Search

In this paper, we present the study on negative refractive index in an inhomogeneously broadened four-level inverted-Y atomic medium based on electromagnetically induced transparency (EIT). The expressions of the relative permittivity and the relative permeability are derived under Doppler broadening. For this four-level system, we have found two frequency bands of negative refractive index in an optical region corresponding to two EIT windows. The influences of coupling and signal laser fields as well as temperature on frequency bands of negative refractive index are investigated. Our research can be convenient for experimental implementation with real atomic media under different temperatures.

Three-dimensional particle-in-cell simulations are conducted to investigate the self-compression of a sub-terawatt (TW) laser pulse when propagating in a thin, dense gas target. In this way, the interplay of ionization of gas atoms, defocusing and diffraction of laser fields cause the phase modulation and spectral broadening required to realize a short duration for the output pulse 1 , 2 .

Spectroscopy is the most established diagnostic method in astrophysics and plasma physics. Radiating ions can provide insight into z-pinch plasma parameters such as temperature, density, implosion velocity and magnetic field. Visible spectroscopy offers the possibility of accurate determinations of spectral line widths, shifts, polarization, and/or absolute intensity calibrations. In high current (>1MA) z-pinches, the plasmas are highly ionized and most of the emitted photons are in the XUV or x-ray region. It is a challenge to use visible spectroscopy there because there are not many detectable visible lines. We have identified several visible spectral lines in gas puff z-pinches on COBRA. These are pairs of lines at 5174 and 5193A from Ar VIII 4d-4f transitions, as well as 5801.3 and 5812.0A lines from C IV 3s-3p transitions, which are emitted from the imploding plasma shell. The Doppler shift due to the implosion is also observed. Zeeman splitting of these lines is measured to infer the magnetic field in the plasma. We have also observed several Ar II lines at a radius smaller than the imploding plasma shell, where both the Zeeman and Doppler shifts are absent. Using a triple-nozzle gas puff z-pinch on COBRA, we are able to measure the Zeeman splitting and Doppler broadening of the Ar VIII and C IV lines about 15 ns before the pinch time. After that, the plasma is so hot that the visible line radiation is washed out by the intense continuum radiation.

We have studied the THz generation in DSTMS using a quasi-3D model that takes into account high-order nonlinear effects beyond the χ(2) nonlinearity. The nonlinear refractive index (Kerr effect) and three-photon absorption (3PA) coefficients for the wavelength of 1.4 μm utilized in the model were determined by the Z-scan method. The simulation results are in good agreement with the experimentally measured THz and near-infrared spectra, confirming the validity of the proposed model. Our analysis shows that the 3PA process mainly limits THz generation in DSTMS at high pump fluences. The cascaded optical rectification further decreases THz conversion efficiency due to the spectral broadening of the excitation pulse.

A silicon nitride horizontal slot waveguide with silicon dioxide cladding is designed to achieve flat negative dispersion ( $\text {nm}\cdot \text {km}$ )) over an octave-spanning bandwidth (782 to 2100 nm). Simulations show that, when a 10-fs pulse with 1-W peak power is input into a 5-cm long waveguide, the pulse width becomes 448 times wider and its peak power is reduced to $2.23\times 10^{-3}$ W. After passing through another ideal positive dispersion element, we find that the output pulse peak power can be recompressed close to 1 W in the case of different input pulse widths. Moreover, for the input pulse with a much higher peak power, the pulse width of the output pulse can be compressed to a much narrower level because of the spectral broadening introduced by optical nonlinearity. A 100-fs input pulse with 10-kW peak power can be compressed to 12.5 fs with a peak power of 63 kW. Considering the time-domain broadening and the peak power reduction, such a slot waveguide with a flat negative dispersion could be used as a time stretcher for 10-fs level pulse in on-chip chirped pulse amplification (CPA) system.

