Chirped Bragg gratings (CBGs) recorded in photo-thermo-refractive (PTR) glass provide a very efficient and robust way to stretch and compress ultra-short laser pulses. These gratings offer the ability to stretch pulses from hundreds of femtoseconds, to the order of 1 ns and then recompress them. However, in order to achieve pulse stretching of this magnitude, 100 mm thick CBGs are needed. Using these CBGs to both stretch, and re-compress the pulse thus requires propagation through 200 mm of optical glass. This therefore demands perfect control of the glass homogeneity, as well as the holographic recording process of the CBG. In this thesis, we present a study of the CBG parameters that lead to distortions in the quality of diffracted beams. We first present the challenges associated with measuring the quality of these beams and we show that such measurements are not easily achieved using commercial systems that rely on the ISO standard M2 method. Thus, we introduce a new metric of beam quality, which we have coined S2 , that is a combination of both the M2 and power in the bucket metrics. Subsequently, we investigate the influence of the CBG parameters on the quality of diffracted beams. In particular, we examine the impact of small optical heterogeneities known as striae, as well as the impact of the optically and thermally induced distortions in the grating. We then use this data to improve the fabrication and characterization of 100 mm long CBGs. Finally, we characterize the performance of CBGs recorded in PTR for stretching and compression of femtosecond pulses using a custom autocorrelation system. We present data on high quality 100 mm long CBGs and an analysis on the correlation between beam quality and the final pulse duration after stretching and re-compressing the pulse.