Sensitive and Rapid Analysis of Inorganic Arsenic in Food and Animal Feed Samples Using LC-ICP-MS

Sensitive and Rapid LC-ICP-MS Analysis of Inorganic Arsenic in Food and Feed: Summary of Application and Performance

Significance of the Topic
Monitoring inorganic arsenic (iAs) in food and animal feed is critical because inorganic species (As(III), As(V)) are substantially more toxic than common organic arsenic forms. Regulatory limits are tightening worldwide (EU Regulations, updated seafood limits effective 2025, and U.S. Baby Food Safety Act), so analytical methods must deliver accurate speciation, low detection limits, and robustness across diverse matrices that may contain large amounts of organic arsenic or halide interferences.

Objectives and Study Overview
The study evaluated a Shimadzu LC-ICP-MS workflow for rapid, sensitive quantification of total inorganic arsenic (converted to As(V)) in 14 food and feed matrices, following EN 16802:2016 (food) and EN 17374:2020 (feed) principles. Key aims were to: separate iAs from organic arsenic and chlorine-derived interferences in a short runtime, verify complete oxidation of As(III) to As(V) during sample pretreatment, demonstrate low limits of detection adequate for regulatory thresholds, and validate the method with spike recoveries and certified reference materials (CRMs).

Methodology
- Samples: 14 matrices including baby foods (rice cereal, apple jelly, infant formula), seafood (sea bream, tuna, squid), beverages (orange juice, sake), ham, animal feed, and three CRMs (brown rice NMIJ 7532-a/7533-a, hijiki NMIJ 7405-b).
- Sample preparation: Solid samples (approx. 0.5 g) were extracted with extraction solution 1 (0.1 M HNO3, 3% H2O2) at ~90 °C for 60 min; liquid samples were mixed 1:1 with extraction solution 2 (0.2 M HNO3, 6% H2O2) and treated identically. Extracts were centrifuged and filtered (0.45 µm PTFE). Additional dilution was applied for high-matrix samples (e.g., hijiki 200×, some liquids 5×). Spiked samples were prepared to assess recovery.
- Oxidation: EN-based pretreatment intentionally oxidizes As(III) to As(V) so total iAs is measured as As(V); verification experiments confirmed near-complete conversion (measured 4.83 µg/L from a 5 µg/L As(III) standard after pretreatment).

Used Instrumentation
- LC: Nexera XS inert HPLC with HAMILTON PRP-X100 anion-exchange column (250 × 4.1 mm, 10 µm). Mobile phase: 100 mM ammonium carbonate with 3% v/v methanol and 400 mg/L sodium sulfate (pH 9.3). Flow 1.2 mL/min, column 30 °C, injection 50 µL, runtime 6.5 min/sample.
- ICP-MS: Shimadzu ICPMS-2040/2050 with Nebulizer DC04, mini-torch (argon saving), cyclone cooled spray chamber, nickel cones, online Ga internal standard, He collision mode (cell gas 3.0 mL/min) to reduce 40Ar35Cl interference at m/z 75. Software: LabSolutions ICPMS TRM for integrated LC and ICP-MS control and data processing.

Main Results and Discussion
- Chromatography and separation: The method achieved baseline separation of iAs (As(V)) from organic arsenic species (AsB, AsC, DMA, MMA) and from a chloride peak, enabling reliable quantification of iAs even in matrices rich in organic arsenic (seafood). Full separation of AsC and AsB was not the primary focus here; DMA/MMA separation may require method adaptation if those species must be quantified accurately in high-organic matrices.
- Interference control: He collision mode removed 40Ar35Cl interference; chromatographic resolution between Cl and iAs was confirmed. Matrix-induced retention time shifts (observed in orange juice) could be mitigated by post-filtration dilution (example: 5×), restoring correct peak alignment.
- Sensitivity and detection limits: Calibration linearity for iAs was excellent (r = 0.99996). The solution LOD was determined from ten replicates of 0.1 µg/L standard; corresponding sample LODs after dilution factors met or exceeded regulatory requirements, including the strict baby-food limit (0.01 mg/kg). Typical LODs in samples were in the 0.00004–0.04 mg/kg range depending on matrix and dilution factors.
- Accuracy and robustness: Spike recovery tests on 11 matrices produced recoveries of 96–113%, demonstrating accuracy across a variety of food/feed types. Results for the three CRMs fell within certified ranges (brown rice and hijiki), validating the sample prep and analysis chain.
- Representative findings: Rice cereal and brown rice CRMs showed relatively high iAs (rice cereal ~0.086 mg/kg; brown rice CRM values ~0.298 and 0.530 mg/kg), consistent with grain-derived samples being prone to elevated iAs. Many other samples were below LODs.

Benefits and Practical Applications
- Rapid throughput: 6.5-minute runtime per sample supports relatively high throughput for routine monitoring and compliance testing.
- High sensitivity: Achieves LODs sufficient for current and anticipated regulatory limits (including baby food thresholds).
- Robust across matrices: Effective separation and interference control allow quantification in complex matrices (seafood with abundant organic arsenic, halide-rich samples).
- Cost efficiency: Use of a mini-torch reduces argon consumption by roughly one-third versus conventional torches, lowering operating costs for routine LC-ICP-MS assays.
- Integrated workflow: Single software control of LC and ICP-MS simplifies automation, data acquisition, and quality control in regulated laboratories.

Future Trends and Applications
- Expanded speciation: Adapting chromatographic conditions to routinely separate all common arsenic species (AsC, AsB, DMA, MMA, As(III), As(V)) will broaden applications where full speciation is required for toxicological interpretation.
- Lower LODs and faster runtimes: Ongoing improvements in instrumentation and columns may reduce detection limits further and shorten analysis time while maintaining resolution.
- Automation and sample prep miniaturization: Automated digestion/oxidation and on-line dilution systems will improve reproducibility and throughput for high-volume testing labs.
- Standardization and regulatory alignment: Continued harmonization of methods across regions (EN, ISO, national regulations) will facilitate global compliance testing and data comparability.

Conclusion
The described Shimadzu LC-ICP-MS method, aligned with EN 16802 and EN 17374 principles, provides a rapid, sensitive, and robust solution for quantifying total inorganic arsenic in diverse food and animal feed matrices. The approach reliably oxidizes As(III) to As(V), separates iAs from organic arsenic and chloride-related interferences, achieves regulatory-level detection limits, and yields accurate results validated by spike recoveries and CRMs. It is suitable for routine monitoring and regulatory compliance testing, particularly where time, sensitivity, and matrix robustness are priorities.

Reference

1. COMMISSION REGULATION (EU) 2023/915 of 25 April 2023 on maximum levels for certain contaminants in food.
2. DIRECTIVE 2002/32/EC OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 7 May 2002 on undesirable substances in animal feed.
3. Commission Regulation (EU) 2025/1891 of 17 September 2025 amending maximum levels of inorganic arsenic in fish and other seafood.
4. U.S. House of Representatives, The Baby Food Safety Act of 2021.
5. EN 16802:2016 Determination of inorganic arsenic in foodstuffs of marine and plant origin by anion-exchange HPLC-ICP-MS.
6. EN 17374:2020 Determination of inorganic arsenic in animal feed by anion-exchange HPLC-ICP-MS.
7. T. Taniguchi et al., J. Anal. At. Spectrom., 1999, 14, 651–655.
8. T. Nakazato et al., J. Anal. At. Spectrom., 2000, 15, 1546–1552.