Two feasible schemes are proposed by laser cooling of alkali atoms to realize the narrow-linewidth, continual superradiant lasing. Utilizing the D1 gray molasses to realize a cloud of cold Cs atoms for the two-level Faraday active optical clock (AOC), the effect of Doppler broadening on the transmission bandwidth of Faraday atomic filter will be effectively suppressed to the level of natural linewidth. Thus, the influence of mechanical or thermal vibrations of cavity mirrors on the clock laser will be decreased by about 102 times, which is expected to further narrow the laser linewidth by two orders of magnitude compared with the thermal atomic Faraday AOC to Hz level. Moreover, the four-level AOC based on the narrow linewidth cooling of Cs atoms using the 6S-7P transition is proposed. The 455 nm laser at 6S 1/2 -7P 3/2 transition plays two roles simultaneously, one is the cooling laser for a magneto-optical trap (MOT), another is the pumping laser for population inversion of 1470 nm clock transition, which is helpful to improve the compactness of the system. In the four-level structure, the frequency of pumping laser is far detuning from that of clock laser, therefore, the light shift resulting from the 455 nm laser is negligible. This four-level AOC based on cold atoms will further reduce the collision shift in thermal atom system to achieve a high-precision optical frequency reference.

The ultrabroadband supercontinuum (SC) source possesses a number of unique characteristics which have had a major impact in many research fields, ranging from precision metrology to high resolution imaging and ultrafast spectroscopy [1] . The generation of a supercontinuum is a highly complex process resulting from the propagation of intense light pulses in a nonlinear medium such as an optical fibre or waveguide. As a result of the very rich landscape of nonlinear dynamics, the input pulses injected into the fibre experience massive spectral broadening that can expand an initially narrowband input field to an output spanning from the visible to infrared [1] .

Gas-based systems (e.g. based on hollow-core fibres) are extensively used over a wide spectral range to temporally compress ultrashort pulses. Recent progress in solid-state lasers and compression schemes has enabled the generation of sub-femtosecond pulses at unprecedented energies and average powers [1] – [3] . However, the latest generation of lasers deliver mJ pulses at hundreds-of-kHz repetition rates [4] , which raises new challenges for gas-based pulse compression. At high repetition rates, even weak single-shot energy deposition in the gas (e.g. by ionisation-driven heating) is sufficient to cause a transversely non-uniform gas density depression to build up pulse-by-pulse [5] , leading to thermal instabilities. Using lighter noble gases is often insufficient to mitigate this, despite the weaker nonlinear absorption. This is because sufficient nonlinearity for spectral broadening requires substantially higher pressures, resulting in much slower gas relaxation (the thermal diffusivity scales inversely with gas density).

On-chip supercontinuum generation (SCG) combines the advantage of broad spectral coverage with extremely low footprint, offering an integrated solution for spectroscopy, frequency metrology and sensing applications [1] . SCG has been extensively studied in the past two decades and several platforms based on different materials have been demonstrated. For on-chip SCG, silicon nitride (Si 3 N 4 ) is one of the most promising materials owing to the mature fabrication technology, CMOS compatibility and large transparency window from the visible to the mid-IR [2] . There is a rich landscape of nonlinear dynamics leading to SCG and pumping in the anomalous dispersion regime generally yields larger spectral broadening but with extreme noise-sensitivity. All-normal dispersion regime (ANDi) is not as efficient but this regime increases significantly the phase-stability and coherence property of the supercontinuum, making it more suitable for pulse compression applications. To date, there have been very few demonstrations of ANDi on-chip SCG [3] , [4] . Here, we demonstrate fiber-pumped short-wave infrared SCG in a dispersion engineered uncladded Si 3 N 4 waveguide with extreme polarization sensitivity. Specifically, the waveguide is designed to support both ANDi and anomalous pumping SCG depending on the initial polarization of the pump field (transverse magnetic (TM) or transverse electric (TE) polarization). The uncladded design further offers the possibility to deposit a substance directly on-chip making it a promising approach for sensing.

Wavelength conversion of ultrashort laser pulses towards mid-IR range has been recently demonstrated by means of chirp-assisted stimulated Raman scattering (SRS) in gas-filled hollow-core fibers (HCF) [1] – [3] . These advances open up new opportunities for the development of the fiber-based mid-IR supercontinuum (SC) sources, since efficient SRS could significantly improve the spectral power density in the mid-IR spectral range.

Efficient supercontinuum generation and Raman red shifting for sub-300 fs pulses was recently demonstrated in hollow core fibers [1] , [2] . Theses sources serve as a highly promising alternative to optical parametric amplifiers and as wavelength tunable sources for further frequency conversion into deep UV or THz ranges. Yetterbium based systems, due to their high average power are of key interest to drive the supercontinuum generation. Alternative approach relying on gas filled multi-pass cells (MPC) showed strong spectral broadening via self-phase modulation in a single or cascaded stages [3] . The technique uses a pair of concave mirrors arranged in a resonator like geometry with a nonlinear medium in between, either bulk or gas. It proved large compression ratios as well as high peak and average power scalability reaching down to sub-20 fs durations [4] and 1 kW average power [5] . In this work, for the first time, we demonstrate experimentally that super continuum generation strongly enhanced by rotational nonlinearities can be efficiently driven in a single MPC. This kind of systems hold promise to serve as a next-generation power-scalable, simple and stable approach of wavelength tuning and supercontinuum generation for a plethora of applications.

