Chapter 16 · Optics

16.4 Wavefronts, Huygens' Principle and Interference

Treat light as a wave and two ideas take over: every point of a wavefront acts as a new source (Huygens), and two coherent waves add with a phase difference that decides the brightness. Together they explain refraction, the bright and dark fringes of Young's double slit, the fringe width $\beta=\lambda D/d$, and how a thin sheet or a liquid moves and squeezes the pattern.

Shine a laser pointer through two scratches a fraction of a millimetre apart on a blackened glass slide, and the wall does not show two bright lines. It shows a long row of evenly spaced bright and dark bands. Adding a second opening has made some places on the wall darker than they were with one opening alone — light added to light has given darkness. No stream of particles can do that; waves can. When Thomas Young showed this effect around 1801, he settled a century-old argument: light is a wave, and its wavelength could now be measured with a ruler.

In this section we build the wave picture of light from the ground up. We describe a wave by its wavefronts and use Huygens' principle to rederive the laws of reflection and refraction. Then we ask when two light waves can interfere in a steady pattern (they must be coherent), work out how their intensities combine, and apply everything to Young's double-slit experiment: the positions of the fringes, the fringe width, the intensity across the screen, and what happens when a thin transparent sheet covers one slit or the whole apparatus is put under water. These are among the most frequently examined ideas in the optics unit of JEE Main.

Wavefronts and Huygens' principle

Drop a stone into a still pond. Circular ripples spread out, and every point on one ripple crest was disturbed at the same moment and is moving in step with every other point on that crest. A light wave spreading from a small lamp is the same in three dimensions, with the oscillating electric field in place of the water surface. Surfaces joining points that oscillate in step are the natural way to describe how the wave travels.

📘 Definition 16.13 — Wavefront

A wavefront is a surface on which every point of the wave has the same phase at a given instant. The wavefront moves outward with the wave speed $v$, and the direction of travel of the energy — the ray — is everywhere perpendicular to the wavefront (in an isotropic medium).

  • A point source gives spherical wavefronts; the rays are radial lines.
  • A line source (a long narrow lit slit) gives cylindrical wavefronts.
  • Far from any source a small part of a sphere is practically flat: a plane wavefront, with parallel rays. Light from the Sun or a distant star, and light from a point source at the focus of a convex lens, arrives as plane wavefronts.

Two wavefronts one wavelength apart differ in phase by $2\pi$. More generally, two points on the same ray separated by a distance $\Delta x$ differ in phase by

$$\Delta\phi=\frac{2\pi}{\lambda}\,\Delta x ,$$

where $\lambda$ is the wavelength in the medium the light is travelling through. This one relation, which converts a path difference into a phase difference, is the key to everything about interference later in this section.

How does a wavefront move forward? In 1678 Christiaan Huygens proposed a geometrical rule that answers this without any detailed theory of what is vibrating.

📐 Law 16.19 — Huygens' principle
  1. Every point on a wavefront acts as a source of secondary wavelets, which spread out in the forward direction with the speed of the wave in that medium.
  2. After a time $t$, the new position of the wavefront is the surface that touches all these wavelets in the forward direction — their forward envelope.

Each wavelet has radius $vt$. The backward-travelling part of the wavelets is ignored (a fuller theory shows that its amplitude is zero).

For a plane wavefront in a uniform medium the wavelets all have the same radius $vt$, so their envelope is a parallel plane a distance $vt$ ahead; for a spherical wavefront of radius $R$ the envelope is a sphere of radius $R+vt$. Nothing new so far — but at a boundary between two media the wavelets on the two sides grow at different speeds, and the construction then gives the laws of reflection and refraction.

Reflection. A plane wavefront $AB$ approaches a plane mirror at angle of incidence $i$ (the angle between the wavefront and the mirror equals the angle between the ray and the normal). End $A$ touches the mirror first. In the time $t$ that end $B$ takes to travel $BC=vt$ to the mirror, the wavelet from $A$ grows to radius $AE=vt$, back in the same medium. The reflected wavefront is the tangent $CE$ from $C$ to this wavelet. Triangles $ABC$ and $CEA$ are right-angled (at $B$ and $E$), share the hypotenuse $AC$ and have $BC=AE$, so they are congruent and the angle the reflected wavefront makes with the mirror equals the angle the incident one made: the angle of reflection equals the angle of incidence. The incident and reflected rays and the normal lie in one plane — the plane of the figure.

Refraction. Now let the boundary separate medium 1 (speed $v_1$) from medium 2 (speed $v_2$). While $B$ travels $BC=v_1t$ in medium 1, the wavelet from $A$ grows to radius $AE=v_2t$ in medium 2. The refracted wavefront is the tangent $CE$, and the refracted ray $AE$ makes angle $r$ with the normal. In the right-angled triangles $ABC$ and $AEC$:

$$\sin i=\frac{BC}{AC}=\frac{v_1t}{AC},\qquad \sin r=\frac{AE}{AC}=\frac{v_2t}{AC}\qquad\Rightarrow\qquad \frac{\sin i}{\sin r}=\frac{v_1}{v_2}.$$
📐 Result 16.20 — Refraction on the wave picture

When light passes from medium 1 into medium 2,

$$\frac{\sin i}{\sin r}=\frac{v_1}{v_2}=\frac{n_2}{n_1}=\frac{\lambda_1}{\lambda_2},$$

where the refractive index of a medium is $n=c/v$. The frequency does not change at the boundary (the field on both sides must oscillate together), so the wavelength scales with the speed: in a medium of index $n$, $\ v=\dfrac cn$ and $\lambda=\dfrac{\lambda_0}{n}$, where $\lambda_0$ is the wavelength in vacuum. Light slows down and bends towards the normal on entering a denser medium ($n_2\gt n_1$).

This was a real test of the wave theory. The particle (corpuscular) model explained refraction towards the normal only by assuming light speeds up in glass and water; the wave model needs it to slow down. When Foucault measured the speed of light in water in 1850 and found it smaller than in air, the wave picture won.

Optical path. In a length $L$ of a medium of index $n$, light takes time $L/v=nL/c$ — the same time it would take to cover a length $nL$ in vacuum. The product $nL$ is called the optical path. It also fits $nL/\lambda_0=L/\lambda$ wavelengths, so the phase change along a path is $\dfrac{2\pi}{\lambda_0}\times(\text{optical path})$. Comparing two routes by their optical paths, with the vacuum wavelength, is the safe way to handle interference when glass sheets or liquids are involved.

