Full chapter
Superposition
All 13 topics and the revision summary on one page.
01
The principle of superposition
Make a guess. You are not marked.
Two pulses travel towards each other on a rope. One is 3 cm up and the other is 3 cm down, with the same shape. What happens when they overlap exactly?
Show the answer
The rope is flat for an instant, then both pulses carry on unchanged.
By the principle of superposition, the displacements add: +3 cm and -3 cm give zero. The pulses pass through each other, so each continues afterwards with its original shape.
When waves meet, the displacement at each point is the sum of the displacements of the separate waves. Keep the signs: up is plus and down is minus.
Adding waves
The principle of superposition says this. When two or more waves meet at a point, the resultant displacement there is the vector sum of the displacements of the separate waves.
Only add values for the same point at the same instant. For light, the quantity that adds is the electric field.
Worked example
Two pulses pass through each other
Two pulses on a rope meet. One is 3.0 mm up and the other is 2.0 mm down. Where their peaks overlap:
= +1.0 mm
Add the displacements at each point
Brown long dashes: the +3.0 mm pulse. Green short dashes: the -2.0 mm pulse. Solid blue: their sum. Each row has its own zero line.
After they pass, both pulses carry on unchanged. Neither one was destroyed.
The pulses do not bounce off each other. Each one carries on afterwards as if the other had not been there.
Check your understanding
At one position and instant, two overlapping string pulses contribute displacements of +3.0 mm and -2.0 mm. What is the resultant displacement in the linear model?
Show the answer
+1.0 mm
The algebraic sum is +3.0 + (-2.0) = +1.0 mm. The positive contribution is larger, so the resultant remains on the positive side of equilibrium.
Interference
Interference is the effect of waves superposing. When the displacements add to a larger one, the interference is constructive. When they partly or fully cancel, it is destructive.
Cancelling at one point does not destroy energy. In a steady pattern, the energy goes to the places where the waves add.
A real example. Noise-cancelling headphones use superposition. A microphone picks up the noise around you. A small amplifier then makes a sound wave exactly out of phase with it, so the two cancel at your ear. Bose was one of the first companies to sell them, starting with headsets for aircraft pilots. They work best on steady, low-pitched noise such as engine hum.
Add displacements, not amplitudes. Adding the amplitudes is only right when the waves are in phase.
02
How a standing wave forms
Two waves of the same frequency, travelling in opposite directions, add to make a standing wave. Its nodes and antinodes stay in the same places.
A standing wave, or stationary wave, needs two waves with the same frequency, speed and wavelength, and similar amplitude, travelling in opposite directions. Usually one of them is the reflection of the other.
Add the two waves at three times
Each wave here has amplitude 2.0 mm, wavelength 0.80 m and period 0.20 s. Add their displacements point by point.
Two travelling waves add to make a standing wave
Wave 1 travels right and wave 2 travels left, each with amplitude 2.0 mm. Brown long dashes and green short dashes are the two waves; solid blue is their sum. All panels share the same axes.
t = 0
Here the two waves line up exactly, so their dashed lines overlap. The blue sum is twice either wave.
t = T/4 = 0.050 s
The arrows show which way the antinodes are moving. The string is flat for an instant, but only the nodes stay still all the time.
t = T/2 = 0.100 s
Here the two waves line up exactly, so their dashed lines overlap. The blue sum is twice either wave.
- At t = 0, the resultant at x = 0, 0.20, 0.40, 0.60 and 0.80 m is 0, +4.0, 0, -4.0 and 0 mm.
- At t = T/4 = 0.050 s, the waves cancel at every point, so the whole string is flat.
- At t = T/2 = 0.100 s, every displacement has the opposite sign to the one at t = 0.
Nodes and antinodes
A node is a point that never moves: its amplitude is zero. Here the nodes are at x = 0, 0.40 and 0.80 m. An antinode is a point of largest amplitude. Here the antinodes are at x = 0.20 and 0.60 m, with amplitude 4.0 mm.
Antinode to next antinode = λ/2
Node to nearest antinode = λ/4
All the points between two neighbouring nodes move in phase. Points in neighbouring loops move in antiphase: one loop goes up while the next goes down.
Check your understanding
Points P and Q on a standing wave are 0.30 m apart, in neighbouring loops. The wavelength is 0.80 m. What is the phase difference between their oscillations?
Show the answer
180 degrees: they move in antiphase.
All the points in one loop move in phase, and neighbouring loops move in antiphase. In a standing wave the phase does not change steadily with distance.
How it differs from a progressive wave
A progressive wave carries energy along, and every point has the same amplitude. In a standing wave there is no net flow of energy along the string. The amplitude changes from zero at a node to a maximum at an antinode.
A quarter period later, the nodes are still in the same places. Do not draw the wave shifted sideways, as if it were moving along.
