Quantum Entanglement
Entanglement is a property of a composite state that cannot be written as a tensor product of states of its parts; it yields correlations that violate Bell inequalities but cannot be used to signal.
A pure state of two systems $A$ and $B$ is called separable, or a product state, if it can be written $|\psi\rangle=|a\rangle\otimes|b\rangle$; otherwise it is entangled. The best-known example is the Bell state $$|\Phi^+\rangle=\frac{1}{\sqrt2}(|00\rangle+|11\rangle),$$ prepared with h(0) followed by cx(0,1). When both qubits are measured in the computational basis we get 00 or 11, each with probability $\tfrac12$, and never 01 or 10.
For a two-qubit state $a|00\rangle+b|01\rangle+c|10\rangle+d|11\rangle$ there is a simple test: it is a product state if and only if $ad-bc=0$. In general, entanglement in pure states is measured by the entanglement entropy, i.e. the von Neumann entropy of the reduced state of one of the two parts; it is one full bit for Bell states.
Correlation alone is not evidence of entanglement, since classically correlated coins give correlation in one basis. What distinguishes entanglement is that the correlations stay strong in several incompatible bases, violating the CHSH inequality that all local hidden-variable theories respect; experiments have confirmed this violation, including a loophole-free experiment in 2015.
A crucial point: entanglement does not allow information to be sent faster than light. This is the no-communication theorem: whatever party $B$ does to its qubit, the reduced density matrix of $A$ does not change, so the distribution of $A$'s outcomes alone stays completely random ($\tfrac12,\tfrac12$ for a Bell state). The correlation appears only when the outcomes are compared, which requires a classical channel. That is why teleportation requires sending two classical bits.
Entanglement is a computational resource: without it, a pure circuit of qubits can be simulated efficiently classically. But its presence is not enough to guarantee an advantage; Clifford circuits, for example, generate a lot of entanglement and can still be simulated efficiently.
In the lab you can observe these properties in practice: measuring both qubits of $|\Phi^+\rangle$ in the $Z$ basis and then in the $X$ basis shows perfect correlation in both cases, while the marginal distribution of each qubit stays uniform. This combination of strong correlation with a completely random marginal distribution is the signature of maximal entanglement.
Mathematical formula
Worked example
Test two states with the condition $ad-bc$:
- $|\Phi^+\rangle$: $a=d=\tfrac{1}{\sqrt2}$, $b=c=0$, so $ad-bc=\tfrac12\neq0$; the state is entangled.
- $\tfrac12(|00\rangle+|01\rangle+|10\rangle+|11\rangle)$: $ad-bc=\tfrac14-\tfrac14=0$; a product state, indeed $|+\rangle\otimes|+\rangle$.
No signalling: in $|\Phi^+\rangle$ the probability that $A$ measures 0 is $\tfrac12$ whether or not $B$ measures first, and whether or not $B$ applies an $X$ gate; $A$ sees no trace of $B$'s choice.
Test your understanding
The quiz needs JavaScript.
Related topics
References
- Michael A. Nielsen; Isaac L. Chuang (2010). Quantum Computation and Quantum Information (10th Anniversary Edition). Cambridge University Press doi:10.1017/CBO9780511976667The standard reference for quantum computing; the Crossref record gives 2012 for the Cambridge Core online edition.
- John Watrous / IBM Quantum (2025). Basics of Quantum Information (course). IBM Quantum Learning quantum.cloud.ibm.com
- J. F. Clauser; M. A. Horne; A. Shimony; R. A. Holt (1969). Proposed Experiment to Test Local Hidden-Variable Theories. Phys. Rev. Lett. 23(15):880–884 doi:10.1103/PhysRevLett.23.880
