Jones Matrix

Light is not just a ray traveling in a straight line; it possesses an internal, dynamic property called polarization, which describes the orientation of its oscillating electric field. Manipulating this property is fundamental to countless technologies, but to control it, we first need a language to describe it with precision. This presents a key challenge: how can we create a predictive mathematical framework to choreograph the complex "dance" of light as it passes through various optical materials?

The answer lies in the elegant Jones calculus, a formalism developed by R. Clark Jones. This article will guide you through this powerful framework. In the "Principles and Mechanisms" chapter, you will learn the fundamental building blocks: the Jones vectors that describe polarization states and the Jones matrices that represent optical elements like polarizers and wave plates. Following that, the "Applications and Interdisciplinary Connections" chapter will reveal how these simple mathematical rules are combined to design sophisticated technologies, from LCD screens and gravitational wave detectors to models used in modern ophthalmology.

Imagine you're watching a dancer on a stage. From a distance, you might only see their path across the stage—forward, backward, left, right. But if you look closer, you see the intricate movements of their arms, the tilt of their head, the spin of their body. Light is much the same. As a light wave travels from point A to B, it's not just moving; it's also "dancing." Its electric field oscillates—up and down, side to side, or even in a spinning, spiraling motion. This dance is called ​​polarization​​.

To understand and control this dance, we need a language, a form of choreography. This is precisely what the ​​Jones calculus​​ gives us. It's a wonderfully elegant mathematical framework, developed by the American physicist R. Clark Jones in the 1940s, that allows us to describe the dance of light and predict how it will change when it passes through different materials. It's like a choreographer's notebook, filled with precise instructions for creating any polarization effect we can imagine.

First, how do we write down a description of the light's dance? We simplify. We look at the light head-on as it travels towards us, and we see its electric field tracing a shape in a two-dimensional plane. We can describe any possible motion in this plane as a combination of two basic, perpendicular movements: a horizontal oscillation (along an x-axis) and a vertical oscillation (along a y-axis).

In the Jones calculus, we represent the state of polarization with a simple two-number column vector, the ​​Jones vector​​.

• Purely horizontal polarization is the fundamental step $\begin{pmatrix} 1 \\ 0 \end{pmatrix}$.
• Purely vertical polarization is its partner, $\begin{pmatrix} 0 \\ 1 \end{pmatrix}$.

What about more complex dances? Any polarization state can be described as a combination of these two. For example, light polarized at a $45^\circ$ angle is an equal mix of horizontal and vertical, written as $\frac{1}{\sqrt{2}}\begin{pmatrix} 1 \\ 1 \end{pmatrix}$. The magic ingredient that allows for spinning, circular, and elliptical dances is the use of complex numbers. A phase difference between the horizontal and vertical components, represented by an imaginary number $i$, can turn a simple back-and-forth oscillation into a graceful pirouette. For instance, $\frac{1}{\sqrt{2}}\begin{pmatrix} 1 \\ i \end{pmatrix}$ describes light that is spinning in a perfect circle!

Now that we can describe a polarization state, how do we describe changing it? We use $2 \times 2$ matrices, the ​​Jones matrices​​. A Jones matrix is an operator, a set of instructions that transforms an incoming Jones vector into a new, outgoing Jones vector. Each optical element—a filter, a crystal, a piece of plastic—has its own characteristic Jones matrix.

Let's meet the cast of our optical ballet.

The Filter: The Linear Polarizer

The most straightforward operation is simply to block one direction of polarization while letting the other through. This is what a ​​linear polarizer​​ does. Think of it as a gate with vertical slots. Only the vertical component of the light's dance can pass through. If the incoming light has components $\begin{pmatrix} E_x \\ E_y \end{pmatrix}$, a perfect vertical polarizer will block the horizontal part ($E_x$) and transmit the vertical part ($E_y$). The outgoing light will be $\begin{pmatrix} 0 \\ E_y \end{pmatrix}$. What matrix accomplishes this? As shown in, the instruction is beautifully simple:

Multiplying this matrix by any input vector instantly sets the top component (horizontal) to zero, just as we wanted. Similarly, a horizontal polarizer that lets only the x-component pass is described by $\begin{pmatrix} 1 0 \\ 0 0 \end{pmatrix}$.

