In undergraduate quantum mechanics you typically compute tunneling probabilities for simple barriers — a rectangular “top hat” or a parabola. The WKB (Wentzel–Kramers–Brillouin) approximation lets you handle an arbitrary potential barrier V(x) with the same machinery. The key is the transfer matrix that connects the wave amplitudes on either side.
Consider a particle with energy E hitting a barrier V(x). The classical turning points are at x=α and x=β (where E=V(x)), dividing space into three regions:
Region I (x<α): classically accessible, oscillatory wavefunction
Region II (α<x<β): classically forbidden, exponentially decaying/growing
Region III (x>β): classically accessible again, transmitted wave
Define k(x)=2m(E−V)/ℏ2 in the accessible regions and κ(x)=2m(V−E)/ℏ2 in the forbidden region. The general WKB solutions in each region are:
ψ1(x)=kaexp(−i∫xαkdx)+kbexp(i∫xαkdx)
ψ2(x)=κcexp(−∫αxκdx)+κdexp(∫αxκdx)
ψ3(x)=kfexp(i∫βxkdx)+kgexp(−i∫βxkdx)
The goal is to find (ab)=M(fg) — the matrix relating incoming coefficients to outgoing ones.
Define θ≡exp(∫αβκdx) — the exponential of the total barrier integral. This is the key quantity controlling tunneling.
Rewriting the Region II wavefunction in terms of integrals running from β (to match the right connection formula), and comparing with the standard form, gives the simple relations:
A′=Bθ,B′=A/θ
The growing exponential across the barrier picks up a factor θ, and the decaying one picks up 1/θ.
For a particle incident from the left we set g=0 (no incoming wave from the right). From the transfer matrix:
a=21(2θ+2θ1)f=(θ+4θ1)f
The transmission coefficient T=∣f∣2/∣a∣2:
T=∣θ+4θ1∣21=θ2∣1+4θ21∣21
In the deep tunneling limitθ≫1 (thick or tall barrier), the term 4θ21→0:
T≃θ−2=exp(−2∫αβκ(x)dx)
This is the standard WKB tunneling formula — the transmission probability is the exponential of −2 times the integral of κ(x) through the classically forbidden region. The shape of the barrier only matters through this single integral, which is why WKB is so powerful for estimating tunneling rates in nuclear physics, alpha decay, scanning tunneling microscopy, and field emission.