Abedi et al. 2025 - Emerging versus conventional bleaching methods for edible oils

Abedi and colleagues review industrial oil bleaching (IB) and four emerging alternatives — ultrasound-assisted bleaching (UAB), high-voltage electric field-assisted bleaching (HVAB), microwave-assisted bleaching (MWAB), and membrane-assisted bleaching (MAB) — across pigment removal, sterol and tocopherol retention, oxidative indices, heavy metals removal, clay consumption, and adsorption kinetics and thermodynamics. Heavy-metal-relevant content is confined to four short subsections (one per bleaching technology) plus scattered table entries. The authors summarise that bleaching clays remove trace metals (notably Fe, Cu, and a broader set including Pb, Cd, Ni, Cr, Co, Al) from crude vegetable oils by adsorption, that higher temperatures and ultrasonic or electric-field assistance enhance the metal-removal efficiency relative to conventional bleaching, and that incomplete neutralisation upstream can leave residual Ca, Mg, Pb, and P in the bleached oil. The review reports no original measurements and presents no occurrence distributions for heavy metals in finished retail edible oils.

Key numbers

All values below are reported by the review as drawn from cited primary studies. The review does not contribute new measurements.

Section 2.4 — Industrial bleaching effect on heavy metals (p. 8, citing Silva et al. 2014 and Abedi et al. 2021)

The review further states that increasing bleaching temperature by about 10 °C can roughly double the reaction rate and enhance trace-metal adsorption onto acid-activated bleaching clay (Abedi et al. 2021).

Section 3.1.5 — Ultrasound-assisted bleaching (UAB) effect on heavy metals (p. 15, citing Abedi et al. 2015, 2021)

The review states that applying ultrasound during soybean oil bleaching produced a “notable decrease in the levels of Fe and Cu compared to the IB method” via cavitation-induced abrasion of the clay surface, exposing fresh adsorption sites. No numeric Fe or Cu reduction value is supplied for the UAB-vs-IB comparison in this subsection beyond the IB values above; the review notes a comparable mechanism for phosphorus and soap removal (Abedi et al. 2015, 2021).

Section 3.2.7 — HVEF-assisted bleaching effect on heavy metals (p. 19, citing Abedi et al. 2016, 2020 and Mousavifard et al. 2024)

The review states that the mechanism for metal removal under HVAB relates to the medium’s ionic strength in the presence of Fe²⁺ and Cu²⁺ and that convection in the electric field enhances cation transport to the electrode surfaces. No quantitative reduction percentages or before/after values are reported in this subsection.

Section 3.3.4 — Microwave-assisted bleaching (MWAB) effect on heavy metals (p. 20, citing Sarafraz Ardakani et al. 2023)

The review reports that microwave treatment at 600 W of sesame oil during refining was associated with removal of metal cations (Al, K, Fe, Pb, Cd, Ni, Cu, Cr, Co) via interaction with bleaching clays. Post-bleaching analysis in the cited primary study revealed elevated concentrations of Mg, Ca, Pb, and P ions, which the review attributes to potential contributions from wash water and adsorbents and to incomplete removal of these elements during the neutralisation stage. No concentration values, sample sizes, or analytical limits of detection are quoted in the review for the Sarafraz Ardakani et al. 2023 measurements; the review’s heavy-metal-specific content for MWAB consists of a single paragraph and a qualitative mechanism statement.

Tabulated entries with explicit heavy-metal numbers in the review’s data tables

Table 2 (USAB conditions and outcomes, pp. 9–13) lists three rows that include explicit metal-reduction percentages copied from the underlying primary studies:

(The “300%” phosphorus reduction in the first row reproduces what the review prints; a reduction greater than 100% is not possible in the literal sense and is presumably a typesetting carry-over from the review’s table — see Verification notes.)

Table 1 (IB conditions and outcomes, pp. 5–6) lists two rows that include metal-reduction figures:

The review’s body text in Section 2.4 paraphrases these two table entries.

Other elements named in the review

The text and tables also mention Mn (magnesium increase patterns in UAB), Ca, Mg, K, and P. Of these, only K is a candidate HMI analyte category and only when speciated as a context element; Ca, Mg, and P are nutritional minerals outside the HMI corpus and are tracked here only because they co-occur with the Fe, Cu, Pb, Cd, Ni, Cr, Co, Al cations in the cited primary work. The review does not measure or quantify them.