The Twister: The Optical Rotator

Some materials have a fascinating, chiral structure that causes them to twist the plane of polarization as light passes through. This is like making the dancer do a full-body turn. If we want to rotate the polarization by an angle $\theta$, the Jones matrix is exactly the standard 2D rotation matrix you might have learned in geometry class:

This is a profound connection. The abstract "polarization space" behaves just like our familiar physical space, and the same mathematical tools apply.

The Phase Shifter: The Wave Plate

The most subtle and powerful choreographers are the ​​wave plates​​ or ​​retarders​​. These elements don't block light or twist its plane; instead, they introduce a time delay, or ​​phase shift​​, between the two orthogonal components. They make one component of the dance lag behind the other.

A ​​quarter-wave plate (QWP)​​, for example, introduces a quarter-cycle ($\pi/2$ radians or $90^\circ$) phase shift. Its simplest form, with its "fast" axis horizontal, has a Jones matrix like this:

This matrix leaves the horizontal component alone but multiplies the vertical component by $-i$. This subtle phase shift has dramatic consequences. If you send linearly polarized light at $45^\circ$ (the vector $\frac{1}{\sqrt{2}}\begin{pmatrix} 1 \\ 1 \end{pmatrix}$) into this QWP, the output is $\frac{1}{\sqrt{2}}\begin{pmatrix} 1 \\ -i \end{pmatrix}$—circularly polarized light! The linear back-and-forth motion has been transformed into a spinning motion.

The real power of the Jones calculus shines when we start combining optical elements. If light passes through element 1 (with matrix $J_1$) and then through element 2 (with matrix $J_2$), the total effect on the light is described by a single matrix, $J_{\text{total}}$, which is simply the product of the individual matrices. But be careful! Just like putting on your shoes and then your socks doesn't work well, the order of optical elements matters. The matrix for the element the light hits last comes first in the multiplication:

Let's see a beautiful example of this symphony in action. What happens if we place two identical quarter-wave plates back-to-back? The total matrix is:

The resulting matrix introduces a phase shift of $-1$, which is equivalent to a phase shift of $\pi$ radians ($180^\circ$). This is the matrix for a ​​half-wave plate (HWP)​​! So, two quarter-delays in a row create one half-delay. The math doesn't just give an answer; it reveals a deep physical connection.

This naturally brings up the question: does the order ever not matter? When does $J_1 J_2 = J_2 J_1$? The mathematics gives a clear answer. For two wave plates, the system's behavior is independent of their order if (1) one of the plates has no effect (it's a full-wave plate), or (2) their fast axes are either perfectly aligned or perfectly perpendicular. In any other orientation, swapping the two elements will change the final polarization of the light.

So far, we've mostly considered elements aligned neatly with our x and y axes. But what about a polarizer at $17^\circ$ or a wave plate at $63^\circ$? The Jones calculus handles this with supreme elegance using a concept known as a ​​similarity transformation​​. The logic is simple and beautiful:

1. ​​Rotate into the element's frame:​​ Apply a rotation matrix $R(\theta)$ to the incoming light, to see it from the element's point of view.
2. ​​Apply the simple element:​​ Use the standard, simple Jones matrix for the element, $J_0$.
3. ​​Rotate back to the lab frame:​​ Apply the reverse rotation, $R(-\theta)$, to the outgoing light to see the final result in our original coordinate system.

The complete matrix for the rotated element becomes $J(\theta) = R(-\theta) J_0 R(\theta)$. This powerful three-step recipe allows us to construct the Jones matrix for any basic element at any orientation, creating a truly universal toolkit.

This leads us to an even deeper and more beautiful idea. Any optical system, no matter how complex—a stack of 10 different plates and polarizers—has its own special, characteristic polarizations. These are its ​​eigenpolarizations​​ (from the German word eigen, meaning "own" or "characteristic").

An eigenpolarization is a state of light that passes through the entire complex system without changing its form. It might be dimmed, brightened, or phase-shifted, but it emerges with the exact same polarization shape it had going in. It is the system's "natural" mode. Most polarization states will be twisted and transformed into something completely different, but the two eigenpolarizations for any given system are unique. The factor by which they are scaled is their corresponding ​​eigenvalue​​.

