Skip to content

Coherence | Physics - Wyatt's Notes

Coherence time τc\tau_c: the time over which the wave maintains a well-defined phase.

Coherence length: Lc=cτcL_c = c\tau_c.

For a source with spectral width Δν\Delta\nu:

τc1Δν,LccΔν=λ2Δλ\tau_c \approx \frac{1}{\Delta\nu}, \quad L_c \approx \frac{c}{\Delta\nu} = \frac{\lambda^2}{\Delta\lambda}

A sodium lamp (Δλ0.6\Delta\lambda \approx 0.6 nm at λ=589\lambda = 589 nm) has Lc0.6L_c \approx 0.6 mm. A laser (Δλ106\Delta\lambda \approx 10^{-6} nm) has Lc300L_c \approx 300 m.

Worked Example: Coherence length and fringe visibility

Problem. A mercury lamp emits light at λ=546.1\lambda = 546.1 nm with a spectral width Δλ=0.025\Delta\lambda = 0.025 nm. (a) Find the coherence length. (b) In a Michelson interferometer, at what Path difference does the fringe visibility drop to 1/e1/e? (c) How many fringes are visible before they Wash out?

Solution.

(a) Lc=λ2/Δλ=(546.1×109)2/(0.025×109)=1.19×102L_c = \lambda^2/\Delta\lambda = (546.1 \times 10^{-9})^2/(0.025 \times 10^{-9}) = 1.19 \times 10^{-2} m =11.9= 11.9 mm.

(b) For a Gaussian spectrum, visibility drops to 1/e1/e when Δx=Lc=11.9\Delta x = L_c = 11.9 mm.

(c) The number of fringes: Nfringes=Lc/λ=(11.9×103)/(546.1×109)=2.18×104N_{\mathrm{fringes} = L_c/\lambda = (11.9 \times 10^{-3})/(546.1 \times 10^{-9}) = 2.18 \times 10^4}. Over 20000 fringes are visible — a large number, but far fewer than for a laser.

The van Cittert-Zernike theorem states that the spatial coherence of light from an extended Incoherent source is given by the Fourier transform of the source intensity distribution.

For a circular source of angular diameter θs\theta_sThe transverse coherence length is:

lc1.22λθsl_c \approx \frac{1.22\lambda}{\theta_s}

The mutual coherence function quantifies the correlation between the wave field at two spacetime points:

Γ12(τ)=E(r1,t)E(r2,t+τ)\Gamma_{12}(\tau) = \langle E^*(r_1, t) E(r_2, t + \tau) \rangle

The complex degree of coherence is the normalised quantity:

γ12(τ)=Γ12(τ)Γ11(0)Γ22(0)\gamma_{12}(\tau) = \frac{\Gamma_{12}(\tau)}{\sqrt{\Gamma_{11}(0) \Gamma_{22}(0)}}

For quasi-monochromatic light, the visibility of interference fringes equals γ12(τ)|\gamma_{12}(\tau)|. Fringes are visible when 0<γ10 < |\gamma| \leq 1, with γ=1|\gamma| = 1 for perfectly coherent light and γ=0|\gamma| = 0 for incoherent light.

8.4 First-Order Coherence and the Wiener-Khinchin Theorem

Section titled “8.4 First-Order Coherence and the Wiener-Khinchin Theorem”

The Wiener-Khinchin theorem relates the power spectral density S(ω)S(\omega) of a stationary random process to the autocorrelation function via Fourier transform:

Γ11(τ)=S(ω)eiωτdω\Gamma_{11}(\tau) = \int_{-\infty}^{\infty} S(\omega) e^{-i\omega\tau} d\omega

The coherence time is inversely related to the spectral width: τc=γ11(τ)2dτ\tau_c = \int_{-\infty}^{\infty} |\gamma_{11}(\tau)|^2 d\tau. For a Lorentzian line shape, τc=1/(πΔν)\tau_c = 1/(\pi\Delta\nu).

8.5 Partial Coherence and the Wolf Equations

Section titled “8.5 Partial Coherence and the Wolf Equations”

Partially coherent light is described by the cross-spectral density function W(r1,r2,ω)W(r_1, r_2, \omega), which is the Fourier transform of the mutual coherence function:

W(r1,r2,ω)=12πΓ(r1,r2,τ)eiωτdτW(r_1, r_2, \omega) = \frac{1}{2\pi} \int_{-\infty}^{\infty} \Gamma(r_1, r_2, \tau) e^{i\omega\tau} d\tau

The Wolf equations govern the propagation of the cross-spectral density, generalising the Helmholtz equation to partially coherent fields.

