Sallam et al. 2025 — Heavy metals and fatty acids in Libyan olive oil from traditional vs modern extraction

Sallam and colleagues compared olive oil and olive mill wastewater (OMWW) from one traditional (manual/mechanical pressing) and one modern (three-phase centrifugation) mill in northwestern Libya. Heavy metals (Pb, Cd, As, Hg) were measured by ICP-OES; fatty acid composition by GC-FID. All four sample types (oil-traditional, oil-modern, OMWW-traditional, OMWW-modern) returned cadmium below 0.1 ppm and lead, arsenic, and mercury below the method’s limit of detection. The contribution to the Heavy Metal Index corpus is a small regional null finding for Pb/Cd/As/Hg in Libyan olive oil and a qualitative observation that traditional-mill OMWW carries higher turbidity and visual coloration (qualitative organic-load proxy) than modern-mill OMWW.

Key numbers

Heavy metal concentrations by ICP-OES (Table 2, p. 252; Table 3, p. 252):

• Olive oil, modern extraction: Cd <0.1 ppm; Pb <LOD; As <LOD; Hg <LOD
• Olive oil, traditional extraction: Cd <0.1 ppm; Pb <LOD; As <LOD; Hg <LOD
• OMWW, modern extraction: Cd <0.1 ppm; Pb <LOD (ND); As <LOD (ND); Hg <LOD (ND)
• OMWW, traditional extraction: Cd <0.1 ppm; Pb <LOD (ND); As <LOD (ND); Hg <LOD (ND)

Detection threshold: methods text (p. 251) states LOD <0.01 ppm for all four metals analyzed. Table 2 (oil + OMWW heavy-metal table) reports Cd as “<0.1 ppm” in all four sample rows. Table 3 (OMWW-only table) lists 0.1 ppm in its dedicated detection-limit column for Cd, with the Pb/As/Hg detection-limit cells left blank. The paper does not reconcile the methods-text claim (<0.01 ppm LOD) with the tables’ “<0.1 ppm” reporting convention. The page reports “<0.1 ppm” as the paper’s reporting notation for Cd in both matrices; whether 0.1 ppm is the analytical LOD or only the reporting precision is unclear from the source.

Qualitative OMWW comparison (Table 3, p. 252): traditional-mill OMWW showed “slightly higher” turbidity and visual coloration than modern-mill OMWW; the paper interprets this as a proxy for higher organic load and possible greater metal retention in traditional effluent, but does not quantify it.

Fatty acid composition by GC-FID (Table 1, p. 251) — modern vs traditional, %:

• Myristic acid (C14:0): 0.001 vs 0.02
• Palmitic acid (C16:0): 20.32 vs 16.15
• Heptadecanoic acid (C17:0): 0.03 vs 0.04
• Stearic acid (C18:0): 2.50 vs 3.00
• Arachidic acid (C20:0): 0.27 vs 0.00
• Lignoceric acid (C24:0): 0.01 vs 0.05
• Palmitoleic acid (C16:1): 2.00 vs 1.50
• Heptadecenoic acid (C17:1): 0.008 vs 0.05
• Oleic acid (C18:1): 59.01 vs 57.03
• Gadoleic acid (C20:1): 0.11 vs 0.47
• Linoleic acid (C18:2): 16.72 vs 19.45
• Linolenic acid (C18:3): 0.43 vs 0.24

Sampling design: triplicates per facility per matrix; n=3 oil samples and n=3 OMWW samples per facility; one traditional + one modern facility. Statistical analysis by one-way ANOVA in SPSS v26 at p<0.05.

Physicochemical parameters (Discussion p. 253; the values Figure 2 actually plots despite its heavy-metal caption) — modern vs traditional:

• Saponification value: 201 vs 197.7
• pH: 6.03 vs 5.92
• Density: 0.918 vs 0.913 g/cm³
• Peroxide value: 9.83 vs 9.71 meq O₂/kg
• UV absorbance at 232 nm: 2.3 vs 2.28
• UV absorbance at 270 nm: 0.21 vs 0.19
• Iodine value: 80.3 vs 80.0

These are not on a numbered table in the paper; they appear only in the Discussion text and in Figure 2 (which is mis-captioned as a heavy-metal figure but plots these seven physicochemical parameters as a grouped bar chart).

Methods

ICP-OES (instrument vendor not specified by the paper) was used for heavy metal quantification per AOAC Official Method 2015.01 and ISO 11885:2007. OMWW samples were filtered, acidified to pH<2 with HNO₃, and digested by microwave-assisted acid digestion with an HNO₃/H₂O₂ mixture. Olive oil samples were digested with the same HNO₃/H₂O₂ mixture and diluted in ultrapure water. The ICP-OES system was calibrated with multi-element standard solutions and quality controlled with certified reference materials (CRMs); CRM identity not specified. The methods text reports LOD <0.01 ppm for all metals; Table 3 quotes the Cd detection limit as 0.1 ppm.

