Adler et al. 2019 - selected metals in coffee from Bosnia and Herzegovina
This study measured selected metals in green coffee beans, roasted coffee beans, and Turkish-style coffee infusions prepared from roasted coffee purchased in Sarajevo, Bosnia and Herzegovina. The study is direct coffee occurrence evidence for Cd, Pb, Ni, Al, Cu, Zn, Fe, and Mg, with transfer-to-infusion percentages for six roasted-coffee samples.
Key numbers
Coffee beans (Tables 1a and 1b; microgram/g coffee):
| Matrix | n | Al mean | Ni mean | Pb mean | Cd mean | Cu mean | Zn mean | Fe mean | Mg mean |
|---|---|---|---|---|---|---|---|---|---|
| Green coffee beans | 2 | 4.2 | 0.415 | 0.076 | 0.015 | 14.0 | 3.6 | 60.0 | 1758 |
| Roasted coffee beans | 6 | 4.19 | 0.88 | 0.0169 | 0.0140 | 11.4 | 5.41 | 41.1 | 1992 |
The source also reports major-element means: green beans had Na 18.6, K 19898, and Ca 789 microgram/g; roasted beans had Na 23, K 23817, and Ca 869 microgram/g.
Ranges in roasted coffee beans (Table 1b; microgram/g coffee):
| Metal | Range |
|---|---|
| Al | 1.74-6.01 |
| Ni | 0.69-1.16 |
| Pb | 0.0115-0.0215 |
| Cd | 0.0111-0.0222 |
| Cu | 6.5-16.5 |
| Zn | 4.96-6.18 |
| Fe | 35.2-49.1 |
| Mg | 1800-2127 |
Coffee infusions (Table 2; n=6; microgram/100 mL):
| Metal | Mean | Range |
|---|---|---|
| Al | 5.63 | 2.03-8.43 |
| Ni | 1.29 | 0.78-1.82 |
| Pb | 0.257 | 0.133-0.558 |
| Cd | 0.048 | 0.036-0.061 |
| Cu | 3.04 | 1.20-6.86 |
| Zn | 7.62 | 5.53-13.17 |
| Fe | 15.33 | 8.93-24.50 |
| Mg | 7000 | 4600-8463 |
Major-element infusion means were Na 44.8, K 99200, and Ca 1378 microgram/100 mL. The highest average transfer from roasted coffee to infusion was reported for K and Mg (>83% and >70% respectively); intermediate transfer was reported for Na, Pb, Ca, Ni, Zn, and Al (about 38% to 27%); the lowest transfer was for Fe, Cu, and Cd (<11% each).
Penetration from roasted coffee to infusion (Table 3; percent):
| Metal | Mean transfer pattern from source table |
|---|---|
| Pb | 3.2-10.3% across six infusions |
| Cd | 21.5-47.6% across six infusions |
| Ni | 13.6-55.3% across six infusions |
| Al | 15.4-39.2% across six infusions |
The authors concluded that measured metal intakes at current Bosnia and Herzegovina coffee-consumption levels remained within the limits they considered safe for health.
Methods (brief)
Coffee beans were purchased in small local stores in Sarajevo, Bosnia and Herzegovina. Two green coffee samples and six roasted coffee samples were tested; two green/roasted pairs represented the same beans before and after roasting. Coffee infusions were prepared by boiling 10 +/- 0.1 g roasted ground coffee in 200 mL water for 5 minutes, cooling to 20°C, and mineralizing a 25 mL aliquot.
Coffee beans (1 +/- 0.01 g) were digested with concentrated nitric and perchloric acids, then diluted to 25 mL. Infusion aliquots were mineralized with nitric and perchloric acids and diluted to 50 mL. Mineralization used a Speedwave Xpert high-pressure microwave mineralizer. Metals were measured with a Hitachi Polarized Zeeman Atomic Absorption Spectrometer ZA3000 Series. Ca, K, Na, and Mg were measured by FAAS; Fe, Cu, Zn, Al, Pb, Cd, and Ni were measured by graphite-furnace AAS. Calibration used 1000 mg/L standardized solutions; ERM-BB422 fish muscle reference material gave recoveries of 95-105% and precision of 1.2-10.1%.
Implications
Standards work: This source provides coffee occurrence evidence on both a dry-bean and infusion basis. The infusion results are especially useful for distinguishing product-as-sold concentrations from drink-as-consumed exposure.
Courses: Useful for teaching process effects: roasting changed some bean concentrations, but the more important consumer-exposure step is transfer into the brewed infusion.
App: Supports coffee Cd/Pb/Ni/Al context with an as-consumed infusion basis. The page should not be used to infer instant-coffee powder concentrations, because the study used roasted beans prepared as Turkish-style coffee.
Microbiome: Not addressed.
Wiki pages this source may touch
Verification notes
- Fresh auto-fetch ingest 2026-05-19 from the gap-driven coffee-Cd wishlist PDF.
- Strict brand firewall: the paper reports anonymized sample codes only. This page reports aggregate matrix-level values and does not reproduce sample-code rows as product identifiers.
- Methods vendor/reference-material names are retained under Part 12 Exception 2 for instrument, calibration, and certified-reference-material documentation.
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.