Skip to content

Separable Mixed States

A separable mixed state is a bipartite density operator that can be prepared as a classical random mixture of product states. It may contain ordinary correlations between subsystems, but it contains no entanglement across the specified subsystem split.

For a bipartite system

HAB=HA⊗HB,\mathcal H_{AB} = \mathcal H_A\otimes\mathcal H_B,

a mixed state is separable if it can be written as

ρAB=∑rpr ρA(r)⊗ρB(r),pr≥0,∑rpr=1.\rho_{AB} = \sum_r p_r\, \rho_A^{(r)}\otimes\rho_B^{(r)}, \qquad p_r\ge0, \qquad \sum_r p_r=1.

If no such decomposition exists, the mixed state is entangled.

The simplest unentangled mixed state is a product density operator:

ρAB=ρA⊗ρB.\rho_{AB} = \rho_A\otimes\rho_B.

This represents independent subsystem states. For local observables,

Tr⁡[(ρA⊗ρB)(OA⊗OB)]=Tr⁡(ρAOA) Tr⁡(ρBOB).\operatorname{Tr} \bigl[ (\rho_A\otimes\rho_B) (O_A\otimes O_B) \bigr] = \operatorname{Tr}(\rho_A O_A)\, \operatorname{Tr}(\rho_B O_B).

Thus product mixed states have factorized expectation values for product observables. They have no correlations between AA and BB.

Product mixed states are separable, but they are not the only separable states.

A separable state may be a mixture of different product preparations. Operationally, one can imagine a classical random variable rr being sampled with probability prp_r, after which subsystem AA is prepared in ρA(r)\rho_A^{(r)} and subsystem BB is prepared in ρB(r)\rho_B^{(r)}.

The resulting density operator is

ρAB=∑rpr ρA(r)⊗ρB(r).\rho_{AB} = \sum_r p_r\, \rho_A^{(r)}\otimes\rho_B^{(r)}.

The shared label rr can create correlations. Those correlations are classical in the sense that they arise from common randomness in the preparation, not from nonseparability of the quantum state.

In finite-dimensional systems, one may also write separable states as mixtures of pure product projectors:

ρAB=∑rpr ∣αr⟩⟨αr∣⊗∣βr⟩⟨βr∣.\rho_{AB} = \sum_r p_r\, \lvert\alpha_r\rangle\langle\alpha_r\rvert \otimes \lvert\beta_r\rangle\langle\beta_r\rvert.

This follows by decomposing the local density operators in each product term into pure-state ensembles.

The standard two-qubit example is

ρcc=12∣00⟩⟨00∣+12∣11⟩⟨11∣.\rho_{\mathrm{cc}} = \frac12 \lvert00\rangle\langle00\rvert + \frac12 \lvert11\rangle\langle11\rvert.

This state is separable because it is already written as a convex mixture of product projectors:

ρcc=12(∣0⟩⟨0∣⊗∣0⟩⟨0∣)+12(∣1⟩⟨1∣⊗∣1⟩⟨1∣).\rho_{\mathrm{cc}} = \frac12 \bigl( \lvert0\rangle\langle0\rvert \otimes \lvert0\rangle\langle0\rvert \bigr) + \frac12 \bigl( \lvert1\rangle\langle1\rvert \otimes \lvert1\rangle\langle1\rvert \bigr).

It is not a product state. The reduced density operators are

ρA=ρB=12I.\rho_A = \rho_B = \frac12 I.

If it were the product of its reduced states, it would be

ρA⊗ρB=14I⊗I,\rho_A\otimes\rho_B = \frac14 I\otimes I,

which assigns probability 1/41/4 to each computational-basis outcome. Instead, ρcc\rho_{\mathrm{cc}} assigns probability 1/21/2 to 0000, probability 1/21/2 to 1111, and zero to 0101 and 1010.

Thus

P(0,0)=12,PA(0)PB(0)=14.P(0,0) = \frac12, \qquad P_A(0)P_B(0) = \frac14.

The state is correlated, but not entangled.

For pure bipartite states, the definition is simple: product means unentangled, and not product means entangled.

For mixed states, “not product” is too weak. A mixed state can fail to factor as ρA⊗ρB\rho_A\otimes\rho_B simply because it contains classical correlation. The state ρcc\rho_{\mathrm{cc}} above is the basic example.

The correct mixed-state distinction is:

product⟹separable⟹not necessarily product.\text{product} \quad\Longrightarrow\quad \text{separable} \quad\Longrightarrow\quad \text{not necessarily product}.

Entangled mixed states are precisely the states outside the separable set.

A density operator can have many different ensemble decompositions. This matters because seeing an entangled-state ensemble does not prove that the density operator is entangled.

For example, the maximally mixed two-qubit state is

ρ∗=14IAB.\rho_* = \frac14 I_{AB}.

It can be written as a product state:

ρ∗=(12IA)⊗(12IB).\rho_* = \left(\frac12 I_A\right) \otimes \left(\frac12 I_B\right).

So it is separable. But it can also be written as an equal mixture of Bell projectors:

ρ∗=14∑μ∣Bμ⟩⟨Bμ∣,\rho_* = \frac14 \sum_{\mu} \lvert B_\mu\rangle\langle B_\mu\rvert,

where {∣Bμ⟩}\{\lvert B_\mu\rangle\} is the Bell basis. The presence of entangled vectors in one ensemble decomposition does not make the density operator entangled. Entanglement of a mixed state is a property of the density operator itself, not of one chosen story about how it was prepared.

