Chetima et al. 2024 - Activated carbons from neem and cotton husks as a bleaching-earth alternative for cottonseed oil decolorization

Chetima and colleagues prepare activated carbons (ACs) from neem and cotton husks by phosphoric-acid impregnation followed by carbonization, comparing 6 h conventional (open-air) and 30 min microwave-assisted activation routes. The four resulting ACs (AC1, AC2 from cotton husks; AC3, AC4 from neem husks) are characterised for burn-off, iodine index, methylene blue index, specific surface area, and surface functional groups, then evaluated as decolorizing adsorbents for neutral cottonseed oil at 70, 80, and 90 °C against commercial bleaching earth. The paper reports adsorption kinetics (intraparticle diffusion and pseudo-second-order models) and equilibrium isotherms (Langmuir and Freundlich) for cottonseed-oil pigment uptake. Heavy-metal content is absent: the study quantifies no analyte in HMI’s panel (Pb, Cd, iAs, tAs, tHg, MeHg, Ni, Al, Cr, Cr-VI, Sn, Sb, U), in cottonseed oil, in the activated carbons, in the spent adsorbents, or in any other food matrix. The source supports a single corpus-level claim only: activated carbons derived from neem and cotton husks can decolorize crude cottonseed oil with 30-80 % higher efficiency than the bleaching earth currently used in industrial refining, providing process context for the heavy-metal-vector role that bleaching earth plays in conventional vegetable-oil refining.

Key numbers

All values are reported by the authors from their own laboratory measurements. No heavy-metal concentrations are reported anywhere in the paper.

Table 1 — Iodine Index, Methylene Blue Index, Burn-off of the four activated carbons (p. 4)

Authors report specific surface areas derived from the iodine index of 843.32, 1002.92, 1481.72, and 1162.52 m²/g for AC1-AC4 respectively (Section 3.1.2, p. 4). Per the Lua-Gu (cited as Ref. [25]) classification used by the authors, an AC is microporous when burn-off is below 50 % and macroporous when above 75 %; AC1-AC4 fall in the mixed-pore 50-75 % band (AC4 marginally below at 50.83 %).

Table 2 — Surface functional groups of the four activated carbons (p. 5)

Values are reported with the column headers “Carboxylic”, “Lactone”, “Phenolic”, “Acid”, and “Basic” without explicit units in the table; the cited prior method (Ref. [24], Chetima et al. 2018) reports these as mmol/g.

The authors note that surface functional-group distribution depends on starting material but not on the impregnation method, while iodine and methylene blue indices and pore structure depend on both.

Pseudo-second-order kinetic equilibrium adsorption q_e (mg pigment/g AC) for cottonseed-oil decolorization at 70, 80, 90 °C (Table 3, p. 6)

R² values for the pseudo-second-order model range from 0.7305 (BE at 80 °C) to 0.9932 (AC3 at 70 °C); the authors conclude that pseudo-second-order kinetics describe the adsorption adequately for ACs and acceptably for BE.

Decolorization-relevant qualitative findings (Sections 3.2.1-3.2.3, p. 4-7, and Conclusion, p. 11)

• At 70 °C, only AC3 (neem, conventional) shows materially higher pigment uptake than BE; AC1, AC2, AC4 are comparable to or below BE.
• At 80 °C, microwave-activated cotton AC (AC2) shows the highest pigment uptake; conventional cotton AC1 follows; both microwave-activated samples outperform their conventional counterparts at this temperature.
• At 90 °C, neem-derived ACs (AC3 and AC4) outperform cotton-derived ACs and BE. Authors attribute the temperature effect to chemisorption increasing with temperature and to lower oil viscosity facilitating pigment diffusion.
• The temperature effect is statistically significant (p < 0.05) for all adsorbents.
• Adsorbent-dose comparison at 80 °C with microwave-activated ACs (1 %, 2 %, 3 %): for neem AC there is no significant difference between 2 % and 3 %, both higher than 1 %; for conventional cotton AC there is no significant difference between the three doses. Authors recommend 2 % as the optimum.
• Stated optimum decolorization conditions in the Conclusion: 90 °C, 40 min, 2 % adsorbent loading.
• Stated efficiency advantage in the Conclusion: prepared ACs are 30-80 % more efficient than the bleaching earth currently used in cottonseed-oil decolorization industries.
• Microwave activation is preferred to conventional activation because it requires 1 h compared with 6 h for ambient-air activation while delivering comparable or better decolorization performance.

