Single-Cycle Light Transients
- Single-cycle light transients are ultrashort electromagnetic waveforms lasting roughly one optical cycle, enabling critical carrier-envelope phase control.
- Their generation employs advanced techniques such as fiber-comb soliton self-compression, gap solitons, and nonlinear pulse shaping to concentrate most energy in the main cycle.
- These pulses underpin applications in high-resolution spectroscopy, broadband supercontinuum generation, and ultrafast control of light–matter interactions in diverse media.
Single-cycle light transients are electromagnetic waveforms whose temporal extent approaches a single oscillation of the carrier field, or in closely related cases a half-cycle or sub-cycle interval. In the literature represented here, the term spans several experimentally and theoretically distinct regimes: all-fiber frequency combs at compressed to $6.8$–$7.1$ fs, single-cycle or sub-cycle gap solitons in resonant structures, mid-IR single-cycle light bullets in LiF, radially polarized single-cycle THz pulses near laser-produced plasmas, and optical-cycle refractive-index transients in transparent conducting oxides (Xing et al., 2021, Zhang et al., 2024, Xie et al., 2010, Chekalin et al., 2015, Kalmykov et al., 2020, Choi et al., 31 Dec 2025). Across these settings, the central technical theme is that carrier-envelope phase, higher-order dispersion, nonlinear phase accumulation, and the distinction between envelope and carrier dynamics all become operationally decisive on the scale of one optical cycle.
1. Conceptual and physical foundations
A single-cycle transient is not defined solely by pulse duration in femtoseconds; it is defined relative to the carrier period at the relevant center wavelength or center frequency. In the all-fiber comb systems, $6.8$ fs at $1930$ nm corresponds to $1.05$ optical cycles, while $7.1$ fs at $1970$ nm corresponds to $1.1$ optical cycles (Xing et al., 2021, Zhang et al., 2024). In the mid-IR LiF filament experiments, a light bullet at $6.8$0 nm reaches a duration of approximately $6.8$1 fs at maximum compression, again near a single optical cycle (Chekalin et al., 2015). In the TCO dynamic-photonics study, a $6.8$2 response time as short as $6.8$3 fs is described as approaching a single optical cycle and tunable to sub-cycle timescales (Choi et al., 31 Dec 2025).
At this limit, the electric field waveform itself becomes the relevant dynamical object. The abstract of the 2024 GHz comb study states that single-cycle optical pulses offer “a strong carrier-envelope-offset (CEO) dependent electric field and the highest peak intensity for a given pulse energy” (Zhang et al., 2024). The 2021 all-fiber comb paper similarly emphasizes deterministic carrier-envelope phase and long-term stable carrier-envelope phase as enabling study and control of light-matter interactions at the sub-cycle timescale and efficient generation of low-noise multi-octave frequency combs (Xing et al., 2021). In the nanotip-based Homochromatic Attosecond Streaking (HAS) work, the target quantity is explicitly the “instantaneous field waveform of the single-cycle transient” (Kim et al., 2024).
Several theoretical mechanisms recur. In fiber systems, pulse evolution is modeled by the generalized nonlinear Schrödinger equation, with anomalous dispersion, Kerr nonlinearity, Raman response, self-steepening, and higher-order dispersion determining whether a chirped few-cycle or tens-of-femtoseconds input self-compresses into a clean single-cycle main lobe (Xing et al., 2021, Zhang et al., 2024). In LiF filaments, the slowly-varying-envelope approximation is used together with Kerr self-focusing, anomalous group-velocity dispersion, ionization, and plasma defocusing to describe formation of a near-single-cycle light bullet (Chekalin et al., 2015). In resonant two-level media, full Maxwell–Bloch equations without the slowly varying envelope approximation and rotating wave approximation are required because intrapulse four-wave mixing and carrier-resolved dynamics become essential for sub-cycle pulse formation (Xie et al., 2010, Arkhipov et al., 2024).
