Diattenuation

In the study of light, few properties are as fundamental as polarization. Yet, beyond the simple concept of a perfect polarizing filter, there lies a more subtle and universal phenomenon: diattenuation. This is the property by which nearly all real-world materials and optical systems transmit light of different polarizations with slightly different efficiencies. Often dismissed as a minor imperfection or a source of noise in sensitive systems, this effect is, in fact, a rich source of information and a principle that connects seemingly disparate fields of science. This article addresses the knowledge gap between the textbook definition of polarization and the profound real-world consequences of its imperfect, differential transmission.

Across the following sections, we will embark on a journey to understand this powerful concept. The first chapter, "Principles and Mechanisms," will deconstruct diattenuation, moving from intuitive analogies to the elegant mathematical formalisms of the Mueller calculus that allow us to precisely quantify it, revealing its deep relationship with other optical properties. Subsequently, the chapter on "Applications and Interdisciplinary Connections" will showcase the astonishing breadth of diattenuation's impact, demonstrating how this single principle is at once a challenge for engineers, a tool for medical diagnosticians, and a key to understanding phenomena in the quantum and cosmic arenas.

Imagine you're trying to slide a long, flat ruler through a picket fence. If you hold the ruler vertically, it slips right through the gaps. But if you turn it horizontally, it clangs against the pickets and stops. The fence, in essence, is a filter that cares deeply about the orientation of objects passing through it. In the world of optics, many materials act like a microscopic picket fence for light. They let light with a certain polarization pass through more easily than light with an orthogonal polarization. This fundamental property is called ​​diattenuation​​.

After our introduction, you might be thinking of a perfect polarizer, like the lens of your sunglasses, as an all-or-nothing filter. But nature is rarely so absolute. Most of the time, the effect is more subtle. An optical element might transmit, say, 99% of vertically polarized light but only 95% of horizontally polarized light. Diattenuation is the concept that precisely quantifies this preference. It's not just about what's blocked, but about the difference in what gets through.

How significant is a small difference in transmission? In engineering fields like fiber optics, this effect is often measured as ​​Polarization-Dependent Loss (PDL)​​, which is just diattenuation expressed on the logarithmic decibel (dB) scale. Let's say an engineer tells you a component has a PDL of just $0.1$ dB. That sounds minuscule, almost negligible. But what does it actually mean for the light?

The decibel scale is defined such that $PDL = 10 \log_{10}(P_{max}/P_{min})$, where $P_{max}$ and $P_{min}$ are the maximum and minimum transmitted powers as you cycle through all possible input polarizations. A quick calculation shows that a $0.1$ dB PDL means the ratio $P_{max}/P_{min}$ is $10^{0.1/10} = 10^{0.01}$, which is about $1.0233$. This means the maximum power getting through is about $2.3\%$ higher than the minimum power. So, a seemingly tiny $0.1$ dB loss actually corresponds to a measurable few-percent flicker in intensity if the input polarization changes. In high-speed communication systems where every photon counts, this "flicker" can introduce errors and degrade performance.

To get a more universal and intuitive physical measure, we often use the normalized diattenuation parameter, $D$, defined as:

Here, $T_{max}$ and $T_{min}$ are the maximum and minimum intensity transmittances (the fraction of power that gets through). This value ranges from $0$ for an element that treats all polarizations equally (like a clear window pane) to $1$ for a perfect polarizer that completely blocks one polarization. An element with a $D$ of $0.6$ is a fairly strong diattenuator, while one with a $D$ of $0.01$ is very weak.

So, what is the underlying mechanism that gives rise to this behavior? Let's peel back the layers. The polarization of light is described by its electric field vector. We can think of this vector in terms of its components along two orthogonal axes, say $x$ and $y$. When light enters a material, these two components can be treated differently.

Imagine an optical element that has two special "principal axes". Light polarized along one axis might experience an amplitude reduction by a factor $p_1$, while light polarized along the other axis is reduced by a different factor, $p_2$. This difference in amplitude reduction is called ​​linear dichroism​​. The element might also delay the phases of these two components by different amounts ($\phi_1$ and $\phi_2$), a property known as ​​birefringence​​, which can change the shape of the polarization (e.g., turning linear into elliptical).

