Great — let’s build the Refraction at a Spherical Surface formula from first principles. The goal is to understand why the relation
vn2−un1=Rn2−n1
holds, where:
- n1 = refractive index of the first medium (where the object lies)
- n2 = refractive index of the second medium (where the image lies)
- u = object distance from the pole (sign convention: negative for real object)
- v = image distance from the pole (sign convention: positive for real image on the opposite side)
- R = radius of curvature of the spherical surface (positive if centre of curvature is on the image side)
1. The core idea: Snell’s law at a curved interface
At any point on the spherical surface, the incident ray and refracted ray obey Snell’s law:
n1sini=n2sinr
For small angles (paraxial approximation — rays close to the principal axis), sinθ≈θ (in radians). So:
n1i=n2r
This linearisation is the key that lets us turn geometry into algebra.
2. Geometry of a single ray
Consider a point object O on the principal axis. A ray from O strikes the spherical surface at point P (height h above the axis). Let:
- C = centre of curvature of the spherical surface
- M = pole of the surface (vertex)
- I = image point formed after refraction
Draw the normal at P — it passes through C (since the surface is spherical). The angles:
- i = angle between incident ray OP and the normal PC
- r = angle between refracted ray PI and the normal PC
3. Relating angles to distances (paraxial approximation)
Because h is small compared to u, v, and R:
- Angle between OP and the axis: α≈uh (with sign)
- Angle between PC (normal) and the axis: θ≈Rh
- Angle between PI and the axis: β≈vh
Now, from the geometry of the triangle formed by the ray, the normal, and the axis:
- Incident angle i = angle between OP and the normal = θ−α (if θ>α)
- Refracted angle r = angle between PI and the normal = θ−β
Check the sign convention carefully — the exact relation depends on whether the ray bends toward or away from the normal. For a convex surface (centre on the image side), the standard result is:
i=α+θandr=θ−β
But the difference that matters is:
i−r=α+β
4. Applying Snell’s law
From n1i=n2r, we can write:
n1i=n2(i−(i−r))or directly:
n1i=n2r⟹n1i−n2r=0
But it’s more useful to express r in terms of i and the geometry:
r=i−(α+β)
Substitute into Snell’s law:
n1i=n2[i−(α+β)]
Simplify:
n1i=n2i−n2(α+β)
(n1−n2)i=−n2(α+β)
Now, i≈α+θ (from geometry). For small angles, α≈h/u, β≈h/v, θ≈h/R.
5. Substituting the small-angle approximations
Let’s do it step by step:
(n1−n2)(α+θ)=−n2(α+β)
Replace α, β, θ:
(n1−n2)(uh+Rh)=−n2(uh+vh)
Cancel h (non-zero):
(n1−n2)(u1+R1)=−n2(u1+v1)
6. Rearranging to the standard form
Expand the left side: …