3
[2026-08-28] · src Lourette, Jarmola, Chathanathil, Acosta, Birdwell, Blumler, Budker, Carrasco, Ivanov, Kolkowitz, Malinovsky; Phys. Rev. Applied (2026); arXiv:2601.00157
Nitrogen-vacancy centers in diamond make compact, chip-scale clocks, but their frequency depends on temperature through the same strong spin-lattice coupling that makes them good sensors, so thermal drift has been the limiting noise source. This group built a composite frequency reference from two transitions in the same NV center - the electronic zero-field splitting D, which is temperature-sensitive, and the ambient nuclear quadrupole splitting of the center's own 14N nucleus, which responds to temperature differently - and combined the two to cancel the first-order temperature dependence rather than stabilizing the temperature itself. Benchmarked against a rubidium vapor-cell standard over ten days, the composite clock reaches fractional instability below 5x10-9 at 200 seconds of averaging (4x better than the D transition alone) and below 1x10-8 at 2x105 seconds, roughly 2.3 days (200x better).
This does not touch the NV clock's absolute accuracy, and the paper reports that once temperature stops dominating, magnetic-field, optical-power and RF sensitivities are what limits it next - so the ceiling has moved rather than disappeared. What it does establish is that a solid-state clock can cancel its own leading systematic using only the internal structure of one defect, with no active temperature control loop, which is the difference between a sensor that needs a laboratory environment and one that can sit on a chip next to whatever it is timing.
[2026-08-28] · src Cornell, Xu, et al. (Loncar group, Harvard); Nature Physics (2026); arXiv:2508.13356
Solid-state spin qubits usually get protected from low-frequency environmental noise by dynamical decoupling: sequences of microwave pulses that periodically flip the spin. This group instead drove a silicon-vacancy spin in diamond continuously with a mechanical (phonon) field, putting it into a "dressed" basis defined by that driving rather than by the bare spin states, and performed optical initialization, control and readout entirely within that dressed basis. Because the dressed states are immune to the same low-frequency noise to first order, no separate decoupling sequence is needed - the drive that protects the qubit is the same drive used to control it, reaching Rabi frequencies up to 800 MHz, which the authors report as a record for this platform. The scheme extends the spin's coherence time by roughly a factor of three over the undriven case.
A phononic quantum network needs spins that store information and phonons that move it between nodes, and this closes the gap between those two roles for a single defect: the mechanical field is no longer just the carrier between nodes, it is also what keeps the memory coherent while it waits. An all-mechanical control-and-protection scheme is also compatible by construction with the phononic cavities such a network would use, where an equivalent microwave decoupling sequence is not always straightforward to apply. It is a threefold coherence gain on one defect type in one lab, not yet a demonstrated network.
[2026-08-28] · src Bennecke, Dinh, Bange, Schmitt, Merboldt, Weinhagen, van Wingerden, Frassetto, Poletto, Reutzel, Steil, Luke, Mathias, Jansen; Nature Communications (2026); arXiv:2502.18269
Photoemission orbital tomography reconstructs a molecular orbital's shape from the angle-resolved momentum distribution of photoelectrons it emits, but recovering a full 3D orbital this way needs photoelectron data across a range of photon energies, which in turn has needed a tunable synchrotron beamline. This group built a table-top source instead - a femtosecond extreme-UV pulse from high-harmonic generation, feeding an ultrafast momentum microscope - and paired it with a reconstruction algorithm redesigned to need far fewer energy points and less data than the synchrotron method. Applied to PTCDA molecules adsorbed on Ag(110), it recovered the full 3D shape of the molecule's highest occupied orbital at a spatial resolution finer than the distance between neighboring carbon atoms, matching what synchrotron-based 3D tomography has achieved.
Access to 3D orbital tomography has been rationed by synchrotron beamtime, which is scarce and fixed to a facility; a table-top version run on a femtosecond source removes both constraints and, because the pulses are femtosecond to begin with, opens the door to watching an orbital's shape change during a chemical reaction rather than only imaging it at rest - stroboscopic orbital videography, which the synchrotron version cannot offer regardless of beamtime. This paper demonstrates the static reconstruction only; the time-resolved extension is stated as future work, not shown here.
2
[2026-08-28] · src Arban, Complete Conservatory Method — 14 Characteristic Studies
Trumpet takes today's slot — three days since its last session, the longest gap of the three instruments right now, continuing the alto → piano → trumpet three-day cycle this rotation has held since early August. Five sessions of drills (tone, thirds, time feel, wide leaps, trills) led into the 14 Characteristic Studies three days ago, which the brief treats as repertoire rather than exercise; today continues that thread with No. 2 instead of opening new ground. Where No. 1 asked for tone that stays full to the last note, No. 2 asks for something closer to musicianship: holding a phrase together through real dynamic contrast instead of just enduring it.
6 minutes: lip slurs on the overtone series from low C, slurred with no tongue between adjacent partials, quarter=66, four slurs per breath, working up through the first five partials and back down. The point is smooth connection between registers before the study asks for dynamic change across register, not raw range.
Characteristic Study No. 2, played straight through once, honoring the written dynamics rather than settling everything into a safe mf — the study moves between piano and forte inside single phrases, and it's the contrast, not the notes, that makes it hard. Success criterion: at least two clearly different dynamic levels audible within the same phrase, each held long enough to read as a choice rather than smoothed into one general loudness.
Art of Phrasing, Melody No. 15. Play it once for notes, then once shading the dynamics the same way the technical target just asked for — the same skill, transferred to a simpler line at a smaller scale.
One clean pass at the Characteristic Study is the assignment, same as No. 1 — a second pass right behind it trains fatigue, not contrast. Stop there, or sooner if the tone starts to spread or the dynamic shading collapses back into one flat loudness; a tired embouchure can't produce contrast at all, so pushing past that point doesn't train what today is asking for.
[2026-08-28] · src —
hearBunny Berigan — I Can't Get Started (Presenting Bunny Berigan) · Tidal · Spotify
This is the historical benchmark for exactly today's technical target: control that reads as musicianship rather than as a display of stamina or range. The Victor recording opens with an extended unaccompanied cadenza over four sustained chords, and it's remembered as a trumpet "concerto" — jazz historian Dick Sudhalter's word for it — precisely because Berigan's "impeccable command of his instrument" never announces itself as difficulty. That's the same restraint today's Characteristic Study is asking a phrase full of dynamic contrast to have: control that reads as shape, not as a show of force.