Double-slit experiment

• 1. Introduction

  ›

  In 2002, readers of Physics World voted the single-electron double-slit experiment "the most beautiful experiment". The double-slit experiment sends light or matter toward a barrier with two narrow openings and then watches what arrives on a screen behind it. Richard Feynman said it holds "the heart of quantum mechanics". The apparatus is plain, yet its results unsettle anyone who expects a speck of matter to behave like a tiny pebble. How can one particle at a time build a pattern that only waves should make? What happens to that pattern when someone tries to learn which opening a particle used? And what do the answers say about the nature of reality? Physicists have pushed these questions from beams of light to electrons, antimatter and giant molecules, and they still disagree about the replies.

  343556

• 2. Young Against Newton

  ›

  In 1801, Thomas Young presented a paper to the Royal Society titled "On the Theory of Light and Colours". It explained interference phenomena such as Newton's rings as the meeting of waves. Young's first printed statement of what he called his general law appeared in January 1802, in his book A Syllabus of a Course of Lectures on Natural and Experimental Philosophy. He compared two coincident undulations that either cooperate or destroy each other to two musical notes producing the alternating swell and fade of beating in an imperfect unison. In 1803, Young followed up with demonstrations of optical interference that used sunlight, pinholes and cards. These demonstrations played a crucial role in the case for the wave theory of light. They challenged the corpuscular theory proposed by Isaac Newton, which had been the accepted model of light in the 17th and 18th centuries. While discussing the demonstrations, Young described a double-slit experiment. There is some question as to whether he ever actually performed one. The photoelectric effect later showed that, under different circumstances, light can behave as if it were made of discrete particles. Together, the two sets of findings are now called wave-particle duality, and they forced physics beyond its classical rules.

  232512628

• 3. Building the Pattern

  ›

  A coherent light source such as a laser, a plate pierced by two parallel slits, and a screen behind the plate make up the basic experiment. Light passing through the slits interferes with itself and paints alternating bright and dark bands on the screen. Classical particles would not be expected to produce such bands. A single slit makes a useful comparison. If light were a stream of ordinary particles, a single slit should leave a patch matching its own size and shape. Instead the light spreads into a diffraction pattern, and the smaller the slit, the wider the spread. A red laser through one slit shows a bright central band and, if you look carefully, two faint side bands. A more refined apparatus reveals more bands. Yet the screen never records a smooth smear of light. It absorbs light at discrete points, as individual particles, and the bands appear only as a varying density of those hits. In 1909, G. I. Taylor turned the light down until emission and absorption events were mostly non-overlapping. With a low-intensity beam, each particle shows up as a single white dot. Remarkably, the interference pattern still emerges as the dots accumulate one by one. The wave itself is not seen. Its modulus squared gives the probability of detecting a particle at each spot, so every individual landing is inherently probabilistic, which classical mechanics cannot explain. Even entangled photons, in groups of up to four, can show interference patterns.

  291449526

• 4. Matter Takes Its Turn

  ›

  In 1927, Davisson and Germer, and independently George Paget Thomson with his research student Alexander Reid, showed that electrons behave the same way as light. A slit interference experiment with anything other than light, however, had to wait until 1961. That year Claus Jönsson of the University of Tübingen used coherent electron beams and multiple slits. In 1974, the Italian physicists Pier Giorgio Merli, Gian Franco Missiroli and Giulio Pozzi used single electrons from a coherent source with a biprism beam splitter. They showed the statistical build-up of the pattern that quantum theory predicted. In 1991, Carnal and Mlynek ran the classic Young arrangement with metastable helium atoms passing through micrometer-scale slits in gold foil. In 2012, Stefano Frabboni and co-workers sent single electrons through nanofabricated slits about 100 nm wide. A single-electron detector let them watch a double-slit pattern build up. Researchers at the University of Nebraska-Lincoln also did the experiment as Feynman had described it. Their electron gun fired through one or two slits that were 62 nm wide and 4 μm tall, while the team controlled each slit's transmission and monitored single-electron detection events. Experiments of this kind are the basis of modern electron diffraction, microscopy and high resolution imaging. In 2018, a group led by Marco Giammarchi showed single particle interference for antimatter. The work took place in the Positron Laboratory of Rafael Ferragut at the Politecnico di Milano campus in Como, Italy. Size then became the challenge. In 1999, buckyball molecules of 60 carbon atoms passed through a diffraction grating instead of two slits. A buckyball is about 0.7 nm across, nearly half a million times larger than a proton, and large enough to be seen in an electron microscope. Buckminsterfullerene followed in 2001, and in 2011 two molecules of 430 atoms showed the same behavior. In 2013, molecules of 810 atoms and over 10,000 daltons took part. The record rose in 2019 to 2000 atoms, with a total mass of 25,000 daltons. In 2025, scientists at the Massachusetts Institute of Technology released ultra-cold atoms from an optical lattice, where they had formed a Mott insulator. Single photons first scattered coherently from the atoms and formed an optical interference pattern. As the released wavepackets expanded, the pattern faded, confirming incoherent scattering. The result showed wave interference using only quantum atoms and photons.