Put these in order
A string is fixed at both ends and driven at one of its natural frequencies. Put the explanation of how the standing wave forms in order.
Show the answer
1. A wave travels along the string and is reflected at the fixed end. The reflected wave has the same frequency, wavelength and speed as the incident wave, but travels the other way.
2. The incident and reflected waves travel in opposite directions through each other and superpose. At each point the two displacements add according to the principle of superposition.
3. Where the two waves always arrive in antiphase, the displacements cancel: a node. Halfway between two nodes the waves always arrive in phase.
4. Where they always arrive in phase, the displacements add to give the maximum amplitude: an antinode. Because the nodes and antinodes stay in the same places, the pattern does not travel.
5. The pattern stays in place, with adjacent nodes half a wavelength apart.
03
Standing waves on strings and with microwaves
Make a guess. You are not marked.
On a vibrating string, adjacent nodes are 0.25 m apart. What is the wavelength of the travelling waves that form the pattern?
Show the answer
0.50 m, because adjacent nodes are half a wavelength apart.
In a standing wave, the distance from one node to the next is half a wavelength. 2 x 0.25 = 0.50 m.
Standing waves on a string and with microwaves give you a wavelength. Neighbouring nodes are half a wavelength apart.
A stretched string
Fix a string between two supports. A vibrator near one end, run by a signal generator, shakes it. Waves travel along the string, reflect at the ends and overlap. At certain frequencies a steady pattern of loops appears.
Both fixed ends are nodes. So the length L between the supports holds a whole number n of loops, each λ/2 long:
Microwaves and a metal sheet
Point a microwave transmitter at a metal sheet. The sheet reflects the waves back along the same line, and the two waves overlap. Move a probe detector slowly along the line between them. Its reading rises and falls: largest at the antinodes and smallest at the nodes.
Find several nodes or minima and count the gaps
A stretched string
The vibrator shakes the string only a little, so both ends act as nodes. Subtract the ruler readings, then divide by the two gaps.
Microwaves and their reflection
Keep the transmitter, the reflector and the direction of the probe fixed. Real minima may not fall all the way to zero.
Worked example
Count the gaps, not the nodes
On the string, three nodes are at 13.0, 53.0 and 93.0 cm. That is two gaps.
One gap = 80.0/2 = 40.0 cm = λ/2
λ = 2 × 40.0 = 80.0 cm = 0.800 m
With the microwaves, four minima are at 8.0, 9.5, 11.0 and 12.5 cm. That is three gaps.
One gap = 4.5/3 = 1.5 cm = λ/2
λ = 2 × 1.5 = 3.0 cm
Measuring across several gaps makes the reading error a smaller fraction of the length.
Check your understanding
A microwave probe moves between a transmitter and a metal sheet. It finds six minima in a row, and the first and sixth are 7.0 cm apart. What is the wavelength?
Show the answer
2.8 cm
Six minima have five gaps, so one gap is 7.0/5 = 1.4 cm. Neighbouring minima are half a wavelength apart, so the wavelength is 2 x 1.4 = 2.8 cm.
A real example. A microwave oven uses microwaves of about 2.45 GHz, which have a wavelength of 12.2 cm. They reflect off the metal walls and overlap, so the oven has hot spots and cold spots. In a simple standing wave, neighbouring hot spots would be half a wavelength apart, about 6 cm. A turntable carries the food through them, so it heats more evenly.
Neighbouring minima are half a wavelength apart, not a whole wavelength. Double the gap to get λ.
04
Standing sound waves in air columns
Make a guess. You are not marked.
A pipe closed at one end has a displacement node at the closed end. What is at the open end?
Show the answer
A displacement antinode, where the air is free to move most.
At the open end the air can move freely, so the standing wave has a displacement antinode there. The simplest pattern is a quarter wavelength long.
Sound in a tube forms standing waves too. A closed end is a displacement node, and an open end is a displacement antinode.
Displacement and pressure
Sound is a longitudinal wave, so the air moves back and forth along the tube. You can describe a standing sound wave in two ways. One is how far the air moves: its displacement. The other is how much the pressure differs from normal air pressure.
- At a closed end, the air cannot move, so it is a displacement node. The air is squeezed and stretched most here, so it is a pressure antinode.
- At an open end, the air meets the room, so the pressure stays near normal. It is a pressure node and a displacement antinode.
So every displacement node is a pressure antinode, and every displacement antinode is a pressure node. A displacement node and the nearest pressure node are λ/4 apart.
A tube closed at one end
The simplest pattern has a node at the closed end and an antinode at the open end, which is a quarter of a wavelength. Each next pattern adds half a wavelength, so the length is an odd number of quarter wavelengths:
Second resonance: L = 3λ/4
Then 5λ/4, 7λ/4, ...
