Path Integral Formulation of Quantum Mechanics

The standard Schrödinger equation provides a powerful differential framework for quantum mechanics, but it obscures a profound physical picture: how does a particle actually get from point A to point B? Richard Feynman's path integral formulation answers this by revealing that a quantum particle does not take a single path, but rather samples every conceivable trajectory simultaneously. This sum-over-histories approach is not just a mathematical curiosity; it provides an intuitive bridge to classical physics, revolutionizes the treatment of field theories and quantum statistical mechanics, and offers a fundamentally different perspective where probability amplitudes, not particle positions, are the primary reality.

The Core Idea: Summing Over Histories

At the heart of the formulation is a radical postulate. The probability amplitude for a particle to travel from an initial point $(x_i,t_i)$ to a final point $(x_f,t_f)$ is found by summing contributions from all possible paths connecting these two events. Crucially, each path contributes a phase factor (a complex number) whose angle is proportional to the classical action $S$ for that path. The action is a functional of the path $x(t)$ and is defined as the time integral of the Lagrangian, $S[x(t)]={\int }_{t_i}^{t_f}L(x,\dot{x},t)dt$, where $L=T-V$ is kinetic energy minus potential energy.

Therefore, the total amplitude $K$—called the propagator or kernel—is formally written as: $$K(x_f,t_f;x_i,t_i)=\mathcal{N}\int \mathcal{D}x(t)e^{\frac{i}{\hslash }S[x(t)]}.$$ Here, the symbol $\int \mathcal{D}x(t)$ represents the daunting, yet conceptually simple, instruction to integrate (sum) over all continuous functions $x(t)$ that satisfy the boundary conditions $x(t_i)=x_i$ and $x(t_f)=x_f$. The factor $\mathcal{N}$ is a normalization constant. The quantity $e^{\frac{i}{\hslash }S}$ is the complex phase weight for each path. This elegant statement encodes the entire quantum dynamics of a single particle.

Defining the Quantum Propagator

The propagator $K(x_f,t_f;x_i,t_i)$ is the central object in the path integral formalism. Its physical meaning is direct: it is the probability amplitude to find the particle at position $x_f$ at time $t_f$, given that it was definitely at $x_i$ at an earlier time $t_i$. If you know the initial wavefunction $\psi (x_i,t_i)$, the path integral tells you how to evolve it: $$\psi (x_f,t_f)={\int }_{-\infty }^{\infty }K(x_f,t_f;x_i,t_i)\psi (x_i,t_i)d{x}_i.$$ This equation shows that $K$ acts as the Green's function or the "influence function" for the wavefunction, propagating the amplitude forward in time. It satisfies the same Schrödinger equation as the wavefunction and contains the complete dynamical information of the quantum system.

Derivation for a Free Particle

To make sense of the formal integral, we must define it precisely. The standard technique is to discretize time into $N$ small slices of duration $ϵ=(t_f-t_i)/N$. The continuous path $x(t)$ is replaced by a series of positions $x_0,x_1,...,x_N$ at successive times, with $x_0=x_i$ and $x_N=x_f$. The action becomes a sum, and the integral over all paths becomes an $N-1$ fold integral over the intermediate $x_k$ positions.

For a free particle ($V=0$), the Lagrangian is $L=\frac{1}{2}m{\dot{x}}^2$. The discretized action is: $$S_N=\sum _{k=0}^{N-1}\frac{m}{2}\frac{(x_{k+1}-x_k{)}^2}{ϵ}.$$ The propagator becomes a product of Gaussian integrals: $$K=\underset{N\to \infty }{\lim }{\left(\frac{m}{2\pi i\hslash ϵ}\right)}^{N/2}{\int }_{-\infty }^{\infty }d{x}_1...\int d{x}_{N-1}\exp \left(\frac{im}{2\hslash ϵ}\sum _{k=0}^{N-1}(x_{k+1}-x_k{)}^2\right).$$ Each Gaussian integral can be performed sequentially. The final, celebrated result for the free particle propagator is: $$K_{\text{free}}(x_f,t_f;x_i,t_i)=\sqrt{\frac{m}{2\pi i\hslash (t_f-t_i)}}\exp \left(\frac{im(x_f-x_i{)}^2}{2\hslash (t_f-t_i)}\right).$$ This function shows how amplitude from a point source spreads diffusively like a wave, with a phase that depends on the classical action for the straight-line path.

