Chapter 16 · Optics

16.2 Refraction, Total Internal Reflection, Lenses and Prisms

Light bends at a boundary according to Snell's law $n_1\sin i=n_2\sin r$. From that one law come apparent depth, total internal reflection and optical fibres, the image formed by a curved surface, the lens maker's and thin-lens formulas, the power of a lens and of lenses in contact, and the deviation of light by a prism.

A swimming pool always looks shallower than it is, and a straw in a glass of water looks broken at the surface. A diamond flashes with light from inside, a hot road seems to shine with water that is not there, and a hair-thin glass fibre carries thousands of telephone calls under the ocean. Your spectacles, the camera in your phone and the eye itself focus light with curved transparent surfaces. All of these are consequences of refraction — the change in direction of light as it passes from one medium into another.

Refraction happens because light travels at different speeds in different materials. In this section we state Snell's law and use it for plane surfaces (real and apparent depth, glass slabs), for total internal reflection and its applications, for a single curved surface and then for thin lenses — the lens maker's formula, the thin-lens formula, magnification, power and lenses in contact — and finally for prisms. We use the Cartesian sign convention of Section 16.1 throughout; microscopes and telescopes are built from these pieces in Section 16.3.

Refraction at a plane surface

📘 Definition 16.5 — Refractive index

The (absolute) refractive index of a medium is the ratio of the speed of light in vacuum to its speed $v$ in the medium:

$$n=\frac{c}{v}.$$

It is a pure number, $n\ge1$ ($n=1$ for vacuum, $1.0003$ for air — taken as $1$). The relative refractive index of medium 2 with respect to medium 1 is $n_{21}=\dfrac{n_2}{n_1}=\dfrac{v_1}{v_2}$. Of two media, the one with the larger $n$ is called optically denser. Typical values: water $1.33\ (\approx\tfrac43)$, crown glass $1.50$–$1.52$, dense flint glass $1.6$–$1.7$, diamond $2.42$. The value depends slightly on the colour (wavelength) of the light.

📐 Law 16.6 — Laws of refraction (Snell's law)
  1. The incident ray, the refracted ray and the normal at the point of incidence lie in one plane.
  2. For light passing from medium 1 into medium 2, with angle of incidence $i$ and angle of refraction $r$ (both from the normal),
$$n_1\sin i=n_2\sin r .$$

Light entering a denser medium bends towards the normal ($r\lt i$); entering a rarer medium, it bends away from it. A ray along the normal is not bent. The path of light is reversible. On crossing a boundary the frequency stays the same (it is fixed by the source); the speed becomes $c/n$ and the wavelength $\lambda=\lambda_0/n$, where $\lambda_0$ is the wavelength in vacuum.

Through a parallel-sided slab. A ray entering a glass slab of thickness $t$ at angle $i$ refracts to $r$, crosses the slab, and at the far face is refracted back to $i$ (Snell's law at the two faces, with the same pair of media, in reverse). So the emergent ray is parallel to the incident ray but shifted sideways. The path inside has length $t/\cos r$ and makes angle $i-r$ with the original direction, so the lateral shift is

$$d=\frac{t\sin(i-r)}{\cos r}.$$

Real and apparent depth. A coin at depth $h$ under water is seen from above. A ray from the coin meets the surface a small distance $x$ from the point vertically above it, making angle $i$ with the normal in water, and leaves at angle $r$ in air. Traced back, the emerging ray seems to come from a point at depth $h'$ on the same vertical, with $x=h\tan i=h'\tan r$. For viewing nearly along the normal the angles are small, $\tan\approx\sin$, and Snell's law $n\sin i=\sin r$ gives $\dfrac{h'}{h}=\dfrac{\tan i}{\tan r}\approx\dfrac{\sin i}{\sin r}=\dfrac1n$. The coin appears raised.

📐 Result 16.7 — Apparent depth and normal shift

An object in a medium of refractive index $n_1$, at real depth $h$ below a plane surface, viewed nearly along the normal from a medium of index $n_2$, appears at depth

$$h'=h\,\frac{n_2}{n_1}\qquad(\text{object in a medium }n\text{, viewed from air: }h'=h/n).$$

For several parallel layers viewed from air, $h'=\sum\dfrac{h_k}{n_k}$. A slab of thickness $t$ and index $n$ placed between an object and the eye (in air) shifts the image towards the eye by $t\left(1-\dfrac1n\right)$, whatever the distance of the object.

The slab result is the basis of the JEE Main experiment on the refractive index of a glass slab with a travelling microscope: focus on a mark, then on the mark seen through the slab, then on the top of the slab; the two differences of the readings are the real and apparent thicknesses, and $n=\dfrac{\text{real}}{\text{apparent}}$.

✏️ Example 1 — Light in glass, and a coin under two liquids
(a) Light of vacuum wavelength $600\ \text{nm}$ enters glass of refractive index $1.50$. Find its speed, wavelength and frequency in the glass. (b) A ray strikes the glass from air at $60^\circ$ to the normal. Find the angle of refraction. (c) A beaker holds $20\ \text{cm}$ of water ($n=\tfrac43$) with a $9.0\ \text{cm}$ layer of oil ($n=1.5$) floating on it. Seen from directly above, how deep does a coin at the bottom appear, and by how much is it raised?
(a) $v=\dfrac{c}{n}=\dfrac{3.0\times10^8}{1.50}=2.0\times10^8\ \text{m/s}$; $\lambda=\dfrac{600}{1.50}=400\ \text{nm}$; frequency unchanged, $\nu=\dfrac{c}{\lambda_0}=\dfrac{3.0\times10^8}{600\times10^{-9}}=5.0\times10^{14}\ \text{Hz}$.
(b) $1\times\sin60^\circ=1.50\sin r$: $\sin r=\dfrac{0.866}{1.50}=0.577$, $r=35.3^\circ$ — bent towards the normal.
(c) $h'=\dfrac{9.0}{1.5}+\dfrac{20}{4/3}=6.0+15=21\ \text{cm}$. The real depth is $29\ \text{cm}$, so the coin appears raised by $8.0\ \text{cm}$.
Check Each layer is shortened by its own factor $1/n$; the order of the layers does not matter. The shift is $9.0\left(1-\tfrac{1}{1.5}\right)+20\left(1-\tfrac34\right)=3.0+5.0=8.0\ \text{cm}$ ✓.

