logo
ResearchBunny Logo
Discovery of carbon nanotubes in sixth century BC potteries from Keeladi, India

Chemistry

Discovery of carbon nanotubes in sixth century BC potteries from Keeladi, India

M. Kokarneswaran, P. Selvaraj, et al.

This groundbreaking research conducted by Manivannan Kokarneswaran, Prakash Selvaraj, Thennarasan Ashokan, Suresh Perumal, Pathikumar Sellappan, Kandhasamy Durai Murugan, Sivanthan Ramalingam, Nagaboopathy Mohan, and Vijayanand Chandrasekaran explores ancient black coatings on pottery sherds from Keeladi, India, revealing advanced nanomaterial synthesis with remarkable stability over 2600 years.... show more
Introduction

The study investigates the nature of a unique, shiny black coating found on the inner walls of pottery sherds excavated from Keeladi, Tamil Nadu, India, dated by radiocarbon to the sixth–third century BC. The research question centers on whether the coating contains nanostructured carbon, such as carbon nanotubes (CNTs), and what its presence implies for ancient materials technology. Contextually, prior reports have identified nanomaterials in historical artifacts (e.g., metallic nanoparticles in lusterware and CNTs in Damascus steel), indicating ancient craftspersons exploited special material properties without formal scientific understanding. This work provides evidence that the Keeladi black coating contains CNTs, making it, to the authors’ knowledge, the oldest artifact exhibiting such nanostructures, and highlighting the potential intentional or incidental production and use of mechanically robust coatings in antiquity.

Literature Review

Earlier studies documented nanomaterials in ancient artifacts: metallic nanoparticles (e.g., Ag, Cu) in Islamic and Renaissance luster pottery, used to enhance visual luster (10th–11th century AD). Multi-walled carbon nanotubes (MWCNTs) were reported in Damascus sabres (16th–18th century), linked to superior mechanical strength. These findings demonstrate that nanostructured materials occurred in historical crafts. However, no prior report documented CNTs in artifacts as early as the Keeladi potteries (6th century BC).

Methodology
  • Samples: Inner black coating on Keeladi pottery sherds was visually identified as shiny, hard, and enduring. Up to 1 mm of the inner black layer was removed with a surgical/diamond knife.
  • Raman spectroscopy: Initial spectra were acquired directly on the cleaned inner surface using a LABRAM-HR confocal micro-Raman spectrometer with 532 nm excitation. Separately, the inner black coating was ground to fine powder, sonicated in 1 ml water (probe sonicator), and fractionated: centrifuged in 50:50 v/v glycerol for 10 min at 2500 RPM; the top layer was collected, washed with acetone, then centrifuged at 1000 RPM for 10 min; this was repeated three times. The resulting suspension was drop-coated on a gold-coated silicon wafer. Micro-Raman measurements on drop-cast material used a LAB-EE 102B-100E CCD (Princeton Instruments) with an Acton SP2500 monochromator and confocal microscope; 532 nm CW laser excitation; room temperature; low laser intensity to avoid damage.
  • TEM: Black powder was sonicated in acetone and deposited on carbon-free Pd and Cu holey grids to avoid carbon contamination; imaged at 200 kV using an FEI Tecnai T20 electron microscope. Minimal sample preparation was employed to preserve original features, with electron-transparent regions used for imaging.
  • XPS: X-ray photoelectron spectroscopy performed on cleaned samples using PHI VersaProbe III with Al Kα source. Survey and high-resolution scans (C 1s, Fe 2p) were collected.
  • Data analysis: Raman peak positions, intensity ratios (ID/IG, I2D/IG), and radial breathing mode (RBM) assessed; crystallite size Le calculated from ID/IG using λ = 532 nm. TEM used to measure tube diameters, wall spacing, and morphology. XPS peak deconvolution identified bonding states and elemental oxidation states.
Key Findings
  • Raman spectroscopy on the inner coating showed characteristic carbon features: G band at 1589 cm⁻¹ (sp² E2g), D-band components at 1354, 1234, and 1473 cm⁻¹ (D1, D2, D3 defect-related), and a broad 2D band around 2600 cm⁻¹. Intensity ratios: ID/IG = 1.28; I2D/IG ≈ 0.6. The G band blue shift and broad 2D are consistent with oxygen functionalization (graphene oxide-like characteristics). Crystallite size Le calculated as ~15 nm (λ = 532 nm).
  • Raman on drop-cast tubular features showed a clear radial breathing mode (RBM) at 236 cm⁻¹, confirming the presence of single-walled carbon nanotubes (SWCNTs).
  • TEM revealed: bundles of SWCNTs; MWCNTs with curling/damage; and stacked sheet-like structures attributed to graphene oxide. Measured metrics: average SWCNT diameter 0.6 ± 0.05 nm (near theoretical lower limit ~0.4 nm); SWCNT wall expansion 0.2 ± 0.1 nm; MWCNT interwall spacing ~0.34 nm (consistent with graphite (002)); MWCNT inner diameter 3 ± 0.15 nm.
  • XPS survey detected C, O, and elements such as Fe, Al, and Si (likely as oxides). C 1s deconvolution indicated sp² carbon domains and oxygen-containing functional groups (carbonyl, ether, carboxyl, alcohol), corroborating Raman. Fe 2p analysis showed Fe in 2+ and 3+ oxidation states; Fe 2p3/2 to 2p1/2 intensity ratio was 1.97 (~2 theoretical).
  • The inner black coating has remained smooth, adherent, and stable for ~2600 years on pottery dated to the sixth–third century BC, indicating robust mechanical/chemical stability.
  • The presence of Fe, Al, and Si suggests potential catalytic roles (notably Fe) in CNT formation during high-temperature pottery firing, with plant-derived carbon as a plausible precursor.
Discussion

