Refraction through glass block

Look through a rectangular glass block at an angle and the world beyond seems to jump sideways — a pencil pushed through the block appears to have a kink in it, even though the block's two faces are flat and parallel. The reason is refraction happening twice: once as light enters the glass, and once as it leaves. The two bends do not cancel each other's direction, but they do something almost as neat — the ray that comes out is parallel to the ray that went in, just shifted sideways. This shift is called lateral displacement, and it is one of the cleanest pieces of geometry in the ICSE Class 10 Light chapter: two applications of Snell's law, one parallel-line argument, and a right-triangle formula that tells you exactly how far the ray moves.

The core insight

Light bends towards the normal entering a glass slab and away from the normal leaving it, so the emergent ray ends up exactly parallel to the incident ray but laterally shifted, by an amount fixed by the slab's thickness, the angle of incidence, and the refractive index.

The core idea

Two refractions, one straight run. A ray in air meets the first face of the slab at angle of incidence i1i_1 and bends towards the normal (air is rarer, glass is denser), continuing inside the glass at angle of refraction r1r_1:

sin⁡i1sin⁡r1=μ\frac{\sin i_1}{\sin r_1} = \mu

where: μ\mu is the refractive index of the glass relative to air, i1i_1 is measured from the normal at the first face, and r1r_1 is measured from the same normal, on the glass side.

Why the second surface repeats the trick in reverse. Because the two faces of a rectangular slab are parallel, their normals are parallel too. So the ray travelling in a straight line inside the glass strikes the second face at exactly the same angle r1r_1 it left the first face with (angles between parallel lines and a common transversal are equal). At the second face the ray goes from glass to air — denser to rarer — so it bends away from the normal, emerging at angle of emergence ee. Snell's law applied at this surface, worked backwards, forces e=i1e = i_1: the emergent ray leaves at precisely the angle the original ray arrived at, just at a different point on a different, parallel face.

Parallel but displaced. Since the incident and emergent rays make the same angle with parallel normals, they are themselves parallel lines — but not the same line. The perpendicular gap between them is the lateral displacement, and it grows with the slab's thickness tt, with i1i_1, and with μ\mu; it vanishes only when i1=0i_1 = 0, since a ray straight along the normal never bends at all.

Figure
Ray diagram of light entering a rectangular glass slab bending towards the normal then leaving the far face bending away from the normal, with the emergent ray parallel to the incident ray but shifted sideways by the lateral displacement
Refraction through a rectangular glass slab glass, refractive index mu normal 1 normal 2 incident ray, angle i1 inside glass, angle r1 emergent ray, angle e = i1 original direction extended lateral displacement
Derivation ALateral displacement of light through a glass slab

The goal is to find a single formula for the perpendicular sideways shift dd between the incident ray's original direction and the emergent ray, in terms of the slab's thickness tt, the angle of incidence i1i_1, and the angle of refraction r1r_1 at the first surface, using nothing beyond one right triangle drawn inside the ray path itself.

Step 1. Isolate the straight run inside the glass. The ray travels in a straight line from the point where it enters the first face to the point where it strikes the second face, at angle r1r_1 to the normal. If the slab has thickness tt (the perpendicular distance between the two faces), this straight run inside the glass has length tcos⁡r1\dfrac{t}{\cos r_1}, because tt is the "adjacent" side of the right triangle formed by the ray and the normal, with the ray as the hypotenuse.

Step 2. Locate the sideways gap between the two parallel rays. At the point where the ray leaves the slab, drop a perpendicular from that exit point onto the extended line of the original incident ray. The angle between the actual path inside the glass and the incident ray's original direction is i1−r1i_1 - r_1 (the amount by which the ray bent towards the normal at entry, which is exactly how far it now sits off the original line, since the exit path emerges parallel to the entry direction).

Step 3. Read off the lateral displacement from the right triangle formed by the glass-path length and this angle. The perpendicular distance dd from the extended incident-ray line to the exit point is the "opposite" side of a right triangle whose hypotenuse is the glass-path length from Step 1 and whose angle is i1−r1i_1 - r_1 from Step 2:

d=tcos⁡r1 sin⁡(i1−r1)d = \frac{t}{\cos r_1}\, \sin(i_1 - r_1)

Step 4. Read the physics out of the formula. dd grows with tt (a thicker slab gives the ray more distance to drift sideways), and it grows with the gap i1−r1i_1 - r_1 (a bigger bend at entry pushes the ray farther off its original line). Because r1r_1 itself depends on μ\mu through Snell's law, a denser slab (bigger μ\mu) also increases dd for a fixed i1i_1. When i1=0i_1 = 0, r1=0r_1 = 0 too, so d=0d = 0: a ray straight along the normal passes through undeviated and undisplaced.

