CO adsorption on Pt-induced Ge nanowires

Using density-functional theory, we investigate the possible adsorption sites of CO molecules on the recently discovered Pt-induced Ge nanowires (NWs) on Ge(001). Calculated scanning tunneling microscope (STM) images are compared to experimental STM images to identify the experimentally observed adsorption sites. The CO molecules are found to adsorb preferably onto the Pt atoms between the Ge nanowire dimer segments. This adsorption site places the CO molecule in between two nanowire dimers, pushing them outward along the NW direction, blocking the nearest equivalent adsorption sites. This explains the observed long-range repulsive interaction between CO molecules on these Pt-induced nanowires.

Figure 1

(Color online) Adsorption sites of CO on (a) a solitary NW geometry and (b) an array NW geometry. Green (red) balls indicate the positions of the germanium (platinum) atoms. Black rectangles indicate surface unit cells. The circles indicate top adsorption sites while ellipses indicate bridge configurations with the binding surface atoms at the apexes of the ellipses. In case of the NW2 surface, the location of the extra Pt atom is indicated (Refs. 19, 21). (c) Ball and stick representations of the two typical CO adsorption configurations: t(op) on the left and b(ridge) on the right. The bond lengths and angles discussed in the text and Tables are indicated.

Figure 2

(Color online) Ball-and-stick representations of the relaxed (a) A3a and (b) A3b adsorption structures: top view on the left, and side views parallel (center) and orthogonal (right) to the NW direction. The QDRs and troughs are indicated for the top view and the orthogonal side view. Bond lengths of interest are given. For the A3b configuration, the three atoms on which the CO molecule is adsorbed are indicated. Also note the Ge NW dimer atom ejected onto the QDR in the A3b configuration.

Figure 3

(Color online) Experimental (a) filled- and (b) empty-state STM images of two CO molecules adsorbed on solitary nanowires. The used sample bias is $-0.5\text{V}$ for the filled state image and 0.2 V for the empty state image. In both cases a tunneling current of 0.44 nA is used. The circles and rectangles indicate the adsorption locations of the CO molecules. The dashed line in the rectangle gives the position of the line scan shown in Figs. 5a, 5b. A single NW dimer is indicated in the filled state image, as is the distance between neighboring NWs. (c) Zoom out of a section of Fig. 1 in Ref. 16, an empty state STM image ($+1.8\text{V}$; 0.5 nA) of CO-decorated NWs, recorded at 77 K. We have indicated three CO adsorption sites. Site 1 shows a large depression in the wire. Sites 2 and 3 show large protrusions. In case of site 2, the protrusion is centered on the NW while for site 3 the protrusion is centered either left or right of the NW. The NWs in the array are separated 1.6 nm.

Figure 4

(Color online) Calculated filled- and empty-state STM images for the A2 [filled (a) and empty (b)], A3a [filled (c) and empty (d)], A3b [filled (e)], and A3b1 [filled (f)] adsorption geometries. The simulated biases are for (a), (e), and (f) $-1.50\text{V}$, (b) $+1.50\text{V}$, (c) $-0.70\text{V}$, and (d) $+0.30\text{V}$. Contours are added to guide the eye, they are separated $0.3\text{Å}$. Colored disks indicate atomic positions: green and red represent Ge and Pt atom positions in the two top layers of the surface, yellow gives the position of the Ge atoms forming the NW, and cyan and fuchsia disks give the C and O positions.

Figure 5

(Color online) Line scans of the (a) filled- and (b) empty-state STM images of the CO adsorption site observed by Öncel et al. as shown in Figs. 3a, 3b (Ref. 34). A single pristine NW dimer is indicated with a curly bracket. The arrows indicate the CO location based on the calculations performed in this work. (c) Line scan images along the NW for the clean NW at a simulated bias of $-1.50\text{V}$ (solid black curve), the CO molecule adsorbed at the A3a site at a simulated bias of $-0.70\text{V}$ (dash-dotted red curve) and $+0.30\text{V}$ (dashed blue curve). For each system $z=2.5\text{Å}$ is used to generate the calculated STM images from which the line scans are taken. The lines are shifted along the [001] axis. The position of the CO molecule along the [110] axis is indicated.

Figure 6

(Color online) Calculated STM images of CO molecules adsorbed at the B1 [(a) and (b)], B2 [(c) and (d)], and B3 sites [(e) and (f)]. The distance above the highest atom was chosen $z=2.0\text{Å}$. Filled state images (a), (c), and (e) are at a simulated bias of $-1.80\text{V}$ while for the empty state images (d) and (f) a simulated bias of $+1.80\text{V}$ was used. For the empty state image (b) a simulated bias of $+1.50\text{V}$ was used. Contours are added to guide the eye. All contours are separated $0.2\text{Å}$ in the $z$ direction. Colored disks indicate the atom positions in the top layers. Ge and Pt atoms in the two top layers are shown in green and red, respectively. Yellow disks are used to indicate the Ge NW atoms while cyan and fuchsia is used for the C and O atoms, respectively.

Figure 7

(Color online) Ball-and-stick representation of a relaxed NW2 unit cell with CO molecules adsorbed at sites (a) B1, (b) B2, and (c) B3; top views are shown on the right and side views on the left. Arrows and ellipses indicate broken bonds between the Ge NW atoms and the extra Pt atom for the (a) B1 and (c) B3 adsorption sites. For the B2 adsorption site (b) these bonds, indicated with the arrows, are not broken. The QDR and the trough regions are shown for the top-view images. The three-atom monatomic Pt wire is indicated with a blue rectangle.

Figure 8

(Color online) Calculated STM images of CO molecules adsorbed in bridging configurations B5 [(a) and (b)], A3b2 [(c) and (d)], and $\text{A}{7}_{2\text{CO}}$ [(e) and (f)]. In each case the CO molecule gives no direct image in the STM pictures, effectively remaining invisible for STM. The distance above the highest atom is chosen $z=2.0\text{Å}$. Filled state pictures (a), (c), and (e) are at a simulated bias of $-1.50\text{V}$ while for the empty state pictures (b), (d), and (f) a simulated bias of $+1.50\text{V}$ is used. Contours are added to guide the eye and are separated $0.2\text{Å}$. Colored disks indicate the atom positions in the top layers. Ge and Pt atoms in the two top layers are shown in green and red, respectively. Yellow disks are used to indicate the Ge NW atoms while cyan and fuchsia is used for the C and O atoms, respectively. Blue arrows show the position and direction of the line scans shown in Fig. 10.

Figure 9

(Color online) (a) Ball-and-stick representation of the two types of bridging CO configurations. In case of a t(op) configuration the C atom is bound to a single substrate atom while in a b(ridge) configuration the C atom is bound to two substrate atoms. Geometrical parameters given in Table are indicated, where values for different CO densities and substrates are presented. (b) and (c) show ball-and-stick representation of two systems with a t configuration and different CO density: (b) the A7 and (c) $\text{A}{7}_{2\text{CO}}$ structures. Arrows in (b) show the drift direction of the NW dimers.

Figure 10

(Color online) Calculated (a) line scan images along and (b) orthogonal to the trough/NW for the B5 (solid black curve), the A3b2 (red dashed curve), and the $\text{A}{7}_{2\text{CO}}$ (green dash-dot curve) configuration. The line scans are taken along the arrows shown in Figs. 8a, 8c, 8e. The arrows in (a) indicate the positions of the CO molecules while C and O show the positions of the C and O atom in (b).