Analyzing SCERPA and ToPoliNano Simulators for the Design of Molecular Field-Coupled Nanocomputing Circuits

Moore's Law, proposed by the Intel co-founder Gordon Moore in 1965, forecast that the number of transistors on a microchip would double approximately every two years, paving the way to significant advancements in computing technology, efficiency, and performance. Initially, the projection proved to be highly accurate, describing the rapid progression of the semiconductor industry for over five decades. However, as the physical limitations of current transistor technology have been approached, researchers and engineers have questioned the long-term viability of Moore's Law, prompting the exploration of alternative nanoscale computing approaches, such as the "Beyond CMOS". The paradigm involves the development of unconventional materials and device structures that are beyond traditional silicon technology to favour a continuous evolution of next-generation systems. Molecular Field-Coupled Nanocomputing (mFCN) is one of the most promising "Beyond CMOS" implementations. It relies on molecules arranged in nanolevel structures, with their charge distributions encoding logic information. Computation and signal propagation are enabled by exploiting Coulomb electrostatic interactions across molecules in cooperation with external electric fields. The mFCN approach offers several advantages over traditional CMOS digital systems, including significantly reduced power consumption, higher operating frequency, and highly scalable device sizes. To study the paradigm, the VLSILAB group of the Politecnico di Torino has developed a collection of nanoelectronic applications: MagCAD, a graphical 3D editor used to design emerging computing devices; ToPoliNano, a CAD tool for studying and simulating these systems, recently embedded with SCERPA (Self-Consistent Electrostatic Potential Algorithm), an academic MATLAB procedure for molecular circuit analysis; and FCNviewer, a 3D visualization program used to represent and check for correct information propagation and elaboration. The thesis aims at reviewing the integrability of the research-based algorithm in the ToPoliNano software and assessing the accuracy of the CAD simulation results. To achieve the goal, a comprehensive verification and validation strategy, supported by a series of comparing and testing scripts, is implemented and applied to a range of increasingly complex molecular circuits. Furthermore, new features and enhancements are introduced to both SCERPA and ToPoliNano suite, maximizing their compatibility, usability, and functionality. One specific improvement of the SCERPA algorithm is the introduction of a dynamic damping feature that accelerates the convergence of the method, reducing the time required for molecular analysis. The study also explores the possibility of designing novel and more complex molecular devices by using the newly verified and improved ToPoliNano package. In particular, it generalizes the 1-bit full adder layout to an n-bit version and successfully converts and validates the ISCAS17 digital circuit into its molecular equivalent. In conclusion, the thesis has contributed to the field of molecular Field-Coupled Nanocomputing by providing researchers with an effective and reliable framework to investigate the behavior of mFCN systems.