Yellow light passes from air into glass of refractive index $1.5$. Which of these is unchanged?

Total internal reflection

When light passes from a denser into a rarer medium (glass to air, water to air), it bends away from the normal: $r\gt i$. As $i$ is increased, $r$ grows faster and reaches $90^\circ$ — the refracted ray skims along the surface — at a particular angle of incidence. For any larger $i$ Snell's law would need $\sin r\gt1$: there is no refracted ray at all, and all the light is reflected back into the denser medium. Some light is always reflected at a boundary, but only now is the reflection total — no polished mirror does better.

📘 Definition 16.6 — Critical angle and total internal reflection

For light going from a denser medium (index $n_1$) towards a rarer one ($n_2\lt n_1$), the critical angle $C$ is the angle of incidence in the denser medium for which the angle of refraction is $90^\circ$. If the angle of incidence exceeds $C$, the light is wholly reflected back into the denser medium: total internal reflection (TIR).

📐 Result 16.8 — Condition for total internal reflection

Total internal reflection occurs if and only if (1) light travels from an optically denser medium towards a rarer one, and (2) the angle of incidence exceeds the critical angle, given by

$$\sin C=\frac{n_2}{n_1}\qquad(\text{denser medium of index }n\text{ in air: }\sin C=\tfrac1n).$$

Glass ($n=1.5$) in air: $C=41.8^\circ$; water: $48.8^\circ$; diamond: $24.4^\circ$. Light going from a rarer into a denser medium is never totally reflected.

Applications.

✏️ Example 2 — Critical angles and a circle of light
(a) Find the critical angle for glass ($n=1.50$) in contact with air, and in contact with water ($n=\tfrac43$). (b) A small lamp is $2.0\ \text{m}$ below the surface of a still pool. Find the radius of the circle on the surface through which light can leave the water.
(a) Glass–air: $\sin C=\dfrac1{1.50}=0.667$, $C=41.8^\circ$. Glass–water: $\sin C=\dfrac{4/3}{1.50}=0.889$, $C=62.7^\circ$ — larger, because the two indices are closer.
(b) For water $\sin C=\tfrac34$, so $\cos C=\dfrac{\sqrt7}{4}$ and $\tan C=\dfrac{3}{\sqrt7}=1.134$. Rays meeting the surface at more than $C$ are totally reflected, so the edge of the circle is where $i=C$: $r=h\tan C=2.0\times1.134=2.27\ \text{m}$.
Check $r=\dfrac{h}{\sqrt{n^2-1}}=\dfrac{2.0}{\sqrt{16/9-1}}=\dfrac{2.0}{0.882}=2.27\ \text{m}$ ✓ — a formula worth remembering.
✏️ Example 3 — How an optical fibre traps light
An optical fibre has a core of refractive index $1.50$ and a cladding of index $1.40$, and its end face is in air. (a) Find the critical angle at the core–cladding boundary. (b) Find the largest angle to the fibre's axis at which light can enter the end face and still be guided along the fibre. (c) What happens to a ray entering at a larger angle?
Find the acceptance angle in (b) before you open the solution:
(a) $\sin C=\dfrac{1.40}{1.50}=0.933$, $C=69.0^\circ$.
(b) To meet the wall at $C$ the ray inside must make $r=90^\circ-69.0^\circ=21.0^\circ$ with the axis. At the end face, $\sin\theta_a=1.50\sin21.0^\circ=0.538$, so $\theta_a=32.6^\circ$. Equivalently $\sin\theta_a=\sqrt{1.50^2-1.40^2}=\sqrt{0.29}=0.539$ ✓.
(c) It is steeper inside, meets the wall at less than $C$, and is partly refracted into the cladding at every reflection — it leaks out within a short distance.
⏱ Shortcut $\sin\theta_a=\sqrt{n_1^2-n_2^2}$ (fibre in air) — the "numerical aperture". If it comes out bigger than 1, every ray that enters is guided.

In Interactive 1 choose glass → air ($n_1=1.50$) and set $i=40^\circ$. Then increase $i$ to $45^\circ$. What do you expect?

The readouts show $C=41.8^\circ$. At $i=40^\circ$, $\sin r=1.50\sin40^\circ=0.964$, so $r=74.6^\circ$ — the ray leaves almost along the surface and the reflected ray is already quite bright. At $45^\circ\gt C$ the readout says "none — total internal reflection": all the light goes into the reflected ray. ($26^\circ$ would be the answer for light going the other way, air into glass; $90^\circ$ happens only at exactly $i=C$.) On the graph, the curve of $r$ against $i$ shoots up to $90^\circ$ at $C$ and stops.
Interactive 1 — Refraction and Total Internal Reflection

Light travels upward from the lower medium into the upper one. Choose the pair of media and the angle of incidence $i$. The brightness of the reflected and refracted rays shows how the light is shared between them. The graph plots the angle of refraction $r$ against $i$; the shaded band is the region of total internal reflection. Press Play to sweep $i$ from $0^\circ$ to $90^\circ$.

30.0°
angle of incidence $i$—
angle of refraction $r$—
critical angle $C$—
what happens—

A ray inside a glass block ($n=1.5$) meets the glass–water boundary ($n=1.33$) at $55^\circ$. Which is correct?

Refraction at spherical surfaces and thin lenses

A lens has curved surfaces, so we first need the image formed by refraction at one spherical surface. Let a spherical surface of radius $R$ separate a medium of index $n_1$ (where the light comes from) from one of index $n_2$. A point object $O$ on the axis sends a ray to a point $N$ of the surface at height $h$; the normal at $N$ is the radius through the centre of curvature $C$, and the refracted ray crosses the axis at the image $I$. Call the angles that $ON$, $CN$ and $NI$ make with the axis $\alpha$, $\gamma$ and $\beta$. For a surface convex towards the object, the angle of incidence is $i=\alpha+\gamma$ and the angle of refraction $r=\gamma-\beta$ (exterior-angle theorem in the triangles $ONC$ and $NCI$). For paraxial rays all angles are small: $\alpha\approx\dfrac{h}{-u}$, $\beta\approx\dfrac{h}{v}$, $\gamma\approx\dfrac{h}{R}$, and Snell's law becomes $n_1i=n_2r$:

$$n_1\left(\frac{h}{-u}+\frac hR\right)=n_2\left(\frac hR-\frac hv\right)\quad\Longrightarrow\quad\frac{n_2}{v}-\frac{n_1}{u}=\frac{n_2-n_1}{R}.$$
📐 Law 16.9 — Refraction at a spherical surface

For paraxial rays passing from a medium of index $n_1$ into a medium of index $n_2$ across a spherical surface of radius of curvature $R$ (Cartesian convention: $R\gt0$ if the centre of curvature lies on the side the light goes to),

$$\frac{n_2}{v}-\frac{n_1}{u}=\frac{n_2-n_1}{R},\qquad m=\frac{n_1v}{n_2u}.$$

For a plane surface ($R\to\infty$) it gives $v=\dfrac{n_2}{n_1}u$ — the apparent-depth result 16.7.

