Light-driven states: transient evidence and equilibrium limits

Scope

Light can change carrier populations, excite lattice motion and alter collective electronic response. Persistent photodoping, a transient pump–probe state and a periodically driven model are different protocols. Specify wavelength, polarization, pulse duration, fluence, repetition rate, base temperature and probe delay. A temperature at which a driven response is observed is not automatically an equilibrium superconducting transition temperature. R1 R2 R4 R6

The evidence now extends beyond optical conductivity. Electrical measurements in K3C60, transient magnetic measurements in YBa2Cu3O6.48 and a later sensitivity-limited magnetic experiment on K3C60 ask different questions. Preserve the specimen, geometry and measured observable rather than combining them into one universal light-induced transition. R5 R9 R10 R11

Mathematical assumptions

With the $e^{-i\omega t}$ convention, a normal Drude response is $\sigma(\omega)=D/(\Gamma-i\omega)$, where $D$ is the Drude weight and $\Gamma>0$ the relaxation rate. Thus $\sigma_2=D\omega/(\Gamma^2+\omega^2)\approx D/\omega$ when the measured frequencies greatly exceed $\Gamma$, although the DC conductivity $D/\Gamma$ remains finite. A finite-band inductive trend alone cannot distinguish this limit from a condensate. This elementary counterexample is not a fit to, or refutation of, any particular experiment. R4 R7

Depth reconstruction also matters. For Dodge's illustrative saturating profile, $\Delta\sigma(z)=\Delta\sigma_{\rm sat}f e^{-z/\Lambda}/(1+f e^{-z/\Lambda})$, with $f=F/F_{\rm sat}$, integration gives sheet photoconductance $\Delta G=\Delta\sigma_{\rm sat}\Lambda\ln(1+f)$. The effective thickness grows with fluence even while the surface response saturates. Assuming a fixed exponential profile can therefore bias the inferred surface conductivity. This integral checks the stated profile, not a full electromagnetic inversion; the experimental interpretation is disputed below. R7 R8

Kennes's minimal quadratic electron–phonon model gives oscillator frequency $\omega(n)=\omega_0\sqrt{1+2gn}$ for site occupation $n=0,1,2$, requiring $g>-1/4$. At fixed phonon occupation $n_B$, its induced interaction is $\Delta U=(n_B+1/2)\hbar\omega_0[\sqrt{1+4g}+1-2\sqrt{1+2g}]$. Concavity makes this nonpositive for either sign of stable $g$, with leading term $-(n_B+1/2)\hbar\omega_0g^2$. The author manuscript sets Planck's constant to one; it is restored here. This is a model mechanism, not a measured coupling or proof of a stable superconducting phase. R6

Experimental criteria

Fausti studied stripe-ordered La1.675Eu0.2Sr0.125CuO4 at a 10 K base temperature. The reported signal is a transient c-axis plasma edge and optical response interpreted as restored interlayer coherence. Figures 2–3 specify a 16 micrometre pump, approximately 1 mJ/cm² fluence and a 5 ps delay for the transient spectrum. The pump excites roughly 200 nm in depth, while the THz probe samples about 10 micrometres; the transformed-layer reconstruction is essential. A near-infrared plateau observed to 100 ps does not directly measure a many-nanosecond lifetime. R1

Hu's YBa2Cu3O6.5 specimen has an equilibrium transition near 50 K. Its approximately 300 fs, 15 micrometre pump modifies both inter- and intra-bilayer optical response. The effective-medium analysis separates transformed and untransformed regions; screened plasma-edge positions are not the same as fitted unscreened plasma frequencies. Broadband data at 60 K and narrower-band measurements at 100 and 300 K have different information content. Mankowsky's associated structural study combines diffraction with calculations to connect driven distortions with electronic changes; structural response does not uniquely identify pairing. R2 R3

Mitrano's K3C60 report compares a transient optical response above the roughly 20 K equilibrium transition with the low-temperature spectrum. Its title deliberately says “Possible”. The reported optical-gap fit and enhanced carrier mobility are not a direct DC resistance or magnetic measurement. Temperature, pump–probe delay and model-dependent gap extraction must remain attached to that evidence. R4

