On the origin of core-to-core variations in multi-core fibre Bragg gratings

Fibre Bragg gratings have been demonstrated to be a powerful tool with which to filter atmospheric emission lines from astronomical spectra. Multicore fibre technology has the potential to simplify the fabrication of fibre Bragg gratings, since all cores can be inscribed simultaneously rather than individually which is both time consuming and expensive to do. Solving the multicore challenge has fundamental implications for many fields outside of astrophotonics. To realise a working multicore fibre Bragg grating (MCFBG), all cores must be written with identical gratings providing uniform depth, Bragg wavelength and bandpass. However, to date, all multicore fibre Bragg gratings display a variation in the Bragg wavelength of the central cores compared to the outer cores. This seems to be a property of the multicore fibre itself, and is not due to the Bragg grating writing process. We investigate the origin of these core-to-core variations using finite difference time domain and finite element simulations, combined with analysis of fabricated multicore fibre. We find that the ellipticity of the core, the size of the core, and the coupling between cores all affect the propagation constants. However, the dependence on ellipticity is very weak, and cores would have to be highly deformed in the manufacturing process for this to be a concern. A variation in radius of similar to 2.5% could account for the observed variation in propagation constants. However, the measured variation in the fabricated MCF is too small and does not display any radial trend. The coupling between cores is too small to change the propagation constants significantly, but even if it were significant any effect would be expected increase the Bragg wavelengths of the central cores, the opposite of what is observed.