Born approximation and the Lippmann-Schwinger equation

Intuition Beginner

Picture standing in a calm lake and dropping a small stone. Ripples spread outward in clean circles. Now slip a rock just under the surface off to one side. The incoming ripples still pass through, but a new set of ripples appears to radiate from the rock. By watching how strong these new ripples are in each direction, you learn the rock's size and shape without touching it. Quantum scattering works the same way: a steady beam of particles is the incoming ripple, the target is the rock, and the outgoing pattern carries the information.

Far from the target the particle's wave splits into two pieces. One piece keeps marching forward as if nothing happened. The other piece is a spherical wave spreading out from the target, but with different strength in different directions. That direction-dependent strength is the scattering amplitude $f$. It is the single most important quantity in this story.

How loud the scattered wave is at angle $\theta$ tells you how likely a particle is to be deflected that way. Squaring the amplitude gives the differential cross-section, the chance per unit solid angle of finding a deflected particle in a detector placed at that angle. Strong scattering means a large amplitude; no scattering means the amplitude vanishes and the wave passes straight through.

The Born approximation is the shortcut that makes this calculable. When the target is a weak bump rather than a brick wall, the scattered wave is faint, and the amplitude turns out to be the Fourier transform of the potential. A wide, gentle bump scatters into small angles; a sharp, narrow bump scatters into wide angles. The angular pattern is a direct readout of the target's shape.

Visual Beginner

ASYMPTOTIC FORM OF THE SCATTERING WAVE
======================================

    incoming plane wave            outgoing spherical wave
    e^{ikz}                        f(theta) e^{ikr} / r
                                          \
    --->  --->  --->                       \   theta
    --->  --->  --->  ----[ V(r) ]---->------O detector
    --->  --->  --->                       /
                                          /
    (marches straight on)          (faint, spreads from target,
                                    strength = f(theta))

    far away:   psi  ~  e^{ikz}  +  f(theta) e^{ikr} / r
    measured:   d(sigma)/d(Omega)  =  | f(theta) |^2

BORN APPROXIMATION = FOURIER TRANSFORM OF THE POTENTIAL
=======================================================

    f(q)  proportional to  integral of  V(r) * (phase factor)

    wide gentle bump  ->  narrow forward beam of scattered particles
    sharp narrow bump ->  wide spray of scattered particles

    momentum transfer  q = 2 k sin(theta/2)

Quantity	Symbol	Meaning
Scattering amplitude	$f(\theta )$	Direction-dependent strength of the outgoing wave
Differential cross-section	$d\sigma /d\Omega$	Probability per solid angle; equals $
Momentum transfer	$q$	$2k\sin (\theta /2)$; how much momentum the kick changes
Potential	$V(r)$	The target, the bump the particle scatters off

Worked example Beginner

Problem. A particle of mass $m$ scatters off a small uniform ball of potential: $V(r)=V_0$ for $r<a$ and $V(r)=0$ outside, with $V_0$ weak. Using the Born shortcut, find how the scattered intensity depends on angle at very low energy, where the momentum transfer $q$ is small enough that $qa$ stays well below 1.

Solution.

Step 1. The Born shortcut says the amplitude is a fixed constant, $-m/(2\pi {\hslash }^2)$, multiplied by the total of the potential added up over all of space, where each piece of the potential is weighted by a phase factor $e^{i\mathbf{q}\cdot \mathbf{r}}$. This weighted total is the Fourier transform of the potential.

Step 2. Because the potential equals the single value $V_0$ everywhere inside the ball and zero outside, only the ball contributes. The weighted total is $V_0$ multiplied by the phase-weighted volume of the ball.

Step 3. At low energy the phase factor $e^{i\mathbf{q}\cdot \mathbf{r}}$ stays close to 1 everywhere inside the ball, because $qa$ is small. So the weighted total is approximately $V_0$ times the plain volume of the ball, $\frac{4}{3}\pi a^3$:

$$f\approx -\frac{m}{2\pi {\hslash }^2}{V}_0\frac{4}{3}\pi a^3=-\frac{2m{V}_0{a}^3}{3{\hslash }^2}.$$

Step 4. This number does not depend on the angle $\theta$. So the intensity $|f{|}^2$ is the same in every direction:

$$\frac{d\sigma }{d\Omega }\approx {\left(\frac{2m{V}_0{a}^3}{3{\hslash }^2}\right)}^2.$$

What this tells us. At low energy a small target scatters equally in all directions: the pattern is a uniform sphere of outgoing particles. The particle's wavelength is far longer than the target, so the wave cannot resolve any internal structure and sees only a featureless point. Fine angular detail in a scattering pattern appears only when the wavelength shrinks to the size of the target.

