Graphene Inkjet Integrated Intelligent Sensors and Systems

Increasing awareness of personal health conditions has emerged as a key driver in point-of-care breath analysis for early diagnosis. This also demands real-time monitoring of air pollution based on internet of things (IoT) networks. Both systems rely on the development of scalable mobile-embeddable devices. Conventional devices face challenges like high power consumption, severe baseline drifts, and cross-analyte interferences, compromising the detection accuracies. Recently, 2D nanomaterials and their functionalization with metal oxides (MOx) have become attractive for high performing sensors. Solution processing of these materials coupled with inkjet printing can be promising for fabricating low-cost devices. I have addressed the abovementioned challenges from the perspectives of materials, device fabrication, and algorithmic approaches. I develop an inkjet-printed array of graphene-ZnO and graphene-WO₃ system integrated onto miniaturized CMOS platforms to detect NH₃ and acetone for potential diagnosis of diseases. I also develop a machine-intelligent classification system based on inkjet-integrated α-Fe₂O₃-rGO CMOS devices to selectively measure NO₂ air pollutants. The ink formulation and the theoretical model I develop facilitate uniform printed morphology, achieving excellent device reproducibility. Coupled with temperature modulation (TM) algorithms I develop, the devices achieve high responsivity, eliminated baseline drift, reduced response/recovery time, high SNR, and extraordinary repeatability. Moreover, a sensor array is established for compensation of cross-analyte interference. The influences of interferents are addressed by machine learning (ML) approaches. The predictive system quantifies NO₂ concentrations under different humidity conditions, with excellent separation between classes. Furthermore, the fully inkjet-printed room temperature NO₂ sensors assembled within the developed multi-sensory hub have performed reliable measurements in a real-world environment. My strategy to combine the sensing material, inkjet printing onto the CMOS platform, and TM/ML algorithms promises a robust system that outperforms conventional devices. The versatile technologies enable reliable multi-material array for potential multi-disease diagnostics. The systems developed also pave the way for ubiquitous implementation of adaptive wireless-connected monitoring systems.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

2020

Author dc:creator

• Wu, Tien-Chun

Advisor dc:contributor.advisor

• Hasan, Tawfique

• Graphene
• Inkjet
• CMOS
• Machine Learning
• Temperature Modulation
• Disease Diagnostics
• Air Quality Monitoring
• Internet of Things
• Intelligent Sensors
• Electronic Nose

Licence

• https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/58dcfbec-7c28-4f29-932f-67a560d0b708/download
• https://www.rioxx.net/licenses/all-rights-reserved/

Language dc:language

en

Author Identifier

0000-0002-6185-3256

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/302006