SEMI-AUTOMATIC AND INTERACTIVE VISUALIZATION OF QUANTUM DOT NANO-STRUCTURES

 
 
 
 

SEMI-AUTOMATIC AND INTERACTIVE VISUALIZATION OF QUANTUM DOT NANO-STRUCTURES

Show full item record


Title: SEMI-AUTOMATIC AND INTERACTIVE VISUALIZATION OF QUANTUM DOT NANO-STRUCTURES
Author: Ye, Xinying
Description: This thesis presents a visual analytics system for semi-automatic and interactive visualization of quantum dot (QD) nano-structures. It presents the rationale, design, implementation and testing of the system. QDs are artificial atoms where the movements of the real atoms in them are confined. Scientists have used Monte Carlo simulations to predict the growth of self-organized QDs in strained semiconductors. The visual analytics system presented in this thesis is a set of post-simulation visual analysis tools. It consists of four modules: Input, Clustering, Identification and Examination. It provides the following tools for the scientists to visualize and analyze the formation and the structures of the QDs. • Timed animation: shows the formation and growth of the QD islands. • Clustering: internally identifies QD islands as spatial clusters of atoms. • Color-coding: renders each QD island in a distinctive color so that its structure can be easily visualized. • Auto-identification: automatically identifies the islands of particular interest (IOPI); • Auto-zooming: allows the scientists to have close examinations of the QD islands of particular interests. • Interactive visualization: allows the scientists to view the overall picture of the QD distribution or an IOPI from different vantage points. • Triangulation: forms an elevation surface from the QDs so that their height profile can be visualized. • Cutting: constructs a cross section view of the triangulated elevation surface from any given cutting plane so that the scientists can further analyze the height profile at any cross sections. Initial tests have shown that the tools are easy to use and the system is valuable to the scientists for post-simulation analysis of the nano-structures of the self-organized QDs. Further work can be done to improve the performance and the user-friendliness of the system. For examples, Quad-trees can be incorporated into the spatial clustering algorithms to speed the clustering process by only considering the atoms in the same and neighboring quads when making distance calculations. Labeled axes can be added to the height profile display to make it easier for scientists to analyze the profile more accurately.
Permanent Link: http://rave.ohiolink.edu/etdc/view?acc_num=akron1195496291
http://hdl.handle.net/2374.OX/3788
Date: 2007

Files in this item

Files Size Format View

There are no files associated with this item.

This item appears in the following Collection(s)

Show full item record