✏️ Example 4 — A fish in a round bowl
A spherical fish bowl of radius $15\ \text{cm}$ is filled with water ($n=\tfrac43$); the thin glass wall can be ignored. A small fish is $5.0\ \text{cm}$ from the wall nearest to an observer, on the line from the observer to the centre. (a) Where does the fish appear to be? (b) Where does it appear when it swims to the centre? (c) Is it magnified?
Signs Light goes from the fish (in water, $n_1=\tfrac43$) to the observer (air, $n_2=1$); take that direction as positive, with the pole at the nearer wall. The centre of curvature is inside the bowl, on the side the light comes from: $R=-15\ \text{cm}$.
(a) $u=-5.0\ \text{cm}$: $\dfrac1v-\dfrac{4/3}{-5.0}=\dfrac{1-4/3}{-15}=\dfrac1{45}$, so $\dfrac1v=\dfrac1{45}-\dfrac{4}{15}=-\dfrac{11}{45}$ and $v=-4.09\ \text{cm}$: the fish appears $4.1\ \text{cm}$ inside the wall (virtual image), slightly nearer than it really is.
(b) $u=-15\ \text{cm}$: $\dfrac1v=\dfrac1{45}-\dfrac{4}{45}=-\dfrac1{15}$, $v=-15\ \text{cm}$ — it appears exactly at the centre. (Rays from the centre cross the surface along the normal and are not bent.)
(c) At the centre $m=\dfrac{n_1v}{n_2u}=\dfrac{(4/3)(-15)}{(1)(-15)}=\dfrac43$: it looks $1.33$ times larger. In (a) $m=\dfrac{(4/3)(-4.09)}{-5.0}=1.09$.

Thin lenses. A lens is a piece of transparent material bounded by two surfaces, at least one curved. A thin lens is one whose thickness is negligible compared with its radii of curvature and with the object and image distances, so all distances can be measured from one point.

📘 Definition 16.7 — Thin lens: optical centre, foci and power
  • The optical centre $O$ is the point of the lens on its principal axis through which a ray passes undeviated.
  • A lens thicker at the middle (biconvex, plano-convex, concavo-convex in air) is converging: parallel paraxial rays meet at the second principal focus $F_2$ behind it. A lens thinner at the middle is diverging: they appear to come from a focus $F_2$ in front of it. A lens has a focus on each side, at equal distances when the same medium surrounds it.
  • The focal length $f$ is positive for a converging lens and negative for a diverging one (Cartesian convention).
  • The power of a lens is $P=\dfrac1f$ with $f$ in metres; its unit is the dioptre ($1\ \text{D}=1\ \text{m}^{-1}$). Converging lenses have positive power.

The lens maker's formula. A thin lens of index $n_2$ in a medium of index $n_1$ has surfaces of radii $R_1$ (the first surface the light meets) and $R_2$. The first surface alone would form an image at $v_1$: $\dfrac{n_2}{v_1}-\dfrac{n_1}{u}=\dfrac{n_2-n_1}{R_1}$. That image is the object for the second surface (the lens is thin, so the distance is still $v_1$), which forms the final image: $\dfrac{n_1}{v}-\dfrac{n_2}{v_1}=\dfrac{n_1-n_2}{R_2}$. Adding the two equations eliminates $v_1$:

$$\frac{n_1}{v}-\frac{n_1}{u}=(n_2-n_1)\left(\frac1{R_1}-\frac1{R_2}\right).$$

For an object at infinity the image is at the focus, $v=f$; so the right-hand side divided by $n_1$ is $1/f$.

📐 Law 16.10 — Lens maker's formula

A thin lens of refractive index $n_2$ in a medium of index $n_1$, with radii of curvature $R_1$ and $R_2$ (Cartesian signs), has focal length

$$\frac1f=\left(\frac{n_2}{n_1}-1\right)\left(\frac1{R_1}-\frac1{R_2}\right).$$

(For a lens in air, $n_2/n_1=n$.) For an equiconvex lens in air ($R_1=R$, $R_2=-R$): $f=\dfrac{R}{2(n-1)}$, which is $f=R$ for $n=1.5$. For a plano-convex lens: $f=\dfrac{R}{n-1}$. If the lens is put in a medium denser than itself ($n_1\gt n_2$), $f$ changes sign: a convex glass lens in a denser liquid diverges.

📐 Law 16.11 — Thin-lens formula and magnification

For a thin lens and paraxial rays, with the Cartesian sign convention,

$$\frac1v-\frac1u=\frac1f,\qquad m=\frac{h'}{h}=\frac vu=\frac{f}{f+u}.$$

A real image (on the far side, $v\gt0$) of a real object is inverted ($m\lt0$). A converging lens forms a virtual, erect, magnified image only for an object inside its focus; a diverging lens always forms a virtual, erect, diminished image of a real object, between the lens and its focus.

To draw ray diagrams for lenses use: a ray parallel to the axis passes (or appears to pass) through $F_2$; a ray through the optical centre goes straight on; a ray through (or towards) the first focus $F_1$ emerges parallel to the axis. The images of a convex lens follow the same pattern as a concave mirror, with "beyond $C$" replaced by "beyond $2F$" — but real images now form on the far side of the lens.

