Determination of Mercury in Food by Atomic Fluorescence Spectrometry (AFS)

Mercury is one of the most strictly regulated toxic contaminants in the global food supply, and atomic fluorescence spectrometry (AFS) has become one of the most sensitive and cost-effective techniques for its determination at trace levels. This article explains the principle of cold vapor generation atomic fluorescence spectrometry (CVG-AFS) for measuring total mercury in food, covering sample digestion, instrumentation, calibration, quality control, and relevant standards such as GB 5009.17 and GB 2762. It is intended for food manufacturers, regulators, and accredited third-party laboratories that require reliable mercury testing.

Technical Overview

Mercury (Hg) is a highly toxic heavy metal that enters the food chain through industrial emissions, contaminated water and soil, and agricultural inputs. In food it occurs mainly as inorganic mercury and as organomercury compounds such as methylmercury, which bioaccumulates strongly in fish and is a potent neurotoxin. Because adverse health effects occur at very low exposure levels, national and international frameworks impose strict maximum limits on mercury in food, making accurate trace-level measurement essential.

Principle of Atomic Fluorescence Spectrometry

In the cold vapor generation AFS technique specified in GB 5009.17, an acidified sample digest is mixed on-line with an alkaline potassium borohydride (KBH4) solution. The borohydride rapidly reduces dissolved Hg(II) to elemental mercury vapor, which is stripped from the liquid phase and swept by a stream of high-purity argon into a quartz atomizer. There, mercury atoms are excited by characteristic radiation at 253.7 nm emitted from a mercury hollow cathode lamp. As the excited atoms return to the ground state, they emit atomic fluorescence at the same wavelength, and the fluorescence intensity recorded by the detector is directly proportional to the mercury concentration in the sample over the working range.

Why AFS Is Well Suited to Mercury Determination

• Exceptional sensitivity, with detection limits at the low microgram-per-kilogram (ppb) level, far below regulatory limits
• Cold vapor generation physically separates mercury from the sample matrix, minimizing spectral and chemical interferences
• Lower instrument acquisition and operating costs compared with ICP-MS
• Simple operation and rapid measurement, typically under one minute per reading
• Wide linear dynamic range and excellent short-term repeatability

Regulatory Framework

In China, GB 2762 establishes maximum levels of mercury in foods, for example on the order of 0.02 mg/kg in rice and other grains, 0.01 mg/kg in vegetables, fruits, and fresh milk, 0.05 mg/kg in meat and eggs, and up to 1.0 mg/kg expressed as methylmercury for predatory fish. Similar limits are set by the Codex Alimentarius General Standard for Contaminants and Toxins in Food and Feed and by other national regulations. Compliance verification depends on validated analytical methods such as the AFS procedure described in GB 5009.17.

Testing Samples

Grains, Cereals, and Pulses

Rice, wheat flour, corn, oats, and legumes are staple commodities routinely screened for mercury because paddy soils irrigated with contaminated water can transfer mercury to the edible grain. Samples are homogenized, and representative portions of 0.2–0.5 g are taken for digestion.

Vegetables, Fruits, and Edible Fungi

Leafy vegetables, root crops, fruits, and mushrooms are analyzed as the edible portion after washing and homogenization. Edible fungi deserve particular attention because several species accumulate mercury efficiently from substrates and the environment.

Meat, Poultry, Eggs, and Dairy

Animal-derived foods are tested to verify that feed-borne mercury exposure does not translate into non-compliant residue levels. Muscle, organ tissues, whole eggs, liquid milk, and milk powder can all be processed by closed-vessel acid digestion prior to AFS measurement.

Fish and Other Aquatic Products

Aquatic foods carry the highest risk because methylmercury biomagnifies through the food chain. Predatory species such as tuna, swordfish, and shark, as well as common freshwater and farmed fish, crustaceans, and shellfish, are frequently monitored under both total mercury and methylmercury limits.

Beverages, Condiments, and Processed Foods

Drinking water, fruit and vegetable juices, alcoholic beverages, edible salts and sauces, canned goods, and infant foods are also tested. Liquid samples may be acidified and measured directly or after gentle digestion, while processed foods are homogenized and digested in the same manner as raw commodities.

Testing Items

The primary testing item is the total mercury content of the food, reported in mg/kg. Using a sample mass of 0.5 g diluted to a final volume of 25 mL, the method detection limit is typically on the order of 0.003 mg/kg with a limit of quantification of about 0.010 mg/kg, comfortably below all current statutory maximum limits.

