# Single and Dual Mode SMR Sensors for Pest Detection in Plant Health Monitoring

## TL;DR

This paper explores using specialized sensors to detect plant-emitted chemicals for monitoring plant health, showing promising results for future applications.

## Contribution

The study introduces a novel classification framework using dual-mode SMR sensors and Python-based signal processing for VOC detection in plant health monitoring.

## Key findings

- SMR sensors successfully detected and classified VOCs like linalool, trans-2-hexenal, and D-limonene under varying humidity.
- A 2D polar classification framework effectively separated VOCs using frequency shift data from UWAR and Sorex sensors.
- The Freundlich isotherm model accurately described adsorption behavior on CH3-terminated surfaces.

## Abstract

This study presents the development and evaluation of surface functionalized solidly mounted resonators (SMRs), including custom developed at the University of Warwick (UWAR) devices and commercial Sorex sensors, for the detection and classification of plant-emitted volatile organic compounds (VOCs). The sensors were tested against linalool, trans-2-hexenal (T2H), and D-limonene at different concentrations under both dry and humid conditions (30% ± 3% RH). A Python-based (v3.13.5) signal-processing workflow was established to filter frequency responses and extract key features, such as baseline, saturation point, and frequency shift (Δf). Adsorption behaviour was modelled using the Freundlich isotherm, showing good agreement with experimental data and suggesting heterogeneous, multilayer adsorption on CH3-terminated EC surfaces. A 2D polar classification framework combining vector-normalized Δf values from UWAR and Sorex sensors enabled a clear separation of the VOCs. The results highlight the complementary performance of the two types of SMR sensors and demonstrate that feature-engineered resonant devices, combined with computational classification, offer strong potential for future use in plant health monitoring systems.