Spatially resolved measurements of coherent waves are often read as direct maps of a system's eigenmodes. In magnonics, where spin waves are seen as information carriers, this interpretation matters critically: you need to know not only the wavelength and amplitude, but also the phase, the direction, and the scattering into competing channels. In the near field, however, where the probe senses the total microwave field rather than the population of a single mode, this simple picture breaks down. Neighboring papers in the archive deal with magnon traps, diamond sensors, and optical imaging of magnons at ultralow temperatures. All of them touch in one way or another on the delicate diagnostics of spin waves, but this is the one that exposes a fundamental catch in the interpretation.
The experiment was performed on a yttrium iron garnet (YIG) film 140 nm thick with record-low damping (Gilbert parameter 8×10⁻⁵). Spin waves were excited by a gold microstrip antenna 10 µm wide, and the local microwave field was measured by a single nitrogen-vacancy center in the diamond probe of an atomic force microscope. An external field of 15.9 mT was applied parallel to the antenna, creating a Damon-Eshbach geometry in which the waves run perpendicular to the magnetization. By varying the probe-sample distance from 200 to 50 nm, the authors tuned the sensitivity to short-wavelength components. For the interpretation they built an analytical model of near-field diffraction, decomposing the total field into three contributions: the direct field of the antenna, the field of the primary wave, and the field of the scattered waves. The contrast of optically detected magnetic resonance is proportional to the square of the total field, which generates three interference terms, including a cross magnon-magnon one.
At a distance of 200 nm the Fourier image shows only a single pair of peaks, corresponding to the wave allowed by the dispersion. At 50 nm a "butterfly" appears, a two-dimensional isofrequency surface, but with anomalous enhancement near zero wave vector and a steeper bending of the contours than theory predicts. The model reproduces these features only when the cross term between the primary and scattered waves is included. This term produces a spatial modulation with a difference vector $k_{exc}$ − $k_{scat}$, which can go to zero even though both waves carry finite momentum. Numerical work shows that the term dominates when the field of the scattered waves exceeds the direct antenna field, which is typical of the Damon-Eshbach geometry. In the backward volume wave geometry, where excitation is inefficient, the spurious signals vanish. Additional experiments at a frequency below ferromagnetic resonance but in resonance with the backward volume branch revealed scattered waves even without a primary wave; their source turned out to be roughness at the antenna edge. Simulations with a corrugated antenna edge showed that even 1% of the width (100 nm) is enough for noticeable scattering, while 4% (400 nm) markedly redistributes power into high-momentum modes.
The work shifts the emphasis from "mapping modes" to "interferometry of fields" in near-field magnetometry. Unlike neighboring studies that use nitrogen-vacancy centers to image waves or examine magnons optically, this one for the first time systematically decomposes the contribution of magnon-magnon interference and shows that it can completely dominate the spectrum. This is not merely a warning but a quantitative criterion: spurious zero components appear when the field of the scattered waves is comparable to the antenna field. For the design of magnonic circuits this means that lithographic defects are unavoidable and must be taken into account at the design stage rather than written off as noise.
The method can be extended to other geometries and types of transducers, especially to grounded coplanar lines, where the direct antenna field is suppressed and the cross term comes to the fore. It is promising to use nitrogen-vacancy centers not only as passive sensors but also as addressable quantum objects: standing scattering waves create sub-50-nm field gradients suitable for local control of spin states. A path also opens toward tomography of scattering sources: by varying the probe height and the frequency, one can reconstruct a map of defects from their caustics.
The results matter for magnonic logic and wave computing, where phase coherence and spectral purity are critical. They also refine the calibration protocols of scanning quantum sensors in magnetism.
The next step is to separate experimentally the contributions of film defects and antenna roughness, for instance by comparing samples with different densities of growth pits. It is also worth testing the prediction that the cross term dominates for volume waves when grounded transducers are used.
In magnonics, the nature of excess losses and phase noise in the linear regime remains an open question. This work links them to unavoidable multimode excitation caused by defects: energy is spread over many modes already below the nonlinearity threshold. Understanding this mechanism is necessary for building stable magnonic interferometers and delay lines.
This section is addressed to the author of the paper and their colleagues. Neighbouring works were found by meaning-based search across our archive, not by word overlap. We do not review: we show the field around the work.
What the work does
This work focuses on scanning magnetometry using nitrogen-vacancy centers in diamond to image spin waves in a yttrium iron garnet film. The authors experimentally discovered spurious signals with zero wave vector and explained them with an analytical near-field diffraction model that accounts for interference between the primary and scattered waves.
Where its strength is
A fundamental artifact in the interpretation of near-field measurements has been uncovered: the cross term between the primary and scattered waves can produce a false zero momentum. A method for identifying scattering is proposed, which is important for diagnosing nanoscale magnonic devices.
Where it could go next
Worth trying
Why this may matter
The work warns against naive interpretation of spectra in near-field magnetometry, which is critically important for the development of magnonics as an information-processing platform, where phase coherence and precise knowledge of wave vectors determine device functionality.
Nearby works in our archive
The archive is sparse around this work: the closest review is only 0.53, against a usual 0.64. It looks like few have drilled nearby.
The YIG film used in the work is so perfect that its damping parameter is a hundred times smaller than that of typical ferrites. That is precisely what made it possible to see the subtle interference effects that losses mask in ordinary materials.
The idea of wave computing on spin waves echoes the "magnonic logic" of Greg Egan's novel Schild's Ladder, where computation is performed on quantum excitations in a solid. There, as here, the key problem is control over parasitic modes.