Use this URL to cite or link to this record in EThOS:
Title: Data mining support for high-throughput discovery of nanomaterials
Author: Yang, Yang
ISNI:       0000 0004 2746 0306
Awarding Body: University of Leeds
Current Institution: University of Leeds
Date of Award: 2011
Availability of Full Text:
Access from EThOS:
Nanotechnology is becoming a promising technology due to its potential to dramatically improve the effectiveness of a number of existing consumer and industrial products, such as drug delivery systems, electronic circuits, catalysts and light-harvesting materials. However, the ability of industry and academia to accelerate the discovery of new nanomaterials is severely limited by the speed at which new compositions can be made and tested for suitable properties. A promising alternative approach for nanophotocatalyst discovery, currently under development at University College London (UCL) and University of Leeds, utilizes recent advances in nanomaterials synthesis and automation to implement a high-throughput (HT) experimental system to enable rapid exploration of materials space. The HT nanocatalyst discovery is an automated continuous process using hydrothermal synthesis which can synthesize a large number of nanoparticles in a short time. The nanoparticles formulated are characterized and tested on as many samples as possible for indicative properties rather than conduct comprehensive characterisation on each sample. This thesis describes the development of chemome~ric and inductive data-mining tools to support the HT nanomaterial discovery process. The work will be reported mainly in five parts, including a data management system structured to reflect the flow of HT work flow, an information flow management system designed for correspondence between UCL and Leeds, a prototype data mining system tailored for HT experimental data processing and analysis as well as its application, a new Design of Experiments (DoE) using genetic algorithm which was proved to be able to handle variables effectively, and robust quantitative structure-activity relationship (QSAR) models with genetic parameter optimization for HT catalyst discovery. In contrast to the enormous benefits, nanoparticles also bring adverse effects to the biological environment and human health. Due to the diversity of nanoparticles, as well as the dependence of toxicity on the physico-chemical properties of nanoparticles, it is not possible to test every particle. A rational way without testing ii every single nanoparticle and its variants is to relate the physicochemical characteristics of nanoparticles with their toxicity in a QSAR model. This thesis also researches the measurement methods of structural and composition properties including size distribution, surface area, morphological parameters and assessment methods for comparative toxicity of the nanoparticles in a panel of 18 nanoparticles which were chosen by University of Edinburgh, UBI Institute and University of Leeds. The major contribution of this work is in using data mining methods to analyze toxicity and measured structural and composition properties to find the relationships between the toxicity and physico-chemical properties of nanoparticles. Structure-activity relationship (SAR) analysis focused on identifying the possible structural and compositional properties that determine the cytotoxicity of nanoparticles. iii
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available