Temperature-dependent electron spin resonance (ESR) measurements are reported on stage-1 potassium-doped graphite, a model system of biased graphene. The ESR linewidth is nearly isotropic, and although the g factor has a sizable anisotropy, its majority is shown to arise due to macroscopic magnetization. Even though the homogeneous ESR linewidth shows an unusual, nonlinear temperature dependence, it appears to be proportional to the resistivity which is a quadratic function of the temperature. These observations suggests the validity of the Elliott-Yafet relaxation mechanism in KC 8 and allows us to place KC 8 on the empirical Beuneu-Monod plot among ordinary elemental metals.

We study the electron spin resonance (ESR) signal of pristine and potassium doped SWCNTs. We identify signals of a superparamagnetic background, a low intensity impurity, and of the conduction electron spin resonance (CESR). The latter only appears upon the alkali atom doping. To identify the CESR signal, we critically assess whether it could come from residual graphitic carbon, which we clearly exclude. We give accurate values for the signal intensities and the corresponding concentration of spins and for the g-factors. The CESR signal intensity allows to determine the density of states on the SWCNT assembly.

Electron spin resonance (ESR) spectroscopy is an important tool to characterize the ground state of conduction electrons and to measure their spin-relaxation times. Observing ESR of the itinerant electrons is thus of great importance in graphene and in single-wall carbon nanotubes (SWCNTs). Often, the identification of CESR signal is based on two facts: the apparent asymmetry of the ESR signal (known as a Dysonian lineshape) and on the temperature independence of the ESR signal intensity. We argue that these are insufficient as benchmarks and instead the ESR signal intensity (when calibrated against an intensity reference) yields an accurate characterization. We detail the method to obtain the density of states from an ESR signal, which can be compared with theoretical estimates. We demonstrate the success of the method for K doped graphite powder. We give a benchmark for the observation of ESR in graphene.