Pure-state entanglement has strong tools: coefficient rank, Schmidt decomposition, and reduced-state entropy.

Mixed-state entanglement is harder because one must rule out every possible separable decomposition:

ρAB≠∑rpr ρA(r)⊗ρB(r)\rho_{AB} \ne \sum_r p_r\, \rho_A^{(r)}\otimes\rho_B^{(r)}

for all choices of probabilities and local states.

Useful criteria include partial-transpose tests, entanglement witnesses, semidefinite programs, and special formulas for low-dimensional systems. These belong to a later mixed-state entanglement toolkit. The definition, however, is already fixed: separable means convex mixture of product states.

A product mixed state is

ρA⊗ρB.\rho_A\otimes\rho_B.

It is separable and uncorrelated.

A diagonal classical two-qubit state

ρ=∑i,j=01pij∣ij⟩⟨ij∣,pij≥0,∑ijpij=1,\rho = \sum_{i,j=0}^{1} p_{ij} \lvert ij\rangle\langle ij\rvert, \qquad p_{ij}\ge0, \qquad \sum_{ij}p_{ij}=1,

is separable because it is a mixture of computational-basis product projectors. It can still be classically correlated if pijp_{ij} does not factor as piApjBp_i^A p_j^B.

A Bell pure state such as

∣Φ+⟩=12(∣00⟩+∣11⟩)\lvert\Phi^+\rangle = \frac{1}{\sqrt2} \bigl( \lvert00\rangle+\lvert11\rangle \bigr)

is not separable, because for pure states separability reduces to productness.

  • Calling every nonproduct mixed state entangled.
  • Calling every correlated mixed state entangled.
  • Treating one ensemble decomposition as the unique meaning of a density operator.
  • Thinking that a mixture of entangled kets is automatically an entangled mixed state.
  • Forgetting to specify the subsystem split A∣BA|B.
  • Assuming that separable means uncorrelated; separable states may have classical correlations.
  • R. F. Werner, “Quantum States with Einstein-Podolsky-Rosen Correlations Admitting a Hidden-Variable Model,” Physical Review A 40, 4277-4281, 1989.
  • A. Peres, Quantum Theory: Concepts and Methods, Kluwer, 1995.
  • M. A. Nielsen and I. L. Chuang, Quantum Computation and Quantum Information, Cambridge University Press, 2010.
  • R. Horodecki, P. Horodecki, M. Horodecki, and K. Horodecki, “Quantum Entanglement,” Reviews of Modern Physics 81, 865-942, 2009.
  • L. E. Ballentine, Quantum Mechanics: A Modern Development, 2nd ed., World Scientific, 2014.
  1. Show that
ρcc=12∣00⟩⟨00∣+12∣11⟩⟨11∣\rho_{\mathrm{cc}} = \frac12\lvert00\rangle\langle00\rvert + \frac12\lvert11\rangle\langle11\rvert

is separable but not product.

Solution

It is separable because it is a convex mixture of product projectors. Its reduced states are

ρA=ρB=12I.\rho_A=\rho_B=\frac12 I.

The product of the reduced states is

ρA⊗ρB=14I⊗I,\rho_A\otimes\rho_B = \frac14 I\otimes I,

which is not equal to ρcc\rho_{\mathrm{cc}}. Therefore the state is separable but not product.

  1. Compute the computational-basis probabilities for ρcc\rho_{\mathrm{cc}} and show that the outcomes are correlated.
Solution

The nonzero joint probabilities are

P(0,0)=12,P(1,1)=12.P(0,0)=\frac12, \qquad P(1,1)=\frac12.

The marginal probabilities are

PA(0)=PB(0)=12.P_A(0)=P_B(0)=\frac12.

If the outcomes were independent, P(0,0)P(0,0) would equal PA(0)PB(0)=1/4P_A(0)P_B(0)=1/4. Instead P(0,0)=1/2P(0,0)=1/2, so the outcomes are correlated.

  1. Explain why an equal mixture of Bell projectors can still be separable.
Solution

The equal mixture of the four Bell projectors is the maximally mixed state:

ρ∗=14IAB.\rho_* = \frac14 I_{AB}.

But this also factors as

ρ∗=(12IA)⊗(12IB).\rho_* = \left(\frac12 I_A\right) \otimes \left(\frac12 I_B\right).

Since it is a product density operator, it is separable. A density operator can have many ensemble decompositions, so one decomposition using entangled kets does not prove mixed-state entanglement.

  1. For
ρp=p∣00⟩⟨00∣+(1−p)∣11⟩⟨11∣,0≤p≤1,\rho_p = p\lvert00\rangle\langle00\rvert + (1-p)\lvert11\rangle\langle11\rvert, \qquad 0\le p\le1,

for which pp is the state product?

Solution

The state is separable for every pp because it is a convex mixture of product projectors. Its reduced states are

ρA=ρB=p∣0⟩⟨0∣+(1−p)∣1⟩⟨1∣.\rho_A=\rho_B = p\lvert0\rangle\langle0\rvert + (1-p)\lvert1\rangle\langle1\rvert.

For 0<p<10<p<1, the product ρA⊗ρB\rho_A\otimes\rho_B contains nonzero probabilities for ∣01⟩\lvert01\rangle and ∣10⟩\lvert10\rangle, while ρp\rho_p does not. Thus ρp\rho_p is product only at p=0p=0 or p=1p=1.