Check your understanding Beginner

Formal definition Intermediate+

Consider a particle of mass $m$ and energy $E={\hslash }^2{k}^2/2m$ scattering off a localized potential $V(\mathbf{r})$ that decays faster than $1/r$ at large distance. The stationary Schrödinger equation

$$\left(-\frac{{\hslash }^2}{2m}{\nabla }^2+V(\mathbf{r})\right)\psi =E\psi$$

is reorganized into $({\nabla }^2+k^2)\psi =U\psi$, where $U(\mathbf{r})=(2m/{\hslash }^2)V(\mathbf{r})$ is the reduced potential. The physically relevant solution carries an incident plane wave and a purely outgoing scattered wave. Its asymptotic form is

$$\psi (\mathbf{r})\stackrel{r\to \infty }{\to }{e}^{ikz}+f(\theta ,\varphi )\frac{e^{ikr}}{r}.$$

The function $f(\theta ,\varphi )$ is the scattering amplitude; it has dimensions of length. Matching incident and scattered probability fluxes gives the differential cross-section $d\sigma /d\Omega =|f(\theta ,\varphi ){|}^2$ and the total cross-section ${\sigma }_{\text{tot}}=\int |f{|}^{2}d\Omega$.

The outgoing free Green's function. The free Helmholtz operator ${\nabla }^2+k^2$ has a Green's function $G_0$ obeying $({\nabla }^2+k^2)G_0(\mathbf{r},{\mathbf{r}}^{\prime })={\delta }^3(\mathbf{r}-{\mathbf{r}}^{\prime })$. Two solutions exist, distinguished by boundary condition. The one carrying outgoing waves (the $+iϵ$ prescription, equivalently the Sommerfeld radiation condition) is

$$G_0^{+}(\mathbf{r},{\mathbf{r}}^{\prime })=-\frac{1}{4\pi }\frac{e^{ik|\mathbf{r}-{\mathbf{r}}^{\prime }|}}{|\mathbf{r}-{\mathbf{r}}^{\prime }|}.$$

In operator form this is the resolvent $G_0^{+}=(E-H_0+iϵ{)}^{-1}$ of the free Hamiltonian $H_0=-{\hslash }^2{\nabla }^2/2m$, with the $iϵ$ pushing the pole off the real axis to select retarded (outgoing) propagation.

The Lippmann-Schwinger equation. Rewriting $({\nabla }^2+k^2)\psi =U\psi$ as an integral equation using $G_0^{+}$ gives

$$\psi (\mathbf{r})=\varphi (\mathbf{r})+\int G_0^{+}(\mathbf{r},{\mathbf{r}}^{\prime })U({\mathbf{r}}^{\prime })\psi ({\mathbf{r}}^{\prime })d^3{r}^{\prime },$$

with $\varphi =e^{ikz}$ the incident free wave annihilated by $({\nabla }^2+k^2)$. In Dirac notation, $|{\psi }^{(+)}⟩=|\varphi ⟩+(E-H_0+iϵ{)}^{-1}V|{\psi }^{(+)}⟩$. The boundary condition is built into the Green's function, not imposed afterward.

First Born approximation. Iterating once — replacing $\psi$ under the integral by the incident wave $\varphi$ — and reading off the coefficient of $e^{ikr}/r$ at large $r$ yields

$$f^{(1)}(\mathbf{q})=-\frac{m}{2\pi {\hslash }^2}\int V(\mathbf{r})e^{i\mathbf{q}\cdot \mathbf{r}}{d}^{3}r=-\frac{m}{2\pi {\hslash }^2}\tilde{V}(\mathbf{q}),$$

the Fourier transform of the potential at momentum transfer $\mathbf{q}=\mathbf{k}-{\mathbf{k}}^{\prime }$, with magnitude $q=2k\sin (\theta /2)$ for elastic scattering ($|\mathbf{k}|=|{\mathbf{k}}^{\prime }|=k$). The treatment follows the integral-equation development in ^{[jimmyqin Lippmann-Schwinger equation, optical theorem]}.