✏️ Example 1 — Light entering glass
Light of wavelength $600\ \text{nm}$ in air falls on a glass block ($n=1.5$) at an angle of incidence of $60^\circ$. Find (a) the angle of refraction, (b) the speed of light in the glass, (c) its frequency and its wavelength in the glass, (d) the time taken to cross $3.0\ \text{cm}$ of the glass along the normal. Take the refractive index of air as $1$ and $c=3.0\times10^8\ \text{m/s}$.
(a) $\sin r=\dfrac{\sin60^\circ}{1.5}=\dfrac{0.866}{1.5}=0.577$, so $r=35.3^\circ$.
(b) $v=\dfrac cn=\dfrac{3.0\times10^8}{1.5}=2.0\times10^8\ \text{m/s}$.
(c) $f=\dfrac{c}{\lambda_0}=\dfrac{3.0\times10^8}{600\times10^{-9}}=5.0\times10^{14}\ \text{Hz}$, the same in the glass. $\lambda=\dfrac{\lambda_0}{n}=\dfrac{600}{1.5}=400\ \text{nm}$.
(d) $t=\dfrac{L}{v}=\dfrac{0.030}{2.0\times10^8}=1.5\times10^{-10}\ \text{s}$ — the same as the time to cover the optical path $nL=4.5\ \text{cm}$ in vacuum.
Check $v=f\lambda=5.0\times10^{14}\times400\times10^{-9}=2.0\times10^8\ \text{m/s}$ ✓. Colour is set by the frequency, which is why a red beam stays red under water although its wavelength shrinks.
✏️ Example 2 — Huygens' construction with numbers
A plane wavefront $AB$, $4.0\ \text{cm}$ wide, travelling in air, meets a flat water surface ($n=\tfrac43$) at an angle of incidence of $53^\circ$ (take $\sin53^\circ=0.8$, $\cos53^\circ=0.6$). End $A$ touches the water first. Find (a) the distance $BC$ that end $B$ must still travel to reach the water, and the time it takes; (b) the radius $AE$ of the wavelet from $A$ in the water in that time; (c) the angle of refraction; (d) the width $EC$ of the refracted wavefront.
Geometry The wavefront makes angle $i=53^\circ$ with the surface. In right-angled triangle $ABC$ (right angle at $B$), $AC=\dfrac{AB}{\cos i}=\dfrac{4.0}{0.6}=6.67\ \text{cm}$ along the surface.
(a) $BC=AC\sin i=AB\tan i=4.0\times\tfrac43=5.33\ \text{cm}$; time $t=\dfrac{0.0533}{3.0\times10^8}=1.78\times10^{-10}\ \text{s}$.
(b) In water $v=\dfrac{3c}{4}$, so $AE=\dfrac{BC}{n}=\dfrac{5.33}{4/3}=4.0\ \text{cm}$.
(c) $\sin r=\dfrac{AE}{AC}=\dfrac{4.0}{6.67}=0.60$, so $r=37^\circ$. Check with Snell: $\dfrac{\sin53^\circ}{\sin37^\circ}=\dfrac{0.8}{0.6}=\tfrac43$ ✓.
(d) $EC=AC\cos r=6.67\times0.8=5.33\ \text{cm}$ — the refracted beam is wider than the incident one because it travels more steeply into the water.

A beam of red light passes from air into water. Which of these is the same in the water as in the air?

Coherent sources and the superposition of light waves

When two light waves arrive at the same point, the principle of superposition applies: the resultant electric field is the sum of the two fields. Let the two waves at a point be $E_1=a_1\sin\omega t$ and $E_2=a_2\sin(\omega t+\phi)$, with the same angular frequency $\omega$ and a phase difference $\phi$. Adding them (as two vectors of lengths $a_1$ and $a_2$ at angle $\phi$ — the same algebra as adding two SHMs) gives a single oscillation $E=A\sin(\omega t+\alpha)$ with

$$A^2=a_1^2+a_2^2+2a_1a_2\cos\phi .$$

The intensity of light (power per unit area) is proportional to the square of the amplitude, $I\propto A^2$. Writing $I_1\propto a_1^2$ and $I_2\propto a_2^2$ gives the central result of interference.

📐 Law 16.21 — Intensity of two superposed waves

Two waves of the same frequency, with intensities $I_1$ and $I_2$ and a phase difference $\phi$ at a point, produce there the intensity

$$I=I_1+I_2+2\sqrt{I_1I_2}\,\cos\phi .$$
  • Constructive interference (bright): $\phi=0,\ 2\pi,\ 4\pi,\dots$, i.e. path difference $\Delta x=0,\ \lambda,\ 2\lambda,\dots$: $\ I_\text{max}=\left(\sqrt{I_1}+\sqrt{I_2}\right)^2$.
  • Destructive interference (dark): $\phi=\pi,\ 3\pi,\dots$, i.e. $\Delta x=\tfrac{\lambda}{2},\ \tfrac{3\lambda}{2},\dots$: $\ I_\text{min}=\left(\sqrt{I_1}-\sqrt{I_2}\right)^2$.
  • For equal intensities $I_1=I_2=I_0$: $\ I=4I_0\cos^2\dfrac{\phi}{2}$, ranging from $4I_0$ (bright) to $0$ (completely dark).

Phase and path differences are linked by $\phi=\dfrac{2\pi}{\lambda}\Delta x$ (or $\dfrac{2\pi}{\lambda_0}\times$ optical path difference).

If the amplitude ratio is $a_1/a_2=r$, then $\dfrac{I_\text{max}}{I_\text{min}}=\left(\dfrac{r+1}{r-1}\right)^2$. Equal amplitudes give the best contrast — perfectly dark minima. Notice also that the bright places get more than $I_1+I_2$ and the dark places less: interference does not create or destroy energy, it only redistributes it. Averaged over many fringes, $\cos\phi$ averages to zero and the mean intensity is exactly $I_1+I_2$.

Why two torches never show fringes. An ordinary source emits light in short, independent bursts from countless atoms, each lasting about $10^{-8}\ \text{s}$ or less. Two separate lamps therefore have a phase difference that jumps randomly millions of times a second. The term $2\sqrt{I_1I_2}\cos\phi$ then averages to zero during any observation, and we simply see $I=I_1+I_2$ everywhere: the intensities add. A steady pattern needs the phase difference at each point to stay fixed in time.

