Position Representation

In the abstract realm of quantum mechanics, a particle's state is encapsulated by a vector in an infinite-dimensional Hilbert space. While mathematically elegant, this description feels distant from the tangible world of positions and momenta that we measure. How do we bridge this gap between abstract theory and experimental reality? The answer lies in the concept of a ​​representation​​—a 'coordinate system' for quantum states. This article delves into the most fundamental of these: the position representation. It addresses the crucial problem of how to translate abstract quantum information into a concrete, functional form that we can calculate with and understand. In the chapters that follow, we will first explore the principles and mechanisms of the position representation, learning how state vectors become wavefunctions and abstract operators become concrete actions. Then, we will journey beyond quantum mechanics to uncover the surprising and profound interdisciplinary connections, revealing how this same idea unifies our understanding of everything from molecules to the cosmos.

Now that we have a taste of what quantum mechanics is about, let's roll up our sleeves and look under the hood. The abstract world of state vectors in Hilbert space is mathematically pristine, but how do we connect it to the world we actually measure—a world of positions, momenta, and energies? The bridge between the abstract and the concrete is the concept of a ​​representation​​. And the most intuitive of all is the ​​position representation​​.

Imagine a single particle moving in one dimension. Its quantum state is an abstract vector, which we'll call $|\psi\rangle$. This vector lives in an infinite-dimensional space, containing all possible information about the particle. How can we get a handle on such a thing?

In ordinary three-dimensional space, we can describe any vector $\vec{v}$ by breaking it down into its components along three perpendicular axes, $\hat{i}$, $\hat{j}$, and $\hat{k}$. We write $\vec{v} = v_x \hat{i} + v_y \hat{j} + v_z \hat{k}$. The triplet of numbers $(v_x, v_y, v_z)$ is the representation of the vector $\vec{v}$ in the basis $\{\hat{i}, \hat{j}, \hat{k}\}$.

Let's try the same trick for our quantum state. What would be a natural "axis" to use? A fantastically useful choice is the set of all possible positions. Let's imagine a basis state for every single point $x_0$ on the line, a state where the particle is known to be exactly at position $x_0$. We'll denote this idealized state by the ket $|x_0\rangle$.

Just as we can write our 3D vector as a sum over basis vectors, we can express our general state $|\psi\rangle$ as a continuous "sum"—that is, an integral—over all these position basis states:

This equation says that any quantum state can be thought of as a superposition of states of definite position, with each position $x_0$ contributing a certain amount specified by the "coefficient" $c(x_0)$. But what are these coefficients? This is where the magic happens.

To find the $x$-component of a normal vector $\vec{v}$, we take its dot product with the basis vector $\hat{i}$: $v_x = \vec{v} \cdot \hat{i}$. We do the same in quantum mechanics, using the inner product. The coefficient for a particular position, say $x$, is found by taking the inner product of $|\psi\rangle$ with the basis state $|x\rangle$, which we write as $\langle x | \psi \rangle$.

Let's calculate it:

What is the inner product $\langle x | x_0 \rangle$? These are basis vectors for different positions. Just like $\hat{i} \cdot \hat{j} = 0$, they must be "orthogonal". But since the positions form a continuum, the inner product is not a simple zero. It's zero everywhere except when $x = x_0$, where it becomes infinitely sharp. This object is the famous ​​Dirac delta function​​, $\delta(x - x_0)$.

Plugging this in, we get:

The sifting property of the delta function tells us that this integral simply picks out the value of the function $c(x_0)$ at $x_0 = x$. So, we find:

This is a profound result. The coefficient function $c(x)$ is nothing other than the inner product $\langle x | \psi \rangle$. We give this function a special name: the ​​wavefunction​​, and denote it by $\psi(x)$.

So, the wavefunction $\psi(x)$ is not the state itself. It is the ​​representation​​ of the abstract state vector $|\psi\rangle$ in the basis of definite positions. It is the continuous list of components of the state vector along every "position axis" $|x\rangle$. This simple and beautiful idea is the gateway to all of wave mechanics. The completeness of this basis is expressed by the ​​closure relation​​, $\int |x\rangle\langle x| dx = \hat{I}$ (the identity operator), which is the formal statement that summing up the projections onto every possible position gives you back the whole space.