Counterexamples to common slips

• The Coulomb potential $1/r$ does not decay faster than $1/r$, so the asymptotic form above is modified by logarithmic phase distortion; the bare $f^{(1)}$ Fourier transform still gives the Rutherford magnitude, but the exact phase is altered.
• The $iϵ$ sign is not cosmetic: choosing $-iϵ$ produces $G_0^{-}$ with an incoming spherical wave $e^{-ikr}/r$, the time-reversed boundary condition, which describes a different physical state.
• The first Born amplitude is real for a real potential, so it satisfies the optical theorem only at the next order; a real $f^{(1)}$ has zero imaginary part at $\theta =0$.

Key derivation with proof Intermediate+

Theorem (outgoing free Green's function). The kernel $$G_0^{+}(\mathbf{r},{\mathbf{r}}^{\prime })=-\frac{1}{4\pi }\frac{e^{ik|\mathbf{r}-{\mathbf{r}}^{\prime }|}}{|\mathbf{r}-{\mathbf{r}}^{\prime }|}$$ solves $({\nabla }^2+k^2)G_0^{+}={\delta }^3(\mathbf{r}-{\mathbf{r}}^{\prime })$ and carries only outgoing waves at infinity. Iterating the Lippmann-Schwinger equation to first order then gives the Born amplitude $f^{(1)}(\mathbf{q})=-(m/2\pi {\hslash }^2)\tilde{V}(\mathbf{q})$.

Proof. Set $\mathbf{s}=\mathbf{r}-{\mathbf{r}}^{\prime }$ and seek $G_0(s)$ depending only on $s=|\mathbf{s}|$. For $s\ne 0$ the radial Helmholtz equation $\frac{1}{s}\frac{d^2}{d{s}^2}(s{G}_0)+k^2{G}_0=0$ has general solution $G_0=(A{e}^{iks}+B{e}^{-iks})/s$. The outgoing boundary condition discards the incoming branch, so $B=0$. To fix $A$, integrate $({\nabla }^2+k^2)G_0={\delta }^3(\mathbf{s})$ over a small ball of radius $\eta$ around the origin. The $k^2{G}_0$ term contributes $O({\eta }^2)\to 0$. By the divergence theorem the Laplacian term gives the flux $\oint \nabla G_0\cdot d\mathbf{A}$ through the sphere $s=\eta$. With $G_0=A{e}^{iks}/s$, the radial derivative near the origin behaves like $-A/s^2$, and the surface integral over $4\pi {\eta }^2$ tends to $-4\pi A$. Matching to the unit weight of the delta function, $-4\pi A=1$, hence $A=-1/4\pi$. This establishes the kernel.

For the amplitude, substitute $\psi \to \varphi =e^{i\mathbf{k}\cdot {\mathbf{r}}^{\prime }}$ on the right of the Lippmann-Schwinger equation:

$$\psi (\mathbf{r})\approx e^{i\mathbf{k}\cdot \mathbf{r}}-\frac{m}{2\pi {\hslash }^2}\int \frac{e^{ik|\mathbf{r}-{\mathbf{r}}^{\prime }|}}{|\mathbf{r}-{\mathbf{r}}^{\prime }|}V({\mathbf{r}}^{\prime })e^{i\mathbf{k}\cdot {\mathbf{r}}^{\prime }}{d}^3{r}^{\prime },$$

where the prefactor combines $-1/4\pi$ from $G_0^{+}$ with $U=2mV/{\hslash }^2$. For large $r$ with ${\mathbf{r}}^{\prime }$ confined to the support of $V$, expand $|\mathbf{r}-{\mathbf{r}}^{\prime }|\approx r-\hat{\mathbf{r}}\cdot {\mathbf{r}}^{\prime }$, so $e^{ik|\mathbf{r}-{\mathbf{r}}^{\prime }|}/|\mathbf{r}-{\mathbf{r}}^{\prime }|\approx (e^{ikr}/r)e^{-i{\mathbf{k}}^{\prime }\cdot {\mathbf{r}}^{\prime }}$ with ${\mathbf{k}}^{\prime }=k\hat{\mathbf{r}}$ the outgoing wavevector. Reading off the coefficient of $e^{ikr}/r$:

