Transport Spectroscopy of Sublattice-Resolved Resonant Scattering in Hydrogen-Doped Bilayer Graphene

We report the experimental observation of sublattice-resolved resonant scattering in bilayer graphene by performing simultaneous cryogenic atomic hydrogen doping and electron transport measurements in an ultrahigh vacuum. This allows us to monitor the hydrogen adsorption on the different sublattices of bilayer graphene without atomic-scale microscopy. Specifically, we detect two distinct resonant scattering peaks in the gate-dependent resistance, which evolve as a function of the atomic hydrogen dosage. Theoretical calculations show that one of the peaks originates from resonant scattering by hydrogen adatoms on the $\alpha$ sublattice (dimer site) while the other originates from hydrogen adatoms on the $\beta$ sublattice (nondimer site), thereby enabling a method for characterizing the relative sublattice occupancy via transport measurements. Utilizing this new capability, we investigate the adsorption and thermal desorption of hydrogen adatoms via controlled annealing and conclude that hydrogen adsorption on the $\beta$ sublattice is energetically favored. Through site-selective desorption from the $\alpha$ sublattice, we realize hydrogen doping with adatoms primarily on a single sublattice, which is highly desired for generating ferromagnetism.

Figure 1

(a) Schematic of bilayer graphene, depicting $\alpha$ and $\beta$ sites. (b) Scanning electron microscopy (SEM) image of the typical device used for in situ hydrogenation and transport measurements in this study.

Figure 2

(a) Gate-dependent resistance of undoped (black curve) bilayer graphene and after exposure to 1, 8, 24, 48, and 96 s of total atomic hydrogen dosing times at 21 K. (b) The same experimental data set in (a), with the evolution of bilayer graphene resistance as a function of $V_g-V_{\mathrm{CNP}}$. The bottom curve is for pristine graphene, and the top curve is for 96 s of hydrogen dosage. The dashed arrow lines show the progression of the two peaks labeled as $\alpha$ peak and $\beta$ peak in resistance with increasing hydrogen dosage.

Figure 3

(a)–(c) Tight-binding model calculation of the DOS, respectively, for 0.07% hydrogen doping of the $\alpha$ sublattice (dimer site) and $\beta$ sublattice (nondimer site) and corresponding momentum relaxation rates. Insets in (a) and (b) show tight-binding parametrization: $ϵ$ is the on-site energy of the hydrogen orbital, $\omega$ is the hybridization between hydrogen $s$ and carbon $p_z$ orbitals, and $t_0$ and $t_1$ are the intralayer and interlayer nearest neighbor hoppings, respectively, in the graphene bilayer. (d) Model-computed data versus experiment for $\alpha :\beta$ ratio $45:55$, where the gray curve is the experimental data in Fig. 2 with a total hydrogenation of 96 s. (e) Calculation of the bilayer graphene resistance versus carrier density ($n$ is positive for electrons) for different relative occupancies of the $\alpha$ sublattice and $\beta$ sublattice for $\alpha :\beta$ ratios ranging from $45:55$ to $0:100$ to simulate the hydrogenation process. The bottom curves are of lower total hydrogen concentration, and the top curves are of higher hydrogen concentration.

Figure 4

(a) Diagram of the annealing process for the dehydrogenation study. (b)–(d) Gate-dependent resistance for annealing temperatures from 21 to 100 K (b), from 100 to 180 K (c), and from 180 to 300 K (d). All measurements are taken at 21 K.