Ultrafast temporal spectroscopy based on parametric mixing time-lens

Abstract

With the increased requirement on the ultrafast measurement technology, how to resolve the spectral dynamics has the top priority on the research list, since spectrum is an essential carrier for most of the physical or chemical phenomena. According to the Nyquist-Shannon sampling theorem, if a function 𝑥(𝑡) contains no frequencies higher than 𝐵 Hz, it is completely determined by giving its ordinates at a series of points spaced 1/(2𝐵) seconds apart. Since most of the conventional optical spectrum analyzers (OSAs) are operated with the sampling rate (or frame rate) of 5 Hz, it results in the resolvable bandwidth of the spectrum dynamic is less than 2.5 Hz. With the development of the space-time duality, the analogy transforms the conventional spatial dimension into the time axis, and the well-known spatial models inspire us in performing their counterparts in the time domain. As one of the most powerful tools in achieving ultrafast time axis information, time-lens plays a more and more important role in the single-pixel imaging system.

By fully analyzing the diploma of the previous optical spectrum resolving mechanisms, in this thesis, for the first time, we raised up the concept the parametric spectrotemporal analyzer (PASTA), which is based on the time-lens focusing mechanism. Here the spectrum resolving frame rate is increased to 100 MHz, then the observable spectrum dynamic bandwidth could be 50 MHz, which is sufficient for most of the ultrafast phenomena. In the PASTA system, the time-lenses are implemented with the fiber optical parametric amplifier (FOPA) based parametric mixer, which provides higher conversion efficiency and repetition rate. On the other hand, the dispersion based dispersive Fourier transformation (DFT) technology generates the swept-pump for the FOPA, as well as the temporal dispersion medium.

This research in this thesis is a fundamental study on the newly PASTA system, from its origin and the theoretical background, to the implementation techniques and operation performances. From its implementation, its principles are strongly related with the combination of the dispersion and the Kerr nonlinear effects, especially the swept-pump FOPA in the time-lens part. The DFT technique, in generating the fast swept-source, has also find its applications in the ultrafast serial time-encoded amplified microscopy (STEAM) and swept-source optical coherence tomography (SS-OCT) systems.

Finally, the single-lens PASTA prototype is capable of resolving 5-nm wavelength range with 0.03-nm resolution under 100-MHz frame rate. Moreover, besides the singlelens PASTA, the telescope/wide-angle configurations have also been investigated experimentally to achieve the spectrum zoom in/out ratio as high as 17 times, here we have obtained the sharpest resolution of 5 pm (<1 GHz) with the telescope configuration, and the widest observation range of 9 nm with the wide-angle configuration.

My research efforts presented in this thesis mainly leverage the ultrafast characteristics of the time-lens system, from theory to implementation, and achieve the real-time optical spectrum analysis – the PASTA system. PASTA is not only essential in observing some non-repetitive ultrafast phenomena, but also provides a potential solution for the frequency to time transformation in some ultrafast bio-medical imaging systems.