The mathematical properties of Jones matrices give us incredible physical insight. For instance, what kind of element is described by a matrix that is both ​​Unitary​​ (conserves the light's total energy) and ​​Hermitian​​ (a type of mathematical symmetry)? Starting from these abstract conditions, a bit of mathematical detective work leads to an astonishingly specific conclusion: the element must be a ​​half-wave plate​​. The mathematical constraints force the matrix to have a determinant of $-1$ and eigenvalues of $+1$ and $-1$. This means there is one polarization (the fast axis) that passes through unchanged, and its orthogonal partner (the slow axis) that gets its phase flipped by $180^\circ$.

This is the true beauty of the Jones matrix formalism. It is not just a computational tool. It is a language that unifies the geometry of rotation, the algebra of complex numbers, and the physics of light. It shows us that abstract mathematical properties like being Hermitian or Unitary are not just curiosities; they are the bedrock of physical laws like energy conservation. With this elegant notebook, we can not only choreograph the dance of light but also understand the fundamental rules that govern the performance.

We have spent some time learning the rules of the game—how to write down these curious two-by-two matrices for polarizers and wave plates. But the real fun, the true power of a physical theory, comes not from describing the pieces, but from seeing what happens when you put them together. The Jones calculus is not just a bookkeeping method; it is a design tool, a lens through which we can understand, predict, and invent. It is by multiplying these matrices, by cascading one optical element after another, that we transform a set of simple rules into a powerful engine for discovery and technology. Let us now take a journey through some of the remarkable places this engine can take us.

Imagine you have a box of simple building blocks—linear polarizers and wave plates. What can you build? A novice might just stack them, but a master architect sees how they can be combined to create functions far more subtle and powerful than the individual parts.

A straightforward, yet essential, construction is a ​​circular polarizer​​. We know how to make linearly polarized light, but what if we need circularly polarized light? The Jones calculus provides the recipe. We take a linear polarizer to create a well-defined initial state, and then we pass this light through a quarter-wave plate, oriented just right—at a $45^\circ$ angle to the polarizer's axis. When we multiply the Jones matrix for the quarter-wave plate by the matrix for the polarizer, the resulting matrix is no longer a polarizer or a wave plate. It is a new object, a matrix that takes any incoming light and transforms it into a right-circularly polarized state. By simply reversing the order of the components, we can build a ​​circular analyzer​​, a device that selectively transmits only one circular polarization state while extinguishing the other. This is the essence of optical design: combining simple, known elements to engineer a desired outcome.

Now for a more beautiful and surprising trick. A half-wave plate is a fascinating device. If you send in linearly polarized light at an angle $\theta$ to its fast axis, the output light is also linearly polarized, but at an angle $-\theta$ on the other side of the axis. It acts like a polarization mirror. So, what happens if we put two half-wave plates in a row? Your intuition might suggest that two reflections should bring you back to where you started. But the matrix multiplication reveals a wonderful secret. A system of two half-wave plates with their fast axes at angles $\theta_1$ and $\theta_2$ is perfectly equivalent to a single, pure ​​optical rotator​​ that rotates the plane of polarization by an angle of $2(\theta_2 - \theta_1)$. This is a profound result! Two components that individually "flip" polarization combine to produce a smooth, continuous "rotation." The final orientation depends only on the relative angle between the two plates, giving us a simple, mechanical way to build a variable rotator.

The real world of optics is filled with more than just simple, straight paths. Light bounces off mirrors, resonates in cavities, and sometimes, it even travels down a one-way street. The Jones calculus is our reliable guide through these more complex journeys.

Consider what happens when light passes through a quarter-wave plate, reflects off a normal mirror, and travels back through the same wave plate. This double-pass configuration is common in many laboratory setups. Since the light's direction is reversed, its "handedness" with respect to the lab frame flips. The Jones formalism handles this with elegant simplicity. The round-trip journey is not a return to the start. Instead, the combination of a quarter-wave plate and a mirror acts as a rotator whose angle is twice the angle of the wave plate's axis. This is an incredibly useful trick, for example, to rotate polarization by a precise amount without introducing other effects.

