Fiber Optical Prametric Amplifiers with Pump Resonators

Fibre OPAs are a new type of optical amplifier, which can boost the power of optical signals travelling in optical fibres. They have several advantages compared to existing optical amplifiers, such as Erbium-doped fibre amplifiers (EDFAs), and Raman amplifiers (RAs). In particular, they can amplify a wider range of wavelengths, which is important for the development of high-capacity communication networks suitable for accommodating the ever-increasing Internet traffic. OPAs are based on the utilization of the third-order nonlinearity of fibres. If an electric field E is injected into the fibre, then at the output there appears a term which is proportional to the cube of the input filed, i.e. to E3. If the input field consists of two monochromatic waves at the frequencies f1 and f2, at the output there will appear terms at new frequencies, and in particular at 2f1 - f2, symmetric of f1 with respect to f2. One generally arranges to have one input wave much stronger than the other one. The former is called the pump, and the latter the signal. If the pump is at f1 and the signal at f2, then the wave at 2f1 - f2 is called the idler. The longer the fibre, the larger the output idler is. Signal and idler grow at exactly the same rate along the fibre, and if the pump is strong, and the fibre is long and highly nonlinear, this arrangement leads to amplification of the signal. The gain can be adjusted, and can easily reach 100 or 1000. Such gains are sufficient to provide amplification for modulated signals in practical optical communication systems.An important feature of fibre OPAs is that they can amplify signals with widely different wavelengths. While EDFAs are limited to a bandwidth of the order of 40 nm, OPAs can provide gain over as much as 400 nm. However, a disadvantage of current fibre OPAs is that in order to achieve such gains and bandwidths, pump powers of the order of 10 Watts are required. This is much larger that the pump power required for EDFAs, and is not suitable for practical applications, for reasons of cost, safety, and reliability.Therefore, to bring fibre OPAs closer to practicality, it is desirable to reduce the pump power, while maintaining high gain and large bandwidth. At first sight it appears that these are conflicting requirements, so that there is little hope of fulfilling them simultaneously.In this proposal we propose to meet these goals by modifying the basic structure of the fibre OPA, by adding an optical resonator. The basic idea is that if a pump laser with modest power is incident on the resonator, with a frequency which matches that of the resonator, it will excite within the resonator a wave with an intracavity power far larger than that of the pump wave outside the cavity. The use of resonators in nonlinear optics is not new. To our knowledge, however, little work has been done for fiber nonlinearities. One reason for this is that in fibres the cavity resonances are very narrow, and one needs a narrow-linewidth laser in order to efficiently excite them. Maintaining the resonance condition is also a challenge, and that must be accomplished by means of some feedback system.In the Case for Support we present in detail the design of the cavities that we plan to use. They consist of two fiber Bragg gratings (FBGs) acting as mirrors, and placed at the two fibre ends. A major part of the work will consist of the design, fabrication and testing of FBGs and cavities built with them.The work is challenging because of the many obstacles that need to be overcome in order for this approach to succeed, as described in detail in the Case for Support. However, we believe that by attacking these problems as outlined, we should be able to demonstrate significant improvements to fiber OPAs by the end of the programme. We therefore hope that this work will bring fibre OPAs much closer to being used in future fibre communication systems.