THEORYMERIDIAN

Quantum Spin Lab · Entangled pairs

Can entangled spins
send instant messages?

Two electrons share a spin-singlet state. Measured along the same axis, their answers always come out opposite, however far apart they are. Yet each side alone sees a random 50/50 list.Be Alice and Bob: choose the axes, send pairs, then compare.

Ideal singlet model
01 / Fresh pairs, two records
Analyzer axis and resultsOpposite answers (after comparing)Ordinary signal used to compare

Simulated from quantum probabilitiesEvery trial is a new pair. Each side keeps its own record. Comparing records takes an ordinary signal, no faster than light. This is not a Bell experiment. The bench is an illustration; each dial shows its analyzer's axis, both in the same frame.

Bob alone: share of +

50%
Pairs in this run
0
Alice alone: share of +
—
Angle between the axes
0°
Opposite answers (after comparing)
not compared
Correlation E (after comparing)
not compared

Start here: pick a story (sends 100 pairs), then compare

Send some pairs. Each side sees only its own random list.

Changing either axis starts a fresh run, so pairs measured with different settings are never pooled.

03Send pairs

Clear records starts a fresh run; the finished run stays in the table below.

Alice cannot choose her results. They are random, so she has nothing to encode. Whatever axis she picks, Bob's own list stays 50/50. The pattern appears only when both lists are brought together.

Can Alice send a message?

Every run, Bob sees
the same 50/50.

Each row is one run with fixed settings. Change Alice's axis between runs and look at Bob's column: it does not carry her choice. Only the compared column depends on both settings.

Runs in this session (simulated)
RunAlice axisBob axisPairsBob: share of +Compared: opposite
No runs yet.

Real experiments

Correlations tested,
signals ruled out.

Bell showed that correlations like these exceed what any local hidden-variable model allows. Experiments confirm the quantum predictions under their stated assumptions. None of them sends a message faster than light.

Real experiment

Electron spins 1.3 km apart

In 2015 Hensen and colleagues entangled electron spins at two sites 1.3 km apart and tested Bell's inequality, closing the detection and locality loopholes in one experiment. In 245 trials they measured S = 2.42 ± 0.20, above the local-realist limit of 2.

B. Hensen et al., Nature 526, 682 (2015), DOI 10.1038/nature15759.

Theory

Bell, 1964

For the singlet, quantum theory predicts a CHSH value of 2√2 ≈ 2.83 in size for suitable axes; local hidden-variable models stay at or below 2. The random draws on this page come from the quantum prediction, so they test nothing.

J. S. Bell, Physics 1, 195 (1964).

The reasoning, in the open

Look under the model.

The calculation

For the spin singlet and ideal analyzers along unit axes a (Alice) and b (Bob), the chance of results A and B (each +1 or −1) is

P(A, B) = (1 − A B a·b) / 4, E(a, b) = −a·b

Along the same axis the answers are always opposite. Summing over Alice's results gives Bob 1/2 for + whatever axis Alice uses, and the same holds the other way round: neither side's choice changes the other side's own odds. With the axes 90° apart all four combinations have probability 1/4. The CHSH combination S = E(a,b) + E(a,b′) + E(a′,b) − E(a′,b′) reaches −2√2 for a = 0°, a′ = 90°, b = 45°, b′ = −45°.

Each simulated pair is drawn at random from P(A, B) with a seeded generator. All axes lie in one plane for clarity.

Read the numerical model · Validation record

Sources
Limits of this model

Ideal singlets and perfect analyzers. The on-screen pairs are drawn from the quantum prediction; they are not a Bell test and do not exclude anything. The animation shows pairs leaving the source and an ordinary signal used for comparison; nothing travels between Alice and Bob at the moment of measurement.

Keyboard, motion and privacy

Use Tab to reach each control, arrow keys for sliders, and Enter or Space for buttons. Pause freezes the illustration only. Motion starts paused when your system asks for reduced motion. Results are written as text. The page runs entirely in your browser with no accounts, trackers or external code. See the privacy notice.