Phase Space Optics

For centuries, our understanding of light has been fragmented, split between the intuitive paths of geometric rays, the intricate patterns of wave interference, and the strange probabilities of quantum mechanics. While each model is powerful in its own domain, they often feel like separate languages describing the same reality. Phase-space optics offers a powerful Rosetta Stone—a unified framework that translates these disparate views into a single, cohesive picture by simultaneously considering a light ray's position and its direction of travel. This approach resolves the apparent conflict between classical intuition and quantum weirdness, revealing deep connections across physics and engineering. This article bridges the gap between these different optical models by exploring how a simple map of position and momentum provides a complete description of a light field's state.

The journey begins in the first chapter, ​​"Principles and Mechanisms,"​​ where we will uncover the fundamental laws governing light in phase space, from the conservation of etendue to the introduction of the powerful Wigner Distribution Function. We will see how this elegant tool turns complicated wave diffraction into simple geometric transformations and reveals the signatures of quantum mechanics itself. In the second chapter, ​​"Applications and Interdisciplinary Connections,"​​ we will see these principles in action, exploring how phase-space thinking revolutionizes optical engineering, enhances microscopy, and provides a startlingly clear window into the quantum world. Prepare to see light not just as rays or waves, but as a dynamic dance in the rich landscape of phase space.

Imagine you are trying to describe a swarm of fireflies on a summer night. To know everything about the swarm at a given moment, you would need to know not just where each firefly is, but also which way it is going. A list of positions isn't enough; you need the positions and the velocities. This combined space of information—position and momentum for every particle—is what physicists call ​​phase space​​. It’s the true "state" of the system.

Now, what if we tried to do the same for light? For centuries, we’ve pictured light as a collection of rays, zipping around in straight lines, bouncing off mirrors, and bending through lenses. In this picture, a ray of light, just like a firefly, has a ​​position​​ (where it hits a screen) and a ​​momentum​​ (which is just its ​​angle​​ of propagation, scaled by the refractive index). The phase space of light, then, is a map where every point represents a ray at a specific location, traveling in a specific direction. This simple idea is the key that unlocks a profound and beautiful perspective on optics.

Let’s start with a simple bundle of rays. Imagine we isolate a group of them that, at one plane, fill a neat rectangular shape in our phase space: their positions are within a range $\Delta x$, and their angles are within a range $\Delta \theta$. Now, we let this bundle of rays propagate through empty space for a distance $L$. What happens?

A ray that starts at position $x_i$ with angle $\theta_i$ will arrive at a new position $x_f = x_i + L\theta_i$. Its angle, of course, doesn't change: $\theta_f = \theta_i$. If you plot the new positions and angles of all the rays in our bundle, you’ll find that the neat rectangle has been sheared into a parallelogram. The range of positions has stretched, but the range of angles has remained the same. If you calculate the area of this new parallelogram in phase space, you will find something astonishing: it is exactly the same as the area of the original rectangle. This conserved area, a product of spatial spread and angular spread, is a fundamental quantity known as ​​etendue​​.

This isn't just a trick of free space. It turns out that any ideal, lossless optical system—no matter how complex, with any combination of lenses and curved mirrors—obeys this law. As long as the system can be described by the standard ​​ABCD matrix​​ of paraxial optics, the etendue of a beam passing through it is conserved. This is the optical version of a deep principle in classical mechanics called ​​Liouville's theorem​​, which states that the volume of phase space occupied by a cloud of particles is constant as they evolve in time. For light, propagation through an optical system is its "evolution," and etendue is its "phase-space area."

But why should nature follow such a rule? Is it just a mathematical curiosity? No, it's something much deeper, tied to the most unshakable laws of physics. Imagine a hypothetical "magic box" optics that could violate this rule—a device that could take a wide, diffuse bundle of light and squeeze it into a narrow, collimated beam, reducing its etendue. If you used this box to image a warm object (a blackbody) onto itself, you could funnel the faint glow from the entire object into a tiny spot, making that spot hotter than the original object. You could then use this temperature difference to run an engine. You would have created a machine that extracts useful work from a single heat source, a blatant violation of the Second Law of Thermodynamics! The conservation of etendue, therefore, is not just an optical rule; it is a direct consequence of thermodynamics.

