Perturbation Methods

Systematic approximations for nearly solvable quantum systems. Theory, formulas, worked examples and interactive graphs.

Conceptual context

Perturbation methods are used when a quantum problem cannot be solved exactly but lies close to one that can. The method does not guess the answer; it arranges the unknown effect into controlled layers. One starts from the $H0$, adds the $V$ scaled by $\lambda$, and studies how the energy and state change. The central reading is how $\delta_{E1}$ and $\delta_{E2}$ vary as $\lambda$ changes, while checking that the effective coupling stays small compared with the Delta_E_mn.

🟢 Essential level

Suppose you know a simple quantum system exactly through its $H0$. A small extra influence, the $V$, is then switched on with strength $\lambda$.

Perturbation methods ask how much the energy levels move without solving the whole problem again. The first response is usually $\delta_{E1}$, which measures how strongly the original state feels the perturbation inside itself through $V_nn$. A finer response, $\delta_{E2}$, comes from indirect mixing with other states through $V_mn$. The Delta_E_mn decides whether that mixing stays controlled or becomes dangerous for the approximation and for the graph reading. If $\lambda$ becomes too large, the correction series no longer behaves like ordered physical response layers; it warns that the original solvable system is no longer the right anchor.

The Stark effect, the shift of atomic levels under an external electric field discovered by Johannes Stark in 1913, and the Zeeman effect, the splitting of levels under a magnetic field discovered by Pieter Zeeman in 1896, are two of the most important historical applications of perturbation theory: both treat the external field as a small modification of the isolated atom's $H0$.

🔵 Formal level

Basic formulas

The formal starting point splits the complete problem into a solvable operator and a controlled perturbation.

Perturbed Hamiltonian

The $H$ is built from $H0$ and $V$. For an unperturbed state n, the first energy response comes from the expectation value of the perturbation in that same state.

First-order energy correction

Thus $\delta_{E1}$ is proportional to $\lambda$ and $V_nn$. Its graph against $\lambda$ is linear in the valid regime, and the physical slope is the diagonal perturbation element.

Real student questions

Why not solve $H$ directly? In many real systems it is too complex; perturbation theory uses the already solved $H0$.

Is a negative $\delta_{E1}$ wrong? No. It indicates that $V$ lowers the level according to the sign of $V_nn$.

What if Delta_E_mn is very small? Mixing grows, and the non-degenerate method loses validity; nearby states must be reorganized.

Why keep $\lambda$ if it is later set to one? It lets us study continuity before evaluating the physical case.

Final synthesis

Perturbation methods turn a difficult quantum problem into a controlled expansion. The key is hierarchy: coupling, gap, sign, mixing, and validity. When $\eta$ is small, the physics speaks in ordered layers.