  316303236

• 5. Watching the Slits

  ›

  Put a detector at each slit, and every detected photon is found passing through just one slit, as a classical particle would. It never seems to go through both, as a wave would. The catch is that the interference pattern then disappears. Niels Bohr read this through the concept of complementarity. In his view, quantum systems are not classical, but measurements can only return classical results. Certain pairs of classical properties are never observed together. For the double slit, this means particle behavior and wave behavior cannot both show at once. Modern quantitative versions allow a continuous tradeoff between the visibility of the fringes and the probability of detecting a particle at a slit. The textbook version of the thought experiment has a practical flaw. A photon cannot be detected without being absorbed, so naive implementations are impossible. Technically feasible realizations were not proposed until the 1970s. An experiment in 1987 showed that partial information about a particle's path could be gained without destroying the interference altogether. This wave-particle trade-off takes the form of an inequality relating the visibility of the pattern to the distinguishability of the paths. Other experiments probed what counts as a choice of slit. In 1967, Pfleegor and Mandel produced two-source interference from two separate lasers. In 1972, an experiment kept only one slit open at any moment. Interference still appeared, provided the path difference allowed the detected photon to have come from either slit, even though the photon density in the system was much less than 1. Researchers have also built the slits out of emission sites. In 2002, a coherent electron wave leaving two closely spaced sites on a needle apex acted as a double slit in a vacuum. In 2017, light-induced field electron emitters let researchers select emission sites optically on a scale of ten nanometers. Switching off one of the two sites made the interference pattern disappear. The Mach-Zehnder interferometer strips the idea down further. Photons travel along just two paths and land on two discrete photodetectors, so simple linear algebra in dimension 2 replaces differential equations. A photon meeting the first beam splitter enters a superposition of both paths. At the second beam splitter the paths interfere, and the right-hand photodetector is hit with probability one and the bottom one with probability zero. Block one path, or detect which path the photon took, and each detector is hit with probability 1/2.

  1350569465

• 6. Erasing and Delaying

  ›

  John Wheeler's delayed-choice experiments take the puzzle one step further. They show that extracting which-path information after a particle has passed the slits can seem to alter its earlier behavior retroactively. Quantum eraser experiments show the reverse. Wave behavior returns when the which-path information is erased or made permanently unavailable. An article in Scientific American described a version that can be done at home. Place a polarizer before each slit, with their axes orthogonal to each other, and the interference pattern vanishes, because the polarizers stamp which-path information onto each beam. Then add a third polarizer in front of the detector, with its axis at 45 degrees to the others. It erases the information, and the pattern reappears. A classical wave account can explain this, and it also covers circular polarizers and single photons. Polarizers built from entangled photon pairs, however, have no classical explanation. In a highly publicized experiment in 2012, researchers claimed to have identified the path each particle took without any adverse effect on the interference pattern. Their particles came not from a point-like source but from a source with two intensity maxima. Svensson and other commentators pointed out that weak measurements do not conflict with the Heisenberg uncertainty principle. Weak measurement followed by post-selection did not allow position and momentum to be measured together for each particle. It measured the average trajectory of particles arriving at different positions. In effect, the experimenters were drawing a statistical map of the whole trajectory landscape.