Langmuir and Freundlich isotherm parameters (Table 4, p. 9-10)

The R² values for both isotherm models exceed 0.91 across the three temperatures, with the authors reporting that the dual fit indicates a mix of physical and chemical (monolayer plus multilayer) adsorption. The Freundlich constant n is reported as less than 1 for several entries at 70 °C, which the authors interpret as weak adsorbate-adsorbent forces. The full parameter set is in the source Table 4 and is not reproduced here because the heavy-metal-relevant claims of the paper do not turn on the isotherm constants.

Process-context statements

• “Currently, in most tropical countries, the decolorization of cottonseed oil is done using imported bleaching earth. Importing this material is expensive and adds significantly to the overall cost of producing vegetable oil … It would therefore be interesting to look for potential sources of bleaching agents capable of carrying out the decolorization of vegetable oils at a reduced overall cost, and which are renewable and biodegradable” (p. 2).
• “All the activated carbons prepared in this work were 30-80 % more efficient in pigment adsorption than bleaching earth that is normally used in decolorizing neutral cotton seed oil in industries” (abstract; restated in Conclusion).
• “The prepared activated carbons have great potential to replace bleaching earth currently used in the cotton oil industry and should be tested in column reactors and on a pilot scale to ascertain these results. The quality of the decolorized oil with these ACs should also be investigated” (Conclusion, p. 11).
• No quantitative claim about heavy-metal levels in the oil, the bleaching earth, the activated carbons, or the spent adsorbents is made anywhere in the paper.

Evidence Fitness

This is a primary laboratory adsorption study about decolorization performance, not a heavy-metal occurrence study. The paper measures pigment adsorption (UV-visible absorbance at 664 nm) and adsorbent physical-chemical properties (burn-off, iodine index, methylene blue index, surface functional groups by Boehm titration as cited from Ref. [24]). It does not measure any HMI analyte in any food matrix.

Public evidence label “Context only” is appropriate. The source contributes one corpus-level claim: activated carbons from neem and cotton husks are a viable alternative to bleaching earth for cottonseed-oil decolorization, achieving 30-80 % higher pigment uptake than bleaching earth under the reported conditions. This is heavy-metal-relevant only by adjacency: bleaching earth is the conventional vegetable-oil bleaching agent and is documented elsewhere in the corpus (see sigauke2024-spent-bleaching-earth-review, abedi2025-edible-oils-bleaching-review) as a trace-metal vector through the refining unit operation. The Chetima et al. study does not quantify whether the husk-derived ACs themselves transfer or remove any heavy-metal load, and that question would need to be tested in a follow-on study with ICP or similar elemental analysis before any HMTc or threshold inference could be drawn.

The source does not support occurrence claims for Pb, Cd, As, Hg, Ni, Al, Cr, Sn, Sb, or U in any food matrix and should not be used as the basis for HMTc threshold work or ingredient contamination_profile values.

Methods (brief)

Activated-carbon preparation. Neem husks (Mora locality, Cameroon) and cotton husks (SODECOTON, Maroua, Cameroon) were washed, dried, crushed, and sieved to particle size below 50 µm. Charring was performed in a Nabertherm GmbH (USA) muffle furnace at 600 °C for 3 h on 50 g batches. Charred samples were then treated with 14 N and 15 N phosphoric acid for 2, 6, 12, or 24 h at room temperature (conventional activation) or for 15, 30, 45, 60, 90, or 120 min in a microwave (microwave activation); the source does not state explicitly which acid normality was used with which feedstock. After activation, ACs were washed with distilled water and dried at 105 °C for 24 h. Preliminary bleaching screening at 80 °C for 20 min selected the four ACs reported (AC1-AC4) for the full kinetic and isotherm experiments.

AC characterization. Burn-off, methylene blue index, iodine index, specific surface area, and surface-group concentrations were measured following the protocol described in the authors’ prior publication (Ref. [24], Chetima et al. 2018, Processes 6(3), doi: 10.3390/pr6030022); the present paper does not re-state instrumental detail. Surface functional groups are reported under headings “Carboxylic”, “Lactone”, “Phenolic”, “Acid”, and “Basic”, consistent with Boehm titration as described in the cited 2018 paper.