Fatty acid composition was measured by GC-FID per IOC method COI/T.20/Doc. No. 33/Rev.2 (2017). Fatty acids were transesterified to methyl esters with methanolic KOH and n-heptane, separated on a 30 m × 0.25 mm × 0.25 µm capillary column (column phase not specified) with N₂ carrier at 1 mL/min, oven program 180 °C for 10 min then ramped to 220 °C at 3 °C/min, FID detection. Identification and quantification against retention times of certified FAME standards.

Sample collection: 2023-2024 olive harvest, northwestern Libya. Three replicates per facility per matrix. Oil samples stored in sterilized dark-glass bottles at 4 °C; OMWW in polyethylene containers acidified to pH<2 with HNO₃, stored at 4 °C until analysis.

Analytical scope: the paper reports total Pb, total Cd, total As, and total Hg by ICP-OES; no speciation between inorganic and organic arsenic, no methylmercury/inorganic mercury speciation, and no Cr(VI)/Cr(III) discrimination is reported. ICP-OES is not a routine technique for mercury at trace levels — sensitivity is typically inferior to cold-vapor AAS or AFS — but the paper claims an LOD of <0.01 ppm for Hg without reporting the spectral line or method-specific Hg validation.

Implications

Certification: Provides a small regional null finding (Pb, Cd, As, Hg all at or below 0.1 ppm) for Libyan olive oil and OMWW. The IOC and EU/Codex regulatory ceiling for Pb in olive oil is 0.1 mg/kg = 100 µg/kg = 0.1 ppm; this study’s reporting limits do not enable a meaningful comparison against that ceiling (the data are consistent with 0 µg/kg, with 50 µg/kg, or with 99 µg/kg — all would read as “<LOD” given the stated 0.01 ppm method floor for Pb and the absence of any quantified value). The study cannot support certification-relevant statements beyond “no exceedance was detected.”

Courses: A pedagogically useful caution about the difference between “below LOD” and “below regulatory ceiling”: null findings at a 10 µg/kg (0.01 ppm) detection floor cannot distinguish between “essentially zero” and “two-thirds of the regulatory limit.” For Libyan/North African regional context the paper documents that one modern three-phase mill and one traditional press in the northwestern region both produce oil that does not exceed reporting limits.

App: Single-study regional data point for Libyan olive oil; treat as confirmatory of the broader Mediterranean literature pattern (Pb/Cd typically <100 µg/kg in commercial olive oil) rather than as primary evidence. The qualitative OMWW turbidity/coloration finding is consistent with the broader OMWW literature (Achak et al. 2009; Tarchouna et al. 2010) that traditional-mill effluent carries higher organic loads.

Speciation caution: All four analytes are reported as totals (ICP-OES inherently does not speciate). The page records them as Pb, Cd, tAs, tHg accordingly.

Wiki pages this source may touch

• olive-oil
• olives
• olive-oil
• lead
• cadmium
• arsenic-total
• mercury-total