Now that we have components for our vectors, what happens to operators? An operator, like the position operator $\hat{x}$ or the Hamiltonian $\hat{H}$, is an abstract instruction that transforms one state vector into another. In a representation, this abstract instruction becomes a concrete mathematical action on the wavefunction.

Let's start with the easiest one: the position operator $\hat{x}$. By definition, it has the position states $|x_0\rangle$ as its eigenvectors: $\hat{x}|x_0\rangle = x_0|x_0\rangle$. What does this imply for the wavefunction of a state $|\phi\rangle = \hat{x}|\psi\rangle$? The new wavefunction is $\phi(x) = \langle x | \phi \rangle = \langle x | \hat{x} | \psi \rangle$.

Here we use a standard trick. Since $\hat{x}$ represents a physical observable, it must be a ​​self-adjoint​​ operator. This means we can let it act either to its right (on $|\psi\rangle$) or to its left (on $\langle x |$). When it acts to the left on its own eigenstate, it just pulls out the eigenvalue: $\langle x | \hat{x} = x \langle x |$. So,

Look at that! The abstract operation of "applying the position operator $\hat{x}$" has become the ridiculously simple action of "multiplying the wavefunction by the number $x$".

Other operators become more interesting. Consider the operator that projects any state onto our specific state $|\psi\rangle$, defined as $\hat{P}_{\psi} = |\psi\rangle\langle\psi|$. Its action on some other state $|f\rangle$ is represented by the function $\langle x | \hat{P}_{\psi} | f \rangle$. We can split this up:

Recognizing that $\langle \psi | x' \rangle = \langle x' | \psi \rangle^* = \psi^*(x')$ and $\langle x' | f \rangle = f(x')$, we get:

The abstract projection operator has become an ​​integral operator​​, whose action is determined by a ​​kernel​​ $P(x, x') = \psi(x)\psi^*(x')$. Every linear operator in quantum mechanics can be represented in this way as an integral operator in the position basis.

This all seems quite neat. But where is momentum? Finding the representation of the momentum operator, $\hat{p}$, reveals the strange and beautiful heart of quantum mechanics.

Unlike position, momentum doesn't have a simple multiplicative form in the position basis. Why? Because position and momentum are inextricably linked by the uncertainty principle. They don't share a common basis. We must find the representation of $\hat{p}$ from a deeper principle: ​​momentum is the generator of spatial translations​​.

What does this mean? It means that if you want to shift a particle's state by a tiny amount $\epsilon$, the operator that does this is related to momentum: $\hat{U}(\epsilon) \approx \hat{I} - \frac{i}{\hbar}\epsilon\hat{p}$. The new wavefunction is $\psi_{\text{new}}(x) = \langle x | \hat{U}(\epsilon) | \psi \rangle$. But we also know that translating the state should be the same as evaluating the original wavefunction at a shifted point: $\psi_{\text{new}}(x) = \psi(x-\epsilon)$.

Let's equate the two views:

Now, we use a Taylor expansion for the left side: $\psi(x-\epsilon) \approx \psi(x) - \epsilon \frac{d\psi}{dx}$. Comparing the two expressions, we are forced into an astonishing conclusion:

In the position representation, the abstract momentum operator $\hat{p}$ becomes the ​​differential operator​​ $-i\hbar \frac{d}{dx}$!

This is not just a convenient choice; it is essentially forced upon us. For the total energy of a system, given by the Hamiltonian $\hat{H} = \frac{\hat{p}^2}{2m} + V(\hat{x})$, to be a real, physical observable, $\hat{H}$ must be a self-adjoint operator. This requirement places strict constraints on the mathematical forms of $\hat{x}$ and $\hat{p}$, and the pair we have found—multiplication by $x$ and differentiation—is the unique choice (up to some simple transformations) that works. It is this differential form that makes the famous ​​canonical commutation relation​​, $[\hat{x}, \hat{p}] = \hat{x}\hat{p} - \hat{p}\hat{x} = i\hbar \hat{I}$, come to life. You can check it for yourself! This relation is the algebraic soul of quantum mechanics, and the position representation gives it concrete flesh and blood.