A second foundation is the distinction between group and phase transport. In the LiF light-bullet study, regular oscillations of the light-bullet intensity are traced to the difference of the wave packet group velocity and the carrier wave phase velocity; the breathing length is given by
$6.8$4
This phase slippage periodically moves a carrier crest, zero crossing, and inverted crest beneath the compressed envelope peak, modulating the instantaneous field amplitude during propagation (Chekalin et al., 2015).
2. Fiber-comb implementations and soliton self-compression
The most explicit realization of practical single-cycle optical transients in the data is the all-fiber frequency-comb platform. In “Single-cycle all-fiber frequency comb,” a commercial $6.8$5, $6.8$6 MHz menlo Figure-9 mode-locked laser at $6.8$7 mW is amplified in PM Er:fiber, then sent through $6.8$8 m of PM HNLF to generate a tunable $6.8$9 to $7.1$0 seed by soliton self-frequency shift. The seed is stretched in PM2000D fiber, amplified in a $7.1$1 cm double-clad, heavily-doped Tm:silica fiber held at $7.1$2 to suppress ASE, pre-compressed in $7.1$3 m of PM1550, and finally self-compressed in $7.1$4 cm of elliptical-core HNLF$7.1$5 (Xing et al., 2021). The output is $7.1$6 fs pulses with $7.1$7 kW peak power and $7.1$8 mW average power, and the spectrum covers more than two octaves, from below $7.1$9 nm up to 0 nm (Xing et al., 2021).
The 2024 GHz system pushes the same architecture to a higher repetition rate. A commercial 1 GHz Er:fiber comb at 2 nm produces 3 fs pulses of 4 pJ, 5 cm of HNLF6 shifts the soliton to 7 nm, and 8 cm of heavily Tm-doped fiber with net normal dispersion functions as amplifier and stretcher. Dual-stage compression then uses 9 cm PM1550 fiber and $6.8$0 cm HNLF$6.8$1, producing $6.8$2 fs pulses centered around $6.8$3 nm at $6.8$4 W average power (Zhang et al., 2024). The study identifies this as the first GHz single-cycle source (Zhang et al., 2024).
The soliton-compression criterion is stated in both fiber papers through the soliton order
$6.8$5
In the GHz comb, the compressed $6.8$6 fs chirped pulse enters HNLF$6.8$7 with $6.8$8, undergoing soliton-effect compression to $6.8$9 fs (Zhang et al., 2024). In the $1930$0 MHz system, higher-order dispersion and Raman terms are minimized by using a very short doped-fiber length and cut-back-optimized HNLF$1930$1, yielding a clean main lobe with $1930$2 of total energy and negligible satellites (Xing et al., 2021). The GHz study reports that $1930$3 of the total pulse energy lies in the central cycle peak; the abstract summarizes this as $1930$4 of the pulse energy concentrated in the pulse center (Zhang et al., 2024).
Both systems also emphasize coherence metrics. The $1930$5 MHz comb yields a measured carrier-envelope-offset beat-note SNR $1930$6 dB at $1930$7 kHz RBW and relative integrated intensity noise $1930$8 over $1930$9 Hz–$1.05$0 MHz (Xing et al., 2021). The $1.05$1 GHz seed laser demonstrates a $1.05$2 dB SNR for the CEO frequency at $1.05$3 kHz RBW and a free-running CEO linewidth of $1.05$4 kHz at $1.05$5 kHz RBW, facilitating comb stabilization and CEO control (Zhang et al., 2024).
