The main objective of this thesis was to investigate how it is possible to extend the master curveapproach to complex components, in order to calculate annealing conditions and giving a life timeprediction according those annealing conditions. We studied the stability of a simple Bragg grating (noapodisation, constant pitch, no tilt). Then, we have been able to study the problem of the stability(according to specific criteria) of complex components: CDC and GFF monofilters. This thesis allowed toshow precisely what is the VAREPA framework and how to apply it to optical telecommunicationsproblems. To give an answer to the question of the possibility of extending the master curve approach tocomplex components, series of accelerate aging studies have been carried out. For this purpose, ameasurement protocol including the determination of the measurement uncertainty was developed. Thisprotocol allow, to determine confidence intervals on the life time prediction.We studied the kinetics of photo induced index growth in the optical fiber dedicated forfabrication of CDC (Coreactive). This study allowed us to demonstrate that the predominant reaction atthe origin of changes in refractive index is activated by a one photon absorption. On the other hand wehave extended the VEREPA framework used to predict the Bragg gratings erasure to take into account thegrowing kinetics. In a second step we tried to calculate annealing conditions in the presence of residualmolecular hydrogen and associated Bragg grating life time prediction. For that purpose we established themaster curve of photo-induced Bragg gratings without annealing, then we validated the annealingconditions and lifetime predictions by building a new master curve from aging of annealed Bragg gratings(in presence of residual hydrogen).We studied the thermal stability of hydrogenated fiber Bragg gratings dedicated to futuremonofiltres GFF in similar conditions to those components. We used master curves from basic Bragggratings (ie no apodisation, constant pitch, no tilt) simply out-gazed at "cold" temperature (here 2 days at50 ° C and 2 days at 110 ° C). Then the aging of the index profile of a GFF monofilter have beensimulated (index profile of the core and of the cladding, each of them having a different master curve).For that purpose the profile have been separated into 35 "basic" filters. Then the spectral response of thisaged monofilter (e.g. 30 days at 200°C) have been calculated and compared with experimentalmeasurements. The advantage of this approach is that it is general and allow to change the annealingconditions if needed. Firstly, this study allowed to highlight that the stability of photo-induced indexchanges in the core depend little on the initial amplitude modulation for values between several 10-5 and10-3. Secondly, due to the presence of a highly sensitive optical cladding (and therefore highly Ge doped),Master curves, of photo-induced index change in the core and the cladding have been established forvalues ranging up to 10-3 in the core and 3.10-3 in the cladding. Master Curves with the same k0 havebeen obtained but the elbow and the slopes are different. Stability of index variations in the core is better(less Ge doped) than in the photosensitive cladding. In a second step, from the basic Bragg gratings, theannealing conditions and the lifetime have been determined. For the first time to our knowledge, thereliably of the annealing conditions and the life time predictions have been validated by building a newmaster curve on annealed GFF monofilters. According to the theory, this master curve show the same k0as the one determined using non-annealed gratings. More beyond the elbow, the two curves are same,which validates the reliability of our predictions.