The Harmonic Oscillator and Euclidean Methods

For the quintessential harmonic oscillator, with $V(x)=\frac{1}{2}m{\omega }^2{x}^2$, the path integral can also be computed exactly. The direct evaluation is more complex due to the presence of the potential, but it remains a Gaussian-type integral because the action is quadratic in $x(t)$. The result is: $$K_{\text{ho}}(x_f,t_f;x_i,t_i)=\sqrt{\frac{m\omega }{2\pi i\hslash \sin \omega T}}\exp \left\{\frac{im\omega }{2\hslash \sin \omega T}\left[(x_f^2+x_i^2)\cos \omega T-2x_f{x}_i\right]\right\},$$ where $T=t_f-t_i$.

A powerful technique for handling oscillatory integrals like $e^{iS/\hslash }$ is the Wick rotation to imaginary time. By substituting $t\to -i\tau$ (or $\hslash \beta$ for statistical mechanics), the phase factor becomes a real decaying exponential: $e^{iS/\hslash }\to e^{-S_E/\hslash }$, where $S_E$ is the Euclidean action. This transforms the wildly oscillating path integral into a well-behaved statistical sum, analogous to a partition function in classical statistical mechanics. This is indispensable for practical calculations in quantum field theory and finite-temperature physics.

The Classical Limit and Stationary Phase

The most profound conceptual link lies in recovering classical mechanics. In the formal path integral, paths contribute with a wildly oscillating phase $e^{iS/\hslash }$. In the limit where the action $S$ is much larger than $\hslash$ (the classical limit $\hslash \to 0$), these oscillations become infinitely rapid. Paths that are not close to each other will have uncorrelated phases and their contributions will cancel out upon summation through destructive interference.

The only paths that survive are those for which the action is stationary—that is, where a small variation in the path $\delta x(t)$ produces no first-order change in the action: $\delta S=0$. This is precisely Hamilton's principle of stationary action, which yields the Euler-Lagrange equations of motion. Thus, the classical trajectory $x_{\text{cl}}(t)$ is the path of constructive interference in the quantum amplitude. This result, derived from the method of stationary phase applied to the functional integral, elegantly shows how classical determinism emerges from quantum probability as a dominant pathway.

Common Pitfalls

1. Interpreting Paths as Real Trajectories: A common misconception is to think the particle physically travels along each infinite set of paths. The paths are not real trajectories but represent contributions to a probability amplitude. The formulation calculates "how much" each hypothetical path contributes to the total amplitude for the transition.
2. Misunderstanding the Stationary Phase Approximation: The classical path is not the only contributor, even in the classical limit. It is the only path whose contribution coherently adds with its immediate neighbors. The approximation involves integrating over small fluctuations around the classical path, which gives quantum corrections (like zero-point energy).
3. Forgetting the Measure $\mathcal{D}x(t)$: The notation $\int \mathcal{D}x(t)$ hides a careful definition involving discretization and normalization. Ignoring the subtleties of the functional measure, especially the $\mathcal{N}$ factor, can lead to missing crucial terms (like the determinant prefactor in the harmonic oscillator result).
4. Applying the Euclidean Formulation Incorrectly: Wick rotation ($t\to -i\tau$) is a formal analytic continuation. It is a computational trick to handle convergence, not a physical transformation of time for all purposes. Results in Euclidean time must often be rotated back to real time to extract physical observables like scattering amplitudes.

Summary

• The path integral formulation posits that the quantum amplitude for a particle to go from one point to another is computed by summing (integrating) over all possible spacetime paths connecting the points, with each path weighted by the complex phase $e^{iS/\hslash }$, where $S$ is the classical action for that path.
• The central object is the propagator $K$, which evolves the wavefunction in time. It can be computed exactly for systems with quadratic actions, leading to explicit results for the free particle and the harmonic oscillator.
• A key technical tool is the Wick rotation to imaginary time, which converts the oscillatory path integral into a statistically convergent Euclidean integral, crucial for advanced applications.
• The classical limit emerges via the stationary phase approximation: in the limit $\hslash \to 0$, only paths where the action is stationary ($\delta S=0$) contribute coherently, recovering the principle of least action and classical trajectories.
• This formulation provides a powerful and intuitive framework that generalizes seamlessly to quantum field theory, where the paths become field configurations over all of spacetime.