Now for a wonderfully simple insight: the diattenuation of the element depends only on the difference in amplitude reduction, not on the phase delays. It doesn't matter how much one component is delayed relative to the other; the overall power loss is determined solely by the attenuation factors. The maximum and minimum transmittances are simply $T_{max} = p_1^2$ and $T_{min} = p_2^2$ (assuming $p_1 > p_2$). Plugging this into our definition gives the diattenuation as:

This elegant formula tells us that diattenuation and birefringence are fundamentally distinct phenomena, even if they often occur in the same device.

This idea leads to a profound and beautiful result from linear algebra called the ​​Singular Value Decomposition (SVD)​​. The SVD theorem tells us that the action of any complex, non-depolarizing optical element can be broken down into a simple, universal three-step recipe. First, the light passes through a "pure retarder," which changes its polarization state without any loss. Second, it passes through a simple "picket fence" diattenuator, aligned perfectly with our coordinate axes, which is responsible for all the polarization-dependent loss. Finally, it passes through a second pure retarder.

This is remarkable! It means that no matter how bizarre or complicated an optical element seems, its diattenuating behavior is equivalent to that of a simple, idealized partial polarizer sandwiched between two lossless polarization transformers. The strength of this central elemental polarizer, its diattenuation $D$, is the diattenuation of the entire complex device. Physics has a wonderful habit of hiding such simplicity within apparent complexity.

The Jones calculus we've alluded to is perfect for fully polarized light where we know the phase. But what about partially polarized light from a distant star, or light that has been scrambled after reflecting off a rough surface? For these cases, we need a more robust framework: the ​​Stokes-Mueller calculus​​.

Here, the polarization state is described not by a two-component complex vector, but by a four-component real vector called the ​​Stokes vector​​, $S = [S_0, S_1, S_2, S_3]^\text{T}$. $S_0$ is the total intensity, and the other three components describe the tendency towards horizontal/vertical, $\pm 45^\circ$, or right/left circular polarization. The optical element is now a $4 \times 4$ real matrix, the ​​Mueller matrix​​ $M$.

The magic happens when we look at how the output intensity, $S'_{0}$, is calculated:

Notice that the output intensity depends only on the first row of the Mueller matrix. The element $m_{00}$ represents the average transmittance for unpolarized light. The other three elements, $(m_{01}, m_{02}, m_{03})$, form what we can call the ​​diattenuation vector​​. This vector lives in the same abstract space as the polarization part of the Stokes vector.

To find the maximum possible transmission, you simply need to send in light whose polarization state (in Stokes space) is perfectly aligned with this diattenuation vector. The minimum transmission occurs when you send in light with the opposite polarization state. This geometric picture gives us an incredibly powerful and direct formula for diattenuation, derived directly from the elements of the Mueller matrix that are easiest to measure:

This means we no longer have to painstakingly search through all possible input polarizations to find the maximum and minimum. We can characterize the diattenuating properties of an entire system—even a "black box"—just by measuring that first row of its Mueller matrix. The system reveals its fundamental polarization preference through these four numbers. Furthermore, when light makes multiple passes through a diattenuating system, like in a laser resonator, the effect compounds dramatically, scaling with the number of passes.

You would be forgiven for thinking that diattenuation is only about, well, attenuation. It's about how much light you lose. But the universe is more subtle and interconnected than that. One of the most beautiful and counter-intuitive consequences of diattenuation appears when we consider how it might change with the color, or frequency, of light.

Imagine sending a short pulse of light, which is composed of many different frequencies, through an optical fiber. The fiber not only has some diattenuation but also ​​Polarization Mode Dispersion (PMD)​​, which means different polarizations travel at different speeds. Now, let's add a twist: what if the diattenuation itself is frequency-dependent? For instance, perhaps blue light is attenuated more strongly than red light for a specific polarization. This "slope" of the PDL with respect to frequency has a startling effect.

It turns out that this frequency-dependent loss can actually change the arrival time of the pulse. A pulse component that experiences a loss that increases with frequency will have its "center of mass" in time shifted, making it appear to arrive earlier. It's a ghostly effect—a property related to loss is directly influencing a property related to timing.