📐 Result 16.12 — Thin lenses in contact

Thin lenses of focal lengths $f_1,f_2,\dots$ placed in contact on a common axis act as a single thin lens of focal length $F$ and power $P$, where

$$\frac1F=\frac1{f_1}+\frac1{f_2}+\cdots,\qquad P=P_1+P_2+\cdots$$

(powers with their signs). Proof for two lenses: the first lens forms an image at $v_1$ with $\dfrac1{v_1}-\dfrac1u=\dfrac1{f_1}$; this is the object for the second, $\dfrac1v-\dfrac1{v_1}=\dfrac1{f_2}$; adding gives $\dfrac1v-\dfrac1u=\dfrac1{f_1}+\dfrac1{f_2}$. For lenses separated along an axis, apply the lens formula lens by lens; the total magnification is $m=m_1m_2$.

🧭 Problem-solving strategy — Surfaces and lenses
  1. Draw the light travelling left to right and mark the pole (or optical centre). Give $u$ its sign (real object: $u\lt0$).
  2. For a single surface use $\dfrac{n_2}{v}-\dfrac{n_1}{u}=\dfrac{n_2-n_1}{R}$, with $n_1$ the medium the light comes from and $R\gt0$ if $C$ is on the far side.
  3. For a lens, find $f$ from the lens maker's formula if needed — $R_1$ is the first surface met, and both radii carry signs — then use $\dfrac1v-\dfrac1u=\dfrac1f$ and $m=\dfrac vu$.
  4. For lenses in contact add powers; for separated lenses or several surfaces, use the image of one as the object of the next (a virtual object has $u\gt0$).
  5. Check: $v\gt0$ means a real image on the far side; a diverging lens must give $0\lt m\lt1$.
✏️ Example 5 — Making a lens, dipping it, pairing it
A biconvex lens of glass ($n=1.50$) has radii of curvature $20\ \text{cm}$ and $30\ \text{cm}$. (a) Find its focal length and power in air. (b) Find its focal length when immersed in water ($n=\tfrac43$). (c) In air, it is placed in contact with a diverging lens of focal length $40\ \text{cm}$. Find the focal length and power of the combination, and the position and magnification of the image of an object $90\ \text{cm}$ in front of it.
Find the focal length in water (b) before you open the solution:
(a) Light from the left: the first surface bulges towards the light, its centre is on the far side, $R_1=+20\ \text{cm}$; the second has its centre on the near side, $R_2=-30\ \text{cm}$. $\dfrac1f=(1.50-1)\left(\dfrac1{20}+\dfrac1{30}\right)=0.50\times\dfrac{5}{60}=\dfrac1{24}$: $f=+24\ \text{cm}$, $P=\dfrac{1}{0.24}=+4.2\ \text{D}$.
(b) $\dfrac{n_2}{n_1}-1=\dfrac{1.50}{4/3}-1=0.125$: $\dfrac1f=0.125\times\dfrac1{12}=\dfrac1{96}$, $f=+96\ \text{cm}$ — four times longer; the lens is much weaker under water.
(c) $\dfrac1F=\dfrac1{24}+\dfrac1{-40}=\dfrac{5-3}{120}=\dfrac1{60}$: $F=+60\ \text{cm}$, $P=+1.7\ \text{D}$ (check: $4.17-2.50=1.67\ \text{D}$ ✓). With $u=-90\ \text{cm}$: $\dfrac1v=\dfrac1{60}+\dfrac1{-90}=\dfrac{1}{180}$, $v=+180\ \text{cm}$; $m=\dfrac{v}{u}=\dfrac{180}{-90}=-2$: a real, inverted image twice the size, $1.8\ \text{m}$ behind the lenses.
⏱ Shortcut For the same lens in two media, $\dfrac{f_\text{medium}}{f_\text{air}}=\dfrac{n-1}{n/n_\text{m}-1}=\dfrac{0.5}{0.125}=4$.

A thin convex lens made of glass ($n=1.5$) has focal length $f$ in air. When it is completely immersed in a transparent liquid of refractive index $1.6$, it

Refraction through a prism

📘 Definition 16.8 — Prism, angle of prism and angle of deviation

A prism is a block of transparent material with two plane refracting faces inclined to each other. The angle $A$ between the two faces is the angle of the prism (refracting angle); the line where they meet is the refracting edge. A ray passing through the prism is turned towards its base; the angle between the emergent ray and the original direction of the incident ray is the angle of deviation $\delta$.

A ray in a plane perpendicular to the edge meets the first face at angle of incidence $i$ and refracts to $r_1$; it meets the second face at $r_2$ and emerges at angle $e$. The normals at the two faces meet at an angle $180^\circ-A$ (the quadrilateral formed by the edge, the two points of incidence and the meeting point of the normals has two right angles). In the triangle formed by the two normals and the ray inside the prism, $r_1+r_2+(180^\circ-A)=180^\circ$. The ray is turned through $i-r_1$ at the first face and $e-r_2$ at the second, both towards the base. So:

$$r_1+r_2=A,\qquad \delta=(i-r_1)+(e-r_2)=i+e-A .$$

As $i$ is increased from small values, $\delta$ first decreases, reaches a minimum $\delta_m$ and then increases again. Because light paths are reversible, if a ray entering at $i$ leaves at $e$, a ray entering at $e$ leaves at $i$ with the same deviation; so a deviation above the minimum occurs for two angles of incidence (whenever both lie in the range in which light emerges). The minimum is the one case where these two coincide, $i=e$: the ray passes through the prism symmetrically, parallel to the base for an isosceles prism. Interactive 2 plots the full curve — the graph you plot in the JEE Main experiment on the $\delta$–$i$ curve of a prism.

📐 Result 16.13 — Minimum deviation

At minimum deviation $i=e$ and $r_1=r_2=\dfrac A2$, so $i=\dfrac{A+\delta_m}{2}$, and the refractive index of the prism (in air) is

$$n=\frac{\sin\left(\dfrac{A+\delta_m}{2}\right)}{\sin\left(\dfrac A2\right)} .$$

Light emerges from the second face only if $r_2\le C$, where $\sin C=1/n$. Since $r_1\le C$ as well, no ray at all can pass through a prism with $A\gt2C$. For $A\le2C$ the smallest angle of incidence that still gives an emergent ray (grazing emergence, $e=90^\circ$) satisfies $\sin i_\text{min}=n\sin(A-C)$.