Decolorization experiments. Neutral cottonseed oil (27 mL, 15 mg AC mass interpretation per Section 2.4.1 phrasing — see Verification notes) was bleached at 70, 80, or 90 °C with continuous stirring for contact times of 5-40 min. Adsorbent-to-oil mass ratios of 1, 2, and 3 % were screened. After each run the oil was filtered through Whatman N°1 filter paper and absorbance measured at 664 nm with a UV-visible spectrophotometer. Decolorization efficiency E was computed as E = (A₀ − A)/A₀ × 100 (Eq. 1, p. 3). Replicate handling and replicate count are not stated in the paper.

Kinetic and isotherm modelling. Intraparticle-diffusion (Eq. 4) and pseudo-second-order (Eq. 5) kinetic models, and Langmuir (Eq. 6) and Freundlich (Eq. 7) isotherm models were fit to the time- and concentration-resolved decolorization data. Statistical software, replicate counts, and error-propagation methods are not stated.

Heavy-metal analysis. None. No elemental analysis (ICP-MS, ICP-OES, AAS, XRF) of oil, adsorbent, or residue is performed or referenced for this study; the only mention of heavy metals in the entire paper appears in the title of Ref. [4] (Hoang et al., Chemosphere) cited in the introduction’s general statement that activated carbons are used for remediation of heavy-metal-polluted waters.

Implications

Certification: Process context only. The source documents that a viable, biomass-derived alternative to imported bentonite bleaching earth exists for cottonseed-oil decolorization, but provides no heavy-metal occurrence data that should inform HMTc thresholds for finished cottonseed oil or vegetable oils generally. Any HMTc question about the relative trace-metal load contributed by activated-carbon-based bleaching vs. bentonite-based bleaching would require a primary elemental-analysis study, which this paper does not provide.

Courses: Useful as a worked example for the edible-oil-refining module — specifically the bleaching unit operation — to illustrate that the choice of bleaching adsorbent is not fixed and that biomass-derived ACs are an active research area in tropical-oil-producing regions where imported bleaching earth carries cost and supply-chain disadvantages. Should be cross-referenced with sigauke2024-spent-bleaching-earth-review (bentonite-based SBE characterisation) and abedi2025-edible-oils-bleaching-review (industrial and emerging bleaching methods, including a four-bleaching-technology comparison that does not include biomass-AC bleaching).

App: No ingredient contamination_profile impact. The paper measures no HMI analyte in any matrix.

Microbiome: Not addressed.

