Ulrike A. Schröder

Oxygen Intercalation under Graphene on Ir(111): Energetics, Kinetics, and the Role of Graphene Edges.

Using X-ray photoemission spectroscopy (XPS) and scanning tunneling microscopy (STM) we resolve the temperature-, time-, and flake size-dependent intercalation phases of oxygen underneath graphene on Ir(111) formed upon exposure to molecular oxygen. Through the applied pressure of molecular oxygen the atomic oxygen created on the bare Ir terraces is driven underneath graphene flakes. The importance of substrate steps and of the unbinding of graphene flake edges from the substrate for the intercalation is identified. With the use of CO titration to selectively remove oxygen from the bare Ir terraces the energetics of intercalation is uncovered. Cluster decoration techniques are used as an efficient tool to visualize intercalation processes in real space.

Structure and Growth of Hexagonal Boron Nitride on Ir(111)

Using the X-ray standing wave method, scanning tunneling microscopy, low energy electron diffraction, and density functional theory, we precisely determine the lateral and vertical structure of hexagonal boron nitride on Ir(111). The moiré superstructure leads to a periodic arrangement of strongly chemisorbed valleys in an otherwise rather flat, weakly physisorbed plane. The best commensurate approximation of the moiré unit cell is (12 × 12) boron nitride cells resting on (11 × 11) substrate cells, which is at variance with several earlier studies. We uncover the existence of two fundamentally different mechanisms of layer formation for hexagonal boron nitride, namely, nucleation and growth as opposed to network formation without nucleation. The different pathways are linked to different distributions of rotational domains, and the latter enables selection of a single orientation only.

Xe Irradiation of Graphene on Ir(111): From Trapping to Blistering

Using x-ray photoelectron spectroscopy, thermal desorption spectroscopy, and scanning tunneling microscopy, we show that upon keV Xe+ irradiation of graphene on Ir(111), Xe atoms are trapped under the graphene. Upon annealing, aggregation of Xe leads to graphene bulges and blisters. The efficient trapping is an unexpected and remarkable phenomenon given the absence of chemical binding of Xe to Ir and to graphene, the weak interaction of a perfect graphene layer with Ir(111), as well as the substantial damage to graphene due to irradiation. By combining molecular dynamics simulations and density functional theory calculations with our experiments, we uncover the mechanism of trapping. We describe ways to avoid blister formation during graphene growth, and also demonstrate how ion implantation can be used to intentionally create blisters without introducing damage to the graphene layer. Our approach may provide a pathway to synthesize new materials at a substrate-2D material interface or to enable confined reactions at high pressures and temperatures. (Less)

Core level shifts of intercalated graphene

Through intercalation of metals and gases the Dirac cone of graphene on Ir(111) can be shifted with respect to the Fermi level without becoming destroyed by strong hybridization. Here, we use x-ray photoelectron spectroscopy to measure the C 1s core level shift (CLS) of graphene in contact with a number of structurally well-defined intercalation layers (O, H, Eu, and Cs). By analysis of our own and additional literature data for decoupled graphene, the C 1s CLS is found to be a non-monotonic function of the doping level. For small doping levels the shifts are well described by a rigid band model. However, at larger doping levels, a second effect comes into play which is proportional to the transferred charge and counteracts the rigid band shift. Moreover, not only the position, but also the C 1s peak shape displays a unique evolution as a function of doping level. Our conclusions are supported by intercalation experiments with Li, with which, due to the absence of phase separation, the doping level of graphene can be continuously tuned.

From Permeation to Cluster Arrays: Graphene on Ir(111) Exposed to Carbon Vapor

Our scanning tunneling microscopy and X-ray photoelectron spectroscopy experiments along with first-principles calculations uncover the rich phenomenology and enable a coherent understanding of carbon vapor interaction with graphene on Ir(111). At high temperatures, carbon vapor not only permeates to the metal surface but also densifies the graphene cover. Thereby, in addition to underlayer graphene growth, upon cool down also severe wrinkling of the densified graphene cover is observed. In contrast, at low temperatures the adsorbed carbon largely remains on top and self-organizes into a regular array of fullerene-like, thermally highly stable clusters that are covalently bonded to the underlying graphene sheet. Thus, a new type of predominantly sp2-hybridized nanostructured and ultrathin carbon material emerges, which may be useful to encage or stably bind metal in finely dispersed form.