8.6 Second-Order Coherence and Photon Bunching

Section titled “8.6 Second-Order Coherence and Photon Bunching”

Second-order coherence measures intensity correlations:

g(2)(τ)=I(t)I(t+τ)I(t)2g^{(2)}(\tau) = \frac{\langle I(t) I(t+\tau) \rangle}{\langle I(t) \rangle^2}

  • Thermal light (chaotic): g(2)(0)=2g^{(2)}(0) = 2, exhibiting photon bunching.
  • Coherent light (laser): g(2)(τ)=1g^{(2)}(\tau) = 1 for all τ\tau.
  • Antibunched light (single-photon source): g(2)(0)<1g^{(2)}(0) < 1.

The Hanbury Brown-Twiss (HBT) interferometer measures g(2)(τ)g^{(2)}(\tau) and was originally used to measure the angular diameter of stars. The HBT effect demonstrated that intensity correlations contain information about source size even when first-order coherence is absent.

8.7 Worked Example: Fringe Visibility of Two Spectral Lines

Section titled “8.7 Worked Example: Fringe Visibility of Two Spectral Lines”

Problem. A source emits two equal-intensity spectral lines at λ1\lambda_1 and λ2\lambda_2 with Δλ=λ2λ1λ\Delta\lambda = |\lambda_2 - \lambda_1| \ll \lambda. Find the fringe visibility in a Michelson interferometer as a function of path difference.

Solution

The interference pattern is the sum of patterns from each line:

I=I0[2+cos ⁣(2πΔxλ1)+cos ⁣(2πΔxλ2)]I = I_0\left[2 + \cos\!\left(\frac{2\pi\Delta x}{\lambda_1}\right) + \cos\!\left(\frac{2\pi\Delta x}{\lambda_2}\right)\right]

Using the sum-to-product identity:

I=2I0[1+cos ⁣(πΔxλ1+πΔxλ2)cos ⁣(πΔxλ1πΔxλ2)]I = 2I_0\left[1 + \cos\!\left(\frac{\pi\Delta x}{\lambda_1} + \frac{\pi\Delta x}{\lambda_2}\right) \cos\!\left(\frac{\pi\Delta x}{\lambda_1} - \frac{\pi\Delta x}{\lambda_2}\right)\right]

For Δλλ\Delta\lambda \ll \lambda, this becomes:

I=2I0[1+cos ⁣(2πΔxλˉ)cos ⁣(πΔxΔλλˉ2)]I = 2I_0\left[1 + \cos\!\left(\frac{2\pi\Delta x}{\bar{\lambda}}\right) \cos\!\left(\frac{\pi\Delta x\,\Delta\lambda}{\bar{\lambda}^2}\right)\right]

where λˉ\bar{\lambda} is the mean wavelength. The fringe visibility is cos(πΔxΔλ/λˉ2)|\cos(\pi\Delta x\,\Delta\lambda/\bar{\lambda}^2)|, which drops to zero when Δx=λˉ2/(2Δλ)\Delta x = \bar{\lambda}^2/(2\Delta\lambda). This is the coherence length for a two-line source.

\blacksquare

8.8 Worked Example: Michelson Stellar Interferometer

Section titled “8.8 Worked Example: Michelson Stellar Interferometer”

Two separated mirrors direct light from a distant star into a single telescope. Fringes are observed When the mirror separation dd satisfies:

d<1.22λθsd \lt \frac{1.22\lambda}{\theta_s}

The first disappearance of fringes gives the angular diameter of the star: θs=1.22λ/d\theta_s = 1.22\lambda/d.

8.9 Coherence of Laser Light vs Thermal Light

Section titled “8.9 Coherence of Laser Light vs Thermal Light”
PropertyLaserThermal source
Spectral width Δν\Delta\nu1\sim 1 MHz or less1014\sim 10^{14} Hz
Coherence time τc\tau_c1 μ\sim 1\ \mus or more1014\sim 10^{-14} s
Coherence length LcL_c300\sim 300 m or more1 μ\sim 1\ \mum
Spatial coherenceFull (across beam)Limited by van Cittert-Zernike
g(2)(0)g^{(2)}(0)11 (coherent)22 (chaotic)

8.10 Worked Example: Spatial Coherence of Sunlight

Section titled “8.10 Worked Example: Spatial Coherence of Sunlight”

Problem. The sun has an angular diameter of approximately 0.530.53^\circ as seen from Earth. What is the transverse coherence length of sunlight at λ=550\lambda = 550 nm?