$$f^{(1)}=-\frac{m}{2\pi {\hslash }^2}\int e^{-i{\mathbf{k}}^{\prime }\cdot {\mathbf{r}}^{\prime }}V({\mathbf{r}}^{\prime })e^{i\mathbf{k}\cdot {\mathbf{r}}^{\prime }}{d}^3{r}^{\prime }=-\frac{m}{2\pi {\hslash }^2}\int V({\mathbf{r}}^{\prime })e^{i\mathbf{q}\cdot {\mathbf{r}}^{\prime }}{d}^3{r}^{\prime },$$

with $\mathbf{q}=\mathbf{k}-{\mathbf{k}}^{\prime }$. This is the Fourier transform of the potential. $■$

Bridge. The first Born amplitude is the leading iterate of the Lippmann-Schwinger integral equation, and that integral equation builds toward the entire formal apparatus of scattering: each further substitution of $\psi$ into its own right-hand side adds one more factor of $G_0^{+}V$, generating the Born series whose convergence and resummation appears again in the Master treatment. The foundational reason the amplitude is a Fourier transform is that the free Green's function is a convolution kernel, so first-order scattering reads the potential off in momentum space; this is exactly the same Fourier structure that, putting these together with the partial-wave expansion of $\mathrm{12.08.01}$, links the momentum-transfer picture to the angular-momentum picture. The bridge is the identity between the resolvent $(E-H_0+iϵ{)}^{-1}$ as an operator and its position-space kernel $G_0^{+}$ as a wave: one statement generalises perturbation theory from bound states to the continuum.

Exercises Intermediate+

Advanced results Master

The first Born amplitude is the leading term of a systematic expansion. Iterating the Lippmann-Schwinger equation $|{\psi }^{(+)}⟩=|\varphi ⟩+G_0^{+}V|{\psi }^{(+)}⟩$ generates the Born series

$$f=f^{(1)}+f^{(2)}+f^{(3)}+\cdots ,f^{(n)}\propto ⟨{\mathbf{k}}^{\prime }|V(G_0^{+}V{)}^{n-1}|\mathbf{k}⟩,$$

the Neumann series of the integral operator. Each term has a transparent physical reading: $f^{(2)}$ is scattering off the potential twice with free propagation by $G_0^{+}$ in between, $f^{(3)}$ thrice, and so on. The series is the perturbative solution of the operator equation $T=V+V{G}_0^{+}T$ for the transition operator $T$, whose on-shell matrix elements are the exact amplitude $f=-(m/2\pi {\hslash }^2)⟨{\mathbf{k}}^{\prime }|T|\mathbf{k}⟩$.

Convergence. The Neumann series converges in operator norm when the spectral radius of $G_0^{+}V$ is below one. For a potential of depth $V_0$ and range $a$, the natural dimensionless coupling is $g=m|V_0|a^2/{\hslash }^2$ at low energy; the series converges for $g$ sufficiently small and diverges once $V$ supports a bound state, at which point $T$ develops a pole at the bound-state energy and no finite truncation captures it. At high energy the convergence improves: the effective coupling is reduced by a factor $\sim 1/(ka)$, so the high-energy validity condition relaxes to $(m|V_0|a/{\hslash }^{2}k)\ln (ka)\ll 1$. The logarithm reflects the slow forward-direction accumulation of phase along the straight-line trajectory, the seed of the eikonal approximation.

Born validity criterion. The single-iterate approximation is reliable when the correction to the incident wave inside the scattering region is small. At low energy this gives the criterion $|V_0|\ll {\hslash }^2/m{a}^2$; at high energy it weakens as above. The criterion is a statement about the smallness of $G_0^{+}V$, the same object whose spectral radius governs the full series.