We experimentally report on the spectral broadening of lasers pulses from an Ytterbium (Yb) laser with 0.8μJ energy level via self-phase modulation in cadmium sulfide (CdS) semiconductor crystal. Then, by frequency doubling the pulse into a beta barium borate crystal (BBO), the spatial beam profile is cleaned out while maintaining the bandwidth enhancement. At the end, the laser pulse, around 512 nm, is compressed up to 45 fs using two standard transmissive grating resulting in a pulse compression factor of 6.13.

An investigation of nonlinear optical effects in a photonic crystal fiber was performed to obtain spectral broadening of 1030 nm and 250 fs duration pulses. The effect of fiber length was investigated at different pulse energies and durations. Spectral broadening of 230 nm was achieved with a pulse energy of 218 nJ.

Supercontinuum (SC) sources based on nonlinear spectral broadening of ultrashort pulses in specialty optical fibers have become an indispensable tool in a diverse range of application fields such as biomedical imaging, precision spectroscopy, and ultrafast photonics. After the initial race to maximize their spectral bandwidth and coverage, the noise properties of SC sources have recently shifted into focus. The large pulse-to-pulse fluctuations and correspondingly large relative intensity noise (RIN) present in commercial SC sources strongly limit the sensitivity, precision or resolution of many applications [1] . Recent work suggests that the dispersion engineering of the nonlinear fiber plays an important role for the resulting SC noise properties, as all-normal dispersion (ANDi) fibers suppress the noise-amplifying nonlinear processes that typically dominate the spectral broadening in their anomalously pumped counterparts [2] , [3] .

We study the intensity and phase noise as well as the spectral broadening of narrowband single-frequency fiber laser emission after amplification in a polarization-maintaining Raman fiber amplifier. The optical properties of a 10-W 1266-nm single-frequency source, based on a single-frequency low-noise fiber laser amplified in a Raman fiber amplifier, will be presented.

Filamentation of femtosecond optical vortex at central wavelengths of 1800 nm and higher in fused silica and fluorides was experimentally and numerically studied. The features of broadening of frequency and frequency-angular spectra are considered.

We present a scheme for pulse compression which consists of stage with negative cascading quadratic nonlinearity, which is effectively negative cubic, followed by stage with normal cubic nonlinearity. Compression of 20 fs pulses down to 6 fs is demonstrated by numerical simulation. Study of focusing and coherent combining of such pulses is presented.

In this paper, we firstly study the effect of Doppler broadening on deposited energy and Compton scattering angle for the 4.44 MeV gamma-rays. We measured and fitted the models of the Doppler broadening effect, as well as compared them with theoretical calculations. Then we presented two kinds of approaches (i.e. unidirectional and bidirectional) to correct the Doppler broadening effect via OE with different sampling distribution. Finally, we evaluated the performance of these correction methods for prompt gamma imaging. The corrections with Lorenz distribution of methods □ and □ (i.e. unidirectional and bidirectional) have the best performance on the Doppler broadening correction compared with Gaussian and uniform distribution. Both the calculated model obtained by theoretical formula and fitting model based on the measurement data can realize effective correction. After the correction, the signal-to-noise ratio of the reconstructed image is improved, and the peak prediction is closer to the true value. Moreover, the results of the Doppler broadening correction have a narrower full width at half maximum, which is corresponding to the expected theoretical prediction. The results show that the Doppler broadening correction can further reduce the inherent point spread function of the Compton reconstruction, and improve the accuracy of the reconstruction.

A multiple octave-spanning supercontinuum generation spanning 400-5000 nm is generated in a Xe-filled hollow-core anti-resonant fiber pumped at 3 μm with pulses of 100 fs duration and 15 μJ energy. A significant reduction of the spectral broadening under tight bending conditions has been demonstrated.

