Login

Concepts

Read through concept notes with worked explanations and diagrams, chapter by chapter.

Concept 07 / 07Wave Optics

Polarization: Malus's Law, Brewster's Angle and Polarization by Scattering

Polarization proves light is a transverse wave: a polaroid halves unpolarized light, Malus's law gives I = I0 cos-squared theta, and reflection at tan(theta) = mu produces completely polarized light.

Figure 1
Interference and diffraction establish that light is a wave, but they work just as well for sound, so they cannot tell us whether the oscillation is along the direction of travel or across it. Polarization settles that question: light is a transverse wave, with E⃗\vec{E} (and B⃗\vec{B}) perpendicular to the direction of propagation. Longitudinal waves such as sound can never be polarized, because there is nothing to orient. Unpolarized, polarized and partially polarized light. An ordinary source is a jumble of independent atomic emitters, each radiating a short wave train with its own orientation of E⃗\vec{E} in the plane perpendicular to the ray. Averaged over even a microsecond, all orientations occur equally often: this is unpolarized (natural) light. If E⃗\vec{E} stays along one fixed direction, the light is plane (linearly) polarized; the plane containing E⃗\vec{E} and the ray is called the plane of vibration. If one direction is favoured but not exclusively, the light is partially polarized, and its quality is measured by the degree of polarization P=Imax−IminImax+IminP = \dfrac{I_{max} - I_{min}}{I_{max} + I_{min}} obtained by rotating an analyser. In diagrams, dots on a ray mean E⃗\vec{E} perpendicular to the page and short bars mean E⃗\vec{E} in the plane of the page, as in the diagram above. Polaroids and the factor of one half. A polaroid is a sheet of long aligned molecules that absorbs the component of E⃗\vec{E} along the molecular chains and transmits the perpendicular component; the transmitted direction is the transmission axis. Send unpolarized light of intensity I0I_0 into it. A wave whose field makes an angle θ\theta with the axis contributes I0cos⁡2θI_0\cos^2\theta, and since all θ\theta are equally likely we must average: ⟨cos⁡2θ⟩=12\langle\cos^2\theta\rangle = \dfrac{1}{2}. So the emerging intensity is I02\dfrac{I_0}{2}, whatever the orientation of the polaroid, and the emerging light is plane polarized along the axis. This is why rotating a single polaroid in front of an ordinary lamp produces no change in brightness at all, a fact worth remembering because it is the standard test for unpolarized light. Malus's law. Now let already-polarized light of intensity I1I_1, with its field along a direction making an angle θ\theta with the transmission axis of a second polaroid (the analyser), fall on it. Resolve the amplitude E0E_0 into a component E0cos⁡θE_0\cos\theta along the axis, which is transmitted, and E0sin⁡θE_0\sin\theta perpendicular to it, which is absorbed. Since I∝E2I \propto E^2, I=I1cos⁡2θI = I_1\cos^2\theta. That is Malus's law. Two important corollaries: with crossed polaroids (θ=90∘\theta = 90^\circ) the output is zero, and for unpolarized light entering a polarizer-analyser pair the final intensity is I02cos⁡2θ\dfrac{I_0}{2}\cos^2\theta. The half applies only once, at the first polaroid; after that the light is already polarized and Malus's law takes over. Identifying a beam with a rotating analyser. Rotate a polaroid through a full turn in the beam. If the intensity drops to zero twice, the light is plane polarized. If it varies between a maximum and a non-zero minimum, it is partially polarized. If it does not vary at all, the light is either unpolarized or circularly polarized, and those two are distinguished by first passing it through a quarter-wave plate, which converts circular into plane polarized light but leaves unpolarized light unpolarized. Polarization by reflection and Brewster's angle. When unpolarized light strikes a transparent surface, the reflected beam is always partially polarized, with the component perpendicular to the plane of incidence reflected more strongly. At one particular angle of incidence the reflected beam becomes 100%100\% plane polarized perpendicular to the plane of incidence. Brewster found experimentally that at this angle the reflected and refracted rays are exactly perpendicular to each other, and that fact gives the law immediately. Let the polarizing angle be θB\theta_B and the refraction angle rr. Since the reflected ray also leaves at θB\theta_B and the reflected and refracted rays are perpendicular, the three angles at the point of incidence satisfy θB+90∘+r=180∘\theta_B + 90^\circ + r = 180^\circ, so r=90∘−θBr = 90^\circ - \theta_B. Substituting into Snell's law, μ=sin⁡θBsin⁡r=sin⁡θBsin⁡(90∘−θB)=sin⁡θBcos⁡θB\mu = \dfrac{\sin\theta_B}{\sin r} = \dfrac{\sin\theta_B}{\sin(90^\circ - \theta_B)} = \dfrac{\sin\theta_B}{\cos\theta_B}, hence tan⁡θB=μ\tan\theta_B = \mu: Brewster's law. For crown glass μ=1.5\mu = 1.5 gives θB=56.3∘\theta_B = 56.3^\circ; for water μ=1.33\mu = 1.33 gives 53.1∘53.1^\circ. Since μ\mu depends on wavelength, so does θB\theta_B, slightly. Why perpendicularity produces complete polarization is worth understanding rather than memorising. The reflected wave is not "bounced off" the surface; it is radiated by the electrons in the medium that the refracted wave sets oscillating. Those electrons oscillate along E⃗\vec{E} of the refracted wave, and an oscillating dipole radiates nothing along its own axis. For the component polarized in the plane of incidence, the electrons oscillate in a direction that, when the reflected and refracted rays are perpendicular, points exactly along the reflected ray. That component therefore cannot be radiated into the reflected beam at all, leaving only the perpendicular component: the reflected light is completely polarized perpendicular to the plane of incidence. A neat consequence: if light enters a glass slab at the Brewster angle, it meets the lower surface at exactly the Brewster angle for the glass-to-air transition, since tan⁡r=tan⁡(90∘−θB)=cot⁡θB=1/μ\tan r = \tan(90^\circ - \theta_B) = \cot\theta_B = 1/\mu. Both faces therefore polarize, which is the basis of a pile-of-plates polarizer. Note also that the transmitted beam at θB\theta_B is only partially polarized, never completely: the surface removes some of one component, not all of the other. Polarization by scattering. When sunlight passes through the atmosphere, molecules absorb and re-radiate it. The scattered field must be perpendicular to the scattered ray, so if you look at right angles to the incident beam, only the component of the molecular oscillation perpendicular to both directions can radiate towards you, and the light you receive is completely plane polarized. Scattered light at 90∘90^\circ from the sun is therefore strongly polarized, as a polaroid held up to a patch of clear blue sky shows dramatically. The same Rayleigh scattering, with its 1/λ41/\lambda^4 dependence, scatters blue about (700400)4≈9\left(\dfrac{700}{400}\right)^4 \approx 9 times more strongly than red, making the sky blue and the setting sun red. Practical uses. Glare reflected from a horizontal surface such as a road or a lake is polarized horizontally, so polaroid sunglasses are made with a vertical transmission axis to block it. Photographic polarizing filters darken skies and cut reflections. Stress patterns in transparent models show up in polarized light (photoelasticity), liquid crystal displays switch polarized light on and off, and sugar solutions rotate the plane of polarization in proportion to concentration (optical activity), which is how sugar content is measured commercially. Key Takeaways and Formulae: Polarization proves light is transverse; unpolarized light contains all orientations of E⃗\vec{E} equally. Unpolarized light through one polaroid: I=I02I = \dfrac{I_0}{2}, output plane polarized, independent of the polaroid's orientation. Malus's law for already polarized light: I=I1cos⁡2θI = I_1\cos^2\theta, θ\theta measured between the vibration direction and the transmission axis. Brewster: tan⁡θB=μ\tan\theta_B = \mu, with θB+r=90∘\theta_B + r = 90^\circ; reflected light is completely polarized perpendicular to the plane of incidence; refracted light is only partially polarized. Brewster angle from denser to rarer is the complement of the one from rarer to denser. Degree of polarization P=Imax−IminImax+IminP = \dfrac{I_{max} - I_{min}}{I_{max} + I_{min}}. Scattered light at 90∘90^\circ to the beam is completely plane polarized; Rayleigh scattering goes as 1λ4\dfrac{1}{\lambda^4}. Common mistake: applying the factor 12\dfrac{1}{2} more than once, or applying Malus's law to unpolarized light. Common mistake: saying the reflected light at θB\theta_B is polarized in the plane of incidence. It is polarized perpendicular to it. Brain Teaser 1: Unpolarized light of intensity I0I_0 falls on two crossed polaroids, so that no light emerges. A third polaroid is now inserted between them with its axis at 45∘45^\circ to the first. Show that light reappears, find its intensity, and find the angle of the middle polaroid that maximises the output. Solution: Step 1. Unpolarized light through the first polaroid: I1=I02I_1 = \dfrac{I_0}{2}, polarized along the first