intermediate · Physics · Density Matrices & Mixed States
Mixed States from Entanglement
Entanglement and mixedness are two sides of the same coin. Here we make the link precise: a subsystem of
an entangled pure state is always mixed, and the amount of mixedness measures the amount of
entanglement.
The Schmidt decomposition
Any pure bipartite state can be written in a special diagonal form, the Schmidt decomposition:
∣ψ⟩AB=k∑λk∣k⟩A∣k⟩B,
with Schmidt coefficients λk≥0 summing to 1, and orthonormal sets {∣k⟩A},
{∣k⟩B}. The number of nonzero λk is the Schmidt rank.
Reduced states from the Schmidt form
Tracing out either subsystem of the Schmidt form gives a diagonal density matrix whose entries are the
Schmidt coefficients:
Both reduced states share the same spectrum {λk}. Two immediate consequences:
Product (unentangled) state ⟺ pure reduced state. A single nonzero λk=1 means
∣ψ⟩=∣k⟩A∣k⟩B factorizes, and ρA is the pure state
∣k⟩A⟨k∣A.
Entangled state ⟺ mixed reduced state. Two or more nonzero λk means the state cannot
factorize, and ρA is mixed.
Mixedness measures entanglement
The purity of the reduced state quantifies the entanglement: Tr(ρA2)=∑kλk2, which is 1 for a product state and drops as entanglement grows. The
entanglement entropyS(ρA)=−∑kλklog2λk packages the same information,
reaching its maximum for a maximally entangled state like ∣Φ+⟩, whose reduced state is
I/2.
Try it
For the partially entangled state ∣ψ⟩=cost∣00⟩+sint∣11⟩ at
t=π/6, the Schmidt coefficients are cos2t and sin2t, so ρA=diag(cos2t,sin2t).
Compute the purity Tr(ρA2). A value below 1 confirms the subsystem is mixed —
the fingerprint of entanglement.
Run your code to see the quantum state.
Sign in on the full site to ask questions and join the discussion.