The Signal
Delft Researchers Resolve Nuclear Spin Transitions in One Titanium Atom
A Delft-led team used scanning tunnelling microscopy and electron–nuclear double resonance to resolve transitions in the nucleus of a single titanium-47 atom. Measurements across magnetic-field strengths revealed how hyperfine and quadrupol
Quantum.world Team · Written with AI · fact-checked ·
A single titanium-47 atom can reveal nuclear magnetic resonance transitions, a Delft-led team reports in a paper published on 24 October in Nature Communications. The researchers used scanning tunnelling microscopy (STM) and an electron–nuclear double-resonance scheme to address the atom’s nucleus, whose spin is 5/2. Their measurements resolve nuclear-spin transitions at an individual surface atom and help characterize the interactions that shape its spectrum.
A nucleus read through its electron
“In this work, using scanning tunnelling microscopy, we reveal nuclear magnetic resonance on a single on-surface atom.”
The authors write that the experiment uses STM to study a single atom on a surface. The team worked with titanium atoms placed at oxygen binding sites on a layer of magnesium oxide grown on silver. The titanium atom carries an electron spin as well as a nuclear spin. The instrument’s spin-polarized tunnelling current provides a signal that changes with the electron’s state, which is linked to the nucleus through their interaction.
For the double-resonance measurement, the researchers applied radio-frequency signals at two frequency ranges: a gigahertz signal to drive electron-spin resonance and a megahertz signal to drive nuclear transitions. They used the relative heights of electron-spin resonance peaks to track the populations of nuclear-spin states. When the nuclear-frequency signal matched a transition, the measured populations changed, giving the researchers a way to identify nuclear resonance through the electron’s response.
The method addresses a measurement challenge: nuclear spins are comparatively isolated from their surroundings, a property associated with long-lived states, but that isolation also makes them harder to read and control. Earlier STM measurements had observed nuclear-resonance-like signals when electron and nuclear spins were strongly hybridized, meaning their states were mixed. The new study’s double-resonance approach separates nuclear driving from that hybridization-dependent signal.
Hyperfine and quadrupole effects shape the spectrum
The team combined measurements with numerical analysis based on an effective spin Hamiltonian. This analysis identified hyperfine coupling between the electron and nucleus and quadrupole coupling associated with the nuclear spin. The hyperfine interaction splits the electron-spin resonance according to the nuclear-spin state. The quadrupole interaction further separates some nuclear transitions, allowing the researchers to distinguish multiple resonances rather than one undifferentiated signal.
In measurements of the titanium-47 atom, the team observed nuclear transitions near 50 and 85 megahertz for two states, and additional features near 65 and 75 megahertz for another pair. Fits to the measured spectrum gave a hyperfine coupling of 132.4 ± 0.4 megahertz and a quadrupole coupling of −2.9 ± 0.8 megahertz. The authors note that the latter differs modestly from earlier reports and suggest the electric-field gradient, which can depend on the STM tip shape, as a possible explanation.
The researchers also varied the magnetic field from 200 to 1,400 millitesla. Across that range, they continued to observe nuclear transitions while electron–nuclear hybridization decreased by about a factor of ten. They extracted a nuclear g-factor of 0.37 ± 0.04, slightly different from the value measured in bulk titanium-47. Changes in signal strength at higher fields were also affected by limits on the radio-frequency amplitude delivered through the experimental setup.
Addressing the nucleus is not yet coherent control
Nuclear spins interest quantum-information researchers because their relative isolation can support long-lived states. The same property can make it difficult to address them. The study reports that the team could address the nucleus in the weak-hybridization limit, which the authors identify as a requirement for direct control in a long-lifetime regime. The experiment resolves and drives nuclear transitions, but does not demonstrate coherent quantum operations on the nucleus.
The result follows earlier work by overlapping authors on reading a single surface-bound titanium nucleus. In that 2025 study, the team demonstrated single-shot readout of a titanium-49 nuclear spin and measured an intrinsic lifetime of 5.3 ± 0.5 seconds for a selected nuclear state. The newer paper tackles a different step: resolving nuclear resonance transitions and examining the couplings that govern them.
Together, the studies move from detecting a nuclear state to addressing its transitions at atomic scale. The new measurements provide a way to examine how an individual atom’s surroundings affect its nuclear spin. Whether this approach can be used to perform coherent operations while preserving long-lived behavior remains a further experimental question.