At this point, you might think that the position representation is the way quantum mechanics works. But it is just one "coordinate system" we can use to view the abstract reality of the state vector. We could have used the basis of momentum eigenstates, $|p\rangle$, to express our vector. The components in that basis, $\tilde{\psi}(p) = \langle p | \psi \rangle$, would be the "momentum-space wavefunction" (which happens to be the Fourier transform of $\psi(x)$).

A monumental result, the ​​Stone-von Neumann theorem​​, tells us something remarkable: for any quantum system with a finite number of parts (like an atom or a molecule), any representation that correctly implements the canonical commutation relation $[\hat{x}, \hat{p}]=i\hbar$ is fundamentally the same as the position representation we have been using. They are all just "unitarily equivalent," which is the mathematical way of saying they are the same picture, just rotated. This assures us that our choice is not arbitrary but universal.

This deep idea—of an invariant, abstract object versus its coordinate-dependent components—is not an oddity of quantum theory. It is one of the great unifying principles of modern physics and mathematics, found most clearly in geometry.

Imagine mapping the wind currents on the surface of the Earth. The wind field itself is a real, physical thing, existing independently of any map. This is our invariant object, like the state $|\psi\rangle$. To describe it, we might lay down a latitude-longitude grid. At each point, we can then write down the wind's components: "15 kph north, 5 kph east." These components are the representation of the wind vector, just like $\psi(x)$ is the representation of the state vector. If we were to use a different coordinate system (say, one tilted relative to the pole), the numerical components would change, but they would still be describing the very same wind. The rules for how the components change when you change coordinates are dictated by the chain rule—exactly the same mathematics that governs transformations between different bases in quantum mechanics.

Let's go one step further. In Einstein's theory of general relativity, the curvature of spacetime is described by a geometric object called the ​​metric tensor​​, $g$. The tensor itself is the "real thing"; it's what tells matter how to move. To do calculations, we must choose a coordinate system and write down the components of the tensor, a matrix of functions $g_{ij}$. If we change our coordinate system, the numbers in this matrix change according to a specific transformation law. However, physical quantities that we can actually measure, like the length of a path or the proper time elapsed on a clock, are scalar invariants. They are calculated by combining the tensor components with vector components (e.g., length-squared is $g_{ij}v^i v^j$), and the final number is the same no matter which coordinate system we used for the calculation.

This is a perfect analogy for quantum mechanics. The abstract state $|\psi\rangle$ and operator $\hat{H}$ are the invariant objects. The wavefunction $\psi(x)$ and the matrix of operator elements $H_{mn} = \langle m | \hat{H} | n \rangle$ are coordinate-dependent representations. But a physical prediction, like an energy expectation value $\langle \psi | \hat{H} | \psi \rangle = \int \psi^*(x) \hat{H} \psi(x) dx$, is a scalar invariant. Its value is absolute, a bedrock reality independent of our descriptive choices.

The position representation, then, is more than a calculational tool. It is a window into a profound unity of thought, connecting the quantum world of probability waves to the geometric world of invariant objects. It provides a reliable, powerful, and deeply beautiful language for describing physical reality, resting assured on a rigorous mathematical foundation.

In the previous chapter, we became acquainted with a cornerstone of the quantum worldview: the position representation. We learned that in the strange, probabilistic universe of quantum mechanics, a particle's state is not a point, but a landscape of possibility—a wavefunction, $\psi(x)$, spread across the stage of space. This might seem like a peculiar feature of the subatomic domain, a departure from the "common sense" of classical physics. But is it?

What if I told you that this way of thinking—describing a system's properties as fields or functions on a space of coordinates—is not an esoteric quantum trick, but one of the most powerful and unifying ideas in all of science? The concept of a "position representation" echoes far beyond its quantum birthplace. It is a master key that unlocks doors in chemistry, mathematics, cosmology, and even engineering. On our journey through this chapter, we will see how this single, elegant idea provides a common language to describe the dance of electrons in a molecule, the fundamental symmetries of motion, the very fabric of a dynamic, curved universe, and the stresses within a piece of solid steel.