Worked example: Yukawa and the Coulomb limit. For the screened-Coulomb (Yukawa) potential $V(r)=V_0{e}^{-\mu r}/r$ the first Born amplitude is the Lorentzian

$$f(q)=-\frac{2m{V}_0}{{\hslash }^2}\frac{1}{q^2+{\mu }^2},\frac{d\sigma }{d\Omega }=\frac{4m^2{V}_0^2}{{\hslash }^4}\frac{1}{(q^2+{\mu }^2{)}^2}.$$

Setting $V_0=Z_1{Z}_2{e}^2/4\pi {ϵ}_0$ and removing the screening, $\mu \to 0$, gives

$$\frac{d\sigma }{d\Omega }={\left(\frac{2m{Z}_1{Z}_2{e}^2}{4\pi {ϵ}_0{\hslash }^2{q}^2}\right)}^2={\left(\frac{Z_1{Z}_2{e}^2}{16\pi {ϵ}_{0}E}\right)}^2\frac{1}{{\sin }^4(\theta /2)},$$

the Rutherford cross-section, where $E={\hslash }^2{k}^2/2m$ and $q=2k\sin (\theta /2)$. That the first Born approximation reproduces the exact classical and quantum Rutherford result is a coincidence of the $1/r$ potential: the screening length $1/\mu$ acts as a physical regulator of the otherwise-divergent total cross-section, restored by atomic-electron screening in real targets.

Optical theorem. Probability conservation forces the exact amplitude to satisfy

$${\sigma }_{\text{tot}}=\frac{4\pi }{k}\mathrm{Im}f(0).$$

The flux removed from the forward beam — measured by $\mathrm{Im}f(0)$ — must reappear as total scattered flux. This is an exact unitarity identity, and it constrains the exact amplitude, not any truncation. A real first Born amplitude has $\mathrm{Im}{f}^{(1)}(0)=0$, so the optical theorem is satisfied only once $f^{(2)}$ supplies the forward imaginary part. The treatment of the unitarity identity follows ^{[jimmyqin Lippmann-Schwinger equation, optical theorem]} and the Born-series structure follows ^{[jimmyqin Born approximation, Fourier transform of the potential]}.

Synthesis. The transition operator $T=V+V{G}_0^{+}T$ is the foundational reason the Born series exists: its Neumann expansion is exactly the multiple-scattering picture, and its on-shell matrix element is the exact amplitude, so the first Born approximation generalises naturally to all orders by reinserting $T$ for $V$. Putting these together, the optical theorem identifies the forward imaginary part of $f$ with the integrated cross-section, which is dual to the spatial Fourier transform of the potential read off at momentum transfer $q$; the central insight is that one nonperturbative constraint (unitarity) ties together every order of a perturbative series that, taken term by term, has no business respecting it. This is exactly the structure that appears again in $\mathrm{12.07.02}$, where Fermi's golden rule expresses the same on-shell delta function $\delta (E-H_0)$ that supplies $\mathrm{Im}{f}^{(2)}$, and it builds toward the relativistic $S$-matrix of $\mathrm{12.12.01}$, where the same resolvent-and-unitarity machinery becomes the optical theorem of quantum field theory.

Full proof set Master

Proposition (the Born series as a Neumann series for the transition operator). Define the transition operator $T$ by $T|\varphi ⟩=V|{\psi }^{(+)}⟩$, where $|{\psi }^{(+)}⟩$ solves the Lippmann-Schwinger equation. Then $T$ satisfies $T=V+V{G}_0^{+}T$, and whenever $\Vert G_0^{+}V\Vert <1$ in operator norm the unique solution is the convergent Neumann series $T={\sum }_{n=0}^{\infty }V(G_0^{+}V{)}^n$.

Proof. From $|{\psi }^{(+)}⟩=|\varphi ⟩+G_0^{+}V|{\psi }^{(+)}⟩$, apply $V$ on the left: $V|{\psi }^{(+)}⟩=V|\varphi ⟩+V{G}_0^{+}V|{\psi }^{(+)}⟩$. Substituting the definition $T|\varphi ⟩=V|{\psi }^{(+)}⟩$ on both sides gives $T|\varphi ⟩=V|\varphi ⟩+V{G}_0^{+}T|\varphi ⟩$. Since $|\varphi ⟩$ ranges over a dense set of free states (plane waves of every momentum direction), the operator identity $T=V+V{G}_0^{+}T$ holds. Rearranging, $(1-V{G}_0^{+})T=V$, so formally $T=(1-V{G}_0^{+}{)}^{-1}V$. The inverse exists as a norm-convergent geometric series exactly when $\Vert V{G}_0^{+}\Vert <1$ (equivalently $\Vert G_0^{+}V\Vert <1$, since the two operators have the same spectral radius): then $(1-V{G}_0^{+}{)}^{-1}={\sum }_{n\ge 0}(V{G}_0^{+}{)}^n$, and

