Zeolites investigation by machine learning

The purpose of this study is to find out if there are any ways to create synthetic zeolites base on the only information in topological features, pore volume, topological density and unit cell parameter to achieve the require type of zeolite material which may contribute to new discovery of zeolite...

Full description

Saved in:
Bibliographic Details
Main Author: Loh, Yu Wei
Other Authors: Su Haibin
Format: Final Year Project
Language:English
Published: 2016
Subjects:
Online Access:http://hdl.handle.net/10356/67304
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:The purpose of this study is to find out if there are any ways to create synthetic zeolites base on the only information in topological features, pore volume, topological density and unit cell parameter to achieve the require type of zeolite material which may contribute to new discovery of zeolite structure. However, most people produce the type of zeolite materials are based on the elements that are present and neglect certain properties to achieve new recipes for new zeolite structure or the structure they wanted. Due to the facts that some of the properties can be adjusted or controlled during the synthesis of synthetic zeolites to fit into different application, people tends to overlook on such properties which are amendable. This report covers on how significant of those properties are effecting the classification of the type of zeolite material which would relate to the new discovery of zeolite structure in the future. Experiment and investigate are done by using machine learning. Specific algorithm is used to obtain accurate prediction results. The finding and results discovered in this project has proved that topological features, pore volume, topological density and unit cell parameter properties hold some significant contribution toward the synthesis to achieve the required type of synthetic zeolite material, although they are not as highly significant as the element presences. The limitation of this project is that all investigates are done by using machine learning and not by the actual hand-on experiment. This could affect the differences between theoretical and physical results. However the advance knowledge from machine learning could provide a good head start in doing any physical experiment if required.