Spatial coherence

Spatial coherence is a fundamental property of light that describes the degree of correlation, or fixed phase relationship, between waves at different points in space. This property is what allows light to produce the distinct patterns of bright and dark bands in interference experiments. While a highly ordered source like a laser is intuitively coherent, a profound puzzle arises when considering chaotic sources like stars. How can the jumbled, independent light emissions from a star produce interference fringes on Earth, a hallmark of coherence? This apparent paradox highlights a gap in our simple understanding of light propagation.

This article delves into the physics of spatial coherence to resolve this puzzle. The first chapter, "Principles and Mechanisms," will introduce the core concepts of coherence, fringe visibility, and the pivotal van Cittert-Zernike theorem, which explains how order emerges from chaos as light travels through space. The second chapter, "Applications and Interdisciplinary Connections," will then explore the far-reaching impact of these principles, demonstrating how spatial coherence is not just a theoretical curiosity but a powerful tool used in astronomy, materials science, quantum physics, and even cosmology.

Imagine you are standing by a calm lake at night. If you drop a single pebble into the water, a perfect series of concentric circular ripples expands outwards. The crests and troughs move in a predictable, orderly procession. Now, imagine a friend drops another pebble a short distance away. Where the two sets of ripples meet, they will interfere, creating a complex but still predictable pattern of higher peaks and deeper troughs. This orderliness, this fixed relationship between the phases of the waves at different points, is the essence of ​​coherence​​.

Now, instead of two pebbles, imagine it starts to rain. Thousands of raindrops hit the surface of the lake, each one a tiny, independent source of ripples. The surface becomes a chaotic mess of motion. If you were to look at two nearby points on the water, you would find no predictable relationship between their up-and-down movements. The waves are ​​incoherent​​.

Light, being a wave, behaves in much the same way. A laser produces light like the single pebble: all the waves march in lockstep, perfectly coherent. A candle flame or a star, on the other hand, is like the rain-swept lake. It consists of countless atoms, each emitting light independently and randomly. It is a fundamentally incoherent source. This brings us to a profound puzzle. In 1801, Thomas Young performed his famous double-slit experiment, showing that light from a single source passing through two pinholes creates an interference pattern of bright and dark stripes, or "fringes." This pattern is the hallmark of coherence. But later experiments showed that you could get interference fringes even from the light of a distant star! How can this be? How can the chaotic, incoherent light from a star produce the orderly pattern of interference? The answer lies not in the source itself, but in the magic of propagation through space.

Let's return to Young's experiment, the ultimate probe of coherence. We shine light on a screen with two narrow, parallel slits. On a second screen far behind the first, we observe the pattern. If the light arriving at both slits is coherent, the two slits act like our two pebbles, creating a beautiful interference pattern. The "quality" of this pattern—how sharp and distinct the bright and dark fringes are—is quantified by its ​​visibility​​, $V$, defined as $V = \frac{I_{max} - I_{min}}{I_{max} + I_{min}}$. For perfect interference, the dark fringes have zero intensity ($I_{min}=0$), and the visibility is 1. For no interference at all, the intensities just add up, $I_{max}=I_{min}$, and the visibility is 0.

It turns out that this visibility is a direct measure of the coherence between the light fields at the two slits. We define a quantity called the ​​complex degree of spatial coherence​​, denoted by $\mu_{12}$, which measures the correlation between the light wave's vibrations at two points, say point 1 and point 2. This is a complex number, but its magnitude, $|\mu_{12}|$, is what matters for interference. It ranges from 1 (perfectly coherent) to 0 (perfectly incoherent). In a Young's double-slit experiment with equally illuminated slits, the relationship is beautifully simple: the fringe visibility is exactly equal to the magnitude of the complex degree of spatial coherence between the slits.

$V = |\mu_{12}|$

Imagine the light illuminating the slits is only partially coherent. Perhaps the coherence decreases as the distance between the two points increases. A simple and common model for this is an exponential decay, where the coherence between two points separated by a distance $d$ is given by $|\mu| = \exp(-d/\sigma_c)$, where $\sigma_c$ is a characteristic "coherence length". If we perform a Young's experiment with this light, the visibility of the fringes we see will be exactly $V = \exp(-d/\sigma_c)$. If the slits are very close together ($d \ll \sigma_c$), the visibility is nearly 1. As we move the slits further apart, the visibility fades, and the interference pattern washes out. This gives us a direct, experimental way to measure the spatial coherence of a light field.

We are now equipped to tackle our central paradox: the coherence of starlight. The solution is one of the most elegant and surprising results in optics, the ​​van Cittert-Zernike theorem​​, named after the Dutch physicists Pieter Hendrik van Cittert and Frits Zernike.

