CMMMech
CMMMech > Primary resource > Dunite

Dunite primary resource

Dunite is an ultramafic rock and geological feed, not a mineral species or elemental magnesium. BGS Rock Classification Scheme code DUN defines this peridotite type by more than 90% olivine in the assemblage of olivine, pyroxene and hornblende. The curated experiments used three powders from a North Carolina dunite deposit, including strongly altered material with much lower remaining olivine fractions. Their laboratory results do not describe every dunite, a pure forsterite feed, a waste tailing, or a traded product grade. No criticality designation is inherited from constituent elements.

Microbial & Biochemical Mechanisms (1)

Conditional metal leaching from dunite-deposit powders with bacterial biolixiviant

bioleaching Microbial

Associated Organisms: Gluconobacter oxydans (NCBITaxon:442)

Glucose-grown Gluconobacter oxydans NRRL B-58 produced an acidic biolixiviant that mobilized Mg, Mn, Fe, Co and Ni from the tested rock powders. Compared with deionized water, release increased; at 1% pulp density the advantage over initially pH-matched gluconic acid was small, and removing cells after biolixiviant production did not significantly change short-term release. The supported mechanism is biologically produced leaching chemistry, without demonstrated cell-contact necessity, an identified additional active metabolite, or mineral-specific flux.

Experimental Context & Analytical Limits:
Weathered exterior was cut from North Carolina Day Brook dunite-deposit samples before tungsten-carbide milling. Main Methods report DUN-1 as 64 wt% olivine, 27% lizardite, 5% antigorite and 4% talc; DUN-2 as 37% olivine, 59% lizardite, 2% talc and 2% meixnerite; and DUN-3 as 92% olivine, 5% antigorite, 2% lizardite and 1% talc. These are author-reported XRD proportions, not newly confirmed phase assignments. DUN-1 supplies main Figures 2-6, DUN-2 Figure 7, and DUN-3 the early-time Figure S3 according to main Methods. B58 was grown in yeast-peptone-mannitol medium at 30 degrees C and 200 rpm, back-diluted to OD590 0.05 and grown to about 1.5 over 48 h. Splitting culture into 7.5-mL portions and adding equal volumes of 40% glucose gave 15 mL at 20% glucose, followed by 48 h biolixiviant production. The main short assays used 1% wt/vol powder, 22 degrees C and 24 h with shaking; Figure S1 depicts 200 rpm during leaching. Wild-type initial biolixiviant pH was 2.39 +/- 0.02. Analytical samples were centrifuged through 0.45-micrometre filter plates at 1500 x g for five minutes, acid-diluted and measured by ICP-MS. Thus results concern operational filtrate elemental concentrations, not oxidation-state-resolved ions or a weighed recovered product. This is laboratory work.
Curated Evidence Points:
Support Source: lee2025
Results and Figure 2 show higher Mg, Mn, Fe, Co and Ni in the DUN-1 biolixiviant arm than in the water control after 24 h. The authors report approximately 300 ppm Mg and 11% extraction, with extraction defined relative to bulk-rock Mg content times pulp density. Mg was only about 6% higher than in initially pH-matched gluconic acid. Figure 3 supports further release over 24-96 h; the stated 73% increase averages five metals and is not a Mg-only recovery yield.
Partial Source: lee2025
Figure 4 compares 0.22-micrometre-filtered biolixiviant with unfiltered material at the same initial pH; the authors report no significant difference, with p-values 0.2-0.9. Its control is described as sugar water, distinct from the Figure 2 water control. Main Methods describe splitting cultures into triplicates, Figure 3 calls its measurements duplicates, and Figure 4 mentions duplicate colonies while the dataset lists three rows per treatment. Independent biological replication is therefore not uniformly established. This experiment does not prove microbial necessity during the leaching stage.
Figure S5 finds no significant 24-h effect of stoppering, with p-values 0.3-0.6; gas restriction is not verified anoxia. Figure S4 preserves much poorer Cr extraction and high variability in the cellulosic-hydrolysate arm, including negative corrected estimates. Tables S1-S2 describe bulk oxide/trace-element composition rather than mineral-specific release; Table S2 defines n.d. as not determined, not measured zero. The paper's opening greater-than-90%-olivine characterization cannot describe the reported DUN-1/DUN-2 powders.
Support Source: lee2025
Figure 7 compares DUN-2 at 5%, 10% and 60% pulp over 24-72 h with initially pH-matched gluconic acid. At 60% pulp, the reported final Mg concentration ratio is about 3.25 to one. This does not isolate pulp density across the DUN-1 versus DUN-2 studies: their phase proportions differ. Equal initial pH does not match acid inventory, buffering or subsequent pH trajectories. Additional biologically produced leaching agents remain proposed rather than chemically identified.
Partial Source: lee2025
Figure 6 reports an average approximately 12% increase over 24-96 h with the B58 delta-pstS, P112:mgdh derivative; its initial pH was 2.01 +/- 0.02 versus wild-type 2.39 +/- 0.02. The double modification and unequal chemistry do not isolate either gene's mineral-specific causal effect. Figure 5 reports lower mean Mg release with corn-stover cellulosic hydrolysate despite a nonsignificant Mg comparison; small sample size and variability do not establish equivalence. Hydrolysate controls were subtracted to account for feedstock metals.
Dataset S1 has treatment, time, dilution and feed metadata, but stale worksheet labels: Figure 4 contains high-pulp DUN-2 rows, Figure 5 filtered/unfiltered DUN-1 rows, Figure 6 glucose/hydrolysate rows, and Figures 3&7 the DUN-1 wild-type/mutant time series. These are matched by their actual content, not by worksheet title alone. Concentrations are instrument ug/L with separate dilution factors; they are not native mg/kg observations. Censored tokens such as less-than-zero displays are not imported as numerical zero.
Figure S2 assigns DUN-2 to Figure 4, whereas main Methods and supplementary Table S1 assign it to Figure 7. Figure S1 shows 48 h for initial culture growth while main Methods state 24 h. Figure 8 displays 12 attempts but main Methods/Table S1 use labels extending to 14. These source inconsistencies remain unresolved; no precise long-duration attempt-to-feed quantitative claim is curated.
Partial Source: lee2025
Figure 8 and its carbon-investment equations estimate potential carbonation assuming that each mobilized Mg can later store one CO2. They do not measure carbonate production, net atmospheric removal, life-cycle emissions, or industrial operation. The authors state that post-leaching pH around 4-5 is unfavorable for carbonate formation and that complete starting/ending mineral mass balances and surface area measurements remain future work. Solution release here establishes neither isolated-metal recovery nor an integrated sequestration process.