This connection also reveals an even more powerful conserved quantity. The brightness of a source is measured by its ​​radiance​​, $L$, which is the power per unit area per unit solid angle. By combining the conservation of energy (power) with the conservation of etendue, one can derive a beautiful law: the quantity $L/n^2$, where $n$ is the local refractive index of the medium, is absolutely constant along any ray path. This is the ​​conservation of basic radiance​​. It’s why you can’t use a magnifying glass to make the image of the sun hotter than the sun's surface itself. The lens increases the solid angle of the light, but the radiance cannot increase. This single, elegant principle governs the flow of light energy through the entire universe.

The ray picture is powerful, but we all know it's an approximation. Light is fundamentally a wave. So, the big question is: can we keep this beautiful phase-space picture when we account for the wave nature of light? The answer is yes, and the tool that lets us do it is one of the most elegant and strange objects in all of physics: the ​​Wigner Distribution Function (WDF)​​.

The Wigner function, let's call it $W(x, p)$, is a kind of phase-space hologram of a light field. It answers the question: "At position $x$, how much light is traveling with momentum $p$?" But because of the uncertainty principle, you can't know both position and momentum perfectly. So the Wigner function is a ​​quasi-probability distribution​​—it behaves almost like a probability, but it can sometimes dip and become negative!

So, what’s the payoff for accepting this weirdness? The magic is in how the WDF propagates. While the light field itself diffracts and interferes in a complex way described by integral equations, its Wigner function simply flows like a fluid of classical particles. The value of the Wigner function at a point in phase space is simply carried along the ray path defined by geometrical optics. An entire, complicated paraxial optical system, described by an ABCD matrix, doesn't diffract the Wigner function; it just performs a linear coordinate transformation on it.

Consider a simple lens that performs a Fourier transform. This is a cornerstone of signal processing and imaging. In the wave picture, it's a complicated integral. But in phase space? It’s just a rotation by 90 degrees! The position axis becomes the momentum axis, and the momentum axis becomes the negative position axis. The generator of this fundamental transformation is simply proportional to the squared radius in phase space, $-(q^2+p^2)$, hinting at the deep Hamiltonian structure familiar from classical and quantum mechanics. This is the ultimate power of phase-space optics: it turns wave propagation into simple geometry.

Now for the best part: those strange negative values. They are not a flaw; they are the Wigner function’s most important feature. They are the signature of wave interference, appearing right in the phase space.

Imagine we have two parallel, separate laser beams. Each beam on its own has a nice, positive, bell-shaped Wigner function, looking much like a classical cloud of particles. But what happens when we combine them coherently? We don't just get the sum of the two bell shapes. In the phase space region directly between them, a ghostly pattern appears: an ​​interference term​​. This term oscillates rapidly between positive and negative values. It is the phase-space picture of the familiar interference fringes you’d see on a screen. This is a phenomenon that a classical distribution of rays could never, ever produce.

These negative regions are the smoking gun for "non-classicality." For any light field that could be described by a classical, stochastic model (like light from a lightbulb or a noisy laser), the Wigner function is always non-negative. It can be interpreted as a genuine probability distribution of rays. But for truly quantum states of light—a single photon, for instance, or a highly organized "Fock state" with exactly $n$ photons—the Wigner function must have negative parts. The volume of these negative regions is a direct, quantitative measure of how "quantum" a state is. The Wigner function of a two-photon state, for example, looks like a central peak surrounded by a negative ring, like a moat of impossibility around a classical castle.

Despite its ghostly nature, the Wigner function is firmly connected to reality. If you want to know the probability of detecting light at a certain position, you simply integrate (sum up) the Wigner function over all possible momenta (angles). If you want to know the probability distribution for a generalized measurement, you just integrate along the appropriate lines in phase space. The negative parts and positive parts conspire to always give a positive, sensible probability for any real-world measurement. It is just one of several ways to represent a quantum state in phase space; it can be seen as a "smoothed-out" version of other, more singular distributions like the Glauber-Sudarshan P-function, related by convolution with a Gaussian kernel.

