Optimization of Cadmium Isotope Analysis in Marine Sediments Using a Standard Pneumatic Nebulizer Coupled with MC-ICP-MS

Cadmium (Cd) is a highly toxic heavy metal released into marine environments through anthropogenic activities such as mining, smelting, and industrial waste. Tracking these pollution sources requires precise isotope analysis, yet marine sediments present significant challenges due to low Cd concentrations (<0.2 mg/kg) and complex matrix interferences. While conventional methods often utilize desolvation systems to enhance sensitivity, they are frequently limited by instrumental mass bias instability and significant memory effects.

In this study, we optimized a robust Cd stable isotope analytical procedure using multi-collector inductively coupled plasma mass spectrometry (MC-ICP-MS) equipped with a standard pneumatic nebulizer (SPN). To achieve high-precision data, a two-step ion-exchange chromatography process (AGMP-1M and TRU-SPEC resins) was established, ensuring >90% Cd recovery and effective removal of major matrix elements(Al, Fe, Ca) and isobaric/molecular interferents (Sn, Mo, Nb, Zr) even without the help of a desolvator.

Our results define the optimal analytical thresholds for reliable δ^114/110Cd measurements: a minimum Cd concentration of 50 ng/g (approx. 150 ng total) is required, making this protocol applicable to sediments with Cd levels as low as 0.15 mg/kg. Strict interference control limits were established, maintaining matrix-to-Cd ratios (M/Cd ≤ 5) and suppressing molecular interferences from Mo, Nb, and Zr below ratios of 0.01, 0.0005, and 0.0001, respectively.

The data quality of the method was extensively validated through a comprehensive assessment of isotope fractionation and comparison with reference values for standard materials. The entire dataset (n=733), encompassing 12 types of reference materials (RMs) (e.g., NIST SRM 2711a, NRCC PACS-3, USGS NOD-A-1), verification standards (BAM-I012), and the zero-delta standard (NIST 3108), exhibited exceptional mass fractionation linearity. The slope between δ^114/110Cd and δ^113/110Cd was 0.7544 (R²=0.996) and that between δ¹¹⁴Cd and δ¹¹¹Cd was 0.2545 (R²=0.944). The slope between δ¹¹⁴/¹¹⁰Cd and δ¹¹³/¹¹⁰Cd was 0.7544 (R² = 0.996), and that between δ¹¹⁴Cd and δ¹¹¹Cd was 0.2545 (R² = 0.944). These slopes agree with predicted values based on mass-dependent fractionation, demonstrating that potential isobaric and molecular interferences were effectively eliminated and ensuring the fundamental reliability of the data. The measured δ^114/110Cd values for the 12 RMs showed an average absolute deviation of only 0.079±0.086‰ compared to previously reported literature values. Furthermore, a rigorous inter-laboratory cross-validation was conducted between Chungnam National University (using the SPN and external mass-bias correction with Ag-doping) and Kyoto University (using the conventional desolvator and double-spike mass-bias correction). This comparison, involving 5 RMs and 5 surface sediment samples from Onsan Port collected in April 2025 (OS4, OS5, OS8, OS14, OS35), yielded a high correlation coefficient (R²=0.944) and a small absolute deviation (0.037±0.015).

These findings demonstrate that our optimized analytical framework ensures international-level reliability and precision. This protocol provides a powerful tool for environmental forensics, successfully differentiating natural background levels from anthropogenic inputs—such as zinc smelting and coal—in complex marine ecosystems.