Portable spectroscopy, digital imaging colorimetry and multivariate statistical tools in contaminant identification: A case study of mint (Mentha) and basil (Ocimum basilicum)

Abstract

The advent of portable Fourier-Transform Infrared (FTIR) and Raman spectrometers has revolutionized analysis capabilities, presenting the possibility of on-site contaminant identification without the need for specialized laboratory settings. Compared to laboratory instrumentation, portable spectroscopy is more prone to noise, and appropriate spectral processing procedures need to be established. This paper introduces a comprehensive methodology that integrates acquisition techniques, spectral analysis, and mathematical tools necessary for utilizing handheld spectrometers to diagnose plant contamination. It focuses on determining the efficacy of handheld FTIR, Raman spectroscopy, and digital imaging for detecting contaminants in two food plants, Basil (Ocimum basilicum) and Mint (Mentha). The study examines the impact of three pollutants: iron (II) sulphate ($FeS{O}_4$), zinc (II) sulphate ($ZnS{O}_4$), and copper (II) sulphate ($CuS{O}_4$), on these plants, but also the necessary amount of measurements to spot the pollutants' effects. Measurements were conducted at the start, after 24 hours, and after 48 hours of exposure, on both fresh and dried plant leaves, as well as in solution. Spectral effects of each of the pollutants were identified with the use of multivariate statistical process control techniques. With the help of the developed methodologies, researchers can identify in-situ contaminant effects, exposure times and run diagnostics directly on the leaf both in alive and dried plants.

Keywords: Colorimetry; Fourier-transform infrared (FTIR) spectroscopy; Handheld instrumentation; Plant contamination; Raman spectroscopy.

Figure 1

Experimental design for mint and basil plants. The solution is analyzed with the Raman instrument, the live plant with Raman, FTIR and digital imaging, and the dried plant is analyzed with the FTIR spectrometer.

Figure 2

Principles of handheld FTIR instrumentation and spectral acquisition.

Figure 3

Principles of Raman spectroscopy and spectral acquisition using a right-angle cap.

Figure 4

Digital image acquisition example for mint leaves 24 h Cu²⁺ exposure: (a) the raw digital image and the (b) color-corrected image. (c) Showcases the three-dimensional distribution of colours in the RGB colour space.

Figure 5

Workflows for identifying changing variation in wavenumber reflectance using Statistical Process Control methodologies.

Figure 6

Visual changes of plants during experiments.

Figure 7

Example of pretreatment intermediate results for healthy Basil plant Raman observations: from the raw spectra (a), after the baseline correction (b), after normalization (c) and after smoothing and ROI-cropping (d). The spectra utilized in the PCA models are showcased in (d). (e) showcases the spectra for the control plants and the Copper exposure experiments.

Figure 8

FTIR spectral pretreatment of Basil plant: (a) original FTIR reflectance spectra of the dried healthy leaf; (b) Smoothed, normalized, ROI-cropped spectra of a dried leaf; (c) Original reflectance FTIR spectra of an alive leaf; (d) Smoothed and normalized spectra of an alive leaf. (b) and (c) are the spectral products utilized in PCA modelling. (e) and (f) showcase the alive and respectively dried plant FTIR processed spectra for the Cu²⁺ exposure.

Figure 9

(a) Variation of the dataset as a function of the number of random samples included (b) and variation of the re-projected dataset as a function of the number of random samples included. The black line represents the number of samples chosen in the model; (c) variogram in the PCA model with three principal components.

Figure 10

Heterogeneity testing and effect of experimental setup on the Basil Alive Control plant. (a) Explained variance by the model (b) Scatter plot of the first PCs scores (c) T2 control chart for the 4 PCs control model.

Figure 11

Control charts for identifying polluted samples, in the case of Cu²⁺ solution exposure, measured with FTIR in the alive state (a) and in dried samples (b). The scale of the plots is logarithmic due to the large out-of-control values.

Figure 12

(a) Contributions to the T² control chart of individual wavenumbers (b) wavenumbers with exceeding contributions on the normalized spectra.

Figure 13

(a) Mean FTIR spectra differences between the healthy plant and the polluted one with Cu in the region of interest of the dried basil plant.

Figure 14

Digital image averages of individual leaves for the (a) Cu²⁺ solution experiment and (b) Zn²⁺ solution experiment. The T² control chart (c) represents the alive-plant FTIR model for Zn²⁺ solution exposure.