Jones Vectors

Light polarization is a fundamental property that describes the intricate oscillation of its electric field. While easy to observe with polarizing sunglasses, describing this behavior with precision requires a dedicated mathematical language. This is the gap filled by the Jones calculus, an elegant and powerful system that provides a complete description of polarized light using simple vectors and matrices. This article delves into the world of Jones vectors, offering a comprehensive overview of this essential tool in optics. The first chapter, "Principles and Mechanisms," will lay the foundation, explaining how complex numbers are used to define linear, circular, and elliptical polarization states. Following this, the "Applications and Interdisciplinary Connections" chapter will demonstrate the immense practical utility of the Jones calculus, from engineering everyday LCD screens to exploring the profound mysteries of quantum entanglement. By the end, the reader will understand not just what Jones vectors are, but why they are an indispensable language in modern science and technology.

Imagine trying to describe a dance. You could say the dancer moved "forward," but that's incomplete. Did they leap? Did they slide? Did they twirl? To capture the full richness of the motion, you need more information. The polarization of light is much the same. It’s not enough to know the direction a light wave is traveling; we must also describe the intricate dance its electric field performs in the plane perpendicular to its motion. The Jones calculus, named after its inventor R. Clark Jones, gives us a wonderfully elegant and powerful language to do just that.

As a light wave zips along, say, the $z$-axis, its electric field oscillates in the transverse $xy$-plane. This oscillation isn't random; it traces a specific, repeating pattern. The simplest patterns are straight lines. If the electric field oscillates only along the $x$-axis, we call it ​​horizontally polarized​​. If it oscillates only along the $y$-axis, it's ​​vertically polarized​​.

How can we write this down? Well, any vector in a 2D plane can be described by its components along two perpendicular axes. So, we can represent the polarization state by a two-element vector, or a "shopping list" with two items: how much "horizontal" it has, and how much "vertical" it has. For purely horizontal light, the list would be (All, None). For vertical, it's (None, All). We can represent this with simple column vectors:

These are the foundational ​​Jones vectors​​. They are like the "North" and "East" on our map of polarization. Notice something beautiful: these two states are not just different, they are fundamentally independent, or ​​orthogonal​​. In the language of vectors, this means their inner product is zero. For Jones vectors, the inner product is a little special (it involves taking the complex conjugate transpose, denoted by $\dagger$), but the result is the same: $J_H^\dagger J_V = (1^* \times 0) + (0^* \times 1) = 0$. This orthogonality is not just a mathematical curiosity; it's the physical reason we can separate light into horizontal and vertical components using a polarizing filter.

But what about other kinds of polarization? What if the light is polarized linearly, but at a 45-degree angle? Its electric field has equal parts horizontal and vertical motion, happening at the same time. Using the principle of superposition, we can just add the vectors:

(The $\frac{1}{\sqrt{2}}$ is just a convention to keep the total intensity equal to one, a process we call normalization).

This seems simple enough. But here we come to a crucial point. The components of our electric field are waves. And waves have not just an amplitude, but also a ​​phase​​—a timing offset. What if the vertical component's oscillation lags slightly behind the horizontal one? Real numbers alone can't keep track of both amplitude and phase simultaneously.

This is where the genius of the Jones calculus shines. We use ​​complex numbers​​. A complex number $A \exp(i\phi)$ is a beautiful little package that contains both a magnitude (amplitude, $A$) and a phase angle ($\phi$). By allowing the components of our Jones vector to be complex, we can describe the full dance of the electric field. Our vector is now:

where $A_x$ and $A_y$ are the real amplitudes, and $\phi_x$ and $\phi_y$ are the phases. The crucial physical information is in the relative phase, $\delta = \phi_y - \phi_x$.

With complex numbers in our toolbox, we can now describe the entire zoo of polarization states.

​​Linear Polarization:​​ As we saw, this occurs when there is no phase difference between the components ($\delta=0$ or $\delta=\pi$). The complex numbers have the same phase angle (or opposite signs), so they can be written as real numbers (after factoring out a common phase). The vector tip just moves back and forth along a line.

​​Circular Polarization:​​ Here's where the magic happens. What if the amplitudes are equal ($A_x = A_y$), and the phase difference is exactly a quarter of a cycle, $\delta = \pm \pi/2$? A phase shift of $\pm \pi/2$ is mathematically represented by multiplying by $\pm i$. For example, consider the vector $\begin{pmatrix} 1 \\ i \end{pmatrix}$. Here, the $y$-component has the same amplitude as the $x$-component but is "ahead" in phase by $\pi/2$. As the $x$-component oscillates like $\cos(\omega t)$, the $y$-component oscillates like $\sin(\omega t)$. The tip of the electric field vector $(\cos(\omega t), \sin(\omega t))$ traces out a perfect circle. This is ​​circularly polarized light​​.