This leads us to an even more profound concept: ​​non-reciprocity​​. For most optical paths, the journey forward is the same as the journey backward. But it doesn't have to be. The Faraday effect, where a magnetic field rotates the polarization of light, is non-reciprocal. A clockwise rotation for forward-propagating light remains a clockwise rotation from the observer's perspective even when the light travels backward. This breaks the symmetry of the path. By combining a Faraday rotator with polarizers, one can construct an ​​optical isolator​​, a device that acts as a one-way valve for light. It lets light pass in one direction but blocks it completely in the reverse direction. This is not just a curiosity; it is a critical component in modern laser systems, protecting the delicate laser source from its own destabilizing back-reflections.

The pinnacle of using polarization to control light's path is found in modern interferometers, such as the LIGO gravitational wave detectors. In one common design, light from a laser is split by a polarizing beam splitter (PBS), which transmits one polarization ($p$) and reflects another ($s$). One of these beams travels down an arm of the interferometer, where it passes through a quarter-wave plate, reflects off a mirror, and comes back through the wave plate again. As we've seen, this double pass rotates the polarization. The genius of this design is that the rotation is precisely $90^\circ$. An initially $p$-polarized beam becomes $s$-polarized upon its return. When it arrives back at the PBS, it is now in the state that gets reflected, not transmitted, sending it towards a detector. The Jones matrix calculation for this entire round trip beautifully confirms this result, showing how polarization can be used to route beams and create an exquisitely sensitive instrument. The same logic can be extended to analyze resonant optical cavities like Fabry-Pérot interferometers, where polarization effects within the cavity can create highly selective filters for both frequency and polarization state.

The elegance of the Jones calculus is that it is not confined to the pristine environment of the optics bench. The principles of polarization are universal, and this mathematical language can describe phenomena in materials science, technology, and even biology.

Perhaps the most ubiquitous application is sitting right in front of you: a ​​Liquid Crystal Display (LCD)​​. These screens work by manipulating light's polarization on a massive scale. A typical LCD pixel contains a thin layer of a twisted nematic liquid crystal sandwiched between two crossed polarizers. In its "off" state, the rod-like liquid crystal molecules are arranged in a gentle spiral. As polarized light enters, its polarization axis is "guided" by this spiral structure, rotating along with the molecules. It arrives at the second polarizer with its polarization rotated by $90^\circ$, allowing it to pass through, and the pixel appears bright. When a voltage is applied, the molecules untwist and align with the electric field. Now, the light's polarization is no longer guided; it arrives at the second polarizer unrotated and is blocked. The pixel appears dark. Analyzing this requires a more advanced form of the Jones calculus, where the matrix evolves continuously through the material, but the fundamental principle is the same. It is a spectacular marriage of materials science and polarization optics.

The journey doesn't stop with technology. It extends into us. Our own bodies contain optically active structures. In a simplified but powerful model of the ​​human eye​​, the cornea and the Henle fiber layer near the fovea both exhibit birefringence—they act as linear retarders. Light traveling to our photoreceptors passes through this cascade of two biological wave plates, each with its own retardance and orientation. By multiplying their respective Jones matrices, we can calculate the net polarization effect of the eye's anterior structures. This is not just an academic exercise; such models are crucial in ophthalmology for techniques like scanning laser polarimetry, which can measure the thickness of the nerve fiber layer to help diagnose diseases like glaucoma.

From designing a simple filter to modeling the screen you are reading this on, from protecting a laser to detecting gravitational waves and understanding the human eye, the Jones calculus provides a single, coherent, and surprisingly simple mathematical framework. Its power lies in its composition rule—the simple act of matrix multiplication that allows us to build the world's most complex optical systems from a handful of basic principles.

Of course, this is not the end of the story. The Jones calculus is perfect for describing fully polarized light and its interaction with non-depolarizing elements. But what about scattered light, or light from an incandescent bulb, which is unpolarized? What about optical elements that scramble polarization? For these more complex situations, the Jones formalism must be extended to a more general framework known as the ​​Mueller calculus​​, which uses $4 \times 4$ matrices to describe intensity and depolarization. Yet, the Jones calculus remains the essential starting point—the clear, elegant, and powerful language for the world of coherent polarization. It is a testament to the beauty of physics that such a wide array of phenomena can be understood with these simple two-by-two arrays of numbers.