Quantum Experiments

Quantum Experiments · II · Paths, information, and time · Chapter 10

Quantum Erasure and Conditional Fringes

Where are the fringes hiding when path information is available?

In this chapter
  1. Entangled path markers and coincidence sorting
  2. Basis rotation of the marker and antiphase fringes
  3. No-signalling and coincidence-window discipline
  4. Interactive experiment
  5. Problems and solutions

10.1 Entangled path markers and coincidence sorting

10.1.1 Tagging interferometer arms with orthogonal marker states

Suppose we entangle the two paths \(\lvert L\rangle\) and \(\lvert R\rangle\) of an interferometer with two orthogonal states \(\lvert H\rangle\) and \(\lvert V\rangle\) of an ancillary marker system (such as the polarization of the photon itself or of a twin herald photon). Because \(\langle V\vert H\rangle=0\), the marker carries complete which-path information, and the unconditional signal counts at the interferometer output become completely flat (\(P(0)=1/2\) for all \(\phi\)), showing zero interference visibility.

10.1.2 Sorting signal arrivals by diagonal marker outcomes

Now, instead of measuring the marker in the path-distinguishing basis \(\{\lvert H\rangle,\lvert V\rangle\}\), rotate the marker analyzer to the diagonal superposition basis \(\lvert\pm\rangle=(\lvert H\rangle\pm\lvert V\rangle)/\sqrt{2}\) that projects equally onto both path labels. When each signal detection is paired in coincidence with the marker outcome \(+\) or \(-\), the two sorted sub-ensembles recover full-contrast interference fringes—shifted in phase by \(\pi\) relative to one another!

10.1.3 Complementary sub-ensembles versus the flat marginal sum

In the interactive plot below, three curves appear simultaneously: the blue fringe \(P(0\mid +)\) conditioned on marker outcome \(+\), the dashed gold antiphase fringe \(P(0\mid -)\) conditioned on marker outcome \(-\), and the flat dotted violet line \(P(0)=1/2\) representing all unsorted signal detections. Quantum erasure never alters the flat unsorted signal distribution; it extracts complementary conditional correlations by cross-tabulating signal and marker records.

Visual animation · Section 10.1

Entangled path markers and coincidence sorting

Entangled path markers and coincidence sortingSchematic apparatus; geometry and moving indicators are explanatory, not measured trajectories.
\(\mathrm{BS}_1\)
\(\mathrm{BS}_2\)
Sourceupper armlower arm
\(D_0\)
port 0
\(D_1\)
port 1
Phase \(\phi\)
Path marker
\(|H\rangle / |V\rangle \to |\pm\rangle\)
Section 10.1 Live State:
\(P(D_0\mid \pm) = \frac{1}{2}(1 \pm \cos\phi), \quad P_{\mathrm{unsorted}}(D_0) = \frac{1}{2}\)
Stage 1 (10.1.1): Tagging interferometer arms with orthogonal marker statesReady

Interactive visual model for 10.1 Entangled path markers and coincidence sorting (Tagging interferometer arms with orthogonal marker states · Sorting signal arrivals by diagonal marker outcomes · Complementary sub-ensembles versus the flat marginal sum). Use Animate or Step to follow the physical state evolution and quantitative predictions.

10.2 Basis rotation of the marker and antiphase fringes

10.2.1 Rewriting the path–marker state in the diagonal basis

Before the recombining beam splitter, the maximally entangled path–marker state is

\[\lvert\Psi(\phi)\rangle = \frac{1}{\sqrt{2}}\bigl(\lvert L\rangle\lvert H\rangle + e^{i\phi}\lvert R\rangle\lvert V\rangle\bigr).\]

Substituting \(\lvert H\rangle=(\lvert +\rangle+\lvert -\rangle)/\sqrt{2}\) and \(\lvert V\rangle=(\lvert +\rangle-\lvert -\rangle)/\sqrt{2}\) rewrites the exact same state vector in the diagonal marker basis without any physical disturbance to the signal arm:

\[\lvert\Psi(\phi)\rangle = \frac{1}{2}\bigl(\lvert L\rangle + e^{i\phi}\lvert R\rangle\bigr)\lvert +\rangle + \frac{1}{2}\bigl(\lvert L\rangle - e^{i\phi}\lvert R\rangle\bigr)\lvert -\rangle.\]

10.2.2 Opposite-phase conditional fringes and exact marginal cancellation

After the signal modes \(\lvert L\rangle\) and \(\lvert R\rangle\) recombine at the output beam splitter \(H\), the conditional probabilities of a signal click at port \(D_0\) given marker outcomes \(+\) and \(-\) are

\[P(0\mid +) = \frac{1+\cos\phi}{2}, \qquad P(0\mid -) = \frac{1-\cos\phi}{2} = \frac{1+\cos(\phi+\pi)}{2}.\]