📐 Result 16.14 — Thin prism

For a prism of small angle $A$ (a few degrees) and light incident at small angles, all the sines may be replaced by the angles: $i=nr_1$, $e=nr_2$, so

$$\delta=i+e-A=n(r_1+r_2)-A=(n-1)A ,$$

independent of the angle of incidence. Two thin prisms placed together with their edges opposite deviate light by $(n_1-1)A_1-(n_2-1)A_2$; choosing $(n_1-1)A_1=(n_2-1)A_2$ gives no net deviation.

🚀 Beyond JEE Main (JEE Advanced) — dispersion by a prism

Dispersion is not in the JEE Main 2026 syllabus list (it is listed for JEE Advanced 2026 as "deviation and dispersion of light by a prism"). This box is for interest only; nothing here is used in the exercises, the Exam corner or the quiz.

The refractive index of glass is slightly larger for violet light than for red (for crown glass about $1.532$ and $1.514$). A prism therefore deviates violet more than red, and white light spreads into a spectrum — Newton's famous experiment. For a thin prism the angular spread is $\delta_v-\delta_r=(n_v-n_r)A$, and the dispersive power of the glass is $\omega=\dfrac{n_v-n_r}{n_y-1}$, the spread divided by the mean deviation. Two thin prisms of different glasses can be combined to cancel the dispersion while keeping a net deviation (an achromatic combination), the principle behind the colour-corrected lenses of good cameras.

✏️ Example 6 — Measuring a prism
An equilateral glass prism ($A=60^\circ$) gives a minimum deviation of $30^\circ$. (a) Find its refractive index and the angle of incidence at minimum deviation. (b) A thin prism of angle $5.0^\circ$ is made of the same glass. Find the deviation it produces. (c) Find the smallest angle of incidence on the equilateral prism for which light still emerges from the second face.
Find the refractive index in (a) first:
(a) $n=\dfrac{\sin45^\circ}{\sin30^\circ}=\dfrac{0.7071}{0.5}=1.414\ (=\sqrt2)$. At minimum deviation $i=\dfrac{A+\delta_m}{2}=45^\circ$ and $r_1=r_2=30^\circ$.
(b) $\delta=(n-1)A=0.414\times5.0^\circ=2.07^\circ$.
(c) $\sin C=1/\sqrt2$, $C=45^\circ$. Grazing emergence needs $r_2=C=45^\circ$, so $r_1=60^\circ-45^\circ=15^\circ$ and $\sin i_\text{min}=\sqrt2\sin15^\circ=0.366$: $i_\text{min}=21.5^\circ$. For smaller $i$ the ray is totally reflected at the second face.
⏱ Shortcut For $A=60^\circ$: $\delta_m=30^\circ\Leftrightarrow n=\sqrt2$, and $\delta_m=60^\circ\Leftrightarrow n=\sqrt3$. These two pairs appear again and again in JEE Main.

In Interactive 2 set $A=60^\circ$, $n=1.50$ and $i=30^\circ$: the ray emerges at $e=77.1^\circ$ with deviation $47.1^\circ$. Predict the deviation when you set $i=77.1^\circ$.

At $i=77.1^\circ$ the readouts show $r_1=40.5^\circ$, $r_2=19.5^\circ$, $e=30.0^\circ$ and $\delta=47.1^\circ$ — the path of $i=30^\circ$ run backwards, so the deviation is the same. In between the graph dips to its minimum, $\delta_m=37.2^\circ$ at $i=48.6^\circ$, where $r_1=r_2=30^\circ$ and the ray inside runs parallel to the base. Below $i=27.9^\circ$ (with these settings) the light is totally reflected at the second face.
Interactive 2 — Prism: Deviation and the Minimum

A ray (red) crosses a prism of angle $A$ and refractive index $n$; the dashed line is its original direction, so the angle between it and the emergent ray is the deviation $\delta$. The graph shows $\delta$ against the angle of incidence $i$ for this prism, with the minimum marked. Press Play to sweep $i$ and watch the dot run along the curve.

60°1.5040.0°
$r_1$, $r_2$—
emergence $e$—
deviation $\delta$—
minimum $\delta_m$—
⚠️ JEE Main traps

1. Apparent depth depends on who looks from where. Object in water seen from air: $h/n$ (looks shallower). Object in air seen by a fish: $h\times n$ (looks further). Always $h'=h\,n_\text{observer}/n_\text{object}$.

2. TIR needs both conditions. Denser to rarer, and $i\gt C$ with $\sin C=n_\text{rarer}/n_\text{denser}$ — not $1/n$ when the outside medium is not air.

3. Lens signs differ from mirror signs. Lens: $\dfrac1v-\dfrac1u=\dfrac1f$, $m=+\dfrac vu$, convex $f\gt0$. Mirror: $\dfrac1v+\dfrac1u=\dfrac1f$, $m=-\dfrac vu$, concave $f\lt0$. In the lens maker's formula $R_1$ is the surface the light meets first, and an equiconvex lens has $R_2=-R_1$.

4. The medium matters. In the lens maker's formula use $n_\text{lens}/n_\text{medium}$. A lens can become weaker, or even change from converging to diverging, when immersed.

5. Prism formulas have conditions. $n=\sin\frac{A+\delta_m}{2}/\sin\frac A2$ only at minimum deviation; $\delta=(n-1)A$ only for a thin prism at small incidence. And power adds only for lenses in contact, with $f$ in metres.

For a ray passing through a prism at minimum deviation, which statement is correct?

Exercises

Take $c=3.0\times10^8\ \text{m/s}$, $n_\text{water}=\tfrac43$ (or $1.33$ where stated) and $n_\text{air}=1$. Use the Cartesian sign convention with the light travelling from left to right; objects are real unless stated otherwise.