| Platform | Reported output | Distinctive feature |
|---|---|---|
| All-fiber $1.05$6 MHz comb | $1.05$7 fs, $1.05$8 kW, $1.05$9 mW | first all-fiber configuration that generates single-cycle pulses |
| All-fiber $7.1$0 GHz comb | $7.1$1 fs, $7.1$2 W, $7.1$3 kW | first GHz single-cycle source |
The $7.1$4 MHz paper extends the source beyond the near-IR. By extending HNLF$7.1$5 to $7.1$6 cm, a smooth, coherent supercontinuum from $7.1$7 to $7.1$8 with $7.1$9 mW integrated power is generated directly in silica fiber, with a smaller fraction extending to $1970$0. Using intra-pulse difference-frequency generation in CdSiP$1970$1 and orientation-patterned GaAs, few-cycle pulses from $1970$2 to $1970$3 are produced, and the abstract additionally states that few-cycle pulses extending from $1970$4 to beyond $1970$5 with long-term stable carrier-envelope phase are created using intra-pulse difference frequency (Xing et al., 2021). This suggests that the single-cycle optical transient in fiber can act as a pump waveform for coherent spectral translation across a much wider band than the primary $1970$6 output itself.
3. Propagation in bulk media, thin layers, and resonant periodic structures
Single-cycle light transients also appear as self-organized propagation states in bulk and structured media. In LiF, mid-IR femtosecond pulses at $1970$7–$1970$8 nm and power slightly exceeding the critical power for self-focusing generate single-cycle light bullets in a filament (Chekalin et al., 2015). The experimental signature is a strictly periodic structure of color centers with section lengths about $1970$9, which increase with decreasing wavelength. The reported periods are $1.1$0 at $1.1$1 nm, $1.1$2 at $1.1$3 nm, and $1.1$4 at $1.1$5 nm (Chekalin et al., 2015). The interpretation is not that the envelope repeatedly breaks apart, but that the field amplitude of an extremely compressed single-cycle wave packet undergoes regular oscillation because the envelope travels at $1.1$6 while the carrier travels at $1.1$7 (Chekalin et al., 2015).
The companion LiF study on self-reconstruction adds a second propagation property: robustness across discontinuities. A single-cycle light bullet formed before an air gap up to $1.1$8 mm width “completely recovered” after passing some distance in LiF after the gap (Chekalin et al., 2020). The bullet in the gap has a strongly convergent wave front with focusing radius of $1.1$9–$6.8$00, and its divergence after the waist is considerably less than that of a Gaussian beam. In simulation, the far-field divergence is $6.8$01 rad, about four times smaller than a Gaussian beam with the same waist, $6.8$02 rad (Chekalin et al., 2020). The reconstruction length grows with gap width and with the pre-gap propagation length, indicating that nonlinear phase accumulation before the gap conditions later self-reformation (Chekalin et al., 2020).
A different route is provided by resonant periodic media. Xie and Macovei show that a pulse with a few optical cycles penetrating through resonant two-level dense media with a subwavelength structure can generate a single sub-cycle optical pulse, observed as a single-cycle gap soliton in the full Maxwell–Bloch equations without the frame of slowly varying envelope and rotating wave approximations (Xie et al., 2010). The subwavelength structure suppresses the frequency shift caused by intrapulse four-wave mixing in continuous media and supports the formation of single-cycle gap solitons even when the structure period breaks the Bragg condition (Xie et al., 2010). Representative numerical results quoted in the summary give, for exact Bragg structure $6.8$03, a transmitted pulse with $6.8$04 fs, $6.8$05 cycles, and for broken Bragg $6.8$06, $6.8$07 fs, $6.8$08 cycles (Xie et al., 2010).
Related resonant-medium control appears in the 2024 study of colliding single-cycle self-induced transparency pulses. There, two counter-propagating single-cycle attosecond self-induced-transparency pulses in a two-level medium generate a population-difference grating whose first-collision spatial period is $6.8$09, halving to $6.8$10 after the $6.8$11th collision (Arkhipov et al., 2024). Through the induced refractive-index modulation, this produces a microcavity with Bragg-like mirrors. The parameters of the structure can be quickly adjusted by increasing the number of collisions, demonstrating control of dynamic properties of the medium on a sub-cycle time scale by using attosecond pulses (Arkhipov et al., 2024).
