Derive the path integral propagator. Start from the amplitude, insert complete position…
- User
Derive the path integral propagator. Start from the amplitude, insert complete position states, and show the exponent becomes the classical action as time steps approach zero.
- Assistant
Deriving path integral propagator
The derivation starts from the quantum amplitude and inserts complete position states to split the time interval into small steps. Evaluating each short-time kernel via momentum integration yields a Gaussian integral producing a factor with an exponent that discretizes the classical action. As the time step goes to zero, the sum converges to the time integral of the classical Lagrangian, so the exponent becomes the classical action, giving the path integral representation with measure and classical action exponent as desired. A convenient starting point is the transition amplitude written as a matrix element of the time–evolution operator, [ K(x_{\rm f},t_{\rm f};x_{\rm i},t_{\rm i})=\langle x_{\rm f},|,{\rm e}^{-\frac{i}{\hbar}H(t_{\rm f}-t_{\rm i})},|,x_{\rm i}\rangle,, ] with (H=p^2/2m+V(x)) for a single particle. Split the time interval into (N) pieces of length (\varepsilon=(t_{\rm f}-t_{\rm i})/N), so that [ {\rm e}^{-\frac{i}{\hbar}H(t_{\rm f}-t_{\rm i})}=\left({\rm e}^{-\frac{i}{\hbar}H\varepsilon}\right)^N,, ] and insert resolutions of unity (\int!{\rm d}x_n,|x_n\rangle\langle x_n|=1) between the factors. This gives [ K=\int\prod_{n=1}^{N-1}{\rm d}x_n;\prod_{n=0}^{N-1}\langle x_{n+1}|,{\rm e}^{-\frac{i}{\hbar}H\varepsilon},|x_n\rangle, ] with (x_0=x_{\rm i},,x_N=x_{\rm f}). For small (\varepsilon) one can separate the kinetic and potential parts (Trotter formula) and evaluate the short–time kernel by inserting a momentum resolution (\int!{\rm d}p,(2\pi\hbar)^{-1}|p\rangle\langle p|): [ \langle x'|{\rm e}^{-\frac{i}{\hbar}H\varepsilon}|x\rangle\simeq{\rm e}^{-\frac{i}{\hbar}V(x)\varepsilon}\int\frac{{\rm d}p}{2\pi\hbar},{\rm e}^{\frac{i}{\hbar}p(x'-x)-\frac{i}{\hbar}\frac{p^2}{2m}\varepsilon}=\sqrt{\frac{m}{2\pi i\hbar\varepsilon}}, {\rm e}^{\frac{i}{\hbar}\left[\frac{m}{2}\left(\frac{x'-x}{\varepsilon}\right)^2 -V(x)\right]\varepsilon}. ]
Inserting this into the product gives [ K=\left(\frac{m}{2\pi i\hbar\varepsilon}\right)^{!N/2}\int\prod_{n=1}^{N-1}{\rm d}x_n;\exp\left{\frac{i}{\hbar}\sum_{n=0}^{N-1}\left[\tfrac{1}{2}m\left(\frac{x_{n+1}-x_n}{\varepsilon}\right)^2-V(x_n)\right]\varepsilon\right}. ]
The sum in the exponent is a Riemann sum for the action, [ S[x]=\int_{t_{\rm i}}^{t_{\rm f}}!{\rm d}t;\left[\tfrac{1}{2}m\dot x^2(t)-V\bigl(x(t)\bigr)\right], ] with (\dot x\approx (x_{n+1}-x_n)/\varepsilon). In the limit (N\to\infty), (\varepsilon\to 0) the prefactor can be absorbed into a formal measure ({\cal D}x(t)), and one arrives at the path–integral representation [ K(x_{\rm f},t_{\rm f};x_{\rm i},t_{\rm i})=\int_{x(t_{\rm i})=x_{\rm i}}^{x(t_{\rm f})=x_{\rm f}}{\cal D}x(t);{\rm e}^{\frac{i}{\hbar}S[x]},, ] i.e. the exponent becomes the classical action functional in the continuum limit.
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.