Verification notes

• 2026-06-01 ingest (Claude Opus 4.7, MFK_najsp32025-54t2): fresh page from the Manual Fetch Kimi / June 1 batch. Three identity checks (DOI — none declared in source; raw_handle MFK_najsp32025-54t2; cite-key sallam2025-olive-oil-libya-traditional-modern) all negative against wiki/sources/; no merge-enhance target.
• DOI: NAJSP does not appear to assign DOIs to its 2025 issue 3(1) articles based on the article PDF and journal masthead; doi: left null. E-ISSN 2959-4820 and the volume/issue/pages string preserve the citation identifier.
• License: NAJSP appears to be an open-access journal (full text freely downloadable from najsp.com without paywall) but the PDF does not declare a specific Creative Commons license. Recorded as “open-access (no explicit CC license declared)” pending verification against the journal’s policy page.
• Analyte vocabulary: metals: [Pb, Cd, tAs, tHg]. ICP-OES does not speciate arsenic or mercury. The paper writes “As” and “Hg” throughout without speciation; the wiki page uses tAs and tHg per the speciation-discipline rule (CLAUDE.md Part 14).
• Paper-internal inconsistency 1 (Cd detection limit): Methods text on p. 251 states “the limits of detection (LOD) were <0.01 ppm for all metals analyzed.” Table 3 on p. 252 lists the cadmium detection limit as 0.1 ppm. The page records both as the source reports them and flags the inconsistency under Key numbers rather than silently resolving it.
• Paper-internal inconsistency 2 (oleic acid in Results text): Page 251 Results paragraph quotes “oleic acid (C18:1) at 57.76%” as the modern-extraction value, but Table 1 on the same page lists modern=59.01% and traditional=57.03%. The Abstract (p. 248) and Discussion (p. 253) both cite 59.01% and 57.03%. The 57.76% figure in the Results paragraph appears to be a typo for the modern value. The wiki page uses the Table 1 / Abstract / Discussion values (59.01 / 57.03) and flags the Results-text figure as inconsistent.
• Paper-internal inconsistency 3 (palmitic acid in Results text): Page 251 Results paragraph quotes palmitic acid at 19.07%; Table 1 modern row lists 20.32% and traditional row 16.15%. The wiki page uses the Table 1 values.
• Paper-internal inconsistency 4 (linoleic acid in Results text): Page 251 Results paragraph quotes linoleic acid at 16.41%; Table 1 modern row lists 16.72% and traditional row 19.45%. The wiki page uses the Table 1 values.
• Paper-internal inconsistency 4b (additional Results-text vs Table-1 fatty-acid mismatches, surfaced by 2026-06-01 audit subagent): Page 251 Results paragraph also quotes palmitoleic acid (C16:1) at 2.46%, stearic acid (C18:0) at 2.77%, and linolenic acid (C18:3) at 0.83% as the modern-extraction values. Table 1 modern column lists 2.00%, 2.50%, and 0.43% for these same three acids. The page Key numbers section uses the Table 1 values throughout; the Results-text figures appear to be alternate-reading or earlier-draft values that the authors did not reconcile with the final table.
• Paper-internal inconsistency 5 (Figure 2 caption): Figure 2 (p. 252) is captioned “Heavy Metal Concentrations in Olive Oil Samples Extracted by Modern and Traditional Methods (ppm)” but the bar chart it shows displays physicochemical parameters (Saponification Value, pH, Density, Peroxide Value, UV 232 nm, UV 270 nm, Iodine Value), not heavy metal concentrations. The heavy metal data are in Table 2 only. Caption/figure mismatch noted; the page uses Table 2 for heavy metals as Tables are the authoritative report.
• Paper-internal inconsistency 6 (arachidic acid in traditional oil): Table 1 reports C20:0 = 0.00% for traditional but Results text (p. 251) lists arachidic acid among “minor components … detected” in trace amounts (<0.5%). The page reproduces the table value as written.
• Mercury at trace levels via ICP-OES: ICP-OES sensitivity for Hg is generally inferior to cold-vapor AAS or AFS. The paper claims LOD <0.01 ppm for Hg without showing Hg-specific method validation. The Verification note flags this but the page preserves the source’s claim.
• Brand firewall (CLAUDE.md Part 12): The paper describes “one traditional facility” and “one modern facility” in northwestern Libya without naming brands, mill operators, or commercial product identifiers. No brand-firewall edits required.
• Sample size n=12 (3 oil + 3 OMWW per facility × 2 facilities). The Key numbers section notes the small n and single-facility design as material limitations.
• Slug check against taxonomy snapshot (2026-05-18): olive-oil exists in both ingredients and products; olives exists in ingredients; lead, cadmium, arsenic-total, mercury-total exist in metals. olive-mill-wastewater is used as a matrix descriptor only (matrices is bare strings, not wikilinks).
• Evidence tier C rationale: regional journal (NAJSP, founded 2023, modest indexing); very small sample (n=3 replicates × 2 facilities × 2 matrices); single-facility-per-category design; multiple text/table/figure-caption inconsistencies within the paper (see notes above); all heavy metals at or below stated detection thresholds (null findings only). Contribution to the corpus is confirmatory regional context rather than primary literature.
• 2026-06-01 audit application (Claude Opus, this session, applying findings from a fresh-context general-purpose audit subagent): subagent verdict REVISE, with three ⚠️ concerns and no ❌ definite errors. Findings applied: (a) added the Discussion’s seven physicochemical parameters (saponification, pH, density, peroxide, UV232, UV270, iodine — values Figure 2 actually plots despite its heavy-metal caption) to the Key numbers section, closing the figure-caption reporting gap; (b) tightened the Cd LOD framing in Key numbers to acknowledge that Table 3’s 0.1 ppm column is the OMWW table’s detection-limit column, and that whether 0.1 ppm is the analytical LOD or only the reporting precision is unclear from the source; (c) added three additional Results-text vs Table-1 fatty-acid inconsistencies (palmitoleic 2.46% vs 2.00%, stearic 2.77% vs 2.50%, linolenic 0.83% vs 0.43%) to the existing inconsistency-4 verification note. Findings rejected (false positives): none. The Check-5 soft observation on “Achak et al. 2009; Tarchouna et al. 2010” being paper-own references rather than wiki-side synthesis was already correctly handled in the source page (those citations sit in the App paragraph as the paper’s own context, not as wiki-side synthesis), so no change was needed.

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.