A more unusual single-cycle transformation is proposed by Arkhipov et al. for reflection from an optically thin metallic or dielectric layer. In the strictly one-dimensional geometry considered there, the emitted field is proportional to the velocity of oscillating medium charges rather than their acceleration, and reflection of a single-cycle optical pulse can therefore yield an approximately unipolar half-cycle pulse (Arkhipov et al., 2017). The reflected field for a $6.8$12-thin polarization sheet is written as
$6.8$13
and the physical consequence is that the emitted field can keep constant sign throughout the pulse duration if the carrier velocity does not cross zero during the second half of the incident waveform (Arkhipov et al., 2017).
4. Interaction with matter and dynamic photonic media
Single-cycle light transients are used not only as ultrashort probes or propagation states, but as drivers of nonequilibrium matter. In correlated organic conductors, nearly single-cycle $6.8$14 fs, $6.8$15-cycle near-infrared pulses with peak fields up to $6.8$16 MV/cm induce transient charge localization effects (Kawakami et al., 2018). In $6.8$17-(BEDT-TTF)$6.8$18I$6.8$19, a transient short-range charge-order state is induced in a metallic phase, while in $6.8$20 the field-induced reduction of a transfer integral is captured as a red shift of the plasma-like reflectivity edge (Kawakami et al., 2018). The minimal theory summarized there uses a time-dependent tight-binding Hamiltonian with Peierls substitution and Coulomb terms, together with dynamical localization: $6.8$21 For typical parameters in that review, a field of order $6.8$22 MV/cm would drive $6.8$23, while the experimental peak fields $6.8$24 MV/cm are estimated to produce a $6.8$25 reduction of $6.8$26 (Kawakami et al., 2018).
In the THz domain, single-cycle pulses can drive structural order. In Nb$6.8$27Sn, above-threshold single-cycle THz pulses described by a sin-Gaussian waveform with $6.8$28 MV/cm, $6.8$29 meV, and $6.8$30 ps induce a light-induced metastable Martensitic anomaly (Yang et al., 2020). The THz spectrum overlaps the $6.8$31 phonon around $6.8$32–$6.8$33 meV, and the paper attributes the response to coherent driving of $6.8$34, lifting the $6.8$35 degeneracy non-thermally (Yang et al., 2020). Experimentally, the nonequilibrium signature persists up to $6.8$36 K, more than twice the equilibrium $6.8$37 K, shows a threshold field $6.8$38 kV/cm, and survives on $6.8$39 ns timescales (Yang et al., 2020).
An even more direct form of optical-cycle matter control appears in transparent conducting oxides. Choi et al. find oscillatory, sign-reversing dynamics on a few optical-cycle timescale under extreme electron temperatures and construct an inverse-designed multilayer cavity incorporating an ultrathin TCO layer that supports this behavior (Choi et al., 31 Dec 2025). In the intraband regime, the time-dependent permittivity is written as
$6.8$40
with $6.8$41 governed by a two-temperature model (Choi et al., 31 Dec 2025). The inverse-designed ENZ cavity produces transmittance oscillations with a period of $6.8$42 fs, corresponding to three optical cycles of the probe beam, and the acceptor-layer design yields a $6.8$43 response time as short as $6.8$44 fs, tunable to sub-cycle timescales (Choi et al., 31 Dec 2025). The paper explicitly links these timescales to time-reflection, time-refraction, and related dynamic phenomena from the visible to the infrared (Choi et al., 31 Dec 2025).
