The Ghost in the Machine
The double-slit experiment forms the bedrock of quantum mechanics. When single particles, like photons, are fired at a barrier with two vertical slits, they create an interference pattern on a detector screen behind it. This pattern of light and dark bands is a sign of waves interfering with each other. It suggests each individual photon travels through both slits simultaneously, like a wave, and interferes with itself. However, if a detector is placed at the slits to observe which path the photon takes, the interference pattern vanishes. The act of measuring the photon's path forces it to behave like a definite particle, passing through one slit or the other, not both. This "which-path" information destroys the wave-like behavior.
The quantum eraser experiment takes this a step further. It asks: what if we could record the which-path information and then, after the photon has already hit the detector screen, choose to erase that information? The concept was first proposed by Marlan Scully and Kai Drühl in 1982. The most famous version, the "delayed-choice" quantum eraser, was experimentally realized in 1999 by Yoon-Ho Kim and his colleagues. In this setup, a special crystal creates pairs of entangled photons. One photon, the "signal," is sent through the double slits to a primary detector, while its entangled twin, the "idler," is sent along a different path. The path of the idler photon can be used to reveal the which-path information of the signal photon.
Erasing a Choice Already Made
The timing is the timing. The decision to measure or erase the idler's which-path information happens long after the signal photon has already been detected. In the Kim et al. experiment, the idler's optical path was 2.5 meters longer, meaning its detection happened about 8 nanoseconds after its twin's. When experimenters use the idler to record the signal photon's path, the signal photons show no interference. But when the idler's path is "erased" using a device called a beam splitter—making it impossible to know which path its twin took—the interference pattern for the signal photons reappears.
This result appears to show a future event (erasing information) influencing a past one (the signal photon's behavior at the slits). Research groups in Vienna, particularly at the University of Vienna and the Institute for Quantum Optics and Quantum Information (IQOQI), have been central to pushing the boundaries of these foundational experiments. Anton Zeilinger's group performed a quantum eraser experiment where the choice was causally disconnected, or "space-like separated," from the interference, using a 144 km free-space link. This rules out any signal from the "eraser" influencing the photon's behavior. The experiment does not allow for faster-than-light communication, as the interference pattern only becomes visible when the results from both the signal and idler detectors are correlated after the experiment is complete. It challenges our classical intuition about cause and effect.