📘 Definition 16.14 — Coherent sources

Two sources are coherent if they emit waves of the same frequency with a constant phase difference (which may be zero). Waves from coherent sources produce a stationary interference pattern, in which the intensities combine according to Law 16.21. Waves from incoherent sources have a randomly changing phase difference; their intensities simply add, $I=I_1+I_2$.

In practice two coherent sources are made from one source, so that any random jump in phase happens in both copies at once and cancels from their phase difference. There are two ways to do it:

A laser emits light whose phase stays regular over long stretches, so a laser beam is a highly coherent source and makes interference easy to see.

🧭 Problem-solving strategy — Conditions for sustained interference
  1. Coherent sources — derived from a single source, so the phase difference stays constant. This is essential: without it there are no fringes at all.
  2. Same frequency (monochromatic light, or one colour at a time); otherwise the phase difference drifts.
  3. Nearly equal amplitudes, so that the minima are dark and the fringes have good contrast ($I_\text{min}=0$ when $a_1=a_2$).
  4. Narrow sources, close together, and a distant screen, so that the fringes are sharp and wide enough to see (fringe width $\propto1/d$).
  5. The two waves should travel nearly the same direction and have the same state of polarisation.
✏️ Example 3 — Two unequal coherent sources
Two coherent sources produce, at points of a screen, waves of intensities $I_0$ and $4I_0$. Find (a) the maximum and minimum intensities and their ratio; (b) the intensity at points where the phase difference is $\pi/2$, and where it is $2\pi/3$; (c) the intensity everywhere if the two sources were incoherent; (d) the average intensity over many fringes in the coherent case.
Find $I_\text{max}/I_\text{min}$ in (a) before you open the solution:
(a) $\sqrt{I_1}=\sqrt{I_0}$, $\sqrt{I_2}=2\sqrt{I_0}$. $I_\text{max}=(1+2)^2I_0=9I_0$; $I_\text{min}=(2-1)^2I_0=I_0$; ratio $9$.
(b) $I=I_0+4I_0+2\sqrt{4I_0^2}\cos\phi=5I_0+4I_0\cos\phi$. For $\phi=\pi/2$: $5I_0$. For $\phi=2\pi/3$: $5I_0-2I_0=3I_0$.
(c) Incoherent: no interference term, $I=I_0+4I_0=5I_0$ everywhere — a uniform patch.
(d) $\cos\phi$ averages to zero, so the mean is $5I_0$ — the same as (c). Interference moves energy from the dark places to the bright ones; it does not change the total.
⏱ Shortcut Amplitude ratio $r=\sqrt{4}=2$: $\dfrac{I_\text{max}}{I_\text{min}}=\left(\dfrac{r+1}{r-1}\right)^2=\left(\dfrac31\right)^2=9$. A common slip is to use the intensity ratio $4$ in place of $r$.

Two identical sodium lamps are placed side by side and shine on a white wall. No interference fringes are seen because

Young's double-slit experiment

In Young's arrangement, monochromatic light from a narrow slit $S$ falls on two parallel narrow slits $S_1$ and $S_2$, a small distance $d$ apart and equidistant from $S$. The two slits are lit by the same wavefront, so they act as two coherent sources in phase. The light from them overlaps on a screen at a distance $D$ from the slits, with $D\gg d$ (typically $D\sim1\ \text{m}$ and $d\sim0.1$–$1\ \text{mm}$). Let $O$ be the point of the screen on the perpendicular bisector of $S_1S_2$, and $P$ a point a distance $y$ from $O$.

Path difference. Using Pythagoras for the two paths to $P$,

$$S_2P^2-S_1P^2=\left[D^2+\left(y+\tfrac d2\right)^2\right]-\left[D^2+\left(y-\tfrac d2\right)^2\right]=2yd .$$

So $S_2P-S_1P=\dfrac{2yd}{S_2P+S_1P}$. For $y$ and $d$ much smaller than $D$, $S_2P+S_1P\approx2D$, and the path difference is

$$\Delta x=S_2P-S_1P=\frac{yd}{D}=d\sin\theta\approx d\,\theta ,$$

where $\theta$ is the angle that $OP$ subtends at the midpoint of the slits.

📐 Law 16.22 — Fringes in Young's double-slit experiment

For two coherent in-phase slits a distance $d$ apart, a screen a distance $D$ away ($D\gg d$) and light of wavelength $\lambda$:

  • bright fringes ($\Delta x=n\lambda$) at $\ y_n=\dfrac{n\lambda D}{d}$, $\ n=0,\pm1,\pm2,\dots$ — the central fringe ($n=0$) at $O$ is bright;
  • dark fringes ($\Delta x=(2n-1)\tfrac{\lambda}{2}$) at $\ y'_n=\dfrac{(2n-1)\lambda D}{2d}$, $\ n=1,2,3,\dots$ on each side;
  • the fringe width (distance between consecutive bright, or consecutive dark, fringes) is $$\beta=\frac{\lambda D}{d},$$ the same for all fringes, and the angular fringe width is $\ \beta/D=\lambda/d$.

Combining the path difference with Law 16.21 for two equal waves (each of intensity $I_0$) gives the whole intensity pattern:

📐 Result 16.23 — Intensity on the screen
$$I(y)=4I_0\cos^2\!\left(\frac{\pi y}{\beta}\right)=I_\text{max}\cos^2\!\left(\frac{\pi\,d\,y}{\lambda D}\right),\qquad \phi=\frac{2\pi}{\lambda}\cdot\frac{yd}{D}.$$

Bright and dark fringes have equal widths, and in the ideal experiment all the bright fringes are equally bright. (In reality each slit has a finite width, and diffraction at each slit makes the fringes fade gradually away from the centre — see §16.5.)