I
Refraction at plane surfaces.
  • 1 Find the speed of light in water.
  • 2 Light of wavelength $589\ \text{nm}$ in air enters glass of refractive index $1.50$. Find its wavelength (in nm) and frequency in the glass.
  • 3 A ray in air strikes a water surface ($n=1.33$) at $45^\circ$ to the normal. Find the angle of refraction.
  • 4 Find the refractive index of glass ($n=1.50$) relative to water.
  • 5 A swimming pool is $2.4\ \text{m}$ deep. How deep does it look when viewed from directly above?
  • 6 A glass slab $4.5\ \text{cm}$ thick ($n=1.50$) is laid on a printed dot. By how much is the dot raised when viewed from above?
  • 7 A ray strikes a glass slab ($n=1.50$, thickness $6.0\ \text{cm}$) at $60^\circ$. Find the lateral shift of the emergent ray.
  • 8 (Experiment) A travelling microscope reads $2.150\ \text{cm}$ when focused on a mark on the table, $3.050\ \text{cm}$ when focused on the same mark through a glass slab placed over it, and $4.850\ \text{cm}$ when focused on the top surface of the slab. Find the refractive index of the glass.
1 $v=\dfrac{c}{n}=\dfrac{3.0\times10^8}{4/3}=2.25\times10^8\ \text{m/s}$.
2 $\lambda=\dfrac{589}{1.50}=393\ \text{nm}$; $\nu=\dfrac{3.0\times10^8}{589\times10^{-9}}=5.09\times10^{14}\ \text{Hz}$, the same as in air.
3 $\sin r=\dfrac{\sin45^\circ}{1.33}=0.532$, $r=32.1^\circ$.
4 $n_\text{gw}=\dfrac{1.50}{4/3}=1.125$.
5 $h'=\dfrac{2.4}{4/3}=1.8\ \text{m}$ — a real danger for people who dive in expecting shallow water, or jump in expecting to stand.
6 Shift $=t\left(1-\dfrac1n\right)=4.5\times\dfrac13=1.5\ \text{cm}$.
7 $\sin r=\dfrac{0.866}{1.50}$, $r=35.26^\circ$; $d=\dfrac{6.0\sin(24.74^\circ)}{\cos35.26^\circ}=\dfrac{6.0\times0.4185}{0.8165}=3.07\ \text{cm}$.
8 Real thickness $4.850-2.150=2.700\ \text{cm}$; apparent thickness $4.850-3.050=1.800\ \text{cm}$; $n=\dfrac{2.700}{1.800}=1.50$.
II
Total internal reflection.
  • 1 Find the critical angle for diamond ($n=2.42$) in air.
  • 2 Find the critical angle for flint glass ($n=1.60$) immersed in water ($n=1.33$).
  • 3 Inside a glass block ($n=1.50$) a ray meets the glass–air surface at (a) $40^\circ$, (b) $45^\circ$. What happens in each case?
  • 4 The critical angle of a medium in air is $37^\circ$. Find its refractive index (take $\sin37^\circ=0.6$).
  • 5 A point source is $80\ \text{cm}$ below the surface of water ($n=\tfrac43$). Find the radius of the circle on the surface through which light emerges.
  • 6 Find the least refractive index of a $45^\circ$–$90^\circ$–$45^\circ$ prism that turns a beam through $90^\circ$ by total internal reflection. Why are such prisms preferred to mirrors in binoculars?
  • 7 A fibre has core index $1.48$ and cladding index $1.46$. Find its acceptance angle in air.
  • 8 Explain, using total internal reflection, why a mirage is seen on a hot road and why a well-cut diamond sparkles.
1 $\sin C=\dfrac1{2.42}=0.413$, $C=24.4^\circ$.
2 $\sin C=\dfrac{1.33}{1.60}=0.831$, $C=56.2^\circ$.
3 $C=41.8^\circ$. (a) $40^\circ\lt C$: refracted, $\sin r=1.50\sin40^\circ=0.964$, $r=74.6^\circ$ (with a weaker reflected ray). (b) $45^\circ\gt C$: totally reflected.
4 $n=\dfrac{1}{\sin C}=\dfrac1{0.6}=1.67$.
5 $r=h\tan C=\dfrac{h}{\sqrt{n^2-1}}=\dfrac{0.80}{0.882}=0.907\ \text{m}$.
6 The light meets the long face at $45^\circ$, so we need $C\lt45^\circ$: $n\gt\dfrac1{\sin45^\circ}=\sqrt2=1.414$. TIR reflects all the light, gives no double images from a silvered back, and does not tarnish.
7 $\sin\theta_a=\sqrt{1.48^2-1.46^2}=\sqrt{0.0588}=0.242$, $\theta_a=14.0^\circ$.
8 Mirage: the air near a hot road is less dense and has a smaller $n$; light from the sky entering these layers at a shallow angle bends away from the normal until it is turned back up (TIR at grazing incidence), so the sky appears below the road like a pool of water. Diamond: its small critical angle ($24.4^\circ$) means that most light entering the top is totally reflected at the back facets and leaves only through the top, after several reflections, so the stone looks brilliant.
III
Spherical surfaces and thin lenses.
  • 1 An object is $30\ \text{cm}$ in front of a convex lens of focal length $20\ \text{cm}$. Find $v$ and $m$.
  • 2 An object is $30\ \text{cm}$ in front of a concave lens of focal length $15\ \text{cm}$. Find $v$ and $m$.
  • 3 Find the focal length of an equiconvex lens with radii $15\ \text{cm}$ made of glass of refractive index $1.60$.
  • 4 A plano-convex lens ($n=1.50$) has a curved surface of radius $10\ \text{cm}$. Find its focal length and power.
  • 5 A convex lens of focal length $25\ \text{cm}$ is in contact with a concave lens of focal length $40\ \text{cm}$. Find the power and focal length of the combination.