Comment on "interfacial carbon nanoplatelet formation by ion irradiation of graphene on iridium(111)".

’ In a recent article by Herbig et al., some of us reported the formation of bulges in a graphene (Gr) sheet on Ir(111) after conducting Xeþ irradiation (energy range of 0.1 5 keV) at 300 K and subsequent annealing to 1000 K. Additional X-ray photoelectron experiments after irradiation and annealing now invalidate the following assessment in the article: “We also rule out that the bulges are agglomerations of implanted noble gas atoms. Although noble gas is certainly implanted into the Ir crystal, trapped in bulk vacancies as well as bulk vacancy clusters, and partially released during annealing to 1000 K, the Gr cover will not protect it from desorption.” Our additional experiments show that indeed the Gr cover protects Xe efficiently from desorption. In a first experiment, we exposed the bare Ir(111) sample to 3 keV Xeþ at 300 K and conducted successive annealing to 1000 and 1300 K. The ion fluence selected was 0.1 MLE, where 1 MLE is 1.57 10 ions/m, that is, numerically identical to the surface atomic density of Ir(111). After irradiation at 300 K, the bottom spectrum in Figure 1a displays an Ir 4s core level peak together with the Xe 3d3/2/Xe 3d5/2 core level doublet. Since Xe only physisorbs on Ir(111), consistent with desorption around 100 K and a binding energy of 0.21 eV as obtained by our density function theory (DFT) calculations, the Xe 3d signal after irradiation must be attributed to Xe implanted into the Ir sample. Upon annealing to 1000 K and subsequently to 1300 K, the Xe signal diminishes (compare middle and top spectra in Figure 1a, as well as solid squares in the inset). We explain these changes as follows: due to thermal excitation, Xe is partially released from its trapping sites inside the crystal, diffuses to the surface, and desorbs to the vacuum, consistent with our statement in ref 1. The residual Xe signal after annealing to 1300 K is due to Xe aggregates trapped in bulk vacancy clusters, as found also in previous studies for similar systems. Conducting precisely the same irradiation experiment, but for Ir(111) covered by a complete monolayer of Gr, yields very different photoelectron spectra, as shown in Figure 1b. Already after ion exposure at 300 K (bottom spectrum of Figure 1b), the integrated Xe 3d intensity is higher by a factor of 2 compared to irradiation of bare Ir(111) (see also inset of Figure 1a). All of the additional intensity must be due to Xe trapped between the Gr cover and the Ir substrate. Annealing to 1000 and 1300 K leads to a slight increase and a substantial increase, respectively, of the integrated Xe 3d peak intensity. The interpretation is straightforward: Instead of being released to the vacuum, the Xe diffusing out from the Ir bulk becomes trapped under the Gr cover and thereby enhances the Xe 3d intensity (smaller photoelectron attenuation due to smaller depth). By comparison with the Xe 3d signal of a saturated Xe 3dmonolayer adsorbed to Ir(111), it is found that the amount of trapped Xe is on the order of 10%of a saturated Xe layer. The efficient trapping of Xe during room temperature irradiation and subsequent annealing is remarkable since (i) a Xe atom does not bind chemically to Ir(111) or to Gr/Ir(111) (DFT binding energy 0.17 eV) but instead causes a DFTcalculated energy penalty of 2.9 eV in the trapped state due to elastic deformation of the Gr on Ir(111); (ii) the Gr cover was heavily damaged and partially sputtered (about 15% of the Gr area is removed according to the sputtering yield obtained from our molecular dynamics simulation); and (iii) the perfect Gr layer adheres only weakly to the substrate with an average height of 3.4 Å. The latter two factors make the situation different from two previously reported cases, where trapping was observed in the limits of very small fluence and very low energy (marginal damage and sputtering) and for strong adhesion of the 2D material to the substrate [a monolayer of hexagonal boron nitride on Rh(111) and Gr on Ru(0001)]. Given the large amount of Xe trapped under Gr, it appears that the bulges observed in ref 1 are formed predominantly due to Xe aggregates rather than interfacial Gr nanoplatelets as was proposed by Herbig et al. In light of the experiments reported here, bulges observed by Herbig et al. after 0.3 keV Neþ or Arþ room temperature irradiation and 1000 K annealing are most likely due to trapped Ne or Ar.