M.Sc Thesis

M.Sc StudentDuong Tuan
SubjectNanomaterial-Based Sensors to Detect Chemical and Biological
Threats Inside Containers
DepartmentDepartment of Chemical Engineering
Supervisor PROF. Hossam Haick


Global economics today relies on a vast number of imported and exported goods and products, which leads to thousands of shipments are being transferred daily worldwide. As a result, border security has become a priority of worldwide countries especially those which are prone to terrorism. There are various threats that require special attention including chemical, biological, nuclear, radiological, explosives, etc. Each has its own limitation and detection. Out of these threats, chemical and biological agents could be potentially identified based on chemical sensing of the molecules or simulates of the hazardous molecules. However, one of the challenges is to effectively scan the containers at the borders without slowing down the supply chain. Sniffer dogs - “amongst the most efficient customs technology” - have been trained to detect a variety of substances, but many of them are odorless or have similar odor with the unharmful ones. Therefore, there is an increasing demand on better solution, which is both user-friendly and more efficient at detecting forbidden substances.

In this work, we demonstrate the use of molecularly modified gold nanoparticles as sensing materials for chemiresistors to detect specific well-known chemical and biological warfare agents via the Volatile Organic Compounds (VOCs) generated from the simulants/ bacterial solutions in lab environment. Five out of sensors we tested have shown the efficacy in detecting simulants at low concentrations, which are diethyl malonate (DEM) at 500 ppm, 2-chloroethyl ethyl sulfide (2-CEES) at 5 ppm and dimethyl methyl phosphonate (DMMP) at 5 ppm. In experiments with biological substances, two sensors could detect VOC from E.coli at 2?107 to 1.6?108 CFU/ml and three other sensors could detect VOC from B.subtilis at 1.5?105 to 7?107 CFU/ml.

The collective data of sensor exposure was then re-analyzed using multi-variant pattern recognition algorithm principle component analysis (PCA) to better visualize the discrimination using all five sensors. With chemical compounds, PCA target compounds could be clearly separated into corresponding groups. Not only could sensor data separate the different chemicals, but also could separate the same chemical (2-CEES) in two different concentrations (5 ppm and 10 ppm). With biological compounds, although the PCA cannot separate two bacteria clearly, it showed discrimination between media with and without bacteria. Potentially, in the future, the results are to develop a system which would be used in the “point-of-care” on border lines and aid in fast identification and prevention of worldwide terror attacks.