Graphitic carbon nitride

C₃N₄ (idealised); real materials contain hydrogen

Also called melon, polymeric carbon nitride

nanosheet semiconductor

A metal-free yellow polymer made by simply heating urea or melamine, which uses visible light to split water into hydrogen or to break down pollutants. It is cheap, non-toxic and very stable, which explains its enormous research popularity; its weakness is that it converts light into chemical energy far less efficiently than practical solar fuels would need.

Key properties

  • Band gap ~2.7 eV with a conduction band edge suitable for hydrogen evolution
  • Made from abundant precursors – urea, melamine, dicyandiamide – at ~500–600 °C in air
  • Visible-light hydrogen evolution first shown in 2009 with a platinum co-catalyst; quantum efficiencies of standard preparations are low
  • Chemically stable in acids, bases and under prolonged illumination

How it is made

  • Thermal polycondensation of urea, melamine or dicyandiamide at ~550 °C – kilogram batches, poorly crystalline
  • Thermal or liquid exfoliation of bulk material into nanosheets with higher surface area
  • Ionothermal synthesis in molten LiCl/KCl – more crystalline poly(triazine imide) and poly(heptazine imide)

Uses, and how close they are

  • Photocatalytic hydrogen evolution and CO2 reductionlab
  • Pollutant degradation and water disinfectionprototype
  • Metal-free electrocatalyst supports and sensorslab

Readiness runs lab → prototype → pilot → deployed.

Open problems

  1. What is the true structure of typical ‘g-C3N4’ samples, and how much does incomplete condensation limit charge transport?
  2. Can quantum efficiency be raised enough, without platinum co-catalysts, for practical solar hydrogen?
  3. Are reported performance gains comparable between labs, given differences in light sources and measurement protocols?

Going deeper

Short notes for specialists. Choose a lens in the header and yours comes first.

For theoreticians · your lens

Idealised heptazine or triazine sheets give gaps and band edges that differ from real, hydrogen-rich melon, so models should include buckling, stacking and defects. Excitons are strongly bound and charge separation is slow, which means band-edge alignment from DFT is necessary but not sufficient to predict photocatalytic activity.

For experimentalists · your lens

Report precursor, heating rate, temperature and atmosphere – all change the structure. Characterise with XRD (the (002) stacking peak near 27.4° 2θ for Cu Kα), FTIR (heptazine breathing mode near 810 cm−1), XPS N 1s and elemental analysis. For photocatalysis, report apparent quantum yield at a defined wavelength, not only µmol per hour per gram.

For engineers · your lens

The cheapest and most scalable photocatalyst in this catalogue, but low efficiency limits solar-fuel applications. Near-term niches are water treatment and self-cleaning or antibacterial surfaces.

Recent news

The newest items tagged g-C3N4, from the news feed updated 5 Oct 2026.

All 3 items tagged g-C₃N₄ in the news feed  ·  RSS feed for g-C₃N₄

Key references

  1. Unmasking melon by a complementary approach employing electron diffraction, solid-state NMR spectroscopy, and theoretical calculations — structural characterization of a carbon nitride polymerLotsch et al. · Chemistry – A European Journal 13, 4969 (2007)cited by 970doi:10.1002/chem.200601759
  2. A metal-free polymeric photocatalyst for hydrogen production from water under visible lightWang et al. · Nature Materials 8, 76 (2009)cited by 12,603doi:10.1038/nmat2317
  3. Exfoliated graphitic carbon nitride nanosheets as efficient catalysts for hydrogen evolution under visible lightYang et al. · Advanced Materials 25, 2452 (2013)cited by 2,535doi:10.1002/adma.201204453
  4. Graphitic carbon nitride (g-C3N4)-based photocatalysts for artificial photosynthesis and environmental remediation: are we a step closer to achieving sustainability?Ong et al. · Chemical Reviews 116, 7159 (2016)cited by 7,161doi:10.1021/acs.chemrev.6b00075