Prediction of Soil Mineralogy and CEC Using Near-Infrared Reflectance Spectroscopy.

Donald Campbell1 (Don.Campbell@panalytical.com), Daniel Shiley1 (Dan.Shiley@panalytical.com) and Brian Curtiss1 (Brian.Curtiss@panalytical.com)1 ASD Inc./PANalytical NIR Excellence Center, 2555 55th St., Boulder, Colorado 80301, USA Abstract Measurement of soil properties by cation-exchange capacity (CEC) and X-Ray diffraction (XRD) are often time consuming and expensive processes. CEC assays can be time consuming and provide additional challenges with regards to technician exposure to hazardous chemicals and the proper handling and disposal of those spent solutions and significant expertise is required to produce quality X-ray diffraction results. Near-infrared reflectance (NIR) spectroscopy can be used to create calibration models based on CEC values and XRD mineralogy. These calibrations can then be utilized to predict both exchange capacities and mineral composition of soil samples in near real-time without the need for additional costly and time consuming laboratory measurements. For mineral soils, CEC values can be interpreted in terms of total abundance of swelling clays. NIR has the added benefits of providing a field portable, non-destructive sample measurement. Multiple soil properties can also be simultaneously predicted via a single spectrum. Quantitative calibrations for NIR analyzers are developed using statistical methods that relate measured NIR reflectance to the material properties of interest. This poster examines the quantitative determination of soil mineralogy based on XRD kaolinite, montmorillonite and mica, and swelling clays determination by CEC measurement with multivariate calibrations using NIR spectroscopy.