Supercontinuum generation is an attractive method for the production of extremely short optical pulses with coherent spectra extending far beyond the bandwidth of any known gain material.With bandwidths exceeding one optical octave, compression of these white-light continua to single-cycle pulse durations is obviously a challenging prospect. However, the multitude and the complexity of nonlinear optical mechanisms may give rise to strongly varying spectral phase or amplitude of these continua, thwarting any attempt of their compression to a single short pulse.
Among the different options for white-light generation, noble-gas filled hollow fibers still offer the simplest nonlinear optical environment. Raman processes are absent inside the noble gas. The dispersion of the gas is positive, and linear and nonlinear contributions of the waveguide are widely absent. For compression to a short pulse, a dispersive delay line is needed that compensates for all linear contributions to the dispersion (e.g. by the cell windows) and also for nonlinear contributions to the spectral phase from self-phase modulation. For this purpose, we use specifically designed chirped mirrors, which cover a wavelength range of about 500 – 1000 nm, i.e. one optical octave.
The working principle of these mirrors and a 4.3-fs pulse compressed with this technique are shown in fig. 1. For measuring the short pulse, we employ spectral phase interferometry for direct electric-field reconstruction (SPIDER). In the future, it is planned to explore new generations of chirped mirrors or adaptively compress these white-light continua for even shorter pulse durations. Nonetheless, the currently reached pulse duration of 4.3 fs corresponds to only about 1.5 cycles of the optical field, which is close to the shortest laser pulse duration ever generated.
Supercontinuum generation (SC) in microstructure fibers (MF) is another method for white-light generation, which is currently a topic of great interest, because of its unique properties, such as the more than two octave broad spectrum generated by pulses with only nJ energy. Due to these properties, SC in MFs is very interesting for applications such as frequency metrology, femtosecond phase stabilization, optical coherence tomography, spectroscopy, pulse compression and as novel light source. The physical mechanism of its generation was theoretically eludicated in this project group of the MBI, and also the first experimental evidence for this novel mechanism has been provided here. Pulse propagation in MFs in the anomalous dispersion region leads to the formation of several fundamental solitons with different central frequencies, which emit blue-shifted phase-matched non-solitonic radiation. An experimental evidence is the fact that longer input pulses yields a broader spectrum than shorter pulses with the same input intensity (fig. 2). This is in direct contrast to SC generated by self-phase modulation in hollow fibers, where shorter pulses yield a broader spectrum. In addition to these results, we studied also other effects in MFs, such as four-wave mixing, comb generation in multicore MFs, and SC in planar rib waveguide, and in highly nonlinear MFs with two zero-dispersion wavelengths.


Recently also another new method to generate single intense supershort pulses, using molecular phase modulation has been proposed and realized at the MBI. This method, named molecular phase modulator, gives a possibility to control a phase imposed to the pulses as well as theit frequency and chirp. Based on this method pulses as short as 3.8-fs at 400nm were in SF6. More recently we have applied the same technique also to pulses in the ultraviolet (266 nm).

Photonic crystals are an another class of microstructured materials which has recently attracted much attention as they can be tailored by design to achieve novel physical properties such as the control of the flow of light on the wavelength scale or an effective negative refraction indices. This lead to a number of new interesting linear and nonlinear phenomena and enable many applications in science and technology. In this project group it was shown that light beams can be focused below the diffraction limit of half of the wavelength by the combination of two main elements: an aperture which creates week seed evanescent waves and a photonic crystal with negative refraction which amplifies the evanescent waves (fig. 3). Besides we have shown that focusing of scanning light beams below the diffraction limit is possible without near-field spatial control by the use of a light-controlled saturable absorber which creates the seed evanescent waves from the beam and a layer of negative refraction material which amplifies the evanescent waves (fig. 4).