PROPOSAL

Abstract

The rare earth elements (REEs) include scandium (Sc), yttrium (Y) and the 15 lanthanide elements. The relevance of these elements is due to the excellent electronic, optic, magnetic and catalytic properties which give them an important role in emerging clean technologies. As the demand of REEs grows annually, coal fly ash (CFA) has attracted attention as a cheap and abundant waste stream containing appreciable REEs concentrations. In this study, we aim to develop a method, using ICP-MS to determine the recovery rate of REEs extracted from CFA at high accuracy and precision.

Acid leaching of CFA typically dissolves substantial concentrations of major cations, particularly Ca, Al, and Fe, which can cause interference and necessitate high dilution factors that compromise REE detectability. In this study, we try to investigate the influence of acid concentration, sample preparation protocols, and ICP-MS operating conditions on ICP-MS response during REE measurement. Solutions prepared from controlled acid leaching experiments and synthetic cation-rich matrices to assess recovery rate of REEs under varying dilution schemes, allowing us to distinguish chemical and instrumental contributions to matrix-induced biases. By quantifying the extent of signal suppression and proposing correction strategies, this work provides analytical guidance for reliable ICP-MS determination of REEs in CFA leachates. These results may support both accurate REE resource assessment and the optimization of acid leaching protocols for the valorization of CFA.

Research Objective

The objective of this research is to develop a method for accurately testing REEs concentration in the leachates of coal fly ash using ICP-MS. To achieve this, we focus on sample preparation protocols, and ICP-MS operating conditions (especially CRC reaction gas mode) in order to minimize matrix effects introduced by the dissolution of major cations and to increase accuracy. Controlled experiments will evaluate how varying acid strength, dilution schemes and calibration approaches impact analytical accuracy. By assessing the combined effects of acid strength and cation abundance on ICP-MS signal fidelity, this work aims to establish a robust analytical framework for REE determination in complex coal fly ash leachates.

Discussion of Significance

In recent years, REEs have been in high demand due to their critical roles in clean energy technologies, defense applications, and electronics. Identifying inexpensive and abundant secondary resources for REEs is therefore of growing importance. CFA represents a promising feedstock. According to the American Coal Ash Association, approximately 32 million short tons of CFA are produced every year, yet only about 60% is reused, mainly in building materials, zeolite synthesis, concrete and agricultural applications. The remainder is typically disposed in landfills, where heavy metals, such as mercury, cadmium, arsenic, and lead pose risks of long-term contamination to surface water or groundwater. But excitingly, this waste material contains REEs at an average concentration of ~400 ppm, corresponding to ~400 g of REEs per ton, with certain samples enriched in economically valuable elements such as Dy and Nd.

During acid leaching, major cations (Ca, Al, Fe) are inevitably released at concentrations hundreds to thousands of times higher than REEs. Because standard ICP-MS detection requires dilution to ~25 ppm for Ca and ~1 ppm for Al, the dilution factor as high as 1,000–10,000 may be necessary, introducing substantial errors in REE quantification. Additional matrix effects may further obscure REE signals. Thus, developing an analytical method with sufficient sensitivity and matrix-tolerance is essential to predict REE recovery efficiency from CFA and to enable downstream optimization of chelating agents and extractants.