New methods for the analysis of electrically-evoked compound action potentials (eCAPs) are described. Mammalian nerves tend to have broad multi-modal distributions of fibre diameters, which translates into a spread of conduction velocities. The method of velocity selective recording (VSR) is unable to distinguish between this spectral spread and the transfer function of the system. The concept of the velocity impulse function (VIF) is introduced as a tool to differentiate between these signal and system attributes. The new methods enable separate estimates of velocity spectral broadening and signal-to-noise ratio (SNR) to be obtained.

Optical fiber networks form the major part of the modern telecommunication infrastructure and are carrying over 95% of the Internet digital data. The increasing demand for higher data rates has promoted the development of higher-order modulation formats, denser wavelength multiplexing and more powerful signal processing. With the entire suppression of the chromatic dispersion, the polarization mode dispersion and the laser phase noise using digital signal processing, Kerr fiber nonlinearities become the main capacity barrier for optical communication systems, and digital back-propagation is generally applied to overcome such distortions. Kerr effects also result in the broadening of spectra of signals propagating in optical fibers, owing to intra- and interchannel nonlinear interactions. This report shows that the spectral broadening effect is crucial for achieving an optimal performance of the digital back-propagation in wideband optical transmission systems.

In this paper, simulated and experimental supercontinuum (SC) generation in the chalcogenide commercial fiber IRF-S-5 are presented. A spectral broadening in the $1.9-3\ \mu\mathrm{m}$ wavelength range is obtained with an average power up to 180 mW by pumping the fiber set-up with a femtosecond Cr:ZnS laser operating at the wavelength $\lambda=2.36\ \mu\mathrm{m}$ . The reported results are obtained by varying the input pulse energy $\mathrm{E}_{0}$ from 0.24 nJ to 1.23 nJ with a repetition rate $R_{\mathrm{R}}=144.55\ \text{MHz}$ and a pulse duration $T_{0}=100\ \text{fs}$ . Experimental results are compared with the results of simulation obtained via a numerical model that takes into account the fiber geometrical parameters, the dispersion behavior, the nonlinear parameters and the input pulse characteristics. A good correspondence is found.

International audience; We aim to demonstrate the first mise en pratique of the kelvin, within the framework of the new International System of Units (SI), with a direct link to the atomic fountains of LNE-SYRTE laboratory, the French time-frequency metrology institute. The Boltzmann constant being fixed in the new SI, the mise en pratique of the kelvin can be cast as a frequency measurement. One method considered, Doppler Broadening thermometry (DBT), originally proposed and initially demonstrated at Laboratoire de Physique des Lasers is being developed by us and others. We report our recent efforts towards implementing DBT with a direct link to the primary frequency standard in the 300-430K range to demonstrate its potential.

We investigate the effect of temperature on the supercontinuum generation of CS2 single-hole liquid core optical fiber (LCOF). By solving the generalized nonlinear Schrödinger equation, the effect of the temperature on spectral broadening, coherence, soliton effects, and modulation instability was analyzed. The results turn out that the supercontinuum generation by pumping the LCOP with 1.95-μm laser is robust over a broad temperature range (15-45 °C). The optimal temperature to generate the broadest spectrum of SC via LCOF is 15 °C, which is independent on the pump power. The coherence of the generated SC is independent on the temperature at low pump power (such as P0 = 4 kW), while at a high pump power (such as P 0 = 40 kW) the coherence of the long wave (around 2600 nm) tends to be deteriorated at low temperature (e.g., 10 °C). The underlying reason for such phenomenon has been discussed as well.

Accurate estimation of maximum velocities in Doppler blood flow spectrograms is vital in diagnosis of cardiovascular anomalies such as stenosis. However, obtaining accurate maximum velocity is not straightforward due to intrinsic spectral broadening and variance in the power spectrum estimate. Therefore, employment of Doppler indices such as pulsatility index is commonplace. Moreover, the results attained by conventional blood spectrogram estimators fail when beam to flow angle is near to 90°. In this work, an approach for estimation of pulsatility index for a fully transverse blood flow is presented. The method has been validated using simulated Doppler signal with scatterers possessing the pulsating flow of femoral artery in Womersley-Evan’s model. By investigation of three scenarios of a normal vessel, and vessel with 20% and 40% stenosis, the possibility of stenosis diagnosis is discussed.