This reveals a deep connection in physics, where the absorptive (lossy) and refractive (phase/delay) properties of a medium are inextricably linked. They are like two sides of the same coin. The most complete modern theories describe the evolution of a pulse's polarization using a single differential equation that combines a PMD vector, $\vec{\tau}$ (describing the speed differences), and a PDL vector, $\vec{\alpha}$ (describing the loss differences), into one unified mathematical object.

So, diattenuation is far more than a simple filter. It is a fundamental property of how light interacts with matter, a property that is elegantly described by our mathematical formalisms and one that has surprising and profound consequences, linking the intensity of light to its very timing. It is a perfect example of the hidden unity and beauty that physics strives to uncover.

Having journeyed through the principles and mechanisms of diattenuation, you might be left with the impression that it's a somewhat specialized topic, a curious detail of polarization optics. But to leave it there would be like learning the rules of chess without ever seeing the beauty of a grandmaster's game. The real magic of a physical principle is not in its definition, but in the astonishing breadth of its consequences. Diattenuation is no exception. It is a thread that, once you start pulling on it, unravels connections across engineering, biology, quantum mechanics, and even the cosmos. It is at once a practical nuisance, a powerful diagnostic tool, and a window into the fundamental fabric of reality.

Let's start in a very practical place: the global network of optical fibers that underpins our internet. We pump vast amounts of information through these glass threads as pulses of light. Ideally, the power that goes in is the power that comes out (minus some uniform, predictable loss). But real-world fibers are not perfect. They contain microscopic, frozen-in stresses and imperfections from the manufacturing process. These imperfections make the glass slightly anisotropic, meaning it responds differently to different polarizations of light.

This leads to what engineers call Polarization-Dependent Loss (PDL). Imagine sending a signal of initially unpolarized light down a long fiber. Due to various environmental effects, the light's polarization state will randomly scramble as it travels. If the fiber system contains elements with even a small amount of diattenuation, the output power will flicker unpredictably as the light's polarization state changes. For a high-speed communication system trying to distinguish between a '1' and a '0', this unwanted power fluctuation is a serious problem, a source of noise that can corrupt data and limit performance. The origin of this effect can be traced back to the very nature of light scattering within the glass. The tiny, anisotropic stress domains that cause the diattenuation do so by scattering light differently depending on its polarization, a beautiful link between a macroscopic engineering problem and the microscopic physics of the material.

But here, as is so often the case in science, one person's noise is another person's signal. If the diattenuation is sensitive to the environment, can we use it to measure that environment? The answer is a resounding yes. Consider a cleverly designed device called a Tilted Fiber Bragg Grating (TFBG). By etching periodic lines at an angle into the fiber's core, we create a device that is exquisitely sensitive to the polarization of light passing through it—it is a strong diattenuator. The strength of this diattenuation, however, depends critically on the refractive index of the medium surrounding the fiber. If you dip this fiber into a liquid, the diattenuation changes. By precisely measuring this change, we can build an incredibly sensitive sensor for chemical concentrations, biological processes, and more. A frustrating bug in telecommunications becomes a brilliant feature in sensing technology.

The power of diattenuation extends far beyond fibers. It allows us to characterize materials in ways that would be impossible with simple imaging. Many biological tissues, for instance, are full of oriented fibrous structures, such as the highly aligned collagen in a tendon or the cytoskeleton within a cell. These structures are, by their very nature, optically anisotropic. They exhibit diattenuation.

By using a technique called Mueller matrix polarimetry, we can get a complete fingerprint of how a sample transforms polarized light. We shine light with various known polarizations onto a tissue sample and measure the full polarization state of the light that passes through. The result is a $4 \times 4$ matrix—the Mueller matrix—that contains a wealth of information. Certain elements of this matrix directly correspond to the amount of linear diattenuation in the sample, while other elements reveal properties like birefringence. By mapping these values, we can create images that reveal the structural integrity and alignment of tissues, potentially helping to diagnose diseases or study tissue development without the need for chemical stains or invasive methods.

