Complex Degree of Coherence

The wavelike nature of light reveals itself most dramatically through interference, the intricate patterns of light and dark formed when waves combine. While textbooks often begin with idealized, perfectly coherent sources, real-world light—from a distant star to a simple lightbulb—is far more complex. It exists in a state of partial coherence, a middle ground between perfect order and complete randomness. This raises a critical question: how can we quantitatively describe and predict the behavior of light from such sources? The answer lies in a single, powerful mathematical tool: the complex degree of coherence. This concept provides a complete description of the correlation between light waves, serving as the Rosetta Stone for translating hidden properties of a light source into observable interference effects. This article will guide you through this fundamental concept. First, in "Principles and Mechanisms," we will dissect the complex degree of coherence, exploring how its magnitude and phase manifest as the visibility and position of interference fringes and uncovering the profound Fourier relationships that link coherence to the source's spectral and spatial properties. Then, in "Applications and Interdisciplinary Connections," we will see how this theoretical framework becomes a practical tool for astronomers to measure stars, for engineers to design optical systems, and for physicists to unify disparate concepts like polarization and coherence.

Imagine you are standing on the shore of a calm lake. If you dip a single finger into the water, a perfect, orderly set of circular ripples expands outwards. Now, imagine dipping two fingers in. The two sets of ripples will cross and combine, creating a beautiful and intricate pattern of high crests and calm troughs—an interference pattern. Light, being a wave, does exactly the same thing. But what happens if the source of the waves is not a single, orderly disturbance, but a chaotic jumble of countless, independent agitations, like rain falling on the lake's surface? This is the world of partial coherence, and to navigate it, we need a guide. That guide is the ​​complex degree of coherence​​, a single, elegant mathematical object that tells us everything we need to know about how well waves can interfere.

The classic stage for observing the wavelike nature of light is ​​Young's double-slit experiment​​. When light passes through two nearby pinholes, it creates a pattern of bright and dark bands, or ​​fringes​​, on a screen behind them. But the quality of these fringes tells a deep story. Sometimes they are sharp and distinct, with brilliant brights and inky blacks. Other times, they are washed-out and murky, barely visible against a uniform background glow.

This fringe "contrast," which we quantify as ​​visibility​​ ($V$), gives us the first piece of our puzzle. It turns out that the visibility is a direct measure of the magnitude of the complex degree of coherence between the light arriving at the two slits, a quantity we denote as $|\gamma_{12}|$. If the light illuminating both slits is equally intense, the relationship is beautifully simple: the visibility is exactly equal to the magnitude of the coherence, $V = |\gamma_{12}|$. A value of $|\gamma_{12}| = 1$ means perfect coherence, yielding the highest possible contrast. A value of $|\gamma_{12}| = 0$ means complete incoherence, and the fringes vanish entirely. For intermediate values, like $|\gamma_{12}| = 0.671$, the fringes are present but have reduced contrast.

Of course, the real world is rarely so perfectly balanced. What if one slit is more brightly illuminated than the other? The ability of the waves to interfere is still governed by $|\gamma_{12}|$, but the unequal brightness "dilutes" the visibility. The general relationship, a testament to how nature combines different effects, is $V = \frac{2|\gamma_{12}|\sqrt{I_1 I_2}}{I_1 + I_2}$, where $I_1$ and $I_2$ are the intensities from each slit alone. You can see that even if the coherence is perfect ($|\gamma_{12}|=1$), if one slit is much brighter than the other (say, $I_1 \gg I_2$), the visibility will be poor. It's like trying to hear a whisper in a shouting match; the fainter wave's contribution to cancellation gets lost.

This explains the contrast of the fringes, but what about their position? If the light reaching the two slits is perfectly "in-step," the central fringe—the brightest of them all—will appear exactly in the middle of the screen, at the point equidistant from both slits. But what if there's a systematic delay, a subtle "stutter" in the phase relationship between the two points? This is where the "complex" part of the complex degree of coherence comes into play.

The complex degree of coherence, $\gamma_{12}$, is not just a magnitude but also has a ​​phase​​, or argument, $\alpha = \arg(\gamma_{12})$. This phase angle directly corresponds to the intrinsic phase difference between the light fields at the two slits. A non-zero phase $\alpha$ causes the entire fringe pattern to shift sideways on the screen. The amount of the shift, $\Delta y$, is directly proportional to this phase: $\Delta y = -\frac{\alpha \lambda D}{2\pi d}$.