What the formula $\beta=\lambda D/d$ tells you:

✏️ Example 4 — Fringe positions
In a Young's double-slit experiment, light of wavelength $600\ \text{nm}$ falls on slits $0.30\ \text{mm}$ apart, and the screen is $1.5\ \text{m}$ away. Find (a) the fringe width; (b) the distance of the third bright fringe and of the second dark fringe from the centre; (c) the distance between the third bright fringe on one side of the centre and the second dark fringe on the other side; (d) the angular fringe width; (e) the path difference at the third bright fringe.
(a) $\beta=\dfrac{\lambda D}{d}=\dfrac{600\times10^{-9}\times1.5}{0.30\times10^{-3}}=3.0\times10^{-3}\ \text{m}=3.0\ \text{mm}$.
(b) Third bright: $y_3=3\beta=9.0\ \text{mm}$. Second dark: $y'_2=\tfrac32\beta=4.5\ \text{mm}$ (the first dark is at $\beta/2$).
(c) On opposite sides the distances add: $9.0+4.5=13.5\ \text{mm}$.
(d) $\dfrac{\lambda}{d}=\dfrac{600\times10^{-9}}{0.30\times10^{-3}}=2.0\times10^{-3}\ \text{rad}$ (about $0.11^\circ$).
(e) $\Delta x=3\lambda=1800\ \text{nm}=1.8\ \mu\text{m}$. Check: $\dfrac{yd}{D}=\dfrac{9.0\times10^{-3}\times0.30\times10^{-3}}{1.5}=1.8\times10^{-6}\ \text{m}$ ✓.
✏️ Example 5 — Two wavelengths at once
A double slit with $d=0.50\ \text{mm}$ and $D=1.0\ \text{m}$ is lit by light containing two wavelengths, $650\ \text{nm}$ and $520\ \text{nm}$. (a) Find the fringe width for each. (b) Find the least distance from the central maximum at which a bright fringe of one wavelength falls exactly on a bright fringe of the other. (c) Is there any point where dark fringes of both wavelengths coincide? (d) The whole apparatus is now immersed in water ($n=\tfrac43$). Find the new distance in (b).
Find the distance in (b) before you open the solution:
(a) $\beta_1=\dfrac{650\times10^{-9}\times1.0}{0.50\times10^{-3}}=1.30\ \text{mm}$; $\ \beta_2=1.04\ \text{mm}$.
(b) Need $n_1\beta_1=n_2\beta_2$, i.e. $n_1\times650=n_2\times520$, so $\dfrac{n_1}{n_2}=\dfrac{520}{650}=\dfrac45$. Smallest: $n_1=4$, $n_2=5$: $y=4\times1.30=5\times1.04=5.20\ \text{mm}$.
(c) Dark fringes coincide if $(2n_1-1)\times650=(2n_2-1)\times520$, i.e. $5(2n_1-1)=4(2n_2-1)$. The left side is odd and the right side is even, so this never happens: no point is dark for both colours.
(d) Every fringe width is divided by $n$: $y=\dfrac{5.20}{4/3}=3.90\ \text{mm}$.
⏱ Shortcut First coincidence of bright fringes: $y=\dfrac{\text{LCM}(\lambda_1,\lambda_2)\,D}{d}$; here $\text{LCM}(650,520)=2600\ \text{nm}$, so $y=\dfrac{2600\times10^{-9}\times1.0}{0.50\times10^{-3}}=5.2\ \text{mm}$.

In Interactive 1 start from $\lambda=600\ \text{nm}$, $d=0.30\ \text{mm}$, $D=1.5\ \text{m}$ in air (fringe width $3.00\ \text{mm}$). Predict the fringe width after you change the slit separation to $d=0.60\ \text{mm}$ and the screen distance to $D=2.0\ \text{m}$.

At the start the readout shows $\beta=3.00\ \text{mm}$ and angular width $2.00\times10^{-3}\ \text{rad}$. With $d=0.60\ \text{mm}$ and $D=2.0\ \text{m}$ it reads $\beta=\lambda D/d=600\times10^{-9}\times2.0/0.60\times10^{-3}=2.00\ \text{mm}$: doubling $d$ halves $\beta$, and the larger $D$ then multiplies it by $4/3$. Now switch the medium to water: $\beta$ drops to $1.50\ \text{mm}$ ($=2.00/\tfrac43$), because the wavelength in water is only $450\ \text{nm}$. Press Play to send the probe along the screen and watch the path difference pass through whole numbers of wavelengths at every bright fringe.
Interactive 1 — Young's Double-Slit Lab

Light from slit $S$ reaches two slits $S_1$, $S_2$ (separation drawn much larger than life) and the fringes form on the screen; the graph on the right is the intensity $I/I_\text{max}$ along the screen ($\pm12\ \text{mm}$). Change the wavelength, slit separation, screen distance and medium. Press Play to move a probe point $P$ along the screen and read the path difference $S_2P-S_1P$ in wavelengths.

600 nm0.30 mm1.5 m
fringe width $\beta$—
angular width—
3rd bright fringe—
probe $P$—

In a Young's double-slit experiment the fringe width is $\beta$. The wavelength is increased by $20\%$ and the slit separation is doubled, with the screen unchanged. The new fringe width is

A thin sheet over one slit, and a liquid

Cover slit $S_1$ with a thin transparent sheet (mica, glass or plastic film) of thickness $t$ and refractive index $n$. Light from $S_1$ now crosses a thickness $t$ of the sheet instead of $t$ of air, so its optical path is longer by

$$\Delta=(n-1)\,t .$$

At a point $P$ at height $y$ the optical path difference between the two waves becomes $\dfrac{yd}{D}-(n-1)t$ (measured as path via $S_2$ minus path via $S_1$). The central fringe — zero optical path difference — is no longer at $O$ but where $\dfrac{y_0d}{D}=(n-1)t$, i.e. on the side of the covered slit: there the geometric path via $S_1$ is shorter, which makes up for the delay in the sheet. Every other fringe moves with it by the same amount, because the extra term is the same for every point.

📐 Result 16.24 — Shift of the fringes by a thin sheet

A thin transparent sheet of thickness $t$ and refractive index $n$ placed over one slit of a Young's double slit (in air) shifts the whole fringe pattern towards the covered slit by

$$y_0=\frac{(n-1)\,t\,D}{d}=\frac{(n-1)\,t}{\lambda}\,\beta ,$$

i.e. by $N=\dfrac{(n-1)t}{\lambda}$ fringes. The fringe width does not change. If the experiment is done in a medium of index $n_\text{m}$, replace $(n-1)$ by $(n-n_\text{m})$ and use the wavelength in the medium: $y_0=\dfrac{(n-n_\text{m})\,t\,D}{n_\text{m}\,d}$.