Displacement and pressure in a tube closed at one end
First resonance: air length 0.160 m
Drawn from the closed end at x = 0. Grey dashed line: the top of the tube. Brown dotted line: where the antinode really is, 0.015 m beyond the top (the end correction).
A pressure node is where the pressure stays at normal air pressure, not a vacuum. Each displacement node is a pressure antinode.
Second resonance: air length 0.510 m
Drawn from the closed end at x = 0. Grey dashed line: the top of the tube. Brown dotted line: where the antinode really is, 0.015 m beyond the top (the end correction).
A pressure node is where the pressure stays at normal air pressure, not a vacuum. Each displacement node is a pressure antinode.
The antinode at an open end is really a little outside the tube. This extra distance is called the end correction, e. So the first resonance is at L + e = λ/4. A tube open at both ends has a displacement antinode at each end.
Check your understanding
A tube closed at one end resonates at its lowest frequency. Where is the amplitude of the pressure variation largest?
Show the answer
At the closed end, a displacement node.
The air cannot move at the closed end, so it is squeezed and stretched most there. Every displacement node is a pressure antinode.
A pressure node is not a vacuum. It is a place where the pressure stays at normal air pressure.
05
Measuring the wavelength of sound
Make a guess. You are not marked.
A tuning fork gives its first resonance when the air column in a tube is 0.16 m long. The water level is then lowered slowly. At about what air length is the next resonance?
Show the answer
About 0.5 m, three times as long.
The first resonance holds a quarter of a wavelength and the next holds three quarters. So the next length is about three times the first; the end correction makes it a little more, about 0.51 m.
A resonance tube measures the wavelength of sound. The air lengths of two resonances in a row differ by half a wavelength.
The resonance tube
Hold a vibrating tuning fork over the open top of a tube that stands in water. Change the water level to change the length of the air column. At certain lengths the sound gets much louder. This is resonance: a standing wave builds up because the fork's frequency matches a natural frequency of the air column.
The water surface is the closed end. Measure each air length from the water surface to the top of the tube.
A resonance tube
Measure each air length from the water surface to the top of the tube.
The top of the tube stays fixed; the water level moves
Worked example
Two resonances remove the end correction
A 480 Hz fork gives resonances at air lengths L1 = 0.160 m and L2 = 0.510 m. The end correction e is the same both times.
λ = 2(0.510 - 0.160) = 0.700 m
v = fλ = 480 × 0.700 = 336 m/s
The end correction cancels when you subtract. You can then find it: λ/4 = 0.175 m, so e = 0.175 - 0.160 = 0.015 m.
Check your understanding
A 480 Hz fork gives successive resonances at physical air lengths 0.160 m and 0.510 m in the same closed/open tube. Assume a common end correction. What are the wavelength and wave speed?
Show the answer
Wavelength 0.700 m; speed 336 m/s.
The length difference is 0.350 m = lambda/2, so lambda = 0.700 m. The wave speed is 480 x 0.700 = 336 m/s.
Compare with real data. The worked numbers are exam-style and give 336 m/s. The speed of sound in air at 20 °C is about 343 m/s. It rises with temperature, to about 349 m/s at 30 °C, so note the room temperature with your result.
Getting a good reading
The loudest point is broad. Find it coming from a longer length and from a shorter one, then take the average. Read the scale with your eye level with the water surface. Check that the two resonances are next to each other: skipping one adds another half wavelength.
Do not use λ = 4L1 on its own. It leaves out the end correction and gives a wavelength that is too small, here 0.640 m.
06
Coherence, phase and path difference
Make a guess. You are not marked.
Two loudspeakers driven in phase are 3.0 m and 4.5 m from a listener. The wavelength of the sound is 1.0 m. What does the listener hear?
Show the answer
A quiet spot, because the path difference is one and a half wavelengths.
The path difference is 4.5 - 3.0 = 1.5 m, which is 1.5 wavelengths. The waves arrive in antiphase and interfere destructively.
Whether two waves add or cancel at a point depends on their phase difference when they arrive. That depends on the path difference and on how the sources start.
Four terms
- Coherent sources have the same frequency and a constant phase difference. They do not have to be in phase.
- Phase difference compares the stages of two oscillations, in degrees or radians.
- Path difference is the difference between the distances from the two sources to a point.
- Diffraction is the spreading of waves through a gap or round an edge. It lets the waves from two slits overlap. See diffraction.
The path rule for sources in phase
An extra path of one wavelength delays a wave by one whole cycle. An extra half wavelength delays it by half a cycle. So for two sources in phase:
Minimum: path difference = (n + ½)λ
n = 0, 1, 2, ...
A complete cancellation also needs the two waves to have equal amplitudes where they meet. If one is weaker, the minimum is quiet but not silent.
Worked example
Sources in antiphase
Two loudspeakers are wired in antiphase: one pushes while the other pulls. Their paths to point P differ by λ/2.
Source phase and path difference both count
Two loudspeakers in antiphase
One speaker is wired the other way round, which gives half a cycle. The extra half wavelength of path gives another half cycle, so the waves arrive in phase and P is loud.
The sources start half a cycle apart. The extra path adds another half cycle. Together that is one whole cycle, so the waves arrive in phase and P is loud.
Check your understanding
Two coherent sound sources of equal amplitude are in antiphase at emission. Their paths to P differ by half a wavelength. What happens at P?
Show the answer
Constructive interference: the arrivals are in phase.
The half-cycle source difference and half-cycle path contribution together give a whole-cycle difference, equivalent to zero phase difference. Both contributions matter.
The rule "path difference nλ gives a maximum" is only for sources in phase. Check how the sources start before you use it.
07
Two-source interference experiments
Make a guess. You are not marked.
In a double-slit experiment, the laser is replaced by two identical red lamps, one behind each slit. What appears on the screen?
Show the answer
Even light with no steady fringes, because the lamps are not coherent.
Light from two separate lamps has a phase difference that changes at random, far too fast to see. Any fringes keep moving, so the screen looks evenly lit.
Interference from two sources can be shown with water waves, sound, microwaves and light. Each needs two coherent sources whose waves overlap.
Four demonstrations
- Water. In a ripple tank, two dippers on one vibrating bar send out ripples. Where they overlap, lines of still water lie between lines of strong ripples.
- Sound. Two loudspeakers are run by one signal generator. As you walk across in front of them, the sound gets loud and quiet in turn. Echoes from walls can fill in the quiet spots.
- Microwaves. One transmitter faces a metal sheet with two slits. A probe moved across the far side finds maxima and minima.
- Light. One laser shines on two narrow slits. Bright and dark fringes appear on a screen, as in the double-slit experiment.
One driver feeds both sources
Water: two dippers on one drive
This is a snapshot of crests and troughs. Watch it over time to see the fixed lines of still water and of strong ripples.
Microwaves: one transmitter, two slits
Keep the frequency and the direction of the probe fixed. Turning the probe changes its reading for a different reason.
Conditions for seeing fringes
- The sources must be coherent. In practice, one source is split into two, such as one laser and two slits.
- The waves must overlap. Each slit must be narrow enough to diffract the waves.
- For light and other transverse waves, the two waves should be polarised in the same plane.
- The amplitudes should be similar, so the dark fringes are really dark.
- The fringes must be far enough apart to see: a small slit separation and a large screen distance help.
Two separate lamps are not coherent. The phase difference between them changes at random, far too fast to see. So the fringes never stay still, and the screen looks evenly lit.
Check your understanding
Two dippers on one bar make a steady interference pattern in a ripple tank. Each dipper is then given its own motor, set to the same frequency, and the pattern drifts. Why does it drift?
Show the answer
The phase difference between the two dippers no longer stays constant.
Two separate motors never stay exactly in step. As the phase difference wanders, the lines of still water move, so the sources are no longer coherent.
A real example. Thomas Young used water waves to explain interference. His published lectures included a sketch of the pattern made by two sources of water waves.
Coherent does not mean in phase. Sources with a fixed phase difference, even antiphase, are coherent and give a steady pattern.
08
The double-slit experiment
Make a guess. You are not marked.
In a double-slit experiment with red light, the two slits are replaced by a pair that are closer together. Nothing else changes. What happens to the bright fringes on the screen?
Show the answer
They move further apart, because fringe spacing is inversely proportional to slit separation.
Fringe spacing x = lambda D / a. A smaller slit separation a gives a larger spacing, so the fringes spread out across the screen.
In a double-slit experiment, the fringe separation x depends on the wavelength λ, the slit separation a and the screen distance D: λ = ax/D.
A laser shines on two narrow slits close together. Light diffracts from each slit, and the two beams overlap on a screen. They make evenly spaced bright and dark fringes.
The quantities
- a: the slit separation, from the centre of one slit to the centre of the other.
- D: the distance from the slits to the screen, at right angles.
- x: the fringe separation, from the centre of one bright fringe to the next.
- y: a position on the screen, measured from the central bright fringe.
When D is much larger than a and y, the path difference at position y is about ay/D. Neighbouring bright fringes differ by one wavelength of path, so:
λ = ax/D
Worked example
From fringes to wavelength
The slit separation is 0.30 mm and the screen is 2.0 m away. Nine bright fringes in a row span 32 mm.
The slits and the fringes are measured in different places
One laser, two slits, one screen
The screen, enlarged
Divide the span by eight gaps, not by nine fringes. The fringe separation x is on the screen; the slit separation a is at the slits.
λ = ax/D = (0.30 × 10-3)(4.0 × 10-3)/2.0
= 6.0 × 10-7 m = 600 nm
Nine fringes have eight gaps between them. At y = 10 mm, the path difference is (0.30 × 10-3)(0.010)/2.0 = 1.5 × 10-6 m. That is 2.5 wavelengths, so this point is dark.
Try it step by step
Work it out in steps: Slit separation a in m, then Fringe spacing x = lambda D / a in mm.
Show the answer
Problem 1.
Slit separation a in m: a = 0.50 mm = 5.0 x 10-4 m
Fringe spacing x = lambda D / a in mm: x = 6.0 x 10-7 x 2.0 / (5.0 x 10-4) = 2.4 x 10-3 m = 2.4 mm
Problem 2.
Slit separation a in m: a = 0.25 mm = 2.5 x 10-4 m
Fringe spacing x = lambda D / a in mm: x = 4.5 x 10-7 x 1.5 / (2.5 x 10-4) = 2.7 x 10-3 m = 2.7 mm
Problem 3.
Slit separation a in m: a = 0.40 mm = 4.0 x 10-4 m
Fringe spacing x = lambda D / a in mm: x = 6.5 x 10-7 x 3.0 / (4.0 x 10-4) = 4.9 x 10-3 m = 4.9 mm (2 sf)
Check your understanding
Nine consecutive bright centres span 32 mm in a double-slit pattern. Slit separation is 0.30 mm and perpendicular screen distance is 2.0 m. Under the small-angle approximation, what is the wavelength?
Show the answer
600 nm.
The eight intervals give x = 4.0 x 10-3 m. Then lambda = ax/D = (0.30 x 10-3)(4.0 x 10-3)/2.0 = 6.0 x 10-7 m.
A real example. In 1803 Thomas Young described an experiment with two slits. Its fringes were strong evidence that light is a wave. Many school labs now use a red helium-neon laser, with a wavelength of 632.8 nm. NIST lists this neon line at 632.816 nm in air; exam questions often round it to 633 nm.
Making each slit narrower does not change the fringe separation, which depends on a. The fringes get dimmer. Do not mix up the slit width with the slit separation.
09
The diffraction grating
A diffraction grating sends light only in a few sharp directions, given by a sin θ = nλ.
A diffraction grating is a plate with many parallel slits, evenly spaced, often hundreds to the millimetre. Light diffracts at each slit. In most directions the waves from all the slits cancel. In a few directions they all arrive in phase and add to give a bright principal maximum.
The grating equation
Let light hit the grating at right angles. Waves from neighbouring slits that leave at angle θ have a path difference a sin θ, where a is the slit separation. For a maximum this must be a whole number of wavelengths:
The number n is the order. Order 0 goes straight through, and the other orders come in pairs, one on each side. Measure θ from the straight-through direction, not from the grating surface.
Path difference and the allowed orders
Path difference between neighbouring slits
Seven allowed directions
Worked example
How many orders?
A grating has 600 lines per mm. Light of wavelength 500 nm hits it at right angles.
sin θ = nλ/a = 0.300n
| Order n | sin θ | θ / ° |
|---|---|---|
| 0 | 0 | 0 |
| 1 | 0.300 | 17.46 |
| 2 | 0.600 | 36.87 |
| 3 | 0.900 | 64.16 |
sin θ cannot be more than 1, so n can be at most a/λ = 3.33. Round down: the highest order is 3. Order 4 would need sin θ = 1.2, which is impossible. Counting both sides and the centre gives seven directions: 0, ±1, ±2 and ±3.
Try it step by step
Work it out in steps: Slit spacing d = 1 / (lines per metre) in micrometres, then sin theta = n lambda / d for the second order, then Angle theta in degrees.
Show the answer
Problem 1.
Slit spacing d = 1 / (lines per metre) in micrometres: d = 1 / 300 mm = 3.33 x 10-6 m = 3.33 micrometres
sin theta = n lambda / d for the second order: sin theta = 2 x 600 x 10-9 / (3.33 x 10-6) = 0.360
Angle theta in degrees: theta = inverse sin of 0.360 = 21.1 degrees (3 sf)
Problem 2.
Slit spacing d = 1 / (lines per metre) in micrometres: d = 1 / 500 mm = 2.00 x 10-6 m = 2.00 micrometres
sin theta = n lambda / d for the second order: sin theta = 2 x 500 x 10-9 / (2.00 x 10-6) = 0.500
Angle theta in degrees: theta = inverse sin of 0.500 = 30.0 degrees (3 sf)
Problem 3.
Slit spacing d = 1 / (lines per metre) in micrometres: d = 1 / 400 mm = 2.50 x 10-6 m = 2.50 micrometres
sin theta = n lambda / d for the second order: sin theta = 2 x 650 x 10-9 / (2.50 x 10-6) = 0.520
Angle theta in degrees: theta = inverse sin of 0.520 = 31.3 degrees (3 sf)
Check your understanding
Normally incident 500 nm light falls on a grating with 600 lines/mm. What is the highest geometrically allowed principal order and the total number of allowed order directions?
Show the answer
Highest order 3; seven directions in total.
The sine bound gives n no greater than 3.33, hence integer orders through 3. Counting zero and both sides gives 1 + 2 x 3 = 7 directions. This does not imply equal brightness.
A real example. A CD acts like a grating. Its data track, in the format Philips and Sony launched in Japan in October 1982, is a tight spiral with a pitch of 1.6 µm. That is 625 tracks per mm. Each colour has its maxima at its own angle, so white light reflects off the disc as a spectrum. In first order, violet light at 400 nm leaves at 14° and red light at 700 nm at 26°.
Lines per mm is not the slit separation. Turn it into a first: 600 lines per mm gives a = 1/600 mm = 1.67 µm.
10
Measuring wavelength with a grating
Make a guess. You are not marked.
A grating with 300 lines per mm is replaced by one with 600 lines per mm. The laser stays the same. What happens to the first-order spots?
Show the answer
They move further from the centre, because the slits are closer together.
More lines per mm means a smaller slit separation a. From sin theta = lambda/a, a smaller a gives a larger angle.
To measure a wavelength with a grating, measure the angle of a maximum and use λ = a sin θ/n.
The method
- Shine light of one wavelength, such as a laser or a sodium lamp, onto a grating at right angles. You need to know its number of lines per mm.
- Find the central bright maximum, order 0.
- Measure the angle between the two first-order maxima, one on each side, and halve it to get θ.
- Work out a from the lines per mm, then λ = a sin θ/n.
Measuring both sides and halving reduces the error if the grating does not quite face the light. A higher order gives a larger angle, so the same reading error is a smaller fraction of it.
Using a screen
With a screen a distance D from the grating, the first-order spot is a distance y from the centre. Then y/D is tan θ, not sin θ. Find θ first, then take its sine.
Measuring on a screen
First-order spots on a screen 1.000 m away
Find θ = tan-1(0.3145/1.000) first, then use λ = a sin θ. Measuring both spots checks that the grating faces the beam squarely.
Worked example
From a screen to a wavelength
The grating has 600 lines per mm, so a = 1.667 × 10-6 m. D = 1.000 m and y = 0.3145 m.
λ = a sin θ/n = 1.667 × 10-6 × sin 17.46°
= 5.00 × 10-7 m = 500 nm
Using sin θ = y/D instead would give 1.667 × 10-6 × 0.3145 = 524 nm, about 5% too big.
Check your understanding
A grating with 500 lines per mm is 1.00 m from a screen. The first-order spots are 0.30 m either side of the centre. What is the wavelength?
Show the answer
575 nm
tan theta = 0.30/1.00, so theta = 16.7 degrees and sin theta = 0.2873. With a = 1/500 000 = 2.00 x 10-6 m, lambda = a sin theta = 2.00 x 10-6 x 0.2873 = 5.75 x 10-7 m.
A real example. A sodium lamp is a common source in school labs. NIST lists its two brightest lines at 588.995 nm and 589.592 nm, and school data round them to 589 nm. Take a 300 lines per mm grating. Then sin θ = 589 × 10-9 × 300 000 = 0.177, so the first order is at 10.2°.
Measure θ from the centre. If you measure from one first-order spot to the other, halve the angle first.
11
Diffraction
Make a guess. You are not marked.
Ripples pass through a gap 3 cm wide. The dipper frequency is lowered, so the wavelength gets longer. The gap stays the same. What happens to the waves beyond the gap?
Show the answer
They spread out more, as the wavelength is now closer to the gap size.
Diffraction is greatest when the gap is comparable to the wavelength. A longer wavelength through the same gap spreads out more.
Diffraction is the spreading of waves through a gap or round an edge. It is greatest when the gap is comparable to the wavelength.
Diffraction in a ripple tank
Send straight ripples at a gap in a barrier. Keep the depth and the frequency the same, so the wavelength does not change. Then compare a wide gap with a narrow one.
- Wide gap, many wavelengths across: the waves go mostly straight through and spread only a little at the edges.
- Narrow gap, about one wavelength across: the waves spread out in arcs into the space behind the barrier.
The wavelength is the same before and after the gap. Only the direction of travel changes.
Wide and narrow gaps, same wavelength
Both tanks use the same scale and the same wavelength. Each blue line is a wavefront.