  • 6 A biconvex lens ($n=1.50$) has focal length $20\ \text{cm}$ in air. Find its focal length when immersed in a liquid of refractive index $1.60$.
  • 7 One end of a long glass rod ($n=1.50$) is ground to a convex hemispherical surface of radius $10\ \text{cm}$. An object in air is on the axis $50\ \text{cm}$ from this surface. Find the position of the image.
  • 8 A small air bubble is inside a glass sphere ($n=1.50$) of radius $6.0\ \text{cm}$, $2.0\ \text{cm}$ from the centre. Where does it appear when seen from the side of the sphere nearer to it?
  • 9 A convex lens forms a real image, twice the size of the object, on a screen $90\ \text{cm}$ from the object. Find the focal length.
  • 10 An equiconvex lens of focal length $10\ \text{cm}$ is cut into two identical halves by a plane perpendicular to its axis (through the middle of the lens). Find the focal length of each half.
1 $\dfrac1v=\dfrac1{20}+\dfrac1{-30}=\dfrac1{60}$, $v=+60\ \text{cm}$; $m=\dfrac{60}{-30}=-2$ (real, inverted, magnified).
2 $\dfrac1v=\dfrac1{-15}+\dfrac1{-30}=-\dfrac{3}{30}$, $v=-10\ \text{cm}$; $m=\dfrac{-10}{-30}=+\dfrac13$ (virtual, erect, diminished).
3 $\dfrac1f=0.60\left(\dfrac1{15}+\dfrac1{15}\right)=0.080$, $f=12.5\ \text{cm}$.
4 $R_1=+10$, $R_2=\infty$: $\dfrac1f=0.50\times\dfrac1{10}$, $f=20\ \text{cm}$; $P=+5.0\ \text{D}$.
5 $P=\dfrac1{0.25}-\dfrac1{0.40}=4.0-2.5=+1.5\ \text{D}$; $F=\dfrac1{1.5}\ \text{m}=66.7\ \text{cm}$ (converging).
6 In air $0.50\,K=\dfrac1{20}$, where $K=\dfrac1{R_1}-\dfrac1{R_2}=0.10\ \text{cm}^{-1}$. In the liquid $\dfrac1f=\left(\dfrac{1.50}{1.60}-1\right)(0.10)=-\dfrac1{160}$: $f=-160\ \text{cm}$, a weak diverging lens.
7 $n_1=1$, $n_2=1.50$, $u=-50$, $R=+10$: $\dfrac{1.50}{v}=\dfrac{0.50}{10}-\dfrac1{50}=0.030$, $v=+50\ \text{cm}$ — a real image inside the rod, $50\ \text{cm}$ from its end.
8 The bubble is $6.0-2.0=4.0\ \text{cm}$ from the nearer surface. $n_1=1.50$, $n_2=1$, $u=-4.0$, $R=-6.0$: $\dfrac1v=\dfrac{1-1.50}{-6.0}+\dfrac{1.50}{-4.0}=0.0833-0.375=-0.292$, $v=-3.43\ \text{cm}$: it appears $3.43\ \text{cm}$ inside the surface.
9 $m=-2$: $v=-2u$ and $v-u=90$, so $u=-30\ \text{cm}$, $v=+60\ \text{cm}$; $\dfrac1f=\dfrac1{60}+\dfrac1{30}=\dfrac1{20}$, $f=20\ \text{cm}$.
10 Each half is plano-convex with one surface of the original: $\dfrac1{f'}=(n-1)\dfrac1R$, half of $\dfrac1f=(n-1)\dfrac2R$. So $f'=2f=20\ \text{cm}$ (the two halves together, in contact, give back $10\ \text{cm}$).
IV
Prisms and harder problems.
  • 1 A prism of angle $60^\circ$ gives a minimum deviation of $40^\circ$. Find its refractive index.
  • 2 Find the angle of minimum deviation of a $60^\circ$ prism of refractive index $1.50$.
  • 3 Find the deviation produced by a thin prism of angle $6.0^\circ$ and refractive index $1.52$.
  • 4 Light falls normally on one face of a prism of angle $30^\circ$ and refractive index $1.50$. Find the angle of emergence and the deviation.
  • 5 What is the largest angle a prism of refractive index $1.50$ can have if any light is to pass through it?
  • 6 A thin prism of angle $4.0^\circ$ and refractive index $1.54$ is combined with a thin prism of refractive index $1.72$, edges opposite, to give no net deviation. Find the angle of the second prism.
  • 7 The plane face of a plano-convex lens ($n=1.50$, curved-face radius $20\ \text{cm}$) is silvered. Light entering the curved face passes through the lens, reflects and passes through again. Find the focal length of the equivalent mirror.
  • 8 Explain why, for a given prism, two different angles of incidence give the same deviation (except at minimum deviation).
1 $n=\dfrac{\sin50^\circ}{\sin30^\circ}=\dfrac{0.766}{0.5}=1.53$.
2 $\sin\dfrac{A+\delta_m}{2}=1.50\sin30^\circ=0.75$, so $\dfrac{60^\circ+\delta_m}{2}=48.6^\circ$ and $\delta_m=37.2^\circ$.
3 $\delta=(n-1)A=0.52\times6.0^\circ=3.12^\circ$.
4 $i=0$, so $r_1=0$ and $r_2=A=30^\circ$ (less than $C=41.8^\circ$). $\sin e=1.50\sin30^\circ=0.75$, $e=48.6^\circ$; $\delta=0+48.6^\circ-30^\circ=18.6^\circ$.
5 $r_1\le C$ and $r_2\le C$, so $A=r_1+r_2\le2C=2\times41.8^\circ=83.6^\circ$.
6 $(1.54-1)(4.0^\circ)=(1.72-1)A_2$, so $A_2=\dfrac{2.16^\circ}{0.72}=3.0^\circ$.
7 The lens alone has $f_L=\dfrac{R}{n-1}=40\ \text{cm}$. The light is refracted by the lens twice and reflected once by a plane mirror (power $0$), so the power of the system is $2P_L=\dfrac{2}{40\ \text{cm}}$: it acts as a concave mirror of focal length $20\ \text{cm}$.
8 Reversibility: a ray entering at $i$ and leaving at $e$ can be run backwards, entering at $e$ and leaving at $i$, with the same deviation $i+e-A$. The two angles coincide only for the symmetric path, $i=e$, which is the minimum.
🎯 Exam corner JEE Main · MCQJEE Main · Numerical