This is a sensitive way to measure the thickness or refractive index of a thin film: with white light, find the new position of the white central fringe and count how many fringes it has moved. Notice also what happens at $O$: the waves arrive there with phase difference $\phi=\dfrac{2\pi}{\lambda}(n-1)t$, so $O$ is bright when $(n-1)t$ is a whole number of wavelengths and dark when it is an odd number of half-wavelengths. The least thickness that makes $O$ dark is $t=\dfrac{\lambda}{2(n-1)}$.

Liquid between the slits and the screen. Filling the space with a liquid of index $n$ changes every path into $n$ times its length, so all fringes come closer: $\beta'=\beta/n$. The central fringe stays at $O$ (the two paths are still equal there). This gives a neat method of measuring the refractive index of a liquid: $n=\beta_\text{air}/\beta_\text{liquid}$.

✏️ Example 6 — A plastic film over the upper slit
A double slit with $d=0.25\ \text{mm}$ and $D=1.0\ \text{m}$ is lit with light of wavelength $600\ \text{nm}$. (a) Find the fringe width. (b) A plastic film of refractive index $1.6$ and thickness $4.0\ \mu\text{m}$ is placed over the upper slit. By how much, and in which direction, does the central fringe move? Is the point $O$ now bright or dark? (c) What is the least thickness of the same plastic that would make $O$ dark? (d) With the $4.0\ \mu\text{m}$ film in place, the whole apparatus is immersed in water ($n=\tfrac43$). Find the new fringe width and the new shift of the central fringe.
Find the shift of the central fringe in (b) first:
(a) $\beta=\dfrac{600\times10^{-9}\times1.0}{0.25\times10^{-3}}=2.4\times10^{-3}\ \text{m}=2.4\ \text{mm}$.
(b) Extra optical path $(n-1)t=0.6\times4.0=2.4\ \mu\text{m}=4\lambda$. Shift $y_0=\dfrac{2.4\times10^{-6}\times1.0}{0.25\times10^{-3}}=9.6\ \text{mm}=4\beta$, upwards (towards the covered slit). At $O$ the phase difference is $4\times2\pi$, so $O$ is bright — it now carries the fourth-order bright fringe.
(c) Dark at $O$ needs $(n-1)t=\tfrac{\lambda}{2}$: $t=\dfrac{600\times10^{-9}}{2\times0.6}=5.0\times10^{-7}\ \text{m}=0.50\ \mu\text{m}$.
(d) In water $\beta'=\dfrac{2.4}{4/3}=1.8\ \text{mm}$. The film now replaces water, not air: extra optical path $(1.6-\tfrac43)\times4.0=1.07\ \mu\text{m}$. Zero optical path difference where $n_\text{m}\dfrac{y_0d}{D}=1.07\ \mu\text{m}$: $y_0=\dfrac{1.07\times10^{-6}\times1.0}{\tfrac43\times0.25\times10^{-3}}=3.2\ \text{mm}$ (about $1.8$ fringes).
Check In (d), $N=\dfrac{(n-n_\text{m})t}{\lambda_0}=\dfrac{1.07}{0.60}=1.78$ fringes and $1.78\times1.8\ \text{mm}=3.2\ \text{mm}$ ✓.

In Interactive 2 (light of $600\ \text{nm}$, $d=0.25\ \text{mm}$, $D=1.0\ \text{m}$, in air) set the sheet's index to $1.6$ and increase its thickness from $0$ to $3.0\ \mu\text{m}$. What happens to the pattern?

With $n=1.6$ and $t=3.0\ \mu\text{m}$ the readouts show extra optical path $(n-1)t=1.80\ \mu\text{m}=3.00\,\lambda$, a shift of $7.20\ \text{mm}$ upward, fringe width $2.40\ \text{mm}$ and intensity at $O$ equal to $1.00\,I_\text{max}$ — three whole fringes have passed $O$, so it is bright again. Only $(n-1)t$ counts, not $nt$: the sheet replaces air that the light would have crossed anyway. Switch to water and the shift falls to $2.40\ \text{mm}$ while $\beta$ falls to $1.80\ \text{mm}$, and $O$ drops to $0.25\,I_\text{max}$.
Interactive 2 — A Thin Sheet Over One Slit

A transparent sheet covers the upper slit $S_1$ (light of $600\ \text{nm}$, $d=0.25\ \text{mm}$, $D=1.0\ \text{m}$). The faint dashed curve is the pattern without the sheet; the coloured curve is the pattern with it, and the arrow marks where the central (zero-path-difference) fringe has moved to. Press Play to thicken the sheet steadily and watch fringes march past the point $O$.

0.0 μm1.60
extra optical path—
shift of central fringe—
fringe width $\beta$—
intensity at $O$—
⚠️ JEE Main traps

1. Intensity ratios need square roots. $I_\text{max}/I_\text{min}=\left(\dfrac{\sqrt{I_1}+\sqrt{I_2}}{\sqrt{I_1}-\sqrt{I_2}}\right)^2$. Intensities $9:1$ give amplitudes $3:1$ and $I_\text{max}/I_\text{min}=4$, not $(10/8)^2$.

2. Counting dark fringes. The first dark fringe is at $\beta/2$, the $n$th at $(n-\tfrac12)\beta$. Bright and dark fringes on opposite sides of the centre: add their distances.

3. In a liquid the fringes shrink. $\beta'=\beta/n$ because the wavelength becomes $\lambda/n$; the frequency is unchanged. Don't multiply by $n$.

4. The thin sheet. The extra path is $(n-1)t$, not $nt$; the pattern moves towards the covered slit; the fringe width stays the same. In a liquid use $(n-n_\text{m})t$.

5. Incoherent sources. Two independent lamps (or two halves of a lamp filament) give $I=I_1+I_2$ — no fringes, no factor $4$. And $d$ is the separation of the slit centres, not the width of a slit.

In a Young's double-slit experiment with light of wavelength $\lambda$, a thin glass plate of refractive index $1.5$ covering the lower slit shifts the pattern by $6$ fringes. The thickness of the plate is

Exercises

Take $c=3.0\times10^8\ \text{m/s}$ and the refractive index of air as $1$ unless told otherwise. In Young's experiment $d$ is the slit separation and $D$ the slit-to-screen distance.

