Seeing Wave and Particle Together: A Quantum Microscope’s First Image
29 May 2026, Yanjiang
A quantum-light microscope images a single polariton’s self-interference fringes in MoS₂, directly visualizing wave-particle duality at the nanoscale for the first time.
What if you could watch a single quantum of light behave like a wave one moment and a particle the next — and even see it interfere with itself? That is the question a team led by D. N. Basov at Columbia University has not merely asked, but answered. In a preprint (arXiv:2605.28987) they unveil a quantum-light microscope that, for the first time, photographs a hybrid light-matter quasiparticle in the very act of its wave‑particle duality — directly in space, at a resolution far below the wavelength of light.
It is the kind of experiment that reopens a century‑old argument. Ever since Einstein insisted that light comes in lumps and de Broglie whispered that everything waves, we have known — in the abstract — that quantum objects inhabit both identities at once. But to snap a picture of a single quasiparticle’s interference as it moves through a crystal, and to do so with a camera sensitive enough to work with a trickle of deliberately faint photons, is something else entirely. This is no philosophical cartoon. It is a direct visualization of the central strangeness of quantum mechanics.
The problem they had to solve was double‑edged. First, quantum correlations inside solids — the kind that give rise to superconductivity, magnetism, and exotic topology — live at length scales of atoms and time scales of femtoseconds, far below what any conventional microscope can resolve. Second, the most exquisitely controlled quantum light, the sort produced by spontaneous parametric down‑conversion, arrives in pairs but at an intensity so feeble that most of it is lost before it can even touch the sample. To make matters worse, the near‑field interaction that carries the light through the scanning tip into the material is notoriously inefficient — like trying to hear a whisper by pressing your ear to a door while a hurricane howls on the other side.
Basov and his colleagues — Michael Dapolito, Matthew Fu, Fuyang Tay, and a large team spanning physics, chemistry, and engineering departments at Columbia and Stony Brook — found a way to turn that weakness into a signature. Their new instrument, which they call a quantum-light scattering‑type scanning near‑field optical microscope, or q‑SNOM, exploits the very property that makes entangled photons seem so useless for imaging: their pairwise correlation.
Here is the trick. One photon from an entangled pair is sent to a sharp metallic tip hovering just above a flake of molybdenum disulfide, a van der Waals semiconductor. There it can either bounce off uselessly, or — with just the right tip‑sample coupling — launch a single polariton, a hybrid particle that is part light, part material vibration. The other photon, the “idler,” travels a separate path and serves as a kind of clock. Because the two photons were born together, the arrival of the idler tells you exactly when its twin interacted with the material, even if the signal photon itself is buried in noise. The team could then reconstruct, photon by correlated photon, how that lone polariton moved.
What they saw was unmistakable, and eerily beautiful. As the launched polariton spread out from the tip, it encountered the edges of the MoS₂ flake and reflected, creating a standing‑wave pattern of bright and dark fringes — a self‑interference pattern. This is the very fingerprint of wave behavior, but it was produced by exactly one quantum of the hybrid excitation at a time. The fringes emerged with a clarity that had previously been the exclusive province of classical light, even though the photon flux was a modest twenty‑five million photons per second. What matters, it turns out, is not the number of photons but their quantum coherence — each photon’s ability to interfere with itself.
This is not a metaphor — but let me offer one anyway, because the aridity of a pure measurement can sometimes obscure its magnitude. Think of a dancer on a stage. If you photograph her a thousand times in rapid succession, you can trace her trajectory and call it a wave. But what if you could take a single snapshot and already see the ghost of all her possible positions, spread out like a ripple? The q‑SNOM is that single snapshot. It sees the polariton not as a dot, but as an interference pattern — and it does so without averaging, without statistics, but through the brute logic of quantum superposition itself. (E001: Unlike a dancer, of course, a polariton is genuinely in all those positions at once — not just ungraspably fast, but fundamentally.)
The philosophical payoff is not just that wave‑particle duality has been imaged at the nanoscale. It is that the imaging was made possible by the same quantum effect that creates the confusion in the first place. Entanglement, the spooky connection that so troubled Einstein, was the bridge that allowed the team to listen to a single excitation amid a cacophony. They stood on the shoulders of paradox to see the thing that creates paradox.
But honest dialectic cannot stop at admiration. A critical reader will ask: is this truly a new microscope, or a demonstration of principle, built on a single material? The polaritons in MoS₂ are well understood; the interference fringes confirm what we already knew. And the setup requires a fragile coincidence‑counting scheme that cannot yet be scaled to arbitrary samples. Those cautions are fair — the instrument is, in many ways, a proof of concept.
Yet the question it asks is not “what does the instrument do today?” but “what does it enable tomorrow?” The team also demonstrates something pointing toward a genuinely new capability: polaritonic time‑of‑flight metrology. By measuring the temporal delay between the detection of the idler photon and the arrival of the signal photon after it has propagated through the material, they can clock the quasiparticle’s journey with femtosecond precision. This is entirely different from standard pump‑probe experiments; here, the entangled photon pair itself plays the role of both excitation and ultrafast stopwatch. It turns the microscope from a static camera into a strobe that can watch excitations move, collide, and perhaps even entangle in real time.
And that, finally, is where the truly unsettling possibility lies. If you can launch one polariton with a known birth time and watch its interference in space, what stops you from launching two, and then observing whether their wave functions combine in ways that classical physics cannot simulate? The q‑SNOM is arguably the first tool that could directly image quantum coherence between engineered solid‑state excitations — not indirectly through transport measurements or spectroscopic dips, but by photographing the fringes of mutual entanglement.
The tension between what this preprint shows and what it implies is itself worthy of the dialectical method. On one hand, it is a careful technological achievement: a clever use of coincidence detection to overcome the signal‑to‑noise ratio that has long crippled quantum‑illuminated near‑field optics. On the other hand, it is a provocation. What does it mean to watch a quantum system being quantum, at the scale of a single quasiparticle, with your own eyes — or at least with a detector whose every count is a certificate of spooky action? We are accustomed to talking about the measurement problem as if it were a philosophical footnote. This instrument makes it possible, perhaps, to be genuinely unsettled again by the fact that looking at something changes what it does.
What the team has built is not just a better microscope. It is a camera that sees the quantumness itself — the inherent ambiguity of a photon that is both here and there, of a hybrid particle that is both a propagating ripple and a localized excitation, of a world where the distinction between wave and particle is finally, irrevocably, a matter not of theory but of what you choose to record. And if that doesn’t make you question the solidity of your own reality, then nothing will.
The Basil‑Dapolito instrument will not, of course, answer the ancient riddle of the observer. It may, however, give us the images that remind us the riddle is still alive — and that the gap between our equations and our experience is not a defect, but the very signature that we are on the right track.
— Yanjiang
Yanjiang is the founding editor of LoomSci.com, specializing in physics and science communication.
References
- Michael Dapolito et al., Quantum Light Nano-Imaging, arXiv:2605.28987