The sign of the phase shift determines the direction of rotation. By convention:

• ​​Left-Circular Polarization (LCP):​​ $y$ leads $x$ by $\pi/2$. Jones vector: $\frac{1}{\sqrt{2}}\begin{pmatrix} 1 \\ i \end{pmatrix}$. The vector rotates counter-clockwise when you look towards the light source.
• ​​Right-Circular Polarization (RCP):​​ $y$ lags $x$ by $\pi/2$. Jones vector: $\frac{1}{\sqrt{2}}\begin{pmatrix} 1 \\ -i \end{pmatrix}$. The vector rotates clockwise.

You can even create circular polarization by cleverly combining two linear polarizations. If you superimpose a wave polarized at $30^\circ$ with another of equal amplitude polarized at $120^\circ$ (which is orthogonal to it), and you give the second wave a $\pi/2$ phase advance, the result is perfectly left-circularly polarized light.

​​Elliptical Polarization:​​ The most general case is ​​elliptical polarization​​. This is what you get for any other combination of amplitudes and phases. The tip of the electric field vector traces out an ellipse. For instance, a vector like $\begin{pmatrix} 2 \\ 3i \end{pmatrix}$ describes elliptically polarized light, since the amplitudes are unequal (2 and 3) and the phase difference is $\pi/2$. In fact, we can "reverse-engineer" the Jones vector for any ellipse we can imagine. If you want an ellipse whose major axis is twice its minor axis, tilted at $45^\circ$, and rotating clockwise, a little bit of trigonometry and complex algebra will tell you that the ratio of the vector's components, $\chi = E_y / E_x$, must be precisely $\frac{3}{5} - \frac{4}{5}i$.

The Jones vector isn't just a descriptive label; it's a mathematical object we can operate on. This gives us a powerful calculus for predicting how polarized light behaves.

​​Intensity:​​ The brightness, or intensity, of the light is proportional to the total energy carried by the wave. This corresponds to the "length squared" of the Jones vector. For a vector $\mathbf{J} = \begin{pmatrix} E_x \\ E_y \end{pmatrix}$, the relative intensity is given by $I = |E_x|^2 + |E_y|^2$. For the vector $\begin{pmatrix} 2 \\ 3i \end{pmatrix}$, the intensity is $|2|^2 + |3i|^2 = 4 + 9 = 13$ units. This gives us a direct physical meaning for the magnitudes of the components. If you pass this light through a horizontal polarizer, you are selecting only the $E_x$ component, and the intensity you measure will be proportional to $|E_x|^2$. This link between vector components and measurement is direct and powerful.

​​Superposition:​​ As we've hinted, if you combine two coherent light beams, the Jones vector of the resulting beam is simply the sum of the individual Jones vectors. For instance, combining a beam of LCP light with a stronger beam of RCP light (with a specific phase shift) results in a new, elliptically polarized state whose properties can be calculated precisely by adding their vectors.

​​Orthogonality:​​ We saw that horizontal and vertical polarizations are orthogonal. The same is true for LCP and RCP. But the concept is even more general: every single polarization state has a unique orthogonal partner. For any given Jones vector $\mathbf{J}_1$, there exists a vector $\mathbf{J}_2$ such that their inner product $\mathbf{J}_1^\dagger \mathbf{J}_2 = 0$. This orthogonal state represents a polarization that can be completely blocked by a filter designed to pass the original state. For instance, the state orthogonal to the elliptical polarization $\begin{pmatrix} 1 \\ -i \tan(\alpha) \end{pmatrix}$ can be found and is given by a vector proportional to $\begin{pmatrix} \tan(\alpha) \\ i \end{pmatrix}$. This principle is the foundation for technologies like polarization multiplexing in fiber optics, where two independent data streams are sent down the same fiber on orthogonal polarization states, effectively doubling the channel's capacity.

This landscape of polarization states has a stunningly beautiful geometric interpretation known as the ​​Poincaré sphere​​. Every possible fully polarized state corresponds to a unique point on the surface of this sphere. The north and south poles could be LCP and RCP, while points on the equator represent all the linear polarizations. And what about our orthogonal states? They are always ​​antipodal points​​—diametrically opposite each other on the sphere. If you have the Stokes vector $\vec{s}_1$ (a 3D real vector that describes the state's position on the sphere) for one state, its orthogonal partner will have the Stokes vector $\vec{s}_2 = -\vec{s}_1$, so their sum is always zero. This provides a deep, intuitive connection between the abstract algebra of Jones vectors and a tangible geometric space.

