A University of Toronto experiment showing that photons can spend a negative amount of time inside a cloud of atoms has been published in Physical Review Letters, clearing peer review more than a year and a half after the result first circulated as a preprint. The work, led by Aephraim Steinberg with first author Daniela Angulo and Griffith University theorist Howard Wiseman, found that photons sent straight through a cloud of cold rubidium can register a negative atomic-excitation time, appearing to exit before they enter, and that the atoms themselves report the same negative figure when probed during transit.
The peer-reviewed paper, published in April, drops the somewhat provocative framing of the September 2024 arXiv draft for a narrower title, with the team again stressing that the effect is fully explained by standard physics, with no faster-than-light signaling involved.
When a photon passes through the rubidium cloud, it can be briefly absorbed, stored as an atomic excitation, then re-emitted. To time that storage without destroying it, the team fired a second, weak laser beam through the cloud and read tiny phase shifts in that beam to track whether the atoms were excited from moment to moment.
Because any single measurement of a quantum system disturbs it, each run was buried in noise, and a clear figure only emerged after averaging roughly 1 million runs across about seven sets of parameters, totaling some 70 hours of data collection. Physics World reported that the measured excitation time for transmitted photons fell as low as minus 0.82 times a baseline value, with that baseline, the excitation time averaged over all photons running between 10 and 20 nanoseconds.
A 1993 experiment that Steinberg also co-authored showed transmitted photons reaching a detector before the center of their own pulse had entered the medium, an effect tied to the long-known optical quantity called group delay. Physicists had largely written off the negative values as an artifact, on the reasoning that only the leading edge of a pulse tends to survive the trip, which makes the survivors look early. Measuring the atoms directly removes that escape hatch, because the excitation stored in the atoms, read out independently, also comes back negative.
"This doesn't mean that we're on the verge of building a time machine or anything like that," Wiseman told Live Science, adding that the result is one more unexpected quantum property rather than a break with known physics. Steinberg said back in 2024 that he saw no path from the work to practical applications, a caveat that still stands even as photonic systems remain central to light-based quantum computing and to schemes for a quantum internet.
The next test, according to Wiseman, targets the scattered photons that never make it through the cloud: theory predicts they carry enough extra positive excitation time to keep the beam's overall average at zero or above, a prediction that hasn’t yet been measured.
Follow Tom's Hardware on Google News, or add us as a preferred source, to get our latest news, analysis, & reviews in your feeds.