Wiki pages this source may touch

• vegetable-oil
• vegetable-oils
• cooking-oils-other

Verification notes

• DOI 10.1016/j.heliyon.2024.e24060 prints on the first page. Received 15 July 2023; received in revised form 6 December 2023; accepted 3 January 2024; available online 5 January 2024. Published as Open Access under CC BY-NC-ND 4.0 (license URL printed on p. 1). 2405-8440 ISSN. Published by Elsevier Ltd.
• “This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors” (Funding statement, p. 11).
• Two of HMI’s matrices vocabulary slugs (cottonseed-oil and activated-carbon-adsorbent) are descriptive analogues used in the absence of controlled-vocabulary slugs for them in the matrices list; both should be confirmed against the matrices vocabulary on Karen’s next taxonomy review. The umbrella vegetable-oil slug is included alongside.
• metals: [] is correct. The paper measures no heavy metal in any matrix. The only occurrence of the phrase “heavy metals” in the entire article is in the title of cited Reference [4] (Hoang et al., Chemosphere) in the Introduction; the Chetima et al. study itself does not analyse, quantify, or estimate any heavy-metal content.
• No food brand names are present in the body of the paper. The only commercial entities named are activated-carbon feedstock suppliers (SODECOTON for cotton husks; women processors in Mora locality for neem fruits) and laboratory-equipment vendors (Nabertherm GmbH, USA, for the muffle furnace; Whatman for filter paper). Per the Part 12 brand-firewall scientific-method exception, instrument and reagent vendor names are permitted in Methods sections; no Part 12 calls were required during ingest.
• The activated-carbon feedstocks (neem husks, cotton husks) are agricultural waste streams from neem-oil processing and cotton-seed processing respectively. They are not food ingredients and do not warrant their own wiki/ingredients/ slugs.
• Cottonseed oil is the matrix bleached in this study but is not currently a controlled-vocabulary ingredient slug; vegetable-oil and vegetable-oils umbrellas are used. A cottonseed-oil ingredient slug would be appropriate if Karen later opts to disaggregate vegetable-oil pages by source oil. Surfaced here as a slug-proposal note, not created during ingest (per CLAUDE.md Part 10: ingredients are auto-stubbed at freq-2 by the system, not by the model during ingest).
• Table 1 prints the iodine index of AC3 as 1117.6 and AC4 as 914.40, with AC1 as 711.20 and AC2 as 812.80. Section 3.1.2 (p. 4) restates these as “711.2; 812.8; 1117.6; 914.4 mg/mL”. The values agree across table and text; units of mg/mL for iodine index are unusual (mg/g or mg AC consumed per g sample are more standard) and follow the authors’ prior 2018 publication’s convention. Reproduced as printed; the unit oddity is a property of the source’s reporting convention, not an ingest error.
• The abstract states “iodine index (300-5000 mg/g)” and “methylene blue index (300-5000 mg/g)” as the range of values “indicating the potential of the prepared activated carbons”. The phrase “300-5000” does not match the Table 1 ranges (iodine 711-1118, MBI 307-7309). This appears to be an abstract-side approximation or typographic slip in the source; the Table 1 values are reproduced verbatim and the abstract figure is not propagated into the wiki page.
• Section 2.4.1 reads “decolorization of oil was carried out with 27 mL (15 mg) of neutral cotton seed oil”; the parenthetical “(15 mg)” is inconsistent with 27 mL of oil (which would mass roughly 24 g for cottonseed oil at density ~0.91 g/mL) and may refer instead to the AC mass loading or to a typo in the source. The same paragraph states adsorbent doses of 1, 2, and 3 % were evaluated. Reproduced as printed in the Methods (brief) section with the inconsistency flagged here; the kinetic and isotherm data tables do not depend on resolving it because per-percent-dose results are reported.
• Audit subagent (2026-06-02) flagged AC2 q_e at 80 °C in Table 3 (pseudo-second-order block) as 69.44 in the initial ingest; independently re-verified against PDF Table 3 on p. 6 — the source prints AC2 q_e=66.67 at 80 °C (k_i=0.006, R²=0.9879) and q_e=69.44 at 90 °C (k_i=0.0062, R²=0.99). The wiki value at 80 °C was a transcription error in the initial ingest and has been corrected to 66.67. The earlier verification note that framed the value as a source-side typographic duplication has been removed because the duplication did not exist in the source.
• Audit subagent (2026-06-02) flagged the methods-section attribution of “14 N (cotton) or 15 N (neem)” phosphoric acid normality as over-specified. Independently re-verified against PDF Section 2.2 (p. 2): the source states “treated with 14 N and 15 N phosphoric acid for 2, 6, 12, and 24h at room temperature (conventional activation) and 15, 30, 45, 60, 90, and 120 min in a microwave (microwave activation) respectively” — the per-feedstock acid-normality mapping is not stated. Methods section softened to reproduce the undifferentiated phrasing.
• Langmuir and Freundlich parameter values in Table 4 include several entries with negative K_L or K_F (e.g., AC1 K_L = −0.293 at 70 °C; AC2 K_F = −95.35 at 70 °C). Negative isotherm constants are physically meaningless and indicate poor model fit at the low-temperature end. The R² values reported for the same rows (0.979 and 0.915) suggest the linearized regression converged on a mathematical fit that does not correspond to a meaningful adsorption model. Reproduced as printed; the authors’ own discussion (Section 3.3.2) acknowledges that Freundlich n less than 1 implies weak adsorbate-adsorbent forces but does not directly address the negative-constant entries.
• Jurisdictions are left empty. The study is performed in Cameroon (University of Maroua, University of Bamenda) but the conclusions are framed as a general process recommendation for cottonseed-oil refining, not a Cameroon-specific regulatory claim.
• Near-duplicates: none identified. The same author group’s prior paper (Chetima et al. 2018, Processes 6(3)) on activated-carbon decolorization of cottonseed oil is cited as Reference [24] but is a distinct earlier study using different activation methods; not present in the wiki corpus at the time of this ingest.

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.