Evidence Fitness

This is a narrative review of edible oil bleaching technologies. It contributes no primary heavy-metal measurements and no occurrence distributions. The quantitative heavy-metal figures it presents are second-hand summaries of the underlying primary studies (Silva et al. 2014, Abedi et al. 2021, Sarafraz Ardakani et al. 2023, De Jesús-Hernández et al. 2023, Gharsalli et al. 2025, Essid et al. 2016, and others). Public evidence label “Context only” is appropriate.

The source supports process-context claims about how bleaching unit operations in edible-oil refining adsorb trace metals from crude oils into bleaching clay residues, and points to specific primary studies where the underlying observations live. It does not support occurrence claims for Pb, Cd, Ni, Cr, Co, Al, Cu, or Fe in finished retail edible oils and should not be used as the basis for HMTc threshold work or for ingredient contamination_profile values on any vegetable-oil ingredient page. Threshold work on edible oils should rely on the primary occurrence studies referenced here.

Methods (brief)

Narrative literature review without an explicit search-and-selection protocol or PRISMA-style inclusion criteria. The review compiles published characterisation and process-outcome data on industrial and emerging bleaching technologies from cited primary studies; analytical methods for the underlying primary measurements are not re-described in detail in the review’s tables and would need to be retrieved from each cited study (e.g., Silva et al. 2014 for palm oil P/Fe figures, Abedi et al. 2021 for soybean oil Fe/Cu figures, Sarafraz Ardakani et al. 2023 for sesame oil Pb/Cd/Ni/Cr/Co/Al figures). The review separately describes the adsorption isotherm, kinetics, and thermodynamic frameworks applied across the cited bleaching studies (Langmuir, Freundlich, Temkin, Hill-de Boer, Toth, BET isotherms; pseudo-first-order, pseudo-second-order, Elovich, Weber–Morris intraparticle diffusion kinetics; Van’t Hoff thermodynamic parameters), but again does not contribute new measurements.

The review’s data tables present results in mixed bases (mg/kg, ppm, percentage reductions, and percentage increases for some mineral elements). No single analyte distribution is reported on a consistent unit basis suitable for cross-study pooling within the review itself.

Implications

Certification: Process context only. The review supports the general claim that the bleaching step of edible-oil refining adsorbs trace Fe and Cu (and, where neutralisation is upstream and complete, Pb, Cd, Ni, Cr, Co, and Al) from crude oil into the bleaching clay, and that ultrasound and electric-field assistance can enhance this adsorption relative to conventional bleaching. The review provides no occurrence values for finished retail edible oils that should inform HMTc thresholds; primary occurrence studies of bottled or bulk retail oils remain the appropriate basis.

Courses: Useful as background reading for explaining the edible-oil refining sequence (degumming, neutralisation, bleaching, deodorisation) and the role of bleaching clay as a trace-metal sink. The Sarafraz Ardakani et al. 2023 reference is the only Pb/Cd/Ni/Cr/Co-quantifying study cited in the review and is worth pursuing as primary evidence for edible-oil heavy-metal occurrence in sesame oil refining.

App: No ingredient contamination_profile impact. The review contributes no concentration values usable for the consumer-app pipeline.

Wiki pages this source may touch

• lead
• cadmium
• nickel
• chromium
• cobalt
• aluminum
• copper
• iron
• vegetable-oil
• vegetable-oils
• soybean-oil
• sunflower-oil
• sesame-oil
• olive-oil
• rapeseed-oil
• corn-oil
• rice-bran-oil
• olive-oil
• cooking-oils-other

