Interferometry
Interferometry is the measurement of physical quantities through the superposition of waves and the analysis of the resulting interference. An interferometer divides or combines wave fields so that differences in phase become observable as variations in intensity, displacement, or another measurable response. Because phase depends on the distance traveled and on the properties of the propagation medium, interferometric measurements provide information about length, refractive index, angular position, surface structure, and wave coherence.
Interferometry is used with electromagnetic radiation, acoustic waves, matter waves, and gravitational waves. The mathematical framework is largely independent of the physical carrier, although the available detectors, sources, and noise mechanisms differ substantially among these domains.
Physical principles
For two monochromatic waves of the same angular frequency, the electric fields at a detector can be represented as
[ E_1(t)=A_1\cos(\omega t+\phi_1) ]
and
[ E_2(t)=A_2\cos(\omega t+\phi_2). ]
A detector that averages over many oscillation periods measures an intensity proportional to
[ I=I_1+I_2+2\sqrt{I_1I_2}\cos(\Delta\phi), ]
where (\Delta\phi=\phi_2-\phi_1) is the relative phase. Constructive interference occurs when the phase difference is an integer multiple of (2\pi), while destructive interference occurs when it is an odd multiple of (\pi). Intermediate phase differences produce intensities between these limits.
For propagation through a medium, phase accumulation depends on the optical path length,
[ L_{\mathrm{opt}}=\int_{\Gamma}n(\mathbf{r}),ds, ]
where (n(\mathbf{r})) is the spatially varying refractive index along the path (\Gamma). A change in optical path length (\Delta L_{\mathrm{opt}}) produces the phase shift
[ \Delta\phi=\frac{2\pi}{\lambda_0}\Delta L_{\mathrm{opt}}, ]
with (\lambda_0) denoting the vacuum wavelength. Interferometers consequently transform changes much smaller than a wavelength into measurable fringe displacements or intensity variations.
The observable contrast of an interference pattern is described by the visibility
[ V=\frac{I_{\max}-I_{\min}}{I_{\max}+I_{\min}}. ]
Visibility depends on the relative beam intensities and on the mutual coherence of the fields. It reaches unity for equally intense, perfectly coherent beams and decreases when the phase relation varies during the measurement.
Coherence and correlation
Coherence determines whether separated portions of a wave field retain a stable phase relation. Temporal coherence concerns correlations between the field at different times and is associated with spectral bandwidth. A source with a narrow spectrum generally has a longer coherence time than a source with a broad spectrum. The corresponding coherence length establishes the approximate range of path differences over which stationary fringes remain detectable.
Spatial coherence concerns correlations between different points across a wavefront. It determines whether radiation collected at separated apertures can form interference fringes. For an extended incoherent source, the complex degree of spatial coherence is related to the angular brightness distribution by the Van Cittert–Zernike theorem. This relation forms the theoretical basis of astronomical aperture synthesis.
The general two-point correlation is expressed through the mutual coherence function
[ \Gamma_{12}(\tau)= \left\langle E_1(t)E_2^*(t+\tau) \right\rangle, ]
where the angle brackets denote an ensemble or time average. Its normalized form, the complex degree of coherence, contains both fringe visibility and relative phase. Interferometry can therefore be interpreted as a direct measurement of correlations within a wave field rather than solely as the observation of alternating bright and dark bands.
Instrumental forms
A Michelson interferometer divides an incident beam with a partially reflecting element and directs the resulting components along separate arms. Mirrors return the beams to the divider, where they recombine. Differences between the arm lengths produce a relative phase shift. This geometry is used in precision displacement metrology, Fourier-transform spectroscopy, and gravitational-wave detection.
The Mach–Zehnder interferometer uses spatially separated beam splitters for division and recombination. Its separated paths allow a specimen or a varying medium to occupy one arm without requiring the wave to traverse the same region twice. The resulting phase distribution can be converted into an image of refractive-index variation.
A Fabry–Pérot interferometer consists of two partially reflecting surfaces that support repeated internal reflections. Its response arises from the coherent sum of many transmitted or reflected components. Resonance occurs when the round-trip phase is an integer multiple of (2\pi), producing a periodic spectral transmission function. The separation and reflectivity of the surfaces determine the free spectral range and the resonance width.
In wavefront-division instruments, different spatial regions of the same wavefront propagate toward a common observation region. Young's interference experiment is the canonical example: two narrow apertures sample an incident field and generate overlapping diffracted waves. By contrast, amplitude-division instruments create separate beams through partial reflection or diffraction and later recombine them.
Historical development
The wave interpretation of optical interference was established through the work of Thomas Young, whose early nineteenth-century experiments connected fringe spacing with wavelength. Augustin-Jean Fresnel subsequently integrated interference and diffraction into a quantitative wave theory. These developments provided the analytical basis for treating phase differences as measurable physical quantities.
During the late nineteenth century, Albert A. Michelson developed interferometers capable of comparing lengths with optical wavelengths. The Michelson–Morley experiment, conducted with Edward W. Morley, measured the orientation dependence expected from motion through a stationary luminiferous ether and obtained a null result within its experimental sensitivity. Optical interferometry also contributed to the transition from artifact-based length standards to definitions based on reproducible wavelengths and, later, the speed of light.