Our first foray was with a single particle. But the world is not so lonely. How do we describe a helium atom, with its two electrons, or a water molecule, with its ten? We cannot simply assign each electron its own independent wavefunction, for their fates are deeply entangled by their mutual repulsion and the iron-clad rules of quantum statistics.

The answer is to expand our stage. Instead of a wavefunction on the three-dimensional space of a single position $\mathbf{r}$, we must imagine a wavefunction that lives on a higher-dimensional configuration space. For two particles, the "position" is a single point in a six-dimensional space $(\mathbf{r}_1, \mathbf{r}_2)$, and the wavefunction $\Psi(\mathbf{r}_1, \mathbf{r}_2)$ tells us the probability amplitude for finding particle 1 at $\mathbf{r}_1$ and particle 2 at $\mathbf{r}_2$ simultaneously.

This is the foundation of ​​quantum chemistry​​. While the full many-electron wavefunction is forbiddingly complex, chemists build it from more manageable pieces. They start with single-electron orbitals, which are just functions in the familiar position representation, like $\varphi(\mathbf{r})$. They then combine these building blocks into a sophisticated, antisymmetrized structure known as a Slater determinant, which beautifully enforces the Pauli exclusion principle—the rule stating that no two identical fermions can occupy the same quantum state. The entire machinery of computational chemistry, which allows us to predict molecular structures and reaction rates, is built upon calculating matrix elements of operators like the Hamiltonian within this position-based framework. The position representation is, quite literally, the canvas on which we draw our picture of the chemical world.

But a picture is static. What about the dynamics? How does a particle get from here to there? Richard Feynman's profound insight was that it takes every possible path at once. The evolution of a wavefunction is governed by the propagator, $\langle x | U(t) | x' \rangle$, which you can think of as the amplitude for a particle that starts at position $x'$ to arrive at position $x$ after a time $t$. Even for the simplest case of a free particle, the propagator is a complex number, its phase swirling with a rhythm dictated by the laws of quantum mechanics. This complex phase is not a mere mathematical accessory; it is the engine of all quantum interference, the very essence of quantum dynamics, all written in the language of a function of initial and final positions.

Let's step back from specific systems and ask a deeper question. Why does this position representation work so well? The answer lies in the fundamental symmetries of nature. The bedrock of quantum mechanics is the curious relationship between position $\hat{q}$ and momentum $\hat{p}$, codified in the Heisenberg commutation relation: $\hat{q}\hat{p} - \hat{p}\hat{q} = i\hbar$. This is an abstract algebraic statement. To make it do physics, we need to represent these abstract operators as concrete things acting on a space of states.

The Schrödinger position representation does exactly this: it turns $\hat{q}$ into the simple operator of "multiplication by $x$" and $\hat{p}$ into the differential operator $-i\hbar \frac{d}{dx}$. Suddenly, the abstract algebra becomes a set of instructions for manipulating functions of position, $f(x)$. This translation between the abstract algebra and the concrete functions of position is one of the most crucial links in the entire chain of physical reasoning.

This connection runs even deeper when we consider the full group of physical transformations. The operations of shifting a particle's position and boosting its momentum are governed by a mathematical structure called the ​​Heisenberg group​​. The Schrödinger representation shows how this entire group of symmetries acts on the space of wavefunctions. A famous result, the Stone–von Neumann theorem, assures us that, in a profound sense, this position representation is the only irreducible way to realize the quantum rules of position and momentum. The space of functions of position is not just a convenient choice; it is the natural stage for the drama of quantum mechanics.

Of course, the position representation isn't the only star of the show. There is a perfectly parallel world: the momentum representation, where states are described by wavefunctions of momentum, $\tilde{\psi}(p)$. The two worlds are connected by a beautiful mathematical dictionary known as the ​​Fourier transform​​. This operator translates wavefunctions from the position language into the momentum language and back again. This duality is a profound truth of nature: the more localized a particle is in position, the more spread out it is in momentum, and vice versa. Position and momentum are two sides of the same quantum coin, forever linked by this elegant transform.