Airborne passive radar is faced with doppler broadening direct wave and strong clutter interference, moreover, the random range sidelobes coupling of the strong clutters seriously affects the performance of moving target detection. In this paper, for above problem, the extensive cancellation algorithm batches versions (ECA-B) is used to cancellation the airborne passive radar clutter at the same time. In order to solve the problem of high computational complexity and high hardware storage requirements of ECA-B algorithm, the fast Fourier transform (FFT) is used to process in frequency domain. And the computational complexity of matrix inversion is reduced by using the symmetric Toeplitz matrix. The simulation data show that: under the same simulation conditions, the frequency-domain ECA-B algorithm can effectively cancellation the airborne passive radar clutter, improve the target signal-to-noise ratio and the target detection performance, while reduce the algorithm complexity and improve the real-time processing capability of the system.

The latest version of general-purpose file JENDL-4.0 was released in 2010 with improving fission product and minor actinide data as well as covariance data to meet needs of innovative reactors. After the release of JENDL-4.0, efforts to extend energy range up to 200 MeV have been made in order to meet needs from various accelerator applications such as ADS. They were complied with proton induced reaction data into JENDL-4.0/HE and released in 2015. We also released two special purpose files. One is JENDL/PD-2016 which contains the data of photo-nuclear reaction cross sections covering a wide area of the nuclear chart. The other one is JENDL/AD-2017 which provides production cross sections of radioactive nuclei by neutrons. Regarding the general-purpose file, we are planning to release the next version of JENDL, which would be made available in FY2021 as JENDL-5. To contribute to the important problems concerning the nuclear energy in Japan such as nuclear waste and nuclear safety as well as various application fields of nuclear data, it is intended to raise reliability of simulation calculation of nuclear reactors and neutron transportation in structure materials. Nuclear data processing is an important interface between an evaluated nuclear data and nuclear transport calculation codes. The transport codes require cross section library which is generated by a nuclear data processing system. The nuclear data processing system is not just a converter. It performs many processes, e.g., linearization, reconstruction of the resonance region, Doppler broadening, etc. We developed a new nuclear data processing system FRENDY to process evaluated nuclear data files. FRENDY version 1 was released in March 2019 as an open source software under the 2-clause BSD license. FRENDY can be downloaded from JAEA’s web site.

In this paper, we presented a monolithic high power random fiber laser with a tandem-pumping scheme. The random fiber laser seed centered at 1064 nm was pumped by a 1018 nm fiber laser and reached an output power of 8.66 W. After the two amplifiers, the seed was amplified to 4020 W and the 3 dB width of laser spectrum was broadened from 0.78 nm to 0.99 nm. To increase the efficiency of the 4 kW laser, a specially designed high absorption coefficient (0.6 dB/m @ 1018nm) 30/250 μm Yb doped fiber and high brightness 1018 nm pump laser were adopted. The optical-to-optical efficiency could reach as high as 88.5%. No obvious mode instability was found during the experiment. Furthermore, with the same amplifier stage, we have also built another 4 kW level fiber laser seeded by an oscillator consisting of a pair of fiber Bragg gratings. The comparison of the experimental results showed that high power laser based on RFL seed would have a better control on spectral broadening.

Silicon-on-insulator waveguides covered with graphene in length of 60 μm have been fabricated and investigated. Self phase modulation were carried out, and enhanced nonlinear phase shift, Kerr nonlinearity and figure-of-merit have been confirmed in graphene-silicon waveguides by experiments and simulations.

Recently, femtosecond pulse shortening techniques applicable for high peak power (>1 PW) lasers have been an object of intense research [1]. It is motivated by pushing the frontiers of the peak power. Another reason is increasing the temporal resolution in pump-probe experiments. The temporal contrast of these lasers is extremely high (>107) and preserving it after the pulse shortening is still a challenge. The common approach for femtosecond pulse shortening includes spectral broadening in a third order nonlinear medium and residual chirp compensation in a pulse compressor. An alternative approach is based on cascaded second order nonlinearities. Although both methods have been successfully demonstrated [2,3], there is no comparison of their capabilities to preserve the temporal contrast. In this report, we compare these two techniques: considering two criteria: the output pulse duration and the temporal contrast.