In essence, the theorem says this: the process of propagation itself creates spatial coherence. Even if a source is completely spatially incoherent, the light field it produces will acquire a degree of spatial coherence as it travels away from the source. The farther it travels, the more coherent it becomes over larger and larger areas.

The theorem provides a stunningly precise mathematical formulation for this miracle. It states that the complex degree of spatial coherence between any two points in the far field is given by the normalized Fourier transform of the intensity distribution of the incoherent source.

Let's unpack that. A Fourier transform is a mathematical tool that breaks down a function (or a signal) into its constituent frequencies. Here, it connects the spatial distribution of the source's brightness to the spatial correlation of the distant light field. This is a deep connection, linking the shape of the source to the coherence of its light.

To grasp this, consider a simple, one-dimensional, incoherent source—a uniformly bright, narrow line of light, perhaps a heated filament of width $a$. According to the theorem, if we observe this light from a large distance $z$, the coherence between two points separated by a distance $\Delta x$ will be described by a sinc function: $\mu(\Delta x) = \frac{\sin(\pi a \Delta x / (\lambda z))}{\pi a \Delta x / (\lambda z)}$, where $\lambda$ is the wavelength of the light.

This function starts at 1 for zero separation ($\Delta x=0$) and oscillates as $\Delta x$ increases, passing through zero for the first time when $\Delta x_c = \lambda z / a$. This distance, $\Delta x_c$, is called the ​​transverse coherence length​​. It defines the size of a "coherent patch" in the observation plane. Within this patch, the light is highly coherent and can produce strong interference. For points separated by much more than $\Delta x_c$, the coherence is essentially lost. Notice the beautiful inverse relationship: the smaller the source size $a$, the larger the coherence length $\Delta x_c$. A tiny, distant star (small $a$, large $z$) will produce a light field that is highly coherent over a large area on Earth. This is why starlight can produce interference fringes. It's not that the star is coherent; it's that it is very, very small and very, very far away.

This idea extends to two dimensions. For a circular source, like a star of radius $a$, the coherence function is no longer a sinc function but its two-dimensional cousin, a Bessel function of the first kind, which looks like a decaying ripple pattern. We can define a ​​coherence area​​, $A_c$, over which the light is correlated. The van Cittert-Zernike theorem gives the beautiful result that this area is inversely proportional to the area of the source, $A_s$: $A_c \approx (\lambda z)^2 / A_s$. Again, a smaller source creates a larger area of coherence.

It is absolutely crucial to remember that this theorem applies only to ​​spatially incoherent sources​​. If the source itself is coherent—for example, two tiny emitters driven in perfect lockstep—the situation is completely different. The light is perfectly coherent everywhere, regardless of the distance, and the visibility of interference fringes will always be 1 (assuming equal intensities). The van Cittert-Zernike theorem simply doesn't apply. It is a description of how order arises from chaos through the physics of wave propagation.

The van Cittert-Zernike theorem is not just an academic curiosity; it is the foundation of one of the most powerful techniques in modern astronomy: ​​stellar interferometry​​. If the coherence pattern contains the Fourier transform of the source's shape, then by measuring the coherence, we can work backwards—perform an inverse Fourier transform—to reconstruct an image of the source!

This allows astronomers to "see" details far beyond the resolving power of any single telescope. The resolution of a telescope is limited by its diameter. But by using two (or more) telescopes separated by a large distance, or "baseline," $\Delta x$, we are not building a bigger lens. We are directly probing the coherence function of the starlight at that specific separation. By moving the telescopes and measuring the fringe visibility at many different baselines, we can map out the coherence function $\mu(\Delta x)$.

Imagine an astronomer points their interferometer at what appears to be a single star and measures the coherence function. Instead of a simple sinc-like decay, they find something that looks like a cosine wave under a decaying envelope. What does this mean? A cosine function is the Fourier transform of two sharp spikes. The astronomer has discovered that this is not a single star, but a binary star system! The frequency of the cosine oscillations directly reveals the angular separation of the two stars. By fitting their data to the model predicted by the van Cittert-Zernike theorem, they can measure this separation with astonishing precision.

The same principle applies to more complex sources. If a source were made of four points of light at the corners of a square, its coherence function would be a two-dimensional checkerboard pattern described by the product of two cosines. Every detail of the source's geometry is encoded in the structure of the coherence function. By carefully measuring these coherence patterns, astronomers can image the surfaces of distant stars, observe the swirling disks of gas around black holes, and map the cosmos with a sharpness of vision that would be unimaginable with conventional telescopes.

The van Cittert-Zernike theorem provides a powerful mechanism for generating coherence, but the world of optics is richer still.