Solution

The angular diameter in radians: θs=0.53×π/1809.25×103\theta_s = 0.53^\circ \times \pi/180 \approx 9.25 \times 10^{-3} rad.

Using the van Cittert-Zernike theorem for a circular source:

lc1.22λθs=1.22×550×1099.25×1037.3×105 m73 μml_c \approx \frac{1.22\lambda}{\theta_s} = \frac{1.22 \times 550 \times 10^{-9}}{9.25 \times 10^{-3}} \approx 7.3 \times 10^{-5}\ \text{m} \approx 73\ \mu\text{m}

This means sunlight is coherent over a distance of about 73 μ73\ \mum transverse to the propagation direction. Two pinholes spaced closer than this will produce visible interference fringes.

\blacksquare

flowchart TD
A[8_Coherence] --> B[Key Concepts]
A --> C[Core Principles]
A --> D[Practical Applications]
B --> E[Fundamental definitions]
C --> F[Design patterns]
D --> G[Real-world usage]

Coherence is the predictability of a wave’s phase over time and space. A laser maintains phase relationships for microseconds, allowing interference over meters. A light bulb’s atoms emit independently, so phases randomize in femtoseconds, limiting coherence to micrometers. The coherence length is the maximum path difference where interference fringes remain visible, like the distance over which two runners stay in step. Spatial coherence depends on source size: a point source is perfectly coherent across space, while an extended source like the sun has limited spatial coherence. The van Cittert-Zernike theorem connects source angular size to coherence area.

Mistake 1: Assuming g(2)(0)=2g^{(2)}(0) = 2 means thermal light has “twice the coherence” The second-order correlation function g(2)(0)=2g^{(2)}(0) = 2 for thermal light reflects photon bunching — photons from a chaotic source tend to arrive in pairs. This is an intensity correlation effect, not a measure of coherence in the traditional sense. Students sometimes interpret g(2)(0)>1g^{(2)}(0) > 1 as enhanced coherence, when it actually indicates the opposite: the light is more chaotic.

Mistake 2: Confusing the coherence time with the correlation time of the source The coherence time τc1/Δν\tau_c \approx 1/\Delta\nu characterises how long the field maintains a definite phase, while the source correlation time relates to the atomic emission process. For a laser, τc\tau_c can be microseconds or longer despite atomic transitions occurring on nanosecond timescales. Students often assume these timescales must be comparable.

  • Diffraction: The formation of diffraction fringes depends on the coherence of the source; partially coherent light produces reduced-visibility fringes.
  • Fourier Optics: Uses the Wiener-Khinchin theorem to relate the power spectrum of the source to the mutual coherence function via Fourier transforms.
  • Polarization: The coherence matrix extends to vector fields, connecting coherence theory to the polarisation state of partially polarised light.

Mistake 3: Assuming that spatial coherence requires a point source The van Cittert-Zernike theorem shows that even an extended incoherent source produces spatially coherent light over a finite area Acλ2/ΩA_c \approx \lambda^2/\Omega. The coherence area increases as the source becomes more compact. Students sometimes think only lasers or point sources produce spatially coherent light, when in fact any source produces some degree of spatial coherence over sufficiently small transverse distances.

Coherence is the predictability of a wave’s phase over time and space. A laser maintains phase relationships for microseconds, allowing interference over meters. A light bulb’s atoms emit independently, so phases randomize in femtoseconds, limiting coherence to micrometers. The coherence length is the maximum path difference where interference fringes remain visible, like the distance over which two runners stay in step. Spatial coherence depends on source size: a point source is perfectly coherent across space, while an extended source like the sun has limited spatial coherence. The van Cittert-Zernike theorem connects source angular size to coherence area.

This section provides detailed coverage of advanced concepts, including full derivations, proofs, and extended examples.

Complete mathematical derivations and proofs are provided where appropriate. Each step is explained to ensure understanding of the underlying reasoning.

Advanced examples demonstrate the application of concepts to complex problems. These examples go beyond standard exam questions to develop deeper understanding.

This material connects to current research and advanced applications in the field. Understanding these connections provides context for the study material.

Ensure you have mastered the prerequisite material before attempting this advanced content.