I
Wavefronts, Huygens' principle and refraction.
  • 1 Name the shape of the wavefront in each case: (a) a short distance from a small bulb; (b) light from a distant star reaching the Earth; (c) light from a long, narrow illuminated slit; (d) light leaving a convex lens when a point source is at its focus.
  • 2 Yellow light of wavelength $589\ \text{nm}$ in air enters water ($n=1.33$). Find its speed, its wavelength and its frequency in the water.
  • 3 Light falls on glass ($n=1.5$) at $45^\circ$. Find the angle of refraction.
  • 4 A ray in glass ($n=1.5$) meets a glass–water boundary ($n_\text{water}=\tfrac43$) at $30^\circ$. Find the angle of refraction in the water.
  • 5 Find the optical path of $4.5\ \text{cm}$ of water ($n=\tfrac43$) and the time light takes to cross it.
  • 6 Light of vacuum wavelength $600\ \text{nm}$ travels in glass ($n=1.5$). Find the phase difference, in radians, between two points $0.15\ \mu\text{m}$ apart along a ray.
  • 7 Use Huygens' construction to explain why a wavefront entering a slower medium obliquely turns so that the ray bends towards the normal. What would the construction predict if light travelled faster in glass?
1 (a) Spherical; (b) plane; (c) cylindrical; (d) plane.
2 $v=\dfrac{3.0\times10^8}{1.33}=2.26\times10^8\ \text{m/s}$; $\lambda=\dfrac{589}{1.33}=443\ \text{nm}$; $f=\dfrac{3.0\times10^8}{589\times10^{-9}}=5.09\times10^{14}\ \text{Hz}$ (unchanged).
3 $\sin r=\dfrac{\sin45^\circ}{1.5}=0.471$, $r=28.1^\circ$.
4 $1.5\sin30^\circ=\tfrac43\sin r$: $\sin r=0.5625$, $r=34.2^\circ$ (away from the normal — water is optically rarer than glass).
5 Optical path $nL=\tfrac43\times4.5=6.0\ \text{cm}$; $t=\dfrac{0.060}{3.0\times10^8}=2.0\times10^{-10}\ \text{s}$.
6 $\lambda=\dfrac{600}{1.5}=400\ \text{nm}$; $\Delta\phi=\dfrac{2\pi}{400}\times150=0.75\pi=2.36\ \text{rad}$ ($135^\circ$).
7 The end of the wavefront that enters first sends out wavelets that grow more slowly ($v_2\lt v_1$) while the other end is still moving fast in medium 1; the envelope therefore swings round, and the ray (perpendicular to it) turns towards the normal, with $\sin i/\sin r=v_1/v_2\gt1$. If light were faster in glass the wavelets in the glass would be larger and the ray would bend away from the normal — contrary to observation.
II
Superposition and coherence.
  • 1 Two coherent waves of amplitudes $3$ and $4$ units meet. Find the resultant amplitude when the phase difference is (a) $0$, (b) $\pi$, (c) $\pi/2$.
  • 2 The intensities of two coherent sources are in the ratio $25:9$. Find $I_\text{max}/I_\text{min}$ in their interference pattern.
  • 3 In an interference pattern $I_\text{max}/I_\text{min}=49$. Find the ratio of the amplitudes and of the intensities of the two waves.
  • 4 Two coherent waves, each of intensity $I_0$, meet with a path difference of $\lambda/3$. Find the resultant intensity in terms of $I_0$.
  • 5 Repeat part 4 for a path difference of $\lambda/4$.
  • 6 Explain why the light from the two headlamps of a car never produces interference fringes on a wall.
  • 7 Coherent waves of intensities $I$ and $9I$ meet with a phase difference of $60^\circ$. Find the resultant intensity.
  • 8 For the waves of part 7, find $I_\text{max}$, $I_\text{min}$ and the average intensity over many fringes, and show that energy is conserved.
1 $A=\sqrt{9+16+24\cos\phi}$: (a) $7$; (b) $1$; (c) $5$ units.
2 Amplitude ratio $5:3$: $\dfrac{I_\text{max}}{I_\text{min}}=\left(\dfrac{5+3}{5-3}\right)^2=16$.
3 $\dfrac{r+1}{r-1}=7\Rightarrow r=\dfrac{8}{6}=\dfrac43$: amplitudes $4:3$, intensities $16:9$.
4 $\phi=\dfrac{2\pi}{3}$: $I=4I_0\cos^2 60^\circ=I_0$.
5 $\phi=\dfrac{\pi}{2}$: $I=4I_0\cos^2 45^\circ=2I_0$.
6 The lamps are independent sources: the phase difference between their light changes randomly and very rapidly, so the interference term averages to zero and the intensities just add.
7 $I=I+9I+2\sqrt{9I^2}\cos60^\circ=10I+3I=13I$.
8 $I_\text{max}=(1+3)^2I=16I$, $I_\text{min}=(3-1)^2I=4I$; the average of $10I+6I\cos\phi$ over many fringes is $10I=I+9I$ — the energy is redistributed, not created.
III
Young's double-slit experiment.
  • 1 Light of wavelength $500\ \text{nm}$ falls on slits $0.20\ \text{mm}$ apart; the screen is $1.2\ \text{m}$ away. Find the fringe width.
  • 2 For part 1, find the distances of the fourth bright fringe and of the third dark fringe from the centre.
  • 3 Fringes $2.0\ \text{mm}$ wide are obtained with slits $0.30\ \text{mm}$ apart and a screen $1.0\ \text{m}$ away. Find the wavelength.
  • 4 The angular fringe width for light of $600\ \text{nm}$ is $0.20^\circ$. Find the slit separation in mm.
  • 5 The screen distance is doubled and the slit separation is halved. By what factor does the fringe width change?
  • 6 For part 1, find the distance between the second bright fringe on one side of the centre and the fifth dark fringe on the other side.
  • 7 Describe the pattern when white light is used. Why are only a few coloured fringes seen?
  • 8 In a fixed field of view, $24$ fringes are seen with light of $500\ \text{nm}$. How many are seen with light of $600\ \text{nm}$?
  • 9 The fringe width is $3.0\ \text{mm}$ and the intensity of the central fringe is $I_\text{max}$. Find the intensity at $0.50\ \text{mm}$ from the centre, as a fraction of $I_\text{max}$.
1 $\beta=\dfrac{500\times10^{-9}\times1.2}{0.20\times10^{-3}}=3.0\times10^{-3}\ \text{m}=3.0\ \text{mm}$.
2 Fourth bright: $4\beta=12.0\ \text{mm}$. Third dark: $2.5\beta=7.5\ \text{mm}$.
3 $\lambda=\dfrac{\beta d}{D}=\dfrac{2.0\times10^{-3}\times0.30\times10^{-3}}{1.0}=6.0\times10^{-7}\ \text{m}=600\ \text{nm}$.
4 $\theta=0.20^\circ=3.49\times10^{-3}\ \text{rad}$; $d=\dfrac{\lambda}{\theta}=\dfrac{600\times10^{-9}}{3.49\times10^{-3}}=1.72\times10^{-4}\ \text{m}=0.172\ \text{mm}$.
5 $\beta\propto D/d$: $\times2\times2=4$ times wider.
6 $2\beta+4.5\beta=6.5\beta=6.5\times3.0=19.5\ \text{mm}$.
7 The central fringe is white (zero path difference for all colours); on each side come coloured fringes, violet/blue nearer the centre and red farther out ($\beta\propto\lambda$). Because the fringe widths differ, after a few fringes the bright fringes of one colour fall on the dark fringes of another and the colours mix into white.
8 Number in a fixed width $\propto1/\beta\propto1/\lambda$: $24\times\dfrac{500}{600}=20$.
9 $I=I_\text{max}\cos^2\dfrac{\pi y}{\beta}=I_\text{max}\cos^2\dfrac{\pi}{6}=0.75\,I_\text{max}$.
IV
Thin sheets, liquids and harder problems.
  • 1 The fringe width in air is $2.4\ \text{mm}$. Find it when the whole apparatus is immersed in water ($n=\tfrac43$).
  • 2 A glass sheet ($n=1.5$) $6.0\ \mu\text{m}$ thick covers one slit; $\lambda=600\ \text{nm}$ and the fringe width is $2.0\ \text{mm}$. Find the number of fringes by which the pattern shifts, and the shift in mm.
  • 3 Find the least thickness of a sheet of index $1.5$ which, placed over one slit, makes the centre of the screen dark for light of $500\ \text{nm}$.
  • 4 A film $9.6\ \mu\text{m}$ thick over one slit moves the central bright fringe to the position previously occupied by the eighth bright fringe ($\lambda=600\ \text{nm}$). Find the refractive index of the film.
  • 5 Sheets of the same thickness $6.0\ \mu\text{m}$ but indices $1.6$ (over $S_1$) and $1.4$ (over $S_2$) cover the two slits; $\lambda=600\ \text{nm}$. By how many fringes, and towards which slit, does the pattern move?
  • 6 The fringe width is $1.50\ \text{mm}$ in air and $1.20\ \text{mm}$ when the space between the slits and the screen is filled with a liquid. Find the refractive index of the liquid.
  • 7 The source slit $S$, $20\ \text{cm}$ behind the double slit, is moved $1.0\ \text{mm}$ upward ($D=1.0\ \text{m}$). Where does the central fringe go?
  • 8 With equal slits, the intensity at the centre of the screen is $I_\text{max}$. A film of index $1.5$ and thickness $0.20\ \mu\text{m}$ is placed over one slit ($\lambda=600\ \text{nm}$). Find the new intensity at the centre, as a fraction of $I_\text{max}$.
1 $\beta'=\dfrac{2.4}{4/3}=1.8\ \text{mm}$.
2 $N=\dfrac{(n-1)t}{\lambda}=\dfrac{0.5\times6.0}{0.60}=5$ fringes; shift $5\beta=10\ \text{mm}$, towards the covered slit.
3 $(n-1)t=\dfrac{\lambda}{2}$: $t=\dfrac{500}{2\times0.5}=500\ \text{nm}=0.50\ \mu\text{m}$.
4 $(n-1)t=8\lambda$: $n-1=\dfrac{8\times0.60}{9.6}=0.50$, $n=1.5$.
5 Net extra optical path on the $S_1$ side: $(1.6-1.4)\times6.0=1.2\ \mu\text{m}=2\lambda$. The pattern moves $2$ fringes towards $S_1$ (the slit behind the denser sheet).
6 $n=\dfrac{\beta_\text{air}}{\beta_\text{liquid}}=\dfrac{1.50}{1.20}=1.25$.
7 The central fringe is where the total paths from $S$ are equal: the extra path $\dfrac{s\,d}{L}$ before the slits must be cancelled by $\dfrac{y\,d}{D}$ after them, so $y=\dfrac{sD}{L}=\dfrac{1.0\times1.0}{0.20}=5.0\ \text{mm}$ downward — opposite to the source's movement. The fringe width is unchanged.
8 $\phi=\dfrac{2\pi}{600}\times(0.5\times200)=\dfrac{\pi}{3}$; $I=I_\text{max}\cos^2\dfrac{\pi}{6}=0.75\,I_\text{max}$.
🎯 Exam corner JEE Main · MCQJEE Main · Numerical