Because each marker outcome \(m\in\{+,-\}\) occurs with marginal probability \(P(m)=1/2\), the total unconditional probability at port \(D_0\) is

\[P(0) = P(+)\,P(0\mid +) + P(-)\,P(0\mid -) = \frac{1}{2}\left(\frac{1+\cos\phi}{2}\right) + \frac{1}{2}\left(\frac{1-\cos\phi}{2}\right) = \frac{1}{2},\]

independent of \(\phi\). The crest of the \(+\) fringe sits directly over the trough of the \(-\) fringe, so their sum is identically flat.

10.2.3 Commuting local measurements and post-hoc coincidence sorting

When the signal and marker live on spatially separated photons, their local measurement operators \(M_{\mathrm{sig}}\otimes I\) and \(I\otimes M_{\mathrm{mark}}\) commute: \((M_{\mathrm{sig}}\otimes I)(I\otimes M_{\mathrm{mark}})=M_{\mathrm{sig}}\otimes M_{\mathrm{mark}}\). Consequently, the joint probability \(P(0,\pm)\) is completely invariant under whether the marker is measured before or years after the signal photon hits its detector. Delayed-choice quantum erasure simply sorts already-recorded signal timestamps using the classical log of later marker outcomes.

Visual animation · Section 10.2

Basis rotation of the marker and antiphase fringes

10.2 Basis rotation of the marker and antiphase fringesInteractive geometric state/phasor dial and quantitative prediction curve for Basis rotation of the marker and antiphase fringes.1. State / Amplitude GeometryAmplitude / Bloch projection
Value \(= \) 0.500
2. Quantitative Law (10.2)
\(P(D_0\mid \pm) = \frac{1}{2}(1 \pm \cos\phi), \quad P_{\mathrm{unsorted}}(D_0) = \frac{1}{2}\)

Interactive visual model for 10.2 Basis rotation of the marker and antiphase fringes (Rewriting the path–marker state in the diagonal basis · Opposite-phase conditional fringes and exact marginal cancellation · Commuting local measurements and post-hoc coincidence sorting). Use Animate or Step to follow the physical state evolution and quantitative predictions.

10.3 No-signalling and coincidence-window discipline

10.3.1 Why quantum erasure cannot transmit a faster-than-light signal

Can an experimenter at the marker station send an instantaneous message to the signal station by choosing whether to measure the marker in the \(\{\lvert H\rangle,\lvert V\rangle\}\) basis or the \(\{\lvert +\rangle,\lvert -\rangle\}\) basis? No: in both cases, the signal observer who lacks access to the classical marker outcomes observes the exact same flat marginal probability \(P(0)=1/2\). Extracting either interference fringe requires receiving the classical bit stream of marker outcomes \(\pm\) over an ordinary communication channel.

10.3.2 Reporting both conditional subsets and accidental coincidences

A rigorous quantum-eraser report displays both conditional curves \(P(0\mid +)\) and \(P(0\mid -)\), their relative weights \(P(+)\) and \(P(-)\), the flat unsorted sum \(P(0)\), and the coincidence time window used to pair signal and marker clicks.

10.3.3 Distinguishing coherent basis rotation from deleting classical files

Quantum erasure requires projecting the marker onto the coherent superposition states \(\lvert\pm\rangle\) before the marker decoheres into a macroscopic environment. Merely erasing a computer file after measuring the marker in the \(\{\lvert H\rangle,\lvert V\rangle\}\) basis leaves the signal in an incoherent mixture and never restores fringes.

Visual animation · Section 10.3

No-signalling and coincidence-window discipline

10.3 No-signalling and coincidence-window disciplineCompares ideal prediction against finite efficiency/coherence and audits the experimental assumptions for No-signalling and coincidence-window discipline.1. Ideal Model vs. Imperfect Apparatus
\(V_{\mathrm{cond}} = \eta, \quad P(D_0) = \frac{1}{2}P(D_0\mid +) + \frac{1}{2}P(D_0\mid -) = \frac{1}{2}\)
2. Audit Ledger (10.3)
Fidelity / efficiency \(\eta = \) 0.85
10.3.1 CheckWhy quantum erasure cannot transmit a 10.3.2 CheckReporting both conditional subsets and10.3.3 ScopeDistinguishing coherent basis rotation
Ideal \(= \) 1.00 \(\;\mid\;\) Obs. \(= \) 0.85
Audit status: calibrated

Interactive visual model for 10.3 No-signalling and coincidence-window discipline (Why quantum erasure cannot transmit a faster-than-light signal · Reporting both conditional subsets and accidental coincidences · Distinguishing coherent basis rotation from deleting classical files). Use Animate or Step to follow the physical state evolution and quantitative predictions.