What if a source is neither perfectly incoherent nor perfectly coherent? Many real-world sources, such as light-emitting diodes (LEDs) or multimode lasers, fall into this category. They are best described by models like the ​​Gaussian Schell-Model (GSM)​​. Such a source has both a characteristic size (its intensity profile) and an intrinsic degree of coherence at the source itself. In a GSM beam, for example, the intensity might have a Gaussian profile of width $w_0$, while the coherence between points at the source decays as a Gaussian function with a characteristic width $\sigma_g$. When we measure the coherence between two points in such a beam, we find that it depends on this intrinsic coherence length $\sigma_g$, not on the overall beam width $w_0$. This gives us a more general framework to understand and engineer partially coherent light.

Furthermore, propagation from an incoherent source is not the only way to generate partial coherence. One can also start with perfect coherence and destroy it. Imagine a perfectly flat, coherent plane wave—like the beam from an ideal laser—reflecting off a mirror. If the mirror is perfectly smooth, the reflected wave is still perfectly coherent. But what if the mirror surface is rough and fluctuates randomly in time, like the surface of a simmering liquid? Each part of the wave reflects off a point with a slightly different, random height. This imprints random phase shifts onto the wavefront. The initially pristine, coherent wave becomes scrambled, emerging as a partially coherent field. The degree of coherence in the reflected beam is now determined by the statistical properties of the rough surface—its root-mean-square height fluctuation and the spatial scale over which these fluctuations are correlated.

The principles connecting source and coherence are remarkably robust. The standard van Cittert-Zernike theorem is derived under a "far-field" assumption. One might think that things get hopelessly complicated closer to the source. But in a beautiful demonstration of the power of wave physics, for certain geometries like an incoherent line source, the same simple Fourier transform relationship holds even in the closer Fresnel region. The underlying physics is deeper than the approximations we often use to describe it.

From the chaotic glow of a candle to the sharp interference fringes of starlight measured by telescopes miles apart, spatial coherence is a testament to the elegant and often surprising nature of wave propagation. It is a story of how order emerges from randomness, and how by understanding this process, we can turn the twinkling of a distant star into a detailed portrait of another world.

We have spent some time understanding the machinery behind spatial coherence, particularly the beautiful relationship between an incoherent source and the field it produces far away, as described by the van Cittert-Zernike theorem. Now, where does this idea lead us? What is it for? It is one thing to have a neat mathematical theorem, but it is another entirely for it to open our eyes to the world around us and give us new tools to explore it. As we shall see, the concept of spatial coherence is not some dusty corner of optics; it is a vibrant, essential thread that weaves through astronomy, materials science, quantum physics, and even cosmology. It is the secret behind how we see the stars, design microscopes, and even ponder the structure of the universe itself.

Let's start with the grandest and most familiar light source we know: the Sun. We think of the Sun as the archetype of an incoherent source—a chaotic maelstrom of thermonuclear reactions, with countless atoms emitting light independently. If you were to ask whether the light from the Sun is coherent, the immediate answer would seem to be a resounding "no." And yet, this is where Nature's subtlety first reveals itself.

Imagine setting up a Young's double-slit experiment, not with a laser, but with sunlight. If you place the two slits very close together, you will see interference fringes! This simple fact is a direct consequence of spatial coherence. The van Cittert-Zernike theorem tells us that even a completely incoherent source will generate a field that is correlated, or coherent, over small transverse distances. For the Sun, which appears as a small disk in the sky, we can calculate the distance over which its light remains coherent upon reaching Earth. A careful calculation, which treats the Sun as a distant, incoherent circular disk, reveals that the transverse spatial coherence length of sunlight at a mean wavelength of $\lambda=550$ nm is about 72 micrometers. This means if your two slits are closer than this distance, the light passing through them will interfere. This isn't just a theoretical curiosity; it is the foundational principle behind stellar interferometry, a technique that allows astronomers to measure the angular size of distant stars by measuring the coherence of their light. What began as a simple question about sunlight becomes a powerful tool for mapping the cosmos.

This surprising coherence from an incoherent source is only the beginning of the story. It turns out we are not just passive observers of this phenomenon; we can manipulate and engineer it. Any time we use a lens or a mirror to form an image, we are fundamentally altering the coherence properties of the light.

Consider an optical system, like a telescope or a microscope, forming an image of a diffuse, incoherent source. You might think that the coherence in the image plane would depend on the size of the source itself. But what is remarkable is that for an ideal imaging system, the coherence properties in the image plane are determined not by the source, but by the size of the instrument's aperture—the lens or mirror that collects the light. The aperture acts like a new, effective source, imposing its own coherence signature onto the image. The same principle applies when you look at an object through a simple magnifying glass; the coherence of the light entering your eye is governed by the properties of the lens and the size of the object as seen through that lens.