E1. JEE Main · MCQ In a Young's double-slit experiment in air the fringe width is $0.80\ \text{mm}$. The whole apparatus is immersed in a liquid of refractive index $1.6$ and the slit separation is halved. The new fringe width is (A) $0.64\ \text{mm}$ (B) $0.25\ \text{mm}$ (C) $1.0\ \text{mm}$ (D) $2.56\ \text{mm}$

Idea $\beta=\dfrac{\lambda D}{d}$ with $\lambda\to\lambda/n$ and $d\to d/2$.
Work $\beta'=0.80\times\dfrac{2}{1.6}=1.0\ \text{mm}$.
Answer (C).
⏱ Shortcut (D) multiplies by $n$ instead of dividing; (B) halves $\beta$ for the halved $d$. Two factors: $\times2$, $\div1.6$.

E2. JEE Main · Numerical In the interference pattern of two coherent sources the ratio of maximum to minimum intensity is $25:9$. The ratio of the intensities of the two sources is $x:1$. Find $x$ to the nearest integer.

Amplitudes $\dfrac{a_1+a_2}{a_1-a_2}=\sqrt{\dfrac{25}{9}}=\dfrac53$, so $3a_1+3a_2=5a_1-5a_2$ and $a_1=4a_2$.
Intensities $\dfrac{I_1}{I_2}=\left(\dfrac{a_1}{a_2}\right)^2=16$.
Answer $16$.
⏱ Shortcut Componendo–dividendo: $\dfrac{a_1}{a_2}=\dfrac{5+3}{5-3}=4$, then square.

