discontinuous. These asymmetric boundary conditions — E tangential continuous, D normal continuous — govern how fields refract at dielectric boundaries, determining the direction of field lines as they cross from one material into another.