Super compact and efficient tripler
We have just published a paper on highly efficient and compact cascade third harmonic generator (tripler). It is 3 times more efficient and 1000 times smaller than the previous constructions.
Ultra short ulraviolet pulses are best for precision laser matchining. They are also usefull for ultrafast spectroscopy where chemical reactions are observed in real time. It is, however, not easy to produce ultraviolet light - there are no UV lasers giving ultra short pulses. Therefore, the only way to go is to convert the light of infrared pulses into UV.
The tripler is used for conversion of light color. We start from infrared, convert it to visible green light through summing up energy of the beam photons. Than again we sum up the energy of newly created green light and the IR to produce ultraviolet. This summing up or "nonlinear conversion" occurs in crystals, it is however usually very inefficient. Fortunately the efficiency depends on light intensity. In Laser Centre we have at our disposal ultrashort pulses. Although, these have small energies (around 0.000000004 Joule), they are also very short (0.0000000000002 second), therefore, their peak powers are enormous, additionally if we focus them tightly in the crystal, their's intensity becomes just the right one for generation (gigawatts per centimeter square). Anyway, even with such pulses feeding the previously existing triplers the efficiencies of these devices were at a level of 10%. The devices themselves had sizes similar to that of a lunch box.
What was our idea for improvement? First of all we decided to miniaturize the device. Instead of separating the green and red beams after the first conversion stage we have decided to keep them together. The separation requires quite a few optical elements like lenses and special mirrors that can reflect one and transmit the other color. So why was separation done in previous construction in the first place? As we are dealing with ultra short pulses, focused to small sizes it takes some effort to overlap the beams' spots in the second nonlinear crystal both spatially and in time. The common approach till now was to split the beams and solve the delay and spatial position problems with bulk optics.
Our idea was to fix the pulse delayed and the beam spot position by using additional crystals: calcite - here, as opose to other crystals, the red pulse can travel slower than the green pulse, and highly birefringent YVO crystal - it allows spatial shift of one of the beams with respect to the other. By using the additional crystals we were able to minimize the tripler to the size of a coin..
Well this is all very well, but we did also something more, we knew that when small beams nd short pulses are used the conversion process becomes really complex and difficult to understand. Simple models and intuition can fail us in these conditions. We have therefore created a simulation tool - a 3D pulse propagation software Hussar. And we have used that tool to find the perfect thicknesses of the crystals used in the experiment (see the video at the top of the page for simulation results - apparently pulses can be regarded as colorful balloons).
Want to read more? See our article in: Scientific Reports and the press release on the Physics Department website.