In the end, phase-space optics gives us a unified canvas. On it, the geometric paths of rays, the wavelike nature of interference, and the strange rules of quantum mechanics are all painted with the same brush. It shows us that these are not separate subjects, but different facets of a single, coherent, and breathtakingly beautiful reality.

We have now learned the rules of this beautiful game, the principles of phase space optics. We have seen that by considering a light beam's position and direction simultaneously, we can create a complete map—a Wigner Distribution Function—that tells us everything there is to know about the beam's state. We've also seen that as light travels through optical systems, this map transforms in wonderfully simple ways.

But learning the rules is one thing; playing the game is another. What can we do with this knowledge? As it turns out, this perspective is not merely an elegant mathematical reformulation. It is an intensely practical tool that unlocks a deeper understanding of the world around us. It allows us to analyze, design, and even correct the instruments we use to explore our universe, from the lenses in our cameras to the microscopes that reveal the machinery of life, and it even gives us a startlingly clear window into the quantum world itself. Let us now explore this vast landscape of applications and see the power of this unified vision in action.

At the heart of optical engineering lies the control of light. The phase space picture provides a new and powerful blueprint for this task, transforming complex wave phenomena into intuitive geometric transformations.

Imagine the simplest optical component: a thin lens. What does it do? In the old view, we might say it bends parallel rays to a focal point. In the phase space view, its action is even more fundamental. A lens imparts a "kick" to each ray, changing its direction (its momentum, $p_x$) by an amount that depends on where it hits the lens (its position, $x$). A ray passing through the center is undeflected, while a ray hitting the edge is bent sharply. In our phase space map, this action is a shear. The entire distribution of points representing the beam is skewed, with the vertical momentum axis tilting into a sloped line. This simple geometric shear is the very essence of focusing.

This idea is astonishingly powerful. It turns out that any ideal paraxial optical system—no matter how complex, with its train of lenses, mirrors, and empty spaces—can be described by a simple linear transformation of the phase space coordinates. This transformation is encapsulated in a small table of four numbers, the famous $2 \times 2$ ray-transfer matrix, or ABCD matrix. This matrix is not just a convenient bookkeeping tool for tracing rays; it is the mathematical operator that transforms the Wigner function from the input to the output of the system. This provides a profound link between the worlds of geometrical ray tracing and physical wave optics. It means we can now design an optical system in reverse: if we know the phase space transformation we want to perform, we can directly calculate the ABCD matrix, and therefore the physical system, required to achieve it.

Of course, the real world is never quite so perfect. Real lenses suffer from aberrations, which are a kind of fundamental "imperfection" in their focusing action. In the phase space picture, aberrations appear as nonlinear warpings of our perfect map. Spherical aberration, coma, astigmatism—each of these classic defects has a unique, recognizable signature as a distortion in phase space. For example, a system with primary coma will produce a local distribution of ray directions that is lopsided and skewed, a feature we can precisely quantify by calculating the moments of the Wigner function at each point in the pupil. Similarly, astigmatism's characteristic distortion can be directly related to its description in the standard language of optical testing, the Zernike polynomials, providing a direct bridge between the abstract phase space and the practical measurements of the optical engineer.

And here is where the true engineering power reveals itself: if we can map an aberration, we can also design its opposite. If a flawed lens introduces an unwanted nonlinear shear in phase space, we can design a special "compensator" plate—a thin piece of glass with a precisely sculpted surface—that introduces the exact inverse shear. By placing this compensator in the beam, we can pre-correct the light, so that when it passes through the aberrated lens, the two distortions cancel each other out, and the final output is that of a perfect, ideal system. This is the fundamental principle behind the sophisticated field of adaptive optics, which uses deformable mirrors to correct for atmospheric turbulence in telescopes, allowing us to see distant stars with breathtaking clarity.