For all its power, the Jones calculus has a fundamental limitation: it can only describe ​​fully polarized light​​. The light is assumed to be coherent and to have a stable, well-defined polarization state.

But what about the light from the sun, or from an incandescent bulb? This light is ​​unpolarized​​. Its electric field vector changes direction randomly and chaotically on incredibly short timescales. There is no fixed phase relationship between the $x$ and $y$ components. You cannot write down a single, time-independent Jones vector to represent this state. Any specific Jones vector, like $\frac{1}{\sqrt{2}}\begin{pmatrix} 1 \\ 1 \end{pmatrix}$ (linear at 45°) or $\frac{1}{\sqrt{2}}\begin{pmatrix} 1 \\ i \end{pmatrix}$ (LCP), describes a perfectly ordered, fully polarized state. Unpolarized light is the antithesis of this order.

To handle unpolarized or partially polarized light, we must move beyond the Jones vector to a statistical description, using tools like the ​​coherency matrix​​ or the related ​​Stokes parameters​​ and ​​Mueller calculus​​. This doesn't diminish the power of the Jones formalism; it simply defines its proper domain. For the vast world of lasers, optical components, and quantum optics where light is coherent and polarized, the Jones vector remains an indispensable and exquisitely beautiful tool for understanding and manipulating the dance of light.

After all our discussion on the principles and mechanisms of Jones vectors, a practical person might begin to wonder, "This is all very elegant, but what is this mathematical machinery good for?" It is a fair question. The answer, it turns out, is that this simple two-component vector is not just a clever bit of bookkeeping; it is a key that unlocks a vast landscape of technology, explains fundamental physical phenomena, and even provides a bridge to the strange world of quantum mechanics. The true beauty of the Jones calculus lies in its power to transform the complex physics of light's interaction with matter into the straightforward, almost mechanical, operations of matrix algebra.

Imagine you are an optical engineer. Your job is to tame light—to bend it, filter it, and shape its properties for a specific purpose. Your tools are not wrenches and screwdrivers, but pieces of specially crafted glass and crystal: polarizers, wave plates, and rotators. When you place one of these elements in the path of a light beam, it changes the light's polarization. But what happens when you use two, three, or even more elements in a sequence?

This is where the power of Jones calculus shines. Each optical element can be represented by a unique 2x2 matrix, its "Jones matrix." A device that simply rotates the plane of linear polarization by an angle $\theta$, for example, is described by a simple rotation matrix that you may recognize from geometry. The act of light passing through the element is reduced to a simple mathematical operation: multiplying the Jones vector of the incoming light by the Jones matrix of the device. If the light passes through a second device, you simply multiply by its matrix, and so on. The final polarization state is found by a chain of matrix multiplications.

This matrix approach allows us to ask and answer very precise questions. For instance, what happens when circularly polarized light, where the electric field vector spins like a propeller, encounters a linear polarizer, which only allows light oscillating in one direction to pass? The calculus gives a definite answer: the light that emerges is linearly polarized, but with its intensity cut in half.

We can also ask a deeper question: is there any type of light that can pass through a device without changing its form of polarization? In the language of linear algebra, we are asking for the "eigenstates" of the device's matrix. For a simple horizontal polarizer, the answer is intuitively obvious: horizontally polarized light passes through perfectly, while vertically polarized light is completely blocked. The Jones calculus confirms this, showing that the vector for horizontal polarization is an eigenstate with an eigenvalue of 1 (it gets through), while the vector for vertical polarization is an eigenstate with an eigenvalue of 0 (it is extinguished).

With this toolkit, an engineer can design complex optical systems. By calculating the effect of a sequence of components, one can determine the exact orientation of a half-wave plate needed to transform an initial polarization state into a final one that maximizes the light transmitted through a subsequent polarizer. This is not merely an academic exercise; it is the fundamental principle behind technologies we use every day, from the liquid crystal displays (LCDs) on our phones and televisions to the technology behind modern 3D cinema.

Let us now leave the engineer’s workshop and enter the physicist’s playground, where the goal is to understand nature at its most fundamental level. Here, too, the Jones calculus is an indispensable tool.

Light is a wave, and one of the defining properties of waves is that they can be added together—a principle called superposition. If two beams of light are combined, their electric fields add. The Jones calculus handles this beautifully. The resulting polarization is simply the vector sum of the individual Jones vectors. One can, for example, coherently combine right-circularly polarized light with horizontally polarized light of a different intensity and phase to create a completely new state of elliptical polarization, described by a new, unique Jones vector.