Verification notes

• 2026-06-02 fresh ingest (Claude Opus 4.7, manual-fetch ingest from raw/Manual Fetch Kimi /June 1 more/): DOI 10.1002/fsn3.71121 verified against the article’s first page; submission 30 May 2025, revision 16 September 2025, acceptance 5 October 2025; published in Food Science & Nutrition 2025, 13:e71121. The article carries the “Open Access” badge and the Creative Commons Attribution licence statement at the bottom of page 1; license recorded as CC BY 4.0.
• The review is structured around five bleaching technologies (IB, UAB, HVAB, MWAB, MAB), with one short heavy-metals subsection per technology and supporting table entries. Total heavy-metals-relevant body content in the review is approximately one paragraph per technology plus two rows in Table 1 and three rows in Table 2.
• The review distinguishes the broad term “heavy metals” loosely, sometimes including essential trace elements (Fe, Cu, Mn, Co) and macromineral elements (Ca, Mg, K, P) alongside HMI’s primary toxic-metal analytes (Pb, Cd, Ni, Cr, Al). The Section 3.3.4 paragraph on Sarafraz Ardakani et al. 2023 is the only place in the review that explicitly lists Pb, Cd, Ni, Cr, Co, and Al as a combined cation set; the frontmatter metals: array reflects this list plus Fe and Cu from the other subsections.
• The review does not separate inorganic from total arsenic (no arsenic measurement is discussed at all) and does not separate Cr-VI from total Cr. The frontmatter uses Cr (total) accordingly and does not list arsenic or mercury.
• The Table 2 USAB row attributed to De Jesús-Hernández et al. 2023 prints “Reduction in phosphor by 300%“. A reduction greater than 100% is not physically possible; the same row reports “Increase in magnesium by 150%” which is plausible as a percentage gain. The 300% figure is reproduced verbatim from the review’s table as printed in the PDF; the primary De Jesús-Hernández et al. 2023 paper would resolve whether this is a typo for “300% increase in phosphor reduction relative to a baseline”, a unit-label error, or a printing artefact in the review. Flagged here because the value is implausible at face.
• No brand names are reported in the review’s heavy-metals body content. The review does name several bleaching-clay product lines and process equipment by trade designation (Calriant Tonsil 4120AFF, Clariant Supreme 112FF, Taiko Omega 1, Amcol Mineral Bent Actigel) in Table 1; these are scientific-method materials (the adsorbent doing the bleaching, not the oil being measured) and fall under the Part 12 Exception 2 carve-out for instrument and reference-material vendor names. They are retained in the source page only by reference to Table 1 and not reproduced as a list here.
• jurisdictions: is left empty: this is a global narrative review citing primary studies from multiple regions. The Iranian institutional affiliations of the authors do not bound the scope of the review.
• The review covers palm, sunflower, soybean, olive, rapeseed/canola, sesame, corn, rice bran, hempseed, linseed, safflower, cottonseed, and post-fermentation corn oils. The taxonomy snapshot includes slugs for soybean-oil, sunflower-oil, olive-oil, rapeseed-oil, sesame-oil, corn-oil, and rice-bran-oil; palm-oil, hempseed-oil, linseed-oil, safflower-oil, and cottonseed-oil are not present in the current ingredient taxonomy. Frontmatter therefore uses the umbrella vegetable-oil / vegetable-oils slugs plus the seven per-oil slugs that exist; the missing slugs (palm, hempseed, linseed, safflower, cottonseed) are not invented here.
• evidence_tier: B reflects the review-of-reviews / narrative-review character of this source. evidence_fitness: EF-4 (Context only) and public_evidence_label: Context only reflect that the source contributes no primary heavy-metal occurrence values.
• 2026-06-02 fresh-context audit (general-purpose subagent) — verdict REVISE. Two ❌ findings on Check 1 numerical fidelity, both targeting the Silva et al. 2014 palm oil row in Section 2.4. Independent re-verification via pdftotext -layout extraction of page 8 resolves the exact source text as: “Silva et al. (2014) reported a reduction in phosphorus (to 2.7 ± 0.1 mg/kg) and iron (to 0.2 ± 0.1 mg/kg) contents after bleaching crude palm oil with 3% acid-activated bleaching clay.” The “to” sits inside the parenthetical and specifies the post-bleaching final value, not a from→to range across crude/bleached; the ± is a real standard deviation. The original wiki page misread this as a from→to range between P (2.7) and Fe (0.2) values; the audit subagent misread it as a from-to range “2.7 to 0.1” / “0.2 to 0.1” within each element. Both readings were wrong. Corrected to the actual reading: P post-bleaching 2.7 ± 0.1 mg/kg; Fe post-bleaching 0.2 ± 0.1 mg/kg; pre-bleaching crude values are not quoted in this review. Consistent with Table 1’s “phosphorus (<3 mg/kg) and iron (<0.3 mg/kg)” entries for the same Silva et al. 2014 row.
• All other audit checks returned ✅ on substantive findings: slug vocabulary clean (Check 2), speciation and methods clean (Check 3), Part 12 brand firewall clean with Exception 2 correctly invoked for the bleaching-clay material names (Check 4), Part 2 wiki/HMTc firewall clean (Check 5). One minor ⚠️ on matrices: edible-oil (not in the standard matrices vocabulary but plausible as a parent of vegetable-oil) — retained without change pending matrices-vocabulary review.

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.