Worked example

A ray strikes a glass slab of refractive index 1.51.5 and thickness 6 cm6\ \text{cm} at an angle of incidence of 50°50°. Find the lateral displacement of the emergent ray.

Step 1. Find r1r_1 using Snell's law at the first surface, exactly as before: sin⁡r1=sin⁡50°1.5=0.7661.5=0.511\sin r_1 = \dfrac{\sin 50°}{1.5} = \dfrac{0.766}{1.5} = 0.511, so r1=sin⁡−1(0.511)≈30.7°r_1 = \sin^{-1}(0.511) \approx 30.7°.

Step 2. Compute the bending angle used in the displacement formula: i1−r1=50°−30.7°=19.3°i_1 - r_1 = 50° - 30.7° = 19.3°, so sin⁡(i1−r1)=sin⁡19.3°≈0.331\sin(i_1 - r_1) = \sin 19.3° \approx 0.331.

Step 3. Compute cos⁡r1\cos r_1: cos⁡30.7°≈0.860\cos 30.7° \approx 0.860.

Step 4. Substitute into d=tcos⁡r1sin⁡(i1−r1)d = \dfrac{t}{\cos r_1}\sin(i_1 - r_1): d=60.860×0.331=6.98×0.331≈2.31 cmd = \dfrac{6}{0.860} \times 0.331 = 6.98 \times 0.331 \approx 2.31\ \text{cm}. The lateral displacement is about 2.31 cm2.31\ \text{cm}, and the emergent ray leaves the far face at e=50°e = 50°, parallel to the original incident ray, just shifted sideways by this amount.

Competitive stretch

If the same slab is doubled in thickness to 12 cm12\ \text{cm} while i1i_1 and μ\mu stay the same, r1r_1 is unchanged (it depends only on i1i_1 and μ\mu, not tt), so the displacement simply scales with tt: d=120.860×0.331≈4.62 cmd = \dfrac{12}{0.860}\times 0.331 \approx 4.62\ \text{cm} — exactly double, confirming that dd is directly proportional to thickness for fixed i1i_1 and μ\mu.

Common traps

  • Thinking the emergent ray changes direction, not just position — it is parallel to the incident ray (e=i1e = i_1 always, for parallel faces), only laterally shifted.
  • Using tt itself as the length of the path inside the glass instead of t/cos⁡r1t/\cos r_1 — the ray travels along the slanted path, not straight across the thickness.
  • Plugging i1−r1i_1 - r_1 into the displacement formula in degrees while leaving cos⁡r1\cos r_1 or sin⁡(i1−r1)\sin(i_1-r_1) computed with a calculator in radian mode, giving a wrong numeric answer despite the correct formula.
  • Assuming displacement depends on thickness alone — it depends on tt, i1i_1, and μ\mu together (through r1r_1), and is zero at normal incidence regardless of thickness.

Ask the tutor: Want to try the displacement formula with a different thickness or angle of incidence, or see how it changes for a slab of a different refractive index?

Why does the displacement formula have cos r1 in the denominator instead of just multiplying by t

Because the ray does not travel straight across the thickness tt — it travels along the slanted path inside the glass, which is the hypotenuse of a right triangle whose short side is tt and whose angle from the normal is r1r_1. That slanted path length is t/cos⁡r1t/\cos r_1, always a bit longer than tt itself. The displacement is then found by resolving that slanted path against the direction i1−r1i_1 - r_1, which is where the sin⁡(i1−r1)\sin(i_1-r_1) factor comes from. Skipping the cos⁡r1\cos r_1 step is the single most common arithmetic slip in this topic.

Does the refractive index alone tell me how big the lateral displacement will be

Not by itself — μ\mu only fixes r1r_1 for a given i1i_1 through Snell's law; the actual displacement d=(t/cos⁡r1)sin⁡(i1−r1)d = (t/\cos r_1)\sin(i_1-r_1) also needs the thickness tt and the angle of incidence i1i_1. A very high μ\mu with i1i_1 close to 0°0° still gives almost no displacement, because r1r_1 stays close to 0°0° too and the ray barely bends. You always need all three quantities together, not μ\mu in isolation.

Is the lateral displacement formula the same as what happens with a prism

No — a prism has two faces that are not parallel to each other, so the angle of emergence does not equal the angle of incidence, and the ray does not come out parallel to how it went in; instead a prism produces an overall angular deviation, plus dispersion of colours. The neat result e=i1e = i_1 and the simple sideways-shift formula both rely on the slab's two faces being parallel, which is only true for a rectangular block, not a prism.

Before this

Where this leads