Double-slit photoelectron interference in strong-field ionization of the neon dimer

Imagine taking Young's double-slit experiment, shrinking it down to the size of a molecule, and letting an intense laser do the work of pulling an electron free. That's the neon dimer: two neon atoms bound loosely together, acting like a pair of slits for an electron wave in a strong field. The trick that makes it an interferometer is indistinguishability. If the electron could have come from either atom without leaving which-way clues, the two quantum paths add coherently. There's a dial you can turn that sets the phase of that interference: the parity of the molecular orbital the electron comes from. In neon, a gerade orbital starts the two paths in step, while an ungerade orbital starts them out of step by half a cycle. Flip the orbital, and you flip the fringes. Kunitski and colleagues built that interferometer in the lab by crafting a clean, cold source of neon dimers and then measuring everything that flew out, both ions and electrons. They created the dimers in a supersonic beam cooled to about 60 kelvin, tuned the backing pressure to 3 bar to hit the sweet spot for dimer formation, and then passed the beam through a grating that is 100 nanometres in size. That matter-wave selection boosted the dimer fraction from roughly 2 percent to about 20 percent relative to the monomer, which is a significant gain for signal without contaminating the physics with heavier clusters. They then hit the dimers with a short, strong pulse—40 femtoseconds long at 780 nanometres. Two intensity settings allowed them to test different regimes while staying in strong-field ionization: around 7.3 times ten to the fourteenth watts per square centimeter for circular polarization and about 1.2 times ten to the fifteenth watts per square centimeter for linear. In both cases, they are in the tunneling regime, where the laser field bends the Coulomb barrier enough for the electron to slip through during the pulse. To see interference, you need the electron's momentum in the molecule's own frame, not just in the lab. That's where coincidence detection comes in. In a Cold Target Recoil Ion Momentum Spectrometer, known as COLTRIMS, they measured the three-dimensional momenta of the outgoing electron and the ionic fragments simultaneously. Modest guiding fields—16 volts per centimeter for the circular case and 23 volts per centimeter for the linear one—shepherded the ions to a position and time sensitive detector, while magnetic fields of 12.5 and 9 gauss helped collect the electrons nearly isotropically. The payoff of all that hardware is simple to say and hard to achieve: for each electron, you also know the direction the dimer was pointing when it broke apart. Now, the dimer doesn't just ionize and stop. It breaks. The way it breaks tells you which orbital the electron came from. The team saw two clean kinetic energy release peaks when neon dimer cations dissociated into neon cations and neutral neon: one near 0.15 electronvolts and another around 1.3 electronvolts. Gating on those energies isolates two distinct pathways. The low-energy peak corresponds to direct dissociation along the one-half g potential, which maps back to ionization from the gerade two p sigma g orbital. The high-energy peak tags an indirect route: after initially moving on the one-half u curve, the ion absorbs a photon and switches to the one-half g state, linking the channel to the ungerade two p sigma u orbital. In momentum space, they reinforced that separation by selecting ion momenta between about 3.5 and 17 atomic units for the direct branch and between 37 and 46 atomic units for the indirect one. With those gates, the breakup channel becomes a symmetry filter. So, what do the electrons look like when you view them in the molecular frame? For circular polarization, there's a background complication: tunneling ionization from atoms produces a doughnut-shaped momentum distribution. To get a clean look at interference, the team divided the dimer spectrum by the monomer spectrum, which scrubs away that intrinsic weighting. After that normalization, a set of bright and dark stripes appears, perpendicular to the internuclear axis. And here's the flip promised at the start. In the direct, gerade channel, there’s a minimum at the center of the pattern along the axis perpendicular to the bond when the component of the electron momentum along the bond is zero. In the indirect, ungerade channel, that same spot becomes a maximum. Same molecule, same laser, same geometry—just opposite parity, and the fringes reverse. That inversion is the headline for circular light, but it isn't a quirk of circular driving. With linearly polarized pulses, the story is the same, once you hold the geometry steady. By postselecting events where the dimer axis lay within roughly fifteen degrees of the field direction, the two-center stripes pop out again—even for low-energy electrons, with momenta below about half an atomic unit, where you might worry that Coulomb forces would smear everything out. They don't. The pattern persists. Behind those pictures sits a simple, powerful model. Imagine two atomic electron waves launched from positions separated by the internuclear vector R. In momentum space, that separation produces phase factors of plus or minus k dot R over two. Add those two waves and include an extra phase shift, delta phi, that encodes the orbital parity—zero for gerade, pi for ungerade. Square the result to get an intensity, and you have an interference term that's a cosine whose argument