• Total mercury in grains, vegetables, fruits, and other plant foods
• Total mercury in meat, eggs, dairy, and other animal-derived foods
• Total mercury in fish, shellfish, and processed aquatic products
• Total mercury in beverages, condiments, and infant foods
• Methylmercury speciation in predatory fish by HPLC coupled with AFS detection, when required by GB 2762
• Assessment of results against the applicable maximum limits of GB 2762, Codex, or the destination market

Testing Methods

Sample Preparation

Samples are collected with clean, mercury-free tools and homogenized thoroughly using a blender or grinder. All apparatus is soaked in dilute nitric acid and rinsed with deionized water to eliminate background contamination. Liquids are shaken or stirred before subsampling, and solids are reduced to a uniform paste or powder to ensure representativeness.

Digestion Procedures

Microwave digestion is the preferred approach: 0.2–0.5 g of homogenized sample is weighed into a PTFE vessel, 5–8 mL of high-purity nitric acid is added (optionally with 1–2 mL of hydrogen peroxide), and a temperature-programmed ramp to 180–200 °C completes the decomposition. Alternatives include high-pressure closed-vessel digestion in a laboratory oven at 140–160 °C and hot-plate wet digestion with nitric acid. Closed-vessel digestion is strongly recommended because mercury and its compounds are volatile and can be lost, causing low recoveries. A reagent blank is prepared through every step of the procedure.

Instrumentation and Operating Conditions

• Hydride generation atomic fluorescence spectrometer (HG-AFS) equipped with a mercury hollow cathode lamp
• Photomultiplier negative high voltage: approximately 250–300 V; lamp current: approximately 15–30 mA
• Carrier gas argon (purity ≥ 99.99%) at roughly 300–400 mL/min, with shield gas at 600–1000 mL/min
• Carrier solution: 3–5% (v/v) hydrochloric acid
• Reducing agent: 5–10 g/L potassium borohydride dissolved in 5 g/L potassium hydroxide, freshly prepared and filtered

Calibration and Measurement

A series of mercury working standards, for example 0, 0.10, 0.20, 0.40, 0.80, and 1.0 µg/L, is prepared by serial dilution of a certified mercury stock solution in 5% hydrochloric acid. The calibration curve must show a correlation coefficient of at least 0.999. Reagent blank, standards, quality control samples, and sample digests are then measured sequentially, and the mercury concentration in the digest is converted to a mass fraction in the food by applying the sample mass, final volume, and any dilution factor. Results are reported on the appropriate basis (fresh weight or dry weight) as specified by the client or the regulation.

Quality Assurance and Quality Control

• Reagent and digestion blanks are run with every batch to monitor background contamination
• Certified reference materials, such as national standard rice flour or fish muscle, are analyzed to confirm accuracy
• Matrix spike recoveries are typically required to fall within 80–120%
• Parallel duplicate analyses are performed, with relative deviations generally not exceeding 10%
• Continuing calibration verification checks are carried out at set intervals during the sequence
• Additional rinsing is applied after high-concentration samples to control mercury memory effects in the tubing and atomizer

Applications

• Regulatory compliance testing and market surveillance against GB 2762, Codex, and other national maximum limits
• Import and export food inspection for customs clearance and international trade documentation
• Raw material and finished product screening within food manufacturer food safety management systems
• Monitoring of farmland, irrigation water, and aquaculture environments where mercury transfer to food is suspected
• Investigation of contamination incidents and verification of corrective actions
• Scientific research on mercury speciation, bioaccumulation, and dietary exposure assessment

FAQ

What is the detection limit of the AFS method for mercury in food?

With a typical sample intake of 0.5 g diluted to 25 mL, the method detection limit is approximately 0.003 mg/kg and the limit of quantification is approximately 0.010 mg/kg, which is far below all statutory maximum levels for mercury in food.

Why is closed-vessel digestion recommended?

Mercury and many of its compounds are volatile at elevated temperatures. Open hot-plate digestion can cause mercury losses and falsely low results, whereas microwave or high-pressure closed vessels retain all mercury until the digest is quantitatively transferred, ensuring reliable recovery.

Which standards govern this testing?

The analytical procedure follows GB 5009.17, the national food safety standard method for the determination of total mercury and organomercury in food, while the applicable maximum limits are set by GB 2762. International trade may also reference Codex contaminant limits or the destination market's regulations.

How are total mercury and methylmercury distinguished?

Total mercury is measured directly by cold vapor generation AFS. When methylmercury-specific limits apply, such as for predatory fish, methylmercury is extracted from the sample, separated by high-performance liquid chromatography, and quantified by on-line coupling of the HPLC system to the AFS detector as described in GB 5009.17.

What common problems affect AFS mercury results?

The main issues are background contamination from reagents or glassware, mercury carryover (memory effect) after concentrated samples, incomplete digestion, and calibration drift. These are controlled through acid-cleaned vessels, blank correction, routine rinse steps, matrix spikes, certified reference materials, and continuing calibration verification.

How much sample is needed and how long does testing take?

Laboratories generally request 50–100 g of homogenizable sample, although the method itself consumes only 0.2–0.5 g portions. In accredited third-party laboratories, standard turnaround is typically five to seven working days, with expedited options available for urgent projects.