$$T=\sum _{n=0}^{\infty }(V{G}_0^{+}{)}^{n}V=\sum _{n=0}^{\infty }V(G_0^{+}V{)}^n,$$

where the last equality regroups the alternating product. Each partial sum is bounded by the geometric majorant $\Vert V\Vert {\sum }_n\Vert G_0^{+}V{\Vert }^n=\Vert V\Vert /(1-\Vert G_0^{+}V\Vert )$, so the series converges absolutely in operator norm, and its limit solves $T=V+V{G}_0^{+}T$. Uniqueness follows because the difference $\Delta$ of two solutions obeys $\Delta =V{G}_0^{+}\Delta$, forcing $\Vert \Delta \Vert \le \Vert V{G}_0^{+}\Vert \Vert \Delta \Vert <\Vert \Delta \Vert$ unless $\Delta =0$. The amplitude $f=-(m/2\pi {\hslash }^2)⟨{\mathbf{k}}^{\prime }|T|\mathbf{k}⟩$ then has the term-by-term expansion $f=f^{(1)}+f^{(2)}+\cdots$ with $f^{(1)}$ the first Born amplitude. $■$

Proposition (optical theorem from unitarity of the transition operator). The exact forward amplitude satisfies ${\sigma }_{\text{tot}}=(4\pi /k)\mathrm{Im}f(0)$.

Proof. The on-shell relation $T=V+V{G}_0^{+}T$ together with the Hermiticity of $V$ gives, on subtracting the adjoint relation $T^{†}=V+T^{†}{G}_0^{-}V$ and using $G_0^{+}-G_0^{-}=-2\pi i\delta (E-H_0)$ (the limiting-absorption jump of the resolvent across the continuous spectrum),

$$T-T^{†}=T^{†}(G_0^{+}-G_0^{-})T=-2\pi i{T}^{†}\delta (E-H_0)T.$$

Taking the diagonal forward matrix element $⟨\mathbf{k}|\cdots |\mathbf{k}⟩$ on the energy shell, the left side is $2i\mathrm{Im}⟨\mathbf{k}|T|\mathbf{k}⟩$ and the right side is $-2\pi i{\int }_{|{\mathbf{k}}^{\prime \prime }|=k}|⟨{\mathbf{k}}^{\prime \prime }|T|\mathbf{k}⟩{|}^{2}d{\Omega }^{\prime \prime }$ weighted by the on-shell density of states. Converting $T$-matrix elements to amplitudes via $f=-(m/2\pi {\hslash }^2)⟨T⟩$ and the density of states $mk/(2\pi {)}^2{\hslash }^2$, the forward equation reads $\mathrm{Im}f(0)=(k/4\pi )\int |f{|}^{2}d\Omega =(k/4\pi ){\sigma }_{\text{tot}}$, which rearranges to ${\sigma }_{\text{tot}}=(4\pi /k)\mathrm{Im}f(0)$. The identity holds for the exact $T$ and is independent of any perturbative truncation. $■$

Connections Master

• The survey unit 12.08.01 develops the partial-wave expansion, phase shifts, and the $S$-matrix; the Born amplitude derived here is its momentum-space complement, and the optical theorem proved here is identical to the one stated there in the angular-momentum basis. The two pictures are Fourier-dual descriptions of the same scattering operator.

• Time-dependent perturbation theory and Fermi's golden rule 12.07.02 supply the on-shell delta function $\delta (E-H_0)$ that reappears in the imaginary part of the second Born term; the golden-rule transition rate into a continuum of final states is the time-dependent face of the optical theorem's unitarity sum.

• Time-independent perturbation theory 12.07.01 is the bound-state analogue of the Born series: both expand a solution in powers of $V$ using the free resolvent, but the scattering problem replaces the discrete resolvent $(E_n-H_0{)}^{-1}$ by the continuum resolvent $(E-H_0+iϵ{)}^{-1}$ with its outgoing-wave $iϵ$ prescription.