This sensitivity, however, can also be a source of subtle deception in other high-precision measurements. In the world of modern astronomy, telescopes use sophisticated adaptive optics systems to correct for the blurring caused by Earth's atmosphere. These systems often use a wavefront sensor, like a Shack-Hartmann sensor, to measure the shape of the incoming light wave. But what if one of the optical components in the telescope—a mirror or a lens—has a slight diattenuation that varies across its surface? It turns out that this gradient in diattenuation can, through the intricate dance of vector diffraction, create a signal in the wavefront sensor that looks exactly like a physical aberration, such as astigmatism. The instrument is fooled into "seeing" a distortion that isn't really there in the incoming starlight. For designers of the next generation of extremely large telescopes, accounting for these "polarization aberrations" is a critical and fascinating challenge, revealing a deep and often-overlooked coupling between the polarization and spatial properties of light.

So far, our examples have all involved classical light waves. But the principle of anisotropic attenuation is far more universal. It echoes in the quantum world and in the most extreme environments of the cosmos.

Consider the strange and delicate resource of quantum entanglement. In a quantum communication system, we might send a pair of photons that are entangled in their polarization—measuring one instantly affects the other, no matter how far apart they are. What happens if these photons travel through a medium with diattenuation? The diattenuation acts like a "polarization-selective filter" on the quantum state. If, for example, vertically polarized photons are attenuated more than horizontally polarized ones, the part of the entangled state corresponding to vertical polarization is suppressed. This act of "selective listening" damages the delicate superposition, reducing the fidelity of the entanglement. Understanding and mitigating these effects is paramount for building robust quantum networks. Diattenuation, a classical concept, becomes a direct source of quantum decoherence.

Now, let's take a truly giant leap—from the quantum realm to the bizarre physics of superfluids. Below about 2.17 Kelvin, liquid helium enters a quantum state of matter known as a superfluid. In this phase, it can be described by a "two-fluid model," behaving as if it's a mix of a normal, viscous fluid and a frictionless superfluid. One of the most curious phenomena in this state is "second sound," which is not a wave of pressure like normal sound, but a wave of temperature and entropy.

What happens if we rotate a container of superfluid helium? The superfluid component forms a dense, regular array of tiny quantum whirlpools, or "vortex lines," all aligned with the axis of rotation. This vortex lattice creates an anisotropic medium. Now, if we send a wave of second sound through this medium, it experiences an attenuation due to "mutual friction" with the vortex lines. And—here is the beautiful part—the amount of attenuation depends on the direction the wave is traveling relative to the aligned vortices. A wave traveling perpendicular to the vortices is attenuated differently from a wave traveling at an angle. This is a perfect analog of diattenuation! The vortex lattice acts as a "polarizer" for thermal waves. The same mathematical structure describing the anisotropic absorption of polarized light applies to the absorption of thermal waves in a quantum fluid. This principle holds true even if the vortices are in a disordered, anisotropic "tangle" rather than a perfect lattice, and it similarly affects conventional sound waves (phonons) propagating through the fluid.

Finally, let us cast our gaze to the heavens, to the most powerful magnets in the universe: magnetars. The magnetic fields around these neutron stars are so stupendously strong that they warp the very fabric of spacetime and make the vacuum itself behave like an optical medium. According to the laws of quantum electrodynamics (QED), a high-energy photon traveling through such a field can split into two lower-energy photons—a process forbidden in empty space. This process leads to an attenuation of the original photon beam. Crucially, the probability of this splitting happening depends on the polarization of the photon relative to the external magnetic field. A photon polarized parallel to the magnetic field might pass through unscathed, while one polarized perpendicularly is heavily attenuated. The vacuum of space, in the presence of an extreme magnetic field, becomes a diattenuator.

From the practical hum of a fiber optic network to the silent dance of quantum vortices and the violent heart of a distant star, the principle of diattenuation reappears. It is a testament to the profound unity of physics: a simple idea—that interaction can depend on orientation—plays out on vastly different stages, with a cast of characters ranging from photons and phonons to the quantum vacuum itself. It is a beautiful and powerful refrain in the grand symphony of the universe.