Now, let's zoom out—way out. What happens when the stage itself, the very fabric of spacetime, is not a static, flat background? This is the realm of Albert Einstein's ​​General Relativity​​. Here, the idea of a "position representation" evolves into the more general concept of describing the world with local coordinates on a curved manifold. Just as we can label any point on the surface of the Earth with latitude and longitude, we can label any event in spacetime with a set of four coordinates, for instance, $(t, x, y, z)$.

In relativity, the geometry of spacetime is encoded in the metric tensor, $g_{\mu\nu}$, whose components tell us the distances between nearby points. The specific values of these components depend entirely on the coordinate system we choose. Consider, for example, a point on the surface of a cylinder. We can describe its position using an angle and a height, and these coordinates define a natural basis for vectors on that surface.

This idea is central to ​​cosmology​​. In the Friedmann-Robertson-Walker (FRW) model of our expanding universe, we use "comoving" coordinates $(t, x, y, z)$. In this system, a distant galaxy can have a fixed coordinate position $(x_0, y_0, z_0)$, yet the physical distance to it grows over time. This cosmic stretch is captured by the scale factor $a(t)$ in the metric: $ds^2 = -dt^2 + a(t)^2(dx^2+dy^2+dz^2)$. The coordinate representation allows us to disentangle the fixed "address" of an object from the dynamic expansion of the space itself. The mathematical tools we use, such as the relationship between a basis vector and its dual, are directly shaped by this dynamic geometry.

The same principle gives us our eyes on the cosmos. When a ​​gravitational wave​​—a ripple in spacetime itself—passes by, it causes the metric to oscillate. Consider a free-falling observer carrying a set of gyroscopes that define their personal, local sense of direction (an orthonormal basis). As the wave passes, their "x-direction" gyroscope will no longer point along the background coordinate $x$-axis. The relationship between the observer's physical frame and the background coordinate grid is distorted by the wave's polarization amplitudes, $h_+$ and $h_\times$. It is by measuring this minuscule, oscillating mismatch—this stretching and squeezing of space described perfectly in a coordinate representation—that detectors like LIGO and Virgo can "hear" the cataclysmic collisions of black holes billions of light-years away.

The power of coordinate representations is not confined to the cosmic scale. Let's bring our attention back down to Earth, to the tangible world of ​​continuum mechanics and engineering​​. When a steel beam bends under a load, how do we describe its state? We can think of the deformation as a mapping, $\varphi$, from an initial, undeformed "reference" configuration to a final, deformed "spatial" configuration. Every point $X$ in the original body is mapped to a new point $x = \varphi(X)$ in space.

The key object describing this local deformation is the deformation gradient, $F$. This mathematical object acts as a dictionary, translating vectors and tensors between the material's original frame and its new frame in space. Physical quantities like stress or the microscopic orientation of crystal grains, represented by a tensor $A$ in the material's reference frame, are "pushed forward" to a new tensor $a$ in the spatial frame using the formula $a = F A F^T$. This formalism, which allows engineers to track and calculate stress and strain in complex, deforming bodies, is mathematically analogous to the way relativists transform tensors between different coordinate systems. It is the bedrock of the Finite Element Method, the computational workhorse that enables the design of everything from bridges and buildings to car bodies and aircraft wings.

Our journey is complete. We began with the ghostly probability wave of a single electron, and we have ended with the tangible stress in a deforming solid. Along the way, we saw how the same fundamental idea—describing physical reality through functions and tensors defined on a coordinate space—provides the essential language for quantum chemistry, the symmetries of quantum theory, and the geometry of spacetime.

It serves as a remarkable testament to the unity of physics that what we call "position representation" is not one idea, but a golden thread running through the entire tapestry of science. Choosing "where" things are turns out to be the first and most crucial step toward understanding "what" they are and "how" they behave. The simple act of setting a stage and labeling its locations gives us the power to write down the laws of the universe, from the smallest scales to the largest.