Let's consider a striking thought experiment. What if the coherence between the slits was described by a purely imaginary number, say $\gamma_{12} = i\beta$? In the language of complex numbers, this corresponds to a phase shift of $\alpha = \pi/2$, or a quarter of a full wave cycle. The consequence? The entire interference pattern shifts by exactly one-quarter of a fringe width. The point that used to be the peak of the central bright fringe is now a point of average intensity, and the new peak is shifted to where the slope was steepest. So, the complex number $\gamma_{12}$ is a wonderfully compact package of information: its magnitude tells you how strong the interference is, and its phase tells you where the pattern is centered.

We've seen what coherence does, but where does it come from? How can a star, a chaotic ball of hot gas with atoms emitting light at random, produce any coherence at all? The answer lies in two profound relationships, both linking the properties of the light source to the coherence of the field it produces. These relationships reveal a beautiful unity in optics, connecting seemingly disparate concepts through the magic of the Fourier transform.

Temporal Coherence: The Cosmic Drumbeat

Let's first consider the light at a single point in space and ask how the field at one moment in time, $E(t)$, is related to the field at a slightly later time, $E(t+\tau)$. This property is called ​​temporal coherence​​, and it is a measure of how "monochromatic" or spectrally pure the light is. A perfect, unending sine wave is perfectly predictable; its temporal coherence is absolute for any time delay $\tau$. But the light from real sources is never perfectly monochromatic. It's a jumble of different frequencies, a "song" with a certain spectral shape, $S(\omega)$.

The deep connection, known as the ​​Wiener-Khinchin theorem​​, states that the temporal coherence function, $\gamma(\tau)$, is the Fourier transform of the source's power spectrum, $S(\omega)$. This means that by measuring how the light correlates with itself over time, we can deduce its spectral composition without ever using a prism or a diffraction grating!

Imagine a simple source, like a collection of atoms that behave like tiny, damped oscillators. They emit light with a characteristic frequency but the oscillations die out. This gives rise to a ​​Lorentzian​​ spectral line. The Wiener-Khinchin theorem tells us that the temporal coherence of this light will decay exponentially: $\gamma(\tau) \propto \exp(-\beta|\tau|)$, where $\beta$ is related to the width of the spectral line. The broader the spectrum (more "colors" mixed in), the faster the light "forgets" its own phase, and the faster the coherence decays.

Now for a more intricate example. Suppose our light source contains two distinct types of atoms, or a single type that can emit at two closely spaced frequencies. The spectrum would have two peaks. What would the temporal coherence look like? The Fourier transform predicts something remarkable. The coherence will still decay, but this decay will be modulated by a cosine term, like $|\gamma(\tau)| \propto (\text{decay envelope}) \times |\cos(\delta\omega\tau)|$, where $\delta\omega$ is half the frequency separation between the spectral peaks,. This is a beat phenomenon, right in the coherence function! For certain time delays $\tau$, the cosine term goes to zero, and the light appears temporarily incoherent, only to have its coherence "revive" at a slightly later time. This pattern of coherence revivals is a direct fingerprint of the bimodal nature of the source's spectrum.

Spatial Coherence: The Fingerprint of a Star

Now let's turn to ​​spatial coherence​​, the correlation of light between two different points in space at the same time. It seems counter-intuitive that a completely disordered, or ​​spatially incoherent​​, source like the surface of a star could produce light that is spatially coherent thousands of light-years away. But it does, and this phenomenon is the key to measuring the size of distant stars.

The principle that governs this apparent magic is the ​​Van Cittert-Zernike theorem​​. It states that the complex degree of spatial coherence, $\gamma_{12}(\Delta \vec{p})$, between two points in an observation plane is given by the Fourier transform of the intensity distribution of the distant, incoherent source, $I(\vec{s})$. This is an astonishing parallel to the Wiener-Khinchin theorem. One relates time-coherence to the frequency spectrum, the other relates space-coherence to the source's spatial "spectrum" (its shape).

Let's see this in action. Imagine we are observing a distant, uniformly bright filament of light. The Van Cittert-Zernike theorem predicts that the spatial coherence function we measure will have the shape of a sinc function, $\text{sinc}(z) = \sin(z)/z$. This is profoundly analogous to Fraunhofer diffraction, where a single slit aperture produces a sinc-squared intensity pattern. By measuring the width of this sinc function in our coherence measurement, we can directly calculate the physical width of the unresolved filament! This is the fundamental principle of stellar interferometry: we measure the coherence pattern of starlight on Earth to "image" the star's surface.