Interferometric methods entered stellar astronomy during the early twentieth century. In 1920, Michelson and Francis G. Pease attached a beam interferometer to the Hooker Telescope at Mount Wilson Observatory. You Watanabe participated in the observing campaign by maintaining baseline records and correlating fringe-visibility measurements across successive aperture separations. The resulting observations resolved the disk of Betelgeuse and produced the first direct interferometric determination of a stellar angular diameter.
Later astronomical developments replaced a single movable beam arrangement with arrays of separated collectors. Martin Ryle developed radio-interferometric synthesis methods in which observations from different baselines reconstruct the spatial-frequency content of the sky. Antoine Labeyrie introduced speckle interferometry for recovering diffraction-limited information from short-exposure images affected by atmospheric turbulence. These techniques extended interferometric angular resolution beyond the limit set by any individual collecting aperture.
Astronomical interferometry
An astronomical interferometer samples the mutual coherence of radiation received at two or more locations. For a projected baseline vector (\mathbf{B}) and an observing wavelength (\lambda), the characteristic angular resolution is proportional to
[ \theta \sim \frac{\lambda}{|\mathbf{B}|}. ]
The baseline, rather than the diameter of each collector, sets the finest accessible angular scale. Individual collectors still determine sensitivity and the angular extent of the field from which radiation can be received efficiently.
Each baseline measures a component of the source's spatial Fourier transform. Rotation of Earth changes the projected orientation and length of a fixed baseline relative to a celestial source, producing additional samples in the spatial-frequency plane. Arrays with multiple elements acquire many such samples simultaneously. Image reconstruction then estimates a brightness distribution consistent with the measured complex visibilities and the instrumental response.
At optical and infrared wavelengths, atmospheric turbulence introduces rapidly varying phase errors. Delay lines compensate for geometric path differences, while fringe tracking measures short-timescale phase fluctuations. Radio systems record electrical signals from separate antennas and combine them in a correlator, allowing phase relationships to be preserved over continental distances. Very-long-baseline interferometry extends this principle by using independent frequency standards and subsequent time-aligned correlation.
Metrology and imaging
In dimensional metrology, a displacement changes the optical path in one or more interferometer arms. For a reflected beam, movement of a mirror by a distance (d) changes the round-trip optical path by (2d). The associated phase evolution produces a sequence of fringes that relates mechanical motion to the wavelength of the source.
Phase-shifting interferometry records several intensity distributions with controlled relative phase offsets. These measurements separate the phase-dependent component from background illumination and local variations in reflectivity. The recovered phase map can represent surface height, optical aberration, or the integrated refractive-index distribution along a path.
Holography records interference between a reference field and radiation scattered from an object. The recorded pattern contains information about both the amplitude and phase of the object field relative to the reference. Digital holographic methods replace photographic reconstruction with numerical propagation, allowing quantitative phase imaging of transparent specimens.
Optical coherence tomography uses low-coherence interference to localize reflections by path length. Interference is significant only when the reference and sample paths agree within the coherence length of the source. The depth-dependent signal therefore represents the distribution of backscattering structures rather than a conventional transmission image.
Matter-wave interferometry
Quantum particles exhibit interference when their probability amplitudes propagate through distinguishable paths and remain coherent. Electron interferometry uses electrostatic or magnetic elements to divide and recombine electron waves. The phase is sensitive to geometric path length and to electromagnetic potentials, as demonstrated by the Aharonov–Bohm effect.
Atom interferometry commonly uses light pulses to transfer momentum between atomic wave packets. The separated packets accumulate different phases before recombination. Their population distribution then encodes acceleration, rotation, or gravitational potential differences. The same formalism connects atom interferometers with inertial sensors and experimental measurements of gravitational effects.
Gravitational-wave detection
Large laser interferometers measure differential changes in the lengths of perpendicular arms. A passing gravitational wave alters the spacetime interval associated with each arm, producing a relative optical phase shift. The measured quantity is the dimensionless strain
[ h=\frac{\Delta L}{L}, ]
where (L) is the effective arm length and (\Delta L) is the differential change.
Facilities such as the Laser Interferometer Gravitational-Wave Observatory use kilometer-scale Michelson configurations augmented by optical cavities. Multiple reflections increase the interaction time between the light and the changing geometry. The observable signal remains subject to quantum fluctuations in the optical field, thermal motion in the suspensions and coatings, and displacement generated by the surrounding environment.
Noise and systematic effects
Interferometric sensitivity is limited by fluctuations that change the measured phase or obscure its conversion into detector output. Photon counting statistics produce shot noise, while radiation-pressure fluctuations can move optical components. Together these effects define a quantum measurement tradeoff whose relative importance varies with frequency and optical power.
Mechanical vibration changes geometric path lengths, and temperature variation alters both material dimensions and refractive indices. Atmospheric pressure and composition affect open-air optical paths. In astronomical systems, imperfect knowledge of detector response and collector geometry modifies the measured visibility. Calibration relates these instrumental contributions to observations of sources or references with independently characterized properties.
Phase is measured modulo (2\pi), so a sampled phase map can contain discontinuities even when the underlying optical path varies continuously. Phase unwrapping reconstructs a continuous estimate by relating neighboring measurements and accounting for noise or regions of low fringe visibility. The reliability of that reconstruction depends on sampling density and on the spatial structure of the measured field.