Robust Floquet-induced gap in irradiated graphite

Abstract: Floquet engineering provides an emerging pathway for tailoring the electronic states of quantum materials through time-periodic drive. A critical step along this direction is achieving light-induced modifications of the dynamical electronic structure, such as avoided-crossing gap at the Floquet Brillouin zone boundary, via efficient coupling of electrons with the coherent light-field. Here, we report robust Floquet-induced gap in bulk graphite that persists despite the presence of interlayer coupling and photo-excitation. Using time- and angle-resolved photoemission spectroscopy with intense mid-infrared pumping, we directly reveal Floquet-induced gaps at resonance points both in the valence and conduction bands, accompanied by coherent Floquet sidebands. The gap and sidebands coexist with photo-excited carriers, yet their distinct timescales allow us to disentangle their origins. Our demonstration of robust Floquet-induced gaps establishes graphite as a platform for coherent manipulation of Dirac fermions and realization of light-engineered quantum phases.

• The paper demonstrates Floquet-induced avoided-crossing gaps in bulk graphite using time- and angle-resolved photoemission spectroscopy, measuring gaps of 158 ± 15 meV and 220 meV near the H point.
• The authors distinguish coherent light-field dressing from slower photoexcited-carrier relaxation through their different temporal signatures, showing that the gap appears only during the approximately 114 fs pump–probe overlap.
• The findings indicate that short, intense mid-infrared pulses can protect Floquet hybridization against interlayer coupling and unblocked interband excitation, while leaving longer-pulse stability, momentum dependence, and transport effects unresolved.

Overview

This paper reports time- and angle-resolved photoemission spectroscopy (TrARPES) evidence for Floquet-induced avoided-crossing gaps in bulk graphite under intense mid-infrared (MIR) pumping (2603.28724). The central result is that light-field dressing of the Dirac cone near the H point survives despite two conditions long considered detrimental: interlayer coupling, which introduces additional bands and scattering channels absent in monolayer graphene, and direct photo-excitation across the Dirac cone, which generates a hot-carrier population that could in principle decohere the driven states. The authors disentangle coherent dressing from incoherent carrier dynamics by exploiting their distinct timescales and extract gap magnitudes of $\Delta = 158 \pm 15$ meV and 220 meV at resonance points on opposite sides of the zone.

Experimental approach

The measurements use an exfoliated graphite flake (~1 mm flat region) held at 80 K under ultra-high vacuum (<3.6 × 10⁻¹¹ Torr). A high-harmonic generation probe (21.7 eV, 66 fs) is combined with a $p$-polarized MIR pump ($\lambda$ = 3 μm, $\hbar\omega$ = 415 meV) delivered at a fluence of 5.2 mJ/cm². The pump photon energy was deliberately chosen so that first-order Floquet sidebands ($n = \pm 1$) of the Dirac cone overlap the equilibrium dispersion at momenta satisfying the resonance condition $E_{CB,k} - E_{VB,k} = \hbar\omega$, maximizing hybridization at the Floquet-zone boundary. The probed momentum cut intersects the Dirac cone near H ($k_z = \pi/c$), whose dispersion closely resembles that of monolayer graphene; the split parabolic-touching bands near K are not directly addressed.

Disentangling dressing from photo-excitation

Because charge-neutral graphite permits interband absorption at 415 meV, photo-excited carriers and Floquet-dressed states coexist. The key methodological advance is temporal separation of their signatures:

• Photo-excited conduction band: rises sharply after time zero and relaxes over several hundred femtoseconds (up to picoseconds near $E_F$).
• Floquet sidebands: appear only within the pump–probe temporal overlap, peaking exactly at time zero — consistent with a convolution of pump and probe pulses.
• Valence-band suppression at resonance points: fits with two components, one tracking the Gaussian Floquet dynamics and one tracking the slower carrier-relaxation exponential.

The two-component decay of the valence-band depletion is the strongest claim in the dynamical analysis: it implies that suppression at the resonance point is not merely hole-burning from interband excitation but contains a genuine contribution from gap opening. This inference rests on the assumption that the two fitted components can be cleanly assigned to dressing versus relaxation, which the authors support by comparison with sideband dynamics but do not independently verify (e.g., via polarization or fluence scaling).

Gap extraction and confirmation of Floquet origin

Energy distribution curve (EDC) analysis at $\Delta t = 0$ shows peak splitting at resonance momenta in both valence and conduction bands, while EDCs away from resonance remain unchanged. The extracted gaps — 158 ± 15 meV at $k_3$ and 220 meV at $p$0 — are comparable to those reported in strongly electron-doped epitaxial graphene at similar fluence, where interband excitation is Pauli-blocked. That a similar gap magnitude is achieved here with unblocked photo-excitation constitutes the paper's principal evidence for robustness.

Temporal evolution provides independent confirmation: second-derivative maps show gapless Dirac dispersions at ±200 fs and clear gaps only within a ~114 fs window around time zero, matching the sideband temporal profile. Since this window equals the pump–probe cross-correlation, the gap exists strictly when the driving field is present, establishing its Floquet character from the time domain.

Limitations and open questions

Several caveats qualify these results. First, the observation window is limited to ~114 fs; whether the gap survives under longer pulses — where accumulated scattering would more severely compete with dressing — remains untested, and the authors pose this explicitly as an open question. Second, the measurement probes only the $p$1 slice resembling monolayer graphene; the effect of the K-point band splitting on Floquet gaps at other $p$2 values is unresolved. Third, the two-component decomposition of valence-band suppression relies on fitting assumptions rather than a microscopic model, so the relative weight of the "gap-opening" component is not quantified against theory. Finally, all observations are spectroscopic; transport consequences (e.g., light-induced anomalous Hall response) are not measured here.

Conclusion

This work demonstrates that Floquet-induced hybridization gaps open and persist in bulk graphite under conditions — interlayer coupling, extra scattering channels, and simultaneous interband photo-excitation — previously expected to disrupt coherent light-field dressing. The timescale hierarchy between sub-cycle Floquet formation (~114 fs) and carrier relaxation (≥ several hundred fs) is identified as the operative protection mechanism, with short pump and probe pulses essential to detection. The result establishes graphite as a viable platform for Floquet engineering of Dirac fermions in three dimensions and leaves open the questions of gap survival under longer drives, $p$3 dependence, and emergent light-induced phases specific to the multilayer band structure.