Ever shorter, more broadband and higher peak power laser sources at high repetition rates are in strong demand, as they can benefit researches such as nonlinear spectroscopy. The all-bulk spectral broadening approach, arranged for example as a multi-pass cell, has seen rapid development recently. The scheme involves self-phase modulation (SPM) and compression of high-power laser pulses, either emitted from MOPA-type systems [1] or oscillator devices [2]. In contrast to waveguides, the separation of nonlinear and dispersive elements allows extensive control over the net dispersion, making it particularly interesting for the realizing of different nonlinear effects, such as soliton Raman self-frequency shift (SRSFS) effect. In comparison to the symmetric broadening of SPM, SRSFS can generate a power-tunable, asymmetric red-shift [3]. This technique has frequently been applied in optical fibers at a nJ-level of pulse energy. Here we present, to the best of our knowledge, the first implementation of this process to high-energy (>1 μJ) pulses in a Herriott-type multi-pass cell (HC).

Frequency combs are used in optical communications, arbitrary RF waveform generation and astronomical spectrograph calibration [1]. Combs with tunable repetition rate can be generated by electro-optic modulation of lasers but are limited in bandwidth (BW) [2]. Wide band combs can be generated by cascaded four-wave mixing with two lasers in highly nonlinear fibre (HNLF) [3], but the spectral spacing of the two pumps determines both bandwidth and repetition rate limiting the independent control of the two. Spectral broadening of electro-optic combs can achieve tunable repetition rate and wide bandwidth. However, the temporal and/or spectral profile is not suitable for spectral broadening. Different methods have been used such as pulse compression in SMF [4] and pulse shaping prior to nonlinear broadening. However, such open loop methods resulted in poor flatness in the central region of comb and are not robust to varying operating and drive conditions of the modulators, initial spectral phase profile, modulator bias, length, dispersion profile of HNLF and power. In this work, we utilise an adaptive spectral phase optimization based on adaptive step size random search algorithm in a closed loop to optimize the spectral phases to obtain the broadest comb. We demonstrate bandwidth scaling by over 13 times of an initial 25GHz comb with 9lines (0.2THz) to 121 lines (3THz).

The first nonlinear effect that appears when an ultrashort laser pulse propagates through an amorphous medium, such as a gas or a glass, is related to an intensity-dependent refractive index: the optical Kerr effect. One of the consequences of the Kerr effect is the self-focusing of the beam, which takes place for input peak powers above a threshold value, the critical power (P CT ) [1,2]. Another consequence of the optical Kerr effect is the spectral broadening related to the self-phase modulation. This spectral broadening is used to generate ultrashort pulses in post-compression schemes, most of them based on the nonlinear propagation of a pulse inside a hollow-core fiber (HCF) [3].

The higher-order solitons are the soliton pulses with the energy higher than that of the fundamental soliton. Although the higher-order solitons' temporal and spectral shapes are not constant, such pulses reproduce their shapes after a soliton period. Previous research assumed that higher-order solitons and high peak power pulses could be obtained in femtosecond Tm-doped all-fiber lasers. For many of these works, the higher-order dispersion and nonlinear effects lead to a pulse break up and higher-order solitons are not able to reproduce their shape during the propagation in the fiber. Such effects as stimulated Raman scattering, self-steepening and higher-order dispersion appear in the cases of a propagation of high energy pulses and can lead not only to a spectral broadening, but also to a soliton fission. That is why it is important to model pulse dynamics in the external fiber and carefully conduct the measurements of the pulse parameters, especially autocorrelation trace and spectrum, while registration of higherorder solitons. In this work, we study the propagation of higher-order solitons assuming the presence of higher-order effects and their absence near $1.9\ \mu \mathrm{m}$ in the SMF-28 fiber, as it's a common waveguide for pulse parameter measurements.

Spectral broadening and pulse compression of ultrashort pulses from high peak power lasers (10 MW-GW) is usually performed in gas-filled, photonic bandgap or Kagome-type hollow core fibers [1]. Recently, a new method relying on a waveguide-like periodic structure consisting of focusing elements (curved dispersive mirrors) and nonlinear media was demonstrated [2, 3]. Both all bulk and gas filled geometries [4, 5] were successfully realized in high average and peak power regimes pushing the broadened spectrum to sub-10 fs Fourier limit [6]. In contrast to fibers, the sign of the overall net dispersion and its profile including higher order dispersion terms can be readily engineered in the quasi-waveguide approach. This advantage becomes increasingly interesting at high peak power levels in combination with nonlinear effects such as soliton supercontinuum generation, soliton self-frequency shifting and self-compression which has been shown for example in [7], were self-compression in a multi-pass cell has been demonstrated at a wavelength of 1550 nm with a pulse compression ratio of 2.8. We report on self-compression in an all bulk multi-pass geometry at a wavelength of 1030 nm by utilizing dispersive mirrors for operation in the net-negative dispersion regime, yielding a temporal compression factor of just below 10.

