Ray Optics – Advanced Study Notes
Prepared for an MIT‑level Physics course (Manipal MET – Physics)
Goal: Provide a self‑contained, rigorous treatment of geometric (ray) optics starting from first principles, integrating wave‑optical background, and extending to edge cases such as gradient‑index media, negative‑index materials, and perfect imaging.
Format: Markdown; each section contains definitions, derivations, examples, and key insights.
1. Foundations
1.1 Light as a Wave and as a Ray
| Wave view | Ray view |
|---|---|
| Maxwell’s equations → electromagnetic wave equations | High‑frequency limit (wavelength ≪ characteristic scale) |
| Wavefront: surface of constant phase | Ray: normal to wavefront, direction of energy flow (Poynting vector) |
| Eikonal equation: ( (\nabla \Phi)^2 = n^2(\mathbf{r}) ) | Fermat’s principle → path of stationary optical length |
Key idea: In the short‑wavelength limit, the wavefronts become locally planar and the field can be described by rays that propagate along normal directions.
1.2 Optical Path Length (OPL)
For a medium with spatially varying refractive index (n(\mathbf{r})),
[ \text{OPL} = \int_{\mathcal{C}} n(\mathbf{r}), ds, ]
where ( \mathcal{C} ) is a curve connecting two points.
Fermat’s principle: the actual ray path makes the OPL stationary (δOPL = 0).
2. Reflection and Refraction
2.1 Derivation of Snell’s Law from Fermat
Consider two homogeneous media with indices (n_1) and (n_2), separated by a planar interface.
Let a ray strike the interface at point (P) from medium 1 at an angle (\theta_1) (to the normal) and refract into medium 2 at (\theta_2).
Take a reference point (O) on the interface. The total optical path from (O) to (P) and then to a point (Q) in medium 2:
[ \text{OPL} = n_1, a + n_2, b, ]
with (a = |OP|), (b = |PQ|).
Let (x) be the coordinate along the interface. Using right triangles:
[ a = \sqrt{x^2 + (h_1)^2}, \qquad b = \sqrt{(L-x)^2 + (h_2)^2}, ]
where (h_1, h_2) are perpendicular distances from (O) to the two media.
Stationarity of OPL with respect to (x) yields:
[ \frac{n_1 x}{\sqrt{x^2 + h_1^2}} = \frac{n_2 (L-x)}{\sqrt{(L-x)^2 + h_2^2}}. ]
Taking ( \sin \theta_1 = x / \sqrt{x^2 + h_1^2}) and similarly for (\theta_2), we obtain
[ n_1 \sin \theta_1 = n_2 \sin \theta_2, ]
which is Snell’s law.
2.2 Reflection
From the same variational principle, the extremum occurs when the derivative of OPL with respect to the point of reflection is zero. The result is the familiar law:
[ \theta_{\text{inc}} = \theta_{\text{refl}}. ]
2.3 Edge Cases
| Situation | Effect | Key Insight |
|---|---|---|
| Total Internal Reflection (TIR) | No refracted ray if ( \theta_1 > \theta_c ). | ( \theta_c = \arcsin(n_2/n_1) ) when (n_1>n_2). |
| Evanescent Wave | Exponential decay beyond interface in TIR | Sensitive to surface perturbations (frustrated TIR). |
| Negative‑Index Media | Snell’s law with (n<0) → refraction on same side of normal | Leads to reversed phase velocity; “left‑handed” media. |
| Anisotropic Media | Snell’s law replaced by Fresnel equations with tensor (n_{ij}) | Wave normals differ from Poynting vectors (walk‑off). |
3. Wavefront Curvature and Ray Equations
3.1 The Eikonal Equation
Let the optical phase be ( \Phi(\mathbf{r}) ). In a lossless medium, the wave equation reduces to
[ (\nabla \Phi)^2 = n^2(\mathbf{r}). ]
Set ( \mathbf{s} = \nabla \Phi ) – the ray direction. Taking gradient of both sides: