Intuition The one core idea
When light hits a thin transparent film it splits into two waves — one bouncing off the top, one off the bottom — and these two waves add up or cancel out depending on how far out of step they got. This whole topic is just careful bookkeeping of "how out of step" using two things: the extra distance the second wave travelled, and a sneaky half-step "hiccup" some reflections give.
Before you can read the parent note Thin film interference — reflected and transmitted , you must own every symbol it throws at you. Below, each symbol is built from plain words → a picture → why the topic needs it , in an order where each one rests on the previous.
Picture a rope you shake up and down. A travelling wave is a repeating pattern of crests (high points) and troughs (low points) moving along.
λ
The distance from one crest to the next crest of a wave. Say it "lambda". For visible light it is tiny — around 400 to 700 nanometres (a nanometre, nm , is a billionth of a metre, 1 0 − 9 m ).
Look at the figure: the horizontal red arrow spans exactly one λ . Notice the green wave : it is an identical wave shifted sideways. Two waves are:
in step (in phase) if their crests line up — they add to make a bigger wave (bright ),
out of step (out of phase) if one's crest sits on the other's trough — they cancel (dark ).
λ is the ruler for everything here
"How out of step are two waves?" only has meaning compared to λ . A shift of one whole λ puts them back in step (crest on crest again). A shift of half a λ puts them exactly out of step. So we always measure extra distances in units of λ . This is the single most important habit of the whole topic.
Cover and recall:
One whole λ of shift means the two waves are... back in step → bright .
Half a λ (λ /2 ) of shift means... exactly out of step → dark .
When two waves overlap and their heights add point-by-point, we call the result interference . This is the beating heart of Interference of light and you meet it first in Young's double slit .
Two waves are coherent if they keep a fixed, unchanging out-of-step relationship over time. Only coherent waves make a stable bright/dark pattern you can actually see.
In a thin film the two rays came from the same original wave (it split), so they are automatically coherent — that is why films show steady colours and don't just flicker into grey.
Two runners who started from the same starting gun stay in a fixed rhythm relationship — that's coherence. Two runners started by two different random guns drift in and out of rhythm — incoherent, no stable pattern.
Now the key measurable quantity.
Definition Path difference
Δ
The extra distance one wave travels compared with the other before they meet again. Symbol Δ (Greek capital "Delta", meaning "difference").
In the figure the lower wave takes a detour and travels an extra length marked Δ . The rule that ties Δ to bright/dark is the entire game:
Intuition Why whole numbers vs half numbers?
If the extra distance is a whole number of wavelengths, the delayed wave has slipped by complete cycles and lands crest-on-crest again — bright. If it's a whole number plus a half , it lands crest-on-trough — dark. The half is what flips bright to dark.
Why does the topic need m ? Because many thicknesses give bright — one for each whole number of wavelengths that fit. m counts which one.
Definition Refractive index
n
A number telling how much a material slows light down compared to vacuum. Air: n ≈ 1.0 . Water: 1.33 . Glass: 1.5 . Higher n = "denser" for light = slower. See Refractive index .
Inside a medium light slows down, so its crests bunch up closer — the wavelength inside shrinks:
λ film = n λ
Look at the figure: same wave, but in the teal medium (n > 1 ) the crests are squeezed together — more crests fit in the same physical distance.
Intuition Optical path length — why we multiply by
n
What actually decides "in step or out" is the number of crests a wave passes through, not the raw distance. Since crests are packed n times tighter inside the film, a physical distance d inside contains n times as many crests as the same d in air. We book-keep this by defining the optical path = n × d . That is why the parent's formula carries an n : it converts real distance into "crest count" so we can compare fairly against the air wavelength λ .
Common mistake "Just use the shrunk wavelength
λ / n everywhere."
Why it feels right: the light really is shorter-waved inside. The fix: the factor n inside 2 n t cos θ r already does that conversion. Keep λ (the air value) on the other side of the equation. Do it once , not twice.
Recall:
Inside a film of index n , the wavelength becomes... λ / n (crests squeeze closer).
"Optical path" of a real distance d inside the film is... n d (real distance × index = crest count).
Light rarely hits straight on. We need to measure directions from the normal — an imaginary line perpendicular to the surface.
Definition Angle of incidence
θ i and refraction θ r
θ i = angle between the incoming ray and the normal (measured outside the film). θ r = angle of the bent ray inside the film. Bending toward the normal happens because light slows.
