Journal
SENSORS AND ACTUATORS B-CHEMICAL
Volume 206, Issue -, Pages 471-487Publisher
ELSEVIER SCIENCE SA
DOI: 10.1016/j.snb.2014.09.102
Keywords
MIP; QCM sensor; Aldehydes; Body odor characterization; Chemical vapor pattern recognition
Funding
- JSPS KAKENHI [25420409]
- [24.02367]
- Grants-in-Aid for Scientific Research [25420409] Funding Source: KAKEN
Ask authors/readers for more resources
Molecularly imprinted polymers (MIPs) have been prepared using the polyacrylic acid (PAA) as host polymer and hexanal, heptanal, and nonanal as pattern molecules. MIPs were employed as selective coating layer of quartz crystal microbalance (QCM) sensors. Hexanal, heptanal, and nonanal were opted as target chemicals after gas chromatography-mass spectrometer (GC-MS) characterization of body odor samples. Transient and static responses of four QCM sensors (three coated with MIPs and one with non-MIP) to target aldehydes in singly, binary and tertiary mixtures, and water at distinct concentrations have been measured. Transient responses were analyzed to compute the response time (t(on)), and recovery time (t(off)) of sensors. This result average values of t(on) approximate to 5 s and t(off) approximate to 10 s to typical concentrations of target odors. The sensitivity and baseline drift of sensors were also calculated using their static response. The heptanal template molecule based MIP coated QCM exhibit improved sensitivity, reproducibility and faster response, than the rest two MIPs, and non-MIP coated QCMs. Static sensors response matrices were further processed with principal component analysis (PCA) for qualitative (visual) discrimination and support vector machine (SVM) classifier for quantitative recognition (in %) of target aldehydes: in singly, binary and tertiary mixtures. Aldehydes odor was effectively identified in principal component (PC) space. Maximum recognition rate of 89% has been achieved for three classes of binary odors, and 79% for the combination of single, binary and tertiary odor classes in 3-fold cross-validation of SVM classifier. (C) 2014 Elsevier B.V. All rights reserved.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available