Dual-comb spectroscopy (DCS) has experienced a remarkable growth propelled by recent advances in fiber and semiconductor technology. A key enabler for DCS is mutual stability between two repetition-rate-mismatched frequency combs, which conventionally relies on feedback loops employing additional optics and electronics to synchronize a pair of sources. This limitation can be overcome in several ways, of which one of the most promising is the generation of two frequency combs in a single free-running fiber laser cavity [1–3]. To date, however, the only configuration proven to be compatible with Doppler-limited spectroscopy was the dual-color laser presented in Ref. [2] requiring external spectral broadening, thus adding an extra layer of complexity.

Supercontinuum (SC) sources are nowadays widely used as fundamental building blocks in systems enabling a multitude of technologies such as optical frequency metrology, optical coherence tomography and broadband spectroscopy to name a few [1]. Highly nonlinear fibers (HNLF) make up an excellent platform for SC generation as they not only offer high nonlinearity, but also dispersion-engineering and broadband single-mode operation. Broad SC are easily obtained by injecting ultrashort pulses into the anomalous dispersion regime of the HNLF, where the initial propagation is characterized by temporal compression and spectral broadening through solitonic effects. However, after compression each launched pulse tend to fission into several pulses, a process that is very sensitive to noise and can therefore lead to increased shot-to-shot fluctuations [2].

Efficient, diode-pumped high energy femtosecond laser systems around 1 μm based on Yb-gain media are readily commercially available. However, owing to the gain bandwidth limitations, the pulses generated in such lasers are substantially longer than the ones generated in Ti:Sapphire systems. A simple, energy-scalable pulse self-compression scheme for the pulses around 1 μm thus would be of great interest for many applications, including time-resolved pump-probe spectroscopy, high-harmonics generation, etc. The self-phase modulation during nonlinear propagation in filaments in gasses is often employed for pulse self-compression [1,2]. Such schemes typically require rather bulky setups and careful control of group velocity dispersion. Some years ago it has been shown theoretically that Raman-active molecules in gaseous form could be used for mid-infrared pulse compression [3].

Nonlinear spectral broadening is widely used for applications like attosecond pulse generation, XUV sources, supercontinuum generation or ultrafast spectroscopy. In these cases the laser is typically spectrally broadened and compressed to reach pulse durations much shorter than originally available. For pulse energies in the millijoule regime, spectral broadening is routinely implemented using capillary optical fibers [1] with remarkable results in terms of output pulse durations and beam quality. In addition to fiber-based approaches, the application of multipass cells for spectral broadening [2] recently attracted increasing interest. These cells are capable of preserving a nearly Gaussian mode in the presence of nonlinear effects without a spatial filter or optical fiber, they are robust to input pointing, have a high throughput of more than 90% and can be used with high average powers and energies. In a multipass cell the nonlinear interaction with either gas [3] or a thin solid [2] provides a small amount of spectral broadening within each pass. Based on this concept we recently demonstrated output energies of almost 18 mJ by guiding near infrared pulses with a pulse duration of 1.3 ps and a repetition rate of 5 kHz through an argon filled multipass cell [4]. We obtained a broadening factor of over 33 and could show compressibility down to 41 fs. The energy was limited by the breakdown intensity of argon and the damage threshold of the used optics.

Microresonator-based frequency combs have potential to achieve on-chip frequency metrology. To obtain high-power comb lines at f and 2f for self-referenced stabilization, we evaluate the feasibility of building a fully integrated comb generator with spectral broadening. Comparing two types of nonlinear waveguides based on Si and chalcogenides, we find that the system feasibility relies on integrated waveguide amplifiers. If with a high-gain amplifier, a dispersion-engineered chalcogenide waveguide is preferred. Otherwise, an Si waveguide performs better. Considering the immaturity of both on-chip amplification technologies for ultrafast pulses and hybrid integration of multiple nonlinear materials, it could be concluded that in near future it is not practical to achieve fully on-chip f-2f self-referenced comb stabilization via subsequent spectral broadening of a Kerr frequency comb. Alternatively, direct generation of octave-spanning Kerr combs via dispersion engineering still remains attractive.