Paramagnetic Behavior of Fullerene-Derived Nanomaterials and Their Polymer Composites: A Comprehensive EPR Study

Background

Carbon‑based nanomaterials—graphene, nanotubes, fullerenes, onion‑like carbon (OLC), nanodiamonds (ND), and carbon dots—have attracted intense research interest over the last decade. Their diverse sizes and hybrid sp¹, sp², and sp³ bonding configurations underpin unique electronic, mechanical, and chemical properties that are exploited across materials science, energy storage, biomedicine, and environmental remediation [1–10].

Fullerene derivatives, in particular, occupy a pivotal position among nanocarbon materials due to their versatile synthesis routes and functionalization capabilities. They can undergo structural transformations, such as the conversion of soot or fullerene black into OLC, or the integration of ND within OLC shells [11–14].

Recent developments have expanded fullerene applications into biology, medicine, nanocomposite fabrication, electromagnetic shielding, and beyond [15–20]. Their physical‑chemical behavior is governed by electronic structure, defect density, surface area, and other parameters. For instance, OLC nanoparticles synthesized in the presence of oxygen exhibit superior microwave‑absorption performance [18]. The high defect density and non‑planar “pyramidalization” of fullerene‑type structures markedly influence reactivity [21–24].

EPR spectroscopy is a primary tool for probing electron spin dynamics in fullerene‑like materials. Previous studies of FS and FB revealed EPR signatures with g‑values ranging from 2.0022 to 2.0023 and linewidths ΔH_pp≈2 G [25–28]. Paramagnetic radical concentrations were estimated at N_s≈10²¹ g⁻¹ for FS and 3 × 10¹⁸ g⁻¹ for FB [25,27]. These parameters remained largely unchanged in the presence of molecular oxygen, except for FB, where N_s increased by an order of magnitude after evacuation at 150 °C [27].

Understanding the interaction of fullerene‑derived materials with oxygen is essential, especially in light of recent reports linking EPR behavior to gas‑defect coupling [29,30]. The present work aims to elucidate the nature of paramagnetic defects in FS and FB, the mechanisms of oxygen interaction, and the influence of a phenolic polyamide matrix (PhC‑2) on these processes. PhC‑2, characterized by strong hydrogen‑bonded interactions, has shown promise for enhancing heat resistance and mechanical strength in space‑grade composites [31,32].

Methods

Fullerene C₆₀, FS, and FB were sourced from NeoTechProduct (St. Petersburg, Russia) and used without further modification. FS was produced via arc‑evaporation of graphite, yielding a black powder with a bulk density of ~0.25 g cm⁻³ and ~10 % fullerene content. FB was derived by extracting fullerenes from FS using o‑xylol followed by steam post‑treatment, resulting in a product with ≤0.3 % C₆₀ [https://www.neotechproduct.ru/main_page].

The PhC‑2 matrix is a linear heterocyclic copolymer containing amide groups (–HNCO–) flanked by phenyl rings, synthesized by emulsion polycondensation of metaphenylene‑diamine with a 3:2 molar mixture of isophthalic and terephthalic anhydrides.

Composite samples (PhC‑2/FS and PhC‑2/FB) were prepared by mixing in a rotating electromagnetic field, followed by compression molding at 598 K and 40 MPa. Filler loadings were 1.5 wt.% and 3 wt.%.

EPR measurements were performed at X‑band (ν ≈ 9.4 GHz) using a Radiopan X‑2244 spectrometer with 100 kHz field modulation. The g‑factor accuracy was ±2 × 10⁻⁴ for linewidths ΔH_pp≤10 G. Spin density N_s was accurate to ±50% absolute and ±20% relative. Samples were studied in ambient air and under controlled oxygen partial pressures achieved by vacuum pumping at temperatures ranging from 20 °C to 170 °C. After evacuation, samples were transferred to the spectrometer cavity without altering the vacuum conditions.

Results and Discussion

Figure 1a displays the room‑temperature EPR spectra of C₆₀, FS, and FB. All spectra share a g‑factor of 2.0024 ± 2 × 10⁻⁴. C₆₀ shows a single Lorentzian line, whereas FS and FB require a sum of two Lorentzian components. Spin concentrations and relative contributions are tabulated in Table 1. The inset presents the FB spectrum at 30 K.

EPR spectra of fullerene, FS, FB, and Phenylon C‑2 composites. a. Initial samples: 1 = C₆₀, 2 = FS, 3 = FB at room temperature. Dashed lines are fitted signals (Table 1). ν = 9350 MHz. Inset: FB spectrum at 30 K. b. Phenylon C‑2 composites (3 % filler): 1 = C₆₀, 2 = FS, 3 = FB. The supplementary broad line (1) belongs to Phenylon C‑2. ν = 9375 MHz, gain ×5. Inset: PhC‑2/FS at 30 K (fitted).