• Spherical harmonics and Legendre polynomials 12.05.02 furnish the angular basis in which the Born amplitude's dependence on momentum transfer $q=2k\sin (\theta /2)$ is re-expressed as a sum over partial waves, the explicit dictionary between the Fourier-transform amplitude and the phase-shift series.

• Canonical quantum field theory 12.12.01 generalises the transition operator $T$ and its unitarity relation $T-T^{†}=T^{†}(G_0^{+}-G_0^{-})T$ into the relativistic $S$-matrix and the field-theoretic optical theorem; the Born series becomes the Feynman-diagram expansion, with $G_0^{+}$ promoted to the Feynman propagator.

Historical & philosophical context Master

Max Born introduced the perturbative scattering amplitude in 1926, in the two-part paper that also launched the probabilistic interpretation of the wavefunction ^{[Born 1926]}. The scattering calculation came first and the probability rule second: Born computed the cross-section for a beam scattering off a potential, found that the squared amplitude gave the observed deflection statistics, and inferred that $|\psi {|}^2$ must be a probability density. The interpretation that won him the 1954 Nobel Prize was extracted from a concrete collision problem, not postulated abstractly.

The integral-equation form, with its explicit outgoing-wave Green's function and the $iϵ$ boundary prescription, was given by Bernard Lippmann and Julian Schwinger in 1950 ^{[Lippmann 1950]}. Their reformulation moved the boundary condition from an external constraint into the structure of the equation itself, making the scattering problem a fixed-point equation $\psi =\varphi +G_0^{+}V\psi$ amenable to operator methods. Murray Gell-Mann and Marvin Goldberger supplied the time-dependent derivation in 1953 ^{[GellMann 1953]}, deriving the Lippmann-Schwinger states as the adiabatic limits of free states evolved with the interacting Hamiltonian — the Møller wave operators — and thereby connecting the stationary integral equation to the asymptotic-condition formulation of scattering. Landau and Lifshitz consolidated the practical apparatus in Volume 3 of their Course, Chapter XVII §§125-127, which remains the canonical physicist's reference for the Born approximation and its validity criteria.

Bibliography Master

@article{Born1926,
  author  = {Born, Max},
  title   = {Zur Quantenmechanik der Sto{\ss}vorg{\"a}nge},
  journal = {Zeitschrift f{\"u}r Physik},
  volume  = {37},
  pages   = {863--867},
  year    = {1926}
}

@article{Born1926b,
  author  = {Born, Max},
  title   = {Quantenmechanik der Sto{\ss}vorg{\"a}nge},
  journal = {Zeitschrift f{\"u}r Physik},
  volume  = {38},
  pages   = {803--827},
  year    = {1926}
}

@article{LippmannSchwinger1950,
  author  = {Lippmann, B. A. and Schwinger, Julian},
  title   = {Variational Principles for Scattering Processes. I},
  journal = {Physical Review},
  volume  = {79},
  number  = {3},
  pages   = {469--480},
  year    = {1950}
}

@article{GellMannGoldberger1953,
  author  = {Gell-Mann, Murray and Goldberger, Marvin L.},
  title   = {The Formal Theory of Scattering},
  journal = {Physical Review},
  volume  = {91},
  number  = {2},
  pages   = {398--408},
  year    = {1953}
}

@book{LandauLifshitz1977QM,
  author    = {Landau, L. D. and Lifshitz, E. M.},
  title     = {Quantum Mechanics: Non-Relativistic Theory},
  edition   = {3},
  series    = {Course of Theoretical Physics},
  volume    = {3},
  publisher = {Pergamon Press},
  year      = {1977}
}

@book{Newton1982,
  author    = {Newton, Roger G.},
  title     = {Scattering Theory of Waves and Particles},
  edition   = {2},
  publisher = {Springer-Verlag},
  year      = {1982}
}

@book{Taylor2006,
  author    = {Taylor, John R.},
  title     = {Scattering Theory: The Quantum Theory of Nonrelativistic Collisions},
  publisher = {Dover Publications},
  year      = {2006}
}

@book{SakuraiNapolitano2017,
  author    = {Sakurai, J. J. and Napolitano, Jim},
  title     = {Modern Quantum Mechanics},
  edition   = {2},
  publisher = {Cambridge University Press},
  year      = {2017}
}