Entanglement

Quantum entanglement is one of the most counterintuitive features of quantum mechanics and one of its most useful. When two qubits are entangled, their states cannot be described independently. There is no state for qubit A and a separate state for qubit B; there is only a joint state for the pair. Measuring one qubit instantly constrains what you will find when you measure the other, regardless of the distance between them.

Einstein called this “spooky action at a distance” and argued it proved quantum mechanics was incomplete. Bell inequality tests, beginning with Aspect’s experiments in 1982 and culminating in loophole-free tests in 2015, proved him wrong. The correlations are real, and no classical model can reproduce them.

The details

The simplest entangled state is the Bell state:

$|\Phi^+\rangle = \frac{|00\rangle + |11\rangle}{\sqrt{2}}$

Measuring both qubits always yields 00 or 11, never 01 or 10, even though each individual measurement is equally likely to give 0 or 1. This perfect correlation cannot be explained by any classical probability distribution over hidden variables, as Bell’s theorem proves.

A state is unentangled (a product state) if it can be written as:

$|\psi\rangle = |\psi_A\rangle \otimes |\psi_B\rangle$

For example, $|+\rangle \otimes |0\rangle = (|0\rangle + |1\rangle)/\sqrt{2} \otimes |0\rangle = (|00\rangle + |10\rangle)/\sqrt{2}$ is a product state. The Bell state above cannot be written this way.

Entanglement is quantified by entanglement entropy. For a bipartite state $|\psi_{AB}\rangle$, compute the reduced density matrix $\rho_A = \text{Tr}_B(|\psi\rangle\langle\psi|)$ and calculate the von Neumann entropy $S = -\text{Tr}(\rho_A \log_2 \rho_A)$. A product state has $S = 0$; a maximally entangled Bell state has $S = 1$ ebit. For $n$ qubits, the maximum entanglement entropy grows as $n/2$ ebits.

Creating entanglement in a circuit requires at least one two-qubit gate. The standard recipe uses a Hadamard followed by a CNOT:

from qiskit import QuantumCircuit

qc = QuantumCircuit(2)
qc.h(0)
qc.cx(0, 1)
# State is now (|00> + |11>) / sqrt(2)

Why it matters for learners

Entanglement is the resource that makes quantum computation qualitatively different from classical computation. Three applications are critical to understand:

Algorithms: Shor’s algorithm and Grover’s algorithm both require entanglement between qubits. Without it, the interference effects that give quantum speedups cannot occur.

Error correction: Quantum error correction encodes one logical qubit into an entangled state spread across many physical qubits. The entanglement is what allows errors to be detected without measuring the protected state directly.

Communication: Quantum teleportation consumes one entangled pair to transfer one qubit’s worth of state. Quantum key distribution uses entangled pairs to detect eavesdropping with physical certainty.

Common misconceptions

Misconception 1: Entanglement allows faster-than-light communication. It does not. The correlation exists, but neither party can control the outcome of their local measurement. The outcome is random. To extract the correlation, the parties must compare their results over a classical channel, which cannot exceed the speed of light.

Misconception 2: Entanglement means the qubits are physically connected. Entangled qubits can be separated by arbitrary distances. The entanglement is a property of the joint quantum state, not a physical link. Once established through a prior interaction, the correlations persist until one of the qubits is measured or decoheres.

Misconception 3: More entanglement is always better. Highly entangled states are computationally powerful but also harder to prepare, more sensitive to noise, and harder to simulate classically. The right amount of entanglement for a given algorithm is specific to that algorithm’s structure.

See also

• Bell State
• Qubit
• Quantum Teleportation
• Quantum Error Correction
• CNOT Gate