Wide gap: 5 wavelengths
Narrow gap: 1 wavelength
Sound spreads round an edge
A narrow gap spreads waves more, but it does not let more energy through. A wide gap still spreads a little at its edges.
Sound round corners
You can hear someone talking round a corner without seeing them. The sounds of speech have wavelengths from tens of centimetres to a few metres, similar to a doorway. So sound diffracts strongly through it. Light has a wavelength about a million times smaller, so it hardly spreads.
A loudspeaker also diffracts the sound it sends out. Low notes have long wavelengths and spread widely. High notes spread less, so they sound quieter off to one side.
Check your understanding
You can hear a radio through an open doorway, but you cannot see it. Which statement explains this?
Show the answer
Sound's wavelength is near the doorway's width; light's is far smaller.
Diffraction is large when the gap is comparable to the wavelength. Sound wavelengths are tens of centimetres to metres. Light's is under a micrometre, so it barely spreads through a doorway.
In an exam, say the gap is "comparable to" the wavelength, not "equal to" it. And a narrow gap spreads the waves more without letting more energy through.
12
The single-slit pattern
Make a guess. You are not marked.
Light passes through a single narrow slit onto a screen. The slit is made narrower. What happens to the central bright band?
Show the answer
It becomes wider, because diffraction increases as the slit narrows.
The first minimum is at sin theta = lambda / b. A smaller slit width b gives a larger angle, so the central band spreads out, though it becomes dimmer.
Light through one narrow slit makes a wide, bright central band. Its edges are the first minima, where b sin θ = λ.
Shine a laser through a single slit of width b. On a screen you see a bright central band, with weaker, narrower bands on each side.
The first minima
The first dark minimum on each side of the centre is at angle θ, where:
This equation gives a minimum, not a maximum. The two first minima are the edges of the central band, so its full width is twice the distance from the centre to one minimum.
Worked example
The width of the central band
A slit of width b = 0.15 mm is lit by light of wavelength 600 nm. The screen is D = 2.0 m away.
Centre to first minimum: y = D tan θ ≈ 2.0 × 0.0040 = 0.0080 m = 8.0 mm
Full width = 2 × 8.0 = 16 mm
The angle is small, so tan θ is almost the same as sin θ.
The first minima mark the edges of the central band
For b = 0.15 mm, λ = 600 nm and D = 2.0 m. The graph shows brightness as a fraction of the centre's, not displacement.
The 8.0 mm is from the centre to one minimum. The whole central band is 16 mm wide.
Check your understanding
A single slit of width 0.15 mm is illuminated by 600 nm light. A screen is 2.0 m away. The first minimum is about 8.0 mm from the centre on each side. What is the central maximum's full width?
Show the answer
16 mm.
The central maximum extends from the first minimum at -8.0 mm to the first minimum at +8.0 mm: a total width of 16 mm.
Changing the slit or the light
From sin θ = λ/b, a narrower slit gives a larger angle. Halving b roughly doubles the width of the central band. Less light gets through, so the band is also dimmer. A shorter wavelength makes the band narrower.
b sin θ = λ looks like the grating equation, but it gives a dark minimum, not a bright one. And b is the width of one slit, not the distance between two slits.
13
Resolving two sources
Make a guess. You are not marked.
Two close stars look like a single blur through a telescope. Which change could let the telescope resolve them?
Show the answer
A wider aperture, which reduces the diffraction spread of each image.
The smallest angle that can be resolved is about lambda / b. A larger aperture b narrows each star's diffraction pattern, so the two images can separate.
Diffraction blurs each point of light into a small patch. Two sources can just be told apart when their angular separation is about λ/b.
Light entering a telescope, a camera or your eye diffracts at the aperture, the opening it passes through. So even a perfect lens makes each point of light into a small diffraction pattern. When the patterns of two sources overlap too much, they look like one.
The Rayleigh criterion
The Rayleigh criterion gives the limit. Two sources are just resolved when the central maximum of one diffraction pattern falls on the first minimum of the other. For an aperture of width b, the smallest angle that can be resolved is:
A smaller θ means better resolution. A wider aperture or a shorter wavelength gives it.
Worked example
Is the aperture wide enough?
Two distant lamps give light of wavelength 550 nm. They are 0.20 mrad apart as seen from the aperture.
The lamps are 0.20 mrad apart, less than 0.275 mrad, so they are not resolved.
Now 0.20 mrad is more than the limit, so the lamps are resolved. The aperture that just resolves them is b = λ/θ = (550 × 10-9)/(0.20 × 10-3) = 2.75 mm.
Same two sources, three aperture widths
The two sources stay at -0.10 and +0.10 mrad, 0.20 mrad apart. Brown long dashes and green short dashes are the pattern of each source; solid blue is the total brightness.