The rule linking them is Snell's law :
sin θ i = n sin θ r
Intuition Why the topic uses
θ r , not θ i
The extra path is travelled inside the film, where the ray runs at angle θ r . So every geometry step must use the inside angle. θ i is only what you happen to measure with a protractor outside. At straight-on hitting (normal incidence) both angles are 0 , so cos θ r = 1 and the messy angle factor vanishes — that's the simplest case, used in most worked examples.
cos θ — the "how much survives straight down" number
On a right triangle, cos θ = hypotenuse adjacent side . It answers: of a slanted length, how much of it points along the reference direction? When the ray is straight (θ = 0 ) all of it survives: cos 0 = 1 . When fully sideways (θ = 9 0 ∘ ) none survives along the normal: cos 9 0 ∘ = 0 .
Why the topic needs cos θ r : the film thickness t is measured straight across (along the normal), but the ray travels slanted. The projection of the slanted trip onto the normal direction is what counts, and projecting = multiplying by cos θ r . That is exactly why the extra path is 2 t cos θ r and not 2 t / cos θ r .
Definition Film thickness
t
The straight-across depth of the film, from top surface to bottom surface, measured along the normal. Typically tens to hundreds of nanometres — comparable to λ , which is exactly why interference is visible.
The ray goes down and back up , so it crosses the thickness twice — that's where the factor 2 in 2 n t cos θ r comes from: 2 crossings × index n × depth t × the projection cos θ r .
This is the trickiest symbol, so build it slowly. See Phase change on reflection .
Definition Phase change of
π (a half-wavelength flip)
When a wave reflects off a surface going from a ==rarer (lower n ) into a denser (higher n )== medium, the reflected wave gets flipped upside down — crests become troughs. This flip is worth an extra half wavelength, λ /2 . Reflecting the other way (denser → rarer) gives no flip.
In the figure the top wave (rarer→denser) comes back inverted; the bottom one (denser→rarer) comes back the same way up.
Intuition Why this matters so much
A flip is worth λ /2 of "out-of-step-ness" for free , without any distance travelled. This is why a soap film thinner than a wavelength — where the geometric path → 0 — can still look dark : the lone flip alone puts the two waves half a step apart. Miss the flip and every single bright/dark prediction inverts.
"Reflect Rarer → Denser? Phase Reverses." All three R's. Count how many surfaces do this. Odd number of flips → add one λ /2 (conditions swap). Even (0 or 2) → they cancel, no net shift.
Recall:
Reflection from rarer→denser gives a phase flip of... π , i.e. an extra λ /2 .
Two flips (both surfaces) together give a net shift of... zero (they cancel to a full λ ).
Some light bounces back (reflected ), some passes through (transmitted ). By Conservation of energy , light that isn't reflected must be transmitted — so the two patterns are complementary (photographic negatives). Where reflection is bright, transmission is dark. Keep this fact ready; it's why the parent gives two swapped formula boxes.
Wave and wavelength lambda
Interference in step or out
Path difference Delta in units of lambda
Optical path n times distance
Angles theta i and theta r
Cosine projects onto normal
Geometric path 2 t cos theta r
Reflected and transmitted complementary
Read each; only click when you can answer without hesitation.
I can explain what λ is and why we measure shifts in units of it λ = crest-to-crest distance; a shift of λ restores "in step", a shift of λ /2 makes "out of step".
I know what "in step" (bright) and "out of step" (dark) mean In step = crests align, waves add = bright. Out of step = crest on trough, cancel = dark.
I can state the master path-difference rule Δ = mλ → bright; Δ = ( m + 2 1 ) λ → dark, for whole m .
I know what refractive index n does to light and to wavelength Slows light; wavelength shrinks to λ / n ; crests bunch up.
I understand "optical path" and why we multiply distance by n Optical path = n d counts crests fairly against the air wavelength.
I can tell θ i from θ r and know Snell's law θ i outside, θ r inside; sin θ i = n sin θ r ; the topic uses θ r .
I know why cos θ r appears (not 1/ cos θ r ) It projects the slanted in-film travel onto the straight-across normal direction.
I know the phase-flip rule and when it happens Rarer→denser reflection flips by π (=λ /2 ); denser→rarer does not.
I can count flips and decide if conditions swap Odd flips → add λ /2 , swap bright/dark. Even flips → no change.
I know why reflected and transmitted patterns are complementary Energy conservation: unreflected light is transmitted, so the patterns are negatives.