Review of relevant work

Currently, the quality of REE concentration or isotope ratio determination by ICP-MS is influenced by multiple factors. In addition to general issues relevant to any analytical measurement or instrumental conditions (e.g., signal drift, detector dead time, and plasma stability), there are specific factors associated with the physical and chemical processes occurring in the ICP source that lead to interference and fractionation. The m/z range for REEs spans from 139 to 175, and the oxides and hydroxides of LREEs may interfere with the detection of HREEs. Previous studies have demonstrated that mass spectrometric interference increases significantly when the LREE/HREE ratio exceeds 50 (Lee 2023, Dong 2025). For example, the formation of oxides and hydroxides of LREEs, such as La, Ce, Pr, and Nd, can substantially affect the accurate determination of HREEs, including Gd, Tb, and Dy. When using¹⁵⁷Gd for mass spectrometric quantification,¹⁴¹Pr¹⁶O typically requires correction. Other forms of interference, such as¹⁴⁰CeOH,¹³⁹La¹⁸O, Nd–C, and Nd–N, also occur. Similarly, when ¹⁶⁰Gd is selected for analysis, it is subject to interference from¹⁴⁴Nd¹⁶O and¹⁶⁰Dy, with additional contributions from Ce and Pr hydroxides, as well as¹⁴⁴Sm¹⁶O. To address these challenges, some researchers have applied correction factors derived experimentally to account for the formation yields of interfering ions in the plasma. Others have investigated double-focusing ICP-SFMS, where increasing resolution (m/Δm) to the range of 8,000–10,000 allows REE oxide ions to be separated from atomic REE ions, though at the cost of significantly reduced sensitivity (Becker, 2007; Norbert, 1998). Several authors have also explored the use of high-resolution modes, collision cell technology, and the management of doubly charged ions to minimize interference (Houk, 2000; Leikin, 2015). In addressing Ba-related interferences, researchers have proposed a sample preparation approach that adjusts HNO₃ concentration to precipitate Ba(NO₃)₂, thereby reducing interference (Liu, 2023).

Besides, since REEs have among the highest metal-oxide (M-O) bond strengths of any element, and the oxide ions of the low mass REE interfere with the mid-mass and high-mass REEs in the ICP-MS spectrum increasing challenge in trace measurement of higher-mass REEs in a low-mass REE matrix. For example, in analysis of trace REEs in high-purity Sm₂O₃, ¹⁴⁷Sm¹⁶O⁺ overlaps the preferred isotope of ⁶³Dy⁺ and ¹⁴⁹Sm¹⁶O+ overlaps the only isotope of ⁶⁵Ho⁺ (Naoki, 2012). Separation of the trace REE analytes from the REE matrix can be performed using either off-line or on-line matrix removal utilizing a chelating resin, but this technique is time consuming and the method needs to be customized according to the matrix element being separated. In order to direct analyze trace REEs in a variety of high-purity REE matrices, researchers have compared the ability of O2 mass-shift mode and NH3 on mass mode. Results show that overlaps from SmO⁺, SmOH⁺ and SmOH2⁺ on Dy, Ho, Er, Tm and Yb were reduced by around two orders of magnitude using O2 mass-shift mode, but O2 mass-shift mode was not as effective for the removal of the GdO⁺ interference on Yb⁺. NH3 on-mass mode was also effective in reducing the interference of GdO⁺ on Yb⁺, but La, Ce, Sm, Gd, Tb and Lu react with NH3 so efficiently that this NH3 mode could not be used for the on-mass measurement of those analytes.

For analyzing silicate samples, another problem is matrix effects as coal fly ash leachates typically contain high concentrations of Ca, Al, and Fe. Rock solution with high concentrations of Al, Fe, Ca and Mg can cause a reduction in response for trace elements, and may partially block the ICP-MS orifice (Beauchemin, 1987; Kawaguchi, 1987). To reduce matrix effects, dilution factors of 500–1000 (potentially higher to 1000–10,000 for CFA are commonly applied (Jarvis, 1990; Longerich, 1990). Makishima (2007) found that the suppression of signals for ⁸⁸Sr, ¹⁴⁰Ce and ²³⁸U in rock solution caused by rock matrix effect in ICP-MS (matrix effects) was reduced at high power operation of the ICP. They proposed a method optimizing dilution factor and plasma voltage to make the signal suppression by the matrix negligible and precisely determinate Rb, Sr, Y, Cs, Ba, REEs, Pb, Th and U using flow injection-ICP-MS.

Materials and Methods

Sample Preparation

A coal fly ash samples from powder river basin is used in this work. Total REEs concentration is ~400 ppm with silicon and aluminum concentration (presented in the form of oxides) typically ~20 wt% and iron concentration (presented in the form of oxides) ~5 wt%. Acid leaching experiment of the CFA is conducted with 12M nitric acid. After acid leaching, the leachate is transferred to a clean centrifuge tube and centrifuged for 10min at 8000 rpm to separate liquid and solid phase. The resulting leachates will be stored in acid-cleaned centrifuge tubes for subsequent analysis.

