Phonons in potassium-doped graphene: The effects of electron-phonon interactions, dimensionality, and adatom ordering

Abstract

Graphene phonons are measured as a function of electron doping via the addition of potassium adatoms. In the low doping regime, the in-plane carbon G-peak hardens and narrows with increasing doping, analogous to the trend seen in graphene doped via the field-effect. At high dopings, beyond those accessible by the field-effect, the G-peak strongly softens and broadens. This is interpreted as a dynamic, non-adiabatic renormalization of the phonon self-energy. At dopings between the light and heavily doped regimes, we find a robust inhomogeneous phase where the potassium coverage is segregated into regions of high and low density. The phonon energies, linewidths and tunability are remarkably similar for 1-4 layer graphene, but significantly different to doped bulk graphite.PACS numbers: 63.22. Rc,78.67.Wj,81.05.ue Due to the intense scientific interest in graphene over the past few years, many of its basic properties have been determined. Now much of the effort in graphene research is devoted to tuning its properties in order to search for exotic physics and to extend and improve its potential for applications [1,2]. The properties of graphene can be tuned both by varying the number of layers in the graphene stack and via doping [2][3][4][5][6][7][8][9][10]. The current method of choice for doping graphene is via the electric field effect [3,4]. In this way the Fermi level can be controllably tuned to a maximum of E D =-0.3 eV away from the Dirac point (about 0.002 e − /C atom) giving carrier densities of ∼10 13 cm −2 . Similar levels of doping have also As the electronic structure is modified, so too is the electron-phonon interaction (EPI) [3,4,9,14,15]. A detailed understanding of this interaction is of great importance as it not only governs electronic transport, and hence the performance of graphene based electronic devices, but can also mediate exotic ground states such as superconductivity and charge density waves. At light doping levels a small (0.3%) hardening in the in-plane carbon phonon energies and narrowing in their linewidth have been reported [3][4][5][6][7][8][9][10]. This is due to a reduction in the electron-phonon scattering as the Kohn anomaly found in pure graphene at Γ is gradually removed to finite q [3, 4]. Here we extend the investigation of graphene phonons to higher dopings where we discover both a strong (3%) softening and significant linewidth broadening of the in-plane carbon phonons. We argue these effects are due to a novel, dynamic EPI arising from the 2D metallic nature of heavily doped graphene. In addition, we find that the tunability, phonons and EPI are remarkably similar for 1-4 layer doped graphene, but these systems exhibit significantly different behavior to doped bulk graphite.Graphene was prepared by micromechanical exfoliation of natural graphite onto an oxidized Si substrate (275 nm SiO 2 ) [16]. The substrate was then loaded into a sealed borosilicate tube with an optical window, evacuated and outgassed at 250 ○ C for 24 hours. An ingot of potassium metal was then ...

Superconductivity in graphite intercalation compounds

“…While this model is now has strong experimental support, interesting inconsistencies remain, for example the reported calcium isotope effect [43]. Furthermore the large electron-phonon coupling between * electrons and Cxy phonons measured in ARPES [51,52], is in contrast with DFT predictions despite being consistent with linewidths measured in Raman Spectroscopy [50,65], an effect also shown to evolve in monolayer and few layer graphene with increasing doping [66]. However despite these interesting discrepancies more recent ARPES reconcile the picture arising from DFT [54,45].…”

Raman spectroscopy of graphite intercalation compounds: Charge transfer, strain, and electron–phonon coupling in graphene layers

“…An example of the latter can be depicted by different experimental and theoretical studies reported with a wide range of different G-line positions and line-shapes between ∼1400 cm −1 and ∼1600 cm −1 : that is, at ∼1500 cm −1 [2], between 1400 cm −1 and 1550 cm −1 [17], 1534 cm −1 [69], 1547 cm −1 [21], 1420 cm −1 and 1582 cm −1 [76]. The incongruity of these results is directly related to an incomplete intercalation as reported recently in potassium doped graphene [12] and stage I GICs [70]. On top of this, defect modulation of GICs is crucial when studying the G-line response from superconducting GICs faces.…”

“…Common intercalants are alkali-metals (K, Li, Cs, Rb), alkali-earth-metals (Be, Ca, Ba), rare-earth elements or molecular groups such as FeCl 3 , AsF 5 , H 2 SO 4 , HNO 3 , etc. All the above form different tri-dimensional macro-molecules, each one with intrinsic electronic, physical and chemical properties [2,3,[9][10][11][12][13][14][15]. The properties that distinguish every GIC directly depend on the specific chemical species which occupy the carbon interlayer space.…”

Probing the charging mechanisms of carbon nanomaterial polyelectrolytes

“…Ideally, the reactivities of KC 8 and KC 24 should fall within these Fermi energy shifts for partial and complete charge transfer. Potassium GICs were produced via the vapour transport method, 20 to give the corresponding characteristic bronze stage 1 KC 8 and steel blue stage 2 KC 24 compounds (Fig. 1B).…”

“…Fractions of the supernatant were then taken for subsequent reactions. KC 8 and KC 24 were made by CAH and PLC, respectively, using the vapour transport method from natural flake graphite (Madagascar) 20 and were confirmed to be phase pure by X-ray diffraction (within the limits of the measurement). For each experiment, 20 mL, NMP (anhydrous grade, 99.5%, further dried by 4 Å molecular sieves, both purchased from SigmaAldrich, UK), was added to 10 mg KC x in a 100 mL Young's tap Schlenk tube.…”