[Paper Review] Criticism of "Asking Photons Where They Have Been"

MMarcin Wieśniak|arXiv (Cornell University)|Jul 7, 2014

Quantum Mechanics and Applications3 citations

TL;DR

This paper challenges the interpretation of Danan et al.'s experiment claiming photons traverse multiple paths simultaneously, arguing that the observed interference patterns—detected via Fourier analysis of time-resolved quad-cell signals—can be fully explained by classical wave optics and mechanical instability, not the two-state vector formalism (TSVF). The authors show that spurious spectral peaks at mirror E, not predicted by TSVF, arise from acoustic vibrations and path-length sensitivity, undermining the claim of quantum multilocality.

ABSTRACT

I stress that [PRL 111,240402(2013)] contains no result that need to be explained by so-called Two State Vector Formalism, and closely inspected data are in disagreement with claims of Danan, Farfurnik, Bar-Ad, and Vaidman

Motivation & Objective

• To challenge the claim that photons in a Mach-Zehnder interferometer traverse multiple paths simultaneously, as proposed by Danan et al. using path markers and Fourier analysis.
• To demonstrate that mechanical instability from mirror oscillations disrupts interference patterns, invalidating the experimental setup’s reliability.
• To argue that observed spectral peaks at frequencies f_E, f_A, and f_B are better explained by classical wave optics and acoustic eigenmodes than by the two-state vector formalism (TSVF).
• To show that the TSVF interpretation fails to account for the dominance of f_E in Figure 2(c), where the lower arm is blocked.
• To uphold Ockham’s razor by rejecting TSVF as unnecessary, given that all data can be explained by established physical theories.

Proposed method

• Fourier analysis of time-resolved quad-cell detector (QCD) signals to extract frequency components corresponding to oscillating mirrors (f_A, f_B, f_C, f_E).
• Modeling mechanical displacement of mirrors due to piezoelectric actuation: 10 nm amplitude at 1 cm from axis causes 100 nm beam spot displacement over 10 cm path length.
• Analyzing spectral structure in Figure 2(c), where the lower arm is blocked, to identify non-uniform peaks not predicted by TSVF.
• Comparing TSVF predictions—overlap of forward and backward evolving wavefunctions on mirrors A, B, and C—with observed spectral dominance at f_E.
• Assessing the impact of acoustic eigenmodes on the system, particularly the 280–310 Hz peaks observed in the power spectrum.
• Evaluating visibility of interference fringes (95% reported by authors) to assess experimental stability and alignment sensitivity.

Experimental results

Research questions

• RQ1 Can the observed spectral peaks in Danan et al.'s experiment be explained by classical wave optics and mechanical instability, rather than quantum multilocality?
• RQ2 Why is the peak at f_E dominant in Figure 2(c), despite TSVF predicting no overlap on mirror E?
• RQ3 To what extent do mirror oscillations disrupt interference patterns, and does this invalidate the experiment’s conclusions?
• RQ4 Are the spectral features in the blocked-arm scenario (Figure 2(c)) consistent with the TSVF interpretation or with acoustic vibrations?
• RQ5 Does the data support the claim that photons were in three locations simultaneously, or is this an artifact of experimental noise and alignment?

Key findings

• The observed spectral peaks at f_A, f_B, and f_C in the unblocked case are consistent with the TSVF interpretation, but only if mechanical instability is ignored.
• In the blocked-arm scenario (Figure 2(c)), the dominant spectral peak at ~280–310 Hz corresponds to f_E, the farthest mirror from the detector, contradicting TSVF predictions.
• The mechanical displacement caused by mirror oscillation (10 nm amplitude over 10 cm path) leads to a 100 nm beam spot shift, sufficient to disrupt interference and alter fringe visibility.
• The 95% fringe visibility reported by Danan et al. is insufficient to rule out mechanical perturbations as the source of spectral features.
• Acoustic eigenmodes of the setup, particularly at f_E, explain the observed spectral structure in Figure 2(c), not quantum path overlap as claimed by TSVF.
• No experimental evidence supports the existence of quantum multilocality; all results are explainable by classical optics and mechanical noise, rendering TSVF unnecessary.