Ben Moussa et al. 2022 — Natural Tunisian clays as adsorbents for heavy-metal removal from phosphogypsum waste (Gabes)

Four-page Springer book chapter from a Tunisian–Belgian group testing whether three local clay deposits in the Gabes region (Haidoudi, Romana, Hmaymet) can adsorb heavy metals (the authors emphasise Cd and Cr) from phosphogypsum — the by-product of phosphoric-acid manufacture at the Groupe Chimique Tunisien plant. The paper is qualitative in published form: it asserts that phosphogypsum from Gabes carries “high” concentrations of Cd, Cr, and U “compared to other wastes of phosphogypsum in the world and to the soil remediation limits” without printing the mg/kg values, and reports column-leaching results as “the concentration of metals decreased … all the way down to 0 mg/kg” for the Haidoudi and Romana clays without printing breakthrough curves or numeric removal efficiencies. The Hmaymet clay is reported as “less substantial” in adsorption performance, again without numbers.

Phosphogypsum is upstream of the agricultural-soil cadmium pathway (phosphate-fertiliser raw-material chain), and the paper documents a remediation technology rather than a food-occurrence finding. It is retained in the Heavy Metal Index as a C-tier remediation-leads source for the Tunisian phosphogypsum-storage context, not as occurrence evidence for any food matrix.

Key numbers

The paper does not print the underlying ICP-OES values for Cd, Cr, U, or Th in either the phosphogypsum or the leachate. The numerical statements that survive in the published chapter are:

• Phosphogypsum sampling year: April 2018 (Groupe Chimique Tunisien, Gabes).
• Three clay deposits tested: Haidoudi, Romana, Hmaymet (all from the Gabes area, southern Tunisia).
• Qualitative XRD mineralogy: all three clays contain kaolinite, gypsum, and calcite; quartz and mica appear as accessory minerals. Silicon oxide is the major constituent and is highest in the Hmaymet clay.
• FT-IR signal: the band at 1430 cm⁻¹ is reported as indicating the disappearance of carbonate after purification.
• Qualitative column-experiment outcome: post-leaching metal concentration “decreased … all the way down to 0 mg/kg” for Haidoudi and Romana columns; Hmaymet column gave a positive but “less substantial” reduction.
• The authors flag Cd, Cr, and U as the elements present at the highest concentration in Gabes phosphogypsum relative to international comparators and to soil-remediation limits (Rutherford et al. 1994), but the comparator values and the Gabes values are not printed.
• Radioactivity statement: Ur and Th activity (Bq/kg) in the Gabes phosphogypsum samples was reported as “acceptable and not dangerous to human health” and “lower” than other phosphogypsum samples cited from other countries. Numeric activity values are not printed.

Methods (brief)

Three Tunisian clay deposits (Haidoudi, Romana, Hmaymet) and phosphogypsum waste from the Groupe Chimique Tunisien (Gabes) were sampled, air-dried, and characterised. Mineralogy: X-ray diffraction (Bruker D8 ECO Advance, Cu tube anode, Lynxe Eye XE energy-dispersive position-sensitive detector). Functional-group characterisation: FT-IR (EQUINOX 55 spectrometer, KBr pellet technique). Both XRD and XRF measurements were carried out at the Department of Geology, Research Group Mineralogy and Petrology, Faculty of Sciences, Ghent University, Belgium. Column experiments used three laboratory columns, each with a clay layer, a transition layer, and a phosphogypsum waste layer above the clay; leachate composition was measured by inductively coupled plasma–optical emission spectrometry (ICP-OES, Varian Vista MPX, Varian, Palo Alto, California, USA). The paper does not report column dimensions, packing density, hydraulic loading, leaching solution chemistry, contact time, pH, replicate structure, detection limits, or quantitative removal-efficiency calculations.

Limitations

C-tier conference / book-chapter contribution; four printed pages; no quantitative data tables for either the input phosphogypsum metal content or the post-leaching effluent metal content. Removal claims are stated only as “decreased … down to 0 mg/kg” without breakthrough curves, isotherm fits, kinetic models, or replicate measurements. No detection limits or method-blank data are reported, so the “0 mg/kg” claim cannot be distinguished from “below detection by ICP-OES under the unreported analytical conditions”. Sample size (single composite per matrix?) is not specified. The chapter’s framing emphasises Cd and Cr as “the most hazardous metals” but the underlying ICP-OES dataset for those elements is not printed; the reader is told that Gabes phosphogypsum exceeds soil-remediation limits but neither the Gabes concentration nor the soil-remediation limit is printed. No statistical analysis. No comparison against a control column without clay. No food matrix is measured. Sampling year is 2018; chapter published 2022. Two of five authors are co-surnamed Moussa; the Belgian co-author (J. De Grave) is associated with the analytical-instrumentation contribution at Ghent University rather than the field sampling. Reference list is short (nine references — Beveridge & Pickering 1983; Briffa et al. 2020; Dammak et al. 2014; Eloussaief et al. 2011; Futalan et al. 2011; Masindi & Muedi 2018; Rutherford et al. 1994; Rutherford et al. 1995; Tran et al. 1999) and does not include the Tunisian-clay sorption literature that would normally be expected for a paper of this kind.