This is a profound shift in perspective. Coherence is not just a passive property of light to be measured; it is a design parameter that an optical engineer can control. But how do we measure it? The most direct way is to look at the visibility of interference fringes, defined as $V = (I_{max} - I_{min})/(I_{max} + I_{min})$. In an ideal experiment, the visibility of the fringes created by interfering two parts of a wavefront is precisely equal to the magnitude of the complex degree of coherence between them. If you take a partially coherent beam and send it into a Michelson interferometer where one mirror is slightly tilted, the two sheared beams will interfere. The contrast of the resulting fringes directly reveals the beam's spatial coherence length. A larger tilt creates a larger shear, and as this shear exceeds the coherence length, the fringes wash out and disappear. This provides a direct, tangible link between the abstract mathematical definition of $\mu$ and a measurable physical effect.

The ability to understand, measure, and control coherence is a cornerstone of modern experimental science. The principles we've discussed apply far beyond visible light, finding critical applications in fields that probe the very structure of matter.

Consider the advanced light sources used in materials science, such as synchrotrons. These particle accelerators produce incredibly bright beams of X-rays that can be used to study materials at the atomic scale. One such technique, Small-Angle X-ray Scattering (SAXS), relies on the interference of X-rays scattered by nanoscale structures in a sample. To get a clean signal, do you need a perfectly coherent beam? Not necessarily. In fact, using a beam with partial transverse coherence can be advantageous, as it averages out speckle noise and gives a clearer picture of the material's average structure. Furthermore, SAXS can even be performed with a "pink" beam that has a relatively broad spectrum, because the path length differences for small-angle scattering are tiny and still fall within the beam's longitudinal coherence length. In contrast, another technique, X-ray Absorption Spectroscopy (XAS), which probes the electronic environment of specific atoms, is fundamentally an absorptive process and does not rely on the coherence of the incoming photons at all. For XAS, what matters is the precise energy of the photons, not their phase relationship. This example beautifully illustrates how a deep understanding of coherence is essential for choosing the right tool for a scientific job.

The same principles extend to even more exotic light sources. In plasma wakefield accelerators, electrons can be made to oscillate and emit intense X-rays, a process known as betatron radiation. This source, arising from the complex motion of an electron bunch, can be modeled as spatially incoherent. The van Cittert-Zernike theorem once again allows physicists to predict the transverse coherence length of the resulting X-ray beam, a critical parameter for designing future experiments with these compact, ultra-bright sources. We can even define a useful quantity called the "coherence area," $A_c$, which is essentially a measure of the size of the "patch" over which a thermal source, like a black body, produces coherent light. This area is elegantly given by $A_c = (\lambda z)^2 / (\pi R^2)$, where $R$ is the source radius and $z$ is the distance, showing that coherence is a resource that can be quantified and optimized.

The concept of spatial coherence is so fundamental that it appears in the most advanced and mind-bending areas of physics. It is not limited to classical waves but is central to our understanding of quantum and even cosmological phenomena.

In the field of nonlinear optics, physicists have developed techniques to manipulate light in extraordinary ways. One such method is phase conjugation, where a special medium can generate a light beam that is the exact time-reversed replica of an incoming probe beam. What happens to the coherence? It turns out that the generated phase-conjugate beam perfectly inherits the spatial coherence properties of the probe beam, but with a conjugated phase factor. This ability to "copy" and "reverse" the coherence of a light field opens up possibilities for correcting distortions in optical systems and for advanced information processing.

The idea of coherence even applies to fields that don't propagate in the usual sense. Near the surface of a material, there can exist "evanescent" electromagnetic fields that decay exponentially away from the interface. These fields can be excited by thermal fluctuations, creating a rich tapestry of near-field phenomena. The spatial coherence of these thermal evanescent fields is directly linked to the material's properties, such as the existence of surface polaritons (hybrid light-matter waves). For a sharp surface resonance, the spatial coherence function takes the elegant form of a Bessel function, $\mu(\Delta\rho) \propto J_0(k_{res} \Delta\rho)$, directly encoding the resonant wavevector $k_{res}$ of the surface mode. Measuring this coherence allows us to probe the physics of surfaces at the nanoscale.

Finally, we take our journey to its grandest and most abstract conclusion: the universe itself. According to quantum field theory, the vacuum is not empty. It is a seething sea of quantum fluctuations. In an expanding universe, such as the one described by the de Sitter metric, these vacuum fluctuations have a specific correlation structure. One can actually calculate the "spatial coherence" of a quantum scalar field in this cosmological setting. This correlation function tells us how the random quantum jitters at two different points in the expanding universe are related. Just as with light, the correlation falls off with distance, and we can define a regularized degree of coherence that depends on the Hubble constant—the rate of cosmic expansion. That the same mathematical concept used to describe interference fringes from sunlight can be adapted to characterize the quantum vacuum of our expanding universe is a stunning testament to the unity and power of physics. From a double slit to the dawn of time, the story of coherence is the story of how different parts of the universe are connected.