The phase space view also illuminates the wave-like phenomena of diffraction and interference with a new light. A diffraction grating, which splits a single beam of light into many, acts in phase space by taking the initial distribution and "slicing" it into a series of parallel sheets. Each sheet corresponds to one of the discrete diffraction orders, separated in the momentum (angle) dimension, vividly showing how the periodic structure has imparted discrete momentum "kicks" to the light wave. Interference, too, finds a natural home here. When two waves overlap, the Wigner function of the combined field is not just the sum of the individual Wigner functions. An additional "cross-term" appears, oscillating between positive and negative values. This strange, non-positive feature is the signature of wave coherence, the "ghost" of interference captured in phase space, painting a map of the beautiful fringe patterns that are the hallmark of wave mechanics.

Perhaps the most profound beauty of the phase space perspective is its universality. The rules of the game do not just apply to photons; they describe any system governed by Hamiltonian dynamics. The dance of waves and rays is a universal one.

Let us switch from photons to electrons. In a scanning electron microscope (SEM), a beam of electrons is focused down to a tiny spot to scan a specimen. The physics governing the electron trajectories through the electromagnetic lenses is identical in form to the paraxial optics of light. Consequently, Liouville's theorem applies, and there must be a conserved quantity analogous to the radiance of light. This quantity is called the ​​reduced brightness​​. It is a fundamental property of the electron source, representing its phase space density, and it remains constant throughout an ideal microscope column. The total current you can deliver into a spot of a given size and convergence angle is directly proportional to this reduced brightness. This single concept elegantly explains why a high-end microscope with a "bright" field-emission source can achieve far higher resolution and signal than one with a conventional thermionic source. The difference is not just technology; it is the fundamental phase space density of the source itself. The same principle that governs the light in a telescope dictates the ultimate performance of a machine used to see the nano-world.

This way of thinking also inspires new ways to overcome old limitations. For nearly a century, the diffraction limit stood as a seemingly unbreakable barrier, stating that a microscope could never resolve details smaller than about half the wavelength of light. But phase space thinking reveals a loophole. In Structured Illumination Microscopy (SIM), a clever trick is played in frequency space. The sample is illuminated not with uniform light, but with a fine striped pattern. This patterned light mixes with the fine details of the sample. In the frequency domain, this corresponds to a convolution: copies of the sample's spatial frequency spectrum (which contains the information about its structure) are created and shifted. High-frequency information, corresponding to the tiny details normally lost to the diffraction limit, gets shifted down into the frequency range that the microscope can detect. By taking several images with different patterns and then using a computer to "un-mix" the frequency information, a final image is reconstructed with up to twice the resolution of a conventional microscope. We are not breaking the laws of physics, but by intelligently manipulating the information in phase space, we are finding a clever way around them.

Finally, let us close the circle, returning to the very origins of the Wigner function: quantum mechanics. Can we use this tool, which has served us so well in classical optics, to visualize the bizarre world of quantum light? Consider a single two-level atom trapped in a mirrored cavity, interacting with a single mode of the light field—a system described by the Jaynes-Cummings model. The light is initially in a coherent state, the most "classical-like" of quantum states, which appears in phase space as a simple, localized Gaussian blob. As the atom and field interact, the Wigner function evolves in a spectacular fashion. The single blob may split into two, which rotate around the origin in opposite directions, creating a "Schrödinger's cat" state—a quantum superposition of two distinct classical states. Yet, if we wait for a specific time, the so-called "revival time," something magical happens: the complex quantum state re-phases and the light returns to a simple, localized coherent state, just as it started (albeit with a phase shift). The Wigner function allows us to watch this entire quantum revival unfold, providing an intuitive, visual map of one of the deepest and most beautiful phenomena in quantum optics.

From designing better lenses to building sharper microscopes and visualizing the quantum nature of reality, the phase space view proves itself to be far more than an academic exercise. It is a unifying perspective, revealing the deep connections between seemingly disparate fields of science and engineering. It shows us that whether we are guiding a ray of starlight, a beam of electrons, or the quantum state of a photon, the fundamental choreography—the elegant and predictable dance in phase space—is always the same.