Implications

Minimal direct value for the Heavy Metal Index. The wiki’s scope is heavy-metal occurrence in food and personal-care supply chains; this paper is a remediation-engineering study on an industrial waste stream. It is retained as a leads document for two narrow purposes:

Phosphate-rock and phosphogypsum supply-chain context: Phosphogypsum is the principal solid by-product of phosphoric-acid manufacture, and phosphoric acid feeds the phosphate-fertiliser supply chain that is upstream of cadmium accumulation in agricultural soils and cereal grains worldwide. This paper documents that the Gabes (Tunisia) phosphogypsum stream is enriched in Cd and Cr relative to international comparators (per the authors’ qualitative claim; the underlying numbers would need to be retrieved from prior Rutherford et al. or Tunisian-Chemical-Group characterisation papers cited in this chapter). The Tunisian Chemical Group is among the major phosphate-rock processors in the Mediterranean basin, and Tunisian phosphate rock is exported into European and African fertiliser supply chains — this is supply-chain provenance context, not occurrence data.

Adsorbent-material leads: Kaolinite-bearing natural clays from arid regions are documented here as candidate low-cost sorbents for Cd and Cr from phosphogypsum leachates. This is methodological-leads context for the broader low-cost-adsorbent literature already represented in the corpus (e.g., shamkhi2021-wastewater-adsorbents-review) and for environmental-remediation context on cadmium and chromium, not direct occurrence evidence.

Wiki pages this source may touch

• cadmium
• chromium
• uranium

Verification notes

• Ingest mode: NEW page (no prior wiki source page). DOI grep (10.1007/978-3-031-00808-5_99), raw_handle grep (MFK_natural-clays-as-adsorbents-for-the-removal-of-hea), and cite-key glob (moussa2022-*, benmoussa*, *phosphogypsum*, *clay*) all returned no matching wiki source page. The cite-key collision check against the only existing moussa* page (moussa2024-spices-herbs-lebanon) is clean (different year, different topic, unrelated author group).
• Scope rationale. The paper does not measure heavy metals in any food, water, or biological matrix, so matrices: [] is correct — matching the corpus convention for non-food regulatory / methodology / remediation sources (e.g., the toys-and-childcare-articles, hard-goods, and methodology-only subcorpus). products: [] and ingredients: [] are also correct: no food product or food ingredient is sampled. The HMI relevance is supply-chain context for the phosphate-rock → fertiliser → cereal-grain Cd pathway, plus a methodological lead for low-cost adsorbents; neither line of relevance routes to a wiki product or ingredient page.
• Metals list. metals: [Cd, Cr, U] reflects the three elements the chapter reports on either as concentration (Cd, Cr — the focus heavy metals; “high” concentrations stated qualitatively in Section 3.2) or as activity (U — Uranium-238 radioactivity, Bq/kg, Section 3.2). The chapter does not speciate Cr, so the plain Cr tag is used rather than Cr-VI even though chromate species are the more usual concern for clay-sorbent studies — the analytical method (ICP-OES) is total-element by default. No As, Pb, Hg, Ni, Al, Sb, or Sn data are presented despite the broad “heavy metals” framing in the title.
• Thorium handling (audit-driven 2026-06-01). The chapter discusses Thorium-232 in the radioactivity paragraph (Bq/kg activity, never mg/kg concentration). Initial frontmatter included Th in metals:; this was removed after the audit subagent flagged that (a) Th is not in the controlled metal-abbreviation vocabulary used elsewhere in the source-page corpus, (b) wiki/metals/thorium.md does not exist, and (c) the paper reports only activity, not concentration. Th is now mentioned in the Methods and Key numbers sections (radioactivity context) but not asserted as a measured metal in the frontmatter. If Thorium ever needs corpus-wide representation, a metals/thorium page and a Th abbreviation entry should be added as a corpus-vocabulary change before any source page re-asserts Th in metals:.
• Source-content firewall (Part 12). No brand-level food-product values; the Tunisian Chemical Group (Groupe Chimique Tunisien) is named as the operator of the phosphogypsum source plant, which is a public-record industrial-source identification rather than a food-product brand attribution — analogous to the regulatory-event carve-out for naming the operator of a public-record contamination source. Method-instrument vendors (Bruker D8 ECO Advance, EQUINOX 55, Varian Vista MPX) are named per the scientific-reproducibility carve-out for methods.
• Wiki/HMTc firewall (Part 2). No synthesis claims, no HMTc threshold proposals, no consumer translations. Implications section is scoped to supply-chain provenance and adsorbent-material leads, not to “what this means for any specific food category”.
• License. Springer Nature exclusive license (©The Author(s), under exclusive license to Springer Nature Switzerland AG 2022). Not open access; treat full-text reuse beyond fair-use quotation as license-restricted.
• Numerical fidelity. All quantitative statements in this page that originate in the source chapter are restatements of qualitative/descriptive text or of single named numbers (1430 cm⁻¹ FT-IR band, the three clay-deposit names, the 2018 sampling year, the eight-reference list). The chapter does not print mg/kg tables for input or output streams; this page therefore does not assert mg/kg occurrence values.

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.