Can visible-near-infrared (vis-NIR) spectroscopy detect toxic trace elements in soil and sediments effectively?
In a recent review article published in the journal Applied Spectroscopy Reviews, a team of researchers at the Institute of Oceanographic Instrumentation and Qilu University of Technology documented how traditional laboratory analytical methods are being replaced by visible-near-infrared (vis-NIR) spectroscopy as an alternative for detecting potentially toxic trace elements (PTEs) in soil and sediments.1 The lead authors Huimin Qiu and Pingping Fan highlighted that the current state of vis-NIR technology is helping to predict PTE concentrations, identifying both its demonstrated capabilities and the technical barriers limiting wider adoption.
Spectroscopic techniques are routinely used for trace element analysis in soil, include X-ray fluorescence (XRF) for rapid screening.2,3 Other highly sensitive, laboratory-based techniques are also used, such as inductively coupled plasma–mass spectrometry (ICP-MS), inductively coupled plasma–optical emission spectrometry (ICP-OES), and atomic absorption spectroscopy (AAS).4 These atomic spectroscopy methods detect heavy metals and nutrients by measuring the light emission, absorption, or fluorescence.2–4
PTEs are environmental contaminants that contain heavy metals such as cadmium, lead, and arsenic.1 Traces of these elements in the environment have had a known negative impact on contaminating agricultural land, waterways, and coastal sediments through industrial activity, mining, and agricultural runoff.2 Traditional detection methods, while accurate, are time-consuming, costly, and generate chemical waste, making large-scale monitoring logistically difficult.1
The researchers highlighted in their review article that vis-NIR spectroscopy is a chemical-free, nondestructive technique. For this reason, vis-NIR can analyze sediment and soil samples much more quickly without destroying the sample or harming the environment that the samples are collected from.
Currently, vis-NIR spectroscopy is being used in conjunction with spectral data processing and machine learning (ML) modeling techniques to link spectral signatures to PTE concentrations.1
The authors identify several factors that compromise prediction accuracy, including variations in soil moisture, organic matter content, and particle size, all of which can obscure the indirect spectral signals associated with PTEs.1 Because many trace elements do not directly absorb vis-NIR radiation, models rely on correlations with other soil constituents, introducing uncertainty.1
There are five areas, according to the researchers, that should be further explored to improve vis-NIR spectroscopy and ML modeling techniques to analyze PTEs in soil and sediments. These areas include the following:
For environmental monitoring agencies and soil scientists, the review provides a consolidated technical reference and signals where investment in method development is most needed to make field-deployable spectroscopy a routine tool in contamination assessment.1
