|q⟩ Bad Qubits

beginner · Physics · The Schrödinger Equation (Intro)

Boundary Conditions

The time-independent Schrödinger equation,

22md2ψdx2+V(x)ψ=Eψ,-\frac{\hbar^2}{2m}\frac{d^2\psi}{dx^2} + V(x)\psi = E\psi,

is a second-order ordinary differential equation. Its general solution in any region where VV is constant contains two free constants, which are fixed by imposing boundary conditions — rules that ψ\psi must satisfy wherever the potential changes.

Why continuity is required

The Schrödinger equation fixes d2ψ/dx2=2m2(VE)ψd^2\psi/dx^2 = \frac{2m}{\hbar^2}(V-E)\psi. Wherever VV is finite, the right-hand side is bounded, so d2ψ/dx2d^2\psi/dx^2 is bounded. Integrating a bounded function across an infinitesimal interval gives a change that vanishes as the interval shrinks, so dψ/dxd\psi/dx is continuous; and because dψ/dxd\psi/dx is finite, integrating once more makes ψ\psi itself continuous. These requirements follow from the equation's structure:

  1. ψ(x)\psi(x) must be continuous everywhere.
  2. dψ/dxd\psi/dx must be continuous wherever VV is finite (see the caveat below).

Together, conditions 1 and 2 are the standard boundary conditions applied at every point where two regions of different potential meet.

Matching across a boundary

Consider a particle at a boundary x=ax = a between two regions. Label the solution in the left region ψL\psi_L and in the right region ψR\psi_R. Continuity of ψ\psi requires

ψL(a)=ψR(a).\psi_L(a) = \psi_R(a).

Continuity of the derivative requires

ψL(a)=ψR(a).\psi_L'(a) = \psi_R'(a).

These two equations — sometimes called the junction conditions — are the practical tool for constructing solutions that span multiple regions. They reduce the total number of free constants from (2 per region) to a manageable set, and combined with normalization they fix the energy eigenvalues.

The infinite-potential exception

What happens when the potential is infinite, as at the hard walls of an infinite square well? Rewriting the Schrödinger equation as

d2ψdx2=2m2(VE)ψ,\frac{d^2\psi}{dx^2} = \frac{2m}{\hbar^2}(V - E)\psi,

reveals that if VV \to \infty inside a region, the only way to keep d2ψ/dx2d^2\psi/dx^2 finite is to require ψ=0\psi = 0 throughout that region. At the wall itself, continuity of ψ\psi then demands

ψ=0at the wall.\psi = 0 \quad \text{at the wall.}

The derivative dψ/dxd\psi/dx, however, need not be continuous at an infinite-potential wall, because the second derivative is singular there anyway. This is the one exception to condition 2 above.

Physical meaning of the conditions

The two conditions encode real physics. Continuity of ψ\psi is required because ψ2|\psi|^2 is a probability density: a discontinuous probability density would assign different probabilities to the same point from the left and the right, which is meaningless. Continuity of ψ\psi' follows from the finite-kinetic-energy condition: the kinetic energy operator involves d2ψ/dx2d^2\psi/dx^2, and a jump in ψ\psi' would make d2ψ/dx2d^2\psi/dx^2 a Dirac delta, which would require an infinitely sharp spike in the potential to balance it.

Worked example: finite step potential

As a concrete illustration, consider a particle of energy EE travelling from left to right, with V=0V = 0 for x<0x < 0 and V=V0V = V_0 (finite, with E>V0>0E > V_0 > 0) for x0x \geq 0.

Left region (x<0x < 0): The Schrödinger equation reads

d2ψdx2=k12ψ,k1=2mE.\frac{d^2\psi}{dx^2} = -k_1^2\psi, \quad k_1 = \frac{\sqrt{2mE}}{\hbar}.

The general solution is ψL=Aeik1x+Beik1x\psi_L = Ae^{ik_1 x} + Be^{-ik_1 x}, combining an incident wave and a reflected wave.

Right region (x0x \geq 0): With potential V0V_0,

d2ψdx2=k22ψ,k2=2m(EV0).\frac{d^2\psi}{dx^2} = -k_2^2\psi, \quad k_2 = \frac{\sqrt{2m(E-V_0)}}{\hbar}.

The transmitted wave moving to the right is ψR=Ceik2x\psi_R = Ce^{ik_2 x}.

Applying junction conditions at x=0x = 0:

Continuity of ψ\psi: A+B=C.A + B = C.

Continuity of ψ\psi': ik1(AB)=ik2C.ik_1(A - B) = ik_2 C.

These two equations relate BB and CC to AA. Solving,

BA=k1k2k1+k2,CA=2k1k1+k2.\frac{B}{A} = \frac{k_1 - k_2}{k_1 + k_2}, \qquad \frac{C}{A} = \frac{2k_1}{k_1 + k_2}.

The reflection coefficient R=B/A2R = |B/A|^2 and transmission coefficient T=(k2/k1)C/A2T = (k_2/k_1)|C/A|^2 satisfy R+T=1R + T = 1, confirming probability is conserved. Notice that even when E>V0E > V_0 — the classically unrestricted case — there is a non-zero reflected amplitude whenever k1k2k_1 \neq k_2. This quantum reflection has no classical analogue and is a direct consequence of the wave nature of ψ\psi.

Summary

| Condition | When it applies | Mathematical form | |-----------|----------------|-------------------| | Continuity of ψ\psi | Always | ψL(a)=ψR(a)\psi_L(a) = \psi_R(a) | | Continuity of ψ\psi' | VV finite at aa | ψL(a)=ψR(a)\psi_L'(a) = \psi_R'(a) | | ψ=0\psi = 0 at wall | V=V = \infty inside | ψ(a)=0\psi(a) = 0 |

Applying these conditions at every interface is how quantum mechanics turns a differential equation into discrete energy levels, transmission coefficients, and the other quantitative predictions that can be tested against experiment.

Sign in on the full site to ask questions and join the discussion.