We reported on the generation of 99.8 fs, 25 kW peak-power, dispersion-managed pulses directly from a passively mode-locked Yb-fiber laser oscillator with a figure-of-9 configuration. The introduction of strongly injected pump power and optical components with a high damage threshold enables high-power operation, while the polarization-maintaining (PM) fiber supports environmentally stable self-started mode-locking. Mode-locking in the soliton-like and negative-dispersion regime is characterized by the dispersion management via tuning the separation distances between a pair of gratings inside the cavity. The oscillator generates stable pulses with up to 40.10 mW average power at a 16.03 MHz repetition rate, corresponding to a pulse energy of 2.5 nJ. To the best of our knowledge, it is the highest peak-power directly obtained by a laser oscillator with a figure-of-9 configuration.

The multilayer defects of mask blanks in extreme ultraviolet (EUV) lithography may cause severe reflectivity deformation and phase shift. The profile information of a multilayer defect is the key factor for mask defect compensation or repair. This paper introduces an artificial neural network framework to reconstruct the profile parameters of multilayer defects in the EUV mask blanks. With the aerial images of the defective mask blanks obtained at different illumination angles and a series of generative adversarial networks, the method enables a way of multilayer defect characterization with high accuracy.