axis. Step 2. Through the middle polaroid at 45∘45^\circ: by Malus's law, I2=I02cos⁡245∘=I02×12=I04I_2 = \dfrac{I_0}{2}\cos^2 45^\circ = \dfrac{I_0}{2}\times\dfrac{1}{2} = \dfrac{I_0}{4}, and, crucially, the light emerging is now polarized along the middle polaroid's axis, not the first one's. Step 3. Through the third polaroid, which is at 90∘90^\circ to the first and therefore at 45∘45^\circ to the middle one: I3=I04cos⁡245∘=I08I_3 = \dfrac{I_0}{4}\cos^2 45^\circ = \dfrac{I_0}{8}. So inserting an extra absorbing sheet increases the transmitted light from zero to I0/8I_0/8. The paradox dissolves once you see that a polaroid does not merely filter, it re-orients: the middle sheet rotates the plane of vibration part way round, so the light arriving at the last sheet is no longer perpendicular to its axis. General angle: with the middle polaroid at θ\theta to the first, the output is I=I02cos⁡2θ cos⁡2(90∘−θ)=I02cos⁡2θsin⁡2θ=I08sin⁡22θI = \dfrac{I_0}{2}\cos^2\theta\,\cos^2(90^\circ - \theta) = \dfrac{I_0}{2}\cos^2\theta\sin^2\theta = \dfrac{I_0}{8}\sin^2 2\theta. This is maximum when sin⁡2θ=1\sin 2\theta = 1, that is θ=45∘\theta = 45^\circ, giving the maximum possible output I08\dfrac{I_0}{8}, confirming the first calculation. It vanishes at θ=0\theta = 0 and θ=90∘\theta = 90^\circ, as it must, since then the middle sheet is parallel to one of the crossed pair and changes nothing. Brain Teaser 2: Light reflected from the surface of a still lake is found to be completely plane polarized. Take μwater=1.33\mu_{water} = 1.33. (a) Find the angle of incidence and the angle of refraction. (b) How high is the sun above the horizon? (c) Should the transmission axis of polaroid sunglasses be vertical or horizontal, and why? Solution: (a) By Brewster's law, tan⁡θB=μ=1.33\tan\theta_B = \mu = 1.33, so θB=tan⁡−1(1.33)=53.1∘\theta_B = \tan^{-1}(1.33) = 53.1^\circ. Since the reflected and refracted rays are perpendicular at the Brewster angle, r=90∘−53.1∘=36.9∘r = 90^\circ - 53.1^\circ = 36.9^\circ. Check with Snell: sin⁡53.1∘/sin⁡36.9∘=0.7997/0.6004=1.33\sin 53.1^\circ/\sin 36.9^\circ = 0.7997/0.6004 = 1.33. Correct. (b) The angle of incidence is measured from the normal, which is vertical for a horizontal water surface. The sunlight arrives at 53.1∘53.1^\circ from the vertical, hence at 90∘−53.1∘=36.9∘90^\circ - 53.1^\circ = 36.9^\circ from the horizontal. The sun is therefore about 36.9∘36.9^\circ above the horizon. (c) The reflected light is polarized perpendicular to the plane of incidence. The plane of incidence here is vertical (it contains the vertical normal and the incoming ray), so "perpendicular to it" means the electric field of the glare oscillates horizontally. To block the glare, the sunglasses must absorb horizontal vibrations, which means their transmission axis must be vertical. That is exactly how polaroid sunglasses are made, and it is why tilting your head sideways while wearing them makes the glare from a wet road reappear. Brain Teaser 3: A beam of partially polarized light is examined with a rotating analyser. The transmitted intensity varies between a maximum of 5I5I and a minimum of 3I3I. Find the degree of polarization, and find what fraction of the incident power is carried by the polarized part of the beam. Solution: Model the beam as a completely unpolarized part of intensity UU plus a completely plane polarized part of intensity PP, so the total incident intensity is U+PU + P. The unpolarized part always contributes U/2U/2 through the analyser, whatever its orientation. The polarized part contributes Pcos⁡2θP\cos^2\theta, which runs from PP down to 00 as the analyser turns. Therefore Imax=U2+P=5II_{max} = \dfrac{U}{2} + P = 5I and Imin=U2+0=3II_{min} = \dfrac{U}{2} + 0 = 3I. From the second equation U=6IU = 6I; substituting into the first, P=5I−3I=2IP = 5I - 3I = 2I. Total intensity =U+P=6I+2I=8I= U + P = 6I + 2I = 8I, so the polarized fraction is PU+P=2I8I=0.25\dfrac{P}{U+P} = \dfrac{2I}{8I} = 0.25, that is 25%25\% of the beam is polarized. Degree of polarization: Pdeg=Imax−IminImax+Imin=5I−3I5I+3I=28=0.25P_{deg} = \dfrac{I_{max} - I_{min}}{I_{max} + I_{min}} = \dfrac{5I - 3I}{5I + 3I} = \dfrac{2}{8} = 0.25. The two numbers agree, and that is not a coincidence: substituting the general expressions gives Imax−IminImax+Imin=PU+P\dfrac{I_{max} - I_{min}}{I_{max} + I_{min}} = \dfrac{P}{U + P} identically, so for this simple model the measured degree of polarization is precisely the polarized fraction of the power. Two useful limits: Imin=0I_{min} = 0 gives Pdeg=1P_{deg} = 1 (fully plane polarized), and Imax=IminI_{max} = I_{min} gives Pdeg=0P_{deg} = 0 (unpolarized).