Prior to instrumental measurement, each leachate was diluted with 2% v/v HNO₃ (prepared from TraceMetal™ grade nitric acid and 18.2 MΩ·cm DI water) to achieve final dilution factors between 1:1,000 and 1:10,000, ensuring total dissolved solids (TDS) remained below the operational limit of 200 ppm for the Agilent 8900 ICP‑MS.

Calibration and Quality Control

REE quantification targeted ⁴⁵Sc, ⁸⁹Y, ¹³⁹La, ¹⁴⁰Ce, ¹⁴¹Pr, ¹⁴⁶Nd, ¹⁴⁷Sm, ¹⁵³Eu, ¹⁵⁷Gd, ¹⁵⁹Tb, ¹⁶³Dy, ¹⁶⁵Ho, ¹⁶⁶Er, ¹⁶⁹Tm, ¹⁷²Yb, and ¹⁷⁵Lu. Calibration curves were constructed from six standards: 0 (blank, L1), 0.4 (L2), 2 (L3), 20 (L4), 100 (L5), and 200 (L6) µg L⁻¹, spanning the expected concentration range after dilution.

Internal standards ⁷¹Ga, ¹¹⁵In, and ¹⁸⁵Re are introduced online to correct for instrumental drift and matrix effects across the mid‑ to high‑mass REE range; ⁴⁵Sc is quantified without a directly bracketing internal standard but will be monitored for stability through calibration verification and quality control samples. One quality control (QC) sample is prepared as a lab‑fortified matrix spike at a concentration near the midpoint of the calibration range, using REE stock solutions independent of those employed for calibration, to assess recovery and matrix effects.

Matrix effects were evaluated indirectly using internal standard stability and calibration linearity across dilution levels. Variations in internal standard recovery and analyte signal response across different dilution factors were used to quantify matrix suppression.

REE Analysis by ICP-QQQ

REE measurements will be performed in both no gas and O₂ reaction modes to assess interference removal and matrix effects.

No-Gas Mode

In no-gas mode, both quadrupoles (Q1 and Q2) will be set to transmit the analyte ion masses directly without reaction gases. This mode will be used for elements less susceptible to polyatomic interferences, such as Sc, Y, La, and Ce. Operating parameters will be optimized for maximum sensitivity while minimizing oxide formation (CeO⁺/Ce⁺ < 2%) and doubly charged ion production (Ba²⁺/Ba⁺ < 3%). Plasma power and sampling depth will be adjusted to maintain stability under high TDS conditions.

O₂ Reaction Mode

For elements prone to oxide and hydroxide interferences (e.g., Sm, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu), O₂ reaction gas will be introduced into the collision/reaction cell to selectively convert REE⁺ ions into their oxide forms (REEO⁺).

• Q1 will be set to pass the target REE ion mass (M⁺), while
• Q2 will be set to transmit the corresponding product ion mass (M + 16) after reaction with O₂.

This mass-shift approach will separate analyte signals from overlapping oxide or hydroxide interferences (e.g., ¹⁴⁴Nd¹⁶O⁺ on ¹⁶⁰Gd⁺). O₂ flow rate, cell bias, and quadrupole settings will be optimized using real leachates to achieve the best balance between interference removal and signal sensitivity, while maintaining consistent internal standard recovery across different matrices.

Possible Outcomes

The development of the ICP‑MS method will provide a robust, reproducible, and matrix‑tolerant analytical protocol for the quantification of REEs in acid leachates of coal fly ash with high accuracy and precision. Optimized dilution, internal standardization, and interference control are expected to yield low detection limits and a wide linear range. The inclusion of an independently prepared matrix spike will verify recovery and assess matrix effects, ensuring suitability for complex, high‑TDS samples. The method will enable reliable evaluation of REE leaching efficiency, support inter‑laboratory comparability, and provide high‑quality data for advancing resource recovery strategies from coal combustion by‑products.

Timeframe and Budget

Table 1: Cost breakdown and timeline in this proposal. Costs are calculated based on 20 samples^*.

* Analysis will be conducted in the following order: 3 blanks→6 standard→2 blank→1 QC→S1-5→2 blanks→ S6-10→2 blank→1 QC→S11-15→2 blanks→ S16-20→2 blank→1 QC.