E1. JEE Main · MCQ A beaker contains $20\ \text{cm}$ of water ($n=\tfrac43$) with a $15\ \text{cm}$ layer of oil ($n=1.5$) floating on it. A coin at the bottom, viewed from directly above, appears raised by (A) $5\ \text{cm}$ (B) $10\ \text{cm}$ (C) $15\ \text{cm}$ (D) $25\ \text{cm}$

Apparent depth $h'=\dfrac{15}{1.5}+\dfrac{20}{4/3}=10+15=25\ \text{cm}$.
Shift Real depth $35\ \text{cm}$, so the coin is raised by $35-25=10\ \text{cm}$.
Answer (B).
⏱ Shortcut Add the shifts layer by layer: $15\left(1-\tfrac{2}{3}\right)+20\left(1-\tfrac34\right)=5+5=10\ \text{cm}$. (D) is the apparent depth, not the shift.

E2. JEE Main · Numerical A thin equiconvex lens of glass ($n=1.5$) has a focal length of $20\ \text{cm}$ in air. It is immersed in a liquid of refractive index $1.25$. Find its new focal length in cm. Give the answer to the nearest integer.

Ratio Only $\left(\dfrac{n}{n_\text{m}}-1\right)$ changes: $0.5$ in air, $\dfrac{1.5}{1.25}-1=0.2$ in the liquid.
Focal length $f_\text{liquid}=20\times\dfrac{0.5}{0.2}=50\ \text{cm}$.
Answer $50$.
⏱ Shortcut $\dfrac{f_2}{f_1}=\dfrac{n-1}{n/n_\text{m}-1}$ — the radii never need to be found.

E3. JEE Main · MCQ A ray of light travelling in a medium of refractive index $\sqrt2$ strikes its plane boundary with air at an angle of incidence of $60^\circ$. The ray (A) is refracted at $30^\circ$ (B) emerges grazing the surface (C) is totally reflected (D) is refracted at $45^\circ$

Critical angle $\sin C=\dfrac1{\sqrt2}$, $C=45^\circ$.
Compare Denser to rarer and $60^\circ\gt45^\circ$: no refracted ray ($\sin r=\sqrt2\sin60^\circ=1.22\gt1$ is impossible).
Answer (C).
⏱ Shortcut If $n\sin i\gt1$ for a ray leaving into air, it is totally reflected; (B) would need $i=C$ exactly.

E4. JEE Main · Numerical An equilateral prism is made of glass of refractive index $\sqrt3$. Find its angle of minimum deviation in degrees. Give the answer to the nearest integer.

Formula $\sin\dfrac{A+\delta_m}{2}=n\sin\dfrac A2=\sqrt3\times\dfrac12=\dfrac{\sqrt3}{2}$.
Solve $\dfrac{60^\circ+\delta_m}{2}=60^\circ$, so $\delta_m=60^\circ$ (with $i=e=60^\circ$).
Answer $60$.
⏱ Shortcut The thin-prism formula $(n-1)A=43.9^\circ$ is wrong here — a $60^\circ$ prism is not thin.

E5. JEE Main · MCQ A convex lens of focal length $20\ \text{cm}$ is placed in contact with a concave lens of focal length $30\ \text{cm}$. The power of the combination is (A) $+8.33\ \text{D}$ (B) $-1.67\ \text{D}$ (C) $+3.33\ \text{D}$ (D) $+1.67\ \text{D}$

Powers $P_1=\dfrac{1}{0.20}=+5.0\ \text{D}$, $P_2=\dfrac{1}{-0.30}=-3.33\ \text{D}$.
Add $P=5.0-3.33=+1.67\ \text{D}$ (converging, $F=60\ \text{cm}$).
Answer (D).
⏱ Shortcut (A) adds the magnitudes — the concave lens has negative power. The stronger lens decides the sign.

E6. JEE Main · Numerical A convex lens forms a real image three times the size of an object on a screen $80\ \text{cm}$ from the object. Find the focal length of the lens in cm. Give the answer to the nearest integer.

Distances Real image: $m=\dfrac vu=-3$, so $v=-3u$; $v-u=80$ gives $u=-20\ \text{cm}$, $v=+60\ \text{cm}$.
Formula $\dfrac1f=\dfrac1{60}-\dfrac1{-20}=\dfrac{4}{60}$, $f=15\ \text{cm}$.
Answer $15$.
⏱ Shortcut For object–image distance $D$ and magnification $|m|$: $f=\dfrac{|m|D}{(|m|+1)^2}=\dfrac{3\times80}{16}=15\ \text{cm}$.
🔬 Research–Teaching NexusResearch-tutored · optional, about 45 minutes

Research question. How does the refractive index of a sugar solution depend on its concentration — and can total internal reflection be used to measure it?

Why it matters. Fruit growers in Brazil, wine makers in France and sugar mills in India test their products with refractometers, which read the sugar content from the refractive index; hospitals use the same instrument to check the protein in blood serum and urine. Most refractometers work by finding the critical angle (Result 16.8) at a glass–liquid boundary. Fibre-optic sensors that detect leaks or chemical changes use the same physics: a change in the outside index changes how much light the fibre loses.

What to do.

  1. Fill a semicircular transparent dish (or a clear plastic box with a curved side) with tap water. Shine a laser pointer through the curved side towards the centre of the flat side, so the beam meets the flat face from inside at a known angle.
  2. Increase the angle until the refracted beam just skims the surface and vanishes. Record this critical angle $C$ and compute $n=1/\sin C$.
  3. Dissolve measured masses of sugar to make solutions of $10$, $20$, $30$ and $40$ g per $100\ \text{mL}$, and repeat for each.
  4. As a second method, measure the apparent depth of a mark under a known depth of each solution with a travelling microscope, and compare the two values of $n$.
  5. Plot $n$ against concentration and fit a straight line.

Tools: Laser pointer (never look into the beam), semicircular dish, protractor or printed angle scale, sugar, measuring cylinder and balance, ruler, optionally a travelling microscope.

Hand in: A table of $C$ and $n$ for each concentration with uncertainties, the graph of $n$ against concentration with its slope, and a comparison of the two methods.

Going further. Use your line to find the sugar content of a fruit juice or a soft drink and compare with the label. Find out how commercial refractometers express the result on the "Brix" scale.

Instructor note: water gives $C\approx48.8^\circ$ ($n\approx1.33$); strong sugar solutions reach $n\approx1.40$ ($C\approx45.6^\circ$), so the changes are a few degrees — an angle scale printed in half-degrees is needed. The graph is close to linear. Emphasise laser safety and the "denser to rarer" condition: the beam must meet the flat face from inside the liquid.

🧪 Self-check quiz

0/5 correct

Q1. When light passes from air into water, which pair of quantities both decrease?

Q2. Total internal reflection can occur when light travels

Q3. A convex lens of power $+4\ \text{D}$ is placed in contact with a concave lens of power $-6\ \text{D}$. The combination is

Q4. A thin prism of angle $5^\circ$ deviates a ray by $3^\circ$. The refractive index of its material is

Q5. A coin lies at the bottom of a tank of water ($n=\tfrac43$) $1.2\ \text{m}$ deep. Seen from directly above, it appears to be at a depth of