Explore the experiment

Switch among the sorted \(+\) subset, the sorted \(-\) subset, and the unsorted marginal, and sweep \(\phi\) across \([0^\circ,360^\circ]\). Verify that the blue (\(+\)) and dashed gold (\(-\)) curves oscillate \(180^\circ\) out of phase and average to the flat line \(1/2\).

Interactive model · Idealized predictions

Sort the same signal records

Sort the same signal recordsSchematic apparatus; geometry and moving indicators are explanatory, not measured trajectories.
\(\mathrm{BS}_1\)
\(\mathrm{BS}_2\)
Sourceupper armlower arm
\(D_0\)
port 0
\(D_1\)
port 1
Phase \(\phi\)
Path marker
\(|H\rangle / |V\rangle \to |\pm\rangle\)

On a small screen, swipe the diagram to inspect its components. Moving marks illustrate the model; they are not measured trajectories.

Where are the fringes hiding when path information is available? The text and worked examples remain available without JavaScript. Animation starts only when requested; Step gives a motion-free way to follow the model.

A worked example

At phase \(\phi=0^\circ\), calculate the joint probabilities \(P(0,+)\) and \(P(0,-)\) as well as the unconditional probability \(P(0)\) at detector \(D_0\).

At \(\phi=0^\circ\), the conditional probabilities are \(P(0\mid +)=(1+\cos 0^\circ)/2=1\) and \(P(0\mid -)=(1-\cos 0^\circ)/2=0\). Multiplying by the marker probabilities \(P(+)=P(-)=1/2\) gives the joint probabilities \(P(0,+)=P(+)P(0\mid +)=1/2\) and \(P(0,-)=P(-)P(0\mid -)=0\). Summing over the two marker outcomes gives the unconditional marginal \(P(0)=P(0,+)+P(0,-)=1/2+0=1/2\).

Summary

Try it yourself

Problem 10.1

Why can’t the signal observer detect whether the distant marker experimenter chose the \(\{\lvert H\rangle,\lvert V\rangle\}\) basis or the \(\{\lvert +\rangle,\lvert -\rangle\}\) basis without receiving a classical message?

Hint

Sum \(P(0,+)+P(0,-)\) and compare it with \(P(0,H)+P(0,V)\).

Worked solution

In both marker bases, the opposite-phase interference terms \(\pm\frac{1}{4}\cos\phi\) cancel when summed over the marker outcomes, leaving the exact same marginal probability \(P(0)=1/2\) at the signal detector.

Problem 10.2

Suppose a detector measures the marker in the path basis \(\{\lvert H\rangle,\lvert V\rangle\}\) and writes the result to a hard drive, which is later destroyed without being read. Does destroying the hard drive restore interference fringes?

Hint

Ask whether destroying the classical log projects the marker onto \(\lvert\pm\rangle=(\lvert H\rangle\pm\lvert V\rangle)/\sqrt{2}\).

Worked solution

No. Once the marker has been measured in \(\{\lvert H\rangle,\lvert V\rangle\}\) (entangling with macroscopic apparatus degrees of freedom), discarding the classical file simply averages over \(H\) and \(V\), leaving the flat marginal \(P(0)=1/2\).

Problem 10.3

Why must a quantum-eraser publication present both conditional fringes \(P(0\mid +)\) and \(P(0\mid -)\) alongside the unsorted sum rather than plotting only \(P(0\mid +)\)?

Hint

Consider how the two complementary fringes explain the flat unsorted marginal.

Worked solution

Plotting both antiphase subsets \(P(0\mid +)\) and \(P(0\mid -)\) together with their subset rates and unsorted sum verifies that the fringes arise from complementary coincidence sorting rather than from residual single-photon path coherence or biased postselection.

What to carry forward

When path alternatives are entangled with a marker, unconditional fringes vanish while complementary conditional fringes \(P(0\mid\pm)=(1\pm\cos\phi)/2\) remain encoded in joint correlations that can be extracted only by coincidence sorting with a diagonal-basis marker record.

Sources and further exploration

These readings provide historical context, experimental implementations, or deeper derivations. The interactive display above is an educational model, not a reproduction of their measured data.

Search Quantum Experiments

Search all 24 chapters.