The findings address the central question by demonstrating that the Keeladi pottery’s inner black coating contains nanostructured carbon—SWCNTs, MWCNTs, and graphene oxide-like sheets—established via convergent evidence from Raman (including RBM), TEM, and XPS. This implies that ancient firing conditions, possibly involving plant-based carbon sources and iron catalysts (from vegetal matter and/or soil), were sufficient to nucleate and grow CNTs and related carbon structures. The results are significant because they place the emergence of CNTs in human-made artifacts at least as early as the 6th century BC, predating prior reports (e.g., Damascus steel). Such nanostructures could impart enhanced mechanical strength and durability to coatings, possibly offering functional advantages (e.g., robust inner surfaces). The work raises broader implications about ancient materials processing knowledge—whether intentional or incidental—and potential health considerations if vessels with CNT/graphene-based coatings were used for food preparation or storage. The proposed formation mechanism is consistent with known modern CNT synthesis catalyzed by Fe under high temperatures, but remains inferential pending direct experimental replication of the ancient process.

Conclusion

The inner black coating of Keeladi pottery sherds contains bundles of SWCNTs, MWCNTs, and sheet-like structures consistent with graphene oxide. The SWCNTs exhibit an exceptionally small average diameter (~0.6 ± 0.05 nm), approaching theoretical limits, and the coating has remained stable for ~2600 years—making this, to the authors’ knowledge, the oldest known occurrence of such nanostructures in artifacts. These findings suggest that ancient pottery firing conditions, possibly with plant-derived carbon and naturally present iron, fostered CNT formation, potentially conferring mechanical benefits to the coating. Future research should aim to: (i) replicate CNT formation under plausible ancient firing conditions with plant-based precursors and iron-bearing materials; (ii) determine whether such coatings were intentionally engineered for performance; (iii) assess potential biological interactions or toxicity if such vessels were used for food-related purposes; and (iv) broaden analyses across more samples and sites to evaluate prevalence and variability.

Limitations
  • The precise source(s) of carbon and the firing conditions that produced the coating are unknown; formation mechanisms are hypothesized rather than experimentally verified.
  • TEM sample preparation aimed to minimize alteration, but bulk, dark regions limited electron-transparent areas; observed curling and damage in MWCNTs may reflect either original features or preparation-induced artifacts.
  • While XPS indicates the presence of Fe, Al, and Si (likely oxides), their exact mineralogical forms and spatial relationship to CNT nucleation sites were not resolved.
  • The study does not provide direct evidence of the intended function or usage context of the coated vessels, nor does it assess potential health implications of CNT/graphene exposure from such artifacts.
Listen, Learn & Level Up
Over 10,000 hours of research content in 25+ fields, available in 12+ languages.
No more digging through PDFs, just hit play and absorb the world's latest research in your language, on your time.
listen to research audio papers with researchbunny