This has profound consequences for the classic phenomenon of interference. In Young's double-slit experiment, we see a pattern of light and dark fringes because light waves from the two slits interfere constructively or destructively. But what if the light from each slit has a different polarization? Suppose one slit only lets out vertically polarized light and the other only right-circularly polarized light. At the central maximum on the screen, where the path lengths are equal, the light from both slits arrives together. The resulting polarization is not simply one or the other; it is a coherent superposition of the two, a new state of elliptical polarization determined by the vector sum of the two Jones vectors. Polarization is not just a property carried by the wave; it actively participates in the interference.

The formalism also extends to light-matter interactions, such as scattering. When sunlight, which is unpolarized, scatters off the molecules in the atmosphere, it becomes partially polarized. This is why the sky looks different when viewed through polarized sunglasses. The Jones calculus can precisely describe this. Given an incident beam of any polarization—say, elliptically polarized—that scatters from a tiny particle, we can calculate the exact polarization of the light scattered in any given direction. The polarization of the scattered light carries rich information about the scattering process and the incident light itself. Even in the advanced field of nonlinear optics, where multiple light beams mix inside a material to create new frequencies of light, the polarization of the newly generated light is governed by tensor equations whose vector nature is perfectly captured by the Jones formalism.

By now, you might think that Jones vectors are a special trick developed just for light. Nothing could be further from the truth. The mathematics is so powerful because it describes the behavior of any phenomenon that can be represented as a two-dimensional transverse wave.

Imagine a long, taut string. If you pluck it, a transverse wave travels down its length. For a simple, uniform string, a wave polarized in any direction (up-and-down, side-to-side, or diagonal) travels at the same speed. But what if the string is anisotropic—for instance, a ribbon that is much stiffer against vertical bending than horizontal bending? Then, a vertically polarized wiggle will travel at a different speed than a horizontally polarized one.

If you launch a wave that is linearly polarized at a $45^\circ$ angle, it can be thought of as a sum of a vertical and a horizontal component. Since these components travel at different speeds, one will get progressively out of phase with the other as the wave moves down the string. A wave that started as a simple diagonal oscillation will morph into an elliptical motion, then a circular one, then back to a different linear one, and so on. The state of this mechanical polarization can be described at any point along the string by a Jones vector, and its evolution is governed by exactly the same mathematics we use for light passing through a birefringent crystal. The universe, it seems, likes to reuse its best ideas. The same mathematical structure describes light in an optical fiber, mechanical waves in an anisotropic solid, and even certain types of waves in plasma physics.

We have saved the most remarkable connection for last—a leap from the familiar world of classical waves into the strange and wonderful realm of quantum mechanics. Light is not just a wave; it is also made of discrete particles called photons. Each individual photon carries polarization.

Here is the profound connection: the classical Jones vector corresponds directly to the quantum state vector of a single photon’s polarization. A photon in a horizontally polarized state is described by the quantum state $|H\rangle$, which maps to the Jones vector $\begin{pmatrix} 1 \\ 0 \end{pmatrix}$. A photon in a vertical state, $|V\rangle$, maps to $\begin{pmatrix} 0 \\ 1 \end{pmatrix}$. The complex numbers in the Jones vector, which we interpreted as the amplitudes and phases of a classical wave, take on a new meaning in the quantum world: they become probability amplitudes. The squared magnitude of each component gives the probability of measuring the photon to be in the corresponding basis state.

This connection provides a stunningly direct bridge to one of physics' deepest mysteries: quantum entanglement. Certain sources can produce pairs of photons that are linked in a special way. For example, in the Bell state $|\Psi^-\rangle = \frac{1}{\sqrt{2}}(|H_A V_B\rangle - |V_A H_B\rangle)$, if you measure photon A to be horizontal, you are guaranteed that its partner, photon B, is vertical, and vice-versa—no matter how far apart they are.

Now, suppose you decide to measure photon B not in the horizontal/vertical basis, but by projecting it onto some arbitrary elliptical polarization state. The moment you detect photon B in that state, you have instantaneously "prepared" photon A in a specific, corresponding polarization state. What is this state? The Jones calculus gives the answer directly. By applying the rules of quantum measurement, we can derive the precise Jones vector of the heralded photon A. Our simple, two-component vector, originally invented to describe classical waves, becomes the descriptor for a "spooky action at a distance." It is the language of the single-photon polarization qubit, the fundamental unit of information in many quantum computing and quantum communication schemes.

From engineering LCDs to explaining the blue sky, from unifying the physics of different waves to describing the mysteries of quantum entanglement, the Jones vector formalism proves to be far more than a mathematical convenience. It is a testament to the power of finding the right language to describe the world, a language that reveals the deep and beautiful unity of physical law.