Budden extended the response using longer pulses. Figure 4 compares optical extrapolation with electrical measurements at a 100 K base temperature and about 25 mJ/cm² fluence. The optical and transport pump durations are 200 ps and 75 ps, respectively; the electrical voltage probe lasts 1 ns. The electrical experiment is transient two-terminal transport with contact contributions calibrated from static four-terminal measurements; it is not a direct transient four-terminal measurement. The optical error bars describe a fit confidence interval, while the transport shading estimates contact-calibration systematic error. Those uncertainties cannot be interchanged or discarded when assessing compatibility with zero. Recovery over tens of nanoseconds supports the authors' metastable interpretation without establishing a pump-free equilibrium transition at 100 K. The drive, probe and recovery times are different quantities. R5

Adelinia's 2025 thin-film study uses a quasi-DC bias, 300 fs pulses at 7 micrometres and photoconductive switches to resolve voltage dynamics. Granular-film and powder results are not interchangeable. The above-transition voltage response changes sign at 80 K; at 100 K it is positive and linear across the measured current range. Fitted superfluid fractions and weak-link phase-slip dynamics depend on the granular two-fluid model. A 100 K measurement in this paper must not become a 100 K superconducting-transition record. R10

Fava's 2024 YBa2Cu3O6.48 experiment detects a transient magnetic-field enhancement near the excited region's edge using Faraday rotation in a nearby GaP detector. Signals at 100 and 300 K last about 1 ps. The inferred diamagnetic susceptibility of order −0.3 depends on geometry and the magnetic-response model; the paper itself cautions that a quasi-static estimate is not fully justified for this dynamics. This provides a complementary magnetic observable, while remaining distinct from a persistent equilibrium Meissner state. R9

De Vecchi's December 2025 K3C60 study used a 50 K base temperature, about 1 ps pulses at 8 micrometres, 40 mJ/cm² fluence and a 0.5 mT applied field. No statistically significant photo-induced field signal above roughly 50 nT was found out to 1 ns. The authors also quote about 30 nT from the standard error of the mean; this is a different stated sensitivity measure. For their modeled long-lived response, the susceptibility bound is of order $\chi_v>-0.1$. It is not a bound on an arbitrary picosecond waveform: metallic shielding strongly filters short responses, and the detector has an approximately 160 ps response scale. A simulated short-lived response produces only about 2 nT, below detection. The model compares susceptibility statistics from six reference pellets rather than measuring the driven susceptibility of six samples. The authors conclude that these measurements do not decide whether Meissner diamagnetism is present or absent. R11

Compare the observable, the clock and the uncertainty before combining these experiments.

Limits and disputes

Dodge and colleagues argue that saturation-dependent excitation depth can create misleading superconducting-like reconstructed spectra. Buzzi and colleagues' March 2023 preprint response disputes that interpretation, arguing that the inductive signature survives their alternative reconstructions. The local profile and Drude checks above do not resolve this dispute. Compare raw reflected fields, depth assumptions, spectral bandwidth and independent transport or magnetic evidence rather than treating a fitted gap as a unique diagnosis. R7 R8

Long-lived after a short pulse does not mean thermodynamic equilibrium. Quantify absorbed energy, depth gradients, carrier and lattice temperatures, relaxation and repetition-rate heating. Selective excitation of a phonon does not establish that equilibrium pairing is phonon-mediated. Report the detector and circuit response alongside the pump duration: a 1 ps drive does not give every probe 1 ps resolution. A null magnetic result constrains the response only through the measured sensitivity and the assumed geometry and time dependence. Optical, contact-calibrated electrical and magnetic observations should retain their own uncertainty models. The present review covers selected original passages and figures; full supplements, remaining version comparisons, experimental raw-data reproduction and complete scientific acceptance remain open. R3 R5 R6 R10 R11

Discovery implications

Use synchronized optical, electrical and magnetic probes with explicit controls and uncertainty. Report signal duration separately from formation time and instrumental resolution. Keep equilibrium transition records separate from driven base temperatures and model estimates in databases and search filters. Theory should predict observables under the actual drive protocol, including relaxation, rather than equating an attractive model interaction with an experimentally established superconducting phase. R5 R6 R9 R11

Related topics

Compare fullerides, electrostatic gating, magnetic fields, pair-density waves and critical fields.

Material cases

Values describe individual reports, not a ranking. Unknown means the checked source did not supply the field at this review scope.