Birefringence — ordinary and extraordinary rays

The key players

Quantity	Meaning
Optic axis	Direction with no birefringence; $n_o=n_e$ along it
$n_o$	Ordinary index (constant)
$n_e$	Principal extraordinary index (max deviation, when $\vec{E}$ ⟂ optic axis)
Positive crystal	$n_e>n_o$ (e.g. quartz)
Negative crystal	$n_e<n_o$ (e.g. calcite)

Figure — Birefringence — ordinary and extraordinary rays

DERIVATION 1 — Why the o-ray is isotropic but the e-ray is not

Take a uniaxial crystal with optic axis along $z$ . Anisotropy lives in the dielectric tensor: $\varepsilon = \varepsilon_0\begin{pmatrix} n_o^2 & 0 & 0\\ 0 & n_o^2 & 0\\ 0 & 0 & n_e^2\end{pmatrix}$

Why this matrix? "Uniaxial" means the $x$ and $y$ directions are equivalent (same $n_o$ ), and only $z$ (the optic axis) is special (index $n_e$ ).

A plane wave $\vec{E}\propto e^{i(\vec{k}\cdot\vec{r}-\omega t)}$ must satisfy the wave equation, which for a given propagation direction $\hat{k}$ gives the index ellipsoid / Fresnel equation. Solving it for $\hat{k}$ at angle $\theta$ to the optic axis yields two allowed waves:

Solution A (the o-ray): $\vec{E}$ lies in the $xy$ -plane (⟂ optic axis). It always sees index $n_o^2 \quad\Rightarrow\quad v_o=\frac{c}{n_o}\ \ \text{(direction-independent).}$ Why this step? If $\vec{E}$ has no $z$ -component, the $\varepsilon_{zz}=n_e^2$ entry never acts on it; only $n_o$ matters whatever the direction.

Solution B (the e-ray): $\vec{E}$ has a component along the optic axis. Working through the Fresnel equation gives the e-ray index formula (derived next).

DERIVATION 2 — The extraordinary index $n_e(\theta)$

We want $n_e(\theta)$ , the index seen by the e-wave whose wavevector makes angle $\theta$ with the optic axis.

Step 1 — Set up the index ellipsoid. The ellipsoid is $\frac{x^2+y^2}{n_o^2}+\frac{z^2}{n_e^2}=1.$ Why? It encodes how index depends on $\vec{D}$ orientation: $\vec D$ along $z$ → index $n_e$ ; in $xy$ → index $n_o$ .

Step 2 — Slice perpendicular to $\hat k$ . The two allowed indices are the semi-axes of the ellipse cut by the plane ⟂ $\hat k$ through the centre. One semi-axis is always $n_o$ (the o-ray). The other (e-ray) has index $n(\theta)$ found by geometry of the tilted ellipse.

Step 3 — Geometry. Let the e-ray's $\vec D$ make angle $\theta$ with the optic axis equivalently. Its index satisfies the ellipse relation: $\boxed{\ \frac{1}{n_e(\theta)^2}=\frac{\cos^2\theta}{n_o^2}+\frac{\sin^2\theta}{n_e^2}\ }$

Why this form? Resolve the unit displacement direction into a component $\cos\theta$ along the $n_o$ -direction and $\sin\theta$ along the $n_e$ -direction, then add their inverse-square contributions — exactly like combining capacitances/springs in "parallel of $1/n^2$ ".

WORKED EXAMPLES

Worked example Example 1 — e-ray index at

$\theta=45^\circ$ in calcite Calcite: $n_o=1.658$ , $n_e=1.486$ . Find $n_e(45^\circ)$ .

Step 1. $\dfrac{1}{n_e(\theta)^2}=\dfrac{\cos^245^\circ}{n_o^2}+\dfrac{\sin^245^\circ}{n_e^2}$ . Why? Direct plug-in of the derived formula. Step 2. $\cos^245^\circ=\sin^245^\circ=0.5$ . $=\dfrac{0.5}{2.749}+\dfrac{0.5}{2.208}=0.1819+0.2264=0.4083$ . Step 3. $n_e(45^\circ)=1/\sqrt{0.4083}=1.565$ . Why this matters: It lies between $n_e=1.486$ and $n_o=1.658$ , as it must — the e-index always interpolates.

Common mistakes (Steel-manned)

Recall Feynman: explain it to a 12-year-old

Imagine running through a crowd. If everyone is packed the same in all directions, you run at one speed (normal glass). But some crystals are like a crowd that's tightly packed left-right and loosely packed up-down. Now your speed depends on which way you face! Light is the same: in these special crystals it travels at two speeds depending on which way it "wiggles." So one light beam becomes two beams — one is the calm "ordinary" one that follows the normal rules, and the other is the rebel "extraordinary" one that bends in a weird way. The one direction where everyone's packed equally is the "optic axis," and light going that way stays a single beam.