is the geometric path difference plus that initial phase. If you prefer it in wavelength language, the fringe modulation depends on the ratio of the projected slit spacing, the magnitude of R times the sine of the emission angle, to the electron's de Broglie wavelength, which is two pi divided by the magnitude of k. Change delta phi from zero to pi, and the whole pattern shifts by half a period. That's exactly what the data do. Real molecules aren't locked to one orientation in the lab, so the model takes that into account by averaging over a set of fixed angles between the dimer axis and the field—eight angles spanning zero to forty-five degrees. After projecting everything onto the polarization plane and weighting geometrically, the calculated distributions reproduce the main features of the data. A more complete treatment, a time-dependent Schrödinger equation propagation in a single-active-electron picture, lands in the same place: the two-center superposition with the parity phase already captures the physics you need to see the fringes. There's more in the spacing than just a pretty pattern. The distance between fringes tells you about the bond length because the phase difference between the two centers grows with k dot R. If you gate on different kinetic energy releases—remember those two peaks at 0.15 and 1.3 electronvolts—you are effectively selecting different parts of the ionic potential curves and, with them, different internuclear distances at breakup. In the experiment, extracting R from the interference pattern returned values in the ballpark of 2.8 to 3.25 angstroms, depending on channel and momentum slice. That's consistently a bit shorter than the standard ground-state neon dimer separation near 3.337 angstroms measured from spectroscopy. Both moves the same direction: the fringes spread as the bond lengthens and squeeze together as it shortens. The offset between the two readouts—interference versus ion momentum mapping—makes sense in a strong field. The laser dresses the potential energy curves, pushing the one-half g surface downward and the one-half u surface upward relative to their field-free shapes, and the Franck–Condon region you're sampling during ionization isn't quite the field-free one either. Not every fringe is perfect black-to-bright. The contrast is finite, and that, too, is a window into the molecular environment. Part of the loss of visibility likely comes from the electron feeling the Coulomb pull of the neighboring neutral atom as it leaves; that extra phase wash flattens the pattern a bit. Add in mundane effects—variation of intensity across the laser focus and a spread in initial bond lengths from the vibrational ground state—and you get the slightly muted, but still unmistakable, stripes they observed. A subtle test of the two-path picture is what happens when you don't pick a channel at all. If you mix events from the two dissociation energies, you mix delta phi equal to zero with delta phi equal to pi. Those two patterns are out of phase, so when you add them, the oscillations cancel and the interference washes away. That's exactly what the combined spectra show. The interference is conditional on indistinguishability, and postselection is your way of turning which-way information on or off. Step back, and the throughline is crisp. Strong-field ionization does not destroy two-center interference in a molecule as simple as neon dimer. It preserves it—and, thanks to the breakup dynamics, it even labels it. Choose the dissociation path, and you choose the orbital symmetry. Choose the symmetry, and you shift the phase. The laser's polarization doesn't change that logic. Circular or linear, the molecular-frame photoelectron distributions carry a parity-tagged interference that can be read out once you normalize away the atomic ionization background. As a method, this stitches together a source, a drive, and a detector into a compact interferometer that can read orbital parity and measure distance at the same time. The equation behind it is simple enough to say in a sentence, and the experimental control—cold dimers selected by a matter-wave grating, a short pulse to ionize, and coincidence detection to lock electron to bond axis—makes it practical. The interpretation isn't just hand-waving. A time-dependent single-electron calculation that coherently adds contributions from the two centers and integrates over orientations lands squarely on the observed maps. Where does this go next? Carefully, because the paper earns its credit by showing, not speculating. Still, the ingredients on the table—postselection as a parity switch, fringe spacing as a distance meter, and the ability to gate on different parts of a potential curve by kinetic energy release—look like the beginnings of time-resolved molecular imaging. Add a pump pulse to nudge the bond, then probe a fraction of a cycle later, and you could watch the fringe spacing breathe. Do the same across symmetry-selected channels, and you would separate how gerade and ungerade pathways light up as the molecule moves. Even without those extras, the immediate message is powerful. A humble neon dimer can act as a tunable, two-slit interferometer for electrons in a strong laser field. With a symmetry tag from its own breakup, it lets you flip the interference phase on demand and read both the parity of the orbital that fed the continuum and the distance between the two neon atoms. We're used to thinking of strong fields as blunt hammers. Here, they're carving out a ruler and a phase knob at the scale of a chemical bond.