کد مقاله کد نشریه سال انتشار مقاله انگلیسی نسخه تمام متن
5770463 1629411 2017 7 صفحه PDF دانلود رایگان
عنوان انگلیسی مقاله ISI
Potential of visible and near-infrared reflectance spectroscopy for the determination of rare earth elements in soil
ترجمه فارسی عنوان
پتانسیل طیف سنجی بازتابی و نزدیک به مادون قرمز برای تعیین عناصر خاکی کمیاب در خاک
کلمات کلیدی
موضوعات مرتبط
مهندسی و علوم پایه علوم زمین و سیارات فرآیندهای سطح زمین
چکیده انگلیسی


- Spectral features of rare earth elements (REE) were not distinct in soil Vis-NIR spectra.
- REE spectral features were masked by those of soil components, such as iron oxides.
- PLSR modeling could identify the REE features from soil Vis-NIR spectra.
- Acceptable prediction was obtained for La, Nd, Pr, Sm, and total light REE.
- Vis-NIR reflectance spectroscopy is a simple and rapid method for determining REE.

The simple and rapid determination of rare earth elements (REE) in soil is essential for the assessment of their environmental and health risks. Inductively coupled plasma-mass spectrometry (ICP-MS) and -emission spectrometry (ICP-ES) are most commonly used for the determination of REE. However, both require complex sample preparation. In this study, we investigated the potential of visible and near-infrared (Vis-NIR) reflectance spectroscopy for determining REE in soil. We collected 130 soil samples near REE mines. Their Vis-NIR spectra, which range from 350 to 2500 nm, were measured after air-drying and grinding in the laboratory. Partial least squares regression (PLSR) was selected to calibrate the spectra and REE contents measured by ICP-MS. The results showed that the absorption features of REE were not distinct in the soil spectra, such as for Nd at 798 nm, which were masked by soil components, particularly by iron oxides because they have overlapping spectral features with REE. However, when using the spectra in the 400-1000 nm range, good or reasonable accuracy [R2 > 0.5 and residual prediction deviation (RPD) > 1.4] was observed for La, Pr, Nd, Sm, and total light REE (∑ La-Eu excluding Pm) because their spectral features could be identified by PLSR calibration. The spectral pretreatments of the first derivative, standard normal variate transformation, and multiplicative scatter correction did not improve the prediction accuracy of REE compared with Savitzky-Golay smoothing. This study provides a simple and rapid method for determining REE in soil, particularly for La, Pr, Nd, Sm, and total light REE.

ناشر
Database: Elsevier - ScienceDirect (ساینس دایرکت)
Journal: Geoderma - Volume 306, 15 November 2017, Pages 120-126
نویسندگان
, , ,