Solid-phase cadmium speciation in soil using L3-edge XANES spectroscopy with partial least-squares regression

Area of Science:

• Environmental Science
• Analytical Chemistry
• Geochemistry

Background:

• Cadmium (Cd) contamination in soil poses significant environmental risks.
• Accurate determination of Cd speciation is crucial for risk assessment but remains challenging.
• X-ray absorption near-edge structure (XANES) spectroscopy is a powerful tool for elemental speciation.

Purpose of the Study:

• To evaluate the efficacy of partial least-square (PLS) regression for deconvoluting Cd L(3)-edge XANES spectra.
• To develop and validate a PLS model for quantifying Cd compounds in complex soil matrices.
• To apply the validated model for direct determination of Cd speciation in a contaminated soil sample.

Main Methods:

• Acquisition of a spectral library of Cd reference compounds.
• Acquisition of Cd L(3)-edge XANES spectra from a soil sample.
• Application of PLS regression to deconvolute multi-compound Cd mixtures.
• Validation of the PLS model using binary and ternary mixtures.

Main Results:

• High coefficients of determination (R(2)) were achieved for Cd compounds in binary and ternary mixtures, validating the PLS approach.
• The PLS regression model successfully quantified five Cd compounds in a highly Cd-contaminated soil sample: Cd(3)(PO(4))(2) (36.1%), Cd(NO(3))(2)·4H(2)O (24.5%), Cd(OH)(2) (21.7%), CdCO(3) (17.1%), and CdCl(2) (0.4%).
• The model demonstrated the capability to handle complex soil systems by incorporating diverse Cd compounds.

Conclusions:

• Partial least-square (PLS) regression is a robust and promising method for direct determination of cadmium speciation in the solid phase of soil.
• The developed PLS-XANES approach offers a valuable tool for environmental risk assessment of Cd-contaminated soils.
• Further inclusion of diverse Cd compounds in the PLS model can enhance its applicability to complex environmental matrices.