Related Experiment Video
Updated: Jan 13, 2026

Clean Sampling and Analysis of River and Estuarine Waters for Trace Metal Studies
Published on: July 1, 2016
Multivariate optimization for preconcentration and ICP-MS determination of rare earth elements, including
Franciele Castro Novais1, Pedro Victor Bomfim Bahia2, Vanessa Hatje3,2,4
1Centro Interdisciplinar de Energia E Ambiente (CIEnAm), Universidade Federal da Bahia, Salvador, Bahia, 40170-115, Brazil. francielenovais@ufba.br.
Abstract:
Gadolinium (Gd) complexes, used as contrast agents in magnetic resonance imaging (MRI), along with other rare earth elements (REE), are emerging contaminants in natural waters. The quantification of these compounds is a challenge due to their low concentrations. In this study, an extraction and preconcentration method was optimized and validated for the simultaneous determination of REE, including anthropogenic Gd, in tap water by inductively coupled plasma mass spectrometry (ICP-MS). Initially, two solid-phase preconcentration methods (NOBIAS-chelate PA-1® resin columns and SEP-PAK Classic C18 columns) were tested univariately. According to the % recovery, NOBIAS-chelate PA-1® columns was chosen. Chemometric tools, including fractional factorial design and Doehlert matrix, were applied to evaluate the parameters influencing the extraction and preconcentration steps. Under optimal conditions, the method was validated, achieving low limits of detection (0.02-0.14 ng L-1) and quantification (0.06-0.48 ng L-1), as well as good recoveries (92-99%) and precision (0.7-5.9%). The practicality and sustainability of the method were demonstrated using blue applicability grade index (BAGI) and sustainability and practicality metrics system (SPMS) metrics. Finally, the optimized and validated method was applied to six tap water samples from Brazil and Germany. ∑REE concentrations ranged from 37.9 to 596 ng L-1 (Brazil) and 47.6 to 80.5 ng L-1 (Germany), and the method demonstrated its suitability for the precise determination of REE, producing high recoveries and low uncertainties. The approach is straightforward, efficient in reagent and sample use and produces minimal waste.