The symmetries of the source are also beautifully preserved in the coherence function. If a distant source has a symmetric intensity distribution (i.e., $I(\vec{s}) = I(-\vec{s})$), then its Fourier transform is purely real. This means the complex degree of coherence $\gamma_{12}$ will be a real number. A real $\gamma_{12}$ has a phase of either $0$ or $\pi$, meaning it will produce interference fringes that are either perfectly centered or "inverted" (with a dark fringe at the center), but never shifted sideways. The symmetry of the source forbids a fringe shift.

Can the real part of coherence be negative? This would correspond to a phase of $\pi$, or an inversion of the fringe pattern. A thought experiment shows it can. Consider a source made of three points of light: one at the center and two flanking it. If the two outer sources are sufficiently bright compared to the central one (specifically, if the intensity ratio $I_{\text{outer}} / I_{\text{central}} > 1/2$), then for certain separations on our screen, the coherence will indeed become negative. This "anti-correlation" arises from the destructive interference of the coherence contributions from different parts of the source.

In the end, the complex degree of coherence reveals itself not as an arcane abstraction, but as a powerful and unifying lens. Through its magnitude and phase, it translates the invisible dance of wave correlations into the visible drama of interference fringes. And through the elegant language of the Fourier transform, it allows us to read the hidden stories written in a beam of light—decoding the spectral song of atoms and mapping the shapes of stars across the universe.

We have arrived at a truly remarkable conclusion: a collection of independent, randomly jiggling emitters—a completely incoherent source—can somehow conspire to produce a light field that has a predictable and structured spatial coherence. The Van Cittert-Zernike theorem gives us the key, revealing that the complex degree of coherence in the far field is nothing but the Fourier transform of the source's intensity distribution. This is not merely a mathematical curiosity; it is a profound principle that unlocks a vast array of applications across science and engineering. Now that we have this peculiar and powerful tool, let's explore the world with it. Where does this principle show up, and what can we do with it?

Let us begin with the simplest possible extended source: a glowing hot wire, like the filament in an old incandescent bulb. If we model this as a one-dimensional, incoherent line source, the theorem tells us that the coherence between two points on a distant screen will follow a specific mathematical form known as the sinc function, $\sin(x)/x$. If we expand our source to a two-dimensional incoherent square, we find, just as beautifully, that the coherence pattern is simply the product of two such sinc functions, one for each dimension. In both cases, a simple geometric shape for the source is transformed into a simple, predictable coherence function in the field. The light field, in a sense, remembers the shape of its origin.

This relationship is more than just a neat trick; it is the foundation of a revolution in astronomy. A distant star, for all practical purposes, is a uniform, incoherent circular disk of light. While it may appear as a point to our eyes or even to a moderately sized telescope, the Van Cittert-Zernike theorem insists that its finite size is encoded in the coherence of its light. If we set up two small detectors on Earth, separated by a baseline distance $d$, and measure the coherence between them, we are effectively sampling one point on the Fourier transform of the star's disk shape. As we vary the baseline $d$, we trace out this transform. For a circular source, the mathematics gives us a beautiful result involving the first-order Bessel function, $J_1(x)$. The coherence starts at 1 for zero separation and then decreases, eventually hitting zero at a specific baseline. This first zero is a magic point! Its location depends directly on the star's angular size and the wavelength of light.

Think about what this means. We can measure the diameter of a star that is light-years away—an object far too small to be resolved by any single conventional telescope—simply by measuring how the "sameness" of its light fades as we move two collectors apart. This is the principle behind the Michelson stellar interferometer, and it transforms the concept of coherence from a physicist's abstraction into an astronomer's measuring tape. The same logic applies to more complex source shapes, like an annular ring, which can be handled by cleverly combining the transforms of simpler shapes, demonstrating the power of this Fourier-space thinking.

So far, we have mostly spoken of the magnitude of the coherence, which relates to the visibility of interference fringes. But the degree of coherence is a complex number; it has a phase as well. What is the physical meaning of this phase?