  6162636413

• 7. The Arithmetic of Fringes

  ›

  Take two slits separated by 0.5 mm, a laser with a wavelength of 0.6 μm, and a screen 1 m away. The fringes then sit 1.2 mm apart. Classical wave theory predicts this with the Huygens-Fresnel principle, in which each point on a wavefront generates a secondary wavelet. The disturbance at any later point is found by adding the wavelets, counting both phase and amplitude. Only intensity can be measured, and it is proportional to the square of the amplitude. If the slits are much narrower than the wavelength, they diffract the light into cylindrical waves. The phase difference between the two waves depends on how much farther one has travelled than the other. Where the path difference is a whole number of wavelengths, the waves are in phase and the intensity peaks. Where the difference is half a wavelength, or one and a half, the waves cancel and the intensity falls to zero. Far from the slits, the fringe spacing equals the wavelength times the viewing distance, divided by the slit separation. If the slit width is appreciable compared with the wavelength, the Fraunhofer diffraction equation applies. It combines a fine cosine structure from the two slits with a coarser envelope from the diffraction of each single slit. Closer to the slits, the Fresnel equation applies. As the observation plane approaches the slits, the area where the two diffracted patterns overlap shrinks, and interference may vanish entirely. The slits need not even be in space. In 2023, an experiment recreated an interference pattern in time. A pump laser pulse hit a screen coated in indium tin oxide, and the Kerr effect turned it from transparent to reflective for around 200 femtoseconds. A probe beam saw that brief change as a slit in time, and two such changes acted as a double slit. Similar results have been obtained classically with water waves. Feynman offered a different mathematics, the path integral formulation. It replaces a single trajectory with a sum over all possible trajectories, each contributing equally. The phase of each contribution is set by the action along its path. Adding the contributions and squaring the magnitude gives the probability of finding the particle at a given position. The differences in accumulated action between paths create the observed fringes. Feynman stressed that this is only a mathematical description, not an attempt to describe a real process that we can measure.

  228990521

• 8. Competing Readings

  ›

  The double-slit experiment is often paired with the Schrödinger's cat thought experiment, because both expose the differences between interpretations of quantum mechanics. The standard interpretation treats the pattern as a wave phenomenon, the interference of two probability amplitudes, one for each slit. Low-intensity runs show the pattern filling in one detection at a time. Any change that lets the apparatus detect a particle at a slit alters the amplitudes, and the interference goes away. This account does not depend on a conscious observer. The term Copenhagen interpretation was apparently coined by Heisenberg during the 1950s for ideas developed in the 1925-1927 period. His choice glossed over his disagreements with Bohr, so no definitive historical statement of the interpretation exists. The views, from Bohr, Heisenberg, Max Born and others, share a few features. Quantum mechanics is intrinsically indeterministic, with probabilities from the Born rule, and measuring an object is irreversible. In a Copenhagen-type view, asking which slit a particle used has no meaning when no detector is present. Carlo Rovelli's relational interpretation locates the effect in the interaction between measuring device and object. An electron observed at a slit carries position information from that interaction, which partly constrains where it lands. Observed only at the screen, it carries no which-path information, so its landing spot follows its probability function, as if it had passed through both slits. The many-worlds interpretation identifies reality with a wavefunction that always evolves following the Schrödinger equation, with no collapses. David Deutsch argues that in each universe the particle takes a specific slit, while interference with particles in other universes creates the fringes. David Wallace, another advocate, writes that in the familiar setup the two paths are not separated enough for a parallel-universes description to make sense. The De Broglie-Bohm theory gives particles precise locations at all times. A particle goes through one slit, while the pilot wave guiding it passes through both. Chris Dewdney first calculated the two-slit trajectories while working with Chris Philippidis and Basil Hiley at Birkbeck College in London. Though deterministic, the theory matches standard quantum mechanics statistically under the quantum equilibrium hypothesis. It is incompatible with Lorentz invariance. A silicone oil droplet bouncing along a liquid surface offers a tabletop echo of the pilot-wave idea. It sloshes the liquid with every bounce, and ripples from past bounces steer its course. This system mimics single-particle diffraction, tunneling, quantized orbits, orbital level splitting, spin and multimodal statistics. No hydrodynamic analog of entanglement has been developed, although optical analogs are possible.

  9553548284