Non-linear optics is not a new field in physics, and even Maxwell in the mid-nineteenth century was aware that his equations defining electro-magnetic fields could be non-linear. But it was until the advent of lasers, when the necessary density of the electric field could be achieved, that it was possible to study non-linear phenomena. Non-linear effects arise from the equation that defines the polarization density as a function of the electric field, which can be seen as a Taylor expansion with increasing higher orders terms of the electric field multiplied by a diminishing factor, specific to each type of medium. All dielectric media can exhibit non-linear behavior. But only when the electric field amplitude reaches the magnitude of the inner-atomic field can we easily observe and use it.

There are many applications of non-linear behavior, such as: Second Harmonic Generation, Parametric Amplification, Frequency Up and Down Conversion, Sum and Difference Frequency Generation, etc. Sum Frequency Generation can be easily visualized as the absorption by an atom of two photons of frequency f1 and f2 followed by the emission of a photon with the frequency f3 = f1 + f2. The medium that we are proposing to use for this process is a non-linear crystal such as BBO (Beta-Barium Borate) or LBO (Lithium Triborate). These crystals were considered because they are the most efficient at the wavelengths that we will use. The light sources that we will use are a Verdi laser at 532 nanometers with a power output of 10 Watts and a diode laser at 1447 nanometers with P > 200 milli-Watts. The diode laser will be connected to a rare-earth fiber amplifier to increase its power to > 1 Watt. These two beams will then be phase-matched -- a process which maximizes the interaction of the beams and the crystal -- and directed through the crystal.

At this point, we will have to deal with some new problems, such as stabilizing the infrared diode laser (we are proposing to use a fiber Bragg grating for this), maximizing the efficiency of the fiber amplifier, and correctly phase-matching the optical fields within the crystal. This latter task is necessary to reduce the destructive interference of the produced light of frequency f3 which has been created at different points along the beam path. We are expecting to finish addressing the inquiries about the equipment and materials that will be needed in a few weeks. As soon as we receive these, we will begin work on integrating all the parts and then on optimizing the process.

On a side note, we are studying the possibility of using a Highly Non-Linear Fiber (HNLF) instead of a non-linear crystal. The advantage of this method is the high confinement of the optical field within the fiber which will increase by a large factor the efficiency of the whole process. However, there is not much literature available on this process and not all of the common applications available to non-linear crystals are mentioned in the few papers available.

This research was supported by NSF, ONR and AGEP.