de Souza et al. 2022 - trace elements in fruit juices by ICP OES and ICP-MS
This study measured essential and toxic trace elements in 21 fruit juice and nectar samples from Spain and Portugal. Commercial matrices included orange, tangerine, red berries, peach, pineapple, apple, grape-containing blends, mango-orange juice, passion-fruit nectar, and purple-pear nectar, with two fresh-squeezed orange juices and one fresh-squeezed lemon juice prepared in the laboratory. The strongest HMT&C signal is a single apple-juice Pb exceedance, while As and Sb were not detected and Cd was only quantified in two samples.
Key numbers
Minor/toxic trace elements in juice and nectar samples (Table V; µg/L):
| Element | Detection pattern / range | Highest or notable value |
|---|---|---|
| As | Not detected in all 21 samples; sample LOD 2.1 µg/L | all <2.1 µg/L |
| Sb | Not detected in all 21 samples; sample LOD 1.6 µg/L | all <1.6 µg/L |
| Pb | Detected only in Apple 1 | 61.7 ± 7.2 µg/L per results text (Table V rounds to 62 ± 7); the authors state this value exceeds the 0.03-0.05 mg/kg regulatory limits cited from Codex, EU Reg 2021/1317, and ANVISA IN 88/2021 |
| Cd | Quantified only in Apple 1 and Pineapple 1 | Apple 1: 2.6 ± 0.1 µg/L; Pineapple 1: 2.8 ± 0.1 µg/L |
| Co | Detected in 13 of 21 samples | 1.2-37.8 µg/L per results text; maximum in passion-fruit nectar (Table V reports 38 ± 2 for the passion-fruit-nectar maximum; the small text/table discrepancy is preserved as a source-internal note) |
| Cr | Detected in all samples except laboratory-squeezed Orange 4 and Orange 5 | 3.1-18.9 µg/L |
| Cu | Detected in 19 samples | 99.2-501.4 µg/L; maximum in Pineapple 2 |
| Mo | Quantified in 20 samples | 0.8-5.9 µg/L; maximum in Pineapple 2 |
| Ni | Detected in 14 samples | 16.6-113.9 µg/L; maximum in passion-fruit nectar |
| V | Quantified in juice samples | 11.7-91.4 µg/L; maximum in Orange 1 |
Major trace / essential elements (Table VI and results text):
| Element | Concentration pattern |
|---|---|
| Al | Quantified in 7 of 21 samples, 0.4-1.5 mg/L; maximum in Orange 1 |
| Fe | Quantified in 20 of 21 samples, 0.16-3.7 mg/L; maximum in a pineapple-apple-grape blend |
| Mn | Quantified in all samples, 0.114-12 mg/L; highest values in pineapple and pineapple-containing juices |
| Zn | Quantified in 18 of 21 samples, 0.08-0.9 mg/L |
| Mg | 17.5-104.7 mg/L; high in pineapple, orange, tangerine, and pineapple-apple-grape juices |
| Ca | 10.2-96.5 mg/L in the results text; high in orange and pineapple-containing juices |
The authors conclude that toxic elements did not pose a general health risk in this sample set, but they explicitly flag the Apple 1 Pb concentration as above current regulations.
Methods (brief)
Commercial juice bottles were manually shaken immediately before subsampling. Each 2.5 mL aliquot was digested with 3.0 mL 69% nitric acid, 1.0 mL 33% hydrogen peroxide, and 2.0 mL ultrapure water in 100 mL closed Teflon microwave vessels. The microwave program ramped to 90°C, then 140°C, then 180°C and held at 180°C for 15 minutes. Digests were diluted to 25 mL and analyzed by ICP OES for Ca, Mg, and Zn and by ICP-MS for Al, As, Cd, Co, Cr, Cu, Fe, Mn, Mo, Na, Ni, Pb, Sb, and V.
Major-element analysis used a Perkin Elmer Optima 3300 DV ICP OES. Trace-element analysis used a Perkin Elmer NexIon 300X ICP-MS with kinetic energy discrimination and Rh/Y internal standards. The method validation covered linearity, precision, recovery, and limits of detection/quantification; analytical recoveries ranged from 89% to 108% for the main method-validation table (Mg recovery 108 ± 5).
Implications
Standards work: This source is useful low-to-mid occurrence evidence for fruit juices and nectars, with a concrete single-sample Pb outlier in apple juice. It is best used as direct occurrence evidence for fruit-juice rows and as a cautionary apple-containing juice example rather than as a high-percentile distribution source, because n=21 is small and fruit types are heterogeneous.
Courses: Useful for explaining that “fruit juice” evidence can mix commercial products, nectars, and laboratory-squeezed samples. It also shows why a single exceedance should be preserved without overstating the whole set.
App: Supports fruit-juice risk flags for Pb, Cd, Ni, Cr, Al, and matrix-specific essential-element context. The app should preserve the source’s total Cr measurement as Cr, not chromium(VI).
Microbiome: Not addressed.
Wiki pages this source may touch
- aluminum
- arsenic-total
- cadmium
- cobalt
- chromium
- copper
- iron
- magnesium
- manganese
- molybdenum
- nickel
- lead
- antimony
- vanadium
- zinc
- fruit-juice
- apple-juice
- orange-juice
- grape-juice
- mango-juice
- fruit-juice-not-canned
- fruit-juices-apple-containing
- fruit-juices-non-apple
Verification notes
- Fresh auto-fetch ingest 2026-05-19 from the gap-driven fruit-juice Co wishlist PDF.
- Numerical values are preserved in the source units: µg/L for Table V and mg/L for Table VI. Because juice density is close to 1 kg/L, these are approximately ppb and ppm respectively, but no conversion is used in the primary table.
- The paper reports total As, total Cr, and total Sb only; no arsenic, chromium, or antimony speciation was performed.
- Strict brand firewall: the paper refers to fruit types and sample labels only; no commercial juice brands are named in this source page.
- Methods vendor/equipment/reagent names are retained under Part 12 Exception 2.
- Fresh-context audit (Claude subagent, 2026-05-19): REVISE → applied. Three findings corrected: (1) Pb row leads with the precise 61.7 ± 7.2 µg/L text value with Table V rounding noted, plus explicit citation of the EU 2021/1317 and ANVISA IN 88/2021 comparators the paper invokes; (2) Co row notes the text-vs-Table-V discrepancy (37.8 vs 38 ± 2); (3) recovery range corrected from 89-103% to 89-108% to capture the Mg 108 ± 5 maximum. Slug vocabulary clean, speciation discipline correct (tAs/Cr-total/Sb explicitly labeled), brand firewall clean (sample labels anonymized as in source), HMTc firewall clean.
Page history
The five most recent substantive edits to this page. The full version history lives in git; when DOI minting comes online (see schema docs), each entry below will also link to a version-pinned DataCite DOI.