Figure 1b shows the ESR spectra of Phenylon C‑2 composites with 3 % C₆₀, FS, and FB. The inset displays the FB composite at 30 K. Parameters extracted from fitting are provided in Table 1. All samples were also examined under vacuum at temperatures 20–300 °C.

Figure 2a reveals a substantial EPR intensity increase for FS upon evacuation at room temperature. Evacuation at higher temperatures amplifies the signal by over 30×, primarily due to broad wings in the spectrum (Table 1). A comparable, albeit weaker, effect is observed in composites after high‑temperature evacuation (Figure 2b, Table 1).

EPR spectra of FS under varying oxygen pressures and PhC‑2/FS composites. a. Pressure series (1–0.001 atm) during 0.5 h pumping at 160 °C; dashed line—fit with 3 Lorentzians. Inset: total EPR intensity vs. oxygen pressure. b. PhC‑2/FS before (1) and after (2) 1 h pumping at 160 °C. Dashed lines—fit (Table 1).

The decay of the signal upon re‑exposure to ambient air was monitored for FB after 0.5 h evacuation at 300 °C. Figure 3 separates the spectrum into L1, L2, and L3 components, each characterized by distinct linewidths (0.9, 3.0, and 24 G). L3 was modeled using the theoretical 2D electron spin system (see Discussion).

Decomposition of the FB EPR spectrum after evacuation. Components L1, L2, and L3 are labeled. ν = 9375 MHz.

Figure 4 tracks the intensity decay of L1, L2, and L3 over time after air contact. The initial rapid decline (seconds to minutes) is followed by a slower, hours‑long relaxation toward equilibrium.

Decay of L1, L2, and L3 intensities after air exposure. t = 0 corresponds to the state in Figure 3.

Bulk composites (∼1.5 × 3 × 3 mm³) exhibit a markedly slower decay (Figure 5), with characteristic times exceeding those of powders by an order of magnitude due to limited gas permeability.

Decay of the total EPR signal in a bulk PhC‑2/FS composite after 1 h evacuation at 160 °C.

Annealing FS at 550 °C under weak vacuum (Figure 6) yielded a drastically different EPR profile: a single Lorentzian line with ΔH≈7–8 G and negligible intensity change upon subsequent evacuation. This contrasts with the pronounced pumping‑out effect observed at 20–300 °C, underscoring the temperature‑dependent evolution of paramagnetic characteristics [33,34].

EPR spectra of FS before and after 1 h annealing at 550 °C, and after 24 h ambient exposure.

Detailed analysis (Section 6) elucidates the origins of L1, L2, and L3. L1, with g = 2.0024 and ΔH = 0.9 G, arises from fullerene‑derived sp²/sp³ defect sites (C₁₂₀O). L2, broader (ΔH ≈ 3 G) and g = 2.0025, is attributed to edge sp³ dangling bonds on carbon flakes. L3, the broadest (ΔH ≈ 24 G) and g = 2.0025, originates from 2D electrons on the flake surfaces, consistent with dipole‑dipole broadening indicative of ~1 nm spin separations. The time‑dependent intensities of each component fit bi‑exponential decays, reflecting distinct binding energies of adsorbed oxygen to the various defect sites.

Composite behavior (Section 7) shows that the gas‑permeability of PhC‑2 dramatically slows the pumping‑out effect compared to the fillers alone. The L3 component remains prominent in composites, yet its amplitude and linewidth are moderated by the polymer matrix. This suggests that the matrix can mitigate oxygen diffusion and preserve the paramagnetic landscape.

Conclusions

Fullerene soot and fullerene black exhibit strong, reversible interactions with environmental gases, leading to near‑complete quenching (~95 %) of paramagnetic signals upon air exposure. Evacuation between 20 °C and 300 °C restores the full EPR signature, revealing a three‑component spectrum linked to specific structural features: L1 (fullerene‑derived defects), L2 (edge sp³ sites), and L3 (2D surface electrons). Theoretical modeling of L3 aligns well with experimental data. Decay kinetics of each component were quantified, demonstrating rapid initial relaxation followed by slower, long‑term equilibration.

In bulk PhC‑2 composites, the same phenomena occur but with reduced clarity due to limited oxygen diffusion. Nonetheless, the presence of fullerene fillers enhances mechanical and electronic performance while preserving a distinctive paramagnetic response.

These findings open avenues for deploying fullerene‑based nanomaterials as high‑sensitivity oxygen sensors in biomedical and environmental applications.

Abbreviations

EPR:

Electron paramagnetic resonance

ESR:

Electron spin resonance

FB:

Fullerene black

FS:

Fullerene soot

ND:

Nano‑diamond

N_s:

Spin concentration

OLC:

Onion‑like carbon

PC:

Paramagnetic center

PhC-2:

Phenylon C‑2

RT:

Room temperature