Sources & References

  1. bgs_dunite: British Geological Survey Rock Classification Scheme, Dunite, code DUN; description and definition referencing Research Report RR/99/06, section 5.1 and Figures 15-16. Online entry checked 11 October 2026.
    https://webapps.bgs.ac.uk/bgsrcs/rcs_details.cfm?code=DUN (accessed 2026-10-11)
  2. lee2025: Lee et al. (2025). Bio-accelerated weathering of ultramafic minerals with Gluconobacter oxydans. Scientific Reports 15:15134, published 30 April 2025; DOI 10.1038/s41598-025-99655-9. Results, Figures 2-8, Materials and methods, and Limitations of the study.
    https://doi.org/10.1038/s41598-025-99655-9 (accessed 2026-10-11)
  3. lee2025_si: Lee et al. (2025), Supplementary Information, 14 pages; Figures S1-S5, Tables S1-S4 and Note S1. Same study as DOI 10.1038/s41598-025-99655-9.
    https://media.springernature.com/original/springer-static/esm/art%3A10.1038%2Fs41598-025-99655-9/MediaObjects/41598_2025_99655_MOESM2_ESM.pdf (accessed 2026-10-11)
  4. lee2025_data: Lee et al. (2025), Dataset S1 XLSX; nine worksheets of ICP-MS measurements, dilution, duration, pulp density and substrate metadata. Worksheet names do not consistently match the published main figure numbers.
    https://media.springernature.com/original/springer-static/esm/art%3A10.1038%2Fs41598-025-99655-9/MediaObjects/41598_2025_99655_MOESM1_ESM.xlsx (accessed 2026-10-11)
  5. ncbi442: NCBI Taxonomy 442, Gluconobacter oxydans, species rank; authority XML checked 11 October 2026. Species identity does not identify the B58 mutant.
    https://www.ncbi.nlm.nih.gov/Taxonomy/Browser/wwwtax.cgi?id=442&mode=Info (accessed 2026-10-11)
  6. ncbi_b58: NCBI BioProject PRJNA761721, Gluconobacter oxydans strain NRRL B-58 genome sequencing and assembly; organism taxon 442 and strain NRRL B-58. This is strain context, not a genome identity assertion for the leaching mutant.
    https://www.ncbi.nlm.nih.gov/bioproject/761721 (accessed 2026-10-11)

Curation History

Timestamp Curator Action Summary
2026-10-11T11:13:17Z codex (GPT-6; codex) CREATE Created dunite resource record with primary bacterial-biolixiviant evidence, altered-feed identity boundaries, control nulls and source uncertainties.

Record source: data/records/dunite.yaml