Optical components produced with inverse design for compact and less power consuming electro-optical modulators

The use of electro-optic modulators is limited in size and efficiency. Waveguide length, which is among the performance criteria of electro-optic modulators, and constructive and destructive interferences created by the electric field applied to this waveguide should be examined with electro-optical effects, the most important feature of these devices. Mach-Zehnder Modulator, one of the known electro-optic modulators, can be examined in this context. Mach-Zehnder Modulators can be formed from Y-junctions. It is important to be able to shorten the arm length of the Y-junction without compromising the performance of the Y-junction structures. Likewise, the required activation voltage will change as the size changes. With the decrease in the surface area, the power consumption will also decrease and thus more profitable structures can be produced in terms of cost. At this point, designs should be able to be taken to more reasonable levels in terms of size and power. The resulting performance results were analyzed with photonic designs made by intuitive method. However, photonic systems designed by trial and error using heuristic methods can be inadequate in terms of performance and cannot be smaller than a certain size. Today, the variety of designs that can be made can be increased by developing photonic designs with reverse design methods. Devices with levels similar to the performances obtained by heuristic methods can be designed in smaller dimensions and less power consumption. For this, it is necessary to change the optical and geometric parameters and change the design style by making a difference in the configuration. By making topology changes, more successful structures in terms of geometry can be achieved. By releasing the refractive index distribution, by observing the changes on the electric field distribution, the desired levels of performance can be reached in desired dimensions with iterative algorithms. Normally, the solution set of designs that can be made with the heuristic method is limited by the parameters, on the contrary, the parameter space of the designs prepared using the inverse design topology is much more. Therefore, within the scope of the thesis, the basic functions of electro-optic modulators have been examined, and the reduction and efficiency problems that arise here have been solved by the inverse design method. Inverse design methods are examined and the difference between objective-first algorithm and SPINS, which are two of these methods, is explained. Two different structures were designed with the SPINS used in this thesis study. One of the structures designed in the studies was produced with a 3D printer and conducting an experiment, promising results were achieved. These designs can pave the way for different configurations and successful, compact and low power consuming designs in the modeling of electro-optic modulators.