Researchers have traced several long-debated spectroscopic signals in carbon materials to specific atomic structures, challenging a common interpretation of X-ray photoelectron spectroscopy (XPS) data and providing new context for Raman peak assignments.
The study, published in the Journal of Materials Science, combined Raman spectroscopy, infrared (IR) spectroscopy, XPS, elemental analysis, X-ray diffraction, and density functional theory (DFT) calculations. The team used isotropic pitch-based carbon fiber as a model for carbon materials produced at temperatures of 1,473 K or higher.
Carbon defects affect mechanical strength, thermal conductivity, electrical behavior, and chemical reactivity. Yet analysts often struggle to connect broad or overlapping spectral peaks with exact local structures. Differences in sample composition, peak-fitting methods, and spectral interpretation add further uncertainty.
“The exact atomic-level origins of the specific peaks derived from these defects have long remained a ‘black box,’” remarks Yasuhiro Yamada, associate professor at Chiba University, in a press release describing the research.
Working in analytical science?
Register for a FREE Separation Science account to subscribe to the Separation Science Newsletter.
Reconsidering the 285 eV XPS Peak
The study’s most consequential result concerns the C1s XPS peak at approximately 285 eV. Researchers have often assigned this signal to sp³-hybridized carbon. The new analysis indicates that this interpretation may be incorrect for some high-temperature carbon materials.
The researchers calculated spectra for 34 large graphene models containing oxygen-bearing functional groups, sp³ carbon, vacancy defects, and non-hexagonal rings. They then compared those results with experimental spectra collected from carbon fibers treated across a range of temperatures.
The results linked a C1s signal near 285 eV to carbon atoms surrounded by three rings that include at least one heptagon, octagon, or larger vacancy defect. This assignment applies when other possible sources—including sample charging, C–N bonding, and adventitious carbon—have been excluded.
The distinction matters because sp³ carbon can prove unstable at the high temperatures used to produce these materials. Assigning the peak by convention could therefore create an inaccurate picture of the defect population. The findings support a more conditional interpretation that accounts for ring geometry and local electronic structure.
The calculations also showed that C1s binding energy increased as the total size of the three rings surrounding a carbon atom increased. A related trend appeared in calculated N1s spectra: the binding energy associated with tertiary nitrogen rose as the combined size of its three surrounding rings increased.
Connecting Raman Peaks to Local Structure
The team also examined Raman signals between 1,500 and 1,550 cm⁻¹. It attributed these peaks to C=C bonds in hexagonal rings affected by nearby non-hexagonal rings and oxygen-containing groups, including cyclic ethers.
This result adds structural detail to a region that can prove difficult to interpret. Raman analysis of carbon materials often centers on the D and G bands and their intensity ratio. That approach offers a useful measure of disorder, but it does not identify every defect responsible for a signal. Peak positions can shift with strain, edge structure, heteroatoms, vacancy defects, non-hexagonal rings, and excitation wavelength. Researchers also use different numbers of fitted peaks during deconvolution, which complicates comparisons among studies.
To address this complexity, the team deconvoluted spectra from the lower-temperature carbon fibers into 17 peaks. It fixed peak positions and peak numbers using reference spectra and added peaks as they emerged across the temperature series. DFT calculations then helped connect those experimental features to candidate structures.
The analysis identified C=C bonds at DUO-type edges and pentagons within basal planes or at edges as principal contributors to the Raman D′ band. It also showed that non-hexagonal rings and oxygen-containing functional groups influence the G⁻ and A bands. Changes in C=C bond length helped explain corresponding shifts in Raman and IR spectra.
A Multi-Technique View of Carbon Defects
No single method supplied the complete structural picture. Raman spectroscopy captured changes in carbon bonding and disorder, IR spectroscopy provided information on edge configurations and functional groups, and XPS probed surface bonding states. DFT calculations created the link between observed peaks and plausible atomic arrangements.
Together, these methods indicated that isotropic pitch-based carbon fibers treated at 1,873–2,773 K contained pentagons, heptagons, octagons, a small proportion of edge sites, and cyclic ether groups. The researchers described edges, cyclic ethers, and non-hexagonal rings as the main defects in fibers heated to at least 1,873 K.
The authors limit the proposed analytical framework to carbon materials prepared at high temperatures. Materials produced at 1,073 K or below contain different and more complex functional structures, so they require an approach that follows spectral changes from a well-characterized precursor through carbonization.
Toward More Reliable Spectral Assignments
The work does not reduce carbon defect analysis to a universal set of peak labels. Instead, it shows why analysts should test conventional assignments against material history, complementary measurements, and realistic structural calculations.
That shift could improve comparisons between carbon materials and help researchers relate processing conditions to performance. More reliable defect identification may also support the controlled design of carbon fibers, battery materials, catalysts, adsorbents, filtration media, and other carbon-based technologies.
For analytical scientists, the broader lesson extends beyond carbon characterization: a familiar peak position may describe several chemical environments. Combining orthogonal techniques with computation can turn a tentative assignment into a structure-based interpretation.