A related, but distinct, current-driven single-cycle transient arises in plasma columns. Photoionization by a femtosecond terawatt laser pulse creates a plasma column whose surface rotational current supports a broadband, evanescent THz signal accompanying the wake; a few millimeters away from the column, rapid evanescence of high-frequency components reshapes this signal into a radially polarized, single-cycle pulse (Kalmykov et al., 2020). In the example given, the near-field THz pulse at $6.8$45–$6.8$46 mm is single-cycle, $6.8$47 ps, $6.8$48–$6.8$49 THz, and $6.8$50–$6.8$51 kV/m (Kalmykov et al., 2020). This suggests that “single-cycle light transient” is not restricted to free-space optical pulses launched from conventional lasers; it also includes near-field and plasma-mediated waveforms produced by current reorganization and spectral filtering during propagation.
5. Metrology and field-resolved characterization
Because the observable of interest is the field waveform, metrology must be carrier-resolved or sub-cycle-resolved. In the all-fiber $6.8$52 comb, all-reflective SHG-FROG yields a $6.8$53 retrieval error and confirms a $6.8$54 fs single-cycle envelope with negligible pedestal (Xing et al., 2021). In the GHz comb, all-reflective SHG-FROG records a spectrogram over delays $6.8$55 fs, shows an octave of SHG signal from $6.8$56 to $6.8$57 nm in phase, and yields a retrieval error of $6.8$58 (Zhang et al., 2024). These measurements establish not only duration but also whether satellites and pedestals are sufficiently suppressed that most of the usable electric field lies within one cycle.
For mid-IR extensions of the $6.8$59 MHz source, dual-comb electro-optic sampling with two asynchronous near-IR combs and a thin GaP crystal measures the mid-IR electric field $6.8$60 with sub-cycle resolution (Xing et al., 2021). The Fourier transform of $6.8$61 at $6.8$62 GHz resolution shows comb-tooth structure spaced by $6.8$63 MHz and resolves molecular absorption lines such as CO$6.8$64 at $6.8$65 (Xing et al., 2021). This establishes a direct link between single-cycle pumping, field-resolved detection, and comb-resolved spectroscopy.
Homochromatic Attosecond Streaking extends field sampling to non-phase-stabilized single-cycle transients. In HAS, a weak replica of the same pump pulse drives the streaking gate, and the cutoff-energy shift yields an effective vector potential
$6.8$66
Inverting this relation gives $6.8$67, and differentiation gives $6.8$68, which is the actual instantaneous field waveform of the single-cycle transient (Kim et al., 2024). The key practical result is that, owing to the exponential sensitivity of nanotip field emission to the instantaneous peak field, the averaged random-CEP spectrogram is dominated by the “optimal CEP” trace in the cutoff region. The measured fields agree with synthesized waveforms with similarity $6.8$69–$6.8$70, and even with CEP unlocked the retrieved field remains identical to the optimal-CEP case with $6.8$71 and attosecond timing precision $6.8$72 as (Kim et al., 2024).
A chip-scale alternative is strong-field nonlinear photocurrent sampling in silicon. In the CMOS-chip method of Liu et al., strong-field mid-IR excitation creates a sub-cycle optical gate through the nonlinear dependence of photocurrent on instantaneous field, and a crossing-angle geometry maps time delay onto a transverse sensor coordinate (Liu et al., 2021). The delay varies as
$6.8$73
with $6.8$74 and pixel pitch $6.8$75, giving $6.8$76 fs/pixel (Liu et al., 2021). The demonstrated retrieval is a $6.8$77-cycle, $6.8$78 fs mid-IR waveform including absolute CEP, with single-shot SNR $6.8$79 (Liu et al., 2021).
Sub-cycle field-resolved detection also underpins single-nanowire multi-THz spectroscopy. Difference-frequency generation in GaSe produces intrinsically phase-locked, few-cycle multi-THz pulses tunable from $6.8$80 THz to $6.8$81 THz, while electro-optic sampling with a synchronized $6.8$82 fs gate pulse yields $6.8$83 fs time resolution (Eisele et al., 2016). Combined with s-NSOM, this enables reconstruction of the time-dependent dielectric function at the surface of a single photoexcited InAs nanowire with $6.8$84 nm spatial resolution and reveals ultrafast $6.8$85 fs) formation of a local carrier depletion layer (Eisele et al., 2016). Although the THz pulses there are described as few-cycle or $6.8$86-cycle, the metrological framework is directly relevant to single-cycle transients because the measurement condition is sub-cycle resolution.