Active-recall flashcards

#flashcards/physics

What defines a birefringent crystal?

It has two distinct refractive indices (

n_o

and

n_e

), so an unpolarized beam splits into ordinary and extraordinary rays.

What is the optic axis?

The direction along which

n_o=n_e

, so o- and e-rays travel at the same speed and do not split.

Which ray obeys Snell's law?

The ordinary (o) ray; the extraordinary ray does not.

How is the o-ray polarized?

\vec E

perpendicular to the principal plane (the plane containing the optic axis and the ray).

Formula for the extraordinary index at angle

\theta

\dfrac{1}{n_e(\theta)^2}=\dfrac{\cos^2\theta}{n_o^2}+\dfrac{\sin^2\theta}{n_e^2}

What does

n_e(\theta)

equal at

\theta=0

and

\theta=90^\circ

n_e(0)=n_o

(along optic axis);

n_e(90^\circ)=n_e

(principal value, max splitting).

Positive vs negative uniaxial crystal?

Positive:

n_e>n_o

(e.g. quartz). Negative:

n_e<n_o

(e.g. calcite).

Phase difference between o and e rays after thickness

d

\Delta\phi=\frac{2\pi}{\lambda}|n_o-n_e|d

Thickness of a quarter-wave plate?

d=\dfrac{\lambda}{4|n_o-n_e|}

(thinnest, for

\Delta\phi=\pi/2

Why does the o-ray have a constant index?

Its

\vec E

has no component along the optic axis, so the anisotropic

\varepsilon_{zz}=n_e^2

never acts — only

n_o

matters in every direction.

Birefringence vs dispersion — difference?

Birefringence splits by polarization (same color); dispersion splits by wavelength.

Connections

Polarization of light — birefringence is the mechanism behind polarizers and wave plates.
Snell's Law — the o-ray obeys it; the e-ray's violation defines "extraordinary."
Wave plates (quarter and half wave) — direct application of $\Delta\phi$ .
Dielectric tensor and anisotropic media — origin of two indices.
Polaroid and Nicol prism — devices using calcite birefringence.
Optical activity — related but distinct rotation of polarization.

Concept Map

Hinglish (regional understanding)

Intuition Hinglish mein samjho

Dekho, normal glass me light har direction me same speed se chalti hai kyunki uska structure har taraf ek jaisa hota hai — isliye sirf ek refractive index. Lekin calcite jaise kuch crystals "lopsided" hote hain: unka atomic structure kisi ek khaas direction (optic axis) me alag hota hai. Iska matlab light ka electric field jis taraf wiggle kar raha hai, uske hisaab se uska index badalta hai. Yahi hai birefringence — ek beam andar jaate hi do rays me split ho jaati hai.

In do rays ko hum ordinary (o-ray) aur extraordinary (e-ray) kehte hain. O-ray seedha Snell's law follow karta hai aur uska index hamesha constant $n_o$ rehta hai, kyunki uska $\vec E$ optic axis ke perpendicular hota hai — anisotropy use chhuti hi nahi. E-ray "rebel" hai: woh Snell's law todta hai aur uska index angle $\theta$ ke saath badalta hai, formula $\frac{1}{n_e(\theta)^2}=\frac{\cos^2\theta}{n_o^2}+\frac{\sin^2\theta}{n_e^2}$ se. Optic axis ke along ( $\theta=0$ ) dono ka index barabar ho jaata hai, toh wahan splitting nahi hoti.

Yeh important kyun hai? Kyunki yahi principle wave plates (quarter-wave, half-wave plates), polarizers aur Nicol prism ke peeche kaam karta hai. O aur e ray ke beech phase difference $\Delta\phi=\frac{2\pi}{\lambda}|n_o-n_e|d$ banta hai — yeh control karke hum linear polarization ko circular me convert kar sakte hain. Exam me yaad rakho: "O obeys, E is eccentric", aur calcite negative crystal hai ( $n_e<n_o$ ).

Birefringence — ordinary and extraordinary rays

WHAT is birefringence?

The key players

DERIVATION 1 — Why the o-ray is isotropic but the e-ray is not

DERIVATION 2 — The extraordinary index $n_e(\theta)$

WORKED EXAMPLES

Common mistakes (Steel-manned)

Active-recall flashcards

Connections

Concept Map

Hinglish (regional understanding)

Go deeper — visual, from zero

Test yourself — Optics

Connections