Magnifying Atomic Images
To gain in-depth understanding of quantum matter, researchers need to probe it at the microscopic level. Ultracold atoms—ensembles of atoms cooled to near absolute zero—offer an exceptionally clean and controllable platform for exploring collective quantum phenomena. Over the past two decades, researchers have sought to take in situ “snapshots” in which every single atom is individually resolved in position and, when needed, in spin. Recent advances have brought this vision to life and have significantly accelerated our understanding of collective quantum behaviors. Yet an important challenge remains: In a number of situations, the typical spacing between particles is smaller than the resolution limit of conventional optical imaging. Now Selim Jochim and his group at Heidelberg University in Germany have introduced a method to overcome this barrier by making the system “self-magnify” before imaging [1]. The approach could be a boon to the use of ultracold atoms as simulators of complex quantum phenomena, from supersolidity to topologically correlated states.
Resolving each atom of a quantum gas does not mean imaging at the true atomic scale—typically a few angstroms, which would be a daunting if not impossible task—but rather at a scale smaller than the typical interparticle spacing. In many ultracold-atom systems, this spacing is about 0.5–1 µm, which is a fortunate range for two reasons. First, the resolution needed to probe such scales can be achieved with visible or near-infrared light, the wavelength range used in everyday cameras and biological microscopy. Second, the atoms in quantum simulators have optical transitions in this range, meaning that lasers change their state and make them fluoresce (or emit light). The matching of interparticle distance to atomic-transition wavelengths and optical resolution is no accident but stems from a deliberate strategy for achieving microscopic access to these systems.
A major breakthrough in studying atomic many-body systems came with the advent of quantum gas microscopes in optical lattices [2–4]—periodic “egg boxes” of light that trap atoms in well-defined sites, mimicking electrons in a crystal. More recently, similar techniques have been adapted to atoms in the continuum (see Viewpoint: A Glimpse at the Quantum Behavior of a Uniform Gas), a situation in which particles move freely rather than stepping between lattice sites. However, this favorable matching of atom spacing and optical resolution is not always possible. Imaging constraints may limit resolution. For example, geometric constraints of the apparatus may prevent using optimal optical systems, or the system itself may be too dense, with atoms separated by much less than a micron.
In this context, Jochim and colleagues present a striking advance: A technique that prompts the system to self-magnify by a factor of more than 50 before imaging, thereby overcoming both the optical-resolution limits of the imaging system and high-density challenges. This allows them to capture single-atom-resolved images of the magnified system.
The team used ultracold lithium-6 atoms confined to a plane. Within that plane, the atoms were held in a harmonic trap—a perfectly parabolic potential well. To build intuition, let us momentarily simplify the picture by stepping out of the quantum world and into the classical one and by considering motion in just a single spatial dimension. In this classical analogy, a particle in a harmonic well executes perfectly regular, undamped oscillations between two turning points—much like a frictionless swing moving back and forth.
To make the system “self-magnify,” the researchers used a precisely timed sequence of experimental steps. First, using a technique that is common in ultracold-atom experiments, they abruptly “turned off” any interactions between particles such that the atoms would not perturb each other during the following steps. Then, they let the atoms evolve freely in a relatively tight trap for a quarter of the oscillation period—equivalent, in the swing analogy, to traveling from the highest point down to the lowest point. At that moment, they transferred the atoms into a much shallower harmonic trap—as if the swing’s chain were suddenly lengthened—and let the atoms evolve for a further controlled time. In an ideal harmonic system, this sequence produces a particle distribution with an exact scaling of the particles’ positions relative to the trap center (Fig. 1). The magnification factor is determined by the oscillation periods of the two traps and by the final evolution time. At the end of this sequence, an image of the magnified spatial distribution can be taken, resolving each atom individually with established optical techniques, which also determines its spin.
Now, returning to the quantum realm, an important difference emerges: In quantum mechanics, particles do not trace well-defined trajectories from well-defined initial positions. Instead, the experimental sequence scales up the entire wave function. The final image then represents a realization of the scaled quantum state. One can show that this realization has the same scaling factor predicted by the classical model.
While the underlying mathematics is straightforward in both the classical and quantum cases, implementing the method in the lab demands great care. The system size at each instant of the sequence must be chosen to match optical limitations, the harmonic potential must be nearly perfect, and the timing of the trap modification must be precisely controlled. Jochim and colleagues demonstrate this level of control by applying their method in progressively more complex settings: from a single atom, to six noninteracting atoms, to one pair of interacting atoms (one with spin-up and the other with spin-down), and finally to six spin-up atoms interacting with six spin-down atoms. In doing so, the researchers also provide a benchmarking protocol to certify the reliability of the magnifier.
This approach can, in principle, be applied to systems containing many particles. Particularly compelling applications lie in the study of quantum systems exhibiting fascinating phenomena such as supersolidity [5, 6] and the emergence of self-bound quantum droplets [7]. These states have been observed in several experiments in recent years and are now being explored with analogous systems made of ultracold dipolar molecules [8], an increasingly attractive platform for quantum simulation [9]. The wave-function magnifier introduced here could be essential for accessing the microscopic organization of such states, where the interparticle spacing is on the order of just 0.1 µm.
References
- S. Brandstetter et al., “Magnifying the wave function of interacting fermionic atoms,” Phys. Rev. Lett. 135, 103401 (2025).
- W. S. Bakr et al., “A quantum gas microscope for detecting single atoms in a Hubbard-regime optical lattice,” Nature 462, 74 (2009).
- J. F. Sherson et al., “Single-atom-resolved fluorescence imaging of an atomic Mott insulator,” Nature 467, 68 (2010).
- C. Gross and W. S. Bakr, “Quantum gas microscopy for single atom and spin detection,” Nat. Phys. 17, 1316 (2021).
- L. Chomaz et al., “Dipolar physics: A review of experiments with magnetic quantum gases,” Rep. Prog. Phys. 86, 026401 (2022).
- A. Recati and S. Stringari, “Supersolidity in ultracold dipolar gases,” Nat. Rev. Phys. 5, 735 (2023).
- C. R. Cabrera et al., “Quantum liquid droplets in a mixture of Bose-Einstein condensates,” Science 359, 301 (2018).
- S. Zhang et al., “Observation of self-bound droplets of ultracold dipolar molecules,” arXiv:2507.15208.
- S. L. Cornish et al., “Quantum computation and quantum simulation with ultracold molecules,” Nat. Phys. 20, 730 (2024).