The phase holds the key to uncovering asymmetries in a source. Imagine our star is not alone but is part of a binary system, with a fainter companion orbiting it. A single, centered circular star is perfectly symmetric. Its Fourier transform, and thus its coherence function, is purely real—it has zero phase everywhere. But the presence of an off-center companion breaks this symmetry. The Van Cittert-Zernike theorem tells us that this asymmetry introduces a non-zero, spatially varying phase into the complex degree of coherence. By measuring not just the fringe visibility (magnitude) but also the fringe position (which reveals the phase), an interferometer can detect the tell-tale signature of the unseen companion. We can determine the separation of the stars and the ratio of their brightness just by carefully analyzing the complex-valued "visibility" data.

This principle—that magnitude encodes size and phase encodes structure—is the absolute heart of modern interferometry. It is how radio astronomers, using arrays of telescopes spanning continents, can reconstruct breathtaking images of phenomena like the shadow of a black hole. They are not building a lens the size of the Earth; they are painstakingly measuring the complex degree of coherence and using a computer to perform the inverse Fourier transform, turning coherence data back into a picture.

Our journey has taken us to the stars, but the light must travel from there to us, passing through our instruments and our turbulent atmosphere. Coherence is not immune to this journey.

First, consider the telescope itself. An ideal telescope with an angular magnification $M$ doesn't just make things look bigger; it also transforms the coherence of the light. It turns out that a separation $d$ in the telescope's exit pupil corresponds to a larger separation, $M \cdot d$, at the entrance pupil where the light first arrived. This means the coherence pattern from the source appears "compressed" at the back end of the telescope. Instrument designers must account for this scaling when building interferometers that combine light from multiple telescopes.

Second, no real-world lens is perfect. They all suffer from aberrations. Let's say our lens has spherical aberration, a common defect where rays hitting the edge of the lens focus at a slightly different point than rays hitting the center. This aberration is not just a geometric problem; it is a phase problem. A lens with spherical aberration imparts a non-ideal phase shift across the wavefront. This, in turn, directly modifies the phase of the complex degree of coherence of the light passing through it. An aberration, therefore, doesn't just blur an image; it fundamentally alters the coherence properties of the field, a crucial consideration in high-performance imaging systems.

Finally, and most vexingly for ground-based astronomers, is the Earth's atmosphere. The air above us is a swirling, turbulent soup of temperature and pressure variations, which means its refractive index is constantly changing. As a wavefront from a distant star travels through it, it gets randomly and continuously distorted. This is why stars twinkle. We can model this effect as the light passing through a "random phase screen." This screen scrambles the phase of the light, and in doing so, it severely degrades the spatial coherence. The effect is catastrophic: the coherence of the starlight is multiplied by a rapidly decaying factor that depends on the strength of the turbulence, characterized by a parameter called the Fried parameter, $r_0$. Beyond a separation of roughly $r_0$ (which can be just a few centimeters on a bad night), the beautiful coherence endowed by the star is almost completely wiped out. This is the fundamental reason it is so hard to build large ground-based telescopes that achieve their theoretical resolution. The battle against this atmospheric decoherence is the primary motivation for developing adaptive optics—systems of deformable mirrors that try to undo the phase scrambling in real time—and for placing telescopes in space, above the troublesome atmosphere altogether.

The concept of coherence has one more surprise for us, connecting it to another fundamental property of light: polarization. So far, we have discussed spatial coherence—the correlation of the field at two different points in space. But what if we stay at a single point and ask about the relationship between the orthogonal components of the electric field vector, $E_x$ and $E_y$?

This relationship is also one of coherence! The state of polarization of a light beam is entirely determined by the intensities of its two orthogonal components and the complex degree of coherence, $\mu_{xy}$, between them. If the light is perfectly linearly polarized, it means $E_x$ and $E_y$ oscillate perfectly in-phase (or exactly out-of-phase), implying $|\mu_{xy}| = 1$. If the light is completely unpolarized, like the light from a thermal source, it means $E_x$ and $E_y$ are completely uncorrelated—their coherence is zero. The vast middle ground of partially polarized light corresponds to intermediate values of $|\mu_{xy}|$.

This is a beautiful unification. The very same mathematical framework that allows us to measure the size of stars and design better telescopes also provides the precise language to describe the polarization state of light. The degree of polarization is not a separate ad-hoc concept; it is a direct consequence of the degree of coherence between the field's vector components.

From the glowing filament of a lamp to the turbulent air above us, from the majestic dance of binary stars to the fundamental vector nature of light itself, the concept of the complex degree of coherence provides a unifying thread. It reveals the hidden order within randomness and gives us the tools not just to understand the universe, but to measure and image it in ways that would have been unimaginable without it.