6. Applications, misconceptions, and research directions
The applications explicitly identified in the sources are diverse but technically coherent. The fiber-comb papers point to coherent multi-octave supercontinuum generation in silica fiber, high-brightness comb-resolved spectroscopy in the $6.8$87–$6.8$88 atmospheric window, strong-field and sub-cycle light-matter interactions, spectroscopy, microscopy, and CEO-sensitive nonlinear optics (Xing et al., 2021, Zhang et al., 2024). The TCO work identifies time-refraction, time-reflection, temporal Bragg mirrors, photonic time crystals, and temporal metasurfaces as target regimes once $6.8$89 exceeds $6.8$90 on optical-cycle timescales (Choi et al., 31 Dec 2025). The LiF studies suggest control of absolute CEP changes on a tens-micron scale and possible use in high-harmonic generation or precision micro-structuring through engineering of the breathing length $6.8$91 (Chekalin et al., 2015). The THz work on Nb$6.8$92Sn shows that single-cycle pulses can stabilize hidden structural order on technologically relevant nanosecond timescales (Yang et al., 2020).
A common misconception is that “single-cycle” is synonymous with “maximum compression” alone. The data do not support that simplification. Several papers make clear that usable single-cycle behavior depends on how much energy sits in the central cycle, on suppression of higher-order-dispersion-induced satellites, and on CEP control or at least CEP-selective metrology (Xing et al., 2021, Zhang et al., 2024, Kim et al., 2024). Another misconception is that such sources must be free-space, alignment-sensitive Ti:sapphire or OPCPA systems. The 2021 and 2024 comb papers explicitly present compact, turn-key, all-fiber implementations, including a $6.8$93 breadboard in the $6.8$94 MHz case and all-PM-fiber branches in the GHz case (Xing et al., 2021, Zhang et al., 2024).
The sources also distinguish several neighboring concepts that should not be conflated. A single-cycle pulse, a sub-cycle pulse, an approximately unipolar half-cycle pulse, a single-cycle light bullet, and a single-cycle refractive-index transient are related but not identical objects (Arkhipov et al., 2017, Chekalin et al., 2015, Choi et al., 31 Dec 2025). The first is a free electromagnetic waveform occupying about one period; the second is shorter; the third is a waveform with one dominant sign; the fourth is a spatiotemporally confined propagation structure; and the fifth is a material modulation rather than a free field. This suggests that “single-cycle light transients” functions best as an umbrella category for carrier-resolved ultrafast electromagnetic or optically induced temporal phenomena, rather than a single experimental platform.
Future directions are stated explicitly in several studies. The $6.8$95 MHz fiber-comb paper proposes scaling to sub-cycle $6.8$96 fs) operation via custom fibers, higher repetition rates or GHz combs, and integration with soft-glass or photonic-crystal fibers for extended mid-IR coverage (Xing et al., 2021). The GHz paper numerically predicts scalability to $6.8$97 GHz by doubling the Tm-fiber length from $6.8$98 m to $6.8$99 m, while preserving the optimal HNLF$7.1$00 regime (Zhang et al., 2024). The TCO study points toward visible-to-infrared time-varying photonic media based on thermionic carrier injection (Choi et al., 31 Dec 2025). The HAS paper anticipates compact field monitoring in light-field synthesizers and possible extension to mid-infrared or terahertz regimes through different tip materials or plasmonic enhancement (Kim et al., 2024). Taken together, these directions indicate that the field is moving simultaneously toward higher repetition rate, broader spectral reach, stronger waveform control, and more compact carrier-resolved diagnostics.