Each pattern is drawn with the same peak height, so that the widths can be compared.
b = 2.0 mm: Not resolved
Each first minimum is 0.2750 mrad from its own source. The sources are still 0.20 mrad apart.
b = 2.75 mm: Just resolved
Each first minimum is 0.2000 mrad from its own source. The brown square marks one pattern's first minimum, right under the other's peak.
b = 4.0 mm: Resolved
Each first minimum is 0.1375 mrad from its own source. The sources are still 0.20 mrad apart.
Check your understanding
Two equal incoherent sources of wavelength 550 nm have angular separation 0.20 mrad. Using the stated Rayleigh criterion theta approximately lambda/b, what aperture width just resolves them?
Show the answer
2.75 mm.
At the limit, b = lambda/theta = (5.50 x 10-7)/(2.0 x 10-4) = 2.75 x 10-3 m. Widening the aperture narrows the diffraction pattern; it does not change source separation.
A real example. NASA's Hubble Space Telescope has a main mirror 2.4 m across. For 550 nm light, λ/b = (550 × 10-9)/2.4 = 2.3 × 10-7 rad. NASA says Hubble can pick out objects just 0.05 arcsecond across in visible light. That is 2.4 × 10-7 rad, close to λ/b.
A smaller aperture gives a larger θ, so poorer resolution. Link the two in your answer, and keep θ in radians.
14
Revision summary
Match each equation to its pattern. A standing-wave gap, a fringe separation and a diffraction minimum measure different things.
Superposition and standing waves
- Superposition: add the displacements at the same point and instant, with their signs.
- Standing waves: two waves with the same frequency, speed and wavelength pass through each other in opposite directions. Nodes never move. Antinodes move the most.
- Strings and microwaves: neighbouring nodes or minima are λ/2 apart. Count gaps, not nodes.
- Air columns: at a closed end the air cannot move, so it is a displacement node. An open end is a displacement antinode.
- Resonance tube: two resonances in a row differ in length by λ/2. Taking one from the other cancels the end correction.
Node to node = λ/2; node to antinode = λ/4
λ = 2(L2 - L1); v = fλ
Interference
- Coherence and path difference: coherent means the same frequency and a fixed phase difference. For sources in phase, a path difference of nλ gives a maximum and (n + ½)λ a minimum. Sources in antiphase swap these.
- Experiments: water, sound, microwaves and light. To see fringes, the sources must be coherent and the waves must overlap. For light, they need the same polarisation. Similar amplitudes give dark minima.
- Double slit: a is the slit separation, x the fringe separation and D the screen distance. Nine fringes have eight gaps.
Gratings, diffraction and resolution
- Grating: a sin θ = nλ gives the maxima. Turn lines per mm into a first. For the highest order, round a/λ down, then count both sides and the centre.
- Measuring with a grating: halve the angle between matching orders. On a screen, y/D is tan θ.
- Diffraction: spreading through a gap or round an edge, greatest when the gap is comparable to λ.
- Single slit: b sin θ = λ gives the first minima. The central band is twice as wide as the centre-to-minimum distance.
- Resolution: just resolved when one central maximum falls on the other's first minimum, at θ ≈ λ/b in radians.
Symbols that are easy to mix up
| Symbol | Meaning | Unit |
|---|---|---|
| a | Slit separation (double slit or grating) | m |
| b | Width of one slit or aperture | m |
| x | Fringe separation on the screen | m |
| y | Position on the screen from the centre | m |
| D | Distance from the slits to the screen | m |
| n | Order of a grating maximum | no unit |
| θ | Angle from the straight-through direction | rad, or degrees for trigonometry |
| e | End correction of an air column | m |
Practise exam questions
Write your answer first, then tick the mark points it makes.
Exam-style question
State three differences between a stationary (standing) wave and a progressive wave. [3]
Saved on this browser only. Anyone using this browser can see it.
Exam-style question
State the principle of superposition. [2]
Saved on this browser only. Anyone using this browser can see it.
Exam-style question
Light of wavelength 633 nm passes through two narrow slits 0.50 mm apart. Fringes are seen on a screen 2.0 m away. Calculate the fringe separation. [3]
Saved on this browser only. Anyone using this browser can see it.
Exam-style question
A string is fixed at one end and vibrated at the other. Explain how a stationary wave forms on the string. [3]
Saved on this browser only. Anyone using this browser can see it.
Exam-style question
A graph of displacement against distance along a string shows a stationary wave at one instant. Adjacent nodes are 0.30 m apart. The vibrator frequency is 50 Hz. Find the speed of the progressive waves on the string. [3]
Saved on this browser only. Anyone using this browser can see it.
Quick recall
At one position and instant, two overlapping string pulses contribute displacements of +3.0 mm and -2.0 mm. What is the resultant displacement in the linear model?