E3. JEE Main · MCQ When a thin transparent film of refractive index $1.5$ is placed over one slit of a Young's double slit lit by light of wavelength $500\ \text{nm}$, the central bright fringe moves to the position of the seventh bright fringe. The thickness of the film is (A) $7.0\ \mu\text{m}$ (B) $3.5\ \mu\text{m}$ (C) $2.3\ \mu\text{m}$ (D) $10.5\ \mu\text{m}$

Idea A shift of $7$ fringes means the extra optical path is $7\lambda$: $(n-1)t=7\lambda$.
Work $t=\dfrac{7\times500\times10^{-9}}{0.5}=7.0\times10^{-6}\ \text{m}$.
Answer (A).
⏱ Shortcut (C) comes from $nt=7\lambda$, (B) from $t=7\lambda$; with $n=1.5$, $t=2N\lambda$.

E4. JEE Main · Numerical A Young's double slit with slit separation $0.50\ \text{mm}$ and screen distance $1.0\ \text{m}$ is lit by a mixture of light of wavelengths $500\ \text{nm}$ and $600\ \text{nm}$. Find the least distance from the central maximum, in mm, at which bright fringes of the two wavelengths coincide. Give the answer to the nearest integer.

Condition $n_1\times500=n_2\times600$: smallest $n_1=6$, $n_2=5$, common path difference $3000\ \text{nm}$.
Distance $y=\dfrac{3000\times10^{-9}\times1.0}{0.50\times10^{-3}}=6.0\times10^{-3}\ \text{m}=6.0\ \text{mm}$.
Answer $6$.
⏱ Shortcut $y=\text{LCM}(\lambda_1,\lambda_2)\times\dfrac{D}{d}$, and $D/d=2000$ here: $3\ \mu\text{m}\times2000=6\ \text{mm}$.

E5. JEE Main · Numerical In a Young's double-slit experiment with two identical slits the fringe width is $6.0\ \text{mm}$ and the intensity at the central maximum is $100$ units. Find the intensity, in the same units, at a point $1.0\ \text{mm}$ from the central maximum. Give the answer to the nearest integer.

Phase A shift of one fringe width changes the phase by $2\pi$: $\phi=2\pi\times\dfrac{1.0}{6.0}=\dfrac{\pi}{3}$.
Intensity $I=I_\text{max}\cos^2\dfrac{\phi}{2}=100\cos^2 30^\circ=75$.
Answer $75$.
⏱ Shortcut $I=I_\text{max}\cos^2\left(\pi y/\beta\right)$ directly: $\cos^2(\pi/6)=3/4$. Forgetting the half gives $\cos^2 60^\circ$ and $25$ — the commonest slip.

E6. JEE Main · MCQ A plane wavefront passes from medium 1 into medium 2. The angle of incidence is $45^\circ$ and the angle of refraction is $30^\circ$. The ratio of the wavelength in medium 1 to that in medium 2 is (A) $1/\sqrt2$ (B) $\sqrt3/\sqrt2$ (C) $1$ (D) $\sqrt2$

Huygens $\dfrac{\sin i}{\sin r}=\dfrac{v_1}{v_2}=\dfrac{\lambda_1}{\lambda_2}$ (the frequency is the same in both media).
Work $\dfrac{\lambda_1}{\lambda_2}=\dfrac{\sin45^\circ}{\sin30^\circ}=\dfrac{1/\sqrt2}{1/2}=\sqrt2$.
Answer (D).
⏱ Shortcut Light bends towards the normal ($r\lt i$), so it slows down and its wavelength shrinks: $\lambda_1/\lambda_2\gt1$ rules out (A) and (C) at once.
🔬 Research–Teaching NexusResearch-tutored · optional, about 60 minutes

Research question. How accurately can you measure the wavelength of a laser pointer with a home-made double slit, a wall and a tape measure?

Why it matters. Interference turns lengths far too small to see into lengths you can measure with a ruler. The same principle, refined, is used in interferometers that check the flatness of telescope mirrors to a few nanometres, in the fibre-optic sensors that monitor strain in bridges and aircraft wings, and in the kilometre-long detectors in the USA, Italy, Japan and (soon) India that sense gravitational waves by measuring changes in length far smaller than an atomic nucleus.

What to do.

  1. Coat a microscope slide with soot from a candle (or use the dark part of an exposed photographic film). Holding two razor blades taped together, scratch two parallel lines; the blade thickness sets $d$ (about $0.1$–$0.2\ \text{mm}$).
  2. Measure $d$: project the slits' shadow, or view them through a magnifier against a fine scale; repeat and average.
  3. In a dark room, shine a red laser pointer through the slits onto a wall $2$–$4\ \text{m}$ away. Never look into the beam or its reflection. Measure $D$.
  4. Mark the centres of as many bright fringes as you can see and measure the distance spanned by $10$ or more fringe widths; divide to get $\beta$.
  5. Compute $\lambda=\beta d/D$ and compare with the value printed on the pointer. Repeat for three values of $D$ and plot $\beta$ against $D$.

Tools: Laser pointer (red, and green if available), glass slide, candle or marker, two razor blades and tape, magnifying glass or travelling microscope, measuring tape, clamp or books to steady the pointer.

Hand in: A table of $D$, the measured span of fringes and $\beta$; the graph of $\beta$ against $D$ with its slope; your value of $\lambda$ with an uncertainty estimate (which measurement contributes most?); and a comparison of red and green fringe widths.

Going further. Look closely: the fringes fade in and out in groups. Measure the width of the whole bright patch — it is set by the width of each slit, not their separation (§16.5). Can you estimate the slit width from it?

Instructor note: with $d\approx0.15\ \text{mm}$ and $D=3\ \text{m}$ a red pointer gives $\beta\approx13\ \text{mm}$ — easy to measure. The largest error is usually in $d$; students who measure $d$ by projecting the slits with a lens, or who use a commercial double-slit slide as a check, typically get within $5\%$. The fading in groups (diffraction envelope) is a good bridge to the next section.

🧪 Self-check quiz

0/5 correct

Q1. Two sources of light are said to be coherent when they have

Q2. In a Young's double-slit experiment the wavelength is changed from $500\ \text{nm}$ to $600\ \text{nm}$, with everything else unchanged. The fringe width

Q3. A thin transparent sheet is placed over one slit in Young's experiment. Then

Q4. Two coherent sources have intensities in the ratio $9:1$. The ratio $I_\text{max}/I_\text{min}$ in their interference pattern is

Q5. Light from a point source placed at the focus of a convex lens emerges from the lens with wavefronts that are